Bismuth oxyselenide

Bi₂O₂Se

Also called bismuth oxychalcogenide

nanosheet semiconductor

The high-mobility 2D semiconductor that is not a van der Waals crystal. Its layers are charged and hold on to each other electrostatically, with no van der Waals gap to peel apart, so it is grown rather than exfoliated – and it grows as a single crystal across a whole wafer. In return it offers electron mobilities far above those of the sulfide and selenide monolayers, survives in air, and oxidises on purpose into Bi₂SeO₅, a high-κ dielectric that can gate the layer underneath it.

Cell: Square, a = 3.89 Å
Atoms per cell: 5
Bi–O bond: 2.32 Å
Bi–Se bond: 3.29 Å
Height: 4.31 Å between the outer atom centres

A [Bi₂O₂]²⁺ sheet – a square net of oxygen with bismuth above and below it – and a square sheet of Se²⁻ on top. Unlike a van der Waals layer, the selenium sheet bonds to bismuth on both sides, so in the crystal each Se sheet is shared between two [Bi₂O₂] sheets and there is no gap to peel apart. The model shows one repeat, the ~0.6 nm usually called a monolayer; the bismuth on the bottom face is missing the selenium that sits below it in the crystal. One repeat of bulk Bi₂O₂Se (Boller, Monatshefte für Chemie 104, 916, 1973; COD 1541953): a = 3.89 Å, c = 12.21 Å. The COD file places oxygen at z = 0, inside the selenium plane; the model puts it on the site (0, ½, ¼) established for this structure type, where Bi–O comes out at 2.32 Å.

Data sheet

Show all 9 entriesHide the data sheet

Kind of sheet: nanosheet
Made by chemistry or cut from a solid, not peeled as a clean crystal layer
Behaves as: semiconductor
Band gap: ~0.8 eV in bulk, widening as the crystal is thinned

In colour: this family, 0.8 eV. In grey: the other 39 families on the catalogue’s chart, each at the middle of its range.
Thickness: ~0.6 nm per layer
Structure: [Bi₂O₂]²⁺ layers alternating with square [Se]²⁻ layers, held together by electrostatic attraction rather than van der Waals forces (tetragonal, I4/mmm)
Space group: Bi₂O₂Se: bulk I4/mmm
Bulk: the Materials Project crystal linked below
Stability: Stable in air, which is why mobility can be measured on bare, unencapsulated flakes. The reactive direction is oxidation: the surface converts to Bi₂SeO₅, useful when intended and a source of drift when not.
First made: Known as a bulk oxychalcogenide long before 2D work; atomically thin crystals grown by CVD and measured in 2017 (Wu and colleagues)
Materials Project: Bi₂O₂Se
Computed crystal structure and properties

Compare with other materials

Key properties

Hall mobility above 20,000 cm² V⁻¹ s⁻¹ at 2 K in atomically thin crystals, with quantum oscillations in samples that need no encapsulation
Air-stable, unlike black phosphorus, so devices can be made and measured without a glovebox
Wafer-scale single-crystal films grow on perovskite oxide substrates, giving ~150 cm² V⁻¹ s⁻¹ at room temperature with on/off ratios above 10⁵
Its own oxide, Bi₂SeO₅, forms a native high-κ gate dielectric on top of the channel
The band gap widens as the crystal thins, by quantum confinement rather than a change of stacking
No van der Waals gap: ultrasonication breaks crystals into nanosheets, but they do not peel cleanly like graphite

How it is made

Chemical vapour deposition on mica, giving atomically thin single crystals down to a bilayer and up to ~200 μm across
Wafer-scale growth of single-crystal films on SrTiO₃, LaAlO₃ and similar perovskite oxides, where lattice matching aligns every seed the same way
Controlled surface oxidation to convert the top of a flake into Bi₂SeO₅

Uses, and how close they are

High-mobility transistor channels for 2D electronicslab
Native high-κ gate dielectrics grown from the channel itselflab
Infrared photodetectorslab
Thermoelectricslab

Readiness runs lab → prototype → pilot → deployed.

