Quantum critical behavior of a three-dimensional superfluid-Mott glass transition

TL;DR

This study investigates the quantum phase transition from superfluid to insulator in a disordered three-dimensional bosonic system, revealing how disorder influences critical behavior and universality classes through large-scale Monte Carlo simulations.

Contribution

It provides the first detailed numerical analysis of the superfluid-Mott insulator transition in 3D disordered bosons, demonstrating the relevance of disorder and characterizing the critical exponents.

Findings

Disorder changes the critical behavior from mean-field to power-law.

Critical exponents are dilution-independent below the percolation threshold.

Transition behavior at the percolation threshold is governed by classical percolation exponents.

Abstract

The superfluid to insulator quantum phase transition of a three-dimensional particle-hole symmetric system of disordered bosons is studied. To this end, a site-diluted quantum rotor Hamiltonian is mapped onto a classical (3+1)-dimensional XY model with columnar disorder and analyzed by means of large-scale Monte Carlo simulations. The superfluid-Mott insulator transition of the clean, undiluted system is in the 4D XY universality class and shows mean-field critical behavior with logarithmic corrections. The clean correlation length exponent $\nu = 1/2$ violates the Harris criterion, indicating that disorder must be a relevant perturbation. For nonzero dilutions below the lattice percolation threshold of $p_c = 0.688392$, our simulations yield conventional power-law critical behavior with dilution-independent critical exponents $z=1.67(6)$, $\nu = 0.90(5)$, $\beta/\nu = 1.09(3)$, and $\gamma/\nu = 2.50(3)$. The critical behavior of the transition across the lattice percolation threshold is controlled by the classical percolation exponents. Our results are discussed in the context of a classification of disordered quantum phase transitions, as well as experiments in superfluids, superconductors and magnetic systems.