ICTP School in Computational Condensed Matter Physics: From Atomistic Simulations to Universal Model Hamiltonians,
Trieste, Italy, September 2015
Anders W. Sandvik, Boston University
Here I provide two programs (Fortran90) implementing the SSE and projector QMC methods for the S=1/2 Heisenberg model, following the algorithms discussed in the lectures.
Lecture slides
Lecture 1
Lecture 2
Lecture 3
Lecture 4
SSE QMC for S=1/2 Heisenberg model
Code: qmc_sse.f90 (simulation program; output should be processed by 'res_sse.f90')
Code: res_sse.f90 (program to compute averages and error bars from output of qmc_sse.f90)
Instructions: Compile the f90 programs using, e.g., gfortran. A read file 'read.in' with run parameters should be provided, with
two lines as follows:
Lx,Ly,Beta (system lengths in two dimensions, inverse temperature)
nbins,msteps,isteps (number of bins, number of sweeps for measurements and equilibration)
The boundary conditions in both the x and y directions are periodic, except for Ly=1 (a chain) and Ly=2 (a ladder), for which
only the x boundary is periodic. The random number generator needs a single seed integer in a file 'seed.in'. Output for each bin is
written to a file 'res.dat', one line of data for each bin. Averages and statistical errors are computed using this file as input
by res_sse, which produces output as follows:
m.dat: staggered magnetization and uniform susceptibility with error bars
e.dat: energy per site and specific heat with error bars
Projector QMC for S=1/2 Heisenberg model Code: qmc_pro.f90 (simulation program; output should be processed by 'res_pro.f90') Code: res_pro.f90 (program to compute averages and error bars from output of qmc_pro.f90)
Some of this material is based upon work supported by the National Science Foundation (USA) under Grants DMR-0803510, DMR-1104708, and DMR-1410126. Any opinions, findings, and conclusions or recommendations expressed in this material a re those of the authors and do not necessarily reflect the views of the National Science Foundation