In 1980 Kroemer, Polasko, and Wright proposed that the (110) orientation should be the preferred orientation for molecular beam epitaxy (MBE) of zincblende compounds (GaAs, GaP) on diamond-structure substrates (Ge, Si), arguing that (a) non-(110) heterovalent interfaces must reconstruct and, reconstructing incompletely, leave hard-to-control interface charges, and (b) the bond-rotation reconstruction of the diamond (110) surface should template the overgrowth so that no antiphase domain boundaries form. The issues that remain unresolved are: quantifying residual interface fields at incompletely reconstructed non-(110) interfaces; the size and controllability of reconstruction domains on clean Si(110)/Ge(110) and whether large single-domain surfaces can be prepared; whether the simple Harrison bond rotation survives within the complex reconstructions actually observed by LEED; and the surface structure of Si(110) and Ge(110) at growth temperature under As and P beams. Later work resolved much of the surface science: first-principles structural models now exist for Si(110)-(16x2), the atomic geometry of the Ge(110)-c(8x10) reconstruction has been determined, micrometer-wide single-domain Si(110)-16x2 surfaces have been fabricated, the As-terminated Si(110) surface has been determined, and antiphase disorder was instead eliminated on (001) substrates. However, the group-V-covered Ge(110) surface and P-covered Si(110) remain undetermined, while the central claim of the 1980 proposal - that (110) is the preferred orientation for zincblende-on-diamond growth free of antiphase domain boundaries and of uncompensated heterovalent interface charge - is still to be established.

Background

GaAs and GaP crystallize in the zincblende structure: two interpenetrating face-centered-cubic sublattices, one occupied by the group-III (or group-II) atom and one by the group-V atom, with each atom tetrahedrally bonded to unlike neighbors. Si and Ge crystallize in the diamond structure, which is the same lattice with both sublattices occupied by the same element. Growing a compound (zincblende) semiconductor on an elemental (diamond) substrate is attractive for devices - GaAs on Ge is nearly lattice matched, and GaP on Si permits combining III-V optoelectronics with Si technology - but it creates two intrinsic problems. First, the interface is heterovalent (valence mismatched): Harrison, Kraut, Waldrop, and Grant showed in 1978 that a mathematically planar Ge/GaAs(001) interface contains Ge-As bonds that each carry a fractional donor-like charge and Ge-Ga bonds that each carry a compensating acceptor-like charge; if the two do not balance locally, a huge electric field (of order 4×1074\times10^{7} V/cm for a fully uncompensated Ge/GaAs(001) interface) forces atomic rearrangements during growth, which stop short of completion and leave locally fluctuating residual charges. Second, the compound can nucleate with either of two sublattice allocations (Ga on sublattice A and As on sublattice B, or vice versa); regions of different allocation meet at antiphase domain boundaries, planar defects lined with wrong (Ga-Ga or As-As) bonds. Molecular beam epitaxy (MBE) - ultra-high-vacuum deposition of elemental beams (Ga, As, P) onto a heated substrate, monitored in situ by reflection high-energy electron diffraction (RHEED) - was the emerging technique expected to give the atomic-scale control needed. Kroemer, Polasko, and Wright (Applied Physics Letters 36, 763 (1980)) argued that the (110) orientation is special: it is the nonpolar plane of the zincblende lattice, containing equal numbers of the two atomic species, so an abrupt (110) interface is charge neutral without reconstruction; and the reconstruction of the clean diamond (110) surface, described by the bond-rotation model of Harrison (Surface Science 55, 1 (1976)) in which surface atoms rotate about their remaining bonds, makes the two sublattice sites inequivalent and should therefore template a unique sublattice allocation, suppressing antiphase domains. The proposal rested on then-current surface science: low-energy electron diffraction (LEED) studies by Olshanetsky, Repinsky, and Shklyaev had shown structural phases and phase transitions on clean Ge(110) and Si(110) that were already more complex than the simple bond-rotation picture, and domain formation on those surfaces could itself seed antiphase boundaries. Establishing the proposal raised issues that had to be settled before the (110) preference could be regarded as established. Of those issues, the single-domain question is only partly resolved: no single-domain preparation for Ge(110) has been shown, and the electromigration-based single-domain preparation for Si(110) has not been tested under subsequent group-V exposure and growth.

