N-Alliloksi piridinyum tuzlarının katyonik polimerizasyonda başlatıcı olarak kullanımı

Abstract

N-Alliloksi piridinyum tuzu, 2-etoksikarbonil-2-propenilpikolinyum heksafloroantimonat, ve N-alllloksi pikolinyum tuzu, N-[2-(metil)alliloksi]-o> pikolinyum heksafloroantimonat, radikalik başlatıcılar olan 2,2'-azobisisobutironitril, AIBN, benzoil peroksit, BPO, fenilazotrifenilmetan, PAT, ile ısısal katyonik polimerizasyonda çok iyi bir etkinlik göstermişlerdir. Radikalik başlatıcılann ısı etkisiyle, (70°C), oluşturduklan radikaller allilik tuzların çift bağına katılarak tuzların piridinyum tipi radikal katyonlara parçalanmalarına sebep olurlar. Sağlanan polimerizasyon hızı PAT > BPO > AIBN şeklinde artış gösterir. PAT ile sağlanan polimerizasyon hızındaki artış, ısısal etki ile oluşan trifenilmetil radikallerinin elektron transferi sonucunda katyona yükseltgenerek başlatıcı parçacıkları vermesiyle açıklanır. Alliloksi pikolinyum tuzunun monomer varlığında 280 nm'de fotopolimerizasyonu uygulanmıştır. Ayrıca serbest radikal kaynağı benzoin ile 370 nm'de yapılan fotopolimerizasyona, katılma-parçalanma mekanizması ile açıklık getirilmiştir. Kullanılan bu tuzların başlatıcı etkinlikleri; sikloheksenoksit, CHO, butilvinileter, BVE, bifonksiyonel (3,4-epoksisikloheksilmetil)-3',4'-sikloheksan karboksilat, EEC, gibi çeşitli monomerler ile gösterilmiştir.

Simple N-alkoxy pyridinium salts are thermally stable. Therefore, using these initiators no polymerization takes place upon heating in the presence of monomers prone to cationic polymerization. However, these salts may be reduced by electron donor radicals produced by thermolysis, for example, by the thermolysis of phenylazotriphenylmethane. The enusing carbocation is capable of initiating cationic polymerizations. OCH2CH3 r + 0CH2CH3 T Monomer Polymer Sheme 1 Initiation Mechanism with Electron Donor Radicals On the contrary, radicals originating from the thermolysis of common initiators such as benzoyl peroxide (BPO) and 2,2'-azobisisobutyronitrile (ABN) are not suitable for a polymerization according to Scheme 1, since these radicals are not oxidizable by pyridinium ions. Hydrojen donors are required to produce initiating species. Typical results of thermally induced polymerizations with the allyloxy pyridinium salt 2-ethoxycarbonyl-2-prophenyl picolinyum hexafluoroantimonate (EAPT), are compiled in Table 1. In these experiments, cyclohexene oxide (CHO) was deliberately chosen as a cationicalry polymerizable monomer to evaluate the polymerization mechanism. CHO can not be polymerized by a radical mechanism. VI Table 1. Bulk polymerization of CHO using the allyloxy pyridinium salt EAP+. [rad. source] = 5 x 10"3 mol l"1, [IJ = 5 x 1(T mol l"1 As İt is seen in Table 1, the thermolysis of AIBN and BPO readily promotes the cationic polymerization of CHO in the presence of the allyoxy pyridinium salt (EAP+). In this case, initiating species are produced by a radical addition fragmentation mechanism (see Scheme 2) İn a manner similar to that described for allyl sulphonium salts. The rupture of the N-0 bond as a relatively weak chemical bond is followed by the generation of various low molecular weight compounds which were identified by GC-MS analyses. Radical generation CH3 CH3- Cjî- N=h CN AIBN BPO ÇH3 A ' *? 2 CH3- Ç' + N2 CN 1 A ?*» 2\(~j) - c-ö J2 2 S> + co2 + R 4 vu er o N' SbF* Initiation 8 + Monomer Polymer Scheme 2. Radical addition-fragmentation. R' hydrojen donor, monomer or solvent 1, AIBN; 2 or 3, BPO. D-H = Together with compounds 5-6, pyridinium radical cations 8 are presumably formed. These species are potentially able to initiate the polymerization of any cationicaUy polymerizable monomer present. As was deduced from laser flash photolysis studies, nitrogen centered radical cations are reactive towards nucleophilic monomers. The bimolecular rate constants were found to be k = 106 - 107 and 5 x 109 1 mol"1 s"1 for the reaction with CHO and BVE, respectively. As can be inferred from GC-MS investigations a small portion of the radicals 4 undergo fragmentations other than those depicted in Scheme 2. For example, a-scission leads to the detected compound with olefrnic double bonds. Besides the direct initiation by radical cations 8, the initiation by Bnjmsted acid generated after hydrogen abstraction has also to be taken into account. This initiation route has been proven experimantally for other pyridinium salts. SbR= + H' vm H + Monomer -»- Polymer Scheme 3. D-H, hydrogen donor, monomer or solvent Apart from AIBN