Characteristics of high-quality HfSiON gate dielectric prepared by physical vapour deposition

This paper presents a method using simple physical vapour deposition to form high-quality hafnium silicon oxynitride (HfSiON) on ultrathin SiO₂ buffer layer. The gate dielectric with 10? (1?= 0.1 nm) equivalent oxide thickness is obtained. The experimental results indicate that the prepared HfSiON gate dielectric exhibits good physical and electrical characteristics, including very good thermal stability up to 1000℃, excellent interface properties, high dielectric constant (k=14) and low gate-leakage current (I_g=1.9×10^-3A/cm² @V_g=V_fb-1V for EOT of 10?). TaN metal gate electrode is integrated with the HfSiON gate dielectric.The effective work function of TaN on HfSiON is 4.3eV, meeting the requirements of NMOS for the metal gate. And, the impacts of sputtering ambient and annealing temperature on the electrical properties of HfSiON gate dielectric are investigated.     

Comparative study of electrical characteristics for n-type 4H–SiC planar and trench MOS capacitors annealed in ambient NO

The interface properties and electrical characteristics of the n-type 4H-SiC planar and trench metal–oxide–semiconductor(MOS) capacitors are investigated by measuring the capacitance voltage and current voltage. The flat-band voltage and interface state density are evaluated by the quasi-static method. It is not effective on further improving the interface properties annealing at 1250℃ in NO ambient for above 1 h due to the increasing interface shallow and fast states.These shallow states reduce the effective positive fixed charge density in the oxide. For the vertical MOS capacitors on the(1120) and(1100) faces, the interface state density can be reduced by approximately one order of magnitude, in comparison to the result of the planar MOS capacitors on the(0001) face under the same NO annealing condition. In addition, it is found that Fowler–Nordheim tunneling current occurs at an oxide electric field of 7 MV/cm for the planar MOS device.However, Poole–Frenkel conduction current occurs at a lower electric field of 4 MV/cm for the trench MOS capacitor. This is due to the local field crowded at the trench corner severely causing the electrons to be early captured at or emitted from the SiO_2/Si C interface. These results provide a reference for an in-depth understanding of the mobility-limiting factors and long term reliability of the trench and planar SiO_2/Si C interfaces.     

Fabricating GeO2passivation layer by N2O plasma oxidation for Ge NMOSFETs application

In this paper, oxidation of Ge surface by N2O plasma is presented and experimentally demonstrated. Results show that1.0-nm GeO2is achieved after 120-s N2O plasma oxidation at 300?C. The GeO2/Ge interface is atomically smooth. The interface state density of Ge surface after N2O plasma passivation is about ～ 3 × 1011cm-2·eV-1. With GeO2passivation,the hysteresis of metal–oxide–semiconductor(MOS) capacitor with Al2O3serving as gate dielectric is reduced to ～ 50 mV,compared with ～ 130 mV of the untreated one. The Fermi-level at GeO2/Ge interface is unpinned, and the surface potential is effectively modulated by the gate voltage.     

Fermi Level Unpinning and Schottky Barrier Modification by Ti, Sc and V Incorporation at NiSi2/Si Interface

A new method is proposed to modify the Schottky barrier height （SBH） for nickel silicide/Si contact. Chemical and electrical properties for NiSi2/Si interface with titanium, scandium and vanadium incorporation are investigated by first-principles calculations. The metal/semiconductor interface states within the gap region are greatly decreased, which is related to the diminutions of junction leakage when Ti-cap is experimentally used in nickel silicide/Si contact process. It leads to an unpinning metal/semiconductor interface. The SBH obeys the Schottky-Mort theory. Compared to Ti substitution, the SBH for electrons is reduced for scandium and increases for vanadium.     

