Organocatalytic [10+4] cycloadditions for the synthesis of functionalised benzo[a]azulenes

Article abstract of DOI:10.1039/C8CC08551J

A direct and mild strategy for the synthesis of benzo[a]azulenes based on an organocatalytic [10+4] cycloaddition reaction is described. The strategy enables a diversity-oriented approach for the synthesis of various poly-functionalised azulenes from easily accessible starting materials.

Full text of DOI:10.1039/C8CC08551J

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Organocatalytic [10+4] cycloadditions for the
synthesis of functionalised benzo[a]azulenes†
Cite this: Chem. Commun., 2019,
55, 202
Maxime Giardinetti, Nicolaj Inunnguaq Jessen, Mette Louise Christensen and
Received 25th October 2018,
Accepted 29th November 2018
Karl Anker Jørgensen
*
DOI: 10.1039/c8cc08551j
rsc.li/chemcomm
A direct and mild strategy for the synthesis of benzo[a]azulenes based
on an organocatalytic [10+4] cycloaddition reaction is described. The
strategy enables a diversity-oriented approach for the synthesis of
various poly-functionalised azulenes from easily accessible starting
materials.
Azulenes belong to poly-aromatic hydrocarbons and are com-
posed of a 5-membered ring fused to a 7-membered ring. This
platform has recently gathered attention, as azulene derivatives
show promising biological activities.1–3 Furthermore, these
compounds demonstrate interesting electronic properties,
rendering them attractive as potential materials for various
applications,4–7 and are often coloured compounds important
for imaging,8 dyes9 and indicators.10
Scheme 1 The approach by Hong and Sun to azulenes and indole-fused
azulenes (top) and this work involving organocatalytic reactions of indene-
2-carbaldehydes with chromen-4-ones or a-pyrones (bottom).
Benzo[a]azulenes are a subclass of azulenes based on a
tricyclic benzo-fused azulene moiety. The first synthesis of
benzo[a]azulenes dates back to 1947 and involves vapour of azulenes and indole-fused azulenes from fulveneketene
dehydrogenation of an already built tricyclic core.11 Since acetal and a-pyrones (Scheme 1, top).19 They later developed
then, several approaches for the synthesis of various types a variant of this reaction using 6-aminofulvenes.20 Recently, we
of benzo[a]azulenes have been reported. Some rely on the have developed an enantioselective organocatalytic [10+4]
construction of the 5-membered ring by metal catalysis12 cycloaddition based on the reaction of indene-2-carbaldehydes with
or by cascade reactions from the tropylium cation.13 The electron-deficient dienes affording a 7-membered ring in tricyclic
6-membered ring of benzo[a]azulenes can also be constructed dihydrobenzo[a]azulenes.21 Inspired by these results, we have been
via Knoevenagel condensation,14 whereas the 7-membered ring seeking a direct approach towards functionalised aromatic
can be obtained by ring expansion of a benzene ring using benzo[a]azulenes from easily accessible substrates. In the present
carbene chemistry.15 In a more direct strategy, all three cycles work, we report the first organocatalytic [10+4] cycloaddition
have been constructed in a Pt-catalysed cascade sequence.16 reaction between indene-2-carbaldehydes and chromen-4-one
However, all these routes involve complex starting materials.
Cycloaddition reactions provide an easy access to cyclic substituted azulene derivatives (Scheme 1, bottom).
compounds. 6-Membered rings can be accessed by the Diels– The investigations were initiated by reacting 3-phenylindene-
derivatives or a-pyrones enabling the mild synthesis of poly-
Alder reaction involving 6p-electrons,17 while higher-order cyclo- 2-carbaldehyde 1a with ethyl (E)-3-(4-oxo-4H-chromen-3-yl)-
additions (46p-electrons) have demonstrated the potential for acrylate 2a in the presence of 3a as a catalyst, applying the
the construction of larger cyclic compounds.18 Hong and Sun reported conditions for the enantioselective [10+4] cyclo-
have described a [6+4] cycloaddition strategy for the synthesis addition.21 We were pleased to observe an elimination reaction
following the [10+4] cycloaddition, thus rendering traces of
the desired aromatic benzo[a]azulene 4aa (Table 1, entry 1).
