Preparation, crystal structure, thermal analysis, scanning electron microscopy and optical band-gaps of Cu₂GeTe₄ and Cu₂SnTe₄ alloys

Pedro Grima-Gallardo¹*, Ricardo Peña¹, Luis Nieves¹, Gustavo Marcano¹, Miguel Quintero¹, Ekadink Moreno¹, Jian-Han Zhang², Jacilynn A. Brant² & Jennifer A. Aitken²

1: Centro de Estudios en Semiconductores (C.E.S.). Departamento de Física, Facultad de Ciencias, Universidad de Los Andes, Mérida 5101, Venezuela.

2: Department of Chemistry and Biochemistry, Duquesne University, Pittsburgh, Pennsylvania 15282, USA.

*e-mail: peg@ula.ve

ABSTRACT

Polycrystalline samples (weight ~ 1g) of Cu₂GeTe₄ and Cu₂SnTe₄ alloys were prepared by the usual melt and anneal method and the products characterized by X-Ray Diffraction (XRD), Differential Thermal Analysis (DTA), Scanning Electron Microscopy (SEM) and Optical Diffuse Reflectance UV/VIS/NIR Spectroscopy techniques. It was found that: a) Cu₂GeTe₄ and Cu₂SnTe₄ crystallize in an orthorhombic structure (s.g. Imm2; Nº 44) with lattice parameters a=5.9281(4) Å, b=4.2211(6) Å, c=12.645(5) Å and a=6.0375(6) Å, b= 4.2706 (3) Å, c=12.844(1) Å, respectively; b) both alloys show two thermal transitions: 762 and 636K upon heating and; 700 and 578K upon cooling for Cu₂GeTe₄; 702 and 636K upon heating and; 650 and 590K upon cooling for Cu₂SnTe₄; c) both alloys present large deviations of stoichiometry for the cations Cu (~35%), Ge (7.2%) and Sn (26.4%) and minor deviation within the experimental error, for the anion Te; and, d) the measured optical band gaps were 0.63 and 0.53 eV for Cu₂SnTe₄ and Cu₂GeTe₄, respectively.

Keywords: Semiconductor alloys, X-Ray Diffraction, Differential Thermal Analysis (DTA), Scanning Electron Microscopy (SEM), Optical Diffuse Reflectance UV/VIS/NIR Spectroscopy, Cu₂GeTe₄ and Cu₂SnTe₄

Preparación, estructura cristalina, análisis térmico, microscopia electrónica de barrido y brecha óptica de energía de las aleaciones Cu₂GeTe₄ y Cu₂SnTe₄

RESUMEN

Se prepararon muestras policristalinas (peso ~ 1 g) de las aleaciones Cu₂GeTe₄ y Cu₂SnTe₄ por el método de fusión y recocido y los productos caracterizados por las técnicas de Difracción de Rayos X (DRX ) , Análisis Térmico Diferencial (ATD), Microscopía Electrónica de Barrido (MEB) y espectroscopía de reflectancia óptica difusa UV / VIS / CIR. Se encontró que : a) Cu₂GeTe₄ y Cu₂SnTe₄ cristalizan en una estructura ortorrómbica ( g.e. Imm2; Nº 44 ) con parámetros de red a = 5,9281 (4) Å , b = 4.2211 (6) Å , c = 12.645 (5) Å y a = 6.0375 (6) Å , b = 4,2706 (3) Å , c = 12.844 (1) Å , respectivamente ; b) ambas aleaciones muestran dos transiciones térmicas : 762 y 636K al calentar y ; 700 y 578K tras el enfriamiento para Cu₂GeTe₄ ; 702 y 636K al calentar y ; 650 y 590K tras el enfriamiento para Cu₂SnTe₄ ; c) ambas aleaciones presentan importantes desviaciones estequiométricas en sus cationes: Cu (~35%), Ge (7.2%) and Sn (26.4%) y menor que el error experimental para el anión Te; y d) las brechas ópticas de energía medidas fueron 0.63 y 0.53 eV para Cu₂SnTe₄ y Cu₂GeTe₄, respectivamente.

