Applied Radiation and Isotopes 82 (2013) 126129

Contents lists available at ScienceDirect

Applied Radiation and Isotopes

journal homepage: www.elsevier.com/locate/apradiso

A search for novel thermoluminescent radiation dosimeter media

Khalid H. Al-Hinai a,n, Nadjima Benkara Mohd a, Nurul Rozullyah Zulkepely a,
Roslan Md. Nor a, Yusoff Mohd. Amin a, D.A. Bradley a,b
a
b

Department of Physics, University of Malaya, 50603 Kuala Lumpur, Malaysia

Department of Physics, University of Surrey, Guildford GU2 7XH, United Kingdom

H I G H L I G H T S






ZnS:Mn nanophosphors and CVD diamond lms as new TL dosimeter media.

The TL response of both increases linearly with radiation dose.
For ZnS:Mn doped to 2 moles of Mn is found to be the most sensitive.
For CVD diamond lms there is a need to adjust parameters such as nanoparticles size.

art ic l e i nf o

a b s t r a c t

Article history:
Received 2 July 2013
Received in revised form
12 July 2013
Accepted 15 July 2013
Available online 2 August 2013

We describe two example pilot efforts to help dene new thermoluminescent dosimeter media. The rst
concerns ZnS:Mn nanophosphors, prepared by chemical precipitation using zinc and sodium sulfate,
doped with manganese sulfate at concentrations varying from 1 to 3 mol. The second concerns chemical
vapor deposited diamond, produced as a thin lm or as amorphous carbon on a single-crystal silicon
substrate, each deposited under the same conditions, use being made of the hot lament-chemical vapor
deposition (HFCVD) technique. The gas concentrations used were 1% CH4 in 99% H2 and 25% CH4 in
75% H2. Characterization of formations used FESEM, XRD and EDX. The nanophosphors consisted of
particles of sizes in the range 85150 nm, the thermoluminescence (TL)-based radiation detection
medium giving rise to a single peaked glow curve of maximum yield at a temperature of 250 1C at a
heating rate of 5 1C/s. The TL response increased linearly with radiation dose, ZnS doped to 2 mol of Mn
being found the most sensitive. Regarding chemical vapor deposited (CVD) carbon, inappreciable TL was
found for the resultant ball-like amorphous carbon lms, graphite, and the silicon substrate, whereas
CVD diamond lms showed a promising degree of linearity with dose. For both the ZnS and diamond
samples, TL signal fading was appreciable, being some 40% per day for ZnS and450% per day for CVD
lms even under storage in the dark at room temperature, making it apparent that there is need to adjust
parameters such as the size of nanoparticles.
& 2013 Elsevier Ltd. All rights reserved.

Keywords:
ZnS:Mn nanophosphors
Diamond
Radiation detectors
Thermoluminescence dosimetry

1. Introduction
Common TL dosimeter (TLD) materials such as LiF:Mg,Ti (TLD100) are most often applied in measuring low or high doses. One
motivation is that most TLD materials suffer from considerable
supralinearity at high doses, reducing the choice of TLD materials
for dosimetry applications such as in megavoltage radiotherapy.
As such, we are currently seeking materials that offer wellbehaved response to allow measurement of radiation doses from
moderate doses of a few Gy up to 100 Gy.

Corresponding author.
E-mail address: yusoffmohdamin@um.edu.my (K.H. Al-Hinai).

0969-8043/$ - see front matter & 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.apradiso.2013.07.013

As a result of their wide band gap, group IIIV compounds are

of interest for phosphor production (Hartmann et al., 1982).
Compared to bulk materials, nanocrystalline ZnS are known to
have rather attractive characteristics, including their photoluminescence (PL) and thermoluminescence (TL) properties (Wu et al.,
2006). Size-dependent effects of nanostructures are expected to
bring about novel changes in the electronic, optical and mechanical properties of materials, mainly due to quantum phenomena.
In regard to chemical vapor deposited (CVD) diamond, an
attractive insulator in modern high-power electronic and optoelectronic devices (large-area CVDs are now available with thermal
conductivities ca. 20 W cm 1 K 1) (Descamps et al., 2006), its
radiation resistance and tissue equivalence makes it highly interesting for radiation dosimetry. To-date, studies of CVD diamond as
TLD materials have tended to focus on TL response as a function of

