The Use of Inverse Gas Chromatography (IGC) to Determine the Surface Energy of RDX

The surface properties of RDX play an important role in enhancing mechanics performances of the propellants and explosives. In this work, thereby, inverse gas chromatography (IGC) using various probe liquids as the medium was used to determine the surface energy components of RDX containing both dispersive and polar components, which were acquired respectively from neutral probe liquids (such as n‐hexane, n‐heptane, n‐octane) and polar probe liquids (such as chloroform, benzene, methanol). The results show that RDX located in different column temperatures has difference in the surface energy and possesses more surface energy when there is high temperature. The calculated formula of the total surface energy with temperature is: , and it is also found that dispersive, polar, electron acceptor, and electron donor components of RDX are , , , and , respectively. These results demonstrate that the dispersive component is the primary part of the total surface energy, and RDX has an acid performance.     

Elongational flow studies on the phase separation of hydroxypropylcellulose solution

The phase separation of hydroxypropylcellulose (HPC) in a mixed solvent of glycerol and water was investigated by an elongational flow birefringence method. In the one‐phase region, the elongational flow birefringence had the characteristics of a typical coil‐stretch transition‐like pattern with a critical elongational strain rate $\dot \varepsilon_c.$ $\dot \varepsilon_c.$ increased monotonously with temperature, but in the vicinity of the phase‐separation point, $\dot \varepsilon_c.$ began to decrease even in the one‐phase region. In the two‐phase region, the flow‐induced birefringence pattern contained both a rigid rod‐like response and the coil‐stretch transition‐like response of a flexible polymer. The appearance of the rod‐like birefringence pattern indicates the association of HPC chains to form a precursor of the liquid‐crystalline phase formation. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 86: 2984–2991, 2002     

Use of various processes for pilot plant treatment of wastewater from a wood‐processing factory

The wastewater from a wood‐processing factory is characterized by a high COD, chlorides and nitrogen content. Various treatment processes were applied to treat this wastewater in pilot‐scale units. By applying one‐stage denitrification–activated sludge biological treatment it was not possible to remove nitrogen. Nitrification was inhibited by wastewater compounds. By applying a second stage of a nitrification biofilter it was possible to have a high degree of nitrification. The denitrification was complete. With biological methods the reduction of COD, and ‐N and ‐N concentrations to acceptable values was not achievable. Physical–Chemical methods as H₂O₂/UV, electrolysis and ozonation were used as post‐treatment of effluents from the biological system. Radical degradation, initiated by the powerful hydroxyl radicals which are generated from H₂O₂ by UV activation, is used for wastewater post‐treatment. The combination of H₂O₂/UV was not suitable for post‐treatment of this wastewater. With electrolysis, ‐N and COD removal can be complete. The total amount of ammonia and organic nitrogen converted to nitrate nitrogen for current density of 1.15 Adm^?2 and energy consumption of 71.6 kWhm^?3 was 0.35 gdm^?3. Further biological denitrification is required for ‐N removal to permitted values. Energy consumption for the elimination of 1 kg COD was 40.4 kWh and 35.8 kWh for current densities of 0.7 Adm^?2 and 1.15 Adm^?2 respectively. The energy required to reach the limit value of COD equal to 150 mgdm^?3 for current density of 1.15 Adm^?2 was 71.6 kWhm^?3. With ozonation, the COD removal can be complete. Further biological nitrification–denitrification is required to remove ‐N and ‐N to permitted values. At pH 7.0, in order to reach the limit value of COD equal to 150 mgdm^?3, specific ozone dose was 6.0 g per g of COD removed and the total amount of ammonia and organic nitrogen converted to nitrate nitrogen was 0.25 gdm^?3. The total equivalent energy required is estimated to be 75.0 kWhm^?3. © 2001 Society of Chemical Industry     

Versatility of a succinimidyl‐ester functional alkoxyamine for controlling acrylonitrile copolymerizations