Problem Statement

Establish, on complete physical evidence, whether the (110) orientation is the preferred orientation for heteroepitaxial growth of zincblende compounds (GaAs, GaP) on diamond-structure substrates (Ge, Si) that is free of antiphase domain boundaries and of uncompensated heterovalent interface charge, addressing the following open issues: (1) the magnitude of the residual interface fields left behind by incomplete reconstruction at non-(110) heterovalent zincblende-on-diamond interfaces, estimated quantitatively for the real, incompletely reconstructed interfaces of the original GaAs/Ge and GaP/Si class rather than only for idealized models; (2) the size of the surface-reconstruction domains on clean Si(110) and Ge(110), whether reconstruction-domain boundaries seed antiphase boundaries in the overgrown zincblende layer, and whether large single-domain surfaces can be reliably generated by annealing, off-cut steps, or other surface treatments; (3) the atomic geometry of the clean reconstructed Si(110) and Ge(110) surfaces and whether the simple Harrison bond-rotation geometry persists within the complex reconstructions actually observed, reported as the clean-surface reference against which the growth-condition reconstructions of item (4) are compared; and (4) the surface reconstruction of Si(110) and Ge(110) at elevated temperatures under exposure to As and P molecular beams as in MBE pregrowth and growth conditions, for which only the As/Si(110) case has been determined. Items (1)-(4) are each mandatory and jointly in the service of one objective - deciding the (110)-preference question, namely whether, for each defect class the proposal identifies (uncompensated heterovalent interface charge; antiphase domain boundaries), (110) is preferable to the alternative orientations weighed for this system - the (001) and (111) orientations considered in the proposal and the later (211) route. Each item is resolved on the available evidence, or shown on that evidence with reasons to be unresolvable.

References

  1. H. Kroemer, K. J. Polasko, S. C. Wright, 'On the (110) orientation as the preferred orientation for the molecular beam epitaxial growth of GaAs on Ge, GaP on Si, and similar zincblende-on-diamond systems', Applied Physics Letters 36, 763-765 (1980). DOI 10.1063/1.91643, https://doi.org/10.1063/1.91643
  2. W. A. Harrison, 'Surface reconstruction on semiconductors', Surface Science 55, 1-19 (1976). DOI 10.1016/0039-6028(76)90372-1, https://doi.org/10.1016/0039-6028(76)90372-1
  3. W. A. Harrison, E. A. Kraut, J. R. Waldrop, R. W. Grant, 'Polar heterojunction interfaces', Physical Review B 18, 4402-4410 (1978). DOI 10.1103/PhysRevB.18.4402, https://doi.org/10.1103/PhysRevB.18.4402
  4. B. Z. Olshanetsky, S. M. Repinsky, A. A. Shklyaev, 'LEED investigation of germanium surfaces cleaned by sublimation of sulphide films; structural transitions on clean Ge(110) surface', Surface Science 64, 224-236 (1977). DOI 10.1016/0039-6028(77)90268-0, https://doi.org/10.1016/0039-6028(77)90268-0
  5. B. Z. Olshanetsky, A. A. Shklyaev, 'Phase transitions on clean Si(110) surfaces', Surface Science 67, 581-588 (1977). DOI 10.1016/0039-6028(77)90016-4, https://doi.org/10.1016/0039-6028(77)90016-4
  6. R. S. Bauer, J. C. McMenamin, 'Ge-GaAs(110) interface formation', Journal of Vacuum Science and Technology 15, 1444-1449 (1978). DOI 10.1116/1.569804, https://doi.org/10.1116/1.569804
  7. S. L. Wright, M. Inada, H. Kroemer, 'Polar-on-nonpolar epitaxy: Sublattice ordering in the nucleation and growth of GaP on Si(211) surfaces', Journal of Vacuum Science and Technology 21, 534-539 (1982). DOI 10.1116/1.571755, https://doi.org/10.1116/1.571755
