Photoconductivity relaxation processes in AgCd2GaS4 single crystals

Materials Chemistry and Physics 200 (2017) 250e256

Contents lists available at ScienceDirect

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

Photoconductivity relaxation processes in AgCd2GaS4 single crystals G.L. Myronchuk a, M. Piasecki a, b, *, A.S. Krymus a, I.V. Kityk a, c, R.O. Vlokh d, A.O. Fedorchuk e, V.R. Kozer f, O.V. Parasyuk f a

Department of Physics, Eastern European National University, 13 Voli Avenue, 43025 Lutsk, Ukraine Institute of Physics, J. Dlugosz University, Ul. Armii Krajowej 13/15, 42-201 Czestochowa, Poland c Institute of Optoelectronics and Measuring Systems, Faculty of Electrical Engineering, Czestochowa University Technology, Armii Krajowej 17, PL-42-217, Czestochowa, Poland d Vlokh Institute of Physical Optics, 23 Dragomanov St., 79005, Lviv, Ukraine e Department of Inorganic and Organic Chemistry, Lviv National University of Veterinary Medicine and Biotechnologies, Pekarska St. 50, 79010 Lviv, Ukraine f Department of Inorganic and Physical Chemistry, Eastern European National University, 13 Voli Avenue, 43025 Lutsk, Ukraine b

h i g h l i g h t s
Optical and photoelectric properties of AgCd2GaS4 single crystal were explored.
Temperature dependence of the time kinetics was investigated.
AgCd2GaS4 single crystals was grown by Bridgman-Stockbarger method.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 8 January 2017 Received in revised form 20 April 2017 Accepted 13 July 2017 Available online 28 July 2017

Optical and photoelectric spectral features of AgCd2GaS4 single crystal were explored. Optical band gap energy was estimated from the absorption and was varied within 2.18e2.28 eV (at ambient temperature) for the samples cut from different parts of the single-crystalline specimens. The results of the photoconductivity relaxation study within the temperature range 100e280 K were explored within a framework of the adhesion trapping level model. It was established that the AgCd2GaS4 crystals exhibit long-term relaxation of photoconductivity with adhesion levels at energies about 0.2 eV. © 2017 Elsevier B.V. All rights reserved.

Keywords: Single crystal growth Bridgman technique Carrier relaxation

1. Introduction The AICIIIX2 crystalline compounds (AI ¼ Cu, Ag; CIII ¼ Ga, In; X ¼ S, Se, Te) are studied recently quite intensively. During the past decades due to their promising parameters and wide application, e.g. in non-linear optics (AgGaS2, AgGaSe2) [1,2] or photovoltaics (CuInSe2, CuIn1-xGaxSe2 etc) [3,4] they are of particular interest. They all (at least their low-temperature modifications) crystallize in the chalcopyrite structure (CuFeS2, SG I-42d) which is related to the crystal structure of their isoelectronic analogs BIIX (BII ¼ Zn, Cd, Hg). The unit cells of chalcopyrite and sphalerite (ZnX, CdTe, HgX, SG F43 m) are related by the ratios ach ~ as and cch~2cs [5], i.e. the

* Corresponding author. Department of Physics, Eastern European National University, 13 Voli Avenue, 43025 Lutsk, Ukraine. E-mail address: [email protected] (M. Piasecki). http://dx.doi.org/10.1016/j.matchemphys.2017.07.053 0254-0584/© 2017 Elsevier B.V. All rights reserved.

chalcopyrite cell results from doubling of the sphalerite cell along c axis. A consequence of this similarity is a formation of wide solid state alloy solution ranges for both components in the AICIIIX2eBIIX systems, up to continuous solid state alloys series in the cases when AICIIIX2 possess a high-temperature modification with the sphalerite structure [6e10]. A somewhat different situation is observed for the systems possessing CdS and CdSe favoring the crystal structure of wurtzite (SG P63mc). Previously we demonstrated that a continuous range of solid solutions (ternary compounds with the sphalerite structure) exists between CdS and the high-temperature modification of CuInS2 [11] and similarly is for in CuInSe2eCdSe system [12]. Additionally the intermediate phases were observed in the systems AgGaS2eCdS (phase diagram is of eutectic type) [13] and AgGaSe2eCdSe [14]. The formation of AgCd2GaS4 was found from the investigation of the AgGaS2eCdS section [13] for the quasi-ternary system Ag2SeCdSeGa2S3 [15]. The compound formed incongruently at

