Alfred Rotimi Akisanya

Initiation of fracture at the interface corner of bi-material joints

Abstract Within the context of linear elasticity, a stress singularity of the form Hr λ −1 may exist at the interface corner of a bi-material joint, where r is the radial distance from the corner, H is the stress intensity factor and λ −1 is the order of the singularity. Recent experimental results in the literature support the use of a critical value of the intensity factor H = H c as a fracture initiation criterion at the interface corner. In this paper, we examine the validity and limitations of this criterion for predicting the onset of fracture in a butt joint consisting of a thin layer of an elastic–plastic adhesive layer sandwiched between two elastic adherends. The evolution of plastic deformation at the corner is determined theoretically and by the finite element method, and the solution is compared with the extent of the elastic singular field. It is shown that H c is a valid fracture parameter if h > B ( H c / σ Y ) 1/(1− λ ) where the non-dimensional constant B =100 for β =0 and B =13 for β = α /4. Here, h is the thickness of the adhesive layer, σ Y is the uniaxial yield stress of the bulk adhesive and ( α , β ) are Dundurs’ parameters (Dundurs, J., J. Appl. Mech. 36 (1969) 650). Experimental results for aluminium/epoxy/aluminium and brass/solder/brass sandwiched joints are used to assess the role of plastic deformation on the validity of the failure criterion.

An investigation of the stress singularity near the free edge of scarf joints

The singular stress field which develops at the interface corner of a scarf joint between two bonded elastic solids is investigated. Depending upon the scarf angle and the material elastic properties, the singular stresses at a radial distance r from an interface corner may be expressed in the form Hr 1 ,w here 1 is the order of the stress singularity and H is the intensity of the singularity. The intensity H of the singularity at the interface corner of scarf joints subjected to a remote uniform tension is evaluated for various material combinations and a range of scarf angles usin g a combination of the finite element method and a path independent contour integral. Two types of scarf joints are considered: (i) a scarf joint between two long bi-material strips and (ii) a scarf joint consisting of a thin elastic layer sandwiched between two substrates. The role of the intensity H ,w hich determines the amplitude of the stress field within the singularity zone, is examined and the implications of the results for the initiation of joint fa ilure are discussed.

Finite element modelling of cold isostatic pressing

Cold isostatic compaction, where a shaped rubber bag is filled with powder, sealed and then subjected to high all-round pressure to produce a compacted green component, is a common processing route for ceramic components. The key to isostatic pressing is the design of the rubber bag which is in general both different in size and shape from the green body. This paper presents: the results of experiments to measure the powder and elastomer properties; finite element simulations of cold isostatic pressing; and comparisions between the two. The finite element simulations use an elasto-plastic, volume hardening plasticity model for the compacting powder and a finite deformation hyperelastic model for the rubber. The simulations give excellent agreement with experimental results for the pressed component shape, and highlight the importance of including the elastomeric bag within the simulations.

Stage I compaction of cylindrical particles under non-hydrostatic loading

Abstract Constitutive models are developed for stage I compaction of a two-dimensional array of cylindrical particles under general loading. Densification is assumed to occur by plastic deformation at particle contacts, and yield surfaces are constructed as functions of two state variables using slip-line field and finite element methods. The shape and size of the yield surface are found to depend upon the loading history and the relative density of the array. The effects of non-hydrostatic loading and material hardening upon the pressure-density response of the array of particles are examined by the finite element method. The numerical predictions of the pressure-density response are compared with experimental results for hydrostatic and closed-die compaction of plasticene cylinders.

Micro-mechanical modelling of powder compaction

This paper compares critically a trio of models of the compaction of granular materials in processes of industrial interest. The simplest model assumes an isotropic material with a spheroidal yield surface in principal stress space. The shape of this yield surface is constant but the size is a simple function of the volume strain. The other two models attempt to capture the anisotropic nature of compaction by assuming initially spherical granules that are deformable. One anisotropic model (kinematic) assumes an affine deformation of the centres of the spheres and gives relatively poor quantitative predictions. The other anisotropic model (static) assumes a simple approximation for the values of the contact forces and can be made to give adequate simulations of the compaction of at least some granular materials. Comparison with previously published experimental results shows that at least for some powders the history of anisotropic compaction is carried, not in the overall deformation, but in the maximum force seen by the contacts. Another important new result is that for the case of proportional loading the results of the isotropic model and the static model are in close, but not perfect agreement.