Styrene/acrylonitrile (S/AN) and tert‐butyl methacrylate/acrylonitrile (tBMA/AN) copolymers were synthesized in a controlled manner (low polydispersity $ {{\overline M _w } / {\overline M _n }} $ with linear growth of number average molecular weight $ \overline M _n $ vs. conversion X) by nitroxide mediated polymerization (NMP) with a succinimidyl ester (NHS) terminated form of BlocBuilder unimolecular initiator (NHS‐BlocBuilder) in dioxane solution. No additional free nitroxide (SG1) was required to control the tBMA‐rich copolymerizations with NHS‐BlocBuilder, a feature previously required for methacrylate polymerizations with BlocBuilder initiators. Copolymers from S/AN mixtures (AN molar initial fractions f_AN,0 = 0.13–0.86, T = 115°C) had $ {{\overline M _w } / {\overline M _n }} $ = 1.14–1.26 and linear $ \overline M _n $ versus conversion X up to X ≈ 0.6. tBMA/AN copolymers (f_AN,0 = 0.10–0.81, T = 90°C) possessed slightly broader molecular weight distributions ( $ {{\overline M _w } / {\overline M _n }} $ = 1.23–1.50), particularly as the initial composition became richer in tBMA, but still exhibited linear plots of $ \overline M _n $ versus conversion X up to X ≈ 0.6. A S/AN/tBMA terpolymerization (f_AN,0 = 0.50, f_S,0 = 0.40) was also conducted at 90°C and revealed excellent control with $ \overline M _n $ = 13.6 kg/mol, $ {{\overline M _w } / {\overline M _n }} $ = 1.19, and linear $ \overline M _n $ versus conversion X up to X = 0.54. Incorporation of AN and tBMA in the final copolymer (molar composition F_AN = 0.47, F_tBMA = 0.11) was similar to the initial composition and represents initial designs to make tailored, acid functional AN copolymers by NMP for barrier materials. © 2012 Wiley Periodicals, Inc. J. Appl. Polym. Sci., 2013     

Use of amidoximated acrylonitrile/N‐vinyl 2‐pyrrolidone interpenetrating polymer networks for uranyl ion adsorption from aqueous systems

Poly(N‐vinyl 2‐pyrrolidone) (PVP)/acrylonitrile (AN) interpenetrating polymer networks (IPNs) were synthesized and amidoximated for the purpose of uranyl ion adsorption. The adsorption of amidoximated IPNs was studied from different uranyl ion solutions (850, 1000, 1200, 1400, and 1600 ppm). The result of all our adsorption studies showed that the bonding between UO‐amidoxime groups complied with the Langmuir‐type isotherm. The adsorption capacity was found as 0.75 g UO/g dry amidoximated IPN. In order to increase the UO ion adsorption capacity the amidoximated IPN was treated with alkali, but no significant increase could be observed. © 2001 John Wiley & Sons, Inc. J Appl Polym Sci 81: 2324–2329, 2001     

Shear rate dependence of the viscosity of cellulose nitrate solutions in the dilute concentration regime revisited

A detailed rheological study of cellulose nitrate in ethylacetate had been carried out in the dilute concentration (c) regime, covering a degree of polymerization (DP) range between 300 < DP_η < 7000 and shear rates ($ \dot \gamma $) between 100 s^?1 < $ \dot \gamma $ < 2000 s^?1. The results show a strong dependence of the transition Newtonian to non-Newtonian behavior on the three variables $ \dot \gamma $, DP, and c, similar to that found recently on solutions of synthetic polymers. Emphasis has been put on the critical concentrations corresponding to the standard shear rate 1000 s^?1 to correspond to the standard conditions ($ \dot \gamma $ ? 1000 s^?1; 0.3 < [η] · c < 0.6; DS = 2.90 ± 0.02) proposed for the determination of the intrinsic viscosity [η] of cellulose nitrates. It is shown that solutions with concentrations adjusted according to the above given conditions still exhibit Newtonian behavior, up to the highest range of DP. It follows, therefore, that applying the standard conditions, an extrapolation to $ \dot \gamma $ = 0 as has been proposed often for the intrinsic viscosity determination of cellulose nitrate is not advisable and results in considerable error. Considering the relationship between [η] and DP, the present results indicate that the decrease of the exponent ( a ) from a = 1.0 to a = 0.76, taking place above a DP ? 1000, is not a consequence of the applied shear rate but rather of the molecular properties of the solutes themselves.     