K.H. Al-Hinai et al. / Applied Radiation and Isotopes 82 (2013) 126129

127

doping (Descamps et al., 2006; Barboza-Flores et al., 2007; Nam

et al., 1991), comparison between undoped and N doped CVD
diamond generally lacking any clear trends in TL yield. Although
doping levels due to inclusion of N may have contributed to TL
response, direct comparison with undoped CVD diamond is
difcult due intrinsic defect levels present in all CVD diamond.

2. Methodology
Chemical precipitation was used to produce ZnS nanophosphor
samples, adding ZnSO4 to Na2S, continuously reuxed to obtain a
colloidal form of ZnS. MnSO4 injected into the colloid yielded
ZnS:Mn nanophosphors, with samples of 1 M, 2 M and 3 M MnSO4
being prepared in this manner. The physical properties of the
nanophosphors have been studied using a eld emission scanning
electron microscope (FESEM), X-ray diffraction (XRD) and energydispersive X-ray (EDX) analysis, providing morphological, structural and elemental analyses respectively. For TL study, the
samples were rst annealed using step-wise heating of 500 1C,
300 1C and 100 1C, for 30 min per step. Some 30 mg of sample
was then irradiated to Co-60 gamma rays from a Co-60 irradiator,
providing various doses, read-out being performed using a
Harshaw (model 3500) TLD reader.
The diamond lms were grown in a conventional hot-lament
CVD reactor, the chamber was evacuated to 2  10 3 mbar, before
performing a hot-lament CVD deposition process. A mirrorpolished Si wafer with dimensions of 2  3 mm2 was used as
substrate. With the substrate mounted at about 3 mm below the
lament, hydrogen gas was run into the reaction chamber with
the lament temperature set at about 1800 1C for 30 min to clean
the substrate surfaces. Then, reaction gas of a mixture of H2 and
CH4, at 1% methane concentration was introduced into the reaction chamber. The total gas pressure was 10 mbar and the lament
was coiled tungsten (0.5 mm in diameter) heated in a methane
hydrogen atmosphere. The substrate holder temperature was
650 7 50 1C measured by thermocouple mounted on the substrate holder and the substrate-to-lament distance was 3 mm.
Following 6 h deposition, the methane gas was switched off and
then the hydrogen gas was left to ow for 45 min to etch the
graphite performed during the process. Again, several techniques
have again been used in characterization of the lm, including
Scanning FESEM to determine the morphology/topography and
XRD for crystallographic study of the lms. In addition, Raman
spectroscopy has been used to determine the quality of the lms.
For TL studies, the samples were rst annealed at 400 1C for 1 h
before being irradiated by a Co-60 irradiator and readout as before.

Fig. 1. Micrograph of ZnS:Mn nanophosphors.

Fig. 2. Elemental makeup of the Mn-doped ZnS nanophosphors.

3. Results and discussion

The typical morphology and chemical compositions of the ZnS
samples are depicted in Figs. 1 and 2 respectively.
As seen in Fig. 1, the nanophosphors obtained are mostly
spherical with diameters in the predominant range 85150 nm.
The presence of Mn is evident in Fig. 2. The typical form of TL glow
curve is shown in Fig. 3, an intense broad feature, peaking at
250 1C, that points to the potential suitability of ZnS:Mn nanophosphors as TLD material. The glow curve shape is typical of
semiconductor material (Barboza-Flores et al., 2007). In regard to
dose sensitivity, the TL response is linear up to a dose of 100 Gy,
the sensitivity increasing with Mn molarity as shown in Fig. 4.
Fig. 5 shows a rather poor thermal stability, the results being for
irradiated samples kept in a darkened room at room temperature.
The initial fading rate is appreciable, with a 40% loss in signal
1 day post irradiation, with only  30% residual signal remaining