G.L. Myronchuk et al. / Materials Chemistry and Physics 200 (2017) 250e256

1284 K and has a homogeneity region for 64e79 mol.% CdS (at 870 K) localized along the AgGaS2eCdS section [13]. It was determined by the X-ray single crystal method, that AgCd2GaS4 crystallizes in an orthorhombic symmetry (wurtzite-stannite structure type), acentric space group Pmn21, and the lattice parameters were equal to a ¼ 8.1395 (9) Å, b ¼ 6.9394 (8) Å, c ¼ 6.6014 (7) Å [16]. The pffiffiffi unit cell parameters are bound by the ratios aws~2aw, bws ~ aw 3, cws ~ cw. The electron structure of AgCd2GaS4 was studied in Refs. [17e19]. Ab initio calculations of linear and non-linear optical susceptibilities by FT-LAPW method show strong negative uniaxial anisotropy [20]. Optical properties of the crystals obtained by Bridgman-Stockbarger method were investigated in Refs. [21,22]. The growth of the crystals with excess AgGaS2 as the solvent results in the change of the composition of the crystal along the boule due to the space homogeneity region. The crystals are photosensitive and exhibit intensive photoluminescence. The energy band gap of AgCd2GaS4 was determined from the edge of the optical absorption and was equal to 2.15e2.28 eV [21], and the IR transparency region reaches wavelengths up to 13 mm [22]. The AgCd2GaS4 crystals are n-type high-resistance semiconductors; their dark specific conductivity is equal to 106 U1cm1. All these characteristics indicate that AgCd2GaS4 may be promising material for non-linear optics and optoelectronics. One of the principal difference of the titled chalcogenide crystals with respect to the just known is a fact that they possess very large electron-phonon anharmonicities [23e25]. The trapping levels which are originated due to this process may substantially change the relaxation processes in the case of such interactions. The development of the technological processes of the growth of quality crystals of compounds and solid solutions required a complex electrical and optical studies to assess the influence of deviations from the stoichiometry on the principal optical parameters of the material. As a result of the research, one can determine and subsequently monitor these parameters during the process. In this work we have studied band gap energy and the relaxation of photoconductivity of the samples cut from different parts of the AgCd2GaS4 specimens. In Section 2 is presented technology of crystal growth. The crystallochemistry is given in Section 3. The principal results are given in Section 3.

2. Experimental Bridgman-Stockbarger method was used for the growth of AgCd2GaS4 single crystals. The detailed crystal growth process was earlier described by us in detail in Refs. [21,22]. Taking into account the incongruent type of the formation of the quaternary compound, the starting batch composition was selected from the field of its primary crystallization for 60 mol.% CdS. The alloy was synthesized from high-purity elements [13], then crushed into powder and poured into a quartz container with a conical bottom. To prevent the interaction of the melt with quartz, the container was previously graphitized by acetone pyrolysis in the flame of oxygen-gas burner. The upper zone temperature (growth zone) was equal to 1350 K, that of bottom (annealing) zone was 1020 K, temperature gradient at the solid-melt interface was 35 K/cm. The rate of temperature cooling of the ampoule was 5 mm/day. After the finishing of the crystallization the ampoule was transferred to the annealing zone and kept for 100 h. Then both furnaces were synchronously cooled to ambient temperature also during 100 h. The obtained boule consisted of two parts. The bottom conical part was the single crystal of the quaternary compound while the top portion contained the eutectic AgGaS2þAgCd2GaS4. The quantitative and qualitative EDAX analysis of the single

