Alkylidene and metalacyclic complexes of tungsten that contain a chiral biphenoxide ligand. Synthesis, asymmetric ring-closing metathesis, and mechanistic investigations

Article abstract of DOI:10.1021/ja0210603

Two complexes that contain the racemic or enantiomerically pure (S) form of the 3,3′-di-tertbutyl-5,5′,6,6′-tetramethyl-1, 1′-biphenyl-2,2′-diolate (Biphen^2-) ligand, W(NAr)(CHCMe₂Ph)(Biphen) (2a) and W(NAr′)(CHCMe₂Ph)(Biphen) (2b) (Ar = 2,6-i-Pr₂C₆H₃; Ar′ = 2,6-Me₂C₆H₃), were prepared and shown to be viable catalysts for several representative ring-closing reactions to give products in good yields in most cases and high % ee in asymmetric reactions. Exploration of the reaction between 2a and a stoichiometric amount of one desymmetrization substrate allowed two intermediate tungstacyclobutane complexes to be observed, in addition to the final and quite stable tungstacyclobutane complex formed in a reaction between the ring-closed product and a tungsten methylene complex. Reactions involving ¹³C labeled ethylene allowed for the observation of an unsubstituted tungstacyclobutane complex, an ethylene complex, an unsubstituted tungstacyclopentane complex, and a heterochiral dimeric form of a methylene complex. The tungstacyclopentane complex was found to catalyze the dimerization of ethylene to 1-butene slowly.

Full text of DOI:10.1021/ja0210603

Published on Web 02/05/2003
Alkylidene and Metalacyclic Complexes of Tungsten that
Contain a Chiral Biphenoxide Ligand. Synthesis, Asymmetric
Ring-Closing Metathesis, and Mechanistic Investigations
W. C. Peter Tsang,† Kai C. Hultzsch,†,§ John B. Alexander,†
Peter J. Bonitatebus, Jr.,†,| Richard R. Schrock,*,† and Amir H. Hoveyda‡
Contribution from the Department of Chemistry, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139, and Department of Chemistry, Merkert Chemistry Center,
Boston College, Chestnut Hill, Massachusetts 02467
Received August 7, 2002 ; E-mail: rrs@mit.edu
Abstract: Two complexes that contain the racemic or enantiomerically pure (S) form of the 3,3′-di-tert-
butyl-5,5′,6,6′-tetramethyl-1,1′-biphenyl-2,2′-diolate (Biphen2-) ligand, W(NAr)(CHCMe2Ph)(Biphen) (2a) and
W(NAr′)(CHCMe2Ph)(Biphen) (2b) (Ar ) 2,6-i-Pr2C6H3; Ar′ ) 2,6-Me2C6H3), were prepared and shown to
be viable catalysts for several representative ring-closing reactions to give products in good yields in most
cases and high % ee in asymmetric reactions. Exploration of the reaction between 2a and a stoichiometric
amount of one desymmetrization substrate allowed two intermediate tungstacyclobutane complexes to be
observed, in addition to the final and quite stable tungstacyclobutane complex formed in a reaction between
the ring-closed product and a tungsten methylene complex. Reactions involving 13C labeled ethylene allowed
for the observation of an unsubstituted tungstacyclobutane complex, an ethylene complex, an unsubstituted
tungstacyclopentane complex, and a heterochiral dimeric form of a methylene complex. The tungstacyclo-
pentane complex was found to catalyze the dimerization of ethylene to 1-butene slowly.
Introduction
systems were present in very low concentrations and that the
carbene complexes known at the time (compounds that con-
Catalytic alkene metathesis in its simplest form consists of a
redistribution of alkylidene fragments of olefins to give, at
equilibrium, a mixture of all possible cis and trans olefins.1 The
reaction was known to be promoted by tungsten, molybdenum,
and rhenium, although little was known about the catalysts in
detail except that a carbene ligand almost certainly was present
in the species that re-formed in this reaction.2-7 It was clear
that the carbene complexes generated in these “classical”
tained a heteroatom-stabilized carbene ligand) did not react with
olefins in a catalytic metathesis-like fashion via propagating
carbenes derived from the olefins.3,8,9 Therefore, it was highly
desirable to learn how to synthesize new types of carbene
complexes that would react with olefins in a metathesis manner
and how to prepare stable and active complexes, to be able to
control the metathesis reaction at a molecular level. The
discovery that alkoxide ligands would “turn on” metathesis of
olefins by well-characterized “high oxidation state” tantalum
alkylidene complexes10,11 and the discovery of related high
oxidation state oxo alkylidene complexes of tungsten12 ulti-
mately led to the conclusion that high oxidation state com-
plexes13 almost certainly were by far the most common, if not
the only catalysts containing Mo or W that would sustain olefin
metathesis. In 1986, a method of synthesizing stable tungsten
imido alkylidene complexes of the type W(NAr)(CH-t-Bu)-
(OR′)2 (Ar ) 2,6-i-Pr2C6H3 and OR′ ) O-t-Bu, OCMe(CF3)2,
† Massachusetts Institute of Technology.
‡ Merkert Chemistry Center.
§ Present address: Institut fu¨r Organische Chemie, Universita¨t Erlangen
Henkestrasse 42, D-91054 Erlangen, Germany.
| Present address: Emerging Technologies Laboratory PSCT, GE Global
Research Center CEB 138, 1 Research Circle, Niskayuna, NY 12309.
(1) Ivin, K. J.; Mol, J. C. Olefin Metathesis and Metathesis Polymerization;
Academic Press: San Diego, 1997.
(2) (a) Casey, C. P.; Burkhardt, T. J. J. Am. Chem. Soc. 1973, 95, 5833. (b)
Casey, C. P.; Burkhardt, T. J. J. Am. Chem. Soc. 1974, 96, 7808. (c) Casey,
C. P.; Tuinstra, H. E.; Saemen, M. C. J. Am. Chem. Soc. 1976, 98, 608.
(d) Casey, C. P.; Burkhardt, T. J.; Bunnell, C. A.; Calabrese, J. C. J. Am.
Chem. Soc. 1977, 99, 2127. (e) Casey, C. P.; Polichnowski, S. W.;
Shusterman, A. J.; Jones, C. R. J. Am. Chem. Soc. 1979, 101, 7282
(3) (a) Katz, T. J.; McGinnis, J. J. Am. Chem. Soc. 1975, 97, 1592. (b) Katz,
T. J.; Lee, S. J.; Acton, N. Tetrahedron Lett. 1976, 4247. (c) Katz, T. J.;
Rothchild, R. J. Am. Chem. Soc. 1976, 98, 2519. (d) Katz, T. J.; Acton, N.
Tetrahedron Lett. 1976, 4251. (e) Katz, T. J.; McGinnis, J.; Altus, C. J.
Am. Chem. Soc. 1976, 98, 606. (f) Katz, T. J.; McGinnis, J. J. Am. Chem.
Soc. 1977, 99, 1903. (g) Katz, T. J.; Hersh, W. H. Tetrahedron Lett. 1977,
585.
(8) (a) Fischer, E. O. Pure Appl. Chem. 1970, 24, 407. (b) Fischer, E. O. Pure
Appl. Chem. 1972, 30, 353. (c) Fischer, E. O. AdV. Organomet. Chem.
1976, 14, 1.
(9) (a) Cardin, D. J.; Cetinkaya, B.; Doyle, M. J.; Lappert, M. F. Chem. Soc.
ReV. 1973, 2, 99. (b) Cardin, D. J.; Cetinkaya, B.; Lappert, M. F. Chem.
ReV. 1972, 72, 545. (c) Cotton, F. A.; Lukehart, C. M. Prog. Inorg. Chem.
1972, 16, 487.
(10) Schrock, R. R. J. Organomet. Chem. 1986, 300, 249.
(11) Schrock, R. R.; Rocklage, S. M.; Wengrovius, J. H.; Rupprecht, G.;
Fellmann, J. J. Mol. Catal. 1980, 8, 73.
(12) Wengrovius, J. H.; Schrock, R. R.; Churchill, M. R.; Missert, J. R.; Youngs,
W. J. J. Am. Chem. Soc. 1980, 102, 4515.
(13) Schrock, R. R. In Reactions of Coordinated Ligands; Braterman, P. R.,
Ed.; Plenum: New York, 1986.
(4) Tebbe, F. N.; Parshall, G. W.; Ovenall, D. W. J. Am. Chem. Soc. 1979,
101, 5074.
(5) Grubbs, R. H.; Carr, D. D.; Hoppin, C.; Burk, P. L. J. Am. Chem. Soc.
1976, 98, 3478.
(6) Dolgoplosk, B. A.; Makovetsty, K. L.; Golenko, T. G.; Korshak, Y. V.;
Timyakova, E. I. Eur. Polym. J. 1974, 10, 901.
(7) He´rrison, J. L.; Chauvin, Y. Makromol. Chem. 1971, 141, 161.
9
2652
J. AM. CHEM. SOC. 2003, 125, 2652-2666
10.1021/ja0210603 CCC: $25.00 © 2003 American Chemical Society

