C-C coupling reactivity of an alkylgold(III) fluoride complex with arylboronic acids

Article abstract of DOI:10.1021/ja106257n

Previously, alkylgold(III) fluorides have been proposed as catalytic intermediates that undergo C-C coupling with reagents such as arylboronic acids in Au(I)/Au(III) cross-coupling reactions. Here is reported the first experimental evidence for this elementary mechanistic step. Complexes of the type (NHC)AuMe (NHC = N-heterocyclic carbene) were oxidized with XeF₂ to yield cis-(NHC)AuMeF₂ products, which were found to be in equilibrium with their fluoride-dissociated, dimeric [(NHC)AuMe(μ-F)] ₂[F]₂ forms. In one case, a monomeric cis-(NHC)AuMeF ₂ complex was favored exclusively in solution, and it was found to react with a variety of ArB(OH)₂ reagents to yield Ar-CH₃ products.

Full text of DOI:10.1021/ja106257n

Published on Web 08/20/2010
C-C Coupling Reactivity of an Alkylgold(III) Fluoride Complex with
Arylboronic Acids
Neal P. Mankad and F. Dean Toste*
Department of Chemistry, UniVersity of California, Berkeley, California 94720
Received July 14, 2010; E-mail: fdtoste@berkeley.edu
solution of a 2/3 mixture resulted in crystals of 3 as judged by
Abstract: Previously, alkylgold(III) fluorides have been proposed
as catalytic intermediates that undergo C-C coupling with
reagents such as arylboronic acids in Au(I)/Au(III) cross-coupling
reactions. Here is reported the first experimental evidence for this
elementary mechanistic step. Complexes of the type (NHC)AuMe
X-ray diffraction (Figure 2). Attempts to isolate 2 by fractional
crystallization led only to 3; apparently 3 crystallizes preferentially
due to its lower solubility and crystallographically imposed inver-
sion symmetry. The Au-F bond distance trans to the SIPr ligand
in 3 (2.034(3) Å) is slightly longer than that in (SIPr)AuF (2.021(17)
(
NHC ) N-heterocyclic carbene) were oxidized with XeF
cis-(NHC)AuMeF products, which were found to be in equilibrium
with their fluoride-dissociated, dimeric [(NHC)AuMe(µ-F)] [F]
forms. In one case, a monomeric cis-(NHC)AuMeF complex was
favored exclusively in solution, and it was found to react with a
variety of ArB(OH) reagents to yield Ar-CH products.
2
to yield
6
Å) and significantly shorter than the Au-F bond distance trans
2
to the methyl ligand in 3 (2.124(3) Å). The Au · · ·Au distance in 3
2
2
of 3.1789(6) Å is relatively short and could be indicative of
2
8
8
8
significant d -d interaction but also simply could be a result of
9
the small fluoride bridging ligands. Reversible formation of
2
3
2 2
Au (µ-F) cores such as that in 3 could contribute to catalyst stability
and therefore might account for the superior performance of
1
a,b
dinuclear Au compounds in cross-coupling catalysis.
Recently there has been an emerging literature of catalytic
reactions exploiting Au(I)/Au(III) redox cycles to achieve cross-
Scheme 1. Synthesis of Alkylgold(III) Fluoride Complexes
1,2
coupling reactivity, which complements the more established role
of cationic Au(I) and Au(III) catalysts as redox-neutral, carbophilic
3
electrophiles. Traditional cross-coupling methodologies, for ex-
ample with Pd(0)/Pd(II) cycles, involve the well-established
mechanistic series of C-X oxidative addition, transmetalation, and
4
C-C reductive elimination. On the other hand, gold(I) catalysts
5
tend to be unreactive toward C-X oxidative addition and instead
Measurement of Keq for the 2/3 equilibrium over the temperature
range 295-320 K allowed for approximation of the thermodynamic
parameters ∆H ) -13 kJ/mol and ∆S ) -56 J/mol ·K for the
dimerization process. We thus determined that a relatively minor
perturbation to this near-thermoneutral equilibrium could allow a
monomeric Au(III) difluoride species to be favored exclusively by
slight modification of the ancillary ligand. Accordingly, we found
that oxidation of (IPr)AuMe (4, IPr ) 1,3-bis(2,6-diisopropylphen-
require a sacrificial two-electron oxidant to access reactive gold(III)
1
intermediates poised for C-C coupling. Selectfluor, a source of
electrophilic fluorine, has met with great success as a common
1
a-e
sacrificial oxidant in these Au-catalyzed coupling reactions.
