Tuning Electrons Between Layers: Controlled Charge Transfer in Moiré Superlattices
11 May 2026, Yanjiang
A vertical electric field precisely controls the transfer of electrons between layers in a MoSe₂/WS₂ moiré superlattice.
In a cleanroom at the University of California, Berkeley, researchers built a structure only a few atoms thick. They placed a single layer of molybdenum diselenide directly on top of a single layer of tungsten disulfide — two different semiconductor crystals, each just three atoms tall. Then they applied an electric field across the stack and watched what happened to the electrons inside. What they found was a system that can be tuned with remarkable precision, switching electrons from one layer to the other and back again on command.
The work, led by physicist Feng Wang and his colleague Michael P. Zaletel, appears in a preprint on arXiv (arXiv:2605.05571). It offers a detailed look at how electrons arrange themselves in these artificial crystalline structures, and how an external voltage can control their behavior.
A natural superlattice
When two crystals with slightly different lattice constants are stacked together, their atomic patterns do not line up exactly. The mismatch creates a larger repeating pattern a regular moiré structure. Think of two overlapping window screens at a slight angle: the dark and light bands that appear are a moiré pattern. In the quantum world, this pattern acts as a new periodic potential for electrons.
The team used a specific combination: MoSe₂ on WS₂, with the crystals aligned at a 60‑degree twist. Earlier studies had shown that this pair forms a type of band alignment called Type‑I at zero electric field, meaning the lowest energy states for electrons are located entirely within the MoSe₂ layer. But the new work reveals that this alignment can be flipped.
By applying a vertical electric field across the bilayer, the researchers could switch the system from Type‑I to Type‑II band alignment. In Type‑II, the lowest conduction band states shift into the WS₂ layer, while the highest valence band states remain in MoSe₂. As the authors write, “Moiré superlattices formed by transition metal dichalcogenide heterobilayers provide a versatile platform for studying strongly correlated electronic, excitonic, and topological phenomena in solids.” Their experiment makes that platform even more powerful.
Electrons that move between layers
The key observation was a series of optical signals that appeared only when the electric field crossed a certain threshold. These signals come from interlayer charge-transfer transitions — events in which an electron moves from one layer to the other, leaving behind a hole. The team detected four such transitions, corresponding to doping levels from one to four times the fundamental moiré density (n/n₀ = 1, 2, 3, and 4). Each transition occurs at a specific electric field strength, as if the system has built‑in energy levels that can be addressed one by one.
This behavior is a direct consequence of the moiré potential. At low doping, the added electrons localise in the MoSe₂ layer, forming what are called moiré trions — bound states of two electrons and a hole. When the field is reversed, those electrons are pulled into the WS₂ layer, and the trions transform into neutral excitons (electron-hole pairs) in the same layer. The entire process is reversible and continuous.
The researchers describe this as realising a Fermi‑Hubbard model on an effective honeycomb lattice. The Hubbard model is a simplified description of electrons hopping between lattice sites, with an energy penalty when two electrons occupy the same site. Here, the moiré pattern creates the lattice, and the vertical electric field controls the hopping and interaction parameters. The result is a tunable system that can mimic strongly correlated materials.
Order from disorder
The team did not stop at optical spectroscopy. They also performed Monte Carlo simulations to understand how the electrons organise themselves as the doping increases. The simulations predict that at both integer and fractional fillings — such as n/n₀ = 1, 2, 1/2, and others — the system develops ordered charge states. These are not the simple uniform filling one might expect but rather patterned arrangements where some moiré sites hold electrons while others are empty.
This is significant because such charge‑ordered phases are central to many exotic states in condensed matter, including superconductivity and topological order. A tunable platform that can access these phases on demand would allow researchers to study them under controlled conditions, rather than relying on happenstance in real materials.
The authors note that the electric‑field susceptibility — how easily the charge distribution responds to the field — changes dramatically with doping. At n/n₀ = 1, the susceptibility is large, indicating the electrons are highly mobile. At higher fillings, it drops, suggesting the system becomes more rigid as ordered structures form. The Monte Carlo results match the optical measurements remarkably well, giving confidence that the theoretical model captures the essential physics.
A bridge to the future
What makes this work stand out is the degree of control it offers. In most moiré systems, the lattice structure is fixed once the stack is built. Here, the interlayer charge transfer can be tuned in real time using an external voltage, without changing the geometry. That opens the door to studying transitions between different ordered states — for instance, switching from a charge‑density wave to a different phase by turning a knob.
The team’s successful observation of four distinct charge‑transfer transitions is a clear demonstration of the system’s tunability. Their Monte Carlo predictions suggest that even richer behaviour awaits at fractional fillings, where strong correlations can give rise to exotic liquids and spin states.
For now, the work establishes MoSe₂/WS₂ moiré bilayers as a laboratory for the Hubbard model — a model that has been studied theoretically for decades but rarely realised so cleanly in experiment. The next steps will involve probing those predicted ordered phases directly, perhaps by measuring electrical transport or using scanning probes.
The road ahead is clear, even if the exact timeline remains uncertain. What matters is that we have a new tool: a tiny sandwich of two materials, controlled by a simple voltage, that can teach us how electrons dance to the beat of a moiré rhythm.
Yanjiang is an online editor of LoomSci
References
- Michael P. Zaletel et al., Tunable Interlayer Charge-transfer States in MoSe$_2$/WS$_2$ Moiré Superlattices, arXiv:2605.05571