Ruthenium-catalyzed tandem ring closing metathesis (RCM) - Atom transfer radical cyclization (ATRC) sequences

Article abstract of DOI:10.1016/j.jorganchem.2005.06.042

α-ω-Dienes bearing a pendant trichloroacetoxy group undergo a tandem RCM - radical cycloisomerization sequence leading to bicyclic γ-butyrolactones, with both steps of the sequence being catalyzed by ruthenium.

Full text of DOI:10.1016/j.jorganchem.2005.06.042

Journal of Organometallic Chemistry 690 (2005) 5552–5555
www.elsevier.com/locate/jorganchem
Ruthenium-catalyzed tandem ring closing metathesis (RCM) –
atom transfer radical cyclization (ATRC) sequences
*
Bernd Schmidt , Michael Pohler
University of Dortmund, FB Chemie, Organische Chemie, Otto-Hahn-Strasse 6, D-44227 Dortmund, Germany
Received 25 May 2005; accepted 30 June 2005
Available online 19 August 2005
Abstract
a-x-Dienes bearing a pendant trichloroacetoxy group undergo a tandem RCM – radical cycloisomerization sequence leading to
bicyclic c-butyrolactones, with both steps of the sequence being catalyzed by ruthenium.
Ó 2005 Elsevier B.V. All rights reserved.
Keywords: Ruthenium; Metathesis; Radical cyclization; Tandem sequence
The enormous progress achieved on the field of ruthe-
nium-catalyzed olefin metathesis [1] over the past few
years has been accompanied by the discovery of novel
and unexpected non-metathesis reactivity patterns of
ruthenium carbene complexes [2]. These include hydro-
genation [3], isomerization [4], hydrosilylation [5], atom
transfer radical addition [6], and, as recently demon-
strated by Quayle et al. [7], the radical cycloisomeriza-
tion of polyhalogenated compounds, which had
previously been catalyzed by other metal complexes
[8]. The novel reactivity patterns of ruthenium carbene
complexes hold a number of opportunities for organic
synthesis. For instance, they open up perspectives
towards diversity oriented synthesis [9], ‘‘catalyst econ-
omy’’ [10b], and novel catalyzed tandem reactions [11–
13]. However, reaction conditions have to be found that
allow for the strict separation of metathesis and non-
metathesis reactivity in order to obtain high selectivities.
For radical reactions this has recently been achieved by
us [10a] and by Quayle et al. [10b] In this contribution,
we report our preliminary results on a tandem ring clos-
ing metathesis – radical cyclization reaction as a tool for
the stereoselective synthesis of bicyclic lactones.1 We
started our investigation with diene 1a. The RCM prod-
uct of 1a is trichloroacetoxy-substituted cyclohexene 2a,
which is one of the substrates investigated by Quayle
and co-workers in their investigation of radical cycloiso-
merization reactions catalyzed by first generation
1
Representative procedure and analytical data: (3aR , 4R , 6R ,
*
*
*
7aR )-6-Benzyloxy-3,3,4-trichloro-hexahydro-benzofuran-2-one (8a).
*
To a solution of 7a (300 mg, 0.8 mmol) in toluene was added
precatalyst C (33 mg, 4.9 mol%). The mixture was stirred at ambient
temperature until the starting material was fully consumed (TLC
hexanes–MTBE (5:1), approximately 1 h), and then heated to reflux
until the intermediate RCM product was fully consumed (TLC
hexanes–MTBE (5:1), approximately 4 h). The solvent was evapo-
rated, and the residue was purified by chromatography on silica to give
8a (170 mg, 62%) as a colourless solid, mp 110 °C. 1H NMR
(600 MHz, CDCl3) d 7.36–7.28 (5H, Ph), 5.10 (ddd, 1H, J = 4.4, 4.0,
4.0 Hz, H-7a), 4.57 (d, 1H, J = 11.7 Hz, –OCH2Ph), 4.52 (d, 1H,
J = 11.7 Hz, –OCH2Ph), 3.82 (ddd, 1H, J = 12.1, 9.2, 4.8 Hz, H-4),
3.58 (dddd, 1H, J = 10.6, 10.6, 4.0 Hz, 4.0, H-6), 3.12 (dd, 1H, J = 9.2,
4.5 Hz, H-3a), 2.66-2.58 (2H, H-5eq, H-7eq), 1.82 (ddd, 1H, J = 12.1,
10.6, 10.6 Hz, H-5ax), 1.78 (ddd, 1H, J = 14.7, 10.3, 4.0 Hz, H-7ax). 13
C
NMR (150 MHz, CDCl3)d 166.8 (0, C@O), 137.5 (0, ipso-C, Ph), 128.6
(1, Ph), 128.0 (1, Ph), 127.6 (1, Ph), 80.8 (0, CCl2), 77.2 (1, C-7a), 71.0
(2, –OCH2Ph), 69.8 (1, C-6), 58.5 (1, C-3a), 52.8 (1, C-4), 40.7 (2, C-5),
33.1 (2, C-7). IR (neat, NaCl) m [cmÀ1] 2921 (m), 1802 (s), 1095 (s), 956
(s).
