Si Pairs Research Articles

Design and performance of the variable-wavelength Bonse–Hart ultra-small-angle neutron scattering diffractometer KOOKABURRA at ANSTO

The double-crystal ultra-small-angle neutron scattering (USANS) diffractometer KOOKABURRA at ANSTO was made available for user experiments in 2014. KOOKABURRA allows the characterization of microstructures covering length scales in the range of 0.1–10 µm. Use of the first- and second-order reflections coming off a doubly curved highly oriented mosaic pyrolytic graphite premonochromator at a fixed Bragg angle, in conjunction with two interchangeable pairs of Si(111) and Si(311) quintuple-reflection channel-cut crystals, permits operation of the instrument at two individual wavelengths, 4.74 and 2.37 Å. This unique feature among reactor-based USANS instruments allows optimal accommodation of a broad range of samples, both weakly and strongly scattering, in one sample setup. The versatility and capabilities of KOOKABURRA have already resulted in a number of research papers, clearly demonstrating that this instrument has a major impact in the field of large-scale structure determination.

Computer simulation of CaSiO3 glass under compression: correlation between Si–Si pair radial distribution function and intermediate range order structure

The structural organization of CaSiO3 glass at 600 K and under pressure of 0–100 GPa is investigated by molecular dynamics simulation (MDS). Results show that the atomic structure of CaSiO3 comprises SiOn and CaOm units considered as basic structural polyhedra. At low pressure, most of the basic structural polyhedra are SiO4, CaO5, CaO6 and CaO7. At high pressure most of the basic structural polyhedra are SiO5, SiO6 and CaO9, CaO10 and CaO11. The distribution of basic structural polyhedra is not uniform resulting in formation of Ca-rich and Si-rich regions. The distribution of SiO4, SiO5 and SiO6 polyhedra is also not uniform, but it tends to form SiO4-, SiO5-, and SiO6-clusters. For the Si–O network, under compression there is a gradual transition from the tetrahedral network (SiO4) to the octahedral network (SiO6) via SiO5 polyhedra. The SiO5-clusters are the same as immediate-phase in the transformation process. The size and shape of SiO4 tetrahedra change strongly under compression. While the size of SiO5 and SiO6 has also changed significantly, but the shape is almost unchanged under compression. The SiOn polyhedra can connect to each other via one common oxygen ion (corner-sharing bond), two common oxygen ions (edge-sharing bond) or three common oxygen ions (face-sharing bond). The Si–Si bond length in corner-sharing bonds is much longer than the ones in edge-sharing and face-sharing bonds. The change of intermediate range order (IRO) structure under compression relating to edge- and face-sharing bonds amongst SiOn at high pressure is the origin of the first peak splitting of the radial distribution functions of Si–Si pair. Under compression, the number of non-bridging oxygen (NBO) decreases. This makes the Si–O network more polymerized. At low pressure, most of the Ca2+ ions incorporate into the Si–O network via NBOs. At high pressure, the amount of NBO decreases, Ca2+ ions mainly incorporate into the Si–O network via bridging oxygen (BO) that belongs to SiO5 and SiO6 with a negative charge. And this is the principle for immobilization of heavy metal as well as fissile materials in hazardous waste (nuclear waste).

Initial surface silicidation on Ni(110)

Initial silicide formation on a Ni(110) surface was studied by scanning tunneling microscopy (STM) in an ultrahigh vacuum. Less than 0.5ML of Si deposition initiated a Si-Ni mixed layer by displacing substrate Ni, and dark sites were formed in the STM images. A 0.5ML-Si deposited surface showed that Si and Ni were alternately aligned in a close-packed [11¯0] row whereas Si pairs aligned along the [001] direction forming p(1×2), obliquely aligned forming c(2×2), or even straightly-and-obliquely aligned forming c(4×2) superstructures. A first-principles total energy calculation showed that the p(1×2) and c(4×2) structures had almost the same energy while the c(2×2) structure gave 13meV/1×1 higher energy. Because a Si-Si bond in the close-packed [11¯0] row is energetically unfavorable, Si deposition of more than 0.5ML did not further replace the substrate Ni, but silicide islands were nucleated along with a trench structure.

