WANG Litian;GE Tingsui (T. S. Ke) Institute of Solid State Physics, Academia Sinica, HefeiInstitute of Metal Research,Academia Sinica,Shenyang

Effect of austenite strengthening on M_s and B_s temperatures in Fe-Ni-C alloys and some commercial steels have been studied. M_s temperature lowers1inearly with the increase of the yield strength of the austenite, while B_s seems tobe independent of the austenite strengthening. By plotting B_s temperature of thethree Fe-Ni-C alloys versus D~γ(C) and D~γ(Fe), a linear relation is obtained. The△G_v and D~γ(Fe) of the alloys were calculated at the nose temperatures in TTT di-agrams, and the incubation periods are found to be directly proportional to thevalues deduced from Feder's equation for diffusional transformations, implying thatthe incubation period at the nose temperature of bainitic transformation is gov-erned by the change in free energy and the diffusion of iron atoms in austenite.

The formation of the lamellar ferrite-martensite dual-phase structuredeveloped by prequenching and subcritical quenching in 09SiCrMnMoRE steel hasbeen studied. It is shown that during subcritical annealing, austenite pools are pref-erentially nucleated at prior austenite boundaries and martensite packet boundaries,and then austenite lamellae form at lath boundaries accompanied by cementite pre-cipitation and dissolution. The lamellar austenite growth along the lath and themartensite packet boundaries is beneficial to the decrease of both the interface andthe strain energies which are necessary for austenite formation. As a result, thelamellar ferrite-martensite dual-phase structure is developed after quenching.

Fe_3Al precipitates with three morphologies have been observed in themicrostructure of the brass containing aluminium and iron. Studies of the micro-structural development show that these three morphologies reflect three stages ofthe precipitate growth. The initially formed Fe_3Al precipitates are coherent withthe matrix. The growth of precipitates is along <111> direction during the coars-ening, which results in the change in morphology (from cubic to dendritic) and theloss of coherency of the precipitates.

Effects of high temperature gas charging of hydrogen and tensile de-formation on the martensite transformation and mechanical properties of three Fe-Cr-Ni-Mn-N steels with various stabilities of austenite resulted from the differenceof compositions have been investigated by means of optical metallography, X-raydiffraction, TEM technique and magnetic determination combined with tensile tests.The results show that tensile deformation promotes the formation of ε- and α'-martensite in the steels in which the austenite is not stable enough. However theexisting of hydrogen enhances the γ→ε transformation but suppresses the formationof α'- phase. It is considered that the formation of ε- martensite is the main rea-son of the hydrogen susceptibility of unstable austenitic steels.

The void swelling and grain boundary segregation induced by elec-tron irradiation in Fe-15%Cr-30%Ni alloy has been studied, using the high-voltageelectron microscope (HVEM) as the source of electron beam. The results show thatat 723 K and maximum dose of 15 dpa (displacement per atom), the bulk swellingis 0.62%. Ni enriched at the grain boundaries while Cr become depleted. Withincreasing irradiation temperature, the segregation increases. The maximum segrega-tion peak of Ni appears at 723K, accompanied by maximum bulk swelling. A modelof irradiation-induced grain boundary segregation has been proposed to correlatethe segregation with void formation.

Permeability, diffusivity and solubility of hydrogen in austenitic stain-less steels 316L, 316LN, 21-6-9, 21-9-9, 304 and 1Cr18Ni9Ti have been measuredby gaseous permeation technique over the temperature range of 200 to 430℃. Theeffects of composition, cold-work and heat treatment on hydrogen petmeation havebeen investigated. The results indicate that permeability and diffusivity of hydro-gen in various alloys over the experimental temperature range follow arrheniusequations and hydrogen permeation behaviour is not obviously influenced by cold-work and heat treatment of materials, but slightly influenced by the composition.The difference of hydrogen permeation behaviour between pure iron, low alloysteels and austenitic stainless steels has also been discussed.

