See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/336852008

Effect of Teflon Transmittance on Sensitivity of Thermoluminescence

Dosimeter Cards

Article in Health Physics · October 2019

DOI: 10.1097/HP.0000000000001149

CITATION READS

1 315

4 authors, including:

Alex Romanyukha
Navy
179 PUBLICATIONS 3,166 CITATIONS

SEE PROFILE

All content following this page was uploaded by Alex Romanyukha on 02 January 2020.

The user has requested enhancement of the downloaded file.

 Paper
Downloaded from https://journals.lww.com/health-physics by Ujb9cAENh456icZPuIBIhgqSEajluOB62t08RJSeLOrGrJpQa5UiUG9U/DNZDJoD+uKVClQlPKRDNYbONZNi018XtwE939n/WDK85WxpHzZci25oWCLnOSKnNpgGRzn3Pi9vo1RgNXncFZlOxLl/NGjZlfqZVNDmNlDSIW/FiNWlyOKmXq2WzU3v+RW8IHokNn8MxRoAc/0= on 11/12/2019

EFFECT OF TEFLON TRANSMITTANCE ON SENSITIVITY OF

THERMOLUMINESCENCE DOSIMETER CARDS

V.B. Podobedov,1 A. Romanyukha,2 C.C. Miller,1 and A. Hoy2

dosimeters can measure lower doses. But significant varia-

Abstract—Thermoluminescence dosimeter cards purchased by the
US Navy in recent years have different radiation sensitivities, e.g., tion of the radiation sensitivity in the population of dosime-
they exhibit a different amount of light per dose unit. Presented tests ters could mean that there are significant variations of other
indicate that the optical transparency of the Teflon encapsulation is properties affecting stability of calibration, dose fading, and
partially responsible for the significant variation of the DT-702/PD uncertainty. In the worst-case scenario, dosimeters with dif-
radiation sensitivity. It was confirmed also that the Teflon transpar-
ency is in fact a primary cause of the radiation sensitivity increase ferent sensitivity may even require different reading condi-
in the most recently produced dosimetric cards. This conclusion is tions or so-called time-temperature profiles for optimal
based on the correlation found between the calibrated radiation sen- readout and calibration. That is why the Naval Dosimetry
sitivity of the dosimeter card element and the optical transparency of Center (NDC) has established acceptable limits for TLD
its Teflon encapsulation. The transparency measurements were per-
formed at the wavelength of 400 nm within a 10 nm spectral interval radiation sensitivity. These acceptable limits of radiation
effectively covering the spectral range of the thermoluminescence. It sensitivity are verified during acceptance testing of new
is anticipated that the experimentally determined correlation will dosimeters following the requirements established by
help to approve the acceptance of new thermoluminescence dosime- the International Organization for Standardization (ISO)/
ter cards in the Naval Dosimetry Center inventory as well as improve
the production process. International Electrotechnical Commission (IEC) document
Health Phys. 00(00):000–000; 2019 standard ISO/IEC 17025 (2005).
Key words: calibration; dosimetry, personnel; dosimetry, exter- In recent years, dosimeters have exhibited quite differ-
nal; dosimetry, thermoluminescence ent radiation sensitivities although they seem to be produced
in the same way and from the same components. Variations
occur typically because of an uncontrolled change in the
materials used to produce them. It was noticed in 2015 that
INTRODUCTION newly purchased dosimeters have a significantly higher ra-
diation sensitivity (>20%) than those that were purchased
SINCE 1998, the US Navy monitors personnel radiation earlier. The radiation sensitivity of a TLD is defined as do-
doses with the Harshaw model 8840/8841 thermolumines- simeter light output per dose unit. It depends on many fac-
cent dosimeters (TLD), which are read with Harshaw Model tors, including the dosimeter’s usage. In 2011, Voss et al.
8800PC readers (Cassata et al. 2002; St. John et al. 2006; (2011) investigated the effect of multiple read-irradiation
Romanyukha et al. 2012). The Harshaw 8841 dosimetric cycles on TLD card sensitivity. According to this research,
card has four lithium fluoride (LiF:Mg,Cu,P) elements 1,000 heat-irradiation cycles causes about 10% sensitivity
which are Teflon encapsulated on an aluminum substrate. loss for card elements 1, 2, and 4 if reader is functioning
The radiation sensitivity of TLDs is one of the most impor- properly. A typical NDC dosimeter issue period is 90 d.
tant parameters of dosimeters. Beneficially, more sensitive When accounting for shipping and processing time this
means that on average TLD cards undergo 3–5 heatings
1
National Institute of Standards and Technology, Gaithersburg, MD; per year. Thus, for a 20-y-old card the number of heatings
2
Naval Dosimetry Center, Bethesda, MD. will be below 100. According to Voss et al. (2011), 100
The authors declare no conflicts of interest.
For correspondence contact V.B. Podobedov, National Institute of heatings would cause about 1% of sensitivity loss but not
Standards and Technology, 100 Bureau Drive, Gaithersburg, MD 20899, the 20% difference that was seen for new and old TLD
or email at [email protected]. cards. Furthermore, according to NDC internal policy, ev-
(Manuscript accepted 25 June 2019)
0017-9078/19/0 ery 2 y every TLD card is the subject of new calibration.
Copyright © 2019 Health Physics Society All historical calibration data are available for review, and
DOI: 10.1097/HP.0000000000001149 they do not indicate a significant change of sensitivity for
www.health-physics.com 1

