Mechanistic insight into fragmentation reactions of titanapinacolate complexes

Article abstract of DOI:10.1021/ja0271577

Reactions between terminal alkynes or aromatic ketones and titanapinacolate complexes (DMSC)-Ti(OCAr₂CAr₂O) (2, Ar = Ph, and 3, Ar = p- MeC₆H₄; DMSC = 1,2-alternate dimethylsilyl-bridged p-tert-butylcalix[4]arene dianion) occur via rupture of the C-C bond of the titanacycle. Thus, reactions of 2 and 3 with terminal alkynes produce 2-oxatitanacyclopent-4-ene or 2-oxatitanacycloheptadiene complexes along with free Ar₂CO. These compounds have been characterized spectroscopically and by X-ray crystallography. Because metallapinacolate intermediates have been implicated in important C-C bond-forming reactions, such as pinacol coupling and McMurry chemistry, the mechanism of the fragmentation reactions was studied. Analysis of the kinetics of the reaction of (DMSC)Ti {OC(p-MeC₆H₄)₂C (p-MeC₆H₄)₂O} (3) with Bu¹C≡CH revealed that the fragmentation reactions proceed via a preequilibrium mechanism, involving reversible dissociation of titanapinacolate complexes into (DMSC)Ti(η²-OCAr₂) species with release of a ketone molecule, followed by rate-limiting reaction of (DMSC)Ti (η₂-OCAr₂) species with an alkyne or ketone molecule.

Full text of DOI:10.1021/ja0271577

Published on Web 09/20/2002
Mechanistic Insight into Fragmentation Reactions of
Titanapinacolate Complexes
Jesudoss V. Kingston, Oleg V. Ozerov,† Sean Parkin, Carolyn P. Brock, and
Folami T. Ladipo*
Contribution from the Department of Chemistry, UniVersity of Kentucky,
Lexington, Kentucky 40506-0055
Received June 3, 2002
Abstract: Reactions between terminal alkynes or aromatic ketones and titanapinacolate complexes (DMSC)-
Ti(OCAr2CAr2O) (2, Ar ) Ph, and 3, Ar ) p-MeC6H4; DMSC ) 1,2-alternate dimethylsilyl-bridged p-tert-
butylcalix[4]arene dianion) occur via rupture of the C-C bond of the titanacycle. Thus, reactions of 2 and
3 with terminal alkynes produce 2-oxatitanacyclopent-4-ene or 2-oxatitanacycloheptadiene complexes along
with free Ar2CO. These compounds have been characterized spectroscopically and by X-ray crystallography.
Because metallapinacolate intermediates have been implicated in important C-C bond-forming reactions,
such as pinacol coupling and McMurry chemistry, the mechanism of the fragmentation reactions was studied.
Analysis of the kinetics of the reaction of (DMSC)Ti{OC(p-MeC6H4)2C(p-MeC6H4)2O} (3) with ButCtCH
revealed that the fragmentation reactions proceed via a preequilibrium mechanism, involving reversible
dissociation of titanapinacolate complexes into (DMSC)Ti(η2-OCAr2) species with release of a ketone
molecule, followed by rate-limiting reaction of (DMSC)Ti(η2-OCAr2) species with an alkyne or ketone
molecule.
Scheme 1
Introduction
The pinacol coupling reaction (reductive coupling of carbonyl
compounds to yield 1,2-diols, mediated by a variety of metals
in low oxidation state) and the McMurry reaction (reductive
coupling of carbonyl compounds to yield alkenes, promoted by
low-valent titanium reagents) are among the most powerful
methods for constructing carbon-carbon bonds.1-3 Metallapi-
nacolate intermediates have been implicated in both of these
reactions (Scheme 1).1,4 For example, Villiers and Ephritik-
hine4a,4c have isolated and characterized a pinacolate intermedi-
* To whom correspondence should be addressed. E-mail:
fladipo0@uky.edu.
† Current address: Department of Chemistry, Brandeis University,
MS#015, 415 South Street, Box 9110, Waltham, MA 02553-2728.
(1) (a) Eisch, J. J.; Gitua, J. N.; Otieno, P. O.; Shi, X. J. Organomet. Chem.
2001, 624, 229. (b) Ephritikhine, M. J. Chem. Soc. Chem. Commun. 1998,
2549. (c) Wirth, T. Angew. Chem., Int. Ed. Engl. 1996, 35, 61. (d) Furstner,
A.; Bogdanovic, B. Angew. Chem., Int. Ed. Engl. 1996, 35, 2443. (e)
McMurry, J. E. Chem. ReV. 1989, 89, 1513. (f) Dushin, R. G. In
ComprehensiVe Organometallic Chemistry II; Hegedus, L. S., Ed.; Perga-
mon: Oxford, U.K., 1996; Vol. 12, p 1071.
(2) (a) Kammermeier, B.; Beck, G.; Jacobi, D.; Jendralla, H. Angew. Chem.,
Int. Ed. Engl. 1994, 33, 685. (b) Robertson, G. M. In ComprehensiVe
Organic Synthesis I; Trost, B. M., Ed.; Pergamon: New York, 1991; Vol.
3, p 563. (c) Bandini, M.; Cozzi, G.; Morganti, S.; Umani-Ronchi, A.
Tetrahedron Lett. 1999, 40, 1997.
(3) (a) Nicolaou, K. C.; Yang, Z.; Liu, J. J.; Ueno, H.; Nantermet, P. G.; Guy,
R. K.; Claiborne, C. F.; Renaud, J.; Couladouros, E. A.; Paulvannan, K.;
Sorenson, E. J. Nature 1994, 367, 630. (b) Nicolaou, K. C.; Liu, J. J.;
Yang, Z.; Ueno, H.; Sorenson, E. J.; Claiborne, C. F.; Guy, R. K.; Hwang,
C.-K.; Nakada, M.; Nantermet, P. G. J. Am. Chem. Soc. 1995, 117, 634.
(c) Nicolaou, K. C.; Yang, Z.; Liu, J. J.; Nantermet, P.; G.; Claiborne, C.
F.; Renaud, J.; Guy, R. K.; Shibayama, K. J. Am. Chem. Soc. 1995, 117,
645.
(4) (a) Ephritikhine, M.; Maury, O.; Villiers, C.; Lance, M.; Nierlich, M. J.
Chem. Soc., Dalton Trans. 1998, 3021. (b) Maury, O.; Villiers, C.;
Ephritikhine, M. Angew. Chem. 1996, 108, 1215. (c) Villiers, C.; Ephri-
tikhine, M. Chem.s Eur. J. 2001, 7, 3043.
ate {UCl3(thf)2}2(µ-OCMe2CMe2O) in uranium-mediated reduc-
tive coupling of acetone at 20 °C, and deoxygenation of the
metallapinacolate into tetramethylethylene was observed at
higher temperature. We recently reported the synthesis of
titanapinacolate complexes (DMSC)Ti(OCAr2CAr2O) (2, Ar )
Ph, and 3, Ar ) p-MeC6H4; DMSC ) 1,2-alternate dimethyl-
silyl-bridged p-tert-butylcalix[4]arene dianion, eq 1).5 Structural
characterization of 2 by X-ray crystallography revealed that the
unit cell contained two independent molecules and that the
OCPh2CPh2O fragment of each molecule possessed an unusually
long C-C bond [1.628(6) and 1.652(5) Å].5 Whether this rather
(5) Ozerov, O. V.; Parkin, S.; Brock, C. P.; Ladipo, F. T. Organometallics
2000, 19, 4187.
