The development of the family of scanning probe microscopes started with the original

invention of the STM in 1981. Gerd Binnig and Heinrich Rohrer developed the first working
STM while working at IBM Zurich Research Laboratories in Switzerland. This instrument
would later win Binnig and Rohrer the Nobel prize in physics in 1986.

The Quantum Corral

The STM image below shows the direction of standing-wave patterns in the local density
of states of the Cu(111) surface. These spatial oscillations are quantum-mechanical
interference patterns caused by scattering of the two-dimensional electron gas off the Fe
atoms and point defects.

Courtesy of International Business Machines

Corporation. Unauthorized use not permitted

3D rendered
Scanning Tunneling Microscope image of atoms.

How an STM Works

The scanning tunneling microscope (STM) works by scanning a very sharp metal wire tip
over a surface. By bringing the tip very close to the surface, and by applying an electrical
voltage to the tip or sample, we can image the surface at an extremely small scale – down
to resolving individual atoms.

The STM is based on several principles. One is the quantum mechanical effect of
tunneling. It is this effect that allows us to “see” the surface. Another principle is the
piezoelectric effect. It is this effect that allows us to precisely scan the tip with angstrom-
level control. Lastly, a feedback loop is required, which monitors the tunneling current and
coordinates the current and the positioning of the tip. This is shown schematically below
where the tunneling is from tip to surface with the tip rastering with piezoelectric
positioning, with the feedback loop maintaining a current setpoint to generate a 3D image
of the electronic topography:

Schematic of
scanning tunneling microscopy (STM).

Tunneling

Schematic of electron wavefunction.

Tunneling is a quantum mechanical effect. A tunneling current occurs when electrons
move through a barrier that they classically shouldn’t be able to move through. In classical
terms, if you don’t have enough energy to move “over” a barrier, you won’t. However, in
the quantum mechanical world, electrons have wavelike properties. These waves don’t
end abruptly at a wall or barrier, but taper off quickly. If the barrier is thin enough, the
probability function may extend into the next region, through the barrier! Because of the
small probability of an electron being on the other side of the barrier, given enough
electrons, some will indeed move through and appear on the other side. When an
electron moves through the barrier in this fashion, it is called tunneling.

Quantum mechanics tells us that electrons have both wave and particle-like properties.
Tunneling is an effect of the wavelike nature.

The top image shows us that when an electron (the wave) hits a barrier, the wave doesn’t
abruptly end, but tapers off very quickly – exponentially. For a thick barrier, the wave
doesn’t get past.

The bottom image shows the scenario if the barrier is quite thin (about a nanometer). Part
of the wave does get through and therefore some electrons may appear on the other side
of the barrier.

Because of the sharp decay of the probability function through the barrier, the number of
electrons that will actually tunnel is very dependent upon the thickness of the barrier. The
current through the barrier drops off exponentially with the barrier thickness.

To extend this description to the STM: The starting point of the electron is either the tip or
sample, depending on the setup of the instrument. The barrier is the gap (air, vacuum,
liquid), and the second region is the other side, i.e. tip or sample, depending on the
experimental setup. By monitoring the current through the gap, we have very good control
of the tip-sample distance.

Piezoelectric Effect

The piezoelectric effect was discovered by Pierre Curie in 1880. The effect is created by
squeezing the sides of certain crystals, such as quartz or barium titanate. The result is the
creation of opposite charges on the sides. The effect can be reversed as well; by applying a
voltage across a piezoelectric crystal, it will elongate or compress.

These materials are used to scan the tip in an scanning tunneling microscopy (STM) and
most other scanning probe techniques. A typical piezoelectric material used in scanning
probe microscopy is PZT (lead zirconium titanate).
 Feedback loop and
electron tunneling for scanning tunneling microscopy (STM).

Feedback Loop

Electronics are needed to measure the current, scan the tip, and translate this information
into a form that we can use for STM imaging. A feedback loop constantly monitors the
tunneling current and makes adjustments to the tip to maintain a constant tunneling
current. These adjustments are recorded by the computer and presented as an image in
the STM software. Such a setup is called a constant current image.

