Intramolecular alkyne hydroalkoxylation and hydroamination catalyzed by iridium hydrides

Article abstract of DOI:10.1021/ol052186i

(Chemical Equation Presented) Iridium(III) hydrides prove to be air-stable active catalysts for intramolecular hydroalkoxylation and hydroamination of internal alkynes with proximate nucleophiles. The cyclization follows highly selective 6-endo-dig regiochemistry when regioselectivity is an issue.

Full text of DOI:10.1021/ol052186i

ORGANIC
LETTERS
2
005
Vol. 7, No. 24
437-5440
Intramolecular Alkyne Hydroalkoxylation
and Hydroamination Catalyzed by
Iridium Hydrides
5
Xingwei Li, Anthony R. Chianese, Tiffany Vogel, and Robert H. Crabtree*
Department of Chemistry, Yale UniVersity, 225 Prospect Street,
New HaVen, Connecticut 06520-8107
robert.crabtree@yale.edu
Received September 11, 2005
ABSTRACT
Iridium(III) hydrides prove to be air-stable active catalysts for intramolecular hydroalkoxylation and hydroamination of internal alkynes with
proximate nucleophiles. The cyclization follows highly selective 6-endo-dig regiochemistry when regioselectivity is an issue.
intramolecular examples are particularly useful in giving
heterocycles. Metal-catalyzed hydroamination of alkynes has
wide synthetic applications.
1
15
Iridium-mediated activation of alkynes toward nucleophilic
16
attack is uncommon, especially for weak nucleophiles such
3,17
as esters or nitro compounds.
We recently reported a
stoichiometric intramolecular nitro oxygen transfer to the
CtC bond of o-RCtCC NO upon coordination to
H
6 4
2
cationic iridium hydrides and proposed electrophilic activa-
6
tion of the alkynes on binding to Ir(III). Strong product
binding inhibited catalysis in this case. Moving to nucleo-
philes stronger than the nitro group, we now report catalytic
intramolecular hydroamination and hydroalkoxylation of
Chem. Soc. 2004, 126, 16520.
(
(
(
4) Wang, Z.; Lu, X. Tetrahedron Lett. 1997, 38, 5213.
5) Asao, N.; Sato, K.; Yamamoto, Y. Tetrahedron Lett. 2003, 44, 5675.
6) Li, X.; Incarvito, C. D.; Vogel, T.; Crabtree, R. H. Organometallics
(12) (a) Sherry, B. D.; Toste, F. D. J. Am. Chem. Soc. 2004, 126, 15978.
(b) Maeyama, K.; Iwasawa, N. J. Am. Chem. Soc. 1998, 120, 1928.
(13) (a) Asao, N.; Nogami, T.; Takahashi, K.; Yamamoto, Y. J. Am.
Chem. Soc. 2002, 124, 764. (b) Patil, N. T.; Yamamoto, Y. J. Org. Chem.
2004, 69, 5139.
(14) (a) Lewis Acids in Organic Synthesis; Yamamoto, H., Ed.; Wiley-
VCH: Weinheim, 2000; Vols. 1 and 2. (b) Asao, N.; Aikawa, H.;
Yamamoto, Y. J. Am. Chem. Soc. 2004, 126, 7458. (c) Yue, D.; C a` , N. D.;
Larock, R. C. Org. Lett. 2004, 6, 1581. (d) Huang, Q.; Hunter, J. A.; Larock,
R. C. J. Org. Chem. 2002, 67, 3437. (e) Barluenga, J.; V a´ zquez-villa, H.;
Ballesteros, A.; Gonz a´ lez, J. M. Org. Lett. 2003, 5, 4121. (f) Barluenga, J.;
V a´ zquez-villa, H.; Ballesteros, A.; Gonz a´ lez, J. M. J. Am. Chem. Soc. 2003,
125, 9028.
2
005, 24, 3066.
7) (a) Dai, G.; Larock, R. C. Org. Lett. 2001, 3, 4035. (b) Roesch, K.
