A rapid and simple cleanup procedure for metathesis reactions

Article abstract of DOI:10.1021/ol0631399

Figure presented A new method for easy removal of ruthenium from metathesis reactions by using a polar isocyanide is reported. This protocol removed most ruthenium byproducts from a variety of synthetically useful metatheses. Moreover, the isocyanide-promoted carbene insertion results in rapid destruction of carbene reactivity, demonstrated in the commonly used first- and second-generation Grubbs' carbenes.

Full text of DOI:10.1021/ol0631399

ORGANIC
LETTERS
2007
Vol. 9, No. 7
1203-1206
A Rapid and Simple Cleanup Procedure
for Metathesis Reactions
Brandon R. Galan, Kyle P. Kalbarczyk, Steven Szczepankiewicz,†
Jerome B. Keister, and Steven T. Diver*
Department of Chemistry, UniVersity at Buffalo, The State UniVersity of New York,
Amherst, New York 14260
diVer@buffalo.edu
Received December 28, 2006
ABSTRACT
A new method for easy removal of ruthenium from metathesis reactions by using a polar isocyanide is reported. This protocol removed most
ruthenium byproducts from a variety of synthetically useful metatheses. Moreover, the isocyanide-promoted carbene insertion results in rapid
destruction of carbene reactivity, demonstrated in the commonly used first- and second-generation Grubbs’ carbenes.
Olefin metathesis has become an important method for
carbon-carbon double bond construction in organic syn-
thesis.1 Metathesis is widely employed in a variety of fields
of chemistry, using the mild, functional group tolerant
ruthenium carbenes discovered by Grubbs.2 Metathesis has
been used in large-scale reactions and continued use in
pharmaceutical, small molecule, and parallel synthesis is
immediately foreseeable. Despite the widespread use of
metathesis procedures, a difficulty common to all metathesis
reactions is the removal of the ruthenium at the end of the
reaction. In this report, we detail a new procedure using a
polar isocyanide that accomplishes the rapid quenching of
carbene activity and leads to the ready removal of the
resulting polar coordination complex (Scheme 1). This
represents a practical, simple-to-use procedure for the
removal of ruthenium in the purification of the organic
products of metathesis. The protocol works efficiently for
all commonly used carbene catalysts in a variety of alkene
and enyne metathesis applications.
Several methods for the removal of ruthenium have been
reported. First, Grubbs and Maynard described the use of a
water-soluble phosphine for the removal of ruthenium in the
first generation complex 1.3a This requires 86 equiv of a
water-soluble phosphine and extractive workup. The reported
large-scale procedure used a 12 h treatment.3b Next, Georg
et al.3c reported a mild oxidative procedure to convert 1 into
a polar, undefined product that can be removed by chroma-
tography after 24 h of treatment with 50 equiv of Ph3PdO
or DMSO. This procedure is mild and is widely used. Third,
Paquette et al.3d disclosed a rapid oxidative procedure using
Pb(OAc)4 that oxidizes the ruthenium carbene to polar,
undefined products that are then removed by silica gel
Scheme 1. Isocyanide-Promoted Ligand Insertion: An
Effective Cleanup Protocol
† Present address: Department of Chemistry, Canisius College, Buffalo,
NY.
(1) Grubbs, R. H., Ed. Handbook of Metathesis; Wiley-VCH: Weinheim,
Germany, 2003; Vols. 1-3.
(2) (a) Nguyen, S. T.; Johnson, L. K.; Grubbs, R. H.; Ziller, J. W. J.
Am. Chem. Soc. 1992, 114, 3974-3975. Complex 1: (b) Schwab, P.;
Grubbs, R. H.; Ziller, J. W. J. Am. Chem. Soc. 1996, 118, 100-110.
Complex 2: (c) Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett.
1999, 1, 953-956. (d) Trnka, T. M.; Morgan, J. P.; Sanford, M. S.; Wilhelm,
T. E.; Scholl, M.; Choi, T.-L.; Ding, S.; Day, M. W.; Grubbs, R. H. J. Am.
Chem. Soc. 2003, 125, 2546-2558.
10.1021/ol0631399 CCC: $37.00
© 2007 American Chemical Society
Published on Web 02/28/2007

chromatography. During our investigation, Crudden et al.3e
reported a mesoporous silicate for the removal of ruthenium
applied to 1 and homogeneous catalysis by ruthenium
hydrides. This requires a large excess of the silicate but is
simple since only a filtration is required. For synthetic
chemists, the most widely used procedure is that of Georg
et al.,3c yet this suffers from lengthy reaction times.
