Surface Space Charge Layer Research Articles

In situ Hall measurements of Si(1 1 1)/Cr, Si(1 1 1)/Fe and Si(1 1 1)Mg disordered systems at submonolayer coverages

In situ Hall measurements of submonolayer and monolayer Si(111)/Fe, Si(111)/Cr and Si(111)/Mg disordered systems are presented. It was shown that for understanding conductivity mechanism in a system adsorbed metal (Fe, Cr, Mg)/Si(111) 7×7, it is necessary to take into account three effects: (a) destruction of the superstructure Si(111) 7×7; (b) formation of donor-type surface states of Fe (Cr) in the band gap of silicon and (c) growth mode of the silicide or metal (Mg) clusters on Si surface. Destruction of the localized surface states of superstructure Si(111) 7×7 plays the greatest role, if the density of donor-type surface states is very small. It corresponds to adsorption of Cr and Mg on Si(111) of p-type. The greatest contribution of donor-type surface states to the conductivity of surface space-charge layer is observed with adsorption of Fe on Si(111) of n-type. Conductivity through the adsorbed iron (chromium) layer begins with thickness more than three monolayers and is supported by electrons.

In Situ Hall Measurements of Fe and Cr Submonolayers on Si(111) of n- and p-Type Conductivity

By the method of in situ Hall effect measurements at room temperature, it was shown that the formation of a clean Si(111)(7×7) surface at high temperature annealing (T=1250°, t=120–180 s) of n-type conductivity silicon results in a change of the majority carriers on a surface. But on a surface of p-type conductivity silicon it results in the formation of a hole accumulated layer. Various dynamics of changes in the Hall voltage and resistivity voltage within the limits of the first monolayer of room temperature adsorption of iron and chromium on substrates with a p–n junction and with a hole accumulated layer are not connected with conductivity along an absorbed layer. To understand the mechanism of conductivity in a system adsorbed layer Fe(Cr)/Si(111)(7×7), it is necessary to take into account two effects: (a) destruction of the superstructure Si(111)(7×7); (b) formation of donor-type surface states of Fe(Cr) in the band gap of silicon. Destruction of the localized surface states of the superstructure Si(111)(7×7) plays the greatest role, when the density of donor-type surface states is very small. It corresponds to adsorption of Cr on Si(111) of p-type conductivity. The greatest contribution of donor-type surface states to the conductivity of the surface space-charge layer was observed with adsorption of Fe on Si(111) of n-type conductivity. Conductivity through the adsorbed iron (chromium) layer began with a thickness more than three monolayers.

IN SITU HALL MEASUREMENTS OF Fe AND Cr SUBMONOLAYERS ON Si(111) OF n- AND p-TYPE CONDUCTIVITY

By the method of in situ Hall effect measurements at room temperature, it was shown that the formation of a clean Si (111)(7×7) surface at high temperature annealing (T=1250°, t=120–180 s) of n-type conductivity silicon results in a change of the majority carriers on a surface. But on a surface of p-type conductivity silicon it results in the formation of a hole accumulated layer. Various dynamics of changes in the Hall voltage and resistivity voltage within the limits of the first monolayer of room temperature adsorption of iron and chromium on substrates with a p–n junction and with a hole accumulated layer are not connected with conductivity along an absorbed layer. To understand the mechanism of conductivity in a system adsorbed layer Fe(Cr)/ Si (111)(7×7), it is necessary to take into account two effects: (a) destruction of the superstructure Si (111)(7×7); (b) formation of donor-type surface states of Fe(Cr) in the band gap of silicon. Destruction of the localized surface states of the superstructure Si (111)(7×7) plays the greatest role, when the density of donor-type surface states is very small. It corresponds to adsorption of Cr on Si (111) of p-type conductivity. The greatest contribution of donor-type surface states to the conductivity of the surface space-charge layer was observed with adsorption of Fe on Si (111) of n-type conductivity. Conductivity through the adsorbed iron (chromium) layer began with a thickness more than three monolayers.

