Designation: E 1225 – 04

Standard Test Method for

Thermal Conductivity of Solids by Means of the Guarded-
Comparative-Longitudinal Heat Flow Technique1
This standard is issued under the fixed designation E 1225; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (e) indicates an editorial change since the last revision or reapproval.

1. Scope C 408 Test Method for Thermal Conductivity of Whiteware

1.1 This test method describes a steady state technique for Ceramics
the determination of the thermal conductivity, l, of C 1045 Practice for Calculating Thermal Transmission
homogeneous-opaque solids (see Notes 1 and 2). This test Properties Under Steady-State Conditions
method is for materials with effective thermal conductivities in D 4351 Test Method for Measuring the Thermal Conduc-
the approximate range 0.2 < l < 200 W/(m·K) over the tivity of Plastics by the Evaporation-Calorimetric Method
approximate temperature range between 90 and 1300 K. It can E 220 Test Method for Calibration of Thermocouples by
be used outside these ranges with decreased accuracy. Comparison Techniques
E 230 Temperature-Electromotive Force (EMF) Tables for
NOTE 1—For purposes of this technique, a system is homogeneous if Standardized Thermocouples
the apparent thermal conductivity of the specimen, lA, does not vary with
F 433 Practice for Evaluating Thermal Conductivity of
changes of thickness or cross-sectional area by more than 65 %. For
composites or heterogeneous systems consisting of slabs or plates bonded Gasket Materials
together, the specimen should be more than 20 units wide and 20 units
thick, respectively, where a unit is the thickness of the thickest slab or 3. Terminology
plate, so that diameter or length changes of one-half unit will affect the 3.1 Descriptions of Terms and Symbols Specific to This
apparent lA by less than 65 %. For systems that are non-opaque or Standard:
partially transparent in the infrared, the combined error due to inhomo- 3.1.1 Terms:
geneity and photon transmission should be less than 65 %. Measurements
3.1.1.1 thermal conductivity, l—the time rate of heat flow,
on highly transparent solids must be accompanied with infrared absorption
coefficient information, or the results must be reported as apparent thermal under steady conditions, through unit area, per unit temperature
conductivity, lA. gradient in the direction perpendicular to the area;
3.1.1.2 apparent thermal conductivity—when other modes
NOTE 2—This test method may also be used to evaluate the contact of heat transfer through a material are present in addition to
thermal conductance/resistance of materials. conduction, the results of the measurements performed accord-
1.2 This standard does not purport to address all of the ing to this test method will represent the apparent or effective
safety concerns, if any, associated with its use. It is the thermal conductivity for the material tested.
responsibility of the user of this standard to establish appro- 3.1.2 Symbols:
priate safety and health practices and determine the applica-
bility of regulatory limitations prior to use.
lM (T) = thermal conductivity of meter bars (reference
2. Referenced Documents materials) as a function of temperature, (W/
2.1 ASTM Standards: 2 (m·K)),
lM1 = thermal conductivity of top meter bar (W/
C 177 Test Method for Steady-State Heat Flux Measure-
(m·K)),
ments and Thermal Transmission Properties by Means of
lM2 = thermal conductivity of bottom meter bar
the Guarded-Hot-Plate Apparatus
(W/(m·K)),
lS (T) = thermal conductivity of specimen corrected
for heat exchange where necessary, (W/(
1
This test method is under the jurisdiction of ASTM Committee E37 on Thermal m·K)),
Measurements and is the direct responsibility of Subcommittee E37.05 on Thermo- l8S (T) = thermal conductivity of specimen calculated
physical Properties.
Current edition approved May 1, 2004. Published June 2004. Originally
by ignoring heat exchange correction, (W/
approved in 1987. Last previous edition approved in 1999 as E 1225–99. (m·K)),
2
For referenced ASTM standards, visit the ASTM website, www.astm.org, or lI (T) = thermal conductivity of insulation as a func-
contact ASTM Customer service at service@astm.org. For Annual Book of ASTM tion of temperature, (W/(m·K)),
Standards volume information, refer to the standard’s Document Summary page on
the ASTM website.
T = absolute temperature (K),

Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.

