PhD projects 2013 cohort

Development of a technique for detection of kissing bonds in composites

Jonathan AlstonStudent: Jonathan Alston
Supervisors: Anthony Croxford and Dmitry Ivanov

The development of low-cost bonded assembly of composite aerospace structures (UAVs for example) ideally requires an NDT method to detect the presence of poor quality, weak bonds or kissing bonds (bonds which are in contact but have no strength). Bonding for these structures is via the use of a paste adhesive. Weak bonds could typically arise, for example, from the presence of contaminants, the presence of moisture on the surface or simply a poorly cured adhesive. Such interfaces can introduce nonlinearity as a result of contact nonlinearity where an ultrasonic wave is distorted when it interacts with the interface, such as present with kissing bonds. In general, the non-linear elastic behaviour of such interfaces will result in the generation of harmonics, non-linear attenuation, resonance frequency shifts, etc. when an ultrasonic wave is incident. Some work has been undertaken into the fundamental performance of these approaches, but their ability to detect kissing bonds in these complex components is an open question. A technique that has shown potential for investigating small material changes is the non-collinear method. This approach has extremely high sensitivity and many modes of operation, however its performance and optimisation for composite structures has not been explored. This and other nonlinear methods will be investigated with comparison to traditional linear methods for their ability to detect weakened or kissing bonds. The project will require understanding and development of fabrication techniques for weak bonds.

The main difficulties with the use of non-linear methods arise because harmonic generation is small and it may be lost in the noise level within multi-layered composites. New methods based on non-collinear mixing have the potential to improve the detectability and could also have particular application for the composite structures under consideration here. Specifically the ability to fully characterise the nonlinear response of different bond types will allow changes to be directly related to the bond effects, and not a function of the overall material. Currently the simplest interaction cases have been explored, and these are not necessarily the most suitable for this type of inspection. In addition to the obvious and significant challenges on the NDT side, controlled methods need to be developed which can be used to make composite bonds with defined strengths, such as kissing bonds, or weakened bonds based on known concentrations of contaminant, moisture or poor cure.

The objective will be to develop and use the non-collinear mixing methodology for nonlinear ultrasonic detection and assessment of kissing bonds with application to paste adhesive bonded composite structures. This can be broken down into the following major tasks:

      1. Understand the performance of each different non-collinear interaction type in simple components
      2. Develop the non-collinear inspection method for composite materials
      3. Develop a model to predict and understand the non-linear ultrasonic response
      4. Compare results to current best practise in linear ultrasonics and other NDT approaches
      5. Develop repeatable well defined techniques for the manufacture of specimens with controlled bond strength
      6. Ruggedise developed techniques for industrial applications

Back to top

Functionalised graded composites for induction processing in manufacture

Giampaolo AriuStudent: Giampaolo Ariu
Supervisors: Dmitry Ivanov, Ian Hamerton, Richard Trask (University of Bath) and Paul Williams (Rolls-Royce plc)

The increasing use and provision of propulsion systems in aerospace and maritime sectors drives towards the investigation of effective through service re-manufacturing technologies. With over 50% of aircraft propulsion revenues coming from aftermarket servicing, a significant commercial opportunity exists in repairing OMC components and systems. Significant opportunities in manufacturing advanced multi-functional composites with enhanced performance together with repairing these structures with the aerospace sector as an early adopter but opportunities potential defence and marine application are expected.

The research proposal focuses on the optimisation of the manipulation of short fibres such as carbon nanotubes (CNTs) through external energy fields. These refer to electromagnetic and electric fields. The work to be developed will involve the formulation of superparamagnetic SAMs on glass particles, iron oxide-capped Janus particles/rods with one magnetic hemisphere for the assessment of short fibre manipulation methods and the development of computer modelling methods for optimisation.

Methods for the discretely ordered alignment of magnetically functionalised short conductive fibres (e.g. CNT or graphene ribbons) within low viscosity matrices using electromagnetic, electrical and other fields will be evaluated. High magnetic fields of around 1 T are currently required for the mentioned alignment due to the reinforcement diamagnetic nature. The reduction of the magnetic field used for the alignment represents an important target for the initial part of the project. Current research at ETH Zurich has also shown the potential for super para-magnetic nanoparticles to be aligned by magnetic fields of 0.8 mT.