Palabras Claves: Aleaciones semiconductoras, Difracción de Rayos X (DRX), Análisis Térmico Diferencial (ATD), Microscopía Electrónica de Barrido (MEB), Espectrocopía de Reflectancia Óptica Difusa, Cu₂GeTe₄ y Cu₂SnTe₄

Recibido: 05-06-2014; Revisado: 15-12-2014

Aceptado: 07-01-2015; Publicado: 01-02-2015

1. INTRODUCTION

The Cu₂-IV-VI₃ and Cu₂-IV-VI₄ alloys (IV: Ge, Sn; VI: Se, Te) belong to the general Cu-IV-VI system and are located on the (Cu₂IV)_1-xVI_x tie line, at x=3/4 and x=4/5, respectively, as it is showed in Figure 1. These alloys are candidates for applications in solar cells, thermoelectrics conversion and electro-optic devices [1-11].

Recently, our group [12] has investigated the Cu₂(Ge_1-xSn_x)Se₄ alloys system and determined that Cu₂GeSe₄ and Cu₂SnSe₄ crystallize in orthorhombic and cubic structures, respectively. The crystal system and lattice parameters of Cu₂GeSe₄ coincide with the high-temperature phase of Cu₂GeSe₃, whereas the cubic phase of Cu₂SnSe₄ coincides with the room temperature cubic phase of Cu₂SnSe₃. According to the phase diagram of the (Cu₂Ge)_1-x(Se)_x tie line (Berger et al [13]; also see Figure 8 in [12] ) the region in the composition range 3/4≤x≤4/5 is single phase, i.e. Cu₂GeSe₃ and Cu₂GeSe₄ have the same crystal structure. However, there is a problem with this phase diagram: Cu₂GeSe₃ has a solid-to-solid phase transition at high temperature from the tetragonal structure (labeled as t₁ in the phase diagram) to an orthorhombic structure (that henceforth we will call t₂) that has not been taken into account in the diagram.

The analogous tellurium systems have also been studied. The structural properties of the Cu₂GeTe₃ alloy have been recently published by Delgado et al [14] and Villarreal et al [15] with the observation of an orthorhombic crystal structure, crystallizing in the space group Imm2 (Nº 44), with lattice parameters a=5.9261(2)Å, b=4.2115(2)Å, and c=12.641(1)Å; on the other hand, the preparation and crystal structure of Cu₂SnTe₃ has been published by Delgado et al [16] who also reported an orthorhombic crystal structure in the space group Imm2 (Nº 44), with lattice parameters a=6.043(1)Å, b=4.274(1)Å, and c=12.833(4)Å. Previously, Sharma et al (1977) [17] reported that Cu₂GeTe₃ and Cu₂SnTe₃ are two-phases solids with eutectic type microstructure; for Cu₂GeTe₃ they found a tetragonal structure with lattice parameters a=5.959Å and c=11.858Å and for Cu₂SnTe₃ a cubic disordered structure, with lattice parameter a=6.094Å.

From thermal analysis, Dovletov et al [18] reported that no ternary compounds were found in the Cu₂Te- SnTe binary system, but Palatnik et al (1961) [19], Rivet et al (1963) [20], Averkieva et al (1965) [21], Rivet et al (1965) [22] and Carcaly et al (1975) [23], (1977) [24] affirm the existence of a ternary phase which forms peritectically at 780-785K.

In this work we report the preparation and characterization of polycrystalline samples of Cu₂GeTe₄ and Cu₂SnTe₄.

2. EXPERIMENTAL PART

2.1 Preparation

Cu₂GeTe₄ and Cu₂SnTe₄ were synthesized using the melt and anneal technique. Stoichiometric quantities of Cu, Ge, Sn and Te elements with purity of 99.99% were charged in an evacuated synthetic silica glass ampoule, which was previously subjected to pyrolysis in order to avoid reaction of the starting materials with silica glass. Then, the ampoule was sealed under vacuum (~10^-4 Torr) and the fusion process was carried out inside a furnace (vertical position) heated up to 1500K at a rate of 20K/h, with a stop of 48 h at 722.5K (melting temperature of Te) in order to maximize the formation of binary species at low temperature and minimize the presence of unreacted Te at high temperatures. The ampoule was shaken using a mechanical system during all the heating process in order to help the complete mixing of all the elements. The maximum temperature (1500K) was kept for other 48 hours with the mechanical shaking system on. Then, the mechanical shaking system was turning off and the temperature was gradually lowered, at the same rate of 20K/h, until 873K. The ampoule was held at this temperature for a period of 30 days. Finally, the sample was cooled to room temperature at a rate of 10K/h. The obtained ingots were bright gray in color and homogeneous to the eye.

