Platinum-catalyzed formation of cyclic-ketone-fused indoles from N-(2-alkynylphenyl)lactams

Article abstract of DOI:10.1002/anie.200702931

(Chemical Equation Presented) An oxygen atmosphere aids the efficient formation of highly substituted ring-fused indoles by the versatile title reaction through an initial cyclization followed by the sequential migration of two groups. The cycloisomerizat

Full text of DOI:10.1002/anie.200702931

Communications
DOI: 10.1002/anie.200702931
Transition-Metal Catalysis
Platinum-Catalyzed Formation of Cyclic-Ketone-Fused Indoles from
N-(2-Alkynylphenyl)lactams**
Guotao Li, Xiaogen Huang, and Liming Zhang*
ortho-Heterosubstituted aryl alkynes are excellent substrates
for Au- and Pt-catalyzed cycloisomerization.[1] Substituted
benzofurans, indoles, and benzothiophenes have been syn-
thesized by intramolecular carboalkoxylation,[2] carboamin-
[3b]
ation,[2a] acylamination,[3a] sulfonylamination,
carbothiol-
ation,[4a] and silylthiolation[4b] of the C C triple bond. A
common feature of this versatile chemistry is that the
substituent (allyl, benzyl, a-alkoxyalkyl, acyl, sulfonyl, or
silyl) on the heteroatom undergoes selective 1,3-migration in
the forming heteroaromatic ring via an alkenyl metal
intermediate A (Scheme 1). A salient feature of gold and
ꢀ
the 1,2-migration process could compete with 1,3-migration, it
was apparent that controlling elements other than steric
factors would be needed to favor substantially the desired 1,2-
migration. We reasoned that, in the case of 2-alkynyl anilines,
appropriate tethering of the two substituents on the
nitrogen atom should abolish the 1,3-migration and
promote the 1,2-migration. We selected lactams 1[3] as
substrates and envisaged that Au/Pt-catalyzed sequen-
tial cyclization, 1,2-acyl migration, and 1,2-migration
of R to the metallocarbenoid moiety[7] would result in
the formation of highly substituted cyclic-ketone-fused
indoles 2 (Scheme 2).[8] Alternatively, the 1,2-acyl
migration could proceed stepwise via the acylium
intermediate C (route b). Interestingly, this reaction
can be viewed as an intramolecular insertion of one
Scheme 1. Pt/Au-catalyzed reactions of ortho-heterosubstituted aryl alkynes.
ꢀ
end of the C C triple bond into the lactam amide bond
with concurrent 1,2-migration of the substituent on the
triple bond.
platinum chemistry is that alkenyl gold and alkenyl platinum
species can react with electrophiles at the distal end of the
We examined the reaction of g-lactam 3 in the presence of
Au/Pt catalysts under a variety of conditions (Table 1). The
anticipated cyclic-ketone-fused indole 4 was formed upon the
treatment of 3 with IPrAuNTf2 (5 mol%) in 1,2-dichloro-
ethane at reflux (Table 1, entry 1). Two major side products
were identified, one of which was inseparable from compound
[5]
ꢀ
C C double bond to form metallocarbenoids. We were
surprised that no electrophilic migration of R2 in A to the 2-
position had been reported, although a metallocarbenoid
species B would be a plausible intermediate. Herein, we
report the synthesis of substituted indoles fused to cyclic
ketones by a Pt-catalyzed 1,2-acyl migration of this type.
We speculated that steric interactions may play a role in
these highly selective 1,3-migrations, as internal alkynes had
always been used (i.e., R1 ¼
6
H). Indeed, the treatment of 2-
ethynyl-N-benzyl-N-methylaniline (R1 = H) with IPrAuNTf2
(5 mol%; Tf = trifluoromethanesulfonyl) led to the formation
of an inseparable mixture of 3-benzyl-1-methylindole and 2-
benzyl-1-methylindole in a combined yield of 50% [Eq. (1);
Bn = benzyl].[6] Although this initial result did suggest that
[*] Dr. G. Li, Dr. X. Huang, Prof. Dr. L. Zhang
Department of Chemistry, University of Nevada, Reno
1664 North Virginia Street, Reno, NV 89557 (USA)
Fax: (+1) 775-784-6804
E-mail: lzhang@chem.unr.edu
Homepage: http://wolfweb.unr.edu/homepage/lzhang/
[**] We thank the University of Nevada, Reno for support.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Scheme 2. Proposed Pt/Au-catalyzed formation of highly substituted
indoles fused to cyclic ketones.
