Combinatorial optimization problems arise in different fields of science and engineering. There exist some general techniques coping with these problems such as simulated annealing (SA). In spite of SA success, it usually requires costly experimental studies in fine tuning the most suitable annealing schedule. In this Letter, the classical integrated circuit placement problem is faced by Thermodynamic Simulated Annealing (TSA). TSA provides a new annealing schedule derived from thermodynamic laws. Unlike SA, temperature in TSA is free to evolve and its value is continuously updated from the variation of state functions as the internal energy and entropy. Thereby, TSA achieves the high quality results of SA while providing interesting adaptive features.