Insights on the limiting factors of Cu2ZnGeSe4 based solar cells

Insights on the limiting factors of Cu2ZnGeSe4 based solar cells I. Anefnafa,b, S. Aazoua,b,*, Y. Sánchezc, P.Vidal-Fuentes c, R. Fonoll-Rubio c, K. J. Tiwari c, S. Giraldoc, Z. Jehl Li-Kaod, M. Guc c, E. Saucedod and Z. Sekkata,b,e a Department of Chemistry, Faculty of Sciences, Mohammed V University in Rabat, Morocco b Optics & Photonics Center, Moroccan Foundation for Advanced Science, Innovation and Research -MAScIR, Rabat, Morocco c Catalonia Institute for Energy Research - IREC, Sant Adrià de Besòs, Barcelona, Spain d Department Enginyeria Electrònica, Universitat Politècnica Catalunya (UPC), Barcelona, Spain e Department of Applied Physics, Osaka University, 2-1 Yamadaoka, Suita, Osaka, Japan Abstract: Germanium-based wide band gap Kesterite semiconductor Cu2ZnGe(S,Se)4 (CZGeSSe) is considered a very promising absorber compound as top cell in tandem devices. Autonomy to tailor the band gap from ~1.47 eV (Cu2ZnGeSe4-CZGeSe) to ~2.2 eV (Cu2ZnGeS4-CZGeS), as well as non-toxic constituents makes this compound a strong candidate for further scientific exploration. However, the record efficiency of Cu2ZnGeSe4 solar cells is still significantly lower than those of its predecessors Cu2ZnSn (SxSe1-x)4 (CZTSSe), Cu(In,Ga)Se2 (CIGS) and CdTe thin-film solar cells. The comprehensive understanding of the factors limiting the performance of Cu2ZnGeSe4-based solar cells is the purpose of this work, by combining a complete characterization of the morphological, structural, compositional and optoelectronic properties of Cu2ZnGeSe4 absorbers and devices. Besides, an in-depth investigation of the main limitations is carried out, specifically focusing on studying the origin of the large VOC deficit, the main recombination mechanisms, electric transport properties, band tails and possible Cu2ZnGeSe4/CdS band offset effects. The champion CZGeSe solar cell device reported in this work shows an efficiency of 6.42%, Voc of 606 mV, JSC of 17.77mA/cm2 and FF of 59.65%. The results presented here demonstrate that the large voltage deficit of CZGeSe solar cells could be mainly ascribed to a Fermi level pinning at the interface, and modifications of the buffer layer to induce a “spike” at the pn junction could be beneficial. Additionally, low carrier diffusion lengths and lifetimes, along with possible back contact recombination, are identified as the main culprits for the limited carrier collection for low-energy photons. Finally, some strategies are proposed to face and overcome most of these issues and to help improving the CZGeSe performance. 1. INTRODUCTION Cu2ZnSn(SxSe1-x)4 (CZTSSe) is a promising candidate for thin-film photovoltaic applications1, thanks to their advantageous properties including its ideal direct bandgap that can be tuned between 1.0 eV and 1.5 eV, high absorption coefficient (~104 cm-1) and non toxicity make it suitable for solar cell application (single and tandem solar cells)2–6. However, the best reported efficiency, so far, for CZTSSe is 12.6% with bandgap energy of Eg=1.13 eV7, which is about half of the efficiency of the close-cousin CIGS-based solar cell and much lower than the theoretical Shockley-Quiesser limit for CZTSSe, (32.8%).8 One of the focal challenges so far for CZTSSe based solar cells is the poor open circuit voltage (VOC).9–11 The maximum VOC depends mostly on the absorber bandgap but can be strongly affected in case of high recombination or defects pinning the Fermi level. Thus, to compare absorbers with different bandgap, the so called VOC-deficit is introduced, defined as Eg/q-VOC (where Eg is the absorber bandgap energy, VOC is the open circuit voltage and q is the electron charge) and offering a figure of comparison between materials of different bandgaps, and it is around 0.6 eV for the record CZTSSe based solar cell.7 Some studies identify that the main culprits for the large VOC deficit are the intrinsic defects and the band tail.12–14 Furthermore, several causes for the