251

crystals was performed using Philips 515-PV9800 SEM. Three different parts of the single-crystalline boule were analyzed e the beginning of the cone (sample 1), exactly the middle (sample 2), and the portion that immediately bordered the eutectic (sample 3). The change of the composition of the AgCd2GaS4 crystal is caused by a wide homogeneity region of the compound itself [13] and solution-melt method of its production. Optical studies were performed on 0.08 mm thick parallel-plane plates with polished optical-quality surfaces. The measurement of the absorption spectra were carried out MDR-206 monochromator with spectral resolution 1 nm. The absorption coefficient was determined from LambertBouger law. Taking into account the reflection of light from internal surfaces of the sample and from the incident surface:

I ¼ I0

ð1  RÞ2 ead 1  R2 e2ad

(1)

If d is large, and neglecting the second member of the denominator, we obtain:

IzI0 ð1  RÞ2 ead

(2)

where І is the intensity of the transmitted light; І о is the intensity of the incident light; R is the reflection coefficient for the normal incidence; a is the absorption coefficient; d is the thickness of the sample. The investigation of photoelectric properties was performed of the samples shaped as regular parallelepipeds. The average dimensions of the samples were 6  2  1 mm3. The surfaces were polished by diamond abrasives. Relaxation and photoconductivity spectra were studied in the temperature range 100e300 K. Kinetics of the rise and decay of the photoconductivity was recorded after the illumination of the samples by a KLM-H980-200-5 diode laser (l ¼ 980 nm, Р ¼ 150 mW). Electric transport measurements were performed using a Keithley 6430 Sub-Femtoamp SourceMeter electrometer. 3. Crystallographic features of the AgCd2GaS4 structure The crystal structure of the AgCd2GaS4 compound may be presented as the packing of the tetrahedra of sulfur atoms surrounding Ga atoms, while Ag and Cd atoms are located in the voids between these tetrahedra (Fig. 1a). The cuboctahedral second coordination surrounding (SCS) [26] of GaS5 4 anions (Fig. 2) confirms the close packing of these anions. Ag atom 4s occupy tetrahedral voids within the SCS, while Cd atoms take octahedral voids. The solid solution range of AgCd2GaS4 at the AgGaS2eCdS

Fig. 1. Photo of the studied AgCd2GaS4 single crystal.

252

G.L. Myronchuk et al. / Materials Chemistry and Physics 200 (2017) 250e256

a)

b)

Fig. 2. a) Packing of the tetrahedra of sulfur atoms around Ga atoms in the structure of the AgCd2GaS4 compound. b) First and second coordination surrounding for GaS5 4 anions in the AgCd2GaS4 crystalline structure.

section may be presented by the formula Ag1-xCd2þ2xGa1-xS4. According to the SCS (Fig. 1b), incorporated Cd atoms will occupy the site-positions of Ag and Ga atoms. Such heteroatomic isovalent substitution results in the charge transfer from the anions to the cation sub-lattice which may show up as a difference in the physical properties within the solid solution. Presenting the crystal structure of AgCd2GaS as the packing of S2 anions, their SCS forms hexagonal analogs of cuboctahedra (Fig. 3) where Ga, Ag and Cd atoms are located in the tetrahedral voids. Inter-atomic distances Ag-S and Cd-S have close values (Fig. 3) which may result in a disordered mutual substitution of these atoms in this structure. The defects in these crystals will also appear due to the substitution Agþ þ Ga3þ 4 2Cd2þ within the solid solution Ag1-xCd2þ2xGa1-xS4 where large Cd2þ atoms occupy the positions of small atoms Ga3þ. 4. Results and discussion The EDX data on the distribution of the chemical composition along the grown crystal are presented in Table 1 [21]. The variation of the stoichiometry leads to the deviation of the long-range ordered periodicity of the potential energy for the electrons in the lattice which imposes on the AgCd2GaS4 crystals the features of disordered systems and consequently leads to the appearance of the tails of the density of electron states in the forbidden energy gap [27,28]. Other defects may provide additional contribution to the infringement of long-range order.