R.; Larock, R. C. J. Org. Chem. 2002, 67, 86.
8) (a) Hua, R.; Tian, X. J. Org. Chem. 2004, 69, 5782. (b) Sashida, H.;
Kawamukai, A. Synthesis 1999, 1145.
9) (a) Nan, Y.; Miao, H.; Yang, Z. Org. Lett. 2000, 2, 297. (b) Hu, Y.;
Yang, Z. Org. Lett. 2001, 3, 1387.
10) (a) Asao, N.; Aikawa, H.; Yamamoto, Y. J. Am. Chem. Soc. 2004,
26, 7458. (b) Asao, N.; Takahashi, K.; Lee, S.; Kasahara, T.; Yamamoto,
(
(
(
(
1
Y. J. Am. Chem. Soc. 2002, 124, 12650. (c) Iwasawa, N.; Shido, M.;
Kusama, H. J. Am. Chem. Soc. 2001, 123, 5814. (d) Jiang, D.; Herndon, J.
W. Org. Lett. 2000, 2, 1267. (e) Kusama, H.; Funami, H.; Takaya, J.;
Iwasawa, N. Org. Lett. 2004, 6, 605.
(15) (a) Beller, M.; Seayad, J.; Tillack, A.; Jiao, H. Angew. Chem., Int.
Ed. 2004, 43, 3368. (b) M u¨ ller, T. E.; Beller, M. Chem. ReV. 1998, 98,
675. (c) Alonso, F.; Beletskaya, I. P.; Yus, M. Chem. ReV. 2004, 104, 3079.
(16) Genin, E.; Antoniotti, S.; Michelet, V.; Gen eˆ t, J.-P. Angew. Chem.,
Int. Ed. 2005, 44, 4949.
(
11) Kennedy-Smith, J. J.; Staben, S. T.; Toste, F. D. J. Am. Chem. Soc.
2
004, 126, 4526.
1
0.1021/ol052186i CCC: $30.25
© 2005 American Chemical Society
Published on Web 10/27/2005

internal alkynes using closely related hydridoiridium cata-
lysts. The reactions even proceed in undistilled solvents,
without exclusion of air and water. The catalytic process is
limited to internal alkynes, as terminal alkynes rearrange to
vinylidenes, which are rapidly trapped by pendant nucleo-
philes to afford highly stable Fischer carbene complexes
stoichiometrically.18
Table 1. Cyclization of Ortho-Substituted Internal Alkynesa
The iridium hydride catalysts (1-5, Figure 1) are obtained
1
9
by a previously described cycloiridation reaction. The
Figure 1. Hydridoiridium complexes screened for cyclization of
6
.
complexes are stable to air and moisture in both the solid
form and in solution. These complexes were initially studied
for the cyclization of 6 (eq 1). Each complex is an active
catalyst for selective conversion of the starting alcohol to
the isochromene 7 by 6-endo-dig cyclization, and screening
revealed a reactivity order: 1 > 2 > 3 g 4 > 5.
[
IrH
tion, but the reaction is not as clean (80 °C, full conversion,
0% NMR yield). We therefore chose 1 as a catalyst for
2 2 3 2 6
(acetone) (PPh ) ]SbF also catalyzes the same cycliza-
8
further studies.
6 4 2
Various 2-alkynylbenzyl alcohols o-RCtC(C H )CH OH
undergo regioselective cyclization with 4 mol % loading of
n
a
as the catalyst. b CHCl3. c CH2Cl2. d PhMe. e Isolated yields.
1
1
(Table 1). For 6 (R ) Pr), the starting material was fully
converted to the corresponding isochromene 7 in high yield
(
94% NMR; 77% isolated), slightly better than that reported
-KI as catalyst.2 For 8 (R ) Ph), the starting
a
isochromene 9 was observed in only 70% NMR yield,
together with unidentified decomposition products. Cycliza-
tion of 8 has been reported under basic conditions using
with PdI
2
material was also fully converted, but the corresponding
20
20
21
2a
(
17) (a) Field, L. D.; Messerle, B. A.; Wren, S. L. Organometallics 2003,
2, 4393. (b) Burling, S.; Field, L. D.; Messerle, B. A.; Turner, P.
