ATLAS User’s Manual

DEVICE SIMULATION SOFTWARE

SILVACO International
4701 Patrick Henry Drive, Bldg. 1 October 2004
Santa Clara, CA 95054
Telephone (408) 567-1000
FAX: (408) 496-6080
Internet: [Link]
E-Mail: support@[Link]
 ATLAS User’s Manual

Copyright 2004
SILVACO International
4701 Patrick Henry Drive, Building 1
Santa Clara, CA 95054

Phone: (408) 567-1000

FAX: (408) 496-6080
Internet: [Link]
E-Mail: support@[Link]

ii SILVACO International
 Notice

The information contained in this document is subject to change without notice.

SILVACO International MAKES NO WARRANTY OF ANY KIND WITH REGARD TO

THIS MATERIAL, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTY
OF FITNESS FOR A PARTICULAR PURPOSE.

SILVACO International Inc. shall not be held liable for errors contained herein or for
incidental or consequential damages in connection with the furnishing, performance, or use of
this material.

This document contains proprietary information, which is protected by copyright laws of the
United States. All rights are reserved. No part of this document may be photocopied,
reproduced, or translated into another language without the prior written consent of SILVACO
INTERNATIONAL.

Simulation Standard, TCAD Driven CAD, Virtual Wafer Fab, Analog Alliance, Legacy,
ATHENA, ATLAS, MERCURY, VICTORY, VYPER, ANALOG EXPRESS, RESILIENCE,
DISCOVERY, CELEBRITY, Manufacturing Tools, Automation Tools, Interactive Tools,
TonyPlot, TonyPlot3D, DeckBuild, DevEdit, DevEdit3D, Interpreter, ATHENA Interpreter,
ATLAS Interpreter, Circuit Optimizer, MaskViews, PSTATS, SSuprem3, SSuprem4, Elite,
Optolith, Flash, Silicides, MC Depo/Etch, MC Implant, S-Pisces, Blaze/Blaze3D, Device3D,
TFT2D/3D, Ferro, SiGe, SiC, Laser, VCSELS, Quantum2D/3D, Luminous2D/3D, Giga2D/3D,
MixedMode2D/3D, FastBlaze, FastLargeSignal, FastMixedMode, FastGiga, FastNoise,
Mocasim, Spirit, Beacon, Frontier, Clarity, Zenith, Vision, Radiant, TwinSim, UTMOST,
UTMOST II, UTMOST III, UTMOST IV, PROMOST, SPAYN, UTMOST IV Measure,UTMOST
IV Fit, UTMOST IV Spice Modeling, SmartStats, SDDL, SmartSpice, FastSpice, Twister, Blast,
MixSim, SmartLib, TestChip, Promost-Rel, RelStats, RelLib, Harm, Ranger, Ranger3D Nomad,
QUEST, EXACT, CLEVER, STELLAR, HIPEX-net, HIPEX-r, HIPEX-c, HIPEX-rc, HIPEX-crc,
EM, Power, IR, SI, Timing, SN, Clock, Scholar, Expert, Savage, Scout, Dragon, Maverick,
Guardian, Envoy, LISA, ExpertViews and SFLM are trademarks of Silvaco International.

All other trademarks mentioned in this manual are the property of their respective owners.

© 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 2000, 2002, 2004 by SILVACO
International Inc.

SILVACO International iii

 Reader Comment Sheet

We welcome your evaluation of this manual. Your comments and suggestions help us to improve
our publications. If you have any responses to the questions below, please let us know. Please
write your bservations down and send your complaints, bug reports, suggestions or comments to
the e-mail address listed below.
• Is this manual technically accurate?
• Are the concepts and wording easy to understand?
• Is the size of this manual convenient for you?
• Is the manual's arrangement convenient for you?
• Do you consider this manual to be easily readable?

Please add any additional relevant comments and fax your comments to:

SILVACO International

Attention: Technical Publications

4701 Patrick Henry Drive, Building 1

Santa Clara, CA 95054

You can also e-mail us at support@[Link] or visit our website at [Link]

iv SILVACO International
Intended Audience

The information presented is based on the assumptions that the reader is (1) familiar with the
basic terminology of semiconductor processing and semiconductor device operation, and (2)
understands the basic operation of the computer hardware and operating system being
employed.

Introduction

ATLAS is a modular and extensible framework for one, two and three dimensional
semiconductor device simulation. It is implemented using modern software engineering
practices that promote reliability, maintainability, and extensibility. Products that use the
ATLAS Framework meet the device simulation needs of all semiconductor application areas.

SILVACO International makes no warranty of any kind with regard to this material, including,
but not limited to, the implied warranty of fitness for a particular purpose.

SILVACO International shall not be liable for errors contained herein or for incidental or
consequential damages in connection with furnishing, performance, or use of this material. This
document contains proprietary information protected by copyright. All rights are reserved. No
part of this document may be photocopied, reproduced, or translated into another language
without the prior written consent of SILVACO International.

The edition number and the issue date is shown below.

History
• Edition 1 - July 1, 1993

SILVACO International v
Technical Support

If you have difficulties or questions relating to the usage of any Silvaco product, contact the
Silvaco support personnel at support@[Link].

When you send us an e-mail message, please:

• Explain the problem or question as fully as possible.
• Include any input files that you have created and error messages that were generated.
• Provide us with the version number of ATLAS, the version numbers of the VWF INTERACTIVE
TOOLS that you are using, and the details of the hardware platform on which the problem was
encountered.
• Include your company affiliation, business telephone number, and fax number.

You will be contacted promptly and your problem will be resolved as quickly as possible.

Your feedback will help us further develop and improve ATLAS. Please send your comments to
our e-mail address.

vi SILVACO International
 How to Read this Manual

Style Conventions

Font Style/Convention Description Example

• This represents a list of items or • Bullet A
terms. • Bullet B
• Bullet C

1. This represents a set of directions to To open a door:

2. perform an action. 1. Unlock the door by inserting
3. the key into keyhole.
2. Turn key counter-clockwise.
3. Pull out the key from the
keyhole.
4. Grab the doorknob and turn
clockwise and pull.

→ This represents a sequence of menu File→Open

options and GUI buttons to perform
an action.

Courier This represents the commands, HAPPY BIRTHDAY

parameters, and variables syntax.

New Century Schoolbook This represents the menu options and File
Bold buttons in the GUI.

New Century Schoolbook This represents the equations. abc=xyz

Italics

This represents the additional

Note: important information. Note: Make sure you save often while run-
ning an experiment.

NEW CENTURY SCHOOLBOOK This represents the names of the ATHENA, ATLAS, EXPERT,
INSMALL CAPS SILVACO/SIMUCAD AUTOMATION GATEWAY, HIPEX, SMARTSPICE,
DESIGN Products. STELLAR, and UTMOST.

SILVACO International vii

 This page is intentionally left blank.

viii SILVACO International

 Table of Contents

Chapter 1:
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
1.1: ATLAS Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
1.2: Features And Capabilities of ATLAS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
1.2.1: Comprehensive Set of Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
1.2.2: Fully Integrated Capabilities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
1.2.3: Sophisticated Numerical Implementation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
1.3: Using ATLAS With Other Silvaco Software. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
1.4: The Nature Of Physically-Based Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
Chapter 2:
Getting Started with ATLAS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
2.1: Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
2.2: ATLAS Inputs and Outputs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
2.3: Modes of Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
2.3.1: Interactive Mode With DeckBuild . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
2.3.2: Batch Mode With DeckBuild. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
2.3.3: No Windows Batch Mode With DeckBuild . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
2.3.4: Running ATLAS inside Deckbuild . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
2.3.5: Batch Mode Without DeckBuild . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
2.4: Accessing The Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-5
2.5: The ATLAS Syntax. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-7
2.5.1: Statements and Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-7
2.5.2: The Order of ATLAS Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-7
2.5.3: The DeckBuild Command Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-8
2.5.4: PISCES-II Quick Start. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-8
2.6: Defining A Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
2.6.1: Interface From ATHENA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
2.6.2: Interface From DevEdit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-11
2.6.3: Using The Command Language To Define A Structure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-11
2.6.4: Automatic Meshing (Auto-meshing) Using The Command Language . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-15
2.6.5: Modifying Imported Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-22
2.6.6: Remeshing Using The Command Language. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-23
2.6.7: Specifying 3D Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-24
2.6.8: General Comments Regarding Grids. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-25
2.6.9: Maximum Numbers Of Nodes, Regions, and Electrodes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-26
2.7: Defining Material Parameters And Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-27
2.7.1: Specifying Contact Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-27
2.7.2: Specifying Material Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-30
2.7.3: Specifying Interface Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-31
2.7.4: Specifying Physical Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-31
2.7.5: Summary Of Physical Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-33
2.8: Choosing Numerical Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-38
2.8.1: Numerical Solution Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-38
2.9: Obtaining Solutions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-42
2.9.1: DC Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-42
2.9.2: The Importance Of The Initial Guess . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-43
2.9.3: Small-Signal AC Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-45

SILVACO International ix
ATLAS User’s Manual

2.9.4: Transient Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-46

2.9.5: Advanced Solution Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-47
2.9.6: Using DeckBuild To Specify SOLVE Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-49
2.10: Interpreting The Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-50
2.10.1: Run-Time Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-50
2.10.2: Log Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-51
2.10.3: Parameter Extraction In DeckBuil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-52
2.10.4: Functions In TonyPlot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-53
2.10.5: Solution Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-54
2.10.6: Technology Specific Issues in ATLAS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-56

Chapter 3:
Physics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
3.1: Basic Semiconductor Equations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
3.1.1: Poisson’s Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
3.1.2: Carrier Continuity Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
3.1.3: The Transport Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
3.1.4: Displacement Current Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
3.2: Basic Theory of Carrier Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-5
3.2.1: Fermi-Dirac and Boltzmann Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-5
3.2.2: Effective Density of States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-5
3.2.3: Intrinsic Carrier Concentration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-6
3.2.4: Evaluation of Fermi-Dirac Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
3.2.5: The Energy Bandgap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
3.2.6: Bandgap Narrowing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
3.2.7: The Universal Bandgap Narrowing Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-9
3.2.8: Passler's Model for Temperature Dependent Bandgap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-9
3.3: Space Charge from Incomplete Ionization, Traps, and Defects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-10
3.3.1: Incomplete Ionization of Impurities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-10
3.3.2: Low Temperature Simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-11
3.3.3: Traps and Defects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-12
3.4: The Energy Balance Transport Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-19
3.4.1: The Energy Balance Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-19
3.4.2: Density of States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-22
3.4.3: Energy Density Loss Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-22
3.4.4: Temperature Dependence of Relaxation Times . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-23
3.4.5: Energy Dependent Mobilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-24
3.5: Boundary Physics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-25
3.5.1: Ohmic Contacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-25
3.5.2: Schottky Contacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-25
3.5.3: Floating Contacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-29
3.5.4: Current Boundary Conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-30
3.5.5: Insulating Contacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-30
3.5.6: Neumann Boundaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-30
3.5.7: Lumped Element Boundaries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-31
3.5.8: Distributed Contact Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-32
3.5.9: Energy Balance Boundary Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-33
3.6: Physical Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-34
3.6.1: Mobility Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-34
3.6.2: Mobility Model Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-63
3.6.3: Carrier Generation-Recombination Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-64
3.6.4: Impact Ionization Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-70
3.6.5: Gate Current Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-81
3.6.6: Device Level Reliability Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-93

x SILVACO International
 Table of Contents

3.6.7: The Ferroelectric Permittivity Model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-94

3.6.8: Polarization in Wurtzite Materials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-95
3.7: Quasistatic Capacitance - Voltage Profiles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-97
3.8: Conductive Materials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-98
3.9: Optoelectronic Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-99
3.9.1: The General Radiative Recombination Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-99
3.9.2: The Default Radiative Recombination Model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-100
3.9.3: The Standard Gain Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-100
3.9.4: The Empirical Gain Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-101
3.9.5: Tayamaya's Gain Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-102
3.9.6: Band Stucture Dependent Optoelectronic Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-102
3.9.7: Yan's and Li's Models for Gain and Radiative Recombination in Zincblende Material . . . . . . . . . . . . . . . . . . . . . 3-103
3.9.8: Chuang's Three Band Model for Gain and Radiative Recombination in Wurtzite Materials . . . . . . . . . . . . . . . . . 3-106
3.9.9: Lorentzian Gain Broadening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-108
3.9.10: Ishikawa's Strain Effects Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-109

Chapter 4:
S-Pisces: Silicon Based 2D Simulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1
4.1: Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1
4.2: Simulating Silicon Devices Using S-Pisces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2
4.2.1: Simulating MOS Technologies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2
4.2.2: Simulating Silicon Bipolar Devices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-5
4.2.3: Simulating Non-Volatile Memory Technologies (EEPROMs, FLASH Memories) . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-7
4.2.4: Simulating SOI Technologies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-8

Chapter 5:
Blaze: Compound Material 2D Simulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1
5.1: Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1
5.1.1: Basic Heterojunction Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1
5.1.2: Alignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
5.1.3: The Drift Diffusion Transport Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-12
5.1.4: The Thermionic Emission and Field Emission Transport Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-13
5.2: The Physical Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-16
5.2.1: Common Physical Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-16
5.2.2: Recombination and Generation Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-18
5.3: Material Dependent Physical Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-19
5.3.1: Gallium Arsenide (GaAs) Physical Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-19
5.3.2: Al(x)Ga(1-x)As System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-21
5.3.3: In(1-x)Ga(x)As(y)P(1-y) System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-23
5.3.4: The Si(1-x)Ge(x) System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-25
5.3.5: Silicon Carbide (SiC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-27
5.3.6: GaN, InN, AlN, AlGaN, and InGaN System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-28
5.3.7: The Hg(1-x)Cd(x)Te System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-34
5.4: Simulating Heterojunction Devices with Blaze. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-35
5.4.1: Defining Material Regions with Positionally-Dependent Band Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-35
5.4.2: Defining Materials and Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-36

SILVACO International xi
ATLAS User’s Manual

Chapter 6:
3D Device Simulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1
6.1: 3D Device Simulation Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1
6.1.1: DEVICE3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1
6.1.2: BLAZE3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1
6.1.3: GIGA3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1
6.1.4: TFT3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
6.1.5: MIXEDMODE3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
6.1.6: QUANTUM3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
6.1.7: LUMINOUS3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
6.2: 3D Structure Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3
6.3: Model And Material Parameter Selection in 3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4
6.3.1: Mobility. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4
6.3.2: Recombination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4
6.3.3: Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4
6.3.4: Carrier Statistics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4
6.3.5: Boundary Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4
6.3.6: Interface and Bulk Traps (Discrete and Continuous) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-5
6.3.7: Single Event Upset Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-5
6.3.8: Boundary Conditions in 3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-7
6.3.9: BLAZE3D Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-8
6.3.10: TFT3D Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-8
6.3.11: QUANTUM3D Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-8
6.3.12: LUMINOUS3D Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-8
6.4: Numerical Methods for 3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12
6.4.1: DC Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12
6.4.2: Transient Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12
6.4.3: Obtaining Solutions In 3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12
6.4.4: Interpreting the Results From 3D. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12
6.4.5: More Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12

Chapter 7:
Giga: Self-Heating Simulator. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1
7.1: Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1
7.1.1: Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1
7.1.2: Numerics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1
7.2: Physical Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-2
7.2.1: The Lattice Heat Flow Equation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-2
7.2.2: Non-Isothermal Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-4
7.2.3: Heat Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-5
7.2.4: Thermal Boundary Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-6
7.2.5: Temperature Dependent Material Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-7
7.2.6: C-Interpreter Defined Peltier Coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-8
7.3: Applications of GIGA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-9
7.3.1: Power Device Simulation Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-9
7.3.2: More Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-9

xii SILVACO International

 Table of Contents

Chapter 8:
Laser: Edge Emitting Simulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1
8.1: Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1
8.2: Physical Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2
8.2.1: Helmholtz Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2
8.2.2: Local Optical Gain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-3
8.2.3: Stimulated Emission. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-4
8.2.4: Photon Rate Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-4
8.2.5: Spontaneous Recombination Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-6
8.2.6: Optical Power. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-6
8.2.7: Gain Saturation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-6
8.3: Solution Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-7
8.4: Specifying Laser Simulation Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-8
8.4.1: LASER Statement Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-8
8.4.2: Numerical Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9
8.5: Semiconductor Laser Simulation Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-10
8.5.1: Generation of Near-Field and Far-Field Patterns. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-10

Chapter 9:
VCSEL Simulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1
9.1: Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1
9.2: Physical Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-2
9.2.1: Reflectivity Test Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-2
9.2.2: Helmholtz Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-4
9.2.3: Local Optical Gain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-6
9.2.4: Photon Rate Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-6
9.3: Simulating Vertical Cavity Surface Emitting Lasers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-8
9.3.1: Specifying the Device Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-8
9.3.2: Specifying VCSEL Physical Models and Material Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-12
9.3.3: Enabling VCSEL Solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-12
9.3.4: Numerical Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-13
9.3.5: Alternative VCSEL Simulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-14
9.4: Semiconductor Laser Simulation Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-16
Chapter 10:
Luminous: Optoelectronic Simulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1
10.1: Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1
10.2: Simulation Method. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-2
10.2.1: Ray Tracing in 2D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-2
10.2.2: Ray Tracing in 3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-3
10.2.3: Reflection and Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-4
10.2.4: Matrix Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-9
10.3: Generation of Photocurrent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-12
10.3.1: Light Absorption and Photogeneration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-12
10.3.2: Photocurrent and Quantum Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-13
10.4: Simulating Photodetectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-15
10.4.1: Defining Optical Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-15
10.4.2: Defining Optical Properties of Materials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-20
10.4.3: Extracting Dark Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-21
10.4.4: Extracting Detection Efficiency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-23
10.4.5: Obtaining Quantum Efficiency versus Bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-23
10.4.6: Obtaining Transient Response to Optical Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-23
10.4.7: Obtaining Frequency Response to Optical Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-24

SILVACO International xiii

ATLAS User’s Manual

10.4.8: Obtaining Spatial Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-24

10.4.9: Obtaining Spectral Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-24
10.5: Simulating Solar Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-26
10.5.1: Obtaining Open Circuit Voltage and Short Circuit Current. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-26
10.6: Beam Propagation Method in 2D. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-27
10.6.1: Using BPM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-27
10.6.2: Light Propagation In A Multiple Region Device. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-28
10.6.3: Fast Fourier Transform (FFT) Based Beam Propagation Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-28

Chapter 11:
LED: Light Emitting Diode Simulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-1
11.1: Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-1
11.2: Defining Light Emitting Devices (LEDs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-2
11.3: Specifying Light Emitting Diode Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-4
11.3.1: Specifying Polarization and Piezoelectric Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-4
11.3.2: Choosing Radiative Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-4
11.3.3: Using k.p Band Parameter Models in Drift Diffusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-5
11.4: Data Extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-6
11.4.1: Extracting Luminous Intensity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-6
11.4.2: Extracting Emission Spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-6
11.4.3: Extracting Emission Wavelength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-7
11.5: Reverse Ray-Tracing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-8
Chapter 12:
MixedMode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-1
12.1: Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-1
12.1.1: Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-1
12.1.2: Advantages of MixedMode Simulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-2
12.2: Using MixedMode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-3
12.2.1: General Syntax Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-3
12.2.2: Circuit and Analysis Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-4
12.2.3: Device Simulation Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-7
12.2.4: Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-7
12.3: A Sample Command File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-12
12.4: MixedMode Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-14
12.4.1: Circuit Element Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-14
12.4.2: Control and Analysis Statements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-24
12.4.3: Transient Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-33
12.4.4: User-Defined Two-Terminal Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-36

Chapter 13:
Quantum: Quantum Effect Simulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-1
13.1: Self-Consistent Coupled Schrodinger Poisson Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-1
13.2: Density Gradient (Quantum Moments Model) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-3
13.3: Bohm Quantum Potential (BQP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-5
13.3.1: Calibration against Schrodinger-Poisson Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-5
13.3.2: Post Calibration runs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-7
13.4: Quantum Correction Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-10
13.4.1: Hansch’s Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-10
13.4.2: Van Dort’s Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-10
13.5: General Quantum Well Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-12
13.6: Multiple Quantum Well Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-13
13.6.1: Specifying MQW Location and Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-13

xiv SILVACO International

 Table of Contents

Chapter 14:
TFT: Thin-Film Transistor Simulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-1
14.1: Polycrystalline and Amorphous Semiconductor Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-1
14.2: Simulating TFT Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-2
14.2.1: Defining The Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-2
14.2.2: Defining The Defect States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-2
14.2.3: Density of States Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-2
14.2.4: Trapped Carrier Density. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-3
14.2.5: Steady-state Trap Recombination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-5
14.2.6: Transient Traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-6
14.2.7: Continuous Defects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-9
14.2.8: Discrete Defects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-9
14.2.9: Plotting The Density Of States Versus Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-10
14.2.10: Using the C-Interpreter to define DEFECTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-10
14.2.11: Setting Mobility and Other Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-10

Chapter 15:
NOISE: Electronic Noise Simulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-1
15.1: Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-1
15.2: Simulating Noise in ATLAS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-2
15.3: Circuit Level Description of Noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-3
15.3.1: Noise “Figures of Merit” for a Two-Port Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-4
15.4: Noise Calculation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-6
15.4.1: The Impedance Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-6
15.4.2: Microscopic Noise Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-7
15.4.3: Local Noise Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-7
15.5: ATLAS Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-8
15.5.1: Diffusion Noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-8
15.5.2: Generation-Recombination Noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-9
15.5.3: Flicker Noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-11
15.6: Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-13
15.6.1: Log Files. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-13
15.6.2: Structure Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-14

Chapter 16:
Thermal 3D: Thermal Packaging Simulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-1
16.1: Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-1
16.1.1: 3D Structure Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-1
16.2: Model and Material Parameter Selection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-2
16.2.1: Thermal Simulation Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-2
16.2.2: Setting Thermal Conductivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-2
16.2.3: Suggested Parameters For Thermal Conductivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-3
16.3: Numerical Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-4
16.4: Obtaining Solutions In THERMAL3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-5
16.5: Interpreting The Results From THERMAL3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-6
16.6: More Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-7

SILVACO International xv
ATLAS User’s Manual

Chapter 17:
Numerical Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-1
17.1: Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-1
17.2: Numerical Solution Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-2
17.3: Meshes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-3
17.3.1: Mesh Regridding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-3
17.3.2: Mesh Smoothing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-4
17.4: Discretization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-5
17.4.1: The Discretization Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-5
17.5: Non-Linear Iteration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-6
17.5.1: Newton Iteration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-6
17.5.2: Gummel Iteration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-6
17.5.3: Block Iteration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-7
17.5.4: Combining The Iteration Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-7
17.5.5: Solving Linear Subproblems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-7
17.5.6: Convergence Criteria for Non-linear Iterations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-8
17.5.7: Error Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-8
17.5.8: Terminal Current Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-9
17.5.9: Convergence Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-10
17.5.10: Detailed Convergence Criteria. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-11
17.6: Initial Guess Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-17
17.6.1: Recommendations And Defaults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-18
17.7: The DC Curve-Tracer Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-19
17.8: Transient Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-20
17.9: Small Signal and Large Signal Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-21
17.9.1: Frequency Domain Perturbation Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-21
17.9.2: Fourier Analysis Of Transient Responses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-22
17.9.3: Overall Recommendations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-23
17.10: Differences Between 2D and 3D Numerics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-24
Chapter 18:
Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-1
18.1: Input Language . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-1
18.1.1: Syntax Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-1
18.2: BEAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-4
18.3: COMMENT, # . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-12
18.4: CONTACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-13
18.5: CURVETRACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-19
18.6: DBR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-22
18.7: DEFECTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-26
18.8: DEGRADATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-29
18.9: DOPING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-30
18.10: ELECTRODE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-44
18.11: ELIMINATE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-48
18.12: EXTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-50
18.13: [Link]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-51
18.14: FOURIER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-53
18.15: GO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-55
18.16: IMPACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-56
18.17: INTDEFECTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-63
18.18: INTERFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-66
18.19: INTTRAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-70

xvi SILVACO International

 Table of Contents

18.20: LASER. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-73

18.21: LOAD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-78
18.22: LOG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-80
18.23: [Link], [Link]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-84
18.24: MATERIAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-85
18.25: MEASURE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-101
18.26: MESH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-104
18.27: METHOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-107
18.28: MOBILITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-116
18.29: MODELS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-129
18.30: MQW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-151
18.31: OPTIONS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-155
18.32: OUTPUT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-156
18.33: PROBE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-162
18.34: QUIT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-168
18.35: REGION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-169
18.36: REGRID. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-175
18.37: SAVE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-179
18.38: SET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-183
18.39: SINGLEEVENTUPSET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-184
18.40: SOLVE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-186
18.41: [Link], [Link] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-200
18.42: SYMBOLIC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-201
18.43: SPREAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-202
18.44: SYSTEM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-205
18.45: THERMCONTACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-206
18.46: TITLE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-208
18.47: TONYPLOT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-209
18.48: TRAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-210
18.49: UTMOST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-212
18.50: VCSEL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-216
18.51: [Link], [Link], [Link] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-218
Appendix A:
C-Interpreter Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-1
A.1: Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-1
Appendix B:
Material Systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1
B.1: Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1
B.2: Semiconductors, Insulators, and Conductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-2
B.2.1: Semiconductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-2
B.2.2: Insulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-2
B.2.3: Conductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-2
B.2.4: Unknown Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-2
B.3: ATLAS Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-3
B.3.1: Specifying Compound Semiconductors Rules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-4
B.4: Silicon and Polysilicon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-5
B.5: The Al(x)Ga(1-x)As Material System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-8
B.6: The In(1-x)Ga(x)As(y)P(1-y) System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-10
B.7: Silicon Carbide (SiC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-11

SILVACO International xvii

ATLAS User’s Manual

B.8: Material Defaults for GaN/InN/AlN System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-12

B.9: Material Defaults for Compound Semiconductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-14
B.10: Miscellaneous Semiconductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-19
B.11: Insulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-23
B.12: Metals/Conductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-24
B.13: Optical Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-25
B.14: User Defined Materials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-26
Appendix C:
Hints and Tips. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-1
Appendix D:
ATLAS Version History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-1
D.1: Version History. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-1
D.1.1: Version 5.8.0.R . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-1
D.1.2: Version 5.6.0.R . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .D-3
D.1.3: Version 5.2.0.R . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-4
D.1.4: Version 5.0.0.R . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-5
D.1.5: Version 4.3.0.R . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-8
D.1.6: Version 4.0.0.R . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-9
D.1.7: Version 3.0.0.R . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-10
D.1.8: Version 2.0.0.R . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-14

xviii SILVACO International

 Chapter 1:
Introduction

1.1: ATLAS Overview

ATLAS provides general capabilities for physically-based two (2D) and three-dimensional
(3D) simulation of semiconductor devices. If you’re new to ATLAS, read this chapter and Chapter 2:
“Getting Started with ATLAS” to understand how ATLAS works. Chapter 2 is a tutorial for ATLAS.
Once you’ve read these chapters, the remaining chapters can then be referred to for a detailed
understanding of the capabilities of each ATLAS product.
Those who have used earlier versions of ATLAS may find it helpful to review the updated version
history in Appendix D: “ATLAS Version History”.
ATLAS is designed to be used in conjunction with the VWF INTERACTIVE TOOLS. The VWF
INTERACTIVE TOOLS, which include DECKBUILD, TONYPLOT, DEVEDIT, MASKVIEWS, and OPTIMIZER, are
documented in the VWF INTERACTIVE TOOLS USER MANUALS. See Section 1.3: “Using ATLAS With
Other Silvaco Software” for more information about using ATLAS with other Silvaco tools.
If you’re new to ATLAS, also read the tutorial chapter in the VWF INTERACTIVE TOOLS USER’S MANUAL
to get a basic idea of how the user interface works.
ATLAS is supplied with numerous examples that can be accessed through DECKBUILD. These
examples demonstrate most of ATLAS’s capabilities. The input files that are provided with the
examples are an excellent starting point for developing your own input files. To find out how to access
the example, see Chapter 2: “Getting Started with ATLAS”, Section 2.4: “Accessing The Examples”.

SILVACO International 1-1

ATLAS User’s Manual

1.2: Features And Capabilities of ATLAS

1.2.1: Comprehensive Set of Models
ATLAS provides a comprehensive set of physical models, including:
• DC, AC small-signal, and full time-dependency.
• Drift-diffusion transport models.
• Energy balance and Hydrodynamic transport models.
• Lattice heating and heatsinks.
• Graded and abrupt heterojunctions.
• Optoelectronic interactions with general ray tracing.
• Amorphous and polycrystalline materials.
• General circuit environments.
• Stimulated emission and radiation
• Fermi-Dirac and Boltzmann statistics.
• Advanced mobility models.
• Heavy doping effects.
• Full acceptor and donor trap dynamics
• Ohmic, Schottky, and insulating contacts.
• SRH, radiative, Auger, and surface recombination.
• Impact ionization (local and non-local).
• Floating gates.
• Band-to-band and Fowler-Nordheim tunneling.
• Hot carrier injection.
• Quantum transport models
• Thermionic emission currents.

1.2.2: Fully Integrated Capabilities

ATLAS works well with other software from SILVACO. For example, ATLAS
• Runs in the DECKBUILD interactive run-time environment.
• Is interfaced to TONYPLOT, the interactive graphics and analysis package.
• Accepts input from the ATHENA and SSUPREM3 process simulators.
• Is interfaced to UTMOST parameter extraction and device modeling software.
• can be used in experiments with the VWF AUTOMATION TOOLS.

1.2.3: Sophisticated Numerical Implementation

ATLAS uses powerful numerical techniques, including:
• Accurate and robust discretization techniques.
• Gummel, Newton, and block-Newton nonlinear iteration strategies.
• Efficient solvers, both direct and iterative, for linear subproblems.
• Powerful initial guess strategies.
• Small-signal calculation techniques that converge at all frequencies.
• Stable and accurate time integration.

1-2 SILVACO International

 Introduction

1.3: Using ATLAS With Other Silvaco Software

ATLAS should only be used in conjunction with the VWF INTERACTIVE TOOLS. These include
DECKBUILD, TONYPLOT, DEVEDIT, MASKVIEWS, and OPTIMIZER. DECKBUILD provides an interactive
run time environment. TONYPLOT supplies scientific visualization capabilities. DEVEDIT is an
interactive tool for structure and mesh specification and refinement. MASKVIEWS is an IC Layout
Editor. The OPTIMIZER supports black box optimization across multiple simulators.
ATLAS is often used in conjunction with the ATHENA process simulator. ATHENA predicts the
physical structures that result from processing steps. The resulting physical structures are used as
input by ATLAS, which then predicts the electrical characteristics associated with specified bias
conditions. The combination of ATHENA and ATLAS makes it possible to determine the impact of
process parameters on device characteristics.
The electrical characteristics predicted by ATLAS can be used as input by the UTMOST device
characterization and SPICE modeling software. Compact models based on simulated device
characteristics can then be supplied to circuit designers for preliminary circuit design. Combining
ATHENA, ATLAS, UTMOST, and SMARTSPICE makes it possible to predict the impact of process
parameters on circuit characteristics.
ATLAS can be used as one of the simulators within the VWF AUTOMATION TOOLS. VWF makes it
convenient to perform highly automated simulation-based experimentation. VWF is used in a way that
reflects experimental research and development procedures using split lots. It therefore links
simulation very closely to technology development, resulting in significantly increased benefits from
simulation use.

SILVACO International 1-3

ATLAS User’s Manual

1.4: The Nature Of Physically-Based Simulation

ATLAS is a physically-based device simulator. Physically-based device simulation is not a familiar
concept for all engineers. This section will briefly describe this type of simulation.
Physically-based device simulators predict the electrical characteristics that are associated with
specified physical structures and bias conditions. This is achieved by approximating the operation of a
device onto a two or three dimensional grid, consisting of a number of grid points called nodes. By
applying a set of differential equations, derived from Maxwells laws, onto this grid you can simulate
the transport of carriers through a structure. This means that the electrical performance of a device
can now be modeled in DC, AC or transient modes of operation.
There are three physically-based simulation. These are:
• It is predictive.
• It provides insight.
• It conveniently captures and visualizes theoretical knowledge.
Physically-based simulation is different from empirical modeling. The goal of empirical modeling is to
obtain analytic formulae that approximate existing data with good accuracy and minimum complexity.
Empirical models provide efficient approximation and interpolation. They do not provide insight, or
predictive capabilities, or encapsulation of theoretical knowledge.
Physically-based simulation has become very important for two reasons. One, it is almost always much
quicker and cheaper than performing experiments. Two, it provides information that is difficult or
impossible to measure.
The drawbacks of physically-based simulation are that all the relevant physics must be incorporated
into a simulator, and numerical procedures must be implemented to solve the associated equations.
These tasks have been taken care of for users of ATLAS.
Those who use physically-based device simulation tools must specify the problem to be simulated. In
ATLAS, specify device simulation problems by defining:
• The physical structure to be simulated.
• The physical models to be used.
• The bias conditions for which electrical characteristics are to be simulated.
The subsequent chapters of this manual describe how to perform these steps.

1-4 SILVACO International

 Chapter 2:
Getting Started with ATLAS

2.1: Overview
ATLAS is a physically-based two and three dimensional device simulator. It predicts the electrical
behavior of specified semiconductor structures and provides insight into the internal physical
mechanisms associated with device operation.
ATLAS can be used standalone or as a core tool in Silvaco’s VIRTUAL WAFER FAB simulation
environment. In the sequence of predicting the impact of process variables on circuit performance,
device simulation fits between process simulation and SPICE model extraction.
This chapter will show you how to use ATLAS effectively. It is a source of useful hints and advice. The
organization of topics parallels the steps that you go through to run the program. If you have used
earlier versions of ATLAS, you will still find this chapter useful because of the new version.
This chapter concentrates on the core functionality of ATLAS. If you’re primarily interested in the
specialized capabilities of a particular ATLAS tool, read this chapter first, and then read the chapters
that describe the ATLAS tools you wish to use.

SILVACO International 2-1

ATLAS User’s Manual

2.2: ATLAS Inputs and Outputs

Figure 2-1 shows the types of information that flow in and out of ATLAS. Most ATLAS simulations use
two input files. The first input file is a text file that contains commands for ATLAS to execute. The
second input file is a structure file that defines the structure that will be simulated.
ATLAS produces three types of output files. The first type of output file is the run-time output, which
gives you the progress and the error and warning messages as the simulation proceeds. The second
type of output file is the log file, which stores all terminal voltages and currents from the device
analysis. The third type of output file is the solution file, which stores 2D and 3D data relating to the
values of solution variables within the device at a given bias point.

DevEdit

(Structure and
Mesh Editor) Runtime Output

Structure Files

ATHENA ATLAS Log Files

Device Simulator
(Process Simulator)
TonyPlot

Command File (Visualization

Tool)
Solution Files

DeckBuild

(Run Time Environment)

Figure 2-1: ATLAS Inputs and Outputs

2-2 SILVACO International

 Getting Started with ATLAS

2.3: Modes of Operation

ATLAS is normally used in conjunction with the DECKBUILD run-time environment, which supports
both interactive and batch mode operation. We strongly recommend that you always run ATLAS
within DECKBUILD. In this section, we present the basic information you need to run ATLAS in
DECKBUILD. The VWF INTERACTIVE TOOLS USER’S MANUALS provides a more detailed description of
the features and capabilities of DECKBUILD.

2.3.1: Interactive Mode With DeckBuild

To start ATLAS in DECKBUILD, type:
deckbuild -as
at the UNIX system command prompt. The command line option, -as, instructs DECKBUILD to start
ATLAS as the default simulator.
If you want to start from an existing input file, start DECKBUILD by typing:
deckbuild -as <input filename>
The run-time output shows the execution of each ATLAS command and includes error messages,
warnings, extracted parameters, and other important output for evaluating each ATLAS run. When
ATLAS runs in this mode, the run-time output is sent to the output section of the DeckBuild Window
and can be saved as needed. Therefore, you don’t need to save the run-time output explicitly. The
following command line, however, specifies the name of a file that will be used for storing the run-time
output.
deckbuild -as <input filename> -outfile <output filename>
In this case, the run-time output is sent to the output file and to the output section of the DeckBuild
Window.

2.3.2: Batch Mode With DeckBuild

To use DECKBUILD in a non-interactive or batch mode, add the -run parameter to the command that
invokes DECKBUILD. A prepared command file is required for running in batch mode. We advise you to
save the run-time output to a file, since error messages in the run-time output would otherwise be lost
when the batch job completes. For example:
deckbuild -run -as <input filename> -outfile <output filename>
Using this command requires a local X-Windows system to be running. The job runs inside a
DECKBUILD icon on the terminal and quits automatically when the ATLAS simulation is complete. You
can also run DECKBUILD using a remote display. For example:
deckbuild -run -as <input file> -outfile <output file> -display<hostname>:0.0

2.3.3: No Windows Batch Mode With DeckBuild

For completely non-X Windows operation of DECKBUILD, use the -ascii parameter. For example:
deckbuild -run -ascii -as <input filename> -outfile <output filename>
This command directs DECKBUILD to run the ATLAS simulation without any display of the DeckBuild
Window or icon. This is useful for remote execution without an X Windows emulator or for replacing
UNIX-based ATLAS runs within framework programs.
When using batch mode, use the UNIX command suffix, &, to detach the job from the current command
shell. To run a remote ATLAS simulation under DECKBUILD without display and then logout from the
system, use the UNIX command, nohup , before the DECKBUILD command line. For example:
nohup deckbuild -run -ascii -as <input filename> -outfile <output filename> &

SILVACO International 2-3

ATLAS User’s Manual

2.3.4: Running ATLAS inside Deckbuild

Each ATLAS run inside DECKBUILD should start with the line:
go atlas
A single input file may contain several ATLAS runs each separated with a go atlas line. Input files
within DECKBUILD may also contain runs from other programs such as ATHENA or DEVEDIT along
with the ATLAS runs.

Running a given version number of ATLAS

The go statement can be modified to provide parameters for the ATLAS run. To run version 4.3.0.R,
the syntax is:
go atlas simflags=”-V 4.3.0.R”

Starting Parallel ATLAS

The -P option is used to set the number of processors to use in a parallel ATLAS run. If the number set
by -P is greater than the number of processors available or than the number of parallel thread
licenses, the number is automatically reduced to this cap number. To run on 4 processors, use:
go atlas simflags=”-V 4.3.2.C -P 4”

2.3.5: Batch Mode Without DeckBuild

You can run ATLAS outside the DECKBUILD environment. But this isn’t recommended by Silvaco. If
you don’t want the overhead of the DeckBuild Window, use the No Windows Mode. Many important
features such as variable substitution, automatic interfacing to process simulation, and parameter
extraction are not available outside the DECKBUILD environment. To run ATLAS directly under UNIX,
use the command:
atlas <input filename>
To save the run-time output to a file, don’t use the UNIX redirect command (>). Simply specify the
name of the output file. For example:
atlas <input filename> -logfile <output filename>

Note: The standard examples supplied with ATLAS will not run correctly outside of DECKBUILD.

2-4 SILVACO International

 Getting Started with ATLAS

2.4: Accessing The Examples

ATLAS has more than 300 standard examples that demonstrate how the program is used to simulate
different technologies. These examples are a good starting point for creating your own simulations.
The examples are accessed from the menu system in DECKBUILD. To select and load an example:
1. Start DECKBUILD with ATLAS as the simulator, which is described in the previous section.
2. Use left mouse button to pull down the Main Control menu.
3. Select Examples. An index will then appear in a Deckbuild Examples Window (see Figure 2-2).

Figure 2-2: Examples Index in DeckBuild

The examples are divided by technology or technology group. For instance, the most common
technologies are individually listed (e.g., MOS, BJT), while others are grouped with similar devices
(e.g., IGBT and LDMOS are under POWER, and solar cell and photodiode are under
OPTOELECTRONICS).
4. Choose the technology by double clicking the left mouse button over that item. A list of examples
for that technology will appear. These examples typically illustrate different devices, applications,
or types of simulation.
You can also search for an example by selecting the Index button. Wildcards can be used in the
search.

SILVACO International 2-5

ATLAS User’s Manual

5. Choose a particular example by double clicking the left mouse button over that item in the list. A
text description of the example will appear in the window. This text describes the important
physical mechanisms in the simulation, and details of the ATLAS syntax used. You should read
this information before proceeding.
6. Press the Load Example button. The input command file for the example will be copied into your
current working directory together with any associated files. A copy of the command file will be
loaded into DECKBUILD. Note that the Load Example button remains faded out until this step is
performed correctly.
7. Press the Run button in the middle frame of the DECKBUILD application window to run the
example. Alternatively, most examples are supplied with results that are copied into the current
working directory along with the input file. To view the results, select (highlight) the name of the
results file and select Tools-Plot. Details on using TONYPLOT can be found in the VWF
INTERACTIVE TOOLS USER’S MANUALS.

2-6 SILVACO International

 Getting Started with ATLAS

2.5: The ATLAS Syntax

An ATLAS command file is a list of commands for ATLAS to execute. This list is stored as an ASCII
text file that can be prepared in DECKBUILD or using any text editor. Preparing an input file in
DECKBUILD is preferred, which can be made easier by using the DeckBuild Commands menu in the
DeckBuild Window.

2.5.1: Statements and Parameters

The input file contains a sequence of statements. Each statement consists of a keyword that identifies
the statement and a set of parameters. The general format is:
<STATEMENT> <PARAMETER>=<VALUE>
With a few exceptions the input syntax is not case sensitive. One important exception is that
commands described in this manual as being executed by DECKBUILD rather than ATLAS are case
sensitive. These include: EXTRACT, SET, GO, and SYSTEM. Also, filenames for input and output under
UNIX are case sensitive.
For any <STATEMENT>, ATLAS may have four different types for the <VALUE> parameter. These are:
Real, Integer, Character, and Logical.
An example of a statement line is:
DOPING UNIFORM [Link] CONCENTRATION=1.0e16 REGION=1 OUTFILE=[Link]
The statement is DOPING. All other items are parameters of the DOPING statement. UNIFORM and
[Link] are logical parameters. Their presence on the line sets their values to true, otherwise they
take their default values (usually false). CONCENTRATION is a Real parameter and takes floating
point numbers as input values. REGION is an Integer parameter taking only integer numbers as input.
OUTFILE is a Character parameter type taking strings as input.
The statement keyword must come first but after this, the order of parameters within a statement is
not important.
You only need to use enough letters of any parameter to distinguish it from any other parameter on the
same statement. Thus, CONCENTRATION can be shortened to CONC. REGION, but can’t be shortened to
R, since there’s a parameter called RATIO associated with the DOPING statement.
Logical parameters can be explicitly set to false by preceding them with the ^ symbol.
Any line beginning with a # is ignored. These lines are used as comments.
ATLAS can read up to 256 characters on one line. But it is best to spread long input statements over
several lines to make the input file more readable. The \ character at the end of a line indicates
continuation.
For more information about statements and parameters in ATLAS, see Chapter 18: “Statements”.

2.5.2: The Order of ATLAS Commands

The order in which statements occur in an ATLAS input file is important. There are five groups of
statements that must occur in the correct order (see Figure 2-3). Failure to do so usually causes an
error message to appear, which could lead to incorrect operation or termination of the program. For
example, if the material parameters or models are set in the wrong order, then they may not be used in
the calculations.
The order of statements within the mesh definition, structural definition, and solution groups is also
important. Failure to place these statements in proper order may also cause incorrect operation or
termination of the program.

SILVACO International 2-7

ATLAS User’s Manual

Group Statements
MESH
1. Structure Specification REGION
ELECTRODE
DOPING

MATERIAL
2. Material Models Specification MODELS
CONTACT
INTERFACE

3. Numerical Method Selection METHOD

LOG
4. Solution Specification SOLVE
LOAD
SAVE

5. Results Analysis EXTRACT

TONYPLOT

Figure 2-3: ATLAS Command Groups with the Primary Statements in each Group

2.5.3: The DeckBuild Command Menu

The DeckBuild Command Menu (Command Menu) can help you to create input files. This menu is
found under the Commands button on DECKBUILD’s main screen. The Commands Menu is configured
for ATLAS whenever ATLAS is the currently active simulator in DECKBUILD. When ATLAS is active,
which is indicated in the lower bar of the DeckBuild Window, an ATLAS command prompt will appear
in the DECKBUILD output section.
The Command Menu gives you access to pop-up windows where you type information. When you select
the Write button, syntactically correct statements are written to the DECKBUILD text edit region. The
DeckBuild Command Menu does not support all possible ATLAS syntax, but aims to cover the most
commonly used commands.

2.5.4: PISCES-II Quick Start

This section is a quickstart for those who may be familiar with the syntax and use of the Stanford
University PISCES-II program or other device simulators derived from this program.
The major differences between ATLAS and PISCES-II are:
• all graphics are handled by a separate interactive graphics program, TONYPLOT. The PISCES-II
graphics commands: PLOT.1D, PLOT.2D, CONTOUR, VECTOR, and so on are not required. By using
TONYPLOT, it is no longer necessary to run the device simulator simply to plot or alter graphics.
• no need to separate individual ATLAS simulations into separate input files. Multiple runs of
ATLAS are possible in the same input file separated by the line go atlas. There’s also no need to
separate process and device simulation runs of Silvaco products into separate input files. A single
file containing ATHENA and ATLAS syntax is permitted in DECKBUILD.
• the interface from process to device simulation is handled though a single file format compatible
with other programs. The file read by ATLAS is the default output file format of ATHENA. No
special file format for the interface is required.

2-8 SILVACO International

 Getting Started with ATLAS

• when defining a grid structure within ATLAS, the NODE and LOCATION syntax to define exact grid
line numbers in X and Y is not recommended. A more reliable and easier to use syntax using
LOCATION and SPACING is available.
• using the REGRID command is not recommended due to the creation of obtuse triangles. A
standalone program, such as DEVEDIT, can be used as a grid pre-processor for ATLAS.
• all numerical method selection commands and parameters are on the METHOD statement. The
SYMBOLIC statement is not used. Historically, SYMBOLIC and METHOD were used as a coupled pair
of statements, but it is more convenient to use a single statement (METHOD) instead. Most of the old
parameters of the SYMBOLIC statement have the same meaning and names, despite this move to a
single statement. One notable change in ATLAS is that numerical methods can be combined
together.
See the “Pisces-II Compatibility” section on page 2-41 for more information concerning the
translation of PISCES-II numerics statements.
• various general purpose commands are actually part of the DECKBUILD user environment. These
include: SET, EXTRACT, GO, SYSTEM, and SOURCE. These commands can be interspersed inside
ATLAS syntax.
• Variable substitution is supported for both numerical and string variables using the SET statement
and $symbol. To avoid confusion, the # symbol is preferred over the $symbol for comment
statements.
In addition to these changes, the physical models are generally different in ATLAS. Most of the
original PISCES-II models have been preserved but often are not the default or the recommended
models to use. See the on-line examples for technology specific information about models.

SILVACO International 2-9

ATLAS User’s Manual

2.6: Defining A Structure

A device structure can be defined in three different ways for use in ATLAS.
The first way is to read an existing structure from a file. The structure is created either by an earlier
ATLAS run or another program such as ATHENA or DEVEDIT. A MESH statement loads in the mesh,
geometry, electrode positions, and doping of the structure. For example:
MESH INFILE=<filename>
The second way is to use the Automatic Interface feature from DECKBUILD to transfer the input
structure from ATHENA or DEVEDIT.
The third way is create a structure by using the ATLAS command language. See Chapter 18:
“Statements” for more information about the ATLAS syntax.

2.6.1: Interface From ATHENA

When ATHENA and ATLAS are run under DECKBUILD, you can take advantage of an automatic
interface between the two programs. Perform the following steps to load the complete mesh, geometry,
and doping from ATHENA to ATLAS.
1. Deposit and pattern electrode material in ATHENA
2. Use the ELECTRODE statement in ATHENA to define contact positions. Specify the x and y
coordinates as cross-hairs to pin-point a region. The whole region is then turned into electrode. In
many cases, only the x coordinate is needed. For example:
ELECTRODE NAME=gate X=1.3 [Y=-0.1])
There is a special case to specify a contact on the bottom of the structure. For example:
ELECTRODE NAME=substrate BACKSIDE
3. Save a structure file while ATHENA is still the active simulator. For example:
STRUCTURE OUTF=[Link]
4. Start ATLAS with the go atlas command written in the same input deck. This will automatically
load the most recent structure from ATHENA into ATLAS.
If you need to load the structure saved in step 4 into ATLAS without using the auto-interface
capability, use the MESH command. For example:
MESH INF=[Link]
ATLAS inherits the grid used most recently by ATHENA. With a careful choice of initial mesh or by
using the grid manipulation techniques in ATHENA, you can produce a final mesh from ATHENA
that will give good results in ATLAS. But, a grid that is appropriate for process simulation isn’t always
appropriate for device simulation. If the final ATHENA mesh is inappropriate for ATLAS, use either
DEVEDIT to re-mesh the structure or the REGRID command.

Note: There’s no need to specify a MESH command in ATLAS when using the Automatic Interface of Deckbuild.

2-10 SILVACO International

 Getting Started with ATLAS

2.6.2: Interface From DevEdit

A 2D or 3D structure created by DEVEDIT can be read into ATLAS using the following statement.
MESH INF=<structure filename>
This statement loads in the mesh, geometry, electrode positions, and doping of the structure. ATLAS
will automatically determine whether the mesh is 2D for S-PISCES or BLAZE, or 3D for DEVICE3D
or BLAZE3D.
If the structure coming from DEVEDIT were originally created by ATHENA, the electrodes should be
defined in ATHENA as described in the previous section. If the structure is created in DEVEDIT, the
electrode regions should be defined in the Region/Add menu of DEVEDIT.

2.6.3: Using The Command Language To Define A Structure

To define a device through the ATLAS command language, you must first define a mesh. This mesh or
grid covers the physical simulation domain. The mesh is defined by a series of horizontal and vertical
lines and the spacing between them. Then, regions within this mesh are allocated to different
materials as required to construct the device. For example, the specification of a MOS device requires
the specification of silicon and silicon dioxide regions. After the regions are defined, the location of
electrodes is specified. The final step is to specify the doping in each region.
When using the command language to define a structure, the information described in the following
four sub-sections must be specified in the order listed.

Specifying The Initial Mesh

The first statement must be:
MESH [Link]=<VALUE>
This is followed by a series of [Link] and [Link] statements.
[Link] LOCATION=<VALUE> SPACING=<VALUE>
.
[Link] LOCATION=<VALUE> SPACING=<VALUE>
.
The [Link] parameter value is used as a scaling factor for the mesh created by the [Link] and
[Link] statements. The default value is 1. Values greater than 1 will create a globally coarser mesh
for fast simulation. Values less than 1 will create a globally finer mesh for increased accuracy. The
[Link] and [Link] statements are used to specify the locations in microns of vertical and horizontal
lines, respectively, together with the vertical or horizontal spacing associated with that line. At least
two mesh lines must be specified for each direction. ATLAS automatically inserts any new lines
required to allow for gradual transitions in the spacing values between any adjacent lines. The [Link]
and [Link] statements must be listed in the order of increasing x and y. Both negative and positive
values of x and y are allowed.
Figure 2-4 illustrates how these statements work. On the left hand plot, note how the spacing of the
vertical lines varies from 1 µm at x=0 and x=10 µm to 0.5 µm at x=5 µm. On the right hand plot, note
how specifying the [Link] parameter to have a value of 0.5 has doubled the density of the mesh
in both the x- and y-directions.

SILVACO International 2-11

ATLAS User’s Manual

Figure 2-4: Non-uniform Mesh Creation using ATLAS Syntax

After an initial mesh has been defined, you can remove grid lines in specified regions. This is typically
done in regions of the device where a coarse grid is expected to be sufficient such as the substrate. The
removal of grid lines is accomplished using the ELIMINATE statement. The ELIMINATE statement
removes every second mesh line in the specified direction from within a specified rectangle. For
example, the statement:
ELIMINATE COLUMNS [Link]=0 [Link]=4 [Link]=0.0 [Link]=3
removes every second vertical grid line within the rectangle bounded by x=0, x=4, y=0 and y=3
microns.

Specifying Regions And Materials

Once the mesh is specified, every part of it must be assigned a material type. This is done with REGION
statements. For example:
REGION number=<integer> <material_type> <position parameters>
Region numbers must start at 1 and are increased for each subsequent region statement. You can have
up to 200 different regions in ATLAS. A large number of materials is available. If a composition-
dependent material type is defined, the x and y composition fractions can also be specified in the
REGION statement.
The position parameters are specified in microns using the [Link], [Link], [Link], and [Link]
parameters. If the position parameters of a new statement overlap those of a previous REGION
statement, the overlapped area is assigned as the material type of the new region. Make sure that
materials are assigned to all mesh points in the structure. If this isn’t done, error messages will appear
and ATLAS won’t run successfully.
You can use the MATERIAL statement to specify the material properties of the defined regions. But you
must complete the entire mesh and doping definition before any MATERIAL statements can be used.
The specification of material properties is described in Section 2.7.2: “Specifying Material Properties”.

2-12 SILVACO International

 Getting Started with ATLAS

Cylindrical Coordinates
Cylindrical coordinates are often used when simulating discrete power devices. In this mode, ATLAS
operates with x=0 as the axis of symmetry around which the cylindrical geometry is placed. Many of
the default units change when cylindrical coordinates are used. The calculated current is in Amps
rather than the usual Amps per micron. External elements are specified in absolute units (e.g., Farads,
not Farads/micron for capacitors).
The MESH statement must be used to specify cylindrical symmetry. The following statement creates a
mesh, which contains cylindrical symmetry.
MESH NX=20 NY=20 CYLINDRICAL
There are 20 mesh nodes along the x-axis and 20 mesh nodes along the y-axis.
The following statement imports a mesh, which contains cylindrical symmetry.
MESH INF=[Link] CYLINDRICAL

Note: The CYLINDRICAL parameter setting isn’t stored in mesh files. Therefore, this parameter must be specified each
time a mesh file, which contains cylindrical symmetry, is loaded.

Specifying Electrodes
Once you have specified the regions and materials, define at least one electrode that contacts a
semiconductor material. This is done with the ELECTRODE statement. For example:
ELECTRODE NAME=<electrode name> <position_parameters>
You can specify up to 50 electrodes. The position parameters are specified in microns using the [Link],
[Link], [Link], and [Link] parameters. Multiple electrode statements may have the same electrode
name. Nodes that are associated with the same electrode name are treated as being electrically
connected.
Some shortcuts can be used when defining the location of an electrode. If no y coordinate parameters
are specified, the electrode is assumed to be located on the top of the structure. You also can use the
RIGHT, LEFT, TOP, and BOTTOM parameters to define the location. For example:
ELECTRODE NAME=SOURCE LEFT LENGTH=0.5
specifies the source electrode starts at the top left corner of the structure and extends to the right for
the distance LENGTH.

Specifying Doping
You can specify analytical doping distributions, or have ATLAS read in profiles that come from either
process simulation or experiment. You specify the doping using the DOPING statement. For example:
DOPING <distribution_type> <dopant_type> <position_parameters>

SILVACO International 2-13

ATLAS User’s Manual

Analytical Doping Profiles

Analytical doping profiles can have uniform or Gaussian forms. The parameters defining the
analytical distribution are specified in the DOPING statement. Two examples are shown below with
their combined effect shown in Figure 2-5.
DOPING UNIFORM CONCENTRATION=1E16 [Link] REGION=1

DOPING GAUSSIAN CONCENTRATION=1E18 CHARACTERISTIC=0.05 [Link] \

[Link]=0.0 [Link]=1.0 PEAK=0.1

The first DOPING statement specifies a uniform n-type doping density of 1016 cm-3 in the region that
was previously labelled as region #1. The position parameters: [Link], [Link], [Link], and [Link] can
be used instead of a region number.

The second DOPING statement specifies a p-type Gaussian profile with a peak concentration of 1018
cm-3. This statement specifies that the peak doping is located along a line from x = 0 to x = 1 microns.
Perpendicular to the peak line, the doping drops off according to a Gaussian distribution with a
standard deviation of 0.05 mm. At x < 0 or x > 1, the doping drops off laterally with a default standard
deviation that is 70% of CHARACTERISTIC. This lateral roll-off can be altered with the
[Link] parameter. If a Gaussian profile is being added to an area that was already defined
with the opposite dopant type, you can use the JUNCTION parameter to specify the position of the
junction depth instead of specifying the standard deviation using the CHARACTERISTIC parameter.

Figure 2-5: Analytical specification of a 2D Profile

2-14 SILVACO International

 Getting Started with ATLAS

Importing 1D SSUPREM3 Doping Profiles

One-dimensional doping profiles can be read into ATLAS from a SSUPREM3 output file. The doping
data must have been saved from SSUPREM3 using the statement:
STRUCTURE OUTFILE=<output filename>
at the end of the SSUPREM3 run.
In ATLAS, the MASTER parameter of the DOPING statement specifies that a SSUPREM3 file will be
read by ATLAS. Since this file will usually contain all the dopants from the SSUPREM3 simulation,
the desired dopant type must also be specified. For example, the statement:
DOPING MASTER INFILE=[Link] BORON REGION=1
specifies that the boron profile from the file [Link] should be imported and used in region #1.
SSUPREM3 profiles are imported into ATLAS one at a time (i.e., one DOPING statement is used for
each profile or dopant). The statements:
DOPING MASTER INFILE=[Link] BORON OUTFILE=[Link]
DOPING MASTER INFILE=[Link] ARSENIC [Link]=0.8 RATIO=0.75
DOPING MASTER INFILE=[Link] ARSENIC [Link]=2.2 RATIO=0.75
offset the arsenic doping from boron to create a 2-D doping profile from a single SSUPREM3 result.
It is advisable to include the OUTFILE parameter on the first DOPING statement to create a 2-D doping
file. This file will then be used in the next section to interpolate doping on a refined mesh after a
REGRID. This file, however, can’t be plotted in TONYPLOT. The position parameters and the
[Link] parameter are used in the same manner as for analytical doping profiles to set the
extent of the 1-D profile.

2.6.4: Automatic Meshing (Auto-meshing) Using The Command Language

Automatic meshing provides a simpler method for defining device structures and meshs than the
standard method described in Section 2.6.3: “Using The Command Language To Define A Structure”.
Auto-meshing is particularly suited for epitaxial structures, especially device structures with many
layers (for example, a VCSEL device). Auto-meshing unburdens you from the delicate bookkeeping
involved in ensuring that the locations of mesh lines in the Y direction are consistently aligned with
the edges of regions. This is done by specifying the locations of Y mesh lines in the REGION statements.
The following sections will show how auto-meshing is done, using a few simple examples.

Specifying The Mesh And Regions

In the first example, we use a simple device to show you the fundimental concepts of auto-meshing.
The first statements in this example are as follows:
MESH AUTO
[Link] LOCATION=-1.0 SPACING=0.1
[Link] LOCATION=1.0 SPACING=0.1
These statements are similar to the ones used in the standard method that described a mesh using the
command language as shown in Section 2.6.3: “Using The Command Language To Define A
Structure”. There are, however, two key differences. The first difference is the inclusion of the AUTO
parameter in the MESH statement. You need this parameter to indicate that you want to use auto-
meshing. The second and more important difference is that in this example we will not specify any
[Link] statements. This is because the locations of Y mesh lines will be automatically determined by
the parameters of the REGION statements.
You can still specify one or more [Link] statements. Such defined mesh lines will be included in the
mesh. But including [Link] statements is optional in auto-meshing.

SILVACO International 2-15

ATLAS User’s Manual

In the next few statements of the example, we will show you several new concepts that will explain
how auto-meshing works. The following four lines describe the regions in the example device:
REGION TOP THICKNESS=0.02 MATERIAL=GaN NY=5 DONOR=1E16
REGION BOTTOM THICKNESS=0.1 MATERIAL=AlGaN NY=5 DONOR=1E17 [Link]=0.2
REGION TOP THICKNESS=0.08 MATERIAL=AlGaN NY=4 ACCEPTOR=1E17 [Link]=0.2
REGION BOTTOM THICKNESS=0.5 MATERIAL=AlGaN NY=10 DONOR=1E18 [Link]=0.2

New Concepts
First, it appears that composition and doping are being specified in the REGION statement. This is the
case for the DONOR, ACCEPTOR, [Link] and [Link] parameters in the REGION
statement that specify uniform doping or composition or both over the specified region. These
parameters are also available to the standard methods described in Section 2.6.3: “Using The
Command Language To Define A Structure”, but are more amenable to specification of epitaxial
structures such as we are describing in this example.
Next, you should notice several other new parameters. These are the TOP, BOTTOM, THICKNESS and NY
parameters. All of these are used to describe the relative locations and thicknesses of the layers as well
as the locations of the Y mesh lines. The most intuitive of these parameters is the THICKNESS
parameter, which describes the thickness in microns, in the Y direction of each layer. As for the extent
in the X direction, in the absence of any specified [Link] or [Link] parameters, it is assumed that the
region extends over the full range of the X mesh described above in the [Link] statements.
The NY parameter describes how many Y mesh lines are contained in the region so that the Y mesh
lines are evenly spaced over the region. You can use the SY parameter instead of NY to specify the
spacing in microns between Y mesh lines in the region. Make sure the value of SY does not exceed the
value of THICKNESS. Note that in general, the relationship between SY, NY and THICKNESS can be
expressed by the following:
SY = THICKNESS/NY
Figure 2-6 shows the meanings of the TOP and BOTTOM parameters.

2-16 SILVACO International

 Getting Started with ATLAS

a) Structure after 1st REGION Statement b) Structure after 2nd REGION Statement

c) Structure after 3rd REGION Statement d) Structure after 4th REGION Statement

Figure 2-6: Simple Example of Auto-meshing Showing Sequence of Structure Development.

This figure shows a progression of representations of how the structure’s mesh and region outlines
appear after each REGION statement. This figure should give you an intuitive feel of how the regions
are alternately placed on the top or bottom of the structure according to the specification of the TOP or
BOTTOM parameters. It is important to keep in mind that the ATLAS coordinate convention for the Y
axis is that positive Y is directed down into the device. This is similar to using the TOP and BOTTOM
parameters of the ELECTRODE statement.
One thing you might notice in this figure is that the number of Y mesh lines in each region does not
always match the number specified. This is because at each interface between regions, the Y mesh line
spacing is ambiguously defined and the auto-meshing algorithm will always pick the smaller spacing
between the two at each interface. Then, the spacing between Y mesh lines varies continuously
between subsequent interfaces in a similar way as it does for mesh spacings specified by the LOCATION
and SPACING parameters in the [Link] and [Link] statements.
The auto-meshing algorithm maintains the "notion" of the current Y locations of the "top" and
"bottom". Let’s call these locations "Ytop" and "Ybottom".

SILVACO International 2-17

ATLAS User’s Manual

Before any REGION statements are processed, "Ytop" and "Ybottom" are both defined to be equal to
zero. As the REGION statements are processed, the following cases are addressed:
• If you place the region on the top, as specified by the TOP parameter, the region will extend from
"Ytop" to "Ytop"-THICKNESS (remember positive Y points down) and "Ytop" will move to the
new location "Ytop"-THICKNESS.
• If you place the region on the bottom, as specified by the BOTTOM parameter, the region will extend
from "Ybottom" to "Ybottom"+THICKNESS and "Ybottom" will move to the new location
"Ybottom"+THICKNESS.
The auto-meshing algorithm will ensure that all regions are perfectly aligned to their neighboring
regions and there are no inconsistencies between the locations of Y mesh lines and the region edges
that they resolve.

Non-uniformity In The X Direction and Auto-meshing

In some cases, you may want to define a device with material discontinuities in the X direction. Such
discontinuities may represent etched mesa structures or oxide aperatures. There are a couple of ways
to do this in auto-meshing. Let’s say for example, you want to etch out a region from the structure of
the previous example, from X=0 to the right and from Y=0 to the top. You can add the statement
REGION MATERIAL=Air [Link]=0.0 [Link]=0.0
In this statement, we are not using the auto-meshing features but are using syntax of the standard
meshing methods described in Section 2.6.3: “Using The Command Language To Define A Structure”.
ATLAS supports mixing the standard syntax with auto-meshing but be careful when doing this as we
will explain shortly. Figure 2-7 shows the resulting structure and mesh after adding this statement.
In this figure, we see that we have obtained the desired result.

Figure 2-7: Structure Etch Example in Auto-meshing

2-18 SILVACO International

 Getting Started with ATLAS

The potential pitfall in using absolute Y coordinates in auto-meshing is that the location you choose,
for example, by summing up thicknesses may not match the location your computer has calculated
with its inherent numerical imprecision. The result is not only that the resulting structure may not
exactly match what you desire, but more importantly, you may end up accidently creating a mesh with
Y mesh lines closely spaced (usually in the order of the machine precision). This can cause numerical
instability (poor convergence) and can overflow or underflow in certain models. What’s worse is that
this situation is difficult to detect.
There is, however, one situation where we can absolutely predict the location of an edge or Y mesh line.
That is at Y=0. This is exactly what we have done in our example. So if you want to use auto-meshing
with specifications of an absolute value of Y, then arrange your device structure so that the
specification of Y will be at zero.
This also applies to the location of a recessed electrode. Make sure it is located at Y=0 if you are using
auto-meshing.
There is another method of providing for discontinuities in material in the X direction that is
absolutely safe from the previously discussed problems. In this approach, we use another parameter
called STAY in the REGION statement in conjunction with the TOP or BOTTOM parameters.
The following describes the effect of the STAY parameter in the REGION statement.
• If you place the region on the top, as specified by the TOP parameter, and the STAY parameter is
specified, the region will extend from "Ytop" to "Ytop"-THICKNESS and "Ytop" will remain in
its current position.
• If you place the region on the bottom, as specified by the BOTTOM parameter, and the STAY
parameter is specified, the region will extend from "Ybottom" to "Ybottom"+THICKNESS and
"Ybottom" will remain in its current position.
The use of the STAY parameter can best be illustrated by the following example. In this example, we
will reproduce the same structure discussed in the last example but this time using only STAY
parameters and by not using any specification of absolute Y coordinates. The new REGION
specifications are as follows:
REGION BOTTOM THICKNESS=0.1 MATERIAL=AlGaN NY=5 DONOR=1E17 [Link]=0.2
REGION BOTTOM THICKNESS=0.5 MATERIAL=AlGaN NY=10 DONOR=1E18 [Link]=0.2
REGION TOP STAY THICKNESS=0.02 MATERIAL=GaN NY=5 DONOR=1E16
REGION TOP THICKNESS=0.02 MATERIAL=Air NY=5 [Link]=0.0
REGION TOP STAY THICKNESS=0.08 MATERIAL=AlGaN NY=4 ACCEPTOR=1E17 [Link]=0.2
REGION TOP THICKNESS=0.08 MATERIAL=Air NY=4 [Link]=0.0
In this example, we slightly rearranged the REGION statements for clarity and split one region into
two. Figure 2-8 shows the results.

SILVACO International 2-19

ATLAS User’s Manual

Figure 2-8: Sturcture Etching Example Using The STAY Parameter

The following describes the operation of the STAY parameter in this example:
• The STAY parameter in the third REGION statement means that the fourth REGION will start at the
same Y coordinate as the third.
• Since the THICKNESS and NY parameters are the same in the third and fourth regions, they will
have the same range of Y values and the same Y mesh.
• The specification of [Link] in the fourth REGION statement means that the region will not
completely overlap the third region but will only overlap to a minimum X value of 0.
• The lack of a STAY parameter in the fourth REGION statement means that the fifth region will lie
directly atop the third and fourth regions.
• The STAY parameter in the fifth REGION statement means that the sixth REGION will start at the
same Y coordinate as the fifth.
• Since the THICKNESS and NY parameters are the same in the fifth and sixth regions, they will have
the same range of Y values and the same Y mesh.
• The specification of [Link] in the sixth REGION statement means that the region will not
completely overlap the fifth region but will only overlap to a minimum X value of 0.
As you can see, using the STAY parameter carefully avoids the problems of specifying values of Y
coordinates and the potential problems involved. By doing so, you can specify arbitrary stepped
structures.

2-20 SILVACO International

 Getting Started with ATLAS

Grading of Compositional and Doping

We have provided another method for specifying composition or doping or both in the REGION
statement that is available for both auto-meshing and meshing using the standard method described
in Section 2.6.3: “Using The Command Language To Define A Structure”. This method allows linear
grading of rectangular regions. Eight new parameters of the REGION statement support this function.
They are [Link], [Link], [Link], [Link], [Link], [Link], [Link], and
[Link]. In the syntax of these parameter names, "TOP" refers to the "top" or extreme Y
coordinate in the negative Y direction, and "BOTTOM" refers to the "bottom" or extreme Y coordinate in
the positive Y direction. With this in mind, the following rules apply:
• If you specify [Link] and [Link] in a REGION statement, the donor doping in the region will
vary linearly from the value specified by [Link] at the "top" of the device to the value specified by
[Link] at the "bottom" of the device.
• If you specify [Link] and [Link] in a REGION statement, the acceptor doping in the region
will vary linearly from the value specified by [Link] at the "top" of the device to the value
specified by [Link] at the "bottom" of the device.
• If you specify [Link] and [Link] in a REGION statement, the X composition fraction
in the region will vary linearly from the value specified by [Link] at the "top" of the device to
the value specified by [Link] at the "bottom" of the device.
• If you specify [Link] and [Link] in a REGION statement, the Y composition fraction
in the region will vary linearly from the value specified by [Link] at the "top" of the device to
the value specified by [Link] at the "bottom" of the device.

Note: Any subsequent DOPING statements will add to the doping specified by the doping related parameters on the
REGION statement.

Superlattices and Distributed Bragg Reflectors DBRs

For auto-meshing, we have provided a convenient way of specifying a certain class of superlattices or
as they are commonly used distributed Bragg reflectors (DBRs). This class of superlattice includes any
superlattice that can be described as integer numbers of repetitions of two different layers. By
different, we mean that the two layers may have different thicknesses, material compositions, or
dopings, or all. These "conglomerate" structures are specified by the DBR or SUPERLATTICE
statement. Actually, SUPERLATTICE is a synonym for DBR. Therefore, in the following discussion we
will use the DBR syntax and recognize that SUPERLATTICE and DBR can be used interchangeably.
Most of the syntax of the DBR statement is similar to the syntax of the REGION statement, except the
syntax is set up to describe two regions (or two sets of regions). As you will see, we differentiate these
regions (or sets of regions) by the indices 1 and 2 in the parameter’s name.
The following example should make this concept clear.
MESH AUTO
[Link] LOCATION=-1.0 SPACING=0.1
[Link] LOCATION=1.0 SPACING=0.1

DBR TOP LAYERS=4 THICK1=0.1 THICK2=0.2 N1=2 N2=3 MAT1=GaAs MAT2=AlGaAs \

[Link]=0.4
DBR BOTTOM LAYERS=3 THICK1=0.05 THICK2=0.075 N1=3 N2=3 MAT1=AlGaAs \
MAT2=GaAs [Link]=0.4

SILVACO International 2-21

ATLAS User’s Manual

In this example, we can see we are using auto-meshing because of the appearance of the AUTO
parameter in the MESH statement. The DBR statements should be familiar since the functionality and
syntax are similar to the REGION statement. We will not discuss those similarities here. We will only
discuss the important differences. For more information about the similarities, see Chapter 18:
“Statements” .
The main new parameter is the LAYERS parameter. This parameter specifies the total number of
regions added. The region indexed "1" is always added first, then the region indexed "2" is added,
and depending on the value of LAYERS, the sequence of regions continues 1, 2, 1, 2, 1, ...
Figure 2-9 shows the functionality of the DBR statement, which gives the resulting structure and mesh
from the example.

Figure 2-9: Supperlattice Example Structure

2.6.5: Modifying Imported Regions

If you import a device structure from a file using the INFILE parameter of the MESH statement, you
may want to modify some of the characteristics of one or more of the regions in the structure. To do
this, specify a REGION statement with the MODIFY parameter and the NUMBER parameter assigned to
the region number of interest. You can specify/respecify any of the following REGION statement
parameters: DONORS, ACCEPTORS, STRAIN, [Link], [Link], [Link], [Link],
[Link], [Link], [Link], [Link], [Link], [Link], [Link],
[Link], QWELL, LED, [Link], and POLARIZATION.

2-22 SILVACO International

 Getting Started with ATLAS

2.6.6: Remeshing Using The Command Language

It can be difficult to define a suitable grid when specifying a structure with the command language.
The main problem is that the meshes required to resolve 2-D doping profiles and curved junctions are
quite complicated and simple rectangular meshes require an excessive number of nodes to resolve such
profiles. If a device structure only includes regions of uniform doping, then there’s usually no need to
regrid. But when realistic 2-D doping profiles are present, a regrid may be necessary.

Note: The recommended solution for defining complex mesh structures for ATLAS is to use the standalone program,
DEVEDIT.

Regrid On Doping
ATLAS includes a regridding capability that generates a fine mesh only in a localized region. You
specify a quantity on which the regrid is to be performed. The mesh is then refined in regions where
the specified quantity varies rapidly. Whenever a specified quantity (usually doping) changes quickly,
the regridding will automatically grade the mesh accordingly. You can get a regrid on doping before
any solutions are obtained. You can do this by using the statement:
REGRID LOGARITHM DOPING RATIO=2 [Link]=4 DOPFILE=<filename1> \
OUTFILE=<filename2>
This statement must be used after the MESH, REGION, MATERIAL, ELECTRODE, and DOPING statements
described previously. The effects of this REGRID statement on a simple diode structure are shown in
Figure 2-10.

Figure 2-10: Regrid on doping provides improved resolution of junction

SILVACO International 2-23

ATLAS User’s Manual

In this statement, the regridding will resolve doping profiles to two orders of magnitude in change.
The doping file, , must be specified in the first DOPING statement with the OUTFILE
parameter. The results of the regrid are saved in the file, . The [Link] parameter value
selects a smoothing algorithm. A value of 4 is typically best as this algorithm tends to produce the
fewest obtuse triangles. For a complete description of the various smoothing algorithms, see Chapter
17: “Numerical Techniques”.

Regrid Using Solution Variables

The REGRID statement can use a wide range of solution variables as the basis for mesh refinement.
Note that regrid on solution variables can only be used after a solution has already been obtained.
After a regrid on a solution variable, the solution must be re-solved at the same bias in ATLAS.
For example:
REGRID POTENTIAL RATIO=0.2 [Link]=1 SMOOTH.K=4 DOPFILE=<filename1>
SOLVE PREV
Regrid on potential is often used for high voltage power devices.

Note: The REGRID statement can be used any number of times on a structure. But we advise you to quit and restart
ATLAS between regrids on electrical quantities. You can use the go atlas statement to do this. This should be followed
by a MESH statement loading the output file of the REGRID command and a re-setting of all material and model parameters.

2.6.7: Specifying 3D Structures

The syntax for forming 3-D device structures is an extension of the 2-D syntax described in the
previous section. The MESH statement should appear as:
MESH THREE.D
The THREE.D parameter tells ATLAS that a three dimensional grid will be specified. The other
statements used to specify 3-D structures and grids are the same as for 2-D with the addition of z
direction specifications. The statements:
MESH THREE.D [Link] LOCATION=0 SPACING=0.15
[Link] LOCATION=3 SPACING=0.15
[Link] LOCATION=0 SPACING=0.01
[Link] LOCATION=0.4 SPACING=0.01
[Link] LOCATION=3 SPACING=0.5
[Link] LOCATION=0 SPACING=0.1
[Link] LOCATION=3 SPACING=0.1
define a 3-D mesh that is uniform in the x and z directions and varies in the y direction.
Position parameters for the z direction ([Link] and [Link]) are also used on REGION, ELECTRODE, or
DOPING statements.

2-24 SILVACO International

 Getting Started with ATLAS

2.6.8: General Comments Regarding Grids

Specifying a good grid is a crucial issue in device simulation but there is a trade-off between the
requirements of accuracy and numerical efficiency. Accuracy requires a fine grid that resolves the
structure in solutions. Numerical efficiency is greater when fewer grid points are used. The critical
areas to resolve are difficult to generalize because they depend on the technology and the transport
phenomena. The only generalization possible is that most critical areas tend to coincide with reverse-
biased metallurgical junctions. Typical critical areas are:
• High electric fields at the drain/channel junction in MOSFETs
• The transverse electric field beneath the MOSFET gate
• Recombination effects around the emitter/base junction in BJTs
• Areas of high impact ionization
• Around heterojunctions in HBT’s and HEMTs.

The CPU time required to obtain a solution is typically proportional to Nα, where N is the number of
nodes and α varies from 2 to 3 depending on the complexity of the problem. Thus, the most efficient
way is to allocate a fine grid only in critical areas, and a coarser grid elsewhere.
The three most important factors to look for in any grid are:
• Ensure adequate mesh density in high field areas
• Avoid obtuse triangles in the current path or high field areas
• Avoid abrupt discontinuities in mesh density
For more information about grids, see Chapter 17: “Numerical Techniques”, Section 17.3: “Meshes”.

SILVACO International 2-25

ATLAS User’s Manual

2.6.9: Maximum Numbers Of Nodes, Regions, and Electrodes

ATLAS sets some limits on the maximum number of grid nodes that can be used. But this shouldn’t be
viewed as a bottleneck to achieving simulation results. In the default version, 2-D ATLAS simulations
have a maximum node limit of 20,000, and 3-D ATLAS simulations have an upper limit of 200,000
nodes with no more than 20,000 in any one plane.
This limit is high enough that for almost all simulations of conventional devices, running out of nodes
is never an issue. For most 2-D simulations, accurate results can be obtained with somewhere between
2,000 and 4,000 node points properly located in the structure.
If the node limits are exceeded, error messages will appear and ATLAS will not run successfully. There
are two options to deal with this problem. The first option is to decrease the mesh density, because
simulations with the maximum nodes take an extremely long time to complete. The second option is to
contact your local Silvaco office (support@[Link]). Versions of 2-D ATLAS with up to 100,000
nodes have been distributed.
A node point limitation below these values might be seen due to virtual memory constraints on your
hardware. For each simulation, ATLAS dynamically allocates the virtual memory. See the Silvaco
INSTALLATION GUIDE for information about virtual memory requirements. The virtual memory used by
the program depends not only on the number of nodes but also on such items as the models used and
the number of equations that are solved.
Also, there is a node limit for the number of nodes in the X or Y directions in 2-D and 3-D ATLAS. In
the standard version, this limit is 5,000 nodes. This is applicable to meshes defined in the ATLAS
syntax using [Link] and [Link] statements.
The maximum number of regions defined in both 2-D and 3-D ATLAS is 200. The maximum number of
definable electrodes is 50. Again, if it is deemed necessary to include more than the maximum number
of regions or electrodes, contact your local Silvaco office (support@[Link]).

2-26 SILVACO International

 Getting Started with ATLAS

2.7: Defining Material Parameters And Models

Once the mesh, geometry, and doping profiles are defined, you can modify the characteristics of
electrodes, change the default material parameters, and choose which physical models ATLAS will use
during the device simulation. These actions are accomplished using the CONTACT, MATERIAL, and
MODELS statements respectively. Impact ionization models can be enabled using the IMPACT
statement. Interface properties are set by using the INTERFACE statement.
Many parameters are accessible through the SILVACO C-INTERPRETER (SCI), which is further described
in Appendix A: “C-Interpreter Functions”. This allows you to define customized equations for some
models. For more information about the INTERFACE and MODELS statements, see Chapter 18:
“Statements”, Sections 18.18: “INTERFACE” and 18.29: “MODELS”.

2.7.1: Specifying Contact Characteristics

Workfunction for Gates or Schottky Contacts
An electrode in contact with semiconductor material is assumed by default to be ohmic. If a work
function is defined, the electrode is treated as a Schottky contact. The CONTACT statement is used to
specify the metal workfunction of one or more electrodes. The NAME parameter is used to identify
which electrode will have its properties modified.
The WORKFUNCTION parameter sets the workfunction of the electrode. For example, the statement:
CONTACT NAME=gate WORKFUNCTION=4.8
sets the workfunction of the electrode named gate to 4.8eV. The workfunctions of several commonly
used contact materials may be specified using the name of the material. Workfunctions for ALUMINUM,
[Link], [Link], TUNGSTEN, and [Link] can be specified in this way.
The following statement sets the workfunction for a n-type polysilicon gate contact.
CONTACT NAME=gate [Link]
Aluminum contacts on heavily doped silicon is usually ohmic. For this situation, don’t specify a
workfunction. For example, MOS devices don’t specify:
CONTACT NAME=drain ALUMINUM /* wrong */
The CONTACT statement can also be used to specify barrier and dipole lowering of the Schottky barrier
height. Barrier lowering is enabled by specifying the BARRIER parameter, while dipole lowering is
specified using the ALPHA parameter. For example, the statement:
CONTACT NAME=anode WORKFUNCTION=4.9 BARRIER ALPHA=1.0e-7‘
sets the work function of the Schottky contact named anode to 4.9eV enables barrier lowering and sets
the dipole lowering coefficient to 1 nm.