*
Corresponding author. Fax: +49 231 755 5363.
E-mail address: bernd.schmidt@uni-dortmund.de (B. Schmidt).
0022-328X/$ - see front matter Ó 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.06.042

B. Schmidt, M. Pohler / Journal of Organometallic Chemistry 690 (2005) 5552–5555
5553
GrubbsÕ catalyst (A) [14a] (Fig. 1) [7]. Three prerequi-
sites are required for the combination of ring closing
metathesis and cycloisomerization to a tandem se-
quence: (1) the olefin metathesis step must proceed effi-
ciently at temperatures where no radical reaction is
initiated by the catalyst; (2) carbene complex B, which
is formed from A after the first catalytic cycle, must be
an equally efficient catalyst precursor for the actual
[10b] radical cycloisomerization catalyst as A; (3) other
non-metathesis side reactions, especially olefin isomeri-
zation [2b], must not interfere with either metathesis step
or radical reaction.
identical conditions. Surprisingly, not only the RCM
step proceeded smoothly at ambient temperature, but
the subsequent radical cyclization step was also effi-
ciently catalyzed upon heating. 4b, presumably formed
from 3b via dehalogenation and double bond migration,
was the sole product isolated from the reaction mixture
(Scheme 1).
We tried to observe the proposed intermediate 3b by
monitoring the reaction with NMR-spectroscopy. To
this end, the reaction was repeated in d6-benzene. After
1 h at room temperature, 1b was completely consumed
1
and 2b was the only product observed by H and 13C
We found that 1a is cleanly converted to 2a in the
presence of 5 mol% A at ambient temperature within
3 h. Heating the reaction to reflux induces formation
of the lactone 4a, which could be isolated in 63% yield.
This finding is in accord with QuayleÕs report [7], who
also did not observe the primary radical cyclization
product 3a in this case. We wanted to extend the concept
to other examples and were particularly interested in
fivemembered bicyclic lactones, because this is an impor-
tant structural pattern in prostacycline natural products.
To this end, 1b was subjected to the same reaction con-
ditions as 1a, however, even after 12 h at ambient tem-
perature no significant conversion to the intermediate
RCM product 2b was observed. We hoped that upon
heating ring closing metathesis would occur sufficiently
faster than radical cyclization to allow clean conversion
of 1b to a bicyclic lactone. This is unfortunately not the
case: although 4b is the major product of the reaction,
conversion is rather sluggish and several unidentified
byproducts are formed, presumably resulting from rad-
ical reactions that occurred prior to the RCM step. An
obvious way to solve reactivity problems in olefin
metathesis reactions is to switch from first to second
generation precatalysts, such as C [14b], which is well-
known for its enhanced activity in metathesis reactions.
It has, however, recently been stated by Snapper et al.
[15] that it is ‘‘not nearly as active in Kharasch addi-
tions’’ compared to the first generation catalyst A. This
statement is in accord with our observations that A cat-
alyzes efficiently atom transfer radical cyclizations of
certain a,x-dienes, while C catalyzes selectively the ole-
fin metathesis of the same substrates [10a]. Nevertheless,
the attempted conversion of 1b was repeated using
5 mol% of second generation catalyst C under otherwise
NMR spectroscopy. The reaction mixture was then
warmed to 40 °C for 1 h, but no change was observed
in the NMR-spectra. Further heating to reflux (80 °C)
induced consumption of 2b, as monitored by TLC. In
the H NMR-spectrum three additional signals appeared
at low field from the aliphatic region, however, these
were broad and shifted significantly compared to the
spectra previously obtained from the reaction mixture.
