Selective area epitaxy and growth over patterned substrates by chemical beam epitaxy - Electronics Letters

conductivity of the sample and was measured by determining the change in the optical sampling beam polarisation due to a linear electro-optic effect in the cell. The intrinsic time resolution of this experiment is defined by the laser pulse duration. It was however a little worse than 7ps because of the poor impedance matching at the place where the sample was connected into the stripline structure. The investigated samples were prepared from &doped GaAs epitaxial layers as grown by metalorganic chemical vapour deposition on semi-insulating GaAs substrates. Trimethylgalliumetherate and 5% arsine in hydrogen were used as G a and As sources and the dopant source was 0.024% silane in argon. Two d-layers, separated by 0.05pm thick undoped GaAs layer, were grown. The spike doping was achieved by interrupting the growth of undoped layer, purging the reactor with hydrogen, then introducing silane into the reactor for a short time, purging the reactor again, and resuming the growth. The concentrations of 2D electrons in d-layers from 8 x 10”cm-’ to 2 x 10 i2cm-2 were measured by the Hall-effect technique. Their distribution across the thickness of an epitaxial layer was determined by the CjV method on a layer with a single &doped region which was grown at the same conditions. The profile widths of IO nm were obtained. The measured low-field Hall mobilities were 2-2.5 x lo3 cm2/ V s at 300 K. All measurements were performed at room temperature.