Hidefumi Date

Microstructure and bonding strength of impact-welded aluminium–stainless steel joints

Abstract An aluminium projectile was impact-welded on a stainless steel target using a nitrogen gas gun at an impact velocity over 250 m s−1. Effects of surface roughness of the impact face of the target on the bonding area and the strength of the bonding area were examined. The microstructure and element distribution in the joint were analyzed and the experimental results of concentrations of the elements in the compound layer at the interface were compared with the theoretical values. The following results were obtained. It was clarified that the bonding strength of the area which increased with the decrease of surface roughness was much lower than that of the region which was independent of surface roughness. The thickness of the compound layer formed at the interface increased with the impact velocity. The experimental results of the concentrations of the elements in the layer hardly depended on impact velocity and were close to the theoretical results in which equivalent heat was generated at impact faces of the projectile and the target.

Estimation of Mechanical Properties of Al/Cu Compound Layer Formed by Impact Welding

The impact welding of aluminum onto copper was carried out using a gas gun and the mechanical properties of Al/Cu joint were investigated by tensile tests and micro hardness indentation tests. The strength measured by tensile test decreased with increasing of the impact velocity. The results of the tensile test suggested that it was necessary to make a microscopic survey of the joint interface. Then, the inverse analysis with FEM analysis was applied to the load and depth curves measured by the indentation technique to identify the material constants in the constitutive equations of aluminum, copper and the compound layer. In addition, the numerical simulation for the tensile test was carried out using the identified material constants of aluminum, copper and the compound layer. The nominal stress-strain curve of the compound layer obtained by the numerical simulation showed the typical feature of brittle material. The ultimate tensile stress of the compound layer was about 1.2 GPa and ten times larger than that of aluminum. It was concluded that the bonding strength of Al/Cu joint was dependent on the integrity of the compound layer.

Simple Estimation of Deformation-Induced Martensite in Stainless Steel

To estimate the volume fraction of martensite induced in 304ss type austenitic stainless steel during tensile deformation, the electric resistance of the specimen was measured using the four-point-probes method at the temperatures of 77, 196 and 293K during the deformation. The magnetic force of the deformed specimen was also measured using a permanent magnet to determine the strain at which the martensite was induced initially in the specimen. The parallelepiped model was suggested to separate the effects of the deformation and transformation on the electric resistivity because the resistivity was influenced by the evolution of the martensite and the growth of the defect in the matrix at a constant temperature. The parallelepiped model consisted of m columns with n pieces of the cubic element and was assumed to be a group of small electric resistors. The volume fraction of the martensite estimated using the measured resistivity and the model was compared with the experimental results reported by other researchers and then it was clarified that the volume fraction of the martensite estimated by the model was in agreement with the volume fraction measured by the experiment.

Effect of Hydrogen Absorption on the Mechanical Properties of Palladium

The mechanical properties of palladium (Pd) wire absorbed hydrogen were evaluated by the quasi-static tension test and indentation test. The electrolytic method was used for hydrogen absorption. Pd wire with a diameter of 1mm was used. The electrolyte was a sulfuric acid solution and the current density used in the electrolytic method was 200mAcm-2. The hydrogen absorption ratio defined by the molecular ratio (H/Pd) of hydrogen and palladium was controlled by the absorption time. The gauge length for the tension test was 20mm. The ultimate tensile strength increased with the increase of the absorption ratio. On the other hand, the increase of the ratio decreased the strain hardening parameter and fracture strain. A model considering the specimen absorbed hydrogen and a composite material constructed in a concentric configuration was suggested to estimate the hydrogen absorption area and mechanical properties. The indentation test was conducted to clarify the evolution of the embrittlement due to the hydrogen absorption microscopically and determine the absorption area precisely. Vickers hardness clearly increased with the increase of the hydrogen absorption ratio. The hardness distribution was measured to detect the boundary of the absorption and non-absorption area using a Berkovich indenter that is smaller than a Vickers indenter. The hardness boundary of the absorption and non-absorption of the specimen with the hydrogen absorption ratio of 22 percent was observed experimentally at the position around 100-150μm from the outside of the specimen. The position of the boundary estimated using the model was 85μm from the outside of the specimen. When the stress-strain curves of the specimen with the unknown hydrogen absorption ratio were measured, the hydrogen absorption ratio could be estimated using the proposed model.

Effects of Deformation Mode on Resistivity Curve Used for Estimating Martensite Induced in Stainless Steel

The martensite induced in three types of austenitic stainless steel, which indicate the different stability of the austenitic phase (γ), were estimated by the resistivity measured during the tensile deformation or compressive deformation at the temperatures 77, 187 and 293 K. The resistivity curves were strongly dependent on the deformation mode. The volume fraction of the martensite (α’) was also affected by the deformation mode. The ε phase, which is the precursor of the martensite and is induced from the commencement of the deformation, decreased the resistivity. However, lots of defects generated by the deformation-induced martensite increased the resistivity. The experimental facts and the results shown by the modified parallelepiped model suggested a complicated transformation process depending on each deformation mode. The results shown by the model also suggested a linear relation between the resistivity and the martensite volume at the region of the martensite formation. The fact denoted that the resistivity is mostly not controlled by the austenite, ε phase and martensite, but by the defects induced due to the deformation-induced martensite.

