首页   按字顺浏览 期刊浏览 卷期浏览 Chapter 11. Post reactions of polymers
Chapter 11. Post reactions of polymers

 

作者: L. P. Ellinger,  

 

期刊: Annual Reports Section "B" (Organic Chemistry)  (RSC Available online 1973)
卷期: Volume 70, issue 1  

页码: 322-347

 

ISSN:0069-3030

 

年代: 1973

 

DOI:10.1039/OC9737000322

 

出版商: RSC

 

数据来源: RSC

 

摘要:

11 Post Reactions of Polymers By L. P. ELLINGER British Petroleum Company Limited Group Research and Development Department Epsom Division Epsom. Surrey 1 Introduction By post reactions of polymers we mean reactions of the preformed polymer in contrast to reactions leading to the formation of polymers. The variety of syn-thetic and natural organic polymers is great; polymers may contain any of the substituents and structural features known in smaller molecules. In most respects the reactions of these groups in polymers are very similar to their reactions in smaller molecules. An extensive account about ten years ago’ has been followed by further advances along established lines and important exten- sions into new directions. The chemical modification especially of the natural polymers proteins,2 and carbohydrates is the subject of an extensive literature.Within the very limited space available an attempt is made to select recent progress particularly where the reaction shows features ascribable to a polymer rather than to a small molecule. Technologically polymers are used in a wide variety of applications. Their suitability for particular applications can often be improved by modification of their molecular weight molecular weight distribution crystallinity stereo- chemistry and structure and may include structural modification by chemical reactions. Polymers are often solid and less soluble than the corresponding smaller molecules; particular substituents groups and even whole molecules can be rendered less soluble by attachment to suitable polymeric molecules.The features of polymers which distinguish them from small molecules are mainly associated with their high molecular weight and in linear polymers at least with the frequently regular and repetitive arrangement of smaller units corresponding closely to individual small molecules throughout the extent of the polymer mole- cule only the terminal groups differing somewhat from the centre units. Owing to their high molecular weight and viscosity polymer reactions are often viscosity controlled the solubility and the rate of diffusion of reagents and the mobility of the smaller units all being slowed. Many polymer molecules such as the poly- olefins are partially amorphous and crystalline.The reactivity of the same groups ’ ‘Chemical Reactions of Polymers’ ed. E. M. Fettes Interscience New York-London- Sydney 1964. * G. E. Means and R. E. Feeney ‘Chemical Modifications of Proteins’ Holden-Day San Francisco-Cambridge-London-Amsterdam 1971. 322 Post Reactions of Polymers in the amorphous and in the crystalline domains may differ sharply mainly again owing to differences of accessibility to the reagent. In many polymers for instance in synthetic polymers made with reagents exerting steric control (Ziegler-Natta catalysts anionic initiators) and in most natural polymers the substituents are sited in a regular sequence with often a very high degree of steric regularity along the main structure. The reactivity of regularly sited substituents may be greatly enhanced by this regularity which promotes concerted reactions over a range not often feasible between small separate molecules.The regular recurrence of particular substituents along a molecular chain affects the kinetics of their reactions. Kinetic expressions have been derived for the probability of a reaction involving a particular pair of substituents depending on whether 0 1 or 2 of the adjacent pairs have already reacted and the results have been compared with the experimental results for certain polymer reaction^.^ Such effects are thought to account very largely for the great catalytic efficiency and specificity of enzyme action and of protein and nucleic acid syntheses and reproduction. In view of limitations of experience and space these cannot be considered here but the design of polymeric supports of increasing regularity for groups known to be catalytically active and the study of the catalytic behaviour of the products is discussed.The chemical modification of both synthetic and natural polymers with a view to improved technological performance includes grafting cross-linking and curing. Only grafting in which it is aimed to attach identical or different polymer chains to the original molecule is considered here. Close control over the modification of product properties may be achieved and there have been important recent advances. Finally the use of polymer molecules as solid supports in stepwise synthesis for catalytic groups or for compounds of pharmacological photochemical or antioxidant reactivity is described.2 Complex Formation Complex formation is the simplest type of post reaction. Some complex types have been known for many years and recent work has led to better understanding of their nature ;a few examples are discussed. Oxygen shifts the optical absorption of liquid nucleophilic organic compounds to longer wavelengths ;the effect is ascribed to donor-acceptor interaction. Light absorption in the range 285-340 nm by polypropene is greater in an oxygen than in a nitrogen atm~sphere.~ The extra absorption has been correlated with an e.s.r. ~ignal.~.~charge-transfer band has also been observed for poly-4- A methylpent-l-ene-oxygen.6 A similar red shift by up to 350nm of the absorp- tion spectrum of polystyrene appears reversibly in the presence of oxygen.It is ' E. A. Boucher J.C.S. Faraday I 1972,68 2281. ' K. Tsuji and T. Seiki J. Polymer Sci.,Part B Polymer Letters 1970,8 517. V. K. Milinchuk Vysokomol. Soedineniya 1965 7 1923. ' G. W. Wood and T. M. Kollman Chem. and Ind. 1972,424. 324 L. P.Ellinger postulated that on irradiation the complex disintegrates with singlet oxygen formation.' This accounts for the photodegradation of polystyrene in the presence of oxygen but not in its absence by light in the range 280-340 nm. The complexes of iodine with oxygen-containing polymers particularly with starches are wellknown. The effect of the degree of polymerization of the amylose has been studied upon the absorption maximum of its iodine complex; this shifts with increasing degree of polymerization from 500 to 620 nm ;the extinction coefficient increases sharply between P, = 25-65 and the optical rotatory dis- persion and the circular dichroism reach maxima in the range P, = 70-100.8 The kinetics of amylose-iodine complex formation have been studied by the stopped-flow technique' and by pulse radiolysis' on aqueous amylose-KI solutions the change of absorption at 600 nm being followed.The detailed kinetic pattern depends on both the relative and absolute concentrations of the com- ponents ; at low iodine concentrations the rate of growth by successive uptake of iodine molecules is faster than the rate of chain initiation which is thought to involve the formation of stable nuclei in the he lice^.^ The conductivities and optical properties of complexes have been determined in which one of the components is a polymer containing a recurring regularly disposed substituent.The syntheses of two polymers containing a polyethyl- eneimine backbone carrying either N-4-[4-(methylthio)phenoxy]butyryl(1) or N-4-[(1O-methyl-3-phenothiazinyl)]butyryl (2) substituents on the backbone / McS~OCH2CH2CH2CO-N\ /CH CH (1) \2 Me / aynCH,CH,CH2CO-N \ nitrogens have been described.' With the ethyleneimine repeating unit in the polymer backbone the repeat distance of the substituent is 0.64-0.66nm. This together with the substituent design (a nucleophilic aromatic ring-system linked through a flexible short aliphatic chain perpendicularly to the backbone) B.RAnby J. F. Rabek and Z.Joffe presented at the conference on 'Degradability of Polymers and Plastics' The Plastics Institute London 1973 preprint 3/7. B. Pfannemuller H. Mayenhofer and R. C. Schulz Makromol. Chem. 1969 121 147. J. C. Thompson and E. Hamori J. Phys. Chem. 1971,75,272. M. Gratzel A. Henglein M. Scheffler H. M. Bossler and R. C. Schulz Ber. Bunsenge- sellschaft phys. Chem. 1972 76 72. M. H. Litt and J. W. Summers J. Polymer Sci. Polymer Chem. Edn. 1973 11 1339 1379. Post Reactions of' Polymers gives good orbital overlap and allows complex formation with organic electron- acceptor molecules in such a way that they are neatly accommodated between the donor branches. 