Search

Generation and growth of whiskers and nodules from electroplated tin and tin-lead films on copper or nickel substrates were studied by scanning electron microscopy and X-ray diffraction. In the case of copper substrates, whiskers were formed in 0.3 Ms on the tin film whose thickness was 1 mm. On the other hand, tin-lead films on copper substrates showed only nodules even after 13 Ms. Residual stress of the tin film (1 mm) and the number of whiskers increased with the amount of copper-tin intermetallic compounds (Cu6Sn5) which developed between the plated film and the copper substrate. Although residual stress and the amount of Cu6Sn5 also increased in the tin-lead system, the morphology of the layer of Cu6Sn5 was more uniform compared with the case of tin films on copper substrates. When the tin-lead film was subjected to the compression testing by a ball of zirconium oxide (1 mm-diameter, 2.94 N–0.605 Ms), the edge of the indentation did not show any whiskers but diffusion of lead was observed from right under the zirconium oxide ball to the fringe, as well as Ostwald growth of lead. In the case of the tin film on the nickel substrate, whose residual stress was weakly tensional, nickel-tin intermetallic compounds (Ni3Sn, Ni3Sn2 ,N i 3Sn4) uniformly developed immediately after electroplating, and no whiskers were observed even after the compression testing. While growth of whiskers is considered to be due to diffusion of tin atoms induced by inhomogeneous strain field in the electroplated film, lead atoms in tin-lead system is considered to diffuse rapidly toward the free surface to release residual stress and to generate many nodules. [doi:10.2320/matertrans.M2009292]

Corrosion resistance of anodized surfaces on high-purity magnesium (99.95 mass%), rolled sheets of ASTM AZ31B (Mg-2.9Al-0.85Zn) magnesium alloy and die-cast plates of ASTM AZ91D (Mg-9.1Al-0.75Zn) magnesium alloy has been studied. Anodization was conducted by environment-friendly electrolysis whose electrolyte consists of phosphate and ammonium salt. The anodized surface was covered with amorphous film, and showed only discoloration during salt spray test where formation of corrosion product (magnesium hydroxide) was well suppressed within 605 ks. Even when the anodized surfaces were trenched with ceramic knife to form locally exposed substrate, corrosion was well suppressed by formation of new type of dense protective films for each substrate which consist of oxygen, magnesium, aluminum and phosphorus. Anodic polarization curves indicate that the anodized surfaces show sacrificial function due to the thermodynamically unstable state of phosphorus in the anodized layers and its resulting release of electrons. From the viewpoint of kinetics in corrosion on the anodzed surfaces, the curves show that the anodized layers dissolve quite slowly into the electrolyte compared with the case of the untreated substrates. The excellent corrosion protection obtained by the anodization is considered to be based on the formation of a dense protective film on the exposed area, as well as sacrificial function of the amorphous anodized layer.

Generation and growth of whiskers on tin electroplating was studied by scanning electron microscopy and X-ray diffraction. When the surface of the plating, whose thickness was 1 mm, was etched by argon ion beam to remove native oxide scale and kept in a vacuum, whiskers were rarely observed but many nodules were formed. While residual stress of tin plating (1 mm) and the number of whiskers increased with the amount of Cu6Sn5 formed between the tin plating and copper substrate, the 10 mm thickness specimen showed no meaningful change and no whiskers were found. The orientation of whiskers was shown to be identical throughout a crystal by electron backscattering diffraction even when kinked. The mechanism of generation and growth of whiskers is considered to be due to diffusion of tin atoms induced by inhomogeneous strain field, formation of vacancies, and the resulting increase in volume of the grain receiving the tin atoms. Energy change in the growth of whiskers was modelled and calculated, and the main factor determining the growth was found to be the formation enthalpy of a vacancy and the entropy of configuration. [doi:10.2320/matertrans.MRA2008236]

