A research team led by National Taiwan Normal University physicists and National Cheng Kung University microscopy specialists grew 1H-stacked bilayer molybdenum disulfide (MoS₂) by chemical vapor deposition, reporting a direct band gap and stronger valley polarization than conventional bilayer forms in the journal Nature, Taiwan's National Science and Technology Council (NSTC) said in an Oct. 2 folksonomy briefing InfoHandle reviewed.
Why stacking geometry matters
Monolayer MoS₂ is prized for direct-gap photoluminescence, but most bilayer stacks adopt 2H or 3R arrangements that push carriers toward indirect transitions, dimming light emission. The NTNU–NCKU collaboration used a two-step CVD recipe to lock 1H alignment, suppressing the thermodynamic drift toward 3R stacking during growth.
Angle-resolved photoemission and density functional theory showed the valence band maximum staying at the K point roughly 57 millielectronvolts above the Γ point—experimental proof of direct-gap behavior in bilayers long treated as dark for optoelectronics.
Optical and valley signatures
Photoluminescence intensified excitonic emission without the indirect shoulder seen in 2H bilayers. Under resonant excitation at room temperature, circular polarization reached about 21 percent, comparable to monolayers; under off-resonance pumping, valley polarization in the A exciton nearly doubled monolayer values because intervalley scattering weakened in the top layer.
Who did what
Corresponding authors include NTNU professors Lan Yann-wen and Lu Ting-hua, postdoctoral fellow Yang Hung-wei, and NCKU core-facilities lead Lin Kuang-i. The Nature paper title is "Stacking-induced direct band gap in CVD-grown 1H MoS₂ bilayers," with publication timed to early October campus releases.
Industrial read-through
Foundry researchers watch bilayer transition-metal dichalcogenides as potential channel or photodetector materials once wafer-scale uniformity improves. A repeatable 1H bilayer process gives designers a knob—stacking registry—that changes band structure without altering chemical composition.
Limits the team noted
Yield and domain size still trail silicon CMOS; the paper emphasizes model-system validation rather than immediate smartphone shipments. Transfer to insulating substrates without wrinkle defects remains an engineering hurdle for valleytronic prototypes.
What comes next
NSTC said follow-on grants will test electrical gating of valley degrees of freedom and explore encapsulation stacks compatible with back-end-of-line temperatures. For Taiwan's research universities, the headline is a Nature proof that CVD growth can engineer quantum properties atom-by-atom—not only shrink feature sizes on silicon.
Synchrotron and ARPES cross-checks
Collaborators at the National Synchrotron Radiation Research Center supplied beam time for reciprocal-space maps that confirmed 1H registry across millimeter-scale domains, while ARPES at partner labs pinned the direct-gap offset reporters can cite without parsing supplementary figures.
From lab to foundry curiosity
TSMC and UMC researchers attending NTNU briefings asked whether CVD recipes can integrate with existing wafer-transfer tools; the team cautioned that bilayer TMD yields still trail silicon pilot lines but offer a testbed for stacking-aware device simulators.