Si Dangling Bond States Research Articles

Electronic phase transitions in quasi-one-dimensional atomic chains: Au wires on Si(553)

The Si(553) surface, covered with half a monolayer of gold, forms a double strand of well-ordered atomic Au chains in each miniterrace. It represents one of the smallest possible realizations of quasi-one-dimensional (1D) systems. In this prototype system, by combining DC conductance and low-energy electron diffraction measurements with density functional calculations and ab initio Monte Carlo simulations, we demonstrate that nonlocal electronic correlations, together with thermal excitations, lead to peculiar phase transitions without long-range order. They are characterized by marginal (average) geometric relaxations, but with huge variations of the electronic band structure and concomitant strong temperature-dependent modifications of the density of states close to the Fermi level. Similar phenomena are expected in the large class of quasi-1D conductors and open a wide range of possibilities for their controlled manipulation. It is the increasing hybridization between spin-polarized Au and Si edge states that makes the Si dangling bond states at the step edge conducting, first as a transient between two insulating phases and finally opening a permanent new 1D conduction channel at high temperatures.

Charge-transfer transition in Au-induced quantum wires on Si(553)

The Si(553)-Au system resembles a heteroatomic chain ensemble with a delicate spin-charge interplay. The ordering of the $\ifmmode\times\else\texttimes\fi{}3$ reconstruction vanishes via a phase transition taking place at ${T}_{c}=100$ K. Our directional-dependent surface transport measurements showed that this order-disorder phase transition is not driven by the formation of a charge-density wave, as previously suggested. Instead, at 65 K there is a pronounced increase of the surface-state conductivity along the wires. We attribute this to activated charge transfer between the localized Si dangling bond states and the proximate Au bands revealing a $\ifmmode\times\else\texttimes\fi{}2$ periodicity. Apparently, a quasiorthogonality between the wave functions of the two proximal reconstructions is also responsible for a missing $\ifmmode\times\else\texttimes\fi{}6$ periodicity along the wires. The electronic charge transfer is in agreement with recent band-structure calculations.

Interface-Induced Room-Temperature Ferromagnetism in Hydrogenated Epitaxial Graphene

We show ferromagnetic properties of hydrogen-functionalized epitaxial graphene on SiC. Ferromagnetism in such a material is not directly evident as it is inherently composed of only nonmagnetic constituents. Our results nevertheless show strong ferromagnetism with a saturation of 0.9μ(B)/hexagon projected area, which cannot be explained by simple magnetic impurities. The ferromagnetism is unique to hydrogenated epitaxial graphene on SiC, where interactions with the interfacial buffer layer play a crucial role. We argue that the origin of the observed ferromagnetism is governed by electron correlation effects of the narrow Si dangling bond states in the buffer layer exchange coupled to localized states in the hydrogenated graphene layer. This forms a quasi-three-dimensional ferromagnet with a Curie temperature higher than 300 K.

Theoretical investigation of atomic structure and electronic properties of Ca/Si(110)-(2 × 1) reconstruction

We have presented first-principles total-energy calculations for the adsorption of Ca metals onto a Si(110) surface. The density functional method was employed within its local density approximation to study the atomic and electronic properties of the Ca/Si(110) structure. We considered the (1×1) and (2×1) structural models for Ca coverages of 0.5 monolayer (ML) and 0.25ML, respectively. Our total-energy calculations indicate that the (1×1) phase is not expected to occur. It was found that Ca adatoms are adsorbed on top of the surface and form a bridge with the uppermost Si atoms. The Ca/Si(110)-(2×1) produces a semiconducting surface band structure with a direct band gap that is slightly smaller than that of the clean surface. One filled and two empty surface states were observed in the gap; these empty surface states originate from the uppermost Si dangling bond states and the Ca 4s states. It is found that the Ca–Si bonds have an ionic nature and complete charge being transferred from Ca to the surface Si atoms. Finally, the key structural parameters of the equilibrium geometry are detailed and compared with the available results for metal-adsorbed Si(110) surface, Ca/Si(001), and Ca/Si(111) structures.

Ab initio calculations of surface structure and electronic properties caused by adsorption of Ca atoms on a Si(1 1 0) surface

The adsorption of Ca metals onto a Si(1 1 0) surface has been theoretically investigated by first-principle total-energy calculations. We employed a local density approximation of the density functional theory as well as a pseudopotential theory to study the atomic and electronic properties of the Ca/Si(1 1 0) structure. The (1×1) and (2×1) surface structures were considered for Ca coverages of 0.5 and 0.25 ML, respectively. It is found that the (1×1) phase is not expected to occur even for rich Ca regime. It was found that Ca adatoms are adsorbed on top of the surface and form a bridge with the uppermost Si atoms. The most stable structure of Ca/Si(1 1 0)–(2×1) surface produces a semiconducting surface band structure with a direct band gap that is slightly smaller than that of the clean surface. We have observed one filled and two empty surface states in the gap region. These empty surface states originated from the uppermost Si dangling bond states and the Ca 4 s states. Furthermore, the Ca–Si bonds have an ionic nature with almost complete charge transfer from Ca to the surface Si atoms. The structural parameters of the ground state atomic configuration are detailed and compared with the available results of metal-adsorbed Si(1 1 0) surface, Ca/Si(0 0 1), and Ca/Si(1 1 1) structures.

