Novel, efficient, palladium-based system for the polymerization of norbornene derivatives: Scope and mechanism

Article abstract of DOI:10.1021/om010232m

[(1,5-Cyclooctadiene)(CH₃)Pd(Cl)], when reacted in situ with 1 equiv of a monodentate phosphine ligand and 1 equiv of the complex Na⁺[3,5-(CF₃)₂C₆H₃]₄ B^- was found to catalyze the vinyl addition polymerization of norbornene derivatives, including those with pendant oxygen functionalities. For norbornene, a polymerization rate of 1000 tons norbornene/mol Pd·h was observed at 25 °C. For several norbornene derivatives, the molecular weight of the polymer was found to decrease with increasing amounts of added 2-propanol. Mechanistic data confirm a vinyl insertion mechanism for these polymerizations. The polymerization rate was found to decrease dramatically for norbornene derivatives with pendant oxygen functionalities. The effect of coordinating solvents and the uptake of endo vs exo isomers for functionalized norbornenes was tested. Experiments show that (a) the endo isomer reacts more slowly than the exo isomer and (b) both isomers react much more slowly compared to norbornene derivatives lacking coordinating functionalities. Reaction of 5-norbornene carboxylic acid ethyl ester with the [(Et₃P)₂Pt(H)]⁺ fragment yields the endo-inserted product exhibiting intramolecular coordination of the ester functionality to the platinum center. The formation of chelates, both upon the coordination of the endo-functionalized norbornene and in the endo-inserted product, appears to be responsible, in part, for the observed decrease in polymerization rate for functional norbornene derivatives. A further reason for the diminution of activity of both the endo- and the exo-functionalized isomers is simply the coordination of the functionality. Of the two factors, the latter is the dominant one.

Full text of DOI:10.1021/om010232m

2802
Organometallics 2001, 20, 2802-2812
Novel, Efficien t, P a lla d iu m -Ba sed System for th e
P olym er iza tion of Nor bor n en e Der iva tives: Scop e a n d
Mech a n ism
April D. Hennis, J ennifer D. Polley, Gregory S. Long, and Ayusman Sen*
Department of Chemistry, The Pennsylvania State University,
University Park, Pennsylvania 16802
Dmitry Yandulov, J ohn Lipian, George M. Benedikt, and Larry F. Rhodes*
The BF Goodrich Company, 9921 Brecksville Road, Brecksville, Ohio 44224
J ohn Huffman
Molecular Structure Center, Indiana University, Bloomington, Indiana 47405
Received March 26, 2001
[(1,5-Cyclooctadiene)(CH3)Pd(Cl)], when reacted in situ with 1 equiv of a monodentate
phosphine ligand and 1 equiv of the complex Na+[3,5-(CF3)2C6H3]4B- was found to catalyze
the vinyl addition polymerization of norbornene derivatives, including those with pendant
oxygen functionalities. For norbornene, a polymerization rate of 1000 tons norbornene/mol
Pd‚h was observed at 25 °C. For several norbornene derivatives, the molecular weight of
the polymer was found to decrease with increasing amounts of added 2-propanol. Mechanistic
data confirm a vinyl insertion mechanism for these polymerizations. The polymerization
rate was found to decrease dramatically for norbornene derivatives with pendant oxygen
functionalities. The effect of coordinating solvents and the uptake of endo vs exo isomers for
functionalized norbornenes was tested. Experiments show that (a) the endo isomer reacts
more slowly than the exo isomer and (b) both isomers react much more slowly compared to
norbornene derivatives lacking coordinating functionalities. Reaction of 5-norbornene
carboxylic acid ethyl ester with the [(Et3P)2Pt(H)]+ fragment yields the endo-inserted product
exhibiting intramolecular coordination of the ester functionality to the platinum center. The
formation of chelates, both upon the coordination of the endo-fuctionalized nobornene and
in the endo-inserted product, appears to be responsible, in part, for the observed decrease
in polymerization rate for functional norbornene derivates. A further reason for the
diminution of activity of both the endo- and the exo-functionalized isomers is simply the
coordination of the functionality. Of the two factors, the latter is the dominant one.
In tr od u ction
entry into other kinds of polymeric materials. Transition
metal-catalyzed addition polymerization of functional
norbornenes, particularly those containing oxygen func-
tionalities, has proved to be difficult. This is especially
true for the endo-functionalized norbornenes, which
Poly(norbornene)s are of considerable interest because
of their unique physical properties, such as high glass
transition temperature, optical transparency, and low
birefringence. Norbornene is known to polymerize by
several mechanisms (Figure 1). Ring-opening meta-
thesis polymerization (ROMP) yields poly(1,3-cyclopen-
tylenevinylene), which retains one double bond in each
polymeric repeat unit.1 Cationic polymerization involves
rearrangement of the norbornene framework and gener-
ally produces moderate yields of poly(2,7-bicyclo[2.2.1]-
hept-2-ene) oligomer.2 Vinyl addition polymerization of
the monomer produces poly(2,3-bicyclo[2.2.1]hept-2-
ene), a saturated polymer.1g,2a,3
One current focus of research is the vinyl addition
polymerization of norbornene derivatives with pendant
functionalities. The resultant polymers are important
because of the novel combination of properties they
permit. Furthermore, certain functionalities allow for
postpolymerization modification, thus providing an
(1) (a) Grubbs, R. H. In Comprehensive Organometallic Chemistry;
Wilkinson G., Stone, F., Abel, E., Eds.; Pergamon Press: Oxford, U.K.,
1982; Vol. 8, p 499. (b) Schrock, R. R. Acc. Chem. Res. 1990, 23, 158.
(c) Hillmyer, M.; Lepetit, C.; McGrath, D. V.; Novak, B. M.; Grubbs,
R. H. Macromolecules 1992, 25, 3345. (d) Muhleback, A.; Bernhard,
P.; Bu¨hler, N.; Karlen, J .; Ludi, A. J . Mol. Catal. 1994, 90, 143. (e)
Kanaoka, S.; Grubbs, R. H. Macromolecules 1995, 28, 4707. (f)
Mashima, K.; Tanaka, Y.; Kaidzu, M.; Nakamura, A. Organometallics
1996, 15, 2431. (g) Okoroanyanwu, U.; Shimokawa, T.; Byers, J . D.;
Willson, C. G. J . Mol. Catal. A: Chem. 1998, 133, 93. (h) Mashima,
K.; Kaidzu, M.; Tanaka, Y.; Nakayama, Y.; Nakamura, A.; Hamilton,
J . G.; Rooney, J . J . Organometallics 1998, 17, 4183. (i) Delaude, L.;
Demonceau, A.; Noels, A. F. Macromolecules 1999, 32, 2091. (j) Lim,
N. K.; Yaccato, K. J .; Dghaym, R. D.; Arndtsen, B. A. Organometallics
1999, 18, 3953. (k) Barrett, A. G. M.; Cramp, S. M.; Roberts, R. S.
Org. Lett. 1999, 1, 1083. (l) Brumaghim, J . L.; Girolami, G. S.
Organometallics 1999, 18, 1923.
(2) (a) Gaylord, N. G.; Deshpande, A. B.; Mandal, B. M.; Martan,
M. J . Macromol. Sci. Chem. 1977, A11 (5), 1053. (b) Sagane, T.; Mizuno,
A. Makromol. Chem. 1993, 194, 37.
10.1021/om010232m CCC: $20.00 © 2001 American Chemical Society
Publication on Web 05/18/2001

Polymerization of Norbornene Derivatives
Organometallics, Vol. 20, No. 13, 2001 2803
thus far (1000 tons norbornene/mol Pd‚h at 25 °C).
Additionally, we delineate for the first time the relative
preference for the incorporation of the exo versus endo
isomer in a norbornene polymerization process. Finally,
using a model platinum-based system, we demonstrate
the endo, endo insertion of an endo-functionalized
norbornene ester derivative into a metal-hydride bond
in tandem with intramolecular coordination of the ester
moiety to the metal center. The isolation of this species
has bearing on the polymerization rate depression
observed for ester-functional norbornenes.
Exp er im en ta l Section
Gen er a l P r oced u r es a n d Ma ter ia ls. All work involving
air and moisture sensitive compounds was carried out using
standard Schlenk or drybox techniques (dinitrogen atmo-
sphere). NMR analyses of polymers were performed on a
Bru¨ker DPX 300 NMR spectrometer at ambient temperature,
using CDCl3 as solvent unless otherwise noted. Other NMR
data [1H, 31P{1H}, and 13C] were obtained on a Bru¨ker AMX
500 NMR spectrometer operating at 500.14, 202.47, and 125.77
MHz, respectively. 31P NMR was referenced to external 85%
H3PO4. Size exclusion chromatography data were obtained on
a Waters SEC System using a three-column bank (Styragel
7.8 × 300 mm columns, 100-5000 D, 500-30 000 D, 2000-
4 000 000 D), a Waters 410 differential refractometer, and a
Waters 600 HPLC pump/controller. Size exclusion chroma-
tography was performed in chloroform at ambient temperature
and calibrated to poly(styrene) standards. Thermal gravimetric
F igu r e 1. Polymerization mechanisms for norbornene.
constitute the vast majority of available functional
norbornenes.3f,h,l,q It has been presumed that this is due
to catalyst inhibition by coordination of the functionality
to the metal center. A related issue is the stereochem-
istry of the alkene insertion step in the metal-catalyzed
norbornene polymerizations. While published studies on
the insertion of norbornene into metal-carbon bonds
indicate an exo, exo insertion stereochemistry,4 the
possibility exists for an endo, endo insertion stereo-
chemistry if the metal is coordinated to the functional
group in an endo-functionalized norbornene.