There are several shortcomings of the reported procedures.
First, from an operational standpoint, they take too long: for
a 1-4 h cross enyne metathesis, stirring 12-24 h with an
oxidant is necessary. This slow oxidation has not been shown
to immediately stop metatheses, which is essential for kinetic
analysis4 and to prevent unwanted, secondary metatheses.5
Second, the oxidative procedures reported previously were
shown to be effective for complexes 1 or 2, but not all
metathesis catalysts. For example, phosphine-free initiators,
which are increasingly used in challenging applications, have
not been evaluated in any cleanup protocol. Third, though
the reported procedures remove ruthenium, the specific
reaction is undefined. Recognizing some of the current trends
in metathesis chemistry, we felt that a full range of cross
metatheses should be evaluated. A new procedure would be
desirable if it was fast, efficient for the range of carbene
complexes used, applicable to a variety of metatheses, and
proceeded by known chemistry.
it to evaluate the effectiveness of the quenching method (eq
3). Formation of the formamide and dehydration with POCl3
gave high chemical yield of the ethyl ester 5. Saponification
with KOH (THF-water) and evaporation of the solvent
mixture deposited an odorless8 off-white amorphous solid
(mp 164-165 °C) that was used for quenching. The solid
was not hygroscopic and could be handled in the atmosphere
during weighing. The effectiveness of the quench was
evaluated in a cross enyne metathesis between 1-hexene and
1-benzoyloxy-2-propyne. At about 50% conversion, 44 mol
% isocyanide 5 was added (8.8 equiv based on catalyst;
solution in ca. 1 mL of toluene).9 The reaction immediately
changed yellow with complex 2.10 Further conversion was
not observed, suggesting that catalysis was rapidly stopped.
The effectiveness of the isocyanide quench and optimiza-
tion of the quenching procedure was evaluated in a typical
ring-closing metathesis (RCM). Similar to previous inves-
tigations,3 we chose an inductively coupled plasma (ICP)
analytical method to quantitate ruthenium content in crude
and purified metathesis samples.11 These data were used to
determine the efficiency of the cleanup procedure. The
theoretical amount of ruthenium that should be left at the
end of the reaction is 100 µg/5 mg of sample. We found an
average of 97 µg/5 mg of sample after a 2 h reaction time
using complex 2 (Table 1, entry 1). If the crude reaction
was passed through silica gel, about half the ruthenium was
removed (entry 2). In Table 1, the third column reports the
concentration of ruthenium determined in the assay for a
diluted sample. The last column accounts for the mass of
the sample digested. In the previous literature, concentration
of ruthenium is reported as “µg of Ru found per 5 mg of
crude sample”. The effectiveness of the cleanup should be
evident from the last column; comparisons to literature
methods can also be made from these data.12
We previously observed a unique ligand insertion pro-
moted by carbon monoxide.6 This reaction resulted in
destruction of carbene activity and formed a moderately polar
bis(carbonyl)ruthenium(II) complex. For carbene 2, we
observed Buchner product 3 arising from benzylidene
insertion into a mesityl group (Scheme 2). We observed the
Scheme 2. Donor-Induced Ligand Insertion (Galan et al., Ref 6)
Next, the stoichiometry of the isocyanide addition was
studied. It was expected that at least 2 equiv would be
required based on ruthenium carbene. Each of these runs
same ligand insertion shown in eq 2 using aryl isocyanides.
We reasoned that a polar isocyanide would rapidly quench
the carbene through ligand insertion and confer polarity to
the resulting ruthenium coordination complex. Polarity of
the complex would then permit removal of ruthenium from
the organic products of a typical metathesis reaction. This
process represents a “well-defined” reaction of the no-longer-
wanted carbene catalyst.
(4) For fast enyne metathesis, a rapid and irreversible quench is desirable.
Kinetic studies in enyne metathesis led us to develop the technique described
in this paper.
(5) If a small amount of active catalyst remains during solvent evaporation
or during crude sample storage, the product can undergo carbene-promoted
polymerization (e.g., for a RCM, the reverse reaction: ROMP).
(6) Galan, B.; Gembicky, M.; Dominiak, P. M.; Keister, J. B; Diver, S.
