Skip to main content
v2026.11,986 entries · CC-BY 4.0

Molecular-Beam Epitaxy (MBE): Principles, Parts and Lab Setup

What molecular-beam epitaxy (MBE) is, how its sources, shutters and RHEED work, and how five published university MBE systems compare.

Written and maintained by CASRAI Editorial Board

Published

On this page

Molecular-beam epitaxy (MBE), also written molecular beam epitaxy, is a thin-film deposition method in which one or more beams of atoms or molecules travel through ultra-high vacuum and condense on a heated crystalline substrate, so that the new layer grows in registry with the crystal beneath it. Because the beams are produced from separately heated sources and switched on and off with shutters, growth can be controlled down to a single atomic layer. Stanford’s ARPES group describes MBE as an advanced thin-film deposition system used in research and industry for over 40 years.

Definition and scope

The word epitaxy refers to growth of a crystalline film on a crystalline substrate so that the film follows the substrate’s crystal structure. Scienta Omicron, a manufacturer of MBE systems, summarises the method this way: epitaxial, layered structures are grown under ultrahigh-vacuum conditions on a crystalline substrate, where one or more molecular beams hit the surface and solidify. The two spellings, molecular-beam epitaxy and molecular beam epitaxy, name the same technique; the 1975 review by Cho and Arthur, titled “Molecular beam epitaxy” (Progress in Solid State Chemistry, volume 10, pages 157-191, DOI 10.1016/0079-6786(75)90005-9).

MBE is a physical deposition method: the film is built from beams of the elements themselves (or from gas sources in some variants), not from a chemical reaction of precursors in a flowing carrier gas. Wikipedia’s overview notes that the absence of carrier gases together with the ultra-high-vacuum environment gives the highest achievable film purity, and that the word “beam” reflects the fact that atoms do not interact with each other or with chamber gases before reaching the wafer, because mean free paths are long.

In scope for this page: what the technique is, what the main parts of an MBE system do, how published laboratory systems are configured, which variants exist, and what MBE means for laboratory planning. Out of scope: process recipes, source-handling safety procedures and equipment purchasing advice, none of which are given here.

How MBE works in practice

Sources and beams

Ultra-pure elements such as gallium and arsenic are heated in quasi-Knudsen effusion cells or in electron-beam evaporators until they sublime (Wikipedia, citing the primary literature). The vapour condenses on the substrate, where the species react; for gallium and arsenic the product is single-crystal gallium arsenide. The source temperature sets the flux that reaches the substrate, and the substrate temperature affects how atoms hop across the surface and whether they desorb.

Omicron notes that effusion cells are typically mounted at the bottom of the vacuum vessel with the substrate usually mounted face down, and that the substrate holder allows in-situ heating and rotation. Stanford’s chalcogenide system illustrates how many sources a research system can carry: it holds up to eight elements, using four effusion cells plus a four-pocket electron-beam evaporator, and includes a cracker effusion cell used for topological-insulator films. Its oxide system uses an ozone gas source for cuprates. The University of Groningen’s oxide MBE group likewise describes oxidising gas (oxygen, ozone or NO2) supplied through a tube directed at the sample.

Shutters and layer control

A computer controls a shutter in front of each source, which is what allows thickness control down to a single atomic layer (Wikipedia). Stanford’s group reports that shutter control combined with stable fluxes gives roughly one atomic layer every few dozen seconds in its systems.

Vacuum

Growth takes place in high or ultra-high vacuum, given in Wikipedia’s overview as 10-8 to 10-12 Torr. Stanford keeps its MBE and ARPES chambers in the low 10-11 Torr range so films can be studied in situ without leaving vacuum. Cryopumps and cryopanels cooled with liquid nitrogen (close to 77 K) trap impurities on cold surfaces. Wikipedia also states that the typical deposition rate is below 3,000 nm per hour, which is about 0.83 nm per second, and that higher rates would require proportionally better vacuum to keep the same impurity levels.

Watching growth with RHEED

Reflection high-energy electron diffraction (RHEED) is the standard in-situ monitor. Omicron calls it an established in-situ method to check film quality. Stanford adds the useful detail that RHEED intensity oscillations indicate layer-by-layer growth, and that for SrTiO3 one oscillation period corresponds to one unit cell. Groningen makes the same point: thin-film growth can be controlled with atomic precision by monitoring the RHEED intensity oscillations. The technique itself dates to the late 1960s, when Alfred Y. Cho used RHEED for in-situ observation (Wikipedia).

What each part of an MBE system does

The table below is a function cross-walk compiled from the facility and reference pages listed under Sources. Each row names the part, what it does, and where the statement comes from.

