Hunting for Organic Molecules with Artificial Intelligence: Molecules Optimized for Desired Excitation Energies

Abstract

This work presents a proof-of-concept study in artificial-intelligence-assisted (AI-assisted) chemistry where a machine-learning-based molecule generator is coupled with density functional theory (DFT) calculations, synthesis, and measurement. Although deep-learning-based molecule generators have shown promise, it is unclear to what extent they can be useful in real-world materials development. To assess the reliability of AI-assisted chemistry, we prepared a platform using a molecule generator and a DFT simulator, and attempted to generate novel photofunctional molecules whose lowest excited states lie at desired energetic levels. A 10 day run on the 12-core server discovered 86 potential photofunctional molecules around target lowest excitation levels, designated as 200, 300, 400, 500, and 600 nm. Among the molecules discovered, six were synthesized, and five were confirmed to reproduce DFT predictions in ultraviolet visible absorption measurements. This result shows the potential of AI-assisted chemistry to discover ready-to-synthesize novel molecules with modest computational resources.

Li-Ion Conductive Li3PO4-Li3BO3-Li2SO4 Mixture: Prevision through Density Functional Molecular Dynamics and Machine Learning

“…In particular, many new materials have recently been predicted through the effective combination of DFT calculations and machine learning. [14][15][16][17][18] However, these examples are limited in that the physical values are obtained through static calculations. If we focus on dynamic values such as Li-ion conductivity, molecular dynamic calculations are basically required.…”

Data integration for accelerated materials design via preference learning

“…In figure 1, We created our own small database of 94 organic molecules with their HOMO-LUMO gaps and absorption wavelength computed via TD-DFT (supplementary table 1). See [4] for computational details. To test the performance statistically, 50 candidate sets are created, each of which has 40 randomly selected molecules.…”

“…A substantial amount of materials data are accumulated in public databases [1][2][3], and machine-learning-based design of materials is increasingly common in recent years [4][5][6]. The problem of materials design is mathematically formulated as a black-box optimization problem, where a large number of candidates are available and the goal is to find the candidate with best target property via a minimum number of observations.…”

Designing metamaterials with quantum annealing and factorization machines

“…Automated materials discovery based on black-box optimization is an iterative process to select one candidate from a massive number of candidates (i.e., design space) for experimental investigations [10][11][12]. From existing materials properties data, machine learning predicts the properties of unobserved candidates and defines an acquisition function in design space.…”

“…The statistical barrier is reduced by using a fast simulator, which calculates the properties of the target materials, instead of an experiment. As several recent studies show, a simulation is not as overwhelming as the computational barrier [10,12]. To circumvent the difficulty with global optimization, exhaustive searches have been replaced by methods such as local searches, tree searches [13], and genetic algorithms [14].…”