Following our initial work on the addition polymeri-
zation of norbornene by [Pd(CH3CN)4](BF4)2,3b,c there
have been several reports on palladium(II)- and related
nickel(II)-catalyzed polymerizations of norbornene.1g,3d-r
Herein, we report a ligand-deficient, cationic, palla-
dium(II)-based system for the addition polymerization
of a wide variety of norbornene derivatives under mild
conditions. To our knowledge, this system displays the
highest rate of polymerization for norbornene reported
analysis data were obtained on
a Universal V1.13A TA
instrument in dinitrogen with a 10 °C/min ramp rate to 800
°C.
The exo and endo isomers of norbornenes were separated
and quantified by gas chromatography. GC data were obtained
on a Varian Model 3700 instrument fitted with a Waters DB-5
capillary column and an FID detector. GC was carried out with
a ramp rate of 10°/min, from a beginning temperature of 35
°C to a final temperature of 300 °C. Baseline separation of
endo and exo isomers of functionalized norbornenes was
evident by GC. Note that control experiments confirmed that
no isomerization occurred in the GC. The percentage of each
unreacted isomer during the course of a polymerization
reaction was determined mathematically. From these data, the
relative uptake of the two isomers in the polymerization
reaction was assessed. These reactions were relatively large
scale (25 g of monomer and 50 mL of solvent) with ap-
proximately 1% chlorobenzene used as an internal standard.
(3) (a) Tanielian, C.; Kiennemann, A.; Osparpucu, T. Can. J . Chem.
1979, 57, 2022. (b) Sen, A.; Lai, T.-W. Organometallics 1982, 1, 415.
(c) Sen, A.; Lai, T.-W.; Thomas, R. Organomet. Chem. 1988, 358, 567.
(d) Mehler, C.; Risse, W. Makromol. Chem., Rapid Commun. 1991, 12,
255. (e) Mehler, C.; Risse, W. Macromolecules 1992, 25, 4226. (f)
Breunig, S.; Risse, W. Makromol. Chem. 1992, 193, 2915. (g) Melia,
J .; Connor, E.; Rush, S.; Breunig, S.; Mehler, C.; Risse, W. Macromol.
Symp. 1995, 89, 433. (h) Reinmuth, A.; Mathew, J . P.; Melia, J .; Risse,
W. Macromol. Rapid Commun. 1996, 17, 173. (i) Goodall, B. L.;
Benedikt, G. M.; McIntosh, L. H., III; Barnes, D. A. US Patent 5 468
819, 1995. (j) Goodall, B. L.; Benedikt, G. M.; McIntosh, L. H., III;
Barnes, D. A.; Rhodes, L. F. PCT Int. Appl., WO 95 14048, 1995 (k)
Safir, A. L.; Novak, B. M. Macromolecules 1995, 28, 5396. (l) Mathew,
J . P.; Reinmuth, A.; Risse, W.; Melia, J .; Swords, N. Macromolecules
1996, 29, 2755. (m) Haselwander, T. F. A.; Heitz, W.; Kru¨gel, S. A.;
Wendorff, J . H. Macromol. Chem. Phys. 1996, 197, 3435. (n) Heinz, B.
S.; Heitz, W.; Kru¨gel, S. A.; Raubacher, F.; Wendorff, J . Acta Polym.
1997, 48, 387. (o) Abu-Surrah, A. S.; Rieger, B. J . Mol Catal. A: Chem.
1998, 128, 239. (p) Ruchatz, D.; Fink, G. Macromolecules 1998, 31,
4674. (q) Heinz, B. S.; Alt, F.; Heitz, W. Macromol. Chem., Rapid
Commun. 1998, 19, 251. (r) Heinz, B. S.; Heitz, W.; Kru¨gel, S. A.;
Raubacher, F.; Wendorff, J . H. Acta Polym. 1997, 48, 385. (s) Schultz,
R. G. Polym. Lett. 1966, 4, 541.
(4) (a) Sen, A. Acc. Chem. Res. 1993, 26, 303. (b) Markies, B. A.;
Kruis, D.; Rietveld, M. H. P.; Verkerk, K. A. N.; Boersma, J .; Kooijman,
H.; Lakin, M. T.; Speck, A. L.; van Koten, G. J . Am. Chem. Soc. 1995,
117, 5263. (c) van Asselt, R.; Gielens, E. E. C. G.; Ru¨lke, R. E.; Vrieze,
K.; Elsevier, C. J . J . Am. Chem. Soc. 1994, 116, 977. (d) Zocchi, M.;
Tieghi, G. J . Chem. Soc., Dalton Trans. 1979, 944. (e) Carr, N.; Dunne,
D. J .; Orpen, A. G.; Spencer, J . L. J . Chem. Soc., Chem. Commun. 1988,
926. (f) Delis, J . G. P.; Aubel, P. B.; Vrieze, D.; van Leeuwen, P. W. N.
M.; Veldman, N.; Spek, A. L. Organometallics 1997, 16, 4150. (g) Li,
C. S.; J ou, D. C.; Cheng, C. H. Organometallics 1993, 12, 3945. (h)
Hughes, R. P.; Powell, J . J . Organomet. Chem. 1973, 60, 427, and
references therein.
Mass spectra were recorded on a Finnegan MAT 95Q mass
spectrometer. Infrared spectra were recorded on a Perkin-
Elmer Model 1800 spectrometer using polystyrene film as a
standard. Samples were prepared on diamond cells in the
drybox.
Dichloromethane was obtained from Aldrich and dried over
CaH2, distilled and degassed, and stored over molecular sieves
in the glovebox prior to use. 2-Propanol was dried similarly
and stored under a dinitrogen atmosphere. PPh3 was obtained
from Aldrich and used without further purification. Nor-
bornene was obtained from ICN Pharmaceuticals. Bicyclo-
[2,2,1]hepta-2,5-diene (98%, inhibited), dicyclopentadiene, 5-eth-
ylidene-2-norbornene (99%), 5-vinyl-2-norbornene (95%, mixture
of endo and exo), and cis-5-norbornene-endo-2,3-dicarboxylic
anhydride were obtained from Aldrich. 5-Triethoxysilyl-2-
norbornene was obtained from Gelest. 5-Norbornenecarboxylic
acid ethyl ester (73% endo, 27% exo), 5-norbornenecarboxylic
acid tert-butyl ester (64% endo, 36% exo), and butylnorbornene
(81% endo, 19% exo) were provided by BF Goodrich. All
monomers were degassed with prepurified nitrogen prior to
use.

2804 Organometallics, Vol. 20, No. 13, 2001
Hennis et al.
Ta ble 1. Cr ysta llogr a p h ic Da ta for
Ta ble 2. Selected Bon d Dista n ces a n d An gles for 1
Bond Distances (Å)
[P t(C7H10C(O)OEt)(P Et3)2][B(3,5-(CF 3)2C6H3)4] (1)
Pt(1)-P(2)
Pt(1)-P(9)
2.355(3)
2.211(3)
2.173(24)
2.083(12)
1.26(4)
1.46(3)
1.61(4)
1.89(3)
1.44(4)
1.58(4)
1.63(4)
1.61(3)
1.521(18)
1.543(27)
1.461(15)
1.394(25)
1.420(13)
formula: C54H57BF24O2P2Pt
color of crystal: colorless
crystal dimens:
Z (molecules/cell) ) 2
volume ) 3145.26 Å3
Pt(1)-O(16)
Pt(1)-C(24)
O(16)-C(17)
C(17)-C(18)
C(18)-C(19)
C(18)-C(22)
C(19)-C(20)
C(2)-C(21)
C(20)-C(23)
C(21)-C(22)
C(22)-C(24)
C(23)-C(24)
C(26)-C(27)
O(25)-C(17)
O(25)-C(26)
calcd density ) 1.544 g/cm3
0.05 × 0.26 × 0.40 mm
space group: P1h
a ) 13.274(2)
wavelength ) 0.71069
molecular weight ) 1461.85
temperature ) -71 °C
R(F) ) 0.0623
b ) 19.020(2)
c ) 13.074(2)
â ) 104.72(1)
Rw(Fo) ) 0.0449
[(1,5-Cyclooctadiene)Pd(CH3)(Cl)] (1,5-cyclooctadiene ) 1,5-
COD) was prepared as previously reported5 and stored under
vacuum prior to use. Na+[3,5-(CF3)2C6H3]4B- was prepared
according to the literature method published by Brookhart et
al.6 or purchased from Boulder Scientific. Wilkinson’s catalyst,
RhCl[P(C6H5)3]3, was obtained from J ohnson Matthey.
