In Tutorial 1, see Sec. 7, we have computed exchange constants for KCuF3 . Here we go further and will use fully automated regime auto to calculate magnetic susceptibility. The jass.inp is nearly the same as in Tutorial 1:

[main]
ncore = 32
[exchanges]
mode = auto
S = 0.5
magion = Cu

If you run JaSS now it will first compute exchange constants, then construct spin lattice and execute solver for the solution of the Heisenberg model and generate file chi.dat, where magnetic susceptibility is written.

Figure 1: Magnetic susceptibility of KCuF3 as obtained by JaSS.

In such a run we chose default method for the solution of the Heisenberg model (quantum Monte-Carlo qmc) and used default number of sweeps (Sweeps), thermalization sweeps (Thermalization), size of the cell (L), minimal (Tmin) and maximal temperature (Tmax) and temperature step (dT). Alternatively, one may specify them in jass.inp in section Heisenberg:

[main]
ncore = 32
[exchanges]
mode = auto
S = 0.5
magion = Cu
[heisenberg]
method = qmc
sweeps = 60000
thermalization = 2000
tmax = 600
tmin = 20
L = 8

Magnetic susceptibility calculated with this set of parameters is shown in Fig. 1. One may see that it has maximum at ∼250 K. This is in a fairly good agreement with experimental χ(T ) having peak at 243 K [1]. All files are available at tutorial archive.

To do: To get better impression how stable these results are and what parameters can be tuned try to change number of cells used in the QMC method and study low temperature behaviour of χ(T ). Also try to change number of sweeps and choice of the solver of the Heisenberg model.

1. S. Kadota, I. Yamada, S. Yoneyama, and K. Hirakawa, J. Phys. Soc. Jpn. 23, 751 (1967)

Back to top