Acid-catalyzed acylation reaction via C-C bond cleavage: A facile and mechanistically defined approach to synthesize 3-acylindoles

Article abstract of DOI:10.1039/c4cc05047a

A facile acid-catalyzed acylation of indoles with 1,3-dione as an eco-friendly acylating agent was developed. This protocol combines C-C bond cleavage and heterocyclic C-H bond functionalization to form new C-C bonds. Based on the detailed mechanistic studies, a credible mechanistic pathway was proposed. This journal is

Full text of DOI:10.1039/c4cc05047a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Acid-catalyzed acylation reaction via C–C bond
cleavage: a facile and mechanistically defined
approach to synthesize 3-acylindoles†
Cite this: Chem. Commun., 2014,
50, 12181
Received 2nd July 2014,
Accepted 21st August 2014
Qi Xing,ab Pan Li,a Hui Lv,a Rui Lang,a Chungu Xiaa and Fuwei Li*a
DOI: 10.1039/c4cc05047a
www.rsc.org/chemcomm
A facile acid-catalyzed acylation of indoles with 1,3-dione as an
eco-friendly acylating agent was developed. This protocol combines
C–C bond cleavage and heterocyclic C–H bond functionalization to
form new C–C bonds. Based on the detailed mechanistic studies, a
credible mechanistic pathway was proposed.
Catalytic unstrained C–C bond cleavage, similar to the emerging
C–H bond functionalization,1 has attracted much attention because
of its fundamental scientific appeal and potential application in
organic synthesis.2 Among the reported catalytic systems, transition
metal catalysis has found increasing applications in the C–C bond
Scheme 1 C–C bond cleavage of 1,3-diones to form aryl ketones.
cleavage.2–9 However, such reaction has rarely been realized by a the corresponding carbonyl products has not been achieved.
simple and metal-free catalytic system. It is known that both C–C Herein, we describe an acid-catalyzed acylation reaction of
bond cleavage and C–H functionalization changed conventional indoles with 1,3-diones under mild and solvent-free conditions
synthetic pathways by nontraditional C–X (X = C or H) bond (Scheme 1). Based on the isolated intermediates, this method
disconnection and reorganization of existing molecular skeletons. was experimentally found to involve two C–C bond formation
Therefore, the development of a novel synthetic methodology processes and two C–C bond cleavage processes with the release
combining these two strategies for the preparation of important of R5COCH2R4 as a leaving group. Moreover, H2O generated in
organic skeletons is more scientifically interesting and desirable. the first C–C bond formation reaction also played an important
Lei et al. reported an efficient Cu-catalyzed arylation of ketones role in the second C–C bond cleavage and its oxygen atom was
via C–C cleavage of 1,3-diones (Scheme 1).8a In such a process, finally incorporated into the 3-acylindole product. Notably, such
functionalized ArX acted as an electrophile to react with 1,3-diones tandem transformation could be facilely scaled up to ten grams
generating Cu(III) intermediates, which underwent C–C bond and the HOTf catalyst was recycled when using water as the
cleavage to yield a-aryl ketone products and release an equivalent solvent. This protocol provides a practical and mild approach to
of R3COOK with the assistance of external H2O. It is worth synthesize 3-acylindoles, which are key structural units in many
pointing out that H2O is not incorporated into the desired biologically active compounds and versatile feedstocks for the
product. Furthermore, the C–C bond cleavage of 1,3-diones with preparation of alkaloids and other heterocyclic intermediates.11
alcohols or amines as partners was also intensively investigated
Initially, with the reaction of 3-methylpentane-2,4-dione (1a)