Surface-state electrical conduction on the [formula omitted] surface with noble-metal adatoms

The Si(111)- 21 × 21 - (Ag+Cu) superstructure was found to be induced by adsorption of about 0.14 monolayers (ML) of copper atoms on the Si(111)- 3 × 3 - Ag surface at room temperature (RT), which was quite similar to the cases of the 21 × 21 superstructures induced by gold adsorption at RT and by silver adsorption at low temperature on the 3 × 3 - Ag surface. Photoemission spectroscopies showed that the high electrical conductance observed for this copper-induced superstructure was due to a two-dimensional intrinsic dispersive surface-state band crossing the Fermi level, not due to the surface space-charge layer, which is again similar to the gold-induced 21 × 21 surface. Each noble-metal adatom (gold, silver, copper) of less than ca. 0.1 ML coverage on the 3 × 3 - Ag surface was found to exist commonly as two-dimensional adatom gas (2DAG) phase before they turned into the 21 × 21 structure. The enhancement in surface conductance by the 2DAG phase was concluded to result from the donation of the adatoms' valence electrons into the surface-state band of the 3 × 3 - Ag substrate. The difference in stability of the 21 × 21 superstructures among gold-, silver- and copper-induced ones is discussed in terms of the first atomic ionization energy of the adatoms.

Structure and electrical conductance of Pb-covered Si(111) surfaces

A dense commensurate )3) phase was obtained by additional 0.7 ML Pb adsorption onto the Si~111!-)3)-Pb surface ~ 1 3 ML Pb! at room temperature ~RT!. Its electronic structure was similar to that of the hexagonal and striped incommensurate ~HIC and SIC! phases. A high surface electrical conductance observed for this phase was attributed partly to an increase of carrier concentration in the surface space-charge layer, and more importantly to the formation of a metallic surface state. The HIC phase, on the other hand, was found to convert to the SIC phase with only 0.08 ML additional Pb adsorption at RT, which was accompanied with a steep increase in conductance. This was due to an increase of carrier concentration in the metallic surface state, and also possibly due to an enhancement of its carrier mobility. Both enhancements are suggested to come from the reduced domain-wall density in the SIC phase compared with the HIC phase. @S0163-1829~99!09731-3#

Photoconductivity of the Si(111)-7×7 and [formula omitted]-Ag surfaces

The photoconductivity of a Si(111) wafer was found to strongly depend on the surface superstructures on it. Positive photoconductivities were measured on the clean 7×7 surface with photon energies larger than the band gap energy, which was attributed to the creation of excess electron-hole pairs in the bulk. On the 3 × 3 -Ag surface, however, negative photoconductivities were measured with light of the same energy range. This phenomenon was qualitatively explained as a result of a reduction of the conduction electrons in a surface-state band, which was due to an enhanced recombination rate between the surface-state electrons and the excess holes created in the surface space-charge layer by illumination.

Structure-dependent electrical conduction through indium atomic layers on the Si(111) surface

We have characterized the changes in surface electrical conductance induced by additional indium depositions onto the Si (111)-7×7 clean and the Si(111)- 3 × 3 -In surfaces at various substrate temperatures by combining in situ electron diffraction and photoemission spectroscopy measurements. On the 7×7 surface, two-dimensional percolation conduction among In microcrystals densely grown on the surface set on around 2.3 ML (monolayer) coverage only below 160 K without any changes in surface structures, while no conductance increase was observed above 160 K up to 40 ML coverage for a lack of percolating paths among sparsely distributed In crystals. On the 3 × 3 -In surface at room temperature, a drastic increase in conductance was observed accompanying the successive structural changes of 3 × 3 →2×2→1×1→ 7 × 3 →1×1 structures with an increase of In coverage. This increase was found to be mainly due to the conduction through the In atomic layers grown in a layer-by-layer mode up to 3 ML, though the conduction through the surface space-charge layer also partly contributed to the conductance increase. At lower temperatures, such structural transformations as well as conductance increases were suppressed.