A first-principles model of copper–boron interactions in Si: implications for the light-induced degradation of solar Si

The recent discovery that Cu contamination of Si combined with light exposure has a significant detrimental impact on carrier life-time has drawn much concern within the solar-Si community. The effect, known as the copper-related light-induced degradation (Cu-LID) of Si solar cells, has been connected to the release of Cu interstitials within the bulk (2016 Sol. Energy Mater. Sol. Cells 147 115–26). In this paper, we describe a comprehensive analysis of the formation/dissociation process of the CuB pair in Si by means of first-principles modelling, as well as the interaction of CuB defects with photo-excited minority carriers. We confirm that the long-range interaction between the cation and the anion has a Coulomb-like behaviour, in line with the trapping-limited diffusivity of Cu observed by transient ion drift measurements. On the other hand, the short-range interaction between the d-electrons of Cu and the excess of negative charge on produces a repulsive effect, thereby decreasing the binding energy of the pair when compared to the ideal point-charge Coulomb model. We also find that metastable CuB pairs produce acceptor states just below the conduction band minimum, which arise from the Cu level emptied by the B acceptor. Based on these results, we argue that photo-generated minority carriers trapped by the metastable pairs can switch off the Coulomb interaction that holds the pairs together, enhancing the release of Cu interstitials, and acting as a catalyst for Cu-LID.

Stability and electronic structures of isoelectronic impurity complexes in Si: First-principles study

The stability and electronic structures of various isoelectronic cation+anion pairs in Si were studied by first-principles calculations. It was shown that the nearest-neighboring substitutional configuration is the most stable structure for most cation+anion pairs, while B+N, Mg+O, and Be+O pairs preferentially occupy a single Si site owing to the small atomic radii of B, N, and O atoms. We found that only Al+N, Ga+N, and In+N pairs produce electron-unoccupied weakly localized states in the band gap of Si, reflecting the large negativity of N atoms. We also showed that single N doping produces N+N pairs and that such pairs induce no electronic states in the band gap of Si. Therefore, the codoping of Al and N atoms is essential to produce an electronic state in the Si band gap and increase the tunneling current in Si tunneling field-effect transistors.

Density Functional Theory Study on Frenkel Pair Formation from Oxygen Clusters in Si Crystal

