Data Science Applied to Sustainability Analysis by Unknown

Author:Unknown
Language: eng
Format: epub
Publisher: Elsevier
Published: 2021-07-15T00:00:00+00:00

Fig. 7.16 Workflow used to generate the joint experimental and computational database of optical absorption properties of molecules by Cole et al. Reproduced with permission from ref. (Beard et al., 2019). Copyright Springer Nature 2019.

High-throughput computational search for hybrid halide semiconductors

As discussed in Section 2.4, hybrid organic-inorganic perovskite solar cells based on lead halides have notably high efficiencies. However, their lead constituent poses a toxicity issue. Moreover, they are liable to decompose quickly, and they can be a potential environmental hazard under wet conditions. Many alternatives to methylammonium lead halide (MAPbX3) have been proposed and experimentally explored, such as MASnI3, (Kaltzoglou et al., 2015; Gu et al., 2018) Cs2InBiCl6 and Cs2InSbCl6. (Xiao et al., 2017) The latter two materials are of the “double perovskite” class, of the form A2BB'X6, where B and B' are ions with 1+ and 3+ charges, respectively (whilst B has a 2+ charge in a regular ABX3 perovskite). Many of the alternative perovskites explored have favourable electronic properties, but also still suffer from a tendency to undergo decomposition or to separate into multiple lower-energy phases. In principle, organic-inorganic metal halide structures other than a perovskite with greater stability may exist with equally good electronic properties. This inspired Yang and coworkers to design a high-throughput computational study to evaluate such structures and their suitability for use in solar cells. (Yang and Li, 2019) Firstly, they identified 24 unique prototype crystal classes of the form CsxByXz in the AFLOWLIB and Materials Project databases. (Jain et al., 2013; Curtarolo et al., 2012) The Cs (cesium) was chosen as a placeholder whilst searching the databases owing to the large size of its ion: its sites in the crystal could therefore be exchanged subsequently to hold small organic polyatomic cations instead. Given this set of 24 unique crystal classes, they combinatorially generated crystal structures by substituting 5 possible alternative options for Cs, 65 possible options for B and 3 possible options for X. The 5 organic cations chosen to potentially replace Cs were methylammonium (MA), formamidinium (FA), aziridinium (AD), dimethylammonium (DMA) and ethylammonium (EA), shown in Fig. 7.17. The three options for X were the three common halide ions Cl-, Br- and I-. For the remaining B positions, metal ions with 1+, 2 + 3+ and 4+ charges were allowed, which determined overall ratios of the stoichiometric compositions (subscripts x, y and z) in order to meet the requirement that the crystal be neutral overall. A total of 4,507 distinct chemical compounds were created in this way, which were then computationally screened for stability and desirable electronic behaviour. (Yang and Li, 2019) Since both photovoltaic devices and light emitting devices rely on such properties being finely tuned, the group investigated suitability of generated compounds for both of these applications. However, we focus here on the chosen criteria for photovoltaics.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.