The deformation and densification of an array of metal-coated fibres

Abstract The densification of an array of metal-coated fibres (MCFs) by rate-independent plasticity is examined in this paper. The finite element method is used to simulate both hydrostatic and closed-die compaction of the MCFs. The slip-line fields of the deformation mechanisms are compared with the finite element predictions and used to address the current disagreement in the literature as to the source of the softening in the contact pressure of an array of MCFs. The accuracy of the relative density based on the current densification model of uniform redistribution of material on the free surface of a particle is also examined. We find that this assumption overestimates the relative density for both hydrostatic and closed-die compaction. The evolution of the relative density for both modes of compaction is compared for a strain-hardening metal coating. The prediction of the relative density based on J2 flow theory is consistently lower than that based on deformation theory for all values of the strain-hardening exponent considered.

Bag design in isostatic pressing

Abstract This paper presents the results of experiments and finite element simulations of the cold isostatic pressing of refractory tubes. Here a shaped rubber bag is filled with powder, sealed and subjected to high all-round pressure to produce a compacted green component with near net-shape. The stiffness of the rubber bag has an influence on the final compacted shape of the powder and this can lead to delays in developing the correct shape for the bag. The objective of this work was to predict the final shape of the compacted ceramic from the initial shape of the bag and the properties of the rubber and powder. The finite deformation simulations use a rigid-plastic, volume hardening plasticity model for the compacting powder and a hyperelastic model for the rubber. This has been implemented through a user subroutine within the commercial finite element code ABAQUS. The simulations give excellent agreement with experimental results for the pressed component shape.

Drifting Impact Oscillator With a New Model of the Progression Phase

In this paper, a new model of the progression phase of a drifting oscillator is proposed. This is to account more accurately for the penetration of an impactor through elasto-plastic solids under a combination of a static and a harmonic excitation. First, the dynamic response of the semi-infinite elasto-plastic medium subjected to repeated impacts by a rigid impactor with conical or spherical contacting surfaces is considered in order to formulate the relevant force-penetration relationship during the loading and unloading phases of the contact. These relationships are then used to develop a physical and mathematical model of a new drifting oscillator, where the time histories of the progression through the medium include both the loading and unloading phases. A nonlinear dynamic analysis of the system was performed and it confirms that the maximum progressive motion of the oscillator occurs when the system exhibits period one motion. The dynamic response for both contact geometries (conical or spherical) show a topological similarity for a range of the static loads.

The Mechanical Behavior of a 25Cr Super Duplex Stainless Steel at Elevated Temperature

Super duplex stainless steel (SDSS) is a candidate material for production tubing in oil and gas wells and subsea pipelines used to transport corrosive hydrocarbon fluids. The suitability of this material for high temperature applications is examined in this article. The uniaxial tensile properties are determined for a 25Cr SDSS over a range of temperature relevant to high pressure-high temperature oil and gas wells. It is shown that there is a significant effect of temperature on the uniaxial tensile properties. Elevated temperature was shown to reduce the Young’s modulus and increase the strain hardening index; temperature effects on these two parameters are usually neglected in the design of subsea pipelines and oil well tubulars, and this could lead to wrong predictions of the collapse pressure. The manufacturing process of the super duplex tubular did not lead to significant anisotropy in the hardness and the ultimate tensile and uniaxial yield strengths.

The yield behaviour of an array of metal-coated fibres

The biaxial compaction of a square array of metal-coated fibres (MCFs) is examined using the finite element method. The array is first compacted to a certain density using biaxial loading paths consistent with those for hydrostatic and closed-die compaction. The yield surfaces are then constructed at various stages of the densification. The size and shape of the yield surface depend on the loading history, relative density and on the fibre volume fraction. The yield surface after hydrostatic compaction has a corner at the initial loading point. However, the location of the corner on the yield surface after closed-die compaction depends on the relative density and the fibre volume fraction. The numerical results of the hydrostatic yield surface over a range of loading path are in good agreement with Fleck (1995) micro-mechanical model originally developed for the compaction of a random array of homogeneous particles.