Note: When a Schottky barrier is defined at a contact, we recommend that a fine y mesh is present just beneath the contact,
inside the semiconductor. This allows the Schottky depletion region to be accurately simulated.

SILVACO International 2-27

ATLAS User’s Manual

Setting Current Boundary Conditions

The CONTACT statement is also used to change an electrode from voltage control to current control.
Current controlled electrodes are useful when simulating devices, where the current is highly
sensitive to voltage or is a multi-valued function of voltage (e.g., post-breakdown and when there is
snap-back).
The statement:
CONTACT NAME=drain CURRENT
changes the drain electrode to current control. The BLOCK or NEWTON solution methods are required for
all simulations using a current boundary condition. For more information about these methods, see
Chapter 17: “Numerical Techniques”, Section 17.5: “Non-Linear Iteration”.

Defining External Resistors, Capacitors, or Inductors

Lumped resistance, capacitance, and inductance connected to an electrode can be specified using the
RESISTANCE, CAPACITANCE, and INDUCTANCE parameters in the CONTACT statement. For example,
the statement:
CONTACT NAME=drain RESISTANCE=50.0 CAPACITANCE=20e-12 INDUCTANCE=1e-6
specifies a parallel resistor and capacitor of 50 ohms and 20 pF respectively in series with a 1 µH
inductor. Note that in 2D simulations, these passive element values are scaled by the width in the
third dimension. Since in 2D ATLAS assumes a 1µm width, the resistance becomes 50 Ω-µm.
Distributed contact resistance for an electrode can be specified using the [Link] parameter. For
example, the statement:
CONTACT NAME=source [Link]=0.01

specifies that the source contact has a distributed resistance of 0.01 Ωcm2.

Note: Simulations with external resistors, capacitors, or inductors must be solved using the BLOCK or NEWTON solution
method.

Floating Contacts
The CONTACT statement is also used to define a floating electrode. There are two distinctly different
situations where floating electrodes are important. The first situation is for floating gate electrodes
used in EEPROM and other programmable devices. The second situation is for contacts directly onto
semiconductor materials such as floating field plates in power devices.
Floating gates are enabled by specifying the parameter FLOATING on the CONTACT statement. For
example, the statement:
CONTACT NAME=fgate FLOATING

specifies that the electrode named fgate will be floating and that charge boundary conditions will
apply.
For contacts directly onto semiconductor, the FLOATING parameter cannot be

2-28 SILVACO International

 Getting Started with ATLAS

specifies current boundary conditions for the electrode named drain. On subsequent SOLVE
statements, the drain current boundary condition will default to zero current, therefore floating the
contact.
You can also make a floating contact to a semiconductor using a very large resistor attached to the
contact instead. For example:
CONTACT NAME=drain RESIST=1e20
Note that extremely large resistance values must be used to keep the current through the contact
insignificant bearing in mind the tolerance on potential will allow the contact voltage to move slightly
above zero. For example, if the tolerance is 10-5V and the defined resistance was only 10MΩ.µm, then a
current of 10-12 A/µm may flow through the contact, which is probably significant in breakdown
simulations.

Shorting Two Contacts Together

It is possible in ATLAS to tie two or more contact together so that voltages on both contacts are equal.
This is useful for many technologies for example dual base bipolar transistors. There are several
methods for achieving this depending on how the structure was initially defined.
If the structure is defined using ATLAS syntax, you can have multiple ELECTRODE statements with
the same NAME parameter defining separate locations within the device structure. In this case, the
areas defined to be electrodes will be considered as having the same applied voltage. A single current
will appear combining the current through both ELECTRODE areas.
Also, if two separate metal regions in ATHENA are defined using the ATHENA ELECTRODE statement
to have the same name, then in ATLAS these two electrodes will be considered as shorted together.
If the electrodes are defined with different names the following syntax can be used to link the voltages
applied to the two electrodes.
CONTACT NAME=base1 COMMON=base
.
SOLVE VBASE=0.1

Here, the electrode, base1, will be linked to the electrode, base The applied 0.1V on base will then
appear on base1 ATLAS, however, will calculate and store separate currents for both base and
base1. This can be a useful feature but in some cases, such as where functions of the currents are
required in EXTRACT or TONYPLOT, it is undesirable. You can add the SHORT parameter to the
CONTACT statement above to specify so that only a single base current will appear combining the
currents from base and base1.
When loading a structure from ATHENA or DEVEDIT where two defined electrode regions are
touching, ATLAS will automatically short these and use the electrode name that was defined first.

SILVACO International 2-29

ATLAS User’s Manual

Making an Open Circuit Contact

It is often required to perform a simulation with an open circuit on one of the defined electrodes. There
are three different methods to make an open circuit contact. The first method is to entirely deleting an
electrode from the structure file. The second method is to add an extremely large lumped resistance.
For example, 1020Ω onto the contact to be made open circuit. The third method is to switch the
boundary conditions on the contact to be made open circuit from voltage controlled to current
controlled and then specifying a very small or zero current through that electrode.
Each of these methods are feasible but if a floating region is created within the structure, then
numerical convergence may be affected. As a result, we normally recommend that you use the second
method because it ensures better convergence.

2.7.2: Specifying Material Properties

Semiconductor, Insulator, or Conductor
All materials are split into three classes: semiconductors, insulators and conductors. Each class
requires a different set of parameters to be specified. For semiconductors, these properties include
electron affinity, band gap, density of states and saturation velocities. There are default parameters
for material properties used in device simulation for many materials.
Appendix B: “Material Systems” lists default material parameters and includes a discussion on the
differences between specifying parameters for semiconductors, insulators, and conductors.

Setting Parameters
The MATERIAL statement allows you to specify your own values for these basic parameters. Your
values can apply to a specified material or a specified region. For example, the statement:
MATERIAL MATERIAL=Silicon EG300=1.12 MUN=1100
sets the band gap and low field electron mobility in all silicon regions in the device. If the material
properties are defined by region, the region is specified using the REGION or NAME parameters in the
MATERIAL statement. For example, the statement:
MATERIAL REGION=2 TAUN0=2e-7 TAUP0=1e-5
sets the electron and hole Shockley-Read-Hall recombination lifetimes for region number two (see
Chapter 3: “Physics”, the “Shockley-Read-Hall (SRH) Recombination” section on page 3-64 for more
information about this type of recombination). If the name, base, has been defined using the NAME
parameter in the REGION statement, then the statement:
MATERIAL NAME=base NC300=3e19
sets the conduction band density of states at 300 K for the region named base.
The description of the MATERIAL statement in Chapter 18: “Statements”, Section 18.24: “MATERIAL”
provides a complete list of all the material parameters that are available.

2-30 SILVACO International

 Getting Started with ATLAS

Heterojunction Materials
The material properties of heterojunctions can also be modified with the MATERIAL statement. In
addition to the regular material parameters, composition dependent material parameters can be
defined. These include composition dependent band parameters, dielectric constants, saturation
velocities, and so on.
For heterojunction material systems, the bandgap difference between the materials is divided between
conduction and valence bands. The ALIGN parameter specifies the fraction of this difference that is
applied to the conduction band edge. This determines the electron and hole barrier height and
overrides any electron affinity specification. For example, the statement:
MATERIAL MATERIAL=InGaAs ALIGN=0.36
MATERIAL MATERIAL=InP ALIGN=0.36

specifies that 36% of the band gap difference between InGaAs and InP is applied to the conduction
band and 64% is applied to the valence band. For example, if the band gap difference ( ∆ Eg) for this
material system is 0.6 eV, then the conduction band barrier height is 0.216 eV and the valence band
barrier height is 0.384 eV.
For heterojunction devices, the transport models may be different for each material. These models and
their coefficients may be specified for each material using the MODEL statement. See Section 2.7.4:
“Specifying Physical Models” for a description of this option.

2.7.3: Specifying Interface Properties

The INTERFACE statement is used to define the interface charge density and surface recombination
velocity at interfaces between semiconductors and insulators. For example, the statement:
INTERFACE QF=3e10

specifies that all interfaces between semiconductors and insulators have a fixed charge of 3.1010cm-2.
In many cases, the interface of interest is restricted to a specific region. This can be accomplished with
the [Link], [Link], [Link], and [Link] parameters on the INTERFACE statement. These parameters
define a rectangle, where the interface properties apply. For example, the statement:
INTERFACE QF=3e10 [Link]=1.0 [Link]=2 [Link]=0.0 [Link]=0.5
restricts the interface charge to the semiconductor-insulator boundary within the specified rectangle.
In addition to fixed charge, surface recombination velocity and thermionic emission are enabled and
defined with the INTERFACE statement. For more information about this statement, see Chapter 18:
“Statements”, Section 18.18: “INTERFACE”.

2.7.4: Specifying Physical Models

Physical models are specified using the MODELS and IMPACT statements. Parameters for these models
appear on many statements including: MODELS, IMPACT, MOBILITY, and MATERIAL. The physical
models can be grouped into five classes: mobility, recombination, carrier statistics, impact ionization,
and tunneling. Chapter 3: “Physics”, Section 3.6: “Physical Models” contains details for each model.
Tables 2-1 to 2-5 give summary descriptions and recommendations on the use of each model.
Table 2-6 is a guide for compatibility between models.
All models with the exception of impact ionization are specified on the MODELS statement. Impact
ionization is specified on the IMPACT statement. For example, the statement:
MODELS CONMOB FLDMOB SRH FERMIDIRAC
IMPACT SELB

SILVACO International 2-31

ATLAS User’s Manual

specifies that the standard concentration dependent mobility, parallel field mobility, Shockley-Read-
Hall recombination with fixed carrier lifetimes, Fermi Dirac statistics and Selberherr impact
ionization models should be used.
ATLAS also provides an easy method for selecting the correct models for various technologies. The
MOS, BIP, PROGRAM, and ERASE parameters for the MODELS statement configure a basic set of
mobility, recombination, carrier statistics, and tunneling models. The MOS and BIP parameters enable
the models for MOSFET and bipolar devices, while PROGRAM and ERASE enable the models for
programming and erasing programmable devices. For example, the statement:
MODELS MOS PRINT
enables the CVT, SRH, and FERMIDIRAC models, while the statement:
MODELS BIPOLAR PRINT
enables the CONMOB, FLDMOB, CONSRH, AUGER, and BGN.

Note: The PRINT parameter lists to the run time output the models and parameters, which will be used during the
simulation. This allows you to verify models and material parameters. We highly recommend that you include the PRINT
parameter in the MODEL statement.

Physical models can be enabled on a material by material basis. This is useful for heterojunction
device simulation and other simulations where multiple semiconductor regions are defined and may
have different characteristics. For example, the statement:
MODEL MATERIAL=GaAs FLDMOB EVSATMOD=1 ECRITN=6.0e3 CONMOB
MODEL MATERIAL=InGaAs SRH FLDMOB EVSATMOD=1 \
ECRITN=3.0e3
change both the mobility models, and critical electric field used in each material. For devices based on
advanced materials, these model parameters should be investigated carefully.

Energy Balance Models

The conventional drift-diffusion model of charge transport neglects non-local effects such as velocity
overshoot and reduced energy dependent impact ionization. ATLAS can model these effects through
the use of an energy balance model, which uses a higher order approximation of the Boltzmann
Transport Equation (see Chapter 3: “Physics”, Section 3.1.3: “The Transport Equations”). In this
equation, transport parameters such as mobility and impact ionization are functions of the local
carrier temperature rather than the local electric field.
To enable the energy balance transport model, use the HCTE, [Link], or [Link] parameters in
the MODELS statement. These parameters enable the energy transport model for both carriers,
electrons only, or holes only respectively. For example, the statement:
MODELS MOS HCTE
enables the energy balance transport model for both electrons and holes in addition to the default
MOSFET models.

2-32 SILVACO International

 Getting Started with ATLAS

2.7.5: Summary Of Physical Models

Table 2-1. Carrier Statistics Models

Model Syntax Notes

Boltzmann BOLTZMANN Default model

Fermi-Dirac FERMI Reduced carrier concentrations in

heavily doped regions (statistical
approach).

Incomplete Ionization INCOMPLETE Accounts for dopant freeze-out.

Typically, it is used at low temperatures.

Silicon Ionization Model IONIZ Accounts for full ionization for heavily
doped Si. Use with INCOMPLETE.

Bandgap Narrowing BGN Important in heavily doped regions.

Critical for bipolar gain. Use Klaassen
Model.

Table 2-2. Mobility Models

Model Syntax Notes

Concentration Dependent CONMOB Lookup table valid at 300K for Si and GaAs
only. Uses simple power law temperature
dependence.

Concentration and Temperature ANALYTIC Caughey-Thomas formula. Tuned for 77-

Dependent 450K.

Arora’s Model ARORA Alternative to ANALYTIC for Si

Carrier-Carrier Scattering CCSMOB Dorkel-Leturq Model. Includes n, N and T

dependence. Important when carrier con-
centration is high (e.g., forward bias power
devices).
Parallel Electric Field FLDMOB Si and GaAs models. Required to model any
Dependence type of velocity saturation effect.

Tasch Model TASCH Includes transverse field dependence. Only

for planar devices. Needs very fine grid.

Watt Model WATT Transverse field model applied to surface

nodes only.
Klaassen Model KLA Includes N, T, and n dependence. Applies
separate mobility to majority and minority
carriers. Recommended for bipolar devices

Shirahata Model SHI Includes N, E . An alternative surface

⊥
mobility model that can be combined with
KLA.

SILVACO International 2-33

ATLAS User’s Manual

Table 2-2. Mobility Models

Model Syntax Notes

Modified Watt [Link] Extension of WATT model to non-surface

nodes. Applies constant E ⊥ effects. Best
model for planar MOS devices

Lombardi (CVT) Model CVT Complete model including N, T, E//, and

E⊥ effects. Good for non-planar devices.

Yamaguchi Model YAMAGUCHI Includes N, E// and E ⊥ effects. Only cali-

brated for 300K.

Table 2-3. Recombination Models

Model Syntax Notes

Shockley-Read-Hall SRH Uses fixed minority carrier lifetimes. Should

be used in most simulations.

Concentration Dependent CONSRH Uses concentration dependent lifetimes. Rec-

ommended for Si.

Auger AUGER Direct transition of three carriers. Important

at high current densities.

Optical OPTR Band-band recombination. For direct materi-

als only.

Surface S.N Recombination at semiconductor to insulator

S.P interfaces. Set on the INTERFACE statement.

Table 2-4. Impact Ionization

Model Syntax Notes

Selberherr’s Model IMPACT SELB Recommended for most cases. Includes temper-
ature dependent parameters.

Grant’s Model IMPACT Similar to Selberherr’s model but with different

GRANT coefficients.

Crowell-Sze IMPACT Uses dependence on carrier

CROWELL scattering length.

Toyabe Model IMPACT Non-local model used with Energy Balance.

TOYABE Any IMPACT syntax is accepted.

Concannon [Link] Non-local model developed in Flash EEPROM

[Link] technologies.

2-34 SILVACO International

 Getting Started with ATLAS

Table 2-5. Tunneling Models and Carrier Injection Models

Model Syntax Notes

Band-to-Band (standard) [Link] For direct transitions. Required with very high
fields.

Concannon Gate Current [Link] Non-local gate model consistent with

Model [Link] Concannon substrate current model.

Direct Quantum Tunnelling [Link] Quantum tunnelling through conduction band

(Electrons) barrier due to an insulator.

Direct Quantum Tunnelling [Link] Quantum tunnelling through valence band

(Hole) barrier due to an insulator.

Fowler-Nordheim FNORD Self-consistent calculation of tunneling

(electrons) through insulators. Used in EEPROMs.

Fowler-Nordheim (holes) FNHOLES Same as FNORD for holes.

Klaassen Band-to-Band [Link] Includes direct and indirect transitions.

Hot Electron Injection HEI Models energetic carriers tunneling through
insulators. Used for gate current and Flash
EEPROM programming.

Hot Hole Injection HHI HHI means hot hole injection.

Note: In the notes in Tables 2-1 through 2-5, n is electron concentration, p is hole concentration, T is lattice temperature, N is
dopant concentration, Ell is parallel electric field, and E⊥ is perpendicular electric field.

SILVACO International 2-35

ATLAS User’s Manual

Table 2-6. Model Compatibility Chart

CONMOB FLDMOB TFLDMB2 YAMAGUCHI CVT ARORA ANALYTIC CCSMOB SURMOB LATTICE H [Link]

CONMOB [CM] — OK OK YA CV AR AN CC OK OK OK

FLDMOB [FM] OK — TF1 YA CV OK OK OK OK OK OK

TFLDMB2 [TF] OK TF1 — YA CV OK OK TF TF OK OK

YAMAGUCHI YA YA YA — CV YA YA YA YA NO NO
[YA]

CVT [CV] CV CV CV CV — CV CV CV CV OK OK
ARORA [AR] AR OK OK YA CV — AR CC OK OK OK

ANALYTIC [AN] AN OK OK YA CV — CC OK OK OK

CCSMOB [CC] CC OK TF YA CV CC CC — OK OK OK

SURFMOB [SF] OK OK TF YA CV OK OK OK — OK OK

LATTICE H [LH] OK OK OK NO OK OK OK OK OK — OK

[Link] OK OK OK NO OK OK OK OK OK OK 2
[EB]

Key To Table Entries

MODEL ABBREVIATION = The model that supersedes when a combination is specified. In some cases, but not all, a warning message is issued when
a model is ignored.
OK = This combination is allowed.
NO = This combination is not allowed.
NOTES:
[Link] internal model similar to FLDMOB
[Link] models including a parallel electric field dependence are used with energy balance the electric field term is replaced by a
function of carrier temperature.

2-36 SILVACO International

 Getting Started with ATLAS

Using the C-Interpreter to Specify Models

One of the ATLAS products is a C language interpreter that allows you to specify many of the models
used by ATLAS. To use these functions, implement the model in C as equations in a special file called
an . You can access the default ATLAS template file by typing:
atlas -T <filename>
at the UNIX command prompt. This creates a default template file with the name specified by the
<filename> parameter. See Appendix A: “C-Interpreter Functions” for a listing of the default
C-INTERPRETER functions. To use the interpreter functions, give the corresponding parameters in the
statements containing the name of the C language file with the model given as the parameter value.
For example, the statement:
MATERIAL NAME=Silicon [Link]=[Link]
specifies that the file, , contains the C-INTERPRETER function for the specification of the
parallel field dependent electron mobility model.

SILVACO International 2-37

ATLAS User’s Manual

2.8: Choosing Numerical Methods

2.8.1: Numerical Solution Techniques
Several different numerical methods can be used for calculating the solutions to semiconductor device
problems. Numerical methods are given in the METHOD statements of the input file. Some guidelines
for these methods will be given here, but for full details, see Chapter 17: “Numerical Techniques”.
Different combinations of models will require ATLAS to solve up to six equations. For each of the
model types there are basically three types of solution techniques: (a) decoupled (GUMMEL), (b) fully
coupled (NEWTON) and (c) BLOCK. The GUMMEL method will solve for each unknown in turn keeping the
other variables constant, repeating the process until a stable solution is achieved. The NEWTON method
solve the total system of unknowns together. The BLOCK methods will solve some equations fully
coupled, while others are de-coupled.
Generally, the GUMMEL method is useful where the system of equations is weakly coupled, but has only
linear convergence. The NEWTON method is useful when the system of equations is strongly coupled
and has quadratic convergence. The NEWTON method may however spend extra time solving for
quantities which are essentially constant or weakly coupled. NEWTON also requires a more accurate
initial guess to the problem to obtain convergence. Thus, a BLOCK method can provide for faster
simulations times in these cases over NEWTON. GUMMEL can often provide better initial guesses to
problems. It can be useful to start a solution with a few GUMMEL iterations to generate a better guess
and then switch to NEWTON to complete the solution. Specification of the solution method is carried out
as follows:
METHOD GUMMEL BLOCK NEWTON
The exact meaning of the statement depends on the particular models it is applied to. This will be
discussed in the following sections.

Basic Drift Diffusion Calculations

The isothermal drift diffusion model requires the solution of three equations for potential, electron
concentration, and hole concentration. Specifying GUMMEL or NEWTON alone will produce simple
Gummel or Newton solutions as detailed above. For almost all cases, the NEWTON method is preferred
and it is the default.
Specifying:
METHOD GUMMEL NEWTON
will cause the solver to start with GUMMEL iterations and then switch to NEWTON, if convergence is not
achieved. This is a robust, yet a more time consuming way of obtaining solutions for any device. But
this method is highly recommended for all simulations with floating regions such as SOI transistors. A
floating region is defined as an area of doping which is separated from all electrodes by a pn junction.
BLOCK is equivalent to NEWTON for all isothermal drift-diffusion simulations.

Drift Diffusion Calculations with Lattice Heating

When the lattice heating model is added to drift diffusion, an extra equation is added. The BLOCK
algorithm solves the three drift diffusion equations as a NEWTON solution and follows this with a
GUMMEL solution of the heat flow equation. The NEWTON algorithm solves all four equations in a
coupled manner. NEWTON is preferred once the temperature is high, however BLOCK is quicker for low
temperature gradients. Typically, the combination used is:
METHOD BLOCK NEWTON

2-38 SILVACO International

 Getting Started with ATLAS

Energy Balance Calculations

The energy balance model requires the solution of up to 5 coupled equations. GUMMEL and NEWTON have
the same meanings as with the drift diffusion model (i.e., GUMMEL specifies a de-coupled solution and
NEWTON specifies a fully coupled solution).
But BLOCK performs a coupled solution of potential, carrier continuity equations followed by a coupled
solution of carrier energy balance, and carrier continuity equations.
It is possible to switch from BLOCK to NEWTON by specifying multiple solution methods on the same
line. For example:
METHOD BLOCK NEWTON
will begin with BLOCK iterations then switch to NEWTON if convergence is still not achieved. This is the
most robust approach for many energy balance applications.
The points at which the algorithms switch is predetermined, but can also be changed in the METHOD
statement. The default values set by Silvaco work well for most circumstances.

Energy Balance Calculations with Lattice Heating

When non-isothermal solutions are performed in conjunction with energy balance models, a system of
up to six equations must be solved. GUMMEL or NEWTON solve the equations iteratively or fully coupled
respectively. BLOCK initially performs the same function as with energy balance calculations, then
solves the lattice heating equation in a de-coupled manner.

Setting the Number Of Carriers

ATLAS can solve both electron and hole continuity equations, or only for one or none. This choice can
be made by using the CARRIERS parameter. For example:
METHOD CARRIERS=2
specifies a solution for both carriers is required. This is the default. With one carrier, the ELEC or HOLE
parameter is needed. For example, for hole solutions only:
METHOD CARRIERS=1 HOLE
To select a solution for potential only specify:
METHOD CARRIERS=0

Note: Setting the number of carriers using the syntax, MODEL NUMCARR=<n>, is obsolete and should not be used.

SILVACO International 2-39

ATLAS User’s Manual

Important Parameters Of the METHOD Statement

You can alter all of the parameters relevant to the numerical solution process. This, however, isn’t
recommended unless you have expert knowledge of the numerical algorithms. All of these parameters
have been assigned optimal values for most solution conditions. For more information about numerical
algorithms, see Chapter 17: “Numerical Techniques”.
The following parameters, however, are worth noting at this stage:
• CLIMIT or [Link] specify minimal values of concentrations to be resolved by the solver.
Sometimes you need to reduce this value to aid solutions of breakdown characteristics. A value of
CLIMIT=1e-4 is recommended for all simulations of breakdown, where the pre-breakdown current is
small. [Link] is equivalent to CLIMIT but uses the more convenient units of cm-3 for the critical
concentration. [Link] and CLIMIT are related by the following expression.

CLIM DD = CLIMIT∗ N c N v 2-1

·

• DVMAX controls the maximum update of potential per iteration of Newton’s method. The default
corresponds to 1V. For power devices requiring large voltages an increased value of DVMAX might be
needed. DVMAX=1e8 can improve the speed of high voltage bias ramps.
• [Link] controls the cut-off carrier concentration below which the program will not consider the
error in the carrier temperature. This is applied in energy balance simulations to avoid excessive
calculations of the carrier temperature at locations in the structure, where the carrier concentration
is low. But if this parameter is set to too high, where the carrier temperature errors for significant
carrier concentrations are being ignored, unpredictable and mostly incorrect results will be seen.

Restrictions on the Choice of METHOD

The following cases require METHOD NEWTON CARRIERS=2 to be set for isothermal drift-diffusion
simulations:
• current boundary conditions
• distributed or lumped external elements
• AC analysis
• impact ionization
Both BLOCK and/or NEWTON are permitted for lattice heat and energy balance.

Note: Simulations using the GUMMEL method in these cases may lead to non-convergence or incorrect results.

2-40 SILVACO International

 Getting Started with ATLAS

Pisces-II Compatibility
Previous releases of ATLAS (2.0.0.R) and other PISCES-II based programs, use the SYMBOLIC
command to define the solution method and the number of carriers to be included in the solution. In
this version of ATLAS, the solution method is specified completely on the METHOD statement.
The COMB parameter, which was available in earlier ATLAS versions, is no longer required. It has been
replaced with either the BLOCK method or the combination of GUMMEL and NEWTON parameters. The
following table identifies direct translations of old syntax to new.

Note: These are direct translations and not necessarily the best choices of numerical methods.

Table 2-7. Parameter Syntax Replacements

Old Syntax (V2.0.0.R) New Syntax

symbolic newton carriers=2 method newton

symbolic newton carriers=1 elec method newton carriers=1 electron

symbolic gummel carriers=0 method gummel carriers=0

symbolic newton carriers=2 method gummel newton
method comb
models [Link] models [Link]
symbolic newton carriers=2 method block
method comb

models hcte models hcte

symbolic gummel carriers=2 method block
method comb

SILVACO International 2-41

ATLAS User’s Manual

2.9: Obtaining Solutions

ATLAS can calculate DC, AC small signal, and transient solutions. Obtaining solutions is rather
analogous to setting up parametric test equipment for device tests. You usually define the voltages on
each of the electrodes in the device. ATLAS then calculates the current through each electrode. ATLAS
also calculates internal quantities, such as carrier concentrations and electric fields throughout the
device. This is information that is difficult or impossible to measure.
In all simulations, the device starts with zero bias on all electrodes. Solutions are obtained by stepping
the biases on electrodes from this initial equilibrium condition. As will be discussed, due to the initial
guess strategy, voltage step sizes are limited. This section concentrates on defining solution
procedures. Saving results using the LOG or SAVE statements, and analyzing and displaying these
results, is described in Section 2.10: “Interpreting The Results”.

2.9.1: DC Solutions
In DC solutions, the voltage on each electrode is specified using the SOLVE statement. For example, the
statements:
SOLVE VGATE=1.0
SOLVE VGATE=2.0
solves a single bias point with 1.0V and then 2.0V on the gate electrode. One very important rule in
ATLAS is that when the voltage on any electrode is not specified in a given SOLVE statement, the
value from the last SOLVE statement is assumed.
In the following case, the second solution is for a drain voltage of 1.0V and a gate voltage of 2.0V.
SOLVE VGATE=2.0
SOLVE VDRAIN=1.0
When the voltage on a particular electrode is never defined on any SOLVE statement and voltage is
zero, you don’t need to explicitly state the voltage on all electrodes on all SOLVE statements. For
example, in a MOSFET, if VSUBSTRATE is not specified, then Vbs defaults to zero.

Sweeping The Bias

For most applications, a sweep of one or more electrodes is usually required. The basic DC stepping is
inconvenient and a ramped bias should be used. To ramp the base voltage from 0.0V to 1.0V with
0.05V steps with a fixed collector voltage of 2.0V, use the following syntax:
SOLVE VCOLLECTOR=2.0
SOLVE VBASE=0.0 VSTEP=0.05 VFINAL=1.0 NAME=base
The NAME parameter is required and the electrode name is case-sensitive. Make sure the initial
voltage, VSTEP and VFINAL, are consistent. A badly specified ramp from zero to 1.5V in 0.2V steps will
finish at 1.4V or 1.6V.

2-42 SILVACO International

 Getting Started with ATLAS

Generating Families Of Curves

Many applications such as MOSFET Id/Vds and bipolar Ic/Vce simulations require that a family of
curves is produced. This is done by obtaining solutions at each of the stepped bias points first, and
then solving over the swept bias variable at each stepped point. For example, in MOSFET Id/Vds
curves, solutions for each Vgs value are obtained with Vds=0.0V. The output from these solutions are
saved in ATLAS solution files. For each gate bias, the solution file is loaded and the ramp of drain
voltage performed.
The family of curves for three 1V gate steps and a 3.3V drain sweep would be implemented in ATLAS
as follows:
SOLVE VGATE=1.0 OUTF=solve_vgate1
SOLVE VGATE=2.0 OUTF=solve_vgate2
SOLVE VGATE=3.0 OUTF=solve_vgate3
LOAD INFILE=solve_vgate1
LOG OUTFILE=mos_drain_sweep1
SOLVE NAME=drain VDRAIN=0 VFINAL=3.3 VSTEP=0.3
LOAD INFILE=solve_vgate2
LOG OUTFILE=mos_drain_sweep2
SOLVE NAME=drain VDRAIN=0 VFINAL=3.3 VSTEP=0.3
LOAD INFILE=solve_vgate3
LOG OUTFILE=mos_drain_sweep3
SOLVE NAME=drain VDRAIN=0 VFINAL=3.3 VSTEP=0.3
The LOG statements are used to save the Id/Vds curve from each gate voltage to separate files. We
recommend that you save the data in this manner rather than to a single LOG file (see Section 2.10:
“Interpreting The Results” ).

2.9.2: The Importance Of The Initial Guess

To obtain convergence for the equations used, supply a good initial guess for the variables to be
evaluated at each bias point. The ATLAS solver uses this initial guess and iterates to a converged
solution. For isothermal drift diffusion simulations, the variables are the potential and the two carrier
concentrations. If a reasonable grid is used, almost all convergence problems in ATLAS are caused by
a poor initial guess to the solution.
During a bias ramp the initial guess for any bias point is provided by a projection of the two previous
results. Problems tend to appear near the beginning of the ramp when two previous results are not
available. If one previous bias is available, it is used alone. This explains why the following two
examples eventually produce the same result. The first will likely have far more convergence problems
than the second.
1. SOLVE VGATE=1.0 VDRAIN=1.0 VSUBSTRATE=-1.0
2. SOLVE VGATE=1.0
SOLVE VSUBSTRATE=-1.0
SOLVE VDRAIN=1.0
In the first case, one solution is obtained with all specified electrodes at 1.0V. In the second case, the
solution with only the gate voltage at 1.0V is performed first. All other electrodes are at zero bias.
Next, with the gate at 1.0V, the substrate potential is raised to -1.0V and another solution is obtained.
Finally, with the substrate and the gate biased, the drain potential is added and the system solved

SILVACO International 2-43

ATLAS User’s Manual

Generally the projection method for the initial guess gives good results when the I-V curve is linear.
But it may encounter problems if the IV curve is highly non-linear or if the device operating mode is
changing. Typically, this might occur around the threshold or breakdown voltages. At these biases
smaller voltage steps are required to obtain convergence. As will be described, ATLAS contains
features such as the TRAP parameter and the curve tracer to automatically cut the voltage steps in
these highly non-linear area.
Numerical methods are described in Section 2.8: “Choosing Numerical Methods” and Chapter 17:
“Numerical Techniques”.
In many cases, these methods are designed to overcome the problems associated with the initial guess.
This is particularly important in simulations involving more than the three drift diffusion variables.
Generally, coupled solutions require a good initial guess, whereas de-coupled solutions can converge
with a poor initial guess.

The Initial Solution

When no previous solutions exist, the initial guess for potential and carrier concentrations must be
made from the doping profile. This is why the initial solution performed must be the zero bias (or
thermal equilibrium) case. This is specified by the statement:
SOLVE INIT
But if this syntax isn’t specified, ATLAS automatically evaluates an initial solution before the first
SOLVE statement. To aid convergence of this initial guess, the solution is performed in the zero carrier
mode solving only for potential.

The First and Second Non-Zero Bias Solutions

From experience with ATLAS, it is found that the first and second non-zero bias solutions are the most
difficult to obtain good convergence. Once these two solutions are obtained, the projection algorithm
for the initial guess is available and solutions should all have a good initial guess.
These first two solutions, however, must use the result of the initial solution as the basis of their initial
guess. Since the initial solution is at zero bias, it provides a poor initial guess.
The practical result of this is that the first and second non-zero bias solutions should have very small
voltage steps. In the following example, the first case will likely converge, whereas the second case
may not.
1. SOLVE INIT
SOLVE VDRAIN=0.1
SOLVE VDRAIN=0.2
SOLVE VDRAIN=2.0
2. SOLVE INIT
SOLVE VDRAIN=2.0

The Trap Parameter

Although ATLAS provides se1.2 25a0 9.96 68.4 184.6801 [Link]

2-44 SILVACO International

 Getting Started with ATLAS

If the second SOLVE statement does not converge, TRAP automatically cuts the bias step in half and
tries to obtain a solution for Vd = 1.0V. If this solution does not converge, the bias step will be halved
again to solve for Vd = 0.5V. This procedure is repeated up to a maximum number of tries set by the
METHOD parameter MAXTRAPS. Once convergence is obtained, the bias steps are increased again to
solve up to 2.0V. The default for MAXTRAPS is four and it is not recommended to increase it since
changing the syntax to use smaller bias steps is generally much faster.
This trap facility is very useful during bias ramps in overcoming convergence difficulties around
transition points such as the threshold voltage. Consider the following syntax used to extract a
MOSFET Id/Vgs curve.
SOLVE VGATE=0.0 VSTEP=0.2 VFINAL=5.0 NAME=gate
Assume the threshold voltage for the device being simulated is 0.7V and that ATLAS has solved for the
gate voltages up to 0.6V. The next solution, at 0.8V, might not converge at first. This is because the
initial guess was formed from the two sub-threshold results at Vgs=0.4V and 0.6V and the solution has
now become non-linear. The trap facility will detect the problems in the 0.8V solution and cut the bias
step in half to 0.7V and try again. This will probably converge. The solution for 0.8V will then be
performed and the bias ramp will continue with 0.2V steps.

2.9.3: Small-Signal AC Solutions

Specifying AC simulations is a simple extension of the DC solution syntax. AC small signal analysis is
performed as a post-processing operation to a DC solution. Two common types of AC simulation in
ATLAS are outlined here. The results of AC simulations are the conductance and capacitance between
each pair of electrodes. Tips on interpreting these results is described in Section 2.10: “Interpreting
The Results”.

Single Frequency AC Solution During A DC Ramp

The minimum syntax to set an AC signal on an existing DC ramp is just the AC flag and the setting of
the small signal frequency. For example:
SOLVE VBASE=0.0 VSTEP=0.05 VFINAL=1.0 NAME=base AC FREQ=1.0e6
Other AC syntax for setting the signal magnitude and other parameters are generally not needed as
the defaults suffice. One exception is in 1D MOS capacitor simulations. To obtain convergence in the
inversion/deep depletion region, add the DIRECT parameter to access a more robust solution method.

Ramped Frequency At A Single Bias

For some applications, such as determining bipolar gain versus frequency you need to ramp the
frequency of the simulation. This is done using the following syntax:
1. SOLVE VBASE=0.7 AC FREQ=1e9 FSTEP=1e9 NFSTEPS=10
2. SOLVE VBASE=0.7 AC FREQ=1e6 FSTEP=2 MULT.F NFSTEPS=10
The first case ramps the frequency from 1GHz to 11GHz in 1GHz steps. A linear ramp of frequency is
used and FSTEP is in Hertz. In the second example, a larger frequency range is desired and so a
geometrical step of frequency is used. The MULT.F parameter is used to specify that FSTEP is a
unitless multiplier for the frequency. This doubles the frequency in successive steps from 1MHz to
1.024GHz.
The syntax described below for ramping the frequency of the AC signal can potentially be combined
with that for ramping the bias. The frequency ramps are done at each bias point during the DC ramp.
SOLVE VBASE=0.0 VSTEP=0.05 VFINAL=1.0 NAME=base AC FREQ=1.0e6 \
FSTEP=2 MULT.F NFSTEPS=10

SILVACO International 2-45

ATLAS User’s Manual

2.9.4: Transient Solutions

Transient solutions can be obtained for piecewise-linear, exponential, and sinusoidal bias functions.
Transient solutions are used when a time dependent test or response is required. To obtain transient
solutions for a linear ramp, the TSTART, TSTOP, TSTEP, and RAMPTIME parameters should be
specified.
Figure 2-11 shows the syntax of the transient voltage ramp. The TSTART parameter specifies the time
that the linear ramp should start. The RAMPTIME specifies the time that the linear ramp should obtain
its final value. TSTOP specifies the time that solutions will stop. TSTEP specifies the initial step size.
Subsequent time steps are calculated automatically by ATLAS. For example, the statement:
SOLVE VGATE=1.0 RAMPTIME=1E-9 TSTOP=10e-9 TSTEP=1e-11,
specifies that the voltage on the gate electrode will be ramped in the time domain from its present
value to 1.0V over a period of 1 nanoseconds.
Time domain solutions are obtained for an additional 9 nanoseconds. An initial time step of 10
picoseconds is specified. It is important to remember that if subsequent transient solutions are
specified, don’t reset the time to zero.

Vg

RAMP TIME TSTOP

1.0

VGATE

TSTEP

1 2 3 4 5 6 7 8 9 10
time (ns)

Figure 2-11: Diagram showing syntax of Transient Voltage Ramp in ATLAS

2-46 SILVACO International

 Getting Started with ATLAS

2.9.5: Advanced Solution Techniques

Obtaining Solutions Around The Breakdown Voltage
Obtaining solutions around the breakdown voltage can be difficult using the standard ATLAS
approach. It requires special care when choosing voltage steps and interpreting the results. The curve
tracer described in “The Curvetrace Capability” section on page 2-48 is the most effective method in
many cases.
A MOSFET breakdown simulation might be performed using the following standard syntax for
ramping the drain bias. Note that the setting of CLIMIT is recommended for breakdown simulations
when the pre-breakdown leakage is low.
IMPACT SELB
METHOD CLIMIT=1e-4
SOLVE VDRAIN=1.0 VSTEP=1.0 VFINAL=20.0 NAME=drain
If the breakdown were 11.5V, then convergence problems will be expected for biases higher than 11.0V.
Although it depends on technology used, it is common for the breakdown curve to be flat up to a
voltage very close to breakdown and then almost vertical. The current changes by orders of magnitude
for very small bias increments.
This produces some problems for ATLAS using the syntax described above. If the breakdown occurs at
11.5V, there are no solutions for voltages greater than this value. ATLAS is trying to ramp to 20.0V so
it is inevitable that ATLAS will fail to converge at some point. This is usually not a problem since by
that point the breakdown voltage and curve have been obtained.
Above 11V, bias step reduction will take place due to the TRAP parameter. ATLAS will continually try
to increase the drain voltage above 11.5V and those points will fail to converge. But it will solve points
asymptotically approaching Vds=11.5V until the limit set by the MAXTRAPS parameter is reached. If
the default of 4 traps is used it is clear that the minimum allowed voltage step is 1.0x(0.5)4 or 0.004V.
This is normally enough accuracy for determining the breakdown point. But the simulation might not
allow the current to reach a sufficiently high level before MAXTRAPS is needed.
Typically in device simulation, the breakdown point is determined once the current is seen to increase
above the flat pre-breakdown leakage value by two orders of magnitude in a small voltage increment.
If you want to trace the full breakdown curve up to high current values, apply more advanced
techniques than the simple voltage ramp. Two of these techniques are described in the following
subsections. These methods may use extra CPU time.

Using Current Boundary Conditions

In all of the examples considered in the basic description of the SOLVE statement, it was assumed that
voltages were being forced and currents were being measured. ATLAS also supports the reverse case
through current boundary conditions. The current through the electrode is specified in the SOLVE
statement and the voltage at the contact is calculated. Current boundary conditions are set using the
CONTACT statement as described in Section 2.7: “Defining Material Parameters And Models”.
The syntax of the SOLVE statement is altered once current boundary conditions are specified.
SOLVE IBASE=1e-6
The syntax above specifies a single solution at a given current.
SOLVE IBASE=1e-6 ISTEP=1e-6 IFINAL=5e-6 NAME=base
This sets a current ramp similar in syntax to the voltage ramp described earlier.
SOLVE IBASE=1e-10 ISTEP=10 IMULT IFINAL=1e-6 NAME=base

SILVACO International 2-47

ATLAS User’s Manual

This is similar to the previous case, but the IMULT parameter is used to specify that ISTEP should be
used as a multiplier for the current rather than a linear addition. This is typical for ramps of current
since linear ramps are inconvenient when several orders of magnitude in current may need to be
covered.
Important points to remember about current boundary conditions are that the problems of initial
guess are more acute when very small (noise level) currents are used. Often it is best to ramp the
voltage until the current is above 1pA/µm and then switch to current forcing.
When interpreting the results, it is important to remember the calculated voltage on the electrode
with current boundary conditions is stored as the internal bias (i.e., base [Link] in TONYPLOT or
[Link] in DECKBUILD’s extract syntax).

The Compliance Parameter

Compliance is a parameter used to limit the current or voltage through or on an electrode during a
simulation. An electrode compliance can be set. After it is reached, the bias sweep will stop. This is
analogous to parametric device testing when we stop a device from being over stressed or destroyed.
The compliance refers to the maximum resultant current or voltage present after a solution is
obtained. If an electrode voltage is set, the compliance will then refer to the electrode current. If
current boundary conditions are used, then you can set a voltage compliance.
The statements:
SOLVE VGATE=1.0
SOLVE NAME=drain VDRAIN=0 VFINAL=2 VSTEP=0.2 COMPL=1E-6 CNAME=drain
solve for IV on the gate and then ramps the drain voltage towards 2V in 0.2V steps. The simulation
will stop if 1 µ A/ µ m of drain current is reached before Vd = 2V. Thus, as in parametric testing, a
particular level can be defined and the simulation can be set to solve up to that point. Once the
compliance limit is reached, ATLAS will simulate the next statement line in the command file.

The Curvetrace Capability

The automatic curve tracing algorithm can be invoked to enable ATLAS to trace out complex IV
curves. The algorithm can automatically switch from voltage to current boundary conditions and vice
versa. A single SOLVE statement can be used to trace out complex IV curves such as breakdown curves
and CMOS latch-up including the snapback region and second breakdown. The algorithm is based
upon a dynamic load line approach.
For example, typical CURVETRACE and SOLVE statements to trace out an IV curve for the breakdown of
a diode would look like:
CURVETRACE [Link]=cathode [Link]=0.5 [Link]=1.2 \
MINCUR=1e-12 [Link]=1e-3 [Link]
SOLVE CURVETRACE
The name of the electrode, which is to be ramped, is specified using [Link]. [Link]
specifies the initial voltage step. [Link] specifies the factor used to increase the voltage step in
areas on the IV curve away from turning points. MINCUR can be used to set a small current value above
which the dynamic load line algorithm is activated. Below the MINCUR level, the [Link] and
[Link] are used to determine the next solution bias. [Link] is used to stop the tracing if the
voltage or current of the ramped electrode equals or exceeds [Link]. Either [Link] or
[Link] is used to specify whether [Link] is a voltage or current value.
When plotting the log file created by the curve trace statement in TONYPLOT, you need to select the
internal bias labeled [Link] for the ramped electrode instead of plotting the applied bias, which is
labeled Voltage.

2-48 SILVACO International

 Getting Started with ATLAS

2.9.6: Using DeckBuild To Specify SOLVE Statements

The DeckBuild Solve menu can be used generate SOLVE statements. The menu has a spreadsheet style
entry. To access this menu, select the Command/Solutions/Solve... button in DECKBUILD. To define
a test, click the right mouse button in the worksheet and select the Add new row option. This will
add a new row to the worksheet. This procedure should be repeated once per electrode in your device
structure. The entry for each cell can then be edited to construct a SOLVE statement.
Some cells require the selection using a pop menu or the entry of numerical values. The electrode
name is specified in the first cell. To edit the electrode name, use a popup menu by pressing the right
mouse button on the cell. The second cell specifies whether the electrode will be a voltage (V), current
(I) or charge (Q) controlled. The third cell specifies whether the SOLVE statement is to be a single DC
solve (CONST), a swept DC variable (VAR 1), a stepped DC variable (VAR2), or a transient solution
(PULSE). The remaining cells specify the parameter values that are required for the type of solution
desired.
The pop-up window to specify the solution file names are accessed through the Props... button.
Several SOLVE statements may be constructed to create solve sequences which define a test. This test
may be saved in a file and read in using the Save... and Load... buttons.

SILVACO International 2-49

ATLAS User’s Manual

2.10: Interpreting The Results

As already described in Section 2.2: “ATLAS Inputs and Outputs”, ATLAS produces three different
types of output files. These files are also described in the following sections.

2.10.1: Run-Time Output

Run-time output is provided at the bottom of the DeckBuild Window. If it’s run as a batch job, the
run-time output can be stored to a file.
Errors occurring in the run-time output will be displayed in this window. Note that not all errors will
be fatal (as DECKBUILD tries to interpret the file and continue). This may cause a statement to be
ignored, leading to unexpected results. We recommend that you check the run-time output of any
newly created input file the first time it runs to intercept any errors.
If you specify the PRINT option within the MODELS statement, the details of material parameters and
constants and mobility models will be specified at the start of the run-time output. This is a useful way
of checking what parameters values and models are being applied in the simulation. We recommend
that you always specify MODELS PRINT in input files.
During SOLVE statements the error numbers of each equation at each iteration are displayed (this is a
change from the previous ATLAS version.) It isn’t vital for you to understand the iteration information
but it can provide important insights in the case of convergence problems.
The output can be interpreted as follows:
proj psi n p psi n p
direct x x x rhs rhs rhs

i j m -5.00* -5.00* -5.00* -26.0* -17.3* -17.3*

- - - - - - - - - - - - - - - - - - - - - - - - - - -
1 N -1.932 -2.934 -1.932 -25.2 -10.19 -9.876
2 N -4.741 -5.64* -4.267 -28.8* -16.67 -15.47
3 A -11.3* -11.7* -9.63* -28.8* -16.67 -18.0*

Electrode Va(V) Jn(A/um) Jp(A/um) Jc(A/um) Jt(A/um)

=======================================================================
gate 0.000e+00 -0.000e+00 -0.000e+00 0.000e+00 0.000e+00
source 0.000e+00 -3.138e-13 -1.089e-35 -3.138e-13 -3.138e-13
drain 1.000e-01 3.139e-13 1.076e-23 3.139e-13 3.139e-13
substrate 0.000e+00 -6.469e-19 -8.853e-17 -8.918e-17 -8.918e-17

The top left value, proj, indicates the initial guess methodology used. The default projection method
is used here. Alternatives are previous, local, or init. The second value, direct, indicates the
solver type. This will either be direct or iterative.
The first three column headings: i, j, and m indicate the iteration numbers of the solution and the
solution method. i indicates the outer loop iteration number for decoupled solutions. j indicates the
inner loop number. m indicates the solution method by a single letter, which are:
G = gummel
B = block
N = newton
A = newton with autonr
S = coupled Poisson-Schrodinger solution

2-50 SILVACO International

 Getting Started with ATLAS

The remaining column headings indicate which column lists the XNORM and RHSNORM errors for the
equations being solved. See Chapter 17: “Numerical Techniques”, Section 17.5.7: “Error Measures” for
a full description of these errors. The values printed in each error column under the hashed line are
the logarithm to base 10 of the error. Earlier PISCES versions would print the floating point value.
The values printed above the dashed line in each column are the tolerances used.
When the star * symbol appears as the least significant digit in the number, it means this error
measure has met its tolerance.
After convergence is achieved, ATLAS lists the results by electrode. The column, , lists the voltage
at the contact surface. This will differ from the applied voltage if external resistors or the curvetracer
are used. All relevant current components are listed. Here, only electron, hole, conduction, and total
currents are given. In other modes of simulation, these columns may differ.
The output of AC analysis, MIXEDMODE, and 3-D simulations differ from this standard. See Chapter
12: “MixedMode: Mixed Circuit and Device Simulator” for more information about MIXEDMODE.
ATLAS may produce a very large amount of run-time output for complex simulations. Run-Time
output may be saved to a file as shown in Section 2.3: “Modes of Operation”.

2.10.2: Log Files

Log files store the terminal characteristics calculated by ATLAS. These are current and voltages for
each electrode in DC simulations. In transient simulations, the time is stored. In AC simulations, the
small signal frequency and the conductances and capacitances are saved. For example, the statement:
LOG OUTF=<FILENAME>
is used to open a log file. Terminal characteristics from all SOLVE statements after the LOG statement
are then saved to this file along with any results from the PROBE statement.
The only way to stop the terminal characteristics being saved to this file is to use another LOG
statement with either a different log filename or the OFF parameter.
Typically, a separate log file should be used for each bias sweep. For example, separate log files are
used for each gate bias in a MOS Id/Vds simulation or each base current in a bipolar Ic/Vce simulation.
These files are then overlaid in TONYPLOT.
Log files contain only the terminal characteristics. They are typically viewed in TONYPLOT. Parameter
extraction of data in log files can be done in DECKBUILD. Log files cannot be loaded into ATLAS to
re-initialize the simulation.

Units Of Currents In Log files

Generally, the units of current written into the log file and therefore seen in TONYPLOT is Amperes per
micron. This is because ATLAS is a two-dimensional simulator. It sets the third dimension (or z-
direction) to be one micron. Thus, if you compare ATLAS 2D simulation results for a MOSFET versus
the measured data from a MOSFET of width 20 micron, you need to multiply the current in the log file
by 20.
There are four exceptions:
• In the 3D modules of ATLAS, the width is defined in the 3-D structure and so the units of the
current are Amperes.
• In MIXEDMODE, set the width of devices. The current is also in Amperes.
• When cylindrical coordinates are used, the current written to the log file is integrated through the
cylinder and is also in Amperes.
• When the WIDTH parameter in the MESH statement is used, the current is scaled by this factor and
is in Amperes.

SILVACO International 2-51

ATLAS User’s Manual

Similar rules apply for the capacitance and conductance produced by AC simulations. These are
usually in 1/([Link]) and Farads/micron respectively.

UTMOST Interface
ATLAS log files can be read directly into the batch mode, UTMOST. The following commands in
UTMOST are used to read in a set of IV curves stored in separate log files.
INIT INF=<filename> MASTER
INIT INF=<filename> MASTER APPEND
The UTMOST statement in ATLAS is no longer recommended for interfacing to UTMOST.

2.10.3: Parameter Extraction In DeckBuild

The EXTRACT command is provided within the DECKBUILD environment. It allows you to extract
device parameters. The command has a flexible syntax that allows you to construct specific EXTRACT
routines. EXTRACT operates on the previous solved curve or structure file. By default, EXTRACT uses
the currently open log file. To override this default, supply the name of a file to be used by EXTRACT
before the extraction routine. For example:
EXTRACT INIT INF="<filename>"
A typical example of using EXTRACT is the extraction of the threshold voltage of an MOS transistor. In
the following example, the threshold voltage is extracted by calculating the maximum slope of the Id /
Vg curve, finding the intercept with the x-axis and then subtracting half of the applied drain bias.
EXTRACT NAME="nvt" XINTERCEPT(MAXSLOPE(CURVE (V."GATE", (I."DRAIN"))) \
-(AVE(V."DRAIN"))/2.0)
The results of the extraction will be displayed in the run-time output and will by default also be stored
in the file [Link]. You can store the results in a different file at the end of EXTRACT
command by using the following option:
EXTRACT....DATAFILE="<filename>"
Cut-off frequency and forward current gain are of particular use as output parameters. These
functions can be defined as follows:
# MAXIMUM CUTOFF FREQUENCY
EXTRACT NAME="FT_MAX" MAX(G."COLLECTOR""BASE"/(6.28*C."BASE""BASE"))
#FORWARD CURRENT GAIN
EXTRACT NAME="PEAK GAIN" MAX(I."COLLECTOR"/ I."BASE")

Note: Over 300 examples are supplied with ATLAS to provide many practical examples of the use of the EXTRACT
statement.

2-52 SILVACO International

 Getting Started with ATLAS

AC Parameter Extraction
Basic analysis of the capacitance and conductance matrix produced by ATLAS can be done using
DECKBUILD or TONYPLOT. The capacitance between gate and drain will be labeled as Cgate>drain in
TONYPLOT or c."gate""drain" in DECKBUILD’S EXTRACT.
The total capacitance on any electrode is defined as Celectrode>electrode. Thus, the magnitude of
Cgate>gate is the total gate capacitance.
The LOG statement also includes options for small-signal, two-port RF analysis including s-parameter
extraction. The solutions are saved into the log file and in the run-time output. The list of options for
RF analysis are:
[Link], [Link], [Link], [Link], [Link] gains
Terminal impedance and parasitics are accounted for by adding any of the following:
impedance=<val>, rin=<val>, rout=<val>, rcommon=<val> or
rground=<val>, lin=<val>, lout=<val>,
lcommon=<val> or lground=<val>, width=<val>
The width defaults to 1µm and impedance defaults to 50Ω. All parasitics default to zero.
The Stern stability factor, k, is calculated along with current gain (h21), GUmax, and GTmax when the
GAINS option is added to the LOG statement.
The run-time output for AC analysis has been modified to only list the analysis frequency and
electrode conductance/capacitance values. If one of the two-port options is added to the LOG statement
(e.g., [Link]), the two-port parameters are also included in the run-time output.
AC parameter conversion utilities in the UTMOST statement have been discontinued.

Additional Functions
EXTRACT has two additional important functions. The first function is to provide data for the VWF
database (i.e., to store device parameters to the VWF database, they must be evaluated using
EXTRACT). The second function is when using the DeckBuild Optimizer to tune parameters, use the
EXTRACT statements as the optimization targets.

2.10.4: Functions In TonyPlot

The Functions Menu in TONYPLOT allows you to specify and plot functions of the terminal
characteristics in the Graph Function text fields. For example transconductance can be calculated
using the following function:
dydx (drain current, gate voltage)
Current gain can be evaluated as:
collector current / base current
When creating functions, the key to correct syntax is that the name for any variable in a function is the
same as that in the Y Quantities list on the Display menu.

SILVACO International 2-53

ATLAS User’s Manual

2.10.5: Solution Files

Solution files or structure files provide an image of the device at a particular bias point (DC solution or
transient solution point). This gives you the ability to view any evaluated quantity within the device
structure in question, from doping profiles and band parameters to electron concentrations and
electric fields. These files should be plotted using TONYPLOT.
The syntax used to generate these files is of two forms:
(1) SAVE OUTFILE=<filename>
(2) SOLVE .... OUTFILE=<filename>.sta MASTER [ONEFILEONLY]
Here in the second form, a file named <filename> will be saved with data from the previously solved
bias point.
In this case, a structure file will be saved at each bias point solved in the solve statement. The last
letter of the file name will be automatically incremented alphabetically (i.e., *.sta, *.stb,
*.stc..., and so on). If the solution for the last bias point is required, you can add the onefileonly
parameter to the command. The <filename>.sta file will be overwritten at each solution point.
Structure files can be very large (1 - 2 MB), depending on the mesh density and quantities saved. It is
recommended that unwanted structure files be deleted.
If many solution files have been written from a long simulation, it is often confusing to find out which
solution file belongs to which bias point or transient time. The solution files should be plotted in
TONYPLOT. In TONYPLOT, you can create 2-D contour plots and 1-D cutlines. To find out the bias of any
solution file in TONYPLOT, select the plot, and press b on the keyboard.

Interpreting Contour Plots

Some quantities saved in the solution files are not evaluated at the node points during solutions They
are evaluated at the center of the sides of each triangle in the mesh. Values of quantities at each node
are derived from averaging the values from the sides of triangles connected to that node. The
weighting method used to do the averaging can be selected with options in the OUTPUT statement. It is
possible that for some meshes, smoother contour plots can be obtained by choosing a non-default
averaging method.
When interpreting the contour plots, it’s important to remember that the solution file contains values
only at each node point. The color fills seen in TONYPLOT are simply interpolations based on the node
values. This may lead to occasional strange contour values. In these cases, check the node values using
the probe in TONYPLOT.
The primary solution variables (potential, carrier concentration, lattice temperature and carrier
temperatures) are calculated on the nodes of the ATLAS mesh and therefore are always correct in
TONYPLOT

2-54 SILVACO International

 Getting Started with ATLAS

Customizing Solution Files (OUTPUT Statement)

Several quantities are saved by default within a structure file. For example, doping, electron
concentration, and potential. You can also specify additional quantities (such as conduction band
potential) by using the OUTPUT statement. This must precede the SAVE statement in question. For
example, to save the conduction and valence band potentials, the following command would be used at
some point before the relevant SAVE.
OUTPUT [Link] [Link]

Saving Quantities from the Structure at each Bias Point (PROBE statement)
Structure files provide all data from the structure at a single bias point. The log files provide terminal
characteristics for a set of bias points. Use the PROBE statement to combine these and allow certain
structural quantities to be saved at each bias point.
The PROBE statement allows you to specify quantities to be saved at given XY locations. There is also a
facility to save the maximum or minimum of certain quantities. The value from the PROBE at each bias
point in DC or timestep in transient mode is saved to the log file. The syntax:
PROBE NAME=mycarriers [Link] X=1 Y=0.1
saves the electron concentration at (1, 0.1) for each solution in the log file. When the log file is
displayed in TONYPLOT, the value will be labelled mycarriers. It will be possible to plot mycarriers
versus terminal bias or current or other probed quantities.
Certain direction dependent quantities, such as electric field and mobility, can be probed. In these
cases, a direction for the vector quantity must be specified using the DIR parameter.
The PROBE statement provides the only way to extract the actual values of quantities that are
calculated along the sides of each triangle in ATLAS. The PROBE statement actually stored the triangle
side value closest to the probed location, while taking into account the direction for vector quantities.

Note: Specifying the probe location exactly at a material or region interface will often lead to erroneous results. It is best to
very slightly offset the location of the probe inside the material or region of interest.

Re-initializing ATLAS at a Given Bias Point

Each SOLVE statement will begin with the device biased at the previous value solved. If you wish to
begin a solution at a previously solved bias point, re-load the structure file saved at that point. This is
accomplished in the following manner:
LOAD INFILE=<filename> MASTER
Information about that solution point will be displayed in the Output Window.
This command is useful for solving a set of I/V curves. For example, to solve a family of Id / Vd (at
various Vg) ramp the gate with zero drain bias. A structure file is then saved at each desired value of
Vg. These structure files can be reloaded in turn while a Vd sweep is performed.

Note: An ATLAS input file cannot start with a LOAD statement. Before loading the structure file, make sure that the device
mesh for the same structure has been loaded using the MESH statement. Also, the same MODELS, MATERIAL, and
CONTACT settings are required when the files are saved by ATLAS.

SILVACO International 2-55

ATLAS User’s Manual

2.10.6: Technology Specific Issues in ATLAS

This chapter was designed to give an overview to the basic use of ATLAS without regard to the details
required for a given technology. Requirements for ATLAS simulation vary considerably. The needs of
the sub-micron MOS device engineer, the 1000V power device engineer, and the III-V RF device
engineer differ and cannot all be covered in one chapter. Silvaco provides many references to
individual technology problems using ATLAS. These are:
• A library of over 500 examples that can be accessed on-line from DECKBUILD. Look at these
examples not only for their technology but also related ones. For example, different aspects of high
frequency analysis is covered in the MESFET and silicon bipolar example sections.
• The chapters for each ATLAS module in this manual.
• The Simulation Standard, a newsletter distributed by Silvaco. To make sure you’re on the mailing
list, contact your local Silvaco office or go to [Link].
• The Silvaco website also provides detailed information. It contains on-line versions of the articles
in our newsletter, on-line searchable index of the examples, links to other TCAD web sites, and a
section on solutions to known problems with all Silvaco programs.
• For more information about suggested technology specific strategies, contact your local Silvaco
support engineer at support@[Link].

2-56 SILVACO International

3.1: Basic Semiconductor Equations
Years of research into device physics have resulted in a mathematical model that operates on any
semiconductor device [1]. This model consists of a set of fundamental equations which link together
the electrostatic potential and the carrier densities, within some simulation domain. These equations,
which are solved inside any general purpose device simulator, have been derived from Maxwell’s laws
and consist of Poisson’s Equation (see Section 3.1.1: “Poisson’s Equation”), the continuity equations
(see Section 3.1.2: “Carrier Continuity Equations”) and the transport equations (see Section 3.1.3: “The
Transport Equations”). Poisson’s Equation relates variations in electrostatic potential to local charge
densities. The continuity and the transport equations describe the way that the electron and hole
densities evolve as a result of transport processes, generation processes, and recombination processes.
This chapter describes the mathematical models implemented in ATLAS. Note that a discretization of
the equations is also performed so that they can be applied to the finite element grid used to represent
the simulation domain.

3.1.1: Poisson’s Equation

Poisson’s Equation relates the electrostatic potential to the space charge density:
( ε∇ψ ) –ρ 3-1

where ψ is the electrostatic potential, ε is the local permittivity, and ρ is the local space charge density.
The reference potential can be defined in various ways. For ATLAS, this is always the intrinsic Fermi
potential ψi which is defined in the next section. The local space charge density is the sum of
contributions from all mobile and fixed charges, including electrons, holes, and ionized impurities.
The electric field is obtained from the gradient of the potential (see Equation 3-2).

–∇ ψ 3-2

3.1.2: Carrier Continuity Equations

The continuity equations for electrons and holes are defined by equations:
∂
– 3-3
∂

3-4

where n and p are the electron and hole concentration, and are the electron and hole current
densities, and are the generation rates for electrons and holes, Rn and Rp are the recombination
rates for electrons and holes, and q is the magnitude of the charge on an electron.
By default ATLAS includes both Equations 3-3 and 3-4. In some circumstances, however, it is
sufficient to solve only one carrier continuity equation. The specification of which continuity equations
are to be solved is performed in the METHOD statement by turning off any equation that is not to be
solved. The syntax, ^ELECTRONS or ^HOLES, turns off the electron continuity equation and the hole
continuity equation respectively.

SILVACO International 3-1

ATLAS User’s Manual

3.1.3: The Transport Equations

Equations 3-1, 3-3, and 3-4 provide the general framework for device simulation. But further

secondary equations are needed to specify particular physical models for: , , , , and .

The current density equations, or charge transport models, are usually obtained by applying
approximations and simplifications to the Boltzmann Transport Equation. These assumptions can
result in a number of different transport models such as the drift-diffusion model, the Energy Balance
Transport Model or the hydrodynamic model. The choice of the charge transport model will then have
a major influence on the choice of generation and recombination models.
The simplest model of charge transport that is useful is the Drift-Diffusion Model. This model has the
attractive feature that it does not introduce any independent variables in addition to ψ, n and p. Until
recently, the drift-diffusion model was adequate for nearly all devices that were technologically
feasible. The drift-diffusion approximation, however, becomes less accurate for smaller feature sizes.
More advanced energy balance and hydrodynamic models are therefore becoming popular for
simulating deep submicron devices. ATLAS supplies both drift-diffusion and advanced transport
models.
The charge transport models and the models for generation and recombination in ATLAS make use of
some concepts associated with carrier statistics. These concepts are summarized in a further section of
this chapter that deals with the carrier statistics.
Drift-Diffusion Transport Model
Derivations based upon the Boltzmann transport theory have shown that the current densities in the
continuity equations may be approximated by a drift-diffusion model [2]. In this case, the current
densities are expressed in terms of the quasi-Fermi levels φn and φp as:

= – µ ∇φ 3-5

= – µ ∇φ 3-6

where µn and µp are the electron and hole mobilities. The quasi-Fermi levels are then linked to the
carrier concentrations and the potential through the two Boltzmann approximations:

(ψ – φ )
= ------------------------- 3-7

– (ψ – φ )
= ---------------------------- 3-8

where nie is the effective intrinsic concentration and is the lattice temperature. These two
equations may then be re-written to define the quasi-Fermi potentials:

φ = ψ – ------------ -------- 3-9

φ = ψ + ------------ -------- 3-10

3-2 SILVACO International

 Physics

By substituting these equations into the current density expressions, the following adapted current
relationships are obtained:

= ∇ – µ ∇ψ – µ ( ∇( )) 3-11

– ∇ – µ ∇ψ µ ( ∇( )) 3-12

The final term accounts for the gradient in the effective intrinsic carrier concentration, which takes
account of bandgap narrowing effects. Effective electric fields are normally defined whereby:

– ∇ ψ 
 3-13

– ∇ ψ –  3-14
 

Which then allows the more conventional formulation of drift-diffusion equations to be written
(see Equations 3-15 and 3-16).

= µ + ∇ 3-15

= µ – ∇ 3-16

It should be noted that this derivation of the drift-diffusion model has tacitly assumed that the
Einstein relationship holds. In the case of Boltzmann statistics this corresponds to:

µ 3-17

µ 3-18

If Fermi-Dirac statistics are assumed for electrons, Equation 3-17 becomes:

  

µ  ⁄ 
[ε – ε ]
 
3-19
 
– ⁄  [ε – ε ]
 

where α is the Fermi-Dirac integral of order α and ε is given by φ . An analogous expression is

used for holes with Fermi-Dirac statistics.

Note: See Section 3.2.6: “Bandgap Narrowing” for more information on the effects resulting from bandgap narrowing and their
implementation.

SILVACO International 3-3

ATLAS User’s Manual

Energy Balance Transport Model

A higher order solution to the general Boltzmann Transport Equation consists of an additional
coupling of the current density to the carrier temperature, or energy. The current density expressions
from the drift-diffusion model are modified to include this additional physical relationship. The
electron current and energy flux densities are then expressed as:

= ∇ – µ ∇ψ + ∇ 3-20

δ
= – ∇ –  ---------- 3-21
 

= – ∇ – µ ∇ψ – ∇ 3-22

δ
= – ∇ –  ---------- 3-23
 

where Tn and Tp represent the electron and hole carrier temperatures and Sn and Sp are the flux of
energy (or heat) from the carrier to the lattice. The energy balance transport model includes a number
of very complex relationships and therefore a later section of this chapter has been devoted to this
model.

3.1.4: Displacement Current Equation

For time domain simulation, the displacement current is calculated and included in the structure, log
file and the run time output. The expression for displacement current is given as:

∂ 
ε  . 3-24
∂ 

3-4 SILVACO International

 Physics

3.2: Basic Theory of Carrier Statistics

3.2.1: Fermi-Dirac and Boltzmann Statistics
Electrons in thermal equilibrium at temperature TL with a semiconductor lattice obey Fermi-Dirac
statistics. That is the probability f(ε) that an available electron state with energy ε is occupied by an
electron is:
1
( ε ) = ------------------------------------------ 3-25
ε – 
 
 

where ΕF is a spatially independent reference energy known as the Fermi level and is Boltzmann’s
constant.
In the limit, ε - Equation 3-25 can be approximated as:

 – ε
(ε) =   3-26
 

Statistics based on the use of Equation 3-26 are referred to as Boltzmann statistics [49, 50]. The use of
Boltzmann statistics instead of Fermi-Dirac statistics makes subsequent calculations much simpler.
The use of Boltzmann statistics is normally justified in semiconductor device theory, but Fermi-Dirac
statistics are necessary to account for certain properties of very highly doped (degenerate) materials.
The Fermi-Dirac statistics have been implemented in ATLAS in a similar form to Boltzmann statistics.
The remainder of this section outlines derivations and results for the simpler case of Boltzmann
statistics which are the default in ATLAS. You can have ATLAS use Fermi-Dirac statistics by
specifying the FERMIDIRAC parameter in the MODEL statement.

3.2.2: Effective Density of States

Integrating the Fermi-Dirac statistics over a parabolic density of states in the conduction and valence
bands, whose energy minimum is located at energies and respectively, yields the following
expressions for the electron and hole concentrations:

 – 
3-27
⁄  


 – 
⁄   3-28


where η is referred to as the Fermi-Dirac integral of order . If Equation 3-26 is a good

approximation, then Equations 3-27 and 3-28 can be simplified to

 – 
  3-29
 

 – 
  3-30
 

which are referred to as the Boltzmann approximation.

SILVACO International 3-5

ATLAS User’s Manual

and are referred to as the effective density of states for electrons and holes and are given by:

 π 
( )     NC300 3-31
   
 

3 3
--- ---
 2π * 2 2
( ) = 2  ------------------------------ =  --------- NV300 3-32
 2   300
 
where NC300 and NV300 are user-definable on the MATERIAL statement as shown in Table 3-1.
In some circumstances, the lattice temperature, , is replaced by the electron temperature, , in
Equation 3-31 and hole temperature, , in Equation 3-32.

Table 3-1. User-Definable Parameters for the Density of States

Statement Parameter Default Units

MATERIAL NC300 2.8×1019 cm-3

MATERIAL NV300 1.04×1019 cm-3

3.2.3: Intrinsic Carrier Concentration

Multiplying Equations 3-29 and 3-30 yields:
2
= 3-33

where is the intrinsic carrier concentration and is given for Boltzmann statistics by:

 – 
= exp  --------------- 3-34
 2 

is the band-gap energy.

For intrinsic (undoped) material, By equating Equations 3-29 and 3-30 and solving for ΕF yields:

+  
= = – ψ = ------------------------ +  – ------------ 1  -------- 3-35
2 2  

where Εi is the Fermi level for intrinsic doped silicon, and ψi is the intrinsic potential. Equation 3-35
defines the intrinsic potential under non-equilibrium conditions also. As indicated previously, for
ATLAS the ψ used in Equation 3-1 is the intrinsic potential.
The electron and hole concentrations can be expressed in terms of the intrinsic carrier concentration
as:

(ψ – φ )
= exp ------------------------- 3-36

3-6 SILVACO International

 Physics

– (ψ – φ )
= exp ---------------------------- 3-37

where ψ is the intrinsic potential and φ is the potential corresponding to the Fermi level (i.e., φ).
The expression for intrinsic carrier concentration, , can be generalized to Fermi-Dirac statistics
using Equations 3-27 and 3-28. Specifying the [Link] parameter in the MODELS statement will
cause ATLAS to calculate using Fermi-Dirac statistics.

3.2.4: Evaluation of Fermi-Dirac Integrals

Most programming languages have a built in exponential and logarithmic function, making it
straightforward to implement the Boltzmann approximation. Evaluation of the Fermi-Dirac integrals
are more problematic. There are simple inaccurate approximations available. ATLAS, however, uses a
Rational Chebyshev approximation that gives results close to the exact values. The Fermi-Dirac
– –
integrals of order are calculated and at worst differ from the exact values by only 1 part in

3.2.5: The Energy Bandgap

The temperature dependence of the bandgap energy is modeled in ATLAS as follows [9]:

EGALPHA  
 
( ) = ( ) – ----------------------------------------------- = EG300 + EGALPHA ---------------------------------------- – --------------------------------------- 3-38
+ EGBETA + EGBETA + EGBETA

You can specify EG300, EGALPHA and EGBETA parameters in the Material statement (see Table 3-2).

Table 3-2. User-Specifiable Parameters for Equation 3-38

Statement Parameter Default Units

MATERIAL EG300 1.08 eV

MATERIAL EGALPHA 4.73×10-4 eV/K

MATERIAL EGBETA 636 K

MATERIAL NC300 2.8×1019 cm-3

MATERIAL NV300 1.04×1019 cm-3

The default values are material dependent, which can be found in Appendix B: “Material Systems”.
Table 3-2 displays the defaults for Silicon only.

3.2.6: Bandgap Narrowing

In the presence of heavy doping, greater than 1018cm-3, experimental work has shown that the pn
product in silicon becomes doping dependent [3]. As the doping level increases, a decrease in the
bandgap separation occurs, where the conduction band is lowered by approximately the same amount
as the valence band is raised. In ATLAS this is simulated by a spatially varying intrinsic
concentration nie defined according to Equation 3-39.

2 2 ∆
= exp  ----------- 3-39

SILVACO International 3-7

ATLAS User’s Manual

Bandgap narrowing effects in ATLAS are enabled by specifying the BGN parameter of the MODELS
statement. These effects may be described by an analytic expression relating the variation in bandgap,
∆ , to the doping concentration, N. The expression used in ATLAS is from Slotboom and de Graaf
[109]:

 
 
 
 
 2 
∆ = BGN • E  -------------------- +  
-------------------- + BGN • C  3-40
 BGN • N  BGN • N  
 
 
 
 
You can specify the BGN.E, BGN.N, and BGN.C parameters in the MATERIAL statement. The default
values from Slotboom [109] and Klaassen [24] are shown in Table 3-3-3. The Klaassen defaults will be
used if you specify the [Link] parameter in the MODELS statement. Otherwise, the Slotboom values
will be used (by default). You can select a second set of defaults by specifying [Link].

Table 3-3. User-Definable Parameters of Bandgap Narrowing Model

Statement Parameter Defaults Defaults Units
(Slotboom) (Klaassen)

MATERIAL BGN.E 9.0×10-3 6.92×10-3 V

MATERIAL BGN.N 1.0×1017 1.3×1017 cm-3

MATERIAL BGN.C 0.9 0.5 —

The variation in bandgap is introduced to the other physical models by subtracting the result of
Equation 3-40 from the bandgap, Eg. In addition an adjustment is also made to the electric field terms
in the transport models as described earlier. The adjustment takes the form:

= – ∇ ψ + ------------  3-41
 

= – ∇ ψ – ------------  3-42
 

The variation in the energy bandgap is also applied partially to the electron affinity, χ . The effective
electron affinity, χ given as follows:

χ = χ+∆ × ASYMMETRY 3-43

where ASYMMETRY is a user-specifiable asymmetry factor. You can specify the value of the asymmetry
factor using the ASYMMETRY parameter of the MATERIAL statement.

3-8 SILVACO International

 Physics

Note: In addition to this in-built model for bandgap narrowing, ATLAS allows you to use its C-INTERPRETER module. You
can write an external file with C-code that defines an equation for bandgap narrowing. The filename is specified with the
[Link] parameter in the MODEL statement. See Appendix A: “C-Interpreter Functions” for more information on C-
INTERPRETER.

3.2.7: The Universal Bandgap Narrowing Model

A universal bandgap narrowing model has been suggested [163]. This model given in Equation 3-44
below relates the change in bandgap to the material effective masses, and , and the static
dielectric constant ε . The UBGN.C and UBGN.B parameters are fitting parameters to define in the
MATERIAL statement. To enable the model, specify UBGN on the MODEL statement.
1
– ---
ε ε 4
∆ = – UBGN · C  + UBGN · B  3-44
 

3.2.8: Passler's Model for Temperature Dependent Bandgap

An alternative temperature dependent bandgap model by Passler [158] can be enabled by the
PASSLER parameter of the MODELS statement.
The bandgap is given by

A · PASSLER ⋅ T · PASSLER
( ) = ( ) – --------------------------------------------------------------
- 3-45

P · PASSLER
 ⋅
 ----------------------------
1 ⁄ P · PASSLER 
 - + – 
 T · PASSLER
 

where is the lattice temperature dependent bandgap, is the lattice temperature, [Link],
[Link] and [Link] are user-specified parameters on the MATERIAL statement and is
given by

A · PASSLER ⋅ T · PASSLER
( ) = EG300 – --------------------------------------------------------------
- 3-46

P · PASSLER
 ⋅ 300  1 ⁄ P · PASSLER
 ---------------------------- 
 - + – 
 T · PASSLER
 

where EG300 is a user specified parameter on the MATERIAL statement.

SILVACO International 3-9

ATLAS User’s Manual

3.3: Space Charge from Incomplete Ionization, Traps, and Defects

Poisson’s Equation (Equation 3-1) including the carrier concentrations, the ionized donor and acceptor
impurity concentrations and , charge due to traps and defects, QT, has the form:

 – + – 
( ε∇ψ ) =  – + + 3-47

In ATLAS the default is to assume full impurity ionization (i.e., and - ), and to
set QT equal to zero.

ATLAS also provides the options of accounting for incomplete ionization of impurities and accounting
for additional charge associated with traps and defects.

3.3.1: Incomplete Ionization of Impurities

ATLAS can account for impurity freeze-out [4] with appropriate degeneracy factors GCB and GVB for
conduction and valence bands. The ionized donor and acceptor impurity concentrations are then given
by:

+ = ------------------------------------------------------------------------------------- 3-48
ε –( + EDB )
1 + GCB  
 ------------------------------------------------
 

–
= ------------------------------------------------------------------------------ 3-49
 + EAB – ε 
+ GVB  ----------------------------------------
 
 

where EDB is the donor energy level, EAB is the acceptor energy level and ND and NA are net
compensated n-type and p-type doping, respectively. Net compensated doping is defined as follows:
If
≡ 3-50

then
= and = 3-51

Otherwise
= and = 3-52

The INCOMPLETE parameter of the MODELS statement is used to select incomplete ionization and the
parameters.

3-10 SILVACO International

 Physics

Table 3-4. User-Specifiable Parameters for Equations 3-48 and 3-49

Statement Parameter Units

MATERIAL GCB

MATERIAL EDB eV

MATERIAL GVB

MATERIAL EAB eV

To properly handle incomplete ionization in silicon for high doping levels, a new incomplete ionization
model has been added.
The models that form incomplete ionization of impurities given by Equations 3-48 and 3-49 give good
physical results for low to moderately doped semiconductors. For heavily (greater than 3×1018/cm3)
doped semiconductors, these models fail to predict experimental results of complete ionization. For
silicon, an optional model has been introduced that better matches experimental results. This model is
set by the IONIZ parameter of the MODELS statement.
In this model, the activation energies of the dopants in Equations 3-48 and 3-49 have been modified for
doping dependence as given in the following equations:

EDB × 3-53
EAB × 3-54
At doping concentrations above 3×1018 cm-3,
the model predicts complete ionization. At doping
concentrations between 10 cm and 3×10 cm-3, the model predicts a linearly interpolated value
18 -3 18
between the above expressions and complete ionization.

3.3.2: Low Temperature Simulations

In conjunction with Fermi-Dirac statistics and impurity freeze-out, ATLAS simulates device behavior
under low operating temperatures. In general, simulations can be performed at temperatures as low
as 50K without loss of quadratic convergence. Below this temperature, carrier and ionization statistics
develop sharp transitions which cause slower convergence. Since many more iterations will probably
be required if temperatures below 50K are specified, the ITLIMIT parameter of the METHOD statement
should be increased.
Due to the limited exponent range on some machines, ATLAS can have trouble calculating the quasi-
Fermi level of minority carriers. As the temperature decreases, the intrinsic carrier concentration also
decreases. In quasi-neutral regions, the minority carrier concentration can easily underflow. Such
situations were handled in the past by setting these concentrations to zero. This method does not allow
an accurate subsequent calculation of minority carrier quasi-Fermi levels. In order to accurately

calculate quasi-Fermi levels, majority carrier concentration and the relation, 2 is used to obtain

minority carrier concentrations in case of an underflow. Despite these efforts, spurious glitches are
occasionally observed at low temperatures in the minority quasi-Fermi levels.

SILVACO International 3-11

ATLAS User’s Manual

3.3.3: Traps and Defects

Semiconductor materials exhibit crystal flaws which can be caused by dangling bonds at interfaces or
by the presence of impurities in the substrate. The presence of these defect centers, or traps, in
semiconductor substrates may significantly influence the electrical characteristics of the device. Trap
centers, whose associated energy lies in a forbidden gap, exchange charge with the conduction and
valence bands through the emission and recombination of electrons. The trap centers change the
density of space charge in semiconductor bulk and influence the recombination statistics.
Device physics has established the existence of three different mechanism which add to the space
charge term in Poissons’ equation, in addition to the ionized donor and acceptor impurities. These are
interface trapped charge, interface trap states and bulk trap states. Interface trapped charge is
modelled as a sheet of charge at the interface and therefore is controlled by the interface boundary
condition. Interface traps and bulk traps will add space charge directly into the right hand side of
Poisson’s equation. This section describes the definition of bulk trap states and the implementation of
these bulk trap states into ATLAS for both steady state and transient conditions.
Semiconductor flaws have two possible states which are called “empty” and “filled”. When empty a
flaw has a particular cross-section for capturing an electron. A flaw can only either emit or capture an
electron. When the charge on the center has been changed by by addition of an electron, a flaw is
filled, and has a new cross section for hole capture. Two basic types of trap have been found to exist.
These are donor-like and acceptor-like traps. The charge contained within each type of trap will
depend upon whether or not an electron or hole fills the trap.
A donor-type trap can be either positive or neutral like the donor. An acceptor-type trap can be either
negative or neutral like the acceptor. A donor-like trap is positively charged (ionized) when empty and
neutral when filled (with an electron). An empty donor-type trap, which is positive, can capture an
electron or emit a hole. A filled donor-type trap, which is neutral, can emit an electron or capture a
hole. An acceptor-like trap is neutral when empty and negatively charged (ionized) when filled (with
an electron). A filled acceptor-like trap can emit an electron or capture a hole. An empty acceptor-like
trap can capture an electron or emit a hole. Unlike the donor, donor-like traps usually lie near the
valence band. Likewise, acceptor-like traps usually lie near the conduction band.
Figure 3-1 shows the terminology used within ATLAS to define the type of trap. The position of the
trap is defined relative to the conduction or valence bands using [Link] so for instance, an acceptor
trap at 0.4eV would be 0.4eV below the conduction band.

