The microstructure evolution during deformation of undercooled austenite in three low carbon steels with much close contents of Si and C but different contents of Mn has been investigated by means of hot compression simulation experiment at 760℃ and strain rate of 1 s-1. The effects of the content of Mn on the deformed behavior of undercooled austentie, the kinetics characteristics of deformation enhanced ferrite transformation and the ferrite grain size have been studied. It was found that the addition of Mn delays the transformation during deformation of undercooled austenite and higher true strain for the transformation is needed with increasing Mn content. Transformation kinetics can be divided into three stages, longer time and higher true strain are needed for each stages with increasing Mn content in the steels. Deformation enhanced transformation induces forming ultra--fine equiaxed ferrite grains with sizes of 3.57±1.60, 2.00±1.05 and 2.29±1.02 um in steels with 0.48, 0.84 and 1.29 Mn respectively, and homogeneously distributed minor second phases.

A two--dimensional cellular automaton model is developed to simulate the transformation from austenite to ferrite in low carbon steels during continuous cooling.The model, with a local rule--based scheme involving nucleation and growth of ferrite,incorporates the change of the solute field into a nucleation or growth function whichis utilized by the automaton in a probabilistic fashion. The competition betweennucleation and growth of ferrite is described by the changes of a nucleationprobability and a growth probability in this cellular automaton lattice. The procedureof this cellular automaton modelling is generalized, and the effect of coolingconditions on this transformation is also discussed.

A series of Al--Cu--Fe alloys with compositions ofAl48-60Cu33-50Fe0-10 was prepared and the phase constituents in thesealloys quenched from various temperatures were identified by using XRD, SEM, EDXS and EBSD attached to SEM, and TEM (including HRTEM). Compared with the polythermal projection proposed by Gayle et al, a main amendment is to divide the previous φ--region into β+φ two regions. A new ternary reaction is the quasiperitectic at U8: L+βIQC+φ.EDXS reveals that the composition of φ phase with Ni2Al3 structure is in Al47.3-50.6Cu45.4-48.1Fe4.5-5.2,and its center composition is Al47.9Cu47.1Fe5.0.

The directionally solidified microstructures of peritectic alloys are affected markedly by the nucleation and growth conditions. By using nucleation and constitutionalsupercooling criterion and assuming that the maximum interface growth temperature of phase growth is more stability in the directional solidification process, the composition range is theoretically developed for predicting the nucleation at the plane-front growth of properitectic and peritectic phases in an initial transient zone. The band--like microstructure occurs as a function of solidification distance and composition range is described on the basis of the purely diffusivemodel. The calculated results of Fe--Ni and Pb--Bi peritectic alloys are in agreement with their experimental data.

A numerical model describing the microstructure evolution under the common action of the nucleation, the diffusional growth, the collisions and coagulations of the minority phase droplets and the spatial separation of the two liquid phases has been developed. The microstructure development in an unidirectionally solidified Al--Pb alloy has been simulated by coupling the calculated temperature and concentration fields with the kinetic equation which controls the microstructure evolution. It is indicated that the liquid--liquid transformation region moves from the bottom to the top of the sample during cooling. The Marangoni migration and the Stokes movement of droplets may cause a decrease of the local number density of droplets and an increase of the local supersaturation. The repeated nucleation of minority phase droplets may, therefore, occur locally.

A new way to improve computational efficiency of phase field is used to simulate the evolution of the equiaxed dendritic growth morphology in pure metal melt with initial dimensionless supercooling of 0.55. Much information about equiaxed dendritic growth has been got, which is in good agreement with experimental observation. The births of second arms and third arms have been simulated successfully by adding appropriate thermal noise. Furthermore, the dimensionless velocity and radius of the equiaxed dendritic axis' tip, and the Peclet number are computed, and the computational results by phase--field method are in good agreement with the ones by the microsolvability theory and the Ivantsov theory.

The diffusion barriers for the single helium atom in vanadium are studied by effective medium theory, in which the structure relaxation caused by the embedded atom is considered. The most possible diffusion path and diffusion barrier are determined. A single helium atom has the minimum energy about 4.37 eV in the tetrahedral interstitial site of bcc vanadium, and the embedding energy in octahedral interstitial site is larger than that in tetrahedral interstitial site. Helium atom in the bcc vanadium crystal diffuses along the path linked with positions (1, 1/4, 1/2), (1, 1/2, 1/2) and (1, 3/4, 1/2) on the (1 0 0) plane or the equivalent paths on corresponding planes.

The influences of high intensity magnetic field on the solidification structure of Bi--Mn alloys in semi--solidified as well as melted state have been investigated experimentally. It is found that under the magnetic field of 10T, MnBi phase moves towards to and accumulates at the periphery of the specimen, forming a ring--like MnBi phase--rich layer where the rod--like MnBi phase aligned along the magnetic direction, and no primary MnBi in the center of the sample is found. Slow solidification above the Curie point makes the rod--like MnBi phases evolve into single crystal gradually. The above phenomena is analyzed and discussed.

