Single bifunctional ruthenium catalyst for one-pot cyclization and hydration giving functionalized indoles and benzofurans

Article abstract of DOI:10.1002/chem.201001075

Chemical equation Presented Bifunctional is more than twice as fun! At low loading, catalyst 1 (see scheme) can form two important heterocycle classes, apparently by attack of XH on a vinylidene intermediate. Aza- and nitroindoles can be formed, and all N-protecting groups tested (alkyl, allyl, sulfonyl) were tolerated. The newly formed ring can be deuterated in one step, and for substrates with two terminal alkynes, cyclization can be followed by hydration, making this catalyst uniquely versatile.

Full text of DOI:10.1002/chem.201001075

DOI: 10.1002/chem.201001075
Single Bifunctional Ruthenium Catalyst for One-Pot Cyclization and
Hydration giving Functionalized Indoles and Benzofurans
Reji N. Nair,[a] Paul J. Lee,[a] Arnold L. Rheingold,[b] and Douglas B. Grotjahn*[a]
Indole[1] and benzofuran[1d,2] heterocycles are key structur-
al units in a variety of biologically active natural products
and unnatural synthetic materials. Over the last few years,
transition-metal catalysts have been extensively used in the
cyclization of o-alkynylarylamines[3a–l] and phenols[3m–p] to
construct benzoheterocycles through intramolecular[3] and
intermolecular reactions.[4] Many of these reactions required
the use of a nitrogen protecting group in the case of indo-
les[3g,i,k] or stoichiometric amounts of the metal.[3l] McLory
and Trost[3d] synthesized both unprotected indoles or N-
benzyl analogues by using rhodium–phosphine complexes,
Scheme 1. The use of bifunctional catalyst 1 for hydration and cycliza-
but to optimize the benzofuran synthesis, large amounts of
phosphine (0.6 equiv) were needed. Saa and co-workers[3q]
recently reported the use of 10 mol% catalyst and amine
solvents in benzofuran synthesis at 908C. Herein, we report
that the bifunctional ruthenium catalyst 1 which has been
used for anti-Markovnikov alkyne hydration[5] emerges as a
versatile choice for both indole and benzofuran formation,
with a number of unique advantages, including the use of
only 2 mol% catalyst in most cases and the unprecedented
ability of a cycloisomerization catalyst to perform hydration
or deuteration in the same reaction.
We envisioned that the use of alkyne hydration catalyst 1
on substrates, such as 2, (Scheme 1) would generate the al-
dehyde 3, on which cyclization and dehydration would then
enable facile synthesis of heterocycles. Indeed, 2a cyclized
to hemiaminal 4a,[5a] with some equilibrium amount of 3a,
tion; Cp=cyclopentadienyl.
but elimination of water from 4a required heating in second
step, forming 5a in 46% overall yield. The same two-step
process could be used on the homologous substrate 2b, to
form 5b in 56% overall yield.
Encouraged by these results, and because of the impor-
tance of indoles, we shifted our attention to substrates de-
rived from o-ethynylaniline. The first four entries of Table 1
demonstrate that 1 cyclizes a wide range of aniline deriva-
tives; most significantly, the simplest, unprotected parent
compound 6 cyclized to indole 17 in 99% yield (Table 1,
entry 1). Sulfonamide 7 (Table 1, entry 2) was cyclized to
give the N-tosyl (Ts) protected indole in just 2 h. Entries 3
and 4 in Table 1 show the ability of the catalyst to form ni-
trogen analogues with different alkyl substituents. Especially
notable is the selectivity of 1: the N-allyl group (Table 1,
entry 3) is fully tolerated without the decomposition by iso-
merization seen when using a Rh-based catalyst.[3d]
The fact that a wide variety of nucleophilic nitrogen cen-
ters could be used in indole formation encouraged us to ex-
plore the scope of 1 not only in making more complex
indole derivatives, but also in cyclizing phenol derivatives to
benzofurans. Gratifyingly, 1 forms both 7-aza- and 6-nitroin-
doles (Table 1, entries 5 and 6), showing tolerance of various
ring substituents, including those that might coordinate to
the catalyst. Furthermore, entry 9 (Table 1) shows the ability
of 1 to form the benzofuran nucleus in essentially quantita-
[a] R. N. Nair, P. J. Lee, Prof. D. B. Grotjahn
Department of Chemistry and Biochemistry
San Diego State University
5500 Campanile Drive, San Diego, CA 92182-1030 (USA)
Fax : (+1)619-594-4634
E-mail: grotjahn@chemistry.sdsu.edu
[b] Prof. A. L. Rheingold
Department of Chemistry and Biochemistry
University of California–San Diego
La Jolla, CA 92093-0358 (USA)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201001075.