346
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 346 –349

Angewandte
Chemie
Table 1: Optimization of the reaction conditions for the Pt/Au-catalyzed
synthesis of indoles fused to cyclic ketones.
These studies led us to suspect that O2 might play a role in
the catalytic process.[14] Indeed, the reaction in the presence of
PtCl2 (5 mol%) was very slow under nitrogen: After 15 h,
77% of 3 remained (Table 1, entry 9). When this PtCl2-
catalyzed reaction was carried out under an atmosphere of O2
with a substrate concentration of 0.01m, the yield of 4
increased to 82% (Table 1, entry 10). However, the reaction
took 42 h. The need to shorten the reaction time and the
observed benefit of O2 led us to test PtCl4.[15] To our delight,
the desired indole 4 was formed in 72% yield in 1 h, and the
excellent selectivity was preserved (Table 1, entry 11). This
observation suggests that PtIV may be the real catalytic species
even when PtCl2 is used. Finally, the yield of 4 was increased
to 83% with 10 mol% of PtCl4 (Table 1, entry 12).
Entry[a] Catalyst[b] Reaction conditions
t
Yield
[h] of
4 [%][c]
4/5/6
1
2
3
4
5
6
7
8
9
IPrAuNTf2 DCE, 808C[d]
15 62
1:0.24:0.06
–
1:0.29:1
1:0.07:0.07
1:0.03:0.08
1:0:0
1:0.22:0.31
–
–
AuIII
DCE, 808C[d]
15 0[e,f]
15 28[g]
PtCl2
PtCl2
PtCl2
PtCl2
PtCl2
PtCl2
PtCl2
PtCl2
PtCl4
PtCl4
toluene, 808C[d]
DCE, 808C[d]
6
4
73
anisole, 808C[d]
33[g]
DCE, reflux, MS (4 Š), N2 15 20[h]
We set out to probe the scope of this chemistry under the
optimized conditions described in Table 1, entry 12. The
reaction worked well when alkyl groups were present at the
alkyne terminus. For example, substrates with a methyl group
(Table 2, entry 1) or a functionalized propyl group, such as 3-
bromopropyl (Table 2, entry 3) or 3-benzyloxypropyl
(Table 2, entry 4), were transformed into the corresponding
highly substituted indoles 8 in excellent yields. Surprisingly,
the bromide product 8c was contaminated with the corre-
sponding chloride in about 20% yield.[16] The chlorine atom
DCE, reflux, CO (1 atm)
toluene, cod, 808C, N2
DCE, reflux, N2
DCE, reflux[j]
9
37[g]
15 0[f]
15 10[i]
42 82[k]
10
11
12
1:0.03:0.02
1:0.04:0.05
1:0.03:0.02
DCE, reflux[l]
1
1
72
DCE, reflux[j]
83[k]
[m]
[a] Substrate concentration: entries 1–9, 0.03m; entries 10–12, 0.01m.
[b] Catalyst loading: 5 mol%. [c] Yield determined by NMR spectroscopy
with diethyl phthalate as the internal reference. [d] The reaction was
carried out in a capped vial. [e] Decomposition of the catalyst was
observed. [f] The starting material remained unchanged. [g] No starting
material remained. [h] Much of substrate 3 (80%) remained. [i] Much of
substrate 3 (77%) remained. [j] The reaction was carried out under O2
(1 atm). [k] Yield of the isolated product. [l] A drying tube was mounted
on the top of the condenser. [m] Catalyst loading: 10 mol%. DCE=1,2-
dichloroethane, MS=molecular sieves, cod=1,5-cyclooctadiene.
