ASTM E668-20

Standard Practice for Application of Thermoluminescence-Dosimetry (TLD) Systems for Determining Absorbed Dose in Radiation-Hardness Testing of Electronic Devices

SIGNIFICANCE AND USE
4.1 Absorbed dose in a material is an important parameter that can be correlated with radiation effects produced in electronic components and devices that are exposed to ionizing radiation. Reasonable estimates of this parameter can be calculated if knowledge of the source radiation field (that is, energy spectrum and particle fluence) is available. Sufficiently detailed information about the radiation field is generally not available. However, measurements of absorbed dose with passive dosimeters in a radiation test facility can provide information from which the absorbed dose in a material of interest can be inferred. Under certain prescribed conditions, TLDs are quite suitable for performing such measurements.
Note 2: For comprehensive discussions of various dosimetry methods applicable to the radiation types and energy and absorbed dose-rate range discussed in this practice, see ICRU Reports 14, 17, 21, and 34.
SCOPE
1.1 This practice covers procedures for the use of thermoluminescence dosimeters (TLDs) to determine the absorbed dose in a material irradiated by ionizing radiation. Although some elements of the procedures have broader application, the specific area of concern is radiation-hardness testing of electronic devices. This practice is applicable to the measurement of absorbed dose in materials irradiated by gamma rays, X rays, and electrons of energies from 12 to 60 MeV. Specific energy limits are covered in appropriate sections describing specific applications of the procedures. The range of absorbed dose covered is approximately from 10−2 to 104 Gy (1 to 106 rad), and the range of absorbed dose rates is approximately from 10−2 to 1010 Gy/s (1 to 1012 rad/s). Absorbed dose and absorbed dose-rate measurements in materials subjected to neutron irradiation are not covered in this practice. (See Practice E2450 for guidance in mixed fields.) Further, the portion of these procedures that deal with electron irradiation are primarily intended for use in parts testing. Testing of devices as a part of more massive components such as electronics boards or boxes may require techniques outside the scope of this practice.
Note 1: The purpose of the upper and lower limits on the energy for electron irradiation is to approach a limiting case where dosimetry is simplified. Specifically, the dosimetry methodology specified requires that the following three limiting conditions be approached: (a) energy loss of the primary electrons is small, (b) secondary electrons are largely stopped within the dosimeter, and (c) bremsstrahlung radiation generated by the primary electrons is largely lost.  
1.2 This standard does not purport to address all of the safety concerns, if any, associated with its use. It is the responsibility of the user of this standard to establish appropriate safety, health, and environmental practices and determine the applicability of regulatory limitations prior to use.  
1.3 This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.

General Information

• Status: Published
• Publication Date: 30-Jun-2020
• ICS: 31.020 - Electronic components in general
• Technical Committee: E10 - Nuclear Technology and Applications
• Drafting Committee: E10.07 - Radiation Dosimetry for Radiation Effects on Materials and Devices

Relations

• Replaces: ASTM E668-13 - Standard Practice for Application of Thermoluminescence-Dosimetry (TLD) Systems for Determining Absorbed Dose in Radiation-Hardness Testing of Electronic Devices
• Effective Date: 01-Jul-2020
• Refers: ASTM E170-17 - Standard Terminology Relating to Radiation Measurements and Dosimetry
• Effective Date: 01-Jun-2017
• Refers: ASTM E170-16a - Standard Terminology Relating to Radiation Measurements and Dosimetry
• Effective Date: 01-Oct-2016
• Refers: ASTM E170-16 - Standard Terminology Relating to Radiation Measurements and Dosimetry
• Effective Date: 15-Feb-2016
• Refers: ASTM E170-15a - Standard Terminology Relating to Radiation Measurements and Dosimetry
• Effective Date: 01-Sep-2015
• Refers: ASTM E170-15 - Standard Terminology Relating to Radiation Measurements and Dosimetry
• Effective Date: 15-Mar-2015
• Refers: ASTM E170-14a - Standard Terminology Relating to Radiation Measurements and Dosimetry
• Effective Date: 15-Oct-2014
• Refers: ASTM E170-14 - Standard Terminology Relating to Radiation Measurements and Dosimetry
• Effective Date: 01-Sep-2014
• Refers: ASTM E2450-11 - Standard Practice for Application of CaF₂(Mn) Thermoluminescence Dosimeters in Mixed Neutron-Photon Environments
• Effective Date: 01-Nov-2011
• Refers: ASTM E170-10 - Standard Terminology Relating to Radiation Measurements and Dosimetry
• Effective Date: 01-Jun-2010
• Refers: ASTM E170-09a - Standard Terminology Relating to Radiation Measurements and Dosimetry
• Effective Date: 15-Aug-2009
• Refers: ASTM E170-09 - Standard Terminology Relating to Radiation Measurements and Dosimetry
• Effective Date: 15-Jun-2009
• Refers: ASTM E666-09 - Standard Practice for Calculating Absorbed Dose From Gamma or X Radiation
• Effective Date: 01-Jun-2009
• Refers: ASTM E666-08 - Standard Practice for Calculating Absorbed Dose From Gamma or X Radiation
• Effective Date: 01-Nov-2008
• Refers: ASTM E170-08d - Standard Terminology Relating to Radiation Measurements and Dosimetry
• Effective Date: 01-Nov-2008

Overview

ASTM E668-20: Standard Practice for Application of Thermoluminescence-Dosimetry (TLD) Systems for Determining Absorbed Dose in Radiation-Hardness Testing of Electronic Devices provides essential guidance for using thermoluminescent dosimeters (TLDs) to determine the absorbed dose of ionizing radiation in materials, particularly for radiation-hardness testing of electronic devices. Accurate measurement of absorbed dose is critical in ensuring the reliability and performance of electronic components exposed to radiation environments. This standard covers procedures applicable to gamma rays, X-rays, and electrons within specified energy ranges and dose limits, supporting industries such as aerospace, defense, nuclear energy, and research laboratories.

ASTM E668-20 is intended to ensure consistency and reliability in radiation dosimetry, boosting confidence in radiation-hardness test results for sensitive electronic components.

Key Topics

• Absorbed Dose Determination
  The standard details practices for quantifying absorbed dose in materials, which is pivotal for correlating radiation effects on electronic devices. TLDs are used as passive dosimeters to provide reliable dose measurements when energy spectrum and particle fluence information may not be fully available.

• Radiation Types and Energy Ranges
  ASTM E668-20 applies to exposure from gamma rays, X-rays, and electrons with energies ranging from 12–60 MeV. The absorbed dose covered by this practice ranges from approximately 10⁻² to 10⁴ Gy (1 to 10⁶ rad), and dose rates from 10⁻² to 10¹⁰ Gy/s (1 to 10¹² rad/s).

• Thermoluminescence Dosimetry Systems
  The practice outlines handling, preparation, annealing, storage, calibration, and readout processes for TLDs. Emphasis is placed on using batches with consistent dosimetric response and controlling environmental factors such as temperature, humidity, and light exposure.

• Performance Testing
  Requirements for verifying TLD system performance, including batch uniformity, reproducibility, dose-rate independence, energy response, and environmental stability, are specified. Routine and batch-specific performance evaluation help ensure accurate dose measurement.

• Calibration and Documentation
  The standard highlights the importance of calibration traceability, consistent protocols, and clear reporting to ensure that results are reliable and meet regulatory standards.

Applications

• Radiation-Hardness Testing
  The primary application is in the assessment of electronic components and systems that must function reliably in radiation-rich environments, such as satellites, aerospace equipment, nuclear reactors, and military applications.

• Quality Assurance
  By standardizing the measurement techniques for absorbed dose, ASTM E668-20 supports quality assurance and risk mitigation in manufacturing and testing of radiation-sensitive electronics.

• Research and Development
  Laboratories and institutions involved in radiation effects research use this standard to benchmark materials and investigate radiation-induced failure mechanisms.

• Regulatory Compliance
  Organizations conducting radiation testing can use ASTM E668-20 to demonstrate compliance with international and national safety and performance standards.

Related Standards

• ASTM E170 – Terminology Relating to Radiation Measurements and Dosimetry
  Provides foundational definitions for terms used in radiation measurements.

• ASTM E666 – Practice for Calculating Absorbed Dose From Gamma or X Radiation
  Supports absorbed dose computation in different radiation fields.

• ASTM E2450 – Practice for Application of CaF2 (Mn) Thermoluminescence Dosimeters in Mixed Neutron-Photon Environments
  Addresses TLD use in environments not covered by E668-20.

• ICRU Reports (International Commission on Radiation Units and Measurements)
  Including Reports 14, 17, 21, 33, and 34 for comprehensive discussions on radiation dosimetry.

Conclusion

ASTM E668-20 offers comprehensive guidance for the precise application of TLD systems in determining absorbed dose during radiation-hardness testing of electronic devices. By following these standardized procedures, organizations ensure reliable, repeatable, and traceable results, supporting advancements in radiation-resilient electronics for critical applications across aerospace, defense, and scientific research sectors.

Standards Content (Sample)