The study of the spectral dependence shows that the absorption coefficient in the transmission window hv < 1.8 eV has close values a ¼ 7e10 cm1 in different samples (Fig. 4) which is probably caused by the scattering and absorption of light due to various defect complexes and other structure damage of the crystal lattice. The dependence a(hv) below the spectral region of strong absorption is exponential. This region is called “Urbach's tail” [29]. Several mechanisms were suggested for explanations for this exponential dependence such as variations of the band gap energy caused by the fluctuations of the density of states, or widening of the band edge due to internal electric fields. Refs. [30,31] suggest that the absorption edge reflects the appearance of the tails of the density of states due to the fluctuation of bond angles and lengths. Other authors [32] consider the electron transitions between localized states in the tails of the band edges to be responsible for the absorption tails, and suggest that the density of such states decreases exponentially with the photon energy. The dependence a(hv) in the exponential section follows Urbach's rule:



a ¼ a0 exp

E  E0 EU

 (3)

where EU is Urbach's energy (EU ¼ kBT/s(T), kB is Boltzmann's constant, s(T) is a parameter that characterizes the slope of the absorption edge). Urbach's energy is generally a function of static and dynamic disorder EU ¼ FðD2D þ D2S Þ where F is a constant independent of hv. The studies performed in ref. [27] have shown that Urbach's rule is versatile and describes the spectral dependence of the absorption coefficient at hv > Eg in numerous semiconductors regardless of their type of chemical bonding. For the pure and flawless crystals even at the lower temperature EU is proportional to temperature which indicates the prevailing role of dynamic disorder over static disorder in these crystals. The spectral dependence of the absorption coefficient near the intrinsic absorption edge in many heavily-doped or defect semiconductors also follows Urbach's rule but only at sufficiently high temperatures (e.g. Т >100 K for doped gallium arsenide). The EU value at lower temperatures is almost independent on temperature but depends on the concentration of impurities increasing with the concentration of defects. This one indicates on the dominant role of static disorder. We have established that EU value are varied within the 0.07e0.09 eV. It should be noted that the low-defect single crystals of CdS which is the electron analog of the investigated compound

Fig. 3. First and second coordination surrounding of S2 anions and inter-atomic distances to the cations for the AgCd2GaS4 structure.

G.L. Myronchuk et al. / Materials Chemistry and Physics 200 (2017) 250e256

253

Table 1 Calculated and experimental composition of the AgCd2GaS4 crystals by EDX data. Composition

Stoichiometric 66.7 mol.% CdS

Boundary 52 mol.% CdS

Experiment 60 mol.% CdS (starting composition)

Measurement point

calculated

calculated

Bottom (sample 1)

Middle (sample 2)

Top (sample 3)

12.50 25.00 12.50 50.00

16.22 17.57 16.22 50.00

13.19 20.23 12.81 53.78

14.37 19.48 13.42 52.72

16.17 13.06 16.67 54.10

Element content (at.%)

400

Ag Cd Ga S

generally given as Dnst ¼ tstbkI, where Dnst is the stationary concentration of the non-equilibrium carriers [36]. For linear recombination tst is a constant independent on the excitation intensity, and therefore the stationary non-equilibrium concentration is proportional to the intensity (Fig. 5): Dnst ~ I. The presented relaxation processes have complex long-term character. It is known [37] that the curves of the photoconductivity rise in semiconductors for impurity excitation are described as:

sample 1 sample 2 sample 3



300 -1

, cm

200

100

0



t

Ds ¼ Dsst $ 1  exp  t1

1.8

2.0 hv, eV

2.2

Fig. 4. Spectral dependence of the absorption coefficient for the AgCd2GaS4 single crystals at 292 K.