Table 2: Scope of the reaction with g-lactam substrates.
Entry[a]
7
R
R’
t [h] Yield of 8/9[c]
8 [%][b]
1
2
7a
7b
H
H
Me
1.5 83
28:1
18:1
14
75[d]
4. A singlet at d = 6.39 ppm and an aromatic doublet at d =
3
4
5
7c
7d
7e
H
H
H
11
2
23
88[e]
29:1
19:1
14:1
7.95 ppm (shifted downfield relative to the aromatic signals
81
70
1
for 4) in the H NMR spectrum of the mixture led to the
Ph
assignment of this by-product as the Friedel–Crafts product
5.[9] The other was the uncyclized carboxylic acid 6. The
formation of 5 and 6 indicates that the 1,2-acyl migration is a
stepwise process via the acylium intermediate C. In an
attempt to limit the side reactions, we screened quickly
various conditions in capped vials. Although the results with
other Au catalysts, such as dichloro(pyridine-2-
carboxylato)gold(III)[10] (Table 1, entry 2), were disappoint-
ing, we observed much better selectivities for 4 over 5 with
PtCl2 in 1,2-dichloroethane (Table 1, entry 4) or anisole[3]
(Table 1, entry 5). The result in 1,2-dichloroethane was
difficult to reproduce on a larger scale. Nevertheless, we
continued our study with reactions in flasks, which allowed
better control of the reaction conditions. The presence of
additives known to improve PtCl2 catalysis (e.g., CO
(1 atm),[11] 1,5-cyclooctadiene[12]) led to either low yield and
selectivity (Table 1, entry 7) or no reaction (Table 1, entry 8).
Surprisingly, compound 5 was not observed when the reaction
was carried out in the presence of 4-Š molecular sieves under
nitrogen (Table 1, entry 6), and the conversion of the
substrate was low.[13]
6
7
8
7 f
H
H
H
11
34
14
73
8:1
7g
7h
70
10:1
60
>30:1
>30:1
9
10
11
12
13
7i
7j
7k
7l
H
48
22
1
1
1
52[f]
73/95[g] 1:0
89
71
65
H
H
nBu
nBu
6-MeO[h]
4-Br[i]
1:0
>30:1
7m 4-EtO2C[i] nBu
>30:1[j]
[a] Substrate concentration: 0.01m. [b] Yield of the isolated product.
[c] The product ratio was determined by 1H NMR spectroscopy or
separation of the products by column chromatography. The uncyclized
acid was present in less than 5% yield. [d] Value includes a 4% yield of
the corresponding by-product 9. [e] Value includes a 20% yield of the
corresponding chloride. [f] Some of substrate 7i (20%) remained.
[g] PtCl2 (5 mol%) was used as the catalyst (reaction time: 22 h). [h] The
methoxy substituent is ortho to the nitrogen atom. [i] The substituent is
para to the nitrogen atom. [j] The uncyclized acid was formed in 13%
yield.
Angew. Chem. Int. Ed. 2008, 47, 346 –349ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
347

Communications
probably comes from PtCl4, as the reaction of 7c catalyzed by
IPrAuNTf2 (5 mol%) did not provide a detectable amount of
the chloride. Remarkably, lactam 7b with a cyclopropyl
substituent underwent smooth cyclopropyl migration to yield
the 3-cyclopropylindole product 8b in 71% yield. The
reactions of substrates with aryl groups having different
electronic characteristics at the alkyne terminus led to the
corresponding 3-aryl indoles in good yields (Table 2,
entries 5–7). Even an aldehyde substituent on the aromatic
ring was tolerated (Table 2, entry 7). Moreover, alkenyl
groups, such as trans-b-styryl (Table 2, entry 8) and trans-
hex-1-en-1-yl (Table 2, entry 9), were also suitable: the 3-
alkenyl indoles 8h and 8i were isolated in fairly good yields
with retention of the trans geometry. The reaction of 7i was
rather sluggish; 20% of the substrate remained after 2 days.
