Solvent-free generation of poly(vinyl acetals) directly from poly.
Poly(vinyl butyral) (PVB) is produced industrially by the acid-catalyzed acetalization of poly(vinyl alcohol). Because vinyl alcohol is an unstable compound, poly(vinyl alcohol) must first be produced by the base or acid catalyzed hydrolysis/alcoholysis of poly(vinyl acetate). This two-stage or sequential procedure for the production of poly(vinyl butyral) is inherently solvent and time intensive. While a simultaneous reaction procedure is currently used to produce poly(vinyl formal) directly from PVAc, it has been suggested that sequential alcoholysis and acetalization are required to produce PVB with low residual acetate contents (1).
Methods by which poly(vinyl acetals) may be produced from poly(vinyl alcohol) (PVOH) tend to vary primarily as to the solvent used (1-3). Homogeneous solution techniques are preferred industrially because of superior reaction kinetics and a reduction in cross-linking relative to heterogeneous techniques (2). Utilizing a solvent in which PVOH and the initial reactants are soluble, but where the poly(vinyl acetal) is insoluble, also provides for easy purification and separation of the product. These requirements necessitate the use of polar organics or water, which ultimately limits the types of aldehydes that may be used in the acetalization to those with sufficient solubility in these polar media. This is one reason that the most widely produced poly(vinyl acetals) are poly(vinyl formal) (PVF) and poly(vinyl butyral) (PVB). Approximately 100,000 metric tons of PVB are produced per year worldwide for applications ranging from automobile safety glass to adhesives for printed circuit boards and reprographic toners (1, 4-6).
Given the relative simplicity of the chemistry involved, one could in principle produce poly(vinyl butyral) (PVB) from poly(vinyl alcohol) (PVOH) (or directly from poly(vinyl acetate) (PVAc)) using a continuous bulk process such as that employed for condensation polymer production or via reactive extrusion. Use of such solvent-free techniques would require thermal stability of the polymers at processing temperatures, and a homogeneous polymer phase to prevent mass transport restrictions and to maximize conversions. While PVAc and PVB are both amorphous with relatively low glass transition temperatures ([T.sub.g] [similar to] 40 and 70 [degrees] C respectively), PVOH is a semicrystalline polymer with a high melting point ([T.sub.m] [similar to] 230 [degrees] C) and degrades rapidly in the melt. Therefore a melt-phase process producing PVB from PVAc should be designed to circumvent the viscous PVOH intermediate
In the hydrolysis/alcoholysis of PVAc, the viscosity of the polymer during transformation to PVOH will increase dramatically, similar to that observed in current processes in alcoholic media (7). With increasing extent of OH substitution the polymer phase will tend to form a gel and finally a solid. This results in a dramatically reduced rate of hydrolysis/alcoholysis towards the latter stages of conversion, apparently due to diffusional restrictions. Consequently, in a simultaneous reaction scheme where the poly(vinyl acetal) is produced directly from PVAc, the rate of acetalization should be significantly higher than that of the initial hydrolysis/alcoholysis to ensure that the concentration of OH groups remains low. In this manner the polymer phase would remain plasticized and the viscosity remain low.
The objective of this study was to determine the feasibility of transforming poly(vinyl acetate) to poly(vinyl acetals) in solvent-free conditions and also during use of supercritical C[O.sub.2] as a processing aid. This study presents the rate of acid-catalyzed methanolysis of PVAc at C[O.sub.2] pressures ranging from 0 to 5000 psia and temperatures from 60 [degrees] C to 90 [degrees] C, Butyralizations of PVAc/PVOH copolymers are used in attempts to quantify the rate of butyralization of PVOH. Simultaneous methanolysis and butyralization reactions are used to quantify the conversion rates for PVB from PVAc and to determine the maximum degree of conversion possible with this technique. Finally, 3, 3-dimethylbutyraldehyde, 2, 3, 5, 6-tetramethylbenzaldehyde, dodecylaldehyde, and benzaldehyde are used to determine the ability of the simultaneous technique to produce various poly(vinyl acetals).