In addition, for very flat surfaces, the feedback loop can be turned off and only the current
is displayed. This is a constant height image.

2. Scanning Tunneling Microscopy (STM)

Scanning Tunneling Microscopy allows researchers to map a conductive sample’s surface

atom by atom with ultra-high resolution, without the use of electron beams or light, and
has revealed insights into matter at the atomic level for nearly forty years. This blog post
will provide a high-level overview of what scanning tunneling microscopy is, how it works,
and how it has been applied in research.

What is Scanning Tunneling Microscopy (STM)?

Scanning Tunneling Microscopy, or STM, is an imaging technique used to obtain ultra-high
resolution images at the atomic scale, without using light or electron beams. STM was
invented in 1981 by two IBM scientists named Gerd Binnig and Heinrich Rohrer. Five years
later, the invention earned them the Nobel Prize in Physics.

STM was the first technique developed in the larger class of scanning probe microscopy
(SPM) imaging modes. At the time, it allowed researchers to capture significantly more
detail than any previous microscopy—all the way down to atoms and inter-atomic spacing.
This ultra-high resolution capability allowed researchers to precisely map the three-
dimensional topography and electronic density of states of conductive materials, and even
manipulate individual atoms on the surface of these materials. Over the following
decades, STM has revolutionized the field of nanotechnology and continues to play an
important role in fundamental and industrial research across a variety of disciplines.

STM image of highly-oriented pyrolytic graphite (HOPG)

Acquired in air on an Asylum Research Cypher S scanning probe microscope,
demonstrating the ability to resolve the atomic lattice structure. (5 nm scan size)

How does STM Work?

STM is a remarkable and rare example of harnessing a quantum mechanical process

(electron tunneling) in a real-world practical application. The term “tunneling” refers to
the situation where electrons traverse a barrier (in this case, a tiny gap between the tip
and surface) that initially seems like it should be impenetrable—like throwing a ball
against a wall. The physics that describes this ball-wall interaction is called the “classical
paradigm,” and the ball will never tunnel through the wall. Electrons, by contrast, have a
quirky wave-like character that makes them a “fuzzy” cloud (unlike a ball), so they can
actually exist on both sides of the barrier simultaneously and therefore have a non-zero
probability of moving across the barrier even if the barrier energy is higher than the total
energy of the electron.

STM works by scanning a sharp conductive probe very close to the surface of a conductive
specimen and forcing electrons to traverse the gap between them. When the tip is
sufficiently near the surface (usually <1 nm away), the fuzzy electron cloud of the first
atom of the tip and surface begin to overlap. Applying a bias voltage between the tip and
the surface in this configuration produces a current because electrons are driven to tunnel
through the potential barrier from the tip to the surface via the overlapping electron
cloud. This tunneling current is highly sensitive to the gap between the probe tip and
surface, varying exponentially with the tip-sample distance. As the tip scans line by line
across the surface of the sample, the intensity of the tunneling current maps the sample’s
electronic density of states.

The STM operates in two distinct modes: constant height mode and constant current
mode. Constant height mode is generally used when the sample surface is very smooth. In
this mode, the probe tip stays at a set height while it quickly raster scans across the
sample. By measuring changes in the intensity of the tunneling current as a function of
(x,y) position and bias voltage, researchers can construct an image of the electronic
density of states of the sample surface, defects, frontier molecular orbitals, and more.

The more popular mode is known as the constant current mode. In this mode, the
tunneling current is held constant by the use of a feedback loop system that adjusts the
distance between the tip and the surface. That is to say, if the tunneling current exceeds
the target value, the feedback control system will increase the distance between the tip
and the sample; if the tunneling current is less than the target current value, the feedback
control system will bring the tip closer to the surface of the sample. The resulting three-
dimensional distance profile as a function of (x,y) position can help researchers measure a
wide array of characteristics, including surface roughness, defects, and the size and
conformation of molecules on the surface.

STM can operate under a wide range of different environmental conditions. Often, STMs
are categorized as either ambient STMs or ultra-high vacuum (UHV) STMs. Ambient STMs
typically operate in air or other gases at near room temperature. With specialized
insulated tips, ambient AFMs can also be operated in liquids. Oxford Instruments Asylum
Research offers ambient STM capabilities on both our MFP-3D SPMs and Cypher SPMs.