Corrosion resistance of anodized surfaces on high-purity magnesium (99.95 mass%), rolled sheets of ASTM AZ31B (Mg-2.9Al-0.85Zn) magnesium alloy and die-cast plates of ASTM AZ91D (Mg-9.1Al-0.75Zn) magnesium alloy has been studied. Anodization was conducted by environment-friendly electrolysis whose electrolyte consists of phosphate and ammonium salt. The anodized surface was covered with amorphous film, and showed only discoloration during salt spray test where formation of corrosion product (magnesium hydroxide) was well suppressed within 605 ks. Even when the anodized surfaces were trenched with ceramic knife to form locally exposed substrate, corrosion was well suppressed by formation of new type of dense protective films for each substrate which consist of oxygen, magnesium, aluminum and phosphorus. Anodic polarization curves indicate that the anodized surfaces show sacrificial function due to the thermodynamically unstable state of phosphorus in the anodized layers and its resulting release of electrons. From the viewpoint of kinetics in corrosion on the anodzed surfaces, the curves show that the anodized layers dissolve quite slowly into the electrolyte compared with the case of the untreated substrates. The excellent corrosion protectivity obtained by the anodization is considered to be based on the formation of a dense protective film on the exposed area, as well as sacrificial function of the amorphous anodized layer.

The high temperature deformation behavior of a fine-grained and particle-dispersed V-2.3%Y-4%Ti-3%Mo (mass%) alloy was investigated. The alloy was fabricated by powder metallurgical methods utilizing mechanical alloying and hot isostatic pressing (HIP), followed by annealing at 1273 K for 3.6 ks. Tensile tests were performed at temperatures from 873 to 1273 K at initial strain rates from 2:5 � 10 � 5 to 1:0 � 10 � 1 s � 1 . X-ray diffraction analyses show that the HIPed and annealed specimens contain a significant amount of (Y, Ti)2O3 and small amounts of YN, Ti2O3 and TiO. Transmission electron microscopy observations show that the matrix contains approximately 3%Ti and 3%Mo in solution and that the average diameters of the matrix grains and dispersoids are about 620 and 42 nm, respectively. Tensile test results show that the yield stress depends strongly on test temperature and strain rate. From the dependence of the yield stress measured at 1073 and 1273 K on plastic strain rate, it is found that the dependence is divided into three regions with different deformation controlling mechanisms: a recovery controlling process of a long range internal stress field associated with dispersed particle (the high-strain-rate region), grain boundary sliding (the medium-strain-rate region) and presumably solute atmosphere dragging (the low-strain-rate region). Effects of 4%Ti addition on the microstructures and high temperature deformation behavior are discussed. [doi:10.2320/matertrans.47.2497]

Intragranular bainite (α), which consists of 24 variants formed at random, in steels has been investigated in order to develop both strength and toughness. The orientation relationship, {011}α//{111}γ, α// γ and α// γ, between intragranular bainite and austenite (γ) has been analyzed to be an important controlling factor for toughness. It could be suggested that the increase in the number density of nucleation site for lath of intragranular bainite increases in both strength and toughness of bainitic steels.

Amorphous and crystallized calcium phosphate (Ca-P) coatings on Ti substrate have been investigated by X-ray photoelectron spectroscopy (XPS). The Ca-P coating on Ti substrate was performed by radio frequency magnetron sputtering of a sintered hydroxyapatite (HAp) target. As-sputtered Ca-P coating having 1.81 of Ca/P ratio was amorphous. Annealing the as-sputtered amorphous coating at 600°C resulted in crystallized coating into HAp and tetracalcium phosphate (TTCP). The XPS profile of amorphous coating is almost the same as that of annealed coating. The XPS profile of oxygen in the amorphous coating shows that oxygen atom combines with both phosphorus and calcium atoms, indicating the amorphous coating contains fundamental chemical bonds of both HAp and TTCP.

The cohesiveness of films on titanium was evaluated by a scratch test. The films were formed on titanium by immersion in an aqueous solution of NaOH (hereafter referred as "alkali-treated") followed by heating at 500-700°C for 3.6 ks. The cohesion of the alkali-treated film was increased by the heat treatment, and the maximum cohesion was obtained by heating at 500°C. The film on alkali- and heat-treated titanium consisted of both microporous films and dense films. The microporous film furthest from the titanium surface was barely cohesive, whereas the dense film showed cohesion with titanium. The cohesion of film formed by alkali-treatment was increased on heating, but the formation of TiO 2 (rutile) at the film-titanium interface on heating at and above 600°C caused a deterioration in the cohesion.

