1. Dielectrophoresis (DEP)-Enabled Transfer Printing for MicroLED Displays and Multifunctional Semiconductor Devices Micro/nano-scale devices (LED emitters, nanowires, nanotubes, 2D materials) often require deterministic placement with wafer- or panel-scale throughput. We develop DEP- based transfer printing/assembly platforms that can selectively capture, align, and place building blocks from liquid-phase suspensions onto electrodes/backplanes—enabling both next-generation displays and heterogeneous semiconductor/sensor integration.

Core research topics

• Electric-field engineering for high-yield placement: fringing-field and electrode/insulator co-design, self-limiting DEP capture, and pixel-/site-confined assembly to enable deterministic transfer of microLEDs and heterogeneous semiconductor building blocks (nanorod LEDs, nanowires, and 2D flakes) with minimized misalignment and defectivity

• Heterogeneous device transfer: DEP placement of nanowires/CNT/2D materials for electronics and sensors; integration strategies to interface with CMOS and advanced packaging

• Scalable microLED transfer to backplanes: placement accuracy, yield modeling, redundancy/repair strategies, and process windows for panel-scale manufacturing

2. STC-Based Moiré Interferometry for Package Solder Deformation, Reliability, and Strain–Electronics Coupling in 3D Semiconductors As packaging evolves toward chiplets, 2D/3D integration, fine-pitch interconnects, and heterogeneous stacks, thermo-mechanical deformation becomes a leading cause of performance drift and reliability failure. We develop moiré interferometry–based metrology combined with STC-type computational strain reconstruction to quantify deformation at the package and device scales, and connect mechanical strain to electrical behavior.

Core research topics

• Thermo-mechanical deformation of solder/interconnects: displacement/strain mapping during thermal loading and cycling; fatigue-driven failure indicators

• Reliability physics and model validation: correlation of measured strain fields with finite-element models to build predictive lifetime models and design rules

• 3D semiconductor strain analysis: extracting strain distributions in advanced devices/3D architectures and linking them to electrical parameter shifts (mobility, threshold, variability)

• Electro–mechanical co-analysis: building compact models that translate strain maps into reliability predictions

3. Silicon Bottom-Cell Engineering for Perovskite/Si Tandem Solar Cells Using TOPCon and HJT Perovskite/Si tandem solar cells are widely viewed as a promising route to surpass the practical efficiency limits of single-junction c-Si. Accordingly, we focus on industrial passivated-contact Si platforms and the bottom-cell changes needed to make tandem integration practical and reliable.

Core research topics

• Bottom-cell performance preservation under tandem constraints: maximizing bottom-cell V OCand fill factor by maintaining excellent surface passivation and low contact resistivity, even after replacing conventional antireflection/passivation stacks with tandem-compatible interlayers.

• Parasitic optical loss minimization: reducing absorption losses in TCO, transport, and contact layers while preserving sufficient electrical conductivity, interfacial quality, and long-term operational stability.

• Texture–optics compatibility for industrial silicon: addressing the challenges of depositing functional layers on micrometer-scale pyramidal textures used in industrial Si wafers, and optimizing light management without sacrificing manufacturability.

• Interface robustness and reliability: ensuring chemical and thermal compatibility with low-temperature perovskite top-cell processing, together with long-term stability against heat, humidity, UV exposure, and thermal cycling.

4. Intelligent Reliability & Manufacturing: Machine Learning for Semiconductor Defect Analytics and Hybrid Neural Networks for Property Simulation Modern semiconductor and optoelectronic systems demand rapid feedback from inspection and simulation. We focus on ML-based image analytics and hybrid neural network models that couple physics and data to accelerate diagnosis, optimization, and reliability prediction.

Core research topics

• Defect analytics from images: automated inspection using optical/SEM/PL/EL/Xray images; segmentation/classification; anomaly detection

• Reliability prediction with AI: data-driven lifetime modeling, uncertainty quantification, and predictive maintenance for manufacturing and packaging