​On November 22nd KSC's Rahim Munir presented his research and successfully defended his Ph.D. thesis.

Regarding his time at KSC, Rahim commented, "Being inside the boundaries of KAUST actually eliminates all the boundaries of research with ample opportunity to utilize top-notch equipment.....  Especially, I feel home at KAUST Solar Center. I am glad to defend my thesis and will be sad to leave this place."

Congratulations to Dr. Rahim Munir and we wish him all the very best for his future career.

PhD Dissertation

Hybrid Perovskite Thin Film Formation: From Lab Scale Spin Coating to Large Area Blade Coating

Rahim Munir

​ABSTRACT: Our reliance on semiconductors is on the rise with the ever growing use of electronics in our daily life. Renewable energy solutions such as solar water heaters, solar cells, etc. also fall under the umbrella of semiconductor applications. The amount of input energy required to produce these semiconductors, such as Silicon, GaAs, CuInGaSe, CdTe, and InGaAs, is humungous as the process requires deep vacuum and high temperatures. Moreover, many of the high-performance inorganic semiconductors use scarce elements like indium, gallium, and tellurium. These problems provide an opportunity for the research community to hunt for alternative low-cost solutions. Organic-inorganic hybrid lead halide perovskites have emerged as a prime alternative to current standard semiconductors because of its use of abundant elements and the ease of solution processing. 

Organic-inorganic hybrid lead halide perovskite is generally labelled as an AMX₃ compound, where A is an organic cation, most often methylammonium (CH₃NH₃⁺) or formamidinium (HC(NH₂)₂⁺), M is a metal, such as lead (Pb) or tin (Sn), and X is a iodine (I), bromine (Br) and/or chlorine (Cl). In 2012, there were only 18 publications in the field of hybrid perovskites, with the highest reported photovoltaic power conversion efficiency of 10.9 %. In 2016, more than 2200 research articles were published in the field of hybrid perovskites, with the highest power conversion efficiency reported of more than 22.1%. This is one of the fastest developments in device performance among all other photovoltaic devices. This material has received significant attention in the semiconductor research community because of its excellent optoelectronic properties. Its high charge mobility and high diffusion length make it an ideal candidate for photovoltaic applications, while its ability to be processed from solution opens up opportunities to use this material in flexible electronics.

This thesis has shed light on the ink-to-solid conversion during the one-step solution process of hybrid perovskite formulations from DMF. We utilize a suite of in situ diagnostic probes including high speed optical microscopy, optical reflectance and absorbance, and grazing incidence wide angle x-ray scattering (GIWAXS), all performed during spin coating, to monitor the solution thinning behavior, changes in optical absorbance, and nucleation and growth of crystalline phases of the precursor and perovskite. The starting formulation experiences solvent-solute interactions within seconds of casting, leading to the formation of a wet gel with nanoscale features visible by in situ GIWAXS. The wet gel alters the thinning behavior of the blank solvent most likely due to changing viscosity and reduced effective evaporation rate of the solvent. The wet gel subsequently gives way to the formation of ordered precursor solvates (equimolar iodide and chloride solutions) or disordered precursor solvates (equimolar bromide or 3:1 chloride), depending upon the halide and MAI content. The ordered precursor solute phases are stable and retain the solvent for long durations, resulting in consistent conversion behavior to the perovskite phase and solar-cell performance. However, in the 3:1 MAI:PbCl₂ case, the excess MAI leads to more solvent retention and disrupts the precursor order, leaving some of the solvent in a volatile state. The as-cast precursor film mesostructure and composition, therefore, evolve with time, making the timing of the conversion process crucial for high-quality film formation, as reflected in the sensitivity of solar-cell performance to the timing of thermal conversion.

In this thesis, we develop a firm understanding of the solvent engineering process in which an antisolvent is used during the coating process through the solvent mixture of GBL and DMSO in different ratios. It has been shown that solvent engineering produce pin hole-free films, justifying its wide adoption across the field. 

Our investigations reveal that the choice of processing solvent, namely DMSO vs. GBL, strongly impacts the effectiveness of the anti-solvent drip. In all instances, we show that the drip must be applied prior to solidification and crystallization of the drying ink to prevent excessive crystallization of the precursor solvates, making the timing of the drip extremely crucial to adjust with respect to the onset of solidification and crystallization of the perovskite ink. We explore mixed cations systems and found out that these systems are dominated by hexagonal phase rather than the solvates. However, antisolvent drip, in both cases, helps nucleate the perovskite phase. 

We then translate our learnings from the lab scale spin coating process to the industrial friendly blade coating process. Spin coating is a wasteful process which cannot be easily scaled up to continuous large area fabrication, where existing solvent engineering methods, such as anti-solvent dripping, are also unlikely to work. Here we compare the ink solidification and film formation mechanisms of CH₃NH₃PbI₃ in solutions we used to understand the key scientific insights through spin coating. We show significant differences in the process kinetics and formation of complex intermediate phases between the two processes at room and intermediate temperatures. To overcome these challenges in the context of blade coating, the sample is heated during deposition. We observe high-quality film formation for T > 100^oC, namely in conditions which inhibit the formation of the crystalline intermediate complex phases. In doing so, we achieve fast and direct formation of the perovskite phase with solar cells yielding PCE > 17%.

We believe that the insights present in this thesis will help improve thin film manufacturability both from a performance perspective and by increasing the yield and reproducibility of perovskite based films and related devices.