UHV STMs differ, obviously, in that they operate under very high vacuum. This is often
done in highly specialized UHV systems where the sample is grown or etched and then
imaged in situ. Their operation in vacuum allows them to operate across a very wide
temperature range, from near zero Kelvin to over 1000°C. Asylum Research does not offer
UHV STMs, though the Oxford Instruments Nanoscience group produces UHV equipment
sometimes used with STM.
What are the Uses of STM in Research?

Since its discovery, the STM has been responsible for major breakthroughs in
nanotechnology and has enabled novel research across various disciplines, including
semiconductor science, electrochemistry, surface chemistry, and more.

Initially, the STM was used to characterize the topology of different metals and delineate
the atomic structure of their surfaces. For the first time, researchers were able to discern
the atomic-scale properties of materials, including surface roughness, defects, and surface
reaction mechanisms. By investigating the atomic lattices of materials, researchers could
begin to understand properties relevant to the fabrication of electronic components,
including conductivity, distributions of frontier molecular orbitals and their energies, and
reaction dependencies on crystal facet orientations, to name a few.

Over the years, STM began to be employed for a variety of applications outside of atomic-
scale imaging. It has been used to assemble and manipulate individual atoms on a surface.
This opened up new vistas for nanotechnology, including the generation of nanostructures
such as quantum corrals and molecular switches. Contacts on nanodevices can also be
constructed with STM by depositing metals (such as gold, silver, or tungsten) in a specified
pattern. Researchers have also used STM to induce chemical reactions and study the
subsequent reaction mechanisms at the molecular level.

Ambient STMs are commonly used to investigate the structure of self-assembled

molecules on surfaces because they can often resolve single molecules and even sub-
molecular structure. For example, the image below clearly shows single molecules of 5-
octadecoxyisophthalic acid in a self-assembled monolayer on HOPG.
STM image of a self-assembled monolayer of 5-octadecoxyisophthalic acid on highly-
oriented pyrolytic graphite.
Molecular models are overlaid to emphasize the molecular ordering. Imaged on a Cypher
AFM, 15 nm scan. Sample courtesy of Matt Blunt, University of Nottingham.

More recently, low-current STM was used on the Cypher ES to resolve single molecule
level detail in self-assembled layers of nickel octaethylporphyrin (NiOEP) on HOPG. The
low-current imaging capability can provide higher resolution by allowing the STM to
operate at tunneling currents as low as 300 femtoamps. The images below show constant-
height STM images, but the full application note demonstrates similar resolution in
constant current STM mode. Though the Cypher provides exceptional resolution using
STM, it is worth noting that Cypher using tapping mode atomic force microscopy (AFM)
can actually rival that resolution.
Constant height STM current images of the 2D lattice of NiOEP on HOPG
(A) 50 nm survey scan showing the NiOEP grain boundary (white arrows), zoom regions,
and moiré pattern imaged with a 6.4 pA setpoint. (B) Zoomed 10 nm region imaged at 300
fA setpoint. (C) Zoomed 5 nm region showing sub-nm molecular resolution imaged with a
60 pA setpoint. Inset: the CPK molecular model of NiOEP.

References

1. Binnig, G., & Rohrer, H. (1983). Scanning tunneling microscopy. Surface Science,
126(1-3), 236–244.

2. Chen, Julian. (1994). Introduction to Scanning Tunneling Microscopy: Second

Edition. American Journal of Physics

Scanning Tunneling Microscopy (STM)

STM is the oldest of the scanning probe microscopes and involves bringing an atomically
sharp tip within a couple of nanometres of the sample. When a bias is applied between the
tip and sample, electrons will tunnel from the one to the other and this current can be used
to measure the topography of the sample. The unique aspect of STM is that is can readily be
used to acquire images with atomic resolution which is not generally true for the other
SPMs. However, because a current is being measured only conducting samples can be
examined and hence this limits the range of possible samples.