As is well known, the production of PVOH from PVAc proceeds by acid or base-catalyzed hydrolysis, where base catalysis is generally more rapid (8). The conversion of PVAc to PVOH shows a sigmoidal profile with time as a result of an auto-catalytic effect (7, 9), where hydroxyls may bind catalyst and increase the reactivity of neighboring acetate functionalities by [approximately]2 orders of magnitude. Sakurada (9) related the rate of conversion, dx/dt, to an auto-acceleration constant m as:
dx/dt = [k.sub.0] (1 + m(x/a)) (a - x) (b - x) =
[k.sub.0](1 + my) (1 - y),
where y is the degree of hydrolysis. The value of m has been determined to be as small as [approximately] 1.7 (7) to as large as 42 (9). In general the extent of conversion from PVAc to PVOH will have a dramatic effect on the viscosity of the bulk polymer phase. Thus it is observed that the rate of alcoholysis is initially rapid but becomes significantly reduced with increasing conversion, as viscosity of the polymer phase increases significantly (7).
Studies have shown that the method of preparation of PVOH has a direct effect on the physical character of the polymer (10). For example, saponification with NaOH in acetone/water solvent showed a high degree of "block" structure to the PVOH even at relatively low percentage conversions to PVOH. Methoxide catalyzed transesterification by NaOH in either anhydrous or aqueous MeOH led to a much lower degree of block structure to the product. Finally, acid catalyzed (HCl) hydrolysis produced an essentially random structure to the PVOH-PVAc copolymer. It has also been shown experimentally via infrared analysis (9) and by 13C NMR analysis (11) that the hydroxyl functionalities are much more structured when produced by base catalyzed saponification relative to acid catalyzed hydrolysis. The chemically dissimilar nature of PVAc and PVOH will likely result in phase separation of the two polymers when block-like copolymer is produced.
In the presence of acid, aldehyde is transformed to its conjugate acid analog. The reaction of the conjugate acid with PVOH produces the hemi-acetal in what is known to be the rate determining step for the production of poly(vinyl acetal) from PVOH. The formation of the hemi-acetal is reversible, but the reverse reaction occurs at a significantly slower rate. Protonation of the hemi-acetal hydroxyl, followed by the loss of water, produces an oxonium ion which quickly attacks the adjacent hydroxyl on PVOH to yield the cyclic acetal. Intermolecular reactions between the hemi-acetal and a hydroxyl function on a separate chain may lead to a cross-linked product, although this is not generally seen when acetalizations are performed in solution or when the bulk polymer phase is plasticized (1). Commercially produced poly(vinyl acetals) are commonly claimed to be [approximately]80-88% acetal, 10-20% alcohol, and 1-3% acetate.
Acetalization reactions generally follow second order kinetics; but experimental evidence suggests that the rate dependence of the formation of polyvinyl butyral on the concentration of vinyl alcohol repeats is first order, since the formation of the hemi-acetal is rate limiting. Statistical calculations suggest that the highest possible yield from the acetalization of PVOH is 86.5%. Steric and electrostatic effects have been shown to play roles in the extent of acetalization of PVOH. Substitution of the hydroxyl for bulky or ionizable substituents can lead to incomplete substitution ([less than]50%) by the aldehyde (12).
Methanolysis During Exposure to Carbon Dioxide
Polymer modifications in SC-C[O.sub.2] were performed in a 50 mL high pressure vessel (Pressure Products Inc.) capable of pressures up to 6000 psia and temperatures of 350 [degrees] C. The reactor was equipped with a high speed (max [approximately]2500 rpm) magnetically coupled impeller. The details of this apparatus are outlined elsewhere (13). In a typical experiment, 10 g of PVAc (Air Products and Chemicals VINAC B-15; 7.5 x [10.sup.4] g/mol), 0.22g (1.0% molar equivalent/VAc monomer units) of p-toluenesulfonic acid monohydrate (Aldrich, 98.5%), and 8.0 mL (1.7 molar equivalent/VAc monomer units) of anhydrous methanol (Sigma-Aldrich, HPLC grade) were introduced to the reactor. Stirring rate was 2000 rpm for all reactions. For reactions with C[O.sub.2], pressure was increased incrementally such that the required pressure was reached at thermal equilibrium which required approximately 15 minutes; reaction time was measured starting from the point at which thermal equilibrium was attained.
Reactions were halted by temperature quench via cold water piped through the heating jacket of the high pressure vessel. Initial cooling rate was [approximately] 10 [degrees] C/min, and [approximately] 10 minutes was required to reach room temperature. Products were dissolved in chloroform, filtered, and precipitated in a 20-fold excess of hexanes. All samples were dried for at least 48 hours under vacuum.
