The investigation on the microstructure of the residual strain, dislocation and twin nearby interfaces resulted from the thermal stress in Gr(C)/Mg composite has been carried out in this work. The dislocations sprout at interfaces between Gr(C) and the matrix and move towards matrix under the axial, radial and tangential stresses. There is a strain layer with a thickness of about 150 nm nearby the Gr(C) fiber, and the strain is concentrated in narrow areas of the matrix. The formation of twins occurs at one interface and usually ends at a neighbor interface. As the twin is thick, it may cause the neighbor interface apart.

Bulk nano-scale multilayers have been prepared by repeated pressing-rolling the initially macroscopically thick layered sample. The SEM and TEM observations revealed that the final layer thickness in the regular laminates is about 10 nm. The Fe/Cu multilayers exhibited a clear magnetoresistance (MR) ratio at 77 K. The tensile strength is over 1500 MPa and its elongation is about 0.8% at room temperature.

On the base of sub-regular melt model (SELF-SReM4.0 model), a model SELF-SReM4.1 was designed to solve the precipitate of carbide. Taking C-Mn-Fe-Si quarternary system as an example, the calculated results of component activities by above mentioned two models are introduced and some equilibrium rules of Si-Mn and Mn-Fe systems are also analysed.

The pilot test of bath smelting reduction process producing stainless master alloy was carried out in a 15 t complex blowing converter. It was found that the yield of chromium is not high and the master alloy contains too much phosphorous. It indicates that the bath smelting reduction process producing stainless master alloy is indeed a process of [C] - [Cr] selective oxidational dephosphorization procedure. Essentially, it is a selective oxidation among [C]-[Cr]-[P]elements. Based on the model of SELF-SReM4, these two problems were discussed quantitatively in this paper.

The X-ray diffraction and neutron diffraction patterns of Y3(Fe, Mo)29 compound have been analysed by Rietveld method. The structural relationship of the 3:29 phase with 2:17 and 1:12 phases has been studied by coordinate transformation. The results show that the 3:29 compound belongs to A2/m space group and the 3:29 compound is a kind of mixed structure of 2:17 and 1:12 structures.

The corrosion behavior of a steel 1Cr11Ni2W2MoV has been studied in the presence of NaCl and water vapor at 500 - 700 ℃. The results indicate that corrosion of the steel 1Cr11Ni2W2MoV was not obvious in the atmospheres without NaCl, while corrosion was severely speeded up when the atmosphere contained NaCl and water vapor. Optical microscope (OM),X-ray diffraction (XRD), scanning electron microscope (SEM) and electron probe microanalysis (EPMA) were used to analyse the corrosion products. The mechanism of the effect of water vapor and NaCl on corrosion of this steel was proposed.

Co85Cu15 hypoperitectic alloy melt was undercooled by 351 K (0.21TL ) with glass fluxing technique. The hypercooling limit of this alloy was determined to be 379 K. With increasing undercooling temperature, the microstructure transformed from dendrites into equiaxed grains and was refined drastically. Significant solute trapping has taken place during the rapid solidification of highly undercooled Co85Cu15 alloy, and it resulted in the remarkable extension of solubility of Cu in Co phase and much weaker intragranular segregation. It was also revealed by X-ray diffraction that the microstructures were comprised mainly by metastable supersaturated α-Co and Cu solid solutions together with a small amount of ε-Co phase. In addition,the theoretical calculation for nucleation rate indicates that heterogeneous nucleation has still taken place at an undercooling of 351 K. Calculations of dendrite growth velocity, solubility of Cu in α-Co, partial undercoolings, and dendrite tip radius were also performed on the basis of LKT/BCT theory.

The desulphurization process by powder injection and blowing in RH refining of molten steel and its mechanism have been considered and analyzed. Based on the two-resistance mass transfer theory and the mass balance of sulphur in the system, a kinetic model for the process has been developed. The ralated parameters of the model have been determined. Calculating and modelling for the process by injecting and blowing the lime based powder flux under the assumed operating modes in a RH degasser of 300 t capacity have been carried out using the model. The corresponding circulation rate and the powder injection and blowing rate were taken as 100 t/min and 150 kg/min, respectively. The initial contents of sulphur and the amounts of powder injection and blowing were respectively assumed to be 70, 60, 50, 40, 30, 20×10-4% (massfraction) and 10, 8, 6, 5, 4, 3 kg/t-steel. The total treatment time for desulphurization under each mode was set up to be 24 min that is equivalent to eight circulation cycles of liquid steel to be treated. The results indicated that the prediction made by this model is in good agreement with some data of the industrial experiments and production practice, injecting and blowing the lime-based powder flux(85%CaO+15%CaF2)of 3-5 kg/t-steel with the total treatment time of 12-20 min, it is entirely possible to make the sulphur content in molten steel decrease to the ultra-low level below (5- 10)× 10-4% from (60-80)×10-4%; intensifying the powder injection and blowing operation and increasing the circulation rate of liquid steel may effectively raise the rate of the process in RH refining.

