Synthesis of Polyheterocyclic Tropones by [2 + 2 + 2 + 1] Carbonylative Cycloaddition of Triynes

Article abstract of DOI:10.1021/acs.orglett.8b01496

A direct synthesis of tropones (2,4,6-cycloheptatrienes) from simple preorganized triynes has been elaborated. This simple rhodium-catalyzed domino strategy allows one-pot access to fully substituted tropones in a 6-6-7-5 tetracyclic core in average to hi

Full text of DOI:10.1021/acs.orglett.8b01496

Letter
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
pubs.acs.org/OrgLett
Synthesis of Polyheterocyclic Tropones by [2 + 2 + 2 + 1]
Carbonylative Cycloaddition of Triynes
̈
Laura Salacz, Nicolas Girard, Gaelle Blond,* and Jean Suffert*
Universite
́
de Strasbourg, CNRS, LIT UMR 7200, F-67000 Strasbourg, France
*
S Supporting Information
ABSTRACT: A direct synthesis of tropones (2,4,6-cyclo-
heptatrienes) from simple preorganized triynes has been
elaborated. This simple rhodium-catalyzed domino strategy
allows one-pot access to fully substituted tropones in a 6−6−
7
−5 tetracyclic core in average to high yields.
2
4−26
aining access to medium-sized carbocycles has been a
long-standing challenge for synthetic chemists, and it is
cycloheptatrienes from enediynes,
but led to ambivalent
27
G
results with modest documentation in the case of triynes. We
have therefore endeavored to devise an efficient method to
synthesize tropones from triynes and carbon monoxide in the
presence of a rhodium catalyst. In order to increase yields as well
as molecular complexity, we have chosen to use a cyclohexene
core bearing tethered alkynes. By the strategic placement of the
triynes around the core and the tethers, we could impose the
preorganization of the system in order to facilitate the successive
cyclization reactions. Herein, we report a mild [2 + 2 + 2 + 1]
carbonylative cycloaddition, giving access to unprecedented
polyheterocyclic scaffolds of high molecular complexity
containing a fully substituted tropone ring through the closure
of three cycles in a single step (Scheme 1).
well established that the formation of these cycles poses a
hardship due to entropic and enthalpic factors and transannular
interactions. In the past decades, many new methodologies have
been developed in order to bypass these difficulties. One
approach that has given good results is the use of carbonylative
1
cycloaddition. Through the presence of a carbonyl function,
medium-sized carbocycles formed by carbonylative cyclo-
addition grant access to a wide panel of functionalities, thereby
allowing access to new scaffolds.
The synthesis of fused medium-sized carbocycles by metal-
catalyzed domino reactions to access high molecular complexity
2
−5
has been a continued interest in our group. Part of our efforts
has been employed toward the synthesis of cycloheptatrienes,
which we recently achieved using palladium catalysis, albeit with
Scheme 1. [2 + 2 + 2 + 1] Synthesis of Tropones
6
moderate results. With the idea to further investigate the
possibilities in the field of seven-membered rings, we have
become interested in a different type of compounds: 2,4,6-
cycloheptatrieneones. These compounds, more commonly
referred to as tropones, are nonbenzenoid aromatic seven-
membered rings that have attracted considerable interest due to
7
original electronic properties and their presence in several
The major challenge for the formation of carbonylated
compounds with the [2 + 2 + 2 + 1] carbonylative cycloaddition
from triynes is the competition with the well-established
rhodium-catalyzed [2 + 2 + 2] cycloaddition of triynes, which
8
natural products. Moreover, tropones have often been used as
9
−16
building blocks in cycloadditions.
With this in mind, a
number of syntheses of the tropone ring have been devised, and
28
1
7−19
leads to the stable aromatic benzenoid compound. The main
goal for our optimization process was therefore to find
conditions which allowed the synthesis of the tropone as a
major product. Based on preliminary results in our group, we
decided to optimize the reaction using model substrate 1a.
Triyne 1a was obtained from 2-cyclohexenone using a six-step
synthesis with 35% overall yield (Scheme 2). 2-Iodocyclohex-2-
enone 4 was obtained from 2-cyclohexenone using a described
among those, cycloaddition reactions such as [5 + 2],
[3 +
2
0
21−23
2
], and [4 + 3]
cycloadditions are well represented.
