Uniform nucleation and epitaxy of bilayer molybdenum disulfide on sapphire
Summary¶
This Nature article addresses a practical bottleneck for two-dimensional transition-metal dichalcogenide (TMD) transistors: although monolayer TMD growth has matured, uniform epitaxial bilayers with wafer-scale homogeneity have remained elusive, even though bilayer channels are argued to be an attractive compromise between electrostatic control, density of states, and mobility for deeply scaled devices. The authors report chemical vapor deposition (CVD) of bilayer molybdenum disulfide (MoS₂) on c-plane sapphire where engineering atomic terrace heights steers an edge-nucleation pathway and enables >99% uniform bilayer nucleation with coalescence into centimeter-scale continuous films. Fabricated field-effect transistors (FETs) on bilayer channels show field-effect mobility up to 122.6 cm² V⁻¹ s⁻¹ with improved device-to-device variation relative to monolayer-film controls under the reported processing. Short-channel FETs further report on-state current 1.27 mA μm⁻¹, which the authors compare against an industry 2028 high-performance FET roadmap target cited in the abstract.
Methods¶
The growth strategy couples CVD with substrate engineering on c-sapphire to manipulate terrace structure and favor bilayer nucleation and orientation over the fragmented bilayer flakes typical of prior attempts. The paper’s introduction contrasts layer-by-layer epitaxy routes that may be too slow for manufacturing with the targeted CVD throughput. Thermodynamic arguments in the article use density functional theory (DFT) estimates of edge energies and surface energies to rationalize why bilayer domain evolution is nontrivial on vdW-like substrates. Device fabrication spans multiple channel lengths to probe mobility statistics and short-channel drive current; detailed fabrication stacks, dielectric integration, and metrology appear in the full text and Extended Data.
Findings¶
Experimentally, the work demonstrates large-area bilayer MoS₂ films with high bilayer yield and FET metrics that improve on monolayer baselines in the reported cohort, supporting the claim that bilayer epitaxy can combine uniformity and competitive transport for beyond-silicon exploratory technology. The short-channel current benchmark is presented as exceeding a cited roadmap figure, framing bilayer TMDs as candidates for high-drive, scaled transistors if contact and integration challenges continue to improve. The study is experimental and DFT-informed rather than a reactive MD paper; it complements atomistic simulation literature on MoS₂ in this corpus by establishing a growth/device benchmark.
Limitations¶
Wafer-edge defect maps, transfer-integration yield, and long-term reliability are outside the scope of the abstract-level summary and require the full article. This entry is not a ReaxFF study; mechanistic atomistic claims should not be imported from unrelated force fields.
Relevance to group¶
Experimental 2D TMD epitaxy benchmark; complements computational MoS₂ ReaxFF papers (e.g., Ponomarev et al.) but is not an atomistic simulation study.