The Future of Material Scientists in an Age of Artificial Intelligence.

Material science has historically evolved in tandem with advancements in technologies for characterization, synthesis, and computation. Another type of technology to add to this mix is machine learning (ML) and artificial intelligence (AI). Now increasingly sophisticated AI-models are seen that can solve progressively harder problems across a variety of fields. From a material science perspective, it is indisputable that machine learning and artificial intelligence offer a potent toolkit with the potential to substantially accelerate research efforts in areas such as the development and discovery of new functional materials. Less clear is how to best harness this development, what new skill sets will be required, and how it may affect established research practices. In this paper, those question are explored with respect to increasingly more sophisticated ML/AI-approaches. To structure the discussion, a conceptual framework of an AI-ladder is introduced. This AI-ladder ranges from basic data-fitting techniques to more advanced functionalities such as semi-autonomous experimentation, experimental design, knowledge generation, hypothesis formulation, and the orchestration of specialized AI modules as stepping-stones toward general artificial intelligence. This ladder metaphor provides a hierarchical framework for contemplating the opportunities, challenges, and evolving skill sets required to stay competitive in the age of artificial intelligence.

Data-Driven Tunnel Oxide Passivated Contact Solar Cell Performance Analysis Using Machine Learning.

Tunnel oxide passivated contacts (TOPCon) have gained interest as a way to increase the energy conversion efficiency of silicon solar cells, and the International Technology Roadmap of Photovoltaics forecasts TOPCon to become an important technology despite a few remaining challenges. To review the recent development of TOPCon cells, this work has compiled a dataset of all device data found in current literature, which sums up to 405 devices from 131 papers. This may seem like a surprisingly small number of cells given the recent interest in the TOPCon architecture, but it illustrates a problem of data dissemination in the field. Notwithstanding the limited number of cells, there is a great diversity in cell manufacturing procedures, and this work observes a gradual increase in performance indicating that the field has not yet converged on a set of best practices. By analyzing the data using statistical methods and machine learning (ML) algorithms, this work is able to reinforces some commonly held hypotheses related to the performance differences between different device architectures. This work also identifies a few more unintuitive feature combinations that would be of interest for further experimentally studies. This work also aims to inspire improvements in data management and dissemination within the TOPCon community.

A systematic discrepancy between the short circuit current and the integrated quantum efficiency in perovskite solar cells.

Halide perovskites solar cells are now approaching commercialisation. In this transition from academic research towards industrialisation, standardized testing protocols and reliable dissemination of performance metrics are crucial. In this study, we analyze data from over 16,000 publications in the Perovskite Database to investigate the assumed equality between the integrated external quantum efficiency and the short circuit current from JV measurements. We find a systematic discrepancy with the JV-values being on average 4% larger. This discrepancy persists across time, perovskite composition, and device architecture, indicating the need to explore new perovskite physics and update reporting protocols and assumptions in the field.

Direct Measurements of Interfacial Photovoltage and Band Alignment in Perovskite Solar Cells Using Hard X-ray Photoelectron Spectroscopy.

A heterojunction is the key junction for charge extraction in many thin film solar cell technologies. However, the structure and band alignment of the heterojunction in the operating device are often difficult to predict from calculations and, due to the complexity and narrow thickness of the interface, are difficult to measure directly. In this study, we demonstrate a technique for direct measurement of the band alignment and interfacial electric field variations of a fully functional lead halide perovskite solar cell structure under operating conditions using hard X-ray photoelectron spectroscopy (HAXPES). We describe the design considerations required in both the solar cell devices and the measurement setup and show results for the perovskite, hole transport, and gold layers at the back contact of the solar cell. For the investigated design, the HAXPES measurements suggest that 70% of the photovoltage was generated at this back contact, distributed rather equally between the hole transport material/gold interface and the perovskite/hole transport material interface. In addition, we were also able to reconstruct the band alignment at the back contact at equilibrium in the dark and at open circuit under illumination.

Big data driven perovskite solar cell stability analysis.

During the last decade lead halide perovskites have shown great potential for photovoltaic applications. However, the stability of perovskite solar cells still restricts commercialization, and lack of properly implemented unified stability testing and disseminating standards makes it difficult to compare historical stability data for evaluating promising routes towards better device stability. Here, we propose a single indicator to describe device stability that normalizes the stability results with respect to different environmental stress conditions which enables a direct comparison of different stability results. Based on this indicator and an open dataset of heterogeneous stability data of over 7000 devices, we have conducted a statistical analysis to assess the effect of different stability improvement strategies. This provides important insights for achieving more stable perovskite solar cells and we also provide suggestions for future directions in the perovskite solar cell field based on big data utilization.