This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the
Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.
Designation: E668 − 20
Standard Practice for
Application of Thermoluminescence-Dosimetry (TLD)
Systems for Determining Absorbed Dose in Radiation-
Hardness Testing of Electronic Devices
This standard is issued under the fixed designation E668; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision.Anumber in parentheses indicates the year of last reapproval.A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.
This standard has been approved for use by agencies of the U.S. Department of Defense.
1. Scope priate safety, health, and environmental practices and deter-
mine the applicability of regulatory limitations prior to use.
1.1 Thispracticecoversproceduresfortheuseofthermolu-
1.3 This international standard was developed in accor-
minescencedosimeters(TLDs)todeterminetheabsorbeddose
dance with internationally recognized principles on standard-
in a material irradiated by ionizing radiation. Although some
ization established in the Decision on Principles for the
elements of the procedures have broader application, the
Development of International Standards, Guides and Recom-
specific area of concern is radiation-hardness testing of elec-
mendations issued by the World Trade Organization Technical
tronic devices. This practice is applicable to the measurement
Barriers to Trade (TBT) Committee.
of absorbed dose in materials irradiated by gamma rays, X
rays, and electrons of energies from 12 to 60 MeV. Specific
2. Referenced Documents
energy limits are covered in appropriate sections describing
2.1 ASTM Standards:
specific applications of the procedures. The range of absorbed
−2 4 6 E170Terminology Relating to Radiation Measurements and
dose covered is approximately from 10 to 10 Gy (1 to 10
Dosimetry
rad), and the range of absorbed dose rates is approximately
−2 10 12 E380Practice for Use of the International System of Units
from 10 to 10 Gy/s (1 to 10 rad/s). Absorbed dose and
(SI) (the Modernized Metric System) (Withdrawn 1997)
absorbed dose-rate measurements in materials subjected to
E666Practice for CalculatingAbsorbed Dose From Gamma
neutron irradiation are not covered in this practice. (See
or X Radiation
Practice E2450 for guidance in mixed fields.) Further, the
E2450Practice for Application of CaF (Mn) Thermolumi-
portion of these procedures that deal with electron irradiation
nescence Dosimeters in Mixed Neutron-Photon Environ-
are primarily intended for use in parts testing. Testing of
ments
devices as a part of more massive components such as
2.2 International Commission on Radiation Units and Mea-
electronics boards or boxes may require techniques outside the
surements (ICRU) Reports:
scope of this practice.
ICRU Report 10eRadiobiological Dosimetry
NOTE 1—The purpose of the upper and lower limits on the energy for
ICRU Report 14Radiation Dosimetry: X Rays and Gamma
electron irradiation is to approach a limiting case where dosimetry is
RayswithMaximumPhotonEnergiesBetween0.6and50
simplified.Specifically,thedosimetrymethodologyspecifiedrequiresthat
MeV
the following three limiting conditions be approached: (a) energy loss of
ICRU Report 17Radiation Dosimetry: X Rays Generated at
the primary electrons is small, (b) secondary electrons are largely stopped
within the dosimeter, and (c) bremsstrahlung radiation generated by the
Potentials of 5 to 150 keV
primary electrons is largely lost.
ICRU Report 21Radiation Dosimetry: Electrons with Initial
1.2 This standard does not purport to address all of the
Energies Between 1 and 50 MeV
safety concerns, if any, associated with its use. It is the ICRU Report 33Radiation Quantities and Units
responsibility of the user of this standard to establish appro-
ICRU Report 34The Dosimetry of Pulsed Radiation
For referenced ASTM standards, visit the ASTM website, www.astm.org, or
This practice is under the jurisdiction of ASTM Committee E10 on Nuclear contact ASTM Customer Service at [e-mail]. For Annual Book of ASTM
Technology and Applicationsand is the direct responsibility of Subcommittee Standards volume information, refer to the standard’s Document Summary page on
E10.07 on Radiation Dosimetry for Radiation Effects on Materials and Devices on the ASTM website.
Materials and Devices. The last approved version of this historical standard is referenced on
Current edition approved July 1, 2020. Published August 2020. Originally www.astm.org.
approved in 1978. Last previous edition approved in 2013 as E668–13. DOI: Available from International Commission on Radiation Units and
10.1520/E0668-20. Measurements, 7910, Woodmont Ave., Suite 800, Bethesda, MD 20814.
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States
E668 − 20
ICRU Report 37Stopping Powers for Electrons and Posi- incremental volume within the material in which the condition
trons of secondary-electron equilibrium exists.
ICRU Report 90Key Data for Ionizing-Radiation Dosim- 3.1.9.1 Discussion—Additional definitions can be found in
etry: Measurement Standards and Applications ICRU Report 33.
3.1.10 secondary electrons— for the case of electron
3. Terminology
irradiation, electrons knocked out of the electron shells of the
3.1 Definitions:
material being irradiated by the primary electron. For the case
3.1.1 absorbed dose, D—thequotientofdɛ¯bydm,wheredɛ¯
of photon irradiation, energetic electrons (photoelectrons,
is the mean incremental energy imparted by ionizing radiation
Auger electrons, and Compton electrons) produced within the
to matter of incremental mass dm. Thus:
material being irradiated by the action of the incident photons.
3.1.10.1 Discussion—Secondary electrons are produced by
dε¯
D 5 (1)
the interaction of the primary electrons with the atoms of the
dm
material being irradiated. This interaction is a principal means
3.1.1.1 Discussion—Previously,thespecialunitofabsorbed
of energy loss for the primary electrons. The kinetic energy of
dose was the rad; however, the gray (Gy) has been adopted as
a secondary electron is typically much lower than that of the
the official SI unit (see Practice E380).
primary electron which creates it.
21 2
1Gy 51 J·kg 510 rad (2)
3.1.11 test conditions—the normal environmental condi-
tions prevailing during routine hardness-test irradiations such
3.1.2 absorbed-dose rate—the absorbed dose per unit time
as the ambient temperature, humidity, and lighting.
interval.
3.1.12 thermoluminescence dosimeter (TLD)—dosimeter
3.1.3 annealing—thermal treatment of a TLD prior to irra-
made of a material that stores energy when irradiated by
diation or prior to readout.
ionizing radiation and then releases that energy in the form of
3.1.3.1 Discussion—Pre-irradiation annealing of TLDs is
visible light when heated.
usually done to erase the effects of previous irradiation and to
readjust the sensitivity of the phosphor; pre-readout annealing
3.1.12.1 Discussion—TheTLphosphorcanbeusedaloneor
usually is done to reduce low-temperature TLD response.
incorporated into a material such as a TFE-fluorocarbon
3.1.4 calibration conditions—the normal environmental
matrix.
conditions prevailing during routine calibration irradiations
3.1.13 thermoluminescence dosimeter (TLD) batch—a
such as the ambient temperature, humidity, and lighting.
group of TLDs, generally originating from a single mix or lot
of TL phosphor, having similar TL responses and similar
3.1.5 equilibrium absorbed dose—the absorbed dose at
thermal and irradiation histories.
some incremental volume within the material which the
conditionofelectronequilibrium(asmanyelectronsofagiven
3.1.14 thermoluminescence dosimeter (TLD) reader—anin-
energy enter as leave the volume) exists (1) (see Appendix
strument used to measure the light emitted from a TLD
X1).
consisting essentially of a heating element, a light-measuring
device, and appropriate electronics.
3.1.6 exposure, X—the quotient of dQ by dm, where dQ is
the absolute value of the total charge of the ions of one sign
3.1.15 thermoluminescence dosimeter (TLD) response—the
producedinairwhenalltheelectrons(negatronsandpositrons)
measured light emitted by the TLD and read out during its
liberated by photons in a volume element of air having mass
heating cycle consisting of one of the following: (a) the total
dm are completely stopped in air.
lightoutputovertheentireheatingcycle,(b)apartofthattotal
light output, or (c) the peak amplitude of the light output.
dQ
X 5 (3)
dm 3.1.16 thermoluminescence (TL) phosphor—a material that
stores, upon irradiation, a fraction of its absorbed dose in
−1
UnitC·kg
various excited energy states. When thermally stimulated, the
3.1.6.1 Discussion—Formerly the special unit of exposure
material emits this stored energy in the form of photons in the
was the roentgen (R).
ultraviolet, visible, and infrared regions.
24 21
1 R 52.58 310 C·kg ~exactly! (4)
3.1.17 TLD preparation—the procedure of cleaning,
3.1.7 primary electrons—for the case of electron
annealing, and encapsulating the TLphosphor prior to irradia-
irradiation, the electrons introduced into the device under test
tion.
by the irradiation source.
3.2 For units and terminology in reports of data, Terminol-
3.1.8 secondary-electron equilibrium—for the case of elec-
ogy E170 may be used as a guide.
tron irradiation, the condition where as many secondary
electrons of a given energy enter a given volume as leave it.
4. Significance and Use
3.1.9 secondary-electron equilibrium absorbed dose—for
4.1 Absorbed dose in a material is an important parameter
the case of electron irradiation, the absorbed dose at some
that can be correlated with radiation effects produced in
electroniccomponentsanddevicesthatareexposedtoionizing
radiation. Reasonable estimates of this parameter can be
Theboldfacenumbersinparenthesesrefertothelistofreferencesattheendof
this practice. calculated if knowledge of the source radiation field (that is,
E668 − 20
energy spectrum and particle fluence) is available. Sufficiently cycle of the annealing oven, accurate timing of the annealing
detailed information about the radiation field is generally not period, and a reproducible cooling rate.
available. However, measurements of absorbed dose with
6.4 For low absorbed-dose measurements (<1 Gy (100
passive dosimeters in a radiation test facility can provide
rad)), dry nitrogen should be flowed through the heating
information from which the absorbed dose in a material of
chamberoftheTLDreaderduringreadout.Thissuppressesthe
interest can be inferred. Under certain prescribed conditions,
spurious TLD response that occurs in most forms of TLDs as
TLDs are quite suitable for performing such measurements.
a result of absorbed oxygen on the phosphor surface. If the
TLD reader uses hot gas to heat the TLDs, nitrogen should be
NOTE 2—For comprehensive discussions of various dosimetry methods
applicable to the radiation types and energy and absorbed dose-rate range
used.
discussed in this practice, see ICRU Reports 14, 17, 21, and 34.
