According to Kirchhoffs approximation，the fourth order moment statistical characteristic of the pulsed wave scattering by random rough surfaces is investigated numerically. As a special application，the fourth order moment （FOM） is simplified to obtain mutual correlation functions of the the two positions and two frequencies of pulsed wave scattering by random rough surface. It is found that the distributions of FOM vary with the coherence bandwidth frequency difference and scattering angle for different incidence angles. The numerical results show that the mean of fluctuating heights and coherent length of random rough surfaces have also an important influence on FOM. Generally，the largest scattering value and smallest coherence bandwidth of the FOM occur at the specific directions. That is to say，the FOM will have a lower coherent component and its distributions of scattering angles will be broadened in relation to the rugged surface. Meanwhile，the FOM decreases rapidly when there is a bigger coherence bandwidth frequency difference.

Based on equivalent particulate model and fractal theory，using the Mie theory and discrete-dipole approximation，the scattering by aggregated soot particles of the 06328?μm incident wave is computed. Then based on radiation transportation theory，and numerically calculating the transfer-characteristics of electromagnetic wave propagating in soot by using Monte Carlo method，the author gets the relationships of the reflectivity and transmissivity of soot with the changing incident angles，soot thickness and density，and analyses these factors affecting the reflectivity and transmissivity. This provides an efficient approach to the research of transfer-characteristics of electromagnetic wave in poly-disperse highly dense media.

The magnetic field microgravity effect is the phenomenon that a secondary microgravity is produced when a magnetic field is introduced into an electricconducting liquid, leading to the decline or extinction of the natural convection driven by buoyancy. The effect level could be described by the formula g_{m}＝（β_{0}/β_{m}）（ν_{0} /ν_{m}）^{2}g_{0}, according to the change of Grashof number. If neglecting β changes, the formula is simplified to g_{m}＝（ν_{0}/ν_{m}）^{2}g_{0}, by which the effect levels corresponding to different horizontal magnetic field strengths were estimated on the basis of the measurement of the viscosities of silicon melt under different conditions of the magnetic field. The results showed that the estimated results agree with the experimental data.

A new model of double-axicon was proposed for the first time. Based on the traditional axicon，a gradient on the vertex of an axicon was designed. When a plane wave illuminates the gradient axicon，two Bessel beams are formed and a Bottle beam generated between the Bessel beams. Bottle beams could be used as optical pipes，optical tweezers，optical spanners, and so on.Because the dark focus is surrounded in all directions by regions of higher intensity，and it shows important application perspective in life science and nanotechnology. The formation mechanism of the single Bottle beam was analyzed by geometrical optics. The transverse and longitudinal intensity distribution were also calculated and simulated using diffraction and interference theory，respectively. The result shows its significance in providing a guidance for the design and application of the gradient axicon.

Principle of the measurement of the upper atmospheric wind field using a Michelson interferometer with wide field, chromatic compensation and thermal compensation is briefly described. The modulation depth of the selfdeveloped wind imaging interferometer is analyzed and calculated, and the expression of the modulation depth is derived. Computer simulation is adopted, and the dependence of the modulation depth on the angle is analyzed. The field of view of the wind imaging interferometer is 129°when modulation depth is greater than 075, which fully shows the advantages of wide field and high modulation depth. The dependence of the modulation depth on the refractive index of glass is analyzed, and the optimized material is given. The dependence of the modulation depth on temperature is analyzed, and the curve is given. The research has important theoretical significance and practical value both for the research and development of the Michelson Interferometer with wide field, chromatic compensation and thermal compensation and the detection of the upper atmospheric wind field.

Diffraction enhanced imaging （DEI） is a powerful phasesensitive technique that generates improved contrast of soft tissues compared to the conventional Xray imaging. In this paper, an information extraction method is investigated based on diffraction enhanced imaging, which simultaneously produces three parametric images, depicting separately an object's projected Xray absorption, refraction and ultrasmallangle Xray scattering （USAXS） properties. The information extraction is resolved using analytic reconstruction formulae, and the quality of the obtained parametric images depends on the input images acquired at three distinct analyser crystal orientations. The stability of the reconstruction procedure is discussed using the condition number of matrix, and the theoretical maxima of the refraction angle and USAXS distribution width are given. Finally, validation studies are conducted by use of guinea pig cochleae DEI image.

The geometric phase of two-level atom, under the non-Markovian effect of environment, is investigated from the microscopic Hamiltonian. The results show that the geometric phase of atom strongly coupled with cavity field is larger than that of atom weakly coupled with cavity field, and the difference between them becomes larger as the loss of environment increases. In the case of relatively small environment loss, the geometric phase of atom shows continuous or discontinuous evolution with time depending on different initial atom states, and the discontinuity range increases when the environment loss increases. In a word, the geometric phase exhibits complicated characteristics under non-Markovian environment.

The closed-form expression for the mean squared beam width of annular beams propagating through atmospheric turbulence is derived by using the integral transform technique. The influence of turbulence on the spreading of annular beams is studied. On the other hand，the range of turbulence-independent propagation is obtained by examining the mean squared beam width，which indicates under what circumstances annular beams will be less affected by turbulence. It is shown that annular beams with larger obscure ratio ε，larger beam order M（N），larger wave length λ，and smaller outer radius w_{0} are less sensitive to the turbulence than those with smaller ε，M（N），λ，and larger w_{0}. The main results obtained in this paper are explained physically.

Based on the principle of crossspectral density of partially coherent beam，we have obtained the analytical expression of longterm intensity distribution，longterm beam width and quality factor on the receiver for J_{0}correlated partially coherent beams propagating in the slant atmosphere，then analyzed the effects of zenith angle，propagation distance，coherence of source and turbulence outer scale on intensity distribution and beam width on the receiver. Results show that when the zenith angle and propagation distance are fixed，we can adjust the coherence of source to obtain the flattened optical intensity distribution or central maximum intensity value on the receiver. When the propagation is fixed，with the increase of zenith angle or the outer scale of turbulence，the intensity distribution varies from central dip to Gaussian distribution. Near the focal plane for short focal length，the centraldip form of intensity distribution becomes deep first，then shallow with the propagation distance increasing. The real position of focal spot for J_{0}correlated partially coherent beam moves toward the emitter with the increase of zenith angle and turbulence outer scale, or with the decrease of the source coherence.

Taking the HermitecoshGaussian（HChG） beam as a typical example, the beamwidth spreading and angular spread of HChG beams propagating through atmospheric turbulence are throughly studied. The relative beamwidth and relative angular spread instead of the beamwidth and angular spread have been used to study the sensitivity of a beam to the effect of turbulence. It is found that the smaller the refraction index structure constant C^{2}_{n} is, the less the beamwidth spreading and angular spread of HChG beams are. The angular spread of HChG beams with larger beam orders m, n, smaller parameter Ω_{0} and smaller waist width w_{0} is less affected by turbulence. This conclusion also holds true for the beamwidth spreading of HChG beams for the case of the propagation distance is sufficiently long. The variation of the relative beamwidth of HChG beams versus Ω_{0} and w_{0} is analyzed if the propagation distance being not so long. The results are illustrated by numerical examples and the validity of the results is explained physically. The spreading of HermiteGaussian, coshGaussian and Gaussian beams in atmospheric turbulence can be treated as special cases of HChG beams.

The M^{2} factor matrix for twodimensional HermiteGaussian field is proposed. By introducing coupled beams half width and coupled M^{2} factor, the M^{2} factor matrix of the mode after being rotated around the propagation axis by some angle has been derived theoretically in the same coordinate. The rules of change of parameters related to M^{2} factor matrix varying with the modes' rotation angle have been simulated numerically; and the track of M^{2} factor vectors versus the modes' rotation angle has been given. Calculation results are in good agreement with theoretically derived ones, which proves that by using M^{2} factor matrix, M^{2} factors of a HermiteGaussian mode after rotation by a certain angle can be related with those of the same mode before rotation. This method is universally applicable to the beam quality analysis for asymmetric higher Gaussian beams, and is very important in guiding the mesurement of beam quality in practice.

We propose and demonstrate a novel technique that uses a FabryPerot laser diode （FPLD） to perform simultaneous alloptical individual channel clock extraction and wavelength conversion. The nonlinear dynamical period1 oscillation can occur when a semiconductor laser is optically injected. By utilizing the crossgain modulation effect and the period1 oscillation harmonic frequencylocked in an optically injected semiconductor laser, we can extract the wavelength conversional individual channel optical clock from the opticaltimedivisionmultiplexing （OTDM） signal. In a FPLD, we numerically simulate the extraction of 20 GHz optical clock at 1550 nm from the 2×20 Gb/s OTDM signal at 1555 nm, and experimentally obtain the -105 dBc/Hz phase noise frequency division clock of 1236 GHz to 618 GHz with simultaneous wavelength conversion from 155024 nm to 154591 nm. Moreover, we investigate the influence of the injection optical power level, wavelength detuning, the FPLD's bias current level and the longitudinal mode selection on the optical clock extraction. This method is potential to technologically apply in the alloptical demultiplexing and wavelength routing in the OTDM mixed wavelengthdivisionmultiplexing （WDM） communication system.

A method for solving the inverse problem of optical parametric chirped pulse amplification, i.e., for a given shape of the output signal pulse, how to calculate the shape of the input signal pulse, is proposed. Based on numerical techniques such as split-step Fourier transform and fourth-order Runge-Kutta, the quantitative relation between the input signal intensity and the output signal intensity was obtained by numerical fitting. Taking the chirped Gaussian pulse and the shaping pulse with a certain shape as the required output signal pulse, the input signal pulse shape can be worked out by the inverse calculation. The results indicate that the inverse problem can be solved quickly and precisely with the method proposed in this paper, and this method is useful for pulse shaping.

