TY  - THES
AU  - De Lannoye, Karen
TI  - The tube furnace as a new bench scale experiment for pyrolysis
VL  - 72
PB  - Wuppertal
VL  - Dissertation
CY  - Jülich
M1  - FZJ-2025-03242
SN  - 978-3-95806-839-1
T2  - Schriften des Forschungszentrums Jülich IAS Series
SP  - ix, 70
PY  - 2025
N1  - Dissertation, Wuppertal, 2024
AB  - Electrical cables are a potential source of fire in many different application areas, ranging from residential buildings to spacecrafts. The current state of the art models for fire spread on a cable or cable tray are using a prescribed spread rate. Many efforts have been made within the research community, to enhance the understanding of cable fires. Nevertheless, many challenges still remain [1]. In order to better understand the influence of certain boundary conditions (e.g. composition and flow conditions of the surrounding atmosphere), a new designed set-up for gram scale pyrolysis experiments is proposed. In fire safety science, the thermogravimetric analyser (TGA) and the cone calorimeter are both often used for pyrolysis and combustion experiments. The TGA allows examining milligram-scale samples under well-defined boundary conditions, while the cone calorimeter allows for larger (gram-scale) samples but with less well known boundary conditions. Within this thesis, experiments with polymethyl methacrylate (PMMA) are conducted in the TGA and the cone calorimeter. In the TGA, the effect of several different experiment and material parameters on the pyrolysis behaviour is investigated. Including, the effect of different sample colour and different atmospheres is included in this study. In the cone calorimeter, the effect of different sample colour is studied. PMMA was chosen as a material since its fire behaviour is often studied in the literature. The newly designed experiment is based on the ISO-19700 standard for the steady state tube furnace. The set-up consists of a tube furnace which is combined with an online mass loss measurement. The advantage of the new set-up is that it provides well-defined boundary conditions, as well as radially symmetric heating. The mass loss measurement is enabled using a cantilever mechanism. This cantilever connects the specimen in the oven with a load cell outside the high temperature region of the oven. Additionally, the set-up is connected to a gas analyser with CO, CO2 and O2 measurements. Temperatures at the in- and outside of the set-up are recorded using K-type thermocouples. The functionality of the balance is demonstrated by comparing the mass loss rate of the balance with the rate obtained from the measurements by the gas analyser. A very good agreement was found between the results from the gas analyser and the balance, showing the excellent functionality of the new balance. A comparison between the new device; the TGA and the cone calorimeter is done using PMMA. A difference in mass loss rate and heat release rate was found between black and transparent PMMA in the cone calorimeter. This difference becomes larger for higher external heat fluxes. It is believed that this might be due to a difference in absorption coefficient for both materials, however additional exiperiments would have to be conducted to confirm this hypothesis. When comparing black and transparent PMMA in the tube furnace no difference in mass loss rate was found. This could indicate that the involved heat transfer process in the tube furnace is less dominated by radiation or this might be due to the lower temperatures in the tube furnace. When comparing the mass loss rate results of the tube furnace and the TGA, it was found that changing the atmosphere from nitrogen to air has a similar effect in both devices, namely the mass loss rate starts earlier under air atmosphere. Additional experiments should be considered to further compare the three devices. The scale of the tube furnace does not allow to ignore heat and mass transfer. Additionally, a temperature gradient is present throughout the length of the furnace. This makes the tube furnace a more complicated experiments than e.g. the TGA, more experiments should be conducted to further characterize the boundary conditions in the tube furnace. In future, it should be considered to make a computational model of the tube furnace which can be used to determine kinetic parameters by inverse modelling. This would allow to further evaluate the added value of the tube furnace to predicting parameters needed for fire spread modelling.
LB  - PUB:(DE-HGF)3 ; PUB:(DE-HGF)11
DO  - DOI:10.34734/FZJ-2025-03242
UR  - https://juser.fz-juelich.de/record/1044505
ER  -