Molybdenum disulfide

MoS₂

Also called molybdenite (mineral)

van der Waals crystal semiconductor

The most studied 2D semiconductor. As a single layer it becomes a direct-gap semiconductor that emits light and makes a transistor that switches properly on and off – what graphene cannot do. It is the leading candidate for ultra-thin transistor channels in future chips, and in bulk form it has been used for decades as a dry lubricant and a refinery catalyst.

Cell: Hexagonal, a = 3.16 Å
Atoms per cell: 3
Mo–S bond: 2.42 Å
Height: 3.17 Å between the outer atom centres

A plane of molybdenum between two planes of sulfur. Upper and lower sulfur atoms sit directly above one another, so each Mo is inside a trigonal prism of six S atoms. A single layer has no inversion centre, which is why its K and K′ valleys can be addressed separately with circularly polarised light. Geometry of bulk 2H-MoS₂ at room temperature: a = 3.16 Å, sulfur planes 1.59 Å above and below the Mo plane.

Data sheet

Show all 10 entriesHide the data sheet

Kind of sheet: van der Waals crystal
Peels off a layered crystal, one sheet at a time
Behaves as: semiconductor
Band gap: Indirect ~1.2–1.3 eV in bulk; direct in the monolayer, with an optical gap (A exciton) of ~1.85–1.9 eV and a quasiparticle gap of roughly 2.2–2.5 eV depending on the substrate

In colour: this family, 1.2–2.5 eV. In grey: the other 39 families on the catalogue’s chart, each at the middle of its range.
Thickness: ~0.65 nm per layer
Structure: Each layer is a sheet of Mo atoms between two S sheets in trigonal-prismatic coordination (1H); bulk is usually 2H, sometimes 3R. A metallic octahedral 1T/1T′ phase forms on lithium intercalation
Space group: MoS₂ (2H): bulk P6₃/mmc, monolayer P6m2
Bulk: the Materials Project crystal linked below. Monolayer: the 2DMatPedia layer, as a periodic slab
Stability: Stable in air for months at room temperature, although monolayers slowly oxidise at defects and edges, faster under light and humidity. Oxidation in air accelerates above roughly 300 °C.
First made: Few-layer flakes cleaved with adhesive tape in 1966 (Frindt); single layers by lithium intercalation in 1986 (Joensen, Frindt and Morrison); monolayer direct gap reported in 2010 (Mak and colleagues; Splendiani and colleagues)
Materials Project: MoS₂ (2H)
Computed crystal structure and properties
2DMatPedia: MoS₂
Computed monolayer, exfoliated from a bulk crystal of the same compound

Compare with other materials

What it looks like

A silvery, mirror-bright crystal with a six-sided outline and a flat, slightly flaking face, sitting in white quartz. — Molybdenite, the mineral form of MoS₂, on quartz from the Molly Hill mine in Quebec. The crystal is 15 mm across, and its flat six-sided face is the top of the stack of layers. Image: John Chapman (Pyrope), CC BY-SA 4.0; resized.

A black-and-white honeycomb pattern of bright spots, five places in it ringed with green dashed circles, and a yellow scale bar. — A single layer of MoS₂ in a scanning transmission electron microscope. Each bright spot is a column of atoms seen end-on, and the circles mark sulfur sites where atoms are missing. Scale bar: 1 nm. Image: Jinhua Hong and colleagues, Nature Communications 6, 6293 (2015), CC BY 4.0; resized.

Key properties

Monolayer photoluminescence is roughly 10⁴ times more efficient than bulk – the signature of the indirect-to-direct crossover
A and B excitons near 1.88 and 2.03 eV (monolayer, room temperature)
Room-temperature electron mobility typically 10–100 cm²/(V·s) in devices, limited by defects, interfaces and contacts
Valence-band spin–orbit splitting at K of ~150 meV; valley-selective excitation with circularly polarised light
Young’s modulus ~270 GPa; monolayers break at roughly 6–11% strain
Contact resistance as low as 42 Ω·µm with semimetallic antimony contacts (2023), after 123 Ω·µm with bismuth (2021)

How it is made

Mechanical exfoliation of natural molybdenite or synthetic crystals – reference flakes, although natural crystals carry many defects
CVD from MoO₃ and sulfur at ~650–850 °C, often salt-assisted – triangular single-crystal domains that merge into polycrystalline films
MOCVD from Mo(CO)₆ and a sulfur precursor – uniform wafer-scale monolayers
Epitaxy on vicinal sapphire with aligned domains – wafer-scale single-orientation films
Sulfurisation of pre-deposited Mo or MoO₃ films – fab-friendly but polycrystalline
Lithium-intercalation or electrochemical exfoliation – solution-processable nanosheets, partly converted to the metallic 1T phase

Uses, and how close they are

Transistor channels for advanced logic and back-end-of-line circuitsprototype
Photodetectors and flexible optoelectronicsprototype
Electrocatalytic hydrogen evolution at edge siteslab
Dry lubricant and hydrodesulfurisation catalyst (bulk MoS₂)deployed

Readiness runs lab → prototype → pilot → deployed.

Open problems

What limits mobility in wafer-scale films – chalcogen vacancies, grain boundaries, interface traps or remote phonons – and which of these can be engineered away?
Can a stable p-type MoS₂, or a well-matched p-type partner, be achieved for complementary 2D logic?
How many reported ‘sulfur vacancies’ are really oxygen substitutions or antisite defects?

