Regiosequence

Cais, R.E. Kometani, J.M. Polymerization of vinylidene fluoride-d2. Minimal regiosequence and branch defects and assignment of preferred chain-growth direction from the deuterium isotope effect. Macromolecules 1984, 17, 1887. 

General Features of PVF Spectra The proton- and fluorine-decoupled 22.62 MHz carbon-13 NMR spectra of PVF prepared from PVCF (a) and commercial PVF (b) are shown in Figure 3. There are five additional peaks present in spectrum (b) from the commercial polymer which are absent in spectrum (a). These are due to aregic monomer sequences, which have been assigned according to Tonelli et al. (16). Monomer sequence triads are resolved and are denoted by the binary regiosequence pentad notation in Table 1 (l = CFH, O = CH2). 

The probabilities of the regiosequence pentads for commercial PVF and urea PVF are shown in Table III. For the former sample it is apparent simply by inspection that the regiosequence distribution is not Bernoullian, since Pobs(C5) and Pobs (D5) are different (2). The distributions conform to first-order Markov statistics, characterized by two reactivity ratios r0 and r 5 where r0 = k /lq, and rj — ku/k10 (kjj is the rate constant for monomer addition to terminal radical i which generates the new terminal radical j). The present pentad data is insufficient to check the validity of this model, but it is unlikely that there is any deviation, as the same model has been tested and found adequate to describe the regiosequence distribution in PVF2 (2). 

Table IV shows the reactivity ratios rG and r, derived from the probabilities in Table III in accord with a first-order Markov model (2), where it is assumed that the more likely propagating terminal radical structure is 1 (—CHF-) and not 0 (—CH2). This assumption is consistent with gas phase reactions of VF with mono-, di-, and trifluoromethyl radicals, which add more frequently to the CH2 carbon than to the CHF carbon (20). The reactivity ratio product is unity if Bernoullian statistics apply, and we see this is not the case for either PVF sample, although the urea PVF is more nearly Bernoullian in its regiosequence distribution. Polymerization of VF in urea at low temperature also reduces the frequency of head-to-head and tail-to-tail addition, which can be derived from the reactivity ratios according to %defect — 100(1 + ro)/(2 + r0 + r,). Our analysis of the fluorine-19 NMR spectrum shows that commercial PVF has 10.7% of these defects, which compares very well with the value of 10.6% obtained from carbon-13 NMR (13). Therefore the values of 26 to 32% reported by Wilson and Santee (21) are in error.

Whereas the stereosequence distribution in isoregic and aregic PVF is nearly ideal random (Bernoullian with p(m) — 0.5), the latter has a first-order Markov regiosequence distribution. Accordingly the monomer sequence isomerism in PVF cannot be described by a single parameter such as the % defect, and requires two reactivity ratios for complete specification. 

Table HI. Observed Probabilities of Regiosequence Pentads (P0bS(S5) as defined in reference 2) for Aregic PVF Samples ...

Aregic PVF, regiosequence distributions, 160-63 Aromatic C-1 resonance, epimerized isotactic PS, 202-11 Atactic polypropylene, 3C spectra, 7 Attached proton test (APT), optimizing sensitivity, 99... 

The matter of the head-to-head, tail-to-tail polymerization of vinyl fluoride, vinylidene fluoride, and trifluoroethylene and the copolymerization of vinyl fluoride with vinylidene chlorofluoride and l-chloro-2-fluoroethylene has been extensively studied by Cais and Kometani [24-27] and by Bruch, Bovey, and Cais [28]. The synthesis of pure head-to-tail poly(trifluoroethylenes) is described in Ref. [25]. Isomers of poly(vinyl fluoride) with controlled regiosequence microstructure are discussed in Ref. [27]. 

---