Modulating the activity and selectivity of an immobilized ruthenium-porphyrin catalyst using a fluorous solvent

Article abstract of DOI:10.1002/adsc.200505456

A ruthenium porphyrin catalyst with styrene side chains was incorporated into a highly cross-linked polymer by co-polymerization with ethylene glycol dimethacrylate in the presence of a chloroform porogen. Oxidation reactions catalyzed by the resulting polymer were accelerated when perfluoromethylcyclohexane (PFMC) was used as a co-solvent. Moreover, the PFMC co-solvent was found to change the substrate selectivity of the catalytic reactions. Both effects could be explained by a PFMC-induced partitioning of substrates and oxidant into the polymeric, catalyst containing matrix.

Full text of DOI:10.1002/adsc.200505456

FULL PAPERS
DOI: 10.1002/adsc.200505456
Modulating the Activity and Selectivity of an Immobilized
Ruthenium-Porphyrin Catalyst using a Fluorous Solvent
a
b
a,
b,
Estelle Burri, Shannon M. Leeder, Kay Severin, * and Michel R. GagnØ *
a
Institut des Sciences et IngØnieries Chimiques, École Polytechnique FØdØrale de Lausanne (EPFL), 1015 Lausanne,
Switzerland
Fax : (+41)-21-693-9305; e-mail: kay.severin@epfl.ch
Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-3290, USA;
b
e-mail: mgagne@unc.edu
Received: November 25, 2005; Accepted: April 15, 2006
Abstract: A rutheniumporphyrin catalyst with sty-
to change the substrate selectivity of the catalytic re-
rene side chains was incorporated into a highly actions. Both effects could be explained by a PFMC-
cross-linked polymer by co-polymerization with eth- induced partitioning of substrates and oxidant into
ylene glycol dimethacrylate in the presence of a the polymeric, catalyst containing matrix.
chloroformporogen. Oxidation reactions catalyzed
by the resulting polymer were accelerated when per-
fluoromethylcyclohexane (PFMC) was used as a co- Keywords: fluorous solvent; immobilization; oxida-
solvent. Moreover, the PFMC co-solvent was found tion; polymers; porphyrin; ruthenium
Introduction
quantified.[7] For less polar substances such as anthra-
cene, the partitioning was lower indicating that fluo-
Macroporous, highly cross-linked organic polymers rophobic effects are the likely cause. Permanently
are increasingly being used as supports for immobi- porous networks are uniquely effective in this context
lized transition metal catalysts.[1] These materials dis- since poor solvents collapse lightly cross-linked mate-
play a number of favorable characteristics. First of all, rials into impenetrable gels.
they typically show
a
high surface area (50–
The consequences of a strong partitioning of some
2
ꢀ1
[2]
500 m g ) and a distribution of pores that ensure analytes into poly-EGDMA is a high local concentra-
efficient access to catalysts within the polymer. Cata- tion within the pores of the polymer. This leads to the
lytic transformations are thus not restricted to sites on intriguing possibility that the activity of a catalyst em-
the exterior surface of the polymer particle. Secondly, bedded in such a support may be enhanced by a fa-
the high content of cross-links results in a permanent vorable substrate concentration gradient when fluori-
[
3]
[8]
pore structure, which allows the use of polar and nated solvents are employed. First results with an
non-polar solvents for catalysis. This is in contrast to immobilized rhodium(I) catalyst suggested that this
lightly cross-linked supports such as Merrifield resins, was indeed possible; the rates for a hydrogenation re-
for which swelling is necessary for access to the interi- action were found to increase with the fluorous con-
[
4]
[9,10]
or volume. Finally, highly cross-linked organic poly- tent of the solvent.
In the following we provide
mers are amenable to molecular imprinting. In the evidence that such rate enhancements are likely to be
case of immobilized metal catalysts, this technique a more general phenomena for catalysts immobilized
[
11]
can be used to modulate the microenvironment and in highly cross-linked polymers.
