The fix heat command (https://lammps.sandia.gov/doc/fix_heat.html) mentions that heat in this context is an ‘extensive’ property.
I want to confirm if the following is consistent:
If the entries are given as follows:
fix 1 hot heat 1 $(HEATIN)
fix 2 cold heat 1 $(HEATOUT)
do we interpret heats in cold and hot atom-groups to have the following relation?
HEATOUTnatom(cold) = HEATINnatom(hot)
If otherwise, let me know what is the right way to evaluate the total externally-applied heats.
Should we assume the same to be true when using the ‘ehex’ as well?
The behavior depends on the type of variable being supplied; if its not a per atom variable the value supplied is the heat given to the whole group of atoms. If it is a per atom variable then the value supplied is heat used to rescale each individual atom according to that per atom variable; in this case the number of atoms in the group and your variable function’s form becomes relevant to determining the total heat, otherwise you only need to be concerned with the input heat scalar if your variable is not a per atom variable.
Thanks Adrian for your reply,
Following the manual, the non-atom-style variables are deemed similar to a numeric constant too, when assigning flux, and in that regard, a numeric constant would also pertain to “extensive” and hence be scaled appropriately to get the total heat flux?
If there are (say) 500 atoms in group ‘hot’ and 200 in group ‘cold’ and we have the following command
fix 1 hot heat 1 1.115e-1,
would the total heat flux into the hot end be 1.115e-1*500 (units)?
And if so, to maintain constant heat flux throughout the system, we’d have to force the heat flux out as
fix 2 cold heat 1 (1.115e-1*500/200) ?
I’ll appreciate if that could be clarified.