Scanning Tunnelling Microscopy (STM) relies on “tunnelling current” between the probe and
the sample to sense the topography of the sample. The STM probe, a sharp metal tip (in the
best case, atomically sharp), is positioned a few atomic diameters above a conducting
sample which is electrically biased with respect to the tip. At a distance under 1 nanometre,
a tunnelling current will flow from sample to tip. In operation, the bias voltages typically
range from 10 to 1000 mV while the tunnelling currents vary from 0.2 to 10 nA. The
tunnelling current changes exponentially with the tip-sample separation, typically
decreasing by a factor of two as the separation is increased 0.2 nm. The exponential
relationship between the tip separation and the tunnelling current makes the tunnelling
current an excellent parameter for sensing the tip-to-sample separation. In essence, a
reproduction of the sample surface is produced by scanning the tip over the sample surface
and sensing the tunnelling current. The first STM operated in ultrahigh vacuum on
cryogenically cooled samples. In the years following its invention in 1981, many variations on
the STM theme appeared. P.K. Hansma and J. Tersoff wrote a good review article on the
subject, containing experiments, theory, and over 100 references for The Journal of Applied
Physics, 61 pp. R1-23, 1987.

STM relies on a precise scanning technique to produce very high-resolution, three

dimensional images of sample surfaces. The STM scans the sample surface beneath the tip
in a raster pattern while sensing and outputting the tunnelling current to the computer
control station. The digital signal processor (DSP) in the workstation controls the Z position
of the piezo based on the tunnelling current error signal. The STM can operate in both
constant height and constant current modes. The DSP always adjusts the height of the tip
based on the tunnelling current signal, but if the feedback gains are low the piezo remains at
a nearly constant height while tunnelling current data is collected. This is constant height
mode. With the gains high, the piezo height changes to keep the tunnelling current nearly
constant, and changes in piezo height are used to construct the image. This is constant
current mode. The exponential relationship between tip-sample separation and tunnelling
current allows the tip height to be controlled very precisely. For example, if the tunnelling
current stays within 20 percent of the set-point value (the current to be maintained by the
feedback system), the variation in the tip-sample separation is less than 0.02 nm.

Background information
The Atomic Force Microscope (AFM) operates by scanning an AFM probe across a sample
surface. The AFM probe consists of a sharp tip at the end of a flexible cantilever that
protrudes from a holder plate also called a holder [Link] tip is typically pyramidal or
conical in shape and is four to five µm in height with a diameter at the apex of 10 to 20 nm,
and is positioned at the end of a cantilever which is typically 100 to 200 μm long. The AFM
probe is usually made from silicon or silicon nitride, with cantilever spring constants ranging
from 0.05 to 50 N/m depending on the AFM mode of operation being employed. Either the
probe or sample is mounted on a piezoelectric scanner which can move in the x,y, and z
directions, and is used to raster scan the probe across the sample surface to acquire an
image in 3 dimensions.

A schematic illustration of an atomic force microscope connected to a computer.

How does an AFM operate?

The AFM functions by scanning the sharp tip over a surface much the same way as in the
olden day record player needle did to produce music. The tip is held at the end of a
cantilever shaped like a diving board (figure_1). As the tip is repelled by or attracted to the
surface by the intermolecular interactions (forces) between the atoms of the tip and the
surface, the cantilever is deflected. The magnitude of deflection is monitored by a detection
system, commonly a laser beam that reflects off the end of the cantilever at an oblique angle
and onto a photodiode (figure_1). A feedback mechanism from the detector system to
adjust the piezoelectric scanner is used to maintain either a constant force or a constant
deflection (height) between the tip and sample surface, depending on the AFM set-up.

Imaging: Interactions

Interactions between the probe and sample

At very close tip-sample distances (a few angstroms), a strong repulsive force appears
between the tip and the sample surface due to the overlap of atomic orbitals. The repulsive
force increases as the distance between the tip and sample surface decreases.

