Effect of hydrogenation on B/Si(0 0 1)-(1 × 2)

Ab initio calculations, based on pseudopotentials and density functional theory, have been performed to investigate the effect of hydrogenation on the atomic geometries and the energetics of substitutional boron on the generic Si(0 0 1)-(1 × 2) surface. For a single B atom substitution corresponding to 0.5 ML coverage, we have considered two different sites: (i) the mixed Si-B dimer structure and (ii) boron substituting for the second-layer Si to form Si-B back-bond structure, which is energetically more favorable than the mixed Si-B dimer by 0.1 eV/dimer. However, when both of these cases are passivated by hydrogen atoms, the situation is reversed and the Si-B back-bond case becomes 0.1 eV/dimer higher in energy than the mixed Si-B dimer case. For the B incorporation corresponding to 1 ML coverage, among the substitutional cases, 100% interdiffusion into the third layer of Si and 50% interdiffusion into the second layer of Si are energetically similar and more favorable than the other cases that are considered. However, when the surface is passivated with hydrogen, the B atoms energetically prefer to stay at the third layer of the Si substrate.     

Atomic and electronic structure of Sr/Si(0 0 1)-(2 × 2)

The adsorption of Sr on the Si(0 0 1) surface with the semiantiphase dimer (2 × 2) reconstruction is studied, based upon the ab initio pseudopotential calculations. It is calculated that the semiantiphase dimer (2 × 2) reconstruction (2 dimers per unit cell) is more favorable than the (2 × 1) phase (1 dimer per unit cell) by an energy of about 0.24 eV/dimer. Considering the energetically more stable reconstruction, we have assumed four possible locations for 1/4 monolayer (ML) Sr adsorption on this surface: (i) bridge, (ii) cave, (iii) pedestal, and (iv) valley-bridge. We find that Sr adsorption on the valley-bridge site is energetically more favorable than all other cases studied here. Interestingly, one of the dimers becomes symmetric, but the other one is still asymmetric with the buckling angle reduced from 18° to 14°, when compared with the clean Si(0 0 1)-(2 × 2) surface. The calculated bond length between Sr and Si in the case of valley-bridge adsorption site is 3.05 Å, and in good agreement with other theoretical calculations. We also present and compare the electronic band structures for the clean and covered surfaces as well as the corresponding charge density plots.     

DFT study on dissociative adsorption of SiH4 and GeH4 on SiGe(1 0 0)-2 × 1 surface

SiH₄ and GeH₄ dissociative adsorptions on a buckled SiGe(1 0 0)-2 × 1 surface have been analyzed using density functional theory (DFT) at the B3LYP level. The Ge alloying in the Si(1 0 0)-2 × 1 surface affects the dimer buckling and its surface reactivity. Systematic Ge influences on the reaction energetics are found in SiH₄ and GeH₄ reactions with four dimers of Si^∗-Si, Ge^∗-Si, Ge^∗-Ge, and Si^∗-Ge (∗ denotes the protruded atom). On a half H-covered surface, the energy barriers for silane and germane adsorption are higher than those on the pristine surface. The energy barrier for silane adsorption is higher than the corresponding barrier for germane adsorption. Rate constants are also calculated using the transition-state theory. We conclude that the SiGe surface reactivity in adsorption reaction depends on the Ge presence in dimer form. If the surface Ge is present in form of Ge^∗-Ge, the surface reactivity decreases as the Ge^∗-Ge content increases. If the surface Ge prefers to be in form of Ge^∗-Si at low Ge contents, the surface reactivity increases first, then decreases at high Ge surface contents when Ge^∗-Ge prevails. The calculated rate constant ratio of GeH₄ adsorption on Si^∗-Si over Ge^∗-Ge at 650 °C is 2.1, which agrees with the experimental ratio of GeH₄ adsorption probability on Si(1 0 0) over Si(1 0 0) covered by one monolayer Ge. The experimental ratio is 1.7 measured through supersonic molecular beam techniques. This consistency between calculation and experimental results supports that one monolayer of Ge on Si(1 0 0) exists in form of Ge^∗-Ge dimer.     