Alkylidene and Metalacyclic Complexes of Tungsten
A R T I C L E S
O-2,6-i-Pr2C6H3, etc.) was discovered.14,15 If the alkoxide is
relatively electron withdrawing (e.g., OR′ ) OCMe(CF3)2), these
complexes are fully active for the metathesis of internal alkenes.
Propagating alkylidene complexes and intermediate tungsta-
cyclobutane complexes could be observed and, in some cases,
isolated and characterized.14 The relatively high stability of
unsubstituted tungstacyclobutane complexes was believed to
limit the utility of tungsten catalysts for the metathesis of
terminal olefins at room temperature, although the metathesis
of terminal olefins was not investigated extensively and, in
particular, not at elevated temperatures. Recent developments
in high oxidation state alkylidene chemistry can be found in a
comprehensive review.16
In the last five years, a variety of molybdenum imido
alkylidene catalysts of the type Mo(NAr)(CHR)(diolate) (R )
CMe3 or CMe2Ph) have been prepared in which the diolate is
optically pure.26 When the diolate is a biphenolate or binaph-
tholate, these species have been shown to promote a variety of
asymmetric metathesis reactions efficiently.26a-p Mo(NAr)-
(CHR)(diolate) complexes have a “modular” character in the
sense that one now can choose from several imido groups and
diolates and thereby optimize catalyst activity and selectivity.26i
Complexes 1a and 1b were the first in the class of biphenolate
or binaphtholate catalysts to be prepared.26f Because the first
well-characterized olefin metathesis catalysts contained tungsten
rather than molybdenum, we felt compelled to attempt to
synthesize and test tungsten analogues of 1a and 1b, 2a and
2b. In this paper, we report the preparation of 2a and 2b and
examine their viability in representative asymmetric metathesis
reactions. Because ethylene is generated in many metathesis
reactions and may play a significant role in determining catalyst
longevity, we also have explored reactions between 2a or 2b
and ethylene.
In the mid 1980s, a relatively efficient synthesis of Mo(NAr)-
(CH-t-Bu)(OR′)2 catalysts from various molybdates was devel-
oped.17 It was believed, by analogy with earlier work concerning
alkyne metathesis by tungsten18 and molybdenum19 alkylidyne
trialkoxide complexes via intermediate metalacyclobutadiene
complexes, that molybdacyclobutane complexes would be more
likely to lose an olefin readily and re-form an alkylidene than
would tungstacyclobutane complexes. Therefore, the possibility
that metal would remain largely sequestered as an unsubstituted
metalacyclobutane complex, which would be formed in any
system in which a terminal olefin was present, would be
diminished by employing molybdenum. In fact, it was found
that even terminal alkenes could be metathesized readily at room
temperature by Mo-based catalysts such as Mo(NAr)(CH-t-Bu)-
[OCMe(CF3)2]2.20-22 Applications to organic chemistry using
Mo(NAr)(CHCMe2Ph)[OCMe(CF3)2]2 also began to appear,
most notably in 1992 in the form of the ring-closing metathesis
(RCM) reaction.23 It appeared that molybdenum catalysts would
always be preferred over tungsten catalysts, in part because of
the relative ease of synthesizing them, and because of their
supposed lower sensitivity to certain functionalities, water,
oxygen, etc. The development of well-characterized ruthenium-
based catalysts in the mid 1990s led to a dramatic further
expansion of interest in the application of metathesis to organic
chemistry.24,25
Results
Synthesis and Characterization of New Tungsten Bipheno-
late Complexes. Two precursors to tungsten diisopropylphenyl-
imido alkylidene catalysts, W(NAr)(CHCMe2Ph)Cl2(DME) (3a,
(25) (a) Scholl, M.; Trnka, T. M.; Morgan, J. P.; Grubbs, R. H. Tetrahedron
Lett. 1999, 40, 2247. (b) Huang, J.; Stevens, E. D.; Nolan, S. P.; Peterson,
J. L. J. Am. Chem. Soc. 1999, 121, 2674. (c) Ackermann, L.; Fu¨rstner, A.;
Weskamp, T.; Kohl, F. J. Tetrahedron Lett. 1999, 40, 4787. (d) Weskamp,
T.; Kohl, F. J.; Hieringer, W.; Gliech, D.; Herrmann, W. A. Angew. Chem.,
Int. Ed. 1999, 38, 2416. (e) Sanford, M. S.; Ulman, M.; Grubbs, R. H. J.
Am. Chem. Soc. 2001, 123, 749. (f) Kingsbury, J. S.; Harrity, J. P. A.;
Bonitatebus, P. J.; Hoveyda, A. H. J. Am. Chem. Soc. 1999, 121, 791. (g)
Garber, S. B.; Kingsbury, J. S.; Gray, B. L.; Hoveyda, A. H. J. Am. Chem.
Soc. 2000, 122, 8168. (h) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res.
2001, 34, 18.
(14) Feldman, J.; Schrock, R. R. Prog. Inorg. Chem. 1991, 39, 1.
(15) Schaverien, C. J.; Dewan, J. C.; Schrock, R. R. J. Am. Chem. Soc. 1986,
108, 2771.
(16) Schrock, R. R. Chem. ReV. 2002, 102, 145.
(17) Schrock, R. R.; Murdzek, J. S.; Bazan, G. C.; Robbins, J.; DiMare, M.;
O’Regan, M. J. Am. Chem. Soc. 1990, 112, 3875.
(26) (a) Alexander, J. B.; La, D. S.; Cefalo, D. R.; Hoveyda, A.; Schrock, R. R.
J. Am. Chem. Soc. 1998, 120, 4041. (b) La, D. S.; Alexander, J. B.; Cefalo,
D. R.; Graf, D. D.; Hoveyda, A. H.; Schrock, R. R. J. Am. Chem. Soc.
1998, 120, 9720. (c) Zhu, S. S.; Cefalo, D. R.; La, D. S.; Jamieson, J. Y.;
Davis, W. M.; Hoveyda, A. H.; Schrock, R. R. J. Am. Chem. Soc. 1999,
121, 8251. (d) La, D. S.; Ford, J. G.; Sattely, E. S.; Bonitatebus, P. J.;
Schrock, R. R.; Hoveyda, A. H. J. Am. Chem. Soc. 1999, 121, 11603. (e)
Weatherhead, G. S.; Ford, J. G.; Alexanian, E. J.; Schrock, R. R.; Hoveyda,
A. H. J. Am. Chem. Soc. 2000, 122, 1828. (f) Alexander, J. B.; Schrock,
R. R.; Davis, W. M.; Hultzsch, K. C.; Hoveyda, A. H.; Houser, J. H.
Organometallics 2000, 19, 3700. (g) Dolman, S. J.; Sattely, E. S.; Hoveyda,
A. H.; Schrock, R. R. J. Am. Chem. Soc. 2002, 124, 6991. (h) Weatherhead,
G. S.; Houser, J. H.; Ford, J. G.; Jamieson, J. Y.; Schrock, R. R.; Hoveyda,
A. H. Tetrahedron Lett. 2000, 49, 9553. (i) Hoveyda, A. H.; Schrock, R.
R. Chem.-Eur. J. 2001, 7, 945. (j) Cefalo, D. R.; Kiely, A. F.; Wuchrer,
M.; Jamieson, J. Y.; Schrock, R. R.; Hoveyda, A. H. J. Am. Chem. Soc.
2001, 123, 3139. (k) Aeilts, S. L.; Cefalo, D. R.; Bonitatebus, P. J.; Houser,
J. H.; Hoveyda, A. H.; Schrock, R. R. Angew. Chem., Int. Ed. 2001, 40,
1452. (l) La, D. S.; Sattely, E. S.; Ford, J. G.; Schrock, R. R.; Hoveyda, A.
H. J. Am. Chem. Soc. 2001, 123, 7767. (m) Hultzsch, K. C.; Bonitatebus,
P. J., Jr.; Jernelius, J.; Schrock, R. R.; Hoveyda, A. H. Organometallics
2001, 20, 4705. (n) Tsang, W. C. P.; Schrock, R. R.; Hoveyda, A. H.
Organometallics 2001, 20, 5658. (o) Hultzsch, K. C.; Jernelius, J. A.;
Hoveyda, A. H.; Schrock, R. R. Angew. Chem., Int. Ed. 2002, 589. (p)
Kiely, A. F.; Jernelius, J. A.; Schrock, R. R.; Hoveyda, A. H. J. Am. Chem.
Soc. 2002, 124, 2868. (q) Fujimura, O.; Grubbs, R. H. J. Am. Chem. Soc.
1996, 118, 2499. (r) Fujimura, O.; Grubbs, R. H. J. Org. Chem. 1998, 63,
824.
(18) (a) Listemann, M. L.; Schrock, R. R.; Dewan, J. C.; Kolodziej, R. M. Inorg.
Chem. 1988, 27, 264. (b) Schrock, R. R.; Listemann, M. L.; Sturgeoff, L.
G. J. Am. Chem. Soc. 1982, 104, 4291. (c) Churchill, M. R.; Ziller, J. W.;
Freudenberger, J. H.; Schrock, R. R. Organometallics 1984, 3, 1554. (d)
Freudenberger, J. H.; Schrock, R. R.; Churchill, M. R.; Rheingold, A. L.;
Ziller, J. W. Organometallics 1984, 3, 1563. (e) Listemann, M. L.; Schrock,
R. R. Organometallics 1985, 4, 74. (f) Schrock, R. R.; Freudenberger, J.
H.; Listemann, M. L.; McCullough, L. G. J. Mol. Catal. 1985, 28, 1. (g)
Freudenberger, J. H.; Pedersen, S. F.; Schrock, R. R. Bull. Soc. Chim. Fr.
1985, 349.
(19) McCullough, L. G.; Schrock, R. R.; Dewan, J. C.; Murdzek, J. S. J. Am.
Chem. Soc. 1985, 107, 5987.
(20) Schrock, R. R. In Topics in Organometallic Chemistry. Alkene Metathesis
in Organic Synthesis; Fu¨rstner, A., Ed.; Springer: Berlin, 1998; Vol. 1, p
1.
(21) Schrock, R. R. Tetrahedron 1999, 55, 8141.
(22) Schrock, R. R. J. Chem. Soc., Dalton Trans. 2001, 2541.
(23) (a) Fu, G. C.; Grubbs, R. H. J. Am. Chem. Soc. 1992, 114, 7324. (b) Fu,
G. C.; Grubbs, R. H. J. Am. Chem. Soc. 1992, 114, 5426.
(24) (a) Fu¨rstner, A. Top. Catal. 1997, 4, 285. (b) Fu¨rstner, A. Angew. Chem.,
Int. Ed. 2000, 39, 3012. (c) Grubbs, R. H.; Miller, S. J.; Fu, G. C. Acc.
Chem. Res. 1995, 28, 446. (d) Schuster, M.; Blechert, S. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 2037. (e) Schwab, P.; France, M. B.; Ziller, J. W.;
Grubbs, R. H. Angew. Chem., Int. Ed. Engl. 1995, 34, 2039. (f) Schwab,
P.; Grubbs, R. H.; Ziller, J. W. J. Am. Chem. Soc. 1996, 118, 100. (g)
Grubbs, R. H.; Chang, S. Tetrahedron 1998, 54, 4413.
9
J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003 2653

A R T I C L E S
Tsang et al.
Scheme 1. Synthesis of W(NAr′)(CHCMe2Ph)Cl2(DME) (Ar′ )
2,6-Me2C6H3)
Ar ) 2,6-i-Pr2C6H3) and W(NAr)(CHCMe2Ph)(OTf)2(DME)
(4), can be synthesized from WCl6 by methods reported in the
literature.27 Only 4 was employed in this work. A complex
related to 3a, W(NAr′)(CHCMe2Ph)Cl2(DME) (3b, Scheme 1,
Ar′ ) 2,6-Me2C6H3), was prepared following a procedure similar
to that used to synthesize 3a. Addition of benzyl potassium28,29
or potassium hydride to H2(Biphen) (where Biphen2- ) 3,3′-
di-t-Bu-5,5′,6,6′-tetramethyl-1,1′-biphenyl-2,2′-diolate) gave the
dipotassium salt, K2(Biphen). W(NAr)(CHCMe2Ph)(Biphen)
(2a) was then prepared by adding K2(Biphen) to 4, while
W(NAr′)(CHCMe2Ph)(Biphen) (2b) was prepared by adding K2-
(Biphen) to 3b.
The racemic complex, rac-2b,30 was obtained free of coor-
dinated THF, while rac-2a was isolated as a THF adduct. (Their
molybdenum analogues were both isolated as THF-free species.26f)
Attempts to synthesize THF-free 2a by carrying out the reaction
between K2(Biphen) and 4 in toluene were unsuccessful.
However, THF-free 2a could be isolated by redissolving the
THF adduct in toluene and removing all solvents in vacuo. The
enantiomerically pure complexes, (S)-2a and (S)-2b, were
prepared in a manner analogous to the methods used to
synthesize the racemic compounds. Unfortunately, each could
be isolated only as a “foam” by removing all solvent in vacuo.
(It is not unusual to find that enantiomerically pure compounds
are much more difficult to crystallize than racemic compounds.)
The enantiopure complexes are essentially free of THF.
Both 2a and 2b exist as mixture of syn and anti isomers in
benzene or toluene, with the syn isomer predominating in each
case. (In the syn isomer, the alkylidene substituent points toward
the imido ligand; in the anti isomer, the substituent points away
Figure 1. Variable-temperature 500 MHz 1H NMR spectra of rac-2a(THF)
in toluene-d8. (Units are ppm; all temperatures are reported in °C.)
resonance was observed at 245.8 ppm (1JWC ) 202.1 Hz, 1JCH
) 113.4 Hz), while a similar downfield resonance was observed
at 247.7 ppm (1JCH ) 115.6 Hz) in the spectrum of 2b. The
1JCH value in each case is characteristic of that for a syn
isomer.14 The anti alkylidene R carbon resonance is too weak
to be observed in the spectrum of either 2a or 2b.
Variable-temperature 1H NMR studies of rac-2a(THF)
(Figure 1) suggest that THF is not strongly bound to the metal.
At 20 °C, a relatively sharp syn-alkylidene proton resonance
(with 183W and 13C satellites) is observed at 7.9 ppm (2JWH
)
1
14.2 Hz; JCH ) 113 Hz). (The 13C satellites are observable
only after many transients.) On the basis of the chemical shifts
of the alkylidene R proton for the THF-free syn isomer of 2a,
the predominant syn isomer of 2a(THF) is close to being THF-
free at 20 °C at the concentrations employed. The anti alkylidene
resonance at 11.15 ppm (weak) is broad and shifted slightly
downfield from where it is found in THF-free 2a. When the
sample was warmed to 80 °C, THF dissociation and intercon-
version of the syn and anti isomers on the NMR time scale led
to a reversible broadening of both resonances. When the sample
was cooled to -80 °C, the alkylidene R proton resonances for
two diastereomeric THF adducts of the anti isomer could be
observed in roughly equal amounts near 11.6 ppm. (Two
diastereomeric THF adducts form as a consequence of THF
coordinating to the diastereotopic CNO faces.) Resonances were
observed at 9.6 ppm (sharp, weak) and ∼9.8 ppm (broad, weak)
for the much more labile diastereomeric syn-THF adducts. At
-80 °C, the amount of the syn isomer is much less than the
amount of anti isomer. As the sample was warmed from -80
to 20 °C, the resonances for the syn diastereomers merged and
1
from the imido ligand.) In the H NMR spectrum of THF-free
2a at 20 °C, the syn alkylidene HR resonance was found at 7.91
ppm (2JWH ) 14.1 Hz), while the broad and weak anti resonance
was located at 10.76 ppm; the syn/anti ratio was 22. The
analogous syn and anti resonances for 2b were found at 7.99
(2JWH ) 16.4 Hz) and 9.06 (2JWH ) 15.0 Hz) ppm, respectively;
the syn/anti ratio was 88. The larger syn/anti ratio for 2b is
consistent with the presence of smaller ortho-substituents in the
arylimido ligand. These alkylidene resonances are found at much
higher field (by ∼3 ppm) than the syn and anti resonances in
1a.26f In the 13C NMR spectrum of 2a, a sharp downfield
(27) 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.
(28) Lochmann, L.; Pospisil, J.; Lim, D. Tetrahedron Lett. 1966, 2, 257.
(29) Lochmann, L.; Trekoval, J. J. Organomet. Chem. 1987, 326, 1.
(30) For simplicity, we will use the rac or S label for the entire complex, even
though it refers to the Biphen2- ligand in the complex.
9
2654 J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003