2.2 X-Ray Powder Diffraction

A small amount of each compound was gently ground in an agate mortar and sieved to a grain size of less than 38 μm. Each sample was mounted on a zero-background specimen holder for the respective measurement. X-ray powder diffraction patterns of the samples were recorded using a D8 FOCUS BRUKER diffractometer operating in Bragg- Brentano geometry and equipped with a copper Xray tube (CuKα radiation: λ=1.5406 Å, 40 kV and 40 mA) using a nickel filter and one the dimensional LynxEye detector. A fixed antiscatter slit of 8 mm, receiving slit of 1 mm, soller slits of 2.5° and a detector slit of 3 mm were used for the diffraction optics. Data were collected from 2 to 140° (2θ) with a step size of 0.02° (2θ) and a counting time of 0.4 s/step.

2.3 Differential Thermal Analysis

Phase transition temperatures were obtained from differential thermal analysis (DTA) measurements, in the temperature range of 300 to 1500K, using a Perkin-Elmer DTA-7. The instrument was calibrated using aluminum and gold as references. The charge was a powdered alloy of approximately 100-mg in weight. Both heating and cooling runs were carried out on each sample, the average rates of these runs were approximately 10 K/min. The error in determining these temperatures is about ±10K. The temperature values of the thermal transitions were obtained using the intercept of the base line with the beginning of the corresponding peak.

2.4 Scanning Electron Microscopy (SEM) and Energy Dispersive Spectroscopy (EDS)

SEM/EDS was performed on a Hitachi S-3400N scanning electron microscope equipped with a Bruker Quantax model 400 energy dispersive spectrometer using an XFlash^® 5010 EDS detector with a 129 eV resolution. Samples were mounted on double-sided carbon tape affixed to an aluminum specimen holder. EDS spectra were collected using a working distance of 10 mm and an accelerating voltage of 15 kV for 3 min live time.

2.5 Optical Diffuse Reflectance UV/VIS/NIR Spectroscopy

Optical diffuse reflectance spectra of the Cu₂GeTe₄ and Cu₂SnTe₄ were obtained using a Varian Cary 5000 UV/VIS/NIR spectrometer equipped with a Harrick Praying Mantis diffuse reflectance accessory that uses elliptical mirrors. Each sample was ground and placed into a sample holder to a depth of 3 mm. Barium sulfate (Fisher, 99.92%) was used as a 100% reflectance standard. Data were collected from 2500 to 200 nm at a scan rate of 600 nm/min. Using the Kubelka-Munk transformation, α_KM / s = (1-R)²/(2R), the raw reflectance (R) was converted to a relative absorption (α_KM) since the scattering coefficient, s, is unknown [25]. The Urbach energy was also obtained by fitting the optical data to the functional form α= A exp(E-E_g/ E_u), where A is a constant, E is the photon energy in eV, E_g is the band gap energy, and E_u is the Urbach energy [26].

3. RESULTS AND DISCUSSION

In Figures 2 and 3, the experimental X-ray powder diffraction patterns of Cu₂GeTe₄ and Cu₂SnTe₄ are displayed. Sharp diffraction peaks were observed in the range of 20-140º 2Q indicating good crystallinity and that the samples had reached thermal equilibrium during synthesis.

For both alloys, the diffraction patterns can be fully indexed with two phases: one with the orthorhombic crystal structure, space group Imm2, N^o 44, Z=2, and a secondary one identified as Cu₂Te. The lattice parameters (Tables I and II) were calculated using the software Dicvol04 [27]. Those obtained for the orthorhombic phase are very close with those reported previously for the analogous ternaries Cu₂GeTe₃ and Cu₂SnTe₃ [2-4] (see Table III).

