Living polymerization of (o-(trimethylsilyl)phenyl)acetylene by molybdenum imido alkylidene complexes

Article abstract of DOI:10.1021/ja954155w

syn-Mo(CHCMe₂Ph)(NAd)[OCH(CF₃)₂]₂(2,4-lutidine) (2a; Ad = 1-adamantyl) is a distorted trigonal bipyramid in which 2,4-lutidine occupies an axial position, a structure that results from addition of 2,4-lutidine to the CNO face of unstable pseudotetrahedral syn-Mo(CHCMe₂Ph)(NAd)[OCH(CF₃)₂]₂. 2a reacts with (o-(trimethylsilyl)phenyl)acetylene (o-TMSPA) solely via formation of an α-substituted metallacyclobutene intermediate (α addition) that opens to give a single rotamer of a disubstituted alkylidene complex. o-TMSPA is smoothly polymerized at a rate k(2a)[2a]₀[o-TMSPA] when [2a] < 1 mM with a propagation rate constant k(2a) = 0.30 s^-1 M^-1. Additional studies confirmed that the disubstituted alkylidene propagating species is essentially base-free (K(2a) = 62 M^-1) and that the propagating species is stable under catalytic conditions (25°C). Other versions of the Mo(CHCMe₂Ph)(NAd)[OCH(CF₃)₂]₂(base) catalyst are either inactive (base = pyridine) or unstable (base = 2-(3-pentyl)pyridine). Mo(CHCMe₂Ph)(NAr')(OC₆F₅)₂(quinuclidine) (7; Ar' = 2,6-Me₂C₆H₃) will also react smoothly with (o-(trimethylsilyl)phenyl)acetylene to give poly(o-TMSPA) with K₇ = 1380 M^-1 and k₇ = 0.23 s^-1 M^-1. Low-polydispersity polyenes containing up to 150 equiv of o-TMSPA can be obtained readily using either catalyst. The thermodynamically most stable form of poly(o-TMSPA), which contains ~25 double bonds, is air-sensitive and has a significantly red-shifted λ(max). o-t-BuPA also can be polymerized to give highly conjugated polyenes, but o-iPrPA, o-MePA, and phenylacetylene itself add to initiator 2a with decreasing α regiospecificity (73%, 60%, and 56%, respectively). A lack of regiospecificity we propose leads to polymers that do not have a pure head-to-tail structure, have a lower degree of conjugation, and have a progressively more blue-shifted λ(max).

Full text of DOI:10.1021/ja954155w

J. Am. Chem. Soc. 1996, 118, 3883-3895
3883
Living Polymerization of (o-(Trimethylsilyl)phenyl)acetylene by
Molybdenum Imido Alkylidene Complexes
Richard R. Schrock,* Shifang Luo, Jesse C. Lee, Jr., Nadia C. Zanetti, and
William M. Davis
Contribution from the Department of Chemistry 6-331, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139
ReceiVed December 11, 1995X
Abstract: syn-Mo(CHCMe2Ph)(NAd)[OCH(CF3)2]2(2,4-lutidine) (2a; Ad ) 1-adamantyl) is a distorted trigonal
bipyramid in which 2,4-lutidine occupies an axial position, a structure that results from addition of 2,4-lutidine to
the CNO face of unstable pseudotetrahedral syn-Mo(CHCMe2Ph)(NAd)[OCH(CF3)2]2. 2a reacts with (o-
(trimethylsilyl)phenyl)acetylene (o-TMSPA) solely via formation of an R-substituted metallacyclobutene intermediate
(R addition) that opens to give a single rotamer of a disubstituted alkylidene complex. o-TMSPA is smoothly
polymerized at a rate k2a[2a]0[o-TMSPA] when [2a] < 1 mM with a propagation rate constant k2a ) 0.30 s-1 M-1
.
Additional studies confirmed that the disubstituted alkylidene propagating species is essentially base-free (K2a ) 62
M-1) and that the propagating species is stable under catalytic conditions (25 °C). Other versions of the Mo-
(CHCMe2Ph)(NAd)[OCH(CF3)2]2(base) catalyst are either inactive (base ) pyridine) or unstable (base ) 2-(3-
pentyl)pyridine). Mo(CHCMe2Ph)(NAr′)(OC6F5)2(quinuclidine) (7; Ar′ ) 2,6-Me2C6H3) will also react smoothly
with (o-(trimethylsilyl)phenyl)acetylene to give poly(o-TMSPA) with K7 ) 1380 M-1 and k7 ) 0.23 s-1 M-1. Low-
polydispersity polyenes containing up to 150 equiv of o-TMSPA can be obtained readily using either catalyst. The
thermodynamically most stable form of poly(o-TMSPA), which contains ∼25 double bonds, is air-sensitive and has
a significantly red-shifted λmax. o-t-BuPA also can be polymerized to give highly conjugated polyenes, but o-i-
PrPA, o-MePA, and phenylacetylene itself add to initiator 2a with decreasing R regiospecificity (73%, 60%, and
56%, respectively). A lack of regiospecificity we propose leads to polymers that do not have a pure head-to-tail
structure, have a lower degree of conjugation, and have a progressively more blue-shifted λmax
.
Introduction
conjugated. Polyenes prepared from monosubstituted acetylenes
also tend to be poorly conjugated in general and relatively poorly
characterized at a molecular level. But an overwhelming
problem is the paucity of liVing polymerization systems that
would lead to polyenes with a narrow molecular weight
distribution about a known number average molecular weight
(low polydispersity). Soluble substituted polyenes have been
prepared indirectly by ring opening substituted cyclooctatetra-
enes, but the polydispersities of these polyenes were not
consistent with a living polymerization.10
There has been significant progress in understanding the
principles of alkyne polymerization in the last few years. For
example, there is now good evidence that alkynes can be
polymerized in a living manner via alkylidene complexes,11-19
and polyenes prepared from certain substituted phenylacetylenes,
in particular (o-(trimethylsilyl)phenyl)acetylene (o-TMSPA),
Organic polymers that contain a π conjugated backbone have
significant potential as nonlinear optical (NLO) materials.1-5
Among the advantages that such materials might have over
inorganic crystals that display similar NLO properties are a sub-
picosecond response time and the ability to be subtly “tuned”
by systematically varying organic groups within the material.
Although polyenes fall into this category of potentially interest-
ing NLO materials, several fundamental problems have pre-
vented establishment of the necessary relationship between chain
length, chain structure, and degree of conjugation, with an NLO
property such as the third-order molecular hyperpolarizability,
γ. One problem is the fact that an unsubstituted polyene that
contains more than ∼15 double bonds tends to be insoluble.6
Polyacetylene itself is a black, intractable, air-sensitive solid
that remains relatively poorly characterized at a molecular level.
Although polyenes can be prepared by polymerization of
disubstituted acetylenes,7-9 the resulting polymers are not highly
(8) Masuda, T.; Higashimura, T. AdV. Polym. Sci. 1986, 81, 122.
(9) Masuda, T.; Higashimura, T. Acc. Chem. Res. 1984, 17, 51.
(10) Gorman, C. B.; Ginsburg, E. J.; Grubbs, R. H. J. Am. Chem. Soc.
1993, 115, 1397.
(11) Masuda, T.; Fujimori, J. I.; Abrahman, M. Z.; Higashimura, T.
Polym. J. 1993, 25, 535.
(12) Wallace, K. C.; Liu, A. H.; Davis, W. M.; Schrock, R. R.
Organometallics 1989, 8, 644.
(13) Schlund, R.; Schrock, R. R.; Crowe, W. E. J. Am. Chem. Soc. 1989,
111, 8004.
X Abstract published in AdVance ACS Abstracts, April 15, 1996.
(1) Lytel, R.; Lipscomb, G. F.; Thackara, J.; Altman, J.; Elizondo, P.;
Stiller, M.; Sullivan, B. In Nonlinear Optical and ElectroactiVe Polymers;
Prasad, P. N., Ulrich, D. R., Eds.; Plenum Press: New York, 1987; p 415.
(2) Kobayashi, T.; Hattori, T.; Terasaki, A.; Kurokawa, K. In Nonlinear
Optical and ElectroactiVe Polymers; Prasad, P. N., Ulrich, D. R., Eds.;
Plenum Press: New York, 1987; p 137.
(3) Zyss, J.; Chemla, D. S. In Nonlinear Optical Properties of Organic
Molecules and Crystals; Chemla, D. S., Zyss, J., Eds.; Academic Press:
Orlando, FL, 1987; Vol. 1, p 23.
(14) Fox, H. H.; Schrock, R. R. Organometallics 1992, 11, 2763.
(15) Makio, H.; Masuda, T.; Higashimura, T. Polymer 1993, 34, 2218.
(16) Fox, H. H.; Wolf, M. O.; O’Dell, R.; Lin, B. L.; Schrock, R. R.;
Wrighton, M. S. J. Am. Chem. Soc. 1994, 116, 2827.
(17) Mizumoto, T.; Masuda, T.; Higashimura, T. J. Polym. Sci., Part A:
Polym. Chem. 1993, 31, 2555.
(18) Masuda, T.; Mishima, K.; Fujimori, J.; Nishida, M.; Muramatsu,
H.; Higashimura, T. Macromolecules 1992, 25, 1401.
(19) Masuda, T.; Yoshida, T.; Makio, H.; Rahman, M. Z. A.; Higash-
imura, T. J. Chem. Soc., Chem. Commun. 1991, 503.
(4) Prasad, P. N. Polymer 1991, 32, 1746.
(5) Prasad, P. N.; Williams, D. J. In Introduction to Nonlinear Optical
Effects in Molecules and Polymers; John Wiley and Sons, Inc.: New York,
1991; p 132.
(6) Knoll, K.; Schrock, R. R. J. Am. Chem. Soc. 1989, 111, 7989.
(7) Masuda, T.; Tachimori, H. J. Macromol. Sci., Pure Appl. Chem. 1994,
A31, 1675.
0002-7863/96/1518-3883$12.00/0 © 1996 American Chemical Society