The quantification of percent conversion to PVOH was determined by mass balance using IR to quantify residual acetate content of the product through Beer's Law. Calibration of the carbonyl stretch of acetate was done using PVAc homopolymer standards which were purified by dissolution in CH[Cl.sub.3] and precipitation into hexanes. Standards were prepared from 0.005-0.05 g in 10 mL of spectrophotometric grade dimethyl sulfoxide (Aldrich). Analysis of copolymer samples were performed in triplicate on 0.3 g to 0.8 g polymer in 10 mL solvent.
Experimental evidence exists that indicates that hydrogen bonding between hydroxide and carbonyl functionalities can affect peak shapes and positions for PVAc-PVOH copolymers (11). For all copolymers studied here, peak shape was narrow, singular, and independent of copolymer content. Because peak position varied from 1736.5 [cm.sup.-1] for pure PVAc to 1732.5 [cm.sup.-1] for high PVOH content copolymers ([approximately]90 mole % VOH), the peak maximum was used for calibration. Accuracy of the calibration using PVAc homopolymer was verified using a commercial PVAc-PVOH copolymer sample (Air Products Airvol 502, nominally 87.7% PVOH), where quantitation by IR indicated a PVOH content of 89.5 mole %.
Butyralizations were performed in both a simultaneous reaction, producing PVB directly from PVAc via simultaneous addition of methanol and butyraldehyde, and sequentially from PVOH-PVAc copolymers produced by methanolysis. Butyralizations were performed with identical reagent loadings to methanolysis, with butyraldehyde (Aldrich, 99.5+%) at various loadings from 1.2 to 2.8 stoichiometric excess added to the reaction vessel in addition to the polymer, methanol, and acid catalyst. As such, simultaneous methanolysis and butyralization was performed at the vapor pressure of the reactant mixture at each temperature. All copolymers produced were dissolved in chloroform for 12 hours prior to filtering and precipitating into a 20-fold excess of hexanes. For butyralization of preformed PVAc/PVOH copolymers, 1 g polymer sample were used and all reagent quantities were reduced by a factor of 10. Elemental analysis indicated that the majority of the acid catalyst remained with the PVAc/PVOH copolymer after purification, so that only a small further addition of catalyst ([approximately]0.25 mole %) was required to ensure catalyst was present during butyralization.
Quantification of the extent of butyralization was determined by 1H-NMR (Bruker 300 MHz) using internal standards of dimethyl formamide, DMF (J.T Baker, 99.9+%) and tetramethyl silane, TMS (Aldrich, 99.9+%). Samples of [approximately] 0.10 g were dissolved in [approximately] 3 mL of CD[Cl.sub.3]. Prior to analysis, 10.0 [[micro]liter] of both DMF and TMS were introduced to the sample. Conversion to PVB was determined by integration of standards versus the -[CH.sub.3] absorption of the butyl functionality at [approximately] +0.91 ppm from TMS. Residual carbonyl content was determined from IR analysis as indicated above. Residual PVOH content was determined by mass balance given the PVB and PVAc content of the product.
Acetalizations were also performed with 3,3-dimethyl butyraldehyde, 2, 3, 5, 6-tetrafluorobenzaldehyde, dodecylaldehyde, and benzaldehyde. All aldehydes were obtained from Aldrich Chemicals Inc. and used as received. Reactions were performed as indicated above, starting from PVAc, at 90 [degrees] C for 6-7 hours with 2.0 mole % ([approximately]4.5 wt %) p-TSA [multiplied by] [H.sub.2]O catalyst and 1.2 equivalents of both methanol and aldehyde. Products were analyzed both by FT-IR, 1HNMR, and DSC.
RESULTS AND DISCUSSION
In the commercial production of PVOH from PVAc, a basic catalyst (typically aqueous NaOH) is used to saponify a 40 wt% PVAc/methanol solution, given that this combination produces a high rate of reaction under mild conditions. However, for simultaneous methanolysis and butyralization to be successful, the rate of methanolysis of PVAc should be the limiting step of the reaction. This will in turn lead to relatively [TABULAR DATA FOR TABLE 1 OMITTED] low steady state concentration of hydroxyl groups during the reaction, thus reducing the chance for the gelation of the polymer phase to occur, and reducing or eliminating viscosity increases. Consequently, we explored the use of acid catalysts (plus various ROH reagents) to conduct the alcoholysis/hydrolysis of PVAc at reasonable, yet somewhat slower rates than may be achieved using base catalysts.
