Investigation of an Intumescent Coating System in Pilot and Laboratory-scale Furnaces

Steel is an incombustible substrate, but at elevated temperatures structural steel suffers from a drastic reduction in mechanical strength. In the event of a fire, the reduced strength may lead to collapse of the structure. A method to prolong the time before steel reaches the critical temperature (450 - 600 °C), at which the collapse may occur, is the use of a fire protective intumescent coating, which swells when exposed to temperatures above about 200 °C. The swelling of the intumescent coating happens according to a complex sequence of chemical reactions, whereby the coating forms a porous char, which thermally insulates the substrate. In addition to the coating itself, several process parameters influence the performance of the intumescent coating. Such parameters may for instance be the interaction with an underlying anticorrosive primer, the heating rate employed, or the oxygen content in the fire. In this work, focus has been on process parameters for an intumescent coating for so-called cellulosic fires.
The thesis contains five chapters, where Chapter 1 is a literature survey providing background knowledge on coatings, intumescent coatings in particular, and fire scenarios. In Chapter 2, the effects of coating thickness and gas-phase oxygen concentration on two epoxy primers used in an intumescent coating system were investigated. It was found that primers with a too high thickness failed in the presence of oxygen. In nitrogen, the primer did well, except for a single case, which showed a minor delamination at the edges. In addition, it was shown that the thermogravimetric behavior of the primer and intumescent coating alone could not be used for explaining the entire coating system performance. A novel experimental method, which may potentially be developed into a fast screening method of primers for intumescent coatings, is also described. Upon heating in nitrogen, a color change of the primer from red to black was observed. Potentially, this may be used as an indicator to whether a primer under an intumescent coating has been exposed to oxygen or not in gas-fired furnace experiments.
In Chapter 3, a mathematical model of an intumescent coating exposed to heating in a pilot-scale gas-fired furnace is presented. The model takes into account convective heat transfer to the char surface, conduction inside the char, and the char expansion rate. Model validation was done against experimental char expansion rates and temperatures of the steel substrate and at intra-char positions. The model was solved in a discretized and non-discretized version and a good qualitative description of the temperature curves was found. An important learning was that temperatures measured inside the char are very important for a proper model validation. Due to its simplicity and few input parameters, the model (non-discretized version) shows a good potential as a practically applicable engineering model. Results suggest that oxygen mass transport is not a limiting factor for the char oxidation reactions. An investigation of the repeatability of the experimental temperatures showed that temperatures close to the char surface were somewhat more uncertain than the steel temperature and char temperatures close to the steel substrate.
Chapters 4 and 5 are concerned with the development of a fast screening method for the extent of expansion and char strength of intumescent coatings. The method is relevant for investigation of special cases, where the char is damaged by moving objects during a fire. The method uses the concept of shock heating to avoid long heating up and cooling down times of a furnace. In Chapter 4, it was found that for measuring char strength reliably at room temperature, dried samples were required. Chapter 5 discusses shock heating in various oxygen concentrations and verified that the expansion is affected by the gas composition. Experimental data showed that under a high heating rate, the char strength could not meaningfully be correlated to the degree of expansion. Furthermore, it was found that at the high heating rates employed thin films (147 µm) would contract horizontally while expanding vertically. This was not the case with a coating thickness of 598 µm. The strength of the char in the vertical direction was also investigated. It was found that the outer crust of the char had the highest mechanical strength and a weak zone, in the central region of the char, was identified.