With respect to the secondary phase, Cu₂Te does not belongs to the (Cu₂Ge)_1-xTe_x or (Cu₂Ge)_1-xTe_x tie lines, so it is probable that the mechanical shaking during the heating process was not enough to dissolve it.

DTA measurements are showed in Figure 4. Two thermal transitions are observed for both alloys, Cu₂GeTe₄ and Cu₂SnTe₄. In the case of Cu₂GeTe₄, the thermogram shows, transitions occurring at 762 and 636K in the heating cycle, and transitions at 700 and 578K in the cooling cycle. Thermal transitions for Cu₂SnTe₄ were observed at 702 and 636K upon heating and 650 and 590K upon cooling. It is evident that the overheating (or supercooling) effect produces the positive difference of (T_f -T_s) where T_f and T_s are the fusion and solidification temperatures, respectively.

The first thermal event corresponds to the solid-tosolid transition, i.e. from the room temperature (tetragonal?) to the high temperature phase (orthorhombic?), and the second thermal event correspond to the melting (or solidification) of the solid (liquid) phase. The shape of the peaks in the cooling cycle suggests that the sequence of transitions for Cu₂GeTe₄ are liquid-solid₁-solid₂, whereas for Cu₂SnTe₄ are liquid-solid₁+liquid-solid₂.

The well-known Grimm-Sommerfeld condition for semiconductors, states that a compound must have an average of 4 valence electrons per atom. In the case of Cu₂GeTe₄ and Cu₂SnTe₄, Cu is bivalent (Cu⁺²), Ge (or Sn) is tetravalent (Ge⁺⁴ or Sn⁺⁴) and Te is hexavalent (Te⁺⁶). The sum gives 32 electrons in total. However, there are only seven atoms (2Cu + 1 Ge + 4 Te); therefore, it is necessary to have a vacancy in the crystal structure in order to have a total 8 “atoms” for the 32 electrons, giving an average of 4 electrons per atom. Thus, the formula units must be written as □-Cu₂GeTe₄ and □-Cu₂SnTe₄.

Scanning electron microscopy (SEM) coupled with energy dispersive spectroscopy was used to compare the experimental composition to the nominal stoichiometry. For Cu₂GeTe₄, measurements were performed in three different points of the sample while for Cu₂SnTe₄ data were collected on four points. The results are displayed in Table V where the column “experimental” represents the mean value of all measurements for each sample.

We observe stoichiometric deviations that are larger than the experimental error generally accepted for this technique which is ~ 10%. Moreover, both alloys show a coincidence of ~35% excess Cu which discards any experimental error. Cu₂GeTe₄ has a Ge-excess of 7.2% whereas Cu₂SnTe₄ has a Sn-excess of 26.4%. With respect to Te both alloys have a little deficiency that is less than the experimental error. These deviations from stoichiometry must be studied in relation to a better understanding of the phase diagrams of these alloys, research that is only just now beginning.

Finally, the optical energy gaps were obtained by Optical Diffuse Reflectance UV/VIS/NIR Spectroscopy. Optical diffuse reflectance data were collected for ground samples of Cu₂GeTe₄ and Cu₂SnTe₄. The spectra exhibit absorption edges indicative of narrow bandgap semiconductors in agreement with their black color, see Figure 5. Tails commonly observed on optical absorption edges originate from defects within the crystal structure that induce defect states in the electronic band structure. These defect states near the valence band maximum and conduction band minimum create a smearing of the band edge that is termed the Urbach tail [28-29]. When estimating the bandgap of a semiconductor it is proper to model the Urbach tail and exclude this region from the fitting of the band edge when determining the bandgap [30]. The Urbach tail appears as a linear region at the low energy region of the absorption edge when the log of the absorption is plotted as a function of energy [26]. Fitting the slope of this linear region yields the Urbach energy, which was determined to be 0.288(1) eV and 0.3795(5) eV for Cu₂GeTe₄ and Cu₂SnTe₄ respectively.