9
10.1021/ja0271577 CCC: $22.00 © 2002 American Chemical Society
J. AM. CHEM. SOC. 2002, 124, 12217-12224
12217

A R T I C L E S
Kingston et al.
powder isolated by filtration was dried under vacuum. Yield: 1.042 g
(92%). Both 1H and13C NMR data are identical to that previously
reported.5
(DMSC)2Ti (4). (DMSC)H2 (0.176 g, 0.250 mmol) was dissolved
in 15 mL of toluene, and (PhCH2)2Mg‚2THF (0.0875 g, 0.250 mmol)
was added to it. The mixture was stirred for 15 min, during which
time (DMSC)Mg started to precipitate. (DMSC)TiCl2 (0.205 g, 0.250
mmol) was added to this heterogeneous mixture, along with 15 mL of
THF. The resulting mixture was stirred at room temperature for 24 h.
Next, the insolubles were filtered off and the amount of solvent was
reduced to 10 mL. Pentane (7 mL) was then added to the solution, and
the resulting precipitate was allowed to settle at -20 °C for 4 h. A
bright-yellow precipitate was collected and dried. Yield: 0.220 g (60%).
1H NMR (C6D6): δ 7.34 (d, 2H, arom CH), 7.09-7.25 (m, 10H, arom
CH), 7.07 (d, 2H, arom CH), 6.93 (d, 2H, arom CH), 4.77 (d, J ) 14.5
Hz, 2H, calix-CH2), 4.41 (d, J ) 15 Hz, 2H, calix-CH2), 4.18 (d, J )
17 Hz, 2H, calix-CH2), 4.01 (d, J ) 17 Hz, 2H, calix-CH2), 3.79 (d, J
) 15 Hz, 2H, calix-CH2), 3.77 (d, J ) 16 Hz, 2H, calix-CH2), 3.53 (d,
J ) 17 Hz, 2H, calix-CH2), 3.44 (d, J ) 16 Hz, 2H, calix-CH2), 1.51
(s, 18H, t-Bu), 1.30 (s, 18H, t-Bu), 1.27 (s, 18H, t-Bu), 1.00 (s, 18H,
t-Bu), 0.07 (s, 6H, exo-SiCH3), -1.32 (s, 6H, endo-SiCH3). 13C NMR
(CDCl3): δ 156.0 (TiOC), 159.5 (TiOC), 149.7, 149.2, 144.9, 143.8,
143.1, 143.0, 135.2, 131.1, 129.8, 129.2, 128.0 (br s), 127.3, 127.2,
126.9, 126.6, 125.9, 125.8, 125.6 (br s), 125.4, 124.8, 40.1 (calix-CH2),
37.7 (calix-CH2), 36.7 (calix-CH2), 34.5 (calix-CH2), 34.1 (C(CH3)3),
34.0 (C(CH3)3), 33.9 (C(CH3)3), 33.5 (C(CH3)3), 31.9 (C(CH3)3), 31.5
(br s, C(CH3)3), 30.9 (C(CH3)3), 2.4 (SiCH3), -2.2 (SiCH3). MS (m/
z): M+ (1454).
peculiar structural feature portends intriguing reaction chemistry
for titanapinacolate complexes 2 and 3 was of interest to us,
especially since 2 and 3 are titanium derivatives of V (Scheme
1). We have found that DMSC-based titanapinacolates 2 and 3
undergo remarkably facile fragmentation of the metallacyclic
C-C bond upon reaction with unsaturated organic molecules,
such as terminal alkynes and ketones. Whereas reversible
cleavage of the C-C bond of some metallapinacolate intermedi-
ates has been invoked to explain their behavior in McMurry
reactions,4c,6-8 facile fragmentation of well-characterized met-
allapinacolate complexes by terminal alkynes is unprecedented
to the best of our knowledge. More importantly, there is
currently very little understanding of the mechanism(s) of
metallapinacolate C-C bond rupture. In this paper, we present
an analysis of the mechanism of fragmentation reactions of
titanapinacolate compounds 2 and 3 on the basis of their
structural properties and kinetic studies.
(DMSC)Ti{OC(p-MeC6H4)2C2ButH} (5). (DMSC)Ti{OC(p-Me-
C6H4)2C(p-MeC6H4)2O} (3) (1.32 g, 1.13 mmol), ButCtCH (0.55 mL,
4.50 mmol), and 25 mL of heptane were charged into a 50 mL reaction
vessel equipped with a Teflon stopcock. The vessel was closed off and
heated at 75 °C for 2 h. The yellow solution was then allowed to
gradually cool to ambient temperature and was left standing for 18 h.
The volatiles were removed under reduced pressure, and the residue
was dissolved in a minimal amount of pentane and placed in the
glovebox freezer at -15 °C for 24 h. During this time, (p-MeC6H4)2-
CO precipitated and was filtered off. The filtrate was stripped under
reduced pressure, and the residue was dissolved in 3 mL of Et2O and
cooled at -15 °C. After 3 days, yellow crystalline precipitate was
collected and recrystallized from pentane to give yellow crystals of
pure 5. Yield: 0.35 g (30%). The rather low isolated yield is primarily
Experimental Section
General Methods. All experiments were performed under dry
nitrogen atmosphere using standard Schlenk techniques or in a Vacuum
Atmospheres, Inc., glovebox. Tetrahydrofuran, ether, and toluene were
distilled twice from sodium benzophenone ketyl. Pentane was distilled
twice from sodium benzophenone ketyl with addition of 1 mL/L of
tetraethylene glycol dimethyl ether as a solubilizing agent. Benzene-d6
was distilled from sodium benzophenone ketyl. All solvents were stored
in the glovebox over 4 Å molecular sieves that were dried in a vacuum
oven at 150 °C for at least 48 h prior to use. Ph213CO and alkynes
were purchased from Aldrich. All of the alkynes were distilled from
CaH2 prior to use. 1H and 13C NMR spectra were recorded on a Varian
Gemini-200 spectrometer or a Varian VXR-400 spectrometer at ca. 22
°C. 1H and 13C chemical shifts were referenced to residual solvent peaks.
GC-MS analyses were performed on a Hewlett-Packard 5890 series II
gas chromatograph with a Hewlett-Packard 5972 series mass selective
detector at an ionizing potential of 70 eV. Alternately, mass spectral
data were obtained from the University of Kentucky Mass Spectrometry
Center. Elemental analyses were performed by Complete Analysis
Laboratories, Inc., Parsippany, NJ. The kinetic data were fitted using
the MacCurveFit program (version 1.1).
1
due to the very high solubility of the product. H NMR (C6D6): δ
7.68 (s, 1H, Câ-H), 7.31 (d, 2H, J ) 8 Hz, p-tolyl), 7.30 (d, 2H, J )
2 Hz, calix arom), 7.27 (d, 2H, J ) 2 Hz, calix arom), 7.18 (d, 2H, J
) 2 Hz, calix arom), 7.08 (d, 2H, J ) 8 Hz, p-tolyl), 7.00 (d, 2H, J )
2 Hz, calix arom), 4.86 (d, 1H, J ) 13 Hz, calix-CH2), 4.27 (d, 1H, J
) 16 Hz, calix-CH2), 4.09 (d, 2H, J ) 17 Hz, calix-CH2), 3.90 (d, 2H,
J ) 17 Hz, calix-CH2), 3.83 (d, 1H, J ) 16 Hz, calix-CH2), 3.52 (d,
1H, J ) 13 Hz, calix-CH2), 2.22 (s, 6H, MeC6H4), 1.27 (s, 18H, t-Bu),
1.19 (s, 18H, t-Bu), -0.21 (s, 3H, exo-SiMe), -1.57 (s, 3H, endo-
SiMe). 13C NMR (C6D6): δ 221.1 (TiCR), 161.0 (TiOC), 151.3, 150.5,
145.4, 144.8, 144.5, 135.6, 130.8, 130.7, 129.0, 128.3 (br), 127.8 (br),
126.7, 125.9, 125.8, 85.9 (C(MeC6H4)2), 39.9 (calix-CH2), 38.4 (calix-
CH2), 37.8 (calix-CH2), 34.1 (C(CH3)3), 33.9 (C(CH3)3), 31.54 (C(CH3)3),
31.46 (C(CH3)3), 21.0 (MeC6H4), 1.4 (exo-SiMe), -3.6 (endo-SiMe).