The influence of the reinforcement type and geometry on the active localised curing is a relevant aim of the project. Optimised systems including the mentioned conductive fillers can be rapidly cured by using induction heat energy sources, with curing times of around two hours. This leads to interesting properties for rapid reuse of the repaired structure.

The first two research steps will lead to the development of computational modelling for the simulation of conductive nano-fibres alignment within an uncured carbon composite matrix for the homogeneous through thickness cure via induced eddy current. This step will be experimentally validated.

This initial research will be applied as pioneering new technology for fibre placement and control of the component curing process during on-platform repair. These repairs do not lend themselves to use of autoclaves or microwave heating; selective uniform through thickness heating is needed for polymer resin curing processes but avoiding re-heating, which can potentially compromise the original un-damaged structure.

Back to top

Synthesis and characterisation of photoresponsive thermoplastic liquid crystal elastomers

Laura BeckettStudent: Laura Beckett
Supervisors: Ian Manners, Annela Seddon, Valeska Ting and Richard Trask (University of Bath)

Self-structuring and self-organization are the driving principles of structure formation in nature, and therefore of great interest in the design of novel intelligent materials. Due to an inhomogeneous expansion/shrinking of different regions, polymer bilayer films are able to spontaneously form complex structures by wrinkling, creasing and folding. By controlling the architecture and composition of the material, the force or energy of activation can be programmed within the design phase such that the component can respond to an external stimuli, e.g. light, heat or mechanical stress, to transform flat sheets into a prescribed three-dimensional shape.

The grand challenge for this approach is the direct integration of these concepts into a 3D printing environment to exploit the design flexibility this manufacturing method allows for. The proposed research will help advance the underpinning engineering science, and demonstrate the potential of ‘bottom-up’ manufacturing of metamaterials by offering a greater understanding of novel, smart or multifunctional features and their integration within structural materials.

Back to top

Development of well-behaved nonlinear structures

Bradley CoxStudent: Bradley Cox
Supervisors: Alberto Pirrera, Paul Weaver, Rainer Groh, Daniele Avitabile (The University of Nottingham) and Giuliano Allegri (Imperial College London)

Structures that utilize composite materials intelligently are inherently efficient, however as structure utilize their constituent materials more efficiently their sensitivity to defects, impact, variability, fatigue, etc. increase significantly. The analysis of some of these problems is not currently possible, and for those that are possible the analyses is often time-consuming, complex and computationally expensive.

The long-term aim of this project is to extend the current analysis capabilities for buckling, post-buckling and other structurally-nonlinear phenomena such as fracture mechanics, crack propagation, delamination or contact problems. This will be achieved with the employment of numerical continuation; a technique commonly used to solve complex partial differential equations, but is yet to be exploited in structural problems.

Numerical continuation is a method of computing solutions of a system of parametrised nonlinear equations. The method is independent from the source of nonlinearity - it can be geometric, inertial, material, or can arise from other phenomena such as contact, damage, delamination, buckling driven delamination, etc. This numerical technique offers a number of advantages, most significantly it allows for a systematic exploration of the design space, and it makes cumbersome and expensive nonlinear calculations, often requiring onerous user intervention, relatively straightforward.

The immediate aim of this research is to devise a robust and efficient tool for the analysis of nonlinear structural phenomena such as buckling, post-buckling and crack propagation. The tool will result from the combination of numerical continuation techniques with the finite element method.

The number of potential applications for this type of tool is constantly increasing; a further understanding of morphing structures, post-buckling behaviour and damage growth, each can be enabled only by the unprecedented capabilities yielded by the proposed method.

