Cyclical deformation in symmetrical push-pull mode was carriedout on a Cr-Mn-N stainless steel at room temperature in air using a Schenck Hy-dropuls machine. The total strain amplitude which was kept constant during thetest ranges from ±0.004 to ±0.012. The 0.2% offset yield stress σ_(0.2f) in tensile andσ_(0.2r) in compressive and peak stress were measured after various cycles. The Bausch-inger strength differential factor(BSDF) was then calculated from σ_(0.2f) and σ_(0.2r). TheΔE was also calculated from hysteresis loop areas for various cycles. These parame-ters, BSDF, σ and ΔE appear to have two distinctively different stages. The dislocation structures were observed using TEM in specimens deformed cy-clically for various cycles. The observation shows that the dislocations pile-up main-ly at grain boundaries and there exists large amount of deformation twins. Thenitrogen content of 0.25 wt-% places the alloy in the low stacking fault energyregime. Cross-slip and climb are thereform rather difficult to occur during the cy-clic deformation at room temperature, and well-defined dislocation cells and wallscan only be seen at the final stage of fatigue.

The correlation between the shape memory effect (SME) and theordering degree of martensite formed during various heat treatment processes, e. g.ice water quenching, step-quenching and quenching plus aging, has been studied ina Cu-Zn-Al alloy. The ordering degree of martensite is characterized by the spacingdifference, Δd, of some pairs of diffracting planes, the indices of which satisfy theequation (h_1~2-h_2~2)/3=(k_2~2-k_1~2)/n. It is shown that the SME depends on Δd monoto-nically. The ordering extent of the DO_3 parent phase and of the martensite in turnincreases with the holding duration in the course of step-quenching, which may bethe main factor for enhancing the shape recovery rate η. The stabilization of themartensite may be associated with the clustering of vacancies which results in thelowering of the ordering degree. The decrease of the SME upon aging at highertemperature or holding at step-quenching temperature for longer time is due to theaccumulation of solute atoms and the precipitation of α-phase, which lowers theordering extent.

The substructural development of martensite plates at different stageof the transformation for an Fe-21Ni-4Mn alloy has been investigated by usingtransmission electron microscope. Twinning was found in small thin platelets atthe initial stage. Fine twins extend to the whole platelet. The dislocations and twinswere found in the thick platelets at a later stage. Bending of the twins was alsoobserved frequently at this stage. Finally, several platelets coalesce with each otherto forma macroscopic (252)_f martensite "plate". The original habit plane of theindividual platelet was measured to be close to (121)_f.

The transformation process and the products of long term isothermaltransformation around the kinetic Bs temperature (the bay on TTT diagram) in Fe-C-Mo alloys have been investigated. It was found that, at the final stage of inter-mediate transformation, a sort of granular structure might form accompanying withthe growth of bainitic ferrite. In comparision with the carbide Mo_2C in the bainiticferrite plates and blockes, the carbide in granular eutectoid has been determined tobe (Fe, Mo)_3C type.

The precipitation behaviour of TiC in the Ti-3.84Al-6.24Mo-10.43V-1.85Fe alloy has been investigated. TiC layer was found to exist at the α/βinterface boundaries in the sample solution treated just below β transus and quenched.The morphology of TiC is similar to that of the interface Phase (IFP) TiH_2 inthe α-β two phase alloy. The IFP TiC provides a site for crack initiation as wellas an easier path for crack propagation. The morphology of fracture surface is alsosimilar to that caused by TiH_2. Arch diffraction patterns were found in some speci-mens after solution treatment and isothermal aging. The structure and lattice para-meters were analyzed and it was shown that the particles which bring about thearch diffraction might be TiC, instead of the "type 2" α phase.

The thermal fatigue behaviour and microstructure in a direction-ally solidified DZ38G alloy have been investigated. It was found that DZ38G alloyhas larger thermal fatigue resistance than that of original M38G alloy because of theelimination of the transverse grain boundaries and the increasing of elastic modulus.Detailed observations show that the microcrack is initiated at the grain boundariesor the phase interface and propagates along interdendrites, grain boundaries and,sometimes, the defined crystallographic directions. After thermal fatigue test, theamount of carbides at grain boundaries increases and the morphology of γ′ phasebecome irregular. Both γ′denudation and recrystallization have been observed onboth sides of the crack.

