THE

INDUSTRIAL

GRAPHITE

ENGINEERING

HANDBOOK
Important:

The information contained herein has been compiled by UCAR Carbon Company Inc. from what it
believes are authoritative sources, and it is offered solely as a convenience to its customers. While UCAR
believes that this information is accurate and factual as of the date printed, this information is intended to be
used only as a guide in providing general information with respect to the products mentioned; and, therefore, it
is not to be taken as a warranty or representation for which UCAR assumes legal responsibility, nor is it to be
understood as permission to practice any patented invention without a license.

Since the user’s specific use application and conditions of use are beyond the control of UCAR, UCAR
Carbon Company Inc. makes no warranty or representation regarding the results which may be obtained by
the user. It is the responsibility of the user to determine the suitability of any of the products mentioned for
the user’s specific application.

Copyright © 2001 UCAR Carbon Company Inc.

GRAFOIL is a registered trademark of Graftech Inc..

UCAR Carbon Company Inc.

12900 Snow Road
Parma, OH 44130
Chapter One

THE MANUFACTURE OF CARBON

AND GRAPHITE
Industrial Graphite Engineering Handbook
 TABLE OF CONTENTS

The Manufacturing of Carbon and Graphite

Introduction .................................................................................................................. 1-1
Filler Effect ................................................................................................................... 1-3
Particle Size ................................................................................................................. 1-3
Forming Method ........................................................................................................... 1-4
Extrusion and Molding ................................................................................................. 1-4
Isomolding ................................................................................................................... 1-5
Baking and Densification ............................................................................................. 1-6
Graphitization ................................................................................................................ 1-7
Purification ................................................................................................................... 1-9
Process ................................................................................................................. 1-10
Analytical Methods ...................................................................................................... 1-12
Concluding Comments................................................................................................ 1-13
The Manufacture of Carbon and Graphite

Introduction

The use of graphite in many diverse areas of industrial activity is due directly to a unique combination of physi-
cal and chemical properties. Graphite is a good conductor of electricity, a superior refractory, has high thermal
conductivity, high thermal shock resistance, excellent high temperature strength, excellent machinability, high
sublimation temperature, and a low modulus of elasticity. In certain applications the superior hardness of
carbon may be preferred over that of graphite. Therefore, there will be continual reference to both monolithic
carbon and monolithic graphite as specialty industrial materials.

The graphite used in industry is manufactured from carbon-base materials, rather than mined as the natural
substance. For this reason, it is frequently called “artificial graphite.” Manufactured graphite is superior to
natural graphite, in that it does not have the disadvantages of mechanical weakness and high ash content that
are commonly associated with natural graphite, and it can be made in a wide variety of monolithic forms.

Manufactured graphite is not one specific material, but a range of materials. Each is essentially pure carbon,
but each varies from the other in such things as the orientation of the crystallites, the size and number of
pore spaces, the degree of graphitization, etc. Consequently, there is a corresponding variation in physical
properties. Specific graphite grades have, therefore, been established on the basis of these controlled
characteristics. Each grade is adapted to particular areas of service, with quality and price limitations that are
dictated by the requirements of the particular application. Proposed new uses of manufactured graphite must
thus be considered from a knowledge of the many grades available, all the way from the lowest cost, coarse-
grained, relatively weak materials to the ultrafine-grained, strongest and purest materials, which require many
times the production effort.

This wide variation in graphite properties is due largely to controlled variation of the production process ---
the choice of raw materials, filler particle sizing, and in the case of the fine- and ultrafine-grained material the
degree of milling and classification, forming method, conditions of the first bake (or gas bake), densification,
graphitizing, and purification. In addition, the manufacturing procedures essential to the production of tonnage
quantities of many grades of graphite at a moderate cost introduce further property variations from piece-to-
piece and within-a-piece.

The typical steps in processing coarse, medium, fine, superfine, and ultrafine grain carbon and graphite are
outlined in Figure 1.1, and briefly discussed in the balance of this section.

1-1
Figure 1.1 Processing Specialty Graphites
Raw Materials: Petroleum Cokes, Binders, Coal Tar Pitches

Fine Grain Medium Grain Coarse Grain

<0.004 inches <0.16 inches >0.16 inches
(<100 µm) (<4 mm) (>4 mm)

Crushing, Milling, Crushing, Milling, Crushing, Milling,

and Sizing and Sizing and Sizing

Weighing Weighing Weighing

Mixing Mixing Mixing

Forming: Isostatic Extrusion Press Extrusion

Pressing Press Molding Press

Carbonization: Baking Baking Baking

Densification: Pitch Pitch Pitch

Impregnation Impregnation Impregnation

Conversion to Graphitization Graphitization Graphitization

Graphite:

Special Oxidation Retardant Oxidation Retardant Oxidation Retardant

Impregnations: Resin Resin

Finishing: Machining Machining Machining

Purity: Purification Purification

1-2
Filler Effects

Graphite can be manufactured from almost any organic material that leaves a high carbon residue when
heated in the absence of air. Petroleum coke, a petroleum refinery by-product of the delayed coking process,
is a major raw material filler. The filler raw materials typically consisting about 70%, by weight, of the final
graphite tend to dominate the properties of the final graphite, see Table 1.1. Delayed coke in large supply
and relatively free from impurities is usually calcined as a preliminary to the carbon and graphite manufacture
steps. Other important filler raw materials include calcined coal tar pitch, delayed cokes, carbon black, lamp
black, raw petroleum coke, and recycled artificial graphite.

To illustrate the effect of raw material, a number of common room-temperature properties of a series of
graphite lots is shown in Table 1.1. Each sample was identical in all respects -grade, manufacture, size -
except for the petroleum cokes used in the raw materials. The wide range in property values is then directly
attributable to the raw material. Varied filler raw materials permit the manufacturer to produce graphites with
specific properties by appropriate raw material selection combinations.
Table 1.1 Typical Room-Temperature Physical Properties of Extruded Graphite
C o ke
Property Source With-Grain Against-Grain
Specific Resistance, A 8.4 11.7
µΩm B 7.5 11.7
C 8.8 12
D 5.7 12
Thermal Conductivity, A 0.4 0.3
cal/cmoC sec B 0.4 0.32
D 0.56 0.29
Thermal Expansion, A 2.4 3.8
10-6/oC (RT-60oC) B 1.5 3.6
C 1.9 3.2
D 1.1 2.7
Young’s Modulus, A 1.43 0.95
1 0 6 p si B 1.49 0.88
C 1.54 1.07
D 1.99 0.79

Particle Size

In manufacturing graphite, the raw delayed coke is first calcined in large rotary or shaft kilns to shrink it and
to drive out the volatile content. The calcined coke is then broken down by crushers and sized into a series of
carefully controlled fractions. The coarsest fraction in extruded graphites may be over 1” (25mm). The finest
fraction termed “flour”, in extruded and molded grades, typically has a maximum particle size of 0.0083”
(210 µm). The flour is usually produced in a mill with a built-in classifier. In normal manufacture, selected
size fractions are recombined with coke flour to produce a dry aggregate. Recycle artificial graphite is
similarly prepared.

Isomolded grades are usually all-flour compositions. Flours are milled to controlled maximum and median
sizes using built-in or stand-alone classifiers. Typical maximum particle size varies from about 75 µm to about
5 µm, depending on the grade produced. Recycle artificial graphite is similarly prepared.

The proportion of fractions and the fraction size is varied, within limits, to control the properties of the end
product.

1-3
Forming Method
Extrusion and Molding
Prior to forming the pieces that are to be graphitized, the selected combination of dry aggregate and coal tar
pitch binder is heated during mixing to assure homogeneity. The mixture is then cooled and the warm mix
formed either by extrusion or molding. Forming is another variable that affects the final graphite properties.
The magnitude of the effects is leveraged by the type of coke, be it needle, sponge, or isotropic. When
the calcined needle (premium) coke is crushed or milled to obtain the dry aggregate, the individual particles,
although irregular in shape, tend to have one dimension larger than the other two. In other words, particles
shaped like splinters or needles, are obtained. The shapes are more distinct in the finer size fractions. The
pre-graphitic layer planes are preferentially aligned along the major axes of these splinter and needle-shaped
particles. At the other extreme, the isotropic cokes usually have no internal preferred orientation and have
more of a rounded cube shape. The sponge cokes exhibit intermediate levels of preferred orientation and
shape.

In forming, the long axes tend to take a preferred orientation - either with the direction of extrusion or
perpendicular to the direction of molding. The final graphite product retains the same pattern of grain orienta-
tion. The with-grain direction is parallel to the direction of extrusion in extruded graphite and perpendicular
to the direction of molding pressure in the molded piece (i.e. perpendicular to the major vertical axis). The
against-grain direction is perpendicular to the with-grain. Thus, with the typical anisotropic lattice arrangement
of graphite crystallites coupled with the level of preferred orientation in the starting coke there is a difference
in the property values measured “with-grain” (WG) and “against-grain” (AG). These effects are most marked
with the needle cokes which are used to make graphite electrodes for electric steel making. Intermediate
effects are evidenced in sponge coke graphites, see Table 1.2. With-grain and against-grain effects are
minimal when isotropic cokes are used.

Other factors influencing isotropy are the sizing of the coarser fractions in the dry aggregate and the ratio of
the coarse fractions to the fine fractions. These effects are illustrated in Table 1.3 for three extruded products,
all with a common base and manufactured under the same process conditions.

Table 1.2 WG and AG Physical Properties of Molded and Extruded Graphite Produced from a Sponge
Coke
Property Extruded Stock Molded Stock
Density, g/cc 1.64 1.75
Specific Resistance, µΩm
With-grain 8.6 9.6
Across-grain 16.2 13.2
Ratio 0.53 0.73
Thermal Expansion, 10-6/oC
With-grain 1.1 1.9
Across-grain 4.1 3.2
Ratio 0.27 0.6
Young's Modulus, 106 psi
With-grain 1.84 1.39
Across-grain 0.78 0.96
Ratio 2.4 1.45
Flexural Strength, psi
With-grain 3420 3550
Across-grain 2370 2750
Ratio 1.44 1.29

1-4
 Isomolding
In the case of isomolding, the hot mix of flour and coal tar binder pitch is cooled and milled into a molding
powder. The molding powder is loaded into a rubber bag and shaped at an elevated pressure in an isostatic
press.

Isostatic molding, where pressures during molding are equalized on all surfaces, tends to produce near
isotropic properties irrespective of the starting coke type. However, the WG and AG jargon is carried over
from molding so that the minor axis, usually horizontal, is referred to as WG, and the major axis (usually in
the vertical) referred to as AG. The isotropy of isostatically molded materials is higher than with any other
forming method, see Table 1.3.

Table 1.3 The Effect of Forming Method and Grain Size on Isotropy

Maximum Grain Size, Extruded Average Grain Size, Isomolded

Property Ratio, WG/AG 0.010 inches 0.060 inches 0.250 inches 0.001 inches 0.0002 inches 0.0008 inches
(Isotropy) (0.25mm) (1.52mm) (6.35mm) (25 µm) (2 µm) (2 µm)
Specific Resistance 0.51 0.79 0.85 0.9 1.0 1.0

Young’s Modulus 2.91 1.71 1.4 1.0 1.0 1.0

Thermal Expansion 0.34 0.63 0.65 0.8 1.0 1.0

1-5
Baking and Densification

The next processing step is baking in which the pitch binder in the formed piece is carbonized to develop an
infusible carbon bond between the filler particles. Baking is done by bringing the product to a temperature
of 750oC to 900oC in oil or gas-fired furnaces. The pieces must be carefully arranged and spaced, and
tightly surrounded by a granular carbonaceous packing material since the binder softens in the early stages
of the bake and the product becomes plastic. Then, as gases are released from the binder, the piece
expands. Therefore, the product must be well supported so that dimensional tolerances are retained. The
carbonaceous packing material also provides oxidation protection.

Large furnaces are commonly used for baking. A high degree of control and temperature uniformity is
necessary. These furnaces will range in size from 35,000 to 200,000 pounds product capacity. Since the load
is large and has poor thermal conductivity, slow baking rates are necessary to keep temperature differences at
a minimum. Baking cycles from 20 to 72 days, depending on furnace size and product mix, are not unusual.

In the course of baking the coal tar pitch binder is converted to binder coke. Approximately one-third of
the pitch binder is driven off in this destructive distillation process. The baked product may show a loss in
apparent density, and the development of porosity and permeability to liquids and gases.

The product after first baking is called carbon. It is characterized as being hard, abrasive, and difficult
to machine (diamond tooling is required). Compared to graphite, it also has poor thermal properties, poor
electrical characteristics, good mechanical properties, and is generally lower in cost.

Porosity and permeability can be reduced, and density and mechanical properties increased by pitch impreg-
nation and rebaking prior to graphitization. Impregnation is generally accomplished by vacuum/pressure
techniques, assuring complete penetration of the liquid pitch through the cross-section of all pieces. The
impregnant is converted to carbon in another baking step often referred to as rebaking. This rebaking
is typically accomplished in large furnaces similar in size and design to those used for the initial baking.
However, baking rates for rebaking are significantly faster because the artifact is already rigid, and it is a
minor carbonaceous component within the porosity which is converted to carbon. Whilst the impregnated
bulk artifact does not require support during the rebaking, protection from burning is essential. The cycle of
impregnation and rebaking is repeated until the desired properties are achieved.

1-6
Graphitization
The final step in graphite manufacture is the conversion of the carbon to graphite. This is known as graphitiz-
ing and requires a temperature in the 2600oC to 3300oC range. In graphitization, the tiny two-dimensional
crystallites already present in random arrangement in the baked carbon piece, rearrange to very large three-
dimensional graphite crystallites. The development of three dimensional order which heralds graphitization is
accompanied by abrupt changes in physical properties.

Tests have shown that the properties of any piece of graphite are directly dependent upon the highest
temperature reached in graphitization. In Figures 1.2, 1.3, and 1.4, the changes in electrical resistivity,
flexural strength, and thermal expansion that occur in a fine-grain isomolded material as a function of final
temperatures are shown.
Figure 1.2 Electrical Resistivity As a Function of Final Graphite Temperature
35

30
Specific Resistance, µΩm

25

20

15

Radial Direction
10 Axial Direction

0
500 1000 1500 2000 2500 3000
o
Final Graphite Temperature, C

Figure 1.3 Flexural Strength As a Function of Final Graphite Temperature

10000

9000

8000

7000
Flexural Strength, psi

6000

5000

4000

3000 Radial Direction

2000
Axial Direction

1000

0
500 1000 1500 2000 2500 3000
o
Final Graphite Temperature, C

1-7
Figure 1.4 Thermal Expansion As a Function of Final Temperature
4.0

3.8

Coefficient of Thermal Expansion x 10 -6 / oC

3.6

3.4

3.2

3.0
Radial Direction
2.8 Axial Direction
o
CTE = RT to 100 C
2.6

2.4

500 1000 1500 2000 2500 3000

o
Final Graphite Temperature, C

Within practical limits, higher temperatures and greater temperature uniformity throughout the furnace can be
obtained by increasing the electrical power input in the graphitizing operation. However, heat losses increase
exponentially with increasing temperatures, which means that each temperature increment used becomes
increasingly expensive. So, again, it is a question of economics with regard to product application in choosing
a graphite for particular uses.

1-8
Purification
Even after graphitization at around 3000oC most bulk commercial graphites contain small amounts of metallic
impurities. These impurities originate in the raw material calcined cokes, recycled graphites, and coal tar
and petroleum pitches. The total impurities may be up to 2000 ppm as measured by combustion ash, see
Table 1.4. The ash level will be higher if inorganic compounds were purposely added at the mixing stage
to control puffing.

In the carbon-rich environment many of the metal species are likely present as carbides, many of which are
extremely stable, see Table 1.5. Scanning Electron Microscopy reveals that high concentrations of some
metals can be present in the final graphite as micron-sized pools of apparently complex alloys.

The use of graphite as a moderator for thermal neutrons in water- and gas-cooled nuclear fission reactors
require high purity levels. Some impurities require removal because of their ability to absorb neutrons,
whilst others because of their undesirable radioactive derivatives. Amongst the former elements, boron is
noteworthy because it substitutes in the carbon lattice. There is a concomitant difficulty in expelling boron
from the lattice.

Table 1.4 Typical Room Temperature Properties of UCAR Grades

Group UCAR Largest Cross Section Grain Size,(1) A sh
Grade inches mm inches mm ppm
Extruded Medium Grain
C BY 24 600 0.03 0.8 800
CS 30 600 0.03 0.8 800
SLWX 24 600 0.03 0.8 <5
CHV 24 x 48 600 x 1200 0.06 1.7 1.50%
Molded Medium Grain
PGW 52 1300 0.03 0.8 1000
PGX 73 1825 0.03 0.8 700
Extruded Fine Grain
FGE 8.5 x 8.5 215 x 215 0.006 0.150 1000
FGEX 8.5 x 8.5 215 x 215 0.006 0.150 1000
Isomolded Fine Grain
ATJ 24 600 0.001 0.025 900
CGW 16 x 16 400 x 400 0.001 0.025 2000
Isomolded Superfine Grain
DM306 8 x 16 200 x 400 0.0002 0.005 2000
DM300 8 x 16 200 x 400 0.0002 0.005 1600
DM301 8 x 16 200 x 400 0.0002 0.005 1300
DM302 8 x 16 200 x 400 0.0002 0.005 1100
DM303 8 x 16 200 x 400 0.0002 0.005 1500
DM342 8 x 16 200 x 400 0.0002 0.005 1500
DM332 24 600 0.0002 0.005 600
Isomolded Ultrafine Grain
DM362 6 x 12 150 x 300 0.00008 0.002 1500
(1) For the extruded medium and fine grain and molded medium grain, the
grain size is the maximum; for all the isomolded, the grain size is the average.

1-9
Table 1.5 Molecular Weight and Boiling Points of Carbides

Metal Formula MW [Link]., oC [Link]., oC Name

Boron B 4C 55.29 2350 >3500 Boron carbide

Sodium Na2C2 70 ~700 Sodium carbide
Magnesium
Aluminum Al4C3 143.91 2100 decomp > 2200 Aluminum carbide
Silicon Si C 40.07 sublimes ~2700 Silicon carbide
Potassiuim
Calcium C aC 2 64.1 2300 Calcium carbide
Titanium TiC 59.89 3140 +/- 90 4820 Titanium carbide
Vanadium VC 62.95 2810 3900 Vanadium carbide
Chromium Cr3C2 180.01 1980 3800 Trichromium dicarbide
CrC 64.02 1550 Chromium monocarbide
Manganese Mn3C 176.83 1520 Manganese carbide
Iron F e 3C 179.55 1227 Iron carbide
Nickel Ni3C 188.08 Nickel carbide
Barium B aC 2 161.35 1780 Barium carbide
Strontium SrC2 111.64 1950 Strontium carbide
Zirconium ZrC 103.23 3540 5100 Zirconium carbide
Molybdenum MoC 107.95 2692 Molybdenum carbide
Mo2C 203.89 2687 Molybdenum dicarbide

References:
Merck Index, 11th Edition, 1989
CRC Handbook of Chemistry and Physics, 75th ed., 1994
PlenumPress Handbook of High-Temperature Materials, No. 1: Materials Index by P.T.B. Shaffer, 1964

Purification Process
Purification processes involving halogen gas purges either during graphitization, or as a subsequent step,
developed for the production of spectroscopic-grade electrodes for chemical analysis, were enhanced in the
1940’s to cope with monolithic nuclear blocks. Chlorine is an effective agent except for boron. Boron removal
to below 1 ppm requires fluorine which is generated in-situ by the reduction of either carbon tetrafluoride, or
chlorofluorocarbons. The reaction between graphite and a stream of neat fluorine is too virulent.

In considering the likely effectiveness of purification, it is helpful to be mindful of the volatility of chlorides and
fluorides, see Table 1.6 and 1.7, and to run the appropriate thermodynamic calculations.

Although the demand for the purification of monolithic graphite moderator blocks has waned, the 1990’s saw
a rising need both quantitatively, and qualitatively, from the semiconductor sector. A current typical limit of
10 ppm ash is met with high temperature chlorine flushes of machined parts. Interestingly, the machining
of nuclear reactor graphite components is commonly done in an appropriately “clean” environment after
purification.

Continuous improvement and new semiconductor materials pushing towards impurity levels a thousand times
lower (to 10 ppb) require more effective purification processes as well as the development of enhanced
analytical techniques.

1-10
Table 1.6 Molecular Weight and Boiling Points of Metal Chlorides

Metal Formula MW [Link]., oC [Link]., oC Name

Boron B2Cl4 163.47 -92.6 various Diboron tetrachloride

BCl3 117.19 -107 12.5 Boron trichloride
Sodium NaCl 58.45 804 1413 Sodium chloride
Magnesium MgCl2 95.23 712 1412 Magnesium chloride
Aluminum AlCl3 133.34 (Note 1) decomposes 262 Aluminum chloride
Silicon SiCl4 169.89 -70 59 Silicon tetrachloride
Potassium KC I 74.55 773 sublimes 1500 Potassium chloride
Calcium CaCl2 110.99 772 >1600 Calcium chloride
Titanium TiCl2 118.81 1035 Titanium dichloride
decomp. >
TiCl3 154.27 Titanium trichloride
500
TiCl4 189.73 -24.1 136.4 Titanium tetrachloride
Vanadium VCl2 121.85 Vanadium dichloride
VCl3 157.3 decomposes Vanadium trichloride
VCl4 192.75 -28 +/- 2 148.5 Vanadium tetrachloride
Chromium CrCl2 122.92 824 Chromous chloride
CrCl3 158.38 1152 dissociates >1300 Chromic chloride
Manganese MnCl2 125.84 58 (differing values) Maganese chloride
650 1190 (differing values)
MnCl3 161.3 slight decomp. Maganese trichloride
Iron FeCl2 126.76 674 1023 Ferrous chloride
FeCl3 162.22 300 316 Ferric chloride
Nickel NiCl2 129.61 1001 sublimes 973 Nickel chloride
C opper CuCl 99 430 1490 Cuprous chloride
CuCl2 134.45 (Note 2) Cupric chloride
Zinc ZrCl2 136.29 290 732 Zinc chloride
Barium BaCl2 208.27 963 1560 Barium chloride
Strontium SrCl2 158.52 875 1250 Strontium chloride
Zirconium ZrCl2 162.13 decomp. 350 Zirconium dichloride
ZrCl3 197.58 decomp. 350 Zirconium trichloride
ZrCl4 233.05 437 sublimes 331 Zirconium tetrachloride
Molybdenum MoCl2 166.85 decomposes Molybdenum dichloride
MoCl3 202.3 decomposes Molybdenum trichloride
MoCl4 237.75 decomposes volatile Molybdenum tetrachloride
MoCl5 273.2 194 268 Molybdenum pentachloride

References:
Merck Index, 11th Edition, 1989
CRC Handbook of Chemistry and Physics, 75th ed., 1994
PlenumPress Handbook of High-Temperature Materials, No. 1: Materials Index by P.T.B. Shaffer, 1964
Note1: Fumes in air; volatilizes without melting when heated in small quantities.
Note2: Partial decomposition >300oC to CuCl + Cl M Pt. (extrapolated) is 630oC. Melting point of 498oC
usually reported for this compound is the melting point of the CuCl + Cl2 mixture.

1-11
Table 1.7 Molecular Weight and Boiling Points of Metal Fluorides

Metal Formula MW [Link]., oC [Link]., oC Name

Boron B F3 67.82 -127.1 (differing values) Boron trifluoride

-100.4 (differing values)
Sodium NaF 42 993 1704 Sodium fluoride
Magnesium MgF2 62.32 1248 2260 Magnesium fluoride
Aluminum AlF3 83.89 sublimes 1272 Aluminum fluoride
Silicon S i F4 104.06 -90.2 sublimes -95.7 Silicon tetrafluoride
S i 2F 6 170.16 -18.7 -18.5 Hexafluorodisilane
Potassiuim KF 58.1 859.9 1505 Potassium fluoride
Calcium C aF 2 78.08 1403 2500 Calcium fluoride
Titanium TiF3 104.88 1200 1400 Titanium trifluoride
TiF4 123.9 >400 sublimes > 284 Titanium tetrafluoride
Vanadium V F3 107.95 ~1406 (differing values) Vanadium trifluoride
>800 (differing values) sublimes
V F4 126.95 decomposes 325 Vanadium tetrafluoride
V F5 145.95 19.5 47.9 (differing values) Vanadium pentafluoride
111.2 differing values)
Chromium CrF2 90.01 894 Chromous fluoride
CrF3 109.01 1100 sublimes 1100-1200 Chromic fluoride
CrF4 128.01 estimate 277 ~400 Chromic tetrafluoride
Manganese MnF2 92.93 856 Manganese difluoride
MnF3 111.93 stable to 600 Manganese trifluoride
Iron F eF 2 93.85 sublimes >1100 Ferrous fluoride
F eF 3 112.85 sublimes >1000 Ferric fluoride
Nickel NiF2 96.69 sublimes >1000 (in HF) Nickel fluoride
C opper CuF 82.54 980 sublimes > 1100 Copper (l) fluoride
CuF2 101.54 decomp.@950 Cupric fluoride
Zinc ZnF2 103.38 872 1500 Zinc fluoride
Barium B aF 2 175.36 1353 2260 Barium fluoride
Strontium SrF2 125.63 ~1400 2460 Stronium fluoride
Zirconium ZrF4 167.22 sublimes > 600 Zirconium fluoride
Molybdenum MoF6 209.95 17.5 35 Molybdenum hexafluoride

References:
Merck Index, 11th Edition, 1989
CRC Handbook of Chemistry and Physics, 75th ed., 1994

Analytical Methods
The chemical analysis of nuclear graphite traditionally consisted of combustion ash and wet chemical analysis
of selected elements, especially boron (see ASTM C560). Improvements in analysis techniques have allowed
for more completed analyses in a shorter period of time.

