Search

The first χ2-based AlGaAs pulsed PSA with substantial confinement. This PSA achieves on chip in-phase gain of 30 dB, a sensitivity of 0.005 photons per pulse, and approaches the genuine quantum noise figure (NF) of 0 dB in near single mode operation for all degrees of freedom.

Mature B cell expressing CD180 molecules was first reported showing a protective effect against radiation- or dexamethasone-induced apoptosis in vitro. CD180 is a homologue of toll-like receptor (TLR) 4 and a pivotal regulator of cell proliferation and death. In SLE patients, number of CD180-negative B cells showed to increase in peripheral blood, and the change in the number of CD180-negative B cells appeared to be associated with the activity of SLE. In addition, such negative cells collected from SLE patients produced autoantibodies in vitro. We previously showed that the radiosensitivity of CD180-negative cells was greater than that of CD180-positive cells in vivo. In SLE-model NZBWF1 mice, we also reported the increase in number of CD180-negative B cells in parallel with development of SLE morbidity, the production of autoantibodies by CD180-negative cells, and pathogenesis of lupus-like nephritis caused by CD180-negative cells. We therefore, here, examine the effects of Xray-irradiation targeted CD180-negative B cells in NZBWF1 female mice. After the onset of symptoms, mice were paired according to their conditions. In each pair, one was irradiated with 4 Gy, and the other was sham-irradiated. Compared to the sham-irradiated mice, the number of CD180-negative cells was significantly decreased, the mean life span was markedly extended, and the nephritis and the morbid conditions were remarkably suppressed in irradiated mice. These findings suggest that radiation could effectively eliminate CD180-negative B cells, resulting in reduction of autoantibody production, and thus relieve SLE-like morbidity in SLE-model NZBWF1 mice. The findings also suggest a possibility that the radiotherapy for autoimmune diseases associated with the expression of CD180-negative cells, such as SLE and Sögren syndrome, might be a novel candidate for practical application., The 16th International Congress of Radiation Research (ICRR2019）

We have examined the reaction characteristics of the shift reaction for Ni/Mo/TiO 2 catalyst (Ni/Mo/Ti) which is a sulfur-tolerant water gas shift catalyst (sour shift catalyst). We used a bench-scale apparatus and the actual coal gasification gas (actual gas) which included COS and HCN as the competing reaction materials for the shift reaction. From the catalyst inlet temperature dependence of the shift reaction at the H 2 O/CO ratio of 1.2 mol/mol, it appeared that the reaction starting temperature was less than 200 °C. In addition, for Co/Mo/Al 2 O 3 catalyst (Co/Mo/Al) as a comparison catalyst, gas hour space velocities (GHSVs) to reach chemical equilibrium was calculated by static characterization analysis using E a and A obtained from the actual experiment for the catalyst inlet temperature of 200 °C and the H 2 O/CO ratio of 1.2 mol/mol. As a result, we confirmed that GHSV of Ni/Mo/Ti was 5 times more than that of Co/Mo/Al, that is, the catalyst quantity of Ni/Mo/Ti can be decreased to 1/5 that of Co/Mo/Al for the above conditions. Furthermore, we confirmed that COS and HCN were sufficiently reacted on Ni/Mo/Ti.

Quasi-planar segmented-channel MOSFETs (SegFETs) with gate lengths down to $\sim$ 45 nm are fabricated using a conventional process flow by starting with a corrugated-silicon substrate. In comparison with control devices (fabricated using the same process flow, but with a planar-silicon substrate), the SegFETs show reduced short-channel effect due to enhanced electrostatic integrity. Despite having a smaller physical channel width, the SegFET can achieve comparable drive current per unit layout area as the conventional planar MOSFET design.

A nonheme chloroperoxidase gene (cpo-p) from Pseudomonas pyrrocinia, a growth inhibitor of mycotoxin-producing fungi, was introduced into peanut via particle bombardment. The expression of the cpo-p gene is predicted to increase pathogen defense in peanut. Embryogenic peanut tissues were bombarded with gold particles coated with plasmid pRT66 carrying the cpo-p and hygromycin phosphotransferase (hph) genes, under the control of a double CaMV 35S and a single CaMV 35S promoter, respectively. Selection for hygromycin-resistant somatic embryos was performed on a liquid medium containing 10–20 mg/L hygromycin 3–4 days after bombardment. The integration and expression of the cpo-p gene was confirmed by Southern, Northern and Western blot analyses. In vitro bioassay using crude protein extracts from transgenic T0, T1, and T4 plants showed inhibition of Aspergillus flavus hyphal growth, which could translate to a reduction in aflatoxin contamination of peanut seed.