Effects of the annealing and deforming temperatures, thermal cycle, and the stress--strain cycle on the transformation and deformation characteristics of Ti49.4Ni50.6 superelastic (SE) coil spring have been investigated by means of differential scanning calorimetry, tensile test, and stress--strain cycling test. The phase--transformed type of the cold worked plus intermediate temperature annealed TiNi alloy is parent phase B2 rightleftharpoons R phase rightleftharpoons martensite B19'during cooling rightleftharpoons heating cycle. The martensitic transformationtemperatures increase and the R--phase transformationtemperatures decrease with increasing the annealingtemperature. The SE property can be obtained at roomtemperature for the 623---773 K annealing Ti49.4Ni50.6alloy spring, and the rigidity of the SE spring increaseswith increasing the deformed temperature. The SE propertyof Ti49.4Ni50.6alloy spring weakens while theannealed temperature is over 823 K. The smaller the takenshear strain, the higher the strain recovery rate ofTiNi SE spring is during the thermal cycle. The priorcyclic training can enhance the SE stability of TiNi

The modified Wagner's equation is deduced based on some assumptions and analysis of the Al content profiles in aluminide coatings formed on K465 alloy and pure Ni at 850℃, 950 ℃ and 1050 ℃ for 5 h, respectively. The complicated composition and microstrurcture of K465 alloy are taken into account when deducing the modified Wagner's equation, by which the interdiffusion coefficients are calculated as functions of Al content in β--NiAl phase as aluminide coatings. Calculated results show that the interdiffusion coefficients in β--NiAl phase formed on K465 alloy are evidently less than those on pure Ni, and the interdiffusion coefficients in β--NiAl phase formed on K465 alloy strongly depends on Al content. Effects of alloying elements and precipitates on the interdiffusion coefficients in β--NiAl phase formed on K465 alloy are briefly discussed.

The influence of high temperature heat treatment on microstructure and microhardness of (DS)NiAl--Cr(Mo)--Hf eutectic alloy fabricated with directional solidification under high temperature gradient was investigated. The results indicate that there is little change in the morphology of NiAl and Cr(Mo) phases after high temperature heat treatment. But the semicontinuous Heusler phase which is originally located on the grain boundaries is partially reduced and fine Heusler particles re--precipitated within NiAl matrix. Furthermore, NiAl precipitates coarsen in the Cr(Mo) phase and dislocation appears in the Cr(Mo) phase. Microhardness of as--fabricated (DS)NiAl--28Cr--5Mo--1Hf alloy is much higher than that of (DS)NiAl--28Cr--6Mo and basically unchanged after heat treatment

Bulk Ti--Zr--Ni quasicrystals (QCs) were prepared by using conventional suction casting method. The characterization on their mechanical properties has been carried out by means of indentation and compression tests at room temperature. The strength indices of the icosahedral phase (I--phase)—based duplex phase alloy (α--Ti(Zr)+I phase) are 5.5 GPa in hardness and 1000 MPa in fracture strength, respectively. The Young's modulus is in the range of 23---36 GPa. The as--castTi60Zr25Ni15 QC--based composite alloy exhibits an elastic deformation capability of 4.2 % under a monaxial compression, which is about 4 times as large as that of Al—based QCs. The quasi--cleavage fracture morphologies indicate the bulk QC--based composite alloys to be brittle fracture.

The effect of phosphorous addition on microstructure and mechanical properties of NiAl alloy has been investigated. The results indicate that a certain level of P additions can induce softening of polycrystalline NiAl, i.e., decreases yield stress and hardness and increases room temperature compressive ductility. Trapping of interstitial atoms and reducing of Peierls stress are proposed to be responsible for the softening. In addition, the macro--deforming behavior (compressive properties) of the alloy results from a competitive process of softening of grain interiors and hardening of grain boundaries.

Phase transformation behavior of the constrained Fe--Mn--Si--Cr--Ni shape memory alloy has been studied in detail through measuring the resistance and the recovery stress as afunction of temperature simultaneously during heating and cooling. On cooling, the recoverystress increases with lowering temperature. At the temperature that the recovery stress isgreater than the yield strength, plastic deformation will first take place. With the further decreaseof the temperature, the ε martensite will be induced by the recovery stress at the temperature thatthe recovery stress reaches to the critical stress for stress--inducedεmartensite. Below this temperature, the recovery stress decreases with loweringtemperature. The above plastic deformation and the εmartensitic transformation induced by the recovery stress drastically relaxthe recovery stress. The recovery stress equations during heating and cooling have beenestablished respectively. The design principle of compositions for Fe--Mn--Si--Cr--Ni shape memory alloys pip couplings has been put forward.

