We propose a new adaptation minority game for understanding of the complex dynamical behaviour characterized by agent interactions competing limited resource in many natural and social systems. Intelligent agents may modify a part of strategies held by them periodically, depending on the strategy performances. In the new model, the strategies will be updated according to uniform-crossover variation process inspired by genetic evolution algorithm in biology. The performances of the agents in the new model are calculated for different parameter conditions. It has been found that the new system may evolve via the strategy uniform crossover adaptation mechanism into a frozen equilibrium state in which the performance of the system may reach the best limit, implying the strongest cooperation among agents and the most effective utilization of the social resources.

The thermodynamic properties of proton transport along hydrogen-bonded systems at finite temperatures have been studied by our model. We first derive the dynamic equations of the proton transport and find the solutions and the free energy of the systems. Finally we obtain the specific heats of the hydrogen bonded systems, resulting from the motion of the soliton, by using transfer integral way. The theoretical value is basically consistent with the experimental data.

We present an “all versus nothing”to show Bell's theorem without inequalities involving 2n observers. As a by-product, we also shows that quantum mechanics can violate Bell's inequality by a constant instead of by an amount that grows exponentially with 2n.

We study the discrimination of quantum states from the other way around, i.e. the likeness of two quantum states. The fidelity is used to describe the likeness of two quantum states. Then we present a scheme to obtain the fidelity of two unknown qubits directly from the integral area of the spectra of the assistant qubit (spin) on a nuclear-magnetic-resonance quantum information processor. Finally we demonstrate the scheme on a three-qubit quantum information processor. The experimental data are consistent with the theoretical expectation with an average error of 0.05, which confirms the scheme.

The particle-number fluctuations due to collective excitations are calculated for a one-dimensional interacting Bose-condensed gas confined in a harmonic potential. By using the Thomas-Fermi
approximation, the behaviour of the anomalous fluctuations is found to be <δ²N₀> ～ N^10/3/( ln N)² with N being the total number of particles.

Some locally rotationally symmetric Bianchi type-I cosmological solutions for a cloud string with bulk viscosity are presented. In the first case, an equation of state ρ = kλ and the relation between metric potentials R = ASⁿ are considered, and the solution represents shearing non-rotating model with the bulk viscosity ∝ ρ^1/2, where ρ is the rest energy density of the cloud of strings with particles attached to them, λ is the tension density of the cloud of strings, is the coefficient of the bulk viscosity, R and S are only the functions of time t, while A and k are constant. In the second case, the constant coefficient of bulk viscosity is considered.

The property and gravitational field of global string of tachyon matter are investigated in four-dimentional approximately cylindrically-symmetric spacetime with a deficit angle. In particular, we give an exact solution of the tachyon field in the flat spacetime background and we also find the solution of the metric in the linearized approximation of gravity.

We study the effects of correlations between the real and imaginary parts of quantum noise on the intensity fluctuation for a single-mode laser. The analytic expressions of the intensity correlation function C(τ) and the associated relaxation time T_c in the case of a stable locked phase resulting from the cross-correlation λ_q between the real and imaginary parts of quantum noise are derived by means of projection operator method. Based on numerical computations it is found that the presence of cross-correlations between the real and imaginary parts of quantum noise causes the intensity fluctuation to increase. A slowing-down phenomenon exists in the sense that T_c increases as a function of |λ_q|. Thus the decay of intensity fluctuation becomes slower.

The main problem of the traditional radiation pyrometry is that fatal errors will be caused by the unknown or varying emissivity. Based on the combined neural networks (CNNE model), we propose an improved method for emissivity modeling. The model structure and the optimum algorithm are described. This method being used, the spectral emissivity and temperature can be fast computed accurately from the spectral radiation measured.

We treat Haldane’s model by the standard Dirac procedure. After canonical quantization, the coordinates of the particle become noncommutative, and the commutation relations for the coordinates coincide exactly with those defining a fuzzy two-dimensional sphere.

We study the energy density in quark-gluon plasma. At the very high temperature, the quark matter is a hot and dense matter in the color deconfinement condition, and quarks can coalescent diquarks. Energy density of this system is figured out and compared with the energy density in the other two kinds of situations. Possible energy density is about ε₀ ≈ 2.4 GeV/fm³ according to our estimation for quark matter including diquarks.

The potential energy surface for superheavy nucleus has been studied within the framework of the constrained relativistic mean field theory, and the shell correction energy as a function of deformation has been extracted by the Strutinsky shell correction procedure. Contrary to the usual expectation, the shell correction energy at the saddle point is too important to be neglected, and it has essential contribution to the fission barrier in superheavy nucleus.