Fig. 3. Typical TL glow curve of ZnS:Mn nanophosphors.

after 4 days. Clearly, to be a useful basis for a radiation dosimeter,

the fading rate would need to be reduced perhaps by investigation
of the affect of annealing or adding of different rare earth
elements. It may also be possible to us the photo transfer
thermoluminescence (PPTL) system to provide for a more stable
situation, the PPTL technique typically using a UV source to photostimulate deep trap electrons into more supercial levels (see for
instance, Kharita et al., 1994).

128

K.H. Al-Hinai et al. / Applied Radiation and Isotopes 82 (2013) 126129

For the CVD lm deposited with 1% CH4 in H2 for 6 h, a high

diamond nucleation density has been obtained, leading to a
continuous lm on the abraded silicon substrate. Diamond crystals
with rough faces were observed, the crystals size being  10 mm
with an estimated density of 2.25  1010 crystals per cm2.
The XRD results from samples obtained using 1% CH4 in 99% H2
for a deposition time of 6 h onto a silicon substrate (Fig. 6) shows
clear and sharp diffraction peaks of diamond at 43.851, 75.251 and
91.51 as well as a silicon peak observed at 56.351. The XRD
spectrum for the substrate alone shows the main peak to be at
40.31 due to the SiC (1013) plane giving the interplanar spacing of
2.2383 compared with the published value of 2.2306 from JCPDS
standard card nos. 22-1319 and 20-0258.

Fig. 4. TL sensitivity of ZnS nanophosphors for different Mn molarities.

14

For TL studies, samples were annealed for 1 h at 400 1C. XRD

analysis shows an absence of crystallinity, suggesting that with
high temperature annealing, samples become amorphous (Fig. 7).
Fig. 8 shows the Raman spectra obtained from CVD diamond on
silicon substrates using a probe beam at 532 nm. Fig. 8(a) shows
the as deposited CVD diamond where two distinct sharp peaks are
evident. The rst peak at about 509 cm 1 was due to scattering
from crystalline silicon. The second peak at about 1332 cm 1 was
the typical Raman shift of crystalline diamond (Staryga and Bak,
2005). The sharp peak which is characteristic single crystal
diamond was due to the symmetric bonding of the carbon atoms
by strong covalent sp3 bonds, forming a cubic structure belonging
to the O7h (Fd3m) space group. Also barely visible is a small feature
at about 1560 cm 1 which can be attributed to amorphous carbon
or diamond like carbon (Harshavardhan et al., 1990). These are
carbon forms with a mixture of sp2 and sp3 bondings, lacking the
long range order to form proper crystals. In the spectra for the CVD

Fig. 7. Post irradiation/annealing XRD of diamond lm on silicon substrate using

HFCVD 1% CH4 in 99% H2 and substrate holder temperature 650 7 501 C. The
highest annealing temperature was 400 1C.

1 mol Mn

12

2 mol Mn

10

3 mol Mn
none Mn

8
6
4
2
0

2
3
Fading time (day)

Fig. 5. Fading rate of  250 1C TL peak at room temperature of nanophosphors with

different Mn concentration.

Fig. 6. XRD of amorphous carbon lm on a silicon substrate using HFCVD 25% CH4
in 75% Ar and substrate holder temperature 650 750 1C.

Fig. 8. Raman spectra of CVD diamond lm on silicon substrate; (a) as deposited

and; (b) annealed at 400 1C for 1 h.

K.H. Al-Hinai et al. / Applied Radiation and Isotopes 82 (2013) 126129

129

detectors strongly dependent on the lm quality (2). A typical TL

glow curve is shown in Fig. 9(b).
Fig. 10 shows the fading of CVD lm as a function of storage
time, when kept in a darkened room at room temperature, being
up to 50% over the period of  25 h. It is apparent that in its
present form the diamond lm is unsuitable as a short term TLD,
with several factors, such as electron tunneling, that are responsible for the fading (3).