Evaluation of compound layer formed by impact welding using phase transformation technique

When a cylindrical projectile is impact-welded to a flat target, a compound layer is usually observed at the joining interface as a result of the impact welding. In this study, the formation process of the compound layer was formulated as a moving boundary problem, which is a phase transformation technique. The numerical results were compared with the experiment results obtained using an aluminum projectile and stainless steel target. Numerical analysis shows that the melting area is similar to the temperature profile given at the boundary face. The area of the compound layer formed at the joining interface almost agrees with the melting area of the target. The profile of the compound layer is similar to the triangular temperature profile in the given temperature profiles. The mixing ratio of the melting weights of aluminum and stainless steel obtained by the numerical analysis strongly depends on the temperature rise at the interface. The melted weight of aluminum in the experiment is somewhat greater than that in the numerical analysis. The heat conduction analysis including deformation of the projectile and target make the results of the numerical analysis closer to the experimental results.

Strain Rate Effects on Strain-Induced Martensitic Transformation and Electrical Resistivity of Deforming Stainless Steel

Austenitic stainless steel was compressed at a strain rates of 103 s-1 using a Hopkinson pressure bar apparatus at temperatures of 77 K and 293 K. The electrical resistivity was measured to determine the volume fraction of martensite of a deforming specimen. A compressive specimen of the dumbbell type was suitable for attaching the lead-in wires of four-point probes to the specimen. The volume fraction of martensite formed at a strain rate of 103 s-1 was lower than that formed at a low strain rate regardless of the temperature, and the effect of the strain rate on the electrical resistivity was slight. However, since the volume fraction of martensite is expressed as a linear function of the electrical resistivity ratio as well as in the results obtained by the tensile test, the electrical resistivity was available as an index for estimating the volume fraction of martensite induced by dynamic deformation. The duration of the input wave was approximately 150 μs, and the appearance of the peak value of transient resistivity was approximately 1ms after the arrival of the input wave at the specimen. These results showed that the structure change evaluated using electrical resistivity was not completed in the time required for the stress wave to pass through the specimen, although the electrical resistivity immediately after dynamic deformation closely approached that obtained by the static test.

Effect of Strain Rates on the Transformation Behavior of Ni-Ti Alloy

In order to clarify the effect of strain rates on phase transformation behaviors of Ni-Ti alloy, a compressive test using a cylindrical specimen of polycrystalline Ni-Ti alloy of Ti-50.69 at% Ni was carried out at a high strain rate and a low strain rate. The transformation temperatures were determined by a differential scanning calorimeter (DSC) using a sample cut from a compressed specimen. The transformation temperatures of the specimens before deformation were Ms= 303 K, Mf = 287 K, As = 297 K and Af = 319 K, respectively. The compressive test was carried out using specimen heated from liquid nitrogen temperature to room temperature. A universal testing machine as a static test apparatus and a Split Hopkinson Bar apparatus for a dynamic test were used. The specimen had a reoriented martensite phase after deformation because the superelastic effect was not observed upon unloading. Two reverse transformations during heating and a forward transformation during cooling were observed by DSC measurement. The first reverse transformation corresponds to that of thermal-induced martensite by immersion in liquid nitrogen and the second reverse transformation corresponds to that of reoriented martensite with slips in a polycrystalline matrix introduced by plastic deformation. The reverse transformation of the martensite phase with a slip exhibited strong strain rate dependency. Plastic strains and strain rate had strong influence on the shape recovery. The interaction between the temperature elevation by a conversion of plastic work and slip generated by dynamic plastic deformation is a complicated problem.

Deformation Process at Impact Face of Lead Projectile during Normal Impact on Target.

When a cylindrical projectile is collided normally with a target at high velocity, plastic deformation and/or failure occur on both bodies. If the relative impact velocity between both bodies is not high, a sharrow crater appears on the impact face of the target and a mushrooming profile on the projectile after impact. However, when the relative velocity is very high and the deformation resistance of the projectile is much lower than that of the target, the deformation of the impact face of the projectile becomes large and the outer part of the face is separated from the target by bending moment. Therefore, it is considered that the deformation process at the impact face of projectile is complicated.In this study, a lead projectile of 11mm dia., which has lower deformation resistance than that of the target, was collided normally with a stainless or polycarbonate target, and the deformation process at the impact face of projectile was recorded by an image converter camera from the side and the front. Additionally, the process was also examined based on the profile of craters on both targets and the distribution of the ring cracks on the polycarbonate target. The following results were obtained.The region from the center to about 7mm dia. of the projectile during impact was loaded compressively, and the projectile material flowed radially in the region from about 7mm to 14mm dia. In the outer region from 14mm, the material of the projectile was bent and separated from the target. These results show that the impact face of the projectile during impact can be characterized by three different regions described above.