Thermally stimulated current observation of trapping centers and their distribution in as-grown TlGaSeS layered single crystals

Materials Chemistry and Physics 118 (2009) 32–36

Contents lists available at ScienceDirect

Materials Chemistry and Physics journal homepage: www.elsevier.com/locate/matchemphys

Thermally stimulated current observation of trapping centers and their distribution in as-grown TlGaSeS layered single crystals Tacettin Yıldırım a,b,∗ , Nizami M. Gasanly b a b

Department of Physics, Nevs¸ehir University, 50300 Nevs¸ehir, Turkey Department of Physics, Middle East Technical University, 06531 Ankara, Turkey

a r t i c l e

i n f o

Article history: Received 11 January 2009 Received in revised form 19 May 2009 Accepted 27 June 2009 PACS: 71.55.−I 77.20.Jv 72.80.Jc Keywords: Semiconductors Chalcogenides Defects Electrical properties

a b s t r a c t Thermally stimulated current (TSC) measurements were carried out in as-grown TlGaSeS layered single crystals. The investigations were performed in temperatures ranging from 10 to 160 K with heating rate of 0.8 K s−1 . The analysis of the data revealed three electron trap levels at 13, 20 and 50 meV. The activation energies of the traps have been determined using various methods of analysis, and they agree with each other. The calculation for these traps yielded 3.3 × 10−24 , 1.0 × 10−24 , and 1.1 × 10−24 cm2 for capture cross-sections and 1.9 × 1012 , 2.9 × 1011 and 4.5 × 1010 cm−3 for the concentrations, respectively. It was concluded that in these centers retrapping was negligible, as confirmed by the good agreement between the experimental results and the theoretical predictions of the model that assumes slow retrapping. An exponential distribution of electron traps was also revealed from the analysis of the TSC data obtained at different light excitation temperatures. This experimental technique provided values of 9 and 77 meV/decade for traps distribution of peaks A and C, respectively. © 2009 Elsevier B.V. All rights reserved.

1. Introduction The layered semiconducting crystal TlGaSeS is formed from TlGaSe2 and TlGaS2 crystals by replacing half of the selenium (sulfur) atoms with sulfur (selenium) atoms. The crystal lattice has two-dimensional layers arranged parallel to the (0 0 1) plane [1,2]. The bonding between Tl and Se(S) atoms in TlGaSeS is an interlayer type whereas the bonding between Ga and Se(S) is an intralayer type. The optical and the electrical properties of TlGaSe2 and TlGaS2 crystals were studied in Refs. [3–8]. The optical properties of TlGaSeS were reported in Ref. [9]. The direct and indirect band gap energies were found as 2.58 and 2.27 eV at room temperature. These crystals are useful for optoelectronic applications as they have high photosensitivity in the visible range of the spectra and high birefringence in conjunction with a wide transparency range of 0.5–14.0 ␮m [8]. The study of the dielectric constant dispersion has revealed that TlGaSe2 layered crystals exhibits two structural phase transition: the first at Ti ≈ 120 K, representing a second-order phase transition to an incommensurate phase; the second at Tc ≈ 107 K, representing a first-order phase transition to a commensurate ferroelectric

∗ Corresponding author at: Department of Physics, Nevs¸ehir University, Avanos Road, 50300 Nevs¸ehir, Turkey. Tel.: +90 384 2153900; fax: +90 384 2153948. E-mail address: [email protected] (T. Yıldırım). 0254-0584/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.matchemphys.2009.06.039