The β-NMR (nuclear magnetic resonance) and β-NQR (nuclear quadrupole resonance) technique and its experimental set-up have been developed for the first time in China. The lifetime, magnetic moment and spin polarization of ¹²B were determined. The experimental results show the reliability of this newly developed β-NMR and β-NQR set-up.

The structure of superheavy nuclei has been studied using the
macroscopic-microscopic model. The macroscopic energy was calculated with the continuous medium model in which the energy is expressed as a function of nucleon densities. The deformations and structures of superheavy nuclei were systematically investigated. Calculations reproduce well the available data of experimental α-decay energies and half-lives. The investigation of single-particle levels shows that the shell structure is deformation and isospin dependent. Potential-energy-surface calculations display that superheavy nuclei have in general harder shapes than the nuclei of other mass regions.

The mass dependence of the limiting temperatures for finite nuclear systems is investigated. Our study shows that the different parameter sets of Skyrme interaction and the form of surface tension can influence the limiting temperature obviously.

The full-potential linearized augmented plane wave method was applied to study the electronic and the magnetic properties of the compound pipz-H₂[MnF₄(HF₂)](pipz=piperazine). The band structure, the total density of states, the partial density of states and the electron density were calculated to explain the electronic and the magnetic properties of pipz-H₂[MnF₄(HF₂)] in the ferromagnetic state. It is found that the magnetic moment of the molecule mainly comes from the Mn atoms with partial contribution from the F atoms. The symmetrical σ/σ bonds via H atoms along Mn-F-H-F-Mn chains and the weak direct-exchange interaction between F(2), F(3) and Mn atoms have effect on the electronic structure and the magnetism of pipz-H₂[MnF₄(HF₂)].

We introduce a new method of frequency-shifting for a diode laser in laser cooling experiments, the method is based on the Zeeman effect of ⁸⁷Rb atoms. The laser frequency is stabilized by absorption spectrum line of atoms in magnetic field. We show that a magnetic field can be added up to 10^-2T. The corresponding frequency shift is 10²MHz and the response time is about 1 ms. The large range of the frequency shift is sufficient for laser-cooling experiments.

The electron energy loss spectrum of argon in the energy region of 24.5-30.5 eV was measured at 2.5 keV impact energy. The line profile parameters of the optically forbidden excitations of 3s^-1 ns (n = 4-6) and 3s^-1 nd (n = 3-7) of argon, i.e., E_γ, Γ, q and μ, were determined.

One-dimensional photonic crystals with two or more structural defects are studied. We observed an interesting characteristic of transmission band structure of photonic crystals with defects using the transmission-matrix-method simulation. The transmission states in the wide photonic band gap caused by defects reveal degeneracy and split in certain conditions. Every split state is contributed by coupling of all defects in a photonic crystal. Using the tight-binding method, we obtain an approximate analytic expression for the split frequency of photonic crystals with two structural defects.

We show theoretically that the group velocity of light pluses can be reduced significantly by use of the steep dispersion properties of the phase coupling effect in the photorefractive two-wave mixing process. The group velocity of light pulses of the order of 0.1m/s can be achieved in typical photorefractive BSO crystals with an appropriate externally applied electric field and moving gratings of appropriate speeds. It is also shown that the slowly propagating light pulses can be set to be amplified after passing through the photorefractive material.

A spatiotemporal non-paraxial correction to the paraxial solution of ultrashort pulsed beam is obtained by using the Fourier transform and the Taylor expansion. By studying the propagation of an isodiffracting pulsed Gaussian beam with different pulse shapes, we find that there are spectrum-induced changes in the non-paraxial propagation of the pulsed beam. We analyse the influence of pulse spectrum on the non-paraxial property of the ultrashort pulsed beam and explain it base on the paraxial approximation condition.

A novel photopolymer storage medium for holographic storage is presented. In particular, the refractive index of the monomer methyl methacrylate is much lower than that of the binder epoxy resin solidified by m-xylylenediamine, and the material can be developed into samples with dimension stability and high optical quality and thickness of several millimeters or more. Some none-polymeric macromolecule impurity is uniformly mixed into the medium to aid the monomer diffusing and to enhance the diffraction efficiency. A holographic recording medium of 0.8 mm thickness has been fabricated, and relative theoretical analysis and experiments are presented.

We investigate continuous-wave four-wave mixing via electromagnetically induced transparency in double-Λ configuration by using Schrödinger-Maxwell formalism. We derive the corresponding explicit analytical expressions.

