Investigations of microstructural evolution in undercooled metallic melts have been extensively conducted in different alloy systems, such as eutectic, peritectic and metallic compound, however, few reports about undercooled superalloys were found up to date.  In this paper, DZ125 superalloy melt was undercooled by the method of molten salt purification combined with cycled superheating under Ar atmosphere, and a substantial undercooling up to 180 K was obtained. Based on the microstructural observation as well as the calculations of the dendritic tip radius, the dendrite growth velocity and undercooling with BCT (Boettinger, Coriell, Trivedi) dendrite growth model, the solidification behavior and structure transition mechanisms were systematically investigated. The results indicate that the microstructure undergoes three sequent transitions in the achieved range of undercooling range ΔT. When ΔT<48 K, the dendrite growth is mainly controlled by solute diffusion, and the solidification structures are the common dendrites. When 48 K≤ΔT< 85 K, the dendrites is remelted and the first kind of granular grains forms because of the serious remelting produced by recalescence. When 85≤ΔT<160 K, the effect of solute diffusion is weakened by solute trapping that results from high dendrite growth velocity, thermal diffusion plays the dominate role at the same time, the remelting effect decreases to a low level, and ripenings of dendrite are restrained, the granular grains turn into highly developed fine dendrites. WhenΔT>160 K, the second kind of granular grains is formed, the transition is caused by the stress originating from the rapid solidification contraction, which results in the recrystallization of the fine dendrite.

The effects of aging temperature on the microstructure and mechanical property of ultra–low–carbon acicular ferrite steel was studied by utilizing OM, SEM, TEM, tensile and impact tests. The strength of the specimen is not obviously changed due to little precipitating amount of the second phase and little coarsening of microstructure when aged from 450 to 500 ℃. The rapid increase in strength appears and is attributed to the dispersive precipitations of  ε–Cu and Nb(C, N) when aged from 550 to 650℃. When aged above 650 ℃, the abrupt decrease in strength appears and is caused by over–aging, i.e., the coalescence of ferrite plates and coarsening of precipitates. When aged at 750℃ (in two–phase field), the toughness decreases abruptly, which is related to the formation of martensite/austenite (M/A) ithigher carbon content during cooling.

The internal friction behaviors of B2 Fe--Al alloys were examined to understand the correlation between the ordering process and the atomic movement. An internal friction peak was found at around 570 ℃ in the oil--cooled Fe₄₇Al₅₃ alloy, while the peak in the furnace--cooled specimens disappeared. The peak has the typical phase transition character. Combined with analyses of XRD and DSC, it is proposed that the peak originates from the ordering process, which is caused by the jump of antisite atoms to their own sublattice sites through the nearest neighboring vacancies. Similar internal peak was observed also at 540℃ for Fe₅₇Al₄₃ alloy.

Dislocation and monovacancy (V₁) are the fundamental defects in Si. The interaction between them is concerned with the electronic and optical properties of electronic devices. In this paper, the interactions of 30^o partial dislocation with V₁ in Si were investigated by the molecular dynamics simulation method based on the Stillinger–Weber (SW) potential. The simulations were conducted under different temperature and shear stress conditions. The results show that 30^o partial dislocation is pinned when dislocation encounters the V₁ under the conditions that shear stress is relatively low and the temperature is kept constant. When the shear stress increases to a critical value, the dislocation can overcome the pin and the V₁ is left in the crystal. As the temperature increases, the critical shear stress decreases approximately as a linear function. Moreover, the values of the critical shear stress corresponding to different dislocation kinks also show that the abilites of kinks to overcome the pin are determinded by the migration barriers of kinks. In comparison of the locations of dislocation core in two models with and without V₁, it is found that the V₁ can make the 30^o partial dislocation move faster once the dislocation glides away from the V_{1 .}

By using Sn–Pd catalyst system, the electroless plating Cu on the surface of Mo powder was performed to fabricate Cu/Mo composite powder. Composition, morphology and formation process of Cu/Mo particles were analyzed by XRD, SEM/EDS and XPS. The formation mechanism of Cu/Mo particles can be described as follows: the PdCl₂ (activator) deposited on the surface of Mo particles firstly and then was reduced by SnCl₂ (sensitizer) to nano Pd particles, which are the nucleation sites for Cu depositionfinally Cu coating formed.

