Architectural Control in УLivingФ Free Radical Polymerizations Preparation of Star and Graft Polymers.код для вставкиСкачать
COMMUNICATIONS stacking ofconsecutive layers is governed by a combination of probability rules and well-defined nearest neighbor relationships. the crystal shows the observed combination of diffuse and sharp scattering. [I21 a) J. D. Dunitz i n X - r u j Anuli..ii.\ und the Siriii f i m of Orguiiic Moiecuk2s, Cornell University Press. Ithaca. NY, 1979. pp 58-59: b) K DornbergerSchiff in Lr/ir,pung iiher OD-Sfrirklurm, Akademie, Berlin, 1966. 113) L. Pauling. T/IPNulure of h, Clieni~d Bond. 3rd ed., Cornell University Press, Ithaca, NY, 1961. pp. 239%40. (141 Alternatively. the total bond order in 3 is 9.75. less than the bond order in benzene (10) but greater than that in hexasubstiluted benLenes (9.54) based on the average bond lengths. [IS] The mean values and their estimated standard deviations listed i n ref.  are 1.554(21) 8, for cyclobutane (any substitution) and 1.510(14) A for R2CH-C=C. 1161 The angle distortion effect seems to be nullified in benzenes with monocyclic annelations, like I . perhaps due l o "bent bonds" in those systems, the angular effect for bicyclics appears to be more effective. For an example of :I related distortion in the norbornyne trimer, see: N . L. Frank. K . K. Baldridge, P. GantLel, J. S. Siege1 7?~/ru/ie~lron Lrfr. 1995. in press.  Computational details: The molecular structure of 3 wits determined a t a variety of theoretical methods to determine self-consistency. Reported herc are the split-valence 6-31G(D)  and triple-zeta valence TZV(D)  baaia sets. employed at the restricted Hartree-Fock ( R H F ) self-consistent field (SCF) level of theory. These basis set5 include a set of SIX d polarization functions o n all hexvy atoms. These calculations were performed with the aid of the aiialycically ana1ytic;illy deterinined gradients and aearch algorithins contained i n GAMESS . Additional calculations at the MP2;6-31G(D) and Density Functional Theory levels were performed to determine the effects ofdyiiainioal correlation. Calculations using the former method. a post-RHF method that incorporates correlation in terms of Moller-Plesset theory of order 2 (MP2) [Zl].were performed using the GAUSSIAN 92 wile ofprogi-ams [ 2 2 ] .Calculations using Density Functional Theory (DFT) Methods. which inherently incorporate effects of correlation in their development. were performed uith the aid of the numerical methods within Dmol . A double numeric;tl basis set augmented by polarization functions, comparable in size to the 6-31G(D) basis set of the traditional Harti-ee- Fock methods. was chosen for the D F T calculations.  a) P. C. Hariharan, J. A. Pople, Uicor. C/i;m. .4rru 1973. 38, 213: b) M. S. Gordon. Cheiii. Phys. Lcfr. 1980. 76. 163. [I91 a ) T H. Dunning. J. C/ic,tii.P l i n 1971, 55. 716. b) A . D. McLenn. G. S. Chandler. J. Choi~.P l i y . ~1980. 72, 5639; c) A. J. H W'ichters. J C/wiii. P/ii,.\.1970. .52. 1033. 1201 M . W. Schmidt, K. K. Baldridge. J. A. Boatr. J. H. .lensen. S. Koseki. M . S. Gordon, K. A. Nguyen. T. L. Windus. S. T, Elbert, QCPE Bid1 1990. 10. 52. (211 J. A. Pople, J. S. Binkley. R. Seeger. /nr. J. Qiuiiiriuii C/ic,iii. Syinp. 1976. 10. 1  Gaussian 92. Revision C : M. J. Frisch. G. W. Trucks. M. Head-Gordon, P M . W. Gill. M. W. Wong. J. B. Foresiixiii. B. G . Johiiaon. H. B. Schlegel. M . A. Robb, E. S. Replogle. R. Gomperts. J. L. Andres. K . Raghawichari. J. S. Binkley. C. Gonzalez. R. L. Martin, D. 1. Fox. D. J. Defrees. J Baker. J. J. P. Stewart. J. A. Pople. Gaussian Inc.. Pittsburgh. PA, 1992.  a ) B. Delley. J. C/i<mi.P/iy.\. 1990, Y3. 508. DMol is available cornmerically from BIOSYM Technologies, San Diego. CA. b) B. Del1ey.J C / i ~ i i iP. h r ~1991. Y4. 7245.  a) R. C. Haddon, K Ragha\achari. J. Am. C/ie/ii SO<..1982. 104. 3516-3518: b) J. Am. C / i w i . S o t . 1985, 107, 289-297. c) 1'.Xie. H. F. Schaefer 111. G. Lidng, J. P. Bowen. J. Am. (%on.Sor. 1994. 1 t 6 . 1442- 1449.  a ) M. Nishio. M. Hirota, 7i~rr.ulic~Iron1989. 45. 7201 7245; b) F. COLLI.J. S. Siegcl, Piiw Appl. (%em. 1995. 67.683. Architectural Control in "Living" Free Radical Polymerizations: Preparation of Star and Graft Polymers** Craig J. Hawker* The ability to accurately control macromolecular architecture is becoming an increasingly important theme in polymer science with the interest being driven by the desire to prepare materials with new and/or improved properties.['] One way to achieve these goals is to introduce branches into the polymer backbone. For example, graft and star systems have been shown to be useful as rheology control agents, compatabilizers for polymer blends, and emulsifiers.r21Traditionally, graft and star polymers have been prepared by anionic[31and cationic polymeri~ation,[~l or by group transfer technique^.'^' However these techniques suffer from rigorous synthetic conditions, an inability to form many random copolymer systems, and incompatibility with a wide range of monomer units. One goal of synthetic polymer chemistry is to devise a free radical approach to the preparation of graft and star systems which has the same degree of macromolecular control as the above techniques but does not suffer from their synthetic drawbacks. Previous attempts at using free radical methods have not been entirely successful due to a lack of macromolecular control and combination reactions. which result in network formation.[61In this report we describe a novel "living" free radical methodology which overcomes these problems and allows the synthesis of star and graft copolymers with controlled molecular weights and low polydispersities under mild conditions. This new polymerization method is based on the use of novel initiators containing a covalent adduct of styrene and 2,2,6,6tetramethylpiperidinyloxy (TEMPO) .['I Previously we have demonstrated that the molecular weight and chain ends of polystyrene can be accurately controlled by the use of 1 as an initiator.[*]These results, coupled with the pioneering work by Georges et aLL9Iand Rizzardo et aI.["] suggest that the polymerization proceeds with little or no termination and may be considered living in nature. It is this lack of termination reactions, coupled with a high degree of macromolecular control, which may allow the synthesis of well-defined star and graft polymer systems using TEMPO-based initiators. To investigate this question we examined the synthesis of a trifunctional initiator, 2, which contains three initiating styreneTEMPO groups. It was hoped that each of the three styreneTEMPO groups of 2 would independently initiate a growing polymer chain and that these individual polymer chains would grow at approximately the same rate to give well-defined star macromolecules. It was found that the benzyl ester group of the starting material 1 could be hydrolyzed with potassium hydroxide to give the alcohol 3 in excellent yields. Reaction of 3 with 1.3,5-benzenetricarbonyl chloride (4) in the presence of 4-dimethylaminopyridine proceeded smoothly to give the desired trifunctional initiator 2 in 71 '/o yield after purification (Scheme 1 ) . Under the conditions developed previously,"~" bulk polymerization of 200 equivalents of deuterated styrene with 2 at 130 "C for 72 h was found to give the polystyrene 5 in 84%) yield with no detectable amounts of crosslinked or insoluble inaterial (Scheme2). Analysis of 5 revealed a molecular [*] Dr. C. J. Hawker IBM Research Center, Alinaden