3-12 SILVACO International

 Physics

EC
[Link] for acceptor traps

EtA

EtD
[Link] for donor traps

EV

Figure 3-1: Definition of the trap energy level for acceptor and donor traps in reference to the
conduction and valence band edges.
Calculation of Trapped Charge in Poisson’s Equation
The total charge caused by the presence of traps is added into the right hand side of Poisson’s
equation. The total charge value is defined by:
= ( – ) 3-55

where nt and pt are the densities of trapped charge for donor-like and acceptor-like traps respectively.
The trapped charge depends upon the trap density DENSITY and its probability of occupation, . For
donor-like and acceptor-like traps, respectively, the trapped charge is calculated by the equations:
= DENSITY × 3-56

= DENSITY × 3-57

In the case where multiple traps at multiple trap energy levels are defined the total charge becomes:

= ∑ α
, = ∑ β
3-58
α=1 β=1

where k is the number of acceptor-like traps and m is the number of donor-like traps.
The probability of occupation assumes that the capture cross sections are constant for all energies in a
given band and follows the analysis developed by Simmons and Taylor [5]. The probability of
occupation is defined by the following equations for donor and acceptor-like traps.
SIGN +
= -------------------------------------------------------------------------------------------- 3-59
SIGN + SIGP + +

SILVACO International 3-13

ATLAS User’s Manual

SIGP +
= -------------------------------------------------------------------------------------------- 3-60
SIGN + SIGP + +

where SIGN and SIGP are the carrier capture cross sections for electrons and holes respectively, and
are the thermal velocities for electrons and holes and the electron and hole emission rates, and
, are defined by:

E . LEVEL –
= [Link] SIGN ------------------------------------------- 3-61

– E .LEVEL
1
= ------------------------------------------ SIGP ------------------------------------------- 3-62
[Link]

where is the intrinsic Fermi level position, [Link] is the energy level in the bandgap of each
discrete trap center and [Link] is the degeneracy factor of the trap center. The latter term takes
into account that spin degeneracy will exist, that is the “empty” and “filled” conditions of a flaw will
normally have different spin and orbital degeneracy choices.

Table 3-5. User-Specifiable Parameters for Equations 3-56 to 3-62

Statement Parameter Units

TRAP [Link] eV

TRAP DENSITY cm-3

TRAP [Link]

TRAP SIGN cm2

TRAP SIGP cm2

Trap Implementation into Recombination Models

To maintain self-consistency, you need to take into account that electrons are being emitted or
captured by the donor and acceptor-like traps. Therefore, the concentration of carriers will be affected.
This is accounted for by modifying the recombination rate in the carrier continuity equations.
The standard SRH recombination term (see “Shockley-Read-Hall (SRH) Recombination” section on
page 3-64) is modified as follows:

∑ α ∑ β
3-63
α β
where is the number of donor-like traps, is the number of acceptor-like traps, and the function is:

–
= --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 3-64

– E · LEVEL
  E · LEVEL – 
TAUN   TAUP DEGEN · FAC  ------------------------------------------------
DEGEN · FAC    
 

3-14 SILVACO International

 Physics

The electron and hole lifetimes TAUN and TAUP are related to the carrier capture cross sections SIGN
and SIGP through the equations:

TAUN = --------------------------------------------------------- 3-65

SIGN DENSITY

TAUP = ---------------------------------------------------------- 3-66

SIGP DENSITY

The thermal velocities and are calculated from the following electron and hole effective masses.

⁄
= ( ⁄ ) 3-67

⁄
= ( ⁄ ) 3-68

To specify the effective masses directly, use the [Link] and [Link] parameters from the MATERIAL
statement. If [Link] or [Link] or both are not specified, the density of states effective mass is
extracted from the density of states ( or ) using Equations 3-31 and 3-32. In the case of silicon if
[Link] or [Link] are not specified, the effective masses are calculated from:
–
× 3-69

– –
× – × 3-70

where is the lattice temperature.

Table 3-6. User-Specifiable Parameters for Equations 3-64- 3-66

Statement Parameter Units

TRAP TAUN s

TRAP TAUP s

The TRAP statement activates the model and is used to:

• Specify the trap type DONOR or ACCEPTOR
• Specify the energy level [Link] parameter
• Specify the density of the trap centers DENSITY
• Specify the degeneracy factor [Link]
• Specify either the cross sections, SIGN and SIGP, or the electron and hole lifetimes TAUN and TAUP

SILVACO International 3-15

ATLAS User’s Manual

Trap-Assisted Tunneling
Trap-Assisted Tunneling models the trap-to-band phonon-assisted tunneling effects for Dirac wells. At
high electric fields, tunneling of electrons from the valence band to the conduction band through trap
or defect states can have an important effect on the current.

Trap-assisted tunneling is modeled by including field-effect enhancement terms [113] (Γ and

Γ ) in the trap lifetimes in Equation 3-64. These enhancement terms modify the lifetimes so that
they include the effects of phonon-assisted tunneling on the emission of electrons and holes from a
trap. This model is enabled by specifying [Link] in the MODELS statement.
The recombination term for traps becomes:

–
,
= ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
– E LEVEL E LEVEL
-
 ·   · 
TAUN TAUP
  DEGEN · FAC  ------------------------------- 3-71
DEGEN · FAC    
Γ Γ

The field-effect enhancement term for electrons is given by:

∆ ∆ ⁄ 
Γ ∫ 
–

3-72

While the field-effect enhancement term for hole is:

∆ ∆ ⁄ 
Γ ∫ 
–

3-73

where is the integration variable, ∆ is the energy range where tunneling can occur for electrons,
∆ is the tunneling energy range for holes, and and are defined as:

MASS · TUNNEL ∆
3-74
h

MASS · TUNNEL ∆
3-75
h
h is the reduced Planck's constant, is the rest mass of an electron and [Link] is the effective
mass You can specify [Link] by setting the [Link] parameter in the MODELS statement.
Table 3-7 shows the user-specifiable parameter for Equations 3-74 and 3-75.

Table 3-7. User-Specifiable Parameters for Equations 3-74 and 3-75

Statement Parameter Default Units

MATERIAL [Link] 0.25

3-16 SILVACO International

 Physics

Poole-Frenkel Barrier Lowering for Coulombic Wells

The Poole-Frenkel barrier lowering effect enhances the emission rate for trap-to-band phonon-assisted
tunneling as well as pure thermal emissions at low electric fields. The Poole-Frenkel effect occurs
when the Coulombic potential barrier is lowered due to the electric field and only occurs in traps with
a Coulombic potential (such as traps that are neutral when filled).
The following interactions are Coulombic.
• Between an empty (positive) donor-type trap and an electron.
• Between a filled (negative) acceptor-type trap and a hole.
The following interactions are short-range (Dirac).
Between a filled (neutral) donor-type trap and a hole.
Between an empty (neutral) acceptor-type trap and an electron.
The Poole-Frenkel effect is modeled by including field-effect enhancement terms for Coulombic wells
(Γ and Γ ) and thermal emission (χ ) [135] in the trap lifetimes in Equation 3-64. The
model also includes the trap-assisted tunneling effects in the Dirac wells. To enable this model, specify
[Link] in the MODELS statement.
The recombination term for traps now becomes:
2
–
= --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 3-76
– E · LEVEL E LEVEL
TAUN   TAUP  · 
  DEGEN · FAC  -------------------------------
DEGEN · FAC
χ Γ   Γ  

2
–
= --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 3-77
– E · LEVEL E LEVEL
TAUN   TAUP  · 
  DEGEN · FAC  -------------------------------
DEGE N · FAC    
Γ χ Γ

where the Poole-Frenkel thermal emission enhancement factor, χ , is defined as:

∆ 
χ   3-78
 

∆ is the barrier lowering term for a Coulombic well (see Equation 3-79 ).

∆ 3-79
πε

The Coulombic field-enhancement terms, Γ and Γ , are defined as:

3-80

SILVACO International 3-17

ATLAS User’s Manual

∆ ∆ ⁄ ∆  ⁄ 
Γ ∫ 

– –
 ∆




3-81
∆
∆

Transient Traps
In the time domain the acceptor and donor traps do not reach equilibrium instantaneously but require
time for electrons to be emitted or captured. This is taken account of inside ATLAS by solving an
additional differential rate equation whenever a transient simulation is performed. The probability of
occupation becomes linked to the time dependent emission of electrons through the following rate
equation [52]:

------ ( DENSITY , ) = ( )– ( ) 3-82

where and are the recombination rates of electrons and holes respectively. The terms
describing the emission and capture processes and take form:

 [Link] – 
= DENSITY SIGN (1 – )– [Link]  ------------------------------------------- 3-83
 

1  – E . LEVEL
= DENSITY SIGP (1 – ) – ------------------------------------------  ------------------------------------------- 3-84
[Link]  

A transient trap simulation using this model is more time consuming than using the static model but
gives a much more accurate description of the device physics. It may sometimes be acceptable to
perform transient calculations using the static trap distribution and assume that traps reach
equilibrium instantaneously. If this is the case, a flag (FAST) on the TRAP statement will neglect the
trap rate equation from the simulation.

3-18 SILVACO International

 Physics

3.4: The Energy Balance Transport Model

The conventional drift-diffusion model of charge transport neglects non-local transport effects such as
velocity overshoot, diffusion associated with the carrier temperature and the dependence of impact
ionization rates on carrier energy distributions. These phenomena can have a significant effect on the
terminal properties of submicron devices. As a result ATLAS offers two non-local models of charge
transport, the energy balance, and hydrodynamic models.
The Energy Balance Transport Model follows the derivation by Stratton [77,85] which is derived
starting from the Boltzmann Transport Equation. By applying certain assumptions this model
decomposes into the hydrodynamic model [115,138, 139].
The Energy Balance Transport Model adds continuity equations for the carrier temperatures, and
treats mobilities and impact ionization coefficients as functions of the carrier temperatures rather
than functions of the local electric field.

3.4.1: The Energy Balance Equations

The Energy Balance Transport Model introduces two new independent variables and , the
carrier temperature for electrons and holes. The energy balance equations consist of an energy balance
equation with the associated equations for current density and energy flux .
For electrons the Energy Balance Transport Model consists of:

1 3 ∂ ∗
= --- ⋅ – – ------  λ 
 3-85
2∂

= ∇ – µ ∇ψ + ∇ 3-86

δ
= – ∇ –  ---------- 3-87

and for holes:

1 3 ∂ ∗
= --- ⋅ – – ------  λ  3-88
2∂  

= – ∇ – µ ∇ψ – ∇ 3-89

δ
= – ∇ –  ---------- 3-90
 

where and are the energy flux densities associated with electrons and holes, and µn and µp are
the electron and hole mobilities.
The remaining terms, and , are the thermal diffusivities for electrons and holes as defined in
Equations 3-91 and 3-98 respectively. and are the energy density loss rates for electrons and
holes as defined in Equations 3-114 and 3-115 respectively. and are the thermal conductivities of
electrons and holes as defined in Equations 3-95 and 3-102 respectively.

SILVACO International 3-19

ATLAS User’s Manual

µ
= ------------------- λ ∗ 3-91

(1 ⁄ 2 )( η ) ε –ε
–1  
λ ∗ = --------------------------------,
–1 ⁄ 2( η )
η = ---------------------- = -------
1 ⁄ 2 
3-92

=  µ 2 – λ ∗ µ  3-93

+ ⁄ (η )
ξ
µ = µ  + ξ  ---------------------------------------- 3-94
ξ + ⁄ (η )

= µ   ∆ 3-95
 

ξ + 5 ⁄ 2(η ) ξ + ⁄ (η )
∆ = δ  ξ +  ---------------------------------------- –  ξ +  ---------------------------------------- 3-96
   
ξ + 3 ⁄ 2(η ) ξ + ⁄ (η )

µ
δ = ----------- 3-97
µ

Similar expressions for holes are as follows:

µ
= ------------------λ ∗ 3-98

(1 ⁄ 2)(η ) ε –ε
–1  
λ ∗ = -------------------------------, η = -------------------- = 1 ⁄ 2 3-99
–1 ⁄ 2 ( η ) 

=  µ – λ ∗ µ  3-100

+ 3 ⁄ 2(η )
ξ
µ  
= µ  + ξ  --------------------------------------- 3-101
ξ + 1 ⁄ 2(η )

= µ   ∆ 3-102

3-20 SILVACO International

 Physics

ξ + 5 ⁄ 2(η ) 5 ξ
+ 3 ⁄ 2(η )
∆ = δ ξ + 7  
--- --------------------------------------- – ξ + --- --------------------------------------- 3-103
 2 + 3 ⁄ 2(η )  2 + 1 ⁄ 2(η )
ξ ξ

µ2
δ = ---------- 3-104
µ

If Boltzmann statistics are used in preference to Fermi statistics, the above equations simplify to:

λ ∗ = λ ∗ = 1 3-105

5
∆ = δ =  --- + ξ  3-106
2 

5
∆ = δ =  --- + ξ  3-107
2 

( µ ) ∂µ
ξ = ----------------------- = -------- ----------- 3-108
( ) µ ∂

( µ ) ∂µ
ξ = ----------------------- = ------- ----------- 3-109
( ) µ ∂

The parameters ξn and ξp are dependent on the carrier temperatures. Different assumptions
concerning ξn and ξp correspond to different non-local models. In the high-field saturated-velocity
limit, that corresponds to velocity saturation, the carrier mobilities are inversely proportional to the
carrier temperatures.
ξ = ξ = –1 3-110

and this corresponds to the Energy Balance Transport Model.

If instead the choice ξn=ξp=0 was chosen this would correspond to the simplified Hydrodynamic Model.
The parameters, ξn and ξp, can be specified using the KSN and KSP parameters on the MODELS
statement.
Boundary conditions for n, p, and ψ are the same as for the drift diffusion model. Energy balance
equations are solved only in the semiconductor region. Electron and hole temperatures are set equal to
the lattice temperature on the contacts. On the other part of the boundary, the normal components of
the energy fluxes vanish.
Hot carrier transport equations are activated by the MODELS statement parameters: [Link]
(electron temperature), [Link] (hole temperature), and HCTE (both carrier temperatures).

SILVACO International 3-21

ATLAS User’s Manual

3.4.2: Density of States

The calculation of the effective density of states is modified for the Energy Balance Transport Model.
The electron and hole temperatures replace the lattice temperature in Equations 3-31 and 3-32. For
example:
3
∗ ---
 π 2
=  ------------------------------ =   ( ) 3-111
   
 

∗
 π 
    ( ) 3-112
   
 

3.4.3: Energy Density Loss Rates

The energy density loss rates define physical mechanisms by which carriers exchange energy with the
surrounding lattice environment. These mechanisms include carrier heating with increasing lattice
temperature as well as energy exchange through recombination processes (SRH and Auger) and
generation processes (impact ionization). If the net generation-recombination rate is written in the
form:

= + + – – 3-113

where Rsrh is the SRH recombination rate, are Auger recombination rates related to electrons
and holes, Gn and Gp are impact ionization rates, then the energy density loss rates in Equations 3-85
and 3-88 can be written in the following form:

3 ( – ) 3
= --- ------------------------------------λ + --- λ +  –  3-114
2 TAUREL • EL 2  

3 ( – ) 3
= --- ------------------------------------λ + --- λ +  –  3-115
2 TAUREL • HO 2  

where

λ (η ) ⁄ (η ) 3-116
2

λ (η ) ⁄ (η ) 3-117
2

(and are equal to 1 for Boltzmann statistics), [Link] and [Link] are the electron and hole
energy relaxation times, Eg is the bandgap energy of the semiconductor. The relaxation parameters
are user-definable on the MATERIAL statement, which have their defaults shown in Table 3-8.
The relaxation times are extremely important as they determine the time constant for the rate of
energy exchange and therefore precise values are required if the model is to be accurate. But, this is an
unmeasurable parameter and Monte Carlo analysis is the favored method through which values can
be extracted for the relaxation time.

3-22 SILVACO International

 Physics

It’s also important to take into consideration that different materials will have different values for the
energy relaxation time but within ATLAS, the relaxation time will always default to the value for
silicon.

Table 3-8. User-Specifiable Parameters for Equations 3-114 and 3-115

Statement Parameter Default Units

MATERIAL [Link] 2.5×10-13 s

MATERIAL [Link] 2.5×10-13 s

3.4.4: Temperature Dependence of Relaxation Times

ATLAS doesn’t provide an explicit default model for the temperature dependence of energy relaxation
times. Two methods exist, however, to make the relaxation time a function of carrier energy.
For the first method, use a built-in model by specifying the TRE.T1, TRE.T2, TRE.T3, TRE.W1,
TRE.W2, and TRE.W3 parameters in the MATERIAL statement. Then, activate the electron temperature
(electron energy) dependent energy relaxation time by using [Link] in the MODELS statement.
Electron energy relaxation time will then be:


 TRE · T1 TRE · W1

τ  TRE · T2 TRE · W2 3-118

 TRE · T3 TRE · W3


where:

= 3-119

For TRE.W1<W<TRE.W2 the energy relaxation time varies quadratically between TRE.T1 and TRE.T2.
For TRE.W2<W<TRE.W3 energy relaxation time varies quadratically between TRE.T2 and TRE.T3. The
corresponding parameter for hole energy relaxation time in the MODELS statement is [Link].
Other parameters are listed in Table 3-9.

Table 3-9. User- Specifiable Parameters for Variable Energy Relaxation Time.

Statement Parameter Default Units

MATERIAL TRE.T1 8x10-13 s

MATERIAL TRE.T2 1.92×10-12 s

MATERIAL TRE.T3 1×10-12 s

MATERIAL TRE.W1 0.06 eV

MATERIAL TRE.W2 0.3 eV

MATERIAL TRE.W3 0.45 eV

SILVACO International 3-23

ATLAS User’s Manual

Table 3-9. User- Specifiable Parameters for Variable Energy Relaxation Time.

Statement Parameter Default Units

MATERIAL TRH.T1 1×10-12 s

MATERIAL TRH.T2 1×10-12 s

MATERIAL TRH.T3 1×10-12 s

MATERIAL TRH.W1 1×1010 eV

MATERIAL TRH.W2 1×1010 eV

MATERIAL TRH.W3 1×1010 eV

For the second method, use the C-INTERPRETER to apply a user-defined model for energy relaxation
time as a function of carrier energy. In the MODELS statement, assign the [Link] and [Link]
parameters with the names of external files that contain the user-defined C-INTERPRETER function for
the energy relaxation time. You should also specify [Link] and [Link] flags in the MODELS
statement when using C-I NTERPRETER functions for the energy relaxation times.

3.4.5: Energy Dependent Mobilities

The Energy Balance Transport Model requires the carrier mobility to be related to the carrier energy.
This has been achieved through the homogeneous steady state energy balance relationship that
pertains in the saturated velocity limit. This allows an effective electric field to be calculated, which
causes the carriers in a homogeneous sample to attain the same temperature as at the node point in
the device. The effective electric fields, Eeff,n and Eeff,p, are calculated by solving the equations:

( – )
µ ( ) 3-120
[Link]

( – )
µ ( ) 3-121
[Link]

for and . These equations are derived from the energy balance equations by stripping out all
spatially varying terms. The effective electric fields are then introduced into the relevant field
dependent mobility model. A full description of the available models is given in Section 3.6.1: “Mobility
Modeling”.
ATLAS2D provides a general C-interpreter function allowing you to specify carrier mobility as a
function of Perpendicular field, carrier temperature, Lattice temperature, carrier concentration and
donor and acceptor concentrations.
For electron mobility, the parameter is [Link], which can be set in either the MATERIAL or
MOBILITY statement. The corresponding parameter for hole mobility is [Link], which can be set in
either the MATERIAL or MOBILITY statement. The respective C-functions are endepmun() and
endepmup(). The examples of these functions are provided in ATLAS.

3-24 SILVACO International

 Physics

3.5: Boundary Physics

ATLAS supports several boundary conditions: Ohmic contacts, Schottky contacts, insulated contacts,
and Neumann (reflective) boundaries. Voltage boundary conditions are normally specified at contacts.
Current boundary conditions can also be specified. Additional boundary conditions have been
implemented to address the needs of specific applications. You can connect lumped elements between
applied biases and semiconductor device contacts. A true distributed contact resistance is included to
account for the finite resistivity of semiconductor contacts.

3.5.1: Ohmic Contacts

Ohmic contacts are implemented as simple Dirichlet boundary conditions, where surface potential,
electron concentration, and hole concentrations (ψs, , ) are fixed. Minority and majority carrier
quasi-Fermi potentials are equal to the applied bias of the electrode (i.e., φn=φp=Vapplied). The
potential ψs is fixed at a value that is consistent with space charge neutrality. For example:

3-122

Equation 3-122 can be solved for ψs, , and , since φn and φp are known. If Boltzmann statistics are
used, the substitution of Equations 3-36 and 3-37 into Equation 3-122 will yield:

2
 + –  +  + –

 2
=   4 3-123
  +

3-124

ψ φ φ – 3-125

Note: If you don’t specify a work function, the contacts will be Ohmic regardless of its material.

3.5.2: Schottky Contacts

ATLAS allows you to model Schottky contacts in two different ways [99]. You can either define the
work function of the electrode metal or specify a finite surface recombination velocity at the electrode.
The work function method defines the surface potential at a Schottky contact according to:

ψ AFFINITY – WORKFUN 3-126

where AFFINITY is the electron affinity of the semiconductor material, Eg is the bandgap, NC is the
conduction band density of states, NV is the valence band density of states, and TL is the ambient
temperature. This model is turned on by setting the WORKFUN parameter equal to some value on the
CONTACT statement. In practice, this value is defined as:
WORKFUN = AFFINITY + φ 3-127

SILVACO International 3-25

ATLAS User’s Manual

where φB is the barrier height at the metal-semiconductor interface in eV.

Table 3-10. User-Specifiable Parameters for Equation 3-127

Statement Parameter Units

CONTACT WORKFUN eV

MATERIAL AFFINITY eV

For example, if the Schottky contact were aluminum with a workfunction difference to the silicon of 4.2
eV and a barrier height of 0.7eV, then you would define the Schottky contact with the statement:
CONTACT NAME=GATE WORKFUN=4.9
The Finite Surface Recombination Velocity Method of defining a Schottky contact can be used at a
contact by specifying the [Link] parameter on the CONTACT statement. In this case, the quasi-
Fermi levels, φn and φp, are no longer equal to Vapplied. Instead, these parameters are defined by a
current boundary conditions at the surface [8]:
∆φ
VSURFN ( – )  ----------  3-128
 

∆φ
VSURFP ( – )  ---------- 3-129
 

Table 3-11. User-Specifiable Parameters for Equations 3-128 to 3-129

Statement Parameter Units

CONTACT VSURFN cm/s

CONTACT VSURFP cm/s

where and are the electron and hole currents at the contact, , is the surface electron
concentration and is the surface hole concentrations. The terms, and , are the equilibrium
electron and hole concentrations assuming infinite surface recombination velocity (φn = φp = Vapplied).
If VSURFN and VSURFP aren’t specified on the CONTACT statement, their values will be calculated
using:

ARICHN
VSURFN

3-130

ARICHP
VSURNP 3-131

3-26 SILVACO International

 Physics

Table 3-12. User-Specifiable Parameters for Equations 3-130 to 3-131

Statement Parameter Default Units

MATERIAL ARICHN 110 A/cm2/K2

MATERIAL ARICHP 30 A/cm2/K2

Here, ARICHN and ARICHP are the effective Richardson constants for electrons and holes, taking
account of quantum mechanical reflections and tunneling, and are the conduction and valence
band density of states. The ARICHN and ARICHP parameters are user-definable as shown in
Table 3-12 and NC and NV are functions of the lattice temperature, , according to Equations 3-31
and 3-32.
The Surface Recombination Schottky Model also accounts for field-dependent barrier-lowering
mechanisms. These mechanisms are caused by image forces and possible static dipole layers at the
metal-semiconductor interface [9.]. If the barrier height is defined as:
φ = WORKFUN – AFFINITY 3-132

φ = AFFINITY + ------- – WORKFUN 3-133

With barrier lowering, the amount of energy by which these barrier heights are lowered is defined by:
1
---
2 1⁄2
∆φ = ------------ + ALPHA × 3-134
4πε

Table 3-13. User-Specifiable Parameters for Equation 3-134

Statement Parameter Units

CONTACT ALPHA cm

Here, is the magnitude of the electric field at the interface and ALPHA is the linear, dipole barrier
lowering coefficient. The Barrier Lowering Model can be turned on with the BARRIER parameter in the
CONTACT statement. Typical values of ALPHA may be found in [10]. Note that the term with the square
root dependence on electric field corresponds to the image force, while the linear term corresponds to
the Dipole Effect [9].
Surface recombination is implemented on a triangle-by-triangle basis, which means using the surface
recombination velocity and geometrical data. A recombination component is then calculated for each
triangle so that an element of interest is connected. Using the electric field for each triangle, an
adjusted recombination term can be computed if barrier lowering is incorporated. This is in contrast to
where a single field value for the electrode node is used to compute total recombination value [61].
You can take electron tunneling through the barrier into account by specifying the [Link]
parameter in the CONTACT statement. The electron tunneling current ( ) is given by the Tsu-esaki
Model (Equation 3-135) and is applied to the contact as a current boundary condition.

SILVACO International 3-27

ATLAS User’s Manual

φ
4 π [Link] 0
= – --------------------------------------------------------------------------------------
3 ∫ ( ) ( ) 3-135
0

Here, φ is the barrier height, [Link] is the relative effective mass for electrons, is the
electron rest mass, is Planck’s constant, is Boltzmann’s constant, is the electron charge, and
is given by:

  –  
  ----------------------- 
   
 1+ 
( ) = ln  ------------------------------------------------------- 3-136
  – – 
  ------------------------------------- 
  
1 + 

The transmission probability is given by the [9] approximation for a triangular barrier and
is expressed as:

 
 
 – π ( [Link] ) (φ – ) 
( )   3-137
 
 
 

The [Link] parameter is user-definable in the CONTACT statement.

For example, to specify a Thermoinic Emission Schottky Contact with tunneling and barrier lowering,
the command would be:
CONTACT NAME=GATE [Link] [Link] BARRIER

Note: At room temperature, both methods for Schottky contacts should yield similiar results. If the ambient temperature
changes, then you should apply the Thermoinic Emission Model. Also, when applying either Schottky contact model, we
recommend that you place a fine grid underneath the contact to model the depletion region and that you specify FERMI
statistics in the MODELS statement.

Note: Tunneling is implemented as an addition to Equations 3-128 and 3-129.

Parabolic Field Emission Model

For wide bandgap materials such as SiC the parabolic field emission model is suggested (1,2).
Equation 3-138 describes the current as a function of applied bias voltage . Equation 3-139 describes
the tunneling probability .

3-28 SILVACO International

 Physics

– – 
 
∗ ∞  
∫ ( )
– – – 
3-138
 
 

 ∗ ∗ – 
  – ∗ ∗ – 
( ) =     ( ∗ ) 3-139
 
  ( ∗ )

Here is Richardson's constant, is the barrier height on the metalside, is the semiconductor
doping concentration, is the quasi fermi level energy is the
built in voltage , and is the barrier height on the semiconductor side.

π
Richardson's constant is given by where is the effective mass.

The effective mass is generally taken from the density of states. You can, however, specify a mass for
this tunneling calculation only by the [Link] and [Link] parameters of the MATERIAL
statement. To enable the model, use the PARABOLIC parameter of the CONTACT statement. To turn on
the tunneling components independently for electrons and holes, specify the [Link] and [Link]
parameters of the CONTACT statement. To enable the model, specify the [Link] or [Link]
parameters of the CONTACT statement.

3.5.3: Floating Contacts

A contact that isn’t connected to a current or voltage source and is totally insulated by dielectric is
called a floating contact. ATLAS accounts for floating contacts such as floating gates in EPROM
devices by using a distributed charge boundary condition, which is applied to all nodes of the floating
electrode (see Equation 3-140).

∫ 3-140

where is the electric displacement vector, represents the external surface of the floating gate, and
is the injected charge.
ATLAS performs an integration over the entire surface of the electrode and forces the potential on the
floating nodes to produce the correct total charge on the electrode. The total charge when performing a
simulation is by default zero but can be defined using the SOLVE statement. The total charge may
change if you activate a Charge Injection Model. For more information about this model, see Section
3.6.5: “Gate Current Models”. To define a contact as a floating contact, use:
CONTACT NAME=fgate FLOATING

Note: When specifying a floating contact, use the Newton scheme as the numerical technique. See Chapter 17: “Numerical
Techniques”, Section 17.5.1: “Newton Iteration”.

SILVACO International 3-29

ATLAS User’s Manual

3.5.4: Current Boundary Conditions

In some devices, the terminal current is a multi-valued function of the applied voltage. This means
that for some voltage boundary conditions, the solution that is obtained depends on the initial guess.
An example of this is the CMOS latch-up trigger point. At the trigger point the I-V curve changes from
being flat to vertical and may exhibit a negative slope. The solution will then have three different
solutions of current for one applied bias. The particular solution which the model finishes in will
depend upon the initial conditions.
This trigger point is difficult to determine using a simple voltage boundary condition. In addition, it is
almost impossible to compute any solutions in the negative resistance regime when using voltage
boundary conditions. Some of these problems can be overcome using current boundary conditions.
Calculation of current boundary conditions is activated by the CURRENT parameter in the CONTACT
statement.
The voltage boundary condition should be used in regions where dI/dV is small. The current boundary
condition may be preferable for operating regimes where dI/dV is large. It is common for the negative
resistance regime of a device to have a slope dI/dV very close to 0. Such behavior should be considered
when using a current source to trace out an entire I-V curve. In these cases, use the CURVETRACE
option in ATLAS.

Note: When a current boundary condition has been specified, choose the Newton numerical scheme on the METHOD
statement. You can perform ac small signal analysis with current boundary conditions.

3.5.5: Insulating Contacts

Insulating contacts are contacts that are completely surrounded by insulator. They may be connected
to a voltage source (tied contact) or they may be floating.
Insulating contacts that are connected to a voltage source generally have a work function that dictates
a value for ψs similar to that given by Equation . Electron and hole concentrations within the insulator
and at the insulating contact are forced to be zero (i.e., ).

3.5.6: Neumann Boundaries

Along the outer (non-contact) edges of devices, homogeneous (reflecting) Neumann boundary
conditions are imposed so that current only flows out of the device through the contacts. In the absence
of surface charge along such edges, the normal electric field component becomes zero. Current isn’t
permitted to flow from the semiconductor into an insulating region except via oxide tunneling models.
At the interface between two different materials, the difference between the normal components of the
respective electric displacements must be equal to any surface charge according to:

ˆ ⋅ ε ∇ψ – ˆ ⋅ ε ∇ψ ρ 3-141

where is the unit normal vector, ε and ε are the permittivities of the materials on either side of the
interface and ρs is the sheet charge at the interface.

3-30 SILVACO International

 Physics

3.5.7: Lumped Element Boundaries

ATLAS supports some simple configurations of lumped elements. These are indicated in Figure 3-2.

Figure 3-2: The Lumped elements supported by ATLAS

Lumped elements can be extremely useful when simulating CMOS or MOSFET structures. For
example, the p-tub contact in the CMOS cross-section might be tens or hundreds of microns away from
the active area of an embedded vertical npn bipolar transistor, which may only be 10-20 µm on a side.
If the whole structure were simulated, a tremendous number of grid points (probably more than half)
are essentially wasted in accounting for a purely resistive region of the device.
In the case of a MOSFET substrate, you would not want to include grid points all the way to the back
side of a wafer. In either of these cases, a simple lumped resistance can be substituted [63].
Table 3-14 shows the user-specifiable parameters for Figure 3-2.

Table 3-14. User-Specifiable Parameters for Figure 3-2.

Symbol Statement Parameter Units

C CONTACT CAPACITANCE F/µm

R CONTACT RESISTANCE Ωµm

L CONTACT INDUCTANCE Hµm

Resistance is specified in Ω m, capacitance is specified in F/µm, and inductance is specified in H m.

The following combinations are possible: resistance only, resistance and capacitance, inductance and
resistance and inductance, capacitance and resistance. For the case of inductance or capacitance only,
ATLAS adds a small resistance as well. For more complicated circuit configurations, use the
MIXEDMODE module of ATLAS.

SILVACO International 3-31

ATLAS User’s Manual

Note: Capacitance increases with device width (into the z-plane) while resistance decreases. Except for the case of extremely
large resistances where the arrangement becomes similar to a pure current source, no convergence degradation has been
observed for a lumped element boundary in comparison to a simple Ohmic contact. Transient simulation therefore becomes
easier and is more well-defined.

You should use the simulator to calculate any resistance (or capacitance) components that might be
included as lumped elements. When performing CMOS simulations, you could simulate just the p-tub
with Ohmic contacts at either end. From the plot of terminal current (in A/µm) versus voltage,
resistance can be directly extracted from the slope. Be very careful to consider any three-dimensional
effects (e.g., current spreading) before using a resistance value in further simulations.
When looking at the results of simulation with lumped elements it’s important to distinguish between
the applied voltage ( ) and the internal bias (φ) in the log file produced by ATLAS.

Note: AC small signal analysis can’t be performed when any lumped elements have been specified. AC small signal analysis,
however, can be performed when you specify the resistance in the LOG statement. Also, when lumped elements have been
defined, use the Newton numerical scheme in the METHOD statement.

3.5.8: Distributed Contact Resistance

Since contact materials have finite resistivities, the electrostatic potential isn’t always uniform along
the metal-semiconductor surface. To account for this effect, a distributed contact resistance can be
associated with any electrode. This is in contrast to the lumped contact resistance described in the
previous section.
ATLAS implements this distributed contact resistance by assigning a resistance value to each surface
element associated with the contact. If each surface element is a different length, a different value for
the contact resistance will be assigned. ATLAS calculates a resistance value of R for each surface
element from the value of [Link], as specified on the CONTACT statement. The units of
[Link] are Ωcm2 and R calculated as:

= [Link]
-------------------------------------------- 3-142
WIDTH

where R is the resistance at node , is specified by the [Link] parameter, is the length of
the contact surface segment associated with node and WIDTH is the width of the device. The effect of
the resistance, R is to add an extra equation to be satisfied to node . This equation is given by:

1
------ –  ψ ± ------------ ln ( ⁄ ) – ( + + )= 0 3-143
 

where V is the external applied voltage, Ψi is the surface potential, is the net doping, is the
intrinsic electron concentration, and are the electron, hole and displacement currents at
node i. This equation simply balances the current in and out of the resistor added to each node.
As with the case for lumped elements, ATLAS can print out a value of contact resistance for each
contact in the run time output. Since the actual value depends on the length of each surface segment
for distributed contacts, ATLAS prints out the value of [Link]/WIDTH which is the same for all
contact surface segments. This runtime output is enabled by adding the PRINT option on the MODELS
statement.

3-32 SILVACO International

 Physics

Table 3-15. User-Specifiable Parameters for Equation 3-142

Statement Parameter Units

CONTACT [Link] Ω-cm2

MESH WIDTH µm

Note: AC small signal analysis cannot be performed when any distributed contact resistance has been specified. Also, only the
NEWTON numerical scheme should be chosen on the METHOD statement.

3.5.9: Energy Balance Boundary Conditions

When the Energy Balance Transport Model is applied, special boundary conditions are applied for
carrier temperatures. By default at the contacts, Dirichlet boundary conditions are used for carrier
temperatures:
= = 3-144

You can treat contacts as Neumann (reflective) boundaries with respect to carrier temperature by
specifying REFLECT on the CONTACT statement.
Elsewhere on the boundary, the normal components of the energy fluxes vanish. The boundary
conditions for (ψ, n, p) are the same as for the drift-diffusion model.

SILVACO International 3-33

ATLAS User’s Manual

3.6: Physical Models

3.6.1: Mobility Modeling
Electrons and holes are accelerated by electric fields, but lose momentum as a result of various
scattering processes. These scattering mechanisms include lattice vibrations (phonons), impurity ions,
other carriers, surfaces, and other material imperfections. Since the effects of all of these microscopic
phenomena are lumped into the macroscopic mobilities introduced by the transport equations these
mobilities are therefore functions of the local electric field, lattice temperature, doping concentration,
and so on.
Mobility modeling is normally divided into: (i) low field behavior, (ii) high field behavior, (iii) bulk
semiconductor regions and (iv) inversion layers.
The low electric field behavior has carriers almost in equilibrium with the lattice and the mobility has
a characteristic low-field value that is commonly denoted by the symbol µn0,p0. The value of this
mobility is dependent upon phonon and impurity scattering. Both of which act to decrease the low field
mobility.
The high electric field behavior shows that the carrier mobility declines with electric field because the
carriers that gain energy can take part in a wider range of scattering processes. The mean drift
velocity no longer increases linearly with increasing electric field, but rises more slowly. Eventually,
the velocity doesn’t increase any more with increasing field but saturates at a constant velocity. This
constant velocity is commonly denoted by the symbol νsat. Impurity scattering is relatively
insignificant for energetic carriers, and so νsat is primarily a function of the lattice temperature.
Modeling mobility in bulk material involves: (i) characterizing µn0 and µp0 as a function of doping and
lattice temperature, (ii) characterizing νsat as a function of lattice temperature, and (iii) describing the
transition between the low field mobility and saturated velocity regions.
Modeling carrier mobilities in inversion layers introduces additional complications. Carriers in
inversion layers are subject to surface scattering, extreme carrier-carrier scattering, and quantum
mechanical size quantization effects. These effects must be accounted for in order to perform accurate
simulation of MOS devices. The transverse electric field is often used as a parameter that indicates the
strength of inversion layer phenomena.
You can define multiple non-conflicting mobility models simultaneously. You also need to know which
models are over-riding when conflicting models are defined.
Low Field Mobility Models
The low field carrier mobility can be defined in five different ways.
The first way is use the MUN and MUP parameters to set constant values for electron and hole
mobilities. The second way is by using a look-up table model (CONMOB) to relate the low field mobility
at 300K to the impurity concentration. The third way is by choosing the analytic low field mobility
models, ANALYTIC and ARORA, to relate the low field carrier mobility to impurity concentration and
[Link] fourth way is by choosing a carrier-carrier scattering model (CCSMOB) that relates the
low field mobility to the carrier concentrations and temperature. The fifth way is to use a unified low
field mobility model (KLAASSEN) that relates the low field mobility to donor, acceptor, lattice, carrier-
carrier scattering, and temperature.

3-34 SILVACO International

 Physics

Constant Low Field Mobility Model

In ATLAS, the choice of mobility model is specified on the MODELS statement. The parameters
associated with mobility models are specified on a separate MOBILITY statement. One or more
mobility models should always be specified explicitly. The default is to use constant low field mobilities
within each region of a device. This default model is independent of doping concentration, carrier
densities and electric field. It does account for lattice scattering due to temperature according to:
– TMUN
µ = MUN  ---------- 3-145
0  300

– TMUP
µ 0 = MUP  ---------- 3-146
 300

where is the lattice temperature. The low field mobility parameters: MUN, MUP, TMUN and TMUP can
be specified in the MOBILITY statement with the defaults as shown in Table 3-16.

Table 3-16. User-Specifiable Parameters for the Constant Low Field Mobility Model

Statement Parameter Default Units

MOBILITY MUN 1000 cm2/(V·s)

MOBILITY MUP 500 cm2/(V·s)

MOBILITY TMUN 1.5

MOBILITY TMUP 1.5

Concentration-Dependent Low Field Mobility Tables

ATLAS provides empirical data for the doping dependent low-field mobilities of electrons and holes in
silicon at =300 only. This data is used if the CONMOB parameter is specified in the MODELS
statement. The data that is used is shown in Table 3-17.

Table 3-17. Mobility of Electrons and Holes in Silicon at T=300K

Concentration (cm-3) Mobility (cm2/V⋅s)

Electrons Holes

1.0x1014 1350.0 495.0

2.0x1014 1345.0 495.0

4.0x1014 1335.0 495.0

6.0x1014 1320.0 495.0

8.0x1014 1310.0 495.0

1.0x1015 1300.0 491.1

SILVACO International 3-35

ATLAS User’s Manual

Table 3-17. Mobility of Electrons and Holes in Silicon at T=300K

Concentration (cm-3) Mobility (cm2/V⋅s)

Electrons Holes

2.0x1015 1248.0 487.3

4.0x1015 1200.0 480.1

6.0x1015 1156.0 473.3

8.0x1015 1115.0 466.9

1.0x1016 1076.0 460.9

2.0x1016 960.0 434.8

4.0x1016 845.0 396.5

6.0x1016 760.0 369.2

8.0x1016 720.0 348.3

1.0x1017 675.0 331.5

2.0x1017 524.0 279.0

4.0x1017 385.0 229.8

6.0x1017 321.0 2103.8

8.0x1017 279.0 186.9

1.0x1018 252.0 178.0

2.0x1018 182.5 130.0

4.0x1018 140.6 90.0

6.0x1018 113.6 74.5

8.0x1018 99.5 66.6

1.0x1019 90.5 61.0

2.0x1019 86.9 55.0

4.0x1019 83.4 53.7

6.0x1019 78.8 52.9

8.0x1019 71.6 52.4

1.0x1020 67.8 52.0

2.0x1020 52.0 50.8

3-36 SILVACO International

 Physics

Table 3-17. Mobility of Electrons and Holes in Silicon at T=300K

Concentration (cm-3) Mobility (cm2/V⋅s)

Electrons Holes

4.0x1020 35.5 49.6

6.0x1020 23.6 48.9

8.0x1020 19.0 48.4

1.0x1021 17.8 48.0

The Analytic Low Field Mobility Model

The following analytic function based upon the work of Caughey and Thomas [11, 54] can be used to
specify doping- and temperature-dependent low-field mobilities.

T ALPHAN • CAUG
µ 0 = MU1N • CAUG ⋅  --------------- 3-147
300K

T BETAN • CAUG T ALPHAN • CAUG

MU2N • CAUG ⋅  --------------- – MU1N • CAUG ⋅  ---------------
300K 300K
+ ---------------------------------------------------------------------------------------------------------------------------------
T GAMMAN • CAUG
N DELTAN • CAUG
1 +  --------------- ⋅  -------------------------------------
300K NCRITN • CAUG

T ALPHAP • CAUG
µ 0 = MU1P • CAUG ⋅  --------------- 3-148
300K

T BETAP • CAUG T ALPHAP • CAUG

MU2P • CAUG ⋅  --------------- – MU1P • CAUG ⋅  ---------------
300K 300K
+ ---------------------------------------------------------------------------------------------------------------------------------
T GAMMAP • CAUG
N DELTAP • CAUG
1 +  --------------- ⋅  -------------------------------------
 300K  NCRITP • CAUG

where is the local (total) impurity concentration in cm-3 and is the temperature in degrees
Kelvin.
This model is activated by specifying both the CONMOB and ANALYTIC parameters in the MODELS
statement. The parameters of this model are specified in the MOBILITY statement. The default
parameters are for silicon at = 300K.

SILVACO International 3-37

ATLAS User’s Manual

Table 3-18. User-Specifiable Parameters for Equations 3-147 and 3-148

Statement Parameter Default Units

MOBILITY [Link] 55.24 cm2/(V ⋅ s)

MOBILITY [Link] 49.7 cm2/(V ⋅ s)

MOBILITY [Link] 1429.23 cm2/(V ⋅ s)

MOBILITY [Link] 479.37 cm2/(V ⋅ s)

MOBILITY [Link] 0.0 arbitrary

MOBILITY [Link] 0.0 arbitrary

MOBILITY BETAN. CAUG -2.3 arbitrary

MOBILITY BETAP. CAUG -2.2 arbitrary

MOBILITY [Link] -3.8 arbitrary

MOBILITY [Link] -3.7 arbitrary

MOBILITY [Link] 0.73 arbitrary

MOBILITY [Link] 0.70 arbitrary

MOBILITY [Link] 1.072×1017 cm-3

MOBILITY [Link] 1.606×1017 cm-3

The Arora Model for Low Field Mobility

Another analytic model for the doping and temperature dependence of the low field mobility is
available in ATLAS. This model, which is due to Arora [12], has the following form:
T [Link]
T ALPH [Link] [Link] RA  ----------
 300
µ 0 = [Link]  ---------- + --------------------------------------------------------------------------------------------------------- 3-149
 300 N
1 + -------------------------------------------------------------------------------------------------
T [Link]
NCRITN . ARORA ⋅  ----------
300 

T [Link]
L
T A LPHAP.A RORA [Link]  ----------
L 300
µp0 = MU1P.A RORA  ---------- + ------------------------------------------------------------------------------------------------------------------------------------------------------- 3-150
300 N
1 + --------------------------------------------------------------------------------------------------------------------------------------------
T L [Link]
NCRI [Link] Þ  ----------
300

This model is used if CONMOB and ARORA are specified in the MODELS statement. The parameters of the
model are specified in the MOBILITY statement. The default parameters are for silicon at TL=300K.

3-38 SILVACO International

 Physics

Table 3-19. User-Specifiable Parameters for Equations 3-149 and 3-150

Statement Parameter Default Units

MOBILITY [Link] 88.0 cm2/(V·s)

MOBILITY [Link] 54.3 cm2/(V·s)

MOBILITY [Link] 1252.0 cm2/(V·s)

MOBILITY [Link] 407.0 cm2/(V·s)

MOBILITY [Link] -0.57

MOBILITY [Link] -0.57

MOBILITY [Link] -2.33

MOBILITY [Link] -2.33

MOBILITY [Link] 2.546

MOBILITY [Link] 2.546

MOBILITY [Link] 1.432×1017 cm-3

MOBILITY [Link] 2.67×1017 cm-3

The Albrecht Model for Low Field Mobility in GaN

You can choose to model low field mobility following the work of Albrecht [Link] (160) by specifying
ALBRCT on the MODEL statement or ALBRCT.N or ALBRCT.P or both on the MOBILITY statement for
separate control over electrons and holes. This model is described as follows:

AN · ALBRCT ⋅ N  –3 ⁄ 2
-------------------------------------------- -----------------------------------------
T
3-151
µ( ) NON ⋅ ALBRCT  TON ⋅ ALBRCT
2 –2 ⁄ 3
 ----------------------------------------
T
-  -----------------------------------------
N
 TON ⋅ ALBRCT  NON ⋅ ALBRCT
3⁄2
+ BN · ALBRCT ⋅ N  -----------------------------------------
T
 TON ⋅ ALBRCT
CN ALBRCT
+ -------------------------------------------------------------------
· -
( TIN ⋅ ALBRCT ⁄ T ) –
where is the mobility as a function of doping and lattice temperature, is the total doping
concentration, and is the lattice temperature. [Link], [Link], [Link], [Link],
[Link] and [Link] are user-specifiable parameters on the MOBILITY statement. You can
use a similar expression for holes with the user-defined parameters [Link], [Link],
[Link], [Link], [Link] and [Link].

SILVACO International 3-39

ATLAS User’s Manual

The Carrier-Carrier Scattering Model For Low Field Mobility

The Dorkel and Leturcq Model [13] for low field mobility includes the dependence on temperature,
doping, and carrier-carrier scattering. This model is activated by specifying the CCSMOB parameter of
the MODELS statement. The parameters of the model are specified in the MOBILITY statement. This
model has the form:

 –
 
 
 
µ µ   3-152
  µ  
   
   
 µ  

where L is the lattice scattering, I is the ionized impurity scattering, and C is the carrier-carrier

scattering. Here, µ is defined as:

–
µ 3-153
µ µ

where:
⁄
⋅  
 
µ 3-154
  ( – ⁄
⋅  
)

⁄
[Link]  
 [Link]  
 
µ 3-155

⁄
[Link]  
 [Link]  

µ 3-156

Here, is the total concentration, is the lattice temperature and are the electron and hole
carrier concentrations respectively.
–
( ) ( )– 3-157

The values of the lattice scattering terms, are defined by Equations 3-145 and 3-146.

3-40 SILVACO International

 Physics

Table 3-20. User-Specifiable Parameters for Equations 3-155 and 3-156

Statement Parameter Default Units

MOBILITY [Link] 4.61×1017 cm-3

MOBILITY [Link] 1.0×1017 cm-3

MOBILITY [Link] 1.52×1015 cm-3

MOBILITY [Link] 6.25×1014 cm-3

Klaassen’s Unified Low Field Mobility Model

The model by D. B. M. Klaassen [14, 15], provides a unified description of majority and minority
carrier mobilities. In so doing, it includes the effects of lattice scattering, impurity scattering
(with screening from charged carriers), carrier-carrier scattering, and impurity clustering effects at
high concentration. The model shows excellent agreement between the modeled and empirical data
for:

• majority electron mobility as a function of donor concentration over the range of 1014 cm-3 to 1022
cm-3
• minority electron mobility as a function of acceptor concentration over the range of 1017 cm3 to
1020 cm-3
• minority hole mobility as a function of donor concentration from 1017 cm-3 to 1020 cm-3
• temperature dependence over the range of 70 K to 500 K
The Klaassen Model accounts for a broader set of effects and has been calibrated over a wider range of
conditions than any other of the low field bulk mobility models. This is the recommended model for
both MOS and bipolar simulation and is the default model for silicon when you set MOS2 or BIPOLAR2
in the MODELS statement. You can enable or disable the model by using the KLA parameter in0 the
MODELS statement, or independently for electrons and holes by the KLA.N and KLA.P parameters of
the MOBILITY statement.
The total mobility can be described by its components using Matthiessen’s rule as:
–1 –1 –1
µ = µ +µ 3-158
0

–1 –1 –1
µ 0 = µ +µ 3-159

µn and µp are the total low field electron and hole mobilities, µ and µ are the electron and hole
mobilities due to lattice scattering, µnDAP and µpDAP are the electron and hole mobilities due to donor
(D), acceptor (A), screening (P) and carrier-carrier scattering.
The lattice scattering components, µ and µ are given as:

300 [Link]
µ = [Link]  ---------- 3-160
 

SILVACO International 3-41

ATLAS User’s Manual

300 [Link]
µ = [Link]  ---------- 3-161
 

where TL is the temperature in degrees Kelvin. [Link], [Link], [Link], and

[Link] are user-definable model parameters which can be specified as shown in Table 3-21.

Table 3-21. User-Specifiable Parameters for Equations 3-160 and 3-161

Statement Parameter Default Units

MOBILITY [Link] 1417.0 cm2/(V⋅s)

MOBILITY [Link] 470.5 cm2/(V⋅s)
MOBILITY [Link] 2.285

MOBILITY [Link] 2.247

The impurity-carrier scattering components of the total mobility are given by:

NREF1N . KLA [Link] +

µ = µ , -----------------------  ---------------------------------------------- +µ  ----------------------
- 3-162
  
, ,

[Link] [Link] +
µ = µ , -----------------------  --------------------------------------------- +µ  ----------------------
- 3-163
  
, ,

Table 3-22. User-Specifiable Parameters for Equations 3-162 and 3-163

Statement Parameter Default Units

MOBILITY [Link] 0.68

MOBILITY [Link] 0.719

MOBILITY [Link] 9.68×1016 cm3
MOBILITY [Link] 2.23×1017 cm3

The impurity scattering components, µ ,n and µ ,p, are given by:

2 [Link] – 1.5
[Link]
µ , = -------------------------------------------------------------------------------------------------------  ---------- 3-164
[Link] – [Link]  300

2 [Link] – 1.5
[Link]
µ , = ------------------------------------------------------------------------------------------------------  ---------- 3-165
[Link] – [Link]  300

where is the temperature in degrees Kelvin. [Link] and [Link] are user-defined
parameters shown in Table 3-23, and the other parameters are as described in Tables 3-3-21 and
Table 3-22.

3-42 SILVACO International

 Physics

Table 3-23. User-Specifiable Parameters for Equations 3-164 and 3-165

Statement Parameter Default Units

MOBILITY [Link] 52.2 cm2(V⋅s)

MOBILITY [Link] 44.9 cm2(V⋅s)

The carrier-carrier scattering components, µ and µ , are given by:

[Link] × M [Link] 300 0.5

µ = -----------------------------------------------------------------------------------------------------------  ---------- 3-166
[Link] – M [Link]  T 
L

[Link] × [Link] 300 0.5

µ = ----------------------------------------------------------------------------------------------------------  ---------- 3-167
[Link] – M [Link]  T 
L

The N and N parameters of Equations 3-162 and 3-163 are given by:

= + + 3-168

= + + 3-169

where is the donor concentration in cm-3, is the acceptor concentration in cm-3, n is the electron
concentration in cm-3 and p is the hole concentration in cm-3.
The parameters of Equations 3-164 and 3-165 are given by:

, = + ( ) +  ----------------- 3-170
 ( )

, = + ( ) +  ---------------- 3-171
 ( )

where is the donor concentration in cm-3, is the acceptor concentration in cm-3 and n is the hole
electron concentration in cm-3 and p is the hole concentration in cm-3. The two functions, and
, are functions of the screening factors, and , for electrons and holes. The function, , in
Equations 3-170 and 3-171 are given by:
[Link] [Link]
( ) = 1 – ----------------------------------------------------------------------------------------------------------------------------------- + ---------------------------------------------------------------------------------------------------- 3-172
  [Link] [Link]   [Link] [Link]
[Link] +  °- ----------
-------  °- 300
------- ---------
- 
,  300 ,  

[Link] [Link]
( ) = 1 – ----------------------------------------------------------------------------------------------------------------------------------- + ---------------------------------------------------------------------------------------------------- 3-173
  [Link] [Link]   [Link] [Link]
[Link] +  °- ----------
--------  °- 300
-------- ---------
- 
,  300 ,  

Here, is the temperature in degrees Kelvin, and are the electron and hole masses and the
parameters [Link] through [Link] are user-specifiable model parameters as shown in Table 3-24.

SILVACO International 3-43

ATLAS User’s Manual

Table 3-24. User-Specifiable Parameters for Equations 3-172 and 3-173

Statement Parameter Default Units

MOBILITY [Link] 0.89233

MOBILITY [Link] 0.41372

MOBILITY [Link] 0.19778

MOBILITY [Link] 0.28227

MOBILITY [Link] 0.005978

MOBILITY [Link] 1.80618

MOBILITY [Link] 0.72169

The functions, and , in Equations 3-170 and 3-171 are given by:

[Link] [Link] + [Link] + [Link] --------

( ) = ------------------------------------------------------------------------------------------------------------------------------------- 3-174
[Link] + [Link] + [Link] --------

[Link] [Link] + [Link] + [Link]

F ( P ) = -------------------------------------------------------------------------------------------------------------------------------------- 3-175
[Link] + [Link] + [Link]

where the parameters, [Link] through [Link], are user-specifiable as shown in Table 3-25.

Table 3-25. User-Specifiable Parameters for Equations 3-174 and 3-175

Statement Parameter Default Units

MOBILITY [Link] 0.7463

MOBILITY [Link] 2.2999

MOBILITY [Link] 6.5502

MOBILITY [Link] 2.3670

MOBILITY [Link] -0.8552

MOBILITY [Link] 0.6478

The screening parameters, and , used in Equations 3-174 and 3-175 are given by:

[Link] [Link] – 1
= -------------------------------- + ------------------------------- 3-176
, ,

3-44 SILVACO International

 Physics

[Link] [Link] – 1
= -------------------------------- + ------------------------------- 3-177
, ,

Here, the [Link] and [Link] parameters are user-specifiable model parameters as shown in Table
3-26.

Table 3-26. User-Specifiable Parameters for Equations 3-176 and 3-177

Statement Parameter Default Units

MOBILITY [Link] 2.459

MOBILITY [Link] 3.828

The functions, and , , and are given by the following equations:

×10   
  
3-178
 

×10   
  
3-179
 

2
---
 
3 3
13  1 
, = 3.97 × 10  ------------------ ---------  3-180
 300
 3 
 

2
---
 3  3
13  1   
, = 3.97 × 10  -----------------  ---------  3-181
 3 300 
 

where is the temperature in degrees Kelvin, and are the normalized carrier effective
masses, and n and p are the electron and hole concentrations in cm-3.

Also here, and are the donor and acceptor concentrations in cm-3, is the temperature in
degrees Kelvin, and and are clustering functions given by:

= + ----------------------------------------------------------------------------------- 3-182
[Link]
[Link] +  -----------------------------------------
 

3-183
[Link]  [Link]
 

SILVACO International 3-45

ATLAS User’s Manual

where and are the donor and acceptor concentrations in cm-3 and [Link], [Link],
[Link], and [Link] are user-definable parameters as given in Table 3-27.

Table 3-27. User-Specifiable Parameters for Equations 3-182 and 3-183

Statement Parameter Default Units

MOBILITY [Link] 0.21

MOBILITY [Link] 0.50

MOBILITY [Link] 4.0×1020 cm3

MOBILITY [Link] 7.2×1020 cm3

Note: When the Klaassen low field mobility is used, remember that it has been calibrated to work with Klaassen’s models for
bandgap narrowing, KLAAUG recombination, and KLASRH recombination. These models are described in the Section 3.6.3:
“Carrier Generation-Recombination Models”.

Inversion Layer Mobility Models

To obtain accurate results for MOSFET simulations, you need to account for the mobility degradation
that occurs inside inversion layers. The degradation normally occurs as a result of the substantially
higher surface scattering near the semiconductor to insulator interface.
This effect is handled within ATLAS by three distinct methods:
• a surface degradation model SURFMOB
• a transverse electric field model SHIRAHATA
• specific inversion layer mobility models CVT, YAMAGUCHI, and TASCH
The CVT, YAMAGUCHI, and TASCH models are designed as stand-alone models which incorporate all the
effects required for simulating the carrier mobility.

Lombardi CVT Model

The inversion layer model from Lombardi [16] is selected by setting CVT on the MODEL statement. This
model overrides any other mobility models which may be specified on the MODELS statement. In the
CVT model, the transverse field, doping dependent and temperature dependent parts of the mobility
are given by three components that are combined using Matthiessen’s rule. These components are µ ,
µ and µ and are combined using Matthiessen’s rule as follows:

– – – –
µ µ µ µ 3-184

The first component, µ , is the surface mobility limited by scattering with acoustic phonons:

[Link]
[Link] [Link]
µ = ------------------------------- + ---------------------------------------------------------------- 3-185
[Link] [Link]
⊥ ⊥

3-46 SILVACO International

 Physics

[Link]
[Link] [Link]
µ = ------------------------------ + --------------------------------------------------------------------- 3-186
[Link] [Link]
⊥ ⊥
where is the temperature, ⊥ is the perpendicular electric field, and is the total doping
concentration. You can define the parameters [Link], [Link], [Link], [Link], [Link], [Link],
[Link], and [Link] in the MOBILITY statement (see Table 3-28 for their defaults).
The second component, sr, is the surface roughness factor and is given by:

[Link]
µ = ------------------------------------ 3-187
KN · CVT
⊥

[Link]
µ = ----------------------------------- 3-188
KP · CVT
⊥

The [Link], [Link], [Link], and [Link] parameters are user-definable in the MOBILITY
statement (see Table 3-28 for their defaults).
The third mobility component, µ , is the mobility limited by scattering with optical intervalley
phonons. This component is given by:

T – G [Link]
[Link]  ---------- – [Link]
– [Link] 300
µ = [Link] exp  ----------------------------------- + ------------------------------------------------------------------------------------------------------------------------------------------------------------- 3-189
,  N  [Link]
N
1 +  --------------------------------
C [Link]

[Link]
– -----------------------------------------------------------------------------------------
 C [Link] [Link]
1 + -------------------------------
 N 

T – G [Link]
MUMAX .CVT  ---------- – [Link]
– PCP .CVT 300
µ = [Link] exp  ----------------------------------- + ------------------------------------------------------------------------------------------------------------------------------------------------------------ 3-190
,  N  ALPHP. CVT
N
1 +  --------------------------------
C [Link]

[Link]
– ----------------------------------------------------------------------------------------
C [Link] [Link]
1 +  -------------------------------
 N 

Here, is the total density of impurities and is the temperature in degrees Kelvin.

Table 3-28. User-Specifiable Parameters for Equations 3-185 to 3-190

Statement Parameter Default Units

MOBILITY [Link] 4.75×107 cm/(s)

MOBILITY [Link] 9.925×106 cm/(s)

MOBILITY [Link] 1.74×105

SILVACO International 3-47

ATLAS User’s Manual

Table 3-28. User-Specifiable Parameters for Equations 3-185 to 3-190

Statement Parameter Default Units

MOBILITY [Link] 8.842×105

MOBILITY [Link] 0.333

MOBILITY [Link] 0.333

MOBILITY [Link] 2.0

MOBILITY [Link] 2.0

MOBILITY [Link] 0.125

MOBILITY [Link] 0.0317

MOBILITY [Link] 2.5

MOBILITY [Link] 2.2

MOBILITY [Link] 52.2 cm2/(V·s)

MOBILITY [Link] 44.9 cm2/(V·s)

MOBILITY [Link] 43.4 cm2/(V·s)

MOBILITY [Link] 29.0 cm2/(V·s)

MOBILITY [Link] 1417.0 cm2/(V·s)

MOBILITY [Link] 470.5 cm2/(V·s)

MOBILITY [Link] 9.68×1016 cm-3

MOBILITY [Link] 2.23×1017 cm-3

MOBILITY [Link] 3.43×1020 cm-3

MOBILITY [Link] 6.10×1020 cm-3

MOBILITY [Link] 0.680

MOBILITY [Link] 0.71

MOBILITY [Link] 2.00

MOBILITY [Link] 2.00

MOBILITY [Link] 0.0 cm-3

MOBILITY [Link] 9.23×1016 cm-3
MOBILITY [Link] 5.82×1014 V/s

MOBILITY [Link] 2.0546×1014 V2/s

MOBILITY [Link] 1.0

MOBILITY [Link] 1.0

3-48 SILVACO International

 Physics

Note: The CVT model when activated will also, by default, apply the Parallel Electric Field Mobility Model which is described in
the “Parallel Electric Field-Dependent Mobility” section on page 3-59. In this model, the low field mobility is supplied from the
CVT model.

Extended CVT Model

The CVT model already described has been successfully used in many device simulations. Lately, an
improved version of this model has been proposed by Darwish et al. [17] where two modifications have
been implemented. The first modification is the bulk mobility that is calculated using Klaassen’s
model (see “Klaassen’s Unified Low Field Mobility Model” section on page 3-41) to take into account
coulomb screening effects. The second modification is a new expression for surface roughness is used.
The new model for surface roughness replaces Equations 3-187 and 3-188 with:

·
µ 3-191
γ
⊥

DELP · CVT
µ = ------------------------------------ 3-192
γ
⊥
where:
ALN · CVT ×
γ = AN · CVT + -------------------------------------------------------- 3-193
ETAN · CVT

ALP · CVT ×
γ = AP · CVT + ------------------------------------------------------- 3-194
ETAP · CVT

Here, is the total doping density and is the total carrier concentration .
The Extended CVT Model has been implemented so that the two additional effects can be enabled
independently. By setting both CVT and KLA parameters in the MODELS statement, the CVT model is
enabled with the bulk impurity model of Klaassen and with the original surface roughness model of
CVT. To enable the new surface roughness calculation from Equations 3-193 and 3-194, set the
NEWCVT.N and NEWCVT.P parameters in the MOBILITY statement.
Table 3-29 shows the default parameters for the new equations.

Table 3-29. Default parameters for the surface roughness components of new CVT model.
Parameter Type Default Units
[Link] Real 2.58
[Link] Real 2.18
[Link] Real 6.85×10-21

SILVACO International 3-49

ATLAS User’s Manual

Table 3-29. Default parameters for the surface roughness components of new CVT model.
Parameter Type Default Units
[Link] Real 7.82×10-21
[Link] Real 0.0767
[Link] Real 0.123

Yamaguchi Model
The Yamaguchi Model [18] is selected by setting YAMAGUCHI in the MODELS statement. This model
overrides any mobility model specifications other than the CVT model. The model consists of
calculating the low field, doping dependent mobility. Surface degradation is then accounted for based
upon the transverse electric field before including the parallel electric field dependence.
The low field part of the Yamaguchi Model is given as follows:

µ [Link]
[Link] 3-195
[Link]

µ [Link] 3-196
[Link]
[Link]

where is the net impurity concentration. The equation parameters: [Link], [Link],
[Link], [Link], [Link], and [Link] are user-definable in the MOBILITY statement
(see Table 3-30 for their defaults).
The transverse electric field dependence is accounted for as follows:

–
µ µ ( [Link]
⊥ ) 3-197

–
µ µ ( [Link] ⊥) 3-198

where Ε⊥ is the perpendicular electric field and the equation parameters, [Link] and [Link],
are user-definable in the MOBILITY statement (see Table 3-30 for their defaults).
The final calculation of mobility takes into account the parallel electric field dependence which takes
the form:

–
µ µ – µ
µ µ      
 [Link]  [Link] [Link]  [Link] 3-199

3-50 SILVACO International

 Physics

–
µ µ – µ
µ µ    [Link]    3-200
 [Link]  [Link]  [Link]

where E is the parallel electric field and the equation parameters: [Link], [Link], [Link],
[Link], [Link], and [Link] are user-definable in the MOBILITY statement (see Table 3-30 for
their defaults).

Table 3-30. User-Specifiable Parameters for Equations 3-195 to 3-200

Statement Parameter Default Units

MOBILITY [Link] 350.0

MOBILITY [Link] 81.0

MOBILITY [Link] 3.0×1016 cm-3

MOBILITY [Link] 4.0×1016 cm-3

MOBILITY [Link] 1400.0 cm2/(V·s)

MOBILITY [Link] 480.0 cm2/(V·s)

MOBILITY [Link] 1.54×10-5 cm/V

MOBILITY [Link] 5.35×10-5 cm/V

MOBILITY [Link] 1.036×107 cm/s

MOBILITY [Link] 1.2×107 cm/s

MOBILITY [Link] 4.9×106 cm/s

MOBILITY [Link] 2.928×106 cm/s

MOBILITY [Link] 8.8

MOBILITY [Link] 1.6

The Tasch Model

S-PISCES includes an improved local field-dependent mobility model. This model, which was originally
derived and published by Tasch et. al [44,46] has been designed explicitly for MOSFETs. It defines the
mobility as a function of the perpendicular and parallel electric fields, the interface charge, the lattice
temperature and the doping concentration. To activate this model, is activated use the TASCH
parameter on the MODELS statement. This mobility model is given by the following expressions:

dΓ
µ = Γ +( – 0) 3-201
d

dΓ
µ = Γ +( – 0) 3-202
d

SILVACO International 3-51

ATLAS User’s Manual

where is the transverse electric field and is the transverse electric field at the edge of the
inversion layer. The functions Γ are defined as:

µ ,
Γ = ----------------------------------------------------------------------------------------------------------------- 3-203
µ
,
 +  --------------------------------- ||  BETAN  1 ⁄ BETAN
-
  VSATN  

, µ
Γ = ------------------------------------------------------------------------------------------- 3-204
µ 1 ⁄ BETAP
, ||  BETAP 
 +  ------------------------------
-
  VSATP  

The carrier mobilities µeff,n and µeff,p are defined by three components µph, µsr and µc which are
combined by Mathiessen’s rule according to:

1 1 1 –1
µ = ----------- + --------- + ------ 3-205
µ µ µ

The term µph takes account of the degradation in mobility due to acoustic phonon scattering through
the expressions:

– T MUBN . TAS – 1 ⁄ –1
µ– , =  [Link]  ---------  + ⁄ [Link]  ---------  3-206
  300    300 

– T MUBP . TAS – 1 ⁄ –1
µ– =  [Link]  ---------  + ⁄ [Link]  ---------  3-207
,   300    300 

The function, , is defined as:

–
–
= [Link]  --------- + [Link] 3-208
 300 , ,

–
–1
= [Link]  --------- , + [Link] , 3-209
 300

where:
( + ( RN ⋅TAS – 1 ) ⋅ )
= ------------------------------------------------------------------------------ 3-210
RN ⋅ TAS

( + ( RP ⋅TAS – 1 ) ⋅ )
= ------------------------------------------------------------------------------ 3-211
RP ⋅ TAS

Also, the function , is defined as:

[Link] + [Link]  –1
= [Link]  --------- + [Link] --------
 300 - 3-212
300

3-52 SILVACO International

 Physics

[Link] + [Link]  –1
= [Link]  --------- + [Link] --------- 3-213
 300  300

Mobility degradation due to surface roughness is accounted for by the term µsr which is calculated
according to:
[Link]
µ =  [Link]
----------------------------------- 3-214
,  , 

[Link]
µ , =  [Link]
---------------------------------- 3-215
 ,


The final term, µC, models Coulombic scattering with the expressions:

[Link] ⋅  ---------
 300
µ , = ------------------------------------------------------------------------------------------ 3-216
γ ,
(1 + γ , ) – ---------------------------------
(1 + γ , )

[Link] ⋅  ---------
 300
µ , = ------------------------------------------------------------------------------------------ 3-217
γ ,
(1 + γ , ) – --------------------------------
-
(1 + γ )
,

Here:
[Link]
[Link]
γ , = ------------------------------ ⋅  --------- 3-218
n  300

[Link]
[Link]
γ , = ----------------------------- ⋅  --------- 3-219
 300

is the lattice temperature in degrees Kelvin, is the fixed interface charge at the gate dielectric-
silicon interface (cm-2), is the channel acceptor doping concentration in cm-3, is the channel
donor doping concentration in cm-3,
and are the electron and hole concentrations per unit volume
-3
in the inversion layer (cm ). The default parameters within each equation is defined in the MOBILITY
statement. The default parameters are shown in Table 3-30.

Table 3-31. Parameters for Equations 3-201 through 3-219

Statement Parameter Default Units

MOBILITY [Link] 2

MOBILITY [Link] 3

MOBILITY BETAN 2

MOBILITY BETAP 1

SILVACO International 3-53

ATLAS User’s Manual

Table 3-31. Parameters for Equations 3-201 through 3-219

Statement Parameter Default Units

MOBILITY [Link] 1150

MOBILITY [Link] 270

MOBILITY [Link] 2.5

MOBILITY [Link] 1.4

MOBILITY [Link] 3.2×10-9

MOBILITY [Link] 2.35×10-9

MOBILITY [Link] 0.09

MOBILITY [Link] 0.334

MOBILITY [Link] 1.75

MOBILITY [Link] 1.5

MOBILITY [Link] 4.53×10-8

MOBILITY [Link] 3.14×10-7

MOBILITY [Link] -0.25

MOBILITY [Link] -0.3

MOBILITY [Link] 0.0388

MOBILITY [Link] 0.039

MOBILITY [Link] 1.73×10-5

MOBILITY [Link] 1.51×10-5

MOBILITY [Link] 2.449×107

MOBILITY [Link] 1.0×108

MOBILITY [Link] 2

MOBILITY [Link] 1

MOBILITY [Link] 1.1×1021

MOBILITY [Link] 1.4×1018

MOBILITY [Link] 2.0×1019

MOBILITY [Link] 8.4×1016

MOBILITY [Link] 2

MOBILITY [Link] 3.4

3-54 SILVACO International

 Physics

Perpendicular Electric Field-Dependent Mobility

The Watt Model

A surface mobility model derived by [Link] [19] is available in ATLAS. This mobility model is
activated when the parameter SURFMOB is specified on the MODELS statement. The default model
parameters are tuned to describe the measured mobilities in silicon at 300K. You can modify model
parameters by using the MOBILITY statement.
The Watt model takes into consideration the following primary scattering mechanisms in the inversion
layer:
1. Phonon scattering which results primarily from the interaction between two-dimensional
inversion layer carriers and bulk phonons.
2. Surface roughness scattering caused by the interaction between inversion layer carriers and
deviations from ideal planarity at the interface.
3. Charged impurity scattering caused by the interaction between inversion layer carriers and ions
located in the oxide, at the interface, or in the bulk
The phonon and surface roughness components are functions of effective electric field. The charged
impurity component is a function of the channel doping density.
The effective mobilities for electrons and holes are given by Equations 3-220 and 3-221.

1 1 1 [Link]
----------------- = ---------------------------------------------------  ------------------ 3-220
µ , [Link]  
,
1 1 [Link]
+ ---------------------------------------------------  ------------------
[Link]
,
1 – 1 [Link]
+ ---------------------------------------------------    
[Link]    

1 [Link]
----------------- = ---------------------------------------------------  

3-221
µ [Link]
1 [Link]
+ ---------------------------------------------------  

[Link]
1 –1 [Link]
+ ---------------------------------------------------    
[Link]    

Here, is the surface trapped charge density, is the inversion layer charge density and is the
effective electric field given by:
E , = ⊥ + [Link] ( – ⊥) 3-222

E = + [Link] ( – ) 3-223
, ⊥ ⊥

Here, ⊥ is the electric field perpendicular to the current flow and is the perpendicular electric field
at the insulator-semiconductor interface. The equation parameters and their defaults are listed in
Table 3-32.

SILVACO International 3-55

ATLAS User’s Manual

The terms on the right side of Equations 3-220

3-56 SILVACO International

 Physics

The logical parameters [Link].N and [Link].P of the MOBILITY statement can also be used to
enable an additional modification to the Watt model. When these parameters are enabled the effective
normal electric field becomes a function of the depth beneath the silicon/oxide interface according to:
–( y – y )
= ---------------------------------------------------- 3-224
⊥, [Link]

–( y – y )
= --------------------------------------------------- 3-225
⊥, [Link]

where ⊥ is the perpendicular electric field, is the perpendicular electric field at the interface, is
the local y-coordinate and is the y-coordinate of the silicon/oxide interface. The [Link] and
[Link] parameters are user-definable in the MOBILITY statement.

Table 3-33. User-Specifiable Parameters for Equations 3-194 and 3-195

Statement Parameter Default Units

MOBILITY [Link] -1.0×1032 microns

MOBILITY [Link] 1.0×1032 microns

MOBILITY [Link] -1.0×1032 microns

MOBILITY [Link] -1.0×1032 microns

MOBILITY [Link] 1.0×1032 microns

MOBILITY [Link] -1.0×1032 microns

MOBILITY [Link] 1.0×1032 microns

MOBILITY [Link] 1.0×1032 microns

Shirahata’s Mobility Model

The Shirahata Mobility Model [20] is a general purpose MOS mobility model that takes into account
screening effects in the inversion layer as well as improved perpendicular field dependence for thin
gate oxides. In the original paper, the authors present the model as a combination of portions of
Klaassen’s model for low field mobility contributions and an empirically fit expression for the
perpendicular field dependent mobility in the inversion layer. In this implementation, for any given
location the lesser of the low field mobility and the mobility due to the Shirahata Mobility Model is
used. If the Klaassen Low-Field Model is used with the Shirahata model, the lattice scattering term in
the Klaassen model is omitted.
The Shirahata Mobility Model is enabled by the SHI parameter of the MODELS statement or can be
enabled individually for electrons and holes using the SHI.N and SHI.P parameters of the MOBILITY
statement.
The Shirahata models for electrons and holes are given by:

T – [Link]
[Link]  ----------
 300
µ = ----------------------------------------------------------------------------------------------------------------------------------------------- 3-226
[Link] [Link]
⊥ ⊥
+ ---------------------------- + ----------------------------
[Link] [Link]

SILVACO International 3-57

ATLAS User’s Manual

T – [Link]
[Link]  ----------
 300
µ = --------------------------------------------------------------------------------------------------------------------------------------------- 3-227
[Link] [Link]
⊥ ⊥
1 + ---------------------------- + ----------------------------
[Link] [Link]

where ⊥ is the perpendicular electric and the equation parameters: [Link], [Link],
[Link], [Link], [Link], [Link], [Link], [Link], [Link], [Link], [Link]
and [Link] are user-definable in the MOBILITY statement (see Table 3-34 for their defaults).

Table 3-34. User-Specifiable Parameters for Equations 3-226 and 3-227

Statement Parameter Default Units

MOBILITY [Link] 1430.0 cm2/(V·s)

MOBILITY [Link] 500.0 cm2/(V·s)

MOBILITY [Link] 6.3×103 V/cm

MOBILITY [Link] 8.0×103 V/cm

MOBILITY [Link] 0.77×106 V/cm

MOBILITY [Link] 3.9×105 V/cm

MOBILITY [Link] 0.28

MOBILITY [Link] 0.3

MOBILITY [Link] 2.9

MOBILITY [Link] 1.0

MOBILITY [Link] 2.285

MOBILITY [Link] 2.247

Note: If the maximum low field mobility has been user-defined, then it’s important to also define this value inside the Shirahata
model with the [Link] and [Link] parameters in the MOBILITY statement.

3-58 SILVACO International

 Physics

Parallel Electric Field-Dependent Mobility

As carriers are accelerated in an electric field their velocity will begin to saturate when the electric
field magnitude becomes significant. This effect has to be accounted for by a reduction of the effective
mobility since the magnitude of the drift velocity is the product of the mobility and the electric field
component in the direction of the current flow. The following Caughey and Thomas Expression [11] is
used to implement a field-dependent mobility. This provides a smooth transition between low-field and
high field behavior where:

------------------------
1 BETAN
µ ( ) = µ 0 ---------------------------------------------------------------- 3-228
µ BETAN
0
+  -----------------------
 VSATN

BETAP
µ ( ) µ 3-229
µ BETAP
 
 VSATP

Here, is the parallel electric field and µn0 and µp0 are the low field electron and hole mobilities
respectively. The low field mobilities are either set explicitly in the MOBILITY statement or calculated
by one of the low field mobility models. The BETAN and BETAP parameters are user-definable in the
MOBILITY statement (see Table 3-35 for their defaults).
The saturation velocities are calculated by default from the temperature-dependent models [43]:

ALPHAN . FLD
VSATN = -------------------------------------------------------------------------------------------------
THETAN . FLD  ------------------------------------ 3-230
 [Link]

ALPHAP . FLD
VSATP = ------------------------------------------------------------------------------------------------ 3-231
THETAP . FLD  -----------------------------------
 [Link]

but can be set to constant values on the MOBILITY statement using the VSATN and VSATP parameters.
In this case, no temperature dependence is implemented. Specifying the FLDMOB parameter on the
MODELS statement invokes the field-dependent mobility. FLDMOB should always be specified unless one
of the inversion layer mobility models (which incorporate their own dependence on the parallel field)
are specified.
You can invoke a C-INTERPRETER function for the saturation velocities. The [Link] and [Link]
parameters in the MATERIAL statement can be set to provide the filenames of two text files containing
the particular functions. These functions allow you to include the temperature dependence. See
Appendix A: “C-Interpreter Functions” for more details.

SILVACO International 3-59

ATLAS User’s Manual

Table 3-35. User-Definable Parameters in the Field-Dependent Mobility Model

Statement Parameter Default Units

MOBILITY BETAN 2.0

MOBILITY BETAP 1.0

MOBILITY VSATN cm/s

MOBILITY VSATP cm/s

MOBILITY [Link] 2.4×107 cm/s

MOBILITY [Link] 2.4×107 cm/s

MOBILITY [Link] 0.8

MOBILITY [Link] 0.8

MOBILITY [Link] 600.0 K

MOBILITY [Link] 600.0 K

Note: Equation 3-231, which was derived for the drift-diffusion approximation, ensures that velocity overshoot cannot occur. To
model velocity overshoot in silicon, apply the Energy Balance Transport Model. This model follows the above implementation
but with the electric field term replaced by a new “effective” field calculated from the carrier temperature see the following
section for more details.

Note: BLAZE includes a different field dependent mobility model that does simulate velocity overshoot in GaAs. See
Chapter 5: “Blaze: Compound Material 2D Simulator” for more information.

Carrier Temperature Dependent Mobility

The Energy Balance Transport Model allows the carrier mobility to be related to the carrier energy.
This has been achieved through the homogeneous steady state energy balance relationship that
pertains in the saturated velocity limit. This allows an effective electric field to be calculated, which is
the uniform electric field value that causes the carriers in a homogeneous sample to attain the same
temperature as at the node point in the device. The effective electric fields, and , are
calculated by solving the equations:
( – )
µ ( ) 3-232
[Link]

2 3 k ( Tp – TL )
q µp ( E , ) , = --- ------------------------------------------- 3-233
2 [Link]

for and . These equations are derived from the energy balance equations by stripping out all
spatially varying terms. The effective electric fields are then introduced into the relevant field

3-60 SILVACO International

 Physics

dependent mobility model. The [Link] and [Link] parameters can be set on the MODELS
statement and have the default values shown in Table 3-36.

