Intermediate Si Research Articles

Structural changes in Zr–Si–N films vs. their silicon content

Zr–Si–N films are deposited on silicon and steel substrates by DC magnetron sputtering of a Zr–Si composite target in Ar–N 2 reactive mixture, with different Si contents. At very low silicon content, no Si–N bond is observed and Si atoms may be inserted in the lattice of ZrN, as confirmed by the strong increase in the compressive stresses. For intermediate Si contents, nanocomposite films are synthesised: [100] oriented ZrN grains are embedded in an amorphous SiN x phase. For a silicon content higher than 6 at.%, amorphous films are deposited. The variation of the films hardness vs. the silicon concentration is discussed as a function of the films structure and compressive stress level. Finally, the relationship between the hardness and Young's modulus values is also presented.

Assessment of MRI and dynamic contrast-enhanced MRI in the differential diagnosis of adenomatoid odontogenic tumor

The radiographical differentiation of adenomatoid odontogenic tumor (AOT) from dentigerous cysts, calcifying odontogenic cysts, calcifying epithelial odontogenic tumors, odontogenic keratocysts and amelobastomas is sometimes difficult. We attempted to differentiate AOT from other lesions similar to AOT in radiographic findings using MRI. The MRI features of AOT in our three cases included homogeneous low SI in the cystic portion and homogeneous intermediate SI in the solid portion on T1WI, homogeneous high SI in the cystic portion and intermediate to slightly high SI in the solid portion on T2WI and enhancement of only the solid portion on CE-T1WI although none of the sequences included SI of calcifications. The contrast index curves in the three cases of AOT showed a gradual increase to 300 s, which signified a benign tumor. These MRI features were characteristic features of AOT and might be a basis for differentiating AOT from the above possible lesions in radiographic examinations.

Thermal decomposition of NH 3 on the Si(1 0 0) surface

The adsorption and thermal reaction of NH 3 on the Si(1 0 0) surface are investigated by high-resolution core-level photoemission spectroscopy using synchrotron radiation. The existence of different reaction products and their chemical bonding configurations at different substrate temperatures are revealed from N 1s and Si 2p core-level spectra. We clearly identified a series of Si–NH 2, Si 2NH and Si 3N species in N 1s spectra indicating a successive N–H bond dissociation during thermal decomposition. The depth distribution and the population changes of each N species with annealing suggest that (i) the intermediate Si 2NH species include insertion into the back-bond site between the first and the second Si layers as well as bridging Si dimer site and (ii) the fully dissociated N atoms are incorporated into the Si subsurface layers first. At a high temperature above 900 K, the incorporated N atoms partly segregate back to the surface to form stoichiometric silicon nitride patches. The Si 2p core levels consistently show progressive changes in subnitride formation and the liberation of H atom upon increase of annealing temperature. The implication of the present result on the proposed reaction mechanism is discussed.

The growth of high-quality SiGe films with an intermediate Si layer

The growth of high-quality relaxed SiGe epilayer has been achieved by introducing an intermediate Si layer in the SiGe film. It was found that intermediate Si in the SiGe film changed the mechanism of strain relaxation during the growth so that the shallow pits related to strain relief on the surface were suppressed. Such a SiGe/Si/SiGe heterostructure has a threading dislocation density of 8.9×10 5 cm −2 and a root mean square roughness of 3 nm. The intermediate Si layer was also demonstrated to act as effective nucleation site for dislocation loops to relax the mismatch strain. A mechanism of strain relaxation for this intermediate Si layer is proposed. Compared with the conventional compositionally graded buffer layer, it has the advantages of having thinner buffer layer for required degree of relaxation, smoother surface, and maintaining the threading dislocation density at the same level.