Hom op olym er iza tion of Nor bor n en e. In an inert (N2)
atmosphere, 0.001 g (3.02 × 10-7 mol) of [(1,5-COD)Pd(CH3)-
(Cl)], 1 equiv of PPh3, and 1 equiv of Na+[3,5-(CF3)2C6H3]4B-
were placed into a small vial. To this was added approximately
5 mL of dichloromethane, and the mixture gently shaken to
dissolve the solids. A 100 µL portion of this stock solution was
then added to 10 g (0.106 mol) of norbornene dissolved in
approximately 5 mL of dichloromethane in a 100 mL GC vial
containing a stir bar. The solution was placed on a magnetic
stir plate and stirred rapidly for about 3 min, after which time
polymerization was complete (i.e., a solid mass of polymer
formed in the reaction vessel). The product was washed in
methanol and dried in vacuo. The resultant polymer was found
to be insoluble in all common laboratory solvents. TGA: 5%
mass loss at 403 °C, 10% at 422 °C.
Bond Angles (deg)
P(2)-Pt(1)-P(9)
P(2)-Pt(1)-O(16)
P(9)-Pt(1)-C(24)
O(16)-Pt(1)-C(24)
101.36(11)
86.8(6)
85.6(3)
87.4(7)
Ester s. In an inert (N2) atmosphere, the appropriate amount
of [(1,5-COD)Pd(CH3)(Cl)] (see Table 3), 1 equiv of PPh3, and
1 equiv of Na+[3,5-(CF3)2C6H3]4B- were placed into a 100 mL
GC vial. To this was added approximately 5 mL of dichlo-
romethane and the mixture swirled to dissolve the catalyst.
The appropriate monomer was then added to the solution. The
flask was sealed with a rubber septum, taken out of the
glovebox, and placed into a 46 °C (or 60° for the ester
monomers) oil bath. After 22 h (or 24 h for the ester
monomers), the product was obtained by precipitation in
methanol.
Hom op olym er iza tion of Dicyclop en ta d ien e. In an inert
(N2) atmosphere, 0.003 g (1.13 × 10-5 mol) of [(1,5-COD)Pd-
(CH3)(Cl)],
1
equiv of PPh3, and
1
equiv of Na+[3,5-
Poly(bicycloheptenyltriethoxysilane): Mn(SEC) 8500; Mw-
(SEC) 12 700; MWD 1.5. 1H NMR (CDCl3) (ppm): 3.72 (b, 6H,
Si(OCH2CH3)3), 1.12 (b, Si(OCH2CH3)3), 0.4-2.56 (b).
Poly(5-norbornenecarboxylic acid tert-butyl ester): Mn(SEC)
(CF3)2C6H3]4B- were placed into a 100 mL GC vial. To this
was added approximately 5 mL of dichloromethane, and the
mixture gently shaken to dissolve the catalyst. Dry, degassed
dicyclopentadiene (0.106 mol) was added to the solution, and
the mixture was stirred rapidly. The polymerization was
complete within approximately 5 min. The product was washed
with methanol and dried in vacuo. The resultant polymer was
found to be insoluble in all common laboratory solvents.
Hom op olym er iza tion s of 5-Eth ylid en e-2-n or bor n en e,
5-Vin yl-2-n or bor n en e, a n d 5-Meth ylen e-2-n or bor n en e. In
an inert (N2) atmosphere, [(1,5-COD)Pd(CH3)(Cl)], 1 equiv of
PPh3, and 1 equiv of Na+[3,5-(CF3)2C6H3]4B- were placed into
a 100 mL GC vial containing a magnetic stir bar. To this was
added approximately 5 mL of dichloromethane. Monomer was
then added to the flask in the appropriate amount (see Table
3). The flask was sealed with a rubber septum and stirred.
Polymer product was obtained by dissolving the resulting
yellow solid in chloroform, then precipitating the polymer in
acidified (3 M HCl) methanol to obtain a white solid.
1
3500; Mw(SEC) 6400; MWD 1.8. H NMR (CDCl3) (ppm): 1.5
(s, b, -C(CH3)3), 0.8-3.5 (b).
Poly(5-norbornenecarboxylic acid ethyl ester): Mn(SEC)
1
4300; Mw(SEC) 6500; MWD 1.5. H NMR (CDCl3) (ppm): 4.3
(b, COOCH2CH3), 1.3 (b, COOCH2CH3), 1.0-3.1 (b).
Hyd r ogen a tion of P oly(5-vin yl-2-n or bor n en e). In an
inert (N2) atmosphere, 0.337 g (2.81 × 10-3 mol monomer
units) of poly(5-vinyl-2-norbornene), 0.052 g (5.6 × 10-5 mol)
of RhCl(PPh3)3, and approximately 10 mL of dichloromethane
were combined in an autoclave. The autoclave was charged to
approximately 50 psi with hydrogen and stirred in a 50 °C oil
bath for 70 h. Product was obtained in quantitative yield by
precipitation in methanol. 1H NMR (CDCl3) (ppm): 0.93 (b,
CH2CH3), 1.63 (b, CH2CH3), 0.09-2.68 (m, b).
Cop olym er iza tion of 5-Eth ylid en e-2-n or bor n en e a n d
5-Tr ieth oxysilyl-2-n or bor n en e. In an inert (N2) atmosphere,
0.020 g (7.6 × 10-5 mol) of [(1,5-COD)Pd(CH3)(Cl)], 0.020 g
(7.6 × 10-5 mol) of PPh3, and 0.068 g (7.7 × 10-5 mol) of Na+-
[3,5-(CF3)2C6H3]4B- were placed in a round-bottom flask. To
this was added approximately 5 mL of dichloromethane and
the mixture gently shaken to dissolve the solids. To this was
added 8.39 × 10-3 mol of 5-ethylidene-2-norbornene and 1.81
× 10-3 mol of 5-triethoxysilyl-2-norbornene. The contents of
the flask were stirred in the glovebox for approximately 5 min
and then sealed with a rubber septum. The flask was removed
from the glovebox and stirred in a 46 °C oil bath for 24 h.
Product was obtained by precipitation in methanol and dried
in vacuo to obtain 68% yield. Resultant polymer composition:
1 silyl norbornene:3.7 ethylidene norbornene. Mn(SEC) 8600;
Poly(5-ethylidene-2-norbornene): Mn(SEC) 21 600; Mw(SEC)
48 600; MWD 2.3. 1H NMR (CDCl3) (ppm): 5.33 (b, 1H,
dCHCH3), 1.53 (b, dCHCH3), 0.72-3.09 (b).
Poly(5-vinyl-2-norbornene): Mn(SEC) 28 100; Mw(SEC)
1
40 100; MWD 1.4. H NMR (CDCl3) (ppm): 5.78 (b, 1H), 4.91
(b, 2H), 0.27-2.58 (b, 9H). TGA: 5% mass loss at 381 °C, 10%
loss at 406 °C.
Poly(5-methylene-2-norbornene): Mn(SEC) 6400; Mw(SEC)
20 500; MWD 3.2. 1H NMR (CDCl3) (ppm): 4.76 (b, 2H), 0.61-
3.09 (b, 8H).
Hom opolym er ization of 5-Tr ieth oxysilyl-2-n or bor n en e
a n d 5-Nor bor n en eca r boxylic Acid ter t-Bu tyl a n d Eth yl
(5) Rulke, R. E.; Ernsting, J . M.; Speck, A. L.; Elsevier: C. J .; van
Leeuwen, P. W. N. M.; Vrieze, K. Inorg. Chem. 1993, 32, 5769.
(6) Brookhart, M.; Grant, B.; Volpe, A. F., J r. Organometallics 1992,
11, 3920.
1
Mw(SEC) 12 300; MWD 1.4. H NMR (CDCl3) (ppm): 5.18 (b,
CHCH3), 3.75 (b, Si(OCH2CH3)3), 1.12 (b, Si(OCH2CH3)3), 1.42
(b, CHCH3), 0.39-2.91 (b).

Polymerization of Norbornene Derivatives
Organometallics, Vol. 20, No. 13, 2001 2805
Ta ble 3. Typ ica l P olym er iza tion Resu ltsa
a
Reactions were performed under N2 atmosphere in ca. 5 mL of CH2Cl2, L ) ancillary ligand (1 equiv), Na+[3,5-(CF3)2C6H3]4B- (1
b
equiv). Obtained relative to poly(styrene) standards in CHCl3.