Somewhat surprisingly, PtCl2 was a better catalyst than PtCl4
for the reaction of the ethynyl substrate 7j. The 3-unsub-
stituted indole 8j was formed in 95% yield in the presence of
PtCl2 (5 mol%; Table 2, entry 10).
competitive, as it results in the formation of a 6-membered
ring. Interestingly, the d-lactam 12 also underwent the desired
reaction to afford the indole 13 with a fused seven-membered
cyclic ketone, albeit in only 45% yield [Eq. (4)].
Further studies revealed that substitution of the benzene
ring of the substrate was generally allowed, regardless of the
electronic nature of the substituent (Table 2, entries 11–13),
and that the reaction was more efficient with an electron-
donating substituent (Table 2, entry 11). The uncyclized acid
was isolated from the reaction of 7m in 13% yield (Table 2,
entry 13), which is much higher than the yield of the acid in all
other cases (< 5%). Good to excellent selectivities were
observed for the formation of the desired product 8 over the
Friedel–Crafts acylation (i.e., the formation of compound 9)
in most reactions described in Table 2. Moreover, in the case
of substrate 7j (R’ = H; Table 2, entry 10) the by-product 9j
was not detected, which indicates that the cyclization of the
acylium species to the 2-position of 3-platinoindole C (M =
Pt) is much faster than its cyclization to the 7-position when
the steric environments of the two positions are similar. Not
surprisingly, the formation of 9 was not observed when the 7-
position was substituted (Table 2, entry 11).
These experimental results support the general mecha-
nism via a Pt-containing acylium intermediate C that is
outlined in Scheme 2.[17] Our attempt to trap the Pt carbenoid
(D in Scheme 2) with either Ph2SO[18] or styrene failed,
presumably because the migration of R in D and subsequent
rearomatization is facile.[19]
In conclusion, we have developed a PtCl4/PtCl2-catalyzed
cycloisomerization of N-(2-alkynylphenyl)lactams to form
substituted indoles fused to cyclic ketones. Various substitu-
ents at the alkyne terminus and on the benzene ring are
tolerated, and lactams with different ring sizes are accom-
modated. This transformation, which is a net intramolecular
insertion of one end of the alkyne into the lactam amide bond
with concurrent migration of the substituent at the alkyne
terminus, offers efficient access to ring-fused, highly substi-
tuted indoles.
We attempted to extend this chemistry to lactams with
different lactam-ring sizes. The b-lactam 10 with an ethynyl Experimental Section
PtCl4 (10 mol%) or PtCl2 (if specified; 10 mol%) was added to a
substituent on the aromatic ring reacted efficiently to yield
selectively the benzene-fused pyrrolizinone 11 in 85% yield;
the corresponding Friedel–Crafts acylation product was not
observed [Eq. (2)]. PtCl2 catalyzed this reaction better than
0.01m solution of the lactam in anhydrous 1,2-dichloroethane under
an atmosphere of O2, and the resulting mixture was refluxed for the
time indicated. Upon the completion of the reaction, the solvent was
removed under vacuum, and the residue was purified by flash column
chromatography on silica gel (eluent: hexanes/ethyl acetate 3:1) to
yield the desired product.
Received: July 2, 2007
Revised: September 1, 2007
Published online: November 12, 2007
Keywords: carbenes · homogeneous catalysis · insertion ·
.
platinum · rearrangement
PtCl4, as observed for the g-lactam substrate 7j, to provide 11
in 99% yield. b-Lactams with either a phenyl or a cyclohexyl
substituent at the alkyne terminus underwent smooth PtCl4-
catalyzed transformations. Although the combined yield of
the desired product and the Friedels–Craft product was close
to quantitative in each case, the desired products were
favored only slightly [Eq. (3)]. In these cases the cyclization
of the acylium species to the benzene ring becomes more
[1] For selected reviews on Au and/or Pt catalysis, see: a) A. S. K.