Mechanical and surface properties and hydrolytic stability of cycloaliphatic polyester‐based waterborne polyurethanes modified with fluoro oligomer

Polyurethane dispersions (PUDs) have been designed and synthesized based on different types of soft segments, namely, poly(2,4‐diethyl‐1,5‐pentamethylene adipate) glycol (PDPAD; = 1000) and poly(2,4‐diethyl‐1,5‐pentamethylene‐1,4‐cyclohexane dicarboxylate) glycol (PDPCD; = 1000), and were subsequently modified with fluoro oligomer (BA‐N fluoroalcohol). It was found that the PDPCD segments improved the hydrolytic stability, whereas the contact angle with water drop was significantly increased with the addition and increasing amount of Zonyl for PDPAD‐based PU and it was marginally increased with PDPCD‐based PU. In addition, the PDPCD provided PUD with enhanced mechanical properties compared with the PDPAD. © 2008 Wiley Periodicals, Inc. J Appl Polym Sci, 2009     

Direct pyrolysis mass spectrometry analysis of fresh and aged $\hbox{PF}_{6}^{-}$‐doped polythiophenes

In this study, pyrolysis mass spectrometry has been applied for the investigation of thermal characteristics of ‐doped polythiophene (PTh) films. Thermal degradation of PTh showed behaviour identical with ‐doped polythiophene (BF₄‐PTh), indicating that the effects of the dopants in question on the thermal characteristics of the PTh are not significant. Besides the oligomer peaks (up to hexamer), peaks indicating decomposition of the thiophene ring have also been detected as in the case of BF₄‐PTh. However, the data also indicated that the oxidation of the dopant was much more effective, yielding mainly POF₃, H₃PO₄ compounds even for the freshly prepared samples. Upon ageing, the yield of these products increased and exposure to air for more than 50 days also caused degradation of the PTh chains. Furthermore, substitution of the thiophene (Th) ring by fluorine has been recorded and was associated with reactions of Th with the reactive oxidation products of the dopant. Analyses of the de‐doped samples showed that although de‐doping was effective to a certain extent, inward diffusion of the counter ion (C₄H₉)₄N⁺ has also occurred. This diffusion was more efficient than that was observed for BF₄‐PTh samples. Copyright © 2004 Society of Chemical Industry     

Vapour pressure and vapourisation enthalpy of pyridine N‐oxide

The vapour pressure of pyridine N‐oxide was measured using a boiling point method at different temperatures from 389.02 to 546.33 K and the constants of the Antoine equation for the substance were determined. The vapourisation enthalpy of pyridine N‐oxide at the normal boiling point is calculated to be 58.8 ± 1.8 kJ mol^?1. Based on Othmer's method, the standard enthalpy of vapourisation is estimated to be 66.6 ± 0.1 kJ mol^?1 by using pyridine as the reference substance. Furthermore, Watson equation is employed to verify the reliability of and the results demonstrate that the value of is acceptable. © 2011 Canadian Society for Chemical Engineering     

Kinetic study of the effect of poly(phenyl sulfone) on the curing of an epoxy/amine resin by conventional and by temperature‐modulated differential scanning calorimetry

Kinetic studies of the curing reaction of semi‐interpenetrating polymer networks (semi‐IPNs) based on diglycidyl ether of bisphenol‐A (DGEBA) and 4,4′‐diaminodiphenylmethane (DDM), containing poly(phenyl sulfone) (PPSU), were carried out using differential scanning calorimetry (DSC) and temperature‐modulated DSC (TMDSC), under both isothermal and dynamic conditions. The curing kinetics is discussed in the framework of three kinetic models: the Kissinger and the Flynn–Wall–Ozawa models, and the autocatalytic model developed by Kamal. To describe the cured reaction in its latter stage, we used the semi‐empirical relationship proposed by Chern and Poehlein to consider the influence of diffusion on reaction rate. The cure mechanism for the system studied remained broadly autocatalytic regardless of PPSU content, and it became far more diffusion controlled at higher PPSU content and lower cure temperatures. The vitrification time of the resins was obtained with TMDSC by following the changes on the complex modulus of heat capacity, , and exhibited a strong dependence on the PPSU content in the semi‐IPN systems. Copyright © 2005 Society of Chemical Industry     

In‐Line Monitoring and Control of Conversion and Weight‐Average Molecular Weight of Polyurethanes in Solution Step‐Growth Polymerization Based on Near Infrared Spectroscopy and Torquemetry

Summary: It is well known that the weight‐average molecular weight ( ) is strictly dependent on conversion in step‐growth polymerizations performed in batch and that the is very sensitive to impurities and molar imbalance. This makes the work of controlling a non trivial job. In this paper a new methodology is introduced for in‐line monitoring and control of conversion and of polyurethanes produced in solution step‐growth polymerizations, based on near‐infrared spectroscopy (NIRS) and torquemetry. A calibration model based on the PLS method is obtained and validated for monomer conversion, while the weight‐average molecular weight is monitored indirectly with the relative shear signal provided by the agitator. Control procedures are then proposed and implemented experimentally to avoid gelation and allow for maximization of . The proposed monitoring and control procedures can also be applied to other step growth polymerizations.