Thus, it has been speculated that alkylgold(III) fluoride species are
key intermediates despite the paucity of known gold fluoride
6
1a
complexes. Indeed, DFT studies have supported this mechanistic
proposal and indicated that alkylgold(III) fluoride intermediates
would be competent to undergo C-C coupling with arylboronic
yl)imidazol-2-ylidene) with XeF
2
in CDCl
3
yielded the spectro-
1
a-c,e
scopically pure monomer species cis-(IPr)AuMeF
2
(5), which
acid reagents often employed as coupling partners.
Here we
present the first experimental corroboration for this hypothesis by
reporting the synthesis of a methylgold(III) fluoride complex and
its reactivity with arylboronic acids to generate C-C coupled,
Ar-Me products.
Because the only known molecular gold-fluoride complex is
Sadighi’s (SIPr)AuF (SIPr ) 1,3-bis(2,6-diisopropylphenyl)imida-
6
,7
zolin-2-ylidene), as a starting point we sought to use SIPr as an
ancillary ligand to stabilize a potential alkylgold(III) fluoride target
2 3
molecule. Indeed, oxidation of (SIPr)AuMe (1) with XeF in CDCl
1
19
yielded evidence for Au(III)-F bonds. The H and F NMR spectra
of the product mixture indicated the presence of two species
assigned as cis-(SIPr)AuMeF
2
2 2
(2) and [(SIPr)AuMe(µ-F)] [F] (3),
which presumably are related by reversible fluoride dissociation
1
9
(
Scheme 1). Monomer 2 has characteristic peaks in the F NMR
2
2
spectrum at δ -201 (d, JFF ) 84.9 Hz) and δ -232 (dq, JFF
)
3
8
5.1 and JHF ) 8.6 Hz), consistent with a cis geometry, while the
bridging fluoride ligands in dimer 3 display a singlet resonance at
δ -253 (Figure 1a). Diffusion of pentane vapors into a CDCl
Figure 1. Au-F region of the 19F NMR spectra of (a) a mixture of 2 and
3, (b) 5, and (c) 9 in CDCl solution at room temperature.
3
3
10.1021/ja106257n  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 12859–12861 9 12859

C O M M U N I C A T I O N S
exhibits characteristic 19F NMR features at δ -200 (d, JFF ) 81.3
Hz) and δ -233 (dq, JFF ) 81.3 and JHF ) 8.6 Hz). Only 5 was
observed in typical NMR experiments (Figure 1b); no evidence
2
with no formation of toluene (Table 1, entry 8). Thus, the presence
of Au(III)-F bonds appears to be key to unveiling C-C coupling
reactivity, both by preventing unproductive C-X elimination and
by activating reactivity toward arylboronic acids.
2
3
for dimeric [(IPr)AuMe(µ-F)]
troscopy except at near-saturation concentrations. Nonetheless, slow
diffusion of pentane vapors into CDCl solutions of 5 yielded only
2 2
[F] (6) was found by NMR spec-
a
2
Table 1. C-C Coupling with Various ArB(OH)
3
X-ray quality crystals of 6. The solid-state structure of 6, which
also possesses crystallographically imposed inversion symmetry,
is very similar to that of 3 and has been placed in the Supporting
Information.
b
b
entry
Ar
yield (%)
entry
Ar
yield (%)
1
2
3
4
C
6
H
5
45
25
28
30
5
6
7
8
4-MeO
2
CC
6 4
6
H
4
68
51
49
4-MeOC
4-MeC
4-BrC
6
H
4
4-O
2
NC H
6
H
4
C
C
6
F
5
c
6
H
4
6
H
5
0
a
2
8 mM 5, 46 mM ArB(OH) H
, room temp. Based on in situ 1
b
1
NMR integration against an internal standard (1,3-dinitrobenzene).
c
2
Used (IPr)AuMeI instead of 5.
Recently, it has been proposed that C-C coupling between
alkylgold(III) fluoride complexes and arylboronic acids proceeds
not by a traditional transmetalation/reductive elimination sequence,
but rather by a concerted process wherein C-C and B-F bonds
1
a,b
form simultaneously (Scheme 2).