Oxidation reac-
thus the activity and selectivity of the catalyst in a tions catalyzed by a poly-EGDMA-supported rutheni-
controlled fashion.[
5,6]
umcomplex are shown to be faster in PFMC-contain-
An interesting feature of highly cross-linked poly- ing solvents. Furthermore, it is demonstrated that dif-
acrylates) is their ability to act as potent sorbents for ferential partitioning propensities of various sub-
(
polar organic compounds dissolved in fluorinated sol- strates can predictively influence the substrate selec-
vents. For example, when a homopolymer of ethylene tivity of the reactions.
glycol dimethacrylate (EGDMA) was suspended in a
solution of 9-anthracenemethanol in perfluoromethyl-
cyclohexane (PFMC):hexane (1:1), a strong partition-
ing of the alcohol into the polymer was observed and
1640
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1640 – 1644

Activity and Selectivity of an Immobilized Ruthenium-Porphyrin Catalyst
FULL PAPERS
Results and Discussion
temperature of 558C, all solvent mixtures form a
monophasic system. The conversion for the first
Recently, we have shown that the co-polymerization 1.25 h of these reactions is depicted in Figure 1. A
of vinyl-substituted rutheniumporphyrin co mp lexes pronounced increase in catalytic activity was observed
with a large excess of EGDMA can be used to gener- upon increasing the concentration of PFMC. After
[6a,12]
ate potent heterogeneous catalysts.
For the oxi- 1.25 h, the conversions for reactions with 0 and 40%
dation of alcohols and alkanes by 2,6-dichloropyridine PFMC differed by a factor of 10.5. A short induction
N-oxide (Cl pyNO), these catalysts were found to be period was evident fromthe time course of the reac-
2
significantly more active than the corresponding ho- tions. This could be explained by noting that the im-
mogeneous catalysts. A likely explanation for this en- mobilized Ru(CO) complex is only a catalyst precur-
hanced activity is the site-isolation of the catalyst sor, fromwhich the catalytically active Ru =O species
within the highly cross-linked polymeric support. Due must be generated.[14]
to the low concentration of the porphyrin complex in
the polymer (0.25 mol% with respect to the cross-
linking monomer), destructive self-oxidation reactions
as observed for homogeneous catalysts[ are prohib-
13]
ited.
In a continuation of these studies, we have investi-
gated whether fluorous solvent-induced partitioning
effects can be used to modulate the activity and selec-
tivity of such catalysts. For this purpose, we prepared
the polymeric catalyst P1 by AIBN initiated co-poly-
[6a]
merization of complex
1
with EGDMA
(
[1]:[EGDMA]=1:400) in the presence of chloroform
as the porogen (Scheme 1). The resulting dark-red po-
lymer P1 was ground in a mortar, washed extensively
Figure 1. Oxidation of 1-indanol by Cl pyNO with the poly-
2
meric catalyst P1 in solvents containing various amounts of
PFMC (~: 0%; ~: 10%; *: 20%; *: 40%). The reactions
were performed in mixtures of benzene, hexane and PFMC
at 558C with a substrate/Cl pyNO/catalyst molar ratio of
2
1
00:100:1. The data points represent averaged values from
two independent experiments.
Scheme 1. Synthesis of the polymeric catalyst P1 by co-poly-
merization of complex 1 with EGDMA.
A priori, the observed rate enhancements could be
due to a PFMC-induced partitioning of the alcohol
with acetone and finally dried in vacuum. The nearly and/or the N-oxide into the polymeric matrix. To gain
colorless washing solutions indicated a quantitative further insight, we performed oxidation reactions with
incorporation of the metallomonomer 1 in the poly- different starting concentrations of N-oxide or 1-inda-
meric EGDMA matrix. From N2 adsorption iso- nol. These experiments showed that the reactions are
therms, a Brunauer–Emmett–Teller (BET) surface approximately first order with respect to the N-oxide
2
ꢀ1
area for P1 of 409 m g and an average pore size of and zero order with respect to indanol. The increased
4 Š were determined. rates can therefore be explained by an increased local
To investigate the influence of a fluorous solvent concentration of the oxidant, a partitioning of the al-
5
on the catalytic activity of polymer P1, we followed cohol is thus not expected to accelerate the reaction.
the time course of the oxidation of 1-indanol using 1 Although control experiments with the homogeneous
mol% Ru and Cl pyNO as the oxidant. The amount catalyst 1 were not possible due to catalyst precipita-
2
[
15]
of polymer that was required was calculated based on tion in PFMC-containing solvent mixtures,
the data
the assumption that the metallomonomer 1 was incor- suggest that the ~10-fold increase in rate on going
porated quantitatively. A mixture of benzene and from0 to 40% PFMC results froma fluorophobic in-
hexane with various amounts of PFMC (0, 10, 20 and duced ~10-fold increase in the local concentration of
40%) was employed as the solvent. At the reaction the oxidant.
Adv. Synth. Catal. 2006, 348, 1640 – 1644
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1641

Estelle Burri et al.