The influence of grating period and noncollinear geometry on the broadly tunable bandwidth of optical parametric amplification are investigated theoretically and numerically for quasiphasematched crystal. The concept of maximum grating period is proposed to achieve the phase matching and groupvelocity matching simultaneously in a wide range. By employing the maximum grating period, geometry Ⅰ is recommended due to the much wider signal tunable range compared with geometry Ⅱ. An expression is proposed to calculate the maximum grating period for congruent periodically poled LiNbO_{3} with 800 nm pump wave. A feasible scheme is presented to determine the working temperature and noncollinear angle α, maximize the tunable range and simplify the tuning by rotating angle θ only. Finally, the parametric bandwidth is studied with consideration of the influences of the maximum grating period and noncollinear geometry.

The model used for analyzing the gain characteristics of Erdoped waveguide amplifier （EDWA） is based on the ratepropagation equations with intensity profiles of pump and signal light. The intensity profiles of pump and signal light for different channel opening width （DCOW） of Erdoped waveguide is discussed. And the numerical method （NM） considering the difference of intensity profiles between signal and pump light in channel waveguide with DCOW is used to solve the gain characteristics of signal light in EDWA. An overlap factor Γ_{sp} between light intensity profiles of signal and pump is introduced to modify the conventional simplified method （CSM） in EDWA, which is a revised version of the conventional simplified method （RCSM）. Results show that difference of 297 dB in gain for -10 dBm signal light at 1534 nm due to DCOW of waveguides is obtained for a 4 cm long waveguide amplifier pumped by 80 mw at 980 nm. Comparing with CSM, the gain characteristics of EDWA from RCSM is simpler and better approaches the result from NM. Therefore, the RCSM is valuable in the research of EDWA.

The eigenvalue equation associated with seven-core photonic crystal fiber was derived from coupled-mode equation. Mode properties were analyzed by its eigenvalues and eigenvector. Using the frequency domain finite difference method, the relationship between fiber structure and coupling coefficient was fully discussed through changing the wavelength, hole pitch and core distance. Mode shaping and higher-order supermode cut off was realized. The results were helpful to the fiber structure design and in-phase supermode selection.

Evanescent wave fiber-optic sensors （EWFS） with acicular encapsulation were fabricated using silicon photolithography technology and silica wet-etching technology. This type of EWFS is small in size, consumes little reagent, suffers less from deformation during the detection process and also protects the sensors from pollution. The results obtained from the EWFS with different sensing fiber lengths were studied. The effect of the fiber length on the sensors absorbance were analyzed theoretically and experimentally. It was shown that the properties of the EWFS could be improved greatly by choosing suitable length of the sensing fiber, and the contribution of unit length of sensing fiber to the absorbance becomes less as the sensors total length becomes longer.

Influence of cavity wall elasticity on resonant frequency of small underwater cylindrical Helmholtz resonator is studied theoretically and experimental. Based on the theory of electroacoustic analogy, the simplified low frequency lumpedparameter model of the Helmholtz resonator is constructed. A general, convenient for calculation expression of the resonant frequency is given by circuit analysis. The influence of the thickness and the material of the resonator on the resonant frequency is investigated by numerical method. And the approximate rigid conditions for small underwater cylindrical Helmholtz resonators of different sizes are given. Small cylindrical Helmholtz resonators of different wall thickness and material were tested in a standingwave tube filled with water. Experimental results well testified the theoreticl results and the approximate rigid condition. This paper is useful for the design and application of the underwater cylindrical Helmholtz resonators.

Nanofluids have enhanced heat transfer capacity compared to conventional fluids. In the present paper, the effects of the thermophysical properties and thermal dispersion on the convective heat transfer coefficients of nanofluids have been investigated. The analytical results show that the effective thermal conductivities of a nanofluid is enhanced with the addition of nanopaticles into the base fluid. The collisions between nanoparticles, fluid molecules, and the wall of flow-tube are intensified, leading to enhanced mixing agitation and turbulence. Improved thermophysical properties and thermal dispersion result in the convective heat transfer coefficient enhancement of nanofluids.

This paper compares and discusses the surface energy and its related factors between different kinds of micro-component materials commonly used in MEMS, and analyzes the nano-adhesion behaviors of their surfaces. The results by the Owens method show that the surface energies of the silicon-based materials are between 60—75 mJ/m^{2} and the presence of surface roughness would result in higher nominal surface energy when the contact angle is less than 90°. In contrast, the nominal surface energy would go much lower when the surface is rough with the coverage of a SAM film. There is a certain dependence of nano-adhesion on surface energy, and the effect of surface roughness on nano-adhesion can be neglected in the present study.

In this paper for the mechanisms of pool boiling heat transfer a mathematical model is derived based on statistical treatment. No additional new empirical constant is introduced. The proposed model for pool boiling heat flux is found to be a function of wall superheating, minimum and maximum active cavity sizes, the contact angle, and physical properties of fluid. It also turns out that the present model well explains the mechanism of how wettability affects the boiling heat flux. The model predictions are compared with the existing experimental data, and fair agreement is found for different contact angles.

By using a twodimensional numerical simulation of the fully hydrodynamic equations, the dynamics of the traveling wave convection in binary fluid mixtures with defects was studied. For the separation ratio Ψ=-06, the system changes from localized traveling wave to traveling wave （TW） convection with the defects with increasing reduced Rayleigh number r. Then, the defects disappear and it changes to TW without defects with further increasing r. For different separation ratio Ψ, the periodicity of appearance of defects increases with reducing separation ratio and increasing r. The increase rate for a small negative separation ratio is larger than that for separation ratios. The range of existence of defects Δr remarkably decreases with the absolute value of Ψ. At Ψ=-011, the defects are annihilated and the TW with defects is not found. The system turns to the TW without defects from the TW with defects where r is larger than the upper limit of Δr. However, different values of Ψ correspond to different TW patterns when r is less than the lower limit of Δr. Therefore, it is obvious that the Ψ has influence on the pattern formation and on the transition between the patterns, and the transition between the patterns is also complex.

CONDENSED MATTER: STRUCTURAL, MECHANICAL, AND THERMAL PROPERTIES

The center points of the thermomechanical fatigued single crystal Nibased superalloys were measured by neutron diffraction and the results have been treated in two methods: two overlapped peaks and a single peak. The macroscopic equivalent stress, deviatoric stress components, γ/γ′ phase equivalent stress and lattice mismatch are determined according to the experimental results. The stress results show that the macroscopic equivalent stresses are basically consistent when using two analysis methods, the maximum of macroscopic equivalent stress and deviatoric stress components appeared at thermomechanical fatigue that of 100 cycles, and at this point, the microdefects， such as dislocations, are saturated and the stress of γ′ phase away from the center was reduced remarkably. The calculated results also indicate that the negative lattice mismatch is reduce markedly at the beginning of the thermomechanical fatigue. Subsequently, it increases linearly with the fatigue cycles at the speed 81×10^{-6} per cycle because of the accumulated plastic strain in the matrix.

According to the classical crystallization theory, the influences of kinetics and thermodynamics on glassforming ability （GFA） for bulk metallic glasses has been investigated. The theoretical analysis shows that crystallization resistance is in proportion to the viscosity of “nose” temperature （T_{n}） while crystallization driving force is inversely proportional to the viscosity of crystallization onset temperature （T_{x}） on reheating in timetemperaturetransformation （TTT） curve, and therefore a GFA parameter ω_{0}, defined as （T_{g}-T_{0}）/（T_{x}-T_{0}）-（T_{g}-T_{0}）/（T_{n}-T_{0}）, was proposed （wherein T_{g} and T_{0} are glass transition temperature and Vogel temperature respectively）. The parameter ω_{0} shows an excellent correlation with the critical cooling rate for glass formation of bulk metallic glasses, with the statistical correlation factor of R^{2}=09626. Furthermore, the relationships between the GFA and T_{rg}, ΔT_{x}, γ, γ_{m}, ΔT_{rg}, α, β, δ and φ, as well as the fragility of liquid could be explained reasonably based on our proposed ω_{0} parameter.

Hydrogenated silicon films were prepared by conventional radio frequency plasma-enhanced chemical vapor phase deposition technique at a high deposition rate at temperatures from 100 to 350℃, which were studied by Fourier transform infrared spectrum and Raman scattering spectrum. The results showed that the hydrogen content and the silicon-hydrogen bonding configurations of the films were closely related to their crystallization properties. When the films changed from amorphous to nanocrystalline phase, the hydrogen content decreased by over a half, and the Si—H bonding configuration was mainly SiH_{2}. With the increase of substrate temperature and crystallinity, the hydrogen content and the structural factor of the nanocrystalline silicon films was reduced gradually.

Using Sn powders as source material, different SnO_{2} nanostructures, such as nanowires and nanorods, nanoflowers were synthesized successfully by simple chemical vapor deposition in a low temperature. SnO_{2} submicron rings were obtained by reducing the oxygen gas concentration in carrier gas and the effect of oxygen gas on controlling the morphology of products was demonstrated by adjusting the quantity of Sn powders and the rate of heating. The products were characterized by X-ray diffraction, scanning electron microscopy and energy-dispersive X-ray spectrometry. The growth mechanism of the products was further discussed.

