The preparation of sterically stabilized polymer dispersions

The emulsion polymerizations of styrene (St) and butyl acrylate (BA) stabilized by nonionic polyoxyethylene type emulsifiers did not show the long stationary rate interval. This was discussed in terms of two opposing effects: 1) the decreased monomer concentration at the reaction loci due to the depletion of monomer droplets or depressed monomer droplet degradation and 2) the increased number of polymer particles with increasing conversion. The continuous particle nucleation is attributed to the continuous release of emulsifier from the emulsifier saturated monomer droplets and/or the presence of monomer swollen micelles (microdroplets). The limited particle flocculation operative at lower emulsifier concentrations increases the nonstationary-state polymerization. The particle agglomeration is accompanied by the increased reaction order x (N_p vs. [E]^x) above 0.6. The increased uniformity of monomer emulsion stabilized by Tween 20 by homogenization of monomer emulsion increased the final conversion and the polymerization rate as well. The polymerization rate vs. conversion curve of the homogenized emulsion characterized with broader stationary rate interval reminds the four rate intervals system typical for miniemulsion. The accumulation of polymer and nonionic emulsifier within the monomer phase preserves the monomer droplets up to high conversion. The decreased monomer droplet degradation rises the monomer-starved condition or the depressed transport of both monomer and emulsifier to the reaction loci.     

Effect of temperature on styrene emulsion polymerization in the presence of sodium dodecyl sulfate. II

The batch emulsion polymerization kinetics of styrene initiated by a water‐soluble peroxodisulfate at different temperatures in the presence of sodium dodecyl sulfate was investigated. The curves of the polymerization rate versus conversion show two distinct nonstationary‐rate intervals and a shoulder occurring at a high conversion, whereas the stationary‐rate interval is very short. The nonstationary‐state polymerization is discussed in terms of the long‐term particle‐nucleation period, the additional formation of radicals by thermal initiation, the depressed monomer‐droplet degradation, the elimination of charged radicals through aqueous‐phase termination, the relatively narrow particle‐size distribution and constant polydispersity index throughout the reaction, and a mixed mode of continuous particle nucleation. The maximum rate of polymerization (or the number of polymer particles nucleated) is proportional to the rate of initiation to the 0.27 power, which indicates lower nucleation efficiency as compared to classical emulsion polymerization. The low activation energy of polymerization is attributed to the small barrier for the entering radicals. The overall activation energy was controlled by the initiation and propagation steps. The high ratio of the absorption rate of radicals by latex particles to the formation rate of radicals in water can be attributed to the efficient entry of uncharged radicals and the additional formation of radicals by thermally induced initiation. © 2000 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 38: 1477–1486, 2000     

Microemulsion polymerization of styrene in the presence of a cationic emulsifier

The principal subject discussed in the current paper is the radical polymerization of styrene in the three- and four component microemulsions stabilized by a cationic emulsifier. Polymerization in the o/w microemulsion is a new polymerization technique which allows to prepare the polymer latexes with the very high particle interface area and narrow particle size distribution. Polymers formed are very large with a very broad molecular weight distribution. In emulsion and microemulsion polymerizations, the reaction takes place in a large number of isolated loci dispersed in the continuous aqueous phase. However, in spite of the similarities between emulsion and microemulsion polymerization, there are large differences caused by the much larger amount of emulsifier in the latter process. In the emulsion polymerization there are three rate intervals. In the microemulsion polymerization only two reaction rate intervals are commonly detected: first, the polymerization rate increases rapidly with the reaction time and then decreases steadily. Essential features of microemulsion polymerization are as follows: (1) polymerization proceeds under non-stationary state conditions; (2) size and particle concentration increases throughout the course of polymerization; (3) chain-transfer to monomer/exit of transferred monomeric radical/radical re-entry events are operative; and (4) molecular weight is independent of conversion and distribution of resulting polymer is very broad. The number of microdroplets or monomer-starved micelles at higher conversion is high and they persist throughout the reaction. The high emulsifier/water ratio ensures that the emulsifier is undissociated and can penetrate into the microdroplets. The presence of a large amount of emulsifier strongly influences the reaction kinetics and the particle nucleation. The mixed mode particle nucleation is assumed to govern the polymerization process. At low emulsifier concentration the micellar nucleation is dominant while at a high emulsifier concentration the interaction-like homogeneous nucleation is operative. Furthermore, the paper is focused on the initiation and nucleation mechanisms, location of initiation locus, and growth and deactivation of latex particles. Furthermore, the relationship between kinetic and molecular weight parameters of the microemulsion polymerization process and colloidal (water/particle interface) parameters is discussed. In particular, we follow the effect of initiator and emulsifier type and concentration on the polymerization process. Besides, the effects of monomer concentration and additives are also evaluated.     