K-shell ionization of high Z elements with low to intermediate velocity Si ions

We report here the measurement of the absolute cross-sections for the K-shell ionization of high atomic number elements (Au and Bi) induced by low to intermediate velocity Si ions. The relativistic wavefunction effects on the ionization cross-sections have been studied by comparing the measured data with different theoretical calculations with and without including the relativistic corrections. The relativistic effect on the K-ionization cross-section is found to depend on the projectile atomic number and velocity.

Enhanced radiative transition inSinGemnanoclusters

Using an ab-initio molecular dynamics scheme (the Fireball scheme), we determined the equilibrium structure of intermediate size Si_nGe_m (n+m=71) nanoclusters with/without hydrogen passivation on the surface. Due to the strong surface distortion, defect states are found to permeate the energy gap of Si_nGe_m clusters. However, the defect states are removed by adding H atoms on the surface of Si_nGe_m clusters, and the gap opens up to a few eV, indicating a blueshift for photoluminescence. It is also found that the radiative transition between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) states is enhanced by one to two orders of magnitude for Si_nGe_m nanoclusters with respect to the corresponding pure Si clusters. This significant increase of the emission probability is attributed to the strong overlap of HOMO and LUMO wavefunctions that are centered mostly on the Ge atoms.

PVD grown (Ti,Si,Al)N nanocomposite coatings and (Ti,Al)N/(Ti,Si)N multilayers: structural and mechanical properties

In the last few years a considerable effort has been undertaken in order to optimise the production techniques of thin films and improve their quality. In this work, nanocomposite films resulting from Si additions to a (Ti,Al)N matrix have been prepared by RF and/or DC magnetron sputtering, with deposition rates varying from 0.21 μm/h to 4.6 μm/h. Rutherford Backscattering (RBS) and Electron Microprobe Analysis (EMPA) were used in order to access the chemical composition as well as the density of the films. For samples prepared with low deposition rates (deposited by a combination of RF and DC reactive magnetron sputtering) both symmetric and asymmetric XRD scans showed the development of crystalline phases whose structure is very similar to that of bulk TiN. The peak positions revealed changes of the lattice parameter from 0.420 to 0.428 nm with an increase of Si content dependent on the deposition rate. The lowest lattice parameter corresponds to a Ti–Si–Al–N phase where some of the Si and Al atoms are occupying Ti positions in the f.c.c. TiN lattice, while the highest lattice parameter corresponds to a system where at least a partial Si segregation can be enough to nucleate and develop the Si 3N 4 phase that forms a layer on the growth surface, covering the (Ti,Al)N nanocrystallites and limiting their growth. As for the (Ti,Al,Si)N crystalline texture evolution, a (111) preferential growth for (Ti,Al)N and for low Si content was observed, while at intermediate Si content the texture changed to (200). With the increase of the Si content there is a corresponding decrease in the size of the diffracting grains. For samples prepared with high deposition rates (DC sputtered samples) High-Resolution Transmission Electron Microscopy (HRTEM) micrographs revealed a columnar growth associated with the f.c.c.-type structure of both phases. Small crystallites with sizes between ±7 and ±10 nm were observed. The use of (Ti,Al) and (Ti,Si) targets, relatively high deposition rates and an alternate deposition resulted in a multilayer of (Ti,Si)N/(Ti,Al)N. This system was produced with modulation periods between 5 and 10 nm, as shown by HR-TEM results, when the samples were grown with a deposition rate between 2 and 4.6 μm/h, respectively. Their average ultramicrohardness can be as high as 50 GPa. The residual stress values for the multilayer system are significantly lower than that of (Ti,Si,Al)N nanocomposite coatings.