6.5 Calibration-irradiated TLDs and all subsequent test-
5. Apparatus irradiated TLDs from the same batch should be read out with
the same reader using the same readout techniques and reader
5.1 The TLD System consists of the TLDs, the equipment
parameters. The calibration is valid only for that batch used in
used for preparation of the TLDs, and the TLD reader.
that particular reader. Readers that are different from the one
5.2 Calibration Facility delivers a known quantity of radia-
used for calibration, including those of the same make and
tion to materials under certain prescribed environmental and
model,donotnecessarilyindicatethesameresponseforTLDs
geometrical conditions. Its radiation source is usually a radio-
irradiated to the same absorbed dose.
60 137
activeisotope,commonlyeither Coor Cs,whoseradiation
6.6 TLDs may be used either as reusable or as single-use
output has been calibrated by specific techniques to some
dosimeters.Single-usedosimetersareirradiatedonce,readout,
specified uncertainty (usually to within 65%) and is traceable
and then discarded; they are generally used as received from
to national standards.
themanufacturer.Dosimetersthatarereusedarecycledrepeat-
5.3 Storage Facility provides an environment for the TLDs
edly through an anneal-irradiation-readout procedure.
before and after irradiation, that is light tight and that has a
6.7 The statistical methods specified in the following sec-
negligiblebackgroundabsorbed-doserate.ATLDstoredinthe
tions are optimal if the response of a batch of TLDs to a given
facility for the longest expected storage period should absorb
radiation dose is normally distributed. However, it has been
no more than 1% of the lowest absorbed dose expected to be
demonstrated that TLD distributions can be severely skewed,
measured in hardness-testing applications.
so that the sample mean may not be a suitable metric for small
5.4 Environmental Chamber is used in testing the effects of
sample sizes(2). In this case TLDs should be fielded in groups
temperature and humidity on TLD response. The chamber
of three, with either the lowest reading or the two extremes
should be capable of controlling the temperature and humidity
discarded. Whatever procedure is adopted, it must be applied
within 65% over the range expected under both calibration
consistently for all calibrations and routine measurements.
and test conditions.
NOTE 3—Adequately determining the normality of a TLD distribution
requires a large sample size.
6. Handling and Readout Procedures
7. Summary of Requirements for Performance Testing of
6.1 Bare TLDs should not be handled with the bare fingers;
a TLD System
dirtorgreaseontheirsurfacescanaffecttheirresponseandcan
contaminatetheheatingchamberoftheTLDreader.Avacuum 7.1 The performance of a specific TLD system should be
pen or tweezers coated with PTFE should be used in handling. evaluated to determine its suitability for use in a specific
If required, the TLDs can be cleaned by using the procedures radiation-hardness test. Acceptable performance of the TLD
in accordance with Appendix X2. system should be verified before applying the system in a
particular radiation-hardness-testing facility. Specific perfor-
6.2 TLDs, especially those with high sensitivity, should be
mance criteria are discussed in Section 8.
protected from light having an appreciable ultraviolet
component, such as sunlight or fluorescent light. Prolonged 7.2 Performance tests should be repeated whenever a sig-
exposure to ultraviolet light, either before or after irradiation, nificant change is made in the TLD system or in the specific
can cause spurious TLD response or enhanced post-irradiation application. Examples of such changes are: a change in the
fading. Incandescent lighting should be used for the TLD physical form or type of phosphor in theTLD, a change in any
preparation and readout areas. However, brief exposures of a critical component or in any adjustable readout factor of the
fewminutestonormalroomfluorescentlightingisnotlikelyto TLD reader, or a change in the irradiation source characteris-
significantly affect theTLD response except for low absorbed- tics.
dosemeasurements(<1Gyor<100rad)ormeasurementswith
7.3 Aparticular performance test may be omitted if widely
high-sensitivity TLDs.
accepted documentation exists in the scientific and technical
6.3 Preparation of the TLDs for irradiation consists of literature to show that the performance of the TLD system is
cleaning the TL phosphor (if required), annealing (if reusable satisfactory for that specific requirement. For example, if
TLDs are employed), and encapsulating the TL phosphor. previously accepted studies document that a particular TLD
Reusable TLDs require careful treatment during annealing in has no absorbed-dose-rate dependence for the expected condi-
order to obtain the best results in dose measurements. The tions of irradiation, then performance testing for absorbed-
annealingprocedureshouldincludeareproducibletemperature dose-rate dependence of that TLD system is unnecessary. All
E668 − 20
reportsoftestresultsshouldincludeappropriatereferencesthat withinthebatcharenotidentified.Totestthereproducibilityof
substantiate the performance of the system and thereby justify theresponseofanindividualreusabledosimeter,thefollowing
the omission of such performance tests. procedures should be followed:
8.2.1 Select the individual TLD to be tested, prepare it,
7.4 IfaparticularTLDsystemfailstomeettheperformance
irradiateitinthecalibrationfacilitytoaspecificabsorbed-dose
specification of any performance test, then use of that TLD
level (for example, at the midpoint of the absorbed-dose range
system is not recommended. Such a system may be used only
of interest), and read it out. In an identical manner, repeat this
if appropriate corrections to the TLD response can be deter-
procedure 30 times. Determine the variance, s , of the re-
mined sufficiently well in order that the results of the specific
sponses and estimate the standard deviation of the TLD
radiation-hardness test can be determined within the required
~ !
response distribution σ5=s . The standard deviation, σ,
uncertainty.
¯
should not exceed 5% of the mean response value, Y , that is
7.5 The number of TLDs, or the number of replicates of
¯
σ≤ (0.05)Y .
measurements with a single TLD, used for each test should be
sufficient to assure that the test results are significant at the
8.2.2 Some types of TLDs may exhibit a change in sensi-
95%confidencelevel.SeeRef(3)fordetailsoftheprocedures
tivity (that is, response per unit absorbed dose) with repeated
usedtoselectrandomsamplesandtodeterminethesamplesize
anneal-irradiation-readout cycling. This effect is most pro-
required.
nouncediftheTLDisnotannealedthoroughly.Thetestresults
in accordance with 8.2.1 may not show such a change in
NOTE4—Ifasampleof nmeasurements Y , Y ,., Y istaken,thebest
1 2 n
responsesensitivity.However,ifsuchachangeisshowninthat
estimate of the population mean, m, of a normal distribution is given by
¯
the mean value, Y, of the sample: test or if it appears after a larger number of cycles than
specified in that test, then a different analysis of the data is
n
¯
required. In this case, a curve should be fitted to the data of
Y 5 Y (5)
i
(
n
i51
responseversusnumberofcyclesbyaleast-squaresmethod.A
The best estimate of the variance, σ , of the distribution is given by the
2 measure of reproducibility would then be given by the average
variance, s , of the sample:
standard deviation of the data points from the least-squares
n
¯ curve. The performance criterion is the same as in 8.2.1.
s 5 ~Y 2 Y ! (6)
( i
n 21
i51
8.2.3 Since the identity of each TLD is maintained when it
The quantity σ ~5=s ! is called the standard deviation of the distri-
is utilized as an individual dosimeter, it is not necessary that
bution. The degree to which s is a best estimate of σ depends on the
groupsofsuchindividualTLDsmeetthebatchrequirementsin
sample size and, as might be expected, s becomes a better estimate of σ
accordance with 8.1. However, for the other performance tests
as the sample size increases.
andcorrectionfactorsdiscussedinSection8,itisassumedthat
8. Specific Performance Tests and Correction Factors
such tests and factors are evaluated by utilizing TLDs in a
batch mode.
8.1 Uniformity of TLD Response Within a Batch:
8.1.1 Select a random sample of 30 TLDs from a batch.
8.3 Dependence of TLD Response on Absorbed-Dose Rate:
TreatingallofthesampleTLDsinanidenticalmanner,prepare
8.3.1 From a TLD batch meeting the requirements in
them, irradiate them in the calibration facility to the same
accordance with 8.1.1, select a number of TLDs. Divide the
absorbed-dose level, and read them out. Determine the
TLDs into x number of groups, each group containing n
variance, s , of the sample and estimate the standard deviation
samples. Determine the absorbed-dose-rate range of interest
of the TLD response distribution ~σ5=s !. the standard
for the intended application and divide this range into x
deviation, σ, should not exceed 8% of the sample mean value,
intervals(forexample,oneintervalperdecade).Prepareallthe
¯ ¯
Y ; that is,σ≤ (0.08)Y . The sample size specified (30) is the
0 0
TLDs in an identical manner and irradiate each group to the
number of measurements required to estimate the standard
same dose level, but at a different absorbed-dose rate for each
deviation, σ, of the TLD response distribution within 25% of
x group, covering the absorbed-dose-rate range of interest.
its true value at a 95% confidence level (see 2.4 of Ref 3).
¯
Read out theTLDs. Determine the mean response, Y, for each
i
¯
8.1.2 For reusable TLDs that have been subjected to a xgroupofnsamples.Determineanoverallmeanvalue,Y ,for
number of anneal-irradiation cycles, the uniformity of the
all x group means. Then the absolute difference between any
batch response should be verified periodically by repeating the group mean and the overall mean should not exceed 20% of
test in accordance with 8.1.1. The frequency required for the
the overall mean. That is,
test depends on the type of TLD and on its previous anneal-
¯ ¯ ¯
Y 2 Y # ~0.2!Y (7)
? i 0? 0
irradiation history. Retesting of the batch uniformity becomes
particularly important for TLDs irradiated to high-dose levels ¯ ¯ ¯
8.3.2 If |Y − Y | > (0.05)Y , then appropriate correction
i 0 0
2 4
(>10 Gy (10 rad)). See, however, X2.2.2.
factorstotheTLDresponseasafunctionofabsorbed-doserate
should be determined by the procedures that follow.
8.2 ReproducibilityofTLDResponseofIndividualReusable
Dosimeters—Certain types of TLDs may be utilized as indi- 8.3.3 Determinethenumberofsamples nrequiredineach x
¯
vidual reusable dosimeters. In this case, the identity of each group in order to detect a difference of δ = (0.05)Y between a
individual dosimeter is maintained during repeated measure- group mean and the overall mean for a confidence level of
ment cycles throughout its useful life. This is in contrast to 95% and a probability of 0.50 of failing to detect such a