FULL PAPERS
Next, we investigated the oxidation of indane in a
mixture of benzene/hexane containing 0 and 40%
PFMC. This reaction proceeded in a two-step fashion
to give first indanol, which could be further oxidized
to indanone. At the beginning of the reaction, both
products were present in comparable amounts. Inter-
esting differences, however, were found for the rela-
tive amounts of these products for the two reaction
conditions. At a total conversion of ~5%, indanol
was the dominant reaction product in reactions per-
formed in benzene/hexane. On the other hand, when
the solvent contained 40% PFMC, it was indanone
that was dominant (Figure 2). These results are ex-
Figure 2. Oxidation of indane by Cl pyNO with the polymer-
2
ic catalyst P1 in solvents containing 0% or 40% of PFMC.
The reactions were performed in mixtures of benzene,
hexane and PFMC at 558C with a substrate/Cl pyNO/cata-
lyst molar ratio of 100:200:1. The data represent averaged
values fromtwo independent experiments.
2
Figure 3. Simultaneous oxidation of two different secondary
alcohols by Cl pyNO with the polymeric catalyst P1 in sol-
2
vents containing 0% or 40% of PFMC. The yields and the
ratios of the ketone products are given. The reactions were
performed in mixtures of benzene, hexane and PFMC at
plainable by the higher partition efficiency of the
polar indanol as compared to indane. Since the prod-
uct determining step (not turnover limiting) involves
a direct competition between catalyst and indane or
indanol, the relative concentration of these substrates
will directly influence the selectivity, even though the
5
58C with a substrate/substrate/Cl pyNO/catalyst molar
2
ratio of 100:100:100:1. The data points represent averaged
values fromtwo independent experiments.
In reactions with the substrates 1-indanol and 2-oc-
overall reaction is zero order in substrate. In the 40% tanol, we observed a strong preference for the oxida-
PFMC case, the local concentration of the more polar tion of indanol when PFMC was used as the co-sol-
indanol in the catalyst phase is higher, which leads to vent. At 15 and 60 min, respectively, 2-octanone was
a more favorable reaction cross-section. The inversion formed in approximately equal amounts. For reactions
in product distribution is therefore a direct conse- with PFMC, however, more than twice as much inda-
quence of their polarity differences rather than their none was formed when compared to the reaction per-
inherent chemical reactivities.
formed in benzene/hexane. This finding is in agree-
We have additionally investigated this effect in the ment with the observation that non-aromatic hydro-
oxidation of three substrate mixtures with variable carbons of low polarity have a small partitioning effi-
[
7]
differences in polarity. In all cases, equimolar amounts ciency. A similar but less pronounced trend was ob-
of two secondary alcohols were oxidized to the corre- served for reactions with 1-tetralol and 4-chromanol.
sponding ketones using again solvent mixtures of ben- In the presence of PFMC, the selectivity for the more
zene/hexane containing 0 and 40% PFMC. The re- polar chromanol increased. No differences in selectiv-
sults are summarized in Figure 3.
ity were observed for 2-phenylethanol and 2-phenyl-
1642
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1640 – 1644

Activity and Selectivity of an Immobilized Ruthenium-Porphyrin Catalyst
FULL PAPERS
pentanol, two substrates of rather similar overall po- analyzed by GC. For the competition experiments, two sub-
ꢀ3
strates were added at the same time (6.25·10 mmol each).
larity.
Conclusions
Determination of the Reaction Order
We have shown that oxidation reactions catalyzed by To determine the reaction order of the oxidation of 1-inda-
nol by Cl pyNO in the presence of 40% PFMC, a series of
the immobilized ruthenium porphyrin catalyst P1 are
significantly accelerated in the presence of the fluori-
nated solvent PFMC. This rate acceleration is ex-
plained by invoking a consequently heightened local
2
reactions with constant concentrations of Cl pyNO
12.5 mM) and catalyst (25 mM), but varying concentrations
of 1-indanol (2.50 to 6.25 mM) was performed. The initial
rates of the reactions as a function of the 1-indanol concen-
2
(
concentration of the Cl pyNO oxidant, which is an in-
ꢀ1
2
tration were found to be constant (TOF=139ꢁ3 h ; calcu-
tegral component in the turnover-limiting step (rate
lated fromthe yield after 20 mi n). This indicates a zero-
order dependence with respect to the substrate. In addition,
/
[Cl pyNO]). Additionally, it is demonstrated that
2
this change in solvent also affects the substrate selec- a series of reactions with constant substrate (1.25 mM) and
tivity of the catalytic reaction. The rate enhancements catalyst (12.5 mM) concentrations but varying Cl pyNO con-
2
and change in selectivity are taken as evidence sup- centrations (625 mM to 2.50 mM) was carried out. A first
order dependence of the initial rates as a function of the
Cl pyNO concentration was observed.