Whereas p-toluene sulfonic acid/methanol was an obvious choice for study, attempts were also made to estimate the rate of conversion to PVOH using a variety of acidic catalysts other than from p-TSA [multiplied by] [H.sub.2]O, including glacial acetic acid and carbonic acid. The goal was a cursory study of some possible acidic systems which might be more amenable to processing and removal in a bulk phase system with C[O.sub.2] as an extractant. The conversion was found to be minimal relative to p-TSA at similar conditions (Table 1). These initial results suggest that the acetate group is not susceptible to
attack by weak acid catalysts. It may also be inferred from these results that the auto-acceleration effect observed previously is not likely to be caused by the production of acetic acid from hydrolysis of PVAc by the water introduced with p-TSA [multiplied by] [H.sub.2]O.
Alcoholysis was also performed with ethanol as reagent to observe its effects on reaction rate. Consistent with the literature (3, 8), the conversion rate to PVOH is reduced by replacing methanol with ethanol. The rate of conversion to PVOH is still significant, however, and the by-product of this reaction is ethyl acetate, as compared with the methyl acetate that is produced from methanolysis. The results summarized in Table 1 show that the combination of methanol plus p-toluene sulfonic acid provided the best results, and so this pair was used in subsequent work.
Methanolysis Using p-TSA
PVAc-PVOH copolymer products of methanolysis were collected by dissolution in chloroform, and reprecipitation in a 10:1 excess of hexanes. The copolymer is completely soluble in chloroform for average molar conversions less than [approximately]60%. Polymers with higher conversions (up to [approximately]80%) are partially insoluble yet highly swollen by chloroform. Conversions for these polymers were estimated from mass fraction averages of the respective percentage conversions determined for each fraction by IR quantitation in DMSO, a solvent for these copolymers to very high percentage OH. The production of partially soluble products, with large variation in the carbonyl content of each fraction (i.e. the soluble fraction of such high PVOH-content copolymers was usually 40-50% PVOH, and the insoluble fraction [greater than] 70%) suggests that percentage conversion varies dramatically among polymer chains producing a chemically polydisperse copolymer system.
Rationale for Use of C[O.sub.2]
Plasticizers can be used to reduce the temperature required for reasonable flow rates in reactive extrusion, yet the presence of plasticizer in the final product may be undesirable and its removal (devolatilization) may not always be straightforward. Use of a high pressure gas as plasticizer would avoid the problem of plasticizer residues. For example, carbon dioxide can exhibit high solubility in polymers (14-20), plasticizing the polymer matrix by increasing the free volume of the mixture relative to the pure polymer. When in contact with C[O.sub.2] these polymers will swell and a concomitant reduction in the [T.sub.g] of the polymer phase is observed. As a result, the increased mobility of molecules within the plasticized matrix allow for enhanced diffusion rates in the polymer phase and potentially eliminate mass transport limitations that may exist during polymer processing (15, 18, 19). The plasticization effect can then be reversed by depressurization, Relative to other commodity polymers C[O.sub.2] has high solubility in both PVAc (14, 15) and PVB (21). For the reaction process studied here C[O.sub.2] may also act as a compatibilizing agent, allowing for various aldehydes and mixtures of aldehydes to be used in the production of novel poly(vinyl acetals) (22-24). In the absence of diffusional restrictions, the use of C[O.sub.2] will lower the rate of reaction (relative to the bulk or melt-phase case) via dilution of the reactants, yet a decreased tendency for cross-linking during acetal formation may be observed as well.
Although the solubility of C[O.sub.2] in PVOH has not been quantified to our knowledge, because of the incompatibility of the two phases (i.e. the nonpolar nature of C[O.sub.2] versus the polar, hydrogen bonded, and semicrystalline PVOH phase) it is expected that the degree of swelling of PVOH will be low and thus significant plasticization of PVOH is not expected (25). Consequently, in a simultaneous reaction scheme [TABULAR DATA FOR TABLE 2 OMITTED] where the poly(vinyl acetal) is produced directly from PVAc, even in the presence of C[O.sub.2], the rate of acetalization should be significantly higher than that of the initial hydrolysis/alcoholysis to ensure that the concentration of OH groups remains low. In this manner the polymer phase would remain plasticized and the viscosity remain low.