Effective dielectric constant model of electromagnetic backscattering from stratified air–sea surface film–sea water medium

Studies of surface film medium on the sea surface are carried out in this paper for developing the technology to automatically detect and classify sea surface films, and an effective dielectric constant model of electromagnetic backscattering from a stratified air–ocean interface. Numerical results of the new model show the characteristics of effective dielectric constants for the air–sea surface film–sea water medium as follows. The effective dielectric constants decrease with increasing relative dielectric constants of the sea surface films. The effective dielectric constants decrease in horizontal polarization(abbr. HH polarization) and increase in VV vertical polarization(abbr. VV polarization) with increasing radar incident angle. Effective dielectric constants vary with relative sea surface film thickness as a cosinusoidal function of sea surface film thickness. Effective dielectric constant of VV polarization is larger than that of HH polarization. Two potential applications are found with our model, i.e., the retrieval of dielectric constants from the sea surface film, and the film thickness retrieval with our model. Our model has a highly significant influence on improving the technology related to the remote sensing of sea surface films.     

Quantum-Mechanical Study on Surrounding-Gate Metal-Oxide-Semiconductor Field-Effect Transistors

As the channel length of metal-oxide-semiconductor field-effect transistors （MOSFETs） scales into the nanometer regime, quantum mechanical effects are becoming more and more significant. In this work, a model for the surrounding-gate （SG） nMOSFET is developed. The SchrSdinger equation is solved analytically. Some of the solutions are verified via results obtained from simulations. It is found that the percentage of the electrons with lighter conductivity mass increases as the silicon body radius decreases, or as the gate voltage reduces, or as the temperature decreases. The eentroid of inversion-layer is driven away from the silicon-oxide interface towards the silicon body, therefore the carriers will suffer less scattering from the interface and the electrons effective mobility of the SG nMOSFETs will be enhanced.     

Effects of electron–optical phonon interactions on the polaron energy in a wurtzite ZnO/Mg_xZn_(1-x)O quantum well

We investigated the properties of polarons in a wurtzite ZnO/MgxZn1-xO quantum well by adopting a modified Lee–Low–Pines variational method, giving the ground state energy, transition energy, and phonon contributions from various optical-phonon modes to the ground state energy as functions of the well width and Mg composition. In our calculations, we considered the effects of confined optical phonon modes, interface-optical phonon modes, and half-space phonon modes, as well as the anisotropy of the electron effective band mass, phonon frequency, and dielectric constant. Our numerical results indicate that the electron–optical phonon interactions importantly affect the polaronic energies in the ZnO/MgxZn1-xO quantum well. The electron–optical phonon interactions decrease the polaron energies. For quantum wells with narrower wells, the interface optical phonon and half-space phonon modes contribute more to the polaronic energies than the confined phonon modes. However, for wider quantum wells, the total contribution to the polaronic energy mainly comes from the confined modes. The contributions of the various phonon modes to the transition energy change differently with increasing well width. The contribution of the half-space phonons decreases slowly as the QW width increases, whereas the contributions of the confined and interface phonons reach a maximum at d ≈ 5.0 nm and then decrease slowly. However,the total contribution of phonon modes to the transition energy is negative and increases gradually with the QW width of d.As the composition x increases, the total contribution of phonons to the ground state energies increases slowly, but the total contributions of phonons to the transition energies decrease gradually. We analyze the physical reasons for these behaviors in detail.     

Effect of PECVD SiN_x/SiO_yN_x–Si interface property on surface passivation of silicon wafer

It is studied in this paper that the electrical characteristics of the interface between Si O_y N_x/Si N_x stack and silicon wafer affect silicon surface passivation. The effects of precursor flow ratio and deposition temperature of the Si O_y N_x layer on interface parameters, such as interface state density Ditand fixed charge Qf, and the surface passivation quality of silicon are observed. Capacitance–voltage measurements reveal that inserting a thin Si O_y N_x layer between the Si N_x and the silicon wafer can suppress Qfin the film and Ditat the interface. The positive Qfand Ditand a high surface recombination velocity in stacks are observed to increase with the introduced oxygen and minimal hydrogen in the Si O_y N_x film increasing. Prepared by deposition at a low temperature and a low ratio of N_2O/Si H_4 flow rate, the Si O_y N_x/Si N_x stacks result in a low effective surface recombination velocity(Seff) of 6 cm/s on a p-type 1 ?·cm~(–5) ?·cm FZ silicon wafer.The positive relationship between Seffand Ditsuggests that the saturation of the interface defect is the main passivation mechanism although the field-effect passivation provided by the fixed charges also make a contribution to it.     