7992
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 7992 – 7995

COMMUNICATION
Table 1. Results of heterocycle formation.[a]
Our ongoing mechanistic studies of bifunctional alkyne
hydration[5b] catalysts provide direct evidence for the inter-
mediacy of vinylidenes, such as B (Scheme 2). Here, indirect
Substrate
Product
Method[b]
t
Yield[c]
[h] [%]
A
B
A
B
A
A
17 99
1
2
6, R1 =H
7, R1 =Ts
17, R1 =H
18, R1 =Ts
7
2
7
87
86
76
3
4
8, R1 =allyl
9,
19, R1 =allyl
20,
10 74
20 82
R1 =CH (C6H4)CH3 R1 =CH
2ACHTUNGTRENNUNG 2CAHTUNTGRNE(NUGN C6H4)CH3
5
6
B
7
94
A
B
2
2
91
91
(90)
Scheme 2. Probable mechanism and evidence against protic catalysis
(X=NH, O); conditions: a) CF3SO3H (5 mol%), 708C, 40 h.
evidence for the intermediacy of vinylidenes comes from the
inertness of an internal alkyne (Table 1, entry 7) and the
sluggish reaction of a silylated alkyne (Table 1, entry 8), in
which the latter compound may suffer slow protiodesilyla-
tion to 6 followed by rapid transformation to 17. These and
all of the other cyclizations discussed below were conducted
in NMR spectroscopy tubes to gain the maximum informa-
tion about the course of the reaction. None of the possible
hydration products of general form E (Scheme 2) were seen
(estimated detection limit usually 1%). This could be a
result of initial formation of E and its more rapid cyclization
and water elimination to give the final product D or because
of direct cyclization of vinylidene B. Evidence for the direct
cyclization of B includes high yield cyclizations under anhy-
drous conditions (entries B in Table 1). Moreover, successful
reactions of the nitro analogue 11 (entries 6A and 6B) sug-
gest that alkyne hydration via E is not a significant pathway,
7
NR[d]
A
40 NR
400 84
8
9
17
A
A
1
1
99
99
(92)
10[e]
B[f]
98
(75)
11
B[f]
1
[a] See Supporting Information for substrate preparation, yield determi-
nations, and product characterization. [b] Method A: with 5 equiv of
H2O, in [D6]acetone or [D8]THF. Method B: no water added, in dry
because
a previously attempted hydration of 4-nitro-
(ethynyl)benzene by using 1[5a] led to the formation of less
than 1% aldehyde and complete catalyst inactivation by
pathways known for several other reactions of metals, al-
kynes, and water.[6] Control experiments[7] ruling out protic
catalysis by using TfOH (5 mol%, 708C, 40 h) on 6 or 14
showed that the aniline formed known[8a,b] quinoline deriva-
tive Y, whereas the phenol underwent Markovnikov hydra-
tion to give ketone Z.[8c]
[D8]THF. Reactions were done on
a 0.05–0.15 mmol scale with 1
(2 mol%) at 708C, unless otherwise noted. [c] Yield calculated from
NMR spectroscopy integrations against an internal standard. In parenthe-
sis are the yields for the respective isolated products. [d] NR=no reac-
tion. [e] BR2 =9-borabicyclononane. [f] Reactions done in acetone or
[D6]acetone.
tive yield with the same low catalyst loading as in entries 1–
6, and with only 1 h reaction time. The mildness of the cycli-
zation conditions and the tolerance of diverse functional
groups in benzofuran formation is highlighted by entries 10
and 11 (Table 1), which show that low catalytic loads are re-
quired to convert boryl-protected o-ethynyltyrosine (15) to
its benzofuran derivative 24 and o-ethynylated vanillin (16)
to 25. Thus, low catalyst loadings of 1 cyclize both ethynylat-
ed nitrogen and oxygen analogues to the corresponding
indole and benzofuran derivatives in high yields.