Methanolysis was performed at three conditions: (a) no C[O.sub.2] where the pressure was simply the vapor pressure of methanol at the reaction temperature, (b) 850 psi C[O.sub.2], and (c) 5000 psi C[O.sub.2] and at multiple temperatures for each condition (Table 2). In each case the molar ratios of methanol:acetate and catalyst:acetate were held constant. As expected, the rate of reaction was highest for reactions without C[O.sub.2], for three basic reasons. First, the presence of a slight excess of methanol and temperatures in excess of the glass transition of PVAc were clearly sufficient to plasticize the system and eliminate any mass transfer limitations. Second, C[O.sub.2] is known to swell PVAc extensively; literature indicates that at 80 [degrees] C and 850 psia that C[O.sub.2] is soluble to about 12 wt% in PVAc (14). Thus, exposure of the polymer to C[O.sub.2] will lead to swelling and dilution of the reactants, slowing the rate accordingly. Third, the system present in the reactor will be two-phase, where a polymer-rich phase exists in equilibrium with a C[O.sub.2]-rich phase. Methanol will partition between the phases, with preference for the C[O.sub.2]-phase increasing upon raising pressure which lowers reactant concentration in the polymer as pressure increases. However, as the reaction proceeds, the methyl acetate by-product will also partition into the vapor phase, and as such the reaction rate may be enhanced by increasing pressure through the removal of this by-product. Without C[O.sub.2] present methanol and methyl acetate will again partition between polymer and vapor phases, although to a much lesser extent than for analogous reactions under C[O.sub.2] pressure. In any case, while C[O.sub.2] may slow the reaction, it may also allow the reaction to attain maximum conversion to product.
At 5000 psia the initial conversion rate increases from about 8%/hr at 70 [degrees] C to almost 60%/hr at 90 [degrees] C. This corresponds to reaction rates of 7.2 x [10.sup.-5] M/sec and 5.5 x [10.sup.-4] M/sec, respectively. Because the exact partitioning of the methanol and carbon dioxide between polymer and gas phases is at present unknown, it is not possible to extract rate constants from this data at this time.
Butyralizations of PVAc/PVOH Copolymers
PVOH/PVAc copolymers produced and characterized in this study were utilized in the subsequent butyralizations. As can be seen from the Table 3, conversion to butyral remains unchanged after short reaction times at relatively low temperatures. The maximum [TABULAR DATA FOR TABLE 3 OMITTED] [TABULAR DATA FOR TABLE 4 OMITTED] conversion of PVOH to PVB for these copolymers is approximately 60-65% of the available hydroxyl groups, lower than the statistical maximum of 86.5%. This result is likely a consequence of the random nature of PVOH/PVAc copolymers produced with acid catalysts versus the more blocky structures generated using base catalysts (10). This data shows that, as desired, the rate of butyralization is significantly faster than the rate of methanolysis. For example, while it requires less than 1 hour at 60 [degrees] C to achieve maximum conversion to butyral for the copolymer, several hours at 90 [degrees] C are needed to obtain high conversion of PVAc to PVOH. These results are in agreement with literature data on the single-stage formalization of PVAc, where it is known that the rate of hydrolysis/alcoholysis is rate determining (8, 26).
Simultaneous Reaction Process for Poly(Vinyl Butyral) in the Absence of C[O.sub.2]
The simultaneous reaction procedure involves introducing butyraldehyde to the identical reaction system used for methanolysis, where the effect of various aldehyde excesses on the reaction rate was measured. The reaction product was slightly tan colored, with the amount of color increasing with reaction time. The products were all soluble in chloroform, suggesting negligible build-up of PVOH during the process. Those products which were found to have high conversions to butyral were also found to exhibit a rubbery consistency after precipitation in hexanes.
The rate of butyralization was studied with 1) 1.7 equivalents of methanol and 2.8 equivalents of butyraldehyde, 2) 1.2 and 2.4 equivalents, and 3) 1.2 and 1.2 equivalents respectively. If one assumes that the volume change upon mixing ([Sigma][V.sub.m]) is small for the PVAc-methanol-butyraldehyde ternary, then the amounts of the various reactants can be used with their respective densities to determine the initial concentration of each reactant under the conditions described above. The results of these calculations are shown in Table 4.