In a recent comparison of five modern instrumental methods for the analysis of halogen-purified graphite,
three suitable techniques were reported(1). Each technique has multi-element capability and detection limits
(DL) which vary by element. Direct current arc-optical emission spectroscopy (DCA-OES or “DC-arc”)
provides timely and cost effective analyses, but has an average (DL) of 0.2 ppm (200 ppb), the poorest DL
of the three techniques. Inductively coupled plasma-atomic emission spectroscopy (ICP-AES) has the highest
accuracy and acceptable DL of 0.04-3.0 ppm (40-300 ppb), but cannot analyze volatile elements such as
boron, phosphorus, or sulfur, as they are lost upon sample preparation. The most capable technique is glow
discharge-mass spectrometry (GD-MS) which can detect 75 elements with DL of 0.005-0.05 ppm (5-50 ppb).

1-12
Concluding Comments

Thus, it is seen that properties and uniformity of the final graphite product are greatly influenced by material
and manufacturing variables. Where product uniformity is important, and special properties are sought, the
manufacturing variables can be controlled, and product meeting certain specifications selected. The important
thing is to determine the graphite best suited to the application from the standpoint of properties and
economics.

The manufacture of artificial graphite is characterized by the following:

• Long processing cycles;

• Energy intensive with respect to raw materials and processes;
• Wide range of grades and property combination through appropriate selection of raw materials
and processing.

1-13
 References

1. Mayer, H.K., Proceedings of the 24th Conference on Carbon, 1999, Vol. 2, p. 644-645.
Chapter Two

TYPICAL GRADES AND ROOM

TEMPERATURE PROPERTIES
Industrial Graphite Engineering Handbook
 TABLE OF CONTENTS

Typical Grades and Room Temperature Properties

Introduction ................................................................................................................... 2-1
Extruded Medium Grain Graphites................................................................................ 2-2
UCAR Grades CBY, CS, SLWX, CHV.................................................................... 2-2
Molded Medium Grain Graphites .................................................................................. 2-2
UCAR Grades PGW and PGX............................................................................... 2-2
Extruded Fine Grain Graphites...................................................................................... 2-2
UCAR Grades FGE and FGEX.............................................................................. 2-2
Isomolded Fine Grain .................................................................................................... 2-3
UCAR Grades ATJ and CGW ................................................................................ 2-3
Isomolded Superfine Grain............................................................................................ 2-3
UCAR Grades DM306, 300, 301, 302, 303, 342, and 332 .................................... 2-3
Ultrafine Grain ............................................................................................................... 2-3
UCAR Grade DM362 ............................................................................................. 2-3
Typical Grades and Room Temperature Properties

Introduction
The worldwide consumption of about 150,000 tons of specialty bulk carbons and graphites is satisfied by a
wide variety of materials differing in billet size, grain size, property combination, and other characteristics.
In selecting a grade for a particular end-use many factors are considered to ensure the choice of the most
viable material for the application.

In this chapter, a number of frequently used carbon and graphite grades are briefly reviewed in general terms
in the following groupings:

• Extruded Medium Grain

• Molded Medium Grain
• Extruded Fine Grain
• Isomolded Fine Grain
• Isomolded Superfine Grain
• Isomolded Ultrafine Grain

The grain size of Specialty Carbons and Graphites is always important. For instance, there are numerous
applications where the machined graphite, in use as a die, directly affects the surface finish and value of
the manufactured product. Then again, the strength properties of carbons and graphites are directly related
to grain size.

Unfortunately, there is no agreed industry standard definition of grain size so the approach here is to use
the National Electrical Manufacturer’s Association (NEMA) definitions as set out in Table 2.1. The NEMA
definitions focus on the size of the largest particle in the starting ingredients. Hence, the extruded and
pressure molded graphites with the largest particles greater than 100 µm (0.004 inches) are appropriately
described in this Handbook as medium grain. The fine grain extruded graphites with a maximum grain size
of 150 µm (0.006 inches) should strictly be defined as medium grain, but are so distinct from the relative
coarseness of the medium grain grades that the NEMA definition has been stretched.

The isomolded materials are well-covered by the NEMA definitions for fine (maximum particle is 50 µm-100
µm, 0.002 inches - 0.004 inches), superfine (maximum particle is 10 µm-50 µm, 0.0004 inches - 0.002 inches),
and ultrafine (maximum particle is less than 10 µm, 0.0004 inches).

Table 2.1 Classification of Grain Size

Graphite material can be classified by grain size as follows:

a. Ultrafine (or micrograin graphite ) - less than 10 microns
maximum grain size;
b. Superfine - between 10 and 50 microns maximum grain
size;
c. Fine - greater than 50 but less than or equal to 100 micron
maximum grain size; or
d. Medium -greater than 100 microns maximum grain size.
The above classifications are based on size measuring
techniques of the initial raw materials. It should be noted that
the measuring techniques involved are not applicable to the
finished product.
"Semiconductor Graphite"" CB-4-1989, National Electrical
Manufacturers Association, Washington, DC, 1989"
For more detail see: Norley, J., Proceedings of the 24th
Conference on Carbon, 1999, Vol. 2, p. 646.

2-1
For convenience, typical properties of extruded and molded medium grain, extruded fine grain, and isomolded
fine, superfine, and ultrafine grain grades are summarized in English units in Table 2.2 and in SI units in Table
2.3. An indication of billet size in terms of largest cross-section is also given in the tables. The structure
of selected grades is highlighted using photomicrographs, see Figures 2.1 to 2.5. Please note that this
chapter is not a comprehensive compilation. UCAR Carbon Company offers over one thousand combinations
of grade, size, machined shape, purity, oxidation-retardant, and other post-treatments. Please contact your
UCAR Carbon Company representative for detailed information.

Extruded Medium Grain Graphites

UCAR Grades CBY, CS, SLWX, CHV

Grades CBY, CS, and SLWX are extruded graphites of good structural quality, having uniform medium grain
size throughout the range of sizes in which they are made. The maximum particle size is 0.03” (0.8 mm).
Grade CBY is not pitch-impregnated, while CS is pitch-impregnated. Grades CBY and CS are frequently
referred to as “Mold Grades” because they impart a smooth finish to castings. They are available in a variety
of rounds with diameters up to 31” (775 mm) and in several rectangular configurations. CBY and CS are
extensively used for handling, containment, and shaping hot metals, ceramics, and refractories.

Grade SLWX is a 24” extruded round, densified, and purified grade developed for use in silicon crystal
growing applications.

Grade CHV is a slightly coarser grade with a maximum particle size of 0.06 inches (1.5 mm) produced
as a large rectangle 24” x 48” (600 mm x 1200 mm) in cross-section for applications such as pressure
casting of steel.

For typical room temperature properties of these grades, see Tables 2.2 and 2.3.

Molded Medium Grain Graphites

UCAR Grades PGW and PGX

The grades PGW and PGX are uniaxially pressure molded graphites having a uniform medium-size grain
(0.03 inches, 0.8mm, maximum) throughout the range of sizes in which they are made.

Grade PGW, a molded version of grade CBY, is available in diameters up to 50.8” (1270 mm). This material
finds application as a casting mold, crucible, heater, and susceptor.

Grade PGX molded to diameters of 36” to 74” (900 mm to 1850 mm) is used whenever large rounds are
required such as chemical reactors, induction furnaces, and heat exchangers.

Typical room temperature properties of these grades are shown in Tables 2.2 and 2.3.

Extruded Fine Grain Graphites

UCAR Grades FGE and FGEX

The grades FGE and FGEX are economically priced extruded graphites in 8.6 inch (215 mm) squares with a
fine grain structure. The maximum particle size is 0.006 inches (150 µm). Grade FGEX is a denser version
of FGE. These grades are used for handling, containing, and shaping hot metals, ceramics, refractories,
and semi-conductor materials.

Typical room temperature properties are summarized in Tables 2.2 and 2.3. The fineness of the grain of the
FGEX is shown in the photomicrographs of Figure 2.1.

2-2
Isomolded Fine Grain

UCAR Grades ATJ and CGW

These isostatically molded graphites are produced using fine powders with a maximum particle size of 0.003”
(75 µm).

Grade ATJ is a high strength fine grain flagship graphite first introduced in the 1940’s that can be machined to
very close tolerances and sharp detail with a fine surface finish. It is available in a variety of sizes of rounds
up to 24” (600 mm), and in rectangles up to 18 inch x 18 inch (460 mm x 460 mm). Applications include dies
and molds, rocket motor nozzle and thrust inserts, and tooling for forming composite parts.

Grade CGW is a densified version of ATJ and is used where high strength, low permeability, and erosion
resistance are required.

Typical room temperature properties of ATJ and CGW are shown in Tables 2.2 and 2.3. The fine grain
structure of ATJ is shown in the photomicrographs in Figure 2.2.

Isomolded Superfine Grain

UCAR Grades DM306, 300, 301, 302, 303, 342, and 332
This family of isostatically molded carbons and graphites is produced using superfine powders with a
maximum particle size below 0.002” (50 µm) and an average particle size of 0.0002” (5 µm). All are capable
of an excellent surface finish. These grades are available in a variety of sizes up to 11 inch x 14 inch (275
mm x 600 mm) in cross sections.

Grade DM306 is a hard carbon of high strength for use in aggressive and demanding applications such as
hot glass bottle handling, permanent molds, cooling plates, and mechanical carbons. Grades DM300, 301,
302, and 303 are a group of general purpose graphites representing a continuum of properties. Applications
include continuous casting molds, silicon crystal growing, and fiber optic furnace parts.

Graphite grade DM342 is suitable where strength is a major requisite.

Grade DM332 is a graphite having reduced CTE which is ideal for silicon carbide coating. This grade is
available in rounds up to 24 inches (600 mm) and rectangles up to 11 inch x 24 inch (275 mm x 600 mm).

Typical room temperature properties of all these superfine grades are summarized in Tables 2.2 and 2.3. The
typical microscopic structure of grade DM302 is shown in Figure 2.3, and grade DM342 in Figure 2.4.

Ultrafine Grain

UCAR Grade DM362

This isostatically molded graphite is produced using ultrafine powders with a maximum particle size below
0.0004” (10 um). The average particle size is 0.00008” (2 um). This graphite represents the high end
of surface finish and can be machined to thicknesses below 0.005”. Applications include fine detail finish
EDM, and aerospace.

Typical properties and microscopic structure are shown in Tables 2.2 and 2.3, and Figure 2.5, respectively.

2-3
Table 2.2 Typical Room Temperature Properties for UCAR Grades, English Units

UCAR Largest Grain Bulk Specific

Cross
Group Grade Size,(1) Density, Resistance
Section
inches inches lb/ft3 10−4 Ω inches
Extruded Medium Grain
C BY 24 0.03 103 3.2
CS 30 0.03 107 3.0
SLWX 24 0.03 112 2.8
CHV 24 x 48 0.06 110 3.2
Molded Medium Grain
PGW 52 0.03 102 3.7
PGX 73 0.03 107 4.0

Extruded Fine Grain

FGE 8.5 x 8.5 0.006 105 2.8
FGEX 8.5 x 8.5 0.006 108 3.2
Isomolded Fine Grain
ATJ 24 0.001 110 4.6
CGW 16 x 16 0.001 113 4.6
Isomolded Superfine Grain
DM306 8 x 16 0.0002 108 14.2
DM300 8 x 16 0.0002 106 5.7
DM301 8 x 16 0.0002 112 4.9
DM302 8 x 16 0.0002 116 4.5
DM303 8 x 16 0.0002 115 5.5
DM342 8 x 16 0.0002 117 4.7
DM332 24 0.0002 113 4.4
Isomolded Ultrafine Grain
D B 362 6 x 12 0.00008 117 7.3
(1) For the extruded medium and fine grain and molded medium
grain, the grain size is the maximum; for all the isomolded, the grain
size is the average.

2-4
Table 2.3 Typical Room Temperature Properties for UCAR Grades, SI Units

UCAR Largest Grain Bulk Specific

Group Grade Cross Section Size(1) Density Resistance
mm mm g/cm3 µΩm
Extruded Medium Grain
C BY 600 0.8 1.64 8.5
CS 600 0.8 1.74 7.5
SLWX 600 0.8 1.80 7.0
CHV 600 x 1200 1.7 1.76 8.0
Molded Medium Grain
PGW 1300 0.8 1.63 9.6
PGX 1825 0.8 1.73 10.2
0.150
Extruded Fine Grain
FGE 215 x 215 0.150 1.68 7.0
FGEX 215 x 215 0.150 1.74 8.0
Isomolded Fine Grain
ATJ 600 0.025 1.76 11.7
CGW 400 x 400 0.025 1.80 11.7
Isomolded Superfine Grain
DM306 200 x 400 0.005 1.73 36
DM300 200 x 400 0.005 1.70 14.6
DM301 200 x 400 0.005 1.79 12.4
DM302 200 x 400 0.005 1.86 11.4
DM303 200 x 400 0.005 1.85 14.0
DM342 200 x 400 0.005 1.87 12.0
DM332 600 0.005 1.81 11.2
Isomolded Ultrafine Grain
DM362 150 x 300 0.002 1.88 15.5
(1) For the extruded medium and fine grain and molded medium grain,
the grain size is the maximum; for all the isomolded, the grain size is the
average.

2-5
 Figure 2.1 Optical structures of UCAR Grade FGEX

200x Bright Field 400x Polarized Light

Here is a visual of the porosity. The black area is Here is a visual of the solid. Focus on the texture
porosity. (The gray area is the solid). Notice the of the white and gray areas, and note the degree
size and distribution of the porosity. of fineness. The largest solid particles can
usually be distinguished.

2-6
 Figure 2.2 Optical Structures of UCAR Grade Isomolded Fine Grain ATJ

200x Bright Field 400x Polarized Light

Here is a visual of the porosity. The black area is Here is a visual of the solid. Focus on the texture
porosity. (The gray area is the solid). Notice the of the white and gray areas, and note the degree
size and distribution of the porosity. of fineness. The largest solid particles can
usually be distinguished.

2-7
 Figure 2.3 Optical Structures of UCAR Grade DM-302

200x Bright Field 400x Polarized Light

Here is a visual of the porosity. The black area is Here is a visual of the solid. Focus on the texture
porosity. (The gray area is the solid). Notice the of the white and gray areas, and note the degree
size and distribution of the porosity. of fineness. The largest solid particles can
usually be distinguished.

2-8
 Figure 2.4 Optical Structures of UCAR Grade DM-342

200x Bright Field 400x Polarized Light

Here is a visual of the porosity. The black area is Here is a visual of the solid. Focus on the texture
porosity. (The gray area is the solid). Notice the of the white and gray areas, and note the degree
size and distribution of the porosity. of fineness. The largest solid particles can
usually be distinguished.

2-9
 Figure 2.5 Optical Structures of UCAR Grade DM-362

200x Bright Field 400x Polarized Light

Here is a visual of the porosity. The black area is Here is a visual of the solid. Focus on the texture
porosity. (The gray area is the solid). Notice the of the white and gray areas, and note the degree
size and distribution of the porosity. of fineness. The largest solid particles can
usually be distinguished.

2-10
Chapter Three

MECHANICAL PROPERTIES

Industrial Graphite Engineering Handbook

 TABLE OF CONTENTS

Mechanical Properties
Bulk Density ................................................................................................................. 3-1
Tensile Strength ............................................................................................................ 3-1
Flexural Strength........................................................................................................... 3-1
Creep ................................................................................................................. 3-5
Compressive Strength .................................................................................................. 3-9
Impact Strength........................................................................................................... 3-10
Young’s Modulus, Modulus of Rigidity, and Poisson’s Ratio ....................................... 3-11
Permeability ............................................................................................................... 3-13
Coefficient of Static Friction ....................................................................................... 3-19
Machinability ............................................................................................................... 3-20
Hardness ............................................................................................................... 3-22
Stress-Strain Relationships......................................................................................... 3-25
Mechanical Properties

Bulk Density
Bulk density (or apparent density) is defined as weight per unit volume. The methodology for bulk carbon
and graphite is described in ASTM C559.

Average values for several typical manufactured carbon and graphite grades are listed in Tables 3.1 and 3.2.

In the graphite family itself, the level of bulk density has some significance other than being purely descriptive.
For example, with major material and manufacturing factors remaining reasonably the same, a higher appar-
ent density will be reflected in improved mechanical strengths and higher thermal conductivity. Such properties
as electrical resistance and permeability will be lowered. In nuclear applications, a shorter diffusion length will
be realized. The above effects are not necessarily directly proportional to bulk density changes.

Tensile Strength

Tensile strength is defined as the pure tensile force per unit area required to rupture a material by pulling
it apart. The room-temperature measurement is described in ASTM C565; measurement at elevated tem-
peratures up to 2400oC is covered in ASTM C749. Average room-temperature values of several typical
manufactured carbon and graphite grades are indicated in Tables 3.1 and 3.2.

The variation in tensile strength with temperature of isomolded fine grain UCAR grades ATJ, superfine grades
DM302, DM332, and ultrafine grade DM362 has been measured to give the relative value shown in Figure
3.1. Note the increase in tensile strength as temperatures increase. This is generally true of most graphites. At
about 2500°C, strength can be expected to increase 80 to 100 percent from room temperature values.

Figure 3.1 % Increase in Tensile Strength vs. Temperature for UCAR Grades ATJ, DM302, DM332,
and DM362
70

60

DM302
50
DM332
DM362
Increase, %

40

30

20

10 ATJ

0
0 500 1000 1500 2000 2500
o
Temperature, C

3-1
In a similar situation, good correlation has been found between the apparent density and tensile strength, as
shown by Figure 3.2(2). Conversion of data to a constant density results in a “normalized” curve, and this is the
one reported for UCAR Grade ATJ. This correlation, which holds for graphite derived from a common base
stock, is not necessarily valid when graphites of different base compositions are compared.
Figure 3.2 Bulk Density (at Room Temperature) vs. Tensile Strength (at 1200°C) for UCAR Grades
ATA, ATR, ATJ
1.85

1.80

1.75
Bulk Density. g/cc

1.70

1.65

1.60

1.55

1.50
2500 3000 3500 4000 4500 5000 5500
2
Tensile Strength, lb/in

Figure 3.3(1) compares the strength-to-weight ratio of graphite at high temperatures with that of alloys and
ceramics that are conventionally used in high-temperature applications.

Figure 3.3 Tensile Strength-to-Weight Ratios vs. Temperature for High-Temperature Materials
(including UCAR Grade ATJ)
4
10
2

Best High-Temperature Alloy

Short-Time Tensile Strength, lb/in

Pure Sintered Pure Sintered Alumina

MgAl2O 4
Specific Gravity

Grade ATJ Graphite

Sintered
BeO

3
10
0 1000 2000 3000 4000 5000 6000
o
Temperature, F

3-2
Table 3.1 Typical Room Temperature Properties UCAR Grades, English Units

Largest Grain Bulk Flexural Tensile Compressive Young's

Cross
Group UCAR Size(1) Density Strength Strength Strrength Modulus Permeability Hardness
Section
Grade inches inches lb/ft3 4-Point p si p si 1 0 6 p si Darcy Rockwell Shore
p si
WG AG WG AG WG WG AG WG AG
Extruded Medium Grain
C BY 24 0.03 103 2400 1600 1700 1150 5000 1.32 0.96 0.07 0.05 70R ---
CS 30 0.03 107 3500 2300 2600 1700 7250 1.74 1.2 0.03 0.03 83R ---
SLWX 24 0.03 112 3300 3000 2300 2200 8500 1.48 1.35 0.01 0.01 85R ---
CHV 24 x 48 0.06 110 2600 1900 1650 1150 5200 1.9 0.97 0.02 0.02 --- ---
Molded Medium Grain
PGW 52 0.03 102 --- 1960 --- 1255 4500 --- --- --- 0.17 --- 32
PGX 73 0.03 107 1600 1300 1600 1300 4350 1 0.7 0.04 0.05 --- ---
Extruded Fine Grain
FGE 8.5 x 8.5 0.006 105 3190 1740 2540 1015 6525 1.86 0.96 0.004 0.003 --- ---
FGEX 8.5 x 8.5 0.006 108 3625 2465 2900 1850 8410 1.75 1.12 0.002 0.001 --- ---
Isomolded Fine Grain
ATJ 24 0.001 110 4250 3740 9700 1.4 0.002 60L ---
CGW 16 x 16 0.001 113 4900 4500 11,200 1.55 0.0001 90L ---
Isomolded Superfine Grain
DM306 8 x 16 0.0002 108 7100 6000 22,000 1.6 <0.002 105H ---
DM300 8 x 16 0.0002 106 4450 3400 9,500 1.45 <0.006 37H 43
DM301 8 x 16 0.0002 112 6000 4400 12,500 1.6 <0.001 55H 50
DM302 8 x 16 0.0002 116 7500 6000 14,000 1.7 <0.0003 75H 60
DM303 8 x 16 0.0002 115 9100 7900 18,000 1.7 <0.0006 90H 75
DM342 8 x 16 0.0002 117 9200 8000 18,000 1.8 <0.0001 --- 68
DM332 24 0.0002 113 6200 4900 13,000 1.7 <0.0007 53H ---
Isomolded Ultrafine Grain
DM362 6 x 12 0.00008 117 12,000 9000 24,000 1.9 --- 75
(1) For the extruded medium and fine grain and molded medium grain, the grain size is the maximum; for all the isomolded, the grain
size is the average.

3-3
Table 3.2 Typical Room Temperature Properties UCAR Grades, SI Units

UCAR Largest Grain Bulk Flexural Tensile Compressive Young's

Cross
Group Grade Size,(1) Density, Strength, Strength, Strength Modulus Permeability Hardness
Section,
mm mm g/cm3 4-Point MPa MPa GPa Darcy Linear
MPa Scale
WG AG WG AG WG AG WG AG WG AG MPa
Extruded Medium Grain
C BY 600 0.8 1.64 16.5 11 12 8.0 35 --- 9.1 6.6 0.07 0.05 125
CS 600 0.8 1.74 24.0 16 18 12.0 50 --- 10.3 8.0 0.03 0.04 200
SLWX 600 0.8 1.80 22.8 20.7 15.9 15.2 58.6 --- 10.2 9.3 0.01 0.01
CHV 600 x 1200 1.7 1.76 18.0 13.0 11.5 8.0 36 --- 10.1 6.7 0.02 0.02 130
Molded Medium Grain
PGW 1300 0.8 1.63 --- 11 --- 8.7 31 --- --- --- --- 0.17 43
PGX 1825 0.8 1.73 11.7 10.3 11.6 9.5 30 --- 7.6 5.5 0.04 0.05 ---
Extruded Fine Grain
FGE 215 x 215 0.150 1.68 22 12 17.5 7 45 --- 12.8 6.6 0.004 0.003 200
FGEX 215 x 215 0.150 1.74 25 17 20 12.5 58 --- 12.1 7.7 0.002 0.001 310
Isomolded Fine Grain
ATJ 600 0.025 1.76 29.5 25.5 67 9.6 0.002 60
CGW 400 x 400 0.025 1.80 34 31 78 10.7 0.001 90
Isomolded Superfine Grain
DM306 200 x 400 0.005 1.73 49 41.4 151.7 11 <0.002 105
DM300 200 x 400 0.005 1.70 30.7 23.4 65.5 10 <0.006 35
DM301 200 x 400 0.005 1.79 41.4 30.3 86.2 11 <0.001 55
DM302 200 x 400 0.005 1.86 51.7 41.4 96.5 11.7 <0.0003 75
DM303 200 x 400 0.005 1.85 62.7 54.9 124.1 11.7 <0.0006 90
DM342 200 x 400 0.005 1.87 63.4 55.2 124.1 12.4 <0.0001
DM332 600 0.005 1.81 42.7 33.8 89.6 11.7 <0.0007
Isomolded Ultrafine Grain
DM362 150 x 300 0.002 1.88 82.7 --- --- 13.1 ---
(1) For the extruded medium and fine grain and molded medium grain, the grain size is the maximum; for all the isomolded, the
grain size is the average.

3-4
Flexural Strength
Flexural strength (modulus of rupture) is the maximum fiber stress that a material will withstand before rupture
in bending. The room-temperature measurement using four-point loading is described in ASTM C1025. The
alternative three-point loading (see ASTM C1161) yields values for flexural strength about 25% higher than the
four-point loading.(17) However, the three-point loading has the higher variance(17). Both loadings are affected
by specimen size, shape, and surface finish.(17) For elevated temperature methodology, see ASTM C1366.
Average room-temperature flexural strengths for several typical carbon and graphite grades using four-point
loading are listed in Tables 3.1 and 3.2.

The flexural strength of UCAR Grade ATJ graphite has been measured at elevated temperatures(3), to
show, as expected, that this increases markedly with increasing temperature, just as do the other strength
characteristics of graphite. The results are plotted in Figure 3.4.

Figure 3.4 Modulus of Rupture vs. Temperature for UCAR Grade ATJ Graphite Stressed Against-
the-Grain

8000
Flexural Strength, lb/in 2

6000

4000
0 500 1000 1500 2000 2500
o
Temperature, C

These results were measured on 0.5 inch x 0.5 inch x 8.0 inch samples, mounted horizontally in the center of
a 40 inch I.D. graphite tube furnace. A four-point loading method was employed, the piece resting on supports
six inches apart, with the stress applied from above on two rounded knife edges 1 inch apart. With graphite,
this has been found to give results more consistent than with single-point loading, although the values are
always of lower magnitude, by about 25 percent. The upper portion of the curve is shown as a dashed line,
since plastic flow is quite significant at 2500oC, and a more precise control of the rate of loading is needed
to improve the reliability of this last point.

3-5
Creep
Creep is defined as plastic flow under constant stress. Graphite displays extremely low creep at room
temperature, its flow characteristics being comparable to those of concrete. Graphite is not perfectly elastic,
even at room temperature, and yields plastically under stress as the stress is increased. The stress-strain
curves for graphite are not linear, and there is a residual strain in the material after stressing, which increases
as the maximum stress is increased. However, this plastic straining is not generally considered to be creep.
Genuine creep over periods up to several hours has been observed in graphite only at temperatures of
approximately 1700°C and above.

Carbon products that have been processed to temperatures of about 900oC show much more creep than
graphite, and at much lower temperatures. They are much stronger than graphite and have higher Young’s
Moduli at room temperature, but creep more easily at high temperatures. Creep in carbon has been observed
at a temperature as low as 1000°C. As the processing temperature of carbon is raised, its creep tends to
decrease. In general, a graphite displays less creep as the crystal habit becomes more highly ordered.

