Hello, I encountered an interesting issue while using HPhi, and I'm not sure whether this problem was caused by me or by HPhi itself.

[shuailiu199966]

I calculated the energy of a 24-site honeycomb cluster for the pure Kitaev model, pure Gamma model, Kitaev + magnetic field, and Gamma model + magnetic field.

I get the correct results when I compute the pure Kitaev model, pure Gamma model, and pure Kitaev + magnetic field. However, when I copy the magnetic field file to the pure Gamma model, the result is incorrect. What could be the reason for this?

The files for the following four models.

gamma_E.zip gamma+h_E.zip kitaev+h_E.zip

If possible, I would like to know how to calculate the pure Gamma model in standard mode.

[shuailiu199966]

gamma+h_E.zip

[tmisawa]

What criteria do you use to determine that the results are correct? If it is based on energy comparison, I would appreciate it if you could provide the correct energy values.

Anyway, Gamma term can be specified in Standard mode. If we regard J0 as x-bond, J1 as y-bond, and J2 as z-bond, the Gamma term can be specified as

J0yz=G J0zy=G J1zx=G J1xz=G J2xy=G J2yz=G

For details of the definitions of J, please see the manual.

http://issp-center-dev.github.io/HPhi/manual/master/en/html/filespecification/standardmode_en/Parameters_for_the_Hamiltonian_en.html

[shuailiu199966]

I first used the standard mode to calculate the energy of the pure Gamma model, and the result was the same as the energy obtained using the expert mode. Then, I provided the magnetic field file to both modes for calculation, but the energies obtained were completely incorrect. The results from the standard mode with the magnetic field file match the results from my own ED code and DMRG calculations. gamma+h_S.zip gamma_S.zip These two files are for the pure Gamma model and the Gamma model with a magnetic field in standard mode. The model energies are located in the out folder.

[k-ido]

I performed the HPhi calculation in the expert mode using input files generated by the standard mode, but the energy is the same as just using the standard mode. So the problem does not seem to originate from using the expert mode.

I guess that your input files for the Gamma terms in the expert mode are inconsistent with those generated by the standard mode. By using the following parameters in the standard mode, I can reproduce the energy in gamma+h_E.zip and gamma_E.zip.

J0yz = -1.0
J0zy = -1.0
J1xz = -1.0
J1zx = -1.0
J2xy = 1.0
J2yx = 1.0

[shuailiu199966]

My biggest confusion right now is why the input file for the pure Gamma model and the magnetic field file both work correctly when used separately, but when used together, they only produce incorrect results?