As the probe tip scans the surface at this close distance, a feedback system to the
piezoelectric scanner raises and lowers the sample to keep a constant repulsive force
between the tip and the sample surface. A plot of this upward and downward motion (z), as
a function of the tip x - y position on the sample surface, provides a high-resolution image of
the surface topography. This mode of operation is called “contact mode”. When the
repulsive force is in place, the tip and sample are considered to be in 'contact'. High
resolution is possible because very small changes in separation (z) (see figure_below), lead
to large changes in force which is the parameter being monitored. However, for some
samples such as biological specimens or soft material such as some polymers, 'contact' of
the surface with the tip can damage the sample.

To overcome this problem, an alternative set-up is that the tip vibrates rapidly up and down
and only 'taps' or comes very close to the sample surface while at the bottom of its
oscillation. This mode, referred to as “tapping mode”, “semicontact mode”, or “dynamic
mode”, is the most common mode of operation as it prevents sample damage. However,
now changes in separation (z) lead to smaller changes in force (see figure_below), and hence
the resolution of images from tapping mode is not as high as images from contact mode.

“Non-contact” mode AFM uses longer range forces which are attractive. These attractive
forces occur at tip-sample surface distances of > 10 nm.
Various forces curves at play in AFM. The force curve for a spring (the cantilever) is a straight
line (purple). The force curve for interactions between two atoms (brown) is the normal
Lennard-Jones type curve*. The net force curve (blue) shows the changes in forces measured
by AFM.
* An explanation of the Lennard-Jones curve: over short distance (z separation) the forces are
repulsive due to overlapping electron orbits (e.g. Pauli repulsion); whereas at longer
distances, the forces are attractive e.g. van der Waals force or dispersion force).

What can you measure?

Measuring forces in the tip-sample space

Force spectroscopy

In addition to imaging the surface of samples, another major application of AFM is force
spectroscopy. Force spectroscopy involves the direct measurement of forces between the tip
and the sample surface as a function of the distance between the two. The result is called a
force-distance curve.

The deflection of the cantilever is monitored as it is moved towards and away from the
surface, and the deflection plotted as a function of piezoelectric displacement gives the
force-distance curve. The horizontal axis in such a curve is typically the vertical distance
moved by the sample stage. The vertical axis is the deflection of the cantilever as the tip is
moved towards the sample surface, contacts and pushes against the surface and then away
from the surface. The vertical axis can then be converted to force by multiplying the
deflection of the cantilever by the spring constant of the cantilever (Hooke’s law).

An example of a force distance curve is displayed in the figure_below. The blue curve is the
approach cycle and the orange curve is the retract cycle. The different sections of the curve
are defined as follows.

1. The probe is moving towards the sample surface and does not experience any force
and therefore no change in deflection. Therefore the curve is flat. When the tip
contacts the surface there is often observed a small snap-on section.

2. The probe is then pushed against the surface. This is known as the compliance
regime and can be used to determine the mechanical properties of surfaces.

3. The approach cycle is completed and the retract cycle begins.

4. The snap-off section of the retract curve. This can yield information on tip-sample
adhesion and has been used to determine bond strengths.

5. The tip has completely disengaged from the surface and the probe is now moving
away from the surface. It now experiences no surface forces and therefore the
deflection of the cantilever is zero and the curve is flat.
A typical force distance curve acquired on silicon in air

Calibration

In order to determine the force experienced by an AFM probe important calibrations must
be performed. The first is converting the photodiode output from mV or nA (depending on
the instrument manufacturer) to nm. This is usually done by performing force-distance
curves against a surface much stiffer than the cantilever. The slope of the force curve in the
compliance regime will then give the conversion factor or sensitivity. This is also sometimes
referred to as the inverse optical lever sensitivity or INVOLS. Surfaces that can be used for
such a calibration are typically mica, glass or silicon. For very stiff cantilevers sapphire is a
good choice. The calibration surfaces should also be flat and clean and an average of at least
10 force curves on at least 3 separate locations on the sample surface should be acquired.

The next calibration that must be performed is determining the AFM cantilever spring
constant. A large number of techniques now exist to calibrate this value. Methods are based
on:
 • Beam theory: These require knowledge of the cantilever dimensions and material
properties.

• Resonance methods: Require knowledge of the resonance frequency of the

cantilever and/or the thermal noise response of the cantilever.

