Catalyst recycling via hydrogen-bonding-based affinity tags

Article abstract of DOI:10.1021/ol0607387

A novel procedure for catalyst recycling is described. Copper(I)-based catalysts, equipped with an affinity tag, are isolated from crude reaction mixtures on the basis of quadruple hydrogen-bonding interactions using a resin functionalized with complement

Full text of DOI:10.1021/ol0607387

ORGANIC
LETTERS
2006
Vol. 8, No. 15
3163-3166
Catalyst Recycling via
Hydrogen-Bonding-Based Affinity Tags
Bas W. T. Gruijters,† Maarten A. C. Broeren,‡ Floris L. van Delft,†
Rint P. Sijbesma,‡ Pedro H. H. Hermkens,§ and Floris P. J. T. Rutjes*,†
Institute for Molecules and Materials, Radboud UniVersity Nijmegen, ToernooiVeld 1,
NL-6525 ED Nijmegen, The Netherlands, EindhoVen UniVersity of Technology,
P.O. Box 513, NL-5600 MB EindhoVen, The Netherlands, Organon N.V., P.O. Box 20,
NL-5340 BH Oss, The Netherlands
f.rutjes@science.ru.nl
Received March 26, 2006
ABSTRACT
A novel procedure for catalyst recycling is described. Copper(I)-based catalysts, equipped with an affinity tag, are isolated from crude reaction
mixtures on the basis of quadruple hydrogen-bonding interactions using a resin functionalized with complementary affinity tags. Recycled
catalysts were successfully used to catalyze a tandem Sonogashira coupling/5-endo-dig cyclization and a Cu-catalyzed [3
+2] Huisgen
cycloaddition reaction in high yields.
The use of transition-metal-based catalysts in chemical
processes is widespread because of an increasing amount of
applications and a generally favorable atom economy.
Drawbacks, however, are the high prices of the metal
complexes and the often laborious removal of the metals
from reaction mixtures. Facile isolation of catalysts and
subsequent reuse is therefore an attractive goal to reduce both
cost and waste. As a result, the development of new
technology for catalyst recycling has been of interest for
many researchers in the past decades.1 A number of
techniques to recycle metal-based catalysts have been
developed over the years, such as covalent catalyst im-
mobilization,2 fluorous phase separation,3 nanofiltration,4 and
ligand immobilization.5
catalyst and leaching of metal ions. In addition to the existing
techniques, we designed a novel noncovalent linking concept
for the recycling of catalysts on the basis of quadruple
hydrogen bonding (Figure 1).6 A ligand, generally used in
homogeneous catalysis, is equipped with a hydrogen-
bonding-based affinity tag (AT). The ligand is complexed
to a metal, and the resulting catalyst can be applied in a
homogeneous catalytic reaction in an organic solvent,
assuming that the affinity tag does not affect its catalytic
activity. After the catalytic reaction, the hydrogen-bonding
properties of the AT allow facile isolation of the complex
by binding it to a resin that contains a complementary
hydrogen-bonding array (1). A simple filtration can be used
to separate the soluble reaction products from the catalyst-
containing solid support. Rinsing the polymer with a protic
solvent causes cleavage of the hydrogen bonds between resin
and the AT so that a second filtration yields only the tagged
catalyst, which can then be directly used in the next catalytic
reaction.
Although significant progress has been made, there are
still major shortcomings, including decreased activity of the
† Radboud University Nijmegen.
‡ Eindhoven University of Technology.
§ Organon N.V.
(1) For recent overviews, see, e.g.: (a) Catalyst Separation, RecoVery
and Recycling-Chemistry and Process Design; Cole-Hamilton, D. J., Tooze,
R. P., Eds.; Springer, New York, 2006. (b) Yoshida, J.; Itami, K. Chem.
ReV. 2002, 102, 3693-3716.
The AT that was selected to demonstrate the viability of
this new purification method was the ureido[1H]pyrimidinone
10.1021/ol0607387 CCC: $33.50
© 2006 American Chemical Society
Published on Web 06/22/2006

The different resins were added to a stock solution of 7
containing 4 equiv of binding sites relative to the amount of
7. After 2 h of shaking, the resins were filtrated and the
concentration of 7 in the filtrate was determined using UV
measurements. The results are summarized in Table 1. The
Table 1. Initial Complexation Studiesa
Figure 1. Principle of recycling by hydrogen bonds.
