The standard metric of a photovoltaic device is how efficient it is at converting the incident solar irradiance into useful electric power, which referred to as the power conversion efficiency (PCE). PCE determines the amount of material and active area required to deliver a specified power. Therefore optimizing PCE is critical aspect of developing any photovoltaic technology. This has traditionally been accomplished through top-down engineering approaches where a large parameter space (i.e. processing) is systematically varied until a local maximum may be achieved. This process can be both time and labor intensive and does not guarantee that all the interrelated parameters have been globally optimized. Therefore, I am broadly interested in developing a bottom-up engineering approach, where a fundamental understanding of material properties drives rational material design. Thus far, this methodology has yielded some of the highest quality solution-processed semiconducting materials to date (Braly, I.; deQuilettes, D.W. et al. 2017, Nature Photonics)
Many promising photovoltaic materials that are compatible with high-throughput manufacturing are deposited through solution-based processes (i.e. inkjet printing) or vapor techniques and self-assemble into polycrystalline and disordered nanostructured materials.
Figure 1 shows a top-view scanning electron microscope image of a potentially disruptive class of materials called metal halide perovskites. Note that the scale bar is 1 µm (that’s one millionth of a meter!)
The bulk performance and power output of a solar cell is dependent on this spatially varying structure and composition. I use specialized microscopes to glimpse into the nanoscale world and watch energy being transported and lost across defective surfaces, grain boundaries, and interfaces. I am particularly enthusiastic about understanding how local structure determines material and device function and have utilized confocal fluorescence microscopy techniques to non-destructively probe local variations in electronic properties. In my graduate studies at the University of Washington, we leveraged this understanding to locally control the extent of energy loss as well as modify the way that energy transports by changing the surface chemistry of metal halide perovskites (deQuilettes, D.W. et al. 2015, Science, 348, 683-686).
My future goals are to conduct in-situ microscopic measurements of energy loss and transport and further correlate this information with lattice motion, crystal structure, and surface chemistry. I view linear and non-linear optical microscopy as promising platforms to tackle these important questions and have therefore started to develop and implement both experimental methods and analytical techniques to gain further insight into the interrelation of electronic and structural dynamics.
When a solar cell absorbs a photon with energy (hν) greater or equal to the bandgap (Eg), a coulombically bound electron-hole pair is formed (exciton). If the photon energy is greater than the bandgap energy (hν>Eg), the electron and hole have excess kinetic energy before they relax to the conduction and valence band edges and are considered “hot”. In most solar cells, relaxation times of hot charge carriers are faster than the charge transfer processes that generate photocurrent. This limits the amount of energy that can be harnessed and results in low PCE’s. Shockley and Queisser predicted the upper theoretical PCE of a single p/n junction solar cell to be 33% under the assumption that all carriers relax to the ground state before charge separation.1 Solar cells currently on the market have PCE’s around 20% and prototypes are slowly reaching the 33% theoretical limit. The idea of building a hot carrier solar cell and exceeding 33% PCE is not new, but hot carrier extraction has proven to be exceedingly difficult.2 Upon photoexcitation, charge carriers with a wide range of energy values are produced and must match an energy selective acceptor to be extracted. One exotic method to harness hot carriers is by utilizing a highly efficient, dopant-mediated Auger de-excitation process to up-convert charge carriers to a precise energy value that can be efficiently extracted with an energy appropriate acceptor. In collaboration with the Gamelin group at the University of Washington, we have demonstrated a prototype device utilizing this concept (DOI: 10.1021/acs.jpclett.6b02219).
These proof-of-concept studies focused on injecting charge carriers and internal quantum efficiency values were below optimal. I am interested in developing new synthetic approaches to make “dual-doped” semiconductors where one dopant facilitates energy storage in a long-lived d-d transition and another alio-valent dopant increases the intrinsic carrier population and participates in the Auger de-excitation process. Quantum dot semiconductors are a useful platform to demonstrate this concept, where it is expected that these ideas can be expanded to a host of semiconducting materials.