Functional imaging of photovoltaic materials at the nanoscale

Abstract

The ideal photovoltaic technology for global deployment must exhibit two key attributes: (i) high power-conversion efficiency, enabling a solar panel with a large power output per area and, (ii) low-cost/W, due to either being derived from earth-abundant materials and/or ease of fabrication. For the past two decades, extensive efforts have been made to boost the efficiency of some of the most promising high performance and low-cost photovoltaic materials, such as CdTe, Cu(In,Ga)Se2 (CIGS), and hybrid organic-inorganic perovskites, to achieve higher efficiency devices. However, improvement in the overall performance is still limited by the open-circuit voltage (Voc). All of the solar cell materials listed above are composed of grains and grain boundaries on the order of micro- and nanometers, respectively, and their nanoscale interfaces can cause electrical charge carriers to become trapped and recombine non-radiatively, reducing the Voc. Therefore, in this thesis, I implement high spatial resolution functional imaging techniques to resolve the local voltage

variations in the thin-lm polycrystalline and hybrid perovskites materials for photovoltaic applications. First, I spectrally and spatially resolve the local photovoltage of CIGS solar cells through confocal optical microscopy to build a qualitative voltage tomography.

From these photovoltage results, I discover variations in the electrical response of >20% that are also on the same length scale as the grains composing the CIGS material. Therefore, by enhancing the spatial resolution beyond the diffraction limit, the electronic properties of individual grains and the interfaces between the grains can be fully resolved. For this, I implement Kelvin probe force microscopy (KPFM), and demonstrate a universal method to directly map the Voc of any photovoltaic material with nanoscale spatial resolution. Next, we extend this ability of KPFM to rapidly image (16 sec/map) the real-time dynamics of perovskite solar cells, which are notorious for their slow and unstable electrical output. Through fast-KPFM imaging, we discover regions within a single grain that show a residual Voc response which pervades for ~9 min, likely caused by a slow ion migration process. Finally, to understand how dierent perovskite compositions influence the behavior of the nanoscale electrical response, I utilize KPFM to realize both irreversible and reversible

Voc signals. Compiling all these results discussed above, throughout my Ph.D. I have yielded the following contributions: (i) evidence that the photovoltage of polycrystalline solar cell materials varies at the same length scale as the grains composing them, (ii) a nanoscale imaging platform to directly map the Voc with unprecedented spatial resolution, and (iii) a technique to map the real-time voltage response of many perovskite compositions, ultimately indicating that the elements constituting the perovskite cation and halide positions are both directly related to their reversible vs. irreversible electrical nature. From these contributions, I foresee the functional imaging methods developed in this thesis to be widely implemented as a diagnostic tool for the rational design of photovoltaics with enhanced electrical performance and lower cost.