A convenient method for the efficient removal of ruthenium byproducts generated during olefin metathesis reactions.

Article abstract of DOI:10.1021/ol010045k

[reaction in text] An efficient method for removing ruthenium byproducts generated during olefin metathesis reactions with Grubbs catalysts is described. Treatment of the crude reaction products with triphenylphosphine oxide or dimethyl sulfoxide, followe

Full text of DOI:10.1021/ol010045k

ORGANIC
LETTERS
2001
Vol. 3, No. 9
1411-1413
A Convenient Method for the Efficient
Removal of Ruthenium Byproducts
Generated during Olefin Metathesis
Reactions
Yu Mi Ahn, KyoungLang Yang, and Gunda I. Georg*
Department of Medicinal Chemistry, UniVersity of Kansas, Lawrence, Kansas 66044
georg@ukans.edu
Received March 8, 2001
ABSTRACT
An efficient method for removing ruthenium byproducts generated during olefin metathesis reactions with Grubbs catalysts is described.
Treatment of the crude reaction products with triphenylphosphine oxide or dimethyl sulfoxide, followed by filtration through silica gel, was
found to be a practical and effective method to remove colored ruthenium byproducts.
Because of the availability of efficient ruthenium and
molybdenum catalysts, the olefin metathesis reaction has
emerged as a new and powerful method in organic chem-
istry.1 Grubbs’ ruthenium catalysts 1 and 2 have been used
often requiring multiple chromatographic purifications. In-
complete removal of the ruthenium byproducts is known to
cause complications such as double bond isomerization
during distillation or decomposition of the reaction products.3
Since metathesis reactions are expected to be used in
pharmaceutical processes, efficient and convenient methods
are also needed to remove potentially toxic ruthenium
byproducts from the reaction products.
The groups of Grubbs and Paquette have recently reported
efficient methods for the removal of ruthenium byproducts
from metathesis reactions.3,4 The Grubbs method relies
on removing the ruthenium byproducts as water-soluble
widely for ring-closing metathesis (RCM), cross-metathesis,
and ring-opening metathesis polymerization (ROMP) reac-
tions.2 One drawback of this method is the formation of
highly colored ruthenium byproducts during the reactions,
which are difficult to remove from the reaction products,
(2) (a) Schuster, M.; Blechert, S. Angew. Chem., Int. Ed. Engl. 1997,
36, 2037-2056. (b) Ivin, K. J.; Mol, I. C. Olefin Metathesis and Metathesis
Polymerization; Academic Press: San Diego, 1997. (c) Fu¨rstner, A.
Ruthenium-Catalyzed Metathesis Reactions in Organic Synthesis. In Topics
in Organometallic Chemistry. Alkene Metathesis in Organic Synthesis;
Fu¨rstner, A., Ed.; Springer: Berlin, 1998; pp 37-72. (d) Trnka, T. M.;
Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18-29. (e) Fu¨rstner, A. Angew.
Chem., Int. Ed. 2000, 39, 3012-3043.
(3) Maynard, H. D.; Grubbs, R. H. Tetrahedron Lett. 1999, 40, 4137-
4140.
(4) Paquette, L. A.; Schloss, J. D.; Efremov, I.; Fabris, F.; Gallou, F.;
Mendez-Andino, J.; Yang, J. Org. Lett. 2000, 2, 1259-1261.
(1) (a) Fu, G. C.; Nguyen, S. T.; Grubbs, R. H. J. Am. Chem. Soc. 1993,
115, 9856-9857. (b) Grubbs, R. H.; Chang, S. Tetrahedron 1998, 54, 4413-
4450. (c) Grubbs, R. H.; Miller, S. J.; Fu, G. C. Acc. Chem. Res. 1995, 28,
446-452. (d) Schrock, R. R. Olefin Metathesis by Well-Defined Complexes
of Molybdenum and Tungsten. In Topics in Organometallic Chemistry.
Alkene Metathesis in Organic Synthesis; Fu¨rstner, A., Ed.; Springer: Berlin,
1998; pp 1-36.
10.1021/ol010045k CCC: $20.00 © 2001 American Chemical Society
Published on Web 04/12/2001

