Synthesis of fused multicyclic compounds containing macrocycles by dienyne ring-closing metathesis and Diels-Alder reactions

Article abstract of DOI:10.1039/c1ob05683b

Fused bicyclic compounds comprising small and large rings were synthesised by dienyne ring-closing metathesis (RCM) using Grubbs' catalyst. By taking advantage of faster small ring cyclisation compared with macrocyclisation, single isomers were obtained r

Full text of DOI:10.1039/c1ob05683b

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 5871
www.rsc.org/obc
PAPER
Synthesis of fused multicyclic compounds containing macrocycles by dienyne
ring-closing metathesis and Diels–Alder reactions†
Hyeon Park and Tae-Lim Choi*
Received 2nd May 2011, Accepted 26th May 2011
DOI: 10.1039/c1ob05683b
Fused bicyclic compounds comprising small and large rings were synthesised by dienyne ring-closing
metathesis (RCM) using Grubbs’ catalyst. By taking advantage of faster small ring cyclisation
compared with macrocyclisation, single isomers were obtained rather than mixtures of two isomers
with different ring sizes. Using this process, various fused bicyclic compounds comprising small rings
(5–7- membered) and large rings (14–17- membered) were obtained. By increasing reaction temperature
and catalyst loading, the product conversion was improved in a predicted manner. This method
produced E-olefins on the macrocycles with high selectivity. Also, the selectivity issues of tandem RCM
for the synthesis of fused bicyclic compounds comprising small and medium rings were investigated.
Lastly, the prepared bicyclic compounds with small and large rings contained 1, 3-dienes that
underwent a further modification reaction, such as Diels–Alder, to produce more complex compounds.
These Diels–Alder reactions produced tri- and tetracyclic compounds containing a macrocycle with
single diastereomers, suggesting that the methodology demonstrated here could be a powerful tool for
rapid preparation of highly complex molecules.
tions, such as Yamaguchi macrolactonisation,4 radical-mediated
Introduction
macrocyclisation,5 and Prins macrocyclisation,6 have been used
The olefin metathesis (OM) reaction has become one of the most
to synthesise various macrocycles. However, previous methods
important reactions in organic synthesis because it provides an
suffered from problems such as low productivity or the use of
easy and mild method to catalytically generate new carbon–
toxic reagents or excess chemical reagents. As an alternative
carbon double bonds.1 This has led to the development of various
method, the RCM reaction has emerged as an effective tool for
macrocyclisation (Scheme 1).7 This was well demonstrated in the
recent work from E. Lee’s lab, who utilised macro-RCM as the key
reaction for the total synthesis of (+)-Exiguolide.8
reactions, such as cross metathesis, ring-opening metathesis, and
ring-closing metathesis (RCM), which have been widely used
among both the organic and polymer communities. Among these
reactions, many organic chemists have been attracted to RCM
because of its easy access to various ring structures.2 Furthermore,
the development of highly active catalysts with N-heterocyclic
carbene (NHC) ligands3 broadened the substrate scope allowing
diverse functional groups into the ring with high productivity.
Scheme 1 Macrocyclic RCM reaction using a ruthenium catalyst.
Macrocyclisation is an important synthetic methodology be-
cause many natural products contain macrocycles. Many reac-
Many fused cyclic compounds comprising rings of various
sizes, from small to large,9 were synthesised by tandem radical
cyclisation. However, controlling reactive intermediates to give the
desired product can be difficult, and generation of toxic residues
remains a problem. Again, the RCM reaction has been recognised
as an alternative method since Grubbs’ first report on the synthesis
of fused bicyclic compounds by a tandem dienyne RCM reaction
(Scheme 2).10 This tandem RCM reaction was proved to be a
Department of Chemistry, Seoul National University, Gwanak 599, Gwanak-
ro, Gwanak-gu, Seoul, 151-742, Korea. E-mail: tlc@snu.ac.kr; Fax: +82
2-889-1568; Tel: +82 2-880-6658
† Electronic supplementary information (ESI) available: 1H, 13C NMR
spectrum data, refractive index, and HRMS spectrum data of substrates
are available. See DOI: 10.1039/c1ob05683b
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 5871–5878 | 5871
©

View Article Online
Scheme 2 Dienyne RCM reaction for synthesis of bicyclic compounds.
powerful tool because it was used to build the core skeletons in a
number of total syntheses of natural products.11
Although many useful molecules have been synthesised by
RCM, a further post-modification reaction on the RCM products
can provide more molecular diversity. For example, after an enyne
metathesis reaction, products with 1,3-dienes were synthesised,
and with appropriate dienophiles, Diels–Alder reaction was
frequently used as the post-modification reaction.12 However, ex-
amples of Diels–Alder reactions on bicyclic compounds produced
by dienyne RCM reactions are rare.13
Recently, we demonstrated a versatile tandem reaction com-
bining macrocyclisation and dienyne RCM to give bicyclic com-
pounds comprising small and large rings.14 Herein, we report the
details of our preliminary results of the tandem RCM reaction. The
resulting fused bicycles containing macrocycles proved to be good
dienes for the Diels–Alder reaction, producing fused multicyclic
compounds containing macrocycles.