The molecular dynamics method is used in this paper to investigate the effect of vacancy defects on the elastic properties of armchair and zigzag singlewalled carbon nanotubes （SWCNTs）.The results show that the Young's moduli of armchair （5,5）, （10,10） and zigzag （9,0）, （18,0） singlewalled carbon nanotubes are 948, 901 GPa, and 804, 860 GPa,respectively. The armchair and the zigzag SWCNTs Young's moduli decrease and increase with the increase of the nanotubes' diameter, respectively. With the increase of the vacancy defect ratio, the Young's moduli will decrease. When the vacancy defect achieves a certain ratio, there appears a sudden slowdown in the curves of Young's modulus vs. vacancy defect ratio and a platform emerges. The influence of the bivacancy defects on the Young's moduli of the carbon nanotubes depends on the orientation of the defects. As the number of atoms between the two singlevacancy defects increases, when the defects are located in the direction of the axis, the Young's modulus decreases to certain value and fluctuates around it; but when the defects are located in the direction of the circumference, the Young's modulus decreases first, then ascends and finally tends to a constant, and the Young's moduli will decrease a little with increasing of the distance between the two singlevacancies. The reasons are analyzed by considering the features of the bond σ and bond π between the molecules and the theory of the shortrange coupling of the electron clouds between two defects as well as the theory of 51DB defect formation in the vacancy defects.

Using molecular dynamics method with many-body potentials, we have studied the thermal stability and melting mechanism of platinum nanocrystal of truncated octahedron enclosed by {111} and {100} surfaces, and investigated its microstructure and shape by statistical radius and Lindemann index. The results show that the nanocrystal first shows shape changes at 1500 K, then transforms to a spherical one at 1700 K. The surface first shows the premelting behavior at 1500 K, and the melt completely into liquid state. The melting process starts from the surface into the interior at this temperature, resulting in the complete melting of the nanocrystal at 1730 K. The surface premelting is advantageous to shape transformation of octahedron-truncated nanocrystal.

The structure study for large Si_{80} cages using ab initio calculations based on density functional theory is reported. The results show that Si cages are distorted from perfect I_{h} symmetry to T_{h} symmetry. The geometry and electronic properties of Si_{80} cages are discussed.

Silicon carbide （SiC） nanorods were synthesized via catalyst-assistant crystallization of amorphous silicon carbonitride. An intensive sharp photoluminescence peak at 378 nm was observed from the SiC nanorods. SEM, TEM, HRTEM and XRD was used to characterize the structure. XRD pattern showed that the nanorods are pure 3C-SiC containing stacking faults. The HRTEM image showed that the stacking faults were threefold stacking faults, which resembled the structure of 6H-SiC. We attribute the emission to these stacking faults, and discussed the emission mechanism.

Nanopores with diameters between 30 nm and 180 nm have been fabricated by inducing latent track with fast heavy ions and etching process in 25 μm thick, singlecrystal muscovite mica. For short etching time, the nanopores are columns with circular cross section. For long etching time the cross section of nanopores becomes rhombic. Thus the shape of nanopores depends on the etching time. Cu nanowires have been fabricated with controlled dimensions by electrodeposition into the nanopores. The ultravioletvisible light absorption spectra of Cu nanowires embedded in mica templates show that the circular Cu nanowires with diameter smaller than 60 nm exhibit one intense resonance peak and one smaller peak. With increasing diameter of the nanowires， the intense peak is redshifted while the smaller peak strengthens gradually. The diameter and shape can tune the optical properties of Cu nanowires. The morphology and crystallinity of the Cu nanowires were studied by means of scanning electron microscopy and Xray diffraction.

Temperature is an important factor for the behavior of He in metals. In this paper, we write a LKMC （lattice kinetic Monte Carlo） program to simulate the behavior of He in bbc Fe under the temperature range of 298—1298 K. The simulation results show that the temperature influence on He behavior can be divided into four different stages: 1）298—598 K；2）598—798 K；3）798—998 K；4）998—1298 K. In the first stage: with the increase of temperature, the concentration of He in the grain has a little decrease, but the average number of He in He bubbles increases greatly; In the second stage, with the increase of temperature, the concentration of He in the grain decrease rapidly, but the average size of He bubbles almost keeps constant. In the third stage, with the increase of temperature, the concentration of He in grains and the average numbers of He in He bubbles both decrease rapidly. In the third stage: with the increase of temperature, the average numbers of He atoms in the grain and the average size of He bubbles both increase. At 1298 K, the average number of He atoms is almost equal to that at room temperature, and almost no He escapes from Fe grains.

In this paper, seven constitutive models for oxygenfree high conductivity copper at high pressure and high strain rate are constructed based on Y/G=constant and on G/B=constant, respectively （Y denoting the yield strength, G the shear modulus, B the bulk modulus）. The variations of longitudinal stress, transverse stress and yield strength of oxygenfree high conductivity copper with time under planar shock loading are obtained using the manganin stress gauge and compared with the predicted values by the constructed seven constitutive models. It is indicated that the pressure, density, temperature and plastic strain dependence of the yield strength for oxygenfree high conductivity copper under planar shock loading is essential to the constitutive description. It seems that the strength models of oxygenfree high conductivity copper obtained from SHPB tests and torsion tests are unavailing to the planar shock loading.

A novel photothermal microactuator based on microscopic photothermal expansion is developed. A theoretical model of photothermal expansion for photothermal microactuator is established. Using finite element analysis and numerical method, the expressions of the temperature distribution and the linear thermal expansion are deduced. A photothermal microactuator of 1200?μm length is fabricated by using an excimer laser micromachining system. According to the parameters of the laser and the photothermal microactuator, a theoretical relationship between the expansion and the laser power is calculated. The results indicate that the photothermal microactuator deflection （corresponding to expansion） is approximately linear with the laser power irradiating the expansion arm, according with theoretical predictions quite well. As an application, a Biswitch like photothermal microactuator is built and the successful photothermal microactuation experiment shows the Biswitch function can be realized.

A stressinduced voiding model based on the NabarroHerring mechanism has been proposed. The stressinduced voiding phenomena in Cu interconnects have been studied by the FIB crosssection technique and stress modeling. The driving force for the formation of stressinduced voids has been investigated. The relationship between stressinduced voiding, temperature, stress gradient and the dominant diffusion path are discussed. The results show that stress and stress gradient reach their peak values at the top surfaces of Cu M1 lines underneath the corner of the vials where voids are observed. Stress gradient shows crucial effect on the failure spot and the voiding rate. Stress migration is basically a diffusion and nucleation process of vacancies through the main diffusion path under the force of the stress gradient. The stress gradient and the diffusion terms vary oppositely with temperature and the maximum voiding rate is reached at a medium temperature.

Based on the nanometer CMOS technology, a novel parallel RLC coupling interconnect analytic model is presented. Based on the function approach and reduced order techniques, an analyzable expression for the outlying terminal of the disturbed line is derived by the model in the off-angle step input signal. In the 90 nm and 65 nm CMOS process, the proposed RLC coupling interconnect analytic model enables to estimate the crosstalk voltage within 4% errors compared with Hspice simulation for various interconnect coupling size. The proposed analytic model can be applied to the design of nanometer SOC and optimizing the design for VLSIs.

The diffusion processes of single adatoms Ag,Pd and Cu on Cu（001） surface are investigated by molecular dynamics. By observing the mechanisms of selfdiffusion and heterogeneous diffusion, the diffusion frequencies of three kinds of adatoms under different temperatures are counted and diffusion barriers and the preexponential factors of diffusion frequency are fitted. The results of diffusion barriers are compared with those from the static calculation. The results show that: Under 800 K, the dominant diffusion mechanism of adatom Ag,Pd and Cu is the simple hopping mechanism. The hopping frequency of adatom Ag is the biggest, and that of adatom Pd is the smallest. For homogeneous and heterogeneous adatom diffusion processes, the relationship between diffusion frequency and temperature obeys the Arrhenius formula. The diffusion barriers fitted from Arrhenius formula are relevant to the surface structure and the cohesive energy of adatom on the surface. The transition temperatures of diffusion from the jumpmechanismdominant to the exchangemechanismdominant are about 825 K and 937 K for Pd and Cu adatoms respectively.

The ntype GaN films have been grown on cplane sapphire with different small misorientation（0°—03°）by metalorganic chemical vapor deposition. It was observed by atomic force microscopy that the ntype GaN has the step flow growth mode, the flow steps of the ntype GaN surface are uniformly distribution on 02° and 03° misorientation sapphire substrate, it was observed clearly that random and poor distribution of the flow steps was caused by the step reconstruction on 0° misorientation sapphire substrate. The image quality parameter of electron backscatter diffraction indicated that the strains increase as the ntype GaN epilayer thickness increases on 0° misorientation sapphire substrate but do not vary obviously on 02° and 03° misorientation sapphire substrates. Electrical and optical properties demonstrated the ntype GaN grown on the 02° and 03° misorientation sapphire substrates have higher electron concentration and lower ratio of the intensity of yellow band to near band edge.

ZnO thin films and ZnOSiO_{2} composite films were deposited on glass substrates by solgel method. Atomic force microscope images showed that ZnO nanoparticles havd spherical shape of relativly small size. The ultravioletvisible transmittance spectra showed that the transmittance of ZnOSiO_{2} composite film is about 85% in visible range and the transmittance begins to drop at about 330 nm and reaches zero around 290 nm. Due to quantum effect, the absorption edge has an obvious blue shift. Photoluminescence spectra measurement indicated that ZnOSiO_{2} composite film excited by a Xe lamp with different wavelength has a UV emission around 290 nm coming from combination of free excitons, which is in agreement with the ultravioletvisible transmittance measurement result. Meanwhile, two photon and three photon absorption and upconversion luminescence were observed in the ZnOSiO_{2} composite film.