Back to top

Crashworthiness improvements to automotive sandwich composites using tufting

Jamie HartleyStudent: Jamie Hartley
Supervisors: Carwyn Ward and Ivana Partrige  

Strict legislation on fuel efficiency and emissions has begun to force the automotive industry to focus on novel, cleaner power systems. This technology requires much lighter vehicles to be effective enough to meet the expectations of the customer and composite structures are seen as a solution. It is highly likely that such structures will be sandwich based, however it is increasingly apparent that little understanding is available to guide the design and performance requirements when loaded in edgewise failure. In fact, these structures are not currently used for this type of loading because of poor interfacial properties. Through-thickness reinforcement is an attractive solution to improve this performance. Tufting offers a localised reinforcement that can increase the adhesion between sandwich skins and core impact, stabilising skin disbonding, and contain failure. Potentially, tufting can be optimised to tailor ply-slippage, based on the impact performance requirements of the structure, allowing for an optimised structural performance and multi-functional structure development. But, predicting the mechanisms of failure of these tufted structures is presently an unknown, and so the design space remains unresolved.

The main objective of this research is to characterise and optimise the failure mode and energy absorption of composite sandwich structures for automotive crashworthiness in a rather idealised case, via simulation and mechanical performance testing as a means of validation. The core objective that will enable this to be realised is to deliver a unique design space of a sandwich structure, locally reinforced with through thickness reinforcement, that can be applied to both design and manufacturing, to not only inform on the predicted performance outcomes but also how to optimise performance. The second core objective is to consider the impact of material variation on the design space, by exploring the impacts of employing continuous to discontinuous (recycled) fibre, as well as hybrids, as these variations will no doubt have a larger future role in composites. Such variations could be in the skins as well as the reinforcing yarns and so how such variations can be optimised for crashworthiness needs careful consideration. Discontinuous fibre composite materials have shown similar stiffness values as traditional continuous fibre materials, and should offer progressive slipping at ply interfaces to absorb additional energy. Given this potential, an opportunity exists to manufacture sustainable CFRP automotive crash structures that exceed the current state-of the-art metallic or composite components.

Back to top

Developments in advanced composites skills training and manufacturing through gamification and an artificially intelligent layup agent

Shashitha KularatnaStudent: Shashitha Kularatna
Supervisors: Kevin Potter, Carwyn Ward and Stefano Cosentino (GE Aviation)

For the vast majority of companies employing composite materials there is only one primary inspection step for their product during manufacturing. At this inspection phase the decision whether the product is acceptable to the engineering standards, or if it is not and requires repair or to be ‘scrapped’, is made. However due to the nature of most manufacturing processes defects or manufacturing variations are realised well before this inspection point. This issue can result in significant process costs or loss of value.

These issues are far more pronounced in composite structures employing a range of materials (such as composite cloth and core material) and where complex geometry is required. Such forms of composite also make up a significant volume of composite parts employed over a range of industrial sectors, and as such this research will focus on them. The main objective is to be able to identify/predict at early a stage as possible potential critical defects such as wrinkles, material dis-bonds, delamination, and inclusions. Such capability should allow for reductions in final inspection time, and possibly a reduction in the number of parts requiring inspection.

The first step in meeting the objective will be to build up a thorough understanding of the current manufacturing process and inspection techniques used for a candidate composite structure. This will involve working side by side with a manufacturing team at a collaborating company in order to analyse the existing manufacturing and inspection techniques, by employing an array of multi-disciplinary research methods. Parallel to this will be a comprehensive review of literature to identify contactless inspection techniques such as chromatography, eddy current, optical, and thermal inspection methods. Following this the next step will be to essentially work backwards through the manufacturing process to ascertain how, where, and why certain defects arise, what is their impact structurally (and can this be accounted for in simulation), how their formation can be alleviated, what is the limitations in detecting them, and how these limitations can be minimised/overcome. The final step will be demonstrate the new understanding in the candidate structure, to show major advances in composite manufacturing that can deliver defect free right first time manufacture with reduction in inspection and associated cost savings.

Back to top

Development of bio-inspired cellulosic smart composites for morphing

Manu MulakkalStudent: Manu Mulakkal
Supervisors: Annela Seddon, Ian Manners, Valeska Ting and Richard Trask (University of Bath)

Materials capable of self-actuation and architectures facilitating re-configuration are highly valued and profoundly sought-after for numerous applications in the fields of robotics, deployable and morphing structures. Here, a bio-inspired approach to realising a programmable materials system capable of morphing was explored and implemented with a specific focus on the sustainability of the material components.

