Atomic structure of the ∑=5B(210)/[001] symmetrical tilt bound-ary in Cu containing different amount of segregated Bi has been investigated. Themost important structural effects found in this study are strong anisotropy of thesegregation energy for different sites and formation of an ordered alloy structurein the region of the grain boundary. A saturation level of 1.25 monolayers beyondwhich no further segregation can occur, was found.

Atomic structures of four ∑=17 (530)/[001] symmetrical bound-aries in Cu containing segregated Bi atoms at different sites have been studied. Thesegregation induced structural transformations from one structure to another accom-panying with the displacement of the upper grain in respect to the lower grainand the migration of the grain boundary plane have been observed. A shear mecha-nism has been proposed in order to explain such foundings on basis of the grainboundary structural unit model and the structural multiplicity.

The hydrogen induced cold work peaks (H-CWP) related to hydrogen-dislocation interaction for Fe-0.5 wt-% Ti and Fe-0.5 wt-% V alloys have beeninvestigated and the activation energies for the peaks obtained. The binding energiesbetween hydrogen and dislocation have also been estimated. Significant differenceon H-CWP for two alloys has been observed, that implies the effect of elements onH-CWP of Fe is significant. The stronger the interaction between hydrogen andalloy element is, the more significant the effect is. It is noticed that there issome correlation between the results of internal friction ditermination and themechanical properties.

The retained austenite at the fatigue plastic zone in a low-carbon mar-rensitic ultra-high strength steel transforms to martensite was observed. The influenceof this strain induced transformation on the fatigue behavior of the steel has beendiscussed.

Low temperature embrittlement has been observed for some austen-itic Fe-Mn alloys. The transition temperature of embrittlement raises with theincrease of Mn content, and the fracture model becomes intergranular at the lowtemperatures. The segregation of Mn at grain boundaries has been detected and theextent of it directly relates to the Mn content in the alloys. It is considered thatthis is the reason of the low temperature embrittlement of the alloys. No evidencewas found to prove a fcc-fct transition occurring at about-l00℃ as mentioned byBogachev.

On the basis of the work about notch sensitivity of aggregativematerials in fracture tests by Alberto Carpinteri, the relationships between relativecrack depth and brittleness number, crack extension force as well as the force ofcollapse at ultimate strength for cylindrical specimens with ring-shaped cracks havebeen obtained. It was found that the parameter brittlencss number can be extensi-vely applied to metallic materials. The possibility of the use of brittleness numberto deal with the ascast magnesium alloy, medium or high strength steels under thecondition of fracture at low stress has been disccused.

Combined corrosion-erosion (C-E) behavior of some commercial steelscontaining 2.25--25 wt-%Cr have been investigated at 850℃ using air as carryinggas and 130μm dia. SiC or Al_2O_3 as erodents. It was found that C-E behaviorstrongly depends on the corrosion resistance of steels. The C-E resistance increaseswith increasing Cr content, and a drastic raise of the resistance occurs at about9 wt-%Cr. Obvious influences of the nature and shape of the particles on the C-Erate of steels was noticed. Mechanisms of material degradation have also beendiscussed.

The microstructures of T6, CgS3 and T73 aged 7075 aluminiumalloy have been investigated using TEM and related to the yield strength and thestress-corrosion resistance (SCR) of the alloy, which were shown to be affectedmainly by the type, size and density of matrix precipitates (MPt). After aged bythe newlydeveloped CgS3 process, the majority of MPt is η' transition phase withabout 5×10~(16)/cm~3 in volume density and 5-7nm in size. The alloy subsequentlyshows an optimum combination of yield strength and SCR, which has been attributed to the increased tendency of dislocations by-passing MPt particles.

The tendency of spalling of nitrided layer of an austenitic heatresistant steel 4Cr14Ni14W2Mo has been studied experimentally. It was found thatthe residual deformation retained in the rolled steel is the root that causes thespalling of nitrided layer of the steel. If the work hardening state is removed by asuitable heat treatment before nitriding, the spalling of nitrided layer can be com-pletely avoided.

Creep crack growth of an Fe-Ni base alloy GH901 (similar to Incoloy901) was studied at 650--715℃ by electrical potential technique. The crack initiationand growth can be described by stress intensity K_I. When the initial stressintensity K_(Ii) was smaller than a threshold value K_(It), no crack growth was detectedin the testing time (200h). When K_(Ii) was larger than K_(It), the creep crack growthrate can be represented by an Arrhenius type equation. The activation energy forcreep crack growth was found in agreement with that for grain boundary diffusion.Fractography by SEM shows that the path of crack growth is along grain bound-aries. The physical origin of the threshold stress intensity K_(It) and the microscopicprocess of creep crack growth was explained in the light of grain boundary diffusionmechanism.

Hydrogen-induced fracture of electrolytic charged, sharp notched specimens ofsuper-high strength steel 30CrMnSiNi2A has been studied by means of cyclic ladder waveformloading tests and SEM examination. The results show that the time to fracture decreases withthe decrease of dwell period of the load. It was considered that hydrogen transported by disloca-tions significantly effects the hydrogen-induced delayed fracture of the steel.

