🔬 materials science

Alloy engineering of excitonic properties in TMD monolayers

This study demonstrates that alloy engineering in monolayer MoS $_{2x}$ Se $_{2(1-x)}$ enables continuous tuning of the optical gap and systematic control of excitonic properties, including B–A splitting and circular polarization, through composition-dependent modifications of the electronic structure and phonon energies.

Original authors: Eirini Katsipoulaki, Adlen Smiri, Panagiotis Spiliotakis, Konstantinos Mourzidis, Danae Katrisioti, Takashi Taniguchi, Kenji Watanabe, Georgios Kopidakis, Zdenek Sofer, Gang Wang, Emmanuel Stratakis

Published 2026-08-05

📖 3 min read☕ Coffee break read

CC BY 4.0

Original authors: Eirini Katsipoulaki, Adlen Smiri, Panagiotis Spiliotakis, Konstantinos Mourzidis, Danae Katrisioti, Takashi Taniguchi, Kenji Watanabe, Georgios Kopidakis, Zdenek Sofer, Gang Wang, Emmanuel Stratakis, George Kioseoglou, Iann C. Gerber, Xavier Marie, Ioannis Paradisanos

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 a world where the tiny building blocks of our future electronics aren't just solid bricks, but flexible, atom-thin sheets that can be stretched, twisted, and mixed like paint. This is the realm of two-dimensional materials, specifically a family of semiconductors called Transition Metal Dichalcogenides (TMDs). Think of these materials as microscopic stages where electrons perform a complex dance. In this dance, the electrons have a "valley" they live in (like a valley on a map) and a "spin" (like a tiny internal compass pointing up or down). Usually, you can't easily change the rules of this dance once the material is made. But what if you could mix two different types of these materials together, like blending red and blue paint to get purple, to create a brand new material with custom-made properties? This is the question scientists have been asking: Can we "tune" these atomic sheets to control how they handle light, heat, and information just by changing their recipe?

In this study, researchers decided to play chef with a specific recipe: mixing Molybdenum Diselenide (MoSe₂) and Molybdenum Disulfide (MoS₂). They created a whole series of "alloys" by swapping sulfur atoms for selenium atoms in varying amounts, creating a smooth gradient from one material to the other. They wanted to see if they could continuously dial up or down the material's properties, like turning a volume knob, rather than just switching between two fixed settings.

The team discovered that they could indeed "tune" the material with incredible precision. By changing the ratio of sulfur to selenium, they were able to shift the color of light the material emits across a range of about 0.35 electron volts (eV). To put that in perspective, it's like being able to smoothly slide a light switch from a deep red glow to a bright blue glow without any jumps or gaps. They found that as they added more sulfur, the energy gap between two specific electron states (called the A and B excitons) shrank steadily, a behavior that matched perfectly with their computer simulations.

But the magic didn't stop at just changing colors. The researchers also looked at how the atoms in the material vibrate (phonons). They found that the average energy of these vibrations increased as the material became richer in sulfur, rising from about 17 meV in the selenium-rich mix to roughly 22 meV in the sulfur-rich mix. This change followed a simple rule based on the weight of the atoms involved, much like how a heavier guitar string vibrates at a different pitch than a lighter one.

Perhaps the most exciting finding involved the "spin" of the electrons. When they shone circularly polarized light on the samples, they measured how well the material kept the "handedness" of that light. In the selenium-rich samples, the material almost completely forgot the direction of the spin, resulting in nearly zero polarization. However, as they added more sulfur, the material got better and better at remembering, reaching about 15% polarization in the pure sulfur version. The scientists believe this happens because mixing the atoms changes the electronic structure just enough to stop the electrons from getting confused and losing their spin direction.

In short, this paper shows that by simply mixing two ingredients, scientists can create a "dial" to control the color, the vibrations, and the spin-memory of these ultra-thin materials. It proves that alloy engineering is a powerful tool for designing future electronic and optical devices, offering a way to customize materials for specific jobs without needing to invent entirely new substances from scratch.

Technical Summary: Alloy Engineering of Excitonic Properties in TMD Monolayers

Problem Statement
While semiconductor alloying is a well-established method for tailoring properties in conventional bulk materials (e.g., AlGaAs, SiGe), its application to two-dimensional transition metal dichalcogenides (TMDs) requires a more comprehensive understanding. Previous studies on TMD alloys, such as MoS $_{2x}$ Se $_{2(1-x)}$ , have primarily focused on bandgap engineering and excitonic transition energies. However, the systematic evolution of vibrational properties (phonon energies), spin-orbit-related excitonic splittings, and valley polarization across the full alloy composition range has remained less explored within a unified experimental framework. This study addresses the need to understand how chalcogen substitution modulates the interplay between electronic structure, lattice dynamics, and spin-valley physics in monolayer TMDs.