In light of this intriguing
3
proposal involving direct attack of the arylboronic acid on the C(sp )
center rather than on the Au center in alkylgold(III) fluorides, we
have begun to examine the mechanism of C-C coupling with the
isolable alkylgold(III) fluoride compounds reported here.
Figure 2. Solid-state structure of dicationic 3 (50% probability ellipsoids).
Hydrogen atoms and cocrystallized anions and solvent molecules have been
omitted for clarity. Selected bond lengths (Å) and angles (deg): Au(I)-C(1),
Scheme 2. Two Possible C-C Coupling Pathways
1
2
.968(4) Å; Au(1)-C(28), 2.029(4); Au(1)-F(1), 2.034(3); Au(1)-F(1#),
.124(3); Au(1)-Au(1#), 3.1789(6); Au(1)-F(1)-Au(1#), 99.69(12).
The ability to observe pure solutions of 5 is significant not only
because the complexes reported here are the first examples of
7
Au(III) fluorides but also because simple monoalkyl complexes
1
0,11
of Au(III) are very rare
despite the efforts of many researchers
to synthesize them. Bercaw and co-workers have observed closely
related complexes (IPr)AuMeX (X ) Cl, Br, and I), which at
12
2
10a
ambient conditions are unstable toward Me-X elimination. The
higher barrier for C-F elimination relative to other types of C-X
elimination is therefore critical to the stability of 5 and might
contribute to why fluorine-containing oxidants have proven to be
1
a-e,13
so successful in Au(I)/Au(III) catalysis.
First, we sought to examine the effect of electronics on the C-C
coupling reaction by conducting competition experiments in which
Complex 5 is closely related to complexes of the type LAu-
R)(F)(X) (X ) Cl or Br) that are proposed as the key intermediates
(
complex 5 was allowed to react with 1:1 mixtures of PhB(OH)
and p-RC B(OH) . Interestingly, there was relatively little
electronic bias in the resulting PhCH /p-RC CH ratios (1.0 and
0.72 for R ) Br and CO Me, respectively). This observation is
2
that undergo C-C coupling in various Au(I)/Au(III) catalytic
6
H
4
2
1
a-e
cycles.
Therefore, it was of fundamental interest to examine
3
6
H
4
3
the C-C coupling reactivity of 5 with weak nucleophiles such as
2
arylboronic acids that are often used in Au-mediated cross-coupling
possibly consistent with the proposed concerted bimolecular
1a-c,e
reactions.
PhB(OH)
formation of toluene, albeit in modest 45% yield (Table 1, entry
We were pleased to observe that addition of excess
elimination mechanism. Such a pathway is conceptually related to
0
2
to CDCl solutions of 5 at room temperature led to rapid
3
concerted σ-bond metathesis reactions of d metal-alkyl complexes
with arenes, whose rates also are relatively insensitive to the
1
4
15
17
1
). The Au-containing product in this reaction was (IPr)AuPh,
electronics of the arene substrates. By contrast, mechanisms in
which presumably formed from reaction between PhB(OH)
2
and
which either transmetalation or reductive elimination is rate limiting
might be expected to exhibit significant sensitivity to the electronics
of the transferring aryl group.
the initial elimination product (IPr)AuF, which was not observed
1
6
18
directly. The remainder of the mass balance was comprised of
trace amounts of biphenyl, benzene, and a significant amount of
insoluble material that was not characterized. It is possible that these
side reactions stem from arylboronic acid oxidation: accordingly,
Second, we sought to examine the effect of sterics on the C-C
3
coupling reaction by examining a tertiary rather than primary C(sp )
center. Oxidation of (IPr)Au(t-Bu) (7) with XeF
dimeric complex [(IPr)Au(t-Bu)(µ-F)] [F] (9), with no evidence
by NMR spectroscopy for significant population of the monomer
(IPr)Au(t-Bu)F (8) (Figure 1c). Though 9 is much less stable than
its less bulky relatives 3 and 6, it persists for several hours in CDCl
solution. When 9 was thus generated in situ and then added to
2
provided the
yields of Ar-CH
Table 1, entries 2-3) and slightly higher for electron-deficient
boronic acids (Table 1, entries 5-7). Following the procedure of
3
were lower for more electron-rich boronic acids
2
2
(
2
1
0a
Bercaw and co-workers,
we generated (IPr)AuMeI
2
in the
3
presence of PhB(OH) and observed its decay to (IPr)AuI and MeI
2
1
2860 J. AM. CHEM. SOC. 9 VOL. 132, NO. 37, 2010

C O M M U N I C A T I O N S
PhB(OH)
2
, no evidence for tert-butylbenzene formation was
Y.; Cui, L.; Zhang, L. Angew. Chem., Int. Ed. 2009, 48, 3112. (f) Iglesias,
A.; Mu n˜ iz, K. Chem.sEur. J. 2009, 15, 10563. (g) Kar, A.; Mangu, N.;
Kaiser, H. M.; Beller, M.; Tse, M. K. Chem. Commun. 2008, 3, 386.