To further highlight the unique synthetic potential of cata-
lyst 1, reactions on doubly ethynylated substrates 26, 27, and
28 were performed (Table 2). Of note, except for entry 2, all
of the yields in Table 2 refer to isolated and purified prod-
ucts. The case of the nitrogen analogue 26 provides good
mechanistic insight: indole formation (29) proceeded
smoothly (Table 2, entry 1), but hydration of the alkynyl
indole to 30, (Table 2, entry 2) was surprisingly slow and
could not be driven to completion. Gratifyingly, the phenolic
analogue 28 cyclized to alkyne derivative 33 under anhy-
Chem. Eur. J. 2010, 16, 7992 – 7995
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7993

D. B. Grotjahn et al.
Table 2. One-pot cyclization and hydration.[a]
Substrate Product
Method[b] t [h] Yield[d] [%]
B
A
B
20
22
9
60[e]
11[e]
67
Scheme 3. Synthesis of polydeuterated indole and benzofuran. [a] For 14:
5 equiv, [D6]acetone, 1 h; for 6: 28 equiv, [D8]THF, 68 h); [b] Maximum
deuteration theoretically possible 95% (see Supporting Information for
details); [c] Maximum deuteration theoretically possible 80% (see Sup-
porting Information for details).
1
of vinylidene intermediates, such as B,[12] here allowing for
facile deuterium incorporation and cyclization in one pot.
Given the recent interest in deuteration of organic mole-
cules[13] and the attractiveness of using inexpensive and safe-
to-handle D2O as a source of deuterium, these results sug-
gest additional roles for 1 and related species[14] in the selec-
tive labeling of organic molecules.
3
4
A
24
95[f]
To summarize, compound 1 is a unique catalyst of broad
application because 1) cyclic enamides, indoles, and benzo-
furans can be made; 2) similar low catalyst loadings are
used in each case; 3) several classes of substituents are toler-
ated on either the benzo ring or on the nitrogen atom in-
volved in the cyclization; 4) one-pot reactions of doubly
ethynylated species can lead to significant increases in mo-
lecular complexity impossible with other cycloisomerization
catalysts incapable of alkyne hydration; 5) both carbons of
the newly formed heterocycle can be deuterated simply by
adding D2O. Available evidence rules out protic catalysis,
but is consistent with indole and benzofuran formation from
direct attack of the heteroatom, rather than water, on a vi-
nylidene intermediate, which suggests additional applica-
tions for bifunctional catalyst 1 and its analogues in organic
synthesis.
5
6
B
10
8
64
74
A
[a] See Supporting Information for substrate preparation, yield determi-
nations, and product characterization. [b] Method A: with 5 equiv of
H2O, in [D6]acetone or [D8]THF. Method B: no water added, in dry
[D8]THF. Reactions were done on 0.2–0.5 mmol scale with 1 (4 mol%) at
708C, unless otherwise noted. [c] Isolated and purified product. [d] Based
on recovered starting material. [e] NMR spectroscopy yield. [f] 20 mol%
catalyst used.
drous conditions and also could be hydrated to 34 in the
presence of water. It may be noted that the newly formed
-CH2CHO substituent in 34 could be elaborated into a vari-
ety of other functionalized side chains.
Because hydration of the sterically similar alkynyl benzo-
furan 33 to 34 was so much more facile, our hypothesis was
that the indole NH must play a role; in putative intermedi-
ate F (Scheme 2), hydrogen bonding of the pyridinium
moiety[5b] to the acyl could be precluded by competitive in-
teraction with the indole NH. Consistent with this, not only
did the N-methylated substrate 27 (Table 2, entry 3) cyclize
to indole derivative 31 quickly, but also hydration (forming
32) was more facile, though 20 mol% catalyst was necessary
(Table 2, entry 4).
In a further demonstration of the unique features of 1,
the use of D2O (Scheme 3) resulted in deuteration of indole
and benzofuran at position 2 and 3,[9] which otherwise re-
quires strong base conditions and protecting groups.[10] Liter-
ature on other alkyne hydration catalysts[11] would suggest
that only position 3 would be deuterated, but the bifunction-
al ligands of 1 promote extensive H/D exchange at the stage
Experimental Section
Preparative procedure for cyclization of 11: In a glove box, a scintillation
vial with stir bar was charged with 11 (0.0993 g, 0.6123 mmol), which was
dissolved in dry and deoxygenated THF (5 mL). To this was added 1
(0.0131 g, 0.0131 mmol), the vial was sealed with its cap and the reaction
mixture was subjected to heating at 708C in an oil bath for 2 h. The reac-
tion was monitored by TLC for complete disappearance of starting mate-
rial following which the solvents were evaporated by rotary evaporation
and the product was purified by column chromatography (2:1 hexanes/
ethyl acetate) to give 22 as an orange solid (0.092 g, 92%).
Preparative procedure for cyclization of 16: In a glove box, a scintillation
vial with stir bar was charged with 16 (0.0943 g, 0.535 mmol), which was
dissolved in deoxygenated acetone (5 mL). To this was added 1 (0.0106 g,
0.0106 mmol), the vial was sealed with its cap and the reaction mixture
was subjected to heating at 708C in an oil bath for 1 h. The reaction was
monitored by TLC for complete disappearance of starting material fol-
lowing which the solvents were evaporated by rotary evaporation and the
product was purified by column chromatography (2:1 hexanes/ethyl ace-
tate) to give 25 as an off-white solid (0.071 g, 75%).
7994
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 7992 – 7995

Bifunctional Ruthenium Catalyst
COMMUNICATION
Varela-Fernandez, C. Gonzalez-Rodriguez, J. A. Varela, L. Castedo,
C. Saa, Org. Lett. 2009, 11, 5350.
Acknowledgements
[4] For recent literature, see: a) L. Ackermann, R. Sandmann, M. V.
Kondrashov, Synlett 2009, 1219, and references therein; b) G. R.
Humphrey, J. T. Kuethe, Chem. Rev. 2006, 106, 2875.
The NSF is thanked for support of this work, and Dr. LeRoy Lafferty is
thanked for assisting with NMR spectroscopy analysis.
[5] a) D. B. Grotjahn, D. A. Lev, J. Am. Chem. Soc. 2004, 126, 12232.
Catalyst 1 is now available from Strem Chemicals. For mechanistic
studies, see: b) D. B. Grotjahn, E. J. Kragulj, C. D. Zeinalipour-
Yazdi, V. Miranda-Soto, D. A. Lev, A. L. Cooksy, J. Am. Chem. Soc.
2008, 130, 10860; c) For a recent review, see: D. B. Grotjahn, Dalton
Trans. 2008, 6497, and references therein; d) A. Labonne, T. Krib-
ber, L. Hintermann, Org. Lett. 2006, 8, 5853; e) A. Labonne, L. Hin-
termann, Synthesis 2007, 1121.
[6] a) C. Bianchini, J. A. Casares, M. Peruzzini, A. Romerosa, F. Zano-
bini, J. Am. Chem. Soc. 1996, 118, 4585; b) S. Ogo, K. Uehara, T.
Abura, Y. Watanabe, S. Fukuzumi, J. Am. Chem. Soc. 2004, 126,
16520.
Keywords: alkynes · benzofurans · deuterium · hydration ·
indoles
[1] a) S. Hibino, T. Choshi, Nat. Prod. Rep. 2002, 19, 148; b) R. R.
Gupta, Heterocyclic Chemistry, Vol. 2, Springer, New York, 1999, p
193; c) A. Rahman, A. Basha, Indole Alkaloids, Harwood Academic
Publishers, Amsterdam, 1998, p. 141; d) F. Diederich, P. J. Stang,
R. R. Tykwinski, Acetylene Chemistry, Wiley-VCH, Weinheim, 2005,
Chapter 2, p. 51.
[7] See the Supporting Information for experimental details.
[8] a) N. Sakai, K. Annaka, T. Konakahara, J. Org. Chem. 2006, 71,
3653; b) N. Sakai, K. Annaka, A. Fujita, A. Sata, T. Konakahara, J.
Org. Chem. 2008, 73, 4160; c) J. Y. Park, P. R. Ullapu, H. Choo, J. K.