Organometallics 2004, 23, 1714.
18) Li, X.; Vogel, T.; Incarvito, T.; Crabtree, R. H. Organometallics
005, 24, 62.
19) Li, X.; Chen, P.; Faller, J. W.; Crabtree, R. H. Organometallics
005, 24, 4810.
NaOH, KH, and TBAF or a Pd catalyst, but with only
2
20,21
2a
5
-exo-dig
or poor regioselectivity. In the present work,
(
2
2
(20) Weingarten, M. D.; Padwa, A. Tetrahedron Lett. 1995, 36, 4717.
(21) Hiroya, K.; Jouka, R.; Kameda, M.; Yasuhara, A.; Sakamoto, T.
Tetrahedron 2001, 57, 9697.
(
5438
Org. Lett., Vol. 7, No. 24, 2005

we consistently obtain 6-endo regiochemistry for both R )
alkyl and aryl groups. Cyclization of 10, having two OH
groups, gives the two regioisomeric spiroacetals 11 and 12
in a ratio of 11:1, presumably by nucleophilic addition of
of Ir-D was 94%. The cyclization of 23 catalyzed by 10 mol
% 1-d (with 94% Ir-D) in CD Cl was monitored by H
8 2 2
1
NMR spectroscopy at -10 °C. All resonances for 24
maintain the same ((4%) integration throughout, indicating
2
6
the pendant -CH
2 2 2
CH CH OH group to an isochromene or
essentially no deuterium incorporation. Meanwhile, at no
point was there any change ((4%) in integration observed
for the residual Ir-H signal and for the proton residue of the
acetophenone backbone. Even assuming an error of 10%
given the weak signal for residual Ir-H, the maximum
hydrogen incorporation into the iridium center can therefore
be reliably estimated to be <1% (10% error × 6% Ir-H)
with respect to total catalyst (Ir-D + Ir-H) and <0.1% with
respect to substrate. If a mechanism involving initial insertion
of the alkyne into Ir-H were operative, significant deuterium
incorporation at the 3-position of 24 would be expected at
low conversion, coupled with replacement of Ir-D by Ir-H,
which would be the highest at full conversion. As this was
not observed, a mechanism involving initial insertion can
be ruled out. Similar results were observed for substrate 27
benzofuran intermediate. Analogously, 12 cyclizes to spiroac-
2
2
etal 13 much more efficiently than in a previous route.
Alcohol 30 failed to cyclize (CDCl , 85 °C, 12 h), however.
Similarly, 2-alkynylphenols [RCtC(C )OH] undergo
3
6 4
H
cyclization to give benzofurans, where no regiochemical
complication applies. Compound 15 cyclizes cleanly (91%
NMR yield) at room temperature with 3% loading of 1. The
cyclization of 17 is more difficult, but catalyst 1 efficiently
gives the benzofuran 18 in 72% yield in refluxing toluene.
Previously reported methods for the cyclization of 2-alky-
23a,b
23c
nylphenols include strong bases,
Lewis aicds,
or
palladium catalysis.9
Although the cyclization of 2-alkynylbenzoic acids o-RCt
C(C )COOH can be catalyzed by strong acids, palladium,
6 4
H
(-35 °C). Given this result, the most straightforward
or silver compounds, the previously obtained products are
mechanism involves electrophilic activation of the alkyne
toward nucleophilic attack by binding to the Ir(III) center,
followed by direct selective protonolysis of the Ir-C (indole)
bond (Scheme 1).
often mixtures from both 6-endo-dig and 5-exo-dig path-
ways.8b,24
n
Here, for both R ) Pr (19) and Ph (21), we
observed only the isocoumarins 20 and 22, respectively, from
selective 6-endo-dig cyclization, in high isolated yields.
