Water accelerates various modes of degradation in silicon photovoltaic (PV) modules including encapsulant yellowing, delamination, and contact corrosion. Moisture primarily enters silicon PV modules through the polymeric module components (encapsulant and backsheet). To mitigate moisture-induced degradation, we must understand the dynamics of moisture in state-of-the-art encapsulants and module architectures. In this work, we present a robust optical method to quantify water content on the front and rear side encapsulants in bifacial silicon PV modules and use such measurements to validate a model for simulating water concentration in these modules. First, we quantify the solubility and diffusivity of water in four modern encapsulants: ethylene vinyl acetate (EVA) and polyolefin (POE) both with and without UV-blocking additives. Second, we measure water concentration within glass-glass and glass-backsheet using water reflectometry detection (WaRD), tracking the diffusion of moisture throughout the modules as a function of time and environmental condition. Crucially we separate the water content from the front encapsulant and rear encapsulant and backsheet within the glass-backsheet modules. Third, we present a model of moisture dynamics in bifacial silicon PV modules and show it to be consistent with our measurements. Finally, we apply this model to simulate the concentration profile of water typical fielded climates as a function of time, environmental conditions, cell size, and module architecture. Overall, our work presents: 1) a quantitative picture of water dynamics in industrially relevant module architectures and field conditions, and 2) a framework to extend this approach other encapsulants and module designs.

With the focus of global energy production shifting towards renewables, efforts have been made to develop sustainable, low-cost technologies to cater to our needs while protecting our environment. The past decade has seen an exponential increase in the popularity of perovskite based solar cells with a push towards commercialization mainly due to its low processing cost, tunability and high efficiency. Despite the attractive advantages offered by perovskite solar cells, one of its shortcomings is its long-term stability. The research community has been invested in tackling issues related to understanding this water-perovskite interaction which poses a threat to device stability. It is known that moisture is one of the most challenging factors to tackle when it comes to degradation of these devices. The presence of water has a detrimental impact on the film’s composition, structure as well as optical properties. Thus, monitoring the uptake of water without destroying the nature of the film altogether is extremely crucial since it directly translates to optimizing the levelized cost of energy efficiency. Researchers have extensively studied the effects of this complex interaction, but there has not been an attempt to isolate the presence of moisture in these films in order to study the cause itself - Water. This thesis aims to study the interaction between water and perovskites using an in-situ and non-destructive technique based on the short-wave infrared (SWIR) spectroscopy.

In recent years, hybrid halide perovskite solar cells have attracted much attention due to their excellent optoelectronic performance. Leveraging the structural flexibility of perovskites, composition engineering appears to be an effective approach to improve material and device stability but can also raise durability challenges such as phase separation. In this presentation, I will share insights into the composition-performance relationships in these emerging materials resulting from correlative X-ray, electron, and optical microscopy investigations. Encompassing a wide range of perovskite chemistries, this work reveals the challenges and benefits of increasing compositional complexity and elucidates the roles of substituents and additives in these complex material systems. By identifying potential pitfalls in tuning the materials chemistry of halide perovskites, we provide a path forward for further improving their optoelectronic performance and making them commercially viable.

Electrochemical CO2 conversion into value-added chemical feedstocks, if powered with renewable electricity, is a promising path to approach carbon neutrality. However, developing active and selective catalysts is necessary to enable energy efficient conversion. In this thesis, I provide strategies to design active electrocatalysts with enhanced selectivity for CO2 electrolysis via tuning of the local reaction environment. First, I show that modulating the surface strain of a model Cu (001) catalyst epitaxially grown on a single-crystal Si substrate results in an increase of the Cu 3d band center position. This change in the electronic structure causes a suppression of the CO production pathway, increasing selectivity to multi-carbon products. I demonstrate that multi-carbon products can be further selected for by an operando restructuring of the micro- and nanoscale morphology of Cu-based catalysts. An electrochemically reduced Cu(OH)2 nanowire catalyst enhances the selectivity of multi-carbon products at moderate electrolysis potentials where a hierarchical morphology evolves. A final level is the tuning of the chemical state of the catalyst surface. By switching the initial surface oxidation state of tin oxide catalysts to be SnII-rich, the selectivity and energy efficiency of formate generation are promoted, offering a possible nearest-term path to carbon-negative CO2 electrolysis. Optimizing electrolyzer design is also crucial to facilitate mass transport of CO2 to reach industrial relevance. The importance of the mass transport of CO2 to enhance overall activity is also demonstrated by controlling flow rate of partially concentrated CO2 stream. These findings reveal the importance of surface strain, morphology and chemical state on designing efficient CO2 catalysts, providing fundamental guidelines and direction toward carbon neutral CO2 conversion.