and BPO, phenylazotriphenylmethane (PAT) was used as a radical source for the cationic polymerization. PAT has the highest initiation activity of the three thermal initiators under investigation. This compound decomposes upon heating yielding phenyl and triphenylmethyl radicals. Whilst phenyl radicals are expected to participate in addition-fragmentation type initiation (see Scheme 2 and 3), triphenylmethyl radicals are oxidized by the pyridinium cation thus forming additional initiating species (see Scheme 4). The predicted formation of triphenylmethyl carbocation 10 was proven upon thermolysis of monomer free solutions of PAT and the allyloxy pyridinium salt(EAP") (see Fig 1). In the course of heating (3h at 70°C) the characteristic triphenylmethyl carbocation absorbtion bands at Xmax = 400 and 430 nm (S430nm = 3.9x1 04 1 mol"1 cm"1) appeared in the UV spectrum. Moreover, following the oxidation of triphenylmethyl radicals 9 further radicals 12 are formed following the electron transfer reaction. These may also add to the double bond initiating an addition-fragmentation event. o o 9 + EAPT SbF 10 + Monomer 11 11 Polymer Scheme 4. Formation of Triphenylmethyl Carbocation IX Apart from CHO, some other cationically polymerizable monomers, namely butylvinyl ether (BVE), p-methoxy styrene (pMeOSt) and (3,4- epoxycyclohexylmethyl)-3',4'-cyclohexane carboxylate (EEC) were examined. Like CHO, these monomers did polymerize readily in bulk or solutions containing BPO and the pyridinium salt EAP+. Typical results are compiled in Table 2. In the case of the bifunctional monomer EEC, an insoluble polymer network was formed. It should be noted that a slightly higher temperature (85°C) was chosen for the polymerization of EEC since at 70°C after 6 h of heating no gel formation was monitored. Most probably, the relatively high stationary concentration of radicals at higher temperature is a prerequisite for the formation of polymer. Table 2. Polymerization of various monomers using the salt EAP+ and BPO as a radical source. [EAP+] = 5xlO"3 mol \'\ [BPO] = SxlO"3 mol l"1. p. =; - C 300 400 500 Wavelength (nm) Fig. 1. UV Spectrum of the Monomer Free Solution of PAT and (EAP) Addition-Fragmentation Reactions For Cationic Polymerization Using a Novel Allyloxy-picolinium Salt As seen from Table 3, systems consisting of N-[2-(Methyl)allyioxy]-a-picolinyum hexafluoroantimonate (MAP*) in conjunction with various thermal free radical initiators were found to be efficient in polymerizing CHO. In these investigations CHO was chosen as a model monomer for it is not polymerizable by MAP+ in the dark at room temperature and cannot be polymerized by a radical mechanism. Furthermore, CHO does not form oxidizable radicals in the course of initiation. Cationically polymerizable monomers, such as vinyl ethers or tetrahydrofuran, were found to produce easily oxidizable free radicals while interacting with radicals stemming from external radical sources. In the precence of onium salts these monomer based radicals may be oxidized to the respective cations and initiate cationic polymerization. Although all free radical sources chosen are efficient in accelerating the polymerization, there are significant differences in polymerization rate dropping in the order PAT > BPO > AIBN, surely reflecting the different reactivity of the respective radicals formed upon thermolysis towards the allylic double bond of MAP+. In the case of PAT, no evidence for the formation of triphenylmethyl cations was found upon heating dichloromethane solutions containing MAP+ and PAT. For a similar initiator, EAP* (see Table 2), the formation of absorption bands at 400 and 430 nm during heating in the precence of PAT was observed previously. The absorption given owes to initiating triphenylmethyl cations that are generated upon oxidation of triphenylmethyl radicals by the pyridine moiety. Since no absorption is formed for the system MAP* / PAT, it has to be concluded that triphenylmethyl radicals are not oxidized by MAP+. Table 3. Bulk polymerization of CHO using MAP+, at 70 °C, [MAP+] = 6x lO^molT1, [Rad. source] = 5 x 10'3 mol l"1, nitrogen saturated For the sake of comparison, the well-known cationic photoinitiator N-ethoxy picolinium hexafluorophosphate (EMP*) was heated in the presence of CHO. At 70°C, even after several hours no polymer was formed indicating that the mechanism XI of thermal initiation does not involve the direct formation of picolinium type radical cations via rupture of the N-0 bond. If the N-0 bond