An alloy having ultra-fine grains with dispersed particles is expected to have the resistance of irradiation embrittlement. We examine the microstructures of V-1.6 and 2.6Y alloys, having ultra-fine grains with dispersed particles, after neutron irradiation at temperatures between 290 and 800°C. The grain size was the level of a few hundreds of nanometers, and the size in diameter of V-2.6Y alloy was much smaller than that of V-1.6Y alloy, where the irradiation did not change the size. The number density of particles in V-2.6Y alloy was higher than that in V-1.6Y alloy. Voids were formed only at 290°C, where the number density of voids in V-2.6Y alloy was smaller than that in V-1.6Y alloy. The void formation was efficiently suppressed due to fine-grains and dispersed particles.

Decomposition processes of β phase in a Ti-15Zr-4Nb-4Ta alloy during isothermal holding have been examined mainly by transmission electron microscopy. Specimens solution-treated at 1000°C in β phase field were held at temperatures between 400 and 600°C. a phase lath formed in β phase during isothermal holding at 600°C. After holding at 600°C for a short duration, remaining β phase, untransformed β phase into a phase at 600°C, was transformed into a' martensite during quenching into ice-brine. As increasing holding time at 600°C, remaining β phase was transformed into a" or athermal ω phases during quenching. After prolonged holding at 600°C, remaining β phase was not transformed during quenching. Enrichment of Nb, Ta and Zr into remaining β phase occurred with increasing holding time at 600°C, resulting in the variation of transformation of remaining β phase during quenching. Microstructures evolved at 400 and 500°C were almost same as those at 600°C.

The effects of isothermal annealing on the magnetic properties of Pr4.5Fe77B18.5 amorphous ribbons have been investigated. Two separate reactions increasing coercive force ( Hc) were recognized during the progress of annealing. In the first reaction, hard magnetic Pr 2Fe14B particles with nanoscale diameters precipitated in the vicinity of Fe 3B and Pr2Fe23B3 particles. The precipitation of fine Pr 2Fe14B particles increased Hc largely but reduced saturation magnetization ( Ms) significantly. In the second reaction, the Pr 2Fe23B3 particles decomposed into Pr2Fe14B particles containing very fine α-Fe particles within them. This resulted in the second increase in Hc due to both the volume fraction increase of Pr2Fe14B particles and the exchange-coupling interaction between the Pr2Fe14B grains and α-Fe particles within them. This phenomenon was observed originally in the present study. In the final stages of annealing, both Hc and Ms were decreased substantially by the decomposition of Pr2Fe14B into PrFe4B4 particle in addition to the coarsening of α-Fe particles.

study will be necessary to specify the structures. It should also be noted that the precipitation sequence has often been examined by differential scanning calorimetry (DSC) during continuous heating and that the results have been compared in many cases with those obtained by transmission electron microscopy (TEM) for the isothermally aged specimens. The processes in continuous heating, however, will be significantly different from those in isothermal ageing. The purpose of the present study is, therefore, to clarify the effects of excess Si and Mg on the ageing behaviour during continuous heating by means of DSC, TEM and Vickers hardness measurements, systematically.

Phase decomposition in a Ti–13 mass%Nb–13 mass%Zr alloy during isothermal aging at 600°C has been investigated by means of transmission electron microscopy and Vickers hardness measurements. Specimens solution-treated at 1000°C in β phase field were quenched and aged at 600°C in (α+β) region. Beta phase was fully transformed into martensitic α′ laths by quenching from 1000°C. Formation of β phase on tempered α′ lath interfaces occurred during aging. Early in the stage of aging, β phase transformed into α′′ martensite completely by quenching. As increasing in aging time, athermal ω phase was formed in β phase by quenching. Beta phase was stabilized by prolonged aging. Enrichment of Nb content into β phase occurred with increasing in aging time, resulting in the formations of athermal ω phase and (α+β) two-phase structure. The hardness increased with the ω phase formation in β phase, followed by the decrease of hardness due to suppression of ω phase formation induced by enrichment of Nb in β phase.

