Perovskite solar cells (PSCs) based on organometal halide light absorbers have hit the limelight in recent years owing to their low temperature solution processability, material abundance and rapidly rising efficiency. To rival the leading photovoltaic technologies, efficiency and long-term stability of PSCs represent two prime facets of the challenges currently facing the research community. Herein we summarize the strategies for improving efficiency and stability of PSCs by drawing on our recent work. Emphasis is given to the importance of perovskite film growth, electron/hole transport materials and interface materials in cell performance. We also discuss possible degradation mechanisms of PSCs.

The study of the chemical stability of solar selective coatings (SSC) for concentrated solar power (CSP) becomes essential for their use at high temperatures. In this paper, the short range order around Mo in Mo-Si3N4 cermets is studied for the first time by X-ray absorption spectroscopy. The information obtained by extended X-ray absorption fine structure (EXAFS) and X-ray absorption near edge structure (XANES) spectroscopies gives new insights of the origin of the optical behavior of the cermets cermets after vacuum and air annealing treatments. The established optical and structural correlation becomes of great importance for the design and optimization of SSC for practical applications.

Hybrid organic-inorganic perovskites (HOIPs) have been previously compounded with hydrophobic polymers in order to improve the resilience of HOIPs in humid environments. In this study HOIP particles were synthesized and dispersed into a polymer solution. The weight loading of the HOIP phase in the composite was increased until an abrupt change in the electrical conduction was observed indicating a percolation threshold was approached. Additionally, the CH3NH3PbI3/ polystyrene composite media was characterized to assess morphology and the effect it has on the observed electrical properties.

Hybrid perovskite solar cells (PSCs) under normal operation will reach a temperature above ∼ 60 °C, across the tetragonal-cubic structural phase transition of methylammonium lead iodide (MAPbI3). Whether the structural phase transition could result in dramatic changes of ionic, electrical and optical properties that may further impact the PSC performances should be studied. Herein, we report a structural phase transition temperature of MAPbI3 thin film at ∼ 55 °C, but a striking contrast occurred at ∼ 45 °C in the ionic and electrical properties of MAPbI3 due to a change of the ion activation energy from 0.7 eV to 0.5 eV. The optical properties exhibited no sharp transition except for the steady increase of the bandgap with temperature. It was also observed that the activation energy for ionic migration steadily increased with increased grain sizes, and reduction of the grain boundary density reduced the ionic migration.

In this work, a planar heterojunction superstrate n-i-p device based on Zn(O,S) electron transport layer and CsPbI2Br absorber material at 1.93 eV bandgap is presented. The CsPbI2Br films are deposited using a 2-step atmospheric solution deposition process and characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), UV-vis spectroscopy and photoluminescence (PL). Best device with an efficiency of 12.34 % and 11.94% in reverse and forward scans respectively and stabilized power output of 12.14 mW/cm2 has been demonstrated via atmospheric solution processing with minimal hysteresis between forward and reverse scans. The devices show voltage dependent current collection as well as light-dark crossover in forward bias. Light soaking tests at 65 °C and 1-sun at Voc, resulted in open-circuit voltage and fill-factor degradation. Electroluminescence (EL) after 100 hours of light soaking shows a reduction in overall EL intensity as well a shift in emission to lower wavelength. The devices exhibit a positive temperature coefficient of about 0.14 %/°C. It is found that Zn(O,S) is a viable alternative electron transport layer to replace TiO2. By replacing methylammonium cation with cesium and addition of Br has improved the stability of the perovskite phase.

Most research of texturization of solar cells has been devoted to Si based cells. For perovskites, it was assumed that texturization would not have much of an impact because of the relatively low refractive indexes lead to relatively low reflection as compared to the Si based cells. However, our optical modeling shows that a significant gain in perovskite (1.55 Ev) absorption from 84.6% to 93.5% for the wavelength range of 400 nm up to 800 nm. The largest gain in absorption is achieved between wavelengths of 700 nm and 800 nm. Because this is a range with a high photon density, the current density increases up to 10rel.%.

We have modeled different sine texture sizes and show generic trends in performance with texture. Moreover, by introducing a texture, the light is locally concentrated, depending on the texture configuration. This offers new cell architectures with optimized front and back contacts.

Electric field-modulated photoluminescence (PL) of perovskite solar cells is investigated to gain deeper insight about the role of the mobile ions in organometal halide perovskites. The PL intensity of perovskite solar cells show significant dependence on the polarity of the external electric field and the voltage scanning direction. This phenomenon is discussed in the framework of an ion migration mechanism, which has been widely accounted for the current density-voltage (J-V) hysteresis in perovskite solar cells. The result suggests that the mobile ions not only change the internal electric field of perovskite solar cells, but also have an effect on the recombination of photogenerated charge carriers.

