I recall another user working on that at [1].
I don't know if they resolved it. I also don't know if they did a dimensional analysis [2] on what they did or not.
From looking at [3], VAMPIRE should output a mean-specific-heat or a material-mean-specific-heat in kB/spin. The kB is the Boltzmann constant [4].
Since 1 kB = 1.380649×10^(-23) J/K, that should make it easy to convert kB/spin to J/K/spin.
The more challenging part might be how to get spin to kg.
Maybe you could assume 1 spin equals 1 atom. If so, from the geometry of the cell that you've specified in the VAMPIRE simulation, you likely could determine the atomic density
[5] in units of atoms/cell_volume.
From that, I think could get the volumetric heat capacity. Then, according to [6], the volumetric heat capacity can multiplied by the material or mass density to express the heat capacity per unit of mass (i.e., units of JK-1kg-1).
Hopefully that can help.
Kind Regards,
Gavin
VAMPIRE user