[Paper Review] AI-enhanced discovery and accelerated synthesis of metal phosphosulfides

Javier Sanz Rodrigo, Nicholas A. Kryger-Nelson+4 more|arXiv (Cornell University)|Jan 23, 2026

Inorganic Chemistry and Materials0 citations

TL;DR

The paper combines DFT, multi-fidelity ML, and high-throughput synthesis to explore 909 ternary metal phosphosulfides, discovering 19 new stable compounds and demonstrating rapid synthesis of four phosphosulfide systems in four combinatorial experiments.

ABSTRACT

Metal phosphosulfides have emerged as unique multifunctional materials, but they present unique synthesis challenges compared to more established material classes such as oxides and nitrides. As a consequence, experimental development and theoretical understanding of phosphosulfides have focused on individual compounds rather than on accelerated broad-range exploration. In this work, we first evaluate the synthesizability and band gaps of 909 hypothetical ternary phosphosulfides by density functional theory. We find 19 previously unknown thermodynamically stable compounds, including the first Si- and Ge-based phosphosulfides. For rapid band gap prediction, we then develop a multi-fidelity machine learning model to translate semilocal density functional theory band gaps into experimentally calibrated band gaps. Importantly, we extend the accelerated material development workflow to the experimental domain by demonstrating a route to high-throughput synthesis and characterization of virtually any phosphosulfide material system. The method is based on thin-film combinatorial libraries and yields over 100 unique compositions in each experiment, enabling us to synthesize four distinct phosphosulfide compounds in only four combinatorial experiments without prior synthesis recipes and without compromising on material quality. Thus, we argue that accelerated materials development workflows combining theory, artificial intelligence, synthesis, and characterization can be viable even for experimentally challenging inorganic materials.

Motivation & Objective

• Evaluate synthesizability and band gaps of 909 hypothetical ternary phosphosulfides using theory.
• Identify thermodynamically stable phosphosulfide compounds, including previously unknown ones.
• Extend accelerated materials development to experimental synthesis for phosphosulfides.

Proposed method

• Compute thermodynamic stability and band gaps for 909 ternary phosphosulfides with density functional theory (DFT).
• Develop a multi-fidelity machine learning model to predict experimentally calibrated band gaps from semilocal DFT band gaps.
• Construct and test high-throughput experimental routes using thin-film combinatorial libraries for rapid exploration of composition space.
• Demonstrate synthesis of four phosphosulfide compounds using four combinatorial experiments without pre-existing synthesis recipes.

Figure 1: Summary of the structural and compositional diversity of phosphosulfides. The structures are visualized with VESTA [ Momma2011 ] . Assignment of oxidation states is discussed in the SI. The ”number of materials” (and the corresponding ”% of total”) refers to the number of phosphosulfides c

Experimental results

Research questions

• RQ1 How do synthesizability and band gaps vary with composition and structure across 909 phosphosulfides?
• RQ2 Can high-throughput synthesis workflows enable rapid discovery of phosphosulfide materials despite volatile and toxic precursors?
• RQ3 What new thermodynamically stable phosphosulfide compounds can be predicted beyond known literature?
• RQ4 Can theory-guided, AI-augmented workflows be translated effectively into experimental phosphosulfide synthesis?

Key findings

• 19 previously unknown thermodynamically stable phosphosulfides identified by DFT across 909 candidates.
• First Si- and Ge-based phosphosulfides predicted among stable compounds.
• A multi-fidelity ML model translates semilocal DFT band gaps into experimentally calibrated gaps.
• Four distinct phosphosulfide compounds synthesized in four combinatorial experiments with high crystalline quality.
• High-throughput thin-film libraries enable exploration of many compositions in parallel, accelerating discovery.

Figure 2: Oxidation state of phosphorus as a function of the atomic P/S ratio in the ternary phosphosulfides calculated in this study (only lowest-energy polymorphs within stability tolerance). The P/S ratios are discrete and are linked to the generalized compositions shown on the top x axis. The da