Cop olym er iza tion of 5-Eth ylid en e-2-n or bor n en e a n d
5-Vin yl-2-n or bor n en e. In an inert (N2) atmosphere, 0.020 g
(7.6 × 10-5 mol) of [(1,5-COD)Pd(CH3)(Cl)], 0.020 g (7.6 × 10-5
mol) of PPh3, and 0.068 g (7.7 × 10-5 mol) of Na+[3,5-
(CF3)2C6H3]4B- were placed in a round-bottom flask. To this
was added approximately 5 mL of dichloromethane and the
mixture gently shaken to dissolve the solids. To this was added
8.33 × 10-3 mol of 5-vinyl-2-norbornene and 8.32 × 10-3 mol
of 5-ethylidene-2-norbornene. The contents of the flask were
stirred in the glovebox for approximately 5 min and sealed
with a rubber septum. The flask was removed from the
glovebox and stirred in a 46 °C oil bath for approximately 24

2806 Organometallics, Vol. 20, No. 13, 2001
Hennis et al.
h. Product was obtained by precipitation in methanol and dried
in vacuo to obtain 65% yield. Resultant polymer composition:
1 (5-ethylidene-2-norbornene):0.9 (5-vinyl-2-norbornene). Mn-
(SEC) 5800; Mw(SEC) 19 800; MWD 3.4. 1H NMR (CDCl3)
(ppm): 5.69 (b, CHdCH2), 5.27 (b, CHCH3), 4.89 (b, CHdCH2),
0.60-3.32 (b, 17H).
P olym er iza tion s in th e P r esen ce of 2-P r op a n ol. In the
glovebox, appropriate quantities of [(1,5-COD)Pd(CH3)(Cl)],
PPh3, and Na+[3,5-(CF3)2C6H3]4B- were placed in a 25 mL
round-bottom flask. The flask was then sealed with a rubber
septum, taken out of the glovebox, and connected to a nitrogen
bubbler. The monomer dissolved in approximately 5 mL of dry,
degassed methylene chloride was injected into the flask. The
flask was gently shaken, and then appropriate equivalents of
dry, degassed 2-propanol were injected. The dinitrogen purge
was removed and the reaction allowed to proceed.
P olym er iza tion of cis-Nor bor n en e-exo-2,3-d im eth yl-
ester . To a GC vial was added [(1,5-COD)Pd(CH3)(Cl)] (0.020
g, 7.6 × 10-5 mol) and 0.020 g (1 equiv) of triphenylphosphine.
The two were dissolved in approximately 5 mL of CH2Cl2, and
then monomer (0.150 g, 7.2 × 10-4 mol) was added to the
solution. The flask was briefly shaken, and then Na+[3,5-
(CF3)2C6H3]4B- (0.068 g, 7.7 × 10-5 mol) was added. The flask
was sealed, taken out of the glovebox, and placed into a 60 °C
oil bath for 24 h. The polymer was then precipitated with
methanol, filtered, and dried under vacuum to afford 0.027 g
of polymer product (18%).
P olym er iza tion of cis-Nor bor n en e-en d o-2,3-d im eth yl-
ester . To a GC vial was added [(1,5-COD)Pd(CH3)(Cl)] (0.020
g, 7.6 × 10-5 mol) and 0.020 g (1 equiv) of triphenylphosphine.
The two were dissolved in approximately 5 mL of CH2Cl2, and
then monomer (0.150 g, 7.2 × 10-4 mol) was added to the
solution. The flask was briefly shaken, and then Na+[3,5-
(CF3)2C6H3]4B- (0.068 g, 7.7 × 10-5 mol) was added. The flask
was sealed, taken out of the glovebox, and placed into a 60 °C
oil bath for 24 h. Addition of the dark yellow mixture to
methanol resulted in no precipitation of polymer.
Syn th esis of P tHCl(P Et 3)2. This procedure is a modifica-
tion of a published method.8 To a mixture of PtCl2(PEt3)2 (0.500
g, 1.00 × 10-3 mol) and HNEt3 (5 mL) in MeOH (12 mL) was
added NaBH4 (0.040 g, 1.5 × 10-3 mol). The mixture was
stirred for 1 h and the solvent removed in vacuo. The resulting
solid was sublimed (75 °C at 10 mmHg). The sublimed solid
was extracted with pentane. The extract was filtered through
Celite filter aid and dried in vacuo to give a white solid.
Yield: 0.43 g (93%). 1H NMR (toluene-d8, ppm): 1.63 (m, 12H),
Isom er iza tion of cis-5-Nor bor n en e-en d o-2,3-d ica r box-
ylic An h yd r id e. The isomerization is a modification of a
published procedure.7 cis-5-Norbornene-endo-2,3-dicarboxylic
anhydride (35.0 g, 0.213 mol) was weighed into a round-bottom
flask and heated to 190 °C, melting the white solid material
into a yellowish liquid. After 3 h at this temperature, the liquid
was cooled to room temperature, solidifying the material into
a yellowish solid. To this solid was then added 100 mL of
toluene, and the mixture was slurried for 1 h. A yellowish
solution with white insoluble material was formed. The slurry
was then filtered through a porous glass frit. The resulting
filtered solid was washed twice with toluene, filtered, and dried
in vacuo (35 °C/0.01 Torr). A yield of 7.5 g (21%) of 91% pure
cis-norbornene-exo-2,3-dicarboxylic anhydride was obtained,
the remainder was due to the unreacted endo isomer. 1H NMR
(CDCl3) (ppm): 6.34 (s, 2H), 3.45 (s, 2H), 3.01 (s, 2H), 1.67 (d,
1H), 1.45 (d, 1H).
1.00 (overlapping d of t, 18 H), -16.90 (t, J PH ) 15 Hz, J PtH
)
1265 Hz). 31P NMR (toluene-d8, ppm): 23.9 (s, J PtP ) 2734 Hz).
P u r ifica tion of Na +[3,5-(CF 3)2C6H3]4B-. Commercially
available Na+[3,5-(CF3)2C6H3]4B- was dissolved in methylene
chloride and filtered. The methylene chloride of the filtrate
was allowed to evaporate over the course of several days. Large
white crystals of the hydrate of Na+[3,5-(CF3)2C6H3]4B- formed,
which were covered with beige impurities. The crystals were
washed with dry methylene chloride under inert atmosphere
to remove the impurities. The crystals were dried in vacuo
overnight at 90 °C and then ground with a mortar and pestle
in the drybox. The nearly white powder was then dried further
under vacuum at 90 °C for 70 h.
Syn th esis of cis-Nor bor n en e-exo-2,3-d im eth ylester . cis-
5-Norbornene-exo-2,3-dicarboxylic anhydride (7.5 g, 0.046 mol)
was dissolved in 30 mL of methanol in a round-bottom flask
equipped with a reflux condenser. To this was added a solution
of four drops of fuming sulfuric acid in 10 mL of methanol.
The mixture was heated to reflux for 4 h. A yellowish solution
resulted after cooling to room temperature. The methanol was
then removed in vacuo on a rotary evaporator, giving a
yellowish viscous liquid. This liquid was then dissolved in
toluene and treated with 50 mL of 0.1 M NaOH solution in
water. The aqueous layer was decanted, and the organic layer
was washed three times with 100 mL of distilled water. The
resulting light yellow solution was then treated with activated
carbon, filtered, and stripped of toluene in vacuo; 1.0 g (10%
yield) of cis-norbornene-exo-2,3-dimethylester was recovered,
which was determined to be 91% pure, the remainder being
the endo isomer. 1H NMR (CDCl3) (ppm): 6.21 (s, 2H), 3.65
(s, 6H), 3.10 (s, 2H), 2.62 (s, 2H), 2.11 (b d, 1H), 1.50 (b d, 1H).
Syn th esis of [P tH(η2-C7H9C(O)OEt)(P Et3)2][B(3,5-(CF3)2-
C6H3)4]. To a fluorobenzene (0.6 mL) solution of PtHCl(PEt3)2
(0.010 g, 2.1 × 10-5 mol) and Na+[3,5-(CF3)2C6H3]4B- (0.020
g, 2.3 × 10-5 mol) was added ethyl ester of 5-norbornene
carboxylic acid (20 µL). The solution became yellow-brown. The
1
volatiles were removed in vacuo to give a yellow oil. H NMR
(CD2Cl2, ppm): -7.62 (t, J PH ) 11 Hz, J PtH ) 855 Hz,
approximately 75% of the total, endo isomer), -7.63 (t, J PtH
)
835 Hz, approximately 25% of the total, exo isomer). 31P NMR
(CD2Cl2, ppm): 11.26 (s, J PtP ) 2368 Hz, approximately 75%
of the total, endo isomer), 11.34 (s, J PtP ) 2364 Hz, ap-
proximately 25% of the total, exo isomer).
Syn th esis of cis-Nor bor n en e-en d o-2,3-d im eth ylester .
cis-5-Norbornene-endo-2,3-dicarboxylic anhydride (20.0 g, 0.122
mol) was dissolved in 100 mL of methanol in a round-bottom
flask equipped with a reflux condenser. To this was added a
solution of two drops of fuming sulfuric acid in 10 mL of
methanol, and the mixture heated to reflux for 4 h. A
yellowish-green solution resulted after cooling to room tem-
perature. The methanol was then removed in vacuo on a rotary
evaporator, giving an orange viscous liquid. This liquid was
then dissolved in toluene and treated with 50 mL of 0.1 M
NaOH. The aqueous layer was decanted, and the organic layer
extracted two times with 100 mL of distilled water. The
resulting light pink solution was then treated with activated
carbon, filtered, and stripped of toluene in vacuo; 13 g (51%
yield) of cis-norbornene-endo-2,3-dimethylester was recovered
Syn th esis of [P tH(η2-C7H10)(P Et3)2][B(3,5-(CF 3)2C6H3)4].
