THE SCHOOL OF MATERIALS SCIENCE AND ENGINEERING  

GEORGIA INSTITUTE OF TECHNOLOGY    

Under the provisions of the regulations for the degree 
  

DOCTOR OF PHILOSOPHY 
  

on Tuesday, May 3, 2022 

3:00 PM 

MRDC 3515

will be held the 

DISSERTATION PROPOSAL DEFENSE 
 
for 

Andres Felipe Castro Mendez

"Thermal Evaporation for Deposition of Metal Halide Perovskites" 

Committee Members: 

Prof. Juan-Pablo Correa-Baena, Advisor, MSE 

Prof. Mark D. Losego, MSE 

Prof. Lauren Garten, MSE 

Prof. Natalie Stingelin, MSE/ChBE 

Prof. Angus P. Wilkinson, ChBE 

Abstract:

Hybrid organic-inorganic perovskites (HOIPs) are a family of materials with a structure ABX3 where A is an organic or inorganic cation, B is a metal cation, and X is a halide anion. These HOIP materials have exceptional optoelectronic properties such as a tunable bandgap, high absorption coefficient, and defect tolerance, which make them a great material for photovoltaic applications. Perovskite solar cells (PSCs) have achieved a record efficiency of 25.6% by using spin coating for the deposition of most of the layers involved. Careful control of the phase purity, morphology, and thus optoelectronic properties of the HOIP polycrystalline layer has been crucial to achieve these results. However, there are many drawbacks of the spin coating deposition approach, such as low throughput, extensive amounts of toxic solvents, low quality of films near the edges of the substrates, and the presence of large defects in the film. These drawbacks translate into roadblocks for thin film processing in large scale. Thermal evaporation is a promising deposition technique to address these concerns. However, there is a lack of understanding of the mechanisms that lead to evaporated HOIP films with optimal optoelectronic properties for solar cells, which results in devices with lower efficiencies than those obtained by solution processes. The goal of this thesis is to provide new understanding of the growth mechanisms of HOIPs films via evaporation to control the morphology, phase purity, stoichiometry, structural defects, and crystallographic orientation. The control of these characteristics is vital to obtain beneficial optoelectronic properties for PSCs such as an optimal bandgap, higher charge carrier mobility, and lower charge carrier recombination. 

This thesis is divided into three research thrusts, addressing three main concerns related to thermal evaporation of HOIP. The first thrust of the thesis focuses on understanding how to control the phase purity of the methylammonium lead triiodide (MAPbI3) HOIP in the thermal co-evaporation of lead iodide (PbI2) and methylammonium iodide (MAI). This thrust shows that the MAI evaporation rate has a key role in the stoichiometry of the film and the formation of secondary phases. The work shows a correlation between the formation of these phases and the performance of PSCs, identifying one of the main roadblocks to obtain high-performing devices via vapor deposition. The second thrust will study the co-evaporation of formamidinium lead triiodide (FAPbI3), a promising composition for high power conversion efficiency (PCE) and thermal stability. Unlike MAI, formamidinium iodide (FAI) has a high sticking coefficient, which allows for higher control of the stoichiometry of the film, and thus, better control of the properties of the material. This thrust will focus on the role intermolecular forces between the substrate and the perovskite layer play in the growth and stability of the corner-sharing perovskite photoactive FAPbI3 (alpha phase). The third thrust will exploit the advantages of thermal evaporation to fabricate stacks of multiple HOIP layers and HOIP layers with a gradient composition where the A-site cation (in the ABX3 structure) is modified. This thrust will study the phase purity and transport of charge carriers in a HOIP film with a composition CsxFA(1-x)PbI3. CsI is commonly used in HOIPs to suppress the non-perovskite delta phase of FAPbI3, but the bandgap increases from 1.43 eV to 1.74 eV for FA-based to Cs-based HOIPs, respectively. This bandgap shift is undesirable as the maximum achievable efficiency, the Shockley-Queisser limit, is at around 1.1 and 1.4 eV, and it drops as the bandgap increases. This thrust hypothesizes that the fabrication of HOIP film with a gradient composition of Cs will allow the epitaxial growth of the cubic phase, as it minimizes the strain in the structure, and will create a graded band structure that reduce the energy losses in the PSC.