We report a cw Er³⁺:Yb³⁺ co-doped phosphate glass laser pumped by a laser diode. The maximum output power of 78.3 mW and a slope efficiency of 15.25% were achieved. The laser spectral region was from 1532 nm to 1535 nm, with the peak laser wavelength at 1534 nm. The laser modes and time stability were also measured. The thermal effect had little influence on the output in our experiment.

A passively Q-switched Yb: YAG microchip laser has been constructed by using a doped GaAs as the saturable absorber as well as the output coupler. At 13.5 W of pump power the device produces high-quality 3.4μJ 52 ns pulses at 1030 nm with a pulse repetition rate of 7.8 kHz in a TEM₀₀-mode.

Optical nonlinearities of new organophosphorus fullerene derivative were determined by the Z-scan method with a pulsed Q-switch Nd:YAG laser at 532 nm. The experimental results demonstrated that the derivative has much larger excited-states nonlinear absorption and nonlinear refraction than C₆₀. A five-level model was utilized to fit the experiemntal data, and a good agreement is reached. Some parameters such as excited-state absorption cross and refraction cross were obtained. To our knowledge, the excited-state cross section of new organophosphorus fullerene derivative and its effective ratio to the ground-state cross section are the largest values among the fullerene derivatives reported to date.

We propose a novel approach to visualize the light-induced refractive index changes in photorefractive crystals employing digital holography. The holograms formed in a Mach-Zehnder interferometer are recorded by a two-dimensional CCD camera. From these holograms, the phase differences, which contain the information of the index changes in photorefractive crystals, are determined by utilizing digital holographic interferometry. Then the two-dimensional visualizations of index changes in the crystals can be obtained. This method is successfully emonstrated in LiNbO₃:Fe, KNSBN:Ce and SBN:Cr crystals.

Colloidal Au less than 20 nm was prepared in an aqueous solution of HAuCl₄ by irradiation with a near-infrared femtosecond laser. The absorption peak around 526 nm in the absorption spectra and the peaks 2θ = 38.3°and 2θ = 24°in the x-ray diffraction (XRD) pattern indicated the formation of Au colloid. Nonlinear optical properties of Au colloid were investigated by using the Z-scan technique with 8 ns pulses at 532 nm. The transition characteristics from self-defocusing to self-focusing and from saturable absorption (SA) to reverse saturable absorption (RSA) were observed for the Au colloid.

We analyse the relationship of conversion efficiency with the inter-crystal phase shift by the heuristic theory and propose a novel configuration of two cascaded nonlinear crystals for the second-harmonic generation with the polarization modulation. With this configuration, 70% external doubling efficiency is obtained, which is, to the best of our knowledge, the highest conversion efficiency with LBO crystal external frequency doubling. This configuration provides a simple and effective method to improve the second harmonic conversion efficiency.

We have measured the third-order optical nonlinearity of the organic-inorganic hybrid materials of supermolecules (C₁₉H₂₅N₂)₇H_n[P₂Mo_18-nV_nO₆₂] (n =1, 2, 4, 5) and their reactants (C₁₉H₂₅N₂)Br and H_n+6[P₂Mo_18-nV_nO₆₂] by using the ultrafast optical Kerr effect at 830 nm wavelength, with time resolution of 115 fs. By comparing the γ values of the supermolecules and their reactants, it is concluded that the charge-transfer enhanced remarkably the third-order optical nonlinearity of the compounds. With the increment of the number of the vanadium atoms, the electron distribution changed strongly and the γ value decreased dramatically.

Optical absorption, electron spin resonance and photoluminescence spectra were used to study the defect formation in high purity fused silica induced by a focused infrared femtosecond laser with the input intensity below damage threshold. Si E' centres were formed in the silica. The number of Si E' centre was found increasing linearly with power density of the fs laser, deviated from the increment photo-induced free carrier density. We concluded that the colour centres were formed at the defect sites that were newly generated by radiolysis of silica tetrahedral network and displacement of oxygen between two silicon atoms. Material structure is already modified even though the irradiation fs laser power was well below the damage threshold.

We propose a model for two-dimensional photonic materials constructed on a kind of fractal structures. Total energy flows of the photonic materials constructed with either dielectric cylinders or air holes show that both of them possess absolute gaps for the electromagnetic waves.

By taking advantage of the approximate approach of small amplitude soliton, we study the higher-order nonlinear Schrödinger equation in an optical fibre. Our results show that the bright and dark solitons of small amplitude can appear on the background of a continuous wave in normal dispersion regime or in anomalous dispersion regime simultaneously due to the higher order effects. Interesting connection between the higher-order nonlinear Schrödinger equation and the Korteweg-de Vries equation is also demonstrated.