Part Function as described in the sources Source
Effusion (Knudsen) cell Heated cell that sublimes or evaporates an element to form a molecular beam aimed at the substrate Omicron; Wikipedia; Groningen
Electron-beam evaporator Bombards source material with high-energy electrons; several pockets in some systems (Stanford: four-pocket; Ohio State: three-pocket) Groningen; Stanford; Ohio State
Shutter Computer-controlled blocker in front of each source that starts and stops a beam, giving single-atomic-layer thickness control Wikipedia; Stanford
Cryopump / cryopanel Pumps or cold surfaces near 77 K that trap impurities and maintain ultra-high vacuum Wikipedia; Texas
Load lock Separate entry chamber, pumped independently, in which samples can be outgassed (Texas: halogen lamps to 150 °C) before moving to the main chamber Texas; Ohio State; Emory
Substrate heater and holder Heats and rotates the substrate during growth; heater limits differ by system (Texas: up to 1000 °C; Emory: 1500 °C) Omicron; Texas; Emory
RHEED gun and screen Electron-diffraction monitor of surface structure; oscillations track layer-by-layer growth Stanford; Omicron; Groningen; Texas
Gas supply (oxidant, cracker) Oxygen, ozone or NO2 for oxide films; Stanford also uses a cracker effusion cell for topological-insulator films Groningen; Stanford

How published laboratory systems are configured

MBE systems are built around the chemistry of the films they grow, so no two published systems are identical. The data table below this page compares five university systems on base pressure, number of sources, source types, in-situ monitoring and sample transfer, using each facility’s own description. A few patterns stand out from it:

  • Base pressure is reported in the 10-10 to 10-11 Torr range. The University of Texas states 1.5×10-10 Torr and Emory states less than 3×10-10 Torr, while Stanford keeps its chambers in the low 10-11 Torr range. All sit inside the 10-8 to 10-12 Torr envelope given for MBE generally.
  • Source counts run from four to nine cells. Ohio State lists nine shuttered effusion cells plus a three-pocket electron-beam evaporator; Texas and Emory list six effusion cells each.
  • Sample transfer is multi-chamber. Groningen separates growth, preparation and analysis into three UHV chambers (analysis uses LEED and XPS); Ohio State describes a load-locked entry/exit chamber; Texas pumps its load lock separately with a turbopump.
  • RHEED is universal. Every system in the table lists RHEED as the in-situ growth monitor.

Growth-rate cross-check

Published rate statements look inconsistent until the basis of each is clear. Wikipedia gives an upper bound of below 3,000 nm per hour for typical deposition, which is a rate in length units. Stanford instead describes about one atomic layer every few dozen seconds for its films, a rate in layers. The two converge only after a material’s layer spacing is known, which is why a monolayer-per-second figure from one paper and a nanometres-per-hour figure from a reference page should not be compared directly. The December 2025 arXiv paper by Stolyarov and colleagues on GaAs-based microcavities (arXiv:2512.03203) is a good example of why growth parameters are specific to the structure: it reports MBE of high-Q planar microcavities in which the AlGaAs layers of the Bragg mirrors are replaced by short-period GaAs/AlAs superlattices, and obtains a polariton-lasing threshold of about 570 W/cm2 and a quality factor of about 5.4 x 104.

Solid-source MBE

The classic arrangement described above, with elemental solid sources in effusion cells or electron-beam evaporators.

Gas-source MBE

Wikipedia describes gas-source MBE as resembling chemical vapour deposition, with gaseous precursors supplying part of the beam. Stanford’s oxide and chalcogenide systems already mix solid sources with ozone or cracker cells.

Oxide MBE

Oxygen sources, including oxygen gas, radical or ozone sources, allow oxide deposition (Wikipedia). Stanford grows SrTiO3, LaTiO3, rutile TiO2 and cuprates this way, and Ohio State describes a Veeco GEN 930 oxide MBE system.

Organic MBE

Organic semiconductor MBE evaporates molecules rather than atoms (Wikipedia).

MBE is one of several ways to make thin films. For a mass-sensing technique used in thin-film work, see our guide to the quartz crystal microbalance, and for how shared materials-growth facilities are organised as user platforms, see PARADIM, the NSF Materials Innovation Platform for interface materials.

Where it matters for labs and facility planning

The published facility descriptions point to several practical consequences for anyone planning access to, or work in, an MBE laboratory.

  • Vacuum integrity is the core constraint. Since the achievable purity depends on ultra-high vacuum and on trapping impurities with cold surfaces, systems separate sample entry (load lock) from growth chambers. Texas, for instance, outgasses samples in the load lock before transfer to the main line.
  • Source count defines what can be grown. Stanford’s chalcogenide system, with up to eight elements, carries more elements in one system than the six-cell systems at Texas and Emory.
  • Growth and analysis can be coupled. Stanford’s MBE and ARPES chambers share ultra-high vacuum so films are studied in situ; Groningen couples growth with LEED and XPS chambers. For a research office, this means an instrument may be a growth-plus-analysis cluster rather than a single tool.
  • Growth rate sets vacuum demand. Wikipedia notes that higher deposition rates require proportionally better vacuum for the same impurity levels, so throughput and vacuum quality are linked rather than independent.

