Unexpected C-C bond cleavage and C-C bond formation observed in the reaction of a cationic iridium complex with heteroatom-substituted cyclopropanes

Article abstract of DOI:10.1021/ja068312a

Iridium complex Cp*(PMe₃)Ir(CH₃)OTf (1) and alkylaminocyclopropanes react via a well-understood initial C-H bond activation with a subsequent C-C bond-breaking step to yield a variety of π-allyl complexes [Cp*(PMe₃)Ir(η³-C₃H₄NR¹R²)][OTf] (R¹ = R² = Bn; R¹ = Bn, R² = Ph) and methane. However, diphenylamino, alkoxy, or siloxycyclopropanes, when treated with complex 1, yield a new π-allyl complex [Cp*(PMe₃)Ir(η³-C₃H₄CH₃)][OTf] (8) and the corresponding amine or alcohol. The methyl group initially bound to iridium is no longer extruded as methane, but instead is incorporated into the allyl moiety to give a new carbon-carbon bond. A detailed mechanistic study provides evidence in support of an initial C-C bond activation mechanism as opposed to the initial C-H bond activation observed with other known substrates. Copyright

Full text of DOI:10.1021/ja068312a

Published on Web 01/06/2007
Unexpected C-C Bond Cleavage and C-C Bond Formation Observed in the
Reaction of a Cationic Iridium Complex with Heteroatom-Substituted
Cyclopropanes
Mitchell R. Anstey, Cathleen M. Yung, Juana Du, and Robert G. Bergman*
Department of Chemistry, UniVersity of California and DiVision of Chemical Sciences, Lawrence Berkeley National
Laboratories, Berkeley, California 94720
Received November 20, 2006; E-mail: rbergman@berkeley.edu
Scheme 1
Metal-mediated cyclopropane activation, by reaction at either
C-H or C-C bonds, is a potentially convenient and direct process
to obtain useful C3 moieties for organic synthesis.1,2 Both types of
reaction are known to occur at late metal centers, but the factors
that control the favored mode of attack in a particular case are still
not clear. We wish to report a series of substituted cyclopropane
activation reactions at a Cp*LIr(III) center (Cp* ) pentamethyl-
cyclopentadienyl, L ) PMe3). Although the products are superfi-
cially similar (allyl complexes), a detailed mechanistic study reveals
that certain types of heteroatom substituents shift the initial
activation process dramatically from the previously observed C-H
to a new C-C bond activation pathway. In addition, the new
Scheme 2
pathway exhibits, on the way to products, an unusual series of
mechanistic steps, including overall methyl migratory insertion
which has been very rarely observed at Cp*LIr centers.2c,3
Previously, Cp*(PMe3)Ir(CH3)OTf (1) (OTf ) OSO2CF3) was
reported by our group to activate cyclopropane, yielding π-allyl
complex 2 (Scheme 1).2c Subsequent kinetic isotope effect (KIE)
experiments demonstrated that this reaction has a kH/kD ) 3.8 (
0.3.4 This indicates that the reaction might proceed by initial C-H
bond activation, followed by extrusion of methane, to afford
cyclopropyliridium complex 3, which parallels extensive C-H bond
activation studies by our group with complex 1.3
With the intent of exploring the scope of this reactivity and
isolating model intermediates, several heteroatom-substituted cy-
clopropanes were examined. Treatment of complex 1 with N,N-
dibenzylcyclopropylamine (4) or N-benzyl-N-phenylcyclopropyl-
amine (5) furnished [Cp*(PMe3)Ir(η3-C3H4NBn2)][OTf] (6) and
[Cp*(PMe3)Ir(η3-C3H4NBnPh)][OTf] (7), respectively, analogous
to the product observed with parent cyclopropane. We assume that
these reactions proceed by the well-precedented pathway illustrated
in Scheme 1.
In contrast to the results summarized above, alkoxy and siloxy-
substituted cyclopropanes produced unexpected products. Instead
of methane loss, alcohol is eliminated to form [Cp*(PMe3)Ir(η3-
C3H4CH3)][OTf] (8) (Scheme 2). Interestingly, reaction of N,N-
diphenylaminocyclopropane (12) with 1 also eliminates secondary
amine Ph2NH rather than methane to afford methyl-migrated
product 8. Apparently, cyclopropanes bearing better anionic leaving
groups favor heteroatom elimination to furnish the unexpected
methyl-migrated products.