Correlations among multiple noise perturbations are considered in a soliton system. It is found that these perturbations lead to disintegration of soliton, their correlations play important roles and reinforce their effects. Furthermore, there is a limiting case among the multiple noise perturbations in the system, here their effect is the largest. Finally the nonlinear gain and the narrow band filter are introduced to suppress these effects effectively.

Gain and noise figure (NF) are the most important two parameters of an erbium-doped fibre amplifier (EDFA) for a multi-channel wavelength division multiplexing (WDM) transmission system. A simple bi-wavelength method for accurate gain and NF spectrum measurement of EDFA for WDM applications is proposed. A saturating input signal, whose power equals to the sum of all the WDM signal power and whose wavelength is determined by the channel numbers, and the wavelength of the WDM input signals saturates the EDFA in a degree as the same as the WDM signals input. Meanwhile, a small power probe signal scans and measures the gain and the NF value at every wavelength of the WDM input signals. Investigative results by numerical simulation show that the gain and the NF spectra measured by this method have good agreement with the real spectra of the WDM signal input in a large total input power range. The maximum errors of the gain and the NF are less than 0.2 dB and 0.16 dB, respectively, for a 50-channel input case. The method is competent for the accurate gain and the NF spectrum measurement of the fibre preamplifier and the line-amplifier for WDM applications and has the advantages of simplicity, convenience and easy implement.

By applying ion Bernstein wave (IBW) heating into the lower hybrid current drive (LHCD) plasma, improved confinements have been obtained in the HT-7 tokamak. The central electron temperature was doubled and the storage energy was increased significantly. The core electron density and temperature were broadened and their profiles near the edge were steepened. A transport barrier has been formed in the vicinity of the limiter radial location. An enhanced shear in poloidal phase velocity was found in the same region with reduction of the fluctuation levels and the coherences between fluctuations. The results suggest that the improved confinement in the IBW and LHCD plasma is at least partially due to the modification of shear in poloidal velocity and then the suppression of fluctuations and fluctuation induced fluxes via de-correlation effect.

Based on the Born approximation, we reduce the approximate analysis solution to the normal and oblique incident electromagnetic wave scattering from the weakly ionized plasma layer shielded by a conducting surface. The solution is closely related to the density profile of the plasma layer. Employing the self-consistent base function, we yield the optimal density profile for the nonuniform plasma layer with the frequencies of incident electromagnetic waves ranging from 4-10GHz. Numerical studies illustrate the optimal density profile can “survive”wide ranges of the plasma parameters. Different from the validity condition for the Wenzell-Kramers-Brillouin-Jeffreys (WKBJ) approximation, the Born approximation is feasible even if the scale length is smaller than the wavelength. Therefore, the Born approximation is universal against the scattering problem from the weakly ionized plasma.

We report the experimental results of picosecond pulse laser microstructuring (pulse duration 35 ps, wavelength 1.06 μm, repetition rate 10 Hz) of silicon using the direct focusing technique. Arrays of sharp conical spikes located below the initial surface have been formed by cumulative picosecond pulsed laser irradiation of silicon in SF₆. Irradiation of silicon surface in air, N₂, or vacuum creates ripple-like patterns, but does not create the sharp conical spikes.

The characteristics of the microwave discharge are studied in a modified surfaguide with a large diameter. Experimental results show that there exist three discharging modes, one is the plasma mode, and the others are the waveguide modes. The discharge can jump between one of the waveguide modes and the plasma mode, and the corresponding hysteresis loop is influenced by the discharging pressure. In the higher pressure region, the hysteresis loop is wide enough so that the discharge in each mode is stable. In the middle pressure region, the discharge becomes unstable as a result of the hysteresis loop being sufficiently narrow. When the gas pressure is further decreased, the plasma mode disappears, while the mode jumps between the two waveguide modes always appear and are stable in the discharge region we have explored.

Neutral beam injection (NBI) experiments have been carried out with two operation modes of a bucket ion source in the HL-1M tokamak. During the first mode, more than 30% rise in ion temperature above the Ohmic level is routinely achieved after NBI power about 0.5 MW is injected. Ion temperature only increases 20-30% for the second operation mode, which is often limited by current termination. The heating effects of the NBI have been analysed experimentally and theoretically. The performance of the NBI system is well described.