The effects of excess Mg and Si contents on the isothermal ageing processes of Al-Mg2Si alloys have been investigated by means of transmission electron microscopy and the following results were obtained. After the formation of β′′ needles inducing large age hardening, cuboid β particles precipitate in both the excess Mg and the quasi-binary alloys, but fine Si particles nucleate in the excess Si alloys. In the quasi-binary alloy, the following reaction is the precipitation of β′ rods and then β plates form. In the excess Si alloys, various rodlike precipitates form after the precipitation of β′′ needles and Si particles in the sequence: Type-A rods→Type-B rods→β′ rods. This precipitation sequence can be understood by considering the chemical compositions of them determined by Matsuda et al.

The phase transformations in a Ti-10 mass%Zr alloy during continuous cooling and isothermal holding have been investigated mainly by means of hot stage optical microscopy. During cooling from β phase at 100°C/s, a' plates formed martensitically at temperatures around 180°C together with sharp and coarse surface reliefs. The Widmanstatten a laths mostly nucleated directly on a β phase grain boundary and grew into one of the grains during isothermal holding at temperatures between 650 and 850° C. At β grain boundaries, a allotriomorphs often nucleated and Widmanstatten a laths grew from them during isothermal holding at 800°C. Widmanstatten a plates exhibited sharp surface reliefs. The precipitation of a phase exhibited the C-curve with the nose at around 775°C. The Ms temperature so far reported to be ∼ 805°C is not the martensite start temperature but the nose temperature for Widmanstatten a laths.

The exothermic reaction occurring after β precipitation in Al-Mg-Si alloys has been investigated by means of differential scanning calorimetry (DSC) and transmission electron microscopy (TEM). The reaction peak height observed in DSC increases with increasing excess Mg or decreasing excess Si content from Al-Mg 2 Si quasi-binary composition. It was confirmed that this reaction peak arises from the precipitation of cuboid β-Mg 2 Si particles related to the matrix with the cube-cube orientation relationship. At the later stage of ageing, thin β-Mg 2 Si plates nucleate at the cuboid β-phase particles. The orientation relationship of them is rotated 45° or 18.4° in the further aged specimens about the direction normal to the {001} Al II {001} β habit plane from the cube-cube relationship. The changes in morphology and the crystallography of the β-Mg 2 Si precipitates with the progress of ageing can be explained qualitatively in terms of coherency and interface energy of the precipitates.

The isothermal crystallisation processes and the decomposition behaviours of the crystallised phases in Pr 4.5 Fe 77 B 18.5 amorphous ribbons prepared by a melt quench technique have been investigated by means of differential thermal analysis, X-ray diffraction, and high resolution transmission electron microscopy, and the following results were obtained. The phase initially crystallised is Fe 3 B and the crystallisation of Pr 2 Fe 23 B 3 particles occurs immediately after the Fe 3 B formation. This is probably due to the fact that Fe 3 B/amorphous interphase boundaries provide potent nucleation sites for Pr 2 Fe 23 B 3 particles in view of chemistry in the vicinity of Fe 3 B particles and the structure of the interfaces. Then, fine hard magnetic Pr 2 Fe 14 B particles nucleate at Fe 3 B interfaces directly. In the following processes, Pr 2 Fe 14 B and Pr 2 Fe 23 B 3 particles grow by the coalescence of fine particles. In the later annealing stages, Pr 2 Fe 23 B 3 particles decompose into Pr 2 Fe 14 B networks involving spherical α-Fe particles in a eutectoid-like mode. At the final stage of annealing these Pr 2 Fe 14 B networks transform in-situ into those of PrFe 4 B 4 .

The decomposition processes of a phase into massive, Widmanstatten and lamellar structures in a Ti-48 at%Al alloy have been investigated in view of morphology and crystallography. In the case of the most rapid quenching from the a phase field, disordered massive γ structure (MS) nucleates at grain boundaries and grows accompanying transformation twins, pseudo-twins and stacking faults, the structure being ordered subsequently. The growth of some antiphase domain boundaries preferentially initiates at the stacking faults. The MS/α-interfaces often act as nucleation sites for fine Widmanstatten y structure (FWS). By reducing the quenching rate, Widmanstatten y structures (WS's) directly nucleate at the grain boundaries apart from both the MS and the FWS. Within the MS, the FWS and the WS, α 2 phase is formed to relax transformation strains. The retained regions decomposes into the (α 2 + y) lamellar structure (LS) exhibiting similar characteristics to those in Ti-40 at%Al alloys. Both the FWS and the WS form in the a phase with the Blackburn orientation relationship, (0001) α //(111) γ and (2110) α //(110] γ , the habit planes being (111) γ //(0111) α . The formation of the MS, the FWS and the WS is suppressed with increasing the degree of order of the a phase matrix into α 2 phase.

