Anita Buxton obtained an honours degree in Materials Technology from the University of Surrey and was subsequently awarded a PhD on 'Interfacial failure phenomena in carbon/epoxy composites' from the University of Sydney. She joined TWI in 2000 having sailed a two person yacht from Australia to England. As a Senior Project Leader in the Polymer Technology Section, Anita led a range of projects on adhesive technology and welding process selection. She is now pursuing her interest in finding novel solutions to difficult joining issues within the Electron Beam Section. Specifically she is involved in the development and promotion of a novel electron beam surface processing technique.
Composite materials are increasingly being used in the manufacture of cars, aeroplanes and other products where high strength, low weight materials offer benefits. But, what happens when an aircraft goes out of service? The disposal of the composite becomes a significant problem. The current solution is to put it to landfill. But as Anita Buxton reports this not only represents an environmentally unfriendly option, it is also costly to the aircraft manufacturers.
TWI has been working with Hodkin and Jones on an idea which could revolutionise the disposal of composite waste.
Concern over the increasing quantities of waste composite material that is being generated led TWI to investigate the feasibility of using discarded composite products in the reinforcement of composite structures. The benefits are threefold. Not only does it serve to prevent the landfill of bulky waste, it also makes use of the superior stiffness and corrosion resistance of composite laminates in applications where steel reinforcement has perhaps proved inadequate. Since a waste material is used to add value to products increasing costs associated with steel reinforcement are avoided. A feasibility study was performed to determine the potential of the concept.
Objective
To assess the feasibility of using strips of composite as a replacement for steel bar in the reinforcement of a concrete drainage channel lid.
Product
Following discussions between Hodkin and Jones and TWI, it was decided that the Decathlon range of high performance drainage systems manufactured by Hodkin and Jones would make a good application with which to test the concept. These drainage systems are commonly found incorporated into the paving of areas such as supermarket car parks. They consist of a concrete channel, which is embedded into the ground, and a lid containing drainage slots which sits over the channel. The complete system is shown in Fig.1. The range is produced in five channel sizes in metre lengths. The lid of the largest system, the Decathlon 375, was selected as a demonstrator product for the case study.
Manufacture
The current lid is manufactured from concrete containing a welded steel bar reinforcement cage, see Fig.2. The cost of the steel frame is estimated to be in the region of £12 per lid. The composite reinforced lid was manufactured using the same route, and was cast upside down in a steel mould. However, the steel frame was replaced with a reinforcement matrix constructed from a composite material.
A carbon-fibre/epoxy composite panel of 6mm thickness was generously donated to the project by Airbus. The panel was cut into strips 60mm wide into which slots were cut. Two long strips ( Fig.3a) were designed to run the length of the lid with the other strips ( Fig.3b) forming bridges between them. Shorter bridging strips ( Fig.3c) were required for either end of the assembly. A frame was constructed from the individual components as shown in Fig.4. The frame was designed to provide maximum reinforcement whilst fitting around the slots in the lid and priority was placed on having a frame that could be easily manipulated into the mould.
Whilst cutting composite materials is not an efficient process, at this initial stage, it was decided that an engineered frame would provide a controlled experiment. In practise a more random form of reinforcement may be viable.
The drainage channel lid was cast at Hodkin and Jones's Sheffield facility. The reinforcement frame was placed in the mould with 25mm spacers to prevent the reinforcement from showing in the cast product ( Fig.5). Self-compacting concrete was poured into the mould as shown in Fig.6.
Initially the reinforcement tended to float to the surface and had to be held down with strips of wood. The lid was removed from the mould after 24 hours and stored for a further month to allow the concrete to reach its 28 day strength. The final product looked identical to a standard product.
Testing
The composite reinforced drainage channel lid was tested at TWI using an industry standard punch test. The lid was supported over a 100mm width along each edge to replicate the loading it undergoes during service. It was subjected to a punching load via a 250mm diameter disc at a rate of 1mm/min. The test set-up is shown in Fig.7. The load and displacement were recorded over a period of 20 minutes during the progressive failure of the lid. The mechanism and position of failure were monitored throughout.
The resulting load/displacement is shown in Fig.8. It can be seen that the product supported a maximum load of 18200 kg (18.2 tonnes). The jagged nature of the curve demonstrates the incremental nature of the failure process. This correlates with observations made during the test. The first visual signs of failure occurred close to the supporting beams at either end of the lid ( Fig.9). This was followed by the appearance of a crack in the concrete below the punch. It was not until later in the test that the punch penetrated the lid ( Fig.10).
Discussion
The lid supported a load of 18.2 tonnes. This shows promise compared to the steel reinforced version of the Decathlon 375 system which handles a load of 33 tonnes. Although the composite reinforced lid did not initially match the performance of its counterpart, it only represents the first iteration in a development process. Examination of the failure process has led to a number of suggestions as to how the trial lid could be improved.
There was no sign of tensile failure in the concrete i.e. no cracking on the underside of the concrete was observed, which demonstrates the reinforcement's considerable capability within the composite. It appears that the cracks that emerged at the corners of the lid on the side face, were a result of the inadequate length of the test support rails, and therefore a function of the test set-up rather than lid construction. However, the cracks that were seen on the end face of the lid ( Fig.9) appeared to coincide with the position of the longitudinal strips of reinforcement. Consequently, the reinforcement was not taking the bending load. It is expected that better performance (in terms of load capacity) would be achieved by ensuring that the longitudinal struts of the reinforcement were positioned over the loading supports.
Examination of the underside of the lid at the completion of the test revealed that the reinforcement was pushed out of the base of the concrete casting ( Fig.11). The transverse struts below the loading disc appeared to have failed in shear, with cracks extending from the pre-cut slots. These struts were then pushed beyond the longitudinal struts and out through the base of the concrete. Measurements indicated that the reinforcement had been only 12mm from the bottom of the casting.
It is suggested that reducing the height of the composite strips to get a greater coverage of concrete would be beneficial in addressing this issue. A cover of at least 25mm of concrete is achieved in the steel reinforced lid. It is thought that the minimal cover obtained with the composite reinforcement was a result of the reinforcement floating to the surface during the casting process, a problem that could be overcome through clamping. Since the composite reinforcement showed no signs of failing in bending, it is expected that a reduced strip height would provide sufficient strength.
Another method of improving the structure would be to improve the transfer of stress from the reinforcement to the concrete. Fig.11 shows clearly that the composite panel that has been pushed out of the lid is smooth. One way of improving the stress transfer would be to improve the bond between the two, perhaps through creating a rougher, better keyed surface.
Acknowledgement
TWI and Hodkin and Jones were able to pursue this innovative application through a feasibility study under the Joining Forces Yorkshire and Humber programme, funded by Yorkshire Forward and Objective 1.