Note: The [Link] and [Link] parameters are distinct from the energy balance relaxation time parameters:
[Link] and [Link]. But, if the [Link] and [Link] parameters are set on the MODELS
statement, then [Link] and [Link] are used in place of [Link] and [Link].

Table 3-36. User-Specifiable Parameters for Equations 3-232 and 3-233

Statement Parameter Default Units

MODELS [Link] 2.5×10-13 s

MODELS [Link] 2.5×10-13 s

Four different models have been implemented into the ATLAS Energy Balance Transport Model which
can be chosen by the parameter EVSATMOD on the MODELS statement. These models shall be described
next.
Setting EVSATMOD=0 implements, the default model for silicon based upon the Caughey-Thomas field-
dependent mobility model in Equation 3-228. The resultant relationship between the carrier mobility
and the carrier temperature is in the forms:
µ 0
µ = --------------------------------------------------------------------- 3-234
------------------------
BETAN BETAN
 
 

µ
0
µ = ------------------------------------------------------------------- 3-235
-----------------------
BETAP BETAP
 
 

BETAN 1 BETAN
= --- ( α BETAN ( – ) 3-236
2
2 BETAN BETAN BETAN 
+ α 2 BETAN ( – ) – 4α ( – ) 

BE TAP 1 BETAP
= --- ( α
2
^ BETAP ( – ) 3-237
2 BETAP BETAP B ETAP 
+ α 2 BETAP ( – ) – 4α ( – ) 

µ
α 3-238
VSATN TAUREL • EL

SILVACO International 3-61

ATLAS User’s Manual

µ 0
3
α = --- --------------------------------------------------------------------- 3-239
2 2
VSATP TAUREL • HO
where µn0 and µp0 are the low field carrier mobilities and VSATN and VSATP are the saturated
velocities for electrons and holes. The VSATN, VSATP, BETAN, and BETAP parameters are user-definable
in the MOBILITY statement. The terms, [Link] and [Link], are the energy relaxation times
for electrons and holes and can be defined in the MATERIAL statement.
Setting EVSATMOD=0 with the additional parameter, [Link], allows you to apply a simplified
form of the above model. This form is:
µ 0
µ = ------------------------------------------------------- 3-240
2 2
1+α ( – )

µ 0
µ = ------------------------------------------------------ 3-241
2 2
1+α ( – )

where µ and µ are again the low field carrier mobilities and are as defined above.
Setting EVSATMOD=1 implements the GaAs carrier temperature dependent mobility model. See
Chapter 5: “Blaze: Compound Material 2D Simulator” for more information about this model.
Setting EVSATMOD=2 will apply the simple velocity limiting model based upon the electric field. In
other words, the temperature dependent mobility is turned off and the standard electric field based
mobility model is applied.

Note: If the YAMAGUCHI or TASCH mobility models are chosen in the MODEL statement, then no energy dependence is
applied. No energy dependence is included in any perpendicular electric field model, such as SHIRAHATA or SURFMOB.

Table 3-37. User-Specifiable Parameters for Equations 3-234– 3-241

Statement Parameter Units

MOBILITY MUN cm2/(V⋅s)

MOBILITY MUP cm2/(V⋅s)

MATERIAL VSATN cm/s

MATERIAL VSATP cm/s

3-62 SILVACO International

 Physics

3.6.2: Mobility Model Summary

Tables 3-3-38 and 3-3-39 shows a brief description of the mobility models.

Table 3-38. Mobility Models Summary

Model Syntax Notes
Concentration Dependent CONMOB Lookup table valid at 300K for Si and GaAs only. Uses simple power law
temperature dependence.
Concentration and ANALYTIC Caughey-Thomas formula. Tuned for 77-450K.
Temperature Dependent
Arora’s Model ARORA Alternative to ANALYTIC for Si.
Carrier-Carrier Scattering CCSMOB Dorkel-Leturq Model. Includes n, N and T dependence. Important when
carrier concentration is high (e.g., forward bias power devices).
Parallel Electric Field Dependence FLDMOB Si and GaAs models. Required to model any type of velocity saturation
effect.
Tasch Model TASCH Includes transverse field dependence. Only for planar devices. Needs very
fine grid.
Watt Model WATT Transverse field model applied to surface nodes only.
Klaassen Model KLA Includes N, T, and n dependence. Applies separate mobility to majority and
minority carriers. Recommended for bipolar devices
Shirahata Model SHI Includes N, E⊥ . An alternative surface mobility model that can be
combined with KLA.
Modified Watt [Link] Extension of WATT model to non-surface nodes. Applies constant
E⊥effects. Best model for planar MOS devices
Lombardi (CVT) Model CVT Complete model including N, T, E// and E⊥ effects.
Good for non-planar devices.
Yamaguchi Model YAMAGUCHI Includes N, E// and E⊥ effects. Only for 300K.

Table 3-39. Mobility Models Summary

CONMOB FLDMOB TFLDMB2 YAMAGUCHI CVT ARORA ANALYTIC CCSMOB SURFACE LATTICE H [Link]

CONMOB [CM] — OK OK YA CV AR AN CC OK OK OK
FLDMOB [FM] OK — TF1 YA CV OK OK OK OK OK OK
TFLDMB2 [TF] OK TF1 — YA CV OK OK TF TF OK OK
YAMAGUCHI [YA] YA YA YA — CV YA YA YA YA NO NO
CVT [CV] CV CV CV CV — CV CV CV CV OK OK
ARORA [AR] AR OK OK YA CV — AR CC OK OK OK
ANALYTIC [AN] AN OK OK YA CV — CC OK OK OK
CCSMOB [CC] CC OK TF YA CV CC CC — OK OK OK
SURFMOB [SF] OK OK TF YA CV OK OK OK — OK OK
LATTICE H [LH] OK OK OK NO OK OK OK OK OK — OK
[Link] [EB] OK OK OK NO OK OK OK OK OK OK 2

SILVACO International 3-63

ATLAS User’s Manual

Table 3-39. Mobility Models Summary

CONMOB FLDMOB TFLDMB2 YAMAGUCHI CVT ARORA ANALYTIC CCSMOB SURFACE LATTICE H [Link]

Key to Table Entries:

MODEL ABBREVIATION = The model that supersedes when a combination is specified. In some cases, but not all, a warning message is issued
when a model is ignored.
OK = This combination is allowed.
NO = This combination isn’t allowed.
NOTES:
1. Uses internal model similar to FLDMOB
2. When models including a parallel electric field dependence are used with energy balance the electric field term
is replaced by a function of carrier temperature.

3.6.3: Carrier Generation-Recombination Models

Carrier generation-recombination is the process through which the semiconductor material attempts
to return to equilibrium after being disturbed from it. If we consider a homogeneously doped
semiconductor with carrier concentrations n and p to the equilibrium concentrations n0 an p0 then at
equilibrium a steady state balance exists according to:

3-242

Semiconductors, however, are under continual excitation whereby n and p are disturbed from their
equilibrium states: n0 and p0. For instance, light shining on the surface of a p type semiconductor
causes generation of electron-hole pairs, disturbing greatly the minority carrier concentration. A net
recombination results which attempts to return the semiconductor to equilibrium. The processes
responsible for generation-recombination are known to fall into six main categories:
• phonon transitions
• photon transitions
• Auger transitions
• surface recombination
• impact ionization
• tunneling
The following sections describes the models implemented into ATLAS that attempts the simulation of
these six types of generation-recombination mechanisms.
Shockley-Read-Hall (SRH) Recombination
Phonon transitions occur in the presence of a trap (or defect) within the forbidden gap of the
semiconductor. This is essentially a two step process, the theory of which was first derived by Shockley
and Read [21.] and then by Hall [124]. The Shockley-Read-Hall recombination is modeled as follows:

–
3-243
  – 
   

where ETRAP is the difference between the trap energy level and the intrinsic Fermi level, TL is the
lattice temperature in degrees Kelvin and TAUN0 and TAUP0 are the electron and hole lifetimes. This
model is activated by using the SRH parameter of the MODELS statement. The electron and hole

3-64 SILVACO International

 Physics

lifetime parameters, TAUN0 and TAUP0, are user-definable in the MATERIAL statement. The default
values for carrier lifetimes are shown in Table 3-40. Materials other than silicon will have different
defaults. A full description of these parameters are given in Appendix B: “Material Systems”.

Table 3-40. User-Specifiable Parameters for Equation 3-243

Statement Parameter Default Units

MATERIAL ETRAP 0 eV

MATERIAL TAUN0 1×10-7 s

MATERIAL TAUP0 1×10-7 s

Note: This model only presumes one trap level which, by default, is ETRAP=0 and it corresponds to the most efficient
recombination centre. If the TRAP statement is used to define specific trap physics then separate SRH statistics are
implemented as described earlier in “Trap Implementation into Recombination Models” section on page 3-14.

SRH Concentration-Dependent Lifetime Model

The constant carrier lifetimes that are used in the SRH recombination model above can be made a
function of impurity concentration [22,93,110] using the following equation:

–
3-244
τ   τ – 
   

where:
TAUN0
τ = ------------------------------------------------- 3-245
+ ⁄ ( NSRHN )

TAUP0
τ = ------------------------------------------- 3-246
+ ⁄ ( NSRHP )

Here, is the local (total) impurity concentration. The TAUN0, TAUP0, NSRHN, and NSRHP parameters
can be defined on the MATERIAL statement (see Table 3-41 for their default values). This model is
activated with the CONSRH parameter of the MODELS statement.

Table 3-41. User-Specifiable Parameters for Equations 3-244 to 3-246

Statement Parameter Default Units

MATERIAL TAUN0 1.0×10-7 s

MATERIAL NSRHN 5.0×1016 cm-3

MATERIAL TAUP0 1.0×10-7 s

MATERIAL NSRHP 5.0×1016 cm-3

SILVACO International 3-65

ATLAS User’s Manual

Klaassen’s Concentration Dependent Lifetime Model

The Klaassen Concentration and Temperature-dependent SRH Lifetime Model [15.] is enabled by
setting the KLASRH logical parameter in the MODELS statement. The lifetimes for electrons and holes in
this model are given by the equations:

–1 –1 300 KSRHGN
TAUN0 = ( KSRHTN + KSRHCN × )  ---------- 3-247
 

–1 –1 300 KSRHGP
TAUP0 = ( KSRHTP + KSRHCP × N )  ---------- 3-248
T

Here, is the local (total) impurity concentration. You can define the KSRHTN, KSRHTP, KSRHCN,
KSRHCP, KSRHGN, and KSRHGP parameters in the MATERIAL statement. Their default values are given
in Table 3-42.

Table 3-42. User-Specifiable Parameters for Equations 3-247 to 3-248

Statement Parameter Default Units

MATERIAL KSRHTN 2.5×10-3 s

MATERIAL KSRHTP 2.5×10-3 s

MATERIAL KSRHCN 3.0×10-13 cm3/s

MATERIAL KSRHCP 11.76×10-13 cm3/s

MATERIAL KSRHGN 1.77

MATERIAL KSRHGP 0.57

Trap-Assisted Tunneling
In a strong electric field, electrons can tunnel through the bandgap via trap states. This trap-assisted
tunneling mechanism is enabled by specifying [Link] on the MODELS statement and is
accounted for by modifying the Schockley-Read-Hall recombination model.

–
3-249
TAUP0  ETRAP TAUN0  – E TRAP
   
Γ Γ

Here, Γ is the electron field-effect enhancement term for Dirac wells, and Γ is the hole
field-effect enhancement term for Dirac wells. Γ and Γ are defined in Equations 3-72 and
3-73.

3-66 SILVACO International

 Physics

Poole-Frenkel Barrier Lowering for Coulombic Wells

The Poole-Frenkel effect can enhance the emission rate for Coulombic wells. the [Link]
parameter is specified in the MODELS statement, the Schockley-Read-Hall electron and hole
recombination model becomes:

–
. 3-250
TAUP0  ETRAP TAUN0
   – ETRAP
Γ χ Γ  

–
3-251
TAUP0  ETRAP TAUN0
   – ETRAP
χ Γ Γ  

The Poole-Frenkel thermal emission factor, χF, is defined in Equation 3-78. The Coulombic field-
enhancement terms, Γ and Γ are defined in Equations 3-80 and 3-81.

Optical Generation/Radiative Recombination

The next physical mechanisms we have to consider for generation/recombination are photon
transition. This mechanism occurs primarily in one step and is therefore a direct generation/
recombination mechanism. There are two partial processes involved. For radiative recombination, an
electron loses energy on the order of the band gap and moves from the conduction band to the valence
band. For optical generation, an electron moves from the valence band to the conduction. In silicon,
band to band generation/recombination is insignificant. This effect, however, is important for narrow
gap semiconductors and semiconductors whose specific band structure allows direct transitions. By
assuming a capture rate CcOPT and an emission rate CeOTP, the involved partial processes can be
written as

3-252

for recombination and

3-253

for generation.
These rates must be equal in thermal equilibrium so that

2
. 3-254

The total band to band generation/recombination is the difference of the partial rates, which equates to

( – ). 3-255

SILVACO International 3-67

ATLAS User’s Manual

In ATLAS, COPT is used express CcOPT and can be defined away from default values on the materials
statement or implemented using a C-Interpreter routine. To turn on the optical recombination/
generation model, define the OPTR keyword on the MODELS statement.
Auger Recombination
Auger recombination occurs through a three particle transition whereby a mobile carrier is either
captured or emitted. The underlying physics for such processes is unclear and normally a more
qualitative understanding is sufficient [2].

Standard Auger Model

Auger Recombination is commonly modeled using the expression [111]:

2 2 2 2
= AUGN  – + AUGP  – 3-256
   

where the model parameters AUGN and AUGP are user-definable in the MATERIAL statement (see Table
3-43 for its default value). You can activate this model with the AUGER parameter from the MODELS
statement.

Table 3-43. User-Specifiable Parameters for Equation 3-256

Statement Parameter Default Units

MATERIAL AUGN 8.3×10-32 cm6/s

MATERIAL AUGP 1.8×10-31 cm6/s

Klaassen’s Temperature-Dependent Auger Model

The Klaassen Auger Recombination Model [24] is activated by specifying the KLAAUG parameter of the
MODELS statement. The form of this model is:

 2 2  2 2
= – + – 3-257
   

where the Auger coefficients are temperature dependent according to:

KAUGDN
KAUGCN   3-258
 

KAUGDP
KAUGCP   3-259
 

Here, the KAUGCN, KAUGCP, KAUGDN, and KAUGDP parameters are user-definable in the MATERIAL
statement and have the defaults shown in Table 3-44.

3-68 SILVACO International

 Physics

Table 3-44. User-Specifiable Parameters for Equation 3-258 and 3-259

Statement Parameter Default Units

MATERIAL KAUGCN 1.83×10-31 cm6/s

MATERIAL KAUGCP 2.78×10-31 cm6/s

MATERIAL KAUGDN 1.18

MATERIAL KAUGDP 0.72

Narrow Bandgap Auger Model

An alternative model for the Auger recombination coefficients that is more suitable for modelling
Auger processes in narrow bandgap semiconductors can be enabled by setting the parameters AUGKN
and AUGKP on the MODELS statement. The model in ATLAS is a simplification of that by Beattie [125]
and takes the form:

 2 2  2 2
= – + – 3-260
   

where the Auger coefficients are concentration dependent according to:

AUGN
= ----------------------------------------- 3-261
+ AUGKN

AUGP
3-262
AUGKP

Here, n and p are the electron and hole carrier concentrations and the new parameters, AUGKN and
AUGKP, are user-definable on the MODELS statement.
Surface Recombination
In addition to generation-recombination within the bulk of the semiconductor, electrons or holes may
recombine or be generated at interfaces. The rate of surface recombination may be even greater than
within the bulk. The standard method is to model interface recombination in a similar manner as the
bulk generation-recombination rate [48] where:
2
–
= -------------------------------------------------------------------------------------------------------------------------------------------------------------- 3-263
τ +  ETRAP
------------------- + τ + – E TRAP
---------------------------
   

Here:

1 1
---------- = ------ + ------ S.N 3-264
τ τ

and

SILVACO International 3-69

ATLAS User’s Manual

1 1
---------- = ------ + ------ S.P 3-265
τ τ

τin is the bulk lifetime calculated at node i along the interface and which may be a function of the
impurity concentration as well. The di and Ai parameters are the length and area of the interface for
node i. The S.N and S.P parameters are the recombination velocities for electrons and holes
respectively, which are user-definable in the INTERFACE statement. The [Link], [Link], [Link], and
[Link] parameters can also be set in the INTERFACE statement to define the region, where the specified
values of the surface recombination velocities apply. This model is activated by the presence of the
recombination velocities in the INTERFACE statement.

Table 3-45. User-Specifiable Parameters for Equations 3-264 to 3-265

Statement Parameter Default Units

INTERFACE S.N 0 cm/s

INTERFACE S.P 0 cm/s

3.6.4: Impact Ionization Models

In any space charge region with a sufficiently high reverse bias, the electric field will be high enough to
accelerate free carriers up to a point where they will have acquired sufficient energy to generate more
free carriers when in collision with the atoms of the crystal. In order to acquire sufficient energy, two
principle conditions must be met.
First, the electric field must be sufficiently high. Then, the distance between the collisions of the free
carrier must be enough to allow acceleration to a sufficiently high velocity
In other words, the carrier must gain the ionization energy Ei between collisions. If the generation rate
of these free carriers is sufficiently high this process will eventually lead to avalanche breakdown.
The general impact ionization process is described by the Equation 3-266.

G = α +α 3-266

Here, G is the local generation rate of electron-hole pairs, αn,p are the ionization coefficient for
electrons and holes and Jn,p are their current densities. The ionization coefficient represents the
number of electron-hole pairs generated by a carrier per unit distance travelled. The accurate
calculation of this parameter has been researched because it is vital if the effects related to impact
ionization, such as substrate current and device breakdown, are to be simulated. These models can be
classified into two main types: local and non-local models.
The former assume that ionization at any particular point within the device is a function only of the
electric field at that position. Non-local models, however, perform a more rigorous approach by taking
into account the energy that the carrier gains.

3-70 SILVACO International

 Physics

G = α n ( E tri ) J ntri + α p ( Etri ) J ptri 3-267

Etri Jntri Jptri

[Link] IMPACT
[Link] IMPACT
G = α n ( E side ) J nside + α p ( E side ) J pside 3-268

Eside Jnside
Jpside

[Link] IMPACT

 E tri ⋅ J ntri   Etri ⋅ J ntri 

G = α n  --------------------------
-  J ntri + α p  --------------------------
-  J ntri 3-269
 J ntri   J ntri 

Local Electric Field Models for Impact Ionization

Selberherr’s Impact Ionization Model

SELB IMPACT

BN  BETAN
αn AN exp –  3-270
 E 

BP  BETAP
αp AP exp –  3-271
 E 

AN AP BN BP BETAN BETAP IMPACT

AN AP BN BP
EGRAN EGRAN AN1 AP1, BN1 BP1
EGRAN AN2 AP2 BN2 BP2

SILVACO International 3-71

ATLAS User’s Manual

Here, E is the electric field in the direction of current flow at a particular position in the structure and
the parameters AN, AP, BN, BP, BETAN, and BETAP are defined on the IMPACT statement and have the
default values shown in Table 3-46. In the case of AN, AP, BN, and BP you can define a value of electric
field, EGRAN V/cm, where for electric fields, >EGRAN V/cm, the parameters are: AN1, AP1, BN1, BP1,
while for electric fields, <EGRAN V/cm, the parameters become AN2, AP2, BN2, and BP2.
The AN and BN parameters are also a function of the lattice temperature in this model [87]. The
temperature dependence of these coefficients is defined as follows:
[Link]
AN = AN 1, 2  + [Link]   –  3-272
   

[Link]
AP = AP 1, 2  + [Link]  
 – 
 3-273

[Link]
BN = BN  + [Link]   –  3-274
1, 2    

[Link]
BP = BP1, 2  + [Link]   –  3-275
   

The parameters associated with these equations are shown in Table 3-46.

Table 3-46. User-Definable Parameters in the Selberherr Impact Ionization Model

Statement Parameter Default

IMPACT AN1 7.03×105 cm-1

IMPACT AN2 7.03×105 cm-1
IMPACT AP1 6.71×105 cm-1

IMPACT AP2 1.58×106 cm-1

IMPACT BN1 1.231×106V/cm

IMPACT BN2 1.231×106 V/cm
IMPACT BP1 1.693×106 V/cm
IMPACT BP2 2.036×106 V/cm
IMPACT BETAN 1.0

IMPACT BETAP 1.0

IMPACT EGRAN 4×105 V/cm

3-72 SILVACO International

 Physics

Table 3-47. Temperature Coefficient Parameters of the Selberherr Impact Ionization Model for
Silicon in Equations 3-272 to 3-275

Statement Parameter Default

IMPACT [Link] 0.588

IMPACT [Link] 0.248

IMPACT [Link] 0.588

IMPACT [Link] 0.248

IMPACT [Link] 1.0

IMPACT [Link] 1.0

IMPACT [Link] 1.0

IMPACT [Link] 1.0

Valdinoci Impact Ionization Model

Valdinoci et al. [27] reported on a measurement technique to calibrate the temperature dependence of
impact ionization models and proposed a new model based on their study. Their model of both electron
and hole ionization coefficients was calibrated for silicon for temperature ranging from 25 to 400°C.
This model is based on the following:

α 3-276
( ) ( ) [ ( )⁄( ( ))]

where is the electric field along the current flow lines. The parameters in Equation 3-276 depend on
temperature as follows:
VAL · AN2
( ) VAL · AN0  VAL AN1 ⋅ 
 · 
( ) VAL · BN0  VAL BN1 ⋅ 
 · 
3-277
VAL · CN2
( ) VAL · CN0  VAL CN1 ⋅  VAL · CN3 ⋅
 · 
( ) VAL · DN0 VAL · DN1 ⋅ VAL · DN2 ⋅

SILVACO International 3-73

ATLAS User’s Manual

VAL · AP2
( ) VAL · AP0  VAL · AP1 ⋅ 
 
( ) VAL · BP0  VAL BP1 ⋅ 
 · 
3-278
VAL · CP2
( ) VAL · CP0  VAL CP1 ⋅  VAL · CP3 ⋅
 · 
( ) VAL · DP0 VAL · DP1 ⋅ VAL · DP2 ⋅

Our default parameters for silicon are shown in Table 3-48. These parameters are also taken from
Valdinoci et al. [27]. To enable this model, specify the logical parameter, VALDINOCI, in the IMPACT
statement.

Table 3-48. Default Parameters for Valdinoci Impact Ionization Model

Statement Parameter Type Default Units

IMPACT VAL.AN0 Real 4.3383

IMPACT VAL.AN1 Real -2.42e-12

IMPACT VAL.AN2 Real 4.1233

IMPACT VAL.BN0 Real 0.235

IMPACT VAL.BN1 Real 0.0

IMPACT VAL.CN0 Real 1.6831e4

IMPACT VAL.CN1 Real 4.3796

IMPACT VAL.CN2 Real 1.0

IMPACT VAL.CN3 Real 0.13005

IMPACT VAL.DN0 Real 1.233735e6

IMPACT VAL.DN1 Real 1.2039e3

IMPACT VAL.DN2 Real 0.56703

IMPACT VAL.AP0 Real 2.376

IMPACT VAL.AP1 Real 0.01033

IMPACT VAL.AP2 Real 1.0

IMPACT VAL.BP0 Real 0.17714

IMPACT VAL.BP1 Real -0.002178

IMPACT VAL.CP0 Real 0.0

3-74 SILVACO International

 Physics

Table 3-48. Default Parameters for Valdinoci Impact Ionization Model

Statement Parameter Type Default Units

IMPACT VAL.CP1 Real 0.00947

IMPACT VAL.CP2 Real 2.4924

IMPACT VAL.CP3 Real 0.0

IMPACT VAL.DP0 Real 1.4043e6

IMPACT VAL.DP1 Real 2.9744e3

IMPACT VAL.DP2 Real 1.4829

Grant’s Impact Ionization Model

The second ionization model has the same form as the Selberherr model but a simpler implementation
where:

BN
α = AN –  ----------  3-279
 

BP
α = AP –  ---------  3-280
 

This implementation has three key differences:

• The model has a low field, an intermediate field and a high field region.
• The coefficients for silicon are different.
• There is no temperature dependence.
This model was developed after investigations by Baraff [28] suggested the existence of a low,
intermediate and high field response region for electron and hole ionization rates. The coefficients
implemented into this model match the experimental data of Grant [29], which suggested that the
three different regions existed.
This model is activated with the GRANT parameter of the IMPACT statement. The model parameters:
AN, AP, BN, and BP aren’t user-definable. Instead, the three electric field regions have in-built values
as follows:

1) Low Electric Field × 3-281

× ×

× ×

2) Intermediate Electric Field × × 3-282

× ×

× ×

3) High Electric Field × 3-283

× ×

× ×

SILVACO International 3-75

ATLAS User’s Manual

Crowell-Sze Impact Ionization Model

Crowell and Sze [30] have proposed a more physical relationship between the electric field and the
ionization rates. This model represents ionization coefficients as follows:

α ( ) ( ) ( ) 3-284
λ

– – 3-285

–
( ) – ×10 – 3-286

–
( ) ×10 – 3-287

where:

3-288
λ

3-289


 3-290


[ ⁄ ]
λ LAMDAE 3-291
[ ⁄ ]

[ ⁄ ]
λ LAMDAH 3-292
[ ⁄ ]

The Crowell-Sze Model for impact ionization is selected by setting the CROWELL parameter of the
IMPACT statement.

Table 3-49. Crowell-Sze Impact Ionization Model Parameters.

Statement Parameter Default

IMPACT LAMDAE 6.2×10-7cm

IMPACT LAMDAH 3.8×10-7 cm

3-76 SILVACO International

 Physics

Non-Local Carrier Energy Models for Impact Ionization

All local electric field based models will normally overestimate the rate of impact ionization. This
occurs because lucky electron theory inherently assumes that a carrier is traveling through a constant
electric field E. As a result it will predict a distance ∆x=Ei/qE over which the carrier will gain the
ionization energy Ei. In real devices, however, the electric field is never constant but is normally
sharply peaked at metallurgical junctions. Therefore, as a carrier passes through the peaked electric
field the lucky electron model will predict the ionization distance ∆x to be too small. As a result the
ionization rate is overestimated. The effect of this is that all the simulated breakdown voltages will be
underestimated and substrate currents overestimated.
The Energy Balance Transport Model can be used to improve the simulation of impact ionization by
implementing ionization models based upon the carrier temperature instead of the electric field. The
carrier temperature is a more meaningful basis as the velocity-field relationship is more closely
modeled. This allows a non-local dependence on the electric field within the impact ionization model.
Energy Balance Transport Models will therefore result in more accurate simulations of breakdown
voltage and substrate current. Two different impact ionization models have been implemented into
ATLAS, the first is based upon the classical Chynoweth relationship, modified to include carrier
temperature, but the second is a more advanced non-Maxwellian approach based upon carrier
temperatures.
When the energy balance transport model is applied, only two impact ionization models are available.
These are the Toyabe model and the Concannon model.

Toyabe Impact Ionization Model

The temperature dependent impact ionization model is founded on the Selberherr model and is similar
to that suggested by Toyabe [80]. The carrier temperature is used to calculate an effective electric field
based upon the homogeneous temperature-field relation. To maintain self-consistency within the
Energy Balance Transport Model this is the same relationship used for the effective electric field
within the carrier temperature dependent mobility. This model is the default model for impact
ionization with energy balance transport and is activated with the TOYABE or SELB parameters on the
IMPACT statement. The ionization rates now have the form:

α AN  – BN  3-293
 

α AP  – BP  3-294
 

where the model parameters: AN, AP, BN, and BP are user-definable in the IMPACT statement (see
Table 3-46 for their default values). The effective electric field is calculated according to:

3
, = --- --------------------------------- 3-295
2 [Link]

3
, = --- ----------------------------------- 3-296
2 [Link]

where the energy relaxation lengths, [Link] and [Link], can be explicitly defined in the IMPACT
statement, or can be calculated according to:

[Link] = VSATN * TAUSN 3-297

[Link] = VSATP * TAUSP 3-298

SILVACO International 3-77

ATLAS User’s Manual

VSATN and VSATP are the saturation velocities for electrons and holes, and TAUSN and TAUSP
correspond to the electron energy relaxation times ([Link] and [Link]) in Equations 3-238
and 3-239. You must set the [Link] flag to use the values of [Link] and [Link] specified in
the IMPACT statement.

Table 3-50. User-Specifiable Parameters for Equations 3-295-3-298

Statement Parameter Units

IMPACT [Link] µm

IMPACT [Link] µm
MATERIAL VSAT cm/s

MATERIAL VSATN cm/s

MATERIAL VSATP cm/s

IMPACT TAUSN s

IMPACT TAUSP s

Note: As an added level of flexibility, the relaxation times used for the energy balance equation and those used in the impact
ionization model are separated into two user-definable parameters. In contrast to [Link] and [Link], which are
used in different formulae, the TAUSN and TAUSP parameters are only applicable in the impact ionization expression in
Equations 3-294 and 3-295. By default, [Link]=TAUSN and [Link]=TAUSP.

It can also be argued that the AN, AP, BN, and BP parameters should also be a function of the carrier
temperature. But, no clear theoretical basis for this has been proposed and accepted. Instead, the C-
INTERPRETER within ATLAS has been extended to include two C-INTERPRETER functions. These
functions are specified via the [Link] and [Link] parameters of the IMPACT statement. These
parameters specify the filename of a text file containing a C-INTERPRETER function that describes the
dependence of the model parameters AN, AP, BN and BP as a function of the carrier temperatures.
These values will then be used within Toyabe’s energy dependent impact ionization model.

Concannon’s Impact Ionization Model

The previous non-local impact ionization model inherently assumes a Maxwellian shape to the
distribution of hot carriers. Recent work by Fiegna et. al. [32] using Monte Carlo simulations suggests
a non-Maxwellian high energy tail to the energy distribution function. To accurately model these
effects, a non-Maxwellian based model from Concannon [33] has been implemented. Based upon this
energy distribution the model calculates the probability of a carrier having sufficient energy to cause
impact ionization. The model results show good agreement with measured results for a 0.9µm flash
EPROM device [33].
To enable the Concannon substrate current for the electron and hole continuity equations, specify the
[Link] and [Link] parameters in the IMPACT statement.
The generation rate is a function of the carrier temperature and concentration is given by:
∞
( ) = CSUB.N × ∫ (ε ( ))) ε 3-299
ETH.N

3-78 SILVACO International

 Physics

∞
( ) = CSUB.P × ∫ (ε ( ))) ε 3-300
ETH.P

where and are the electron and hole carrier concentrations within the semiconductor, ε is
energy, and are the electron and hole carrier temperatures in the semiconductor, is
given in Equation 3-231, CSUB.N, CSUB.P, ETH.N, and ETH.P are user-specifiable parameters as
given in Table 3-51.

Table 3-51. User-Specifiable Parameters for Equations 3-299 and 3-300

Statement Parameter Default Units

IMPACT CSUB.N 2.0×1014

IMPACT CSUB.P 4.0×1014

IMPACT ETH.N 1.8 eV

IMPACT ETH.P 3.5 eV

The function, (ε, ), in Equations 3-299 and 3-300 is given by the product of the density of states
function, g(ε), and the energy distribution function (ε) as:
(ε) (ε)
= ----------------------------- 3-301
∞
∫
°
(ε) (ε)

The density of states function is given by:

1.25
(ε) = ε 3-302

The energy distribution functions for electrons and holes are:

   
(ε) = – CHIA ε
----------------------------- + C0 – CHIB ε
----------------------------- 3-303
   
   

 3
(ε) = – [Link] ε 
--------------------------------------------------
- 3-304
 1.5
 

Here, ε is energy, are the carrier temperatures, CHIA, CHIB, and C0 are user-specifiable
parameters (see Table 3-52).

SILVACO International 3-79

ATLAS User’s Manual

Table 3-52. User-Definable Parameters for the Energy Distribution Functions

Statement Parameter Default Units

IMPACT CHIA 3.0×105

IMPACT CHIB 5.0×104

IMPACT C0 2.5×10-10

IMPACT [Link] 4.6×104

Two other parameters in the IMPACT statement are user-definable, which may affect the result of the
numeric integration. The [Link] parameter specifies the energy step size in eV used during the
numeric integration. The default step size is 25 meV. The INFINITY parameter sets the upper limit of
the integration and specifies ratio of the increment added to the integral divided by the current value
of the integral. The default value of the INFINITY parameter is 0.001.

Note: To maintain self-consistent results, implement this model if the Concannon model is being used for the simulation of gate
current.

Band-to-Band Tunneling
If a sufficiently high electric field exists within a device local band bending may be sufficient to allow
electrons to tunnel, by internal field emission, from the valence band into the conduction band. An
additional electron is therefore generated in the conduction band and a hole in the valence band. This
generation mechanism is implemented into the right-hand side of the continuity equations. The
tunneling generation rate is [6,89, 144] as:

[Link]  – BB.B
= BB.A 3-305
 

where is the magnitude of an electric field and BB.A, BB.B, and [Link] are user-definable
parameters. In ATLAS there are three different sets of values that may be applied to the model
parameters.
The model parameters can be set to the standard model [35] by specifying [Link] on the MODELS
statement. The parameter defaults for the standard model are as follows:
BB.A = 9.6615e18 V/cm BB.B= 3.0e7 [Link]= 2.0
The model parameters may also be set to the Klaassen Model [6,89,144] by specifying [Link] on the
MODELS statement. The parameter defaults for the Klaassen model are as follows:
BB.A = 4.00e14 V/cm BB.B = 1.9e7 [Link]= 2.5
In application, use the standard model with direct transistors while using the “klaassen” model with
indirect transistors.
You can modify the basic band-to-band tunneling given in Equation 3-305 by the generation rate GBBT
with the D factor as suggested in [6]. The D factor is given by:
– –
( [φ – ψ ⁄ ] ) –( [φ – ψ ⁄ ] ) 3-306

3-80 SILVACO International

 Physics

where φ and φ are the electron and hole quasi-Fermi levels as given in Equations 3-9 and 3-10 and ψ
is the electrostatic potential. To enable this modification, specify [Link] in the MODELS statement.
Another modification allows these model parameters to be calculated from the first principles by
specifying the AUTOBBT parameter in the MODELS statement. In this case, the parameters are
calculated according to:

( × [Link] 0 )
BB.A = ------------------------------------------------------------------------------------------- 3-307
EG300

[Link] 0
π EG300 --------------------------------------------------------------
BB.B = ---------------------------------------------------------------------------------------------------- 3-308

Here:
[Link] = 2 3-309

where is the electronic charge, h is Planck’s constant, is the energy bandgap, is the rest mass
of an electron and [Link] is the effective mass. The parameter [Link] may be set on the
MODELS statement and the bandgap at 300K, EG300, is defined on the MATERIAL statement.

Table 3-53. User-Definable Parameters in the Band-to-Band Tunneling Model

Statement Parameter Default Units

MODEL BB.A 4.0×1014 V-2s-1cm-1

MODEL BB.B 1.9×107 V/cm

MODEL [Link] 2.5

3.6.5: Gate Current Models

In devices that have a metal-insulator-semiconductor (MIS) formation, the conductance of the
insulating film would ideally be considered as zero. But, for the sub 0.5um generation of MOS devices
there is now considerable conductance being measured on the gate contacts. This gate current has
resulted in two major consequences, one negative and one positive.
On the negative side, the gate current is responsible for the degradation in device operating
characteristics with time. This reliability issue is important because the lifetime of electronic parts has
to be guaranteed. You can simulate reliability within the Silvaco suite of tools for device level
reliability which are described in a later chapter.
On the positive side, the existence of this gate current has caused the proliferation of the non-volatile
memory market. These devices use the existence of gate current to program and erase the charge on a
“floating” contact. This concept has resulted in a variety of different devices such as FLASH, FLOTOX,
and EEPROM. All such devices rely on the physics of the gate current process for their existence.

There are a variety of different conduction mechanisms within an insulating layer [9], but in the case
of nonvolatile memory, only two mechanism are relevant: Fowler-Nordheim tunneling and hot carrier
injection. Models for these two injection processes are described in the following sections. In the case of

SILVACO International 3-81

ATLAS User’s Manual

hot electron injection, two models are available: the lucky electron model and the Concannon gate
current model.
Fowler-Nordheim Tunneling
If the electric field across an insulator is sufficiently high, then it may cause tunneling of electrons
from the semiconductor (or metal) Fermi level into the insulator conduction band. This process is
strongly dependent on the applied electric field but is independent of the ambient temperature.
The Fowler-Nordheim Equation [37] expresses tunnel current density through the oxide as:

[Link]  – [Link] 3-310

 

2  – [Link]
= [Link] 3-311
 

where specifies the magnitude of the electric field in the oxide. The model parameters: [Link], [Link],
[Link], and [Link] can be defined on the MODELS statement. The default values for these parameters,
obtained from Keeney, Piccini, and Morelli [35], are shown in Table 3-54.

Table 3-54. User-Specifiable Parameters for Equations 3-310 and 3-311

Symbol Statement Parameter Default Values

AFN MODELS [Link] 1.82×10-7

BFN MODELS [Link] 1.90×108

AFH MODELS [Link] 1.82×10-7

BFH MODELS [Link] 1.90×108

The Fowler-Nordheim model in ATLAS has been implemented as either a post processing function or
as a self-consistent scheme with the other equations. The post processing option may be chosen by
specifying the parameter, FNPP, on the MODEL statement. The self-consistent scheme is activated by
specifying the parameter, FNORD, on the MODEL statement.
For either model, the implementation scheme is the same. Each electrode-insulator and insulator-
semiconductor interface is divided into discrete segments which are based upon the mesh. For each
insulator-semiconductor segment, the Fowler-Nordheim current is calculated as described above. This
current is then added to a segment on the electrode-insulator boundary. Two schemes have been
implemented to find out to which segment this current should be added.
The default model that calculates which electrode segment receives the Fowler-Nordheim current
follows the path of the electric field vector at the semiconductor-insulator interface. The first electrode-
insulator segment that is found along this trajectory, provided no other semiconductors or metals are
found along the trajectory, receives the Fowler-Nordheim current.
A second model may be chosen using the NEARFLG parameter of the MODEL statement. In this case, the
electrode-insulator segment found closest to the semiconductor-insulator segment receives the Fowler-
Nordheim current.
The total current on the gate electrode is then the sum of the currents from all the individual segments
around the electrode boundary.

3-82 SILVACO International

 Physics

Note: Since Fowler-Nordheim tunneling current is responsible for EPROM and EEPROM cell erasure, this model should
always be specified when performing erasure simulation. We also recommend that you model the band-to-band tunneling
model if you model Fowler-Norheim tunneling.

Note: When simulating EPROM erasure in a transient analysis with this model, the floating contact charge becomes a function
of the gate current. In this case, the total current flowing into the floating electrode is multiplied by the time step to calculate the
charge added to the electrode during that time step. The new value of the charge is then used as the boundary condition for the
next time step.

Lucky Electron Hot Carrier Injection Model

In the Lucky-Electron Hot Carrier Injection Model it is proposed that an electron is emitted into the
oxide by first gaining enough energy from the electric field in the channel to surmount the insulator/
semiconductor barrier. Once the required energy to surmount the barrier has been obtained the
electrons are redirected towards the insulator/semiconductor interface by some form of phonon
scattering. When these conditions are met, the carrier travelling towards the interface will then have
an additional probability that it will not suffer any additional collision through which energy could be
lost.
The model implemented into ATLAS is a modified version of the model proposed by Tam [38] and is
activated by the parameters HEI and HHI, for electron and hole injection respectively, on the MODELS
statement. The gate electrode-insulator interface is subdivided into a number of discrete segments
which are defined by the mesh. For each segment the lucky electron model is used to calculate the
injected current into that segment. The total gate current is then the sum of all of the discrete values.
If we consider a discrete point on the gate electrode-insulator boundary we can write a mathematical
formula for the current injected from the semiconductor. The formula calculates the injected gate
current contribution from every node point within the semiconductor according to:

∫∫ ( ) ( ) ∫∫ ( ) ( ) 3-312

where Jn,p (x,y) are the electron and hole current densities at a point (x,y) within the semiconductor,
and Pn,p (x,y) are the probabilities that a fraction of this current reaches the gate oxide and is injected
across into the gate electrode. The total probability Pn,p (x,y) is defined by:

( )
φ ⁄ [Link] 3-313

( ) φ ⁄ [Link] 3-314

where is the electric field parallel to the current flow, [Link] and [Link] are the electron and
hole mean free path lengths between redirecting collisions. The three probability factors will now be
described.
The probability φ is the probability of a carrier gaining the energy φ by moving in, and parallel to,
an electric field , without suffering energy loss by optical phonon scattering and is given by:

[Link] φ
 –  3-315
φ  φ   [Link]

SILVACO International 3-83

ATLAS User’s Manual

φ
[Link] ,
φ , = 0.25  --------------------------------------- exp  – --------------------------------------- 3-316
 φ ,   [Link]

where [Link] and [Link] are the mean free path lengths of electrons and holes for scattering
by optical phonons. The barrier heights φBn,p are defined according to:

⁄
φ IG.EB0 – [Link] ⊥ – [Link] ⊥ – ∆ψ ( ) 3-317

⁄
φ IG.HB0 – [Link] ⊥ – [Link]⊥ – ∆ψ ( ) 3-318

where Ε⊥ is the electric field perpendicular to the semiconductor-insulator interface. The traditional
barrier heights, IG.EB0 and IG.HB0, are reduced to take account of three effects. The first effect is
due to Schottky barrier lowering which depends on the perpendicular electric field at the
semiconductor-insulator interface. The second effect takes account of tunneling through the gate oxide
by reducing the barrier height. The third effect takes into account that a potential difference exists
between the semiconductor-insulator interface and the starting position of the hot carrier. By default,
this last effect is disabled. But you can enable it by specifying the [Link] and [Link]
parameters for electrons and holes respectively.
The second probability P1 is the probability that no energy is lost by optical phonon scattering as the
hot carrier travels towards the semiconductor-insulator interface after being redirected, and is given
by:

–  3-319
 [Link]

–  3-320
 [Link]

where r is the distance from point of redirection to the semiconductor-insulator interface.

The final probability P2 accounts for the probability of scattering in the image force potential well in
the gate oxide and is given by:

 ------------------------------
 16πε 
=  – ---------------------------------- for θ > THETA.N 3-321
2,  PATH.N 
 
 

= 0 for θ < THETA.N 3-322

2,

 ------------------------------
 16πε 
=  – ---------------------------------- for θ > THETA.P 3-323
2,  PATH.P 
 
 

2, = 0 for θ < THETA.P 3-324

3-84 SILVACO International

 Physics

Here, PATH.N and PATH.P are the electron and hole mean free path lengths within the oxide, εox is the
oxide permittivity and is the electric field in the oxide. The angle θ introduces an angle dependence
which is based upon the work of Wada [39]. His experiments indicate a critical rejection angle,
THETA.N and THETA.P, between the angle θ formed between the semiconductor-insulator interface
and the electric field in the oxide. If the angle θ is less than the rejection angle then the electrons are
repelled back to the substrate.
Table 3-55 lists the user-definable model parameters that can be set in the MODELS statement, their
default values, and their units.

Table 3-55. User-Definable Parameters in Concannon’s Gate Current Model

Statement Parameter Default Units

MODELS IG.EB0 3.2 eV

MODELS [Link] 6.16×10-6 cm

MODELS [Link] 6.16×10-6 cm

MODELS [Link] 9.2×10-7 cm

MODELS IG.HB0 4.0 eV

MODELS [Link] 9.2×10-7 cm

MODELS [Link] 2.59×10-4 (Vcm)1/2

MODELS [Link] 2.59×10-4 (Vcm)1/2

MODELS [Link] 2.0×10-5 V1/3cm2/3

MODELS [Link] 2.0×10-5 V1/3cm2/3

MODELS [Link] 3.0×10-6 [Q=1] cm

MODELS [Link] 2.0×10-6 [Q=1] cm

MODELS PATH.N 3.4×10-7 cm

MODELS PATH.P 2.38×10-7 cm

MODELS THETA.N 60 degrees

MODELS THETA.P 60 degrees

The implementation of this model is similar to that for Fowler-Nordheim tunneling. Each electrode-
insulator and insulator-semiconductor interface is divided into discrete segments, which are based
upon the mesh. For each insulator-semiconductor segment the Fowler-Nordheim current is calculated
as described above. This current will then be added to a segment on the electrode-insulator boundary.
Two schemes have been implemented to find out to which segment this current should be added.
The default model that calculates which electrode segment receives the hot carrier injected current
follows the path of the electric field vector at the semiconductor-insulator interface. The first electrode-
insulator segment that is found along this trajectory, provided no other semiconductors or metals are
found along the trajectory, will receive the current.

SILVACO International 3-85

ATLAS User’s Manual

A second model may be chosen using the NEARFLG parameter of the MODELS statement. In this case,
the electrode-insulator segment found closest to the semiconductor-insulator segment will receive the
hot carrier injected current.
The total current on the gate electrode is then the sum of the currents from all the individual segments
around the electrode boundary.
The lucky electron hot carrier injection model can be used to include the electron or hole carrier
temperature in the solution because the carrier temperature does not directly enter the equations.

The one exception is in ATLAS2D where the electric field parallel to the current flow is calculated as
E = 1.5 KB Tn / [Link]
for electrons if [Link] is specified and
E = 1.5 KB Tp / [Link]
for holes if [Link] is specified. K_{B} is Boltzmanns constant in units of eV/Kelvin. If [Link] is
set to zero then E will be calculated the same way as if [Link] is unspecified. The same is true for
[Link] and [Link]. In ATLAS3D, this model is unavailable and E is calculated the same way
regardless if HCTE is specified.

Note: When simulating EPROM programming with this model, the floating contact charging is simulated in the transient mode.
In this case, the total current flowing into the floating electrode is multiplied by the time step to calculate the charge added to the
electrode during that time step. The new value of charge is then used as the boundary condition for the next time step.

Concannon’s Injection Model

The implicit assumption in the lucky electron approach is a Maxwellian shape for the energy
distribution of the hot carriers. Recent work by Fiegna [32] using Monte Carlo simulations suggests a
non-Maxwellian high energy tail to the distribution function. To accurately model these effects, a non-
Maxwellian based model from Concannon [33] has been implemented. This model requires the
solution to the energy balance equation for the carrier temperatures but has been implemented in a
similar manner to the lucky electron model. The Concannon gate injection model may be specified with
the parameters [Link] and [Link] on the MODELS statement. This choice of parameters
automatically activates the Energy Balance Transport Model.
The Concannon injection model has a similar form to the lucky electron model. The injected current is
calculated according to:

∫∫ ( ) ( ) ∫∫ ( ) ( ) 3-325

where and are the carrier concentrations within the semiconductor. The probability
functions and are now defined by:

( , ) = – CGATE.N φ , 1, 2, 3-326

( ) CGATE.P φ 3-327

where is the electronic charge and the parameters CGATE.N and CGATE.P are user-definable on the
MODEL statement. The three probability functions in Equations 3-325 and 3-326 shall now be
described.

3-86 SILVACO International

 Physics

The probability that a carrier has sufficient energy to surmount the insulator-semiconductor barrier of
height φB is now defined as a function of energy. The probability now has the form:

∞
φ , = ∫φ ,
⊥ ( ε ) ( ε, ( , )) ε 3-328

∞
φ , = ∫φ ,
⊥ ( ε ) ( ε, ( , )) ε 3-329

Here, v⊥(ε) is the perpendicular velocity of a hot carrier and defines the probability of a hot carrier
with an energy ε travelling in the direction of the insulator-semiconductor. The barrier heights φBn,p
are defined according to:
2⁄3
φ , = PEFF.N – [Link] ⊥ – [Link] ⊥ – ∆ψ ( , ) 3-330

2⁄3
φ , = PEFF.P – [Link] ⊥ – [Link] ⊥ – ∆ψ ( , ) 3-331

where E⊥ is the electric field perpendicular to the semiconductor-insulator interface. The traditional
barrier heights, IG.EB0 and IG.HB0, are reduced to take account of three effects. The first effect is
due to Schottky barrier lowering which depends on the perpendicular electric field at the
semiconductor-insulator interface. The second effect takes account of tunneling through the gate oxide
by reducing the barrier height. The third effect takes into account that a potential difference exists
between the semiconductor-insulator interface and the starting position of the hot carrier. By default,
this last effect is disabled. But you can enable it by specifying the [Link] and [Link]
parameters for electrons and holes respectively.
The carrier velocity model follows the approach of Fiegna et. al. [32.] where velocity is proportional to
energy according to:
0.25
⊥∼ ε 3-332

The function, F(ε, Tn,p(x,y)), is determined by the density of states and the energy distribution function
according to:
( ε ) ( ε)
( ε, ( , ) ) ∼ ------------------------------------ 3-333
∞
° ∫ (ε) (ε) ε

The density of states g(ε) follows the analysis of Cassi [40] where:

(ε) ∼ ε 3-334

Finally the energy distribution functions for electrons and holes are defined by:

   
(ε) ∼ – ETH.N ε 
--------------------------------- – ETH.P ε 
 + C0 --------------------------------
- 3-335
  
   

SILVACO International 3-87

ATLAS User’s Manual

 
(ε) ∼ – CHI.A ε
------------------------------- 3-336
 
 
where ETH.N, ETH.P, CHI.A, and C0 are user-definable constants found from fitting to measured
data. The terms: and are the mean carrier temperatures for electrons and holes, which are
calculated from the Energy Balance Transport Model.
Normalization in all of the above equations is accounted for in the combined constants of
proportionality, CGATE.N and CGATE.P.
The second probability is the probability that no energy is lost by optical phonon scattering as the
hot carrier travels towards the semiconductor-insulator interface after being redirected and is given
by:

∼  – -------------------------------- 3-337
1,  [Link]

–  3-338
 [Link]

where is the distance from point of redirection to the semiconductor-insulator interface.

The final probability accounts for the probability of scattering in the image force potential well in
the gate oxide and is given by:

 ------------------------------
 16πε 
=  – ---------------------------------- for θ > THETA.N 3-339
2,  PATH.N 
 
 

2, = 0 for θ > THETA.N 3-340

 ------------------------------
 16πε 
=  – ---------------------------------- for θ > THETA.P 3-341
2,  PATH.P 
 
 

= 0 for θ < THETA.P 3-342

2,

Here, PATH.N and PATH.P are the electron and hole mean free path lengths within the oxide, εox is the
oxide permittivity and is the electric field in the oxide. The angle θ introduces an angle dependence
which is based upon the work of Wada [39]. His experiments indicate a critical rejection angle,
THETA.N and THETA.P between the angle θ formed between the semiconductor-insulator interface and
the electric field in the oxide. If the angle θ is less than the rejection angle, then the electrons are
repelled back to the substrate.

3-88 SILVACO International

 Physics

Note: The current implementation of the Concannon model for hot carrier injection is that only carriers along the
semiconductor-insulator interface are significant and as a result the probability P1 is assumed unity. This also means that the
integration is only applied to those node points along the semiconductor-insulator interface.

Two other parameters of the MODELS statement that may affect the result of the numeric integration
are user-definable. The [Link] parameter specifies the energy step size in eV used during the
numeric integration. The default step size is 25 meV. The INFINITY parameter sets the upper limit of
the integration and specifies ratio of the increment added to the integral divided by the current value
of the integral. The default value of the INFINITY parameter is 0.001.
The implementation of this model is similar to that for Fowler-Nordheim tunneling. Each electrode-
insulator and insulator-semiconductor interface is divided into discrete segments which are based
upon the mesh. For each insulator-semiconductor segment the Fowler-Nordheim current is calculated
as described above. This current will then be added to a segment on the electrode-insulator boundary.
Two schemes have been implemented to find out to which segment this current should be added.
The default model that calculates which electrode segment receives the hot carrier injected current
follows the path of the electric field vector at the semiconductor-insulator interface. The first electrode-
insulator segment that is found along this trajectory, provided no other semiconductors or metals are
found along the trajectory, will receive the current.
A second model may be chosen using the NEARFLG parameter of the MODELS statement. In this case
the electrode-insulator segment found closest to the semiconductor-insulator segment will receive the
hot carrier injected current.
The total current on the gate electrode is then the sum of the currents from all the individual segments
around the electrode boundary.

Note: To maintain self-consistent results, it’s important that this model is implemented if the Concannon model is being used
for the simulation of substrate current.

Direct Quantum Tunneling Model

For MOS devices with very thin oxide layers, the Fowler-Nordheim formula (Equations 3-310 and 3-
311) for calculating the tunnelling current gives incorrect results. A more accurate approximation in
this case is to treat the oxide as a potential barrier in the envelope approximation and solve for the
quantum tunnelling probability from an occupied state α on one side of the barrier to an unoccupied
state α on the other side of the barrier using Fermi's Golden Rule.

π
α →α = ----- 〈 α α 〉 3-343
h

SILVACO International 3-89

ATLAS User’s Manual

Figure 3-3: Typical conduction and valence band profiles of a MOS capacitor. The right region
represents the MOS bulk, the left region represents the gate.
The tunnelling current ( → ) is then calculated by integrating the tunnelling probability over all
occupied states on side of the barrier and all empty states on side of the barrier. Similarly, the
tunnelling current ( → ) is calculated by integrating the tunnelling probability over all occupied states
on side of the barrier and all empty states on side of the barrier.
The difference between the ( → ) contribution and the ( → ) contribution is the net tunnelling current
through the barrier.
If the carrier densities and potential have been calculated using a semi-classical model, then the
formula for current for a given band minimum is

 [( – ) ⁄ ]

π h
∫ ( ) 
 [( – )⁄

]
3-344

where and are the effective masses at the band minimum in a direction perpendicular to the
tunnelling direction. For electrons in silicon, with 6 band minima, we obtain

 [( – ) ⁄ ]

π h
( + ) ∫ ( ) 
 [( – )⁄

]
3-345

and for holes we obtain

 [( – ) ⁄ ]

π h
( ) ∫ ( ) 
 [( – ) ⁄

]
3-346

where is the tunneling probability for incident energy and and are the quasi-Fermi levels
on either side of the barrier. If then there is no tunnelling current flowing through the barrier.

3-90 SILVACO International

 Physics

If we have strong quantum confinement on one side of the barrier and the carrier densities and
potential have been calculated using the self-consistent Schrodinger-Poisson model, then the tunneling
current is obtained from

 [( – ) ⁄ ]

πh
2 ∑ ( ) ( ) 
 [( – )⁄

]
3-347

where the sum is over all bound state energies, , for a particular band minimum.
For electrons in silicon, with 6 conduction band minima, we obtain

 [( – )⁄ ]
πh
2 ∑ ( ) 
 [( – )⁄ ] 
3-348

 [( – )⁄ ]
+
πh
2 ∑ ( ) 
 [( – )⁄ ] 
where the sums over longitudinal bound eigenstates and transverse bound eigenstates are done
separately. For holes, the current is calculated using the equation

 [( – )⁄ ]

πh
2 ∑ ( ) 
 [( – )⁄ ] 
3-349

 [( – )⁄ ]
+
πh
2 ∑ ( ) 
 [( – )⁄ ] 
where the sums over light hole and heavy hole bound eigenstates are done separately.
The expression υ is called the attempt frequency and is given by

h Ψ 2
Ψ( ) + ( ) 3-350

It is evaluated at the semiconductor-oxide interface on the side of the barrier where the carrier
confinement occurs.
The tunnelling probability is calculated by using a transfer matrix method to solve the
Schrodinger equation through the barrier. The method approximates the potential as piecewise
constant. Therefore, the best results will be obtained if the oxide layer contains at least several mesh
points in the tunnelling direction.
To enable the quantum tunnelling model for electrons, specify [Link] in the MODELS statement. To
enable the quantum tunnelling model for holes, specify [Link] in the MODELS statement. To enable
the quantum tunnelling model for both, specify QTUNN in the MODELS statement.

Table 3-56. MODELS Statement

Parameter Type Default

QTUNN Logical False

[Link] Logical False

[Link] Logical False

SILVACO International 3-91

ATLAS User’s Manual

It can be used with the semi-classical model for CARR=0, CARR=1 or CARR=2, or with the Schrodinger-
Poisson model with CARR=0. It is a post-processing model only and is calculated after obtaining the
solution for a given bias state.
Implementation details differ depending on whether the mesh is generated by ATLAS itself or is
created by ATHENA or DEVEDIT. If you’re using an ATLAS mesh, then the x-y direction of tunnelling
current will be automatically determined according to the device geometry. If you’re using an ATHENA
or DEVEDIT mesh, you can specify a rectangular mesh to use for calculating the quantum tunnelling
current by including the [Link] and [Link] statements. The direction of tunnelling will be
assumed to be in the y-direction unless the [Link] parameter in the MODELS statement is set to 1.
In which case, the tunnelling current will be assumed to be in the x-direction.

Table 3-57. [Link] and [Link] Statements

Parameter Type Units Default

NODE Int -999

LOCATION Real microns -999

X Real microns -999

RATIO Real microns 1.0

SPACING Real microns -999

A rectangular mesh must be specified if the [Link] parameter is specified to solve the
Schrodinger equation on a mesh generated from the [Link] and [Link] statements. In this
case, there must also be a degree of overlap between the SP mesh and QT mesh with the QT mesh
encompassing the oxide layer. We strongly recommend that the Schrodinger-Poisson mesh ends at the
semiconductor-oxide interface. If the potential profile near to the interface is one without bound states
for a given carrier type, then the contribution of that carrier type to the quantum tunnelling current is
zero (unless the semi-classical model is specified).
It is unnecessary to specify this extra mesh for a semi-classical simulation on a general unstructured
mesh. When the oxide geometry is non-planar, we recommend not to use [Link] and [Link] for
a semi-classical simulation. The quantum tunnelling model is essentially one dimensional. Therefore,
the default behavior for a non-planar oxide geometry is to consider the tunnelling direction as being
perpendicular to the semiconductor/oxide interface segment it is being calculated across.
To enable an alternative algorithm, specify SHAPEOX in the MODELS statement. In this case, the
shortest distance from a given interface point to the contact is used as the direction of tunnelling.
Since the tunnelling current depends strongly on the oxide thickness, this model will generally give a
higher value of tunnelling current than the default model.

Table 3-58. MODELS Statement

Parameter Type Default

[Link] Real 0
SHAPEOX Logical False

3-92 SILVACO International

 Physics

Carriers that tunnel through an oxide are added to the current of the electrode into or from which they
flow. If they tunnel into a polysilicon region with several contacts attached, then the tunnel current is
added to the electrode that is nearest to the segment of the oxide/polysilicon interface across which the
current is being calculated. The NEARFLG parameter in the MODELS statement is automatically set for
quantum tunnelling. Therefore, the algorithm used to obtain the nearest electrode is the same as when
you set NEARFLG. To exclude any electrode from this algorithm, set the EXCLUDE_NEAR flag in the
CONTACTS statement.
You can also set the effective mass to use inside each contact for the tunnelling current calculation
using the [Link] (electrons) and [Link] (holes) parameters in the CONTACTS statement.
If the contact is polysilicon, then the default effective mass is either the conduction band or valence
band density of states effective mass depending on its dopant specification.

The tunnelling current depends strongly on the value of the effective mass in the oxide region. This
quantity (MC for electrons and MV for holes) can be used as a parameter to calibrate the model. To set
the values, use the MATERIAL statement. For example:
MATERIAL MATERIAL=OXIDE MC=0.6 MV=0.2
The direct quantum tunnelling current is output as a variable called (A) or (A/µ ). The
current obtained using the Fowler-Nordheim model is also output as (A) or (A/µ ). It is
impossible to specify direct quantum tunnelling and the Fowler-Nordheim model on the same MODELS
statement. Therefore, or will refer to the current obtained using whichever of the two

SILVACO International 3-93

ATLAS User’s Manual

where N(x,t) represents the trapped electron (hole) density, at the interface point x, at time=t during a
transient simulation. The NTA and NTD parameters represent the acceptor and donor-like trap
densities at time=0. The Jinj,n(x,t) and Jinj,p(x,t) parameters are the injected electron and hole current
densities, SIGMAE and SIGMAH are the capture cross section of electrons and holes.
To activate this model, use the DEVDEG, DEVDEG.E, and DEVDEG.H parameters in the MODELS
statement (to account for both hot electron and hole injection, hot electron or hot hole injection,
respectively). The model parameters are user-definable on the DEGRADATION statement.

Table 3-60. User-Definable Parameters for Equations 3-351 and 3-352

Statement Parameter Units

DEGRADATION SIGMAE cm2

DEGRADATION SIGMAH cm2

DEGRADATION NTA/[Link] cm2

DEGRADATION NTD/[Link] cm2

The results of stress simulation can be used to calculate the characteristics of the degraded device (the
shift of the threshold voltage, transconductance degradation, and so on). You can view the distribution
of traps, hot electron (hole) current density, and trapped electron (hole) distribution by using
TONYPLOT.
The model parameters: NTA, NTD, SIGMAE, and SIGMAH can also be defined through the C-
INTERPRETER functions: [Link], [Link], [Link], and [Link]. This allows you to define these
values as functions of their position (x,y) along the insulator-semiconductor interface. These C-
function libraries are also defined on the DEGRADATION statement. More information on the C-
INTERPRETER functions can be found in Appendix A: “C-Interpreter Functions”.

3.6.7: The Ferroelectric Permittivity Model

Ferroelectric materials exhibit high dielectric constants, polarization and hysteresis. Such materials
are finding more and more applications in integrated memory devices. To simulate these effects, a
modified version of the ferroelectric model from Miller [117.] has been implemented.
To enable the Ferroelectric Model, set the FERRO parameter in the MODELS statement. In this model
the permittivity used in Poisson’s Equation (Equation 3-1) is given the following functional form:

– [Link]
ε( ) = [Link] + [Link] • δ • ---------------------------------------------- 3-353
δ

where [Link] is the permittivity, is the electric field and δ is given as follows:

+ [Link] ⁄ [Link] – 1
δ = [Link] -------------------------------------------------------------------------------------- 3-354
– [Link] ⁄ [Link]

The [Link], [Link], [Link], and [Link] parameters can be modified in the
MATERIAL statement (see Table 3-61).
The permittivity in Equation 3-352 can be replaced with a user-defined expression with the
C-INTERPRETER. The [Link] parameter of the MATERIAL statement (see Chapter 18: “Statements”,
Section 18.24: “MATERIAL”) defines the file that contains the C-function. This function allows the
permittivity to be position and field dependent.

3-94 SILVACO International

 Physics

For more information about C-Interpreter, see Appendix A: “C-Interpreter Functions”.

The derivative of the dipole polarization with respect to electric field is given by:

----------- = Γ----------------- 3-355

where is the position dependent dipole polarization. A numeric integration of this function is
carried out in ATLAS to determine the position dependent dipole polarization.
For saturated loop polarization, the Γ function is equal to unity, which corresponds to the default
model. If you specify the [Link] parameter in the MODELS statement, the Γ function will take
on a more general form suitable for simulation of unsaturated loops. In this case, the Γ function is
given by:

 – 1 ⁄ 2
Γ = –  --------------------------
-  3-356
ξ – 

where ξ for increasing fields and ξ for decreasing fields.

Table 3-61. User-Specifiable Parameters for Equations 3-353 to 3-354

Statement Parameter Default Units

MATERIAL [Link] 0.0 V/cm

MATERIAL [Link] 1.0

MATERIAL [Link] 0.0 C/sqcm

MATERIAL [Link] 0.0 C/sqcm

3.6.8: Polarization in Wurtzite Materials

Polarization in wurtzite materials is characterized by two conponents, spontaneous polarization, ,
and piezoelectric polarization, . Therefore, the total polarization, , is given by:

PSP 3-357

where PSP is specified on the MATERIAL statement and specifies the total spontaneous polarization,
, for the given material(s). The piezoelectric polarization, , is given by:

–
 E31 – C13 E33 3-358
 C33 

where E31 and E33 are piezoelectric constants, and C13 and C33 are elastic constants all specified in
the MATERIAL statement. The parameter is the lattice constant of the material layer in question,
which can be specified by the ALATTICE parameter of the MATERIAL statement. The parameter is
the average value of the lattice constants of the layers directly above and below the layer in question.
To enable the polarization model, specify POLARIZATION in the REGION statement for the region for
which you wish to characterize polarization effects. Typically, this will be a quantum well layer or
active layer.

SILVACO International 3-95

ATLAS User’s Manual

The polarization enters into the simulation as a positive and negative fixed charges appearing at the
top (most negative Y coordinate) and bottom (most positive Y coordinate) of the layer in question. By
default, the positive charge is added at the bottom and the negative charge is added at the top. You
can modify the sign and magnitude of this charge by specifying [Link] in the REGION
statement. This parameter is multiplied by the polarization determined by Equation 3-357 to obtain
the applied charge. The default value for [Link] is 1.0.
In some cases, the introduction of polarization charges may introduce difficulties with convergence due
to problems with initial guess. If these problems arise, you can use the PIEZSCALE parameter of the
SOLVE statement to gradually introduce the effects of polarization. This parameter defaults to 1.0 and
is multiplied by the net charge given by the product of the results of Equation 3-358 and the
[Link] parameter.
Table 3-62 shows the parameters of the wurtzite polarization model.

Table 3-62. User Specifiable Parameters of the Wurtzite Polarization Model

Statement Parameter Default Units

MATERIAL PSP see Table B-21 cm-2

MATERIAL ALATTICE see Table B-21 Å

MATERIAL E13 see Table B-21 cm-2

MATERIAL E33 see Table B-21 cm-2

MATERIAL C13 see Table B-20 GPa

MATERIAL C33 see Table B-20 GPa

3-96 SILVACO International

 Physics

3.7: Quasistatic Capacitance - Voltage Profiles

Quasistatic Capacitance is calculated by specifying the QSCV parameter in the SOLVE statement. The
quasistatic capacitance is obtained for the electrode (whose bias is being ramped) by subtracting the
electrode charge on the electrode at one bias from that at the adjacent bias and dividing by the Voltage
increment. The charge density is calculated by applying Gauss' Flux Theorem to the electrode. This
gives the capacitance at the midpoint between the two bias points. ATLAS does not correct the output
for this because the voltage increment should be sufficient fine. Therefore, this small shift is negligible.
A fine voltage increment will also give a good approximation to the continuous derivative.
QSCV will work with CARRIERS=0, 1 or 2 specified in the METHOD statement. If you link electrodes
using the COMMON parameter of the CONTACT statement, then the capacitance for each one and the sum
of capacitances will be calculated. If you specify MULT and FACTOR for a linked electrode, then the
capacitance will be calculated for the appropriate bias points, but will be stored as a function of the
bias applied to the COMMON electrode. In this case, scaling or shifting of the C-V curve may be
necessary.

SILVACO International 3-97

ATLAS User’s Manual

3.8: Conductive Materials

In certain cases, it may be advantageous to simulate metal conductivity directly rather than handling
electrodes as boundaries. You can define metal regions as “conductive” by specifying the CONDUCTOR
parameter of the REGION statement. This might be useful, for example in simulating self-heating in
metal regions.
When the metal is treated as a conductor, the conduction equation for all points in the region are
solved as follows:

⁄ 3-359

where is the current density, is the electric field, and RESISTIVITY is the metal resistivity in
metal resistivity in Ω/cm. To specify the metal resistivity, use the RESISTIVITY parameter from the
MATERIAL statement. To specify the thermal coefficient of resistivity, use the DRHODT parameter from
the MATERIAL statement.

3-98 SILVACO International

 Physics

3.9: Optoelectronic Models

This section discusses various physical models used to simulate optoelectronic devices. These models
predict fundamental optoelectronic processes, such as absorption and gain and radiative
recombination rates versus material composition, temperature and optical wavelength.
These models are based on various band theories and account for the following:
• the existence of multiple valence bands supporting multiple optical transitions,
• asymmetry in conduction band effective masses,
• the effects of strain on the band parameters,
• the effects of quantum confinement on allowable transitions.
In the following paragraphs, we will discuss three optoelectronic models.
With respect to their various principal investigators they are named the Yan model, the Li model and
the Chuang model. To activate these models, specify YAN, LI or CHUANG in the MODELS statement. Yan
and Li's models are applicable to zincblende materials while Chuang's model applies to wurtzite
materials. Yan's model is based on a single valence band, Li's is based on two valence bands (light
holes and heavy holes), and Chuang's model is based on three valence bands (light holes, heavy holes
and crystal split-off holes).
In addition to the band structure based models, there are certain, more general but less physical
models that are also mentioned. All of these models are used in the following applications:
• General drift-diffusion to account for radiative recombination for any semiconductor device,
• Laser and VCSEL simulation in both stimulated emission gain and spontaneous emission (see
Chapters 8: “Laser: Edge Emitting Simulator” and 9: “VCSEL Simulator”),
• LED light emission (see Chapter 11: “LED: Light Emitting Diode Simulator”).

3.9.1: The General Radiative Recombination Model

At the most, fundamental level the radiative recombination model described in Equation 3-245
describes all the most salient features of radiative recombination except spectral content. You can,
however, use this model in all three applications mentioned above. It has the advantages of being fast,
simple and easily calibrated.
To enable the general model, specify OPTR in the MODEL statement. This will enable radiative
recombination using the general model in the drift diffusion part of the simulation. To enable this
model in LASER or VCSEL, disable the default model for LASER and VCSEL by specifying
^SPONTANEOUS in the LASER statement. To use the general model for LED, make sure no other
competing mechanisms are enabled. The disadvantage of using this model for any light emission
application is that it lacks spectral information.
When used in LASER and VCSEL the spontaneous recombination rate is given by:

EMISSION FACTOR ⋅ COPT ⋅  ⋅ – 


r ( , ) = ------------------------------------------------------------------------------------------------------------------------------------ 3-360

where and are the electron and hole concentrations, is the intrinsic concentration, is the
number of longitudinal modes, EMISSION_FACTOR and COPT are user-defined parameters on the
MATERIAL statement. COPT accounts for the radiative rate in all directions and energies.
EMISSION_FACTOR represents the fraction of energy coupled into the direction of interest and in the
energy range of interest. Note that Equation 3-360 contains no spectral information.

SILVACO International 3-99

ATLAS User’s Manual

3.9.2: The Default Radiative Recombination Model

The default spontaneous radiative recombination model is given by

3-361

where:
• is the speed of light,
• h Planck's constant,
• Boltzman's constant,
• is the energy bandgap,
• and are the conduction and valence band edge energies,
• is the lattice temperature,
• and are the electron and hole quasi-Fermi energies,
• ω is the emission frequency that co

3-100 SILVACO International

 Physics

• ω is the emission frequency,

• and are the valence and conduction band edge energies,
• the function is defined in Equation 3-363,
• GAMMA and GAIN0 user-definable parameters specified on the MATERIAL statement.
Table 3-63 describes the user defined parameters of Equation 3-364.

Table 3-63. User-Specifiable Parameters for Equation 3-364

Statement Parameter Default Units

MATERIAL GAIN0 2000.0 cm-1

MATERIAL GAMMA

If GAMMA in Equation 3-364 is not specified, it is automatically calculated from the following
expression:

GAMMA 3-365

 
 
 

where and are the conduction and valence band densities of states.

3.9.4: The Empirical Gain Model

The emipircal gain model is enabled by specifying [Link] in the MODELS statement. The model
is described by the following expression:

( ) = GAIN00 + GAIN1N ⋅ + GAIN1P ⋅ 3-366

+ GAIN2NP ⋅ + GAIN1MIN ⋅ ( )

where and are the electron and hole concentrations, and GAIN00, GAIN1N, GAIN1P, GAIN2NP and
GAIN1MIN are user-specified parameters from the MATERIAL statement. Note that the empirical model
contains no spectral dependency and should not be used for LASER or VCSEL simulations with multiple
logitudinal modes. Table 3-64 shows the user-definable parameters for Equation 3-366.

Table 3-64. User-Specifiable Parameters for Equation 3-366

Statement Parameter Default Units

MATERIAL GAIN00 -200.0 cm-1

MATERIAL GAIN1P 0 cm2

MATERIAL GAIN2NP 0 cm2
MATERIAL GAIN1NP 0 cm5
MATERIAL GAIN1MIN 3.0×10-16 cm2

SILVACO International 3-101

ATLAS User’s Manual

3.9.5: Tayamaya's Gain Model

Tayamaya's gain model [128] is enabled by specifying TAYAMAYA in the MODELS statement.

3-102 SILVACO International

 Physics

Figure 3-4: Simulation Flow For Physically Based Optoelectronic Models

First, strain is introduced either directly or calculated from lattice mismatch and used to calculate the
strain effects on band calculations. The strain is also introduced to polarization calculations that enter
directly into the drift diffusion calculations as polarization fields.
Next, the band parameters (band edges and effective masses) are used directly in the drift diffusion
simulations as well as feeding into the solutions of Schrodinger's equations to calculate the bound
state energies. The band parameters and bound state energies are then used to calculate gain,
radiative recombination rate and optical absorption.

3.9.7: Yan's and Li's Models for Gain and Radiative Recombination in Zincblende
Materials
You can select one of two models for optical gain and spontaneous recombination. If the LI parameter
is specified, the model by Li [136] will be used. If the YAN parameter is specified the model by Yan [137]
will be used. The difference between the models is that in the YAN Model only one valence band is
accounted for, whereas the LI Model accounts for both light and heavy holes in the valence band.
The first step in calculating the gain and radiative recombination is to calculate the bound state
energies. The band edge energies and effective masses are used to calculate the quantum well bound
state energies through the Schrodinger's Equation (See Chapter 13: “Quantum: Quantum Effect
Simulator”, Equations 13-1 and 13-2). This calculation is performed in the same manner as the Self-
Consistent Coupled Schrodinger Poisson Model. The calculation of the bound state energies is
performed over a discrete domain that isn’t the same as the device simulation mesh. The discrete
domain is specified by the [Link] and [Link] parameters in the REGION statement.
The [Link] and [Link] parameters specify the number of uniform mesh locations in the bounding
box where the solution of Schrodinger's Equation is performed to extract bound state energies.
Typically, [Link] should be set to a large number so that there will be several samples per well.
[Link] should be comparable to the number of grid lines in the device mesh over the same extent in
the X direction.

SILVACO International 3-103

ATLAS User’s Manual

Once the bound state energies are calculated, the bulk momentum matrix element is calculated as
shown in Equation 3-368.

  ( DELTA )
 –  3-368
 ∗   DELTA 
 
where is the effective mass, is the bandgap, and [Link] is the user-specifiable spin-orbital
splitting energy as described in Table 3-66.

Table 3-66. User-Specifiable parameters for Equation 3-368

Statement Parameter Default Units
MATERIAL [Link] 0.341 eV

Next, the bulk momentum matrix element is used to calculate the quantum well matrix element. If the
YAN Model is specified, Equation 3-369 will be used to calculate the quantum well momentum matrix
element.

3-369

If the LI Model is specified the matrix elements for light and heavy holes are given by Equation 3-370.

3-370

where and are anisotropy factors for heavy and light holes.
The anisotropy factors depend upon whether TM or TE modes are the dominant modes [136]. The
dominant mode is user-specifiable by using the [Link] parameter of the MATERIAL statement.
When this parameter is true (default), the TE mode models are used and when false the TM mode
model is used.
For TE modes, the values of the anisotropy factors are given by Equation 3-371.
3+3 ⁄ 5–3 ⁄
= ------------------------------, = ----------------------------- ( > )
4 4 3-371
= 3 ⁄ 2, = 1⁄2 ( ≤ )

For TM modes, the values of the anisotropy factors are given by Equation 3-372.
3–3 ⁄ 1+3 ⁄
= ----------------------------- = ----------------------------- ( > )
2 2
= 0 = 2 ( < )
3-372
where is the transition energy and is the energy difference between the conduction bound
state and the valence band state.
The bound state energies and effective masses are then used to calculate the Fermi functions given in
Equation 3-373.

3-104 SILVACO International

 Physics

   – 1
′ =  1 + exp  – --------- ( – )– ⁄ 
   
3-373
   – 1
′ =  1 + exp  – --------- ( – )– ⁄ 
   

where is the hole Fermi level, is the electron Fermi level, is the valence band energy, and
is the conduction band energy.
Next, the reduced effective masses are used to calculate the 2D density of states as in Equation 3-374.

ρ = ------------ 3-374
2
πh
where is the quantum well width.
Next, the optical mode density is calculated as given by Equation 3-375.
3 2
( ) = ------------------- 3-375
2 3 3
π h

Optical Gain Models

For the LI Model, the optical gain is given by Equation 3-376.

 0  =  2π 2  ε1 
  -
 ----- ( ′ – ′ )  ------ ρ ⋅ WELL · GAIN
h  
3-376
2 2 2hω  2
=  --------  -----------------------
   
0  4ε ε ω 2
1 0

where is the Hamiltonian matrix element, w is the frequency, and is either or and
[Link] is a user-specifiable scale factor on the REGION statement.
For the YAN Model, the optical gain is given by Equation 3-377.

 2 
π h 2
( ) =  -------------------------- ρ⋅ ( )( ′ – ′) 3-377
 2 
 ε0 0 
Spontaneous Recombination Models
For the LI Model, the spontaneous recombination is given by Equation 3-378.

2π 2
( ) = ∑  -----h- ′ (1 – ′ ) ( )ρ 3-378
,
For the YAN Model, the spontaneous recombination is given by Equation 3-379.

2π 2  2 h ω ⁄ ε 2
( ) =  ------  --------  ----------------- ρ⋅ ( ) ⋅ ′ (1 – ′ ) 3-379
 h    2
0 4ω 

SILVACO International 3-105

ATLAS User’s Manual

3.9.8: Chuang's Three Band Model for Gain and Radiative Recombination in Wurtzite
Materials
Chuang's model [148] is derived from the k*p method for three valence bands in wurtzite crystalline
structure. Given the assumptions of parabolic bands, no valence band mixing and momentum
approaching zero, the following approach can be used. First, we solve the strain tensor as given in
Equations 3-380 through 3-382
–
ε = ε = 3-380

ε – ε 3-381

ε ε ε 3-382

where is the lattice constant in the strained layer, is the bulk layer lattice constant, and
are elastic constants and ε , ε , ε , ε , ε and ε are the strain tensor.
Next, we calculate the parameters θε and λε from Equations 3-383 and 3-384

θε ε (ε ε ) 3-383

λε 3-384

where , , and are shear deformation potentials. Next, we can calculate the valence band
energies from Equations 3-385 through 3-387

3-385

3-386

3-387

where ∆ , ∆ and ∆ are split energies and is the valence band reference level. Next, we can
calculate the hydrostatic energy shift from Equation 3-388
3-388

where and are hydrostatic deformation potentials. From which we can calculate the conduction
band energy as given in Equation 3-389

3-389

where is the energy bandgap.

3-106 SILVACO International

 Physics

Next, we can calculate the effective masses in the various bands using the expressions in Equations 3-
390 through 3-395
–
– ( ) 3-390

–
– ( ) 3-391

 –
– λε 
–   3-392
 
 – 

 –
– λε 
–   3-393
 
 – 

 –
–λ 
 ε
– 3-394
 
 – 

 –
– λε 
–   3-395
 
 – 

where φ is the free space mass of an electron, , , , , and are the

effective masses for heavy holes, light holes and crystal split off holes in the axial and transverse
directions and , , and are hole effective mass parameters.
The momentum matrix elements for the various transitions can be calculated by Equations 3-408
through 3-401
3-396

  3-397
 

  3-398
 

⊥ 3-399

⊥   3-400
 

⊥  
  3-401

SILVACO International 3-107

ATLAS User’s Manual

where and are given by Equations 3-402 and 3-403, and are given by Equations 3-406

and 3-407. The values of and are given by Equations 3-404 and 3-405.

 ⁄ h  3-402


 ⁄ h  3-403


 ( ∆ ∆ )( ∆ )– ∆
h
= ----------  –  -------------------------------------------------------------------------

3-404
( ∆ )
 

h
  [( ∆ ∆ )( ∆ )– ∆ ]
= ----------  –  ------------------------------------------------------------------------------------- 3-405
  ( ∆ ∆ )( ∆ ) –∆

 – λε 
=  

3-406
 – 

 – λε 
=  

3-407
 – 

In Equations 3-402 through 3-407, and are the transverse and axial conduction band effective
masses.
The default values for the parameters of Equations 3-380 through 3-407 are shown in Appendix B:
“Material Systems”, Section B.8: “Material Defaults for GaN/InN/AlN System”.

