回复: thermal conductivity with Muller-Plathe

Dear Steve
Thank you!
I have changed the code, and get the temp profile

So “is it ok” means “is it right in RNEMD principle”
CL Wang

1AB460DA@...4334....png

FF325672@...4334....png

4E967492@...4338....png

MP basically is a RNEMD method with controlled heat flux, so you need the periodic boundary condition, namely, doubling the domain size you are interested in. (for instance, http://iopscience.iop.org/0957-4484/12/1/305)

There are other methods utilizing adiabatic boundary condition, for example, http://prb.aps.org/abstract/PRB/v76/i8/e085424, but you need to modify the code more to meet the method’s requirement.

LC Liu

FF325672@...4334....png

1AB460DA@...4334....png

4E967492@...4338....png