Sterically and electrosterically stabilized emulsion polymerization. Kinetics and preparation

The principal subject discussed in the current paper is the radical polymerization in the aqueous emulsions of unsaturated monomers (styrene, alkyl (meth)acrylates, etc.) stabilized by non-ionic and ionic/non-ionic emulsifiers. The sterically and electrosterically stabilized emulsion polymerization is a classical method which allows to prepare polymer lattices with large particles and a narrow particle size distribution. In spite of the similarities between electrostatically and sterically stabilized emulsion polymerizations, there are large differences in the polymerization rate, particle size and nucleation mode due to varying solubility of emulsifiers in oil and water phases, micelle sizes and thickness of the interfacial layer at the particle surface. The well-known Smith-Ewart theory mostly applicable for ionic emulsifier, predicts that the number of particles nucleated is proportional to the concentration of emulsifier up to 0.6. The thin interfacial layer at the particle surface, the large surface area of relatively small polymer particles and high stability of small particles lead to rapid polymerization. In the sterically stabilized emulsion polymerization the reaction order is significantly above 0.6. This was ascribed to limited flocculation of polymer particles at low concentration of emulsifier, due to preferential location of emulsifier in the monomer phase. Polymerization in the large particles deviates from the zero-one approach but the pseudo-bulk kinetics can be operative. The thick interfacial layer can act as a barrier for entering radicals due to which the radical entry efficiency and also the rate of polymerization are depressed. The high oil-solubility of non-ionic emulsifier decreases the initial micellar amount of emulsifier available for particle nucleation, which induces non-stationary state polymerization. The continuous release of emulsifier from the monomer phase and dismantling of the non-micellar aggregates maintained a high level of free emulsifier for additional nucleation. In the mixed ionic/non-ionic emulsifiers, the released non-ionic emulsifier can displace the ionic emulsifier at the particle surface, which then takes part in additional nucleation. The non-stationary state polymerization can be induced by the addition of a small amount of ionic emulsifier or the incorporation of ionic groups onto the particle surface. Considering the ionic sites as no-adsorption sites, the equilibrium adsorption layer can be thought of as consisting of a uniform coverage with holes. The de-organization of the interfacial layer can be increased by interparticle interaction via extended PEO chains--a bridging flocculation mechanism. The low overall activation energy for the sterically stabilized emulsion polymerization resulted from a decreased barrier for entering radicals at high temperature and increased particle flocculation.     

Kinetics and mechanism of emulsifier-free emulsion copolymerization: Styrene-methyl methacrylate-acrylic acid system

The emulsifier-free emulsion copolymerization of styrene (St) and methyl methacrylate (MMA) in the presence of functional monomer acrylic acid (AA) was carried out in batch process. The kinetics was investigated in detail using model function, Integrated Gamma Function. The morphology and size of particles were monitored continuously by TEM all along the polymerization. It was found that the nucleation, polymerization rate increase with increasing concentration of the functional monomer AA, initiator ammonium persulfate (APS), and polymerization temperature T, and APS plays a predominant role in the particle nucleation process. The particle nucleation stage ceased at about 10% conversion and the steady stage can be extended to about 70% conversion. The particle nucleation is likely to yield primary particle via the mechanism of homogeneous coagulative nucleation and coagulation of the primary particle to yield uniform particles. The particle growth in the postnucleation stage is via a shell growth mechanism. © 1999 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 37: 2649–2656, 1999     

Emulsion and miniemulsion polymerizations with an oil‐soluble initiator in the presence and absence of an aqueous‐phase radical scavenger