Corrosion fatigue behaviour of an as-rolled dual-phase steel has beeninvestigated in tap water or 3.5%NaCl agueous solution under full immersion andintermittent wetting. It was found that in comparison with the data in air, expo-sure to either tap water or salt water results in a substantial reduction in fatiguelife and intermittent wetting later shows the most harmful effects. However, underthe same aggressive condition, the corrosion environments become more and moreharmful while the stress level decreases. A combined observation of fracture surfacewith the flat free surface in SEM shows that corrosion fatigue cracks initiateoften at corrosion pits resulted from the anodic dissolution. In addition, the inter-action between hydrogen and crack tip is considered another important factorwhich controls the crack initiation and propagation in corrosion fatigue. The mar-tensite exhibits a higher resistance to corrosion fatigue than the ferrite matrix.

Effect of the interval (△t) of compressive overload peak (S_p) on thefatigue-creep interaction for GH698 alloy at 700℃ has been studied. It was foundthat, under the condition of superimposed S_p, the fatigue damage aggravates withthe decrease of △t, meanwhile, the fraction of fatigue fracture mode on fracturesurface was extended and that of intergranular creep fracture mode was supressed.The effect of superimposed S_p on both the dynamic creep rate ε and fatigue crackgrowth rate dα/dN depend upon the △t. As △t. was getting shorter, the factorwhich acceterates dynamic creep and crack propagation becomes dominant, this re-sults in the decrease of rupture life and vice versa.

The effects of B and C content on the mechanical properties and mi-crostructures of a Ni-Fe-Cr base superalloy have been investigated. It has beenfound that the role of B in strengthing the grain boundaries is much stronger thanthat of C. Both the B dissolved in matrix and precipitated as boride along thegrain boundaries increase the stress rupture properties of the alloy markedly. Ad-dition of B improves the structure of grain boundaries, suppresses the precipitationof η phase and raises the stability of microstructure of the grain boundaries in thealloy. Addition of either B or C does not effect the tensile strength obviously butincrease the tensile ductility of the alloy. In order to obtain the optimum tensile properties, the minimum contents of Band C should be 0.002 and 0.02% respectively.

The transgranular growth of martensite in a CuZnAl shape memory alloy was ob-served by using TEM and optical microscope with heating attachment. The analysis shows that thetransgranular growth seems to be a general behavior of thermoelastic martensite at low anglegrain boundaries.

The crystal texture of secondary recrystallization of 50%Ni-Fe alloy has been inves-tigated quantitatively with X-ray Laue method. The samples used throughout the investigationwere several sheets of 50%Ni-Fe alloy cold rolled with 98% thickness reduction. Samples annealedbelow 1000℃ produced a cubic texture with fine grain size, while samples annealed at highertemperatures (1070, 1150, 1250, and 1240℃) produced secondary recrystallization, the ori-entation of which are different from the cubic one but belong to several definite types of orien-tations. For each annealing temperature, about 30 crystals were selected from a sample sheet forthe orientation determination with Laue method. The percent abundance of the texture componentsat several annealing temperatures has been calculated according to the relative volume of the crys-tals with the different orientations.

A new type of electroslag refining technique has been proposed andthe influence of superimposed DC density on the contents of active metal elementsas well as on the desulphurization and degassing have been studied. The resultsshow that the enerey consumption may be reduced by 41% by using this technique.During the semi-lining ESR under normal atmosphere, there is no evident loss ofactive metal elements, while the contents of O, N and S can be effectively reducedto 10ppm in superalloys GH49 and GH37. The ingot solidified along its axis andhas a compact microstructure. The secondary dendrite arm spacing of the semiliningESR ingot is about 1/3 shorter than that of tbe general ESR ingot.