Vacancy-type defects in MOSFET structures fabricated with high-k (HfSiON) gate dielectrics were studied by monogenetic positron beams. An expansion of open volumes in HfSiON fabricated on Si substrates using atomic layer deposition technique was observed with increasing nitrogen concentration. This fact was discussed in terms of a role of nitrogen in Hf-related oxide using results obtained by first-principles calculation and XPS. MOSFETs fabricated by F-channel implantation technique were also characterized. The major defect species which causes the Vth shift of MOSFETs was identified as vacancy-fluorine complexes (such as V3F2) locate in channel regions of Si substrates. The preset work suggests that positron annihilation can be used to detect microscopic defects in MOSFETs, and it is a useful tool for determining process parameters for MOSFET fabrications with high-k gate dielectrics.

Ionicity has been raised as an important factor in discussing defect formation in Hf-based oxides. It has been elucidated from our first-principles calculations, that the stability of defects is dominated by Coulomb interactions between charged defects and surrounding ions. For instance, the formation energy of positively charged oxygen vacancies (VO+2) is markedly decreased when they are coupled with substitutional N atoms at O sites, as the nominal charges of N3- are negatively greater than those of the oxygen ions O2-. Our computational results have further revealed that the ionic character of Hf causes generation of a low-lying Si dangling bond level in Si doped HfSiOx. These results suggest that one must take into account the ionicity of the Hf-based compounds, which possess totally different properties from the conventional gate insulator SiO2 in terms of defect formation.

High-dielectric constant (high-k) gate materials, such as HfSiOx and HfAlOx, fabricated by atomic-layer-deposition techniques were characterized using monoenergetic positron beams. Measurements of the Doppler broadening spectra of annihilation radiation and the lifetime spectra of positrons indicated that positrons annihilated from the trapped state by open volumes that exist intrinsically in amorphous structures of the films. The size distributions of the open volumes and the local atomic configurations around such volumes can be discussed using positron annihilation parameters, and they were found to correlate with the electrical properties of the films. We confirmed that the positron annihilation is useful technique to characterize the matrix structure of amorphous high-k materials, and can be used to determine process parameters for the fabrication of high-k gate dielectrics. (© 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

Summary The pathogenesis of intestinal metaplasia (IM) of the human stomach remains completely unknown. Several classifications of IM have been suggested. Recent investigators have examined epithelial cell phenotypes in IM through immunohistochemistry, and revealed that gastric IM glands consist of a mixture of gastric and intestinal cell phenotypes. This implies that a heterogeneous cell population with both gastric and intestinal phenotypes would develop into a single intestinal phenotype. Recently, an intestinal stem cell marker was found to be present in the putative position of human gastric glands, whereas it was absent in gastric IM glands. This supports the assumption that IM is a consequence of abnormal stem cell differentiation, leading to lack of differentiation into any of the normal intestinal epithelial phenotypes. Recent advances in molecular biology have revealed intestinal transcriptional factors ( Cdx 1/ Cdx 2), suggesting that upregulation of Cdx 2 may trigger the initiation and development of IM in the stomach. Epidemiological evidence indicates that Helicobacter pylori is a major cause of IM but might not be solely responsible for its induction. A variation in host and bacterial background that predisposes to the development of IM has been revealed.

We consider the cause of different behavior of effective work function of gate metals after high and low temperature treatments. After high temperature treatment, reaction between Hf-based high-k dielectrics and a Si substrate generates oxygen vacancies (Vo) in high-k dielectrics, which leads to the subsequent electron transfer from Vo to gate metals. As a result, effective work function of gate metals becomes small and Fermi level pinning of gate metals occurs. Thus, Fermi level pinning is unavoidable phenomena for gate-first- processes. On the other hand, intrinsic hybridization between metal and high-k wave function at the interface is crucial factor to determine effective work function of gate metals after low temperature treatment. Accordingly, interface structures and band structures of metals govern the intrinsic effective work functions of metals.