(2) Garcia, P.; Malacria, M.; Aubert, C.; Gandon, V.; Fensterback, L.
ChemCatChem 2010, 2, 493.
apparent prior to the thermal decomposition of 9. Direct attack of
an arylboronic acid on the C(sp ) center of an alkylgold(III) fluoride
complex is expected to be attenuated at a tertiary C(sp ) center,
and thus the observed lack of t-Bu-Ar coupling is consistent with
the proposed concerted bimolecular elimination pathway. On the
other hand, an alkylgold(III) aryl intermediate (i.e., L(R)(Ar)Au X)
resulting from transmetalation might be expected to undergo
reductive elimination more rapidly with a bulkier R group.
In conclusion, we have reported the first isolable gold(III) fluoride
3
3
(
(
(
3) Recent reviews: (a) Shapiro, N. D.; Toste, F. D. Synlett 2010, 675. (b)
F u¨ rstner, A. Chem. Soc. ReV. 2009, 38, 3208.
4) Crabtree, R. H. The Organometallic Chemistry of the Transition Metals,
4
th ed.; Wiley-Interscience: NJ, 2005.
III
5) Lauterbach, T.; Livendahl, M.; Rosell o´ n, A.; Espinet, P.; Echaverren, A. M.
Org. Lett. 2010, 12, 3006.
1
8,19
(6) Laitar, D. S.; M u¨ ller, P.; Gray, T. G.; Sadighi, J. P. Organometallics 2005,
2
4, 4503.
(7) Though not isolated, the anions [Au(CF
NMR spectroscopy: Bernhardt, E.; Finze, M.; Willner, H. J. Fluorine Chem.
004, 125, 967.
(8) Mendizabal, F.; Pyykk o¨ , P. Phys. Chem. Chem. Phys. 2004, 6, 900.
3 n
) F
4-n]- have been observed by
III
complexes and noted the stabilizing influence of Au
2
(µ-F)
2
2
moieties. The alkylgold(III) difluoride complexes reported here are
closely related to speculative catalytic intermediates in Au(I)/Au(III)
cross-coupling reactions, and we have shown that they are indeed
competent at performing C-C coupling with arylboronic acids,
reactivity that appears to be unique to gold(III) fluorides among
the halide series. Further experiments aimed at elucidating the
intimate mechanism of C-C coupling are currently underway in
our laboratory.
III
(9) Based on a search of the Cambridge Structural Database, Au
2
(µ-X)
2
cores
with X ) O, OR, or NR
2
have Au · · · Au distances spanning 2.957-3.425
Å. Selected examples: (a) Adams, H.-N.; Grassle, U.; Hiller, W.; Strahle,
J. Z. Anorg. Allg. Chem. 1983, 504, 7. (b) Gabbiani, C.; Casini, A.; Messori,
L.; Guerri, A.; Cinellu, M. A.; Minghetti, G.; Corsini, M.; Rosani, C.;
Zanello, P.; Arca, M. Inorg. Chem. 2008, 47, 2368.
