Crown-ether functionalised second coordination sphere palladium catalysts by molecular imprinting

Article abstract of DOI:10.1039/b309072h

Functionalisation of the second coordination sphere of a molecularly imprinted Pd complex was achieved by localising within the polymeric cavity a crown-ether receptor capable of altering the catalytic activity of the reactive site.

Full text of DOI:10.1039/b309072h

Crown-ether functionalised second coordination sphere palladium
catalysts by molecular imprinting†
Florian Viton, Peter S. White and Michel R. Gagné*
Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-3290, USA.
E-mail: mgagne@unc.edu
Received (in Corvallis, OR, USA) 31st July 2003, Accepted 24th September 2003
First published as an Advance Article on the web 20th October 2003
Functionalisation of the second coordination sphere of a
molecularly imprinted Pd complex was achieved by local-
ising within the polymeric cavity a crown-ether receptor
capable of altering the catalytic activity of the reactive
site.
factors coupled with Cammidge’s recently reported P2Pd(ca-
techolate)-based MIP catalysts for Suzuki reactions,7 stimulated
the design of a synthetically straightforward route to CE
palladium MIP catalysts. Non-covalent interactions enable the
assembly of a polymerisable ternary complex 2 which would
hopefully lead to a hybrid crown-ether functionalised active site
(Scheme 1).8
Treating 1ठwith a primary amine and crown-ether was
expected to generate an ion-pair that acted to electrostatically
associate the metal complex with a polymerisable CE. Indeed,
ternary complex 2 was obtained by combining 1 with 1.0 equiv.
of n-butylamine and 1.0 equiv. of 4A-vinylbenzo-18-crown-6 in
CH2Cl2 which maintained its compositional integrity after
trituration and precipitation.9 In the absence of CE, the ion pair
is nearly insoluble.
1 and 2 were copolymerised with ethylene dimethacrylate (Pd
: EDMA = 1 : 99) in the presence of a porogen (C6H5Cl).10 The
pyrogallol template was then cleaved with HCl and removed11
(Soxhlet, CH2Cl2) giving the white-yellow highly crosslinked
porous polymer precatalysts MIP–Pd and CE–MIP–Pd,12
which should have similar inner-spheres, and different outer-
spheres (Scheme 1). During the preparation of CE–MIP–Pd no
leaching of the polymerisable crown-ether was observed. A
control polymer missing the n-butylamine was also synthesised
to test the effect of unassociated CE sites.
The Suzuki reaction (Table 1) was chosen to assess the
presence and/or effect of the crown-ether in CE–MIP–Pd, since
the effect of coordinating the cation by the associated crown-
ether could be manifested in several points in the catalytic cycle
(highlighted in Fig. 1).13
High activities and selectivities in enzymatic reactions are
normally achieved by imposing precise geometric controls on
the enzyme–substrate complex; in the active site key catalytic
group(s) converge on the substrate to guide and manipulate
chemical reactivity. Metallo-enzymes function similarly and
utilise both their inner and outer spheres to control the metal’s
reaction profile.1 The same manipulation of the second
coordination sphere of synthetic organometallic catalysts, while
desirable, is challenging and often requires considerable
synthetic efforts.2
Molecular imprinting has recently emerged as a promising
alternative method for achieving this goal.3 This technique
relies on a three-dimensional transfer of information from a
polymerisable template to a porous highly cross-linked organic
polymer matrix. To date most MIP transition metal catalyst
reports have dealt with developing methods for associating
specifically shaped cavities to influence chemical reactivity
and/or selectivity.3,4 We sought access to polymer active sites
that additionally contained a receptor (recognition site) dis-
played in the outer sphere of the metal center (reactive site), and
report herein 1) preliminary results validating a synthetic
strategy for achieving this goal and 2) a proof of principle that
functionalised cavities can indeed influence chemical re-
activity.
To test the feasibility of outer sphere recognition occurring in
concert with catalysis, we chose to investigate a hybrid
palladium catalyst/crown-ether receptor system, since crown-
ethers (CE) have rich host–guest properties,5 and Pd-catalysts
have been demonstrated to be influenced by secondary
interactions in the coordination sphere (including CEs).6 These
A series of alkali metal bases were tested with both MIP
catalysts. As reported in Table 1, the reaction rate was closely
related to the nature of the base with the CE-modified catalyst
being more active in every case. More importantly, the
magnitude of the rate increase was strongly dependent on the
nature of the cation. Indeed, while almost no increase was
+
observed with Li2CO3 (gLi ~ 1), a much higher “gain” was
obtained with K2CO3 (gK ~ 2.5). In fact, the order (K+ > Rb+
+
† Electronic supplementary information (ESI) available: synthesis and
characterisation of compounds 1–3, MIP–Pd and CE–MIP–Pd; molecular
structure of 3. See http://www.rsc.org/suppdata/cc/b3/b309072h/
> Cs+ > Na+ > Li+) fits exactly the known affinity of alkali
metal cations for an 18,6-crown-ether.5 Control experiments
Scheme 1 Reagents and conditions: (i) a) 1 or 2, EDMA, C6H5Cl (Pd : EDMA : C6H5Cl = 1 : 99 : 100), AIBN (1%), 60 °C, 24 h; b) CH2Cl2 Soxhlet
extraction, 16 h; (ii) a) HCl in dioxane (4 M), CH2Cl2, rt, 6 h; b) CH2Cl2 Soxhlet extraction, 16 h.
3040
CHEM. COMMUN., 2003, 3040–3041
This journal is © The Royal Society of Chemistry 2003

