Thermal Interference Analysis and Compensation for Monolithic Integrated Thermopile Infrared Sensors

To meet the demand for high-performance, miniaturized, and low-cost sensing systems in the field of temperature monitoring, chip-level monolithic integration of sensors and integrated circuits (ICs) is an effective solution for constructing sensing-processing integrated intelligent microsystems. However, this monolithic integrated structure inherently introduces thermal interference issues, directly limiting the system’s temperature measurement accuracy. In this work, we designed a monolithic integrated thermopile infrared sensor developed using standard complementary metal oxide semiconductor (CMOS) process technology and conducted thermal interference coupling analysis and temperature compensation studies based on its structural characteristics. First, through structural analysis of this type of sensor, we identified the presence of thermal mismatch between the hot and cold ends, as well as thermal transient temperature rise phenomena. The former manifests as spatial temperature imbalance, causing static error, while the latter manifests as temporal temperature changes, causing dynamic error. Subsequently, we experimentally validated these two types of errors and constructed a static reference correction model combined with an RC network dynamic compensation model, achieving synergistic suppression of both types of thermal interference. Experimental results show that after static reference correction, the static deviation caused by thermal mismatch between the hot and cold ends decreased from 0.22 °C to 0.06 °C, with an error reduction of 73%. After dynamic temperature compensation, the dynamic error caused by the self-heating temperature rise of the tube shell decreased from 0.42 °C to 0.08 °C, with an error reduction of 80.9%. This study provides the theoretical basis and compensation methods for improving temperature measurement accuracy in monolithic integrated temperature measurement systems.