Alkylidene and Metalacyclic Complexes of Tungsten
A R T I C L E S
Table 1. Some Simple Ring-Closing Reactions Catalyzed by
rac-2a and rac-2ba
Table 2. Catalytic Asymmetric Ring-Closing Reactions with (S)-2a
and (S)-2b
a Solvent ) benzene-d6; 5% catalyst loading (0.01 M). Conversion was
a Solvent ) benzene-d6; 5% catalyst loading (0.01 M). Conversion was
1
1
determined by the analysis of the 500 MHz H NMR spectrum.
determined by the analysis of the 500 MHz H NMR spectrum. The data
for 1a and 1b were obtained from the literature.26f b Product enantio-
selectivity was determined by GLC analysis (Chiraldex-GTA by Alltech
for entries 1-6, Betadex 120 for entries 7-9) in comparison to the authentic
racemic ring-closed product. In all cases, the (R)-antipode was enriched.
shifted upfield as a consequence of the loss of labile THF and
generation of a significant amount of the THF-free syn complex
(Figure 1). The same phenomenon leads to the resonances for
the anti diastereomers broadening and merging, and the average
resonance shifting upfield, although that process takes place at
slightly higher temperatures than for the syn isomer. The syn/
anti ratio increases significantly as the temperature increases.
In the 13C NMR spectrum of 2a at -80 °C in the presence of
5 equiv of THF, the anti alkylidene R carbon resonances were
found at 283.5 (1JCH ) 141 Hz) and 273.1 (1JCH ) 136 Hz)
ppm. The alkylidene R carbon resonances for the syn isomers
under these conditions could not be located with certainty. In
the presence of 36 equiv of THF, the syn/anti ratio at 20 °C
decreased from 22 to 1.4; at -80 °C, the anti-THF adducts
(∼1:1) again were the dominant species. Under these circum-
stances, all four diastereomeric THF adducts could be observed
at or below -60 °C, with the now sharp HR resonance for the
most labile syn adduct appearing at 9.8 ppm. All of this behavior
for 2a is broadly similar to what has been observed for 1a,26f
although it is clear that THF binds more strongly to W than to
Mo, as one might expect.
Ring-Closing Metathesis Reactions and the Observation
of Tungstacyclobutane Complexes. The results of some simple
ring-closing metathesis reactions catalyzed by racemic 2a and
2b are listed in Table 1. (The THF adduct of 2a is noted as
2a(THF).) N,N-Diallylsulfonamide (5) was cyclized slowly but
completely by 2b in 8 h at 22 °C. At elevated temperatures, 5
was fully ring-closed by 2a (75 °C) and 2b (60 °C) in 0.5 h. At
60 °C, substrates 6 and 7 both were ring-closed smoothly by
2b. A slightly higher temperature (75 °C) was required for 2a
relative to 2b (compare entries 4 and 6 in Table 1). For 7 at 75
°C, the rate of conversion is approximately the same whether
the THF-free complex (2a) or the THF adduct (2a(THF)) was
used (compare entries 7 and 8). Substrate 8 could not be ring-
closed completely by 2a(THF) or 2b at 60 °C; a conversion of
approximately 14% was observed after 0.5 h, and the reaction
did not proceed further even upon heating for an additional 1.5
h. Attempts to ring-close 8 with the THF-free complex 2a did
not lead to significant improvement in the conversion. It seems
likely that 8 and the ring-closed product both bind strongly to
the metal and thereby inhibit reaction. However, we cannot
exclude some destructive reaction between either catalyst and
some component of the reaction. In general, 2b seems to afford
higher conversion at lower temperatures than 2a with these
selected substrates.
The results of some preliminary catalytic desymmetrizations
of 9, 10, and 11 by (S)-2b and (S)-2a are summarized in Table
2. The conversion of 10 to product by (S)-2b at 60 °C was
unexpectedly low (30%), and the enantioselectivity (93% ee)
was lower than that observed when (S)-1b was employed (99%
ee).26b (The low conversion did not improve with time.) Both
conversion and enantioselectivity were high with sterically less
hindered substrates 9 and 11. At 75 °C, substrate 11 was ring-
closed to the six-membered product (>98% conversion and
>98% ee) within 5 h. This result is particularly encouraging,
as previous attempts to ring-close 11 by (S)-1b yielded only
51% product (85% ee) and 28% homocoupled product (“dimer”)
even when the reaction was performed at room temperature for
24 h.26b Multiplets characteristic of an unsubstituted tungsta-
cyclobutane complex (vide infra) were observed at 4.02, 3.91,
and 3.83 ppm when this reaction was periodically monitored
by 1H NMR at 20 °C. Substrate 9 was shown to be ring-closed
by (S)-2a at 75 °C (entry 2, Table 2) to marginally greater
conversion and % ee than by (S)-2b (entry 4, Table 2). On the
basis of what has been observed in a molybdenum binaphtholate
system,31 the higher reaction temperatures are needed in general
to eject ethylene from the unsubstituted tungstacyclobutane
resting state and re-form an alkylidene (methylene) which can
then continue the ring-closing reaction.
To identify possible intermediates, the stoichiometric reaction
between rac-2a and 9 (mixed initially at -78 °C) was monitored
by 1H NMR at -40 °C. No alkylidene resonances, except those
(31) Tsang, W. C. P.; Schrock, R. R.; Hoveyda, A. H. Organometallics 2001,
20, 5658.
9
J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003 2655

A R T I C L E S
Tsang et al.
R-protons (H2a and H1b) were observed at 6.21 and 6.37 ppm.
On the basis of the coupling pattern observed in a homonuclear
1H gCOSY experiment, we conclude that the substituents on
the two R carbons are trans to one another. (The absolute
arrangement is not known.) The equatorial â-proton (H3b) is
through-bond coupled to H3a and H1b; its coupling to H2a is too
weak to be observed. The axial â-proton (H3a) is coupled to
1
H3b and H2a; its coupling to H1b is weak. In a H-13C HMQC
experiment, the 13C NMR resonances for the R carbons were
located at 130.5 and 117.4 ppm; the â carbon resonance could
not be located unambiguously. The resonance for the methine
carbon next to the ethereal oxygen in 12 was shifted downfield
to 90.5 ppm as compared with 85.9 ppm in 9. The resonance
for the methylene carbon attached to the metalacyclobutane R
carbon was found at 75.9 ppm.
Complex 12 was moderately stable at -60 °C, although it
decomposed slowly to yield 13 and 3-methyl-3-phenyl-1-butene
at -40 °C (t1/2 ≈ 200 min). Conversion of 12 to 13 was
complete at 0 °C within 15 min, as judged by the irreversible
disappearance of the characteristic R- and â-proton resonances
and the appearance of 3-methyl-3-phenyl-1-butene. On the basis
of extensive NMR studies, we conclude that 13 is the bicyclic
intermediate in the desymmetrization reaction (Scheme 2). The
1
characteristic multiplet at 4.52 ppm in the H NMR spectrum
at -40 °C was assigned to the methylene protons next to the
ethereal oxygen (H4 and H5) on the basis of a correlation
between these protons and the methylene carbon at 75.8 ppm.
The methylene protons in the tungstacyclobutane ring (H2 and
H3) were observed as multiplets at 3.01 and 3.19 ppm. The
resonance for H1 was obscured by the tert-butyl resonance and
a THF resonance at 1.4 ppm. The other methine proton in the
ether ring (H6) was observed as a broad resonance at approxi-
mately 5.0 ppm, which is partly obscured by an olefinic
resonance from 3-methyl-3-phenyl-1-butene. The methine R
carbon resonance was located at 12.2 ppm, while the resonance
for the methine carbon on the ether ring was found at 93.1 ppm.
The methylene R carbon resonance was located at 79.9 ppm by
1H-13C HMQC. Unfortunately, all attempts to obtain single
crystals of 13 have failed.
Figure 2. 1H NMR spectra of the reaction mixture prepared from 2a and
9 in toluene-d8.
Scheme 2. Synthesis of Tungstacyclobutane 14
Complex 13 showed no signs of decomposition after two
weeks in solution at -30 °C, but it decomposed in a first-order
manner (k ≈ 7.3 × 10-5 s-1, t1/2 ≈ 160 min) at 22 °C to form
the final tungstacyclobutane complex 14. In the absence of
excess 9, complex 14 is stable both in solution and in the solid
state. Resonances characteristic of 14 were still observed upon
heating a sample to 90 °C. Spectroscopic data for 14 are
consistent with a tungstacyclobutane formed between the
tungsten methylene complex and the ring-closed product (Figure
2, Scheme 2); its solid-state structure was determined by X-ray
crystallography (vide infra). The four R-protons were observed
as two pairs of doublets at 2.53 (2JHH ) 9.7 Hz), 1.33 (2JHH
)
9.7 Hz), 1.79 (2JHH ) 10.0 Hz), and 1.17 (2JHH ) 10.0 Hz)
ppm in the 1H NMR spectrum. The two R carbons in the
metalacyclic ring were found at 47.9 and 55.1 ppm in the 13C
NMR spectrum, consistent with a square-pyramidal structure.
of unreacted 2a, were detected in these experiments. Two
intermediates (12 and 13) were observed (Scheme 2, Figure 2).
Complex 12 is believed to be the initial metalacycle formed
When 2a was mixed with stoichiometric amounts of 1,6-
heptadiene or 3-allyl-(2,5-dimethyl-1-hexenyl) ether (15), the
ring-closed product and the tungstacyclobutane complex 16 were
observed as the major products (Scheme 3). Complex 16 is
stable in the solid state but relatively unstable in C6D6; an orange
1
from 2a and 9 on the basis of two high field H resonances at
0.34 (H3a) and -0.17 (H3b) ppm (see Figure 2 for numbering
of protons), which are typical for â-protons in trigonal bipyra-
midal tungstacyclobutane complexes.14 Resonances for the two
9
2656 J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003

Alkylidene and Metalacyclic Complexes of Tungsten
A R T I C L E S
Figure 3. ORTEP diagram of 14. Thermal ellipsoids are displayed at 50%
probability level. Hydrogen atoms were omitted for clarity.
Figure 4. ORTEP diagram of 16. Thermal ellipsoids are displayed at 50%
probability level. Hydrogen atoms were omitted for clarity.
Scheme 3. Synthesis of Tungstacyclobutane 16
Table 3. Crystal Data and Structure Refinement for 14 and 16a
14
16
empirical formula
formula weight
WC45H63NO3
849.81
WC50H65NO2
895.88
crystal system
a
b
c
triclinic, P1h
10.1880(16) Å
10.8869(17) Å
21.512(3) Å
75.914(3)
monoclinic, C2/c
22.2675(15) Å
19.8014(14) Å
22.5546(15) Å
90
R
â
80.904(2)
65.334(2)
113.5690(10)
90
γ
volume, Z
density (calculated)
absorption coefficient
F(000)
2098.7(6) Å3, 2
1.345 Mg/m3
2.790 mm-1
876
9115.3(11) Å3, 8
1.306 Mg/m3
2.572 mm-1
3696
θ range for data
collection
2.47 to 23.34
2.09 to 19.99
index ranges
-11 e h e 11
-7 e k e 12
-18 e l e 23
8327
-21 e h e 21
-12 e k e 19
-21 e l e 21
12 979
solution of 16 immediately darkened at 20 °C, and free
3-methyl-3-phenyl-1-butene was observed within 2 h; the metal-
containing decomposition products could not be identified.
Compound 16 was isolated as orange crystals from pentane at
-30 °C and identified in an X-ray study (vide infra). In a
homonuclear 1H gCOSY experiment, resonances for the R-pro-
tons were found at 2.97, 2.56, 1.32, and 1.26 ppm. The two
downfield resonances (δ 2.97, 2.56) were assigned to the
equatorial protons, and the upfield resonances (δ 1.32, 1.26)
were assigned to the axial protons on the ring, consistent with
previously observed chemical shift data in square pyramidal
tungstacyclobutane complexes.14 The â-proton resonance was
found at 3.18 ppm.
reflections collected
independent relections
5863
[Rint ) 0.0321]
96.2%
4250
[Rint ) 0.0635]
99.8%
completeness to θ
max. and min.
transmission
0.5958 and 0.3937
0.4624 and 0.3419
data/restraints/
5863/0/452
4250/0/482
parameters
goodness-of-fit on F 2
final R indices
[I > 2σ(I)]
1.120
1.427
R1 ) 0.0280,
wR2 ) 0.0699
R1 ) 0.0292,
wR2 ) 0.0705
0.0018(3)
1.304 and
-0.875 e Å-3
R1 ) 0.0769,
wR2 ) 0.1268
R1 ) 0.0878,
wR2 ) 0.1298
0.00000(2)
1.026 and
R indices (all data)
extinction coefficient
largest diff. peak
and hole
-1.651 e Å-3
X-ray Studies of Tungstacyclobutanes 14 and 16. Red
crystals of 14 suitable for X-ray diffraction were obtained in a
reaction between rac-2a and 9 in a concentrated pentane solution
at room temperature. Orange crystals of 16 were prepared in
the reaction between rac-2a and 15 in pentane at -20 °C.
ORTEP diagrams of 14 and 16 are shown in Figures 3 and 4.
Crystal data can be found in Table 3, and selected bond distances
and angles are given in Table 4. Only one diastereomer of 14
was present, consistent with the observation of only one
diastereomer in solution in NMR studies.
a Both experiments were carried out at 183(2) K using Mo KR radiation
(0.71073 Å), the absortpion correction was empirical, and the refinement
method was full-matrix least-squares on F 2.
sites. The octahedral structure of 14 can be viewed as a related
“solvated” square pyramid in which O(3) is coordinated trans
to the imido nitrogen. In each compound, the C(1)-W-C(2)
angle is relatively small (∼64°). The tungsten atom resides
above the least-squares plane defined by O(1), O(2), C(1), and
C(2) by 0.47 Å in 14 and 0.65 Å in 16. Carbon C(3) points
away from the imido ligand in 14 as a consequence of
intramolecular coordination of the dihydrofuran side chain to
the metal through O(3). In 16, C(3) points toward the imido
Complex 16 adopts approximately a square pyramidal
geometry in which the two biphenolate oxygen atoms and the
two metalacyclobutane R carbon atoms occupy the four basal
9
J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003 2657