Loading the cantilever with a known force

The appropriate method needs to be chosen and will depend on the type of cantilever used.
All AFM users endeavouring to calibrate AFM cantilevers must make sure they use all
appropriate corrections as many techniques require corrections due to, for example, the tip-
offset distance, cantilever angle to the surface, or tip-height.

Uses

The forces that can be measured include atomic bonding, van der Waals forces, single
molecule stretching and rupture forces. Force spectroscopy can be used to measure the
dispersion force due to a polymer adsorbed on the surface of a substrate, or in biophysics,
the mechanical properties of biological materials.

Intro

AFM can operate in three modes according to how the AFM probe moves:

• Contact mode is also called static mode, and is where the probe does not oscillate
and is always in very close contact with the sample surface.

• Tapping mode is a dynamic mode also called intermittent contact or semi-contact

mode, and is where the probe oscillates such that the tip only comes into very close
contact with the sample at the bottom of its oscillation.

• Non-contact is also a dynamic mode where the probe oscillates above the sample
without the tip coming into very close contact with the surface.

• Contact mode
•
Contact mode/ constant force AFM operates by scanning the tip across the sample
 surface while monitoring the change in cantilever deflection with a split photodiode
detector. A feedback loop maintains a constant deflection between the cantilever and
the sample by vertically (z) moving the scanner at each (x,y) data point. The vertical
movement is used to build an image. By maintaining a constant cantilever deflection,
the force between the tip and the sample remains constant. Spring constants usually
range from 0.01 to 1.0 N/m, resulting in forces ranging from nN to μN in an ambient
atmosphere. Operation can take place in ambient and liquid environments.

•
•
Contact mode/ constant height AFM operates by keeping the height (z) of the scanner
constant, and using the cantilever deflection, as monitored by the split photodiode
detector, to directly generate the topographical image. Contact mode/ constant height
mode is most useful for atomic-scale images of atomically flat surfaces.
•
Contact mode is useful for obtaining 3D topographical information on nanostructures
and surfaces. A disadvantage of this mode is that, because the tip is in constant 'contact'
with the sample surface, large lateral forces can be exerted on the sample as the tip is
raster scanned over the surface resulting in damaged samples and deformed images.
• Tip choice for contact mode in air
•
Since the tip is in essence in contact with the surface, the spring constant of the
cantilever is not as stiff as for tapping mode operation in air. Spring constants tend to
be <1 N/m, and the cantilever geometry is generally V-shaped to minimize cantilever
twisting and buckling. Resonant frequencies for the cantilevers of contact mode probes
are usually between 10 to 80 kHz. In the figure_below, (a) Is an optical image of a
typical contact mode probe while (b) is an SEM image of a typical contact mode probe
showing the tip in more detail. The probes are usually made of silicon nitride and have
tip diameters of approximately 20 to 40 nm.
•
• (a) Optical image of tapping mode cantilever. (b) SEM image showing the tip in more detail
• Tapping mode
•
Tapping Mode AFM operates by scanning a tip attached to the end of an oscillating
cantilever across the sample surface. The cantilever is oscillated at or near its resonance
frequency with an amplitude ranging from 20 nm to 100 nm. The frequency of
oscillation can be at or on either side of the resonant frequency of the cantilever. The
tip lightly “taps” on the sample surface during scanning, contacting the surface at the
bottom of each oscillation. Changes in the vibration frequency of the cantilever
vibrating near its resonance frequency are measured. The feedback loop maintains a
constant oscillation amplitude, and therefore, a constant tip-sample interaction is
maintained during imaging.
•
Like contact mode, tapping mode can be operated in ambient and liquid environments.
Tapping mode has a number of advantages compared with contact mode, including
greatly reduced lateral and normal forces being applied to the sample surface.
Therefore, tapping mode is preferred to image samples with structures that are weakly
bound to the surface or soft samples. The figure_below is an example of a tapping mode
image in air of carbon nanotubes on a silicon surface.
•
• Tapping mode AFM image of Cr nanoparticles embedded in the surface of a compliant epoxy
substrate. The surface stress induces local curvature on the nanoscale. Image courtesy of Jitesh
Hora and Drew Evans, University of South Australia.
• Tip choice for tapping mode in air
•
The tip taps the surface and must not adhere or stick to the surface due to capillary
forces. The cantilevers are therefore designed to have spring constants, on average,
between, 2 N/m and 50 N/m. Resonant frequencies for tapping mode in air cantilevers
are usually between 50 to 400 kHz. In the figure_below, left is an optical image of a
typical tapping mode probe while in the right is an SEM image of a tapping mode probe
showing the tip in more detail. The probes are usually made of silicon and have tip
diameters of approximately 20 µm or less.

