Perovskite Automated Spin Coat Assembly Line (PASCAL)

Motivation

PASCAL is an automated platform for spin coating and annealing thin films onto small (>2x2 cm) substrates, aimed af increasing experimental throughput in the pursuit of designing better perovskite solar cells. Perovskite solar cells are infinitely tunable, as they can be formed with combinations of nearly half of the periodic table, presenting the excitement of limitless possibilities and the curse of dimensionality. With PASCAL, we aim to increase our experimental search rate by orders of magnitude, enabling a more systematic and exhaustive approach toward exploring the vast compositional space of interest for solar cell design.

Hardware

CAD schematics of PASCAL

Liquid handler preparing precursor solution mixtures for a compositional search

Liquid handler + spincoater working in conjunction to deposit solution onto substrate.

Automated Full Sample Fabrication + Characterization Timelapse

pascal_timelapse.mp4

Experimental Planning

• Computed material properties -> manifold embedding / dimensionality reduction into "behavior space"

Hardware Scheduling

Active Learning

intra-run (narrow scope)

• bayesian approach to evaluating sample space

• update compositions of interest using in-line during experiment

inter-run (wide scope)

• bayesian approach with operator input (XRD, UV-Vis, etc)
• find covariances between composition/fabrication conditions
• point towards broader design philosophies in perovskite solar cells

In-Line Characterization

• Widefield PL camera

• RGB Imaging for color determination

Dependency

• python=3.7.9=h26836e1_0

Addons

See our fenningresearchgroup.com/https://github.com/fenning-research-group/PASCAL_TrainView https://github.com/fenning-research-group/PASCAL_TrainView

and https://github.com/dnzckn/BO-PV repo's for additional capabilities