Butyl acrylate conventional emulsion (macroemulsion) and miniemulsion polymerizations were carried out with an oil‐soluble initiator (azobisisobutyronitrile) in the presence or absence of an aqueous‐phase radical scavenger. For macroemulsion polymerization, in the presence of an aqueous‐phase radical scavenger, no particle nucleation occurred, whereas in the absence of an aqueous‐phase radical scavenger, particle nucleation proceeded as expected. For miniemulsion polymerization, the rate of polymerization was much higher in the absence of an aqueous‐phase radical scavenger than in its presence. Furthermore, in the absence of an aqueous‐phase radical scavenger, the miniemulsion polymerization rate increased with reduced droplet size, whereas in the presence of an aqueous‐phase radical scavenger, the trend was reversed. It is concluded that (1) for macroemulsion polymerization, the contribution from free radicals originating in the aqueous‐phase is predominant in the micellar nucleation of particles; (2) free radicals originating in the particle phase contribute to the rate of polymerization and the contribution increases with an increase in the particle size; and (3) for polymer particles with diameters of up to approximately 100 nm, polymerization is initiated from free radicals originating in the aqueous phase. © 2002 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 40: 3200–3211, 2002     

Kinetics of the batch cationic emulsion polymerization of styrene: A comparative study with the anionic case

An in‐depth study on the kinetics of the cationic emulsion polymerization of styrene in a batch reactor is presented. This study is focused on the effect of the amount of the cationic surfactant dodecyltrimethylammonium bromide (DTAB), using two different cationic initiators: 2,2′‐azobisisobutyramidine dihydrochloride (AIBA), 2,2′‐azobis (N,N′‐dimethyleneisobutyramidine) dihydrochloride (ADIBA), on kinetics and colloidal features such as conversion, number of particles, number average of radicals per particle, mean particle diameter, and particle size distribution (PSD) of the polystyrene latices obtained by emulsion polymerization in a batch reactor. Furthermore, the results of the cationic emulsion polymerization were compared with its homologous anionic case. Using DTAB as cationic surfactant an expected increase in the total rate of polymerization was observed when the DTAB concentration increased. However, the total number of particles increased much more than in the anionic system. On the other hand, a dependence on the particle size of the rate of polymerization per particle together with the average number of radicals per particle was found. These differences between cationic and anionic emulsion polymerizations were explained taking into account the limited particle coagulation observed with cationic surfactants, and the high rate of radical formation of cationic initiators. © 2006 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 44: 4461–4478, 2006     

Microemulsion polymerization of butyl acrylate. IV. Effect of emulsifier concentration

The oil/water microemulsion polymerizations of butyl acrylate initiated by a water (ammonium peroxodisulfate, APS) or oil (dibenzoyl peroxide, DBP) soluble radical initiator at different emulsifier concentrations were investigated. The rate of polymerization vs. conversion curve shows two intervals. The rate of polymerization is found to decrease with the emulsifier concentration. This finding was discussed in terms of the decrease of both radical and monomer concentration, the chain transfer to emulsifier, desorption of chaintransferred radicals, and the contribution of solution polymerization. The polymerization is faster with APS. In the APS system the rate per particle or the number of radicals per particle increases exponentially with increasing particle size. The particle size and number increase during the whole polymerization. This behavior was discussed in terms of the nucleation of monomer-containing micelles and agglomeration of primary particles during the whole polymerization. © 1995 John Wiley & Sons, Inc.     

On the kinetics of styrene emulsion polymerization above CMC. I. A mathematical model

A detailed mathematical model of the kinetics of styrene emulsion polymerization has been proposed. Its main features/assumptions are compartmentalization, micellar and homogeneous nucleation, particle formation by both initiator‐derived and desorbed radicals, dependence on the particle size of the rate coefficients, thermodynamic considerations, and aqueous phase kinetics. The model predicts that micellar nucleation dominates over homogeneous nucleation and that the evolution of the nucleation rate reaches a maximum, where desorbed radicals have an important contribution. Initiator‐derived radicals with only one monomeric unit have also a significant contribution on the rate of capture in particles. The results suggest that the correctness of the instantaneous termination approach depends not only on the size of the particle, but also on the type of entering radical (initiator‐derived or monomeric). © 2000 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 38: 2201–2218, 2000     

