🔬 materials science

Atomically-precise synthesis and simultaneous integration of 2D transition metal dichalcogenides enabled by nano-confinement

This study presents a versatile nano-confinement strategy using van der Waals capping layers to enable the atomically precise synthesis of tailored 2D transition metal dichalcogenides and Janus monolayers while simultaneously creating ultraclean interfaces that preserve air-sensitive properties.

Original authors: Ce Bian, Yifan Zhao, Roger Guzman, Hongtao Liu, Hao Hu, Qi Qi, Ke Zhu, Hao Wang, Kang Wu, Hui Guo, Wanzhen He, Zhaoqing Wang, Peng Peng, Zhiping Xu, Wu Zhou, Feng Ding, Haitao Yang, Hong-Jun Gao

Published 2026-03-03

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Ce Bian, Yifan Zhao, Roger Guzman, Hongtao Liu, Hao Hu, Qi Qi, Ke Zhu, Hao Wang, Kang Wu, Hui Guo, Wanzhen He, Zhaoqing Wang, Peng Peng, Zhiping Xu, Wu Zhou, Feng Ding, Haitao Yang, Hong-Jun Gao

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 you are trying to build a house out of incredibly delicate, single-layer sheets of paper (these are the 2D materials). The problem is, these sheets are so thin and fragile that if you try to build them in the open air, they get crumpled, torn, or covered in dust before you even finish. Furthermore, if you try to stack two different types of paper together, the glue you use often leaves sticky, messy residue that ruins the magic of the paper.

This paper describes a brilliant new "construction method" that solves all these problems. Here is the breakdown of what the scientists did, using simple analogies:

1. The "Under-the-Table" Construction Site

Usually, when scientists grow these materials, they do it in an open oven (like baking cookies on a tray). This leads to uneven layers and contamination.

In this study, the researchers used a clever trick: Nano-confinement.

The Analogy: Imagine placing a clear, protective glass lid (made of graphene or hexagonal boron nitride) over your workbench. You then build your delicate paper house underneath that lid.
Why it works: The space between the lid and the table is tiny. This "nano-confined" space acts like a strict traffic cop. It forces the building blocks (atoms) to line up perfectly in a single layer. If they try to stack up too high, they get squished by the lid. This ensures you get a perfect, single-layer sheet every time, with almost zero mistakes (98% success rate!).

2. The "One-Way Street" for Building Blocks

Normally, building materials rain down from the sky (the oven) onto the top of your house. This causes layers to pile up on top of each other randomly.

The Analogy: In this new method, the "rain" of building blocks can't fall from the top because the glass lid is in the way. Instead, the blocks have to sneak in through the edges of the lid, like water seeping under a door.
The Result: Once inside, the blocks can only attach to the edges of the growing sheet, not the top surface. It's like a game of Tetris where you can only add blocks to the side of the stack, never on top. This guarantees the sheet stays exactly one layer thick, no matter how long you build.

3. The "Janus" Mask (The Two-Faced Coin)

The researchers also wanted to create a special type of material called a "Janus" monolayer. Think of a coin: one side is Silver, the other is Gold. In normal chemistry, it's very hard to turn just the top side of a coin into Gold without accidentally turning the bottom side too.

The Analogy: Because the scientists built their house under the protective glass lid, the "top" side of the material was shielded like a knight in armor. They could then send in a chemical agent that only attacked the "bottom" side (the side touching the table).
The Result: They successfully created a material with a Silver top and a Gold bottom (specifically, Sulfur on top and Selenium on the bottom) with atomic precision. This creates a material with unique electrical properties that didn't exist before.

4. The "Time-Travel" Preservation

Some of these materials are like fresh fruit; they rot (oxidize) instantly when exposed to air. Usually, scientists have to move them into a vacuum chamber immediately, which is risky and messy.

The Analogy: Because the material was built under the glass lid, it is essentially "sealed in a time capsule" the moment it is born.
The Result: The scientists showed that these materials stayed fresh and perfect for over 60 days just sitting on a shelf in normal air. It's like keeping a strawberry fresh for two months without a fridge!

5. The "Superconductor" Superpower

One of the materials they made was Niobium Diselenide (NbSe2), which is a superconductor (it conducts electricity with zero resistance). However, this property is very sensitive to damage.

The Analogy: Because the material was so clean and perfectly preserved by the "glass lid," it became a super-superconductor. It worked better than any other version made by traditional methods, even matching the quality of materials made by expensive, high-tech machines.
Bonus: They could even draw shapes (like rings) directly on the material just by cutting the shape of the glass lid they used, avoiding the need for messy cutting tools that usually damage the delicate sheets.

The Big Picture

This paper introduces a universal "construction kit" for the future of electronics. By building these ultra-thin materials under a protective shield, scientists can:

Build them perfectly (no mistakes).
Stack them cleanly (no sticky glue).
Keep them fresh (no rotting).
Create new, weird shapes and properties that were previously impossible.

It's a shift from "trying to build a sandcastle in a hurricane" to "building it in a calm, controlled room," opening the door to faster, smaller, and more powerful computers and sensors.