Observation of two ion-acoustic waves via Thomson scattering can provide precise measurements of plasma parameters. The conditions for the observation of two ion-acoustic modes in a two-ion plasma are discussed. The ratio of electron temperature T_e to ion temperature T_i is the critical parameter for the presence of two ion-acoustic modes, which should be in the range of 4/Z_L T_e/T_i 2A_H/Z_HA_L, where Z_L,H are the charge states of light and heavy ions, and A_L,H are the atomic numbers of light and heavy ions, respectively. As the temperature ratio varies in this range, the concentration of heavy ions must increase with the ratio T_e/T_i so that the two ion-acoustic modes can have the same fluctuation levels.

Pulsed molybdenum ion beams extracted from a metal vapor vacuum arc ion source at voltage of 25 kV or 48 kV were implanted into H13 steel with a high implantation dose of 5 x 10¹⁷ ions.cm^-2 and a time-averaged ion beam current density of about 300μA.cm^-2. We have investigated the steel implanted for wear resistance by an optical interference microscope and a pin-on-disc apparatus. The Rutherford backscattering spectroscopy demonstrated that rather low energy ions could penetrate quite deep into the substrates. It was observed by x-ray photoelectron spectroscopy and transmission electron microscopy that carbide of molybdenum appeared in the doped region. The results showed that dramatically improved wear resistance of H13 steel after molybdenum ion implantation at 48 kV was attributed to the development of Mo₂C precipitates in the doped zone and to the formation of the implantation affected zone below the doped zone.

An improved apparatus is developed from Ke-pendulum. This new apparatus, resonant absorption mechanical spectrometer (RAMS), can measure the internal friction of solids under a forced vibration mode and the measuring frequency can change quasi-continually from a frequency that is much lower than the resonant frequency of the pendulum system, f_r, to the one that is much higher than f_r. The internal friction measurement is able to cover the frequency range from 10^-3Hz to kHz. The measurement method and the calculation formula of the internal friction measured by a RAMS in the full frequency range are derived. A series of resonant absorption peaks are observed in copper, aluminum, zinc, iron samples by the RAMS. The resonant absorption characteristics of the copper sample are studied in details. The experimental results indicate that the position (frequency) of the resonant absorption peaks are independent of the resonant frequency of the pendulum system. The reality of resonant absorption mechanical spectra is discussed and an inference based on the experimental results is presented such that the RAMS is able to characterize some feature of solid materials.

We report the transmission-electron microscopy study of the defects in wurtzitic GaN films grown on Si(111) substrates with AlN buffer layers by the metal-organic chemical vapor deposition method. The In_0.1Ga_0.9N/GaN multiple quantum well (MQW) reduced the dislocation density by obstructing the mixed and screw dislocations passing through the MQW. No evident reduction of the edge dislocations density by the MQW was observed. It was found that dislocations with screw component can be located at the boundaries of sub-grains slightly in-plane misoriented.

The specific features of electric current variation against time were studied by keeping the anode and cathode pulse voltage to be constant. The maximum thickness of the films fabricated by microarc oxidation was found to be closely related to the voltage. The surface morphology of the coatings can be divided into different stages according to the variation of electric current. A method for obtaining high micro hardness coating has been discussed.

Self-organized InAs quantum wires (QWRs) were fabricated on the step edges of the GaAs (331)A surface by molecular beam epitaxy. The lateral size of InAs QWRs was saturated by the terrace width (i.e., 90 nm) while the size along the step lines increased with the increasing thicknesses of the InAs layers, up to 1100 nm. The height of InAs QWRs varied from 7. nm to 13 nm. The evolution of the morphology of InAs QWRs was attributed to the diffusion anisotropy of In adatoms.

We have measured the fluctuation in local surface potential of GaN epitaxial films having two different types of nanostructure, as-grown islands or etched pits, by Kelvin probe force microscopy. We found that the perimeters of as-grown islands and the internal walls of etchd pits have lower surface potential as compared to the as-grown c-plane. The results show that the crystallographic facets tilted with respect to c-plane have higher work function and are electrically more active than c-surface.

The photoinduced birefringence was observed in the polymer poly[2-(4-(4-cyanophenyl) diazenyl phenyloxy) ethoxyl methacrylate] with a cw 532 nm laser. The azobenzene polymer character has been studied under the conditions of various illuminating time and light intensities. By analysing the processes of reorientation, the effect of laser-induced heating has been introduced to the buildup of photoinduced birefringence in azobenzene-side-chain copolymer. The curves for the buildup of birefringence were fitted with a modified function, i.e., biexponential curves and Gaussian curves. The relationship among all the parameters has also been presented. With the modified fitting function, we obtain a better fitting result.