The conversion data show a number of interesting trends. To illustrate this the data at 80 [degrees] C in the absence of C[O.sub.2] are shown in Fig. 1. The remainder of the data is shown in Tables 5a-c. From Fig. 1, for the most part, we do not observe the presence of significant amounts of hydroxyl repeats in the final product, suggesting that methanolysis is indeed the rate-limiting step in the reaction sequence at 80 [degrees] C. The continuous conversion of PVAc to PVB with only a small build-up of PVOH agrees with the prediction made from the results of separate methanolysis and butyralization reactions. Because of the inherent error in the NMR and IR procedures for determining butyral and acetate concentrations, the mass balance procedure for generating hydroxyl content can sometimes produce negative numbers.
One might initially expect that the overall reaction rate in the direct conversion of PVAc to PVB would be similar to that of methanolysis of PVAc. However, the overall rate is actually reduced because the addition of the aldehyde dilutes all reagent concentrations, reducing the rate of the initial methanolysis. With decreases in the quantities of methanol and butyraldehyde to 1.2 equivalents, there is a concomitant increase in the conversion rate despite the use of smaller reagent excesses. If we assume that methanolysis is the rate determining step, then we should expect that the quantity [PVAc]*[methanol], in addition to temperature and catalyst concentration, should govern the initial rate of reaction of the polymeric acetate groups with methanol. As shown in Table 4, reducing the excess of butyraldehyde acts to concentrate the methanol and acetate, leading to higher overall conversion rates.
In two of the data sets (Tables 5a and 5c), the runs at 70 [degrees] C and three hours suggest that at longer times [TABULAR DATA FOR TABLE 5A OMITTED] at this temperature butyralization is actually rate-limiting, rather than methanolysis, contrary to what is observed at 80 [degrees] C. It is not clear why this behavior has occurred.
One reason given that simultaneous reaction of aldehyde and methanol with poly(vinyl acetate) to produce PVB is not practiced commercially is that the simultaneous process does not allow for generation of PVB with sufficiently low acetate content. Consequently, we compared a sample of PVB generated at 90 [degrees] C and 7 hours, with 1.2 equivalents of both methanol and butyraldehyde to a commercial sample of PVB (Butvar B-79, Monsanto) via elemental analysis performed at Galbraith Laboratories (Table 6), as well as IR and NMR. All analyses indicate that the two materials are nearly identical in composition. It is noteworthy that a 1.2 stoichiometric excess of methanol with PVAc results in a polymer phase composition of 60 wt% PVAc, significantly higher than that used currently, and still allows for complete conversion of PVAc to PVOH.
A potential side reaction, that between methanol and butyraldehyde to produce the methyl-based acetal, did not interfere in the generation of PVB with high butyral content. This is rationalized by the fact that the reaction between methanol and butyraldehyde likely proceeds at a much lower rate than the primary reaction, given that the former is third order. As well, the methyl-based butyral produced from this side reaction may, fortuitously, react with hydroxyls on the polymer to produce the desired poly(vinyl butyral), as shown by Ichimura (27).
Regarding the phase behavior of the system, the reaction system was examined in a high pressure, variable volume view cell (13), simply to confirm the presence of two phases in those reactions employing carbon [TABULAR DATA FOR TABLE 5B OMITTED] [TABULAR DATA FOR TABLE 5C OMITTED] dioxide. In the "no C[O.sub.2]" case, the upper phase contains only methanol and aldehyde vapor, so consequently only a small fraction of the reactants added to the reactor does not affect the outcome of the reaction. When C[O.sub.2] is used at relatively low pressure, the upper phase is again low density, given the temperatures employed, and thus the "loss" of material from the lower, polymer-rich phase, is again probably not significant. Upon use of C[O.sub.2] at high pressure, the partitioning of the reactants between the phases could indeed lower the concentration of the reactants in the polymer-rich phase. Because methanol is likely to partition more strongly to the C[O.sub.2]-phase, such partitioning would tend to reduce the rate of methanolysis more so than acetalization.