However, these syntheses require one or several transformations
following the key step to access the desired tropone ring.
Moreover, surprisingly, these strategies did not make use of the
presence of the carbonyl function in tropones for their synthesis.
Attracted by the scarcity of efficient carbonylative strategies to
access tropones, we have imagined that three alkyne moieties in
the presence of carbon monoxide and a suitable catalyst could
allow the formation of tropones via a [2 + 2 + 2 + 1]
cycloaddition reaction. The [2 + 2 + 2 + 1] carbocyclization of
enediynes has given remarkable results in the synthesis of 1,3-
29
procedure. It then underwent a Sonogashira reaction to yield
Received: May 11, 2018
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b01496
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 2. Typical Synthesis of Substrates 1
when the reaction was conducted under atmospheric pressure of
CO, the isolated yield of tropone decreased (entry 4), but no
notable effect was observed when the pressure was increased
from 2 to 4 atm (entry 5). Further shortening of the reaction
time to 1 h maintained both conversion and selectivity, allowing
us to obtain the fully substituted tropone in 68% yield (entry 6).
The influence of the metal ligands was then studied. Using a
bidentate phosphine ligand not only slowed the reaction
considerably but also lowered the ratio of carbonyl insertion
(
entry 7). When a monodentate phosphine ligand was used in
the shape of Wilkinson’s catalyst, the major product was
benzenoid compound 3a, and the conversion only attained 24%
(
entry 8). This can be slightly improved by the use of a cationic
catalyst, i.e. adding silver triflate to the mixture, affording an 87%
conversion. However, selectivity remained less favorable than
3
1,32
that obtained with our original catalyst (entry 9).
enyne 5. The ketone was reduced using dibutylaluminum
hydride to yield enynol 6. The alcohol was subsequently
propargylated in biphasic media using concentrated sodium
hydroxide and propargyl bromide. The terminal alkyne thus
obtained was then converted to propargyl alcohol 8 by
deprotonating the terminal alkyne using ethylmagnesium
bromide and allowing the formed species to react with
paraformaldehyde at 50 °C over 1 day. Triyne 1a was then
obtained by performing a second propargylation in the same
conditions as previously described. With this substrate in hand,
we undertook the process of optimizing the reaction conditions
for the synthesis of tropone 2a (see Supporting Information).
We were pleased to find that, under 2 atm of CO in the
Ruthenium and cobalt are other metals that are associated
with carbonylative cycloaddition reactions, but control experi-
ments using their carbonyl complexes gave low to no conversion
33,34
(
entries 10 and 11).
Noteworthy, we observed during this
study that the triyne substrates 1 are poorly stable and needed to
be freshly purified prior to cycloaddition experiments to avoid
yield depletion.
We have proposed a mechanism based on the results of a
theoretical study of the rhodium-catalyzed [2 + 2 + 2 + 1]
35
carbonylative cycloaddition of enediynes (Scheme 3). We
propose that the cyclooctadiene ligand is easily decoordinated
and replaced by carbon monoxide. The L′ ligand, on the other
hand, can either be μ-Cl[Rh(COD)] or be replaced by CO. L′ is
then displaced by the alkynes that are ideally preorganized due
to the presence of the cyclohexene ring, to obtain intermediate
A. A [2 + 2 + M] cycloaddition occurs toward five-membered
rhodacycle B. Henceforth, two pathways are possible, the
limiting factor being the kinetics of the chelation of the third
alkyne. If this step is slow with respect to the migratory insertion
of carbon monoxide from the rhodium to the C−Rh bond, the
carbonylated six-membered rhodacyclic intermediate C is
obtained. Here, the triple bond chelates the rhodium and is
inserted in the cycle to obtain eight-membered intermediate D.
The reductive elimination of rhodium gives tropone 2. However,
to some extent, the chelation of the rhodium by the third alkyne
presence of [Rh(COD)Cl] , the triyne 1a underwent cyclization
2
at room temperature with definite selectivity in favor of the
carbonylated product 2a (Table 1, entry 1), whereas using
[
(
RhCl(CO) ] lowered the selectivity of the reaction slightly
entry 2). As the cyclization was very sluggish at room
2 2
temperature with both catalysts, and long reaction times
afforded only poor conversion, we increased the temperature
to reflux of DCE, which afforded complete conversion within 3 h
and yielded 68% of the desired isolated tropone (entry 3).