ABSTRACT OF THE THESIS

Liquid Phase Epitaxy Doping for High-Performance Emitters in Silicon Solar Cells

by

Tulika Rastogi

Master of Science

University of California San Diego, 2018

Professor David Fenning, Chair

Recombination in the emitter is a primary contributor to recombination current in modern industrial silicon solar cells. This loss is primarily due to recombination that occurs because of the inflexible constraints of solid-state diffusion kinetics during conventional emitter formation that result in a heavily-doped, difficult-to-passivate front surface. P-n junction formation by gas diffusion is also one of the costliest steps in the cell process due to equipment capital expenditure, the large thermal budget required, and the low-throughput batch nature of the process [1].

Liquid phase epitaxy doping (LPE-D) is proposed as a facile low-temperature, low-cost replacement for boron and/or phosphorus diffusion. This doping process decouples the emitter formation from the slow solid-state kinetics of gas diffusion. The doping profile formed is largely time-independent – theoretically, doping could be done in a brief firing step or during bulk impurity gettering, allowing throughput to be optimized without constraining cell architecture in next generation solar cells. The liquid phase epitaxy approach results in a nearly uniform doping concentration profile with the potential for significantly lower saturation currents and improved surface passivation in comparison to existing and proposed industrial diffused emitters. We examine the performance of emitters doped via liquid phase epitaxy using simulation and present experimental demonstration.

Perovskite photovoltaics hold promise to accelerate the world’s transition to renewable energy by greatly reducing the cost of solar energy. While promising, perovskite commercialization is hindered by short cell lifetimes, and thus researchers use accelerated degradation testing to quickly engineer improvements. However, commercial test equipment is cost prohibitive, and many research groups instead design custom systems. This work presents the novel design of an accelerated and inexpensive testing enclosure for perovskites, and describes the engineering challenges of light, temperature, and electrical tests. Currently, the system is designed for a capacity of 45, 1 cm2 cells, capable of illuminating at 7.2 times the sun’s peak intensity, up to 85C, with a leak rate of 0.3%/hr, and 4-point electrical connection to a custom maximum power-point (MPP) circuit. Total cost of the system is ~$2.5k and is compact; cells can be transferred directly within a glovebox. This paper presents the thermal design, modeling, and challenges of high temperatures. We then present the structural and electrical design and conclude with a cost and failure analysis. The broader goal of this work is to contribute design guidelines for laboratories designing inexpensive, custom enclosures, in hopes of accelerating perovskites’ commercialization.

Transforming carbon dioxide (CO2) into valuable chemical feedstocks through electrochemical processes powered by renewable electricity shows promise in achieving carbon neutrality. However, the development of effective and selective catalysts is essential for enabling energy-efficient conversion. The activity and selectivity of the reaction towards formic acid production are governed by the stabilization of the *OCHO intermediate on the catalyst surface. This thesis presents strategies for designing active electrocatalysts that exhibit improved selectivity towards formic acid in CO2 electrolysis by manipulating the chemical environment surrounding the catalyst. Introducing a partial positively charged copper species on the surface in the form of copper sulfide stabilizes the *OCHO intermediate on the catalyst surface. The ligand plays a key role in supplying the proton required for the reduction of CO2 to formate, and that an optimum pKa value ligand is beneficial for improved selectivity towards formate. Another aspect involves tuning the chemical state of tin oxide catalyst surfaces. By favoring a Sn (II) rich initial surface oxidation state, the selectivity and energy efficiency of formate generation are improved, offering a potential near-term solution for carbon-negative CO2 electrolysis. Optimal design of electrolyzers is also crucial for facilitating the mass transport of CO2, thereby increasing industrial relevance. These discoveries underscore the importance of chemical environment surrounding the catalyst, and chemical state in the efficient design of CO2 reduction catalysts to formic acid and providing fundamental guidelines and direction towards achieving carbon- neutral CO2 conversion.

