UC Santa Barbara

Among primary energy sources that do not generate carbon dioxide through the oxidation of fossil fuels, photovoltaics (PV)—which convert sunlight directly to electricity—are a promising approach. The solar resource is ample and well-characterized, pollutant emissions from PV power plants are negligible, and the technology is suited to nearly all climates and regions on earth. While the cost of solar electricity has plummeted in the last decade, further cost reductions and efficiency improvements in photovoltaic absorber materials may make this technology more accessible and ubiquitous.

To this end, there is much excitement about main-group halide perovskites, which have garnered intense research attention since 2009 when they were employed successfully in solar cells. These materials, which comprise main-group dications (Ge2+, Sn2+, Pb2+), halide anions (Cl-, Br-, I-), and large countercations (Cs+, [CH3NH3]+, [CH(NH2)2]+) crystallized in the perovskite structure, combine excellent performance in PV and other optoelectronic applications with ease of preparation and abundant constituent elements. Despite some years of in-depth study, key fundamental questions and practical challenges remain. In particular, the remarkable properties of these systems confound the conventional wisdom of what constitutes a high-performance semiconductor, and the origins of these favorable properties remain a matter of significant debate. Concurrently, practical application of devices employing halide perovskites is hampered by stability challenges, and the presence of lead in high-performing formulations raises significant neurotoxicity and environmental contamination concerns.

In this dissertation, we report research in three thematic areas aimed at resolving fundamental questions around the materials chemistry, crystal structure, and plastic crystal dynamics of these halide perovskites. In the first portion, the preparation and characterization of lead-free inorganic bismuth halides illustrate the importance of the electronic configuration of the main-group cation and of highly-connected, high-symmetry crystal structures to the favorable optoelectronic properties of the tin and lead halides. Subsequently, scattering experiments and ab initio calculations reveal an unusual and chemically-tunable form of dynamic disorder arising from an electronic instability associated with the main-group cations which affects thermal, dielectric, and electronic properties. Finally, a range of spectroscopic, computational, and scattering techniques are employed to establish the nature of molecular motion and its effects on the crystal structures of the high performance photovoltaic absorber, formamidinium lead iodide, providing critical context for the evaluation of hypotheses about the origins of the remarkable properties of these materials.

In an effort to understand the crystal chemistry of the halide perovskites, lead-free alkali bismuth iodides and the binary bismuth iodide are prepared in single crystal, bulk, and thin film forms, and their structures resolved via X-ray diffraction and 87-Rb solid-state nuclear magnetic resonance (NMR) spectroscopy. These phases are shown to exhibit strong optical absorption and suitable bandgaps for single junction and tandem solar cells, but photoemission spectroscopy and ab initio calculations based on density functional theory reveal valence band maxima that are deep relative to existing hole transport materials. This poor band alignment is demonstrated to be a consequence of relativistic stabilization of the Bi 6s orbital combined with reduced bandwidths from distorted Bi coordination environments, thus establishing the importance of the high symmetry structures seen for the divalent tin and lead halides.

X-ray total scattering studies reveal local off-centering of the main-group cations (Sn2+, Pb2+) within their coordination octahedra across the halide perovskites, reflecting a preference for lower symmetry coordination than that implied by crystallographic approaches. Taking CsSnBr3 as an exemplar of the broader class of materials, ab initio calculations, photoluminescence measurements, and analogies to existing theory implicate the Sn 5s2 lone pair electrons (equivalently, the pseudo-Jahn–Teller effect) as the origin of this phenomenon, which we propose leads to enhanced defect screening, reduced thermal conductivity, and unusual temperature-dependence of the electronic bandgap. We further demonstrate control of the strength of this phenomenon in the hybrid tin and lead perovskites by chemical substitution on all sites of the crystal, with a lighter carbon-group dication, a lighter halogen, and a larger countercation all leading to more pronounced off-centering. This proximity to a polar phase boundary leads to an elevated lattice polarizability and suggests a possible mechanism for the formation of large polarons, which have been proposed as the origin of long-lived carriers and modest carrier mobilities in these materials.

As plastic crystals—which exhibit translational periodicity but orientational disorder—the hybrid organic–inorganic perovskites display a complex interplay between motion of the molecular cations and the structure of the surrounding anionic inorganic framework. High resolution X-ray diffraction reveals the complete phase evolution with temperature of formamidinium lead iodide, including an unusual reentrant pseudosymmetry at cryogenic temperatures arising from geometric frustration between the molecular symmetry and the favored ground-state tilting of the inorganic octahedra. Solid-state 1-H NMR and dielectric spectroscopies, calorimetry, ab initio calculations, and neutron total scattering establish the full temperature-dependent dynamics of molecular reorientation between 4 K and 400 K. Despite markedly different barriers for molecular rotation compared to those in the homologous methylammonium lead iodide, both systems exhibit similar dynamics at room temperature. Together with the vastly different dipole moments for the two molecules, this result clarifies emerging hypotheses of polaronic transport and fugitive spin polarization, suggesting the primacy of the main-group–halogen sublattice, rather than the molecular cations, for defect-tolerant electronic transport.