Cellulose-based fibers from liquid crystalline solutions. III. Processing and morphology of cellulose and cellulose hexanoate esters

Cellulose and a cellulose hexanoate ester (DS 0.69) exhibited liquid crystalline behavior in dimethylacetamide/lithium chloride and dimethylacetamide, respectively. The experimentally observed critical volume fraction (V^c_p) of cellulose was lower than that predicted by Flory's theory, whereas the experimental and theoretical values of V^c_p were within 70% of prediction for cellulose hexanoate. The V^c_p value obtained for cellulose hexanoate was lower than that previously reported for cellulose acetate butyrate with a maximum degree of butyration (CAB-3). This indicates that bulky substituents may lower V^c_p values. Fibers were spun from isotropic and anisotropic solutions of cellulose and cellulose hexanoate by a dry jet/wet spinning method. There was an increase in mechanical properties through the isotropic to anisotropic transition with moduli reaching 152 g/d (20.8 GPa) for cellulose fibers. The formation of cellulose fibers with high modulus at large extrusion rates and large takeup speeds (draw ratio) is explained with molecular organization prior to coagulation. This unexpected enhancement is attributed to the air gap that exists in the dry jet/wet spinning process. Similar improvements were not observed for cellulose hexanoate fibers. This is explained with incomplete development of liquid crystalline structure at the solution concentrations from which the fibers were spun. © 1993 John Wiley & Sons, Inc.     

Transient currents in charging and discharging modes in ethyl cellulose

Transient current in vacuum aluminized ethyl cellulose foil samples of thickness 75 μm were investigated in charging and discharging modes at different temperatures and fields. The occurrence of a peak in the current–time characteristics of charging current (I_c), and the dependence of such currents on field, characterized by a negative resistance region and followed by a superlinear region, are considered to indicate injection controlled transient currents modified by space-charge arising from the localization of carriers in various traps. Carrier mobility values calculated from a knowledge of the peak in the charging current versus time characteristics were found to range from 10^?11 to 10^?9 cm²/V s. As times larger than the peak time, the charging current follows the relation I ∝ t^?n at low temperatures with different poling fields (V_p). However, the current increases with time and finally tends to saturate for samples poled at higher temperatures with higher values of field. The isochronal characteristics (i.e. current–temperature plots at fixed times) constructed from discharge current versus time plots are characterized by a peak at 70°C or 100°C, depending upon the poling field values.     

A kinetic model for pyrolysis of cellulose

It has been shown that the pyrolysis of cellulose at low pressure (1.5 Torr) can be described by a three reaction model. In this model, it is assumed that an “initiation reaction” leads to formation of an “active cellulose” which subsequently decomposes by two competitive first-order reactions, one yielding volatiles and the other char and a gaseous fraction. Over the temperature range of 259–341°C, the rate constants of these reactions, k_i (for cellulose → “active cellulose”), k_v (for “active cellulose” → “volatiles”), and k_c (for “active cellulose” → char + the gaseous fraction) are given by k_i = 1.7 × 10²¹e^{? (58,000/RT)} min ^?1, k_v = 1.9 × 10¹⁶e^{? (47,300/RT)} min^?1, and k_c = 7.9 × 10¹¹e^{? (36,600/RT)} min^?1, respectively.     

Thermal and structural properties of tri-O-substituted cellulose ethers

Tri-O-substituted cellulose ethers prepared by the use of SO₂–diethylamine (DEA)–dimethylsulfoxide (DMSO) and powdered sodium hydroxide were characterized by measurement of melting points and degradation points and by differential scanning calorimetric (DSC) and X-ray analyses. Their thermal and structural characteristics were ascertained to be dependent on the kinds of substituents. Among the ethers examined, four kinds of arylmethylcelluloses showed the characteristics of thermotropic liquid crystals (smectic). Tri-O-α-naphthylmethyl-cellulose showed the three transformations due to a solid–smectic phase, a smectic–nematic phase, and a nematic–isotropic phase transformation. This thermotropic liquid crystal was enantiotropic below 200°C.     

Electrically conductive composites of poly(urea‐formaldehyde) and cellulose filled with aluminum

This article addresses the preparation and characterization of composite materials obtained with compression molding of mixtures of aluminum powder and a commercial grade thermosetting resin of poly(urea‐formaldehyde) filled with α‐cellulose in powder form. The homogeneity of these composites was checked by the morphologies of the constituents (filler and matrix) by optical microscopy. The density of the composites was measured and compared with values calculated by assuming different void levels within the samples, to discuss the porosity effect, in connection with optical microscopy observations. Then, the dependence of electrical conductivity of the composites on volume fraction of the metal filler was investigated. The conductivity of the composites is <10⁻¹² S/cm unless the metal content reaches the percolation threshold at a volume fraction of V_c = 38.6 vol%, beyond which the conductivity increases markedly by as much as nine orders of magnitude, indicating an insulator–conductor phase transition. The obtained results on electrical conductivity have been well interpreted with the statistical percolation theory. The deduced critical parameters, such as the threshold of percolation, V_c, the critical exponent, t, and the packing density coefficient, F, were in good accord with earlier studies. In addition, the hardness of samples remained almost constant with the increase of metal concentration. POLYM. COMPOS., 2012. © 2012 Society of Plastics Engineers     

Isothermal transient current studies in cellulose acetate films

Step voltage transient current studies have been made in cellulose acetate films as a function of filed and thickness. A logarithmic plot (Scherr-Montroll plot) of the transient current vs. time gives a knee at a time t_T, which is interpreted as the transit time of the charge carrier. The value of the carrier mobility has been estimated to be 3.9 × 10^?9 cm².V^?1.S^?1 in cellulose acetate film. The carrier mobility in iodine-doped (2% w/w) cellulose acetate film has also been determined from Scher-Montroll plot and is found to be 3.3 × 10^?7 cm².V^?1.S^?1.     

Vinylic graft copolymers of cellulose. I. Effects of monomer concentrations and crystallinity index of cellulosic substrate on grafting vinyl acetate onto cotton

Graft copolymerization of vinyl acetate (VA) and methyl acrylate (MA) on cotton cellulose was initiated by the Ce (IV) ion, and ungrafted vinylic polymer was separated from the graft copolymer by acetone extraction. The influence of the ratio aqueous initiator solution volume/monomeric volume (V_aq/V_mon), vinyl acetate volume/methyl acrylate volume (V_VA/V_MA), and the cellulose crystallinity index (CI) on the grafting reaction were studied. To modify the crystallinity of cellulose, native cotton was treated with NaOH in the concentrations 10, 15, and 20% (mercerized). The viscosimetric average molecular weight (M_v), the polymerization degree (PD), and the crystallinity index proposed by Nelson and O'Connor (CI) were determined for native and NaOH-treated cotton. The polymeric side chains grafted were separated from the cellulose backbone by acid hydrolysis in 72% H₂SO₄. The viscosimetric average molecular weight (M_v) was determined, and the number of vinylic chains per cellulosic chain (graft frequency, GF) were calculated. The grafting percentage, %G, was higher for most amorphous cellulose and for a higher methyl acrylate percentage (%MA) in monomeric reaction mixtures (VA-MA). The V_aq/V_mon ratio that yields the highest %G was 70/30. The increase of the %G with the %MA in the VA–MA monomeric mixture seems to be due to both an increase in the length of vinylic grafted chains (as shown by its M_v) and the number of grafted chains (GF). The increase in the %G when the crystallinity index (CI) of the cellulosic substrate decreases seems to be due to an increase in the length of the vinylic grafted chains, but not to an increase in the number of grafted chains, since the M_v increases and GF decreases when the CI of cellulose decreases.     

Viscosity‐molecular weight relationship for cellulose solutions in either NMMO monohydrate or cuen

The intrinsic viscosities, [η], of nine cellulose samples, with molar masses from 50 × 10³ to 1 390 × 10³ were determined in the solvents NMMO*H₂O (N‐methyl morpholin N‐oxide hydrate) at 80°C and in cuen (copper II‐ethlenediamine) at 25°C. The evaluation of these results with respect to the Kuhn–Mark–Houwink relations shows that the data for NMMO*H₂O fall on the usual straight line in the double logarithmic plots only for M ≤ 158 10³; the corresponding [η]/M relation reads log ([η]/mL g⁻¹) = –1.465 + 0.735 log M. Beyond that molar mass [η] remains almost constant up to M ≈ 10⁶ and increases again thereafter. In contrast to NMMO*H₂O the cellulose solutions in cuen behave normal and the Kuhn–Mark–Houwink relation reads log ([η]/mL g⁻¹) = −1.185 + 0.735 log M. Possible reasons for the dissimilarities of the behavior of cellulose in these two solvents are being discussed. The comparison of three different methods for the determination of [η] from viscosity measurements at different polymer concentrations, c, demonstrates the advantages of plotting the natural logarithm of the relative viscosities as a function of c. © 2010 Wiley Periodicals, Inc. J Appl Polym Sci, 2011     

Investigation of the cellulose/LiCl/dimethylacetamide and cellulose/LiCl/N-methyl-2-pyrrolidinone solutions by 13C NMR spectroscopy

The cellulose/lithium chloride/dimethylacetamide (DMAc) and cellulose/lithium chloride/N-methyl-2-pyrroilidinone (NMP) solutions were investigated by ¹³C NMR spectroscopy. Well-resolved spectra were obtained for both solutions and indicated that cellulose was present in these systems in the form of underivatized cellulose. The change in ¹³C chemical shifts of DMAc and NMP in the presence of LiCl and LiBr was compared with that of several salt/aprotic solvents, and the results point to the existence of a cellulose–LiCl–DMAc (or NMP) complex in which the lithium cation is strongly bound to the amide carbonyl oxygen and the chloride anion involved in the dissociation of the cellulose hydrogen bonds. Spin–lattice relaxation times (T₁ of the ¹³C carbons of the solvent molecules, DMAc and NMP, show a large decrease in T¹ for all solvent carbons upon addition of LiCl. Further decrease in T₁ is observed when cellulose is introduced to the LiCl/NMP but not to the LiCl/DMAc systems. These observations are attributed to slower molecular motions of DMAc and NMP in the presence of LiCl, and, in the case of NMP, in the presence of cellulose.     

Characterisation of cellulose treated by the steam explosion method. Part 3: Effect of crystal forms (cellulose I,II and III) of original cellulose on changes in morphology,degree of polymerisaion,solubility and supermolecular structure by steam explosion

An attempt was made to clarify the effect of the crystal form of untreated cellulose on the morphological and structural changes of cellulose during steam explosion treatment (steam pressure P = 2.9MPa (T = 508K), treatment time t = 15-300 s). For this purpose, the crystal form of soft wood pulp (cellulose I) was converted by solid-to-solid transition, with minimal unavoidable change in other structural characteristics including morphology and average degree of polymerisation, into cellulose II or cellulose III. It was proved by both X-ray and solid-state cross-polarisation/magic-angle sample-spinning (CP/MAS) ¹³C NMR analyses that even a simple addition of water at room temperature brought about a significant structural change in the steam-untreated cellulose samples. The solubility towards 9.1 wt% aqueous sodium hydroxide, S_a, of the cellulose samples of crystal forms I and III could be improved from 31-33% up to almost 100% by selecting appropriate steam explosion conditions (for example, P = 2.9MPa, t = 30 s). Such a magnificent increase in S_a by the steam explosion treatment was not observed for the cellulose II sample, even under the rather severe conditions of the steam explosion treatment at which the cellulose III crystal was converted to a large extent to cellulose I, as confirmed by X-ray diffraction. X-ray diffraction analysis showed that crystallisation of samples with cellulose I or II crystal occurred to some extent during the steam explosion treatment. Contrary to this, the degree of breakdown of the intramolecular hydrogen bond O₃…O'₅, as estimated by CP/MAS ¹³C NMR analysis, significantly increased for cellulose I and I11 during the treatment. The decrease in the viscosity-average degree of polymerisation, P, observed for all treated samples can be roughly categorised into two or three steps of the first-order decomposition reaction with different reaction rates.     

Pyridine-catalyzed depolymerization of cellulose during carbanilation with phenylisocyanate in dimethylsulfoxide

Carbanilation reactions of cellulose samples (bleached cotton linters and Avicel) with phenylisocyanate in dimethylsulfoxide (DMSO) were carried out at 60°C in the presence of various pyridine derivatives. The molecular weight distributions of the resulting cellulose tricarbanilates (CTCs) were measured by high-performance size exclusion chromatography. When pyridine or its derivatives were included in the reactions, the CTCs had reduced degree of polymerization (DP) values compared to those of CTCs prepared without the additives. The cellulose depolymerization was promoted by pyridines with electron-donating substituents and was not favored by pyridines with electron-withdrawing substituents nor with groups at positions ortho to the pyridine ring nitrogen atom. For the 3-, 4-, and 3,4- substituted pyridines, there was a linear relationship between log (weight-average CTC DP) and the pK_a (in water) of the pyridine derivative. For 2- and 2,6-substituted pyridines, the DP–pK_a relationships were different, probably because of steric effects of the different substituents ortho to the pyridine nitrogen atom. The optimum DMSO : pyridine solvent ratio for cellulose depolymerization during carbanilation in DMSO : pyridine mixtures was 3 : 1. All three components, phenylisocyanate, pyridine or its derivatives, and DMSO, are required for the depolymerization reaction. It is suggested that the depolymerization may be a consequence of cellulose oxidation by DMSO and/or cleavage of glucosidic bonds of partially carbanilated cellulose by reactions promoted by an enhanced solvent effect of DMSO.     