Common misunderstandings

“MBE only makes compound semiconductors.”

The method began with GaAs (Davey and Pankey, 1968, growing GaAs epitaxial films on single-crystal GaAs), but Stanford grows Bi2Te3, Bi2Se3, monolayer FeSe, oxides and 2D materials, and organic MBE exists.

“The 1958 work was the first MBE.”

K. G. Gunther originated the basic ideas in 1958, but his films were deposited on glass and so were not epitaxial. Davey and Pankey’s 1968 GaAs-on-GaAs films are the first demonstration cited by Wikipedia, with J. R. Arthur’s growth-kinetics studies enabling later development.

“A single growth rate applies to all MBE.”

Rates depend on material, source flux and vacuum; see the cross-check above.

“Layer control means no monitoring is needed.”

Shutters set when beams start and stop, but facilities still use RHEED to confirm that growth is layer-by-layer.

Frequently asked questions

What does molecular-beam epitaxy mean?

It is a method for growing crystalline thin films on a crystalline substrate by directing molecular or atomic beams at it in ultra-high vacuum. “Epitaxy” means the film follows the substrate’s crystal structure.

Is it molecular-beam epitaxy or molecular beam epitaxy?

Both spellings are used for the same technique. The 1975 review by Cho and Arthur is titled “Molecular beam epitaxy”, while the 2025 arXiv paper cited below writes “molecular-beam epitaxy”.

What vacuum does MBE need?

Wikipedia gives 10-8 to 10-12 Torr. Published university systems report base pressures of 1.5×10-10 Torr (Texas), below 3×10-10 Torr (Emory) and the low 10-11 Torr range (Stanford).

What is RHEED and why is it used?

RHEED is reflection high-energy electron diffraction, an in-situ method that checks film quality during growth; its intensity oscillations indicate layer-by-layer growth.

What materials can be grown?

Per the sources read: GaAs, other compound semiconductors, chalcogenides such as Bi2Se3, oxides such as SrTiO3, 2D materials and organic semiconductors.

Who developed MBE?

Basic ideas date to K. G. Gunther (1958); Davey and Pankey demonstrated epitaxial GaAs films in 1968; J. R. Arthur’s kinetics studies and Alfred Y. Cho’s RHEED work followed, and Cho and Arthur published the 1975 review.

Sources

All pages read on 2026-11-13.

Data: Configuration of five published university MBE systems

Compares base pressure, number of sources, source types, in-situ monitoring and sample-transfer features of molecular-beam epitaxy systems, compiled from each facility's own public description page read on 2026-11-13.

Period covered: 2026.

Configuration of five published university MBE systems
FacilityBase pressure statementSourcesIn-situ monitoringSample transfer and other notes
Stanford ARPES group (chalcogenide and oxide systems)Low 10^-11 Torr range (MBE and ARPES chambers)Up to eight elements: four effusion cells plus a four-pocket e-beam evaporator; cracker cell; ozone source in oxide systemRHEEDMBE connected in vacuum to ARPES for in-situ study
University of Texas at Austin1.5x10^-10 TorrSix effusion cells (four standard, two high-temperature)Differentially pumped RHEEDCryopump; turbopumped load lock with lamp heating to 150 C; sample heater to 1000 C
Emory UniversityBelow 3x10^-10 TorrSix effusion cellsRHEED-based deposition feedbackSputtering source in load lock; 1500 C sample heater
Ohio State University (Brillson group)Not statedNine shuttered effusion cells plus a load-locked three-pocket e-beam evaporator1-30 keV RHEED opticsVeeco GEN 930 oxide MBE system; load-locked entry/exit chamber
University of Groningen (Zernike Institute)Not stated (UHV)Several sources: Knudsen cells or e-beam evaporators; oxidising gas supply for oxidesRHEED intensity oscillationsThree UHV chambers: growth, preparation, analysis (LEED, XPS)

Sources

Compiled by CASRAI from the sources above. Free to reuse under CC BY 4.0 with attribution.

Follow CASRAI

Research-administration guidance, standards updates and independent tool reviews.

Ask CASRAI · full access

Ask about Molecular-Beam Epitaxy (MBE): Principles, Parts and Lab Setup

Full access: large monthly usage, up to 10 source passages behind each answer, and MCP and API access for Claude Code, Codex and other CLIs. Cancel any time.

An AI assistant for research: funder rules, compliance, methods and publishing. It links the sources behind its answers and labels answers that come from a web search.

Answers draw on CASRAI's guides and dictionary plus the federal and funder documents we index: Federal Register, Grants.gov, Regulations.gov and UKRI.

Works on this site and inside Claude, Cursor and the AI tools you already use.

Everything CASRAI publishes — this page, the dictionary, the guides and the news — stays open to read, with no account and no card.

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →