Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
ABSTRACTS AND KEVIEWS 287 cellulose m nitrogen and air atmospheres Some chromatographic experiments using flame retarded cellulose (mono ammonium phosphate) are given and some general observations on the action of this flame retardant are recorded Lipska, A. E . and Martin, S. B. (Stanford Research Institute, Menlo Park, Cali- forma) "The Effect of Flame Retardants on Thermal Degradation of a-Cellulose in Nitrogen," Fmal Report, August 1970-August 1971, under Contract No DAHC20-70-C-0219 for Office of Civil Defense (August 1971) Section: H Subjects: Cellulose pyrolysis. Effect of morganic additives Authors' Summary The Problem Cellulose is a major constituent of the bulk of combustible materials available as fuel in destructive fires For more effective prediction, control, and prevention of urban fires, i t is necessary to learn more about the rate and mode of decomposi- tion of cellulosic materials Additionally, to develop more effective flame retardants, i t IS important to understand how these retardants mteract with the decomposition process m cellulose to reduce its flammabihty. The present study seeks to develop a self-consistent model of the decomposition of cellulose and the effect of flame- retardant treatments I t reviews current theories of cellulose decomposition, including the Parker-Lipska model previously developed by one of the authors.^ New data are presented and used to test the capabihty of the current theoretical models to predict the pyrolytic behavior of cellulose The Findings A sigmficant finding of this research is that the specially purified cellulose, a commercial filter paper, used in contemporary studies in other laboratories behaves in isothermal pyrolysis in a manner similar to that of acid-salt-treated alpha cellulose used m this laboratory This finding suggests that the specially piuified material, known to have been treated with strong acids m the purification process, is substantially degraded by the purification process in a manner resembling the degradation induced by the addition of acids or salts of weak bases to alpha cellu- lose With this in mind, i t is reasonable to attribute at least part of the differences in mechanisms of decomposition of cellulose to varying procedures used m the preparation of the material as well as to their residual mineral content Molecular weight determinations of the acid-purified filter paper offer further evidence of the vahdity of this explanation
288 F I B E KESEARCH The second highly significant finding of the current experimental work deals with the important effect of low concentrations of oxygen on the rate of decom- position of cellulose Comparative measurements of rates of decomposition of thick and thin samples indicate no effect of sample thickness if oxygen is carefully excluded. On the other hand, surface adsorption of low concentrations of oxygen appears to increase the rate of decomposition m the initial phase of pyrolysis of the thin (0.0043-in.) samples by a factor of three, whereas no appreciable effect was noted in the case of thick (0 030-m.) samples. The results (1) suggest that the previously observed initial rapid decomposition phase is caused, at least in part, by effects of surface-absorbed oxygen and (2) indicate that there is no appreciable surface effect resulting from mteraction between the particles of the fluidized bed and the surface of the cellulose, thereby proving the effectiveness of the fluidized- bed technique for isothermal-kmetic studies. The experimental results further indicate that the overall pattern for the weight and monomer loss of neutral-salt-treated cellulose is similar to that of untreated samples, but with the rates of weight and monomer loss being, respectively, 25 8 and 25 2 times faster than the untreated samples Observed rates of weight loss agree well with values predicted by the Parker-Lipska model Molecular-weight measurements of the neutral-salt-treated cellulose combined with the correspond- ing data on monomer loss suggest that the dominant reaction that governs the decomposition is simultaneous unzipping of the majority of the molecules. Rates of weight loss for the acidic-salt-treated samples are often orders of magni- tude faster than those of untreated samples; for example, the whole degradation process is finished after 1 minute at a temperature of 276°C in cellulose containing small amounts of ammonium sulfate I n contrast to untreated cellulose, weight loss is not hnear with time over most of the duration of active pyrolysis, similar to the case for acid-purified cellulose. Monomer loss is also changed, asymptotically approaching a linutmg value rather than suffermg depletion in value with the length of pyrolysis as with neutral and basic retardants. Molecular-weight meas- urements of the acid-salt-treated cellulose show a rapid decrease in the degree of polymerization at the beginning of pjTolysis, asymptotically approaching a (DP) value of 10 with subsequent heating. This suggests that acidic salts mcrease the rate of degradation through hydromum-ion-catalyzed hydrolysis. This hydrolytic effect seems to be the primary reason for the failure of the Parker-Lipska model to predict effectively the rates of weight loss of cellulose treated with acid re- tardants. The pattern for the rates of weight and monomer loss of basic-salt-treated cellulose is sinular to that of the untreated cellulose, but with rates of weight and monomer loss bemg 19 and 50 05 times faster, respectively, than for the untreated samples. Results of the molecular weight measurements show a rapid drop in the average degree of polymerization after the imtial heating, and, just as in KBr- treated samples, a slow progressive decrease in the length of the molecules on subsequent heating. Degradation products from acid-, neutral-, and basic-salt-treated cellulose were analyzed with the gc/ms combination instrument and with an infrared spectrometer. Results indicated that the basic salt treatments eliminate the pro- duction of levoglucosan, whereas neutral and acidic salt additions decrease the amount of levoglucosan but do not eliminate i t entirely. Acidic treatment sig- nificantly increases the amount of the furanose derivative, previously identified,
