Ancilla wavefunctions of Mott insulator and pseudogap metal through quantum teleportation

TL;DR

This paper introduces a novel wavefunction for Mott insulators using quantum teleportation with ancilla qubits, capturing both spin liquid and charge fluctuations, and providing new insights into metal-insulator transitions.

Contribution

It presents a new class of wavefunctions for Mott insulators that incorporate quantum teleportation, enabling continuous tuning of the charge gap and revealing potential fractional Fermi liquid phases.

Findings

Wavefunction at inities recovers Gutzwiller wavefunction

Charge gap ollows the parameter 

Numerical confirmation in one dimension

Abstract

Weak Mott regime with finite U is a wonderful region to search for quantum spin liquid, but it is challenging to write down a wavefunction capturing both spin liquid and charge fluctuations. Conventional methods using complicated Jastrow factors have difficulties when the underlying spin liquid has a non-trivial projective symmetry group (PSG). To cure this problem, here we provide a new class wavefunction for Mott insulator through quantum teleportation using ancilla qubits. We primarily focus on half filling of the fermionic Hubbard model. We will prove that a single variation parameter $\Phi$ in our wavefunction tunes the Mott charge gap continuously. On a generic lattice, we show that the wavefunction at $\Phi=+\infty$ recovers the familiar Gutzwiller projectived wavefunction at infinite U. The wavefunction at $\Phi=\frac{U}{2}$ is equivalent to applying the inverse Schrieffer Wolff transformation at linear order of $t/U$, as expected in large but finite U regime. From a gauge theory description we can show that the wavefunction has an electronic sector decoupled from a spinon sector describing localized spin moments. The charge gap $\Delta_c$ can be shown to be $2\Phi$ and we conjecture that the wavefunction works well down to the regime with small charge gap on a generic lattice. We represent the wavefunction using tensor network and numerically confirm this conjecture in one dimension. Beyond the numerical power, the ancilla wavefunction also provides a new conceptual picture to understand the bandwidth tuned metal insulator transition. In this new framework, there can in principle exist a narrow region of fractional Fermi liquid (FL*) phase between the usual Fermi liquid and the Mott insulator, a scenario which is not captured by the conventional slave rotor theory and thus was usually outlooked.