Electronic structure of the Si(111)-21×21-(Ag+Au) surface

Angle-resolved ultraviolet and x-ray photoelectron spectroscopies were used to analyze the electronic band structure of the $\mathrm{Si}(111)\ensuremath{-}\sqrt{21}\ifmmode\times\else\texttimes\fi{}\sqrt{21}(R\ifmmode\pm\else\textpm\fi{}10.89\ifmmode^\circ\else\textdegree\fi{})\ensuremath{-}(\mathrm{A}\mathrm{g}+\mathrm{A}\mathrm{u})$ surface that was induced by Au adsorption of 0.19 atomic layer onto the Si(111)-$\sqrt{3}$\ifmmode\times\else\texttimes\fi{}$\sqrt{3}$-Ag surface at room temperature. We found two intrinsic dispersive surface-state bands crossing the Fermi level, which were considered to originate from an antibonding surface state of the initial $\sqrt{3}\ifmmode\times\else\texttimes\fi{}\sqrt{3}\ensuremath{-}\mathrm{Ag}$ structure. The electrons accumulated in these bands were found to be donated by Au adatoms. The electron transfer from Au adatoms into the substrate bulk (surface space-charge layer) was also confirmed by measuring the changes in band bending. The results seemed to be consistent with an atomic model in which the Au adatoms sit atop the Ag trimers of the $\sqrt{3}\ifmmode\times\else\texttimes\fi{}\sqrt{3}\ensuremath{-}\mathrm{Ag}$ framework. We proposed a kind of atomic bonding mechanism on this surface, referred to as ``parasitic surface bonding,'' where Au adatoms make metallic bonds via a surface-state band of the substrate surface.

Electrical conduction via surface-state bands

We will discuss a two-dimensional electron system originating from a surface-state band and especially its electronic transport properties. We have found a surface superstructure that has a highly conductive surface-state band, the Si(111)-√21 × √21R ± 10.89° structure, induced by adsorption of a 0.14-atomic-layer of Au or Ag onto a Si(111)-√3 × √3-Ag surface. Photoemission spectroscopies showed that this √21 × √21 structure had a highly dispersive surface-state band crossing over the Fermi level, while the surface space-charge layer was a depletion layer. It was then concluded that the observed excess surface conductance was due to the two-dimensional band of the surface electronic state.

Surface electrical conduction due to carrier doping into a surface-state band on Si(111)-3×3-Ag

Photoemission spectroscopy has shown that each Ag atom in its two-dimensional adatom gas (2DAG) phase deposited on the Si(111)-$\sqrt{3}\ifmmode\times\else\texttimes\fi{}\sqrt{3}$-Ag surface at room temperature donates one electron into an antibonding surface-state band of this substrate, resulting in a steep increase in electrical conductance through the band. The surface space-charge layer makes no contribution to the conductance increase by the 2DAG adsorption, estimated from the band-bending measurements. When the 2DAG nucleates into three-dimensional Ag microcrystals by further deposition beyond a critical supersaturation coverage, the carrier-doping effect vanishes, returning to a lower conductance. These results reveal that the surface state acts as a surface conduction band. The electron mobility in this band is estimated to be on the order of 10 cm${}^{2}$/V s.

Electrical functional properties of surface superstructures on semiconductors

We have found that the electrical conductivity parallel to the surface of a bulk silicon crystal is crucially influenced by the surface superstructures formed only on the topmost atomic layers of the surface. We have demonstrated this influence by depositing metal (Au, Ag, Cu, Pb, or In) on Si(111) surfaces that had a variety of surface superstructures such as Si(111)-7 × 7 clean, √3 × √3, 5 × 2, and so on. The increase in surface electrical conductance compared with the clean Si(111) surface were due either to conduction through the surface space-charge layer, to the two-dimensional surface-state bands, or to the conductive metal layers grown on the surface, depending on the combination of the deposit and the substrate-surface structure. This is the first investigation directly correlating the electrical conductivity to the surface structures. This result may lead to nanometer-scale electronic devices that use the electrical conduction restricted within a few atomic layers on the surface with controlled domains of surface superstructures.