First-principles study of adsorption and diffusion on Ge/Si(0 0 1)-(2 × 8) and Ge/Si(1 0 5)-(1 × 2) surfaces

Using first-principles total-energy calculations, we have investigated the adsorption and diffusion of Si and Ge adatoms on Ge/Si(0 0 1)-(2 × 8) and Ge/Si(1 0 5)-(1 × 2) surfaces. The dimer vacancy lines on Ge/Si(0 0 1)-(2 × 8) and the alternate S_A and rebonded S_B steps on Ge/Si(1 0 5)-(1 × 2) are found to strongly influence the adatom kinetics. On Ge/Si(0 0 1)-(2 × 8) surface, the fast diffusion path is found to be along the dimer vacancy line (DVL), reversing the diffusion anisotropy on Si(0 0 1). Also, there exists a repulsion between the adatom and the DVL, which is expected to increase the adatom density and hence island nucleation rate in between the DVLs. On Ge/Si(1 0 5)-(1 × 2) surface, the overall diffusion barrier of Si(Ge) along direction is relative fast with a barrier of ∼0.83(0.61) eV, despite of the large surface undulation. This indicates that the adatoms can rapidly diffuse up and down the (1 0 5)-faceted Ge hut island. The diffusion is also almost isotropic along [0 1 0] and directions.     

Bimodal growth of manganese silicide on Si(1 0 0)

Two different growth modes of manganese silicide are observed on Si(1 0 0) with scanning tunneling microscopy. 1.0 and 1.5 monolayer Mn are deposited at room temperature on the Si(1 0 0)-(2 × 1) substrate. The as-grown Mn film is unstructured. Annealing temperatures between room temperature and 450 °C lead to small unstructured clusters of Mn or Mn_xSi_y. Upon annealing at 450 °C and 480 °C, Mn reacts chemically with the Si substrate and forms silicide islands. The dimer rows of the substrate become visible again. Two distinct island shapes are found and identified as MnSi and Mn₅Si₃.     

SiO adsorption on a p(2 × 2) reconstructed Si(1 0 0) surface

We have investigated the adsorption mechanism of SiO molecule incident on a clean Si(1 0 0) p(2 × 2) reconstructed surface using density functional theory based methods. Stable adsorption geometries of SiO on Si surface, as well as their corresponding activation and adsorption energies are identified. We found that the SiO molecule is adsorbed on the Si(1 0 0) surface with almost no activation energy. An adsorption configuration where the SiO binds on the channel separating the dimer rows, forming a Si-O-Si bridge on the surface, is the energetically most favourable geometry found. A substantial red-shift in the calculated vibrational frequencies of the adsorbed SiO molecule in the bridging configurations is observed. Comparison of adsorption energies shows that SiO adsorption on a Si(1 0 0) surface is energetically less favourable than the comparable O₂ adsorption. However, the role of SiO in the growth of silicon sub-oxides during reactive magnetron plasma deposition is expected to be significant due to the relatively large amount of SiO molecules incident on the deposition surface and its considerable sticking probability. The stable adsorption geometries found here exhibit structural properties similar to the Si/SiO₂ interface and may be used for studying SiO_x growth.     

Adsorption of Al on the Si(0 0 1) surface

The first-principles calculations have been presented to study the adsorption of aluminum (Al) on the Si(0 0 1)(2×1) surface. We have investigated the optimized geometries and electronic structures of the adatom-substrate system. The adsorption energy of the system has been calculated. The most stable adsorption sites were consequently determined to be HH site and T3+T4. It is shown that the Si-Si dimer is asymmetric on the reconstructed bare surface and become symmetric upon Al adsorption. In addition, the bond length of Si-Si was found to be considerably elongated in the adsorption system. It is found that the work function change obtained in our work is different from other previous results on the adsorption of alkali metals on the Si(0 0 1) surface. In order to investigate the relative stability of phases at different coverages, the surface formation energy of the adsorption system was calculated. To shed light on the nature of the Al-Si bond and the character of silicon surface, the density of states (DOS) and difference charge density of the system were evaluated.     

