Si 2p Core Level Research Articles

Atomic layer deposition of nanolaminate oxide films on Si

Among the methods for depositing thin films, atomic layer deposition is unique for its capability of growing conformal thin films of compounds with a control of composition and thickness at the atomic level. The conformal growth of thin films can be of particular interest for covering nanostructures since it assures the homogeneous growth of the ALD film in all directions, independent of the position of the sample with respect to the incoming precursor flow. Here we describe the technique for growing the HfO2/Al2O3 bilayer on Si substrate and our in situ approach for its investigation by means of synchrotron radiation photoemission. In particular, we study the interface interactions between the two oxides for various thickness compositions ranging from 0.4 to 2.7 nm. We find that the ALD of HfO2 on Si induces the increase of the interfacial SiO2 layer, and a change in the band bending of Si. On the contrary, the ALD of Al2O3 on HfO2 shows negligible interaction between layers as the binding energies of Hf4f, Si2p, and O1s core level peaks and the valence band maximum of HfO2 do not change and the interfacial SiO2 does not increase.

High-temperature vesuvianite: crystal chemistry and surface considerations

A multi-methodical approach has been applied for characterizing the bulk and surface crystal chemical features of a high-temperature vesuvianite crystal from skarns of Mount Somma-Vesuvius Volcano (Naples, Italy). Vesuvianite belongs to the space group P4/nnc with unit cell parameters a = 15.633(1) A, c = 11.834(1) A and chemical formula (Ca18.858 Na0.028 Ba0.004 K0.006 Sr0.005 □0.098)19.000 (Al8.813 Ti0.037 Mg2.954 Mn0.008 Fe 0.114 2+ Fe 1.375 3+ Cr0.008 B0.202)13.511 Si18.000(O0.261 F0.940 OH7.799)9.000. Structure refinement, which converges at R = 0.0328, demonstrates a strong positional disorder down the fourfold axes, indicating that the Y1 site is split into two positions (Y1A and Y1B) alternatively occupied. However, because of X4 proximity to Y1B and Y1A, X4 cannot be occupied if Y1B or Y1A are. Overall Y1 occupancy (Y1A + Y1B) reaches approximately 0.5, as common in vesuvianite and occupancy of Y1B site is extremely limited. Moreover, T1 position, limitedly occupied, accommodates the excess of cations generally related to Y position. A small quantity (0.202 apfu) of boron is sited at the T2 site that, like T1, is poorly occupied. The determination of the amount of each element on the (100) vesuvianite surface, obtained through X-ray photoelectron spectroscopy high-resolution spectra in the region of the Si2p, Al2p, Mg1s, and Ca2p core levels, evidences that a greater amount of aluminum and a smaller amount of calcium characterize the surface with respect to the bulk. Although both of these features require further investigation, we may consider the Al increase can be related to preferential orientation of Al-rich sites on the (100) plane. Furthermore, the surface structure of vesuvianite suggests that Al, Ca, and Mg cations maintain coordination features at the surface similar to the bulk. Silica, however, while presenting fourfold coordination, shows also a [1]-fold small coordinated component at binding energy 99.85 eV, due to broken Si–O bonds at the surface. The presence of eight- and nine-coordinated Ca cations is suggested by a large XPS feature resulting from the merging of Ca2p3/2 and Ca2p1/2 peaks at 348.45 and 352.05 eV, respectively.

In situ study on the electronic structure of graphene grown on 6H–SiC [formula omitted] with synchrotron radiation photoelectron spectroscopy

Low energy electron diffraction (LEED) and synchrotron radiation photoelectron spectroscopy (SRPES) were used to study the synthesis process and detailed electronic structures of graphene produced by thermal decomposition of 6H–SiC (0 0 0 1¯) in ultrahigh vacuum (UHV). The LEED results showed that as annealing temperature increased, different reconstructions of 6H–SiC (0 0 0 1¯) appeared and the anisotropic graphene layers were produced finally. The results of C 1s core levels indicated that the component, which was assigned to graphene emerged and the interface interaction between graphene and the substrate was weak. The existence of Si clusters was confirmed by Si 2p core levels and perhaps these clusters led to rough epitaxial graphene surface. The results of valence band spectra showed that at high annealing temperature both of σ and π bonds existed in the grown graphene layers.