A reaction vessel was charged with PtHCl(PEt3)2 (0.010 g, 2
× 10-3 mol), Na+[3,5-(CF3)2C6H3]4B- (0.020 g, 2.3 × 10-5 mol),
norbornene (0.010 g, 1.1 × 10-4 mol), and fluorobenzene (0.6
mL) followed by vigorous shaking. A white precipitate and a
yellow-brown solution resulted. After about 2 h, the volatiles
were removed in vacuo. Spectroscopic characterization of the
resulting white residue is consistent with the formation of
1
[PtH(η2-C7H10)(PEt3)2][B(3,5-(CF3)2C6H3)4]. H NMR (CD2Cl2,
ppm): 7.72 (s, 8H), 7.56 (s, 4H), 4.05 (t, 2H, J PH ) 5 Hz J PtH
) 45 Hz), 2.99 (s, 2H), 2.13 (b s, 6H), 1.67 (s, 2H), 1.60 (s, 1H),
1.30 (s, 1H), 1.18 (d, 2H), 1.07 (t, 9H), -7.90 (t, J PH ) 11.5 Hz,
1
(>98% purity). H NMR (CDCl3) (ppm): 6.26 (s, 2H), 3.61 (s,
J PtH ) 853 Hz, 1H). 31P NMR (CD2Cl2, ppm): 11.0 (s, J PtP
2390 Hz).
)
6H), 3.29 (s, 2H), 3.16 (s, 2H), 1.48 (d, 1H), 1.33 (d, 1H).
(7) Craig, D. J . Am. Chem. Soc. 1951, 73, 4889.
(8) Phillips, J . R.; Trogler, W. C. Inorg. Synth. 1992, 29, 189.

Polymerization of Norbornene Derivatives
Organometallics, Vol. 20, No. 13, 2001 2807
the disorder by modeling with two alternative conformations
for the norbornyl group. These two conformations are shown
in the Supporting Information. Hydrogen atoms were placed
in fixed idealized positions, and they were included as isotropic
contributors in the final cycles of refinement. A partial
occupancy/disordered solvent is present in the cell at a center
of inversion. The solvent could not be identified and was
modeled as partial occupancy carbon atoms.
A final difference Fourier was featureless; the largest peaks,
1.12 e/A3, were located at the metal site. Crystallographic data
are presented in Table 1. Fractional coordinates, isotropic and
anisotropic parameters, and a complete set of bond distances
and angles are found in the Supporting Information. Selected
bond distances and angles are given in Table 2. A complete
set of distances and angles, atomic coordinates, and thermal
parameters are given in the Supporting Information.
F igu r e 2. 1H and 13C NMR assignments for inserted ethyl
ester of the 5-norbornene-2-carboxylic acid portion of the
platinum complex 1.
Syn th esis of [P t(C7H10C(O)OEt)(P Et3)2][B(3,5-(CF 3)2-
C6H3)4] (1). A Schlenk flask was charged with solid PtHCl-
(PEt3)2 (0.235 g, 5.02 × 10-4 mol) and Na+[3,5-(CF3)2C6H3]4B-
(0.489 g, 5.51 × 10-4 mol). To this mixture was added
norbornene carboxylic acid ethyl ester (0.835 g, 5.02 × 10-3
mol) and fluorobenzene (10 mL). The mixture was stirred at
45 °C for 48 h. A yellow solution resulted along with the
precipitation of sodium chloride. The reaction mixture was
filtered through Celite filtering aid. The remaining precipitate
was washed with additional fluorobenzene (3 × 10 mL). The
filtrate was concentrated to about 5 mL and subsequently
layered with pentane (40 mL) to eventually give a white
precipitate and a yellow oil after 2 days. The oil was decanted
along with the solvent, and the solid was dried in vacuo. The
solid was extracted with methylene chloride and filtered
through Celite filtering aid. Volatiles were removed from the
filtrate to give a yellowish oil. The oil was extracted a second
time with methylene chloride, filtered through Celite filtering
aid, and concentrated to about 1 mL volume. The solution was
layered with pentane (25 mL) and allowed to stand at room
temperature overnight to give a white solid. Yield: 0.30 g
Resu lts a n d Discu ssion
1. Ca ta lyst. The catalyst was generated in situ by
the addition of 1 equiv each of a monodentate phosphine
(L) and Na+[3,5-(CF3)2C6H3]4B- to a solution of [(1,5-
cyclooctadiene)(CH3)Pd(Cl)]. The addition of 1 equiv of
the monodentate phosphine generates the known chloro-
bridged dimer, [Pd(L)(CH3)(Cl)]2. This transformation
1
can be followed by H and 31P NMR spectroscopy. For
1
example, in the H NMR spectrum, the methyl group
on the palladium moves from a singlet at 1.12 ppm in
[(1,5-cyclooctadiene)(CH3)Pd(Cl)] to a broad singlet at
0.11 ppm upon the addition of 1 equiv of tricyclohexyl-
phosphine, indicating the formation of [Pd(PCy3)(CH3)-
(Cl)]2. This is accompanied by the appearance of unco-
ordinated 1,5-COD peaks at 5.55 and 2.34 ppm. In the
same vein, for [Pd(PPh3)(CH3)(Cl)]2, the methyl on the
palladium appears as a broad singlet at 0.63 ppm in
the 1H NMR spectrum and the phosphorus atoms
exhibit a singlet at 40.0 ppm in the 31P{1H} NMR
spectrum. Polymerizations are accomplished by the
addition of 1 equiv of Na+[3,5-(CF3)2C6H3]4B- to [Pd-
(L)(CH3)(Cl)]2. This presumably generates the highly
reactive (solvent/monomer-coordinated) cationic species
[Pd(L)(CH3)]+[3,5-(CF3)2C6H3]4B-, which is unstable in
the absence of the monomer, forming metallic palladium
with time.
2. P olym er Syn th esis an d Ch ar acter ization . Table
3 shows typical polymerization results obtained. It is
noteworthy that for norbornene, using a monomer-to-
catalyst ratio of 351,000 to 1, quantitative conversion
to polymer was observed within the time of mixing. To
our knowledge, this makes the catalyst the fastest of
those reported in the literature.3 The molecular weight
and polydispersity values were not obtained because of
its insolubility in traditional SEC solvents. However,
as shown in Table 3, the presence of a butyl substituent
results in the formation of a soluble polymer.
(40%). 31P NMR (CDCl3, ppm): 25.45 (d, J PP ) 13 Hz, J PtP
)
1648 Hz), 12.21 (d, J PP ) 13 Hz, J PtP ) 5008 Hz). IR (diamond
cell): ν(CO) 1629 cm-1. FD-MS: m/z 598 [M+] exhibited
expected isotope pattern for platinum.
Based on extensive 2-D NMR spectral characterization (see
Supporting Information), the 1H and 13C NMR assignments
can be made for the inserted ethyl ester of 5-norbornene-2-
carboxylic acid portion of the platinum complex (Figure 2).
X-r a y Str u ctu r e Deter m in a tion of [P t(C7H10C(O)OEt)-
(P Et3)2][B(3,5-(CF 3)2C6H3)4] (1). A sample of the compound
was dissolved in methylene chloride, filtered through Celite,
concentrated to about 1 mL, and layered with about 20 mL of
toluene and stored at 0 °C. After 3-4 days, some thin, well-
shaped crystals formed. Several of the platelike crystals were
examined at -170 °C and found to have very broad and weak
diffraction. It was finally discovered that the crystals under-
went a phase transition. It was possible, however, to transfer
a crystal to the goniostat at -71 °C without going through a
phase transition.
After transferring the crystal to the goniostat at -71 °C, a
systematic search of a limited hemisphere of reciprocal space
was used to determine that the crystal possessed no symmetry
or systematic absences corresponding to one of the triclinic
space groups. Subsequent solution and refinement confirmed
the centrosymmetric P1h. The data were collected using a
standard moving crystal, moving detector technique with fixed
backgrounds at each extreme of the scan. Data were corrected
for absorption and Lorentz and polarization effects, and
equivalent reflections were then averaged. The structure was
solved using direct methods (MULTAN78) and Fourier tech-
niques. During the initial refinement, the norbornyl group was
badly distorted and exhibited excessive anisotropic thermal
ellipsoids. After numerous attempts it was possible to resolve
Table 4 shows the effect of solvent on the polymeri-
zation of 5-ethylidene-2-norbornene. The common labo-
ratory solvents methylene chloride, toluene, and chlo-
robenzene provide near quantitative yield of polymer,
and there was no practical difference between them with
regard to level of effectiveness. The small deviation from
a quantitative yield was due to loss of product during
workup. In the case where no solvent was employed and
the catalyst components were added to the liquid
monomer, the yield was significantly reduced. This, and
the higher polydispersity value, can be explained by the

2808 Organometallics, Vol. 20, No. 13, 2001
Hennis et al.
