Condensed-phase pyrolysis

Price D, Anthony G, Carty P. Introduction Polymer combustion, condensed phase pyrolysis and smoke formation. In Fire Retardant Materials. Horrocks AR, Price D, Eds. Woodhead Publishing Cambridge, U.K., 2001 chap. 1, pp. 1-30. 

The basic assumption inherent to heat transfer limited pyrolysis models is that heat transfer rates, rather than decomposition kinetics, control the pyrolysis rate. Consequently, thermal decomposition kinetics do not come into play, other than indirectly through specification of Tp. This approximation is often justified on the basis of high activation energies typical of condensed-phase pyrolysis reactions, i.e., the reaction rate is very small below Tj, but then increases rapidly with temperature in the vicinity of Tp owing to the Arrhenius nature, and the high activation energy, of the pyrolysis reaction. 

Essentially, the conveyor belt was approximated as an ensemble of combustible particles. This approach is similar to the treatment of natural fuels such as trees and shrubs in the experimental version of FDS known as Wildland Urban Interface FDS (WFDS) [92], Fowndes et al. [91] found that the model qualitatively replicated the flame spread that was observed experimentally. However, no quantitative comparison between modeled pyrolysis front position or HRR and analogous experimental data was given. Modeling a single continuous surface, such as a conveyor belt, as an ensemble of combustible particles is a novel idea, but it is not clear in this particular case whether the relevant physics of condensed-phase pyrolysis were accurately represented. 

Within experimental error, the activation energy is the same as that measured for the condensed phase pyrolysis of hexanitroethane" , 38.9 kcal.mole, also suggesting a simple rate determining rupture of a C-N bond in the latter reaction. Slightly smaller activation energies and pre-exponential factors have been report-ed for C(N02)4 and C2(N02)e decompositions (Table 30). Trinitroalkanes" also follow first-order kinetics and it appears that C-N cleavage is rate-determining. Table 30 lists the Arrhenius parameters of thermolysis. 

The interest here is focused mainly on heavy hydrocarbons feedstocks, as in the case of certain refinery processes, and on polymer thermal degradation. A radical chain mechanism is also involved in the liquid- or condensed-phase pyrolysis. This is once again characterized by initiation, radical recombination... 

The Beckstead-Derr-Price model (Fig. 1) considers both the gas-phase and condensed-phase reactions. It assumes heat release from the condensed phase, an oxidizer flame, a primary diffusion flame between the fuel and oxidizer decomposition products, and a final diffusion flame between the fuel decomposition products and the products of the oxidizer flame. Examination of the physical phenomena reveals an irregular surface on top of the unheated bulk of the propellant that consists of the binder undergoing pyrolysis, decomposing oxidizer particles, and an agglomeration of metallic particles. The oxidizer and fuel decomposition products mix and react exothermically in the three-dimensional zone above the surface for a distance that depends on the propellant composition, its microstmcture, and the ambient pressure and gas velocity. If aluminum is present, additional heat is subsequently produced at a comparatively large distance from the surface. Only small aluminum particles ignite and bum close enough to the surface to influence the propellant bum rate. The temperature of the surface is ca 500 to 1000°C compared to ca 300°C for double-base propellants. 

Phosphoms-containing additives can act in some cases by catalyzing thermal breakdown of the polymer melt, reducing viscosity and favoring the flow or drip of molten polymer from the combustion zone (25). On the other hand, red phosphoms [7723-14-0] has been shown to retard the nonoxidative pyrolysis of polyethylene (a radical scission). For that reason, the scavenging of radicals in the condensed phase has been proposed as one of several modes of action of red phosphoms (26). 

On the other hand, numerous examples are already known in which monomeric metaphosphoric esters are generated by thermolysis reactions. Most worthy of mention in this context is the gas phase pyrolysis of the cyclic phosphonate 150 which leads via a retro-Diels-Alder reaction to butadiene and monomeric methyl metaphosphate (151) 108,109, no). While most of the phosphorus appears as pyrophosphate and trimeric and polymeric metaphosphate, a low percentage (<5%) of products 152 and 153 is also found on condensation of the pyrolyzate in a cold trap containing diethylaniline or N,N,N, N,-tetraethyl-m-phenylene-diamine. The... 

In this regard, it should be noted at this point that one of the products identified by CGC/MS from these pyrolysis reactions was SbBr3- Furthermore, the data presented concerning the importance of the polymer substrate in the degradation of the DBDPO and the proposed chain radical transfer mechanism [7] would suggest that the condensed phase chemistry could be much more important in antimony oxide/organohalogen flame retardant systems than had been previously thought. 