2 :1 Charge-transfer complexes of the polymers with dichlorodicyanobenzoquinone tetracyanoquinodimethane tetracyanoethylene and 2,4,5,7-tetranitrofluorenone have 3-50 times higher equilibrium constants than the corresponding charge-transfer complexes formed from p-methylthio- anisole or 10-methylphenothiazine and the absorption spectrum maxima of the polymer complexes exhibit some red shifts.Polymer (1) is crystalline; all but one of the polymer (1)and (2)acceptor complexes are amorphous. They are up to 200 times more conducting than the complexes made from the non-polymeric donors. The concentration of unpaired electrons (e.s.r.) are essentially indepen- dent of temperature most of the electrons being trapped. Elongation of one of the amorphous polymer complexes decreases resistivity suggesting that conductivity is parallel to the polymer backbone.The complexes are photoconducting. The charge-transfer-band extinction coefficient of the complex of tetracyano- quinodimethane with polyesters of trans-2,3-dicarboxyspirocyclopropane-1,9'-fluorene and ethylene glycol has been found to increase over a range of 1-4 times with increasing degree of polymerization reflecting most probably an increase in the equilibrium constant of complex formation. Thus conductivity optical absorption maximum and equilibrium constant for complex formation are all affected by the degree of polymerization of at least the donor species. In most polymerization reactions the polymer has little direct effect on the propagation reaction other than that ascribable to increasing viscosity ; it is however possible that polymers derived from highly polar monomers may interact as templates with polar monomers.If the polymer is sufficiently stereo- regular an increase of the propagation rate and a stereoregular product could ensue. Thus poly(methy1 methacrylate) is known to promote polymerization of its monomer. Comparison of the influence of conventional isotactic and syndio- tactic poly(methy1 methacrylate) and of a 1 1 stereo-complex of the latter two upon the bulk polymerization of methyl methacrylate has provided detailed evidence of such effects particularly at low conversion ;l compared with the conventional polymer isotactic poly(methy1 methacrylate) at 90 "C promotes the formation of the syndiotactic polymer with which it forms an acetone-insoluble stereo-complex ; no polymerization occurs under the same conditions in the absence of the polymer.Crown ethers have remarkable complexing power particularly for small metal cations ;crown-ether-substituted polystyrenes are even more efficient complexing agents. The reduced viscosity pattern of complexes based on the polymers resembles that ofpolyelectrolytes the reduced viscosity increasing as the complex concentration falls. l4 '* R. C. Schulz and H. Tanaka Pure Appl. Chem. 1972,30,249. l3 R. Buter Y. Y. Tan and G. Challa J. Polymer Sci. Polymer Chem. Edn. 1973 11 989. l4 S. Kopolov Z. Machacek U. Takaki and J. Smid J. Macromol. Sci. Chem. 1973,7 1015. 326 L. P.Ellinger 3 Reduction and Hydrogenation Lithium aluminium hydride has proved a powerful reducing agent replacing aliphatic chlorine essentially quantitatively by hydrogen.Lithium deuteride is similarly effective at introducing deuterium in the place of chlorine. l5 Various heterogeneous hydrogenation catalysts add hydrogen efficiently to aliphatic double bonds ; the reaction products of lithium trialkyls with nickel salts of organic acids are examples. l6 Co-ordination compounds of lithium have also been used as hydrogenation ~ata1ysts.l~ The complete removal of heterogeneous catalysts from polymers is often difficult but essential if the product is to be stable. An effective homogeneous non-catalytic hydrogenation technique is based on the use oftoluenesulphonyl hydrazide in DMF.Thermally discoloured poly(viny1 chloride) has been decolorized and substantially stabilized against further discoloration by treatment with this reagent ;I8 cis-polypentenamer could be hydrogenated more readily than the trans-isomer and without significant degradation of the polymer backbone;lg at a low degree of hydrogenation of a high trans-material an amorphous product was obtained ;complete hydrogena- tion yielded a product of 85-86 % crystallinity and a crystalline melting point of 130 "C. Polycyclohexa- 1,3-diene has also been hydrogenated. 4 Halogenation The surface of polyethylene film has been fluorinated by exposure to fluorine. Fluorination is thought to eliminate the weak boundary layer by cross-linking or at least increasing the molecular weight in the surface.20 Polyethylene crystal mats or monolayer single crystals became insoluble in xylene when fluorinated to a composition approaching C2F ;despite the essentially complete replacement of hydrogen by fluorine the appearance of the single crystal remained essentially unchanged only the a and b dimensions in the unit cell being somewhat expanded.Thus the crystal structure differs from the hexagonal symmetry of polytetra-fluorethylene.2 Poly(viny1 fluoride) film is much less readily fluorinated than polyethylene only ca. 62% of the hydrogen being replaced by fluorine,22 sug- gesting hindered-type substitution. There is a suggestion that fluorination at a C-H bond loosens the other C-H bond at the same carbon and then at an adjacent carbon but that complete fluorine substitution at the latter retards progress of the reaction.The fluorination of polyolefins is highly exothermic D. Braun and F. Weiss Angew. makromol. Chem. 1970 13 55 67. l6 Shell Oil Co. U.S.P. 3 752 767/1973. J. C. Falk Makromol. Chem. 1972 160 291. T. Nakagawa and M. Okawara J. Polymer Sci. Part A-I Polymer Chem. 1968 6 1795. l9 K. Sanui W. J. McKnight and R. W. Lenz J. Polymer Sci. Polymer Letters Edn. 1973 11 427. H. Schonhorn and R. H. Hansen J. Appl. Polymer Sci.,1968,12 123 1. " H. Schonhorn P. G. Gallagher J. P. Luongo and F. J. Padden jun. Macromolecules 1970 3 800. '' H. Shinohara M. Iwasaki S. Tsujimura K. Watanabe and S. Okazaki J. Polymer Sci. Part A-I Polymer Chem. 1972 10 2129.Post Reactions of Polymers 327 and fluorination of single crystals specifically at chain folds would require very carefully controlled conditions. However a slurry of single crystals has been ~hlorinated~~ or brominated at the chain folds under U.V. irradiation in dry-ice- acetone-cooled Freon 11 solution. The heat of fusion of the chlorinated material as prepared stays constant independently of halogen content but after melt crystallization it drops to a degree dependent on the extent of halogenation. Halogenation appears to result in a block structure consistent with halogenation occurring only at chain folds. With ‘nascent’ polyethylene halogenation also appears to occur preferentially in the non-crystalline region. N.m.r. and i.r.spectroscopic examinations of slurry-phase chlorinated (24.k 45.2 ‘x)high-density polyethylene indicate a hindered-type substitution unaffec- ted by the molecular weight degree of chlorination or residual crystallinity ;24 by contrast the chlorine distribution in chlorosulphonated polyethylene is random.25 Since the stereochemistry of the double bonds in polydienes and poly- alkenamers can be controlled to a high degree it should also be possible to direct the stereochemistry of halogen addition to these. Chlorine addition to the double bonds of high trans-and high cis-polyalkenamers has yielded partially crystalline poly- 1,2-dichloroalkenamers. Chlorine addition appears to proceed over long sequences of double bonds.26 5 Mechanical and Thermal Degradation Degradation can be brought about me~hanically,~’ and has been used for grafting.In high molecular weight polyacrylonitrile application of stress has been found to produce shifts proportionally in the i.r. absorption peaks at 1352 1247 and 1071 cm-’.28 Thermal degradation sets an upper limit to the temperature at which various polymers can be used and there has been much study of the mechanism and prevention of thermal degradation. It has also been used preparatively and for structural study. The preparation of carbon fibres from polymer fibres may involve separate pyrolytic and oxidative stages. The pyrolysis of polyacrylonitrile and of its copolymers with minor proportions of acrylates acrolein methyl vinyl ketone styrene and vinyl acetate has been studied systematically by differential thermal analysis.Polyacrylonitrile pyrolyses to a heterocyclic ladder polymer. The first three of the above comonomers can participate structurally and kinetically in the formation of the ladder structure whereas the latter two comonomers inter- rupt it.29 In air ladder-structure formation prevails up to 350°C the cyclized 23 I. R. Harrison and E. Baer J. Polymer Sci. Part B Polymer Letters 1971,9 843. 24 I. A. Abu-Isa and M. E. Myers jun. J. Polymer Sci. Polymer Chem. Edn. 1973 11 225. 25 E. G. Brarne jun. J.. Polymer Sci. Part A-I Polymer Chem. 1971 9 205. 16 G. Dall’Asta P.Meneghini I. W.Bassi and U. Gennaro Makromoi. Chem. 1973,165 83. 21 A. Casale L. R. Whitlock R. S. Porter and J.F. Julian Amer. Chem. SOC.,Div. Polymer Chem. Polymer Preprints 1971 12 496. 28 S. L. Dobretsov Vysokomol. Soedineniya (B) 1972 14 786. 29 N. Grassie and R. McGuchan European Polymer J. 1973 9 113. 328 L. P.Ellinger 1P-dihydropyridine ladder structures containing oxidizable methylene groups. As the temperature is raised further the dihydropyridine structure is either oxi- datively aromatized or oxidized to the corresponding 4-pyridone in which mole- cular motion is restrained by hydrogen b~nding.