Behaviour of multi-celled GFRP beam assembly with concrete infill: experimental and theoretical evaluations

Behaviour of multi-celled GFRP beam assembly with concrete infill: experimental and theoretical evaluations

Glass Fibre Reinforced Polymer composites (GFRP) have become an attractive construction material for civil engineering applications due to their excellent corrosion resistance, design flexibility, and high stiffness and strength-to-weight ratios. However, research related to the flexural behaviour of concrete filled GFRP tubes is very limited, especially with regards to developing high strength and lightweight composite beams. Therefore, this research project has developed and investigated the behaviour of a new composite beam, termed 'multi-celled GFRP beam', which was made by gluing together a number of pultruded GFRP tubes and then filled with low-strength concrete.

Firstly, the effective elastic properties of the pultruded GFRP tubes were evaluated by testing full-scale specimens with different shear span-to-depth ( ) ratios. The flexural ( ) and shear ( ) moduli of the GFRP tubes were then calculated using back calculation ( ) and simultaneous ( ) methods. The results showed that the method gives a more reliable elastic properties of pultruded hollow GFRP sections compared with SM and coupon tests. In addition, the full-scale test can accurately capture the local buckling failure of the compression flange of the GFRP section and the contribution of shear deformation, which is impossible to capture using a coupon specimen due to the discontinuity of the fibres. The compression buckling failure can significantly affect the ultimate failure load of GFRP sections and result in the section utilising only half of its design capacity.

Secondly, the effect of filling the pultruded GFRP single sections with concrete of different compressive strengths on the flexural behaviour was investigated. Three different compressive strengths of concrete, i.e. 10 MPa, 37 MPa, and 43.5 MPa, were used to fill the hollow pultruded tubes. These beams were then tested under 4-point static bending. The results indicated that the concrete filling improved the flexural behaviour of GFRP tubes. The beams filled with concrete of 10 MPa compressive strength showed a 100% increase in strength, while the beams filled with concrete of 43.5 MPa compressive strength exhibited a 141% increase in strength, compared to the hollow sections. However, both concrete filled beams showed an approximately similar stiffness suggesting that low-strength concrete is a practical solution to filling the GFRP tubes.

Thirdly, multi-celled GFRP beams were developed, and the flexural behaviour of this new beam concept was investigated. Beams with 1, 2, 3, and 4 cells were tested in hollow and concrete-filled configurations. From the experimental outcomes, it was found that gluing the pultruded GFRP profiles together can help stabilise the section and effectively utilise the high strength of the fibre composite materials. Moreover, the provision of a concrete core in the top cells significantly enhanced the bending strength and stiffness of the GFRP sections, due to the concrete supporting the tube walls and delaying the local buckling. Similarly, the increased number of cells in the cross-section changed the failure mode from compression buckling to bearing.

Finally, a simplified prediction equation, based on the maximum stresses of the GFRP materials and incorporating the shear span-to-depth ratio and local buckling, was developed. This model was used in a parametric study to evaluate the effect of the number of cells, shear span-to-depth ratio, and filling percentage on the flexural behaviour of the multi-celled GFRP beams. The results indicated that the increased number of cells enhanced the capacity of the hollow and concrete-filled beams. Similarly, the failure of the hollow beams will be governed by either the buckling failure of the compression flange of the top cell or the bearing failure, while the concrete filled beams will be affected by either bearing failure of the hollow cell under the filled cells or web buckling of the top-filled cell. It was also established that the multi-celled beams with a concrete infill at the top cell only would have a higher strength-to-weight ratio compared to their hollow beam counterparts. Furthermore, a failure mechanism map was developed to help identifying the possible failure mode for any combinations of GFRP tubes and concrete.

An in-depth understanding of the behaviour of multi-celled GFRP beams, with and without concrete infill, was the significant outcome of this study. Moreover, the newly developed multi-celled GFRP section, filled with a low compressive strength concrete at the top cell only, showed high potential for structural applications that need high strength, high stiffness, and lightweight characteristics.