Lee, S.-J. Min, A. N. Pae, Y. Kim, D.-J. Back, Y. S. Cho, Eur. J. Org.
Chem. 2008, 5461.
[9] Control experiments showed that neither indole nor benzofuran
were deuterated at positions 2 or 3 under identical conditions, even
after 168 h reaction time.
[10] a) W-M. Yao, K. Gawrisch, J. Labelled Compd. Radiopharm. 1999,
42, 709; b) S. Anto, G. S. Getvoldsen, J. R. Harding, J. R. Jones, S-Y.
Lu, J. C. Russel, J. Chem. Soc. Perkin Trans. 2 2000, 2208; c) J. S. L.
Ibaceta-Lizana, A. H. Jackson, N. Parsitpan, P. V. R. Shannon, J.
Chem. Soc. Perkin Trans. 2 1987, 1221.
[11] M. Tokunaga, T. Suzuki, N. Koga, T. Fukushima, A. Horiuchi, Y.
Wakatsuki, J. Am. Chem. Soc. 2001, 123, 11917.
[12] D. Grotjahn, unpublished results.
[13] J. Atzrodt, V. Derdau, T. Fey, J. Zimmermann, Angew. Chem. 2007,
119, 7890–7911; Angew. Chem. Int. Ed. 2007, 46, 7744–7765.
[14] G. Erdogan, D. B. Grotjahn, J. Am. Chem. Soc. 2009, 131, 10354.
[2] a) G. D. McGallion, Curr. Org. Chem. 1999, 3, 67; b) M. G. Kadieva,
E. T. Oganesyan, Chem. Heterocycl. Compd. 1997, 33, 1245; c) Com-
prehensive Heterocyclic Chemistry II, Vol. 2 (Eds.: A. R. Katritzky,
C. W. Rees, E. F. V. Scriven), Pergamon Press, New York, 1996,
p. 259.
[3] a) K. Murai, S. Hayashi, N. Takaychi, Y. Kita, H. Fujioka, J. Org.
Chem. 2009, 74, 1418; b) I. Nakamura, Y. Sato, S. Konta, M. Terada,
Tetrahedron Lett. 2009, 50, 2075; c) S. Mehta, J. Waldo, R. C.
Larock, J. Org. Chem. 2009, 74, 1141; d) A. McLory, B. M. Trost,
Angew. Chem. 2007, 119, 2120; Angew. Chem. Int. Ed. 2007, 46,
2074; e) V. Terrasson, J. Michaux, A. Gaucher, J. Wehbe, S. Marque,
D. Prim, J. Campagne, Eur. J. Org. Chem. 2007, 5332; f) S. Tang, Y.
Xie, J. Li, N. Wang, Synthesis 2007, 1841; g) Y. Yin, W. Ma, Z. Chai,
G. Zhao, J. Org. Chem. 2007, 72, 5731; h) N. Sakai, K. Annaka, T.
Konakahara, Tetrahedron Lett. 2006, 47, 631; i) T. Kurisaki, T.
Naniwa, H. Yamamoto, H. Iragawa, M. Nishizawa, Tetrahedron Lett.
2007, 48, 1871; j) K. Hiroya, S. Itoh, T. Sakamoto, J. Org. Chem.
2004, 69, 1126; k) S. Cacchi, G. Fabrizi, L. Parisi, Synthesis 2004,
1889; l) J. Ezquerra, C. Pedregal, C. Lamas, G. Barluenga, M. Perez,
M. Garcia-Martin, J. Gonzalez, J. Org. Chem. 1996, 61, 5804; m) N.
Isono, M. Lautens, Org. Lett. 2009, 11, 1329; n) X. Li, A. R. Chia-
nese, T. Vogel, R. H. Crabtree, Org. Lett. 2005, 7, 5437; o) C. Marti-
nez, R. Alvarez, J. M. Aurrecoechea, Org. Lett. 2009, 11, 1083; p) Y.
Hu, Y. Zhang, Z. Yeng, R. Fathi, J. Org. Chem. 2002, 67, 2365; q) A.
Received: April 23, 2010
Published online: June 11, 2010
Chem. Eur. J. 2010, 16, 7992 – 7995
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7995