phase [10,11]. The results of specific heat measurements on TlGaSe2 and TlGaS2 crystals in the temperature range of 3–300 K have been reported in Ref. [12]. While for TlGaSe2 crystals the Cp (T) curve has singular features at 108.9 and 118.4 K, there were no anomalies in the Cp (T) curve of TlGaS2 crystals. In our previous paper [13], polarized Raman spectra of TlGaS2 crystals have been studied as a function of temperature. The appearance of new lines at T ≈ 180 and 250 K while cooling the crystal we have attributed to the phase transitions in TlGaS2 . Kato et al. [5] observed the anomalies of the temperature coefficient of exciton absorption peak shift of TlGaS2 crystals around 180–190 and 240–250 K. Some Raman lines were found to split at temperatures around 230–260 K. These features were considered by authors to show that in TlGaS2 successive phase transitions occur around 180–190 and 230–260 K. Moreover, the dielectric spectroscopy was used to study the vibrational spectra of TlGaSe2(1−x) S2x (0 ≤ x ≤ 1) mixed crystals [14]. It was established that, when the value of x exceeds 0.25, there were no longer any actual phase transitions in the studied crystals. The influence of defects on the performance of optoelectronic devices is a well-known subject. In optoelectronic devices such as LEDs or lasers, defects may introduce nonradiative recombination centers to lower the internal quantum efficiency or even render light generation impossible, depending on defect density. In the case of electronic devices, defects introduce scattering centers lowering carrier mobility, hence hindering high-frequency operation. Among the several experimental methods for determining

T. Yıldırım, N.M. Gasanly / Materials Chemistry and Physics 118 (2009) 32–36

the properties of trap centers in semiconductors, thermally stimulated current (TSC) measurements are relatively easy to perform and provide detailed information on trap states [15–22]. In TSC experiments, traps are filled by band-to-band excitation of carriers at low temperatures using a suitable light source. If the trapped charge carriers are thermally released to the conduction (valence) band upon heating, they give rise to a transient increase in the conductivity of the sample. A TSC curve for a single trap depth has the form of a slightly asymmetric curve with a fairly sharp maximum at a temperature, which is determined by the trap depth, the capture cross-section of the trap and the heating rate. The purpose of the present work is to obtain detailed information concerning trapping centers in undoped TlGaSeS layered crystals using the well-established technique of TSC measurements. We used the various methods to analyze the measured TSC spectra. The activation energy, attempt-to-escape frequency, the capture cross-section, concentration and distribution of the electron traps in TlGaSeS crystals are reported. 2. Experimental details TlGaSeS polycrystals were synthesized from high-purity elements (at least 99.999%) prepared in stoichiometric proportions. Single crystals of TlGaSeS were grown by the Bridgman method in evacuated (10−5 Torr) silica tubes with a tip at the bottom. The ampoule was moved in a vertical furnace through a thermal gradient of 30 ◦ C cm−1 , between the temperatures 850 and 480 ◦ C at a rate of 1.0 mm h−1 . The resulting ingot appears red in color and the freshly cleaved surfaces were mirrorlike. For TSC measurements a sample with dimensions of 9 mm × 10 mm × 0.8 mm were used. Electrical contacts were made on the sample surface with silver paste according to “sandwich” geometry. In this configuration, the electrodes are placed on the front and back sides of the crystal. Thin copper wires were attached to the electrodes for circuit connection. The TSC measurements were performed in the temperature range from 10 to 160 K using a closed-cycle helium cryostat. The sample was mounted on the cold finger of the cryostat. Constant heating rates in the range of 0.6–1.2 K s−1 were achieved by a Lake-Shore 331 temperature controller. A Keithley 228A voltage/current source and a Keithley 6485 picoammeter were used for the TSC measurements. The nominal instrumental sensitivities of temperature and current measurement devices were about 10 mK and 2 pA, respectively. At low enough temperatures, when the probability of thermal release is negligible, the carriers are photo-excited by using a light emitting diode (LED) generating light at a maximum peak of 2.6 eV. Using the value of absorption coefficient (1440 cm−1 ) at this energy, reported from the measurements of transmission and reflection on TlGaSeS crystals [9], we evaluated the optical absorption length as about 7 ␮m. The trap filling was performed by illumination under bias voltage of V1 = 1 V at the initial temperature T0 = 10 K for about 10 min. Then the excitation was turned off. After an expectation time (≈60 s) the bias voltage of V2 was applied to the sample and temperature was increased at constant rate. In TlGaSeS the dark current contribution is low, therefore the voltage of V2 = 100 V can be applied during heating.

