Noble nanoparticle(NP)-sized electrocatalysts have been exploited for diverse electrochemical reactions, in particular, for eco-friendly hydrogen economy such as water splitting. Recently, minimal amounts of single atoms (SAs) are exploited to...

Machine learning (ML) methods would significantly reduce the computational burden of catalysts screening for nitrogen reduction reaction (NRR).

The catalytic activity and selectivity can be improved for nitrogen fixation by using hollow sites and vacancy defects in 2D materials, while a new machine learning descriptor accelerates screening of efficient electrocatalysts.

Carbon-based transition metal (TM) single-atom catalysts (SACs) have shown a great potential toward electrochemical water splitting and H2 production. Given that two-dimensional (2D) materials are widely exploited for sustainable energy...

MXenes have been widely used as substrates of hybrid electrocatalysts for water splitting due to their stability and metallic properties. However, tuning MXenes towards superb hydrogen/oxygen evolution reaction (HER/OER) activity has remained elusive. Using first‐principles calculations along with machine learning (ML) based descriptors, it is shown that late transition metal doping is able to significantly promote HER/OER activities. Both single‐atom adsorption onto a stable hollow site above the outer oxygen layer single‐atom catalyst 1 (SAC1), and single‐atom replacement at a sub‐surface metal layer (SAC2) are considered. An adsorbate evolving mechanism (AEM) is preferred for SAC1, while the increased M‐O bond covalency for SAC2 makes lattice oxygen mechanism (LOM) favored. It is found that a single Ni or Co atom embedded into MXenes provides a suitable number of electrons for optimal AEM and raises the O 2p band towards activated LOM. The stability and superb bifunctional catalytic capability of MXene combinations (Ni‐doped Sc3N2O2 and Ni‐doped Nb3C2O2) towards both HER and OER are demonstrated. The electronic and geometric descriptors used in the ML analysis work universally for classification of high‐performing HER/OER catalysts. This work provides a rational strategy for promoting bifunctional electrocatalytic activities based on low‐cost MXenes metals.