Anal. Calcd for C67H82O5SiTi: C, 77.13; H, 7.92. Found: C, 77.11; H,
8.01. A single crystal of 5 suitable for an X-ray diffraction study was
obtained by cooling a pentane solution of 5 to -15 °C.
(DMSC)Ti{OCPh2C2(SiMe3)H} (6). (DMSC)Ti(OCPh2CPh2O) (2)
was generated in situ from 0.020 mmol (20.9 mg) of (DMSC)Ti{η6-
1,2,4-(Me3Si)3C6H3} (1) and 0.040 mmol (7.28 mg) of Ph2CO (C6D6
solution). To this solution 0.100 mmol (14.1 µL) of Me3SiCtCH was
added, and the reaction was heated at 80 °C and monitored by 1H NMR.
After 25 min, 2 was completely exhausted and 6 and 9 were observed
in ca. 1:2 ratio as the only DMSC-containing species in solution. After
(DMSC)Ti{OC(p-MeC6H4)2C(p-MeC6H4)2O} (3). To a solution of
(DMSC)Ti{1,2,4-(Me3Si)3C6H3} (1)9 (1.015 g, 0.970 mmol) in dry
heptane (20 mL) was added 4,4′-dimethylbenzophenone (0.408 g, 1.94
mmol). The brown slurry was heated at 65 °C with constant stirring
for 1 h, using an oil bath. The clear yellow solution was cooled to
room temperature, and the solvent was removed under reduced pressure
to give a sticky yellow residue. A 5 mL volume of cold pentane was
added into the flask, and the solution was stirred for 10 min. The yellow
(6) Bogdanovic, B.; Bolte, A. J. Organomet. Chem. 1995, 502, 109.
(7) Coutts, R. S. P.; Wailes, P. C.; Martin, R. L. J. Organomet. Chem. 1973,
50, 145.
(8) Kahn, B. E.; Rieke, R. D. Chem. ReV. 1988, 88, 733.
(9) Ozerov, O. V.; Patrick, B. O.; Ladipo, F. T. J. Am. Chem. Soc. 2000, 122,
6423.
9
12218 J. AM. CHEM. SOC. VOL. 124, NO. 41, 2002

Fragmentation Reactions of Titanapinacolates
A R T I C L E S
2 h, the conversion of 6 into 9 was >95% complete. The resonances
due to 6 were mostly obstructed by those of 9 although certain peaks
could be identified. 1H NMR (unobstructed resonances) (C6D6): δ 8.25
(s, 1H, Câ-H), 1.29 (s, 18H, t-Bu), 1.17 (s, 18H, t-Bu), 0.18 (s, 9H,
SiMe3), -0.23 (s, 3H, exo-SiMe), -1.60 (s, 3H, endo-SiMe).
it. This solution was placed in the freezer at -15 °C for 24 h. The
resulting precipitate was collected, washed with 2 mL of pentane, and
1
dried under vacuum to give 0.56 g (92%) of solid yellow product. H
NMR (C6D6): δ 7.62 (d, 1H, J ) 2 Hz, calix arom), 7.56 (pseudo d,
2H, J ) 7 Hz, Ph), 7.42 (d, 1H, J ) 2 Hz, calix arom), 7.33 (2 AB
doublets, 2H, calix arom), 7.24 (d, 1H, J ) 2 Hz, calix arom), 7.0-
7.2 (m, 12 H), 6.85-6.92 (m, 2H), 6.59 (s, 1H), 4.69 (d, 1H, J ) 14
Hz, calix-CH2), 4.54 (d, 1H, J ) 17 Hz, calix-CH2), 4.50 (d, 2H, J )
15 Hz, calix-CH2), 4.44 (d, 1H, J ) 17 Hz, calix-CH2), 4.35 (d, 1H, J
) 17 Hz, calix-CH2), 3.87 (d, 1H, J ) 17 Hz, calix-CH2), 3.70 (d, 1H,
J ) 14 Hz, calix-CH2), 3.62 (d, 1H, J ) 15 Hz, calix-CH2), 1.32 (s,
9H, t-Bu), 1.30 (s, 9H, t-Bu), 1.252 (s, 9H, t-Bu), 1.249 (s, 9H, t-Bu),
0.14 (s, 3H, exo-SiMe), 0.07 (s, 9H, SiMe3), -0.02 (s, 9H, SiMe3),
-1.40 (s, 3H, endo-SiMe). 13C NMR (C6D6): δ 212.5 (TiCR), 159.9,
159.7, 151.0, 150.9, 149.7, 148.7, 148.6, 146.9, 145.0, 143.9, 143.2,
142.8, 137.3, 133.4, 130.8, 130.1, 128.5, 128.3, 127.9, 127.8, 127.7,
127.4, 127.3, 127.1, 126.6, 126.2, 126.0, 125.6, 125.1, 86.5 (CPh2),
41.2 (calix-CH2), 38.4 (calix-CH2), 37.7 (calix-CH2), 37.6 (calix-CH2),
34.23 (C(CH3)3), 34.18 (C(CH3)3), 34.12 (C(CH3)3), 33.94 (C(CH3)3),
31.82 (C(CH3)3), 31.78 (C(CH3)3), 31.61 (C(CH3)3), 31.55 (C(CH3)3),
2.3 (exo-SiMe), -0.3 (SiMe3), -2.0 (SiMe3), -2.6 (endo-SiMe). Anal.
Calcd for (DMSC)Ti{OCPh2C4(SiMe3)2H2}(C6D6)(pentane)0.5, C77.5H94-
D6O5Si3Ti: C, 74.48; H, 8.55. Found: C, 74.76; H, 8.46.
In a similar reaction, the sample was hydrolyzed while some of 6
was still present in solution. The hydrolysis product Ph2C(OH)CHd
CHSiMe3 was identified by GC-MS. EI-GC-MS (m/z): 282 (10, M+),
267 (4, M+ - CH3), 209 (10, M+ -SiMe3), 192 (100, M+ - HOSiMe3),
182 (40, M+ - HCdCHSiMe3), 105 (75, PhCO+), 73 (60, Me3Si+).
(DMSC)Ti{OC(p-MeC6H4)2C2(SiMe3)H} (7) and (DMSC)Ti-
{OC(p-MeC6H4)2C4(SiMe3)2H2} (10). (DMSC)Ti{OC(p-MeC6H4)2-
C(p-MeC6H4)2O} (3) (23.4 mg, 0.020 mmol) was dissolved in 0.6 mL
of C6D6, and Me3SiCtCH (11.3 µL, 0.080 mmol) was added to it.