A R T I C L E S
Tsang et al.
Table 4. Selected Bond Distances (Å) and Angles (deg) in
Tungstacyclobutanes 14 and 16
14
16a
W(1)-C(1)
2.204(4)
2.160(4)
1.739(3)
1.936(3)
1.975(3)
2.381(3)
1.541(6)
1.526(6)
91.8(2)
2.157(14)
2.173(15)
1.684(12)
1.917(8)
1.983(8)
W(1)-C(2)
W(1)-N(1)
W(1)-O(1)
W(1)-O(2)
W(1)-O(3)
C(1)-C(3)
1.520(19)
1.53(2)
95.7(9)
C(2)-C(3)
W(1)-C(1)-C(3)
W(1)-C(2)-C(3)
W(1)-N(1)-C(10)
W(1)-O(1)-C(22)
W(1)-O(2)-C(33)
C(1)-W(1)-C(2)
C(1)-C(3)-C(2)
N(1)-W(1)-C(1)
N(1)-W(1)-C(2)
N(1)-W(1)-O(1)
N(1)-W(1)-O(2)
N(1)-W(1)-O(3)
O(1)-W(1)-C(1)
O(1)-W(1)-C(2)
O(1)-W(1)-O(2)
O(1)-W(1)-O(3)
O(2)-W(1)-C(1)
O(2)-W(1)-C(2)
O(2)-W(1)-O(3)
C(22)-C(27)-C(28)-C(33)
94.0(3)
94.8(9)
176.0(3)
125.1(2)
112.9(2)
63.75(17)
97.4(3)
168.0(10)
128.4(7)
106.7(7)
63.8(6)
97.2(11)
101.8(5)
97.8(6)
102.79(16)
97.52(17)
104.81(14)
107.30(14)
168.63(13)
151.13(14)
104.04(15)
98.89(11)
80.68(10)
80.71(14)
140.42(14)
81.26(10)
78.8(5)
118.2(4)
111.1(4)
Figure 5. Variable-temperature 1H NMR (500 MHz) spectra of W(NAr′)-
(CH2CH2CH2)(Biphen) (18b) in toluene-d8. (Units are ppm; temperatures
are reported in °C.)
137.0(5)
94.2(5)
99.5(3)
butane complex, W(NAr′)(C3H6)(Biphen) (TBP-18b; eq 1), on
the basis of a comparison of these chemical shift values with
those in unsubstituted metalacyclobutane complexes in the
literature.14 (Minor amounts of other compounds were detected
78.6(4)
136.5(6)
81.1(16)
a Atom numbers were changed to correspond to the atom labels in 14.
ligand. The dihedral angle between the planes defined by C(1),
W, and C(2) and that by C(1), C(3), and C(2) is 37° in 14 and
29.5° in 16 (cf. 25° in W[CH(t-Bu)CH2CH(CO2Me)](NAr)-
[OCMe2(CF3)]2 (17)32). Metal-ligand bond lengths and angles
around the metal are not unusual, and the C-C bond lengths
(1.541(6) and 1.526(6) Å in 14, 1.520(19) and 1.53(2) Å in 16)
in the tungstacyclobutane ring are similar to what have been
observed in other square pyramidal tungstacyclobutane com-
plexes.14 The dihedral angles in the biphenolate ligands are 78.8-
(5)° in 14 and 81.1(16)° in 16, which are similar to angles that
have been observed in five-coordinate molybdenum biphenolate
or binaphtholate alkylidene complexes.26c,f,33,34 The structure of
14 closely resembles that of 17;32 in both complexes, the
tungsten atom adopts a distorted square pyramidal geometry
with an additional oxygen donor binding weakly trans to the
apical imido ligand. The W-O(3) bond lengths in 14 (2.381-
(3) Å) and 17 (2.372(6) Å) are essentially identical. The angles
around tungsten and within the tungstacyclobutane ring are
similar in 14 and 17, although 14 contains a bidentate biphe-
nolate ligand.
that were characterized by employing 13C2H4, as discussed in
the following section.) Five of the six proton resonances are
shown in the spectrum at 0 °C in Figure 5; the one near 3.0
ppm is not shown. Each has a relative intensity of one proton.
All resonances split into two when 13C2H4 was employed. All
six protons on the tungstacyclobutane ring could be observed
1
as discrete contour circles in the H-13C HMQC spectrum at
20 °C, and each carbon was through-bond coupled to two
protons. The proton NMR data for TBP-18b at 20 °C are listed
in Table 5. When a toluene-d8 sample of 18b was cooled (Figure
5), the metalacycle proton resonances first broadened and then
shifted, but sharpened again at -80 °C. Several resonances were
shifted dramatically at -80 °C; only four are shown (each with
relative area 1) in Figure 5. Complex 18b was stable enough to
be precipitated in pentane from a mixture of 2b and 4 equiv of
1,6-heptadiene. (Note that the addition of only 1 equiv of 1,6-
heptadiene to 2a allowed 16 to be isolated, as described earlier.)
In the solid state, 18b is relatively stable under dinitrogen and
could be analyzed satisfactorily.
Three broadened resonances centered at 86.9, 79.4, and 3.7
ppm were observed in the 13C spectrum of a 13C-labeled sample
of 18b at 20 °C (Figure 6); carbon-proton coupling constants
were found to be 136, 149, and 149 Hz, respectively. Upon
cooling the sample to -80 °C, we found that these resonances
broadened, shifted, and sharpened to yield sharp resonances at
99.2, 87.4, and -3.2 ppm with JCH values of 155, 156, and 152
Observation of Unsubstituted Metalacyclobutane Com-
plexes. Because ethylene is formed to a greater or lesser extent
in all ring-closing reactions that have been explored here, we
became interested in exploring reactions between 2a or 2b and
ethylene. When 4 equiv of ethylene was added to a sample of
rac-2b in C6D6, several well-defined resonances were observed
1
in the H NMR spectrum that could be attributed to the six
protons in an unsubstituted trigonal bipyramidal tungstacyclo-
(32) Feldman, J.; Murdzek, J. S.; Davis, W. M.; Schrock, R. R. Organometallics
1989, 8, 2260.
(33) Totland, K. M.; Boyd, T. J.; Lavoie, G. G.; Davis, W. M.; Schrock, R. R.
Macromolecules 1996, 29, 6114.
(34) Schrock, R. R.; Jamieson, J. Y.; Dolman, S. J.; Miller, S. A.; Bonitatebus,
P. J., Jr.; Hoveyda, A. H. Organometallics 2002, 21, 409.
9
2658 J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003

Alkylidene and Metalacyclic Complexes of Tungsten
A R T I C L E S
Table 5. Chemical Shift and Coupling Values for Unsubstituted
Tungstacyclobutane Complexes
complex is essentially a pure TBP species. As the sample is
warmed, increasing amounts of a square pyramidal species forms
(SP-18; eq 2) and the two metalacycles begin to interconvert
on the NMR time scale starting at approximately -60 °C. Free
13C δ (ppm)
1H δ (ppm)
1JCH (Hz)
W(NAr′)(C3H6)(biphen)
86.9 (99.2c)
79.4 (87.4c)
3.7 (-3.2c)
4.02, 3.91
3.83, 3.00
0.70, 0.10
136 (155c)
149 (156c)
149 (152c)
18ba
W(NAr)(C3H6)(biphen)
73.5 (99.1c)
69.1 (86.5c)
9.0 (-2.6c)
3.39, 3.20
3.49, 2.47
1.77, 1.10
147 (153c)
145 (151c)
143 (147c)
18ab
a All values are reported at 20 °C in C6D6 unless otherwise stated; Ar′
) 2,6-Me2C6H3. b All values are reported at 20 °C in toluene-d8 unless
otherwise stated; Ar ) 2,6-i-Pr2C6H3. c Data are reported at -80 °C in
toluene-d8.
ethylene is not involved in the temperature-dependent process.
The square pyramidal form cannot be observed at low temper-
ature, as too little is present, or above -60 °C, as it is
interconverting with the TBP form. Above -30 °C, the three
metalacycle carbon resonances (Figure 6) sharpen and shift
toward the 20-40 ppm region, where resonances for R and â
carbon atoms in unsubstituted square pyramidal metalacycles
are found.14 On the basis of the average chemical shifts at 20
°C, we estimate that roughly 25% of the SP form is present in
the mixture at 20 °C. At 20 °C, the average chemical shifts of
the three broad labeled metalacycle carbon resonances of 18a
are nearly 10 ppm closer to the 20-40 ppm region than the
resonances in 18b, suggesting a more significant contribution
of SP-18a to the interconverting mixture of tungstacyclobutane
complexes (TBP-18a and SP-18a). It is not possible to obtain
spectra at temperatures significantly above 20 °C, as the
metalacycles decompose (vide infra). It should be noted that
the two R carbons and the â carbon retain their identities when
the two metalacycle isomers are interconverting rapidly, which
is inconsistent with interconversion via loss of ethylene.
A second possible explanation for the temperature dependence
is that the TBP metalacycle is in equilibrium with a distinct
ethylene/methylene complex (eq 3). This possibility must be
Figure 6. Variable-temperature 13C NMR (125 MHz) spectra of W(NAr′)-
(CH2CH2CH2)(Biphen) (18b) in toluene-d8. *JCH values were determined
from the 1H-coupled 13C NMR spectrum. (Units are ppm; all temperatures
are reported in °C.)
Hz. The positions of the three sharp resonances observed at
-80 °C are characteristic of a trigonal bipyramidal (TBP)
tungstacyclobutane complex.14 Unfortunately, the resonances at
-80 °C were still too broad to observe 13C-13C coupling. On
the basis of previous studies of tungstacyclobutane complexes,14
we assign the two downfield resonances to the R carbons and
the upfield resonance to the â carbon in a TBP species. A
resonance for free ethylene could be observed at 123 ppm in
all 13C NMR spectra; therefore, ethylene is not a part of the
temperature-dependent process observed in Figure 6.
considered in view of the fact that high oxidation state alkene/
alkylidene complexes have been observed in cationic tungsten
systems explored by Osborn and Kress.35 If a methylene/
ethylene complex were an intermediate, the ethylene could not
rotate by 180° at a rate of the order of the NMR time scale at
20 °C, or else CR and Câ would exchange. On the other hand,
a nonrotating ethylene seems unlikely, because cycloheptene
rotation was found to be rapid above 240 K in the tungsten
system studied by Osborn and Kress.35 Therefore, we prefer
the explanation that the temperature dependence involves the
formation of increasing amounts of a SP tungstacyclobutane
complex and the interconversion of it with the TBP form.
Other Species Observed in Reactions between 2a or 2b
and 13C-Labeled Ethylene. Several other species are present
When a sample of 2a in toluene-d8 was exposed to 2.5 equiv
of 13C2H4, an unsubstituted tungstacyclobutane complex analo-
gous to 18b, W(NAr)(C3H6)(Biphen) (18a), was observed. Three
broadened resonances centered at 73.5, 69.1, and 9.0 ppm were
observed in the 13C spectrum at 20 °C. A list of all of the
1
correlated protons of 18a, obtained from a H-13C HMQC
experiment, is summarized in Table 5. Temperature-dependent
behavior similar to that of 18b was observed when 18a was
cooled to -80 °C. The chemical shifts (99.1, 86.5, -2.6 ppm)
and JCH values (153, 151, 147 Hz) of the three sharp resonances
observed at -80 °C are characteristic of a TBP tungstacyclo-
butane complex.14
We ascribe the temperature dependence in the proton and
the carbon NMR spectra of 18a and 18b to the same phenom-
enon. At the lowest temperature, each tungstacyclobutane
(35) Kress, J.; Osborn, J. A. Angew. Chem., Int. Ed. Engl. 1992, 31, 1585.
9
J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003 2659

