A solid-supported phosphine-free ruthenium alkylidene for olefin metathesis in methanol and water.

Article abstract of DOI:10.1016/S0960-894X(02)00260-3

The synthesis and olefin metathesis activity in protic solvents of 7, a phosphine-free ruthenium alkylidene bound to a hydrophilic solid support are reported. This heterogeneous catalyst promotes relatively efficient ring closing- and cross-metathesis reactions in both methanol and water. The potential utility of homogeneous catalyst 2 for olefin metathesis in methanol is also demonstrated.

Full text of DOI:10.1016/S0960-894X(02)00260-3

Bioorganic & Medicinal ChemistryLetters 12 (2002) 1873–1876
A Solid-Supported Phosphine-Free Ruthenium Alkylidene for
Olefin Metathesis in Methanol and Water
Stephen J. Connon and Siegfried Blechert*
Institut fu¨r Chemie, Technische Universita¨t Berlin, Strasse des 17 Juni 135, 10623 Berlin, Germany
Received 28 November 2001; accepted 21 January2002
Abstract—The synthesis and olefin metathesis activity in protic solvents of 7, a phosphine-free ruthenium alkylidene bound to a
hydrophilic solid support are reported. This heterogeneous catalyst promotes relatively efficient ring closing- and cross-metathesis
reactions in both methanol and water. The potential utilityof homogeneous catal ys t 2 for olefin metathesis in methanol is also
demonstrated. # 2002 Elsevier Science Ltd. All rights reserved.
The development of highlystable and active ruthenium
1
alkylidenes promoted living ROMP in acidic aqueous
10
2
alkylidenes such as 1 and 2 (Fig. 1) bearing N-hetero-
cyclic carbene (NHC) ligands has significantly broad-
solution, and for the first time made possible the
RCM of specific diene substrates in methanol and
3
11
ened the scope of the olefin metathesis reaction.
Consequently, ring closing metathesis (RCM) to form
water. On the other hand, these catalysts are reported
to be extremelyox gy en sensitive, requiring manipu-
lation under an inert atmosphere and rigorous degas-
sing of solvents. In addition theydid not promote the
ring closure of simple a,o-heptadienes (one terminal-
and one internal-olefin component is required) and they
1
,4
tetrasubstituted cycloalkenes,
thesis (CM) with highlyelectron deficient alkenes
selective cross meta-
3
,5
and
even ring opening-cross metathesis (ROM-CM) of
6
unstrained cycloolefins are now possible. Despite its
potential importance in the concise synthesis of biologi-
callyimportant molecules, less progress has been made
with regard to the design of olefin metathesis catalysts
for use in protic solvents such as methanol or water.
The advantages of such a strategyare obvious; for
example tedious protection/deprotection steps that are
3
c
commonplace in carbohydrate metathesis chemistry
could be avoided, and environmental impact could be
7
reduced. Earlyexamples
detailed the ring opening
metathesis polymerisation (ROMP) of strained cyclo-
olefins in aqueous media initiated bypoorlydefined
ruthenium complexes such as RuCl (hydrate) or
3
Ru(H O) (Tos) . However, these polymerisations were
2
6
2
not living, and the water soluble initiators could not be
applied to either RCM or CM. Later, the ROMP of
carbohydrate functionalised monomers in aqueous/
organic mixtures using the well-defined ruthenium
8
initiator (PCy ) Cl Ru¼CHPh were reported.
A
3
2
2
breakthrough came with the preparation byGrubbs et
al. of water soluble metathesis catalysts 3 and 4. These
9
*
3
Corresponding author. Tel.: +49-30-3142-2255; fax: +49-30-3142-
619; e-mail: blechert@chem.tu-berlin.de
Figure 1. Ruthenium alkylidene olefin metathesis catalysts (Cy, cyclo-
hexyl; Mes, 2,4,6-trimethylphenyl).
0
960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00260-3

1874
S. J. Connon, S. Blechert / Bioorg. Med. Chem. Lett. 12 (2002) 1873–1876
possessed negligible CM activity. We therefore sought
to applythe high stabilityof phosphine-free catal ys t
to the problem of RCM and CM activityin methanol
and water.
this solid support swells 4 times more per unit mass in
water than the polystyrene/polyethylene glycol-based
2
1
4
TentaGel resins, and proved perfectlystable under
metathesis conditions. Isopropoxystyrene 5 was pre-
2
a
pared according to the literature procedure and cou-
ꢁ
1
In contrast to 1, phosphine-free catalyst 2 is reasonably
soluble in methanol at room temperature, and readily
pled to PEGA-NH resin (loading 0.40 mmol g ) under
2
standard conditions to give the immobilised ligand 6.