The deployment of manually folded paper architectures using a fluid medium as the morphing stimulus presents a simple and inexpensive methodology capable of self-actuation. The materials-based as well as the stimuli parameters of this system were found to be programmable to control the actuation response of folded paper architectures. These results confirmed the suitability of cellulose as a cost effective and sustainable smart material capable of further functionalisation, and thus justified its consideration in developing programmable morphing systems. Following nature’s inspiration, a strain-gradient mediated methodology for self-folding paper architectures was realised by locally patterning ‘fold-lines’ with a compatible hydrogel system. The architectures were designed to actuate along the principles of paper folding techniques such as origami and kirigami thus providing proven and elegant programmability and actuation attributes. As such, a novel methodology for self-folding and subsequent stimuli responsive deployment of cellulosic architectures was established.

The developments in 3D printing (3DP) technology allow the controlled placement of building materials in 3D space and this feature was harnessed to complement the strain-gradient mediated actuation of the cellulosic substrates. Therefore, a bespoke bio-inspired cellulose-hydrogel (carboxymethyl cellulose - CMC) composite for 3D printing was developed which can morph in the time domain (4D) according to the design rules developed from the previous chapters to mimic the actuation of responsive cellulosic structures observed in nature. Consequently, these responsive materials permit 4D printing - a process which pertains the transformation of 3D printed forms with respect to time. In this material system, the cellulose-hydrogel composite constitutes the programmable substrate and the CMC hydrogel acts as the localised component responsible for actuation. The strategy of drying and crosslinking following 3D printing results in a high fibre volume fraction cellulosic composites. The versatility of the materials and fabrication strategies demonstrated here enables the development of complex morphing architectures from computer aided design files with the tuneable material and structural features permitting programmability of the actuation responses. As such, this project demonstrates the realisation of sustainable cellulosic architectures capable of morphing via 4D Printing.

Back to top

Aeroelastic tailoring of composite aircraft wings with uncertainty quantification for robust and reliable design

Muhammad OthmanStudent: Muhammad Othman
Supervisors: Jonathan Cooper, Alberto Pirrera, Paul Weaver, Gustavo Silva (Embraer) and Pedro Cabral (Embraer)

Composite materials are being used increasingly in primary aerospace structure due to numbers of useful properties including high specific strength and stiffness and anisotropic behaviour. Large amount of research has been undertaken since early 1980s in the field of aeroelastic tailoring. This involves the use of anisotropic composite wing structure to enhance the aerodynamic performance and reduce loads due to gusts and manoeuvres (and hence reduce aircraft weight) through exploitation of aeroelastic deflections. Having said that, aeroelastic tailoring is not used on current aircraft designs mainly due to strict safety requirements, manufacturing constraint and also difficulties in consolidating new advanced methods into company workflow and design practice.

The traditional transport aircraft configuration is approaching its maximum utility. The aircraft industry needs to consider new aircraft configurations in order to move onto a new level of environmental friendly design. Composites materials are seen to be competent for such design. However, the number of design parameters that need to be considered, and current techniques to perform conceptual and preliminary composite tailored aircraft designs do not exist in order to permit rapid evaluation of initial configurations.

The objective of this project is to develop a 2-level aeroelastic tailoring optimization approach that can efficiently and accurately develop conceptual and detailed composite wing designs that minimise wing weight subject to stress, flutter and deflection constraints, performance (aerodynamics and weight) and also manufacturing and certification requirements. Such a methodology does not exist at the moment and detailed modelling techniques are too computationally intensive to consider radically different designs. The development of such an approach will be a major step forward in aircraft design.

Back to top

Realising the potential of carbon fibre composites in compression

Jakub RycerzStudent: Jakub Rycerz
Supervisors: Michael Wisnom and Kevin Potter

Carbon fibre composites have favourable properties under tension, but their performance is greatly hindered when subject to compressive loading. Design for low compressive strength when notched or after impact leads to heavier composite structures and less efficient use of the material. Moreover, the failure in compression tends to be sudden, with no warning before catastrophic collapse. Recent work has shown that carbon fibres themselves are both strong and very ductile in compression, achieving strains of over 15%. The significant decrease of composite properties can be attributed to shear instability driven by misalignment of the fibres and lack of restraint from the matrix, which has been shown to drastically reduce strength even for small misalignments.