The Complex Degradation Mechanism of Copper Electrodes on Lead Halide Perovskites.

Lead halide perovskite solar cells have reached power conversion efficiencies during the past few years that rival those of crystalline silicon solar cells, and there is a concentrated effort to commercialize them. The use of gold electrodes, the current standard, is prohibitively costly for commercial application. Copper is a promising low-cost electrode material that has shown good stability in perovskite solar cells with selective contacts. Furthermore, it has the potential to be self-passivating through the formation of CuI, a copper salt which is also used as a hole selective material. Based on these opportunities, we investigated the interface reactions between lead halide perovskites and copper in this work. Specifically, copper was deposited on the perovskite surface, and the reactions were followed in detail using synchrotron-based and in-house photoelectron spectroscopy. The results show a rich interfacial chemistry with reactions starting upon deposition and, with the exposure to oxygen and moisture, progress over many weeks, resulting in significant degradation of both the copper and the perovskite. The degradation results not only in the formation of CuI, as expected, but also in the formation of two previously unreported degradation products. The hope is that a deeper understanding of these processes will aid in the design of corrosion-resistant copper-based electrodes.

X-ray stability and degradation mechanism of lead halide perovskites and lead halides.

Lead halide perovskites have become a leading material in the field of emerging photovoltaics and optoelectronics. Significant progress has been achieved in improving the intrinsic properties and environmental stability of these materials. However, the stability of lead halide perovskites to ionising radiation has not been widely investigated. In this study, we investigated the radiolysis of lead halide perovskites with organic and inorganic cations under X-ray irradiation using synchrotron based hard X-ray photoelectron spectroscopy. We found that fully inorganic perovskites are significantly more stable than those containing organic cations. In general, the degradation occurs through two different, but not mutually exclusive, pathways/mechanisms. One pathway is induced by radiolysis of the lead halide cage into halide salts, halogen gas and metallic lead and appears to be catalysed by defects in the perovskite. The other pathway is induced by the radiolysis of the organic cation which leads to formation of organic degradation products and the collapse of the perovskite structure. In the case of Cs0.17FA0.83PbI3, these reactions result in products with a lead to halide ratio of 1 : 2 and no formation of metallic lead. The radiolysis of the organic cation was shown to be a first order reaction with regards to the FA+ concentration and proportional to the X-ray flux density with a radiolysis rate constant of 1.6 × 10-18 cm2 per photon at 3 keV or 3.3 cm2 mJ-1. These results provide valuable insight for the use of lead halide perovskite based devices in high radiation environments, such as in space environments and X-ray detectors, as well as for investigations of lead halide perovskites using X-ray based techniques.

SnO x Atomic Layer Deposition on Bare Perovskite-An Investigation of Initial Growth Dynamics, Interface Chemistry, and Solar Cell Performance.

High-end organic-inorganic lead halide perovskite semitransparent p-i-n solar cells for tandem applications use a phenyl-C61-butyric acid methyl ester (PCBM)/atomic layer deposition (ALD)-SnO x electron transport layer stack. Omitting the PCBM would be preferred for manufacturing, but has in previous studies on (FA,MA)Pb(Br,I)3 and (Cs,FA)Pb(Br,I)3 and in this study on Cs0.05FA0.79MA0.16PbBr0.51I2.49 (perovskite) led to poor solar cell performance because of a bias-dependent light-generated current. A direct ALD-SnO x exposure was therefore suggested to form a nonideal perovskite/SnO x interface that acts as a transport barrier for the light-generated current. To further investigate the interface formation during the initial ALD SnO x growth on the perovskite, the mass dynamics of monitor crystals coated by partial p-i-n solar cell stacks were recorded in situ prior to and during the ALD using a quartz crystal microbalance. Two major finds were made. A mass loss was observed prior to ALD for growth temperatures above 60 °C, suggesting the decomposition of the perovskite. In addition, a mostly irreversible mass gain was observed during the first exposure to the Sn precursor tetrakis(dimethylamino)tin(IV) that is independent of growth temperature and that disrupts the mass gain of the following 20-50 ALD cycles. The chemical environments of the buried interface were analyzed by soft and hard X-ray photoelectron spectroscopy for a sample with 50 ALD cycles of SnO x on the perovskite. Although measurements on the perovskite bulk below and the SnO x film above did not show chemical changes, additional chemical states for Pb, Br, and N as well as a decrease in the amount of I were observed in the interfacial region. From the analysis, these states and not the heating of the perovskite were concluded to be the cause of the barrier. This strongly suggests that the detrimental effects can be avoided by controlling the interfacial design.

Degradation Mechanism of Silver Metal Deposited on Lead Halide Perovskites.