T. J. Am. Chem. Soc. 2005, 127, 15702-15703.
(7) Obrecht, R.; Hermann, R.; Ugi, I. Synthesis 1985, 400-402.
(8) We typically run the metatheses and the treatment procedure in the
fume hood. Because of its excellent ligand properties, the isocyanide 6 is
assumed to be toxic. Due care should be used in handling the solid to avoid
breathing any fine particulates.
(9) IR was used to follow alkyne consumption, and methanolic solutions
obscured the alkyne CH stretch. As a result, a toluene solution of ester 5
was used instead of carboxylate 6.
A polar isocyanide was prepared and used to quench a
cross metathesis. We prepared the isocyanide salt 67 and used
(3) (a) Maynard, H. D.; Grubbs, R. H. Tetrahedron Lett. 1999, 40, 4137-
4140. (b) Grubbs, R. H. Org. React. 2004, 22, 123-1234. (c) Ahn, Y. M.;
Yang, K.; Georg, G. I. Org. Lett. 2001, 3, 1411-1414. (d) Paquette, L. A.;
Schloss, J. D.; Efremov, I.; Fabris, F.; Gallou, F.; Mendez-Andino, J.; Yang,
J. Org. Lett. 2000, 2, 1259-1261. (e) McEleney, K.; Allen, D. P.; Holliday,
A. E.; Crudden, C. M. Org. Lett. 2006, 8, 2663-2666. Summary of
methods: (f) Conrad, J. C.; Fogg, D. E. Curr. Org. Chem. 2006, 10, 185-
202.
(10) With the complex 1, the solution turned blue-green.
(11) In this study, we quantitated total ruthenium content in the plasma
and did not use MS to separate isotopes.
(12) 1 µg of Ru detected in a 5 mg sample corresponds to 200 ppm Ru.
1204
Org. Lett., Vol. 9, No. 7, 2007

Scheme 3. Application of the Removal Method to Other
Table 1. Removal of Ruthenium Byproducts from a Typical
Ring-Closing Metathesis Reaction
Metathesis Reactions
time/ concn [Ru]/ concn [Ru]/
entry
1
treatmenta
h
ppm
16.1, 19.91 86.4, 108
(av ) 97)
µg/5 mgb,c
NT
2
NT, chrom.
6.45, 6.25 44.0, 43.8
(av ) 44)
3
4
5
2.2 equiv (11 mol %) 2 0.5 2.21
4.4 equiv (22 mol %) 2 0.5 1.35
16.2
8.31
4.4 equiv (22 mol %) 2
1
1.15, 1.55 10.9, 10.3
(av ) 10.6)
24.2
0.5 0.66, 0.69 4.61, 5.70
4.4 equiv (22 mol %) 1 0.5 0.67 4.87
6
7
8d
8.8 equiv (44 mol %) 2 0.5 3.46
10 equiv (50 mol %) 2
a Treatment to the crude sample. Equivalents are based on ruthenium
carbene used; mole percents are based on malonate reactant. b Ruthenium
concentration in sample; all assayed values are 0.1-100 ppm in the
analytical assay. c µg Ru/5 mg sample, calculated for the original sample
from quantitation of ruthenium by ICP determination. d Complex 1 (Ru gen-
1) was used in this run.
6-8). The enyne metathesis was performed with both
ethylene and 1-hexene, and gave less than 1 µg of Ru/5 mg
of sample in these cases after only 0.5 h of isocyanide
treatment. Last, we investigated a methylene-free ring
synthesis (eq 8) and found that the cleanup procedure was
effective in this case. Yields in all cases were comparable
to those previously reported. Overall, the crude products of
intermolecular metatheses using the commonly used second
generation Grubbs carbene complex 2 were effectively
cleaned up by the isocyanide treatment just before routine
purification. This quenches carbene activity (vide supra) and
obviates the need for long waiting times before purification.
This is an improvement in efficiency that should prove useful
in practical and large-scale metathesis applications.
The effectiveness of the isocyanide cleanup procedure was
next evaluated for several commonly used metathesis
catalysts (Table 2). These experiments were performed by
using the RCM of diallyl malonate and quenched after 2 h.
The isocyanide cleanup procedure was effective for met-
atheses conducted with the first generation Grubbs carbenes.