1
 E 1225 – 04

Z = position as measured from the upper end of

the column, (m),
l = specimen length, (m),
Ti = the temperature at Zi, (K),
q8 = heat flow per unit area, (W/m 2),
dl, dT, etc. = uncertainty in l, T, etc.,
rA = specimen radius, (m),
rB = guard cylinder inner radius, (m),
Tg (z) = guard temperature as a function of position,
z, (K), and
4. Summary of Test Method
4.1 A test specimen is inserted under load between two
similar specimens of a material of known thermal properties. A
temperature gradient is established in the test stack and heat
losses are minimized by use of a longitudinal guard having
approximately the same temperature gradient. At equilibrium
conditions, the thermal conductivity is derived from the mea-
sured temperature gradients in the respective specimens and
the thermal conductivity of the reference materials.
4.2 General Features of Test Method:
4.2.1 The general features of the guarded longitudinal heat
flow technique are shown in Fig. 1. A specimen of unknown
thermal conductivity, lS, but having an estimated thermal
conductance of lS/ lS, is mounted between two meter bars of
known thermal conductivity, lM, of the same cross-section and
similar thermal conductance, lM/lM. A more complex but
suitable arrangement is a column consisting of a disk heater
with a specimen and a meter bar on each side between heater
and heat sink. Approximately one-half of the power would then
FIG. 1(a) Schematic of a Comparative-Guarded-Longitudinal Heat Flow
flow through each specimen. When the meter bars and speci- System Showing Possible Locations of Temperature Sensors
men are right-circular cylinders of equal diameter the tech-
nique is described as the cut-bar method. When the cross-
sectional dimensions are larger than the thicknesss it is
described as the flat slab comparative method. Essentially, any
shape can be used, as long as the meter bars and specimen have
the same conduction areas.
4.2.2 A force is applied to the column to ensure good
contact between specimens. The stack is surrounded by an
insulation material of thermal conductivity, lI. The insulation
is enclosed in a guard shell with a radius, rB, held at the
temperature, Tg (z). A temperature gradient is imposed on the
column by maintaining the top at a temperature, TT, and the
bottom at temperature TB. Tg(z) is usually a linear temperature
gradient matching approximately the gradient established in
the test stack. However, an isothermal guard with Tg (z) equal
to the average temperature of the specimen may also be used.
An unguarded system is not recommended due to the potential
very large heat losses, particularly at elevated temperatures
(1).3 At steady state, the temperature gradients along the
sections are calculated from measured temperatures along the
two meter bars and the specimen. The value of lS , as
uncorrected for heat shunting) can then be determined using
the following equation where the notation is shown in Fig. 1:
Z4 2 Z3 lM
ls 5 T 2 T · 2 ·
4 3
S T2 2 T1 T6 2 T5
Z2 2 Z1 1 Z6 2 Z5 D (1)
FIG. 1(b) Schematic of Typical Test Stack and Guard System Illustrating
Matching of Temperature Gradients
3
The boldface numbers in parentheses refer to a list of references at the end of
this test method.