Effective Dielectric Response of Nonlinear Composites of Coated Metal Inclusions

The effective dielectric response of the composites in which nondilute coated metal particles are randomly embedded in a linear host is investigated. Two types of coated particles are considered, one is that the core is nonlinear, the other is that the shell is nonlinear. We derive general expressions for the effective linear dielectric function and the effective third-order nonlinear susceptibility, and take one step forward to perform numerical calculations on the coated metal/dielectric composites. Numerical results show that the effective linear and nonlinear dielectric responses can be greatly enhanced near the surface plasmon resonant frequency. Moreover, the resonant peaks are found within a range from 0.46ωp to 0.57ωp for spherical particles and from 0.59ωp to 0.70ωp for cylindrical inclusions. In the frequency region, the resonant peak can achieve the maximum, according to an optimal structural parameter and volume fraction. The resonant frequency exhibits a redshift with the increasing structural parameter k or volume fraction f or dimensionality factor D.     

Nonequilibrium kinetics effects in Richtmyer–Meshkov instability and reshock processes

Kinetic effects in the inertial confinement fusion ignition process are far from clear. In this work,we study the Richtmyer–Meshkov instability and reshock processes by using a two-fluid discrete Boltzmann method. The work begins by interpreting the experiment conducted by Collins and Jacobs(2002, J. Fluid Mech. 464, 113–136). It shows that the shock wave causes substances in close proximity to the substance interface to deviate more significantly from their thermodynamic equilibrium state. The ...     

Electronic structure of Yb/H－Si (111) interface

Photoemission spectra are measured for Yb covered surface of wet-chemically-etched H－Si (111). The results reveal that the lattice structure of the H－Si (111) surface is stable against the deposition of Yb atoms. X-ray photoemission spectra indicate the formation of a polarized (dipole) surface layer, with the silicon negatively charged. Ultraviolet photoemission spectra exhibit the semiconducting property of the interface below one monolayer coverage. Work function variation during the formation of the Yb/H－Si (111) interface is measured by the secondary-electron cutoff in the ultraviolet photoemission spectral line. The largest decrease of work function is ～1.65eV. The contributions of the dipole surface layer and the band bending to the work function change are determined to be ～1.15eV and ～0.5eV, respectively. The work function of metal Yb is determined to be ～2.80±0.05eV.     

Binding energies of impurity states in strained wurtzite GaN/Al_xGa_(1-x)N heterojunctions with finitely thick potential barriers

Ground state binding energies of donor impurities in a strained wurtzite GaN/AlxGal_xN heterojunction with a po- tential barrier of finite thickness are investigated using a variational approach combined with a numerical computation. The built-in electric field due to the spontaneous and piezoelectric polarization, the strain modification due to the lattice mismatch near the interfaces, and the effects of ternary mixed crystals are all taken into account. It is found that the binding energies by using numerical wave functions are obviously greater than those by using variational wave functions when impurities are located in the channel near the interface of a heterojunction. Nevertheless, the binding energies using the former functions are obviously less than using the later functions when impurities are located in the channel far from an interface. The difference between our numerical method and the previous variational method is huge, showing that the former should be adopted in further work for the relevant problems. The binding energies each as a function of hydrostatic pressure are also calculated. But the change is unobvious in comparison with that obtained by the variational method.     

Generalized Harmonic Oscillator and the SchrSdinger Equation with Position-Dependent Mass

We study the generalized harmonic oscillator that has both the position-dependent mass and the potential depending on the form of mass function in a more general framework. The explicit expressions of the eigenvalue and eigenfunction for such a system are given, they have the same forms as those for the usual harmonic oscillator with constant mass. The coherent state and its properties corresponding effective potentials for several mass functions, for the system with PDM are also discussed. We give the the systems with such potentials are isospectral to the usual harmonic oscillator.     