utilization in the batch mode where individual dosimeters difference. It is assumed that the variance, σ , of the TLD
E668 − 20
response determined in accordance with 8.1.1, does not vary 8.5.1 If the geometrical orientation of theTLD with respect
with the absorbed-dose rate. Calculate the following param- totheradiation-hardnesstestfieldissignificantlydifferentthan
eter: its orientation with respect to the calibration radiation field,
then any dependency of the TLD response on the direction of
δ δ
d 5 5 (8)
theincidentradiationshouldbedetermined.Selectanumberof
=2σ σ=2
TLDs from a batch meeting the requirements in accordance
with 8.1.1. Divide the TLDs into x number of groups, each
Then the sample size, n, is required for each x group to
group containing n samples. Prepare the TLDs in an identical
satisfy the above parameters is read off the graph of n versus d
manner, and irradiate each group to the same absorbed-dose
(see Fig. X3.1).
level in the following manner: (a) group g , in the usually
8.3.4 Example of Sample Number Determination—If σ =
¯ oriented direction used for routine calibration, and (b) groups
0.03 Y (determined in 8.1.1),
g , g,., g oriented, respectively, at angles θ , θ,., θ ,
1 2 x 1 2 x
¯
0.05 Y
relative to the usually oriented direction with the center of the
d 5 51.18 (9)
¯
=
2 0.03 Y group at the same distance from the source. These angles
should divide, in equal intervals of no more than 30° each, the
From Fig. X3.1, the sample size required is n = 4.4. The
angle between the normal and the maximum possible angle of
samplesizeshouldbe5,obtainedbyroundinguptothenearest
incidence of the radiation-hardness test field. Read out all the
integer.
¯
TLDs. Determine the mean response, Y, for each x group of n
i
¯
NOTE5—Onemethodbywhichthistestrequirementcanbecarriedout samples. Then the absolute difference between the mean, Y ,
isbycomparingtheTLDresponseswiththeresponseofanotherradiation
for the normally used calibration orientation and the mean for
dosimeter whose absorbed-dose-rate dependence is known. A suitable
¯
any other orientation should not exceed 5% of the mean Y .
type of dosimeter for use in most cases would be a calorimeter whose
Thus:
response is absorbed-dose-rate independent and whose radiation absorp-
tion properties are similar to the TLD under test.
¯ ¯ ¯
Y 2 Y #0.05 Y (10)
? i 0? 0
8.4 Dependence of TLD Response on Energy:
To determine the sample size n required for each x group,
8.4.1 The radiation absorption properties of the TLDs em-
use the procedures in accordance with 8.3.3.
ployed in radiation-hardness testing should be similar to those
of the material in which the dose is to be estimated. Calcula-
NOTE 6—This test applies only to a collimated-beam type source
tions can be made to determine the effects of a broad incident
geometry. If the angle of incidence of the radiation from the source is
energyspectrumontheresponseoftheTLDscomparedtothat nearly isotropic, then it is recommended that the TLDs and their
encapsulation material should be as nearly spherical as possible.
ofthematerialofinterest(usuallysilicon).Therequirementsof
7.5 are not applicable to this section.
8.6 Dependence of TLD Response on Time Between Prepa-
8.4.2 If the ratios [(µ /ρ) ]/[(µ /ρ) ] and [(S/ρ) ]/
ration and Irradiation:
en TLD en mat TLD
[(S/ρ) ] are equal to 1.0 within 610% over a significant
mat 8.6.1 A change in TLD sensitivity can occur during the
range of energy spectrum (for both calibration and test irradia-
storage period between preparation and irradiation. This may
tions) incident upon both the TLD and the material of interest,
beasignificanteffectifawiderangeofstorageperiodsisused.
then the energy-response performance of the TLD system is
Usethefollowingproceduretotestforthiseffect.FromaTLD
acceptable. Here, µ /ρ is the mass photon energy absorption
en batchmeetingtherequirementsinaccordancewith8.1.1,select
coefficient and S/ρ is the mass collision electron stopping
two equal groups of n samples each. Prepare the first group of
power. Tables of values of µ /ρ and S/ρ for several materials
en TLDs and place them in the storage facility for a time interval
maybefoundinAppendixX4.Thephrase“significantrangeof
equal to the maximum time interval expected between prepa-
the energy spectrum” means the minimum and maximum
ration and irradiation during routine application in either
energy limits containing those incident radiation particles
calibration or hardness testing. At a later time, prepare the
(eitherphotonsorelectrons)thatcontributeatleast90%ofthe
second group of TLDs, and place them in the storage facility
absorbed dose. In this case, detailed energy spectral informa-
for the minimum time interval expected between preparation
tionisnotrequired;theincidentparticlefluence(eitherphotons
and irradiation. Time the procedures so that the ends of the
or electrons) between the energy limits is sufficient.
storage period for both groups occur simultaneously. Then
8.4.3 If the energy spectrum of the radiation incident upon
irradiate both groups to the same absorbed-dose level in the
the TLD (under both calibration and test conditions) and the
calibration facility and read them all out. The difference
material of interest (under test conditions) is well known, then ¯
between the meanTLD response, Y , of the first group and the
the conversion from absorbed dose in the TLD to absorbed ¯
mean response, Y , of the second group is a measure of the
dose in the material of interest can be calculated from such
effectofstoragetimebetweenpreparationandirradiation.This
data. If this conversion can be made to an uncertainty of
difference should not exceed 20% of the average of the means
610% or less, then the performance of the TLD system is
of the two groups. Thus:
acceptable. In this case, the criteria concerning the ratios of
¯ ¯
Y 1Y
µ /ρ and S/ρ in 8.4.2 need not be met. (See Practice E666 for 1 2
¯ ¯
en
Y 2 Y # 0.2 (11)
~ !
? ?
1 2
more specific guidelines.)
8.5 Dependence of TLD Response on Direction of Incident 8.6.2 If the effect tested for in accordance with 8.6.1
Radiation: exceeds 5% of the average of the group means, then the
E668 − 20
functional dependence of the TLD response on the storage a procedure used that eliminates the need for a correction. A
period should be determined in order that appropriate correc- procedure that achieves the latter would be one in which all
tion factors may be applied. This functional dependence may TLDs are read out at the same elapsed time after the end or
be determined by the procedures that follow. irradiation. Such a procedure is often inconvenient or imprac-
tical. Therefore, it is usually necessary to apply a fading
8.6.3 The range of the elapsed time intervals between
correction to the TLD response. The fading characteristics of
preparation and irradiation of interest is determined from the
theTLD system may be determined by the test procedures that
minimum and maximum intervals utilized in 8.6.1. Tests
follow.
shouldbeperformedataminimumoftwointervalsperdecade
of elapsed time over the entire range. For example, if the 8.7.3 Determine the minimum and maximum elapsed times
minimum elapsed time is 0.1 h and maximum elapsed time is between the end of the irradiation period and readout. Tests
100h,thenanappropriatesetoftestswouldbeatelapsedtimes should be performed at a minimum of two time intervals per
of 0.1, 0.3, 1, 3, 10, 30, and 100 h. From aTLD batch meeting decade of elapsed time over the entire period in accordance
the requirements in accordance with 8.1.1, select as many with 8.6.3. From a TLD batch meeting the requirements in
groups of n samples each as there are elapsed time intervals as accordancewith8.1.1,selectasmanygroupsofnsampleseach
determinedabove.PrepareagroupofTLDs,andplaceitinthe as there are elapsed time intervals as determined above. Each
storage facility for the appropriate preselected test-time inter- group of TLDs should undergo identical preparation and then
val. Repeat this procedure for all preselected storage time should be irradiated in the calibration facility to the same dose
level.The groups ofTLDs are placed in the storage facility for
intervals from the maximum to the minimum elapsed time.
Arrange the storage times so that the ends of all procedures allpreselectedappropriatetimeintervalsfromthemaximumto
theminimumelapsedtime.Arrangethetimeofirradiationsfor
occur simultaneously. Then irradiate all groups to the same
dose level in the calibration facility and read them all out as all the groups so that the ends of their storage periods occur
quickly as possible thereafter. This procedure is designed to simultaneously. Read out all the TLDs. Determine the mean
minimize effects on dosimeter response caused by fading and response for each group of TLDs. A plot of mean TLD
variation in reader output. Determine the mean response for response versus elapsed time provides the fading correction
each group of TLDs. A plot of mean TLD response versus factor. The number of samples n required for each group of
elapsed time provides a correction factor for a change in TLD TLDs should be determined by the procedures in accordance
sensitivity as a function of storage period. The number of with 8.3.3.
samples n required for each group of TLDs should be deter-
8.8 Dependence of TLD Response on Temperature During
mined by the procedure in accordance with 8.3.3.
Storage or Irradiation:
8.7 Dependence of TLD Response on Time Between Irra-
8.8.1 If the storage temperature experienced by the TLDs
diation and Readout:
between preparation and irradiation during routine radiation-
hardness testing differs from the temperature during routine
8.7.1 Significant fading of the TLD response may occur
during the storage period between the end of irradiation and calibration by more than 10°C, the test in accordance with 8.6
should be repeated over the range of temperatures expected
readout. Use the following procedure to test for this effect.
From a TLD batch meeting the requirements in accordance usingtheenvironmentalchamberinsteadofthestoragefacility.
The performance criteria in accordance with 8.6 are applicable
with 8.1.1, select two equal groups of n samples each. Prepare
thefirstgroupofTLDs,irradiatetheminthecalibrationfacility to this section.
to a specific dose level, then place them in the storage facility 8.8.2 If the storage temperature experienced by the TLDs
for an interval equal to the maximum time interval expected
between irradiation and readout during routine radiation-
during routine application (for either calibration or hardness hardness testing differs from the temperature during routine
testing) between the end of the irradiation period and readout.
calibration by more than 10°C, the test in accordance with 8.7
Prepare the second group of TLDs, irradiate them in the
should be repeated over the range of temperatures expected
calibration facility to the same dose level as the first group,