3.9.9: Lorentzian Gain Broadening

Gain broadening due to intra-band scattering can be introduced by specifying LORENTZ in the MODEL
statement. The following equation describes gain broadening when it’s applied.
( )
∫ WELL · GAMMA0 [( – ) ⁄ WELL · GAMMA0 ] 3-408

The Lorentzian shape function is given by Equation 3-409.

1 WELL · GAMMA0
( – ) = --- ⋅ -----------------------------------------------------------------------------------------
- 3-409
π 2 2
( – ) + WELL · GAMMA0
where WELL.GAMMA0 is user-specifiable in the MATERIAL statement.

3-108 SILVACO International

 Physics

3.9.10: Ishikawa's Strain Effects Model

The strained layer InGaAsP and InGaAlAs quantum well material models from [133] are implemented
in the ATLAS simulator. To enable these models, specify the ISHIKAWA parameter in the MQW
statement (see Chapter 18: “Statements”, Section 18.30: “MQW”). Strain percentages are specified by
the STRAIN parameter in the MQW statement. STRAIN between a substrate material with lattice
constant, , and an epitaxial material without STRAIN, with lattice constant, , is ( - )| .
When you enable this model, the band edge parameters for materials in the InGaAsP and InGaAlAs
systems are calculated using Equations 3-410, 3-411,and 3-412.

InGaAs
2
= 1.040 – 0.0474 STRAIN + 0.003303 STRAIN

, = 0.3331 + 0.05503 STRAIN – 0.002212 STRAIN

, = 0.331 – 0.01503ε – 0.003695 STRAIN

3-410
InGaAsP
) STRAIN
= ( 0.6958 + 0.4836 ) – 0.03031 STRAIN
= ( 0.5766 – 0.3439 ) – 0.3031 STRAIN
,
3-411
) STRAIN 0
= ( 0.6958 + 0.4836 ) + 0.003382 STRAIN

, = ( 0.6958 – 0.5164 ) + 0.003382 STRAIN

InGaAlAs
) STRAIN < 0
= ( 0.5766 + 0.6561 ) – 0.02307 STRAIN

, = ( 0.5766 – 0.3439 ) – 0.2307 STRAIN

3-412
) STRAIN > 0
= ( 0.5766 + 0.06561 ) + 0.01888 STRAIN

, = ( 0.5766 – 0.3439 ) + 0.01888 STRAIN

A STRAIN parameter has also been added to the REGION statement to account for strain in the bulk
materials. Note that for the InGaAs material system, the equations ignore composition fraction and
variation in the parameters is accounted strictly through strain (see Equation 3-410). The
compositional variation can be seen in from [136].
These band edge parameters are used through all subsequent calculations. Most importantly, the band
edges are used in solving the Schrodinger's Equation (See Chapter 13: “Quantum: Quantum Effect
Simulator”, Equations 13-1 and 13-2) to obtain the bound state energies in multiple quantum wells.
If you enable the Ishikawa Model, this would include the effects of strain in the valence and conduction
band effective masses as described in [136]. The effects of strain are introduced in Equation 3-413.

SILVACO International 3-109

ATLAS User’s Manual

1 2
----------- = ASTR + BSTRSTRAIN + CSTRSTRAIN
3-413
1 2
------------- = DSTR + ESTRSTRAIN + FSTRSTRAIN

The ASTR, BSTR, CSTR, DSTR, ESTR, and FSTR parameters are user-definable in the MQW statement.
You can also choose the appropriate values for these parameters from Tables 3-3-67 or 3-3-68.

Table 3-67. In-Plane Effective Mass of InGaAsP Barrier (1.2µm)/InGaAs(P) Well and
InGaAlAs Barrier (1.2µm)/InGa(Al)As Well System for 1.55µm Operation [136]

−2.0<Strain(%)<−0.5 0<Strain(%)<2.0

ASTR BSTR CSTR DSTR ESTR FSTR

Lattice-matched InGaAsP −4.238 −6.913 −1.687 4.260 2.253 −0.584

barrier/InGaAs well

InGaAsP (1.6-µm) well −0.936 −4.825 −1.754 2.986 5.505 −1.736

Strain-compensated InGaAsP −7.761 −13.601 −5.100 4.166 1.933 −0.558

barrier/InGaAs well

Lattice-matched InGaAlAs −3.469 −6.133 −1.481 3.9134 2.336 −0.633

barrier/InGaAs well

InGaAlAs(1.6-µm) well −1.297 −5.598 −2.104 2.725 6.317 −1.766

Strain-compensated InGaAlAs −5.889 −9.467 −2.730 4.193 1.075 −-0.155

barrier/InGaAs well

Table 3-68. In-Plane Effective Mass of InGaAsP Barrier (1.1µm)/InGaAs(P) Well and
InGaAlAs Barrier (1.1µm)/InGa(Al)As Well System for 1.30µm Operation [136]

−2.0<Strain(%)<−0.5 −0.5<Strain(%)<2.0

ASTR BSTR CSTR DSTR ESTR FSTR

Lattice-matched InGaAsP −9.558 −8.634 −1.847 4.631 1.249 −0.313

barrier/InGaAs well

InGaAsP (1.4-µm) well −2.453 −4.432 −1.222 3.327 3.425 −1.542

Strain-compensated InGaAsP 4.720 0.421 −0.014

barrier/InGaAs well

Lattice-matched InGaAlAs −8.332 −8.582 −2.031 3.931 1.760 −0.543

barrier/InGaAs well

InGaAlAs(1.4-µm) well −3.269 −5.559 −1.594 3.164 4.065 −1.321

Strain-compensated InGaAlAs 4.078 0.516 −0.0621

barrier/InGaAs well

To choose these values, use the STABLE parameter in the MQW statement. You can set this parameter to
the row number in the tables. The rows are numbered sequentially. For example, the first row in Table
3-67 is selected by specifying STABLE=1. The first row of Table 3-68 is selected by specifying
STABLE=7.

3-110 SILVACO International

 Physics

You can also choose to specify the effective masses directly. The conduction band effective mass is
specified by the MC parameter in the MQW statement. There are two ways to specify the valence band.
One way, is to use the MV parameter to specify a net effective mass. The other way, is to use the MLH
and MHH parameters to specify the light and heavy hole effective masses individually.
If you don’t specify the effective masses by using one of these methods, then the masses will be
calculated from the default density of states for the given material as described in Section 3.4.2:
“Density of States”.
The effective masses described are also used to solve the Schrodinger Equation (see Chapter 13:
“Quantum: Quantum Effect Simulator”, Equations 13-1 and 13-2).

SILVACO International 3-111

ATLAS User’s Manual

This page is intentionally left blank.

3-112 SILVACO International

4.1: Overview
S-PISCES is a powerful and accurate two-dimensional device modeling program that simulates the
electrical characteristics of silicon based semiconductor devices including MOS, bipolar, SOI,
EEPROM, and power device technologies.
S-PISCES calculates the internal distributions of physical parameters and predicts the electrical
behavior of devices under either steady-state, transient, or small signal AC conditions. This is
performed by solving Poisson’s Equation (see Chapter 3: “Physics”, Section 3.1.1: “Poisson’s Equation”)
and the electron and hole carrier continuity equations in two dimensions (see Chapter 3: “Physics”,
Section 3.1.2: “Carrier
arbvltlciy12
pi(n)5.5(g
Continuity9.96
prfl)4-6illsna(n)5.5nd
0 .3 96 0 .t(ion)-3.6(s”)]TJETEMC/Span
the(stducure( of the devl)467(ce
<</MCID113>>BDCBT9.96
a“)1072(y2383(bhe
0 0 9.96
)]T

SILVACO International 4-1

ATLAS User’s Manual

4.2: Simulating Silicon Devices Using S-PISCES

4.2.1: Simulating MOS Technologies
Physical Models for MOSFETs
S-PISCES provides special physical models tuned for simulating MOS devices. Most of these models are
accessed from the MODEL statement. The MOS parameter of the MODEL statement can be specified to
turn on a default set of physical models that are most useful for MOS simulation. The MOS parameter
enables Shockley-Read-Hall (SRH), Fermi Statistics (FERMI), and the Lombardi Mobility model (CVT)
for transverse field dependence. To set the default MOS simulation models , use:
MODEL MOS PRINT
The transverse field dependent mobility models are of particular importance for simulating MOS
devices. S-PISCES currently supports several different transverse field dependent mobility models. The
CVT parameter selects the Lombardi CVT model. The YAMA parameter selects the Yamaguchi model.
The TASCH parameter selects the Tasch model. The WATT parameter selects the Watt Surface Model,
which can be operated in a more accurate mode with the extra parameter, [Link], on the MOBILITY
statement.
You will find that the MOBILITY statement can be used to modify some of the parameters of the
various models, to apply different models to different regions, or to apply different models to electrons
and holes.

Meshing for MOS Devices

In device simulation of MOS devices, the key areas for a tight mesh are:
• very small vertical mesh spacing in the channel under the gate. The exact size of mesh required
depends on the transverse field or surface mobility model chosen. See Figure 4-1 for an example of
the effect of mesh size on drain current.
• lateral mesh spacing along the length of the channel for deep sub-micron devices. This is required
to get the correct source-drain resistance and to resolve the channel pinch-off point.
• lateral mesh at the drain/channel junction for breakdown simulations. This is required to resolve
the peak of electric field, carrier temperature and impact ionization.
• several vertical grid spacings inside the gate oxide when simulating gate field effects such as gate
induced drain leakage (GIDL) or using any hot electron or tunneling gate current models
Figures 4-1 and 4-2 show the effect of mesh size in the MOS channel on IV curves. In Figure 4-1, the
mesh density in the vertical direction is increased. As the mesh density increases, the resolution of the
electric field and carrier concentration is improved. This example uses the CVT mobility model.
Improvements in transverse electric field resolution lead to a reduced mobility in the channel and a
stronger IV roll-off.
But Figure 4-2 shows the effect of surface channel mesh in MOSFETs is model dependent. This result
shows the current at Vds=3.0V and Vgs=0.1V versus the size of the first grid division into the silicon.
Results vary for each model but note that for all models a very fine grid is required in order to reduce
the grid dependence to acceptable levels.

4-2 SILVACO International

 S-Pisces: 2D Device Simulator

Figure 4-1: Effect on MOS IV curve of progressive refinement of the vertical mesh spacing at the surface of
the MOS channel

Figure 4-2: Effect of surface mesh spacing on simulated current for several MOS Mobility Models

SILVACO International 4-3

ATLAS User’s Manual

MOS Electrode Naming

For MOS simulation, S-PISCES allows you to use standard electrode names to reduce confusion with
the use of electrode indices. These names include: source, drain, gate, and substrate. Electrode
names can be defined in ATHENA or DEVEDIT or in the ELECTRODE statement in ATLAS. These
names can be used in the SOLVE statements for setting bias voltages such as:
SOLVE VGATE=1.0 VSTEP=1.0 VFINAL=5.0 NAME=GATE

Gate Workfunction
In MOS simulations, the workfunction of the gate is an important parameter. This must be set in each
input deck using the WORK parameter of the CONTACT statement. For example:
CONTACT NAME=GATE WORK=4.17
would set the workfunction on the gate at 4.17 eV.
Certain material names can also be used to set the workfunction of common gate materials. For
example:
CONTACT NAME=GATE [Link]
would set the workfunction on the gate to that of n type polysilicon.

Note: The gate workfunction should be set on a CONTACT statement even though the material or workfunction might be set
from ATHENA or DEVEDIT.

Interface Charge
For accurate simulation of MOS devices the interface charge at the oxide, specify semiconductor
interface. This can be done by setting the QF parameters for the INTERFACE statement. Typically, a
value of 3x1010 cm-2 is representative for the interface charge found in silicon MOS devices. The
proper syntax for setting the value for this interface fixed charge is:
INTERFACE QF=3e10
You can also try to model the fixed charge more directly by using interface traps to simulate the
surface states. This can be done by using the INTTRAP statement. But this is rarely done in practice.

Single Carrier Solutions

Frequently for MOS simulation, you can choose to simulate only the majority carrier. This will
significantly speed up simulations where minority carrier effects can be neglected. This can be done by
turning off the minority carrier. To do this, use the ATLAS negation character, ^ , and one of the
carrier parameters (ELECTRONS or HOLES) in the METHOD statement. For example, to simulate
electrons only, you can specify one of the following:
METHOD CARRIERS=1 ELECTRONS
METHOD ^HOLES
Single carrier solutions should not be performed where impact ionization, any recombination
mechanism, lumped element boundaries, or small signal AC analysis are involved.

4-4 SILVACO International

 S-Pisces: 2D Device Simulator

Energy Balance Solutions

As MOS devices become smaller and smaller, non-local carrier heating effects becomes important.
Accurate simulation of non-local carrier heating effects can be simulated using the energy balance
model (EBM). As a general rule, the gate length can be used as a gauge to predict when non-local
effects are important. Generally, for drain currents, energy balance should be applied for gate lengths
less than 0.5 microns. For substrate currents, energy balance should be applied for gate lengths less
than 1.0 micron.
To enable energy balance for electrons/holes, the [Link] or [Link] parameters should be set on
the MODEL statement. For example:
MODEL [Link]
enables the energy balance model for electrons.

ESD Simulation
In some cases, lattice heating may be important to MOS simulation. This typically occurs in cases with
very high currents, just like the case with ESD simulation. In these cases, GIGA should be used to
simulate the heat-flow in the device. To enable heat flow simulation, set the [Link] parameter of
the MODEL statement (a license for GIGA is required). For example, the statement:
MODEL [Link]
enables heat-flow simulation.

4.2.2: Simulating Silicon Bipolar Devices

Physical Models for BJTs
S-PISCES provides special physical models for bipolar device simulation. These models can be selected
using the MODEL statement. The BIPOLAR parameter of the MODEL statement enables a reasonable
default set of bipolar models. These include: concentration dependent mobility (CONMOB), field
dependent mobility (FLDMOB), bandgap narrowing (BGN), concentration-dependent lifetime (CONSRH)
and Auger recombination (AUGER).
For the most accurate bipolar simulations the recommended mobility model is the Klaassen model
(KLA). This includes doping, temperature and carrier dependence. It applies separate mobility
expressions for majority and minority carriers. This model should also be used with Klaassen’s Auger
model (KLAAUG) and Klaassen’s concentration dependent SRH model (KLASRH). The mobility model
should be combined with FLDMOB to model velocity saturation. For surface (or lateral) bipolar devices
the Shirahata model (SHI) can be used to extend the Klaassen model with a transverse electric field
dependence. The most accurate and appropriate model statement for bipolar devices is therefore:
MODELS KLA FLDMOB KLASRH KLAAUG BGN FERMI PRINT
You can choose this set of models by using the BIPOLAR2 parameter from the MODEL statement.

Note: For a complete syntax including description of models and method for simulating polysilicon emitter bipolar devices,
see the BJT directory in the on-line examples.

Meshing Issues for BJTs

The most important areas to resolve in bipolar transistors are the emitter/base and base/collector
junctions. Typically, a very fine mesh throughout the base region is required. The gain of the bipolar
device is determined primarily by the recombination at the emitter/base junction or inside the emitter.
Hence, these regions need to be resolved with a fine mesh.

SILVACO International 4-5

ATLAS User’s Manual

BJT Electrode Naming

S-PISCES also provides special electrode names for bipolar simulation that can be used to ease
confusion over electrode indices. These electrode names include: “emitter”, “base”, “collector”, “anode”,
and “cathode”. Electrode names can be defined in ATHENA or DEVEDIT or in the ELECTRODE
statement in ATLAS. The electrode names are used on SOLVE statement for example:
SOLVE VBASE=1.0 VSTEP=1.0 VFINAL=5.0 NAME=BASE

Dual Base BJTs

It is possible in S-PISCES to tie two or more contacts together so that voltages on both contacts are
equal. This is useful for many technologies for example dual base bipolar transistors. There are several
methods for achieving this depending on how the structure was initially defined.
If the structure is defined using ATLAS syntax it is possible to have multiple ELECTRODE statements
with the same NAME parameter defining separate locations within the device structure. In this case the
areas defined to be electrodes will be considered as having the same applied voltage. A single current
will appear combining the current through both ELECTRODE areas.
Similarly if two separate metal regions in ATHENA are defined using the ATHENA ELECTRODE
statement to have the same name, then in ATLAS these two electrodes will be considered as shorted
together.
If the electrodes are defined with different names, the following syntax can be used to link the voltages
applied to the two electrodes.
CONTACT NAME=base1 COMMON=base
.
SOLVE VBASE=0.1
Here, the electrode, base1, will be linked to the electrode, base Then, the applied 0.1V on base will
also appear on base1 But ATLAS will calculate and store separate currents for both base and base1.
This can be a useful feature. In some cases, however, such as where functions of the currents are
required in EXTRACT or TONYPLOT, it is undesirable. The SHORT parameter can be added to the
CONTACT statement above to specify that only a single base current will appear combining the
currents from base and base1.
When loading a structure from ATHENA or DEVEDIT where two defined electrode regions are
touching, ATLAS will automatically short these and use the electrode name that was defined first.

Creating an Open Circuit Electrode

It is often required to perform a simulation with an open circuit, such as for an open-base breakdown
voltage simulation, on one of the defined electrodes. There are three different methods that will
accomplish this. The first method is to entirely deleting an electrode from the structure file. The
second method is to add an extremely large lumped resistance, for example 1020Ω, onto the contact to
be made open circuited. The third method is to switch the boundary conditions on the contact to be
made open circuited from voltage controlled to current controlled and then specify a very small current
through that electrode.
Each of these methods are feasible but if in doing so a floating region is created within the structure,
then numerical convergence may be affected. As a result, it is normally recommended to use the
second method to ensure that no floating region is created.

4-6 SILVACO International

 S-Pisces: 2D Device Simulator

Solution Techniques for BJTs

To obtain bipolar solutions, you almost always need to simulate using two carriers. This is due to the
importance of minority carriers to device operation.
In certain cases, non-local carrier heating may be important to accurately simulate bipolar devices. In
these cases, use the energy balance model. To model non-local carrier heating for electrons/holes, set
the [Link] and [Link] parameters in the MODEL statement. For example, the statement:
MODEL [Link]
invokes the carrier heating equation for electrons.

4.2.3: Simulating Non-Volatile Memory Technologies (EEPROMs, FLASH Memories)

If you want to simulate non-volatile memory devices, first become familiar with the basics of MOSFET
simulation (see Section 4.2.1: “Simulating MOS Technologies”).

Defining Floating Gates

To simulate non-volatile memory technologies such as EEPROMs or FLASH EEPROMs, it is
necessary to specify one or more electrodes as floating gates. This is done by setting the FLOATING
parameter of the CONTACT statement. For example:
CONTACT NAME=fgate FLOATING
This specifies that the electrode named fgate is simulated as a floating gate. This means that the
charge on the floating gate is calculated as the integral of the gate current at that gate during a
transient simulation.
Modeling the correct coupling capacitance ratio between the floating gate and control gate often
requires adding an extra lumped capacitor from the floating gate to the control gate or one of the other
device terminals. This is often required since S-PISCES is performing a 2-D simulation whereas the
coupling of the gates is often determined by their 3-D geometry. Parameters on the CONTACT statement
are used to apply these extra lumped capacitances. For example, to add a capacitor of 1fF/mm between
the control and floating gates the syntax is:
CONTACT NAME=fgate FLOATING [Link]=1.0e-15 [Link]=cgate

Gate Current Models

The gate currents for the floating gate structure can be supplied by one of three sources: hot electron
injection (HEI or [Link]), hot hole injection (HHI or [Link]) and Fowler-Nordheim tunneling
current (FNORD).
These currents are of importance, depending on whether electrons are being moved onto the gate or off
the floating gate. In the case of placing electrons on the floating gate hot electron injection and Fowler-
Nordheim tunneling should be used. In the case of removal of electrons from the floating gate, hot hole
injection and Fowler-Nordheim tunneling should be set.
In drift diffusion simulations hot electron injection is simulated by setting the HEI parameter of the
MODELS statement. Hot hole injection is simulated using the HHI parameter of the MODELS statement.
Fowler-Nordheim tunneling is enabled by setting the FNORD parameter of the MODELS statement. The
following example demonstrated the proper setting for Flash EPROM programming.
MODELS MOS HEI PRINT
This next example is appropriate for EEPROM erasure.
MODELS MOS HHI FNORD PRINT

SILVACO International 4-7

ATLAS User’s Manual

With energy balance simulations, the Concannon Models should be used for EPROM programming
and erasing. For more information about these models, see Chapter 3: “Physics”, the “Concannon’s
Injection Model” section on 3-86.

Note: Writing and erasure of floating gate devices should be done using transient simulation.

Gate Current Assignment (NEARFLG)

The actual calculation of floating gate current magnitude is done at the silicon-oxide interface. The
question of distribution of oxide currents to the various electrodes near the interface is resolved using
one of two models. The actual flow of carriers in oxides is not well known. Accepted physical models of
carrier flow in oxides are still under research. As such, S-PISCES provides two heuristic models to
choose from. The default is to distribute currents calculated at points along the interface to the
electrode in the direction of highest contributing field. This model is somewhat analogous to a purely
drift model of oxide carrier transport. The alternative is to set the NEARFLG parameter of the MODEL
statement. In this case, the currents calculated at points along the interface are distributed to the
geometrically closest electrode. This model is analogous to a purely diffusion model of carrier transport
in oxide.

4.2.4: Simulating SOI Technologies

Silicon substrates are now being produced that contain an oxide layer buried below the surface of the
silicon at some predefined depth. The existence of this buried oxide layer has resulted in a change not
only in the fabrication process used to manufacture a device in the surface silicon, but also in the
challenges facing device simulation.
All of the issues raised previously about MOS device simulation should also be considered with some
extra SOI specific problems.
The most common device technology that uses these SOI substrates is the SOI MOSFET. This section
summarizes the simulation requirements for SOI using this particular technology as a reference.

Meshing in SOI devices

The mesh requirements for SOI MOSFETs is very similar to that described in the previous section for
bulk MOS transistors. In addition to these requirements, there are some additional points to meshing
these devices. These are:
• Two channel regions may exist: one underneath the top (front) gate oxide and one above the
buried (back gate) oxide.
• Inside the buried oxide layer, the mesh constraints can be relaxed considerably compared with the
top gate oxide.
• The active silicon underneath the top gate can act as the base region of a bipolar transistor and as
such may require a finer mesh when compared to bulk MOS transistors.

4-8 SILVACO International

 S-Pisces: 2D Device Simulator

Physical Models for SOI

SOI MOSFET simulations are based upon the physical operation of the device, which exhibits both
MOS and bipolar phenomena. As a result a more complex set of physical models will be required than
for either MOS or bipolar technologies. Table 4-1 shows these models.

Table 4-1: SOI Physical Models

Model Description

Mobility Klaassens Model (KLA) is recommended to account for lattice

scattering, impurity scattering, carrier-carrier scattering and
impurity clustering effects at high concentration. The
Shirahata Mobility Model (SHI) is needed to take into account
surface scattering effects at the silicon/oxide interface which is
a function of the transverse electric field. High electric field
velocity saturation is modelled through the field dependent
mobility model (FLDMOB). You can tune model parameters using
the MOBILTY statement syntax.

Interface Charge In SOI transistors, there exist two active silicon to oxide inter-
faces on the wafer. The top interface, under the top gate, is sim-
ilar to MOS technology. The bottom interface is quite different
and typically contains significantly more charge. Different
interface charges may be set in SPISCES using the INTERFACE
statement with region specific parameters.

Recombination To take account of recombination effects, we recommend the use

of the Shockley-Read-Hall (SRH) model. This simulates the
leakage currents that exist due to thermal generation. It may
also be necessary to simulate the presence of interface traps at
the silicon/oxide interface. Then, turn on the Direct
Recombination Model (AUGER). The parameters for both models
can be tuned in the MOBILITY statement.
Bandgap Narrowing This model (BGN) is necessary to correctly model the bipolar cur-
rent gain when the SOI MOSFET behaves like a bipolar tran-
sistor. The parameters for this model are specified in the
MODELS statement.

SILVACO International 4-9

ATLAS User’s Manual

Table 4-1: SOI Physical Models

Model Description

Carrier Generation Impact ionization significantly modifies the operation of SOI

MOSFETs. To account for this phenomena, switch on the
impact ionization model (IMPACT) and calibrate for SOI tech-
nology. The calibration parameters are set in the IMPACT state-
ment.

Lattice Heating When a device is switched on, there can be significant current
density within the silicon. This could generate a significant
amount of heat. In bulk MOS devices, the silicon substrates
behaves like a good heat conductor and this generated heat is
quickly removed. But, this isn’t the case with SOI substrates as
the buried oxide layer allows this generated heat to be retained.
For SOI MOSFETs, this can be a significant amount and can
drastically affect the operation of the device. In such cases, take
account of this by using the GIGA module. Note that when lat-
tice heating is switched on, by using the [Link] parameter
in the MODELS statement, you also need to specify a thermal
boundary condition with the THERMCONTACT statement. See
Chapter 7: “Giga: Self-Heating Simulator” for more details.

Carrier Heating In deep submicron designs, you may need to switch on the addi-
tional energy balance equations. These take into account the
exchange of energy between carriers and between the carriers
and the lattice. See Section 4.2.1: “Simulating MOS Technolo-
gies” for more information.

An example of a set of typical models for a partially depleted

SOI MOSFET could be:

MODEL KLA SHI FLDMOB SRH AUGER BGN [Link]

INTERFACE QF=1e10 [Link]=0.05
INTERFACE QF=1e11 [Link]=0.05
THERMCONTACT NUM=1 [Link]=4 [Link]=300
IMPACT SELB

Numerical Methods for SOI

One important issue with SOI device simulation is the choice of numerical methods. In SOI technology,
the potential in the channel (or body) region is commonly referred to as “floating”. This occurs because
there exists no direct contact to it by any electrode. As a result, when a bias is applied, or increased, on
a contact there may be some convergence problem. This occurs because the guess used in the
numerical solution scheme for (ψ, n,p) may be poor, particularly in the “floating” region. This is
particularly true if impact ionization is used. To overcome the problem of poor initial guess, the
following numerical methods syntax should be used in isothermal drift-diffusion simulations.
METHOD GUMMEL NEWTON
This method initially performs a GUMMEL iteration to obtain an improved initial guess for the NEWTON
solution scheme. Although this method is more robust, this is slower than using the NEWTON scheme
alone. For more information on the numerical schemes available, see Chapter 17: “Numerical
Techniques”.

4-10 SILVACO International

 S-Pisces: 2D Device Simulator

SOI Physical Phenomena

The physical models and the numerical schemes described above should allow S-PISCES to study all
the important SOI phenomena. These include:
• full or partial depletion effects
• threshold voltage
• subthreshold slopes
• front to back gate coupling effects
• leakage current analysis
• high frequency analysis
• device breakdown
• snapback effects
• the kink effect in the output Ids-Vds characteristics
• negative differential resistance

SILVACO International 4-11

ATLAS User’s Manual

This page is intentionally left blank.

4-12 SILVACO International

 Chapter 5:
Blaze: Compound Material 2D Simulator

5.1: Overview
Before continuing to the sections that follow, become familiar with ATLAS. If not, read Chapter 2:
“Getting Started with ATLAS” before proceeding with this chapter.
BLAZE is a general purpose 2-D device simulator for III-V, II-VI materials, and devices with position
dependent band structure (i.e., heterojunctions). BLAZE accounts for the effects of positionally
dependent band structure by modifications to the charge transport equations. BLAZE is applicable to a
broad range of devices including: HBTs, HEMTs, LEDs, heterojunction photodetectors (APDs, solar
cells, and so on) and heterojunction diodes.
This chapter is composed of several sections. Section 5.1.1: “Basic Heterojunction Definitions”
diagrams the basic heterojunction band parameters and includes a section on heterojunction
alignment. Heterojunction charge transport is covered in Section 5.1.3: “The Drift Diffusion Transport
Model”. This section includes the details of how BLAZE modifies the basic transport models to simulate
heterodevices. Section 5.2: “The Physical Models” covers the physical models unique to BLAZE.
Detailed information about the material systems encountered in heterojunction simulation is covered
in Section 5.3: “Material Dependent Physical Models”. This includes the relationships between the
compound elemental concentrations and bandgap, dielectric constant, low field mobility, and other
important material and transport parameters. Finally, Section 5.4: “Simulating Heterojunction
Devices with Blaze” covers how to define materials and models for heterojunction devices with BLAZE.
Appendix B: “Material Systems” has the defaults for these parameters.

5.1.1: Basic Heterojunction Definitions

Figure 5-1 shows the band diagrams and band parameters for a basic p-n heterojunction device under
equilibrium conditions. This diagram illustrates two materials with different bandgaps and electron
affinities and a Schottky barrier in contact to the n-type material.

E(x)
Eo

qVo El(x)
χp

χn φm
Ec(x)
Εgp
Ei(x) φb
Ef
Ev(x)
Metal
∆Εv
Εgn

x
0
Figure 5-1: Band Diagram of p-n Heterojunction

SILVACO International 5-1

ATLAS User’s Manual

Referring to Figure 5-1:

• Ec(x), Ev(x) and Ei(x)are the spatially dependent conduction band, valence band and intrinsic
energy levels respectively.
• Ef and Eo are the Fermi and Vacuum level energies.
• Egp and Egn are the p-type and n-type material bandgaps.
• ∆Ev is the portion of the difference between Egp and Egn that appears between the valence bands
at the heterojunction interface.
• ∆Ec (not labelled) is the portion of the difference between Egp and Egn that appears between the
conduction bands at the heterojunction interface.
• qVo is the built-in potential of the heterojunction.
• χp and χn are the electron affinities of the p-type and n-type materials.
• φm is the work function of the metal.
• φb is the barrier height between the metal and the semiconductor.
The basic band parameters for defining heterojunctions in BLAZE are bandgap parameter, EG300,
electron parameter, AFFINITY, and the conduction and valence band density of states, NC300 and
NV300. These parameters are defined for each material using the MATERIAL statement. Other
transport parameters relating these basic definitions to compound elemental concentrations
([Link] and [Link]) can also be defined.
See the MATERIALS section of this chapter for a description of these relationships for each material
system and the Simulating Heterojunction Devices with BLAZE section for their usage. The work
function of metals is defined using the CONTACT statement.

5.1.2: Alignment
As can be seen from Figure 5-1, the difference in the two material bandgaps creates conduction and
valence band discontinuities. How the bandgap difference is distributed between the conduction and
valence bands has a large impact on the charge transport in these heterodevices. There are three
methods for defining the conduction band alignment for a heterointerface. These methods are: the
Affinity Rule, using the ALIGN parameter on the MATERIAL statement, and manually adjusting the
material affinities using the AFFINITY parameter on the MATERIAL statement.
The Affinity Rule
The default method in BLAZE for assigning how much of the bandgap difference appears as the
conduction band discontinuity makes use of the affinity rule. The affinity rule assigns the conduction
band discontinuity equal to the difference between the two materials electron affinities (AFFINITY on
the MATERIAL statement). The affinity rule method is used by default for all materials where the
ALIGN parameter has not been defined on the MATERIAL statement.
Using the ALIGN parameter in the MATERIAL statement
Experimental measurements of the actual band discontinuities can differ from what is assigned using
the affinity rule with the standard material electron affinities. Therefore, BLAZE allows ∆Ec to be
calculated by specifying the ALIGN parameter on the MATERIAL statement. ALIGN specifies the
fraction of the bandgap difference which will appear as the conduction band discontinuity. This
bandgap difference is between the material for which the ALIGN parameter is specified and the
smallest bandgap material in the overall structure (the reference material). Internally, BLAZE creates
the desired conduction band offset by modifying the electron affinity of the material for which the
ALIGN parameter is specified.

5-2 SILVACO International

 Blaze: Compound Material 2D Simulator

In many applications, the Schottky barriers or more than two different semiconductor materials are
present. Keep the reference material bandgap and these assigned affinities in mind when defining
offsets for multiple materials or Schottky barrier heights. Examples for multiple materials and
Schottky barriers are given in the examples section.
Manually Adjusting Material Affinity
The AFFINITY parameter on the MATERIAL statement can be used in conjunction with the default
affinity rule alignment method to manually adjust the conduction band offset. In this case, the electron
affinity of the larger bandgap material is adjusted so that the difference between the two materials
affinity equals the desired conduction band offset. When more than two different materials are
present, each material affinity can be adjusted in this manner. This is the easiest method for handling
multiple materials and heterojunctions.
The following examples describe the procedure for aligning heterojunctions using these three methods
in BLAZE.
EXAMPLE 1

Eo

qVo
χ1 El(x)

χ2
Ec(x)
Εg1
Ei(x)
Ef
Ev(x)

Εg2

Figure 5-2: Band diagram of heterojunction with band offset.

Figure 5-2 diagrams a heterojunction consisting of two semiconductors with different bandgaps Eg1
and Eg2 and electron affinities χ1 and χ2. This example is similar to the bandstructure of a HFET or
HEMT. For this example, Eg1 < Eg2 and χ2 < χ1.
Allocating the conduction band offsets using the affinity rule:
5-1

and
5-2

∆Εc is the amount of the conduction band discontinuity at the heterointerface and ∆ is the amount of
the valence band discontinuity.

SILVACO International 5-3

ATLAS User’s Manual

Note: Remember the Affinity Rule is used to calculate the conduction band offset for a material, as long as the ALIGN
parameter is not specified on the MATERIAL statement for that material.

Using the ALIGN parameter on the MATERIAL statement

Let’s assign 80% of the bandgap difference between Material1 and Material2 to the conduction
band offset. Then, define the ALIGN parameter on the MATERIAL statement for Material2 using
MATERIAL NAME=Material2 ALIGN=0.80
Then:
∆ Ec = ( E g2 – E g1 ) ⋅ 0.80 5-3

Internally, the affinity of Material2 is adjusted so that ∆Εc equals this value. This value of electron
affinity will override any electron affinity specification for Material2. This has an impact on any
calculation where this materials electron affinity is used and considered when specifying Schottky
barriers contacted to this materials. See “EXAMPLE 4” on page 5-10 for more details on Schottky
barrier considerations

Manually Adjusting Material Affinity

MATERIAL NAME=Material2 AFFINITY=VALUE
where VALUE is adjusted to provide for the desired conduction band offset as calculated by the affinity
rule, that is, relative to Material1. This value of electron affinity will override any electron affinity
specification for Material2. This has an impact on any calculation where this materials electron
affinity is used and must be considered when specifying Schottky barriers contacted to this materials.
See “EXAMPLE 4” on page 5-10 for more details on Schottky barrier considerations

Note: Remember, if the ALIGN parameter is not specified on the MATERIAL statement, BLAZE will use the Affinity Rule
and either the default electron affinity or the affinity assigned using the AFFINITY parameter on the MATERIAL statement
to calculate the conduction band offsets.

5-4 SILVACO International

 Blaze: Compound Material 2D Simulator

EXAMPLE 2

Eo

qVo12 qVo23
El(x) χ2

χ1 χ3

Εg2
Ec(x)
Ef

∆Εv23
Ei(x) Εg3
∆Εv12
Ε g1

Ev(x)
Material 1 Material 2 Material 3

Figure 5-3: Band diagram of three material system (lowest Eg in center)

Figure 5-3 details a heterostructure device consisting of three semiconductors with different bandgaps
Εg1, Εg2 and Εg3 and electron affinities χ1, χ2 and χ3. This is similar to the band diagram of a Double
Heterojunction Bipolar Transistor. For this example, Εg1 > Εg2 < Εg3 and χ1 < χ2 > χ3.
Allocating the conduction band offsets using the affinity rule:
∆ E c12 = χ 2 – χ 1 5-4

and:
∆ E v12 = ∆ Eg12 – ∆ E c12 5-5

for the heterojunction between Material1 and Material2 and:

∆ E c23 = χ 2 – χ 3 5-6

and:
∆ E v23 = ∆ Eg23 – ∆ E c23 5-7

for the heterojunction between Material2 and Material3.

Using the ALIGN parameter on the MATERIAL statement:
Notice that the reference material, the material with the smallest bandgap in this case, Material2, is
located between the two larger bandgap materials, Material1, and Material3.

SILVACO International 5-5

ATLAS User’s Manual

Let’s assign 80% of the bandgap difference between Material1 and Material2 to the conduction
band offset for this heterojunction. Then, define the ALIGN parameter on the MATERIAL statement for
Material 1 using:
MATERIAL NAME=Material1 ALIGN=0.8
then:
∆ E c12 = ( E g1 – E g2 ) ⋅ 0.80 5-8

Internally, the affinity of Material1 is adjusted so that ∆Εc12 equals this value.
Let’s assign 70% of the bandgap difference between Material3 and Material2 to the conduction
band offset for this heterojunction. Defining the ALIGN parameter on the MATERIAL statement for
Material3 using:
MATERIAL NAME=Material3 ALIGN=0.70
then:
∆ Ec23 = ( E g3 – E g2 ) ⋅ 0.70 5-9

Internally, the affinity of Material3 is adjusted so that ∆Εc23 equals this value.
These new values of electron affinity for Material1 and Material3 will override any electron
affinity specification for these materials. This has an impact on any calculation where these materials
electron affinity is used and must be considered when specifying Schottky barriers contacted to these
materials. See“EXAMPLE 4” on page 5-10 for more details on Schottky barrier considerations.
Manually Adjusting Material Affinity
Assigning the conduction band offsets for each heterojunction is accomplished by setting the electron
affinities for Material1 and Material3 using the AFFINITY parameter on the MATERIAL statement.
The electron affinity for Material1 and Material3 are adjusted relative to Material2 by the
amount of the desired conduction band offset for each heterojunction. Since Material2 affinity is
larger than that for Material1 and Material3, the affinities for Material1 and Material3 are
reduced (relative to Material2) to provide the desired conduction band offset.
Let’s assume an electron affinity for Material2 of 4eV (~ that of GaAs). Let’s decide that between
Material1 and Material2, the conduction band offset is 0.3eV and that between Material3 and
Material2, the conduction band offset is 0.2eV. Then, for Material1:
MATERIAL NAME=Material1 AFFINITY=3.7
and for Material3:
MATERIAL NAME=Material3 AFFINITY=3.8
This is the easiest method to define the conduction band offsets for multiple materials. This value of
electron affinity will override any electron affinity specification. This has an impact on any calculation
where this materials electron affinity is used and must be considered when specifying Schottky
barriers contacted to this materials. See “EXAMPLE 4” on page 5-10 for more details on Schottky
barrier considerations.

Note: The band offsets are always defined with reference to the conduction band. Therefore, if a specific valence band offset
is required, the appropriate conduction band offset should be calculated from the desired valence band offset and the materials
bandgap.

5-6 SILVACO International

 Blaze: Compound Material 2D Simulator

EXAMPLE 3

Eo
qVo12
qVo23
χ1
El(x)
χ2
Ec(x) χ3
Εg1
Ei(x)
Ef
Ev(x)
∆Εv12
Εg2
Εg3

∆Εv23

Material1 Material2 Material3

Figure 5-4: Band diagram of three material system (lowest Eg not in center)
Figure 5-4 details a heterostructure device consisting of three semiconductors with different bandgaps
Εg1, Εg2 and Εg3 and electron affinities χ1, χ2 and χ3. This is similar to the “EXAMPLE 2” on page 5-
5, except that the narrow bandgap material is not located in between the other larger bandgap
materials. As will be seen, this adds extra complexity to the conduction and valence band offset
calculations. For this example, Εg1 < Εg2 < Εg3 and χ3 < χ2 < χ1.
Allocating the conduction band offsets using the affinity rule:
∆ E c12 = χ 1 – χ 2 5-10

and
∆ E v12 = ∆ Eg12 – ∆ E c12 5-11

for the heterojunction between Material1 and Material2 and:

∆ E c23 = χ 2 – χ 3 5-12

and
∆ E v23 = ∆ Eg23 – ∆ E c23 5-13

for the heterojunction between Material2 and Material3.

SILVACO International 5-7

ATLAS User’s Manual

Using the ALIGN parameter on the MATERIAL statement

Notice that the reference material, the material with the smallest bandgap in this case, Material1, is
not shared between the two larger bandgap materials, Material2, and Material3. This will be
important in calculating the conduction band offsets for the heterojunction formed by Material2 and
Material3 (the one where the reference material is not present).
Let’s assign 80% of the bandgap difference between Material1 and Material2 to the conduction
band offset for this heterojunction. Since the reference material is one of the materials of this
heterojunction, we can proceed as before. Define the ALIGN parameter in the MATERIAL statement for
Material2 using:
MATERIAL NAME=Material2 ALIGN=0.8
then
∆ E c12 = ( E g2 – E g1 ) ⋅ 0.80 5-14

Internally, the affinity of Material2 is adjusted so that ∆Εc12 equals this value.
Let’s assign 70% of the bandgap difference between Material3 and Material2 to the conduction
band offset for this heterojunction. Since the reference material is not one of the materials in this
heterojunction, another procedure will be used. Since BLAZE always uses the bandgap of the reference
material (the smallest bandgap material in overall structure) when calculating the conduction band
offset using the ALIGN parameter on the MATERIAL statement, the actual value for the ALIGN
parameter needs to be calculated as follows:
ALIGN = ( ∆ E g32 ⁄ ∆ E g31 ) ⋅ FRACTION 5-15

where FRACTION is the desired fraction of the bandgap difference between Material3 and
Material2 that will appear in the conduction band. ∆Εg23 is the bandgap difference for the actual
heterojunction and ∆Εg31 is the bandgap difference using the reference material. Once calculated, this
value for the ALIGN parameter can be used on the MATERIAL statement for Material3. FRACTION
and ALIGN will only be equal when the reference material is one of the two materials in the
heterojunction. For this example, lets assume that
∆ E g32 = 0.2 5-16

and
∆ E g31 = 0.4 5-17

then for a desired conduction band offset fraction of 0.70:

ALIGN = ( ∆ E g32 ⁄ ∆ E g31 ) ⋅ FRACTION = ( 0.2 ⁄ 0.4 ) × 0.70 = 0.35 5-18

The proper value of the ALIGN parameter reflecting the desired conduction band offset can be assigned
as:
MATERIAL NAME=Material3 ALIGN=0.35
This assigns 70% of the bandgap difference between Material3 and Material2 as the conduction
band offset. Internally, the affinity of Material3 is adjusted so that ∆Εc23 equals this value.

5-8 SILVACO International

 Blaze: Compound Material 2D Simulator

Note: Calculating ALIGN in this manner is only necessary when the reference material is not in contact with the material
where the ALIGN parameter will be specified.

These new values of electron affinity for Material2 (from the first heterojunction band offset
calculation) and Material3 (from the second heterojunction band offset calculation) will override any
electron affinity specification for these materials. This has an impact on any calculation where these
materials electron affinity is used and must be considered when specifying Schottky barriers contacted
to these materials. See “EXAMPLE 4” on page 5-10 for more details on Schottky barrier
considerations.

Manually Adjusting Material Affinity

Assigning the conduction band offsets for each heterojunction is accomplished by setting the electron
affinities for Material2 and Material3 using the AFFINITY parameter on the MATERIAL statement.
The electron affinity for Material2 is adjusted relative to Material1 and Material3 is adjusted
relative to Material2 by the amount of the desired conduction band offset for each heterojunction.
Since Material1 affinity is larger than that for Material2 and Material2 affinity is larger than
that for Material3, the affinities for Material2 and Material3 are reduced to provide the desired
conduction band offsets.
Let’s assume an electron affinity for Material1 of 4eV (~ that of GaAs). Let’s decide that between
Material1 and Material2, the conduction band offset is 0.3eV and that between Material2 and
Material 3, the conduction band offset is 0.2eV. Then, for Material2:
MATERIAL NAME=Material2 AFFINITY=3.7
and for Material3:
MATERIAL NAME=Material3 AFFINITY=3.5
This is the easiest method to define the conduction band offsets for multiple materials. This has an
impact on any calculation where this materials electron affinity is used and must be considered when
specifying Schottky barriers contacted to this materials. See “EXAMPLE 4” on page 5-10 for more
details on Schottky barrier considerations.

Note: The band offsets are always defined with reference to the conduction band. Therefore, if a specific valence band offset
is required, the appropriate conduction band offset should be calculated from the desired valence band offset and the materials
bandgap.

SILVACO International 5-9

ATLAS User’s Manual

EXAMPLE 4

E(x)
Eo

qVo El(x)
χ1

χ2 φm
Ec(x)
Εg1
Ei(x) φb
Ef
Ev(x)
∆Εv
Metal
Εg2

x
0

Figure 5-5: Schematic band diagram for an abrupt heterojunction

Figure 5-5 details a heterostructure device consisting of two semiconductors with different bandgaps
Εg1 and Εg2 and electron affinities χ1and χ2 and a Schottky barrier. For this example, Εg1 < Εg2 and χ2
< χ 1.
This example will first define the heterojunction band offsets and then the Schottky barrier height.
Schottky contact barrier heights are calculated by BLAZE using the metal work function and the
semiconductor electron affinity as:
φb = φm – χs 5-19

where φb is the Schottky barrier height, φm is the work function of the metal, and χs is the
semiconductor electron affinity. φm is set using the WORKFUN parameter in the CONTACT statement.
Therefore, the semiconductor electron affinity as modified or defined during the heterojunction
alignment process plays an important role in determining the value of the metal workfunction needed
to provide the desired barrier height. Let’s assume for this example that a Schottky barrier height of
0.2eV is desired and calculate the appropriate metal workfunction for each case.

Using the Affinity rule for the heterojunction

∆ E c = χ 1 – χ2 5-20

and
∆ Ev = ∆ Eg – ∆ Ec 5-21

∆Εc is the amount of the conduction band discontinuity at the heterointerface and ∆Εv is the amount of
the valence band discontinuity.

5-10 SILVACO International

 Blaze: Compound Material 2D Simulator

Note: Remember the Affinity Rule is used for a material as long as the ALIGN parameter is not specified on the
MATERIAL statement for that material.

Let’s use an electron affinity for Material1 of 4eV and for Material2 of 3.5eV. Since the affinity of
the material on which the Schottky barrier is formed was not modified with this method of alignment,
the metal work function needed to provide for a Schottky barrier height of 0.2eV is:
φ m = 3.5 + 0.2 = 3.7 5-22

This value can now be assigned to the WORKFUN parameter on the CONTACT statement as:
CONTACT NUMBER=1 WORKFUN=3.7
This produces a Schottky barrier height of 0.2eV between the metal and Material 2.
Using the ALIGN parameter on the MATERIAL statement:
Let’s assign 80% of the bandgap difference between Material1 and Material2 to the conduction
band offset, an electron affinity for Material1 of 4eV, and ∆Εg of 0.2eV. Then, define the ALIGN
parameter on the MATERIAL statement for Material2 using:
MATERIAL NAME=Material2 ALIGN=0.80
Then:
∆ E c = ( Eg2 – E g1 ) ⋅ 0.80 = 0.2 ⋅ 0.80 = 0.16 5-23

Internally, the affinity of Material2 is reduced by 0.16eV so:

χ 2 = χ 1 – ∆ E c = ( 4 – 0.16 ) = 3.84 5-24

This value of electron affinity can be used to assign the proper value of WORKFUN on the CONTACT
statement to provide for a Schottky barrier height of 0.2eV.
φ m = 3.84 + 0.2 = 4.04 5-25

This value can now be assigned to the WORKFUN parameter on the CONTACT statement as:
CONTACT NUMBER=1 WORKFUN=4.04
producing a Schottky barrier height of 0.2eV between the metal and Material2.

Manually Adjusting Material Affinity

Let’s assign a conduction band offset between Material1 and Material2 of 0.15eV, an electron
affinity for Material1 of 4eV, and the desired Schottky barrier height of 0.2eV. The affinity for
Material2 is calculated from the affinity of Material1 and the desired conduction band offset as:
χ 2 = χ 1 – 0.15 = 4 – 0.15 = 3.85 5-26

This is then used to assign the value of AFFINITY using,

MATERIAL NAME=Material2 AFFINITY=3.85

SILVACO International 5-11

ATLAS User’s Manual

This value of electron affinity can now be used to calculate the metal workfunction necessary to
produce a Schottky barrier height of 0.2eV as:
φ m = 3.85 + 0.2 = 4.05 5-27

This value can now be assigned to the WORKFUN parameter on the CONTACT statement as:
CONTACT NUMBER=1 WORKFUN=4.05
This produces a Schottky barrier height of 0.2eV between the metal and Material2.

5.1.3: The Drift Diffusion Transport Model

Drift-Diffusion with Position Dependent Band Structure
The current continuity equations for electrons and holes, and the Poisson Equation (see Chapter 3:
“Physics”, Equation 3-1) are the same as for the homogeneous case. Although the changing dielectric
constant is taken into account. The current density expressions, however, must be modified to take
into account the nonuniform band structure [37]. This procedure starts with the current density
expressions:

J n = – µn n ∇φ n 5-28

J p = – µp n ∇φ p 5-29

where φ and φ are quasi-Fermi potentials.

1
φ n = --- E FN 5-30
q

1
φ p = --- E FP 5-31
q

The conduction and valence band edge energies can be written as:
E c = q ( ψ0 – ψ ) –χ 5-32

E v = q ( ψ0 – ψ ) –χ –E g 5-33

where:
• ψ is some reference potential.
• χ is the position-dependent electron affinity.
• is the position-dependent bandgap.
• ψ can be selected in the form:

χ r kT L N χ r + E g kT L N
cr = -------------------
ψ 0 = ----- + --------- ln -------- vr
– --------- ln --------
- - 5-34
q q n ir q q n ir

where is the intrinsic carrier concentration of the arbitrarily selected reference material, and is
the index that indicates that all of the parameters are taken from reference material.

5-12 SILVACO International

 Blaze: Compound Material 2D Simulator

Fermi energies are expressed in the form:

n
E FN = E c + kTL ln ------ – kT L ln γ n 5-35
Nc

n
E FP = E v + kT L ln ------ – kT L ln γ n 5-36
Nv

The final terms in Equations 5-35 and 5-36 are due to the influence of Fermi-Dirac statistics. These
final terms are defined as follows:
F1 ⁄ 2 ( η n ) E FN – E c –1 n
γ n = --------------------------, η n = ------------------------ = F1 ⁄ 2  ------ 5-37
ηn kTL Nc
e
F1 ⁄ 2 ( η ) E v – E FP –1 p
p
γ p = --------------------------, η p = ------------------------ = F 1 ⁄ 2  ------ 5-38
ηp kTL Nv
e

where and are position-dependent and γ γ for Boltzmann statistics.

By combining Equations 5-28 to 5-38, the following expressions for current densities can be obtained.

 kTL χ kTL N 
J n = kT µ ∇ n – q µ n ∇  ψ + --------- ln γ + --- + --------- ln ------c- 5-39
L n n  q n q q n ir

 kTL χ + E g kT L N 
J p = kT L µ p ∇ n – q µ p p ∇  ψ + --------- ln γ p + ----------------- + --------- ln ------v- 5-40
 q q q n ir

5.1.4: The Thermionic Emission and Field Emission Transport Model

You can activate alternative current density expressions for electron and hole current [90,126], which
take into account thermionic emission dominated current in abrupt heterojunctions. These equation
applies only at the node points along the interface of the heterojunction and take the form:

 + –  – ∆E C 
J = q υ ( 1 + δ )  n – n exp  --------------  5-41
n n   kT L  

 + –  – ∆E V 
J p = ( – q ) υ p ( 1 + δ )  p – p exp  --------------  5-42
  kT L  

where and are the electron and hole current densities from the " -" region to the "+" region
ν andν are the electron and hole thermal velocities, ∆ is conduction band energy change going from
the “-” region to the “+” region and ∆ is the valence band energy change going from the “-” region to
the “+” region. The δ parameter represents the contribution due to thermionic field emission
(tunneling).

SILVACO International 5-13

ATLAS User’s Manual

The thermal velocities ν andν are given by:

* 2
An TL
υ n = ----------------- 5-43
qN C

* 2
Ap TL
υ p = ----------------- 5-44
qN V

where is the lattice temperature, is the conduction band density of states, is the valence
* *
band density of states and A n and A p are the electron and hole Richardson constants.

The minimum valued Richardson constants from the “-” region or “+” region are used for the
calculation of the thermal velocities [126]. You can specify the Richardson constants with the ARICHN
and ARICHP parameters of the MATERIAL statement. If the Richardson constants aren’t specified, the
following expressions will be used:
2 *
* 4 π qk m n
A n = ------------------------ 5-45
3
h

2 *
* 4 π qk m p
A p = ------------------------ 5-46
3
h

where and are the electron and hole effective masses. The electron and hole effective masses
can be specified with the [Link] and [Link] parameters of the MATERIAL statement. If the effective
masses aren’t specified, then they will be calculated from the conduction and valence band densities of
states using Equations 3-31 and 3-32 in Chapter 3: “Physics”.
The tunneling factor δ in Equation 5-41 is zero when the tunneling mechanism is neglected. When you
account for tunneling by specifying the THERMIONIC parameter in the INTERFACE statement, the
tunneling factor, δ, is calculated by using the expression:
+
EC XE
 E+ – E  
 –4π 0.5 

1 C x *
δ = ------
kT ∫ 
 kT   ∫
exp -------------------- exp  ---------
h
2m n ( E C – E x ) dx dEx

5-47
   
E min 0

where is the energy component in the direction and as described in

Figure 5-6.

5-14 SILVACO International

 Blaze: Compound Material 2D Simulator

Figure 5-6: Band Diagram of p-n heterojunction

SILVACO International 5-15

ATLAS User’s Manual

5.2: The Physical Models

Chapter 3: “Physics”, Section 3.6: “Physical Models” already describes a comprehensive set of physical
models for silicon. It is known, however, that for III-V, II-VI, and ternary compounds that special
consideration are required such as for mole fraction dependence, material properties, and so on. The
following sections describe the material dependent physical models that have been implemented into
BLAZE to account for these effects.
First, a description of the common mobility model equations is given. These equations are applicable to
all materials unless otherwise stated in the following material sections. Following this are sections
describing the physical models on a material-per-material basis. For each material there are
descriptions of the models for bandgap narrowing, electron affinity, density of states, dielectric
permittivity, and low field mobility.

5.2.1: Common Physical Models

Low Field Mobility Models
The default low field mobility models used for most materials in BLAZE are given by the following:

T L TMUN
µ n0 ( T L ) = MUN  --------- 5-48
300

T L TMUP
µ p0 ( T L ) = MUP  --------- 5-49
300

Where is the temperature in degrees Kelvin and the parameters: MUN, MUP, TMUN and TMUP are
user-definable as shown in Table 5-1.

Note: All the mobility models described in Chapter 3: “Physics”, except for the TASCH model, can be used in BLAZE. But all
default coefficients exist only for Silicon and therefore are not suitable for compound materials.

Table 5-1. User-Specifiable Parameters for Equations 5-48 and 5-49

Statement Parameter Units

MOBILITY MUN cm2/V·s

MOBILITY TMUN

MOBILITY MUP cm2/V·s

MOBILITY TMUP

5-16 SILVACO International

 Blaze: Compound Material 2D Simulator

Parallel Electric Field-Dependent Mobility Models

There are two types of electric field-dependent mobility models used in ATLAS/BLAZE. These models
are called Standard Mobility Model and Negative Differential Mobility Model. Both of these models
contain appropriate default values of parameters for different materials. You need to specify which
type of mobility will be used for each material and which material parameters you want to alter.
The Standard Mobility Model that takes account of velocity saturation is defined according to:
1 ⁄ BETAN
1
µ n ( ) = µ n0 ------------------------------------------------------ 5-50
µ n0 BETAN
1 +  -------------------
 VSATN

1 ⁄ BETAP
1
µ ( ) = µ p0 ------------------------------------------------------ 5-51
µ p0 E BETAP
1 +  -------------------
VSATP

where VSATN and VSATP are the saturation velocities for electrons and holes, BETAN and BETAP are
constants given in Table 5-2, and µ are the electron and hole low field mobilities. This model is
activated by the FLDMOB or the EVSATMOD=0 parameter in the MODEL statement.

Table 5-2. User-Specifiable Parameters for Equations 5-50 and 5-51

Statement Parameter Units

MOBILITY BETAN
MOBILITY BETAP

MOBILITY VSATN cm/s

MOBILITY VSATP cm/s

The Negative Differential Mobility Model of Barnes et. al. [55] has been implemented to account for
certain devices where the carrier drift velocity peaks at some electric field before reducing as the
electric field increases. This model takes account of this through the carrier mobility with equations of
the form:

VSATN GAMMAN
µ n0 + -------------------  -----------------------
ECRITN
µ n ( ) = ----------------------------------------------------------------------------------------- 5-52
GAMMAN
1 +  -----------------------
 ECRITN

VSATP GAMMAP
µ p0 + -------------------  -----------------------
ECRITP
µ p ( ) = ----------------------------------------------------------------------------------------- 5-53
GAMMAP
1 +  -----------------------
 ECRITP

where VSATN and VSATP are the electron and hole saturation velocities, is a constant, and µ are
the low-field electron and hole mobilities. This mobility model is activated by specifying EVSATMOD=1
in the MODEL statement.

SILVACO International 5-17

ATLAS User’s Manual

Table 5-3. User-Specifiable Parameters for Equations 5-52 and 5-53

Statement Parameter Default Units

MOBILITY ECRITN 4.0×103 V/cm

MOBILITY ECRITP 4.0×103 V/cm

MOBILITY GAMMAN 4.0

MOBILITY GAMMAP 1.0

Note: The Negative Differential Mobility Model introduces an instability in the solution process and is not recommended for
general use. Only activate this model in cases where the device operation directly depends on negative differential mobility
(e.g., a Gunn diode).

For both the standard and negative differential models an empirical temperature-dependent model for
saturation velocity in GaAs [56] is implemented according to:
6 4
VSATN = VSATP = 11.3 ×10 – 1.2 ×10 TL 5-54

where VSATN and VSATP are expressed in cm/sec and TL is the temperature in degrees Kelvin.
Alternatively, the saturation velocities can be set to constant values using the VSATN and VSATP
parameters of the MATERIAL statement.
Velocity Saturation with Energy Balance Transport Model
When the Energy Balance Transport Model is activated the mobility can be made a function of carrier
energy. In Chapter 3: “Physics”, Section 3.6.1: “Mobility Modeling” physical models for the dependence
of carrier mobility on carrier energy were introduced. The same models are applicable for use within
BLAZE with one additional model, which applies when the negative differential mobility model is
used.
The carrier temperature dependence is activated when the EVSATMOD=1 parameter is on the MODEL
statement. This model can be derived in a similar fashion as in the case of EVSATMOD=0 described in
Chapter 3: “Physics”, “Parallel Electric Field-Dependent Mobility” on page 3-59. These expressions,
however, require several piecewise approximations, which are too in-depth to show in this manual. To
say that these piecewise expressions provide a continuous velocity saturation model for mobility
versus carrier temperature are completely like the expressions for drift diffusion given in Equations 3-
240 and 3-241 from Chapter 3: “Physics”.

5.2.2: Recombination and Generation Models

The recombination and generation models for compound semiconductors are the same as the models
previously described in Chapter 3: “Physics”, Section 3.6.3: “Carrier Generation-Recombination
Models”. The default parameters for different materials are automatically used, unless new values are
specified. The default parameter values are listed in Appendix B: “Material Systems”.

5-18 SILVACO International

 Blaze: Compound Material 2D Simulator

5.3: Material Dependent Physical Models

5.3.1: Gallium Arsenide (GaAs) Physical Models
Bandgap Narrowing
Following Klausmeier-Brown [88], the bandgap narrowing effects are important only for p-type
regions. By default, BLAZE uses the bandgap narrowing values shown in Table 5-4.

Table 5-4. Default Bandgap Narrowing Values

Concentration Bandgap Narrowing

cm-3 meV

1.0×1018 31.0

2.0×1018 36.0

4.0×1018 44.2

6.0×1018 48.5

8.0×1018 51.7

1.0×1019 54.3

2.0×1019 61.1

4.0×1019 64.4

6.0×1019 61.9

8.0×1019 56.9

1.0×1020 53.2

2.0×1020 18.0

Note: This table is only used for GaAs. No data is available at present for other materials. The C-INTERPRETER function
for bandgap narrowing, however, allows a user-defined model for bandgap narrowing to be applied for these materials. See
Appendix A: “C-Interpreter Functions” for more information about on these functions.

SILVACO International 5-19

ATLAS User’s Manual

Low Field Mobility

The mobility in GaAs can be made concentration dependent by setting the CONMOB parameter of the
MODEL statement.
In this model, mobility is interpolated from the values in Table 5-5.

Table 5-5. Default Concentration Dependent Mobility for GaAs

Mobility in GaAs (cm2/v-s)

Concentration(cm-3)
Electrons Holes

1.0×1014 8000.0 390.0

2.0×1014 7718.0 380.0

4.0×1014 7445.0 375.0

6.0×1014 7290.0 360.0

8.0×1014 7182.0 350.0

1.0×1015 7300.0 340.0

2.0×1015 6847.0 335.0

4.0×1015 6422.0 320.0

6.0×1015 6185.0 315.0

8.0×1015 6023.0 305.0

1.0×1016 5900.0 302.0

2.0×1016 5474.0 300.0

4.0×1016 5079.0 285.0

6.0×1016 4861.0 270.0

8.0×1016 4712.0 245.0

1.0×1017 4600.0 240.0

2.0×1017 3874.0 210.0

4.0×1017 3263.0 205.0

6.0×1017 2950.0 200.0

8.0×1017 2747.0 186.9

1.0×1018 2600.0 170.0

2.0×1018 2060.0 130.0

5-20 SILVACO International

 Blaze: Compound Material 2D Simulator

Table 5-5. Default Concentration Dependent Mobility for GaAs

Mobility in GaAs (cm2/v-s)

Concentration(cm-3)
Electrons Holes

4.0×1018 1632.0 90.0

6.0×1018 1424.0 74.5

8.0×1018 1293.0 66.6

1.0×1020 1200.0 61.0

If MODEL ANALYTIC is specified, the program will use [88]:

–
µ n, p = ----------------------------------------------------- 5-55
 t  0.75
 ---------------------------
 16
×
where is the total impurity concentration.

5.3.2: Al(x)Ga(1-x)As System

The Al(x)Ga(1-x)As material system is commonly used for the fabrication of heterojunction devices.
These materials are available in BLAZE by specifying the material name GaAs, AlGaAs, or AlAs. As a
ternary material system, different material properties are obtained by adjusting the molar fraction of
Aluminum and Gallium. This mole fraction is represented by the as written in Al(x)Ga(1-x)As. GaAs
material parameters are identical to the those of AlGaAs with mole the fraction set equal to zero.
AlAs material parameters are identical to the those of AlGaAs with mole the fraction set equal to
one. Fundamental in the proper simulation with the AlGaAs material system is the relationship
between this mole fraction , and the material parameters for that composition. In the following
sections, the relationship between mole fraction and material parameters for the AlGaAs material
system will be described. You can specify the [Link] of a material in the REGION statement
with the [Link] parameter.

Note: Don’t use this material system to form GaAs by setting x=0, specify GaAs as the material instead.

Bandgap
There are three primary conduction bands in the AlGaAs system that depending on mole fraction,
determine the bandgap. These are named Gamma, L, and X. The default bandgaps for each of these
conduction band valleys are as follows:
E g Γ = EG300 + [Link] ⋅ ( 1.155 + 0.37 ⋅ [Link] ) 5-56

E gL = 1.734 + [Link] ⋅ ( 0.574 + 0.055 ⋅ [Link] ) 5-57

E gX = 1.911 + [Link] ⋅ ( 0.005 + 0.245 ⋅ [Link] ) 5-58

SILVACO International 5-21

ATLAS User’s Manual

The bandgap used for any given Al concentration is the minimum as calculated from these equations.
EG300 is the bandgap at 300K and specified on the material statement. [Link] is the
Aluminum mole fraction and may be user-defined on the REGION statement.
The temperature dependence of the bandgap is calculated according to:

2 2
L
g
( ) =
g
( )+ ⋅ -------------------------------------------- – ------------------------------------------ 5-59

The value of is taken as the minimum of Γ, , and . The default temperature

dependent bandgap parameters for AlGaAs are listed in Table 5-6.

Table 5-6. Default Bandgap Parameters for Al(x)Ga(1-x)As

Statement Parameter Default Units

MATERIAL EG300 1.59 eV

MATERIAL EGALPHA 5.405×10-4 eV/K

MATERIAL EGBETA 204 K

Electron Affinity
As indicated in the introduction, the semiconductor electron affinity χ is a key parameter for
determining the alignment of heterojunctions. For AlGaAs, χ is a function of EgΓ and is given by:

χ AlGaAs = 4.07 – 0.85 ⋅ ( E g Γ ( [Link] ) – E gGaAs ) 5-60

Density of States and Effective Mass

The valence and conduction band densities of states, NC and NV, are calculated from the effective
masses according to the following equations:
3
* ---
 2 π m kT  2
e L
N c = 2  ------------------------ 5-61
 2 
 h 

3
* ---
 2 π m kT  2
h L
N v = 2  ------------------------ 5-62
2
 h 
For the AlGaAs system the conduction band and valence band effective masses, for electrons and holes,
are given by [135]:
0.067 + 0.083x ( 0 < x < 0.45 )
me = 5-63
0.85 – 0.14x ( x > 0.45 )

5-64
l

5-22 SILVACO International

 Blaze: Compound Material 2D Simulator

5-65

5-66

Dielectric Permittivity

SILVACO International 5-23

ATLAS User’s Manual

Note: Don’t use this material system to form GaAs by setting x=1 and y=1, specify GaAs as the material instead.

Bandgap
The default energy bandgap for the InP lattice matched In(1-x)Ga(x)As(y)P(1-y) system used in BLAZE
is given by:

E g ( InGaAsP ) = 1.35 + [Link] ⋅ ( 0.642 + ( 0.758 ⋅ [Link] ) )+ ( 0.101 ⋅ [Link] – 5-69

1.101 ) ⋅ [Link]– ( 0.28 ⋅ [Link] – 0.109 ⋅ [Link] + 0.159 ) ⋅ [Link] ⋅ [Link]

Electron Affinity
The electron affinities for materials in the InP lattice matched InGaAsP system are derived from
conduction band offsets and from the assumption that the affinity of InP is 4.4eV. The default
conduction band edge offset between lattice matched InGaAsP and InP is then:

∆ E c = 0.268 ⋅ [Link] + 0.003 ⋅ ( [Link] ) 2 5-70

Density of States and Effective Mass

The density of states is defined, as before, as a function of the effective masses of electrons and holes
according to Chapter 3: “Physics”, Equation 3-31. For the InGaAsP system, the default conduction and
valence band effective masses, for electrons and holes, are given by the following.
For the conduction band:
*
m e = 0.08 – ( 0.116 ⋅ [Link] ) + ( 0.026 ⋅ [Link] )– 0.059 ⋅ ( [Link] ⋅ [Link] )+ 5-71

( 0.064 – 0.02 ⋅ [Link] ) ⋅ ( [Link] ) 2 + ( 0.06 + 0.032 ⋅ [Link] ) ⋅ ( [Link] ) 2

For the valence band the hole effective mass is defined by:
2
---
*  1.5 1.5 3
m h = m lh + mhh 5-72
 

where the default light hole effective mass is given by:

m lh = 0.120 – ( 0.116 ⋅ [Link] ) + 0.03 ⋅ ( [Link] ) 2 5-73

and the default heavy hole effective mass is a constant and is given by:
m hh = 0.46 5-74

Dielectric Permittivity
The default static dielectric constant for lattice matched InGaAsP to InP is given by
ε InGaAsP = ( 14.6 ⋅ ( 1 – [Link] ) ⋅ [Link] )+ 12.5 ⋅ ( 1 – [Link] ) ⋅ 5-75
( 1 – [Link] )+ 13.18 ⋅ [Link] ⋅ [Link]+ 11.11 ⋅ [Link] ⋅ ( 1 – [Link] )

5-24 SILVACO International

 Blaze: Compound Material 2D Simulator

Low Field Mobility

The default low field mobility parameters for electrons and holes for lattice matched InGaAs are given
by linear interpolations from the binary compounds GaAs and InP. The following formulas are used:
µ n1 = 33000 + ( 8500 – 33000 ) ⋅ [Link] 5-76

µ p1 = 460 + ( 400 – 460 ) ⋅ [Link] 5-77

µ n2 = 4600 + ( 300 – 4600 ) ⋅ [Link] 5-78

µ p2 = 150 + ( 100 – 150 ) ⋅ [Link] 5-79

µ n0 = µ n1 + ( 1 – Y .COMP ) ( µn2 – µ n1 ) 5-80

µ p0 = µ p1 + ( 1 – Y .COMP ) ( µp2 – µ p1 ) 5-81

5.3.4: The Si(1-x)Ge(x) System

Advances in the growth of Silicon and Si(1-x)Ge(x) alloys have allowed the potential for using bandgap
engineering to construct heterojunction devices such as HBTs and HEMTs using these materials.
BLAZE supports the SiGe material system by providing composition dependent material parameters.
These parameters are accessed by specifying the material name SiGe.
The following sections describe the functional relationship between Ge mole fraction x, and the SiGe
material characteristics necessary for device simulation.
Bandgap
Bandgap is one of the most fundamental parameters for any material. For SiGe, the dependence of the
bandgap on the Ge mole fraction, , is divided into ranges as follows:
E g = 1.08 + [Link] ⋅ ( 0.945 – 1.08 ) ⁄ 0.245 ) ; 5-82

for ≤
E g = 0.945 + ( [Link] – 0.245 ) ⋅ ( 0.87 – 0.945 ) ⁄ ( 0.35 – 0.245 ) ; 5-83

for ≤
( [Link] – )⋅( – )⁄ ( – ); 5-84

for ≤
E g = 0.78 + ( [Link] – 0.5 ) ⋅ ( 0.72 – 0.78 ) ⁄ ( 0.6 – 0.5 ) ; 5-85

for ≤
E g = 0.72 + ( [Link] – 0.6 ) ⋅ ( 0.69 – 0.72 ) ⁄ ( 0.675 – 0.6 ) ; 5-86

SILVACO International 5-25

ATLAS User’s Manual

for ≤
E g = 0.69 + ( [Link] – 0.675 ) ⋅ ( 0.67 – 0.69 ) ⁄ ( 0.735 – 0.675 ) ; 5-87

for ≤
E = 0.67 ; 5-88
g

for ≤
The temperature dependence of the bandgap of SiGe is calculated the same as for Silicon using except
that EGALPHA and EGBETA are a function of Ge mole fraction x as follows:
2
2
300
( ) = E g + EGALPHA -------------------------------------- – -------------------------------------- 5-89
300 + EGBETA + EGBETA

4
EGALPHA = ( 4.73 + [Link] ⋅ ( 4.77 – 4.73 ) ) ×10 5-90

EGBETA = 626 + [Link] ⋅ ( 235 – 636 ) 5-91

where is dependent upon the mole fraction as above.

Electron Affinity
The electron affinity χ of SiGe is taken to be constant with respect to composition.
Density of States
The density of states for SiGe is defined differently compared to the previous materials by not being a
function of the effective masses. Instead the density of states have been made to depend upon the Ge
mole fraction, ,according to:
19 19 19
N c = 2.8 ×10 + [Link] ⋅ ( 1.04 ×10 – 2.8 ×10 ) 5-92

19 18 19
N v = 1.04 ×10 + [Link] ⋅ ( 6.0 ×10 – 1.04 ×10 ) 5-93

Dielectric Function
The compositional dependence of the static dielectric constant of SiGe is given by
ε = 11.8 + 4.2 ⋅ [Link] 5-94

Low Field Mobility

No specific SiGe low field mobility models have been implemented into BLAZE.
Velocity Saturation
In SiGe, the temperature dependent velocity saturation, used in the field dependent mobility model is
defined by the following equations.

5-26 SILVACO International

 Blaze: Compound Material 2D Simulator

7 175
VSATN = 1.38∗ 10 ⋅  tan h  ---------  5-95
  T 
L

6 312
VSATP = 9.05∗ 10 ⋅  tan h  ---------  5-96
  T 
L

Note: All other defaults used for SiGe are taken from Silicon

5.3.5: Silicon Carbide (SiC)

Silicon carbide materials are of interest for high power, high temperature applications. The main
characteristics of silicon carbide are that they have a very wide bandgap, high thermal conductivity,
high saturation velocity, and high breakdown strength. Silicon carbide is commercially available in
three polytypes called 6H-SiC, 3H-SiC, and 4H-SiC. ATLAS supports all three these polytypes. The
following paragraphs describe the material defaults for these materials.
Band Parameters for SiC
SiC band parameter equations are identical to those used for Silicon but with the values adjusted for
3C-SiC, 4H-SiC, and 6H-SiC. The physical band parameter values are shown in Tables B-18 and B-19
of Appendix B: “Material Systems”.
SiC Mobility Parameters
Isotropic Mobility
By default mobility is assumed to be entirely isotropic in nature. that is there is no directional
component. The default low field mobilities of electrons and holes for 3C-SiC, 4H-SiC, and 6H-SiC are
shown in Table 5-8.

Table 5-8. Silicon Carbide Low Field Mobility Defaults

Statement Parameter 6H-SiC 3C-SiC 4H-SiC Units

MOBILITY MUN 330 1000 460 cm2/V·s

MOBILITY MUP 60 50 124 cm2/V·s

Anisotropic Mobility
The mobility behavior within SiC is now known to be anisotropic in nature, which dramatically alters
the electrical performance of a device. An isotropic model has been implemented into ATLAS to
correctly model this behavior. Following the ideas of Lindefelt [7] and Lades [134], the mobility within
the drift diffusion equations has been made a tensor property. As a result the mobility has become:

µ1 0 0
µ = 0 µ1 0 5-97
0 0 µ2

SILVACO International 5-27

ATLAS User’s Manual

where µ1 represents the mobility defined in one plane and µ2 the mobility defined in a second plane. In
the case of SiC µ1 represents the mobility of plane <1100> whilst µ2 represents the mobility of plane
<1000>. These mobilities are defined for both holes and electrons.
Defining Anisotropic Mobility in ATLAS
To define a material with anisotropic mobility, specify two MOBILITY statements. In each statement,
the [Link] and [Link] parameters are used to specify the direction where that particular
mobility is to apply. The following example shows how this is done.

# FIRST DEFINE MOBILITY IN PLANE <1100>

#
MOBILITY MATERIAL=3C-SiC VSATN=2E7 VSATP=2E7 BETAN=2 BETAP=2 \
[Link]=10 [Link]=410 [Link]=13E17 \
[Link]=0.6 [Link]=0.0 \
[Link]=-3 [Link]=-3 \
[Link]=20 [Link]=95 [Link]=1E19 \
[Link]=0.5 [Link]=0.0 \
[Link]=-3 [Link]=-3
#
# NOW DEFINE MOBILITY IN PLANE <1000>
#
MOBILITY MATERIAL=3C-SiC [Link]=90.0 VSATN=2E7 VSATP=2E7 BETAN=2 BETAP=2 \
[Link]=5 [Link]=80 [Link]=13E17 \
[Link]=0.6 [Link]=0.0 \
[Link]=-3 [Link]=-3 \
[Link]=2.5 [Link]=20 [Link]=1E19 \
[Link]=0.5 [Link]=0.0 \
[Link]=-3 [Link]=-3
Impact Ionization and Thermal Parameters
The equations governing these effects are identical to those for Silicon but with adjusted coefficients.
See Appendix B: “Material Systems” for a list of all these parameters.

5.3.6: GaN, InN, AlN, AlGaN, and InGaN System

The following sections describe the relationship between mole fraction, [Link], and the material
parameters of the Al/ln/GaN system.
Bandgap
By default, the bandgap for the nitrides is calculated in a two step process. First, the bandgap(s) of the
relevant binary compounds are computed as a function of time, , using Equations 5-98 through 5-100
[150].
–3 2
0.909 × 10 T
Eg ( GaN ) = 3.507 – --------------------------------------- 5-98
T + 830.0

–3 2
0.245 × 10 T
Eg ( InN ) = 1.994 – --------------------------------------- 5-99
T + 624.0

5-28 SILVACO International

 Blaze: Compound Material 2D Simulator

–3 2
1.799 × 10 T
Eg ( AlN ) = 6.23 – --------------------------------------- 5-100
T + 1462.0

Then, the dependence on composition fraction, , is described by Equations 5-101 and 5-102 [153].

Eg ( In x Ga1 – x N ) = Eg ( InN )x + Eg ( GaN ) ( 1 – x ) – 3.8x ( 1 – x ) 5-101

Eg ( Al x Ga 1 – x N ) = Eg ( AlN )x + Eg ( GaN ) ( 1 – x ) – 1.3x ( 1 – x ) 5-102

Electron Affinity
The electron affinity is calculated such that the band edge offset ratio is given by [153].
∆Ec 0.7
---------- = ------- 5-103
∆Ev 0.3

You can overide this ratio by specifying the ALIGN parameter of the MATERIAL statement.
Permittivity
The permittivity of the nitrides as a function of composition fraction, , is given by linear
interpolations of the values for the binary compounds as in Equations 5-104 and 5-105 [155].

E ( In x Ga 1 – x N ) = 19.6x + 10.4 ( 1 – x ) 5-104

E ( Al x Ga1 – x N ) = 10.1x + 10.4 ( 1 – x ) 5-105

Density of States Masses

The nitride density of states masses as a function of composition fraction, , is given by linear
interpolations of the values for the binary compounds as in Equations 5-106 through 5-109 [150].
m e ( In x Ga 1 – x N ) = 0.12x + 0.2 ( 1 – x ) 5-106

m h ( In x Ga1 – x N ) = 0.17x + 1.0 ( 1 – x ) 5-107

m e ( Al x Ga1 – x N ) = 0.314x + 0.2 ( 1 – x ) 5-108

m ( Al Ga N ) = 0.417x + 1.0 ( 1 – x ) 5-109

h x 1–x

Nitride Low Field Mobility

You can use a composition and temperature dependent low field model by specifying the FMCT.N and
FMCT.P in the MOBILITY statement. FMCT stands for Farahmand Modified Caughey Thomas. This
model [164] was the result of fitting a Caughey Thomas like model to Monte Carlo data. The model is
similar to the analytic model described by Equations 3-147 and 3-148 in Chapter 3: “Physics”. This
modified model is described by Equations 5-110 and 5-111 for electrons and holes.

SILVACO International 5-29

ATLAS User’s Manual

T BETAN · FMCT
µ 0 ( T, N ) = MU1N · FMCT  --------- + 5-110
300

T  DELTAN · FMCT
( MU2N · FMCT – MU1N · FMCT )  -------- -
300
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
EPSP · FMCT
ALPHAN · FMCT ( T ⁄ 300 )
N
1 + ----------------------------------------------------------------------------------------------------
T GAMMAN · FMCT
NCRITN · FMCT  ---------
 300

T BETAP · FMCT
µ 0 ( T, N ) = MU1P · FMCT  --------- + 5-111
300

T  DELTAP · FMCT
( MU2P · FMCT – MU1P · FMCT )  -------- -
300
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
EPSP · FMCT
ALPHAP · FMCT ( T ⁄ 300 )
N
1 + ----------------------------------------------------------------------------------------------------
T GAMMAP · FMCT
NCRITP · FMCT  ---------
 300

In these equations, is the lattice temperature and is the total doping. Table 5-9 shows the user-
definable parameters.

Table 5-9. User-specifiable parameters for the Faramand modified Caughey Thomas model for Nitrides.

Parameter Statement Type Units

[Link] MOBILITY Real cm2/(V*s)

[Link] MOBILITY Real cm2/(V*s)

[Link] MOBILITY Real cm2/(V*s)

[Link] MOBILITY Real cm2/(V*s)

[Link] MOBILITY Real

[Link] MOBILITY Real

[Link] MOBILITY Real

5-30 SILVACO International

 Blaze: Compound Material 2D Simulator

Table 5-9. User-specifiable parameters for the Faramand modified Caughey Thomas model for Nitrides.

Parameter Statement Type Units

[Link] MOBILITY Real

[Link] MOBILITY Real

[Link] MOBILITY Real

[Link] MOBILITY Real

[Link] MOBILITY Real

[Link] MOBILITY Real

[Link] MOBILITY Real

[Link] MOBILITY Real cm-3

[Link] MOBILITY Real cm-3

The default parameters as taken from the Monte Carlo fits for various nitride compositions are given
in Tables 5-10a and 5-10b.