to yield the corresponding esters or amides.10 However, to our and N-methylindole (2a) as the model reaction, the reaction
knowledge, the arylation of 1,3-diones with unfunctionalized conditions were optimized (Table S1 in the ESI†). Usually, Lewis
heterocyclic compounds through C–C bond cleavage to obtain acids are appropriate activators for 1,3-diones, among them
Fe(OTf)3 gave the highest yield (91%) of 3aa without any solvent
a State Key Laboratory for Oxo Synthesis and Selective Oxidation,
Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences,
Lanzhou 730000, China. E-mail: fuweili@licp.cas.cn
at 80 1C for 12 h (entries 1–9). Considering that metal triflates
might release HOTf in the presence of H2O, HOTf was then
tested to catalyze such acylation. Surprisingly, 3aa was obtained
in 93% yield (entries 10 and 11). In view of the lower price of
HOTf and avoiding metal contamination of the product, HOTf
was employed as the optimized acid catalyst for the subsequent
b University of Chinese Academy of Sciences, Beijing, 100049, China
† Electronic supplementary information (ESI) available: Detailed experimental
procedures and spectral data for all compounds, including scanned images of 1
and 13C NMR spectra. See DOI: 10.1039/c4cc05047a
H
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 12181--12184 | 12181

View Article Online
Communication
ChemComm
Table 1 Acid catalyzed reactions of 2a with different 1,3-dionesa
Entry
1,3-Diketone
Product
Yield (%)
1
2
3
4
5
6
7
8
9
1b, R4 = H, R3 = R5 = Me
1c, R4 = H, R3 = R5 = Et
3aa, R3 = Me
3ca, R3 = Et
3da, R3 = iPr
3ea, R3 = CF3
3fa, R3 = Ph
3aa, R3 = Me
3aa, R3 = Me
3ea, R3 = CF3
3aa, R3 = Me
81, 58b
75
74
1d, R4 = H, R3 = R5 = iPr
1e, R4 = H, R3 = R5 = CF3
1f, R4 = H, R3 = R5 = Ph
79
38c, 49d
58c
1g, R4 = H, R3 = Me, R5 = Ph
1h, R4 = H, R3 = Me, R5 = OEt
1i, R4 = H, R3 = CF3, R5 = Me
1j, R4Á Á ÁR5 = –(CH2)3–, R3 = Me
34c, 43d
78
78
a
Conditions: 1 (1.5 mmol), 2a (0.5 mmol), 15 mol% HOTf, in air, 24 h,
b
c
80 1C, isolated yield. 2a (20 g), 5 mol% Fe(OTf)3, 36 h. 0.5 mmol
HOTf and 130 1C. 10 mol% Fe(OTf)3 with the addition of 0.5 mmol
d
Scheme 2 HOTf-catalyzed synthesis of indolyl-3-ethanone (3). Condi-
tions: 1a (1.5 mmol), 2 (0.5 mmol), 80 1C, 15 mol% of HOTf, 24 h in air,
isolated yield.
(Tf)2O, 130 1C.
instead, the corresponding b-indolyl-a,b-unsaturated-enone was
generated in 85% yield from their condensation (S-Eq. (2) in the
investigations unless otherwise noted. Prolonging the reaction ESI†). Increasing the HOTf amount and the reaction temperature
time (24 h) could enhance the formation of the desired product could make this reaction happen and the desired 3fa was
in a quantitative HPLC yield (89% isolated yield, entry 12). The isolated in 38% yield, and a higher yield (49%) could be obtained
reaction proceeded even at room temperature to afford 3aa with with Fe(OTf)3 and (Tf)2O as the binary catalyst (entry 5).
a 32% yield (entry 13). Remarkably, an interesting intermediate
When the 1,3-dione was changed to dissymmetric 1g, 3aa
b,b-bisindolyl ketone (D1) was isolated in 49% yield (S-Eq. (1) in rather than 3fa was obtained as the only product in moderate
the ESI†). Interestingly, the HOTf catalyzed acylation even yield (entry 6), possibly owing to the steric effect of the phenyl
worked efficiently affording 91% yield of 3aa (80% isolated group. Similarly, the reaction of 1h also gave 3aa instead of
yield) by using water as the solvent (entry 14), and the HOTf indolyl-3-ester as the product (entry 7). Different from 1h, 3ea
aqueous phase could be facilely separated and reused, providing instead of 3aa was obtained as the sole product with a yield of
72% isolated yield of 3aa in its second run (entry 15).
