Designation: E 1640 – 99 An American National Standard

Standard Test Method for

Assignment of the Glass Transition Temperature By
Dynamic Mechanical Analysis1
This standard is issued under the fixed designation E 1640; 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 E 1356 Test Method for Glass Transition Temperatures by

1.1 This test method covers the assignment of a glass Differential Scanning Calorimetry or Differential Thermal
transition temperature (Tg) of materials using dynamic me- Analysis3
chanical analyzers. E 1363 Test Method for Temperature Calibration of Ther-
1.2 This test method is applicable to thermoplastic poly- momechanical Analyzers3
mers, thermoset polymers, and partially crystalline materials E 1545 Test Method for the Determination of Glass Transi-
which are thermally stable in the glass transition region. tion Temperatures by Thermomechanical Analysis3
1.3 The applicable range of temperatures for this test 2.2 Other Standard:
method is dependent upon the instrumentation used, but, in SRM 18R-94 Recommended Method for Glass Transition
order to encompass all materials, the minimum temperature Temperature (Tg) Determination by DMA of Oriented
should be about −150°C. Fiber-Resin Composites 4
1.4 This test method is intended for materials having an 3. Terminology
elastic modulus in the range of 0.5 MPa to 100 GPa.
1.5 Electronic instrumentation or automated data analysis 3.1 Definition:
and data reduction systems or treatments equivalent to this test 3.1.1 Specific technical terms used in this document are
method may also be used. defined in Terminology D 4092 and E 1142.
3.1.2 dynamic mechanical analyzer—any of various com-
NOTE 1—The user bears the responsibility for determining the preci- mercial or experimental devices used to study the viscoelastic
sion, accuracy, and validity of the techniques and measurements made
response of a specimen under a forced or free resonant
using dynamic mechanical analyzers in accordance with this standard. If
disputes arise, only the manual procedures described in this standard are oscillatory load. The force may be applied in torsion, flexure,
to be considered valid. or a combination of tension and compression.
1.6 This standard does not purport to address all of the 4. Summary of Test Method
safety concerns, if any, associated with its use. It is the
4.1 A specimen of known geometry is placed in mechanical
responsibility of the user of this standard to establish appro-
oscillation at either fixed or resonant frequency and changes in
priate safety and health practices and determine the applica-
the viscoelastic response of the material are monitored as a
bility of regulatory limitations prior to use.
function of temperature. Under ideal conditions, the glass
2. Referenced Documents transition region is marked by a rapid decrease in the storage
modulus and a rapid increase in the loss modulus. The glass
2.1 ASTM Standards:
transition of the test specimen is indicated by the extrapolated
D 4065 Practice for Determining and Reporting Dynamic
onset of the decrease in storage modulus which marks the
Mechanical Properties of Plastics2
transition from a glassy to a rubbery solid.
D 4092 Terminology Relating to Dynamic Mechanical
Measurements in Plastics2 5. Significance and Use
E 691 Practice for Conducting an Interlaboratory Study to
5.1 This test method can be used to locate the glass
Determine the Precision of a Test Method3
transition region and assign a glass transition temperature of
E 1142 Terminology Relating to Thermophysical Proper-
amorphous and semi-crystalline materials.
ties3
5.2 Dynamic mechanical analyzers monitor changes in the
viscoelastic properties of a material as a function of tempera-
1
This test method is under the jurisdiction of ASTM Committee E-37 on ture and frequency, providing a means to quantify these
Thermal Measurements and is the direct responsibility of Subcommittee E37.01 on changes. In ideal cases, the temperature of the onset of the
Test Methods and Recommended Practices.
Current edition approved Aug. 10, 1999. Published November 1999. Originally
published as E 1640–94. Last previous edition E 1640–94
2 4
Annual Book of ASTM Standards, Vol 08.02. Available from Cuppliers of Advanced Composite Materials Association,
3
Annual Book of ASTM Standards, Vol 14.02. Arlington, VA.