Table 5-10a. Default Nitride Low Field Mobility Model Parameter Values [164]
MATERIAL [Link] [Link] [Link] [Link]
(cm2/V.s) (cm2/V.s)

InN 774 3138.4 0.68 -6.39

In0.8Ga0.2N 644.3 1252.7 0.82 -1.30

In0.5Ga0.5N 456.4 758.1 1.04 -1.74

In0.2Ga0.8N 386.4 684.2 1.37 -1.95

GaN 295.0 1460.7 0.66 -3.84

Al0.2Ga0.8N 132.0 306.1 0.29 -1.75

Al0.5Ga0.5N 41.7 208.3 0.12 -2.08

Al0.8Ga0.2N 47.8 199.6 0.17 -2.04

AlN 297.8 683.8 1.16 -1.82

SILVACO International 5-31

ATLAS User’s Manual

Table 5-10b. Default Nitride Low Field Mobility Model Parameter Values [164]
MATERIAL [Link] [Link] [Link]
InN -1.81 8.05 -0.94

In0.8Ga0.2N -1.30 4.84 -0.41

In0.5Ga0.5N -1.74 2.21 -0.22

In0.2Ga0.8N -1.95 2.12 -0.99

GaN -3.84 3.02 0.81

Al0.2Ga0.8N -1.75 6.02 1.44

Al0.5Ga0.5N -2.08 10.45 2.00

Al0.8Ga0.2N -2.04 20.65 0.01

AlN -3.43 3.78 0.86

For composition fractions not listed in Tables 5-10a and 5-10b, the default parameters are linearly
interpolated from the nearest composition fractions on the table. You can override these defaults by
specifying any of the parameters listed in Table 5-9 on the MOBILITY statement. Currently, this model
has only been calibrated for electrons. We do not recommend you to use this model for holes without
careful consideration of a set of sensible user-defined default parameters.
Nitride High Field Mobility
You can select nitride specific field dependent mobility model by specifying GANSAT.N and GANSAT.P
on the MOBILITY statement. This model [164] is based on a fit to Monte Carlo data for bulk nitride,
that is described in Equations 5-112 and 5-113.

N1N · GANSAT – 1
E
µ 0 ( T, N ) + VSATN ----------------------------------------------------------------
N1N · GANSAT
ECN GANSAT
·
µ = -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- 5-112
E N2N · GANSAT N1N · GANSAT
1+ ANN · GANSAT  --------------------- +  ----------E-----------
ECN · GANSAT ECN · GANSAT

N1P · GANSAT – 1
E
µ 0 ( T, N ) + VSATP ----------------------------------------------------------------
N1P · GANSAT
ECP GANSAT
·
µ = -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- 5-113
E N 2P · GANSAT N1 P · GANSAT
1 + ANP · GANSAT  ---------------------
 ECP · GANSAT
+  ----------E-----------
ECP · GANSAT

In these equations µ0(T,N) is the low field mobility and E is the electric field. The user-definable
parameters are listed in Table 5-11.

5-32 SILVACO International

 Blaze: Compound Material 2D Simulator

Table 5-11. User Definable Low Field Nitride Mobility Model Parameters
Parameter Statement Type Units

[Link] MOBILITY Real [Link]

[Link] MOBILITY Real [Link]

[Link] MOBILITY Real [Link]

[Link] MOBILITY Real

[Link] MOBILITY Real

[Link] MOBILITY Real V/cm

[Link] MOBILITY Real V/cm

The default parameters as taken from the Monte Carlo fits for various nitride compositions are given
in Table 5-12.

Table 5-12. Default Nitride Field Dependent Mobility Model Parameter Values [164]
MATERIAL VSATN [Link] [Link] [Link] [Link]
(107cm/s) (kV/cm)

InN 1.3595 52.4242 3.8501 0.6078 2.2623

In0.8Ga0.2N 0.8714 103.4550 4.2379 1.1227 3.0295

In0.5Ga0.5N 0.7973 148.9098 4.0635 1.0849 3.0052

In0.2Ga0.8N 1.0428 207.5922 4.7193 1.0239 3.6204

GaN 1.9094 220.8936 7.2044 0.7857 6.1973

Al0.2Ga0.8N 1.1219 365.5529 5.3193 1.0396 3.2332

Al0.5Ga0.5N 1.1459 455.4437 5.0264 1.0016 2.6055

Al0.8Ga0.2N 1.5804 428.1290 7.8166 1.0196 2.4359

AlN 2.1670 447.0339 17.3681 0.8554 8.7253

For composition fractions not listed in Table 5-12, the default parameters are linearly interpolated
from the nearest composition fractions on the table.
You can override these defaults by specifying any of the parameters listed in Table 5-11 in the
MOBILITY statement. Currently, this model has only been calibrated for electrons. We do not
recommend you use this model for holes unless you define a set of default parameters that somewhat
correspond to physical reality. You should also note that these models exhibit negative differential
mobility and therefore may exhibit poor convergence. In such cases, you may do well to use the simpler
model given in Equations 3-225 and 3-226 with a reasonable value of saturation velocity.

SILVACO International 5-33

ATLAS User’s Manual

5.3.7: The Hg(1-x)Cd(x)Te System

The following sections describe the relationship between mole fraction, [Link], and the material
parameters of the of the Hg(1-x)Cd(x)Te system. This data has been taken from [141].
Bandgap
Equation 5-114 is used in the calculation of bandgap as a function of composition fraction.
2
Eg = – 0.302 + 1.93∗ X · COMP – 0.810∗ X · COMP 5-114

3 –4
+ 0.832∗ X · COMP + 5.354 × 10 ∗  1 – 2∗ X · COMP T L
 
Here, is lattice temperature.

Electron Affinity
Equation 5-115 is used in the calculation of electron affinity as a function of composition fraction.

χ = 4.23 – 0.813∗ ( EG300 – 0.0083 ) 5-115

Permittivity
Equation 5-116 is used in the calculation of relative permittivity as a function of composition fraction.
2
ε = 20.5 – 15.5∗ X · COMP + 5.7∗ X · COMP 5-116

Density of States Masses

Equations 5-117 and 5-118 are used to calculate the density of states effective masses of electrons and
holes.

me 2 1 –1
------- = – 0.6 + 6.333∗  ----------------- + --------------------------- 5-117
m  EG300 EG300 + 1
0

me
------- = 0.55 5-118
m
0

Mobility
Equations 5-119 and 5-120 are used to calculate the electron and hole mobilities as functions of
composition fraction.
7.5 0.6
4 0.2 
– 0.2∗  0.2 ⁄ X · COMP
µ e = 9 × 10  ---------------------- T 5-119
 X · COMP L  

µ h = 0.01∗ µ e 5-120

5-34 SILVACO International

 Blaze: Compound Material 2D Simulator

5.4: Simulating Heterojunction Devices with Blaze

5.4.1: Defining Material Regions with Positionally-Dependent Band Structure
Step Junctions
The easiest way to define a device with positionally dependent band structure is to specify two
adjacent semiconductor regions with dissimilar bandgap. In this case, BLAZE would simulate an
abrupt heterojunction between the two materials. As an example, suppose you wanted to simulate an
abrupt heterojunction parallel to the x-axis at a location of y=0.1 microns. For values of y greater than
0.1 specify, for example, GaAs. For values of y less than 0.1, specify AlGaAs with a composition fraction
of 0.3. The following statements would specify this situation.
REGION [Link]=0.1 MATERIAL=GaAs
REGION [Link]=0.1 MATERIAL=AlGaAs [Link]=0.3
This fragment specifies that the two regions form an abrupt heterojunction at Y=0.1. The first region
is composed of GaAs while the second is composed of AlGaAs.
These two material names are used by BLAZE to choose default material models for the two regions.
For a complete list of the materials available in ATLAS/BLAZE, see Appendix B: “Material Systems”.
For the AlGaAs region a composition fraction of 0.3 is specified.
Graded Junctions
A grading can be applied to this heterojunction with a simple modification. For example:
REGION [Link]=0.1 MATERIAL=GaAs
REGION [Link]=0.1 MATERIAL=AlGaAs [Link]=0.3 GRAD.34=0.01
specifies that the composition fraction of the AlGaAs region decreases from 0.3 at y=0.1 microns to 0.0
at y=0.11 microns. The GRAD. parameter specifies the distance over which the mole fraction reduces to
zero. The GRAD. parameter is indexed such that GRAD.12 corresponds to the [Link] side of the region,
GRAD.23 corresponds to the [Link] side of the region, GRAD.34 corresponds to the [Link] side of the
region, and GRAD.41 corresponds to the [Link] side of the region. In most cases the GRAD.n parameter
acts to increase the size of the region. By default the GRAD.n parameters are set to zero and all
heterojunctions are abrupt. It should also be noted that the GRAD. parameter acts just like the other
region geometry parameters in that later defined regions overlapping the graded part of the region
will overlap the grading. If in the previous example, the grading had been applied to the GaAs region it
would be over lapped by the AlGaAs region. This would have produced an abrupt interface. A solution
would be to limit [Link] in the AlGaAs region to 0.09. Care should always be taken to specify regions
in the proper order to avoid such problems.
Along similar lines, the overlapping of regions can be used to an advantage in forming graded
heterojunctions between two materials in the same system with different non-zero composition
fractions. For example:
REGION [Link]=0.1 MATERIAL=AlGaAs [Link]=0.3 GRAD.12=0.02
REGION [Link]=0.11 MATERIAL=AlGaAs [Link]=0.1
specifies a graded heterojunction with a composition of 0.3 at y = 0.1 falling to 0.1 at y = 0.11.

SILVACO International 5-35

ATLAS User’s Manual

5.4.2: Defining Materials and Models

Materials
For example to set the bandgap for the material, InP, the following syntax should be used:
MATERIAL MATERIAL=InP EG300=1.35
Models
BLAZE has two ways of simulating the physical effects of variations in semiconductor composition. For
relatively gradual variations in composition the standard modifications to the drift-diffusion equations
can be considered adequate for simulation purposes. For abrupt heterojunctions, it has been suggested
that thermionic emission may be the dominant factor in the behavior of heterojunction behavior.
Individual material parameters and models can be defined for each material or region. These models
are set in the MATERIAL, MODEL, and IMPACT statements.
This statement uses the MATERIAL parameter to select all regions composed of the material "InP". The
bandgap in these regions will be set to 1.35. The parameters of a particular region can be set in two
ways. The first is by way of the region index, as in:
MODEL REGION=1 BGN
In this case, the band gap narrowing model is enabled in the region indexed number 1.
The region name may also be used as in the following:
IMPACT NAME=substrate SELB
This example turns on the Selberherr Impact Ionization Model in the region named “substrate”. You
can then set the parameters for all regions and materials by omitting the MATERIAL, REGION, or NAME
parameters, as in the following:
MODEL BGN
This statement sets the bandgap narrowing model for all regions and materials
Parser Functions
To use the C-INTERPRETER functions, you need to know the C programming language. See Appendix A:
“C-Interpreter Functions” for a description of the parser functions.
To specify a completely arbitrary spatial variation of varying composition fraction, use a parser
function. To define the parser function for composition fraction, write a C function describing the
composition fraction as a function of position. A template for the function called COMPOSITION is
provided with this release of ATLAS. Once you define the COMPOSITION function, store it in a file. In
order to use the function for composition, set the [Link] parameter to the file name of the
function.

5-36 SILVACO International

 Chapter 6:
3D Device Simulator

6.1: 3D Device Simulation Programs

This chapter aims to highlight the extra information required for 3-D simulation as compared to 2-D.
You should be familiar with the equivalent 2-D models before reading this chapter.
This chapter describes the set of ATLAS products that extends 2D simulation models and techniques
and applies them to general non-planar 3D structures. The structural definition, models and material
parameters settings and solution techniques are similar to 2D. You should be familiar with the
simulation techniques described in Chapter 2: “Getting Started with ATLAS” and the equivalent 2D
product chapters before reading the sections that follow. The products that form 3D device simulation
in ATLAS are:
• DEVICE3D – silicon simulation equivalent to S-PISCES
• BLAZE3D – compound material and heterojunction simulation
• GIGA3D – non-isothermal simulation
• MIXEDMODE3D – mixed device-circuit simulation
• TFT3D – thin film transistor simulation
• QUANTUM3D – quantum effects simulation
• LUMINOUS3D – photodetection simulation
The 3D modules, THERMAL3D, are described in Chapter 16: “Thermal 3D: Thermal Packaging
Simulator”.
In a similar manner to the 2D products, GIGA3D, MIXEDMODE3D, LUMINOUS3D and QUANTUM3D
should be combined with both DEVICE3D or BLAZE3D depending on the semiconductor materials used.

6.1.1: DEVICE3D
Device3D provides semiconductor device simulation of silicon technologies. Its use is analogous to the
2-D simulations in S-PISCES. See Chapter 4: “S-Pisces: Silicon Based 2D Simulator” for more
information on S-PISCES.

6.1.2: BLAZE3D
BLAZE3D allows you to simulate semiconductor devices with semiconductor compositional variations
(i.e., heterojunction devices). BLAZE3D is completely analogous to BLAZE (see Chapter 5: “Blaze:
Compound Material 2D Simulator”), with some exceptions. For one thing, the current version of
BLAZE3D doesn’t allow or account for compositional variations in the Z direction. It also doesn’t
account for thermionic interfaces at abrupt heterojunctions.

6.1.3: GIGA3D
GIGA3D is an extension of DEVICE3D or BLAZE3D that accounts for lattice heat flow in 3D devices.
GIGA3D has all the functionality of GIGA (see Chapter 7: “Giga: Self-Heating Simulator”) with a few
exceptions. One exception is that additional syntax has been added to account for the three
dimensional nature of thermal contacts. The [Link] and [Link] parameters can be specified on the
THERMCONTACT statement to describe the extent of the contact in the Z direction. Another exception is
that there is no BLOCK method available in the 3D version.

SILVACO International 6-1

ATLAS User’s Manual

6.1.4: TFT3D
TFT3D is an extension of DEVICE3D that allows you to simulate amorphous and polycrystalline
semiconductor materials in three dimensions. TFT3D is completely analogous to the TFT simulator
described in Chapter 14: “TFT: Thin-Film Transistor Simulator”. The complete functionality of the
TFT simulator is available in TFT3D for three dimensional devices.

6.1.5: MIXEDMODE3D
MIXEDMODE3D is an extension of DEVICE3D or BLAZE3D that allows you to simulate physical devices
embedded in lumped element circuits (SPICE circuits). MIXEDMODE3D is completely analogous to the
MIXEDMODE simulator, which is described in Chapter 12: “MixedMode: Mixed Circuit and Device
Simulator”. The complete functionality of the MIXEDMODE simulator is available in MIXEDMODE3D for
three dimensional devices.

6.1.6: QUANTUM3D
QUANTUM3D is an extension of DEVICE3D or BLAZE3D, which allows you to simulate of the effects of
quantum confinement using the Quantum Transport Model (Quantum Moments Model). QUANTUM3D
is completely analogous to the QUANTUM model, but applies to three dimensional devices. See Chapter
13: “Quantum: Quantum Effect Simulator” for more information on QUANTUM.

6.1.7: LUMINOUS3D
LUMINOUS3D is an extension of DEVICE3D that allows you to simulate photodetection in three
dimensions. LUMINOUS3D is analogous to the LUMINOUS simulator, which is described in Chapter 10:
“Luminous: Optoelectronic Simulator” with a few significant differences described later in this
chapter.

6-2 SILVACO International

 3D Device Simulation

6.2: 3D Structure Generation

All 3-D programs in ATLAS supports structures defined on 3D prismatic meshes. Structures may have
arbitrary geometries in two dimensions and consist of multiple slices in the third dimension.
There are two methods for creating a 3D structure that can be used with ATLAS. One way is through
the command syntax of ATLAS. Another way is through an interface to DEVEDIT3D.
A direct interface from ATHENA to 3D ATLAS isn’t possible. But DEVEDIT3D provides the ability to
read in 2D structures from ATHENA and extend them non-uniformly to create 3D structures for
ATLAS.
ATLAS Syntax For 3D Structure Generation
Mesh generation
Chapter 2: “Getting Started with ATLAS”, Section 2.6.3: “Using The Command Language To Define A
Structure” covers the generation of 2D and 3D mesh structures using the ATLAS command language.
The [Link] statement and the NZ and THREE.D parameters of the MESH statement are required to
extend a 2D mesh into 3D.
Conventionally, slices are made perpendicular to the Z axis. The mesh is triangular in XY but
rectangular in XZ or YZ planes.

Region, Electrode, and Doping definition

Chapter 2: “Getting Started with ATLAS”, Section 2.6.3: “Using The Command Language To Define A
Structure” also covers the definition of 2D regions, electrodes and doping profiles. To extend the
regions into 3D, use the [Link] and [Link] parameters. For example:
REGION NUM=2 MATERIAL=Silicon [Link]=0 [Link]=1 [Link]=0 [Link]=1 [Link]=0
[Link]=1
ELECTRODE NAME=gate [Link]=0 [Link]=1 [Link]=0 [Link]=1 [Link]=0 [Link]=1
DOPING GAUSS [Link] CONC=1E20 JUNC=0.2 [Link]=0.0 [Link]=1.0
For 2D regions or electrodes defined with the command language, geometry is limited to rectangular
shapes. Similarly, in 3D regions and electrodes are composed of rectangular parallelopipeds.

DevEdit3D Interface
DEVEDIT3D is a graphical tool that allows you to draw 3D device structures and create 3D meshes. It
can also read 2D structures from ATHENA and extend them into 3D. These structures can be saved
from DEVEDIT3D as structure files for ATLAS. Also, save a command file when using DEVEDIT3D. This
file is used to recreate the 3D structure inside DEVEDIT3D, which is important, since DEVEDIT3D
doesn’t read in 3D structure files.
ATLAS can read structures generated by DEVEDIT3D using the command:
MESH INF=<filename>
The program is able to distinguish automatically between 2D and 3D meshes read in using this
command.

Defining Devices with Circular Masks

DEVEDIT3D makes a triangular mesh in the XY plane and uses z-plane slices. This means, that
normally the Y direction is vertically down into the substrate. But in the case of using circular masks,
you need to rotate the device.
With defining devices using circular masks in DEVEDIT3D, the XY plane should be the surface of the
device and the Z direction should be into the substrate.

SILVACO International 6-3

ATLAS User’s Manual

6.3: Model And Material Parameter Selection in 3D

Models and material parameters are chosen in 3-D in common with other 2-D modules using the
MODELS, IMPACT, MATERIAL, MOBILITY, INTERFACE, and CONTACT statements.
The models available in 3D device simulation programs are:

6.3.1: Mobility
• Table for 300K (CONMOB)
• Thomas (ANALYTIC)
• Arora’s Model (ARORA)
• Klaassen’s Model (KLAASSEN)
• Lombardi’s Model (CVT)
• Yamaguchi Model (YAMA)
• Parallel Field Dependence (FLDMOB)
• Parallel Field Dependence with negative differential mobility (FLDMOB EVSATMOD=1)

6.3.2: Recombination
• Shockley Read Hall (SRH)
• Concentration dependent lifetime SRH (CONSRH)
• Klaassen’s concentration dependent lifetime SRH (KLASRH)
• Auger (AUGER)
• Klaassen’s concentration dependent Auger recombination model (KLAAUG)
• Optical Recombination (OPTR)
• Bulk and interface traps (TRAP, INTTRAP)
• Continuous defect states (DEFECT)

6.3.3: Generation
• Selberherr Impact Ionization (IMPACT SELB)
• Crowell Impact Ionization (IMPACT CROWELL)
• Hot Electron Injection (HEI)
• Fowler Nordheim Tunneling (FNORD)
• Single Event Upset (SINGLEEVENTUPSET)

6.3.4: Carrier Statistics

• Boltzmann (default)
• Fermi (FERMI)
• Band Gap Narrowing (BGN)
• Incomplete Ionization (INCOMPLETE)
• Quantum Mechanical Effects (QUANTUM)

6.3.5: Boundary Conditions

• Ohmic and Schottky
• Current Boundary Conditions

6-4 SILVACO International

 3D Device Simulation

• Lumped Element Boundary Conditions

• Distributed Contact Resistance

6.3.6: Interface and Bulk Traps (Discrete and Continuous)

• Photogeneration with Ray Tracing (LUMINOUS3D):
All these models, with the exception of SINGLEEVENTUPSET, are documented in the Chapter 3:
“Physics” or in the 2D product chapters of this manual.

6.3.7: Single Event Upset Simulation

The capability of single event upset/photogeneration transient simulation is included in 3D using the
SINGLEEVENTUPSET statement (see Chapter 18: “Statements”, Section 18.39:
“SINGLEEVENTUPSET” for more information). It allows you to specify the radial, length, and time
dependence of generated charge along tracks. There can be a single particle strike or multiple strikes.
Each track is specified by an Entry Point Location (x0,y0,z0) and an Exit Point Location (x1,y1,z1).
This is assumed to be a cylinder with the radius defined with the RADIUS parameter.
The entry and exit points are specified by the ENTRYPOINT and EXITPOINT parameters of the
SINGLEEVENTUPSET statement. These are character parameters that represent the ordered triplet
coordinates of the entry and exit points of the particle track.
The electron/hole pairs generated at any point is a function of the radial distance, r, from the center of
the track to the point, the distance l along the track and the time, t. The implementation into ATLAS
allows you to define the generation as the number of electron-hole pairs per cm3 along the track
according to the equation:
6-1
where DENSITY and [Link] are defined as the number of generated electron/hole pairs per cm3.
In radiation studies, the ionizing particle is typically described by the linear charge deposition (LCD)
value, which defines the actual charge deposited in units of pC/µm. You can use this definition within
ATLAS by specifying the PCUNITS parameter in the SINGLEEVENTUPSET statement. If the user-
defined parameter, PCUNITS, is set in the SINGLEEVENTUPSET statement, then [Link] is the
generated charge, in pC/µm, and the scaling factor S is:

6-2

where RADIUS is defined on the SINGLEEVENTUPSET statement. If the PCUNITS parameter isn’t set,
then [Link] is the number of generated electron/hole pairs in cm-3, and the scaling parameter, ,
is unity.
Another common measure of the loss of energy of the SEU particle, as it suffers collisions in a
material, is the linear energy transfer (LET) value which is given in units of MeV/mg/cm2 (or MeV-
cm2/mg). In silicon, energy can be converted to charge by considering that approximately 3.6 eV of
energy is required to generate an electron-hole pair. The conversion factor from the LET value to the

SILVACO International 6-5

ATLAS User’s Manual

B4
L2 ( l ) = B1 ( B2 + l ⋅ B3 ) 6-4

where the parameters: A1, A2, A3, A4, B1, B2, B3 and B4 are user-definable as shown in Table 6-1.

Table 6-1. User-Specifiable Parameters for Equations 6-3 and 6-4

Statement Parameter Default Units

SINGLEEVENTUPSET A1 1

SINGLEEVENTUPSET A2 0 cm-1

SINGLEEVENTUPSET A3 0

SINGLEEVENTUPSET A4 0 cm-1

SINGLEEVENTUPSET B1 1

SINGLEEVENTUPSET B2 1

SINGLEEVENTUPSET B3 0 cm-1

SINGLEEVENTUPSET B4 0

Note: The default parameters in Table 6-1 were chosen to result in constant carrier or charge generation as a function of
distance along the particle track.

The factor R(r) is the radial parameter which is defined by one of two equations. The default is:

r
R ( r ) = exp  – ------------ 6-5
 RADIUS

where r is the radial distance from the centre of the track to the point and RADIUS is a user-definable
parameter as shown in Table 6-2. You can choose an alternative expression if the RADIALGAUSS
parameter is specified on the SINGLEEVENTUPSET statement. In this case, R(r) is given by:

r 2
R ( r ) = exp –  ------------ 6-6
 RADIUS

The time dependency of the charge generation T(t) is controlled with the TC parameter through two
functions.
For TC=0:
T ( t ) = deltafunction ( t – T0 ) 6-7

6-6 SILVACO International

 3D Device Simulation

For TC>0:

t – TO 2
2e –  ---------------
TC
T ( t ) = -------------------------------------------- 6-8
– TO
TC π erfc  ----------
TC

where TO and TC are parameters of the SINGLEEVENTUPSET statement.

Table 6-2. User-Specifiable Parameters for Equations 6-8, 6-2, 6-7, and 6-8

Statement Parameter Default Units

SINGLEEVENTUPSET DENSITY 0.0 cm-3

SINGLEEVENTUPSET [Link] 0.0 cm-3 or pC/µm
SINGLEEVENTUPSET TO 0.0 s

SINGLEEVENTUPSET TC 0.0 s

SINGLEEVENTUPSET RADIUS 0.05 µm

The following example shows a particle strike that is perpendicular to the surface along the z-plane, so
only the z-coordinates change has a radius of 0.1 µm and a LET value of 20 MeV-cm2/mg
([Link]=0.2). The strike has a delay time of 60 ps and the Gaussian profile has a characteristic
time of 10 ps. The generation track for this striking particle is from z=0 to z=10 µm and carrier
generation occurs along its entire length.
single entry="5.5,0.0,0.0" exit="5.5,0.0,10" radius=0.1 \
pcunits [Link]=0.2 t0=60.e-12 tc=10.e-12
User-defined SEU in 3D
In addition to the model described by the SINGLEEVENTUPSET statement, you can use the
C-INTERPRETER to specify an arbitrary generation profile. This is specified using the syntax:
BEAM NUM=1 [Link]=myseu.c
.
SOLVE B1=1.0 RAMPTIME=1e-12 DT=1e-14 TSTOP=1e-7
The [Link] parameter indicates an external c-language subroutine conforming to the template
supplied. The file, , returns a time and position dependent value of carrier generation in 3-D.
The value returned by , is multiplied by the B1 parameter at each timestep.
For more information about the C-Interpreter, see Appendix A: “C-Interpreter Functions”.

SILVACO International 6-7

ATLAS User’s Manual

6.3.8: Boundary Conditions in 3D

External Passive Elements
You can attach external lumped resistors, capacitors and inductors to any contact. The syntax is the
same as for the 2D products, which is:
CONTACT NAME=drain RES=1e3 CAP=1e-12 L=1e-6
You can also apply distributed resistances to contacts. The algorithm used for estimating contact area
for 3D distributed contact resistance multiplies the contact perimeter in a given Z plane by the
displacement in the Z direction. This algorithm will only work properly for planar contacts that do not
vary in the Z direction. They may however abruptly terminate or start in the Z direction.
The units of lumped external passive elements are ohms for resistors, Farads for capacitors and
Henrys for inductors. Distributed contact resistance is defined in ohms.cm3.
Thermal Contacts for GIGA3D
Thermal contacts for non-isothermal simulation in GIGA3D are defined in an analogous manner to the
2D thermal contacts in GIGA. The [Link] and [Link] parameters are used to define the extent of the
thermal contact in the z-plane. The units of the thermal resistance parameter ALPHA are scaled in 3D
to W/(cm.K). For more information about GIGA3D, see Chapter 7: “Giga: Self-Heating Simulator”.

6.3.9: BLAZE3D Models

Models for simulating III-V materials are supported in BLAZE3D. The material parameters for all
compound semiconductors are the same as used in 2-D simulations with BLAZE. Specialized models for
III-V materials including heterojunction transport, optical recombination, and negative differential
field dependent mobility are described in Chapter 5: “Blaze: Compound Material 2D Simulator”.

6.3.10: TFT3D Models

Models for simulating thin-film transistors made from amorphous or polycrystalline semiconductors
are supported in TFT3D. The definition of the continuous defect states in the bandgap is performed
using the same parameters as in 2D simulations with TFT. The models for continuous defect (or trap)
densities are documented in Chapter 14: “TFT: Thin-Film Transistor Simulator”.

6.3.11: QUANTUM3D Models

Models for simulating quantum effects semiconductors are supported in QUANTUM3D. The definition
of the quantum moments solver is the same as in 2D simulations with QUANTUM. The models for
simulating quantum effects and the parameters to control the model are shown in Chapter 13:
“Quantum: Quantum Effect Simulator”.

6.3.12: LUMINOUS3D Models

Many of the models for simulating photodetection in LUMINOUS3D are similar to those for simulating
photodetection in LUMINOUS. For more information about LUMINOUS, see Chapter 10: “Luminous:
Optoelectronic Simulator”. This chapter, however, shows several important differences between
LUMINOUS and LUMINOUS3D.
Optical Source Specification
The specification of the optical source in 3D is similar to the specification of the source described in
Chapter 10: “Luminous: Optoelectronic Simulator”, Section 10.4.1: “Defining Optical Sources” with a
few additions to account for the third dimension.
Figures 6-1 and 6-2 show the complete specif

6-8 SILVACO International

 3D Device Simulation

PHI parameters. The PHI parameter is analogous to the LUMINOUS parameter, ANGLE, and specifies
the direction of propagation relative to the device x axis (see Figure 6-1). In face, PHI and ANGLE are
synonymous. The THETA parameter specifies the rotated angle of propagation relative to the x-y plane
(see Figure 6-2).

Figure 6-1: Source beam coordinate rotation around Z-axis

Figure 6-2: Source beam coordinate rotation around Y-axis

Ray tracing
The discrete sampling of the source beam into rays in LUMINOUS3D is unlike that done in LUMINOUS.
In LUMINOUS, the source beam is automatically broken up into a set of rays that resolve the device
topology and variations in the interior of the device. In LUMINOUS3D, this process is more complex and
is a computational burden. As such in LUMINOUS3D, specify a discrete sampling of the source beam
(see Figure 6-3).

SILVACO International 6-9

 3D Device Simulation

As shown in Figure 6-4, the lenslet is represented by a spherical section atop a plane. The center of the
sphere is specified by the LENS.X, LENS.Y and LENS.Z parameters of the BEAM statement. The radius
of the sphere is specified by the [Link] parameter of the BEAM statement. The location of the
planar part surrounding the lenslet relative to the y=0 plane is specified by the [Link]
parameter. The index of refraction of the lenslet as well as the volume between the lens plane and the
device surface is specified by the [Link] parameter of the BEAM statement.
Note that the volume associated with the lenslet is not meshed or considered in the solution of the
device equations. It is merely used in the ray tracing. Also, you should keep in mind that lenslet can
only be specified in planes perpendicular to the y axis.

Other differences between Luminous and Luminous3D

LUMINOUS3D has several other differences with LUMINOUS. For example, in LUMINOUS the rays
generated during ray tracing are stored in all subsequent structure files. In LUMINOUS3D, this isn’t the
case. To save the rays in LUMINOUS3D, set the RAYTRACE parameter of the BEAM statement to the
name of a file where the results of the ray trace are to be stored.
Another difference is that LUMINOUS3D has an option to cause all metal regions to act as perfect
reflectors. To enable this option, specify [Link] in the BEAM statement.
LUMINOUS3D also allows you to simulate periodicity with respect to ray tracing by specifying
PERIODIC on the BEAM

SILVACO International 6-11

ATLAS User’s Manual

6.4: Numerical Methods for 3D

6.4.1: DC Solutions
There are several differences between the default numerical methods applied in 2D ATLAS and those
applied in 3D ATLAS. For example, with respect to non-linear iteration strategies, the current version
of the 3D simulator does not support the BLOCK method. The NEWTON and GUMMEL iteration strategies
are supported for 3D simulations, whereas NEWTON, GUMMEL and BLOCK are all supported for 2D
simulations.
In solving the linear subproblem, the default approach in 3D is to use an iterative solver. This is
believed to be the most efficient method for general 3D problems. In 2D, the direct solver is used by
default. You may find it desirable to use direct methods in 3D problems due to improved convergence of
computational efficiency. You can select the direct method by specifying DIRECT in the METHOD
statement.
Also, in 3D there are two linear iterative solution methods available. The default is ILUCGS
(incomplete lower-upper decomposition conjugate gradient system) and BICGST (bi-conjugate gradient
stabilized). Historically, tests have shown that the current implementation of ILUCGS is slightly more
stable than BICGST and is the default iterative solver in 3D. You can define the BICGST solver by
specifying BICGST in the METHOD statement.

Note: Iterative solvers are recommended for large problems, typically greater than 5000 node points, due to lower solution
times and memory usage.

6.4.2: Transient Solutions

In transient mode, a semi-implicit scheme is used in addition to the default TR-BDF algorithm [114].
This algorithm is recommended for complex simulations such as Single Event Upset. To select this
method, use:
METHOD HALFIMPL

6.4.3: Obtaining Solutions In 3D

ATLAS3D programs can perform DC and transient analysis in an equivalent manner to 2D. The
SOLVE statement is used to define the solution procedure. The syntax used is described as that in
Chapter 2: “Getting Started with ATLAS”, Section 2.9: “Obtaining Solutions”.

6.4.4: Interpreting the Results From 3D

The log files produced by 3D ATLAS can be plotted in TONYPLOT exactly as those that result from S-
PISCES or BLAZE. The only difference is the units of the currents produced. Log files from 3D
simulations save current in Amperes, whereas the 2D simulations use Amperes/micron.
The solution files produced by 3D ATLAS should be plotted using TONYPLOT3D. These files cannot be
read directly into the 2D TONYPLOT program. TONYPLOT3D contains features that allows you to make
slices of the 3D structure, which can be plotted in 2D TONYPLOT. For more information about

6-12 SILVACO International

 7.1: Overview
GIGA extends ATLAS to account for lattice heat flow and general thermal environments. GIGA
implements Wachutka’s thermodynamically rigorous model of lattice heating [84], which accounts for
Joule heating, heating, and cooling due to carrier generation and recombination, and the Peltier and
Thomson effects. GIGA accounts for the dependence of material and transport parameters on the
lattice temperature, and supports the specification of general thermal environments using a
combination of realistic heat-sink structures, thermal impedances, and specified ambient
temperatures. GIGA works with both S-PISCES and BLAZE, and with both the drift-diffusion and energy
balance transport models. See Chapter 3: “Physics”, the “Drift-Diffusion Transport Mode Traneoo.79f
an d the a79f3a

.If not,( )-6.1 rede

SILVACO International 7-1

Een
,e in foti
egynBbalance
ono (n)-3.4( Trausor
the)-5.9(s)-08(
to ee )]TJETEMC/Span < </MCID1 2>>BDCBT9.96 0 0 9.96 86.4 230.1 Tm0.0 067 Tc0 Tw[ mo dels
ATLAS User’s Manual

7.2: Physical Models

7.2.1: The Lattice Heat Flow Equation
GIGA adds the heat flow equation to the primary equations that are solved by ATLAS. The heat flow
equation has the form:
∂T L
C ---------- = ∇ ( κ∇T L ) + H 7-1
∂t

where:
C is the heat capacitance per unit volume.
κ is the thermal conductivity.
H is the heat generation
TL is the local lattice temperature.
The heat capacitance can be expressed as C = ρCp, where Cp is the specific heat and ρ is the density of
the material.
Specifying the [Link] parameter in the MODELS statement includes the lattice heat flow equation
in ATLAS simulations.
GIGA supports different combinations of models. For example, if the HCTE and [Link] parameters
are specified in the MODEL statement, and both particle continuity equations are solved, all six
equations are solved. If [Link] is specified instead of HCTE, only five equations are solved and the
hole temperature Tp is set equal to lattice temperature TL.

Specifying Heat Sink Layers For Thermal Solutions

Regions may be defined for inclusion only in thermal calculations. These regions will typically consist
of layers associated with heat sinks. They are defined using the REGION statement. Even though in
reality the heat sink materials are typically metal conductors, it is easier to specify these layers with
the material type, INSULATOR. This is because as insulators the program will only solve heat flow and
not attempt to solve current continuity in these layers. The region number is subsequently used as an
identifier when thermal conductivities and heat capacities are assigned to these regions.
The following statements specify two layers of a heat sink for inclusion in the thermal calculation.
REGION NUM=5 [Link]=0.5 [Link]=2.0 INSULATOR
REGION NUM=6 [Link]=2.0 [Link]=3.0 INSULATOR

Specifying Thermal Conductivity

The value of thermal conductivity, k, for each region should be specified in the MATERIAL statement.
Because thermal conductivity is generally temperature dependent, the following four models are
available:

k ( T ) = [Link] ( W ⁄ cm · K ) . 7-2

[Link]
k ( T ) = ( [Link] ) ⁄ ( T ⁄ 300 ) ( W ⁄ cm · K ) . 7-3

2
k ( T ) = 1 ⁄ ( TC.A + ( TC.B )∗ T + ( TC.C )∗ T ) ( W ⁄ cm · K ) . 7-4

7-2 SILVACO International

 Giga: Self-Heating Simulator

k ( T ) = ( TC.E ) ⁄ ( T – TC.D ) ( W ⁄ cm · K ) . 7-5

The [Link], [Link], TC.A, TC.B, TC.C, TC.D and TC.E parameters are all user-specifiable in the
MATERIAL statement. The choice of models is also user-specified in the MATERIAL statement. To choose
the model in Equation 7-2, specify the [Link] parameter. To choose the model in Equation 7-3,
specify the [Link] parameter. To choose the model in Equation 7-4, specify the [Link]
parameter. To choose the model in Equation 7-5, specify the [Link] parameter. The default
model is the polynomial model in Equation 7-4.
The following statements would be used to specify the temperature dependent thermal conductivity of
the regions previously defined.
MATERIAL REGION=5 TC.A=<n> TC.B=<n> TC.C=<n>
MATERIAL REGION=6 TC.A=<n> TC.B=<n> TC.C=<n>

C-Interpreter Defined Thermal Conductivity

You can use the C-INTERPRETER to define the thermal conductivity, TCOND, as a function of the lattice
temperature, position, doping and fraction composition. This is defined using the syntax:
MATERIAL REGION=<n> [Link]=<filename>
where the <filename> parameter is an ASCII file containing the C-INTERPRETER function. For more
information about the C-INTERPRETER, see Appendix A: “C-Interpreter Functions”.

Specifying Heat Capacity

For transient calculations, specify heat capacities for every region in the structure. These are also
functions of the lattice temperature, and are modeled as:

HC ⋅D 3
C = HC ⋅A + HC ⋅ B T + HC ⋅ C T + --------------
- ( J ⁄ cm ⁄ K ) 7-6
2
T
Default values of HC.A, HC.B, HC.C, and HC.D are provided for common materials. These values can
be specified on the MATERIAL statement.
The following statements would be used to specify the temperature dependent heat capacities of the
regions previously defined.
MATERIAL REGION=5 HC.A=<n> HC.B=<n> HC.C=<n> HC.D=<n>
MATERIAL REGION=6 HC.A=<>> HC.B=<n> HC.C=<n> HC.D=<n>

Table 7-1. User Specifiable Parameters for Equation 7-6

Statement Parameter Units

MATERIAL HC.A J/cm3/K

MATERIAL HC.B J/cm3⋅K2

MATERIAL HC.C J/cm3⋅K3

MATERIAL HC.D JK/cm3

SILVACO International 7-3

ATLAS User’s Manual

C-Interpreter Defined Thermal Capacity

You can use the C-INTERPRETER to define the thermal capacity, TCAP, as a function of the lattice
temperature, position, doping and fraction composition. This is defined using the syntax:
MATERIAL REGION=<[Link]> [Link]=<filename>
where the <filename> parameter is an ASCII file containing the C-INTERPRETER function. For more
information about the C-INTERPRETER, see Appendix A: “C-Interpreter Functions”.

7.2.2: Non-Isothermal Models

Effective Density Of States
When lattice heating is specified with the drift-diffusion transport model, the effective density of
states for electrons and holes are modeled as functions of the local lattice temperature as defined by
Equations 3-31 and 3-32 in Chapter 3: “Physics”.
When lattice heating is specified with the energy balance model, the effective densities of states are
modeled as functions of the local carrier temperatures, and , as defined by Equations 3-111 and
3-112 in Chapter 3: “Physics”.

Non-isothermal Current Densities

When GIGA is used, the electron and hole current densities are modified to account for spatially
varying lattice temperatures:

J n = – q µ n n ( ∇φ n + P n ∇T L ) 7-7

J p = – q µ p p ( ∇φ p + P p ∇T L ) 7-8

where:
Pn and Pp are the absolute thermoelectric powers for electrons and holes.
Pn and Pp are modeled as follows:

k n 5 
P n = ---  ln ------ –  --- + KSN  7-9
q  Nc  2  

k p 5 
P p = ---  ln ------- –  --- + KSP  7-10
q N 2 
v
The default values for Equations 7-9 and 7-10 are given in Table 7-2.

Table 7-2. User-Specifiable Parameters for Equations 7-9 and 7-10

Statement Parameter Default Units

MODELS KSN -1 None

MODELS KSP -1 None

7-4 SILVACO International

 Giga: Self-Heating Simulator

7.2.3: Heat Generation

When carrier transport is handled in the drift-diffusion approximation the heat generation term, H,
used in Equation 7-1 has the form [83]:

7-11

In the steady-state case, the current divergence can be replaced by the net recombination and
Equation 7-11 simplifies to:

7-12

where:

is Joule heating term,

is recombination and generation heating and cooling term,

accounts for the Peltier and Thomson effects.

A simple and intuitive form of H that has been widely used in the past is:

7-13

GIGA can use either Equations 7-12 or 7-13 for steady-state calculations. By default, Equation 7-13 is
used. Equation 7-12 is used if the [Link] parameter is specified in the MODELS statement. To
enable/disable the individual terms of Equation 7-12, use the [Link], [Link], and [Link]
parameters of the MODEL statement.
If the general expression shown in Equation 7-11 is used for the non-stationary case, the derivatives

and are evaluated for the case of an idealized non-degenerate semiconductor and

complete ionization.

SILVACO International 7-5

ATLAS User’s Manual

When carrier transport is modeled in the energy balance approximation the following expression is
used for H:
H = W n + W p + E g U, 7-14

where U, Wn and Wp are defined by Equations 3-113 through 3-115 in Chapter 3: “Physics”.

7.2.4: Thermal Boundary Conditions

At least one thermal boundary condition must be specified when the lattice heat flow equation is
solved. The thermal boundary conditions used have the following general form:

u
σ  J tot ⋅ s = α ( T L – T ext ) 7-15
 

u
where σ is either 0 or 1, J t ot is the total energy flux and s is the unit external normal of the boundary.

The projection of the energy flux onto s is:

 u  ∂T L
 J tot ⋅ s = – κ ---------- + ( TL P n + φ n ) J n ⋅ s + ( T L P p + φ p ) ( J p ⋅ s ) 7-16
  ∂n

When σ = 0, Equation 7-15 specifies a Dirichlet (fixed temperature) boundary condition:

T = TEMPER . 7-17
L

where TEMPER can be defined in the THERMCONTACT statement as shown in the next section. Dirichlet
boundary conditions may be specified for an external boundary (which may coincide with an electrode),
or for an electrode that is inside the device.
When σ = 1, Equation 7-15 takes the form:

1
u ⋅ s ) = -------
( J tot - ( T – TEMPER ) 7-18
R th L

where the thermal resistance, Rth, is given by:

1
R th = ------------------------
ALPHA
and ALPHA is user-definable on the THERMCONTACT statement.

Specifying Thermal Boundary Conditions

Setting thermal boundary conditions is similar to setting electrical boundary conditions. The
THERMCONTACT statement is used to specify the position of the thermal contact and any optional
properties of the contact. Thermal contacts can be placed at any position in the device (including
sidewalls). Equation 7-18 is used if a value is specified for α. Otherwise, Equation 7-17 is used.
The following command specifies that thermal contact number 1 is located between x=0 µm and x=2
µm at y=0 µm, and that the temperature at the contact is 300K.
THERMCONTACT NUM=1 [Link]=0 [Link]=2 [Link]=0 [Link]=0 TEMP=300

7-6 SILVACO International

 Giga: Self-Heating Simulator

You can use simpler statement if the coordinates of a thermal contact coincide with the coordinates of
an electrical contact. In this case, it is permissible to specify the location of the thermal contact by
referring to the electrode number of the electrical contact. For example, the statement:
THERMCONTACT NUM=1 [Link]=3 TEMP=400
specifies that thermal contact number 1 is located in the same position as electrode number 3 and that
the contact temperature is 400K.
Specifying the BOUNDARY parameter gives you flexibility in applying thermal boundary conditions.
This parameter is set by default and means that the thermal boundary condition will only be set on the
outside surface of the thermal contact, where it forms the exterior of the device. If the parameter is
cleared by specifying ^BOUNDARY in the THERMCONTACT statement, the boundary conditions will be
applied to the interior of the thermal contact and to the part of the surface of the thermal contact that
forms an interface with the interior of the device. In a 2D model, you may have a thermal contact that
is internal to the device and in this case you would need to specify ^BOUNDARY. Otherwise, the contact
will be ignored.
For example:
THERMCONTACT [Link]=1 ^BOUNDARY TEMPER=450 ALPHA=2.5
where the ELECTRODE extends into the device and will set the interior points of the thermalcontact to
450K and will apply the flux boundary condition on all faces of the thermal contact that interface with
the interior of the device.

Table 7-3. User Specifiable Parameters for Equations 7-17 and 7-18

Statement Parameter Units Default

THERMCONTACT ALPHA W/(cm2K) ∞

THERMCONTACT BOUNDARY - True

THERMCONTACT TEMPER K 300

Representing a thermal environment, in terms of thermal impedances, leads to efficient solutions. But,
thermal impedance representations are typically only approximations. Detailed thermal modeling
(e.g., the effect of heat sink design changes) typically requires using detailed modeling of thermal
regions with specified external ambient temperatures.

Note: You can’t alter the value of a thermal resistor within a sequence of SOLVE statements. Rerun the input file whenever a
thermal resistor is changed.

7.2.5: Temperature Dependent Material Parameters

GIGA automatically uses the built-in temperature dependence of the physical models that are
specified. When lattice heating is specified, temperature dependent values are evaluated locally at
each point in the device. When lattice heating is not solved, the models only provide global
temperature dependence (i.e., all points in the device are assumed to be at the specified temperature).
The non-isothermal energy balance model uses the same carrier temperature dependencies of the
mobility and impact ionization models as in the pure energy balance case, but with coefficients that
depend on the local lattice temperature. Impact ionization coefficients depend on lattice temperature.
Almost all other models and coefficients depend on lattice temperature.

SILVACO International 7-7

ATLAS User’s Manual

When lattice heating is used there is no point in specifying models that do not include temperature
dependence. For mobilities, don’t specify CONMOB. Instead, specify ANALYTIC or ARORA. For impact
ionization coefficients, specify the SELBERHERR model.
GIGA can account for the temperature dependence of the minority carrier lifetimes for electrons and/or
holes. The [Link] (electrons) and [Link] (holes) parameters of the MATERIAL statement are used
to select this model. This model is turned on whenever the value of [Link] or [Link] is greater
than 0, which is the default.
The temperature dependence of electron and hole lifetimes in the SRH recombination model have the
forms:

T LT • TAUN
τ n = TAUN0  --------- 7-19
 300

T LT • TAUP
τ p = TAUP0  --------- 7-20
300

Table 7-4. User-Specifiable Parameters for Equations 7-19 and 7-20

Statement Parameter Default Units

MATERIAL TAUN0 1×10-7 s

MATERIAL TAUP0 1×10-7 s

MATERIAL [Link] 0

MATERIAL [Link] 0

See Equations 3-247 and 3-248 in Chapter 3: “Physics” for information regarding concentration
dependent lifetimes.

7.2.6: C-Interpreter Defined Peltier Coefficients

You can use the C-INTERPRETER to define the Peltier coefficients, KSN and KSP, as a function of the
electron and hole carrier temperatures, Tn and Tp. This is defined using the syntax:
MODELS [Link]=<filename> [Link]=<filename>
where the <filename> parameter is an ascii file containing the C-INTERPRETER function. See
Appendix A: “C-Interpreter Functions” for more information regarding the C-INTERPRETER and its
functions.

7-8 SILVACO International

 Giga: Self-Heating Simulator

7.3: Applications of GIGA

7.3.1: Power Device Simulation Techniques
This section contains a series of techniques, which you may find useful when simulating typical power
device structures. Not all of the features described below are specific to GIGA and are common to the
ATLAS framework.

Floating Guard Rings

No special syntax is needed for the simulation of uncontacted doping areas used in floating guard
rings. The program is able to simulate guard ring breakdown with the standard impact ionization
models. In some extreme cases, convergence may be slow due to poor initial guesses. If convergence is
slow, both GUMMEL and NEWTON should be specified in the METHOD statement.

Floating Field Plates

You should use the ELECTRODE statement to specify the field plate regions as electrodes. If these plates
do not contact any semiconductor, then these electrodes can be set to float in the same manner as
EEPROM floating gates. The following statement line specifies that the field plate region PLATE1 is a
floating field plate.
CONTACT NAME=PLATE1 FLOATING
If the plates do contact the semiconductor, this syntax must not be used. Instead current boundary
conditions are used at the electrode with zero current. See Chapter 2: “Getting Started with ATLAS”,
the “Floating Contacts” section on page 2-28 for more information about floating electrodes.

External Inductors
Inductors are commonly used in the external circuits of power devices. The CONTACT statement can be
used to set an inductor on any electrode. The following statement sets an inductance on the drain
electrode of 3 µH/µm.
CONTACT NAME=DRAIN L=3E-3
The next statement is used to specify a non-ideal inductor with a resistance of 100 Ω/µm.
CONTACT NAME=DRAIN L=3E-3 R=100

7.3.2: More Information

Many examples using GIGA have been installed on your distribution tape or CD. These include power
device examples but also SOI and III-V technologies. You can find more information about the use of
GIGA by reading the text associated with each example.

SILVACO International 7-9

ATLAS User’s Manual

This page is intentionally left blank.

7-10 SILVACO International

 Chapter 8:
Laser: Edge Emitting Simulator

8.1: Overview
LASER performs coupled electrical and optical simulation of semiconductor lasers. LASER works in
conjunction with BLAZE and allows you to:
• Solve the Helmholtz Equation (see Equation 8-1) to calculate the optical field and photon
densities.
• Calculate the carrier recombination due to light emission (i.e., stimulated emission).
• Calculate optical gain, depending on the photon energy and the quasi-Fermi levels.
• Calculate laser light output power.
• Calculate the light intensity profile corresponding to the multiple transverse modes.
• Calculate light output and modal gain spectra for several longitudinal modes.
To perform a LASER simulation, you need to know ATLAS and BLAZE first. If you don’t, read Chapter 2:
“Getting Started” and Chapter 5: “BLAZE”.

SILVACO International 8-1

ATLAS User’s Manual

8.2: Physical Models

To simulate semiconductor lasers, the basic semiconductor equations (See Equations 3-1 to 3-4 in
Chapter 3: “Physics”) are solved self-consistently with an optical equation that determines the optical
field intensity distribution. LASER uses the following coordinate system:
• The x axis is perpendicular to the laser cavity and goes along the surface (from left to right).
• The y axis is perpendicular to the laser cavity and goes down from the surface.
• The z axis goes along the laser cavity.
The x and y axes are the same as in other ATLAS products. The electrical and optical equations are
solved in the x and y plane (i.e., perpendicular to the laser cavity).

8.2.1: Helmholtz Equation

LASER solves a two-dimensional Helmholtz equation to determine the transverse optical field profile
[102]:

ω 
∇ ( )  ε( ) –β  ( ) 8-1
 
 
where:

∂ ∂ 
∇   is the two-dimensional Laplace operator, ω is the frequency corresponding to
∂ ∂ 
longitudinal mode (ω corresponds to OMEGA on the LASER statement for single frequency
simulations), is the velocity of light in vacuum, and ε is the high frequency dielectric permittivity.
Equation 8-1 is a complex eigenvalue problem. LASER solves this equation to determine the set of
complex eigenvalues (β ) and corresponding eigenfunctions . LASER takes into account only the
fundamental transverse mode solution, so the index ( ) will be dropped from subsequent equations.
In principle, Equation 8-1 should be solved for each longitudinal mode that is taken into account. Since
very few longitudinal modes are actually lasing, LASER solves Equation 8-1 only once for the
longitudinal mode with the greatest power and subsequently assumes:

8-2

where is the optical field corresponding to the most powerful longitudinal mode. This
assumption is reasonable, since the shape of the solution is almost independent of the frequency
within the range of interest.
For dielectric permittivity, LASER uses the following model [103]:

ε ( ) ε ( ALPHAA FCN ⋅ FCP ⋅ )

ε( r z) ε ( – ALPHAR ) – 8-3
ω ω

8-2 SILVACO International

 Laser: Edge Emitter Simulation

where:
• ε is the bulk permittivity.
• ALPHAR is a line width broadening factor.
• j= –1
• kω = ω/c
• is the local optical gain.
• ALPHAA is the bulk absorption loss and is specified in the MATERIAL statement.
• ABSORPTION must be specified in the LASER statement to include absorption loss.
• FCN and FCP are the coefficients of the free-carrier loss and are set through the MATERIAL
statement. FCARRIER must be specified on the LASER to include this loss mechanism.

Table 8-1: User-Specifiable Parameters for Equation 8-3

Statement Parameter Default Units

MATERIAL ALPHAR 4.0

MATERIAL ALPHAA 0.0 cm-1

MATERIAL EPSINF

MATERIAL FCN 3.0×10-18 cm2

MATERIAL FCP 7.0×10-18 cm2

LASER ABSORPTION FALSE

LASER FCARRIER FALSE

Bulk Permittivity (ε0)

The following rules apply to the specification of the bulk permittivity.
• If located in an MQW and EPSILON is defined in the MQW statement, that value is used.
• If EPSINF is defined in the MATERIAL statement, that value is used.
• If [Link] defined in the MATERIAL statement, then the bulk permittivity is calculated from
the square of the index returned from the C-Interpreter function defined in the file specified by
[Link].
• If [Link] is defined in the MATERIAL statement, the bulk permittivity is calculated as the
square of the index interpolated from the table specified in the file pointed to by [Link].
• If the material is listed in Appendix B: “Material Systems”, Table B-26, the permittivity is
calculated as the square of index interpolated from the built-in tables for that material.
• If PERMITTI is defined in the MATERIAL statement, that value is used. If it’s not defined, the
default material permittivity is used.

8.2.2: Local Optical Gain

For a discussion of gain models, see Chapter 3: “Physics”, Sections 3.9.3: “The Standard Gain Model”
through 3.9.5: “Tayamaya's Gain Model”.

SILVACO International 8-3

ATLAS User’s Manual

8.2.3: Stimulated Emission

Carrier recombination due to stimulated light emission is modeled as follows:

( ) ∑ NEFF ( ) ( ) ⋅ 8-4

where is the recombination rate due to stimulated light emission, NEFF is the group effective
refractive index, and is the linear photon density. The subscript in this equation and all
subsequent equations refer to a modal quantity. For example, in Equation 8-4 is the photon linear
density for mode . The NEFF parameter is user-specifiable but has a default value of 3.57
(See Table 8-2).

Table 8-2: User-Specifiable Parameters for Equation 8-4

Statement Parameter Default Units

LASER NEFF 3.57

8.2.4: Photon Rate Equations

Optical gain provides a link between optical and electrical models. The optical gain depends on the
quasi-Fermi levels and in turn impacts dielectric permittivity (see Equation 8-3), and by the coupling
between the stimulated carrier recombination rate ( ) and the density of photons ( ) as described by
Equation 8-4.
To determine , LASER solves the system of photon rate equations:

 LOSSES
 – –  8-5
 NEFF τ NEFF 

where the modal gain is given by:

∫∫ ( )⋅ ( ) ⋅ 8-6

and the modal spontaneous emission rate is given by:

∫∫( ( )) ⋅ 8-7

LOSSES is the internal losses and can be specified in the MODEL statement (See Table 8-3). is the
normalized optical field.

8-4 SILVACO International

 Laser: Edge Emitter Simulation

Table 8-3: User Specifiable Parameters for Equation 8-5

Statement Parameter Default Units

LASER LOSSES 0 cm-1

The modal photon lifetime, τphm , in Equation 8-8 represents the losses in the laser. The losses per mode
are given by [104,107]:

(α α α ) 8-8
τ NEFF

αa is the bulk absorption loss, αfc is the free-carrier loss, and αmir is the mirror loss. The models for
absorption and free carrier loss are switched on with the ABSORPTION and FCARRIER parameters on
the LASER statement. The mirror loss is always on. These are defined as:

α ∫ ∫ ALPHAA ⋅ ( ) 8-9

α ∫ ∫ ( FCN FCP ) ⋅ ( ) ⋅ 8-10

The mirror loss, αmir , can be defined in two ways. One way is to define the percentage reflectivity of
both mirrors. The other way is to define the individual facet mirror reflectivities. If the former is
chosen, then set the [Link] parameter on the LASER statement to the percentage reflectivity
for the facet cavity mirrors. This assumes that both front and back mirrors are indentical and gives a
mirror loss calculated by:

 
α   8-11
[Link]  [Link] 

where [Link] is set to the length of the laser cavity on the LASER statement.
If you choose the latter model for αmir, set the RR and RF parameters on the LASER statement instead.
These parameters are the rear and front facet reflectivities respectively. The mirror loss is then
calcualated by:

α 8-12
⋅ CAVITY · LENGTH RF · RR

The user-specified parameters for the loss models in Equations 8-8 to 8-12 are given in Tables 8-1, 8-2,
8-3, and 8-4.

Table 8-4: User-Specifiable Parameters for LASER Loss Models

Statement Parameter Default Units

LASER [Link] 90.0 %

LASER [Link] 100.0 µm

SILVACO International 8-5

ATLAS User’s Manual

Note: Absorption loss and free carrier loss are switched off by default. These are switched on by specifying the
ABSORPTION and FCARRIER parameters in the LASER statement. Mirror loss is switched on by default to 90%.

8.2.5: Spontaneous Recombination Model

For a discussion of spontaneous recombination models, see Chapter 3: “Physics”, Sections 3.9.1: “The
General Radiative Recombination Model” and 3.9.2: “The Default Radiative Recombination Model”.

8.2.6: Optical Power

The optical power emitted from the front mirror is calculated by [104, 156]

ω ( ⁄( ))
------------------------------------------------------------------- 8-13
NEFF ⁄ ( – )⁄( – )
where is the photon density of mode , is the frequency of emitted light, and are the front
and rear mirror reflectivities, and NEFF is a user-defined in the LASER statement.

8.2.7: Gain Saturation

To simulate non-linear gain saturation in LASER, use the simple model described in following equation
[159].
( )
′( ) 8-14
( ) ⁄ GAIN · SAT

where [Link] is a user-specifiable parameter on the MATERIAL statement, ´( , ) is the local gain
and ( , ) is the local intensity given by

Σ
( ) ( ) 8-15

To enable this model, specify GAIN_SAT on the LASER statement. Non-linear, absorption loss is
similarly modeled using the following expression.
α( )
α′ ( ) 8-16
( ) ⁄ ABSORPTION · SAT

where α´( , ) is the local absorption and [Link] is a user-definable parameter on the
MATERIAL statement. To enable this model, specify ABSOR_SAT on the LASER statment.

8-6 SILVACO International

 Laser: Edge Emitter Simulation

8.3: Solution Techniques

LASER solves the electrical and optical equations self consistently. LASER solves the Helmholtz equation
in a rectangular region, which is a subdomain of the general ATLAS simulation region. An eigenvalue
solver provides a set of eigenfunctions and corresponding eigenvectors.
LASER uses boundary conditions in the form, , on the boundaries of the solution region. This
region should be large enough to cover the entire active region of the laser diode with some inclusion of
the passive regions. When the single frequency model is used, the lasing frequency used in
Equation 8-1 is an external parameter that is fixed during the calculation and either model can be
then used for optical gain.
The multiple longitudinal mode model requires you to use a physically-based optical gain model.
Specify an initial estimate of the lasing frequency. This is adjusted during calculations. You can also
specify a frequency range or photon energy range, which LASER will then calculate multiple
longitudinal modes.
LASER may also simulate multiple transverse modes.

SILVACO International 8-7

ATLAS User’s Manual

8.4: Specifying Laser Simulation Problems

The structure of the laser diode and the mesh used to simulate it are specified in the normal way using
the capabilities provided by ATLAS and BLAZE. To enable LASER simulation, use the parameters and do
the following:
1. Activate LASER by specifying a LASER statement.
2. Define a mesh for solution of the Helmholtz equation. The Helmholtz Equation is solved on a
rectangular mesh. The rectangular mesh can be independent of the triangular mesh, which is
used for device simulation. This rectangular mesh is specified by the [Link] and [Link]
statements. The area specified by the laser mesh must be completely inside the ATLAS
simulation domain and should completely cover the active lasing region.
3. For single-frequency calculations, you can select different gain models for different regions/
materials using the REGION or MATERIAL parameters of the MODELS statement.

Note: If multiple longitudinal modes are to be accounted for, then specify GAINMOD=1 for the active lasing region.

8.4.1: LASER Statement Parameters

You can simulate multiple transverse modes by setting the NMODE parameter in the LASER statement.
Set it to the desired number of modes you want to simulate.
There are several kinds of parameters that appear in the LASER statement. There are parameters that
deal with the numerical solution to the photon rate equation, parameters that deal with loss
mechanisms, and parameters that deal with specifying energies to look for longitudinal solutions.
Specify laser physical parameters and models and do the following:
1. Specify the [Link] parameter, which is the length of the laser cavity in the z direction.
2. Specify the [Link] or OMEGA parameters for (initial) photon energy or laser frequency.
3. Specify laser loss mechanisms (MIRROR, FCARRIER, ABSORPTION parameters in the LASER
statement) and any associated constants (FCN, FCP, ALPHAA in the MATERIAL statement).
4. Specify any additional laser losses (LOSSES parameter).
If you to calculate the laser spectrum, specify the multiple longitudinal modes model and additional
parameters. Do the following:
1. Specify the LMODES in the statement. This enables the Multiple Mode Model.
2. Specify the EINIT and EFINAL parameters. These parameters set the photon energy range
within, which LASER will take into account multiple longitudinal modes. Make sure that initial
photon energy is within this range and that is specified for the active lasing region.
3. Specify the photon energy separation (ESEP parameter). If this isn’t specified, LASER will
automatically calculate the number of longitudinal modes based on the cavity length and the
energy range. We recommend that you allow LASER to choose the photon energy separation.
4. Specify the spectrum file name ([Link]). LASER will produce a structure file containing
spectrum data after calculation of each bias point. LASER will automatically append to
the specified file name (where is the number of the bias point) for steady-state solutions or
for a transient simulation. The first bias point where LASER is active will have N=1. This
is often not the first bias point during simulation. These files can be examined using TONYPLOT. If
the SPECSAVE parameter is specified, the spectrum files will only be saved on every
[Link] solution.

8-8 SILVACO International

 Laser: Edge Emitter Simulation

Note: The index in the spectrum file name will still increase by one each time. The spectrum data can be stored in a single
file for the transient simulation, only if MULTISAVE - LASER statement is set to FALSE.

You can save near and far field patterns by specifying a value for the PATTERNS parameter in the SAVE
statement. The value of the PATTERNS parameter is a character string representing the root name of a
file for saving the near and far field patterns. The near field pattern is saved to a file with the string
“.nfp” appended to the root and the far field pattern is saved to a file with the string “.ffp”
appended to the root name. These files can be examined using TONYPLOT.
You can also use the numeric parameters. These parameters are: TOLER, ITMAX, SIN, TAUSS, and
MAXCH. The default values of these parameters have been selected to give a good exchange between
accuracy, efficiency, and robustness for most applications. You can adjust the calculation by specifying
different values. These parameters are discussed in the next section.

8.4.2: Numerical Parameters

The following numerical parameters control the performance of the LASER simulation. All of these
parameters have reasonable default values, but you can specify them.
• TOLER sets the desired relative tolerance of the photon density calculation. The default
value is 0.01. Setting this parameter to a lower value may slow down the calculation
significantly. Using a larger value will result in a faster but less accurate calculations.
• ITMAX sets maximum number of external LASER iterations during photon density
calculation. The default value is 30.
• SIN is the initial photon density used only with simple LASER models. LASER starts the
iteration process for photon density calculation from this value. This parameter influences
only the calculation time for the first bias point after the laser threshold is reached.
• TAUSS is an iteration parameter used in the calculation of photon densities. Using a larger
value of this parameter can speed up the calculation but may cause convergence problems.
• MAXCH is the maximum allowed relative change of the photon density between LASER
iterations. Using a larger value of this parameter can speed up the calculation but may
cause convergence problems.

SILVACO International 8-9

ATLAS User’s Manual

8.5: Semiconductor Laser Simulation Techniques

The most common technique for simulating laser diodes is to simulate forward characteristics with a
gradually increasing bias. The forward voltage is usually specified, but current boundary conditions
can be used, and external elements can be included.
To save computational time, we recommend that you don’t enable LASER models at the start of the
simulation. Ramp the device bias first, and then enable Laser simulation by using the LASER
parameter in an additional MODELS statement.
Generally, the single frequency LASER model, which doesn’t take into account the longitudinal mode
spectrum, is faster. This model provides accurate results, and is recommended for use if the lasing
spectrum isn’t the subject of interest. If the multiple longitudinal mode model is used, computational
time will be longer and strongly dependent on the number of longitudinal modes involved in the
calculation.
Although ATLAS is a two-dimensional device simulator, the laser spectrum and all other laser results
are strongly dependent on cavity length, which is effectively the device length of the z direction. For
example, terminal currents can be scaled to apply a different length in the third dimension by using
simple multiplication. This isn’t the case for LASER, since the cavity length determines energy spacing
between longitudinal modes and influences all other laser characteristics.
Be sure to specify the [Link] parameter in the MODELS statement when the longitudinal
mode spectrum is to be calculated.
For symmetric devices, you can improve computational efficiency by simulating half of the device. To
do this, define the device so that the center of the device is located at X=0 and only half of the device is
defined at coordinate values at X>=0.
For device simulation, the default boundary condition is a mirror boundary, which means symmetry is
implied.
For laser simulation, (Helmholtz Equation, see Section 8.2.1: “Helmholtz Equation”) specify that the
laser mesh starts at X=0. Also, specify the REFLECT parameter in the LASER statement. This tells the
Helmholtz Equation that a mirror boundary condition is to be used at X=0.

8.5.1: Generation of Near-Field and Far-Field Patterns

The intensity profile of the laser at the cleaved surface can be examined in the standard structure file
using TONYPLOT. To generate a structure file, use the OUTFILE parameter of either the SOLVE or SAVE
statements. The intensity far from the laser surface, in the Fraunhofer region, is important for
designing the optical coupling to the laser.
The far-field pattern can be generated by LASER by specifying the PATTERNS parameter of the SAVE
statement. The PATTERNS parameter specifies the file name prefix of two output files. The first of
these files, with the suffix, contains the near-field pattern. This is essentially a copy of the near
field pattern contained in the structure file output. The second of these files, with the suffix,
contains the far-field pattern.
The far-field pattern is essentially a 2D fast Fourier transform (FFT) of the near-field pattern. The
sampling for this transform is set by the minimum spacing in the laser mesh and the overall size of the
laser mesh. The output samplings for the near-field and far-field patterns are controlled by the
[Link], [Link], [Link], and [Link] parameters of the LASER statement. These specify the
numbers of samples in the X and Y directions in the near and far field. By default the values are set to
100.

8-10 SILVACO International

 Chapter 9:
VCSEL Simulator

9.1: Overview
VCSEL (Vertical Cavity Surface Emitting Lasers) performs electrical, thermal, and optical simulation
of vertical cavity surface emitting lasers using accurate, robust, and reliable fully numerical methods
and non-uniform meshes. With VCSEL, you can simulate advanced structures containing multiple
quantum wells and oxide apertures including the effects, such as gain guiding, lattice heating, current
crowding, and spatial hole burning. VCSEL works with BLAZE and GIGA and allows you to do the
following:
• Test whether the structure complies to general VCSEL device requirements by simulating cold
cavity reflectivity experiments. Calculate the reflectivity of a VCSEL cavity as a function of the
incident light wavelength in a user-specified range and determine the resonant frequency of the
cold cavity.
• Solve the Helmholtz equation (see Equation 9-1) in cylindrical coordinates to calculate optical
intensities of the multiple transverse modes for index guiding and gain guiding structures.
• Calculate the optical gain in multiple quantum well systems depending on the photon energies,
quasi-Fermi levels, temperature, and optical intensity distribution. Calculate the carrier
recombination due to the spontaneous and stimulated emission.
• Solve the photon rate equations (see Section 9.2.4: “Photon Rate Equations”) for multiple
transverse modes to calculate the photon density in each mode and the total photon density.
• Calculate the light output power and the wavelength for each transverse mode.
• Speed up solution of the Helmholtz equation by using perturbational treatment.
• Simulate more general cavities that support multiple longitudinal modes. This obtains modal
gain spectra and determines the dominant mode with maximum gain.
To perform a VCSEL simulation, you need be familiar with ATLAS and BLAZE first. Please see
Chapter 2: “Getting Started” and Chapter 5: “BLAZE” for more information.

Note: To avoid the confusion in terms, we refer to a VCSEL device as VCSEL and to a VCSEL
simulator as VCSEL:ATLAS in this chapter.

SILVACO International 9-1

ATLAS User’s Manual

9.2: Physical Models

When modeling VCSELs, it is essential to take into account interaction of optical, electrical, and
thermal phenomena that occur during the VCSEL operation. Modeling only optical properties is
already an involved task. Different methods developed recently in order to improve the accuracy of
optical solution produce results that vary considerably from method to method [146]. The variation is
especially strong between vectorial models that tend to produce a spread in results much larger than
simpler scalar models.
On the other hand, the complexity of the vectorial methods makes the self-consistent electro-thermo-
optical simulation impractical from the point of view of the calculation time required. For VCSEL
simulation in VCSEL:ATLAS, we adopted an approach that can account for mutual dependence of
electrical, optical, and thermal phenomena.
Basic semiconductor equations (See Equations 3-1 to 3-4 in Chapter 3: “Physics”) are solved self-
consistently with the Helmholtz Equation, Lattice Heat Flow Equation (See Chapter 6, Equation 6-1),
and the Photon Rate Equation (see Equation 9-12). Since we are using a scalar method, we neglect
polarization effects and reduce Maxwell equations to scalar Helmholtz equation. The method used for
the solution of Helmholtz equation is based on a well developed effective index model, which showed
its validity for a much wider range of problems than originally expected [147]. This method fine tuned
for simulation of various VCSEL structures is often referred to as Effective Frequency Method (EFM).
EFM is a fast and flexible method that allows further development to include dispersion, diffraction
losses, and graded interfaces.
VCSEL:ATLAS uses a cylindrical coordinate system to take advantage of the cylindrical symmetry of
the VCSEL devices. Figure 9-1 shows the relationship between ATLAS’s X-Y coordinate system and
the r-θ-z cylindrical system used in VCSEL:ATLAS. Note that the X coordinate axis is equivalent to
the radius r, and the Y coordinate axis is equivalent to the z axis in the cylindrical system.

9.2.1: Reflectivity Test Simulation

Reflectivity experiment is an important test conducted to evaluate the quality of a VCSEL device. In
this experiment, the light is normally incident on a VCSEL device and its reflectivity is measured as a
function of wavelength. Most manufactured VCSEL resonators are tested this way to ensure adequate
optical performance. Numerical simulation of this experiment enables you to calibrate the material
parameters used in the optical model and eliminate possible specification errors at the early stage of
numerical analysis of a VCSEL device.
Another purpose of the numerical reflectivity test is to ensure that the analyzed device complies with
general VCSEL requirements. These requirements include the following:
1. Presence of a high reflectivity band in the reflectivity spectrum of a VCSEL cavity.
2. The reflectivity over 99% in the high reflectivity band.
3. Presence of a drop in reflectivity within the high reflectivity band. The drop in reflectivity
(increase in absorptance) occurs at the resonant wavelength of the cavity.
VCSEL:ATLAS calculates reflectivity and absorptance using transfer matrix approach (Chapter 10:
“Luminous: Optoelectronic Simulator”, Section 10.2.4: “Matrix Method”). In the simulation, the
incident light propagates along the axis of symmetry of the device.
The [Link] parameter specifies the position of the light origin with respect to the device
(top of bottom). Specific output associated with the reflectivity test includes reflectivity and
absorptance spectra plotted in the corresponding files, reflectivity(absorption)_top.log and
reflectivity(absorption)_bottom.log.
To run the reflectivity test simulation, specify the [Link] parameter and other parameters in
Table 9-1 on a VCSEL statement. Section 9.3.3: "Enabling VCSEL Solution" will discuss parameter
specification in detail.

9-2 SILVACO International

 Laser:VCSEL Simulation

Table 9-1: User-Specifiable Parameters for the Reflectivity Test

Statement Parameter Default Units

VCSEL [Link] False

VCSEL [Link] 1

VCSEL OMEGA s-1

VCSEL EINIT eV

VCSEL EFINAL eV

VCSEL [Link] eV

VCSEL [Link] 1.0

VCSEL [Link] 1.0

VCSEL NSPEC 100

If the structure does not meet requirements 1-3, VCSEL:ATLAS will close with an error. Otherwise,
VCSEL:ATLAS uses Brent's minimization procedure to find the resonant frequency of the cold cavity.
VCSEL:ATLAS uses the obtained value as a reference frequency in the solution of the Helmholtz
equation.

Figure 9-1: The relationship between rectangular and cylindrical coordinate systems used by VCSEL

SILVACO International 9-3

ATLAS User’s Manual

9.2.2: Helmholtz Equation

VCSEL:ATLAS solves the Helmholtz Equation in cylindrical coordinates , , and φ using effective
frequency method [147].

2
2 ω
∇ E ( r, z, ϕ ) + ------- ε ( r, z, ϕ, ω )E ( r, z, ϕ ) = 0 9-1
2
c
where ω is frequency, ε( , , ϕ, ω) is the complex dielectric permittivity, ( , , ϕ) is the optical electric
field, and is the speed of light in vacuum.
Using the expansion around a reference frequency ω :

2 2
ω ω0 ω0
------- ε ( r, z, ϕ, ω ) ≈ ---------- ε ( r, z, ϕ, ω ) + 2 ------- ε ( r, z, ϕ, ω ) ( ω – ω ) 9-2
2 2 0 2 0 0
c c c
we transform the Helmholtz equation into:

2 2 2
∇ + k 0 ε ( r, z, ϕ ) E ( r, z, ϕ ) = ν k 0 ε ( r, z, ϕ )E ( r, z, ϕ ) 9-3

where =ω / and ν is a dimensionless frequency parameter given by:

ω0 – ω λ – λ 0 2 ⋅ Im ( ω )
ν = 2 ----------------- = 2 --------------- – j ----------------------- 9-4
ω0 λ ω0

We assume that the field is separable:

E ( r, z, ϕ ) = f ( r, z, ϕ ) Φ ( r, ϕ ) 9-5

This assumption of separability is the main approximation of the method. The applicability of the
method can be extended beyond this approximation by adding a non-separable component of the field
on the right hand side of Equation 9-5. For dispersive materials, it is possible to make another
modification of the effective frequency method by taking into account material dispersion in Equation
9-2. For separable fields, we obtain the longitudinal wave equation for function :

2
∂ 2 2
-------- + k 0 ε ( r, z, ϕ ) f ( r, z, ϕ ) = v eff ( r, ϕ )k 0 ε ( r, z, ϕ ) f ( r, z, ϕ ) 9-6
2
∂z

at each lateral position ( , φ). Due to the cylindrical symmetry, we can ignore the azimuthal variations.
In this case, we only need to solve Equation 9-6 for each cylindrically symmetric region characterized
by a particular distribution of ε in the vertical direction . Complex Newton method combined with the
characteristic matrix approach (Chapter 10: “Luminous: Optoelectronic Simulator”, Section 10.2.4:
“Matrix Method”) yields eigenvalues ν and eigenfunctions of the longitudinal wave equation.
The transverse wave equation takes the form:
2
1 d  d l 2 2
--- ----- r ----- – ----- + v ( r )k 〈 ε〉
  eff 0 r Φ lm ( r ) = ν lm K 0 〈ε 〉 r Φ lm ( r ) 9-7
r dr dr 2
r

9-4 SILVACO International

 Laser:VCSEL Simulation

where

L
2
〈 ε〉 r = ∫ ε ( r, z )f ( r, z ) dz 9-8
0

VCSEL:ATLAS uses standard eigentechniques to solve an ordinary differential Equation 9-7.

Solutions of this equation are modes.
For each mode, the imaginary part of the eigenvalue determines the overall mode gain(loss) factor,
which is negative below the lasing threshold. The lasing occurs when:

Im ( ν l m ) = 0 9-9
th

The dielectric permittivity is given by:

ε 0 g ( r, z ) ε 0 ( ALPHAA + FCN ⋅ n + FCP ⋅ p )

ε ( r, z ) = ε 0 + ( – ALPHAR + j ) -------------------------
- – j ------------------------------------------------------------------------------------------------------ 9-10
k 0 k 0

where:
• ε is the bulk permittivity.
• ALPHAR is a line width broadening factor.
0 = ω0/c
• is the local optical gain.
• ALPHAA is the bulk absorption loss and is specified in the MATERIAL statement (specify
ABSORPTION in the VCSEL statement to include absorption loss).
• FCN and FCP are the coefficients of the free-carrier loss and are set using the MATERIAL
statement (specify FCARRIER in the VCSEL statement to include this loss mechanism).

Table 9-2: User-Specifiable Parameters for Equation 9-10

Statement Parameter Default Units

MATERIAL ALPHAR 4.0

MATERIAL ALPHAA 0.0 cm-1

MATERIAL EPSINF
MATERIAL FCN 3.0×10-18 cm2
MATERIAL FCP 7.0×10-18 cm2

VCSEL ABSORPTION FALSE

VCSEL FCARRIER FALSE

SILVACO International 9-5

ATLAS User’s Manual

Bulk Permittivity (ε0)

The following rules apply to the specification of the bulk permittivity.
• If located in an MQW and EPSILON is defined in the MQW statement, that value is used.
• If EPSINF is defined in the MATERIAL statement, that value is used.
• If [Link] defined in the MATERIAL statement, then the bulk permittivity is calculated from
the square of the index returned from the C-Interpreter function defined in the file specified by
[Link].
• If [Link] is defined in the MATERIAL statement, the bulk permittivity is calculated as the
square of the index interpolated from the table specified in the file pointed to by [Link].
• If the material is listed in Appendix B: “Material Systems”, Table B-26, the permittivity is
calculated as the square of index interpolated from the built-in tables for that material.
• If PERMITTI is defined in the MATERIAL statement, that value is used. If it’s not defined, the
default material permittivity is used.

9.2.3: Local Optical Gain

For a discussion of gain models, see Chapter 3: “Physics”, Sections 3.9.3: “The Standard Gain Model”
through 3.9.5: “Tayamaya's Gain Model”.

Stimulated Emission
Carrier recombination due to stimulated light emission is modeled as follows:

c 2
R st ( r, z ) = ∑ NEFF
------------------- g m ( r, z ) E m ( r, z ) ⋅ Sm 9-11
m
where is the recombination rate due to stimulated light emission, NEFF is the group effective
refractive index, and is the photon number. The m subscript in this equation and all subsequent
equations refers to a modal quantity. For example, in Equation 9-11 is the photon number for mode
. The NEFF parameter is user-defined but has a default value of 3.57 (see Table 9-3).

Table 9-3: User-Specifiable Parameters for Equation 9-11

Statement Parameter Default Units

VCSEL NEFF 3.57

9.2.4: Photon Rate Equations

Optical gain provides the link between optical and electrical models. The optical gain depends on the
quasi-Fermi levels and in turn impacts dielectric permittivity (see Equation 9-12), and by the coupling
between the stimulated carrier recombination rate ( ) and the density of photons ( ) as described by
Equation 9-11.
To determine , VCSEL solves the system of photon rate equations:

dS m  c 1 c LOSSES
------------ =  ------------------- G m – --------------- – ----------------------------------- S m + R sp 9-12
dt  NEFF τ ph m NEFF  m

9-6 SILVACO International

 Laser:VCSEL Simulation

where the modal gain is given by

2
Gm = ∫ ∫ ∫ gm ( r, z ) ⋅ Em ( r, z ) rdθdrdz 9-13

and the modal spontaneous emission rate R spm is given by

R sp
m
= ∫ ∫ ∫ rsp ( r, z )m rdθdrdz . 9-14

is the normalized optical field.

Table 9-4: User Specifiable Parameters for Equation 9-12

Statement Parameter Default Units

VCSEL LOSSES 0 cm-1

The modal photon lifetime, τ phm , in Equation 9-11 represents the losses in the laser. The losses per
mode are given by

1 c c
------------ = ------------------- ( α a + α fc + α mir ) = ------------------- G m – ω 0 ⋅ ν l m 9-15
τ ph NEFF m m NEFF
m

αa is the bulk absorption loss, α is the free-carrier loss, and α is the mirror loss. These are defined
as:
2
αa =
m ∫ ∫ ∫ ALPHAA ⋅ E m ( r, z ) × rdθdrdz 9-16

2
α fc =
m ∫ ∫ ∫ ( FCN n + FCP p ) ⋅ E m ( r, z ) rdθdrdz 9-17

c
α mir = Gm – α a – α f c – ω 0 ⋅ ν l m ------------------- 9-18
m m NEFF

Spontaneous Recombination Model

For a discussion of spontaneous recombination models, see Chapter 3: “Physics”, Sections 3.9.1: “The
General Radiative Recombination Model” and 3.9.2: “The Default Radiative Recombination Model”.