OPTICAL-PHONON MODES AND ELECTRON-OPTICAL-PHONON INTERACTION IN A COUPLED QUANTUM WELL

By applying the dielectric continuum model, optical-phonon modes of the lattice vibration and a complete interaction Frohlich-like Harniltonian between an electron and the optical phonons including the interface phonons, the confined LO phonons and the half-space LO phonons are derived for a general coupled quantum well (GCQW) structure of polar crystals. The dispersion curves of the interface modes and the electron-interface-phonon coupling function as functions of coordinate z and wavenumber k are given and discussed for a GCQW. We find that there are eight (not ten) frequency solutions for the interface optical-phonon modes in GCQW and that, in the long-wavelength limit, the longitudinal and transverse modes in the two side materials 1 and 5 are forbidden and two new frequency solutions ω± are obtained instead. Moreover, we also find that the electron-interface-phonon coupling functions are complicated functions of k and that the phonons with long wavelengths are important and the higher-frequency modes are more important than the lower-frequency modes for the electron-phonon interaction.     

Electrical and dielectric properties of Al/p-Si and Al/perylene/p-Si type diodes in a wide frequency range

The perylene(C20H12) layer effect on the electrical and dielectric properties of Al/p-Si(MS) and Al/perylene/p-Si(MPS) diodes have been investigated and compared in the frequency range of 0.7 kHz–2 MHz. Experimental results show that C–V characteristics give an anomalous peak for two structures at low frequencies due to interface states(Nss) and series resistance(Rs). The increases in C and G/ω at low frequencies confirm that the charges at interface can easily follow an ac signal and yield excess capacitance and conductance. The frequency-dependent dielectric constant(ε) and dielectric loss(ε) are subtracted using C and G/ω data at 1.5 V. The ε and ε values are found to be strongly dependent on frequency and voltage, and their large values at low frequencies can be attributed to the excess polarization coming from charges at traps. Plots of ln(σac)–ln(ω) for two structures have two linear regions, with slopes of 0.369 and 1.166 for MS, and of 0.077 and 1.061 for MPS, respectively. From the C 2–V characteristics, the doping acceptor atom concentration(NA) and barrier height(ΦB) for Schottky barrier diodes(SBDs) of MS and MPS types are also obtained to be 1.484 × 1015 and 1.303 × 1015cm 3, and 1.10 and 1.13 eV, respectively.     

Influence of spherical inclusions on effective thermoelectric properties of thermoelectric composite materials

A homogenization theory is developed to predict the influence of spherical inclusions on the effective thermoelectric properties of thermoelectric composite materials based on the general principles of thermodynamics and Mori–Tanaka method. The closed-form solutions of effective Seebeck coefficient, electric conductivity, heat conductivity, and figure of merit for such thermoelectric materials are obtained by solving the nonlinear coupled transport equations of electricity and heat. It is found that the effective figure of merit of thermoelectric material containing spherical inclusions can be higher than that of each constituent in the absence of size effect and interface effect. Some interesting examples of actual thermoelectric composites with spherical inclusions, such as insulated cavities, inclusions subjected to conductive electric and heat exchange and thermoelectric inclusions, are considered, and the numerical results lead to the conclusion that considerable enhancement of the effective figure of merit is achievable by introducing inclusions. In this paper, we provide a theoretical foundation for analytically and computationally treating the thermoelectric composites with more complicated inclusion structures, and thus pointing out a new route to their design and optimization.     

Probe the 2s（1/2） and 1d（3/2） state level inversion with electron-nucleus scattering

The neutron-rich even-even nuclei^26–40Mg,^28–46Si,^30–48S, and ^32–56Ar are calculated with the RMF model and the phase-shift electron scattering method. Results show that level inversion of the 2s1/2 and 1d3/2 proton states may occur for the magnesium, silicon, sulphur, and argon isotopes with more neutrons away from the stability line. Calculations show that the variation of the central charge densities for30–48S, and32–56Ar are very sensitive to the 2s1/2 and 1d3/2 proton state level inversion, and the level inversion can lead to a large measurable central charge depletion to the charge density distributions for the neutron-rich isotopes. Calculations also show that the charge density diferences between the isotopes with and without central charge depletion can reveal not only the level inversion of the 2s1/2 and 1d3/2 proton states but also the behavior of the proton wave functions of both states. The results can provide references for the possible study of the nuclear level inversion and nuclear bubble phenomenon with electron scattering of short-lived nuclei at RIKEN or/and GSI in the future. In addition, direct nuclear reaction 44S（n, d）43P or44S（3H, α）43P might also be a possible way to study the 2s1/2 and 1d3/2 proton state level inversion.     