Adsorption and reactions of HN3 on Si(1 0 0)-2 × 1: A computational study

We have studied the adsorption and decomposition of HN₃ on Si(1 0 0)-2 × 1 surface using the hybrid density functional B3LYP method and effective core potential basis, LanL2DZ, with Si₁₅H₁₆ as a double dimer surface model for cluster calculations. The result shows that the barriers for the dissociative adsorption of HN₃ forming HN(a) + N₂(g) are quite low by stepwise dissociation processes occurring either on a dimer or across the dimers. The low activation energies are consistent with previous experimental observations that the molecularly adsorbed HN₃(a) can undergo decomposition producing HN(a) at low surface temperatures. On the other hand, the predicted activation energies for the N₃(a) + H(a) formation processes are all relatively higher. These results also explain the absence of the N₃(a) species in HREELS measurements following each annealing experiment. Several selected reaction paths were also confirmed with slab model calculations using an optimization approach coupling the energy and gradient calculations by the slab model with the geometrical optimization using Berny algorithm.In addition, the adsorbate effect was examined for the end-on and side-on molecular configurations. For the side-on adsorption configuration, all possible combinations with 1-4 adsorbates can exist on the four surface Si sites of the double dimers, with adsorption energies lying closely to the multiples of that of a single side-on adsorbate (LM2); i.e., adsorption energies are nearly additive. Interestingly, for the end-on adsorption, only 1 and 2 HN₃ molecules can adsorb on a dimer due to the presence of the negative charges on the remaining Si sites in the neighboring dimer. For the two end-on adsorbates on the same dimer, the total adsorption energy is close to two times that of HN₃(a) or LM1. For the mixed end-on/side-on configurations, only one of each type can co-exist on a single dimer pair (Si₁-Si₂ or Si₃-Si₄) sites with adsorption energy close to the sum of those of one end-on and one side-on adsorbate. Finally, the predicted reaction routes and vibrational frequencies showed good agreement with previous experimental results. The stabilities of many ad-species involved in these reactions with end-on and/or side-on configurations have been predicted together with the transition states connecting those species.     

Selforganized nanopatterning of Si(0 0 1)

The (2 × n) superstructure of Si(0 0 1) consists of elongated (2 × 1) reconstructed stripes separated by a dimer-vacancy line every few nanometers, thus offering a means to obtain a nanopattered Si(0 0 1) surface. Scanning tunneling microscopy (STM) investigations of Si(0 0 1) substrates that were deoxidized at 880-920 °C reveal that the formation of the (2 × n) depends strongly on the Si coverage of the topmost surface layer. It forms only in a narrow coverage window ranging from 0.6 to 0.8 ML. Systematic Monte Carlo simulations by an algorithm that combines the diffusion of monomers and dimers with the simultaneous deposition of Si onto the Si(0 0 1) surface, corroborate the STM results and suggest Si deposition as a viable alternative for introducing dimer vacancies in a well-defined manner.     

Methanol adsorption on silicon (0 0 1)

Using a first-principles pseudopotential technique, we have investigated the adsorption of CH₃OH on the Si(0 0 1) surface. We have found that, in agreement with the overall experimental picture, the most probable chemisorption path for methanol adsorption on silicon (0 0 1) is as follows: the gas phase CH₃OH adsorbs molecularly to the electrophilic surface Si atom via the oxygen atom and then dissociates into Si-OCH₃ and H, bonded to the electrophilic and nucleophilic surface silicon dimer atoms, respectively. Other possible adsorption models and dissociation paths are also discussed. Our calculations also suggest that the most probable methanol coverage is 0.5 ML, i.e., one molecule per Si-Si dimer, in agreement with experimental evidences. The surface atomic and electronic structures are discussed and compared to available theoretical and experimental data. In addition, we propose that a comparison of our theoretical STM images and calculated vibrational modes for the adsorbed systems with detailed experimental investigations could possibly confirm the presented adsorption picture.     

STM observation of initial growth of Sn atoms on Ge(0 0 1) surface

We have studied initial growth of Sn atoms on Ge(0 0 1) surfaces at room temperature and 80 K by scanning tunneling microscopy. For Sn deposition onto the Ge(0 0 1) substrate at room temperature, the Sn atoms form two kinds of one-dimensional structures composed of ad-dimers with different alignment, in the 〈3 1 0〉 and the 〈1 1 0〉 directions, and epitaxial structures. For Sn deposition onto the substrate at 80 K, the population of the dimer chains aligning in the 〈3 1 0〉 direction increases. The diffusion barrier of the Sn adatom on the substrate kinetically determines the population of the dimer chain. We propose that the diffusion barrier height depends on surface strain induced by the adatom. The two kinds of dimer chains appearing on the Ge(0 0 1) and Si(0 0 1) surfaces with adatoms of the group-IV elements are systematically interpreted in terms of the surface strain.     

Mixed PbSi dimer chains on Si(1 0 0): a first-principles study

It has been a common belief that the one-dimensional structures observed by STM at low coverage of Pb on Si(1 0 0) are buckled Pb-Pb dimer chains. However, using first-principles density functional calculations, we found that it is energetically favorable for Pb adatoms to intermix with Si atoms to form mixed dimer chains on Si(1 0 0), instead of Pb-Pb dimer chains as assumed in previous studies. Up to a Pb coverage of 0.125 ML, mixed PbSi dimer chain is 0.19 eV per Pb atom lower in energy than Pb dimer chain.     