Typical tensile creep curves for UCAR grade ATJ graphite are shown in Figure 3.5. The curves are
characterized by an initial elastic strain, which is followed by a region of decreasing creep rate. In the relatively
short-time creep tests of 90 minutes, there is a region of steady-state creep in which the creep rate is constant
over approximately the last 30 minutes. In some graphites, however, the creep rate has been observed to
continue to decrease over a testing period of 6 to 7 hours. When the stress is removed from a specimen that
has been creeping for some time, there is an immediate recovery of the elastic part of the strain and a gradual
recovery of part of the creep strain which can extend for several hours. Mainly because of its behavior in
recovery, graphite can be considered to be a viscoelastic material at high temperatures.

Creep in graphite is strongly dependent on the grain orientation. As Figure 3.5 shows, the creep of UCAR
grade ATJ graphite is greater in the “against-the-grain” direction than in the “with-the-grain” direction. Figure
3.6 compares the tensile creep of two other graphites. For UCAR grade ZTA graphite, which has a higher
density and a much higher degree of anisotropy than UCAR grade ATJ graphite, the creep is very low in the
“with-the-grain” direction and relatively high “against-the-grain.” The creep of an isotropic, molded lampblack-
base graphite (UCAR grade CEP), having a relatively low density, is almost independent of the orientation of
the specimen relative to the direction of the applied force in molding. Lampblack-base graphite displays much
less creep than UCAR grade ATJ graphite under the same conditions of stress and temperature.

3-6
Figure 3.5 Tensile Creep of UCAR Grade ATJ Graphite
12

o
Temperature=2600 C
2
10 Stress=3200 lb/in
Bulk Density=1.76 g/cc

8 Against-the-Grain
Elongation, %

With-the-Grain
4

0
0 20 40 60 80 100

Time, min

Figure 3.6 Tensile Creep of UCAR Grade ZTA Graphite and a Lampblack-Base Graphite

6
ZTA, against-the-grain
2000 lb/in2
Bulk Density = 1.95 g/cc
5

4
Elongation, %

o
Temperature=2600 C

3
CEP
3200 lb/in2
Bulk Density = 1.58 g/cc
2

1
ZTA, with-the-grain, 4400 lb/in2
Bulk Density = 1.99 g/cc

0
0 20 40 60 80 100

Time, min

3-7
The stress dependence of the steady-state creep rate of UCAR grade ATJ specimens oriented “with-the-
grain” is shown in Figure 3.7. At constant temperature, the steady-state creep rate varies with the fourth
power of the stress. The same stress dependence was found for specimens oriented “against-the-grain.”
Temperature dependence of the steady-state creep rate of UCAR grade ATJ specimens oriented “with-the-
grain” is shown in Figure 3.8, and an activation energy for the creep process of 124 kcal/mole was calculated
from these results. “Against-the-grain” specimens yielded the same activation energy, even though under
identical conditions of temperature and stress the steady-state creep rate was 1.8 times as great as that of
the “with-the-grain” specimen. Tensile creep data of UCAR grade ZTA graphite and lampblack-base graphite
yielded the same activation energy.
Figure 3.7 Tensile Creep of UCAR Grade ATJ Oriented “With-the-Grain”
-2
10

C
o

27 C

C
o

o
00

C
00

00

C
o
29

28

00

00
26
-3

25
10
Creep Rate, min -1

C
o
00
24
C
o
00
-4
10

23

-5
10
1000 10000

Stress, Logarithmic Scale

Figure 3.8 Tensile Creep of UCAR Grade ATJ Oriented “With-the-Grain”

-2
10
2
Stress, lb/in
40
39 00
28 36 00 44
00 00 00

24
Creep Rate, min -1

-3
00
10
20
00

16
00

-4
10
12
00

-5
10
0.00030 0.00032 0.00034 0.00036 0.00038 0.00040

1/Temperature, (1/K)

3-8
The mechanisms for creep in graphite have not been definitely established. For practical purposes, creep of a
coke-base graphite decreases in the direction “with-the-grain” as the density increases and as the anisotropy
of the graphite increases. There is a corresponding increase in creep in the direction “against-the-grain.” A
lampblack-base graphite has relatively low creep in all directions.

Compressive Strength
Compressive strength is defined as the maximum strength that can be developed under a compressive
load, the load being referred to is the original cross-sectional area and expressed as load per unit area.
Compressive strength is not a fundamental constant of a material. Different unit values result from different
sizes and shapes of sample, different rates of loading, etc. Average compressive strengths measured using
ASTM C695 for several typical carbon and graphite grades are shown in Table 3.1 and 3.2.
The temperature variation of compressive strength of a medium grain graphite UCAR grade AGSR has been
measured using in-house methodology on samples one-inch in diameter to two inches long, cut from a 20-inch
diameter piece with the length taken both from the diameter of the piece (transverse) and from the length of
the piece (longitudinal)(4). The results are shown in Table 3.3, where the compressive strength of graphite is
seen to increase with increasing temperature.

The following relationship has been found to hold between crushing strength and the more commonly
measured flexural strength for a variety of miscellaneous graphite samples(5).
Crushing Strength
Flexural Strength = 2.07 (Range 1.61 to 2.90)

Table 3.3 Compressive Strength of 20-inch Extruded Graphite

Compressive Strength, psi

With-the-Grain Across-the-Grain
Direction a t 2 5 oC a t 1 6 0 0 oC
Longitudinal 2840 3260
Transverse 2450 3050

3-9
Impact Strength
The impact strength of a material is a measure of its ability to withstand a sudden blow, a property convention-
ally measured with a pendulum device, such as that described in ASTM test method D256. However, with
rigid plastics, there are many instances where the strengths so measured are not correlated with a practical
experience(6). A comparison of dynamic with static stress-strain behavior using oscillographic techniques has
shown that rupture (failure) under dynamic stress-strain conditions results from the same basic considerations
as those which govern static or flexural failure. Therefore, the conventional static flexural test technique can
be utilized in evaluating the energy required for fracture and the notch resistivity. These values are a much
better measure of the service characteristics of brittle materials under impact loading than is given by the
pendulum test methods.

For graphite, the static test data are found to give a much improved correlation of the impact behavior
as compared with pendulum test results(7). The work done to failure (the recommended measure of impact
strength) is given by the area under the stress-strain diagram. Since, with graphite, this curve approximates
a straight line, it can be shown that:
2
Impact Strength (ft-lb) = 0.00463 F
E
where:

F = Flexural Strength, psi

E = Young’s Modulus, psi

Impact strength values calculated using average F and E from the room-temperature property Tables 3.1 and
3.2 are given in Table 3.4 for UCAR grades CS and ATJ.

Another conventional measure of the static and impact behavior of brittle materials is “notch sensitivity,” which
is defined as the ratio of the strength of a 0.400 inch thick test bar to that of a 0.500 inch thick bar with a
0.100 inch deep notch cut into it. Because of the inherent high modulus of elasticity of carbon and graphite,
the impact strength is “notch” and shape sensitive. For this reason, the test values apply directly only to
specimens of identical geometry.

Table 3.4 Impact Strengths of UCAR Grades

Impact Strength, ft-lb

Grade and Size With-the-Grain Across-the-Grain
CS
3 - 9 inch, Diameter 0.018 0.016
10 - 24 inch, Diameter 0.018 0.014
ATJ
Rectangles 0.036 0.045

3-10
Young’s Modulus, Modulus of Rigidity, and Poisson’s Ratio
Young’s modulus of elasticity is a measure of the resistance to change in length per unit length of a sample
subjected to tensile or compressive stress. It is expressed as the ratio of the stress (load per unit area) to the
resulting strain (change in length per unit length). Typically, Young’s Modulus is estimated from measurement
of the sonic resonance in a specimen, see ASTM C747. An approximate value for the Young’s Modulus (sonic
modulus) is estimated from measurement of the sonic velocity in a specimen, ASTM C769.

The modulus of rigidity (or shear modulus) of a material is a measure of the resistance to angular deformation
which occurs when the material is subject to pure shear stress. It is expressed as the ratio of shearing stress
(load per unit area) to the resulting shear strain.

When a material is subjected to a tensile (or compressive) stress, there is a transverse contraction (or
expansion) which accompanies the longitudinal expansion (or contraction). The ratio of the transverse strain
to the longitudinal strain is a dimensionless number known as Poisson’s ratio. For graphite, which has a
nonlinear stress-strain relationship, Poisson’s ratio is taken as the ratio of the transverse to longitudinal strains
in the limit of zero strain. Three Poisson’s ratios may be given for graphite. When the stress is in the with-the-
grain direction, νww is the ratio of the transverse strain in the across-the-grain direction to the longitudinal
strain in the with-the-grain direction. When the stress is in the across-the-grain direction, νaw is the ratio of the
transverse strain in the with-the-grain direction to the longitudinal strain in the across-the-grain direction. Table
3.5 lists the Young’s moduli, modulus of rigidity, and Poisson’s ratios at room temperature of several grades
of molded graphites(13). Room temperature Young’s moduli for typical carbon and graphites are reported in
Table 3.1 and 3.2.

Table 3.5 Young’s Moduli, Modulus of Rigidity, and Poisson’s Ratios of UCAR Molded Graphites at
Room Temperature
Young's Modulus, 10-6 psi Modulus of Rigidity, 106 psi Poisson's Ratios

With-the-Grain Across-theGrain

Grade Ew Ea G νww νwa νaw

ATJ 1.40 1.40 0.57 0.10 0.16 0.11

ZTA 2.85 0.70 0.50 0.07 0.25 0.05

The Young’s moduli of isomolded UCAR grades ATJ, DM302, DM332, and DM362 have been measured as a
function of temperature in the range from 25 to 2700°C, see Figure 3.9. Young’s modulus was determined by
measuring the resonant frequency of the fundamental flexural mode of vibration of a graphite bar or rod.

3-11
Figure 3.9 Young’s Modulus vs Temperature for UCAR Isomolded Graphites
2.5

2.0

Young's Modulus, 10 6lb/in 2

ATJ
1.5

1.0
DM302
DM332
DM362
0.5

0.0
0 500 1000 1500 2000 2500 3000
o
Temperature, C

Single Crystal Elastic Constants

The elastic constants of the graphite crystal may be given either by a set of stiffnesses, cij, or a set
of compliances, sij. The components of stress σj are related to the components of strain εj through the
relationships:

6 6
σi = ∑ cij εj and εj =∑ sijσi
j=1 i=j

The values of the elastic constants, with their estimated uncertainties, are given in Table 3.6 (14). The Young’s
moduli of the single crystal are Ew = 1/S11 and Ea = I/S33. The modulus of rigidity of the graphite crystal,
corresponding to shear parallel to the basal planes, is G = I/S44 = C44.

Table 3.6 Elastic Constants of Graphite

(Units: Cij and 1 /Sij in 1011 dynes/cm2, Sij in 10-11 cm2/dyne)

C11 = 106 ± 2 S11= 0.0098 ± 0.0003
C 12 = 1 8 ± 2 S 12 = -0.0016 ± 0.0006
C13 = 1.5 ± 0.5 S 13 = -0.0033 ± 0.0008
C33 = 3.65 ± 0.10 S 33 = 0.275 ± 0.010
C44 = 0.40 ± 0.04 S 44 = 2.50 ± 0.25
C 66 = 4 4 ± 2 S 66 = 0.023 ± 0.002
1/S11 = 102± 3
1 /S33 = 3.65 ± 0.10
Poisson's Ratios:
vww = S12/S11 = 0.16 ± 0.06
vwa = S13/S11 = 0.34 ± 0.08
vaw = S13/S33 = 0.012 ± 0.003

3-12
Permeability

Liquid and gas permeabilities of unconsolidated media are well understood and remain simple in consolidated
media of high permeability below the velocity range in which critical or turbulent flow come into play(8).
However, as pore size becomes smaller, details of pore shape become increasingly important and gas
flow and liquid flow are not simply interrelated. In low-permeability graphites, liquid and gas flow should
be discussed separately. Gas flow in manufactured carbon and graphite is determined by ASTM C577 and
adequately described by Darcy’s Law, in the form:
A <p>
Equation (1) Qv = K L η ∆p

where K is Darcy’s constant, A and L are, respectively, the sample area (cm2) and length (cm), η the viscosity
in centipoises, <p> the average pressure and ∆p the pressure differential in atmospheres. K is thus defined
to be in Darcy units and the volume flow, Qv, is expressed in ml/sec at one atmosphere pressure. The room
temperature permeabilities for some commonly used graphites have been measured using nitrogen as the
fluid, (helium for the finer isomolded grades), and at small pressure gradients (Tables 3.1 and 3.2).

Graphite permeability is affected by temperature as a result of the change in viscosity of the test gas. Gas
viscosities increase as the temperature increases and, as indicated in Equation (1), there is a corresponding
reduction in the volume flow rate. Darcy’s Law is valid at elevated temperatures so long as the flow of the
test gas is predominantly viscous in nature. The effective pore size of the solid must be large compared to
the mean-free path of the gas molecules. Since the mean-free path increases with temperature, it is possible
to encounter diffusive flow at elevated temperatures. In such cases, Equation (1) is no longer adequate. The
pressure dependence of the flow rate is a reasonable indication that the test conditions lie in the region
where Darcy’s Law applies.

Gas flow in low permeability graphites is more complex than that defined by Darcy’s equation and may be
described as viscous flow with Slip(9). This type of flow is given by an equation of the form:

A
Qpv = F L ∆p

Where Qpv is the pressure volume flow in milliliter-atmospheres per minute, F is an admittance (ml-cm/min),
A and L are, respectively, the sample area (cm2) and length (cm) and ∆p the pressure differential (atm). The
admittance, F, in turn consists of two terms in which K0 and K1 are(10):

F = F0 + F1 <p> = K0
T
+ K1<p>
η
M

constants characteristic of the material, (strictly speaking, K0 is a weak function of pressure in the transition
region from viscous to diffusive flow.), T is the absolute temperature (K), M the molecular weight of the
gas, <p> the average pressure (atm), and η the viscosity (poise). The coefficient, K1, is similar to that
appearing in Darcy’s equation above. This relationship has been carefully verified both for various gases and
at temperatures up to 1000oC(11).

The constants K0and K1 may be expressed in terms of effective capillary dimensions:

3
K0 = 1.48 x 106 < Nr
x
> ml • min-1 • cm-1 • (K)-1/2
4
K1 = 23.9 x 106 < Nr
x
> ml • poise • min-1 • cm-1 atm-1

3-13
in which N is the number of pores of radius r and effective length xL, x being a tortuosity factor and L the
sample length as above. The angular brackets indicate that appropriate averages over the pore distribution
are to be taken. From the ratio K0/K1, an effective pore radius for flow may be computed, and hence an
effective factor N/x. Without more exact calculation the values are only qualitative, but give some idea of the
orders of magnitude involved. Parameters are given in Table 3.7 for two graphites. The fact that the effective
pore radius occurs at high exponents in K0 and K1 indicates the sensitivity of the admittance factor to small
changes in pore radius. Finally, it should be mentioned that BET surface area measurements in conjunction
with the porosimetry results indicate the tortuosity factor “x” is of the order two to ten and increases as the
effective pore radius decreases.

Table 3.7 Permeability Coefficients

Unimpregnated Impregnated
Units
Stock Stock
Fo (Helium) 25.5 1.53 ml•min-1•cm-1
F1 (Helium) 78.1 0.62 ml•min-1•cm-1•atm-1
Effective Pore Radius 3.2 0.42 Microns
4 6
Effective N/x 6 x 10 1.6 x 10 cm -2

Heterogeneous liquid flow is related in a complex way to both static and dynamic phenomena. If there were
purely liquid flow, the Darcy equation would hold:

Qv = K A <p> ∆p
L η

in which Qv, is the volume flow, h the viscosity, K’ a permeability coefficient similar to the K1 above for gas flow.
With liquid moving against a gas, however, it is necessary for the applied pressure to overcome the capillary
force arising from an interface arrested at sharp discontinuities in pore diameter. This defines a critical radius
rc for each pore and, until the pressure difference ∆p exceeds the value given by:
2γ cos Θ
∆p = rc
where γ is the surface tension, the pore cannot support flow. As a result, liquid permeability through the
material will vary sharply with the sample thickness, attaining a cutoff at which no flow will occur at reasonable
∆p. This behavior is difficult to measure due to the tendency toward bubble formation, but the effect is shown
in Figure 3.10 for water through an impregnated graphite. The S-shape of the curve correlates with the
transition from macro- to micro-pore structure. For water (θ = 0, γ = 72.8 dynes/cm), the critical radius is given
below, where ∆p is in atmospheres:

rc = 1.5 microns
∆p

which gives a value of rc = 0.15 µ at the cutoff with ∆p = 150 psi. Since the minimum pore radius decreases
sharply as the length increases, the sharp drop-off in K’ results.

3-14
Figure 3.10 Water Permeability vs. Sample Thickness at ∆p= 150 psi
1.2

1.0

Permeability Coefficient, K'

0.8

0.6

0.4

0.2

0.0
0.0 0.5 1.0 1.5 2.0

Sample Thickness, cm

At very small pore radii, wall friction may be more important than viscosity in determining the liquid flow
characteristics. In a practical sense, complete impermeability to liquid flow can be supplied by a sufficient
thickness of graphite still permeable to gases.

To achieve low-permeability materials, selected base materials are subjected to additional densifications. The
conventional pitch-type impregnation has the effect of increasing density while reducing permeability.

The effect of pitch impregnation on the mercury intrusion pore spectrum of graphites are shown in Figures
3.11 for UCAR grade CBY, and in Figure 3.12 for UCAR grade CS. The UCAR grade ATJ is shown in
Figure 3.13. Table 3.8 summarizes these investigations.

Table 3.8 Porosimetry Analysis of UCAR Grades(16)

ATJ CS C BY
Maximum Particle Size, inches. 0.001(1) 0.03 0.03
Bulk Density, g/cc 1.76 1.73 1.66
Median Pore Diameter, µm
Volume 1.714 4.347 4.866
Area 0.012 0.164 0.183
Average Pore Diameter, µm 0.040 0.953 1.238
(1)
Average Grain Size

3-15
Figure 3.11 Porosimetry - UCAR Grade CBY

Cumulative Pore Volume (cc/g) vs. Pore Diameter

100% = + 0.1293

0% 20% 40% 60% 80% 100%

|...................|...................|...................|...................|...................|
+300- x
+235- x
+184- x
+144- x
+113-
+88.1- x
+69.0- x
+54.0- x
+42.3- x
+33.1- x
+25.9- x
+20.3- x
+15.9- x
+12.4- x
+9.72- x
+7.61- x
+5.96- x
+4.66- x
+3.65- x
+2.86- x
+2.24- x
+1.75- x
+1.37- x
+1.07- x
+0.839- x
+0.657- x
+0.514- x
+0.402- x
+0.315- x
+0.247- x
+0.193- x
+0.151- x
+0.118- x
+0.0926- x
+0.0725- x
+0.0567- x
+0.0444- x
+0.0347- x
+0.0272- x
+0.0213- x
+0.0167- x
+0.0130- x
+0.0102- x
+0.00799- x
+0.00626- x
+0.00490- x
+0.00383- x
+0.00300-

3-16
Figure 3.12 Porosimetry - UCAR Grade CS

Cumulative Pore Volume (cc/g) vs. Pore Diameter

100% = + 0.1293

0% 20% 40% 60% 80% 100%

|...................|...................|...................|...................|...................|
+300- x
+235- x
+184- x
+144- x
+113-
+88.1- x
+69.0- x
+54.0- x
+42.3- x
+33.1- x
+25.9- x
+20.3- x
+15.9- x
+12.4- x
+9.72- x
+7.61- x
+5.96- x
+4.66- x
+3.65- x
+2.86- x
+2.24- x
+1.75- x
+1.37- x
+1.07- x
+0.839- x
+0.657- x
+0.514- x
+0.402- x
+0.315- x
+0.247- x
+0.193- x
+0.151- x
+0.118- x
+0.0926- x
+0.0725- x
+0.0567- x
+0.0444- x
+0.0347- x
+0.0272- x
+0.0213- x
+0.0167- x
+0.0130- x
+0.0102- x
+0.00799- x
+0.00626- x
+0.00490- x
+0.00383- x
+0.00300-

3-17
Figure 3.13 Porosimetry - UCAR Grade ATJ

Cumulative Pore Volume (cc/g) vs. Pore Diameter

100% = + 0.1244

0% 20% 40% 60% 80% 100%

|...................|...................|...................|...................|...................|
+300- x
+235- x
+184- x
+144- x
+113- x
+88.1- x
+69.0- x
+54.0- x
+42.3- x
+33.1- x
+25.9- x
+20.3- x
+15.9- x
+12.4- x
+9.72- x
+7.61- x
+5.96- x
+4.66- x
+3.65- x
+2.86- x
+2.24- x
+1.75- x
+1.37- x
+1.07- x
+0.839- x
+0.657- x
+0.514- x
+0.402- x
+0.315- x
+0.247- x
+0.193- x
+0.151- x
+0.118- x
+0.0926- x
+0.0725- x
+0.0567- x
+0.0444- x
+0.0347- x
+0.0272- x
+0.0213- x
+0.0167- x
+0.0130- x
+0.0102- x
+0.00799- x
+0.00626- x
+0.00490- x
+0.00383-
+0.00300-

3-18
Coefficient of Static Friction

In certain applications, graphite is used in contact with other surfaces in such a manner that the static friction
between the two surfaces becomes an important consideration. The actual degree of this friction is expressed
by the coefficient of static friction which is defined by:

µ= F
P

where F = the frictional force required to put one body in motion over another, P = the normal force with
which the two bodies are pressed together, and µ = the coefficient of static friction. The value of the coefficient
is dependent upon: (1) the composition of the two surfaces in contact; (2) the presence of any surface
contaminants or oxides; (3) the temperature existent before sliding occurs; (4) condition of the surrounding
atmosphere; and (5) the load on the surfaces.

In Table 3.9, average values of coefficients of static friction are given for monolithic graphite (UCAR grade
CS) in contact with a number of surfaces. The values presented are indicative of the range of frictional
coefficients for graphite. Generally, the finer the grain size, the lower the coefficient. As temperatures increase,
the coefficient also increases. Graphite to graphite static friction measurements on coarse grained electrode
graphite run between 0.85 and 0.15 depending on the machined finish.

Table 3.9 Static Values of Coefficients of Friction of UCAR Grade CS

Stationary Coefficient of Contact Surface
Material Sliding Material Friction Temp., oC Atmosphere
 
Graphite (UCAR CS) Graphite (UCAR CS) 0.15 25 Room Air
 
Graphite (UCAR CS) 18-8 Stainless Steel 0.35 204 Room Air
Carbon
Graphite (UCAR CS) Cold-Rolled Steel 0.24 204
Dioxide
The important characteristics of graphite pipe slides are the following:

• Low Friction Allow uniform thermal expansion in all directions

• Corrosion Resistance Bearing surfaces that are permanently lubricated
• Long Life, Durable Maintenance-free and lower costs
• Economical Lower installation costs
• Temperature Stability 1000oF air atmosphere;
3000oF inert atmosphere
• Thermal Stability No cold flow deformation

3-19
Machinability

Graphite is machined with ease into almost any conceivable shape and contour. As an index of machinability,
graphite is usually comparable to hardwood or aluminum. Graphite can be sawed, turned, milled, bored,
drilled, planed, shaped, ground, tapped, reamed, honed and lapped. Tools will not seize or bind, and no
buffs are left. Because of inherent characteristics, such operations as knurling and thread rolling cannot be
performed.

Standard machining equipment generally used in woodworking and metal machining is easily adapted to
graphite machining with only slight modification. Graphite may be cut rapidly, using cutting speeds employed
with hardwood; cutting speeds are not critical and surface rates as high as 3000 ft/min have been successfully
used. Generally, the limiting factors are the built-in range of the machine and tool and the ability to capture
the dust and chips. Feeds and speeds normally associated with machining metals can be dramatically
increased in machining graphite. It is not uncommon to run machinery at its rated capacity feed and speed
in machining graphite.

Common, ordinary tooling can be used. The choice of economic tooling will depend on such factors as the
number of pieces to be machined, tolerances, and surface finish. The most common and economical material
is hardened tool steel. It is followed in cost by hardened high-speed steel, the “Stellite” alloy group (solid or
tipped), tungsten carbide and diamond tools. Tungsten carbide and diamond are used in situations where long
machining runs and minimum tool wear is desirable, and for the hard carbon and hard graphite such as the
isomolded superfine and ultrafine grades.

With regard to tool design for machining graphite, the cutting and clearance angles agree closely with
woodworking practice. A top-rake angle is of no particular advantage.

Depth of cut and feed rates are almost infinitely variable and are, generally, determined by tolerance, surface
finish, and sturdiness of the machine tool. Feed rates may be as high as one-half inch per revolution,
depending on the depth of cut. On the other hand, depths of cut may range up to two inches depending
on feed rates. In general, it may be said that feed rates may vary over the entire range built into standard
metal shaping or woodworking equipment; the depth of cut will depend upon the feed rate and the strength
of the particular machine.

Since graphite is easily machinable, the use of lubricating oils to facilitate machining and reduce tool wear
is not required. It is recommended that graphite be machined dry, since oils, greases, etc., can act as
contaminants to the graphite.

Dust from synthetic graphite contains no significant amount of extraneous material and is classified as a
nuisance dust by the American Conference of Governmental Industrial Hygenists (ACGIH). Because of the
black color, it is more noticeable in the air and on the skin than most other industrial dusts. A threshold
limit value (TLV) of 10 mg/m3 has been established by the ACGIH. Prolonged or repeated breathing of dust
can result in pneumoconiosis. For nuisance dust concentrations above the TLV, use a NIOSH-approved
respirator.

Cold creams or soaps containing cold cream seem to be more effective in removing graphite from the skin
than are the abrasive cleansers and soaps, and they are certainly gentler to the skin.

3-20
In the operation of graphite machining facilities, the following steps should be taken to insure economical
and safe machining:

1 .Wear adequate eye protection.