usingtheenvironmentalchamberinsteadofthestoragefacility.
then place them in the storage facility for an interval equal to
The performance criteria in accordance with 8.7 are applicable
the minimum time interval expected between the end of
to this section.
irradiation and readout.Time the procedures so the ends of the
8.8.3 IfthetemperatureexperiencedbytheTLDsduringthe
storageperiodsforbothgroupsoccursimultaneously.Readout
irradiation period during routine radiation-hardness testing
all of the TLDs. The absolute difference between the mean
differsfromthetemperatureduringroutinecalibrationbymore
¯ ¯
TLDresponse, Y ,ofthefirstgroupandthemeanresponse, Y ,
1 2 than 10°C, then the effect on TLD response should be
of the second group is a measure of the effect of storage time
determined by the following procedure: Select a number of
between the end of irradiation and readout. The difference
TLDs from a batch meeting the requirements in accordance
shouldnotexceed20%oftheaverageofthemeansofthetwo
with 8.1.1, prepare them in an identical manner, and separate
groups. Thus:
themintotwoequalgroupsofnsampleseach.Irradiatethefirst
group in the calibration facility to a specific dose level,
¯ ¯
Y 1Y
1 2
¯ ¯
Y 2 Y # 0.2 (12)
~ ! maintaining the temperature of the TLDs at the minimum
? ?
1 2
temperature expected during routine hardness-test irradiation.
¯ ¯
8.7.2 If the fading effect is greater than (0.05)(Y + Y )/2, Irradiatethesecondgroupinthecalibrationfacilitytothesame
1 2
theneitheracorrectionshouldbemadetotheTLDresponseor dose level, maintaining the temperature of the TLDs at the
E668 − 20
maximum temperature expected during routine hardness-test absorbed-dose level to define accurately the shape of the
irradiations. Readout all of the TLDs. The difference between characteristic response curve. The number of TLD samples
¯
the mean TLD response, Y , of the first group and the mean requiredtodeterminethemeanresponseateachabsorbed-dose
¯
response, Y , of the second group is a measure of the effect of level is given by the following procedures:
temperaturevariationduringirradiation.Thisdifferenceshould ¯
9.3.1 In order to determine the mean TLD response, Y ,
notexceed20%oftheaverageofthemeansofthetwogroups.
within 65% at a 95% confidence level, the number of TLD
Ifthemagnitudeoftheeffectisgreaterthan5%oftheaverage
samplesrequiredforagivenabsorbed-doselevelisasfollows:
of the means, then appropriate corrections to the TLD re-
2 2
2.045 s
~ !
sponses should be determined by procedures analogous to
n 5 (13)
¯
~0.05 Y !
those in accordance with 8.6.
8.9 Dependence of TLD Response on Humidity—Ingeneral,
where s is the estimate of the standard deviation σ of the
the responses of the most widely used TLDs have not been TLD response distribution as determined by the procedures in
¯
shown to be sensitive to changes in relative humidity.
accordance with 8.1.1. For example , if s = (0.06) Y , then:
However, if a TLD that is hygroscopic is being considered for
¯
2.045 ~0.06 Y !
~ !
application in radiation-hardness testing, then the performance
n 5 56.0 (14)
¯
~0.05 Y !
tests in accordance with 8.8 should be repeated with the
humidity as the variable parameter and the temperature main-
(See Section 2.3.2 of Ref (3) for more details.)
tained at the maximum value used in the temperature tests.
9.3.2 The procedures described in 9.3.1 assume that the
NOTE 7—Once a TLD system of a particular TL-phosphor type and
standarddeviationoftheTLDresponsedistributionisconstant
physicalconfigurationhasmettheperformancerequirementsofSection8,
for all absorbed-dose levels measured. This assumption gener-
new batches of the same type need only be tested for the requirements of
ally is valid over most of the usable absorbed-dose range for
8.1 (batch uniformity) and 8.7 (post-irradiation fading). See also 7.2 and
mostTLDsbutmaynotbecorrectforveryhigh-absorbed-dose
7.3.
3 5
levels of approximately 10 Gy(TLD) (10 rad(TLD)) or
9. Calibration of the TLD System
higher.IftheTLDsystemisusedattheseabsorbed-doselevels,
then redetermine the standard deviation of the response distri-
9.1 CalibratetheTLDsysteminamannersuchthattheTLD
bution at these levels by repeating the procedures in accor-
response can be related directly to the absorbed dose in the
dance with 8.1.1.
TLD phosphor. Use a suitable, well-characterized radiation
source in the calibration. Radioactive isotope sources such as
9.4 During a calibration irradiation, encapsulate the TL
60 137
Co or Cs are generally used for this purpose. Exposure
phosphor in a material with a thickness just sufficient to
rates (or absorbed-dose rates) from such sources should be
product electron equilibrium in the phosphor (see Appendix
known to be better than 65% at all locations normally used
X1). If possible, the encapsulation material should have the
for calibration irradiations. The methods used for determining
same thickness on all sides of the dosimeter.
the output rates from such calibration sources include the use
NOTE 8—The encapsulation material should resemble the phosphor
ofsecondarystandardradiationmeasuringinstruments,suchas
material as closely as possible with respect to radiation absorption
air-ionization chambers or transfer dosimeters, whose calibra-
properties. For example, if the TLphosphor is CaF , acceptable encapsu-
tion is traceable to the National Institute of Standards and
lation material would be calcium fluoride, 1000-series aluminum, or
Technology (NIST) or other recognized calibration laboratory.
silicon (see Appendix X2). If the calibration source is Co, then a
Other types of dosimeters whose responses are absolute thickness of 2.2 mm of aluminum (equal to the practical range of the
highest-energy secondary electrons produced) could establish electron
(require no calibration), such as calorimeters, may also be
equilibrium in the CaF phosphor. This thickness is sufficient to stop
employed for calibration of such sources.
secondary electrons that might be generated by the source photons in
material other than the encapsulation material.
9.2 The response of most types of TLDs generally is not
linear as a function of absorbed dose (4). The response of a
9.5 Correct for attenuation of the photons from the source
typical TLD is nearly linear from low-absorbed-dose levels to
by the layer of material used to establish electron equilibrium,
approximately 10 Gy(TLD) (10 rad (TLD)), then becomes
using the following formula:
2 3 4
supralinear up to approximately 10 to 10 Gy(TLD) (10 to
5 µ
en
10 rad (TLD)) where saturation effects become evident.
X 5 X exp 2 ρx (15)
F S D G
ρ
Exercise care in the use of the TLD system for absorbed-dose
3 5
levels of approximately 10 Gy(TLD) (10 rad(TLD)) or
where:
highertoensurethatthechangeinthesystemresponseperunit
X = unattenuated exposure at the position of the TLD
absorbed dose is adequate in order that the absorbed dose can
phosphor, roentgens,
be determined within the required uncertainty.
X = unattenuated exposure, roentgens,
µ /ρ = mass energy absorption coefficient of the encapsu-
9.3 Theabsorbed-doserangeofcalibrationshouldcoverthe en
lation material for the effective source photon
maximum absorbed-dose range of interest for the intended
energy, cm /g,
application.Measureaminimumofthreeabsorbed-doselevels
ρ = density of the encapsulation material, g/cm , and
per decade of absorbed dose covered. Since the TLD response
x = thickness of the encapsulation material, cm.
versus absorbed dose for most types of TLDs generally is not
linear, make a sufficient number of measurements at each Values of µ /ρ may be found in Appendix X4 and Ref (5).
en
E668 − 20
NOTE 9—The attenuation formula given is not rigorously correct for a
linear accelerators (linacs). Maximum absorbed-dose rates
3 10
broad-beam geometry as it does not include a buildup factor. Buildup
range from about 10 Gy(Si)/s (10 rad (Si)/s) to about 10
factors generally are not available for a wide range of energies, materials,
Gy(Si)/s (10 rad(Si)/s).
and geometries. However, the formula gives the results that are in
reasonable agreement with more rigorous treatments for materials of low
10.2 Characterization of Radiation Field—TLDs irradiated
to medium atomic number of relatively thin sections over the range of
in various locations in the test facility under free-field condi-
photon energies that are applicable to this practice
tionscanbeusedtocharacterizetheradiationfield.Inaddition,
9.6 Once the exposure has been determined, the absorbed
it may be desirable and practical to monitor the radiation field
dose(grays)totheencapsulatedTLphosphorisfoundfromthe
of the source during actual radiation-hardness testing of
formula:
electronic devices. When there is a significant variation of the
source output from irradiation to irradiation, use TLDs as
~µ /ρ!
en TLD
D 5 0.877 310 X (16)
~ !
TLD
monitors.
µ /ρ
~ !
en
air
10.2.1 TLD Use for Irradiation with Gamma Sources with
−2
The factor (0.877 × 10 ) is used to convert exposure
Energies Above 300 keV—For irradiation by relatively high
(roentgens)toabsorbeddoseinair(grays).Thesubscriptsrefer
energy gamma rays (e.g. Co gamma rays), encapsulate the
to the material of interest. As in 9.5, the µ /ρ values are
en
TL phosphor in material with sufficient thickness to produce
evaluated at the effective calibration source photon energy.
electron equilibrium conditions in the TL phosphor (see 9.4
This formula is valid only if electron equilibrium exists in the
and Appendix X1 for details).The equilibrium material should
TL phosphor. It is assumed that the incident photon fluence is
have radiation absorption properties similar to theTLD.When
essentially monoenergetic. If this is not the case, then average
the TLD material is CaF :Mn, 1000-series aluminum is an
all of the energy-dependent energy absorption coefficients of
acceptable equilibrium material.
9.5 and 9.6 over the appropriate energy spectrum.
10.2.2 TLD Use for Irradiation with Pulsed X-Ray Sources:
−2
NOTE 10—The value of 0.877 × 10 Gy(air)/roentgen is based on an 10.2.2.1 Pulsed X-ray sources provide a particularly diffi-
average energy of 33.97 6 0.20 eVrequired to produce an ion pair in dry
cult problem because they have a wide range of photon
air (see ICRU Report 90).
energies. The appropriate treatment of dosimetry for sources
9.7 The absorbed dose calibration results of the procedures
depends on the peak electron energy which is used to generate
of 9.3 – 9.6 are valid only for a given batch of TLDs. A the X-rays. A summary of some relevant considerations for
different batch generally will have a different radiation sensi-
doseinthedosimeter(anddoseinthedeviceundertest(DUT)
tivity.However,thisdifferenceisusuallyaconstantfactorover is provided in Table 1 for flash X-ray sources lying in three
the entire absorbed-dose range. Therefore, it may not be
energy bands.
necessary to generate a new calibration curve over the entire
10.2.2.2 For irradiation with pulsed X rays, encapsulate the
absorbed-dose range covered. Measurements at a minimum of
...