We calculate the excitonic transition energies and exciton binding energies in ZnO/Mg_xZn_1-xO quantum-well heterostructures with Mg composition x varied from 0.08 to 0.36. The effect of the exciton-phonon interaction on the exciton binding energies is taken into account in the model. For the ZnO/Mg_0.12Zn_0.88O quantum-well structure, we compare the calculated result with the available experimental data at 5 K, and a good agreement is achieved. The excitonic transition energies at room temperature in ZnO/Mg_xZn_1-xO quantum-well heterostructures are also calculated. The results show that when the well width exceeds 50Å, the quantum size effect is neglectable and the excitonic transition energies in ZnO/Mg_xZn_1-xO (with x varied from 0.08 to 0.36) quantum-well heterostructures are close to the value of bulk ZnO. The maximum exciton binding energy as large as 121.1 meV is obtained for the well width of 12.5Å in the ZnO/Mg_0.36Zn_0.64O quantum-well heterostructures.

The master equation for the cluster-size distribution function is solved numerically to investigate the electron-hole droplet formation claimed to be discovered in the direct-gap CuCl excited by picosecond laser pulses [Nagai et al Phys. Rev. Lett. 86 (2001) 5795; J. Lumin. 100 (2002) 233]. Our result shows that for the excitation in the experiment, the average number of pairs per cluster (ANPC) is only around 5.2, much smaller than that (10⁶ typically for Ge) of the well studied electron-hole droplet in indirect-gap semiconductors such as Ge and Si. These results indicate that what measured in CuCl by Nagai et al. may not come from the EHD formed from exciton gas, instead possibly come from some bubbles of excitons in metallic liquid.

Spectral properties of Er³⁺-doped aluminium fluorophosphate glasses are experimentally investigated and calculated. The full width at half maximum is 53 nm and the emission cross section is 0.6 x 10^-20cm². The measured lifetime of the ⁴I_13/2 level is 8±0.4 ms and the temperature difference is about 160°C. These parameters are comparable with those in other glasses, which indicates that Er³⁺-doped aluminium fluorophosphates glass is a promising material for Er³⁺ to realize broadband amplification.

We investigate the absorption, stimulated emission cross section and potential laser parameters of Yb³⁺ doped gernano-silicate glasses (GSY glass). The emission cross section is evaluated by using the measured absorption spectra and the principle of reciprocity. It is found that Yb³⁺ in GSY glasses has high stimulated emission cross section of 0.5-0.65 pm² near 1020 nm and exists long measured fluorescence lifetime of 1.20-2.00 ms. Compared with other glass hosts, the GSY glass exhibits excellent general spectroscopic properties for Yb³⁺-doped double-cladding fibres.

The surface properties of GaN films grown by plasma-assisted molecular beam epitaxy were investigated by using x-ray photoelectron spectroscopy (XPS) and Auger electron spectroscopy, while the depth profile was analysed by the Ar ion sputtering method. The contaminative carbon and silicon are chiefly adsorbed onto the surface while oxygen and aluminum diffuse into the bulk to distribute in a certain depth. The mixture oxides is roughly 0.1μm in thickness. Based on the analytical results of XPS of the GaN films, the Ni/Cr/Au interdigital metal-semiconductor-metal (MSM) structure has been fabricated. It has been found that the contact behaviour of the Ni/Cr/Au/undoped GaN exhibits a linear I-V characteristic under dark and 362-nm light excitation without annealing treatment. The lower resistance of the MSM structure has also been observed.

Several theoretical models are established to simulate the interface roughness in a quantum well. The numerical result shows that the roughness correlation function always deviates from the extensively used Gaussian form to some extent, which depends on what a model is used. The influence of such a deviation on the electronic transport property is investigated by assuming several different analytical forms of the correlation function. It is found that the Fermi wave vector is crucial to determine whether the conductivity depends sensitively on the details of the correlation function.

The injection efficiency of spin-polarized quasiparticles in a superconducting Y-Ba-Cu-O (YBCO) film is investigated by means of current injection into the Nd_0.7Sr_0.3MnO₃/SrTiO₃/YBa₂Cu₃O_7-δ heterostructures. We have successfully prepared six injection junction windows in the YBCO thin film strip, which are all 5 μm wide, but 80 μm, 40 μm, 20 μm, 10 μm, 5 μm, and 2 μm long, respectively. At 16 K, our YBCO thin film strip (80 nm thick) has J_c = 2 x 10⁵A/cm². Under the spin-polarized injection current of I_inj = 0.5 mA, with the decrease of injection junction length L, injection efficiency η(η ΔJ_c/ΔJ_inj; ΔJ_c: the suppression of critical current density due to finite ΔJ_inj increases gradually, till L is not more than 20 μm. In that case, η hardly varies with the change of L, reaching up to about 6. We think that these results are related to the spin diffusion length of spin-polarized quasiparticles in the superconducting thin film. In our heterostructures, the YBCO thin film keeps good superconductivity and the injection junction windows are of proper size, therefore, we have obtained relatively high injection efficiency of spin-polarized quasi-particles.