2
 E 1225 – 04
This is a highly idealized situation, however, since it must be thoroughly tested. Requirements for qualifying an
assumes no heat exchange between the column and insulation apparatus are outlined in Section 10.
at any position and uniform heat transfer at each meter
bar-specimen interface. The errors caused by these two as- 6. Requirements
sumptions vary widely and are discussed in Section 10.
6.1 Meter Bar Reference Materials:
Because of these two effects, restrictions must be placed on this
test method, if the desired accuracy is to be achieved. 6.1.1 Reference materials or transfer standards with known
thermal conductivities must be used for the meter bars. Since
the minimum measurement error of the method is the uncer-
tainty in lM, it is preferable to use standards available from a
national standards laboratory. Other reference materials are
available because numerous measurements of l have been
made and general acceptance of the values has been obtained.
Table 1 lists the currently available recognized reference
materials including those available from National Institute of
Standards and Technology. Fig. 2 shows the approximate
variation of lM with temperature.
6.1.2 Table 1 is not exhaustive and other materials may be
used as references. The reference material and the source of lM
values shall be stated in the report.
6.1.3 The requirements for any reference material include
stability over the temperature range of operation, compatibility
with other system components, reasonable cost, ease of ther-
mocouple attachment, and an accurately known thermal con-
ductivity. Since heat shunting errors for a specific lI increase as
lM/ls varies from unity, (1) the reference which has a lM
nearest to lS should be used for the meter bars.
6.1.4 If a sample’s thermal conductivity ls is between the
thermal conductivity values of two types of reference materi-
als, the reference material with the higher lM should be used to
reduce the total temperature drop along the column.
6.2 Insulation Materials:
6.2.1 A large variety of powder, particulate, and fiber
materials exists for reducing both radial heat flow in the
column-guard annulus and surrounds, and for heat shunting
along the column. Several factors must be considered during
selection of the most appropriate insulation. The insulation
NOTE 1—The material selected for the meter bars should have a thermal must be stable over the anticipated temperature range, have a
conductivity as near as possible to the thermal conductivity of the low lI, and be easy to handle. In addition, the insulation should
unknown.
not contaminate system components such as the temperature
FIG. 2 Approximate Values for the Thermal Conductivity of
Several Possible Reference Materials for Meter Bars sensors, it must have low toxicity, and it should not conduct
electricity. In general, powders and particulates are used since
they pack readily. However, low density fiber blankets can also
5. Significance and Use be used.
6.2.2 Some candidate insulations are listed in Table 2.
5.1 The comparative method of measurement of thermal
conductivity is especially useful for engineering materials 6.3 Temperature Sensors:
including ceramics, polymers, metals and alloys, refractories, 6.3.1 There shall be a minimum of two temperature sensors
carbons, and graphites including combinations and other com- on each meter bar and two on the specimen. Whenever
posite forms of each. possible, the meter bars and specimen should each contain
5.2 Proper design of a guarded-longitudinal system is diffi- three sensors. The extra sensors are useful in confirming
cult and it is not practical in a method of this type to try to linearity of temperature versus distance along the column, or
establish details of construction and procedures to cover all indicating an error due to a temperature sensor decalibration.
contingencies that might offer difficulties to a person without 6.3.2 The type of temperature sensor depends on the system
technical knowledge concerning theory of heat flow, tempera- size, temperature range, and the system environment as con-
ture measurements, and general testing practices. Standardiza- trolled by the insulation, meter bars, specimen, and gas within
tion of this test method is not intended to restrict in any way the the system. Any sensor possessing adequate accuracy may be
future development by research workers of new or methods or used for temperature measurement (2) and be used in large
improved procedures. However, new or improved techniques systems where heat flow perturbation by the temperature

3
 E 1225 – 04
TABLE 1 Reference Materials For Use as Meter Bars
Percentage
Material Temperature Range (K) Uncertainty lM(W/m·K) Material Source
in l (6 %)
A
Electrolytic Iron SRM 734 To 1000 2 NISTA
Tungsten SRM 730 4 to 300 2 lM Dependent on TA NISTA
300 to 2000 2 to 5
>2000 5 to 8
Austenitic Stainless SRM 4 to 1200 <5 % lM = 1.22T0.432 T > 200KA NISTA
735
Iron 80 to 1200 2 lM should be calculated from ...
measured valuesBC
Copper 90 to 1250 <2 lM = 416.3 − 0.05904T + 7.087 manufacturer
3107/T3D
EF
Pyroceram Code 9606 90 to 1200 ... manufacturer
Fused SilicaG 1300 <8 lM = (84.7/T) + 1.484 + 4.94 3 10−4 manufacturer
Up to 900 K T + 9.6 3 10−13T4HI
EF
Pyrex 7740 90 to 600 6 manufacturer
A
National Institute of Standards and Technology, Washington, D.C. 20234. See Special Publications 260-52 and 260-46.
B
Fulkerson W., et al., Physics Review 167, p. 765, (1968).
C
Lucks C. F., Journal of Testing and Evaluation, ASTM 1 (5), 422 (1973).
D
Moore, J. P., Graves, R. S. and McElroy, D. L., Canadian Journal of Physics, 45, 3849 (1967).
E
“Thermal Conductivity of Selected Materials,” Report NSRDS-NBS 8, National Bureau of Standards, 1966.
F
L. C. Hulstrom, R. P. Tye, and S. E. Smith, Thermal Conductivity 19, Ed. D. W. Yarbrough, Plenum Press, New York, In Course of Publication (see also High
Temperature-High Pressures, 17, 707, 1985.
G
Hust J. G., Cryogenics Division; NBS, Boulder, Colorado 80302.
H
Above 700 K a large fraction of heat conduction in fused silica will be by radiation and the actual effective values may depend on the emittances of bounding surfaces
and meter bar size.
I
Recommended values from Table 3017 A-R-2 of the Thermophysical Properties Research Center Data Book, Vol. 3, “Nonmetallic Elements, Compounds, and
Mixtures,” Purdue University, Lafayette, Indiana.