2. Provide adequate dust collection. This may vary from the use of common shop vacuum canisters from
single machine collection to highly sophisticated central vacuum systems for a series of machines.
Graphite chips and dust should be collected as near the generating source as practical. This generally is
accomplished through collection guides and cups directed at vacuum collection. Generation of chips are
less of a nuisance and will generally not become airborne as will graphite dust or fines. Thus, machining
feeds and speeds will dictate the type of collection needed. Good housekeeping practices are always
recommended.
Where dust collection systems are not available or not adequate to facilitate dust collection, graphite has
been machined using a fine water mist to minimize airborne dust. This procedure has been effective
on short-run or temporary machining situations. The graphite should be adequately dried after any wet
machining.

3. Suitable hoods and guards prevent dust from getting into slides, gears, and feed screws. Since graphite
acts as a lubricant, it generally will create little, if any, wear of machine parts. However, normal precautions
in preventing excessive dust accumulations should be used.

4. All electrical equipment should be totally enclosed. Graphite dust is electrically conductive. Care should
be taken to properly ground electrical equipment used in the machining area. Highly sensitive NC and
CNC control systems should utilize positive pressure or vacuum systems to keep sensitive digital control
panels dust-free.

Surface Finish
The surface of machined coarse grain graphite is made up in part of voids or valleys related to fundamental
composition and plateaus carrying the tool marks where the peaks have been removed. A surface analyzer
which measures vertical variations in microinches is meaningless as an expression of machining smoothness
since it combines a variable basic stock factor along with a true machined surface factor. For the same
reason, plus the fact that a graphite surface is not reflective (other than in a few exceptional cases), an optical
method of surface flatness measurement is also not applicable.

It has been found practical to specify machined surface smoothness by another surface roughness method.
This is a comparative method, made by vision and feel, in which the machined surface is judged against
standard metal scales of roughness. Such scales are manufactured by a number of suppliers and are
divided into gradations of average roughness expressed in terms of microinches. Bearing in mind the nature
of the base graphite in terms of its inherent porosity, it is not difficult to distinguish by sight and feel the
characteristic machine cut.
Different machining operations from rough sawing to lapping obviously create extremes in roughness. For
any one operation, a range will be obtained, dependent on tool shape, feed, depth of cut, and sturdiness of
machine. Under normal, good shop practice and tool maintenance, it is possible to predict the expectancy of
the range to be realized in machined surface roughness.

Tolerances
It is not unusual to machine small graphite parts to tolerances as close as 0.0002 inch or less. This may
require higher strength stock, since graphite is more elastic than most commonly used metals and will spring
more under tool pressure. An example where such small tolerances are required is in the machining of
EDM electrodes.

3-21
Hardness
The hardness of a substance may be defined in various ways, depending on the method of testing that is
used and the use to which the results are to be put. In general, hardness is the ability to resist indentation,
abrasion, and plastic deformation.

Resistance to indentation under static loads is measured by the Brinnel, Rockwell (ASTM C748 & E18), Vick-
ers, Tukon, and Knoop methods, Rebound hardness or energy absorption under impact loads is measured
by the Shore Scleroscope (ASTM C886). Resistance to scratching is measured by the Moh scratch hardness
scale. The results of these methods are not translatable to one another and, to compare results, measure-
ments must be made by the same apparatus.

The Rockwell with its fifteen scales (A to V) is in common use. Carbon and graphite are usually on the
scale H (hardest, 1/8” ball, 60 kg load), L (1/4” ball, 60 kg load), or R (softest, 1/2” ball, 60 kg load). The
Rockwell scales are so constructed that no linear correlation exists between the scales. Formulae have been
proposed(12) which correlate measurements of hardness (N) using the 1/2”, 1/4”, and 1/8” balls, irrespective of
the load, L (kg). The load L and the hardness N (irrespective of scale) are converted using the formulae in
Table 3.10 to a Linear Scale Hardness, LSH, having the units of kg/m2.

The Rockwell Hardness for typical carbon and graphites are shown in Table 3.1, and LSH in Table 3.2.
Table 3.10 Conversion Formula for Rockwell and Linear Scale Hardness

Ball Diameter, inches Linear Scale Hardness

1/2 12.7(L)/(130 - N)
1/4 25.5 (L-4)/(130 - N)
1/8 52.8(L-8 )/(130 - N)
L = L o a d , kg
N = Rockwell Hardness
Variation of Hardness with Other Properties

Within any one piece of stock, the hardness is always linearly dependent upon the apparent density. This
correlation also is sometimes evident between different pieces of the same type and sometimes shows up
between different types of stock. A graph illustrating this relationship is shown in Figure 3.14.

3-22
Figure 3.14 Static Values of Coefficients of Friction of UCAR Grades
20

ATA, ATR, ATJ

LSH Hardness, kg/mm 2 CS

15

CBY
10

5
1.55 1.60 1.65 1.70 1.75 1.80
3
Apparent Density, g/cm

All properties which are significantly correlated with density will almost invariably be correlated with hardness
as well. These include specific resistance, flexural strength, machinability, and crushing strength. Graphs
illustrating the relationship between hardness and flexural strength and hardness and machinability are
shown in Figures 3.15 and 3.16. Table 3.11 shows the correlation of hardness with resistivity, and thermal
conductivity, of UCAR grade ATJ.

Table 3.11 Resistivity, Hardness, and Thermal Conductivity for Grade ATJ(15)

Resistivity, µΩm 9 10.4 12.8

Hardness, Rockwell H 30 38 50
Thermal Conductivity, W/mK
Room Temperature 131 119 102
3 0 0 oC 102 93 82
6 0 0 oC 80 71 63
1 0 5 0 oC 47 42 39
1 3 2 5 oC 40 38 4

3-23
Figure 3.15 Mean Flexural Strength vs. Linear Scale Hardness
5000

Mean Flexural Strength (psi)

All Stocks
4000

3000

2000

1000

0
0 5 10 15 20 25
2
Linear Scale Hardness, kg/mm

Figure 3.16 Mean Machining Force vs. Linear Scale Hardness

500

Force Required to
Mean Machining Force, g

Make a 0.010-in. Deep Cut

400 With a 0.018-in. Feed

300

200

100

0
0 2 4 6 8 10 12 14 16 18 20
2
Linear Scale Hardness, kg/mm

3-24
Stress-Strain Relationships
The stress-strain relationships of materials may be depicted graphically by plots of stress (load per unit area)
versus strain (change in length per unit length). For elastic materials, the stress-strain curves are linear and
reversible when the stress is cycled. The stress-strain curves of standard elastic materials do not deviate
from linearity until the yield strength is reached and plastic deformation occurs; plastic deformation is defined
as deformation which is not recovered when the stress is reduced to zero. The stress-strain relationships
of polycrystalline graphites are different than those of standard elastic materials; the stress-strain curves of
polycrystalline graphite are not linear, plastic deformation may be introduced at very low values of stress, and
the stress-strain behavior differs in tension and compression(13).

Representative of the manufactured polycrystalline graphites, some stress-strain curves for UCAR grade
ATJ stressed in compression at room temperature, are shown in Figure 3.17. The two plots in Figure 3.17
correspond to strains in two different directions: the longitudinal strain is the strain in the direction parallel to
the direction of the applied stress and the transverse strain is the strain in a direction perpendicular to the
applied stress. The compressive stress was applied in the with-grain direction, and the transverse strain is
the strain in the across-grain direction. The stress-strain curves which result from an increasing stress are
nonlinear. After the applied stress is reduced to zero, the graphite retains residual strains, and new curves
are obtained when the stress is reapplied. The stress-strain curves were obtained by cyclically increasing
and decreasing the applied stress. This procedure led to the observed hysteresis loops and showed that
the residual strain in each direction increased with the applied stress. The ratio of the transverse strain to
the longitudinal strain (Poisson’s ratio) of the envelope curves in Figure 3.17 is constant over the entire
range of stress.

Figure 3.17 Longitudinal and Transverse Stress-Strain Curves for UCAR Grade ATJ With-the-Grain.
Compression at Room Temperature
12 12

10 10
Stress, 10 3 lb/in 2

8 8

6 6

4 4

2 2

0 0
0.0 -0.4 -0.8 -1.2 -1.6 -2.0 0.2 0.1 0.0
Longitudinal Transverse
Strain, %

3-25
Stress-strain curves for UCAR grade ATJ stressed in tension at room temperature are shown in Figure 3.18.
These curves also show the longitudinal strain in the with-the-grain direction and transverse strain in the
against-the-grain direction. The stress-strain curves for the longitudinal strains are qualitatively similar in both
tension and compression. However, the stress-strain curve for the transverse strain in tension is quite different
from its counterpart in compression. The residual strains introduced by an applied tensile stress are positive
in all directions, i.e., the dimensions of the graphite are increased in all directions after a tensile stress has
been applied and removed.

Figure 3.18 Longitudinal and Transverse Stress-Strain Curves for UCAR Grade ATJ, With-the-Grain.
Tension Test at Room Temperature

4 4

3 3

Stress, 10 3 lb/in 2
2 2

1 1

0 0
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 -0.03 -0.02 -0.01 0.00 0.01
Longitudinal Transverse
Strain, %

Unlike the stress-strain curves of polycrystalline graphites, the stress-strain curves at room temperature for
tensile or compressive stresses applied parallel to the major axes of a good graphite crystal are linear and
reversible. However, the stress-strain curves for the graphite crystal which are obtained by shear parallel
to the basal planes are nonlinear and are qualitatively similar to the stress-strain curves of polycrystalline
graphites in tension and compression.

3-26
 References
1. Malstrom, C., Keen, R., and Green, L., J. Appl. Phys. 22,593 (1951).
2. Gorton, A. F. and Malmstrom, C., North American Aviation Co. Report, NAA-SR-67 (1950).
3. Stieber, H. C., Unpublished Data, National Carbon Company Development Laboratories, NCDL-I (1956).
4. Downey, H. A., Unpublished Data, National Carbon Company Research Laboratories, NCRL-6 (1950)
(ERC-293).

5. Currie, L. M., Hamister, V. C., and MacPherson, H. G., Proceedings of the International Conference on
the Peaceful Uses of Atomic Energy, Geneva, 1955; 8, 451 (New York, United Nations) 1956.

6. Welch, L. E. and Quackenbos, H. M. Jr., ASME Trans. 68, 547 (July 1946)

7. Hatfield, M. R., Unpublished Data, National Carbon Company Research Laboratories, NCRL-47 (1947)
(EC-47).
8. Carmen, P. C., Flow of Gases Through Porous Media, Academic Press, Inc., London(1956).

9. Loeb, L. B., Kinetic Theory of Gases, McGraw-Hill Book Co., Inc., New York (1934).
10. Dushman, Saul, Scientific Foundations of Vacuum Technique, John Wiley & Sons, New York (1949).
11. Hutcheon, J. M., Longstaff, B., and Warner, R. K., The Flow of Gases Through a Fine-Pore
Graphite, Conference On Industrial Carbon And Graphite, London (September 1957).
12. MacPherson, H. G., Unpublished Data, National Carbon Company Research Laboratories, TM-242
(1957).
13. Seldin, E. J., Stress-Strain Properties of Polycrystalline Graphites in Tension and Compression at Room
Temperature, Carbon, 4: 177 (1966).
14. Complete Scientific Report, Union Carbide Corporation to U.S. Atomic Energy Commission, Contact
AT-(40-1) 3237, April 1964-June 30, 1967.

15. Page, D. J., Unpublished Data, UCAR Carbon Co. Inc., Parma, Ohio (1987).

16. Page, D. J., Unpublished Data, UCAR Carbon Co. Inc., Parma, Ohio (1989).
17. Tiwari, R., Ball, D. R., Miller, D. J., and Hudgins, S.A., Proceedings of the 23rd Biennial Carbon
Conference, 1997, p. 274-275.
Chapter Four

THERMAL PROPERTIES

Industrial Graphite Engineering Handbook

 TABLE OF CONTENTS

Thermal Properties
Thermal Conductivity .................................................................................................... 4-1
Thermal Expansion ....................................................................................................... 4-8
Thermal Shock Resistance ......................................................................................... 4-11
Emissivity and Spectral Characteristics ...................................................................... 4-12
Specific Heat ............................................................................................................... 4-14
Vapor Pressure ........................................................................................................... 4-17
Thermal Properties

Thermal Conductivity
The thermal conductivity of a material is defined as the quantity of heat transferred by that material per unit
area per unit temperature gradient. This is written:

K = WL
A∆T

where K is the thermal conductivity, W is the heat transferred per unit time, A is the area at right angles to
the direction of heat flow, L is the length of sample along which the heat flows, and ∆T is the temperature
differential along that length. Usually, the laboratory units of thermal conductivity are cal-cm/sec cm2K or
watt/mK and the engineering units are Btu-ft/hr ft2°F. The conversion factors in Table 4.1 apply.

Measurements between room temperature and 1000oC are typically made using a laser flash thermal
diffusivity method such as ASTM C714.

Table 4.1 Thermal Conductivity Conversion Factors

cal-cm/sec cm2K W/mK Btu-ft/hr ft2ºF

cal-cm/sec cm2K 1 418.6 242
W/mK 0.002389 1 0.578
Btu-ft/hr ft2ºF 0.004135 1.73 1

Manufactured graphite has an unusually high thermal conductivity over a considerable temperature range
near room temperature. With-the-grain, it is comparable to aluminum, and against-the-grain, it is comparable
to brass. Unlike metals, however, graphite behaves as a dielectric solid in that the principal mechanism for
heat transfer is lattice vibrations or phonon waves. The electronic component so important with metals is much
less than one percent at ordinary temperatures with graphite. Table 4.2 shows typical thermal conductivities
of various materials.

Table 4.2 Typical Thermal Conductivities Measured At Room Temperature

Material W/mK
Aluminum 210
Brick 0.6
Bronze (High Copper Content) 65
Carbon 4-30
Copper (Electrolytic) 350
Firebrick 0.1
Graphite 70-200
Iron 75
Molybdenum 125
Nickel 90
Silicon Carbide 15
Silica Glass 2
Silver 420
Tungsten 165

The Weidemann-Franz ratio is useful as a means of relating the thermal conductivity with the more easily
determined electrical conductivity for metals. Figure 4.1 shows the measurements of this ratio for a number of
graphites made by several investigators(2,17). It is apparent that the W-F ratio is not constant with graphite.

4-1
Figure 4.1 Weidemann-Franz Ratio vs. Temperature for Typical Graphites
35
Natural Graphite

30

[(Kρ/T)10 6 volts2/degree2 ]
Weideman-Franz Ratio,
25

20
Pyrolytic
Filament
15

10
Coke-Base Graphite

5
CEQ (Lampblack)
0
0 50 100 150 200 250 300

Temperature, K

The W-F ratio for graphite is several hundred times the free-electron value. In general, samples that have the
highest electrical conductivities show the greatest variations from the free-electron value. Additional evidence
for phonon conductivity is that by the addition of the acceptor impurity, boron, to graphite, the electrical
properties can be varied over a wide range without measurable effect on the thermal conductivity(1).

The general features of the thermal conductivity variation with temperature are shown in Figure 4.2. This is a
composite curve drawn for illustration and does not represent an actual sample. The average crystallite size
can be computed roughly from the position of the maximum conductivity using the relationship:

K = (1/3 Cp) • Vg • λ

where Cp is the specific heat, Vg is an average group velocity, and λ is the average phonon wavelength. At the
maximum, λ approximates the average crystallite diameter. Such calculation can be roughly confirmed in the
crystallite size range where X-ray techniques can be applied.

4-2
Figure 4.2 General Nature of the Temperature Dependence of the Thermal Conductivity of Graphite
1.0

Thermal Conductivity, cal/sec cm2 K

0.1

0.01

0.001
10 100 1000 10000

Temperature, K
At temperatures below the maximum conductivity, where the mean-free path is limited by the crystallite size,
the thermal conductivity of many manufactured graphites has a greater temperature dependence than does
the specific heat(2,3). The temperature dependence of the thermal conductivity in manufactured graphite can
be interpreted(4) by considering the interconnections between crystallites as having a T3 variation, while the
crystallites are considered to have a T2 dependence. Although this model is over-simplified and has defects in
its details(28), the amount of nongraphitic material necessary for its success is roughly in agreement with that
found to be present by measurements of magnetic susceptibility.

Thermal conductivity variation with temperature has been measured by a number of investigators(5,7). The
general shape of the thermal conductivity curve with temperature is shown in Figure 4.3. Throughout this
temperature range, above the maximum shown on Figure 4.2, the thermal conductivity falls off roughly as T-1.
As the vaporization temperature of graphite is approached, the thermal conductivity drops sharply(6,7). Figure
4.4 shows the general shape of the thermal conductivity curve for the UCAR isomolded grades ATJ, DM302,
DM332, and DM362 to 1000ºC.

4-3
Figure 4.3 Thermal Conductivity - Percent of Room Temperature Value with Temperature Increase
110

100

Relative Thermal Conductivity, R.T.=100%

90

80

70

60

50

40

30

20

10

0
0 500 1000 1500 2000 2500 3000
o
Temperature, C
Figure 4.4 Thermal Conductivity of Isomolded UCAR Grades ATJ, DM302, DM332, and DM362
125

ATJ
DM302
Thermal, Conductivity, W/m oC

100 DM332
DM362

75

50

25
0 200 400 600 800 1000
o
Temperature, C
Although a number of thermal conductivities at room temperature and above for manufactured graphite are
reported in the literature, few of the references give sufficient information on the other properties of the
graphite to characterize it properly. The thermal and electrical conductivities of a number of identified graphite
grades are given in Table 4.3 where it is noted that within the accuracy of the measurements, the ratio
between the two is fairly constant. In spite of the wide variation in stock size, mix particles, and pitch
treatment represented by these graphites, the ratios of the two conductivities shown in the last column remain
remarkably constant. A recalculation(8) of the data in Table 4.3 converting all the thermal conductivities to
25oC, yields a value of 0.00030 for the ratio of the thermal conductivity in cal-cm/sec cm2 K to the electrical
conductivity in mho/cm. This product has no relation to the Weidemann-Franz law, because the thermal
conductivity is not electronic in nature.

4-4
Table 4.3 Thermal Conductivity (K) of Graphite (With-the-Grain)
Diameter of
Original Maximum Size Pitch Mean Temperature Thermal Electrical
Extruded Particle in Mix, Impregnated of K Determination, Conductivity, K Conductivity,σ (K/σ) x 103
P i e ce ,
Source inches ºC cal-cm/sec cm2K mho/cm
inches
A 18 18 No 58.8 0.268 917 0.292
A 18 18 No 58.0 0.258 850 0.304
A 12 1/16 No 57.4 0.274 897 0.305
A 3 1/32 No 54.0 0.324 1098 0.295
A 3 1/32 Yes 47.3 0.428 1508 0.284
A 4 1/32 No 59.3 0.257 894 0.287
A 4 1/32 No 50.8 0.382 1337 0.286
A 2 Flour Only No 45.7 0.357 1209 0.295
A 2 Flour Only Yes 46.2 0.37 1269 0.292
B 18 1/8 No 53.3 0.341 1185 0.288
B 12 1/16 No 52.7 0.355 1264 0.281
B 4 1/32 No 51.7 0.364 1241 0.293
B 2 Flour Only No 46.1 0.400 1462 0.274
B 2 Flour Only Yes 42.0 0.510 1832 0.278
C 12 1/16 No 56.8 0.353 1236 0.286
C 4 1/32 No 49.5 0.415 1453 0.286
C 2 Flour Only No 45.1 0.397 1385 0.287
C 2 Flour Only Yes 44.9 0.375 1205 0.311
D 2 Flour Only No 42.6 0.419 1462 0.286
D 2 Flour Only Yes 46.9 0.409 1393 0.294
E 2 Flour Only Five Times 40.1 0.451 1529 0.295

The anisotropy of the thermal and electrical conductivities has also been reported by others(9). Some general
measurements of thermal conductivity so reported are given in Table 4.4, but the accurate determination of the
constancy of the ratio of thermal and electrical conductivities was accomplished by measuring both properties
on a number of individual samples of widely varying conductivity. The ratio of conductivities so determined
is 0.00032 when both are measured at room temperature. The ratio is general enough in validity to justify
the use of the relationship in calculating the room temperature thermal conductivity of graphite from the room
temperature electrical conductivity. The latter property is easy to measure accurately. If the relation:

K (cal-cm/sec cm2K) = 0.00031 σ (mho/cm)

is used for measurements at 25oC, the thermal conductivity so calculated will be accurate within ±5 percent.

The thermal conductivity data for several typical carbon and graphite grades is shown in Tables 4.5 and 4.6.

Table 4.4 Thermal Conductivity of Reactor Graphite Measured at Room Temperature

Thermal Conductivity, Btu-ft/hr ft2oF

Coke Source Parallel to Grain Perpendicular to Grain
B 0.56 0.29
C 0.4 0.3
D 0.44 0.32

4-5
Table 4.5 Typical Room Temperature Properties of UCAR Grades, English Units

Group UCAR Largest Cross Section Grain Size(1) CTE(2) (RT-212°F) Thermal Conductivity
Grade inches inches 10-6/°F Btu · ft / hr · ft2 · °F
WG AG WG AG
Extruded Medium Grain
C BY 24 0.03 1.5 1.9 85 75
CS 30 0.03 1.5 2.0 90 85
SLWX 24 0.03 2.0 2.2 100 85
CHV 24 x 48 0.06 0.6 1.3 80 60
Molded Medium Grain
PGW 52 0.03 --- 2.0 --- 75
PGX 73 0.03 1.2 1.4 72 60
Extruded Fine Grain
FGE 8.5 x 8.5 0.006 1.0 2.3 100 59
FGEX 8.5 x 8.5 0.006 1.2 2.2 88 59
Isomolded Fine Grain
ATJ 24 0.001 1.4 68
CGW 16 x 16 0.001 1.4 71
Isomolded Superfine Grain
DM306 8 x 16 0.0002 3.1 12
DM300 8 x 16 0.0002 2.5 49
DM301 8 x 16 0.0002 2.2 60
DM302 8 x 16 0.0002 2.4 70
DM303 8 x 16 0.0002 2.6 50
DM342 8 x 16 0.0002 2.3 64
DM332 24 0.0002 2.0 64
Isomolded Ultrafine Grain
D B 362 6 x 12 0.00008 2.8 47
(1) For the extruded medium and fine grain and molded medium grain, the grain size is the maximum;
for all the isomolded, the grain size is the average.
(2) CTE = Coefficient of Thermal Expansion

4-6
Table 4.6 Typical Room Temperature Properties of UCAR Grades, SI Units

Group UCAR Largest Cross Section Grain Size(1) CTE(2) (RT-100°C) Thermal Conductivity
Grade mm mm 10-6/°C W/mK
WG AG WG AG
Extruded Medium Grain
C BY 600 0.8 2.5 3.4 145 120
CS 600 0.8 2.5 3.6 175 145
SLWX 600 0.8 3.4 3.9 170 150
CHV 600 x 1200 1.7 1.5 2.6 140 105
Molded Medium Grain
PGW 1300 0.8 --- 2.5 --- 130
PGX 1825 0.8 2.2 2.6 125 104
Extruded Fine Grain
FGE 215 x 215 0.150 1.8 4.1 170 100
FGEX 215 x 215 0.150 2.1 4.0 150 100
Isomolded Fine Grain
ATJ 600 0.025 2.5 115
CGW 400 x 400 0.025 2.5 120
Isomolded Superfine Grain
DM306 200 x 400 0.005 5.5 20
DM300 200 x 400 0.005 4.5 84
DM301 200 x 400 0.005 4.2 105
DM302 200 x 400 0.005 4.3 120
DM303 200 x 400 0.005 4.8 90
DM342 200 x 400 0.005 4.3 110
DM332 600 0.005 3.7 112
Isomolded Ultrafine Grain
DM362 150 x 300 0.002 5.1 80
(1) For the extruded medium and fine grain and molded medium grain, the grain size is the maximum;
for all the isomolded, the grain size is the average.
(2) CTE = Coefficient of Thermal Expansion

4-7
Thermal Expansion
The coefficient of thermal expansion of (CTE) a material is the increase in length per unit length per degree
rise in temperature. For methodologies see ASTM E228 and References 110 and 111. Typical coefficients for
some common graphites are listed in Tables 4.5 and 4.6.

The variation of the thermal expansion coefficient of graphite with temperature is most accurately measured
by the interferometer method in the 0 to 600oC range(10,11). The results of these interferometer mean-
measurements for three very different types of graphite, between room temperature and the final temperature
indicated, are shown in Figure 4.5.

Figure 4.5 Mean Thermal Expansion Coefficient vs. Temperature (Interferometer Method)
/C

25 50
-7 o
Mean Thermal Expansion Coefficient, x10

te
ra phi
20
s eG 45
Ba
bl ack
p
Lam te
ra phi
15 40
aseG
-B
C oke
rol eum te
10 Pet ra phi 35
AG
R
e dG
at
y Tre
ec ialt
Sp
5 30

0 25
0 100 200 300 400 500 600 700 800
o
Final Temperature, C

The points at the higher temperatures are more accurate since their calculation involves a wider temperature
interval. They have been weighed more heavily than the low temperature points in drawing the curves. The
slope of the curves is approximately the same for all three graphites, as well as many others tested.

Thermal expansion measurements at temperatures up to 2500ºC have been made by the optical dilatometer
method on the UCAR extruded graphites(12) L117, AWH, AGR, and ECA, and more recently on the UCAR
isomolded graphites DM302, DM332, and DB362. This was done by mounting 5 to 12 inches long samples
in a 4 inch I.D. graphite resistance tube furnace. Length changes were determined by the telemicroscopic
observation of pointed tungsten probes mounted 10 inches apart in the sample. An apparatus of considerable
physical and thermal stability was required, giving measurements with a precision of +0.002 inches expansion
and ±5oC.

Figure 4.6 shows the results measured parallel to the axis of the extruded grades. Small obvious irregularities
of expansion occur in individual samples, probably as a result of irregular release of internal stresses.
Nevertheless, the parallel nature of the curves is consistent with the idea that the change of the coefficient of
thermal expansion with temperature is approximately the same for all graphites at high temperatures.