The grain-oriented C_xCo_1-x (x = 0.9, 0.5) samples were fabricated by the hot-pressing method. The microstrucre was observed by an x-ray diffractometer and a scanning electron microscope. The resistance against the applied magnetic field was measured by a standard four-point probe method at different temperatures. The magnetoresistance and the magnetization ratio were studied as a function of magnetic field in the range of -1800 kA/m-1800 kA/m at different temperatures from 50 K to 300 K. The magnetoresistance of grain-oriented C_xCo_1-x is positive. The maximum positive MR of 98% at 50 K and 34% at 300 K was obtained under 1800 kA/m magnetic field in the C_0.9Co_0.1 sample.

Highly magnetic and air-stable silica-coated Fe particles have been prepared by a rapid and simple method. The specific magnetization of the sample can be as high as 201 Am²/kg even including silica, which is a non-magnetic material. The iron particle coated with silica is passivated and protected from oxidation. The influences of H₂ reduction temperature on structure and magnetic properties are also studied.

We report the space-selective formation of colour centres and refractive index change in LiF crystals at room temperature by an 800 nm Ti:sapphire femtosecond laser irradiation. Optical absorption and photoluminescence spectra have been measured for the crystal before and after the laser irradiation. The absorption spectra indicate F and F⁺₃ colour centres are induced in LiF after the laser irradiation. When the temperature of heat treatment reaches 300°C, the absorption peak disappears, indicating that the induced colour centres are bleached. The induced refractive index change is also evaluated. A mechanism for the formation of colour centres and refractive index change is proposed.

A saturated red-organic-light emitting diode (OLED) has been realized by doping an emitting material both in the hole-transporting layer (HTL) and the electron-transporting layer (ETL) to form double emitting zone. The red dopant, 4-(Dicyanomethylene)-2-t-butyl-6-(1,1,7,7-tetramethyljulolidyl-9-enyl)-4H-pyran (DCJTB), was doped into the N,N'-bis-(1-naphthyl)-N,N'-diphenyl-1,1'-biphenyl-4,4'-diamine (NPB) layer and the tris (8-quinolinolate) aluminum (Alq₃) layer, both of which act as the emitting layers. The optimal device, with a structure of ITO/CuPc/NPB/NPB:DCJTB/Alq₃:DCJTB/Alq₃/LiF/Al, showed good chromaticity coordinates (x = 0.63, y = 0.36) at 8 V. Uniquely, the current efficiency of the device was relatively independent of the drive voltage in a wide range from 8 V to 20 V. That may be helpful to ameliorate the lifetime of the organic electroluminescent devices and to adjust the gray-scale for the future full-color display panel.

Calculations have been carried out on spin-polarized Auger electron emission from a seven-layer chromium (100)-film. The core and valence states concerning the Auger transition are obtained from a self-consistent full-potential linearized augmented plane wave calculation by using a repeated slab geometry. The calculations refer to experiments on the L₃M₂₃V- and L₃VV-transitions that have recently been carried out by Heinzmann et al. The Auger spectrum obtained for the L₃VV transition agrees relatively well with our calculations whereas the observed L₃M₂₃V-Auger structure is considerably wider than that predicted by our theory. Nevertheless, the spin-polarization in the latter case, which is about -13%, is in fair agreement with the experiment, different from the L₃VV-transition where the experiment yields -10% as opposed to the theoretical value of -25%. We give possible reasons for the origin of these discrepancies.

We report a special phenomenon on switching behaviour and the electroluminescence (EL) spectrum shift of doped diamond thin films. Nitrogen and cerium doped diamond thin films were deposited on a silicon substrate by microwave plasma-assisted chemical vapor deposition system and other special techniques. An EL device with a three-layer structure of nitrogen doped diamond/cerium doped diamond/SiO₂ thin films was made. The EL device was driven by a direct-current power supply. Its EL character has been investigated, and a switching behaviour was observed. The EL light emission colour of diamond films changes from yellow (590nm) to blue (454nm) while the switching behaviour appears.