~' Porous carbon may be obtained by the complete dehydrochlorination of poly(viny1idene chloride) ;the quality of the product depends on the structure and pretreatment of the (co)polymer and on the pyrolysis conditions. The dehydrochlorination (at 150-190 "C) is subject to an induction period; up to about half the hydrogen chloride separates by an unzip reaction.31 Loss of the second hydrogen chloride with complete carbonization requires far more energetic conditions e.g.a temperature above 700 "C. The thermal degradability of PVC seriously impairs the performance of this polymer and is associated with progressive discoloration. The loss of hydrogen chloride32 by an unzipping elimination yields polyenic sequences. Initiation appears to be associated with 2-3 % of the chlorine of the PVC which alone reacts with triphenylaluminium and which is grafted cationically in the presence of certain aluminium compounds ;33 after replacement of this reactive chlorine by a stable substituent the polymer becomes appreciably more resistant to thermal degradation.Part only of the unzipping reaction appears to be pro- moted by HCl.34 From a study at 180 "C in [Me-14C]toluene and in [Me3H]- toluene (in absence of oxygen) in which both tritium and 14C were found to be incorporated in the polymer it was concluded that the initial stages of degradation involve a free-radical mechanism.35 The cause of the enhanced reactivity of part of the chlorine is obscure ;although tertiary chlorine which has been quantitatively introduced into model copolymers of vinyl chloride and 2-chloropropene does impart enhanced degradability and discoloration tertiary chlorine has not been detected spectroscopically," as it should be at a branching level of 5-6 branches per 1000 C atoms.Tertiary deuterium has not been detected by n.m.r. in PVC reduced with lithium deuteride but is easily discerned in the product obtained similarly from vinyl chloride-2-chloropropene copolymer. At least some chloromethyl branches have been detected.36 HCl elimination should be easier when the chlorine is attached to a carbon next to a tertiary rather than to a secondary CH group ;alternatively the reactive chlorine may be activated by a neighbouring double bond. The nature of the reactive chlorine is thus still unestablished. It has even been concluded from another comparison of PVC and vinyl chloride-2-chloropropenecopolymer that 0.1 mol % tertiary chlorine would suffice to account for the thermal decomposition rate of PVC.37 30 J. W. Johnson W.Potter P. G. Rose and G. Scott Brit. Polymer J. 1972 4 527. '*D. H. Davies D. H. Everett and D. J. Taylor Trans. Furuduy SOC.,1971 67 382. 32 A. Guyot and M. Bert J. Appl. Polymer Sci. 1973,17,753. 3J J. P. Kennedy and M. Ichikawa Amer. Chem. SOC.Div. Polymer Chem. Polymer Preprinfs 1973 14 677. J4 M. Carenza Yu.-V. Moiseev and G. Palma J. Appl. Polymer Sci. 1973 17 2685. 35 C. H. Bamford and D. F. Fenton Polymer 1969 10 63. 36 A. Rigo G. Palma and G. Talamini Mukromol. Chem. 1972 153 219. 3' A. R. Berens Amer. Chem. SOC., Div. Polymer Chem. Polymer Preprints 1973 14,671. Post Reactions of Polymers 329 Dehydrochlorination in PVC powder begins just above the glass transition temperature ( -70 0C).32Following some initial acceleration it proceeds with zeroth-order kinetics and an activation energy of 92 kJ mol- The changes in U.V.absorption suggest that some HCl remains dissolved in the polymer and that initially the reaction mechanism may be ionic diffusion controlled and partially reversible termination involving inter- or intra-molecular reactions.32 The kinetic patterns of decomposition by unzipping in solid polymers have been described and the derived equation is applicable to the thermal degradation of PVC.38 Techniques for the study of degradation kinetics and for the identification and even the elucidation of structural details of (co)polymers are being steadily im- proved.The importance of closely defined and reproducible conditions of decomposition has led to the refinement of resistive heater and of Curie-point pyrolysers which have then been coupled to a gas chromatograph or a mass spectrometer.Polyolefins are amongst the (co)polymers for which decomposition patterns have been determined.39 The diad distribution of copolymers has been elucidated for acrylonitrile-methyl methacrylate copolymer^.^^ Direct pyrolysis in the mass spectrometer yields characteristic patterns with poly-p-alanine~.~' 6 Photodegradation The photodegradation of polymers may but need not involve oxidation ; in some instances oxidation takes a progressively increasing share. U.V.irradiation with light of wavelength < 340 nm of PVC results in dis- coloration loss of hydrogen chloride and ~ross-linking.~~ In the absence of oxygen discoloration is preceded by bleaching.In PVC films photodegrada- tion is confined to a thin surface layer in which light-absorbing polyene sequences are formed. Initially dehydrochlorination is subject to an activation energy of 58-75 kJ mol-I and highly dependent on intensity. becoming independent of temperature and light intensity after about one hour. It is postulated that under the latter conditions HCI availability controls the reaction ;the absorbing species appears not to change in kind during this stage.43 With light of 185nm at -196 "C in uucuo an e.s.r. signal develops and has been ascribed to the breaking of a C-Cl bond.44 The initiation mechanism is obscure but may be associated with unsaturated sites or as in thermal degradation with a small proportion of C-CI sites more reactive than the rest.The development of polyenic sequences which may resemble that for thermal degradation is promoted by ferrocene '* J. D. Danforth and T. Takeuchu J. Polymer Sci. Polymer Chem. Edn. 1973 11 2083 209 1. 39 B. R. Northmore Brir. Polymer J. 1972 4 5 11. 40 Y. Yamamoto S. Tsuge and T. Takeuchu. Macromolecules 1972 5 325. 4' A. Leuderwald and H. Ringsdorf Angew. mnkrornol. Chem.. 1973 29/30 453. 42 C. A. Brighton G. C. Marks and J. L. Benton in 'Encyclopaedia of Polymer Science and Technology' ed. H. F. Mark N. G. Gaylord and N. M. Bikales Wiley-Interscience New York 1971 Vol. 14 p. 391. 4J W. H. Gibb and J. R. MacCallum European Polymer J. 1972 8 1223. 44 M. Yamamoto M. Yano and Y. Nishijima Reports Progr.Polymer Phys. Japan 1968 11 495. 330 L. P.Ellinger which is thought to react to yield a polymeric free radical and a ferrocinium cation :45 (C,H,),Fe + +CH2-CHClt + (C,H,)2Fe+ C1- + -CH2-~H+CH,-CHCl~3-,_ Carbonyl groups which are important chromophores in the context of polymer photodegradation can be introduced synthetically either into the main chain (olefinxarbon monoxide copolymers) or attached to the main chain (vinyl ketone copolymers) ; they are formed also through photo- or auto-oxidation. The main photodegradation mechanisms are (i) the Norrish I mechanism which results in free-radical products and (ii) the Norrish I1 mechanism which involves non-radical intermediates and products and is limited to carbonyl groups associated with a y-C-H ;46 sterically the carbonyl-containing segment must conform to a cyclic transition state involving a y-CH group (see Scheme 1).The quantum efficiencies at 25 "C for both mechanisms are listed in Table 1 for methyl ketones having second alkyl 0..... . .. . H // \ R~-CO-CR:-CR:-CHR 3 R~C ,CR: \ CRi-CRj Yo + RiHC-C + CR:=CRi \ R2 Scheme 1 Table 1 R' @l a2 c3 0.023 0.2 1 c 0.003 0.2 1 c7 0.002 0.21 c9 0.0003 0.22 poly(methy1 vinyl ketone) 0.04 0.025 poly(ethy1ene-carbon -0.025 monoxide) copolymer groups R' of various lengths in solution47 and for two carbonyl copolymer^.^^ It is notable that in the copolymers an appreciable quantum efficiency applies to 45 A. W. Birley and D. S. Brackman presented at a conference on 'Degradability of Polymers and Plastics' The Plastics Institute London 1973 preprint 1 /4.46 J. E. Guillet J. Dhanraj F. J. Golemba and G. H. Hartley 'Stabilization of Polymers and Stabiliser Processes' in Adv. Chem. Ser. Amer. Chem. Soc. Washington 1968 No. 85 p. 272. 47 F. J. Golemba and J. E. Guillet Macromolecules 1972 5 63. Post Reactions of Polymers 33 1 the Norrish I mechanism only where the carbonyl groups are attached to the polymer main chain. The quantum efficiencies of the Norrish I1 mechanism of carbonyl-containing polymer [poly(phenyl vinyl ketone)] and copolymer (styrene-phenyl vinyl ketone and methyl methacrylate-methyl vinyl ketone) films increase sharply above the glass transition to the value observed in solution.48 Below the glass transition temperature the quantum efficiencies for the copolymers decrease further with temperature reaching zero at about -150"C.At ambient temperature the non-propagating Norrish I1 reaction makes substantially the major contribution to the photodegradation of carbonyl- containing polymers. Using triplet quenchers triplet carbonyl lifetimes of -1 x lO-*s have been determined for solid and dissolved ethylene-carbon monoxide copolymer ;49 in this polymer in the absence of oxygen 45 % of the Norrish I1 reaction was thought to occur through the triplet state. The Norrish I reaction is subject to a greater activation energy (-21 kJ mol- for pentadecan-8-one 2 1.7 kJ mol- for ethylene-carbon monoxide copolymers) and is much more sensitive to viscosity than the Norrish I1 reaction (3.55 kJ mol- for pentadecan-8-0ne).~~*~'~ In smaller carbonyl compounds the quantum efficiency of the Norrish I reaction increases below -350 nm sharply with decreasing wavelength whereas that of the Norrish I1 reactions is almost unaffected ;this effect has not apparently been studied in polymers.The Norrish I reaction can initiate through its free-radical products subsequent auto-oxidation (Photo-oxidation p. 329). The Norrish I1 reaction is at least partially quenched by oxygen much more efficiently for methyl methacrylate- methyl vinyl ketone than for styrene-methyl vinyl ketone or ethylene carbon monoxide50n copolymer. Quenching affects mainly the triplet excited state of the carbonyl group which is thought to contribute appreciably to the Norrish I1 reaction only at low oxygen concentration^,^' the singlet excited carbonyl making the major contribution to the reaction under normal conditions.When quenching the triplet carbonyl oxygen is raised to the singlet excited state which may contribute to subsequent photo-oxidation. Owing to their sensitivity to sunlight carbonyl-containing polymers provide an approach to photodegradable plastics. It is usually not practicable to produce plastics in which photodegradability is built into the main component which has to be carefully and specifically designed for particular applications. But if the carbonyl- containing polymer can impart photodegradability to a base polymer it can then be used as effective additive.Styrene-methyl vinyl ketone or especially isopropenyl vinyl ketone copolymers appear to be more efficient in this respect than ethylene-carbon monoxide copolymer which correlates with the higher Norrish I quantum efficiency of the copolymers containing a pendant carbonyl 48 E. Dan and J. E. Guillet Macromolecules 1973 6 231. 