This document is not an ASTM standard and is intended only to provide the user of an ASTM standard an indication of what changes have been made to the previous version. Because
it may not be technically possible to adequately depict all changes accurately, ASTM recommends that users consult prior editions as appropriate. In all cases only the current version
of the standard as published by ASTM is to be considered the official document.
Designation: E668 − 13 E668 − 20
Standard Practice for
Application of Thermoluminescence-Dosimetry (TLD)
Systems for Determining Absorbed Dose in Radiation-
Hardness Testing of Electronic Devices
This standard is issued under the fixed designation E668; 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 (´) indicates an editorial change since the last revision or reapproval.
This standard has been approved for use by agencies of the U.S. Department of Defense.
1. Scope
1.1 This practice covers procedures for the use of thermoluminescence dosimeters (TLDs) to determine the absorbed dose in
a material irradiated by ionizing radiation. Although some elements of the procedures have broader application, the specific area
of concern is radiation-hardness testing of electronic devices. This practice is applicable to the measurement of absorbed dose in
materials irradiated by gamma rays, X rays, and electrons of energies from 12 to 60 MeV. Specific energy limits are covered in
appropriate sections describing specific applications of the procedures. The range of absorbed dose covered is approximately from
−2 4 6 −2 10 12
10 to 10 Gy (1 to 10 rad), and the range of absorbed dose rates is approximately from 10 to 10 Gy/s (1 to 10 rad/s).
Absorbed dose and absorbed dose-rate measurements in materials subjected to neutron irradiation are not covered in this practice.
(See Practice E2450 for guidance in mixed fields.) Further, the portion of these procedures that deal with electron irradiation are
primarily intended for use in parts testing. Testing of devices as a part of more massive components such as electronics boards or
boxes may require techniques outside the scope of this practice.
NOTE 1—The purpose of the upper and lower limits on the energy for electron irradiation is to approach a limiting case where dosimetry is simplified.
Specifically, the dosimetry methodology specified requires that the following three limiting conditions be approached: (a) energy loss of the primary
electrons is small, (b) secondary electrons are largely stopped within the dosimeter, and (c) bremsstrahlung radiation generated by the primary electrons
is largely lost.
1.2 This standard dosedoes not purport to address all of the safety concerns, if any, associated with its use. It is the
responsibility of the user of this standard to establish appropriate safety safety, health, and healthenvironmental practices and
determine the applicability of regulatory limitations prior to use.
1.3 This international standard was developed in accordance with internationally recognized principles on standardization
established in the Decision on Principles for the Development of International Standards, Guides and Recommendations issued
by the World Trade Organization Technical Barriers to Trade (TBT) Committee.
2. Referenced Documents
2.1 ASTM Standards:
E170 Terminology Relating to Radiation Measurements and Dosimetry
E380 Practice for Use of the International System of Units (SI) (the Modernized Metric System) (Withdrawn 1997)
E666 Practice for Calculating Absorbed Dose From Gamma or X Radiation
E2450 Practice for Application of CaF (Mn) Thermoluminescence Dosimeters in Mixed Neutron-Photon Environments
2.2 International Commission on Radiation Units and Measurements (ICRU) Reports:
ICRU Report 10e Radiobiological Dosimetry
ICRU Report 14 14—RadiationRadiation Dosimetry: X Rays and Gamma Rays with Maximum Photon Energies Between 0.6
and 50 MeV
This practice is under the jurisdiction of ASTM Committee E10 on Nuclear Technology and Applicationsand is the direct responsibility of Subcommittee E10.07 on
Radiation Dosimetry for Radiation Effects on Materials and Devices on Materials and Devices.
Current edition approved Jan. 1, 2013July 1, 2020. Published January 2013August 2020. Originally approved in 1978. Last previous edition approved in 20102013 as
E668 – 10.E668 – 13. DOI: 10.1520/E0668-13.10.1520/E0668-20.
For referenced ASTM standards, visit the ASTM website, www.astm.org, or contact ASTM Customer Service at [e-mail]. For Annual Book of ASTM Standards
volume information, refer to the standard’s Document Summary page on the ASTM website.
The last approved version of this historical standard is referenced on www.astm.org.
Available from International Commission on Radiation Units and Measurements, 7910, Woodmont Ave., Suite 800, Bethesda, MD 20814.
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States
E668 − 20
ICRU Report 17 17—RadiationRadiation Dosimetry: X Rays Generated at Potentials of 5 to 150 keV
ICRU Report 21 21—RadiationRadiation Dosimetry: Electrons with Initial Energies Between 1 and 50 MeV
ICRU Report 31—Average Energy Required to Produce an Ion Pair
ICRU Report 33 33—RadiationRadiation Quantities and Units
ICRU Report 34 34—TheThe Dosimetry of Pulsed Radiation
ICRU Report 37 37—StoppingStopping Powers for Electrons and Positrons
ICRU Report 90 Key Data for Ionizing-Radiation Dosimetry: Measurement Standards and Applications
3. Terminology
3.1 Definitions:
3.1.1 absorbed dose, D—the quotient of ddɛ¯ε¯ by dm, where ddɛ¯ε¯ is the mean incremental energy imparted by ionizing
radiation to the matter in a volume element and matter of incremental mass dmm. is the mass of matter in that volume
element.Thus:
d¯ε
D 5 (1)
dm
Previously, the special unit of absorbed dose was the rad; however, the gray (Gy) has been adopted as the official SI unit
(see Practice E380).
21 2
1 Gy 5 1 J·kg 5 10 rad (2)
3.1.1.1 Discussion—
Previously, the special unit of absorbed dose was the rad; however, the gray (Gy) has been adopted as the official SI unit (see
Practice E380).
21 2
1 Gy 5 1 J·kg 5 10 rad (2)
3.1.2 absorbed-dose rate—the absorbed dose per unit time interval.
3.1.3 annealing—thermal treatment of a TLD prior to irradiation or prior to readout.
3.1.3.1 Discussion—
Pre-irradiation annealing of TLDs is usually done to erase the effects of previous irradiation and to readjust the sensitivity of the
phosphor; pre-readout annealing usually is done to reduce low-temperature TLD response.
3.1.4 calibration conditions—the normal environmental conditions prevailing during routine calibration irradiations such as the
ambient temperature, humidity, and lighting.
3.1.5 equilibrium absorbed dose—the absorbed dose at some incremental volume within the material which the condition of
electron equilibrium (as many electrons of a given energy enter as leave the volume) exists (1) (see Appendix X1).
3.1.6 exposure, X—the quotient of dQ by dm, where dQ is the absolute value of the total charge of the ions of one sign produced
in air when all the electrons (negatrons and positrons) liberated by photons in a volume element of air having mass dm are
completely stopped in air.
dQ
X 5 (3)
dm
−1
Unit C · kg
3.1.6.1 Discussion—
Formerly the special unit of exposure was the roentgen (R).
24 21
1 R 5 2.58 310 C·kg ~exactly! (4)
3.1.7 primary electrons—for the case of electron irradiation, the electrons introduced into the device under test by the
irradiation source.
3.1.8 secondary-electron equilibrium—for the case of electron irradiation, the condition where as many secondary electrons of
a given energy enter a given volume as leave it.
The boldface numbers in parentheses refer to the list of references at the end of this practice.
E668 − 20
3.1.9 secondary-electron equilibrium absorbed dose—for the case of electron irradiation, the absorbed dose at some
incremental volume within the material in which the condition of secondary-electron equilibrium exists.
3.1.9.1 Discussion—
Additional definitions can be found in ICRU Report 33.
3.1.10 secondary electrons— for the case of electron irradiation, electrons knocked out of the electron shells of the material
being irradiated by the primary electron. For the case of photon irradiation, energetic electrons (photoelectrons, Auger electrons,
and Compton electrons) produced within the material being irradiated by the action of the incident photons.
3.1.10.1 Discussion—
Secondary electrons are produced by the interaction of the primary electrons with the atoms of the material being irradiated. This
interaction is a principal means of energy loss for the primary electrons. The kinetic energy of a secondary electron is typically
much lower than that of the primary electron which creates it.
3.1.11 test conditions—the normal environmental conditions prevailing during routine hardness-test irradiations such as the
ambient temperature, humidity, and lighting.
3.1.12 thermoluminescence dosimeter (TLD)—a TL phosphor, alone, or incorporated in a material, used for determining the
absorbed dose in materials. For example, the TL phosphor is sometimes incorporated in a TFE-fluorocarbon matrix.dosimeter made
of a material that stores energy when irradiated by ionizing radiation and then releases that energy in the form of visible light when
heated.
3.1.12.1 Discussion—
The TL phosphor can be used alone or incorporated into a material such as a TFE-fluorocarbon matrix.
3.1.13 thermoluminescence dosimeter (TLD) batch—a group of TLDs, generally originating from a single mix or lot of TL
phosphor, having similar TL responses and similar thermal and irradiation histories.
3.1.14 thermoluminescence dosimeter (TLD) reader—an instrument used to measure the light emitted from a TLD consisting
essentially of a heating element, a light-measuring device, and appropriate electronics.
3.1.15 thermoluminescence dosimeter (TLD) response—the measured light emitted by the TLD and read out during its heating
cycle consisting of one of the following: (a) the total light output over the entire heating cycle, (b) a part of that total light output,
or (c) the peak amplitude of the light output.
3.1.16 thermoluminescence (TL) phosphor—a material that stores, upon irradiation, a fraction of its absorbed dose in various
excited energy states. When thermally stimulated, the material emits this stored energy in the form of photons in the ultraviolet,
visible, and infrared regions.
3.1.17 TLD preparation—the procedure of cleaning, annealing, and encapsulating the TL phosphor prior to irradiation.
3.2 For units and terminology in reports of data, Terminology E170 may be used as a guide.
4. Significance and Use
4.1 Absorbed dose in a material is an important parameter that can be correlated with radiation effects produced in electronic
components and devices that are exposed to ionizing radiation. Reasonable estimates of this parameter can be calculated if
knowledge of the source radiation field (that is, energy spectrum and particle fluence) is available. Sufficiently detailed information
about the radiation field is generally not available. However, measurements of absorbed dose with passive dosimeters in a radiation
test facility can provide information from which the absorbed dose in a material of interest can be inferred. Under certain
prescribed conditions, TLDs are quite suitable for performing such measurements.
NOTE 2—For comprehensive discussions of various dosimetry methods applicable to the radiation types and energy and absorbed dose-rate range
discussed in this practice, see ICRU Reports 14, 17, 21, and 34.
5. Apparatus
5.1 The TLD System consists of the TLDs, the equipment used for preparation of the TLDs, and the TLD reader.
5.2 Calibration Facility delivers a known quantity of radiation to materials under certain prescribed environmental and
60 137
geometrical conditions. Its radiation source is usually a radioactive isotope, commonly either Co or Cs, whose radiation output
has been calibrated by specific techniques to some specified uncertainty (usually to within 65 %) and is traceable to national
standards.
E668 − 20
5.3 Storage Facility provides an environment for the TLDs before and after irradiation, that is light tight and that has a
negligible background absorbed-dose rate. A TLD stored in the facility for the longest expected storage period should absorb no
more than 1 % of the lowest absorbed dose expected to be measured in hardness-testing applications.
5.4 Environmental Chamber is used in testing the effects of temperature and humidity on TLD response. The chamber should
be capable of controlling the temperature and humidity within 65 % over the range expected under both calibration and test
conditions.
6. Handling and Readout Procedures
6.1 Bare TLDs should not be handled with the bare fingers; dirt or grease on their surfaces can affect their response and can
contaminate the heating chamber of the TLD reader. A vacuum pen or tweezers coated with PTFE should be used in handling. If
required, the TLDs can be cleaned by using the procedures in accordance with Appendix X2.
6.2 TLDs, especially those with high sensitivity, should be protected from light having an appreciable ultraviolet component,
such as sunlight or fluorescent light. Prolonged exposure to ultraviolet light, either before or after irradiation, can cause spurious
TLD response or enhanced post-irradiation fading. Incandescent lighting should be used for the TLD preparation and readout areas.
However, brief exposures of a few minutes to normal room fluorescent lighting is not likely to significantly affect the TLD response
except for low absorbed-dose measurements (<1 Gy or <100 rad) or measurements with high-sensitivity TLDs.
6.3 Preparation of the TLDs for irradiation consists of cleaning the TL phosphor (if required), annealing (if reusable TLDs are
employed), and encapsulating the TL phosphor. Reusable TLDs require careful treatment during annealing in order to obtain the
best results in dose measurements. The annealing procedure should include a reproducible temperature cycle of the annealing oven,
accurate timing of the annealing period, and a reproducible cooling rate.
6.4 For low absorbed-dose measurements (<1 Gy (100 rad)), dry nitrogen should be flowed through the heating chamber of the
TLD reader during readout. This suppresses the spurious TLD response that occurs in most forms of TLDs as a result of absorbed
oxygen on the phosphor surface. If the TLD reader uses hot gas to heat the TLDs, nitrogen should be used.
6.5 Calibration-irradiated TLDs and all subsequent test-irradiated TLDs from the same batch should be read out with the same