3884 J. Am. Chem. Soc., Vol. 118, No. 16, 1996
Schrock et al.
have been found to be soluble and highly conjugated.8,9,11,20,21
Significant progress also has been made in the synthesis of a
wide variety of molybdenum alkylidene complexes of the type
Mo(NR)(CHR′)(OR′′)2 (and sometimes Mo(NR)(CHR′)(OR′′)2-
(L), where L is a donor ligand),22 and to some extent their
tungsten relatives,23,24 and their use as olefin metathesis catalysts.
One of the most interesting applications so far is their use as
initiators for the living polymerization of strained cyclic olefins
such as norbornenes and substituted norbornadienes.25-28 There-
fore a study of the polymerization of o-TMSPA by a well-
defined alkylidene initiator seemed appropriate, especially in
view of the paucity of hard data concerning the mechanism of
polymerization of terminal alkynes and the paucity of polyenes
prepared by a living polymerization method. We hoped that
head-to-tail low-polydispersity poly(o-TMSPA) could be pre-
pared, that absolute molecular weights could be determined, and
that ultimately we could begin to address questions concerning
the correlation of γ with chain length and structure. It should
be noted that the synthesis of soluble polyenes containing up
to 200 double bonds by the living cyclopolymerization of diethyl
dipropargylmalonate16 has allowed the observation that γ/N,
where N is the number of double bonds, begins to maximize at
approximately 50 double bonds in this particular polymer.29
Therefore an important goal is to prepare other types of polyenes
that contain greater than 15 double bonds and that have a PDI
(Mw/Mn) low enough to be useful for NLO studies.
or â addition took place exclusively.31) However, initial studies
showed that all initiators known in our labs at that time did not
yield low-polydispersity poly(o-TMSPA).30 We then realized
that R addition could be favored if steric interaction between
the growing polymer chain (P) and a bulky R′ group were
severe, and if the metal were less crowded than normal in
complexes of this type. If our initial assumption that a
disubstituted alkylidene would be unreactive were also incorrect,
then it might be possible to polymerize a terminal alkyne via R
addition. This strategy led to the development of initiators that
contain relatively “small” alkoxides, which we report here. A
portion of this work has appeared in the form of a preliminary
communication.32
Results
Synthesis and Characterization of Hexafluoroisopropoxide
Complexes. Addition of 3 equiv of LiOCH(CF3)2 to
Mo(NAd)(CHCMe2Ph)(OTf)2(DME)33 in diethyl ether yielded
the off-white, crystalline, pentane-soluble “ate” complex, Li-
(DME)Mo(NAd)(CHCMe2Ph)[OCH(CF3)2]3 (1a; eq 2). The
Investigations into the possibility of polymerizing o-TMSPA
by well-defined complexes initially focused on pseudo-four-
coordinate molybdenum complexes of the type Mo(NAr)(CHR)-
(OR′′)2 (e.g., Ar ) 2,6-diisopropylphenyl).30 Such species are
stable toward bimolecular decomposition by virtue of the bulky
nature of the Ar group, R (usually CMe3 or CMe2Ph initially),
and OR′′ (e.g., hexafluoro-tert-butoxide).22 At the time, we
believed that â addition of the terminal alkyne (eq 1b) would
be most desirable, as the bulky R′ group thereby would interact
minimally with the bulky ligands in the coordination sphere.
We assumed that for Mo in a crowded coordination sphere the
intermediate molybdacyclobutene would be relatively unstable
compared to the “open” tautomer, the vinyl-substituted alkyli-
dene.22 Also, we felt that the terminal alkylidene ligand that
would result from rearrangement of the molybdacyclobutene
intermediate would be inherently more reactive than the
disubstituted alkylidene that would result from R addition (eq
1a). (It should be noted at this point that, strictly speaking, the
narrowest molecular weight distribution would be possible only
if R and â addition pathways were of equal rate or if either R
reaction is relatively complex if less than 3 equiv of LiOCH-
(CF3)2 is employed, most likely because Mo(NAd)(CHCMe2-
Ph)[OCH(CF3)2]2 is unstable. (No isolable pseudo-four-
coordinate complex of this general type is known in which the
alkoxides are not relatively bulky, e.g., tert-butoxide, 2,6-
disubstituted phenoxides, etc.,22 and high-oxidation-state alkyli-
dene complexes are known in several cases to decompose
bimolecularly by alkylidene coupling to give the olefin.22)
Proton and carbon NMR spectra of 1a show the neophylidene
HR resonance at 12.59 ppm and the CR resonance at 297.9 (JCH
) 119 Hz) ppm, respectively. The relatively small value for
JCH suggests that the neophylidene ligand is syn, i.e., the
CMe2Ph group points toward the nitrogen atom of the adamantyl
imido ligand. There is no evidence for an HR resonance in an
analogous anti complex, which typically is found ∼1 ppm
downfield of HR in the syn complex and has a value for JCH of
130-150 Hz.31 The proton NMR spectrum of 1a at room
temperature also shows three resonances at 6.11, 4.21, and 3.87
ppm for the methine protons in three inequivalent hexafluoro-
isopropoxide ligands.
(23) Schrock, R. R.; DePue, R. T.; Feldman, J.; Yap, K. B.; Yang, D.
C.; Davis, W. M.; Park, L. Y.; DiMare, M.; Schofield, M.; Anhaus, J.;
Walborsky, E.; Evitt, E.; Kru¨ger, C.; Betz, P. Organometallics 1990, 9,
2262.
(24) Wu, Z.; Wheeler, D. R.; Grubbs, R. H. J. Am. Chem. Soc. 1992,
114, 146.
(25) Schrock, R. R. Acc. Chem. Res. 1990, 23, 158.
(26) Schrock, R. R. In Ring-Opening Polymerization; Brunelle, D. J.,
Ed.; Hanser: Munich, 1993; p 129.
(27) Novak, B. M.; Risse, W.; Grubbs, R. H. In Polymer Synthesis
Oxidation Processes; Springer-Verlag: Berlin, 1992; Vol. 102, p 47.
(28) Grubbs, R. H.; Tumas, W. Science 1989, 243, 907.
(29) Samuel, I. D. W.; Ledoux, I.; Dhenaut, C.; Zyss, J.; Fox, H. H.;
Schrock, R. R.; Silbey, R. J. Science 1994, 265, 1070.
(30) Fox, H. H. Ph.D. Thesis, Massachusetts Institute of Technology,
1993.
(31) Oskam, J. H.; Schrock, R. R. J. Am. Chem. Soc. 1993, 115, 11831.
(32) Schrock, R. R.; Luo, S.; Zanetti, N.; Fox, H. H. Organometallics
1994, 13, 3396.
(33) Oskam, J. H.; Fox, H. H.; Yap, K. B.; McConville, D. H.; O’Dell,
R.; Lichtenstein, B. J.; Schrock, R. R. J. Organomet. Chem. 1993, 459,
185.
(20) Masuda, T.; Hasegawa, K.; Higashimura, T. Macromolecules 1974,
7, 728.
(21) Masuda, T.; Sasaki, N.; Higashimura, T. Macromolecules 1975, 8,
717.
(22) Feldman, J.; Schrock, R. R. Prog. Inorg. Chem. 1991, 39, 1.

LiVing Polymerization of Substituted Phenylacetylene
J. Am. Chem. Soc., Vol. 118, No. 16, 1996 3885
Table 2. Selected Bond Distances and Angles in 2a
Distances (Å)
Mo-C(6)
Mo-N(1)
Mo-N(2)
1.880(4)
1.729(3)
2.254(3)
Mo-O(1)
Mo-O(2)
N(1)-C(18)
2.024(3)
2.012(3)
1.450(5)
Angles (deg)
145.5(3)
166.1(3)
100.7(1)
110.7(2)
102.9(2)
94.2(1)
Mo-C(6)-C(7)
Mo-N(1)-C(18)
C(6)-Mo-O(1)
C(6)-Mo-O(2)
C(6)-Mo-N(1)
C(6)-Mo-N(2)
Mo-O(1)-C(17)
N(1)-Mo-O(1)
N(1)-Mo-O(2)
N(1)-Mo-N(2)
N(2)-Mo-O(1)
N(2)-Mo-O(2)
O(1)-Mo-O(2)
Mo-O(2)-C(14)
99.8(1)
145.5(1)
90.9(1)
159.2(1)
79.1(1)
82.0(1)
130.2(3)
131.4(3)
is isolated initially. This compound loses 1 equiv of DME under
dynamic vacuum to afford what we presume to be the “ate”
complex, K(DME)Mo(NAd)(CHCMe3Ph)[OCH(CF3)2]3 (1b).
1b could only be isolated as a yellow oil. Its proton and carbon
NMR spectra are analogous to those of 1a, including an HR
resonance at 12.51 ppm and a CR resonance at 296.6 ppm (JCH
) 119 Hz). The proton NMR spectrum of 1b at 25 °C in C6D6
only shows two resonances for hexafluoroisopropoxide ligands
at 4.07 ppm (two alkoxides) and 6.20 ppm (one alkoxide). We
assume that the structure of 1b is analogous to that of 1a and
that two of the alkoxide methine resonances cannot be resolved
from one another in the proton NMR spectrum.
Figure 1. Chem 3D view of the structure of Mo(NAd)(CHCMe2Ph)-
[OCH(CF3)2]3Li(MeOCH2CH2OMe) (1a).
Table 1. Selected Bond Distances and Angles in 1a
Distances (Å)
Mo-C(1) 1.87(2) Mo-O(2) 2.115(9) Li-O(2) 1.87(3)
Mo-N
1.67(1) Mo-O(3) 2.00(2)
Li-O(4) 1.95(3)
Li-O(5) 1.98(3)
Mo-O(1) 2.11(1) Li-O(1)
1.92(2)
Angles (deg)
Addition of 1 equiv of boron trifluoride etherate to a cold
(-40 °C) solution of 1b in a 2:1 mixture of ether and pentane
produced a white precipitate and a yellow solution. After 2,4-
lutidine was added and the white precipitate (which is assumed
to be KBF3[OCH(CF3)2]) was filtered off, off-white crystalline
Mo(NAd)(CHCMe2Ph)[OCH(CF3)2]2(2,4-lutidine) (2a; eq 3)
could be isolated. All attempts to isolate “Mo(NAd)(CH-
Mo-C(1)-C(21)
Mo-N-C(11)
N-Mo-C(1)
N-Mo-O(1)
N-Mo-O(2)
N-Mo-O(3)
O(1)-Li-O(2)
Mo-O(3)-C(2)
Mo-O(1)-C(7)
152(2)
165(1)
103.3(8)
120.3(6)
105.7(5)
99.1(6)
83(1)
C(1)-Mo-O(1)
C(1)-Mo-O(2)
C(1)-Mo-O(3)
O(1)-Mo-O(2)
O(1)-Mo-O(3)
O(2)-Mo-O(3)
O(4)-Li-O(5)
Mo-O(2)-C(4)
136.2(7)
92.6(6)
97.3(8)
72.8(4)
80.8(4)
150.3(4)
82(1)
136(1)
127.4(8)
124.3(8)
The structure of 1a was established by X-ray crystallography.
(A drawing can be found in Figure 1 and selected bond distances
in Table 1). The structure is not of high quality (see supporting
information), so bond distances and angles are not known to a
high degree of accuracy. The coordination geometry around
Mo can be described as either a distorted trigonal bipyramid or
a square pyramid. In the trigonal bipyramid description, the
two axial ligands are OCH(CF3)2 groups containing O(2) and
O(3), while the adamantylimido nitrogen atom, the neophylidene
R carbon atom (C(1)), and the oxygen of the third OCH(CF3)2
group (O(1)) lie in the equatorial plane. The sum of the bond
angles in the equatorial plane is 360°. The CMe2Ph group points
toward the nitrogen atom of the adamantylimido group, as
suspected on the basis of NMR data. The ModN distance in
1a is 1.67(1) Å, 0.1 Å shorter than the ModN distance (1.767
Å) in syn-Mo(CHCMe3)(N-2,6-C6H3-i-Pr2)[OCMe(CF3)2]2-
(PMe3).34 The short ModN distance in 1a may be attributed
to the better electron-donating ability of the adamantyl group
than the (2,6-diisopropylphenyl)imido group and the presence
of three electron-withdrawing OCH(CF3)2 groups; both are likely
to lead to stronger σ and π bonds in the M-N pseudo triple
bond. The ModC(1) distance (1.87(2) Å) and the ModC(1)s
C(2) angle (152°) are normal.22 Two OCH(CF3)2 groups (one
in the equatorial plane, the other one in the axial position) are
bridging to the lithium atom. Two bridging OCH(CF3)2 groups
and one CH3OCH2CH2OCH3 (DME) produce a distorted
tetrahedral geometry about lithium.
CMe2Ph)[OCH(CF3)2]2” by adding boron trifluoride etherate to
1b have failed. Evidence presented later suggests that Mo-
(NAd)(CHCMe2Ph)[OCH(CF3)2]2 decomposes readily, even in
dilute solution (<1 mM). Compound 2a is prepared most
conveniently simply by reacting Mo(NAd)(CHCMe2Ph)(OTf)2-
(DME) with 2 equiv of KOCH(CF3)2 in the presence of 2 equiv
of 2,4-lutidine. Proton and carbon NMR spectra of 2a (Table
3) show the neophylidene HR resonance at 13.76 ppm and the
CR resonance at 297.8 ppm (JCH ) 121 Hz), consistent with a
syn orientation of the neophylidene ligand. In the presence of
1 equiv of 2,4-lutidine at room temperature, resonances for both
free and bound 2,4-lutidine are observed, consistent with slow
exchange on the NMR time scale between free and bound 2,4-
lutidine. 2a (1-10 mM) reacts rapidly with benzaldehyde to
yield PhMe2CHdCHPh (97% trans) quantitatively.
The results of an X-ray crystallographic study of 2a are shown
in Figure 2 and Table 2. The coordination geometry around
Mo in 2a is similar to what it is in 1a. In the trigonal
bipyramidal description the 2,4-lutidine ligand (N(2)) and one
OCH(CF3)2 group (O(1)) occupy the two “axial” positions with
an N(2)-Mo-O(1) angle of 159.2°, while N(1), C(6), and O(2)
lie in the equatorial plane. (The sum of the three angles in the
equatorial plane is 359°.) The neophylidene ligand is syn, as
When 3 equiv of KOCH(CF3)2 is added to Mo(NAd)(CHCMe2-
Ph)(OTf)2(DME), K(DME)2Mo(NAd)(CHCMe3Ph)[OCH(CF3)2]3
(34) Schrock, R. R.; Crowe, W. E.; Bazan, G. C.; DiMare, M.; O’Regan,
M. B.; Schofield, M. H. Organometallics 1991, 10, 1832.

