Occlusion of Grubbs' catalysts in active membranes of polydimethylsiloxane: Catalysis in water and new functional group selectivities

Article abstract of DOI:10.1021/ja0642212

The Grubbs' first and second generation catalysts were occluded into cross-linked slabs of polydimethylsiloxane with volumes from 1 mm³ to 1 cm³ by swelling the polymer with catalyst and methylene chloride. Methylene chloride was evaporated under vacuum to yield occluded catalysts where their solvent was polydimethylsiloxane. These occluded catalysts were reacted with alkenes dissolved in H₂O or H₂O/MeOH mixtures that diffused into the polydimethylsiloxane to react by ring-closing metathesis and cross metathesis. Control experiments revealed that the catalysts remained occluded and metathesis did not occur in the solvent. Occlusion of these catalysts allowed commercially available Grubbs' catalysts to be used with H₂O as the solvent while isolating the H₂O sensitive ruthenium methylidene from exposure to H₂O. Functional group selective experiments were carried out where the polydimethylsiloxane was an "active" membrane to exclude salts. Polydimethylsiloxane is a hydrophobic polymer, so the deprotonated salt of diallylmalonic acid did not diffuse into it while a diallylether diffused into it and reacted by metathesis. Thus, by controlling the polarity of reagents their reactivity can be controlled owing to the properties of polydimethylsiloxane rather than those of the Grubbs' catalysts. Occlusion of catalysts in polydimethylsiloxane has been shown to add new selectivities to mature catalysts. Copyright

Full text of DOI:10.1021/ja0642212

Published on Web 10/26/2006
Occlusion of Grubbs’ Catalysts in Active Membranes of
Polydimethylsiloxane: Catalysis in Water and New Functional Group
Selectivities
Martin T. Mwangi, M. Brett Runge, and Ned B. Bowden*
UniVersity of Iowa, Department of Chemistry, 423K Chemistry Building, Iowa City, Iowa 52242
Received June 15, 2006; E-mail: ned-bowden@uiowa.edu
Homogeneous catalysts exhibit terrific selectivities, and their
reactivities can be readily tuned by altering the ligand or metal,
but these catalysts are often difficult to remove after a reaction is
1
complete and can be sensitive to solvents. Heterogeneous catalysts
are easily removed from a reaction medium and are easy to recycle,
2
but their active sites are challenging to study and alter. Although
designing catalysts that combine the best aspects of homogeneous
and heterogeneous catalysis is an active area of research, it has
Figure 1. How the PDMS slabs were occluded with catalysts. A PDMS
met with limited success. Typically, homogeneous catalysts are
attached to surfaces or polymers which lower their selectivities or
require additional synthetic steps that limit their applications.3
In this communication we report a new method to make
homogeneous catalysts heterogeneous without requiring additional
synthesis on the catalyst. We developed a new method to occlude
the Grubbs’ first and second generation catalysts in a slab of
polydimethylsiloxane (PDMS) such that the reactions they catalyze
occur within the slab.4 The idea is simple: reagents dissolve in
an aqueous solvent and diffuse into PDMS slabs to react with
slab was swelled in CH2Cl2 with the Grubbs’ first generation catalyst. The
Grubbs’ catalyst diffused into the PDMS slab with CH2Cl2. The CH2Cl2
was removed under vacuum and the PDMS slabs were briefly rinsed. The
PDMS slabs were cut into mm-sized pieces and used in our reactions.
such that there is only one report of cross metathesis with a
7
homogeneous metathesis catalyst in water. Ring-closing metathesis
reactions are easier to catalyze than cross metathesis reactions, yet
there are only a few reports of these reactions catalyzed by
homogeneous catalysts in aqueous solvents.7
,5
,9
6
occluded catalysts. Occluded catalysts are insoluble in the aqueous
We addressed each of these limitations in our work. We used
commercially available catalysts without modifying their structures
or adding extra synthetic steps. In addition, we used aqueous
solvents for reagents but the catalysts remained occluded in PDMS
solvents used for reagents so they do not diffuse out of the PDMS
slab. This method is exciting as it offers a chance to combine
homogeneous and heterogeneous catalysis without modifying the
catalysts, and the PDMS slab can function as an “active membrane”
to exclude polar reagents to alter the reactivity of occluded catalysts.