Butyl acrylate batch emulsion polymerization in the presence of sodium lauryl sulphate and potassium persulphate

The batch emulsion polymerization of butyl acrylate in the presence of sodium lauryl sulphate as emulsifier and potassium persulphate as initiator was investigated. The effects of emulsifier concentration, initiator concentration, and monomer/water ratio on the kinetic features were studied. The kinetic data showed that at the conditions studied, the number of particles is proportional to [KPS]^0.39 and [SLS]^0.54. The number of particles did not practically vary with monomer concentration at the high range of monomer and emulsifier concentrations. At low emulsifier concentration, particle coagulation occurred in the course of reaction, which increased with monomer concentration. Particle nucleation was found to occur during Interval III of the batch process if undissociated micelles exist. It was also confirmed that the zero-one kinetics system can better fit the experimental results, compared to the pseudobulk kinetics. © 1999 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 37: 3957–3972, 1999     

Role of the nonionic surfactant Triton X-405 in emulsion polymerization. I. Homopolymerization of styrene

The emulsion polymerization of styrene was studied using the nonionic surfactant Triton X-405 (octylphenoxy polyethoxy ethanol). Two separate nucleation periods were noted in these polymerizations resulting in bimodal final latex particle size distributions. The partitioning of the surfactant between the phases was found to play the major role in determining the nucleation mechanism(s) in these polymerizations. Although the total concentration of the emulsifier was always added at a level above its critical micelle concentration (CMC) based on the water phase in the recipe, it was found that the portion of the surfactant initially present in the aqueous phase was below its CMC due to the partitioning. This CMC was also found to increase with increasing total surfactant because the distribution of the surfactant (varying ethylene oxide chain length) depended on the partitioning between the phases. Under these conditions, the first of the two nucleation periods was attributed to homogeneous nucleation, while the second was attributed to micellar nucleation. © 1997 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 35: 3813–3825, 1997     

Radical polymerization in homogenized butyl acrylate emulsion

The addition of a small amount of monomer strongly decreased the clouding temperature of nonionic emulsifier (Tween 20). The clouding temperature of the Tween 20 aqueous solution was independent of emulsifier concentration but it strongly varied in the presence of monomer. The decreased cloud temperature was attributed to the penetration of monomer molecules into the interfacial layer that increased the flocculation of microdroplets (monomer-swollen micelles). The surface tension of homogenized ((mini)emulsion) butyl acrylate aqueous emulsion was much smaller than that estimated at or above CMC of Tween 20. The polymerization rate vs. conversion curve of the (mini)emulsion deviates from the three rate intervals typical for the emulsion polymerisation. The shape of the rate-conversion curve reminds more the four rate intervals curve. Interval 2 is overlapped with the initial maximal rate and rate shoulder at higher conversion. The initial maximal polymerization rate (R_p,max,1) is attributed to the abrupt increase in polymer particles, the polymerization under monomer saturated condition and emulsifier containing peroxide groups (Tw_peroxid 20). The rate of emulsion polymerization of BA initiated by ammonium peroxodisulphate (APS) is ca. by one order of magnitude larger than that of blank polymerization (without APS). The second maximal rate (rate shoulder) can result from the gel effect. The more pronounced increase in R_p,max,1 with Tw 20 concentration supports the presence of peroxide groups. The slight dependence of R_p,max,2 on [Tw 20] for both APS and DBP (dibenzoyl peroxide) is discussed in terms of the depressed radical entry rate into the close packed surface later of polymer particles. The low activation energy is attributed to the decreased barrier for entering radicals into the polymer particles with increasing temperature. This is more pronounced with the accumulation of covalently bound emulsifier moieties (resulting from Tw_peroxid 20) at the particle surface. The ratio of the final number of polymer particles to the initial number of monomer droplets (N_p/N_drop) promotes the partial monomer droplet nucleation. The dye approach indicates that the degree of depletion of monomer droplets decreases from the classical emulsion polymerization to the polymerization in pre-homogenized emulsions and the emulsion polymerization with a prolonged-emulsification interval.     