30
Seminal experiments in our group and data in the literature
indicated that increasing the pressure of carbon monoxide too
much (10 atm) seemed to hinder the catalysis, leading us to
conduct this study preferably using low pressures. However,
Table 1. Optimization Reaction
b
b
a
a
cat./L or add.
time (h)
t (°C)
conv (%)
2a/3a
2a (%)
3a (%)
1
2
3
4
5
6
7
8
9
1
1
[Rh(COD)Cl] /−
40
23
3
3
3
1
3.5
1
1
1
rt
rt
65
35
70/30
60/40
67/33
65/35
68/32
68/32
60/40
41/59
55/45
45/55
0/100
40
nd
68
58
66
68
46
nd
nd
nd
−
18
nd
30
35
34
30
27
nd
nd
nd
nd
2
[Rh(CO) Cl] /−
2
2
[Rh(COD)Cl] /−
84
84
84
84
84
84
84
84
84
100
100
100
100
90
24
87
78
<10
2
c
[Rh(COD)Cl] /−
2
d
[Rh(COD)Cl] /−
2
[Rh(COD)Cl] /−
2
[Rh(COD)Cl] /dppp
2
e
RhCl(PPh ) /−
3
3
e
RhCl(PPh )/AgOTf
3
0
1
Co (CO) /−
2
8
Ru (CO) /−
1
3
12
a
b
1
c
d
e
Isolated yields. H NMR ratio. P = 1 atm. P = 4 atm. 10 mol % cat. was used.
CO
CO
B
DOI: 10.1021/acs.orglett.8b01496
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Possible Mechanism
Scheme 4. Scope of the Reaction
can be faster than the migratory insertion. In this case, from B we
obtain chelated intermediate C′, followed by insertion of the
triple bond yielding seven-membered rhodacyclic intermediate
D′. Final reductive elimination then produces benzenoid
compound 3.
With this proposed mechanism in hand, we have investigated
2
the scope of the reaction (Scheme 4). When R is a different
alkyl chain, the yields are similar (2a/2c); however, in the
presence of a terminal alkyne the yield lowers to 53% (2b). In
order to investigate the consequences of electronic effects on the
triple bond, we synthesized a number of substrates possessing
differently substituted aryls using a Sonogashira reaction on 1b
(1e−1k). We hoped that a visible effect on the selectivity would
be seen depending on the donating or withdrawing quality of the
aryl substituent, which would favor or disfavor the rapid
chelation of the third alkyne. However, when substrates 1e−1k
were exposed to the reaction conditions, the efficiency of the
desired [2 + 2 + 2 + 1] reaction lowered. Moreover, no definite
effect of the substituent could be observed. We hypothesize that
rather than the electronic effects of the substituents, it is the
steric hindrance of aryl groups at this position which has the
strongest repercussions and interferes with the migratory
insertion of carbon monoxide during the catalytic cycle (B to
C or C′). Our efforts to synthesize a substrate possessing an ester
the reaction produces 2o as a major compound in 47% yield with
an alkyl R chain, but with a phenyl terminal group, the
2
formation of tropone remarkably decreases and the desired
compound becomes a minority product (2p). Inversely, when
the nitrogen was in the Y position, full conversion was achieved,
and the yield was restored to a reasonable 48% (2q).
Good results were also obtained when a geminal diester was
used either as Y or both X and Y tethers, in which case full
conversion was achieved and good yields were obtained for the
formation of tropones 2r and 2s. In all cases the sole byproducts
were the benzenoid compounds 3, and full conversion was
achieved in 1 h for a vast majority of the substrates (see
Supporting Information for the yields and structures of the
byproducts).
2
as an electron-withdrawing group at the R position to withdraw
the steric factor have unfortunately been unsuccessful.