3886 J. Am. Chem. Soc., Vol. 118, No. 16, 1996
Schrock et al.
JCH
1
Table 3. Selected H and 13C NMR Data of New Compoundsa
compound
rotamer
δ (HR)
δ (CR)
Li(DME)Mo(NAd)(CHCMe2Ph)(OR)3 (1a)
K(DME)Mo(NAd)(CHCMe2Ph)(OR)3 (1b)
Mo(NAd)(CHCMe2Ph)(OR)2(2,4-Me2Py) (2a)
Mo(NAd)(CHCMe2Ph)(OR)2(Py) (2b)
syn
syn
syn
syn
anti
syn
syn
12.59
12.51
13.76
13.76
14.08
13.76
12.81
297.9
296.6
297.8
301.3
119
119
121
121
Mo(NAd)(CHCMe2Ph)(OR)2[2-(3-pentyl)NC5H4] (2c)
Mo(NAd)(CHCMe2Ph)(OR)2(Quin) (2d)
296.7
121
119
Mo(NAd)[C(Ph-o-TMS)(CHCHCMe2Ph)(OR′)2 (3a/b)
Mo(NAd)(CMePh)(OR)2(2,4-Me2Py) (5a)
Mo(NAd)(CPh2)(OR)2(2,4-Me2Py) (5b)
Mo(NAd)[C(Ph)C(Ph)CHCMe2Ph)(OR)2(2,4-Lut) (6)
Mo(N-2,6-C6H4Me2)(CHCMe2Ph)(OC6F5)2(Quin) (7a)
283.6
302.4
310.0
305.1
syn
anti
syn
13.02
14.16
13.80
115
145
127
Mo(N-2,6-C6H4-i-Pr2)(CHCMe2Ph)(OR)2(2,4-Lut) (8)
295.9
a Ad ) 1-adamantyl; OR ) OCH(CF3)2, OR′ ) OCMe(CF3)2; Quin ) quinuclidine; 2,4-Lut ) 2,4-dimethylpyridine.
stabilizes the remaining 2a (usually ∼80% of the amount added)
against further decomposition. Irradiation of the resonance for
the proton in the 3-position of ∼1 equiv of added 2,4-lutidine
(at ∼6.5 ppm) did not alter the resonance for H3 in coordinated
lutidine in 2a (at ∼6.1 ppm).
Base adducts analogous to 2a, Mo(NAd)(CHCMe2Ph)[OCH-
(CF3)2]2(L) (L ) Py (2b), 2-(3-pentyl)NC5H4 (2c), or quinu-
clidine (2d)), also could be prepared by methods analogous to
that employed for preparing 2a. An attempt to synthesize an
adduct containing 2,6-lutidine yielded decomposition products
only, presumably because this base is too bulky to bind strongly
to Mo(NAd)(CHCMe2Ph)[OCH(CF3)2]2. It is also clear that
2,4-lutidine in 2a is readily replaced by a smaller base. For
example, addition of 1 equiv of pyridine to a benzene-d6 solution
of 2a yielded Mo(NAd)(CHCMe2Ph)[OCH(CF3)2]2(Py) (2b)
immediately, along with resonances characteristic of free 2,4-
lutidine. Compounds 2b and 2d are stable both in solution and
as a solid, while 2c is stable in solution for only a short time,
decomposing to a significant extent after 24 h in solution. 2c
also was found to have decomposed to a significant degree in
the solid state after 2 weeks at -40 °C under dinitrogen. We
attribute the instability of 2c to the lability of 2-(3-pentyl)-
pyridine and the instability of Mo(NAd)(CHCMe2Ph)[OCH-
(CF3)2]2.
Figure 2. Chem 3D view of the structure of Mo(NAd)(CHCMe2Ph)-
[OCH(CF3)2]2(2,4-lutidine) (2a).
predicted on the basis of NMR data. The ModN distance in
2a (1.729 Å) is longer than it is in 1a, but the MosC(6) distance
(1.880(4) Å) is approximately the same. The equatorial N(1)-
Mo-O(2) angle in 2a (145.5°) is larger than that (120°) in 1a,
possibly because of steric interaction between the o-Me group
in the 2,4-lutidine ligand and the equatorial AdN and OCH-
(CF3)2 groups. Such an interaction would help account for the
lability of the 2,4-lutidine in 2a (see below).
A closer examination of proton NMR spectra of dilute
solutions (∼0.05-1 mM) of 2a revealed that some free 2,4-
lutidine is always present (∼20%, depending on the concentra-
tion of 2a), and that some weak resonances could be observed
in the alkylidene region and elsewhere. For some time we
entertained the idea that observation of free lutidine was
indicative of an equilibrium between 2a and Mo(NAd)-
(CHCMe2Ph)[OCH(CF3)2]2 and therefore that K1 (k1/k-1 in eq
4) could be measured directly. (An equilibrium constant as large
NMR data for compounds 2b-d are listed in Table 3.
Adducts 2c and 2d exist as syn rotamers according to the small
values for JCH, while 2b is a 9:1 mixture of syn and anti rotamers
in solution. Only at temperatures below -20 °C do the 1H and
13C NMR spectra for 2c show sharp alkylidene proton and
carbon resonances. At temperatures above -20 °C, all reso-
1
nances in the H NMR spectrum of 2c are broad, consistent
with the relatively rapid exchange of bound 2-(3-pentyl)pyridine
in 2c with free 2-(3-pentyl)pyridine, most likely via formation
of Mo(NAd)(CHCMe2Ph)[OCH(CF3)2]2.
Reactions between Alkylidene Complexes and o-TMSPA.
Compounds 1a, 1b, and 2b do not react readily with o-TMSPA,
but 2a, 2c, and 2d do. The product of the reaction of o-TMSPA
with 2a is only that arising from R addition (eq 1a), as is readily
determined from the presence of two olefinic resonances at 8.80
and 4.90 ppm for Hâ and Hγ, respectively,34 in the first insertion
product (eq 5). (The product of â addition would be a terminal
Mo(NAd)(CHCMe2Ph)[OCH(CF3)2]2 + lutidine { k } 2a
1
k-1
(4)
as 105 or 106 could be measured by high-field NMR at
concentrations below ∼1 mM.) We finally came to the
conclusion that loss of 2,4-lutidine from 2a does yield Mo-
(NAd)(CHCMe2Ph)[OCH(CF3)2]2 rapidly on the chemical time
scale, but that even in dilute solution Mo(NAd)(CHCMe2Ph)-
[OCH(CF3)2]2 decomposes readily. The liberated base then
alkylidene with an HR resonance in the usual region.) The CdC
bond in the vinyl group is trans, according to the value for JHâHγ
(∼16 Hz). Only 3 equiv of o-TMSPA is required to consume

LiVing Polymerization of Substituted Phenylacetylene
J. Am. Chem. Soc., Vol. 118, No. 16, 1996 3887
Table 4. GPC Data for Poly(o-TMSPA) Prepared Employing 2a
Initiator 2c also reacts with o-TMSPA via solely R addition
and initiates the relatively fast polymerization of o-TMSPA in
toluene at room temperature compared to the rate of polymer-
ization by initiator 2a. However, a 45-mer obtained after 3 h
of polymerization time shows a bimodal molecular weight
distribution by GPC with Mn ) 565 200 (PDI ) 1.06) and Mn
) 20 020 (PDI ) 1.07), respectively, for the two peaks (by
viscometry; theory ) 7960 absolute molecular weight). Since
the predictable propagating species must be the same type as
shown in eq 5, the faster polymerization initiated by 2c can be
attributed to formation of higher molecular weight polymer by
some more active polymerization catalyst. When the same
polymerization reaction was run for 23 h the Mn values of the
two distributions were found to be 387 200 (PDI ) 1.27) and
17 250 (PDI ) 1.09), respectively. When more than 1 equiv
of 2,4-lutidine was added to a toluene solution of 2c before
addition of 40 equiv of o-TMSPA, a polymer having a unimodal
molecular weight distribution and low PDI was formed (Mn )
17 000 by viscometry; PDI ) 1.08). We also do not observe
bimodal molecular weight distributions and polymer degradation
in reactions employing 2a as the initiator in toluene. We
conclude that 2,4-lutidine prevents some decomposition that
produces a more active catalyst or “poisons” a more active
polymerization catalyst that is formed under the reaction
conditions. A more active catalyst could arise through decom-
position of a fraction of 2c (as a consequence of the high lability
of the base in 2c) or through reactions of 2c with impurities or
water. A small amount of relatively active catalyst would lead
to the formation of high molecular weight polymer and
secondary metathesis (chain degradation). Evidently the de-
composition product or products that are formed when 2a is
employed as an initiator are likewise “deactivated” by 2,4-
lutidine and do not further complicate polymerizations.
a
a
equiv
Mn
PDI
equiv
Mn
PDI
10
20
30
40
50
1 410
2 680
4 400
5 750
6 930
1.07
1.07
1.05
1.05
1.06
60
70
80
90
100
8 810
9 940
11 650
12 500
13 900
1.04
1.05
1.05
1.04
1.04
a By GPC versus polystyrene standards.
2a. Interestingly, the first R addition product (the disubstituted
alkylidene) does not appear to bind 2,4-lutidine to any significant
degree, since 2,4-lutidine resonances are observed essentially
where they are found in free 2,4-lutidine, not shifted toward
where they are found in the lutidine bound in 2a. This proposal
is corroborated by kinetic studies described later. The bulkier
disubstituted alkylidene must block coordination of 2,4-lutidine
and also stabilize the four-coordinate complex against decom-
position.
Addition of excess o-TMSPA to 2a in either THF or toluene
leads to formation of poly(o-TMSPA). Polymerization rates
in THF are approximately half the rates in toluene under the
same experimental conditions, and poly(o-TMSPA) prepared
in THF generally had a higher PDI (1.10-1.50) compared to
that prepared in toluene (PDI ) 1.04-1.07). Therefore, in all
subsequent polymerizations we employed toluene as the solvent.
Table 4 lists the number average molecular weight (by GPC
versus polystyrene) of poly(o-TMSPA) samples prepared in
toluene using initiator 2a. In initial studies benzaldehyde was
added as the “quenching” agent, and the polymer was precipi-
tated with methanol, collected, and dried in vacuo. (In ring
opening metathesis polymerization reactions employing initiators
of the type described here and norbornadienes or norbornenes
as the monomers, benzaldehyde is typically employed to cap
the polymer in a Wittig-like reaction with the alkylidene.26) The
polymers were obtained in >85% yields. A plot of the number
average molecular weights (Mn) vs N (the number of equivalents
of monomer added) shows a linear relationship (Figure 3),
consistent with a living polymerization. Polymers prepared
simply by precipitation with methanol were essentially identical
to those isolated from reactions that had been quenched with
benzaldehyde.
Evidence for quenching of the propagating alkylidene can
be obtained indirectly by NMR by observing consumption of
o-TMSPA. Consumption of o-TMSPA ceases ∼20 min after
addition of ∼100 equiv of benzaldehyde at ∼20 °C, presumably
as a consequence of conversion of propagating alkylidene
complexes to analogous oxo species via a Wittig-like reaction.
If only a few equivalents of benzaldehyde is added, then at the
low concentrations of the catalyst in a typical polymerization
reaction, the capping reaction could be incomplete, even after
several hours. These results contrast with the rapid reaction of
2a with benzaldehyde, but are consistent with what we suspect
to be a much lower reactivity of a disubstituted alkylidene in
general (i.e., toward benzaldehyde or o-TMSPA).
The rate of polymerization of o-TMSPA by initiator 2a could
be followed by proton NMR, although we found gas chroma-
tography to be more accurate and reliable. Plots of ln(C/C0)
versus time (after initiator was consumed) in the presence of
80 equiv of o-TMSPA were found to be linear through more
than 3 half-lives, and the observed rate was found to depend to
the first power upon the total initiator present ([2a]0). (See Table
9a for individual values.) If we assume that the rate of loss of
base from a base adduct (k-2[Catn(B)]; eq 6) is larger than
k2a[Catn][M] (M ) monomer) (eq 7), and [Catn] ≈ [2a]0, the
rate of disappearance of monomer is approximately equal to
k2a[2a]0[M]. From 13 runs (Table 9a) we get a value of k2a
)
k2
Catn + B { } Catn(B)
(6)
k-2
k2a
98 Catn+1
Catn + M
(7)
0.30 with a range of 0.24-0.36 s-1 M-1 (Table 5). Other
assumptions are the following: (i) the reaction of M with Catn
is irreversible; (ii) the reactivities of Catn and Catn+1 alkylidene
intermediates are identical; and (iii) there is only one mechanism
or mode of chain growth. (The possible involvement of
rotamers will be considered later.)
When a significant amount of base (B ) 2,4-lutidine) is added
(10-75 equiv) the rate of disappearance of o-TMSPA is again
found to be cleanly first order, but the rate in the presence of
∼75 equiv of added base slows to ∼25% of its “base-free”
value. If we make the same assumptions as we did above, then
the rate is that shown in eq 8 (where K2a ) k2/k-2). Therefore
a plot of [2a]0/kobs versus [2,4-lutidine] ([B]) should produce a
An alternative method of quenching polymerization is to add
8 equiv of acetic acid. The amount of Mo in a reprecipitated
poly(o-TMSPA)20 sample was found to be only 14% of the
amount of catalyst 2a that had been employed by elemental
analysis for Mo. Therefore in some studies polymerizations
were terminated by addition of 8 equiv of degassed acetic acid
to the reaction solution and stirring of the mixture at room
temperature for 24 h. The polymers obtained after exposure to
acetic acid for 24 h in the absence of oxygen had UV/vis spectra
that were similar to those of samples that had been allowed to
isomerize thermally in the absence of oxygen, as discussed in
more detail in a later section.