Optical Power
The optical power emitted is given by Equation 9-19.

hω c α
mir
------------------------------------------------------------------
- 9-19
NEFF ⁄ ( – )⁄( – )

SILVACO International 9-7

ATLAS User’s Manual

9.3: Simulating Vertical Cavity Surface Emitting Lasers

The key to simulating a Vertical Cavity Surface Emitting Laser is the input deck. The input deck
describes the physical device structure, the physical models to be used, and the specific measurements
to be performed. The deck itself can be roughly divided into three sections: structure specification,
model definition, and obtaining solutions. The following sections will describe these concepts in detail.

9.3.1: Specifying the Device Structure

VCSEL devices are the most complicated devices to be addressed by device simulation. This is due to
the many layers involved in making Distributed Bragg Reflectors (See “Specifying Distributed Bragg
Reflectors” section on page 9-10) and the possible Multiple Quantum Well (See “Specifying Quantum
Wells” section on page 9-11) active layers. This complexity is mitigated by the fact that these devices
are mostly epitaxial. Also, many of the layers are periodic. Recognizing these simplifications, we’ve
designed a custom syntax for defining VCSEL devices.
This syntax makes specification of periodic structures (e.g., super-lattices) simple, which completely
avoids problems of aligning various layers to each other. This new syntax also simplifies meshing the
structure.
Despite this simplification, specifying a VCSEL device can still be complex. We recommend that you
look at the standard examples. See Chapter 2: “Getting Started with ATLAS”, Section 2.4: “Accessing
The Examples” for more information about these examples.
The order of statements is important when specifying the device structure. The following order of
statements should be strictly followed.
1. Specify the MESH statement. This initializes the structure definition and allows specification of
cylindrical symmetry.
2. Set up the mesh in the X direction by using the [Link] statements. The mesh in the Y direction
is usually specified along with the device layers in the REGION statements. You can, however,
introduce the Y mesh lines by using the [Link] statements after the [Link] statements.
3. Specify the device regions using REGION and DBR statements. These statements specify the layer
thickness, composition, doping, order of placement, and meshing information in the Y direction.
The order of REGION and DBR statements is up to you and is usually defined by the order the
layers appear through the structure.
4. You can use quantum well models for the gain and spontaneous recombination and then specify
the locations of quantum wells within the device, using the MQW statement. These statements
specify the location, doping and composition of the wells, and certain model information about
the wells.
The wells are centered somewhere within the geometry of the previously specified (using REGION
statements) regions, so that the well barrier layers take on the composition of the underlying
region. The composition of the wells is specified in the MQW statement.
5. Specify the locations of the electrodes using ELECTRODE statements.
6. Use DOPING statements to specify additional doping or composition fraction information. This
usually isn’t necessary since most, if not all doping and composition information can be specified
during the specification of regions.

9-8 SILVACO International

 Laser:VCSEL Simulation

Setting Up Cylindrical Symmetry

VCSELs use cylindrical symmetry to account for the 3D nature of VCSEL devices. To specify
cylindrical symmetry in the MESH statement, use the CYLINDRICAL parameter. An example of
specifying cylindrical symmetry is down below.

Example
MESH CYLINDRICAL
The axis of symmetry for cylindrical is the Y axis at X=0. When using cylindrical symmetry, no X
coordinates should be specified with negative values since X represents radius.
The usual abbreviation rules for parameter names can be used. See Chapter 2: “Getting Started with
ATLAS”, Section 2.5: “The ATLAS Syntax” for more information.

Specifying the Mesh

To specify the mesh in the X direction, use the [Link] statement. Each [Link] statement specifies
the location of one mesh line in the X direction and the spacing between mesh lines at that location.
The location is specified by the LOCATION parameter in microns. The spacing is specified by the
SPACING parameter, which is also in microns. The number of mesh lines specified is limited to built-in
limits in the overall size of the mesh, but there should be at least two [Link] statements. There
shouldn’t be [Link] specifications with negative values in the LOCATION parameter. Such
specification would be ill-defined due to the cylindrical symmetry.
You should also specify mesh lines at any defining edges of regions, electrodes, or other geometrical
aspects in the X direction Also for similar reasons, use the [Link] statements to specify mesh lines in
the Y direction.
Mesh lines can be inserted automatically at defined region edges by using the REGION and DBR
statements. To enable this feature, specify AUTO in the MESH statements. See Chapter 18:
“Statements”, Section 18.26: “MESH” for information about the MESH statement.

Specifying Regions
A region is a volume (in cylindrical coordinates a disk or annulus) that has a uniform material
composition. The region is specified by a REGION statement. In the REGION statement, the material
composition is specified by the MATERIAL parameter.
The MATERIAL parameter can take on values of any material name as described in Appendix B:
“Material Systems”. For terniary and quaterniary materials, you can also specify composition fractions
X and/or Y, using the [Link] and/or [Link] parameters. The DONORS and ACCEPTORS
parameters can be used to specify uniform densities of ionized donors and/or acceptors.
The thickness of the region in the Y direction in microns is specified by the THICKNESS parameter. You
can also use the [Link], [Link], [Link], and [Link] parameters to specify the location and extent of the
region. Except in special cases, we don’t recommend this because of the difficulties aligning the
numerous layers involved.
Unless it’s specified, the region will encompass the entire range of mesh in the X direction as specified
in the [Link] statements. Mesh lines in the Y direction will then be added at the upper and lower
edges of the region. The mesh spacing at the edges is specified in microns by the SY parameter.
The TOP/BOTTOM parameters specifies the relative ordering of regions in the Y direction. Specifying
TOP indicates that the region starts at the top of all previously specified regions (i.e., at the minimum
previously specified Y coordinate) and extends to the thickness in the negative Y direction. Remember
for device simulation, the Y axis conventionally extends down into the wafer for increasing values of Y.
If no previous regions are specified, TOP indicates the region starts at Y=0 and extends to the

SILVACO International 9-9

ATLAS User’s Manual

thickness in the negative Y direction. BOTTOM indicates that the region starts at the bottom of all
previously specified regions (i.e., at the maximum previously specified Y coordinate) and extends to the
thickness in the positive Y direction. If no previous regions are specified, BOTTOM indicates the region
starts at Y=0 and extends to the thickness in the positive Y direction.
Conventionally, we have found it convenient to start the active layers at the Y=0 location (as done in
the standard examples). In ordering REGION statements:
1. Start with the upper DBR by using the TOP parameter.
2. Use the TOP parameter on the top side contact layers.
3. Use the BOTTOM parameter on the active layers.
4. Use the BOTTOM parameter on the lower DBR.
5. Use the BOTTOM parameter on the substrate/lower contact regions.

Specifying Distributed Bragg Reflectors

A Distributed Bragg Reflector (DBR) is a periodic structure composed of regions of two alternating
material compositions. A DBR can be specified by a series of REGION statements alternating between
two different material composition. But DBRs typically consist of many of such layers and specifying
them with REGION statements can be tedious, which could be prone to errors. The DBR statement can
be used to simplify the specification instead. Generally, the DBR statement (alias SUPERLATTICE) can
be used to specify any superlattice composed of layers of two alternating material compositions. For
more information about this statement, see Chapter 18: “Statements”, Section 18.6: “DBR”.
The MAT1 and MAT2 parameters specify the material names of the two materials, which is used the
same way the MATERIAL parameter of the REGION statement is used to specify a single material.
The [Link], [Link], [Link], and [Link] parameters are used to specify the X and Y
composition fractions of the two materials for terniary and quaterniary materials.
The thicknesses of the two layers are specified by the THICK1 and THICK2 parameters of the DBR
statement. The doping for the layers are specified by the NA1 and NA2 parameters for acceptors. The
ND1 and ND2 parameters for donors.
The mesh spacing for the DBR can be specified using the SPA1 and SPA2 parameters, just like the SY
parameter of the REGION statement. You can also use the N1 and N2 parameters to specify the integer
number of the mesh divisions in the introduction of Y direction for each of the layers.
The total number of layers is specified by the integer [Link] parameter. The [Link]
parameter specifies the total number of layers. Thus, an odd values specifies that one more layer of the
material 1 is to be used than of material 2.
The TOP and BOTTOM parameters of the DBR statement are used just like the REGION statement,
except the first material to be added to the top/bottom of the device is always of material 1.
DBR and REGION statements can be intermingled indiscriminately.

Specifying Oxide Apertures

Once all the epitaxial layers of the VCSEL have been specified, you can then specify an oxide aperture
by using a REGION statement with the MATERIAL parameter assigned to an insulator, and the
geometry specified by the [Link], [Link], [Link], and [Link] parameters.

9-10 SILVACO International

 Laser:VCSEL Simulation

Specifying Quantum Wells

The Multiple Quantum Well (MQW) models enable special models for gain and spontaneous
recombination that account for quantum carrier confinement inside of the wells. These models are
described in detail in Chapter 13: “Quantum: Quantum Effect Simulator”, Section 13.6: “Multiple
Quantum Well Model”.
The wells themselves are specified by using the MQW statement. The MQW statement should follow all
the REGION and DBR statements and precede the ELECTRODE statements. For more information about
this statement, see Chapter 18: “Statements”, Section 18.30: “MQW”.
The locations of the wells are specified by the XMIN, XMAX, YMIN, and YMAX parameters of the MQW
statement. These parameters specify a bounding box in units of microns. XMIN and XMAX specify the
extent of the wells in the X direction (i.e., the wells are horizontal and extend across the entire
bounding box in the X direction). The YMIN and YMAX parameters act to define the center of the wells
in the Y direction (i.e., the set of defined wells will be centered in the Y direction at the average
coordinate of YMIN and YMAX. For example, if a single well is defined with YMIN=0 and YMAX=1.0, it
will be centered at Y=0.5. If three wells were defined the middle one would be centered at Y=0.5.
The WW parameter specifies the width of the wells in microns. The WB specifies the width of the barrier
between wells in microns. The number of wells is specified by the NWELL parameter.
The material composition of the wells is specified by the MATERIAL, XCOMP, and YCOMP parameters.
The MATERIAL parameter is assigned one of the valid material names described in Appendix B:
“Material Systems”. The XCOMP and YCOMP parameters are assigned to the X and Y composition
fractions as necessary. The material composition of the barriers between wells is taken from whatever
material regions that may have been defined coincident to the bounding box. This means, define the
barrier material(s) the MQW with the REGION statements.
When MQW is specified mesh lines will be added to the device mesh to resolve the locations of the edges
of the wells and the bounding box. The SY parameter can also be specified to limit the maximum
spacing between mesh lines within the bounding box.
When using MQW, a secondary mesh must be set up for the solution of the Schrodinger’s Equation (See
Chapter 13: “Quantum: Quantum Effect Simulator”, Equation 13-1) in the wells to determine the
bound state energies. This mesh is specified by the NX and NY parameters in the MQW statement.
Typically, NY should be set to some large number so that there will be several samples per well. NX
should be comparable to the number of grid lines in the device mesh over the same extent in the X
direction.
Finally, there are several parameters that describe the details of the physical models. These
parameters are described in detail in Chapter 13: “Quantum: Quantum Effect Simulator”.

Specifying Electrodes
Device electrodes are specified using the ELECTRODE statement. Note that electrodes in device
simulation are treated as boundary conditions. Therefore, there’s no advantage to resolve the full
outline of the electrode. It is usually easiest to simply place electrodes at the top and bottom of the
structure.
This is done by specifying the TOP or BOTTOM parameters of the ELECTRODE statement. The extent
along the top or bottom of the device can be described by the [Link] and/or [Link] parameters. If these
parameters are omitted, the electrode will extend all the way along the corresponding surface of the
device.
Use the NUMBER parameter to assign electrodes to a number. Electrodes should be assigned
consecutive numbers starting at 1. They can also be given a name by using the NAME parameter. They
can also be addressed later by their name or number when modifying the electrode characteristics,
using the CONTACT statement or assigning biasing information in the SOLVE statements.

SILVACO International 9-11

ATLAS User’s Manual

9.3.2: Specifying VCSEL Physical Models and Material Parameters

Much of the simulation results depend on how the physical models are described. For more
information about physical models, see Chapter 3: “Physics”, Section 3.6: “Physical Models”.

Selecting Device Models

For most laser devices including VCSELs, we advise that you at least enable predominant
recombination mechanisms, since these usually define the threshold characteristics and compete with
stimulated emission for carrier recombination. Thus, directly impact the efficiency of the device.
Typically three bulk recombination mechanisms should be addressed: Schockley-Read-Hall (SRH),
Auger, and Radiative Recombination. You can enable these mechanisms by specifying SRH, AUGER, and
OPTR parameters on a MODEL statement.
The parameters of these models are specified on MATERIAL statements. These include TAUN and TAUP,
which specify the SRH lifetimes. AUGN and AUGP, which specify the Auger coefficients. And COPT, which
specifies the radiative recombination rate coefficient.
These models and associated parameters are discussed in more detail in Chapter 3: “Physics”, Section
3.6.3: “Carrier Generation-Recombination Models”.

Specifying Material Parameters

For LASER or VCSEL simulation there are several other material parameters that should be
considered. First and foremost, specify the high frequency dielectric relative permittivity of each
material region. This is specified by the EPSINF parameter in the MATERIAL statement.
The high frequency dielectric enters in the Helmholtz Equation and directly affects which laser mode
may be present. In particular, try to consider the design of the DBRs carefully. Typically, the objective
of the design of the DBRs is to choose the layer thicknesses so that they are one quarter wavelength
thick relative to the local dielectric. During this design process, you need to also carefully consider the
band-gap of the active region, since this affects the energy and wavelength of the maximal gain. A good
design will try to operate near the peak gain, while satisfying the conditions for oscillation between the
DBR mirrors.
Another important consideration for choosing material parameters is the selection of band edge
parameters such as band gap and electron affinity. Although defaults do exist for many material
systems it is almost always better for you to specify the best estimate. Also, consider the mobilities
since they directly affect series resistance.
Chapter 3: “Physics”, Section 3.6.1: “Mobility Modeling” has more information about these parameters.
For more about MATERIAL statements, see Chapter 18: “Statements”, Section 18.24: “MATERIAL”.

9.3.3: Enabling VCSEL Solution

VCSEL Solution Mesh

The solution to Helmholtz equation is performed on a spatially discrete domain or mesh. Use of
matrix method for the solution of the longitudinal wave equation allows the reduction in the number of
mesh points in ( in VCSEL) to the number of material boundaries. This significantly improves
calculation time while maintaining the accuracy. The mesh points in the transverse direction ( in
VCSEL) coincide with the device mesh.

9-12 SILVACO International

 Laser:VCSEL Simulation

Specifying VCSEL Parameters

Specify the VCSEL statement to enable the VCSEL simulator. Once you enable it, the semiconductor
device equations are solved self-consistently with the photon rate equations and the Helmholtz
equation.
To enable reflectivity test simulation do the following:
1. Specify the [Link] parameter in a VCSEL statement.
2. Specify the [Link] parameter to monitor cold cavity reflectivity for light incident on
a structure from either top or bottom of the structure.
• [Link] = 1 - The light incident from the top.
• [Link] = 0 - The light incident from the bottom.
• [Link] = 2 or >2 - Both directions of light incidence are considered. The program
compares cavity resonant frequencies obtained for each direction of incidence. If results do not
agree within a certain tolerance, VCSEL:ATLAS will close with an error.
By default, light is incident from the top of the structure.
3. Specify [Link] for the refractive index of the medium above the structure.
4. Specify [Link] for the refractive index of the medium below the structure. Default
medium above and below the structure is air.
5. Specify the [Link] or OMEGA parameters for (initial) photon energy or frequency.
6. Specify the EINIT and EFINAL paprameters. These parameters set the photon energy range in
reflectivity test. If not specified, they take the following default values: EINIT=0.8
[Link] and EFINAL=1.2 [Link].
7. Specify NSPEC for the numbe2TwPtm-0.0L80024CID 1

SILVACO International 9-13

ATLAS User’s Manual

9.3.5: Alternative VCSEL Simulator

If you need to model a vertical cavity laser that supports multiple longitudinal modes, use the LASER
statement to enable VCSEL solution.
Specify the LASER statement to enable the laser simulator. Once you enable it, the semiconductor
device equations are solved self-consistently with the laser rate equations. The major difference
between this simulator and the one discussed above is in the approach to the solution of the Helmholtz
equation.
The solver used in this case allows multiple longitudinal modes, but it is not as efficient in dealing
with structural variations in the transverse direction.
There are several kinds of parameters that appear in the LASER statement. Most of the parameters
available are discussed above with respect to VCSEL statement.
Specify laser physical parameters and models, and specify the [Link], [Link],
[Link], and [Link] parameters in the LASER statement.
The mirror loss is defined as:
1 1
α mir = ------ ln  ----------- 9-20
2L Rf Rr

where is the cavity length given as:

L=[Link]
and are the front and rear mirror reflectivities and are calculated from the solution of Helmholtz
Equation in the range specified by [Link], [Link], and [Link].
LASER simulates single or multiple longitudinal modes as well as single or multiple transverse modes.
Also with LASER, you can simulate index guiding only or gain guiding. Simulation of gain guiding takes
more computation time but should be considered when spatially varying gain effects are anticipated.
If you to calculate the laser spectrum, specify the multiple longitudinal modes model and additional
parameters. Do the following:
1. Specify the LMODES parameter in the statement. This enables the Multiple Mode Model.
2. Specify the EINIT and EFINAL parameters. These parameters set the photon energy range
within, which LASER will take into account multiple longitudinal modes. Make sure initial
photon energy is within this range and is specified for the active lasing region.
3. Specify the photon energy separation (ESEP parameter). If this isn’t specified, LASER will
automatically calculate the number of longitudinal modes based on the cavity length and the
energy range. We recommend that you allow LASER to choose the photon energy separation.
4. Specify the spectrum file name ([Link]). LASER will produce a structure file containing
spectrum data after calculation of each bias point. LASER will automatically append _dcN.log
to the specified file name (where N is the number of the bias point) for steady-state solutions or
_trN.log for a transient simulation. The first bias point where LASER is active will have N=1.
This is rarely the first bias point during simulation. Use TONYPLOT to examine these files. If you
specify the SPECSAVE parameter, it will only save the spectrum files on every [Link]
solution.
The index in the spectrum file name will still increase by one each time. The spectrum data can be
stored in a single file for the transient simulation, only if the MULTISAVE-LASER statement is set to
FALSE. To enable multiple transverse mode solutions, specify the number of transverse modes by
using the NMODE parameter from the LASER statement.
By default, VCSELs simulate using the Index Guiding Model. To simulate using the gain guiding,
specify [Link]=1.

9-14 SILVACO International

 Laser:VCSEL Simulation

You can save near and far field patterns by specifying a value for the PATTERNS parameter in the SAVE
statement. The value of the PATTERNS parameter is a character string representing the root name of a
file for saving the near and far field patterns. The near field pattern is saved to a file with the string
.ffp appended to the root name. Use TONYPLOT to examine these files.

SILVACO International 9-15

ATLAS User’s Manual

9.4: Semiconductor Laser Simulation Techniques

The most common technique for simulating laser diodes is to simulate forward characteristics with a
gradually increasing bias. The forward voltage is usually specified. You can use, however, current
boundary conditions and include external elements.
To save computational time, we recommend that you use the VCSEL statement. The effective
frequency method employed for the solution of the Helmholtz equation in this case allows you to
consider structures with oxide apertures or other structural variations in transverse direction. We also
recommend that you test the structure using reflectivity experiment simulation prior to applying the
bias.
If you use the multiple longitudinal mode model, computational time will be longer and strongly
dependent on the number of longitudinal modes involved in the calculation.
Simulating multiple transverse modes takes more computation time and should only be preformed
when the higher order modes are of interest.

9-16 SILVACO International

 Chapter 10:
Luminous: Optoelectronic Simulator

10.1: Overview
LUMINOUS is a general purpose ray trace and light absorption program integrated into the ATLAS
framework, which is used to run with device simulation products. When used with S-PISCES or BLAZE,
or device simulators, LUMINOUS will calculate optical intensity profiles within the semiconductor
device, and converts these profiles into photogeneration rates. This unique coupling of tools allows you
to simulate electronic responses to optical signals for a broad range of optical detectors. These devices
include but are not limited to pn and pin photodiodes, avalanche photodiodes, Schottky photodetectors,
MSMs, photoconductors, optical FETs, optical transistors, solar cells, and CCDs.
The following sections address various types of optoelectronic devices. Go to the sections that are most
relevant to your application, but we strongly recommend you that read the other sections as well.

Note: You should be familiar with ATLAS and either S-PISCES or BLAZE before you can use LUMINOUS. If not, read
Chapter 2: “Getting Started with ATLAS” and either Chapter 4: “S-Pisces: Silicon Based 2D Simulator” or Chapter 5: “Blaze:
Compound Material 2D Simulator” first.

SILVACO International 10-1

ATLAS User’s Manual

10.2: Simulation Method

Optoelectronic device simulation is split into two distinct models that are calculated simultaneously at
each DC bias point or transient timestep:
1. Optical ray trace using real component of refractive index to calculate the optical intensity at each
grid point
2. Absorption or photogeneration model using the imaginary component of refractive index to calcu-
late a new carrier concentration at each grid point.
This is followed by an electrical simulation using S-PISCES or BLAZE to calculate terminal currents.

Note: LUMINOUS assumes that the refractive indices, both real and imaginary, are constant within a particular region.

10.2.1: Ray Tracing in 2D

Defining The Incident Beam

An optical beam is modeled as a collimated source using the BEAM statement. The origin of the beam is
defined by parameters [Link] and [Link] (see Figure 10-1). The ANGLE parameter specifies the
direction of propagation of the beam relative to the x-axis. ANGLE=90 is vertical illumination from the
top. [Link]/[Link] parameters specify the illumination window. As shown in Figure 10-
1, the Illumination Window is “clipped” against the device domain so that none of the beam bypasses
the device).
The beam is automatically split into a series of rays such that the sum of the rays covers the entire
width of the illumination window. When the beam is split, ATLAS automatically resolves
discontinuities along the region boundaries of the device.

Figure 10-1: Optical Beam Geometry

Although the automatic algorithm is usually sufficient, you can also split the beam up into a number of
rays using the RAYS parameter. Each ray will then have the same width at the beam origin and the
sum of the rays will cover the illumination window. Even when the RAYS parameter is specified,
ATLAS will automatically split the rays in order to resolve the device geometry.

10-2 SILVACO International

 Luminous: Optoelectronic Simulator

Ray Splitting At Interfaces

Rays are also split at interfaces between regions into a transmitted ray and a reflected ray. Figure 10-
2 illustrates the difference between rays that are split to resolve the geometry and transmitted/
reflected rays split at a region interfaces.

Figure 10-2: Reflected and Transmitted Rays

In Figure 10-2, the incident rays come in from the top left. They intersect a interface between two
material regions with differing refractive indices. Within this interface lies a geometric point where
the normal to the interface changes. This implies that the angles of reflection and transmission will be
different for light incident to the left of the point from light incident on the right. Thus, the incident
rays are split to resolve the interface point. The second level of splitting occurs at the interface itself.
Here, the incident rays are split into reflected and transmitted rays.

10.2.2: Ray Tracing in 3D

In 3D, a simpler algorithm is used for ray tracing because the 2D algorithm takes up more
computational time. Therefore in 3D, the algorithm doesn’t automatically split rays to resolve
topological features of the device.

Note: You must specify enough rays to resolve such features to the desired accuracy. But there’s a trade off between the
computation time and accuracy in specifying the number of rays.

The 3D source geometry is very similar to the geometry shown in Figure 10-1. In 3D, the source origin
is specified by three coordinates: [Link], [Link], and [Link]. Unlike Figure 10-1, there are
two angles that describe the direction of propagation. These are: THETA, which describes the rotation
of the X axis (THETA is an alias of ANGLE), and PHI, which describes the angle of rotation of Z axis.
In 3D, the clipping of the window of propagation is done in two directions. In the X direction, this is
described by XMIN and XMAX (XMIN and XMAX are aliases of [Link] and [Link]). In the
Z direction, the window is described by ZMIN and ZMAX.
The numbers of rays in the X and Z directions are described by NX and NZ. Ray samples are taken at
regular intervals over the source window.
Figure 10-3 shows an example of a 3D ray trace, where TonyPlot3D is used to visualize the rays.

SILVACO International 10-3

ATLAS User’s Manual

Figure 10-3: 3D Ray Trace

10.2.3: Reflection and Transmission

Figure 10-4 shows the relationship between the angles of incidence (θi), reflection (θr), and
transmission (θt) at the interface between two media. These coefficients are calculated as a function of
the refractive indices in the two media.

10-4 SILVACO International

 Luminous: Optoelectronic Simulator

Figure 10-4: Angles of incidence, reflection and transmission

The reflection and transmission coefficients of the light for parallel and perpendicular polarization are
calculated as shown in Equations 10-1 to 10-6.
θ – θ
(parallel polarization) 10-1
θ θ

θ
(parallel polarization) 10-2
θ θ

θ – θ
(perpendicular polarization) 10-3
θ θ

θ
(perpendicular polarization) 10-4
θ θ

 
  10-5
 

 2 θ
  ------------ 10-6
  θi

where is electric field of the incident wave, is the field of the reflected wave, is the field of the
transmitted wave. is the reflection coefficient, is the transmission coefficient, is the refractive
index on the incident side and is the refractive index on the transmission side.
The angles of reflection and transmission are given in Equations 10-7 and 10-8.
θ θ 10-7

SILVACO International 10-5

ATLAS User’s Manual

θ θ 10-8

where θ is the angle of incidence, θ is the angle of transmission and θ is the angle of reflection.

Specifying Reflections
By default no reflections are considered during the ray trace. The REFLECTS=<i> parameter is used to
set an integer number of reflections to consider. Note that setting a very large number of reflections
can lead to extremely long simulation times for the ray trace.
One very convenient way to overcome the long CPU times is to use the [Link] parameter. This
terminates each ray when the optical power falls to the fraction of the original power defined by this
parameter.

Front Reflection
By default, reflection and refraction at the first interface (the initial interface with the device) are
ignored. The first reflection coefficient is zero and the transmission coefficient is one. The polarization
and angle of the transmitted ray at the first interface is identical to the polarization and angle of the
incident beam.
If the [Link] parameter of the BEAM statement is specified, the transmission coefficient is
calculated using Equations 10-1 to 10-6. When the transmission coefficient is calculated, it’s assumed
that the material outside the device domain is a vacuum. The transmitted rays are attenuated by the
transmission coefficient but the reflected ray is not traced.

Back Reflection
By default, the reflection at the back of the device are ignored. No reflected ray is traced once the back
of the device is reached. If the [Link] parameter is specified, the backside reflection coefficient is
calculated (again assuming a vacuum outside the device) and the back-side reflected ray is traced.

Sidewall Reflection
By default, the reflection from the sides of the device are ignored. No reflected ray is traced back into
the structure. As above, [Link] is used to enable the sidewall reflections assuming a vacuum
outside the device.

Discontinuous Regions
You can simulate devices electrically where a single region is defined as two or more separated areas.
The ray tracing algorithm, however, doesn’t support such structures. If a structure has two separate
areas with the same region number, we recommend you using DEVEDIT to renumber the regions,
perhaps even creating a new region number for each area.

Note: This limitation is only for two separated areas with the same region number and not for two regions with different region
numbers of the same material. This latter case can be simulated.

10-6 SILVACO International

 Luminous: Optoelectronic Simulator

Anti-Reflective Coatings
It’s a popular strategy to place anti-reflective (AR) coatings on light detecting devices to improve
device quantum efficiency. Such coatings rely on destructive interference of reflected waves to reduce
overall reflection coefficient of light incident on the detecting device.
Simple AR coatings are typically composed of one layer of a transparent insulating material that is one
quarter optical wavelength thick. Such coatings significantly reduce reflectivity of light at the design
wavelength. Far from the wavelength in question, however, their performance is poor. Currently, more
sophisticated multi-layer AR coatings are used for improved detection of broad-band light.
LUMINOUS enables you to model and design complicated AR coatings optimized for your specific
application. AR coatings are simulated in LUMINOUS using an efficient transfer matrix method. This
method enables you to deal with multi-layer AR coatings with virtually no overhead in computation
time. Section 10.2.4: “Matrix Method” discusses matrix method and its application to wave
propagation in a stratified medium.
Typically, AC coatings affect only optical properties of the device. Therefore, specifying the coating as
a part of the device structure will unnecessarily increase the complexity of the electrical simulation.
In ATLAS coatings are associated with the interfaces of the device instead. This allows you to specify
certain optical properties of any material boundary while not affecting electrical properties of the
structure.
To define optical properties of a coating associated with that interface, use the INTERFACE statement.
Using this statement, you can define an AR coating, a dielectric mirror, or an ideally reflecting surface.
Note that the coating is absent in the structure defined for ATLAS.
The [Link] parameter defines the thickness of the coating layer. The [Link] parameter defines
the refractive index of the layer. For a single-layer coating or for the first layer of a multi-layer coating,
specify the coordinates of the points defining the coated surface P1.X, P1.Y, P2.X, and P2.Y.

Note: These coordinates should define a line, not a rectangular box of thickness ([Link]).

Figure 10-5 shows an example of a single-layer coating. The following shows the syntax used for a
single-layer coating.
INTERFACE [Link]=2.05 [Link]=0.0634 P1.X=0.0 P1.Y=0.0 P2.X=10.0 P2.Y=0.0
This defines a 63.4nm layer of real refractive index 2.05 at Y=0.0 µm in a structure.
You can specify any number of coatings for each device. The COATING parameter in the INTERFACE
statement refers to the number of the coating. If COATING parameter is not set, the first coating will
be used. Coatings should be specified in order (i.e., COATING=3 can not be set before COATING=2 is
defined).
Different coatings should not overlap. If this occurs, the later coating will be used in the overlapping
part.
Typically, the purpose of a single-layer AR coating is to minimize reflectivity of normally incident
monochromatic light at the design wavelength λ.
Equations 10-9 and 10-10 define the parameters of choice for a single-layer AR coating between
materials with refractive indexes and :

· = n1 n 2 10-9

SILVACO International 10-7

ATLAS User’s Manual

λ
AR · THICK = -------------------------------- 10-10
⋅ AR · INDEX

According to Equations 10-9 and 10-10, the coating in the example considered above has optimal
properties for light at 520nm normally incident from air on a silicon detector.
You need several INTERFACE statements to specify a coating composed of multiple layers. Each
statement defines only one layer of a coating. The syntax of the first INTERFACE statement for a
multi-layer coating is identical to the specification of a single-layer coating. Each subsequent layer
must have the number of the coating and the number of the layer specified by COATING and LAYER
parameters. Specify the same COATING parameter for all INTERFACE statements that refer to the
layers of the same AR coating. For example, the following statements specify a two-layer coating:
INTERFACE [Link]=1.5 [Link]=0.06 P1.X=0.0 P1.Y=0.0 P2.X=10.0 P2.Y=0.0
INTERFACE [Link]=2.05 [Link]=0.06 COATING=1 LAYER=2
Here, COATING=1 and LAYER=2 show the second INTERFACE statement describing a second layer of a
two-layer coating number one. Use the LAYER parameter to describe the order of layers in a coating. If
LAYER is not specified, the first (top) layer will be used. Again, you cannot specify LAYER=3 before
LAYER=2. Note that you only need to specify the coordinates of the interface for the first layer of each
coating.
If a coating is made out of an absorbing material, you can use [Link] parameter to take into
account absorption. You can also use totally reflective coatings. The coating will behave as an ideal
reflector if you set the [Link] parameter to a value > 1000.
An alternative way to set the properties of the top interface is to use C-INTERPRETER function with
[Link] specified in the BEAM statement. Using this function, you can specify the reflection
coefficient, angle of transmission, and transmitted polarization as a function of position, wavelength,
angle of incidence, and incident polarization.

Figure 10-5: Single Layer AR Coating Under Normal Incidence

10-8 SILVACO International

 Luminous: Optoelectronic Simulator

Anti-Reflective Coatings in LUMINOUS3D

LUMINOUS3D enables you to model and design multi-layer AR coatings optimized for your specific 3D
application.
As in LUMINOUS, you can define properties of an AR coating using INTERFACE statement in
LUMINOUS3D. The [Link], [Link], [Link], LAYER, and COATING parameters retain their
meaning and functionality in 3D. The only difference in description of an AR coating in 3D pertains to
specification of the coating location. For a single-layer 3D coating or for the first layer of a multi-layer
3D coating, you need to specify coordinates of the bounding box XMIN, XMAX, YMIN, YMAX, ZMIN, and
ZMAX. All interfaces within the bounding box obtain optical properties of the coating. If you don't set
the bounding box coordinates, the default bounding box containing the whole device will be used.
The REGION parameter gives you additional flexibility in the description of an AR coating in 3D.
REGION specifies the number of a region adjacent to the coating. REGION=0 is the defaulf value,
corresponding to the ambient region outside the device boundaries.
When you define a bounding box for an interface, it could contain other interfaces, which you do not
intend to be AR coated. When you use REGION, you can select only those interfaces adjacent to that
region to have AR coating properties.
The following example
INTERFACE [Link]=2.05 [Link]=0.0634 REGION=0
specifies that a 63.4nm layer of real refractive index 2.05 covers the entire outside boundary of a 3D
device. All possible internal boundaries (heterojunctions) are not coated.
In a multi-layer coating, the first layer is assumed to be adjacent to the region specified by REGION.
INTERFACE [Link]=1.5 [Link]=0.06 XMIN=0.0 XMAX=2.0 YMIN=0.0 YMAX=1.0 \
ZMIN=0.0 ZMAX=6.0 REGION=2
INTERFACE [Link]=2.05 [Link]=0.06 COATING=1 LAYER=2
In this example, the first coating layer with refractive index 1.5 is adjacent to the region number 2 on
all boundaries within the specified bounding box. The second layer is next to the first layer at the same
boundaries. Note that you only need to specify the bounding box and REGION only for the first layer of
each coating.

10.2.4: Matrix Method

Matrix method presents a fast and accurate way to simulate electromagnetic wave propagation
through a layered medium. Among a variety of approaches to matrix theory, the most commonly used
are the characteristic matrix and the transfer matrix descriptions.
Using the matrix method is not limited to LUMINOUS only. It also finds its application in VCSEL and
LED modules. ATLAS uses the characteristic matrix approach that relates total tangential components
of the electric and magnetic fields at the multilayer boundaries. More often used transfer matrix,
which relates tangential components of the electric field of left-travelling and right-travelling waves,
has several disadvantages that limit its applicability.
The structure of a multilayer completely determines the characteristic matrix of this multilayer, the
transfer matrix also contains the information about media on both sides of the multilayer. The
characteristic matrix approach is more general and can be expanded to deal with graded interfaces, in
which case, the transfer matrix method is inappropriate to use.

SILVACO International 10-9

ATLAS User’s Manual

Characteristic Matrix
First, we describe the characteristic matrix of a single layer. As mentioned earlier, the matrix relates
tangential components of the electric and magnetic fields at the layer boundaries and

( ) ( d)
= M 10-11
( ) (d)

The matrix itself is

cos ϕ j sin ϕ ⁄ Y
M = 10-12
jY sin ϕ cos ϕ
where

2π
ϕ = ------nd cos Θ 10-13
λ

is the phase shift for the wave propagating through the layer, is the complex refractive index, and Θ
is the angle of wave propagation in the layer (Figure 10-6). is the optical admittance of the layer,
which is for parallel ( ) and perpendicular ( ) polarizations, is given by:

(s) ε0
Y = ----- n cos Θ 10-14
µ0

(p) ε0
Y = ----- n ⁄ cos Θ 10-15
µ0

where ε0 and are permittivity and permeability of free space.

The characteristic matrix of a multilayer is a product of corresponding single layer matrices. If is
the number of layers, then the field at the first ( ) and the last ( ) boundaries are related as
follows:

( ) ( Zm ) cos ϕ i j sin ϕ i ⁄ Y i
= M 1, M 2 …M m , Mi = 10-16
( ) ( Zm ) jY i sin ϕ i cos ϕ i

Reflectivity, Transmissivity, and Absorptance

The detailed derivation of amplitude reflection ( and transmission ( ) coefficients is given in [145].
The resulting expressions are shown in Equations 10-17 and 10-18:

Y 0 M 11 + Y0 Y s M 12 + M 21 – Y s M 22
r = --------------------------------------------------------------------------------- 10-17
Y 0 M 11 + Y 0 Y s M 12 + M 21 + Y s M 22

t = --------------------------------------------------------------------------------- 10-18
Y 0 M 11 + Y 0 Y s M 12 + M 21 + Y s M 22

where and are the characteristic admittances of the media on both sides of the multilayer.
are the elements of the characteristic matrix of the multilayer. Light is incident from the medium with
admittance . The energy coefficients (reflectivity, transmissivity, and absorptance) are given by:

10-10 SILVACO International

 Luminous: Optoelectronic Simulator

10-19

R ( Ys ) 2
T = ----------------- t 10-20
R ( Y0 )

( )
= ( – ) – 10-21
[( )( )∗ ]

These expressions take into account imaginary part of refractive index and are, therefore, more
general than Fresnel formulae (Equations 10-1 through 10-6). LUMINOUS automatically uses these
expressions when appropriate.

Θ
Θ0

Figure 10-6: Refractive index profile of a single-layer coating

SILVACO International 10-11

ATLAS User’s Manual

10.3: Generation of Photocurrent

10.3.1: Light Absorption and Photogeneration
The cumulative effects of the reflection coefficients, transmission coefficients, and the integrated loss
due to absorption over the ray path are saved for each ray. The generation associated with each grid
point can be calculated by integration of the Generation Rate Formula (Equation 10-22) over the area
of intersection between the ray and the polygon associated with the grid point.
∗ λ –α
= η ----------α 10-22

where:
contains the cumulative effects of reflections, transmissions, and loss due to absorption over the
ray path.
ηo is the internal quantum efficiency which represents the number of carrier pairs generated per
photon observed.
is a relative distance for the ray in question.
is Planck’s constant
λ is the wavelength.
is the speed of light.
α is given by Equation 10-23.
π
α = ------ 10-23
λ

where:
α is the absorption coefficient.
λ is the wavelength.
is the imaginary part of the optical index of refraction.

Photogeneration on a Non-uniform Mesh

The photogeneration algorithm used integrates the optical intensity around each node point. This is
done to ensure that the total photogeneration rate isn’t grid sensitive. A uniform photogeneration rate
is defined as a constant value of (photogeneration rate at any node)*(element area around the node).
In TONYPLOT a uniform photogeneration rate may appear to vary across a non-uniform mesh density.

Photogeneration at Contacts
The photogeneration associated with nodes that are also defined as electrodes is a special case. The
electrical boundary conditions require that the carrier concentration at electrode nodes equals the
doping level. This means that photogeneration at nodes that are electrodes must be zero. But just
setting these nodes to zero photogeneration will typically cause an apparent drop in quantum
efficiency.
The photogeneration rate at the contact nodes is calculated as usual. However this photogeneration
rate is applied to the neighboring node inside the semiconductor. This means for a uniform mesh and
photogeneration rate, if the photogeneration rate is 1.0×1017 pairs/cm-3s, then the nodes at the

10-12 SILVACO International

 Luminous: Optoelectronic Simulator

contacts will have zero photogeneration and the next node into the semiconductor will have 2.0x1017
pairs/cm-3s.

User-Defined Arbitrary Photogeneration

An option exists for you to define the photogeneration rate. A C-INTERPRETER function written into a
text file can be supplied to the program using the [Link] parameter of the BEAM statement. For
example, if a file, myoptics.c, was developed using the template C-INTERPRETER functions supplied it
can be referenced by using:
BEAM NUM=1 [Link]=myoptics.c
.
.
SOLVE B1=1.0
SOLVE B1=2.0
The file, myoptics.c, returns a time and position dependent photogeneration rate to the program.
This returned value is multiplied at every node point by the value of B1.
With this option all other parameters of the BEAM statement and all the material refractive indices are
overridden.

SILVACO International 10-13

ATLAS User’s Manual

The available photocurrent can be thought of as a measure of the rate of photo absorption in the device
expressed as a current density. This should be similar but somewhat less than the source
photocurrent. The losses are due to reflection and transmission of light out of the device structure.

λ –α
∑ ∫ α 10-25

Depending how you define it, quantum efficiency can be readily calculated by dividing the current
from one of the device electrodes by either the source photocurrent or the available photocurrent.

10-14 SILVACO International

 Luminous: Optoelectronic Simulator

10.4: Simulating Photodetectors

This section describes techniques simulate photodetectors. This section applies to the simulation of
any of the following devices: p-n and p-i-n photodiodes, avalanche photodiodes, Schottky
photodetectors, CCDs, MSMs, photoconductors, optical FETs and optically triggered power devices.

10.4.1: Defining Optical Sources

Identifying an Optical Beam

You can define up to ten optical sources. Optical sources are described by using the BEAM statement.
All BEAM statements must appear somewhere after the MESH, REGION, DOPING, and ELECTRODE
statements and before any SOLVE statement. The parameters in the BEAM statement describe a single
optical source.
The NUM parameter is used to uniquely identify one optical source. Values between 1 and 10 are valid
for the NUM parameter. Optical sources are subsequently referred to by the value of their NUM
parameter. The power of the optical beam is set by using the B<n> parameter of the SOLVE statement,
where n is the beam number defined by NUM.

Origin Plane of the Beam for 2D Optical Sources

To specify the origin of the optical source, use the [Link] and [Link] parameters. These
parameters describe the origin of the optical beam relative to the device coordinate system. Currently,
it is required that the origin lie outside any device region. The ANGLE parameter specifies the angle of
the direction of propagation of the beam with respect to the device coordinate system. ANGLE=90
specifies vertical (normal) illumination from above.
The width of the optical beam is specified using the [Link] and [Link] parameters. These
parameters specify the limits of the source beam relative to the beam origin. If either of the limits are
omitted, that limit will be clipped to the edge of the device domain.

Note: It’s extremely important that no section of the origin plane of the beam intersects or is inside the simulation grid.
Otherwise, you’ll get incorrect results. This is important to check in cases where the ANGLE isn’t 90 º or 270 º.

Origin Plane of the Beam for 3D Optical Sources

To specify the origin of a 3D source, use the [Link], [Link], and [Link] parameters. These
parameters describe the location of the origin relative to the device coordinate system. The THETA and
PHI parameters describe the direction of propagation relative to the beam origin. The THETA
parameter describes rotation of the X axis about the Z axis. The PHI parameter describes rotation of
the Z axis about the Y axis. As with the 2D for normal illumination, specify a negative [Link] and
THETA=90.
Specify the window of illumination using the XMIN, XMAX, ZMIN, and ZMAX parameters, and specify the
numbers of samples in the X and Z directions using the NX and NZ parameters. These values should be
set to numbers large enough to resolve any salient features of the topology.
To save the rays into a file for visualization, specify the RAYTRACE=<filename> parameter.

SILVACO International 10-15

ATLAS User’s Manual

Reflections
You can also specify whether to ignore the first reflection using the [Link], and the backside and
sidewall reflection using the [Link]. You should activate the backside reflections for devices,
which use a back side reflector to improve collection efficiency. The number of reflections solved is set
by the REFLECTS parameter.
Typically, [Link] should be used if the structure simulated is equivalent to the complete
photodetector geometry as in a discrete device. If the simulation structure is a section of a larger
substrate as in CCD simulation then [Link] shouldn’t be used.
Since reflection and transmission coefficients are used in the ray trace, for arbitrary angles of
incidence, specify the polarization by using the POLARIZATION parameter.
In complex structures, you should limit the ray tracing to trace only those rays with significant optical
power. The [Link] parameter is used to terminate ray traces that drop below [Link]*
(optical source power).

Specifying Lenslets in 3D
In 3D, you can specify one of three lenslets to focus the light into the device. The lenslet has no direct
effect on the device simulation mesh. It is only used in the ray trace. Lenslets are described by user
defined parameters on the BEAM statement.

Spherical Lenslets
The spherical lenslet is described by the intersection of a plane (perpendicular to the Y axis and a
sphere. The Y coordinate of the plane is given by [Link]. The origin of the sphere is described by
LENS.X, LENS.Y, and LENS.Z. The sphere radius is described by [Link].
For ray tracing purposes, the space between the plane and the device surface is uniformly of a single
index of refraction specified by [Link]. The spherical section projecting above the plane in the
negative Y direction is also considered to consist of the same index of refraction. The plane and the
spherical section form the lenslet.

Ellipsoidal Lenslets
To specify an ellipsoidal lenslet, define the axial half lengths using the [Link], [Link] and
[Link] parameters of the BEAM statement. The ellipsoidal lenslet center is specified by the
LENS.X, LENS.Y and LENS.Z parameters. The [Link] parameter defines the location of the lens
plane. The [Link] parameter gives the lens real index of refraction. Equation 10-26 gives the
equation for the lenslet surface above the lens plane.
2 2 2
( x – LENS · X ) ( y – LENS · Y ) ( – LENS · Z )
----------------------------------2 + ----------------------------------
2
+ ---------------------------------
-
2
10-26
X · SEMIAXIS Y · SEMIAXIS Z · SEMIAXIS

Composite Lenslets
You can also specify a composite lenslet. The composite lenslet is composed of a central planar section,
quarter cylinder sections and eigth sphere sections, which in shown in Figure 10-7. The composite
lenslet is defined by the [Link], [Link], [Link], [Link], [Link],
[Link], [Link] and [Link] parameters. Figure 10-7 shows the meanings of these
parameters.

10-16 SILVACO International

 Luminous: Optoelectronic Simulator

Figure 10-7: Composite Lenslet

Specifying Periodicity in the Ray Trace

By default the unspecified boundary conditions for device simulation are mirror or periodic boundary
conditions. For ray tracing, by default, the edges of the device are considered interfaces with a vacuum.
Thus, at the edges of the device by default reflections are calculated. In some cases, it is convenient to
consider the edges of the device to be periodic with respect to ray tracing. This is particularly true
when the beam is not normal to the plane. You should specify PERIODIC in the BEAM statement to
obtain such boundaries for ray tracing.

Defining Luminous beam intensity in 2D

When a beam is defined in LUMINOUS, it is, by default, assumed to have a uniform light intensity
across the width of the beam. The intensity is defined in the SOLVE statement with the B<n>
parameter and is in the units of W/cm2.
The beam intensity distribution can also be defined with a Gaussian profile.

 ( – ) 
( ) –  10-27
π  ( ) 

SILVACO International 10-17

ATLAS User’s Manual

The location of the peak of the Gaussian distribution with respect to the beam coordinates is specified
by the MEAN parameter of the BEAM statement. The standard deviation of the distribution is specified
with the SIGMA parameter.
is the cumulative distance over the beam width where the distribution will be formed. Define the
RAYS parameter to get smooth distribution of intensity profile. This determines the increment of .
Therefore, the RAYS parameter should be set to a large number as the spacing between rays is always
a constant (width of the beam /RAYS).
The resultant distribution has a unity peak value, which is then scaled by the B<n> value specified on
the SOLVE statement. For example, the following BEAM statement will define a beam window 2 µm
wide centered at [Link] and [Link], which has a Gaussian peak of 0.01 W/cm2 and a standard
deviation of 0.05 µm.
beam num=1 [Link]=5.0 [Link]=-1.0 angle=90.0 wavelength=0.6 \
xmin=-1 xmax=1 GAUSSIAN MEAN=0 SIGMA=0.05 RAYS=200 \
SOLVE B1=1E-2

Defining Luminous3D Beam Intensity

A beam defined in LUMINOUS3D is assumed by default to have a uniform light intensity across the
width of the beam. The intensity is defined in the SOLVE statement with the B,n. parameter and is in
the units of W/cm^2.
In LUMINOUS3D, you can also specify a circular or elliptical source. To specify an elliptical source you
should specify the appropriate origin location using [Link], [Link], and [Link]. This will
specify the center of the elliptical source. The major and minor axes of the ellipse are specified by the
XRADIUS and ZRADIUS parameters. Remember you must also specify the NX and NZ parameters. These
will sample uniformly across the diameters of the ellipse.
You may also specify Gaussian profiles in LUMINOUS3D. To do this, specify the location of the beam
origin. You can specify the Gaussian(s) in the X and Z direction. These can be done independently or
together. These Gaussians are specified by the XMEAN, XSIGMA, ZMEAN, and ZSIGMA parameters of the
BEAM statement.

Monochromatic or Multispectral Sources

The optical source can be either monochromatic or multispectral. For monochromatic sources, you can
use the WAVELENGTH parameter to assign the optical wavelength. WAVELENGTH uses the units microns
to be more consistent with the rest of ATLAS. Note this if you’re accustom to the optoelectronic
engineering preference for nanometers.
For multispectral sources, spectral intensity is described in an external ASCII file indicated by the
[Link] parameter. This is a text file that contains a list of pairs defining wavelength and
spectral intensity. The first line of the file gives the integer number of wavelength-intensity pairs in
the file. An example of such a file is shown below.
4
0.4 0.5
0.5 1.0
0.6 1.2
0.7 1.1
This example specifies that there are four samples, and that at a wavelength of 0.4 µm, the intensity is
0.5 Watts per square cm per µm of wavelength, and so on. With multispectral sources, specify a
discretization of the interpolated information. Values must be specified for the [Link],
[Link], and [Link] parameters. These values specify the starting and ending wavelengths
and the number of wavelengths to sample. LUMINOUS uses wavelengths at equal intervals over a
specified range of wavelengths.

10-18 SILVACO International

 Luminous: Optoelectronic Simulator

LUMINOUS performs an independent ray trace at each of the sample wavelengths. For example:
[Link]=0.4 [Link]=0.6 [Link]=2
causes ray traces at wavelengths of 0.45 and 0.55. LUMINOUS obtains the intensity associated with
each sample by integrating the values of the spectral intensity file using a piece wise linear
approximation. Each integral is performed over the range between successive midpoints. In the
preceding example, the integration for the sample at 0.45 would be performed over the range of 0.4 to
0.5.

For a multispectral source, the generation rate (like Equation 10-22) is given by:
WAVEL · END

(λ) λ –
η ∫ α λ 10-28
WAVEL · START

where:
η0 is the internal quantum efficiency.
(λ) is the power spectral density of the source.
is a factor representing the cumulative loss due to reflections, transmissions, and absorption
over the ray path.
λ is the wavelength.
is Planck's constant.
is the speed of light.
α is the absorption coefficient given by Equation 10-23.
is the depth of the device, where , forms the two-dimensional mesh.
[Link] and [Link] are the spectral limits specified on the BEAM statement.

Note: The integral in Equation 10-28 may be inexact due to the discrete sampling in conjunction with wavelength
dependence of the absorption coefficient. For constant absorption coefficient, the integral is exact of the number of discrete
wavelength samples specified by [Link] on the BEAM statement.

The source photocurrent (like Equation 10-24) is given by:

WAVEL · END

∫ ( λ )λ λ 10-29
WAVEL · START

where λ is the power spectral density of the source, and the other parameters have the same
definition as in Equation 10-24.
The available photo current (like Equation 10-25) is given by:

SILVACO International 10-19

ATLAS User’s Manual

WAVEL · END

–
∑ ∫ ∫ (λ) λλ 10-30
WAVEL · START

where λ is the power spectral density of the source, and the other parameters have the same
definition as in Equation 10-25.
For either the monochromatic or multispectral sources, you can uniformly scale the wavelength(s) and
intensities by using the [Link] parameter and the [Link] parameter respectively.
These parameters are useful if the intensities or wavelengths are specified in units other than the
default units.

Note: The units of spectral intensity in the [Link] are W/cm2 per µm of wavelength. The per micron of wavelength
is important to remember when integrating the total power across the spectrum.

10.4.2: Defining Optical Properties of Materials

For ray tracing, the complex index of refraction of the various material regions in the structure must
be specified. For many of the more common semiconductors and insulators, there are built-in tables of
index versus wavelength. You can specify the index for the materials lacking reasonable default
complex index of refraction.

Note: You can add the [Link] parameter to any SOLVE statement to print out the refractive indices being used
for that bias step.

Setting Single Values For The Refractive Index

The [Link] and [Link] parameters of the MATERIAL statement can be used to set the real
and imaginary indices of a specified material, region or regions.
For example, the statement:
MATERIAL MATERIAL=Air [Link]=1.0 [Link]=0.0
would set the index for all material regions composed of “Air”
The statement:
MATERIAL REGION=1 [Link]=1.0 [Link]=0.0
would set the index for region number 1.
The statement:
MATERIAL [Link]=1.0 [Link]=0.0
would set the index for all regions.

10-20 SILVACO International

 Luminous: Optoelectronic Simulator

Setting A Wavelength Dependent Refractive Index

The preceding examples set the complex index of refraction for a material regardless of wavelength.
This is probably adequate for monochromatic sources. For multispectral simulations, the index of
refraction should be modeled as having a dependence on wavelength. There are two ways to do this.

ASCII File Input

The first way is to specify index versus wavelength in a file. This is a text file that contains ordered
triplets of wavelength, real index, and imaginary index. The first entry in the table is the number of
samples. If you set the [Link] parameter of the MATERIAL statement to the name of the index
file, the linear interpolation from this table will complete to obtain the index of refraction as a function
of wavelength.
A valid index file is shown below.
2
0.5 2.0 0.0
0.6 3.0 0.02
In this example, a real index of 2.5, and an imaginary index of 0.015 would be used for a wavelength of
0.55 microns.

C-Interpreter Function
The second way to specify index versus wavelength is by using the C-INTERPRETER function. The
syntax is:
MATERIAL NAME=Silicon [Link]=myindex.c
The file myindex.c is an external file conforming to the template supplied with the program. It
returns wavelength dependent real and imaginary indices. Instructions for using the C-INTERPRETER
and finding the template functions are described in Appendix A: “C-Interpreter Functions”.

10.4.3: Extracting Dark Characteristics

One of the first tasks in analyzing a new detector design is to examine dark current, device
capacitance, and possibly other unilluminated characteristics. This can normally be done without
using LUMINOUS. Extraction of the characteristics is adequately covered in the chapters on S-PISCES or
BLAZE.
The extraction of reverse bias leakage currents for diodes presents some difficult numerical problems
for device simulators. These problems are associated with limitations on numerical precision. ATLAS,
as well as most other available device simulators, use double precision arithmetic to evaluate terminal
currents. Double precision arithmetic provides roughly 16 decimal digits of precision. Internal scaling
allows the measurement of currents down to a level of between about 10-12 A/micron to 10-16 A/micron.
Unfortunately, photodiode leakage currents are often around or below this level. This means that the
currents printed are subject to significant numerical noise, and don’t provide an accurate estimate of
the device leakage currents. There are two ways to estimate reverse leakage current.

Integrated Recombination
From a theoretical standpoint, the reverse behavior of diodes can be dominated by one of two effects:
diffusion currents in the neutral regions or recombination currents inside the depletion region [9].
ATLAS can provide insight into both of these contributing mechanisms. To estimate recombination
current, use the MEASURE statement to calculate the integrated recombination rate. The following
statement can be used:
MEASURE [Link]

SILVACO International 10-21

ATLAS User’s Manual

When this statement is executed, it prints out the total integrated recombination rate. Multiply this
value by the electron charge (1.6023x10-19 coulombs) to obtain an estimate of the recombination
current contribution to the reverse diode leakage current.

Extrapolation from High Temperatures

The diffusion current contribution can be estimated by taking advantage of the non-linear relationship
between the diffusion current and temperature. Referring to the expression for the “Ideal Diode”
current given by Equation 10-31, the dominant temperature dependency arises from the variation of
the intrinsic concentration.

    – 
     10-31
 

where 0 and are thermal-equilibrium minority carrier densities on either side of the junction.
This gives an exponential variation of the diffusion current with temperature as given in the following
equation:

–     – 
≈   ⋅   10-32
   

This relation can be used to estimate the diffusion current contribution at the operating temperature.
The basic idea is to calculate the current at a high temperature where the problem of numerical
precision does not arise, and then scale the current to the operating temperature using Equation 10-
32. For example, if the device is to operate at 300K, the temperature may be set to 450K using the
TEMPERATURE parameter of the MODEL statement. Any temperature dependence of the energy gap
should be disabled by explicitly specifying the band gap using the EG300 parameter and setting
EGALPHA and EGBETA parameters to zero, all on the MATERIAL statement. The following statement
illustrates this approach as it might apply to a silicon diode:
MODEL TEMPERATURE=450
MATERIAL EG300=1.12 EGALPHA=0.0 EGBETA=0.0
ATLAS can then be used to obtain the reverse bias current at the elevated temperature. The following
equation can be applied to obtain the depletion current contribution at the operating temperature:

  – 
     
⋅  –  ⋅   10-33
     – 
   

where Je is the current measured at the elevated temperature, Eg is the bandgap, Te is the elevated
temperature, TL is the operating temperature, V is the operating bias voltage, and J is the current
estimate at the operating temperature. Once you’ve obtained estimates of the recombination and
diffusion contributions, you can obtain the total leakage current by summing the two contributions.

Numerical Solution Parameters

ATLAS uses a cut-off value of carrier concentration below, which solutions are not required to
converge. This limit is set by the [Link]. parameter. See Chapter 17: “Numerical Techniques” for
more details on [Link]. For photodetectors, you often need to reduce [Link] to 105 in order to
resolve carrier concentrations in depleted regions before illumination.

10-22 SILVACO International

 Luminous: Optoelectronic Simulator

10.4.4: Extracting Detection Efficiency

One of the simpler tasks in characterizing a photodetector design is to measure DC detection
efficiency. This will be done typically as a function of bias voltage, optical intensity, or wavelength.
Each of these analyses can be performed using the SOLVE statement. The Bn parameter can be used to
set the optical intensity of the optical sources described in the previous section. The following example
illustrates obtaining a solution with a specified optical intensity:
SOLVE B1=1.0
This specifies that a solution is to be obtained for an optical intensity in the beam numbered “1” of 1.0
Watt/cm2. If this were the first SOLVE statement specified, the ray trace in LUMINOUS would be
initiated. At the start of the solution the optical intensities of each optical source with a positive
intensity is printed. In addition, the available photocurrent and source photocurrent are printed. See
the prior section on Photocurrent for a definition of these two quantities.

Internal and External Quantum Efficiency

The available photocurrent divided by the source photocurrent is a measure of the external quantum
efficiency of the detector. The calculated terminal current can be divided by the source or available
photocurrents is used to evaluate the internal quantum efficiency of the device.

10.4.5: Obtaining Quantum Efficiency versus Bias

The intensities specified in the SOLVE statement apply until another SOLVE statement changes the
intensity of the beam. Sequences of SOLVE statements can be used to vary the optical intensity at
arbitrary intervals. The simple linear ramps of optical intensity can be abbreviated using the
[Link] and NSTEP parameters of the SOLVE statement. The [Link] parameter specifies the size
of the DC step and NSTEP specifies how many steps are desired.
Another option for analyzing DC quantum efficiency is to fix the optical intensity and vary bias
voltages. The bias voltages can be varied in arbitrary discrete steps using several SOLVE statements,
or in a linear ramp using individual SOLVE statements. This is useful for determining the optimum
operating bias of devices such as avalanche detectors and photo transistors.

10.4.6: Obtaining Transient Response to Optical Sources

It is sometimes desirable to examine the time domain response of a detector to time-dependent (e.g.,
ramped or pulsed) optical sources. LUMINOUS provides this capability with the [Link] parameter.
When the [Link] parameter is specified in a SOLVE statement, the optical intensity is changed
linearly from the most recently set intensity to the intensity set in the B parameter. If a particular
source intensity is not set using the corresponding B parameter, its intensity is not varied during the
transient.
The period of the linear ramp is specified by the RAMPTIME parameter. The transient simulation stops
after the time specified by the TSTOP parameter. If the time given by TSTOP is greater than that given
by RAMPTIME, the source intensities remain constant until the time given by TSTOP. For transient
ramps the TSTEP parameter should also be set. TSTEP is typically set to allow several samples within
the RAMPTIME. After the first time step, subsequent time step sizes will be chosen automatically based
on estimates of truncation error. The following is an example of the specification of an optical impulse
transient:
SOLVE B1=1.0 RAMPTIME=1E-9 TSTOP=1E-9 TSTEP=1E-11
SOLVE B1=0.0 RAMPTIME=1E-9 TSTOP=20E-9 TSTEP=1E-11
In this example, a triangular impulse in the intensity of the optical source is simulated. The peak
intensity is 1.0 and the impulse width is 2 ns. The response of the device is simulated for an additional
18 ns. An initial time step of 10 ps is chosen for both parts of the impulse.

SILVACO International 10-23

ATLAS User’s Manual

10.4.7: Obtaining Frequency Response to Optical Sources

You can also simulate small signal response to optical sources. To obtain a solution for small signal
response, set the [Link] parameter in the SOLVE statement. The BEAM parameter must also be
assigned to the specific optical source index for small signal response of that source. A single source
small signal frequency can be specified using the FREQUENCY parameter. You can vary the frequency
within a single SOLVE statement by using the NFSTEP and FSTEP parameters. The NFSTEP indicates
how many frequency steps are to be simulated, while the FSTEP indicates the step size. If the MULT.F
parameter is specified, the start frequency is multiplied by the step size for the specified number of
steps. If not, the step size will be added to the start frequency for the specified number of steps. For
example:
SOLVE [Link] BEAM=1 FREQUENCY=1e6 NFSTEP=5 FSTEP=10 MULT.F
will invoke solutions as optical frequencies at every decade from 1MHz to 100 GHz.
If the small signal parameters are specified in the same SOLVE statement as a DC bias ramp the small
signal response is extracted for each bias voltage for each small signal frequency. This is a useful
strategy for analyzing the frequency response of the device as a function of bias voltage.

10.4.8: Obtaining Spatial Response

To obtain spatial response, move an optical spot along a line segment perpendicular to the direction of
propagation of the source. Each incremental step is equal to the width of the spot. The total distance
over which the source is scanned is defined by the [Link] and [Link] parameters of the
BEAM statement. The number of steps is defined by the RAYS parameter of the BEAM statement. The
spot width is defined by the ratio:
([Link] - [Link])/RAYS
The spot scan is started by the [Link] parameter of the SOLVE statement. This parameter is set
to the beam index of the optical source to be scanned (i.e., the beam defined by the BEAM statement
whose NUMBER parameter is set to the beam index). During the spot scan, ATLAS obtains solutions
and outputs terminal currents, and so on, as well as the relative beam location at each incremental
spot location. This information can be used by TONYPLOT to produce plots of photoresponse as a
function of position.

10.4.9: Obtaining Spectral Response

The spectral response, defined as device current as a function of the wavelength of the optical source
wavelength, can be obtained. The LAMBDA parameter of the SOLVE statement sets the source
wavelength of the beam in microns. Since the wavelength can be set in each SOLVE statement,
successive solutions can be obtained as a function of wavelength. Each time the LAMBDA parameter is
specified, a new ray trace is run for that new wavelength and the electrical solution recalculated.
The following statements could be used to extract terminal currents at a series of discrete
wavelengths:
SOLVE B1=1 LAMBDA=0.2
SOLVE B1=1 LAMBDA=0.3
SOLVE B1=1 LAMBDA=0.4
SOLVE B1=1 LAMBDA=0.5
SOLVE B1=1 LAMBDA=0.6
SOLVE B1=1 LAMBDA=0.7
SOLVE B1=1 LAMBDA=0.8
In this example, the spectral response is obtained for wavelengths from 0.2 microns to 0.8 microns.
When using LAMBDA, the WAVELENGTH parameter of the BEAM statement will overridden, however be
sure to use a monochromatic beam and not a multi-spectral beam for this simulation.

10-24 SILVACO International

 Luminous: Optoelectronic Simulator

Note: The units of LAMBDA and WAVELENGTH in the BEAM statement are in µm.

SILVACO International 10-25

ATLAS User’s Manual

10.5: Simulating Solar Cells

10.5.1: Obtaining Open Circuit Voltage and Short Circuit Current
To obtain VOC and ISS for a solar cell, first define the illumination conditions. This should be done as
discussed above in describing multi-spectral sources. The short circuit current is obtained by defining
the contacts as voltage dependent contacts (default) and obtaining a solution with the device under
zero bias with illumination. This can typically be done as a first step by the following input statement,
for example:
SOLVE B1=1
In this statement, it’s assumed the source, B1, has been already defined and the intensity is expected
in the actual device. When you obtain a solution, the terminal currents represent the short circuit
current.
The open circuit voltage is obtained by defining one or more of the contacts as current controlled. This
is done using the CONTACT statement. For example:
CONTACT NAME=anode CURRENT
defines that electrode number 1 is a current controlled [Link] open circuit voltage is then
obtained by setting the current at this contact to zero, and obtaining a solution. The following
statement illustrates:
SOLVE I1=0.0 B1=1
Once the solution is obtained, the bias associated with I1 is the open circuit voltage.

10-26 SILVACO International

 Luminous: Optoelectronic Simulator

10.6: Beam Propagation Method in 2D

You can simulate most optoelectronic devices using ray tracing based on geometrical optic principles.
But when diffraction or coherent effects are important, ray tracing methods are no longer sufficient.
In this case, you need a method that takes into account the wave nature of light.
The method that does this is called the Beam Propagation Method (BPM). BPMs such as [131] rely on
paraxial appoximation. The BPM in LUMINOUS, however, has been extended to solve a more general
Helmholtz Wave Equation (Equation 10-34).

∂ ( )
∇ ( )– ------------------------ = 10-34
∂
Here, is the electric field of an optical wave, is the complex refractive index of the material, and is
the speed of light in vacuum.
The Helmholtz Wave Equation is consistent with the Rayleigh-Sommerfeld formulation of scalar
diffraction theory [132]. It can also be derived from Maxwell’s equations assuming homogeneous and
isotropic media and neglecting non-linear effects.
These assumptions put certain restrictions on the applicability of BPM. For example, material regions
with gradual change in refractive index or photodetector saturation modeling aren’t supported. But
the framework of the BPM library allows the non-linear effects (e.g., saturation) and treatment of
inhomogeneous media (e.g., graded index regions) to be included in the future.

10.6.1: Using BPM

The BPM in LUMINOUS isn’t designed to completely replace the fast and robust ray tracing algorithm.
In fact, we recommend you start the analysis of your particular application using the ray tracing
method (see Sections 10.2.1: “Ray Tracing in 2D” to 10.2.3: “Reflection and Transmission”).
Only use BPM when you think light diffraction or coherent effects are important. Here are the
following guidelines when to use it:
•The source is monochromatic (no coherent effects are observed for multispectral sources).
•The application is 2D (currently, a 3D BPM isn’t implemented due to calculation time
considerations).
• Spatial distribution of irradiance or photogenerated carrier density is important.
• The structure isn’t periodic in x.
BPM should be useful for:
• Multilayer structures with layer thicknesses comparable to wavelength.
• Narrow incident beams (when transverse beam size is comparable to wavelength).
Currently, BPM doesn’t support either user-specified reflection coefficient models ([Link]) or
photogeneration rate vs. position or time. The [Link] parameter of SOLVE statement (see Chapter
18: “Statements”, Section 18.40: “SOLVE”) and anti-reflection coatings (see “Anti-Reflective Coatings”
on page 10-7) are also not included.
If you need to use BPM, include the BPM parameter in the BEAM statement (see Chapter 18:
“Statements”, Section 18.2: “BEAM”). BEAM parameters related specifically to ray tracing (such as
RAYTRACE, RAYS, THINEST, [Link], and [Link]) are ignored when BPM is specified.
Unlike ray tracing, the incident beam is specified on the top of the device. The origin of the beam is
then assumed to be at the [Link] in the BEAM statement, where [Link] parameter is ignored.
This is done to avoid unnecessary beam propagation in the material surrounding the device.

SILVACO International 10-27

ATLAS User’s Manual

You can specify additional parameters ([Link] and [Link]) in the BEAM statement to set
respective step sizes for the internal BPM grid. If you don’t define these step sizes, the default values
are used. The default step size in both directions is equal to λ/16.0, where λ is the wavelength of light
in vacuum. Don’t use step sizes larger than λ/2.0, especially if coherent effects are of interest. Also, we
don’t recommend extremely small step sizes because it can result in unnecessarily long computation
times. The default values are adequate for most cases.

10.6.2: Light Propagation In A Multiple Region Device

Since Equation 10-34 doesn’t describe light propagation through the region’s boundary, each region
has to be considered separately. Each region’s boundary is composed of straight line segments (edges)
that form a polygon. Light propagating from one of the segments could be reflected back into the
region from other segments of the boundary, while transmitted light enters other regions. Reflected
and transmitted optical fields are calculated using Fresnel formulae described in Equations 10-1 to
10-6 in Section 10.2.3: “Reflection and Transmission”.
Once the field reflected from a certain boundary is known, it then propagates back into the region. The
light then reaches all edges of the polygon. At this point, the field incident upon each boundary is
saved (i.e., added to the total field incident on each boundary). Once all polygon edges are addressed
like this, the algorithm then proceeds to the next region.
When the pass through the device is complete, one reflection of light on each boundary has been
accounted. The REFLECTS parameter specifies the number of passes to be done.
Like in ray tracing, if the REFLECTS parameter is large (e.g., 10), the optical part of the simulation
might take considerable amount of time. To limit the number of propagation routines, use the
[Link] parameter. But unlike in ray tracing, where the default value of this parameter is zero, the
default is [Link]=1e-6. This limit helps avoid unnecessary computations.
Since only propagation through convex polygons can be properly addressed, concave polygons are
automatically split to form two convex regions. If necessary, air polygons are added at the top of the
device structure to fill the empty regions, where the propagation of light also needs to be considered.

10.6.3: Fast Fourier Transform (FFT) Based Beam Propagation Algorithm

Field propagation from a boundary segment through a region is considered in the coordinate system of
that segment. The X axis is along the boundary and Z axis is along the normal (Figure 10-8).

Figure 10-8: Internal Coordinate System

10-28 SILVACO International

 Luminous: Optoelectronic Simulator

We assume the optical field and the device structure are uniform in the Y direction (perpendicular to
the plane of the Figure 10-8). The rectangular grid covering the whole region is chosen according to the
step sizes defined by [Link] (∆Z) and [Link] (∆X) parameters. The Helmholtz Equation
in a 2D cartesian coordinate system (see Equation 10-35) describes field propagation through the
region.

∂ ( ) ∂ ( )
------------------------- + ------------------------- + ( ) = 0 10-35
∂ ∂
Provided that the field on the input plane Z=0 is known, Equation 10-35 allows you to find the field on
any subsequent plane Z=∆Z. The numerical solution of Equation 10-35 is based on the transformation
of input complex field into superposition of plane waves. This superposition represents the direction
cosine spectrum of plane waves, which is obtained by the FFT of the original field (see
Equation 10-36 ).
∞
( ) ∫ ( ) (– ) 10-36
–∞
Each plane wave propagates at a certain angle to the Z axis. The phase accumulated by each plane
wave component before reaching the plane Z=∆Z is given by:

( ∆ ) ( )  2
– ∆  10-37
 

where π /λ, is the complex refractive index. We then can include diffraction and absorption
upon propagation simultaneously. The direction cosine spectrum given by Equation 10-37 at Z=∆Z is
transformed back into the spatial domain by applying an inverse FFT.
∞

( ∆ ) ------
π ∫ ( ∆ ) ( ) 10-38
–∞
The field distribution at Z=∆Z is then found.
These steps are repeated until the field in the whole region is found. The field incident on each
boundary segment and the field at each node within the region are obtained using bi-linear
interpolation.
When the propagation through the same region is considered, the resulting field values at each node
are added to the values saved upon the previous propagation. Since the phase information is retained
in our propagation scheme, adding the field ensures the coherent effects are taken into account.
In ray tracing, the phase information is lost and irradiance values are added upon consecutive runs
through the same region. This approach is only valid for incoherent sources.
The numerical scheme to solve Equation 10-35 requires the rectangular grid in the transverse
direction (X) to cover an area significantly larger than the region of interest.
Zero-padding of the input field distribution is done automatically to specify the input field on the grid.
Super-gaussian filters are used in the spatial domain to suppress possible reflections from numerical
boundary.
Evanescent waves are included in the propagation scheme to avoid discontinuities in the spectral
domain [132].

SILVACO International 10-29

ATLAS User’s Manual

This page is intentionally left blank.

10-30 SILVACO International

 Chapter 11:
LED: Light Emitting Diode Simulator

11.1: Overview
LED is a simulator within the ATLAS framework that provides general capabilities for simulation of
light emitting diode (LED) devices. When used with the BLAZE simulator, LED supports the
simulation of DC and transient electrical and light emitting characteristics of LEDs. When used with
the GIGA simulator, LED permits the self-consistent analysis of thermal effects on both electrical and
optical emission characteristics. You can also use the LED simulator with the MIXEDMODE simulator
to analyze the circuit level performance of LED devices.
The LED simulator supports simulation of both zincblende (e.g., AlGaAs/GaAs, InGaAsP/InP) as well
as wurtzite (e.g., GaN/AlGaN/InGaN) material systems. It can also account polarization effects and
the effects of strain on both emission spectra and piezoelectric polarization.
Extraction of electrical characteristics, such as DC, transient and small signal response, is augmented
by the capability to extract light emitting characteristics, such as emission power versus current,
emission spectra, emission wavelength, efficiency and output coupling (a function of emission angle).
The following sections describe how to specify LED devices, how to select and specify physical models
including most importantly the optical models and how to extract the most pertinent device
characteristics for device performance optimization.

SILVACO International 11-1

ATLAS User’s Manual

11.2: Defining Light Emitting Devices (LEDs)

There are several ways to define an LED device structures in ATLAS/LED. A structure may be defined
using the ATHENA process simulator. ATHENA simulates the "construction" of devices by simulating the
process flow and the physical process steps. Second, you can define the device structure using the
DEVEDIT device builder tool. The DEVEDIT tool allows you to specify the device structure in a simple
visual point and click environment. Finally, you can define device structures using ATLAS framework
commands (statements). For LED devices, the ATLAS command language approach is probably the
most convenient since the device structures are somewhat simple due to their epitaxial nature.
In the following sections, we will concentrate on definition of the device structure using the ATLAS
command syntax. We recommend that you read Chapter 2: “Getting Started with ATLAS” where we
describe general features of the ATLAS simulator, commands and structure definition before you
proceed. Chapter 3: “Physics” also provides insight into the details of the electrical and optical physical
models used by ATLAS/LED. Chapter 5: “Blaze: Compound Material 2D Simulator” describes some
general physics of simulation of compound semiconductors and heterojunctions.
Within the ATLAS command syntax, there are two approaches for defining device structures. The
general purpose structure definition approach (see Chapter 2: “Getting Started with ATLAS”, Section
2.6.3: “Using The Command Language To Define A Structure”) requires more user-specification and
requires that you describe the entire mesh. The auto-mesh approach (see Chapter 2: “Getting Started
with ATLAS”, Section 2.6.4: “Automatic Meshing (Auto-meshing) Using The Command Language”) is
simpler to use, requires less user-specification and is particularly well-suited to specification of
epitaxial (especially multiple-quantum well or super-lattice) devices. It is for these reasons that we
recommend the auto-mesh approach when defining LED devices .
In the following sections, we will focus mainly on this approach. We will point out important
differences with the other methods: ATHENA, DEVEDIT or manual-mesh as these differences may arise.
The first command in any ATLAS input deck is the MESH command. When loading a structure from
ATHENA, DEVEDIT or a previous ATLAS simulation, you should assign the INFILE parameter of the
MESH statement to the name of the file containing the device structure. For example:
MESH INFILE=[Link]
would cause the simulator to load a device structure from the file named [Link] in the current
working directory. When defining the device structure in the ATLAS auto-meshing syntax, the first
statement should contain the AUTO parameter as shown in the following statement:
MESH AUTO
This will invoke the auto meshing capability.

Note: The INFILE parameter is not specified since this would indicate that the device was both being loaded as well as
constructed in the ATLAS syntax and would therefore produce an ambiguity or conflict.

When using auto-meshing, you must only specify the mesh in the X direction using [Link]
statements. The mesh spacing in the Y direction is specified in the REGION statements as described
later. The following example shows how you would specify an X mesh 1 µm long with a uniform mesh
spacing of 0.1 µm.
[Link] LOCATION=0 SPACING=100
[Link] LOCATION=1000 SPACING=100
Using auto-meshing, you should next specify the device regions using REGION statements. The
REGION statement can be used to specify the material types of individual "regions", as characterized
by a spatial extent in the X and Y directions, and their composition, doping, strain and certain other
characteristics describing how the region is to be modeled.

11-2 SILVACO International

 LED Simulation

In the following example, we will describe a simple GaN/InGaN diode structure suitable for LED
device simulation.
REGION MATERIAL=GaN THICKNESS=0.25 ACCEPTOR=1e19 BOTTOM NY=20
REGION MATERIAL=InGaN THICKNESS=0.003 DONOR=2e14 [Link]=0.2 BOTTOM NY=20 LED
REGION MATERIAL=GaN THICKNESS=0.25 DONOR=1e18 BOTTOM NY=20
In this example, we describe a three layer LED. The top layer is composed of GaN (MATERIAL=GaN), it
has a thickness of 0.25 microns (THICKNESS=0.25), and has an acceptor concentration of 1e19
(ACCEPTOR=1e19). The middle layer is composed of InGaN (MATERIAL=InGaN), it has a thickness of
0.003 microns (THICKNESS=0.003) cm-3, and has an In composition fraction of 0.2 ([Link]=0.2), and
has a donor concentration of 2e14 (DONOR=2e14) cm-3. The bottom layer is also composed of GaN
(MATERIAL=GaN), it has a thickness of 0.25 microns (THICKNESS=0.25), and has a donor
concentration of 1e18 (DONOR=1e18) cm-3.
The BOTTOM parameter in each of the REGION statements specifies that each region resides directly
below the preceeding region. NY specifies the minimum number of Y grid lines used to resolve the
region. The LED parameter in the second REGION statement specifies that region will be treated as a
light emitting region during the post-processing stages of the simulation, and certain light emitting
characteristics of that region will be extracted.
The final step in defining the LED using auto-meshing is to specify the electrodes. To specify
electrodes, use the ELECTRODE statement. The following example specifies two electrodes extending all
the way across the top and the bottom of the device.
ELECTRODE NUMBER=1 NAME=anode TOP
ELECTRODE NUMBER=2 NAME=cathode BOTTOM
Here, the NUMBER parameter is used to give the electrodes numerical tags they can be referred to in
subsequent operations. Similarly, the NAME parameter is used for future references (see Chapter 18:
“Statements”, Section 18.40: “SOLVE”).

SILVACO International 11-3

 ATLAS User’s Manual

11.3: Specifying Light Emitting Diode Models

11.3.1: Specifying Polarization and Piezoelectric Effects
In some material systems (such as the GaN/InGaN system), the built-in fields due to polarization and
strain (piezoelectric effect) can play significant roles in the LED emission characteristics. ATLAS
introduces these polarization fields as sheet charges at the top and bottom edges of regions. To include
the effects of polarization on a region, you should specify the POLARIZATION parameter on the
corresponding REGION statement.
This will only include spontaneous polarization. If you want to also include piezoelectric polarization,
you can also specify the [Link] parameter. If you specify [Link], the simulator will
automatically calculate strain from the lattice mismatch and will calculate the piezoelectric
polarization and apply it to the region. You can also specify the value of the STRAIN parameter, which
specifies the axial strain in the region.
With STRAIN and POLARIZATION set, the simulator will apply a piezoelectric polarization calculated
using the strain value assigned by the STRAIN parameter.