펄스 레이저 증착법에 의해 제작된 ZnO-Si-ZnO 다층 박막의 특성 연구

ZnO-Si-ZnO multi-layer thin films have been deposited by pulsed laser deposition (PLD). And then, the films have been annealed at 300<TEX>$^{\circ}C$</TEX> in oxygen ambient pressure. Peak positions of ultraviolet (UV) and visible region were changed by addition of Si layer. Mobility of the films was improved slightly than ZnO thin film without Si layer. The structural property changed by inserting intermediate Si layer in ZnO thin film. The optical properties and structural properties of ZnO-Si-ZnO multi-layer thin films were characterized by PL(Photoluminescence) and XRB(X-ray diffraction) method, respectively. Electrical properties were measured by van der Pauw Hall measurements

One-dimensional ecosystem model of the equatorial Pacific upwelling system. Part II: sensitivity analysis and comparison with JGOFS EqPac data

Abstract A one-dimensional model of the equatorial Pacific upwelling ecosystem that incorporates two phytoplankton components, two grazers, and three nutrients, Si(OH)4, NO3, and NH4 (Chai et al., Deep-Sea Research II (2002) 2713–2745), was designed to consider the effects of Si(OH)4 limitation on the diatom growth and ecosystem functioning. Model output was obtained for a range of source concentrations of Si(OH)4, 3–15 mmol m−3, coinciding with the range measured at 120 m depth during JGOFS EqPac. NO3 was held at 12 mmol m−3, reflecting the relatively greater concentrations of NO3 compared to Si(OH)4 in the JGOFS data. The model was shown to function as a chemostat-like system with the loss rates, provided largely from zooplankton grazing, controlling growth rates of the phytoplankton. When different source concentrations of Si(OH)4 were applied, surface concentrations of Si(OH)4 varied within a narrow range compared to NO3 as would occur in a chemostat with limiting Si(OH)4 and non-limiting NO3 in the feed water. Vertical profiles of nutrients compared well with field data. Model results are compared with field data for new and total nitrogen production and export of N, Si, and C, and with other models, although none consider Si(OH)4 specifically. The model suggests that the stability of the equatorial system with its narrow range of biological and chemical variables is conferred by the action of diatoms providing food for mesozooplankton whose grazing also depletes the picoplankton. Diatoms increase with source Si(OH)4 concentrations, and picoplankton population and NO3 consumption decrease, resulting in a maximum surface TCO2 and increased CO2 flux to the atmosphere at intermediate source Si(OH)4 concentrations. Diatoms function in the equatorial system as a silica pump to export silica. This means that sedimented biogenic silica under the equatorial upwelling area should be viewed as an amplifier of changes in surface properties, with important consequences to paleoequatorial productivity.

DEPENDENCE OF THERMAL OXIDATION BEHAVIOR OF SILICON NANOCRYSTALLITES

Three kinds of nanosilicon crystallites were prepared by different methods in high vacuum. All of them were composed of tiny silicon crystallites which were initially only mildly oxidized before annealing, but their behavior upon annealing in vacuum differed substantially depending on the environment in which they resided. XPS analyses revealed that the unencapsulated nanoparticles tended to oxidize quite quickly, whereas the nanoparticles sandwiched between layers of Al 2 O 3 matrices were oxidized rather slowly even under intense annealing. In the zinc/silicon nanocrystalline mixture, the oxidation of the Si 0 state was even faster than that of the intermediate Si +1,+2,+3 states. Both the stability and formation processes of the Si - O bonds in the partially oxidized states differed considerably with different environmental surroundings. However, in all cases, the Si - O bonds of the fully oxidized Si +4 state remained the most stable, to which the less oxidized states tend to gravitate eventually.

Environmental dependence of thermal oxidation behaviour of silicon nanocrystallites

AbstractThree kinds of nanosilicon crystallites were prepared by different methods in high vacuum. All of them were composed of tiny silicon crystallites that initially were only mildly oxidized before annealing, but their behaviour upon annealing in a vacuum differed substantially, depending on the environment in which they resided. X‐ray photoelectron spectroscopy analyses revealed that the unencapsulated nanoparticles tended to oxidize quite quickly, whereas the nanoparticles sandwiched between layers of Al2O3 matrices were oxidized very slowly even under intense annealing. In the zinc/silicon nanocrystalline mixture, oxidation of the Si0 state was even faster than that of the intermediate Si +1,+2 and +3 states. Both the stability and formation processes of the Si O bonds in the partially oxidized states differed considerably with different environmental surroundings. However, in all cases, the Si O bonds of the fully oxidized Si+4 state remained the most stable, to which the less oxidized states tend to gravitate eventually. Copyright © 2001 John Wiley &amp; Sons, Ltd.