has given the values varying EU ¼ 0.01e0.02 eV (at 77 K). The band gap energy estimated at 293 K by the position of the fundamental absorption edge for a ¼ 400 cm1 is equal to 2.18 eV; 2.22 eV; 2.28 eV for the samples 1, 2, 3 respectively (Fig. 4). It is obvious that the increase of the band gap energy is caused by the distribution of the chemical elements along the AgCd2GaS4 crystal (Table 1). It is known that the band gap energy of CdS is 2.42 eV [33,34]. contrary Eg of AgGaS2, according to various experimental data is varied within 2.51 eVe2.73 eV [5,35]. Typical curves of photoconductivity relaxation for the investigated samples after the excitation by monochromatic illumination of various intensity are presented in Fig. 5. The steady value of non-equilibrium conductivity is achieved only after a certain time after the beginning of the excitation. Similarly, the non-equilibrium conductivity does not disappear immediately after switching off the illumination at stationary and

 (4)

where t1 is the photoconductivity relaxation time after the beginning of illumination (generally speaking t is a measure of the increase or decrease in time of non-equilibrium charge carriers according Eq. (4) or (5), respectively), Dsst is stationary nonequilibrium conductivity. Two linear sections may be marked out for the photoconductivity curves in the coordinates eln(1eDs/Dsst) vs t (Fig. 6). The slopes of these were used to determine photoconductivity relaxation times t1, t2 (Fig. 7). At the initial stage the photoconductivity increase has the relaxation time t1 (Fig. 7a). After about 0.2 s after the start of illumination the photoconductivity relaxation time changes its value to t2 (Fig. 7b), with t2> t1. Both t1 and t2 decrease with increasing temperature. The reason for such behavior of photoconductivity may be a re-distribution of electrons of defect centers. Similar processes were observed in Ge with Cu admixture [36]. High t1 and t2 values confirm the participation in the photoconductivity relaxation of either shallow sticking levels or even recombination barriers. According to reference [38,39], there are currently two mutually complementary models for description of the processes of longterm photoconductivity relaxation. The first model is based on the concept of trapping the non-equilibrium carriers on heavily localized states in the vicinity of point defects (traps) or admixtures with deep energy levels in the forbidden gap. The second model is based on the spatial separation of electrons and holes by potential barriers which are formed by huge space non-uniformities near the surface of the crystal, and also by the clusters of defects generated by light or high-energy charged particles. This potential barrier, often called recombination barriers, would need to be overcome by electrons to gain the spatial region containing holes. Generally the separation of relaxation mechanisms between the two models is quite complicated task. Since we investigated the relaxation of photoconductivity caused by light quanta from the admixture region. It is more reasonable to analyze the experimental results using the model of long-term relaxation of trapping free carriers by point localization centers (traps). In that case the photoconductivity relaxation time will not be equal to the lifetime of photoexcited carriers but will significantly exceed it since some electrons will be stored at t-levels. The higher relaxation time compared to inter-band recombination is also caused by the principal channel of the recombination processes on

G.L. Myronchuk et al. / Materials Chemistry and Physics 200 (2017) 250e256

a

b

0.20

c

0.8 0.6

P = 45 mVt P = 150 mVt

0

0.15

0.01

0.10 P = 45 mVt P = 150 mVt

0.05

0.00

0

2

4

6 t, s

8

0.00

10

ln (

ln (

ln (

0

), arb.units

P = 45 mVt P = 150 mVt

0.02

1.0

0.25

0

), arb.units

0.03

), arb.units

254

0

4

8

12 t, s

0.4 0.2

16

0.0

20

0

4

8 t, s

12

16

Fig. 5. Kinetics of photoconductivity rise and decrease for various excitation intensity: a e sample 1; b e sample 2; c e sample 3.

a 0 0.0

0.2

0.4

0.6

0.8

1.0

t, s

T = 100 K T = 140 K T = 180 K T = 260 K

4 3

1

st

2

), arb.units

), arb.units

1

st

2

3

-ln(1-

), arb.units

-ln(1-

1

st

3

4

-ln(1-

T = 100 K T = 140 K T = 180 K T = 220 K T = 280 K

4

b 0 0.0

0.2

0.4 0.6 t, s

0.8

1.0

T = 100 K T = 140 K T = 180 K T = 260 K

2

c 0 0.0

0.2

0.4 0.6 t, s

0.8

1.0

Fig. 6. Kinetics of photoconductivity rise at various temperature: a e sample 1; b e sample 2; c e sample 3.