3888 J. Am. Chem. Soc., Vol. 118, No. 16, 1996
Schrock et al.
Figure 3. Plot of molecular weight (by GPC vs polystyrene) for poly-
(o-TMSPA) prepared from N equiv of o-TMSPA and 2a (O) or 7 ([)
as the initiator in toluene.
Table 5. Summary of Observed Rate and Equilibrium Constantsa
constant
value
no. of expts
range
k2a
0.30 s-1 M-1
62 M-1
13
5
1
2
1
1
4
4
0.24-0.36
K2a
K2d
k3bf3a
k3a
k3b
k7
(see Figure 4a)
2810 M-1
3.1 × 10-4 s-1
1.2 × 10-3 s-1 M-1
6.3 × 10-3 s-1 M-1
0.23 s-1 M-1
1380 M-1
3.01, 3.19
(see Figure 4b)
(see Figure 4b)
K7
a See Experimental Section for individual values.
straight line with intercept 1/k2a and slope K2a/k2a (eq 9). From
Figure 4. (a) Plot of [2a]0/kobs versus [2,4-lutidine] for polymerization
of o-TMSPA; determination of k2a and K2a. (b) Plot of [7]0/kobs versus
[quinuclidine] for polymerization of o-TMSPA; determination of k7 and
K7.
k2a[2a]0[M]
1 + K2a[B]
d[M]
dt
) -
) -kobs[M]
(8)
(9)
[2a]0
K
1
)
+
2a[B]
K1 (the equilibrium constant for binding 2,4-lutidine to the
initiator), we cannot determine ki.
kobs
k2a k2a
Compound 2d polymerizes o-TMSPA in a manner analogous
to that of 2a. Since the rate constant for propagation (k2a) is
known, K2d can be measured as a consequence of a single rate
measurement using eq 9. K2d was found to be 2810 M-1. The
larger value for K2d versus K2a (62 M-1) is consistent with the
greater basicity of quinuclidine versus 2,4-lutidine in forming
adducts of this type. K2d is also approximately twice the size
of the binding constant in the case of Mo(N-2,6-Me2C6H3)-
(CHCMe2Ph)(OC6F5)2(quinuclidine) (see later).
5 runs (see Table 9b for data and details) we obtained a value
of k2a ) 0.32 s-1 M-1 and K2a ) 62 M-1 (Figure 4a). These
results are consistent with our initial assumptions above, since
if K2a ) 62 M-1, then at a [2a]0 of 1 mM ∼95% of the metal
is in the base-free form. Therefore K2a[B] , 1 in the absence
of added base and the rate of disappearance of monomer equals
k2a[2a]0[M]. In the presence of 17 equiv of added base, for
example ([2a]0 ) 10-3 M), K2a[B] is ∼1 and the rate is
approximately halved.
We can now estimate a value for k-2 (at [2a]0 ) 10-3 M and
[M] ) 0.08 M) since we know that k-2[Catn(B)] should be at
least 10 times k2a[Catn][M]. In order for this to be true, k-2
must be ∼5 s-1 or greater. Resonances are broad in the room
temperature proton and carbon spectra of 2c, a result that
suggests that the rate of loss of 2-(3-pentyl)pyridine from 2c is
approaching the NMR time scale (50-100 s-1). Loss of 2-(3-
pentyl)pyridine from 2c should be faster than loss of 2,4-lutidine
from 2a for steric reasons, so k-2 should be much less than
50-100 s-1. Therefore a value for k-2 of ∼5 s-1 would seem
to be approximately correct.
Isolation of a “First Insertion Product” and Observation
of a “Second Insertion Product” in the Reaction between
Mo(NAd)(CHCMe2Ph)[OCMe(CF3)2]2 and o-TMSPA. Al-
though the “first insertion product” of the reaction of 2a with
o-TMSPA could be observed by proton NMR, all attempts to
isolate it failed, probably because it reacts with additional
o-TMSPA too readily to give mixtures of oligomers. However,
addition of 1 equiv of o-TMSPA to Mo(NAd)(CHCMe2Ph)-
33
[OCMe(CF3)2]2 in pentane at -40 °C leads to a mixture of
two “first insertion products” (3a and 3b) in a variable ratio
(usually approximately 9:1) in favor of 3a. Proton NMR spectra
of the mixture contain two sets of doublets, one set at 7.34 and
5.78 ppm (JHH ) 15.6 Hz, ) 1.56 ppm between resonances)
for 3a, and the other set at 8.44 and 4.30 ppm (JHH ) 15.6 Hz,
) 4.14 ppm between resonances) for 3b. The JHH values
are consistent with a trans CdC bond in each compound. When
o-TMS(C6H4)CtCD was employed, the resonances at 7.34 (3a)
and 8.44 (3b) were absent, indicating that the acetylene proton
from o-TMSPA is found in the â position in the disubstituted
It would be useful to determine the rate at which M reacts
with “Mo(NAd)(CHCMe2Ph)[OCH(CF3)2]2”, formed when 2a
loses 2,4-lutidine, in order to compare it with the rate of
propagation. Unfortunately, the consumption of 2a is simply
too fast at 25 °C to measure using NMR techniques under
pseudo-first-order conditions; for example, initiation is complete
in 1-2 min when [M] ≈ [B] ≈ 1 M. Since we cannot determine

LiVing Polymerization of Substituted Phenylacetylene
J. Am. Chem. Soc., Vol. 118, No. 16, 1996 3889
vinylalkylidene ligand, as expected. Carbon NMR spectra of
the 3a/3b mixture contain a resonance at 283.6 ppm for the
alkylidene carbon of the major product (3a), but the alkylidene
resonance for 3b could not be observed with certainty as a
consequence of its concentration being too low. In 3a the
resonances for Câ and Cγ are at found at 126.3 (JCH ) 150 Hz)
and 122.8 ppm (JCH ) 150 Hz), respectively, consistent with
the proposal that this species is a vinyl-substituted alkylidene
complex.13,34 We propose that 3a and 3b are the two rotamers
shown in eq 10.
conformation found in the OCH(CF3)2 system. The OCMe-
(CF3)2 system would clearly be the more crowded, in general,
whichever rotamer is considered. We assume that greater steric
hindrance in general in the OCMe(CF3)2 system is what prevents
the ready polymerization of o-TMSPA at 25 °C.
A mixture of 3a and 3b reacts slowly with 1 equiv of
o-TMSPA to yield two “second insertion” products, as indicated
by olefinic doublets in the proton NMR spectrum at 5.99 and
5.23 ppm (JHH ) 15.6 Hz) along with a singlet at 8.87 ppm,
and a second set of doublets at 6.11 and 5.27 ppm (JHH ) 15.6
Hz) along with a singlet at 8.12 ppm. When the mixture was
heated to 70 °C, the intensities of the resonances ascribed to
the two second insertion products increased at the expense of
those for 3a and 3b. It is proposed that the second insertion
products are also a mixture of two rotamers in solution (eq 11).
The mixture of 3a and 3b can be isolated in ∼60% yield as
deep red microcrystals. NMR spectra of the isolated material
are identical in all respects to spectra observed upon treating
Mo(NAd)(CHCMe2Ph)[OCMe(CF3)2]2 with o-TMSPA. How-
ever, we noticed that the ratio of 3a to 3b depended upon the
history of the sample and changed with time to an equilibrium
value of 5-7 at 24 °C in favor of 3a. When the reaction time
is shortened to 5 min, and the reaction solution is kept below
∼0 °C during workup, the product is an approximately 1:1
mixture of 3a and 3b. In solvents such as C6D6 and over a
period of ∼1 day at room temperature, the 1:1 mixture returned
to the equilibrium value. The conversion of 3b to equilibrium
The singlet resonances at 8.87 and 8.12 ppm are assigned the
Hâ protons, while the doublets are ascribed to the Hδ and Hꢀ
protons. The rate of disappearance of 3a and 3b at 23 °C was
found (by NMR) to be first order in the presence of ∼25 equiv
of o-TMSPA. The major rotamer reacted with a rate constant
k3a ) 1.2 × 10-3 s-1 M-1 (R2 ) 0.995) while for the minor
rotamer k3b ) 6.3 × 10-3 s-1 M-1 (R2 ) 0.983). The value
for k3b can be compared with the value for k2a in the catalytic
reaction initiated by 2a (k2a ) 0.30 s-1 M-1). The rate of
reaction of what appear to be analogous rotamers with o-
TMSPA in the OCH(CF3)2 system (2a) is ∼50 times faster than
in the OCMe(CF3)2 system, a clear demonstration of the relative
steric influence of OCH(CF3)2 versus OCMe(CF3)2.
Synthesis of Other Initiators Containing OCH(CF3)2
Ligands. Two disubstituted alkylidene complexes, Mo(NAd)-
(CMePh)[OCH(CF3)2]2(2,4-lutidine) (5a) and Mo(NAd)(CPh2)-
[OCH(CF3)2]2(2,4-lutidine) (5b), could be synthesized by
treating 2a with R-methylstyrene or 1,1′-diphenylethylene,
respectively (eq 12). Both 5a and 5b are bright yellow
microcrystalline solids that are soluble in benzene, toluene, and
ether and slightly soluble in pentane. The 13C chemical shift
values for the disubstituted alkylidene carbon atom are 302 ppm
for 5a and 310 ppm for 5b (see also Table 3). Only a single
rotamer is observed for 5a.
was found to be first order with values at 24 °C of k3bf3a
)
3.01 × 10-4 s-1 and K3bf3a ) 7.3.
The rotamer proposal is supported by the fact that irradiation
of the 3a/3b mixture with a medium-pressure mercury lamp at
-60 °C increases the amount of 3b to approximately 25% of
the total after 24-48 h, a phenomenon that has been observed
for rotamers of other alkylidene complexes in this general
class.31 The rate at which equilibrium was reestablished was
followed at 15 °C and found to be first order through 3 half-
lives with k3bf3a ) 1.05 × 10-4 s-1 (K3bf3a ) 6.3 at 15 °C).
At 24 °C k3bf3a ) 3.19 × 10-4 s-1 (K3bf3a ) 5.5), in agreement
with the result quoted above. These rate constants should be
compared to that for conversion of the disfavored anti rotamer
of Mo(NAd)(CHCMe2Ph)[OCMe(CF3)2]2 to the syn rotamer in
toluene at -31.4 °C (5.75 × 10-4 s-1).31 At 15 °C the latter
should be of the order of ∼10-2 s-1; i.e., the rate of conversion
of rotamers of 3b to 3a is approximately 2 orders of magnitude
slower. It is important to note that the reaction between
o-TMSPA and 2a yields only one first insertion product, with
olefinic resonances at 4.90 and 8.80 ppm ( ) 3.90 ppm),
chemical shifts and a value that would make it most analogous
to the minor product (3b) in the 3a/3b mixture (4.50 and 8.44
ppm, ) 4.14 ppm). The other rotamer of the first insertion
product of 2a probably is not formed readily, unless the rotamer
isomerization rate is extremely fast. This seems unlikely in view
of the slow rate of rotamer isomerization in the OCMe(CF3)2
system. Therefore we can have some confidence that only one
rotamer is formed in each insertion step in the polymerization
of o-TMSPA by 2a and that the chain grows in only one way.
It is not clear which rotamer is which in either the OCH(CF3)2
or the OCMe(CF3)2 system. However, the bulkier OCMe(CF3)2
ligands appear to destabilize the sterically more favorable
Diphenylacetylene reacts with 2a to give an isolable “first
insertion product”, Mo(NAd)[C(Ph)C(Ph)CHCMe2Ph][OCH-
(CF3)2]2(2,4-lutidine) (6) as a single rotamer. The 13C chemical
shift value for the disubstituted alkylidene carbon atom is 305

