Se alloying of CdTe absorber is an integral step in achieving high performing devices. This alloying predominantly takes place during the CdCl2 heat treatment making it imperative to understand the combined role they both play in the system. Previous work in literature has focused on understanding macroscopic electrical, optical, and structural properties of Se.-alloyed CdTe absorbers. In this work, we hope to shine light on the role CdCl2 conditions play on the local environment of Se and Cl in the absorber, as well as, on the strain developed in the system at the nanoscale. Through a multimodal X-ray microscopy approach we combine diffraction, fluorescence and absorption to evaluate the evolution and levels of strain in the system and their correlation to the main defects formed under different CdCl2 processing conditions.

X-ray microscopy is a powerful tool to study defect chemistry in solar cells since it is capable of probing structure, chemistry, and opto-electrical properties, correlatively, on a pixel-by-pixel basis. In this work, X-ray absorption near edge structure (XANES) was used to probe the As local structures across the depth of the CdSexTe1-x absorber under (x-ray) illumination and bias conditions. It was found that the As structures located near the front contact, which mostly have the spectral features of oxides, are susceptible to undergo structural change under different biasing conditions. While those located near the back contact largely remain unchanged - notably, Cd2AsCl2 and AsTe+Clx complex defect. Additionally, while we found no spectral evidence of the self-compensating AX center, we do observe a high fraction of AsTe+Clx complex around the middle of the absorber, in both low and high performing areas. This finding could explain why a high percentage of the incorporated As is inactive and could perhaps shed light on the self-compensating effect.

Selenium alloying of the absorber is a key component in achieving high efficiency polycrystalline CdSeTe solar cells. Much work has been carried out in characterizing the macroscopic electrical, optical, and structural properties of this material system. However, the movement of Se in the absorber layer and the role of CdCl2 in this process are still not fully understood. In this study, nano-scale correlative X-ray microscopy has been utilized to study ex-situ and in-situ Se-alloyed CdTe samples before, during and after CdCl2 treatment. Changes in the Se distribution were mapped by nano x-ray fluorescence and detailed information about the Se local environment in the absorber layer was collected through x-ray absorption spectroscopy ex-situ and in-situ during CdCl2 treatment. Complementary nano-Xray diffraction grain maps were collected to properly assess the evolution of stress/strain through the recrystallization process.

Copper has been used as a p-type dopant in cadmium telluride (CdTe) for decades. However, the density of Cu atoms in the finished device is much higher than that of holes, which means that most Cu atoms are not activated as acceptors during incorporation. Furthermore, studies have demonstrated that the distribution of copper (Cu) atoms across the device is highly inhomogeneous, with reports citing Cu substitution on Cd sites and segregation to grain boundaries. Fast diffusion along these boundaries and Cu accumulation at the CdTe/CdS interface have also been observed and validated computationally. These levels of inhomogeneity make it difficult to accurately characterize and correlate the performance with the nature of the Cu atomic species present. To address this challenge, we utilize X-ray microscopy and, specifically, nanoscale fluorescence-mode X-ray absorption near-edge structure to resolve the atomic Cu environment throughout the depth of the CdTe layer. Our results suggest that the majority of Cu atoms are in the form of CuxTe phases (or similar local environments) near the ZnTe|CdTe interface, CuxO phases in the CdTe absorber, and present in various oxidation states, including Cu1+ and Cu2+, near the CdS/CdTe junction. This work also provides experimental evidence for the first time of the presence of CuS around the ZnTe|CdTe interface and the hypothesized Cu-Cd-Cl-i complex in the CdTe absorber.

The selection of arsenic as a dopant has been an integral part of achieving high performing, 20.8%, polycrystalline CdTe solar cells. Although arsenic has significantly improved the long-term performance of the cells and the p-type doping levels reached inside the absorber, the activation ratio of dopant remains quite low, ~1%. Understanding the origins of this activation would aid in further bettering device performance. Herein, devices of two different activation levels but identical arsenic concentration are studied using nanoscale correlative X-ray microscopy in cross-section. Charge collection in cross-section shows a distinctive change between activation levels and the local environment around the As atom, as measured by X-ray absorption, shows the signature of several defects and phases that could be the culprit.