Electronic structure and defect states of transition films from amorphous to microcrystalline silicon studied by surface photovoltage spectroscopy

In this paper, surface photovoltage spectroscopy (SPS) is used to determine the electronic structure of the hydrogenated transition Si films. All samples are prepared by using helicon wave plasma-enhanced chemical vapour deposition technique, the films exhibit a transition from the amorphous phase to the microcrystalline phase with increasing temperature. The film deposited at lower substrate temperature has the amorphous-like electronic structure with two types of dominant defect states corresponding to the occupied Si dangling bond states (D0/D−) and the empty Si dangling states (D+). At higher substrate temperature, the crystallinity of the deposited films increases, while their band gap energy decreases. Meanwhile, two types of additional defect states is incorporate into the films as compared with the amorphous counterpart, which is attributed to the interface defect states between the microcrystalline Si grains and the amorphous matrix. The relative SPS intensity of these two kinds of defect states in samples deposited above 300°C increases first and decreases afterwards, which may be interpreted as a result of the competition between hydrogen release and crystalline grain size increment with increasing substrate temperature.

Band-gap opening in metallic carbon nanotubes adsorbed on H∕Si(001)

A recent scanning tunneling microscope nanolithography technique can fabricate one-dimensional “dangling-bond (DB) wire” by the selective removal of H atoms from a H-passivated Si(001) surface along the Si dimer row. We here theoretically investigate the bonding geometry, band structure, and binding mechanism of an armchair (3,3) single-walled carbon nanotube (CNT) adsorbed on the DB wire. We find that the formation of C–Si bonds between the CNT and the DB wire gives rise to hybridization between the carbon π-bond states and the Si dangling-bond states. This hybridization breaks the rotational symmetry of the (3,3) CNT whose π-bonding and π-antibonding bands cross at the Fermi level. As a result, the adsorbed CNT opens an energy gap of ∼0.1eV, yielding a metal-to-semiconductor transition.

Atomic Structure, Band Offsets and Hydrogen in High K oxide:Silicon interfaces

AbstractThe bonding, electronic structure and valence band offsets are calculated for various atomic models of interfaces between Si and high dielectric constant insulators ZrO2 and SrTiO3. A non-polar face of the oxide does not necessarily give a semiconducting interface, because of the need to fill Si dangling bond states on the Si side. This can be achieved by stoichiometry changes. Band offsets of semiconducting interfaces are generally the same as those found from bulk charge neutrality levels, indicating no dipole layers. Dipole layers are present at some configurations, where the offset is then changed by up to 1 eV. The states of hydrogen in the oxides are also considered.

Atomic Structure, Band Offsets and Hydrogen in High K oxide:Silicon interfaces

ABSTRACTThe bonding, electronic structure and valence band offsets are calculated for various atomic models of interfaces between Si and high dielectric constant insulators ZrO2 and SrTiO3. A non-polar face of the oxide does not necessarily give a semiconducting interface, because of the need to fill Si dangling bond states on the Si side. This can be achieved by stoichiometry changes. Band offsets of semiconducting interfaces are generally the same as those found from bulk charge neutrality levels, indicating no dipole layers. Dipole layers are present at some configurations, where the offset is then changed by up to 1 eV. The states of hydrogen in the oxides are also considered.

ANGLE-RESOLVED PHOTOEMISSION STUDIES OF THE 3C–SiC(001)(2×1) SURFACE

We have investigated the electronic structure of the single-domain 3C–SiC(001)(2×1) using angle-resolved photoemission and synchrotron radiation. Two different surface-state bands are clearly identified within the bulk bandgap. The upper band has a binding energy of 1.4 eV at the center of the surface Brillouin zone (SBZ) and shows a weak dispersion of 0.3 eV in the [Formula: see text] direction, but is nondispersive in the perpendicular direction. It has a polarization dependence suggesting a pz character, as expected for a Si dangling-bond state. The second band is located at 2.4 eV binding energy and is nondispersive. The weak or nonexistent dispersions suggest very localized electronic states at the surface and show poor agreement with calculated dispersions for the proposed models for the 2×1 and c(4×2) reconstructions.