reader using the same readout techniques and reader parameters. The calibration is valid only for that batch used in that particular
reader. Readers that are different from the one used for calibration, including those of the same make and model, do not necessarily
indicate the same response for TLDs irradiated to the same absorbed dose.
6.6 TLDs may be used either as reusable or as single-use dosimeters. Single-use dosimeters are irradiated once, read out, and
then discarded; they are generally used as received from the manufacturer. Dosimeters that are reused are cycled repeatedly through
an anneal-irradiation-readout procedure.
6.7 The statistical methods specified in the following sections are optimal if the response of a batch of TLDs to a given radiation
dose is normally distributed. However, it has been demonstrated that TLD distributions can be severely skewed, so that the sample
mean may not be a suitable metric for small sample sizes(2). In this case TLDs should be fielded in groups of three, with either
the lowest reading or the two extremes discarded. Whatever procedure is adopted, it must be applied consistently for all calibrations
and routine measurements.
NOTE 3—Adequately determining the normality of a TLD distribution requires a large sample size.
7. Summary of Requirements for Performance Testing of a TLD System
7.1 The performance of a specific TLD system should be evaluated to determine its suitability for use in a specific
radiation-hardness test. Acceptable performance of the TLD system should be verified before applying the system in a particular
radiation-hardness-testing facility. Specific performance criteria are discussed in Section 8.
7.2 Performance tests should be repeated whenever a significant change is made in the TLD system or in the specific application.
Examples of such changes are: a change in the physical form or type of phosphor in the TLD, a change in any critical component
or in any adjustable readout factor of the TLD reader, or a change in the irradiation source characteristics.
7.3 A particular performance test may be omitted if widely accepted documentation exists in the scientific and technical
literature to show that the performance of the TLD system is satisfactory for that specific requirement. For example, if previously
accepted studies document that a particular TLD has no absorbed-dose-rate dependence for the expected conditions of irradiation,
then performance testing for absorbed-dose-rate dependence of that TLD system is unnecessary. All reports of test results should
include appropriate references that substantiate the performance of the system and thereby justify the omission of such
performance tests.
7.4 If a particular TLD system fails to meet the performance specification of any performance test, then use of that TLD system
is not recommended. Such a system may be used only if appropriate corrections to the TLD response can be determined sufficiently
well in order that the results of the specific radiation-hardness test can be determined within the required uncertainty.
E668 − 20
7.5 The number of TLDs, or the number of replicates of measurements with a single TLD, used for each test should be sufficient
to assure that the test results are significant at the 95 % confidence level. See Ref (3) for details of the procedures used to select
random samples and to determine the sample size required.
NOTE 4—If a sample of n measurements Y ,Y , . . ., Y is taken, the best estimate of the population mean, m, of a normal distribution is given by the
1 2 n
mean value, Y¯, of the sample:
n
¯
Y 5 Y (5)
( i
n
i51
2 2
The best estimate of the variance, σ , of the distribution is given by the variance, s , of the sample:
n
2 ¯
s 5 ~Y 2 Y ! (6)
( i
n 2 1
i51
The quantity σ ~5=s ! is called the standard deviation of the distribution. The degree to which s is a best estimate of σ depends on the sample size
and, as might be expected, s becomes a better estimate of σ as the sample size increases.
8. Specific Performance Tests and Correction Factors
8.1 Uniformity of TLD Response Within a Batch:
8.1.1 Select a random sample of 30 TLDs from a batch. Treating all of the sample TLDs in an identical manner, prepare them,
irradiate them in the calibration facility to the same absorbed-dose level, and read them out. Determine the variance, s , of the
~ = !
sample and estimate the standard deviation of the TLD response distribution σ5 s . the standard deviation, σ, should not exceed
8 % of the sample mean value, Y¯ ; that is, σ ≤ (0.08)Y¯ . The sample size specified (30) is the number of measurements required
0 0
to estimate the standard deviation, σ, of the TLD response distribution within 25 % of its true value at a 95 % confidence level
(see 2.4 of Ref 3).
8.1.2 For reusable TLDs that have been subjected to a number of anneal-irradiation cycles, the uniformity of the batch response
should be verified periodically by repeating the test in accordance with 8.1.1. The frequency required for the test depends on the
type of TLD and on its previous anneal-irradiation history. Retesting of the batch uniformity becomes particularly important for
2 4
TLDs irradiated to high-dose levels (>10 Gy (10 rad)). See, however, X2.2.2.
8.2 Reproducibility of TLD Response of Individual Reusable Dosimeters—Certain types of TLDs may be utilized as individual
reusable dosimeters. In this case, the identity of each individual dosimeter is maintained during repeated measurement cycles
throughout its useful life. This is in contrast to utilization in the batch mode where individual dosimeters within the batch are not
identified. To test the reproducibility of the response of an individual reusable dosimeter, the following procedures should be
followed:
8.2.1 Select the individual TLD to be tested, prepare it, irradiate it in the calibration facility to a specific absorbed-dose level
(for example, at the midpoint of the absorbed-dose range of interest), and read it out. In an identical manner, repeat this procedure
30 times. Determine the variance, s , of the responses and estimate the standard deviation of the TLD response distribution ~σ
5=s !. The standard deviation, σ, should not exceed 5 % of the mean response value, Y¯ , that is σ ≤ (0.05)Y¯ .
0 0
8.2.2 Some types of TLDs may exhibit a change in sensitivity (that is, response per unit absorbed dose) with repeated
anneal-irradiation-readout cycling. This effect is most pronounced if the TLD is not annealed thoroughly. The test results in
accordance with 8.2.1 may not show such a change in response sensitivity. However, if such a change is shown in that test or if
it appears after a larger number of cycles than specified in that test, then a different analysis of the data is required. In this case,
a curve should be fitted to the data of response versus number of cycles by a least-squares method. A measure of reproducibility
would then be given by the average standard deviation of the data points from the least-squares curve. The performance criterion
is the same as in 8.2.1.
8.2.3 Since the identity of each TLD is maintained when it is utilized as an individual dosimeter, it is not necessary that groups
of such individual TLDs meet the batch requirements in accordance with 8.1. However, for the other performance tests and
correction factors discussed in Section 8, it is assumed that such tests and factors are evaluated by utilizing TLDs in a batch mode.
8.3 Dependence of TLD Response on Absorbed-Dose Rate:
8.3.1 From a TLD batch meeting the requirements in accordance with 8.1.1, select a number of TLDs. Divide the TLDs into
x number of groups, each group containing n samples. Determine the absorbed-dose-rate range of interest for the intended
application and divide this range into x intervals (for example, one interval per decade). Prepare all the TLDs in an identical manner
and irradiate each group to the same dose level, but at a different absorbed-dose rate for each x group, covering the
absorbed-dose-rate range of interest. Read out the TLDs. Determine the mean response, Y¯ , for each x group of n samples.
i
Determine an overall mean value, Y¯ , for all x group means. Then the absolute difference between any group mean and the overall
mean should not exceed 20 % of the overall mean. That is,
¯ ¯ ¯
Y 2 Y # ~0.2!Y (7)
? i 0? 0
8.3.2 If |Y¯ − Y¯ | > (0.05)Y¯ , then appropriate correction factors to the TLD response as a function of absorbed-dose rate should
i 0 0
be determined by the procedures that follow.
E668 − 20
8.3.3 Determine the number of samples n required in each x group in order to detect a difference of δ = (0.05)Y¯ between a
group mean and the overall mean for a confidence level of 95 % and a probability of 0.50 of failing to detect such a difference.
It is assumed that the variance, σ , of the TLD response determined in accordance with 8.1.1, does not vary with the absorbed-dose
rate. Calculate the following parameter:
δ δ
d 5 5 (8)
=2σ σ=2
Then the sample size, n, is required for each x group to satisfy the above parameters is read off the graph of n versus d (see Fig.
X3.1).
8.3.4 Example of Sample Number Determination—If σ = 0.03 Y¯ (determined in 8.1.1),
¯
0.05 Y
d 5 5 1.18 (9)
¯
=2 0.03 Y
From Fig. X3.1, the sample size required is n = 4.4. The sample size should be 5, obtained by rounding up to the nearest integer.
NOTE 5—One method by which this test requirement can be carried out is by comparing the TLD responses with the response of another radiation
dosimeter whose absorbed-dose-rate dependence is known. A suitable type of dosimeter for use in most cases would be a calorimeter whose response
is absorbed-dose-rate independent and whose radiation absorption properties are similar to the TLD under test.
8.4 Dependence of TLD Response on Energy:
8.4.1 The radiation absorption properties of the TLDs employed in radiation-hardness testing should be similar to those of the
material in which the dose is to be estimated. Calculations can be made to determine the effects of a broad incident energy spectrum
on the response of the TLDs compared to that of the material of interest (usually silicon). The requirements of 7.5 are not applicable
to this section.
8.4.2 If the ratios [(μ /ρ) ]/[(μ /ρ) ] and [(S/ρ) ]/[(S/ρ) ] are equal to 1.0 within 610 % over a significant range of
en TLD en mat TLD mat
energy spectrum (for both calibration and test irradiations) incident upon both the TLD and the material of interest, then the
energy-response performance of the TLD system is acceptable. Here, μ /ρ is the mass photon energy absorption coefficient and
en
S/ρ is the mass collision electron stopping power. Tables of values of μ /ρ and S/ρ for several materials may be found in Appendix
en
X4. The phrase “significant range of the energy spectrum” means the minimum and maximum energy limits containing those
incident radiation particles (either photons or electrons) that contribute at least 90 % of the absorbed dose. In this case, detailed
energy spectral information is not required; the incident particle fluence (either photons or electrons) between the energy limits is
sufficient.
8.4.3 If the energy spectrum of the radiation incident upon the TLD (under both calibration and test conditions) and the material
of interest (under test conditions) is well known, then the conversion from absorbed dose in the TLD to absorbed dose in the
material of interest can be calculated from such data. If this conversion can be made to an uncertainty of 610 % or less, then the
performance of the TLD system is acceptable. In this case, the criteria concerning the ratios of μ /ρ and S/ρ in 8.4.2 need not be
en
met. (See Practice E666 for more specific guidelines.)
8.5 Dependence of TLD Response on Direction of Incident Radiation:
8.5.1 If the geometrical orientation of the TLD with respect to the radiation-hardness test field is significantly different than its
orientation with respect to the calibration radiation field, then any dependency of the TLD response on the direction of the incident
radiation should be determined. Select a number of TLDs from a batch meeting the requirements in accordance with 8.1.1. Divide
the TLDs into x number of groups, each group containing n samples. Prepare the TLDs in an identical manner, and irradiate each
group to the same absorbed-dose level in the following manner: (a) group g , in the usually oriented direction used for routine
calibration, and (b) groups g , g ,. . ., g oriented, respectively, at angles θ , θ , . . ., θ , relative to the usually oriented direction
1 2 x 1 2 x
with the center of the group at the same distance from the source. These angles should divide, in equal intervals of no more than
30° each, the angle between the normal and the maximum possible angle of incidence of the radiation-hardness test field. Read
out all the TLDs. Determine the mean response, Y¯ , for each x group of n samples. Then the absolute difference between the mean,
i
Y¯ , for the normally used calibration orientation and the mean for any other orientation should not exceed 5 % of the mean Y¯ .
0 0
Thus:
¯ ¯ ¯
Y 2 Y # 0.05 Y (10)
? ?
i 0 0
To determine the sample size n required for each x group, use the procedures in accordance with 8.3.3.
NOTE 6—This test applies only to a collimated-beam type source geometry. If the angle of incidence of the radiation from the source is nearly isotropic,
then it is recommended that the TLDs and their encapsulation material should be as nearly spherical as possible.
8.6 Dependence of TLD Response on Time Between Preparation and Irradiation:
8.6.1 A change in TLD sensitivity can occur during the storage period between preparation and irradiation. This may be a
significant effect if a wide range of storage periods is used. Use the following procedure to test for this effect. From a TLD batch
meeting the requirements in accordance with 8.1.1, select two equal groups of n samples each. Prepare the first group of TLDs
and place them in the storage facility for a time interval equal to the maximum time interval expected between preparation and
E668 − 20
irradiation during routine application in either calibration or hardness testing. At a later time, prepare the second group of TLDs,
and place them in the storage facility for the minimum time interval expected between preparation and irradiation. Time the
procedures so that the ends of the storage period for both groups occur simultaneously. Then irradiate both groups to the same
absorbed-dose level in the calibration facility and read them all out. The difference between the mean TLD response, Y¯ , of the
first group and the mean response, Y¯ , of the second group is a measure of the effect of storage time between preparation and