Silicon (Si) wafers, which are the most widely used semiconductors for integrated circuit fabrication, are produced by polishing after slicing Czochralski (CZ) grown Si ingots. Oxygen (O) atoms with a concentration of 1×1018 atom/cm3 are incorporated from the melt of Si charged in the SiO2 crucible during crystal growth. Most of the O atoms in Si crystals exist as interstitials (Oi) at the center of Si - Si bonds, which are detected with the Fourier transform infrared (FTIR) method. As the solubility of O atoms in Si crystals is about 4.4×1016 atom/cm3 at 800°C, the typical temperature at which integrated circuits are fabricated, the oxygen in as-grown CZ Si crystals is usually supersaturated. Therefore, the Oi atoms precipitate in the subsequent heat treatments and grow into oxide precipitates (SiO2). It is well known that self-interstitial (I) atoms are emitted from growing oxide clusters to relieve the compressive lattice strain. Since these emitted atoms interact with dopant atoms, impurities, lattice defects, etc., and affect the performance of Si wafers, the mechanism of emission from Oi clusters should be clarified and taken into consideration in the computer simulation of oxygen precipitation. In this study, we analyzed the Frenkel (self-interstitial I and vacancy V) pair formation from oxygen clusters in Si crystal by density functional theory (DFT) study. The program package used in this study was CASTEP code. The ground state of the system in this method was found by solving the Kohn-Sham equation, which is a rule equation of electronic systems for given atom placement. The wave function was expanded as plane waves, and the ultrasoft pseudopotential was used to reduce the plane wave number. A Si 64-atom cubic cell, a 2 × 2 × 2 supercell constructed with a conventional cell, was used in this study. The cell was surrounded by (100), (010), and (001) planes, and each edge of the cell was along the [100], [010], and [001] directions. Three-dimensional periodic boundary conditions were set for each calculation. The cutoff energy for the plane-wave expansion was 340 eV. Generalized gradient approximation (GGA) was used for the exchange-correlation term, and the functional form was of the Perdew-Burke-Ernzerhof (PBE) type. The density mixing method was used to minimize the energy of the electronic system, and the Broyden-Fletcher-Goldfarb-Shanno (BFGS) optimizing structure method was run to optimize atom placement. The 2 × 2 × 2 points were used for k sampling. The binding energies of up to five Oi atoms in Si crystal were calculated. In addition, the formation energy of the Frenkel (I and V) pair was calculated in a perfect cells and Oi cluster involving cells. For the cluster including two Oi atoms, the O2 dimer of Si-Oi-Si-Oi-Si has the lowest binding energy of about 0.45 eV. The binding energy depends not only on the distance between two Oi atoms but also their atomic configuration. From n = 2 to 6, the binding energies of an Oi and On-1 cluster increase with increasing Oi number n. The calculated formation energies of the Frenkel pair from Oi clusters are about 2.30 eV (O2), 1.83 eV (O3), 1.17 eV (O4), 1.48 eV (O5), and 1.56 eV (O6). The formation energy of the Frenkel pair in perfect Si is about 4.87 eV. This result indicates that the formation of I becomes easier by the accumulation of Oi atoms in Si. The formation of Frenkel pairs around dopant atoms is ongoing.

Direct visualization of a cycloaddition reaction on frozen asymmetric Si dimers at room temperature

We firstly report an experimental visualization of a cycloaddition reaction on RT frozen asymmetric Si dimers. The frozen Si dimers with a local c(4 × 2) order were prepared by pinning flip-flopping Si dimers by using molecules. This RT pristine c(4 × 2) structure was used to determine what Si atom of an asymmetric Si dimer bonds to a molecule at the initial stage of the RT cycloaddition reaction, which has been a long-standing puzzling issue. This made it possible to compare directly experimental cycloaddition reactions with theoretical ones. As a prototype for the experiment, a 1,3-butadiene molecule adsorbed between Si dimer rows was used. The 1,3-butadiene molecule was found to prefer a symmetric Si pair on the frozen Si dimers, i.e., two electrophilic lower atoms of asymmetric Si dimers. This result is consistent with the theoretical prediction that a 1,3-diene molecule prefers a symmetric Si pair on the Si(001)c(4 × 2) surface. This experimental approach can also be applied to other studies for the adsorption of a molecule on a Si(001) surface at room temperature.

First-Principles Study of Chemical and Topological Short-Range Orders in the Mg–Si Liquid Alloys

The structures of a series of Mg–Si liquid alloys were investigated by means of ab initio molecular dynamic simulation. The pair distribution function analysis manifests a tendency of aggregating for the Si–Si pairs in the Mg90Si10 liquid alloy. Chemical short-range orders are observed around Si atoms between unlike atoms, and the maximum is observed for the eutectic Mg47Si53 alloy. Furthermore, the topological environment changed abruptly near the eutectic Mg47Si53 alloy according to Voronoi polyhedra analysis. The variation of diffusion coefficients of Mg and Si suggests that the dynamical properties of Mg–Si liquid alloys are more sensitive to temperature than to compositions.

