Fused tetracycles with a benzene or cyclohexadiene core: [2 + 2 + 2] Cycloadditions on macrocyclic systems

Article abstract of DOI:10.1039/b806524a

A series of fused tetracycles with a benzene or cyclohexadiene core (2a-h) is satisfactorily prepared by intramolecular [2 + 2 + 2] cycloadditions of triynic and enediynic macrocycles (1a-h) under RhCl(PPh₃)₃ catalysis; the enantioselective cycloaddition of macrocycles 1b and 1e gives chiral tetracycles with moderate enantiomeric excess. The Royal Society of Chemistry.

Full text of DOI:10.1039/b806524a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Fused tetracycles with a benzene or cyclohexadiene core: [2 + 2 + 2]
cycloadditions on macrocyclic systemsw
a
a
a
a
Sandra Brun,a Lı
´
dia Garcia, Ivan Gonzalez, Anna Torrent, Anna Dachs,
´ ´
Anna Pla-Quintana,a Teodor Parellab and Anna Roglans*a
Received (in Cambridge, UK) 17th April 2008, Accepted 30th May 2008
First published as an Advance Article on the web 16th July 2008
DOI: 10.1039/b806524a
A series of fused tetracycles with a benzene or cyclohexadiene
core (2a–h) is satisfactorily prepared by intramolecular
[2 + 2 + 2] cycloadditions of triynic and enediynic macrocycles
(1a–h) under RhCl(PPh3)3 catalysis; the enantioselective cy-
cloaddition of macrocycles 1b and 1e gives chiral tetracycles
with moderate enantiomeric excess.
enediynes bearing different aryl substituents on the sulfonamide
moieties. The same reaction was also efficiently run in non-
conventional media such as molten TBAB.11
Given that the [2 + 2 + 2] cycloisomerization reaction
inside macrocyclic systems to yield fused tetracycles is an
interesting reaction which is still quite unexplored, we describe
here the synthesis of triynic and enediynic triazamacrocycles
and its cycloisomerization reactions. Preliminary results of the
enantioselective versions of the reaction will also be presented.
Macrocycles 1a–h (Fig. 1) were synthesized and completely
characterized by spectroscopic methods12 (see ESI). They all
have a common 1,6,11-tris(arilsulfonyl)-1,6,11-triazaundeca-
3,8-diyne part but have different chains closing the macrocyclic
rings. The macrocycles differ from each other in the ring size
(15-, 16-, 17-membered), the number and sort of unsaturation
(three triple bonds or two triple bonds and one double bond),
and also the substituents present in the double bond or in the
a-position relative to the triple bond. 1a also has different aryl
units compared to the rest of the compounds.13 We then
proceeded to study the various intramolecular cycloisomeriza-
tion reactions. Wilkinson’s catalyst (RhCl(PPh3)3) was selected
as it had formerly given the best results.10 In all cases, cyclo-
isomerized products (Table 1) were obtained in high yields
(from 71% to 99%).14 The [2 + 2 + 2] cycloaddition of
enediynic macrocycles 1b–g proceeds with total stereoselectivity
(Table 1, entries 2–7), and the relative syn/anti stereochemistry
of cycloisomerized compounds 2b–g was determined from NOE
data and from correlated 1H and 13C chemical shifts (see ESIw).
In order to study the scope of the methodology, we chose
different macrocycles (1b–e) whose later cycloisomerization led
to various fused tetracycles such as 5,5,5-, 5,5,6- and 5,6,6-. As a
general trend for enediynic macrocycles (Table 1, entries 2–5),
the formation of 5,6,6-membered rings fused to a cyclohexa-
dienic core (product 2d) is much faster than the formation of
5,5,6-ring system (products 2b and 2c), which at the same time is
faster than the formation of the 5,5,5-tetrafused structure
(product 2e). Although there is a certain tendency to the
formation of bigger rings giving faster reactions, all the macro-
cycles gave fused tetracycles unlike in other methods of synth-
esis, where the failure to construct 5,5,5- is attributed to ring
constraint.5 We have also made an initial study into the effects
of including substituents at different positions of the ring. The
incorporation of a methyl group at the propargylic position (1h)
does not seem to encumber the reaction. Indeed, the yield,
reaction time, and temperature for both the methyl containing
and the non-methyl-containing10 15-membered macrocycles are
The transition-metal-catalyzed [2 + 2 + 2] cycloaddition of
alkynes is an interesting method for synthesizing polysubsti-
tuted benzene derivatives.1 When running this reaction intra-
molecularly, three fused rings are formed in one synthetic
operation. Furthermore, when it is performed in a closed
system, i.e. a macrocycle, the reaction leads to fused tetra-
cycles, which is a highly attractive synthetic strategy.
