SinoGreenTech Academic Portal
ZL
Verified CAS / Academic Author2 Decoded Studies

Prof. Zihan Li

School of Chemistry and Chemical Engineering, Harbin Institute of Technology

Research Publications & English Decoded Briefs

Showing 2 publications
SCIENCE CHINA Materials2026DOI: 10.1007/s40843-025-3633-5

Phase Penetration: Key Drivers in Barrier Layer Failure of Hf-free Half-Heusler Thermoelectric Modules

High-temperature interfacial diffusion in Half-Heusler (HH) thermoelectric devices poses significant challenges for practical applications, particularly the diffusion of Ag from conventional solders, which degrades material performance and device stability. This study reveals anomalous Ag diffusion through a Cr powder barrier layer into Ti0.5Zr0.5NiSn0.98Sb0.02, driven by Sn phase penetration. In contrast, employing a Cr foil barrier layer pre-densified the material, effectively preventing Sn phase penetration and eliminating Ag diffusion pathways, thereby preserving junction integrity. After aging at 973 K for 30 days, the Cr foil junction maintained a clean interface with a low contact resistivity of 0.27 μΩ cm2. Benefiting from this interfacial design, a Hf-free HH module achieved a high conversion efficiency of 10.4% at a hot-side temperature of 976 K, alongside long-term stability. This work addresses critical bottlenecks in developing high-performance, low-cost HH modules, facilitating their commercial application in waste heat recovery.

SCIENCE CHINA Materials2026DOI: 10.1007/s40843-025-4104-4

COFAP: A Universal Framework for COFs Adsorption Prediction through Designed Multi-Modal Extraction and Cross-Modal Synergy

Covalent organic frameworks (COFs) are promising adsorbents for gas adsorption and separation, yet identifying optimal structures among their vast design space requires efficient high-throughput screening. Conventional machine-learning predictors rely heavily on specific gas-related features, which are time-consuming and limit scalability, leading to inefficiency and labor-intensive processes. Here, we propose COFAP, a universal COFs adsorption prediction framework that extracts multi-modal structural and chemical features via deep learning and fuses these complementary features through a cross-modal attention mechanism. Without relying on explicit gas-specific thermodynamic descriptors, COFAP achieves state-of-the-art prediction performance on the hypoCOFs dataset under the conditions investigated, outperforming existing approaches. Based on COFAP, we found that high-performing COFs for gas separation concentrate within a narrow range of pore size and surface area. A weight-adjustable prioritization scheme is also developed to enable flexible, application-specific ranking of candidate COFs. Superior efficiency and accuracy render COFAP directly deployable in crystalline porous materials.