View Article Online
Table 1 Suzuki reaction catalysed by MIP–Pd or CE–MIP–Pd
It is tempting to ascribe the rate enhancement in the above
experiments to beneficial binding between the CE and the
cation of the nucleophile as has shown by Ito in bifunctional
chiral ligands.6
Together, these results show that molecular imprinting can be
used to functionalise the second coordination sphere of a
transition metal complex and subsequently affect its catalytic
behavior. However, a consideration of the comparative rates of
metal/CE association/dissociation (fast) and polymerisation
(slow), along with the structural impreciseness of ion pairing,
suggest that the imprinted CE effect emanates despite a broad
distribution of structures in the active site of CE–MIP–Pd.
While clearly beneficial in a heterogeneous state, these
considerations suggest that more defined imprinting assemblies
will lead to even better synergism between the metal and the
recognition site, much like metallo-enzymes.
Conversion (%)a
“Pd” =
“Pd” =
+b
Entry
M2CO3
MIP–Pd
CE–MIP–Pd
gM
1
2
3
4
5
K2CO3
Rb2CO3
Cs2CO3
Na2CO3
Li2CO3
26
48
49
57
42
66 (36)c
77
2.5 (1.4)c
1.6
1.4 (1.1)c
1.2
70 (54)c
71
We gratefully acknowledge the Swiss National Science
Foundation for a postdoctoral fellowship (F. V.) and the NIH
(GM-60578). M. R. G. is a Camille-Dreyfus Teacher Scholar.
44
1.1
a Determined by GC after 6 h. b gM = conv[CE–MIP–Pd]/conv[MIP–Pd].
c A control polymer, prepared from 1 and the polymerisable CE in the
absence of n-butylamine, was used.
+
Notes and references
‡ The polymerisable palladium complex 1 with a pyrogallol ligand was
synthesised in high yield from the corresponding (isopropenyl dppe)PdCl2
and 4,6-dinitropyrogallol under biphasic CH2Cl2/aq. KOH conditions. 3,
the non-polymerisable analogue of 1, was structurally characterised (see
Supporting Information).
§ Crystal data for C32H26N2P2O7Pd 3: M = 718.91, orthorhombic, space
group Pbca (no. 61), a = 17.173(3), b = 17.584(3), c = 20.164(3) Å, U =
6088.8(18) Å3, Z = 8, m(Mo-Ka) = 0.77 mm21, T = 173 K, 55123
reflections measured, 5389 unique and 3967 observed [I > 2.5s(I), Rint
=
0.047] which were used in all calculations. The final wR(F2) was 0.053
(observed data). CCDC 216871. See http://www.rsc.org/suppdata/cc/b3/
b309072h/ for crystallographic data in .cif or other electronic format.
1 A. J. Kirby, Angew. Chem., Int. Ed. Engl., 1996, 35, 705.
2 J. K. M. Sanders, Chem. Eur. J., 1998, 4, 1378; R. Breslow and S. D.
Dong, Chem. Rev., 1998, 98, 1997; Y. Murakami, J. Kikuchi, Y.
Hisaeda and O. Hayashida, Chem. Rev., 1996, 96, 721.
3 T. Mizuki and I. Yasuhiro, J. Mol. Catal. A: Chem., 2003, 199, 115; C.