We examined whether soluble mediators regulate the expression of tachykinin receptor mRNAs in synovial fibroblasts of patients with rheumatoid arthritis (RA). mRNAs encoding long and short isomers of neurokinin 1 receptor (NK1R), and neurokinin 2 receptor (NK2R) were confirmed by reverse transcription-polymerase chain reaction (RT-PCR) analysis. Level of long, but not the short, of NK1R mRNA was increased by treatment with 10-100 ng/ml basic fibroblast growth factor (bFGF) or 20 ng/ml tumor necrosis factor-alpha (TNF-alpha), but not with 1ng/ml interleukin 1beta (IL-1beta). TNF-alpha upregulated NK2R mRNA as well as long NK1R mRNA whereas bFGF had no effect on NK2R mRNA. Expression of neurokinin 3 receptor (NK3R) mRNA was not observed in RA fibroblasts, and its expression was not induced by bFGF and TNF-alpha. The basal and increased levels of long NK1R mRNA were inhibited by treatment with 20 microM SU5402, an inhibitor of the tyrosine kinase activity of FGF receptor 1 (FGFR1), or 10 ng/ml transforming growth factor-beta1 (TGF-beta1). SU5402 and TGF-beta1 had no effect on the basal level of short NK1R mRNA. Immunocytochemistry revealed the enhancement by bFGF of immunoreactive NK1Rs in the cells at 24 h after treatment. These results suggest that bFGF, TGF-beta1, and TNF-alpha in synovial tissue and fluid play a role in the regulation of long NK1R expression in synovial fibroblasts of RA patients. It appears that the pathway of downregulation by TGF-beta1 is more dominant in the long NK1R mRNA expression than that of upregulation by bFGF or TNF-alpha. Furthermore, the regulation of short NK1R mRNA expression seems to be performed via a different pathway from that of long isomer mRNA.

Aims : Secretory phospholipase A2 is associated with ischaemic injury in the human heart, but the distribution of type V secretory phospholipase A2 (sPLA2-V) remains unknown. The significance of sPLA2-V in myocardial infarction was investigated histopathologically. Methods : Sequential changes in the localization of sPLA2-V and its mRNA in myocardial tissues obtained from 30 autopsied hearts were examined by immunohistochemistry and in situ hybridization and compared with those of fibronectin, vascular endothelial growth factor (VEGF), interleukin (IL)-1β, tumour necrosis factor (TNF)-α, and cyclooxygenase-2 (COX-2). Results : No expression of sPLA2-V was observed in normal heart, but it was promptly expressed in wavy myofibres positive for fibronectin just after the onset of infarction. sPLA2-V was subsequently expressed in ischaemic cardiomyocytes around the lesion. The expression decreased at the granulation tissue and disappeared at the chronic stage with scar formation. The distribution of the signal for sPLA2-V mRNA paralleled that of the protein. Ischaemic myocytes around the lesion expressed VEGF, IL-1β, TNF-α and COX-2 at all stages. Conclusions : sPLA2-V production in myocardium is limited to the acute phase of infarction. sPLA2-V may play a dual role, acting both to remove degraded cell-membrane through cooperative activity with COX-2 in necrotic areas and to attack ischaemic myocytes around the lesion via degradation of membrane phospholipids.

A hybrid amplifier spanning from 1468 to 1568 nm is constructed by cascading a thulium-doped fiber amplifier pumped at 690 and 1400 nm and a Cerium co-doped high concentration wideband erbium-doped fiber amplifier. Results show seamless small signal net gain of 10 dB and noise figure maintained below 10 dB over the 100-nm gain bandwidth, making the amplifier a promising candidate for coarse wavelength division multiplexing in Metropolitan Area and optical access networks.

The energy transfer rate between Tm3+ ions in ZBLAN glass was estimated from the optical spectra and transition rates on the basis of a method proposed by Kushida. The optical spectra and radiative transition rates were obtained from the structural models prepared by molecular dynamic simulation and the crystal field theory. The lifetimes of 3H4, 1G4 and 1D2 levels were estimated from the calculation and the results were experimentally confirmed. The numerical model was also used to predict emission intensities under dual-wavelength excitation and decay curves under CW excitation.