Electrical conduction through the surface-state band of the Si(111)-√21 × √21-(Ag+Au) structure

The Si(111)-√21 × √21 superstructure, which was induced by 0.14 atomic layer Au adsorption onto the Si(111)-√3 × √3-Ag surface at room temperature, was found to have a very high surface electrical conductance, higher than that of the Si(111)-7 × 7 clean surface by about 3.1 × 10−4A/V. Photoemission spectroscopies showed that this √21 structure had a dispersive surface-state band crossing the Fermi level, while the surface space-charge layer was a depletion layer. It was thus concluded that the observed excess surface conductance was due to the two-dimensional band of the surface electronic state.

Surface conductivity for Au or Ag on Si(111).

The surface electrical conductivity was separated from the bulk one by simultaneous conductivity measurements of two different surface structures formed on a single Si-wafer surface in ultrahigh vacuum. We have found that the surface conductivities for the Si(111)-\ensuremath{\surd}3\ifmmode\times\else\texttimes\fi{}\ensuremath{\surd}3-Ag and -5\ifmmode\times\else\texttimes\fi{}2-Au superstructures are inherently higher than that of the Si(111)-7\ifmmode\times\else\texttimes\fi{}7 clean surface by (11.5\ifmmode\pm\else\textpm\fi{}0.5)\ifmmode\times\else\texttimes\fi{}${10}^{\mathrm{\ensuremath{-}}5}$ A/V and (5\ifmmode\pm\else\textpm\fi{}1)\ifmmode\times\else\texttimes\fi{}${10}^{\mathrm{\ensuremath{-}}5}$ A/V, respectively. These excess conductivities are estimated to originate mainly from the surface space-charge layer, although the surface-state-band conduction is considered to partly contribute, especially on the \ensuremath{\surd}3\ifmmode\times\else\texttimes\fi{}\ensuremath{\surd}3-Ag surface. \textcopyright{} 1996 The American Physical Society.

High Temperature I-V and C-V Characteristics of a Al/n-GaAs/In Schottky Barrier Type Device

AbstractAl/n-GaAs Schottky barrier type diode was fabricated by thermal deposition of Al on chemically etched polished surface of (100) n-GaAs at a pressure of 4×10−6 Torr. Back ohmic contact to n-GaAs was prepared by thermal deposition of In, followed by a 90 min. anneal in Ar atmosphere at 390° C. The C−2 vs V characteristics were quite linear for reverse bias voltages in the range 0-4 V. Over the temperature range 300–360 K, the values of the barrier height and the net carrier density (Nd-Na), obtained from the C−2-V data, were in the range 1.33–1.26 V and 4.3×l0−16-5.0×l0−16 cm−3, respectively. The forward I-V data over the temperature range 300–400 K, indicated that the electrical current across the Al/n-GaAs Schottky junction was transported by the mechanisms of generation-combination (GR) in the space charge, thermionic emission (TE) and ohmic leak current. A value of (1.20±0.04) V for the zero bias barrier height was deduced from the temperature dependence of the TE reverse saturation current. The barrier height deduced from the I-V data was practically independent of temperature. The 20% change in Nd-Na with temperature, obtained from the C-V data, and important contributions of the GR and leak currents to the total forward current, suggested the presence of defect levels in the surface space charge layer in n-GaAs, which may be responsible for the discrepancy in the values of the barrier height obtained from the I-V and C-V data.

Electronic properties of the space charge layer of the clean SnO 2(110) surface

Photoemission Yield Spectroscopy (PYS) associated with Auger Electron Spectroscopy (AES) and Low Energy Electron Diffraction (LEED) control have been used in the studies of the electronic properties of the surface space charge layer of the clean SnO 2(110) surface with (1 × 1) atomic structure obtained by ion bombardment with subsequent annealing at 900 K. The work function φ and absolute band bending eVs were 4.20±0.03 and −1.64±0.13 eV, respectively. Moreover, two filled electronic state bands localized in the band gap below the Fermi level E F were observed. Their origin has also been briefly discussed.