4-8
Figure 4.6 Mean Coefficient of Thermal Expansion of UCAR Graphite vs. Temperature
(Optical Dilatometer Method).

/C
-7 o
80
)
ck (Long.

Mean Thermal Expansion Coefficient, x10

Base-Sto
mpblack
L-117, La DM362

60 DM302

DM332

g.)
tock (Lon
e-Base S
allu rgical Cok
40 AWH Met
g .)
ck (Lon
ase Sto
d Petroleum-B ng.)
o arse -Graine ase S tock (Lo
AGR, C Coke-B
ed Petroleum
20 in e-Grain
ECA, F

0
0 500 1000 1500 2000 2500
o
Final Temperature, C

This leads to a simplification, for engineering purposes, in the calculation of high temperature thermal expan-
sion. The mean linear coefficient of thermal expansion between room temperature and any final temperature
can be obtained from the value for the temperature interval RT-100ºC by adding to it the numbers shown in
Table 4.7. This method is valid for stock of any grain orientation. Using the table, the expansion curves of
individual graphite samples can be computed to obtain good agreement with the measured curve.

The added factor indicated is to be combined with a coefficient determined in the temperature range RT-
100ºC(45), to give the mean coefficient of thermal expansion over the range from RT to the final temperature
of interest.

These factors have been compared with some very accurately measured coefficients of thermal expansion for
graphite, independently determined and published(13), to check their accuracy. Up to the 1000oC limit of these
measurements, the agreement between actual and calculated values is very close. More recent studies(46)
confirm the efficacy of combining CTE measured between RT and 100oC with a factor from Table 4.7 to
estimate CTE at temperatures up to 2500oC.

Since graphite has a low coefficient of thermal expansion (CTE), manufactured parts being cast can be
designed for an essentially finished size. Raw material selection, particle size, and impregnation will affect
CTE.

4-9
Table 4.7 Factors for Calculation of Mean Coefficient of Thermal Expansion

Final Temperature, ºC Added Factor, x 10-7/ºC

100 0
200 2
300 4
400 6
500 7.7
600 9.2
700 10.4
800 11.4
900 12.3
1000 13.2
1500 17.2
2000 21.2
2500 25.2

The effect of raw material selection on thermal expansion is significant in that the lower the CTE of the
calcined coke, the lower the CTE of the finished graphite. The addition of coarser particles to the base formula
also helps decrease CTE.

The CTE of graphite, compared to other structural materials, is illustrated in Table 4.8.

Table 4.8 Thermal Expansion of Various Structural Materials

Material x 10-6/oC
C opper 17
Carbon 4
Glass 0.5-10
Graphite 1-7
Iron 12
Porcelain 1.6-5.5
Molybdenum 5
Silicon Carbide 3.6
Silver 19
Tungsten 4.5

4-10
Thermal Shock Resistance
Resistance to thermal shock is a measure of the ability of a material to resist weakening or fracture when
subjected to sudden heating or cooling. No standard tests exist that will accurately evaluate this property since
shape is an important factor, as in the case of impact strength. Several practical tests are in use for specific
applications and, in the case of graphite, temperature gradients and times on the order of those encountered
in the striking of an arc are necessary, even for relative comparisons.

The exceptional thermal shock resistance of graphite arises from a unique combination of properties. The
properties which appear to be necessary are high tensile strength, high thermal conductivity, low modulus of
elasticity at failure, and low coefficient of thermal expansion. More sophisticated analyses take into account
the temperature dependence of these properties, but for relative comparison of materials, simpler analyses
may be used.

A test for the thermal shock resistance of refractory materials(18) consists of heating 2 inch diameter by 1/4
inch thick discs in a furnace, followed by sudden quenching in an air stream having a velocity of 250 feet per
second past the specimen. The test determined the thermal shock resistance of only refractory oxides and
carbides and expressed it as a parameter Ks/αE for these materials(19). Table 4.9 shows a comparison of the
value for graphite and the other refractory materials using this formula.

Table 4.9 Relative Evaluation of Thermal Shock Resistance

Order of Merit of
Ks
Materials Evaluated in 9 8 0 oC 1100oC 1 2 0 0 oC 1 3 2 0 oC
αE
Thermal Shock
UCAR ATJ Graphite (a) (a) (a) (a) 3,080,000
Titanium Carbide 25 25 25 21 18,200
Beryllium Oxide 25 3 6,390
Zircon 1 2,345
Magnesium Oxide 1/2 641-1,840
Stabilized Zirconia 0 345
(a) Specimens at 1650oC have been quenched in water without damage.
Spalling has been obtained by quenching in liquid sodium.

The table shows the marked superiority of graphite over ceramic materials in resistance to thermal shock. For
a qualitative prediction of the resistance of materials to fracture by thermal shock, the material should have
a high value of the parameter Ks/αE where:

K = Thermal Conductivity;
s = Tensile Strength;
α = Coefficient of Thermal Expansion; and
E = Modulus of Elasticity.

Within the carbon-graphite family of materials, the following are listed in descending order (i.e. best to worst)
by their ability to resist thermal shock.

1. Graphite with large grain size is more resistant than fine grain graphite.
2. Semi-Graphite.
3. Coal-Base Carbon.
4. Coke-Base Carbon.
5. Lampblack-Base Carbon.

For any grade, a thin section is more resistant to thermal shock than a thick section. Carbon or graphite
should be dried thoroughly to prevent internal steam explosions causing cracking. This is sometimes mistaken
for thermal shock spalling.

4-11
Emissivity and Spectral Characteristics
Emissivity is a measure of the heat-radiating ability of a surface. It may be expressed as total emissivity
or spectral emissivity. The total emissivity(14) of a surface of a radiator is the ratio of its radiant flux density
(radiancy) to that of a black body at the same temperature. A representative emissivity value for graphite
is 0.77(15), but is dependent upon the specific grade and surface finish. A true black body has an emissivity
of one.

The spectral emissivity of the surface of a radiator is the ratio of its radiant flux density per unit wavelength
interval (spectral radiancy) at that wavelength to the spectral radiancy of a black body at the same tempera-
ture. The spectral emissivity of graphite at 6500Å has been measured in connection with optical pyrometry
applications. Values of 0.90 for dulled-surface graphite and 0.77 for polished graphite are typical(16). The
emissivity of graphite varies slightly with temperature, having a temperature coefficient of 1.9 x 10-5/K. In the
wavelength range 2000 to 7500Å , graphite emissivity is nearly constant.

The crater of a carbon arc using a pure graphite anode is a useful high-temperature radiation standard. The
brightness of the anode seen through an optical pyrometer (6530Å) matches that of a black body at 3790K
(1948 I.T.S.) within the experimental error(15,20,21,22).

The crater radiation is also useful for spectral radiance calibration throughout the wavelength region 0.25 to
1.8µ. The measurements of MacPherson(15) and Euler(21,22) using a graphite anode agree within six percent
above 3000Å and are shown plotted in Figure 4.7. Although different temperatures and emissivities were
used to specify their data, for practical purposes, both sets of data may be described as the spectral radiance
of a grey body at 3900K with an emissivity of 0.894. The resultant calculated curve (drawn in Figure 4.7)
fits the observed data within six percent throughout the wavelength range 2500 to 18,000Å. (The spectral
radiance using a carbon anode is 10 to 15 percent less than with a graphite anode.)

A value of 0.98 was obtained for the emissivity of the crater of an operating arc based on reflectance
measurements(20). Subsequent data(23) on the arc and on samples at temperatures in the range 500 to 2600K
confirm the low reflectance of graphite and suggest emissivities in the range 0.96 to 0.98. Why reflectance
measurements yield higher emissivity values than are obtained with an optical pyrometer is not understood.

A carbon arc operating in air produces a strong band system due to the CN molecule in the arc plasma and
a C2 system, as shown in Figure 4.7. A strong atomic line due to carbon occurs at 2478Å. Additional weaker
bands due to CN, C2, and C3 molecules are observed(24-27).

4-12
Figure 4.7 Spectral Energy Distribution

1000

Watts/cm 2 /µ/Steradian
2
100 C

1,0 0,0 0,1 v I , v II

CN
H.G. MacPherson
1,0 0,0 0,1 vI , vII J. Euler
Grey Body Curve
T = 3900 K
10 e = 0.894

Carbon Line

1
2000 3000 4000 5000 6000 7000 8000 9000 10000

Wavelength, Å

In general terms, emissivity can be characterized in Table 4.10, as follows:

Table 4.10 Total Emissivity

Material Total Emissivity

Carbon 0.60-0.80
Graphite 0.70-0.95
Lampblack 0.90-0.99
Silver 0.04

4-13
Specific Heat
Specific heat is the quantity of heat required to raise the temperature of a unit mass of material one degree.
The common units of measure are related as follows:

1 BTU cal
lb oF = 1 gºC

Except at temperatures near absolute zero, the values of the specific heat of all types of natural and
manufactured graphites are essentially the same. From an extensive survey(30) of the measurements of the
specific heat reported in the literature, it appears that the difference between different measurements of the
same grade of graphite is as large as the difference between measurements of different grades. Below 20K,
the specific heat of a graphite single crystal is slightly less than that of petroleum coke-base manufactured
graphite; on the other hand, below 20K, lampblack-base manufactured graphite and unannealed pyrolytic
graphite have slightly higher specific heats than petroleum coke-base manufactured graphite(30,31). De Sorbo(32)
found that the specific heat of Ceylon natural graphite is up to nine percent less than that of manufactured
graphite in the temperature interval 120 to 300K. At higher temperatures, the difference between the values of
all types of graphite is less than the experimental error.

A number of individual measurements(29) have been made of the specific heat at constant pressure of natural
and manufactured graphites. Figure 4.8 represents the best fit of the values for manufactured graphite with
proper consideration given to the accuracy of the individual measurements. Below 300K, the average curve
essentially coincides with several individual measurements.

Figure 4.8 Specific Heat of Manufactured Graphite

0.5

0.4
Specific Heat, c p, cal/g oC

0.3

0.2

0.1

0.0
0 200 400 600 800 1000 1200 1400 1600 1800 2000

Temperature, K

The specific heat at constant pressure, cp, may be written as the sum of several terms,

cp = cv + cw + ce where:

cv = specific heat at constant volume of the lattice vibrations;

cw = difference between the total specific heat at constant pressure and at constant volume;
ce = specific heat of the electrons at constant volume.

4-14
Table 4.11 gives the numerical values of cp taken from the average curve for manufactured graphite and the
corresponding values of ∆cp, cw and cv. The quantity ∆cp represents a probable error such that most of
the reliable individual measurements fall within the range cp ± ∆cp. The table also extends the range of
values to 3800K. The difference, cw, has been calculated(28) from the density, coefficient of thermal expansion,
and compressibility. The compressibility at high temperatures has to be roughly estimated from the room
temperature value, so the values of cw at higher temperatures may be in error by a factor of two or three.
The values of the electronic specific heat, ce, have been computed from formulas(29) based on McClure’s
calculation of the electronic band structure of graphite.

The data in Table 4.11 indicate that at about 2200K, the lattice specific heat cv at constant volume, from
cv = cp - cw -ce, becomes greater than the classical Dulong and Petit value, which for graphite is 0.496 cal/goC.
Values of cv greater than 0.496 would indicate the existence of anharmonic atomic forces, but the possible
errors in cp and cw are large enough that this point is not at all certain. The value of cp increases sharply
as the temperature is increased above 3200K. This result is based on the work of Rasor and McClelland(33)
and, according to these authors, is due to the creation of vacancies at temperatures near the sublimation
temperature.

Table 4.11 Specific Heat of Manufactured Graphite

Temperature, K Specific Heat,cp, cal/goC ∆cp cw ce
0 0 0 0 0
50 0.0101 0.0002 0.00002 0.00001
100 0.0335 0.0003 0.00014 0.00002
150 0.0643 0.0007 0.00027 0.00004
200 0.0995 0.001 0.00039 0.00005
250 0.1357 0.0013 0.00054 0.00007
300 0.1723 0.0017 0.00069 0.00009
350 0.209 0.003 0.0009 0.0001
400 0.245 0.004 0.001 0.0001
450 0.276 0.005 0.0012 0.0002
500 0.303 0.006 0.0014 0.0002
550 0.323 0.007 0.0016 0.0002
600 0.34 0.008 0.0017 0.0002
650 0.356 0.009 0.0019 0.0003
700 0.37 0.009 0.0021 0.0003
750 0.382 0.01 0.0023 0.0004
800 0.393 0.01 0.0025 0.0004
850 0.402 0.01 0.0027 0.0004
900 0.409 0.01 0.0029 0.0005
950 0.415 0.01 0.0031 0.0005
1000 0.421 0.011 0.0033 0.0006
1100 0.432 0.011 0.004 0.001
1200 0.443 0.011 0.004 0.001
1300 0.452 0.012 0.004 0.001
1400 0.46 0.012 0.005 0.001
1500 0.468 0.013 0.005 0.001
1600 0.474 0.014 0.005 0.002
1800 0.486 0.015 0.006 0.002
2000 0.496 0.016 0.007 0.002
2200 0.504 0.018 0.007 0.003
2400 0.511 0.02 0.007 0.004
2600 0.516 0.022 0.008 0.004
2800 0.521 0.024 0.008 0.005
3000 0.527 0.026 0.008 0.006
3200 0.536 0.03 0.008 0.006
3400 0.548 0.03 0.01 0.01
3600 0.58 0.04 0.01 0.01
3800 0.69 0.05 0.01 0.01

4-15
The average experimental values of cp can be represented by simple polynomial functions of the absolute
temperature T, with T in K to give cp in cal/goC.

Within the temperature interval 0K to 300K:

cp = 7.2774 x 10-12T4 - 8.5781 x 10-9T3 + 3.6890 x 10-6T2 + 4.4112 x 10-5T -7.9099 x 10-5

Above 40K, the calculated values are within 2.2 percent of the average experimental values. This equation is
not reliable below 40K
.
Within the temperature interval 300K to 3200K:

cp = 0.44391 + 0.30795 x 10-4T - 0.61257 x 105T-2 + 0.10795 x 108T-3.

This equation reproduces the experimental values within 1.5 percent over the entire interval.

These functions can be easily integrated to obtain the heat energy added or given off on changing the tem-
perature of graphite. Rasor and McClelland(33) give a formula for the exponential increases in cp due to the
creation of vacancies at temperatures above 3200K.

4-16
Vapor Pressure

Vapor pressure is defined as the pressure exerted when a solid or liquid is in equilibrium with its own vapor.
The vapor pressure is a function of the substance and of the temperature. Graphite is unique in exhibiting a
vapor pressure that is negligible up to about 2500oC under normal conditions and, for that reason, the vapor
pressure of graphite is not a limiting property in most applications. Even under high vacuum, it is practical
to use graphite up to 2000oC. Other factors may have a dominant influence on weight loss, namely loss
of carbon due to chemical reaction of graphite with container material or with residual gases, notably
oxygen. These reactions will predominate unless the graphite is protected. Also, when graphite is heated to
temperatures above 1000ºC, outgassing occurs, and in order of increasing temperatures, gases H2O, CO,
CO2, and H2 are released. Depending on the previous history of the graphite, these gases may vary from
minute quantities to several hundred times the volume of the graphite. Finally, most graphites, unless purified,
contain small amounts of ash, the constituents of which volatilize at various temperatures, causing a weight
loss.

Standard techniques for vapor pressure determination are based on the classical Clapeyron equation:

dp ∆H
=
dT T∆V
where:

p is the vapor pressure;

∆H is the heat of vaporization;
∆V is the increase in volume from solid or liquid to vapor; and
T is the absolute temperature.

When ∆H and ∆V are not constant, it is necessary that the deviations be known as functions of temperature in
order to integrate this equation. In cases where several species are produced during vaporization, e.g., C, C2,
C3.... it is further necessary that the above equation be applied separately to each constituent. Because the
required information on the variation of ∆H and ∆V with temperature has not been determined experimentally
outside the temperature range of 2500K to 3000K, the following relation is employed to obtain vapor pressures
outside this range:
∆Fτo ∆Hoo
-R log.p = = fef (gas) -fef (solid) +
T T
where:

R is the Universal Gas Constant = 1.987 cal/oC/g mol;

∆Fτo is the free energy change at T K;
∆Hoo is the heat of vaporization at 0 K; and

∆Fτo - ∆Hoo is the free energy function, a quantity that is related to the
fef = entropy by the equation:
T
1
fef = Sτ + (Hτ-Ho)
T

4-17
 As with entropy, free-energy functions in liquid or solid phases are determined from heat capacity data,
whereas in the gaseous phase, they are determined from known or assumed structures, vibrational frequen-
cies, and multiplicities of ground and low-lying excited states of the various molecular species present.

Both monatomic and polyatomic species of carbon result from the vaporization of graphite; molecules
of C through C5 have been observed(34-36). These species were identified by effusion studies and further
characterized by spectroscopic measurements(37,38). A theoretical treatment, based upon the assumption that
the bonding can be described by molecular orbitals, predicts that the larger molecules become relatively more
abundant with temperature and that the resulting molecules are linear chains (ring-shaped molecules are
expected to be considerably less abundant), of which those containing an odd number of atoms are more
abundant than the even numbered ones(39,40). At 2500K, the saturated vapor contains about 60% C3, 10% C2,
30% C, and small amounts of larger molecules. For these species, free-energy functions were obtained and
combined with existing vapor pressure data(34-36) to yield the vapor pressure information listed in Table 4.12
and depicted in Figure 4.9. The equivalent pressure is the pressure which would be observed if the vapor
consisted entirely of monatomic carbon. This quantity was obtained as a weighted sum of partial pressures
for the different species. In turn, the weight loss of material under free vaporization was determined from the
equivalent pressure assuming that Knudsen flow conditions prevailed (essentially ideal gas behavior). This
information is presented in Table 4.12 and Figure 4.10.

Table 4.12 Vapor Pressure of Graphite

Temperature Partial Pressures of Species in Atmospheres Equivalent Free Vaporization

Vapor Pressure, Rate,
K ºC C C2 C3 C5 C5 atm(1) g cm2 sec
-23
1200 927 1.01 x 10 1. 17 x 10-23 5.19 x 10-23
1300 1027 2.81 x 10-21 2.38 x 10-22 3.54 x 10-21 1.51 x 10-20
-19 -21
1400 1127 3.21 x 10 1.015 x 10 4.37 x 10-20 4.54 x 10-19 1.87 x 10-18
1500 1227 1.95 x 10-17 1.34 x 10-19 3.96 x 10-18 3.16 x 10-17 1.26 x 10-16
1600 1327 7.08 x 10-16 8.55 x 10-18 2.04 x 10-16 1.34 x 10-15 5.15 x 10-15
-14 -16
1700 1427 1.68 x 10 3.34 x 10 6.55 x 10-15 3.11 x 10 -20
4.47 x 10 -20
3.72 x 10-14 1.39 x 10-13
1800 1527 2.81 x 10-13 8.65 x 10-15 1.42 x 10-13 1.62 x 10-18 2.33 x 10-18 7.25 x 10-13 2.64 x 10-12
1900 1627 3.47 x 10-12 1.59 x 10-13 2.45 x 10-12 5.52 x 10-17 8.00 x 10-17 1.08 x 10-11 3.81 x 10-11
-11 -12
2000 1727 3.34 x 10 2.21 x 10 3.10 x 10-11 1.03 x 10-15 1.92 x 10-15 1.31 x 10-10 4.61 x 10-10
2500 2227 1.79 x 10-7 4.30 x 10-8 2.62 x 10-7 2.15 x 10-10 3.17 x 10-18 1.05 x 10-6 3.23 x 10-6
3000 2727 5.43 x 10-5 3.03 x 10-5 1.41 x 10-4 6.05 x 10-7 8.97 x 10-7 5.38 x 10-4 1.51 x 10-3
3500 3227 3.18 x 10-3 3.20 x 10-3 1.38 x 10-2 1.68 x 10-4 2.49 x 10-4 5.30 x 10-2 0.138
4000 3727 6.67 x 10-2 0.103 0.47 1.10 x 10-2 1.64 x 10-2 1.83 5.29
4500 4227 0.708 1.52 7.13 0.276 0.408 28.3 65
5000 4727 4.66 12.9 72.5 3.54 5.2 288 627
5500 5227 21.6 72.8 839 27.8 40.6 3000 6240
6000 5727 77.6 307 6390 152 220 21,600 43,000
(1) The equivalent pressure is that pressure which would be observed if the vapor were monatomic carbon.

4-18
Figure 4.9 Vapor Pressure of Graphite
10 2

10 1

10 0

10 -1
Equivalent Vapor
10 -2 *
Pressure
10 -3
Vapor Pressure, atm

C3
10 -4

10 -5
C
-6
10
C2
10 -7
C5
-8
10
C4
10 -9

10 -10 *The pressure that would be observed if the

vapor were monatomic carbon only
10 -11

10 -12
1500 2000 2500 3000 3500 4000 4500 5000 5500

Temperature, K

4-19
Figure 4.10 Free Vaporization Rate of Graphite

10 2

10 1

10 0

10 -1
Vaporization Rate, g/cm 2-sec

10 -2

10 -3

10 -4

10 -5

10 -6

10 -7

10 -8

10 -9

10 -10

10 -11

10 -12
1500 2000 2500 3000 3500 4000 4500 5000

Temperature, K

4-20
All of the measurements used in the above determinations were made under conditions of equilibrium
between solid and vapor phases. Unpublished work on accommodation coefficients (a factor taking into
account the resistance to vaporization due to energy exchange effects at the surface) shows that for high-
density graphites, free vaporization and equilibrium are essentially equivalent(41). In similar work on single
crystals of pyrolytic graphite(42), the accommodation coefficients for the C and C2 species were found to be
normal (approximately 0.3), whereas the value for C3 was surprisingly low (approximately 0.4). This implies
that the free vaporization results given here should be adjusted by reducing the contribution due to the C3
species by a factor of about 10. In general, therefore, the present free vaporization values should be regarded
as upper limits; they will apply to graphites with normal accommodation coefficients and they will be too high
when this quantity is smaller. The equilibrium pressures given here are valid in all cases.

The free-energy functions used for vapor pressure calculation are based on the following considerations.

Graphite
Since pressures of the various constituents in the vapor are obtained by difference of the free energy functions
of gaseous and solid materials, data concerning graphite are necessary. There are two sources of heat
capacity data from which free energy functions of solid may be obtained: either the JANAF tables(43), which
are employed here, or the specific heat. Both these sets of values were obtained by averaging experimental
data for different graphites; although the results differ slightly, sometimes one set is larger and sometimes
the other, so that these effects compensate each other upon integration and the free energy functions differ
by a maximum of two percent.

C and C2
The physical properties of these molecules are well defined; hence, the values of the free energy functions
given in the JANAF tables(43) are regarded as reliable and were used.

C3
At present, the reported physical properties of C3 lead to conflicting results and, unfortunately, this species is
the most important constituent of the vapor. Mass spectrometric effusion studies by several investigators(34-36,39)
are in good agreement with each other in yielding heats of formation, of about 188 kcal/mole. Recent
spectroscopic work(37,44) indicates, however, that there is a low bending vibrational frequency of about
70 cm-1. The resulting free-energy functions in conjunction with the mass spectrometric results lead to heats of
formation of about 230 kcal/mole; moreover, these values are not self-consistent because they increase with
the temperature of the mass spectral measurements. Because of the conflicting data, the spectroscopy-based
free energy functions are employed here in connection with a set of arbitrary temperature dependent heats of
formation, ∆H, obtained by extrapolation of Drowart’s data(34). The resulting vapor pressures appear plausible
and are somewhat higher than the values derived from the free energy functions in the JANAF tables, which
are based on earlier work. Hence, the data reported for temperatures exceeding 3000K should be regarded
as tentative and subject to revision. The values given for temperatures of 2500 to 3000K are experimental
quantities and considered to be exact.

C4 and C5
The free energy functions are based upon theoretical estimates of heat capacities and are regarded as only
approximate. Between 2500 and 3000K, as with C3, the data are experimental quantities.

C6 and C..., Ring Molecules, etc.

Theoretical estimates concerning these molecules have been given by Pitzer(39,40). They have not yet been
observed experimentally, but may appear at temperatures above 3000K. Their partial pressures, however, will
probably not be significant, relative to those of the other species reported here.

Graphite has a measurable vapor pressure at 2400K of 1.6 x 10-7. The normal pressure is only obtained at
3491oC, the sublimation point. The sublimation rate increases rapidly with an increase in temperature. It does
not liquefy until approximately 3700oC and 100 atmospheres.

4-21
 References

1. Bowman, J. C., Krumhansl, J. A., and Meers, J. T., Paper Presented to the Society of Chemical Industry
Conference on Industrial Carbon and Graphite London (September 1957).

2. Smith, A. W. and Rasor, N. S., Phys. Rev. 104, 885 (1956).

3. De Sorbo, W. and Tyler, W. W., J. Chem. Phys. 21, 1660 (1953).

4. Smith, A. W., Phys. Rev. 93, 952; ibid. 95, 1095 (1954).

5. Powell, R. and Schofield, F., Phys. Soc., London Proc. 51, 153 (1939).

6. Euler, J., Nature, 39, 568 (1952).

7. Rasor, N. and McClelland, J., Technical Report, U.S. Wright Air Development Center, WADC-TR-56-400,
Part 1 (1956).

8. Johnson, H. V., Unpublished Data, National Carbon Company Research Laboratories as Quoted in
Currie, L. M., Hamister, V. C., and MacPherson, H. G., Proceedings of the International Conference
on the Peaceful Uses of Atomic Energy, Geneva, 1955; 8, 451 (New York, United Nations) 1956.

9. Neubert, T., Private Communication as Quoted in Currie, L. M., Hamister, V. C., and
MacPherson, H. G., Proceedings of the International Conference on the Peaceful Uses of
Atomic Energy, Geneva, 1955; 8, 451 (New York, United Nations) 1956.
10. MacPherson, H. G., Unpublished Data, National Carbon Company Research Laboratories, NCRL-242
(1946).

11. Meritt, G. E., National Bureau of Standards Scientific Papers, 19, 357 (1923-4).
12. Westbrook, R. D. and Micinski, E., Unpublished Data, National Carbon Company Research Labora-
tories, NCRL-299 (1952).