irradiation. This difference should not exceed 20 % of the average of the means of the two groups. Thus:
¯ ¯
Y 1Y
1 2
¯ ¯
Y 2 Y # 0.2 (11)
~ !
? ?
1 2
8.6.2 If the effect tested for in accordance with 8.6.1 exceeds 5 % of the average of the group means, then the functional
dependence of the TLD response on the storage period should be determined in order that appropriate correction factors may be
applied. This functional dependence may be determined by the procedures that follow.
8.6.3 The range of the elapsed time intervals between preparation and irradiation of interest is determined from the minimum
and maximum intervals utilized in 8.6.1. Tests should be performed at a minimum of two intervals per decade of elapsed time over
the entire range. For example, if the minimum elapsed time is 0.1 h and maximum elapsed time is 100 h, then an appropriate set
of tests would be at elapsed times of 0.1, 0.3, 1, 3, 10, 30, and 100 h. From a TLD batch meeting the requirements in accordance
with 8.1.1, select as many groups of n samples each as there are elapsed time intervals as determined above. Prepare a group of
TLDs, and place it in the storage facility for the appropriate preselected test-time interval. Repeat this procedure for all preselected
storage time intervals from the maximum to the minimum elapsed time. Arrange the storage times so that the ends of all procedures
occur simultaneously. Then irradiate all groups to the same dose level in the calibration facility and read them all out as quickly
as possible thereafter. This procedure is designed to minimize effects on dosimeter response caused by fading and variation in
reader output. Determine the mean response for each group of TLDs. A plot of mean TLD response versus elapsed time provides
a correction factor for a change in TLD sensitivity as a function of storage period. The number of samples n required for each group
of TLDs should be determined by the procedure in accordance with 8.3.3.
8.7 Dependence of TLD Response on Time Between Irradiation and Readout:
8.7.1 Significant fading of the TLD response may occur during the storage period between the end of irradiation and readout.
Use the following procedure to test for this effect. From a TLD batch meeting the requirements in accordance with 8.1.1, select
two equal groups of n samples each. Prepare the first group of TLDs, irradiate them in the calibration facility to a specific dose
level, then place them in the storage facility for an interval equal to the maximum time interval expected during routine application
(for either calibration or hardness testing) between the end of the irradiation period and readout. Prepare the second group of TLDs,
irradiate them in the calibration facility to the same dose level as the first group, then place them in the storage facility for an
interval equal to the minimum time interval expected between the end of irradiation and readout. Time the procedures so the ends
of the storage periods for both groups occur simultaneously. Read out all of the TLDs. The absolute difference between the mean
TLD response, Y¯ , of the first group and the mean response, Y¯ , of the second group is a measure of the effect of storage time
1 2
between the end of irradiation and readout. The difference should not exceed 20 % of the average of the means of the two groups.
Thus:
¯ ¯
Y 1Y
1 2
¯ ¯
Y 2 Y # 0.2 (12)
~ !
? ?
1 2
8.7.2 If the fading effect is greater than (0.05)(Y¯ + Y¯ )/2, then either a correction should be made to the TLD response or a
1 2
procedure used that eliminates the need for a correction. A procedure that achieves the latter would be one in which all TLDs are
read out at the same elapsed time after the end or irradiation. Such a procedure is often inconvenient or impractical. Therefore,
it is usually necessary to apply a fading correction to the TLD response. The fading characteristics of the TLD system may be
determined by the test procedures that follow.
8.7.3 Determine the minimum and maximum elapsed times between the end of the irradiation period and readout. Tests should
be performed at a minimum of two time intervals per decade of elapsed time over the entire period in accordance with 8.6.3. From
a TLD batch meeting the requirements in accordance with 8.1.1, select as many groups of n samples each as there are elapsed time
intervals as determined above. Each group of TLDs should undergo identical preparation and then should be irradiated in the
calibration facility to the same dose level. The groups of TLDs are placed in the storage facility for all preselected appropriate time
intervals from the maximum to the minimum elapsed time. Arrange the time of irradiations for all the groups so that the ends of
their storage periods occur simultaneously. Read out all the TLDs. Determine the mean response for each group of TLDs. A plot
of mean TLD response versus elapsed time provides the fading correction factor. The number of samples n required for each group
of TLDs should be determined by the procedures in accordance with 8.3.3.
8.8 Dependence of TLD Response on Temperature During Storage or Irradiation:
8.8.1 If the storage temperature experienced by the TLDs between preparation and irradiation during routine radiation-hardness
testing differs from the temperature during routine calibration by more than 10°C, the test in accordance with 8.6 should be
repeated over the range of temperatures expected using the environmental chamber instead of the storage facility. The performance
criteria in accordance with 8.6 are applicable to this section.
E668 − 20
8.8.2 If the storage temperature experienced by the TLDs between irradiation and readout during routine radiation-hardness
testing differs from the temperature during routine calibration by more than 10°C, the test in accordance with 8.7 should be
repeated over the range of temperatures expected using the environmental chamber instead of the storage facility. The performance
criteria in accordance with 8.7 are applicable to this section.
8.8.3 If the temperature experienced by the TLDs during the irradiation period during routine radiation-hardness testing differs
from the temperature during routine calibration by more than 10°C, then the effect on TLD response should be determined by the
following procedure: Select a number of TLDs from a batch meeting the requirements in accordance with 8.1.1, prepare them in
an identical manner, and separate them into two equal groups of n samples each. Irradiate the first group in the calibration facility
to a specific dose level, maintaining the temperature of the TLDs at the minimum temperature expected during routine
hardness-test irradiation. Irradiate the second group in the calibration facility to the same dose level, maintaining the temperature
of the TLDs at the maximum temperature expected during routine hardness-test irradiations. Readout all of the TLDs. The
difference between the mean TLD response, Y¯ , of the first group and the mean response, Y¯ , of the second group is a measure
1 2
of the effect of temperature variation during irradiation. This difference should not exceed 20 % of the average of the means of
the two groups. If the magnitude of the effect is greater than 5 % of the average of the means, then appropriate corrections to the
TLD responses should be determined by procedures analogous to those in accordance with 8.6.
8.9 Dependence of TLD Response on Humidity—In general, the responses of the most widely used TLDs have not been shown
to be sensitive to changes in relative humidity. However, if a TLD that is hygroscopic is being considered for application in
radiation-hardness testing, then the performance tests in accordance with 8.8 should be repeated with the humidity as the variable
parameter and the temperature maintained at the maximum value used in the temperature tests.
NOTE 7—Once a TLD system of a particular TL-phosphor type and physical configuration has met the performance requirements of Section 8, new
batches of the same type need only be tested for the requirements of 8.1 (batch uniformity) and 8.7 (post-irradiation fading). See also 7.2 and 7.3.
9. Calibration of the TLD System
9.1 Calibrate the TLD system in a manner such that the TLD response can be related directly to the absorbed dose in the TLD
60 137
phosphor. Use a suitable, well-characterized radiation source in the calibration. Radioactive isotope sources such as Co or Cs
are generally used for this purpose. Exposure rates (or absorbed-dose rates) from such sources should be known to be better than
65 % at all locations normally used for calibration irradiations. The methods used for determining the output rates from such
calibration sources include the use of secondary standard radiation measuring instruments, such as air-ionization chambers or
transfer dosimeters, whose calibration is traceable to the National Institute of Standards and Technology (NIST) or other
recognized calibration laboratory. Other types of dosimeters whose responses are absolute (require no calibration), such as
calorimeters, may also be employed for calibration of such sources.
9.2 The response of most types of TLDs generally is not linear as a function of absorbed dose (4). The response of a typical
TLD is nearly linear from low-absorbed-dose levels to approximately 10 Gy(TLD) (10 rad (TLD)), then becomes supralinear up
2 3 4 5
to approximately 10 to 10 Gy(TLD) (10 to 10 rad (TLD)) where saturation effects become evident. Exercise care in the use
3 5
of the TLD system for absorbed-dose levels of approximately 10 Gy(TLD) (10 rad(TLD)) or higher to ensure that the change
in the system response per unit absorbed dose is adequate in order that the absorbed dose can be determined within the required
uncertainty.
9.3 The absorbed-dose range of calibration should cover the maximum absorbed-dose range of interest for the intended
application. Measure a minimum of three absorbed-dose levels per decade of absorbed dose covered. Since the TLD response
versus absorbed dose for most types of TLDs generally is not linear, make a sufficient number of measurements at each
absorbed-dose level to define accurately the shape of the characteristic response curve. The number of TLD samples required to
determine the mean response at each absorbed-dose level is given by the following procedures:
9.3.1 In order to determine the mean TLD response, Y¯ , within 65 % at a 95 % confidence level, the number of TLD samples
required for a given absorbed-dose level is as follows:
2 2
2.045 s
~ !
n 5 (13)
¯
~0.05 Y !
where s is the estimate of the standard deviation σ of the TLD response distribution as determined by the procedures in
accordance with 8.1.1. For example , if s = (0.06) Y¯ , then:
2 ¯
2.045 ~0.06 Y !
~ !
n 5 5 6.0 (14)
¯
~ !
0.05 Y
(See Section 2.3.2 of Ref (3) for more details.)
9.3.2 The procedures described in 9.3.1 assume that the standard deviation of the TLD response distribution is constant for all
absorbed-dose levels measured. This assumption generally is valid over most of the usable absorbed-dose range for most TLDs
3 5
but may not be correct for very high-absorbed-dose levels of approximately 10 Gy(TLD) (10 rad(TLD)) or higher. If the TLD
E668 − 20
system is used at these absorbed-dose levels, then redetermine the standard deviation of the response distribution at these levels
by repeating the procedures in accordance with 8.1.1.
9.4 During a calibration irradiation, encapsulate the TL phosphor in a material with a thickness just sufficient to product electron
equilibrium in the phosphor (see Appendix X1). If possible, the encapsulation material should have the same thickness on all sides
of the dosimeter.
NOTE 8—The encapsulation material should resemble the phosphor material as closely as possible with respect to radiation absorption properties. For
example, if the TL phosphor is CaF , acceptable encapsulation material would be calcium fluoride, 1000-series aluminum, or silicon (see Appendix X2).
If the calibration source is Co, then a thickness of 2.2 mm of aluminum (equal to the practical range of the highest-energy secondary electrons produced)
could establish electron equilibrium in the CaF phosphor. This thickness is sufficient to stop secondary electrons that might be generated by the source
photons in material other than the encapsulation material.
9.5 Correct for attenuation of the photons from the source by the layer of material used to establish electron equilibrium, using
the following formula:
μ
en
X 5 X exp 2 ρx (15)
F S D G
ρ
where:
X = unattenuated exposure at the position of the TLD phosphor, roentgens,
X = unattenuated exposure, roentgens,
μ /ρ = mass energy absorption coefficient of the encapsulation material for the effective source photon energy, cm /g,
en
ρ = density of the encapsulation material, g/cm , and
x = thickness of the encapsulation material, cm.
Values of μ /ρ may be found in Appendix X4 and Ref (5).
en
NOTE 9—The attenuation formula given is not rigorously correct for a broad-beam geometry as it does not include a buildup factor. Buildup factors
generally are not available for a wide range of energies, materials, and geometries. However, the formula gives the results that are in reasonable agreement
with more rigorous treatments for materials of low to medium atomic number of relatively thin sections over the range of photon energies that are
applicable to this practice
9.6 Once the exposure has been determined, the absorbed dose (grays) to the encapsulated TL phosphor is found from the
formula:
~μ /ρ!
en TLD
D 5 0.873 310 X (16)
~ !
TLD
μ /ρ
~ !
en
air
μ /ρ
~ !
en
TLD
D 5 ~0.877 310 !X (16)
TLD
~μ /ρ!
en air
−2
The factor (0.873(0.877 × 10 ) is used to convert exposure (roentgens) to absorbed dose in air (grays). The subscripts refer to
the material of interest. As in 9.5, the μ /ρ values are evaluated at the effective calibration source photon energy. This formula
en
is valid only if electron equilibrium exists in the TL phosphor. It is assumed that the incident photon fluence is essentially
monoenergetic. If this is not the case, then average all of the energy-dependent energy absorption coefficients of 9.5 and 9.6 over
the appropriate energy spectrum.
−2
NOTE 10—The value of 0.8730.877 × 10 Gy(air)/roentgen is based on an average energy of 33.8533.97 6 0.150.20 eV required to produce an ion
pair in dry air (see ICRU Report 31)).90).
9.7 The absorbed dose calibration results of the procedures of 9.3 – 9.6 are valid only for a given batch of TLDs. A different
batch generally will have a different radiation sensitivity. However, this difference is usually a constant factor over the entire
absorbed-dose range. Therefore, it may not be necessary to generate a new calibration c
...