News & Highlights

We are excited to release ASCICat, an open-source Python framework that transforms how we prioritize electrocatalyst candidates. Traditional volcano plots optimize for activity alone, while Pareto methods identify multiple "equally optimal" solutions without clear prioritization. ASCICat bridges this gap by combining catalytic activity (from Sabatier-optimal binding energies), electrochemical stability (via surface energies), and material cost into a single Activity-Stability-Cost Index (ASCI). Validated on 361 bimetallic HER catalysts and three CO₂RR pathways (CO, CHO, COCOH), ASCICat identifies earth-abundant candidates like Fe₂Sb₄ (ASCI = 0.899) and Cu₃Sb while providing built-in sensitivity analysis to quantify ranking robustness across 171 weight configurations. The package includes a Python API, command-line interface, and graphical user interface for accessible multi-objective catalyst screening.

We just released a new preprint where we combine symmetry-equivariant machine learning with DFT to rethink how we search for electrocatalysts. The EquiformerV2 model allowed us to scan more than 560,000 hydrogen adsorption sites across 439 bimetallic surfaces — an intractable space for pure DFT. Rather than focusing solely on activity, we also incorporated surface stability and material cost into the screening, allowing us to prioritize candidates that are not only efficient but also durable and scalable. This led to a shortlist of low-cost, scalable HER catalysts like Fe₂Sb₄, Cu₆Sb₂, and Cu₆Sn₂. Our automated pipeline bridges AI predictions with DFT validation, offering an adaptable framework for other electrocatalytic systems.

The development of advanced catalysts with innovative nanoarchitectures is critical for addressing energy and environmental challenges such as the electrochemical CO₂ reduction reaction (CO₂ RR). We present an innovative copper–sulfur planar structure, Cu–S–BDC, within a metal–organic framework (MOF) catalyst, which demonstrates 100% selectivity toward formate as the sole carbon product with a narrow bandgap of 1.203 eV and Faradaic efficiency of 92% at −0.4 V overpotential.

Lithium-sulfur (Li–S) batteries are promising for next-generation energy storage due to their high theoretical energy density (~2600 Wh kg⁻¹). We developed a novel sulfur host material that effectively suppresses the shuttle effect by anchoring lithium polysulfides through Li-P and S-P bonds, reducing dissolution into the electrolyte and enhancing sulfur reduction reaction kinetics.

We combined high-throughput DFT screening with few-shot machine learning to predict the hydrogen evolution activity of catalysts supported on 2D Ga₂CoS₄-x. We identified promising alternatives to traditional platinum catalysts and developed an intrinsic descriptor that links atomic properties to catalytic performance.

We report that a 2D MoSi₂N₄/Arsenene van der Waals heterostructure exhibits a type-II band alignment with an indirect band gap of 1.58 eV, achieving a spectroscopic limited maximum efficiency of 27.27%, marking a substantial improvement over conventional 2D monolayers.

The 2D Ge₂Se₂P₄ monolayer exhibits superior HER activity compared to most 2D materials and outperforms reference catalysts IrO₂(110) and Pt(111) in terms of low overpotential values for ORR and OER mechanisms, with very high solar-to-hydrogen efficiency.

We explored the design and implementation of an improved 2D Boron Nitride material as a sulfur electrode supplement to reduce polysulfide incompatibilities and kinetic latency of the cathode in Li–S cells, leveraging 2D BN's strong polysulfide attachment capability.

Through first-principle calculations and molecular dynamics simulations, we examined the effectiveness of graphyne, graphdiyne, and graphtriyne in protecting battery electrodes in solid polymer electrolyte batteries, addressing dendrite growth challenges.

In collaboration with experimental researchers from Austria and Japan, we successfully unraveled the enigma surrounding hydrogen embrittlement in high-strength aluminum alloys. We explored the addition of Zr-containing nanoparticles and discovered that depending on their characteristics, some particles can mitigate HE, some provide hydrogen to vulnerable sites, and some act as crack initiation sites.

Our first front cover publication in the Royal Society of Chemistry (RSC), Journal of Materials Chemistry A (IF: 14.511), promoting our critical overview on the advancement of materials design along with the charge storage mechanisms of organic battery electrodes from monovalent to multivalent alkali ions.

Using DFT, ab-initio molecular dynamics, and Basin-hopping Monte Carlo algorithm, we explored an alternative stable boron nitride structure with highly improved electrical conductivity, achieving theoretical capacity up to 8.7 times higher than commercialized graphite (3239.74 mAh/g vs 372 mAh/g).