Proposed control scheme.     

Elongational flow studies of hinged rodlike molecules

Poly(amino acid) in an intermediate state of its helix-coil transition is known to be in a hinged rodlike conformation. In this work, the responses of poly(amino acids) in the hinged rodlike conformation against an elongational flow field were investigated by monitoring their flow-induced birefringence. Poly(L-glutamic acids) (PGA) and poly(γ-benzyl-L-glutamate) (PBLG) were examined as polyelectrolyte and noncharged poly(amino acids), respectively, and the results were compared. In the plots of flow-induced birefringence, Δn, against strain rate, $ {\dot \varepsilon } $, for hinged rodlike PBLG, there was a critical strain rate, $ \dot \varepsilon_0 $, below which Δn was not observed. Over $ \dot \varepsilon_0 $, the birefringence pattern observed was identical with that of rodlike molecules. The Δn vs. $ {\dot \varepsilon } $ plot for hinged rodlike PGA had characteristics of a rigid rod at any strain rate and there was no $ \dot \varepsilon_0$ observed. The rotational diffusion coefficient, D_r, of PBLG in the hinged rodlike conformation was larger than that for its helical conformation, while D_r, for the hinged-rodlike PGA was smaller than that for its helical conformation. It is concluded that the hinged-rodlike PGA molecule is in an extended form and that the hinged-rodlike PBLG is hydrodynamically more compact and rigid than that in its quiescent state. It is deduced that at $\dot \varepsilon_0$ hinged rodlike PBLG molecules collapse to a conformation optically anisotropic and mechanically rigid. © 1996 John Wiley & Sons, Inc.     

Lewis acid–base properties of cellulose acetate phthalate‐polycaprolactonediol blend by inverse gas chromatography

The net retention volumes, V_N, of n‐alkanes and five polar probes are determined on cellulose acetate phthalate–polycaprolactonediol blend column by inverse gas chromatography in the temperature range 323.15–363.15 K. The dispersive surface energy, $\gamma _{\bf S}^{\bf d}$ , of the blend has been calculated using the V_N values of n‐alkanes and the $\gamma _{\bf S}^{\bf d}$ at 333.15 K is 12.6 mJ/m². The $\gamma _{\bf S}^{\bf d}$ values are decreasing linearly with increase of temperature. The V_N values of the five polar solutes are used to calculate the specific component of the enthalpy of adsorption, ${\Delta }{H}_{\bf a}^{\bf S}$ . The Lewis acid–base parameters, K_a and K_b, are derived using ${\bf \Delta }{H}_{\bf a}^{\bf S}$ values and are found to be 0.019 and 0.403, respectively. The K_a and K_b values indicate that the blend surface contain more basic sites and interact strongly with the acidic probes. The acid–base parameters have been used to analyze the preferential interaction of the solid surface with acidic and basic probes. POLYM. ENG. SCI., 2013. © 2013 Society of Plastics Engineers     

Adsorptive removal of nitrate and phosphate anions from aqueous solutions using functionalised SBA‐15: Effects of the organic functional group