A R T I C L E S
Tsang et al.
Figure 7. Observation of a mixture of 18a, 19a, and 20a generated by
treating 2a with 1 or 1.5 equiv of 13C2H4.
in reactions in which 18a and 18b are observed. They can be
identified most readily by employing 13C2H4.
When 1 equiv of 13C2H4 was added to a sample of rac-2a, a
singlet was observed for the 13C label in the initial cleavage
product, PhMe2CCHd13CH2, along with resonances for two
other compounds (Figure 7). The major species gives rise to a
Figure 8. 13C spectrum (125 MHz) of labeled 20a showing two sets of
13C satellites for the methylene carbon atom.
relatively large chemical shift and low 1JWC values are consistent
with that species being a dimeric complex, [W(NAr)(Biphen)-
(CH2)]2 (20a; Figure 7 and Scheme 4), that contains either a
C2 axis or a center of inversion that makes the two methylenes
equivalent. In Scheme 4, compound 20 has been drawn with
the methylene behaving as a “donor” toward the second tungsten
on one (face “A”) of the two possible CNO faces, where donors
are known to bind. The left half of the molecule contains the
(S)-Biphen2- ligand and the methylene coordinated through face
A, while the right half of the molecule contains the (R)-Biphen2-
ligand and the methylene coordinated through face B. As drawn,
20 is a heterochiral species that contains an inversion center
that makes the two methylene carbons equivalent. Proposals
concerning the mechanism of formation of 19a and other species
(Scheme 4) will be deferred to the discussion section. The
mixture of 19a and 20a was stable for at least 12 h. Because
dimeric methylene complexes of this general type have never
been observed before, it is difficult to know if the different
couplings to tungsten (78.9 and 36.7 Hz) in 20a are characteristic
of a dimeric methylene complex and therefore if they would
be present even if two symmetric alkoxide ligands were present
on each metal, or whether they are a consequence solely of the
unsymmetric nature of the biphenolate ligands in 20a.
1
1
set of two doublets at 55.4 (1JWC ) 59.0, JCC ) 6.9, JCH
)
1
1
132 Hz) and 51.3 (1JWC ) 46.1, JCC ) 6.9, JCH ) 133 Hz)
ppm. Both resonances can be ascribed to methylene carbons
that are through-bond coupled to protons at 2.93 (δ 55.4), 2.48
(δ 51.3), 1.23 (δ 51.3), and 1.21 (δ 55.4) ppm, according to a
1H-13C HMQC experiment. On the basis of comparison with
a crystallographically characterized ethylene complex of this
general type and extensive NMR studies of that species,36 these
doublets are assigned to the ethylene complex, 19a (Scheme
4). (In Scheme 4, the compound number is a generic number
that refers to the species derived either from 2a or from 2b. All
compounds except 20 are drawn with the (S)-Biphen ligand,
although both enantiomers are present.) Note that for 19a to be
formed quantitatively, 1.5 equiv of ethylene would be required.
The second (minor) compound gives rise to a singlet at 185.9
1
ppm. In the H-coupled 13C spectrum, this resonance becomes
two overlapping doublets as a consequence of that carbon being
coupled to two inequivalent protons by different amounts (1JCH
) 148, 131 Hz). The two correlated proton resonances were
located at 8.20 and 7.37 ppm; each was split into two resonances
because of 13C labeling. Importantly, two sets of tungsten
satellites were observed (1JWC ) 78.9, 36.7 Hz) in the
1H-decoupled 13C spectrum (Figure 8), which is possible only
if the carbon is coupled to two different 183W nuclei. The
When 0.92 equiv of 13C2H4 was added to a sample of (S)-
2a, almost exclusively 19a was observed. In particular, no
resonance for 20a was obserVed. The failure to observe any
20a in the enantiomerically pure system leads us to conclude
(36) The isolated ethylene complex that has been crystallographically character-
ized is Mo(N-2,6-Cl2C6H3)(C2H4)(Biphen) (Jamieson, J.; Aeilts, S.; Tsang,
P., unpublished results). These results will be published in due course.
9
2660 J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003

Alkylidene and Metalacyclic Complexes of Tungsten
A R T I C L E S
Scheme 4. A Plausible Relationship between Compounds Formed from Ethylene
that the 20a observed in the racemic system is, in fact, a
heterochiral dimer. (See Discussion section.)
When 1.5 equiv of 13C2H4 was added to a sample of rac-2a,
broad resonances characteristic of tungstacyclobutane complex
18a were observed in addition to resonances for 19a and a trace
of 20a (Figure 7).
of propylene can be generated, while 1-butene can be generated
catalytically.
Introduction of >3 equiv of 13C2H4 into a sample of 2a led
immediately to the formation of some 18a, along with propylene,
1-butene, and 21a. Thereafter, ethylene was slowly and catalyti-
cally dimerized to 1-butene. The resonances for 20a remained
weak throughout this dimerization process.
When 2.5 equiv of 13C2H4 was added to 2a in toluene-d8,
largely 18a was observed. With time, 18a appears to decompose,
and propylene (with resonances at 134, 116, and 20 ppm),
1-butene (with resonances at 141, 114, 27, 14 ppm), and a
compound that is characterized by two small sets of doublets
When a sample of 2b in C6D6 was exposed to 2-4 equiv of
13C2H4, the 18b that was formed was found to be stable for at
least 12 h at room temperature; resonances characteristic of free
propylene were observed to grow in slowly (over a period of
days) thereafter. However, propylene was observed immediately
when 18b was prepared using >4 equiv (e.g., 6 or 8 equiv) of
labeled ethylene. After 24 h at 20 °C, a mixture of 18b,
propylene, 1-butene, and an unsubstituted tungstacyclopentane
complex (21b) was clearly observed (Scheme 4). Two sets of
doublets, which can be ascribed to the two R carbons of 21b,
1
at 78 (1JCC ) 35.4 Hz, JCH ) 126 Hz) and 77 (1JCC ) 34.7
1
Hz, JCH ) 127 Hz) ppm and a non first-order multiplet at 37
ppm are formed. The doublets and multiplet are ascribed to an
unsubstituted tungstacyclopentane complex (21a; Scheme 4).
1-Butene appeared to form more rapidly than propylene,
although the amounts were comparable. A maximum of 1 equiv
9
J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003 2661

A R T I C L E S
Tsang et al.
Table 6. Chemical Shift and Coupling Values for Tungsten
Methylene Dimer, Ethylene Adduct, and Unsubstituted
Metalacyclopentane Complexes
3-phenyl-1-butene and formation of 13. We also were surprised
to find that by simply raising the temperature of a metathesis
reaction that even unsubstituted metalacycles would break up
to give an alkylidene at a rate fast enough to make metathesis
by tungsten viable.
13C δ (ppm) 1H δ (ppm)
1JCC (Hz)
1JCH (Hz)
[W(NAr)(CH2)(biphen)]2
185.9
8.20, 7.37
148, 131
20aa,b
Although several unsubstituted molybdacyclobutane com-
plexes have been observed previously, none has been stable
enough to isolate.17,37 The molybdacycle formed upon treating
Mo(NAr)(CH-t-Bu)[OCMe(CF3)2]2 with ethylene is stable at
25 °C under ethylene for hours, but it decomposes over a period
of ∼12 h under ethylene to yield a molybdacyclopentane
complex, Mo(NAr)[OCMe(CF3)2]2(C4H8) (δ CR ) 76.8 ppm,
δ Câ ) 38.6 ppm, JCRCâ ) 32 Hz, JCRH ) 128 Hz, JCâH ) 141
Hz; cf. 79.9, 76.0, 38.5, and 37.2 ppm in 21b; Table 6).17 On
the basis of the chemical shifts for the R carbon atoms at 104.1
ppm and the â carbon atom at -2.28 ppm, and by analogy with
crystallographically characterized tungstacyclobutane com-
plexes,14 Mo(NAr)[OCMe(CF3)2]2(C3H6) was proposed to have
a trigonal bipyramidal structure. Treatment of Mo(NAr)(CH-
t-Bu)(OAr)2 (Ar ) 2,6-i-Pr2C6H3) with ethylene yields primarily
(95%) square pyramidal Mo(NAr)(OAr)2(C3H6) (δ CR ) 39.9
ppm, δ Câ ) 26.5 ppm) mixed with 5% trigonal bipyramidal
Mo(NAr)(OAr)2(C3H6) (δ CR ) 100.1 ppm, δ Câ ) -0.7 ppm).
Addition of ethylene to Mo(NAr)(CH-t-Bu)(O-t-Bu)2 was
proposed to yield a third molybdacyclobutane complex, square
pyramidal Mo(NAr)(O-t-Bu)2(C3H6) (δ CR ) 34.9 ppm, δ Câ
) 29.1 ppm, JCC ) 32 Hz, JCH ) 140 and 129 Hz) in low yield
(∼25%).37 Mo(NAr)(O-t-Bu)2(C3H6) decomposed to yield crys-
tallographically characterized [Mo(NAr)(O-t-Bu)2]2, in which
the imido ligands bridge symmetrically between the metals.
Molybdacyclobutane complexes have also been observed
recently in binaphtholate imido systems.31 Evidence suggested
that these molybdacyclobutane complexes decomposed (hours
to days) in two ways, either by rearrangement of the metalacycle
to yield propylene or by bimolecular decomposition of the
methylene complex formed upon loss of ethylene from the
metalacycle. The decomposition product was not identified in
this case. It is important to note that decomposition appeared
to be accelerated in the presence of ethylene (possibly by
ethylene-induced â hydride rearrangement of the metalacycle)
and in the absence of ethylene (by loss of ethylene from the
metalacycle and decomposition of the resulting methylene
species).
W(NAr)(C2H4)(biphen)
55.4c
51.3c
2.93, 1.21
2.48, 1.23
6.9
6.9
132
133
19ac
W(NAr′)(C2H4)(biphen)
56.2
51.2
2.92, 1.22
2.60, 1.26
7.0
7.0
132
135
19bd
W(NAr′)(C4H8)(biphen)
79.9
76.0
38.5
37.2
3.43, 2.93 34.5
3.39, 3.04 34.3
3.15, 3.15 34.5, 33.9 126
3.10, 2.99 34.3, 33.9 125
125
125
21bd
a All values are reported at 20 °C in toluene-d8 unless otherwise stated;
b 1
1
Ar ) 2,6-i-Pr2C6H3.
J
values were determined from 13C NMR; JWC
WC
c 1
) 78.9, 36.7 Hz.
J
) 59.0 (δ 55.4), 46.1 (δ 51.3) Hz. d All values are
WC
reported at 20 °C in C6D6 unless otherwise stated; Ar′ ) 2,6-Me2C6H3.
were observed at 79.9 (1JCC ) 34.5 Hz, JCH ) 125 Hz) and
1
76.0 (1JCC ) 34.3 Hz, 1JCH ) 125 Hz) ppm. Two sets of triplets
which can be ascribed to the two â carbons of 21b were
observed at 38.5 (1JCC ) 34.5, 33.9 Hz, JCH ) 126 Hz) and
1
37.2 (JCC ) 34.3, 33.9 Hz, 1JCH ) 125 Hz) ppm. Homonuclear
1
13C gCOSY and H-13C HSQC experiments completely sup-
ported the tungstacyclopentane proposal; all data are gathered
in Table 6. Upon close inspection of the 13C NMR spectrum,
two small doublets, which can be ascribed to a tungsten ethylene
adduct 19b, were observed at 56.2 (1JCC ) 7.0 Hz, 1JCH ) 132
Hz) and 51.2 (1JCC ) 7.0 Hz, JCH ) 135 Hz) ppm. The
1
chemical shifts of the correlated protons are very similar to those
observed for ethylene adduct 19a (Table 6).
In the sample in which 2b was treated with 13C2H4, the
resonances for 1-butene slowly intensified at the expense of
that for ethylene. Warming the reaction mixture to 60 °C for
10 h led to the disappearance of 18b and an increase in the
intensity of propylene, 1-butene, and 21b with respect to the
solvent internal standard; a large decrease in the intensity of
the ethylene resonance was also observed. Some 21b was still
observable after all of the solvent and ethylene were removed
and fresh C6D6 was added. Upon standing this solution at room
temperature for 24 h, we observed a small amount of labeled
ethylene. Warming the sample at 60 °C for 8 h resulted in a
decrease in the intensity of the resonances for 21b and a small
increase in the amount of free ethylene, although 21b was still
the major labeled product.
One of the interesting results reported here is the observation
of 20. In Scheme 4, 20 is drawn as a heterochiral dimer; that
is, in the left half, the Biphen2- ligand has the S configuration,
and the methylene “donor” is bound to one of the two possible
CNO faces (say face A), while in the right half, the Biphen2-
ligand has the R configuration, and the methylene “donor” is
bound to the other possible CNO face (B). As drawn, 20 then
has the configuration “SABR” and an inversion center. The
homochiral dimer would have the configuration SAAS and a
C2 axis that passes through the W2C2 ring. Other possible dimers
in a racemic system would have configurations SABS, SABR,
etc., and in them the methylene carbons would not be equivalent.
The proposal that 20 is a heterochiral dimer is consistent with
the following: (i) donors are known to bind to Mo complexes
through a CNO face in a manner that is determined solely by
Discussion
We were not surprised by the stability of tungstacyclobutane
complexes in general, at least relative to molybdacyclobutane
complexes, as this is the trend that has been observed for imido
bisalkoxide complexes of molybdenum and tungsten that have
been investigated previously.14 However, we were surprised by
the relatively high stability of what must be highly strained 13
(Figure 2). We have assumed in the Mo-catalyzed ring-closing
of 9 that the molybdacycle analogous to 13 is the least stable
metalacycle in the catalytic cycle.26f That is no longer likely to
be the case. It is easier to understand why 12 is relatively
unstable in view of the presence of two R substituents. In a
species in which the CMe2Ph group points toward the imido
ligand, steric interaction between the diisopropylphenylimido
group and the CMe2Ph group would encourage loss of 3-methyl-
(37) Robbins, J.; Bazan, G. C.; Murdzek, J. S.; O’Regan, M. B.; Schrock, R. R.
Organometallics 1991, 10, 2902.
9
2662 J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003