78% in the reaction of asymmetric 1i with 2a (entry 8). For
We then turned to explore the substrate scope of acylation. cyclic 1,3-diones, ring-opening reaction always occurred in the
As shown in Scheme 2, indoles with different N-protecting presence of a Lewis acid catalyst to give the corresponding ester or
groups reacted smoothly with 1a to produce the corresponding amide when alcohol or amine was used as the nucleophiles.10
indole-3-ketones (3aa–3ae) in moderate to good yields. The influence However, 78% yield of 3aa was obtained in the present acid-
of substituents on the benzene ring was also examined. N-Methyl catalyzed reaction of 1j with indole (entry 9). To our delight, the
indoles bearing an electron-donating group (EDG) or an electron- scaling-up (20 g) reaction of 1b and 2a proceeded efficiently (58%
withdrawing group (EWG) afforded 48–86% yield of the desired yield) utilizing Fe(OTf)3 as the catalyst (entry 1). Comparatively,
3ag–3ak. For indoles with some strong EWGs, no expected product HOTf gave a lower (41%) yield of 3aa.
was obtained. The N-allyl and N-benzyl indoles all worked well,
Clarifying the mechanism of a new synthetic process is very
providing good yields of the expected 3an–3as. Fortunately, the important from the viewpoint of its scientific interest and catalytic
reactions of N-unprotected indoles with 1a also proceeded efficiently, application in designing other synthetic strategies. As reported
giving 3at–3av in good yields. It is necessary to note that these earlier, alcohols and amines could react with 1,3-diones to form
products could not be directly produced via classical Friedel–Crafts esters and amides through a retro-Claisen condensation process.10
reaction due to the involvement of acid chloride.12
From the experimental investigation mentioned above, it could be
Subsequently, the HOTf-catalyzed C–C bond cleavage of proposed that the present acid-catalyzed acylation might not proceed
various 1,3-diones with 2a was also investigated. As presented via retro-Claisen condensation because two key intermediates,
in Table 1, with 15 mol% HOTf as the catalyst, 1b and 1c gave b-indolyl-a,b-unsaturated enone and b,b-bisindolyl ketone, were
the desired 3aa and 3ca in 81% and 75% yields, respectively isolated during the synthesis of 3-acylindoles. To figure out this
(entries 1 and 2). More sterically hindered 1d also underwent hypothesis, a control reaction between 2j and 1b was initially
such C–C bond cleavage smoothly, providing 3da in 74% yield performed at room temperature. As shown in Scheme 3, 22%
(entry 3). The reaction of 2a with the 1,3-diones bearing the CF3 yield of b-indolyl-a,b-unsaturated(E)-enone (C2) was generated
group (1e) proceeded more efficiently, affording 3ea in 79% by the condensation of 2j with 1b and 61% yield of b,b-bisindolyl
yield (entry 4). Surprisingly, the reaction of 1f with 2a failed to ketone (D2) was also obtained besides 8% yield of the desired
give the desired 3fa product under the above conditions, 3aj (eqn (1)). Subsequently, the isolated C2 reacted with one
12182 | Chem. Commun., 2014, 50, 12181--12184
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
Scheme 4 A putative reaction pathway for the acid-catalyzed C–C bond
formation and cleavage reactions of indoles and 1,3-diones.
and R4 groups. The weaker electron donation and less steric
hindrance of R4 will generally favor the formation of II, which
undergoes further addition with a second indole producing III,
which undergoes C–C bond cleavage via b-carbon (to the carbonyl)
elimination to possibly give a tertiary alcohol intermediate (IV) with
the release of a ketone using water as the nucleophile. Finally, IV
might undergo another b-carbon (to the hydroxyl) elimination to
give the desired 3-acylindole, generating one molecule of indole to
undergo further condensation with 1,3-diones to yield II again.
To further prove the above mechanism, the control reactions
of 1b and 2j with water were also investigated. As shown in
Scheme 5, in the absence of water, D2 and C2 could be
obtained, but no 3-acylindole was observed (eqn (7)). Moreover,
the direct transformation of D2 to 3aj with the addition of
excess 4 Å molecular sieves gave only a trace amount of 3aj
(eqn (8)). On the other hand, when D2O was added into the
similar conversion of D1, 52% yield of the desired 3aa was
obtained generating the same amount of N-methyl indole, 44%
of which was deuterated through deuterium transfer from the
hydroxyl group of IV to the C3 position of the resultant indole
(eqn (9), Fig. S3 in the ESI†). Accordingly, the reaction of D1 with
H218O afforded 87% yield of 3aa with the 100% 18O labelled
carbonyl group (eqn (10), Fig. S4 and S5 in the ESI†). These
results confirmed again the involvement of water in the C–C
bond cleavage of III to yield 3-acylindole via intermediate IV.