Density functional theory study on the possibility of Si‐, Ge‐, and Sn‐doped carbon nanotubes as efficient support materials for platinum

PBEPBE‐D3 calculations were performed to investigate how platinum (Pt) interacts with the internal and external surfaces of single‐walled pristine, Si‐, Ge‐, and Sn‐doped (6,6) carbon nanotubes (CNTs). Our calculations showed that atomic Pt demonstrates stronger binding strength on the external surfaces than the internal surface adsorption for the same type of nanotube. In cases of external surface adsorptions, Si‐, Ge‐, and Sn‐doped CNTs show comparable binding energies for Pt, at least 1.40 eV larger than pristine CNT. This enhancement can be rationalized by the strong covalent interactions between Pt and XC (X = Si, Ge, and Sn) pairs based on structural and projected density of states analysis. In terms of internal surface adsorptions, Ge and Sn doping could significantly enhance the binding of Pt. Pt atom shows much more delocalized and bonding states inside Ge‐ and Sn‐doped CNTs, indicating multiple‐site interaction pattern when atomic Pt is confined inside the nanotubes. However, the internal surface of Si‐doped CNT presents limited enhancement in Pt adsorption with respect to that of pristine CNT because of their similar binding geometries. © 2016 Wiley Periodicals, Inc.

Critical fluctuations of the proton density in A+A collisions at 158A GeV

We look for fluctuations expected for the QCD critical point using an intermittency analysis in the transverse momentum phase space of protons produced around midrapidity in the 12.5\% most central C+C, Si+Si and Pb+Pb collisions at the maximum SPS energy of 158$A$~GeV. We find evidence of power-law fluctuations for the Si+Si data. The fitted power-law exponent $\phi_{2} = 0.96^{+0.38}_{-0.25}\text{ (stat.)}$ $\pm 0.16\text{ (syst.)}$ is consistent with the va\-lue expected for critical fluctuations. Power-law fluctuations had previously also been observed in low-mass $\pi^+ \pi^-$ pairs in the same Si+Si collisions.

Affect of the graphene layers on the melting temperature of silicon by molecular dynamics simulations

In the paper, molecular dynamics simulations have been used to detect the melting temperature of silicon. The two models have been considered as the graphene/silicon/graphene (GSG) sandwich and the only silicon system. Atoms in the models interact with each other via a 3-body Tersoff potential with a modified part based on a Coulomb potential and the Ziegler–Biersack–Littmark universal screening function. We find that C–Si interaction prevents the melting of the two silicon surfaces adjoining the graphene layers. The melting temperature of the silicon part in the GSG sandwich (Tm=2450K) is 1.6 times higher than that of the only silicon system (Tm,pure=1540K). Difference of these melting temperatures has original from interaction of C–Si pairs, which causes decreasing of the energy of the silicon part in the GSG sandwich during heating process leading to an increase of the melting temperature of this silicon part.

Tunneling spectroscopy of close-spaced dangling-bond pairs in Si(001):H.

We present a combined experimental and theoretical study of the electronic properties of close-spaced dangling-bond (DB) pairs in a hydrogen-passivated Si(001):H p-doped surface. Two types of DB pairs are considered, called “cross” and “line” structures. Our scanning tunneling spectroscopy (STS) data show that, although the spectra taken over different DBs in each pair exhibit a remarkable resemblance, they appear shifted by a constant energy that depends on the DB-pair type. This spontaneous asymmetry persists after repeated STS measurements. By comparison with density functional theory (DFT) calculations, we demonstrate that the magnitude of this shift and the relative position of the STS peaks can be explained by distinct charge states for each DB in the pair. We also explain how the charge state is modified by the presence of the scanning tunneling microscopy (STM) tip and the applied bias. Our results indicate that, using the STM tip, it is possible to control the charge state of individual DBs in complex structures, even if they are in close proximity. This observation might have important consequences for the design of electronic circuits and logic gates based on DBs in passivated silicon surfaces.