Copyright © 2019 Health Physics Society. Unauthorized reproduction of this article is prohibited.
2 Health Physics Month 2019, Volume 00, Number 00

the cards that are used for personnel monitoring. TLD light and calibrated according to the NDC-established proce-
output depends not only on the sensitivity of the TLD ele- dures. As a result of the calibration, the so-called element
ments but also on the transparency of the Teflon used for correction coefficient or ecc was assigned to each of four
their encapsulation. The TLD manufacturer stated that the TLD elements:
Teflon fabrication process was changed because of a termi-
nation of some chemicals and reagents previously used in ecc ¼ ðRCF  DÞ=L; ð1Þ
the production process. The purpose of this work was to in-
vestigate if a change in Teflon could be responsible for in- where RCF (nC mrem−1) is a reader calibration factor, D is
creased sensitivity of TLDs. The basic techniques involve the dose delivered to the element during irradiation (1 mSv =
the comparison of Teflon transparency in the TLD cards cal- 100 mrem), and L (nC) is the response of the element which
ibrated to the same dose. is given by the light output and measured with the TLD
reader (Romanyukha et al. 2018). Note that the delivered
MATERIALS AND METHODS dose is shown in the numerator of eqn (1), and the corre-
sponding TLD light output is in the denominator. Therefore,
The TLD card has four elements (Fig. 1). The TLD TLD cards with higher sensitivities will have smaller ecc
element is a disc made of powdered LiF crystal mixed values. To measure the transparency of the Teflon encapsu-
with required dopants. The disc is produced by a hot pres- lation film, an experimental setup (presented in Fig. 2) was
sure pelleting process. In this study, 51 standard TLD used. The light from an optical radiation source was col-
cards were used. Of these, 26 cards were produced before lected by an optical system and directed to a replaceable
2005, and another 25 cards were produced in 2015–2016 af- aperture. It was placed in front of the TLD card opening
ter the stated change of the Teflon (Chemours, Wilmington, so that only the Teflon-film-covered area was illuminated.
Delaware, US) fabrication. All cards were properly inspected Three apertures were used in this study: the first one had a
collimator hole with a diameter of 1 mm (#1 in Fig. 2), the
second aperture also had a collimator hole but with a diam-
eter of 2 mm; while the third one was a ring-type (#2 in
Fig. 2). The first and third apertures allowed illumination
of only the Teflon while excluding the TLD element
whereas use of the second aperture allows illumination
of the encapsulated TLD element and remaining Teflon.
Similar information was obtained using the first and third
apertures, but the ring-type aperture provides a stronger
signal due to illumination of a larger surface area and more
effective light collection. This was important for higher
accuracy of the transparency measurements because the
source output drops significantly in the range below
400 nm (Fig. 3).
The amount of detected light that passes through the
card depends not only on material transmittance but also
on broad-angle light scattering from both Teflon and the
LiF:Mg,Cu,P materials. Therefore, measured light output
varies with distance L* between TLD card and spectrome-
ter. The accurate measurements of total transmittance be-
come more complicated. In this study, the measurements
were focused on the optical transparency T* defined as the
ratio of detected signal (output) to the reference level (input)
at the fixed distance L* as described below. The reference
level can be established as 100% of the light supplied to
the TLD element area. For this purpose, one element in
the designated TLD card is totally removed (Fig. 1), and
thus the light reference level is established. All other TLD
card elements are measured intact in a highly reproducible
Fig. 1. TLD card. The upper left TLD element was entirely removed optical geometry. The determined reference level is used
for the reference measurements. for normalization of the light output of all intact TLD cards
www.health-physics.com