large VOC deficit are suggested: (i) the disordering and lattice defects are acting as an effective recombination centers, and are resulting from the CZTSSe nature, including anti-sites (e.g., SnCu, SnZn), vacancies (e.g., VZn, VSn) and interstitial (e.g., Zni) defects in te system.15,16 Particularly, the anti-site defects (CuZn and ZnCu) are almost unavoidable in CZTSSe compound due to the similar ionic radius of Cu and Zn as well as their chemical properties.17,18 (ii) the tail state formation which is ascribed to high defect concentrations producing electrostatic potential fluctuations19–21 or spatial variations in the crystalline system and/or composition, driving to a nanoscale bandgap fluctuations.22–24 (iii) the nonideal band alignment between the absorber and the CdS, possibly causing high recombination at the absorber/CdS interface and hampering minority carrier transport.25,26 There is a particular research axis targeting the reduction of defect density to boost device performance by optimizing processes of surface and grain boundary passivation27–29. Additionally, the issues of optical losses30, anti-reflection coating and bulk CZTSSe quality31 are also regularly addressed by the community. Nevertheless, the intrinsic point defects are still a critical issue in the bulk absorber.32 The alternative solution to decrease cation disorder in CZTSSe compound is the cationic substitution: Zn with Cd,33 Zn with Ba34, Cu with Ag35,36 and Sn with Ge..37,38 IREC group38,39 proved that partial substitution of Sn with Ge led to a performance enhancement of these thin film solar cells compared to their previous baseline cells. The crystalline quality and grain size were particularly improved, thus boosting the VOC. Table 1 shows a summary of some successful studies introducing Ge in the CZTSSe solar cells. A significant improvement in the cell’s voltage is obtained from very small quantities ˂0.005% (doping) up to around 40% (substitution of Sn with Ge). Table 1 Summary of the beneficial effects of Ge on thin films kesterite solar cells. Technique Ge/Ge+Sn Eff. (%) FF(%) JSC (mA/cm2) VOC (V) Eg (eV) VOC-deficit a (V) Ref Molecular precursor solutions 0.25 11.0 33.6 33.6 583 1.15 0.31 40 Nanocrystal inks printing 0.30 9.4 63.8 31.9 460 1.19 0.48 41 DC Sputtering ˂ 0.005 11.8 66.3 38.3 463 1.04 0.31 37 DC Sputtering ˂ 0.005 10.6 66.7 33.6 473 1.05 0.30 42 DC Sputtering DC Sputtering ˂ 0.005 0.20 10.1 9.2 66.8 65.1 33.6 29.9 453 471 1.04 1.12 0.31 0.40 39 43 Co-evaporation 0.22 12.32 72.7 32.2 527 1.11 0.35 44 Co-evaporation 0.22 10.03 62.7 29.5 543 1.19 0.39 45 Technique Cu/Zn+Ge Eff. (%) FF (%) JSC (mA/cm2) VOC (V) Eg (eV) VOC-deficit a (V) Ref Evaporation Evaporation 1.0 0.9 5.5 7.6 46 58 16 22.8 744 558 1.4 1.36 0.58 0.58 46 47 DC Sputtering Evaporation 1.0 0.9 6.5 8.5 60 55.7 19.6 24.4 556 625 1.45 1.39 0.58 0.55 48 49 DC Sputtering 0.67 6.42 59.65 17.77 606 1.47 0.51 This work a respected to Shockley–Queisser limit (SQ), and calculated by the authors with the available data in the different references. The best efficiency reported so far for pure Ge-kesterite (CZGeSe) thin film solar cell is 8.5%, using metallic stack precursors annealed under H2Se gas.49 The highest open circuit voltage VOC reported so far for pure Ge-kesterite (CZGeSe) is 744 mV with an efficiency of 5.5%.46 Research studies on pure Ge-kesterite devices could be benefits to broaden the spectrum of kesterite application, and specifically for efficient photovoltaic tandem solar cells in combination with Si or a narrow bandgap chalcogenide compound. There are still several limitations to overcome due to the scarcity of studies of Ge-compounds, comparatively rare when considering their Sn counterpart, besides the first studies recently published about physical and fundamental properties of Ge-compounds.50,51 In this context, it is of prime importance to acquire deeper understanding of the defects existing in CZGeSe-based materials and comprehend the key factors limiting CZGeSe-based thin film solar