Specific features in the olive self-incompatibility system: A method to decipher S-allele pairs based on fruit settings

Olive fruit production is dependent on wind pollination, and recently sporophytic self-incompatibility has been advocated to function with six S-alleles. The aim here is to show fruit setting data gathered from literature can be interpreted and their standardization help to release firm conclusions on the S-allele pair carried by most varieties. Under a bag, fruit settings for mates do not allow to infer inter-compatibility of pairs of mates. Each variety displays a different number of flowers per inflorescence, and for a given proportion of hermaphrodite flowers, fruit sets per hermaphrodite flowers were calculated. Thus, broad fruit set data were standardized per 100 hermaphrodite flowers. The ratio fruit sets from self-pollination and fruit sets from crosses are commonly referred to fruit sets after open pollination, as the index of self-incompatibility (ISI) and the index for cross inter-compatibility (ICOI), respectively. Open pollination rate (OP-R) is considered to be the best rate of fruit sets for a variety. ICOI values were classified with the threshold at 0.6 to rank a pair of varieties as inter-incompatible and inter-compatible. Before, we have proceeded by trials and errors. Here we showed that rationale reasoning is efficient to analyze fruit settings. Computations on ICOI showed that matings that succeeded are more confident than matings that failed, because pollen efficiency appeared insufficient. Computations on ISI revealed high correlations between SI and co-dominant S-allele pairs as for SC and S-allele pairs sharing dominance relationships. Thus, the sporophytic self-incompatibility (SSI) model will help geneticists to unravel the S-allele series and to predict suitable pollinizers in olive orchards. Moreover, compensation for OP-R deficiency in orchards could be achieved by supplementing suitable pollinizers computed in silico.

Adsorption and dissociation of hydrogen peroxide on the defected carbon nanotubes

Density functional calculations have been performed on the structural and electronic properties of the pristine and Stone–Wales-defected (SW) carbon nanotubes (CNT) with and without the adsorption of a H2O2 molecule. H2O2 interacts with the pristine CNT weakly, releasing energy of about 2.9 kcal/mol, but it presents a bit higher reactivity toward the SW-defected CNT. However, the adsorption of the H2O2 molecule on the SW-CNT is still very weak because of its small adsorption energy, large binding distance, and small charge transfer. To enhance the reactivity, a Si–Si pair is doped at the center of a SW defect, forming a Si2-SW-CNT. In this case, a strong adsorption is found with a large E ad of 155.6 kcal/mol, and a short bond length in comparison with the SW-CNT. It also was shown that the H2O2 molecule could be reduced into the OH on the Si2-SW-CNT without significant change in the electronic properties of the tube.

An atomistic picture of the diffusion of two vacancies forming a di‐vacancy in Si

A method is proposed to calculate the formation of a point defect pair in Si based on an atomistic picture. As an illustration, the capturing of a vacancy (V) by another V forming a di‐vacancy is described semi‐quantitatively. It is shown that in Si, the number of most likely diffusion paths of two vacancies V to form a di‐vacancy is limited. The energetically most favorable diffusion paths are the twelve zigzag Si chains surrounding the V along the ⟨110⟩ directions. Systematically calculating all of the irreducible configurations of two vacancies, each with its weight (=total number of each possible configuration), enables determining all possible V diffusion paths with their relative weights. This obtained picture is different from diffusion in the continuum model, which does not take into account the anisotropy of covalent bonds in the diamond lattice. The proposed approach is a first step towards the calculation of the effective “capture cross‐section” for the formation of point defect pairs that can be used in continuum models.

Synergetic effects of Mn and Si in the interaction with point defects in bcc Fe

The interaction of Mn, Si and Cr with a vacancy and self-interstitial defects in BCC Fe has been analyzed using ab initio calculations. While the interaction of the considered solute clusters with a single vacancy is linearly additive, there is a considerable synergetic effect in the case of self-interstitial atoms, found to bind strongly with Mn–Si pairs. The latter therefore act as deep trapping configurations for self-interstitials. At the same time, the presence of the point defects nearby weakly attractive Mn–Si pairs significantly enhances the solute–solute binding. The revealed effects are rationalized on the basis of charge density and local magnetic moment distributions.