Alkylidene and Metalacyclic Complexes of Tungsten
A R T I C L E S
the chirality of the biphenolate or binaphtholate ligand in the
six known adducts;34 (ii) only one methylene carbon atom
resonance is observed in 20; and (iii) 20 was not observed when
reactions were carried out with (S)-2a. Therefore, we suspect
that the homochiral analogue of 20 is unstable and, in particular,
it is unstable with respect to decomposition to yield 19 and “W-
(NAryl)(Biphen).”
It is important to understand the details of the reaction
between 2a or 2b and ethylene and how various species that
are formed in the presence of ethylene decompose, however
qualitative those details might be at this stage. The following
is only one of several consistent scenarios, but it provides a
point of departure and a basis for discussion and future
investigations. In the ethylene reactions, there will be issues
concerning the mixing of ethylene throughout the solution versus
reactions in one portion of the solution. These issues will have
to be ignored for now, along with issues concerning the TBP
or SP nature of various metallacycles, syn versus anti isomer
interconversion, and relative rates of reaction of syn and anti
isomers.
We propose that the (unobserved) initial metalacycle that is
formed upon reaction of 2 (a or b) with ethylene readily loses
3-methyl-3-phenyl-1-butene to give PhMe2CCHdCH2 and
unobservable monomeric W(NAryl)(Biphen)(CH2) (Scheme 4).
W(NAryl)(Biphen)(CH2) then reacts bimolecularly with itself
to give either heterochiral 20 (perhaps entirely reversibly) or to
give 19 and “W(NAryl)(Biphen)” irreversibly via homochiral
20 (not observable). “W(NAryl)(Biphen)” then rapidly scav-
enges any remaining ethylene that is present to yield more 19.
The mechanism of formation of 19 is analogous to that proposed
for the formation of Cp2Ta(CH2CH2)Me upon decomposition
of Cp2Ta(CH2)Me in the presence of ethylene.38
In the presence of a larger amount of ethylene, much of the
W(NAryl)(Biphen)(CH2) is captured by ethylene to give 18.
Compound 18 is quite stable toward rearrangement to give
propylene in the absence of ethylene. Although it can lose
ethylene to regenerate W(NAryl)(Biphen)(CH2), that process
is relatively slow at the temperatures employed in the NMR
experiments, and the equilibrium lies far toward 18. We might
propose that intramolecular rearrangement of a tungstacyclo-
butane complex would proceed via an intermediate allyl hydride
(Scheme 4) that is in equilibrium with 18. We believe that it is
this allyl hydride that is transformed into 19 and propylene (up
to 1 equiv) upon reaction with ethylene. This is the reason more
propylene is generated if a relatively large amount of ethylene
is added initially. This proposal is analogous to that published
recently in which acceleration of â hydride rearrangement of
unsubstituted molybdacyclobutane complexes has been observed
in similar types of complexes.31
termediate has also been proposed in the relatively efficient
dimerization of ethylene to 1-butene when ethylene is added to
Ta(CH-t-Bu)2(CH2-t-Bu)(PMe3)2.44 Therefore, it is not surprising
that tungstacyclopentane complexes can rearrange to give
1-butene. What we do not know is whether the formation of
1-butene from 21 is also accelerated in the presence of ethylene.
How rapidly the reactions in Scheme 4 and analogous
reactions involving other alkylidenes lead to catalyst decomposi-
tion in real catalytic reactions in which ethylene is formed is
not known at present. The answer is likely to depend on
conditions. If it is true that ethylene in fact promotes decom-
position of 18, then carrying out metathesis reactions in which
ethylene is generated in a closed reaction vessel would lead to
low rates (because little W(NAryl)(Biphen)(CH2) is available
from 18) and catalyst decomposition (by ethylene-induced
decomposition of 18). It should be noted that most of the
ethylene would be evolved from solution in a vented reaction
vessel, especially if that reaction is heated. Therefore, heating
tungsten-catalyzed reactions in vessels from which ethylene can
escape might turn out to be relatively beneficial in the long run
for long-lived metathesis by tungsten complexes. Compound
18 is clearly a central feature of tungsten-catalyzed metathesis
chemistry; it is essentially a way of sequestering W(NAryl)-
(Biphen)(CH2) in a usable form and preventing bimolecular
decomposition, if rearrangement of 18 to give propylene is slow
under the reaction conditions relative to loss of ethylene to
regenerate W(NAryl)(Biphen)(CH2).
The tungsten systems examined here and the analogous
molybdenum systems show many similarities. In both Mo and
W systems, rearrangement of unsubstituted metalacyclobutane
complexes to yield propylene appears to be faster in the
sterically more crowded NAr complexes and possibly acceler-
ated in the presence of ethylene. Two notable differences
between tungsten and molybdenum that have surfaced so far
are that 20 can be observed in the tungsten system and that
metalacyclopentane complexes formed in the tungsten systems
act as catalysts for the (slow) formation of 1-butene. It should
be noted that if 1-butene builds up before the metathesis reaction
of interest is complete, then the desired reaction could be
complicated to a significant degree by metathesis reactions that
involve 1-butene.
We assume that no alkylidene can be re-formed once 19 or
21 forms and that catalytic metathesis therefore ceases. How-
ever, it is now known that an olefin complex can rearrange to
an alkylidene in high oxidation state tantalum45-47 and niobium48
chemistry. Therefore, we cannot exclude the possibility com-
pletely at this stage that some tungsten olefin complexes can
rearrange to alkylidene complexes under certain conditions.
Finally, we do not know the nature of the final decomposition
product or products in the tungsten metathesis systems. At this
Compound 19 can react with ethylene to give 21, possibly
in an equilibrium that lies largely toward 21, and 21 can
rearrange slowly to give “W(NAryl)(Biphen)(1-butene)”, from
which 1-butene is displaced by ethylene, etc., thereby allowing
1-butene to be formed catalytically. Documented examples of
alkene dimerization via a metalacyclopentane intermediate are
rare. Perhaps the best examples are monocyclopentadienyl
tantalacyclopentane complexes.39-43 A metalacyclopentane in-
(41) McLain, S. J.; Sancho, J.; Schrock, R. R. J. Am. Chem. Soc. 1979, 101,
5451.
(42) McLain, S. J.; Sancho, J.; Schrock, R. R. J. Am. Chem. Soc. 1980, 102,
5610.
(43) Schrock, R. R.; McLain, S. J.; Sancho, J. Pure Appl. Chem. 1980, 52, 729.
(44) Fellmann, J. D.; Rupprecht, G. A.; Schrock, R. R. J. Am. Chem. Soc. 1979,
101, 5099.
(45) Hughes, R. P.; Maddock, S. M.; Rheingold, A. L.; Guzei, I. A. Polyhedron
1998, 17, 1037-1043.
(46) Freundlich, J.; Schrock, R. R.; Cummins, C. C.; Davis, W. M. J. Am. Chem.
Soc. 1994, 116, 6476.
(38) Schrock, R. R.; Sharp, P. R. J. Am. Chem. Soc. 1978, 100, 2389.
(39) McLain, S. J.; Schrock, R. R. J. Am. Chem. Soc. 1978, 100, 1315.
(40) McLain, S. J.; Wood, C. D.; Schrock, R. R. J. Am. Chem. Soc. 1979, 101,
4558.
(47) Freundlich, J. S.; Schrock, R. R.; Davis, W. M. J. Am. Chem. Soc. 1996,
118, 3643.
(48) Veige, A. S.; Wolczanski, P. T.; Lobkovsky, E. B. Angew. Chem., Int. Ed.
2001, 40, 3629.
9
J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003 2663