Research Center 650 Harry Road. San Jose. CA 95120-6099 (USA) Telef:ix' Int. code +(408)927-3310 [**I Financial support of this work was provided by Nalional Institute of Standards ;tnd Technology through ATP contract No. 70 NANB-3H-365. COMMUNICATIONS arms of the star polystyrene, 5 was hydrolyzed with potassium hydroxide (Scheme 2). Significantly. the hydrolyzed product 6 was found to have a molecular weight M , of 7000 (PD = 1.12) which agrees closely with the theoretical value for one arm of the star polymer ( M , = 7000). These results support the formation of a well-defined three-arm star macromolecule from 2, the molecular weight of which can be accurately controlled by the initial feed ratio. Further evidence for the proposed structures wiis obtained by comparing the ' H NMR spectra of the trifunctional initiator 2 with that of the deuterated star polystyrene 5 m d its hydrolyzed product 6. As c'nn be seen in Figure 1. the signals of the protons 2 9.0 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 -6 5 I 1 Fig. 1. a ) ' H N M R spectrum (300 MHL. CDCl,) of thc trililnctlcinal initiator 2; b) 'HN M R spectrum (300 MHL. CDCI,) of the three-arm dcuteratrd polystyrene star 5 : c) ' H NMR spectrum (300 MHz. CDCI,) of the h y d i ( > l y s product 6 . KOH 6 Schcmc 2 weight M,, of' 16500 amu and a polydispersity (PD) of 1.20.["] From the feed ratio, and assuming a three-arm star polymer is formed. the theoretical molecular weight M , of 5 should be 21 000 with each arm having a number average molecular weight of 7000. This discrepancy in molecular weights is, however, fully consistcnt with the proposed structure since the hydrodynamic volume o f ;I star polymer is less than that of a comparable linear polymer. To determine the molecular weights of the individual Ha, H,, and H,, which stem from the styrene unit in 2. appear as three A B quartets at 6 = 4.30-5.10. and the protons of the symmetrically substituted core unit appear ;IS ii singlet at 6 = 8.20. On reaction with [DJstyrene the resonances for Ha, H,. and H, undergo upfield shifts which correlate with the insertion of deuterated styrene units between the TEMPO and styrenic units of 2, while the resonance for the core unit remains as a singlet at 6 = 8.16. On hydrolysis the resonance for the core unit at about 6 = 8.20 disappears, and the resonances for Ha, H,, and H, undergo shifts which are consistent with the generation of a hydroxy end group. In fact the spectrum of 6 is identical to that obtained when the alcohol 3 is used to initiate the polymerization of deuterated styrene.18] In all three spectra the resonances for the TEMPO group are observed at 6 = 0.0- 1.5 which demonstrates the stability of the TEMPO linkage to hydrolysis conditions. COMMUNICATIONS Repetition of the above polymerization/hydrolysis reactions with up to I000 equivalents of styrene was found to give similar results and illustrates the usefulness of this technique for the preparation of high molecular weight materials. For example. reaction of 2 with 700 equivalents of styrene was found to give a three-arm star polystyrene in 8 8 % yield which had a M , of 53 000 and a P D of 1.I 9. While this value is again less than the theoretical number average molecular weight of 73 000, hydrolysis to give the individual arms resulted in much closer agreement ( M , = 22000; P D =1.09) with the theoretical number average molecular weight of 24000. This high degree of macromolecular control coupled with a lack of crosslinked material supports a “living” nature for this novel polymerization process with no termination due to combination or disproportionation. The versatility and usefulness of this ”living” free radical chemistry is also demonstrated by the ability to use