Encapsulation of Emissive Polymers within a Fluorinated Matrix Guillermo C. Bazan,* Michelle L. Renak, and Benjamin J. Sun Department of Chemistry, University of Rochester, Rochester, New York 14627 Received September 21, 1995 Revised Manuscript Received November 27, 1995 There is a considerable need to develop and fine-tune synthetic methods to prepare block copolymers com- posed of segments of specific and distinct functions. 1 Especially challenging are protocols capable of adjoining two homopolymers containing very different function- alities in a stereoregular manner. Additional consid- erations are required when one of the partners is conjugated. Conjugated chains are insoluble and in- tractable, as a result of their rigid delocalized structure, and their preparation proceeds via a precursor material that must ultimately be transformed to the target structure. 2 Conditions for the final elimination reac- tions are typically elevated temperatures and/or the use of base or acid catalysts that may introduce oxidative impurities or defects. In the case of conjugated emissive polymers, oxidative impurities are effective exciton traps that significantly abate important useful properties such as photoluminescence or electroluminescence efficiency. 3 Procedures that circumvent harsh conversion conditions therefore have a tendency to produce higher quality material. It occurred to us that a potentially interesting hybrid material would be a block copolymer containing poly- (p-phenylenevinylene) (PPV) and trans-syndiotactic- poly(2,3-bis(trifluoromethyl)norbornadiene) (poly1). PPV is of current worldwide interest as it exhibits electroluminescence and can be implemented as an emissive material in light-emitting diodes. 4 The fluori- nated section has, after poling, an unusually high relaxed dielectric constant and demonstrates pyroelec- tric behavior. 5 Both homopolymers are accessible via ring-opening metathesis polymerization (ROMP) but require different molybdenum-based Schrock-type ini- tiators. 6 PPV is derived from the cis-specific living ROMP of 9-((tert-butyldimethylsilyl)oxy)-[2.2]-paracy- clophan-1-ene (2), which works only when very reactive initiators, such as Mo(NAr)(CHCMe 2 Ph)[OCMe(CF 3 ) 2 ] 2 (Ar ) 2,6-diisopropylphenyl), are used. 7 Conversion of poly2 to PPV requires two additional steps, deprotection to poly(9-hydroxy-[2.2]-paracyclophan-1-ene) and finally dehydration using a catalytic amount of HCl. Poly1 requires an all-trans, highly tactic, stereochemistry to maximize its pyroelectric properties. 5 For this purpose, the less reactive but trans-specific Mo(NAr)(CHCMe 2 - Ph)(OCMe 3 ) 2 works well, producing poly1 with a 92% tactic content. The alkoxide-ligand exchange studies of Feast and Gibson suggested that it should be possible to convert quantitatively any Mo(NAr)(CHR)[OCMe(CF 3 ) 2 ] 2 spe- cies, where CHR is a propagating alkylidene, to Mo- (NAr)(CHR)(OCMe 3 ) 2 by simple addition of excess Li- OCMe 3 . 8 The equilibrium in this exchange predomi- nantly favors ligation of the more electron-donating alkoxide and is driven by the electrophilic nature of the metal center. Since these polymerizations are living, the propagating species, after consumption of monomer, can be viewed simply as alkylidenes where R represents a polymer chain and which may undergo ligand ex- change reactions at molybdenum. A second monomer would then be incorporated, but with the kinetic speci- ficity of the modified metal center. We have successfully applied this strategy to the preparation of PPV 20 -block-poly1 y (y ) 50, 100, 200, etc.), and the detailed sequence of steps is shown in Scheme 1. 9 Adding 2 to Mo(NAr)(CHR)[OCMe(CF 3 ) 2 ] 2 results in living cis-poly2 (A in Scheme 1). At this stage, 3-4 equiv of LiOCMe 3 are added which completely replace their fluorinated counterparts on molybdenum to generate a new propagating species B. The alkyl- idene and imido ligands are left unperturbed by this alkoxide exchange. 1 H NMR spectroscopy lends itself well to monitor the changes at the metal and the signals of MoCH R are especially diagnostic. For example, the spectrum in the alkylidene region of the reaction mixture resulting from addition of 20 equiv of 2 to Mo- (NAr)(CHR)[OCMe(CF 3 ) 2 ] 2 is shown in part a of Figure 1. The two singlets at 11.80 and 11.89 ppm are due to the propagating benzylidene in living Mo(NAr)[OCMe- (CF 3 ) 2 ] 2 -poly2 (two signals are observed due to the head- to-head and head-to-tail isomers). A significant upfield shift of the signals occurs after addition of LiOCMe 3 but, instead of two, now four signals are detected. The frequency range of the new alkylidene signals is con- sistent with previously characterized benzylidene com- plexes of the type Mo(NAr)(OCMe 3 ) 2 (CHPh). We are unsure at the present time as to why the number of signals increases after alkoxide addition. Our current thinking is that the signals arise as a result of head- to-head and head-to-tail isomerism and syn and anti rotamers. 10 These new alkylidenes, while stable for Figure 1. 1 H NMR (C6D6) spectra in the alkylidene region for the reaction of (a) [(CF3)2MeCO]2Mo(NAr)(CHRpoly220), (b) [Me3CO]2Mo(NAr)(CHRpoly220), and (c) [Me3CO]2Mo(NAr)(CHR- poly150-block-poly220). 1085 Macromolecules 1996, 29, 1085-1087 0024-9297/96/2229-1085$12.00/0 © 1996 American Chemical Society