Department of Chemistry, Aarhus University Langelandsgade 140, 8000 Aarhus C,
By extending the reaction time to 4 d, the conversion reached
Denmark. E-mail: kaj@chem.au.dk
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c8cc08551j only 50% and compound 4aa was isolated in 15% yield (entry 2).
202 | Chem. Commun., 2019, 55, 202--205
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Communication
ChemComm
Table
1
Optimisation of the reaction between 3-phenylindene-2-
carbaldehyde 1a and ethyl (E)-3-(4-oxo-4H-chromen-3-yl)acrylate 2aa
Entry
Catalyst 3
Temperature
Time (d)
Yieldb (%)
1
3a
3a
3b
3b
3c
3c
RT
RT
RT
40 1C
40 1C
40 1C
1
4
3
2
1
1
Trace
15
69
70
75
2
3c
4c
5
Scheme 2 Scope of the organocatalytic reaction between indene-
2-carbaldehydes 1 and chromen-4-ones 2. Reaction conditions: 1 (0.1 mmol),
2 (0.15 mmol), 3c (15 mol%), pMeOBzOH (15 mol%), CDCl3 (0.4 mL, filtered
over Al2O3), MS 4 Å, 40 1C. Isolated yields.
6d
65
a
Reaction conditions: 1a (0.075 mmol), 2a (0.05 mmol), 3 (15 mol%),
pMeOBzOH (15 mol%), CDCl3 (filtered over Al2O3, 0.2 mL), molecular
b
c
d
sieves (MS) 4 Å. Isolated yields. No acid additive was used. In this
Encouraged by these results, we decided to expand the
potential of the reaction concept to also include ethyl 2-oxo-
2H-pyran-5-carboxylate 5a in a reaction with 1l. In the develop-
ment of the reaction conditions, the reaction was allowed
to reach full conversion before isolation of compound 6la.
We started to screen the reaction conditions with TMS-
protected catalyst 3d in CDCl3 at room temperature. After 3 d,
the expected benzo[a]azulene 6la was obtained in 49% yield
(Table 2, entry 1). We then considered using pyrrolidine as a
catalyst (3c), which then rendered 6la in 52% yield (entry 2).
The use of pMeOBzOH as an additive furnished the product in
42% yield after 1 d (entry 3). We then hypothesised that the
bifunctional catalyst 3b could help increase the reaction rate
case: 1a (0.05 mmol), 2a (0.075 mmol).
Using a bifunctional thiourea catalyst 3b afforded 4aa in 69%
yield after 3 d (entry 3). Similar results were obtained when the
reaction mixture was stirred at 40 1C for 2 d (entry 4). Since
compounds 4 are achiral, we decided to test pyrrolidine 3c as
a simpler catalyst, and 4aa was produced in 1 d at 40 1C in
75% yield (entry 5). Inverting the 1a/2a ratio to 1/1.5 eased the
purification affording 4aa in 65% yield (entry 6). Although the yield
was slightly lower, we retained these conditions for the scope.
We then proceeded to study the scope of this reaction,
utilising various indene-2-carbaldehydes 1 (Scheme 2). Intro-
duction of an electron-rich substituent, as well as an electron-
poor substituent, gave rise to the desired azulenes (4ba, 4ca) in
modest yields, whereas a 4-methyl group in 1 provided 4da in
83% yield. The tetracyclic compound 4ea was produced in high
yield. In comparison, product 4fa, obtained from a more sterically
hindered starting compound (next to the reactive centre), was
formed in 42% yield after 2 d. The presence of other aryl and
hetero-aryl substituents (4ga–4ja) at the 3-position of indene-2-
carbaldehyde was well-tolerated, and a tert-butyl substituent (4ka),
although in this case, a longer reaction time was required. On the
other hand, when no 3-substituent in 1 was present, the desired
azulene (4la) could only be obtained in low yield. We then
subjected several chromen-4-one derivatives 2 to the reaction
conditions. A large variety of electron-withdrawing groups provided
the desired azulenes 4ab–4ae in high yields. Electron-poor substi-
tuents on the benzene ring of chromen-4-one were also viable,
delivering the azulene products 4af and 4ag in good yields. However,
methoxy-substituted 4ah was obtained in only 19% yield.