49 M. Heskins and J. E. Guillet Macromolecules 1970 3 224. '' G. H. Hartley and J. E. Guillet Macromolecules 1968 1 (a) 165 (6) 413. 5' A. M. Trozzolo and F. H. Winslow Macromolecules 1968 1 98. 332 L. P.Ellinger the Norrish I reaction initiating auto-oxidation. It is not known to what extent triplet-carbonyl-photosensitized singlet oxygen contributes to the imparted photo-oxidation.Other types of polymers whose photodegradability in sunlight may be con- trolled are based on the reversible photochemical dimerization of cinnamic acid and its derivatives to the corresponding truxillic acid products. Examples include the poly(viny1 esters) of substituted cinnamylideneacetic acid5’ and polyamides made from diphenyl-a-truxillic or diphenyl-6-truxillic acid.53 7Oxidation In the absence of oxygen high-energy radiation causes loss of hydrogen cross- linking and some double-bond formation in polyolefins ; mercury-lamp U.V. light is sufficiently energetic for a similar effect in polystyrene. In chain-folded polyolefin domains the cross-linking appears to be mainly between neighbouring folds since the effect on solubility and stress-strain characteristics is smaller than can otherwise be readily explained.54 Fuming nitric acid oxidizes crystalline polyethylene at the chain folds with the formation of carboxylic acid groups.The molecular weights of the dicarboxylic acids produced reflect the fold lengths. This oxidation has been used for the preparation from crystalline polyethylene of chain-extendable C oo dicarboxylic acid. Selective nitric acid oxidation of crystalline polyethylene irradiated with up to 6 MJ kg-y-radiation yields dicarboxylic acids of various chain lengths. Gel-permeation chromatography indicates the absence of any cross-linked material in the product; double bonds due to the y-irradiation are also oxidized to carboxylic acid but are randomly distributed.These conclusions are further borne out by a comparison of the effect of y-irradiation upon polyethylenes of varying crystallinity. 56 Auto-oxidation and photo-oxidation are important degradative reactions limiting the useful life of plastics. Polypropene and poly-4-methylpent- 1-ene are particularly prone to oxidative degradation and much effort has been devoted to a better understanding of the degradation mechanisms and the formulation of efficient antioxidant systems. Recently the promotion of photo-oxidation has become an objective with a view to making photodegradable plastics (see p. 331). Access and absorption of light of the appropriate wavelengths is essential to photo-oxidation ; the nature of the oxidation reactions depends on oxygen availability at the site of oxidation.In polymers the solubility and diffusion rate of oxygen are usually lower in crystalline than in amorphous domains. In amorphous polyethylene the solubility is similar to that in alkanes ;it increases with branching. Oxygen is regarded as insoluble in and non-diffusing through the folded lamellar 52 H. Tanaka and Y.Sato J. Polymer Sci. Part A-I Polymer Chem. 1972 10 3279. 53 H. Takahashi M. Sakuragi M. Hasegawa and H. Takahashi J. Polymer Sci. Part A-I Polymer Chem. 1972 10 1399. 54 R. Salovey and A. Keller Bell Syst. Tech. J. 1961,40 1397 1409. 55 D. G. H. Ballard and J. V. Dawkins European Polymer J. 1973,9 21 1. 56 A. Keller G. N. Patel and H. Keller presented at the Polymer Phys. Gr. Biennial Meeting R. M. C.Shrivenham September 1973 Preprint; G. N. Patel and A. Keller J. Polymer Sci. Polymer Letters Edn. 1973 11 737. Post Reactions of Polymers 333 domains of p~lyethylene.~' In polyolefins the difference of density between amor- phous and crystalline domains decreases with increasing substituent size the structure of the crystalline domain becoming more open. Crystalline atactic polypropene has a folded lamellar structure ; the crystalline domain of poly-4- methylpent-1-ene is helical and below 50 "C slightly less dense than the amor- phous domain.58 The differences between the domains as regards oxygen solu- bility and diffusion would be expected to decrease with increasing substituent size but detailed information is not available. Like oxygen antioxidants appear to concentrate in the amorphous domains of p~lyethylene.'~ In polyolefins the size of the crystalline structures may itself be affected by oxidation.The size of spherulites in polypropene has been reported as increasing during oxidation at 190"C.This effect did not occur in the absence of oxygen nor in the presence of antioxidants.60 The scattering of light by crystallites increases with decreasing wavelength. Light of -350 nm penetrates only a few microns into polyethylene polypropene or poly(ethy1ene terephthalate) films. The extent of photo-oxidation in terms of carbonyl formation measured by attenuated total reflection i.r. spectroscopy and of surface damage seen by electron microscopy declines sharply from the surface into the interior of polypropene6' and poly(ethy1ene terephthalate) film;62 it is accompanied by changes in the surface which develops cracks under stress.The effect has been ascribed to the attenuation of the light of shorter wavelengths rather than to inadequate oxygen diffusion. Auto-~xidation~~ share the same free-radical propaga- and photo-~xidation~~ tion and termination mechanisms but differ in respect of initiation. In polyolefins transition metal-ions their ligands and anions may affect every one of these mechanisms. Some insight into the complicated effects observed is being gained grad~ally.~~.~~ Auto-oxidation by non-radical pathways has been suggested with organo- metallic compounds capable of complexing both with oxygen and with the sub- strate.This may possibly occur with natural polymers but at present there is no supporting evidence concerning synthetic polymer^.^' In polyolefins initiation of auto-oxidation comprises the following distinct steps the formation (i) of the propagating alkyl and hydroperoxyl radicals (ii) of 57 A. S. Michaels and H. J. Bixler J. Polymer Sci. 1961 50 393 413. 58 C. E. Wilkes and M. H. Lehr Macromol. Sci. (B) 1973,7,225. 59 D. A. Curson Proc. Roy. Microscop. SOC.,1972,7 96. 6o J. Balteniene and R. Baltenas Polim. Muter. Ikh Issled. Muter. Respub. Nauch Tekh. KonJ 12th 1971 14 (Chem. Abs. 1973,78 125 077). 6' P. Blais D. J. Carlson and D. M. Wiles J. Polymer Sci. Part A-I Polymer Chem. 1972 10 1077. '* P. Blais M. Day and D. M. Wiles J. Appl. Pol-vmer Sci.1973 17 1895. " L. Reich and S. S. Stivala 'Auto-oxidation of Hydrocarbons and Polyolefins' Dekker New York 1969; 'The Mechanism of Pyrolysis Oxidation and Burning of Organic Materials' ed. L. A. Wall Nat. Bur. Stand. Washington Special Publication No. 357 1972. 64 (a) K. Tsuji Adv. Polymer Sci.,1973 12 131; (b)0.Cicchetti ibid. 1970 7 70. " 0. Cicchetti and F. Gratani European Polymer J. 1972 8 561. 66 0. Cicchetti R. De Simone and F. Gratani European Polymer J. 1973 9 1205. 67 J. P. Vollman Accounts Chem. Res. 1968 1 136. 334 L. P.Ellinger hydroperoxide and possibly some peroxide and (iii) the homolysis of hydro- peroxide or peroxide to propagating free radicals. At ordinary temperatures the formation of hydroperoxide is very slow.The oxidation of 2,2,4-trimethylpentane appears to proceed by a bimolecular initiation step involving the formation of radical species RH + 0 -+ R-+ HO .68 The rate constants have also been determined and estimated for secondary and tertiary C-H groups. During the processing of antioxidant-free polyolefins carbonyl groups are formed progre~sively,~~ presumably from hydroperoxide. Photo-oxidation initiation mechanisms may include the following (1) Formation of reactive species following light absorption by a polymer- oxygen charge-transfer complex (p. 323). (2) Photolysis of pre-formed hydroperoxide or peroxide. (3) Photodegradation of pre-formed carbonyl groups by the Norrish I mechanism leading to reactive free-radical products (p.330). (4) Formation of singlet oxygen from ground-state oxygen molecules quenching pre-formed excited singlet carbonyl groups or other photosensitizer and reaction of the singlet oxygen with p~lymer.~.~' A problem is the low excitation energy (-109 kJ mol-') of singlet oxygen but various reactions observed with small unsaturated molecules (olefins 1,3-diene~)~ ' should occur with polymers even though efficiencies may be small. (5) Below a limiting concentration Ti'" compounds which are often present as polymerization catalyst residues have been shown to catalyse auto-oxidation of atactic and isotactic polypropene and of 2,4,6,8-tetramethylnonane7 becoming inhibitors at higher concentrations. The initiation is formulated66 ri = ki[RH] [O,] [Ti4+] As in smaller alkanes the propagation of auto-oxidation usually involves the two stages R'.