🔬 materials science

Stacking-Directed Polarization and Excitonic Engineering in MoS $_2$ /MoSe $_2$ van der Waals Heterostructures

Using GW+BSE many-body perturbation theory, this study reveals that while MoS $_2$ /MoSe $_2$ heterostructures lack switchable interlayer dipoles due to intrinsic chemical potential mismatch, their specific stacking sequences enable deterministic control of photogenerated electrons and interlayer excitonic shifts, establishing them as promising platforms for sliding ferroelectricity and programmable optoelectronics.

Original authors: Mohammed El Amine Miloudi, Oliver Kühn

Published 2026-06-23

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Mohammed El Amine Miloudi, Oliver Kühn

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine two ultra-thin, magical sheets of material, like sheets of graphene but made of specific atoms (Molybdenum, Sulfur, and Selenium). Scientists call these "transition metal dichalcogenides." When you stack these sheets on top of each other, they don't stick together like glue; they float slightly apart, held by weak forces, much like a stack of playing cards.

This paper explores what happens when you slide these cards around relative to each other. The researchers found that how you align these layers changes the way electricity and light behave inside them, almost like tuning a radio to a different station just by shifting the antenna.

Here is a breakdown of their findings using simple analogies:

1. The "Sandwich" Experiment

The scientists built two types of sandwiches:

The Homogenous Sandwich: Two identical layers of Molybdenum Disulfide ( $MoS_2$ ).
The Mixed Sandwich: One layer of $MoS_2$ and one layer of Molybdenum Diselenide ( $MoSe_2$ ).

They discovered that the "flavor" of the sandwich (the chemical makeup) matters just as much as the "arrangement" (the stacking pattern).

2. The "Sliding Ferroelectric" (The Identical Sandwich)

When they stacked two identical $MoS_2$ layers, they found something cool: The stack can act like a switchable magnet for electricity.

The Analogy: Imagine two identical people standing face-to-face. If they stand perfectly aligned (stack A), they are neutral. But if one person slides slightly to the left or right (stack B), their balance shifts.
The Result: This sliding breaks the symmetry. It creates an internal electric field, like a tiny battery built right into the material.
The Switch: If you slide the layers back the other way, the "battery" flips its polarity. The electricity flows in the opposite direction. The researchers call this "sliding ferroelectricity." It's a way to store information (like a 0 or a 1) just by physically sliding the layers.

3. The "Chemical Anchor" (The Mixed Sandwich)

When they mixed the two different materials ( $MoS_2$ and $MoSe_2$ ), the story changed.

The Analogy: Imagine trying to slide a heavy wooden block across a floor. If the floor is made of the same wood, you can slide it easily. But if the floor is made of sticky rubber (a different chemical), the block gets stuck.
The Result: In the mixed sandwich, the difference between Sulfur and Selenium atoms is so strong that it "pins" the electricity in place. No matter how you slide the layers, the internal electric field doesn't flip. The chemical difference acts like an anchor, overriding the sliding effect. The electricity is locked into a specific pattern (Type-II alignment) where electrons and holes are forced to stay on opposite sides of the sandwich.

4. The "Three-Layer Cake" (The Trilayer)

The researchers then added a third layer, creating a $MoS_2$ - $MoS_2$ - $MoSe_2$ stack. This is where things got really interesting.

The Analogy: Think of a three-layer cake where the bottom two layers are identical, but the top layer is different. The bottom two layers can still slide around to change the flavor, but the top layer acts as a heavy lid.
The Result: By sliding the bottom two layers, the scientists could control where the electrons go.
- In one sliding position, the electrons prefer the middle layer.
- In the other sliding position, the electrons get pushed to the bottom layer.
The Effect: This sliding creates an internal "gate" that moves the energy levels of the electrons up or down by a tiny but precise amount (about 60–70 meV). It's like having a dimmer switch for the energy of the electrons, controlled entirely by how you stack the layers.

5. The "Light Show" (Excitons)

Finally, they looked at how these stacks interact with light. When light hits the material, it creates "excitons" (pairs of an electron and a "hole" that act like a single particle).

The Finding: Because the stacking changes the internal electric fields, it also changes the color (energy) of the light the material emits.
The Match: The scientists predicted that sliding the layers would shift the light energy by about 36 meV. This prediction matched almost perfectly with real-world experiments (which showed about 40 meV).
The Takeaway: This proves that the "sliding" mechanism is the real reason the light changes color. It's a precise way to tune the material's optical properties without changing the material itself.

Summary

In simple terms, this paper shows that in these ultra-thin materials, geometry is power.

If the layers are identical, you can slide them to flip an internal electric switch.
If the layers are different, chemistry locks the switch in place.
If you stack three layers, you can use the sliding motion to precisely move electrons around and tune the color of light the material emits.

This isn't just about understanding atoms; it's about discovering a new way to control electricity and light by simply rearranging the order of atomic layers, like shuffling a deck of cards to change the game.