13. Hidnert, P., Journal of Research, U.S. National Bureau of Standards, 13, 37 (1934).

14. ASA, Nomenclature and Photometric Standards, Z-7, 1 (1942).

15. MacPherson, H. G., J. Opt. Soc. Am. 30. 189 (1940).

16. Thom, R. J. and Simpson, O. C., J. Appl. Phys. 24, 633 (1953).

17. Tyler, W. W. and Wilson, A. C. Jr., Phys. Rev. 89, 870 (1953).

18. Gangler, J. J., J. Am. Ceram. Soc. 33, 367-75 (December 1950).

19. Green, L. Jr., J. Appl. Mech. 18, 346 (1951).

20. Null, M. R. and Lozier, W. W., J. Appl. Phys. 29, 1605 (1958).

21. Euler, J., Ann. d. Physik 11, 203-224 (1953).

22. Euler, J., Ann. d. Physik 14, 145-173 (1954).

23. Lozier, W. W., Technical Report, U.S. Wright Air Development Center, WAD-TR-59-789.
24. Pearse, R. W. B. and Gaydon, A. G., The Identification of Molecular Spectra, John Wiley & Sons, New
York (1941).
25. Rosen, B., Constantes Selectionnees 4, Hermannet Cie, Paris (1951).

26. Phillips, J. G., Astrophys. Journ. 108, 434-452 (1948).

27. Marr, G. V., University of Western Ontario, AFCRC-TN-57-247 (1957).

28. Hove, J. E., Proceedings of the 3rd Biennial Carbon Conference, University of Buffalo, June 1957, p.
515 (1957).
29. Spence, G. B., Technical Report WADD TR 61-72, Vol. XLI (1963), “Survey and Analytical Representation
of the Measurements of the Specific Heat of Graphite.”

30. De Sorbo, W. and Nicols, G. E., J. Phys. Chem. Solids 6, 352 (1958).

31. van der Hoeven, B. J. C. and Keesom, P. H., Phys. Rev. 130, 1318 (1963)

32. De Sorbo, W., J. Am. Chem. Soc. 77,4713 (1955).

33. Rasor, N. S. and McClelland, J. D., J. Phys. Chem, Solids 15, 17 (1960).
34. Drowart, J., Bums, R. P., De Maria, G., and Inghram, M. G., J. Chem. Phys. 31, 1131 (1959).

35. Chupka. W. A. and Inghram, M. G., J. Chem. Phys. 59, 100 (1955).

36. Thom, R. J. and Winslow, G. H., J. Chem. Phys. 26,186 (1957).

37. Weltner, W. Jr., Walsh, P. N., and Angell, C. L., J. Chem. Phys. 40, 1289 (1964).

38. Weltner, W. Jr. and McLeod, D. Jr., ibid, 1299.

39. Pitzer, R. S. and Clementi, E., J. Am. Chem. Soc. 81, 447 (1959).

40. Clementi, E. and McLean, A. D., J. Chem. Phys. 36, 45 (1962).

41. Schissel, P. O., Union Carbide Research Institute, Tarrytown, New York, Private Communication.

42. Bums, R. P., Jason, A. J., and Inghrarn, M. G., J. Chem. Phys. 40,1161 (1964).
43. Stull, D. P., Editor, Joint Army-Navy-Air Force Thermochemical Data. Midland, Michigan (1960).

44. Gausset, L., Herzberg, G., Lagerquist, A., and Rosen, B., Dis. Far. Soc. No. 35(1963).
45. Wagoner G., Sprogis G., and Proctor, D. G., Proceedings of the International Carbon Conference,
Baden-Baden, 1986, p. 234-236.
46. Shao, R. L., Proctor, D. G., and Chang, C. F., Proceedings of the 20th Biennial Carbon Conference,
1991, p. 202-203.
Chapter Five

ELECTRICAL PROPERTIES

Industrial Graphite Engineering Handbook

 TABLE OF CONTENTS

Electrical Properties
Graphite Single Crystals...............................................................................................5-1
Electrical Resistivity......................................................................................................5-1
Manufactured Graphites ...............................................................................................5-1
Electrical Contact Resistance.......................................................................................5-6
Electrical Resistance - Designing Elements .................................................................5-8
Power Input ................................................................................................................5-9
Electrical Properties
Electrical Resistivity

Graphite Single Crystals

The resistivity of a pure single crystal of graphite is an intrinsic property and represents the lower limit that
can be achieved in any pure, formed, graphite body. Since graphite crystal resistivity is intermediate between
that of a metal and a semiconductor, it is a strong function of temperature and impurity content. In addition,
the resistivity is anisotropic, being about two orders of magnitude higher in the C-axis direction at room
temperature. Measurements on highly-purified, good, single crystals of graphite(1) represent the lowest values
obtained in the temperature range below 300K. The fact that these values are the lowest of all measurements
reported, that they approach zero resistivity at absolute zero, that they were obtained by trustworthy
experimental techniques, is interpreted to mean that they most truly represent our knowledge of graphite
single crystal resistivity. These data for resistivity in the A-axis direction are shown in Figure 5.1.

Figure 5.1 Graphite Single Crystal Resistivity vs. Temperature (A-axis Direction)

50

40
Resistivity, µ-ohm-cm

30

20

10

0
0 50 100 150 200 250 300 350

Temperature, K
Manufactured Graphites
The values of resistance for polycrystalline manufactured graphites are not expected to have any simple
relation to those for single crystals, although the anisotropy of resistance of manufactured graphite shown
in most stocks is, of course, related to the single crystal anisotropy. For instance, the room-temperature
resistivities of commercial graphites are of the order of 20 times that for a single crystal, while the nature of
the temperature dependence of resistivity of commercial materials is widely different as well. For lampblack-
base, poorly crystalline graphite, the temperature coefficient of resistance is continuously negative to 1000oC,
with evidence of flattening suggesting a reversal at some higher temperature. With coke-base graphites, the
coefficient is negative only to about 500oC, where a minimum resistivity occurs; at higher temperatures, the
coefficient becomes positive. A physical basis for explaining this behavior is readily seen in the combined
effect of the two types of mechanisms contributing to the resistivity of formed graphite. First, the number
of electrons available to carry the current increases with increasing temperature, which should lower the
resistivity; but second, the freedom of these electrons to move in a given direction is restricted by scattering

5-1
effects of various sorts, with different temperature dependences. Thus, while the increasing electron density
always tends to reduce the resistivity as the temperature rises, scattering effects may dominate at any
temperature to modify or even to reverse this trend. With an assembly of disoriented crystallites bonded
together in a matrix with 10%-30% porosity, as in manufactured graphite, it is easily seen that scattering will be
a very important, as well as a very complex factor, in the determination of electrical resistivity.

Typical resistivity data are shown in Figure 5.2 for a coke(2,3,5) and coal-base and lampblack-base graphites.
Independent measurements on a coke-base graphite to about 3700ºC(4) show a smooth continuation of the
upward trend in resistivity to this very high temperature. Table 5.1 references normal resistivities for various
graphites. Figure 5.3 illustrates the temperature resistivity curves for the UCAR Isomolded Grades ATJ,
DM302, DM332, and DM362.

Figure 5.2 Electrical Resistivity vs. Temperature

200

180
Relative Electrical Resistivity, R.T.=100%

160

140
Graphitized Petroleum Coke-Base
120

100

80 Carbon Petroleum Coke-Base

Carbon Lampblack-Base
60 Carbon Anthracite-Base

40
Graphitized Lampblack-Base
20

0
0 400 800 1200 1600 2000 2400 2800
o
Temperature, C

Table 5.1 • Normal Resistivity of Various Graphites

Normal Resistivity at Room Temperature µΩm

Carbon Anthracite-Base 38
Carbon Petroleum Coke-Base 38
Carbon Lampblack-Base 51
Graphitized Petroleum Coke-Base 9.1
Graphitized Lampblack-Base 47

5-2
Figure 5.3 Specific Resistance vs. Temperature for the UCAR Isomolded Grades ATJ, DM302, DM332,
and DB58

16

15

14

Specific Resistance, µ Ω m
13 ATJ
DM302
12
DM332
11 DM362

10

5
0 200 400 600 800 1000 1200 1400 1600 1800 2000
o
Temperature, C

In graphite, which has a defined crystalline structure, numerous electrons are free, providing high conductivity.
Carbon, with undefined monoplanes and relatively few available electrons has high resistance. See Table 5.2
for the specific resistivity of common materials. The room temperature resistivity of several typical carbon and
graphite grades is shown in Tables 5.3 and 5.4.

Table 5.2 Room Temperature Electrical Resistivity

Material µΩm
Aluminum 0.028
Carbon 50
C opper 0.017
Graphite 7
Iron 0.1

5-3
Table 5.3 Typical Room Temperature Properties of UCARGrades, English Units

UCAR Largest Grain Specific

Group Grade Cross Section Size(1) Resistance
inches inches 10−4 Ωinches
WG AG
Extruded Medium Grain
C BY 24 0.03 3.2 4.1
CS 30 0.03 3.0 4.4
SLWX 24 0.03 2.8 3.8
CHV 24 x 48 0.06 3.2 4.7
Molded Medium Grain
PGW 52 0.03 --- 3.7
PGX 73 0.03 4.0 5.6
Extruded Fine Grain
FGE 8.5 x 8.5 0.006 2.8 4.7
FGEX 8.5 x 8.5 0.006 3.2 4.7
Isomolded Fine Grain
ATJ 24 0.001 4.6
CGW 16 x 16 0.001 4.6
Isomolded Superfine Grain
DM306 8 x 16 0.0002 14.2
DM300 8 x 16 0.0002 5.7
DM301 8 x 16 0.0002 4.9
DM302 8 x 16 0.0002 4.5
DM303 8 x 16 0.0002 5.5
DM342 8 x 16 0.0002 4.7
DM332 24 0.0002 4.4
Isomolded Ultrafine Grain
DM362 6 x 12 0.00008 7.3
(1) For the extruded medium and fine grain and molded medium
grain, the grain size is the maximum; for all the isomolded, the grain
size is the average.

5-4
Table 5.4 Typical Room Temperature Properties of UCAR Grades, SI Units

UCAR Largest Grain Specific

Group Grade Cross Section Size(1) Resistance
mm mm µΩm
WG AG
Extruded Medium Grain
C BY 600 0.8 8.5 10.0
CS 600 0.8 7.5 9.0
SLWX 600 0.8 7.0 9.6
CHV 600 x 1200 1.7 8.0 12.0
Molded Medium Grain
PGW 1300 0.8 --- 9.6
PGX 1825 0.8 10.2 13.0
Extruded Fine Grain
FGE 215 x 215 0.150 7.0 12
FGEX 215 x 215 0.150 8.0 12
Isomolded Fine Grain
ATJ 600 0.025 11.7
CGW 400 x 400 0.025 11.7
Isomolded Superfine Grain
DM306 200 x 400 0.005 36
DM300 200 x 400 0.005 14.6
DM301 200 x 400 0.005 12.4
DM302 200 x 400 0.005 11.4
DM303 200 x 400 0.005 14
DM342 200 x 400 0.005 12
DM332 600 0.005 11.2
Isomolded Ultrafine Grain
DM362 150 x 300 0.002 15.5
(1) For the extruded medium and fine grain and molded medium
grain, the grain size is the maximum; for all the isomolded, the grain
size is the average.

5-5
Electrical Contact Resistance
The term electrical contact resistance refers to the resistance occurring between any two contact points in
electrical applications. A resistance characteristic of the two materials making the contact is established.
This contact resistance is dependent upon a number of variables, the most important being the nature of
the materials involved, the pressure at the contact, the nature of the surfaces making the contact, and the
quality of the contact surfaces.

In Figure 5.4, the contact resistance of graphite to graphite and to a number of other frequently encountered
metals is plotted against contact pressure. In making these determinations, all surfaces were smoothed to a
plane and the metal surfaces were cleaned. Table 5.5 shows similar data.

Figure 5.4 Typical Electrical Contact Resistance of Graphite to Various Surfaces

1000
Graphite
Graphite
Surface Contact Pressure, lb/in 2

to
to Steel
Copper
800

600

Graphite Graphite
to to
400
Brass Aluminum

200 Graphite
to
Graphite

0
0.01 0.1 1 10 100
-3 2
Contact Resistance, x 10 ohm-inches
Table 5.5 Typical Contact Resistance of Graphite

Contact Resistance, ohms-inches2

Pressure, Graphite to Graphite to Graphite to Graphite to Graphite to
p si Graphite C opper Steel Brass Aluminum
25 0.000473 0.000704 0.01309 0.00092 0.0448
75 0.000261 0.000424 0.00694 0.000345 0.0159
150 0.000175 0.000315 0.00438 0.000214 0.0067
250 0.000101 0.000237 0.00282 0.000162 0.0043
400 0.000064 0.000162 0.00177 0.000109 0.002
750 0.000036 0.0000755 0.000856
1000 0.000031 0.0000555 0.000737

5-6
The values must be considered as typical only, since the absolute electrical contact resistance for any given
situation will depend upon the individual pieces and their exact surface condition. The contact resistance of
carbon to metals is higher than that of graphite because carbon has a higher specific resistance and because
it is harder. Table 5.6 shows some typical contact resistance of carbon-metal junctions.

Table 5.6 Typical Contact Resistance of Carbon

Contact Resistance, ohms-inches2

Pressure, Carbon to Carbon to Carbon to Carbon to
p si Carbon C opper Brass Aluminum
5 0.0058 0.0133 0.038 0.138
25 0.0026 0.0093 0.012 0.078
75 0.0017 0.0042 0.0052 0.015

5-7
Electrical Resistance - Designing Elements

In considering the design or operation of a furnace provided with graphite resistor elements, the following
comments may be helpful:

Limiting Current Density

The maximum current density that can be put through a graphite resistor in an inert atmosphere is determined
by two factors: (1) the rate of surface evaporation to be allowed for a reasonable life of the resistor; and (2)
the interior temperature of the resistor should not reach such a value that sublimation of the graphite will be
so rapid as to cause the rod to explode. Calculation of the maximum current in cylindrical graphite rods for the
two cases in which each of these two factors, respectively, is limiting is given below.

Limited by Surface Evaporation

The following formula is calculated on the basis of allowing a maximum surface temperature of 2200oC, which
should give a reasonably long life for a graphite resistor in an inert atmosphere:
A
(1) I =
RD

where R is the specific resistance of the rod at 2200oC in ohm-in., D is the diameter of the rod in inches, and I
is the limiting current density in amp/in.2. Table 5.7 lists the factor A for various furnace temperatures.

Table 5.7 Temperature and Corresponding Affect

o
Temperature of Surrounding Environment C “A” Factor
1000 63.9
1200 61.9
1400 58.9
1600 54.1
1800 46
2000 35.4
2100 26.3
2200 0

Limited by Interior Temperature

A conservative calculation for the maximum current density based on the limitation of internal vapor pressure
is given by the formula:

(2) I = 96.7
D R

where R is the specific resistance of the rod at 2200oC in ohm-in., D is the diameter of the rod in inches, and
I is the limiting current density in amp/in.2.

The limiting current density used in any case should be the smaller of the calculations (1) and (2). For rods
smaller than two inches in diameter, (1) will always be the limiting condition. For rods greater in diameter than
two inches, the temperature of the surroundings will determine whether (1) or (2) is the determining factor.

5-8
Power Input
The wattage radiated by the resistor elements, of course, will just equal the rate of energy loss from the
furnace by radiation, conduction, and sensible heat removed in the product tapped minus any heat contained
in the charge or developed by chemical reactions. The relationship between the temperature within the
furnace and the rate of radiation of energy from the resistors is given by the following equation derived from
the Stefan-Boltzmann Law:

W = (36.6 x 10-12) (.8) (T24 - T14)

where:

W = power radiated per unit area of resistor surface in watts/in.2

Stefan-Boltzmann constant in watts/in.2/K4 = 36.6 x 10-12
Emissivity ratio for graphite = 0.8
T1= Furnace temperature in K
T2 = Resistor temperature in K

5-9
 References

1. Soule, D. E., Unpublished Data, National Carbon Company Research Laboratories, NCRL-I
(1955).
2. Dexter, W. B., Unpublished Data, National Carbon Company Research Laboratories, NCRL-218
(1942).
3. Micinski, E., Unpublished Data, National Carbon Company Research Laboratories, NCRL-3
(1949).

4. Rasor, N. S., Private Communication to W. P. Eatherly.

5. Tyler, W. W. and Wilson, A. C. Jr., Phys. Rev. 89, 870 (1953).

6. Shepard, R. L., Unpublished Data, National Carbon Company Research Laboratories, NCRL-6
(1950).
Chapter Six

CHEMICAL PROPERTIES

Industrial Graphite Engineering Handbook

 TABLE OF CONTENTS

Chemical Properties
Ash Content ................................................................................................................6-1
Gas Evolution from Graphite.........................................................................................6-2
Oxidation ................................................................................................................6-5
Oxidation Retardation ...................................................................................................6-8
Resistance to Erosion and Attrition ...............................................................................6-9
Heat of Combustion ....................................................................................................6-10
Chemical Properties

Ash Content
The ash content of all graphite is relatively low since most of the impurities originally present in the raw
materials are volatilized and driven off at the temperatures encountered in the graphitization of the product.

In Table 6.1, the total ash contents of several typical carbon and graphite grades are presented.

The total ash content of graphite can have an important influence on oxidation characteristics (see the section
on oxidation of graphite later in this chapter). Low-ash graphites, such as CBY and CS, are examples of
graphites exhibiting good oxidation characteristics due to low ash.

Table 6.1 Typical Ash Content of UCAR Grades

Group UCAR Largest Cross Section Grain Size,(1) A sh

Grades inches mm inches mm ppm
Extruded Medium Grain
C BY 24 600 0.03 0.8 800
CS 30 600 0.03 0.8 800
SLWX 24 600 0.03 0.8 <5
CHV 24 x 48 600 x 1200 0.06 1.7 1.50%
Molded Medium Grain
PGW 52 1300 0.03 0.8 1000
PGX 73 1825 0.03 0.8 700
Extruded Fine Grain
FGE 8.5 x 8.5 215 x 215 0.006 0.150 1000
FGEX 8.5 x 8.5 215 x 215 0.006 0.150 1000
Isomolded Fine Grain
ATJ 24 600 0.001 0.025 900
CGW 16 x 16 400 x 400 0.001 0.025 2000
Isomolded Superfine Grain
DM306 8 x 16 200 x 400 0.0002 0.005 2000
DM300 8 x 16 200 x 400 0.0002 0.005 1600
DM301 8 x 16 200 x 400 0.0002 0.005 1300
DM302 8 x 16 200 x 400 0.0002 0.005 1100
DM303 8 x 16 200 x 400 0.0002 0.005 1500
DM342 8 x 16 200 x 400 0.0002 0.005 1500
DM332 24 600 0.0002 0.005 600
Isomolded Ultrafine Grain
DM362 6 x 12 150 x 300 0.00008 0.002 1500
(1) For the extruded medium and fine grain and molded medium grain, the
grain size is the maximum; for all the isomolded, the grain size is the average.

6-1
Gas Evolution from Graphite
The many fine pores which characterize the structure of industrial graphite usually contain chemically- and
physically-absorbed gases, some of which are retained despite prolonged evacuation of room temperature.
Moreover, very small amounts of certain metallic carbide impurities will react with moisture to produce gases
not originally present in the graphite. Such impurity centers, trapped inside the graphite structure, can operate
to evolve gas for a long time. The use of graphite to form the plates and grids of high-power radio and rectifier
tubes first called attention to the fact that removal of these gases and gas-forming impurities was necessary to
ensure long-term maintenance of a good vacuum and to prevent filament poisoning.

The “gas content” of graphite has come to be defined as the volume of gas evolved from graphite when
it is heated to a specified high temperature after it has been evacuated at a lower temperature, i.e., room
temperature. In order to determine this value, the graphite sample is mounted inside a quartz bulb in a
vacuum system and pumped out at room temperature for at least one hour, reaching a pressure well below
0.001 mm of Hg. Then the system is sealed off and the graphite heated by induction to the specified high
temperature. The gas evolved increases the pressure. The volume of gas evolved is calculated from the
pressure rise. Thus, the term “gas content” includes the gas which cannot be removed by prolonged pumping
at the lower temperature as well as the gas which is produced by the decomposition of chemical impurities in
the graphite at the higher temperature. Table 6.2 shows typical results with two standard grades, not intended
for use where gas content is critical. Large variations in these values, dependent upon the particular sample
and the environmental conditions, can occur with these grades.

Table 6.2 Gas Content of UCAR Medium Grain Extruded Graphite

UCAR Gas Liberated, mm3/g

Grade 1 4 0 0 oC 2 0 0 0 oC Total
AGSR 400 90 490
AGSX 484 80 564

Typical composition of gas liberated by UCAR grade AGSR is shown in Table 6.3

Table 6.3 Percentage Composition of Gases from Medium Grain Extruded AGSR Graphite

Composition, %
o
Gas 1400 C 2 0 0 0 oC
O2 0 0
H2O 0 0
CO2 13 10
CO 34 52
H2 Trace 0
N2 52 38

The nature of the impurities in graphite is known to have a marked effect on the gas content. Calcium is
an especially critical material(1). Initially present as the carbide, it hydrolyzes to Ca(OH)2 on standing in air,
and then partially converts to CaCO3 in contact with CO2. With even a fraction of one percent of calcium
present, graphite heated under vacuum will evolve water vapor, CO2 and CO for a long time, making a stable
“clean-up” very difficult. In most graphite applications, this is of no concern, but where low gas content is
required, special grades in which particular attention has been paid to the choice of materials and to the
processing conditions which influence gas content should be used.

6-2
The gas content of graphite may also be of importance in nuclear reactors. If the graphite is to be canned
in metal containers, an excessive evolution of gas could create certain pressure and embrittlement problems.
Low gas content graphites are obviously recommended for such applications. Figure 6.1 shows gas evolution
versus temperature data on two samples of nuclear UCAR grade AGOT graphite(2). In a similar study of
purified grade TSP graphite, considerably less gas was evolved. The composition of the gas for UCAR AGOT
graphite was found to be predominantly CO with a lesser amount of H2, while that for TSP was primarily H2
with a lesser amount of CO (see Table 6.5)(3).

Figure 6.1 Total Gas Desorbed as a Function of Temperature for Various Types of Graphite

200
Total Volume Gas, mm 3/g

AGOT-KC

100

AGOT-CSF

0
0 200 400 600 800 1000 1200 1400 1600 1800
o
Outgassing Temperature, C

Table 6.4 Percentage Composition of Gases from UCAR Grades AGOT and TSP Graphites at 1750ºC

Composition, %

Gas UCAR AGOT UCAR TSP
H2 30.1 81.5
CO ~67 16.9
N2 ~3 1.4
CO2 0.2 0.1
CH4 0.1 0.1

While the origin of the reported H2 and CO is not positively known, it may result from the absorption of water
vapor from the atmosphere after the stock is exposed to air. For example, the gas content of grade TSP
graphite has been found to be very much less when stored in a desiccator with P2O5 than when exposed to
moist air(4). In this experiment, a given sample of UCAR grade TSP graphite was exposed in turn to a series
of wet and dry storage conditions, the gas content being measured at the conclusion of each exposure. The
sample thus entered each new exposure in a relatively gas-free condition (see Table 6.5).

6-3
Table 6.5 Gas Content of TSP Graphite

Storage Period, Gas Evolution,

Exposure No. Type Atmosphere Days mm3/g at 1800oC
0 As Received 62
1 Air + H2O 8 104
2 Air + H2O 5 85
3 Air + P2O5 23 27
4 Air + P2O5 25 23

Nuclear irradiation has also been shown to have a marked effect on the amount of gas evolved for reactor
graphite(5). Typical data, such as are plotted in Figure 6.2, show that prior irradiation almost tripled the amount
of gas evolved at 1000ºC and almost doubled it at 1400ºC. The tendency for the curves to come closer at high
temperatures suggests that both temperature and irradiation act to accelerate the processes which release
gas for the same potential sources in the graphite.

Figure 6.2 Outgassing of Graphite vs. Temperature

500

Approx Weight - 0.47 g

Volume of Evolved Gas, mm 3/g

400
Approx Weight - 0.66 g

300

Irradiated to
17
200 10 NVT (Fast)

100 Unirradiated

0
0 500 1000 1500
o
Temperature, C

6-4
Oxidation
The fact that graphite burns to form CO and CO2 is very well known. There is a tremendous quantity of
literature dealing with this apparently simple phenomenon. The reason relates directly to the form in which
industrial graphite exists - a porous mass of solid particles of assorted kinds and sizes, held together with
a bond carbon formed during fabrication of the piece. Relatively little agreement exists among the findings
of many investigators who have studied the graphite-oxygen (or carbon-oxygen) reactions, because the rate,
and perhaps even the mechanism of the reaction, depends strongly on the nature of the carbon, the impurities
present, degree of graphitization, particle size, etc. These vary widely from one graphite grade to another.

Graphite is actually less reactive to oxygen at low temperatures than many metals, but since the oxide is
volatile, no protective oxide film is formed. Still, the rate of oxidation of graphite is slow enough to permit its
very effective use in oxidizing atmospheres at extremely high temperatures where a modest consumption rate
can be tolerated. For example, a one-half inch diameter all-carbon arc lamp electrode operates in the open air
with the tip at about 4000K and a consumption rate of no more than three inches per hour - half of which
is the result of direct volatilization from the arcing tip and half is due to oxidation. When the arc current is
interrupted, the electrode promptly cools, showing that ignition or self-sustained combustion does not occur
even after heating to this very high temperature.

The differences in oxidation behavior of the various grades of graphite are widest at the lowest temperatures,
tending to disappear as the temperature increases. If “threshold oxidation temperature” is defined as that at
which a sample loses approximately one percent of its weight in 24 hours, pure graphite has a threshold
oxidation temperature of 520 to 560ºC, depending on coke source and processing variables. Small amounts of
catalysts are very effective in increasing the oxidation rate at low temperature, and there are enough of these
catalysts in commercial graphite to lower the threshold oxidation temperature by as much as 100oC. Tests of
individual catalysts as additives to the graphite have shown, for example, that 20 to 40 parts per million of
sodium, potassium, vanadium, or copper will increase the rate of oxidation by a factor of six at 550oC, thereby
reducing the time for a five percent oxidation from 120 hours to 20 hours.

