Mapping the Mottness under Magnetic Field

Lei Wang,LingNan Wei,Zhao-Long Gu,Ammon Fischer,Yuchi He,Qiaoling Xu,Augusto Ghiotto,Clara S. Weber,Kenji Watanabe,Takashi Taniguchi,Martin Claassen,Ángel Rubio,Andrew J. Millis,Abhay Pasupathy,Lede Xian,Daniel Rhodes,Dante M. Kennes,Jian-Xin Li
DOI: https://doi.org/10.21203/rs.3.rs-3385782/v1
2023-01-01
Abstract:Abstract Charge and spin are two intrinsic degrees of freedom of an electron. In a Mott insulator (MI), they are separated in that the charge degree is frozen due to strong Coulomb interaction U while the spin degree can still fluctuate and is governed by the energy scale of spin exchange J ≈ 4t 2 /U ( t , hopping integral). Usually, a magnetic field can only manipulate the spin degree as the attainable Zeeman energy Ez can only compete with J at the scale of tens of milli-electron volts, while it does not affect the charge degree with U being at least two orders of magnitude larger [1–3]. Therefore, the complete evolution of a Mott insulator under magnetic field with both charge and spin impacted has yet to be explored. With recent progress in two-dimensional (2D) moiré systems, Mott insulators have been realized with significantly reduced U [4–9], providing an opportunity to investigate this problem. Here, we map out the full picture of Mottness under magnetic field by transport measurements on twisted bilayer (TB) WSe 2 with twist angle θ ranging from 2.4° to 3.2°. We achieve the tuning of the charge dynamics of the Mott insulator using magnetic fields and observe unexpected Mott insulator-metal-Mott insulator (IMI) transitions. We theoretically establish that these IMI transitions are driven by an exotic form of spectral weight transfer (SWT) between the spin-split Hubbard bands due to carrier occupancy variations of different spins. We further identify the reentrant insulating phase at higher magnetic fields as a Mott-Zeeman insulator (MZI) with its gap determined collectively by U and Ez . Remarkably, with SWT, the observed critical magnetic field for this reentrance is as low as 5 T, which corresponds to a Zeeman energy approximately an order of magnitude smaller than U . The unveiled IMI transitions experimentally and theoretically evidence SWT as a crucial driving mechanism for the evolution of a Mott insulator in response to control parameters, which offers an unprecedented perspective into the nature of Mott physics.
What problem does this paper attempt to address?