interface heat transfer

Dear lammps user:
I have a problem in my heat transfer simulation.I use the NEMD method .
First,I create a AlNi structure along the (111) direction and the eam potential is the NiAlH_jea.eam.alloy in the lammps.The system temperature is at the 300K. The timestep is 1fs and the system get the equilibrium state after two million steps.
Second,In the NVE ensemble ,I give the left end(Al region) 320K and give the right end(Ni region) 280K.The process lasted ten millions step.However,the temperature distribution is not satisfactory.The phenomenon of temperature jump happened between the Al material near the interface ,and,the temperature jump didn’t happen between Al and Ni .I am puzzled for a long time.
Thank you very much.
I am very grateful.

Xiaobai

How about posting a plot of the temperature
profile thru the system and indicating how
you computed it, and which atoms you thermostatted?

Steve