Study of the density of states in a-Si:H using the Si/electrolyte system

Localized states in a-Si:H are studied by pulsed measurements on the a-Si:H/electrolyte (S/E) system. The S/E interface is essentially blocking to current flow and, as a result, surface space-charge layers, ranging from large depletion to very strong accumulation conditions, can be induced and studied. Measurements in the depletion range under illumination yield directly the total density of occupied states in the entire energy gap. This is useful in obtaining a quick and reliable assessment of the quality of the amorphous films. In high-grade films we find that the total density of occupied states is around 10 18 cm −3. The data in the accumulation range, on the other hand, provide useful information on unoccupied states near the conduction band edge. The S/E system is utilized also to apply a sweep-out technique for an accurate determination of μτ, the product of the electron mobility and lifetime, even when this value is very low.

Some comments on the electronic properties of the surface space charge layer of copper phthalocyanine thin films

A simple analysis, using a theory of the surface space charge layer of semiconductors, of the published values of the work function φ and surface ionization energy Φ s of copper phthalocyanine (CuPc) thin films was performed. Using a well known position of the Fermi level E F within the band gap E g the values of its absolute band bending eV s and surface electron affinity X s were determined. A small negative value of the absolute band bending eV s = −0.17 ∓ 0.15 eV has been interpreted by the existence of the filled electronic surface states localized in the band gap below the Fermi level E F. Such states were predicted theoretically for thin films and the crystalline surface of CuPc, and attributed to surface lattice defects of a high concentration.

Study of a-Si:H Using the Solid/Electrolyte System

Pulsed measurements on the solid-electrolyte system, which proved very useful in the study of crystalline semiconductors, have been found to be equally effective when applied to hydrogenated amorphous Si films. Here, as well, the aSi:H/electrolyte interface is essentially blocking to current flow and, as a result, surface space-charge layers, ranging from large depletion to very strong accumulation conditions, can be induced and studied. In particular, valuable information can be gained on the density of the localized bulk states. Measurements in the depletion range under illumination yield directly the total density of occupied states in the entire energy gap. This is very useful in obtaining a quick and reliable assessment of the quality of the amorphous films. In high-grade films we find that the total density of occupied states is around 1018 cm−3. The data in the accumulation range, on the other hand, provide useful information on unoccupied states near the conduction band edge. The blocking nature of the amorphous Si/electrolyte interface is utilized also to apply a sweep-out technique for an accurate determination of μτ the product of the electron mobility and lifetime, even when this value is very low. In a rather poor-quality film, for example, we find μτ to be 5 × 10−8 cm2/V.

Characterization of MOVPE-grown GaAs layers by C-V analysis

A computer simulation study of the capacitance of a surface space charge layer in undoped n-GaAs grown by metalorganic vapour phase epitaxy is presented. The effect of the deep donor level EL2 on the surface capacitance of epilayers with an ideal free surface is estimated. In order to approach the as-grown layer surface the model used is extended considering MIS and Schottky-barrier structures and their voltage-capacitance curves are analysed. The theoretical C-V dependences are compared with experimental C-V curves of a real structure including N+-GaAs substrate, undoped n-GaAs epitaxial layer containing EL2 levels and thin native oxide. Conditions are determined at which the EL2 levels as well as the native oxide film may influence the capacitance characteristics.

Impact of illumination level and oxide parameters on Shockley–Read–Hall recombination at the Si-SiO2 interface

The experimentally observed dependence of effective surface recombination velocity Seff at the Si-SiO2 interface on light-induced minority carrier excess concentration is compared with theoretical predictions of an ‘‘extended Shockley–Read–Hall (SRH) formalism.’’ The calculations of SRH-recombination rates at the Si-SiO2 interface are based on the theory of a surface space charge layer under nonequilibrium conditions and take into account the impact of illumination level, gate metal work function, fixed oxide charge density, and the energy dependence of capture cross sections σn, σp and interface state density Dit. Applying this theory to p-type silicon surfaces covered by high quality thermal oxides, the experimentally observed strong increase of Seff with decreasing minority carrier excess concentration could quantitatively be attributed to the combined effect of the σn/σp ratio of about 1000 at midgap and the presence of a positive fixed oxide charge density Qf of about 1×1011 charges/cm2. Due to the favorable work function of aluminum, surface recombination velocities below 1 cm/s can be obtained at Al-covered Si-SiO2 interfaces for minority carrier densities above 1013 cm−3.