3. Results and discussion 3.1. Determination of the trap type and minimum excitation time When the sample is illuminated, both types of carriers are created in this region. Either type of carrier may be driven out from this region by applying the bias voltage leading to different charge distribution in the sample that is affected by the type of trapping centers. The type of trap can be determined by its filling under photo-excitation through the positive and negative contacts. In our experiments, the samples were illuminated through the negative contact. In this case, mainly the electrons are distributed in the crystal and then trapped. Therefore, the observed peaks in TSC spectra of TlGaSeS crystal are assigned to electrons traps (Fig. 1). It is useful to know the minimum excitation time necessary for carrying out the TSC experiments. To determine this, we illuminated the sample during different time periods between 50 and 900 s and recorded the TSC curves in the temperature range of 10–160 K at constant heating rate of ˇ = 0.8 K s−1 . We found the minimum illumination (saturation) time as 600 s.

33

Fig. 1. Experimental TSC spectrum of TlGaSeS crystal registered at the excitation temperature T0 = 10 K and decomposition of this spectrum into two separate peaks (A and B). Open circles are experimental data. Dashed curves represent decomposed peaks. Solid curve shows total fit to the experimental data. Inset: Experimental TSC spectrum of TlGaSeS crystal obtained after the thermal cleaning at T ≈ 55 K and decomposition of this spectrum into two separate peaks (B and C). Open circles are experimental data. Dashed curves represent decomposed peaks. Solid curve shows total fit to the experimental data.

3.2. Activation energy determination There are several methods to evaluate the trapping parameters from the experimental TSC spectra. The applicability of some methods is restricted if the spectra consist of a number of overlapped peaks. We have used the curve fitting and initial rise methods for the analysis of the present data. 3.2.1. Curve fitting method The curve fitting method is used for decomposition of the TSC spectra into separate peaks associated with the charged traps in TlGaSeS crystals. For the monomolecular conditions (i.e., slow retrapping), the TSC curve of a discrete set of traps with trapping level Et is described by [15]: I = nt e

V  2

L

 exp

−

Et − kT



T

T0



Et  exp − kT ˇ



 dT

,

(1)

where I is the thermally stimulated current, nt is the initial density of filled traps,  is the lifetime of a free carrier, e is the electronic charge,  is the carrier mobility,  is the attempt-to-escape frequency of a trapped carrier, A and L are the area and length of the sample, respectively, T0 , the temperature at which heating begins after filling of the traps. If we assume  to be independent of T and ignore the variation of  and  with T over the temperature span of the TSC curve, Eq. (1) can be rewritten as [23]:





t

I = A0 exp −t + B

 exp(−t˜)t˜−2 dt˜ ,

(2)

t0

where t = Et /kT, and A0 and B are constants: A0 = nt e

V  2

L

A

and

B=

Et ˇk

(3)

If Eq. (2) is differentiated and equated to zero to find the maximum of the curve, which occurs at t = tm = Et /kTm , then 2 . B = exp(tm )tm

(4)

In order to analyze all peaks of spectra simultaneously, the fitting function comprising the sum of all features of the TSC spectra was

34

T. Yıldırım, N.M. Gasanly / Materials Chemistry and Physics 118 (2009) 32–36

Table 1 The activation energy (Et ), attempt-to-escape frequency (), capture cross-section (St ) and concentration (Nt ) of traps for three TSC peaks of TlGaSeS crystal. Peak

Tm (K)

A B C

36.3 55.0 101.8

Et (meV) Curve fitting method

Initial rise method

13 20 50

13 20 47

built as:

m

I(T ) =

Ii (T ),

(5)