A R T I C L E S
Tsang et al.
stage, we assume that a W(IV) complex that contains bridging
imido ligands is formed from “W(NAryl)(Biphen)”, as has been
documented in one molybdenum system, and that this dimeric
species does not readily react with ethylene to give 19.37 We
also do not know how reactions involving other types of alkenes
or donor functionalities might alter decomposition pathways.
Both tungsten(VI)49 and molybdenum(VI)50,51 phenylimido
dineopentyl and neopentylidene complexes (as PMe3 adducts)
have been prepared that contain bisamido ligands of the type
[o-(Me3SiN)2C6H4]2-. The tungsten dineopentyl complex will
react slowly with ethylene at temperatures above 70 °C to give
neopentane (from the initial R hydrogen abstraction reaction),
tert-butylethylene (from the reaction of the intermediate neo-
pentylidene complex with ethylene), and a structurally charac-
terized (distorted square pyramidal) tungstacyclopentane com-
plex with carbon resonances at 61.25 (CR) and 35.23 ppm (Câ).
These chemical shifts are similar to what we have found in the
complexes described here. In reactions between W(NPh)[o-(Me3-
SiN)2C6H4](CH-t-Bu)(PMe3) and ethylene, the initial R tert-
butyl-substituted tungstacyclobutane complex was observed in
solution, as were the unsubstituted tungstacyclobutane and
ethylene complexes. All were converted into the tungstacyclo-
pentane complex with time in the presence of ethylene. These
results are similar to what we propose in Scheme 4, although
the reactions of the bisamido complexes are much slower than
those in the biphenoxide complexes as a consequence of
deactivation of the metal by σ and π electron donation from
the amido nitrogens. (Molybdenum imido alkylidene complexes
that contain a N,N′-disubstituted-2,2′-bisamido-1,1′-binaphthyl
ligand also were shown to be unreactive toward ethylene and
even benzaldehyde at room temperature.52) However, in the
Boncella complexes, the tungstacyclobutane complex did not
appear to rearrange to propylene, and the tungstacyclopentane
complex did not appear to rearrange to 1-butene. Therefore, it
appears that both metathesis and â hydride processes are slowed
in bisamido complexes relative to bisphenoxide complexes.
Tungsten has potential as an asymmetric metathesis catalyst
in the right circumstances for several reasons. First, no five-
coordinate tungstacyclobutane complex that we have observed
so far is too stable for ready turnover in ring-closing reactions
at some readily accessible temperature where catalyst decom-
position is not rapid. Second, the enantioselectivities and
conversions can be high. Third, there are some hints that
tungsten catalyst systems may be more stable at higher
temperatures than molybdenum catalyst systems, and possibly
longer lived, and that substrates that are resistant to reactions
with Mo catalysts therefore may be transformed successfully
with W catalysts. We look forward in future studies to an
exploration of asymmetric reactions involving viable substrates
by the catalysts described here, and others. We also hope to
devise conditions that would allow us to isolate and structurally
characterize 20 (especially), or (more realistically) 19 or 21, or
related species. If the nature of the metal (Mo or W) along with
the diolate and the imido group is a variable that must be
considered in a search for a suitable catalyst for a given
substrate, then the number of available catalysts in theory
increases from approximately two dozen (for Mo) to twice that.
Experimental Section
General. All reactions were conducted in oven- or flame-dried
glassware under an inert atmosphere of nitrogen or argon. Commercially
available chemicals were obtained from Aldrich Co. or Lancaster
Synthesis. 2,6-Dimethylphenyl isocyanate was stirred over P2O5 for
24 h, vacuum distilled, and stored over molecular sieves (4 Å) at -25
°C. Liquid reagents were distilled from CaH2 under nitrogen and stored
over molecular sieves (4 Å) before use. Ether, pentane, toluene, benzene,
THF, and DME were dried with columns of activated alumina.
Dichloromethane was distilled from calcium hydride. Benzyl potas-
sium,28,29 W(O)Cl4, W(NAr′)Cl4, and W(NAr)(CHCMe2Ph)(OTf)2-
(DME)27,53 were synthesized according to published procedures.
Conversions were determined by 1H NMR of the unpurified reaction
mixtures. Enantiomeric ratios were determined by chiral GLC analyses
with an Alltech Associates Chiraldex GTA column (30 m × 0.25 mm)
or Betadex 120 column (30 m × 0.25 mm) in comparison with authentic
samples. Microanalyses were performed by Kolbe Microanalytical
Laboratories (Mu¨lheim an der Ruhr, Germany).
Representative Procedures for Ring-Closing Reactions in Tables
1 and 2. Diallyl sulfonamide (25 mg, 0.10 mmol) was loaded in a
J-Young tube. W(NAr′)(CHCMe2Ph)(Biphen) (12 mg) was dissolved
in benzene-d6 (1.5 mL) to obtain a 0.01 M catalyst solution. The catalyst
solution (0.5 mL) was added to the substrate via syringe, and the
J-Young tube was sealed, attached to a dinitrogen line, and immersed
into a 60 °C oil bath. The tube was opened to the dinitrogen line, and
the ethylene was vented to a fast flowing stream of dinitrogen through
the line. Periodically, the tube was resealed, and the reaction was
1
monitored by H NMR at room temperature.
Representative Procedure for Reactions Involving Ethylene. A
toluene-d8 solution (0.6 mL) of W(NAr)(CHCMe2Ph)(Biphen) (20 mg)
was loaded in a J-Young tube, sealed, and attached to a gas transfer
setup equipped with a 760 Torr gauge. The volume of the setup was
calibrated immediately before the experiment (24.23 mL). The sample
was degassed by freeze-pump-thaw and resealed with the J-Young
cap. Ethylene (45 Torr) was passed into the setup by slowly opening
the regulator attached to the ethylene tank. All of the ethylene in the
setup was condensed into the frozen sample (with liquid nitrogen) by
opening the J-Young seal. The sample was then brought to the
spectroscopy facility under liquid nitrogen before it is thawed and
immediately inserted into the NMR probe.
rac-W(NAr)(CHCMe2Ph)(Biphen)(THF) (rac-2a-THF). Potassium
hydride (470 mg, 11.7 mmol) was added in small portions to a solution
of H2(Biphen) (1.17 g, 3.3 mmol) in THF (20 mL). The solution was
cooled to -20 °C after being stirred for 2 h and was added to a chilled
solution (-20 °C) of W(NAr)(CHCMe2Ph)(OTf)2(DME) (2.9 g, 3.3
mmol) in THF (20 mL). The reaction mixture was warmed to room
temperature and stirred for 12 h. Concentration in vacuo gave a brown
solid, which was extracted with pentane (4 × 10 mL) and filtered. The
filtrate was concentrated in vacuo to approximately 5 mL. Upon being
cooled at -25 °C, a yellow-brown crystalline solid was obtained, yield
1
2.05 g (68%). H NMR (C6D6): δ 11.04 (s, 1, anti WdCH), 7.90 (s,
1, syn WdCH, JWH ) 14.2 Hz, JCH ) 113 Hz), 7.50 (d, 2, aryl, JHH
)
7.6 Hz), 7.41 (s, 1, aryl), 7.21 (t, 2, aryl, JHH ) 7.8 Hz), 7.04 (t, 1,
aryl, JHH ) 7.4 Hz), 7.01 (d, 2, aryl, JHH ) 7.3 Hz), 6.93 (m, 1, aryl),
3.61 (sept, 2, CH(CH3)2, JHH ) 6.7 Hz), 3.57 (s, br, 4, THF), 2.10 (s,
3, Ar-CH3), 2.03 (s, 3, Ar-CH3), 1.87 (s, 3, CHC(CH3)2Ph), 1.69 (s,
3, Ar-CH3), 1.61 (s, 3, Ar-CH3), 1.60 (s, 9, C(CH3)3), 1.50 (s, 9,
C(CH3)3), 1.17 (s, 3, CHC(CH3)2Ph), 1.14 (d, 6, CH(CH3)2, JHH ) 6.7
Hz), 0.92 (d, 6, CH(CH3)2, JHH ) 6.7 Hz). 13C NMR (100 MHz,
(49) Wang, S. S.; VanderLende, D. D.; Abboud, K. A.; Boncella, J. M.
Organometallics 1998, 17, 2628.
(50) Cameron, T. M.; Ortiz, C. G.; Ghiviriga, I.; Abboud, K. A.; Boncella, J.
M. Organometallics 2001, 20, 2032.
(51) Ortiz, C. G.; Abboud, K. A.; Boncella, J. M. Organometallics 1999, 18,
4253-4260.
(52) Jamieson, J. Y.; Schrock, R. R.; Davis, W. M.; Bonitatebus, P. J.; Zhu, S.
S.; Hoveyda, A. H. Organometallics 2000, 19, 925.
(53) Schrock, R. R.; DePue, R.; Feldman, J.; Schaverien, C. J.; Dewan, J. C.;
Liu, A. H. J. Am. Chem. Soc. 1988, 110, 1423.
9
2664 J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003

