Solution reactivity ratios

An example of a commercial semibatch polymerization process is the early Union Carbide process for Dynel, one of the first flame-retardant modacryhc fibers (23,24). Dynel, a staple fiber that was wet spun from acetone, was introduced in 1951. The polymer is made up of 40% acrylonitrile and 60% vinyl chloride. The reactivity ratios for this monomer pair are 3.7 and 0.074 for acrylonitrile and vinyl chloride in solution at 60°C. Thus acrylonitrile is much more reactive than vinyl chloride in this copolymerization. In addition, vinyl chloride is a strong chain-transfer agent. To make the Dynel composition of 60% vinyl chloride, the monomer composition must be maintained at 82% vinyl chloride. Since acrylonitrile is consumed much more rapidly than vinyl chloride, if no control is exercised over the monomer composition, the acrylonitrile content of the monomer decreases to approximately 1% after only 25% conversion. The low acrylonitrile content of the monomer required for this process introduces yet another problem. That is, with an acrylonitrile weight fraction of only 0.18 in the unreacted monomer mixture, the low concentration of acrylonitrile becomes a rate-limiting reaction step. Therefore, the overall rate of chain growth is low and under normal conditions, with chain transfer and radical recombination, the molecular weight of the polymer is very low. 

GopolymeriZation Initiators. The copolymerization of styrene and dienes in hydrocarbon solution with alkyUithium initiators produces a tapered block copolymer stmcture because of the large differences in monomer reactivity ratios for styrene (r < 0.1) and dienes (r > 10) (1,33,34). In order to obtain random copolymers of styrene and dienes, it is necessary to either add small amounts of a Lewis base such as tetrahydrofuran or an alkaU metal alkoxide (MtOR, where Mt = Na, K, Rb, or Cs). In contrast to Lewis bases which promote formation of undesirable vinyl microstmcture in diene polymerizations (57), the addition of small amounts of an alkaU metal alkoxide such as potassium amyloxide ([ROK]/[Li] = 0.08) is sufficient to promote random copolymerization of styrene and diene without producing significant increases in the amount of vinyl microstmcture (58,59). 

Epichlorohydrin Elastomers without AGE. Polymerization on a commercial scale is done as either a solution or slurry process at 40—130°C in an aromatic, ahphatic, or ether solvent. Typical solvents are toluene, benzene, heptane, and diethyl ether. Trialkylaluniinum-water and triaLkylaluminum—water—acetylacetone catalysts are employed. A cationic, coordination mechanism is proposed for chain propagation. The product is isolated by steam coagulation. Polymerization is done as a continuous process in which the solvent, catalyst, and monomer are fed to a back-mixed reactor. Pinal product composition of ECH—EO is determined by careful control of the unreacted, or background, monomer in the reactor. In the manufacture of copolymers, the relative reactivity ratios must be considered. The reactivity ratio of EO to ECH has been estimated to be approximately 7 (35—37). 

AGE-Gontaining Elastomers. The manufacturing process for ECH—AGE, ECH—EO—AGE, ECH—PO—AGE, and PO—AGE is similar to that described for the ECH and ECH—EO elastomers. Solution polymerization is carried out in aromatic solvents. Slurry systems have been reported for PO—AGE (39,40). When monomer reactivity ratios are compared, AGE (and PO) are approximately 1.5 times more reactive than ECH. Since ECH is slightly less reactive than PO and AGE and considerably less reactive than EO, background monomer concentration must be controlled in ECH—AGE, ECH—EO—AGE, and ECH—PO—AGE synthesis in order to obtain a uniform product of the desired monomer composition. This is not necessary for the PO—AGE elastomer, as a copolymer of the same composition as the monomer charge is produced. AGE content of all these polymers is fairly low, less than 10%. Methods of molecular weight control, antioxidant addition, and product work-up are similar to those used for the ECH polymers described. 

Similar anomalies have been encountered by several workers in the bulk and solution polymerization of this monomer induced by classical free-radical initiators84-86) also, particularly low rates of conversion were observed. The most thorough kinetic study was carried out by Aso and Tanaka86) who again found normal results and a value of k jkt much lower than that for styrene. Copolymerization studies of 2-vinylfuran (Mj) have given the following values of the reactivity ratios ... 

Numerical approaches for estimating reactivity ratios by solution of the integrated rate equation have been described.124 126 Potential difficulties associated with the application of these methods based on the integrated form of the Mayo-kewis equation have been discussed.124 127 One is that the expressions become undefined under certain conditions, for example, when rAo or rQA is close to unity or when the composition is close to the azeotropic composition. A further complication is that reactivity ratios may vary with conversion due to changes in the reaction medium. 