i=1

where Ii (T) denotes the current contribution of each peak, calculated by means of Eq. (2), and m denotes the number of trap levels involved in the calculation. Fig. 1 shows the experimental TSC spectrum of TlGaSeS crystal obtained at the excitation temperature T0 = 10 K using the heating rate of ˇ = 0.8 K s−1 . Attempts to fit the theoretical curve to experimental data with only single peak were not successful. This fact forced us to fit the data by means of two peaks (designated A and B). As a result we have obtained a good fit for the experimental data. The activation energies for these peaks were found to be EA = 13 meV and EB = 20 meV. Thereafter, we decided to use so-called “thermal cleaning” procedure to verify the presence of the additional weak peaks in the high-temperature part of the spectra along with peaks A and B. “Thermal cleaning” procedure was applied as follows: the sample was cooled and irradiated at T0 = 10 K. Then it was taken through the same heating cycle as before but was stopped at a temperature near Tmax = 55 K of peak B. This way, the traps responsible for the current peaks A and B were substantially emptied. Then, the crystal was recooled in the dark down to 10 K and reheated at the same rate. The quenching the high-intensity peaks A and B enables us to observe the small details of the remaining part (inset of Fig. 1). However, we could not register the third peak C alone, since the initial intensity of the second peak B is much higher (almost 10 times) than that of the third peak C. We noticed a weak shoulder due to the second peak B on the low-temperature side of TSC spectrum. This may be due to the fact that the trapping levels corresponding to the second peak B could not be emptied entirely. It would be possible to empty completely the traps associated to the peak B by applying higher thermal cleaning temperatures. However we have observed that such a process drastically decreases the intensity of the third peak C as well. Attempts to fit the theoretical curve to the experimental data of thermally cleaned spectrum using only single peak were not successful either. Therefore, we have retained to fit the observed TSC spectrum using two peaks (B and C). The theoretical curve calculated for two peaks (with activation energies EB = 20 meV and EC = 50 meV) is compatible with the thermally cleaned experimental data in the 10–160 K temperature range (inset of Fig. 1). Good agreement has been obtained between the experimental TSC data and theoretical curves, computed with the assumption of slow retrapping (Fig. 1). This suggests that retrapping does not occur for the traps of TlGaSeS studied in the present work. As a result, we have determined three trapping centers in TlGaSeS crystal with activation energies of 13, 20 and 50 meV (Table 1). 3.2.2. Initial rise method The initial rise method, valid for all types of recombination kinetics [15], is based on the assumption that the TSC is proportional to exp (−Et /kT) when the traps begin to empty with temperature. Thus, a semi-logarithmic plot of the current versus 1/T gives a straight line with a slope of (−Et /k). The plots for TSC peaks of TlGaSeS crystal

 (s−1 )

St (cm2 )

Nt (cm−3 )

3.9 2.8 9.9

3.3 × 10−24 1.0 × 10−24 1.1 × 10−24

1.9 × 1012 2.9 × 1011 4.5 × 1010

are shown in Fig. 2. The activation energies of the traps calculated by this procedure are listed in Table 1. Due to the lack in the literature of experimental data on temperature-dependent electrical conductivity of TlGaSeS crystals, we are forced to compare our obtained results on activation energies with those reported for TlGaSe2 and TlGaS2 crystals. Since these crystals have the high resistivity, the dark electrical conductivity measurements were possible only down to 200 and 100 K, respectively [24,25]. The analysis of temperature-dependent conductivity of TlGaSe2 (200–350 K) and TlGaS2 (100–350 K) revealed the existence of moderate depth levels with activation energies of 330 meV (TlGaSe2 ) and 240 and 360 meV (TlGaS2 ). The authors of Ref. [3] also studied the electrical conductivity of TlGaSe2 and TlGaS2 crystals as a function of temperature (100–300 K). They obtained the activation energies of 207 and 370 meV for TlGaSe2 and TlGaS2 crystals, respectively. Thus, the measurements of electrical conductivity were enabled to expose only the existence of moderate levels, while in our TSC experiments we revealed the existence of shallow levels with activation energy of 13, 20 and 50 meV. 3.3. Determination of the traps distribution The information about the characteristic features of traps distribution can be obtained from the analysis of TSC data. The experimental TSC procedure for the analysis of traps distribution was applied as follows: the sample was excited by light at different excitation temperatures to allow trapping of the photoproduced electrons. Then the light was turned off and the sample was recooled to initial temperature (T = 10 K) in darkness. Thereafter the sample was heated with a constant heating rate ˇ = 0.8 K s−1 to excite the trapped electrons into the conduction band. Figs. 3 and 4 show the experimental TSC spectra obtained at different excitation temperatures T0A = 20, 23, 26, 29 and 31 K (peak A) and T0C = 55, 57, 60, 65 and 70 K (peak C), respectively. The spectra decreased in intensity and shifted towards higher temperatures with increasing the light excitation temperature. These shifts support the valid-

Fig. 2. Thermally stimulated current versus 1000/T for all three peaks in the TSC spectra of TlGaSeS crystal. Open circles are experimental data; solid lines show fits to the experimental data for A, B and C peaks.

