Growth on (001) and vicinal (001) GaAs surfaces in a combined scanning tunneling microscope/molecular-beam epitaxy system

To study molecular-beam epitaxy of III–V surfaces, we have added a small growth chamber to our existing ultrahigh vacuum (UHV) scanning tunneling microscope (STM) system. The growth chamber accommodates four effusion cells with pneumatic shutters, and is equipped with a reflection high-energy electron diffraction (RHEED) system with the ability to measure RHEED oscillations. On completion of growth, the specimen is removed from the growth chamber and transferred under UHV into the analytical chamber for study by STM. We have compared a flat GaAs(001) surface grown under arsenic rich conditions with previously studied GaAs(001) surfaces that were grown in a separate system and arsenic capped. The (2×4) unit cell structure is the same, however, there are some interesting differences. First, the surface is generally much smoother with flat planes several hundred angstroms across. Second, the surface reconstruction is predominantly c(2×8) and not (2×4). Third, there are a large number of kinks in the missing dimer rows which are regularly spaced along the 2x direction, forming a domain like structure on the surface. We have obtained the first images from a vicinal GaAs(001) surface. GaAs was grown under step edge growth conditions on a substrate misoriented from the (001) plane by 2° towards the (111)A plane. The STM images show the steps to be relatively straight with some variation in the terrace width. The step structure is made up from complete (2×4) unit cells.