Copyright © 2019 Health Physics Society. Unauthorized reproduction of this article is prohibited.
 Effect of Teflon transmittance on TLD cards c V.B. PODOBEDOV ET AL. 3

Fig. 2. Experimental setup for transparency measurements.

used in this study because of the high reproducibility of the chosen experimental setup. Therefore, most of the data
optic measurements. The reproducibility was achieved by were collected in the 10 nm spectral interval at 400 nm
use of a specially designed holder shown in Fig. 2. Posi- where the light source output was stronger.
tioning uncertainty was estimated as 0.1 mm based on
the actual dimensions of the holder and TLD card. This RESULTS AND DISCUSSION
uncertainty has smaller effect for the L direction because
L >> 0.1 mm. Depending on the aperture used, the trans- Typical results of the transparency T* measurements
parency of only the Teflon film (excluding the TLD ele- are shown in Fig. 4. As expected, the transparency of Teflon
ment) or the area including the TLD element covered by only (solid line) is higher because any TLD element has
Teflon were measured. both Teflon and LiF:Mg,Cu,P material contributing to the
Optical measurements were performed by a multi- loss of optical signal. Quantitative analysis of 204 spectra
channel spectrometer equipped with a holographic grating from 51 TLD cards was performed with a specially de-
and a charge-coupled device (CCD). Simultaneously covered signed computer code. It first selects the transparency data
spectral range was available between 200 nm and 850 nm. in the range between 395 nm and 405 nm and then averages
The maximum of the thermoluminescence emission of them over 20 pixels of the charge-coupled device and pre-
the LiF:Mg,Cu,P is about 350 nm (McKeever et al. sents results as a table of T* values vs. the element position
1995; Yang et al. 2005). Ideally, the transparency must and the card serial number. As estimated from multiple
be measured in the whole spectral range of the thermolu- measurements of individual TLD elements, the expanded
minescence light. However, the source that was used has uncertainty of the measured transparency T* of Teflon film
very low intensity at 350 nm (Fig. 3). The comparison was about 2% (k = 2).
of the different sets of data obtained between 350 nm The correlation between Teflon transparency T* and ecc
and 450 nm showed that the range in the vicinity of of TLD element is shown in Fig. 5. The sequence of the mea-
400 nm provides nearly the same qualitative information surements was designed so that a smaller TLD element
as a whole range between 350 nm and 450 nm but allows
improvement of the signal-to-noise (S/N) ratio for the

Fig. 4. Typical spectral dependence of transparency T* for Teflon en-

Fig. 3. Spectrum of the source radiation. capsulation film (solid line) and for a whole TLD element (dashed line).
www.health-physics.com

Copyright © 2019 Health Physics Society. Unauthorized reproduction of this article is prohibited.
4 Health Physics Month 2019, Volume 00, Number 00

