Cu2ZnSnS4 (CZTS) shows great potential for cheap, efficient photovoltaic devices. However, one problem during synthesis of CZTS films is the loss of Sn as a result of decomposition and evaporation of SnS. This paper uses kinetic models to show that the mechanism of the decomposition reaction probably occurs in at least two stages; first, a loss of sulfur which causes dissociation of the structure into binary sulfides, and only then the evaporation of SnS. Knowledge of the reaction mechanism helps to identify the driving force for decomposition as arising from the relative instability of Sn(IV) in CZTS against reduction; this theory is backed up by thermodynamic data. The volatility of SnS further exaggerates the decomposition by rendering it irreversible. This insight, alongside experimental data, allows prediction of the annealing conditions required to stabilize CZTS surfaces. A fundamental incompatibility of CZTS with high-temperature, vacuum-based processing is exposed, distinguishing it from related indium-containing compounds. This offers an explanation as to why the most efficient CZTS devices to-date all arise from “two-stage” fabrication processes involving low temperature deposition followed by annealing at high pressure, and provides key information for designing successful annealing strategies.

International audienceCu2ZnSnS4 (CZTS) is an interesting material for sustainable photovoltaics, but efficiencies are limited by the low open-circuit voltage. A possible cause of this is disorder among the Cu and Zn cations, a phenomenon which is difficult to detect by standard techniques. We show that this issue can be overcome using near-resonant Raman scattering, which lets us estimate a critical temperature of 533 +/- 10K for the transition between ordered and disordered CZTS. These findings have deep significance for the synthesis of high-quality material, and pave the way for quantitative investigation of the impact of disorder on the performance of CZTS-based solar cells. (C) 2014 AIP Publishing LLC

Experimental proof is presented for a hitherto undetected solid-state reaction between the solar cell material Cu(2)ZnSn(S,Se)(4) (CZTS(e)) and the standard metallic back contact, molybdenum. Annealing experiments combined with Raman and transmission electron microscopy studies show that this aggressive reaction causes formation of MoS(2) and secondary phases at the CZTS|Mo interface during thermal processing. A reaction scheme is presented and discussed in the context of current state-of-the-art synthesis methods for CZTS(e). It is concluded that alternative back contacts will be important for future improvements in CZTS(e) quality.

Cu 2 ZnSnS 4 (CZTS) is a promising material for thin film solar cells based on sustainable resources. This paper explores some consequences of the chemical instability between CZTS and the standard Mo "back contact" layer used in the solar cell. Chemical passivation of the back contact interface using titanium nitride (TiN) diffusion barriers, combined with variations in the CZTS annealing process, enables us to isolate the effects of back contact chemistry on the electrical properties of the CZTS layer that result from the synthesis, as determined by measurements on completed solar cells. It is found that instability in the back contact is responsible for large current losses in the finished solar cell, which can be distinguished from other losses that arise from instabilities in the surface of the CZTS layer during annealing. The TiN-passivated back contact is an effective barrier to sulfur atoms and therefore prevents reactions between CZTS and Mo. However, it also results in a high series resistance and thus a reduced fill factor in the solar cell. The need for high chalcogen pressure during CZTS annealing can be linked to suppression of the back contact reactions and could potentially be avoided if better inert back contacts were to be developed.

Tandem solar cells based on hybrid organic−inorganic metal halide perovskites have reached efficiencies up to 28%, but major concerns for long-term stability and the presence of Pb have raised interest in searching for fully earth-abundant, intrinsic chemically stable, and nontoxic alternatives. With a direct band gap around 1.8 eV and stability in air up to at least 500 °C, BaZrS 3 is a promising candidate. This work presents the first approach of synthesizing a thin film of such compound by sputtering at ambient temperature with a subsequent rapid thermal process. Despite the short fabrication time, the width of the XRD diffraction peaks and the energy and distribution of the photoluminescence response show comparable crystalline quality to that from bulk synthesis methods. Good crystallization required around 900 °C. Such a high temperature could be incompatible with fabrication of tandem solar cells.

Cu 2 ZnSnS 4 (CZTS) is a promising thin-film absorber material that presents some interesting challenges in fabrication when compared with Cu(In,Ga)Se 2 . We introduce a two-step process for fabrication of CZTS films, involving reactive sputtering of a Cu-Zn-Sn-S precursor followed by rapid annealing. X-ray diffraction and Raman measurements of the sputtered precursor suggest that it is in a disordered, metastable CZTS phase, similar to the high-temperature cubic modification reported for CZTS. A few minutes of annealing at 550 C are sufficient to produce crystalline CZTS films with grain sizes in the micrometer range. The first reported device using this approach has an AM1.5 efficiency of 4.6%, with J sc and V oc both appearing to be limited by interface recombination.

Tin monosulfide (SnS) is a promising light-absorbing material with weak environmental constraints for application in thin film solar cells. In this paper, we present low-temperature atomic layer deposition (ALD) of high-purity SnS of both cubic and orthorhombic phases. Using tin­(II) 2,4-pentanedionate [Sn­(acac)2] and hydrogen sulfide (H2S) as precursors, controlled growth of the two polymorphs is achieved. Quartz crystal microbalance measurements are used to establish saturated conditions and show that the SnS ALD is self-limiting over temperatures from at least 80 to 160 °C. In this temperature window, a stable mass gain of 19 ng cm–2 cycle–1 is observed. The SnS thin film crystal structure and morphology undergo significant changes depending on the conditions. High-resolution transmission electron microscopy and X-ray diffraction demonstrate that fully saturated growth requires a large H2S dose and results in the cubic phase. Smaller H2S doses and higher temperatures favor the orthorhombic phase. The optical properties of the two polymorphs differ significantly, as demonstrated by spectroscopic ellipsometry. The orthorhombic phase displays a wide (0.3–0.4 eV) Urbach tail in the near-infrared region, ascribed to its nanoscale structural disorder and/or to sulfur vacancy-induced gap states. In contrast, the cubic phase is smooth and void-free and shows a well-defined, direct forbidden-type bandgap of 1.64 eV.

Zn 1-x Sn x O y (ZTO) has yielded promising results as buffer material for the full sulfur Cu 2 ZnSnS 4 (CZTS), with efficiencies continuously surpassing its CdS-references. ZTO can be deposited by atomic layer deposition (ALD) enabling tuning of the conduction band position through choice of metal ratio or deposition temperature. Thus an optimization of the conduction band alignment between ZTO and CZTS can be achieved. The ZTO bandgap is generally larger than that of CdS and can therefore yield higher currents due to reduced losses in the short wavelength region. Another advantage is the possibility to omit the toxic Cd. In this study the ALD process temperature was varied from 105 to 165 °C. Current-blocked devices were obtained at 105 °C while the highest open-circuit voltage and device efficiency was achieved for 145 °C. The highest fill factor was seen at 165 °C. The best efficiency reached in this study was 9.0 %, which, to our knowledge, is the highest efficiency reported for Cd-free full-sulfur CZTS. We also show that the effect of heat needs to be taken into 2 account. The results indicate that part of the device improvement comes from heating the absorber, but that the benefit of using a ZTO-buffer is clear.