In conclusion, we have developed a novel and practical acid-
catalyzed synthesis of 3-acylindoles from indoles and 1,3-diones
through C–C bond cleavage. The present transformation works
Scheme 3 Control reactions of the intermediates: reactions of eqn (1) to
(6) were conducted in air. H2O in eqn (4) and (5) was from the hydrous
MeCN.
equivalent amount of 2j to give D2 in 88% yield (eqn (2)). As
expected, such conversion could not occur in the absence of
HOTf (eqn (3)). Under the optimized reaction conditions, D2
could be quantitatively converted into the desired 3aj (eqn (4)).
However, only a trace amount of the product was detected
without the acid catalyst (eqn (5)). Another possible pathway
for present acylation is that water instead of a second indole
adds to C to give the corresponding b-hydroxy ketone, which
releases the desired acylindole via retro-aldol reaction. As displayed
in eqn (6), only a trace amount of 3aa was obtained in the reaction of
C3 with H2O although C3 was quantitatively converted, and a large
amount of 2a, CH3COOH and acetophenone were detected by
GC-MS (Fig. S1 in the ESI†). These results experimentally proved
that C2 and D2 were two intermediates in the synthesis of
3-acylindole and the retro-aldol pathway accounted for only a
very small proportion in this transformation.
To probe the role of these intermediates in depth and gain
more kinetic information on this transformation, we tracked
the concentration distribution of the substrate (2j), intermediates
(C2 and D2) and the product (3aj) in the reaction of 1b and 2j by
HPLC analysis of the reaction mixture every three minutes. As
shown in Fig. S2 in the ESI,† D2 was generated with a 65% yield in
three minutes based on 2j. Subsequently, the concentration of D2
was slowly decreased and the yield of 3aj was increased with
prolonged time, possibly suggesting that the C–C bond cleavage
of D2 to yield 3-acylindole is the rate-determining step of such a
process.
Based on the above investigations, a credible reaction path-
way is proposed in Scheme 4. It is reported that Lewis acid or
Bronsted acid could activate 1,3-dione through protonation or
coordination to increase the electrophilicity of the carbonyl
group, which is attacked by indole to give intermediate II with
the generation of one equivalent of water. The selectivity of II is
dependent on both the electronic and steric properties of R3
Scheme 5 All the reactions were conducted in a glove box.
Chem. Commun., 2014, 50, 12181--12184 | 12183
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
5 For Pd-catalyzed C–C bond cleavage, see: (a) A. J. Grenning and
J. A. Tunge, Angew. Chem., Int. Ed., 2011, 50, 1688–1691; (b) S. W. Youn,
B. S. Kim and A. R. Jagdale, J. Am. Chem. Soc., 2012, 134, 11308–11311;
(c) A. Ziadi and R. Martin, Org. Lett., 2012, 14, 1266–1269; (d) J. R. Bour,
J. C. Green, V. J. Winton and J. B. Johnson, J. Org. Chem., 2013, 78,
1665–1669.
6 For Ni catalyzed C–C bond cleavage, see: M. P. Watson and
E. N. Jacobsen, J. Am. Chem. Soc., 2008, 130, 12594–12595.
7 For Mn-catalyzed C–C bond cleavage, see: (a) C. Zhang, Z. J. Xu,
T. Shen, G. L. Wu, L. R. Zhang and N. Jiao, Org. Lett., 2012, 14,
2362–2365; (b) J. X. Liu, Z. T. Du, T. L. Lu and J. Xu, ChemSusChem,
2013, 6, 2255–2258.