3890 J. Am. Chem. Soc., Vol. 118, No. 16, 1996
Schrock et al.
Table 6. Results of Initiation Reactions between 2a and
o-RC6H4CtCH (R ) H, Me, i-Pr, t-Bu)
for λmax in poly(o-i-PrPA) and poly(o-MePA) are markedly and
progressively more blue-shifted in chains of comparable length
versus λmax in poly(o-TMSPA). For a 100-mer, λmax for poly-
(o-TMSPA) is ∼538 nm (see also below), for poly(o-i-PrPA)
it is 456 nm, and for poly(o-MePA) it is 444 nm. The marked
decrease in λmax could be ascribed to progressively less selective
R addition and consequently a less regular structure and less
conjugation in the polymer backbone.
initiator
R
R (%)
â (%)
2a
2a
2a
2a
t-Bu
i-Pr
Me
H
100
73
60
27
40
44
56
ppm (see also Table 3). More diphenylacetylene does not react
readily with 6 at 25 °C.
Initiators That Contain Other Imido or Alkoxide Ligands.
It would be desirable to establish whether initiators that contain
other “small alkoxides” can be prepared and whether they also
will polymerize o-TMSPA smoothly. Another important ques-
tion is whether the adamantylimido ligand is required. Ada-
mantylimido complexes33 have not been explored to any
significant extent as catalysts for more standard polymerization
reactions of norbornenes or substituted norbornadienes, but in
the one instance where an adamantylimido catalyst has been
explored in some detail, it was found to behave in a significantly
different manner than the analogous arylimido catalysts.31
Proton NMR studies in benzene-d6 show that o-TMSPA reacts
with 5a, 5b, and 6 solely via R addition, and all three compounds
serve as initiators for the polymerization of o-TMSPA in toluene.
Poly(o-TMSPA)150 prepared with initiator 5a had PDI ) 1.08;
poly(o-TMSPA)40 prepared with initiator 5b had PDI ) 1.04;
and poly(o-TMSPA)80 prepared with initiator 6 had PDI ) 1.12.
These results lend further support to the proposal that disub-
stituted (vinyl) alkylidenes are the reactive intermediates in the
polymerization of o-TMSPA by 2a.
Reactions between Alkylidene Complexes and Other
Ortho-Substituted Phenylacetylenes (R ) H, Me, i-Pr, t-Bu).
Table 6 lists the results of reactions between 2a and o-
RC6H4CtCH (R ) H, Me, i-Pr, t-Bu). Addition of only 2 equiv
of o-t-BuPA is necessary in order to consume all 2a. Phenyl-
acetylenes that contain progressively smaller ortho substituents
react with 2a to give progressively more â addition in the first
step. If we assume that all alkynes react in fundamentally the
same way, most likely at the CNO face of the pseudo-four-
coordinate initiator,26 we can attribute an increasing amount of
â addition to a decreased steric interaction between the
neophylidene ligand and the ortho-substituted phenyl ring.
The “smaller” phenylacetylenes also react with disubstituted
alkylidenes by both R and â addition to varying degrees. Both
o-t-BuPA and o-TMSPA add to the ModCMePh bond in 5a
100% via R addition. However, o-MePA and o-i-PrPA react
with 5a by both R and â addition, with the amount of â addition
being ∼7% of the total initial concentration of 5a in each case.
(The actual ratio of R vs â first insertion products could not be
determined because of the complexity of the proton NMR
spectra.) Similar studies employing 6 show that o-MePA, o-i-
PrPA and o-TMSPA all add to the ModC(Ph)C(Ph)CHCMe2-
Ph bond 100% via R addition, but phenylacetylene itself reacts
via both R (90%) and â addition. The steric demands of the
disubstituted alkylidene ligand in 6 clearly cause R insertion to
be much more favored. The above data suggest that during
polymerization of o-i-PrPA, o-MePA, and especially phenyl-
acetylene itself, some â addition of the monomer to one or both
rotamers of the propagating ModC(o-RC6H4)(P) species is
likely. Therefore there is a significant possibility that these
polyenes are not completely regular (head to tail), at least when
prepared by the OCH(CF3)2 catalysts, and that the regiospeci-
ficity declines as the size of the R group in the ortho position
decreases.
Addition of 2 equiv of KOC6F5 to Mo(CHCMe2Ph)(NAr′)-
(triflate)2(1,2-dimethoxyethane) (NAr′ ) N-2,6-Me2C6H3) in
1,2-dimethoxyethane followed by 1 equiv of quinuclidine yields
a complex with the composition Mo(CHCMe2Ph)(NAr′)(OC6F5)2-
(quinuclidine) (7) as a mixture of anti and syn isomers having
alkylidene proton resonances at 14.16 and 13.02 ppm, respec-
tively. Initially the ratio of syn to anti was typically 95:5, after
several days ∼75:25, and after 2 weeks ∼55:45. This behavior
is analogous to formation of a syn rotamer of Mo(NAr)(CH-t-
Bu)[OCMe(CF3)2]2(PMe3) first and with time the more stable
anti rotamer (exclusively, in this case, within a few days).34
Such behavior is indicative of a larger binding constant for a
base to the anti form than to the syn form, and it suggests also
a greater inherent reactivity of the anti form than the syn form
toward olefins and acetylenes. (Greater reactivity of Mo(NAr)-
(CH-t-Bu)[OCMe(CF3)2]2 toward several norbornadienes has
been confirmed.31)
When excess o-TMSPA is added to a syn/anti mixture of 7,
a smooth first-order consumption of each initiator can be
observed (unlike the rapid consumption of initiator 2a), but no
insertion products were evident (as they were in the case of
2a). If we assume that loss of quinuclidine from either the syn
or anti form of 7 is fast relative to reaction of Mo(NAd)-
(CHCMe2Ph)(OCH(CF3)2)2 with o-TMSPA, then these data are
consistent with a larger binding constant for quinuclidine in 7
compared to 2,4-lutidine in 2a. The observed rate constant for
disappearance of 7syn under pseudo-first-order conditions was
found to be 7.61 × 10-4 s-1 (R2 ) 0.985), while the observed
rate constant for disappearance of 7anti was found to be 1.13 ×
10-4 s-1 (R2 ) 0.930). A more rapid consumption of 7syn (by
a factor of ∼7) is again consistent with the larger binding
constant for quinuclidine in 7anti, assuming that anti-Mo(CHCMe2-
Ph)(NAr′)(OC6F5)2 would be the most reactive inherently.
Compound 2a serves as an initiator for the polymerization
of o-MePA and o-i-PrPA under conditions used for the
polymerization of o-TMSPA. Poly(o-MePA) and poly(o-i-
PrPA) were isolated as red powders that appear to be analogous
to polymers reported in the literature.7 GPC and UV/vis data
for these polymers are presented in Table 7. The PDI values
are larger than those for poly(o-TMSPA). The polydispersity
of poly(o-i-PrPA) is smaller than that for poly(o-MePA),
possibly as a consequence of a higher R selectivity in the
propagation step, although different rotamers with different
reactivities might also be accessible and could contribute to a
broadening of the molecular weight distribution. The values
Polymerizations of o-TMSPA in toluene using 7 (largely the
syn rotamer) as an initiator proceed smoothly to give low
polydispersity poly(o-TMSPA) that is essentially identical to
poly(o-TMSPA) prepared employing 2a (Table 4). In this case
also the relationship between Mn and the number of equivalents
of o-TMSPA employed is linear (Figure 3). If we assume that
the polymerization of o-TMSPA initiated by 7 can be described
in a manner similar to the reaction initiated by 2a (eqs 6 and
7), and that loss of quinuclidine from the propagating species
is faster than chain growth, then eqs 13 and 14 (cf. 8 and 9) are
valid. (K7 is the binding constant and k7 the propagating rate

LiVing Polymerization of Substituted Phenylacetylene
J. Am. Chem. Soc., Vol. 118, No. 16, 1996 3891
Table 7. GPC and UV-Visible Data of Poly(o-MePA) and
Poly(o-i-PrPA)
k7[7]0[M]
1 + K7[B]
d[M]
dt
) -
) -kobs[M]
(13)
(14)
poly(o-MePA)
poly(o-i-PrPA)
equiv
Mn
PDI
λmax (nm)
Mn
PDI
λmax (nm)
[7]0
K
1
)
+
7[B]
20
30
40
60
70
80
90
100
3 970 1.16
4 550 1.17
5 760 1.16
7 260 1.24
8 580 1.30
10 160 1.24
11 950 1.35
13 570 1.25
426
434
436
440
442
442
440
444
4 790 1.09
5 990 1.09
7 610 1.10
11 020 1.15
13 700 1.16
14 160 1.09
14 900 1.09
16 320 1.09
444
450
454
454
456
456
456
456
kobs k7 k7
constant.) The rate of the consumption of monomer was
followed in the presence of 9-33 equiv of quinuclidine (four
runs; see Table 10), and the resulting values for [7]0/kobs were
plotted against the quinuclidine concentration (Figure 4b). From
the slope and intercept we obtain the values k7 ) 0.23 s-1 M-1
and K7 ) 1380 M-1. The value for k7 is approximately the
same as that for k2a, but K7 is ∼22 times larger than K2a.
Therefore at a concentration of 1 mM ∼50% of the quinuclidine
is bound to the metal, but in the presence of 30 equiv of added
quinuclidine ∼98% of the quinuclidine is bound to the metal,
leading to a significantly slower rate of polymerization in the
absence of added quinuclidine and a sharp decrease in the rate
when quinuclidine is added. Therefore the K7[B] term in the
denominator of eq 13 cannot be ignored at catalyst concentra-
tions of ∼1 mM, even in the absence of added quinuclidine.
Mo(NAr)(CHCMe2Ph)[OCH(CF3)2]2(2,4-lutidine) (8) has also
been prepared from Mo(CHCMe2Ph)(NAr)(OTf)2(DME), KOCH-
(CF3)2, and 2,4-lutidine in ether. (Proton and carbon NMR
spectral data are listed in Table 3.) Only one rotamer (syn; JC-H
) 127 Hz) is readily observable at room temperature. Proton
NMR spectroscopic studies show that 8 reacts with o-TMSPA
100% via R addition. Approximately 3.5 equiv of o-TMSPA
is required to consume all 8. We did not observe the first
insertion product by proton NMR. Poly(o-TMSPA)80 prepared
with initiator 8 is also a purple solid. However, the GPC trace
of the polymer contains two peaks with Mn ) 2.26 × 105 (PDI
) 1.46) and Mn ) 4.47 × 104 (PDI ) 1.14). The bimodal
molecular weight distribution in poly(o-TMSPA) prepared by
catalyst 8 is similar to that observed in poly(o-TMSPA) prepared
by catalyst 2c. The origin of the bimodal distribution is
unknown, and this initiator was not studied further.
a By GPC versus polystyrene standards.
Table 8. λmax of Poly(o-TMSPA)
N
as isolated (no acid)
after 5-6 days
after air
10
20
30
40
50
60
70
80
90
100
526
470, 600
464
512
526
530
534
538
534
536
538
538
526, ∼660 (sh)a
534, ∼660 (sh)
534, ∼660 (sh)
530, ∼660 (sh)
534, ∼660 (sh)
538, ∼660 (sh)
538, ∼660 (sh)
538, ∼ 680 (sh)
538, ∼680 (sh)
505, ∼660
520, ∼670
534, ∼670 (sh)
530, ∼670 (sh)
530, ∼670 (sh)
536, ∼670 (sh)
540, ∼700 (sh)
538, ∼700 (sh)
538, ∼700 (sh)
a Shoulder.
All attempts to prepare a 2,4-lutidine adduct of Mo(NAr′)-
(CHCMe2Ph)(OC6F5)2 failed, suggesting that 2,4-lutidine does
not bind strongly enough to form a stable adduct. These results
are consistent with the finding that K7 ) 1380 M-1 while K2d
is approximately twice that (2814 M-1); i.e., the metal binds
quinuclidine more strongly when OCH(CF3)2 ligands are present
than when OC6F5 ligands are present.
UV/Vis Spectroscopic Properties of Poly(o-TMSPA).
Samples of poly(o-TMSPA) prepared here are initially purple
with a single broad absorption centered at 526 nm for the 10-
mer out to 538 nm for the 100-mer in their UV-vis spectra
(Table 8). If a sample in THF is not exposed to air, the spectrum
changes over a period of several days at 22 °C to one that has
two absorptions, one shifted significantly to lower energy (600-
700 nm) and one shifted to higher energy, as shown for the
10-, 20-, and 30-mers in Figures 5, 6, and 7. The approximate
area fraction of the low-energy absorption decreases with chain
length, but the λmax for the low-energy absorption increases with
chain length to ∼680 nm for the 30-mer. (The location of the
low-energy absorption in the long polymers is difficult to
determine accurately because the intensity is low.) Upon the
exposure of samples in THF to air, the low-energy absorption
disappears immediately and λmax for the remaining absorption
is approximately 538 nm for the longest polymers. However,
the low-energy absorption is least noticeable for the longer
polymers. For example, the main absorption in the 100-mer is
at 538 nm with a shoulder at 680 nm. With time that shoulder
shifts slightly to approximately 700 nm, but the main absorption
Figure 5. UV-vis spectra of (o-TMSPA)10 in THF in the absence of
air.
is still at 538 nm. After exposure to air, only the 538 nm
absorption remains. Therefore only samples containing less than
80 equiv of monomer show the red-shifted absorption to a
significant degree, and only in the absence of air. The UV-
vis spectra of samples that were isolated by treatment with acetic
acid in the absence of air are approximately the same as those
after several days in solution in the absence of air. The
formation of polymers having a low-energy absorption also
seems to depend upon solvent. The low-energy absorption is
not observed in toluene in the absence of air over a period of
several days, although a sample that has been “isomerized” in
THF that is then isolated by precipitation with methanol has