Alexander, L. Davidson and W. Yayes, Tetrahedron, 2003, 59, 2025; G.
Wulff, Chem. Rev., 2002, 102, 1; K. Severin, Curr. Opin. Chem. Biol.,
2000, 4, 710.
4 N. M. Brunkan and M. R. Gagné, J. Am. Chem. Soc., 2000, 122, 6217;
K. Polborn and K. Severin, Chem. Eur. J., 2000, 6, 4604; B. P. Santora,
A. O. Larsen and M. R. Gagné, Organometallics, 1998, 17, 3138.
5 G. W. Gokel (Ed.), Crown Ethers and Cryptands; The Royal Society of
Chemistry: Cambridge; 1991; J. M. Tinko, S. S. Moore, D. M. Walba,
P. C. Hiberty and D. J. Cram, J. Am. Chem. Soc., 1977, 99, 4207.
6 M. Sawamura and Y. Ito, Chem. Rev., 1992, 92, 857; M. Sawamura, H.
Nagata, H. Sakamoto and Y. Ito, J. Am. Chem. Soc., 1992, 114, 2586.
7 A. N. Cammidge, N. J. Baines and R. K. Bellingham, Chem. Commun.,
1999, 2588.
Fig. 1 Catalytic cycle for a typical Suzuki reaction.
with non-associated crowns (Table 1, bracketed entries 1 and 3)
indicate that both the rate acceleration and the metal ion
dependence are attenuated.
To further compare the catalytic behavior of MIP–Pd and
CE–MIP–Pd we examined the allylation of dimethyl malonate
with allyl acetate (Table 2).14 The course of this reaction was
somewhat impaired by deactivation of the catalysts after
prolonged reaction times. However, conversion after short
periods of time clearly showed the higher activity of CE–MIP–
Pd. And more interestingly, a decrease in temperature led to an
increased activity-gap between MIP–Pd and CE–MIP–Pd.15
Table 2 Allylic alkylation catalysed by MIP–Pd or CE–MIP–Pd
8 For a very recent use of a polymerisable crown-ether for MIPs, see: H.
Kim and D. A. Spivak, Org. Lett., 2003, 5, 3415.
9 Low temperature NMR spectra are broad below 240 °C and never
sharpen to reveal potential static structures.
10 For an MIP preparation at 60 °C involving a crown-ether, see: H. S.
Andersson and O. Ramström, J. Mol. Recognit., 1998, 11, 103.
11 For 2, nBuNH3Cl is removed at this stage also.
Conversion
(%)a
Entry
“Pd”
T/°C
t/h
12 For a study describing the surface area effects of porogens that solubilise
organometallic complexes, see: B. P. Santora, M. R. Gagné, K. G.
Moloy and N. S. Radu, Macromolecules, 2001, 34, 658.
13 A. Suzuki, J. Organomet. Chem., 1999, 576, 147; N. Miyaura and A.
Suzuki, Chem. Rev., 1995, 95, 2457.
1
2
3
4
MIP–Pd
CE–MIP–Pd
MIP–Pd
60
60
rt
2
2
8
8
10b
39b
9b
CE–MIP–Pd
rt
59b
14 B. M. Trost and D. L. van Vranken, Chem. Rev., 1996, 96, 395.
15 Binding to crown-ether is known to be highly temperature dependent,
see footnote 5.
a Determined by GC using n-decane as internal standard. b Formation of the
disubstituted product was observed; [mono/di] = 93/7.
CHEM. COMMUN., 2003, 3040–3041
3041