The reaction is slow at 22 °C; the solution contained 3, 7, and 10 in
ca. 50:45:5 ratio after 5 h. Under all the reaction conditions attempted,
the final product mixture was ca. 85% of 10 and 15% of a DMSC-
based compound. Whereas aromatic and calixarene CH2 resonances
are difficult to assign to specific components of the mixture due to
spectra overlap, the remaining peaks could be identified and assigned.
1H NMR (unobstructed resonances of 7) (C6D6): δ 8.24 (s, 1H, Câ-
H), 2.20 (s, 6H, MeC6H4), 1.29 (s, 18H, t-Bu), 1.17 (s, 18H, t-Bu),
0.16 (s, 9H, SiMe3), -0.19 (s, 3H, exo-SiMe), -1.58 (s, 3H, endo-
1
A single-crystal suitable for an X-ray diffraction study was obtained
by inducing crystallization of 9 from hexamethyldisiloxane by addition
of a small amount of benzene at ambient temperature. A sample of 9
was decomposed with H2O in ether. The suspension was allowed to
stand for 15 min, and then the insolubles were filtered off and the filtrate
was analyzed by GC-MS. Ph2C(OH)CHdC(SiMe3)CHdCH(SiMe3)
was the only species observed. EI-GC-MS (m/z): 380 (1, M+), 362 (1,
M+ - H2O), 307 (12, M+ - SiMe3), 281 (5, M+ -HCdCHSiMe3),
259 (M+ - 121), 217 (M+ - OH, -2SiMe3), 187 (9, M+ - 193), 147
(8, M+ - 233), 105 (18, PhCO+), 73 (100, Me3Si+).
SiMe). H NMR (unobstructed resonances of 10) (C6D6): δ 6.59 (s,
1H), 2.27 (s, 6H, MeC6H4), 2.10 (s, 6H, MeC6H4), 1.31 (s, 9H, t-Bu),
1.29 (s, 9H, t-Bu), 1.28 (s, 9H, t-Bu), 1.23 (s, 9H, t-Bu), 0.14 (s, 3H,
exo-SiMe), 0.07 (s, 9H, SiMe3), -0.03 (s, 9H, SiMe3), -1.40 (s, 3H,
endo-SiMe).
(DMSC)Ti{OC(p-MeC6H4)2C4But(SiMe3)H2} (8). Me3SiCtCH
(0.20 mL, 1.415 mmol) and (DMSC)Ti{OC(p-MeC6H4)2C2ButH} (5)
(0.450 g, 0.431 mmol) were dissolved in 10 mL of toluene in a heavy-
walled pressure tube equipped with a Teflon stopcock. The solution
was stirred for 2 days at room temperature, and then the volatiles were
removed under reduced pressure. The residue was washed with pentane
and dried under vacuum to give 0.360 g (74%) of 8 as a pure yellow
solid. 1H NMR (C6D6): δ 7.63 (d, 1H, J ) 2 Hz, calix arom), 7.41 (d,
2H, J ) 8 Hz, p-tolyl), 7.40 (d, 1H, J ) 2 Hz, calix arom), 7.37 (d,
1H, J ) 2 Hz, calix arom), 7.33 (d, 1H, J ) 2 Hz, calix arom), 7.22
(d, 1H, J ) 2 Hz, calix arom), 7.14 (d, 1H, J ) 2 Hz, calix arom),
7.13 (d, 1H, J ) 2 Hz, calix arom), 7.04 (s, 1H), 7.03 (d, 1H, J ) 2
Hz, calix arom), 6.98 (d, 2H, J ) 8 Hz, p-tolyl), 6.96 (d, 2H, J ) 8
Hz, p-tolyl), 6.86 (d, 2H, J ) 8 Hz, p-tolyl), 6.21 (s, 1H), 4.61 (d, 1H,
J ) 14 Hz, calix-CH2), 4.55 (d, 1H, J ) 17 Hz, calix-CH2), 4.49 (d,
2H, J ) 15 Hz, calix-CH2), 4.43 (d, 1H, J ) 17 Hz, calix-CH2), 4.36
(d, 1H, J ) 17 Hz, calix-CH2), 3.88 (d, 1H, J ) 17 Hz, calix-CH2),
3.68 (d, 1H, J ) 14 Hz, calix-CH2), 3.63 (d, 1H, J ) 15 Hz, calix-
CH2), 2.26 (s, 3H, MeC6H4), 2.10 (s, 3H, MeC6H4), 1.31 (s, 9H, t-Bu),
1.29 (s, 9H, t-Bu), 1.29 (s, 9H, t-Bu), 1.21 (s, 9H, t-Bu), 1.06 (s, 9H,
t-Bu), 0.14 (s, 3H, exo-SiMe), -0.03 (s, 9H, SiMe3), -1.40 (s, 3H,
endo-SiMe). 13C NMR (C6D6): δ 214.3 (TiCR), 159.7 (TiOC), 159.5
(TiOC), 155.1, 150.8, 149.6, 146.9, 146.2, 144.9, 144.8, 143.9, 143.2,
142.7, 137.4, 135.5, 135.0, 133.4, 130.9, 130.0, 129.5, 129.1, 128.5,
127.8, 127.7, 127.5, 127.4 (br), 127.0, 126.2, 126.0, 125.9, 125.8, 125.2,
86.0 (CPh2), 41.2 (calix-CH2), 38.4 (calix-CH2), 38.1 (calix-CH2), 37.7
(calix-CH2), 34.39 (C(CH3)3), 34.36 (C(CH3)3), 34.22 (C(CH3)3), 34.15
(C(CH3)3), 33.89 (C(CH3)3), 31.9 (C(CH3)3), 31.8 (C(CH3)3), 31.6
(C(CH3)3), 31.5 (C(CH3)3), 29.4 (C(CH3)3), 21.1 (MeC6H4), 20.9
(MeC6H4), 2.5 (exo-SiMe), -0.3 (SiMe3), -2.6 (endo-SiMe). Anal.
Calcd for C71H89O5Si2Ti: C, 75.70; H, 7.96. Found: C, 75.47; H, 8.06.
(DMSC)Ti{OCPh2C4(SiMe3)2H2} (9). (DMSC)Ti(OCPh2CPh2O)
(2) (0.600 g, 0.535 mmol), Me3SiCtCH (0.5 mL, ca. 3.50 mmol), and
10 mL of heptane were charged into a 50 mL reaction vessel equipped
with a Teflon stopcock. The vessel was closed off and heated at 90 °C
for 3 h. The volatiles were removed under reduced pressure, and then
benzophenone was removed by sublimation. The sublimation residue
was dissolved in 2 mL of pentane, and 0.5 mL of C6D6 was added to
Typical Procedure for Kinetic Study of the Reaction between
(DMSC)Ti{OC(p-MeC6H4)2C(p-MeC6H4)2O} (3) and ButCtCH
under Pseudo-First-Order Conditions. A 0.400 mL (0.0208 mmol)
volume of a 0.0521 M stock solution of (DMSC)Ti{OC(p-MeC6H4)2-
C(p-MeC6H4)2O} (3) in C6D6 was added into a J. Young NMR tube,
followed by 0.200 mL of a 2.00 M stock solution of ButCtCH (0.401
mmol, 19.3 equiv) and 0.300 mL of C6D6. This resulted in 0.900 mL
of a 0.0232 M solution of 3 and a 0.445 M solution of ButCtCH. The
tube was vigorously shaken and placed into the spectrometer at a set
1
temperature (50 °C). The first H NMR spectrum (at time ) 0) was
recorded immediately after inserting the sample in the spectrometer.