The strain fatigue, impact fatigue and rotation beam fatigue behavi-our of granular bainitic structure has been studied. The results show that the strainfatigue properties and the impact fatigue properties of granular bainite are su-perior to that of tempered martensite under the condition that the ultimate tensilestrength is equal. The impact fatigue life increases with increasing amount of gra-nular bainite, because the M-A islands might retard the propagation of fatiguecrack. The rotation beam fatigue properties of granular bainite are similar to thatof tempered martensite. The relationship between fatigue limit S_f, yield strength σ_yand fracture strength S_k may be expressed as S_f=4.651+0.1411 (σ_y+S_k)

The fatigue-creep interaction behaviour of superalloy GH698 with dif-ferent grian sizes has been studied at 700℃. The results show that three zones,F, C and FC, can be distinguished at the fatigue-creep interaction curves. In zonesFC and C, the presence of alternative stresses inhibits the creep deformation pro-cess, therefore delays the fracture of the alloy. This effect will be enhanced bygrain refining. However, in zone F, grain refining exhibits an inverse effect un-der large alternative stresses.

Rupture and tensile properties of K17 before and after thermal expo-sure in vacuum, air and hot corrosion environments have been investigated. Theresults show that vacuum exposure essentially does not affect the rupture lives, butit results in the decrease of high temperature tensile strength and increases of rup-ture and high temperature tensile ductilities; air or hot corrosion environmental ex-posure make the alloy degradation: severely reduce the rupture and high temperaturetensile properties, hot corrosion environmental exposure being more harmful. Theformer causes grain boundary oxidation and depletion of Cr, and the later makesthe sulphurization at grain boundaries. That results in the concentration of S anddepletion of Cr at grain boundaries and, consequently, weakens the grain bounda-ries.

Fracture behaviour of low-all oysteels AISI4340, HY100, WELTEN60, AISIA537 and A131 in artificial seawater under static, cyclic loading and cath-odic protection potential has been investigated by using fracture mechanics, elec-trochemistry and electronfractography. The results reveal that under hydrogen evo-lution potential (cathodic protection potential) the critical yield strength requiredfor the occurrence of SCC decreases from 985 to 872MPa comparing with corrosionpotential (E_c). The effect of cathodic protection on crack propagation of CF is notsimple, it is closely related to the yield strength of steels and their SCC behaviour.For the steels with high yield strength, cathodic protection promotes da/dN

The inclusion parameters, fracture surface morphology and voidgrowth characteristics of tensile and fracture toughness test specimens of 2.25Cr-1Mo steels with and without rare earth (RE) additions have been investigated byusing Quantimet (QTM) and SEM techniques. The results indicate that the lowerfracture toughness of the steel with RE at room temperature may be ascribed to itslarge content of RE-containing inclusions.

The phase boundary of α_2 precipitation at 600℃ in Ti-Al-La ternary systemhas been determined by TEM examination, which shows three kinds of phase regions, namelythe single α phase region, the transitional region where there is superlattice reflection but no α_2precipitation and the α+α_2 region. A point of maximum aluminium content was found to exist onthe α/α+a_2 boundary and the track of the boundary was explained in terms of the internal oxida-tion of lanthanum and the electron concentration rule for α_2 formation

The correlation between the diverse fracture modes and the microstructural va-riation with recrystallization in the Al-Li alloys under heat treatment was observed by means oftriplanar optical and scanning electron microscopy. The growth, propagation and feature of theirfractures were described from the viewpoint of microstructure, energy and geometry.

The step pattern of single crystal growth and the morphology at equilbriiumstate of (Mn, Fe)S on the wall of micro-voids in ZG25 cast steel have been observed using scan-ning electron microscope. The face-centred cubic (Mn, Fe)S single crystal at equilibrium stateis shown to be tetrakaidecahedron consisted of eight {111} planes and six {100} planes, and isa typical example of the O_h-m3m cubic crystal system.