A series of AlN/Si_{3}N_{4} multilayers with different Si_{3}N_{4} thickness were synthesized by reactive magnetic sputtering. The microsructure of the multilayers was characterized with Xray diffraction and highresolution transmission electron microscopy, and nanoindentation was employed to measure their mechanical properties. The crystallization behavior of Si_{3}N_{4} modulation layer in the multilayers and its influences on the microstructure and mechanical properties of AlN/Si_{3}N_{4} multilayers were studied. The results show that when Si_{3}N_{4} thickness is less than about 1 nm, Si_{3}N_{4}, normally amorphous in deposition state, could form a wurtzitetype pseduocrystal structure, same as the structure of hAlN, due to the template effect of AlN crystal layer. Crystallized Si_{3}N_{4} layers and AlN template layers grow coherently into columnar crystals. Correspondingly, the hardness of the films is enhanced, showing a superhardness effect. Further increasing the thickness of Si_{3}N_{4} layers, the coherent interfaces of the multilayers are damaged and Si_{3}N_{4} layers become amorphous, accompanied by the decline in hardness of the films. The discussion indicates that the superhardness effect in AlN/Si_{3}N_{4} nanomultilayers is related to the enhancement of modulus difference between the two different module layers caused by the alternating stress field in the coherent growth structure.

CONDENSED MATTER: ELECTRONIC STRUCTURE, ELECTRICAL, MAGNETIC, AND OPTICAL PROPERTIES

The elastic constants, electronic structure and optical properties of solid krypton are investigated by first-principles plane-wave pseudo-potential method in the scheme of density functional theory within local density approximation and generalized gradient approximation. The calculated elastic constants under pressure are in good agreement with the experimental and theoretical data. The obtained Debye temperature and sound velocity increase with pressure. The electronic structure and optical properties under pressure are also predicted, the energy band gap decreases, optical absorption coefficient increases and absorption peak widens under pressure. This leads to easy electron transition and solid krypton probably becomes semiconductor.

Molecular dynamics simulations are used to investigate the surface melting processes of three low-index surfaces of aluminum. We find that the surface melting is anisotropic for different surfaces by analyzing the variation of the microstructures during melting. Although （001） and （111） surfaces can both sustain overheating, the microstructures of them are different. There is a thin disordered quasi-liquid film on （001） surface when overheated while the （111） surface is crystalline. It is surprising that the melted （001） surface can be overheated. The velocity of melting is calculated for all the surfaces at different temperatures, and the melting point is extrapolated to be 950 K±9 K which agrees with the experimental data of 9335 K.

The generalized gradient approximation （GGA） based on density functional theory （DFT） is used to analyze the evolution of the structure and electronic properties of the fullerene derivatives C_{60}（CF_{3}）_{n} （n=2, 4, 6, 10）. It is found that among the three respective possible stable isomers of C_{60}（CF_{3}）_{4} and C_{60}（CF_{3}）_{6}, the structure C_{60}（CF_{3}）_{4} with ppp addition and C_{60}（CF_{3}）_{6} with pppmp addition are the most stable structures, respectively. By analyzing the structures of C_{60}（CF_{3}）_{2}, C_{60}（CF_{3}）_{4}1，C_{60}（CF_{3}）_{6}1, and C_{60}（CF_{3}）_{10}, it was found that both the average C—C bond length and the C_{C60}—C_{CF3} bond length increase with the increase of the CF_{3} number. It was found from the electronic properties of four compounds that the heat of reaction of the compounds almost linearly increases with the number of the CF_{3}, and its maximum appears at n=6, thus, C_{60}（CF_{3}）_{6} should be the most easily synthesized derivative. It is known from the Mulliken charge that the interaction between CF_{3} and the cage and the electron transference from CF_{3} to the cage increase with the number of the CF_{3} However, the net spins of the compounds are all zero, indicative of their closedshell electronic structures. Finally, the frontier orbital analysis shows that the electron detaching and attaching both occur at carbon sites of the cage.

Applying the method of coherent state orthogonal expansion, an analytical expression of the ground state energy for one-dimensional Holstein model is obtained. We expand the trial wave function of ground state to third order, and calculate the ground state energy for different lattice sites and different coupling intensities. In the third order approximation, the ground state energy agrees with the numerical solution.

The first principles calculation of plane wave ultra-soft pseu-dopotential method based upon the denstiy functional theory and generalized gradient approximation have been adopted to investigate the energy band structure and electronic density of states and Mulilken charge population of Al and N codoped Zn_{1-x}Mg_{x}O. The calculation indicates that the Al and N codoped Zn_{1-x}Mg_{x}O has the trend of forming p-type Zn_{1-x}Mg_{x}O.

By using the freespace terahertz （THz） electrooptic （EO） sampling technique, the THz electromagnetic wave waveforms emitted from intrinsic bulk GaAs photoexcited by femtosecond laser pulses under strong bias electric fields up to 300 kV/cm were recorded. From the experimental data, we can clearly see the THz electromagnetic wave emission waveforms, E_{THz}（t）, which are proportional to the acceleration/deceleration of electrons, have a bipolar feature. Power dissipation spectra of electrons for stepfunctionlike input electric fields have been obtained by calculating Fourier spectra of the measured THz traces. The cutoff frequency, ν_{c}, for negative power dissipation （i.e., gain） due to intervalley transfer is found to gradually increase with increasing bias electric fields, F_{0}, for F_{0} < 50 kV/cm and saturate at 750 GHz above ～50 kV/cm at 10 K.

Doublesubstituted Ba_{x}Ag_{y}Ca_{3-x-y}Co_{4}O_{9} bulks were fabricated by nitric acid solgel and spark plasma sintering, and the phase composition, orientation, texture and electrical transport properties of the bulks were investigated by means of X-ray diffraction,scanning electron microscope and electrical resisitivity measurement. The result shows that silver which does not belong to the stoichiometric compounds Ba_{x}Ag_{y}Ca_{28}Co_{4}O_{9} is distributed in the matrix. The degree of orientation increases with increasing ratio of Ba to Ag for the doublesubstituted bulk. The orientation of Ba_{x}Ag_{y}Ca_{28}Co_{4}O_{9} （y>0） bulks is prominently lower than that of Ca_{3}Co_{4}O_{9} and the orientation of Ba_{02}Ca_{28}Co_{4}O_{9} bulk is higher than that of Ca_{3}Co_{4}O_{9} bulk. The electrical transport mechanism of the bulk with x=y=01 is changed slightly. The Ba_{01}Ag_{01}Ca_{28}Co_{4}O_{9} bulk exhibits simultaneously lowered orientation and moderately low resistivity and the resistivity reaches 73 mΩ·cm at 973 K. The Ag_{02}Ca_{28}Co_{4}O_{9} bulk exhibits lowest orientation and lowest resistivity, reaching 63 mΩ·cm at 973 K.

In-doped ZnO nanoarrays were successfully synthesized by the carbon thermal reduction deposition process, at 850℃ using Au catalyst. The length of legs of T-ZnO nanorods is 2—3 μm and the diameter is ～400 nm, dimensionally uniform and the surface smooth. Energy dispersive spectroscopy and X-ray photoelectron spectroscopy investigations show that the In content of nanorods reaches about 08%. Room temperature photoluminescence spectra of the nanoarrays shows that the ultraviolet emission peak red-shifts and becomes broader after doping. Raman scattering study shows that ZnO peaks have different shift, and a new peak appears, which proved a part of In atoms replacing Zn in the crystal lattice.

Zn_{095}Co_{005}O films were prepared under different oxygen partial pressure P by magnetron sputtering. The effect of P on the magnetic and electrical properties was investigated. The effect of oxygen vacancy on the magnetic properties was also calculated by first-principles calculation. The experimental results indicated that Zn_{095}Co_{005}O films showed room-temperature ferromagnetism and high electron concentration when they were deposited under high vacuum. The ferromagnetism disappeared and the electron concentration decreased sharply when P was increased. The calculated results indicated that the energy of ferromagnetic states could be decreased by introducing oxygen vacancy in Co-doped ZnO system. The stability of ferromagnetism was determined by the distance between oxygen vacancy and Co atoms.

The quantum transportation of electron through equilateral polygonal quantum rings with Rashba spin-orbit interaction is studied. By using the typical method of quantum network and the Landauer-Büttiker formalism, we solve analytically the scattering problem of electron through any equilateral polygonal quantum ring, and obtain the relevant formula for spin transportation conductance. The characters of conductance varying with wave-vector of electron and the strength of spin-orbit interaction are investigated, and the series of zero conductance points originating from spin-orbit interaction is determined. In the limit of infinite number of borders of equilateral polygon, we prove that the formula is consistent with the results obtained directly from the circular model of quantum rings.

Using the slave-boson mean-field approximation, we theoretically study the properties of the ground state of T-shaped-coupled quantum-dot Aharonov-Bohm （A-B） ring in the Kondo regime by means of one impurity Anderson Hamiltonian. It is found that in this system, the complex physical properties are determined by the interplay of the Kondo-Fano effect. When the free electron spin of quantum dot embedded in A-B ring is screened, with the increase of coupling between the dots, a giant increasing sharp persistent current （PC） peak appears, which implies that a giant Fano interference exists in this system. But, in this case, the Kondo PC peak is not affected. Thus this model may be a new candidate Kondo screening cloud.

Taking into account the gradient decay of the real disorder, we explore the effects of the disorder gradient and the structure sizes on persistent currents in the disorderorder twodimensional mesoscopic ring system, within the tightbinding model. In the absence of disorder gradient, the persistent current firstly decreases and then increases with disorder strength, indicating a localizedballisticlike transition existing there. In the presence of disorder gradient, the persistent current is decreased, due to the stronger disorder scattering in the surface disorder region of lower disorder strength. In the case of exponentially decaying disorder, especially, the localizedballisticlike transition disappeared, meaning that its observability depends on the distribution of the gradient disorder. In addition, the effects of both the ring width and the penetration depth of the disorder on persistent currents are explored, showing a strange quantum sizeeffect, which is different from previous results.