Progressive changes in atomic structure and gap states on Si(0 0 1) by Bi adsorption

From ab initio studies employing the pseudopotential method and the density functional scheme, we report on progressive changes in geometry, electronic states, and atomic orbitals on Si(0 0 1) by adsorption of different amounts of Bi coverage. For the 1/4 ML coverage, uncovered Si dimers retain the characteristic asymmetric (tilted) geometry of the clean Si(0 0 1) surface and the Si dimers underneath the Bi dimer have become symmetric (untilted) and elongated. For this geometry, occupied as well as unoccupied surface states are found to lie in the silicon band gap, both sets originating mainly from the uncovered and tilted silicon dimers. For the 1/2 ML coverage, there are still both occupied and unoccupied surface states in the band gap. The highest occupied state originates from an elaborate mixture of the p_z orbital at the Si and Bi dimer atoms, and the lowest unoccupied state has a ppσ* antibonding character derived from the Bi dimer atoms. For 1 ML coverage, there are no surface states in the fundamental bulk band gap. The highest occupied and the lowest unoccupied states, lying close to band edges, show a linear combination of the p_z orbitals and ppσ* antibonding orbital characters, respectively, derived from the Bi dimer atoms.     

Ge(5 × 5) as a wetting layer on Si(1 1 1)

Ab initio total energy methods are used to investigate the effects on a Ge(1 1 1)-5 × 5 surface of the lateral compressive stress that would be due to a Si substrate, and the effects of intermixing at the interface with the substrate. The effects of stress due to the lattice mismatch between Si and Ge are studied on a Ge slab by changing the lattice constant in the surface plane from that of experimental bulk diamond Ge to that of Si. When this is done the height difference of the Ge adatoms in the faulted half-cell from those in the unfaulted half is accentuated. Effects on the Ge surface due to the presence of the Si-Ge interface were studied using a thin Ge layer on a Si substrate. The presence of the substrate leads to corrugations with significant height differences appearing among the faulted adatoms. The energetics of intermixing were investigated for Si-Ge single atom interchanges. Additional corrugations resulted from the shortened bondlengths due to the Si impurity in the wetting layer.     

Surface diffusion of carbon atom and carbon dimer on Si(0 0 1) surface

Carbon (C) atom and carbon dimer (C2) are known to be the main projectiles in the deposition of diamond-like carbon (DLC) films. The adsorption and diffusion of the C adatom and addimer (C2) on the fully relaxed Si(0 0 1)-(2 × 1) surface was studied by a combination of the molecular dynamics (MD) and Monte Carlo (MC) simulation. The adsorption sites of the C and C2 on the surface and the potential barriers between these sites were first determined using the semi-empirical many-body Brenner and Tersoff potential. We then estimated their hopping rates and traced their pathways. It is found that the diffusion of both C and C2 is strongly anisotropic in nature. In addition, the C adatom can diffuse a long distance on the surface while the adsorbed C2 is more likely to be confined in a local region. Thus we can expect that smoother films will be formed on the Si(0 0 1) surface with single C atoms as projectile at moderate temperature, while with C2 the films will grow in two-dimensional islands. In addition, relatively higher kinetic energy of the projectile, say, a few tens of eV, is needed to grow DLC films of higher quality. This is consistent with experimental findings.     

Low reactivity of molecular oxygen with Si(1 1 1)-c(2 × 8)

The adsorption of molecular oxygen on the c(2 × 8) reconstruction of quenched Si(1 1 1) surfaces has been studied at the atomic scale using scanning tunneling microscopy (STM) at room temperature (RT). It has been found that clean well reconstructed c(2 × 8) adatoms do not react with O₂ molecules but that a limited oxidation can start where adatom sites arranged in reconstructed structures are present. Comparison between O₂ adsorption on Si(1 1 1)-c(2 × 8) and Si(1 1 1)-7 × 7 reconstructions coexisting on the same quenched silicon surface has been carried out in detail. For each atomic site present on the surface the variation of reacted sites with exposure has been measured. For low O₂ exposures, bright and dark oxygen induced sites appear on the Si(1 1 1)-7 × 7, while Si(1 1 1)-c(2 × 8) does not oxidized at all. At high O₂ exposures, large oxidation areas have spread on the 7 × 7 reconstruction, preferentially on the faulted halves of the unit cell, and smaller oxidation areas induced by topological defects have grown all around clean un-reacted c(2 × 8) regions.     