monomers such as p-chloromethylstyrene as well as acrylates and methacrylates. For example. previously unreported star polymers containing narrow dispersity arms which are random copolymers of styrene and butyl acrylate can be prepared from 2. This lack of termination steps also opens up the possibility of using this polymerization process to prepare graft systems. In an effort to also investigate the compatibility of the styreneTEMPO group with standard free radical procedures the synthetic strategy for preparation of the graft systems involved initial synthesis of a monomer unit incorporating the styreneTEMPO group, copolymerization with styrene under normal free radical conditions. followed by a “living” free radical polymerization with a second feed of styrene. The synthesis of the polymeric initiator is outlined in Scheme 3. Reaction of 2 with p-chloromethylstyrene in the presence of sodium hydride was found to give the desired monomeric derivative 7 in 71 YOyield after purification. Copolymerization of 7 with styrene was conducted under normal conditions using azobisisobutyronitrile (AIBN) as an initiator in refluxing tetrahydrofuran. Significantly the polymerization proceeded smoothly to give the desired copolymer 8 ( M n = 12000; P D = 1.80) in 729‘0 yield after purification. Analysis of the copolymer by ‘H and I3C N M R spectroscopy showed the expected resonances for the styreneTEMPO group, and comparison with the aromatic styrenic resonances for the backbone polymer permitted the ratio of monomer units to be determined. The experimentally deter- 3 8 Scheme 3 @ mined value of 1 : 19 agrees with the feed ratio and shows that the styrene-TEMPO group has little effect on reactivity ratios. Bulk polymerization of a mixture of the copolymer 8 and 200 equivalents of styrene at 130 ”C for 72 h resulted in polymerization of the added styrene to give the proposed graft system 9. Comparison of the gel permeation chromatographic (GPC) traces for the starting polymer 8 and the graft system 9 clearly shows an increase in molecular weight for 9 and the absence of unreacted starting polymer (Fig. 2). In this case the nature of the grafted polymer chain 10 could not be probed by hydrolysis due to the stability of the ether linkage. and cleavage of this group was therefore accomplished by treatment of 9 with an excess of trimethylsilyl iodide (Scheme4). This results in a dramatic change in the GPC trace for the isolated product, no peak for the starting graft system is observed, and the polydispersity of the sample is lowered from 2.01 to 1.26. Interestingly. the number average molecular weight for 10 was determined to be 23 000, which agrees closely with the theoretical M , for the grafted chains of 21 000. These results demonstrate that a graft sysVImL tem is produced and that the Fig. 2. a ) Gcl permeation chroTEMPO groups attached to matogram of the starting copolymer the polystyrene backbone of 8 8 : b) gel permeation chromatograph are capable of initiating the of the graft polymer 9 : c) gel permepolymerization of styrene to ation chromatogram of the cleaved graft polymer 10. give grafts of controlled molecular weight and low polydispersity. The small shoulder at lower molecular weight in the G P C trace for 10 correlates with the molecular weight of the starting polymer 8. and is due to the cleaved backbone polymer, whose signal underlies the main peak for the grafted chain. This ability to conduct one free radical polymerization and then by simply increasing the temperature and adding new monomer, to conduct a second “living” free radical polymerization is intriguing and may open up new paths to unusual macromolecular architectures. As for the star polymers unusual graft copolymers can be readily prepared