Scheme 4 Selective dienyne RCM containing macrocycles.
Results and discussion
between the small ring (kS) and the large ring (kM). To test this
hypothesis, substrate 3a was treated with catalyst 1 under a diluted
concentration, and indeed the reaction produced 4ah and 4a only
To synthesise fused bicyclic compounds containing small and
large rings by dienyne RCM, dienyne substrates containing short
and long tethered alkenes are required. However, there is one
potential problem in this asymmetric dienyne RCM reaction:
dienyne substrates with different lengths of tethers give two
products with different ring sizes10 (Scheme 3). This is due to the
catalyst reacting with both terminal alkenes with no preference.
To prevent this problem, one terminal alkene was protected as a
disubstituted internal alkene, allowing the catalyst to react with the
less hinderedterminalalkenefirst.10,15 More recently, increasing the
reaction concentration also enhanced the selectivity by increasing
the exchange rate between the two possible carbenes (Scheme 4),16
but this strategy is not suitable for macrocyclisation because low
concentrations are required to avoid undesirable oligomerisation.
1
(Scheme 4). A more detailed study by H NMR revealed that
4ah was the initial intermediate product, which then underwent
the final macrocyclisation to give 4a. This provided an important
insight into the reaction pathway; as the catalyst reacted with
either of the two terminal alkenes, metal carbene intermediate A
or B formed. These metal carbenes underwent an enyne RCM
reaction with the nearby alkyne or an exchange reaction between
A and B. Lastly, dimerisation of 3a might have occurred by a
cross metathesis reaction. Here, only intermediate A provided
productive RCM due to rapid cyclisation, kS. On the other hand,
B did not undergo any productive reaction due to much slower kM
and kD. Instead, the exchange reaction proceeded to form A again,
which then produced 4ah immediately. After the initial enyne RCM
reaction that formed small rings exclusively, intermediate C or D
underwent the final macrocyclisation to form the desired fused
bicyclic product. Overall, these processes (kS, kEx > kM, kD) pushed
the equilibrium to one pathway, leading to the selective formation
of the final product 4a.
Previously, Fogg and co-workers reported that during their
studies of macrocyclisation of dienes, the dimer and the oligomers
formed initially as kinetic products and then reacted further
to yield the desired macrocycles.17 However, from our NMR
monitoring study, no dimer or oligomers were observed; there
was direct conversion to 4ah and 4a. We believe the difference
was due to instantaneous RCM, giving 4ah which contained one
terminal olefin and a less reactive conjugated olefin with a bulky
substituent. This might have reduced the chance of dimerisation
compared to substrates with two readily reactive terminal olefins.
Scheme 3 General reaction pathway in dienyne RCM.
The key to overcome the selectivity issue in the synthesis
of bicyclic compounds is controlling the rates of cyclisation.
To synthesise fused bicyclic compounds comprising small and
large rings, we reasoned that the selectivity issue might be
eliminated by taking advantage of the different cyclisation rates
5872 | Org. Biomol. Chem., 2011, 9, 5871–5878
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 1 Dienyne RCM reaction optimisationa
Table 2 Dienyne RCM reaction of various substratesa
Entry Substrate
1
Product
Temp. (Yield)
55 ◦C (95%)
Entry
Catalyst
Solvent
Temp.
Yield of 4a
4ah : 4ab
45 ◦
45 ◦
55 ◦
55 ◦
70 ◦
C
C
C
C
C
N/A
N/A
N/A
73%
82%
1 : 2
1 : 2
1 : 1.2
1 : 12
1 : 14
2
3
55 ◦C (85%)
55 ◦C (91%)
1
2
3
4
5
1
2
1
1
1
DCM
DCM
1, 2-DCE
Toluene
Toluene
a General reaction conditions: Under Ar atmosphere, 5 mol% catalyst was
added to a substrate in 4 mM degassed solvent. Solution was heated for 24
h under reflux conditions. b Ratio was determined by 1H NMR analysis.
Also, the cyclopentyl moiety might have induced a biased structure
favoring macro-RCM over dimerisation.
4
5
55 ◦C (85%)
70 ◦C (87%)
As shown in Scheme 4, the dienyne RCM reaction of 3a can
produce 4ah or 4a. When substrate 3a reacted with 5 mol% of
catalyst 1 in dichloromethane (DCM) at 45 ◦C, a 1 : 2 mixture
of 4ah and 4a was observed (Table 1, entry 1). The reactivity
toward macrocyclisation did not improve when catalyst 2 was
used instead (Table 1, entry 2). To push the reaction to complete
cyclisation, further optimisation was performed. To raise the
reaction temperature to 55 ◦C, 1,2-dichloroethane (1,2-DCE) was
used as the solvent, but it was less favorable for macrocyclisation,
giving almost equal amounts of 4ah and 4a (1 : 1.2, Table 1, entry
3). However, switching the solvent to toluene greatly improved
the conversion, forming the desired product in 73%◦yield (Table
1, entry 4). A further increase in temperature to 70 C increased
the catalyst activity, giving 4a with 82% yield (Table 1, entry 5).