T. Yıldırım, N.M. Gasanly / Materials Chemistry and Physics 118 (2009) 32–36

35

Table 2 TSC parameters of TlGaSeS crystals for peaks A and C at different excitation temperatures. Peak A a

Curve Excitation temperature (K) Maximum temperature (K) Curve area (a.u.) Activation energy (meV) a

1 20 47.0 78 18

Peak C 2 23 47.7 60 20

3 26 50.1 37 21

4 29 54.2 21 23

5 31 56.9 12 25

1 55 102.6 1807 69

2 57 105.2 1249 76

3 60 110.7 944 85

4 65 119.7 499 111

5 70 127.9 325 128

See Figs. 3 and 4.

excitation, respectively. Insets of Figs. 3 and 4 show the ln[So (Im /Ie )] plotted as a function of the energy Et determined from the analysis of TSC curves, which were registered at different excitation temperatures. The graph obtained for A peak is a straight line with a slope ˛ = 0.25 meV−1 corresponding to 9 meV/decade, an order of magnitude variation in the trap density for every 9 meV. The graph drawn for C peak is also a straight line with a slope ˛ = 0.03 meV−1 corresponding to 77 meV/decade, an order of magnitude variation in the trap density for every 77 meV. 3.4. Determination of capture cross-section and trap concentration

Fig. 3. Experimental TSC spectra (peak A) of TlGaSeS crystals at different excitation temperatures T0 . Inset: ln[So (Im /Ie )] plot as a function of activation energy Et .

ity of a quasi-continuous traps distribution [26–30]. The activation energies, obtained by curve fitting method, and the maximum temperatures of thermo-current curves registered at different excitation temperatures were listed in Table 2. The activation energies range from 18 to 25 meV at T0 = 20 and 31 K, and from 69 to 128 at 55 and 70 K for peaks A and C, respectively. We note that above T0 = 31 K (peak A) and T0 = 70 K (peak C), the TSC spectra could not be analyzed due to their low intensity. By assuming an exponential trap distribution whose density at energy Et will be given by N = A1 exp(−˛Et ), we can write for the traps filled at the excitation temperature T0 [26]: So

I  m

Ie

∝ A1 exp(−˛Et ),

where ˛ is the energy parameter which characterizes the distribution, So is the area of the TSC peak, Im and Ie are the currents flowing in the crystal at the peak temperature in darkness and during light

Once the TSC curve was fitted and the values of Et and Tm were determined (Table 1), Eqs. (4) and (3) were used to calculate B and the attempt-to-escape frequency , respectively. Knowing the value of  (see Table 1), one can calculate the capture cross-section of the traps according to following expression: St =

(6)

where Nc is the effective density of states in the conduction band and th is the thermal velocity of a free electron. The calculated values of St were found to be 3.3 × 10−24 , 1.0 × 10−24 , and 1.1 × 10−24 cm2 for peaks A, B and C, respectively (Table 1). The small values of the capture cross-section justify the assumption of monomolecular kinetics. The concentration of the traps was estimated using the relation [31]: Nt =

Q , ALeG

(7)

where Q is the amount of charge released during a TSC experiment and can be calculated from the area under the TSC peaks; G is the photoconductivity gain. Determination of the photoconductivity gain for a sample under the specified experimental conditions is a particular problem. Here we present an approach to evaluate photoconductivity gain by utilizing photoconductivity decay experiments [23]. When the light falls on the sample, electron–hole pairs are generated which change the material conductivity. The electric field in the sample causes the electrons and holes to move in opposite directions leading to current. The carriers are present in the system until they either recombine or are collected at the contacts. The photoconductivity gain in the samples arises because the electron goes around the circuit several times before it can recombine with a photo-generated hole. Each time the electron goes through the circuit it contributes to the current. The photoconductivity gain G was evaluated from the expression [32]: G=

Fig. 4. Experimental TSC spectra (peak C) of TlGaSeS crystals at different excitation temperatures T0 . Inset: ln[So (Im /Ie )] plot as a function of activation energy Et .