Because methane is not a byproduct of these eliminations, Cp*-
(PMe3)Ir(13CH3)OTf was used to track the position of the iridium-
bound methyl in the methallyl moiety. With tert-butyl cyclopropyl
ether 10, the 13C label is transferred to the terminal methyl of the
allyl moiety in complex 8, as verified by 1H and 13C NMR
spectroscopy. Additional positional information was obtained by
reaction of complex 1 with (1-methylcyclopropyl)(methyl) ether
13, which yielded [Cp*(PMe3)Ir(η3-CH3C3H3CH3)][OTf] (9). This
labeling study shows the final position of the tertiary cyclopropyl
carbon once complex 8 has been formed.
Although we do not observe scrambling of the 13C label in this
reaction, KIE measurements are complicated by H/D scrambling
of the cyclopropyl and iridium-methyl hydrogens.5 Cyclopropane
10 was treated with complex 1-d3, and deuterium was found not
only in the methyl position of the allyl group but also in the 1 and
3 positions. Surprisingly, no deuterium was incorporated at the
central carbon of the allyl in complex 8. Control experiments
revealed that H/D scrambling is an intramolecular exchange process,
and the H/D composition of complex 8 is static once it is formed
(see Supporting Information).
On the basis of these observations, we hypothesized that the
reaction between 1 and a 1:1 mixture of siloxy ethers 11/11-d5
should result in an artificially large KIE for this reaction on the
basis of the additional hydrogen introduced by the iridium methyl
into the allyl. The kH/kD of 1.22 ( 0.03 that we measured in this
way is therefore a maximum value for this reaction. The modest
magnitude of this effect, which is dramatically different from the
primary isotope effects observed with arenes3 and parent cyclo-
propane, argues against a C-H bond activation occurring during
or before the rate-determining step. Therefore, on the basis of the
type of reactivity observed for cyclopropanes with metal systems
other than Cp*(L)Ir,2d,e,i,j the most reasonable mechanistic alternative
for this reaction involves initial C-C bond cleavage to yield
metallocyclobutane 14 (Scheme 3).
With the knowledge that alcohol is a product of this reaction, it
was of interest to determine a mechanism for its elimination.
Because the complex is cationic, it seemed unlikely that an alkoxide
anion would dissociate to make a dicationic complex. Therefore,
we considered the possibility that adventitious acid was involved
9
776
J. AM. CHEM. SOC. 2007, 129, 776-777
10.1021/ja068312a CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
Scheme 3
Figure 1. ORTEP diagram of Cp*(PMe3)Ir(C4H8OC4H9)Cl (19). Thermal
ellipsoids are shown at the 50% probability level. Hydrogen atoms have
been omitted for clarity.
istry of the overall methyl-migration and isotope-labeling phenom-
ena we observe.
In summary, we have shown that several heteroatom-substituted
cyclopropanes react with Cp*(PMe3)Ir(CH3)OTf via C-C bond
cleavage and formation. This is an unexpected mode of reactivity
for a complex known to favor reaction by C-H bond activations
and to avoid methyl-migratory insertion reactions.2c,3 On the basis
2
of H and 13C isotope-labeling studies, dependence on acid, and
literature precedent, we propose the mechanism pictured in Scheme
3. Further experiments and computations directed at understanding
the selectivity of these reactions and the divergent reactivities
observed for amino- and alkoxy-substituted cyclopropanes are in
progress.
in the ROH elimination step. To test this hypothesis, hindered 2,6-
disubstituted pyridine bases were added to the reaction between 1
and cyclopropane 10. This changed the observed product dramati-
cally: instead of an allyl complex, a diastereomeric mixture of the
thermally sensitive hydride complex 18 was observed. Further
investigation revealed this complex to be a transient intermediate
in the reaction of 1 and cyclopropane 10 in the absence of base as
observed by NMR spectroscopy. Cp*(PMe3)Ir(13CH3)OTf was
subjected to the same basic conditions, and the 13C carbon was
tracked by 13C NMR spectroscopy to the terminal olefinic methylene
group of the carbon chain in 18.
Complex 18 is most probably reached by C-C reductive
elimination in the initial metallacycle intermediate 14 to afford the
cationic complex 15, followed by C-H bond activation of the
terminal methyl to yield 16. Reductive elimination and â-hydride
elimination would then furnish complex 18 as a mixture of alkene
diastereomers. To transfer the iridium center from one end of the
carbon chain to the other, we propose this C-H bond activation/
reductive elimination scheme (rather than the more common chain-
walking mechanism) to account for the lack of H/D scrambling
into the 2-position of allyl complex 8. Complex 17, the proposed
immediate precursor to 18, can also be trapped with a number of
nucleophiles. When chloride is used, a single isomer 19 (Cp*(PMe3-
Ir(C4H8OC4H9)Cl, Figure 1) is formed. Supporting our hypothesis
of acid catalysis, hydride complex 18 proceeds cleanly to allyl
complex 8 when an excess of triflic acid is added, demonstrating
its validity as an intermediate in the mechanism.
