STShih-Wen Tseng

Shih-Wen Tseng

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Shih-Wen Tseng has a total of 32 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Surface-Treated InP Quantum Dots for Selective CO2 Photoreduction

    by Le-Chun Li, Chia-Ying Wu, Tsukasa Takanashi, Shih-Wen Tseng, Tetsu Yonezawa, Akira Yamakata, Chen-Hao Yeh, Ying-Chih Pu

    Photocatalytic CO2 reduction over semiconductor quantum dots (QDs) offers a promising strategy for solar fuel production, yet the influence of surface atomic structure on reaction pathways and product selectivity remains poorly understood. Herein, surface-treated indium phosphide (InP) QDs with tetrapod and tetrahedral morphologies were synthesized through a temperature-controlled hot-injection method followed by hydrofluoric acid post-treatment. Removal of the native surface oxide layer effectively suppresses charge-carrier trapping and prolongs carrier lifetimes, resulting in significantly enhanced photocatalytic CO2 reduction activity. More importantly, the product selectivity is governed by the exposed crystal facets: (110)-oriented tetrapod preferentially produces CO through a two-electron pathway, whereas (111)-oriented tetrahedron promotes further hydrogenation toward CH4. This work establishes a direct structure–reactivity relationship between surface atomic arrangement and CO2 photoreduction selectivity, providing a design principle for environmentally benign QD photocatalysts for solar fuel production.

  • Open Access

    Redox-Tuned Ce-MOF Phosphides as Efficient Electrocatalysts for Anion Exchange Membrane Water Electrolysis

    by Yin-An Yang, Zhi-Rou Liang, Shih-Wen Tseng, Wei-Chung Chen, Cheng-Han Lee, I-Wen Peter Chen

    Designing efficient and durable platinum-group-metal-free electrocatalysts is crucial for advancing anion-exchange membrane water electrolysis (AEMWE) toward practical green hydrogen production. Herein, we report a binder-free cerium-based metal−organic framework (Ce-MOF) electrode that is transformed into a porous phosphide-derived electrocatalyst through a simple phosphidation strategy while preserving the parent MOF architecture on nickel foam. The robust Ce−O coordination network formed between Ce ions and benzene-1,3,5-tricarboxylic acid (BTC) effectively maintains the structural integrity and hierarchical porosity of the framework during phosphidation, thereby retaining abundant accessible active sites. Meanwhile, the reversible Ce3+/Ce4+ redox couple acts as an intrinsic electron reservoir to promote charge transfer, whereas phosphorus incorporation simultaneously enhances electrical conductivity and tailors the local electronic structure, resulting in synergistically improved electrocatalytic activity. Consequently, the optimized Ce-BTC-P300 electrode exhibits outstanding bifunctional performance, requiring overpotentials of only 340 mV for the oxygen evolution reaction (OER) and 231 mV for the hydrogen evolution reaction (HER) to achieve an industrially relevant current density of 100 mA cm−1 in 1 M KOH. The electrode also demonstrates excellent durability, retaining more than 98% of its initial activity after 80 h for the OER and 100 h for the HER. When employed as both the anode and cathode in an AEMWE device, the catalyst delivers a high current density of 1 A cm−2 at a low cell voltage of 2.04 V under ambient conditions while maintaining stable long-term operation. This work establishes a rare-earth MOF-derived phosphidation strategy that integrates intrinsic redox buffering, robust structural stability, and enhanced electronic conductivity into a single catalyst platform, providing new insights into the rational design of high-performance, platinum-group-metal-free bifunctional electrocatalysts for practical alkaline water electrolysis.

  • Open Access

    AI-Assisted Materials Characterization: A Curriculum-Guided Multimodal Framework for Transmission Electron Microscopy

    by Chong-Ren Tu, Hung-Wei Hsueh, Wei-Che Chang, You-Ting Lin, Yu-Hsuan Cheng, Wen-Huei Chu, Shih-Wen Tseng, Shu-Han Hsu

    Scientific imaging techniques such as transmission electron microscopy (TEM) offer unparalleled nanoscale insights, but knowledge extraction remains severely limited by the scarcity of structured datasets and the extensive expertise required for interpretation. To overcome this bottleneck, we present an automation-driven framework for multimodal scientific image understanding, demonstrated in TEM as a representative case. Our pipeline integrates automated literature mining, figure segmentation, and GPT-based knowledge distillation to generate 216,448 question–answer pairs across 40,000 TEM images, thereby establishing an end-to-end data collection framework for constructing a large-scale visual question answering (VQA) dataset in this domain. This automated process achieves >1000× acceleration in data curation. To effectively train on this complex data, we introduce a difficulty-aware stratification strategy that organizes samples by intrinsic cognitive complexity rather than task type, mimicking the pedagogical progression of a human expert. The resulting model, TEM-LLM, achieves a 119% relative gain in human-aligned GPT scores and high accuracy in microstructural feature detection. By bridging visual perception and linguistic reasoning, this methodology establishes a scalable foundation for AI-assisted characterization, and accelerated materials discovery.