We demonstrate error-free dense-wavelength-division multiplexing (DWDM) transmission of 40 40-Gb/s channels with 100-GHz spacing over 10 000 km dispersion-managed fiber using carrier-suppressed return-to-zero differential-phase-shift keying (CSRZ-DPSK), enhanced foward-error correction, and all-Raman-amplified spans with 100-km terrestrial length.

The paper describes a planar measurement technique for droplet size and velocity vector for applications to transient flow fields. Interferometric laser imaging for droplet sizing (ILIDS) is a novel measurement technique that gives an instantaneous spatial size distribution of droplets. The inter-fringe spacing of the ILIDS interferogram is correlated with individual particle sizes and the measurement accuracy depends upon the image processing technique for estimating the fringe spacing. The relationship between the diameter and fringe spacing was found by considering the angular phase difference of the external reflection and the direct refraction. Conventional ILIDS, which observes a circular image with fringes on a film, has difficulties in evaluating the fringe spacing accurately due to overlapping of the circular image at high concentration. The objective of the present study is to develop a simultaneous measurement technique of size and velocity vector of individual spherical particles by using anamorphic image acquisition optics that have different focal lengths along mutually perpendicular radii and accelerated data processing on a fully automated computing system. In the present technique, the circular image with fringes is optically compressed into linear images that are horizontally defocused and vertically focused keeping the information of the location and the size of droplets. The present "squeezing" technique can reduce the fringe overlapping and avoid complexity for the evaluation of fringe spacing. In the case of 100 droplets in a 1 M-pixel image plane, the particle validation ratio of the present technique is more than 95% in contrast with the validation ratio of the conventional technique that is less than 20%. Moreover, the vertical integration of the image can eliminate noise and consequently the signal-to-noise ratio of the interference signals becomes much higher than that of the conventional technique.

Theory shows that the gain bandwidth of a one-pump fiber optical parametric amplifier (OPA) using highly nonlinear fiber (HNLF) could be more than 200 nm. Under these circumstances, the OPA gain would overlap the pump-induced Raman gain. We have studied the combined effects of OPA and Raman gain theoretically and experimentally. The experimental results demonstrate a 200-nm bandwidth from a single fiber-optical amplifier and also verify that the influence of the Raman effect is relatively small, as predicted by the theory.

A data acquisition IC has been developed for digital storage oscilloscopes (DSOs). The entire DSO front-end except an input attenuator was integrated using 1-/spl mu/m double-poly, double-metal (DPDM) CMOS process technology. In the analog-to-digital conversion, a time-interleaved successive approximation architecture effectively enables both 25 Ms/s 8-b and 10 Ms/s 10-b operation. The input signal conditioner consists of a variable gain amplifier (VGA) and a second-order programmable low-pass filter (LPF) using folded-cascode structures with current feedback circuits. The overall gain is externally controllable from 12 dB to 38 dB, and the bandwidth is programmable at 500 kHz, 5 MHz, and 25 MHz. The chip consumes 340 mW at the 25 Ms/s operating condition and less than 8 mW in the power-down mode from a single 5 V supply.

A W/WN x /poly-Si composite structure (poly-metal) has been proposed as a low resistivity gate material. It has been found that an ultrathin WSiN layer forms at the W/Si interface after annealing, and as a result, the W/WSiN/poly-Si structure is very stable up to 950°C. In this paper, the formation mechanism of the ultrathin WSiN layer was studied. It was found that a 1 nm WSiN layer forms by solid state reaction between the WN x and poly-Si during annealing. A part of nitrogen atoms originally incorporated in the WN x film react with Si to form a 1 nm WSiN layer during annealing. Chemical bonds in the ultrathin WSiN layer consists of SiN bonds and metallic W bonds. Metallic W bonds are attributed to WW or WSi bonds. There is no WN bonds. Therefore, it is speculated that the WSiN consists of Si 3 N 4 and W or WSi x , and stabilizes the W/poly-Si interface. Since the WSiN layer acts as an excellent barrier metal for W and Si diffusion, the sheet resistivity of the poly-metal structure (where W thickness is 100 nm) can be maintained to be lower than 1.5 ω/sq.