Open problems

Without a van der Waals gap, can Bi₂O₂Se be thinned reliably to a chosen number of layers, or does everything depend on growth?
Can contacts and doping be controlled well enough to turn high Hall mobility into short-channel transistor performance?
How reliable is the native Bi₂SeO₅ dielectric – leakage, breakdown, interface traps – compared with deposited high-κ oxides?
What sets the electron density in as-grown crystals, and can selenium vacancies be tuned or compensated?

Going deeper

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

For theoreticians · your lens

Electrostatically bonded layers put Bi₂O₂Se outside the usual van der Waals picture: interlayer coupling is strong, so the Madelung energy matters more than dispersion corrections, and the gap changes with thickness through quantum confinement rather than through stacking. The conduction band is light and free-electron-like, which is where the high mobility comes from, and selenium vacancies act as donors, so crystals are n-type without deliberate doping. The oxidised phase Bi₂SeO₅ keeps the same [Bi₂O₂]²⁺ framework, which is why the interface between the two can be sharp.

For experimentalists · your lens

Growth, not exfoliation, sets the thickness – calibrate optical contrast against AFM for your substrate, because these are not the familiar graphite-like steps. Say which mobility you are quoting: Hall mobility on a bare flake at low temperature runs far above the field-effect mobility of a finished transistor at room temperature. Watch for unintended surface oxidation between growth and measurement, and check the phase by Raman, because Bi₂SeO₅ forms easily.

For engineers · your lens

Attractive for the reasons TMDC channels are, plus two of its own: wafer-scale single crystals instead of polycrystalline films, and a native high-κ oxide that removes the hardest step in a 2D transistor stack. Against that, it cannot be picked up and stacked like a van der Waals layer, it is n-type only so far, and wafer-scale mobility (~150 cm² V⁻¹ s⁻¹) is well below the low-temperature record.

In the research tracks

Growth methods

Powder-source CVD of TMDC monolayers

Readiness assessments

Recent news

The newest items tagged Bi₂O₂Se, from the news feed updated 5 Oct 2026.

Preprintnot yet peer reviewed 14 Jun 2026 arXiv

Confined Oxygen-Vacancy Migration Drives Ferroelectric Switching

Conventional ferroelectricity arises from intrinsic lattice distortions, whereas oxygen vacancies are generally regarded as detrimental because their migration induces leakage currents and polarization degradation. However, the recently discovered ultrathin van der Waals ferroelectric Bi2SeO5 exhibits robust…

ExperimentBi₂O₂Se

Preprintnot yet peer reviewed 11 Dec 2025 arXiv

Towards ultra-scaled nanoelectronics using the zipper material system Bi₂O₂Se/Bi₂SeO₅

Two-dimensional (2D) materials could overcome the scaling bottleneck of nanoelectronics by enabling atomically thin channels, superior electrostatic control, and reduced short-channel effects. However, progress is limited by the lack of semiconductor–insulator interfaces being simultaneously scalable, stable, and…

EngineeringExperimentBi₂O₂Se

Preprintnot yet peer reviewed 1 Dec 2025 arXiv

Extraordinary cation-replace-cation antisite defect predominate in Bi2SeO5

As a newly identified single-crystalline van der Waals dielectric with a high dielectric constant, Bi2SeO5 plays a pivotal role in advancing 2D electronic devices. In this work, we systematically investigate the defect properties of Bi2SeO5 using first-principles calculations based on a hybrid functional. Although…

TheoryBi₂O₂Se

All 3 items tagged Bi₂O₂Se in the news feed · RSS feed for Bi₂O₂Se

Key references

High electron mobility and quantum oscillations in non-encapsulated ultrathin semiconducting Bi₂O₂SeWu et al. · Nature Nanotechnology 12, 530 (2017)cited by 755doi:10.1038/nnano.2017.43
Controlled synthesis of high-mobility atomically thin bismuth oxyselenide crystalsWu et al. · Nano Letters 17, 3021 (2017)cited by 313doi:10.1021/acs.nanolett.7b00335
Wafer-scale growth of single-crystal 2D semiconductor on perovskite oxides for high-performance transistorsTan et al. · Nano Letters 19, 2148 (2019)cited by 119doi:10.1021/acs.nanolett.9b00381
A native oxide high-κ gate dielectric for two-dimensional electronicsLi et al. · Nature Electronics 3, 473 (2020)cited by 306doi:10.1038/s41928-020-0444-6