Silver nano-films were prepared by thermal evaporation. Scanning electron microscope observations showed that the silver films have nano-structures in morphology. The optical properties of silver nano-structures were obtained by ultraviolet-visible spectrometry, which shows the resonance wavelength of surface plasmon. Precise control of thermal deposition parameters allows the fabrication of silver films on glass substrates with tunable surface plasmon resonance wavelength, which makes the films to have tunable optical properties of transmittance and reflection. In this work, we have prepared silver nano-films with over 90% transmittance in the near-IR region and near 50% reflection in the visible region near the surface plasmon resonance wavelength. Due to the novel optical properties, silver nano-films have an enormous light management potential for use as an interlayer in thin-film solar cells.

Ni/4H-SiC Schottky barrier diodes （SBDs） were fabricated, and irradiated with the ^{60}Co gamma-ray source to the accumulated dose of 1 Mrad（Si）. The 0 V and -30 V bias voltage were applied to the SBDs during irradiation. After 1Mrad（Si） radiation, the Schottky barrier height and ideality factor of the Ni/4H-SiC SBDs under different bias voltages basically remain at the same values, and the minority carrier lifetime of the epitaxial layer also has no degradation. The reverse current decreases after radiation, which can be explained by the negative surface charge increase. The results show radiation bias voltage has no obvious influence on the radiation effect of the Ni/4H-SiC SBD.

The effect of thickness on transformation temperature of the NiTi thin films has been studied by X-ray diffraction and X-ray photoelectron spectroscopy. Results show that the crystallization temperature for 3?μm-thick film is higher than that for 18?μm thick film at the same growth temperature and post annealing. With the substrate temperature increasing, the start temperature （A_{s}） of austenite phase is lowered after annealing at 763 K for 1?h. There is an oxide layer （TiO_{2}） on the film surface, which prevents the Ni atom from coming onto the surface. There is an oxide layer of a mixture Ti_{2}O_{3} with NiO on the film /substrate interface. The oxide layers affect the transformation temperature by changing the Ni atomic content in the interior of the film.

The effects of thermomagnetic pretreatments on the magnetism and mechanical properties of Ni_{50.3}Mn_{28.7}Ga_{21} single crystal were studied. Two samples A and B were heated to a temperature above their Curie temperature （T_{C}） and then cooled down to room temperature via different cooling processes. An external magnetic field was applied during cooling of sample A from T>T_{C} to room temperature. For sample B, the same magnetic field was applied at T below its T_{C} but higher than its austenite to martensitic transformation temperature during cooling. The reversible strain and change in magnetization （ΔM） measured at room temperature for sample A are much smaller than those for sample B. It was supposed that with the appearance of the magnetic field during the transformation from austenite to martensitic phase, the field tends to induce a favorable single martensitic domain in sample B, leading to the greater reversible strain and ΔM. However, preferential magnetic domains may develop in sample A by applying the same magnetic field during the transition from paramagnetism to ferromagnetism, leading to the different alignments of martensitic domains from those in sample B. This might be the reason for the smaller reversible strain and ΔM in sample A.

The structural stability and magnetism of Ag atomic chains are studied by the firstprinciples calculations. The chain structures, such as the linear, zigzag, ladder and Tstructure which are formed by three linear chains, are studied. The results show that the linear structure is the most unstable geometry with the lowest cohesive energy. The Tstructure is the most stable geometry, showing the largest cohesive energy. The bond lengths in all the atomic chains studied are shorter than that in the bulk, indicating a stronger bonding strength in the atomic chains as compared with the bulk material. Besides, the study on the magnetism of Ag linear chain reveals that the linear structure could exhibit magnetism when the bond length is compressed by as large as 52%, although the equilibrium linear chain is nonmagnetic. By employing the Stoner criteria and the interaction picture of atomic orbitals, the reason for the possible magnetism of Ag linear chain are explained.

The ferromagnetic domainwall model is used to investigate the influence of the cooling field （including its magnitude and orientation） on the exchange bias（h_{E}） in ferromagnet/antiferromagnet bilayers with uncompensated interface. The results show that h_{E} is independent of the magnitude of the cooling field when it is applied parallel to the easy axis of the antiferromagnet. When the cooling field deviates from the easy axis direction, h_{E} changes slowly with increasing field angle, but a sudden change is observed when the angle goes up to a critical angle γ_{c}, which γ_{c} increases with the cooling field increaseing. Especially, h_{E} has a transition from negative to positive when the field angle is beyond γ_{c}, and the transition point also relates to the magnitude of the cooling field. Moreover, the thickness dependence of the exchange bias exhibits a liner dependence on the inverse ferromagnet layer thickness, but yields a crossover from h_{E}∝N^{-1}_{F} to h_{E}∝N^{-λ}_{F} at a certain thickness （here λ>1）. This behavior is strongly dependent on the ferromagnet thickness, the magnitude and the angle of the cooling field.

The new-domain nucleation by applying a weak electric field is an important process for the ferroelectric domain inversion in ferroelectrics. We studied the process of the new-domain nucleation in ferroelectrics using the classical model, and found that the rate of nucleation in the electric field did not agree with the experimental result. Based on Tagantsevs model, we improved the classical model of the new-domain nucleation using a long flat dagger model instead of the ellipse model. Using the improved model, the result of the theoretical calculation agreed well with the experimental measurement.

Rare earth doped yttrium oxide （RE^{3+}:Y_{2}O_{3}） nano powder was fabricated by wet chemical method. The influence of doping concentration and different doping elements on the microstructure, phase composition and spectral properties was analyzed. It was shown that by changing the doping element and doping concentration, the spectral properties, including spectral coefficients and upconversion spectrum, can be controlled, and the material can be made to have better absorption properties for 106 μm laser light.

The variable relation of the coefficient variance of wavelet transformation of 1/f fractal signal in white noise versus the scale is adequnately modified to develop a novel method based on least\|squares to estimate the parameters of 1/f noise of semiconductor laser diodes （LDs）. The measured data indicate that this method can effectively extract the 1/f noise submerged in white noise of LDs, and the estimated 1/f signal is in better accordance with the measured results of the contrast apparatus.

The diammonium phosphate was added as start material to synthesize Pdoped BaMgAl_{10}O_{17}:Mn^{2+} phosphors, which were prepared by high temperature solid state technology. The effect of Pdoping on the structure, morphology and luminescent properties of BaMgAl_{10}O_{17}:Mn^{2+} were determined by Xray diffraction, scanning electron microscopy, Fouriertransform infrared spectroscopy, energy dispersive Xray spectroscopy and vacuum ultraviolet （VUV） spectra. The results show that P doping is conducive to the crystallization of the phosphors and improving the morphology. The lattice parameters become smaller when the phosphor is doped with P. The P doping makes the emission spectra to blueshift and the CIE x value to decrease. At appropriate P concentration, the absorption of the matrix in VUV range is improved, and the emission intensity under VUV excitation is enhanced.

An organic vertical cavity surface emitting laser has been fabricated, using a Forster energy transfer system of organic pyridium salt dye ASPI and Alq_{3} doped in PMMA film as the gain medium, which was sandwiched between a high-reflectance distributed Bragg reflector and a silver reflector. The stimulated emission properties were investigated by nanosecond-pulsed laser pumping. The lasing phenomenon was observed with a full width at half maximum of 243 nm at the lasing wavelength of 600 nm, and the threshold energy for lasing was estimated to be about 8?μJ per pulse.

Zinc oxide （ZnO） thin films have been grown on n type Si （100） substrate using femtosecond pulsed laser deposition （PLD）. The effect of change in parameters, including substrate temperature, laser energy and oxygen pressure, on the structure and optical properties of ZnO films is discussed. The Xray diffraction results show that the ZnO films are highly caxis oriented when deposited at substrate temperature of 80?℃ with laser energy of 1\^5?mJ under oxygen pressure of 10 mPa. The field emission scanning electron microscopy indicates that the mean grain size increases with the increase of temperature, but decreases with the increase of laser energy. The ultravioletvisible transmissivity shows that the annealed films have a transmittance of 90% in visible rang. The photoluminescence spectra of ZnO films are discussed. In comparison, the structure and photoluminesceme properties of ZnO films produced by using nanosecond PLD are also studied.

In this paper, ZnO thin film was first grown on the sapphire fiber-ending by electron-beam evaporation, which has good morphology and high crystallization quality. The transmittance spectra showed that the sharp optical absorption edge has a red shift with increasing temperature, and the obtained temperature-dependent band gap energy for ZnO follows to the linear relation E_{g}（T）=340-491×１０^{-4}T. This formula presents the key to fabricate new broad\|measurement range fiber-optic temperature sensors based on semiconductor ZnO by utilizing the optical absorption spectra at different temperatures.

INTERDISCIPLINARY PHYSICS AND RELATED AREAS OF SCIENCE AND TECHNOLOGY

By using the KKSO multiphase field model, the fixedvelocity growth and velocitychanging growth of directional solidification for eutectic CBr_{4}C_{2}Cl_{6} alloys are simulated in three dimensions to reproduce the morphology evolution and selection. The diagram of the eutectic morphology selection is further established. The changes of the average interface velocity and the average interface undercooling during velocitychanging growth are also studied. The simulation results show that the eutectic morphology selections both before and after velocity change are the same as the morphology selection in the case of fixed pulling velocity. For all of the eutectic morphology selections, changes of the average interface velocity and the average interface undercooling show the hysteresis effect. It's also found that the relationship between the average interface velocity and the average interface undercooling is in good agreement with the theoretical prediction.