  • Open Access

    Universal Interfacial Engineering via Amorphous Inorganic Binders: Passivating Surface States and Accelerating Hole Transfer across Metal Oxide Photoanodes in Photoelectrochemical Water Oxidation

    by Po-Keng Hsiao, Yen-Lun Kung, Yun-Pei Liu, Shih-Wen Tseng, Tetsu Yonezawa, Chun-Hu Chen, Ying-Chih Pu

    Interfacial engineering of photoanodes is essential for efficient solar water splitting; however, many existing strategies rely on complex, multistep processes or hydrothermal growth, limiting scalability and reproducibility. Here, we present a universal interfacial engineering platform based on a sub-2 min acidic redox-assisted deposition (ARD) of an amorphous inorganic binder, cobalt-manganese oxyhydroxide (CMOH). Using two-step-fabricated BiVO4 as a literature-standard benchmark, we demonstrate that an ultrathin CMOH overlayer markedly enhances photoelectrochemical (PEC) water oxidation performance. The conformal CMOH coating (∼3.5 nm) is deposited under ambient conditions via a simple dip-coating process and is broadly applicable to metal oxide photoanodes, including BiVO4, WO3, and ZnO. The CMOH-modified BiVO4 photoanode delivers a high photocurrent density of ∼6.0 mA cm-2 and a hole-transfer efficiency of ∼82% at 1.23 VRHE, together with a remarkable applied bias photon-to-current efficiency (ABPE) of 1.68% at 0.66 VRHE. These enhancements originate from effective surface-state passivation by the CMOH overlayer, leading to over a 4-fold reduction in the surface recombination rate constant, as revealed by intensity-modulated photocurrent spectroscopy (IMPS). In addition, the reduced Tafel slope (78.5 mV dec-1) suggests that the CMOH layer promotes interfacial charge transfer and enhances the water oxidation kinetics of the BiVO4 photoanode. This work elucidates the critical role of amorphous CMOH oxygen evolution catalysts in mediating photoinduced charge-carrier dynamics and establishes a facile and scalable strategy to broadly enhance the PEC performance of metal oxide photoanodes for large-scale solar fuel production.


  • Open Access

    Free-Standing Boron-Doped Graphene Hydrogel Buckypaper Fabricated via Zinc-Plate Reduction for High-Performance Supercapacitors

    by Jia-Yu Ji, Jeremiah Hao Ran Huang, Shih-Wen Tseng, Yi-Wen Chen, Jhao-Rong Huang, I-Wen Peter Chen

    Graphene offers exceptional mechanical flexibility and superior electrical conductivity that are highly desirable for next-generation supercapacitors; however, its intrinsic strong π-π stacking interactions severely hinder energy-storage performance by blocking active redox sites and restricting ion-accessible pathways. To address these limitations, we developed a binder-free, free-standing, boron-doped graphene hydrogel buckypaper with a hierarchical architecture. In this study, a mixed suspension of thermally activated boron-doped expanded graphene (BT-rGO) and graphene oxide (GO) was integrated via a redox reaction using a Zn plate as a substrate, enabling the instantaneous fabrication of a free-standing BT-rGO/Zn-rGO hydrogel buckypaper. The boron-doped graphene framework exhibits enhanced wettability, facilitating efficient electrolyte ion transportation, while the heteroatom-induced uneven charge distribution modulates the local electronegativity and improves electroconductivity throughout the free-standing hydrogel buckypaper network. The electrochemically enhanced E-BT-rGO3/Zn-rGO1 hydrogel buckypaper (BT-rGO:GO mass ratio = 3:1) delivers a high specific capacitance of 443.63 F g-1 at 2 mA cm-2 and maintains 88% capacitance retention after 10,000 cycles at 50 mA cm-2. Moreover, a flexible solid-state supercapacitor assembled from the BT-rGO3/Zn-rGO1 hydrogel buckypaper achieves an energy density of 16.96 Wh kg-1 and a power density of 5.21 kW kg-1, demonstrating its practical applicability for powering LED lightbulbs. Hence, the binder-free strategy creates a mechanically robust, porous framework with fully accessible ion pathways, delivering a purely carbon-based, metal-oxide-free electrode platform with an exceptional electrochemical performance.


  • Open Access

    A Lightweight Data-Augmented Deep Learning Framework for Real-Time Instance Segmentation in Liquid-Phase In Situ Transmission Electron Microscopy

    by Ming-Hao Shen, Wei-Che Chang, Wen-Huei Chu, Yu-Hsuan Cheng, Shih-Wen Tseng, Shu-Han Hsu

    Liquid-phase transmission electron microscopy (LPTEM) offers critical insights into the dynamic behaviors of materials, but automated analysis is hindered by the scarcity of annotated data and the high cost of image acquisition. To address these challenges, we propose an LPTEM deep learning framework that (i) closes the annotation gap, (ii) localizes nanoparticles through real-time detection, and (iii) preserves per-pixel detail. Through CycleGAN, we are able to generate a large number of images that are style-consistent with real LPTEM frames, effectively expanding the data set without manual labeling. After augmenting the data set, our fine-tuned YOLOv11n model achieves 97.66% precision and 99.05% mAP50 for detecting nanoparticles, closely approaching the performance of models trained on real data. Three Mobile-UNet variants, optimized for different computational efficiency and accuracy trade-offs, demonstrate instance segmentation for in situ TEM with intersection over union (IoU) scores ranging from 0.8685 to 0.9207 (on a scale where 1.0 indicates perfect overlap with the ground truth), indicating highly accurate particle shapes. Our integrated framework significantly enhances real-time detection and segmentation performance of nanoparticles in LPTEM, with YOLO + Mobile-UNet Slim achieving a real-time processing speed of 102.34 FPS. This framework establishes a scalable, annotation-efficient route to high-precision, real-time LPTEM analysis, opening the door to autonomous in situ experiments and accelerated materials discovery.