Structural analysis of the c(4 × 2) reconstruction in Si(0 0 1) and Ge(0 0 1) surfaces by low-energy electron diffraction

The c(4 × 2) structures in (0 0 1) surfaces of Si and Ge have been studied by low-energy electron diffraction (LEED). Using a proper cleaning method for the Si surface, we were able to observe clear c(4 × 2) LEED patterns up to incident energy of ∼400 eV as well as the Ge surface. Extensive experimental intensity-voltage curves allowed us to optimize the asymmetric dimer model up to the eighth layer (including the dimer layer) in depth in the dynamical LEED calculation. Optimized structural parameters are almost the same for the Si and Ge except for the height of the buckled-up atom of the asymmetric dimer. For the Ge surface, the structural parameters are in excellent agreement with those obtained by a previous theoretical calculation. The tilt angle and bond length of the dimer are 18 ± 1 (19 ± 1)° and 2.4 ± 0.1 (2.5 ± 0.1) Å for the Si(0 0 1) (Ge(0 0 1)), respectively.     

Surface etching induced by Ce silicide formation on Si(1 0 0)

We investigate the temperature-dependent surface etching process induced by Ce silicide on Si(1 0 0) using scanning tunneling microscopy and X-ray photoelectron spectroscopy. We found that step edges on the Si(1 0 0) surface are gradually roughened due to the formation of Ce silicide as a function of substrate temperature. Unlike the Si(1 1 1) surface, however, terrace etching also occurs in addition to step roughening at 500 °C. Moreover, we found that Si(1 0 0) dimers are released and formed dimer vacancy lines because bulk diffusion of Ce silicide into Si(1 0 0) surface occurs the defect-induced strain at higher temperature (∼600 °C).     

STM study of Si(1 1 3) 3 × 2-Ga surface

The Ga-adsorbed structure on Si(1 1 3) surface at low coverage has been studied by scanning tunneling microscopy (STM). The bright protrusion corresponding to the position of the dimer without the interstitial Si atom of the clean surface disappeared in the filled-state STM image after Ga adsorption, although the protrusion due to the Si adatom still remained. On the basis of the adatom-dimer-interstitial (ADI) model, this result indicates that the Ga atom is adsorbed interstitially at the center of another pentamer that does not have the interstitial Si atom. An ab initio calculation was performed and STM images were simulated.     

A structural parallel between Ge- and Si-induced 4 × 4 and 3 × 3 reconstructions on SiC(0 0 0 1) drawn from comparative RHEED oscillations

The initial Ge growth stages on a (√3 × √3)R30°-reconstructed SiC(0 0 0 1) surface (√3) have been studied using a complete set of surface techniques such as reflection high energy electron diffraction (RHEED), low energy electron diffraction (LEED), atomic force microscopy (AFM) and photoemission and compared with similar Si surface enrichments in place of Ge. The investigations essentially focus on the wetting growth-regimes that are favoured by the use of the √3 surface as a starting substrate, this surface being the closest to a smooth and ideally truncated Si-terminated face of hexagonal SiC(0 0 0 1). Depending on temperature and Ge or Si coverages, varying surface organizations are obtained. They range from unorganized layer by layer growths to relaxed Ge(1 1 1) or Si(1 1 1) island growths, through intermediate attempts of coherent and strained Ge or Si surface layers, characterized by 4 × 4 and 3 × 3 surface reconstructions, respectively. RHEED intensity oscillation recordings, as a function of Ge or Si deposited amounts, have been particularly helpful to pinpoint the limited (by the high lattice mismatch) existence domains of these interesting coherent phases, either in terms of formation temperature or surface coverages. Prominently comparable data for these two Ge- and Si-related reconstructions allow us to propose an atomic model for the still unexplained Ge-4 × 4 one. It is based on a same local organization in trimer and ad-atom units for the Ge excess as admitted for the Si-excess of the 3 × 3 surface, the higher strain nevertheless favouring arrangements, for the Ge-units, in 4 × 4 arrays instead of 3 × 3 Si ones. Admitting such models, 1.25 and 1.44 monolayers of Ge and Si, should, respectively, be able to lie coherently on SiC, with respective lattice mismatches near 30% and 25%. The experimental RHEED-oscillations values are compatible with such theoretical ones. Moreover, these RHEED coverage determinations (for layer completion, for instance) inform us in turn about the initial Si richness of the starting √3 reconstruction and help us to discriminate between earlier contradictory atomic models proposed in the literature.