Intracellular processes of live organisms, which can be described by sets of biochemical reactions, are inherently stochastic, where the fluctuations in molecule abundance inside the cell play a crucial role in the cellular growth and development. For typical building blocks of biochemical reaction networks （including signal transduction networks and metabolic networks）, this paper first gives a unified formulation and then presents a general mechanism of noise propagation by applying the linear noise approximation theory. The main results show that: there is noise propagation in signaling transduction pathways and the noise intensity satisfies noise sum rule, and there is no noise propagation in metabolic pathways and the distributions of the steady state molecules mutually independent. The analytical results lay a theoretical foundation for understanding the intracellular processes.

With the combination of optimal control theory and multi\|configuration time-dependent Hartree（MCTDH） method, the vibrational quantum model for describing the multidimensional planar structural PTCDA molecule is constructed. Within the frame of MCTDH, the dynamics of PTCDA ground state via pump-dump process under an excitation of femtosecond laser pulse is considered. The relation of reduced control yield of target state to the shape of excitation pulses, the propagation time and the effective energy of optimal field are analyzed theoretically. The wavepacket distributions in the involved vibrational coordinates are compared with the vibrational parameters. It is found that with longer propagation time, higher control yield of target state at lower strength of optimal pulse is realized, which provides an effective method to realize the control with low energy laser pulses in experiment.

Ordered colloidal particle films with double layers and dual members could be prepared by stepbystep spincoating method. Monolayer film of larger sized PS or SiO_{2} colloids was utilized as template to selfassemble second layer of SiO_{2} colloidal particles with smaller diameter. The size ratio γ of smaller and larger colloidal particles was 0.20—0.56 in the composite films. Larger （L） and smaller （S） dual member colloidal particles formed composite crystalline doublelayer films with multifarious arraymodes, whose structures could be expressed as LS_{x} （here, L represents larger sized particles, S represents smaller colloidal particles, and x represents the stoichiometry of larger and smaller colloidal particles）, and x=1, 2, …, 13. The performance of composite colloidal films were influenced mainly by several factors, including spincoating velocity, spincoating time, viscosity of suspension medium, number density of colloidal suspension, and the wetting properties of substrates, etc.. On the prerequisite of complete wetting between colloidal suspension and substrate, appropriate number density of colloidal suspension, velocity and time of spincoating were the necessary conditions for ordered selfassembling to crystalline film of colloidal particles by spincoating method.

Indium tin oxide （ITO） compound is widely used as the front contact of silicon solar cells. Its work function Φ_{ITO} is one of the most important factors related to the performance of solar cells. In this paper, we use the AMPS-1D （Analysis of Microelectronic and Photonic Structures） program developed by Pennsylvania State University to analyze the dependence of physical parameters of solar cells on the Φ_{ITO}. Our results show that Φ_{ITO} has greater effect on amorphous silicon solar cells than on the microcrystalline ones. Physical performance of amorphous silicon （the efficiency, filling factor etc.） improves with increasing Φ_{ITO}, which is more obvious than that for microcrystalline silicon. These results provide a useful guide in selecting appropriate p-type silicon materials based on the performance of front contacts.

A function transformation method for constructing exact solutions of the variable coefficient nonlinear evolution equations is proposed. The method together with the the second kind of elliptic equation and the symbolic computation system Mathematica is used to construct the new exact Jacobi elliptic function solutions, the degenerated soliton-like solutions and trigonometric function solutions of the composed KdV equation with forced variable coefficients.

Various phenomena governed by power law distributions are common in nature and in society, thus their study has broad and far-reaching significance. This paper focuses on a kind of power law phenomenon originating from human behaviors, for instance, popular web page or blog distribution and so on. It is pointed out that human behaviors are rational and selfinterested. On the basis of this, we established a model and found that variables of this kind obeyed a new distribution which was named as the Wei distribution rather than Zipf distribution. Based on Wei distribution, Yule distribution was generated through changing the restrictions. Moreover, in the paper, we describe the meaning of Yule distribution and set up a new distribution which would characterize the reallife conditions more faithfully. It is illustrated further that the human behavior theory could be applied generally to explaining this kind of power law phenomenon and the model could be expanded easily. Besides, we put forward a function to compare different kinds of distributions. This paper supports the reductionism in power law phenomena, which means that simple human behaviors lead power law to be universal.

Stochastic fluctuations in the single potential well and probability transitions between the two potential wells exist in the bistable system under the action of noise. These two kinds of movements in different hierarchies have different time scales, and the fluctuations in the low hierarchy potential well have effects on the transitions between high hierarchy potential wells. When time scale of the first input periodical signal and noise-induced probability transitions between the potential wells achieves stochastic synchronization, the phenomenon of stochastic resonance happens. When the second driven periodic signal, the time scales of which matches that of the intrawell fluctuations, is introduced into the system, the phenomenon of frequency absorption happens and the effect of stochastic resonance in bistable system is reinforced.

Mei symmetry and Mei conserved quantity of Nielsen equations for a nonholonomic system of unilateral non-Chetaev's type in the event space are studied. The differential equations of motion for the system are established. The definition and the criteria of Mei symmetry, loose Mei symmetry and strict Mei symmetry for the system are respectively given. The existence condition and the expression of Mei conserved quantity deduced directly from Mei symmetry are obtained. An example is given to illustrate the application of the results.

The dynamic behaviors of a relative-rotation nonlinear dynamic system with cubic coupled terms are studied. First, the dynamic equation of coupled system with nonlinear elastic force and generalized friction and harmonic excitation is deduced. The approximate solution of coupled unautonomous equation under harmonic excitation is obtained by the method of multiple scales. The effect of coupled terms on system resonance is analyzed in respect of principal resonance and internal resonance. The singularity stability of bifurcation function of principal resonance is studied by singularity theory, and the transfer concourse and topological structure of bifurcation function are obtained.

In this paper, the F-expansion method is revised by adding a term with negative index with constant c，then the irregular wave solutions are avoided. By using this method， different new periodic wave solutions expressed by the generalized KdV equation are obtained.In the special cases，these periodic wave solutions degenerate to the corresponding Jacobi elliptic function solutions，solitary wave solutions or trigonometric function solutions.

Using the coordinate and momentum transformation theory and the trial function methed, the exact wave function of the coupled harmonic oscillator with time-dependent mass and frequency is derived.

The method of solving the radial Schrdinger equation is studied by under the tight coupling condition of several positivepower and inversepower potential functions. The precise analytic solutions and the conditions that determine the existence of analytic solution are searched when the potential of the radial Schrdinger equation is V（r）=α_{1}r^{8}＋α_{2}r^{3}+α_{3}r^{2}+β_{3}r^{-1}＋β_{2}r^{-3}＋β_{1}r^{-4}. According to the single valued, bounded and continuous stipulations of wave function in a quantum system, firstly, the asymptotic solution is solved when the radial coordinate r→∞ and r→0; secondly, the asymptotic solutions are combined with the series solutions in the neighborhood of irregular singularities; and then the power series coefficients are compared. A series of analytic solutions of the stationary state wave function and the corresponding energy level structure are deduced by tight coupling between the coefficients of potential functions for the radial Schrdinger equation. And the solutions are discussed and the conclusions are made.

A two-mode integral form projection unitary operator is introduced. Using the technique of integration within an ordered product （IWOP） of operators, the transformation property of the integral form projection operator is studied, and its normal ordered expression is derived. Then the Hamiltonian of two-mode quantum harmonic oscillator system with coordinate-momentum coupling is diagonalized by virtue of the integral form projection operator, there by the eigenenergy and eigenfunction of the system are exactly solved. Lastly, a special case is discussed.

On the basis of the new introduced two-mode coherent entangled state representation and the technique of integration within an ordered product of operators, we derived some operator identities of the operator functions about the quadrature of two-mode optical field, such as （X_{1}+X_{2}）/2^{1/2} and （μX_{1}+νX_{2}）/λ, which should be helpful to further studying the high order squeezing properties of optical field and to constructing some generalized-squeezed states.

In order to obtain steady key rate for a continuous variable quantum key distribution system, Alice should adaptively adjust the intensity of optical pulses she sents. In this paper, an optical fiber channel is taken as the additive bosonic quantum Gaussian channel. It is proven that a Gaussian channel transforms Gaussian states into Gaussian states. The maximum-likelihood method is used to estimate the channel parameters after balanced homodyne detection. Then, Alice can change the intensity of optical pulses adaptively according to the estimated noise level, thereby a steady key rate can be obtained.

The security of quantum-key-distribution system is analyzed using the decoy state protocol with two weak decoy states and one signal state. The comparison is based on the key generation rate as a function of distance for two types of decoy states: the vacuum and a weak decoy state, asymptotically approaching the theoretical limit of the most general type of decoy protocol, one-decoy-state protocol. We studied the two-decoy-state protocol in connection with the intensity of signal state, and the gain of signal states, the overall quantum bit error rate, the gain of single-photon states and the error rate of single-photon states and arrived at the optimal condition which maximizes the key generation rate.

Applying the new equation of state density motivated by the generalized uncertainty relation in the quantum gravity field，we investigate the black hole entropy on the background of three-dimensionnal BTZ. When the λ with the Planck scale，and relative to the dimensions of spacetime introduced in generalized uncertainty relation has a fixed value，we obtain the Bekenstein-Hawking entropy of BTZ black hole and the correction term. Because we have used the new equation of state density，in our results the divergence term appearing in the brick-wall model is removed，at the same time with holding the small mass approximation. Thus the correction value of the Bekenstein-Hawking entropy of the black hole is derived from the quantum statistical view. It depends the understanding of the correction value of the black hole entropy.

Using a method and its results which deals with the geometrical operators consistently，a complete area spectrum is obtained. At the same time，a different interpretation and result is given for the action of the Euclidean term of Thiemanns Hamiltonian constraint. In the graphical scheme of the grasping action，using the simplification of any triple of grasping actions of the volume operator，the associated recoupling matrices are demonstrated. Employing the movements of grasps，a complete and exact general expression of recoupling matrix for any triple of grasp actings on any valent vertex is detailed.