The activity of CaO and La_2O_3 in the liquid ternary slag CaO-SiO-La_2O_3 has been determined by equilibrating Sn with the slag phase in a graphitecrucible under 1 atm of CO at 1600℃. The experimental results showed that theactivity of CaO and La_2O_3 changes slowly while the ratio of mole fraction, x_(CaO)/x_(SiO_2), is less than unit, but varies sharply while the ratio increases from 1 to 1.2.In addition, by equilibrating Sn with the binary slag B_2O_3-La_2O_3 saturated with solidLa_2O_3 at three different temperatures, the relation between lg γ°_(La) and 1/T isdetermined as follows: lg γ°_(La)=-29400/T+11.37(1500-1600℃) A random network model of molten silicate structure was applied to CaO-SiO-La_2O_3. It was found that the experimentally determined activity-composition curvesof CaO are qualitatively reproduced by the curves calculated with the model, butthe curves of La_2O_3 are considerably different from the calculated ones. It impliesthat the model used is over-simplified and that the difference in interaction energybetween the metal cations of different valence and silicate anions should be consid-ered.

A criterion of the metastable extension of solid solubility was propo-sed. When the eutectic concentration arm ratio parameter of an alloy, J>10, thehypereutectic extension and the metastable solid solubity limit beyond eutecticconcentration may be obtained. When J≤10, the hypoeutectic extension may be obtained. The extension parameter, C_(eq)~s/C_(eu), is directly proportional to the parameter J.Thus, it seems easy to predict the tendency of the solid solubility extension underrapid solidification on the basis of the eutectic phase diagram.

The phase diagram of system NdCl_3-LiCl-KCl has been investigated byDTA. Six surfaces corresponding to the primary crystallization of NdCl_3, LiCl, KCl,γ-KCl·NdCl_3, β-3KCl·NdCl_3, and 2KCl·NdCl_3 respectively, eleven univariant linesrelated to the secondary crystallization, two ternary eutectics E_1 (70.5wt-% NdCl_3+8.0wt-%3KCl, 360℃) and E_2 (5.0wt-%NdCl_3+53.0wt-%KCl, 355℃) and one ternaryperitectic P (67.0wt--%NdCl_3+10.0wt-% KCl, 375℃) are found. It is also clarifiedthat this study on the system NdCl_3-KCl differs from previous work in the twotransition temperatures: e_1 (77.0wt-%NdCl_3, 484℃), e_2 (40.0wt-%NdCl_3, 620℃), p(70.0wt-%NdCl_3, 610℃), 3KCl·NdCl_3 (690℃) and two reactions: α-3KCl·NdCl_3?β-3KCl·NdCl_3(355℃) β-3KCl·NdCl_3?γ-3KCl·NdCl_3(450℃)The eutectic point e_5 (31.0wt-%LiCl, 460℃) is determined for the pseudobinary sys-tem 3KCl·NdCl_3-LiCl.

The mathematical models of the kinetics of the layer growth at differ-ent ion nitriding conditions of armco iron, steels 45, 40Cr, 42CrMo and 38CrMoAlhave been established. Based on these models the expression of nitrogen concentra-tion profile of ion nitrided layers have been deduced with profile simulating method.They are C_β=C_(min)~ε+P~4/ξ_(10)(ξ_1-x)+P~5/ξ_(10)~2(ξ_1-x)~2 C_γ~′=C_(min)~γ′+P_1/ξ_(21)(ξ_2-x)+P_2/ξ_(21)~2(ξ_2-x)~2 C_α=C_(min)~α+C_(33)/ξ_(32)~3(ξ_3-x)~3Using these models, the kinetics of layer growth and the nitrogen concentration pro-file of ion nitrided layers were simulated on Apple-Ⅱ computer. Results show thatthe simulated curves coincide quite well with the experimental data.

Study on the temperature variation of domain revealed that the maindomains remain the same without change during heating, but the supplementary do-mains, such as the cone like reversal domains in the surface, would be changedboth in the size and shape, or even disappeared, while temperature is higher than240℃. Since the nucleation sites and the spontaneous orientation of nucleated do-main has some randomness, so the domain patterns should be changed every timeafter the thermal demagnetization higher than Curie point.

The diffusion process of Sn in the transformation of Nb_6Sn_5 to Nb_3Snhas been studied. The experimentel results show that the growth of Nb_3Sn layer iscontrolled by two processes, i. e. the short range diffusion of Sn which is respon-sible for the inner layer with large grains, and the long range diffusion of Sn whichresults in the outer layer with fine grains. It was found that grain size depends onthe reaction temperature. A model of Nb_3Sn growth based on the above processeswas established and its solution was found to be in good agreement with experimen-tal results.