The flame retardant mechanism of PC/ABS compositions using bisphenol A bis(diphenyl phosphate) (BDP) and zinc borate have been investigated (54). BDP affects the decomposition of PC/ABS and acts as a flame retardant in both the gas and the condensed phase. The pyrolysis was studied by thermogravimetry coupled with fourier transform infrared spectroscopy (FUR) and nuclear magnetic-resonance spectroscopy. Zinc borate effects an additional hydrolysis of the PC and contributes to a borate network on the residue. 

Flash vacuum pyrolysis of 357 (R = H, Me) at 530°C/0.01 mmHg gave imidazo[2,l-a]isoquinolines (368) with only traces of other products provided that the sublimation temperature was maintained about 25 °C below the melting point (92AJC1811 93T8147). If compounds 357 (R = H) were allowed to melt during the flash vacuum pyrolysis, or if the pyrolysis was carried out in the condensed phase, a number of products was obtained ethyl 2-hydroxy-4-oxo-4//-pyrimido[2,l- ]isoquinoline-3-carboxylate (20) was a major component (93T8147). 

In this sequence a radical, possibly a btradical derived from unpairing the electrons of the oxazirane oxygen-nitrogen bond, abstracts the a-hydrogen atom of the A -alkyl group to form (XXI) which subsequently isomerizes to (XXII). Alternatively the formation of (XXII) may take place directly by a concerted, reaction. In either event the iminoaikoxy radical (XXII) carries the chain. The ammonia which is formed presumably comes from aldol-like condensations of the imine (XXIV). The fact that vapor-phase pyrolysis does not take this course simply reflects tbo low probability of a chain reaction in the vapor phase. 

The condensed phase mechanism was explained taking into account the decrease of the pyrolysis rate of polypropylene BiCl3 could catalyze the condensation between chloroparaffin and polypropylene by addition to chain end double bonds (Equation 4.25) formed either in reaction (Equation 4.22) or in chain scission occurring during volatilization of polypropylene 31... 

The zinc species (such as, zinc chloride and zinc hydroxychloride) in the condensed phase can alter the pyrolysis chemistry by catalyzing the dehydrohalogenation and promoting cross-linking, resulting in increased char formation and a decrease in both smoke production and flaming combustion. 

This type of model works well at high applied heat flux levels, where the pyrolysis front is thin. Simplicity is its advantage it is not necessary to specify any parameters related to the decomposition kinetics. A large body of flame spread modeling work has applied this type of model, but there is a tendency to focus with great detail on gas-phase phenomena (i.e., full Navier-Stokes, detailed radiation models, multistep combustion reactions) and treat the condensed-phase fuel generation process in an approximate manner. 

In kinetically limited models, the pyrolysis rate is no longer calculated solely from a heat balance at the pyrolysis front. Instead, the rate at which the condensed-phase is volatilized depends on its temperature. This gives a local volumetric reaction rate (kg/m3-s) by assuming that all volatiles escape instantaneously to the exterior gas-phase with no internal resistance, the fuel mass flux is obtained by integrating this volumetric reaction rate in depth. One consequence is that the pyrolysis reaction is distributed spatially rather than confined to a thin front as with heat transfer limited models and the thickness of the pyrolysis front is controlled by decomposition kinetics and heat transfer rates. For a pyrolysis reaction with high activation energy or for very high heat transfer rates, the pyrolysis zone becomes thin, and kinetically limited models tend toward heat transfer limited models. 

The FDS5 pyrolysis model is used here to qualitatively illustrate the complexity associated with material property estimation. Each condensed-phase species (i.e., virgin wood, char, ash, etc.) must be characterized in terms of its bulk density, thermal properties (thermal conductivity and specific heat capacity, both of which are usually temperature-dependent), emissivity, and in-depth radiation absorption coefficient. Similarly, each condensed-phase reaction must be quantified through specification of its kinetic triplet (preexponential factor, activation energy, reaction order), heat of reaction, and the reactant/product species. For a simple charring material with temperature-invariant thermal properties that degrades by a single-step first order reaction, this amounts to -11 parameters that must be specified (two kinetic parameters, one heat of reaction, two thermal conductivities, two specific heat capacities, two emissivities, and two in-depth radiation absorption coefficients). 

---