Plastic Zr_58.5Ti_14.3Nb_5.2Cu_6.1Ni_4.9Be_11.0 bulk metallic glass matrix composites containing uniformly distributed dendrites in the glass matrix were synthesized by the Bridgman solidification methold. Through tailoring the withdrawal velocity, the volume fraction of dendrites with characteristic spanning length, as well as the mechanical properties of the samples can change. The characteristic spanning length of the individual dendrites roughly obeys linear relationship with the withdrawal velocity. The compressive ultimate strength and the fracture strain of the sample reached 1930 MPa and 11.3%, respectively, when the withdrawal velocity was 1.0 mm/s.

A multidimensional numerical model was developed to investigate the temperature field, fluid field of liquid phase in the molten pool, and microstructure evolution in the plasma deposition manufacturing (PDM) process. A level--set approach was used to track the evolution of free surface of the molten pool, and an enthalpy--porosity model was introduced to deal with the transformation of solid and liquid phases. To understand the physical mechanism of thermal impact on the microstructure of the deposited layer, a Monte Carlo method combined with thermal--fluid analysis was applied to track the grain growth process in the PDM process. A numerical experiment of nickel--based alloy thin wall parts by PDM was implemented. The numerical results show that the microstructure of the deposited layer mainly depends on frequency  and amplitude of thermal impact, which is also influenced by variable processing parameters such as plasma power, scanning speed, and powder feed rate. Therefore, under full melting of fed powder, an increase of scanning speed could make the grain size of final microstructure finer to some extent.

Mechanical behaviors of miniature solder joint interconnections with different scale matches, diameter (d) in the range of 200—575 μm and length (l) of 75—525 μm, were investigated by using quasi–static micro–tensile testing. The results show that the joint geometry scaling factor (d/l) plays an important role in influencing the mechanical constraint level and tensile strength of the solder joints. However, with increasing d/l, the tensile strength of the joints does not always increase and not always exhibit a similar trend to that predicted by Orowan approximation equation, but there is an inverse size effect on the solder joint strength. For both lead–free and lead–containing solders, the correlation between the tensile strength and volume of the solder joints largely follows inverse proportion function equation, i.e., σ_F−Joint = 1/(Ad²l) + B, where tensile strength of the solder joints increases obviously with the decrease of the solder joint volume and there is a so–called "solder joint volume effect", that is, the smaller the solder joint volume, the higher the solder joit strenth.

Since ultrasonic fatigue test has been used to study the very high cycle fatigue (10⁷—10⁹ cyc), the difference between ultrasonic and conventional fatigue test methods should be evaluated in order to ensure the validity of ultrasonic fatigue result. By comparing some results of other researchers, it is found that the frequency effect is negligible, and the loading condition is the main reason for the difference. A comparison of the fatigue strengths of the 54SiCr6 high strength spring steel under three kinds of cyclic loading conditions, rotating bending (RB), tension compression (TC) and ultrasonic (UL), was reported. The results reveal that the three kinds of fatigue specimens display different fracture features, and the fatigue strength of RB is the highest, TC is the lowest, and UL is somewhere in between. The difference in the fatigue strengths is mainly attributed to the distinctions of stress gradient and the size of specimens. By taking highly stressed cross–section area (HSCA) into consideration, a relationship of the fatigue strength and loading condition was proposed, and the two constants σ_lim,0 and σ_ΞA in the equation of HSCA are mainly dependent on material strength and inclusion size, respectively. A relationship of fatigue strengths between RB and TC is also discussed specifically.