cell performance to push up the CZGeSe solar cells efficiency to a more competitive level. To the best of our knowledge, no previous studies investigating and identifying the defects and limitations for CZGeSe-based thin film solar cell are available. For that purpose, our work intent is to assess and unveil limitations on complete CZGeSe solar cells. In the present study, a complete analysis of morphological, structural and compositional characterization of the prepared CZGeSe thin film based solar cell is presented. The highest performance achieved in this study is 6.42%, with VOC of 606 mV, JSC of 17.77 mA/cm2 and FF of 59.65%. Furthermore, a broad range of material characterizations provides valuable insights on the factor limiting state-of-the-art devices, and specifically on the defects and recombination mechanisms. 2. EXPERIMENTAL SECTION Precursor deposition: Cu2ZnGeSe4 thin-films are prepared by sequential deposition of Cu, Zn and Ge stack precursors by DC-magnetron sputtering (Alliance AC450), with the elemental layer stack order Cu/Zn/Ge onto Mocoated soda-lime glass substrates, 750nm Mo is deposited by DC-magnetron sputtering. The absorber composition is selected to obtain Cu-poor and Zn-rich precursors (Cu/(Zn+Ge) = 0.67 and Zn/Ge = 1.07) as confirmed by XRF measurement. Reactive annealing: The optimized thermal annealing process to growth Cu2ZnGeSe4 absorbers is carried out in a dedicated tubular furnace using graphite box with 100 mg elemental Se and 5 mg of GeSe2. The thermal treatment used consists of a two-step annealing process, first step at 330°C for 30 min, with 20 °C/min ramping, in Ar dynamic pressure (1.5 mbar), followed by a second step at 480°C for 15 min, 20 °C/min ramping, in Ar (1bar) static pressure. Solar cells elaboration: The solar cells are completed after etching the synthesized absorbers with KCN solution (2% w/v, 2min) to clean and remove the secondary phases. Immediately after the chemical etching, ~50 nm of CdS is deposited by chemical bath deposition (CBD) from nitrate salts.52 Then the transparent front electrode i-ZnO (50 nm) and In2O3:SnO2 (ITO, 200 nm, 60Ω/sq) is deposited by DC-pulsed sputtering (Alliance Concept CT100). Finally, the solar cells are mechanically scribed (3×3 mm2) with a microdiamond scriber (OEG MR200). Neither antireflective coating nor metallic grids are used for the optoelectronic characterization of the fabricated devices. Films and devices characterization: The composition and precursor thicknesses is characterized by using X-ray fluorescence (XRF) system (Fischercope XDV) previously calibrated by inductively coupled plasma mass spectrometry (ICP). The scanning electron microscope (SEM) micrographs are acquired with a ZEISS Series Auriga microscope using an acceleration voltage of 5 kV. Raman scattering spectra are performed by FHR 640 and iHR 320 monochromators from Horiba Jobin Yvon both coupled with CCD detectors. The first system is previously optimized for UV-visible spectral range and used with 442 nm and 532 nm excitation wavelengths. The second system is optimized for NIR range and used with 785 nm excitation wavelength. The lasers power densities are in the range 100-150 Wcm-2. The characterizations are performed in a backscattering configuration though probe designed at IREC. Moreover, the in-depth chemical composition profiles of -Ge-kesterite absorber layer is investigated by Glow Discharge Optical Emission Spectroscopy (GDOES) using Horiba Jobin Yvon GD Profiler 2 spectrometer with 4 mm anode diameter. J-V measurements are performed on the completed devices using a calibrated Sun 3000 class AAA solar simulator (Abet Technologies, 25°C, AM1.5G illumination). The EQE spectra of the elaborated solar cells are measured by Bentham PVE300 system calibrated with Si and Ge photodiodes. The capacitance-voltage measurements are done in the dark, at room temperature with a modulation voltage of 50 mV with a Novocontrol Technologies impedance analyzer. J–V–T measurements are performed by a cryostat (Cold-Head model RDK101D Sumitomo Heavy Industries Ltd.) cooled by helium closed cycle compressor (Zephyr HC-4A from Sumitomo cryogenics). 