(10) (a) Scott, V. J.; Labinger, J. A.; Bercaw, J. E. Organometallics, accepted.
b) Bennett, M. A.; Hoskins, K.; Kneen, W. R.; Nyholm, R. S.; Hitchcock,
(
P. B.; Mason, R.; Robertson, G. B.; Towl, A. D. C. J. Am. Chem. Soc.
1971, 93, 4591. (c) Vincente, J.; Chicote, M. T.; Lozano, M. I.; Huertas,
S. Organometallics 1999, 18, 753. (d) Cinellu, M. A.; Cocco, F.; Minghetti,
G.; Stoccoro, S.; Zucca, A.; Manassero, M. J. Organomet. Chem. 2009,
Acknowledgment. Funding was provided by the Director,
Office of Science of the U.S. Department of Energy under Contract
No. DE-AC02-05CH11231 and NIHGMS (R01 GM073932).
N.P.M. is grateful for an NIH Kirchstein-NRSA postdoctoral
fellowship (1F32 GM93654-01). Dr. Antonio DiPasquale assisted
with crystallography. We thank Prof. John Bercaw (Caltech) and
Valerie Scott for discussing unpublished results.
6
94, 2949. (e) Sanner, R. D.; Satcher, J. H.; Droege, M. W. Organometallics
1989, 8, 1498.
(
11) Monoaryl complexes of Au(III) are more well known. Selected references: (a)
Vicente, J.; Arcas, A.; Mora, M.; Solans, X.; Font-Altaba, M. J. Organomet.
Chem. 1986, 309, 369. (b) Uson, R.; Laguna, A.; Bergareche, B. J.
Organomet. Chem. 1980, 184, 411. (c) Vaughan, L. G.; Sheppard, W. A.
J. Am. Chem. Soc. 1969, 91, 6151.
(
12) For example: Komiya, S.; Huffman, J. C.; Kochi, J. K. Inorg. Chem. 1977,
16, 1253.
(
13) Use of an NHC ligand apparently is also a key factor to stabilizing 5, as
3 2 2 6
we found that oxidation of Ph PAuMe with XeF yielded C H and
Note Added after ASAP Publication. This communication was
published on the Web August 20, 2010 without a Supporting Informa-
tion file. The correct version posted August 25, 2010 is correct.
+
[
3 2
(Ph P) Au] . Similar reactivity has been observed previously in phosphine-
containing organogold systems upon fluorination: (a) Hashmi, A. S. K.;
Ramamurthi, T. D.; Rominger, F. J. Organomet. Chem. 2009, 694, 592.
(b) ref 1e.
(
14) Similar results were obtained by adding PhB(OH)
2
to a mixture of 2 and
Supporting Information Available: Synthetic and crystallographic
details. This material is available free of charge via the Internet at http://
pubs.acs.org.
3.
(15) Hashmi, A. S. K.; Lothsch u¨ tz, C.; D o¨ pp, R.; Rudolph, M.; Ramamurthi,
T. D.; Rominger, F. Angew. Chem., Int. Ed. 2009, 48, 8243.
(
16) We note that reaction of PhB(OH)
2
and isolated (SIPr)AuF rapidly produced
(
SIPr)AuPh.
References
(17) Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.; Nolan, M. C.;
Santarsiero, B. D.; Schaefer, W. P.; Bercaw, J. E. J. Am. Chem. Soc. 1987,
109, 203.
(
1) (a) Brenzovich, W. E., Jr.; Benitez, D.; Lackner, A. D.; Shunatona, H. P.;
Tkatchouk, E.; Goddard, W. A., III; Toste, F. D. Angew. Chem., Int. Ed.
(18) Hartwig, J. F. Inorg. Chem. 2007, 46, 1936.
2010, 49, 5519. (b) Melhado, A. D.; Brenzovich, W. E., Jr.; Lackner, A. D.;
(19) One must be cautious in drawing conclusions from this experiment,
however, as it is possible that more facile access to monomer 9 is necessary
for C-C coupling to proceed by any mechanism.
Toste, F. D. J. Am. Chem. Soc. 2010, 132, 8885. (c) Zhang, G.; Cui, L.;
Wang, Y.; Zhang, L. J. Am. Chem. Soc. 2010, 132, 1474. (d) Hopkinson,
M. N.; Tessier, A.; Salisbury, A.; Giuffredi, G. T.; Combettes, L. E.; Gee,
A. D.; Gouverneur, V. Chem.sEur. J. 2010, 16, 4739. (e) Zhang, G.; Peng,
JA106257N
J. AM. CHEM. SOC. 9 VOL. 132, NO. 37, 2010 12861