To ensure complete incorporation of the styrene moiety
this coupling step was repeated until the resin gave a
ꢀ
dissolves in this solvent at 50 C. However, 2 is com-
pletelyinsoluble in water or water–methanol mixtures.
We have recentlyreported two pol ym er bound variants
of 2 which had excellent activityand rec yc labilityin
1
4
negative Kaiser test. As a precautionarymeasure the
resin was then subjected to acylation conditions. Treat-
ment of 6 with excess 1 in the presence of CuCl as a
phosphine scavenger gave the solid-supported catalyst 7
1
2
RCM and CM processes, and therefore to circumvent
this solubilityproblem, the idea of preparing a suitably
substituted analogue of 2 immobilised on a hydrophilic
solid support seemed attractive. It was hoped that with
such a heterogeneous species, metathesis reactions in
water could occur to some extent in the resin pores,
where the pre-catalyst concentration is relatively high,
therebymore efficientlyexposing the substrate to the
hydrophobic ruthenium moiety. A similar strategy for
heterogeneous catalysis in methanol has been briefly
1
5
as a deep green resin (Scheme 1).
The activityof both 2 and 7 in RCM was tested, using
methanol and water as solvents. From the outset it was
decided that for the use of these catalysts to be con-
sidered advantageous compared to 3 or 4, theymust be
capable of promoting the RCM of a,o-heptadienes
(Scheme 2) in the presence of adventitious oxygen, and
so no attempt was made to degas solvents and the
reactions were performed under an air atmosphere. The
results of these experiments are outlined in Table 1.
1
3
described recentlybyDowden
with promising results.
Commerciallyavailable 14 PEGA-NH resin (Fig. 2) was
2
chosen as the hydrophilic solid support (Scheme 1).
Consisting of amino functionalised dimethyl acrylamide
and mono-2-acrylamidoprop-1-yl polyethyleneglycol,
Gratifyingly both 2 and 7 served as active catalysts for
RCM in methanol under these conditions, with 7 also
exhibiting activityin water. Ammonium salt 8, which
has proven incompatible with NHC ligand-based cata-
1
6
lysts even in CH Cl
2
could be cyclised cleanly in
2
moderate yields with either 7 or 2 in methanol, and even
underwent some reaction in water. It is noteworthythat
for metathesis reactions promoted by 7, methanol is
actuallya superior solvent to CH Cl .
2
2
Heptadienes 9 and 10 gave higher conversions to RCM
products 13 and 14, respectively, but were also able to
1
7
undergo a competing cyclo-isomerisation pathway to
afford 17 and 18. In an attempt to understand the ori-
gins of these byproducts, the RCM of 9 in CD OD was
3
1
monitored by H NMR spectroscopyat room tempera-
ture. It was found that formation of 17 was less
Figure 2. PEGA-NH
2
resin.
favoured at the lower temperature, onlyoccurring after
70% of 9 had been consumed, and coinciding with the
appearance of a signal at ꢁ5.05 ppm. Interestingly,
conversion to 17 continued even after the disappearance
1
8
of the alkylidene signal at 16.63 ppm (ca. 130 min) and
ceased onlyafter the highfield signal could no longer be
observed (ca. 240 min), indicating that this catalytic
isomerisation is promoted bya decomposition product
of 2, and not by 2 itself. The RCM of 11 in D O was
2
Scheme 1. Reagents and conditions: (a) DIC (5.0 equiv), HOBt
ꢀ
(5.0 equiv), 5 (5.0 equiv), PEGA-NH
then repeat; (b) 6 (1.0 equiv) Ac
2
(1.0 equiv), DMF, 25 C, 12 h,
O (30 equiv), TEA (35 equiv) DMAP
2
ꢀ
(
3.0 equiv), CH
1.3 equiv), CH
2
Cl
Cl
2
, 25 C, 5 h; (c) 6 (1.0 equiv), 1 (1.3 equiv), CuCl
ꢀ
(
2
2
, 45 C, 4 h.
