1. Significant achievements have been made in preparing high-strength and tough coatings, researching substrate damage induced by coating cracking, and developing hydrogen embrittlement-resistant metallic materials. The theory of coating-induced cleavage cracking in ductile metallic substrates has been proposed, and a protective method against substrate damage caused by coating cracking has been established.

2. These advancements have significantly improved the fatigue resistance of coated metallic materials and resolved the technical challenge of brittle cracking in coated metallic structural components. Furthermore, a theoretical framework for deep hydrogen traps formed by misfit dislocations at semi-coherent interfaces has been developed, enabling the controllable preparation of high-density and uniformly distributed deep hydrogen traps. A series of high-strength and tough hydrogen-resistant steels, including heavy-haul railway wheel steel, spring steel, and marine equipment steel, have been successfully developed. The overall research technology has reached an internationally leading level.

Fig.1. (a) High-resolution transmission electron microscopy (HRTEM) analysis of NbC and the Fe matrix, indicating a K–S semi-coherent orientation relationship between NbC and α-Fe, with two sets of misfit dislocations, b(1) and b(2), present at the semi-coherent interface. (b) Thermal desorption spectroscopy (TDS) experiments combined with density functional theory (DFT) calculations reveal that misfit dislocations are the essence of hydrogen traps. (c) Comparison of deep hydrogen trapping capabilities of NbC with other defects. (d) Comparison of hydrogen embrittlement susceptibility. (e) HRTEM images and fast Fourier transform (FFT) analysis of B2-Cu, BCC-Cu, 9R-Cu, and FCC-Cu precipitates in the Fe matrix. (f) Three-dimensional atom probe tomography (3D-APT) analysis of the composition of 9R-Cu precipitates. (g) DFT calculations demonstrate that semi-coherent 9R-Cu precipitates exhibit stronger hydrogen trapping capabilities. (h) Comparison of hydrogen embrittlement susceptibility, showing that martensitic stainless steel containing 9R-Cu precipitates exhibits the best resistance to hydrogen embrittlement.

Fig.2. Fundamental characterization of fatigue failure mechanisms: (a) Cross-sectional morphology of the alloy matrix interface after fatigue testing at 600 MPa, (b) S-N curve, fatigue fracture morphology, and fatigue crack initiation diagram, (c) XRD pattern of the material and HRTEM image of the interface with <200> orientation, (d) Microstructural image of the coating after fatigue testing at 350 MPa, highlighting the diffusion barrier layer.

3. A material design strategy combining machine learning and high-throughput experiments is proposed to rapidly screen ultrahard high-entropy ceramic coatings. The composition and process parameters of the coating are used to predict the hardness. The coating hardness predicted by this method is 9% higher than in the quinary system from the original training dataset. These coatings were successfully synthesized through high-throughput experiments. Machine learning combined with high-throughput experimental methods can effectively accelerate the composition design of multi-component materials.

(a) Schematic design of a high-entropy ceramic coating with high hardness, (b) Schematic diagram of magnetron co-sputtering process,(c) Global interpretation using SHAP,(d) Elemental components of high-entropy nitride coatings, including strong nitride formers, weak nitride formers, and p-elements that form a covalent bond with N