Probe the 2s_(1/2) and 1d_(3/2) state level inversion with electron-nucleus scattering

The neutron-rich even-even nuclei26–40Mg,28–46Si,30–48S,and32–56Ar are calculated with the RMF model and the phase-shift electron scattering method.Results show that level inversion of the 2s1/2and 1d3/2proton states may occur for the magnesium,silicon,sulphur,and argon isotopes with more neutrons away from the stability line.Calculations show that the variation of the central charge densities for30–48S,and32–56Ar are very sensitive to the 2s1/2and 1d3/2proton state level inversion,and the level inversion can lead to a large measurable central charge depletion to the charge density distributions for the neutron-rich isotopes.Calculations also show that the charge density diferences between the isotopes with and without central charge depletion can reveal not only the level inversion of the 2s1/2and 1d3/2proton states but also the behavior of the proton wave functions of both states.The results can provide references for the possible study of the nuclear level inversion and nuclear bubble phenomenon with electron scattering of short-lived nuclei at RIKEN or/and GSI in the future.In addition,direct nuclear reaction44S(n,d)43P or44S(3H,α)43P might also be a possible way to study the 2s1/2and 1d3/2proton state level inversion.     

Effect of overdrive voltage on PBTI trapping behavior in GaN MIS-HEMT with LPCVD SiNx gate dielectric

The effect of high overdrive voltage on the positive bias temperature instability(PBTI)trapping behavior is investigated for GaN metal–insulator–semiconductor high electron mobility transistor(MIS-HEMT)with LPCVD-SiNx gate dielectric.A higher overdrive voltage is more effective to accelerate the electrons trapping process,resulting in a unique trapping behavior,i.e.,a larger threshold voltage shift with a weaker time dependence and a weaker temperature dependence.Combining the degradation of electrical parameters with the frequency–conductance measurements,the unique trapping behavior is ascribed to the defect energy profile inside the gate dielectric changing with stress time,new interface/border traps with a broad distribution above the channel Fermi level are introduced by high overdrive voltage.     

Enhanced interface properties of diamond MOSFETs with Al_2O_3 gate dielectric deposited via ALD at a high temperature

The interface state of hydrogen-terminated(C–H) diamond metal–oxide–semiconductor field-effect transistor(MOSFET) is critical for device performance. In this paper, we investigate the fixed charges and interface trap states in C–H diamond MOSFETs by using different gate dielectric processes. The devices use Al_2O_3 as gate dielectrics that are deposited via atomic layer deposition(ALD) at 80℃ and 300℃, respectively, and their C–V and I–V characteristics are comparatively investigated. Mott–Schottky plots(1/C~2–VG) suggest that positive and negative fixed charges with low density of about 1011 cm~(-2) are located in the 80-℃-and 300-℃ deposition Al_2O_3 films, respectively. The analyses of direct current(DC)/pulsed I–V and frequency-dependent conductance show that the shallow interface traps(0.46 e V–0.52 e V and0.53 e V–0.56 e V above the valence band of diamond for the 80-℃ and 300-℃ deposition conditions, respectively) with distinct density(7.8 × 10~(13) e V~(-1)·cm~(-2)–8.5 × 10~(13) e V-1·cm~(-2) and 2.2 × 1013 e V~(-1)·cm~(-2)–5.1 × 10~(13) e V~(-1)·cm~(-2) for the80-℃-and 300-℃-deposition conditions, respectively) are present at the Al_2O_3/C–H diamond interface. Dynamic pulsed I–V and capacitance dispersion results indicate that the ALD Al_2O_3 technique with 300-℃ deposition temperature has higher stability for C–H diamond MOSFETs.