ruthenium tris(hydroxymethyl)phosphine complexes,3 and the
procedure from the Paquette laboratory uses oxidation with
Pb(OAc)4 for metal byproducts removal.4 We herein report
yet another convenient and efficient method to remove
colored metal byproducts associated with RCM reactions
using the inexpensive reagents triphenylphosphine oxide
(Ph3PdO) or dimethyl sulfoxide (DMSO). Our studies
showed that colored ruthenium byproducts could be removed
effectively by treatment of the crude reaction mixtures with
Ph3PdO or DMSO followed by filtration through silica gel
or column chromatography on silica gel.5
5.0 mg of 5, respectively. The residual ruthenium levels
achieved with this method are very similar to those reported
by Grubbs (1.03 µg in 5.0 mg of 5)3 and Paquette (1.55 µg
in 5.0 mg of product).4
Similar results were obtained when we used catalyst 2 in
the RCM reaction of diethyl diallylmalonate (3).
To find the optimal reaction time needed for effective
purification using the Ph3PdO/chromatography protocol,
individual reactions were examined and measurements of the
ruthenium levels in purified product 5 were made (Table 2).
Diethyl diallylmalonate (3) was chosen as the substrate
to examine the novel purification method (Scheme 1 and
Table 2. Ruthenium Levels of 5 (µg/5 mg) after Different
Exposure Times of Crude Reaction Product 4 to 50 equiv of
Ph3PdO at Room Temperature, Followed by Column
Chromatography on Silica Gel
Scheme 1
entry
exposure time (h)
ruthenium (µg/5 mg)
1
2
3
4
5
6
1
2
4
8
12
24
16.4 ( 0.16
13.5 ( 0.21
10.2 ( 0.15
2.30 ( 0.02
1.20 ( 0.01
1.27 ( 0.01
Table 1).3 RCM of 3 using ruthenium catalyst 1 provided
the crude reaction product 4, which was subsequently treated
with Ph3PdO or DMSO for a minimum of 12 h, followed
by column chromatography to provide purified 5 as a
colorless oil (Scheme 1). We examined the ruthenium levels
of the purified products after adding increasing amounts of
Ph3PdO or DMSO (5, 10, 20, 50, and 100 equiv relative to
1) to the crude reaction mixture 4 (Table 1). The ruthenium
levels in the purified RCM product 5 were analyzed by
inductively coupled plasma mass spectrometry (ICP-MS).6
As shown in Table 1, the ruthenium levels decreased with
increasing amounts of the additives Ph3PdO and DMSO.
In both cases, optimal results were observed when 50 equiv
of Ph3PdO or DMSO relative to catalyst 1 was added to
the crude product mixture (entry 5). Under those conditions,
the ruthenium levels were reduced to 1.2 µg and 1.8 µg in
We found that efficient removal of ruthenium byproducts
was achieved with a minimum treatment time of 8 h (entry
4). However, optimum results were achieved after 12 h (entry
5).
We next examined the novel purification method with
different substrates (Scheme 2) under the conditions that have
Scheme 2
Table 1. Ruthenium Levels of 5 (µg/5 mg) after Exposure of
Crude Product 4 to Different Amounts of Ph3PdO and DMSO
for at Least 12 h at Room Temperature Followed by Column
Chromatography on Silica Gel5
Ph3PO
(equiv)a
ruthenium
DMSO
(equiv)a
ruthenium
entry
(µg/5 mg)
(µg/5 mg)
1
2
3
4
5
6
0b
5c
10
20
50
59.7 ( 0.50
2.77 ( 0.02
2.11 ( 0.02
1.85 ( 0.02
1.20 ( 0.01
1.91 ( 0.01
0b
5c
10
20
50
59.7 ( 0.50
6.99 ( 0.02
5.43 ( 0.02
2.60 ( 0.02
1.81 ( 0.02
1.34 ( 0.02
100
100
been reported for the synthesis of 7, 9, and 11.7-9 The RCM
transformations were carried out under argon. Once the RCM
reactions were completed (monitored by TLC), Ph3PdO (50
equiv relative to 1) was introduced into the reaction mixture
and the solutions were stirred at room temperature overnight,
while maintaining the argon atmosphere. Subsequent column
a Equivalent of Ph3PO or DMSO relative to 1. b The ruthenium content
of the crude product was determined without purification by column
chromatography. After digestion of the crude sample, insoluble material
was filtered off, which may account for the observed decreased amount of
ruthenium compared to the theoretical amount of ruthenium at 5 mol %
(90 µg/5 mg of product). c The addition of 5 equiv did not remove all colored
products.
1412
Org. Lett., Vol. 3, No. 9, 2001