There are some well-known fused tetracycles with benzene
cores such as trindane which has applications in organometallic
complex stabilization.2 The radioactive technetium complexes
of trindane have also been tested as myocardial imaging agents
in rats.3 Analogous fused tetracycles with a benzene core have
been prepared using routes that afford only symmetrical pro-
ducts.4 Ma et al. accomplished the synthesis of asymmetrically
substituted tetrafused benzenes by triple Mizoroki–Heck5 or
Suzuki–Miyaura6 reactions on hexasubstituted benzenes
although the processes of synthesis are quite laborious. Trian-
nulated benzenes with attached nitrogen functional groups are
also known compounds, particularly benzotripyrrolium ca-
tions, which are used in the synthesis of a number of zeolites,7
and the mellitic triimides, which are used as C3-symmetric
supramolecular building blocks.8 To the best of our knowledge,
fused tetracycles with a cyclohexadiene core have not been
previously described, even though it is likely that they may be
useful as substrates for Diels–Alder reactions.
We have previously reported the [2 + 2 + 2] cycloisomeriza-
tion of 15-membered macrocyclic triynes catalyzed by different
transition metals.9,10 Wilkinson catalyst [RhCl(PPh)3)3] was
found to give the best results and the same catalytic system
was applied to a series of macrocyclic triynes and cis and trans
a Department of Chemistry, Universitat de Girona, Campus de Montilivi,
s/n., E-17071-Girona, Spain. E-mail: anna.roglans@udg.edu;
Fax: 0034 972 418150; Tel: 0034 972 4182 75
b
´
Servei de RMN, Universitat Autonoma de Barcelona, Cerdanyola,
E-08193-Barcelona, Spain. E-mail: teodor.parella@uab.es;
Fax: 0034 93 5812291; Tel: 0034 93 5812291
w Electronic supplementary information (ESI) available: Experimental
procedures and full spectroscopic data for all new compounds. See
DOI: 10.1039/b806524a
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 4339–4341 | 4339

View Article Online
Fig. 1 Macrocycles 1 prepared with their differential chains.
Table 1 [2 + 2 + 2] Cycloaddition reactions of triynic and enediynic macrocycles 1
Entry Substrate T/1C t/h Product Yield (%) Entry Substrate T/1C
t/h Product
Yield (%)
98
1a
1a
90
28
81
5ac
1e
90
24
2b
3b
4b
1b
1c
1d
80
80
60
5
5
4
90
87
98
6a
7a
8b
1f
Reflux 24
95
71
99
1g
Reflux 24
1h
60
24
a
b
c
5 mol% RhCl(PPh3)3 catalyst used. 10 mol% RhCl(PPh3)3 catalyst used. Reaction published elsewhere (see ref. 10).
almost equivalent. Hence, the cycloaddition reaction does not
seem to be affected by the steric hindrance introduced in the
propargylic position. The situation is slightly different for
substituents introduced at the double bond of the enediynic
macrocycles (1f, 1g). If we compare entries 5–7 (Table 1), it is
observed that harsher conditions were required to cycloisome-
rize enediynes containing a phenyl substituent on the double
bond (reflux toluene, entries 6 and 7) as compared to the non-
substituted macrocycle 1e (90 1C, entry 5). In conclusion, the
success of the cycloisomerization reaction may be attributed to
the macrocyclic nature of the reagent, which brings the un-
saturated bonds close together with an orientation that is
favourable to the process.
the next aspect to be evaluated. The intramolecular enantio-
selective cycloaddition of triynes has been reported in just
three cases using Ni,15 Ir,16 and Rh17 catalysts. Furthermore,
two examples of enantioselective intramolecular cycloaddition
of enediynes have been reported,18 one of them including a
case involving macrocycle 1e.10 The study of the enantioselec-
tive version of the cycloaddition reaction was undertaken
starting with macrocycle 1e (Table 2). None of the reactions
using [RhCl(COD)]2 and chiral phosphanes with different
steric hindrances [(S)-BINAP (L1), (S)-(+)-neomenthyldiphe-
nylphosphane (L2) or (2S,3S)-(À)-2,3-bis(diphenylphosphi-
no)butane (L3), Table 2, entries 1–3] led to chirality
induction, but the use of a cationic catalyst [Rh(hpd)L3]ClO4
(hpd = bicyclo[2.2.1]hepta-2,5-diene) with the chiral ligand
already coordinated (i.e. not formed in situ) induced up to a
44% ee, whether or not activated by hydrogen gas (entries 4
and 5). Using the cationic complex and changing toluene for
The [2 + 2 + 2] intramolecular cycloaddition reactions
studied, with the exception of macrocycle 1a, led to tetracyclic
products containing one (2h) or two asymmetric carbons.