Signal amplification in the S- and S/sup +/-band is demonstrated for the first time by simultaneous pumping of thulium-doped fiber amplifier (TDFA) with 690 and 1050 nm (or 1400 nm). Both pumping schemes are investigated and shown to be highly efficient: Roughly 20 dB of signal gain is available with only 80 mW of 1050 nm plus 42 mW of 690 nm, while signal input power is fixed at -35 dBm. Furthermore, these excitations can take advantage of low-cost readily available pump laser diodes at 690 nm for digital video disk applications. This makes TDFA a promising candidate for coarse wavelength-division-multiplexing applications in metropolitan area network and access network environment.

A new low-resistivity poly-metal gate structure, W/WSiN/poly-Si, is proposed, A uniform ultrathin (

Upconversion pumping in the 14xx-nm range gives the thulium-doped fiber amplifier (TDFA) a 3-dB dynamic range of 15 dB, and transient power excursion much lower than the conventional erbium-doped fiber amplifier during dynamic wavelength add-drop, due to the energy transitions inherent to the Tm/sup 3+/ ion. For specific pumping scheme and operating conditions, gain increases with increasing signal input, a behavior significantly different from conventionally optical amplifiers. This phenomenon is studied experimentally and theoretically, and promises to provide insight for gain control algorithms for TDFAs under bursty traffic conditions in a metropolitan area network environment.

A Raman fiber oscillator used for optical amplification is demonstrated to have lower double Rayleigh scattering, transient spikes, cross-phase modulation, and higher saturation input threshold compared with a conventional discrete Raman amplifier at similar operating conditions. This could be a promising technology for deployment in practical systems.

We designed and evaluated a fluoride-based high concentration erbium-cerium codoped fiber amplifier. It is suitable for metropolitan area networks due to faster transient, flatter (unfiltered) gain, smaller footprint, and gain excursion than its silica-based counterpart. A detailed study of its transient dynamics has been performed.

To facilitate continued CMOS technology scaling, thin-body transistor structures such as the FinFET [1] and fully depleted silicon-on-insulator (FD-SOI) MOSFET [2] have been proposed to better suppress short-channel effects (SCE) than the conventional MOSFET structure in the sub-25 nm gate length (L g ) regime. However, these structures require either more challenging fabrication processes or more expensive silicon-on-insulator substrates. Recently, a segmented-channel bulk MOSFET (SegFET) structure [3] was proposed as a more evolutionary solution that offers the advantages of a thin-body MOSFET (reduced variability in performance and improved scalability) together with the advantages of a conventional planar MOSFET (low substrate cost and capability for dynamic threshold-voltage control).

A p-MOSFET structure with solid-phase diffused drain (SPDD) is proposed for future 0.1- mu m and sub-0.1- mu m devices. Highly doped ultrashallow p/sup +/ source and drain junctions have been obtained by solid-phase diffusion from a highly doped borosilicate glass (BSG) sidewall. The resulting shallow, high-concentration drain profile significantly improves short channel effects without increasing parasitic resistance. At the same time, an in situ highly-boron-doped LPCVD polysilicon gate is introduced to prevent the transconductance degradation which arises in ultrasmall p-MOSFETs with lower process temperature as a result of depletion formation in the p/sup +/-polysilicon gate. Excellent electrical characteristics and good hot-carrier reliability are achieved. >

The salicide technology using rapid thermal annealing was applied to MOSFETs on thin-film SOI. Since the SOI film was limited to a thickness of less than 100 nm, the silicidation reaction between Ti and Si atoms on the SOI surface exhibited new features that depended on the initial thickness of the deposited Ti. There was an optimum thickness of as-deposited Ti on silicidation due to the restricted thickness of the Si layer. Beyond the optimum point, the region adjacent to the silicided Si layer works as a Si source to assure stoichiometric TiSi/sub 2/. The subthreshold slopes and carrier mobilities were not changed by the salicide process. Junction leakage characteristics were slightly degraded; however, the change was small enough for device application. The influence on AC characteristics was well demonstrated for a high-speed CMOS ring oscillator with a gate length of 0.7 mu m. The minimum delay time/stage was 46 ps/stage at 5 V. This gives 1.8 times higher speed operation than the controlled bulk CMOS ring oscillators with the same design rule. >

The structure of a ternary 40AlF3 · 20YF3 · 40CaF2 glass (AYC glass) was examined by a combination of X-ray and neutron diffraction analysis and molecular dynamics simulations. The comparison with the structure of 40AlF3 · 20BaF2 · 40CaF2 glass (ABC glass) [T. Nanba et al., Mater. Sci. Forum 32&33 (1988) 385] indicated that the network structures of both glasses were similar to each other, with AlF6 octahedra joining at their corners to form chains with branches. Although AlF7 polyhedra were not found in ABC glass, they existed in AYC glass, as one third of the AlF6 octahedra, and these polyhedra were distorted due to the presence of Y3+ ions.