Spectra were recorded every 10 min thereafter. The dependence of the
reaction on [3] was determined by varying the concentration of 3 while
conducting each experiment in C6D6 at 50 °C, using an identical amount
of ButCtCH (0.200 mL, 0.401 mmol) and the same total volume (0.900
mL).
Typical Procedure for Determining the Reaction Dependence on
the Concentration of ButCtCH. A 0.200 mL volume of a 0.0508 M
stock solution of (DMSC)Ti{OC(p-MeC6H4)2C(p-MeC6H4)2O} (3) in
C6D6 (0.0102 mmol) was added into a J. Young NMR tube, followed
by 0.500 mL (0.500 mmol, 49.2 equiv) of a 1.00 M stock solution of
ButCtCH and then 0.100 mL of C6D6. This resulted in 0.800 mL of
a 0.0127 M solution of 3 and a 0.626 M solution of ButCtCH. The
tube was vigorously shaken and placed into the spectrometer at 50 °C.
1
The first H NMR spectrum (at time ) 0) was recorded immediately
after inserting the sample in the spectrometer. Spectra were recorded
for every 10 min thereafter. The dependence of the reaction on [ButCt
CH] was obtained by varying the concentration of ButCtCH while
conducting each experiment in C6D6 at the same temperature (50 °C),
using an identical amount of 3 (0.200 mL, 0.0102 mmol) and the same
total volume (0.800 mL).
Determining the Effect of Added (p-MeC6H4)2CO on the Reaction
between (DMSC)Ti{OC(p-MeC6H4)2C(p-MeC6H4)2O} (3) and ButCt
9
J. AM. CHEM. SOC. VOL. 124, NO. 41, 2002 12219

A R T I C L E S
Kingston et al.
Table 1. Crystallographic Data for 5‚C5H12 and 9‚2C6H6
5‚C H
Scheme 3
9‚2C H
5
12
6
6
formula
fw
T, K
C72H94O5SiTi
1115.46
173(1)
C81H100O5Si3Ti
1285.78
173(2)
cryst system
space group
Z
monoclinic
Pn
2
triclinic
P1h
2
a, Å
b, Å
c, Å
13.986(1)
11.690(1)
19.815(2)
90
97.94(1)
90
11.242(2)
13.483(3)
25.368(5)
79.350(10)
84.900(10)
76.400(10)
3668.8(13)
1.164
R, deg
â, deg
γ, deg
V, Å3
3208.6(5)
1.155
d
calcd, g/cm3
final R indices [I > 2σ(I)]: R1, wR2
0.047, 0.109
0.113, 0.053
0.0970, 0.2522
0.2558, 0.1071
80 °C is (DMSC)2Ti (4),10 which can be obtained in a more
wR2, R1 (all data)
direct fashion by reaction between (DMSC)Mg11 and (DMSC)-
12
TiCl2 (Scheme 2). Compound 4 is quite soluble in aromatic
Scheme 2
and ethereal solvents but is only modestly soluble in aliphatic
hydrocarbons. Both solution 1H and 13C NMR data are consistent
with the existence of the DMSC ligand of 4 in 1,2-alternate
conformation.13 In the 1H NMR spectrum of 4, four But
resonances of equal intensity, eight doublets of equal intensity
(for calix[4]arene methylene protons), eight aromatic resonances
of equal intensity, and two resonances (corresponding to exo-
and endo-SiMe groups)13 are observed (see Experimental
Section). Each DMSC unit is therefore absent of any local
symmetry, but the molecule possesses a C2 axis which renders
the two DMSC units equivalent. Thus, the geometry about
titanium is pseudotetrahedral, with the two DMSC units oriented
perpendicular to one another.
Reactions between (DMSC)Ti(OCAr2CAr2O) (2, Ar ) Ph,
and 3, Ar ) p-MeC6H4) Complexes and Terminal Alkynes.
Reactions of terminal alkynes with titanapinacolates 2 and 3
proceed in a rather unusual manner, wherein alkynes displace
one of the Ar2CO units of the titanapinacolate (2,5-dioxatitana-
cyclopentane) complex to form 3,5-disubstituted 2-oxatitana-
cyclopent-4-enes14 (5-7, Scheme 3). The reactions are quite
slow at room temperature but proceed to completion in under
CH. A 0.200 mL (0.0101 mmol) volume of a 0.0507 M stock solution
of (DMSC)Ti{OC(p-MeC6H4)2C(p-MeC6H4)2O} (3) in C6D6 was added
into a J. Young NMR tube, followed by 0.250 mL of a 2.00 M stock
solution of ButCtCH (0.501 mmol, 49.5 equiv), 0.050 mL of a 0.203
M stock solution of (p-MeC6H4)2CO (0.0101 mmol), and 0.300 mL of
C6D6. This resulted in 0.800 mL of a 0.0127 M solution of 3, a 0.626
M solution of ButCtCH, and a 0.0127 M solution of (p-MeC6H4)2-
CO. The tube was vigorously shaken and placed into the spectrometer
at 50 °C. The first 1H NMR spectrum (at time ) 0) was recorded
immediately after inserting the sample in the spectrometer. Spectra were
recorded for every 5 min thereafter. A second experiment was conducted
in C6D6 and at 50 °C, using identical amounts of 3 and ButCtCH as
above, 0.0303 mmol (3 equiv) of (p-MeC6H4)2CO, and the same total
volume (0.800 mL).
(10) Whether 4 was formed via comproportionation reaction of 3 was of interest
to us. Thus, a toluene-d8 solution of 3 was heated at 95 °C in a screw-
1
capped NMR tube and monitored periodically by H NMR. After ∼16 h,
1H NMR revealed a complete disappearance of 3 and formation of several
DMSC-containing products: 4 was the major DMSC-containing product
(>50%). Importantly, only two singlets were observed in the p-tolyl region
of the spectrum at δ 2.06 ppm (p-MeC6H4)2CO and 1.94 ppm in ∼2:3
ratio. We tentatively assign the latter singlet to the p-tolyl methyl groups
of the presumed comproportionation coproduct Ti{OC(p-MeC6H4)2C(p-
MeC6H4)2O}2. Consistent with this proposal, GC-MS analysis of the reaction
mixture (after hydrolysis and filtration through a plug of silica) revealed
(p-MeC6H4)2CdC(p-MeC6H4)2, (p-MeC6H4)2CO, and HOC(p-MeC6H4)2-
C(p-MeC6H4)2COH in ∼1:3:10 ratio as the only p-tolyl-containing organic
products.
Crystallographic Studies. The crystal data for 5‚C5H12 and 9‚2C6H6
are collected in Table 1. Further details of the crystallographic study
are given in the Supporting Information.
(11) Mg can be generated in solution, but it tends to precipitate out of solution
after a short period of time. We have not been able to redissolve this
precipitate.
(12) Ozerov, O. V.; Ladipo, F. T.; Patrick, B. O. J. Am. Chem. Soc. 1999, 121,
7941.
Results and Discussion
(13) That the DMSC ligand exists in 1,2-alternate conformation is apparent from
the NMR resonances of the endo- and exo-methyls of the bridging SiMe2
unit. Invariably, 1H and 13C NMR resonances for the endo-SiMe group
(located inside of the calix[4]arene cavity) are strongly shielded compared
to corresponding signals for the exo-SiMe group (located outside of the
calix[4]arene cavity), due most probably to ring current effect. For a
discussion of ring current effect, see: Gunther, H. NMR Spectroscopy: An
Introduction; Wiley: New York, 1980; pp 77-86.