11.3.2: Choosing Radiative Models

The next step in simulating LED devices is the selection of appropriate radiative models. For LED
devices, the most relevant radiative model is the model for spontaneous (radiative) recombination. For
radiative recombination, ATLAS/LED offers several options. Chapter 3: “Physics”, Section 3.9:
“Optoelectronic Models” discusses these options in more detail.
First, is the general radiative model (see Chapter 3: “Physics”, Equation 3-245). To enable this
equation, specify OPTR in a MODELS statement associated with the active region(s) of the LED device
(i.e., those regions with the LED parameter specified in the REGION statement). For our example, the
LED discussed is shown in the following statement:
MODELS MATERIAL=InGaN OPTR
will enable the standard radiative recombination model. The disadvantage of using the standard
model is that it provides no information about the spectral content of the LED light emission. In fact
to properly specify emission intensity, you must specify the emission wavelength in the SOLVE
statement for extraction of LED luminous intensity. To specify the emission wavelength for the general
radiative recombination model given by Equation 3-245 in Chapter 3: “Physics”, assign the value of the
wavelength to the [Link] parameter as shown in the following SOLVE statement, where [Link] is in
units of microns.
SOLVE [Link]=0.51
The COPT parameter of the MATERIAL statement assignes the rate constant for the general radiative
recombination model. For example:
MATERIAL COPT=1.25e-10

COPT assigned a value of 1.25×10-10 cm

11-4 SILVACO International

The
 LED Simulation

The two band zincblende model by . [136] accounts for optical transitions between light and
heavy hole valence bands and the conduction band. To activate this model, specify LI in the MODELS
statement. See Chapter 3: “Physics”, Section 3.9.7: “Yan's and Li's Models for Gain and Radiative
Recombination in Zincblende Materials” for a discussion of the Li model.
The three band wurtzite model by Chuang and Chang [149], is based on k.p modeling and accounts for
optical transitions between the conduction band and heavy and light hole and the crystal field split-off
valence bands. See Chapter 3: “Physics”, Section 3.9.8: “Chuang's Three Band Model for Gain and
Radiative Recombination in Wurtzite Materials” for a discussion of the Chuang model. To activate this
model, specify CHUANG in the MODELS statement. For the on-going example, we chose the following
CHUANG model (since InGaN is a wurtzite material).
MODEL MATERIAL=InGaN CHUANG
You should note that any of the YAN, LI or CHUANG model can be used to analyze spectral
characteristics, such as emission wavelength and emission spectra.
You can also choose to analyze LED efficiency degradation due to competing recombination
mechanisms by enabling these other non-radiative mechanisms. Particularly, you can enable
Schockley-Read-Hall or Auger mechanisms. To activate these models, specify SRH and AUGER in the
MODELS statement as follows:
MODELS SRH AUGER
Make sure you do not enable two radiative mechanisms for the same region, such as OPTR and one or
more of the YAN, LI or CHUANG models. If you do, you may over account for the same radiative
recombination. For more information about recombination models, see Chapter 3: “Physics” Section
3.6.3: “Carrier Generation-Recombination Models”.
As mentioned before, you can calibrate the radiative rate constant (using COPT of the MATERIAL
statement) of the basic radiative recombination model (enabled by the OPTR parameter of the MODELS
statement) to the more physical models (YAN, LI or CHUANG). To do this, first prepare a simulation
with the activated physical model. In this simulation, save a structure file around the device operating
conditions. You can then extract the proper value of COPT to use with the OPTR model by dividing the
radiative recombination rate at any point by the product of the electron and hole concentrations.

11.3.3: Using k.p Band Parameter Models in Drift Diffusion

For the Chuang model, the band calculations performed in Equations 3-386, 3-387, 3-388 and 3-389
can be used in the drift diffusion part of the simulation. To use this capability, you should specify the
K.P parameter on the MODELS statement as follows:
MODELS CHUANG K.P

SILVACO International 11-5

ATLAS User’s Manual

11.4: Data Extraction

11.4.1: Extracting Luminous Intensity
When the LED parameter is specified on any REGION statement, the LUMINOUS intensity emitted by
the LED device is automatically calculated and written to the log file. The file that this information is
output to is specified by the OUTFILE parameter of the LOG statement. The following example outputs
the log file to the file [Link].
LOG OUTFILE=[Link]
You can then load [Link] into TONYPLOT for post-processing visualization, where you can plot
characteristics, such as luminous intensity versus current (see Figure 11-1).

Figure 11-1: Example Luminous Intensity Plot

11.4.2: Extracting Emission Spectra

The LED simulator can save multiple spectrum files from the SOLVE statement. To save a spectrum
file, specify the SPECTRUM parameter after each solution. SPECTRUM is assigned to the file name prefix.
Each subsequent file will be written to a file named by SPECTRUM and appended with a version
number starting at 0.
The LED simulator can output spectral intensity as a function of energy and wavelength integrated
over all wells specified as LEDs. To output spectral intensity, specify the SPECTRUM parameter of the
SAVE statement. SPECTRUM is assigned to the name of a file that the LED spectrum is written.

11-6 SILVACO International

 LED Simulation

You must also specify either LMIN and LMAX or EMIN and EMAX. LMIN and LMAX specify the minimum
and maximum values of wavelength in microns to be captured. EMIN and EMAX specify the minimum
and maximum values of energy in eV to be captured. Specify either energy range or wavelength range
but not both. To capture the number of descrete samples, specify the NSAMP parameter of the SAVE
statement.
Figure 11-2 shows an example spectrum file plotted in TONYPLOT.

Figure 11-2: Plot of Spectrum File for LED

11.4.3: Extracting Emission Wavelength

In the LED simulator, you can use the probe to measure wavelength for LED devices. This will extract
the wavelength from the peak of the spectral response over a specified range. To use this capability,
specify WAVELENGTH in the PROBE statement. Use either EMIN and EMAX or LMIN and LMAX in the
PROBE statement to specify the search range. EMIN and EMAX specify the range in terms of energy in
eV, while LMIN and LMAX specify the range in terms of wavelength in microns.

SILVACO International 11-7

ATLAS User’s Manual

11.5: Reverse Ray-Tracing

Reverse ray-tracing is a technique that allows you to obtain optical output characteristics of an active
optoelectronic device based on material properties and device geometry. Ray-tracing is commonly used
for modeling of passive optoelectronic components, such as photo-detectors.
Rays originating at the external light source are traced into the device and are absorbed to form
electron/hole pairs, which are subsequently detected (using the ray-tracing method in LUMINOUS is
described in Chapter 10: “Luminous: Optoelectronic Simulator”, Sections 10.2.1: “Ray Tracing in 2D”
and 10.2.2: “Ray Tracing in 3D”). Unlike direct ray-tracing, the rays in the reverse method originate
inside the active region and are traced until they exit the device. One or multiple origin points (user
specified) for rays are considered, and interference effects for rays originating at the common sourse
point can be taken into account. When you enable interference, you can analyze the spectral selectivity
of the device structure by performing ray-tracing at multiple wavelengths.
You can assess light extraction from the device by using the SAVE statements (after biasing). Five
different sets of parameters for reverse ray-tracing are used in the examples throughout the section to
demonstrate the features available for modeling LEDs. The parameter sets are given in the order of
increasing complexity of simulation. Computation times also vary. It takes approximately 200 times
longer to run the last SAVE line compared to the first one.
Set the ANGPOWER parameter in the SAVE statement to start the reverse ray-tracing algorithm. The
name of the output file containing the angular power density vs. output angle dependence is specified
as a value for ANGPOWER. The information in the angpower file includes the angular power density for
TE- and TM- polarized light, and total angular power density and total flux angular density vs. output
angle.

Note: In TonyPlot, polar charts show the y-axis directed upward (in the opposite direction to the internal coordinate system
used in ATLAS). Therefore, the plots will appear to be flipped around x-axis (top of the structure is at the bottom of the chart).

The RAYPLOT parameter specifies the name of the output file containing the information on each ray
exiting the device. This file is only created when single origin for all rays is assumed. The information
includes ray output angle, relative ray power (TE-, TM-polarization, and total), and initial internal
angle at the origin (only if INTERFERE parameter is not specified). 0° angle corresponds to the rays in
the X axis direction. 90° angle corresponds to the rays in the Y axis direction.
To start ray-tracing from one point of origin, specify the origin’s coordinates for rays X and Y and the
wavelength [Link]. It is important to choose the origin in the active region of the device. Rays are not
traced if the radiative recombination is zero at the ray origin. All remaining parameters outlined
below are optional and their default values are assumed if the parameters are not specified.
REFLECTS specifies a number of reflections to be traced for each ray originating at the point (X,Y)
(similar to REFLECTS parameter in the BEAM statement). The default value (REFLECTS=0) provides for
a quick estimate of the coupling efficiency. To obtain a more accurate result, use REFLECTS>0,
especially if you specify [Link] or [Link]. The choice of this parameter is based on a
compromise between calculation time and accuracy. The maximum allowed value is REFLECTS=10.
Number of reflections set to 3 or 4 is often a good choice. Example 1 produces a simple ray-tracing
analysis of an LED. Figure 11-3 shows the resulting angular distribution of the emitted light
intensity.

Example 1
SAVE ANGPOWER=OPTOEX18ANG_1.LOG RAYPLOT=OPTOEX18RAY_1.LOG X=2.0 Y=1.01
[Link]=0.9 REFLECTS=4

11-8 SILVACO International

 LED Simulation

Figure 11-3: Emitted Light Intensity vs Angle for Example 1

Specify the DIPOLE parameter to turn on a particular angular distribution of the internal radiating
field that corresponds to a preferred in-plane orientation of dipoles. This orientation is often found in
polymer based OLEDs and results in higher optical output coupling. The default setting
(DIPOLE=false) corresponds to isotropic distribution of emitting dipoles and spherically symmetric
radiation pattern.
[Link] specifies that the top surface of the device be treated as an ideal mirror.
[Link] specifies that the bottom surface of the device be treated as an ideal mirror.
SIDE specifies that the rays reaching the sides of the device terminate there and do not contribute to
the total light output. This is often a good assumption for realistic LEDs as these rays tend to be either
absorbed internally or blocked by the casing of the device.
TEMPER is the temperature (needed for using appropriate refractive indexes of the materials). The
default setting of 300 K will be used if TEMPER is not specified.
POLAR specifies polarization of the emitted photons in degrees (linearly polarized light is assumed).
Parallel (TM-mode, POLAR=0.0) and perpendicular (TE-mode, POLAR=90.0) polarizations result in
significantly different output coupling values.
You should use the default value (POLAR=45.0) if there is no preferred direction of polarization of
emitted photons (unpolarized light emission).
[Link] specifies the minimum relative power of a ray (similar to [Link] parameter in the
BEAM statement). The ray is not traced after its power falls below [Link] value. This is useful for
limiting the number of rays traced. The default value is [Link]=1e-4.
NUMRAYS specifies the number of rays starting from the origin. The default is 180. The acceptable
range 36-3600.

SILVACO International 11-9

ATLAS User’s Manual

Example 2 shows how some of the parameters described above are used for modeling of a realistic
LED. The angular distribution of light power in Figure 11-4 exhibits almost a Lambertian pattern for
this simple planar LED geometry. Note that optical coupling coefficient produced in this calculation
reflects a 2D nature of the example (the light origin is not a point but rather an infinite line in z-
direction).
Simultaneously, you can calculate the optical coupling efficiency for an axially symmetric 3D device
(Normally, this is the value to be compared with experimental results). To do this, specify the
COUPLING3D parameter (while SIDE is set). For this calculation, the light source is assumed to be a
point located on the axis of symmetry.

Example 2
SAVE ANGPOWER=OPTOEX18ANG_2.LOG RAYPLOT=OPTOEX18RAY_2.LOG X=2.0 Y=1.01
[Link]=0.9 SIDE [Link] NUMRAYS=360 REFLECTS=4

Figure 11-4: Emitted Light Intensity vs Angle for Example 2

The rays are assumed to be incoherent by default. This is a good approximation if the thickness of the
active layer of the device is on the order of a wavelength/index. For layer thicknesses, smaller than the
wavelength coherent effects might be important. When INTERFERE is set, the rays originating at the
common source point are taken to be 100% coherent. In this case, the phase information upon
propagation is preserved. Phase change upon reflection is also considered. Thus, interference of rays
exiting the device at the same angle is taken into account. In this case, the internal angle information
is not written to the output rayfile.
Example 3 takes into account interference. Figure 11-5 shows the result. You can take absorption of
rays into account by setting the ABSORB parameter. The absorption is assumed to be specified for each
material by the imaginary part of the refractive index. Carrier density dependent absorption and
photon-recycling are not considered at this point.

11-10 SILVACO International

 LED Simulation

Example 3
SAVE ANGPOWER=OPTOEX18ANG_3.LOG RAYPLOT=OPTOEX18RAY_3.LOG X=2.0 Y=1.01
[Link]=0.9 INTERFERE SIDE [Link] REFLECTS=4

Figure 11-5: Emitted Light Intensity vs Angle for Example 3

Although ray-tracing from one point of origin can give a reasonable estimate of optical output coupling
and angular distribution of light power, it is often desirable to consider multiple points within the
active layer of the device to obtain more accurate results. To set multiple origin points, use the
following parameters.
• XMIN, XMAX, YMIN, and YMAX define a rectangular area containing all origin points.
• XNUM and YNUM specify the number of points along x and y axes within the rectangular area. If
XNUM=1, the x-coordinate of all origin points will be set to XMIN so that the points are chosen
along the line X=XMIN. Similarly, you can choose points along the specific y-line.
Ray-tracing from multiple origins is realized by repeating single origin algorithm for each point and by
adding up the normalized angular power density values thus obtained. The luminous power assigned
to each source (origin) is proportional to the radiative recombination at that point. The luminous
power of all sources adds up to the value obtained by integration of radiative recombination over the
entire device. Rays originating at different source points are completely incoherent (even when
INTERFERE is set), which is consistent with the spontaneous character of the radiation produced by an
LED. The rayplot file is not written for multiple origins (even if RAYPLOT parameter is set). Example 4
shows how multiple origin simulation can be done. Figure 11-6 shows angular distribution of light
obtained in this case.

Example 4
SAVE ANGPOWER=OPTOEX18ANG_4.LOG XMIN=1.0 XMAX=3.0 XNUM=3 YMIN=1.01 YMAX=1.09
YNUM=9 INTERFERE SIDE [Link]=0.9 [Link] REFLECTS=4

SILVACO International 11-11

ATLAS User’s Manual

Figure 11-6: Emitted Light Intensity vs Angle for Example 4

You can also take spectral selectivity of optical output coupling for LEDs into account. Reverse ray-
tracing at multiple wavelengths is considered if the SPECTRUM parameter specifies a filename for
spectral selectivity output while the ANGPOWER parameter is set.
EMIN and EMAX, or LMIN and LMAX specify the energy or wavelength range respectively.
NSAMP specifies the number of spectral components to be considered. We suggest that you specify the
SPECTRUM file only when INTERFERE is set. Otherwise, a warning will be issued during ray-tracing.
When ANGPOWER and SPECTRUM are set in the SAVE statement, the resulting optical output coupling is
averaged over the entire energy (wavelength) range from EMIN to EMAX (from LMIN to LMAX). The same
applies to the quantities in the output angular distribution file. The spectrum file only shows how
output coupling changes with wavelength. Currently, the shape of the gain curve is not taken into
account (flat gain).
Example 5 shows how to use multiple spectral components. Figure 11-7 shows that the results
obtained after averaging over spectral components and multiple origin points while taking
interference into account are similar to the single point source model, where interference and multiple
spectral content are ignored (Example 2). This demonstrates the applicability of a simpler model in
this case.

Example 5
SAVE ANGPOWER=OPTOEX18ANG_5.LOG SPECTRUM=OPTOEX18SP_5.LOG XMIN=1.0 XMAX=3.0
XNUM=3 YMIN=1.01 YMAX=1.09 YNUM=5 INTERFERE SIDE LMIN=0.86 LMAX=0.94 NSAMP=15
[Link] REFLECTS=4

11-12 SILVACO International

 LED Simulation

Figure 11-7: Emitted Light Intensity vs Angle for Parameter Set 5

SILVACO International 11-13

ATLAS User’s Manual

This page is intentionally left blank.

11-14 SILVACO International

 Chapter 12:
MixedMode: Mixed Circuit and Device Simulator

12.1: Overview
MIXEDMODE is a circuit simulator that can include elements simulated using device simulation, as
well as compact circuit models. It combines different levels of abstraction to simulate relatively small
circuits where compact models for single devices are not available or sufficiently accurate.
MIXEDMODE also allows you to also do multi-device simulations. MIXEDMODE uses advanced
numerical algorithms that are efficient and robust for DC, transient, small signal AC and small signal
network analysis.
MIXEDMODE is typically used to simulate circuits that contain semiconductor devices for accurate
compact models don’t exist or circuits, where devices that play a critical role must be modeled very
accurately. Applications of MIXEDMODE include: power circuits which may include diodes, power
transistors, IGBTs, and GTOs, optoelectronic circuits, circuits subject to single event upset, thin film
transistor circuits, high-frequency circuits, precision analog circuits, and high performance digital
circuits.
MIXEDMODE circuits can include up to 100 nodes, 300 elements, and up to ten numerical simulated
ATLAS devices. These limits are reasonable for most applications, but they can be increased in custom
versions on request to Silvaco. The circuit elements that are supported include dependent and
independent voltage and current sources, as well as resistors, capacitors, inductors, coupled inductors,
MOSFETs, BJTs, diodes, and switches. Commonly used SPICE compact models are available, and the
SPICE input language is used for circuit specification.
This chapter describes circuit simulation capabilities rather than device simulation capabilities. The
first part of the chapter contains introductory and background information. Then, it describes presents
and explains MIXEDMODE syntax. This is followed by some sample input decks. The final sections
contain a statement reference and a detailed description of the provided electrical compact models for
diodes, BJTs, and MOSFETs.

12.1.1: Background
Circuit simulators such as SPICE [97] solve systems of equations that describe the behavior of
electrical circuits. The devices that are of interest to circuit designers are normally well characterized.
Compact or circuit models are analytic formulae that approximate measured terminal characteristics.
Advanced compact models provide high accuracy with minimum computational complexity. Device
modeling, device characterization and parameter extraction are concerned with the development and
use of accurate and efficient compact models.
Physically based device simulation solves systems of equations that describe the physics of device
operation. This approach provides predictive capabilities, and information about the conditions inside
a device, but it can require significant amounts of CPU time. Information is usually transferred from
device simulation to circuit simulation as follows: Electrical characteristics are calculated using a
physically-based device simulator. These calculated electrical characteristics are then used as input by
a device modeling and parameter extraction package such as UTMOST [98]. The extracted parameters
are used to characterize a compact model used by the circuit simulator.
This approach is adequate for many purposes, but has limitations. It requires that satisfactory
compact models already exist. The use of compact models always introduces some error, and models
that are adequate for digital circuit simulation may be inadequate for other applications. Applications
and devices for which compact modeling is not always satisfactory includes: Precision low power, high
power, high frequency circuit simulation, SOI, IGBT, GTO, TFT, and optoelectronic devices.

SILVACO International 12-1

ATLAS User’s Manual

12.1.2: Advantages of MixedMode Simulation

The limitations of compact models can be overcome by using physically-based device simulation to
predict the behavior of some of the devices contained in a circuit. The rest of the circuit is modeled
using conventional circuit simulation techniques. This approach is referred to as mixed-mode
simulation, since some circuit elements are described by compact models, and some by physically-
based numerical models.
MIXEDMODE simulation provides several worthwhile advantages. No compact model need be specified
for a numerical physically-based device. The approximation errors introduced by compact models can
be avoided particularly for large signal transient performance. You can also examine the internal
device conditions within a numerical physically-based device at any point during the circuit
simulation. But the cost is increased CPU time over SPICE as CPU time is comparable to a device
simulation excluding the external circuit nodes. MIXEDMODE simulation normally uses numerical
simulated devices typically only for critical devices. Non-critical devices are modeled using compact
models.

12-2 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

12.2: Using MixedMode

Input file specification for MIXEDMODE is different in many respects to the rest of ATLAS. But if you’re
familiar with SPICE and ATLAS syntax, you should have little difficulty understanding how these two
syntax styles are joined in MIXEDMODE.
Each input file is split in two parts. The first part is SPICE-like and describes the circuit netlist and
analysis. The second part is ATLAS-like and describes the device simulation model parameters. These
two sections of an input file are separated as described in the next section.
The circuit description includes the circuit topology (called the netlist) and the electrical models and
parameters of the circuit components. The simulation conditions specify the types of analysis to be
performed. These items are described using syntax based on SPICE.
The ATLAS device descriptions provide information about device geometry, doping distribution,
meshes. Device descriptions can be prepared using the built-in ATLAS syntax, the ATHENA process
simulator, or the DEVEDIT structure specification and meshing tool. You can also read-in previously
calculated device solutions. The device data is read-in from standard structure format files.
When a simulation has finished, the following information is available:
• I-V data (voltages in all circuit nodes and currents in all circuit branches).
• Internal distributions of solution variables (such as electron, hole, and potential distributions)
within the numerical devices.
The results of previous runs of MIXEDMODE can be used as initial guesses for future simulations. This
is particularly helpful when multiple simulations must be performed from the same starting point.
The accessing and running of examples for ATLAS are documented in the DECKBUILD chapter of the
VWF INTERACTIVE TOOLS USER’S MANUAL. We recommend that you run at least one MIXEDMODE
example provided on the distribution tape before trying their own simulations.

12.2.1: General Syntax Rules

The SPICE-like part of any MIXEDMODE input file starts with the .BEGIN parameter. The SPICE-like
part of the input file ends with .END. All parameters related to the device simulation models appear
after the .END statement.
The first non-comment statement after initializing ATLAS (go atlas) has to be .BEGIN. The order of
the following netlist and control statements is arbitrary, but the last SPICE-like statement has to be
.END.
Unlike the rest of ATLAS, for SPICE-like statements the exact command has to be used, unique
abbreviations are not accepted. Statements are not case sensitive.
There has to be at least one numerical ATLAS device (“A” device) within the netlist.
Comment characters are # and $, but not *.
All ATLAS statements specifying the parameters for the numerical device simulation have to be
specified after .END
After all ATLAS statements the simulation has to be explicitly terminated (quit, go <simulator>)
These rules do not apply to the SET statement for parameterization of the input file, since it is
interpreted by DECKBUILD only.
EXTRACT statements are also an exception similar to SET. Since MIXEDMODE input files are parsed
completely before execution (see Section 12.2.4: “Recommendations” for more information), extractions
can only be done after completion of the simulation. To extract results from a MIXEDMODE simulation,
EXTRACT should be specified after re-initialization of ATLAS (go atlas).

SILVACO International 12-3

ATLAS User’s Manual

12.2.2: Circuit and Analysis Specification

The SPICE-like MIXEDMODE statements can be divided into three categories:
• Element Statements: These statements define the circuit netlist.
• Simulation Control Statements: These statements specify the analysis to be performed.
• Special Statements: These statements typically related to numerics and output (first
character being a dot “.”).
The specification of the circuit and analysis part has to be bracketed by [Link] and [Link]
statements (i.e., all MIXEDMODE statements [Link] or after .END will be ignored or regarded as
an error). The order is [Link] [Link] is arbitrary.
Netlist Statements
Each device in the circuit is described by an element statement. The element statement contains the
element name, the circuit nodes to which the element is connected and the values of the element
parameters. The first letter of an element name specifies the type of element to be simulated. For
example, a resistor name must begin with the letter, R, and can contain one or more characters. This
means that R1, RSE, ROUT, and R3AC2ZY are all valid resistor names. Some elements, such as diodes
and transistors, must always refer to a model. A set of elements can refer to the same model. For some
elements, such as resistors and capacitors, model referencing is optional. Each element type has its
own set of parameters. For example, a resistor statement can specify a resistance value after the
optional model name. The bipolar transistor statement (Q) can specify an area parameter. All
parameters have corresponding default values. Independent voltage and current sources have
different specifications for transient, DC, and AC phases of simulation. Transient specifications use
the keywords: EXP, PULSE, GAUSS, SFFM, SIN, and TABLE. AC parameters start with the keyword,
AC.
Elements to be simulated numerically are defined as “A” devices (ATLAS devices). At least one ATLAS
device in a circuit is mandatory.
MIXEDMODE supports the use of the following circuit elements:
• Numerically simulated ATLAS devices (“A” devices)
• User-defined two-terminal elements (“B” devices)
• Capacitors (“C” devices)
• Diodes (“D” devices)
• Voltage controlled voltage source (“E” devices)
• Current controlled current source (“F” devices)
• Voltage controlled current source (“G” devices)
• Current controlled voltage source (“H” devices)
• Independent current sources (“I” devices, may be time dependent)
• JFETs (“J” devices)
• Coupled (mutual) inductors (“K” devices)
• Inductors (“L” devices)
• MOSFETs (“M” devices)
• Optical sources (“O” devices)
• Bipolar junction transistors (“Q” devices)
• Resistors (“R” devices, may be time dependent)
• Lossless transmission lines (“T” devices)
• Independent voltage sources (“V” devices, may be time dependent)

12-4 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

• MESFETs (“Z” devices)

The physical models for linear elements (resistors, capacitors, sources, and so on) are described Section
12.4.1: “Circuit Element Statements”. The models for diodes: BJTs, JFETs, MESFETs and MOSFETs
are described in Section 12.4.2: “Control and Analysis Statements”, but more extensive documentation
can be found in the SMARTSPICE/UTMOST MODELING MANUALS Volumes 1, 2, and 3.
A node is a point in the circuit where two or more elements are connected. A node can be described
either in terms of a node name or as a node number. The node names and numbers are arbitrary with
the exception of the ground node. The ground node is set either by specifying “0” as the node number or
using the name GND. All voltages at the nodes are calculated with respect to the voltage at the ground
node.
Example
The netlist for the circuit shown in Figure 12-1 is represented by the following MIXEDMODE input deck
fragment.
# independent voltage source, 0.1V, connected to node 0 (GND) and 1:
V0 1 0 0.1
# 1kOhm resistor, connected to node 1 and 2
R1 1 2 1K
# ATLAS device, connected to node 2 (anode) and 0 (cathode),
# current scaled by 5e7, mesh from file [Link]
ADIO 2=anode 0=cathode WIDTH=5e7 INFILE=[Link]

SILVACO International 12-5

ATLAS User’s Manual

Figure 12-1: Schematic of Primitive Example Circuit

Control Statements
Control statements are used to specify the analysis to be performed in MIXEDMODE. These take the
place of the SOLVE statements in a regular ATLAS input file. At least one of these statements must
appear in each MIXEDMODE input file. These statements are:
• .DC: steady state analysis including loops
• .TRAN: transient analysis
• .AC: small signal AC analysis
• .NET: small signal parameter extraction (e.g., s-parameters)
For more information about these statements, see Section 12.4.2: “Control and Analysis Statements”.

12-6 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

Special statements
Other statements beginning with a dot “.” specify special parameters for the circuit simulation. These
include numerical options, file input and output, and device parameter output. These statements are
listed below.
• compact device models (.MODEL)
• the output files (.LOG,.SAVE)
• initial conditions settings (.NODESET,.IC)
• initial conditions from a file (.LOAD)
• numerics (.NUMERIC,.OPTIONS)
• device parameter output (.PRINT)
• miscellaneous (.OPTIONS)
Full descriptions of each statement and associated parameters are found in Section 12.4.2: “Control
and Analysis Statements”.

12.2.3: Device Simulation Syntax

The second part of a MIXEDMODE command file ([Link]) is used to define physical models, material
parameters, and numerical methods for ATLAS devices referenced in the “A”-element statements. The
following statements may appear in this part of the command file: BEAM, CONTACT, DEFECT, IMPACT,
INTERFACE, INTTRAP, MATERIAL, MOBILITY, METHOD, MODELS, OUTPUT, PROBE, TRAP, and
THERMCONTACT.
You always need to include an indicator to the circuit element name in each device simulation
statement even if there is only one A-device in the circuit. All statements specifying the device
properties and models are just supplemented by the DEVICE=name parameter, where name is the
circuit element in the netlist, and name will always begin with the letter “A”. This makes it possible to
define different material properties and model settings for different devices within the circuit.
We recommended that you specify the REGION parameter referring to only one region in IMPACT,
MATERIAL, and MODELS statements. If the device consists of more than one region, several statements
with the same device parameters and different region parameters are recommended.
For example to specify the bipolar set of models to a device the syntax used might be:
MODEL DEVICE=AGTO REGION=2 BIPOLAR PRINT

12.2.4: Recommendations
Input Parsing
In regular ATLAS (non-MIXEDMODE) simulations, the input is interpreted line by line and each
statement is executed immediately. This is very useful and nicely supported by DECKBUILD for the
interactive development of the input. Circuit simulations, however, require the complete input before
any simulation can be performed. Consequently, the following occur:
• The complete input is read and parsed before any simulation is initiated.
• An explicit termination of a simulation is required (quit).
• All post processing (extraction and plotting) has to be done after re-initializing ATLAS again.
No simulation is started until either a QUIT statement or a GO statement is seen in the input file. Post-
processing can be done by restarting ATLAS.

SILVACO International 12-7

ATLAS User’s Manual

Scale and Suffixes

In the MIXEDMODE part of the input, numerical values of parameters are represented in standard
floating-point notation. The scale suffix may be followed by a unit suffix (e.g., A for Ampere, V for Volt,
and so on). Using a unit suffix can increase the clarity of a command file. The unit suffix is ignored by
the program. The scale suffixes are shown in Table 12-1.

Table 12-1. Scale Suffixes

Factor Name Suffix
10-15 femto- F

10-12 pico- P

10-9 nano- N

10-6 micro- U

10-3 milli- M

103 kilo- K

106 mega- MG

109 giga- G

1012 tera- T

Numerics
MIXEDMODE solves circuit and device equations simultaneously using fully coupled algorithms. This
provides better convergence and requires less CPU time than alternative approaches. The number of
circuit variables is often small in comparison with the number of device variables. In this case the
CPU time required for simulation performed using MIXEDMODE does not increase drastically
compared to the sum of the simulation times required for the individual numerical physically based
devices.
MIXEDMODE uses the Newton algorithm for each bias point during steady-state analysis and for each
time step during transient analysis. Different variants of the Newton algorithm are used depending on
the circumstances [95, 96]. The full Newton method [.OPTIONS FULLN] and a modified two-level
Newton method [.OPTIONS M2LN] are available for steady-state simulation. The full Newton method
provides rapid convergence when a good initial guess is available. The modified two-level Newton
algorithm is less sensitive to the initial guess. For transient simulation a good initial guess always
exists. The full Newton method therefore works very well, and is therefore always used for transient
simulation.
When using MIXEDMODE3D, it is recommended that you specify the DIRECT or GMRES solver in the
ATLAS part of the MIXEDMODE input deck on the METHOD statement. Also, use the NOPROJ parameter
in [Link] statement in the MIXEDMODE of the input deck.
Multi-Device Structure Representation
If more than one ATLAS device is defined in a MIXEDMODE simulation, the structures are merged
together internally. The output solution file is a single file which contains both structures. The first
structure referenced will be on top, all other structures will be attached below.
Example
A diode and a bipolar transistor are specified as numerical devices with the following element
statements:
ABJT 1=BASE 2=EMITTER 4=COLLECTOR WIDTH=1E4 INFILE=[Link]

12-8 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

ADIO 3=ANODE 4=CATHODE WIDTH=1.5E5 INFILE=[Link]

After outputting the solution with:
.SAVE MASTER=mas
The solution file for the first DC-point, mas_dc_1, contains both structures with the second ATLAS
device (diode) shifted downwards (see Figure 12-2).
This coordinate shift has to be accounted for eventually when extracting position dependent solution
quantities or when defining spatially dependent properties with the C-INTERPRETER (See Appendix A:
“C-Interpreter Functions” for more information).

Figure 12-2: Display of a MixedMode solution with two Numerical Devices

SILVACO International 12-9

ATLAS User’s Manual

Extraction of Results
By default, EXTRACT reads its data from the currently opened log-file when executed along with
ATLAS. Since the extraction of MIXEDMODE log-files require a re-initialization of ATLAS (see “Input
Parsing” section on page 12-7), EXTRACT has to be initialized explicitly with the correct name of the
MIXEDMODE log file. To extract voltages at specific nodes, use the syntax [Link].”circuit
node”. To extract circuit elements, use [Link].”circuit element”.
Example

Specify the log file:

.LOG OUTFILE=hallo
Subsequent extraction from the transient log-file is done with:
go atlas
extract init inf=”hallo_tr.log”
extract name=”t0” [Link] from curve(time,[Link].”Adio_anode”) \
where [Link]=0
It extracts the time, t0, when the transient of the current from the anode electrode of the adio device
in the circuit crosses zero. For more details for the EXTRACT syntax refer to the VWF INTERACTIVE
TOOLS USER’S MANUAL.
Using MixedMode inside the VWF Automation Tools
Like all other Silvaco products, MIXEDMODE is fully integrated into the VWF framework and can be
used for automated experiments. There are, however, some factors to take into account.
One factor is that the auto-interface feature doesn’t work with MIXEDMODE. All structures have to
be explicitly saved in unique files previous to the MIXEDMODE runs and referred to in the A- element
statements. Another factor is that splits within MIXEDMODE runs are not possible. To overcome this
problem, use the SET statement to define a variable in a process simulator or in the “dummy” internal
run. This variable is used to parameterize the input file.
Example

Capacitance as an independent split variable in a VWF experiment.

go internal
# define the independent split variable in a re-entrant simulator:
set cap=5e-9

go atlas

.BEGIN
# use the variable as parameter in MixedMode:
C1 2 3 $cap
Another factor is that the automation tools only store files opened by the normal ATLAS LOG
statement in the VWF database, but ignore those defined [Link]. To overcome this, re-initialize
ATLAS and open the relevant log file of the previous MIXEDMODE run [Link] as the append option
(so that the file is not reset).

12-10 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

Example

MIXEDMODE log-file definition.

.LOG OUTFILE=hallo
Re-opening the second DC-log file and the transient log file to get them stored in the VWF database.
go atlas

log outfile=hallo_dc_2.log append

log outfile=hallo_tr.log append
Initial Settings
Initial convergence is critically dependent on the initial settings of the node voltages
(i.e,.IC [Link]). There should not be any problem starting from the zero bias case. Just like
starting from a preceding MIXEDMODE solution is simple, since the complete solution of the circuit and
the ATLAS devices is directly available (i.e,.LOAD [Link]). But when loading solutions for the
numerical devices from ATLAS, using .OPTIONS LOADSOLUTIONS, sometimes precise matching of the
initial circuit condition is required. In this case it is practical to extract the relevant properties in the
preceding ATLAS run and use them to parameterize the MIXEDMODE input.
In the following example, the voltages and current of an ATLAS solution is extracted, and the results
are used for the initial definition of the circuit.
The end of the first part is the stand-alone ATLAS simulation:
# extract the final voltage drop on the anode:
extract name="Von" max(vint."anode")
# extract the gate current:
extract name="I_gate" [Link] from curve(vint."anode",i."gate") \
where [Link] = $"Von"
extract name="V_gate" [Link] from curve(vint."anode",vint."gate") \
where [Link] = $"Von"

# now the MIXEDMODE part

go atlas
.BEGIN
# define the gate current source, use extracted value as parameter
I1 0 7 $"I_gate"
#
# use extracted gate bias and other expressions to calculate
# the node settings:
set Rgl = 10.5
set v7= $V_gate + $I_gate * $Rgl
.NODESET V(1)=2000 V(2)=$"Von" V(3)=$"V_gate" V(4)=$"V_gate" V(5)=-25 \
V(6)=-15 V(7)=$"v7"

SILVACO International 12-11

ATLAS User’s Manual

12.3: A Sample Command File

A sample MIXEDMODE command file is shown below. This file is used to simulate the reverse recovery
of a power diode. Several MIXEDMODE examples are provided with the product which can be accessed
using DECKBUILD.
1. go atlas
2. .BEGIN
3. V1 1 0 1000.
4. R1 1 2 1m
5. L1 2 3 2nH
6. R2 4 0 1MG EXP 1MG 1E-3 0. 20NS 10 200
7. IL 0 4 300
8. ADIODE 3=cathode 4=anode WIDTH=5.E7 INFILE=[Link]
9. .NUMERIC LTE=0.3 TOLTR=1.E-5 VCHANGE=10.
10. .OPTIONS PRINT RELPOT WRITE=10
11. $
12. .LOAD INFILE=pdsave
13. .LOG OUTFILE=pd
14. .SAVE MASTER=pd
15. $
16. .TRAN 0.1NS 2US
17. $
18. .END
19. $
20. MODELS DEVICE=ADIODE REG=1 CONMOB FLDMOB CONSRH AUGER BGN
21. MATERIAL DEVICE=ADIODE REG=1 TAUN0=5E-6 TAUP=2E-6
22. IMPACT DEVICE=ADIODE REG=1 SELB
23. $
24. METHOD [Link]=1.E8 DVMAX=1.E6
25. $
26. go atlas
27. tonyplot pd_tr.log
Description
Line 1: All ATLAS input files should begin with go atlas
Line 2: [Link] [Link] statements indicate the beginning and end of the circuit simulation
syntax. These commands are similar to those used in SPICE.
Lines 3-7: Circuit components, topology, and analysis are defined within. Generally, the circuit
component definition consists of three parts: the type of component, the lead or terminal mode
assignments, and the component value or model name. For example, if the first component definition
in this simulation is a DC voltage source, then V1 defines the component as voltage source number
one, 1 and 0 are the two circuit modes for this component, and 1000 indicates that the voltage source
value is 1000 volts. The remaining circuit components are resistors (R1, R2) inductor (L1) and
independent current source (IL).
The reverse recovery of the diode is simulated by dropping the value of output resistor R2 over a small
increment of time. The R2 statement contains additional syntax to perform this task. Here, the resistor
is treated as a source whose resistance decreases exponentially from 1 mOhm to 1 mOhm over the
specified time step. This action essentially shorts out the parallel current source IL which is also
connected to the base of the diode.

12-12 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

Line 8: The ADIODE statement specifies a device to be analyzed by ATLAS. The A part of the ADIODE
command specifies that this is a device statement. The DIODE portion simply defines the device name.
The option INFILE= indicates which device structure file is to be used.
Lines 9-10: These set numerical options for the circuit simulation. WRITE=10 specifies that every
tenth timestep will be saved into the solution file specified on [Link] statement.
Line 12: Specifies a file generated by a previous MIXEDMODE simulation to be used as an initial guess
to the voltage.
Line 13-14: Specifies the output log and solution filenames. These names are root names and
extensions will be added automatically by the program.
Line 16: Indicates the type of analysis required. In th

SILVACO International 12-13

ATLAS User’s Manual

12.4: MixedMode Syntax

This section is split into two parts. The first part describes circuit element statements, which describe
the netlist. The second part describe the control and analysis statements.

12.4.1: Circuit Element Statements

A – ATLAS device to be simulated using device simulation
Syntax

Axxx n1=name1 n2=name2 [n3=name3 ...] infile=filename [width=val]

Description
This statement defines a device to be represented by a numerical ATLAS model. The device
description with all necessary information (geometry, mesh, doping, models, electrode names, and so
on) must be available in a standard structure file prior to starting a MIXEDMODE simulation.
Axxx: Name of the element. It must begin with A.
n1: Circuit node to which the ATLAS device electrode with the name, name1, is connected. The ATLAS
device must have at least two electrodes. The maximum number of electrodes allowed in ATLAS is 55.
This means that up to 25 ATLAS devices can be specified (For example, 25 devices with 2 electrodes
each, or 10 devices with 5 electrodes each can be specified). The ATLAS device models should be used
sparingly, because it can be very time consuming. Use circuit models for less important circuit
components to conserve CPU time.
infile: Name of standard structure format file with device geometry, mesh, doping, electrodes names,
and so on. The number of electrodes and their names should match those mentioned in this statement.
Optionally, this file can contain a solution, which MIXEDMODE will use as an initial guess (see
[Link] statement for more details).
width: Device width. This is an optional parameter (default=1). All currents through ATLAS device
terminals calculated using the 2-D ATLAS model will be multiplied by this parameter to account for
the third dimension of the device. width can still be used as a multiplier to the ATLAS current if a 3D
ATLAS structure is used in MIXEDMODE3D.
Example

ABJT1 3=EMITTER 4=BASE 6=COLLECTOR INFILE=[Link] WIDTH=10

Note: Optional parameters for a statement are shown with square brackets (e.g., [n3=name3]).

12-14 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

B – User-defined two-terminal element

Syntax

Bxxx n+ n- INFILE=file_name FUNCTION=function_name

Description
Bxxx: User-defined two terminal element name. It must begin with B.
n+, n-: Positive and negative terminal nodes.
INFILE: Name of the text file ( ) that contains C source code for a user-defined function that
describes element behavior. This file can contain more than one function description.
function_name: Name of the function (“ ) from the file.
Example

B1 2 3 infile=ud.c function=rc

Note: For more information on User-defined Two-Terminal Elements, see the “User-Defined Model” section on page 12-36.

C – Capacitor
Syntax
Cxxx n+ n- value
Description
Cxxx: Name of a capacitor element. It must begin with “C”.
n+, n-: Positive and negative terminal nodes.
value: Capacitance in farads.
Example

Cload 3 0 1pF
D – Diode
Syntax

Dxxx n+ n- mname [area] [L=val] [W=val] [PJ=val] [WP=val] [LP=val]

[WM=val]
[LM=val] [OFF] [IC=val] [M=val] [TEMP=val] [DTEMP=val]
Description
Dxxx: Name of the diode element. It must begin with D.
n+, n-: Positive (anode) and negative (cathode) terminal nodes.
mname: Diode model name. It must refer to a diode model.
area: Area factor. The default is 1.0.
L: Length of the diode in meters. Used for LEVEL 3 diode model only.
W: Width of the diode in meters. Used for LEVEL 3 diode model only.
PJ: Periphery of the diode junction. Calculated from W and L if they are specified (in the LEVEL 3
diode model). The ISW and CJSW model parameters are affected by the value of PJ.

SILVACO International 12-15

ATLAS User’s Manual

WP: Width of the polysilicon capacitor in meters. Used for LEVEL 3 diode model only. Default is 0m.
LP: Length of the polysilicon capacitor in meters. Used for LEVEL 3 diode model only. Default is 0m.
WM: Width of the metal capacitor in meters. Used for LEVEL 3 diode model only. Default is 0m.
LM: Length of the metal capacitor in meters. Used for LEVEL 3 diode model only. Default is 0m.
OFF: Sets ON/OFF startup condition for DC analysis. Default is ON.
IC: Initial voltage across the diode.
M: Multiplier used to describe multiple parallel diodes.

TEMP: Device operating temperature (oC).

DTEMP: Difference (in oC) between the device operating temperature and the circuit temperature.
Default value is 0.
Example
D1 2 3 dmodel1
Dclmp 3 7 Diol 3.0 IC=0.3

Note: See the SMARTSPICE/UTMOST MODELING MANUAL VOLUME 2 for a complete description of the diode
models.

E – Linear voltage controlled source

Syntax

Exxx n+ n- nc+ nc- gain

Description

Exxx: Name of the linear voltage controlled voltage source. It must begin with E.
n+, n-: Positive and negative terminal nodes. A positive current flows from the node, n+, through the
source to the node, n-.
nc+, nc-: Positive and negative controlling node numbers.
gain: Voltage gain.
The linear voltage-controlled voltage source is characterized by Equation 12-1.
12-1
Example

ER 4 5 6 7 55
F – Linear current controlled current source
Syntax

Fxxx n+ n- vcontrolname gain

Description

Fxxx: Name of the linear current controlled current source. It must begin with F.
n+, n-: Positive and negative terminal nodes. A positive current flows from the node, n+, through the
source to the node, n-.

12-16 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

vcontrolname: Name of the voltage source through which the controlling current flows. The direction
of positive controlling current flow is from the positive node, through the source, to the negative node
of vcontrolname.
gain: Current gain.
The linear current-controlled current source is characterized by Equation 12-2.
12-2
Example

F12 4 5 VIN 0.1

G – Linear voltage controlled current source
Syntax

Gxxx n+ n- nc+ nc- transconductance

Description

Gxxx: Name of the linear voltage controlled current source. It must begin with G.
n+, n-: Positive and negative terminal nodes. A positive current flows from the node, n+, through the
source to the node, n-.
nc+, nc-: Positive and negative controlling node numbers.
transconductance: Transconductance (in 1/Ohms).
The linear voltage controlled current source is characterized by Equation 12-3.
12-3
Example

G2 4 5 6 7 5.5
H – Linear current controlled voltage source
Syntax

Hxxx n+ n- vcontrolname transresistance

Description

Hxxx: Name of the linear current controlled voltage source. Must begin from H.
n+, n-: Positive and negative terminal nodes. A positive current flows from the node, n+, through the
source to the node, n-.
vcontrolname: Name of voltage source through which the controlling current flows. The direction of
positive controlling current flow is from the positive node, through the source, to the negative node of
vcontrolname.
transresistance: transresistance (in Ohms).
The linear current controlled voltage source is characterized by Equation 12-4.
12-4
Example

H12 4 5 V1 0.1K

SILVACO International 12-17

ATLAS User’s Manual

I – Independent current source

Syntax

Ixxx n+ n- value [AC acmag] [transient_parameters]

Description
Ixxx: Name of the independent current source. It must begin with I.
n+, n-: Positive and negative terminal nodes.
value: DC value of the source (ampers).
AC: Keyword for the AC source value.
acmag: AC magnitude.
transient_parameters: The transient parameters are described in Section 12.4.3: “Transient
Parameters”.
Example

I1 2 8 0. PULSE 0 200 0 20ns 20ns 100ns 10 100

I2 1 5 1u AC 2u
J – Junction Field-Effect Transistor (JFET)
Syntax
Jxxx nd ng ns [nb] mname [area] [M=val] [L=val] [W=val] [OFF]
[IC=vds,vgs]
[TEMP=val] [DTEMP=val]
Description
Jxxx: Name of the JFET element. It must begin with J.
nd, ng, ns, nb: Drain, gate, source and bulk terminal nodes. The bulk node doesn’t need to be
specified. If the bulk node is not specified, then the bulk is connected to the source node
mname: Model name. It must refer to a JFET model.
area: Area factor. The default is 1.0.
M: Multiplier used to describe multiple parallel JFETs.
L: Length of the gate in meters.
W: Width of the gate in meters.
OFF: Sets ON/OFF startup condition for DC analysis. Default is ON.
IC: Initial condition specification for and .

TEMP: Device operating temperature (oC).

DTEMP: Difference (in oC) between the device operating temperature and the circuit temperature.
Default value is 0.
Example

J44 1 4 6 jmodel

12-18 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

Note: See SMARTSPICE/UTMOST MODELING MANUAL, VOLUME 2 for a complete description of the JFET
models.

K – Coupling between two inductors

Syntax

Kxxx Lyyy Lzzz kval

Description

This is not a real circuit element. This statement defines only the coupling between two inductors.
Kxxx: Name. This parameter is not important and is used only to distinguish the statement. It must
begin with K.
Lyyy: First inductor element name. It must begin with an L and match one of the inductor names from

This is n31nducit ed iETEMC/Span <</MCID 7>>9DCB

SILVACO International 12-19

ATLAS User’s Manual

M – MOSFET
Syntax

Mxxx nd ng ns [nb] mname [L=val] [W=val] [AD=val] [AS=val]

[PD=val] [PS=val] [NRD=val] [NRS=val] [OFF] [IC=vds,vgs,vbs] [M=val]
[TEMP=val] [DTEMP=val] [GEO=val] [DELVTO=val]
Description

Mxxx: MOSFET element name. It must begin with M.

nd, ng, ns, nb: Drain, gate, source, and bulk terminal nodes. The bulk terminal node name is optional.
If it is unspecified, ground is used.
mname: Model name. It must refer to a MOSFET model.
L=val: Channel length in meters.
W=val: Channel width in meters.

AD: Drain diffusion junction area (meters2). This default is 0.

AS: Source diffusion junction area (meters2). This default is 0.

PD: Drain diffusion junction perimeter (meters). This default is 0.
PD: Source diffusion junction perimeter (meters). This default is 0.
NRD: The Number of squares of drain diffusion for resistance calculations. The default is 0.
NRS: The Number of squares of source diffusion for resistance calculations. The default is 0.
OFF: Sets ON/OFF startup condition for DC analysis. Default is ON.
IC: Initial voltage condition specification for , , and .
M: Multiplier used to describe multiple parallel MOSFETs. The default is 1.

TEMP: Device operating temperature (oC).

DTEMP: Difference (in oC) between the device operating temperature and the circuit temperature.
Default value is 0.
DELVTO: Threshold-voltage shift. When specified on the device line, the value overrides the value of
the model parameter, DELVTO. If not specified, the value of the model parameter is used.
Example

M1 2 4 8 9 mod1
Mout2 19 20 21 0 nmos L=5u W=2u TEMP=50
M22 3 5 7 8 mosmod1 L=10u W=5u AD=150p AS=150p PD=50u
PS=50u NRD=10 NRS=20

Note: See SMARTSPICE/UTMOST MODELING MANUAL, VOLUME 1 for a complete description of the MOSFET
models.

12-20 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

O – Optical source
Syntax

Oxxx beam value [transient_parameters]

Description
Oxxx: Name of an independent optical source. It must begin with O.
beam: Beam number. The beam with this number should be described in the ATLAS section of the
command file. See Chapter 10: “Luminous: Optoelectronic Simulator” for a complete description of
optoelectronic simulation.
value: DC optical intensity value (W/cm2).
transient_parameters: The transient parameters are described in Section 12.4.3: “Transient
Parameters”.

Note: The treatment of optical sources is fully similar to the treatment of independent voltage/current sources (i.e,.DC
statements can be used to simulated DC light responses of the circuit and transient parameters can be used to describe the
transient behavior of the optical sources).

Example
O1 1 0.001 pulse 0.001 0.002 0 2ns 2ns 100ns 10 100
Q – Bipolar junction transistor
Syntax

Qxxx nc nb ne [ns] mname [area] [OFF] [IC=vbe,vce] [M=val] [TEMP=val]

[DTEMP=val]
or
Qxxx nc nb ne [ns] mname [area=val] [areab=val] [areac=val] [OFF]
[IC=vbe,vce] [M=val] [TEMP=val] [DTEMP=val]
Description

Qxxx: Name of a bipolar junction transistor. It must begin with Q.

nc, nb, ne, ns: Collector, base, emitter, and substrate nodes. The substrate terminal node name is
optional. If it is unspecified, ground is used.
mname: Model name. It must refer to a BJT model.
area: Emitter area factor. The default value is 1.0.
areab: Base area factor. The default is area.
areac: Collector area factor. The default is area.
OFF: Sets ON/OFF startup condition for DC analysis. Default is ON.
IC: Initial voltage condition specification for , .
M: Multiplier used to describe multiple parallel BJTs. The default is 1.

TEMP: Device operating temperature (oC).

DTEMP: Difference (in oC) between the device operating temperature and the circuit temperature.
Default value is 0.

SILVACO International 12-21

ATLAS User’s Manual

Example

Q1 2 3 9 npnmod 1.5 IC=0.6,5.0

Q9 10 11 12 20 mod22 OFF TEMP=50

Note: See SMARTSPICE MODELING MANUAL, VOLUME 2 for a complete description of the BJT models.

R – Resistor
Syntax
Rxxx n+ n- value [transient_parameters]
Description
Rxxx: Name of the resistor element. It must begin with “R”.
n+, n-: Positive and negative terminal nodes.
value: Resistance in ohms.
transient_parameters: The transient parameters are described in Section 12.4.3: “Transient
Parameters”.

Note: Unlike the traditional SPICE program, transient parameters are acceptable for resistor elements. This allows
simulation of different kinds of time-dependent resistors and switches in a simple way.

Example

R12 4 5 100k
T – Lossless transmission line
Syntax

Txxx n1 n2 n3 n4 Z0= val TD=val

Description

Txxx: Name of the transmission line element. It must begin with T.

n1, n2: Nodes at port 1.
n3, n4: Nodes at port 2.
Z0: Characteristic impedance.
TD: Transmission delay.
Example

T1 1 0 2 0 Z0=50 TD=10ns

12-22 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

V – Independent voltage source

Syntax

Vxxx n+ n- value [AC acmag] [transient_parameters]

Description
Vxxx: Name of the independent voltage source. It must begin with V.
n+, n-: Positive and negative terminal nodes.
value: DC value of the source in units of volts.
AC: Keyword for the AC source value.
acmag: AC magnitude.
transient_parameters: The transient parameters are described in Section 12.4.3: “Transient
Parameters”.
Example

VCC 5 0 10.5
VIN 2 4 5.5 AC 1
Z – MESFET
Syntax
Zxxx nd ng ns [nb] mname [area] [M=val] [L=val] [W=val] [OFF]
[IC=vds,vgs]
[TEMP=val] [DTEMP=val]
Description
Jxxx: Name of the MESFET element. It must begin with Z.
nd, ng, ns, nb: Drain, gate, source and bulk terminal nodes. The bulk node doesn’t need to be
specified. If the bulk node it not specified, then the bulk is connected to the source node
mname: Model name. It must refer to a MESFET model.
area: Area factor. The default is 1.0.
M: Multiplier used to describe multiple parallel MESFETs.
L: Length of the gate in meters.
W: Width of the gate in meters.
OFF: Sets ON/OFF startup condition for DC analysis. Default is ON.
IC: Initial condition specification for and .

TEMP: Device operating temperature (oC).

DTEMP: Difference (in oC) between the device operating temperature and the circuit temperature.
Default value is 0.
Example

Z44 1 4 6 jmodel

SILVACO International 12-23

ATLAS User’s Manual

Note: See SMARTSPICE MODELING MANUAL, VOLUME 2 for a complete description of the MESFET models.

12.4.2: Control and Analysis Statements

.AC
.AC performs an AC linear small-signal analysis on the circuit. MIXEDMODE first creates a linearized
small-signal model at the operating point of the circuit and the computes the frequency response over
a user-specified range of frequencies.
Syntax
.AC DEC|OCT|LIN nump fstart fstop.
Description

DEC: Sweep frequency by decades.

OCT: Sweep frequency by octaves.
LIN: Linear frequency sweep. This is default.
nump: Total number of points per decade or per octave, or the total number of points of the linear
sweep.
fstart: Starting frequency (Hz).
fstop: Final frequency (Hz).
[Link] statements can be specified in the same command file. In this case, they will be executed
sequentially. Before executing the [Link] statement, the program will execute [Link] statements (if
any), regardless of the order of [Link] [Link] statements in the command file.
Example

.AC DEC 3 1.e3 1.e12

.AC LIN 20 1.e5 2.e6
.BEGIN
.BEGIN indicates the start of the circuit part of a MIXEDMODE command file.
.DC
.DC causes a DC transfer curve to be computed for the circuit with all capacitors opened and all
inductors shorted.
Syntax

.DC DEC|OCT|LIN source_name start stop numbers_steps

source_name2 DEC|OCT|LIN start2 stop2 number_steps2
Description

DEC: Sweep DC bias (voltage or current) by decades.

OCT: Sweep DC bias by octaves.
LIN: Linear DC bias sweep. This is the default.
source_name: Name of the independent voltage or current source to be swept.
start: Starting value of the sweep argument.

12-24 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

stop: Final value of the sweep argument.

number_steps: Number of steps of the inner sweep
source_name2: Name of the secondary sweep source.
start2: start value of the secondary sweep source
stop2: Final value of the secondary sweep source
number_steps2: Number of steps of the secondary sweep.
[Link] statements can be specified in a command file. In this case, they will be executed
sequentially. Before executing the [Link] statement, the program will simulate the circuit with the
independent source values given in the description of those sources.
[Link] statement is also often used to increment the values of independent voltage and current
sources in a circuit to avoid convergence problems.
Examples

.DC VIN 0. 5. 0.25

.DC IE 50 500 50
.END
.END indicates the end of the circuit part of a MIXEDMODE command file.
.IC
.IC sets specified node voltages during the steady-state simulation.
Syntax
.IC [V(I)=val_I...]
Description

This statement forces the specified node voltages to be set to specified values during the steady-state
simulation. These voltages are release when the transient simulation begins.
Example
.IC V(1)=10
.IC V(node1)=-0.5
.LOAD
.LOAD loads a solution file.
Syntax

.LOAD INFILE=filename
Description

INFILE: Name of a file ( ) to be loaded as an initial guess for further simulation. This file
must have been saved during a previous run of MIXEDMODE using [Link] statement.
Example

.LOAD INFILE=pdsave

Note: This statement is not used to load SSF format solution files from ATLAS ([Link] LOADSOLUTIONS)

SILVACO International 12-25

ATLAS User’s Manual

.LOG
.LOG specifies the filename for the circuit voltages and currents that will be saved.
Syntax

.LOG OUTFILE=filename
Description
OUTFILE: Name of a file ( ) for the circuit voltages and currents to be saved in standard
structure format files.
These files will have the following names:
For steady-state analysis:
"filename"_dc_1.log
"filename"_dc_2.log
"filename"_dc_3.log
..
(new file will be created for each .DC statement).
For AC analysis:
"filename"_ac_1.log
..
For network parameter extraction:
"filename"_net_1.log
..
For transient analysis:
"filename"_tr.log
..
To plot results of and entire steady-state analysis simultaneously, load all files related to steady-state
analysis into TONYPLOT.
Example

.LOG OUTFILE=pd
.MODEL
.MODEL specifies the circuit element model to be used for diodes, BJTs, or MOSFETs, and the
numerical values of parameters associated with the model.
Syntax

.MODEL name type <parameters>

Description

name: This is the model name. Circuit element definition statements refer to this name to link
elements to models.
type: This is the model type. This type must be consistent with the type of the circuit elements that
uses the model. The type can be one of the following:
• D - Diode model

12-26 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

• NMOS - n-channel MOSFET model.

• PMOS - p-channel MOSFET model.
• NPN - npn BJT model
• PNP - pnp BJT model
• NJF - n-channel JFET/MESFET model
• PJF - p-channel JFET/MESFET model
parameters: Model parameters. The parameters are described in the SMARTSPICE/UTMOST
MODELING MANUALS VOLUMES 1,2 and 3.
Example

.MODEL MODBJT NPN IS=1.E-17 BF=100 CJE=1F TF=5PS \

CJC=0.3F RB=100 RBM=20
.NET
.NET specifies that a network parameter extraction is to be performed.
Syntax

.NET INPORT OUTPORT DEC|OCT|LIN nump fstart fstop [Z0] [INDIN] [RSIN]
[INDOUT]] [RSOUT] [CIN] [COUT]
Description

INPORT: Input port description. It should be in one of the following formats:

• V(n+,n-): two nodes (positive (n+) and negative (n-)).
• Vxxxx: where Vxxxx is the name of an existing voltage source. The positive terminal of the
source becomes the positive input port node and the negative terminal becomes the negative
input node.
• Ixxxx: where Ixxxx is the name of an existing current source. The positive terminal of the
source becomes the positive input port node and the negative terminal becomes the negative
input node.

Note: If the nodes specified as the input port are the same nodes as an existing current or voltage source, then the name of
the source must be specified as inport. Also, remove all AC parameters from voltage or current sources before using
[Link] statement.

OUTPORT: Output port description. It should be in one of the following formats.

• V(n+,n-): two nodes (positive (n+) and negative (n-)).
• Vxxxx: where Vxxxx is the name of an existing voltage source. The positive terminal of the
source becomes the positive output port node and the negative terminal becomes the negative
output node.
• Ixxxx: where Ixxxx is the name of an existing current source. The positive terminal of the
source becomes the positive output port node and the negative terminal becomes the negative
output node.

Note: If the nodes specified as the output port are the same nodes as an existing current or voltage source, then the name of
the source must be specified as outport.

SILVACO International 12-27

ATLAS User’s Manual

DEC: Sweep frequency by decades.

OCT: Sweep frequency by octaves.
LIN: Linear frequency sweep. This is default.
nump: Total number of points per decade or per octave, or the total number of points of the linear
sweep.
fstart: Starting frequency (Hz).
fstop: Final frequency (Hz).
Additional optional parameters may also be specified on the .NET statement.
Z0: Matching Impedance (default = 50 Ohms).
INDIN: Inductance through which the DC voltage source is connected to the input source (only if
INPORT is given as Vxxxx).
RSIN: Series resistance of INDIN.
INDOUT: Inductance through which the DC voltage source is connected to the output source (only if
OUTPORT is given as Vxxxx).
RSOUT: Series resistance of INDOUT.
CIN: Capacitance through which the S-parameter test circuit is connected to the input port.
COUT: Capacitance through which the S-parameter test circuit is connected to the output port.

Note: The S-parameters will be automatically saved to the log file. The Z, Y, H, ABCD, and gain small-signal parameters can
also be written to the log file. These are selected through [Link] statement. You can also view the default values if
PRINT is specified in [Link] statement

Examples
.NET V1 V2 DEC 10 1e6 1e10
.NET I1 V2 DEC 10 1e6 1e10 Z0=75 RSOUT=100
.NET V(1,0) V(2,3) DEC 10 1e6 1e10
.NODESET
.NODESET sets initial values for circuit node voltages.
Syntax
.NODESET [V(I)=VAL_I ...]
Description

This statement specifies the initial values for circuit node voltages. If a node voltage is not specified,
the program will try to find a solution using zero as an initial guess for this node. This statement can
significantly reduce the CPU time needed to calculate the initial condition.
Example

.NODESET V(1)=50 V(2)=49.4 V(3)=10 V(5)=-1.5

.NODESET V(in1)=0 V(2)=2 V(out1)=-1

12-28 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

.NUMERIC
.NUMERIC specifies special numeric parameters for the circuit analysis.
Syntax

.NUMERIC [parameters]

Parameter Type Default Units

IMAXDC Integer 25

IMAXTR Integer 15

DTMIN Real 1*10-12 s

LTE Real 0.1

TOLDC Real 1*10-4

TOLTR Real 1*10-4

VCHANGE Real 5*107 V

VMAX Real 5*107 V

VMIN Real -5*107 V

Description

IMAXDC: Maximum number of mixed circuit-device iterations to be performed during steady-state

analysis.
IMAXTR: Maximum number of mixed circuit-device iterations to be performed during transient
analysis.
DTMIN: Minimum time step value for transient analysis.
LTE: Local truncation error for transient analysis.
TOLDC: Relative accuracy to be achieved during steady-state analysis for the calculation of voltages
in circuit nodes.
TOLTR: Relative accuracy to be achieved during transient analysis for the calculation of voltages in
circuit nodes.
VCHANGE: Maximum allowable change in circuit node voltages between two mixed circuit-device
iterations. (This parameter can be useful for reaching steady-state convergence with a bad initial
guess.)
VMAX: Maximum value for circuit node voltages.
VMIN: Minimum value for circuit node voltages.
Example

.NUMERIC LTE=0.05 TOLDC=1.*10-8 DTMIN=1ns

SILVACO International 12-29

ATLAS User’s Manual

.OPTIONS
.OPTIONS specifies various circuit simulation options.
Syntax

.OPTIONS [parameters]

Parameter Type Default Units

[Link] Logical False

CNODE Real 1*10-16 F

CYLINDER Logical False

FULLN Logical True

GAINS Logical False

[Link] Logical False

LOADSOLUTIONS Logical False

M2LN Logical False

[Link] Logical False

NOPROJ Logical False

NOSHIFT Logical False

PRINT Logical False

RELPOT Logical False

RV Real 1*10-4 W

TEMP Real 300 K

TNOM Real 300 K

WRITE Integer 1

[Link] Logical False

[Link] Logical False

Description

[Link]: ABCD parameters will be written to the log file. This is used in conjunction with
[Link] statement.
CNODE: A very small capacitance, which for algorithmic reasons automatically connected from each
circuit node to ground. This value can be set to 0.
CYLINDR: Cylindrical coordinate system for all ATLAS devices.
FULLN: Full Newton solution method is used during steady-state simulation.
GAIN: Stability factor (K), unilateral power gain (GU), maximum unilateral transducer power gain
(GTUmax) and |H21|^2 are written to the LOG file. This is used in conjunction with [Link]
statement.

12-30 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

[Link]: H-parameters will be written to the LOG file. This is used in conjunction with [Link]
statement.
LOADSOLUTIONS: Solutions, structures, doping distributions, and meshes, are to be loaded from
standard structure files. The solutions are used as the initial guess or initial conditions for subsequent
MIXEDMODE simulation. To use this feature, you must:
• Calculate a solution for each ATLAS device and save each solution in a separate standard
structure format file using SAVE or SOLVE.... MASTER.
• For each ATLAS device, use the A statement in the MIXEDMODE command file to specify the
associated standard structure format file
• Set node voltages to appropriate values with [Link] statement
• Specify LOADSOLUTIONS in [Link] statement

Note: If using this feature, specify solutions for all ATLAS devices.

[Link] statement must always be used when LOADSOLUTIONS is used. [Link]

statement is used to make the initial circuit voltages match those device solutions that were obtained.
You may also need to specify the NOSHIFT parameter of the OPTIONS statement. By default,
MIXEDMODE shifts device terminal voltages with respect to the voltage on the first terminal that is
specified in the A statement. You must either prepare initial solutions with this terminal grounded, or
specify NOSHIFT in the OPTIONS statement.
M2LN: Uses the modified two-level Newton solution method during steady-state simulation. The full
NEWTON method provides faster solution than the modified two-level Newton method when a good
initial guess is available. The modified two-level method is more reliable when the initial guess is far
from the solution. The default is the full NEWTON method.
[Link]: Uses the modified two-level Newton solution method during transient simulations.
NOPROJ: Disables the initial guess project method for the ATLAS iterations. MIXEDMODE attempts
to extrapolate the values of the ATLAS device variables (such as potential and carrier concentration)
for the each iteration. Specifying NOPROJ disables the extrapolation and the previous values of
potential and carrier concentration are used instead.
NOSHIFT: Disables the shift of voltages for ATLAS device models. MIXEDMODE normally shifts the
voltages on ATLAS device terminals to be referenced to the voltage on the first terminal. From the
physical point of view, the state of the p-n diode is the same for voltages of 0V and 0.5V on the diode
terminals with 1000V and 1000.5V, but the first situation is better for numerical simulation.
PRINT: Enables printing of circuit nodes voltages after the calculation for each bias point (DC
analysis) or time step (transient the analysis).
RELPOT: Enables the use of relative convergence criteria for potential for ATLAS models. By default,
ALTAS models use absolute convergence criteria for potential. When bias voltages are large (a
common situation for power devices), then absolute convergence criteria are not appropriate and this
parameter should be specified.
RV: Defines the ohmic resistance that MIXEDMODE associates with all voltage sources and all
inductances. This value should never be set to 0. The default value is small enough to avoid errors due
to the influence of the internal resistance. Usually, extremely small values of this parameters can
cause convergence problems. It is usually acceptable to decrease this parameter to the range of 1*10-6-
1*10-7. This parameter should not be varied unless there is a compelling reason to do so.
TEMP: Device temperature to be use during the simulation.
TNOM: Circuit temperature to be use during the simulation.

SILVACO International 12-31

ATLAS User’s Manual

WRITE: How often the solution is to be saved in standard structure files during the simulation. For
example, write=3 specifies that the solution will be saved at every third timestep. Specifying this
parameter can help avoid disk overflow.
[Link]: Y-parameters should be written to the log file. This is used in conjunction with [Link]
statement.
[Link]: Z-parameters should be written to the log file. This is used in conjunction with [Link]
statement.
Example

.OPTIONS TNOM=293 FULLN

.PRINT
.PRINT specifies which device output parameters will be printed to the log files.
Syntax
.PRINT [antype] parmeter(device_name) [parmeter2(device_name) ...]
Description
antype: Type of analysis for which the outputs are desired. If antype is unspecified, the outputs of all
simulation types will be printed. antype must be one of the following keywords:
• AC: AC analysis outputs
• DC: DC analysis outputs
• NET: Network analysis outputs
• TRAN: Transient analysis outputs
parmeter: Output variables or expressions to be printed.
device_name: The device name.
Example

.PRINT ic(q1) ib(q1) is(q1)

.PRINT AC ic(q1) ib(q2)
.PRINT DC ic(q1) ib(q2) i(d1)
.PRINT TRAN cd(m1) cg(m1) cs(m1) cb(m1)
.SAVE
.SAVE saves simulation results into files for visualization or for future use as an initial guess.
Syntax

.SAVE OUTFILE=name [MASTER=mname]

Description

OUTFILE: Specifies that after the simulation is finished the solution is to be written to a file called
. The ATLAS model solutions will be written to the file ( ) and the circuit solution will be
written to the file, . These files can be used later for loading solutions to be used as an initial
guess ([Link] statement).
MASTER: Internal states of all ATLAS models should be written during the simulation in standard
structure format for future visualization using TONYPLOT. These files, with the base name, ,
will be written after the calculation of each bias point during DC simulation, and after of each time
step during transient simulation.

12-32 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

The program will automatically add the following suffixes to :

• During DC simulation: , where is the number of the DC point.
• During transient simulation: , where is the number of the time step.
Example
.SAVE OUTFILE=pdsave MASTER=pd
.TRAN
.TRAN specifies that a transient analysis is to be performed.
Syntax

.TRAN tstep tstop

Description

tstep: Time interval in seconds.

tstop: Final time value for which the simulation is to be performed.
Transient analysis is performed only after the execution of [Link] statements. If no DC statements are
used the transient starts after the calculation of the initial circuit state with the values of the
independent sources given in the descriptions of those sources.
[Link] statements are supported. The TSTOP parameter is not reset between [Link]
statement.
Example

.TRAN 1ns 100ns

12.4.3: Transient Parameters

MIXEDMODE allows you to specify transient parameters for voltage sources (Vxxx), current sources
(Ixxx), and resistors (Rxxx). These parameters describe the time development behavior of the source.
EXP
EXP is used to define an exponential waveform. The waveform is specified as follows:
EXP i1 i2 td1 tau1 td2 tau2
where:
i1 is the initial value.
i2 is the pulsed value.
td1 is the rise delay time.
td2 is the fall delay time.
tau1 is the rise time constant.
tau2 is the fall time constant.
The transient behavior is shown in Table 12-2.

SILVACO International 12-33

ATLAS User’s Manual

Table 12-2. Transient behavior for EXP

Time Value

0 < t < td1 i1

td1 ≤ t < td2 i1+(i2-i1).(1-exp[(-t-td1)/tau1])

td2 ≤ t i1+(i2-i1).(1-exp[(-t-td1)/tau1])
+(i1-i2).(1-exp[(-t-td2)/tau2])

GAUSS
GAUSS is used to define a Gaussian waveform. The waveform is specified as follows:
GAUSS i1 i2 td1 tau1 td2 tau2
where:
i1 is the initial value.
i2 is the pulsed value.
td1 is the rise delay time.
td2 is the fall delay time.
tau1 is the rise time constant.
tau2 is the fall time constant.
The transient behavior is shown in Table 12-3.

Table 12-3. The transient behavior for GAUSS

Time Value

0 < t < td1 i1

td1 ≤ t < td2 i1+(i2-i1).(1-exp[((-t-td1)/tau1)2]

td2 ≤ t i1+(i2-i1).(1-exp[((-t-td1)/tau1)2]
+(i1-i2).(1-exp[((-t-td2)/tau2)2]

PULSE
PULSE is used to define a pulse waveform. The waveform is specified as follows:
PULSE i1 i2 td tr tf pw per
where:
i1 is the initial value.
i2 is the pulsed value.
td is the delay time before the pulse is started.
tr is the rise time of the pulse.
tf is the fall time of the pulse.

12-34 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

pw is the pulse length per period

per is the period.
The transient behavior is described in Table 12-4. Intermediate points are found by linear
interpolation.

Table 12-4. Transient behavior for PULSE

Time Value

0 i1

td i1

td + tr i2

td + tr + pw i2

td + tr + pw + tf i1
td + per i1

td + per + tr i2 (second period)

SFFM
SFFM is used to define a modulated sinusoidal waveform. The waveform is specified as follows:
SFFM io ia fc mdi fs
where:
io is the DC offset.
ia is the amplitude.
fc is the carrier frequency.
mdi is the modulation index.
fs is the signal frequency.
The transient behavior will be:
value(t)=io + ia · sin[π · fc · t + mdi · sin(2π · fs · t)]
SIN
SIN is used to define a sinusoidal waveform. The waveform is specified as follows:
SIN io ia freq td theta
where:
io is the offset.
ia is the amplitude.
freq is the frequency.
td is the delay.
theta is the damping factor.

SILVACO International 12-35

ATLAS User’s Manual

The transient behavior is shown in Table 12-5.

Table 12-5. Transient Behavior for SIN

Time Value

t < td value(t) = io

t ≥ td value(t) = i0+[Link][(-t-td)/THETA]
.sin[2π.freq.(t-td)]

TABLE
TABLE is used to define a waveform using a table of values. This parameter is used as follows:
TABLE infile=<table_file_name>
where table_file_name is an ASCII text file that contains the tabulated time-dependence of a
variable in the following format:
t1 v1
t2 v2
t3 v3
...
tN vN
end
Each line contains two numbers. The first number is the time in seconds. The second number is the
time-dependent variable) voltage in volts, the current in amps, or the resistance in ohms). Up to 1000
lines can be used. Input is terminated by the word end.
If during the simulation the transient time becomes larger than the last value in the table, then the
last value will be used for the remainder of the simulation

12.4.4: User-Defined Two-Terminal Elements

MIXEDMODE users who are familiar with C-INTERPRETER can define their own two-terminal elements
using the B statement and a function written in C that defines the behavior of the element.
User-Defined Model
A user-defined model is specified by defining the following:
• The dependencies of the device terminal current on the terminal voltages
• The derivatives of the terminal current with respect to the terminal voltages
• The device current (I) is described by Equation 12-6.

dU
I = F1 ( U, t ) + F2 ( U, t ) ⋅  ------- 12-6
 dt 

where:
U is the device voltage, t is time, and F1 and F2 are functions that determine the behavior of the
device. The first term on the right hand side describes the “DC” current and the second term describes
“capacitive” current.

12-36 SILVACO International

 MixedMode: Mixed Circuit and Device Simulator

The following four functions are user-specified:

• F1(U,t): the DC current.
• F2(U,t): the capacitance.
• dF1(U,t)/dU: the DC differential conductance.
• Q: the charge associated with F2(U,t).
To define the element, prepare a text file that contains an appropriate function written in C. A
template for this user-defined function is shown below:
int udef(double v, double temp, double ktq, double time, double *curr,
double *didv, double *cap, double *charge)
{
/* user-supplied code here */
return(0);
}
Input Parameters
Four input parameters are supplied to the function and can be used in the user-defined code. The input
parameters are:
• v: the voltage across the element (V)
• temp: the temperature (K)
• ktq: the thermal voltage kT/q (V)
• time: transient time (sec); a value of 0 is supplied during DC calculations
Output Parameters
The four output parameters that must be returned by the function are:
• curr: the value of F1 (Amps)
• didv: the value of dF1(v, time)/dU (A/V)
• cap: the value of F2(v, time)
• charge: the value of the charge (Q)
Example

Consider an element that consists of a resistor R and a capacitor C connected in parallel. The equation
for the total current through this combination is:

U dU
I ( U, t ) = ---- + C ⋅  ------- 12-7
R  dt 

The quantities that must be user-defined are:

U
F1 ( U, t ) = ---- 12-8
R

12-9

dF ( U, t ) 1
--------------------- = --- 12-10
dU R

SILVACO International 12-37

ATLAS User’s Manual

12-11

When R=2kΩ and C=100pF, a user-defined function could have the following form:
intrc(double v, double temp, double ktq, double time, double *curr, double
*didv, double *cap, double *charge)
{
*curr = v/2000.0;
*didv = 1.0/2000.0
*cap = 1.0e-10;
*charge=1.0e-10*v;
return(0); /* 0 - ok */
}

12-38 SILVACO International

 Chapter 13:
Quantum: Quantum Effect Simulator

13.1: Self-Consistent Coupled Schrodinger Poisson Model

To model the effects of quantum confinement, ATLAS also allows the solution of Schrodinger’s
Equation along with the fundamental device equations. The solution of Schrodinger's Equation gives a
quantized description of the density of states in the presence of quantum mechanical confining
potential variations.
The calculation of the quantized density of states relies upon a solution of Schrodinger’s Equation.

h∂ ∂Ψ 
–   ( )Ψ Ψ 13-1
∂  ∂ 

h∂ ∂Ψ 
–   ( )Ψ Ψ 13-2
∂  ∂ 

where is the bound state energy for transverse mass, is longitudinal, Ψ is the transverse
wavefunction, Ψ is the longitudinal wavefunction, and is the is the band edge energy. This
expression is for electrons but an equivalent expression exists for holes.
In the case that the mass is isotropic, only one solution to Schrodinger's equation is obtained with the
appropriate mass. To choose the number of directions in space, specify the [Link] parameter.
If you set [Link] to 1, you will obtain the solution for isotropic effective mass. If you set
[Link] is set to 3, you will obtain a solution for a single lateral effective mass and 2 equivalent
transverse masses.
To specify how many valence bands to consider, specify the [Link] parameter. If you set [Link]
to 1, you will obtain a Schrodinger solution for only one valence band. If you set [Link] to 3, you
will cause solutions for heavy holes, light holes and holes in the split off band.
Using Fermi-Dirac statistics, the discrete nature of the quantized density of states reduces the integral
over energy to a sum over bound state energies.
∞
–
–  +
= ----------
πh
∑ Ψ ( )
 
13-3

∞
 – 
----------
πh
∑ Ψ ( ) –




The self-consistent coupled Schrodinger-Poisson model is enabled for electrons by setting the SCHRO
parameter of the MODEL statement. With this parameter set, ATLAS solves the one-dimensional
Schrodinger's Equation along a series of slices in the y direction relative to the device.
The locations of the slices in the y direction is developed in two ways. For rectangular ATLAS-defined
meshes, the slices will automatically be taken along the existing mesh lines in the ATLAS mesh. If the
mesh is non-rectangular and/or not an ATLAS defined mesh, specify a rectangular mesh. To do this,
specify the locations of individual mesh lines and their local spacings using the [Link] and
[Link] statements like that to the specification of a device mesh using the [Link] and [Link] or a
laser mesh using the [Link] and [Link] statements.

SILVACO International 13-1

ATLAS User’s Manual

If you want, specify a mesh for Schrodinger's Equation to solve on an ATLAS rectangular mesh using
the [Link] and [Link] statements. But since, by default, the solver uses the ATLAS mesh, you
must specify [Link] in the MODEL statement to use the separate mesh.
For holes, the Schrodinger-Poisson model is enabled by setting the [Link] parameter in the MODEL
statement.
Once the solution of Schrodinger’s Equation is taken, carrier concentrations calculated from Equation
13-3 are substituted into the charge part of Poisson’s Equation. The potential derived from solution of
Poisson's Equation is substituted back into Schrodinger's Equation. This solution process (alternating
between Schrodinger’s and Poisson’s equations) continues until convergence and a self-consistent
solution of Schrodinger's and Poisson's equations is reached.
Since the wavefunctions diminish rapidly from the confining potential barriers in the Schrodinger
solutions, the carrier concentrations become small and noisy. You can refine these carrier
concentrations by setting a minimum carrier concentration using the QMINCONC parameter from the
MODELS statement. This parameter sets the minimum carrier concentration passed along to the
Poisson solver and the output to the structure files. The transition between the Schrodinger solution
and the minimum concentration is refined between 10xQMINCONC and QMINCONC so that it is
continuous in the first derivative.
Use the SAVE statement or the OUTFILE parameter from the SOLVE statement to write the solutions of
the self-consistent system into a structure file. These structure files will then contain the self-
consistent potential and electron or hole concentrations.
The Eigen energies and functions can also be written to the structure file by specifying the EIGENS
parameter from the OUTPUT statement. This parameter specifies the number of Eigen energies/wave
functions to be written.
The number of Eigen values solved is limited to a number of 2 less than the total number of grid points
in the Y direction. Note that the self-consistent solution of Schrodinger’s Equation with the Poisson’s
Equation doesn’t allow solutions for the electron and hole continuity equations in the current ATLAS
version. Non-self-consistent solutions, however, can be obtained by setting the [Link] parameter
in the MODELS statement. These non-self-consistent solutions are obtained by solving Schrodinger's
Equation only after convergence is obtained. That way, you can obtain Schrodinger solutions with the
electron and hole continuity equations.
Similar results are saved to the structure file and the meaning of the EIGENS parameter in the
OUTPUT statement are the same. In obtaining post-processed ([Link]) solutions for the
Schrodinger's Equation, an assumption is made about the location of the electron or hole quasi-fermi
level. You can set two flags in the MODELS statement to vary the interpretation of these results. These
flags are the [Link] and [Link] parameters, which have interpretations in Table 13-1.

Table 13-1. Interpretations of optional parameters during post-processed Schrodinger solution

[Link] [Link] Quasi-Fermi Level Calculation Method

FALSE FALSE Quasi-Fermi level is calculated from the local electron density
using Equation 3-9.

FALSE TRUE Quasi-Fermi level varies with Y position and is calculated to match
the local classical and quantum-mechanical charge concentration.
TRUE FALSE Quasi-Fermi level is uniformly zero.

TRUE TRUE Quasi-Fermi level is uniform across Y slice and is calculated to

match the classical and quantum-mechanical sheet charge.

13-2 SILVACO International