+02+ R'O,. R'O,. + R2H + R102H + R2. the latter being rate-determining. In alkanes and alkenes the bond energy of the least stable C-H bond is closely related to the propagation rate constant7 and subject to an activation energy of 42-84 kJ mol-'. Similarly the longest wave- length of light capable of effecting photodegradation in terms of a 'damage index' corresponds often to the strength of the weakest C-H or C-C bond in the polymer molecule ;7 this may reflect on the propagation rather than on the initia- tion reaction. 68 T. G. Degtyareva L. N. Denisova and E. T. Denisov Kinetika i Kataliz 1972 13 1400. 69 D. Mellor A. Moir and G.Scott European Polymer J.1973,9,219; G. V. Hutson and G. Scott Chem. and Ind. 1972 725. 70 M. L. Kaplan and P. G. Kelleher J. Polymer Sci.,Pari B Polymer Letters 1971,9 565. " D. R. Kearns Chem. Ret.. 1971 71 395. 72 S. Korcek J. H. B. Chenier J. A. Howard and K. U. Ingold Canad.J. Chem. 1972,50 2285. 73 G. V. Stephenson B. C. Moses and W. S. Wilcox J. Polymer Sci.,1961 55 451. Post Reactions of Polymers 335 In solid polymers the first propagation stage is slowed by restricted oxygen access and by the high viscosity so that termination by recombination becomes relatively important. Detailed kinetic data are not available. Under typical photo-oxidation conditions (20 "C)the propagation reaction should be very much slower absolutely and relative to termination (being subject to a much lower activation energy) than under the higher-temperature conditions characteristic of auto-oxidation.The comparative ineffectiveness of antioxidants in preventing photo-oxidation has been explained in these terms.64b It may also be due to photo-sensitization by the antioxidant or products derived from the latter such as dienone peroxides.74 1,2-Epoxides and 1,3-peroxides between tertiary carbons are formed during the auto-oxidation of polypropene the peroxyl radicals abstracting a hydrogen either from the neighbouring carbon atom or from the weaker tertiary C-H bond. The hydroperoxides introduced into polyolefins are often thermally stable below 100 OC,but are photolysed at wavelengths <340 nm with quantum efficien- cies of 0.7-1 to yield two radicals.This degenerative chain-branching is likely to contribute more to photo-oxidation than to auto-oxidation. The decomposition of the hydroperoxides to products including free-radical species is greatly pro- moted by many transition-metal ions. Alkoxyl radicals are thus formed. The secondary alkoxyl radicals derived from small molecules are subject to dis- sociative propagation which yields an aldehyde and an alkyl radical that of tertiary alkoxyl radicals a ketone and an alkyl radical. These reactions are of interest as carbonyl-forming reactions ;they may be of limited importance only in polymers since introduction of hydroperoxide groups into polypropene by initiated auto-oxidation was found to be unaccompanied by keto-carbonyl and alcoholic hydroxy-group~,~~ presumably owing to the size and viscosity of the polymer molecules ;the matter is however controversial.Termination between two free-radical sites is favoured by the high viscosity of solid polymers assisting the cage effect. It is subject to a much lower activation energy than the propagation reaction. Transition-metal ions also interact with free-radical sites with electron transfer resulting in termination.66 Photo-oxida- tion and auto-oxidation must be regarded as complementary. As regards individual polymers the information regarding polyethylene is still rather incomplete ;owing to the lack of penetration of U.V. light some of the more important work has been on films. The main chain-breaking reaction fol- lows the formation of keto-groups and proceeds mainly by the Norrish I1 mechan- ism.A comparison of the photo-oxidation of branched and linear polyethylene has shown that whereas under daylight conditions branched polyethylene takes up oxygen much more rapidly than linear polyethylene it embrittles more slowly and '4 L. V. Samsonova V. I. Gol'denberg E. V. Bistritskaya G. A. Nikiforov and V. Ya. Shlyapintokh Mitt. Chem. Forschungsinst. Wirt. Oesterr. Oesterr. Kunststofinst 1972 26 242 (Chem. Ah. 1973 78 148 562). 75 c.R.Boss H. Jabloner and E. J. Vandenberg J. Polymer Sci. Part B Polymer Letters 1972 10 915. 336 L. P. Ellinger maintains appreciable elongation and complete solubility to much higher oxygen- uptake levels.76 While auto-oxidation occurs mainly in the amorphous region it is accompanied by some increase in lateral order ascribed to the breaking of interlamellar bonds.Chain breaking may be followed by further crystallization which in turn results in a progressive increase in fusion temperature. With polypropene the formation of hydroperoxide occurs rather easily owing to the weak tertiary C-H bonds; photolysis of this hydroperoxide leads directly to main-chain breakage through the intermediate t-alkoxyl radical (p. 335) with the formation of a penultimate carbonyl. The subsequent Norrish I1 reaction eliminates only a small segment replacing it by a double bond (Scheme 2). Main-chain breakage occurs thus at an earlier stage of the sequence than with poly- ethylene.1 0 FH2 II fCH 3 + CH,-C-CH Scheme 2 The photo-oxidation of polystyrene by light of wavelengths >280 nm is now believed to involve the absorption band of the polystyrene-oxygen charge- transfer complex which in solution and possibly also in films disintegrates upon photoabsorption with formation of singlet oxygen. To account for the yellowing during photo-oxidation it is postulated that the benzene ring is opened by reaction with singlet or triplet oxygen to yield without main-chain fracture the diene dialdehydes corresponding to hexa-2,4-diene- I ,6-dial which is known to be one of the photo-oxidation products of ben~ene.~ Photoisomerization with formation of fulvene and benzvalene groups may contribute to the yellowing.Aromatic rings absorb in the range 25@-280 nm and are raised to the singlet excited state. Following intersystem crossing to the triplet state several reactions are thought to be possible (a)fracture of the bond linking the aromatic ring to the main chain; (b)fracture of main-chain C-C or C-H bonds; (c) quenching by oxygen which is raised to the singlet excited state followed by reaction of singlet oxygen to form main-chain hydroperoxide or to abstract hydrogen from the main chain. 7h F. H. Winslow W. Matreyek E. P. Otocka and P. M. Muglia Helsinki IUPAC 1972 Preprint IV-38. Post Reactions of Polymers 8 Graft Copolymerization The grafting of a polymer backbone with branches formed by the polymerization of one or several monomers has been studied and technically exploited over many years.The grafts are usually selected to differ in their physical properties from those of the backbone and the products may have improved properties and a widened range of ~ompatibility.~~ The main problems in graft polymerization still concern the homogeneity of the product [its freedom from ungrafted backbone and branch (co)polymers] and the separation and identification of the copolymer. The value of much published work and claims for graft polymerization are often impaired by the method used for product isolation. In particular single-stage batch solvent extraction and precipitation unchecked by fractionation with determination of the homogeneity of fractions is not regarded as satisfactory in supporting quantitative claims regarding the grafting.Fractionation followed by degradation is an essential feature of many st~dies.~~,~' The attachment of the graft to the backbone can either be part of the initiation or of the termination step of graft growth; either can fail owing to transfer to monomer resulting in ungrafted polymer. Grafting can involve condensation step or addition polymerization. When the graft is initiated from the polymer backbone and proceeds by addition polymerization reactive sites must be created on the backbone by reagents which are either not themselves active initiators or which are completely consumed before the monomer is introduced. Quite efficient techniques involving free- radical cationic and anionic mechanisms have been reported recently.Radical- initiating sites have been attached to cellulose by converting it into amino- aromatic esters or ethers ;diazotization and reaction of the diazonium salt with ferric chloride created aromatic free-radical sites to which acrylic and methacrylic monomers and vinylpyrrolidone were grafted." Cellulose is activated by water to be grafted during initiation by acrylic and methacrylic esters without added initiator but some homopolymer is formed especially in the presence of carbon tetrachloride.8 Pairs of donor (styrene isoprene) and acceptor (maleic anhydride acrylates acrylonitrile) vinyl monomers often form alternating copolymers. The copoly- merization may be aided by complexing with an electrophilic inorganic or metal- organic additive (ZnC1 ,MgCI ,NiCl ,lithium halide or alkylaluminium halide) and may not (sometimes above a limiting temperature) require a free-radical initiator.According to Gaylord,82 termination involves either a unimolecular or a bimolecular step ;the bimolecular step proceeds through a carbene (involving a " V. Stannett J. Macromol. Sci. Chem. 1970 A4 1177. '13 J. P. Fischer Angew. makromol. Chem. 1973 33,35. Z. A. Rogovin J. Polymer Sci.,Part C Polymer Symposia 1972 37 221. C. I. Simionescu and S. Dumitriu J. Polymer Sci. Part C Polymer Symposia 1972,37 *' 187. M. Imoto K. Takemoto and T. Otsu Makromol. Chem. 1967 104,244. '' N.G. Gaylord J. Polymer Sci. Part C Polymer Symposia 1970 31,247. 