SPECTROSCOPIC STUDY OF A DARK LANE AND A COOL LOOP IN A SOLAR LIMB ACTIVE REGION BYHINODE/EIS

We investigated a cool loop and a dark lane over a limb active region on 2007 March 14 using the Hinode/EUV Imaging Spectrometer. The cool loop is clearly seen in the spectral lines formed at the transition region temperature. The dark lane is characterized by an elongated faint structure in the coronal spectral lines and is rooted on a bright point. We examined their electron densities, Doppler velocities, and nonthermal velocities as a function of distance from the limb. We derived electron densities using the density sensitive line pairs of Mg VII, Si X, Fe XII, Fe XIII, and Fe XIV spectra. We also compared the observed density scale heights with the calculated scale heights from each peak formation temperatures of the spectral lines under the hydrostatic equilibrium. We noted that the observed density scale heights of the cool loop are consistent with the calculated heights, with the exception of one observed cooler temperature; we also found that the observed scale heights of the dark lane are much lower than their calculated scale heights. The nonthermal velocity in the cool loop slightly decreases along the loop, while nonthermal velocity in the dark lane sharply falls off with height. Such a decrease in the nonthermal velocity may be explained by wave damping near the solar surface or by turbulence due to magnetic reconnection near the bright point.

Transport and Photoelectric Effects in Structures with Ge and SiGe Nanoclusters Grown on Oxidized Si (001)

Crystalline germanium nanoclusters (NCs) are grown by a molecular-beam epitaxy technique on chemically oxidized Si (100) surface at 700oC. Deposition of silicon on the surface with Ge nanoclusters leads to surface reconstruction and formation of polycrystalline diamond-like Si coverage, while nanoclusters core becomes tetragonal SiGe alloy. Possible mechanisms for nanoclusters growth are discussed. Selective photoexcitation of Ge or SiGe nanoclusters or space-charge layer of underlying Si allows to observe two non-equilibrium steady-states with higher and lower conductivity values as compared to the equilibrium one. The persistent photoconductivity (PPC) behaviour was observed after excitation of electron-hole pairs in Si (001) substrate. This effect may be attributed to spatial carrier separation by macroscopic fields in the depletion layer of the near-surface Si. Decreasing of surface conductivity, driven by optical recharging of NCs and Si/SiO2 interface states, is observed in the spectral range from 0.6 to 1.0 eV. Conductivity drop is discussed in the terms of hole accumulation by Ge-NC states enhancing the local-potential variations and, therefore, decreasing the surface conductivity of p-Si.

Kookaburra: the ultra-small-angle neutron scattering instrument at OPAL

The new double crystal ultra-small-angle neutron scattering instrument Kookaburra, currently under construction at the ANSTO OPAL reactor, will allow characterization of microstructures covering length scales in the range of 0.1 to 10 µm. Using the 002 and 004 reflections of a doubly curved mosaic highly oriented pyrolytic graphite premonochromator crystal at a fixed Bragg angle of 45° in conjunction with two pairs of Si(111) and Si(311) quintuple-reflection channel-cut crystals will allow operation of the instrument at two different wavelengths, thus optimally accommodating weakly and strongly scattering samples in one sample position. The versatility, the estimated neutron fluxes and the low background noise of Kookaburra suggest that this state-of-the-art instrument will have a major impact in the field of large-scale structure determination.

Impact of isovalent defect engineering strategies on carbon-related clusters in silicon

In electron-irradiated silicon (Si) the formation of oxygen-vacancy pairs (VO or A-centers) is deleterious and for this reason isovalent defect engineering strategies have been proposed to suppress their concentration. Carbon-related clusters such as CiOi and CiCs are also important as they are electrically active and their properties need to be determined. The present study has two aims. The first aim is to review the impact of isovalent doping [germanium (Ge), tin (Sn) and lead (Pb)] on the thermal stability of the CiOi and CiCs pairs in Si by means of infrared (IR) spectroscopy. The second aim is to analyze these results and show that the presence of isovalent dopants in Si reduces the temperature of annealing (Tann) of the CiOi pairs and increases the temperature of annealing of the CiCs pairs. The results are discussed in view of recent experimental and theoretical studies. It is suggested that the change in the thermal stability of the C-related defects is due to the local strains induced in the lattice by the oversized isovalent dopants. It is surmised that these strains have an opposite impact on the Tann of the CiOi and CiCs pairs as a result of their different structures.