Also, the reaction product showed only the trans isomer on the
macrocyclic alkene. This selectivity is noteworthy because E/Z
stereoselectivity in macrocyclic RCM reactions remains a serious
issue.18
a General reaction conditions: Under Ar atmosphere, 5 mol% catalyst 1
was added to substrates in 4 mM toluene. Solution was heated for 24 h
under reflux conditions.
Table 3 Synthesis of 6-membered ring containing fused bicyclea
Entry
Catalyst
Temp.
Yield of 4g
4gh : 4gb
1
2
3
10 mol (%)
10 mol (%)
10 mol (%)
55 ◦
70 ◦
90 ◦
C
C
C
N/A
59%
87%
1 : 1.5
1 : 8
1 : 28
To expand the scope of the reaction, various compounds with
different tether lengths were synthesised to prepare bicyclic [n.3.0]
(n = 12–15) compounds. In particular, fused bicyclic compounds
with 14- to 17-membered rings were prepared with 85–95%
isolated yields at 55 ◦C from substrates containing multiple oxygen
atoms on the long tether (Table 2, entries 1–4). The improved
conversion was due to the pseudo gem-dialkyl effect, in which
the carbon–oxygen–carbon bond angle becomes smaller than the
carbon–carbon–carbon bond angle due to lone pair electrons on
oxygen.19 Therefore, the macro-RCM reaction containing more
oxygen atoms (3b–e) gave better yields than that of 3a.20 Also,
introducing nitrogen resulted in good productivity, although 70 ◦C
was required for high conversion (Table 2, entry 5).
Synthesis of fused bicyclic compounds with 6-membered rings is
important because cyclohexyl derivatives are abundant in natural
products. Therefore, synthesis of 6-membered bicylic compounds
was also optimised. Substrate 3g showed low conversion to 4g
under the same conditions as the 5-membered ring derivative.
Increasing the catalyst loading to 10 mol% gave a 1 : 1.5 mixture
of 4gh and 4g (Table 3, entry 1). Increasing the temperature to
70 ◦C showed an enhanced yield of 4g, and further increase in
the temperature to 90 ◦C gave 87% isolated yield (Table 3, entries
2–3). Since the initial cyclisation of 6-membered rings completed
a General reaction conditions: Under Ar atmosphere, 5 mol% catalyst 1
was added to substrates in 4 mM toluene. Solution was heated for 24 h
under reflux conditions. Additional 5 mol% catalyst was added and the
solution was heated for another 24 h. b Ratio was determined by 1H NMR
analysis.
almost immediately after addition of the catalyst, we concluded
that the macrocyclisation was slower than that of the cyclopentene
containing substrates.
Similar to 5-membered ring containing fused bicycles, introduc-
ing oxygen atoms in the long tether resulted in better productivity
in tandem RCM reaction. Addition of one oxygen atom resulted
in 75% yield of 4h, even with 5 mol% catalyst (Table 4, entry 1).
Adding more oxygen atoms improved the conversion so that the
desired product was synthesised under milder conditions at lower
temperature (Table 4, entry 2). Nitrogen was also introduced in the
small ring (Table 4, entry 3), although the reaction required higher
temperature and higher loading of the catalyst to achieve 71%
yield. A bridgehead fused bicyclic compound was also synthesised
by dienyne RCM (Table 4, entry 4). Although the final product
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 5871–5878 | 5873
©

View Article Online
Table 4 Synthesis of various 6-membered ring containing fused bicyclesa
Table 6 Synthesis of fused bicycles with a mixture of stereoisomersa
Entry Substrate
1
Product
Conditions (Yield)
Entry Substrate
1
Product
Conditions (Yield)
5 mol% 1, 100 ◦C (75%)
5 mol%, 55 ◦C,
4 mM (63%)
5 mol% 1, 55 ◦C (71%)
8 mol% 1, 90 ◦C (71%)
2
5 mol%, 90 ◦C,
4 mM (87%)
2
3b
10 mol%, 70 ◦C,
2 mM (27%)
3b
4b
15 mol%, 100 ◦C,
2 mM (63%)
4b
10 mol% 1, 70 ◦C (69%)
a General reaction conditions: Under Ar atmosphere, 5 mol% catalyst 1
was added to substrates in 4 mM toluene. Solution was heated for 24 h
under reflux conditions. b Additional 3–5 mol% catalyst was added and the
solution was heated for another 24 h.
a General reaction conditions: Under Ar atmosphere, 5 mol% catalyst 1
was added to substrates in 4 mM toluene. Solution was heated for 24 h
under reflux condition. E/Z ratio was determined by 1H NMR analysis.
b Additional 5–10 mol% catalyst was added and the solution was heated
for another 24–48 h.
contained an anti-Bredt olefin, which was not readily formed in
small bridgehead molecules, reduction in strain due to the long
flexible chain on the macrocycle allowed the formation of the
desired product in 69% yield.
compounds was successful, resulting in better conversion when
1,2-DCE was used as the solvent. Similar to previous cases,
the substrate with the 6-membered ring required more forcing
conditions, such as higher catalyst loading and higher temperature,
for good conversion (Table 5, entries 3–4).