Scheme 2. RCM in methanol and water.

S. J. Connon, S. Blechert / Bioorg. Med. Chem. Lett. 12 (2002) 1873–1876
Table 1. RCM in methanol and water promoted by 2 and 7
1875
Diene Solvent Catalyst Temp. Conversion
ꢀ
Product(s)
a
(%)a
(
C)
(%)
8
8
8
8
8
9
9
9
9
1
1
1
1
1
CH
2
Cl
O
O
2
7
7
7
7
2
7
7
2
2
7
2
2
2
7
45
45
rt
45
45
45
45
45
rt
45
45
rt
rt
rt
33
11
0
12 (33)
12 (11)
None
12 (57)
12 (69)
13 (44)+17 (9)
13 (70)+17 (24)
13 (83)+17 (10)
13 (86)+17 (6)
14 (38)+18 (44)
14 (71)+18 (29)
14 (58)
H
H
2
2
MeOH
MeOH
CH Cl
57
69
b
2
2
100
100
b
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
b
100
92
82
100
58
73
96
0
0
0
1
1
15 (73)
15 (96)
D
2
O
Scheme 3. CM (olefin dimerisation) promoted by 2 and 7.
a
1
Relative integration of H NMR spectrum.
Acyclic isomerisation of terminal olefin also evident.
b
Table 2. CM in methanol and water promoted by 2 and 7
Substrate
Solvent
Catalyst
Conversion
a
Product(s)
a
(
%)
(%)
20
20
21
22
23
24
24
25
26
27
28
28
D
2
O
MeOH
7
2
7
7
7
7
2
7
2
2
7
2
80
69
79
83
45
100
100
7
29 (80)
29 (69)
30 (79)
31 (83)
32 (45)
33 (100)
33 (100)
34 (7)
Scheme 4. Attempted selective cross-metathesis.
D
2
D
2
D
2
O
O
O
b
was formed (E/Z=4.1:1) in MeOH using homogeneous
catalyst 2; however, a wasteful alkene isomerisation
reaction was dominant. To our knowledge this is the
first example of CM involving an unprotected glycoside
in pure MeOH. Ammonium salts 26 and 27 either did
not react or underwent a similar isomerisation process.
MeOH
MeOH
D O
2
MeOH
MeOH
2
D O
3
78
0
35 (3)
36 (5)+39 (73)
None
37 (28)+40 (37)
MeOH
65
Recentlyselective cross-metathesis with acrl yi c acid
a
1
Relative integration of H NMR spectrum.
3
,5
b
derivatives has been reported. DisappointinglyCM
between N-isopropyl acrylamide and 21 (Scheme 4) was
unselective, affording a 52:48 mixture of CM product 41
2
Not miscible with D O.
1
also a pleasing result. This diene is immiscible with D O
2
and homodimer 30 by H NMR (23% conversion, E/Z
at room temperature, but gives the soluble product 15
after highlyefficient RCM. We would propose that 11 is
of a sufficient polarityto seep into the resin pores to an
environment of high catalyst concentration, aided by
dissolution of the product in the solvent, which would
account for the high conversion observed under mild
conditions. This hydrophobic effect is currently under
investigation.
=100:0). Poor results were also obtained using other
electron deficient alkenes such as acrylic acid and acry-
lonitrile. It seems probable that the particularlyelectro-
philic alkylidene intermediates in these reactions are of
insufficient stabilityin nucleophilic solvents, resulting in
poor conversions and unselective CM.
It is interesting to note that a heterogeneous system
consisting of 1 in D O failed to promote either RCM or
2
Having established the activityof 2 and 7 in RCM our
CM of 8 or 21, respectively, under standard conditions.
This indicates that metathesis reactions promoted by 7
occur mainlyin the resin pores and not in the bulk sol-
vent. In summary, these findings demonstrate that a
water-insoluble ruthenium alkylidene moiety, when
attached to hydrophilic solid support, can exhibit con-
siderable metathesis activityin protic solvents. While
homo- and heterogeneous catalysts 2 and 7 have some
limitations, these catalysts can promote relatively effi-
cient RCM and CM of hitherto problematic substrates
in methanol and water without requiring either degas-
sing or an inert atmosphere, and as such mayhave
greater synthetic utility than benchmark water-soluble
catalysts 3 and 4.
