Hi again all,
I just had a short question about the TEBD code in ALPS. I would like to study the evolution of an initial state that is a Neel state, but rotated by an angle /alpha about the y-axis, under the XXZ Hamiltonian. A Hamiltonian whose ground state would be the right one would simply have a magnetic field along the direction of /alpha that swaps direction at each site i.e. (-1)^l *(cos /alpha *S^z_l + sin /alpha *S^x_l) where l is the site number, but I can't see a way to put this into the code. Is alternating the sign of the interaction in this way possible? Also, if there is another way of producing such a state in the TEBD code, it would be wonderful if someone pointed it out, I would be very grateful.
Thank you very much,
Joseph Prentice
Hi Joseph,
At present, the only initial states supported by tebd are ground states of translation-invariant Hamiltonians or the 'kink' initial state used in the spin tutorials. You can edit the source file tebd.f90 to write in another product state of your choice. If this is the route you would like to follow I can give you more details. The next generation of MPS codes for ALPS will include much greater flexibility in the choice of initial state.
-Michael
On Tue, Aug 6, 2013 at 6:25 AM, Joseph Prentice < Joseph.Prentice@physics.ox.ac.uk> wrote:
Hi again all,
I just had a short question about the TEBD code in ALPS. I would like to study the evolution of an initial state that is a Neel state, but rotated by an angle /alpha about the y-axis, under the XXZ Hamiltonian. A Hamiltonian whose ground state would be the right one would simply have a magnetic field along the direction of /alpha that swaps direction at each site i.e. (-1)^l *(cos /alpha *S^z_l + sin /alpha *S^x_l) where l is the site number, but I can't see a way to put this into the code. Is alternating the sign of the interaction in this way possible? Also, if there is another way of producing such a state in the TEBD code, it would be wonderful if someone pointed it out, I would be very grateful.
Thank you very much,
Joseph Prentice
comp-phys-alps-users@lists.phys.ethz.ch