the LiF:Mg,Cu,P material and Teflon film. These transpar-

ency measurements were done using the same experimental
setup except the aperture size; e.g., the diameter of circular
aperture was increased to about 2 mm to obtain better
signal-to-noise ratio. Because the aperture diameter is lim-
ited to the central part of the TLD element, the transparency
measured in this configuration accounts for the light loss in
both Teflon and LiF:Mg,Cu,P material. Significant reduc-
tion of the spectral transparency for the whole TLD element
as compared to the transparency of Teflon film (as shown in
Fig. 4) supports this assumption. The results of transpar-
ency measurements of TLD elements are presented in
Fig. 6. Upon weighted fit and normalization of the TLD el-
ement transparency data in the same way as above, the cor-
Fig. 5. Comparison of calibration coefficient ecc (open circles, relation of ecc and transparency T* was also observed.
dashed line) and Teflon transparency T* (solid squares and solid line). Indeed, the normalized transparency values of 204 TLD el-
Presented lines use the weighted (10%) fit of the data.
ements presented in Fig. 6 (bottom curve) indicate similar
but even stronger correlation with the ecc as compared to
number on the x axis of Fig. 5 indicates earlier production of one found earlier for Teflon film alone.
the TLD card. For convenience, both sets of data were nor- Assuming that TLD element transparency is a product of
malized so that the average data value for the most recently those of Teflon and LiF:Mg,Cu,P material, it is possible to re-
produced cards would be unity. The ecc values measured trieve the contribution of the LiF:Mg,Cu,P material only. Pre-
for 204 calibrated TLD elements presented in Fig. 5 as open sumably its transparency is affected by both scattering and
circles indicate significant differences depending on the year absorption of thermoluminescence emission inside the mate-
of the production. The TLD elements produced before 2005 rial. The transparency was evaluated by dividing the TLD ele-
have significantly higher ecc (lower sensitivity) values than ment transparency T* (solid line in Fig. 6) by that of Teflon
TLD elements produced in 2015–2016. The maximum dif- only. From these data, it was found that the LiF:Mg,Cu,P
ference found between any one ecc of old and new TLD card material also demonstrates the dependence on the year of el-
elements was a factor of 2. Corresponding transparency ement production. The data presented in Fig. 7 indicate up
measurements of Teflon film (closed squares in Fig. 5) also to 15% additional reduction of transparency in TLD cards
demonstrate similar but opposite dependence; e.g., the produced before 2005.
transparency of the Teflon film in the most recently pro- Presented optical measurements of both Teflon and
duced TLDs is higher. Significant scatter of the experimen- LiF:Mg,Cu,P elements indicate some correlation of ecc
tal data in Fig. 5 observed for both ecc and transparency T* with transparency of the TLD element. The relative
is due to variation of the TLD element property and limited
accuracy of the measurements.
It is important, however, to notice that the data scatter
observed for transparency measurements is a few times
lower than that for ecc. So, the accuracy of transparency
measurements is sufficient for comparison to ecc. The in-
crease of ecc values and the opposite drop of transparency
for the individual TLD elements are shown in the left part
of the Fig. 5 displaying the aged TLD cards. A more obvi-
ous conclusion becomes available from the weighted fit of
the data sets shown in Fig. 5. This fit is presented by dashed
and solid lines related to ecc and Teflon transparency, re-
spectively. The observed trend of the curves including the
ecc maxima (the lowest sensitivity) and their coincidence
with transparency minimums clearly indicates the correla-
tion of Teflon transparency with ecc values.
Fig. 6. Comparison of calibration coefficient ecc (open circles,
Similar variability of transparency was found when the dashed line), transparency T* for Teflon film (solid squares and solid
TLD element was illuminated in its central area; i.e., the line), and a whole TLD element (triangles and dotted line). All lines
data obtained represented the optical properties of both present the weighted (10%) fit of the data.
www.health-physics.com

Copyright © 2019 Health Physics Society. Unauthorized reproduction of this article is prohibited.
 Effect of Teflon transmittance on TLD cards c V.B. PODOBEDOV ET AL. 5