TABLE 2 Suitable Thermal Insulation Materials 6.3.5 In Fig. 3b, the thermocouple resides in a radial slot,
Typical Thermal Conductivity (W/(m·K)) and in Fig. 3c the thermocouple is pulled through a radial hole
MaterialA
300K 800K 1300K in the material. When a sheathed thermocouple or a thermo-
Poured Powders couple with both thermoelements in a two-hole electrical
Diatomaceous Earth 0.053 0.10 0.154 insulator is used, the thermocouple attachment shown in Fig.
Bubbled Alumina 0.21 0.37 0.41 3d can be used. In the latter three cases, the thermocouple
Bubbled Zirconia 0.19 0.33 0.37
Vermiculite 0.07 0.16 ... should be thermally connected to the solid surface using a
Perlite 0.050 0.17 ... suitable glue or high temperature cement. All four of the
Blankets and Felts procedures shown in Fig. 3 should include wire tempering on
Aluminosilicate 60–120 kg/m3 0.044 0.13 0.33
Zirconia 60–90 kg/m3 0.039 0.09 0.25 the surfaces, wire loops in isothermal zones, thermal wire
A
All materials listed can be used up to the 1300 K limit of the comparative
grounds on the guard, or a combination of all three (5).
longitudinal except where noted. 6.3.6 Since uncertainty in temperature sensor location leads
to large errors, special care must be taken to determine the
correct distance between sensors and to calculate the possible
sensors would be negligible. Thermocouples are normally error resulting from any uncertainty.
employed. Their small size and the ease of attachment are 6.4 Reduction of Contact Resistance:
distinct advantages. 6.4.1 This test method requires uniform heat transfer at the
6.3.3 When thermocouples are employed, they should be meter bar to specimen interfaces whenever the temperature
fabricated from wires which are 0.1 mm diameter or less. A sensors are within a distance equal to rA from an interface (6).
constant temperature reference shall always be provided for all This requirement necessitates a uniform contact resistance
cold junctions. This reference can be an ice-cold slurry (3), a across the adjoining areas of meter bars and specimens. This is
constant temperature zone box, or an electronic ice point normally attained by use of an applied axial load in conjunction
reference. All thermocouples shall be fabricated from either with a conducting medium at the interfaces. Measurements in
calibrated thermocouple wire (4) or from wire that has been a vacuum environment are not recommended, unless the
certified by the supplier to be within the limits of error vacuum is required for protection purposes.
specified in Table 1 of Standard E 230. 6.4.2 For the relatively thin specimens normally used for
6.3.4 Thermocouple attachment is important to this tech- materials having a low thermal conductivity, the temperature
nique in order to ensure that reliable temperature measure- sensors must be mounted close to the surface and in conse-
ments are made at specific points. The various techniques are quence the uniformity of contact resistance is critical. In such
illustrated in Fig. 3. Intrinsic junctions can be obtained with cases, a very thin layer of a compatible highly conductive fluid,
metals and alloys by welding individual thermo-elements to the paste, soft metal foil, or screen shall be introduced at the
surfaces (Fig. 3a). Butt or bead welded thermocouples junc- interfaces.
tions can be rigidly attached by peening, cementing, or welding 6.4.3 Means shall be provided for imposing a reproducible
in fine grooves or small holes (Fig. 3b, 3c, and 3d). and constant load along the column with the primary purpose

4
 E 1225 – 04

3a 3b
Intrinsic weld with separate temperature elements welded to specimen or meter Radial slots on the flat surfaces to hold a bare wire or ceramic insulated
bars so that signal is through the material. thermocouple sensor the may be bonded into slot.