3892 J. Am. Chem. Soc., Vol. 118, No. 16, 1996
Schrock et al.
isomerization is facile. It is difficult to say what the percentage
of all-trans sequences is since the molar extinction coefficient
is likely to be larger for “trans” than for “cis” forms. It is
natural that these “low band gap” “trans” polyene sequences
will be significantly more air sensitive than “cis” polyene
sequences. It is important to note that only polyenes containing
fewer than approximately 50 equiv are obviously air sensitive
and therefore that poly(o-TMSPA) prepared with classical
catalysts,7 which has a relatively high molecular weight, would
not appear to be air sensitive.
Discussion and Conclusions
One of the most important features of the work described
here is that disubstituted alkylidene complexes are the propagat-
ing species in a controlled polymerization of o-TMSPA and
that they form exclusively only when interaction of the
o-(trimethylsilyl)phenyl group with the coordination sphere
around the metal is quantitatively reduced compared to interac-
tion of the o-(trimethylsilyl)phenyl group with the disubstituted
alkylidene. Disubstituted alkylidenes of the type observed here
are significantly less reactive than terminal alkylidenes, but they
still react with a terminal acetylene readily at 25 °C.
A
disubstituted alkylidene complex is also very much more stable
than a terminal alkylidene complex toward bimolecular alkyli-
dene coupling reactions and, therefore, possibly could be
employed in reactions at higher temperatures, or in the presence
of certain functionalities (e.g., a ketone or internal olefin). It
seems likely that successful polymerization of o-TMSPA by
classical alkylidene catalysts7 also depends on selective R
addition of o-TMSPA to a disubstituted alkylidene and on the
stability of disubstituted alkylidene intermediates in general.
A second noteworthy feature of the results reported here is
that base adducts of otherwise unstable alkylidene complexes
can be used successfully as initiators. The success of this
approach depends upon the special circumstances surrounding
polymerization of o-TMSPA with initiator 2a or 7, namely, that
stable, base-free, propagating species are formed, and upon
choosing a base that is labile enough to be lost relatively rapidly
to give the base-free initiator, but not one that is so weakly
bound that catalyst decomposition or side reactions become
significant problems during initiator preparation or polymeri-
zation reactions. Up to this point we have been concerned
largely with designing stable four-coordinate catalysts containing
a primary alkylidene and understanding rotamer interconversion
and reactivity. The stabilization of an initiator by coordination
of base that is reported here should be compared to stabilization
of a propagating species by coordination of base, where the
rate of propagation is thereby reduced relative to the rate of
initiation.13,35 Designing appropriate base adducts for a given
polymerization reaction will likely become an important aspect
of controlled polymerization by well-defined catalysts in the
future.
Figure 6. UV-vis spectra of (o-TMSPA)20 in THF in the absence of
air.
Figure 7. UV-vis spectra of (o-TMSPA)30 in THF in the absence of
air.
Third, we were surprised to find that polymers prepared from
ortho-substituted phenylacetylenes in which the ortho substituent
was not TMS or tert-butyl are almost certainly not regular head-
to-tail polymers, largely because steric repulsion between the
propagating disubstituted alkylidene and the ortho-substituted
phenyl ring of the incoming alkyne is not significant enough to
make the alkyne add R exclusively. At this point we do not
know whether such “mistakes” alone are responsible for the
lower conjugation of polyenes prepared from “smaller” phen-
ylacetylenes, but that is certainly a good possibility. We cannot
comment upon the nature of the poly(phenylacetylenes) made
with catalysts that operate by mechanisms other than the
the same spectrum in toluene as it did in THF. The final
spectrum appears to be an equilibrium state.
Our working hypothesis is that the low-energy absorption can
be ascribed to highly conjugated sequences of poly(o-TMSPA),
and these sequences are not produced initially in the polymer-
ization reaction. The most plausible explanation is that the
highly conjugated form is all-trans and that cis/trans isomer-
ization can take place thermally, or rapidly if catalyzed by acetic
acid. The “all-trans sequence” can of course involve the entire
chain, if the chain length is such that the all-trans form is
significantly more stable than other forms, and if cis/trans
(35) Wu, A.; Grubbs, R. H. Macromolecules 1995, 28, 3502.

LiVing Polymerization of Substituted Phenylacetylene
J. Am. Chem. Soc., Vol. 118, No. 16, 1996 3893
Ph), 6.11 (sept, OCH(CF3)2, JFH ) 7.3), 4.21 (sept, OCH(CF3)2, JFH
)
alkylidene mechanism. We hope in the future to be able to
prepare other phenylacetylenes that we can confirm are exclu-
sively head-to-tail polymers in order to begin to understand the
structure of such polyenes in more detail, how λmax correlates
with polyene structure, and the extent to which the ortho
substituent controls the structure of exclusively head-to-tail poly-
(phenylacetylenes).
Finally, it is worth noting that poly(o-TMSPA) containing
only ∼20 monomers (on the average) isomerizes thermally to
a significant extent to give (presumably all-trans) forms having
λmax between 600 and 700 nm. These forms have not been
observed for poly(o-TMSPA) prepared with classical catalysts,7
either because the molecular weight of such polymers is
generally too high or because such polymers were exposed to
air, or both. Isomerization of polyene oligomers (<15 double
bonds)6 or substituted polyenes prepared by ring opening of
monosubstituted cyclooctatetraenes10 to trans-rich or all-trans
forms has been observed. In neither case does the red-shifted
form have λmax > 600 nm. At this point it is not known whether
the low-band-gap forms of poly(o-TMSPA) that we observe here
are unique to poly(o-TMSPA), or whether they can be observed
in other regular head-to-tail polyenes in this general class when
∼20 double bonds are present.
6.6), 3.87 (sept, OCH(CF3)2, JFH ) 6.2), 2.97 (s, 6, OCH3), 2.73 (s, 4,
OCH2), 2.19 (br, 6, CH2), 1.94 (br, 3, CH), 1.81 (s, 3, CMe2Ph), 1.73
(s, 3, CMe2Ph), 1.48 (q, 6, CH2); 13C NMR (C6D6) δ 297.87 (CHCMe2-
Ph, JCH ) 119), 148.61 (Cipso), 129.08 (Ph), 125.92 (Ph), 126.73 (Ph),
84.72 (m, OCH(CF3)2), 83.13 (m, OCH(CF3)2), 71.75 (m, OCH(CF3)2),
75.59 (AdN), 69.64 (OCH2), 58.71 (OCH3), 50.60 (CMe2Ph), 43.94
(CH2, AdN), 36.06 (CH2, AdN), 29.99 (CH, AdN), 31.97 (CMe2Ph),
31.33 (CMe2Ph); 19F NMR (C6D6) δ 75.71, -75.95, -76.46, -76.64,
-76.90. The elemental analysis sample for 1a, crystals grown in the
presence of DME, has the formula Li(DME)2Mo(NAd)(CHCMe2Ph)-
[OCH(CF3)2]3. Anal. Calcd for C37H50NF18O7Mo: C, 41.98; H, 4.76;
N, 1.32. Found: C, 41.79; H, 4.75; N, 1.47.
Synthesis of K(DME)Mo(NAd)(CHCMe2Ph)[OCH(CF3)2]3 (1b).
Compound 1b was prepared as a yellow oil by a method analogous to
that described for 1a: 1H NMR (C6D6) δ 12.51 (s, 1, CHCMe2Ph),
7.40 (d, 2, Ph), 7.19 (t, 2, Ph), 7.04 (t, 1, Ph), 6.20 (sept, 1, OCH(CF3)2,
JFH ) 7.2), 4.07 (sept, 2, OCH(CF3)2, JFH ) 6.6), 2.84 (s, 6, OCH3),
2.79 (s, 4, OCH2), 2.21 (d, 6, CH2), 1.98 (s, 3, CH), 1.83 (s, 6, CMe2-
Ph), 1.54 (q, 6, CH2); 13C NMR (C6D6) δ 296.62 (CHCMe2Ph, JCH
)
119), 149.44 (Cipso), 130.03 (Ph), 125.43 (Ph), 124.98 (Ph), 84.44 (sept,
JCF ) 29), 73.87 (AdN), 72.33 (sept, JCF ) 29), 71.07 (OCH2), 58.23
(OCH3), 50.04 (CMe2Ph), 44.20 (AdN), 36.22 (AdN), 31.93 (AdN),
29.99 (CMe2Ph); 19F NMR (C6D6) δ -74.38, -75.13, -75.47.
Synthesis of Mo(NAd)(CHCMe2Ph)[OCH(CF3)2]2(2,4-lutidine)
(2a). Solid KOCH(CF3)2 (1.47 g, 7.13 mol) was added to a cold ether
(∼60 mL) slurry of Mo(NAd)(CHCMe2Ph)(OTf)2(DME) (3.00 g, 3.50
mmol) . One hour later, 2,4-lutidine (0.50 mL, 4.32 mmol) was added.
The resulting mixture was stirred at room temperature overnight and
then filtered through Celite. Removal of the ether from the filtrate in
vacuo afforded a yellow solid. The yellow solid was taken up in ∼10
mL of toluene, and the Celite-filtered solution was diluted with ∼80
mL of pentane. This solution was stored in a -40 °C freezer overnight
to give large off-white crystals of 2a; yield 2.51 g (88%): 1H NMR
(C6D6) δ 13.76 (s, 1, CHCMe2Ph), 7.56 (d, 1, 2,4-Lut), 7.34 (d, 2, Ph),
7.13 (t, 2, Ph), 7.02 (t, 1, Ph), 6.20 (sept, 1, OCH(CF3)2, JFH ) 6.6),
6.12 (s, 1, 2,4-Lut), 5.85 (d, 1, 2,4-Lut), 4.50 (sept, 1, OCH(CF3)2, JFH
) 6.0), 2.35 (s, 3, Meo), 1.96 (s, 3, CMe2Ph), 1.89 (br, 6, CH2), 1.77
(br, 3, CH), 1.56 (s, 3, CMe2Ph), 1.40 (s, 3, Mep), 1.36 (br, 6, CH2);
13C NMR (C6D6) δ 297.8 (CHCMe2Ph, JCH ) 121), 158.5 (Py), 155.8
(Py), 150.8 (Py), 148.1 (Cipso, Ph), 130.0 (Ph), 127.4 (Py), 126.6 (Ph),
126.2 (Ph), 122.4 (Py), 83.4 (sept, OCH(CF3)2, JCF ) 30), 75.0 (sept,
OCH(CF3)2, JCF ) 29), 72.8 (AdN), 50.0 (CMe2Ph), 43.8 (AdN), 36.2
(AdN), 32.0 (CMe2Ph), 31.0 (CMe2Ph), 29.9 (AdN), 26.3 (Me, 2,4-
Lut), 20.39 (Me, 2,4-Lut); 19F NMR (C6D6) δ -74.99, -74.63, -74.54.
Anal. Calcd for C33H38N2O2F12Mo: C, 48.42; H, 4.68; N, 3.42.
Found: C, 48.11; H, 4.83; N, 3.56.
The results reported here should be compared with a recent
report that catalysts in this family that contain triphenylacetate
ligands allow only â addition of the propargylic terminal triple
bonds in diethyl dipropargylmalonate and thereby formation of
only cyclohexene rings in the polyene backbone formed in a
cyclopolymerization reaction.36 The inherent regiochemical
problem associated with addition of a terminal alkyne to an
alkylidene would appear to be on its way to being solved to
some degree simply by employing “small” alkoxides or “large”
carboxylate ligands on the metal.
Experimental Section
All manipulations were performed under a nitrogen atmosphere in
a Vacuum Atmosphere drybox or using standard Schlenk techniques.
Reagent grade ether and tetrahydrofuran were distilled from sodium
benzophenone ketyl under nitrogen. Pentane was washed with 5% nitric
acid in sulfuric acid, stored over calcium chloride, and then distilled
from sodium benzophenone ketyl under nitrogen. Dichloromethane
was distilled from calcium hydride under nitrogen. Toluene was
distilled from molten sodium under nitrogen.
All deuterated NMR solvents were passed through a column of
activated alumina. NMR data are listed in parts per million downfield
from tetramethylsilane for proton and carbon, and relative to CFCl3
for fluorine. Coupling constants are quoted in hertz. Variable
temperature NMR spectroscopic studies were recorded without calibrat-
ing the probe temperature.
Gel permeation chromatography (GPC) was carried out using Shodex
KF-802.5, -803, -804, -805, and -800P columns, a Knauer differential
refractometer, and a Viscotek differential refractometer/viscometer
H-500 on samples 0.1-0.3% w/v in THF which were filtered through
a Millex-SR 0.5 mm filter in order to remove particulates. GPC
columns were calibrated versus polystyrene standards (Polymer Labo-
ratories Ltd.) of MW 1206 to 1.03 × 106.
Synthesis of Mo(NAd)(CHCMe2Ph)[OCH(CF3)2]2(Py) (2b). Com-
pound 2b was synthesized in 44% yield by the same method as that
used to prepare 2a: 1H NMR (syn rotamer, C6D6) δ 13.31 (s, 1,
CHCMe2Ph), 8.35 (d, 2, Py), 6.90-7.10 (m, 5, Py + Ph), 6.60 (m, 1,
Ph), 6.25 (m, 3, Py + OCH(CF3)2), 4.37 (sept, 1, OCH(CF3)2, JFH
)
6.6), 1.88 (s, 3, CMe2Ph), 1.84 (b, 6, CH2), 1.78 (b, 3, CH), 1.49 (s, 3,
CMe2Ph), 1.34 (b, 6, CH2); 13C NMR (C6D6) δ 301.2 (CHCMe2Ph,
JCH ) 121), 152.4 (Py), 147.3 (Cipso, Ph), 138.6 (Py), 128.8 (Ph), 126.4
(Ph), 126.0 (Ph), 124.1 (Py), 84.1 (m, OCH(CF3)2), 75.2 (m, OCH-
(CF3)2), 73.4 (AdN), 50.4 (CMe2Ph), 43.8 (AdN), 35.8 (AdN), 31.7
(CMe2Ph), 30.3 (CMe2Ph), 29.5 (AdN); 19F NMR (C6D6) δ -68.85,
-69.32, -69.66, -70.29.
Synthesis of Mo(NAd)(CHCMe2Ph)[OCH(CF3)2]2(2-(3-pentyl)-
NC5H4) (2c). Compound 2c was prepared in 72% yield by a method
analogous to that used to prepare 2a. The compound decomposes
quickly at room temperature and at low temperature (-40 °C) after a
long period of time, even in the solid state: 1H NMR (C6D5CD3, -35
°C) δ 13.76 (s, 1, CHCMe2Ph), 8.46 (d, 1), 7.34 (d, 2, Ph), 7.14 (t, 2,
Ph), 7.03 (t, 1, Ph), 6.73 (t, 1), 6.33 (m, 2), 6.12 (m, 1, OCH(CF3)2),
4.21 (m, 1, OCH(CF3)2), 2.24 (m, 1, CHEt2), 1.94 (b, 9, CMe2Ph +
NAd), 1.82 (br, 3, CH), 1.70 (s, 3, CMe2Ph), 1.40 (m, 6, NAd); 13C
NMR (CD2Cl2, -35 °C) δ 296.76 (CHCMe2Ph, JCH ) 121), 164.7
(Py), 149.6 (Py), 148.3 (Cipso, Ph), 139.5 (Py), 128.8 (Ph), 126.6 (Ph),
125.8 (Ph), 123.6 (Py), 120.7 (Py), 83.1 (sept, OCH(CF3)2, JCF ) 30),
73.7 (sept, OCH(CF3)2, JCF ) 29), 72.9 (AdN), 51.0 (CMe2Ph), 43.0
Synthesis of Li(DME)Mo(NAd)(CHCMe2Ph)[OCH(CF3)2]3 (1a).
Solid LiOCH(CF3)2 (150 mg, 0.862 mmol) was added in one portion
to a cold ether solution (∼10 mL) of Mo(NAd)(CHCMe2Ph)(OTf)2-
(DME) (225 mg, 0.263 mmol). After the solution was stirred for 2 h
at room temperature, the ether was removed from the yellow reaction
solution in vacuo. The yellow residue was extracted with pentane (∼20
mL), and the extract was filtered through Celite to afford a light yellow
solution, from which off-white crystalline 1a was obtained upon cooling
of the solution to -40 °C; yield 190 mg (71%): 1H NMR (C6D6) δ
12.59 (s, 1, CHCMe2Ph), 7.35 (d, 2, Ph), 7.15 (t, 2, Ph), 7.01 (t, 1,
(36) Schattenmann, F. J.; Schrock, R. R.; Davis, W. M. J. Am. Chem.
Soc. 1996, 118, 3296.