Occluded PDMS slabs were readily prepared from commercially
available, uncross-linked PDMS (Figure 1). We first cured PDMS
by mixing its two components and heating at 65 °C for at least 12
1
0
and were exposed to only minute amounts of water. Thus, the
ruthenium methylidene was stable, and we were able to carry out
these reactions at temperatures up to 100 °C with low catalyst
loadings. Our method solves the problems of not having to modify
these catalysts while still using water as the solvent for olefin
metathesis at low catalyst loadings. In addition, we showed that
these catalysts could be recycled (see Supporting Information).
We believe these reactions occur in PDMS rather than the
solvents because the catalysts are not soluble in aqueous solvents
and the PDMS remained colored throughout the reaction while the
solvent remained clear. To prove that the Grubbs’ catalysts remained
occluded in PDMS and metathesis did not occur in the solvent, we
removed an aliquot of the reaction when the cross metathesis of
h. These PDMS slabs were swelled in three batches of CH
h to remove uncross-linked PDMS and Pt used as a cross-linking
catalyst. The PDMS slabs were dried under vacuum, and a solution
of the Grubbs’ catalyst dissolved in CH Cl was added to the PDMS
slabs under an atmosphere of N . CH Cl swelled the PDMS slabs
and allowed the Grubbs’ catalyst to diffuse into it. After removal
of CH Cl under vacuum, rinsing, and cutting the slabs into mm-
2 2
Cl for
4
2
2
2
2
2
2
2
sized pieces, the occluded catalysts were ready.
We carried out several ring-closing and cross metathesis reactions
CH
aliquot was placed in a Schlenk flask under N
The reaction did not proceed any further in this control experiment,
2
dCH(CH
2
)
8
CO
2
H was at approximately 50% completion. The
in H
2
O to MeOH/H
2
O mixtures as shown in Table 1. Conversions
2
and heated as before.
1
of these reactions were all 100% by H NMR spectroscopy, and
isolated yields were high.
but in the Schlenk flask with the remaining CH
2
dCH(CH
2
8 2
) CO H
1
This work is important since water was used as a green solvent
for metathesis reactions with commercially available Grubbs’
catalysts. Olefin metathesis in water is limited by the lack of
solubilities of commercially available Grubbs’ catalysts and by the
low stabilities of ruthenium methylidenes. The first problem was
partly addressed by Grubbs through the synthesis of water soluble
ruthenium metathesis catalysts, but this catalyst requires extra
exposed to occluded catalysts the conversion reached 100% by H
NMR spectroscopy. We repeated these control experiments and
received the same result each time: the solvent did not contain
catalysts and the metathesis reactions were occurring in PDMS.
Occlusion of metathesis catalysts provides a new opportunity to
carry out functional group selective reactions since PDMS can be
an “active” membrane. PDMS is very hydrophobic and although
most organic reagents will diffuse into it, salts are poorly soluble.
This difference in solubility of salts and organic reagents provides
a chance to carry out functional group selective reactions as shown
in Figure 2. Two dienes that readily undergo ring-closing metathesis
7
synthetic steps that limit its potential uses. The low stability of
the ruthenium methylidenes that are generated in the catalytic cycle
8
is a problem that is inherent to these catalysts. These limitations
have hindered the development of metathesis reactions in water
14434
9
J. AM. CHEM. SOC. 2006, 128, 14434-14435
10.1021/ja0642212 CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Metathesis Reactions with Occluded Grubbs’ First and
Second Generation Catalysts
Figure 2. Functional group selective chemistry with occluded catalysts.
These reactions were run under identical conditions with occluded Grubbs’
second generation catalyst; they differed in the presence of an added base
or acid. The base deprotonated the acid and kept 7 from diffusing into the
PDMS and reacting. The numbers in parenthesis refer to the isolated yields.
described and show that Grubbs’ catalysts are stable upon occlusion
and can react by cross or ring-closing metathesis. Furthermore, the
PDMS slab can act as an active membrane to impart functional
group selectivity on occluded catalysts.
Acknowledgment. We thank the Research Corporation, the
American Chemical Society-Petroleum Research Funds, the Ameri-
can Cancer Society, and the University of Iowa for financial support.
Supporting Information Available: Experimental detail and
characterization data. This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(
1) (a) Astruc, D.; Lu, F.; Aranzaes, J. R. Angew. Chem., Int. Ed. 2005, 44,
7
852-7872. (b) Dijkstra, H. P.; Van Klink, G. P. M.; van Koten, G. Acc.
Chem. Res. 2002, 35, 798-810.