Bridging, non-bridging and free (O 2–) oxygen in Na 2O-SiO 2 glasses: An X-ray Photoelectron Spectroscopic (XPS) and Nuclear Magnetic Resonance (NMR) study

High resolution Na 1s, O 1s and Si 2p core level XPS spectra of six Na 2O–SiO 2 glasses ranging in composition from 100 to 45 mol % SiO 2 have been collected using the Kratos Ultra Axis instrument with its unique charge compensation system. The O 1s spectra for the glasses are well resolved so that bridging oxygen (BO, Si– O–Si) and non-bridging oxygen (NBO, Na– O–Si) signals can be accurately fitted and quantified without resorting to constraints or assumptions. The same samples were analysed by 29Si MAS NMR to obtain Q-species abundances from which BO and NBO proportions were calculated. Similar BO:NBO ratios were obtained by both methods over the entire compositional range studied. They are also consistent with most previous XPS and NMR results for glasses containing more than ~ 65 mol % SiO 2. Our XPS and NMR experimental results, however, differ somewhat from previously published XPS and NMR results for glasses containing less than about 65 mol % SiO 2. Na is mobile in the X-ray beam and mobility causes BO:NBO ratios to increase with time of exposure. Na mobility here has been circumvented to yield reliable BO:NBO ratios of the glasses. The ratios are lower than previously reported in XPS studies and are similar to ratios obtained from our 29Si MAS NMR results on the same glasses. The XPS and 29Si MAS NMR results also indicate the presence of a third oxygen species in sodic glasses. As has been proposed for CaSiO 3 glass and for sodic and potassic glasses containing La, we suggest that O 2− is present in sodic glasses at small concentrations. The O 2− content correlates with increased soda content and may be associated with, and instrumental in development of, three dimensional percolation channels in the glasses. The XPS O 1s line width of the BO peak is broader than the NBO peak, indicating more than one contribution to the BO peak. As observed in crystalline Na metasilicate and Na disilicate, BO of Na-silicate glasses may be of two types, one arising from BO bridging two Si atoms, and the second BO signal arising from BO bonded not only to two Si atoms but also to Na.

High oxidation state at the epitaxial interface of γ-Al2O3 thin films grown on Si(111) and Si(001)

High resolution synchrotron radiation x-ray photoelectron spectroscopy allowed us to identify the chemical bonding at the interface between epitaxial γ-Al2O3 and Si substrate. The experiments were performed on 1 nm thick epitaxial γ-Al2O3 layers grown on both Si(111) and Si(001) substrates. In both cases, the Si 2p core level decomposition recorded at photon energy of 160 eV provided evidence for the absence of Si2+ and Si3+ species and the presence of two different Si4+ species. A microscopic model is proposed for the interface obtained with two incomplete SiO2 planes based on the Si 2p3/2 line shape.

Electronic and morphological properties of the electrochemically prepared step bunched silicon (111) surface

AbstractTopographical and electronic and properties of step bunched Si(111), prepared by electrochemical processing in alkaline solution, are analyzed. Tapping mode atomic force microscopy (TM AFM) analysis shows that one bunched step consists of about 15 atomic steps (each 0.314 nm in height) and that the (111) oriented terraces have widths that range from 150 to 250 nm. Scanning tunneling microscopy (STM) experiments show a corrugation of the (111) terraces with an rms roughness of 0.5–0.8 nm, correlated with etch pits in alkaline solution. Low energy electron diffraction (LEED) data show a splitting of the (10) and (01) spot from which a minimum terrace width of 4.8 nm have been calculated in good agreement with the TM AFM data. Kelvin probe force microscopy (KPFM) experiments show a decrease of the contact potential difference (CPD) at and near the edges of steps indicating a more negatively charged surface area. Synchrotron radiation photoelectron spectroscopy (SRPES) on electrochemically and purely chemically prepared step bunched surfaces is compared. From the Si 2p core level shift, and, in particular, from the onset of the valence band emission, an accumulation layer‐type shift is observed on the electrochemically prepared sample that is absent for chemical preparation. The move of the Fermi level toward the conduction band minimum of the electrochemically conditioned samples is interpreted by H incorporation and discussed by a doping model that involves the mechanism of hydrogen evolution.

