High-powered LED arrays have emerged as a viable alternative to halogen lamp-based arrays for heating silicon wafers to 1000C due to their monochromatic radiosity which can be chosen in an efficient bandwidth for high absorption into silicon, removing the requirement for the pre-
...
High-powered LED arrays have emerged as a viable alternative to halogen lamp-based arrays for heating silicon wafers to 1000C due to their monochromatic radiosity which can be chosen in an efficient bandwidth for high absorption into silicon, removing the requirement for the pre-soak anneal in traditional systems. However, a gap in literature is found as there are only low temperature applications for LED arrays found (max. 400C), on which the research is focussed around improving the control system design. The high-temperature RTP application for LEDs has not yet been studied in literature or implemented in industry. Advancements over the past decades have sharply increased the radiometric flux and efficacy of LEDs, enabling more and more high-powered applications. However, due to the exceptionally small size and high heat fluxes of high-powered LED chips, junction cooling has traditionally been a major bottleneck for implementing LEDs in demanding applications. For the power levels required in 1000C RTP, the feasibility of LEDs in terms of power output, wafer temperature uniformity and cooling of LED-based heater arrays remains unproven. Through a systems-based approach, a multi-domain fundamental analysis and simulation in the fields of linear algebra, heat transfer, optics, electrical engineering, semiconductor physics, CAD and FEM has been performed. Resultingly, feasibility has been demonstrated for 1000C LED RTP.