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

1
 E 1640
decrease in storage modulus marks the glass transition. placement (or strain), frequency, and temperature. Tempera-
5.3 A glass transition temperature ( Tg) is useful in charac- tures should be measurable with an accuracy of 60.5°C, force
terizing many important physical attributes of thermoplastic, to 61 %, and frequency to 60.1 Hz.
thermosets (see SRM 18R-94), and semi-crystalline materials 7.2.4 Temperature Controller and Oven, for controlling the
including their thermal history, processing conditions, physical specimen temperature, either by heating, cooling (in steps or
stability, progress of chemical reactions, degree of cure, and ramps), or by maintaining a constant experimental environ-
both mechanical and electrical behavior. T g may be determined ment. The temperature programmer shall be sufficiently stable
by a variety of techniques and may vary in accordance with the to permit measurement of specimen temperature to 60.5°C.
technique. The precision of the required temperature measurement is
5.4 This test method is useful for quality control, specifica- 61.0°C.
tion acceptance, and research. 7.2.5 Output Device, capable of displaying the storage
modulus (either linearly or logarithmically) on the Y axis
6. Interferences increasing in the upward direction and temperature on the X
6.1 Because the specimen size will usually be small, it is axis increasing to the right.
essential that each specimen be homogeneous and/or represen- NOTE 2—Some instruments suitable for this test may display only
tative of the material as a whole. linear or logarithm storage modulus while others may display either linear
6.2 An increase or decrease in heating rates from those and/or logarithm storage modulus. Care must be taken to use the same
specified may alter results. modulus scale when comparing unknown specimens, and in the compari-
6.3 A transition temperature is a function of the experimen- son of results from one instrument to another.
tal frequency, therefore the frequency of test must always be 7.3 Nitrogen, Helium or other gas supplied for purging
specified. (The transition temperature increases with increasing purposes.
frequency.) Extrapolation to a common frequency may be 7.4 Calipers or other length measuring device capable of
accomplished using a predetermined frequency shift factor or measuring dimensions (or length within)6 0.01 mm.
assuming the frequency shift factor of about 8°C per decade of
8. Precautions
frequency.5
8.1 Toxic and corrosive, or both, effluents may be released
7. Apparatus when heating some materials and could be harmful to person-
7.1 The function of the apparatus is to hold a specimen of nel and to apparatus.
uniform dimension so that the sample acts as the elastic and 8.2 Multiple Transitions—Under some experimental condi-
dissipative element in a mechanically oscillated system. Dy- tions it is possible to have transitions secondary to the primary
namic mechanical analyzers typically operate in one of several glass transition. Secondary transitions may be related to the
modes. See Table 1. glass transition of a second polymeric phase, melt processes,
crystallization, chemical reactions, the motion of groups pen-
dent to the main backbone or the crankshaft motion of the
TABLE 1 Modes for Dynamic Mechanical Analyzers
polymer backbone.
Mechanical Response
Mode 9. Samples
Tension Flexural Torsional Compression
Free/dec ... ... X ... 9.1 Samples may be any uniform size or shape, but are
Forced/res/CA ... X X ... ordinarily analyzed in rectangular form. If some heat treatment
Forced/fix/CA X X X X is applied to the specimen to obtain this preferred analytical
Forced/fix/CS X X ... X
form, such treatment should be noted in the report.
Free 5 free oscillation; dec 5 decaying amplitude; forced 5 forced oscillation; 9.2 Due to the numerous types of dynamic mechanical
CA 5 constant amplitude; res 5 resonant frequency; fix 5 fixed frequency;
CS 5 controlled stress. analyzers, sample size is not fixed by this method. In many
cases, specimens measuring between 1 3 5 3 20 mm and
1 3 10 3 50 mm are suitable.
7.2 The apparatus shall consist of the following:
7.2.1 Clamps, a clamping arrangement that permits gripping NOTE 3—It is important to select a specimen size appropriate for both
of the specimen. Samples may be mounted by clamping at both the material and the testing apparatus. For example, thick samples may be
required for low modulus materials while thin samples may be required
ends (most systems), one end (for example, torsional pendu-
for high modulus materials.
lum), or neither end (free bending between knife edges).
7.2.2 Oscillatory Stress (Strain), for applying an oscillatory 10. Calibration
deformation (strain) or oscillatory stress to the specimen. The 10.1 Calibrate the storage modulus and temperature signals
deformation may be applied and then released, as in freely in accordance with manufacturer’s recommended procedures
vibrating devices, or continuously applied, as in forced vibra- and report the method used.
tion devices.
7.2.3 Detector, for determining the dependent and indepen- 11. Procedure
dent experimental parameters, such as force (or stress), dis- 11.1 Mount the specimen in accordance with the procedure
recommended by the manufacturer.
11.2 Measure the length, width, and thickness of the speci-
5
Ferry, D. “Viscoelastic Properties of Polymers,” John Wiley & Sons, 1980. men to an accuracy of 60.01 mm.