It was not possible to unambiguously assign any signal
to either 2b, 3b or 4b. We assume that these problems
can be attributed to the formation of paramagnetic
ruthenium species once the radical reaction is initiated.
Unfortunately, it was also impossible to isolate the inter-
mediate 3b from the reaction mixture, because this com-
pound could not be observed as a defined spot on TLC,
indicating that decomposition occurs upon contact with
silica. Structural assignment of compound 4b is based
on H,H- and H,C-correlation spectroscopy in combina-
tion with one-dimensional gradient-selected NOE-spec-
troscopy. All experiments reveal that the endocyclic
double bond is located in the allylic position relative
to the lactone oxygen. Quayle et al. [16] have recently
addressed the issue of elimination and double bond
isomerization in detail from a mechanistic point of view
for the transformation of 2a to 4a.
We assumed that the double bond migration ob-
served during the conversion of 1a,b to 4a,b should be
suppressed by introducing an additional substituent in
H
O
O
O
i or ii
O
CCl3
n
n
H
Cl
Cl
1a (n = 1)
1b (n = 0)
4a (n = 1)
4b (n = 0)
H
H
O
O
O
PCy3
PCy3
MesN
Cl
NMes
Ph
O
Cl
Cl
Ph
Cl
Cl
CCl3
Ru
Ru CH2
PCy3
n
n
Ru
Cl
Cl
PCy3
Cl
Cl
PCy3
2a (n = 1)
2b (n = 0)
3a (n = 1)
3b (n = 0)
A
B
C
Scheme 1. Reagents and conditions: i, A (5 mol%), toluene, 20–110 °C,
(63% of 4a); ii, C (5 mol%), toluene, 20–110 °C, (61% of 4b).
Fig. 1. Ruthenium carbene complexes.

5554
B. Schmidt, M. Pohler / Journal of Organometallic Chemistry 690 (2005) 5552–5555
the metathesis precursor. As the relative configuration
of these metathesis precursors might have influence on
both cyclization steps, we wanted to have access to both
diastereoisomers. To this end, 7a and 7b were synthe-
sized starting from ethyl acetate, which was reacted with
acroleine in an aldol reaction. The resulting aldol prod-
uct was protected as a THP-ether to give ester 5, which
was converted to a 1:1 mixture of diastereomeric alco-
hols 6a and 6b via a one-pot DIBAL-H reduction of
the ester function, trapping of the resulting aldehyde
with allylmagnesium bromide, protection of the alcohol
as a benzyl ether, and final cleavage of the THP–acetal.
Diastereomers 6a and 6b could be separated by column
chromatography on silica and were separately esterified
using trichloroacetyl chloride to give the desired precur-
sors 7a and 7b as single diastereoisomers (Scheme 2).
We first checked whether or not dienes 7 would un-
dergo ring closing metathesis using the first generation
catalyst A at ambient temperature. To this end, 7b was
treated with 5 mol% of first generation GrubbsÕ catalyst
in toluene at ambient temperature. Unfortunately, un-
der these conditions the reaction stops at approximately
30% conversion after 2 h. Even after 20 h at ambient
temperature no improvement could be observed. We
therefore repeated the experiment using 5 mol% of sec-
ond generation catalyst C. This lead to a rapid con-
sumption of the metathesis precursor within 30 min
and clean formation of the primary metathesis product,
as indicated by TLC. Gratifyingly, heating the mixture
to reflux induces the formation of a bicyclic product
via an atom transfer radical cyclization. Surprisingly,
the trichlorinated compound 8b was obtained as the sole
product in high diastereoselectivity and no elimination
products could be detected. Conversion of the diastereo-
meric precursor 7a to the bicyclic product 8a proceeds
equally efficient, indicating that the different orientation
of the additional benzyloxy group does not influence the
efficiency of either cyclization step. In both cases com-
plete conversion of the intermediate RCM product
was observed within 4 h (Scheme 3).