The aging processes in a 6061 commercial aluminum alloy during continuous heating at various heating rates have been studied by means of DSC measurements. Eight reactions have been detected, six exothermic and two endothermic. The activation energy for the first exothermic peak observed at about 373 K at a heating rate of 10 K min -1 is 79 kJ mol -1 , almost the same as the activation energy for vacancy migration. Considering the excess vacancies exist in the as-quenched condition, this reaction can be explained in terms of vacancy cluster formation. The second and the third exothermic reactions within the large endothermic peak are thought to arise from the formation of solute atom clusters involving silicon and magnesium. The endothermic reaction is probably due to the dissolution of these atomic clusters. The activation energy for this process is close to those for the bulk diffusion of magnesium and silicon atoms in aluminum. A large and sharp exothermic peak at temperatures around 500-550 K is due to the precipitation of intermediate β needles. The activation energy for this process is significantly smaller than that for bulk diffusion. The reason for it is not clear at present, but may arise from the needlelike shape of the precipitates. The following two exothermic reactions which are difficult to separate clearly from the peak at 500-550 K are probably due to the precipitation of β' needles and the stable β phase, respectively. The dissolution of stable β phase occurs at higher temperatures.

Effects of cold rolling on the aging kinetics in an Al-Mg-Si alloy have been examined mainly by means of differential scanning calorimetry and transmission electron microscopy. The first exothermic reaction detected at temperatures around 350 K in the cold rolled specimen is probably due to the migration of quenched-in vacancies to the dislocations introduced by cold rolling and the molar heat for this reaction is decreased by cold rolling. The subsequent diffuse exothermic reaction may arise from the segregation of solute atoms to the dislocations, Cottrell's atmospheres being formed. The needlelike precipitates formed in the non-recrystallized matrix of the cold-rolled specimens, and this reaction was displaced to lower temperature side by cold rolling. The molar heat and the activation energy for the needle formation did not vary with reduction, but the activation entropy increased.

The effects of ω phase precipitation on the decomposition processes of α/β duplex phase structure in a Ti-10V-2Fe-3Al alloy have been investigated by means of both X-ray diffraction and transmission electron microscopy. In the specimen quenched from β phase region, thin α martensite plates form within the β phase matrix, the structure being consisted of both α and β phases. In the retained β phase region, however, the diffuse scattering showing the pre-precipitation stage of ω phase particles can be recognised. By ageing at temperatures from 448 to 673 K, α martensite plates transform gradually to β and spherical ω phase precipitation follows. This ω formation stabilises β phase and retards the β→α transformation during the subsequent cooling. In the temperature range between 573 and 673 K where both ω and α phases form, ω phase particles precipitate first as well-defined and homogeneously dispersed spheres about 10 nm in diameter, and then very fine α phase platelets 100-200 nm long precipitate with replacing ω phase particles. At temperatures above 873 K, coarse monolithic α phase plates precipitate directly within β matrix without ω phase precipitation. The interplanar spacing of the retained β phase varies with the ageing condition due to the partition of alloying elements mainly between α and β phases.

The β/α duplex phase structure formed by quenching from β phase region in a Ti-10V-2Fe-3Al alloy was aged at 673 and 873 K to produce α phase particles. Since the α phase particles exhibit the morphology similar to that of α' martensite, the processes of transformation have been analysed by the crystallographic theory of martensite. At 673 K, very fine spherical ω phase particles precipitate initially and then fine α platelets in specific directions replace with them as bands. At 873 K, on the other hand, coarse monolithic α phase plates precipitate directly in the parent β phase. The orientation relationships between α and β phases in both cases are close to that of Burgers. The habit planes for both the α platelets and the α plates lie close to the (122) β and the (111) β , respectively. The results can be explained consistently in terms of Bowles and Mackenzie theory by considering the condition that the elastic stress due to the shape change is relaxed by slips within the matrix.

