Thermoluminescence of aquamarine: A preliminary study

Radiation Measurements 155 (2022) 106806

Contents lists available at ScienceDirect

Radiation Measurements
journal homepage: www.elsevier.com/locate/radmeas

Thermoluminescence of aquamarine: A preliminary study
Jorge Herrera a, Rafael Cogollo a, *, Omar D. Guti´errez b, M.L. Chithambo c
a

Grupo de Materiales y Física Aplicada, Departamento de Física, Universidad de C´
ordoba, Cra. 6 # 76-103, Montería, 230002, Colombia
Grupo de Química B´
asica, Aplicada y Ambiente, Facultad de Ciencias Exactas y Aplicadas, Instituto Tecnol´
ogico Metropolitano, Calle 73 No. 76A – 354, Medellín,
050034, Colombia
c
Department of Physics and Electronics, Rhodes University, P O Box 94, Grahamstown, 6140, South Africa
b

A R T I C L E I N F O

A B S T R A C T

Keywords:
Beryl
Aquamarine
Thermoluminescence
Kinetic-analysis
Reproducibility

Thermoluminescence of aquamarine, the cyan variety of beryl (Be3Al2(SiO3)6), is reported. Samples were irra­
diated at room temperature using a 90Sr/90Y β source at a rate of 0.10 Gys–1. Measurements made at 1 ◦ C/s after
irradiation to 10 Gy show a prominent glow peak at 75 ◦ C and three secondary peaks at 113, 188 and 306 ◦ C.
Kinetic analysis of the main peak performed using the initial rise, whole glow peak and curve fitting methods
show that the peak follows first order kinetics, that its activation energy is of the order of 1 eV and that it has a
frequency factor of ~1012 s− 1. Reproducibility analysis shows that the material reproduces its response under
identical experimental conditions with a maximum uncertainty of 2.0%.

1. Introduction

major dopants whereas in Co, V and Fe-doped samples the luminescence
is deduced to be from the host lattice as the emission band is not
modified by introduction of dopant as evident from TL and radio­
luminescence spectra (Chithambo et al., 1995). Some reports on the
luminescent and dosimetric properties of beryl reported no emission
(Kati et al., 2012).
In this work the thermoluminescence of the aquamarine variety of
beryl is reported. Samples were irradiated at room temperature using
a90Sr/90Y β source at a rate of 0.10 Gys-1. We use different methods of
kinetic analysis to determine the kinetic parameters and to establish the
influence of the dose on its TL behavior. In addition, one of its dosimetric
characteristics, such as reproducibility, is reported.

Beryl (Be3Al2(SiO3)6) is a ring silicate with a hexagonal crystalline
structure. Varieties of beryl exhibit a wide range of color including green
(for emerald), blue (for aquamarine), pink (for morganite), among
others. From a dosimetry point of view, beryl is of interest as some of its
luminescence features resemble those of quartz (Chithambo et al.,
1995). Both are built of framework structures (but not framework sili­

cates) and they also contain silica tetrahedra linked to each other
(Chithambo et al., 1995). This report is concerned with aquamarine, the
cyan variety of beryl.
Reports of characterization of beryl by thermoluminescence (TL) are
rare. Only one report has been published on the thermoluminescence of
natural beryl doped with transition metals (Chithambo et al., 1995).
That work was the first successful attempt to record the low-temperature
spectra of beryl. This information is useful for the identification of defect
properties. The pale blue hue of aquamarine alluded to earlier is
ascribed to the presence of Fe-related cations within its matrix. By way
of comparison, beryl doped with Fe shows strong thermoluminescence
emission between 300 and 400 nm (Chithambo et al., 1995). In addition,
whereas manganese and chromium dopants are efficient thermolumi­
nescent activators in beryl, the absence of a common feature in the TL
spectra of doped beryl shows how intrinsic and extrinsic defects define
its TL. This is exemplified by the fact that in emerald (Cr doped beryl)
and morganite (Mn doped beryl) the luminescence is attributed to the

2. Materials and methods
Measurements were made on commercially available aquamarine
(African Gems and Minerals Inc., Cape Town, South Africa). The sample,
prepared in coarse grain form, was used as received without any premeasurement treatment. Experiments were performed using a RISØ
TL/OSL DA-20 Luminescence Reader. The luminescence was detected
by an EMI 9235QB photomultiplier tube through a 7 mm Hoya U-340
filter (transmission band 250–390 nm FWHM). Samples were irradiated
at room temperature using a90Sr/90Y β source at a rate of 0.10 Gys-1. All
measurements were carried out at a heating rate of 1◦ Cs− 1.

* Corresponding author.
E-mail address: (R. Cogollo).