The Ga-doped ZnO (GZO) were deposited by magnetron reactive sputtering on glass substrates at room temperature with different deposited times to obtain various thickness. The root-mean-square (RMS) roughness obtained from the atomic force microscopy (AFM) images is observed to shift linearly with the deposited time, the fractal geometry and multi-resolution signal decomposition (MRSD) based on wavelet transform were applied on the surface profiles and the results does not synchronously changes as the thickness, which is related to the profile’s frequency. The calculated compressive in-plane stress of highly c-axis oriented GZO films also shows an irregular variation as the increase of film thickness, what’s more, the in-plane stress and fractal dimension exhibit a polynomial relationship and the two parameters can be used for describing the surface morphology.

ZnO1-xTex thin films were deposited by DC reactive magnetron co-sputtering using pure Zn and Te targets. Te atomic concentration in the films ranged from x=0 to 0.33 by adjusting the applied power on the targets or varying the cathode-substrate distance. Chemical composition and crystalline structure were determined by RBS experiments and X-ray Diffraction, respectively. For low Te atomic concentrations (x≤0.04) the deposited ZnO1-xTex films showed a crystalline structure ZnO wurtzite type however, for increasing Te concentration significant broadening and decreasing intensities of the main peaks belonging to pure ZnO films together with some weak peaks characteristic of crystalline Trizinc Tellurate salt have appeared. For the highest x values some non-identified weak peaks beside to some others peaks corresponding to the crystalline phases mentioned above as well as, a broad band probably associated to amorphous TeO2 phase were observed. A preliminary optical characterization of the samples point out the possibility of different electronic transitions within the ZnO band gap.

We present the optical and electrical properties of ITO and AZO films fabricated directly on silicon substrates under several growth and annealing temperatures, as well as their potential performance when used as low emissivity coatings in hybrid photovoltaic-thermal systems. We use broadband spectroscopic ellipsometry measurements (from 300 nm to 20 μm) to obtain a consistent model for the permittivity of each of the films. The best performance is found using the properties of the ITO film grown at 250 °C, with a state of the art resistivity of 0.2 mΩ-cm and an optimized thickness of 75 nm which leads to an estimated 50% increase in the extracted power compared to a standard diffused silicon solar cell. The Hall mobility and resistivity measurements of all the films are also provided, complementing and supporting the observed optical properties.

Concentrating Photovoltaics (CPV) field aims to integrate expensive high efficiency multi-junction cells into modules with low cost concentrating optics. The choice of the optics depends on different factors: easiness of fabrication and integration process, added costs, optical efficiency and the profile of the spot uniformity reaching the cell. Indeed, previous work has shown a dependence between electrical performance and spectral and spatial uniformities of the light on the cell. To analyze it, a solar CPV test bench is developed at CEA-INES facilities. Lens and cell temperature can be applied separately, in order to evaluate independently different test conditions, while electrical or optical parameters are recorded. The present work shows how temperature and mechanical variations on first stage concentrating optic affects module performances. Several optics and materials are compared, in order to present the tool capabilities.

Novel metrology tool for in-situ characterization of surfaces semiconductor solar cells (both silicon and compound), and Light Emitting Device diffusers is presented. The tool measures the total integrated scattering when measuring forward, or back-reflection at very large angles of incidence. The tool is insensitive to vibrations and stray light. We discuss polarization resolved data and characterize our technique using NIST traceable standards. We discuss it’s applications to semiconductor manufacturing.

It is known that n-Si solar cells have higher efficiency than p-Si solar cells. One of the problems connected with n-Si application for solar cell production is the difficulty of using Czochralski method for growing n-Si ingots, uniform in structure. The present paper examines the possibility of production of n-Si ingots, uniform in resistance, by neutron transmutation doping (NTD) for photovoltaics using the mathematical modeling method. The provided calculation data are obtained by MCU-RFFI/A accounting code with DLC/MCUDAT-1.0 constant library developed by «Kurchatov Institute» Russian Research Center. The MCU accounting code is used for solution of the neutron-transport equation by Monte-Carlo procedure on the basis of estimated nuclear data for arbitrary three-dimensional geometry systems.

The present paper provides the estimation of uniformity of neutron-flux density along the ingot length and radius; dependence of silicon resistance on duration of irradiation. These studies established the neutron flux density distribution along the ingot length and radius; regularities of silicon resistance changes on duration and intensity of irradiation.

The influence of deep level defects lateral distribution in active layers of multicrystalline Si-based standard solar cells is investigated. Multicrystalline p-type Si wafers with 156×156 mm dimensions and 200 μm thickness were used for SCs preparation. One type of solar cells with conversion efficiency 20.4% was studied using capacitance voltage characteristics method (C-V) and by current deep level transient spectroscopy (I-DLTS). From various places along the diagonal of solar cell’s substrate with 20.4% efficiency nine pieces with an area ∼20 mm2 were extracted and studied. I-DLTS spectra of the five pieces from solar cell were measured. The features of deep levels defects concentration lateral distribution along the SC’s surface were studied.