1
9
attention then turned to CM (Scheme 3). As can be
seen from the results presented in Table 2, both cata-
lysts were active in protic solvents; the efficient forma-
tion of dimers 29–33 from simple hydroxy
functionalised olefins 20–24 in D O was possible using
2
heterogeneous catalyst 7. An inverse relationship
between substrate chain length and E/Z selectivitywas
also found; the E/Z content of homodimers 29, 30, and
3
3 were 14.6:1, 6.2:1 and 1.4:1, respectively.
Unfortunately, the high CM activity found using simple
alcohol substrates did not translate to the unprotected
a-C-glycoside 28. A moderate amount of homodimer 37

1876
S. J. Connon, S. Blechert / Bioorg. Med. Chem. Lett. 12 (2002) 1873–1876
Acknowledgements
12. Randl, S.; Buschmann, N.; Connon, S. J.; Blechert, S.
Synlett 2001, 10, 1547.
1
1
´
3. Savovicfa, J.; Dowden, J. Chem. Commun. 2001, 37.
4. For infomation on swelling chacteristics and a simple
We would like to thank the ‘Fonds der Chemischen
Industrie’ for financial support. S. J. C. thanks the
Graduiertenkolleg ‘Synthetische, mechanistische und
reaktionstechnische Aspekte von Metallkatalysatoren’
Kaiser test protocol see: Novabiochem Catalogue, 2000.
5. Synthesis of catalyst 7: PEGA-NH resin (1 g, 0.40 mmol)
1
2
was swollen in DMF for 1 h. In a separate flask, HOBt
307 mg, 2.0 mmol) and 5 (468 mg, 2.0 mmol) were dissolved in
(
(
2001) and the Alexander von Humboldt foundation
2002) for post-doctoral stipends.
(
3
DMF (12 cm ) and then DIC (252 mg, 2.0 mmol) was added.
The resulting solution was allowed to stand for 10 min and
then added to the resin via syringe. After gentle agitation for
1
1
2 h, the slurrywas filtered and washed well with DMF (7 ꢂ
References and Notes
3
0 cm ). This step was repeated until the beads gave a negative
Kaiser test. Next the solvent was exchanged byrepeated
washing with CH Cl . A solution of TEA (1.41 g, 14 mmol)
and DMAP (146 mg, 1.2 mmol) in CH
followed byacetic anh dy ride (1.22 g, 12 mmol) in CH
2
1
1
2
. Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett.
999, 1, 953.
. (a) Garber, S. B.; Kingsbury, J. S.; Gray, B. L.; Hoveyda,
2
2
3
Cl
2
(7 cm ) was added,
Cl
(3 cm ). The resulting mixture was gentlyagitated for 5 h at rt
2
2
3
A. H. J. Am. Chem. Soc. 2000, 122, 8168. (b) Gessler, S.;
Randl, S.; Blechert, S. Tetrahedron Lett. 2000, 41, 9973.
3
and then washed consecutivelywith CH
3
2
Cl
O (2 ꢂ 20 cm ), DMF (2 ꢂ 20 cm ),
2
(2 ꢂ 20 cm ),
3
3
3
Acc. Chem. Res. 2001, 34, 18. (b) Furstner, A. Angew. Chem.,
. (a) For recent reviews, see: Trnka, T. M.; Grubbs, R. H.
¨
DMF (2 ꢂ 20 cm ), H
2
3
3
CH
resulting ligand 6 could be stored indefinitelyat 4 C under N
2
Cl
2
(2 ꢂ 20 cm ) and finallyether (3 ꢂ 20 cm ). The
ꢀ
Int. Ed. Engl. 2000, 39, 3012. (c) Roy, R.; Das, S. J. Chem.
Commun. 2000, 519. (d) Philips, A. D. Aldrichima Acta 1999,
2
.
Coupling to ruthenium was carried out as follows: Ligand 6
was swollen in CH Cl for 1 h, filtered, and then suspended in
(4 cm ). To this was added 1 (441 mg, 0.52 mmol) and
CuCl (51.5 mg, 0.52 mmol) under N and the resulting red
32, 75. (e) Armstrong, S. K. J. Chem. Soc., Perkin Trans. 1
1998, 371. (f) Grubbs, R. H.; Chang, S. Tetrahedron 1998, 54,
4413. (g) Schuster, M.; Blechert, S. Angew. Chem., Int. Ed.