(UPy) group. This self-complementary UPy tag displays a
remarkably high dimerization constant of >107 in chloro-
form.7 We envisioned that functionalization of a solid support
with a UPy tag, in combination with a complementary UPy-
tagged catalyst, should give rise to a distinct affinity, leading
to a reversible binding as described above.
Initially, an optimal AT-equipped resin system had to be
developed. Therefore, both a flexible Merrifield resin (2a,
2% cross-linked) and a highly cross-linked, rigid ArgoPore
resin (2b) were functionalized with different ATs and their
complexation efficiency was evaluated. These complexation
studies with the UV-active UPy-tagged compound 78 were
performed to determine which resin would be most suited
to bind tagged compounds from a solution.
a Conditions: 4 equiv of binding sites on the resin (relative to the amount
of 7) in 1,2-dichloropropane, 2 h, room temperature.
Merrifield resin. b: ArgoPore resin.
b a: 2% cross-linked
(2) See, for example: (a) Cornejo, A.; Fraile, J. M.; Garc´ıa, J. I.; Gil,
M. J.; Luis, S. V.; Mart´ınez-Merino, V.; Mayoral, J. A. J. Am. Chem. Soc.
2005, 70, 5536-5544. (b) Kobayashi, S.; Akiyama, R. Chem. Commun.
2003, 4, 449-460. (c) Nagayama, S.; Kobayashi, S. Angew. Chem., Int.
Ed. 2000, 39 567-569. (d) Annis, D. A.; Jacobsen, E. N. J. Am. Chem.
Soc. 1999, 121, 4147-4154.
high amounts of unbound substrate using Merrifield resins
3a and 4a (entries 3 and 5) led us to conclude that
homodimerization on the solid support was a significant
problem. Conceivably, the relatively flexible linker allowed
the resin-bound ATs to dimerize, making binding of substrate
molecules more difficult. This problem was partially solved
by using a rigid ArgoPore resin with a low degree of
functionalization (0.28 mmol of endgroups/g). In this case,
the substrate was bound somewhat more efficiently, with a
clear difference between the linker-bound UPy moiety (3b,
entry 4) and the UPy that was directly attached to the resin
(4b, entry 6). This observation confirmed the hypothesis of
homodimerization on the resin. Finally, the best results were
obtained using the same highly cross-linked ArgoPore resin
in combination with alkyl-functionalized UPy moieties
(5a,b). Because the side group on these UPy moieties did
not have a large influence on the complexation, i.e., resins
5b and 6b performed equally well in binding substrate 7
(entries 7 and 8), the commercially available isocytosine (R
) Me) was selected as the AT of choice for the catalyst
recycling process. The functionalization procedure involved
(3) For reviews, see: (a) de Wolf, E.; van Koten, G.; Deelman, B. Chem.
Soc. ReV. 1999, 28, 37-41. (b) Curran, D. P. Angew. Chem., Int. Ed. 1998,
37, 1175-1196. For some recent examples, see, e.g.: (a) Matsugi, M.;
Curran, D. P. J. Org. Chem. 2005, 70, 1636-1642. (b) Dinh, L. V.; Gladysz,
J. A. Angew. Chem., Int. Ed. 2005, 44, 4095-4097.
(4) For a review, see: Dijkstra, H. P.; van Klink, G. P. M.; van Koten,
G. Acc. Chem. Res. 2002, 35, 798-810. Recent examples: (a) Witte, P.
T.; Chowdhury, S. R.; ten Elshof, J. E.; Sloboda-Rozner, D.; Neumann,
R.; Alsters, P. L. Chem. Commun. 2005, 9, 1206-1208. (b) Aerts, S.;
Weyten, H.; Buekenhoudt, A.; Gevers, L. E. M.; Vankelecom, I. F. J.;
Jacobs, P. A. Chem. Commun. 2004, 6, 710-711. (c) De Smet, K.; Pleysier,
A.; Vankelecom, I. F. J.; Jacobs, P. A. Chem.-Eur. J. 2003, 9, 334-338.
