Machine-Learning-Assisted Synthesis of Polar Racemates

Abstract

Racemates have recently received attention as nonlinear optical and piezoelectric materials. Here, a machine-learningassisted composition space approach was applied to synthesize the missing M = Ti, Zr members of the Δ,Λ-[Cu(bpy) 2 (H 2 O)] 2 [MF 6 ] 2 • 3H 2 O (M = Ti, Zr, Hf; bpy = 2,2′-bipyridine) family (space group: Pna2 1 ). In each (CuO, MO 2 )/bpy/HF(aq) (M = Ti, Zr, Hf) system, the polar noncentrosymmetric racemate (M-NCS) forms in competition with a centrosymmetric one-dimensional chain compound (M-CS) based on alternating Cu(bpy)(H 2 O) 2 2+ and MF 6 2− basic building units (space groups: Ti-CS (Pnma), Zr-CS (P1̅ ), Hf-CS (P2/n)). Machine learning models were trained on reaction parameters to gain unbiased insight into the underlying statistical trends in each composition space. A human-interpretable decision tree shows that phase selection is driven primarily by the bpy:CuO molar ratio for reactions containing Zr or Hf, and predicts that formation of the Ti-NCS compound requires that the amount of HF present be decreased to raise the pH, which we verified experimentally. Predictive leave-one-metal-out (LOO) models further confirm that behavior in the Ti system is distinct from that of the Zr and Hf systems. The chemical origin of this distinction was probed via fluorine K-edge X-ray absorption spectroscopy. Pre-edge features in the F1s X-ray absorption spectra reveal the strong ligand-to-metal π bonding between Ti(3d − t 2g ) and F(2p) states that distinguishes the TiF 6 2− anion from the ZrF 6 2− and HfF 6 2− anions.

Machine Learning Guided Design of Single–Phase Hybrid Lead Halide White Phosphors

“…Thus, this result strongly suggests that the sample is homogeneous and that the halogen atoms are randomly dispersed within the crystal structure. The same conclusion can be drawn from the analysis of the 13 C and 15 N MAS NMR spectra (Figure S2, Supporting Information). The 207 Pb NMR line width reflects the distribution of angles and lengths of the Pb─X bonds (X = Br or Cl).…”

“…Machine‐learning (ML) models have been shown to be efficient solutions for such optimization problems. Although ML models have recently been used to predict the optical and piezoelectric properties of materials, [ 14 , 15 , 16 ] the reported strategies do not typically incorporate the traditional approach of discovering new materials. [ 17 , 18 ] In this article, we proposed a novel strategy in which ML models substitute the human scientist decisions in step (iii) of the traditional iterative approach (Figure 1 ).…”

Machines for Materials and Materials for Machines: Metal-Insulator Transitions and Artificial Intelligence

“…NLP can then be used to prescribe techniques for the synthesis of new materials [84,85]. BO and other machine learning tools may help optimise the process [86,87].…”

Predicting the Redox Potentials of Phenazine Derivatives using DFT Assisted Machine Learning

“…However, some reports also explore simple features based on molecular structure [37][38][39][40][41][42] .…”