2
2
3
2 2
CH Cl
2
Engl. 1997, 36, 2036.
. Scholl, M.; Trnka, T. M.; Morgan, J. P.; Grubbs, R. H.
Tetrahedron Lett. 1999, 40, 2247.
. (a) Randl, S.; Gessler, S.; Wakamatsu, H.; Blechert, S.
Synlett 2001, 3, 430. (b) Imhof, S.; Randl, S.; Blechert, S.
Chem. Commun. 2001, 1692. (c) Chatterjee, A. K.; Morgan,
J. P.; Scholl, M.; Grubbs, R. H. J. Am. Chem. Soc. 2000, 122,
mixture heated under reflux for 4 h. The reaction mixture was
transferred to a separatoryfunnel and the dense phosphine
salts were removed. The solvent was then removed byfiltra-
4
5
tion and the resin washed with CH
colourless, and then again three times with Et
resin was dried under high vacuum and stored at 4 C.
2 2
Cl until the washings were
O. The green
ꢀ
2
¨
16. Furstner, A.; Thiel, O. R.; Ackermann, L.; Schanz, H. J.;
Nolan, S. P. J. Org. Chem. 2000, 65, 2204.
17. (a) Such isomerisation reactions have been observed
3
6
2
783.
. (a) Randl, S.; Connon, S. J.; Blechert, S. Chem. Commun.
001, 1796. (b) Choi, T. L.; Lee, C. W.; Chatterjee, A. K.;
¨
before: Furstner, A.; Liebl, M.; Lehmann, C.-W.; Picquet, M.;
Kunz, R.; Bruneau, C.; Touchard, D.; Dixneuf, P. H. Chem.
Grubbs, R. H. J. Am. Chem. Soc. 2001, 123, 10417.
¨
¸ etinkaya, B.; Demir, S.; Ozdemir,
7. (a) Novak, B. M.; Grubbs, R. H. J. Am. Chem. Soc. 1988,
110, 960. (b) Novak, B. M.; Grubbs, R. H. J. Am. Chem. Soc.
1988, 110, 7542. (c) Hillmeyer, M. A.; Lepetit, C.; McGrath,
Eur. J. 2000, 6, 1847. (b) C
I.; Toupet, L.; Semeril, D.; Bruneau, C.; Dixneuf, P. H. New.
J. Chem. 2001, 25, 519.
´
D. V.; Novak, B. M.; Grubbs, R. H. Macromolecules 1992, 25,
345. (d) Mortell, K. H.; Weatherman, R. V.; Kiessling, L. L.
J. Am. Chem. Soc. 1996, 118, 2297.
. For an example, see: Kanai, M.; Mortell, K. H.; Kiessling,
L. L. J. Am. Chem. Soc. 1997, 119, 9931.
18. In a stabilitytest, the decomposition of 2 in the absence of
ꢀ
substrate (CD OD, 45 C) was monitored by H NMR. Under
3
1
3
these conditions the alkylidene signal could still be observed
after 26 h.
19. General procedures for metathesis using 7 (note: quanti-
8
3
9
. (a) Mohr, B.; Lynn, D. M.; Grubbs, R. H. Organometallics
996, 15, 4317. (b) Lynn, D. M.; Mohr, B.; Henling, L. M.;
Day, M. W.; Grubbs, R. H. J. Am. Chem. Soc. 2000, 122,
tative catalyst loading is assumed): Solvent (2 cm ) was added
1
to 7 (50 mg, 0.02 mmol), and the resulting suspension agitated
gentlyfor 10 min in air. Substrate (0.4 mmol) was then added
and the mixture was stirred overnight in a closed vessel at
6
1
1
1
601.
ꢀ
0. Lynn, D. M.; Mohr, B.; Grubbs, R. H. J. Am. Chem. Soc.
998, 120, 1627.
1. Kirkland, T. A.; Lynn, D. M.; Grubbs, R. H. J. Org.
either room temperature or 45 C. The solvent was removed in
1
vacuo and conversion determined by H NMR. More con-
venientlyfor volatile substrates, deuterated solvents were used,
1
Chem. 1998, 63, 9904.
making direct H NMR analysis possible after filtration.