For the first time, next to the flammability tests (LOI, UL-94 HB, VBB, TGA), experimental tests and computer simulation have been conducted on the flame spread and combustion of glass fiber-reinforced epoxy resins (GFRER) with 6% graphene and 6% DDM-DOPO flame-retardant additives. The downward rates of flame spread (ROS) in opposed flow with oxidizer and the upward ROS along GFRER composites have been first measured as well as the distribution of temperature over the combustion surface of the composites with flame-retardant additives and without them. The LOI and UL-94 HB tests showed a reduction in the flammability of GFRER when flame retardants were added and predicted a higher effectiveness of DDM-DOPO compared to graphene. Adding DDM-DOPO resulted in increasing the rate of formation of the volatile pyrolysis products and their yield, indicating, together with the other data obtained, the gas phase mechanism of the flame retardant’s action. Adding graphene resulted in an increase in the soot release on the burning surface and an increase in the amount of non-volatile pyrolysis products on the burning surface, reducing the amount of fuel that participated in the oxidation reactions in the gas phase. The developed numerical combustion model for GFRER with a DDM-DOPO additive, based on the action of DDM-DOPO as a flame retardant acting in the gas phase, satisfactorily predicts the effect of this flame retardant on the reduction in downward ROS over GFRER for 45–50% oxygen concentrations. The developed model for GFRER with graphene additive, based on a reduction in the amount of fuel and increase in the amount of incombustible volatile pyrolysis products when graphene is added, predicts with good accuracy downward ROS over GFRER depending on oxygen concentration.

No abstract

No abstract

For the first time, a comprehensive study of downward flame spread over glass-fiber-reinforced epoxy resin (GFRER) slabs in oxidizer flow has been carried out experimentally and numerically. Microthermocouples were used to measure the temperature profiles on the solid fuel’s surface and in the flame, and a video camera was used to measure the rate of flame spread (ROS). The ROS was found to be linearly dependent on the oxygen concentration, to be inversely proportional to the slab thickness and not to depend on the direction of the flame spread over the slab. The absence of the influence of the forced oxidizing flow velocity and the weak influence of the GFRER pyrolysis kinetics on the ROS were observed. For the first time, a numerical model of flame spread over reinforced material with thermal conductivity anisotropy was developed on the basis of a coupled ‘gas–solid’ heat and mass transfer model, using modifications of the OpenFOAM open-source code. The sensitivity analysis of the model showed that the thermal conductivity in the normal direction to the GFRER surface had a much greater effect on the ROS than the thermal conductivity along the direction of flame propagation. The numerical results show good agreement with the experimental data on the dependences of the ROS on oxygen concentration, slab thickness and the N2/O2 mixture flow velocity, as well as temperature distributions on the fuel surface, the maximum flame temperatures and the flame zone length.

No abstract