Low-cost solar energy, a critical component in our future energy infrastructure, requires durable photovoltaic materials with long operating lifetimes. I work to improve this durability in two photovoltaic technologies: silicon and halide perovskites. For silicon, I develop a method to quantify moisture within standard modules, then examine the local impacts of water on performance losses in industrially-relevant module architectures over a range of environmental conditions. For halide perovskites, I develop an automation platform to rapidly synthesize and test absorbers with high precision, and apply it to screen for durable wide bandgap absorbers. The development of this platform yields software components of general interest for automation of materials research. Throughout my work I highlight the development of tailor-made tools that enable high-throughput investigation of photovoltaic materials.

Hybrid perovskite solar cells (PSCs) have received considerable attention in the hopes of a paradigm shift towards low-cost, high-efficiency light energy harvesting. In just a decade, researchers have increased the record certified power conversion efficiency (PCE) of perovskite solar cells (PSCs) from 3.8% to a record 23.3%. Much of this research has been conducted in lab-scale settings. However, to compete in the current energy sector, solutions to large-scale, cost-effective manufacturing must be engineered. Recent progress from various groups have presented viable solutions to scaled perovskite synthesis, often focusing on the absorber layer, while neglecting other necessary layers. An integral part of PSCs are electron transport layers (ETLs). ETLs serve as an electron selective barrier at the cathode and facilitate the ability to extract high currents. To date, the large-area PSC community has systematically researched scalability with insufficient reports on transport layer commercialization, specifically SnO2 ETL industrial manufacturability. Vacuum thermal evaporation can reproducibly deposit high purity thin films, over large areas, at high volume. Combined with ease of automation, it is a highly compatible technique for assembly-line manufacturing. Herein, we report the use of vacuum thermally evaporated (vte) SnO2 ETLs to produce stable and efficient PSCs. The fabricated vte- SnO2 ETLs excellent coverage, confirmed through cyclic voltammetry, on substrates ranging from 5 cm2 up to 100 cm2. Additionally, PSC devices fabricated with VTE-SnO2 ETLs on average display slightly higher PCE, ~16%, than our standard solution processed SnO2 ETLs, ~15.7%. These results help pave the way for commercial manufacturing of perovskite solar modules.

Perovskite materials are emerging as promising top-cell candidates for high-efficiency silicon tandem solar cells due to their sharp absorption onsets, long carrier lifetimes/diffusion lengths, and tunable bandgaps. However, to achieve commercialization, challenges such as their metastability in operating conditions and their irreproducibility due to sensitivity to process variance must be addressed. Hence the Perovskite Automated Spin Coat Assembly Line – PASCAL – is introduced as a materials acceleration platform for the deposition and characterization of spin-coated thin films, with specific application to halide perovskites. We first demonstrate improved consistency of perovskite film fabrication by tightly controlling process parameters uniquely exposed under the automated experimental framework. Next, we conduct 3 case studies using PASCAL to 1) conduct composition screening spanning the triple-cation, triple-halide composition space, producing a machine learning assisted composition with film durability under intense illumination and heat. 2) Leverage PASCAL to conduct series of experiments to conformally coat perovskite on commercially textured silicon enabling a tandem device prototype. 3) Perform standardized tests on PASCAL-produced thin films using a Bayesian optimization framework to accelerate durability research and develop regression models linking PASCAL's cost-effective metrology to costly standard tests. The approach, hardware, and data detailed herein highlight automated platforms as an opportunity to accelerate the identification and discovery of novel thin film materials and demonstrates the efficacy of PASCAL specifically for automation of solution-processed optoelectronic thin film research.