In the temperature range 400 to 650oC, just above the threshold oxidation temperature, the oxidation is
primarily penetrative in character, such that the weight loss is greater than the volume loss. At still higher
temperatures, the oxidation is diffusion-controlled, with virtually all oxidation taking place at the outside
surface. In this diffusion-controlled temperature region, the oxidation rate is extremely sensitive to the velocity
of air movement over the surface.
The oxidation of graphite has been measured between 425 and 575ºC in pure oxygen at pressures of 0.15
to 9.8 cm Hg(13). It was found that the rate at which the samples lost weight with time could be expressed
according to the equation:

W = Kt + Ct2

where t is the time and K and C are constants. The activation energy was found to be 36.7 kcal/mol.

Oxidation tests in air have been carried out on a variety of purified graphite grades at 500, 600, and 700ºC(6).
The rate of oxidation was found to increase with time, but leveled off after the samples had undergone a
weight loss of several percent. The oxidation rates during this latter period of oxidation varied with temperature
in a manner consistent with an activation energy of 55 kcal /mol, averaged for ten different graphite grades.
Since this value refers to a different part of the oxidation curve, it is not inconsistent with the value of 36.7
kcal/mol cited above.

To indicate the order of magnitude of oxidation rates in air, some of the average results on both purified
and general-purpose graphites at 500, 600, and 700oC are cited in Table 6.6. The samples were 1.25 inches
diameter by 2 inches long, and were suspended in a furnace from an automatic recording balance. Individual
values for different graphites deviated from the indicated averages by as much as 30 percent.

6-5
Table 6.6 Typical Values of Oxidation Rates of Graphite in Air

No. of Oxidation Time for 1% Time for 6%

Graphite Varieties in Temperature, Weight Loss, Weight Loss,
o
Sample Average C hr hr
Pure 13 500 200 -
Pure 13 600 7 -
Pure 13 700 0.6 1.8
Commercial 1 500 6 -
Commercial 6 600 0.75 2.5
Commercial 6 700 0.2 0.7

A critical study of the graphite oxidation reaction has been published in a series of papers by Eyring and his
associates(7-9). After studying the heterogeneous gas solid reaction in the range of 1-100 micron pressure and
600 to 1300oC, they report that below 800ºC, the surface reaction is of zero order with an activation energy of
80 kcal/mol. On samples thicker than 0.1 mm, the diffusion of oxygen into the pores of the graphite results in
an observed half order reaction with 42 kcal/mol activation energy. Above 1200oC, the reaction is controlled by
the frequency of oxygen collisions with the surface. Between 800 and 1200ºC, a continuous transition region
exists. It is suggested that whenever an O2 molecule is absorbed on a C surface, an activated complex is
formed which may break up into C (surface) + O2 or into 2 CO, and that the latter reaction is favored by the
potential energy situation prevailing under most circumstances. Eyring et al. point out that a great deal of the
literature on carbon oxidation deals with materials which have been carbonized but not graphitized. These
materials are usually oxidized in first order reactions with an activation energy of 20-30 kcal/mol.

Oxidation of graphite is also produced by steam and carbon dioxide atmospheres, although the threshold
oxidation temperature is higher for these gases. Typical manufactured graphite will have a threshold oxidation
temperature of about 700oC in steam and 900ºC in carbon dioxide, see Table 6.7. Graphite consumption by
hydrogen and nitrogen is significant at temperatures above 1000oC for the former and 1500oC for the latter.

Table 6.7 Threshold Oxidation Temperatures of Graphite in Various Atmospheres

Atmosphere Temperature, oC
Ai r 400
Steam 700
CO2 900

6-6
The amount of total ash and impurities in graphite plays a significant part in determining oxidation characteris-
tics. Low-ash graphite, such as CBY and CS, have significantly lower oxidation rates than do commercial
graphites having 1.5 percent ash. Similarly, purified graphite having ash levels less than 10 ppm have even
lower oxidation levels. Table 6.8 and Figure 6.3 show this comparison(19).

Table 6.8 Oxidation of Graphite

High-Ash Purified
Grade Grouping Commercial CBY-CS CBY-CS
Ash, typical, % 1.5 0.05 0.001
Grain Size, max, inches 0.03 0.03 0.03
(1)
Oxidation Rates
6 0 0 oC 66 38 12
o
650 C 243 126 53
o
700 C 651 343 210
Threshold Oxidation
Temperature(2) oC 470 560 550
(1) (g/m2)/hr
(2) 1% weight loss in 24 hr

Figure 6.3 Oxidation Rates of Graphite

700

600
1.5% Ash
Oxidation Rate, (m 2 )/hr

500
_g

400

300 0.05% Ash

200

0.001% Ash
100

0
500 600 700
o
Temperature, C

6-7
Oxidation Retardation
Various forms of UCAR graphite are used in applications where resistance to oxidation is a primary
consideration. Research on the oxidation characteristics of UCAR graphite has shown the following:

(1) Untreated graphite obeys an Arrhenius activation energy equation within certain temperature ranges
up to 700ºC. Above 700ºC, oxidation is not governed solely by temperature.

(2) The treatment of UCAR graphite materials with an oxidation retardant materially enhances oxidation
resistance. In general practice, a typical commercial oxidation retardant maintains a high degree of
efficiency and is superior to untreated graphite at temperatures up to 750ºC.

Tests on graphite tubes summarized in Table 6.9 show the effectiveness of a commercial oxidation retardant
treatment. The tests were conducted over a 24-hr period in an air atmosphere. The specimens were
thick-walled pipes 2 inches outside diameter, 0.5 inches inside diameter, 5.5 inches long, and weighing
approximately 500 g. The anti-oxidation treated samples are about ten times more resistant to oxidation at
750ºC than untreated specimens.

Table 6.9 Effect of Air Oxidation on Untreated and Treated Graphite

Oxidation Rate, (g/m2))/ hr

Graphite at 650ºC at 750ºC
Untreated
UCAR CBY-Type 230 1800

UCAR CS-Type 340 1900
Treated 23 200

6-8
Resistance to Erosion and Attrition
The ability of graphite to resist erosion or abrasion, either mechanical or chemical, is important in its
applications as structural components in guided missiles and as containers for liquid metals in nuclear reactor
high-temperature heat transfer systems.

The erosive effect of an inert hot gas on graphite has been shown to be negligible, e.g., a helium jet at
2000ºC and a velocity of 900 ft/sec produced, in 24 hours, no more than an initial change in surface quality(10)
attributable to a blasting away of particles previously loosened in machining. However, any oxidizing gas will
react with graphite (even at moderate temperatures) to chemically erode it. A sharp distinction must, therefore,
be made between chemical erosion and mechanical erosion. Molecular nitrogen, which is inert, will act as
the helium did. Nascent nitrogen, on the other hand, will readily react with graphite at high temperatures to
produce cyanogen, and thus erode the graphite.

An example of chemical erosion caused by oxidation is seen in Table 6.10(11). A supersonic kerosene-oxygen
jet torch operating at a chamber pressure of 160 psig, an exhaust gas velocity of 6000 ft/sec, and a
temperature of 2760ºC, was used. At lower chamber pressures, the exhaust velocity and temperature were
somewhat reduced.

Table 6.10 Oxidation Erosion of Graphite Nozzle in Supersonic Kerosene - Oxygen Torch(11)

UCAR Chamber Number Life, seconds

Grade Pressure, psi of Tests Maximum Minimum Average
ATJ 100 5 43 36 40
125 5 18 11 14
CS 100 5 33 15 20
125 5 5 2 4
135 4 4.5 2 3
AGSX 100 3 27 12 19
AGSR 100 5 22 12 17.5

The graphite grade to be tested formed the nozzle for the torch, simulating the role of the rocket nozzle.
The life indicated in seconds is arbitrarily taken as the time required for a five-percent decay in the original
peak chamber pressure of the torch, which is a direct measure of the throat enlargement (due to wear)
of the nozzle.

Similar results are reported in measurements of erosion caused by directing the exhaust jet of a rocket motor
against graphite test specimens(12). The test samples, in this case, were graphite plates 4 1/8 inches square by
7/16 inches thick. The 2700K exhaust jet was adjusted to cover the entire face of the test plate. Dimensional,
weight, and appearance changes following a 15-second exposure to the jet were recorded. These results
are presented in Table 6.11.

6-9
Table 6.11 Oxidation Erosion of Graphite by Rocket Exhaust Jet Directed Against Graphite Surface

Erosion Rate
UCAR Temperature, By Thickness, By Weight, Surface
o
Grade C inches/second g/second Appearance
AGSR 850 0.004 1 Rough, Pitted
CS 750 0.003 0.7 Slightly Rough
ATJ 465 0.002 0.2 Slightly Rough

While densification (impregnation) and other processing modifications effect small increases in abrasion
resistance, the largest single influence for mechanical abrasion is raw materials. Petroleum coke type is
the biggest factor. Table 6.12 shows the influence of raw materials on the mechanical abrasion of various
graphites. Coke Type A is characterized as isotropic and shows the least weight loss on abrasion. Coke
types B and C are anisotropic cokes with relatively low CTE and SR values, where abrasion is simulated
in an in-house test by rotating a cylindrical specimen in a bed of sized aluminum particles, and measuring
the weight loss with respect to time.
Table 6.12 Effect of Raw Materials on Graphite Abrasion Properties

Abrasion Resistance,
C o ke BD , SR, Hardness, % Weight Loss after
Type Densification g/cc µΩm Rockwell “R” 10 30 50
minutes minutes minutes
A No 1.6 12.7 91 2 6 9
A Yes 1.7 11.7 99 2 5 8
B No 1.64 8.9 62 8 18 24
B Yes 1.72 8.6 67 7 18 23
C No 1.7 8.6 65 7 17 22
C Yes 1.77 9.5 79 6 15 20

Heat of Combustion
The heat of combustion of graphite is a reproducible thermochemical quantity that has been determined with
high precision. It measures the heat that is liberated when graphite is burned with oxygen to form CO2. A
U.S. Bureau of Standards(14) compilation gives the “standard heat of formation” of CO2 as 94.0518 kcal/mole.
More precisely, this refers to the “standard heat of formation” and “standard states” as defined and used
in thermochemistry(14). Practically, it agrees within a fraction of a percent with the heat of combustion at 10
atmospheres pressure and 25ºC.

Several high-precision studies have been made at the National Bureau of Standards. Dewey and Harper(15)
found, in 1938, that several low-ash natural graphites and spectroscopic-grade artificial graphite gave the
same heat of combustion. Jessup(16) confirmed this and found that diamond has approximately 0.5 percent
larger heat of combustion than graphite. Prosen and Rossini(17) made refined measurements in 1944 on some
of the same material used earlier by Jessup(16). These men(18) used these results along with a recalculation
of some of the earlier data to obtain the currently accepted value(14). Measurements on anthracite cokes
indicated that the heat of combustion of these depended on the hydrogen content, but in the hydrogen-free
form, it was about 2.5 percent higher than for graphite(15).

Heat of combustion measurements made on reactor-irradiated graphite samples are affected by the amount
of stored energy.

6-10
 References
1. U.S. Patent No. 2,456,099 (1948).
2. Eggleston, R. R., Carter, R. L., Greening, W. J., and Durand, R. F., North American Aviation Co.
Report,NAA-SR-MEMO-1240 (1955).
3. Carter, R. L., North American Aviation Co., Private Communication, NCRL-10.
4. Mitchell, C. V., Unpublished Data, National Carbon Company Research Laboratories, NCRL-366
(1956).
5. Carter, R. L. and Greening, W. J., Paper No. 319 Presented to the American Nuclear Society,
Cleveland,Ohio, December 12-16, 1955.
6. Mitchell, C. V., Unpublished Data, National Carbon Company Research Laboratories, NCRL- I I
(1954-5).

7. Ruoff, A. L. and Eyring, H., University of Utah, Report AFOSR-TN-55-312 (1955).

8. Blyholder, G. D., Binford, J. S., and Eyring, H., University of Utah, Technical Memorandum No. XII
(1955).
9. Blyholder, G. D. and Eyring, H., University of Utah, Report AFOSR-TN-56451 (1956); J. Phys. Chem.
61, 682 (1957).

10. Green, L., North American Aviation Co. Report, NAA-SR-77 (1950).

11. Manofsky, M. B. and Kemp, J. G., Unpublished Data, National Carbon Company Research
Laboratories, NCDL-3.
12. Gates, D. W., Private Communication, U.S. Army Ballistic Missile Agency, Redstone Arsenal, Huntsville,
Alabama (1957).

13. Gulbransen, E. A. and Andrew, K., Ind. Eng. Chem. 44, 1034 (1952).

14. U. S. Bureau of Standards, Circular 500, pp. 98-99 (1952).

15. Dewey, P. H. and Harper, D. R., J. Res. NBS 21, 457-474 (1938).

16. Jessup, R. S., J. Res. NBS 21, 475-490 (1938).

17. Prosen, E. J. and Rossini, F. D., J. Res. NBS 33, 439-446 (1944).

18. Prosen, E. J., Jessup, R. S., and Rossini, F. D., J. Res. NBC 33, 447-449 (1944).

19. Page, D. J., Unpublished Data, UCAR Carbon Co. Inc., Parma, Ohio (1988).
Chapter Seven

INDUSTRIAL APPLICATIONS OF
BULK CARBON AND GRAPHITE
Industrial Graphite Engineering Handbook
 TABLE OF CONTENTS

Introduction.......................................................................................................................7-1
Metallurgical ..................................................................................................................7-1
Boats and Trays ............................................................................................................7-1
Crucibles .......................................................................................................................7-1
Fluxing, Injection Tubes.................................................................................................7-1
Molds and Dies..............................................................................................................7-2
Furnace Parts................................................................................................................7-2
Foundry Accessories .....................................................................................................7-2
Aluminum Processing Parts ..........................................................................................7-2
Chemical Process Equipment .......................................................................................7-3
Electrical........................................................................................................................7-3
Electrical Discharge Machining (EDM)..........................................................................7-4
Resistance Heating Elements .......................................................................................7-4
Electronic.......................................................................................................................7-4
Fiber-Optic.....................................................................................................................7-4
Semiconductors.............................................................................................................7-5
Electrolytic .....................................................................................................................7-6
Aerospace .....................................................................................................................7-6
Rockets and Missiles ....................................................................................................7-6
Tooling ...........................................................................................................................7-7
Nuclear ..........................................................................................................................7-7
Introduction
This chapter is concerned with industrial applications of bulk carbon and graphite. This is not an exhaustive
treatment, but rather a brief survey of metallurgical, chemical process equipment, electrical, electronic,
electrolytic, aerospace, and nuclear applications. For greater detail, and for information about applications
which are not mentioned here, please contact a UCAR Carbon representative.

Metallurgical
Metallurgical applications are defined as those in which graphite comes in contact with metal, normally in
the presence of high temperatures. Since the processing of metals generally involves high temperatures, the
term metallurgical applications has, in many cases, become synonymous with high-temperature applications.
The properties of graphite that make it particularly useful in a broad range of metallurgical applications are
listed below.

• Unreactive with many molten metals (e.g., Zn, Al, Mg, Cu and Cu Alloys) and metal-producing
slags.
• Does not adhere to most castings, billets, and ingots.
• Strength properties increase as temperature increases.
• Low thermal expansion; high dimensional stability.
• High thermal-shock resistance due to a combination of high thermal conductivity, low Young’s
modulus, low coefficient of thermal expansion, and high strength at temperature.
• Will not melt at high temperatures under ordinary pressures.
• Machinability.
• High strength-to-weight ratio.
• Performance at high temperatures.

Boats and Trays

Graphite is used extensively in the sintered carbide and powdered metallurgy fields for molds and crucibles,
pre-sintering and sintering boats, and trays, as well as for hot pressing and other methods of carbide and
sintered carbide production. Graphite is also used widely for trays and reference forms in furnace brazing.

Crucibles
Since graphite may be heated either by induction or electrical resistance, as well as by conduction or radiation
from other sources, it is an excellent crucible material. Customary crucible sizes vary from the small, vacuum-
fusion-type (less than 1-in. diameter) to the large type (up to 73-in. diameter) for melting beryllium and
uranium and other high-melting point metals. Most metals may be melted directly in graphite since graphite is
not attacked, and the metal not contaminated.

Fluxing, Injection Tubes

Graphite fluxing tubes are used (1) for the introduction of gas into the gas purification of molten aluminum and
other nonferrous metals, (2) for the desulfurization of blast furnace iron, and (3) for the injection of graphite
powder into foundry iron. Graphite is well established as the standard fluxing tube material in many industries.
Fluxing tubes can also be treated with oxidation retardants for resistance to oxidation.

7-1
Molds and Dies
Graphite may be adapted to molds for static castings of most metals or to molds or dies for the continuous
casting or drawing of nonferrous metals. Such products vary in size from molds with cavities of one inch
diameter by one inch depth to molds used in the centrifugal casting of bronze and bearing metals with a
73 inch diameter and 72 inch length. Other mold and die applications include molds for the hot pressing of
powdered metals and ceramics, ingot molds, jominy bar molds, pig and over metal molds and precious metal
molds. Graphite is also used in pressure and static casting of carbon steels as a mold to form precision cast
parts or slabs. Graphite can be remachined to produce hundreds of castings.

Furnace Parts
Because of desirable refractory properties, in addition to relatively high resistance, graphite makes excellent
resistor elements where they may be operated in a reducing or non-oxidizing atmosphere, or in a vacuum.
Graphite can be used at temperatures up to 2800°C in inert atmospheres. The temperatures attainable with
graphite are well above those of any other known resistor elements.

Graphite is also used extensively in atmospheric-controlled furnaces as a lining material. Its high thermal
conductivity, low thermal expansion, and ease of machinability are characteristics conducive to lining furnaces
in virtually any shape or size. Graphite slabs can be bolted together with a variety of jointing techniques.
On any structural furnace part (hearth, wall, roof, door) where there is a non-oxidizing atmosphere, graphite
should be considered as a material of construction.

Graphite and semi-graphites are also used in blast furnaces as hearth, bosh, and stack materials. Its high
thermal conductivity, thermal shock resistance, high-temperature dimensional stability, and machinability are
ideal characteristics for cost-efficient linings.

Foundry Accessories
In foundries, graphite plates and rods are used as molds, chills, and cores because of their rapid chill
characteristics, resistance to metal attack and smooth surfaces. Graphite riser rods are used in conjunction
with steel castings to add carbon to the riser metal in order to maintain a fluid liquid head of metal for a longer
period of time to feed the casting during cooling and shrinking. In brass foundries, graphite skimmer floats
are used in pouring crucibles to hold back the dross from the metal being cast. In nonferrous melting pots,
graphite rods are used for stirring since this material does not contaminate the melt, does not crack, and does
not have an affinity for the molten metal.

Aluminum Processing Parts

In the extrusion of small-section aluminum shapes, such as angles, counter edge- concave and convex
beadings, window frame channels, etc., it is general practice to use multiple die openings in each cylinder.
Where as many as four or six such shapes are being extruded simultaneously, the tendency to twist or curl
upon leaving the die causes considerable tangling of the leading ends of the several extrusions and the
subsequent loss of considerable material in removing the tangled ends. Tangling may be prevented by using
graphite extrusion guides which also assure freedom from scratching of the aluminum between exit from the
die and alignment on the runout table. The guide may be a simple graphite cylinder having rectangular or
pie-shaped openings that run the length of the graphite tube between the bolster and runout table for whatever
distance may be required to ensure satisfactory operation.

7-2
Chemical Process Equipment
As a construction material in chemical processes, manufactured graphite offers a unique combination of
properties: resistance to practically all corrosives (except highly oxidizing conditions), inertness to thermal
shock, high thermal conductivity, thermal stability and good machinability - all at moderate cost. Some of the
major applications of graphite, illustrating its wide use in the chemical industry, follow.

Graphite vessels and reactors are used in applications, such as burning hydrogen and chlorine to form
hydrogen chloride, and burning phosphorus and its subsequent hydration product to form phosphoric acid.
Such vessels range in size from 12 inches to 20 feet in diameter by 10 to 30 feet in height and are constructed
from either single graphite shapes or fabricated from premachined graphite blocks. Units can be arranged for
water cooling to remove the heat of reaction. Because of the chemical inertness of graphite, impure reactants
can be handled without concern.

When installed as a vessel lining, graphite provides inertness to chemical attack, resistance to abrasion,
and a heat barrier to protect the vessel proper or its inner lining (either metallic or non-metallic). Graphite
linings can be installed in vessels of any conceivable size and cross section. They can be erected with either
a keyed construction to provide mechanical stability or with suitable chemical and heat-resisting cements.
Graphite-lined vessels of this type are well suited for high-temperature processes, such as chlorination
reactions for the production of boron trichloride and other metallic chlorides.

Because it offers resistance to corrosion, high temperature, and thermal shock, and can be easily fabricated,
graphite has many internal uses in chemical processing equipment. Typical are graphite bubble caps and
trays, sieve plates, and packing support plates. Graphite is employed for constructing submerged guide rolls
used in the production of synthetic fibers and films because of its resistance to corrosive action and freedom
from metallic contamination. Graphite rolls are also used as conductor rolls on electrolytic tin-plating lines
where the electrical conductivity of graphite is an additional advantage.

A complete line of chemical processing equipment items can also be manufactured. In this field of application,
graphite is outstanding in the construction of heat exchangers of all types. Here it offers unusually high heat
transfer rates together with freedom from corrosion, metallic contamination and thermal shock. Because of
its ease in machining and fabrication, graphite is an economic material of construction for centrifugal pumps,
valves, pipe and fittings, towers, hydrochloric acid absorbers, and a variety of process accessories.

Electrical
In electrical and electronic applications, graphite is used primarily for component parts, such as vacuum power
tubes and resistor elements. It is also used to some extent as electrical contacts, brazing tips, etc. The
properties of graphite of particular utility for electrical applications are summarized in the list below:

• Low gas content.

• High thermal emissivity.
• High thermal conductivity.
• Low coefficient of thermal expansion.
• Low vapor pressure.
• Good electrical conductivity
• No electron emission at high temperatures.
• High-temperature strength.
• Machinability.

7-3
Electrical Discharge Machining (EDM)
Electrical discharge machining or EDM, as it is most commonly known, uses the energy contained in an
electrical discharge to remove metal. These discharges are generated many thousands of times per second.
Simple or complex operations, such as through-hole boring or cavity sinking in any electrically-conductive
work material, including carbide, tool steels, high-temperature alloy steels, and metals of all types, are made
possible with this process. The EDM process will machine any electrically-conductive metal, regardless of
its hardness.

Electrical discharge machining is used for a variety of machining applications and is especially noted for its
ability to produce intricate shapes, while holding extremely close tolerances. For many machining jobs, EDM
is the only viable means to manufacture a particular shape. The technique may be used in the manufacture of
forging dies, cavity inserts, plastic injection molds, and many tools.

UCAR Carbon superfine and ultrafine graphites are used in EDM applications for the following reasons:

• Relatively inexpensive
• Available in a variety of sizes
• Readily machinable to complex shapes with conventional tools
• May be used in “No-Wear” (positive electrode) machining
• Provide highest metal removal rates with lowest wear
• Are non-melting.

The EDM operation is a melting or vaporization process: the spark energy jumps across the gap to melt
or vaporize a portion of the workpiece. This energy also affects the electrode. If the electrode is made of
the same material as the workpiece, the electrode will be eroded at the same rate as the work piece - a
“one-to-one” wear ratio. Graphite does not melt. Instead, it sublimes at a temperatures above 3000oC. Thus,
graphite resists the spark action, making wear ratios of 100:1 and greater possible.

Resistance Heating Elements

Resistance heating elements may be fabricated in almost any shape and size from round rods to rectangular
beams joined and fitted to obtain the desired resistance for a given application. Assemblies can be arranged
in banks, series, or in series-parallel. Graphite tubes or slabs may also be machined to form a hairpin resistor
or to form an extended path by alternate saw cuts resembling numerous hairpin resistors in series; or, the
graphite cylinder may be machined into a single or double helix to provide the maximum resistance in a given
space. A resistor suitable for three-phase operation may be machined from a single graphite tube, connections
for such resistors may be made from water-cooled copper parts; or, the graphite may be joined directly by
threading, pressed fit, or spring fit, which is provided by making a saw cut in the end of a graphite rod for
insertion into a drilled hole a few thousandths of an inch smaller in diameter than the resistor rods.

Electronic
Fiber-Optic
High grade quartz is drawn at temperatures around 1700°C to form a cable with fiber-optic qualities. In
one industrial process the furnace container and heater are fabricated from high purity isomolded superfine
graphite such as UCAR DM302. Graphite with its combination of thermal characteristics, chemical inertness,
high purity, superfine grain texture, and structural integrity at elevated temperatures is eminently suitable.

7-4
Semiconductors
There are several major applications for graphite in the semiconductor sector with most requiring purification
to at least 10ppm ash.

Single Crystal Growth - Silicon Carbide: Purified graphite with its enviable thermal properties and inertness
provides a suitable container material in which single crystals of silicon carbide can be grown. The isomolded
grades are ideal for this application.

Single Crystal Growth - Silicon: Silicon is the most widely used material for the manufacture of integrated
circuits and other semiconductor devices. Silicon, the second most abundant element, comprising 28 percent
of the earth’s crust, is a good material for circuits because its electrical properties can be precisely altered
by adding controlled amounts of impurities, called dopants. One of the most essential operations in the
manufacture of an integrated circuit is the conversion of this silicon into single crystal form. The most common
method of converting the polycrystalline silicon to a single crystal ingot is through growing the ingot from
molten silicon by the Czochralski crystal-growing process. This process involves wetting a properly-oriented
seed in a melt and withdrawing it vertically to grow a crystal which might be 12” (300 mm) in diameter.