Hashmi, G. J. Hutchings, Angew. Chem. 2006, 118, 8064; Angew.
Chem. Int. Ed. 2006, 45, 7896; b) L. Zhang, J. Sun, S. A. Kozmin,
Adv. Synth. Catal. 2006, 348, 2271; c) R. A. Widenhoefer, X. Q.
Han, Eur. J. Org. Chem. 2006, 4555; d) E. Jimenez-Nunez, A. M.
Echavarren, Chem. Commun. 2007, 333; e) N. T. Patil, Y.
348
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 346 –349

Angewandte
Chemie
Yamamoto, ARKIVOC 2007, 5, 6; f) N. Marion, S. P. Nolan,
Angew. Chem. 2007, 119, 2806; Angew. Chem. Int. Ed. 2007, 46,
2750; g) D. J. Gorin, F. D. Toste, Nature 2007, 446, 395; h) A.
Fꢀrstner, P. W. Davis, Angew. Chem. 2007, 119, 3478; Angew.
Chem. Int. Ed. 2007, 46, 3410; i) A. S. K. Hashmi, Chem. Rev.
2007, 107, 3180.
[9] This structural assignment is supported by the isolation and
characterization of the side product from the reaction of lactam
7a, in which a methyl group is at the alkyne terminus.
[10] For the first use of the AuIII complex as a precatalyst, see:
A. S. K. Hashmi, J. P. Weyrauch, M. Rudolph, E. Kurpejovic,
Angew. Chem. 2004, 116, 6707; Angew. Chem. Int. Ed. 2004, 43,
6545; for other reactions catalyzed by this complex, see: a) S.
Wang, L. Zhang, J. Am. Chem. Soc. 2006, 128, 8414; b) S. Wang,
L. Zhang, J. Am. Chem. Soc. 2006, 128, 14274.
[2] a) A. Fꢀrstner, P. W. Davies, J. Am. Chem. Soc. 2005, 127, 15024;
b) I. Nakamura, Y. Mizushima, Y. Yamamoto, J. Am. Chem. Soc.
2005, 127, 15022.
[3] a) T. Shimada, I. Nakamura, Y. Yamamoto, J. Am. Chem. Soc.
2004, 126, 10546; b) I. Nakamura, U. Yamagishi, D. Song, S.
Konta, Y. Yamamoto, Angew. Chem. 2007, 119, 2334; Angew.
Chem. Int. Ed. 2007, 46, 2284.
[11] A. Fꢀrstner, P. W. Davies, T. Gress, J. Am. Chem. Soc. 2005, 127,
8244.
[12] I. Nakamura, Y. Mizushima, Y. Yamamoto, J. Am. Chem. Soc.
2005, 127, 15022.
[4] a) I. Nakamura, T. Sato, Y. Yamamoto, Angew. Chem. 2006, 118,
4585; Angew. Chem. Int. Ed. 2006, 45, 4473; b) I. Nakamura, T.
Sato, M. Terada, Y. Yamamoto, Org. Lett. 2007, 9, 4081.
[5] For selected examples, see: a) C. Nieto-Oberhuber, M. P.
Munoz, E. Bunuel, C. Nevado, D. J. Cardenas, A. M. Echavar-
ren, Angew. Chem. 2004, 116, 2456; Angew. Chem. Int. Ed. 2004,
43, 2402; b) K. Miki, K. Ohe, S. Uemura, J. Org. Chem. 2003, 68,
8505; c) M. J. Johansson, D. J. Gorin, S. T. Staben, F. D. Toste, J.
Am. Chem. Soc. 2005, 127, 18002; d) N. Kim, Y. Kim, W. Park, D.
Sung, A. K. Gupta, C. H. Oh, Org. Lett. 2005, 7, 5289; e) L. Peng,
X. Zhang, S. Zhang, J. Wang, J. Org. Chem. 2007, 72, 1192;
f) B. P. Taduri, S. M. A. Sohel, H.-M. Cheng, G.-Y. Lin, R.-S. Liu,
Chem. Commun. 2007, 2530; g) G. Li, L. Zhang, Angew. Chem.
