Metal electrode deposition is universally adopted in the community for optoelectronic device fabrication, inducing hybridization at electrode interfaces, and allows efficient extraction or injection of photocarriers. However, hybridization-induced midgap states increase photocarrier recombination pathways, creating a paradoxical trade-off. Here, we discovered that efficient photocarrier extraction and a long photocarrier lifetime can be achieved simultaneously in MoS 2 /van der Waals Au contact, minimizing photocarrier loss at the interface. Using transient reflection spectroscopy, we quantified an ultrafast hole extraction time of 310 femtoseconds and a photocarrier lifetime of 259 picoseconds. Suppressed exciton formation in diffusion measurement validated ultrafast hole extraction. We further revealed a short photocarrier lifetime of ~8.7 picoseconds in deposited contact and negligible hole extraction in unintimate contact with space gap at the interface. We unraveled the hot electron transfer process from Au to MoS 2 in different contact interfaces. Our results on different metal contact interfaces provide valuable insights into electrode engineering for future optoelectronic devices.
Multi-modal drug discovery frameworks increasingly rely on robust encoders capable of representing chemical structures alongside other data modalities. In this study, we fine-tuned a DeBERTa-based SMILES encoder to improve its predictive capacity for 22 ADMET (Absorption, Distribution, Metabolism, Excretion, and Toxicity) endpoints while maintaining strong comprehension of molecular syntax. Starting from a pretrained ZINC-based DeBERTa checkpoint, we trained on a 300K PubChem-ADMET dataset using a multi-label regression scheme with a focal MAE loss to stabilize learning across diverse properties. Our encoder achieved top-10 rankings on 16 TDC benchmark tasks, including notable improvements of 14–30% on critical endpoints such as bioavailability and CYP2C9-substrate. Compared with BERT- and RoBERTa-based molecular encoders, our approach preserved significantly higher MLM accuracy (>89%) over training, indicating robust retention of chemical language understanding. Additional analysis using an ADMET path length metric revealed that DeBERTa produced more disentangled latent representations, underscoring its suitability for property-specific molecular manipulation. These results demonstrate that a disentangled, ADMET-aware DeBERTa encoder can serve as a powerful component for future multi-modal pipelines in AI-driven drug design, effectively balancing structural fluency with predictive specialization.
14 May 2026 in Bulletin of the Korean Chemical Society