Ti–2Al–2.5Zr alloy with hcp structure is a kind of structural materials used under low temperature condition, e.g. pipeline system of liquid hydrogen and liquid oxygen in missile engine. It is usually serviced in condition of severe low temperature and dynamic loading. Deformation twinning is a common and important plastic deformation mode in the hexagonal close–packed alloy, but will be severely restricted as the grain is refined from tens of microns to a few microns. On the other hand, twinning has a low sensitivity to temperature, consequently becomes a favorable deformation mode at low temperature in comparison with dislocation slip. The objective of this work is to study the coupling effect of grain refinement and testing temperature on twinning behavior and the low–cycle fatigue behavior of Ti–2Al–2.5Zr. Symmetrical push–pull low–cycle fatigue (LCF) tests were performed on Ti–2Al–2.5Zr with grain size of about 5 μm at room temperature (RT) and low temperature (77 K). The results show that the alloy exhibits the higher ductility and the longer low–cycle fatigue life at 77 K than those at RT. The cyclic stress response curves show that the cyclic softening occurs at the low strain amplitudes of 0.5% and 1.0%. However, as strain amplitude increased to 1.5% and 2.0%, cyclic stress saturation appeared. When testing temperature declined to 77 K, the cyclic hardening was observed at all four strain amplitudes. The degree of cyclic  ardening increases as the strain amplitude rises. The fractography analyses suggest that transgranular fracture with well–developed fatigue striations is the predominant failure mode. The amount of secondary cracks is much higher in the alloys deformed at RT than that at 77 K. TEM examination reveals that deformation twins become more active. The primary types of twinning are {1011} and {1121}. The typical deformation microstructures consist of individual dislocation lines together with the tangled dislocation at the low strain amplitudes of 0.5% and 1.0%. As the strain amplitude increased to 1.5% and 2.0%, {1010}prismatic slip and {1121} pyramidal slip were simultaneously activated, the subgrain and dislocation cells were formed. At 77 K and strain amplitude of 2.0%, the parallel dislocation bands distribute along prismatic plane. As the strain amplitude increased to 0.5%, utual perpendicular dislocation lines appeared. The improvement of fatigue life of Ti–2Al–2.5Zr at 77 K is attributed to the constraint of inhomogeneous slip and the activation of deformation twinning.

The weatherability of weathering steel depends on the protection of the compact rust layers on its surface. However, the compact rust layers might be damaged when they were subject to external actions, for example, collision and scrape etc., which will influence the protective effect on steel base. It is deserved to put attention on further corrosion behavior of the samples with damaged rust layer, which will determine the long term corrosion property. In the present study, electrochemical measurement, metallographic observation and energy disperse spectrum analysis were employed to investigate the further corrosion behaviors of a low carbon bainitic steel in the environment containing Cl⁻, after its original rust layers had been damaged on different degrees. It was found that the damnification of rust layers on both the low carbon bainitic steel and low carbon steel as a contrast could be rapidly self–repaired in the further corrosion process. When damnification degree and further corrosion time are the same, the resistance and the repair ratio of damaged rust layers on the low carbon bainitic steel are higher than those of the low carbon steel. The fracture toughness of the steel base/rust layer interface of the low carbon bainitic steel is higher than that of the rust layers, so the rust layers would not be abscised thoroughly from the interface and some residual rust would be remained when the steel is subjected to applied actions. The formation of new rust at the damaged sites can be significantly promoted by the residual rust. The contents of Cu and Cr in the original rust layers are close to those in the newly formed rust layers, and both of them are equivalent to those in the steel base. These results indicate that low carbon bainitic steel with excellent mechanical properties and weld ability is a potential candidate for novel weathering steels with higher strength.

The compressive deformation of the Zr₅₅Cu₃₀Al₁₀Ni₅ bulk metallic glass (BMG) was investigated in the supercooled region. It was found that the deformation behavior of the BMG alloy is strongly dependant on strain rate and temperature. The BMG exhibits Newtonian behavior at low strain rate but becomes non–Newtonian flow at high strain rate. The correlation between flow stress as well as strain rate and temperature was established in terms of stretched exponential function.