5

b

a

0.24

4

sample 1 sample 2 sample 3

0.20

3

1

2

,s

,s

0.16

sample 1 sample 2 sample 3

0.12

2

0.08

1

0.04 120

160

200

240

280

120

160

200

240

280

T, K

T, K

Fig. 7. Temporary parameters of the relaxation processes in AgCd2GaS4: a) at the initial stage, b) after the start of illumination.

defect centers [37,38]; in our case r or s recombination centers. Long-term processes of photoconductivity relaxation were observed in the crystals of other semiconductor compounds [30e42]. The photoconductivity relaxation after the end of photoexcitation is characterized by the presence of at least two recombination channels (Fig. 8). The relaxation process in that case may be presented well by the sum of two exponents:



t





t

þ B$exp  Ds ¼ A$exp  t3 t4

 (5)

where A z B z Dsst. The photoconductivity relaxation times t3 and t4 were determined from the slope of the linear sections. According to the calculations, the photoconductivity relaxation has a “fast” and a “slow” temporary components with typical relaxation times presented in (Fig. 9). The estimations have shown that the fast component of the photoconductivity kinetics decreases and is characterized by the lifetime of non-equilibrium carriers t3 < 0.8 s, whereas for the slow component t4 ¼ 20e30 s. The magnitude of the initial fast decrease of conductivity upon switching off the excitation increases with temperature, whereas the characteristic time of the slow process t4 obtained from the slope of the exponential portion of the relaxation

G.L. Myronchuk et al. / Materials Chemistry and Physics 200 (2017) 250e256

1

a

1

255

1

c

T = 100 K T = 140 K T = 180 K T = 220 K T = 280 K

0.1 0.0

0.5

t, s

1.0

0.1

1.5

, arb.units

st

st

st

, arb.units

, arb.units

b

0.1

T = 100 K T = 140 K T = 180 K T = 220 K T = 280 K

0

1

2

t, s

3

4

T = 100 K T = 120 K T = 140 K T = 160 K T = 200 K

0

1

2

t, s

3

4

Fig. 8. Kinetics of photoconductivity decrease at various temperature: a e sample 1; b e sample 2; c e sample 3.

30 0.8

a

0.7

sample 1 sample 2 sample 3

0.6

3

sample 1 sample 2 sample 3

,s

20

4

,s

0.5

b

25

0.4 0.3

15 10

0.2

5

0.1 120

160

200

240

120

280

160

200

240

280

T, K

T, K

Fig. 9. Time parameters of the relaxation processes in AgCd2GaS4: a) fast , b) slow temporary components, respectively.

3.5

curves (Fig. 10) decreases. The photoconductivity relaxation time in the presence of sticking centers decreases with increasing temperature in accordance with equation [37]:

  Nt Etn ; $exp Nc kT

where tn is the lifetime of free non-equilibrium electrons, Nt is the concentration of the sticking centers, Nc is the effective density of electron states in the conduction band, Etn is the energy distance from the bottom of the conduction band to the particular electron trap level. The photoconductivity relaxation time decreases with increasing temperature approaching tn value. Respectively, the lower temperature and the higher trap concentration in the semiconductor, the more the photoconductivity relaxation time differs from the lifetime of free non-equilibrium carriers. Therefore, considering that t[tn in the investigated temperature range, the previous formula may be re-written as:

  N E t ¼ tn $ t $exp tn : Nc kT

3.0

(6)

(7)

This makes possible to evaluate Etn from the temperature dependence of the relaxation time ln t vs 1/T. Obtained t values of all samples of the investigated compounds well fitted by linear dependence in the coordinates ln t vs 1/T

ln( )



t ¼ tn $ 1 þ

sample 1 sample 2 sample 3

2.5 2.0 1.5 1.0

2

4

6 8 -1 1000/T, K

10

Fig. 10. Dependence of the relaxation time of the slow component of photoconductivity relaxation on temperature in the AgCd2GaS4 crystals.

