Session:

This research proposes a novel approach to streamline the choice of precursors for the sol-gel synthesis of BiFeO3 (BFO) thin films [1], leveraging text-mining and chemical reaction networks. An initial dataset of 340 recipes from 178 scientific papers was manually compiled, providing insight into typical experiment steps. However, the reasonings behind material choices and their interplay remain largely obscure [2]. To clarify these aspects, the study categorizes precursor materials by roles (e.g., metal source, solvent, chelating agent) and analyzes their interactions and influence on the synthesis process. A major challenge is the gelation step, influenced by material selection, which can introduce impurities in the final film [2]. We also highlight the crucial roles of solvent in forming a dense, stable precursor molecule and the chelating agent in enhancing the film's homogeneity. We propose to integrate chemical reaction networks [3] representing all plausible reactions within the materials system, allowing for a more profound understanding of the processes in the precursor. In our study, we employed an integrated automated computational pipeline including a semi-empirical xTB package [4] for high-throughput calculations. This setup facilitated ab-initio calculations revealing that Bi prefers a coordination number of 8 and stabilizes when bidentately coordinated with solvent ligands. To generate a comprehensive chemical reaction network, we utilized Molecular Graphs to represent molecules, creating an exhaustive set of intermediate species for each reaction step. For the reaction between Bi nitrite and 2-methoxyethanol solvent, we identified 634 unique molecular graphs, and fed their 3D representations into the chemical reaction network pipeline. Recovering reactions with chemical reaction networks for various systems (e.g., metal source + solvent, metal source + solvent + chelating agent) is an effective strategy to track final film formation and study the role of chelating agent. This method enhances our understanding of sol-gel precursor processes and promises to facilitate the design and optimization of synthesis methods for high-quality BFO thin films.
References:
[1] JBNJ Wang, JB Neaton, H Zheng, V Nagarajan, SB Ogale, B Liu, D Viehland,
V Vaithyanathan, DG Schlom, UV Waghmare, et al. Epitaxial bifeo3 multiferroic thin film
heterostructures. science, 299(5613):1719–1722, 2003.
[2] Qi Zhang, Nagarajan Valanoor, and Owen Standard. Epitaxial (001) bifeo 3 thin-films with
excellent ferroelectric properties by chemical solution deposition-the role of gelation. Journal
of Materials Chemistry C, 3(3):582–595, 2015.
[3] Daniel Barter, Evan Walter Clark Spotte-Smith, Nikita S Redkar, Aniruddh Khanwale, Shyam
Dwaraknath, Kristin A Persson, and Samuel M Blau. Predictive stochastic analysis of massive
filter-based electrochemical reaction networks. Digital Discovery, 2023.
[4] Christoph Bannwarth, Sebastian Ehlert, and Stefan Grimme. Gfn2-xtb—an accurate and
broadly parametrized self-consistent tight-binding quantum chemical method with multipole
electrostatics and density-dependent dispersion contributions. Journal of chemical theory and
computation, 15(3):1652–1671, 2019.

Co-Author(s): Kevin Cruse (University of California, Berkeley), Carolin Sutter-Fella (Lawrence Berkeley National Laboratory), Anubhav Jain (Lawrence Berkeley National Laboratory), Samuel Blau (Lawrence Berkeley National Laboratory)

Viktoriia Baibakova

Appearances