Stereospecific ring-opening metathesis polymerization of norbornadienes employing tungsten oxo alkylidene initiators

Article abstract of DOI:10.1021/ja506446n

We report here the polymerization of several 7-isopropylidene-2,3- disubstituted norbornadienes, 7-oxa-2,3-dicarboalkoxynorbornadienes, and 11-oxa-benzonorbornadienes with a single tungsten oxo alkylidene catalyst, W(O)(CH-t-Bu)(OHMT)(Me₂Pyr) (

Full text of DOI:10.1021/ja506446n

Communication
pubs.acs.org/JACS
Stereospecific Ring-Opening Metathesis Polymerization of
Norbornadienes Employing Tungsten Oxo Alkylidene Initiators
William P. Forrest, Jonathan G. Weis, Jeremy M. John, Jonathan C. Axtell, Jeffrey H. Simpson,
Timothy M. Swager, and Richard R. Schrock*
Department of Chemistry, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, United States
S
* Supporting Information
dramatically upon addition of B(C6F5)3 as a consequence of its
ABSTRACT: We report here the polymerization of
several 7-isopropylidene-2,3-disubstituted norbornadienes,
7-oxa-2,3-dicarboalkoxynorbornadienes, and 11-oxa-benzo-
norbornadienes with a single tungsten oxo alkylidene
catalyst, W(O)(CH-t-Bu)(OHMT)(Me2Pyr) (OHMT =
2,6-dimesitylphenoxide; Me2Pyr = 2,5-dimethylpyrrolide)
to give cis, stereoregular polymers. The tacticities of the
menthyl ester derivatives of two polymers were
determined for two types. For poly(7-isopropylidene-2,3-
dicarbomenthoxynorbornadiene) the structure was shown
to be cis,isotactic, while for poly(7-oxa-2,3-dicarbo-
menthoxynorbornadiene) the structure was shown to be
cis,syndiotactic. A bis-trifluoromethyl-7-isopropylidene nor-
bornadiene was not polymerized stereoregularly with
W(O)(CHCMe2Ph)(Me2Pyr)(OHMT) alone, but a cis,
stereoregular polymer was formed in the presence of 1
equiv of B(C6F5)3.
binding reversibly to the oxo ligand and thereby enhancing the
electrophilicity of the metal.6,7
We now find that tungsten oxo alkylidene MAP complexes can
be employed for the synthesis of cis and highly tactic polymers
from monomers other than DCMNBD. These monomers
(Figure 1) either have not been polymerized before, or if they
have, the polymer that is formed is not stereoregular.
Figure 1. Norbornadienes explored in this study.
he preparation of highly tactic polymers through ring-
1
Topening metathesis polymerization (ROMP) is a key to
We were drawn to 7-isopropylidenenorbornadienes because
polymers made from them through ROMP have potential as
intermediates in the synthesis of highly conjugated polymers and
because polymerization of 7-isopropylidenenorbornadienes in a
stereospecific manner with selected Mo or W initiators has been
challenging.8−10 We chose first to examine the polymerization of
2,3-dicarbomethoxy-7-isopropylidenenorbornadiene (1a), the 7-
isopropylidene analogue of DCMNBD. A previous attempt to
polymerize 1a with Mo(NAr)(CH-t-Bu)(O-t-Bu)2 (Ar = 2,6-i-
Pr2C6H3) failed, even at 45−55 °C.10 A monoinsertion product
could be isolated and structurally characterized, but it would not
react further with 1a.