Ta ble 4. Solven t Effect on th e Ad d ition
P olym er iza tion of 5-Eth ylid en e-2-n or bor n en ea
solvent
CH2Cl2 C6H5CH3 C6H5Cl
none
42
17 700 24 100
2.42 6.26
yield (%)
molecular weight (Mw)
polydispersity (Mw/Mn) 1.45
95
21 500
91
18 900
1.37
91
a
8.42 × 10-3 mol of (1,5-COD)Pd(CH3)(Cl), 1 equiv of Na+[3,5-
(CF3)2C6H3]4B-, 1 equiv of PPh3, ∼2 g of monomer, 48 °C, 19 h.
F igu r e 4. Plot of unreacted monomer vs time for the
polymerization of 5-norbornene-2-carboxylic acid ethyl
ester.
3. P olym er Mod ifica tion . We have used [RhCl-
(PPh3)3] to hydrogenate poly(5-vinyl-2-norbornene). The
1H NMR spectrum shows the clean disappearance of the
vinyl signals at 4.7-5.1 and 5.5-6.0 ppm and the
appearance of methyl and methylene signals at 0.93 and
1.63 ppm, respectively.
4. P olym er iza tion Mech a n ism . The possibility that
the polymerization proceeds through a mechanism other
than coordination polymerization by vinyl addition was
examined. Ring-opening metatheses polymerization can
be excluded due to the absence of expected olefinic
resonances in the 1H NMR spectra of the polymers.
Radical polymerization of norbornene is not known to
yield high polymers. In any case, radical inhibitors were
not removed from the monomers prior to polymerization.
An ionic, especially cationic, polymerization mechanism
has also been excluded. The addition of 1000 equiv of
water, an ion trap, to the reaction mixture did not retard
the polymerization of norbornene. The polymerization
was also not retarded by the addition 5 equiv (per Pd)
of the cationic inhibitor 2,6-di-tert-butylpyridine.
5. Rea ctivity of exo ver su s en d o Isom er in P o-
lym er iza tion . The activity as well as the molecular
weight of the polymer is drastically attenuated on
moving from norbornene to its derivatives with pendant
oxygen functionalities (Table 3). Based on the 13C NMR
shifts, one can surmise that the electronic effect of the
substituent on the CdC bond is minimal. Thus, the
vinyl carbons of the norbornene resonate at 135.5 ppm,
while those of 5-norbornene-2-carboxylic acid ethyl ester
appear at 137.7 and 132.5 ppm (endo) and 138.1 and
135.9 ppm (exo).
To ascertain whether the observed decrease in po-
lymerization efficiency was due to the slower polymer-
ization of the endo isomer of the functionalized nor-
bornenes, a series of experiments was performed using
known exo, endo mixtures of norbornene carboxylic acid
esters. Figure 4 shows the uptake profile versus time
for the polymerization of 5-norbornene-2-carboxylic acid
ethyl ester starting with a monomer isomer ratio of 22%
exo to 78% endo.
It is very clear that there is a preferential uptake of
the exo isomer in the polymerization reaction. A similar
observation was made when the polymerization of
5-norbornene-2-carboxylic acid tert-butyl ester was car-
ried out (Table 5).
F igu r e 3. Molecular weight vs added 2-propanol for poly-
(2-n-butyl-5-norbornene).
greatly increased viscosity and concomitant reduced
monomer diffusion rate even during the initial polym-
erization period.
An interesting procedure for the control of the mo-
lecular weight of poly(norbornene) derivatives involves
the addition of an alcohol. For example, a soluble, lower
molecular weight poly(norbornene) was formed when
1000 equiv of 2-propanol was added to the reaction
mixture (Table 3). Presumably, chain termination
through protic cleavage of the palladium-carbon bond
occurs in the presence of alcohols. Figure 3 summarizes
the results of a more systematic study on the monotonic
decrease in the molecular weight of poly(5-butylnor-
bornene) with increasing amounts of added 2-propanol.
Increasing the concentration of 2-propanol also gives
lower conversion to polymer.
Thermogravimetric analysis was performed for poly-
(norbornene) and poly(vinylnorbornene). Thermograms
obtained showed a single decomposition temperature
with a steep mass-loss curve. The observed decomposi-
tion temperatures for both poly(norbornene) and poly-
(vinylnorbornene) exceeded 400 °C.
Copolymerizations were also successfully performed
for 5-ethylidene-2-norbornene and 5-vinyl-2-norbornene,
as well as 5-ethylidene-2-norbornene and 5-triethoxysi-
lyl-2-norbornene. The copolymers were characterized by
1H NMR spectroscopy and size exclusion chromatogra-
phy. The 1H NMR spectrum of the ethylidene/vinyl
copolymer showed a broad olefinic signal at 4.4-6.2
ppm. The size exclusion chromatograph was unimodal,
indicating the formation of a copolymer, as opposed to
a mixture of two homopolymers. Likewise, the 1H NMR
spectrum of the 5-ethylidene-2-norbornene/5-triethoxy-
silyl-2-norbornene copolymer showed a broad vinylic
signal at 5.0-5.4 ppm and a resonance at 3.6-3.8 ppm
due to the methylene hydrogens of the silyl ether
functionality. The size exclusion chromatograph of the
copolymer was also unimodal.
To further probe the higher reactivity of the exo
isomer, the pure exo and endo isomers of 5-norbornene-
2,3-dicarboxylic acid dimethyl ester were synthesized.
The starting material, endo-5,6-norbornene dicarboxylic

Polymerization of Norbornene Derivatives
Organometallics, Vol. 20, No. 13, 2001 2809
Ta ble 5. Rea ction Tim e vs exo/en d o Ra tio in th e
Un r ea cted Mon om er , NB-C(O)OtBu a
endo face of the norbornene. For the exo isomer the
disposition of the pendant functionality is such that the
norbornene double bond and ester cannot simulta-
neously bind to the metal, on either the endo or exo face
of the norbornene.
Given the published studies that indicate that the
insertion of nonfunctionalized norbornenes into metal-
carbon bonds occurs with exo, exo stereochemistry,4 we
sought to determine whether the presence of an endo
functionality would change the insertion stereochemis-
try.
elapsed time (h)
% exo
% endo
0
36.3
32.2
29.5
27.2
63.7
67.8
70.5
72.8
1.75
5.83
18.58
a
3.78 × 10-4 mol of (1,5-COD)Pd(CH3)(Cl), 1 equiv of PPh3, 1
equiv of Na+[3,5-(CF3)2C6H3]4B-, 25 mL of CH2Cl2, 50 °C, 10 g of
monomer, 1/5 solution removed at each interval.
Because of the relative instability of the catalytically
active palladium species, detailed studies encompassing
the coordination and insertion of norbornene derivatives
were carried using the [(Et3P)2Pt(H)]+ fragment. This
species was produced in situ via chloride ion abstraction
from trans-[(Et3P)2Pt(H)(Cl)] with Na+[3,5-(CF3)2C6H3]4B-
in fluorobenzene. The formation of the cation is an
equilibrium process; the fragment is present only in a
small steady-state concentration and immediately reacts
with added norbornene, which drives the reaction to
completion.
F igu r e 5. Polymerization of the pure exo and endo isomers
of 5-norbornene-2,3-dicarboxylic acid dimethyl ester.
acid anhydride (endo-nadic anhydride), is readily isomer-
ized to the more thermodynamically stable exo isomer
by thermolysis and crystallization. In a subsequent step
both the endo- and exo-nadic anhydrides were ring-
opened in methanol in the presence of a catalytic
amount of sulfuric acid. In the case of the endo isomer
greater than 98% pure material was obtained. The exo
compound was 91% isomerically pure, the rest being the
endo compound. Figure 5 shows the results for poly-
merizations of the pure isomers using the aforemen-
tioned catalyst. The catalyst shows a complete prefer-
ence for the exo isomer over the endo for this monomer
when only the pure isomer is present.
When reacted with trans-[(Et3P)2Pt(H)(Cl) and 1.05
equiv of Na+[3,5-(CF3)2C6H3]4B-, substituted norbornenes
formed trans-[(Et3P)2Pt(H)(η2-norbornene-R)]+ instan-
taneously and quantitatively (eq 1).
An interesting question involves the extent of conver-
sion of the pure exo isomer of 5-norbornene-2,3-dicar-
boxylic acid dimethyl ester. If chelation of the double
bond and oxygen functionality to the metal center is the
most important reason for the attenuation of the po-
lymerization activity, the exo isomer should be just as
reactive as the parent, nonfunctionalized, norbornene.
Clearly, the incomplete conversion of the exo isomer
must be due to a second factor, presumably the coordi-
nation of the functionality to the metal center. To verify
the possibility of the ester group of the monomer slowing
down the polymerization through coordination, the
polymerization of the parent norbornene was carried out
in the presence of ethyl acetate (reaction conditions: 4
g norbornene, 7.56 × 10-5 mol of (1,5-cyclooctadiene)-
Pd(CH3)(Cl), 1 equiv of PPh3, 1 equiv of Na[{3,5-
(CF3)2C6H3}4B], 5 mL of ethyl acetate, 2 mL of chlo-
robenzene, 60 °C, 24 h). In very sharp contrast to the
high polymerization rate in noncoordinating solvents
(cf., Table 3), the presence of ethyl acetate led to only a
9% yield. Additionally, the polymer obtained had a
significantly lower molecular weight and was soluble
in chloroform (Mw ) 157 000).