Two types of glasses in the ternary Alf 3 BaF 2 CaF 2 system were prepared under different cooling rates to investigate their network structure. It is suggested that the network is constructed from (AlF 5 ) 2− chains and (AlF 6 ) 3− isolated units and many isolated units are present in the rapidly cooled glass from radial distribution analyses. Molecular dynamics simulations were carried out to reproduce the difference in the glass network. The models showed good agreement with the observed RDF and Raman spectrum, but an exact structural difference in the network could not be obtained. However, it is suggested that chains with branches may be formed when the melt is slowly cooled.

A mobility model for MOSFET device simulation is proposed. The model is not only applicable to both inversion layer and source/drain high concentration regions of a MOSFET, but it also takes into account the screening effect in the inversion layer. The model also includes an improved normal-field dependence for thin gate oxide MOSFETs. The low parallel electric field mobility is estimated by adding mobilities due to donor scattering, acceptor scattering and lattice scattering using Matthiesen's rule. Mobilities due to both the donor and the acceptor scattering include the electron screening effect. The mobility due to lattice scattering is formed as a function of normal electric field E/sub n/, including the strong dependence term of E/sub n/, to express surface roughness scattering. Calculation results of the device simulation using the mobility model show good agreement with the experimental data for various channel dopings. >

Recent experimental and theoretical achievements are reviewed on three types of laser oscillation inp-type germanium; the intervalence band (IVB) laser oscillation due to transitions between the light-hole band and the heavy-hole band, the light-hole cyclotron resonance (LHCR) laser oscillation, and the heavy hole cyclotron resonance (HHCR) maser oscillation. Described for the IVB oscillation are fundamental characteristics (such as the wavelength range of oscillation and the cooperation with higherharmonic cyclotron resonances), the polarization characteristics of radiation and influence of uniaxial stress on the oscillation. The present status of the attempt at single line oscillation is also described. As for the LHCR oscillation, the formation of population inversion and the amplification are explained on the basis of quantum mechanical calculations of the valence bands in crossed electric and magnetic fields.

We investigated the basic diagnostic processes used in telepathology with our robotic still-image system, OLMICOS, by analysing the steps and patterns used in 20 consecutive tissue section diagnoses. Three basic approaches were recognized. One was magnifying a suspect finding in a low-powered microscopic image. This approach was used mostly for confirming or characterizing a tumour. The second approach was scanning over a low-powered image by magnifying square images to form a mesh. This was found to be useful to confirm the presence or absence of signs and was mostly used as the initial step in judging the surgical margin of malignant cases or diagnosing lesions. The third approach was a combination of these two and was used for delineating the surgical margin, confirming the absence of metastases or diagnosing difficult lesions. Recognition of these three basic diagnostic approaches is important in making a rapid and correct remote diagnosis.

We have found that effective work functions of high-work function gate metals (p-metals) become small and Fermi level pinning of gate metals occurs after high temperature treatment as the same in the case in p+poly-Si gates. On the contrary, intrinsic hybridization between metal and high-k wave function at the interface is crucial factor to determine effective work function of gate metals after low temperature treatment. As discussed above, metal/high-k interface properties are much different each other after high- and low-temperature treatment.

We developed 10 Gbps burst-mode clock recovery IC with synchronization time of 50 ps. Using the IC with PIN-TIA PDs, an optical packet receiver which requires no preamble is also developed.

Structures of two borate glasses containing heavy metals, SrO·2B 2 O 3 and PbO·2B 2 O 3 , were investigated by using X-ray and neutron diffraction. Structural models were built and radial distribution functions (RDF) were calculated by the pair function method. It is known that SrO·2B 2 O 3 and PbO·2B 2 O 3 crystals have three-coordinated oxygens. Radial distribution functions calculated from crystal structures with the same composition showed poor comparison with the observed glass RDFs, indicating that the structures of SrO·2B 2 O 3 and PbO·2B 2 O 3 glasses consist of different macro anions than those found in the crystals.