Synthesis of water-dispersible photoluminescent silicon nanoparticles and their use in biological fluorescent imaging

Water-dispersible silicon nanoparticles (Si-NPs) are desirable for applications in biological techniques. A simplified method to synthesize such particles is reported here. The resulting Si-NPs are water-dispersible and luminescent. Under the excitation of UV light, the Si-NPs emit strong red light with a peak maximum at 606 nm and a quantum yield of 6%. They are highly stable, and remain so over several weeks. Fourier Transform Infrared (FTIR) spectroscopy shows a visible Si–CH2 scissoring vibration mode. Furthermore, the surface chemical bondings were confirmed by X-ray photoelectron spectroscopy (XPS). In the Si2p and C1s core levels, Si–C components are observed. The diameters of the synthesized Si-NPS as measured by atomic force microscope (AFM) are approximately 5 nm. Furthermore, the nanoparticles can be taken up by cultured cells. Fluorescence images of Si-NPs within MCF-7 human breast cancer cells show they are distributed throughout the cell tissue.

Crystal chemistry, surface morphology and X-ray photoelectron spectroscopy of Fe-rich osumilite from Mt. Arci, Sardinia (Italy)

Microprobe analysis, single crystal X-ray diffraction, X-ray photoelectron spectroscopy, atomic force microscopy, and X-ray absorption spectroscopy were applied on Fe-rich osumilite from the volcanic massif of Mt. Arci, Sardinia, Italy. Osumilite belongs to the space group P6/mcc with unit cell parameters a = 10.1550(6), c = 14.306(1) A and chemical formula (K0.729)C (Na0.029)B (Si10.498 Al1.502)T1 (Al2.706 Fe 0.294 2+ )T2 (Mg0.735 Mn0.091 Fe 1.184 2+ )AO30. Structure refinement converged at R = 0.0201. Unit cell parameter a is related to octahedral edge length as well as to Fe2+ content, unlike the c parameter which does not seem to be affected by chemical composition. The determination of the amount of each element on the mineral surface, obtained through X-ray photoelectron spectroscopy high-resolution spectra in the region of the Si2p, Al2p, Mg1s and Fe2p core levels, suggests that Fe presents Fe2+ oxidation state and octahedral coordination. Two peaks at 103.1 and 100.6 eV can be related to Si4+ and Si1+ components, respectively, both in tetrahedral coordination. The binding energy of Al2p, at 74.5 eV, indicates that Al is mostly present in the distorted T2 site, whereas the Mg peak at 1,305.2 eV suggests that this cation is located at the octahedral site. X-ray absorption at the Fe L2,3-edges confirms that iron is present in the mineral structure, prevalently in the divalent state and at the A octahedral site.

Valence band offset and interface stoichiometry at epitaxial Si3N4/Si(111) heterojunctions formed by plasma nitridation

Ultrathin β-Si3N4(0001) epitaxial films formed by N2-plasma nitridation of Si(111) substrates have been studied by photoelectron spectroscopy using synchrotron radiation. The valence band offset at the β-Si3N4/Si interface was determined by valence-band photoelectron spectra to be 1.8 eV. Furthermore, the Si 2p core-level emissions were analyzed for nitride (Si4+) and subnitride (Si3+ and Si+) components to characterize the interface stoichiometry. In contrast to the interfaces formed by ammonia thermal nitridation and N2-plasma nitridation at room temperature, the interface formed by N2-plasma nitridation at high substrate temperature is very close to subnitride free with an abrupt composition transition.