Graphite is extensively used in this application for a number of reasons. First, graphite has electrical proper-
ties that are necessary for good heater and crucible- operation. Second, it has the required thermal properties.
Material is needed that can withstand temperature recycling from room temperature to above 1400°C,
maintaining strength at these high temperatures. Graphite has excellent thermal conductivity, enabling it to
maintain thermal equilibrium in the growing process. Third, the combination of cost, availability, machinability,
and life of the graphite parts has not been improved upon by other materials.

Purified medium-grain graphites, such as UCARGrades CS and PGW, are used as resistance heaters, which
melt the silicon and maintain it at a temperature of approximately 1400°C. UCARGrades SLWX and purified
CS and purified isomolded superfine are used as holders or crucibles to hold a quartz crucible liner containing
the molten silicon. These same grades are used as a graphite pedestal which support the crucible, rotate it,
and lift it as necessary. The graphite is subjected to the same thermal conditions as the crucible and also acts
as a path for removal of excess heat of fusion from the puller. A graphite heat shield, surrounds the heater.
This shield prevents the loss of heat outward from the crucible area. Lastly, graphite chucks are used in the
polysilicon reactors. These hold filaments upon which the polysilicon is grown.

The choice of grade depends upon the size of the hot zone, customer preferences, and the graphite
performance required.

Epitaxial Deposition: Another application for graphite is epitaxial deposition, a process utilized throughout
this industry for doping silicon or III - V compound substrates. Epitaxy, in silicon semiconductor technology,
refers to the oriented overgrowth of one crystalline material on to another. This process is known as “vapor
phase epitaxy” and uses silicon carbide coated graphite susceptors to hold the silicon wafers. Controlled
doping of “P” or “N” type impurities is also possible in silicon epitaxial technology.

Graphite is used in epitaxial deposition because of its high purity, electrical properties, thermal conductivity,
low gas evolution, and coating characteristics. Graphites are generally superfine grain and possess CTE
characteristics compatible with the coatings.

Glass-To-Metal Sealing: In glass-to-metal sealing, a “chip” mounted to a metal lead frame must be protected,
quite often by glass encapsulation. The device is mounted in a graphite fixture, heated to a desired
temperature and the molten glass applied. Graphites are generally used due to their low gas evolution, high
thermal conductivity, ability to be machined to very close tolerances, excellent resistance to abrasion and
wear, and superior oxidation resistance.

7-5
Coatable Graphites: Coating of graphite with silicon carbide extends the applications of graphite into high
temperature oxidation regimes and eliminates interaction with potentially reactive materials such as silicon.
The UCAR superfine grain grades are eminently suitable substrates for CVD, CVI, PBN, and SiC coatings.

Electrolytic
Several basic chemicals are manufactured by electrolytically decomposing compounds into their elements.
Graphite can be used for either the anode or cathode in electrolytic applications.

Graphite, in performing its function in an electrolytic cell, must meet seven fundamental requirements:

1. High electrical conductivity - high electrical resistance results in expensive power losses. As a
conductor of electricity, graphite holds a position between the lesser conducting metals (iron and
steel) and the best conducting metals (copper and aluminum).
2. Operate at low voltage. Graphite has a lower anode voltage than any other practical material except
precious metals when used for the electrolysis of brine. Graphite possesses a remarkably low
chlorine discharge voltage and a relatively high oxygen discharge voltage. Platinum exhibits these
properties in some degree and lasts longer than graphite, but it operates at a higher voltage and
is much more expensive.
3. High purity - the cell products must be protected from contamination. Graphite can be manufactured
to almost any desired purity.
4. Satisfactory mechanical strength - the mechanical strength of graphite anodes has been found
satisfactory in all electrolytic applications.
5. Easily machined and readily assembled - the remarkable ease of machining graphite permits its
use in any type cell, no matter how intricate the design. Its comparatively light weight makes
assembly relatively easy.
6. Low initial cost - graphite is an inexpensive material, particularly when compared with its nearest
rival (platinum).
7. High durability - with an expected life of 200 to 800 days in modern cells, depending on the type
and the severity of operation, graphite meets this basic practical requirement to a high degree.
Purity of the graphite improves life since iron and other metallic contaminants can act as oxidation
catalysts.

Electrolytic manufacturing of chemicals is conducted by two methods: solution electrolysis and fused-salt
electrolysis. Fluorine, chlorine, chlorate, and manganese dioxide are liberated from water solutions while
magnesium and sodium are generated from molten salt solutions.

Aerospace
Rockets and Missiles
The vast field of rockets and missiles takes advantage of many of the properties of graphite. Applications
include tactical rockets, strategic rockets and missiles, and large, advance-launch systems. Rapid tempera-
ture rise and unusually high operating temperatures are encountered and unusual cone, nozzle and vane
shapes are needed. Graphite is one of the few materials that can reasonably meet the demands encountered
under these conditions.

Of particular importance in this type of application are the excellent thermal properties of graphite - high
thermal shock resistance, high thermal stress resistance, and a strength increase with temperature increase.
In addition, its excellent machinability makes it possible to maintain the required close tolerances.

For most short-time applications, firing times are normally less than one minute, graphite is very cost effective
in the hot, reactive atmosphere.

Tooling

7-6
With a proven application history involving high-temperature casting and shaping, monolithic graphite should
be considered as a viable tooling material for forming composite parts.

While desirable tooling parameters depend on the circumstances of composite part manufacture, the following
are generally agreed upon as being most influential: thermal expansion and thermal conductivity.

The thermal expansion of the tool must be compatible with the composite part. This characteristic becomes
even more critical when the part manufacturer requires higher temperatures or as part sizes become larger.
This compatibility of thermal expansion between tool and part allows the machining of precise dimensions and
tolerance directly eliminating costly intermediate steps.

High thermal conductivities are also necessary. The ability to transmit or dissipate heat rapidly is necessary
for rapid cycle turnaround times. Monolithic graphite thermal properties match these characteristics.

The structural characteristics of the tooling material are equally important. Integrity of the working tooling
surface must be assured by providing a low porosity material void of defects and uniform in physical
properties.

Composite parts require exact dimensions and tolerances, especially where assembly to related parts
is required. Tooling must be machined to precision tolerances with resultant smooth working surfaces.
Multiple axis machining is necessary with integrated CAD/CAM link-ups between the tooling designer, tooling
manufacturer, and machiner. Lead times are significantly reduced if this linkage is available.

Monolithic graphite machines comparable to hard wood with ordinary hardened steel tooling. It must be
machined dry. Stress relieving is not required.

Techniques are regularly employed to repair graphite. Also, since the material is easily machinable, design
changes or contour modifications can be accommodated at a reasonable cost by remachining.

The high-temperature thermal stability and the machining innovation possible with monolithic graphite allow
this material to be used for high-temperature composite resin systems for complex shapes and sizes.

Nuclear
Graphite finds widespread use in many areas of nuclear technology based on its excellent moderator and
reflector qualities, which are combined almost uniquely with strength and high temperature stability. The
function of a moderator is to slow fast neutrons to thermal velocities at which fission in Uranium-235 and
Uranium-233 are most efficient. The reflector serves to reflect neutrons, which otherwise would escape, back
into the active core region.

The UCAR Carbon Company has participated in the development and production of nuclear grade graphite for
fission reactors since the beginnings around the Manhattan project in the 1940s. The low-boron moderator
UCAR grade AGOT was developed for the air-cooled X10 reactors at Oak Ridge, TN, and the first water-
cooled reactors at Hanford, WA. Later water-cooled reactors built at Hanford were moderated with purified
grades CSF, GBF, and TSF, respectively.

Several carbon dioxide cooled graphite moderated commercial reactors were built in the UK, France, and
Japan. UCAR Carbon produced grade UCAR grade PGA as the life-of-reactor moderator for use in the
post-war UK Magnox design. This PGA is notable in that like its UCAR grade AGOT predecessor, it was not
purified, but produced from specific “clean” (low impurity) raw materials. Many of the Magnox reactors built in
the 1950’s and 1960’s are now decommissioned.

The later, higher operating temperature, Advanced Gas Reactors (AGR) built in the 1970’s and 1980’s use a

7-7
UCAR grade GCMB as the life-of-reactor moderator. The AGR design incorporates a replaceable graphite
sleeve, UCAR purified grade PPEA.

Two designs of the most advanced gas cooled reactor type, the so-called High Temperature Reactor (HTR),
use helium as the coolant and graphite as the moderator. In the one design, demonstrated in small reactors
in the UK (Winfrith), and USA (Peach Bottom and Fort St. Vrain), the moderator is the usual prismatic
bed characteristic of the Magnox and AGR types, but it is unique in that it is replaceable. There is a
demonstration unit currently critical in Fuji, Japan. The alternative Pebble Bed Reactor (PBR) of German
design demonstrated in the Jülich and THTR-300 units uses a moving bed of fuel embedded in 50mm
diameter spherical graphite “pebbles”.

7-8
Glossary

Industrial Graphite Engineering Handbook

Glossary
Terminology Relating to Manufactured Carbon and Graphite

Acheson graphite - a synthetic graphite made by the Acheson process.

across grain - synonym for against grain (AG).

activated carbon - a porous carbon material, a char, which has been subject to reaction with gases,
sometimes with the addition of chemicals, e.g. ZnCl2, before, during, or after carbonization in order to increase
its adsorptive properties.

activated charcoal - a traditional term for activated carbon.

against grain - (AG) the direction in a carbon or graphite body with preferred orientation that has the
maximum c-axis alignment (synonym across grain).

agglomerate - a composite particle containing a number of grains.

amorphous carbon - a carbon material without long-range crystalline order. Short-range order exists, but
with deviations of the interatomic distances and interbonding angles with respect to the graphite lattice as
well as to the diamond lattice.

anisotropic - properties are not equal in all directions.

artificial graphite - synonym for synthetic graphite.

ash - the residue remaining after the complete oxidation of a carbon or graphite.

baking - the process in which the carbonaceous binder, usually coal-tar pitch or petroleum pitch, as part of a
shaped carbon mix, is converted to carbon yielding a rigid carbon body by the slow application of heat.

binder - a substance, usually an organic material such as coal tar pitch or petroleum pitch, used to bond the
coke or other filler material prior to baking.

binder coke - a constituent of a carbon (or a ceramic) artifact resulting from carbonization of the binder
during baking.

calcined coke - petroleum coke or coal-derived pitch coke obtained by heat treatment of green coke to about
1600 K. It will normally have a hydrogen content of less than 0.1% wt.

carbon - an element, number 6 of the periodic table of elements, electronic ground state 1s22s2 2p2.

carbon - an artifact consisting predominantly of the element carbon and possessing limited long range order.
The presence of limited long range order is usually associated with low electrical and low thermal conductivity
and difficult machinability when compared with graphite.

carbon artifact - an “artificially produced” solid body which consists mainly of carbonaceous material in a
distinct shape.

carbon black - an industrially manufactured colloidal carbon material in the form of spheres and of their fused
aggregates with sizes below 1,000 nm.

G-1
carbon electrode - an electrode for an electrical application. In its green state, it comprises granular carbon
material bound with pitch.

carbon material - a solid high in content of the element carbon and structurally in a non-graphitic state.

carbon mix - a mixture of filler coke, e.g.. grains and powders of solid carbon materials, and a carbonaceous
binder and selected additives, prepared in heated mixers at temperatures in the range of 410 K - 525 K as a
preliminary step for the formation of shaped green bodies.

carbonization - a process by which solid residues with increasing content of the element carbon are formed
from organic material usually by pyrolysis in an inert atmosphere.

char - a solid decomposition product of a natural or synthetic organic materials.

charcoal - a traditional term for a char obtained from wood, peat, or coal or some related natural organic
materials.

coal-derived pitch coke - the primary industrial solid carbonization product obtained from coal-tar pitch, and
is mainly produced in the delayed coking process.

coal-tar pitch - a residue produced by distillation or heat treatment of coal-tar. It is a solid at room tempera-
ture, consists of a complex mixture of numerous predominantly aromatic hydrocarbons and heterocyclics, and
exhibits a broad softening range instead of a defined melting temperature.

coarse-grained - containing predominantly large grains that are greater than 2mm in size.

coke - a solid high in content of the element carbon and structurally in the non-graphitic state. It is produced
by pyrolysis of organic material which has passed, at least in part, through a liquid or liquid-crystalline state
during the carbonization process. Coke can contain mineral matter.

compressive strength - a property of solid material that indicates its ability to withstand a uniaxial compres-
sive load.

delayed coke - a commonly used term for a primary carbonization product (green coke or raw coke) from
high-boiling hydrocarbon fractions (heavy residues of petroleum or coal processing) produced by the delayed
coking process.

delayed coking process - a thermal process which increases the molecular aggregation or association in
petroleum-based residues or coal-tar pitches leading to extended mesophase domains.

diamond - an allotropic form of the element carbon with cubic structure (space group Oh7 - Fd3m) which
is thermodynamically stable at pressures above 6 GPa at room temperature and metastable at atmospheric
pressure. At low pressures, diamond converts rapidly to graphite at temperatures above 1900 K in an inert
atmosphere. The chemical bonding between the carbon atoms is covalent with sp3 hybridization.

electrographite - a synonym for manufactured graphite.

extruded - formed by being forced through a shaping orifice as a continuous body.

filler - carbonaceous particles comprising the base aggregate in an unbaked green-mix formulation.

flaw - a defect sufficiently greater than those typical of the morphology of a carbon or graphite body to
influence a property.

G-2
flexural strength - a property of solid material that indicates its ability to withstand a flexural or transverse
load.

flow line - a defect induced by discontinuous flow velocities during forming of molded or extruded bodies.

furnace black - a type of carbon that is produced industrially in a furnace by incomplete combustion in an
adjustable and controllable process that yields a wide variety of properties within the product.

glassy carbon - a non-graphitizable carbon with homogenous microstructure which does not exhibit any
structural components distinguishable by optical microscopy.

grade - the designation given a material by a manufacturer such that it is always reproduced to the same
specifications established by the manufacturer.

grain - a region in a carbon or graphite body that is identifiable as being derived from a particle of filler.

graphite - an allotropic crystalline form of the element carbon, occurring as a mineral, commonly consisting
of a hexagonal array of carbon atoms (space group P 63/mmc) but also known in a rhombohedral form
(space group R 3m).

graphite - an artifact consisting predominantly of the element carbon and possessing extensive long range
order. The presence of extensive long range order is usually associated with high electrical and high thermal
conductivity in the hexagonal plane and relatively easy machinability when compared with carbon.

graphitic carbons - all varieties of substances consisting of the element carbon in the allotropic form of
graphite irrespective of the presence of structural defects.

graphitizable carbon - a non-graphitic carbon which upon graphitization heat treatment converts into
graphitic carbon.

graphitization - a solid-state transformation of thermodynamically unstable non-graphitic carbon into graphite

by means of heat treatment.

green carbon - a formed, but unfired carbon body.

green coke (raw coke) - the primary solid carbonization product from high boiling hydrocarbon fractions
obtained at temperatures below 900 K. It contains a fraction of matter that can be released as volatiles during
subsequent heat treatment at temperatures up to approximately 1600 K. This mass fraction, the so-called
volatile matter, is typically between 4% and
15% by weight.

hardness - the resistance of a material to deformation, particularly permanent deformation, indentation, or

scratching.

hexagonal graphite - the thermodynamically stable form of graphite with an ABAB stacking sequence of the
graphene layers. The exact crystallographic description of this allotropic form is given by the space group
D6h4 - P63/mmc (unit cell constants: a=245.6 pm, c = 670.8 pm). Hexagonal graphite is thermodynamically
stable below approximately 2600 K and
6 GPa.

impervious graphite - manufactured graphite that has been impregnated with a resinous material to make
the final article impervious to liquids in the recommended operating range.

G-3
impervious carbon - the same as impervious graphite with the exception that the base stock has not been
graphitized.

impregnation - partial filling of the open pore structure with another material.

intercalation - the insertion of an element or compound between graphitic layers.

isostatic - equal pressure from every direction.

isotropic - equal properties in all directions.

lamination - line of demarcation or elongated void generally parallel to the principal grain direction of a
carbon or graphite body.

lamp black - a special type of carbon black produced by incomplete combustion of a fuel rich in aromatics that
is burned in flat pans. Lamp black is characterized by a relatively broad particle size distribution.

longitudinal, L - direction in a carbon or graphite body coinciding (usually) with the center cylindrical axis, or
plane, in the case of a rectangle. (Same as with grain for an extruded body, maybe either with grain or against
grain, for both mechanically molded and isostatically molded bodies depending upon the orientation of the
body in the press). Transverse (T) is the perpendicular direction to longitudinal.

machinability - a measure of the ease with which a material can be shaped with the aid of cutting or
abrasive tools.

manufactured carbon - a bonded granular carbon body whose matrix has been subjected to a temperature
typically between 1150 and 2650K.

manufactured graphite - a bonded granular carbon body whose matrix has been subjected to a temperature
typically in excess of 2650K and whose matrix is thermally stable below that temperature.

mesophase - see spherical carbonaceous mesophase.

metallurgical coke - produced by carbonization of coals or coal blends at temperatures up to 1400K to

produce a macroporous carbon material of high strength and relatively large lump size.

molded - formed in a closed die by the application of external pressure.

natural graphite - a naturally occurring mineral of varying crystalline perfection always intimately mixed with
inorganic mineral matter.

needle coke - the commonly used term for a special type of coke with extremely high graphitizability resulting
from a strong preferred parallel orientation of its turbostratic layer structure and a particular physical shape
of the grains.

non-graphitic carbons - all variations of solids consisting mainly of the element carbon with two-dimensional
long-range order of the carbon atoms in planar hexagonal networks, but without any measurable crystal-
lographic order in the third direction (c-direction) apart from more or less parallel stacking.

non-graphitizable carbon - a non-graphitic carbon which cannot be transformed into graphitic carbon solely
by high-temperature treatment up to 3300 K under atmospheric pressure or lower pressure.

G-4
nuclear graphite - a polygranular graphite material for use in nuclear reactor cores consisting of graphitic
carbon of very high chemical purity. High purity is needed to avoid absorption of low-energy neutrons and the
production of undesirable radioactive species.

orientation of a crystal - the angular position of a crystal described by the angles which certain crystal-
lographic axes make with the frame of reference.

orientation of a grain - the angular position of a grain described by the angles which a defined set of axes
of the grain make with the frame of reference.

orientation of an object - the angular position of an object described by the angles which a defined set of
axes or surfaces of the object make with the frame of reference.

oxidation of carbon - the chemical combination of carbon with oxygen or oxygen-containing compounds.

particle sizing - segregation of granular material into specified particle size ranges.

penetration - the depths to which one material extends into or penetrates another.

permeability - a property measured by the rate of passage of a fluid under a pressure gradient through
a material.

petroleum coke - a carbonization product of high-boiling hydrocarbon fractions obtained in petroleum

processing (heavy residues). It is the general term for all special petroleum coke products such as green,
calcined and needle petroleum coke.

petroleum pitch - a residue from heat treatment and distillation of petroleum fractions. It is a solid at
room temperature, consists of a complex mixture of numerous predominantly aromatic and alkyl-substituted
aromatic hydrocarbons, and exhibits a broad softening range instead of a defined melting point.

pitch - a residue from pyrolysis of organic material or tar distillation which is solid at room temperature,
consisting of a complex mixture of numerous, essentially aromatic hydrocarbons and heterocyclic compounds.
It exhibits a broad softening range instead of a defined melting point. When cooled from the melt, pitches
solidify without crystallization.

polycrystalline graphite - a graphite material with coherent crystallographic domains of limited size regard-
less of the perfection and preferred orientation (texture) of their crystalline structure.

polygranular carbon - a carbon material composed of grains, which can be clearly distinguished by means
of optical microscopy.

polygranular graphite - a graphite material composed of grains, which can be clearly distinguished by means
of optical microscopy.

porosity - the percentage of the total volume of a material occupied by both open and closed pores.

preferred orientation - an alignment in the crystal or defect structure of a body leading to variations in
physical properties as a function of direction; normally referenced to an orthogonal system where one of the
axes is the working direction.

G-5
premium coke - an extremely well graphitizing carbon with a high degree of optical anisotropy (isochromatic
areas of optical texture above about
100 mm) and is characterized by a combination of the following properties which differ significantly from those
of regular coke : high real density, low reversible thermal expansion, and low ash content combined, in most
cases, with low sulfur content.

puffing - an irreversible expansion of some carbon artifacts during graphitization heat treatment between
1650 K and 2700 K.

puffing inhibitors - metals or metal compounds with a high chemical affinity for the heteroatoms in the
carbons. They are distributed as fine particles within the carbon materials to be graphitized.

pyrolytic carbon - carbon material deposited from gaseous hydrocarbon compounds on suitable underlying
substrates (carbon materials, metals, ceramics) at temperatures ranging from 1000 K to 25000 K (chemical
vapor deposition).

pyrolytic graphite - graphite material with a high degree of preferred crystallographic orientation of the c-axes
perpendicular to the surface of the substrate, obtained by graphitization heat treatment of pyrolytic carbon by
chemical vapor deposition at temperatures above 2500 K.

reactivity - rate at which another material will form compounds with carbon or graphite.

regular coke - a petroleum coke with good graphitizability and is characterized by a combination of properties
which differ significantly from those of metallurgical coke.

rhombohedral graphite - a thermodynamically unstable allotropic form of graphite with an ABCABC stacking
sequence of the layers. The exact crystallographic description of this allotropic form is given by the space
group D3d5 - R-3m, (unit cell constants: a = 256.6 pm, c = 1006.2 pm).

semicoke - a carbonaceous material intermediate between fusible mesophase pitch and a non-deformable
green coke produced by the incomplete carbonization at temperatures between the onset of fusion and
complete devolatilization. Semicoke still contains volatile matter.

single crystal graphite - a graphite exhibiting the perfect crystallographic hexagonal structure, an infre-
quently occurring and rather special natural graphite.

soot - a randomly formed particulate carbon material and may be coarse, fine, or colloidal in proportions
dependent on its origin. Soot consists of variable quantities of carbonaceous and inorganic solids together
with absorbed and occluded tars and resins.

specialty carbon - a carbon manufactured with specific characteristics of low impurity, high density, uniform
strength, and certain grain size to meet specific industry needs.

specialty graphite - graphite manufactured with specific characteristics of low impurity, high density, uniform
strength, and certain grain size to meet specific industry needs (such as semiconductor fabrication process
parts, metal coating dies, crucible and furnace parts, aerospace and nuclear).

spherical carbonaceous mesophase - describes the morphology of carbonaceous mesophase which

is formed in the isotropic pitch matrix. The spherical carbonaceous mesophase usually has a lamellar
structure consisting of flat aromatic molecules arranged in parallel layers which are perpendicular to the
sphere/isotropic phase interface as described by Brooks and Taylor. On coalescence, this spherical meso-
phase loses its characteristic morphology and is converted to the bulk mesophases.

G-6
surface finish - the geometric irregularities in the surface of a solid material.

tensile strength - a property of solid material that indicates its ability to withstand a uniaxial tensile load.

transverse, T - direction perpendicular to longitudinal, L. (Same as against grain or an extruded body, maybe
either against grain, or with grain, for both mechanically molded and isostatically molded bodies depending
upon the orientation of the body in the press.)

turbostratic - a structure in which the stacked layered planes show no crystallographic stacking registry.

G-7
Bibliography

Industrial Graphite Engineering Handbook

Bibliography

ASTM (American Society for Testsing & Materials), especially 05.05 “Test Methods for Rating Motor, Diesel,
and Aviation Fuels; Catalysts; Manufactured Carbon adn Graphite Products” (was 15.01 in 1999 and prior
years).

Blackman, L.C.F., Modern Aspects of Graphite Technology, New York, Academic, 1970.

Burchell, T.D., Carbon Materials for Advanced Technologies, Kidlington, Oxford, Elsevier, 1999.

“Carbon” by various UCAR Carbon authors, Kirk-Othmer Encyclopedia of Chemical Technology, 4th, 4, New
York, John Wiley & Sons, 1992.

Ishikawa, T.; Nagaoki, T. and Lewis, I.C., Recent Carbon Technology; Including Carbon and SiC Fibers,
Cleveland, OH, JEC Press, 1983.

Mantell, C.L., Industrial Carbon, New York, D. Van Nostrand, 1928.

Mantell, C.L., Industrial Carbon; Its Elemental, Adsorptive, and Manufactured Forms, New York, D. Van
Nostrand, 1946.

Mantell, C.L., Carbon and Graphite Handbook, New York, John Wiley, 1968.

Marsh, H., Introduction to Carbon Science, London, Butterworths, 1989.

Marsh, H., Rodriguez-Reinoso, F., Heintz, E.A., Introduction to Carbon Technologies, Spain, University of
Alicante, 1997.

Nightingale, R.E., Nuclear Graphite, New York, Academic, 1962.

Pierson, H.O., Handbook of Carbon, Graphite, Diamond & Fullerenes: Properties, Processing and Applica-
tions, Park Ridge, NJ, Noyes Data, 1993.

Reynolds, W.N., Physical Properties of Graphite, New York, Elsevier, 1968.

Thrower, P.A. (Ed.), Chemistry and Physics of Carbon, 18-, New York, Marcel Dekker Inc., 1982 - onwards.

Ubbelohde, A.R. and Lewis, F.A., Graphite and Its Crystal Compounds, London, Oxford Univ., 1960.

Walker, P.L. (Ed.), Chemistry and Physics of Carbon, 1-17, New York, Marcel Dekker Inc.., 1965-1981.

Conference Proceedings

American Carbon Society, Conference on Carbon Conference (Extended Abstracts), held in odd-numbered
years, 1953 - .

British Carbon Group, Proceedings of the International Conference (Extended Abstracts), held in 1970, 1974,
1978, 1982, 1988, 1996.

Carbon Society of Japan, International Symposium on Carbon (Extended Abstracts), held in 1964, 1982,
1990, 1998.

B-1
Deutsche Keramische Gesellschaft, Proceedings of the International Conference (Extended Abstracts), held
in 1972, 1976, 1980, 1986, 1992.

Eurocarbon 2000 (1st World Conference on Carbon)

Groupe Francais d’Etude des Carbones, Carbone, Eurocarbon/International Carbon Conference (Extended
Abstracts), held in 1984, 1990, 1998.

Grupo Espanol del Carbon, Conference on Carbon (Extended Abstracts), held in 1994.

B-2