(5) For covalent examples, see, e.g.: (a) Belser, T.; Sto¨hr, M.; Pfaltz,
A. J. Am. Chem. Soc. 2005, 127, 8720-8731. (b) Dai, L.-X. Angew. Chem.,
Int. Ed. 2004, 41, 5726-5729. (c) Bra¨se, S.; Dahmen, S.; Lauterwasser,
F.; Leadbeater, N. E.; Sharp, E. L. Bioorg. Med. Chem. Lett. 2002, 12,
1849-1851. (d) Pugin, B. J. Mol. Catal. A 1996, 107, 273-279. For a
noncovalent example, see: Chen, R.; Bronger, R. P. J.; Kamer, P. C. J.;
van Leeuwen, P. W. N. M.; Reek, J. N. H. J. Am. Chem. Soc. 2004, 126,
14557-14566.
(6) For a separation tag based on H-bonding, see: Zhang, K.; Fukase,
M.; Izumi, Y.; Fukase, S.; Kusumoto Synlett 2001, 590.
(7) (a) Folmer, B. J. B.; Sijbesma, R. P.; Versteegen, R. M.; van der
Rijt, J. A. J.; Meijer, E. W. AdV. Mater. 2000, 12, 874-878. (b) Beijer, F.
H.; Sijbesma, R. P.; Kooijman, H.; Spek, A. L.; Meijer, E. W. J. Am. Chem.
Soc. 1998, 120, 6761-6769.
(8) So¨ntjens, S. H. M.; Sijbesma, R. P.; van Genderen, M. H. P.; Meijer,
E. W. J. Am. Chem. Soc. 2000, 122, 7487-7493.
3164
Org. Lett., Vol. 8, No. 15, 2006

treatment of the ArgoPore resin with carbonyl diimidazole,
followed by reaction with isocytosine, allowing a large-scale
preparation of the resin-bound AT 6b.9
Having the optimized resin system in hand, next we
focused on identifying a suitable solution-phase AT. When
using the same UPy group (R ) Me) for this purpose, a
very poor solubility of a variety of tagged compounds in
virtually all common solvents was observed.
Therefore, we felt that switching to isocytosines with less
polar and bulkier R-groups was necessary. The alternative
isocytosines 8a and 8b were synthesized (R ) (1-ethylpentyl)
and 1-adamantyl, respectively) via condensation of the
corresponding â-ketoester with guanidine carbonate (Scheme
1).10 This standard reaction initially did not work in the case
focused on the functionalization of a ligand with the
(1-ethylpentyl)-functionalized UPy tag.
The ligand we chose was 1,10-phenanthroline (phen), a
well-known ligand for Cu-catalyzed cross-coupling reac-
tions.11 1,10-Phenanthroline has been functionalized previ-
ously for immobilization purposes as reported by Canham
et al., and we used their procedure to obtain compound 11.12
The alkylation of the phenolic hydroxyl group initially gave
poor results, but via adjusting Canham’s method by switching
to other conditions (K2CO3 in DMF), the desired compound
12 was obtained in 94% yield (Scheme 2). After cleaving
Scheme 2. Synthesis of the UPy-Tagged Ligand 14
Scheme 1. Synthesis of Isocytosines 8a,b and UPy-Tagged
Compounds 10a,b
the Boc group (TFA in CH2Cl2) to unveil the amine, we
used the previously described carbonyl diimidazole activation
to attach the required UPy tag.9 Thus, the activated isocy-
tosine 13 was obtained from 8a and then reacted with the
deprotected phenanthroline 12 to give the AT-bearing ligand
14 in excellent yield (93% over two steps).13
We decided to first test our new recycling concept on the
copper(I)-catalyzed synthesis of 2-arylbenzo[b]furans as
developed in the group of Venkataraman.11a This tandem
Sonogashira coupling/5-endo-dig cyclization was carried out
using [Cu(phen)(PPh3)2]NO3 as the catalyst and 2 equiv of
Cs2CO3 as a base in toluene (Table 2). Instead of isolating
the copper(I) complex, it was formed in situ by stirring
equimolar quantities of Cu(PPh3)2NO3,14 triphenylphosphine,
and AT-ligand 14 in DMF, which immediately resulted in a
clear yellow solution. To this solution, the other reagents
were added, and using the same conditions as those described
by Venkataraman, we obtained a complete conversion. This
clearly indicates that the UPy tag does not interfere with the
copper(I)-mediated coupling and that DMF can be used as
a solvent for this cyclization process.