Elucidation of the relative configuration of 8a and 8b
was achieved using one- and two-dimensional NMR-
techniques. Especially NOE-experiments and analysis
of vicinal coupling constants turned out to be very use-
ful to prove the cis-junction of cyclohexane and lactone
ring. The trans-orientation of the chloro-substituent rel-
ative to the lactone ring becomes obvious from the com-
paratively large trans-diaxial coupling constants
observed for the proton –CHCl–. These values are in
the range of 8 to 10 Hz. It should be mentioned that
the high diastereoselectivity observed for the ATRC step
is not without precedence in the literature. Nagashima
et al. [8d] demonstrated that the copper-catalyzed atom
transfer radical cyclization of 2a yields 3a as a single dia-
stereoisomer, with the chloro-substituent being in a
trans-arrangement relative to the lactone ring.
In conclusion, we were able to demonstrate that tan-
dem ring closing metathesis-atom transfer radical cycli-
zation reaction sequences are in principle possible. The
second generation GrubbsÕ catalyst, which had previ-
ously been reported by several groups to be less reactive
in ATRA and ATRC reactions than the first generation
catalyst, turned out to mediate both steps of these se-
quences in preparatively useful yields and rates of con-
version. Further investigations into scope and
O
O
OTHP
i, ii
iii-v
OH
OEt
OEt
5
BnO
OH
BnO
+
6a
6b
vi
H
BnO
BnO
O
O
O
BnO
BnO
O
O
i
i
vi
O
O
CCl3
BnO
O
O
BnO
O
O
H
Cl
Cl
Cl
Cl
Cl
CCl3
CCl3
7a
7b
8a
H
O
7b
7a
CCl3
Scheme 2. Reagents and conditions: i, LDA, THF, À78 °C, then add
acroleine; ii, 2,3-dihydropyran, p-TSA (cat.) (58% from ethyl acetate);
iii, DIBAL-H (1.3 equiv.), DCM, À90 °C, then add allylmagnesium
bromide, À90 to 20 °C; iv, NaH, THF, 65 °C, then add benzyl
bromide; v, dissolve in methanol, p-TSA (10 mol%), separate by
chromatography (33% of 6a, 31% of 6b from 5); vi, Cl3COCl, NEt3,
DMAP (cat.), DCM (99% of 7a, 99% of 7b).
H
Cl
8b
Scheme 3. Reagents and conditions: i, C (5 mol%), toluene, 20–110 °C,
(61% of 8a; 62% of 8b).

B. Schmidt, M. Pohler / Journal of Organometallic Chemistry 690 (2005) 5552–5555
5555
(d) Cu-catalyzed cyclizations: H. Nagashima, K. Seki, N. Ozaki,
H. Wakamatsu, K. Itoh, Y. Tomo, J. Tsuji, J. Org. Chem. 55
(1990) 985;
limitations of this sequence are currently in progress in
our laboratory.
(e) Ru-catalyzed cyclizations: H. Nagashima, H. Wakamatsu, N.
Ozaki, T. Ishii, M. Watanabe, T. Tajima, K. Itoh, J. Org. Chem.
57 (1992) 1682;
Acknowledgements
(f) Other Ru-catalyzed ATRA-reactions: H. Matsumoto, K.
Ohkawa, S. Ikemori, T. Nakano, Y. Nagai, Chem. Lett. (1979)
1011;
(g) H. Matsumoto, T. Nakano, K. Ohkawa, Y. Nagai, Chem.
Lett. (1978) 363;
This work has been generously supported by the
Deutsche Forschungsgemeinschaft and the Chemistry
Department of the University of Dortmund. We thank
Prof. Dr. B. Costisella for conducting non-routine
NMR experiments.
(h) H. Matsumoto, T. Nakano, T. Nikaido, Y. Nagai, Chem.
Lett. (1978) 115.
[9] M.D. Burke, S.L. Schreiber, Angew. Chem. Int. Ed. 43 (2004)
46.
References
[10] (a) B. Schmidt, M. Pohler, B. Costisella, J. Org. Chem. 69 (2004)
1421;
(b) J. Faulkner, C.D. Edlin, D. Fengas, I. Preece, P. Quayle, S.N.
Richards, Tetrahedron Lett. 46 (2005) 2381.