3894 J. Am. Chem. Soc., Vol. 118, No. 16, 1996
Schrock et al.
(AdN), 35.8 (AdN), 32.6 (CMe2Ph), 30.3 (CMe2Ph), 29.9 (CHEt2), 29.7
(AdN), 28.7 (CH2CH3), 27.0 (CH2CH3), 11.9 (CH2CH), 10.4 (CH2CH3);
19F NMR (CD2Cl2, -35 °C) δ -76.53, 76.78, 77.77. Anal. Calcd for
MoC36H44N2O2F12: C, 50.24; H, 5.15; N, 3.25. Found: C, 50.15; H,
5.55; N, 3.25.
Synthesis of Mo(NAd)(CHCMe2Ph)[OCH(CF3)2]2(quinuclidine)
(2d). To a 3 mL solution of 39 mg of 2a (0.047 mmol) in toluene was
added 26 mg of quinuclidine (0.24 mmol) at room temperature. All
solvents and excess quinuclidine were removed in vacuo after 1 h, and
the solid residue was recrystallized from 1 mL of pentane at -30 °C
to give 28 mg of off-white crystalline product (70%): 1H NMR (C6D6)
δ 12.81 (s, 1, CHCMe2Ph, JCH ) 119), 7.02-7.10 (m, 4, Ph), 6.92-
6.97 (m, 1, Ph), 6.13 (sept, 1, OCH(CF3)2, JFH ) 7.1), 4.26 (sept, 1,
OCH(CF3)2, JFH ) 6.3), 2.78 (m, 6, quinuclidine), 2.02 (br, 6, CH2),
1.89 (s, 6, CHMe2Ph), 1.45 (s, 6, CH2), 1.39 (br, 3, CH), 1.09 (m, 6,
quinuclidine), 0.93 (br, 6, quinuclidine).
140.8, 139.4, 130.1, 127.6, 127.5, 127.2, 127.0, 126.4, 125.8, 122.9,
83.9 (CMe2Ph), 75.5 (NAd), 43.4 (NAd), 41.8 (CMe2Ph), 35.7 (NAd,
29.8 (CMe2Ph), 29.7 (CMe2Ph), 29.4 (NAd), 27.8 (o-Me), 20.1 (p-
Me); 19F NMR (C6D6) δ -72.94, -73.20, -73.53, -73.95. Anal.
Calcd for MoC47H48N2O2F12: C, 56.63; H, 4.85; N, 2.81. Found: C,
56.58; H, 4.93; N, 2.58.
Synthesis of Mo(N-2,6-Me2C6H4)(CHCMe2Ph)(OC6F5)2(quinucli-
dine) (7). Two equivalents of KOC6F5 was added to a cold solution
(-40 °C) of 0.5 mol of [Mo(CHCMe2Ph)(N-2,6-Me2C6H4)(OSO2CF3)2-
(DME)] in 50 mL of THF. After 5 h at -40 °C, 1 equiv of quinuclidine
was added and the reaction mixture was stirred for 2 h at room
temperature. The solvents were evaporated in vacuo to give an orange
powder. The orange powder was extracted with a minimum amount
of toluene. The extract was filtered, and the solvents were removed
from the filtrate in vacuo. The crude product was recrystallized from
cold ether to give yields in the range 70-90%: 1H NMR (C6D6) δ
14.16 (s, 1, CHCMe2Ph anti, JCH ) 145 (from 13C satellites)), 13.02
(s, 1, CHCMe2Ph syn, JCH ) 115 (from 13C satellites)), 7.12-6.75 (m,
8, aromatic), 2.93 (m, 6, NCH2 anti), 2.75 (m, 6, NCH2 syn), 2.72 (s,
6, NMePh syn and anti), 1.95 (s, 3, CHCMeMe′Ph), 1.73 (s, 3,
CHCMeMe′Ph syn), 1.33 (s, 3, CHCMeMe′Ph anti), 1.10 (s, 3,
CHCMeMe′Ph syn), 1.00 (m, 1, NCH2CH2CH syn and anti), 0.89 (m,
6, NCH2CH2 syn and anti); 13C NMR δ 298.34 (CHCMe2Ph anti),
298.17 (d, J ) 117, CHCMe2Ph syn), 155.51 (Cipso CMe2Ph anti),
153.96 (Cipso CMe2Ph syn), 148.6-136.48 (m, OC6F5), 135.50 (Co NAr),
128.67-125.75 (aromatics), 53.80 (NCH2CH2 anti), 53.27 (NCH2CH2
syn), 52.60 (CMe2Ph syn), 51.47 (CMe2Ph anti), 31.06 (CMePh syn),
30.86 (CMePh anti), 28.71 (CMePh syn), 27.90 (CMePh anti), 25.69
(NCH2CH2 syn and anti), 20.67 (NMePh syn), 19.62 (NCH2CH2CH
syn and anti), 19.19 (NMePh anti). Anal. Calcd for C37H34-
MoN2F10O2: C, 53.89; H, 4.16; N, 3.40. Found: C, 53.84; H, 4.27;
N, 3.22.
Synthesis of Mo(NAd)[C(o-TMS-C6H4)(CHdCHCMe2Ph)][OCMe-
(CF3)2]2 (3a/b). A pentane solution (60 mL) of the compound
Mo(NAd)(CHCMe2Ph)[OCMe(CF3)2]2 (1.00 g, 1.35 mmol) was cooled
to -40 °C. To the cold yellow solution was added o-TMSPA (239
mg, 1.37 mmol). The mixture was then kept cold at -40 °C for 1/2 h.
During this period, the mixture assumed a deep red color. The red
solution was then filtered through Celite, concentrated in vacuo to ∼10
mL, and put into a -40 °C freezer. Deep red microcrystalline solid
was obtained after a few days; yield 658 mg (53%): 1H NMR (C6D6,
minor) δ 8.44 (d, Hâ, J ) 15.6), 6.9-7.7 (m, H’s of Ph’s), 4.30 (d, Hγ,
1
J ) 15.6), 0.34 (TMS); H NMR (C6D6, major) δ 7.34 (d, Hâ, J )
15.6), 6.9-7.7 (m, H’s of Ph’s), 5.78 (d, Hγ, J ) 15.6), 0.22 (TMS);
13C NMR (major, C6D6) δ 283.6, 1, 126.3 (JCH ) 150), 122.8 (JCH
)
150). Anal. Calcd for MoC39H47NO2F12Si: C, 51.26; H, 5.18; N, 1.53.
Found: C, 51.53; H, 5.52; N, 1.69.
Synthesis of Mo(NAd)(CMePh)[OCH(CF3)2]2(2,4-lutidine) (5a).
R-Methylstyrene (0.105 mL, 0.808 mmol) was added to an ether
solution of 2a (578 mg, 0.71 mmol). After 24 h the ether was removed
in vacuo to leave a dark yellow oily residue. Recrystallization of the
residue from pentane gave 5a as bright yellow microcrystals; yield 312
mg (56%): 1H NMR (C6D6) δ 8.45 (d, 1, 6, 2,4-Lut), 7.68 (dd, 2, J )
1.2, 8.4, Ph), 7.25 (m, 2, Ph), 7.01 (t , 1, Ph), 6.34 (sept, OCH(CF3)2),
6.17 (s, 1, 2,4-Lut), 6.02 (d, 1, 6), 4.51 (sept, 1, 6.3, OCH(CF3)2), 2.31
(s, 3, 2,4-Lut), 1.92 (br, 6, AdN), 1.76 (br, 3, AdN), 1.50 (br, 3, 2,4-
Lut), 1.34 (br, 6, AdN); 13C NMR (C6D6) δ 302.4 (CMe2Ph), 160.5
(2,4-Lut), 153.3 (2,4-Lut), 151.1 (Ph), 146.7 (2,4-Lut), 129.2 (Ph), 128.1
(Ph), 126.5 (Ph), 123.3 (2,4-Lut), 83.0 (OCH(CF3)2), 74.9, 74.5 (dNC),
43.6 (NAd), 36.0 (AdN), 29.7 (AdN), 27.4 (Me), 21.4 (Me); 19F NMR
(C6D6) δ -78.96 (1), -79.19 (1), -79.29 (1), -79.70 (1). Anal. Calcd
for MoC31H34N2O2F12: C, 47.01; H, 4.33; N, 3.54. Found: C, 47.26;
H, 4.44; N, 3.49.
Synthesis of Mo(NAd)(CPh2)[OCH(CF3)2]2(2,4-lutidine) (5b).
This compound was prepared in 46% yield by the same procedure as
that used to prepare 5a: 1H NMR (C6D6) δ 8.94 (d, 1, J ) 6), 7.50 (dt,
2, Ph), 7.10-7.26 (m, 6, Ph), 7.00 (m, 2), 6.23 (s, 1, 2,4-Lut), 6.10
(sept, 1, OCH(CF3)2), 6.09 (d, 1, 2,4-Lut), 4.82 (sept, 1, J ) 6.0,
OCH(CF3)2), 2.39 (s, 3, 2,4-Lut), 1.87 (q, 6, AdN), 1.75 (br, 3, AdN),
1.49 (s, 3, 2,4-Lut), 1.3 (s, 6, AdN); 13C NMR (C6D6) δ 310 (CPh2),
160.9, 153.0, 151.7, 151.2, 141.5, 129.5, 129.0, 128.6, 127.9, 127.38,
126.9, 123.2, 82.2 (OCH(CF3)2), 76.1 (dNC), 75.2 (OCH(CF3)2), 43.4
(NAd), 35.9 (NAd), 29.7 (NAd), 28.1 (Me, 2,4-Lut), 20.1 (Me, 2,4-
Lut); 19F NMR (C6D6) δ 73.57 (1), 74.43 (2), 74.52 (1). Anal. Calcd
for MoC36H36N2O2F12: C, 50.71; H, 4.26; N, 3.28. Found: C, 50.65;
H, 4.21; N, 3.31.
Synthesis of Mo(NAd)[C(Ph)C(Ph)CHCMe2Ph][OCH(CF3)2]2-
(2,4-lutidine) (6). To a cold (-40 °C) ether solution (∼60 mL) of 2a
(645 mg, 0.79 mmol) was added diphenylacetylene (142 mg, 0.80
mmol). The resulting yellow mixture was stirred at room temperature
for 6 h and then stripped of solvents in vacuum. The yellow solid
residue obtained was recrystallized from a toluene/pentane mixture to
afford 6 as a bright yellow microcrystalline solid; yield 530 mg
(67%): 1H NMR (C6D6) δ 8.75 (d, 1), 7.64 (d, 2), 6.80-7.40 (m, Ph’s),
6.20 (m, 1, OCH(CF3)2), 6.16 (d, 1), 6.14 (2H, m-H and CHCMe2Ph),
4.89 (m, 1, OCH(CF3)2), 2.30 (s, 3, o-Me), 1.76 (br, 6, CH2’s), 1.64
(br, 3, CH’s), 1.48 (br, 6, CH2’s), 1.29 (s, 3, p-Me), 1.22 (3, 6, CMe2-
Ph); 13C NMR (C6D6) δ 305.1, 160.4, 153.9, 150.9, 150.8, 149.4, 143.6,
Synthesis of Mo(N-2,6-i-Pr2C6H4)(CHCMe2Ph)[OCH(CF3)2]2(2,4-
lutidine) (8). This compound was synthesized in 74% yield by the
same procedure as that used to prepare 2a: 1H NMR (C6D6) δ 13.80
(s, 1H, CHCMe2Ph), 7.57 (d, 2H), 7.49 (d, 1H), 7.36 (t, 3H), 7.23 (t,
1H), 6.14 (s, 1H), 6.10 (d, 1H), 6.00 (m, 1H), 3.95 (m, 2H), 3.80 (m,
1H), 2.28 (s, 3H), 2.18 (s, 3H), 1.71 (s, 3H), 1.55 (s, 3H), 1.48 (d,
6H), 0.99 (d, 6H); 13C NMR (C6D6) δ 295.9 (CMe2Ph, JCH ) 127 Hz),
157.8, 156.3, 152.5, 151.0, 148.9, 147.8, 128.7, 17.8, 127.7, 126.5,
126.2, 123.82, 123.77, 77.4 (m, OCH(CF3)2), 75.6 (m, OCH(CF3)2),
54.6 (CMe2Ph), 31.4 (Me), 29.1 (Me), 28.1 (o-Me), 24.2 (Me), 24.1
(Me), 22.5, 20.1 (Me); 19F NMR (C6D6) δ -73.29, -73.88, -75.45,
-75.68. Anal. Calcd for MoC35H40N2O2F12: C, 49.77; H, 4.77; N,
3.32. Found: C, 49.91; H, 4.65; N, 3.10.
Photocatalyzed Isomerization of the Alkylidene Ligand in
Mo(NAd)[C(o-TMS-C6H4)(CHdCHCHCMe2Ph)][OCMe(CF3)2]2
(3a/b). A sample of the equilibrium mixture of 3a/b (71 mg, 0.078
mmol) and 6.8 µL of diphenylmethane (6.8 mg, 0.040 mmol) were
dissolved in 0.80 mL of toluene-d8. The deep red solution was then
irradiated with a medium-pressure Hg lamp at -60 °C. The concentra-
tion of the minor component increased at the expense of the major
product over a period of 24-48 h. The irradiated mixture was
immediately transferred to a precooled NMR probe, and the disap-
pearance of the minor product was followed by integration (TMS
group).
Insertion Studies with Catalyst 2a. Bulk toluene-d8 solutions of
catalyst 2a (64.7 mM) with Ph2CH2 (92.7 mM) as an internal standard
and monomers PA (2.6 M), o-MePA (2.1 M), o-i-PrPA (2.9 M), o-t-
BuPA (1.9 M) , and o-TMSPA (0.29 mL) were prepared. Typical
insertion studies were carried out by adding monomer to the catalyst
solution containing an internal standard (Ph2CH2). The catalyst
concentration was measured by integration of the alkylidene resonance
versus the CH2 resonance of Ph2CH2. The amount of the R product
was measured by the same method using the intensity of the Hâ and
Hγ resonances in the first insertion product, and the amount of the â
addition product using the intensity of new alkylidene resonances.
Polymerization of o-TMSPA by 2a. Stock toluene solutions were
prepared for both 2a (13.9 mM) and the monomer o-TMSPA (0.23
M). The polymerization reactions were carried out under dinitrogen
in a drybox by quick addition of the o-TMSPA solution to a vigorously
stirred solution of 2a at room temperature. In a typical experiment,