TiN films were deposited onto high speed steel, Si(100) and Si(111) substrates using D. C. PCVD method. The crystallographic structure, surfacemorphology, fracture cross section, microhardness and the chlorine content of thePCVD TiN films were studied. Part of the samples were analyzed using SIMS, AESand ESCA. Experimental results indicated that the structure and properties of theTiN films on different substrates were almost the same when the same depositionparameters were used. At about 500℃, the mode of the film growth undergoes achange probably from layer growth to island growth. The PCVD TiN films had aratio N/Ti of 1, and revealed strong (200) texture. A mixed region existed betweenthe films and the substrates. It could be concluded that PCVD TiN films show goodadherence and excellent wear resistant properties, and are suitable for anti-frictioncoating.

The thermodynamic properties of Fe-C-Pb liquid solution in the tem-perature range 1300--1400℃ have been studied using the equilibrium method betweenPb vapour and liquid Fe in sealed Mo reaction chamber. The temperature depend-ence of the first and second order interaction coefficients between C and Pb areobtained as follows: e_(Pb)~C=9205(1/T)-4.578, γ_(Pb)~C=-1047(1/T)+0.512. The thermody-namic preperties of Fe-C_(sat)-Pb-Ce liquid solution at 1300℃ have been determinedusing the equilibrium method between vapour and liquid in a graphite cruciblesealed in a vacuum quartz tube. The interaction coefficients between Ce and Pb arederived to be: e_(Pb)~(Ce)=-2. 11, e_(Ce)~(Pb)=-1.43.

The thermodynamic properties of Fe-C-Sn liquid solution have beenstudied by using the solubility method in the temperrature range 1300-1500℃ andby using the iso-tin-activity with silver bath at 1550℃. The results are summarizedas follows: 1. The relation between the solubility of C and the content of Sn in liquid Feare: 1300℃ [%C]_(sat)=4.65-0.086[%Sn] 1400℃ [%C]_(sat)=4.86-0.106[%Sn] 1500℃ [%C]_(sat)=5.13-0.131[%Sn] 2. The solubility of C dependence on the content of Sn and the temperature isobtained as [%C]_(sat)=8.41-0.101[%Sn]-5904(1/T) 3. The temperature dependence of interaction coefficients between Sn and C inC saturated liquid Fe are ε_C~(Sn)=-(27170)/T+20.07; e_(Sn)~C=-(51.29)/T+0.041 4. The temperature dependence of interaction coeffients between Sn and C inliquid Fe are derived as e_C~(Sn)=-(194)/T+0.151; e_(Sn)~C=-(1810)/T+1.39 5. The first and second order interaction coefficients between Sn and C in liq-uid Fe at 1550℃ are determined as e_(Sn)~C=0.12; e_C~(Sn)=0.016; γ_(Sn)~C=0

Ca_3P_2 was formed by the reaction of Ca with P, and by the decom-position of Ca_3P_2 at 1600℃, the equilibrium constant between Ca and P, and theinteraction coefficient of activity were obtained as follows: K_(Ca_3P_2)=a_(Ca)~3·a_P~2=7.0×10~(14), e_P~(Ca)=-3.1

A three dimensional partial phase diagram of Fe-Si-C was built basedon the geometrical nature of phase diagram using the data of some pseudo-binarysections of Fe-Si in literatures and experimental data of developing column grainsin Si-steel sheets. Series of pseudo-binary sections for Fe-Si and Fe-C were givenas well.

The hot deformation behaviour and microstructure change of a Cr25 Tiferritic stainless steel under compression at strain rate 5×10~(-2)s~(-1) and 1100℃ byusing formaster press have been studied. It is observed that in the early stage, theminor fine grains were found along the original grain boundaries, and under thefurther increasing strain, the intragranular nucleation became more predominate.The stronger dynamic recovery results in a lot of well-defined subgrains. The strainwill induce the subgrain rotation together with the subboundary coalescence andform new grains. An observation also revealed that lattice orientation betweendynamic recrystallization grains is formed from the progressing rotation in "tilt"or "torsion" of subgrain in the original grains. It seems that the dynamic recrys-tallization nucleation of Cr25Ti steel is similar to several minerals in that bothare nucleated homogeneously by the strain inducing progressing rotation of sub-grains. However, a high degree of "asynchronism" during dynamic recrystallizationcauses no obvious peak on the stress-strain plot.