Fe-6.5%Si samples were prepared from cold-rolled Fe-3%Si sheets by the deposition-diffusion method. The Si concentrations of the deposited samples with and without homogenization were examined by the X-ray diffraction technique. The texture characteristics of the cold-rolled, deposited and homogenized samples were evaluated by the ODF analysis. It was shown that after deposition the typical cold-rolling textures were weakened to a large extent while the η-fiber (<001>//RD)with the maximum density at the {120}<001>component became dominated.After homogenization the {120}<001>component evolved into{130}<001>for the conventional cold-rolled samed sample with 0.9 mm thickness but it kept strongest for the cross-shear cold-rolled sample with 0.35 mm thickness.

The micro-denucleation technique of bulk melt was applied to Ni-Ni3Si alloy system. In two composition ranges of Ni-(6.07 - 19.7)%Si and Ni-(20.5 - 22)%Si, the alloys were undercooled up to the highest undercooling of 344 and 265 K respectively. The highest undercooling could be held for 20 melting-superheating-cooling circles. The experimental results indicated that, during the solidification of the alloys with compositions closed to the eutectic composition in Ni-Ni3Si alloy system, Ni3Si was always nucleated heterogeneously ahead of α(Ni) phase. The catalytic factors f(θ) of Ni3Si and α(Ni) were 0.069 and 0.28 respectively,calculated with the experimental data and the kinetic model of heterogeneous surface nucleation.The temperature vs composition curve with the same nucleation rate for Ni3Si and α(Ni) lay on the left side of eutectic point completely, which indicated that in the undercooled melts closed to the eutectic composition the nucleation of faceted Ni3Si phase was easier than that of α(Ni)phase.

On a small directionally solidified equipment, the technique of double directional solidification with opposite direction was used to produce Ni52Al48 single crystals and the growth rates are 15, 30 and 60 mm/h respectively.

The interfacial desulphurization reaction kinetics between molten Sn-S alloy and ZnCl2-NaCl salt was studied by an AC impedance method and the impedance response of this system was given. By using Randles equivalent circuit, the kinetic factors on electrode reaction between the alloy and the salt were obtained, such as electrolyte resistance Re, double-layer capacitance Cd, charge transfer resistance Rct, rate constant of desulfurization reaction Kf and so on. When overpotential η equals to -85 mV, these values are Re=2.93×10-5 Ω.m2, Rct =2.47×10-5 Ω.m2, Cd=1.01 F/m2, Kf=2.1 × 10-3 s-1, respectively.

The tensile tests of Al67Ti25Mn8 intermetallic compound were performed at 1173 K and at the range of strain rates from 3.34×10-5 to 6.68× 10-4 s-1. The results show that with decreasing strain rates, the yield stress decreases and the tensile ductility increases from 2.8% to 21%. The good high temperature ductility at low strain rates is attributed to dynamic recovery and dynamic recrystallization during deformation. The fractured surfaces of this compound are intergranular, which suggests that the grain boundaries of this alloy are also intrinsic weak.

X-ray diffractometer and scanning electron microscope were applied to examine the specimens processed from CuO/Al powder compacts under different conditions. The influence of temperature and environment medium on the chemical reactions in CuO/Al system is studied,and it is proposed that α-Al2O3 particle reinforced Al-Cu alloy matrix composites can be processed by introducing the chemical reactions in CuO/Al system.

During isothermal decomposition of the supersaturated α2 phase in Ti-45%Al alloy at 950℃, the volume fractions of the decomposed γ phase and the lattice parameters of both α2 and γ phases have been investigated by X-ray diffraction. The results show that α2 phase in the primary lamellar microstructure (PLMS) is a metastable phase, and its composition should be identified by the tangency point on the common tangent of metastable α2 and γ phase free-energy curves. This composition is very important in calculation of free energy of the alloy and the driving force of discontinuous coarsening (DC). There exists transformation stress inside the original α grains at the direction perpendicular to PLMS which is a compress stress of about 50 MPa and will greatly affect the microstructure of the alloy.

Cyclic deformation behavior of [011] multiple-slip-oriented copper single crystals was investigated with the plastic resolved shear strain amplitudes (γp1) in the range of 1.1 × 10-4 - 7.2× 10-3. It is shown that the cyclic deformation behavior of [011] copper single crystal is quite different from that of [001] and [111] multiple-slip-oriented crystals. [011] crystals exhibit rather low initial hardening rate, which does not increase notably even under higher plastic strain amplitudes. The cyclic stress-strain curve (CSSC) of [011] crystal shows a plateau region over the range of 1.1 × 10-4 - 7.2 × 10-3. The occurrence of plateau region in CSSC is mainly determined by the slip characteristics and the corresponding dislocation reactions in the crystal. Moreover,the hysteresis loop shape parameter VH could be efficiently employed to determine the nucleation stress of PSBs in [011] multiple-slip-oriented copper single crystal.

Surface deformation features were investigated on [011] multiple-slip-oriented copper single crystals under different strain amplitudes. It is shown that the fatigue limit of [011]multiple-slip-oriented copper single crystal is lower than that of [001] multiple-slip-oriented copper single crystal. For γpl≥2.5×10-3, two types of deformation bands (DBI and DBII)formed on the specimen surface and their habit planes are perpendicular to each other strictly.This may be caused by the occurrence of irreversible rotation of crystal under cyclic loading. For γpl ≥5.0×10-3, another type of deformation band (DBIII) forms on the specimen surface and its habit plane is (001). The occurrence of the obvious cyclic softening processes, in cyclically deformed [011] copper single crystals with high strain amplitudes, was largely attributed to the