In the first generation complexes, the ancilliary ligand is
Cy3P, which contains no aromatic moieties. In contrast, the
aromatic mesityl groups are found in the second generation
carbene complexes and account for the Buchner insertion
chemistry described in eq 2.6 There is still wide use of the
first generation complex, both as the Grubbs benzylidene 1
and as the Hoveyda chelate 7. We applied the cleanup
procedure to these complexes and found that most of the
ruthenium was removed (entries 1-5, Table 2). Though the
details of this process are still unclear, it is presumed that a
polar coordination complex is formed by application of the
standard procedure.15 Recently, the phosphine-free second
generation carbene complexes 8-10 have become popular
(entries 3-8) was performed by addition of the isocyanide
after 2 h of reaction time and stirring for the time indicated,
followed by plug filtration on silica gel. On addition of the
isocyanide, the dark color of the reaction discharged. With
a 0.5 h reaction time, 4.4-8.8 equiv of isocyanide proved
effective. There was no advantage to longer quench periods.
In the last entry, ruthenium in the first generation complex
1 was removed by the isocyanide treatment.
Next, the cleanup protocol13 was examined for several
different metatheses. Here we wanted to be sure the
optimized procedure could be used reliably for typical
metathesis reactions that are used by synthetic chemists. In
the first case, a cross metathesis of the MeBMT side chain
of cyclosporin and an alkene bearing an active ester was
examined at high catalyst loading (eq 5, Scheme 3).14 After
treatment (4.4 equiv, 22 mol % of 6; 30 min), the product
was isolated in 84% yield with only 0.96 µg of Ru/5 mg of
sample. This is noteworthy due to the potential of the cyclic
undecapeptide to sequester the metal and because the side
chain active ester withstands the quench procedure. The
cleanup was effective for cross alkene and enyne metatheses,
using the standard treatment of 4.4 equiv of 6 for 0.5 h (eqs
(13) Between 4.4 and 8.8 equiv of the isocyanide is used, based on
ruthenium carbene (22-44 mol % based on unsaturated reactant). Repre-
sentative Experimental Procedure (eq 7): Into an oven-dried 100 mL
Schlenk flask were placed 177 mg (1.02 mmol) of butynyl benzoate in 5.0
mL of CH2Cl2, 0.8 mL (6.44 mmol, 6.2 equiv) of 1-hexene, and 43.3 mg
of Grubbs’ complex 2 (0.051 mmol, 5 mol %). The reaction was stirred at
rt for 2 h. After this time, the mixture was treated with a solution of 38.8
mg of 6 (0.315 mmol, 31 mol % or 6.2 equiv based on 2) in 1.0 mL of
methanol. The color of the solution changed from purple to yellow within
5 min. The reaction mixture was stirred for 30 min at rt, concentrated in
vacuo (rotary evaporator), and purified by column chromatography on
approximately 10 g of silica gel (2.0 g of silica gel was used for every 0.01
mmol of catalyst used), using 1:8 ethyl acetate-hexanes as the eluent.
(14) Smulik, J. A.; Diver, S. T.; Pan, F.; Liu, J. Org. Lett. 2002, 4, 2051-
2054.
(15) Further studies are in progress.
Org. Lett., Vol. 9, No. 7, 2007
1205

Table 2. Results of Isocyanide Cleanup Procedure Applied to
RCM of Diethyl Diallylmalonate, Using Additional Metathesis
Catalysts
isocyanide
temp,
time
concn [Ru]/
entrya
Ru carbene
1, Ru gen-1
1, Ru gen-1
1, Ru gen-1
1, Ru gen-1
7, Hov gen-1
8, Hov gen-2
9, Grela gen-2
6b,c
µg/5 mgd
1
2
3
4
5
6
7
8
9
4.4 equiv rt, 0.5 h
8.8 equiv reflux, 0.5 h
8.8 equiv 6 h
4.87
0.8
1.2
1.1
2.05
1.16
0.8
0.6
11
8.8 equiv 12 h
4.4 equiv 0.5 h
4.4 equiv 0.5 h
8.8 equiv 0.5 h
10, Grubbs pyr solv 8.8 equiv 0.5 h
11, vinyl carbene 4.4 equiv 0.5 h
a RCM of eq 4, using carbene at 5 mol % loading. b Isocyanide 6 added
as a solution in methanol. c Equivalents are based on the carbene complex,
e.g., 4.4 equiv of 6 is 22 mol %. d Concentration expressed as µg/5 mg of
organic product: e.g., 1.1 µg/5 mg of sample is 220 ppm.
Figure 1. ORTEP drawing of isocyanide insertion complex 13.