In semiconductor compounds, bandgaps may be of direct or indirect origin, depending on the location(s) of the valence band maximum and the conduction band minimum in k-space. Direct bandgap materials exhibit relatively sharp absorption edges, while indirect bandgap compounds present a more gradual onset of the absorption edge [28]. In order to determine the nature of the bandgap transition in the compounds presented here, the optical absorption edge data, for energies greater than those determined for the Urbach tail region, were fit to the function for a direct bandgap semiconductor, α = A(E-E_g)^1/2/E, and to the function for an indirect bandgap semiconductor, α = A(E-E_g)²/E, where A is a constant, E is the photon energy in eV, and E_g is the band gap energy [28]. Based on these fits, it was found that the absorption edge of Cu₂SnTe₄ showed a larger range of linearity using the direct function rather than the indirect function. Accordingly, Cu₂SnTe₄ is assigned a direct bandgap of 0.63 eV, which corresponds to ~1970 nm in the near infrared region of the electromagnetic spectrum. We are hesitant to definitely designate the bandgap of Cu₂GeTe₄ as direct or indirect, since the band edge lies near the end of the range of our instrument. However, Cu₂GeTe₄ appears to have a direct transition and fitting with the function for a direct bandgap yields 0.53 eV, which corresponds to ~2339 nm.

Telluride compounds in general can be semiconducting or metallic depending upon their composition and structure. The bandgaps reported here are narrower than those of some telluride compounds, such as CdTe, E_g~1.5 eV [28], an ideal value for use in solar cells. On the other hand, Cu₂SnTe₄ and Cu₂GeTe₄ possess wider bandgaps than some other tellurides such as PbTe [28], RbHgSbTe₃ [31], BaBiTe₃[32] and Bi₂Te₃ [32], which all have E_g<0.5 eV. CdAgTe₃ possess E_g=0.65 eV [33], which is very close to that of Cu₂SnTe₄.

4. CONCLUSIONS

Polycrystalline samples of Cu₂GeTe₄ and Cu₂SnTe₄ alloys were prepared by the melt and anneal technique. Both alloys crystallize in an orthorhombic structure together with traces of a Cu₂Te secondary phase. The melting transition of Cu₂GeTe₄ was congruent whereas Cu₂SnTe₄ melts incongruently. While both alloys present large stoichiometric deviations for the cations Cu, Ge and Sn, the stoichiometry of the anion, Te, is close to the nominal composition. The band-gaps are located in the near-IR region of the electromagnetic spectrum. In view of the results these alloys could have applications as detectors in the near infrared radiation.

5. ACKNOWLEDGEMENTS

The Venezuelan authors want to thanks to CDCHTA-ULA grant C-1885-14-05-B.

6. REFERENCES

1. Berger, L.I. and Prochukhan V.D. “Ternary Diamond-like Semiconductors”. Consultants Bureau, New York, 1969, pp. 55-63.        [ Links ]

2. Hamakawa, Y. “Thin-film Solar Cells: Next Generation Photovoltaics and its Applications”. Springer-Verlag Berlin Heidelberg (2004).        [ Links ]

3. Adelifard M., Mohagheghi M.M.B. and Eshghi H. Phys. Scr. 2012; 85: 035603.        [ Links ]

4. Kuku T.A. and Fakolujo O.A. Solar Energy Materials, 1987; 16: 199–204.        [ Links ]

5. Kuku T.A. and Fakolujo O.A., Proceedings of the 3rd International Symposium on Optical and Optoelectronic Sciences of Engineering 14-18 April, 1986. Innsbruck, Austria. SPIE Vol. 653 pp.321-325, (1986).        [ Links ]

6. Tiwari, D. and Chaudhuri, T. K., Solid State Physics, Procceding of the 55th DAE Solid State Physics Symposium, 26-30 December 2010, Manipal, India. AIP Conference Proceedings, Vol. 1349, pp. 1295-1296.        [ Links ]

7. Berg D.M, Djemour R, Gütay L, Zoppi G., Siebentritt S & Dale P.J. Thin Solid Films, 2012; 520: 6291–6294.        [ Links ]

8. Morelli, D.T. and E.J. Skoug. MRS Proceedings Spring Meeting; Symposium N: “Materials and Devices for Thermal to Electric Energy Conversion”. Vol. 1166, Editors: Yang J., Nolas G.S., Koumoto K., Grin Y. (2009).        [ Links ]