Monomer 1a (100 equiv, 0.1 M in CDCl3) is polymerized
smoothly by W(O)(CH-t-Bu)(Me2Pyr)(OHMT)(PMe2Ph)
(A; OHMT = O-2,6-Mes2C6H3) to give cis and highly tactic
poly(1a)A in essentially 100% yield. In the IR spectrum a
relatively intense absorption near 980 cm−1 for the CH
vibrational mode that is characteristic of trans double bonds
between units is absent. Therefore, poly(1a)A contains cis CC
linkages. The 1H NMR spectrum of poly(1a)A contains
characteristic second-order olefinic (at 5.23 ppm) and methine
(at 4.63 ppm) proton resonances, each of which simplifies upon
decoupling the other (Figure 2). The 13C NMR spectrum also
the advancement of many applications involving ROMP with
well-defined catalysts,2 in part because the properties of
stereoregular polymers (melting point, crystallinity, solubilities,
etc.) usually are more sharply defined for tactic polymers relative
to atactic polymers.3 Molybdenum-based imido alkylidene
ROMP initiators that contain a biphenolate ligand were the
first well-defined initiators to be shown to yield stereoregular cis,
isotactic polymers through enantiomorphic site control.4 More
recently, monoaryloxide pyrrolide (MAP) complexes with the
general formula M(NR)(CHR′)(pyrrolide)(OR′′) (M = Mo or
W) have been shown to yield cis,syndiotactic poly(DCMNBD)
(DCMNBD = 2,3-dicarbomethoxynorbornadiene) as a con-
sequence (it is proposed) of inversion of the configuration at the
metal with each monomer insertion (stereogenic metal
control);5 inversion of the configuration at the metal forces the
monomer to approach first one side of the MC bond and then
the other. The most recent variation of MAP complexes are
tungsten oxo alkylidene complexes, W(O)(CHR)(Me2Pyr)-
(OAr) (R = CMe2Ph or t-Bu; Me2Pyr = 2,5-dimethylpyrrolide;
OAr = an aryloxide), or monoadducts thereof that contain a
labile phosphine or acetonitrile.6 Tungsten oxo MAP complexes
have been shown to be efficient initiators for formation of cis,
syndiotactic-poly(DCMNBD).7 A potentially important feature
of polymerization of DCMNBD employing tungsten oxo
initiators is that the rate of polymerization is accelerated
Received: June 26, 2014
Published: July 18, 2014
© 2014 American Chemical Society
10910
dx.doi.org/10.1021/ja506446n | J. Am. Chem. Soc. 2014, 136, 10910−10913

Journal of the American Chemical Society
Communication
protons, two polymer methine protons, and two polymer olefinic
protons (see SI). In toluene-d8 at 80 °C the olefinic protons are
separated from each other and from the four other proton
resonances (Figure 4). The pseudo triplet structure of each (at
Figure 2. (a) 1H NMR (500 MHz) spectrum of the olefinic region of
poly(1a)A with (b) the methine proton selectively decoupled, and with
(c) the olefinic proton selectively decoupled, in CDCl3 at 22 °C.
contains only the number of sharp resonances expected for a
single structure (see Supporting Information (SI)). The
relatively simple and sharp proton and 13C NMR resonances
are both consistent with poly(1a)A having a single tacticity
(>99%). The polymerization of 50 equiv of 1a with W(O)-
(CHCMe2Ph)(Me2Pyr)(OHMT)7 (B) resulted in a polymer
whose proton and 13C NMR spectra are identical to the spectra
obtained for poly(1a)A. Initiator A is known to lose the relatively
labile phosphine to give B, so the propagating species in each case
is the four-coordinate 14e species.
To prove the tacticity of a stereoregular polymer it is necessary
that a chiral group be present. For example, the tacticity of 2,3-
dicarbomenthoxy-7-isopropylidenenorbornadiene could be pro-
ven4,11 if two separate olefinic proton resonances are found in the
1H NMR spectrum of the polymer. If the two olefinic protons are
coupled strongly with JHH ≈ 11 Hz, the structure is cis,isotactic
(Figure 3; Y = CCMe2). If the two olefinic protons are not
coupled strongly, then the structure is cis,syndiotactic (Figure 3; Y
= CCMe2).
Figure 4. 1H/1H COSY NMR spectrum of the 4.8−5.6 ppm region for
poly(1b)B in toluene-d8 at 80 °C.
∼5.48 and ∼5.56 ppm) suggests that Ha and Hb are coupled. This
conclusion is confirmed through a 1H/1H COSY NMR
spectrum, which shows strong cross peaks between Ha and Hb
(lower left in Figure 4). Therefore, the structure of poly(1b)B is
cis,isotactic (Figure 3) when prepared in the presence of 1 equiv of
B(C6F5)3, not cis,syndiotactic, as we initially proposed would be
the case on the basis of the tendency of MAP catalysts to yield cis,
syndiotactic polymers from a small collection of monomers
through what we have called stereogenic metal control.2 We
ascribe the acceleration of the polymerization to binding of
boron to the oxo ligand and a resulting increase in the
electrophilicity of W.6
An important question is whether the structure of poly(1b) is
the same when prepared in the absence of B(C6F5)3. Fortunately,
1b is polymerized completely by B at 55 °C in 16 h. According to
1
the H NMR spectra the structure of the resulting polymer is
identical to the polymer made from B in the presence of
B(C6F5)3. Therefore, the presence of B(C6F5)3 does not alter the
structure of the polymer that is formed.