3. RESULTS AND DISCUSSION The SEM micrographs of Cu2ZnGeSe4-based solar cells with the following configuration SLG/Mo/ Cu2ZnGeSe4/CdS/i-ZnO/ITO are investigated. The cross section of the entire Cu2ZnGeSe4 solar cell, as well as the top view of the absorber surface are presented in Figure 1. The top surface image of the absorber reveals large and closely packed grains well connected with each other’s. The cross sectional image of the full device (after chemical etchings and CdS/TCO deposition) indicates large grains in the bulk extending throughout the whole thickness. Unlike CZTSe, CZGeSe seems to not (or just slightly) form the typical bilayer structure, with very few small crystals segregating at the back. In this sense, the cross sectional morphology of CZGeSe appears much better than CZTSe. The average absorber thickness is around 1.45 µm. However, it is very difficult to define the exact thickness of MoSe2 phase, most probably due to their very thin nature related to the lower temperatures required for the synthesis of CZGeSe when compared with CZTSe (480 ºV vs 550 ºC respectively). Fig. 1. SEM micrographs of the Cu2ZnGeSe4 (a) top surface of the absorber and (b) Cross-section of the SLG/Mo/ Cu2ZnGeSe4/CdS/i-ZnO/ITO. The Raman scattering spectra are measured under different excitation conditions to check the presence of any secondary phases. Analysis of the obtained spectra (Figure 2) showed presence of Cu2ZnGeSe4 related peaks only.53 A narrow width of the Raman peaks corresponds to the high crystalline quality of the absorber surface, an observation consistent with the aforementioned SEM images (Figure 1). Once again, this suggests that secondary phases are less prone to be formed in this system than in CZTSe, suggesting a clear advantage of the Ge-compound. However, additional characterization using X-Ray Diffraction spectroscopy would be needed to assess with certainty the phase purity of the film. In that context, the comparatively lower efficiency of Cu2ZnGeSe4 solar cells may not be ascribed to morphological or phase-composition issues, as both appear fully compatible with high efficiency devices. Fig. 2. Raman scattering spectra of Cu2ZnGeSe4 based device measured under different excitation conditions. Numbers indicate the peaks position. One can notice that from the glow discharge optical emission spectroscopy (GDOES) profiles of Cu, Zn, Ge, Se and Mo (Figure 3), the cation distribution is relatively uniform at the top half of the absorber, whereas Cu tends to decrease and Ge to slightly increase (as compared to the other elements) towards the back region of the absorber. This can suggest that Ge-rich phases are accumulated towards the back region, as it is proven in this system where Ge tends to naturally be present with higher concentration at the back area.54 In such case, some secondary phases can be expected at the back and affect the device performance (mainly the JSC and the FF). Additionally, the combined decrease in Cu and Se composition toward the back interface could be ascribed to the presence of Cu-based secondary phases, though further material analysis (see Raman) do not allow to conclude in that regard. A future XRD analysis would shed light on that question, though the CuGeSe3 secondary phase appears to be the likeliest in that context. Fig. 3. Depth composition profiles of Cu2ZnGeSe4 device performed by a glow discharge optical emission spectrometry analysis (GDOES). To further analyze the best device, the current density-voltage (J-V) characteristics and EQE measurement are performed. The device reveals a 6.42 % efficiency with a VOC of 606 mV (Figure 4(a)), it is very important to highlight that neither anti-reflection coating nor metallic grid are used in this work. The optoelectronic parameters of the device are summarized in Table 2. J-V curves exhibit a distinct cross-over between dark and