was prone to thermally induced bond dissociation at 70°C, one should obtain polymer while heating EMP* in the presence of CHO. Since the reaction of CHO with MAPT at 70°C (without added radical initiator) yielded some polymer, it can be concluded that MAP* is by itself slightly more thermolabile than EMP+. Probably, while heating MAP", radicals are formed that add to the double bond of intact MAP* units and thus initiate addition- fragmentation. The reaction mechanism of the radical promoted polymerization using MAP* involves undoubtedly the production of radicals from the radical initiator and the subsequent addition of these radicals to the allylic double bond of MAP*. The energy rich intermediates thus produced undergo fragmentation yielding picolinium radical cations, that are known to react very efficiently with a variety of cationically polymerizable monomers. Radical Source SbF6 R ^ SbF, -*? R N+ O. T'jO SbF, R SbF6" \{-J\ + monomer N' polymer + Î>k/R Scheme 5. The Reaction Mechanism of the Radical Promoted Polymerization of MAP Salt Interestingly, with the present salt MAP*, epoxy products were not detected by the GC-MS analysis of methanol soluble fractions after polymerization. For EAP*, evidence for the formation of respective epoxy compounds was found under similar reaction and analysis conditions. It seems that, in the case of MAP*, the electron donating methyl substituent at the epoxy group makes the epoxy products more prone to polymerization. In other words, the epoxy type fragmentation products are likely to be incorporated into the polymer in the course of polymerization. As seen from Table 4, upon heating MAP* in the presence of BPO, various monomers may be polymerized. However, the initiation efficiency differs appreciably xn depending on the type of monomer being in accordance with earlier observed monomer selectivity of pyridinium type initiators. Thus, for CHO, the two vinyl ethers under investigation (butylvinyl ether, BVE, and isobutylvinyl ether, IBVE), for p- methoxystyrene, p-MeOSt, and N-vinylcarbazole, NVC, relatively high polymerization rates were found. Phenylglycidyl ether, PGE, on the other hand, could not be polymerized by the system presented. In the case of the afunctional (3,4- epoxycyclohexylmethyl)-3',4'-cyclohexane carboxylate, EEC, higher reaction temperatures had to be applied in order to achieve crosslinking within a reasonable period of time. Table. 4. Thermal Polymerizations at 70°C, [MAP+] = 6 x 10"3 mol l"1, [Radical Source] = 5 x 10"3 mol l"1, nitrogen saturated. (*)... Polymerization at 85°C Upon irradiation with UV light, MAP+ does initiate cationic polymerizations with very satisfactory yields. As isseen in Fig. 2, MAP+ is unambiguously superior to the also pyridinium based EMP+ initiator as far as initiation efficiency is concerned. In the experiments presented in Fig. 2, the absorbencies of MAP+ and EMP+ at the wavelength of incident light (280 nm) are identical. A much fester polymerization with MAP1" in comparison with EMP+ was also detected for other irradiation wavelengths and for both bulk and solution (CH2Cl2) polymerization. This phenomenon surely owes to the contribution of addition-fragmentation type reactions. As it is known from N-alkoxy pyridinium salts, the N-0 bond is photolabile and can easily be cleaved giving rise to initiating pyridinium radical cations and to alkoxy type radicals. The latter do presumably add to the double bond of intact MAP+ units and trigger then- fragmentation according to reactions in scheme 6. xm SbF, O. hv SbF, NT Scheme 6. Irradiation of MAP Salt at 280 nm Notably, following this reaction scheme, the absorption of one photon could create two initiating pyridinium radical cations rather than one as in the case of EMP+. Since MAP+ absorbs light only below 300 nm, it was tested whether polymerizations could be performed by addition of radical initiators with spectral response at higher wavelengths. As seen in Fig. 3, with benzoin as typical a-cleavage photoinitiator, high polymerization rates of CHO were achieved at 370 nm. Thus, an absorption region fitting most practical applications was reached. The high polymerization rates of the system MAPT / benzoin gives rise to the conclusion that the radicals stemming from benzoin are efficiently added to MAP+'s allylic double bond. 1.0i MAP* EMP+ 40 60 80 tin min 120 Fig. 2. Time-Conversion curves for EMP+ and MAP" salts. xiv 1.0-1 0.8- o 0.64 î £ 0.4 o O 0.2H 0.0 t in min Fig. 3. Time-Conversion curve for MAP/ Benzoin system