In this paper, an adaptive-tree-structure-based fuzzy model is applied to predict chaotic time series. The fuzzy partition of input data set is adaptive to the pattern of data distribution to optimize the number of the subsets automatically by binary-tree model. A fuzzy area around every discriminant edge is set up by the membership functions corresponding to every subset of input data. A complex nonlinear function is obtained by piecewise linear approximation and smoothing the discontinuous at the discriminant edges of subsets to reduce the error of approximation. The fuzzy tree model is evaluated using prediction of the Mackey-Glass, Lorenz and Henon chaotic time series. In comparison with some existing methods, it is shown that the FT is also of less computation and high accuracy.

The methods of generating spread-spectrum sequences with Mach-Zehnder interferometer are studied, also the principle and the scheme of a reported Chebyshev chaotic sequence generator is analyzed. To eliminate the limitation caused by the size of optical elements, a new scheme using the properties of cosine function is put forward. By using voltage to adjust the refractive index of the arms of Mach-Zehnder interferometer, a wavelength adaptive generator can be realized, with which different Chebyshev optical chaotic sequences with different input wavelengths could be obtained.

A stablility theorem is proposed for fractional systems whose order is not greater than 1. Based on this theorem, back-stepping approach for designing controller is extended to fractional order chaotic system. And the fractional order Newton-Leipnik chaotic system is synchronized based on the extended backstepping approach. Numerical simulation certifies effectiveness of the method.

A 1-D smooth map constructed from DOG wavelet function is discussed in this paper. With analysis on the fixed points and the constructed iterative curves，its dynamical characteristics are thoroughly studied. It is found that the number of the fixed points will increase or decrease depending on the dilation and translation operation of the wavelet and thus the stable or unstable cross points and tangent point or zero points are produced. Numerical calculations are performed to obtain the dynamic behavior，bifurcation diagrams and Lyapunov spectra. Some nonlinear phenomena，such as period-doubling bifurcation，tangent bifurcation，boundary crisis bifurcation，periodic window，and imperfect Feigenbaum-tree，are revealed and investigated.

Sine inverter is a time varying nonlinear system, for which two scales, namely the fast and slow-scale, can be used to analyses its stability. Based on this, fast- and slow-scale discrete model of H-bridge sine inverter under proportional control are derived respectively. For the fast-scale stability, folded diagram and spectrum analysis are introduced. For the slow-scale stability, slow-scale fixed points and a theorem of slow-scale instability of a discrete-time periodically varying system are proposed. It is shown that slow-scale instability is an effective criterion for chaos motions of discrete-time periodically varying system. Research shows that proposed methods can be used to analyse the fast- and slow-scale instability and chaotic behavior of H-bridge sine inverter.

By differentiating the parameters of a newly proposed chaotic system with invariable Lyapunov exponent spectrum, an improved chaotic system is reported in this paper. The system has three important properties which can be summarized as: invariable Lyapunov exponent spectrum for two parameters, existence of the global linear amplitude adjuster and the phase reversal parameter. Simulation analysis by Lyapunov exponent spectrum and bifurcation diagram combined with theoretical reasoning and theory proving, the dynamical characteristics of the system are investigated and expounded. An analog electronic circuit is designed to implement the improved system, and then the chaotic behavior is verified by physics experiments. Finally, the synchronization of the new chaotic system is realized by using the single variable coupling feedback, and meanwhile the circuit of synchronization is presented and the synchronization condition is verified.

Using the improved NaSch traffic model, we simulate the traffic flow on the crossroad under open boundary condition. The simulations are controled by ordinary traffic light and intelligent traffic light respectively. The results show that the average speed and the traffic flux under control of the later are greater than that controled by the former. Moreover, the vehicles controled by intelligent traffic light can cut down the unnecessary waiting time.

Based on the principle of automatic driving, the mixed traffic flow model of vehicles with different lengths and maximum velocities on one-lane highway is proposed. We obtained the fundamental diagrams of traffic flow model under different parameters via numerical simulation. Moreover, with the help of mean-field theory, the analytical results are given. The theoretical results are in agreement with those from numerical simulations.

In this paper,we study the effects of decelerating lane in the off-ramp system of freeway. Based on the NaSch model of traffic flow, a model of two-lane system with off-ramp is established. From the numerical simulations, we obtain the fundamental diagrams of traffic flow under different parameters. Moreover, we compare the road including decelerating lane with the road without it.

Taking into account three basic pedestrian behaviors, namely the position-exchange, lateral-move and step back, we proposed an improved cellular automaton model to simulated pedestrian counter flow with different velocities in subterranean channel. We compared cellular automaton model with Weng's model. By computer simulation, the improved model is proved to be more efficient. Moreover, pedestrians in the system tend to have higher mean velocity and lower occupancy density.

The method of measuring the intensity and state of polarization of optical radiation using polarization interference imaging spectrometer （PIIS） combines the basic theories of polarization interference imaging spectroscopy and polarization measurement technology. In this way, the PIIS can be used to detect not only the interference characteristics but also the polarization parameters. Based on the study of polarization measurement, this paper compares the effect of the degree of polarization errors for the conventional detection angles of 45°and 60°. The optimization of detection angle is then discussed, which will lead to a better angle choice for polarization measurements that enables the interference imaging spectrometers to be adapted to almost every field in which the spectroscopy has ventured and able to some where field which has not touched before.

The influence of electron gas on atomic nuclei ^{56}Fe and ^{120}Sr in compact matter has been studied by relativistic mean field theory in combination with Wigner-Seitz approximation.Bardeen-Cooper-Schrieffer approximation is used to deal with pairing correlations.The results show that single-particle energy levels, nucleon density distributions and root-mean-square radii are changed due to electron gas. The influence on protons is greater than that on neutrons.

Half-lives of heavy cluster （^{14}C-^{34}Si） emission for heavy nuclei are theoretically calculated in the frame of WKB approximation. The microscopic nuclear potential between heavy cluster and residual nuclei for parent nuclei is obtained by double-folding process by using density-dependent nucleon-nucleon interaction with zero-range exchange term. The results are compared with those of liquid-drop model, systematic formula and experimental data.It is obvious that the present calculation can successfully give the lifetimes of heavy cluster（^{14}C-^{34}Si） emission for heavy nuclei. It also provides reasonable prediction for the other heavy cluster（^{15}N, ^{46}Ar and ^{48}Ca, etc.） emission for some nuclei.

Highdensity SiO_{x} nanowires were fabricated on a largescale using carbonassisted CVD method by Fe—Al—O catalyst at 1140℃ in flowing N_{2}/H_{2},N_{2} and NH_{3} atmospheres.The SiO_{x} nanowires have uniform diameters of 20—300 nm and lengths of up to a few hundred micrometers. SEM, TEM, EDS, FTIR and PL were preformed to characterize the microstructure, composition and optical performance of the nanowires. Energy dispersive Xray spectral analysis reveals that the nanowires consist of Si and O elements in an atomic ratio of approximately 1∶18The nanowires show IR absorption peaks at 482,806,1095 and 1132 cm^{-1}. The PL peak of the nanowires is located at 440 nm （283eV）.The PL inensity of the SiO_{x} nanowire （N_{2}） is 10^{4} times that of the SiO_{x} nanowire（N_{2}/NH_{3}）.

Bremsstrahlung will be produced when the metal target is bombarded by high energy electrons. Bremsstrahlung characteristics is discussed under the condition of optimal bremsstrahlung efficiency for the purpose of improving its quality. This paper simulates the bremsstrahlung characteristics, combining theoretical analysis with MCNP/4C program. Electron yield characteristics, photon yield of conic collimator, angular distributions and energy spectrum distributions are simulated for targets of different thickness and materials for 10 and 20?MeV electrons. The simulation results show that photon efficiency, emitted electron and photon flux distribution is related to material and thickness of the target. And the maximum photon yield of bremsstrahlung is mainly dependent on the electron energe range, and related to atomic number. Thus the relevant boundary conditions and characteristics are obtained under the optimal bremsstrahlung efficiency of different targets for 10 and 20?MeV electrons.

By using a fully relativistic distortedwave method, the electron impact excitation cross sections of Belike N^{3+} and O^{4+} ions from the ground state 2s^{2}^{1}S to 2s2p ^{1,3}P, 2p^{2}^{3}P, ^{1}D and ^{1}S 2s2p and from the metastable 2s2p ^{3}P to 2p^{2}^{3}P have been calculated systematically. Meanwhile, the correlation effects of target states are also discussed. For the low energy collisions the correlation effects play a very important role, and accurate target wavefunctions are needed, but for the high energy collisions, the more correlation effects are not so important. For twoelectron excitations, the configuration interactions are very important and the excitations from ground state to the 2p^{2}^{3}P_{0} and ^{1}S_{0} have larger cross sections than the other tow electron excitations.

In the framework of quantum defect theory, the photofragment yield spectra for F^{-} from F_{2} ion-pair production have been studied. The multiple-scattering self-consistent field method and the theoretical time-dependent wave-packet dynamics method are employed. Based on our calculated results, the vibrational resolved assignment and the calculated relative intensity are provided.

BiVO_{4} powders with different morphologies and structures were synthesized by hydrothermal method by controlling the n_{Bi}/n_{V}，reaction temperature，reaction time and pH value. Xray diffraction，transmission electron microscope，scanning electron microscope，Raman，Fourier transform infrared and ultravioletvisible absorbance spectroscopy were used to characterize the products. The results showed that highly crystalline monoclinic scheelite structure of BiVO_{4} was obtained through increasing the n_{Bi}/n_{V} or the pH value. By controlling different reaction temperature and reaction time，monoclinic scheelite structure BiVO_{4} with different grain sizes and different internal structures were obtained. Besides，BiVO_{4} with different morphologies can be controlably synthesized when using different surfactants.