Kinetics of the WF 6 and Si 2H 6 surface reactions during tungsten atomic layer deposition

The atomic layer deposition (ALD) of tungsten (W) films has been demonstrated using alternating exposures to tungsten hexafluoride (WF 6) and disilane (Si 2H 6). The present investigation explored the kinetics of the WF 6 and Si 2H 6 surface reactions during W ALD at 303–623 K using Auger electron spectroscopy techniques. The reaction of WF 6 with the Si 2H 6-saturated W surface proceeded to completion at 373–573 K. The WF 6 reaction displayed a reactive sticking coefficient of S=0.4 and required an exposure of 30 L (1 L=1×10 −6 Torr s) to achieve saturation at 573 K. The WF 6 exposures necessary to reach saturation increased with decreasing temperature. At surface temperatures <373 K, the WF 6 reaction did not consume all the silicon (Si) surface species remaining from the previous Si 2H 6 exposure. The reaction of Si 2H 6 with the WF 6-saturated W surface displayed three kinetic regimes. In the first region at low Si 2H 6 exposures⩽50 L, the Si 2H 6 reaction was independent of temperature and had a reactive sticking coefficient of S∼5×10 −2. In the second kinetic region at intermediate Si 2H 6 exposures of 50–300 L, the Si 2H 6 reaction showed an apparent saturation behavior with a Si thickness at saturation that increased with substrate temperature. At high Si 2H 6 exposures of 300–1×10 5 L, additional Si was deposited with an approximately logarithmic dependence on Si 2H 6 exposure. The Si 2H 6 reaction in this third kinetic region had an activation energy E=2.6 kcal/mol and the Si thickness deposited by a 1.6×10 5 L Si 2H 6 exposure increased with temperature from 3.0 Å at 303 K to 6.6 Å at 623 K. These kinetic results should help to explain W ALD growth rates observed at different reactant exposures and substrate temperatures.

Bulk and interface properties of low-temperature silicon nitride films deposited by remote plasma enhanced chemical vapor deposition

Hydrogenated silicon nitride (a-SiNx:H) films were deposited at temperatures ranging from 50 to 300 °C with remote plasma enhanced chemical vapor deposition (RPECVD) from NH3 and SiH4. The effect of the operating variables, such as deposition temperature and especially the partial pressure ratio of reactant (R=NH3/SiH4) on the properties of the Sa-SiNx:H interface was investigated. The H* radical was dominantly observed and the deposition rate was proportional to the NH* radical concentration. The density of highly energetic N 2 * radicals increased in the high plasma power regime in which the film surface was roughened, but they promote surface reactions even at low temperature. The refractive index was more closely related to the film stoichiometry than film density. The interface trap density is related to the amount of silicon intermediate species and Si–NH bonds at the Si/SiNx:H interface and it can be minimized by reducing the intermediate Si species and Si–NH bonding state. The films showed a midgap interface trap density of 2 × 1011 - 2 × 1012cm-2. © 2001 Kluwer Academic Publishers

The Ultraviolet and Optical Spectra of Metal‐deficient O Stars in the Small Magellanic Cloud11Based in part on observations with the NASA/ESA Hubble Space Telescope obtained at the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS 5‐26555. Also based on observations obtained at the Anglo‐Australian Observatory, Siding Spring,