9. Cho J.Y., Shi X., Salvador J.R., Meisner G.P., Yang J., Wang H., Wereszczak A.A., Zhou X. and Uher C. Physical Review B, 2011; 84: 085207.        [ Links ]

10. Samanta L.K. Phys. Stat. Sol. (a) 1987; 100: K93.        [ Links ]

11. Ibáñez M., Cadavid D., Anselmitamburini U., Zamani R., Gorsse S., Li W., López A.M., Morante J.R., Arbiol J. and Cabot A.J. Material Chemistry A, 2013; 1421-1426.        [ Links ]

12. Peña R., Grima-Gallardo P., Nieves L., Marcano G., Quintero M., Moreno E., Ramos M.A., Henao J.A. and Briceño J.M. Adv. Mat. Sci. & Technol. 2013; 7: 23-38.        [ Links ]

13. Berger L.I. and Kotina E.K. Inorg. Mater. 1973; 9: 330-322.        [ Links ]

14. Delgado G.E., Mora A.J., Pirela M., Velásquez-Velásquez A., Villarreal M. and Fernandez B.J. Phys. Stat. Sol. (a) 2004; 201: 2900-2904.        [ Links ]

15. Villarreal M., Fernandez B.J., Pirela M. and Velásquez-Velásquez A. Revista Mexicana de Física, 2003; 49 Suplemento 3: 198-200.        [ Links ]

16. Delgado G.E., Mora A.J., Marcano G. and Rincon C. Cryst. Res. Technol. 2008; 43: 433-437.        [ Links ]

17. Sharma B.B., Ayyar R. and Singh H. Phys. Stat. Sol. (a) 1977; 40: 691.        [ Links ]

18. Dovletov K. and Tashliev K. Izv. Akad. Nauk. Turkm. SSR Ser. Fiz.-Tekhn. Khim. Geol. Nauk 1978; 2: 24-27.        [ Links ]

19. Palatnik L.S., Komnik Yu. F., Koshkin V.M. and Belova E.K. Dokl. Akad. Nauk. SSSR 1961; 137: 68-71.        [ Links ]

20. Rivet J. Bull. Soc. Chim. Fr. 1963; 12: 2703.        [ Links ]

21. Averkieva G.K., Vaipolin A.A. and Goryunova N.A. Sov. Research in New Semiconductor Materials, 1965; 26-34.        [ Links ]

22. Rivet J. Ann. Chim. (Paris) 1965; 10 (5-6): 243- 270.        [ Links ]

23. Carcaly C., Rivet J. and Flahaut J.J. Less- Common Met. 1975; 41(1): 1-18.        [ Links ]

24. Carcaly C., Rivet J. and Flahaut J.J. Less- Common Met. 1977; 51(1): 165-171.        [ Links ]

25. Kubelka P., Monk F. Z. Technol. Phys. 1931; 12: 593–601.        [ Links ]

26. Urbach F. Phys. Rev. 1953; 92: 1324.        [ Links ]

27. Boultif A. and Louër D. J. Appl. Cryst. 1991; 24: 987-993.        [ Links ]

28. Pankove J.I. Optical Processes in Semiconductors, Dover Publication, New York, 1971.        [ Links ]

29. Choudhury B., Dey M., Choudhury A. Int. Nano Lett. 2013; 3: 25.        [ Links ]

30. Malingowski A. C., Stephens P. W., Huq A., Huang Q., Khalid S., Khalifah P.G. Inorg. Chem. 2012; 51: 6096-6103.        [ Links ]

31. Li J., Chen Z., Wang X., Proserpio D.M. J. Alloy Compds. 1977; 262-263: 28-33.        [ Links ]

32. Chung D.-Y., Jobic S., Hogan T., Kannewurf C. R., Brec R., Rouxel J., Kanatzidis M.G. J. Amer. Chem. Soc. 1997; 119: 2505-2515.        [ Links ]

33. Li J., Guo H.-Y, Zhang X., Kanatizidis M.G. J. Alloys Compds. 1995; 218: 1-4.        [ Links ]

34. Tauc J. MRS Bull. 1969; 3: 37-46.        [ Links ]