Thermal decomposition of cellulose ethers

The thermostability and thermal decomposition kinetics of methyl cellulose (MC), ethyl cellulose (EC), carboxymethyl cellulose (CMC), hydroxyethyl cellulose (HEC), and hydroxypropyl–methyl cellulose (HPMC) were characterized in nitrogen and air by thermogravimetry (TG). Various methods of kinetic analysis were compared in case of thermal degradation of the five cellulose ethers. The initial decomposition temperature (T_d), temperature at the maximum decomposition rate (T_dm), activation energy (E), decomposition reaction order (n), and pre-exponential factor (Z) of the five cellulose ethers were evaluated from common TG curves and high-resolution TG curves obtained experimentally. The decomposition reactions in nitrogen were found to be of first order for MC, EC, and HPMC with the average E and ln Z values of 135 kJ/mol and 25 min⁻¹, although there were slight differences depending on the analytical methods used. The thermostability of cellulose ethers in air is substantially lower than in nitrogen, and the decomposition mechanism is more complex. The respective average E, n, ln Z values for HEC in nitrogen/air were found to be 105/50 kJ/mol, 2.7/0.5, and 22/8.3 min⁻¹, from constant heating rate TG method. The respective average E, n, and ln Z values for three cellulose ethers (EC/MC/HPMC) in air are 123/144/147 kJ/mol, 2.0/1.8/2.2, 24/28/28 min⁻¹ by using high-resolution TG technique. © 1999 John Wiley & Sons, Inc. J Appl Polym Sci 73: 2927–2936, 1999     

Preparation and properties of chemical cellulose from ascidian tunic and their regenerated cellulose fibers

Chemical cellulose (dissolving pulp) was prepared from ascidian tunic by modified paper‐pulp process (prehydrolysis with acidic aqueous solution of H₂SO₄, digestion with alkali aqueous solution of NaOH/Na₂S, bleaching with aqueous NaOCl solution, and washing with acetone/water). The α‐ cellulose content and the degree of polymerization (DPw) of the chemical cellulose was about 98 wt % and 918, respectively. The Japanese Industrial Standard (JIS) whiteness of the chemical cellulose was about 98%. From the X‐ray diffraction patterns and ¹³C‐NMR spectrum, it was found that the chemical cellulose obtained here has cellulose Iβ crystal structure. A new regenerated cellulose fiber was prepared from the chemical cellulose by dry–wet spinning using N‐methylmorpholine‐ N‐oxide (NMMO)/water (87/13 wt %) as solvent. The new regenerated cellulose fiber prepared in this study has a higher ratio of wet‐to‐dry strength (<0.97) than commercially regenerated cellulose fibers. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 85: 1634–1643, 2002.     

The Analysis of the Electrical and Photovoltaic Properties of Cr/p-Si Structures Using Current-Voltage Measurements

In this work, the electrical and photovoltaic properties of Cr/p -Si structures were investigated using forward and reverse bias current-voltage (I - V ) measurements in dark and under illumination conditions (100 mW/cm²) at room temperature. The forward and reverse bias current–voltage (I - V ) characteristics of the Cr/p-Si structures were analyzed by the thermionic emission theory. For this, the main parameters such as ideality factors (n), barrier heights (Φ_bo), series resistances (R_S), and reverse-saturation currents obtained from different methods using forward and reverse bias I - V measurements were investigated in under dark and illumination conditions at room temperature, respectively. Furthermore, the photovoltaic parameters such as short circuit current (I_sc), open circuit voltage (V _oc), fill factor (FF) and conversion efficiency (η_P) were acquired as 7.43 × 10^-3 A/cm², 0.260 V, 61.5% and 1.18% under 100 mW/cm² light intensity, respectively, and these values are near to a photodiode. Experimental results show that all electrical parameters were found to be strong function of illumination density. Also, this result confirms that Cr/p-type-Si diode can be used as a photodiode in optoelectronic applications.