Cathodoluminescence measurements of suboxide band-tail and Si dangling bond states at ultrathin Si–SiO2 interfaces

We have directly observed cathodoluminescence (CL) in ultrahigh vacuum over a broad spectral range (0.7–4.0 eV) from ultrathin 5 nm layers of remote plasma enhanced chemical vapor deposition grown a-SiO2:H deposited on silicon substrates. In the infrared regime, luminescence is observed at 0.8 eV, consistent with the presence in the as-deposited film of Si dangling bond localized states, as well as at 1.1 eV due to band edge emission. In the optical regime, three peaks (1.9, 2.7, and 3.4 eV) are observed showing evidence for band tail state emission from an amorphous silicon–oxygen bonded suboxide region in the film, with smaller contributions from either substrate related c-Si or defect containing, stoichiometric SiO2. CL spectra with a range of beam voltages indicate that the stoichiometric SiO2 is very close to the free surface of the film, possibly due to oxidation of the air-exposed wafer or due to a nonuniformity in the film. When the films are annealed in situ in stages up to 500 °C, we observe no change in the shape of the a-SiOx:H peak at 1.9 eV, showing the stability of this suboxide to such temperatures. These observations are consistent with CL measurements of thicker films of a-SiO2:H and a-SiOx, and demonstrate the utility of CL spectroscopy for the study of ultrathin dielectric studies.

Gap-State Distributions in Hydrogenated Amorphous Silicon-Germanium Evaluated Using Capacitance–Voltage Method

The density of gap states in hydrogenated amorphous silicon-germanium (a-SiGe:H) alloys deposited at temperatures of 150–250°C has been evaluated by using a capacitance–voltage (C–V ) method. It was found that the gap-state distribution for a-SiGe:H deposited at 150°C has a broad bump around 0.6 eV below the conduction band edge Ec which is associated with Ge dangling bonds. In contrast, a 250°C-deposited film exhibits a peak density of 8×1018 cm-3·eV-1 around 0.75 eV below the conduction band edge Ec, which might originate from Ge and/or Si dangling bond states. After AM1 (200 mW/cm2) light soaking, the gap-state density above the midgap is increased for films deposited below 200°C, while at 250°C deposition, defect states near the midgap are increased.

Interface state-induced shift of the oxide and semiconductor core levels for metal–oxide–semiconductor devices

Measurements of x-ray photoelectron spectra are performed for ∼3-nm-thick Pt/∼3.6-nm-thick silicon oxide/n-Si(100) devices under biases between the Pt layer and the Si substrate. It is observed that the oxide Si 2p peak as well as the substrate peaks is shifted upon applying biases. These shifts are caused by a bias-induced change of the potential drop across the oxide layer due to the change in the amount of the interface state charge. The amount of the shift of the oxide Si 2p peak is well correlated to that of the substrate Si 2p3/2 peak. The energy distribution of the interface states is obtained by analyzing the amount of the shift of the substrate Si 2p3/2 peak measured as a function of the bias voltage. The interface state spectrum has one peak near the midgap, and the peak is attributed to isolated Si dangling bond states.

Characterization of Localized Density of States in Intrinsic a-Si and poly-Si Films by Transient Voltage Spectroscopy

ABSTRACTThe localized density of states (LDOS) at interfaces between intrinsic silicon and silicon nitride (Si3N4 films are studied using transient voltage spectroscopy (TVS). In the TVS technique, the transient of the voltage across a MIS-diode after a trap filling voltage pulse is measured using a high-impedance voltage probe. This allows us to make a precise measurement of the LDOS at undoped Si/insulator interfaces. The LDOS in a-Si:H/Si3N4systems has a broad peak around the energy of 0.9 eV below the conduction-band edge. A modification of the LDOS at a-Si:H/Si3N4 interfaces by bias-annealing is clearly observed using this technique. The results are consistent with the defect pool model. The LDOS in laser annealed poly-Si/Si3N4 systems has a peak centered 0.6eV below the conduction-band edge, which seems to be the Si dangling bond states in the poly-Si films.

Properties of hydrogenated amorphous silicon prepared by alternatively repeating chemical-vapor deposition from disilane and hydrogen plasma treatment

Experimental studies have been carried out to characterize hydrogenated amorphous silicon prepared by alternatively repeating chemical-vapor deposition (CVD) from disilane and hydrogen plasma (HP) treatment (referred to as HP treated CVD a-Si:H). It has been found that hydrogen plasma treatment induces structural relaxation of Si network in addition to the passivation of Si dangling bonds. Two types of defect states with different Gaussian energy distribution exist in this material, and this type of distribution has been found to be common in undoped a-Si:H prepared by a variety of methods. High film quality, i.e., a low defect density (<1016 cm−3) and a fairly long ambipolar diffusion length (0.12 μm), and significant reduction of light-induced changes have been simultaneously achieved in HP treated CVD a-Si:H. Steady state transport of electrons and holes under illumination are controlled by the states other than neutral Si dangling bond state (D0). The behavior of these states with long exposure to light is different from that of D0.

