modeling interface in conjugate heat transfer

Vivek

Hello I am modeling conjugate heat transfer in a channel with thick walls. The fluid is air and solid zone is silicon. The fluid is entering at the 1m/s and 350K. There is no heat flux applied anywhere and all the wall are adiabatic. The solution is initialized with 300K and I expect to get the isothermal temperature distribution in the system but temperature is increasing to 5000K. I think the problem is at the interface between solid and fluid zone. I have coupled the walls at the interface. Can anybody tell what should be the right boundary conditions at the interface?

Vivek