X-ray diffraction and photoelectron spectroscopy study of swift heavy ion irradiated Mn/p-Si structure

The electronic structure and interfacial chemistry of thin manganese films on p-Si (1 0 0) have been studied by photoelectron spectroscopy measurements using synchrotron radiation of 134 eV and from X-ray diffraction data. The Mn/p-Si structures have been irradiated from swift heavy ions (∼100 MeV) of Fe 7+ with a fluence of 1 × 10 14 ions/cm 2. Evolution of valence band spectrum with a sharp Fermi edge has been obtained. The observed Mn 3d peak has been related to the bonding of Mn 3d–Si 3sp states. Mn 3p (46.4 eV), Mn 3s (81.4 eV) and Si 2p (99.5 eV) core levels have also been observed which show a binding energy shift towards lower side as compared to their corresponding elemental values. From the photoelectron spectroscopic and X-ray diffraction results, Mn 5Si 3 metallic phase of manganese silicide has been found. The silicide phase has been found to grow on the irradiation.

Selective Reactions and Adsorption Geometries of a Multifunctional Molecule: cis-2-Butene-1,4-diol on Si(100)-2 × 1

The adsorption geometry of cis-2-butene-1,4-diol (BEDO, HOCH(2)CH=CHCH(2)OH) on Si(100)-2 x 1 was studied using scanning tunneling microscopy (STM), high resolution X-ray photoemission spectroscopy (XPS), and near-edge X-ray absorption fine structure (NEXAFS) spectroscopy. Bias-voltage-dependent STM images exhibited features characteristic of two distinct BEDO adsorption geometries. One feature was a bright protrusion located on the center of a single dimer, indicating an on-top configuration. The low bias-voltage STM image clearly showed dark features indicative of the formation of Si-H bonds on adjacent dimers in the same dimer row. The other feature was a bright protrusion bridged on end between two adjacent dimers in the same dimer row, indicating an end-bridge configuration. Accompanying this feature, two dark features attributed to Si-H bonds were observed on opposite positions to the bridged protrusion. From the XPS results, the Si 2p core level spectra revealed that the dimer atoms are involved in the formation of Si-O and Si-H bonds. On the other hand, carbon K-edge NEXAFS spectra showed that the C=C bond does not participate in the adsorption reaction and remains as an unreacted group. Collectively, the experimental results indicate that the adsorption of BEDO on Si(100)-2 x 1 occurs through the formation of two Si-O bonds via nucleophilic reaction between the two OH groups of BEDO and two Si-Si dimers. Importantly, the maintenance of the C=C bond means that the C=C functional group can be utilized as a new reaction site for further surface chemical reactions.

Hydrogen desorption kinetics and band bending for 6H–SiC(0 0 0 1) surfaces

The desorption kinetics of hydrogen from polished 6H–SiC(0 0 0 1) surfaces exposed to various sources of hydrogen have been determined using temperature programmed desorption (TPD). For (3 × 3) 6H–SiC(0 0 0 1) surfaces prepared via annealing and cooling in SiH 4, desorption of 0.2 ± 0.05 monolayer of molecular hydrogen was observed to occur at ≈590 °C. This β 1 H 2 desorption peak exhibited second order kinetics with an activation energy of 2.4 ± 0.2 eV. For (3 × 3) 6H–SiC surfaces exposed to atomic hydrogen generated via either a hot rhenium filament or remote hydrogen plasma, low energy electron diffraction patterns showed an eventual conversion back to (1 × 1) symmetry. Spectra acquired using Auger electron and X-ray photoelectron spectroscopies revealed that the atomic hydrogen exposure removed the excess Si. Photoelectron spectroscopy results also showed a 0.5 eV increase in binding energy for the Si2p and C1s core levels after removal of the Si–Si bilayer that is indicative of a decrease in band bending at the SiC surface. TPD from the (3 × 3) 6H–SiC(0 0 0 1) surfaces exposed to atomic hydrogen showed substantially more molecular hydrogen desorption (1–2 ML) through the appearance of a new desorption peak (β 2,3) that started at ≈200 °C. The β 2,3 peak exhibited second order desorption kinetics and a much lower activation energy of 0.6 ± 0.2 eV. A third smaller hydrogen desorption state was also detected in the 650–850 °C range. This last feature could be resolved into two separate desorption peaks (α 1 and α 2) both of which exhibited second order kinetics with activation energies of 4.15 ± 0.15 and 4.3 ± 0.15 eV, respectively. Based on comparisons to hydrogen desorption from Si and diamond surfaces, the β and α desorption peaks were assigned to hydrogen desorption from Si and C sites, respectively.