•
• (a) Optical image of tapping mode cantilever. (b) SEM image showing the tip in more detail.
• Non-contact mode
•
As the name implies, the (oscillating) tip is set further from the sample surface where
it is surrounded by attractive forces, and the tip does not contact the surface at any stage.
•
In this mode, magnetic, electrostatic, van der Waals, and other longer range forces can
be measured. Topography can be measured with some difficulty. The choice of constant
force versus constant height experiments is still possible.
• Data display
• How is the Data Displayed?
•
Most Scanning Probe experiments produce data as a function of the sample's x and y
position. This data can range from tunneling current, to force between sample and tip,
to height to reach a certain force etc. In the vast majority of cases, the data is displayed
as a false colour image. Typically high values of the parameters being measured are
assigned lighter colours while low values are given darker colours. An example is
shown in the following:

•
• This is a 3D image in the x, y and z planes. The z plane is shown as a graded colour from dark
(lowest) to light (highest).
• Image artefacts
• Tip convolution
•
Tip convolution is due to the radius of curvature of the tip being similar to or larger
than the width of the feature being imaged. As the tip moves across the feature, the
sides of the tip come into contact with the feature sooner than the apex of the tip, and
the detector and feedback system responds accordingly. The effect is to broaden the x,
y dimensions of the feature being imaged.

•
• An AFM artefact (tip convolution) arises from using a tip with a similar or higher radius of
curvature with respect to the feature which is to be visualized.
•
Tip convolution is an inevitable consequence of AFM imaging, but manufacturers of
AFM probes are continually making tips with smaller diameters and higher aspect
ratios. For the smallest diameter probes, silicon is currently the material of choice.
Manufacturers claim to be able to produce tips with diameters less than 10 nm while
the smallest silicon nitride tips are usually on the order of 20 - 30 nm in diameter.
•
The disadvantage of smaller diameter probes is that the pressure applied to the surface
and the tip increases for a given imaging force. This can potentially result in damage to
the surface or wear of the probe.
•
Carbon nanotubes promise to be able to produce the ultimate AFM tips. Their diameters
can be as small 1 - 2 nm and their aspect ratio (length/width) can be measured in the
thousands. This allows such probes to access parts of a sample ordinary AFM tips
cannot. Carbon nanotubes are also incredibly strong materials that do not wear like
silicon or silicon nitride. The disadvantage is attaching carbon nanotubes to AFM
probes. A number of methods are available but as yet no manufacturer is mass
producing carbon nanotube probes at a wafer scale.
• Thermal drift
•
Thermal drift, as the name suggests, is a distortion of the image due to changes in the
temperature of the probe-sample environment. AFM is sensitive to heat from external
sources, for example lights, and internal self-heating components. Thermal expansion
may result in the sample becoming 'loose' on the stage and moving as the sample is
scanned. Drift results in image distortions including features appearing smaller or
longer in one direction than they actually are.
• Tip artefacts
•
Artefacts result from the tip either becoming contaminated by material on the sample
surface or from wear due to scanning. The resultant image can possess a number of tip
artefacts and it is important for AFM users to recognize these artefacts.
• Worn tip
•
The tip may be worn down due to excessive imaging or using excessive force. A worn
tip reduces overall image quality, and may introduce odd shapes to the image.
•
• An AFM image showing oddly-shaped artifatcs due to a worn tip.
• Tip contamination
•
Tip contamination is due to the imaging tip 'picking up' material from the sample
surface. This is typically more of a problem on soft biological tissues. Tip
contamination can introduce strange, sometimes repeating, shapes to an image.
•
• An AFM image showing repeating blob-shaped artefacts due to tip contamination.
• Double tip
•
If the tip is damaged or picks up contamination, a tip with two or more sharp points
may form. This will show on the image as a doubling of surface features.
•
• A tip with a double point gives AFM images showing double images of surface features.
•
Debris on the surface is a good indicator of tip conditions since these are often spherical
in shape - images of these, with a good tip, should therefore also be nearly spherical.
Changing the scan direction should have no effect on the image artefact - to verify that
these features are due to the tip, the user may physically rotate the sample relative to
the tip and see if the artefacts have also rotated or not.
• Scanner artefacts
•
Piezoelectric elements in scanners are used to position the probe tip relative to the
sample surface at the nanometer scale with great accuracy. However, the behaviour of
the piezoelectric elements does not always meet the theoretical ideal, that is, a linear
increase in movement along with an increase/ decrease in applied voltage. The
movement of these piezoelectric devices is subject to deviations from the ideal
including non-linearity, hysteresis and creep.
•
Scanner sensitivity is the movement (nm/μm) per voltage (V).
•
Because of differences in the material properties and dimensions of each piezoelectric
element, each scanner responds slightly differently to an applied voltage.
• Nonlinearity
•
Nonlinearity means that the sensitivity of the scanner is not a linear function of the
(amount of) applied voltage. Typically, sensitivity is greater at larger applied voltages
than at near zero applied volts.
•
Nonlinearity of the x - y plane displays as surface features appearing stretched or
contracted at the top of the image, and then appearing to correct nearer to the middle
and bottom of the image. Some linear features may appear curved. This artefact will
typically not reduce when the same region is scanned multiple times. Although obvious
in images of linear structures like calibation grids, this artefact may go unnoticed in
images of samples lacking evenly spaced features.