light characteristics (Figure 4(a)), which is often related to defect states at the CZGeSe/CdS interface or in the bulk of CdS layer, and leads to acceptor-like buffer traps states.52 In order to get a deeper insight how to reduce defects and boost device performance, EQE measurement are presented Figure 4(b). In the region 300–500 nm, the collection losses are affected by parasitic absorption from the CdS/TCO. The EQE spectrum reaches a maximum close to 80% at 500 nm, showing that photocarriers generated close to the pn junction are efficiently collected and the p-n interface does not appear to be limiting the current; in the red and the infrared region however, the collection efficiency collapses, indicating either bulk recombination (low carrier diffusion length), incomplete absorption or back contact recombination. Incomplete absorption can be discarded in the largest part of the considered spectral range; indeed, the film’s thickness is close to 1.5 m, which is more than enough for complete absorption considering an absorption coefficient in the range of 104-5. Hence, a low minority carrier diffusion length or back interface recombination appear to significantly hamper the collection of carriers generated deeper in the absorber. As the morphology or the phase purity of CZGeSe appear excellent in both the SEM and Raman analysis, bulk recombination and native defects existing in the material could be important factor limiting the performance of the device. From the EQE results, the optical band gap of CZGeSe is estimated from a Tauc plot (inset Figure 4(b)), and it is found to be 1.47 eV. Fig. 4. J-V characteristics (a) and EQE spectrum with the inset graph shows the band gap extraction by plotting (hυ.ln(1-EQE))2 vs. energy of Cu2ZnGeSe4 (b). To understand further the Cu2ZnGeSe4 absorber nature, the device is measured by C-V characterization under 5.6 KHz. The carrier concentration and the depletion region width are calculated from capacitance–voltage (C-V) profile measured at room temperature (RT) on Cu2ZnGeSe4 device. Figure 5 displays the charge carrier versus the distance to the front interface (NCV (x)). This later is calculated from C-V measurement of the solar cell and can include contributions of superficial and deep defects. The profile demonstrated in Figure 5 reveals a distinct U-shape, this profile was previously attributed to deep defects, interface defects and/or back contact barriers effects55–57. Moreover, the lowest value of U-shape can be taken as an approximation of the charge density. The carrier concentration reflects the properties of Cu2ZnGeSe4 near depletion region is of 1.7. 1015 cm-3 and a charge space region width (SCR) of 311 nm. This carrier concentration is at least one order of magnitude lower than the one reported typically in high efficiency CZTSe devices, indicating that Cu2ZnGeSe4 doping is not as easy. In fact, this lower carrier concentration can explain at least in part the reduction of the VOC. Fig. 5. Charge carrier density versus the distance to the junction interface extracted from CapacitanceVoltage measurements of a Cu2ZnGeSe4 device. Table 2. Summary of the optoelectronic parameters obtained for the best device presented in this work. Parameters Value Units Eff. VOC JSC FF SQ, Voc deficit Voc deficit RS Rsh J0 A Carrier concentration SCR Eg EU Ext. Voc (0 K) 6.5 606 17.8 59.6 0.51 0.81 0.14 8700 1.6.10-4 1.7 1.7.1015 310 1.47 22 0.94 % mV mA/cm2 % V V Ω. cm2 Ω. cm2 mA. cm-2 -- cm-3 nm eV eV eV Table 2 shows a summary of the determined optoelectronic parameters from the J–V characteristics, EQE, temperature-dependent VOC and C–V measurements. The following information on the figures of merit (Figures 4–7) of the fabricated solar cell could be obtained from: References 1. Wong, L. H. et al. Emerging inorganic solar cell efficiency tables (Version 1). J. Phys. Energy 1, 032001 (2019). 2. Guo, Q. et al. Fabrication of 7.2% Efficient CZTSSe Solar Cells Using CZTS Nanocrystals. J. Am. Chem. Soc. 132, 17384–17386 (2010). 