(
(
(
2) Coperet, C.; Chabanas, M.; Saint-Arroman, R. P.; Basset, J.-M. Angew.
Chem., Int. Ed. 2003, 42, 156-181.
3) (a) McMorn, P.; Hutchings, G. J. Chem. Soc. ReV. 2004, 33, 108-122.
(b) Kakkar, A. K. Chem. ReV. 2002, 102, 3579-3587.
4) For a review reactions with different occluded catalysts in PDMS see the
following references. (a) Guedes, D. F. C.; Mac Leod, T. C. O.; Gotardo,
M. C. A. F.; Schiavon, M. A.; Yoshida, I. V. P.; Cuiffi, K. J.; Assis, M.
D. Appl. Catal., A 2005, 296, 120-127. (b) Wolfson, A.; Janssens, S.;
Vankelecom, I.; Geresh, S.; Gottlieb, M.; Herskowitz, M. Chem. Commun.
a
These reactions were run with either the first or second generation
2
002, 388-389. (c) Vankelecom, I.; Vercruysse, K.; Moens, N.; Parton,
b
catalysts. The mole % of catalyst to substrate is shown in parentheses. The
time each of these reactions were run to 100% conversion by H NMR
spectroscopy. Isolated yields are shown. One equiv of p-toluenesulfonic
R.; Reddy, J. S.; Jacobs, P. Chem. Commun. 1997, 137-138. (d)
Vankelecom, I. F. J.; Tas, D.; Parton, R. F.; Van de Vyver, V.; Jacobs, P.
A. Angew. Chem., Int. Ed. 1996, 35, 1346-1348.
1
c
d
acid to the substrate was added to the reaction.
(5) (a) Grubbs, R. H. Handbook of Metathesis Catalysts; Wiley-VCH:
Weinheim, Germany, 2003. (b) Buchmeiser, M. R. Polymeric Materials
in Organic Synthesis and Catalysis; Wiley-VCH: Weinheim, Germany,
2003.
2
were dissolved in 90% MeOH in H O with occluded Grubbs’
(
6) For solubilities of organic compounds and water see the following
second generation catalyst. In one reaction we added NaOH to
deprotonate the acid on 7 and in another reaction we added
p-toluenesulfonic acid (PTSA) to facilitate the diffusion of 7 into
the PDMS. These reactions were run under identical conditions
but yielded dramatically different results. In the reaction with NaOH
we did not observe any of the cyclized products of 7, but 5
cyclized.11 When PTSA was added, both reagents cyclized. Thus,
we were able to control the distribution of products by affecting
the diffusion of reagent 7 into PDMS. These experiments are
important since they demonstrate that occluded catalysts can
demonstrate functional group selectivity without modification of
the catalysts’ structures.
references. (a) Mayer, P.; Vaes, W. H. J.; Hermens, J. L. M. Anal. Chem.
2
000, 72, 459-464. (b) Baltussen, E.; Sandra, P.; David, F.; Janssen, H.-
G.; Cramers, C. Anal. Chem. 1999, 71, 5213-5216. (c) Lee, J. N.; Park,
C.; Whitesides, G. M. Anal. Chem. 2003, 75, 6544-6554.
(
7) Hong, S. H.; Grubbs, R. H. J. Am. Chem. Soc. 2006, 128, 3508-3509.
(8) Kirkland, T. A.; Lynn, D. M.; Grubbs, R. H. J. Org. Chem. 1998, 63,
9904-9909.
(9) (a) Connon, S. J.; Blechert, S. Bioorg. Med. Chem. Lett. 2002, 12, 1873-
1876. (b) Davis, K. J.; Sinou, D. J. Mol. Cat. A: Chem. 2002, 177, 173-
1
78. (c) Zarka, M. T.; Nuyken, O.; Weberskirch, R. Macromol. Rapid
Commun. 2004, 25, 858-862.
(
10) To demonstrate how little water diffuses into PDMS, we cured crushed
sodium bicarbonate into PDMS. The PDMS slab was immersed in water
acidified to pH ) 1 with HCl. After several months the sodium bicarbonate
solid was clearly visible in the PDMS.
(11) In the reaction with NaOH, we did not observe any of the cyclized product
1
of 7 by crude H NMR spectroscopy.
Reactions of catalytically active Grubbs’ first and second
generation catalysts occluded in PDMS using aqueous solvents are
JA0642212
J. AM. CHEM. SOC.
9
VOL. 128, NO. 45, 2006 14435