present method provides about 2% (k = 2) uncertainty for

transparency value measured at the 400 nm wavelength.
Based on the results obtained for 204 calibrated TLD
elements, the direct correlation of radiation sensitivity of
TLD cards and their Teflon encapsulation transparency
has been established. It was found that Teflon transparency
is a primary cause of the radiation sensitivity increase in the
most recently produced dosimetric cards. Practically this
means that transparency of Teflon encapsulation holding
the LiF:Mg,Cu,P element in the TLD card may contribute
to the value of ecc of the TLD element calibrated to the
same radiation dose. Variations of both the ecc and optical
transparency are stronger in the aged TLD cards. Addition-
ally, the transparency of the LiF:Mg,Cu,P material itself
may affect the ecc although the appearance is different
from the related features in Teflon. This effect is also stron-
Fig. 7. Relative transparency of the LiF:Mg,Cu,P element.
ger in the aged TLD cards. As follows from this study that
a Teflon transparency control, if established at the produc-
variation of the Teflon transparency was found to be a few tion site, would significantly improve the ecc uniformity
times higher as compared to that of the LiF:Mg,Cu,P ele- for TLD cards.
ment. Thus, the improved transparency of Teflon encapsula-
tion material can be considered a primary reason for Acknowledgments—Lieutenant Andrew Hoy is a military service member. This
work was prepared as part of his official duties. Title 17, USC, §105 provides
increasing sensitivity of the most recent TLD cards. The that copyright protection under this title is not available for any work of the
evaluation of the physical properties of Teflon responsible US government. Title 17, USC., §101 defines a US government work as a work
for its transparency is beyond the scope of this paper and prepared by a military service member or employee of the US government as
part of that person’s official duties.
may be found elsewhere (White et al. 2016; Lin et al. The views expressed in this article are those of the author and do not nec-
2012; Tsai et al. 2008; Galante et al. 2010). There is an in- essarily reflect the official policy or position of the Department of the Navy, De-
dication that light scattering in Teflon film significantly partment of Defense, nor the US government.
Disclaimer: certain commercial equipment, instruments, or materials are
contributes to transparency of the TLD element. Related identified in this paper to foster understanding. Such identification does not im-
measurements have established the reduction of transpar- ply recommendation or endorsement by the National Institute of Standards and
Technology, nor does it imply that the equipment is necessarily the best avail-
ency with distance L* to spectrometer input (Fig. 2). Be- able for the purpose.
cause in all the transparency measurements the source
beam was collimated, the dependence found suggests the REFERENCES
presence of wide-angle radiation, presumably diffuse
Cassata JR, Moscovitch M, Rotunda JE, Velbeck KJ. A new par-
light-scattering coming from Teflon film. Light scattering adigm in personal dosimetry using LiF:Mg,Cu,P. Radiat Pro-
observed in Teflon film is contributing to TLD card trans- tect Dosim 101:27–42; 2002.
parency and its TL sensitivity. The presence of diffuse scat- Galante AMS, Galante OL, Campos LL. Study on application of
tering in Teflon encapsulation can be observed even PTFE, FEP and PFA fluoropolymers on radiation dosimetry.
Nucl Instrum Methods Phys Res A 619:177–180; 2010.
visually; e.g., Teflon film looks noticeably milky in the International Organization for Standardization/International
cards produced before 2005 compared to the cards pro- Electrotechnical Commission. General requirements for the com-
duced in 2015–2016. petence of testing and calibration laboratories. Geneva: ISO; ISO/
IEC 17025:2005(E); 2005.
Lin GR, Meng FS, Pai YH, Lin YH. Enhanced conversion efficiency
and surface hydrophobicity of nano-roughened Teflon-like film
CONCLUSION coated poly-crystalline Si solar cells. Phys Chem Chem Phys
14:3968–3973; 2012.
We developed a method for optical transparency mea- McKeever SWS, Moscovitch M, Townsend PD. Thermolumines-
cence dosimetry materials: properties and uses. Ashford, Great
surements of Teflon-encapsulated TLD elements. The tech- Britain: Nuclear Technology Publications; 1995.
nique allows light output-input ratio measurements for Romanyukha A, King DL, Kennemur LK. Impact of Fukushima
specified areas of the TLD cards. The light illumination nuclear accident on background radiation doses measured by
was done by a preconfigured beam. The transparency mea- control dosimeters in Japan. Health Phys 102:535–541; 2012.
Romanyukha A, Grypp MD, Trompier F, Thompson AK, Minniti
surements were available in the spectral range from 200 nm R, Debroas J, Williams AS. Performance evolution of TLD-
to 850 nm which effectively covers the entire spectrum 700H/600H dosimetry system at extended issue periods.
of the LiF:Mg,Cu,P thermoluminescence emission. The Radiat Meas 108:45–51; 2018.
www.health-physics.com

Copyright © 2019 Health Physics Society. Unauthorized reproduction of this article is prohibited.
6 Health Physics Month 2019, Volume 00, Number 00

St. John TJ, Cassata J, Blake PK, Wallace WH, Minniti R. Techni- White WH, Trzepla K, Hyslop NP, Schichtel BA. A critical review
cal aspects of the Naval Dosimetry Center quality assurance of filter transmittance measurements for aerosol light absorp-
program. Radiat Protect Dosim 120:273–277; 2006. tion, and de novo calibration for a decade of monitoring on
Tsai BK, Allen DW, Hanssen LM, Wilthan B, Zeng J. A compar- PTFE membranes. Aerosol Sci Technol 50:984–1002; 2016.
ison of optical properties between high density and low-density Yang B, Lu Q, Wang S, Townsend PD. Studies on the thermolumines-
sintered PTFE. In: Gu Z-H, Hanssen LM, eds. Reflection, scat- cence spectra and thermal stability of LiF:Mg,Cu, LiF:Cu,Na and
tering, and diffraction from surfaces; proceedings of SPIE, vol. LiF:Mg,Cu,Si. Nucl. Instr Meth Phys Res B 239:171–178; 2005.
7065. Bellingham, WA: SPIE; 2008: 70650Y.
Voss SP, Sucheta A, Romanyukha A, Moscovitch M, Kennemur
LK, Benevides LA. Effect of TLD-700H (LiF:Mg,Cu,P) sensi-
tivity loss at multiple read-irradiation cycles on TLD reader
calibration. Radiat Meas 46:1590–1594; 2011. ■■

www.health-physics.com

View publication stats Copyright © 2019 Health Physics Society. Unauthorized reproduction of this article is prohibited.