The products were characterized by in situ NMR
spectroscopy. The 1H NMR spectra of the reaction
mixtures with R ) H and COOEt in CD2Cl2 showed
hydride signals as binomial triplets, flanked by 195Pt-
satellites, and shifted ca. 9 ppm downfield from that in
trans-[(Et3P)2Pt(H)(Cl)]. For R ) H, the olefinic protons
of the coordinated norbornene appeared as a broad
triplet, also flanked by Pt-satellites, while the 31P{1H}
NMR spectrum showed a singlet with Pt-satellites. By
virtue of coupling with 195Pt, the above data unambigu-
ously establish the structure of the adduct as that shown
in eq 1. For the general case with R * H the two
phosphines are inequivalent; however, a single peak was
observed by 31P{1H} NMR for each isomer of nor-
bornene-R mixture (exo and endo; ratios of the 31P{1H}
NMR signal intensities corresponded to the nor-
bornene-R isomer ratios). The close similarity of the
31P{1H} NMR chemical shifts of the norbornene adduct
and those of norbornene-R, as well as hydride chemical
shifts for R ) H and COOEt, clearly suggests analogous
structures for norbornene-R adducts. Evidently, the
center of asymmetry (substituted methine carbon at-
tached to the R group) is too remote from the metal
center to significantly influence the environment of the
phosphines. The 31P{1H} NMR data are collected in
Table 6, showing marginal variations of δP and J P-Pt
with R.
6. Syn th esis of th e Mod el System , tr a n s-[(Et3P )2-
P t(H)(R-NB)]+ (R-NB ) 5-n or bor n en e-2-R). One
reason that the ester-functionalized norbornenes are so
reluctant to polymerize may be due to the endo isomer,
which typically represents 70-80% of the total isomer
content. The endo isomer can easily chelate a metal
center since both the norbornene double bond and the
pendant ester functionality are readily accessible on the
7. Rea ctivity of tr a n s-[(Et3P )2P t(H)(R-NB)]+ (R-
NB ) 5-n or bor n en e-2-R). Heating the norbornene-R
adducts in fluorobenzene in the presence of 5-10-fold

2810 Organometallics, Vol. 20, No. 13, 2001
Ta ble 6. 31P {1H} NMR Da ta of
Hennis et al.
+
tr a n s-P tH(η2-RNB)(P Et3)2 in P h F a t RT
a
b
a
b
R
δendo (J Pt-P
)
δexo (J Pt-P)
H
Ph
11.02 (2386)
10.85 (2377)
10.87 (2366)
11.14 (2391)
11.04 (2385)
11.02 (2386)
11.17 (2375)
11.01 (2363)
11.42 (2389)
11.21 (2382)
COOCH2CH3
Si(OCCH2CH3)3
CH2OSi(CH3)3
a
b
In ppm. In Hz.
F igu r e 7. Structure of 1.
Successive recrystallization of the solids from the
reaction mixture gives analytically pure 1 in 40% yield
1
as a colorless, air-stable solid. The H NMR spectrum
of 1 showed signals corresponding to a total of 10
protons in addition to the resonances of six phosphine
ethyl groups, the ethyl group of the ester functionality,
and the aromatic signals of the counterion. No hydride
signal was observed. Extensive one- and two-dimen-
sional proton and carbon NMR spectra further clarified
the structure of 1. With the aid of 2D-COSY and 2D-
NOESY NMR spectra, all 1H and 13C signals were
assigned as shown in Figure 2.
F igu r e 6. Modes of bonding for ester-functionalized nor-
bornene.
Mass spectroscopy of 1 confirmed the elemental
1
composition derived from H NMR spectra. Both FD and
excess of norbornene-R for 18 h led to polymerization
for R ) H, Ph, Si(OEt)3, and CH2OSiMe3. In situ NMR
experiments gave precipitated polymers upon addition
of the NMR tube contents to acetone. 31P{1H} NMR
spectra of the reaction mixtures, recorded prior to
precipitation, showed only the unreacted trans-[(Et3P)2-
Pt(H)(η2-norbornene-R)]+. Scale-up of the polymerization
reactions with norbornene to platinum molar ratios of
1000 to 1 in fluorobenzene for 65 h confirmed polym-
erization activity. The highest yield of polynorbornene
was obtained with R ) H (85%). However, the presence
of the functionality was found to considerably decrease
the catalyst’s activity.
Heating the NB-COOEt adduct of (Et3P)2PtH+ in
fluorobenzene in the presence of 5 equiv of NB-COOEt
led to the appearance of multiple peaks in the 31P{1H}
NMR spectrum, in addition to those of unreacted trans-
[(Et3P)2Pt(H)(η2-NB-COOEt)]+. The major product (1)
formed in up to 80% nonisolated yield (depending on
the reaction conditions) and showed two mutually
coupled doublets, each flanked by Pt-satellites with
unequal J P-Pt. The magnitude of the P-P coupling, 13
Hz, is characteristic of cis-disposition of the phosphines.
The remarkably different 31P{1H} NMR chemical shifts
and platinum-phosphorus coupling constants (PA, 24.76
ppm, 1630 Hz; PB, 11.53 ppm, 4993 Hz) indicate that
ligands of considerably different trans-influence are
coordinated trans to the inequivalent phosphines. Com-
parison of the above 31P{1H} NMR parameters of 1 with
those of [(Et3P)2Pt(η3-C,O-CH2-CH2COOMe)]+ 9 (where
the acrylate moiety is bonded to both carbon and
carbonyl oxygen to form a five-membered metallocycle)
MALDI-TOF MS spectra of 1 showed a peak with the
nominal mass of 598 amu and an isotopic distribution
consistent with the proposed structure.
The IR spectrum of 1 showed a carbonyl stretch at
1629 cm-1, shifted by 105 cm-1 to a lower frequency
from that of pure NB-COOEt. Such a red shift of ν-
(CdO) strongly suggests coordination of the carbonyl
group to the metal.
In summary, the spectroscopic data indicate that the
formation of 1 results from the insertion of a single endo-
NB-COOEt moiety into the Pt-H bond of (Et3P)2PtH+.
The carbonyl oxygen of the ester substituent is coordi-
nated to the metal, forming a bidentate ligand that
occupies two cis sites around platinum. Similar 31P{1H}
NMR patterns were also obtained with pure endo-
dimethyl ester of norbornene. Pure exo-dimethyl ester
of norbornene gave a notably different and more com-
plicated mixture. Under no conditions did ester-substi-
tuted norbornenes polymerize with trans-[(Et3P)2Pt(H)-
(η2-NB-R)]+.
To clarify the stereochemistry around the R-carbon
of the inserted NB-COOEt, the solid-state structure of
1 was determined by X-ray crystallography. Crystals of
1 suitable for X-ray diffraction were grown from CH2Cl2/
toluene. The solid-state structure of 1 suffers a disorder
involving in particular the norbornane fragment. It was
possible to model the disorder with two alternative
conformations for the norbornyl group. An equal oc-
cupancy of each conformer was used in the model to
facilitate resolving the disorder present in the crystal
structure. Figure 8 shows an ORTEP view of one of the
conformations (see Supporting Information for ad-
ditional details). Selected bond distances and angles for
this conformation are found in Table 2.
(PA, 22.5 ppm, 1860 Hz; PB, 7.0 ppm, 4414 Hz; J P-P
)
13.5 Hz) showed notable similarity, indicating a very
similar ligand environment around the metal. This
suggests that the structure of 1 is that shown in Figure
7. Further spectroscopic evidence for this structural
assignment is given below.
Results from the X-ray study clearly show that the
platinum-hydride moiety has inserted into the double
bond on the endo face of the norbornene substrate. This
is highly unusual and unexpected. Typically norbornene
coordinates to metals on the exo face and then undergoes
(9) Sorato, C.; Venanzi, L. M. Inorg. Synth. 1989, 26, 134.

Polymerization of Norbornene Derivatives
Organometallics, Vol. 20, No. 13, 2001 2811
F igu r e 10. Complexes with coordination environments
similar to 1.
F igu r e 8. ORTEP representation of cis-Pt(η2-C,O-(endo,-
Figure 10 contain five-membered metallocyclic rings,
while 1 contains a six-membered metallocycle.
+
endo)-NB-COOCH2CH3)(P(CH2CH3)3)2
.