The research aims to examine the parameters governing shear instability and identify ways to suppress it, either by changing the matrix properties or composite architecture. This is to be achieved by analytical and numerical modelling and experimental investigation. A review of existing literature will be conducted to systematise and quantify the underlying causes of reduced compression performance. Models will be implemented to determine which parameters can be changed in order to maximize strength and ductility. Possible solutions will be examined experimentally.

The focus of this study is on general composite application. Enhanced composites would exhibit higher stress and strain at failure, leading to overall better performing composite structures in all fields of engineering. More efficient design is especially important in aircraft applications, where small weight savings bring significant cost benefits.

Back to top

Applications for scrap material reuse in automotive structures

Emily WithersStudent: Emily Withers
Supervisors: Carwyn Ward, Ian Hamerton and James Kratz

EU and US legislation on fuel efficiency and emissions are forcing the automotive industry to move towards novel power systems and lighter vehicles. Carbon fibre composites have realistically to date been limited to super- or hyper-cars, being considered too expensive for commuter vehicles, where a greater than 95% cost reduction for their use is needed. But, the commercial success of the BMW i3 & i8 models may spur demand across the auto industry as a whole. In addition to lighter structures, composite materials offer higher Specific Energy Absorption (SEA) than metals if the correct failure mode is achieved during impact. The primary absorption mechanisms, of continuous fibre forms of composites, occurs through frictional losses at ply interfaces and deformation of the structure. Currently, structures require more ductility during impact, but must offer high stiffness in regular operation in order to meet the public’s driving performance and comfort expectations.

A promising solution towards these rather disparate aims involves incorporating scrap material into the structure. An estimated 40-60% of virgin continuous carbon fibre is unused during production of automotive components. This material could be recycled into a short fibre mat and used in non-structural applications, such as the roof panels on the BMW i3/i8, but such reclaimed materials use is not value added and underestimates the materials potential. It is possible that the off-cuts could be re-used in a more aligned form and directly incorporated into energy absorbing structures. Re-using off-cut material in this way would be an ideal solution, providing additional ply interfaces to absorb energy during an impact and saving on costly energy-intensive recycling processes.

This work seeks to research this opportunity in terms of side crash structures, and takes reference from recent works where it was demonstrated that SEA improvements for continuous fibre sandwich panels was possible with through-thickness (tufting) reinforcement. Tufting offers a localised reinforcement that increases adhesion between sandwich skins and core impact, stabilising skin disbonding and containing failure. But the effect of tufting off-cut fibre panels is unknown and untried. The main challenge of this research is ‘can scrap dry cloth material, in co-operation with through-thickness reinforcements and sandwich cores, be utilised in a high-performance structure that will successfully challenge conventional automobile design for future vehicles?’

The main objective will be to characterise and optimise the failure mode & energy absorption of composite sandwich structures for automotive crashworthiness, and will be achieved by using a combination of techniques. The first phase will be experimental, in-order to explore the practical aspects of dry cloth material re-use. The second phase will be numerical simulation, in order to optimise the failure of discontinuous or hybrid continuous-discontinuous composite structures. The research will aim to study the following:

      • Scrap material handling and performance in either ply waste or fabric manufacture form
      • Optimal ply overlap length for energy absorption in monolithic skins and sandwiches
      • Influence of through-thickness reinforcement on ply-slippage/energy absorption
      • Hybrid sandwich panel developments featuring continuous and discontinuous plies or recycled fibre mats
      • Elemental mechanical testing leading to numerical simulation of dynamic impact
      • Cost-performance-sustainability relationships for different material combinations

This research project is well suited to the needs of the automotive industry, and will show how to:

      1. Reduce the carbon footprint and improve material efficiency of future structures
      2. Increase energy absorption to protect passengers and high voltage power systems
      3. Apply novel technologies to optimise the failure mode of impact structures and demonstrate the value (cost/performance impact) of smart composite design