The changes in the reduced radial distribution function of metallicglasses TiCu and Zr_(3.12)Ni before and after hydriding have been measured using thewide-angle X-ray scattering. It is shown that if hydrogen atoms mainly occupy thetetrahedral interstices defined by four Ti or Zr atoms, the displacement of the firstpeak can be interpreted quantitatively. Broad peaks with approximate periodicity atlarger r range is also discussed.

A high time resolution thermovision and an infrared detector withthe diameter of 20mm were used to detect the temperature distribution on the sur-face of samples which rotated and bending fatigued at high speed. The relation be-tween infrared radiation and local fatigue damage has been determined. A new non-destructive testing method for fatigue damage has been given.

The sum of squares of deviations of the equation for calculating kwas constructed using the measured data of m samples. By deriving the minimumof the sum of squares of deviations of n equations, i.e. let ?sum from j=1 to m(M_j--sum from i=1 to n I_(ij)k_s~i)~2'/?k_s~i=0,the regressive coefficients of the n equations can be obtained (m>n). The k valuescan be deduced from the matrix equation Δ_k=Δ_I~(-1), Δ_M, and the n unknown phasesmay be calculated subsequently. It is expected that the trouble of error expansionduring solving the polynomial equation might be overcome by minimizing the sumerror through regression. The above method has been used to analyse precipitatesin high speed steel W_9Mo_3Cr_4V.

A high T_c oxide YBaCuO superconducting coating has been prepared by plasmaspraying method. The structure and phase transformation of the coating during spraying and after-annealing treatment process were studied by X-ray diffraction, TEM, SEM and DTA technique.The results show that the as-sprayed coating lost its superconductivity completely at liquid nitro-gen temperature and was composed mainly of microcrystalline and amorphous structure. Its su-perconductivity can be restored by a proper annealing at 850--950℃.

Simulative study for erosion of refractory in VOD or AOD steel refining was car-ried out by the method of rotating cylindrical magnesite-chrome specimen immersed in a synthe-tic molten slag. It was resulted that the dissolution rates of specimens with contents of sameCr_2O_3 but various Al_2O_3 are the same at the ratio Al_2O_3/MgO from 0 to 0.085% and increasesharply from 0.085 to 1.14% then smoothly over 1.14%. With the increasing of Al_2O_3 content inthe slags, the dissolution rates of specimens decrease, and their decrement meet each other at 25%Al_2O_3 in slags. The microcopic observation and EPMA were also made of the eroded specimens.

Young's modulus of Nd-Fe-B magnet within temperature range from 25 to 800℃was measured by means of ultrasonic resonance. E_∥ measured along c-axis is always smaller thanE_⊥, measured perpendicular to c-axis. The relative difference of Young's modulus (E_⊥-E_∥)/E_⊥=2.2% at room temperature remains almost the same till 500℃, whereas increases very sharplywhen T>500℃. Taking the value of aboat 12% at 600℃, i.e. 5 times more than the value atroom temperature. There is a step following a steep fall of both E_∥ and B_⊥ within the tempera-ture range of 550--600℃, which may be attributed to the melting of Nd-rich phase in the mag-net. Since Young's modulus reflects mainly the strength of bonding between atoms, i. e. thestiffness of the materials, the obtained results imply that the anisotropy of stiffness for Nd-Fe-Bmagnets becomes much higher at temperature above 600℃. Both the high anisotropy of stiff-ness and the remarkable softening at T>600℃ provide evidences for the mechanism of inducedtexture due to hot pressing.

In the amorphous alloys (Fe_(80)Ni_(20))_(78)Si_(22-x)B_x, the magnetic moment and the curietemperature were found to increase slowly with increasing boron and relatively decreasing silicon.This phenomenon has been explained qualitatively in terms of the electron transition model andthe molecular field approtimation theory.