The formation process of L1 o γ lamellae in the hcp a phase of a Ti-4Oat.%Al alloy solution-treated at 1473 K for 86.4 ks followed by isothermal decomposition at 1273 K has been analysed in terms of morphology and crystallography. The α phase was ordered to D0 19 α 2 phase during cooling. The γ lamella with the orientation relationships, {111} γ // (0001)α 2 and a 2 , nucleated at a grain boundary γ allotriomorph / α 2 matrix interface and exhibited 90° domain relationship with the γ allotriomorph. Both sharp surface reliefs and the overlapping stacking faults with almost regularly spaced 1/6[112] γ partial dislocations on every two close-packed layers formed during the growth of the γ lamellae, inheritting the anti-phase domain boundaries (APBs) in the α 2 phase. The C-curve for the γ lamellae nucleation indicated the diffusional nature. The phase transition processes are as follows: homogeneous γ phase → homogeneous α 2 phase with thermal APBs → the formation of γ allotriomorph at grain boundary → the nucleation of γ lamella at γ allotriomorph / α 2 matrix interface. The formation of γ allotriomorph induces an elastic strain field around it due to the semicoherency at γ allotriomorph / α 2 matrix interface. The γ lamella with the 90° domain relationship to a γ allotriomorph nucleates at this interface to relax the elastic strain field. The growth of γ lamella is due to the glides of 1/3[1010] α 2 partial dislocations with short-range rearrangements between Ti and At atoms at an (0001)α 2 // {111}γ interphase boundary.

Crystallographic Analysis of Upper Bainite in Fe-9%Ni-C Alloys In order to determine the orientation relationships among ferrite, austenite and cementite in addition to the habit plane of bainite laths, bainite in Fe-9 mass%Ni-C alloys has been examined mainly by means of transmission electron microscopy. Since little austenite has been retained in the present alloys, the orientation of austenite was estimated by assuming the Kurdjumov-Sachs ferrite/austenite relationship with fixing the ferrite/cementite orientation relationship and the habit of upper bainite lath. The results obtained are as follows: An upper bainite lath is formed by the coalescence of needlelike ferrite subunits elongating in a α ∥ γ on a {111} γ plane, the resultant habit of a bainite lath being a γ {111} γ . The ferrite subunit needle exhibits parallelogram cross section. One set of the parallel needle surfaces inclines slightly from the {111} γ plane and the other set lies almost parallel to the plane expected as that for lattice invariant shear plane expected by the application of phenomenological theory of martensite. At lower temperatures, cementite platelets precipitate on the latter sets of the ferrite needle surfaces, and the coalescence of them leads to a lathlike bainite containing cementite platelets in a specific orientation. The ferrite/cementite orientation relationship is close to that of Isaichev and cementite can be related to austenite by the Pitsch relationship within the deviation less than one degree. Thus, the three phases, i.e., austenite, ferrite and cementite, are mutually almost coherent in upper bainite.

The formation mechanism of an unusual structure formed at lower temperatures in the upper C curve where eutectoid reaction occurs in a T-T-T diagram of high Cr steels has been investigated mainly by means of metallographic observations. The morphology of this structure is quite different from those of lamellar eutectoid products, but very similar to that of bainite in optical microscopic levels. TEM observation shows that this strusture was consisted of fine lamellar carbide/ferrite aggregates, and grew preferentially in edgewise directions. Dendritic growth of the eutectoid decomposition products led to the morphology similar to bainite. The interlamellar spacings of them were continuously changed with decreasing transformation temperature and exhibited the same inclination as that constructed for pearlite in plain carbon steels. The carbide particles precipitated during the formation of this structure were not cementite but alloy carbides such as M 23 C 6 . This structure, therefore, is a sort of eutectoid transformation products, and the formation mechanism is essentially diffusional and different from that of bainite where both diffusional and shear processes are thought to operate.

Isothermal transformation characteristics of Ti-6 mass% Al-4 mass%V alloy has been investigated by means of microstructural and surface relief observations. At temperatures between (P-transus and 700 o C, the nucleation of primary α-particles occurs at the grain boundaries with forming a single C-curve. Acicular structures grow subsequently after the primary α-nucleation, but they exhibit two different characteristics, i.e., the α-plates with rather flat surface markings due to the volume change at temperatures above about 930 o C and Widmanstatten laths with invariant plane strain type surface reliefs