4. Conclusions
The work herein represents two example efforts at producing
TL media with unique properties, the rst of the media being ZnS
nanophosphors. This consisted of particles of sizes in the range
85150 nm, the thermoluminescence (TL)-based radiation detection medium giving rise to a single peaked glow curve of maximum yield at a temperature of 2501 C at a heating rate of 51 C/s.
The TL response increased linearly with radiation dose. ZnS doped
with 2 mol of Mn was found to be the most radiation sensitive.
Regarding chemical vapor deposited (CVD) thin lms, inappreciable TL was found for the resultant ball-like amorphous carbon
lms, graphite, and the silicon substrate, whereas CVD diamond
lms showed a promising degree of linearity with dose. For both
the ZnS and diamond samples, TL signal fading was found to be
appreciable, even with storage in the dark at room temperature, at
some 40% per day for ZnS and4 50% per day for CVD lms, making
it apparent that there is need to adjust parameters such as the size
of the nanoparticles.
Fig. 9. TL responses of CVD lm to gamma radiation from a Co-60 irradiator (a) and
typical TL glow curve (b).

TL intensity (arb. units)

Acknowledgments
We wish to acknowledge high impact research grant number
HIR-21001-00-F000033.

1600
1200

References

800
400
0

10

20

30
40
Time (hours)

50

60

Fig. 10. TL fading of CVD diamond lm on silicon substrate.

diamond sample annealed at 400 1C for 1 h, two broad peaks at

about 1360 cm 1 and 1560 cm 1 dominate while the sharp peak
due to the silicon substrate also persists. The broad peaks can be
attributed to graphitic carbon, characteristic of carbon forms such
as graphite and carbon nanotubes (Harshavardhan et al., 1990).
The result indicate that the sample has lost some degree of
crystallinity due to annealing.
CVD diamond lm on silicon substrate is seen to demonstrate
linear sensitivity for doses in the approximate range 30200 Gy,
reducing in response beyond that (Fig. 9a). The polycrystalline
nature of the lms makes the performance of CVD diamond

Barboza-Flores, M., Gastlum, S., Cruz-Zaragoza, E., Melndrez, R., Chernov, V.,
Pedroza-Montero, M., Favallic, A., 2007. Thermoluminescence properties of
undoped and nitrogen-doped CVD diamond exposed to gamma radiation.
Radiat. Meas. 7, 123129.
Descamps, C., Tromson, D., Guerrero, M.J., Mer, C., Rzepka, E., Nesladek, M.,
Bergonzo, P., 2006. Nitrogen-doped diamond: Thermoluminescence and dosimetric applications. Diamond Relat. Mater. 15, 833837.
Hartmann, H., Mach, R., Selle, B., 1982. Gap IIVI compounds as electronic materials.
In: Kaldis, E. (Ed.), Current Topics Materials Science, vol. 1. North Holland
Publishing, pp. 146.
Harshavardhan, K.S., Vijayarahavan, M.N., Chandrabhas, N., Sood, A.K., 1990. Raman
investigations of diamond lms prepared by combustion ames. J. Appl. Phys.
68, 33033306.
Kharita, M.H., Stokes, R., Durrani, S.A., 1994. Phototransferred Thermoluminescence
(PTTL) in LiF (Mg, Cu, P) (GR-200). Radiat. Meas. 23, 493496.
Nam, T.L., Karfunkel, U., Keddy, R.J., 1991. The effects of nitrogen impurity on the
radiation detection properties of synthetic diamond. Radiat. Eff. Defects Solids
116, 233252.
Staryga, E., Bak, G.W., 2005. Relation between physical structure and electrical
properties of diamond-like carbon thin lms. J. Diamond Relat. Mater. 14,
2334.
Wu, X., Lu, G., Li, C., Shi, G., 2006. Room-temperature fabrication of highly oriented
ZnO nanoneedle arrays by anodization of zinc foil. Nanotechnology 17,
49364940.