Adsorption of nitrate and monovalent phosphate anions from aqueous solutions on mono, di‐ and tri‐ammonium‐functionalised mesoporous SBA‐15 silica was investigated. The adsorbents were prepared via a post‐synthesis grafting method, using either 3‐aminopropyltrimethoxysilane (N‐silane) or [1‐(2‐aminoethyl)‐3‐aminopropyl]trimethoxysilane (NN‐silane) or 1‐[3‐(trimethoxysilyl)‐propyl]‐diethylenetriamine (NNN‐silane), followed by acidification in HCl solution to convert the attached surface amino groups to positively charged ammonium moieties. The nominal loading of amino moieties on the SBA‐15 surface was varied from 5% to 20% as organoalkoxysilane/silica molar ratio. The adsorption experiments were conducted batchwise at room temperature. Results showed that adsorption capacity increased with increasing the concentration of monoammonium groups on the SBA‐15 adsorbent. Nitrate adsorption capacity increased from 0.34 to 0.66 mmol ${\rm NO}_{3}^{{-} } /{\rm g}$ adsorbent while phosphate adsorption capacity increased from 0.34 to 0.63 mmol ${\rm H}_{2} {\rm PO}_{4}^{{-} } /{\rm g}$ adsorbent when the molar ratio organoalkoxysilane/silica was varied from 5% to 20%, respectively. Also, for the same organoalkoxysilane/silica molar ratio of 10%, the adsorption capacity increased with the increase of the number of protonated amines in the functional groups. Therefore, maximum adsorption capacities of 0.80, 1.16 and 1.38 mmol ${\rm NO}_{3}^{{-} } /{\rm g}$ adsorbent and 0.72, 0.82 and 1.17 mmol ${\rm H}_{2} {\rm PO}_{4}^{{-} } /{\rm g}$ adsorbent were obtained using mono‐, di‐ and triammonium functionalised SBA‐15 adsorbents, respectively. © 2011 Canadian Society for Chemical Engineering     

Pilot‐scale anaerobic/anoxic/oxic/oxic biofilm process treating coking wastewater

BACKGROUND: Biological treatment efficiency of coking wastewater is rather poor, especially for chemical oxygen demand (COD) and ammonia‐nitrogen (NH$_{4}^{+}$ ‐N) removal due to its complex composition and high toxicity. RESULTS: A pilot‐scale anaerobic/anoxic/oxic/oxic (A²/O²) biofilm system has been developed to treat coking wastewater, focusing attention on the COD and NH$_{4}^{+}$ ‐N removal efficiencies. Operational results over 239 days showed that hydraulic retention time (HRT) of the system had a great impact on simultaneous removals of COD and NH$_{4}^{+}$ ‐N. At HRT of 116 h, total removal efficiencies of COD and NH$_{4}^{+}$ ‐N were 92.3% and 97.8%, respectively, reaching the First Grade discharge standard for coking wastewater in China. Adequate HRT, anoxic removal of refractory organics and two‐step aerobic bioreactors were considered to be effective measures to obtain satisfactory coking effluent quality using the A²/O² biofilm system. The correlation between removal characteristics of pollutants and spatial distributions of biomass along the height of upflow bioreactors was also revealed. CONCLUSION: The study suggests that it is feasible to apply the A²/O² biofilm process for coking wastewater treatment, achieving desirable effluent quality and steady process performance. © 2012 Society of Chemical Industry     

Preparation and Performance of Highly Efficient Au Nanoparticles Electrocatalyst for the Direct Borohydride Fuel Cell

Au nanoparticles supported on Vulcan XC‐72R carbon were prepared by a modified NaBH₄ method in aqueous solution and employed as electrocatalyst of oxidation for the direct borohydride fuel cell (DBFC). The morphology and structure of as‐prepared particles were examined by transmission electron microscopy (TEM) and X‐ray diffraction (XRD). It was found that Au nanoparticles were mainly about 3.0 ± 0.5 nm in size and uniformly distributed on the surface of Vulcan XC‐72R carbon. The electrooxidation behaviors of and fuel cell performances using carbon‐supported Au nanoparticles as catalysts were investigated. Compared with Au/C prepared by conventional reduction method, the kinetics of oxidation on as‐prepared carbon supported 3.0 ± 0.5 nm Au nanoparticles were significantly improved. The DBFC employing carbon supported 3.0 ± 0.5 nm Au nanoparticles showed a maximum power density of 85.3 mW cm^–2 at 60 °C.     