Particle formation under monomer‐starved conditions in the semibatch emulsion polymerization of styrene. I. Experimental

Particle formation and particle growth compete in the course of an emulsion polymerization reaction. Any variation in the rate of particle growth, therefore, will result in an opposite effect on the rate of particle formation. The particle formation in a semibatch emulsion polymerization of styrene under monomer‐starved conditions was studied. The semibatch emulsion polymerization reactions were started by the monomer being fed at a low rate to a reaction vessel containing deionized water, an emulsifier, and an initiator. The number of polymer particles increased with a decreasing monomer feed rate. A much larger number of particles (within 1–2 orders of magnitude) than that generally expected from a conventional batch emulsion polymerization was obtained. The results showed a higher dependence of the number of polymer particles on the emulsifier and initiator concentrations compared with that for a batch emulsion polymerization. The size distribution of the particles was characterized by a positive skewness due to the declining rate of the growth of particles during the nucleation stage. A routine for monomer partitioning among the polymer phase, the aqueous phase, and micelles was developed. The results showed that particle formation most likely occurred under monomer‐starved conditions. A small average radical number was obtained because of the formation of a large number of polymer particles, so the kinetics of the system could be explained by a zero–one system. The particle size distribution of the latexes broadened with time as a result of stochastic broadening associated with zero–one systems. © 2001 John Wiley & Sons, Inc. J Polym Sci Part A: Polym Chem 39: 3940–3952, 2001     

Particle nucleation in emulsion polymerization. IV. Nucleation in monomer droplets

Nucleation in finely dispersed monomer emulsions in competition with homogeneous and micellar mechanism was studied. The emulsions were prepared with a high-pressure homogenizer under varying homogenizing conditions and made stable by the addition of hexadecane to the monomer. Sodium dodecyl sulfate was used as emulsifer. The number of particles in latexes polymerized with potassium persulfate and benzoyl peroxide initiators was measured and plotted as a function of the free emulsifier concentrations. With persulfate initiator the particle number achieved a minimum in the transition region between droplet and water phase–micellar nucleation mechanisms. With very fine emulsions Smith–Ewart case II kinetics with n = 0.5 applied. The reaction rate, which differed from conventional emulsion polymerizations, decreased with time up to the point at which n began to increase. In these runs the particle size distribution became nearly monodisperse.     

Effects of 2‐hydroxyalkyl methacrylates on the styrene miniemulsion polymerizations stabilized by SDS and alkyl methacrylates

The effects of 2‐hydroxyalkyl methacrylates (HEMA and HPMA) on the styrene miniemulsion polymerizations stabilized by SDS/lauryl methacrylate (LMA) or SDS/stearyl methacrylate (SMA) were investigated. A mixed mode of particle nucleation (monomer droplet nucleation and homogeneous nucleation) is operative during polymerization. Homogeneous nucleation plays a crucial role in the polymerizations stabilized by SDS/LMA, whereas monomer droplet nucleation becomes more important in the polymerizations stabilized by SDS/SMA. The polymerization kinetics is insensitive to the type of 2‐hydroxyalkyl methacrylates, but the difference in the relative importance of monomer droplet nucleation and homogeneous nucleation is detected. Incorporation of 1‐pentanol (C₅OH) into the reaction mixture also shows a significant influence on the polymerizations stabilized by SDS/LMA or SDS/SMA. This is attributed to the formation of a close‐packed structure of SDS and C₅OH on the droplet surface, which acts as a barrier to the incoming oligomeric radicals. © 2000 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 38: 3188–3199, 2000     

Sterically stabilized emulsion polymerization of styrene: Pseudo‐semicontinuous approach

