Low molecular fatty acids such as formic acid and acetic acid, and formaldehyde, which are believed to be often contained in fire smoke, are very irritative and could lead to a failure in escaping from fires, causing damage to man's eyes and respiratory organs.
Although it is known that these irritants will be produced by smouldering of wood, very few reports are available about quantitative data of these irritants evolved by burning of wood only wood but also of synthetic polymens.
In the present study, low molecular fatty acids and formaldehyde produced by combustion and pyrolysis of polymers under a variety of conditions were quantitatively measured mainly by either titrating or a colorimetic method to provide fundamental data for the evalutation of toxicity of smokes.
In general, the evolution of both low molecular fatty acids and formaldehyde was the highest at 300-400°C, lower than the temperature at which flaming occurred, when materials were heated in the air current.
These irritative substances were produced in larger quantities from aliphatic compounds and/or oxygencontaining materials than from aromatic compounds.
Flame retardant treatment and halogenated hydrocarbon reduced the evolution of low molecular fatty acids or formaldehyde.
The concentration of low molecular fatty acids will be about 10ppm if wood, polyethylene or polypropylene of 1gr. is burned in a volume of 1m³ under the optimum condition for the production of the acids, while the maximum permissible concentrations adopted by ACGPH of formic and acetic acids are 5 and 10ppm respectively. Low molecular fatty acids consisted mostly of formic and acetic acids, according to the gas chromatographic analysis.
The higher concentrations of formaldehyde, 60, 40 and 12ppm will be reached when polyethylene, polypropylene and wood are subjected to smouldering combustion under the optimum condition respectively.
When we take into consideration the lower maximum permissible concentration of formaldehyde, 2ppm, the evolution of formaldehyde in terms of toxicity is more than 10 times larger than that of low molecular fatty acids.
Since usually a floor space of each 1m² is loaded with tens of kgs of combustibles, smouldering fire could lead to dangerous situations due to above-mentioned irritants' accumulation in the room of fire, even if the dangerous concentration of each toxic gas for short time exposure is assumed about 20-100 times the maximum permissible concentration by ACGIH.

View full abstract

Critical condition for ignition in porous body with oxygen dependent exothermic reaction was investigated by use of the stability in the phase plane composed of temperature and oxygen concentration. The equation which used for the analysis is as follow in dimensionless form :
dθ/dφ = (−Aθ + δφⁿe^θ ) / ((A/ω) (1−φ) − (δ/B )φⁿe^θ )            (1)
Where θ and φ are the tempererature-rise and oxygen concentration in the specimen respectively, and B, δ, ω, A and n are the adiabatic temperature rise, the heat release rate, the ratio of the thermal diffusivities, the effective heat transfer coefficient and the order of reaction respectively. The singular solutions of Eq. (1) were studied in view of the Liapounov stability theorem, and the boundaries between stable and unstable solutions were obtained analytically. As these boundaries correspond to the criticality for the present system, the critical condition for ignition can be determined quantitatively by three dimensinless parameters B, δ and ω. Thus the following results are obtained for n=1.
in the region of B >ω³/(ω-1)
AB (B −2ω−√(B ²−4Bω)) = 2ω²δ exp ((B −√(B ²−4Bω))/2ω)            (2)
in the region of ω+2+2√(ω+1) < B < ω³/(ω-1) and ω > 3/2
AB (B −ω−2√((B −ω)²−4B )) = 2ω δexp ((B +ω− √((B −ω)²−4B ))/2ω)            (3)
and there is no critical condition in the region of B <ω+2+2√(ω+1.) These results show that the effect of oxygen diffusion for ignition becomes large as B(abiabatic temperature-rise) is small and n(order of reaction) is large. The relation between B, δ and ω is also given by a curved surface in B−δ−ω space in general.

View full abstract

The objective of the present study was primarily to pursue the dynamical characteristics of the combustion behaviors of fire-retarded materials and secondarily, to pursue the relationships between the parameter for the smoking behavior C_s^max and other combustion characteristics such as Weightloss rate, Heat evolution, CO₂/CO ratio, Gas evolution rate, and HX evolution rate.
Following results were obtained :
The heat values of the samples were estimated by sumarizing the heat evolution in the sample, heat evolution in gaseous phase and the heat exchanged between sample and gas phase under many conditions of the test furnace.
The approximate coincidence were found among the values of the heat of combustion directly obtained by the oxygen-bomb test and those obtained from the heat values on the basis of CO- and CO₂—evolution at 500°C.
The estimated values of the heat evolution of the fire-retarded samples were lower than those of the untreated ones. This result was due to the suppression of the oxidation of CO in gaseous phase by HX. and this suggested the possibilities to prevent the propagation of fire, the rapid flow of smoke and the extreme growth of smoke thickness flash-over
The smoking potency of the samples were studied in terms of smoke evolution coeftieient C_s^max and the gasification factor (f). The measurement were performed in the accumulation smoke chamber for the electro microscopically observable smokes with radii mostly larger than ca. 0.2μ. They were grown during their travel through the electric furnace where the rapid nucleation and the condensatm reaction within micro and mili-second must be considered close to the combustion surface.
The presence of the slower coagulation stage were found through a change in the distribution of smoke radii and the inhibition of the electrostatic coagulation by HX were confirmed for smoke from the fire-retarded materials. This enhanced C_s^max of the fire-retarded polymers as compared to those of the untreated ones.
The combustion parameters were found to possess approximately linear correlation with weight loss rate. The correlations among Q and C_s^max were represented by Burgess' constants which were peculiar to materials and dependent on A/F ratios of the test Furnace.
The CO evolution were found to be approximately proportional to weight loss rate, where the proportional constant was dependent on A/F ratio.
The smoke evolution (C_s^MAX ) were linearily correlated with the weightloss rate of the sample in terms of the linear operator A (2ρV_sD_vs /3Fm²KT₃) which depends on the characteristic of material but not depend on A/F ratio. Consequently, C_s^max were found be nearly proportional to CO evolution for each A/F ratios of the test furnace. Eventually, the smoke evolution rate (C_s^MAX/T₃) was estimated to be equal to A × dw/dt by operating A(=3Fm²s(1−f)) to the weightloss rate(dw/dt) where the constant A of the combustion systems was peculiar to each materials. S(=ΣK_i d_i ²N_i /w_s ) was the mean scattering cross sectional area per unit weight of the captured smoke and (1−f) was obtianed as the ratio of the weight of the smokes (W_s ) to the weight of loss of the samples (W).

View full abstract