Stress relaxation behaviour of liquid crystalline solutions of ethyl celluloses with different molecular weight in m-cresol

The stress relaxation behaviour of liquid crystal-forming ethyl celllulose (EC) solutions in m-cresol was determined by means of a cone-plate type viscometer at 30°C. The effect of molecular weight (MW) on the behaviour was also determined. The relaxation behaviour could be fitted with the following equation: where σ_i and σ_f are steady-state shear stresses at shear rate $\dot \gamma _{\rm i}$ and $\dot \gamma _{\rm f}$, σ(t) is time- dependent stress, A₁ and A₂ are constants, τ₁ and τ₂ are relaxation times, t is time, and t_c is a characteristic time. When log σ* was plotted against time, one straight line was obtained for isotropic solutions, whereas anisotropic solutions yielded two straight lines. This suggests that the liquid crystalline solutions have two separate relaxation processes: Process 1 has a relatively short relaxation time, and process 2 has a long one. The parameters τ₁, τ₂, and A₂ were greatly dependent on polymer concentration, combination of $\dot \gamma _{\rm i}$ and $\dot \gamma _{\rm f}$, and MW, whereas A₁ was independent thereof and was close to unity. The process 1 was supposed to be valid for individual molecules, and process 2 for liquid crystalline domains or randomly aggregated or entangled molecules.     

Evaluation of the termination mode in the radical polymerization of 2,2,2‐trifluoroethyl α‐fluoroacrylate

The mode of termination of 2,2,2‐trifluoroethyl α‐fluoroacrylate (FATRIFE) in radical polymerization was studied, and only termination by recombination occurred, which led to telechelic macromolecular structures. The radical polymerization in acetonitrile was carried out to synthesize oligomers with a low number average degree of polymerization ( )_cum (about 20), using tert‐butylcyclohexyl peroxydicarbonate (TBCPC) as initiator at 75 °C. The initial [TBCPC]₀/[FATRIFE]₀ molar ratio was monitored to evaluate its influence on the ( )_cum of α‐fluoroacrylic oligomers. The ¹H NMR analysis of the polymers showed that the ( )_cum values obtained were higher than 40, in spite of a high C₀ value. To explain these results, the mode of termination was evaluated using the following kinetic law: . The development of kinetic relationships allowed us to calculate the ratio k_prt/k_i·k_p as about 17–30 mol s l^?1, and to confirm that primary radical termination (PRT) was in competition with bimolecular macromolecular termination (BMT). © 2002 Society of Chemical Industry     

Kinetics of ion exchange in monodisperse resin

Ion-exchange kinetics within a conventional strong base resin, Dowexl-8X®, and a resin with uniform particle size, Dowex® Monosphere® Tough Gel® TG550A®, were investigated using neutron activation analysis and radio-tracer techniques. The kinetics of ion exchange were measured in a batch and in a “shallow-bed” flow system. The experimental data were compared with the results of model computations. The diffusivities of several anions within TG550A and Dowexl-8X were deduced. It was found that at 25°C Br^-, Cl^-, OH^-, and Na⁺ diffuse within TG550A with the diffusion coefficients = 6.0 × 10^-7 cm²/s, = 1.2 × 10^-6 cm²/s, = 7.0 × 10^-8 cm²/s, and = 5 × 10^-7 cm²/s. Diffusion of anions within a conventional resin, Dowexl®, was slower: = 3.5 × 10^-7 cm²/s, = 6 × 10^-7 cm²/s, = 2.7 × 10^-8 cm²/s, and = 5 × 10^-7 cm²/s. A higher rate of ion diffusion and the bead-size uniformity may make monodisperse Dowex Monosphere Tough Gel TG550A resin attractive for analytical applications. The difference in properties between conventional and monodisperse resins is not sufficient to affect the large volume applications of resins. © 1997 John Wiley & Sons, Inc. J Appl Polm Sci 65:1271–1283, 1997     

Atom‐transfer radical polymerization of methyl methacrylate with α,α′‐dichloroxylene/CuCl/N,N,N′,N″,N″‐pentamethyldiethylenetriamine initiation system under microwave irradiation

The atom‐transfer radical polymerization (ATRP) of methyl methacrylate (MMA), using α,α′‐dichloroxylene as initiator and CuCl/N,N,N′,N″,N″‐pentamethyldiethylenetriamine as catalyst was successfully carried out under microwave irradiation (MI). The polymerization of MMA under MI showed linear first‐order rate plots, a linear increase of the number‐average molecular weight with conversion, and low polydispersities, which indicated that the ATRP of MMA was controlled. Using the same experimental conditions, the apparent rate constant (k) under MI (k = 7.6 × 10^?4 s^?1) was higher than that under conventional heating (k = 5.3 × 10^?5 s^?1). © 2004 Wiley Periodicals, Inc. J Appl Polym Sci 92: 2189–2195, 2004