Group IV–V layered semiconductors

SiP, SiAs, GeP, GeAs

Also called GeAs, GeP, SiAs, SiP

van der Waals crystal semiconductor

Anisotropic 2D semiconductors that are not black phosphorus. Their layers have two inequivalent in-plane directions, so conduction, absorption and Raman response all depend on which way the crystal is turned – useful for polarisation-sensitive detectors – and unlike black phosphorus they are stable enough to handle. GeP photodetectors work into the short-wave infrared and have been built directly onto silicon photonics.

Crystal structure

  • P
  • Si
Cell
Rectangular, a = 3.53 Å, b = 20.56 Å
Atoms per cell
24
Si–P bonds
2.26–2.30 Å
Si–Si bond
2.34 Å
Height
4.69 Å between the outer atom centres
Each silicon bonds to three phosphorus atoms and to one other silicon, so Si–Si pairs are built into the layer, and each phosphorus bonds to three silicon. The result is a corrugated sheet with a long repeat in one direction and a short one in the other, which is why its optical and electrical properties depend strongly on direction. SiP is the orthorhombic member of the family; GeP, GeAs and SiAs build the same kind of layer and stack it monoclinically. One layer of bulk SiP (Li and colleagues, CrystEngComm, 2017; COD 7229436): a = 3.53 Å, b = 20.56 Å.

Key properties

  • Strong in-plane anisotropy in both optics and transport, measurable by angle-resolved Raman and reflectance-difference microscopy
  • Micaceous crystals that exfoliate into flakes, with millimetre-sized crystals of GeP, GeAs and SiAs available
  • Few-layer GeAs transistors show clearly direction-dependent transport
  • Multilayer GeP phototransistors respond from the ultraviolet into the short-wave infrared, with ~25 A W−1 at 1,310 nm
  • Resistivity follows two-dimensional variable-range hopping, and the compounds keep their structure down to 5 K

How it is made

  • High-pressure melt growth (0.5–1 GPa) in a cubic anvil press, which gives large shiny crystals of GeP, GeAs and SiAs; SiP comes out small and brittle
  • Chemical vapour transport with iodine from sintered precursors
  • Mechanical exfoliation into flakes, and transfer onto silicon waveguides for photonic devices

Uses, and how close they are

  • Polarisation-sensitive and short-wave infrared photodetectorslab
  • On-chip photodetectors integrated with silicon photonicslab
  • Anisotropic thermoelectric and thermal-management studieslab

Readiness runs lab → prototype → pilot → deployed.

Open problems

  1. How large is the intrinsic mobility, and how much of the measured transport is hopping between defects rather than band conduction?
  2. Can crystals be grown without the high-pressure step, in sizes and quality that support device work?
  3. Do monolayers behave as the calculations predict, and can they be isolated reliably?
  4. Which member of the family is the best infrared detector – the comparison has not been made under common conditions

Going deeper

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

For theoreticians · your lens

Low-symmetry layers with two inequivalent in-plane axes: effective masses, dielectric response and phonons all become tensors, and comparisons with experiment need the crystal orientation stated. The bonding mixes group-IV–group-IV and group-IV–pnictogen bonds in the same sheet, so band edges are not simply pnictogen-derived. Because measured transport shows variable-range hopping, intrinsic-mobility calculations should be compared with care – the samples are probably disorder-limited.

For experimentalists · your lens

Identify the crystal axes before anything else: angle-resolved polarised Raman and reflectance-difference microscopy both work and are quick. Report the angle with every transport or photoresponse number, since the anisotropy is the point. Growth history matters too – high-pressure melt growth and iodine vapour transport give different defect levels, and the hopping transport that dominates below room temperature is a defect signature.

For engineers · your lens

A practical appeal – air-stable, polarisation-sensitive detection in the near and short-wave infrared, demonstrated on silicon photonics – against an impractical supply chain: the best crystals come from a high-pressure press, and nothing here is grown at wafer scale. Treat as a materials option for specialised detectors rather than a platform.

In the research tracks

Key references

  1. High-pressure melt growth and transport properties of SiP, SiAs, GeP, and GeAs 2D layered semiconductorsBarreteau et al. · Journal of Crystal Growth 443, 75 (2016)cited by 170doi:10.1016/j.jcrysgro.2016.03.019
  2. Highly in-plane optical and electrical anisotropy of 2D germanium arsenideYang et al. · Advanced Functional Materials 28, 1707379 (2018)cited by 174doi:10.1002/adfm.201707379
  3. Multilayer 2D germanium phosphide (GeP) infrared phototransistorDushaq et al. · Optics Express 29, 9419 (2021)cited by 23doi:10.1364/OE.420431