chromatography on silica gel (2 g/0.01 mmol of catalyst,
eluent ) hexanes:ethyl acetate ) 4:1) gave a clear, colorless
liquid or a white solid in the reported yields (Scheme 2).7-9
In all three cases, the colored byproducts were successfully
removed from the products [ICP-MS data: 7 (0.5 µg/5 mg),
9 (1.1 µg/5 mg), and 11 (2.1 µg/5 mg)].
In summary, our results demonstrate that Ph3PdO and
DMSO can be used effectively to remove the colored
ruthenium byproducts of ruthenium catalysts 1 and 2, which
are formed after RCM, when followed by simple filtration
through silica gel or column chromatography. With this
method, residual ruthenium levels from metathesis reactions
are lowered to approximately 1-2 µg/5 mg of product. The
advantage of the method is that Ph3PdO and DMSO are
very stable and inexpensive reagents. Both reagents have
limited chemical reactivity and should therefore be compat-
ible with a wide variety of functional groups.
(5) Procedure for RCM of 3 and purification of crude product 4
with triphenylphosphine oxide or dimethyl sulfoxide: A solution of
diethyl diallylmalonate (3, 120 mg, 0.5 mmol) in CH2Cl2 (100 mL) was
degassed with argon for 5 min and then catalyst 1 (20.5 mg, 5 mol %) was
added to the solution. The reaction flask was sealed with a rubber septum.
After the reaction was complete (approximately after 2 h at rt, monitored
by TLC), the reaction mixture was treated with Ph3PdO (350 mg, 50 equiv,
relative to catalyst 1) or DMSO (90 µL, 50 equiv, relative to catalyst 1)
overnight. The solution was concentrated in vacuo and purified by column
chromatography on silica gel (2 g/0.01 mmol of catalyst 1) using hexanes:
EtOAc ) 8:1 as the eluent to provide the RCM product 5 in 95-98% yield.
(6) Samples (approximately 5-10 mg) were accurately weighed, digested
with distilled 6 N nitric acid overnight, and diluted to a total volume of
approximately 100 mL with the distilled acid and distilled deionized water
to a final acid concentration of about 2%. Samples were analyzed on a
PQII+XS inductively coupled plasma mass spectrometer (ICP-MS). Masses
96, 98, 99, 100, 101, 102, and 104 were monitored in all samples and
standards. Drift (corrected according to the method by Cheatham, M. M.;
Sangrey, W. F.; White, W. M. Spectrochim. Acta, Part B 1993, 48B, E487-
E506) was usually less than 5% and always less than 10% during a single
run. The reported concentration is the average of the total Ru concentration
as calculated from each of the individual isotopes after drift correction.
The average usually had a percent relative standard deviation (%RSD) of
less than 2% and always less than 4%.
Acknowledgment. We gratefully acknowledge financial
support for this work by the National Institutes of Health
(CA84173). We wish to thank Dr. Apurba Datta and Dr.
Paul R. Hanson for helpful discussions. We also thank Dr.
Paul R. Hanson for providing RCM substrate 10 in Scheme
2. We are grateful to Professor Gwendolyn L. Macpherson,
Director of the Plasma Analytical Laboratory at the Univer-
sity of Kansas, for carrying out the ICP-MS measurements.
OL010045K
(8) Georg, G. I.; Ahn, Y. M.; Blackman, B.; Farokhi, F.; Flaherty, P.
T.; Mossman, C. J.; Roy, S.; Yang, K. Chem. Commun. 2001, 255-256.
(9) Dougherty, J. M.; Probst, D. A.; Robinson, R. E.; Moore, J. D.; Klein,
T. A.; Snelgrove, K. A.; Hanson, P. R. Tetrahedron 2000, 56, 9781-9790.
(7) Ackermann, L.; Fu¨rstner, A.; Weskamp, T.; Kohl, F. J.; Herrmann,
W. A. Tetrahedron Lett. 1999, 40, 4787-4790.
Org. Lett., Vol. 3, No. 9, 2001
1413