An algorithm is employed to elucidate the molecular bond polarizabilities of methylviologen, including their temporal relaxations，from Raman intensities. The main characteristic of the Raman excited virtual state of MV is that the excited electrons tend to flow to the molecule periphery and to the bond connecting its two rings due to electronic repulsion. The bond electron densities of its ground state can be mapped out by the temporal bond polarizabilities at the final stage of relaxation.

Based on the multi-configuration Dirac-Fock method, the level structures and decay processes of the inner-shell excited states （1s2s^{2}2p^{2} and 1s2s2p^{3}） created by 1s-2p photoexcitation of C Ⅱ ion have been studied systematically, which the correlation and relaxation effects, relativistic radiative corrections have been considered in the calculations. Auger and radiative transition energy, decay rate and linewidth for each inner-shell excited states of C Ⅱ ion has been calculated. Correspondingly, lifetimes of those autoionization states can be obtained via the Heisenberg uncertainty principle. A comparison of the present calculations with the latest experimental and available theoretical results is also presented.

Based on multi-configuration Dirac-Fock method, radiative recombination （RR） process of H-like uranium ion and radiative decay process of the resultant ion have been studied systematically using the recently developed program RERR06, which is based on GRASP92 and RATIP packages. In the studies, the RR cross sections from 1s state of H-like uranium ion to 1snl （1≤n≤8, 0≤l≤6） of He-like uranium ion have been calculated in detail. Meanwhile, the relative intensity of the radiative decay spectra of the RR final states have also been studied. It was found that the RR cross sections to different orbitals decrease with the increasing of principal quantum number. Furthermore, an important contribution of the radiative decay of the RR final states to the relative intensity of Kα spectra has also been found in the present work.

We have studied experimentally the dipole orientation and reorientation dynamics of single dye molecules at room temperature. The dipole rotation of individual dye molecules which are physisorbed to glass and embedded in spin-cast polymer films is measured with the confocal scanning optical microscopy and the method of fluorescence polarization analysis, the probability of reorientation of single molecules found to be about 5%—9% when embedded in polymer, and about 26% when absorbed to glass by using the statistic analysis. It is found the quantum jumps for molecule dipole reorientation are distributed in several different polarization states by measuring the polarization degree of fluorescent photons.

Interaction potential of the OD（X^{2}Π） radical is constructed by employing the CCSD（T） theory in combination with the correlationconsistent quintuple basis set, augccpV5Z, in the valence range. Using the potential, the spectroscopic parameters are accurately determined. The present D_{0}, D_{e}, R_{e}, ω_{e}, ω_{e}χ_{e} and B_{e} values are of 44574, 46225 eV, 009702 nm, 2724923, 453534 and 100096 cm^{-1}, respectively, which are in excellent agreement with the recent measurements wherever available. A total of 23 vibrational states have been found when J = 0 by solving the radial Schrdinger equation of nuclear motion. The complete vibrational levels, classical turning points, initial rotation and centrifugal distortion constants when J = 0 are reported for the first time, which are in good agreement with the available experimental results. The total and various partialwave cross sections are calculated for the elastic collisions of O and D atoms in their ground states at low and ultralow temperatures when the two atoms approach each other along the OD（X^{2}Π） potential energy curve. The impact energy range covers a vange from 10×10^{-11} to 10 ×10^{-3}a.u.. One shape resonance has been found in the total elastic cross sections. Contribution to the total elastic cross sections by each partial wave is investigated carefully. The results show that the shape of the total elastic cross sections is mainly dominated by the s partial wave. The shape resonances coming from the higher partial waves are covered up by the strong s partial wave cross sections.

The intermolecular interaction potential developed in our previous research and close-coupling method are applied to the HI-He system, and the differential cross sections, partial cross sections （PCSs） and integral cross sections （ICSs） at the incident energies from 1 to 140?meV are calculated. The reliability of the interaction potential for He-HI system and the close-coupling calculations have been verified by comparing with the PCSs of He-HX （X=F, Cl, Br） systems. The results show: （1） The probability at small scattering angles is larger than that for large scattering angles. With the increase of collision energy, the scattering probability and the tail effect fall down. （2） Total ICSs mainly come from the elastic collisions. The transitions of 00 → 01and 02 contribute to the total inelastic ICSs predominantly.

Detailed and accurate EMS measurements of complete valence region of N_{2} molecule using a high resolution （ΔE=08?eV，Δp≈01 a.u.） high efficiency electron momentum spectrometer have been performed. Several ionization peaks attributed to inner valence orbitals and correlation （satellite） states are resolved for the first time in the binding energy spectroscopy. By comparing the （e，2e） cross sections of valence orbitals and the correlations states，significant insights into the dominant character of the valence and correlation states of N_{2} are obtained. The EMS momentum distributions show that they provide a sensitive test for orbital wavefunctions from different theoretical models.

PHYSICS OF GASES, PLASMAS, AND ELECTRIC DISCHARGES

We investigates the magnetic field intensity effects on the discharge characterization of Hall thrusters for optimization of the magnetic fields by varying the strength of magnetic fields with its topography holds constant （keep in "focusing" type）, and making an experimental study of the thruster's operating performance working in different regions of magneto-ampere curve through Langmuir probe, spectrograph, etc.. The results shows that when the magnitude of magnetic field intensity is less than its optimal value, near-wall conductivity gives priority to the activity that electrons traverse the magnetic field; contrarily, discharge current varying with magnetic field intensity abnormally, and existing electrons conductive theories are failed to explain this phenomenon.

The IV characteristics of ZnOBi_{2}O_{3}, ZnOBi_{2}O_{3}MnO systems and commercial ZnO varistor ceramics were measured at temperatures in a wide range of -180?℃<T<100?℃. It was found that experimental conductancetemperature curves under various currents can be theoretically reproduced by calculating conductance in parallel or series for mechanisms such as scattering mechanism, thermionic emission and tunneling effect. It was found that if the relation between a singlemechanism conductance and temperature obeys the exponential law with an exponent a little less than 1, then this mechanism will give rise to a high nonlinear exponent. The nonlinear exponent due to tunneling effect was 33, which was similar to the measured value of 43 for commercial ZnO varistor ceramics. This tunneling effect is active even at low electric field, especially in low temperature region.

The anisotropy, twophase pore and nonuniformity must be considered comprehensively in order to describe the oil and gas reservoirs accurately. Consequently a twophase anisotropic random medium model based on the twophase model and random medium theory is presented in this paper. The seismic wavefield numerical modeling was simulated by using the pseudospectral method. The results show that there exist complex wave field features such as wave scattering and traveling time disturbance and these features strongly depend on parameters of the random medium model, the nonuniformity rate mainly impacts the wave traveling time disturbance when spatial scale of nonuniformity is large and mainly impacts wave scattering when spatial scale of nonuniformity is small. The studies make it possible to retrieve the nonuniformity on statistical basis and are useful to understanding more deeply the seismic wave propagation in oil and gas reservoirs.

Using MonteCarlo method, the law of statistics of recordbreaking high temperature events has been investigated based on the statistical characteristics of the temperature data registered in 40 years from 1961 to 2000 in Nanjing area, and the influence on the law of statistics of the global warming （ｖ=0.006?℃/a） during 20th century and Nanjing regional warming （ｖ=0.017?℃/a） in last 40 years has also been compared. Both the theoretic analysis and Monte Carlo simulation results show that the most likely probability of its occurrence intensity of the kth recordbreaking high temperature events takes the form of linear increase with k （k=1,2,3,\:）, and the frequency of the occurrence of recordbreaking high temperature events in a year tends to 1/（t+1）, decreasing with time t, and in years when the average temperature is high, the probability of the occurrence of recordbreaking high temperature events is also high, otherwise, the probability is low. The result also indicates that the velocity of global warming （ｖ=0.006?℃/a） in 20th century and the rate of regional warming （ｖ=0.017?℃/a） of Nanjing area in last 40 years are yet insufficient to alter meaningfully the intensity and frequency of record high temperature events. However, such velocity of warming will ultimately make the frequency of the occurrence of recordbreaking high temperature events decrease gradually towards a constant, being approximately equal to the warming velocities. In addition, we have also studied the effect of variances and autocorrelation between temperatures in two successive days on recordbreaking high temperature events，finding that the effect of different variance and weak autocorrelation on the intensity and probability of occurrence of the recordbreaking high temperature events in a year can be neglected.

The long range correlation of extreme events of the Lorenz system is discussed by using the method of fixed threshold. It turns out that all of the extreme events with different threshold have long range correlation. The scaling exponents are similar, but just smaller than the original series. The long range correlation of extreme events is less effected by the initial value changes, but it decreases distinctly when the parameters increases. The long range correlation of Lorenz systems extreme events series has the traits of memory when compared with Gaussian white noise. Finally, we use the maximal day air temperature data of 194 stations between 1957 and 2004, from the National Climate Center of China, to reveal that the similar law exists in the actual meteorological factors.

Approximate entropy （ApEn） is well known to be an effective abrupt change detection method in dynamic structure. Based on this, we compare the performances between the moving detrended fluctuation analysis （MDFA） and ApEn in detecting abrupt dynamic change. The results show that the MDFA results almost do not depend on length of subseries, and while the ApEn could identify dynamic structure to some extent but still depends on the length of subseries. At the same time, there exists a huge drift for the ApEn results which means the actual time-instants of abrupt change dont coincide with the detected one. Therefore, compared with ApEn, MDFA is more suitable to be used to detect abrupt dynamic change, and the advantage of MDFA is obvious.