 , Nc vth

 V2 = . ttr L2

(8)

Here  is the carrier lifetime, ttr is the carrier transit time between the electrodes and V2 is the applied voltage. The carrier lifetime can be determined from the photoconductivity decay experiments [33]. The experiments are carried out by developing the set-up as

36

T. Yıldırım, N.M. Gasanly / Materials Chemistry and Physics 118 (2009) 32–36

of the traps were calculated to be 3.3 × 10−24 , 1.0 × 10−24 , and 1.1 × 10−24 cm2 . Also the concentrations of the traps were estimated to be 1.9 × 1012 , 2.9 × 1011 and 4.5 × 1010 cm−3 . An exponential distribution of electron traps was also revealed from the analysis of the TSC data obtained at different light excitation temperatures. This experimental technique provided values of 9 and 77 meV/decade for traps distribution of peaks A and C, respectively. Acknowledgment T. Yıldırım acknowledges the support given under the TUBITAKBIDEB (2218) Programmer. References

Fig. 5. The photoconductivity decay curve for the TlGaSeS crystal. Open circles are experimental data. Solid curve shows the theoretical fit to the experimental data.

follows. Using silver paste, we formed ohmic contacts on both sides of the sample according to sandwich geometry and that was illuminated by a high efficiency blue LED controlled by a digital signal generator operating square waves. The photocurrent was amplified by a fast current–voltage converter circuit. The signal was recorded by a fast digital voltmeter and transmitted to the computer. The recorded data were analyzed to determine the decay time of the photocurrent. Since the current decays are nearly exponential after termination of light pulse at t = t0 , the carrier lifetime  is determined by the corresponding output voltage equation:

 t

V = V0 + C exp −



,

(9)

where V0 is the voltage at t = ∞ and C is a constant. Fig. 5 shows the theoretical fit to the experimental data using Eq. (9) for TlGaSeS crystals. The carrier lifetime was obtained as  = 3.8 × 10−2 s from the decay of the photocurrent. The corresponding photoconductivity gain was found to be G = 3622 by means of Eq. (8) using the values V2 = 10 V and  = 61 cm2 V−1 s−1 [34]. The traps concentration (Nt ) in TlGaSeS crystals were evaluated using Eq. (7) as 1.9 × 1012 , 2.9 × 1011 and 4.5 × 1010 cm−3 for peaks A, B and C, respectively (Table 1). 4. Conclusions Three trapping centers at 13, 20 and 50 meV have been detected in as-grown TlGaSeS layered single crystals by the TSC technique. Since the crystals studied are not intentionally doped, the observed levels are thought to originate from defects, created during the growth of crystals, and/or unintentional impurities. The trap parameters were determined by various methods of analysis, and they agree with each other. The retrapping process is negligible for these levels, as confirmed by the good agreement between the experimental results and the theoretical predictions of the model that assumes slow retrapping. The capture cross-sections