2
porting the notion that fluorous solvent-induced parti-
tioning effects can strongly influence reactions with
catalysts that are immobilized in highly cross-linked
organic polymers. The fact that heterogeneous cata-
lysts of this kind not only tolerate fluorinated solvents
but may actually function better in such solvents is a
finding that may be of interest for various applica-
tions involving fluorous solvents.[16]
Acknowledgements
We gratefully acknowledge support of this work by an ERO-
ARO Seed grant (N62558–04m-0010). MRG additionally ac-
knowledges the Department of Energy, Office of Basic
Energy Sciences and the Army Research Office for partial
support.
Experimental Section
General Remarks
References
The syntheses of complex 1 and polymer P1 were performed
as described in ref.[ The BET measurements were carried
out by QUANTACHROME GmbH, Odelzhausen, on a
QUANTACHROME AUTOSORB-3 instrument. The ad-
6a]
[
1] a) C. Saluzzo, M. Lemaire, Adv. Synth. Catal. 2002, 344,
15–928; b) Y. R. De Miguel, E. BrulØ, R. G. Margue,
9
J. Chem. Soc., Perkin Trans. 1 2001, 3085–3094; c) B.
Altava, M. I. Burguete, E. García-Verdugo, S. V. Luis,
M. J. Vicent, J. A. Mayoral, React. Funct. Polym. 2001,
sorption measurements were performed with N at a sample
2
temperature of ꢀ1968C. Prior to the measurements, the
samples were dried under vacuum at 1008C for 2 h. The
average pore diameter was calculated form the pore volume
and the BET surface area using assuming a cylindrical pore
model. The GC analyses were performed with a Varian 3800
spectrometer using a CP-Sil 8 CB column (30 m).
4
8, 25–35; d) B. Clapham, T. S. Reger, K. D. Janda, Tet-
rahedron 2001, 57, 4637–4662; e) Y. R. de Miguel, J.
Chem. Soc., Perkin Trans. 1 2000, 4213–4221; f) S. J.
Shuttleworth, S. M. Allin, R. D. Wilson, D. Nasturica,
Synthesis 2000, 1035–1074; g) S. J. Shuttleworth, S. M.
Allin, P. K. Sharma, Synthesis 1997, 1217–1239.
[
2] a) D. C. Sherrington, Chem. Commun. 1998, 2275–
2286; b) B. P. Santora, M. R. GagnØ, K. G. Moloy, N. S.
Radu, Macromolecules 2001, 34, 658–661.
Catalytic Oxidations
ꢀ
3
The substrate (6.25·10 mmol) was added to a suspension
[3] R. A. Taylor, B. P. Santora, M. R. GagnØ, Org. Lett.
ꢀ5
of the polymer P1 (10.0 mg, 6.25·10 mmol Ru) in different
2000, 2, 1781–1783.
solvent mixtures (A: 4.90 mL benzene, 0% PFMC; B:
[4] a) B. Corain, M. Zecca, K. Je rˇ ꢁbek, J. Mol. Catal. 2001,
177, 3–20; b) A. R. Vaino, K. D. Janda, J. Comb.
Chem. 2000, 2, 579–596; c) P. Hodge, Chem. Soc., Rev.
1997, 26, 417–424.
[5] For reviews see: a) L. L. Welbes, A. S. Borovik, Acc.
Chem. Res. 2005, 38, 765–774; b) J. J. Becker, M. R.
GagnØ, Acc. Chem. Res. 2004, 37, 798–804; c) S. Strie-
gler, J. Chromatogr. B 2004, 804, 183–195; d) M. Tada,
Y. Iwasawa, J. Mol. Catal. A 2003, 199, 115–137; e) C.
Alexander, L. Davidson, W. Hayes, Tetrahedron 2003,
59, 2025–2057; f) G. Wulff, Chem. Rev. 2002, 102, 1–
2.90 mL benzene+1.50 mL hexane+0.50 mL PFMC=10%
PFMC; C: 2.40 mL benzene+1.50 mL hexane+1.00 mL
PFMC=20% PFMC; D: 1.40 mL benzene+1.50 mL
hexane+2.00 mL PFMC=40% PFMC). The mixture was
placed in an oil bath and tempered at 558C for 30 min. The
reaction was then started by addition of 100 mL of a stock
ꢀ1
solution containing Cl pyNO (51.2 mg, 3.12·10 mmol) in
2
benzene (5.0 mL). Samples (100 mL) were removed at regu-
lar intervals, filtered and poured in vials containing 1 mL of
diethyl ether (in order to obtain a monophasic solution) and
Adv. Synth. Catal. 2006, 348, 1640 – 1644
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1643

Estelle Burri et al.
