A Machine Learning Approach for Metal Oxide Based Polymer Composites as Charge Selective Layers in Perovskite Solar Cells

Abstract

A library of metal oxide-conjugated polymer composites was synthesized, encompassing of WO3-polyaniline (PANI), WO3poly(N-methylaniline) (PMANI), WO3-poly(2-fluoroaniline) (PFANI), WO3-polythiophene (PTh), WO3-polyfuran (PFu) and WO3-poly(3,4ethylenedioxythiophene) (PEDOT). These composites were probed as hole selective layers for perovskite solar cells (PSCs) fabrication. We adopted machine learning approaches to predict and compare PSCs performances with the developed WO3 and its composites. The experimental and theoretical results are coherent, when the electrooptical properties of PSC were computed. Notably, for the evaluation of PSCs performance, decision tree model is the ideal for WO3-PEDOT composite, while random forest model was found to be suitable for WO3-PMANI, WO3-PFANI, WO3-PFu. While in the case for WO3, WO3-PANI and WO3-PTh, K Nearest Neighbors model was appropriate. Machine learning models can be a pioneering prediction models for the PSCs performance and its validation.

Molecular Interface Engineering via Triazatruxene‐Based Moieties/NiOx as Hole‐Selective Bilayers in Perovskite Solar Cells for Reliability

“…Triple‐cation‐perovskite [Cs 0.1 (FA 0.9 MA 0.1 ) 0.9 Pb(I 0.9 Br 0.1 ) 3 ] layer was achieved according to the previous report. [ 51 ] The PCBM solution of 15 mg mL −1 in chlorobenzene was spin coated at room temperature on perovskite film at 1000 rpm (500 rpm s −1 ) for 20 s and vacuum dried for 10 min. A thin layer of BCP was deposited atop PCBM by spin coating at 5000 rpm for 40 s followed by evaporating Ag (90 nm, <1 Å s −1 ) in a thermal evaporator under low‐vacuum conditions (10 −7 Torr).…”

Automated Machine Learning Approach in Material Discovery of Hole Selective Layers for Perovskite Solar Cells

“…Here, n‐ and p‐type represent ETM and HTM, respectively, while “i” denotes the light‐harvesting layer. [ 5 ] The subtle difference between these two PSC structures is the carrier extraction direction; [ 6 ] in a typical (n–i–p) PSC, electrons are collected through the transparent conductive oxide electrode while the holes are extracted through the top metal electrode. Conversely, in the p–i–n type PSCs, holes are extracted through the bottom transparent conductive oxide electrode while electrons are extracted through the top metal electrode.…”

Predicting Perovskite Bandgap and Solar Cell Performance with Machine Learning

“…[1] Hybrid perovskite is widely being used in a variety of applications, including solar cells, light-emitting diodes, lasers, and photodetectors, due to its longer electron and hole diffusion lengths and higher carrier mobility and a wide tunable bandgap (E g ). [1] The perovskite structure is represented by ABX 3 , [2,3] where A is an organic or inorganic cation, B is metal, and X is a halogen anion (Figure 1a). The rapid astonishing advances in the power conversion efficiency (PCE) of perovskite solar cells (PSCs) led to a rise in the PCE from 3.8% to 25.5% and currently occupy the center stage for photovoltaic research and development.…”

“…[9,10] ML is a data-driven approach that combines with experimental datasets to predict concealed information and trends. [2] ML approaches are established as influential for properties and performance prediction of materials, which in turn can expedite the material exploration by minimizing the tasks for proposing potentially promising structures. [11] The accuracy of the ML approach is comparable with the DFT calculations and arguably viable to design the materials even from small datasets, while the small and uneven distributions are insufficient for DFT calculations.…”