2007, 119, 5248; Angew. Chem. Int. Ed. 2007, 46, 5156; for a
review, see Ref. [1h].
[13] The reason for this decreased reactivity is not clear, but the
molecular sieves may affect the Pt catalyst and partially trap the
reactive acylium intermediate.
[14] PtCl2-catalyzed reactions in vials have been observed previously
to be more efficient than those in flasks under N2: B. A.
Bhanu Prasad, F. K. Yoshimoto, R. Sarpong, J. Am. Chem. Soc.
2005, 127, 12468.
[15] For selected previous examples of the use of PtCl4, see: a) W.
Baidossi, M. Lahav, J. Blum,J. Org. Chem. 1997, 62, 669; b) C. H.
Oh, S. Y. Bang, C. Y. Rhim, Bull. Korean Chem. Soc. 2003, 24,
887; c) K. Hiroya, S. Matsumoto, M. Ashikawa, K. Ogiwara, T.
Sakamoto, Org. Lett. 2006, 8, 5349; d) J. Marco-Contelles, A. N.
Arroyo, S. Anjum, E. Mainetti, N. Marion, K. Cariou, G.
Lemiꢁre, V. Mouriꢁs, L. Fensterbank, M. Malacria, Eur. J. Org.
Chem. 2006, 4618.
[6] The formation of the known compound 3-benzyl-1-methylindole
[16] The structure of the chloride derivative was assigned on the basis
of the close resemblance of the signals in its NMR spectra to
those of the bromide product as well as the characteristic isotope
pattern in its ESI+ mass spectrum.
[17] Additional evidence for the existence of the acylium intermedi-
ate C and/or the preceding spiro-zwitterionic intermediate is the
formation of the ethyl ester of acid 6 in 30% yield along with the
product 4 (45% yield) when EtOH (1.5 equiv) was added to the
reaction mixture to trap the intermediates. Moreover, when
EtOH (at reflux) was used as the solvent, the ethyl ester was
isolated in 68% yield, and the product 4 was observed (< 2%).
[18] C. A. Witham, P. Mauleon, N. D. Shapiro, B. D. Sherry, F. D.
Toste, J. Am. Chem. Soc. 2007, 129, 5838.
1
was ascertained by comparison of its H and 13C NMR spectra
with reported data: Y. Jia, J. Zhu, J. Org. Chem. 2006, 71, 7826.
The existence of 2-benzyl-1-methylindole was assumed on the
basis of the characteristic singlet at d = 6.27 ppm in the 1H NMR
spectrum of the mixture, as well as the signal at d = 4.14 ppm
expected for the benzylic hydrogen atoms.
[7] For examples of the migration of alkyl groups to Pt carbenoids,
see: a) H. Kusama, Y. Miyashita, J. Takaya, N. Iwasawa, Org.
Lett. 2006, 8, 289; b) J. Sun, M. P. Conley, L. Zhang, S. A.
Kozmin, J. Am. Chem. Soc. 2006, 128, 9705; c) H. Funami, H.
Kusama, N. Iwasawa, Angew. Chem. 2007, 119, 927; Angew.
Chem. Int. Ed. 2007, 46, 909.
[8] For selected examples of the Au/Pt-catalyzed preparation of
indoles, see: a) A. Arcadi, G. Bianchi, F. Marinelli, Synthesis
2004, 610; b) K. Cariou, B. Ronan, S. Mignani, L. Fensterbank,
M. Malacria, Angew. Chem. 2007, 119, 1913; Angew. Chem. Int.
Ed. 2007, 46, 1881.
[19] This result also suggests that the evolution of the Pt carbenoid D
may not be the rate-limiting step, which would explain why
substrates 7 with a primary alkyl group at the alkyne terminus
reacted faster than those with an aryl group, although aryl
groups usually show greater aptitude for nucleophilic migration.
Angew. Chem. Int. Ed. 2008, 47, 346 –349ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
349