Diffusion reactions of superconducting composite NbTi/Cu, diffusioncouple Ti/Cu and Nb/Cu have been studied by means of EPMA. The results indica-te that the initial interface state of the composite NbTi/Cu significantly affect theforming process of intermetallic compounds at the interface, In comparision with themetallurgical binding composite, the intermetallic compounds forms at a higher tem-perature in mechanical binding one. Therefore, the mechanical binding process isbeneficial to the production of NbTi/Cu superconducting composite. No intermetallic compounds have been observed in Nb/Cu system. The morphol-ogy of intermetallic phases and the sequence in which they form in Ti/Cu systemare somewhat different to that in NbTi/Cu. The relationship between the thicknessof compound layer and anealing time obeys the rule of y~(1.5)--t.

The effects of material inhomogenity and weld width of the over-matched welded joints with a longitudinal crack or flaw upon the J-integral, a charac-terizing parameter for elastoplastic fracture, have been studied. The overmatchedweld is idealized to be a parallelly cracked and sandwiched hard layer, the yieldstress of which is higher than that of the base metal. Numerical solutions were ob-tained by using elastoplastic finite element method under the assumption of plane stress.The results show that the values of J-integral are remarkably affected by the mate-rial inhomogenity and the weld width especially when the nominal stress or nominalstrain is comparatively high, which, therefore, should be taken into account whenevaluating the crack driving force for a overmatched joint.

The amorphous Fe_(90-x)V_xZr_(10) alloy after electrolytic hydrogenation pro-cedure was studied by means of Mossbauer effect, magnetic measuement, X-ray dif-fraction and second ion mass spectroscopy (SIMS) techniques. The experimental re-sults show that the magnetic properties of amorphous Fe_(90-x)V_xZr_(10) alloy are highlysensitive to presence of hydrogen. The isomer shift (IS) has a small positive changeand the magnetic properties transfer from paramagnetic to ferromagnetic state byhydrogenation. The effect of hydrogen absorption is also to raise Fe magnetic mo-ment and Curie temperature T_C from 257 to 384 K. Finally, the considerable in-crease of magnetic properties of amorphous Fe_(90-x)V_xZr_(10) alloy by hydrogenation is at-tributable to the change of hydridization effect between outer electrons of Zr atomsand 3d-electrons of Fe atoms. It may be concluded that the hydrogen atoms is loca-lized around Zr atoms when hydrogen atoms enter inside of alloy.

The effects of grain shape on the accommodation velocity of grainboundary sliding, which is the dominant mechanism in superplastic deformation, onthe contribution of each mechanism to the total strain and the region transitionstrain rate have been discussed. A new concept, critical aspect ratio for superplas-ticity, has been proposed. The results predicated by means of the new concept werecompared with those experimentally observed in an Al-Zn-Mg alloy.

& An inflection behaviour was revealed during isothermal annealing under the observa-tion of transition of the coercivity of Fe-Si-B amorphous alloys on local crystallization. It maybe interpreted by the transition of α-Fe grain precipitated from single to mutiple domain. Thus.the theoretical expression of H_0-t_a may be derived. The plot of expression H_0-t_a calculated wellrepresents the feature of the experimental curve.

It is clarified by experimental examination and theoretical analysis that the trans-formation of crystal morphology of the eutectic Al-Cu alloy composite-in-situ seems to be judgedby not only the constitutional supercooling proposed in previous literatures but also the supercoolingdue to the effect of curved surfaces at the solidifying front. The greater entropy of fusion and thetendency to faceted face are important features of the non-metallic phase, which may contributeto the leading role of the phase during solidification. It might be the change of leading phase thatchanges the morphology of crystals.

The space group of PrCo_(12)B_6 compound has been determined using the convergentbeam electron diffraction method. The space group is found to be R3m.

The Bi-Sr-Ca-Cu-O superconducting ceramics with T_c (onset)=112K, T_c (R=0)=85.3K were prepared by quenching process. The Meissner effect of the sample appears at about84.0K.