ABSTRACTAn ultraviolet and optical spectral atlas of 15 O stars in the Small Magellanic Cloud (SMC) is presented and described. The echelle data have resolving powers of order 104; they were obtained with the Hubble Space Telescope Space Telescope Imaging Spectrograph in the UV, and at the Anglo‐Australian Telescope or the European Southern Observatory 3.6 m in the optical. The ultimate objective is to develop metal‐deficient templates for the interpretation of distant starbursts, but here we discuss interesting new properties of the SMC stars themselves, revealed by the high quality of these data. The SMC metal deficiency produces anomalously weak stellar‐wind profiles along the entire O main sequence, as well as at intermediate luminosities; the first intermediate Si iv λ1400 wind profile in the SMC is shown. The second known Of star in the SMC displays wind peculiarities that are identical to those of its spectral classmate, again likely due to the low systemic metallicity. Several objects display marked CNO anomalies, including the first cases of C iii λ4650 emission without N iii λ4640 in O‐type spectra. The N/C ratio appears to increase with mass, extent of evolution away from the zero‐age main sequence, and/or rotational velocity in the young cluster NGC 346. In addition, the first examples of Onfp (Oef) and Of?p spectra in the SMC have been found (the latter being only the fourth member of its peculiar shell category known anywhere). The UV wind characteristics of these objects correlate with their optical peculiarities. All these spectroscopic phenomena provide diagnostics of the evolutionary status of metal‐deficient massive stars.

Optical investigation of modified Stranski–Krastanov growth mode in the stacking of self-assembled Ge islands

The modification of the Stranski–Krastanov growth mode produced by the strain-field from buried islands has been non-destructively investigated by photoluminescence (PL) spectroscopy in lattice-mismatched Ge/Si system. The samples consist of double Ge layers with each coverage of 4.5 monolayers separated by a Si spacer layer with various widths. At an intermediate Si spacer thickness of 100–200 Å, the modification of the growth mode of the second Ge layer was clearly evidenced in the PL spectral feature which was characterized by the superposition of two different PL spectra. In addition, a drastic increase of the density of Ge islands in the second layer was observed by atomic force microscopy. These facts show that the strain-field from the buried islands is very important to control the growth mode of lattice-mismatched semiconductor heterostructures.

Reduced critical thickness and photoluminescence line splitting in multiple layers of self-assembled Ge/Si islands

We present a transmission electron microscopy and photoluminescence (PL) study of stacked Ge island layers embedded in Si. We systematically varied the thickness t s of the Si interlayer and the number of island layers to explore the effect of the interlayer thickness on the growth of Ge islands. For sufficiently thin t s, the Ge islands align in a vertical direction. Analytic transmission electron microscopy reveals that, in the regime of vertical island alignment, the wetting layer (WL) thicknesses decrease in upper Ge layers. This effect is explained by a reduced critical thickness for island nucleation caused by the strain fields of buried islands. At the same time, PL experiments show that the WL energy transition is split into multiple lines, as would be expected due to the different WL thicknesses. The line splitting is dependent on the intermediate Si spacer thickness, reflecting the degree of strain field interaction between the stacked layers. The island-related optical transition is blue-shifted with increasing number of island layers. We attribute this effect to enhanced material intermixing in the second and proceeding island layers.

Structural analysis of Ti1−xSixNy nanocomposite films prepared by reactive magnetron sputtering