(14) Previous work in our group has demonstrated that DMSC ligation favors
introduction of substituents at the 3- (endo-â) and 5- (exo-R) positions of
disubstituted five-membered titanacycles. Such location of one substituent
outside (exo) of the calixarene cavity and the other substituent at the endo-â
(inside the cavity) position apparently reduces unfavorable steric interac-
tions.9
Reaction of (DMSC)Ti{1,2,4-(Me3Si)3C6H3} (1) with (p-
MeC6H4)2CO. We previously reported that reaction of (DMSC)-
Ti{1,2,4-(Me3Si)3C6H3} (1) with Ph2CO or (p-MeC6H4)2CO in
heptane proceeded to completion in under 3 h at 80 °C to
produce titanapinacolates 2 and 3, respectively.5 Whereas these
conditions are efficient for the preparation of 2, we have since
discovered that 3 is more reliably produced by carrying out the
reaction at 65 °C for 1 h (Scheme 2). The predominant product
formed in the reaction of 1 with (p-MeC6H4)2CO (2 equiv) at
9
12220 J. AM. CHEM. SOC. VOL. 124, NO. 41, 2002

Fragmentation Reactions of Titanapinacolates
A R T I C L E S
peculiarity of the structure. It is most likely due to distortion of
the DMSC ligand by unfavorable steric interaction between the
But group of one of the p-tert-butylcalix[4]arene phenol units
and tolyl substituents of the 2-oxatitanacyclopent-4-ene ring.
This evidently causes the phenol unit to bend away and bring
the oxygen close to titanium (Figure 1). The bond lengths within
the five-membered titanacycle are within the expected ranges;15,16
the C(48)-C(49) bond distance of 1.328(6) Å is comparable
to that reported [1.332(6) Å] for the doubly bonded carbons in
the related 2-oxatitanacyclopentene Cp2Ti(CPhCPhCMe2O).16
The distortion of the DMSC ligand of 5, observed in the solid-
1
state structure, is also manifested in solution. In C6D6, the H
NMR chemical shifts of both the exo- and endo-SiMe groups
are unusually upfield at δ -0.21 and -1.57 ppm, respectively.17
The Ti-O(4) interaction apparently draws the endo-SiMe group
deeper inside the calix[4]arene cavity, increasing the ring current
effect, due to the two proximal aromatic rings, that it experi-
ences. On the other hand, the aromatic ring of the phenol unit
that is bent away by unfavorable steric interaction with the Ar2C
unit exerts a certain ring current effect on the exo-SiMe group.
In contrast to 5, exclusive formation of 2-oxatitanacyclopent-
4-enes 6 and 7 did not occur. Instead, 6 and 7 undergo further
reaction with alkyne to yield oxatitanacycloheptadienes 9 and
10, respectively (Scheme 3). Consequently, 2 (3), 6 (7), and 9
(10) are all present in solution for some time. Our attempts to
favor predominant formation of 6 (7) by varying the reaction
conditions have so far met with limited success. Consequently,
Figure 1. Molecular structure of (DMSC)Ti{OC(p-MeC6H4)2C2ButH} (5)
(50% probability ellipsoids).
Table 2. Selected Bond Distances (Å) and Angles (deg) for 5 and
9
5
9
Ti-O(1)
1.801(3)
1.815(3)
2.497(3)
1.835(3)
2.122(4)
1.328(6),
105.43(13)
85.09(11)
108.43(14)
84.57(12)
142.94(13)
83.75(11)
103.19(15)
104.39(16)
165.38(14)
82.20(16)
Ti-O(1)
1.785(5)
1.808(5)
1.834(5)
2.077(8)
1.536(11)
1.314(11)
1.496(11)
1.341(11)
Ti-O(2)
Ti-O(2)
Ti-O(4)
Ti-O(5)
Ti-C(51)
Ti-O(5)
1
a completely unobstructed H NMR spectrum could not be
Ti-C(49)
C(47)-C(48)
C(48)-C(49)
C(49)-C(50)
C(50)-C(51)
O(1)-Ti-O(2)
O(1)-Ti-O(5)
O(2)-Ti-O(5)
C(48)-C(49)
O(1)-Ti-O(2)
O(1)-Ti-O(4)
O(1)-Ti-O(5)
O(2)-Ti-O(4)
O(2)-Ti-O(5)
O(4)-Ti-O(5)
observed for 6 or 7. However, only one Me3Si group belonging
to 6 (7) could be observed by 1H NMR and the chemical shifts
of the endo- and exo-SiMe groups and the But groups of the
DMSC ligand of 6 (7) are very close to those observed for 5,
supporting their structural similarity. Reaction of 2-oxatitana-
cyclopent-4-enes 5-7 with Me3SiCtCH proceeds via formal
insertion into the Ti-C bond to yield the corresponding
2-oxatitanacycloheptadienes 8-10 (Scheme 3). In contrast, 5-7
did not react with ButCtCH under identical conditions. The
latter result probably reflects steric effect of the 2-oxatitanacy-
clopentene R-carbon substituent as well as differences in steric
and electronic properties between the two alkynes.18 While the
conversion of 5 to 8 is quantitative (by 1H NMR), a minor side
product is formed together with 9 and 10 (in <5% and ∼15%
105.6(2)
115.2(2)
118.6(2)
102.3(3)
114.9(3)
99.4(3)
O(1)-Ti-C(51)
O(2)-Ti-C(51)
O(5)-Ti-C(51)
O(1)-Ti-C(49)
O(2)-Ti-C(49)
O(4)-Ti-C(49)
O(5)-Ti-C(49)
3 h at 80 °C. Although the very high solubility of 5 in
hydrocarbon solvents hampered its isolation in high yield, it
was obtained as yellow crystals and fully characterized. Both
microanalysis and solution NMR (1H and 13C) data for 5 are
consistent with the proposed formulation. In the 13C NMR
spectrum, the R-carbon of 5 and the Ar2C carbon (Ar )
p-MeC6H4) resonate at δ 221.1 and 85.9 ppm, respectively.
Analogous data (δ 219.7 ppm for CR and δ 95.8 ppm for Ph2C)
were reported for the related 2-oxatitanacyclopentene (ArO)2Ti-
(CEtCEtCPh2O) (Ar ) 2,6-Ph2C6H3).15
The molecular structure of 5 was determined by single-crystal
X-ray diffraction analysis (Figure 1), and selected metrical
parameters are listed in Table 2. The geometry about Ti is best
described as distorted trigonal bipyramidal with the R-carbon
of the 2-oxatitanacyclopent-4-ene ring [C(49)] occupying one
of the axial positions and one of the silicon-bridged oxygen
atoms [O(4)] occupying the other [O(4)-Ti-C(49) angle )
165.38(14)°]. The Ti-O(4) bond is long at 2.497(3) Å and
indicative of a weak interaction. The Ti center of 5 is not very
electrophilic hence coordination of O(4) to Ti is an unexpected
1
yield, respectively). On the basis of the H NMR data, we
assume that the side product is an isomeric species, possessing
a different disposition of the Me3Si groups on the metallacycle.
The amount of side product formed in each case does not
increase with reaction temperature between 20 and 80 °C. When
an alkyne such as HCtCCH2OSiMe3, with a less bulky
substituent, is allowed to react with 2, a mixture of three
(DMSC)Ti-based compounds is produced along with 1 equiv
of Ph2CO. Analysis of the 1H NMR data revealed that each of
the products possessed two Me3Si groups (from the correlation
of the intensities of the endo-SiMe and the SiMe3 resonances).