Thermal plastic strain fatigue life of the rolls have been approximately estimatedby using Neuber's plastic stress-strain concentration theory, a path-independent nature integralJ defined by Rice and cyclic strain growth rate.

Growth of large single crystal Cu by Bridgman-Stockbarer method was reported.The conditions of growth are: temperature gradient at the level of melting temperature=10℃/cm;the lowering speed of crucible=7-12cm/h. The size of the largest crystal is: diameter=4.1cm; length=18cm; weight=2.1 kg

The microstructure and mechanical properties of 1Cr18Ni9Ti, FeCrAl andCr25Ni20SiMn alloys after injection of oxide particles have been determined. The results showthat the tensile strength and hardness at room temperature or high temperatures, the high temper-ature rupture strength as well as the oxidation-wear resistance at high temperatures of the alloysincrease owing to the injection.

An elastic-plastic wear model for metal abrasion has been built up based oncomparing the dry abrasion behavior and high temperature hardness of Ni-Ti, Co45 and 38CrMoAlA alloys and studying the microstructure and the deformation character of Ni-Ti alloy. It wasused to interprete the nature of were resistance of Ni-Ti alloy which possesses low hardness andhigh elasticity.

The activity of N in α-Fe was derived from the solubility data at580℃ in the Fe-N diagram. The interaction coefficients of Fe-N, Fe-C and C-Nwere obtained and the phase boundaries of single-phase ε-and γ_1-fields were subse-quently calculated. The results show that the γ_1-region of Naumann and LangescheidFe-C-N ternary phase diagram is correct, but the ε-region requires revision. Calcu-lation give an evidence that most of the experimental data on the ε-phase report-ed by Wells and Bell appears reasonable. Calculations taking graphite as a standaredstate confirms the above conclusions.

Ca-O equilibria in liquid Fe and Fe-C (C≤0.83 wt-%) have beendetermined by means of Ca vapor method in a sealed Mo chamber. The differenceof the equilibrium constants between them are quite small. Following results wereobtained for Fe-Ca-O system at 1600℃:K_(Ca-O)=4.3×10~(-8), e_O~(Ca)=-178

The solute distribution in the initial transient during unidirectionalplanar crystal growth has been studied. An expression of solute distributionin liquid phase near liquid-solid interface at the initial growth stage was derived.A method for reducing the length of initial transient was put forward and theshortest initial transient length was determined. The growth velocity function re-quired for realizing the shortest transient length at the initial stage was then ob-tained.

The crystallization behavior of amorphous alloys Fe_(80-x)Cu_xSi_5-B_(15) and(Fe_(1-y) Co_(y))_(82)Cu_4Si_(4.4)B_(13.2), has been studied by X-ray diffraction, Mossbauer γ-rayabsorption and conversion electron emission. Crystallization of amorphous ribbonprepared by means of single roller quenching starts at rough side in contact withroller. The product of crystallization is α-Fe. When a small amount of Fe issubstituted by Cu in amorphous Fe_(80)Si_5B_(15), the crystallization temperature increase.If the content of transition metal is more than 80 at.-%, for example 82 at.-%,a marked drop in crystallization temperature occurs. Three crystalline phases: α-Fe,Fe_2B and Fe_3B are coexistent for these two serises of Si-containing amorphousalloys after annealed 2h at 400--450℃. The metastable phase Fe_3B further trans-forms into Fe_2B and α-Fe at higher temperature.

The change of the characteristic parameters, strain rate sensiti-vity m, activation energy Q_a and grain size exponent P with changing strain rateand temperature and their physical reason have be n analysed according to the mul-tiple machanism effect in superplastic deformation. Their quantitative expressionshave also been given.

Through a study on Na corrosion of Al_2O_3/Nb seal interface, itis shown that the main factor influencing the Al_2O_3/Nb seal and the lifetime ofhigh pressure natrium lamp is the interface bonding between glass seal and Nb. The microstructure of interface of high pressure natrium lamp with differentlifetime has been studied by means of TEM, and the influence of interface struc-ture on the quality of Al_2O_3/Nb seal has been revealed. Based on the thermodyna-mic analysis and Pask theory of metal/ceramic interface bonding, the physicoche-mical process of glass seal/Nb interfaces and bonding mechanism during sealing hasalso been discussed.

The element distribution of the homogeneous binary samples withmechanicaly scraped surfaces have been studied by means of Auger-Ion sputteringtechnique. The resulted profiling curves, generally show that the steady state con-centration usualy deviates more or less from the bulk concentration and that anobservable "spike" appears in the near surface layer. The amplitude of the divia-tion or the "spike" depends on the system and sputtering condition. Therefore, theAuger-Ion sputtering profiling curve could not reflect the original element distribu-tion exactly due to the interaction between energetic ions and atoms on the samplesurface. Some effects leading to the "distortion" and some aspects which must betaken in to account in using the profiling curve have been discussed briefly.