  8. H. Kroemer, 'Polar-on-nonpolar epitaxy', Journal of Crystal Growth 81, 193-204 (1987). DOI 10.1016/0022-0248(87)90391-5, https://doi.org/10.1016/0022-0248(87)90391-5
  9. H. Kroemer, 'Sublattice allocation and antiphase domain suppression in polar-on-nonpolar nucleation', Journal of Vacuum Science and Technology B 5, 1150-1154 (1987). DOI 10.1116/1.583702, https://doi.org/10.1116/1.583702
  10. H. Kroemer, 'Nobel Lecture: Quasielectric fields and band offsets: Teaching electrons new tricks', Reviews of Modern Physics 73, 783-793 (2001). DOI 10.1103/RevModPhys.73.783, https://doi.org/10.1103/RevModPhys.73.783
  11. R. C. Pond, J. P. Gowers, D. B. Holt, B. A. Joyce, J. H. Neave, P. K. Larsen, 'A General Treatment of Antiphase Domain Formation and Identification at Polar-Nonpolar Semiconductor Interfaces', MRS Proceedings 25, 273 (1983). DOI 10.1557/PROC-25-273, https://doi.org/10.1557/PROC-25-273
  12. R. C. Pond, J. P. Gowers, B. A. Joyce, 'Surface structure and the origin of antisite domains in GaAs:Ge epitaxial films', Surface Science Letters 152-153, A169 (1985). DOI 10.1016/0167-2584(85)90209-9, https://doi.org/10.1016/0167-2584(85)90209-9
  13. Y. Yamamoto, S. Ino, T. Ichikawa, 'Surface Reconstruction on a Clean Si(110) Surface Observed by RHEED', Japanese Journal of Applied Physics 25, L331 (1986). DOI 10.1143/jjap.25.l331, https://doi.org/10.1143/jjap.25.l331
  14. H. Noro, T. Ichikawa, 'RHEED Study of Surface Reconstruction at Clean Ge(110) Surface', Japanese Journal of Applied Physics 24, 1288 (1985). DOI 10.1143/jjap.24.1288, https://doi.org/10.1143/jjap.24.1288
  15. D. K. Biegelsen, R. D. Bringans, J. E. Northrup, M. C. Schabel, L.-E. Swartz, 'Arsenic termination of the Si(110) surface', Physical Review B 47, 9589-9596 (1993). DOI 10.1103/PhysRevB.47.9589, https://doi.org/10.1103/PhysRevB.47.9589
  16. M. Menon, N. Lathiotakis, A. N. Andriotis, 'The reconstruction of the Si(110) surface and its interaction with Si adatoms', Physical Review B 56, 1412-1415 (1997). DOI 10.1103/PhysRevB.56.1412, https://doi.org/10.1103/PhysRevB.56.1412
  17. R. Butz, H. Lüth, 'The structure of clean and SiGe-covered Si(110) surfaces', Surface Science 365, 807-816 (1996). DOI 10.1016/0039-6028(96)00709-1, https://doi.org/10.1016/0039-6028(96)00709-1
  18. T. Ichikawa, 'Atomic geometry of the Ge(110)c(8x10) structure', Surface Science 560, 205-212 (2004). DOI 10.1016/j.susc.2004.04.040, https://doi.org/10.1016/j.susc.2004.04.040
  19. Y. Yamada, A. Girard, H. Asaoka, H. Yamamoto, S. Shamoto, 'Single-domain Si(110)-16x2 surface fabricated by electromigration', Physical Review B 76, 153309 (2007). DOI 10.1103/PhysRevB.76.153309, https://doi.org/10.1103/PhysRevB.76.153309
  20. Y. Yamada, A. Girard, H. Asaoka, H. Yamamoto, S. Shamoto, 'Controlling the surface chirality of Si(110)', Physical Review B 77, 153305 (2008). DOI 10.1103/PhysRevB.77.153305, https://doi.org/10.1103/PhysRevB.77.153305
  21. T. Yamasaki, K. Kato, T. Uda, T. Yamamoto, T. Ohno, 'First-principles theory of Si(110)-(16 x 2) surface reconstruction for unveiling origin of pentagonal scanning tunneling microscopy images', Applied Physics Express 9, 035501 (2016). DOI 10.7567/apex.9.035501, https://doi.org/10.7567/apex.9.035501
  22. H. Aoyama, T. Abukawa, 'Complete three-dimensional structure of Bi-adsorbed Si(110) surface: Discovery of heavily reconstructed Si(110) substrate', Physical Review Research 3, 043164 (2021). DOI 10.1103/PhysRevResearch.3.043164, https://doi.org/10.1103/PhysRevResearch.3.043164
