Study of recombination process in Cu2ZnSnS4 thin film using two-wavelength excited photoluminescence

Mohammad Abdul Halim1, Muhammad Monirul Islam1, Xianjia Luo1, Chong Xu1, Takeaki Sakurai1, Noriyuki Sakai2, Takuya Kato2, Hiroki Sugimoto2, Hitoshi Tampo3, Hajime Shibata3, Shigeru Niki3,

Katsuhiro Akimoto1

1Institute of Applied Physics, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki, 305-8573, Japan, 2Energy Solution Business Center, Showa Shell Sekiyu K.K., 2-3-2 Daiba, Minato-ku, Tokyo 135-8074, Japan, 3National Institute of Advanced Industrial Science and Technology (AIST), 1-1-1 Umezono,

Tsukuba, Ibaraki 305-8568, Japan.

Abstract — Room-temperature two-wavelength excited

photoluminescence (PL) measurements have been performed in the kesterite Cu2ZnSnS4(CZTS) and Cu2ZnSn(S,Se)4 (CZTSSe) thin film absorbers. A defect level at 0.8 eV from the valence band and its properties are investigated. Two light sources of 635 nm and 1550 nm diode lasers, respectively, were used for above bandgap and 0.8 eV defect level excitation. The two-wavelength excited PL intensity was stronger than that only above-gap laser irradiation for the CZTS specimen. This phenomenon strongly suggests that the 0.8eV defect level acts as recombination center at room temperature. On the other hand, this defect may act as a trap in lower gap CZTSSe. Key Words — Cu2ZnSnS4, kesterite, recombination center,

thin film, two-wavelength photoluminescence.

I. INTRODUCTION

The promising thin film technologies based on Cu(In,Ga)Se2, (CIGS) and CdTe are recently in the commercial stage. Despite the promise of these absorber materials, the scarcity of elements such as indium and tellurium may limit their potential for the future demand of multi-terawatt level [1, 2] power production. Moreover, indium and gallium are expensive and consequently hard to combine with large scale production at relatively low costs. In order to overcome these limitations, semiconductor materials like copper zinc tin sulfide, Cu2ZnSnS4 (CZTS), in which indium and gallium of CIGS is replaced with the zinc and tin, are of great interest. The quaternary kesterite p- CZTS absorber with high absorption coefficient (~104 cm-1) [3, 4] is an excellent candidate for thin film-solar cells. The direct band gap of CZTS is close to the optimal single-junction value (∼1.5 eV) [5]. Recently the alloy of CZTS and its Se counterpart Cu2ZnSnSe4 (CZTSe), of band gap (∼1.0 eV), [6, 7] Cu2ZnSn(S,Se)4 (CZTSSe) has been used as a solar cell absorber. The tuning band gap of 1.0 to 1.5 eV [6, 7] depending on S, Se composition matches well to the solar spectrum to acquire most of the intensity photons from the solar radiation. All the elements in CZTS are earth-abundant, non-toxic and cost effective which are the main requirements for the production of sustainable solar electricity. With recent development, the efficiency of CZTSSe-based

solar cells has achieved 12.6% in 2013 [8] and for the pure sulfide CZTS solar cells so far have shown 8.4% using thermal co-evaporation of elemental sources [9]. However, these efficiencies are still much lower than the record efficiency of 20.3% for CIGS [10] thin-film solar cells, although a number of material properties, such as crystal structure of kesterite CZTS are very similar to the chalcopyrite CIGS [11]. Therefore CZTS deserves much more attention to improve the efficiency. In general, the efficiencies of thin film solar cells are limited by detrimental defects involved in the absorber materials. By using first principles calculations Chen et al [12] investigated CuZn antisite defect with activation energy of 0.12 eV which is responsible for p-type conductivity. To our best knowledge, a few admittance data with defect energy from 0.12 to 0.2 eV are reported [13, 14, 15] for CZTS and CZTSSe solar cells. But so far experiments on deep defects which strongly limit the efficiency are hardly available. By Transient Photocapacitance Technique (TPC) in our laboratory we observed deep defect centered roughly at 0.8-1 eV for CZTS. In lower band gap of 1.07 eV to 1.20 eV CZTSSe solar cells deep level defect spectra centered roughly at 0.8 eV is reported [16]. Generally, photoluminescence is performed by laser

irradiation with above-gap energy. In this study we have studied recombination properties 0.8 eV from the valence band for both CZTS and CZTSSe specimens, by using the two-wavelength excited PL method. Besides above-gap excitation we introduce another light source corresponding to defect level in the band gap to examine whether it is trap or recombination center. Two- wavelength excited PL, above gap excitation and below gap excitation which enables to saturate the 0.8 eV defect level, was carried out at room temperature.