2
 E 1640
11.3 Maximum strain amplitude should be within the linear
viscoelastic range of the material. Strains of less than 1 % are
recommended and should not exceed 5 %.
11.4 Conduct tests at a heating rate of 1°C/min and a
frequency of 1 Hz. Other heating rates and frequencies may be
used but shall be reported.
NOTE 4—The glass transition temperature measured by dynamic me-
chanical measurements is dependent upon heating rate and oscillatory
frequency. The experimental heating rate and the frequency of oscillation
should be slow enough to allow the entire specimen to reach satisfactory
thermal and mechanical equilibration. When the heating rate or oscillatory
rate is high, the experimental time scale is shortened, and the apparent Tg
is raised. Changing the time scale by a factor of 10 will generally result in
a shift of about 8°C for a typical amorphous material. The effect of these
variables on the temperature of the tangent delta peak may be observed by
running specimens at two or more rates and comparing the results (see
FIG. 1 Storage Modulus
appendix).
NOTE 5—Where possible in automated systems, a minimum of one data
point should be collected for each °C increase in temperature. At low and
high frequencies, use care in the selection of scanning rate and frequency
rate; select test conditions and a data collection rate that will ensure
adequate resolution of the mechanical response of the specimen. For
example, select a heating rate that allows the specimen to complete at least
one oscillation for each° C increase in temperature.
11.5 Measure and record the storage modulus, from 30°C
below to 20°C above the suspected glass transition region.
12. Calculation
12.1 For the purpose of this test method the glass transition
shall be taken as the extrapolated onset to the sigmoidal change
in the storage modulus observed in going from the hard, brittle
region to the soft, rubbery region of the material under test.
NOTE 6—Storage modulus may be displayed on a linear or logarithmic
scale. The reported glass transition temperature will differ depending upon
the scale chosen. The scale type (for example, linear or logarithmic) shall
be reported and must be the same for all parties comparing results.
12.1.1 Construct a tangent to the storage modulus curve
below the transition temperature.
12.1.2 Construct a tangent to the storage modulus curve at
the inflection point approximately midway through the sigmoi-
dal change associated with the transitions.
12.1.3 The temperature at which these tangent lines inter- FIG. 2 Loss Modulus
sect is reported as the glass transition temperature, T g (see Fig.
1).
NOTE 7—Under special circumstances agreeable to all parties, other 12.3 For measurements made at frequencies other than 1
temperatures taken from the storage modulus, loss modulus, or tangent Hz.
delta curve may be taken to represent the temperature range over which 12.3.1 Using a predetermined frequency shift factor (k) (see
the glass transition takes place. Among these alternative temperatures are appendix), calculate the first approximation of the glass tran-
the peak of the loss modulus (Tl ) or tangent delta (Tt ) curves as illustrated sition temperature (Tl8) using equation 1.
in Fig. 2 and Fig. 3, respectively. These temperatures are generally in the
order Tg < Tl < Tt. T2 F
Tl 8 5 T 1 k log 1 Hz (1)
12.1.4 The value of Tg reported from this document shall be
the mean value of duplicate determinations. 12.3.2 Calculate the glass transition temperature using
12.2 For fixed frequency measurements at 1 Hz. equation 2:
12.2.1 Report the mean value of duplicate determinations as T T1 8 F
T1 5 T 1 k log 1 Hz (2)
Tg.