Back to top

Behaviour of pseudo-ductile thin-ply angle-ply laminates under different loading conditions

Xun WuStudent: Xun Wu
Supervisors: Michael Wisnom, Ian Bond and Jonathan Fuller

Carbon fibre reinforced polymer composites show a number of excellent specific properties, but are limited by their brittleness. Achieving ductility in composites could ensure their safe use in critical structures by avoiding catastrophic failure. At the moment, several approaches have been introduced to achieve ductile carbon/epoxy composites such as introducing novel constituents or by designing a novel laminate architecture.

The aim of this project is to overcome the inherent brittle failure of CFRP laminates in structural applications using the thin-ply angle-ply concept. Compared with traditional normal thick prepregs, thin ply has shown the ability of suppressing premature failure such as edge delamination and resulting in an improved strain behaviour. Previous studies achieved pseudo-ductility under uniaxial tension loading by using thin-ply angle-ply laminates, but their structural behaviour under more complex load cases is still unknown. It is envisaged that through a large range of thin-ply and different angle coupling, an optimized design can be made for different structural requirements. In this project, the experimental investigation and failure characterization will be carried out on various structural behaviours and loading conditions, such as notched behaviour, single-lap joints and static indentation etc. Finally, finite element model will be developed to predict the structural behaviour.

Back to top

Development of a composite tape placement technology with advanced fibre steering capability

Evangelos ZympeloudisStudent: Evangelos Zympeloudis
Supervisors: Byung Chul Kim, Kevin Potter and Paul Weaver

The evolution of robotic control technology has enabled automated material placement machines to achieve high productivity and reliable operation. The need for more complex products, however, has placed a lot of attention on the tow steering capabilities of modern machines. This is because, in order to lay-up on curved moulds and maintain a constant fibre orientation, the fibre trajectories have to be steered. Tow steering is also responsible for the expansion of the design space for composites, by enabling the production of Variable Angle Tow (VAT) laminates, where the stiffness can be tailored at every point. Current state-of-the-art material placement machines rely on in-plane bending of the tows, in order to steer the fibre paths, leading to fibre buckling and extensive defects. A novel concept, Continuous Tow Shearing (CTS), was developed in the University of Bristol, where during the placement process the tow is sheared to steer its trajectory. The result is a minimum steering radius, which is an order of magnitude less than that of any existing material placement machine. In order to satisfy the productivity demands of the industry, CMTS (Continuous Multi Tow Shearing) was developed, which can place multiple tows simultaneously. One of the main advantages of the CMTS technology is that the width of the material does not affect the minimum steering radius, by breaking the coupling between steering capability and productivity. The Technology Readiness Level (TRL) of the CMTS is currently 2, as the concept has been presented and validated through a prototype.

The aim of this project is to further develop the CMTS technology to enable high-volume production of large-scaled complex composite structures without process-induced defects. In order to fully exploit the advantages of the CMTS process for lay-up on 3D curved surfaces, design modification of the head mechanism, as well as development of a head control algorithm will be developed. Furthermore, feasibility of producing preforms for the OOA curing process, using an alternative binder application mechanism, will be studied. To achieve the above, the material responses of carbon fabrics and binder materials will be investigated and the placement conditions will be optimised, based on the material characteristics. The head mechanism will be modified to ensure uniform compaction pressure on a curved surface for 3D applications. An algorithm will be developed to control the devices mounted on the placement head itself, as well as the CNC or robot platform, for accurate placement on 3D surfaces. Process reliability and consistency in quality will be demonstrated through layup tests and NDT of the manufactured parts. Whether the manufactured product can meet the designed performance totally depends on layup accuracy and defects included during the layup process, as well as the curing process.

The ultimate goal of this research is to increase the TRL of this technology in order to make it ready for industrial application. By achieving the above objectives, a significant contribution will be made to the current state of the art in material placement technology. The potential applications target primarily the aerospace industry, where many of the components are large in size and slightly curved. This technology can be applied, nonetheless in many fields, such as the wind turbine and the marine industry.

Back to top

Edit this page