Methodology
The authors investigated monolayer MoS $_{2x}$ Se $_{2(1-x)}$ alloys with compositions ranging from pure MoSe $_2$ ( $x=0$ ) to pure MoS $_2$ ( $x=1$ ), including intermediate compositions ( $x \approx 0.3, 0.52, 0.7, 0.95$ ). The samples were fabricated via mechanical exfoliation of bulk crystals grown by flux zone methods and encapsulated between hexagonal boron nitride (hBN) flakes to ensure high optical quality and a clean dielectric environment.

The study employed a multi-faceted approach:

Optical Spectroscopy: Temperature-dependent (8 K to 300 K) and polarization-resolved photoluminescence (PL), photoluminescence excitation (PLE), and differential reflectivity measurements were conducted. Circular polarization was measured under fixed detuning conditions (93 meV) to ensure comparable valley polarization data across compositions.
Theoretical Modeling: Density Functional Theory (DFT) calculations were performed using the VASP package with the Special Quasi-Random Structure (SQS) approach to model substitutional disorder. These calculations provided insights into spin-orbit splittings and electronic band structures.
Data Analysis: Experimental data were analyzed using Vegard's law for bandgap trends, an empirical Varshni-like expression for temperature-dependent bandgap shifts to extract average phonon energies, and a reduced-mass scaling model for vibrational properties.

Key Results

Optical Gap and Vegard's Law: The optical gap was found to be continuously tunable over a range of approximately 0.35 eV (from 1.63 eV in MoSe $_2$ to 1.93 eV in MoS $_2$ at 8 K). The excitonic transition energies exhibit nearly linear behavior with composition, characterized by negligible bowing parameters ( $b_A \approx -0.064$ eV, $b_B \approx -0.069$ eV), indicating limited perturbation to the electronic structure by alloying.
B–A Exciton Splitting: The energy separation between the A and B excitons decreases monotonically from $\sim$ 200 meV in MoSe $_2$ to $\sim$ 150 meV in MoS $_2$ . This reduction correlates with the decreasing spin-orbit coupling strength as sulfur replaces selenium. DFT calculations of the spin-split band-edge states ( $\Delta_v + \Delta_c$ ) closely match the experimental trend, confirming that the splitting is governed by the alloy-induced modification of the valence and conduction band spin-orbit splittings.
Phonon Energy Evolution: Temperature-dependent PL analysis revealed a systematic increase in the average phonon energy ( $\langle \hbar\omega \rangle$ ) from $\sim$ 17 meV in Se-rich alloys to $\sim$ 22.5 meV in S-rich alloys. This trend is accurately described by a reduced-mass scaling model ( $\langle \hbar\omega \rangle \propto 1/\sqrt{\mu_{eff}}$ ), suggesting that the vibrational properties are primarily governed by the change in the effective mass of the Mo-chalcogen bond rather than changes in the force constant.
Circular Polarization: Polarization-resolved spectroscopy at 78 K demonstrated a monotonic increase in exciton circular polarization from nearly 0% in MoSe $_2$ to $\sim$ 15% in MoS $_2$ . This enhancement is attributed to alloy-induced modifications of the conduction band spin splitting, which alters the bright–dark exciton splitting ( $\Delta_{bd}$ ). In MoSe $_2$ , a small $\Delta_{bd}$ facilitates efficient Rashba-assisted mixing and depolarization, whereas in MoS $_2$ , a larger $\Delta_{bd}$ suppresses this mixing, preserving valley polarization.

Significance and Claims
The paper establishes chalcogen alloying as an effective and stable strategy for the controlled tuning of excitonic, vibrational, and spin-valley properties in 2D semiconductors. The authors claim that their work provides a unified experimental and theoretical framework demonstrating that:

Electronic structure modifications via alloying follow predictable, near-linear trends (Vegard-like) with minimal disorder-induced nonlinearity.
Vibrational properties can be engineered simply by adjusting the reduced mass of the constituent atoms.
Valley polarization efficiency can be systematically enhanced by shifting the alloy composition away from the near-degenerate bright–dark exciton regime characteristic of MoSe $_2$ .

These findings offer new insights into the fundamental interplay between lattice dynamics and spin-valley physics in TMD alloys, providing a basis for tailoring excitonic functionalities in two-dimensional materials without relying on external perturbations like strain or doping.

More like this