[1] D. Muller, H. Hahn, Z. Anorg. Allg. Chem. 438 (1978) 258. [2] K.A. Yee, A. Albright, J. Am. Chem. Soc. 113 (1991) 6474. [3] M. Hanias, A. Anagnostopoulos, K. Kambas, J. Spyridelis, Mater. Res. Bull. 27 (1992) 25. [4] I.M. Ashraf, M.M. Abdel-Rahman, A.M. Badr, J. Phys. D: Appl. Phys. 36 (2003) 109. [5] A. Kato, M. Nishigaki, N. Mamedov, M. Yamazaki, S. Abdullaeva, E. Kerimova, H. Uchiki, S. Iida, J. Phys. Chem. Solids 64 (2003) 1713. [6] M.M. El Nahass, M.M. Sallam, S.A. Rahman, E.M. I˙ brahim, Solid State Sci. 8 (2006) 488. [7] V. Grivickas, V. Bikbajevas, P. Grivickas, Phys. Status Solidi (b) 243 (2006) R31. [8] K.R. Allakhverdiev, Solid State Commun. 111 (1999) 253. [9] I. Guler, N.M. Gasanly, J. Korean Phys. Soc. 51 (2007) 2031. [10] A.A. Volkov, Yu.G. Goncharov, G.V. Kozlov, S.P. Lebedev, A.M. Prokhorov, R.A. Aliev, K.R. Allakhverdiev, JETP Lett. 37 (1983) 615. [11] H.D. Hochheimer, E. Gmelin, W. Bauhofer, Ch. Von Schnerig-Schwarz, H.G. von Schnering, J. Ihringer, W. Appel, Z. Phys. B 73 (1988) 257. [12] S.G. Abdullaeva, A.M. Abdullaev, K.K. Mamedov, N.T. Mamedov, Sov. Phys. Solid State 26 (1984) 375. [13] A. Aydinli, R. Ellialtioglu, K.R. Allakhverdiev, S. Ellialtioglu, N.M. Gasanly, Solid State Commun. 88 (1993) 387. [14] A.A. Volkov, Yu.G. Goncharov, G.V. Kozlov, K.R. Allakhverdiev, R.M. Sardarly, Sov. Phys. Solid State 26 (1984) 1668. [15] R. Chen, S.W.S. Mckeever, Theory of Thermoluminescence and Related Phenomena, Word Scientific, Singapore, 1997. [16] G. Kitis, R. Chen, V. Pagonis, Phys. Status Solidi (a) 205 (2008) 1181. [17] E. Borchi, M. Bruzzi, S. Pirollo, S. Sciortino, J. Phys. D: Appl. Phys. 31 (1998) L93. [18] V.M. Skorikov, V.I. Chmyrev, V.V. Zuev, E.V. Larina, Inorg. Mater. 38 (2002) 751. [19] Z.Q. Fang, B. Claflin, D.C. Look, J. Appl. Phys. 103 (2008) 073714. ´ [20] J.M. Wrobel, A. Gubanski, E. Płaczek-Popko, J. Rezmer, P. Becla, J. Appl. Phys. 103 (2008) 063720. [21] J. Schafferhans, A. Baumann, C. Deibel, V. Dyakonov, Appl. Phys. Lett. 93 (2008) 093303. [22] R. Schmechel, H. von Seggern, Phys. Status Solidi (a) 201 (2004) 1215. [23] N.M. Gasanly, H. Ozkan, N.A.P. Mogaddam, Semicond. Sci. Technol. 21 (2006) 1250. [24] A.F. Qasrawi, N.M. Gasanly, Mater. Res. Bull. 39 (2004) 1353. [25] A.F. Qasrawi, N.M. Gasanly, Semicond. Sci. Technol. 20 (2005) 446. [26] A. Serpi, J. Phys. D: Appl. Phys. 9 (1976) 1881. [27] A. Bosacchi, B. Bosacchi, S. Franchia, L. Hernandez, Solid State Commun. 13 (1973) 1805. [28] A. Anedda, M.B. Casu, A. Serpi, I.I. Burlakov, I.M. Tiginyanu, V.V. Ursaki, J. Phys. Chem. Solids 58 (1997) 325. [29] P.C. Ricci, A. Anedda, R. Corpino, I.M. Tiginyanu, V.V. Ursaki, J. Phys. Chem. Solids 58 (1997) 325. [30] S.N. Singh, B.A. Sharma, A.N. Singh, Int. J. Mod. Phys. B 20 (2006) 3307. [31] C. Manfredotti, R. Murri, A. Quirini, L. Vasanelli, Phys. Status Solidi (a) 38 (1976) 685. [32] R. Bube, Photoelectronic Properties of Semiconductors, Cambridge University Press, Cambridge, 1992. [33] T. Pisarkiewicz, Opto-Electron. Rev. 12 (2004) 33. [34] A.F. Qasrawi, N.M. Gasanly, Cryst. Res. Technol. 41 (2006) 174.