3. Liu, F. et al. Kesterite Cu 2 ZnSn(S,Se) 4 Solar Cells with beyond 8% Efficiency by a Sol–Gel and Selenization Process. ACS Appl. Mater. Interfaces 7, 14376–14383 (2015). 4. Bag, S. et al. Low band gap liquid-processed CZTSe solar cell with 10.1% efficiency. Energy Environ. Sci. 5, 7060 (2012). 5. Wang, G. et al. Fabrication of a Cu 2 ZnSn(S,Se) 4 Photovoltaic Device by a Low-Toxicity Ethanol Solution Process. ACS Appl. Mater. Interfaces 5, 10042–10047 (2013). 6. Haass, S. G. et al. 11.2% Efficient Solution Processed Kesterite Solar Cell with a Low Voltage Deficit. Adv. Energy Mater. 5, 1500712 (2015). 7. Wang, W. et al. Device characteristics of CZTSSe thin-film solar cells with 12.6% efficiency. Adv. Energy Mater. 4, 1–5 (2014). 8. Siebentritt, S. Why are kesterite solar cells not 20% efficient? Thin Solid Films 535, 1–4 (2013). 9. Kim, J., Park, S., Ryu, S., Oh, J. & Shin, B. Improving the open-circuit voltage of Cu 2 ZnSnSe 4 thin film solar cells via interface passivation: Open-circuit voltage of Cu 2 ZnSnSe 4 thin film solar cells. Prog. Photovolt. Res. Appl. 25, 308–317 (2017). 10. Tai, K. F. Investigating the Open-Circuit Voltage Deficit in CZTSSe Solar Cells. (2015) doi:10.13140/RG.2.2.20153.98404. 11. Mitzi, D. B., Gunawan, O., Todorov, T. K., Wang, K. & Guha, S. The path towards a high-performance solution-processed kesterite solar cell. Sol. Energy Mater. Sol. Cells 95, 1421–1436 (2011). 12. Gunawan, O., Gokmen, T. & Mitzi, D. B. SunsV OC characteristics of high performance kesterite solar cells. J. Appl. Phys. 116, 084504 (2014). 13. Polizzotti, A., Repins, I. L., Noufi, R., Wei, S.-H. & Mitzi, D. B. The state and future prospects of kesterite photovoltaics. Energy Environ. Sci. 6, 3171 (2013). 14. Mitzi, D. B., Gunawan, O., Todorov, T. K. & Barkhouse, D. A. R. Prospects and performance limitations for Cu–Zn–Sn–S–Se photovoltaic technology. Philos. Trans. R. Soc. Math. Phys. Eng. Sci. 371, 20110432 (2013). 15. Chen, S., Walsh, A., Gong, X.-G. & Wei, S.-H. Classification of Lattice Defects in the Kesterite Cu 2 ZnSnS 4 and Cu 2 ZnSnSe 4 Earth-Abundant Solar Cell Absorbers. Adv. Mater. 25, 1522–1539 (2013). 16. Giraldo, S. et al. Progress and Perspectives of Thin Film Kesterite Photovoltaic Technology: A Critical Review. Adv. Mater. 31, 1806692 (2019). 17. Kattan, N. A., Griffiths, I. J., Cherns, D. & Fermín, D. J. Observation of antisite domain boundaries in Cu 2 ZnSnS 4 by atomic-resolution transmission electron microscopy. Nanoscale 8, 14369–14373 (2016). 18. Chen, S., Gong, X. G., Walsh, A. & Wei, S.-H. Defect physics of the kesterite thin-film solar cell absorber Cu2ZnSnS4. Appl. Phys. Lett. 96, 021902 (2010). 19. Gokmen, T., Gunawan, O., Todorov, T. K. & Mitzi, D. B. Band tailing and efficiency limitation in kesterite solar cells. Appl. Phys. Lett. 103, 103506 (2013). 20. Mendis, B. G. et al. The nature of electrostatic potential fluctuations in Cu 2 ZnSnS 4 and their role on photovoltaic device performance. J. Phys. Conf. Ser. 471, 012014 (2013). 21. Bishop, D. M. et al. Modification of defects and potential fluctuations in slow-cooled and quenched Cu 2 ZnSnSe 4 single crystals. J. Appl. Phys. 121, 065704 (2017). 22. Rey, G. et al. On the origin of band-tails in kesterite. Sol. Energy Mater. Sol. Cells 179, 142–151 (2018). 23. Scragg, J. J. S. et al. Cu-Zn disorder and band gap fluctuations in Cu 2 ZnSn(S,Se) 4 : Theoretical and experimental investigations: Cu-Zn disorder and band gap fluctuations in Cu 2 ZnSn(S,Se) 4. Phys. Status Solidi B 253, 247–254 (2016). 24. Gunawan, O., Todorov, T. K. & Mitzi, D. B. Loss mechanisms in hydrazine-processed Cu2ZnSn(Se,S)4 solar cells. Appl. Phys. Lett. 97, 233506 (2010). 25. Platzer-Björkman, C. et al. Reduced interface recombination in Cu 2 ZnSnS 4 solar cells with atomic layer deposition Zn 1− x Sn x O y buffer layers. Appl. Phys. Lett. 107, 243904 (2015). 26. Yan, C. et al. Band alignments of different buffer layers (CdS, Zn(O,S), and In 2 S 3 ) on Cu 2 ZnSnS 4. Appl. Phys. Lett. 104, 173901 (2014). 