Surface states and Fermi-level pinning at epitaxial Pb/Si(111) surfaces.

The difference in Schottky-barrier height of epitaxial Si(111) (\ensuremath{\surd}3 \ifmmode\times\else\texttimes\fi{} \ensuremath{\surd}3 )R30\ifmmode^\circ\else\textdegree\fi{}-Pb(\ensuremath{\beta}) and Si(111)(7\ifmmode\times\else\texttimes\fi{}7)-Pb interfaces, supports the view that the Schottky-barrier height at these interfaces is not determined by bulk properties of the metal and the semiconductor, but that it depends on the local geometry and electronic structure at the interface. To elucidate the relation between the interface electronic structure and the Schottky-barrier height, we performed an angle-resolved photoemission study on the Si(111)(7\ifmmode\times\else\texttimes\fi{}7)-Pb and the Si(111)(\ensuremath{\surd}3 \ifmmode\times\else\texttimes\fi{} \ensuremath{\surd}3 )R30\ifmmode^\circ\else\textdegree\fi{}-Pb(\ensuremath{\beta}) surfaces. The electronic structures of these two surfaces are rather similar. Two different surface-state bands were resolved. One of them is fully occupied and is situated below the valence-band maximum (VBM). This state is interpreted as a Si dangling-bond state that is hybridized with Pb 6${\mathit{p}}_{\mathit{x}}$,${\mathit{p}}_{\mathit{y}}$ orbitals. The other state pins the Fermi level at approximately 0.1 eV above the VBM. This state has no measurable dispersion and appears in regions of k space where a common gap is present in the projected band structure of Pb and Si. Therefore, this surface state may become a true interface state at thick overlayers, pinning the Fermi level of the Si(111)(\ensuremath{\surd}3 \ifmmode\times\else\texttimes\fi{} \ensuremath{\surd}3 )R30\ifmmode^\circ\else\textdegree\fi{}-Pb(\ensuremath{\beta}) Schottky diodes near the bottom of the energy gap.

Energy Differences Between the Si and the Ge Dangling Bond Defects in a-Si1-xGex Alloys

ABSTRACTWe have investigated the difference in the electronic energies of neutral Si and Ge dangling bond states in undoped a-Si1-xGex alloys as a function of the alloy composition, x, and local bond-angle distortions. The local density of states, LDOS, in a-Si1-xGex alloys has been calculated using nearest-neighbor interactions, and employing the Cluster Bethe Lattice method. We conclude that for ideal, tetrahedrally bonded amorphous semiconductors alloys, the Ge dangling bond energy is lower than that of Si dangling bonds by ∼ 0.13 eV, independent of the specific nearest neighbors to the dangling bond (3 Si-atoms, 2 Si-atoms and 1 Ge-atom, etc.), but that the spread in dangling bond energies associated bond-angle variations of the order of 6–8 degrees can be larger than this energy difference (∼0.3 eV or greater). This means that structural disorder, rather than chemical disorder causes Si and Ge-atom dangling bond states to overlap in their energy distributions.

A model for ac conduction in amorphous silicon nitrides

The frequency-dependent conductivity of amorphous silicon nitride (a-Si1−xNx) films has been reported for the first time. The origin of the ac loss is discussed in terms of nitrogen-related two-level systems and electron transfer between Si dangling bond states.

Evidence for the defect pool concept for Si dangling bond states in a-Si:H from experiments with thin film transistors

We compare the characteristics of a-Si:H TFTs made using silicon nitride and silicon oxide gate insulators, and then determine the energy distribution of states in the amorphous silicon band -gap. We also subject the transistors to positive and negative bias-stress, which creates new states in the gap. We find the majority of intrinsic deep states occur at a higher energy for oxide transistors than for nitride transistors. We also find that the metastable states induced by positive bias-stress occur at a higher energy than those induced by negative bias-stress. The results are consistent with the defect pool model for the Si dangling bond states.

Au adsorbed Si(111) surfaces studied by angle-resolved electron-energy-loss spectroscopy

Electronic structure was investigated by using AR-EELS on some different stages of Au on Si(111): the 5 × 1- and the √3-Au surfaces. The loss peak at 2.2 eV on 5 × 1 was ascribed to the bare Si dangling bond state. The electronic structure of 5 × 1-Au was consistent with the recent structural model in which Au atoms are embedded. Dispersion of surface plasmons on the √3-Au surface was observed.