Production of Poly(Vinyl Acetals) From Various Aldehydes
Commonly used techniques for the production of poly(vinyl acetals) use PVOH as the starting material in homogeneous reaction processes. This necessitates the use of water or polar organics (solvents or swelling agents for PVOH) as media, which ultimately limits the types of aldehydes which can be used in the homogeneous process. For aliphatic aldehydes, butyraldehyde is the highest molecular weight aldehyde which still retains any significant solubility in polar or aqueous media. As the pendant group of the acetal will have a tremendous effect on the physical character of the polymer, it is desirable to have an efficient synthesis route for the production of a diversity of poly(vinyl acetals) (28). For example, various aldehydes and aldehyde mixtures have been used to produce internally plasticized PVB, reducing or eliminating the need for added chemical plasticizers in the production of laminate sheet (24). All poly(vinyl acetals) produced were characterized by IR, NMR, and DSC.
Reactions were performed with various aldehydes by the solvent-free technique without C[O.sub.2] present. Both methanol and aldehyde were held at 1.2 equivalents to PVAc in an attempt to determine the maximum conversion to acetal that could be obtained by this technique (Table 7). In the case of dodecylaldehyde, for example, the final product was golden in color and completely soluble in chloroform. Analysis by 1H-NMR and IR indicate a conversion to PVDD of [greater than]70%, with the majority of the residual being PVAc. The final product was precipitated Into methanol as hexane was not an effective anti-solvent for the product. The purified PVDD was rubbery and somewhat tacky, with a [T.sub.g] of approximately 4 [degrees] C.
Poly(vinyl tetrafluorobenzal) (PVTFBz) was partially insoluble in chloroform, with the insoluble fraction highly swollen by chloroform and insoluble in a variety of other solvents, suggesting that some cross-linking had occurred. The product was reddish in color with a similar consistency to PVB. Characterization of PVTFBz indicated an extent of conversion of almost 60%, also having a large fraction of residual acetate. The results from both the PVDD and PVTFBz experiments suggest that the technique can be used for a variety of aldehydes, and possibly aldehyde mixtures. Further, they indicate that the extent of reaction is not limited by the reactivity of the aldehyde, steric effects from the bulky aldehydes, or mass transport limitations due to the build-up of PVOH, but by the conversion of PVAc to PVOH.
Both isobutyraldehyde and benzaldehyde products were essentially Insoluble in chloroform. While the product from isobutyaldehyde was highly swollen and slightly soluble initially in chloroform, the product isolated could not be redissolved. The final product of the acetalization with benzaldehyde was a black mass of insoluble char, completely insoluble in all solvents. [TABULAR DATA FOR TABLE 6 OMITTED] [TABULAR DATA FOR TABLE 7 OMITTED] Characterization of products from acetalization with isobutyraldehyde and benzaldehyde could not be performed, and suggest that further study need be conducted on the effects of side reactions and/or cross-linking in these systems.
The simultaneous methanolysis and butryalization of poly(vinyl acetate) were conducted via the reaction of methanol and butryaldehyde with polyvinyl acetate under batch conditions at temperatures from 70 [degrees] C to 90 [degrees] C at the vapor pressure of the reactants. It was found that use of an acid catalysis allowed for the methanolysis to be the rate limiting step in the sequential reaction, minimizing the buildup of alcohol repeat units in the polymer during the reaction. As such, use of an excess of aldehyde was counterproductive, in that it served to dilute both the methanol and the acetate groups in the polymer phase, lowering the overall rate of the reaction. Carbon dioxide was evaluated as a reversible plasticizer for PVAc during methanolysis, and results at various pressures were consistent with the expectation that the presence of C[O.sub.2] would lower the rate, primarily because of dilution of reactants. Finally, it was shown that the simultaneous reaction procedure can be used to generate a variety of polyvinyl acetals from aldehydes of various chemical nature. These results suggest that a solvent-free technique, where only slight stoichiometric excesses of reagents are used, may provide a novel method to produce various poly(vinyl acetals) and mixed acetals without use of, or isolation of, a PVOH intermediate polymer phase, which is inherently more difficult to handle. The results also support the use of C[O.sub.2] as a plasticizer for the polymer phase, and extractant for residual reagents and by-products in this process.