The electromechanical coupling coefficient (K₃₃) is a crucial parameter to express the efficiency of the magnetostrictive alloy. Fe_72.5Ga_27.5 magnetostrictive oriented crystal has been prepared by zone melting unidirectional solidification. The K₃₃ values of the directionally solidified sample and its quenched sample(1000℃/3 h, W. Q.) were determined by improved AC impedance resonance frequency testing method. The DC magnetic field was formed by the DC coil and the pre-pressure was given by the press-pack. Since the magnetostriction of Fe–Ga Alloy is less than that of the giant magnetostrictive material of Tb–Dy–Fe by 10 times, the K₃₃ of Fe–Ga alloy cannot be determined by the traditional testing methods. In this paper, a 4–wire configuration connection was used to pickup the signal in experiment, and the AC excitation signal was adjusted to 0.5 A, so that the obvious and stable AC impedance curves of the Fe–Ga alloy were acquired. When the DC magnetic field is 32.7 mT and no compressive pre–stress is applied, the K₃₃ values of the directionally solidified sample and quenched sample are 0.103 and 0.137, respectively. The K₃₃ values of the two kinds of Fe_72.5Ga_27.5 magnetostrictive alloys decrease with increasing DC magnetic field, and increase at first then decrease with increasing pre–pressure, but the K33 value of the quenched sample always higher than that of the oriented sample.

Nanocomposites Co_1-xNi_xFe₂O₄/SiO₂   were prepared by sol–gel method using tetraethyl orthosilicate and nitrates, in which SiO₂ is used to restrict the growth of ferrite crystalline and remove the agglomeration. XRD and TEM are used to characterize structures and morphologies. The composites sintered at 900 ℃consist of spinel phase with size ranging from 15 to 20 nm and amorphous SiO₂. The doping of Ni²⁺ reduces the volume of unit cell of the spinel phase. The magnetic properties of the Co_1-xNi_xFe₂O₄/SiO₂   nanocomposites were measured by vibrating sample magnetometer and Mossbauer spectroscopy. The results indicate that the saturation magnetization decreases from 41.8 A·m²/kg to 23.0 A·m²/kg with increasing Ni²⁺ content. At the same time the coercivity decreases from 104.2 kA/m to 5.4 kA/m, which may originate from the variation of anisotropy induced by Co²⁺ ions in octahedral sites. All the samples are magnetic order state, the hyperfine magnetic field of the nanocomposites decreases with increasing Ni²⁺ content.

In the steel industry, the continuous casting product is usually a pre–produt before being fed to hot rolling mill, which brings in high cost and ollution and low productivity. So a new technology is demanded urgently to produce surface defect free billets and slabs. The success of electromagnetic casting (EMC) technology utilized in aluminum casting to improve surface quality enlightens the metallurgical researchers to apply this novel technology into production of steel. An experiment using high frequency magnetic field (26 kHz) was carried out on a vertical type caster with a segmented copper mold in our laboratory. Carbon structural steel (0.20% C, mass fraction) was used as experimental material. A special power was used to provide high frequency current to an induction coil surrounding the segmented mold. Moreover, the distribution of magnetic field and meniscus shape of Sn–Pb–Bi alloy in the mold were investigated and used to explain the phenomena happened in the initial solidification area in the case of EMC. The result shows that oscillations marks (OMs) can be totally suppressed and the surface of round billets with EMC technology becomes extremely smooth compared to the billets without EMC technology when the output power comes to an optimal one. According to the conventional mechanism of OM formation, the pressure generated in the flux channel due to the relative movement between solid flux rim and molten steel pool plays an important role in the formation of OMs. In the case of EMC, the solid flux rim becomes smaller as temperatures of mold and liquid steel in the initial solidification area increase because of Joule heat generated by the high frequency magnetic field, and the flux channel turns wider as Lorentz force is imposed on the liquid steel. As a result, the flux pressure becomes smaller, which makes the billets have better surface quality. However, when the powder exceeds the optimal value, the distribution of initial solidification starting point along the mold perimeter behaves in a wavy pattern. Simultaneously, the free surface of molten steel fluctuates more greatly, resulting in formation of OMs. As a conclusion, the formations of wavy OMs on the round billet surface are resulted from non–homogeneity of magnetic field between slit center and segment center as well as fluctuation of liquid steel level.