Alkylidene and Metalacyclic Complexes of Tungsten
A R T I C L E S
C6D6): δ 245.5, 153.4, 153.0, 152.2, 151.9, 144.9, 139.7, 137.7, 135.9,
135.2, 131.7, 131.5, 131.0, 130.5, 129.5, 129.4, 127.0, 126.3, 125.8,
123.0, 51.1, 35.6, 35.3, 34.6, 30.5, 30.3, 28.5, 25.7, 24.3, 24.1, 20.4,
20.3, 16.8, 16.3. Anal. Calcd for WC50H69NO3: C, 65.57; H, 7.59; N,
1.53. Found: C, 65.79; H, 7.40; N, 1.56.
and filtered. The filtrate was concentrated in vacuo to give a yellow-
brown foam, yield 676 mg (85%).
W(NAr′)(O-t-Bu)2Cl2(THF). W(NAr′)Cl4 was recrystallized as a
solvent adduct from diethyl ether before use. Lithium tert-butoxide (13.1
g, 164 mmol) suspended in diethyl ether (100 mL) was added to a
chilled solution (-25 °C) of W(NAr′)Cl4(Et2O) (42.5 g, 82 mmol) in
diethyl ether (300 mL) and THF (100 mL) over 20 min. After being
stirred for 20 h, the mixture was filtered through Celite; the solid residue
was washed with diethyl ether (100 mL) until colorless. All volatiles
were removed in vacuo, and the orange solid was extracted with diethyl
ether (250 mL). The solution was concentrated to approximately 200
mL and stored at -25 °C for 18 h. Orange crystals were collected by
rac-W(NAr)(CHCMe2Ph)(Biphen) (rac-2a). To a 25 mL scintil-
lation vial was added W(NAr)(CHCMe2Ph)(Biphen)(THF) (500 mg,
0.55 mmol). The complex was dissolved in toluene (5-10 mL), and
the solution was filtered through Celite. All volatiles were removed in
vacuo. The sample was redissolved in toluene (5-10 mL), and all
volatiles were removed. In the final cycle, benzene (5-10 mL) was
added, and all volatiles were removed in vacuo to obtain a bright
orange-yellow dry foam, yield 480 mg (96%). 1H NMR (C6D6): δ 10.76
(s, br, 1, anti WdCH), 7.91 (s, 1, syn WdCH, JWH ) 14.11 Hz, JCH
) 113.36 Hz), 7.52 (d, 2, aryl), 7.43 (s, 1, aryl), 7.23 (t, 2, aryl), 7.18
(s, 1, aryl), 7.06 (t, 2, aryl), 7.02 (d, 2, aryl), 6.95 (t, 1, aryl), 3.63
(sept, 2, CH(CH3)2), 2.12 (s, 3, Ar-CH3), 2.04 (s, 3, Ar-CH3), 1.89
(s, 3, Ar-CH3), 1.71 (s, 3, Ar-CH3), 1.63 (s, 3, CHC(CH3)2Ph), 1.62
(s, 9, C(CH3)3), 1.52 (s, 9, C(CH3)3), 1.19 (s, 3, CHC(CH3)2Ph), 1.16
(d, 6, CH(CH3)2), 0.94 (d, 6, CH(CH3)2). 13C NMR (125 MHz, C6D6):
δ 245.8, 153.83, 153.39, 152.55, 152.24, 145.31, 140.11, 138.13,
136.34, 135.57, 132.10, 131.94, 131.43, 130.88, 129.94, 129.82, 128.93,
127.43, 126.71, 126.17, 123.41, 51.45, 36.04, 35.71, 34.97, 31.94, 30.93,
30.72, 28.90, 24.69, 24.46, 20.78, 20.71, 17.20, 16.70. Anal. Calcd for
WC46H61NO2: C, 65.45; H, 7.29; N, 1.66. Found: C, 65.58; H, 7.35;
N, 1.58.
1
filtration and dried in vacuo, yield 32.3 g (66%). H NMR (C6D6): δ
6.90 (d, 2, aryl, JHH ) 7.8 Hz), 6.58 (t, 1, aryl, JHH ) 7.8 Hz), 4.20 (t,
br, 4, THF), 3.02 (s, 6, Ar-CH3), 1.46 (s, 18, OC(CH3)3), 1.42 (t, br,
4, THF). 13C NMR (125 Hz, C6D6): δ 151.18, 140.29, 128.68, 128.04,
86.65, 71.21, 31.05, 25.79, 21.16. Anal. Calcd for WC20H35Cl2NO2:
C, 40.56; H, 5.96; N, 2.36. Found: C, 40.39; H, 5.88; N, 2.43.
W(NAr′)(O-t-Bu)2(CH2CMe2Ph)2. An ether solution of PhMe2-
CCH2MgCl (95 mL, 109 mmol, 1.15 M) was slowly added to a cold
solution of W(NAr′)(O-t-Bu)2Cl2(THF) (32.3 g, 54.6 mmol) in diethyl
ether (120 mL, -25 °C). The mixture was stirred for 20 h at room
temperature. The mixture was filtered through Celite, and the residue
was washed with diethyl ether (100 mL) until colorless. The combined
solution was concentrated to approximately 50 mL. Orange needles
were observed upon overnight cooling at -25 °C. The crystals were
1
collected by filtration and dried in vacuo, yield 20 g (51%). H NMR
(S)-W(NAr)(CHCMe2Ph)(Biphen) ((S)-2a). Potassium hydride
(220 mg, 5.48 mmol) was added in small portions to a solution of (S)-
H2(Biphen) (712 mg, 2.01 mmol) in THF (15 mL). After 18 h, the
suspension was cooled to -20 °C and added to a chilled solution (-20
°C) of W(NAr)(CHCMe2Ph)(OTf)2(DME) (1.77 g, 2.01 mmol) in THF
(15 mL). The brown mixture was warmed to room temperature and
stirred for 12 h. Concentration in vacuo gave a brown solid, which
was extracted with benzene (3 × 5 mL) and filtered. The filtrate was
dried to give a brown foam, yield 1.51 g (89%).
(C6D6): δ 7.55 (d, 4, aryl, JHH ) 7.5 Hz), 7.23 (t, 4, aryl, JHH ) 7.5
Hz), 7.05 (t, 2, aryl, JHH ) 7.5 Hz), 6.92 (d, 2, aryl, JHH ) 7.5 Hz),
6.63 (t, 1, aryl, JHH ) 7.5 Hz), 2.45 (s, 6, Ar-CH3), 2.21 (s, br, 4,
CH2CMe2Ph), 1.66 (s, 12, CH2C(CH3)2Ph), 1.34 (s, 18, OC(CH3)3).
13C NMR (125 Hz, C6H6): δ 156.96, 152.99, 137.50, 128.74, 127.84,
126.29, 126.09, 125.78, 42.09, 33.36, 32.09, 19.98, 14.64. Anal. Calcd
for WC36H53NO2: C, 60.42; H, 7.40; N, 1.96. Found: C, 60.58; H,
7.54; N, 2.04.
W(NAr′)(CHCMe2Ph)Cl2(DME) (3b). Phosphorus pentachloride
(4.41 g, 21.2 mmol) was added to a rapidly stirring solution of W(NAr′)-
(O-t-Bu)2(CH2CMe2Ph)2 (14.7 g, 20.6 mmol) in cold (-25 °C) DME
(125 mL). After 1 h, all volatiles were removed in vacuo. The residue
was extracted in diethyl ether (30 mL), and the orange precipitate was
collected by filtration, triturated with diethyl ether (2 × 10 mL), and
rac-W(NAr′)(CHCMe2Ph)(Biphen) (rac-2b). Benzyl potassium
(533 mg, 4.1 mmol) was added to a THF (40 mL) solution of rac-H2-
(Biphen) (708 mg, 2 mmol). After 15 min, W(NAr′)(CHCMe2Ph)Cl2-
(DME) (1.191 g, 2 mmol) was added, and the reaction mixture was
stirred for 2 h at room temperature. All volatiles were removed to give
a yellow powder which was extracted with benzene (20 mL). The
suspension was filtered through Celite, and the residue was washed
with additional benzene (30 mL) until colorless. Volatiles were removed
in vacuo, and the residue was redissolved in diethyl ether (10 mL).
The bright yellow powder precipitated was collected by filtration, yield
1.09 g (69%). 1H NMR (mixture of isomers, Ksyn/anti ) 88): δ 9.06 (s,
1, anti WdCH, JWH ) 15 Hz), 7.99 (s, 1, syn WdCH, JWH ) 16.5 Hz,
JCH ) 115 Hz), 7.44 (s, 1, aryl), 7.40 (d, 2, aryl, JHH ) 7.5 Hz), 7.17
(s, 1, aryl), 7.14 (t, 2, aryl, JHH ) 7.5 Hz), 6.98 (t, 1, aryl, JHH ) 7.5
Hz), 6.83 (d, 2, aryl, JHH ) 7.5 Hz), 6.74 (t, 1, aryl, JHH ) 7.5 Hz),
2.31 (s, 6, Ar-CH3), 2.12 (s, 3, Ar-CH3), 2.00 (s, 3, Ar-CH3), 1.72
(s, 3, Ar-CH3/CHC(CH3)2Ph), 1.67 (s, 3, Ar-CH3/CHC(CH3)2Ph), 1.60
(s, 3, Ar-CH3/CHC(CH3)2Ph), 1.58 (s, 9, C(CH3)3), 1.54 (s, 9,
C(CH3)3), 1.33 (s, 3, CHC(CH3)2Ph). 13C NMR (125 MHz, C6D6): δ
247.70, 155.12, 153.25, 152.67, 152.02, 140.29, 138.32, 136.58, 135.76,
135.48, 132.38, 131.95, 131.28, 130.89, 130.04, 129.86, 128.91, 128.10,
127.30, 126.19, 126.10, 51.75, 36.04, 35.45, 34.91, 33.92, 30.94, 30.85,
20.79, 20.71, 19.60, 17.30, 16.80. Anal. Calcd for WC42H53NO2: C,
64.04; H, 6.78; N, 1.78. Found: C, 64.12; H, 6.71; N, 1.74.
1
dried in vacuo, yield 5.4 g (44%). H NMR (C6H6): δ 10.25 (s, 1,
WdCH), 7.71 (d, 2, aryl, JHH ) 7.7 Hz), 7.27 (t, 2, aryl, JHH ) 7.7
Hz), 7.05 (t, 1, aryl, JHH ) 7.7 Hz), 6.86 (d, 2, aryl, JHH ) 7.7 Hz),
6.76 (t, 1, aryl, JHH ) 7.7 Hz), 3.16 (s, 6, OCH3), 3.12 (s, 4, OCH2),
2.86 (s, 6, Ar-CH3), 1.76 (s, 6, CHC(CH3)2Ph). 13C NMR (125 Hz,
C6D6): δ 282.00, 155.32, 139.97, 128.92, 128.58, 127.13, 127.03,
126.44, 71.94, 62.64, 53.78, 33.00, 21.22.
Procedures for Observation of Intermediates 12 and 13. An NMR
tube was charged with rac-2a (52 mg, 57 µmol) and toluene-d8 (0.5
mL). The tube was cooled to -78 °C, and triene 9 (8 mg, 53 µmol)
was added by syringe. The tube was flame-sealed in vacuo and was
immediately inserted into the NMR probe cooled to -60 °C. The
sample was warmed to -40 °C to allow for complete consumption of
the starting materials. The sample was cooled back to -60 °C after 30
min, and the spectra for 12 were collected. Warming to 0 °C for 15
min resulted in complete decomposition of 12, and spectra for species
13 were collected at -40 °C. The sample was warmed to 20 °C, and
1
the decomposition of 13 was monitored by H NMR spectroscopy.
(S)-W(NAr′)(CHCMe2Ph)(Biphen) ((S)-2b). To a solution of (S)-
H2(Biphen) (357 mg, 1.01 mmol) in THF (10 mL) was added benzyl
potassium (262 mg, 2.02 mmol). After 15 min, the suspension was
added to a solution of W(NAr′)(CHCMe2Ph)Cl2(DME) (602 mg, 1.01
mmol) in THF (10 mL). The yellow-brown solution was stirred for 2
h and chilled at -20 °C for 12 h. Removal of volatiles in vacuo gave
a yellow-brown solid, which was extracted with pentane (3 × 5 mL)
Metalacyclobutane 14. A saturated solution of rac-2a (185 mg, 202
µmol) in pentane (2.5 mL) was added to triene 9 (35 mg, 230 µmol)
at room temperature. The red reaction mixture was left without stirring
for 16 h, after which the supernatant was decanted from the ruby red
cubes. The crystals were dried in vacuo to give 14, yield 137 mg (80%).
1H NMR (tol-d8): δ 7.24 (s, 1, aryl), 7.06 (s, 1, aryl), 7.00 (d, 2, aryl,
JHH ) 7.7), 6.81 (t, 1, aryl, JHH ) 7.7 Hz), 4.88 (m, 1, CH2O), 4.83 (s,
9
J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003 2665

A R T I C L E S
Tsang et al.
br, 1, OCH2CHdCMe), 4.53 (m, br, 1, CHO), 3.67 (m, br, 2,
CH(CH3)2), 3.05 (m, 1, CH2O), 2.53 (d, 1, WCH2, JHH ) 9.7 Hz), 2.23
(s, 3, Ar-CH3), 2.18 (s, 3, Ar-CH3), 1.79 (d, 1, WCH2, JHH ) 10.0
Hz), 1.75 (s, 3, Ar-CH3), 1.73 (s, 3, Ar-CH3), 1.57 (s, 9, C(CH3)3),
1.41 (s, 9, C(CH3)3), 1.37 (m, br, 3, OCHC(CH3)dCH), 1.33 (d, 1,
WCH2, JHH ) 9.6 Hz), 1.27 (s, 3, WCH2C(CH3)(C5H7O)CH2), 1.23
(d, 6, CH(CH3)2, JHH ) 6.8 Hz), 1.17 (d, 1, WCH2, JHH ) 10.0 Hz),
1.11 (d, 6, CH(CH3)2, JHH ) 6.8 Hz). 13C NMR (100 MHz, tol-d8): δ
158.6, 156.7, 150.5, 137.6, 135.8, 134.6, 134.5, 134.0, 131.7, 130.1,
128.9, 128.5, 127.9, 127.4, 126.7, 124.4 (OCH2CHdCMe), 122.5, 99.1
(CHO), 74.3 (CH2O), 55.1, 47.9 (WCH2), 37.5 (WCH2C(CH3)(C5H7O)-
CH2), 35.2, 34.9, 31.6, 31.4 (WCH2C(CH3)(C5H7O)CH2), 30.4, 27.7,
25.0, 24.7, 20.7, 20.4, 17.4, 16.8, 14.1. Anal. Calcd for WC45H63NO3:
C, 63.60; H, 7.47; N, 1.65. Found: C, 63.46; H, 7.58; N, 1.61.
Metalacyclobutane 16. A solution of rac-2a (100 mg, 109 µmol)
in pentane (5 mL) was added to freshly distilled and degassed 1,6-
heptadiene (10.5 mg, 109 µmol) at room temperature. The solution
was stirred for 2 h and filtered through Celite. The red solution was
concentrated to approximately 1 mL and cooled to -30 °C for 12 h.
Some orange crystals of 16 and a small amount of unidentified purple-
black powder were observed in the vial; the total dry weight was 65
mg. The orange crystals were carefully picked out for further analysis.
1H NMR: δ 7.35 (d, 2, aryl), 7.31 (s, 1, aryl), 7.29 (t, 1, aryl), 7.19 (d,
2, aryl), 7.13 (s, 1, aryl), 7.02 (t, 2, aryl), 6.93 (t, 1, aryl), 3.41 (sept,
2, CHCMe2), 3.18 (m, 1, â-CH), 2.95 (m, 1, R-CHeq), 2.55 (m, 1,
R-CHeq), 2.08 (s, 3, Ar-CH3), 2.03 (s, 3, Ar-CH3), 1.65 (s, 3, Ar-
CH3), 1.64 (s, 3, Ar-CH3), 1.52 (s, 9, C(CH3)3), 1.39 (s, 3, CHC-
(CH3)2Ph), 1.38 (s, 9, C(CH3)3), 1.32 (m, 1, R-CHax), 1.29 (s, 3,
CHC(CH3)2Ph), 1.26 (m, 1, R-CHax), 1.22 (d, 6, CH(CH3)2), 1.07 (d,
6, CH(CH3)2). Anal. Calcd for WC48H65NO2: C, 66.10; H, 7.52; N,
1.61. Found: C, 66.22; H, 7.46; N, 1.57.
rac-W(NAr′)(C3H6)(Biphen) (rac-18b). Four equivalents of 1,6-
heptadiene was added to a slurry of rac-W(NAr′)(CHCMe2Ph)(Biphen)
(150 mg, 0.191 mmol) in pentane (1 mL). The vial was immediately
capped, and the suspension became homogeneous. The reaction was
cooled to -25 °C for 6 days. The yellow-orange precipitate was
collected by decanting the supernatant and drying the precipitate in
vacuo, yield 125 mg (95%). 1H NMR (C6D6): δ 7.17 (s, 2, aryl), 6.81
(d, 2, aryl), 6.65 (t, 1, aryl), 4.10-3.80 (m, 3, R-CH2), 3.00 (m, 1,
R-CH2), 2.27 (s, 6, Ar-CH3), 2.15 (s, 6, Ar-CH3), 1.78 (s, 6, Ar-
CH3), 1.54 (s, 18, C(CH3)3), 0.70 (s, br, 1, â-CH2), 0.10 (s, br, 1, â-CH2).
13C NMR (125 MHz, C6D6): δ 152.22, 136.69 (br), 134.83, 129.55
(br), 127.90, 126.14, 86.9 (br), 79.4 (br), 35.82 (br), 34.76, 33.08, 31.74
(br), 30.53, 28.71, 23.48, 23.06, 20.71, 19.76, 19.22, 17.18, 14.63, 3.7
(br), 3.51. Anal. Calcd for WC35H47NO2: C, 60.26; H, 6.79; N, 2.01.
Found: C, 60.44; H, 6.73; N, 1.95.
Acknowledgment. We thank the National Science Foundation
(CHE-9988766 to R.R.S.) and the National Institutes of Health
(GM-59426 to R.R.S. and A.H.H.) for supporting this research.
K.C.H. thanks the Alexander von Humboldt Foundation for
partial support through a Feodor Lynen Research Fellowship.
We also gratefully acknowledge the generous assistance of Dr.
Jeffrey H. Simpson in multidimensional NMR experiments.
Supporting Information Available: Fully labeled ORTEP
drawing, atomic coordinates, bond lengths and angles, and
anisotropic displacement parameters for 14 and 16 (PDF). This
material is available free of charge via the Internet at
http://pubs.acs.org.
JA0210603
9
2666 J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003