II. EXPERIMENTAL

The CZTS and CZTSSe absorber layers were grown by sulfurization and sulfo-selenization respectively after metallic precursor stacking by co-sputtering on Mo coated soda lime glass substrate performed by Energy Solution Business Center,

978-1-4799-4398-2/14/$31.00 ©2014 IEEE 2334

Showa Shell Sekiyu KK. PL measurements were carried out at room temperature with a confocal laser scanning microscope using a 635 nm laser for above-gap excitation and a CCD is used for the detection of luminescence light for CZTS sample. In case of lower gap CZTSSe sample an InGaAs-based photomultiplier tube was used to detect the longer wavelength luminescence signals. During the PL measurements, a 0.8 eV (1550 nm) laser beam corresponding to the defect level was irradiated simultaneously together with the 635 nm laser.

III. RESULTS AND DISCUSSION

The trap center in a semiconductor is assumed to be a defect in which a carrier is trapped or captured for a period of time and then thermally emitted into the band from which it originated. On the other hand prior to thermal emission if the captured carrier is annihilated by recombination with an opposite carrier then it is called recombination center. Photoluminescence (PL) is one of the most important methods for the characterization of semiconductor materials with a particular relevance when they are intended for photonic applications like solar cells. This technique probes the electronic structure revealing important information about the radiative and non-radiative recombination mechanisms in these materials.

Fig. 1 Principle of the two wavelength excited PL. (a) Optical process in case of above-gap excitation, (b) two wavelength excitation. The PL intensity for (b) can be expected to be stronger than (a). Figure 1 shows the schematic diagram showing the principle

of the two-wavelength excited PL. The electron distribution

and its recombination processes under the normal above-gap excitation (635 nm Laser) and two-wavelength excitation (635 nm + 1550 nm) are shown in Fig 1(a) and 1 (b) respectively. Figure 2 shows the room temperature PL spectra of CZTS with Cu/Sn ratio of 1.75 measured for normal excitation and two wavelength excitation. There is clear increase in the PL intensity under 1550 nm laser irradiation which indicates a reduction in the non-radiative recombination by filling electrons at the 0.8 eV defect level that is the 0.8 eV defect level works as a recombination center at room temperature. If we consider 0.8 eV defect level as a trap center then the PL intensity should be the same under normal and two-wavelength excitation.

Fig. 2 PL spectra of CZTS under single and two wavelength excitation. The red line is for two-wavelength and the black for single wavelength excitation. To show the difference in PL intensities we introduce three

parameters R, Idouble and Isingle, where R= Idouble/ Isingle, and Idouble, Isingle are the PL intensities under two-wavelength excitation (635 nm + 1550 nm) and only 635 nm laser excitation, respectively. These are the integrated PL intensities estimated as an area of PL spectrum integrated over photon energy. When the 0.8 eV defect works as a trap center, the R value

should be 1 or less than 1, and when the defect level works as a recombination center, the R value should be more than 1 depending on 0.8 eV light intensity. The variation of R values with the irradiation intensity of 0.8 eV laser for the CZTS sample is shown in Fig. 3. It is seen that the R value increases from 1 with 0.8 eV laser

intensity up to a limit of defect level saturation. Therefore after a maximum limit a decline tendency is also observed. So we can consider the 0.8 eV defect level works as a recombination center. The maximum point of the R value corresponds to the full saturation of the 0.8 eV defect level and further stronger 1550 nm laser light irradiation enhances heat

978-1-4799-4398-2/14/$31.00 ©2014 IEEE 2335

generation by phonon excitation which increases phonon assisted non-radiative recombination and results in the R value to be decline [ 17].

Fig. 3 Variation of R values for the CZTS sample with 1550 nm laser intensity along with fixed 635 nm laser excitation.

Fig. 4 Pl intensity variation of CZTSSe under two-wavelength excitation. I1, I2, I3 and I4 are the 0.8 eV laser intensities. In case of CZTSSe sample there is no increase in PL

intensities rather Idouble is less than Isingle after the introduction of 1550 nm laser excitation. For weaker secondary excitation I2 figure 4 shows that the intensities are almost same in both cases. But the difference in intensities gradually increases with the increase in power of induced secondary 1550 nm laser. Thus for lower band gap (~1.10 eV) CZTSSe as estimated from the external quantum efficiency (EQE) data, the 0.8 eV defect level acts as trap center. While in wider gap (~1.53 eV) CZTS this level acts as recombination center i.e., before the thermal emission of captured minority carrier (electron) to the

conduction band it is recombined with a majority carrier (hole) and recombination takes place. This is may be one of the reasons that the efficiency of CZTSSe based solar cells is higher than CZTS [8, 9, 18]. Due to the position of 0.8 eV defect closer to the mid-gap in CZTS it acts as efficient recombination center which is detrimental to the performance of materials.

IV. CONCLUSION

To examine the properties of a particular defect level we have introduced its equivalent energy as a secondary excitation source along with the above-gap primary excitation. In this study we have investigated the 0.8 eV defect level for wider gap CZTS and smaller gap CZTSSe samples by two-wavelength excitation PL measurement. The results obtained for CZTS in this study suggest that ~0.8 eV defect level works as a recombination center at room temperature while in CZTSSe this level works as a trap center.