Standard k--ε,low Reynolds' number Jones--Launder k--ε and RNG models were adopted to simulate the flow field of metal melt in three different prebaked anode cell designs, i.e.80 kA cell with three anode risers, 160 kA cell with two anode risers and 200 kA cell with four side anode risers. Moreover, the melt velocities were measured using iron rod dissolution method. The simulating results were discussed and were compared with measuring data. It is shown that the turbulent viscosity in low Reynolds' number model is smaller than that in standard k--ε model. Standard k--ε and RNG model predict better the flow field of metal melt. Furthermore, the preliminary conclusion is that the low Reynolds' number model is not suitable to calculate metal melt flow in cells if the current is greater than 80 kA.

Al--O--N and Cr--O--N thin films were deposited on superalloy DSM11 by arc ion plating (AIP). The effects of flow rates of oxygen and nitrogen on phase composition and the interdiffusion of various elements at high temperature between DSM11 substrate and NiCoCrAlY coating were studied in the systems of DSM11/NiCoCrAlY, DSM11/Al--O--N/NiCoCrAlY and DSM11/Cr--O--N/NiCo- CrAlY. The results indicate that the Al(Cr)--O--N films are polycrystalline with phase constitution of α--Al2O3 + hexagonal AlN or Cr2O3+CrN , respectively. The relative intensity of AlN to Al2O3 or CrN to Cr2O3 is changed with the change of flow rates of oxygen and nitrogen. After oxidation at 1050℃ for 100 h, Al--O--N hindered the interdiffusion of elements between DSM11 and NiCoCrAlY more effectively than Cr--O--N film. The diffusion barrier layers hardly influenced the oxidation kinetics of the NiCoCrAlY coating.

Large eddy simulation of vortexing flow of molten steel in the continuous casting mold was conducted. The influence of the SEN (submerged entry nozzle) port angle and the SEN position to the turbulent vortex was analyzed. The mechanism of the turbulent vortex and the biased vortex was analyzed. A new vortex brake was designed to eliminate the vortexing flow, and the fluid flow with and without the vortex brake was simulated. The simulation results show that the turbulent vortex is caused by the turbulent energy; Increase of the SEN port angle and the depth of the SEN below the surface increases the residence time of the turbulent vortex in the free surface. The biased vortex is caused by the off--center of SEN and the turbulent energy. The new vortex brake can eliminate the turbulent vortex and decrease the strength of the ``biased vortex''.

The theory of fractal geometry is used to analyze the corrosion morphology images, which were acquired by scanner. The fractal dimension, generalized dimension, area factor and laucinarity of metallic corrosion images of samples that exposed in seawater were calculated. Taking the obtained characters of metallic samples and their corrosion modality as the knowledge base, the diagnosing system identifying corrosion modality of metallic material in seawater was established according to the theory of fuzzy pattern recognition. The morphology of corrosion images can be identified by acquired fractal characters.

The saturation dislocation microstructures in a [013] orientated coppersingle crystal during fatigue in air and NaCl aqueous solutionwere observed by electron channelingcontrast (ECC) technique in SEM and TEM. The resultsshow that the saturation dislocationstructure during the corrosion fatigue in the aqueous solution of 0.5 mol/L NaCl mainly consists of labyrinth and veins dislocation structures, which differs from the dislocation structures of the PSBs and veins in the air environment. The observation reveals that the influence of environment for fatigued dislocation structures is equivalent to increase the plastic shear strain amplitude to the fatigued sample when other conditions are identical.

The effect of zinc powder content and contamination of epoxy zinc--rich primer/substrate interface on the protective performance of the zinc--rich coatings was investigated by means of electrochemical impedance spectroscopy (EIS) and corrosion potential monitoring．The degradation mechanism of the zinc—rich coatings was elucidated．It was shown that effective barrier can not be formed in the coatings with low zinc content. The attack of zinc particles in zinc--rich coatings is explained in terms of diffusion control through the film of zinc corrosion products around the zinc particles. The surface contamination of the substrate is vital to the performance of the organic coatings protecting the substrate steel. The surface contamination will lead to not only the impairment of the performance of the zinc—rich coatings but also the acceleration of the degradation processes of the steel substrate.

The experiment results show that the combustion products is TiC and very small amount of Ti3AlC2 if using the stoichiometric ratio of formula Ti3AlC2 and Ti, Al and C powders as raw materials. However, Ti3AlC2 become the main phase in the final product as adding intermetallic compound TiAl3 in the raw mixtures without changing overall stoichiometry. The amount of Ti3AlC2 is increased with increasing the amount of TiAl3 in the initial mixtures. The above experimental phenomena were alsoexplained through thermodynamic and kinetic analysis.