•
• Image of a calibration grid showing non-linearity in the x - y plane.
• Hysteresis
•
Hysteresis occurs in piezoelectric elements in the scanners when the response traces a
different path depending on the direction of a voltage change. The magnitude of the
effect depends on many variables including the starting voltage, the size of the voltage
change, the rate of voltage change, and the scan angle.
•
In AFM, rastering across the sample surface (driven by voltage changes to the
piezoelectric elements) means the scanner will move non-uniformly in the x - y plane
if hysteresis is present. This non-uniformity has several effects, including differences
in scanner sensitivity at the beginning and end of a scan. If the scanner moves further
than it should vertically, in the z plane, tall features will appear to have a very 'sharp'
edge in the image.
•
Hysteresis typically occurs when changing the scan size or the centre offset of the
image, and typically only affects the first 10 - 20 lines of an image. Hysteresis can be
reduced or eliminated by re-starting the scan, and/or scanning more slowly.

•
•
• The dashed line represents an ideal linear relationship between applied voltage and the movement
(extension) of the piezoelectric elements in a scanner. The red lines represent the actual nonlinear
relationship that usually exists in piezo-scanners. This non-ideal relationship leads to the scanner
artefacts of nonlinearity where scanner sensitivity depends on the size of the voltage, and hysteresis
where scanner sensitivity depends on the direction of voltage change.
• Scanner creep
•
Creep describes the continued motion of the scanner after a rapid change in voltage, for
example, when moving the scanning position. The piezo scanner does not respond all
at once; it moves the majority of the distance quickly (< 1 ms), then the last part of the
movement is slower. The slow movement is called creep. If large voltage changes occur
during scanning, then the slower movement causes a distortion of the image. Creep may
result in surface features appearing smaller or longer in one direction than they actually
are. Creep accounts for the initial lateral distortion apparent after zooming or moving
to a new scan area, and which settles out after several scan lines have been traced.

•
• The continued slow movement of the piezoelectric scanner after a rapid change in applied voltage
results in creep. Creep is measured as the ratio of the slow dimensional change to the fast
dimensional change.
•
To correct for scanner artefacts, some AFM machines employ a sensor to detect actual
movement of the scanner and compensation adjustments are made to the scanner drive
signal in real time; this is called a 'closed loop' AFM.
Scanning Tunneling Microscope

Atomic Force Microscope