Therefore, the enantioselective version of this reaction was
ꢀc
This journal is The Royal Society of Chemistry 2008
4340 | Chem. Commun., 2008, 4339–4341

View Article Online
Table 2 Enantioselective [2 + 2 + 2] cycloaddition reactionsa
Product Yield (%) eeb (%)
Chem., 2005, 4741–4767; P. R. Chopade and J. Louie, Adv. Synth.
Catal., 2006, 348, 2307–2327; V. Gandon, C. Aubert and M.
Malacria, Chem. Commun., 2006, 2209–2217.
Entry Substrate Catalyst/ligand
2 H. K. Gupta, P. E. Lock and M. J. McGlinchey, Organometallics,
1997, 16, 3628–3634.
3 D. W. Wester, J. R. Coveney, D. L. Nosco, M. S. Robbins and R.
T. Dean, J. Med. Chem., 1991, 34, 3284–3290.
4 Z. Li, W.-H. Sun, X. Jin and C. Shao, Synlett, 2001, 1947–1949; C.
1c
2c
3c
4
1e
1e
1e
1e
1e
1e
1e
1e
1b
1b
[Rh(COD)Cl]2/L1
[Rh(COD)Cl]2/L2
[Rh(COD)Cl]2/L3
[Rh(hpd)L3]ClO4
[Rh(hpd)L3]ClO4
[Rh(hpd)L3]ClO4
2e
2e
2e
2e
2e
2e
96
96
98
95
98
97
96
35
46
24
0
0
0
44
43
26
10
12
41
33
5c
6d
7
Lambert, G. Noll, E. Schmalzlin, K. Meerholz and C. Brauchle,
¨
¨
¨
Chem.–Eur. J., 1998, 4, 2129–2135; H.-J. Choi, Y. S. Park, S. H.
Yun, H.-S. Kim, C. S. Cho, K. Ko and K. H. Ahn, Org. Lett.,
2002, 4, 795–798.
5 S. Ma and B. Ni, J. Org. Chem., 2002, 67, 8280–8283.
6 S. Ma, B. Ni, S. Lin and Z. Liang, J. Organomet. Chem., 2005, 690,
5389–5395.
7 S. L. Lawton and W. J. Rohrbaugh, Science, 1990, 247, 1319–1322;
R. K. Harris, J. Parkinson, A. Samadi-Maybodi and W. Smith,
Chem. Commun., 1996, 593–594; R. K. Harris, A. Samadi-May-
bodi and W. Smith, Zeolites, 1997, 19, 147–155; M. J. Sabater and
G. Sastre, Chem. Mater., 2001, 13, 4520–4526.
8 K. A. McMenimen and D. G. Hamilton, J. Am. Chem. Soc., 2001,
123, 6453–6454; L. Y. Park, D. G. Hamilton, E. A. McGehee and
K. A. McMenimen, J. Am. Chem. Soc., 2003, 125, 10586–10590; J.
B. Carroll, M. Gray, K. A. McMenimen, D. G. Hamilton and V.
M. Rotello, Org. Lett., 2003, 5, 3177–3180.
[Rh(COD)2]BF4/L3 2e
[Rh(COD)2]BF4/L3 2e
8d
9
[Rh(hpd)L3]ClO4
[Rh(hpd)L3]ClO4
2b
2b
10e
a
All reactions were performed with 10 mol% of catalyst loading using
b
toluene as the solvent at 65 1C for 24 h. Determined by chiral phase
HPLC (Kromasil TBB; heptane–THF (80 : 20)). Hydrogen gas was
c
introduced to the catalyst solution prior to substrate introduction.
d
e
The reaction was performed with CH2Cl2 heated to reflux. Reac-
tion time 60 h.
CH2Cl2 in order to reduce the temperature gave lower ee
(entry 6). On observing that the cationic catalyst gave better ee
results, we tested [Rh(COD)2]BF4 with one of the three chiral
phosphanes, specifically (2S,3S)-(À)-2,3-bis(diphenylphosphi-
no)butane (L3). The best reaction conditions with the new
catalytic system (toluene at 65 1C) gave only 10% of ee (entry
7). When we used CH2Cl2 as the solvent, a 35% of yield of 2e
with a 12% of ee was obtained after heating to reflux (entry 8).