Going deeper

Short notes for specialists. Choose a lens in the header and yours comes first.

For theoreticians · your lens

Band edges at K are dominated by Mo d orbitals, so a three-band tight-binding model captures the valley physics; spin–orbit coupling splits the valence band by ~150 meV but the conduction band by only a few meV. PBE gets the direct monolayer gap right in character but wrong in size – GW–BSE with Coulomb truncation is needed for quasiparticle gaps and exciton binding energies (a few hundred meV, strongly screening-dependent). Defect studies must separate sulfur vacancies from substitutional oxygen, which look similar in STM but differ in their gap states.

For experimentalists · your lens

Count layers with Raman: the separation between E′ (~385 cm⁻¹) and A′₁ (~403–405 cm⁻¹) is ~18–20 cm⁻¹ in a monolayer and grows towards ~25 cm⁻¹ in bulk (532 nm excitation). Confirm monolayers by strong PL at ~1.85–1.9 eV and by SHG, which is strong for odd and absent for even layer numbers in 2H stacking. A′₁ responds to doping and E′ to strain, so do not infer layer number from peak positions alone on doped or strained samples.

For engineers · your lens

The most mature 2D semiconductor for integration: MOCVD delivers uniform wafer-scale monolayers, and 300 mm test transistors have been built in industrial research fabs. The hard problems are contact resistance (industry targets are around 100 Ω·µm), growth temperatures above back-end-of-line budgets (typically ≤400–450 °C), dielectric deposition on an inert basal plane, and the lack of a stable p-type counterpart.

In the research tracks

Theory frontiers

Growth methods

Readiness assessments

Recent news

The newest items tagged MoS₂, from the news feed updated 5 Oct 2026.

Journal 5 Oct 2026 Nano Letters

Direct-Write Defects for Bias-Free Two-Dimensional Optical Synapses Towards In-Sensor Motion Detection

Optoelectronic synaptic devices, particularly those based on two-dimensional (2D) transition metal dichalcogenides (TMDs), emulate biological synaptic functions, promising in neuromorphic computing. However, most TMD-based synapses rely on external gate or bias control, leading to complicated device architecture and…

ExperimentMoS₂

Journal 2 Oct 2026 ACS Applied Materials & Interfaces

Decoupling Thermal and Electronic Effects of Metal Contacts in Bilayer MoS2 Transistors

Contact resistance remains a primary bottleneck in two-dimensional (2D) semiconductor field-effect transistors (FETs), yet the concurrent influence of metal work function, adhesion, and deposition temperature obscures the underlying mechanisms. Here, using bilayer MoS2 as a model system, we systematically benchmark…

ExperimentEngineeringMoS₂

Preprintnot yet peer reviewed 1 Oct 2026 arXiv

Layer-by-layer growth of highly aligned MoS₂ nanoribbon arrays

One-dimensional transition metal dichalcogenides (TMD) nanoribbons (NRs) offer a promising route to aggressive channel-width scaling in nanoscale transistors. However, controlling their layer number, a key determinant of device performance, has remained elusive. Here, we demonstrate a chemical vapor deposition (CVD)…

ExperimentOxide nanosheets MoS₂ WS₂

Preprintnot yet peer reviewed 30 Sep 2026 arXiv

Quantum Well Resonant Tunneling Diode Probe of Correlated States in Twisted Bilayer MoS₂

Moiré superlattices formed in transition metal dichalcogenides (TMDs) offer a versatile platform for exploring emergent quantum phases arising from strong electronic correlations. In this work, we develop a new experimental platform, the quantum well resonant tunneling diode (QWRTD), to probe the electronic landscape…

TheoryExperimentMoS₂

Preprintnot yet peer reviewed 30 Sep 2026 arXiv

Qantum Well Resonant Tunneling Diode Probe of Correlated States in Twisted Bilayer MoS₂

TheoryMoS₂

All 115 items tagged MoS₂ in the news feed · RSS feed for MoS₂

Key references

Single crystals of MoS₂ several molecular layers thickFrindt · Journal of Applied Physics 37, 1928 (1966)cited by 352doi:10.1063/1.1708627
Single-layer MoS₂Joensen et al. · Materials Research Bulletin 21, 457 (1986)cited by 1,356doi:10.1016/0025-5408(86)90011-5
Atomically thin MoS₂: a new direct-gap semiconductorMak et al. · Physical Review Letters 105, 136805 (2010)cited by 15,430doi:10.1103/PhysRevLett.105.136805
Emerging photoluminescence in monolayer MoS₂Splendiani et al. · Nano Letters 10, 1271 (2010)cited by 9,520doi:10.1021/nl903868w
Single-layer MoS₂ transistorsRadisavljevic et al. · Nature Nanotechnology 6, 147 (2011)cited by 14,996doi:10.1038/nnano.2010.279
High-mobility three-atom-thick semiconducting films with wafer-scale homogeneityKang et al. · Nature 520, 656 (2015)cited by 1,963doi:10.1038/nature14417
Ultralow contact resistance between semimetal and monolayer semiconductorsShen et al. · Nature 593, 211 (2021)cited by 1,414doi:10.1038/s41586-021-03472-9
Approaching the quantum limit in two-dimensional semiconductor contactsLi et al. · Nature 613, 274 (2023)cited by 620doi:10.1038/s41586-022-05431-4