FULL PAPERS
2
7
8; g) K. Severin, Curr. Opin. Chem. Biol. 2000, 4,
10–714.
[10] For partitioning effects in aqueous solvents see: C. M.
Hartshorn, J. R. Deschamps, A. Singh, E. L. Chang,
React. Funct. Polym. 2003, 55, 219–229.
[11] This is a well-established phenomenon in heterogene-
ous catalysis, see for example: G. Langhendries, D. E.
De Vos, G. V. Baron, P. A. Jacobs, J. Catal. 1999, 187,
[
6] For some selected examples see: a) E. Burri, M. ꢂhm,
C. Daguenet, K. Severin, Chem. Eur. J. 2005, 11, 5055–
5
1
061; b) J.-Q. Liu, G. Wulff, J. Am. Chem. Soc. 2004,
26, 7452–7453; c) Z. Cheng, L. Zhang, Y. Li, Chem.
Eur. J. 2004, 10, 3555–3561; d) J.-Q. Liu, G. Wulff,
Angew. Chem. Int. Ed. 2004, 43, 1287–1287; e) F.
Viton, P. S. White, M. R. GagnØ, Chem. Commun. 2003,
4
53–463.
[
[
12] O. Nestler, K. Severin, Org. Lett. 2001, 3, 3907–3909.
13] E. Gallo, A. Caselli, F. Ragaini, S. Fantauzzi, N. Mas-
ciocchi, A. Sironi, S. Cenini, Inorg. Chem. 2005, 44,
3
040–3041; f) J. H. Koh, A. O. Larsen, P. S. White,
M. R. GagnØ, Organometallics 2002, 21, 7–9; g) A. N.
Cammidge, N. J. Baines, R. K. Bellingham, Chem.
Commun. 2001, 2588–2589; h) K. Polborn, K. Severin,
Chem. Eur. J. 2000, 6, 4604–4611; i) K. Polborn, K. Se-
verin, Chem. Commun. 1999, 2481–2482.
2
039–2049.
[
14] a) J. T. Groves, K. Shalyaev, J. Lee, in: The Porphyrin
Handbook, (Eds.: K. M. Kadish, K. M. Smith, R. Gui-
lard), Academic Press, New York, Vol. 4, 2000, pp. 17–
4
2
0; b) C.-M. Che, W.-Y. Yu, Pure Appl. Chem. 1999, 71,
81–288.
[
7] S. M. Leeder, M. R. GagnØ, J. Am. Chem. Soc. 2003,
1
25, 9048–9054.
[
15] Upon addition of Cl pyNO to a solution of complex 1
2
[
8] For several examples of fluorous accelerated solid-
phase synthesis (coined FAST), see: a) R. Morphy, Z.
Rankovic, M. York, Tetrahedron Lett. 2001, 42, 7509–
in benzene/hexane/PFMC (1.4:1.5:2.0) at 508C, a black
precipitate was observed after a few minutes.
16] a) W. Zhang, Tetrahedron 2003, 59, 4475–4489; b) H.-
U. Blaser, M. Studer, Green Chem. 2003, 5, 112–117;
c) A. P. Dobbs, M. R. Kimberley, J. Fluor. Chem. 2002,
[
7
511; b) R. Morphy, Z. Rankovic, M. York, Tetrahe-
dron Lett. 2002, 43, 6413–6415; c) R. Morphy, Z. Ran-
kovic, M. York, Tetrahedron 2003, 59, 2137–2145;
d) R. S. Roberts, J. Comb. Chem. 2005, 7, 21–32.
1
18, 3–17; d) J. A. Gladysz, D. P. Curran, Tetrahedron
2
002, 58, 3823–3825; e) L. P. Barthel-Rosa, J. A. Gla-
[
9] S. L. Vinson, M. R. GagnØ, Chem. Commun. 2001,
dysz, Coord. Chem. Rev. 1999, 190–192, 587–605.
1
130–1131.
1644
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1640 – 1644