To discover other examples of the stereospecific ROMP of
monomers by A, B, or B in the presence of B(C6F5)3, we chose to
investigate the 7-oxanorbornadiene monomers (Figure 1), 2,3-
dicarbomethoxy-7-oxanorbornadiene (2a), 2,3-dicarbodecyloxy-
7-oxanorbornadiene (2b), and 2,3-dicarbomenthoxy-7-
oxanorbornadiene (2c). Monomer 2a has been polymerized
previously using Mo(NAr)(CHR)(O-t-Bu)2 (R = t-Bu or
CMe2Ph) as the initiator to give poly(2a) that contains both
cis and trans CC bonds.11
Figure 3. Olefinic protons Ha and Hb in cis,syndiotactic and cis,isotactic
polymers, where X* = carbomenthoxy.
Neither A nor B initiates the homopolymerization of 1b at
room temperature in several hours. However, in the presence of 1
equiv of B(C6F5)3 polymerization of 1b by B is complete in
chloroform in ∼3 h under the standard set of conditions (see SI)
The polymerizations of 50 equiv of 2a and 2b by A proceeded
relatively rapidly and smoothly to give CDCl3-soluble polymers
within 5 min at 22 °C. Addition of A to the monomer solutions
caused an initial color change to red-orange which faded to
orange as polymerization proceeded. After poly(2a)A was
to give an all cis, highly tactic polymer, according to IR and 13
NMR spectra. The H NMR spectrum in the olefinic proton
C
1
region contains overlapping resonances for two menthyl OCH
10911
dx.doi.org/10.1021/ja506446n | J. Am. Chem. Soc. 2014, 136, 10910−10913

Journal of the American Chemical Society
Communication
precipitated from methanol, the resulting white polymer was
insoluble in common organic solvents at 22 °C. However, 1H and
13C NMR spectra of poly(2a)A could be obtained in CD2Cl4 at 80
°C. From IR spectroscopy it was determined that isolated
1
poly(2a)A and poly(2b)A are cis. In addition, the H and 13C
NMR spectra reveal that each of the polymers has a highly regular
structure (see SI).
Polymerization of 2c with A was complete within 10 min at 22
°C to give a CDCl3-soluble, white polymer. From IR spectros-
copy it is clear that poly(2c)A has all cis-olefinic linkages. In the
1H NMR spectrum of isolated poly(2c)A, there are two methine
resonances at 5.85 ppm and two resonances for Ha and Hb at 5.50
ppm (Figure 5 and SI). (The olefinic protons were identified
1
Figure 6. H NMR (500 MHz) spectra in THF-d8 at 22 °C of the
olefinic region for poly(1c) formed (a) from initiator A, and (b) from
initiator A in the presence of 2 equiv of B(C6F5)3.
presence of two equivalents of B(C6F5)3 to give a cis, highly tactic
1
polymer in high yield, whose H and 13C NMR spectra are
comparatively simple (see SI). The 4.2−5.8 ppm region of the 1H
NMR spectrum (Figure 6b) reveals only relatively sharp olefinic
(at ∼5.4 ppm) and methine (at ∼4.6 ppm) resonances. We
propose that the poly(1c)A formed with A in the absence of
B(C6F5)3 contains isotactic (r) and syndiotactic (m) dyads, but
also longer range variations in tetrads (i.e., mrm, mrr/rrm, mmr/
rmm, and rmr) that introduce other olefinic proton resonances.
We entertained the possibility that the polymer had undergone
various 1,3-proton shifts of the triply allylic methine proton
during polymerization,13 but at this stage we reject that
possibility in favor of the simpler explanation. We propose that
coordination of B(C6F5)3 to the oxo ligand6,7 leads to a faster
formation of either a rrr (syndiotactic) or a mmm (isotactic)
structure.
