issp-center-dev / HPhi

Quantum Lattice Model Simulator Package
https://www.pasums.issp.u-tokyo.ac.jp/hphi/en/
GNU General Public License v3.0
82 stars 25 forks source link

How should I use the standard mode to generate the following file for the expert mode?Just like the files you displayed #170

Open shuailiu199966 opened 2 weeks ago

shuailiu199966 commented 2 weeks ago

How can I generate the following file for expert mode using standard mode? I want to obtain the InterAll file for the model using standard mode. How should I proceed? I executed HPhi -sdry stan.in, but it did not produce an InterAll file. image

In your InterAll file, it is mentioned that to ensure the generated Hamiltonian is a Hermitian matrix, the input needs to include both image and its Hermitian conjugate image . How should I implement this in the InterAll file? For example, if I first input "2 0 2 1 3 0 3 1 -0.25 -0.25 ", do I need to continue inputting "3 1 3 0 2 1 2 0 -0.25 0.25" Does this operation need to be performed for all interaction terms, or only for off-diagonal terms? For example, should this be done for the x-bond and y-bond directions in the Kitaev model, but not for the interaction terms in the z-bond direction?

tmisawa commented 2 weeks ago

In standard mode, if the specified interactions are represented by standard interactions such as Hund coupling, Exchange coupling, PairLift interactions, etc, input files are generated in the format of the standard interactions.

W = 2 L = 3 model = "SpinGC" method = "CG" lattice = "Honeycomb" J0x = -1.0 J1y = -1.0 J2z = -1.0 2S=1 exct = 8

In this case, the Kitaev interactions are represented by CoulombInter, Hund, Exchange, and PairLift. Thus, if you want to specify the Kitaev interactions in InterAll format, please make an input file manually.

How should I implement this in the InterAll file? For example, if I first input "2 0 2 1 3 0 3 1 -0.25 -0.25 ", do I need to continue inputting "3 1 3 0 2 1 2 0 -0.25 0.25"

You are right. The Hermitian conjugate pair is necessary in InterlAll format.

Does this operation need to be performed for all interaction terms, or only for off-diagonal terms?

You can specify both diagonal terms and off-diagonal terms in InterAll format. In standard mode, the interactions that are not represented by the standard interactions such as J0xy are represented in InterAll format.

For example, should this be done for the x-bond and y-bond directions in the Kitaev model, but not for the interaction terms in the z-bond direction?

You can represent all interaction in InterAll format and you can also mix InterAll format and other formats for the standard interactions. In this case, you can specify x-bond and y-bond interactions in InterAll format and represent z-bond interaction in Ising interaction format.

shuailiu199966 commented 2 weeks ago

Can I manufacture a hexagonal honeycomb lattice using standard mode, as shown in the following figure image

ultimatile commented 2 weeks ago

Please refer to the HPhi manual. The figure below is taken from Figure 4.2 in the manual. スクリーンショット 2024-06-19 2 17 01

If my understanding of your requirements is correct, the specifications for the interactions with the input file provided by tmisawa-san will change as follows according to the figure above:

J1x = -1.0
J2y = -1.0
J0z = -1.0
tmisawa commented 6 days ago

@shuailiu199966 The 24-site honeycomb cluster can be defined in the following input file.

model = SpinGC method = CG lattice = Honeycomb J0x = -1.0 J1y = -1.0 J2z = -1.0 a0W = 2 a0L = 2 a1W = 4 a1L = -2 exct = 1

This 24-site cluster consists of two 12-site loops and topologically equivalent to your lattice. You can check the geometry of this lattice by execting load ./lattice.gp in gnuplot. I attached the lattice structure generated by gnuplot. スクリーンショット 2024-06-28 15 45 49