3c 3d
Small radial hole drilled through the specimen or meter bar and non-insulated Small Radial hole drilled part way through the specimen or meter bar and a
(permitted if the material is an electrical insulator) or insulated thermocouple pulled thermocouple pushed into the hole.
through the hole.

NOTE 1—In all cases the thermoelements should be thermally tempered and/or thermally grounded on the guard to minimize temperature measurement
errors due to heat flow into or out of the hot junction.
FIG. 3 Thermocouple Attachments

of minimizing interfacial resistances at meter bar-specimen 6.6 System Instrumentation:

interfaces. Since the force applied to the column usually affects 6.6.1 The combination of temperature sensor and the instru-
the contact resistance, it is desirable that this force be variable ment used for measuring the sensor output shall be adequate to
to ensure that lS does not change with force variation. This ensure a temperature measurement precision of 60.04 K and
force can be applied either pneumatically, hydraulically, by an absolute error less than 60.5 %.
spring action, or by putting a dead weight on the column. The 6.6.2 Instrumentation for this technique shall be adequate to
above load mechanisms have the advantage of remaining maintain the required temperature control and measure all
constant with change in column temperature. In some cases, pertinent output voltages with accuracy commensurate with the
the compressive strength of the specimen might be so low that system capability. Although control can be manual, a technique
the applied force must be limited to the dead weight of the of this general description can be automated so that a computer
upper meter bar. In this case, special care must be taken to limit carries out all the control functions, acquires all pertinent
errors caused by poor contact, by judicious positioning of voltages, and calculates the thermal conductivity (7).
temperature sensors away from any heat flow perturbation at
the interfaces. 7. Sampling and Conditioning Test Specimens
6.5 Guard Cylinder: 7.1 Test Specimens—This test method is not restricted to a
6.5.1 The specimen-meter bar column shall be enclosed particular geometry. General practice is to use cylindrical or
within a guard tube or pipe normally of right circular symme- square cross-sections. The conduction area of the specimen and
try. This guard cylinder can be either a metal or a ceramic but reference samples must be the same to within 1 % (see Note 3)
its inside radius should be such that the ratio rB/rAwill be and any difference in area shall be taken into account in the
between 2.0 and 3.5 (1). This guard cylinder shall contain at calculations of the result. For the cylindrical configuration, the
least one heater for controlling the temperature profile along radii of the specimen and meter bars must agree to within
the guard. 61 % and the specimen radius, rA, must be such that rB/rAis
6.5.2 The guard shall be constructed and operated so that the between 2.0 and 3.5. Each flat surface of the specimen and
temperature of the guard surface is either isothermal and equal reference must be flat with a surface finish equal to or better
to the approximate mean temperature of the specimen or than 32− and the normal to each end shall be parallel with the
preferably has an approximately linear profile with the top and specimen axis to within 610 min.
bottom ends of the guard matched to corresponding positions NOTE 3—In some cases this requirement is not necessary. For example,
along the column. In each case, at least three temperature some apparatus might consist of meter bars and specimen with high values
sensors shall be attached to the guard at known positions to of l M and lS so that thermal shunting errors would be small for long
measure the temperature profile. sections. These sections might be long enough to permit temperature