Figure 5. gCOSY of the methine (∼5.85 ppm) and olefinic proton
(∼5.50 ppm) region for poly(2c)A in CDCl3 at 22 °C.
2,3-Dicyano-7-isopropylidenenorbornadiene (1d) can be
polymerized in 16 h at 22 °C or 1 h at 45 °C to give a relatively
1
insoluble polymer whose H NMR spectrum in the olefinic
through an HSQC experiment; see SI.) Upon decoupling a given
methine proton, one olefinic resonance sharpens in a manner
analogous to that shown in Figure 2. The absence of olefinic
region is similar to that shown in Figure 6a for poly(1c)A (see SI).
However, an attempted polymerization of 1d in the presence of
B(C6F5)3 did not yield a polymer with a simplified spectrum. We
propose that poly(1d)A is also a mixture of rrr tetrads, mmm
tetrads, and variations, but B(C6F5)3 must bind essentially
exclusively to the cyano groups in the monomer or polymer
instead of to the oxo group, so the structure in this case does not
simplify when the polymerization is run in the presence of
B(C6F5)3. In spite of the fact that poly(1d)A is not stereoregular,
1d simply could not be polymerized by other Mo, W, and Ru
initiators that were examined (see SI).
1
cross-peaks between Ha and Hb resonances in the gCOSY H
NMR spectrum (500 MHz) of poly(2c)A confirmed that Ha and
Hb (with resonances at 5.48 and 5.52 ppm) are not coupled
(Figure 5) and that poly(2c)A therefore has a cis,syndiotactic
structure. The weak cross peaks between the two types of
methine protons (with resonances at 5.83 and 5.88 ppm) on one
furan ring in poly(2c)A are ascribed to a small (∼4 Hz) four-bond
coupling between them.
2,3-Bis(trifluoromethyl)-7-isopropylidenenorbornadiene (1c)
was previously polymerized by Feast and Millichamp12 using a
classical catalyst (MoCl5/SnMe4) in 30 min at 70 °C in
chlorobenzene. The cis content was proposed to be “high”, but
the degree of tacticity of the polymer was not stated. Monomer
1c is polymerized by A in 16 h at 22 °C or 1 h at 45 °C to yield a
THF-soluble polymer whose 1H NMR spectrum in the olefinic
and methine proton region is unexpectedly complex (Figure 6a).
The same is true for poly(1c)C prepared employing W(O)-
(CHCMe2Ph)(Me2Pyr)(ODFT)(PPh2Me)7 (C; ODFT = O-
2,6-(C6F5)2C6H3) as the initiator. IR spectroscopy confirmed
that poly(1c)A contains no trans linkages, so cis/trans isomerism
is not the cause of the relatively complex set of olefinic
resonances around 5.4 ppm in Figure 6a. In contrast, 1c (50 or
250 equiv) is polymerized smoothly by A in 3 h at 22 °C in the
11-Isopropylidenebenzonorbornadiene (3a) has been poly-
merized (48 h at room temperature) previously using Mo-
(NAr)(CH-t-Bu)(OR)2 (R = t-Bu or CMe(CF3)2) initiators.8c
The resulting poly(3a) was found to contain 80% trans-olefinic
bonds. Monomer 3a can be polymerized by A in ∼5 min to give a
highly regular, CDCl3-soluble polymer. IR spectroscopy
confirmed that poly(3a)A is cis, while the relatively sharp
resonances in the proton and 13C NMR spectra are what would
be expected for a tactic structure (see SI), but in view of
poly(1b)B having a cis,isotactic structure, we are reluctant to
assign a tacticity to poly(3a)A.
11-Oxabenzonorbornadiene (3b) (50 equiv) can be poly-
merized by A to yield a polymer that precipitates out of CDCl3
within 30 s and is insoluble in common organic solvents at 22 °C.
However, the proton and 13C NMR spectra of poly(3b)A could
10912
dx.doi.org/10.1021/ja506446n | J. Am. Chem. Soc. 2014, 136, 10910−10913

Journal of the American Chemical Society
Communication
Notes
be obtained in 1,1,2,2-tetrachloroethane-d2 at 120 °C (see SI).
The resonances in both the 1H and 13C NMR spectra are highly
resolved single resonances, indicative of a highly regular
structure. From IR spectroscopy it was again confirmed that
poly(3b)A is cis, but again we are reluctant to assign a tacticity.