The cyclization (intramolecular hydroamination) of o-
1,25
alkynylanilines often involves palladium catalysts. A base-
Scheme 1. Proposed Mechanism for the Cyclization of 23
Catalyzed by 1
t
promoted cyclization using BuOK or KH has also been
reported.2 Hydride 1 is a highly active catalyst to convert
these substrates nearly quantitatively to indoles under mild
conditions (rt or 35 °C). The aniline substrates 23 and 25
both give nearly quantitative yield of the corresponding
indoles, and cyclization of the diarylamine 27 to the indole
3a,b
28 was achieved with even lower catalyst loading (0.5 mol
%). Surprisingly, the cyclization of 29 failed, even with 8%
loading of 1 at 85 °C.
The rarity both of iridium complexes and of metal hydrides
as hydroamination catalysts led us to make some preliminary
16
mechanistic observations. A mechanism involving ami-
noiridation of the alkyne seems unlikely, as only 4-exo
cyclization is feasible, and the observed 5-endo regioselec-
tivity would require subsequent rearrangements. To deter-
mine if the Ir-H was directly involved, we prepared 1-d
containing Ir-D, by reaction (acetone, 65 °C, 5 h) of
acetophenone-d with [IrH (acetone) (PPh ]SbF and an
8
,
8
2
2
3
)
2
6
excess (20 equiv) of tert-butylethylene as a hydrogen
1
acceptor. By H NMR spectroscopy the deuterium content
(
22) Fugami, K.; Hagiwara, N.; Okeda, T.; Masanori, K. Chem. Lett.
998, 81.
23) (a) Koradin, C.; Dohle, W.; Rodriguez, A. L.; Schmid, B.; Knochel,
1
(
P. Tetrahedron 2003, 59, 1571. (b) Rodriguez, A. L.; Koradin, C.; Dohle,
W.; Knochel, P. Angew. Chem., Int. Ed. 2000, 39, 2488. (c) Arcadi, A.;
Cacchi, S.; Fabrizi, G.; Marinelli, F.; Moro, L. Synlett 1999, 1432.
(
24) (a) Bellina, F.; Ciucci, D.; Vergamini, P.; Rossi, R. Tetrahedron
2
5
000, 56, 2533. (b) Mukhopadhyay, R.; Kundu, N. G. Tetrahedron 2001,
7, 9475.
The cyclization reactions of 23 and 27, observed at low
temperature with 10% catalyst loading, showed surprisingly
(25) (a) Patil, N. T.; Pahadi, N. K.; Yamamoto, Y. Tetrahedron Lett.
2
005, 46, 2101. (b) M u¨ ller, T. E.; Berger, M.; Grosche, M.; Herdtweck,
E.; Schmidtchen, F. P. Organometallics 2001, 20, 4384. (c) Iritani, K.;
Matsubara, S.; Utimoto, K. Tetrahedron Lett. 1988, 29, 1799. (d) Rudisill,
D. E.; Stille, J. K. J. Org. Chem. 1989, 54, 5856. (e) Yu, M. S.; de Leon,
L. L.; McGuire, M. A.; Botha, G. Tetrahedron Lett. 1998, 39, 9347.
(26) At early reaction times, big [Ir-D]/[24] ratios makes deuterium
incorporation into 24 easily seen by H NMR spectroscopy given appropriate
precautions. See: Burk, M. J.; Crabtree, R. H. J. Am. Chem. Soc. 1987,
109, 8025.
1
Org. Lett., Vol. 7, No. 24, 2005
5439

different behavior. For 23, the catalyst resting state is
1
13
assigned by H and C NMR as the N-bound substrate
complex 31 (Scheme 1). In a kinetic experiment, zero-order
consumption of substrate was observed for the majority of
the reaction (Figure 2). This is consistent27 with reversible
Figure 3. Kinetic data for the competitiVe cyclization of 23 (open
symbols) and 27 (closed symbols), catalyzed by 10 mol % of 1.