5
 E 1225 – 04
sensor attachment to be far enough away from the interfaces to ensure that the system should be pumped and purged, and the operating
heat flow was uniform. The specimen length should be selected based on gas and pressure established. The predetermined force required
considerations of radius and thermal conductivity. When lM is higher than for reducing the effects of non-uniform interfacial resistance
the thermal conductivity of SRM 735 (stainless steel), long specimens
with length / rA>>1 can be used. These long specimens permit the use of
should be applied to the load column.
large distances between temperature sensors and this reduces the percent- 9.2 Heaters at either end of the column should be energized
age error derived from the uncertainty in sensor position. When l M is (see Note 4) and adjusted until the temperature differences
lower than the thermal conductivity of SRM 735, the sample’s length must between positions Z1 and Z2, Z3 and Z4, and Z5 and Z6 are
be reduced because uncertainty due to the heat shunting becomes too between 200 times the imprecision of the DT measurements
large. and 30 K, and the specimen is at the average temperature
7.2 Sampling and Conditioning—Unless specifically re- desired for the measurement. Although the exact temperature
quired or prescribed, one representative specimen shall be profile along the guard is not important for rB/rA $ 3, the power
prepared from the sample and no preconditioning has to be to the guard heaters should be adjusted until the temperature
undertaken. profile along the guard, Tg (z), is constant with respect to time
to within 60.1 K and either:
8. Calibration and Verification 9.2.1 Approximately linear so that Tg (z) coincides with the
8.1 There are many situations that call for equipment temperature along the sample column at a minimum of three
verifications before operations on unknown materials can be places including the temperature at the top sensor on the top
successfully accomplished. These include the following: meter bar, the bottom sensor on the bottom bar, and the
8.1.1 After initial equipment construction, specimen midplane; or
8.1.2 When the ratio of lM to lS is less than 0.3 or greater 9.2.2 Constant with respect to z to within 65 K and
than 3 and it is not possible to match thermal conductance matched to the average temperature of the test specimen.
values,
NOTE 4—These heaters can either be attached to the ends of the meter
8.1.3 When the specimen shape is complex or the specimen bars or to a structure adjacent to the meter bar. The heaters can be powered
is unusually small, with A.C. or D.C. Several different heater configurations are acceptable.
8.1.4 When changes have been made in the system geom- The power to these heaters shall be steady enough to maintain short term
etry, temperature fluctuations less than 60.03 K on the meter bar thermocouple
8.1.5 When meter bar or insulation material other than those nearest the heater. These two heaters, in conjunction with the guard shell
listed in 5.1 and 5.2 are considered for use, and heater and the system coolant shall maintain long term temperature drift
less than 60.05 K/h.
8.1.6 When the apparatus has been previously operated to a
high enough temperature to change the properties of a compo- 9.3 After the system has reached steady state (T drift <0.05
nent such as thermocouples’ sensitivity. K/h), measure the output of all temperature sensors.
8.2 These verification tests shall be run by comparing at
least two reference materials in the following manner: 10. Calculation
8.2.1 A reference material which has the closest thermal 10.1 Approximate Specimen Thermal Conductivity:
conductivity to the estimated thermal conductivity of the 10.1.1 The outputs from the temperature sensors shall be
unknown sample should be machined according to 6.1, and converted to temperature, and the apparent heat flow per unit
8.2.2 The thermal conductivity l of the specimen fabricated area, q8, in the meter bars shall be calculated using the
from a reference material shall then be measured as described following:
in Section 9, using meter bars fabricated from another refer- T2 2 T1
ence material which has the closest l to that of the specimen. q8T 5 lM · Z 2 Z
2 1
For example, verification tests might be performed on a top bar (2)
Pyroceramy 9606 specimen using meter bars fabricated from T6 2 T 5
SRM 735 stainless steel. If the measured thermal conductivity q8B 5 lM · Z 2 Z
6 5
of the specimen disagrees with the value from Table 1 after bottom bar (3)
applying the corrections for heat exchange, additional effort is In each of these equations, the lM value (see Note 5) to be
required to find the error source(s). inserted shall be obtained from the information of 6.1 for the
9. Procedure average meter bar temperature. Although these two values, q8
T and q8B, should agree with each other to within about 610 %
9.1 Where possible and practical, select the reference speci-
when heat exchange with the insulation is small, good agree-
mens (meter bars) such that the thermal conductance is of the
ment is not a sufficient condition (nor always a necessary
same order of magnitude as that expected for the test specimen.
condition) for low heat shunting error.
After instrumenting and installing the proper meter bars, the
10.1.2 A value for the specimen thermal conductivity at
specimen should be instrumented similarly. It should then be
temperature (T3 + T4)/2, as uncorrected for heat exchange with
inserted into the test stack such that it is at aligned between the
the insulation, can then be calculated using the following:
meter bars with at least 99 % of each specimen surface in
contact with the adjacent meter bar. Soft foil or other contact- ~q8T 1 q8B! ~Z4 2 Z 3!
l8S 5 (4)
ing medium may be used to reduce interfacial resistance. If the 2 ~ T4 2 T 3 !
system must be protected from oxidation during the test or if NOTE 5—This type of calculation procedure actually requires only two
operation requires a particular gas or gas pressure to control lI, temperature sensors on each column section. In this case, the third sensor