(16) Shur, V. B.; Burlakov, V. V.; Yanovsky, A. I.; Petrovsky, P. V.; Struchkov,
Yu. T.; Vol’pin, M. E. J. Organomet. Chem. 1985, 297, 51.
(17) For typical exo- and endo-SiMe group chemical shifts, see for example:
Ozerov, O. V.; Ladipo, F. T.: Rath, N. P. J. Organomet. Chem. 1999,
586, 223.
(18) The polarization in the π* orbital of alkyl- and silylacetylenes has been
calculated (see: Stockis, A.; Hoffmann, R. J. Am. Chem. Soc. 1980, 102,
2952). The π* orbital of terminal alkylacetylene has its largest lobe on the
substituted carbon while the π* orbital of terminal silylacetylene has its
largest lobe on the unsubstituted carbon.
(15) Hill, J. E.; Balaich, G. J.; Fanwick, P. E.; Rothwell, I. P. Organometallics
1993, 12, 2911.
9
J. AM. CHEM. SOC. VOL. 124, NO. 41, 2002 12221

A R T I C L E S
Kingston et al.
Scheme 4
two molecules of benzene. The geometry about Ti is best
described as pseudotetrahedral, and the oxatitanacycloheptadiene
ring is decidedly nonplanar. The solid-state structure confirmed
the head-to-tail regiochemistry of the Me3Si substituents of the
metallacycle, as was deduced from solution NMR data.
Mechanistic Considerations. The precise mechanism of the
metallacyclic C-C bond rupture reported for titanapinacolates
2 and 3 in this study was of interest to us. The relative ease of
fragmentation reactions of 2 and 3 is probably due to the long
metallacyclic C-C bond. In fact, benzopinacol (HOCPh2CPh2-
OH) and its derivatives have been shown to be amenable to
photolytic and thermolytic fragmentation, as well as oxidation
to benzophenone.19 However, L2Ti(OCPh2CMe2O) (L ) N,N’-
dimethylaminotroponiminate) possesses a metallacyclic C-C
bond distance [1.610(2) Å] similar to that for 2, and reactivity
analogous to that observed for 2 and 3 was not reported.20 The
two most probable mechanisms that can be envisioned for
fragmentation reactions of titanapinacolates 2 and 3 with
terminal alkynes are (i) an associative mechanism, involving
coordination of the alkyne to titanium prior to rate-limiting
rupture of the titanacyclic C-C bond, and (ii) a preequilibrium
mechanism, involving reversible formation of a (DMSC)Ti(η2-
OCAr2) intermediate prior to rate-limiting reaction with alkyne
(Scheme 4). In an attempt to differentiate between the two
mechanistic possibilities, we monitored reactions of (DMSC)-
Ti(OCAr2CAr2O) (2, Ar ) Ph, and 3, Ar ) p-MeC6H4) with
Ph213CO (1.1 equiv) at room temperature by 13C NMR
spectroscopy.21 Essentially, statistical scrambling of Ph213CO
into both endo- and exo-positions of the titanapinacolate rings
occurred within 30 min (eq 2). While this result did not allow
the two mechanistic possibilities to be unambiguously differenti-
ated,22 it does demonstrate facile reversible fragmentation of
well-characterized titanapinacolate complexes by benzophenone.
To elucidate the mechanism of these fragmentation reactions,
we conducted a kinetic analysis of the reaction of 3 with ButCt
Figure 2. Molecular structure of (DMSC)Ti{OCPh2C4(SiMe3)2H2} (9)
(50% probability ellipsoids).
It is reasonable to assume that the mixture consists of isomeric
products resulting from different alkyne insertion regiochemistry.
This result suggests that steric interaction between alkyne
substituents, as well as with the Ar2C moiety, influences the
regiochemistry of 5-10. It is worth pointing out that no reaction
occurs between 5-7 and Ar2CO. The Ar2CO moiety that is part
of the metallacycle is apparently not exchangeable. Conse-
quently, 6 does not react with Ph2CO. In addition, Ph2CO is
not incorporated in the products when the transformation of 6
into 9 is carried out in the presence of Ph2CO.
2-Oxatitanacycloheptadienes 8 and 9 were isolated as yellow
solids in high yield, while 10 was studied in situ. The solution
NMR data for 9 are consistent with the existence of the DMSC
ligand in 1,2-alternate conformation. The C1 symmetry of 9 is
maintained in C6D6 solution over the 22-80 °C temperature
range. The 13C NMR resonances of the TiCR carbons of 8 and
9 are observed at δ 214.3 and 212.5 ppm, respectively, and
compare well to the 210.8-215.8 ppm range reported for related
oxatitanacycloheptadienes by Rothwell.15 Only one of the
metallacyclic CH hydrogen signals can be identified by 1H NMR
for 9 and 10 at δ 6.59 ppm (singlet); the other signal is obscured
by the aromatic resonances. Both of these signals could be
identified for 8 at δ 7.04 and 6.21 ppm; their near-zero coupling
constant suggests head-to-tail incorporation of the Me3SiCt
CH units into the metalacycle. Compound 9 is extremely soluble
in hydrocarbon solvents, and it proved impossible to induce its
precipitation from pentane solution even after extended periods
at -78 °C. Remarkably, the incorporation of a six-membered
aromatic ring into the crystal lattice promotes the crystallization
of 9. Thus, addition of small amounts of benzene or pyridine
to pentane solutions of 9 causes (within minutes) precipitation
of crystalline solid. 1H NMR data revealed that pyridine is not
coordinated to 9 in solution. Neither 1,4-dioxane or toluene has
the same effect on the crystallization of 9. The molecular
structure of 9 was determined by X-ray crystallography (Figure
2), and selected metrical parameters are listed in Table 2. The
bond lengths within the titanacycle are within the expected
ranges.15 The asymmetric unit contained one molecule of 9 and
(19) Hillgartner, H.; Schroeder, B.; Neumann, W. P. J. Organomet. Chem. 1972,
42, C83. (b) Weiner, S. A. J. Am. Chem. Soc. 1971, 93, 6978. (c) Michielli,
R.; Elving, P. J. J. Am. Chem. Soc. 1969, 91, 6864.
(20) Steinhuebel, D. P.; Lippard, S. J. J. Am. Chem. Soc. 1999, 121, 11762.
(21) In a typical experiment, 15 mg (0.0134 mmol) of (DMSC)Ti(OCPh2CPh2O)
(2) and 3.00 mg (1.1 equiv) of Ph213CO were dissolved in 0.7 mL of C6D6
in a screw-capped NMR tube under N2 atmosphere. The tube was vigorously
shaken, and the reaction was monitored by 13C NMR spectroscopy (1024
scans were recorded). Two singlet resonances at δ 110.1 and 111.1 ppm
(for endo-Câ and exo-Câ metallacyclic carbons, respectively) and two
doublet resonances centered at 110.1 and 111.1 ppm (2JC-C ) 36.6 Hz)
were observed in approximately 2: 2: 1 ratio. Similar results were obtained
when 3 was utilized.
(22) Had no incorporation of Ph213CO into the titanapinacolate ring been
observed, a mechanism involving reversible formation of a (DMSC)Ti-
(η2-OCAr2) intermediate prior to the rate-limiting reaction with an alkyne
or ketone molecule would have been ruled out.
9
12222 J. AM. CHEM. SOC. VOL. 124, NO. 41, 2002

Fragmentation Reactions of Titanapinacolates
A R T I C L E S
Figure 3. Plot showing the disappearance of (DMSC)Ti{OC(p-MeC6H4)2-
C(p-MeC6H4)2O} (3) with time ([ButCtCH] ) 20 equiv).