DSC and dilatometry were carried out to determine the phase transformation temperature of a high speed steel used for centrifugal cast rolls. Based on the critical transformation temperature, high speed steel was annealed, quenched and tempered in order. The results show that the as–cast high speed steel softens only when annealed at above 630℃ as to be easily machined. Before and after heat treatment, there are only little ifferences between the components, morphologies and contents of MC eutectic carbide. However, more and more metastable M2C eutectic carbides decompose with the increase of quenching temperature; the tempering temperature affects much onthe final hardness of the high speed steel. A relationship between tempering temperature and the secondary hardening is given and the peak temperature of the secondary hardening is determined. And the effects of resolving and precipitating of secondary carbides on the microstructure and properties were also studied.

Mathematical model was developed to study the 3D temperature distribution and heat transfer from superheated liquid steel to the inside of the solidifying shell in the slab continuous casting mold. The effects of some factors, such as submergence depth and port angle of submerged entry nozzle (SEN), mold width, casting speed, superheat temperature, argon gas injection, electromagnetic brake (EMBr) and also including the argon gas flow rate and current intensity etc., on the temperature distribution and heat transfer of superheated liquid steel in the mold were analyzed. The results indicate that the maximum heat input to the solidifying shell forefront occurs near the impingement point of liquid steel on the narrow face of mold, and the most superheat of superheated liquid steel is dissipated near the impingement zone. Heat flux of superheated liquid steel delivered to the shell  surface increases in direct proportion to the casting speed and superheat temperature, respectively. Argon gas injection leads to a substantial increase in superheat flux to the impingement zone of narrow face and the upper region of wide face. EMBr is beneficial in increasing the temperature of upper region of the mold, but has no obvious effect on the heat flux distribution. The double action of argon gas injection and EMBr also produces an increase in heat flux to the upper region of wide face, which has no visible influence for the hat flux distribution of impingement zone.

Liquid flux is an effective lubrication in the conventional continuous casting of steel, which can prevent the breakout effectively, and also the longitudinal cracks that occur on strand surface are decreased obviously. Previously, many studies have been reported on the flux infiltration, some empirical equations for calculating flux consumption have been reported based on the data accumulated through the commercial operation of casters, and the new tools and techniques for estimating the lubrication condition in mold have also been introduced. Especially, the lubrication mechanism of liquid flux has drawn general concerns recently, and some relevant simulation of flux infiltration behavior based on the cold model experiments and mathematical models of flux infiltration derived from theoretical calculations have been conducted. Most of these researches lay particular emphasis on the macroscopical detection and calculation to the infiltration behavior, and some researches related to micro–mechanism lack the discussion of the relevant influence factor yet. In the present work, on the basis of calculation of liquid flux channel pressure in meniscus for slab continuous casting mold, a new mechanism of the liquid flux consumption was proposed by analyzing the deformation behavior of initial solidifying shell during the mold oscillation cycle as high casting speed 2.0 m/min, and the concepts of infiltration time and infiltration intensity were defined for the first time, then the effect of non–sinusoidal oscillation parameters on the liquid flux consumption was discussed. The results show that the periodically continued liquid flux consumption is caused by variety of flux channel pressure which is induced by change of channel width, and the liquid flux will be infiltrated into flux channel from last stage of positive strip time until last stage of negative strip time by negative flux channel pressure. The infiltration time is lengthened and the infiltration intensity is weakened by reducing oscillation frequency. The infiltration intensity is strengthened by improving amplitude, and the infiltration time is slimly influenced. Non–sinusoidal oscillation factor has a little effect on the infiltration intensity, and the infiltration time is increased with the oscillation factor decreasing.