For decades, Cu has been the primary dopant in CdTe photovoltaic absorbers. Typically, Cu acceptor concentrations in these devices are on the order of 1 × 1014 cm−3, which has made it notoriously difficult to directly correlate nanoscale Cu distributions to the local charge transport properties of these devices. To measure and correlate these properties, measurement techniques require high sensitivity to elemental concentration, large penetration depth, and operando compatibility. Techniques such as secondary-ion mass spectroscopy and X-ray energy dispersive spectroscopy are widely adopted to measure Cu concentrations, but they are limited by penetration depth, sensitivity, or spatial resolution. Additionally, they lack the operando capabilities required to correlate one-to-one Cu concentrations to electrical performance. In this work, correlative X-ray microscopy is used to investigate the spatial distribution of Cu and its impact on charge collection through the depth and breadth of CdTe photovoltaic devices. Plan-view, nanoscale X-ray fluorescence maps clearly demonstrate the spatial segregation of copper around regions thought to be CdTe grain boundaries. Complementary cross-section imaging unveils the transition of the maximum charge-collection efficiency from the ZnTe–CdTe interface to the CdS–CdTe interface as a function of Cu incorporation. The copper concentration through the depth of the CdTe layer is characterized by slow and fast diffusion components, and cross-section charge-transport modeling shows that the experimentally obtained charge collection can be explained by the modeled acceptor distribution through the depth of the CdTe layer.

X-ray Absorption Near Edge Structures (XANES) is a powerful tool to unravel chemical environment as it is sensitive to oxidation state and small structural variations. In this work, we measured XANES spectra and simulated structures to fit the Se local structures, including intrinsic defects around Se atoms, inside the absorber layer of Cu(In,Ga)Se2 (CIGS) solar cells. This work reveals for the first time the distributions of point defects in the absorber, across the Mo and CdS interfaces, validating the presence of selenium vacancies (VSe) and copper vacancies (VCu) proposed by multiple authors but also suggesting that there are more clusters of VSe and VCu on the CdS side and more VCu clusters on the Mo side.

X-ray microscopy is a powerful tool to study defects in solar cells as it allows to correlate pixel-by-pixel the local environment of selected atoms and the nanoscale electrical performance. In this work, we used X-ray absorption near edge structures (XANES) and X-ray induced current (XBIC) to track Se local structures, particularly changes in Se-Cd bond lengths, across the CdSexTe(1-x) absorber layer, and contrasts areas with high and low electrical performance. Even though all the experimental Se K-edge XANES clearly show signature of CdSexTe(1-x), there are spectral changes both across the absorber, and at different performing areas, revealing different atomic surrounding of Se atoms which together with XANES at the As K-edge may provide a full picture of the role of defects in the bulk's electrical performance.

It is widely accepted that micro- and nanoscale inhomogeneities govern the performance of many thin-film solar cell absorbers. These inhomogeneities yield material properties (e.g., composition, structure, and charge collection) that are challenging to correlate across length scales and measurement modalities. The challenge is compounded if a correlation is sought during device operation or in conditions that mimic aging under particular stressors (e.g., heat and electrical bias). Correlative approaches, particularly those based on synchrotron x-ray sources, are powerful since they can access several material properties in different modes (e.g., fluorescence, diffraction, and absorption) with minimal sample preparation. Small-scale laboratory x-ray instruments have begun to offer multi-modality but are typically limited by low x-ray photon flux, low spatial resolution, or specific sample sizes. To overcome these limitations, a characterization stage was developed to enable multi-scale, multi-modal operando measurements of industrially relevant photovoltaic devices. The stage offers compatibility across synchrotron x-ray facilities, enabling correlation between nanoscale x-ray fluorescence microscopy, microscale x-ray diffraction microscopy, and x-ray beam induced current microscopy, among others. The stage can accommodate device sizes up to 25 × 25 mm2, offering access to multiple regions of interest and increasing the statistical significance of correlated properties. The stage materials can sustain humid and non-oxidizing atmospheres, and temperature ranges encountered by photovoltaic devices in operational environments (e.g., from 25 to 100 °C). As a case study, we discuss the functionality of the stage by studying Se-alloyed CdTe photovoltaic devices aged in the stage between 25 and 100 °C.

Copper is a traditional dopant for many types of polycrystalline thin-film CdTe photovoltaic devices. However, Cu can easily distribute through the depth and breadth of the device, segregating at interfaces or grain boundaries and leading to metastability of the device. Directly correlating Cu-related defect species to the local (i.e. nanoscale) charge transport in CdTe devices remains challenging due to relatively low Cu concentrations in the CdTe layer. Using nanoscale X-ray microscopy, we simultaneously probe both the elemental copper distribution and electrical performance of the device in cross-section. Complementary charge transport modelling delineates the possible defect distributions that can exist under low and high Cu loading, and how these defects interact with charge carriers at different depths of the device.