     

Preparation of tri-O-benzylcellulose by the use of nonaqueous cellulose solvents

A procedure has been developed for preparing tri-O-benzylcellulose with higher yields than 90%. In this procedure, cellulose was dissolved in a nonaqueous cellulose solvent, SO₂–diethylamine (DEA)–dimethyl sulfoxide (DMSO), and then powdered sodium hydroxide was added. To the suspension thus prepared, benzyl chloride was added in four portions under heating. Chemical and physical properties of tri-O-benzylcellulose were studied, and tri-O-benzylcellulose prepared from microcrystalline cellulose powder shows the characteristic of thermotropic liquid crystal.     

Predictions of nucleation theory applied to ehrenfest thermodynamic transitions: II. The effects of pressure and stress

This paper is a sequel to an earlier one on the applicability of classical nucleation theory to second-order transitions in the Ehrenfest sense (1). In each case the approach was to obtain the critical size r_c and energy barrier ΔG_c for the growth of a nucleus of β-phase in an α-phase matrix by a Maclaurin series expansion of the free-energy-density g = (G_β ? G_α)/v_β as a function of θ (in BC-I) and of ΔP and Δσ in this paper where θ = (T ? T_t) is the degree of undercooling and ΔP and Δσ are analogous terms for the hydrostatic pressure shift and tensile stress shift away from the equilibrium transition. The expansion coefficients were determined by the use of thermodynamic relationships. For second-order transitions, r_c = 4γv_β T_t/ΔC_pθ², r_c = 4γ/Δβ(Δp)², and r_c = 4γ/Y_αY_β(Δσ)², respectively, for the three cases. The terms ΔC_p, Δβ, and ΔY denote the differences in heat capacity, compressibility, and Young's modulus, e.g., ΔY = Y_β ? Y_α. The interfacial energy γ_αβ is denoted by γ. The activation energy barriers for the cases developed in this paper were ΔG_c = (16π/3)γ³/(Δβ)² (Δp)⁴ and ΔG_c = (64π/3)γ³Y_α²Y_β²/(ΔY)²(Δσ)⁴. More complicated expressions are given in the paper for the r_c and ΔG_c for first-order transitions. In the long run, these expressions may prove more useful than the ones for second-order because of the modifications expressions for the kinetics of transformations.     

Preparation of cellulose derivatives containing carbazole chromophore

Cellulose derivatives containing long hydrocarbon side chains and the carbazole chromophore are prepared. N‐4′‐Bromobutylcarbazole is first synthesized from carbazole and 1,4‐dibromobutane. Alkylated carbazole is then reacted with cellulose acetate in dimethyl sulfoxide solution to produce cellulose ethers containing the desired chromophore. Polymers containing a mixture of alkyl side chains are also prepared by the subsequent addition of 1‐bromododecane to the reaction mixture. Characterization of the resulting cellulose derivatives by FTIR spectroscopy indicates that the deacetylation of cellulose acetate and the subsequent etherification are both complete. In addition, the incorporation of the carbazole chromophore is clearly shown by ¹H‐ and ¹³C‐NMR spectroscopy. Polymers of different carbazole content, ranging from 2.9 to 1.1 chromophores per anhydroglucose repeat unit, are obtained by varying the reaction conditions. Substitution is found to be controlled primarily by the quantity of alkylating agent introduced while variation of the reaction time has little effect. This method is used to prepare (dodecyl)_y(N‐4′‐carbazolylbutyl)_xcellulose, (decyl)_y(N‐4′‐carbazolylbutyl)_xcellulose, and (butyl)_y(N‐4′‐carbazolylbutyl)_xcellulose. Cellulose acetate can be replaced by (methyl)cellulose as the starting material to obtain analogous products. © 1999 John Wiley & Sons, Inc. J Appl Polym Sci 74: 2764–2772, 1999     

Cellulose phosphates as biomaterials. I. Synthesis and characterization of highly phosphorylated cellulose gels

