Behaviour of steel reinforcement in composite slabs at elevated temperatures

Abstract

This thesis investigates the potential rupture of modern deformed steel reinforcement around the boundaries of composite or reinforced concrete slabs with continuity over perimeter supporting beams as tensile membrane action develops in fire scenarios. The potential for rupture of modern deformed reinforcement in these conditions is not known from previous research but it is critical to the integrity of reinforced concrete and composite floor structures in fire scenarios. The primary aim of this research is to determine whether modern deformed reinforcement will be more prone to rupture than plain round reinforcement at hogging cracks which form around the perimeter of a slab during the development of tensile membrane action in a fire scenario.

Reinforcement bond has been experimentally investigated at elevated temperatures because bond is one of the main mechanisms which must be understood in order to understand the rupture of the reinforcement at crack locations. Pull-out tests have been conducted at temperatures up to approximately 600˚C on plain round and deformed reinforcement of different diameters. This temperature range is expected to cover the range of bond temperatures which are realistic or applicable from a structural fire design perspective. The response of bond at elevated temperatures has been characterised in terms of stress and slip. A modified local bond stress-slip model which is applicable at elevated temperatures has been developed.

A finite element model has been developed to represent a region of reinforced concrete slab subject to flexural hogging with severe crack development. The purpose of the modelling is to investigate reinforcement behaviour and potential rupture when tensile membrane action develops at elevated temperatures. The model incorporates the stress-slip bond response from the results of the pull-out tests.

It was found that slabs reinforced with deformed bars will perform similarly well to slabs reinforced with plain round bars in facilitating the cracking and rotation which occurs around the perimeter of a slab during the formation of tensile membrane action in a fire scenario. Increased slab depth was found to substantially reduce the ability of a slab to facilitate the cracking and rotation which occurs in this scenario. The effects of bar diameter and temperature were also investigated. It was found that the assumption of perfect bond, with no slip, was appropriately, and not excessively, conservative when modelling hogging reinforcement in this scenario. The implications of these research outcomes for reinforced concrete and composite slab design have been explained.