3
 E 1640
14. Precision and Bias
14.1 An interlaboratory study of the measurement of the
glass transition temperature of an epoxy composite was con-
ducted in 1992. Following temperature calibration using a
polystyrene thermoplastic polymer (a secondary reference
material specifically prepared for this test program) each of 13
laboratories tested 4 test specimens. Seven laboratories used
linear storage modulus while nine laboratories used logarith-
mic storage modulus. Instruments from five manufacturers
were employed. The results were treated by Practice E 691 and
are given in an ASTM Research Report 6.
14.2 Precision:
14.2.1 Using a linear presentation of storage modulus,
r 5 95 % repeatability limit ~within laboratory! 5 4.1°C (3)
R 5 95 % reproducibility limit ~between laboratories! 5 12.3°C
(4)
14.2.1.1 Two values, each the mean of duplicate determina-
tions, should be considered suspect if they differ by more than
the limits described above. The respective standard deviations
among test results, related to the above values by the factor 2.8,
FIG. 3 Tangent Delta are:
s r 5 repeatability standard deviation 5 1.5°C (5)
sR 5 reproducibility standard deviation 5 4.4°C (6)
where:
k 5 Predetermined Frequency Shift Factor (see Appen- 14.2.2 Using a logarithmic presentation of storage modulus,
dix X1.1) r 5 95 % repeatability limit ~within laboratory! 5 2.6°C (7)
F 5 Frequency of Measurement (Hz) R 5 95 % reproducibility limit ~between laboratories! 5 7.7°C (8)
T 5 Glass Transition Temperature Observed at Fre-
quency F (K) 14.2.2.1 Two values, each the mean of duplicate determina-
Tl8 5 First Approximation for the Glass Transition Tem- tions, should be considered suspect if they differ by more than
perature at 1 Hz (K) the limits described above. The respective standard deviations
Tl 5 Glass Transition Temperature at 1 Hz (K) among test results, related to the above numbers by the factor
2.8, are:
Example: s r 5 repeatability standard deviation 5 0.9°C (9)
k 5 −12,417K
F 5 2 Hz sR 5 reproducibility standard deviation 5 2.7°C (10)
T 5 100°C 5 373K 14.3 Bias:
T8 5 ~373K! ~373K! 14.3.1 The glass transition temperatures ( Tg) of the poly-
373K 1 212,417K log 2 5 373K 2 3.37K
5 369.62 K styrene calibrant and epoxy composite used in this study were
T 5 ~373K! ~369.62K! assigned by thermomechanical analysis using Test Methods
373 1 212,417K log 2 5 373K 2 E 1363 and E 1545. Tg for the polystyrene was established to
3.34K be 101.4 6 1.8°C with 16 degrees of freedom (df) while Tgfor
5 369.66 K 5 96.5°C the epoxy composite was established to be 121.2 6 0.4°C with
21 df.
13. Report 14.3.2 Using a linear presentation of storage modulus, the
13.1 The report shall include the following: value for the epoxy composite glass transition by this dynamic
13.1.1 A complete identification and description of the mechanical test method was 120.8 6 4.2°C with 18 degrees of
material testing including dimensions and any pretreatment. freedom.
13.1.2 A description of the instrument used to perform the 14.3.3 Using a logarithmic presentation of storage modulus,
test. the value for the epoxy composite glass transition by this
dynamic mechanical test method was 118.6 6 2.6°C with 24
13.1.3 A description of the temperature calibration proce-
degrees of freedom.
dure used.
13.1.4 Whether linear or logarithmic storage modulus was NOTE 8—The glass transition derived from the linear presentation of
displayed. storage modulus was 2.2°C lower for polystyrene and 2.3°C higher for the
epoxy composite than those obtained for a logarithmic data presentation.
13.1.5 The calculated glass transition temperature.
13.1.6 The frequency of test and any extrapolation proce-
dures used to provide results comparable at 1 Hz.
13.1.7 The dynamic mechanical curves recorded. 6
A Research Report is available from ASTM. Request E37-1015.

4
 E 1640
15. Keywords
15.1 dynamic mechanical analysis; elastic modulus; glass
transition; modulus; storage modulus; temperature; thermal
analysis

APPENDIX

(Nonmandatory Information)

X1. Frequency Shift Factor

X1.1 The transition temperature is a function of experimen- F 2 5 Frequency of Measurement 2 (Hz)

tal frequency with the transition temperature increasing with T1 5 Transition Temperature at Frequency 1 (K)
increasing frequency. This test method requires that results be T2 5 Transition Temperature at Frequency 2 (K)
reported at a frequency of 1 Hz. Experimental data collected at
other frequencies may be extrapolated to 1 Hz through the use Example:
of a Frequency Shift Factor. F1 5 10 Hz
F2 5 2 Hz
X1.2 Determination of the Frequency Shift Factor: Tl 5 108°C 5 381K
T2 5 100°C 5 373K
X1.3 Measure the transition temperature at two or more k 5 ~381K! ~373K! 10 Hz
log 2 Hz
frequencies, according to the method. ~373K 2 381K!
NOTE X1.1—For best accuracy, the two test frequencies should be
separated by a decade of frequency but be as close to 1 Hz as practical. For
example, between 0.1 and 10 Hz. k 5 −12,417K
X1.4 The Frequency Shift Factor is determined using
X1.5 The frequency shift factor is a function of the material
equation A1:
and should be determined individually. The Frequency Shift
T 1 T2 F1 Factors for many thermoplastics and thermosets are nominally
k5T 2 T1 log F2 (X1.1)
2
8°C per decade of frequency change. Values for elastomers are
where: usually higher and may be as much as 40°C per decade of
k 5 Frequency Shift Factor ( K) frequency change.
F1 5 Frequency of Measurement 1 (Hz)

The American Society for Testing and Materials 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 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, 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).