27. Xin, H. et al. Lithium-doping inverts the nanoscale electric field at the grain boundaries in Cu 2 ZnSn(S,Se) 4 and increases photovoltaic efficiency. Phys. Chem. Chem. Phys. 17, 23859–23866 (2015). 28. Courel, M., Andrade-Arvizu, J. A. & Vigil-Galán, O. Towards a CdS/Cu 2 ZnSnS 4 solar cell efficiency improvement: A theoretical approach. Appl. Phys. Lett. 105, 233501 (2014). 29. Kim, J. H. et al. Atomic-Scale Observation of Oxygen Substitution and Its Correlation with HoleTransport Barriers in Cu 2 ZnSnSe 4 Thin-Film Solar Cells. Adv. Energy Mater. 6, 1501902 (2016). 30. Winkler, M. T. et al. Optical designs that improve the efficiency of Cu 2 ZnSn(S,Se) 4 solar cells. Energy Env. Sci 7, 1029–1036 (2014). 31. Tuan, D. A., Ke, N. H., Thi Kieu Loan, P. & Hung, L. V. T. A method to improve crystal quality of CZTSSe absorber layer. J. Sol-Gel Sci. Technol. 87, 245–253 (2018). 32. Levcenko, S., Tezlevan, V. E., Arushanov, E., Schorr, S. & Unold, T. Free-to-bound recombination in near stoichiometric Cu 2 ZnSnS 4 single crystals. Phys. Rev. B 86, (2012). 33. Su, Z. et al. Cation Substitution of Solution-Processed Cu 2 ZnSnS 4 Thin Film Solar Cell with over 9% Efficiency. Adv. Energy Mater. 5, 1500682 (2015). 34. Shin, D. et al. Earth-Abundant Chalcogenide Photovoltaic Devices with over 5% Efficiency Based on a Cu 2 BaSn(S,Se) 4 Absorber. Adv. Mater. 29, 1606945 (2017). 35. Gershon, T. et al. Photovoltaic Materials and Devices Based on the Alloyed Kesterite Absorber (Ag x Cu 1x ) 2 ZnSnSe 4. Adv. Energy Mater. 6, 1502468 (2016). 36. Guchhait, A. et al. Enhancement of Open-Circuit Voltage of Solution-Processed Cu 2 ZnSnS 4 Solar Cells with 7.2% Efficiency by Incorporation of Silver. ACS Energy Lett. 1, 1256–1261 (2016). 37. Giraldo, S. et al. How small amounts of Ge modify the formation pathways and crystallization of kesterites. Energy Environ. Sci. 11, 582–593 (2018). 38. Neuschitzer, M. et al. V oc Boosting and Grain Growth Enhancing Ge-Doping Strategy for Cu 2 ZnSnSe 4 Photovoltaic Absorbers. J. Phys. Chem. C 120, 9661–9670 (2016). 39. Giraldo, S. et al. Large Efficiency Improvement in Cu 2 ZnSnSe 4 Solar Cells by Introducing a Superficial Ge Nanolayer. Adv. Energy Mater. 5, 1501070 (2015). 40. Collord, A. D. & Hillhouse, H. W. Germanium Alloyed Kesterite Solar Cells with Low Voltage Deficits. Chem. Mater. 28, 2067–2073 (2016). 41. Hages, C. J. et al. Improved performance of Ge-alloyed CZTGeSSe thin-film solar cells through control of elemental losses: Improved performance of CZTGeSSe solar cells. Prog. Photovolt. Res. Appl. 23, 376–384 (2015). 42. Giraldo, S. et al. Cu 2 ZnSnSe 4 -Based Solar Cells With Efficiency Exceeding 10% by Adding a Superficial Ge Nanolayer: The Interaction Between Ge and Na. IEEE J. Photovolt. 6, 754–759 (2016). 43. Andrade-Arvizu, J. et al. Rear Band gap Grading Strategies on Sn–Ge-Alloyed Kesterite Solar Cells. ACS Appl. Energy Mater. 3, 10362–10375 (2020). 44. Kim, S., Kim, K. M., Tampo, H., Shibata, H. & Niki, S. Improvement of voltage deficit of Geincorporated kesterite solar cell with 12.3% conversion efficiency. Appl. Phys. Express 9, 102301 (2016). 45. Kim, S. et al. Ge-incorporated Cu2ZnSnSe4 thin-film solar cells with efficiency greater than 10%. Sol. Energy Mater. Sol. Cells 144, 488–492 (2016). 46. Sahayaraj, S. et al. Optoelectronic properties of thin film Cu2ZnGeSe4 solar cells. Sol. Energy Mater. Sol. Cells 171, 136–141 (2017). 47. Choubrac, L. et al. 7.6% CZGSe Solar Cells Thanks to Optimized CdS Chemical Bath Deposition. Phys. Status Solidi A 215, 1800043 (2018). 48. Benhaddou, N. et al. Uncovering details behind the formation mechanisms of Cu 2 ZnGeSe 4 photovoltaic absorbers. J. Mater. Chem. C 8, 4003–4011 (2020). 49. Choubrac, L. et al. Sn Substitution by Ge: Strategies to Overcome the Open-Circuit Voltage Deficit of Kesterite Solar Cells. ACS Appl. Energy Mater. 3, 5830–5839 (2020). 50. Gunder, R., Márquez-Prieto, J. A., Gurieva, G., Unold, T. & Schorr, S. Structural characterization of off-stoichiometric kesterite-type Cu 2 ZnGeSe 4 compound semiconductors: from cation distribution to intrinsic point defect density. CrystEngComm 20, 1491–1498 (2018). 