Table 2 Optimisation of the reaction between indene-2-carbaldehyde 1l
and ethyl 2-oxo-2H-pyran-5-carboxylate 5aa
Entry
Catalyst 3
Time (d)
Additive
Yieldb (%)
1
2
3
4
5
3d
3c
3c
3b
3b
3
3
1
3
6
—
—
49
52
42
41c
64
pMeOBzOH
—
NaOAc
The products obtained are highly colourful and in the ESI†
the UV spectra and data of representative compounds (4aa, 4ba,
4ca (Scheme 2) and 6la (vide infra)) are shown.
a
Reaction conditions: 1l (0.05 mmol), 5a (0.075 mmol), 3 (15 mol%),
additive (15 mol%), CDCl3 (0.2 mL). b Isolated yields. NMR conversion.
c
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 202--205 | 203

View Article Online
ChemComm
Communication
and by employing this catalyst, 6la was obtained in 41% yield 1o and 1p were probed, but a higher 6pa : 6oa ratio of 7.7 : 1 was
after 3 d (entry 4). Finally, the addition of NaOAc as a base obtained. The introduction of an aryl group at position-3 of
additive enabled us to reach 64% yield over 6 d (entry 5).
the indene 1 impaired the reactivity and no product could be
These conditions were then applied to several indene-2- isolated in a significant amount. A study of the scope of
carbaldehydes 1 (Scheme 3). In general, the yields dropped, a-pyrones 5 was attempted but it turned out to be difficult
independent of the electronic effects of the substituents. When to synthesise the variants of 5a bearing another electron-
indene-2-carbaldehydes 1m and 1n were employed, in both withdrawing group at position-5 or a substituent at position-3.
cases, isomerisation of the pure aldehyde towards a 1.9 : 1 The introduction of aryl groups at either position-4 or position-6
1n : 1m mixture occurred, leading to an inseparable mixture of 5a proved to be unsuccessful, as no reactivity was observed
of the respective products 6ma and 6na with the same ratio. when such substrates were employed.
A similar pattern was observed when indene-2-carbaldehydes
The first step in the mechanism for the formation of
azulenes is expected to be the generation of an electron-rich
10p-intermediate I arising from the condensation of the pyrro-
lidine moiety of catalyst 3 with the indene-2-carbaldehyde 1
(Scheme 4). Based on mechanistic studies,21 it is proposed that
the [10+4] cycloaddition occurs as a step-wise process. In the
case of pathway A, involving chromen-4-ones, we surmise that
product 4 is formed via intermediate IIA, involving deprotona-
tion/elimination of the phenolate moiety. When a-pyrone 5a
is used, we propose that the [10+4] cycloaddition produces
intermediate IIB.19 Deprotonation of this intermediate conco-
mitantly releases the catalyst, CO2 and the desired aromatic
structure 6la. In the present stage, we propose that the potential
role of the thiourea moiety of 3b is to activate and direct the
electrophile which is possibly also involved in the CO2 elimina-
tion process.
Scheme 3 Scope of the organocatalytic reaction between indene-2-
carbaldehydes 1 and ethyl 2-oxo-2H-pyran-5-carboxylate 5a. Reaction
conditions: 1 (0.1 mmol), 2 (0.15 mmol), 3b (15 mol%), NaOAc (15 mol%),
CDCl3 (0.4 mL), RT for 6–7 d. Isolated yields for the mixture of
regioisomers.
In summary, a new organocatalytic strategy based on a [10+4]
cycloaddition reaction for the synthesis of benzo[a]azulenes has
been developed. It allows for the formation of poly-substituted
azulenes in low to high yields under mild conditions from simple
and easily accessible substrates.
The authors are grateful for financial support from the
Carlsberg foundation Semper Ardens programme, Aarhus
University and DFF.