  23. M. López, T. Ikei, Y. Takano, K. Pak, H. Yonezu, 'Initial Growth Mechanism of GaAs on Si(110)', Japanese Journal of Applied Physics 29, 551 (1990). DOI 10.1143/jjap.29.551, https://doi.org/10.1143/jjap.29.551
  24. T. Yodo, M. Tamura, M. López, Y. Kajikawa, 'GaAs heteroepitaxial growth on vicinal Si(110) substrates by molecular beam epitaxy', Journal of Applied Physics 76, 7630-7632 (1994). DOI 10.1063/1.357932, https://doi.org/10.1063/1.357932
  25. T. Yodo, 'Characterization of GaAs Films Grown on Vicinal Si(110) Substrates by Molecular-Beam Epitaxy', Japanese Journal of Applied Physics 34, 4631 (1995). DOI 10.1143/jjap.34.4631, https://doi.org/10.1143/jjap.34.4631
  26. M. Meléndez-Lira, M. A. Santana-Aranda, M. López-López, M. Tamura, T. Yodo, M. A. Vidal, 'Effects of the substrate tilting angle on the molecular beam epitaxial growth of GaAs on Si(110)', Journal of Vacuum Science and Technology B 19, 1567-1571 (2001). DOI 10.1116/1.1376387, https://doi.org/10.1116/1.1376387
  27. H. Usui, K. Ishiji, H. Yasuda, H. Mori, 'Growth process and nanostructure of crystalline GaAs on Si(1 1 0) surface prepared by molecular beam epitaxy', Journal of Crystal Growth 289, 76-80 (2006). DOI 10.1016/j.jcrysgro.2005.11.097, https://doi.org/10.1016/j.jcrysgro.2005.11.097
  28. S. M. Ting, E. A. Fitzgerald, 'Metal-organic chemical vapor deposition of single domain GaAs on Ge/GexSi1-x/Si and Ge substrates', Journal of Applied Physics 87, 2618-2628 (2000). DOI 10.1063/1.372227, https://doi.org/10.1063/1.372227
  29. B. Kunert, I. Németh, S. Reinhard, K. Volz, W. Stolz, 'Si (001) surface preparation for the antiphase domain free heteroepitaxial growth of GaP on Si substrate', Thin Solid Films 517, 140-143 (2008). DOI 10.1016/j.tsf.2008.08.077, https://doi.org/10.1016/j.tsf.2008.08.077
  30. T. J. Grassman, M. R. Brenner, S. Rajagopalan, R. Unocic, R. Dehoff, M. Mills, H. Fraser, S. A. Ringel, 'Control and elimination of nucleation-related defects in GaP/Si(001) heteroepitaxy', Applied Physics Letters 94, 232106 (2009). DOI 10.1063/1.3154548, https://doi.org/10.1063/1.3154548
  31. O. Romanyuk, O. Supplie, T. Susi, M. M. May, T. Hannappel, 'Ab initio density functional theory study on the atomic and electronic structure of GaP/Si(001) heterointerfaces', Physical Review B 94, 155309 (2016). DOI 10.1103/PhysRevB.94.155309, https://doi.org/10.1103/PhysRevB.94.155309
  32. P. Farin, H. Eisele, M. Dähne, A. Lenz, 'From surface data to bulk properties: a case study for antiphase boundaries in GaP on Si(001)', Journal of Physics D: Applied Physics 54, 205302 (2021). DOI 10.1088/1361-6463/abdff1, https://doi.org/10.1088/1361-6463/abdff1
  33. D. Gupta, S. Pallikkara Chandrasekharan, S. Thebaud, C. Cornet, L. Pedesseau, 'Stability of monodomain III-V crystals and antiphase boundaries over a Si monoatomic step', Applied Surface Science 678, 161076 (2024). DOI 10.1016/j.apsusc.2024.161076, https://doi.org/10.1016/j.apsusc.2024.161076
  34. O. Rubel, S. D. Baranovskii, 'Formation Energies of Antiphase Boundaries in GaAs and GaP: An ab Initio Study', International Journal of Molecular Sciences 10, 5104-5114 (2009). DOI 10.3390/ijms10125104, https://doi.org/10.3390/ijms10125104
  35. D. E. Aspnes, J. Ihm, 'Biatomic Steps on (001) Silicon Surfaces', Physical Review Letters 57, 3054-3057 (1986). DOI 10.1103/PhysRevLett.57.3054, https://doi.org/10.1103/PhysRevLett.57.3054
  36. E. S. Tok, J. H. Neave, M. J. Ashwin, B. A. Joyce, T. S. Jones, 'Growth of Si-doped GaAs(110) thin films by molecular beam epitaxy; Si site occupation and the role of arsenic', Journal of Applied Physics 83, 4160-4167 (1998). DOI 10.1063/1.367169, https://doi.org/10.1063/1.367169