Compared to fusion welding processes used in joining structural aluminum alloy, the friction stir welding (FSW) is an emerging solid state welding process which can lead to less distortion and residual stress owing to low heat input characteristic. In order to apply FSW in the aerospace field, the 2219 aluminum alloy in Al–Cu series, which has high–strength–weight ratio, resistance to stress corrosion cracking and superior cryogenic properties, and is an ideal material in the construction of liquid cryogenic rocket fuel tanks, has been welded in O condition (full annealing). Microstructures of weld nugget zones and mechanical properties of the slices in the three kinds of thick plate FSW joints, which are the flaw–free joint, the joint with flaw and the flaw–free joint after post weld heat treatment (PWHT), were studied. The tensile testing shows that the tensile strength σ_b and yield strength σ_0.2 decrease from 160.8 and 96.8 MPa of the top part of the flaw free joint to 146 and 86 MPa of the bottom part, respectively, σ_b and σ_0.2 decrease from 132.9 and 94 MPa of the top part of the joint with flaw to 126 and 78.8 MPa of the bottom part, respectively. σ_b of the middle of the flaw–free joint after PWHT is 243.8 MPa and σ_0.2 of the top of the flaw–free joint after PWHT is 123.3 MPa, they are higher than those of the other parts.  increases from 6.7%, 4.8% and 7.5% of the top parts of the flaw–free joint, joint with flaw and flaw–free joint after PWHT to 10.1%, 8.5% and 14% of the bottom parts of the joints, respectively. For the PWHT (500 ℃/50 min + 160℃/20 h) flaw–free joint, grains are more uniform and finer along the thickness direction in weld nugget zone, fractographs present deep dimples, most of the failure is ductile fracture, and the microhardness distribution has no obvious fluctuation. For the joint with flaw the mechanical properties decrease sharply, fractograph exhibits a ductile–brittle fracture morphology and the microhardnesses in slices are lower than those of the flaw–free joints.

The particle size of raw powders to prepare the cemented carbides plays a significant role not only in the properties of the resultant bulk material, but also in the sintering process. WC/Co composite powders with various size matches were prepared by spark plasma sintering (SPS), and the variations of shrinkage rate, which is an important factor describing SPS process, with the sintering temperature and the relationship between the shrinkage rate and the relative density of the sintered sample were analyzed. The results show that the temperature of shrinkage starting, the temperature corresponding to the peak of the shrinkage rate, and the temperature of full densification are almost the same for different sized WC particles, however, all of the above temperatures reduce with the decrease of the Co particle size. It indicates that the shrinkage process of SPS has a rather very weak relationship with the WC particle size, whereas has a close relationship with the Co particle size. As compared with the conventional sintering methods, the above temperatures and the relative density corresponding to the maximum of the shrinkage rate are all lowered. It implies that in the conventional process, at the stage with the maximum of the shrinkage rate, most of the shrinkage has been accomplished. However, in the SPS process, at the same stage the shrinkage degree is lower. Finally, he secial SPS mechanisms for comosite powders with various size matces were analyzed.

To protect carbon/carbon (C/C) composites against oxidation at high temperature, SiC inner layer and SiC–MoSi₂ multi–layer outer coatings were prepared on the surface of C/C composites by a two–step technique of pack cementation and slurry coating. The microstructure of the as–received coating was analyzed by XRD and SEM. The oxidation behaviors of the coated samples at high temperatures in air were studied by isothermal oxidation tests. The results show that the as–received multi–layer SiC/SiC–MoSi₂ coating posseses a crack–free structure and excellent oxidation resistance, which could protect C/C composites from oxidation in air for more than 110 h at 1500 ℃and for more than 100 h at 900 ℃Excellent anti–oxidation ability of the SiC/SiC–MoSi₂ coating is attributed to its multi–layer and multi–phase structure and the formation of protective SiO₂ film on its surface during oxidizing.

Because of the solidifying characteristic of alloy and the difference in thermal conductivity of mould, the microstructures of different zones of semi–solid slurry are different, which is hard to satisfy the requirement of rheoforming. A new process preparing semi–solid slurry is proposed, in which a locally rapid cooling near the center zone of mould can be realized by inserting a copper rod into the slurry and then pulling out repeatedly, except application of a weakly electromagnetic stirring. Semi–solid A356 alloy slurry was prepared by the new process and the effects of the new process on morphology and size of primary α–Al in semi–solid A356 were researched. The results indicate that the nucleation rate, morphology and grain size of primary α–Al are markedly improved by the new process. Primary α–Al with small size presents particle–like or globular–like and distributes uniformly in the slurry even at lower overheating. The grain refining and structural homogeneity are relative to homogenization of the temperature field in the mould.