Molecular beam epitaxy (MBE) is used to produce layers of crystalline material on a crystalline substrate, and the technique is employed in industry to produce advanced opto-electronic devices. Both as a UHV-based technique and as a method caable of controlling crystal growth down to the level of single atomic layers, there is clearly a close connection between MBE and surface science. Two MBE chambers with auxiliary surface science analysis techniques are available in the Warwick group. The first is dedicated to antimonide materials (e.g. the semiconductors InSb and GaSb and the ferromagnetic metal MnSb) while the other is used to grow arsenides (GaAs, InAs) and germanium. More details on MBE can be found here.
In many senses, MBE is the prototypical crystal growth technique: the simple picture of atoms landing on a crystal surface, migrating around it until they bond, tehn being covered by new layers of atoms is remarkably close to the actual mechanism of growth. Crucially, all of the interactions in MBE occur on the surface since under UHV conditions, the incident flux of material consists solely of non-interacting molecules or atoms. This is not the case in many other epitaxial growth techniques which operate at higher gas pressures or with liquid precursors (so flow, viscocity and gas-phase interactions become important) or use non-elemental precursors.
The group's main interests in MBE relate to heteroepitaxy - where the material deposited is different to the substrate crystal. We employ III-V semiconductor wafers as substrates, usually sourced from Wafer Technology Ltd (UK). While rules and recipes for heteroepitaxial combinations of many III-V materials are well established (e.g. AlAs-GaAs, InGaAs-InP, AlSb-GaSb) these are all very similar materials chosen in combinations with nearly identical lattice constants. Under such circumstances very little strain energy builds up in one material forced by the epitaxial relationship to adopt the in-plane lattice constant of the substrate. For larger lattice mismathces, strain energy builds up rapidly leading to the relief of strain by various mechanisms. This can be exploited to produce arrays of three-dimensional nano-islands by self-organised growth. Growth of these islands, especially InAs islands on GaAs substrates, is a major theme of our research. Even greater mismatch occurs when the crystal structures of the substrate and growth layer are different - this extreme epitaxy is the second major theme, especially as it relates to ferromagnet-semiconductor heteroepitaxy.
Scanning electron micrograph of InAs nano-islands produced by self-organised MBE growth.