The polymerization of all monomers presented in this work
(except the dimentholates) were investigated using a variety of
Mo and W imido alkylidene and Ru alkylidene initiators (see SI).
In some cases the monomer is polymerized, but the resulting
polymer is not stereoregular. Monomers 1a, 1b, and 1c were not
polymerized by the Ru initiators G2 or G3.
We conclude that A or B (alone or in the presence of
B(C6F5)3) will initiate the stereospecific polymerization of
several 7-isopropylidenenorbornadienes and 7-oxanorborna-
dienes. We also conclude that cis,syndiotactic and cis,isotactic
structures can both be formed employing the tungsten oxo MAP
initiators, the latter probably without inversion of configuration at
W.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
R.R.S. thanks the Department of Energy (DE-FG02-
86ER13564) for financial support. T.M.S. acknowledges the
financial support of the Chemical and Biological Technologies
Department of the Defense Threat Reduction Agency (DTRA-
CB) via grant BA12PHM123 in the “Dynamic Multifunctional
Materials for a Second Skin D[MS]2” program. We thank Dr.
Benjamin Autenrieth for his assistance with the high-temperature
NMR measurements for poly(3a)A and poly(3b)A.
REFERENCES
■
(1) (a) Ivin, K. J.; Mol, J. C. Olefin Metathesis and Metathesis
Polymerization; Academic Press: San Diego, 1997. (b) Ivin, K. J. Olefin
Metathesis; Academic Press: San Diego, 1983. (c) Buchmeiser, M. R.
Chem. Rev. 2000, 100, 1565. (d) Bielawski, C. W.; Grubbs, R. H. Prog.
Polym. Sci. 2007, 32, 1. (e) Smith, E.; Pentzer, E. B.; Nguyen, S. T. Polym.
Rev. 2007, 47, 419. (f) Feast, W. J.; Khosravi, E. New Methods Polym.
Synth. 1995, 69. (g) Buchmeiser, M. R. Chem. Rev. 2009, 109, 303.
(2) (a) Schrock, R. R. Dalton Trans. 2011, 40, 7484. (b) Schrock, R. R.
Acc. Chem. Res. 2014, DOI: 10.1021/ar500139s.
One piece of evidence in the literature suggests that cis,isotactic
and cis,syndiotactic structures can be formed with different MAP
initiators; although Mo(NC6F5)(CHR)(Me2Pyr)(OHMT) (R =
CMe2Ph) yields >98% cis,syndiotactic-poly(DCMNBD), Mo-
(NC6F5)(CHR)(Me2Pyr)(ODFT) (ODFT = decafluoro-
terphenoxide = O-2,6-(C6F5)2C6H3) yields 95% cis,91%
isotactic-poly(DCMNBD).14
(3) (a) Coates, G. W. Chem. Rev. 2000, 100, 1223. (b) Baugh, L. S.,
Canich, J. M., Eds. Stereoselective Polymerization with Single-Site Catalysts;
CRC Press: Boca Raton, FL, 2007.
We propose that a key issue that determines whether an
isotactic or syndiotactic structure is formed with a MAP catalyst
is the slow rate at which a five-coordinate metallacyclobutane
intermediate rearranges its configuration at the metal (aryloxide/
pyrrolide exchange) through a Berry-type process15 relative to the
rate at which the metallacyclobutane ring opens to give a
propagating species without rearrangement at the metal. In the
case of monomers that create some steric challenges, as with 1b,
both the rate of rearrangement of a five-coordinate metalla-
cyclobutane complex and the rate of opening the metalla-
cyclobutane to give the next alkylidene could be altered
dramatically, perhaps in opposite directions. Some form of
chain end control always remains a possibility in any given
circumstance, but the simplicity of the above proposal is
attractive. The facts that each structure can be formed with a
high degree of precision and specificity employing one MAP
initiator and different monomers, and that B(C6F5)3 can speed
up formation of one of these two stereoregular structures, are
surprising, but fascinating, developments.
(4) (a) McConville, D. H.; Wolf, J. R.; Schrock, R. R. J. Am. Chem. Soc.