Thermal ellipsoids are drawn at 50% probability. See the Supporting
Information for bond angles and bond lengths.
choices for metathesis. As expected, the cleanup/quench
protocol was effective for these other second generation
carbene complexes (entries 6-8), including a vinyl carbene
complex 11.
and a benzylic methine appeared at δ 4.91 ppm, supporting
the cycloheptatrienyl substructure. The 31P NMR spectrum
showed a single resonance at δ 36.3 ppm.17 The similar
spectroscopic properties support the conclusion that the polar
isocyanide promotes a ligand insertion as observed in
complex 13. In terms of the cleanup protocol, the polar
isocyanide ligands confer greater polarity to the hexacoor-
dinate ruthenium complex, facilitating its removal from the
organic products of metathesis. The ligand insertion of eq 9
also results in loss of the metal carbene fragment, which
switches off the catalyst’s metathesis activity.
In conclusion, we have demonstrated a useful cleanup
procedure for metathesis reactions. The isocyanide addition
to a crude metathesis reaction mixture results in “quench”
of the metal carbene by triggering a bond insertion into the
mesityl group. We demonstrated that the quench is effective
for a variety of metatheses and for all of the commonly used
Grubbs’ carbene complexes. The carboxylate salt of isocya-
nide 6 is an odorless solid that is easily obtained and easily
handled. Further study of the insertion processes promoted
by strongly coordinating ligands is under active study in our
group.
The product resulting from a representative isocyanide-
promoted ligand insertion was structurally identified. The
second generation carbene complex 2 was reacted with 2.2
equiv of p-chlorophenyl isocyanide 12 to produce a crystal-
line complex 13, eq 9. X-ray analysis revealed two molecules
of isocyanide 12 had added to the ruthenium center and the
benzylidene had formally inserted into one of the mesityl
groups (Figure 1).16 The observed chemistry is analogous
to our earlier findings with carbon monoxide.6 The cyclo-
heptatrienyl protons of 13 were found as singlets at δ 5.61
and 5.41 ppm, with a benzylic methine appearing at 5.04.
The 31P NMR showed a single resonance at δ 13.9 ppm.
With this structural data in hand, we wanted to know if the
same insertion was triggered by the polar isocyanide 6. The
resulting complex did not provide X-ray quality crystals.
Solution NMR studies were performed on the triphenylphos-
phine complex, (H2IMes)(Ph3P)Cl2RudCHPh, reacted with
2.7 equiv of the polar carboxylate 6 (1:1 v/v CD2Cl2-CD3-
OD). Proton resonances were observed at δ 5.80, 5.79 ppm,
Acknowledgment. The authors thank the Petroleum
Research Fund (PRF AC-44202) and the NSF (CHE-601206)
for financial support of this work. We also thank Dr. Cara
L. Nygren (SUNY at Buffalo) for obtaining the solid-state
structure of 13 and Materia (Pasadena, CA) and Boeringher-
Ingelheim for their generous catalyst support.
(16) X-ray diffraction data were collected on a Bruker SMART APEX2
diffractometer with Mo KR (λ ) 0.71073 Å) radiation, using a colorless
crystal with the dimensions of 0.40 × 0.13 × 0.13 mm3. Intensity data
were integrated with SAINTPLUS program. The crystal structure was solved
by direct methods with SHELXTL NT version 6.14, and refined by full
matrix least-squares against F2. Non-hydrogen atoms were refined aniso-
tropically. Positions of hydrogen atoms were found by difference electron
density Fourier synthesis. Crystal data for 13: C60H73N4Cl4PRu(CH2Cl2)4,
mp 152-154 °C, M ) 1457.83, triclinic, space group P1h, a ) 12.6750(3)
Å, b ) 14.9214(4) Å, c ) 19.4499(5) Å, V ) 3504.87(15) Å3, Z ) 2, T )
90(1) K, Dc ) 1.381 g‚cm-3, 2θmax ) 26.4°, R(F) ) 3.6% for 62709
reflections with Fo > 2σ(Fo) and Rw(F2) ) 0.263 for all 17343 independent
reflections, 802 parameters. GOF(F2) ) 1.02.
Supporting Information Available: Detailed experi-
mental procedures and structure determination for 13. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(17) The same experiment was performed on the Cy3P complex of 2
(Grubbs’ Ru gen-2), see the Supporting Information for details.
OL0631399
1206
Org. Lett., Vol. 9, No. 7, 2007