The synthesis of macrolactones was investigated because they
comprise the core skeletons of many natural products. Interest-
ingly, substrates with ester groups showed a different result for
the dienyne RCM reaction. Unlike the previous reactions, which
showed good conversion using toluene as the solvent, these ester
substrates showed decreased reactivity in toluene (Table 5, entry
1). However, by changing the solvent to 1,2-DCE, conversion
to the 14-membered macrolactone, 4l, greatly increased to 73%
yield (Table 5. entry 2). It seemed that the polar solvent 1,2-
DCE induced faster macrocyclisation to 4l containing an ester
group. Synthesis of a 6-membered ring containing fused bicyclic
Although most macro-RCM reactions produced E-olefins on
the macrocyclic alkenes, mixtures of E/Z isomers were observed
in the following cases. The first case was the synthesis of a
fused bicyclic compound containing a synthetically challenging
tetrasubstituted alkene on the small ring (Table 6, entries 1–2).
Initially, a moderate yield of 4n was obtained with an E/Z ratio
of 5/1. However, increasing temperature not only increased the
isolated yield (63% to 87%), but also improved the stereoselectivity
(E/Z = 5/1 to 8/1). The second example was the synthesis of a
fused bicyclic compound containing a 7-membered ring which
proved to be more challenging than bicycles comprising 5- and 6-
membered rings because the initial RCM for the 7-membered ring
formation was much slower. Therefore, to achieve good conversion
to 4o, ahigher temperature (100◦C) andhigher catalystloading(15
mol%) with longer reaction times were required. At low conversion
(Table 6, entry 3), an E/Z ratio of 5/1 was obtained for the
macrocyclic alkene, but again, higher selectivity (E/Z = 25/1) was
obtained at higher conversion (Table 6, entry 4). In both examples,
E/Z selectivity increased with higher conversion due to enhanced
catalyst activity. As the catalyst became more active at higher
temperature, reversible E/Z isomerisation on the macrocyclic
alkene occurred more frequently, leading to the formation of a
more stable E isomer.7d
Table 5 Synthesis of fused bicycles containing an ester groupa
Entry Substrate Conditions
Yield of 4 4xh : 4xb
1
2
3l
5 mol% 1, toluene, 55 ◦
C
19%
73%
N/A
83%
1 : 0.23
1 : 10
1 : 1
3l
5 mol% 1, 1,2-DCE, 55 ◦
5 mol% 1, 1,2-DCE, 55 ◦
C
C
C
Although the substrate scope for fused bicyclic macrocyclisation
was quite broad as seen in the previous tables, several substrates
did not lead to desired products. Unlike the previous example, in
which the fused bicyclic compound with an additional methyl
substituted olefin on the small ring was successfully prepared
(Table 6, entry 2), synthesising fused cyclic compounds containing
an additional substitution on the alkene of the macrocycles to
3
3m
3m
4c
10 mol% 1, 1,2-DCE, 70 ◦
1 : 10
a General reaction conditions: Under Ar atmosphere, 5 mol% catalyst 1
was added to substrates in 4 mM solvent. Solution was heated for 24
1
h under reflux conditions. b Ratio was determined by H NMR analysis.
c Additional 5 mol% catalyst was added and the solution was heated for
another 24 h.
5874 | Org. Biomol. Chem., 2011, 9, 5871–5878
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 7 Unsuccessful examples of tandem RCMa
Entry Substrate
1
Product
Conditions (Yield)
5 mol% 1, 55 ◦
C
(66%)
2
5 mol% 1, 55 ◦
(74%)
C
3b
Complex mixture
10 mol% 1, 100 ◦
C
a General reaction conditions: Under Ar atmosphere, 5 mol% catalyst 1
was added to substrates in 4 mM toluene. Solution was heated for 24 h
under reflux conditions. b Additional 5 mol% catalyst was added and the
solution was heated for another 24 h
Scheme 5 Attempts to prepare non-fused bicyclic compounds.
make trisubstituted alkenes failed, giving only molecules with
incomplete cyclisation (Table 7, entries 1–2). Because the addi-
tional substitution made it harder for the catalyst to perform the
productive metathesis reaction, the macrocyclisation rate became
even slower, leaving only small ring-closed products. Synthesis
of an 8-membered ring containing fused cyclic compounds was
even more challenging (Table 7, entry 3) due to intrinsically
low reactivity of RCM toward the formation of 8-membered
rings. Even with high catalyst loading at an elevated temperature,
this reaction gave a complex mixture of molecules, and clean
conversion to the initial 8-membered ring was not observed.
In previous examples, the terminal alkynes were used as the key
synthon for tandem RCM reactions. In order to extend the scope of
the reaction, a diene containing an internal alkyne, 5, was prepared
with the expectation of producing a non-fused bicyclic compound
5b via small ring cyclisation with 5a as an intermediate for tandem
macro-RCM reaction (Scheme 5). However, all our attempts
failed and only a complex mixture was obtained. As shown in
previous examples (Table 7, entry 1–2), macrocyclisation of 1,1-
disubstituted alkenes was challenging, so that the intermediate 5a
underwent side-reactions to generate a mixture of other products.