The fact that added base arrests this reaction at compound 18
demonstrates that C-C bond formation occurs before elimination
of the alkoxy-, siloxy-, or diphenylamino-substituents (See Sup-
porting Information for further discussion). The 13C-labeling study
reveals the regiochemistry of the C-C bond formation. Finally,
observing that acid is required to convert intermediate hydride 18
to allyl complex 8, we propose elimination of the heteroatom
substituent as the most probable next step.
Acid-promoted elimination of alcohol from hydride complex 18
should furnish complex 21. Several groups have proposed analogous
structures as intermediates in the hydrovinylation of ethylene by
iridium(III) complexes.6 A pathway from 17 to 21 could also go
through crotyliridium(III) complex 20. As yet, we have no evidence
to rule out either pathway: both would account for the regiochem-
Acknowledgment. We wish to thank Prof. Jack Norton
(Columbia), Mr. Hairong Guan (Columbia), Prof. Charles Casey
(Wisconsin), Prof. Michael Hall (Texas A&M), and Prof. M. Edwin
Webster (Memphis) for helpful discussions, as well as Prof. Peter
Vollhardt for suggesting that alkoxy group elimination might be
acid-mediated. Drs. Fred Hollander and Allen Oliver at the UCB
X-ray Facility are acknowledged for the crystal structure determina-
tions. This work was supported by the Director, Office of Energy
Research, Office of Basic Energy Sciences, Chemical Sciences
Division, of the U.S. Department of Energy under Contract DE-
AC02-05CH11231.
Supporting Information Available: Experimental procedures,
spectral data for unknown compounds, and a comment on the difference
in behavior of alkylamino- and alkoxycyclopropanes. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) ActiVation of UnreactiVe Bonds and Organic Synthesis; Murai, S., Ed.;
Springer: Berlin, 1999.
(2) (a) Janowicz, A. H.; Bergman, R. G. J. Am. Chem. Soc. 1983, 105, 3929.
(b) Periana, R. A.; Bergman, R. G. J. Am. Chem. Soc. 1986, 108, 7346.
(c) Burger, P.; Bergman, R. G. J. Am. Chem. Soc. 1993, 115, 10462. (d)
Tipper, C. F. H. J. Chem. Soc. 1955, 2045. (e) Phillips, R. K.; Puddephatt,
R. J. J. Chem. Soc., Dalton Trans. 1978, 1732. (f) Barrett, A. F. M.; Tam,
W. J. Org. Chem. 1997, 62, 7673. (g) Wick, D. D.; Northcutt, T. O.;
Lachiotte, R. J.; Jones, W. D. Organometallics 1998, 17, 4484. (h) Wender,
P. A.; Pedersen, T. M.; Scanio, M. J. C. J. Am. Chem. Soc. 2002, 124,
15154. (i) Itazaki, M.; Nishihara, Y.; Osakada, K. J. Org. Chem. 2002,
67, 6889 .(j) Bart, S. C.; Chirik, P. J. J. Am. Chem. Soc. 2003, 125, 886.
(3) (a) Bengali, A. A.; Arndtsen, B. A.; Burger, P. M.; Schultz, R. H.; Weiller,
B. H.; Kyle, K. R.; Moore, C. B.; Bergman, R. G. Pure Appl. Chem.
1995, 67, 281-288. (b) Klei, S. R.; Golden, J. T.; Burger, P.; Bergman,
R. G. J. Mol. Catal. 2002, 189, 79-94.
(4) The KIE was determined by competition experiments and analyzing the
ratio of deuterated to protiated products.
(5) The nature and pattern of this H/D scrambling process will be outlined in
a future publication.
(6) (a) Rodriguez, P.; Diaz-Requejo, M. M.; Belderrain, T. R.; Trofimenko,
S.; Nicasio, M. C.; Perez, P. J. Organometallics 2004, 23, 2162. (b) Bhalla,
G.; Oxgaard, J.; Goddard, W. A., III; Periana, R. A. Organometallics 2005,
24, 5499. (c) Oxgaard, J.; Bhalla, G.; Periana, R. A.; Goddard, W. A.,
III. Organometallics 2006, 25, 1618.
JA068312A
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 777