The solvent in a bulk copolymerization comprises the monomers. The nature of the solvent will necessarily change with conversion from monomers to a mixture of monomers and polymers, and, in most cases, the ratio of monomers in the feed will also vary with conversion. For S-AN copolymerization, since the reactivity ratios are different in toluene and in acetonitrile, we should anticipate that the reactivity ratios are different in bulk copolymerizations when the monomer mix is either mostly AN or mostly S. This calls into question the usual method of measuring reactivity ratios by examining the copolymer composition for various monomer feed compositions at very low monomer conversion. We can note that reactivity ratios can be estimated for a single monomer feed composition by analyzing the monomer sequence distribution. Analysis of the dependence of reactivity ratios determined in this manner of monomer feed ratio should therefore provide evidence for solvent effects. These considerations should not be ignored in solution polymerization either. 

Barrett and Thomas (10)proposed that these effects of differential monomer adsorption could be modeled by correcting homogeneous solution copolymerization reactivity ratios with the monomer s partition coefficient between the particles and the diluent. The partition coefficient is measured by static equilibrium experiments. Barrett s suggested equations are ... 

Compositionally uniform copolymers of tributyltin methacrylate (TBTM) and methyl methacrylate (MMA) are produced in a free running batch process by virtue of the monomer reactivity ratios for this combination of monomers (r (TBTM) = 0.96, r (MMA) = 1.0 at 80°C). Compositional ly homogeneous terpolymers were synthesised by keeping constant the instantaneous ratio of the three monomers in the reactor through the addition of the more reactive monomer (or monomers) at an appropriate rate. This procedure has been used by Guyot et al 6 in the preparation of butadiene-acrylonitrile emulsion copolymers and by Johnson et al (7) in the solution copolymerisation of styrene with methyl acrylate. 

It has been found that the monomer reactivities vary by the method of polymerization, even for radical polymerization in different environments, i.e., in bulk, solution and microemulsion (5). The reactivity ratios for different methods of polymerization are given in Table 10.2... 

Although the 5 1 volume ratio solutions can be used for coating applications, the gellike material that results on solvent removal limits the copolymer s utility. Fortunately, the reactive Si—H bonds that remain in the copolymer can be used to further modify polymer properties. The presence of the catalyst in the final reaction solution provides a simple means of further modifying the original copolymer. Thus, after 72 h of reaction,... 

Given the preceding mechanistic discussion, one would not expect the reactivity ratios to represent true kinetic parameters. Indeed, the reactivity ratios are sensitive to monomer and simple electrolyte concentrations, comonomer feed compositions, temperature and whether the reaction was carried out in aqueous solutions or a heterophase system. Clearly the data from the various groups are partly inconsistent, but still some general conclusions regard DADMAC/AAM copolymerization, listed in Table 6, are reasonable. 

A method for calculating apparent reactivity ratios based on run number theory has been applied to "starved-feed" styrene/ ethyl acrylate systems. The reactivity ratios found are in agreement with those determined from solution polymerization data. The further confirmation of the observed agreement between reactivity ratios determined at low conversions and those determined by run number theory in "starved-feed" high conversion copolymerization requires the analysis of other comonomer pairs. 

The initial copolymer composition corresponds well to reactivity ratios measured 03, 7) from bulk or solution copolymerization (r =0.13 Tg = 0.34) taking into account not the whole monomer feed, but its composition within particles. So, the initial copolymer composition is practically kept constant as long as droplets remain. After their disappearance, the polymerization rate remains constant up to about 50 % conversion. 

Assuming copolymerization takes place chiefly in the particles, there is a good agreement between experimental average composition and calculated one using "solution" reactivity ratios (0.13 and 0.34). 

The vinyl chloride and vinyl bromide copolymer samples were prepared as polymer analogs of PVC. Cleavage of the carbon-halogen bond at LNT should result in the same radical for these polymers. Since secondary C-Cl and C-Br bond dissociation energies are 73 and 59 Kcal/mole respectively, a copolymer of vinyl chloride and vinyl bromide could be regarded as a PVC chain with weak points. The feed ratio of VC/VB was 10/1 by volume. Since the reactivity ratios are (in solution, at 40°C), r = 0.825 for VC and r2 = 1.050 for VB (22), the copolymer composition should be 15 units of vinyl chloride for one unit of vinyl bromide on the average. In addition, the values for r and... 

---