Lead halide perovskite solar cells have significantly increased in both efficiency and stability over the last decade. An important aspect of their long-term stability is the reaction between the perovskite and other materials in the solar cell. This includes the contact materials and their degradation if they can potentially come into contact through, e.g., pinholes or material diffusion and migration. Here, we explore the interactions of silver contacts with lead halide perovskites of different compositions by using a model system where thermally evaporated silver was deposited directly on the surface of the perovskites. Using X-ray photoelectron spectroscopy with support from scanning electron microscopy, X-ray diffraction, and UV-visible absorption spectroscopy, we studied the film formation and degradation of silver on perovskites with different compositions. The deposited silver does not form a continuous silver film but instead tends to form particles on a bare perovskite surface. These particles are initially metallic in character but degrade into AgI and AgBr over time. The degradation and migration appear unaffected by the replacement of methylammonium with cesium but are significantly slowed down by the complete replacement of iodide with bromide. The direct contact between silver and the perovskite also significantly accelerates the degradation of the perovskite, with a significant loss of organic cations and the possible formation of PbO, and, at the same time, changed the surface morphology of the iodide-rich perovskite interface. Our results further indicate that an important degradation pathway occurred through gas-phase perovskite degradation products. This highlights the importance of control over the interface materials and the use of completely hermetical barrier layers for the long-term stability and therefore the commercial viability of silver electrodes.

Photoinduced Stark Effects and Mechanism of Ion Displacement in Perovskite Solar Cell Materials.

Organometallic halide perovskites (OMHPs) have recently emerged as a promising class of materials in photovoltaic technology. Here, we present an in-depth investigation of the physics in these systems by measuring the photoinduced absorption (PIA) in OMHPs as a function of materials composition, excitation wavelength, and modulation frequency. We report a photoinduced Stark effect that depends on the excitation wavelength and on the dipole strength of the monovalent cations in the A position of the ABX3 perovskite. The results presented are corroborated by density functional theory calculations and provide fundamental information about the photoinduced local electric field change under blue and red excitation as well as insights into the mechanism of light-induced ion displacement in OMHPs. For optimized perovskite solar cell devices beyond 19% efficiency, we show that excess thermalization energy of blue photons plays a role in overcoming the activation energy for ion diffusion.

Unreacted PbI2 as a Double-Edged Sword for Enhancing the Performance of Perovskite Solar Cells.

Lead halide perovskites have over the past few years attracted considerable interest as photo absorbers in PV applications with record efficiencies now reaching 22%. It has recently been found that not only the composition but also the precise stoichiometry is important for the device performance. Recent reports have, for example, demonstrated small amount of PbI2 in the perovskite films to be beneficial for the overall performance of both the standard perovskite, CH3NH3PbI3, as well as for the mixed perovskites (CH3NH3)x(CH(NH2)2)(1-x)PbBryI(3-y). In this work a broad range of characterization techniques including X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), photo electron spectroscopy (PES), transient absorption spectroscopy (TAS), UV-vis, electroluminescence (EL), photoluminescence (PL), and confocal PL mapping have been used to further understand the importance of remnant PbI2 in perovskite solar cells. Our best devices were over 18% efficient, and had in line with previous results a small amount of excess PbI2. For the PbI2-deficient samples, the photocurrent dropped, which could be attributed to accumulation of organic species at the grain boundaries, low charge carrier mobility, and decreased electron injection into the TiO2. The PbI2-deficient compositions did, however, also have advantages. The record Voc was as high as 1.20 V and was found in PbI2-deficient samples. This was correlated with high crystal quality, longer charge carrier lifetimes, and high PL yields and was rationalized as a consequence of the dynamics of the perovskite formation. We further found the ion migration to be obstructed in the PbI2-deficient samples, which decreased the JV hysteresis and increased the photostability. PbI2-deficient synthesis conditions can thus be used to deposit perovskites with excellent crystal quality but with the downside of grain boundaries enriched in organic species, which act as a barrier toward current transport. Exploring ways to tune the synthesis conditions to give the high crystal quality obtained under PbI2-poor condition while maintaining the favorable grain boundary characteristics obtained under PbI2-rich conditions would thus be a strategy toward more efficiency devices.

Phase Formation Behavior in Ultrathin Iron Oxide.