/>Received 26 November 2021; Received in revised form 20 May 2022; Accepted 26 May 2022
Available online 29 May 2022
1350-4487/© 2022 Elsevier Ltd. All rights reserved.


J. Herrera et al.

Radiation Measurements 155 (2022) 106806

Fig. 1. Glow curve of aquamarine measured at 1 ◦ C/s after irradiation at 8, 40 and 100 Gy of beta dose. The inset shows the lower intensity peaks at 188 and 306 ◦ C
respectively for the same dose.

Fig. 2. Variation of position of the main peak with irradiation dose. The inset shows the influence of irradiation dose on the main peak for glow curves corresponding
at 8, 40 and 100 Gy of beta dose.

3. Results

3.2. Qualitative assessment of order of kinetics

3.1. Glow curve features

3.2.1. Dependence of peak position on dose
The position (TM ) of the main peak is shown in Fig. 2 as a function of
irradiation dose. The position is independent of dose at 74.9 ± 0.8 ◦ C
from 0.1 to 100 Gy. This result suggests that the main glow peak is
consistent with first-order kinetics. Some glow curves corresponding to
various doses are shown in the inset.

Fig. 1 shows examples of glow curves recorded at 1 ◦ C/s after irra­
diation to different doses (8, 40 and 100 Gy). The main peak is at 75 ◦ C.

There are three secondary peaks at 113, 188 and 306 ◦ C. The lower
intensity peaks are shown more clearly in the inset to Fig. 1.

2


J. Herrera et al.

Radiation Measurements 155 (2022) 106806

Fig. 3. Glow curve of aquamarine after various preheating temperatures. All measurements correspond to irradiation to 10 Gy.

Fig. 4. Plots of ln(I) versus 1/kT for the initial rise portion of the main peak of
aquamarine corresponding to 8, 40 and 100 Gy beta dose.

Fig. 5. Plots of ln(I/nb) against 1/kT for some values of b for the glow peak
recorded after beta irradiation to 100 Gy.

3.2.2. Tm-Tstop analysis for the main peak glow curve
The order of kinetics of the main peak was determined using the TmTstop method (McKeever and Chen, 1997). In the Tm-Tstop analysis, the
position of a peak, Tm is monitored with respect to the preheat tem­
perature, Tstop. The position of a first order peak is independent of
preheat temperature while that of a non-first order peak shifts towards
high temperatures with increase in the preheat temperature. In our case,
a sample irradiated to 10 Gy was first partially heated to 20 ◦ C and after
cooling, the complete glow curve was measured from 20 ◦ C to 400 ◦ C.
The procedure was repeated on the same sample, freshly irradiated each
time, but with the Tstop temperature increased by 2 ◦ C from 20 up to
52 ◦ C. Fig. 3 shows that the position of the main peak is independent of
Tstop at 76.2 ± 0.10 ◦ C showing that the main peak follows first order

kinetics. The result is consistent with the one on the dependence of Tm

on dose.
3.3. Kinetic analysis
The kinetic analysis of the main peak was determined using the
initial rise (IR), peak shape (PS), whole glow peak (WGP), curve fitting
(CF) and variable heating rate (VHR) methods described elsewhere
(Pagonis et al., 2006).
3.3.1. Initial rise method
The activation energy was determined using the initial rise method
for measurements corresponding to irradiation to 8, 40 and 100 Gy.
Fig. 4. shows plots of ln(I) versus 1/kT for the initial rise portion of the
main peak recorded after irradiation at 8, 40 and 100 Gy. The values of
3


J. Herrera et al.

Radiation Measurements 155 (2022) 106806

area n under a glow peak is related to the order of kinetics b as (Pagonis
et al., 2006):
( )
′
I
s
E
(1)
ln b = ln −
n

β kT
where s´is the effective frequency factor, β is the heating rate and E, the
activation energy. For an appropriate value of b, the term ln(I /nb ) is a
linear function of 1/kT with a slope − E and an intercept equal to ln(s´/β).
Fig. 5 shows a plot of ln(I /nb ) against 1/kT for some value of b for the
glow peak recorded after irradiation to 100 Gy. The plot for which b =
1.1 is linear. The values of the activation energy and frequency factor are
0.92 ± 0.01 eV and 3.7 × 1011 s− 1respectively, typical of this parameter.
The values of b (1.1) corresponding to 100 Gy dose indicate that the
peak follows first order kinetics. The activation energy is consistent with
the previous initial rise.
3.3.3. Curve fitting technique
The kinetic parameters of the main glow peak were further evaluated
by curve fitting using the expression,
(
)[
E T − TM
2kTM
b
I(T) = IM bb− 1 exp
1 + (b − 1)
kT TM
E
(
(
))]b−− b1
)( 2
2kT
T
E T − TM

exp
(2)
+(b − 1) x 1 −
E
kT TM
TM2

Fig. 6. Results of curve fitting for the main peak in aquamarine irradiated to
100 Gy and read out at 1 ◦ C/s.

