宇晶股份实现45μm超薄半片硅片加工 鼎力支持轻量化HJT组件落地

Core Viewpoint - Hunan Yujing Machinery Co., Ltd. has achieved a technological breakthrough in its dual crystal slicing base in Yancheng, Jiangsu, successfully conducting the first trial cut of 45μm ultra-thin silicon wafers using its self-developed ultra-precision thermal stability cutting technology [1][2] Group 1: Technological Advancements - The company has developed a specialized multi-wire cutting machine and cutting coolant specifically for ultra-thin HJT half wafers, marking a significant advancement in the thin wafer technology field [1] - The first batch of samples underwent a 40-minute high-precision online testing system validation, demonstrating outstanding electrical performance and physical characteristics, which align with the research and development needs of high-end flexible HJT and stacked batteries [1][2] Group 2: Equipment and Materials - The self-developed slicing machine features micron-level axle spacing precision and excellent dynamic stability, providing a solid foundation for ultra-thin cutting [2] - The company's self-developed tungsten wire diameter has surpassed Φ15μm, improving output rates and yield while significantly reducing breakage rates [2] Group 3: Process Optimization - The CrystalMind slicing model optimizes over a hundred cutting parameters in real-time, achieving superior performance with 45μm thick half wafers, including 180° bending without edge chipping and flexible pressure resistance for 30 minutes without hidden cracks [2] - The company emphasizes lean production principles throughout the process, effectively reducing silicon material loss and overall processing costs while enhancing production stability and efficiency through highly automated operations [2] Group 4: Industry Significance - The successful trial cut of 45μm ultra-thin silicon wafers represents a significant breakthrough in domestic photovoltaic equipment manufacturing and precision processing, reflecting the integration of ultra-precision control, micron-level detection, and intelligent parameter optimization across multiple disciplines [2]