The sterically stabilized emulsion polymerization of styrene initiated by a water‐soluble initiator at different temperatures has been investigated. The rate of polymerization (R_p) versus conversion curve shows the two non‐stationary‐rate intervals typical for the polymerization proceeding under non‐stationary‐state conditions. The shape of the R_p versus conversion curve results from two opposite effects—the increased number of particles and the decreased monomer concentration at reaction loci as the polymerization advances. At elevated temperatures the monomer emulsion equilibrates to a two‐phase or three‐phase system. The upper phase is transparent (monomer), and the lower one is blue colored, typical for microemulsion. After stirring such a multiphase system and initiation of polymerization, the initial coarse polymer emulsion was formed. The average size of monomer/polymer particles strongly decreased up to about 40% conversion and then leveled off. The initial large particles are assumed to be highly monomer‐swollen particles formed by the heteroagglomeration of unstable polymer particles and monomer droplets. The size of the “highly monomer” swollen particles continuously decreases with conversion, and they merge with the growing particles at about 40–50% conversion. The monomer droplets and/or large highly monomer‐swollen polymer particles also serve as a reservoir of monomer and emulsifier. The continuous release of nonionic (hydrophobic) emulsifier from the monomer phase increases the colloidal stability of primary particles and the number of polymer particles, that is, the particle nucleation is shifted to the higher conversion region. Variations of the square and cube of the mean droplet radius with aging time indicate that neither the coalescence nor the Ostwald ripening is the main driving force for the droplet instability. © 2003 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 41: 804–820, 2003     

Particle formation and coagulation in the seeded semibatch emulsion polymerization of butyl acrylate

Particle formation and coagulation in the seeded semibatch emulsion polymerization of butyl acrylate were studied under monomer‐starved conditions. To investigate the importance of the kinetics of the water phase in the nucleation process, the monomer feed rate was used as a variable to alter the monomer concentration in the aqueous phase. The emulsifier concentration in the feed was employed to alter the particle stability. Particle formation and coagulation were discussed in terms of critical surface coverage ratios. Particle coagulation occurred if the particle surface coverage dropped below θ_cr1 = 0.25 ± 0.05. The secondary nucleation occurred above a critical surface coverage of θ_cr2 = 0.55 ± 0.05. The number of particles remained approximately constant if the particle surface coverage was within θ_cr1 = 0.25 < θ < θ_cr2 = 0.55. This surface coverage band is equivalent to the surface tension band of 42.50 ± 5.0 dyne/cm that is required to avoid particle formation and coagulation in the course of polymerization. The kinetics of the water phase was shown to play an important role during homogeneous and micellar nucleations. For any fixed emulsifier concentration in the feed and above θ_cr2, the number of secondary particles increased with monomer concentration in the aqueous phase. Moreover, the presence of micelles in the reaction vessel is not the only perquisite for micellar nucleation to occur, a sufficient amount of monomer should be present in the aqueous phase to enhance the radical capture by partially monomer‐swollen micelles. The rate of polymerization increased with the surfactant concentration in the aqueous phase. © 2000 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 38: 3612–3630, 2000     

Transition from microemulsion to emulsion polymerization: Mechanism and final properties

Microemulsion and emulsion polymerization can have some similarities in starting conditions and polymerization mechanisms, but the resulting latices are unalike in particle size and molecular weight. Here we show that polymerizations can be formulated that display the characteristics often separately associated with microemulsion or emulsion polymerization. Kinetic modeling and particle size measurements show that emulsion polymerizations with initial concentrations close to the microemulsion–emulsion phase boundary demonstrate relatively fast consumption of monomer droplets and produce smaller particles. Because of their high surfactant concentrations, none of the emulsion polymerizations examined demonstrate the classical Smith–Ewart kinetics usually associated with emulsion polymerization. Instead these emulsion polymerizations have a long period of particle nucleation that subsides only after the disappearance of monomer droplets. © 2004 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 42: 5253–5261, 2004     

The inverse emulsion polymerization of acrylamide using polystyrene-graft-polyoxyethylene as the stabilizer

Inverse emulsion polymerization of aqueous solution of acrylamide (AM) in toluene is carried out using polystyrene-graft-polyoxyethylene (PSt-g-PEO) as an emulsifier. The kinetics of polymerization, morphology of the particle, and particle size of the inverse emulsion have been investigated. The rates of polymerization are found to be proportional to the initiator concentration, the monomer concentration, and the emulsifier concentration. The morphology of the particle shows a spherical structure. The effects of amphipathic graft copolymer structure on the average molecular weight of polyacrylamide are studied. The mechanism of the inverse emulsion polymerization using amphipathic graft copolymer as emulsifier is proposed. © 1999 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 37: 2719–2725, 1999