by clicking the arrows at the side of the page, or by using the toolbar.
by clicking anywhere on the page.
by dragging the page around when zoomed in.
by clicking anywhere on the page when zoomed in.
web sites or send emails by clicking on hyperlinks.
Email this page to a friend
Search this issue
Index - jump to page or section
Archive - view past issues
Concrete In Australia : September 2014
44 Concrete in Australia Vol 40 No 3 FEATURE: CONCRETE PERFORMANCE IN FIRE Assessing the performance of concrete structures in fires José L. Torero, School of Civil Engineering, The University of Queensland The common methodology used to establish if a concrete structure has an acceptable fire performance is the fire resistance approach. This approach subjects a structural component to a standardised fire within a furnace and a failure criterion is established on the basis of a critical temperature. The critical temperature is commonly associated to the reinforcement. Numerous studies have demonstrated the differences between real fires and standardised fires, highlighted the shortcomings of establishing system performance from component behaviour and emphasised that thermal performance (per the furnace) is not necessarily consistent with structural performance thus casting doubts on the failure criteria. Furthermore, spalling has remained intractable making predictions of structural behaviour in fire always questionable. Numerous computational studies associated with performance of concrete structures in fires have appeared in literature in the last decade attempting to explicitly demonstrate the performance of concrete structures in fires. These studies develop numerical models that are generally compared against furnace test data. The models normally utilise well characterised constitutive models for concrete at high temperature obtained using standardised tests where temperature gradients are purposely avoided as well as empirically based heat transfer coefficients to characterise the furnace boundary conditions. Finally, most computations of structural systems will find ways to by-pass the need to include fire induced spalling. Criticisms towards these approaches are beginning to appear in the literature pointing that empirical heat transfer coefficients and furnace related fires are inaccurate and commonly lead to incorrect interpretation of structural behaviour, constitutive properties obtained with standard tests ignore cracking introduced by differential thermal expansion (that can be very severe in fires) and that it is not possible to ignore spalling while not fully understanding the mechanisms that induce it. This paper does not attempt to provide a literature review of all these issues, instead it summarises the reasons why these issues need to be addressed in a more rigorous manner. 1.0 INTRODUCTION At the core of a fire there is a flame or a reaction front that is effectively a combustion process, and thus is governed by the mechanisms and variables controlling combustion (Williams, 1985). The interaction between the fire and the environment determines the behaviour of the flame and nature of the combustion processes. This is commonly referred to as Fire Dynamics. An extensive introduction to the topic is provided by Drysdale (1998). As indicated by Drysdale (1998), Fire Dynamics involves a compendium of different sub-processes that start with the initiation of a fire and end with its extinction. The onset of the combustion process, i.e. ignition, in a fire is a complex process that implies not only the initiation of an exothermic reaction but also a degradation process that provides the fuel feeding the fire. In a fire it is common to have different materials involved and given the nature of the fire growth many could be involved simultaneously but others sequentially. The sequence of ignition of items in an enclosure will affect the nature of the combustion processes. Thus, ignition mechanisms set the dynamics of the fire and also are affected by the fire itself, creating a feedback loop. The process of ignition is reviewed in detail by Torero (2009). Once a material is ignited, the flame propagates over the condensed fuels by transferring sufficient heat to the fuel until a subsequent ignition occurs. This process is commonly referred to as flame spread and is described in detail by Fernandez-Pello (1995). Flame spread defines the surface area of flammable material that is delivering gaseous fuel to the combustion process. The quantity of fuel produced per unit area is the mass burning rate. The mass burning rate multiplied by the surface area determines the total amount of fuel produced. If the total amount of fuel produced is multiplied by the effective heat of combustion (energy produced by combustion per unit mass of fuel burnt), it yields the heat release rate. The heat release rate is generally considered the single most important variable to describe fire intensity (Babrauskas & Grayson, 1992). Given the nature of the environment, the oxygen supply might not be enough to consume all the fuel, thus in many cases combustion is incomplete and therefore the heat of combustion is not a material property but a function of the interactions between the environment and the fire. In these cases it is usually deemed appropriate to calculate the heat release rate as the energy produced per unit mass of oxygen consumed multiplied by the available oxygen supply. If the fire is within a compartment, smoke will accumulate in the upper regions of the compartment. Hot smoke will radiate and/or convect heat towards all surfaces. If the surfaces are flammable, remote ignition of different materials can occur. If remote ignition occurs in the lower (i.e. cold) layer then the fire tends to suddenly fill the entire compartment. This transition is generally named flashover. Before flashover, the lower layer tends to have enough oxygen to burn the pyrolysing fuel and the heat release rate is determined by the quantity of fuel generated. This period is termed pre-flashover, fire growth CIA 40-3 FINAL.indb 44 CIA 40-3 FINAL.indb 44 26/08/14 9:19 AM 26/08/14 9:19 AM