The present study concerns the investigation of a material expected to be biocompatible and able to promote bone regeneration. For this purpose, cellulose was chemically modified by phosphorylation. Once implanted, phosphorylated cellulose could promote the formation of calcium phosphates, thus having closer resemblance to bone functionality. In a previous investigation, the obtention and the preliminary characterization of cellulose phosphate gels were reported. In the present study, the synthesis by the H₃PO₄/P₂O₅/Et₃PO₄/hexanol method was optimized in terms of reaction parameters. The structure of materials was investigated by FTIR, Raman, and solid‐state ³¹P– and ¹³C–NMR spectroscopic studies, and X‐ray diffraction. Water swelling and stability to sterilization by gamma‐radiation were also assessed. It was demonstrated that the present method allows highly phosphorylated cellulose derivatives to be obtained. Cellulose triphosphates gels are described here for the first time. Products obtained were poorly crystallized monoesters, apparently not significantly affected by gamma sterilization and showed a high water swelling capability. Chemical bonding was confirmed by FTIR, Raman, and both ³¹P– and ¹³C–NMR spectroscopies. It was also shown that the H₃PO₄/P₂O₅/Et₃PO₄/hexanol method provides a versatile and interesting alternative route to some of the more widely used techniques for the phosphorylation of cellulose. © 2001 John Wiley & Sons, Inc. J Appl Polym Sci 82: 3341–3353, 2001     

Competitive removal of heavy metal ions by cellulose graft copolymers

The effect of composition of graft chains of four types cellulose graft copolymers on the competitive removal of Pb²⁺, Cu²⁺, and Cd²⁺ ions from aqueous solution was investigated. The copolymers used were (1) cellulose‐g‐polyacrylic acid (cellulose‐g‐pAA) with grafting percentages of 7, 18, and 30%; (2) cellulose‐g‐p(AA–NMBA) prepared by grafting of AA onto cellulose in the presence of crosslinking agent of N,N′‐methylene bisacrylamide (NMBA); (3) cellulose‐g‐p(AA–AASO₃H) prepared by grafting of a monomer mixture of acrylic acid (AA) and 2‐acrylamido‐2‐methyl propane sulphonic acid (AASO₃H) containing 10% (in mole) AASO₃H; and (4) cellulose‐g‐pAASO₃H obtained by grafting of AASO₃H onto cellulose. The concentrations of ions which were kept constant at 4 mmol/L in an aqueous solution of pH 4.5 were equal. Metal ion removal capacities and removal percentages of the copolymers was determined. Metal ion removal capacity of cellulose‐g‐pAA did not change with the increase in grafting percentages of the copolymer and determined to be 0.27 mmol metal ion/g_copolymer. Although the metal removal rate of cellulose‐g‐p(AA–NMBA) copolymer was lower than that of cellulose‐g‐pAA, removal capacities of both copolymers were the same which was equal to 0.24 mmol metal ion/g_copolymer. Cellulose did not remove any ion under the same conditions. In addition, cellulose‐g‐pAASO₃H removed practically no ion from the aqueous solution (0.02 mmol metal ion/g_copolymer). The presence of AASO₃H in the graft chains of cellulose‐g‐p(AA–AASO₃H) created a synergistic effect with respect to metal removal and led to a slight increase in metal ion adsorption capability in comparison to that of cellulose‐g‐pAA. All types of cellulose copolymers were found to be selective for the removal of Pb²⁺ over Cu²⁺ and Cd²⁺. © 2003 Wiley Periodicals, Inc. J Appl Polym Sci 90: 2034–2039, 2003     

The use of DTA to study spontanceous ignition of cellulose

The use of differential thermal analysis has enabled spontaneous ignition behaviour of cotton cellulose to be investigate. The temperature. T_i, at which the onset of spontaneous ignition occurs is recorded as a function of the oxygen concentration of the flowing oxygen-nitrogen atmosphere to which the cellulose sample is exposed in the DTA furnace, when heated at a defined heating rate. The dependence of T_i, on heating rate has enabled the activation energy, E_p, of the rate-determining flammable pyrolysis product reaction to the determined using both a previously derived simple kinetic model and the technique of Ozawa. E_p, increases from a lower limiting value of 112 kJ mol^?1 at zero oxygen concentration to an asymptote value of 169 kJ mol^?1 at oxygen volume concentrations above 30%. This effect is described in terms of oxygen catalysis of competing pyrolysis routes. At a given heating rate, increased oxygen concentration reduces T_i. A plot of 1/T_i versus In [O₂] gives two liner regions which intersect at an oxygen concentration of about 20%, suggesting that two combustion mechanisms exist, one above and the other below this value. Below this concentration, which is similar to the conventional limiting oxygen for cellulose, significant char remains, suggesting that ignition of gaseous products only occurs. If the difference in slopes is sttributed to the variations in E_p with oxygen concentration, then a value for the activation energy of gaseous product oxidation, E_ox = 215 kJ mol^?1 is derived.