the activation energy corresponding to 8, 40 and 100 Gy are (0.76 ±
0.01), (0.86 ± 0.01) and (0.89 ± 0.01) eV respectively.

where TM is the temperature corresponding to the maximum TL in­
tensity IM (Kitis et al., 1998). The goodness of fit was tested by the
Figure of Merit (FOM) (Balian and Eddy, 1977). The values of FOM are
considered acceptable if they are less than 3%.
Fig. 6 shows a fit of Eq. (3) to the main peak in a measurement
following beta irradiation to 100 Gy. The frequency factor was

3.3.2. Whole glow peak method
The whole glow peak method is based on use of the area under a glow
peak. Using this method, the order of kinetics, activation energy and
effective frequency factor were evaluated. According to the theory, the

Fig. 7. The peak intensity of the main peak of aquamarine for 12 repeated identical measurements for different irradiation doses.
4


J. Herrera et al.

calculated by using the following expression:
)
(
βE
E
( 2kTM )) exp
s= 2 (
kTM
kTM 1 + (b − 1) E

Radiation Measurements 155 (2022) 106806

carried out at a heating rate of 1 ◦ C/s. The glow curve consists of the
main peak at 75 ◦ C and three secondary peaks at 113, 188 and 306 ◦ C.
The position of the main peak is independent of dose between 0.1 and
100 Gy, a behavior suggestive of first order kinetics. Kinetic analysis of
the main peak was done using the initial rise, whole glow peak and curve
fitting methods all of which also show that the peak follows first order
kinetics. Its activation energy is of the order of 1 eV with a frequency
factor of ~1012 s− 1. The reproducibility of the peak is favorable.

(3)

Using the curve fitting method on the main peak as described, the
value of the activation energy is (0.96 ± 0.01) eV; the order of kinetics
(b) is 1.27 ± 0.01 and frequency factor is 7.60 × 1012 s− 1. The best fit
was obtained with an FOM of 2.45%. These values can be compared with
ones from the other methods. The activation energy, frequency factor
and order of kinetics are comparable with those from other methods

used earlier.

Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.

3.4. Reproducibility
For reproducibility analysis, twelve identical measurements were
carried out on a sample after irradiation to a given dose. Fig. 7 shows the
peak intensity against the number of measurements for the main peak
for doses 1, 10, 20 and 40 Gy. The intensity of the peak is consistent for
the 12 cycles of measurements used. For 1, 10, 20 and 40 Gy beta
irradiation, the coefficient of variation (CV) in the results of 12 identical
measurements in each found to be about 2, 0.8, 0.5 and 0.8% respec­
tively. This shows that the material properly replicates its response
under identical experimental conditions. The decrease in CV with in­
crease in irradiation dose indicates that the reproducibility in general
improves with dose.

References
Balian, H., Eddy, N., 1977. Figure of Merit (FOM), an improved criterion over the
normalized Chi-squared Test for assessing goodness of fit of Gamma Ray. Nucl.
Instrum. Methods B 145, 389–395. />Chithambo, M.L., Raymond, S.G., Calderon, T., Townsend, P.D., 1995. Low temperature
luminescence of transition metal-doped beryls. J. Afr. Earth Sci. 20, 53–60. https://
doi.org/10.1016/0899-5362(95)00044-T.
Kati, M.I., Türemis, M., Keskin, I.C., Tastekin, B., Kibar, R., Çetin, A., Can, N., 2012.
Luminescence behaviour of beryl (aquamarine variety) from Turkey. J. Lumin.
2599–2602 />Kittis, G., G´
omez-Ros, J.M., Tuyn, J., 1998. Thermoluminescence glow curve

deconvolution functions for first, second and general order of kinetics. J. Phys. D
Appl. Phys. 31, 2636–2641.
McKeever, S.W., Chen, R., 1997. Luminescence models. Radiat. Meas. 27, 625–661.
/>Pagonis, V., Kitis, G., Furetta, C., 2006. Numerrical and Practical Exercises in
Thermoluminescence, first ed. Springer, New York, NY. />
4. Conclusions
The TL of aquamarine, a pale blue hue variety of beryl (Be3Al2(­
SiO3)6) has been studied. Samples were irradiated at room temperature
using a90Sr/90Y β source at a rate of 0.10 Gys-1. Measurements were

5