1. Problem Statement

Two-dimensional (2D) transition metal dichalcogenides (TMDs) like $NbSe_2$ and $MoS_2$ possess unique properties (e.g., superconductivity, direct bandgaps, piezoelectricity) that are highly sensitive to their atomic-scale structures, specifically layer thickness, atomic composition, and symmetry. However, the field faces three critical challenges:

Precise Monolayer Synthesis: Conventional Chemical Vapor Deposition (CVD) struggles to suppress the nucleation of adlayers (multilayers) due to the delicate balance between precursor deposition rates and surface diffusion.
Atomic-Precision Janus Synthesis: Creating Janus TMDs (asymmetric structures like S-Mo-Se) via one-side substitution is difficult because the opposite chalcogen plane is often unintentionally modified due to poor kinetic control, leading to alloy phases rather than pure Janus structures.
Clean Integration: Fabricating van der Waals (vdW) heterostructures typically requires transfer methods that introduce interfacial contamination, degrading the intrinsic properties of air-sensitive materials (like $NbSe_2$ ) and preventing the observation of emergent phenomena.

2. Methodology: Nano-Confined CVD

The authors introduce a nano-confined CVD approach where a pre-existing 2D capping layer (graphene or hexagonal boron nitride, hBN) is mechanically exfoliated onto a substrate ( $SiO_2/Si$ ). TMD precursors are then introduced into the interface between the capping layer and the substrate.

Growth Mechanism:
- Edge Intercalation: Precursors diffuse directionally from the edges of the capping layer (or through defects) into the confined space.
- Kinetic Control: The confined environment restricts precursor attachment exclusively to the in-plane edges of the growing crystal. This prevents vertical nucleation (adlayer formation), naturally enforcing monolayer growth regardless of macroscopic deposition rates.
- Energetic Favorability: Density Functional Theory (DFT) calculations show that the binding energy of TMD monolayers in the nano-confined state (sandwiched between the capping layer and substrate) is significantly more negative than in open growth, thermodynamically stabilizing the monolayer.
Janus Synthesis Strategy: To create Janus $MoSSe$, $MoS_2$ monolayers are grown under the capping layer. $NbSe_2$ precursors are then introduced. Due to the distinct interfacial properties, precursors selectively intercalate into the bottom $MoS_2/SiO_2$ interface (where the substrate is rough and chemically active) but cannot penetrate the top vdW gap (graphene/hBN-MoS $_2$ ). This allows for the selective substitution of only the bottom sulfur atoms with selenium.

3. Key Contributions

Versatile Nano-Confined Platform: A unified method for synthesizing various 2D TMDs ( $NbSe_2$ , $MoS_2$ ) and their heterostructures with atomic precision.
High-Yield Monolayer Synthesis: Achieved a 98% monolayer yield for $NbSe_2$ , a significant improvement over the ~41% yield of conventional open-growth CVD.
Atomically-Precise Janus TMDs: Successfully synthesized high-quality Janus $MoSSe$ monolayers with true atomic-plane selectivity (bottom S replaced by Se, top S protected), overcoming the thermodynamic preference for alloy phases.
In Situ Ultraclean Integration: The method simultaneously grows the TMD and encapsulates it with the capping layer, creating ultraclean vdW interfaces without post-synthesis transfer contamination.
Intrinsic Patterning: By tuning precursor flux and capping layer geometry, the authors demonstrated the ability to grow intrinsically patterned TMD rings that conform to the edges of the capping layer, eliminating the need for lithography.

4. Key Results

Structural Characterization:
- Cross-sectional STEM and EELS confirmed the formation of single-layer $NbSe_2$ and $MoS_2$ sandwiched between the capping layer and $SiO_2$ .
- Raman spectroscopy showed characteristic monolayer peaks (e.g., large $A_{1g}-E_{2g}$ separation in $NbSe_2$ ) and the absence of multilayer signatures.
Janus $MoSSe$ Verification:
- Raman and Photoluminescence (PL) spectra confirmed complete one-side substitution (disappearance of $MoS_2$ peaks, appearance of Janus-specific modes).
- HAADF-STEM imaging revealed different atomic intensities for the top (S) and bottom (Se) chalcogen layers, confirming the polar structure.
- The vdW gap was measured at ~3.7 Å, indicating an ultraclean interface.
Air Stability and Superconductivity ( $NbSe_2$ ):
- Nano-confined $NbSe_2$ monolayers remained stable in ambient air for over 60 days, whereas open-grown monolayers degraded almost immediately.
- Superconducting Transition: The encapsulated $NbSe_2$ exhibited an onset critical temperature ( $T_c$ ) of 2.8 K, which is significantly higher than conventional CVD-grown samples (~1.2–1.4 K) and comparable to mechanically exfoliated or MBE-grown samples. This demonstrates the preservation of Ising pairing due to the clean interface.
Device Performance:
- Intrinsically-Patterned Rings: $NbSe_2$ rings grown under patterned hBN flakes showed metallic behavior and sharp superconducting transitions, suitable for 2D circuits.
- Optoelectronics: Graphene/Janus $MoSSe$ heterostructures exhibited photoresponse governed by charge transfer, with persistent photoconductivity states, highlighting potential for photomemory applications.

5. Significance

This work establishes a versatile platform that solves the long-standing trade-off between synthesis precision, material quality, and integration cleanliness in 2D materials.

Scalability: By leveraging mature wafer-scale CVD techniques for graphene and hBN, this method offers a pathway to wafer-scale synthesis of high-quality TMD films and patterned circuits.
Material Discovery: The ability to synthesize Janus TMDs with atomic precision opens new avenues for exploring polar 2D materials with tailored piezoelectric, thermoelectric, and catalytic properties.
Quantum Applications: The demonstrated enhancement in superconducting performance and air stability of $NbSe_2$ suggests that this nano-confined approach is critical for developing robust 2D quantum devices and superconducting circuits.
Fundamental Insight: The study elucidates distinct growth kinetics (edge intercalation vs. surface deposition) and provides a kinetic phase diagram for controlling TMD morphology (domains, films, rings) without post-processing.