6
 E 1225 – 04
on each section serves as a test for consistency of the other two. Some since all such specimens should have the same size as the
calculation procedures require more than the two sensors to obtain more specimen with an unknown thermal conductivity and have the
knowledge about dT/dZ. same surface finish.
10.2 Corrections for Extraneous Heat Flow:
10.2.1 Calculation of the specimen thermal conductivity by 11. Report
a simple comparison of temperature gradients in the meter bars 11.1 The report of the test results shall include the follow-
to that in the specimen is less valid when the specimen or meter ing:
bars, or both, have low thermal conductivities relative to that of 11.1.1 Complete specimen identification including shape
the insulation. The apparatus should be designed to minimize and size,
these errors. The deviation from uniform heat flow has been 11.1.2 Complete identification of insulation and source of lI
expressed as follows (1) : values, gas, and gas pressure,
g 5 F g Fl (5) 11.1.3 Statements of thermocouple type, size, and attach-
where Fg is a function of system dimensions, and Fl is a ment procedure,
function of lM, lI, and lS (1). The Fg term has a value between 11.1.4 Complete listing of the geometrical dimensions of
2 and 3 for the ratio of guard radius to column radius specified the system including rA, rB, specimen height, meter bar height,
for this system. The Fl term is shown in Fig. 4 as a function of and distances between temperature sensors,
l/lI for various values of lM/lI for a linear guard. At high 11.1.5 Column force,
ratios of lM/lI and lS/lI, corrections would not be necessary 11.1.6 Meter bar material and source of lM values if other
since the departure from ideal heat flow would be small. For than those listed in Table 1,
example, the product of Fl and Fg would be less than 0.10 11.1.7 Reference to the use of this test method shall include
(10 %) for all measurements where l M/lI and lS/lI are greater a statement of the percentage variation of the qualification
than 30. If the value of FgFl is to be kept below 10 %, the results about the true value. For example, “thermal conductiv-
ratios lM/lI and ls/lI must be within the boundaries on Fig. 4. ity results on Pyroceramy 9606 using SRM 735 meter bars
10.2.2 Measurements on materials where the ratios of lM/lI were within 64 % of the accepted values for Pyroceramy over
and lS/lI do not fall within these boundaries shall be accom- the temperature range from 250 to 900 K,”
panied with corrections for extraneous heat flow. These cor- 11.1.8 Variations, if any, from this test method. If results are
rections can be determined in the following three different to be reported as having been obtained by this method, then all
ways: requirements prescribed by this method shall be met. Where
10.2.2.1 Use of analytical techniques as described by such conditions are not met, the phrase, “All requirements of
Didion (1) and Flynn (8), this method have been met with the exception of ...” shall be
10.2.2.2 Using calculations from finite-difference or finite- added and a complete list of the exceptions included;
element heat conduction codes, and 11.1.9 Measured values of temperature and specimen ther-
10.2.2.3 Determined experimentally by using several refer- mal conductivity.
ence materials or transfer standards of different thermal con-
ductance as specimens. The procedure must be used cautiously 12. Precision and Bias
12.1 Example of Error Estimation:
12.1.1 Assumptions for a system where both meter bars and
the specimen are of equal length is that the sensor spacings are
all 13 mm and lM' l S:

U dll U 5 ? 0.003 ?
M
M
(6)

