Holdbarhed af fiberarmeret beton. Eksperimentelle undersøgelser

Durability studies are carried out at the Department of Structural Engineering and Materials (BKM), DTU as part of the research project "Design Methods for Fibre Reinforced Concrete" (FRC). The project involves BKM, Dept. of Building Technology and Structural Engineering at the University of Aalborg, The Concrete Research Centre at Danish Technological Institute, Rambøll A/S, 4K-Beton A/S and NCC Denmark A/S.

Durability studies are carried out by exposing FRC-beams to combined mechanical and environmental load. Mechanical load is obtained by exposing beams to 4-point bending until a predefined crack width is reached. A test arrangement is developed. The surface crack pattern is characterized using video-scanning and digital image analysis.

As environmental load, exposure to 1) a concentrated chloride solution (16% NaCl) or 2) capillary water suction is used. Additional durability tests - freeze-thaw, water vapour diffusion, chloride permeability - and studies of pore structure are made on specimens drilled or sawed from FRC-beams after removing the mechanical load.

It is the aim to identify important mechanisms for the effect of the fibres on the durability and the pore structure based on these studies. The test programme involves three different concrete qualities (water-cement ratios). Both steel fibres (ZP) and polypropylene fibres (PP) are used in the concrete beams as well as main reinforcement.

Results of the durability tests on cracked FRC-beams are compared with results for uncracked FRC-beams and beams without fibres.A comparison of the casted concretes with and without fibres show that it is difficult to achieve a satisfactory air content in a concrete with a low water/cement ratio without increasing the amount of air entrainment in relation a corresponding concrete without fibres.

No matter the amount or type of fibres an air content of 6-7% by volume measured in the fresh state is needed to secure frost resistance. Thin section analysis show that both ZP- and PP-fibres are well mixed into the concrete. Mechanically induced cracks have a striking effect on the depth of chloride penetration into the concrete and on the amount of water that the concrete is able to absorb by capillary suction.

The tests on the other hand indicates that the size of the cracks is of almost no importance for the chloride penetration when the crack width is larger than 0.05 mm. Tests on mechanically loaded beams show that the effect of fibres increases with the concrete quality, expressed by the water/cement ratio.

The higher the quality (the lower the water/cement ratio) the more can the load level be increased when fibres are added without increasing the crack depth and the depth of chloride penetration. Differences in depth of penetration between the tension and the compression side of the beams (exposed to bending) seem to be explained by the existence of micro cracks in the tension side.

Steel fibres (ZP) seem to more effective than the tested PP-fibres as regards the prevention of chloride penetration or freeze-thaw scaling. Only steel fibres in contact with the chloride exposed surface seem to be corroding. High performance concrete with a water/cement ratio 0.32 and 7% by weight of silica fume exposed to a BS60 standardized Danish fire test at the University of Aalborg are completely cracked.

This result is not affected by the addition of fibres to the concrete and the by the fact that the concrete had been stored at 80°C in several weeks before the fire test to prevent scaling at 250°C. When the PP-fibres melt they leave pockets in the concrete. These pockets are in direct contact with the system of cracks formed in the concrete.

The compressive strength is reduced with 75-80% after a BS60 exposure. Because of the test circumstances it is not possible to confirm or reject the hypotheses that are put forward with regard to the effect of fibres on the fire resistance of concrete. The tests have shown that a BS60 exposure is to strong for this purpose.