 |
Solid Mechanics & Materials Engineering Group |
| Home | Research | Publications | People | The UTC | Seminars | Opportunities | Links |
Solid Mechanics and Materials
Engineering Group Seminars
Hilary 2010
For further information, contact Clive Siviour, Telephone
01865 (2)83473
Natural silk is an important biopolymer with huge potential as it combines superb mechanical properties with environmentally sensitive production methods. With weight-for-weight tensile strength stronger than steel and a toughness greater than Kevlar it is of no surprise that this represents the pinnacle of fibre production. At the Oxford Silk group we believe the ability to control the processing of silk, be it natural or artificial, will be key to the successful commercial application of this biologically important, high-performance biopolymer. One novel and important way to approach silk processing is through the application of rheology. On the one hand, rheological analysis of the flow properties of the silk feedstock during processing is demonstrating how applied shear energy transforms a liquid gel into a final solid fibre with bespoke structure function properties. On the other hand, the deployment of rheometry as a quality control tool allows us to improve artificial production conditions. For more information please refer to: www.oxfordsilkgroup.com
Angelos Mintzas,
University of Oxford Adhesively bonded joints consist of dissimilar material junctions and sharp geometry changes a combination which, within the context of linear elasticity, may give rise to singular stresses. Experiments conducted on a number of different joint configurations show that these bimaterial junctions are sites of failure initiation and the characterization of the singular stress field acting around them is therefore of great importance. Over the last 50 years many scientists have worked on the determination of the order of the stress singularity of such bimaterial junctions, however, this provides only qualitative information about the strength of the joint. In order to fully characterize the singular stress field, the multiplier of the singular stress term of the asymptotic solution needs to be determined. For this, both a path independent contour integral method based on Betti’s reciprocal theorem and an extrapolation method are employed in this work. The stress multiplier or else widely known as generalized stress intensity factor (H) is then used as a fracture initiation parameter in an analogous way to the use of the crack tip stress intensity factor in linear elastic fracture mechanics. Joint configurations with different geometries and isotropic as well as anisotropic adherends are analyzed and the strength predictions obtained from the application of the H-based criterion are compared to experimental results. The conditions under which the proposed criterion is valid are also examined.
This talk will review our recent works on elastic constants measurement by Resonant Ultrasound Spectroscopy (RUS). Free vibration resonance frequencies of a solid depend on its shape, size, mass density and elastic constants. RUS enables us to determine a complete set of the second-order elastic constants from inverse analysis to the resonance frequencies. In the first section of this talk, we will present some theoretical/experimental foundations of RUS: variational formulation within the framework of linear elasticity, numerical analysis by Ritz method, ultrasound spectroscopy experiments and mode identification technique. The second section includes experimental results obtained from low-symmetric crystalline materials (alpha-quartz, langasite and C12A7) and analysis based on the crystallographic group theory coupled with the lattice dynamics perturbation approach. In the last section, we shall, as time allows, address the extension of the basic theory of RUS into a general nonlinear framework. Professor David Barton, School of Mechanical Engineering, University of Leeds Processing and Properties of Highly Oriented Thermoplastic Polymers Substantial improvements in the mechanical and physical properties of thermoplastic polymers as a result of molecular orientation have initiated the development of a number of solid state thermoforming processes such as free tensile drawing, die-drawing, hydrostatic extrusion etc. Though these processes have attracted significant scientific and practical interests over the years, commercialising these processes in a normal industrial environment requires detailed investigations of the various process parameters and the properties of the oriented product. Recently detailed property evaluations and computer simulations have been used at Leeds to predict the response of polymers in solid phase thermoforming processes and also to optimise the thermoforming conditions. However successful quantitative predictions are difficult due to the sensitivity of the polymer to temperature, strain, strain-rate and hydrostatic pressure as well as to microstructural changes. This seminar will present recent results arising from studies on three classes of polymer systems that have successfully undergone solid state orientation at Leeds as follows: • polyoxymethylene (POM) for wire rope applications, looking particularly at the deformation and damage arising