Nanostructured iron oxides, and especially hematite, are interesting for a wide range of applications ranging from gas sensors to renewable solar hydrogen production. A promising method for deposition of low-dimensional films is atomic layer deposition (ALD). Although a potent technique, ALD of ultrathin films is critically sensitive to the substrate and temperature conditions where initial formation of islands and crystallites influences the properties of the films. In this work, deposition at the border of the ALD window forming a hybrid ALD/pulsed CVD (pCVD) deposition is utilized to obtain a deposition less sensitive to the substrate. A thorough analysis of iron oxide phases formation on two different substrates, Si(100) and SiO2, was performed. Films between 3 and 50 nm were deposited and analyzed with diffraction techniques, high-resolution Raman spectroscopy, and optical spectroscopy. Below 10 nm nominal film thickness, island formation and phase dependent particle crystallization impose constraints for deposition of phase pure iron oxides on non-lattice-matching substrates. Films between 10 and 20 nm thickness on SiO2 could effectively be recrystallized into hematite whereas for the corresponding films on Si(100), no recrystallization occurred. For films thicker than 20 nm, phase pure hematite can be formed directly with ALD/pCVD with very low influence of the substrate on either Si or SiO2. For more lattice matched substrates such as SnO2:F, Raman spectroscopy indicated formation of the hematite phase already for films with 3 nm nominal thickness and clearly for 6 nm films. Analysis of the optical properties corroborated the analysis and showed a quantum confined blue-shift of the absorption edge for the thinnest films.

Determination of Thermal Expansion Coefficients and Locating the Temperature-Induced Phase Transition in Methylammonium Lead Perovskites Using X-ray Diffraction.

Lead halogen perovskites, and particularly methylammonium lead iodine, CH3NH3PbI3, have recently attracted considerable interest as alternative solar cell materials, and record solar cell efficiencies have now surpassed 20%. Concerns have, however, been raised about the thermal stability of methylammonium lead iodine, and a phase transformation from a tetragonal to a cubic phase has been reported at elevated temperature. Here, this phase transition has been investigated in detail using temperature-dependent X-ray diffraction measurements. The phase transformation is pinpointed to 54 °C, which is well within the normal operating range of a typical solar cell. The cell parameters were extracted as a function of the temperature, from which the thermal expansion coefficient was calculated. The latter was found to be rather high (αv = 1.57 × 10(-4) K(-1)) for both the tetragonal and cubic phases. This is 6 times higher than the thermal expansion coefficient for soda lime glass and CIGS and 11 times larger than that of CdTe. This could potentially be of importance for the mechanical stability of perovskite solar cells in the temperature cycling experienced under normal day-night operation. The experimental knowledge of the thermal expansion coefficients and precise determination of the cell parameters can potentially also be valuable while conducting density functional theory simulations on these systems in order to deliver more accurate band structure calculations.

A size dependent discontinuous decay rate for the exciton emission in ZnO quantum dots.

The time resolved UV-fluorescence in ZnO quantum dots has been investigated using femtosecond laser spectroscopy. The measurements were performed as a function of particle size for particles between 3 and 7 nm in diameter, which are in the quantum confined regime. A red shift in the fluorescence maximum is seen while increasing the particle size, which correlates with the shift in band gap due to quantum confinement. The energy difference between the UV-fluorescence and the band gap does, however, increase for the smaller particles. For 3.7 nm particles the fluorescence energy is 100 meV smaller than the band gap energy, whereas it is only 20 meV smaller for the largest particles. This indicates a stabilization of the excitons in the smallest particles. The lifetime of the UV fluorescence is in the picosecond time scale and interestingly, it is discontinuous with respect to particle size. For the smallest particles, the exciton emission life time reaches 30 ps, which is three times longer than that for the largest particles. This demonstrates a transition between two different mechanisms for the UV-fluorescence. We suggest that this is an effect of surface trapping and stabilization of the excitons occurring in the smallest particles but not in the larger ones. We also discuss the time scale limit for slowed hot carrier dynamics in ensembles of quantum confined ZnO particles.

Absorption and fluorescence spectroscopy of growing ZnO quantum dots: size and band gap correlation and evidence of mobile trap states.

ZnO nanoparticles constitute a convenient model system for fundamental studies with many possible technical applications in, for example, sensors and the field of catalysis and optoelectronics. A large set of ZnO quantum dots in the size range 2.5-7 nm have been synthesized and analyzed in detail. Time resolved in situ UV-vis absorption measurements were used to monitor the growth of these particles in solution by correlating the optical band gap to particle size given from X-ray diffraction (XRD) measurements. The particles formed were isotropic in shape, but small initial deviations gave indications of a transition from thermodynamic to kinetically controlled growth for particles around 4 nm in diameter. On the basis of this, the behavior and mechanisms for the particle growth are discussed. The fluorescence dependence on particle size was investigated by combining fluorescence and UV-vis measurements on growing particles. This revealed that the positions of the fluorescence trap states are mobile toward the conduction- and valence band. A broadening of the trap states was also found, and a surface dependent mechanism of the trap state shift and broadening is proposed.