LiVing Polymerization of Substituted Phenylacetylene
J. Am. Chem. Soc., Vol. 118, No. 16, 1996 3895
0.24 mL of the o-TMSPA solution (55.2 mmol) was quickly injected
Table 9. Observed Rate Constants for Polymerization of
o-TMSPA by 2a in Toluene (by GC) at 22 °C
(a) No Base Added
into a toluene solution of catalyst (50 mL bulk solution diluted to 3
1
mL). The solution turned purple within /2 h, and the color did not
change significantly thereafter. After 24 h, the reaction was brought
into the air, and MeOH was quickly added to induce the precipitation
of poly(o-TMSPA) as a dark purple solid. The polymer was collected
by filtration, washed with MeOH, and dried in vacuo; yield 9.0 mg
(93%). Polymers that had been exposed to air did not show any red-
shifted absorption.
[2a]0 (mM)
kobs (×10-4 s-1
)
kobs/[2a]0 (s-1 M-1
)
0.41
0.52
0.72
0.98
1.03
1.42
1.89
2.06
2.46
2.51
1.48
0.36
0.25
0.29
0.31
0.26 (0.30)
0.30 (0.28)
0.34
0.31 (0.24)
0.31
2.08
2.10
3.00
3.85 (3.36)
4.26 (4.02)
6.41
6.66 (8.43)
7.71
Reactions could also be quenched with 100 equiv of benzaldehyde
or 8 equiv of acetic acid. Polymers prepared in the absence of air,
which contained 20-30 equiv of o-TMSPA on the average, could be
shown to isomerize to an air-sensitive form as described in the text.
Polymerization of o-TMSPA by 2c. A bulk toluene solution of
o-TMSPA (0.40 M) was prepared. A typical polymerization was the
following: Inside a drybox, catalyst 2c (8 mg, 10 mmol) was dissolved
in 10 mL of toluene; 1.00 mL of monomer solution (400 mmol) was
quickly added to the solution. The polymerization solution turned to
red and then purple within a few minutes at room temperature. After
3 h of stirring, ∼1.5 mL of polymerization solution was put into a
separate vial and quenched with MeOH inside the drybox. The purple
poly(o-TMSPA) precipitated out immediately as a powder and was
collected by filtration, dried in vacuum, and used for GPC measurement.
The same procedure was repeated after 22 h and 45 h stirring.
Kinetics Measurements of the Polymerization of o-TMSPA by
2a. Stock toluene solutions of the catalyst 2a (10.3 mM) and o-TMSPA
(0.41 M) containing internal standard Ph2CH2 (0.12 M) were prepared
separately. Polymerization reactions were carried out by quick addition
of toluene solutions of o-TMSPA and internal standard (Ph2CH2) to
the toluene solutions of initiator 2a, all of known concentration and
volume, under pseudo-first-order conditions ([o-TMSPA]/[2a] ) 80)
inside a drybox. A typical kinetic run consisted of the following: 0.5
mL of catalyst solution was diluted with 3.50 mL of toluene. To this
solution was quickly added 1.00 mL of monomer/internal standard
solution with vigorous stirring. Aliquots of the polymerization solution
(∼0.2 mL) were taken at intervals and quenched with an excess amount
of MeOH (∼1.5 mL) immediately. The quenched reactions were
filtered through Celite to remove the precipitated polymer. The filtrates
were then analyzed by gas chromatography. The temperature inside
the drybox (22 °C) was found to vary less than (1 °C during
polymerization experiments.
Four additional runs were carried out at [2a]0 ) 0.98 mM with base
added (Table 9b). In the run in which no base was added, it was
assumed that [2,4-lutidine] ) 0.98 mM (complete dissociation of
lutidine). In the other runs [2,4-lutidine] was assumed to be the amount
that had been added to the reaction. A plot of [2a]0/kobs versus [2,4-
lutidine] gave k2a ) 0.31 s-1 M-1 and K2a ) 61 M-1 with R2 ) 0.987.
Using K2a ) 61 M-1 the values for [2,4-lutidine] in Table 9b were
recalculated and the plot was redone to give k2a ) 0.32 s-1 M-1 and
K2a ) 62 M-1 with R2 ) 0.996 (Figure 4a).
Polymerization of o-MePA by 2a. Bulk toluene solutions of
o-MePA (2.1 M) and catalyst 1 (12.3 mM) were prepared. A typical
polymerization was the following. Inside a drybox, 0.24 mL of catalyst
(3.0 mmol) solution was diluted to ∼5 mL. To this was quickly added
0.10 mL (210 mmol) of monomer solution. The solutions turned
yellow, orange, and finally red within a few minutes at room
temperature. After the solution was stirred for 24 h, methanol (∼15
mL) was added and the precipitated deep red poly(o-MePA) was
collected by filtration and dried in vacuum overnight; weight 23 mg
(95%).
7.50
0.30
(b) Base Added (Except Run 1)
[2a]0
(mM)
added
adjusted
[2,4-Lut]
kobs
[2a]0/kobs
[2,4-Lut] (mM)
(×10-4 s-1
)
(s-1 M-1
)
0.98
0.98
0.98
0.98
0.98
0
0.93
5.28
9.36
13.3
17.3
3.00
2.31
1.93
1.76
1.50
3.27
4.24
5.08
5.57
6.53
4.53
8.74
12.8
16.9
Table 10. Observed Rate Constants for Polymerization of
o-TMSPA by 7 in Toluene (by GC) at 22 °C with Base Added
[7]0 (mM) [quinuclidine] (M) kobs (×10-5 s-1
)
[7]0/kobs (s-1 M-1
)
1.36
1.36
1.36
1.36
0.0122
0.0190
0.0272
0.0449
1.77
1.14
0.78
0.49
77
119
174
275
Polymerization of o-TMSPA by 7. Stock solutions of the monomer
(1.05 M in toluene) and the catalyst (0.04 M in toluene) were prepared.
The catalyst was added to 5 mL of toluene; 0.5 mL of the monomer
solution was added rapidly to the stirred catalyst solution. The solutions
turned immediately deep red and, after 3 h, purple. After 36 h, 10
equiv of benzaldehyde was added. After an additional 12 h, the purple
solution was passed through alumina and the solvent was removed in
vacuo, affording the polymer as a purple film (88-92 mg; 96-100%).
Kinetics Measurements of the Polymerization of o-TMSPA by
7. The general procedure was the same as in the polymerization using
2a as the initiator. The results are shown in Table 10.
Acknowledgment. R.R.S. thanks the Director, Office of
Basic Energy Research, Office of Basic Energy Sciences,
Chemical Sciences Division of the U.S. Department of Energy
(Contract DE-FG02-86ER13564), for support.
Supporting Information Available: Detailed description
of X-ray data collection, structure solution, and refinement,
labeled ORTEP diagrams, and tables of fractional coordinates
and anisotropic thermal parameters for 1a and 2a (24 pages).
This material is contained in many libraries on microfiche,
immediately follows this article in the microfilm version of the
journal, can be ordered from the ACS, and can be downloaded
from the Internet; see any current masthead page for ordering
information and Internet access instructions.
Polymerization of o-i-PrPA by 2a. Bulk toluene solutions of o-i-
PrPA (3.0 M) and catalyst 1 (7.5 mm) were prepared. Polymerization
and polymer isolation were carried out as described for poly(o-MePA).
A typical run with monomer (630 mmol) and catalyst (7.5 mmol) gave
87 mg (100%) polymer as a deep red solid.
JA954155W