Conflicts of interest
The authors declare no conflict of interest.
Notes and references
1 T. Wada, R. Maruyama, Y. Irie, M. Hashimoto, H. Wakabayashi,
N. Okudaira, Y. Uesawa, H. Kagaya and H. Sakagami, In Vivo, 2018,
32, 479.
2 T. O. Leino, A. Turku, J. Yli-Kauhaluoma, J. P. Kukkonen, H. Xhaard
´
and E. A. A. Wallen, Eur. J. Med. Chem., 2018, 157, 88.
´
´
3 L. P. P. Cano, R. Q. Manfredi, M. Perez, M. Garcia, G. Blustein,
´
R. Cordeiro, C. D. Perez, L. Schejter and J. A. Palermo, Chem.
Biodiversity, 2018, 15, e1700425.
4 H. Xin, C. Ge, X. Jiao, X. Yang, K. Rundel, C. R. McNeill and X. Gao,
Angew. Chem., Int. Ed., 2018, 57, 1322.
5 Y. Yamaguchi and H. Katagiri, J. Synth. Org. Chem., 2018, 76, 820.
6 G.-L. Arnold, I.-G. Lazar, G.-O. Buica, E.-M. Ungureanu and
L. Birzan, U. P. B. Sci. Bull., Series B, 2017, 79, 89.
`
7 J. Hieulle, E. Caronell-Sanroma, M. Vilas-Varela, A. Garcia-Lekue,
˜
E. Pena and J. I. Pascual, Nano Lett., 2018, 18, 418.
8 Y. Zhou, Y. Zhuang, X. Li, H. Ågren, L. Yu, J. Ding and L. Zhu,
Chem. – Eur. J., 2017, 23, 7642.
9 P. Cowper, A. Pockett, G. Kociok-Kohn, P. J. Cameron and
S. E. Lewis, Tetrahedron, 2018, 74, 2775.
Scheme 4 Mechanistic proposal for the formation of benzo[a]azulenes 4
and 6la via [10+4] cycloaddition reaction pathways.
¨
204 | Chem. Commun., 2019, 55, 202--205
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Communication
ChemComm
10 C. M. Lopez-Alled, A. Sanchez-Fernandez, K. J. Edler, A. C. Sedgwick, 16 T. Matsuda, T. Goya, L. Liu, Y. Sakurai, S. Watanuki, N. Ishida and
¨
S. D. Bull, C. L. McMullin, G. Kociok-Kohn, T. D. James, J. Wenk and
M. Murakami, Angew. Chem., Int. Ed., 2013, 52, 6492.
S. E. Lewis, Chem. Commun., 2017, 53, 12580.
17 O. Diels and K. Alder, Justus Liebigs Ann. Chem., 1928, 460, 98.
11 D. H. S. Horn, J. R. Nunn and W. S. Rapson, Nature, 1947, 160, 829. 18 R. Mose, G. Preegel, J. Larsen, S. Jakobsen, E. H. Iversen and
12 D. Sperandio and H.-J. Hansen, Helv. Chim. Acta, 1995, 78, 765. K. A. Jørgensen, Nat. Chem., 2017, 9, 487.
13 K. Yamamura, S. Kawabata, T. Kimura, K. Eda and M. Hashimoto, 19 B.-C. Hong and S.-S. Sun, Chem. Commun., 1996, 937.
J. Org. Chem., 2005, 70, 8902.
14 C.-P. Wu, B. Devendar, H.-C. Su, Y.-H. Chang and C.-K. Ku, Tetrahedron
Lett., 2012, 53, 5019.
20 B.-C. Hong, Y.-F. Jiang and E. S. Kumar, Bioorg. Med. Chem. Lett.,
2001, 11, 1981.
21 B. S. Donslund, N. I. Jessen, G. Bertuzzi, M. Giardinetti, T. A. Palazzo, M. L.
Christensen and K. A. Jørgensen, Angew. Chem., Int. Ed., 2018, 57, 13182.
15 C. Wentrup and J. Becker, J. Am. Chem. Soc., 1984, 106, 3705.
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 202--205 | 205