Fire properties of surrogate refuse-derived fuels

Abstract

SUMMARYThis paper investigates the fundamental fire properties of surrogate refuse-derived fuels (RDF), a class of multicomponent materials characterized by high void fraction, with particles of polydisperse sizes and significant internal porosity. A surrogate RDF was developed to improve the reproducibility of experimental measurements. This surrogate RDF reflects typical municipal solid waste collected in the city of Newcastle, in the state of New South Wales in Australia. The material consists of shredded newspaper, wood, grass and plastic bags, with small amounts of sugar and bread. About 95% of the material passes through 50 mm square screens, as required by ASTM E828 standard for RDF-3 specification. The experiments presented in this paper were performed with the components of the RDF dried in a forced-air oven at 1038 C, except for grass which was dried under nitrogen. The material was found to be very hygroscopic, requiring special care in handling. The experiments performed in the cone calorimeter were designed to measure the heat release rate, total heat release, time to ignition, time to extinction, effective heat of combustion and formation of CO during the combustion process, as a function of sample thickness, sample density and the magnitude of the imposed radiative heat flux. The thermophysical properties of the surrogate material were either measured (solid density, void space, particle density, particle porosity) or extracted from the published data (heat capacity). The present surrogate RDF material was found to ignite easily, within a few seconds of the imposition of the incident heat flux of 40 kW m À2 , and then to reach rapidly the peak heat release rate of 110-165 kW m À2 . The deduced values of the critical heat flux, pyrolysis temperature and effective thermal conductivity are 9-10 ( AE 2) kW m , respectively, depending on the material density. The effective heat of combustion of the RDF was estimated as 15.3 MJ kg À1 . The material produced 1 kg of CO per 18 kg of dried RDF, mostly during smouldering phase after the extinguishment of the flaming combustion. These results indicate that dried RDF pose significant fire risks, requiring that fire safety systems be implemented in facilities handling RDF.

Fire Properties of Refuse Derived Fuels: Measurements of Temperature Profiles and Mass Loss

“…During the pre-ignition period, a steep temperature profile develops in the material near its top boundary, although within the sample itself the temperature remains unchanged. The pre-ignition measurements allow one to confirm the thermally-thick approximation of our previous analysis [2], to provide a complete picture of heat transfer in RDF samples prior to ignition.…”

“…About 95% of the material passes through 50 mm square screens, as required by ASTM E828 standard for RDF-3 specification. Reference [2] contains detailed description of the composition and preparation of the surrogate RDF.…”

“…The material can ignite within a few seconds, especially at higher heat fluxes, such as at 40 kW m -2 [2]. The fire grows rapidly to attain the peak heat release rate of 110-165 kW m -2 .…”

“…The purpose of this contribution is to examine (i) the temperature profiles within beds of non-densified RDF, and (ii) mass changes of RDF samples as a result of pyrolysis and char oxidation. The former aims to confirm the assumptions of an earlier ignition model [2], whereas the latter serves to obtain a critical mass flux of pyrolysates produced by the burning material at extinction of the flaming combustion. The measurements were performed in mass-loss (part i) and cone (part ii) calorimeters, with major modifications performed to both instruments, including adaptations of the sample holder and the data acquisition system.…”

“…The temperature results are complemented by measurements of the mass-loss rate, which reflect the volume-average rate of pyrolysis and char oxidation in the RDF samples. These measurements are performed for three sample densities (consistently with reference [2]), but only for the samples of 40 mm in thickness, to circumvent laboratory difficulties encountered in experiments with thinner samples. The mass-loss measurements are converted to mass-flux data by differentiating the mass-loss curves and then dividing the results by the area of the opening of the sample holder.…”