The cooling phase, a key factor in the post-fire performance of RC columns

Abstract

The focus of recent literature is primarily on the heating phase of concrete columns in relation to possible delayed failure, in combination with a linear descending cooling phase. The research is only recently started, and this is mainly based on a parametric temperature-time curve, including a linear cooling regime as the first kind of approximation suitable for design purposes. Gernay (2019) [1] proposed a straightforward linear relationship between the burn-out resistance of a column subjected to a parametric fire and the ISO fire rating time, valid for heavily loaded columns. However, it is known that the linear cooling curve in the case of the parametric fire does not follow Newton's fundamental law of cooling in a transient situation and, and therefore, is non-natural. Exploring the importance of the cooling regime delivers the need to investigate more profoundly other slopes of, or shapes than, linear cooling, which might be closer to reality. Also, the importance of the load level should carefully be investigated. Indeed, in real buildings, not every column is heavily loaded to obtain some uniformity in dimensions. A simple column is chosen and heated on four, three, or two sides to investigate the effects of the cooling regime and the load level. A combination of thermal loads coming from the parametric fire, the more natural cooling BFD curve, a heavily forced, very slow cooling regime was created. For the material model, a proposed strength loss of 10% (EN 1994-1-2, 2005) is verified and compared against a 20% loss observed in several other experiments out of literature. Explicit formulation of creep will be used to account for the time-dependent loss in strength. Research relevance can be found in the assessment of the residual capacity of concrete structures in post-fire circumstances. This capacity can be expressed in terms of load-bearing capacity or post-fire fire rating; both are important to assess the reliability of a structure for further use after it has been subjected to a fire. It is shown that a difference in the cooling regime, load level, number of exposed sides can cause failure and do influence the post-fire bearing capacity.