Speaker: Dr. Zhengcheng GU   (Chinese Univ. of Hong Kong)

Abstract

In this talk, I will discuss some nontrivial applications for tensor product state(TPS) approach to strongly correlated systems. The first nontrivial example is the spin-1/2 J1-J2 Heisenberg model on square lattice. On small clusters, the TPS variational ground state energy is perfectly consistent with other exact methods,such as exact diagonalization (ED) and DMRG. For larger systems, the TPS approach works much better than other known methods and consequently a DQCP is discovered in the maximally frustrated regime J2~0.5J1 [40]. Remarkably, the derived critical exponents perfectly agree with the known sign problem free J-Q model, which can also realize the DQCP between the Neel order and the valence bond solid (VBS) order.  We have also applied this method to the famous t-J model (the strong coupling limit of Hubbard model) at t/J=3 on honeycomb lattice. At half filling, an anti-ferromagnetic (AF) ground state is found and the results are comparable to the best of Quantum Monte Carlo results. At finite doping, the ground state energy can be obtained up to 0.1%~0.5% accuracy comparing to other unbiased methods for small clusters, such as DMRG and ED. Different from previous studies, the Grassmann TPS approach does not assume any specific ordering pattern a priori, hence it is unbiased in the variational sense. In a wide regime of doping, a d+id chiral topological superconducting phase is discovered.

Time: 10:00am, March 31 (Saturday), 2018

Venue:  Rm. S401, Kavli ITS meeting room [View Map]