CH. This reaction was chosen for study because 5 is the
exclusive product (vide supra) and the reaction proceeds at a
convenient rate over a broad temperature range. In addition, 3
affords two sets of signals through which the reaction could be
easily monitored: p-tolyl methyl resonances, as well as endo-
and exo-SiMe resonances. The kinetic studies were conducted
at 50 °C, under pseudo-first-order conditions, by adding ∼20
equiv of ButCtCH to a benzene-d6 solution of 3 and monitoring
the reactions at various time intervals by 1H NMR spectroscopy.
The concentration of ButCtCH was then varied in a second
set of experiments. Plots of the disappearance of 3 with time
and of the observed rate constants (kobs) versus [ButCtCH] are
depicted in Figures 3 and 4, respectively. The reactions showed
first-order dependence on both [3] and [ButCtCH], confirming
that the rate-limiting step in the reaction involves both ButCt
CH and a titanium species.
Figure 4. Plot of the observed rate constants (kobs) versus [ButCtCH].
Clearly, if fragmentation reactions described in this study
proceed via a preequilibrium pathway, the presumed (DMSC)-
Ti(η2-OCAr2) intermediate (A) is so reactive that it does not
accumulate to an appreciable level. Indeed, (DMSC)Ti(η2-
OCAr2) species will most likely be unstable and can be expected
to react rapidly with the released Ar2CO molecule to regenerate
the titanapinacolate complex. Consistent with this suggestion,
no ligand-free (RO)2Ti(η2-ketone) (R ) alkyl or aryl) complexes
are known and very few well-characterized mononuclear group
4 metal-ketone complexes bearing alkyl or aryl substituents
have been reported, including (TC-3,5)Hf(η2-OC(CH2Ph)2)
(where TC-3,5 ) tropocorand ligand)23 and (OC6H3Ph2-2,6)2-
Ti (η2-Ph2CO)(PMe3).24 Since [ButCtCH] . [3] and [A] ,
[3] under the conditions of our kinetic studies, the steady-state
approximation25 can be applied to the concentration of the
(DMSC)Ti{η2-OC(p MeC6H4)2} intermediate (A) as follows:
d[5]
dt
) k2[A]ss[ButCtCH]
k2k1[3][ButCtCH]
)
k-1[Ar2CO] + k2[ButCtCH]
When k2[ButCtCH] . k-1[Ar2CO],
d[5]
) k1[3]
dt
Ar ) p-MeC6H4
It then follows that if the reaction proceeds via a preequilibrium
pathway, the observed rate constant (kobs) will equal k1 (eq 3)
and the reaction rate will be retarded by added ketone (path a,
Scheme 4). On the other hand, the reaction rate will be
independent of added ketone if the fragmentation reactions occur
via an associative mechanism (path b, Scheme 4). When the
reaction of 3 with ButCtCH (∼50 equiv) was monitored in
the presence of 1 and 3 equiv of (p-MeC6H4)2CO, we observed
d[A]
dt
) k1[3] - k-1[Ar2CO][A] - k2[A][ButCtCH] ) 0
k1[3]
[A]ss )
(3)
k-1[Ar2CO] + k2[ButCtCH]
retardation of the reaction rate. Thus, kobs ) 7.50 × 10-4
(
)
4.84 × 10-5 s-1 in the absence of added ketone while kobs
Thus,
2.34 × 10-4 ( 8.31 × 10-6 s-1 and 9.95 × 10-5 ( 1.08 ×
10-6 s-1 in the presence of one and three equivalents of added
ketone, respectively. This result, together with facile reversible
exchange of Ph213CO into endo- and exo-positions of titanapi-
nacolates 2 and 3 (vide supra), strongly supports a mechanism
(23) Scott, M. J.; Lippard, S. J. J. Am. Chem. Soc. 1997, 119, 3411.
(24) Hill, J. E.; Fanwick, P. E.; Rothwell, I. P. Organometallics 1992, 11, 1771.
(b) Thorn, M. G.; Hill, J. E.; Warantuke, S. A.; Johnson, E. S.; Fanwick,
P. E.; Rothwell, I. P. J. Am. Chem. Soc. 1997, 119, 8630.
(25) See for example: Espenson, J. H. Chemical Kinetics and Reaction
Mechanisms; McGraw-Hill: New York, 1981.
9
J. AM. CHEM. SOC. VOL. 124, NO. 41, 2002 12223

A R T I C L E S
Kingston et al.
involving reversible formation of a (DMSC)Ti(η2-OCAr2)
species prior to rate-limiting reaction with an alkyne or ketone
molecule.26
CH are consistent with a mechanism that involves reversible
dissociation of the titanapinacolate complexes into (DMSC)-
Ti(η2-OCAr2) species with release of a ketone molecule,
followed by rate-limiting reaction of the (DMSC)Ti(η2-OCAr2)
species with an alkyne or ketone molecule (i.e. a preequilibrium
mechanism). Further studies of the reaction chemistry of these
and related compounds with various unsaturated organic sub-
strates are underway in our laboratory.
Conclusions
While fragmentation of metallapinacolate intermediates has
previously been invoked to explain reactivity patterns in pinacol
and McMurry reactions, the present studies demonstrate unam-
biguously the fragmentation of well-characterized titanapina-
colate complexes by terminal alkynes and aromatic ketones.
More importantly, both structural parameters of the titanapina-
colate complexes and kinetic investigations of the reaction of
(DMSC)Ti{OC(p-MeC6H4)2C(p-MeC6H4)2O} (3) with ButCt
Acknowledgment. Thanks are expressed to the U.S. National
Science Foundation (Grant No. CHE-9984776) for financial
support of this work, as well as to the Chemistry Department
and the Graduate School of the University of Kentucky for the
award of fellowships to O.V.O. NMR instruments utilized in
this research were funded in part by the CRIF program of the
U.S. National Science Foundation (Grant No. CHE-9974810).
The authors also thank Professors Jack Selegue (University of
Kentucky), Clark Landis (University of WisconsinsMadison),
and Richard Eisenberg (University of Rochester) for helpful
discussions.
(26) Since the reaction between 3 and (p-MeC6H4)2CO is reversible, this result
also rules out the two alternate mechanisms depicted below: (i) a
mechanism in which reversible fragmentation of 3 to form the (DMSC)-
Ti{η2-OC(p-MeC6H4)2} species (A) occurs but A is not an intermediate in
the reaction between 3 and ButCtCH; (ii) a mechanism in which the
reversible fragmentation of 3, through a transition state species of increased
coordination number, competes with the direct reaction of 3 with ButCt
CH. In either case, the rate of formation of 5 will not be retarded by added
ketone.
(i)
Supporting Information Available: A summary of crystal-
lographic parameters, atomic coordinates and equivalent iso-
tropic displacement parameters, bond lengths and angles,
anisotropic displacement parameters, and hydrogen coordinates
and isotropic displacement parameters for 5 and 9 (PDF). This
material is available free of charge via the Internet at
http://pubs.acs.org.
3 h A + Ar2CO
3 + [ButCtCH] f 5 + Ar2CO
(ii)
3 + Ar2CO h [(DMSC)Ti(OCAr2CAr2O)(OCAr2)]q h 3 + Ar2CO
3 + [ButCtCH] f 5 + Ar2CO
Ar ) p-MeC6H4
JA0271577
9
12224 J. AM. CHEM. SOC. VOL. 124, NO. 41, 2002