by this “living” free radical technique from monomers which cannot be polymerized by better known anionic or cationic techniques. In conclusion we have demonstrated that “living” free radical polymerizations based on TEMPO derivatives allow for the accurate control of macromolecular architecture. Star and graft copolymers can be prepared from the appropriate multifunctional initiators with no observation of crosslinking or termination by combination even under melt conditions. The molecular weights of the arms, or grafts, can be controlled by varying the equivalents of monomer added while maintaining very low polydispersities. We believe that this novel polymerization process offers the architectural control previously obtainable only under synthetically more rigorous anionic or cationic conditions. - COMMUNICATIONS layer extracted nith dichloroincthane ( 2 x 50 i n L ) . The cmibined orpanic layers \\ere dried, ebaporated to dryness and purified b) llash chromatography eluting \ b i t t i 1 . 1 hexane,dichloromethane increasing to dichloromethane This gave the styrene dci-ivative 7 as ii pale yellou oil. Yield 71 '%. 'HN M R (C'DCI,). h = 0.63, 1.01 ( e : i c h b r a . 6 H . C H , ) . 1.15 l.S5(in. I 2 H . 3 x C H 2 a i i d 2 .:C'H,).i65(ABq. ./=~HI.~H.CHH).~~~(~B~.J=~ ( \ .H 7 H~. C.' /~ f , )H. 4.. X~4 (Hd / / ) . ~ o f d . J = 2 and 6 H r . 1 H. CIIH). 5 20 (d o f d . J = 2 iind 7 HI. I H. =C'IIH).5.71 ( d 0 f d . J = 2 a n d 6 HI. 1 H.= C H t I J . 6 . 6 6 ( d o f d . J = h a n d 7 111. I H = C H ) . 7 OX ( d . 2 H . A r l t ) . 7 . 2 5 ~7.52 (m. 7 H . A r l l ) . " C N M R (CDCI.): 0=17.17. 20.56. 33.87. 404X. 50.36. 72.74, 72.83. 85.41. 113.52. 12604. 12723. 127.51. 127x2. 127.XX. 136.62. I3X 14. iind 141 79. m i i s speclruni ( E I J !if :103 Copo1ymerir;ition of7 w,itli styiene A solution of thc atyrene- I'bMPO monomer 7 (450 nig. 1.15 mmolj. styrene (2.40 g. 23.0 nimol. 20 cquir). .ind AIBN (40 mg, 0.23 minol) in dry tetrahydrofuran (20 inLj \ + a s heated at reflux under argon for 24 h. The rcaction inixture wiis evaporated to dryncss. rcdissolved i n dichloromcthane (10 inL) and precipitated i n t o inethanol ( 5 0 0 m L j folloued by h e x m e (500 m L j . The copolymcr 8 w a s isolated iis :I white powder Yield 72"'o: ,M,, = 12000 and P D = 1.80: ' H N M R (CDCI,). ,i= 0.6.'. 0 9 0 1.70. 3 65. 3.95. 4.30. 188.8.131.52-7.25 ( h r m ) : '-'C N M R (CDC'I,)-n=17.25. i9.0 43 5. 125.3 (br). 127.5 ( h r j . 128.30. 144.7- 145 Y (;I number ofreson~iinces\\ere tiio biii:ill t o obserrej. Pi-cparation of graft polqatyrme 9: 4 aolution of the polymerii initiat(ir 8 (200 nig. 0.085 minol. 1.0equiv) i n styrene (1.82 g. 17.5 iniiiol. 2 O O c q i i i ~ )\\'IS heated at 130 C \vith stirring under argon for 72 h During this timc tlic viscosity of thc \elution wiis observed t o gradually increase a n d the clear rciiction mixturc ebentu;ill>wlidilied The reaction mixture H B S then dissolved i n dichl~iroinethonc( 2 5 mL) and prccipit'ited i n t o liexane ( I L ) lotloNed by re-precipitation i n t o mcth,inol(I L j . The gralc polystyrene 9 mas i ~ o l a t e das ii bvhite sdid after dr)ing. Yield 80 )M,, = 8 6 0 0 0 a n d P D = 2 . 0 1 . ' H N M R ( C D C I , j : 6 = 0 . ~ 0 ~ 1 :I)lhi-mj.6.4Ob72 ; "C N M R (CDCI,): 6 = 39.0-44.5. 125.0 ( h r ) . 127 5 ( h r j . I13 5 146.0. Recei\cd. Jiinuary 13. 1995 Revised version: March 17. I995 [Z 7634 IE] Gerinan version: A i f p i i . ( % c i i f . lY95. 107. 1623 1627 Keywords: graft polymers . polymers . radical polymerization star polymers Scheinc -I ~ . \ ~ ~ ~ ~ l ~ ;P l r ~ o~ c~~~dt"7, t ~ i l 2 . To :I w l u t i o i i o f t h e alcohol 3 (1.0 g. 3.6 mmolj and 4 (290 mg, 1 1 mmol) i n dry teti-aliy~lrolui-,iii(20 mL) u'as added dropwise a solution of 4-dimethylaininopyridine ( 5 0 nig. 0 4 ininol) kind pyridine (300 mg, 3.8 mmolj in tetruhydrofuraii (2.0 nil.) The rctiction mixture was stirred at room temperature under argon for 16 h ; i d then cbaporated t o dryness. The residue was partitioned between dich1or~iineth:inc(.'