REFERENCES

[1] B. A. Andersson, “Materials availability for large-scale thin-film photovoltaics”, Prog. Photovolt: Res. Appl. 8, 61–76(2000).

[2] C. Wadia, A.P. Alivisatos and D.M. Kammen, “Materials availability expands the opportunity for large-scale photovoltaics deployment” Environ. Sci. Technol. 43, 2072–2077(2009).

[3] J. S. Seol, S. Y. Lee, J. C. Lee, H. D. Nam and K. H. Kim, “Electrical and optical properties of Cu2ZnSnS4 thin films prepared by rf magnetron sputtering process” Sol. Energy Mater. Sol. Cells 75, 155 (2003).

[4] K. Ito and T. Nakazawa, “Electrical and Optical Properties of Stannite-Type Quaternary Semiconductor Thin Films” Jpn. J. Appl. Phys. 27, 2094 (1988).

[5] W. Shockley and H. J. Queisser, “Detailed Balance Limit of Efficiency of pn Junction Solar Cells” J. Appl. Phys. 32, 510 (1961).

[6] S. Chen, X. G. Gong, A. Walsh and S. H. Wei, “Crystal and electronic band structure of Cu2ZnSnX4, X = S and Se) photovoltaic absorbers: First-principles insights” Appl. Phys. Lett. 94, 041903 (2009).

[7] C. Perrson, “Electronic and optical properties of Cu2ZnSnS4 and Cu2ZnSnSe4” J. Appl. Phys. 107, 053710 (2010).

[8] W. Wang, M. T. Winkler, O. Gunawan, T. Gokmen, T. K. Todorov, Y. Zhu and D. B. Mitzi, “Device Characteristics of CZTSSe Thin-Film Solar Cells with 12.6% Efficiency” Adv. Energy Mater. (2013), doi: 10.1002/aenm.201301465.

[9] B. Shin, O. Gunawan, Y. Zhu, N. A. Bojarczuk, S. J. Chey and S. Guha, “Thin film solar cells with 8.4% power conversion efficiency using an earth-abundant Cu2ZnSnS4 absorber” Prog. Photovolt: Res. Appl. 21, 72–76 (2013).

[10] M. A. Green, K. Emery, Y. Hishikawa, W. Warta and E. D. Dunlop, “Solar cell efficiency tables (Version 40)” Prog. Photovoltaics 20, 606 (2012).

[11] S. Schorr, “Structural aspects of adamantine like multinary chalcogenides” Thin Solid Films 515, 5985 (2007).

[12] S. Chen, J. H. Yang, X. G. Gong, A. Walsh and S. H. Wei, “Intrinsic point defects and complexes in the quaternary kesterite semiconductor Cu2ZnSnS4” Phys. Rev. B 81, 245204 (2010).

978-1-4799-4398-2/14/$31.00 ©2014 IEEE 2336

[13] E. Kask, T. Raadik, M. Grossberg, R. Josepson, and J. Krustok, “Deep defects in Cu2ZnSnS4 monograin solar cells” Energy Procedia 10, 261 (2011).

[14] P. A. Fernandes, A. F. Sartori, P. M. P. Salome, J. Malaquias, A. F. da Cunha, M. P. F. Graca and J. C. Gonza´lez, “Admittance spectroscopy of Cu2ZnSnS4 based thin film solar cells” Appl. Phys. Lett. 100, 233504 (2012).

[15] O. Gunawan, T. Gokmen, C. W. Warren, J. D. Cohen, T. K. Todorov, D. Aaron, R. Barkhouse, S. Bag, J. Tang, B. Shin and D. B. Mitzi, “Electronic properties of the Cu2ZnSn(Se,S)4 absorber layer in solar cells as revealed by admittance spectroscopy and related methods” Appl. Phys. Lett. 100, 253905 (2012).

[16] D. W. Miller, C. W. Warren, O. Gunawan, T. Gokmen, D. B. Mitzi and J. D. Cohen, “Electronically active defects in the

Cu2ZnSn(Se,S)4 alloys as revealed by transient photocapacitance spectroscopy” Appl. Phys. Lett. 101, 142106 (2012).

[17] A. Gupta, N. Hiraoka, T. Sakurai, A. Yamada, S. Ishizuka, S. Niki and K. Akimoto, “Characterization of Cu(In,Ga)Se2 grown by MBE by two-wavelength excited photoluminescence spectroscopy” Journal of Crystal Growth, 378, 162-164 (2013).

[18] T. K. Todorov, J. Tang , S. Bag , O. Gunawan , T. Gokmen , Y. Zhu and D. B. Mitzi, “Beyond 11% Effi ciency: Characteristics of State-of-the-Art Cu2ZnSn(S,Se)4 Solar Cells”, Adv. Energy Mater. 3, 34–38 (2013).

978-1-4799-4398-2/14/$31.00 ©2014 IEEE 2337