High resolution photoemission study of SiOx/Si(111) interface disruption following in situ HfO2 deposition

We report on an in situ high resolution core level photoemission study of the early stages of interface formation between an ultrathin SiOx layer (∼0.3 nm) grown on the atomically clean Si(111) surface and a HfO2 dielectric layer. Si 2p core level spectra acquired at 130 eV photon energy reveal evidence of a chemically shifted component on the lower binding energy side of the substrate peak which is attributed to interface defect states resulting from the incorporation of silicon atoms from the substrate into the interfacial oxide at room temperature. This evidence of Si/SiOx interface disruption would be expected to increase charge carrier scattering mechanisms in the silicon and contribute to the generally observed mobility degradation in high-k stacks with ultrathin silicon oxide interface layers.

Spatially resolved, energy-filtered imaging of core level and valence band photoemission ofhighly p and n doped silicon patterns

An accurate description of spatial variations in the energy levels of patternedsemiconductor substrates on the micron and sub-micron scale as a function of local dopingis an important technological challenge for the microelectronics industry. Spatially resolvedsurface analysis by photoelectron spectromicroscopy can provide an invaluable contributionthanks to the relatively non-destructive, quantitative analysis. We present results on highlydoped n and p type patterns on, respectively, p and n type silicon substrates. Usingsynchrotron radiation and spherical aberration-corrected energy filtering, we have obtaineda spectroscopic image series at the Si 2p core level and across the valence band. Local bandalignments are extracted, accounting for doping, band bending and surface photovoltage.

Core level photoemission and STM characterization of Ta/Si(1 1 1)-7 × 7 interfaces

We present the results of scanning tunneling microscopy (STM) and photoemission spectroscopy (PES) of the Ta/Si(1 1 1)-7 × 7 system after deposition of Ta at substrate temperatures from 300 to 1250 K. The coverage of Ta varied from 0.05 up to 2.5 of a monolayer (ML). STM shows that at 300 K and coverage less than 1 ML, a disordered chemisorbed phase is formed. Deposition on a hot surface (above 500 K) produces round 3D clusters randomly distributed on the surface. Cluster height and their diameter are found to change drastically with annealing temperature and the Ta coverage. Analysis of photoemission data of the Si 2p core levels shows that at room temperature and at coverage ⩽1 ML core level binding energy shifts and intensity variations of Si surface related components are observed, which clearly indicate that the reaction starts already at 300 K. Shifts in the binding energy, changes of the peak shapes and intensity of the Ta 4f doublet at higher temperatures can be explained by the formation of stable silicide on the surface.

XPS Study and Optical Properties of Si Films Electrodeposited in a Room-Temperature Ionic Liquid

The binding state of Si in films electrodeposited in trimethyl-n-hexylammonium bis(trifluoromethylsulfonyl)imide containing 0.1 mol L−1 SiCl4 was investigated by means of X-ray photoelectron spectroscopy (XPS). The bombardment of Ar+ mainly causes the transition of spectrum shape from bimodal to broad in XPS spectra of Si 2p core level when Ar+ etching is conducted. Angle-resolved XPS confirmed that the bimodal spectrum for the electrodeposited film surface is due to the oxidation in the air. The electrodeposited Si film was also characterized by reflectance and photoluminescence (PL) measurements. These measurements demonstrated that the electrodeposited film contains amorphous Si with the band gap of about 1.55 eV and a broad PL peak centered at the photon energy of 1.37 eV.

Dissociation dynamics of anionic and excited neutral fragments of gaseous SiCl4 following Cl 2p and Si 2p core-level excitations

The state-selective dissociation dynamics for anionic and excited neutral fragments of gaseous SiCl4 following Cl 2p and Si 2p core-level excitations were characterized by combining measurements of the photon-induced anionic dissociation, x-ray absorption and UV/visible dispersed fluorescence. The transitions of core electrons to high Rydberg states/doubly excited states in the vicinity of both Si 2p and Cl 2p ionization thresholds of gaseous SiCl4 lead to a remarkably enhanced production of anionic, Si− and Cl−, fragments and excited neutral atomic, Si*, fragments. This enhancement via core-level excitation near the ionization threshold of gaseous SiCl4 is explained in terms of the contributions from the Auger decay of doubly excited states, shake-modified resonant Auger decay, or/and post-collision interaction. These complementary results provide insight into the state-selective anionic and excited neutral fragmentation of gaseous molecules via core-level excitation.