In order to avoid formation of the 1,1-disubstituted alkene
as an intermediate during the tandem macrocyclisation, another
substrate 6 which could lead to a non-fused bicyclic compound
comprising small and large rings by tandem ring-opening ring-
closing metathesis (RO-RCM) reaction was prepared.21 Initially,
we expected that substrate 6 would undergo RO-RCM reaction to
generate an intermediate 6a from small ring cyclisation followed by
the slow macrocyclisation to produce 6b. Again, disappointingly,
only a complex mixture was observed upon adding the catalyst,
suggesting that side-reactions were predominant. As shown in the
previous successful tandem RCM, it seems that the intermediate
with diene linked in the cyclic conformation might be the key for
efficient tandem RCM reaction by facilitating macrocyclisation.
We have investigated the selective synthesis of fused bicyclic
compounds by taking advantage of the rate difference between
cyclisation of small and large rings. This exclusive selectivity was
possible because kS and kEx were far greater than kM (Scheme
4). Thus, further experiments were designed to study selectivity
for the synthesis of fused bicyclic compounds comprising small
and medium rings by estimating the relative rates of kS, kEx and
the rate for cyclisation of the medium ring, kMed. When substrate
7 reacted at 4 to 20 mM, two products with a 1.5 : 1 ratio of
7a and 7b from non-selective RCM were observed during crude
NMR analysis. The 5- and 7-membered ring cyclisations occurred
almost equally at low concentration, implying that the cyclisation
rates of 5-, 6-, and 7-membered rings were faster than the kEx
between two different terminal alkenes at low concentration. At
higher concentration, kEx might dominate over 7-membered ring
cyclisation rate, giving higher selectivity for 7a,16 but side reactions,
such as dimerisation and oligomerisation, would occur. However,
when substrate 8 reacted at low concentration, only 8a, with
a 5-membered ring, was observed during crude NMR analysis.
Although complete tandem RCM did not occur, this suggested
that the ring-closure rate of the 8-membered ring (kMed) was much
slower than kEx, resulting in complete selectivity, similar to the
cases of tandem macrocyclisation.
So far, we have demonstrated that the fused bicyclic compounds
comprising small and large rings were efficiently prepared by
tandem dienyne RCM reactions. This method would be more
useful if further manipulation of the RCM products were possible,
giving more diversity. For a typical dienyne RCM reaction, 1,3-
diene, a potential functional group for Diels–Alder reaction,
is formed at the end of the reaction. However, prior to our
preliminary communication, there have been no previous reports
of Diels–Alder reactions on the dienes of fused bicycles formed
by dienyne RCM reactions because the products were composed
of two small or medium rings (Scheme 5). Thus, there was no
chance of forming s-cis dienes but s-trans dienes only, which cannot
participate in Diels–Alder reactions.10a On the other hand, we
reasoned that dienes on the bicycles comprising small and large
rings might adopt the s-trans conformation as well due to the
presence of flexible chains on the macrocycles. The first evidence
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 5871–5878 | 5875
©

View Article Online
for this came from a nuclear Overhauser effect (NOE) study
on substrate 4a, which showed that vinyl proton H1 interacted
with proton H3 in addition to an adjacent proton H2 (Scheme
7). Also, very crude computational analysis using molecular
mechanics methods suggested that the energy difference between s-
cis and s-trans conformations was quite small. This suggested that
substrate 4a might adopt both s-trans and s-cis conformations,
implying that these dienes could undergo Diels–Alder reactions
with dienophiles. Indeed, when treated with maleic anhydride at
70 ◦C, a cycloaddition product 9a with a single diastereomer was
obtained in 76% isolated yield.
Scheme 8 Origin of stereochemistry for the Diels–Alder reaction.
Scheme 6 Tandem RCM to synthesise fused cyclic compounds with small
and medium rings.
were obtained. To test dienophiles other than maleic anhydride,
dimethyl acetylenedicarboxylate was reacted with 4a. Even at
higher temperatures, this Diels–Alder reaction was much slower
than the previous cases with maleic anhydride as a dienophile.
We expected to obtain the Diels–Alder product with a 1,4-
cyclohexadiene moiety, but 9d with an aromatic ring was the
only isolated product. Presumably, the initial Diels–Alder product
of 1,4-cyclohexadiene quickly underwent aromatisation under
the reaction conditions, giving a tricyclic compound with the
aromatic ring (Table 8, entry 4).23 Since the Diels–Alder reactions
were conducted at elevated temperatures, it was unclear whether
equilibrium between s-cis and s-trans isomers of the dienes
was possible at room temperature. Thus, a stronger dienophile,
tetracyanoethylene, was added to 4l at room temperature, and the
product 9e was isolated with excellent yield (Table 8, entry 5). This
confirmed our initial assumption that the fused cyclic compounds
containing macrocycles could adopt the s-cis diene conformation
at room temperature due to the flexible chains on the macrocycles.