[11] Reviews: (a) A. Ajamian, J.L. Gleason, Angew. Chem. Int. Ed. 43
(2004) 3754;
[1] Reviews: (a) R.H. Grubbs (Ed.), Handbook of Metathesis, Wiley-
VCH, Weinheim, 2003;
(b) A. Furstner, Angew. Chem. Int. Ed. 39 (2000) 3013;
¨
(c) S.J. Connon, S. Blechert, Angew. Chem. Int. Ed. 42 (2003) 1900;
(b) D.E. Fogg, E.N. dos Santos, Coord. Chem. Rev. 248 (2004)
2365.
(d) B. Schmidt, J. Hermanns, Top. Organomet. Chem. 13 (2004)
223.
[12] RCM-hydrogenation: (a) J. Louie, C.W. Bielawski, R.H. Grubbs,
J. Am. Chem. Soc. 123 (2001) 11312;
[2] Reviews: (a) B. Alcaide, P. Almendros, Chem. Eur. J. 9 (2003)
1259;
(b) A. Furstner, A. Leitner, Angew. Chem. Int. Ed. 42 (2003)
¨
(b) B. Schmidt, Eur. J. Org. Chem. (2004) 1865.
308;
[3] S.D. Drouin, F. Zamanian, D.E. Fogg, Organometallics 20 (2001)
(c) B. Scheiper, F. Glorius, A. Leitner, A. Furstner, Proc. Natl.
¨
5495.
Acad. Sci. USA 101 (2004) 11960;
[4] (a) B. Alcaide, P. Almendros, J.M. Alonso, Chem. Eur. J. 9
RCM-isomerization: (d) A.E. Sutton, B.A. Seigal, D.F. Finn-
egan, M.L. Snapper, J. Am. Chem. Soc. 124 (2002) 13390;
(e) B. Schmidt, Eur. J. Org. Chem. (2003) 816;
(f) B. Schmidt, Chem. Commun. (2004) 742;
(g) B. Schmidt, J. Org. Chem. 69 (2004) 7672.
[13] Ru-catalyzed tandem ROMP-radical addition has been reported
for the synthesis of high molecular weight polymers: (a) C.W.
Bielawski, J. Louie, R.H. Grubbs, J. Am. Chem. Soc. 122 (2000)
12872;
(b) H. Katayama, F. Yonezawa, M. Nagao, F. Ozawa, Macro-
molecules 35 (2002) 1133.
[14] (a) P. Schwab, R.H. Grubbs, J.W. Ziller, J. Am. Chem. Soc. 118
(1996) 100;
(2003) 5793;
(b) C. Cadot, P.I. Dalko, J. Cossy, Tetrahedron Lett. 43 (2002)
1839.
[5] (a) S.V. Maifeld, R.L. Miller, D. Lee, Tetrahedron Lett. 43 (2002)
6363;
´
(b) C.S. Arico, L.R. Cox, Org. Biomol. Chem. 2 (2004) 2558;
(c) S.V. Maifeld, M.N. Tran, D. Lee, Tetrahedron Lett. 46 (2005)
105.
[6] (a) J.A. Tallarico, L.M. Malnick, M.L. Snapper, J. Org. Chem.
64 (1999) 344;
(b) F. Simal, A. Demonceau, A.F. Noels, Tetrahedron Lett. 40
(1999) 5689;
(c) B. De Clercq, F. Verpoort, Tetrahedron Lett. 42 (2001) 8959.
[7] P. Quayle, D. Fengas, S. Richards, Synlett (2003) 1797.
[8] Reviews: (a) J. Iqbal, B. Bhatia, N.K. Nayyar, Chem. Rev. 94
(1994) 519;
(b) M. Scholl, S. Ding, C.W. Lee, R.H. Grubbs, Org. Lett. 1
(1999) 953.
´
[15] B.T. Lee, T.O. Schrader, B. Martın-Matute, C.R. Kauffman, P.
Zhang, M.L. Snapper, Tetrahedron 60 (2004) 7391.
[16] C.D. Edlin, J. Faulkner, D. Fengas, C.K. Knight, J. Parker, I.
Preece, P. Quayle, S.N. Richards, Synlett (2005) 572.
(b) B.M. Trost, M.J. Krische, Synlett (1998) 1;
(c) H. Nagashima, in: S.-I. Murahashi (Ed.), Ruthenium in
Organic Synthesis, Wiley-VCH, Weinheim, 2004;