The initial atomic layer deposition of HfO2∕Si(001) as followed in situ by synchrotron radiation photoelectron spectroscopy

We have grown HfO2 on Si(001) by atomic layer deposition (ALD) using HfCl4 and H2O as precursors. The early stages of the ALD were investigated with high-resolution photoelectron spectroscopy and x-ray absorption spectroscopy. We observed the changes occurring in the Si2p, O1s, Hf4f, Hf4d, and Cl2p core level lines after each ALD cycle up to the complete formation of two layers of HfO2. From the analysis of those variations, we deduced the growth properties of HfO2. The first layer consists of a sparse and Cl-contaminated oxide because of the incomplete oxidation, and the second layer is denser than the first one and with an almost stoichiometric O∕Hf ratio. At the completion of the second layer, the x-ray absorption spectra revealed the change of the Hf-oxide chemical state due to the transition from the thin Hf-oxide to the bulklike HfO2.

Structures of clean and oxygen-adsorbed SiC(0 0 0 1)-(3 × 3) surfaces

The structures of clean and oxygen-adsorbed SiC(0 0 0 1)-(3 × 3) surfaces were investigated by high-resolution medium energy ion scattering (MEIS) and photoelectron spectroscopy (PES). The high-resolution MEIS directly evidenced the Si-adatom/trimer/adlayer structure proposed by Starke et al. [U. Starke, J. Schardt, J. Bernhardt, M. Franke, K. Reuter, H. Wedler, K. Heinz, J. Furthmüller, P. Köchkel, F. Bechstedt, Phys. Rev. Lett. 80 (1998) 758]. For the surfaces exposed to dry O 2 gas at room temperature, only Si 4+ and Si + states were observed in the Si 2p core level spectra at an exposure above 10 L (1 L = 1.0 × 10 –6 Torr s) and the ratio of Si 4+ atoms to Si + atoms is almost constant and estimated to be ∼1/3. The MEIS analysis demonstrated that the amount of oxygen adsorbed is increased steeply with increasing O 2-exposure up to 10 L and saturated at ∼50 L with an oxygen coverage of 0.40 ± 0.05 ML (1 ML = 1.22 × 10 15 atoms/cm 2). The above results indicate that the Si-tetramer(adatom/trimer) making the (3 × 3) periodicity is preferentially oxidized to form a SiO 4-like structure. Such an initial oxidation is intimately related to the reconstructed surface structure and different from those of the SiC(0 0 0 1)- 3 × 3 and Si-rich SiC( 0 0 0 1 ¯ )-3 × 3 surfaces, for which all the oxidation states emerge in the former case and Si +, Si 2+ and Si 3+ states in the latter.

Chromium silicide formation by argon irradiation of Cr/Si bilayers

Chromium/silicon interfaces produced by Cr thin film deposition on Si substrates were ion-beam mixed using Ar+ ions at 3 keV. The ion-beam mixing (IBM) has been studied by means of x-ray photoelectron spectroscopy and factor analysis (FA). FA of the Si 2p and Cr 2p core levels shows that the kinetics of interface formation by IBM is characterized by two stages. During the first stage, a strong decrease in metallic Cr species is observed up to ion doses of ∼1.5 × 1016 ions cm−2, not only due to sputtering but also due to the formation of chromium silicide. The second stage is characterized by a similar evolution of the chromium silicide species related to Cr and Si, respectively, suggesting that the stoichiometry of the chromium silicide formed is nearly constant. Finally, with increasing sputtering time, the signals associated with the silicide slowly decrease, and pure Si0 species begin to appear, since no free metallic chromium atoms are available to react with the unlimited silicon supply from the substrate. The comparison of the experimental results with those obtained from TRIDYN simulations suggests that pure ballistic ion mixing mechanisms alone are unable to explain experimental results, and other radiation-enhanced diffusion mechanisms must be considered to explain the IBM of Cr/Si interfaces by low-energy Ar+ ions.