As the Diels–Alder reactions were performed with the same
solvent and the same temperature, the dienyne RCM and Diels–
Alder reactions could be conducted sequentially in a one-pot
reaction. Substrate 3c was treated with catalyst 1 at 70 ◦C, and
after 24 h, addition of two equiv. of maleic anhydride produced
the fused tetracyclic compound 10 with 37% isolated yield (Scheme
9). Although the yield was low, this result demonstrated that the
complexity of the molecules could be rapidly built up by the com-
bination of one-pot tandem dienyne RCM and Diels–Alder reac-
tions, producing the tetracycle from an acyclic starting compound.
Scheme 7 Diels–Alder reaction of a fused cyclic compound synthesised
by the dienyne RCM reaction.
As shown in Scheme 8, the Diels–Alder reaction could give
four different diastereomers depending on how the dienophile
approached the fused bicyclic diene molecule. First, the dienophile
could add to the diene from two different sides, but the dienophile
would preferentially approach from the less sterically hindered
side, Ha, and away from the more sterically hindered ether linkage.
Second, the dienophile could approach the diene with endo
orientation over exo. To determine the molecular structure of
the adduct, a 1D NOE study was conducted and showed that
Ha interacts with neither Hb or Hc, suggesting that they were anti
to one another.22 Also, the NOE study showed that Hb interacted
with Hc, confirming that the product isolated was the predicted 9a
isomer due to endo selective addition of the dienophile from the
less hindered side of the fused cyclic diene.
Reactions between maleic anhydride and various dienyne RCM
products containing 5-, 6-, and 7-membered rings in toluene
at 70 ◦C gave good yields of Diels–Alder products (74–76%)
comprising tetracyclic compounds (Table 8, entries 1–3). In all
cases, single diastereomers with the predicted stereochemistry
Conclusions
We presented a full report on the synthesis of fused cyclic ring
compounds comprising small and large rings with high selectivity,
5876 | Org. Biomol. Chem., 2011, 9, 5871–5878
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 8 Diels–Alder reaction on the dienyne RCM productsa
build complex molecules, especially those multicyclic compounds
containing macrocycles.
Entry Diene
1
Dienophile Product
Yield
76%
Experimental section
Typical procedure for dienyne ring closing metathesis reaction
(Table 1)
To a 2-neck round bottom flask, reaction substrate 3a (28.8 mg,
0.1 mmol) was added. The reaction flask was filled with argon gas,
followed by degassed solvent addition (25 ml, 4 mM). After 5 mo◦l%
catalyst (4.2 mmol) was added, the solution was heated to 70 C
for 24 h. Then, the solvent was removed by evaporation and the
product was purified with silica gel column chromatography (ethyl
acetate–hexane = 1 : 50). Inseparable mixtures of the final product
4a and 4ah with incomplete macrocyclisation were obtained in a
14 : 1 ratio. The reported yield was calculated by subtracting the
portion of the half-closed product. Rf 0.24 (ethyl acetate–hexane =
1 : 50)
2
3
74%
75%
1H NMR (500 MHz, CDCl3, ppm): d 6.120 (1H, dd, J = 0.5,
16.0 Hz), 5.770 (1H, m), 5.674 (1H, t, J = 2.5 Hz), 3.993 (1H, d, J =
1.0 Hz), 3.725–3.314 (2H, m), 2.340–2.038 (2H, d, J = 17.5 Hz),
2.240–2.038 (2H, m), 1.610–1.510 (2H, m), 1.498–1.331 (12H, b),
1.267(2H, s), 1.144 (3H, s), 1.058 (3H, s)
13C NMR (500 MHz, CDCl3, ppm): d 141.72, 132.05, 130.32,
126.85, 91.22, 68.54, 46.96, 42.38, 32.14, 30.26, 29.60, 26.75, 25.93,
25.81, 25.71, 24.66, 24.21, 23.03
4b
66%
83%
5c
HRMS (EI+) calcd. for C18H30O, 262.2297, found, 262.2298
General procedure for Diels–Alder reaction (Scheme 6)
a General information: Under Ar atmosphere, 2 equiv. dienophiles was
added to dienes in 0.3 M toluene. Rea◦ction flask was heated to 70 ◦C for 4
h. b Reaction flask was heated to 100 C for 48 h. c Reaction proceeded at
room temperature for 24 h.
To a 2-neck round bottom flask, fused bicyclic compound 4a
(23.8 mg, 0.09 mmol) was added, followed by addition of maleic
anhydride (17.9 mg, 0.18 mmol). Toluene (1 ml) was used as the
◦
solvent, and the solution was heated to 90 C for 4 h. Then, the
solvent was removed by evaporation and product 9a was purified
by silica gel column chromatography (ethyl acetate–hexane = 1 : 5)
to give 76% yield. Rf 0.28 (ethyl acetate–hexane = 1 : 5)
1H NMR (500 MHz, CDCl3, ppm): d 5.967 (1H, s), 3.690 (1H,
t, J = 4.5 Hz), 3.656 (1H, s), 3.545 (1H, dt, J = 6.5, 2.5 Hz), 3.389
(1H, t, J = 8.5 Hz), 3.302 (1H, dd, J = 6, 3.5 Hz), 2.688 (1H, d,
J = 8 Hz), 2.285 (1H, m), 1.977 (2H, m), 1.820 (2H, m), 1.645 (2H,
m), 1.500 (4H, m), 1.359 (8H, m), 1.048 (3H, s), 0.891 (3H, s)
13C NMR (500 MHz, CDCl3, ppm): d 172.10, 148.89, 124.56,
87.72, 71.58, 47.74, 44.23, 43.09, 38.13, 37.17, 36.45, 29.43, 28.30,
27.35, 27.10, 26.75, 26.57, 26.50, 26.45, 26.42, 21.42
Scheme 9 Sequential RCM–Diels–Alder reaction in one-pot reaction.