256

G.L. Myronchuk et al. / Materials Chemistry and Physics 200 (2017) 250e256

(Fig. 5). The values of Etn calculated from the slope of these lines were ~0.2 eV. Thus obtained energy depth of t-centers is close to the value of the temperature energy of activation of these centers that was determined from thermo-induced conductivity spectra [43]. 5. Conclusions Temperature dependences of the time kinetics were explored for the AgCd2GaS4 single crystals. The photoconductivity relaxation after the end of photoexcitation is characterized by the presence of at least two recombination channels. The relaxation process in that case may be presented well by the sum of two exponents. The estimations have shown that the fast component of the photoconductivity kinetics decreases and is characterized by the lifetime of non-equilibrium carriers t3 < 0.8 s, whereas for the slow component t4 ¼ 20e30 s. The magnitude of the initial fast decrease of conductivity upon switching off excitation increases with temperature, whereas the characteristic time of the slow process t4 obtained from the slope of the exponential portion of the relaxation curves decreases. References [1] David N. Nikogosyan, Nonlinear Optical Crystals: a Complete Survey, Springer, 2005. [2] P.G. Schunemann, S.D. Setzler, T.M. Pollak, J. Cryst. Growth 211 (2000) 257e264. [3] T. Surek, J. Cryst. Growth 275 (2005) 292e304. [4] Y. Ming-Hua, H. Shih-Jung, C. Guang-Hong, Y. Chang-Wei, C. Pin-Ru, C. HsuehShih, Sol. Energy 132 (2016) 547e557. [5] V.B. Lazarev, Z.Z. Kish, E.Yu Peresh, E.E. Semrad, Slozhnye Khal’kogenidy V Sistemakh AI e BIIecvi (Complex Chalcogenides in AI ebiiecvi Systems), Metallurgiya, Moscow, 1993 (in Russian). [6] P. Grima-Gallardo, Phys. Stat. Sol. (A) 134 (1992) 119e125. [7] P. Grima-Gallardo, M. Quintero, A. Baretto, J. Ruiz, Adv. Mat. Sci. Tech. 1 (1998) 1e12. [8] O.V. Parasyuk, S.V. Voronyuk, L.D. Gulay, G.Ye Davidyuk, V.O. Halka, J. Alloy. Comp. 348 (2003) 57e64. [9] G. Wagner, S. Lehmann, S. Schorr, D. Spemann, Th Doering, J. Sol. State Chem. 178 (2005) 3631e3638. [10] L. Roussak, G. Wagner, S. Schorr, K. Bente, J. Sol. State Chem. 178 (2005) 3476e3484. [11] I.D. Olekseyuk, H.Ye Davidyuk, O.V. Parasyuk, S.V. Voronyuk, V.O. Halka, V.A. Oksyuta, J. Alloy. Comp. 309 (2000) 39e44.