An etching model in which topography is derived by solving a modified diffusion equation is introduced. This model is simple and makes it possible to simulate three-dimensional (3-D) topography accurately and quickly. Based on this model, a 3-D topography simulator which can be applied in the development of photolithography and isotropic/anisotropic etching has been developed. With this simulator, it is possible to simulate the series processes and multilayer etching, such as contact hole and trench etching. By simulating photolithography, diffraction and standing-wave effects can be found clearly in the 3-D topography of the developed photoresist. In the case of an etching process which is restricted by diffusion, the dependence of the etch front topography on the window width of the mask is examined. >

Apparent difference of the dielectric breakdown behavior between high-k stack and the conventional SiO2 is investigated by the temperature dependence of TDDB lifetime. It is clarified that the temperature dependence of TDDB lifetime in high-k stack can be described by two parameters, which is identical to the case of conventional SiO2.

A novel technique for the control of nitrogen (N) and oxygen (O) concentrations in the metal/high-k gate stacks is proposed. By inserting a reactive metal (Ti) layer remote from work function metal and high-k insulator, N and O concentration is easily decreased. This technique effectively suppresses the EOT increase after high-temperature annealing. Moreover, improved electron mobility and decreased interfacial trap density can be obtained by this technique. We demonstrated MISFET with extremely high electron mobility and thin EOT can be easily obtained even with relaxed process of HfSiON

Room temperature ferromagnetism of GaMnN thin films is awaked by a mild hydrogenation treatment of samples synthesized by molecular beam epitaxy. Local environment of Mn atoms is monitored by Mn-L2,3 near edge x-ray absorption fine structure (NEXAFS) technique. Doped Mn ions are present at substitutional sites of Ga both before and after the hydrogenation. No secondary phase can be detected. Major valency of Mn changes from +3 to +2 by the hydrogenation. The present result supports the model that the ferromagnetism occurs when Mn2+ and Mn3+ are coexistent and holes in the mid- gap Mn band mediate the magnetic coupling., 12 pages, 5 figures

We have realized a 0.9nm-EOT TaSix/HfSiON gate stack that exhibits the high electron mobility of 264 cm2/Vs @ 0.8MV/cm (86% of thermal SiO2), even after spike annealing at 1000degC. This was achieved by using thermally-stable HfSiON gate dielectrics with plasma nitridation, in which interfacial layer growth due to recoiled oxygen had been successfully suppressed

We have achieved a remarkably wide range (~1.2 V) of differences in flatband voltage (Vfb) by using a combination of metal alloy (Pt-W)/La2O3 gate stacks. The controllable range is about 3 times greater than that of Hf-based gate stacks. The wide range of Vfb can be maintained even after annealing in forming gas and oxidizing gas ambients. We consider that this is attributed to charge transfer at the Pt- La2O3 interfaces inducing increase in the effective work functions, especially at high Pt composition ratios. We believe that La2O3 is one of the most promising high-k dielectric materials for future CMOS devices in 32nm-node and beyond.

We show the new findings in nano-scale interface physics and atomistic behaviors of defects in gate dielectric materials. In this paper, we first discuss the relation between defect behaviors and transistor characteristics. Next, we introduce our newly prosed mechanism of Fermi level pinning governed by the interface reaction. Further, we show that conventional charge neutrality level concept does not applicable to metal/high-k dielectric interfaces, and we propose a generalized charge neutrality level concept that includes both nano-scale interface structures and metal band structures. Finally, we discuss the atomistic investigation on the characteristics of conventional Si/SiO2 nano interfaces.

Fermi-level pinning of poly-Si and metal-silicide gate materials on Hf-based gate dielectrics has been systematically studied theoretically. Fermi-level pinning in high- and low-work-function materials is governed by the O vacancy and O interstitial generation, respectively. From our theoretical considerations, the authors have found that the work-function pinning-free-region generally appears due to the difference in the mechanism of Fermi-level pinning of high- and low-work-function materials. Further, the authors also discuss the interface physics between pure metal gates and high-k dielectrics