Mechanism of the step flow to island growth transition during MBE on Si(001)-2 × 1

• Published in: Surface science Vol. 331-333; pp. 473 - 478
• Main Authors: van Wingerden, J., Wiechers, Y.A., Scholte, P.M.L.O., Tuinstra, F.
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Lausanne Elsevier B.V 01.07.1995; Amsterdam Elsevier Science; New York, NY
• Subjects: Adatoms; Cross-disciplinary physics: materials science; rheology; Diffusion and migration; Epitaxy; Exact sciences and technology; Growth; Materials science; Methods of deposition of films and coatings; film growth and epitaxy; Models of surface kinetics; Molecular beam epitaxy; Molecular, atomic, ion, and chemical beam epitaxy; Physics; Silicon; Single crystal epitaxy; Single crystal surfaces; Surface diffusion; Diffusion and migration; Growth; Single crystal epitaxy; Epitaxy; Molecular beam epitaxy; Adatoms; Models of surface kinetics; Silicon; Single crystal surfaces; Surface diffusion; Crystal growth; Homoepitaxy; Semiconductor materials; Theoretical study; Island structure
• ISSN: 0039-6028; 1879-2758
• DOI: 10.1016/0039-6028(95)00345-2

During homoepitaxial growth of Si on Si(001) steps act as sinks for the diffusing species on the surface. The ratio between step distance and diffusion length determines, whether growth proceeds by incorporation of the diffusing species into steps, step flow, or by nucleation and growth of islands in between the steps, island growth. For Si(001)-2 × 1 adatom diffusion is extremely anisotropic. However, for many growth conditions incorporating the diffusion perpendicular to the easy diffusion direction is essential in calculating the critical values for the step flow to island growth transition. It will be shown that for usual experimental growth conditions a transition from 1D to 2D diffusion occurs for increasing terrace width, w. As a consequence, the critical flux at marginal step flow scales as w−3 for small terraces and as w−4 for large terraces. Furthermore, it has been investigated, whether a stronger anisotropy for dimer than for adatom diffusion may cause any significant contribution of the dimer diffusion in sustaining step flow.