of the bulky adamantyl group but proceeded in excellent yield
after addition of a stoichiometric amount of NaOtBu. The
two new tagged compounds 10a and 10b were synthesized
using these isocytosines in a one-pot procedure from car-
boxylic acid 9 by means of a Curtius rearrangement. The
products were indeed soluble in a range of organic solvents,
which is important for the scope of reactions that can be
performed using UPy-tagged catalysts. When complexation
experiments were carried out in chloroform using resin 6b
(10 equiv of binding sites relative to the amount of 10), the
substrates were bound in a quantitative manner after shaking
for approximately 5 h. Upon washing the resin with
chloroform, no notable leakage of these compounds from
the polymer was observed. Importantly, the tagged com-
pounds 10a and 10b were recovered in yields over 98% by
shaking the resin in a mixture of DMF/MeOH (2:1 v/v) for
2 h, followed by filtration and evaporation.
After an aqueous workup, the crude product was dissolved
in chloroform and the AT-functionalized resin 6b was added
(11) See, e.g.: (a) Bates, C. G.; Saejueng, P.; Murphy, J. M.; Venkat-
araman, D. Org. Lett. 2002, 4, 4727-4729. (b) Gujadhur, R. K.; Bates, C.
G.; Venkataraman, D. Org. Lett. 2001, 3, 4315-4317.
This proof of principle set the stage for application of our
affinity-based recovery method for catalysts, in which we
(12) Slough, G. A.; Krchnak, V.; Helquist, P.; Canham, S. M. Org. Lett.
2004, 6, 2909-2912 and references therein.
(9) Keizer, H. M.; Sijbesma, R. P.; Meijer, E. W. Eur. J. Org. Chem.
2004, 2553-2555.
(10) (a) Alker, D.; Campbell, S. F.; Cross, P. E.; Burges, R. A.; Carter,
A. J.; Gardiner, D. G. J. Med. Chem. 1989, 32, 2381-2388. (b) Snell, B.
K. J. Chem. Soc. C 1968, 2367-2370.
(13) Because of slightly better solubility, we preferred to use the
1-ethylpentyl-substituted isocytosine 8a instead of 8b.
(14) All metal complexes in this article were synthesized as described
by Venkataraman in ref 6 and used without further purification.
Org. Lett., Vol. 8, No. 15, 2006
3165

Table 2. Comparison among the Regular, the AT-Labeled, and
Table 3. Second Comparison among the Regular, the
the Recycled AT-Labeled Copper Catalysts
AT-Labeled, and the Recycled AT-Labeled Catalysts
isolated yield of 17
isolated yield of 19
amount cycle cycle cycle
amount cycle cycle cycle
entry
catalyst
(mol %)
1
2
3
entry
catalyst
(mol %)
1
2
3
1
2
(phen)Cu(PPh3)2NO3
(Upy-phen)Cu(PPh3)2NO3
10
10
>99%
-
-
1
2
(phen)Cu(PPh3)Br
(Upy-phen)Cu(PPh3)3Br
10
10
>99%
>99% >99% >99%
-
-
>99% 91% 68%
to bind the catalyst via hydrogen bonding. After filtration,
the product was isolated from the filtrate via evaporation
and the resin was transferred to a mixture of DMF and
methanol to cleave the hydrogen bonds. After filtering off
the resin, the filtrate was concentrated and the crude catalyst
was used in the next reaction.15 The recycled catalyst
remained active, although the yields dropped somewhat as
indicated in Table 2, probably because of the harsh reaction
conditions that may degrade the copper(I) complex.