1993, 115, 4413. (b) O’Dell, R.; McConville, D. H.; Hofmeister, G. E.;
Schrock, R. R. J. Am. Chem. Soc. 1994, 116, 3414. (c) Totland, K. M.;
Boyd, T. J.; Lavoie, G. G.; Davis, W. M.; Schrock, R. R. Macromolecules
1996, 29, 6114.
(5) (a) Flook, M. M.; Jiang, A. J.; Schrock, R. R.; Muller, P.; Hoveyda,
̈
A. H. J. Am. Chem. Soc. 2009, 131, 7962. (b) Flook, M. M.; Gerber, L. C.
H.; Debelouchina, G. T.; Schrock, R. R. Macromolecules 2010, 43, 7515.
(c) Flook, M. M.; Ng, V. W. L.; Schrock, R. R. J. Am. Chem. Soc. 2011,
133, 1784. (d) Flook, M. M.; Borner, J.; Kilyanek, S.; Gerber, L. C. H.;
̈
Schrock, R. R. Organometallics 2012, 31, 6231.
(6) (a) Peryshkov, D. V.; Schrock, R. R.; Takase, M. K.; Muller, P.;
̈
Hoveyda, A. H. J. Am. Chem. Soc. 2011, 133, 20754. (b) Peryshkov, D.
V.; Schrock, R. R. Organometallics 2012, 31, 7278. (c) Peryshkov, D. V.;
Forrest, W. P., Jr.; Schrock, R. R.; Smith, S. J.; Muller, P. Organometallics
̈
2013, 32, 5256.
(7) Forrest, W. P.; Axtell, J. C.; Schrock, R. R. Organometallics 2014, 33,
2313.
(8) (a) Stelzer, F.; Muelner, R.; Schlick, H.; Leising, G. NATO Sci. Ser.
II 2002, 56, 185. (b) Feast, W. J. NATO Sci. Ser. II 2002, 56, 177.
(c) Schimetta, M.; Stelzer, F. Macromol. Chem. Phys. 1994, 195, 2699.
(9) A rare example of a successful polymerization is that of 11-(α-
phenylbenzylidene)benzonorbornadiene, the diphenylmethylidene ana-
logue of 3a, by Mo(NAr)(CH-t-Bu)[OC(CF3)2Me]2, to yield a polymer
that is entirely cis and that is highly tactic with unusually simple 1H and
13C NMR spectra.8c This specificity must result from a chain end control
mechanism.
(10) Bazan, G. C.; Khosravi, E.; Schrock, R. R.; Feast, W. J.; Gibson, V.
C.; O’Regan, M. B.; Thomas, J. K.; Davis, W. M. J. Am. Chem. Soc. 1990,
112, 8378.
(11) Bazan, G. C.; Oskam, J. H.; Cho, H.-N.; Park, L. Y.; Schrock, R. R.
J. Am. Chem. Soc. 1991, 113, 6899.
Finally, it should be noted that a MAP catalyst is known to
catalyze formation of trans,isotactic-poly((+)-2,3-dicarbo-
methoxynorbornene).5d The mechanism of formation of the
trans,isotactic polymer is proposed to involve formation of a trans-
metallacyclobutane ring that flips over rapidly in a Berry-type
process (without exchanging the aryloxide and pyrrolide) before
it opens to give the propagating alkylidene.
ASSOCIATED CONTENT
■
S
* Supporting Information
(12) (a) Feast, W. J.; Khosravi, E. J. Fluorine Chem. 1999, 100, 117.
(b) Feast, W. J.; Millichamp, I. S. J. Mol. Catal. 1985, 28, 331.
(13) Gerber, L. C. H.; Schrock, R. R. Organometallics 2013, 32, 5573.
Experimental details for the syntheses and characterization of
new monomers and all polymers, along with a description of
NMR studies that prove two tacticities. This material is available
free of charge via the Internet at http://pubs.acs.org.
(14) Yuan, J.; Schrock, R. R.; Gerber, L. C. H.; Muller, P.; Smith, S. J.
̈
Organometallics 2013, 32, 2983.
(15) Moberg, C. Angew. Chem., Int. Ed. 2011, 50, 10290.
AUTHOR INFORMATION
■
Corresponding Author
rrs@mit.edu
10913
dx.doi.org/10.1021/ja506446n | J. Am. Chem. Soc. 2014, 136, 10910−10913