Z2 2 Z1 ; Z4 2 Z 3 ; Z6 2 Z5 5 13 mm;
T2 2 T 1 ; T4 2 T3 ; T 6 2 T5 5 10 K;
d~Z2 2 Z 1! ; d~Z4 2 Z3! ; d~Z 6 2 Z5! 5 0.2 mm; and
d~T2 2 T 1! ; d~T4 2 T3! ; d~T 6 2 T5! 5 0.04 K.
12.1.2 The maximum value of d(Z2 − Z 1) etc. was approxi-
mated by assuming an uncertainty of 60.5 (sensor diameter) at
each temperature measurement position. Therefore, if the
diameter of each sensor is 0.2 mm, the uncertainty in the
difference would be 60.2 mm. The number for d(T2 − T1) etc.
was calculated based on the sensor absolute accuracy.
12.1.3 With these values the fractional uncertainty in l8S
will be |0.069| or 66.9 %.
ls/li 12.2 Indeterminate Errors:
FIG. 4 Fractional Heat Exchange Between the Meter Bar-
12.2.1 There are at least three other errors that can contrib-
Specimen Column and Surrounding Insulation as a Function of ute to total system error and these are (1) non-uniform
lm/l i for Several Values of ls/li interfacial resistance, (2) heat exchange between the column

7
 E 1225 – 04
and the guard, and (3) heat shunting through the insulation 12.3 Overall—An international, inter-laboratory round
around the column. These three errors must be minimized or robin study also involving absolute methods (9) has shown that
appropriate corrections applied to the data if the desired a precision of 66.8 % can be attained over the temperature
accuracy is to be obtained. range 300 to 600 K. Although no definite bias could be
12.2.2 The contributions from the last two errors can be established these are indications that the values were on the
determined approximately using results from appropriate ex- order of 2 % lower than those obtained by absolute methods.
periments carried out at different levels of guard temperature to This cited paper is on file at ASTM as a research report.
specimen stack temperature out of balance.

REFERENCES

(1) Didion, D. A., “An Analysis and Design of a Linear Guarded Cut-Bar (6) Fried, E., Thermal Conductivity, Vol 2, Tye, R. P. (ed.), Academic
Apparatus for Thermal Conductivity Measurements,” AD-665789, Press, New York (1969).
January, 1968, available from the National Technical Information (7) Morgan, M. T., and West, G. A., “Thermal Conductivity of the Rocks
Service, Springfield, VA. in the Bureau of Mines Standard Rock Site,” Thermal Conductivity,
(2) Finch, D. I., “General Principles of Thermoelastic Thermometry,” in 16, Larsen, D. C., ed., Plenum Press, NY, NY, 1983, pp 79–90.
Temperature, It’s Measurement and Control in Science and Industry, (8) Flynn, D. R., “Thermal Conductivity of Ceramics,” Mechanical and
Vol. 3, Part 2, Section 1, Reinhold Publishing Corporation. Thermal Properties of Ceramics, Special Publication 303, National
(3) Caldwell, F. R., “Temperatures of Thermocouple Reference Junctions Bureau of Standards, 1969.
in an Ice Bar,” J. Res. Nat. Bur. Std., 69C (2) (1965). (9) Hulstrom, L. C., Tye, R. P., and Smith, S. E., “Round-Robin Testing of
(4) ASME: PTC 19.3, Temperature Measurement, Part 3, p. 1232, 1974. Thermal Conductivity Reference Materials,” Thermal Conductivity,
(5) Anderson, R. L., and Kollie, T. G., “Problems in High Temperature 19, Yarbrough, D. W., ed., Plenum Press, New York. (See also
Thermometry,” July, 1976, pp. 171–221. High-Temp-High Pressures, Vol 17, p. 707, 1985).

ASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentioned
in this standard. Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk
of infringement of such rights, are entirely their own responsibility.

This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and
if not revised, either reapproved or withdrawn. Your comments are invited either for revision of this standard or for additional standards
and should be addressed to ASTM International Headquarters. Your comments will receive careful consideration at a meeting of the
responsible technical committee, which you may attend. If you feel that your comments have not received a fair hearing you should
make your views known to the ASTM Committee on Standards, at the address shown below.

This standard is copyrighted by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959,
United States. Individual reprints (single or multiple copies) of this standard may be obtained by contacting ASTM at the above
address or at 610-832-9585 (phone), 610-832-9555 (fax), or service@astm.org (e-mail); or through the ASTM website
(www.astm.org).