8 For Cu-catalyzed C–C bond cleavage, see: (a) C. He, S. Guo, L. Huang
and A. W. Lei, J. Am. Chem. Soc., 2010, 132, 8273–8275; (b) F. Chen,
C. Qin, Y. X. Cui and N. Jiao, Angew. Chem., Int. Ed., 2011, 50,
11487–11491; (c) L. H. Zou, D. L. Priebbenow, L. Wang, J. Mottweiler
and C. Bolm, Adv. Synth. Catal., 2013, 355, 2558–2563; (d) C. Zhang,
P. Feng and N. Jiao, J. Am. Chem. Soc., 2013, 135, 15257–15262.
under mild conditions with good substrate tolerance and high
product selectivity. Interestingly, the b-indolyl-a,b-unsaturated
enone and b,b-bisindolyl ketone were isolated and found to be
the key intermediates for this process based on the detailed
experimental investigations. Moreover, the C–C bond cleavage
was demonstrated to be the rate-determining step. Compared
with the reported procedures using Friedel–Crafts reaction,12
complex substrates13 or noble metal catalysts,14 this methodology
provides a greener and cheaper alternative for the synthesis of
3-acylindoles from readily available substrates and will inspire more
interesting designs in the application of C–C bond cleavage.
This work was supported by the Chinese Academy of
Sciences and the National Natural Science Foundation of China
(21133011 and 21373246).
´
9 For Fe-catalyzed C–C bond cleavage, see: (a) J. Mecinovic,
R. B. Hamed and C. J. Schofield, Angew. Chem., Int. Ed., 2009, 48,
2796–2800; (b) H. R. Li, W. J. Li, W. P. Liu, Z. H. He and Z. P. Li,
Angew. Chem., Int. Ed., 2011, 50, 2975–2978; (c) C. Qin, T. Shen,
C. H. Tang and N. Jiao, Angew. Chem., Int. Ed., 2012, 51, 6971–6975;
(d) A. P. Dieskau, M. S. Holzwarth and B. Plietker, J. Am. Chem. Soc.,
2012, 134, 5048–5051.
Notes and references
1 For selected reviews on catalytic C–H bond functionalization, see:
(a) L. Ackermann, Chem. Rev., 2011, 111, 1315–1345; (b) P. B. Arockiam,
C. Bruneau and P. H. Dixneuf, Chem. Rev., 2012, 112, 5879–5918.
2 For selected reviews on catalytic carbon–carbon bond cleavage, see: 10 (a) A. Kawata, K. Takata, Y. Kuninobu and K. Takai, Angew. Chem.,
(a) C. H. Jun, Chem. Soc. Rev., 2004, 33, 610–618; (b) T. Seiser and
N. Cramer, Org. Biomol. Chem., 2009, 7, 2835–2840; (c) K. Ruhland,
Eur. J. Org. Chem., 2012, 2683–2706; (d) J. E. M. N. Klein and
B. Plietker, Org. Biomol. Chem., 2013, 11, 1271–1279.
Int. Ed., 2007, 46, 7793–7795; (b) S. Biswas, S. Maiti and U. Jana, Eur.
J. Org. Chem., 2010, 2861–2866; (c) C. B. Rao, D. C. Rao, D. C. Babu
and Y. Venkateswarlu, Eur. J. Org. Chem., 2010, 2855–2859.
11 I. Nicolaou and V. J. Demopoulos, J. Med. Chem., 2003, 46, 417–426.
3 For Rh-catalyzed C–C bond cleavage, see: (a) T. Seiser and 12 D. M. Ketcha and G. W. Gribble, J. Org. Chem., 1985, 50, 5451–5457.
N. Cramer, J. Am. Chem. Soc., 2010, 132, 5340–5341; (b) Z. Q. Lei, 13 (a) W. J. Anthony, J. Org. Chem., 1960, 25, 2049–2053; (b) S. C.
H. Li, Y. Li, X. S. Zhang, K. Chen, X. Wang, J. Sun and Z. J. Shi,
Angew. Chem., Int. Ed., 2012, 51, 2690–2694.
Eyley, R. G. Giles and H. Heaney, Tetrahedron Lett., 1985, 26,
4649–4652.
4 For Ru-catalyzed C–C bond cleavage, see: T. Kondo, Y. Kaneko, 14 (a) M. M. Faul and L. L. Winnerroski, Tetrahedron Lett., 1997, 38,
Y. Taguchi, A. Nakamura, T. Okada, M. Shiotsuki, Y. Ura, K. Wada
and T. A. Mitsudo, J. Am. Chem. Soc., 2002, 124, 6824–6825.
4749–4752; (b) Y. H. Ma, J. S. You and F. J. Song, Chem. – Eur. J.,
2013, 19, 1189–1193.
12184 | Chem. Commun., 2014, 50, 12181--12184
This journal is ©The Royal Society of Chemistry 2014