338 L. P.Ellinger rearrangement of the propagating site) to yield either a double bond or by insertion a six-membered ring.If a polymer containing reactive C-H bonds (tertiary aliphatic C-H allylic C-H or aldehyde C-H) is added to the alternat- ing copolymerization grafting occurs presumably by termination involving insertion of the intermediate carbene into the reactive C-H group of the back- bone polymer. Grafting-site formation during initiation however cannot be excluded. Grafting on to polystyrene butadiene-acrylonitrile copolymers poly(buty1 acrylate) low-density polyethylene p~lypropene,~ or cellulose has been described. In this work batch extraction-precipitation techniques were used ;the amount of non-grafted copolymers and the characterization of the graft copolymers are not firmly established.An elegant cationic grafting technique has been developed84 in which polymers containing reactive tertiary aliphatic chlorine react in solution (ethyl chloride dichloroethane or chlorobenzene) with AlEt or AlEt,CI which are themselves not active initiators under the conditions. A cationic site is formed on the polymer P which is grafted by a cationically polymerizable monomer P-Cl + AIEt + P+ AIEt,Cl-P+ + M -+ PM+ % graft copolymer About 2-3 % ofthe chlorine contained in poly(viny1 chloride) could be replaced by short polyisobutene or polybutadiene grafts with a dramatic improvement of its thermal stability (p. 328),84 suggesting that the poor thermal stability is associated with the most reactive chlorine. Chlorinated butyl rubber and chlorinated ethylene-propene rubber have been grafted similarly with poly~tyrene.~’ If transfer to monomer to which cationic polymerization is prone can be limited graft copolymers essentially free from homopolymer should be obtainable.A few sufficiently large grafts would yield products approaching ABA block copolymers in structure and physical proper- tiesE5 A related reaction is the phenylation of PVC by triphenylaluminium with replacement of the reactive chlorine by a phenyl substituent also with marked improvement of thermal stability.33 Since anionic polymerization can often be conducted under conditions for which transfer and termination are largely eliminated ;anionic grafting appears attractive but it is subject to experimental difficulties attending the placing of initiating sites on the backbone polymer and the choice of solvent.Monoethyl malonate derivatives of cellulose have been metallated and grafted with acrylo- nitrile.86 Polyamides are quite readily metallated and grafted anionically by 83 N. G. Gaylord A. Takahashi S. Kikuchi and R. A. Guzzi J. Polymer Sci.,Part B Polymer Letters 1972 10 95. 84 N. G. Thame R. D. Lundberg and J. P. Kennedy J. Polymer Sci.,Part A-1 Polymer Chem. 1972 10 2507. J. P. Kennedy J. J. Charles and D. L. Davidsuri Amer. Chem. Soc. Diu.Polymer Chem. Polymer Preprints 1973 14 974; J. P. Kennedy and R. R. Smith ibid. p. 1069. 86 C. I. Simionescu and V. Rusan J. Polymer Sci.,Part C Polymer Symposia 1972 37 173. Post Reactions of Polymers 339 suitable monomers ; poly(ethy1ene oxide) grafts have then been attached to Nylon-6." An interesting possibility in graft copolymerization is that the backbone poly- mer may exert more influence -a kind of template effect -upon the polymeriza- tion of the graft than is provided by the initiating site.Polyvinylidene grafts attached to oriented amorphous poly(methy1 methacrylate) or crystalline polyethylene are both stated to yield heterogeneous films. The supermolecular structure of the polyvinylidene blocks in the polyethylene copolymer is stated to be crystalline and oriented whereas those in the poly(methy1 methacrylate) copolymers are described as amorphous and isotropic. The grafted copolymer is claimed to have increased strength and thermal stability compared with the backbone polymers." There is a great deal of work on radiation grafting.Isotropic low-density polyethylene film swollen with acrylonitrile is grafted upon y-irradiation in the amorphous domains ; on stretching the grafts concentrate in the interfibrillar spaces.89 Graft copolymerization of natural polymers (cellulose wool silk starch) with acrylic monomers or with cyclic monomers such as ethylene sulphide has been studied extensively. It improves crease resistance dimensional stability dye acceptance and resistance to degradation of fibres.g0 The grafting of hydro- phobic non-polar vinyl polymers with hydrophilic or polar monomers results in products which can be used in semi-permeable membranes of value in liquid separation processes and in improved adhesives and films.Better control over the number and size of grafts and over homopolymer in the product should lead to products closely related in structure and properties to block copolymers and liquid rubbers. 9 Polymers as Supports General.-Organic polymers are used increasingly as supports for instance in sequential syntheses (proteins nucleotides polyols oligosaccharides) or for groups which have catalytic photosensitizing antioxidant or pharmaceutical activity. A major advantage is the gain in ease and efficiency with which the supported species is separated. Activity specificity and selectivity are frequently greater for supported than unsupported species and with pharmaceuticals a depot effect providing more extended and uniform dosage is sought.Polymers supporting active groups may be made by polymerization of the appropriate monomers or the supported group may be introduced into pre- formed polymers. Some examples of the latter and some of the reactions in which the supported groups take part will be discussed. '' T. Yamaguchi J. Maezawa E. Sasaki and M. Kawamoto Kobunshi Kuguku 1973,30 331. '' A. I. Kurilenko L. P. Krul V. I. Gerasimov and V. N. Kalinin Dokfudy Akad. Nuuk S.S.S.R. 1973,209,648 (Chem. Abs. 1973,79 54008). 89 A. I. Kurilenko L. P. Krul V. I. Gerasimov and N. F. Bakeev. Doklady AkQd. Nuuk S.S.S.R. 1973 209 144 (Chem. Abs. 1973,79 19 275). 90 'Block and Graft Copolymerisation' ed. R. J. Ceresa Wiley 1973 Vol. 1. 340 L.P.Ellinger Polymeric Catalysts.-Various polymers including especially commercially available macroreticular styrene-divinylbenzene copolymer beads have been used as supports ; certain ion-exchange resins have proved suitable. Inorganic polymers are however often preferred since they may provide better thermal stability and control over pore size more stable bonding to active groups and greater ease of separation. Metals catalytically active in hydrogenations hydroformylations and cyclo- oligomerizations have been attached to chloromethylated ‘Amberlite’ ion- exchange re~in,’~.’~ to cross-linked chloromethylated styrene-divinylbenzene copolymer,’ 1,93-’8 or to polystyrene ring-br~minated’~ by Br,-FeBr through complexing diphenylphosphine or imidazole groups.” The extent of cross-linking” and the bead size96 of supports based on styrene-divinylbenzene copolymer may be varied.The ligand may be attached to halogen-containing polymers before the metal is introduced. The diphenylphosphino-group may be attached using lithi~rn’~ or potassium diphenylphosphine,’ or by lithiation of aromatic bromine groups followed by reaction with Ph,PCl. The metal may then be attached from its complex (palladium rhodium) halide (cobalt nickel) carbonyl or other derivative.’ 1,997100 Rhodium has also been attached by equilibration of the phosphine-substituted polymer with metal bearing a tri- phenylphosphine ligand,93-’s provided the latter is sufficiently readily displaced. Ligand and metal together e.g.RhCl(C0){ Ph2PCH,CH2Si(OEt),} or RhH(C0){ Ph,PCH,CH,Si(OEt),) ,,have been attached to halogen-substituted polymer by displacement of the halogen.” Palladium-(0) and -(II) and platinum- (11) have also been complexed to ‘Amberlite XAD-4’ supported diphenylphos- phine ligand~.~~.~~ Up to 20 % titanocene (TiCp,Cl,) groups have been attached to chloromethylated styrene-divinylbenzene copolymers by a three-step synthesis.’* Imidazole has been attached directly and following condensation with iron rneso-tetraphenylporphyrin and reaction with carbon monoxide a product regarded as a model for deoxymyoglobin was ~btained.’~ Polystyrene reacts with chromium hexacarbonyl to yield a product in which 32 % of the benzene rings carry Cr(CO) groups ;the reaction is not accompanied by degradation so that from a narrow molecular weight distribution polystyrene a product ofnarrow molecular weight distribution could be obtained.Copolymers 91 K.G. Allum R. D. Hancock S. McKenzie and R. C. Pitkethly ‘Proceedings of the 5th International Congress on Catalysis Miami Beach 1972’ p. 477. 92 H. Bruner and J. C. Bailar jun. horg. Chem. 1973 12 1465. 93 R. H. Grubbs and L. C. Kroll J. Amer. Chem. SOC. 1971,93 3062. 94 J. P. Collman L. S. Hegedus M. P. Cooke J. R. Norton G. Dolcetti and D. N. Marquardt J. Amer. Chem. SOC. 1972,94 1789. ” R. H. Grubbs L. C. Kroll and E. M. Sweet J. Macromol. Sci. Chem. 1973 A7 1047. 96 M. Capka P. Svoboda and J. Hetfejs Coll. Czech. Chem. Comm. 1973,38 1242. 97 J. P. Collman and C.A. Reed J. Amer. Chem. SOC. 1973,95,2048. ’* R. H. Grubbs C. Gibbons L. C. Kroll W. D. Bonds jun. and C. H. Brubaker jun. J. Amer. Chem. SOC. 1973 95 2373. 99 N. Higara S. Takahashi and Y. Nonaka Japan Kokai 1973,13,487 (Chem.Abs. 1973. 