The accumulated experimental data allow some con-
clusions to be drawn regarding the formation of 1
(Scheme 1). A general outline of the reaction sequence
leading from [trans-(Et3P)2Pt(H)(Cl)], Na+[3,5-(CF3)2-
C6H3]4B-, and 5-norbornenecarboxylic acid ethyl ester
to the alkene-inserted 1 consists of the following basic
steps. Chloride removal with Na+[3,5-(CF3)2C6H3]4B-
allows for coordination of the ester-functionalized nor-
bornene through the CdC bond leading to a platinum-
norbornene adduct with trans phosphine ligands. The
insertion of the coordinated CdC bond into the Pt-H
bond requires cis-disposition of the hydride and the
coordinated alkene; hence a trans to cis rearrangement
of the primary product must occur prior to the insertion
step. The latter is presumably mediated by phosphine
dissociation: the 14-electron T-shaped transient species
B or B′ in Scheme 1. After the trans to cis rearrange-
ment, insertion occurs either prior to or following
phosphine coordination. The ester functionality is also
coordinated to platinum in 1. While no kinetic experi-
ments have been carried out to establish the postulated
inverse [PEt3] rate dependence, it was found that
increasing the steric bulk of the phosphines accelerates
the insertion process. The reactions of the analogous
trans-(Ph3P)2Pt(H)(Cl) and trans-(Cy3P)2Pt(H)(Cl) do not
require elevated temperatures; the latter species actu-
ally polymerizes NB-COOEt, albeit with low activity.
The discovery that conversion of trans-(Et3P)2Pt(H)(η2-
NB-COOEt)+ to 1 reaches an upper limit is also
consistent with the rate-inhibiting effect of free phos-
phine; the inserted product 1 has cis-configuration of
the two Et3P ligands, one of which is trans to the alkyl
group which has a high trans-influence (cf. J Pt-P ) 1630
Hz). This feature likely facilitates phosphine dissocia-
tion from 1, which, in turn, inhibits the conversion of
trans-(Et3P)2Pt(H)(η2-NB-COOEt)+ to 1.
F igu r e 9. Formation of the endo-σ,π-bonded complex.
insertion chemistry on the same face. There are numer-
ous examples of insertion on the exo face of norbornene
substrates found in the literature.4 Formation of endo-
palladium or platinum-carbon R,π-bonded complexes
are reported in the literature. However, their formation
is typically the result of nucleophilic attack by an
anionic species10 on the exo-face of the norbornene
derivative, such as norbornadiene, in which the metal
is coordinated to the endo face of the norbornene
derivative (Figure 9).
The platinum metal center in complex 1 exhibits
square-planar geometry. The summation of bond angles
of the platinum inner coordination sphere is approxi-
mately 360°. The bond angle (101.36(11)°) between the
two phosphines is splayed out from the ideal. The other
inner coordination angles are as a result less than 90°.
The two Pt-P bond distances in 1 are substantially
different from one another. The Pt-P2 distance is
2.355(3) Å. The Pt-P9 bond distance is significantly
shorter at 2.211(3) Å.
Similar coordination environments are established for
platinum and palladium by X-ray methods11 (Figure 10).
As was found for 1, the complexes in Figure 10 exhibit
square-planar environments and longer metal-phos-
phorus distances for those phosphines that are trans to
carbon and shorter distances for those trans to oxygen.
This is expected due to the stronger trans influence of
a carbon σ-bonded to a metal compared to a neutral,
two-electron oxygen donor. All of the complexes in
By virtue of the endo:exo composition of the NB-
COOEt, the actual sequence of transformations is more
complex. The coordination of endo NB-COOEt to the
(Et3P)2Pt(H)+ fragment needs to occur on the endo face
(A′) in order to form 1. Although the stereochemistry of
NB-COOEt coordination in trans-(Et3P)2Pt(H)(η2-NB-
COOEt)+ has not been determined, it is likely to be on
the exo face. The exo face of norbornene is less sterically
congested than the endo face. Indeed, the previous
reports on the insertion of (unsubstituted) norbornene
(10) (a) Maitlis, P. M. The Organic Chemistry of Palladium; Aca-
demic Press: New York, 1971; Vol. 1, p 161. (b) Hughes, R. P.; Powell,
J . J . Organomet. Chem. 1973, 60, 427, and references therein.
(11) (a) Brock, C. P.; Attig, T. G. J . Am. Chem. Soc. 1980, 102, 1319.
(b) Brumbaugh, J . S.; Whittle, R. R.; Parvez, M.; Sen, A. Organome-
tallics 1990, 9, 1735. (c) Hinkle, R. J .; Stang, P. J .; Arif, A. M.
Organometallics 1993, 12, 2510. (d) Markies, B. A.; Kruis, D.; Rietveld,
M. H. P.; Spek, A. L.; van Koten, G. J . Am. Chem. Soc. 1995, 117,
5263.

2812 Organometallics, Vol. 20, No. 13, 2001
Sch em e 1. Rea ction Equ ilibr ia Lea d in g to th e F or m a tion of 1
Hennis et al.
into metal-carbon bonds indicate an exo, exo stereo-
chemistry.4 Thus, the formation of 1 requires the prior
migration of the (Et3P)2Pt(H)+ fragment from the exo
to the endo face of NB-COOEt. This isomerization
likely occurs via NB-COOEt dissociation and is kineti-
cally slow: 1H and 31P {1H} NMR spectra of both trans-
(Et3P)2Pt(H)(η2-NB-COOEt)+ and trans-(Et3P)2Pt(H)-
(η2-norbornene)+, recorded at 65 °C in chlorobenzene-
d5, show virtually no change with respect to those
recorded at ambient temperature. More importantly, the
likely exo-coordination of NB-COOEt to the (Et3P)2Pt-
(H)+ fragment indicates the nonparticipation of the
carbonyl group in determining the regiochemistry of the
first interaction with the alkene: if the carbonyl group
were to coordinate first to the (Et3P)2Pt(H)+ fragment,
the subsequent coordination of the CdC bond would
occur through the endo face, at least for endo-NB-
COOEt.
An interesting question is why the insertion occurs
at the endo face even though the alkene is initially
coordinated through the exo face. It is likely that
insertion does occur at the exo face, but the resultant
product (C) cannot be stabilized by coordination of the
carbonyl oxygen and â-hydrogen abstraction from C
regenerates the starting alkene adduct. On the other
hand, the product derived through insertion at the endo
face (1) is more stable because of the coordination of the
carbonyl group. Thus, while the exo insertion may be
kinetically fast, the endo insertion product is thermo-
dynamically favored.
The reverse reaction, 1 to trans-(Et3P)2Pt(H)(η2-endo-
norbornene-COOEt)+, takes place, although it is very
slow. Heating a chlorobenzene-d5 solution of 1 in the
presence of 5 equiv of 1-hexene at 65 °C for 18 h led to
nearly complete isomerization of the terminal olefin into
a mixture of internal unsaturated isomers. The 31P{1H}
NMR spectra of the reaction mixture showed that ca.
90% of 1 remained intact, while the remaining 10%
rearranged to trans-(Et3P)2Pt(H)(η2-endo-NB-COOEt)+,
the latter being responsible for isomerization of 1-hex-
ene. A similar 31P{1H} NMR spectrum was obtained
when 5 equiv of NB-COOEt was used instead of
1-hexene. It therefore appears that 1 can â-hydrogen
eliminate the inserted alkene, producing the catalyti-
cally active hydride adduct.
Con clu sion
A palladium(II)-based catalyst system has been de-
scribed for the efficient polymerization of norbornene
derivatives, including those with pendant oxygen func-
tionalities. The very high polymerization rates and the
ability to control polymer molecular weight through the
addition of an alcohol make the system attractive from
a practical standpoint. For functional norbornenes, the
rates of polymerization are substantially slower than
for the nonfunctional monomers. It has been demon-
strated that for such functional norbornenes, the exo
isomer is preferentially incorporated into the growing
polymer chain. Finally, by using a model platinum-
based system, we have for the first time demonstrated
the endo, endo insertion of the endo isomer of an ester-
functionalized norbornene forming a chelated platinum
complex. The formation of chelates, both upon the
coordination of the endo-functionalized norbornene and
in the endo-inserted product, appears to be responsible,
in part, for the observed decrease in polymerization rate
for functional norbornene derivates. A further reason
for the diminution of activity of both the endo- and the
exo-functionalized isomers is simply the coordination of
the functionality to the metal center.
Ack n ow led gm en t. This research was supported by
a grant from the U.S. Department of Energy, Office of
Basic Energy Sciences (DE-FG02-84ER13295), and the
Advanced Technology Program of NIST (ATP Project 95-
05-0038). The authors would like to thank Seth B.
Harkins for synthesizing Na+[3,5-(CF3)2C6H3]4B- used
in portions of this work, Dr. Ian Harrison for usage of
the DSC, Dr. Shahid Murtuza for assistance in its use,
and Dr. Richard Vicari for suggesting an alternative
synthesis of the 5,6-norbornenedicarboxylic acid methyl
esters. We would also like to thank Elf Atochem, North
America, for their analysis of poly(norbornene) and poly-
(vinyl norbornene) samples.
Su p p or tin g In for m a tion Ava ila ble: Details regarding
the H and 13C 2D-NMR spectra and assignments for complex
1
1, ORTEP representations of two alternate solutions to the
disorder found in the X-ray structural study of 1, and a
complete set of distances and angles, atomic coordinates, and
thermal parameters. This material is available free of charge
via the Internet at http://pubs.acs.org.
OM010232M