HRMS (EI+) calcd. for C22H33O4, 361.2379, found, 361.2377.
generality, and predictability by dienyne RCM reactions using
ruthenium catalyst 1. Because the cyclisation rate was significantly
faster for the small rings compared with the macrocycles, the
synthetic pathway was driven to produce single isomers. This
methodology efficiently produced fused bicycles with small rings
from 5- to 7-membered rings and macrocycles from 14- to 17-
membered rings, demonstrating the versatility of the reactions.
Generally, higher conversion to complete macrocyclisation was
achieved with higher temperature or higher catalyst loading. Also,
the RCM reaction products underwent Diels–Alder reactions with
exclusive stereocontrol. The combination of tandem dienyne RCM
and Diels–Alder reactions provided a powerful method to rapidly
Acknowledgements
Financial support from the National Research Foundation of Ko-
rea, BRL, BK21, SNU start-up Fund, and Chungam Fellowship
is acknowledged.
Notes and references
1 (a) R. H. Grubbs and S. Chang, Tetrahedron, 1998, 54, 4413; (b) A.
Fu¨rstner, Angew. Chem., Int. Ed., 2000, 39, 3013; (c) R. H. Grubbs,
Handbook of Metathesis, Vol. 1, 2, Wiley-VCH, Weinheim, 2003; (d) R.
H. Grubbs, Tetrahedron, 2004, 60, 7117.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 5871–5878 | 5877
©

View Article Online
2 (a) G. C. Fu, S. T. Nguyen and R. H. Grubbs, J. Am. Chem. Soc., 1993,
115, 9856; (b) R. H. Grubbs, S. J. Miller and G. C. Fu, Acc. Chem. Res.,
1995, 28, 446; (c) S. K. Armstrong, J. Chem. Soc., Perkin Trans. 1, 1998,
371.
3 For reports about catalyst 1, see: (a) M. Scholl, S. Ding, C. W. Lee
and R. H. Grubbs, Org. Lett., 1999, 1, 953; (b) T. M. Trnka and R.
H. Grubbs, Acc. Chem. Res., 2001, 34, 18; (c) T.-L. Choi and R. H.
Grubbs, Chem. Commun., 2001, 2648. For reports about catalyst 2,
see; (d) J. P. A. Harrity, M. S. Visser, J. D. Gleason and A. H. Hoveyda,
J. Am. Chem. Soc., 1997, 119, 1488; (e) J. P. A. Harrity, D. S. La, D. R.
Cefalo, M. S. Visser and A. H. Hoveyda, J. Am. Chem. Soc., 1998, 120,
2343.
4 A. Nakano, S. Takimoto, J. Inanaga, T. Katsuki, S. Ouchida, K. Inoue,
M. Aiga, N. Okukado and M. Yamaguchi, Chem. Lett., 1979, 1019.
5 For a review on radical cyclisations, see: G. S. C. Srikanth and S. L.
Castle, Tetrahedron, 2005, 61, 10377.
6 (a) P. A. Wender, B. A. DeChristopher and A. J. Schrier, J. Am. Chem.
Soc., 2008, 130, 6658; (b) S. K. Woo, M. S. Kwon and E. Lee, Angew.
Chem., Int. Ed., 2008, 47, 3242.
7 (a) B. Ko¨nig and C. Horn, Synlett, 1996, 1013; (b) A. Fu¨rstner and
K. Langemann, J. Org. Chem., 1996, 61, 3942; (c) A. Fu¨rstner and N.
Kindler, Tetrahedron Lett., 1996, 39, 7005; (d) C. W. Lee and R. H.
Grubbs, Org. Lett., 2000, 2, 2145; (e) A. B. Smith, III, S. A. Kozmin,
C. M. Adams and D. V. Paone, J. Am. Chem. Soc., 2000, 122, 4984;
(f) E. C. Hansen and D. Lee, J. Am. Chem. Soc., 2003, 125, 9582; (g) A.
Gradilas and J. Perez-Castells, Angew. Chem., Int. Ed., 2006, 45, 6086;
(h) K. C. Majumdar, H. Rahaman and B. Roy, Curr. Org. Chem., 2007,
11, 1339.
Yang, Y.-Y. Lai, W.-J. Xiao and H. Alper, Tetrahedron Lett., 2008,
49, 7334; (d) M. Movassaghi, G. Piizzi, D. S. Siegel and G. Piersanti,
Tetrahedron Lett., 2009, 50, 5489.