For 23, the linear portion of the graph (open circles) was used for
linear regression analysis. For 27, the non-zero portion of the graph
(closed circles) was used for curve-fitting.
which strongly supports fast reversible dissociation of 23
from 31 on the time scale of the cyclization reaction. The
consumption of 27 fits a curve defined by eq 2, where A is
derived from the linear fit of the consumption of 23.27
Interestingly, the exponent in eq 2, k23/k27, corresponds to
the ratio of intrinsic turnover frequencies for the consumption
of 23 vs 27, factoring out the unproductive binding by 23.27
The observed value of ca. 4.0 for k23/k27 indicates that 27 is
consumed four times faster than 23. This is inconsistent with
a simple rate-limiting nucleophilic attack of NHR on CtC,
as 27 is less nucleophilic, but may be explained by the
increased NH acidity of 27 (as in a concerted nucleophilic
attack/proton-transfer pathway).
Figure 2. Kinetic data for the cyclization of 23 (open symbols)
and 27 (closed symbols), catalyzed by 10 mol % 1. For 23, the
linear portion of the graph (open circles) was used for linear
regression analysis.
dissociation of 23 from 31 to generate the reactive five-
coordinate intermediate 32. Catalysis may then proceed by
re-coordination of 23 to 32 through the CtC bond to give
33, nucleophilic attack on the alkyne moiety to generate 34,
and direct protonolysis of the Ir-C bond to complete the
cycle (Scheme 1). Coordination of 23 through nitrogen thus
leads to unproductive substrate binding2
5b,28
inhibiting cata-
lytic turnover. Although this kinetic result is also consistent
with saturation behavior, we consider a direct reaction of
the amine complex to be unlikely (see above).
For the cyclization of 27, the observed catalyst resting state
is now the acetone complex 1, and the kinetics is first order
with respect to the substrate (Figure 2). We propose a
mechanism analogous to that for 23, except that here the
secondary amine 27 is a weaker N-donor and it does not
inhibit the catalytic reaction by binding to iridium. The first-
order kinetic behavior is consistent27 with rate-determining
reaction of 27 with a steady-state concentration of the reactive
iridium complex 32, generated by reversible dissociation of
acetone from the resting complex 1.
k27/k23
[
27] ) [27] (1 - At)
(2)
init
In summary, we report a novel class of hydridoiridium-
III) catalysts for the intramolecular hydroalkoxylation and
(
hydroamination of ortho-substituted aryl alkynes, using a
variety of tethered nucleophiles. When regioselectivity is an
issue, highly selective 6-endo-dig cyclization is observed.
Mechanistic experiments indicate that the reaction likely
proceeds via electrophilic activation of the alkyne toward
nucleophilic attack, followed by direct protonolysis of the
resulting Ir-C bond. Moderately basic amines are shown to
bind to iridium though nitrogen, providing a well-character-
ized example of unproductive substrate binding in organo-
metallic catalysis.
To provide support for the hypothesis of unproductive
substrate binding in the cyclization of 23, a competition
experiment was performed with 23 and 27 both present. The
kinetic plots are shown in Figure 3. As expected, the
cyclization of 23 was unaffected by the presence of 27. The
Acknowledgment. We thank the U.S. DOE and the
Johnson Matthey Co. for support and Leah Appelhans for
assistance.
2
7
cyclization of 27, however, behaved consistently with a
model of decreasing inhibition as substrate 23 is consumed,
Supporting Information Available: Syntheses, deriva-
tion of eq 2, NMR spectra, and kinetics data. This material
is available free of charge via the Internet at http://pubs.acs.org.
(
27) See the Supporting Information.
(28) Pirrung, M. C.; Liu, H.; Morehead, A. T. J. Am. Chem. Soc. 2002,
1
24, 1014.
OL052186I
5440
Org. Lett., Vol. 7, No. 24, 2005