Technical Summary: Stacking-Directed Polarization and Excitonic Engineering in MoS2/MoSe2 van der Waals Heterostructures

Problem Statement
While two-dimensional van der Waals (vdW) heterostructures offer versatile platforms for tuning electronic and optical properties via stacking sequence and rotational alignment, a systematic understanding of how atomic registry governs many-body excitonic behavior in MoS2-based systems remains incomplete. Existing theoretical frameworks often rely on mean-field approximations that fail to capture non-local screening and enhanced Coulomb interactions inherent to 2D interfaces. Specifically, the fundamental relationship between local stacking order, interfacial dipoles, and the resulting quasiparticle band alignments at the many-body GW–BSE level requires rigorous elucidation. Furthermore, it is unclear how chemical asymmetry (e.g., in MoS2/MoSe2 heterobilayers) competes with geometric registry to determine polarization and charge transfer.

Methodology
The authors employed a high-fidelity theoretical framework combining Density Functional Theory (DFT) with many-body perturbation theory.

Structural Modeling: Calculations were performed using the Vienna Ab initio Simulation Package (VASP) with projector-augmented wave (PAW) potentials. Exchange–correlation effects were treated using the PBE functional with Grimme's D3 dispersion correction. Commensurate models were constructed for bilayers (MoS2 homobilayer and MoS2/MoSe2 heterobilayer) and trilayers (2L-MoS2/MoSe2) by applying biaxial strain to match lattice constants.
Electronic Structure: To account for many-body effects and reduced dielectric screening, eigenvalue-self-consistent quasiparticle calculations were performed within the EVGW0@PBE framework.
Excitonic Response: Optical properties were evaluated by solving the Bethe–Salpeter equation (BSE) atop the quasiparticle landscape.
Analysis: The study systematically analyzed various stacking configurations (AA, AB, BA for bilayers; AAA, ABA, ABB, BAA, BAB, AAB, BBA for trilayers) to quantify charge density differences (CDD), electrostatic potentials, band alignments, and excitonic binding energies.

Key Contributions and Results

1. Bilayer Systems: Homobilayer vs. Heterobilayer

MoS2 Homobilayer: The study confirms that lateral sliding between AB and BA registries breaks global inversion symmetry, inducing a finite out-of-plane electric polarization ( $P_z$ ). This results in switchable interfacial dipoles of opposite polarity (upward for AB, downward for BA) driven by registry-induced symmetry breaking and asymmetric orbital hybridization. This creates a bistable system with a potential offset of $\sim$ 55 meV and an internal electric field of $\sim$ 0.86 V/nm, establishing the microscopic prerequisite for sliding ferroelectricity.
MoS2/MoSe2 Heterobilayer: In contrast, the hetero-interface remains pinned by the intrinsic chemical potential mismatch between sulfur and selenium. Despite the presence of a type-II band alignment and significant charge transfer, the out-of-plane polarization is not switchable by stacking registry. The chemical asymmetry dominates over geometric registry, suppressing the collective, registry-dependent charge transfer required for bistable dipolar states.

2. Trilayer Systems (2L-MoS2/MoSe2)

Deterministic Control: In trilayer configurations (MoS2/MoS2/MoSe2), the stacking sequence enables deterministic control over the localization of photogenerated electrons. The MoS2/MoS2 interface acts as a tunable degree of freedom, while the MoS2/MoSe2 interface remains pinned.
Band Alignment: Stacking-induced asymmetry allows for the fine-tuning of polarization states. Internal electric fields induce quasiparticle band-edge shifts of 60–70 meV, dictating whether electrons localize in the central or bottom MoS2 layer, while holes remain confined to the MoSe2 layer.
Excitonic Shifts: The many-body excitonic spectra reveal a strong sensitivity to atomic registry. The calculated average interlayer excitonic shift between AB-type (ABA/ABB) and BA-type (BAA/BAB) stacking families is 36 meV. This shift arises from a rigid translation of the electrostatic potential landscape upon dipole reversal, affecting conduction and valence band edges similarly while leaving the exciton binding energy largely unchanged.

Significance and Claims
The paper claims to establish a microscopic link between atomic registry and many-body interactions in transition metal dichalcogenide (TMD) trilayers. The primary findings are:

Validation of Mechanism: The calculated 36 meV interlayer excitonic shift is in remarkable agreement with recent experimental values of approximately 40 meV, confirming that stacking-governed internal dipoles are the dominant mechanism for excitonic energy tuning in these systems.
Platform Potential: The results suggest that TMD trilayers serve as a potential platform for sliding ferroelectricity and programmable optoelectronic functionalities. The ability to reversibly modulate interfacial polarization and excitonic transitions via stacking sequence offers a predictive roadmap for designing non-volatile excitonic transistors and atomically controlled ferroelectric optoelectronic logic.
Methodological Rigor: The work demonstrates the necessity of moving beyond standard approximations (like PBE) to EVGW0+BSE frameworks to accurately quantify the interplay between local stacking order, interfacial dipoles, and quasiparticle band alignments in 2D heterostructures.