Advanced simulations of solar cell devices

The course aims to provide a brief overview of different solar cell technologies, with a focus on direct and indirect perovskite-based solar cells (PSCs).
The aim of the course is to acquaint students with state-of-the-art quantum and classical simulation techniques applied to PSCs, with the objective of performing high-level simulations to unravel the physical-chemical mechanisms underlying the possible improved performance and stability of PSC devices. The course will also provide a solid introduction to how to develop reliable classical force fields for complex multi-component systems. The topic of interatomic potentials with machine learning (ML) will also be addressed.  
An introduction to the main experimental techniques for characterizing PSCs will also be provided, to understand which quantities are most important and how to compare theoretical results with experimental ones.

 

The course is divided into three modules, starting with a brief introduction to the physical, electronic and optical properties of solar cell devices.

 

1. PSC characterization: introduction to the main experimental characterization techniques (I-V curve, PLQY, PL spectra, TRPL, XRD, NMR). Emphasis will be placed on how to use this information to set up meaningful simulation protocols.

 

2. Simulation of PSC devices: students will become familiar with different quantum (CP2K, Quantum Espresso, Gaussian) and classical (LAMMPS, AMBER) simulation software applied to perovskites, different hole transport layers (HTL) and electron transport layers (ETL), also in the case of interaction with passivating molecules. The modeling of perovskite/ETL(HTL) interfaces in the presence/absence of additives will also be addressed. Particular attention will be paid to the simulation of optoelectronic properties (band gaps, band alignment, exciton binding energy, etc.) in the presence/absence of additives. 

 

3. Force field development: in most cases, the size of the system and/or the physical-chemical mechanisms investigated require a number of atoms and/or timescales that can only be achieved with classical MD. In this case, it is essential to have accurate force fields based on ab initio simulations to be sure of correctly capturing all interactions of the system. Students will learn the basic concepts of force field development (including machine learning interatomic potentials) from ab initio data. They will also learn how to parameterize force fields in the presence of composite systems, such as additives and perovskites.