In this communication, we study the pure selenide phase CZTSe thin fi lm solar cells prepared by vacuum co-evaporation and demonstrate a power conversion effi ciency of 11.6%, which is the highest certifi ed effi ciency among CZTSe-based thin-fi lm solar cells reported to date. Device characterizations indicate that the higher effi ciencies appear to be related to signifi cantly enhanced transport properties for photogenerated carriers. The measured minority carrier diffusion length yields a record value of ≈2.1 µm, approaching that of state-of-the-art CIGS devices. The device also shows a record V OC -defi cit of 0.578 V, which can be attributed to reduced band tailing or the occurrence of shallower defect related states in the bandgap of the CZTSe layer. A comparative study of room-temperature photoluminescence (PL) of CZTSe and CZTS devices indicates that the PL emission peak from CZTSe is much closer in energy to the band edge than the case in CZTS. The improved device results are also consistent with the observation of a high-quality microstructure with large grains of ≈2.6 µm size.Polycrystalline CZTSe thin-fi lms with a thickness of ≈2.2 µm were deposited at 150 °C on 0.7-µm-thick Mo-coated soda-lime glass substrates using a vacuum co-evaporation technique, as described in earlier publications. [ 5,7 ] Knudsen-type elemental sources of Cu, Zn, and Sn, and a valved Se cracking source were used for the CZTSe co-evaporation. A ≈30-nm-thick NaF layer was deposited in a separate chamber prior to the CZTSe co-evaporation to promote grain coarsening of CZTSe thinfi lms. [ 16 ] Recrystallization and grain growth were carried out by annealing the co-evaporated CZTSe fi lm at ≈590 °C on a hot plate with excess selenium under N 2 atmosphere; similar to a process described elsewhere. [ 5,17,18 ] Inductively coupled plasma measurements showed a Cu-poor and Zn-rich elemental composition of the annealed fi lm: [Cu] Figure 1 a shows a θ -2 θ X-ray diffraction spectrum of the annealed CZTSe fi lm. The spectrum shows high crystallinity of the annealed fi lm with a major peak of (112)-orientation at ≈27.3°. The inset of Figure 1 a shows a top-view scanning electron microscopy (SEM) image of the annealed CZTSe fi lm, indicating dense and pinhole-free fi lm morphology. The average and standard deviation of grain size are determined to be 2.6 and 1.0 µm, respectively, estimated via digital image processing of SEM images.Using the co-evaporated CZTSe fi lms, solar cells with a total device area ( A ) of ≈0.43 cm 2 were fabricated with the following structure: MgF 2 /indium-tin-oxide (ITO)/ZnO/CdS/CZTSe/ Mo bottom electrode, as shown in Figure 1 b. The thicknesses Kesterite Cu 2 ZnSn( S x S e 1-x ) 4 (CZTSSe) has emerged as a promising candidate for scalable photovoltaic applications due to its earth-abundant elemental constituents and bandgap ( E g ) range between ≈1.0 eV ( x = 0) and ≈1.5 eV ( x = 1) with predicted theoretical maximum effi ciencies of over 30%. [ 1,2 ] To date, among thin-fi lm solar cells based on the kesteri...

narrows the CZTSSe band gap. The low energy barrier to Cu/Zn antisite formation is related to the similarity between the covalent radii of Cu and Zn. [ 4 ] The elevated processing temperatures (550-600 °C) needed to form large-grained CZTSSe fi lms (and peak device effi ciencies) provide the required thermal energy to randomize Cu and Zn in the unit cell, leading to a high density of antisite defects. [ 5 ] If disorder is the primary cause of performance loss in CZTSSe, then suppressing or eliminating it might offer a path to effi ciencies that compete with Cu(In,Ga)Se 2 (CIGS) technology.Ag is an interesting candidate for replacement of Cu since, in addition to belonging to the same chemical group as Cu, it possesses an atomic radius roughly 16% larger. This leads to the intriguing possibility that antisites can be suppressed by increasing the strain required to accommodate each defect (due to larger dissimilarity in radius). Ab initio calculations predict that substitution of Cu with Ag more than doubles the formation energy for antisites, which should result in an order of magnitude lower density of defects for equivalent processing. [ 6 ] Previous studies have examined the optical and crystallographic properties of the mixed Cu-Ag kesterite [ 7 ] and found that introducing 10% or 5% Ag into the CZTS(Se) layer gave 4.4% [ 8 ] or 7.1% [ 9 ] effi ciencies, respectively. These effi ciencies were shown to be an improvement over the baseline pure-Cu material; however, few direct measurements have been made to demonstrate how this substitution impacts the fundamental properties of the material.In this study, we have prepared thin fi lms of the mixed alloy (Ag x ,Cu 1x ) 2 ZnSnSe 4 (ACZTSe) across the full range of Ag/(Ag + Cu) ratios. We show, using Hall effect measurements, that while the pure-Cu kesterite compound is p-type, the carrier density decreases with increasing Ag content. For the highest values of Ag content (>50%), the material inverts to n-type. We also show, using femtosecond ultraviolet photoelectron spectroscopy (fs-UPS) measurements, that unlike in CZTSSe the Fermi level of AZTSe is not pinned near the center of the band gap, indicating that AZTSe does not suffer from the same degree of heavy compensation. Additionally, the energetic difference between the measured band gap and the photoluminescence (PL) peak position approaches zero for the pure-Ag compound. These results imply that the magnitude The photovoltaic absorber Cu 2 ZnSn(S x Se 1-x ) 4 (CZTSSe) has attracted interest in recent years due to the earth-abundance of its constituents and the realization of high performance (12.6% effi ciency). The open-circuit voltage in CZTSSe devices is believed to be limited by absorber band tailing caused by the exceptionally high density of Cu/Zn antisites. By replacing Cu in CZTSSe with Ag, whose covalent radius is ≈15% larger than that of Cu and Zn, the density of I-II antisite defects is predicted to drop. The fundamental properties of the mixed Ag-Cu kesterite compound are reported as a function of ...