Study was made of determination of the deviation from stoichi-ometry on oxygen-deficient MoO_(2-x). The diagram showing true relationships be-tween chemical potential of oxygen and compositions for the non-stoichiometricMo and MoO_2 equilibrium at 750, 850 and 950℃ is given. And the correction val-ues for the standard free energy of formation of MoO_2 are calculated by Smiltens'integration. In order to overcome the difficulty of the integration used for puremetal side, the Sieverts' law constant is introduced. It is shown that the correctionvalues effect little change in the standard free energy of formation of MoO_2.

The thermodynamic properties of nonstoichoimetric niobium mono-carbide (NbC_x) have been investigated by a, galvanic cell technique using calciastabilized zirconia(CSZ) and yttria stabilized zirconia(YSZ) solid electrolytes. EMFmeasurements were made in the temperature range 1100--1360K for compositionsvarying from NbC_(0.72) to NbC_(0.98) and activity relationships and partial molar freeenthalpies for Nb and C and integral molar free enthalpies for NbC_x have beencalculated from the results. The Fe-Nb-C system was assumed to be a regular solutionand the equation for evaluating γ/γ+NbC_x phase boundary in the Fe-Nb-C systemwas evaluated by using the integral molar free enthalpy of NbC_x determined bythis work lnN_(Nb)-N_(Nb)[13990+29390x-29390(1+x)N_c]/T+xlnN_c+ +[(1+x)6995N_(Nb)~2]/T-5284x/T+(13900+A)/T-B/8.314- -N_c[29390-10570x+5284(1+x)N_c]/T=0 When substituted 0.98, 0.945, 0.91 and 0.87 for x in the equation, the calcu-lated results were in excellent agreement with the results determined by Meyer, Chris-tensen, Nordberg, Johanson and SHENG Lizhen, Mori, respectively. The cause ofthe differences amony former data was also shown in present paper.

The rate equation for the reaction of PbO with PbS in liquidstate at 850--1000℃ has been determined as dα/dt=4.73×10~6 exp (-190/RT)in which the activation en ergy E=190kJ/mol. In the solid-liquid reaction system of xPbO. SiO_2-PbS, the rate equation maybe formulated as following: dF(α)/dt=k′(t<20min)+k"(t>20min) (1000℃)where F(α)=α-(N°_(SiO_2/N°_(PbO))ln(1-α) The activities of PbO and SiO_2 in the silicate calculated from the rate dataare in good agreement with those measured thermodynamically.

Based on the works of Zevin and Guo, a quantitative phase analysismethod without standard has been developed. Calculation of absorption coefficientof the sample and addition of extra phase are not necessary. In spite of the num-ber of phases in the sample, only the interesting phases are analysed. The composi-tion is expressed by diffraction intensities from different phases in the same sam-ple, thus, errors caused by the difference between different samples and experimen-tal conditions will be avoided.

The "carbon peaks" and high carbon activity spots appear in molten cast ironduring inoculation by Si. No evidence shows that the number of Si-rich particles whichmight be the heterogeneous nuclei of graphite is increased by inoculation. It is suggested thatthe mechanism of inoculation is the undulation of C concentration and activity in an inocu-lated cast iron. The fading of inoculation is caused by the fading of undulation.

The Raman spectra of molten KNO_3-NaNO_2 solution at 473--623 K have beenmeasured and the regularity of the shape of spectra investigated. The symmetrical stretching frequency peak v_1 of NO_3- or NO_2- becomes broaden as the temperature increases or the concentra-tion of NO_3- or NO_2- ions in KNO_3NaNO_3 mixture decreases. The results have been discussedbased on the view-point of a theoretical model obtained in previous work.

Single phase compounds SmPd_5 and EuPd_5 have been prepared respectively X-raydiffraction has been used to determine the structure and lattice parameters. Both of them exhibita SmPt_5-type structure. Following lattice parameters have been obtained:α=0.5253nm, b=0.9015nm, c=2.568nm for SmPd_5;α=0.5268nm, b=0.8984nm, c=2.553nm for EuPd_5.

Amorphous Fe_(75)Cr_5P_(13)C_7 alloy powder was prepared using internal gyrate solutionmethod. The powder is irregular and east to shape. Corrosion tests of the powder and block inacid and neutral medium environments show that the corrosion resistance of the powder isexcellent.

Magnetic anisotropy and distribution of moments in the ribbon of rapidly quenchedamorphous alloy Fe_(81)B_(13.5)Si_(3.5)C_2 have been studied with~57Fe Mossbauer effect. The results showthat thinning and surface removing of ribbons have little influence on the internal stress systemand distribution of moments in as-quenched sample, and the distribution of moments does notvary along the thickness of the ribbon.

The lifetimes and S parameters of Co-Ga binary alloys with different compositionsat different quenching temperatures have been determined by means of positron annihilationmethod. Two types of β solid solution, i. e., substitutional solid solution at Co-rich side ofthe ideal composition and Vacancy solid solution at Ga-rich side have been observed respec-tively.