1
Surprisingly, in the presence of a siloxyalkyl R chain, an aryl
2
at the R position gave higher conversion and yield than an alkyl
(2l/2m). However, the carbonylative cycloaddition of the
diphenyl substrate 1n was again disfavored by the high steric
hindrance. This goes to confirm that steric hindrance, rather
than favoring the formation of the least hindered species
When the tether was extended by one carbon, however,
conversion was extremely low and no tropone was observed
(Scheme 5), and most of the starting material could be
recovered. Such results have already been observed, notably in
our group during cyclocarbopalladation reactions. This
indicates that the extended distance between the rhodium in
intermediate B or C and the third alkyne likely hampers the
chelation of the latter to the metal, leading to a very sluggish
(
nonbenzenoid compound 2), inhibits the CO insertion and
3
6
consequently the formation of the desired tropone, leading to an
increased amount of benzenoid compound 3. To study the
possibility of enriching the heterocyclic scaffolds, we have
substituted the ether tethers by various atoms. When X = NTs,
C
DOI: 10.1021/acs.orglett.8b01496
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(
7) Sugiyasu, K.; Song, C.; Swager, T. M. Macromolecules 2006, 39,
Scheme 5. Effect of a Six-Membered Tether
5
598−5600.
8) Liu, N.; Song, W.; Schienebeck, C. M.; Zhang, M.; Tang, W.
Tetrahedron 2014, 70, 9281−9305.
9) Kumar, P.; Thakur, A.; Hong, X.; Houk, K. N.; Louie, J. J. Am.
(
(
Chem. Soc. 2014, 136, 17844−17851.
(
(
10) Jia, S.; Su, S.; Li, C.; Jia, X.; Li, J. Org. Lett. 2014, 16, 5604−5607.
́
11) Rivero, A. R.; Fernandez, I.; Sierra, M. A. Org. Lett. 2013, 15,
́
4
(
928−4931.
12) Liu, H.; Wu, Y.; Zhao, Y.; Li, Z.; Zhang, L.; Yang, W.; Jiang, H.;
Jing, C.; Yu, H.; Wang, B.; et al. J. Am. Chem. Soc. 2014, 136, 2625−
629.
reaction. As both 2t and 3t compounds are aromatic, we can
speculate that they have a similar stability, making it complicated
to rationalize the sole formation of the benzenoid compound
without extensive calculations.
2
(13) Mose, R.; Preegel, G.; Larsen, J.; Jakobsen, S.; Iversen, E. H.;
Jørgensen, K. A. Nat. Chem. 2016, 9, 487−492.
(14) Wu, Y.; Liu, H.; Zhang, L.; Sun, Z.; Xiao, Y.; Huang, J.; Wang, M.;
Guo, H. RSC Adv. 2016, 6, 73547−73550.
In summary, we have devised a new method for the one-pot
synthesis of fully substituted polyheterocyclic tropones via the
rhodium-catalyzed [2 + 2 + 2 + 1] carbonylative cycloaddition.
The use of preorganized triynes allows reaching good yields with
respect to the high molecular complexity of the formed
compounds. Indeed, in a single step, four C−C bonds and
three cycles are formed. It is, to the best of our knowledge, one of
the first syntheses of highly substituted tropones using a one-pot
(
(
(
15) Nair, V.; Abhilash, K. G. Synlett 2008, 2008, 301−312.
16) Li, P.; Yamamoto, H. J. Am. Chem. Soc. 2009, 131, 16628−16629.
17) Song, W.; Xi, B. M.; Yang, K.; Tang, W. Tetrahedron 2015, 71,
5
(
1
(
979−5984.
18) Zhang, M.; Liu, N.; Tang, W. J. Am. Chem. Soc. 2013, 135,
2434−12438.
19) Meck, C.; Mohd, N.; Murelli, R. P. Org. Lett. 2012, 14, 5988−
5991.
1
8,37−39
reaction.
The heteroatoms can be modified, and many
(20) Zhang, H. J.; Hu, L.; Ma, Z.; Li, R.; Zhang, Z.; Tao, C.; Cheng, B.;
Li, Y.; Wang, H.; Zhai, H. Angew. Chem., Int. Ed. 2016, 55, 11638−
functionalities are tolerated by the reaction conditions, which
are mild and require only a low pressure of carbon monoxide.
1
1641.
(
21) Noyori, R.; Makino, S.; Okita, T.; Hayakawa, Y. J. Org. Chem.
975, 40, 806−807.
22) Murray, D. H.; Albizati, K. F. Tetrahedron Lett. 1990, 31, 4109−
112.
1
(
4
ASSOCIATED CONTENT
Supporting Information
■
*
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.8b01496.
(23) Davies, H. M. L.; Clark, T. J. Tetrahedron 1994, 50, 9883−9892.