To investigate the scope of this catalyst recycling proce-
dure further, we decided to use AT-ligand 14 in a second
copper(I)-catalyzed reaction. In 2002, both Meldal and
Sharpless independently showed that a regioselective Huisgen
[3+2] cycloaddition reaction could be performed under the
influence of a copper(I) catalyst.16 This method of preparing
1,2,3-triazoles has been widely used since then17 and
therefore forms an attractive application for our new
procedure. For this reaction, we used the copper catalyst
(Phen)Cu(PPh3)Br,14 which proved to be well-suited for the
[3+2] cycloaddition reaction depicted in Table 3, and the
desired product was isolated in quantitative yield.
recycled in the same way as previously described, initially
low yields were encountered in the second reaction cycle.
We reasoned that this could be due to undesired oxidation
of the copper(I) species to inactive copper(II). This was
overcome through the addition of 2 equiv of triphenylphos-
phine to the mixture to stabilize the Cu(I) complex so that
complete conversions and excellent yields were obtained in
the next two cycles as well (Table 3).15 To quantify possible
leakage of the catalyst during the recycling procedure,
quantitative mass spectroscopy measurements were carried
out. This showed that the amount of copper in the crude
product was only 1.7% of the initial quantity, meaning that
98.3% of the catalyst was bound by the AT-bearing resin.
This outcome, combined with the two aforementioned
applications, clearly demonstrates that the viability of
hydrogen-bonding-based catalyst recycling as a concept has
been firmly established.
In summary, we have realized a novel concept of catalyst
recycling based on hydrogen bonding, which is a potentially
powerful method because the AT allows facile purification
without affecting the catalyst. Furthermore, the tagged resin
can be cleaned by simple rinsing with different solvents and
reused without limitations. It is clear that other ligands may
be equipped with the same AT to recycle different catalysts
as well, and studies toward a broader application of this
affinity protocol are currently ongoing in our laboratories.
Analogous to the aforementioned AT catalyst, the tagged
CuBr-based catalyst was generated and used in the “click
reaction’ of substrates 16 and 18. Again, a quantitative yield
of 19 was obtained, showing the minimal influence of the
tag on the outcome of the reaction. When the catalyst was
(15) Following the same procedure with the nonfunctionalized resin 2b
did not result in any catalytic activity.
(16) (a) Tornøe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002,
67, 3057-3064. (b) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless,
K. B. Angew. Chem., Int. Ed. 2002, 41, 2596-2599.
(17) For some recent examples, see, e.g.: (a) Dura´n Pacho´n, L.; van
Maarseveen, J. H.; Rothenberg, G. AdV. Synth. Catal. 2005, 347, 811-
815. (b) Kaleta, Z.; Egyed, O.; Soo´s, T. Org. Biomol. Chem. 2005, 3, 2228-
2230. (c) Lin, H.; Walsh, C. T. J. Am. Chem. Soc. 2004, 126, 13998-
14003. (d) Wu, P.; Feldman, A. K.; Nugent, A. K.; Hawker, C. J.; Scheel,
A.; Voit, B.; Pyun, J.; Fre´chet, J. M. J.; Sharpless, K. B.; Fokin, V. V.
Angew. Chem., Int. Ed. 2004, 43, 3928-3932. (e) Kuijpers, B. H. M.;
Groothuys, S.; Keereweer, A. R.; Quaedflieg, P. J. L. M.; Blaauw, R. H.;
van Delft, F. L.; Rutjes, F. P. J. T. Org. Lett. 2004, 6, 3123-3126. (f)
Dirks, A. J.; van Berkel, S. S.; Hatzakis, N. S.; Opsteen, J. A.; van Delft,
F. L.; Cornelissen, J. J. L. M.; Rowan, A. E.; van Hest, J. C. M.; Rutjes, F.
P. J. T.; Nolte, R. J. M. Chem. Commun. 2005, 4172-4174.
Acknowledgment. These investigations were supported
(in part) by The Netherlands Research Council for Chemical
Sciences (CW) with financial aid from The Netherlands
Technology Foundation (STW). J. Eygensteyn (Radboud
University Nijmegen) is kindly acknowledged for performing
the ICP-MS measurements.
Supporting Information Available: Experimental pro-
cedures and spectroscopic data of new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL0607387
3166
Org. Lett., Vol. 8, No. 15, 2006