From Cells to Residues: Flame-Retarded Rigid Polyurethane Foams

ABSTRACT

Rigid polyurethane foams (RPUFs) exhibit short times to ignition as well as rapid flame spread and are therefore considered to be hazardous materials. This paper focuses on the fire phenomena of RPUFs, which were investigated through a multimethodological approach. Water-blown polyurethane (PUR) foams without flame retardants (FRs) as well as water-blown PUR foams containing triethyl phosphate as a gas phase-active FR were examined. The aim of this study is to clarify the influence of the FR on the fire phenomena during combustion of the foams. Additionally, materials’ densities were varied to range from 30 to 100 kg/m³. Thermophysical properties were studied by means of thermogravimetry; fire behavior and flammability were investigated via cone calorimeter and limiting oxygen index, respectively. During the cone calorimeter test, the temperature development inside the burning specimens was monitored with thermocouples, and cross sections of quenched specimens were examined visually, giving insight into the morphological changes during combustion. The present paper delivers a comprehensive study, illuminating phenomena occurring during foam combustion and the influence of a FR active in the gas phase. The superior fire performance of flame-retarded PUR foams was found to be based on flame inhibition, and on increased char yield leading to a more effective protective layer. It was proven that in-depth absorption of radiation is a significant factor for estimation of time to ignition. Cross sections investigated with the electron scanning microscope exhibited a pyrolysis front with an intact foam structure underneath. The measurement of temperature development inside burning specimens implied a shift of burning behavior towards that of non-cellular materials with rising foam density.

KEYWORDS:

• Polyurethane
• rigid foams
• fire behavior

Acknowledgments

Thanks go to Patrick Klack as well as Sebastian Rabe for technical assistance with the cone calorimeter, Michael Morys for support with the SEM, and to Tobias Kukofka and Philip Nickl for the LOI measurements. The authors thank Aleksandra Sut for her support with TG measurements. Also, the help of Thomas Rybak with the heat capacity measurements was highly appreciated.

Disclosure statement

The authors declare no conflict of interest.

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website.

Additional information

Funding

This research received no external funding.