12 (a) D. A. Heerding, D. T. Takada, C. Kwon, W. F. Huffman and J.
Samaren, Tetrahedron Lett., 1998, 39, 6815; (b) H. Y. Lee, H. Y. Kim
and H. Tae, Org. Lett., 2003, 5, 3439; (c) M. Rosillo, G. Dom´ıniguez,
L. Casarrubios, U. Amador and J. Pe´rez-Castells, J. Org. Chem., 2003,
69, 2084; (d) M. Rosillo, G. Dominquez and L. Casarrubios, J. Org.
Chem., 2004, 6, 2084; (e) D. A. Clark, B. S. Basile, W. S. Karnofel and
S. T. Diver, Org. Lett., 2008, 10, 4927.
13 For an example of Diels–Alder reaction on a non-fused bicyclic ring
compound, see: M. J. Aldegunde, E. M. Codesido, L. Castedo and J.
R. Granja, Tetrahedron Lett., 2006, 47, 6587.
14 H. Park, Y.-L. Hong, Y. B. Kim and T.-L, Choi, Org. Lett., 2010, 12,
3442.
15 (a) C.-J. Wu, R. J. Madhushaw and R.-S. Liu, J. Org. Chem., 2003,
68, 7889; (b) T. Honda, H. Namiki, K. Kaneda and H. Mizutani, Org.
Lett., 2004, 6, 87; (c) S. V. Maifeld, R. L. Miller and D. Lee, J. Am.
Chem. Soc., 2004, 126, 12228; (d) S. V. Maifeld and D. Lee, Chem.–Eur.
J., 2005, 11, 6118.
16 (a) E. C. Hansen and D. Lee, J. Am. Chem. Soc., 2004, 126, 15074;
(b) E. C. Hansen and D. Lee, Acc. Chem. Res., 2006, 39, 509;
(c) Y. J. Kim, J. B. Grimm and D. Lee, Tetrahedron Lett., 2007, 48,
7961.
17 J. C. Conrad, M. D. Eelman, J. A. Duarte Silva, S. Monfette, H. H.
Parnas, J. L. Snelgrove and D. E. Fogg, J. Am. Chem. Soc., 2007, 129,
1024.
18 (a) P. Busca, M. Etheve-Quelquejeu and J.-M. Valre´y, Tetrahedron Lett.,
2003, 44, 9131; (b) Y. J. Kim and D. Lee, Org. Lett., 2004, 6, 4351; (c) J.
C. Conrad, M. D. Eelman, J. A. Duarte Silva, S. Monfette, H. H.
Parnas, J. L. Snelgrove and D. E. Fogg, J. Am. Chem. Soc., 2007, 129,
1024.
8 M. S. Kwon, S. K. Kook, S. W. Na and E. Lee, Angew. Chem., Int. Ed.,
2008, 47, 1733.
9 (a) N. A. Porter, V. H.-T. Chang, D. R. Magnin and B. T. Wright, J.
Am. Chem. Soc., 1988, 110, 3554; (b) C. P. Jasperse, D. P. Curran and
T. L. Fevig, Chem. Rev., 1991, 91, 1237; (c) D. P. Curran, Synlett, 1991,
63; (d) K. S. Feldman, H. M. Berven, A. L. Romanelli and M. Parez, J.
Org. Chem., 1993, 58, 6851; (e) L. Chen, G. B. Gill and G. Pattenden,
Tetrahedron Lett., 1994, 35, 2593; (f) B. P. Haney and D. P. Curran, J.
Org. Chem., 2000, 65, 2007.
19 S. W. Smith and M. S. Newman, J. Am. Chem. Soc., 1968, 90, 1249.
20 (a) A. Fu¨rstner, Top. Catal., 1997, 4, 285; (b) M. Delgado and J. D.
Martin, J. Org. Chem., 1999, 64, 4798.
21 W. J. Zuercher, M. Hashimoto and R. H. Grubbs, J. Am. Chem. Soc.,
1996, 118, 6634.
10 (a) S.-H. Kim, N. Bowden and R. H. Grubbs, J. Am. Chem. Soc., 1994,
116, 10801; (b) S.-H. Kim, W. J. Zuercher, N. B. Bowden and R. H.
Grubbs, J. Org. Chem., 1996, 61, 1073.
11 (a) F.-D. Boyer and I. Hanna, Tetrahedron Lett., 2002, 43, 7469; (b) K.
Shimizu, M. Takimoto and M. Mori, Org. Lett., 2003, 5, 2323; (c) Q.
22 For similar stereochemistry assignation, see: D. Rega´s, M. M. Alfonso,
M. L. Rodr´ıguez and J. A. Palenzuela, J. Org. Chem., 2003, 68,
7845.
23 A. V. Subrahmanyam, K. Palanichamy and K. P. Kaliappan, Chem.–
Eur. J., 2010, 16, 8545.
5878 | Org. Biomol. Chem., 2011, 9, 5871–5878
This journal is
The Royal Society of Chemistry 2011
©