[12] I.D. Olekseyuk, O.V. Parasyuk, O.A. Dzham, L.V. Piskach, J. Sol. State Chem. 179 (2006) 315e322. [13] S.I. Chykhrij, O.V. Parasyuk, V.O. Halka, J. Alloy. Comp. 312 (2000) 189e195. [14] I.D. Olekseyuk, L.D. Gulay, O.V. Parasyuk, O.A. Husak, E.M. Kadykalo, J. Alloy. Comp. 343 (2002) 125e131. [15] I.D. Olekseyuk, O.V. Parasyuk, V.O. Halka, L.V. Piskach, V.Z. Pankevych, YaE. Romanyuk, J. Alloy. Comp. 325 (2001) 167e179. [16] N.V. Pervykhina, V.V. Atuchin, O.V. Parasyuk, Acta Cryst. E 61 (2005) i91ei93. [17] V.V. Atuchin, V.G. Kesler, O.V. Parasyuk, Surf. Rev. Let. 14 (2007) 403e409. [18] A.H. Reshak, S. Auluck, I.V. Kityk, A. Perona, B. Claudet, J. Phys. Condens. Matter 20 (2008) 325213. [19] V.L. Bekenev, V.V. Bozhko, O.V. Parasyuk, G.E. Davydyuk, L.V. Bulatetska, A.O. Fedorchuk, I.V. Kityk, O.Yu Khyzhun, J. Electron. Spectrosc. Relat. Phenom. 185 (2012) 559e566. [20] A.H. Reshak, V.V. Atuchin, S. Auluck, I.V. Kityk, J. Phys. Condens. Matter 20 (2008) 325234. [21] I.D. Olekseyuk, O.V. Parasyuk, O.M. Yurchenko, V.Z. Pankevych, V.I. Zaremba, Y.E. Romanyuk, R. Valiente, J. Cryst. Growth 279 (2005) 140e145. [22] V.V. Atuchin, V.Z. Pankevich, O.V. Parasyuk, N.V. Pervukhina, L.D. Pokrovsky, V.G. Remesnik, V.N. Uvarov, V.I. Pekhnyo, J. Cryst. Growth 292 (2006) 494e499. [23] J. Fan, W. Carrillo-Cabrera, L. Akselrud, I. Antonyshyn, L. Chen, Y. Grin, Inorg. Chem. 52 (2013), 1067e11074. [24] W.G. Zeier, C.P. Heinrich, T. Day, C. Panithipongwut, G. Kieslich, G. Brunkla-Us, G.J. Snyder, W. Tremel, J. Mater. Chem. A 2 (2014) 1790e1794. [25] E.J. Skoug, J.D. Cain, D.T. Morelli, Appl. Phys. Lett. 98 (2011) 261911. [26] A.O. Fedorchuk, O.V. Parasyuk, I.V. Kityk, Mater. Chem. Phys. 139 (2013) 92e99. [27] V.L. Bonch-Bruyevich, I.P. Zvyagin, R. Kayper, Elektronnaya Teoriya Neuporyadochennykh Poluprovodnikov, Nauka, Moskva, 1981. [28] N.F. Mott, E.A. Davis, (M. Mir 1974), 472 p. [29] F. Urbach, Phys. Rev. 92 (1953) 1324. [30] H.P.D. Lanyon, Phys. Rev. 130 (1963) 134e143. [31] J. Tauc, North-Holland, Amsterdam, 1970, p. 277. [32] Z. Pankov, Optical Processes in Semiconductors, Mir, Moscow, 1973, p. 457. [33] T. Nakanishi, K. Ito, Sol. Energy Mater. Sol. Cells 35 (1994) 171. [34] A. Adachi, A. Kudo, T. Sakata, Bull. Chem. Soc. Jpn. 68 (1995) 3283. [35] J.E. Jaffe, A. Zunger, Phys. Rev. B 29 (1984) 1882e1906. [36] Z.U. Jabua, A.V. Gigineishvili, I.G. Tabatadze, I.L. Kupreishvili, PSE 9 (1) (2011) 44e47. [37] R.H. Bube, Photoelectronic Properties of Semiconductors, Cambridge University Press, 1992, ISBN 0-521-40681-1, p. 318. [38] V.S. Vavilov, P.K. Eaimiu, J.E. Zardas, UFN, 169, 1999, pp. 209e210. [39] M.S. Yunusov, M. Karimov, B.L. Oksengendler, FTP 32 (1998) 264e267. [40] U. Pal, S. Sahat, S.K. Datta, A.K. Chaudhurit, Semicond. Sci. Technol. 5 (1990) 429e434. [41] M.A. Iovu, M.S. Iovu, E.P. Colomeico, J. Opt. Adv. Mat. 5 (5) (2003) 1209e1214. [42] A.S. Krymus, G.L. Myronchuk, O.V. Parasyuk, G. Lakshminarayana, A.O. Fedorchuk, A. El-Naggar, A. Albassam, I.V. Kityk, Mater. Res. Bull. 85 (2017) 74e79. [43] L.V. Bulateckaja, V.V. Bozko, G.Ye Davidyuk, O.V. Parasyuk, FTP 42 (5) (2008) 522e527.