The authors wish to acknowledge Air Products and Chemicals, Inc., for funding for this project, and to Frank Prozonic (DSC) and Jack Frost (NMR) of APCI for analyses. In addition, the authors wish to thank Jennifer Stagno for help in the work-up of the polymer samples.
1. T. P. Blomstrom, Encyclopedia of Polymer Science and Technology, Vol. 17, 136-167 (1989).
2. K. Toyoshima, in Polyvinyl Alcohol - Properties and Applications, 1st Edition, 391-411, C. A. Finch, ed., John Wiley and Sons, New York (1973).
3. F. L. Marten, Encyclopedia of Polymer Science and Technology, Vol. 17, 167-198 (1989).
4. S. P. Pontis and B. M. Mehta, Popular Plastics, 12, 25 (1968).
5. C. A. Finch, ed., Polyvinyl Alcohol - Properties and Applications, John Wiley and Sons, New York (1973).
6. C. A. Finch, Polyvinyl Alcohol - Developments, p. 850, John Wiley and Sons, New York (1992).
7. M. Demonie, C. Cincu, C. Oprescu, and G. H. Hubca, J. Polym. Sci.: Polym. Symposium, 64, 245 (1978).
8. F. L. Marten and C. W. Zvanut, in Polyvinyl Alcohol - Developments, pp. 31-56, 57-76, C. A. Finch (ed.), John Wiley and Sons, New York (1992).
9. I. Sakurada, Pure Appl Chem., 16, 263 (1968).
10. I. Ahmed and J. G. Pritchard, Polymer, 20, 1492 (1979).
11. J. R. Isasi, L. C. Cesteros, and I. Katime, Macromolecules, 27, 2200 (1994).
12. G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Comprehensive Polymer Science, 6, Pergamon Press, New York (1989).
13. M. Super. E. Berluche, C. Costello, and E. J. Beckman, Macromolecules, 30, 368 (1997).
14. N. Wang, S. Takishima, and H. Masuoka, Kagaku Kogaku Rombunshu, 16, 931 (1990).
15. A. R. Berens, and G. S. Huvard, Supercritical Fluid Science and Technology, ACS Symposium Series #406, 207-25, K. P. Johnston and J. M. L. Penninger, eds. (1988).
16. R. G. Wissinger and M. E. Paulaitis, J. Polym. Sci.: Part B: Polymer Physics, 29, 631 (1991).
17. P. D. Condo and K. P. Johnston, J. Polym. Sci.: Part B: Polymer Physics, 32, 523 (1994).
18. M. L. O'Neill, PhD dissertation, Carleton University, Ottawa, Canada (December 1994).
19. M. L. O'Neill and Y. P. Handa, American Society for Testing Materials Symposium Series, STP 1249, p. 165, R. J. Seyler, ed. (1994).
20. Y. P. Handa, P. Kruus, and M. L. O'Neill, J. Polym. Sci.: Part B: Polymer Physics, 34, 2635 (1996).
21. Y. Kamiya, T. Hirose, K. Mizoguchi, and K. Terada, J. Poly Sci.: Part B: Polymer Physics, 26, 1409 (1988).
22. H. W. Matheson and G. O. Morrison, U. S. Patent no. 2,116,635, to Shawinagan Chemicals Ltd. (1938).
23. A. F. Fitzhugh and R. N. Crozier, J. Polym. Sci., 8, 225 (1952).
24. D. J. David and N. A. Rotstein, U. S. Patent no. 5,594,069, to Monsanto Co. (1997).
25. J. L. Kerschner, S. H. Jureller, and R. Harris, Polymer Mater Science and Engineering, #74, 246-47, Annual ACS symposium, New Orleans (1996).
26. S. Michitaka, and Z. Yokohama, Chem. High Polymers, 2, 196 (1945).
27. T. Ichimura, J. Polym. Sci.-Polym. Chem. Ed., 20, 1411 (1982).
28. P. Chetri, and N. N. Dass, J. Appl. Polym. Sci., 62, 2139 (1996).
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||vinyl acetate|
|Author:||O'Neill, Mark L.; Newman, Deborah; Beckman, Erik J.|
|Publication:||Polymer Engineering and Science|
|Date:||May 1, 1999|
|Previous Article:||Drawing of high density polyethylene tube: effect of different cooling conditions.|
|Next Article:||Reinforcement of polyamide 6 with thermotropic liquid crystalline polymer.|