51. Courel, M., Sanchez, T. G., Mathews, N. R. & Mathew, X. Cu 2 ZnGeS 4 thin films deposited by thermal evaporation: the impact of Ge concentration on physical properties. J. Phys. Appl. Phys. 51, 095107 (2018). 52. Neuschitzer, M. et al. Optimization of CdS buffer layer for high-performance Cu 2 ZnSnSe 4 solar cells and the effects of light soaking: elimination of crossover and red kink: CdS and effects of light soaking: elimination of crossover and red kink. Prog. Photovolt. Res. Appl. 23, 1660–1667 (2015). 53. Guc, M. et al. Polarized Raman scattering analysis of Cu 2 ZnSnSe 4 and Cu 2 ZnGeSe 4 single crystals. J. Appl. Phys. 114, 193514 (2013). 54. Márquez, J. et al. Chemistry and Dynamics of Ge in Kesterite: Toward Band-Gap-Graded Absorbers. Chem. Mater. 29, 9399–9406 (2017). 55. Yin, L. et al. Limitation factors for the performance of kesterite Cu 2 ZnSnS 4 thin film solar cells studied by defect characterization. RSC Adv. 5, 40369–40374 (2015). 56. Li, J. V. et al. Theoretical analysis of effects of deep level, back contact, and absorber thickness on capacitance–voltage profiling of CdTe thin-film solar cells. Sol. Energy Mater. Sol. Cells 100, 126–131 (2012). 57. Heath, J. T., Cohen, J. D. & Shafarman, W. N. Bulk and metastable defects in CuIn1−xGaxSe2 thin films using drive-level capacitance profiling. J. Appl. Phys. 95, 1000–1010 (2004). 58. Nagai, T. et al. Characterization of Surface and Heterointerface of Cu 2 ZnSn 1– x Ge x Se 4 for Solar Cell Applications. Phys. Status Solidi RRL – Rapid Res. Lett. 14, 1900708 (2020). 59. Giraldo, S. et al. Cu 2 ZnSnSe 4 solar cells with 10.6% efficiency through innovative absorber engineering with Ge superficial nanolayer: Cu 2 ZnSnSe 4 solar cells with 10.6% efficiency. Prog. Photovolt. Res. Appl. 24, 1359–1367 (2016). 60. Giraldo, S. et al. Study and optimization of alternative MBE‐deposited metallic precursors for highly efficient kesterite CZTSe:Ge solar cells. Prog. Photovolt. Res. Appl. 27, 779–788 (2019). 61. Ould Salem, M. et al. Over 10% Efficient Wide Bandgap CIGSe Solar Cells on Transparent Substrate with Na Predeposition Treatment. Sol. RRL 2000284 (2020) doi:10.1002/solr.202000284. 62. Cuevas, A. The Recombination Parameter J0. Energy Procedia 55, 53–62 (2014). 63. Li, J. et al. 10% Efficiency Cu2ZnSn(S,Se)4 thin film solar cells fabricated by magnetron sputtering with enlarged depletion region width. Sol. Energy Mater. Sol. Cells 149, 242–249 (2016). 64. Vidal-Fuentes, P. et al. Efficient Se‐Rich Sb 2 Se 3 /CdS Planar Heterojunction Solar Cells by Sequential Processing: Control and Influence of Se Content. Sol. RRL 4, 2000141 (2020). 65. Nagaya, K. et al. Very small tail state formation in Cu 2 ZnGeSe 4. Appl. Phys. Lett. 113, 093901 (2018). 66. Xie, H. et al. Impact of Na Dynamics at the Cu 2 ZnSn(S,Se) 4 /CdS Interface During Post Low Temperature Treatment of Absorbers. ACS Appl. Mater. Interfaces 8, 5017–5024 (2016). 67. Ojeda-Durán, E. et al. CZTS solar cells and the possibility of increasing VOC using evaporated Al2O3 at the CZTS/CdS interface. Sol. Energy 198, 696–703 (2020). 68. Nadenau, V., Rau, U., Jasenek, A. & Schock, H. W. Electronic properties of CuGaSe2-based heterojunction solar cells. Part I. Transport analysis. J. Appl. Phys. 87, 584–593 (2000). 69. Turcu, M., Pakma, O. & Rau, U. Interdependence of absorber composition and recombination mechanism in Cu(In,Ga)(Se,S)2 heterojunction solar cells. Appl. Phys. Lett. 80, 2598–2600 (2002). 70. Scheer, R. Activation energy of heterojunction diode currents in the limit of interface recombination. J. Appl. Phys. 105, 104505 (2009). 71. Halim, M. A. et al. A comparative study on charge carrier recombination across the junction region of Cu 2 ZnSn(S,Se) 4 and Cu(In,Ga)Se 2 thin film solar cells. AIP Adv. 6, 035216 (2016). 72. Wang, K. et al. Thermally evaporated Cu2ZnSnS4 solar cells. Appl. Phys. Lett. 97, 143508 (2010).