Skip to Main Content
Good control of wafer temperature during the postexposure bake (PEB) process is gaining importance. Although negative feedback techniques can improve temperature control, poor thermal contact between the sensor and the wafer will cause the feedback signal quality and the performance of the closed-loop system to deteriorate. A plausible solution is to predict the wafer temperature using the inverse sensor model constructed via data from a loop current step response (LCSR) test. However, the LCSR test must be completed before the PEB process can commence, resulting in a loss in wafer throughput. In this paper, a sensor parameter estimation algorithm is proposed to enable the LCSR test and the PEB process to be concurrently performed. Simulation results demonstrated that the sensor parameters can be reasonably accurately estimated, provided that the start of the PEB process coincided exactly with the instant at which the resistance temperature detector current is switched from a high value (self-heating mode) back to its nominal value (sensing mode). As this condition is not practical, a workaround is proposed to enable the accurate identification of sensor parameters in practice. Finally, experimental results are presented. They demonstrate that the proposed algorithm is able to identify the sensor parameters. Consequently, consistent control performance can be achieved, regardless of the level of thermal contact between sensor and wafer.