(24) Bennacer, B.; Fujiwara, M.; Ojima, I. Org. Lett. 2004, 6, 3589−
3
(
2
(
591.
Procedures for the synthesis of substrates and final
25) Bennacer, B.; Fujiwara, M.; Lee, S.; Ojima, I. J. Am. Chem. Soc.
005, 127, 17756−17767.
26) Kaloko, J. J.; Teng, Y. H. G.; Ojima, I. Chem. Commun. 2009, 30,
1
13
products and characterization, H and C NMR spectras
of isolated compounds (PDF)
4
569−4571.
AUTHOR INFORMATION
Corresponding Authors
(27) Teng, Y. H. G. Synthesis of Novel Fused Tropones and Colchicinoids
Through Rh(I)-Catalyzed [2 + 2 + 2 + 1] Cycloaddition of Triynes with
Carbon Monoxide, Ph.D. Thesis, 2011, Stony Brook University.
■
*
E-mail: jean.suffert@unistra.fr.
E-mail: gaelle.blond@unistra.fr.
(
28) Shibata, Y.; Tanaka, K. Synthesis 2012, 44, 323−350.
*
(29) Liu, H.; Yan, P.; Li, Y.; Liu, J.; Sun, Q.; Wang, X.; Wang, C.
Monatsh. Chem. 2012, 143, 1055−1059.
ORCID
(
(
30) Wang, Y.; Yu, Z. Acc. Chem. Res. 2015, 48, 2288−2296.
Nicolas Girard: 0000-0003-4610-9872
31) Wender, P. A.; Croatt, M. P.; Deschamps, N. M. J. Am. Chem. Soc.
Gael
Notes
̈
le Blond: 0000-0001-9144-0295
2004, 126, 5948−5949.
(
32) Grillet, F.; Brummond, K. M. J. Org. Chem. 2013, 78, 3737−3754.
(33) Fukuyama, T.; Higashibeppu, Y.; Yamaura, R.; Ryu, I. Org. Lett.
The authors declare no competing financial interest.
2
(
007, 9, 587−590.
34) Son, S. U.; Yoon, Y. A.; Choi, D. S.; Park, J. K.; Kim, B. M.;
ACKNOWLEDGMENTS
■
Chung, Y. K. Org. Lett. 2001, 3, 1065−1067.
The authors gratefully acknowledge the support of the
University of Strasbourg and the MNERT (L.S.) for a
fellowship.
(35) Montero-Campillo, M. M.; Rodrigez-Otero, J.; Cabaleiro-Lago,
E. J. Phys. Chem. A 2008, 112, 2423−2427.
(36) Blouin, S.; Pertschi, R.; Suffert, J.; Blond, G. Adv. Synth. Catal.
2
(
018, 360, 2166.
37) Nair, V.; Sethumadhavan, D.; Nair, S. M.; Rath, N. P.; Eigendorf,
G. K. J. Org. Chem. 2002, 67, 7533−7536.
38) Gao, W.; Xing, X.; Li, Y.; Lan, S. Tetrahedron 2014, 70, 2180−
189.
39) Zhao, J.; Liu, J.; Xie, X.; Li, S.; Liu, Y. Org. Lett. 2015, 17, 5926−
929.
REFERENCES
■
(
2
(
(
1
1) Salacz, L.; Girard, N.; Suffert, J.; Blond, G. Monatsh. Chem. 2018,
49, 671−686.
(
2
(
5
2) Blouin, S.; Gandon, V.; Blond, G.; Suffert, J. Angew. Chem., Int. Ed.
016, 55, 7208−7211.
3) Charpenay, M.; Boudhar, A.; Blond, G.; Suffert, J. Angew. Chem.,
Int. Ed. 2012, 51, 4379−4382.
4) Petrignet, J.; Boudhar, A.; Blond, G.; Suffert, J. Angew. Chem., Int.
Ed. 2011, 50, 3285−3289.
5) Hulot, C.; Blond, G.; Suffert, J. J. Am. Chem. Soc. 2008, 130, 5046−
047.
6) Joussot, J.; Schoenfelder, A.; Suffert, J.; Blond, G. C. R. Chim. 2017,
0, 665−681.
(
(
5
(
2
D
DOI: 10.1021/acs.orglett.8b01496
Org. Lett. XXXX, XXX, XXX−XXX