For decades, copper has been used to improve the performance of cadmium telluride thin film solar cells. However, it has also been shown to be the main cause of metastability in CdTe. Recently a low activation energy has been reported for the thermal diffusion of Cu in CdTe explaining the ease of motion that it has under moderate temperatures. The community consensus is that copper segregates to the absorber grain boundaries, where it's either beneficial or detrimental to device performance depending on its concentration. Using nanoscale X-ray micrsocopy and a two-dimensional drift-diffusion model we present a preliminary correlation between local copper distribution and electrical performance of a single-junction CdTe/CdS solar cell.

II–VI/Si tandem solar cells have strong potential for high efficiency at low cost by combining the two most widely used solar cell materials: silicon and cadmium telluride (CdTe). However, there are challenges with this merger, as loss of minority-carrier lifetime in the silicon bottom cell can be caused by growth of a II–VI cell on top. Silicon lifetime degradation in monolithic II–VI/Si structures is measured here on experimental samples for CdTe deposition temperatures between 400 and 500 °C, with variable In2O3:ZnO (IZO) thickness between the CdTe and silicon, and with and without CdCl2 postdeposition treatment. Results indicate that the CdCl2 treatment has the strongest effect on silicon lifetime reduction, followed by temperature and IZO thickness. Potential causes are discussed, and the effect on monolithic II–VI/Si two-junction solar cells is modeled. Remarkably, many silicon samples in the study were able to maintain >400 μs lifetimes, with some exceeding 1 ms, consistent with >30% projected efficiency in fully integrated II–VI/Si tandem solar cells.

Cadmium telluride and silicon are among the widely used absorber materials in photovoltaic industry. A tandem solar cell of these two can absorb significant portion of solar spectrum to yield high efficiency due to the added voltage of the two solar cells. On basis of low-cost production, a CdTe/Si cell has the potential to produce low-cost and high efficiency tandem PV. The CdTe top cell in a substrate configuration is essential to achieve a tandem between CdTe and Si. A functional CdS/CdTe solar cell in the substrate configuration was fabricated on a Si wafer. Current-Voltage measurements show a diode-like curve with lower J-V parameters compared to standard CdS/CdTe cells. SCAPS simulations were performed to identify possible reasons for poor performance and help improve the device performance.

Passivation of dangling bonds at surfaces and interfaces is an essential feature of high-efficiency solar cells. The high recombination activity of surface states in III-V semiconductors typically requires epitaxial growth of high-bandgap layers to suppress interface recombination. Surface recombination in III-Vs can be reduced by solution deposition of sulfides, however, the resulting passivation can degrade over time, and surface roughness caused by the passivation can degrade cell performance. In this work, the effects of different passivation materials and techniques, such as atomic layer deposition (ALD) and plasma enhanced chemical vapor deposition (PECVD), on GaAs and InP surfaces are analyzed by direct measurement of the minority-carrier lifetime by time-resolved photoluminescence (TRPL). Recombination parameters due to passivation of surface states are characterized for Al 2 O 3 , amorphous silicon (a-Si) and sulfide interfaces on n-type, intrinsic, and p-type GaAs and InP substrates. Al 2 O 3 passivation of n-GaAs has shown 1.35 ns improvement in measured lifetime, while a-Si passivation shows a slight improvement in passivation for n-type GaAs and p-type InP.

An intricate look is taken at the methods used to account for variance in minority-carrier lifetime in the silicon bottom cell of II-VI/Si tandem solar cells. A discussion on the modeling is provided. Lateral wafer variance is determined to be much less than wafer-to-wafer variance. Size testing indicates a minimum size of 4 × 4 cm is necessary for accurate results. The cleaning procedure and photoluminescence testing is described. Despite a small sample size, Si samples with CdTe deposition and CdCl2 treatment maintain over 1 ms lifetimes, enabling the Si bottom cell in II-VI/Si tandem cells to reach state-of-the-art performance.

The analysis of Auger-corrected inverse minoritycarrier lifetime as a function of excess carrier concentration can impart information about many crucial solar cell material properties including: emitter saturation current density (J o ), surface recombination velocity (SRV), Shockley-Read-Hall (SRH) recombination in the bulk, trap density, band bending, surface Fermi-level pinning, and bandgap narrowing. This work demonstrates TaO x as a potential electron selective contact and passivation layer on silicon. Increasing TaO x thickness reduces the measured upper limit of effective surface recombination velocity (S eff, UL ) . The minimum S eff, UL is 55 cm/s for CZ n-Si/ 30 nm TaO x interfaces. S eff, UL increases for Si / TaO x / ITO structures due to unfavorable band bending.