In the conventional crystalline silicon heterojunction solar cell with the intrinsic thin layer structure (the HIT solar cell), a p-doped thin film silicon or its alloy (pDTF-Si/A) is used as the hole collecting window layer. However, the parasitic absorbance in the pDTF-Si/A window layer, and the toxic, explosive diborane gas used for p-doping are limiting factors for achieving HIT cells with reduced processing costs and / or higher efficiencies. In this work, pDTF-Si/A is replaced by V2Ox, which is deposited by a simple physical vapor deposition technique. Due to the wide band gap of V2Ox, the HIT solar cell with the V2Ox window layer generates a higher short-circuit current density than the reference conventional HIT cell under 1 sun, and achieves an open-circuit voltage of 0.7 V. Furthermore, the charge carrier lifetime and pseudo-efficiency values of the HIT solar cell with the V2Ox window layer indicate that this cell has the potential to outperform the conventional HIT cell in terms of the power conversation efficiency under the standard test conditions.

We determined that the conduction band offset (CBO) and the valence band offset (VBO) at the CdS/ Cu2ZnSnSe4 (CZTSe) heterointerface are +0.56 and +0.89eV, respectively, by using X-ray photoemission spectroscopy (XPS), ultraviolet photoemission spectroscopy (UPS) and inversed photoemission spectroscopy (IPES). A positive CBO value, so-called “spike” structure, means that the position of conduction band becomes higher than that of absorber layer. The evaluated CBO of +0.56 eV suggests that the conduction band alignment at CdS/CZTSe interface is enough to become an electron barrier. Despite such a large spike structure in the conduction band at the interface, a conversion efficiency of 8.7 % could be obtained for the CdS/CZTSe heterojunction solar cells.

Copper indium gallium selenium (CIGS) photovoltaic (PV) devices exhibit unique reverse breakdown characteristics in terms of the dependence on temperature, light intensity, photon energy, and buffer layer material. In this work, the theoretical basis of potential reverse breakdown mechanisms are described and compared to available data. Quantitative analysis performed with semiconductor device simulation indicates that none of the conventional reverse breakdown mechanisms can account for the observations. Further work to better understand the reverse current-voltage characteristics of CIGS PV devices will provide insight to improve performance and reliability.

Cu(In,Ga)Se2 (CIGS) photovoltaic absorbers frequently develop Ga gradients during growth. These gradients vary as a function of growth recipe, and are important to device performance. Prediction of Ga profiles using classic diffusion equations is not possible because In and Ga atoms occupy the same lattice sites and thus diffuse interdependently, and there is not yet a detailed experimental knowledge of the chemical potential as a function of composition that describes this interaction. We show how diffusion equations can be modified to account for site sharing between In and Ga atoms. The analysis has been implemented in an Excel spreadsheet, and outputs predicted Cu, In, and Ga profiles for entered deposition recipes. A single set of diffusion coefficients and activation energies are chosen, such that simulated elemental profiles track with published data and those from this study. Extent and limits of agreement between elemental profiles predicted from the growth recipes and the spreadsheet tool are demonstrated.

We show the results of nanotextured device designs combined with carefully placed nanogrids in order to minimize optical losses. Finite element method (FEM) based optical modeling indicates that the reflection of both the layer stack and the metal is diminished by the proposed configuration in which the metallic nanowires at the front of a device are placed into the relatively shallow crevices, whereby the metal is not covered by other materials of the cell stack. The electric field distribution and energy dissipation (i.e. absorption) diagrams of the texture show how the light is distributed and where it is absorbed. It shows that light is ‘concentrated’ in the tips of the texture (depending on the size and wavelength of the light). Simultaneously, for wavelengths above 750 nm there appears to be a reduction of the E-field in the lower part of the texture and, therefore, putting a metallic nanowire in this position has hardly any negative optical effect.

Furthermore, the impact of the texture height up to 1000 nm and the nanowire width up to 150nm was systematically investigated for a texture and wire period of 500 nm. The spectra reveal dimension dependent and wavelength specific optical features. This is the case even if the flat nanowire remains fully exposed to the front glass medium (i.e. not embedded underneath absorbers). At a texture height of 900 nm, the reflection related current loss is reduced by an order of magnitude compared to flat layer stacks, virtually regardless of the width of the metal nanowire. This opens up exciting new ways of creating nano-metal containing devices without the usual optical losses.

Differential scanning calorimetry experiments on mixed Cu2-xS, ZnS, and SnS2 precursors were conducted to better understand how Cu2ZnSnS4 (CZTS) and Cu2SnS3 form. The onset temperatures of Cu2SnS3 reactions and CZTS suggest that the ZnS phase may mediate Cu2SnS3 formation at lower temperatures before a final CZTS phase forms. We also found no evidence of a stable Cu2ZnSn3S8 phase. The major diffraction peaks associated with Cu2ZnSnS4, and Cu2SnS3 (overlaps with ZnS, as well) began to grow around 380 °C, although the final reaction to form Cu2ZnSnS4 probably did not occur until higher temperatures were reached. An exothermic reaction was observed corresponding to formation of this phase. There was some variability in the onset temperature for reactions to form Cu2SnS3. At least 5 steps are involved in this reaction and several segments of the reaction had relatively reproducible energies.