from the die-drawing process • wood-polypropylene (WPP) composites, including micromechanical modelling of the properties • silane-grafted & crosslinked polyethylene (PE), considering the effect of the silane grafts on the practical limits of the drawing process. In each case, the important process parameters are reviewed as well as the tensile, physical and microstructural properties of the highly oriented product. Where appropriate, theoretical and empirical models of the material behaviour are derived and compared with the experimental results. SEM, SAXS and WAXS are used to help in the interpretation of the mechanical property data. Dr Homa Hadavinia, Kingston University XRD, SEM and AFM study of ferroelctric microstructure Evolution of Damage in Ductile Materials and Simulation of Crack Propagation in Laminated Composites In majority of designs, it is desired materials remain undamaged and to be durable, i.e. do their functionality for a long time and ideally for infinite life. But there are occasions that a controlled crack growth is sought. In these cases the designer wants to grow crack(s) in the component to achieve a design objective. Therefore, a detailed knowledge of conditions under which cracks grow has a vital role either to contain and hinder crack propagation or to create suitable environment for the crack to grow in a desired path. A considerable effort has been devoted to the development of computational analysis procedures for crack growth in the materials. Several crack propagation models have been devised by different researchers with the goal to incorporate an increasing amount of physical information about the crack tip processes. Two categories can be identified for crack growth methods. Those which are based on debonding along a predefined path and others which are based on continuum damage of materials. In this seminar we will look at phenomenological approaches of ‘cohesive zone model’ and ‘critical plastic strain fracture model’ from first categories and continuum damage model based on Lemaitre damage model for the second categories. FRP composite materials are ideal for structural applications where high strength to weight and stiffness to weight ratios are required. The impact behaviour of these materials is of great importance. Low velocity impact modelling of these materials will also be discussed. Prof Harm Askes, University of Sheffield Inertia penalties in computational dynamics The simulation of engineering systems often includes the solution of a system of equations. This system normally follows from the discretisation of a set of partial differential equations. Constraints may be added to the system to account for Dirichlet boundary conditions or periodicity effects, e.g. In this talk, the focus will be on the use of penalty functions to enforce such constraints. In particular, the use of inertia penalties in computational dynamics will be discussed. The interpretation of an inertia penalty goes as follows: when a certain degree of freedom is to be fixed, a large artificial mass can be assigned to it so as to penalise acceleration of this degree of freedom. The following aspects of inertia penalties will be discussed: 1. Inertia penalties can be positive or negative. Whereas engineering intuition would suggest that penalties should be positive, it can be proven that negative penalties are as effective as positive penalties. In fact, the approximation errors of positive and negative penalties can be proven to be of opposite sign. Thus, in an incremental time stepping algorithm a significant reduction of error can be achieved by letting the sign of the penalty alternate between positive and negative. 2. In contrast to the more common stiffness-type penalties, inertia penalties have an advantage in explicit time integration algorithms: inertia penalties tend to increase the critical time step of such explicit schemes. This can be exploited in devising so-called mass-scaling algorithms to reduce CPU times of simulations. 3. Inertia penalties can also be applied simultaneously with stiffness penalties. Their combined use leads to greater accuracy than using either penalty on its own. Furthermore, certain combinations of stiffness penalties and inertia penalties have a neutral net effect on the critical time step, which can be of benefit in applications of constraint imposition and interface formulation. Modelling residual stresses and deformation at different scales: eigenstrain and beyond The concept of eigenstrain was extensively used to study the residual elastic strain distributions and their sources. A pseudo-thermal strain FE approach was used to simulate the residual stress state in 1D, 2D and 3D cases. The direct and inverse problems of eigenstrain analysis were carried out and validated by various experimental techniques. To better the understanding of the material deformation history, both polycrystal plasticity and strain gradient crystal plasticity model were employed to simulate the elasto-plastic deformation behaviour of Ti-6Al-4V alloy and pure large grain Ni. Various post-processors were developed to extract the average elastic strains, obtain the forward prediction of diffraction peaks (position, width, shape) and predict the Laue patterns with streaking spots.
Tba Tba
|