78 148 481). loo M. Kraus Chem. listy 1972 66 1281 (Chem. Abs. 1973 78 72 607). Post Reactions of PoIymers 341 carrying the same group were obtained also by free-radical copolymerization of styrenechromium tricarbonyl with styrene or methyl acrylate. lo' Poly-2-vinylpyridine has been used as complexing polymeric support for cobalt@) iron-@) and -@I) nickel(II),lo2 and rhodium. O3 Cations or anions containing catalytically active metals have been supported directly on ion-ex- change resin. Thus tungstate anions have been supported on the quaternary ammonium ion-exchange resin 'Amberlite-IRA 400'.O4 Polymerically supported metal catalysts of a different type are those of the lower alkali metals especially of lithium. They can be made by the reaction usually in an ether solvent of an alkyl derivative of the metal (butyl- or pentyl- lithium) with a polymer containing a sufficiently reactive hydrogen e.g.styrene-p-benzylstyrene c~polymer,'~~ or a reactive halogen ;lo6 with butyl-lithium in tetramethylethylenediamine even the aromatic rings of polystyrenelo7 or poly(pheny1ene oxide)"' are lithiated. The metallated polymer is an anionic initiator and a metallating agent. Polymers as Catalysts.-Catalytically active polymers include ion-exchange resins which are used technically in esterifications and dehydrations.Most of the transition-metal catalysts which have been supported on polymers are -without such support -homogeneous catalysts for hydrogenation hydro- formylation cyclo-oligomerization acetoxylation silylation oxidation or polymerization reactions. Many supported 'homogeneous' hydrogenation and hydroformylation catalysts retain their activity sati~factorily,~~ but others for instance Pdo and Pd" supported on cross-linked polystyrene through di- phenylphosphine ligands lose their catalytic activity for the dimerizing addition of trimethylsilanol to butadiene to yield l-trimethylsiloxy-octa-2,7-dienequite quickly. The palladium complex becomes soluble presumably owing to a competition between ligand and product for the metal.96 Triphenylphosphine also removes the metal from its support. In hydrogenations advantages over the corresponding unsupported catalysts (Ru Rh Pd or Pt) include greater activity and improved specificity. The greater activity is ascribed to the inability of the supported catalytic sites to associate as readily as in the absence of The greater activity ofsupported reduced titanocene dichloride as hydrogenation catalysts for olefins and acetylenes is also ascribed to this.99 The supported catalysts show selectivity for the smaller olefin molecule in dependence on the pore size of the cross-linked support. The effective pore size lo' C. U. Pittman jun. P. L. Grube 0.E. Ayers S. P. McManus M. D. Rausch and G. A. Moser J.Polymer Sci. Part A-I Polymer Chem. 1972 10 379. lo* H.-G. Biedermann E. Griessl and K. Wichmann Mukromol. Chem. 1973,172,49. Io3 L. D. Rollmann Inorg. Chim. Acta 1972 6 137. '04 G. G. Allan and A. N. Neogi J. Catalysis 1970 19 256. '05 M. Kikuchi and T. Kakurai Kobunshi Kagaku 1972 29 911 (Chem.Abs. 1973 78 85 012). '06 M. L. Hallensleben Angew. makromol. Chem. 1973,31 147. lo' A. T. Bullock G. C. Cameron and P. M. Smith Polymer 1973,14,525. lo* A. J. Chalk and A. S. Hay. J. Polymer Sci. Part A-I Polymer Chem. 1969 7,691. 342 L. P.Ellinger in terms of improved selectivity is at 740 pm only 5-10 % of that measured by other techniques. Polar solvents enhance this selectivity effect more than non- polar solvents through reducing effective pore size.95 Activation of supported palladium hydrogenation catalyst by alcohols has been noted.92 Polymer-supported rhodium hydroformylation catalysts have been described.' ' A polymer-supported nickel catalyst has oligomerized phenylacetylene to a mixture containing mainly 1,2,4- and 1,3,5-triphenylbenzenes." The epoxidation of maleic acid on unsupported tungstate and on tungstate supported on a quaternary ion-exchange resin has been compared ;'O4 on the supported catalyst the kinetics appear to be those for a diffusion-controlled catalyst-lined pore model.The possibility of controlling the electronic and steric environment of the co-ordinated catalyst site particularly effectively in supported catalyst both through the polymer backbone and when the metal is attached by more than one ligand through control of the distance between and the relative position of the ligands has been recognized," but has not yet been studied in depth.Certain polymers supporting an alkali metal are in suitable solvents such as aliphatic ethers or aliphatic ditertiary amines initiators of anionic polymerization. They can initiate without transfer so that the newly forming polymer is linked to the supporting polymer. 'O5 Starting from suitably metallated supports graft or block copolymers of quite highly specified structure and composition are at least potentially accessible (p. 341). Esterolytic enzyme function has been related to the imidazole group of histi- dine. Imidazole and other small molecules containing the imidazole group have the same activity but at a much lower level.The factors responsible for the much greater activity of imidazole groups in at least some polymers have been studied extensively' O9 on amongst others poly-4(5)-vinylimidazole,poly-2(N)-methyl-5-vinylimidazole poly-5(6)-vinylbenzimidazole poly-N-vinylimidazole poly-2-methyl-N-vinylimidazole polyethyleneimine backbone containing a lateral N-dodecyl groups supporting imidazole rings ' ' and amongst related polymers poly-3-vinyl- 1,2,4-triazole. Copolymers have been made with vinyl alcohol p-vinylphenol acrylic- maleic- or vinyl-sulphonic acid vinylimidazolium salts,' ''vl ' and vinylpyrroli- done113 designed to introduce into the imidazole polymer substituents of a different charge type and to study the effect of this upon the pattern of the catalysis kinetics.These polymers were made by free-radical polymerization under conditions which do not impart stereoregularity ;amongst related polymers this '09 C. G. Overberger and J. C. Salamone Accounfs Chem. Res. 1969 2 217. H. C. Kiefer W. I. Congdon I. S. Scarpa and I. M. Kletz Proc. Nut. Acud. Sci.U.S.A. 1972,69,2155. J. C. Salamone B. Snider S. C. Israel P. Taylor and D. Raia Amer. Chem. SOC.,Div. Polymer Chem. Polymer Preprints 1972 13 271. l1 ' C. G. Overberger and T. J. Pacansky Amer. Chem. SOC.,Div. Polymer Chem. Polymer Preprints 1973 14 766. K. Uehara Y. Kaji M. Tanaka and N. Murata Kobunshi Kugaku 1973,30,165 (Chem. Ah. 1973 79 5761). Post Reactions of Polymers has been achieved with poly-1-vinyluracil.' l4 The effect of stereoregularity upon catalytic efficiency and pattern appears not to have been studied yet but should be interesting.The catalytic behaviour of derivatives of imidazole is explained by its ampholy- tic character. Three species differing in basicity and connected by two equilibria are involved (see Scheme 3).l1 Their relative proportions and their contributions g1 Scheme 3 to esterolytic reactions depend on the pH and on the substrate. To study these contributions the esterolyses of phenyl acetate substrates of different charge type have been studied the most important being compounds (3)-(6).'09 The less hydrophilic C8 and CI3acyl ester analogues of these have also been used as sub- Me Me Me Me I I I I c=o c=o c=o C=O I I I I +\I$?$ \ \ ' h(CH,) I-NO2 c0,-SO3-Na+ (3) PNPA (4) NABA (5) NABS (6) ANTI strates.'I2 The polymers are insoluble in either pure ethanol or water; binary mixtures of water and (a minor proportion of) methanol ethanol or propanol are used as solvents.The effect of the fraction of the catalytically active charge species upon the rate of esterolysis has been studied in detail for both monomeric and polymeric species. For small molecules this activity increases essentially proportionally with the active imidazole fraction but for the polymeric species much more complex patterns are observed. For instance for PNPA esterolysis by poly-1-vinylimida- zole the catalytic rate constant is below that for imidazole up to a certain fraction but at higher fractions esterolysis by the polymer becomes much faster than that by imida~ole.'~~ Esterolysis rates of NABA and NABS with the polymer pass through maxima at intermediate fractions at which the rates are very much faster than with imidazole.For other imidazole-group-containingpolymers and co- polymers patterns of the relationship of esterolytic rate to fraction of any particular effective charge group type are found which bear some resemblance I14 H. Kaye and S.-H. Chang J. Macromol. Sci. Chem. 1973 AT 1127. "' C. G. Overberger and M. Morimoto J. Amer. Chem. Soc. 1971,93 3222. 344 L. P.Ellinger to either of these patterns. Esterolysis by the polymers proceeds under most conditions much more rapidly than with imidazole ; 1012-fold acceleration has been claimed in one instance.' lo Analysis of these patterns ascribes the enhancement of catalytic activity to the following effects which contribute variously according to conditions (a) Co-operative effects -nucleophilic interactions between neighbouring imidazole groups in the polymer.This effect falls sharply when the fraction of the effective species falls below a certain level. (b) Electrostatic effects -polar interactions involving charged groups in the polymer ;this becomes important at high and low pH. (c) Apolar hydrophobic interactions which are related to the solubility of the polymers in certain alkanol-water mixtures although they are insoluble in either pure solvent component.With polyvinylbenzimidazoles the esterolytic activity depends critically upon the solvent composition and rate enhancements of several thousand times over monomeric imidazole have been observed especially with esters having long alkyl groups. This is ascribed to the extent of coiling of the polymer. ''571 l6 The esterolyses proceed in two steps the acylation of the imidazole by the substrate followed by the rather slower deacylation (Scheme 4).'''9' l7 Some f-CH,-CH+ NABS ,KHz -cH- deacylation) Scheme 4 separation of the kinetics of these reactions has been achieved by studying the initial rates to much lower conversions.' Highly conjugated mainly polyheterocyclic polymers catalyse a variety of reactions. This is ascribed to the rather high concentration of delocalized electrons.