It is well‐known that sodium improves the performance of Cu2ZnSnS4 (CZTS) devices, yet the mechanism of the enhancement is still not fully understood. This work aims to present a unified account of the relationships between grain boundaries in CZTS, sodium content at these boundaries, non‐radiative recombination, and surfactant effects that produce large microstructural changes. Using temperature‐dependent photoluminescence measurements, it is demonstrated that samples containing dramatically different grain sizes display identical radiative and non‐radiative decay characteristics when sufficient sodium is present in the film. It is also shown that the sodium concentration needed to efficiently passivate non‐radiative defects is significantly less that the quantity needed to obtain micrometer‐sized CZTS grains. Finally, the high densities of donor‐acceptor pairs that are observed in CZTS films appear to reside within the grains themselves, rather than at grain boundaries.

lead iodide perovskite top cell on Cu 2 ZnSn(S,Se) 4 kesterite and Si-based bottom cells. [ 10,12 ] The remarkable effi ciencies of perovskite devices above 15%, the possibility to process at temperatures below 150 °C, and the highly tunable band gap range from 1.6 to 2.25 eV make these materials especially attractive for monolithic tandem integration with CIGS. [12][13][14][15][16][17][18][19][20] Here, we report perovskite-CIGS tandem solar cells in which each cell was customized for monolithic integration in the following sequence: transparent conducting electrode (TCE)/phenyl-C61-butyric acid methyl ester (PCBM)/perovskite/poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS)/ITO/CdS/CIGS/Mo/Si 3 N 4 /glass. By modifying our solution-based process, [ 21 ] the CIGS absorber band gap was reduced to 1.04 eV for improved photon management. A 30 nm thick ITO was selected as a transparent recombination layer that was deposited directly onto the CdS without the intrinsic ZnO layer commonly used in CIGS devices. The elimination of ZnO from the device structure was critical for achieving functional perovskite tandems, as we discovered that the presence of ZnO in proximity to the perovskite layer degrades device performance. Perovskite devices with electron-selective ZnO layers have been reported previously, [22][23][24] however we found that processing at temperatures above 60 °C resulted in deterioration of the perovskite layer. In contrast, perovskite layers processed on our ZnO-free CIGS structure could withstand annealing treatments for several hours at 120 °C without any damage.Considering the importance of precise band gap control in monolithic tandem devices, we designed a reactor for continuous in situ monitoring and precise control of the optical properties of the perovskite layer via vapor-based halide exchange reactions, as illustrated in Figure 1 a. The system provided temperature and pressure control as well as spectroscopic measurement capability. In a typical process, a sample of spincoated lead halide PbX 2 (X = I and/or Br) is placed inside the temperature-controlled vacuum chamber and annealed in the presence of methylammonium halide (CH 3 NH 3 X) vapor. A transparent port at the top of the chamber provides an optical path for a broad-spectrum light source to illuminate the sample in the reactor, and a port beneath the sample leads to a spectrophotometer, which collects the transmission spectra. The real-time feedback is used for precise engineering of the perovskite absorber properties by adjusting the temperature, pressure, and precursor type. The process was designed to have better compatibility with the CIGS device than previously reported vapor-assisted approaches that employ temperatures in excess of 120 °C and focus on pure iodide perovskite with fi xed band gap. [ 25,26 ] In our reactor the conversion temperature was