(I mL) and water ( 5 0 m L j . and the aqueous layer extracted u i t h dic1ilor~imctli;inc( 2 x 50 m L ) . T h e coinbiiied organic layers were dried, evaporaled t o dryness. .ind purified by flash chromatography eluting with 1 . 2 hexane: dich1oroiiieth;inc increasing t o dichloromethane. This gave the trifiinctional initiator 2 iis :I p;ik ycllou oil. Yield 71 04: ' H N M R (CDCI,): J = 0.79. 1.03. 1.25. 1.62 (cach hi-,. 12H. C'H ,I. 1.34 1.58 (m. 6 H . CH,). 4.60 (ABq. J = 6 Hr. 1 H. C H H ) . 4.91 (AHq, .I = 0 H/. IH . C H t l ) . 5.14 (t. J = 3 Hz. 1 H. C H ) . 7.25 7.52 (m.5H. .Art<),X 20 (\. 3 11. Ai-Hj. " C N M R (CDCI,): 4 = 17 17. 20.38. 34.12.40 47. 60.15. 67 30. X i XY. 127 74. 128.18. 128.85. 130.92. 131.55. 13453. 140.33. kind 164.54: miis\ spectruin ( 1 . 1 ) iu;: 9x7: anal calcd. for C,,,H,,N,O,,: C 72 9. H 8.26. N 4.25: liiund: ( 7 3 2. I 1 7.99. N 4 43. Preparation 01' three-arm polystyrene 5 from 2 . A solution of the trifunctionul TEMPO initiatoi. 2 (150 mg. 0.152 inmolj i n [DJstyrene (3.40 g, 30.4 minol. 200 e q i i i b ) h i i s hcated a t 130 C uith stirring under argon for 72 h During this time the \ i \ c o \ i t ~ of i l i e \elution \wis observed to gradu;illy iiicreii~eand the cleai- reiict i o i i mixture e\cntually solidified. The reaction mixture was rhcn dissolved in dichl~~r~~inctli.iiic (25 mL) a n d precipitated into hexane (1 L) Vollowed by re-prccipitation into i)ie~Iimol( 1 L ) . The three-arm polystyrene 5 was isolated a b a white xilid alter dr)in:. Yield 84 M,, =16500 and P D = 1 . 2 0 : ' H NMR (CDCI,): d = O40(hi-dJ.O.~lO 1 . 7 0 ( b 2.65nnd2.90(hrd.H,j.4.3S(n~.2H.H,~andH,j. 0.40 7.15 ( h r i n i . X 40 ( s , core H's): "C N M R (CDCI,): 0 =17.10. 20.90, 39.443.X. 125.0 (bri 127.14 ( h r ) . 12X.22. 129.42. 132.73, 144.8-145.X ( a number of i-eum:iiice~U C J C 1 0 0 small to observe). 7 : TO ;I \ ( l l t i t i c i i i < ) t t l l e ;ilcohol 3 (1.0- 3.6 mmol) i n dry tetrahydrofur;in ( 2 0 mL) wii\ addcd \odium Iiydride (200 mg. .0 inniol) and the mixture stirred at rooin teiiiperaturc f o i - 10 min. A \olution otp-vinylhentyl chloride (1.52 g. 10.0 minol, 3.0CCILII\ j \+;I\ aclded a n d the inisture stirred at room tempcratiire for 1 h. then heated .it irellii\ ror I6 ti. The reaction mixture was then evapor'ited to drynesa iind p:irtitioncd hetuscii dichlorometh;iiir (50 mLJand water (50 mL) and the aqueous . [I]J M J. k'r&chet. Sfirrrcc, 1994. 263. 1710. [?I P. F. Rempp, P. J. Lutz i n C'oiiipr.ehc~i.\i~eP i ~ l j n i r rChoffi,ri'l~.h l . 7 ( E d . : S. V. Aggarwal). Pergamon. Oxford. 1988, Chapter 12.  P. J Lutr. G. Beinert, P. F. Rempp. Muuoiiiol. <'li(wf.1982. 1x3. 2787.  H . A. Nguyen. J. P. Kennedy. P o l v n . Bull. 1983. 10. 74 [ S ] 1). Y. Sogah. W. R. Hertler. 0 . W Webster. G . M. Cohen. Miu rr,niohculc,.s 1987. 20. 1473.  G. C. E'istmond. L. U: Harvey. Polrnfrr. 1985. 17. 275.  C. J. Hawker. J .An!. C'l7cw1 So(. 1994. 116. I1 185 [XI C. J. Hawker. J. L. Hedrick. Mu~riJi~olr,~,rler 1995. 28. 2W:. ['I] M. K. Georges. R . P. N . Veregin. P. M. Katmaier. C i K. Hamer. Mocroiiuil(%ilc,\ 1993. 26. 2987. Trcwl.~t'olwi. .Ti? r 7 i ? i ~ i u i d r i i f i i l i i d k i ) 1994. 2. 66. [lo] E. Rirzordo. Chcrii. .Au.sr. 1987.54. 32: D. H . Solomon. f Rizrurdo. P. Ciicioli CSIRO. US-A 4581 429. 1985. [ I I] Number average molecular heights were determined c\periinentally by gel permeation chromatography using commercially availnhlc narrow molecular polystyrene sample a s standards.