* Polymer Supports in Synthesis.-Progress in the technique and its application to peptide synthesis' have been discussed. '2o Chloromethylated lightly cross- linked styrene-divinylbenzene copolymer is still the most important support ; beads are commercially available but product reproducibility could be better ;' they are unsuitable as stationary phase in columns. A new synthesis of an alternative support -cross-linked 4-hydroxy-3-nitropolystyrene-starts with the copolymerization of 4-methoxystyrene with (a little) divinylbenzene. Following C. G. Overberger M. Morimoto I. Cho and J. C. Salamone J. Amer. Chem. Soc. 1971,93,3228. 'I7 Y. Okamoto and C. G. Overberger J. Polymer Sci.,Part A-I Polymer Chem.1972,10 3387. "IJ D. R. Cooper A. M. G. Law and B. J. Tighe Brit. Polymer J. 1973,5 163. 'l9 'Progress in Peptide Research' ed. S. Lande Gordon and Breach New York-London- Paris 1972 Vol. 2. lZo R. B. Merrifield in ref. 119 paper 14. Post Reactions of Polymers demethylation of BBr the product is nitrated.I2' This support can cause undue racemization during peptide synthesis so that care over the selection of amino- acid blocking groups is mandatory. Polymeric supports have found increasing use in polysaccharide synthesis. Poly-p-(prop- l-en-3-ol-l-y1)styrene (7) has been synthesized from chloromethyl- ated styrene by a four-step synthesis and condensed with appropriately substituted CH =CHCH,OH (7) glycosyl halides. 22 In another approach p-vinylbenzoyl derivatives of suitably protected glucopyranoses have been prepared and polymerized or copolymerized with styrene.'23 6-Nitrovanillin has been attached through an ether linkage to chloromethylated styrene-divinylbenzene copolymer.After selective reduction of the aldehyde group by sodium borohydride the alcohol was condensed with a pro- tected p-nitrobenzoyl-substitutedglucopyranosyl bromide. The p-nitrobenzoyl group was removed by sodium ethoxide-ethanol-dioxan after each step. This polymeric support is photosensitive and releases the saccharide upon irradiation ( >320 nm dioxan) ;the recovered resin has typical aldehyde absorption. After hydrogenation isomaltose and glucose could be separated from the carbohydrate component.l2 Chloromethylated styrene-divinylbenzene copolymer has also been used in the stepwise synthesis of ethers.'24 The polymer (8) based on a p-(8) bromostyrene-divinylbenzene-styrene copolymer has been used as macroporous insoluble carrier in nucleotide synthesis.' Chloromethylated cross-linked polystyrene beads react with the sodium salt of 4-hydroxymethyl-2,2-dimethyl-1,3-dioxolan.Acidic hydrolysis yields a supported vicinal diol. Under acetal- fprming conditions this reacted with one of the aldehyde groups of the terephthal- dehyde and isophthaldehyde leaving the other aldehyde free for suitable reactions. 126 12' R. E. Williams J. Polymer Sci.,Part A-I Polymer Chem. 1972 10 2123. lz2 J. M. Frechet and C. Schuerch J. Amer. Chem. SOC.,1971,93,492.'*' R. D. Guthrie A. D. Jenkins and J. Stehlkek J. Chem. SOC.(0,1971 2690. lZ4 U. Zehavi and A. Patchornik J. Amer. Chem. SOC.,1973 95 5673. lZ5 R. Glaser U. Sequin and C. Tamm Helu. Chim. Acta 1973,56 654. C. C. Leznoff and J. Y.Wong Canad. J. Chem. 1973,51 3756. 346 L. P.Ellinger Polymer Supports for Enzymes and Pharmaceuticals.-The advantages of polymeric supports in these cases are similar to those described for otherwise homogeneous active catalyst. With enzymes they facilitate not only the study of their function and their use as chemical and analytical but are beginning to be used in large-scale technical applications. With pharmaceuticals the main advantages sought are long-term and depot effects.129 Complex formation for instance with polyvinylpyrrolidone has found applications but is very limited in scope.Local anaesthetics have been supported on polyamides -presumably through hydrogen bonds. '30 Grafting of deferox- amine B used for its iron-chelating ability on to polyacrolein has been effected without loss of activity.I3' The linking of enzymes and of pharmaceuticals to a polymeric support or its precursor presents problems. Reaction with the appropriate acid chloride is generally too unspecific and is accompanied by racemization and loss of activity. Acid derivatives which may react more specifically under mild conditions include esters with 1-hydroxy-5-methoxybenzotriazole, N-hydroxysuccinimide and 2,4,5-trichlorophenol. Monomeric and polymeric esters derived from acrylic methacrylic N-vinylcarbamic and isopropenylcarbamic acid have been examined in model reactions with nucleophiles.129 Generally the monomeric species which can be polymerized and copolymerized are more reactive and give better yields under mild conditions than the corresponding polymers but the latter retain their specificity. Polymeric supports should prove of particular value with pharmaceuticals providing radiation protection. 29 Polymeric Support of Other Active Molecules-With a rapidly extending scope only examples can be given. The attachment of 3,5-di-t-butylphenol groups to low-density and high-density polyethylene polypropene or polyoxymethylene is effected through the 4-diazo-oxide derivative of the phenol. This is readily access- ible and decomposes under milling conditions to yield a carbene sufficiently reactive to insert into a C-H bond on the polymer ba~kb0ne.l~~ It is claimed that in low-density polyethylene the retarding effect of the polymer-attached antioxidant groups is greater than that produced by an equivalent quantity of di-t-butylphenol alone.Polymers containing nitroxyl groups include poly-(4-methacryloylamino-and 4-methacryloyl-2,2,6,6-tetramethylpiperidine 1-oxyl),'33 prepared by oxidation '" D. Warburton K. Balasingham P.Dunnill and M. D. Lilly Biochem. Biophys. Acta 1972 284 278. G. P. Roger and J. P. Andrews J. Macromol. Sci. Chem. 1973 A7 1167. lz9 H. G. Batz G. Franzmann and H. Ringsdorf Macromol. Chem. 1973 172 27. I3O G. Bauer and E. Ullmann Arch.Pharm. 1973,306 86. ''I R.Kamirez and J. E. Andrade J. Macromol. Sci. Chem. 1973 A7 1035. '" M. L. Kaplan P.G. Kelleher G. H. Bebbington and R.L. Hartless J. Polymer Sci. Polymer Letters Edn. 1973 11 357. '33 T. Kurosaki K. W. Lee and M. Okawara J. Polymer Sci. Part A-1 Polymer Chem. 1972,10 3295. Post Reactions of Polymers with hydrogen peroxide-sodium tungstate-H,edta according to Rozantsev from the corresponding piperidine polymer. Polystyrene substituted in the 4-position by imidazoline 3-oxide 1-oxyl has been prepared in fair yield by condensation in DMF of poly(p-formylstyrene) with 2,3-bis(hydroxylamino)-2,3-dimethylbutanefollowed by oxidation of the poly-[4-(1',3'-dihydroxy-4',4,5',5'-tetramethyltetrahydroimidazol-2-yl)phenyleth-ylene] with lead dioxide in DMF.,The product gives a blue solution in DMF or THF. '34 Lithiation of polystyrene followed by reaction with 2-methyl-2-nitroso- propane yields poly-(2-t-butylnitrosophenylethylene),which is oxidized by silver oxide in toluene to the corresponding nitroxyl radical. The e.s.r. spectrum of the polymer is being studied the nitroxyl group serving as a spin label. lo' Rose Bengal has been supported on chloromethy€ated styrene-divinylbenzene copolymer and the product used to study photo-oxidations involving singlet oxygen as the reactive species.' 35 Poly-p-styryldiphenylalkylidenephosphoranes have been prepared from poly- styrene containing a nuclear diphenylphosphine substituent (p. 340). They react with carbonyl compounds as Wittig reagents.'36 Carbodi-imide groups supported on cross-linked polystyrene (R) RC,H,CH,N=C=NCHMe ,have been used in the Moffat oxidation of alcohols by DMS0.13' N-Chloro-nyIon~'~~ oxidize alcohols and sulphides. Poly-(5-t-butyl-3-vinyl-1,2-benzoquinone) has been made by polymerization of a protected monomer precursor and removal of the protecting group. The product is analogous to 3,5- di-t-butyl- 1,2-benzoquinone in oxidizing primary amines to ketones.' 39 A variety of quinonoid redox polymers have been made and investigated. 140 crPP'-Tribromocumene is the main product obtained from cumene with N-bromopolymaleimide. '41 134 Y. Miura K. Nakai and M. Kinoshita Makromol. Chem. 1973 172 233. 135 E.C. Blossey D. C. Neckers A. L. Thayer and A.P. Schaap J. Amer. Chem. SOC. 1973,95 5820. 13' S. V. McKinley and J. W. Rakhys jun. J.C.S. Chem. Comm. 1972 134. 37 N. M. Weinshenker and C. M. Shen Tetrahedron Letters 1972 3285. '* H. Schuttenberg G. Klump U. Kaczmar S. R. Turner and R. C. Schulz J. Macromol. Sci.,Chem. 1973 A7 1085. W. H. Daly and D. C. Kaufman Amer. Chem. SOC.,Div. Polymer Chem. Polymer Preprints 1973 14 1187. I4O G. Manecke H.-J. Kretzschmar and W. Hubner J. Macromol. Sci. Chem. 1973 A7 1181. 14' C.Yaroslavsky A. Patchornik and E. Katchalski Tetrahedron Letters 1970,42 3629.

 



返 回