We have grown high density Co clusters with a narrow-sized distribution on the Si₃N₄(0001)-(8 x 8) surface. In the submonolayer regime, Co clusters tend to keep a certain size (～ 1.45 nm in diameter) irrespective of coverage. With increasing coverage above 0.92 ML, two new clusters with certain but larger sizes are formed. This novel growth behaviour can be explained by the quantum size effect [Phys. Rev. Lett. 90 (2003) 185506]. It is found that the Co cluster size distribution can be improved by post annealing. Even at high temperature (700°C), no reaction of Co with Si₃N₄ is observed, indicating that Si₃N₄(0001)-(8 x 8) is a promising substrate for growth of magnetic nanostructures.

The reflected differential interference phase contrast microscope is used to study a growing crystal surface. The surface phenomena on the {111} and {100} faces of Sr(NO₃)₂, such as the propagation of steps, the bunches of surface steps, the impurity stopper and the growth hillocks, have been observed during the crystal growth. It was found that: (1) The macrosteps velocity is from 0.86 μm/s to 9.8 x 10^-2μm/s on the {111} face at σ = 5.33 x 10^-3 to 2.13 x 10^-3. (2) If the propagating directions of the steps are in opposition, the velocity of the macrosteps will be increased after they bunched. These phenomena first provide the evidence for the existence of the mutual acceleration effect of macroscopic steps. (3) The growth hillocks include a concentric step which evidently results from successive acts of a two-dimension nucleation on surface.

Polycrystalline silicon film was directly fabricated at 200°C by the conventional plasma enhanced chemical vapor deposition method from SiCl₄ with H₂ dilution. The crystallization depends strongly on the deposition power. The maximum crystallinity and the crystalline grain size are over 80% and 200-500 nm, respectively. The results of energy dispersive spectroscopy and infrared spectroscopy measurements demonstrate that the film is mostly composed of silicon, without impurities such as Cl, N, C and bonded H. It is suggested that the crystallization at such a low temperature originates from the effects of chlorine, i.e., in-situ chemical etching, in-situ chemical cleaning, and the detachment of bonded H.

The conservation of the hydrophobic and the hydrophilic residue sites obtained from 1000 designed sequences with the Z-score method for a four-helix bundle has been studied. The folding dynamic and thermodynamic features of the designed sequences and their different mutations are also studied. It is found that this conservation is related to the stability and the fast folding of the model proteins. Our results are consistent with the experimental results.

A full relativistic detailed-level-accounting approach has been developed independently to deal with the detailed spectral line effects on the opacity of the third most abundant element in stars: oxygen. The atomic energy levels and the oscillator strengths of the radiative transitions between the energy levels are obtained by carrying out the full relativistic one-configuration Dirac-Fock calculations. The photoionization cross sections are obtained via an average atom scheme with a consideration for the splitting of the ionization threshold due to the ionization stages and the term-couplings. As an example, the spectra resolved opacities and the mean opacities of oxygen are calculated to show the importance of the detailed spectral line profiles with the density of the matter.

A wave-particle resonance absorption model in the two-ion plasma is suggested in explanation to the coronal ions preheating in ³He-rich solar particle events. It is found that ³He and Fe ions are preferably preheated by the ion-ion hybrid waves at their fundamental and second harmonic ion cyclotron frequencies, respectively.

The cycle of black hole (BH) spin proposed by Li and Paczynski (henceforth CYCLP) is compared with a more natural model (henceforth CYC03), in which energy and angular momentum are transferred from a rotating BH to a region of some widths by the closed magnetic field lines. It turns out that the efficiency of converting the accreted mass into the radiation energy in the CYC03 is less than that estimated in the CYCLP, while the BH mass and entropy in the CYC03 are greater than those in the CYCLP. It is shown that the features of the CYC03 are insensitive to the power law index indicating the variation of the magnetic field in the disc.

Recent observations favour an accelerating Universe dominated by the dark energy. We take the effective Yang-Mills condensate as the dark energy and couple it to a relativistic matter which is created by the decaying condensate. The dynamic evolution has asymptotic behaviour with finite constant energy densities, and the fractional densities Ω_Λ ～ 0.7 for dark energy and Ω_m ～ 0.3 for relativistic matter are achieved at proper values of the decay rate. The resulting expansion of the Universe is in the de Sitter acceleration.

Recently the distances to type Ia supernova (SN Ia) at z ～ 0.5 have been measured with the motivation of estimating cosmological parameters. However, different sleuthing techniques tend to give inconsistent measurements for SN Ia distances (～0.3 mag), which significantly affects the determination of cosmological parameters. A Bayesian ``hyper-parameter'' procedure is used to analyse jointly the current SN Ia data, which considers the relative weights of different datasets. For a flat Universe, the combining analysis yields Ω_M = 0.20±0.07.