Abstract In this paper, we report on the preparation of thin films resulting from additions of Si to TiN matrix, by r.f. reactive magnetron sputtering. Results of X-ray diffraction (XRD) in both θ –2 θ and α –2 θ scans showed that a mixture of two phases is present, where the first is most likely fcc TiN. The higher lattice parameter of this phase, about 0.429 nm (0.424 nm for bulk TiN), could be explained by taking into account that a correction of the residual stress effect on peak positions might slightly decrease the value of the lattice parameter (around 1%). Regarding phase 2, and although the exact nature of its composition is more difficult to evaluate, we believe that it is also a cubic lattice consisting of TiSiN, where the Si could be occupying Ti positions within the TiN lattice. This would explain the low value of the lattice parameter, which by assuming a cubic structure would be 0.418 nm. Concerning texture evolution, phase 1 revealed some variations in preferential growth, which changed from 〈111〉 for low Si additions to 〈220〉 at intermediate Si additions and finally to a weak 〈200〉 texture for large Si additions. A small amorphous region of silicon nitride for large Si additions was also observed. Fourier analysis of XRD patterns showed a decrease in the size of grains for small Si additions when compared to that of TiN. For higher Si contents, only small changes were observed, although a decrease in grain size seems to be the main tendency. The grains are within the range of 4–6 nm. High-resolution transmission electron microscopy (HRTEM) on Ti 0.63 Si 0.37 N 1.12 confirmed this nanocrystalline nature of the grains, revealing grains with sizes of about 2–3 nm.

Electrical Characterization of a Double Barrier Direct Tunneling Diode Structure

ABSTRACTWe report the electrical characterization of a direct tunneling diode structure that incorporates a multilayer dielectric. The dielectric consists of a stack of two thermally grown, ultrathin SiO2 layers, each ∼3.5 rin thick, separated by a deposited, continuous, undoped, ultrathin nanocrystalline Si layer ∼5.0 nm thick. Electrical measurements of this structure are reported for both n-type and p-type Si substrates. We find that the room temperature transport through this structure is accounted for by describing the intermediate Si layer as a quantum well with a continuum of states, and by otherwise assuming bulk properties for the ultrathin layers, such as the existence of a bandgap in the Si well and the usual Si-SiO2 interface potential barrier height at all interfaces. This structure is expected to be useful as the active dielectric in nonvolatile memory devices.

Si and N K-XANES spectroscopic study of novel Si–C–N ceramics

Abstract X-ray absorption near edge structure (XANES) measurements at the Si and N K-edges of ternary Si–C–N compounds are applied to analyze the changes of structural units owing to the thermally induced ceramization of silicon dicarbodiimide Si(NCN) 2 at temperatures up to 1600°C. Samples synthesized below 800°C show an environment of silicon consisting of carbodiimide (NCN) groups. Further annealing leads to a crystalline phase, Si 2 CN 4 , which combines the structural units (Si–N–Si) and (Si–NCN). Recorded spectra of samples annealed at temperatures higher than 1200°C show the decomposition of Si 2 CN 4 . The material then transforms to a Si 3 N 4 –SiC composite material through an intermediate amorphous Si–C–N phase.

Physical, structural and mechanical characterization of Ti1−xSixNy films

Abstract Within the frame of this work, Ti 1− x Si x N y hard coatings with 0≤ x ≤0.37 and thicknesses ranging from 1.2 to 3.5 μm, were prepared by r.f. reactive magnetron sputtering in an Ar/N 2 gas mixture. X-ray diffraction and Fourier analysis of X-ray profiles were used to investigate the structure and grain size, and its correlation with hardness behaviour, as a function of the Si content, bias voltage and working gas (argon) flow rate. In this respect, the results show that a double cubic phase of NaCl type was developed with lattice parameters of 4.18 and 4.30 A, revealing the (111) orientation for low Si content ( x =0.05), (220) for intermediate Si contents (0.13≤ x ≤0.22) and (200) for the highest Si contents (0.30≤ x ≤0.37). Regarding the results of ultramicrohardness tests, and although all samples with 0.05≤ x ≤0.30 present a hardness value higher than 30 GPa, the Ti 0.85 Si 0.15 N 1.03 revealed the highest hardness value, around 47 GPa, which is more than twice as high as that of common TiN. Furthermore, the study of hardness as a function of the applied bias voltage revealed that best results are achieved between −50 and 0 V. The variation in hardness as a function of the argon flow showed that best results in hardness are obtained when working with flow rates around 110 cm 3 /min.