We then tested macrocycle 1b in the conditions which were
successful for 1e. The cycloisomerized product 2b was ob-
tained with a 41% ee, similar to that obtained for 2e, but in
only a 46% yield (entry 9). In an attempt to increase the yield,
the same reaction was run for a longer reaction time (60 h
instead of 24 h), but more decomposition was obtained giving
a reduced 24% yield and 33% ee (entry 10).
9 A. Pla-Quintana, A. Roglans, A. Torrent, M. Moreno-Manas and
J. Benet-Buchholz, Organometallics, 2004, 23, 2762–2767.
10 A. Torrent, I. Gonzalez, A. Pla-Quintana, A. Roglans, M. Mor-
´
eno-Manas, T. Parella and J. Benet-Buchholz, J. Org. Chem., 2005,
70, 2033–2041.
11 I. Gonzalez, S. Bouquillon, A. Roglans and J. Muzart, Tetrahedron
Lett., 2007, 48, 6425–6428.
´
12 P. Nolis, A. Roglans and T. Parella, J. Magn. Reson., 2005, 173,
305–309; P. Nolis, A. Roglans and T. Parella, Magn. Reson. Chem.,
2005, 43, 979–984.
13 2,4,6-Triisopropylphenyl in 1a is replaced by 4-methylphenyl in
1b–h in order to simplify the 1H NMR spectra in the aliphatic
region.
14 General procedure: In a 50 mL flask, macrocycle 1 (1 equiv) and
RhCl(PPh3)3 (0.1 equiv) were introduced and degassed, and anhy-
drous toluene (10 cm3) was added. The mixture was heated and
stirred under N2 until completion. The solvent was removed and
the residue was purified by column chromatography on silica gel.
In conclusion, a series of fused tetracycles with benzene
cores (5,5,5- and 5,5,6-) and cyclohexadiene cores (5,5,5-,
5,5,6- 5,6,6-) can be conveniently prepared by Rh(I)-catalyzed
[2 + 2 + 2] cycloadditions of macrocyclic systems. When the
reaction was run with a chiral Rh(I) complex, the tetracycles
were obtained in high yields and moderate ee’s.
15 I. G. Stara
Saman, Tetrahedron Lett., 1999, 40, 1993–1996; Z. Alexandrova,
P. Sehnal, I. G. Stara, I. Stary, D. Saman, S. G. Urquhart and E.
Otero, Collect. Czech. Chem. Commun., 2006, 71, 1256–1264; I.
Stary, I. G. Stara, Z. Alexandrova, P. Sehnal, F. Teply, D. Saman
and L. Rulisek, Pure Appl. Chem., 2006, 78, 495–499.
, I. Stary´ , A. Kollarovic, F. Teply´ , S. Vyskocil and D.
´ ´
´
´
´
´
´
16 T. Shibata, K. Tsuchikama and M. Otsuka, Tetrahedron: Asym-
metry, 2006, 17, 614–619; T. Shibata, S. Yoshida, Y. Arai, M.
Otsuka and K. Endo, Tetrahedron, 2008, 64, 821–830.
17 K. Tanaka, H. Sagae, K. Toyoda, K. Noguchi and M. Hirano, J.
Am. Chem. Soc., 2007, 129, 1522–1523; K. Tanaka, H. Sagae, K.
Toyoda and M. Hirano, Tetrahedron, 2008, 64, 831–846.
18 K. Tanaka, G. Nishida, H. Sagae and M. Hirano, Synlett, 2007,
1426–1430. During the continuing study of this project, an en-
antioselective example of Rh-catalyzed intramolecular [2 + 2 + 2]
cycloaddition of macrocycle 1e was reported. Macrocycle 1e was
treated with 10 mol% [Rh(H8-binap)BF4] in dichloromethane at
room temperature to achieve a 73% yield of 2e with an enantio-
meric excess of ca. 20%. See: T. Shibata, H. Kurokawa and K.
Kanda, J. Org. Chem., 2007, 72, 6521–6525.
Financial support from MEC of Spain (CTQ2005-04968,
08797, CTQ2006-01080), ‘‘Generalitat de Catalunya’’
(2005SGR00305, 00238) and UdG (grants to S. B. and I. G.)
is acknowledged. A, R. also thanks Research Technical Ser-
vices of the UdG for analytical data and Johnson and Matthey
for a loan of RhCl3.
Notes and references
1 For recent reviews, see: S. Saito and Y. Yamamoto, Chem. Rev.,
2000, 100, 2901–2916; Y. Yamamoto, Curr. Org. Chem., 2005, 9,
503–519; S. Kotha, E. Brahmachary and K. Lahiri, Eur. J. Org.
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 4339–4341 | 4341