Full text of DOI:10.1016/S0040-4039(01)02099-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 411–414
Metal coordination and stacking effects in supramolecular
catalysis. Effects of structural variations of copper complexes for
the hydrolysis of phosphate esters
Shigeru Negi and Hans-Jo¨rg Schneider*
FR Organische Chemie der Universita¨t des Saarlandes, D 66041 Saarbru¨cken, Germany
Received 18 August 2001; revised 2 November 2001; accepted 5 November 2001
Abstract—Until now barely explored stacking effects with rate constant enhancements by up to two orders of magnitudes and
total accelerations of about 107 are found by introduction of aromatic substituents at the N atoms of ethylenediamine of Cu(II)
complexes. Ligands containing more than two nitrogen atoms show as an often overlooked factor a strong dependence of the
efficiency towards metal coordination. © 2002 Elsevier Science Ltd. All rights reserved.
The rational design of catalysts for the hydrolysis of
phosphate esters is of significance in view of the wide
occurrence of related biocides. In addition, it can help
to develop new chemical nucleases such as artificial
restriction enzymes, and to better understand enzymatic
mechanisms.1 Stacking between aromatic entities has
evolved to be a non-covalent interaction of prime
importance in synthetic2 and biological3 complexes, but
to the best of our knowledge has until now, not been
explored systematically as a possible contributor in
supramolecular catalysts. We wish to report here on
rate enhancements by two orders of magnitude due to
stacking and on related effects by lipophilic interactions
with alkyl groups, with total rate constant enhance-
ments of at least 107 compared to metal-free solutions.
Solvent effects can be used to characterize the underly-
ing mechanisms. In addition, we observe striking differ-
ences due to a variable metal coordination, with rate
constant differences spanning several orders of
magnitude.
up to 120.5 Complexes with macrocyclic ligands can
lead to very significant rate effects, but are also ham-
pered by dimerization.6 If we now increase the size of
the aromatic side group R in the naphthalene derivative
(ligand 6, Scheme 1) we reach a factor of krel=kobs/k0=
400 compared to krel=1.00 with ligand 1 (R=H), and
krel=18 with ligand 2 (R=benzyl) (details see Table 1;
it should be noted that with the ligand 3 for solubility
reasons 7.5% DMSO had to be added, which dimin-
ishes the rates, see below). A similar, although smaller,
increase is found if alkyl instead of aryl side groups are
introduced (ligands 7 and 8). The corresponding
lipophilic interactions are weaker with isopropyl- and
tert-butyl substituents (8 and 9) which may be due to
the fact that only part of the methyl groups can pro-
duce contacts with the phenyl parts of the substrate.
Surprisingly small catalytic effects are seen with the
Copper(II) complexes of ethylenediamine (en) have
been shown4 to enhance the hydrolysis of bis(p-nitro-
phenyl)phosphate (BNPP, see Fig. 1) by a factor which
is difficult to measure due to solubility problems with
the Cu(II) salt alone; introduction of e.g. positively
charged aminomethyl phenyl groups at the nitrogen
atoms of en leads to additional rate enhancements by
Keywords: phosphate ester hydrolysis; catalysis; metal-coordination
effects; stacking effects.
* Corresponding author. E-mail: ch12hs@rz.uni-sb.de
Figure 1. Schematic structure of the ternary complex between
BNPP, a metal(II) cation M and ligand 3.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)02099-8

412
S. Negi, H.-J. Schneider / Tetrahedron Letters 43 (2002) 411–414
Table 1. Hydrolysis of BNPP with Cu2+ complexes
Ligand
kobs [10−6 s−1
]
krel (kobs/ko)
1
6
0.75 (ko)
13.8
75.2
10.6
0.21
300
4.6
8.7
5.7
3.3
0.82
0.14
4.7
0.28
1.8
1.0
18.3
100
14.0
0.3
400
6.1
11.4
7.5
4.0
1.1
0.2
6.3
2
6
3a
6
4
6
5
6
6b
6
7
6
8
6
9
6
10c
11
12
13
14
15
16
17
0.4
2.4
6.5
4.9
B0.1d
Reagents and conditions: pH 7.0; 0.01
M
EPPS buffer; 75°C;
[BNPP]=4.0×10−5 M; [metal complex]=2.0×10−4 M (ligand:metal=
1:1); observed reaction usually about 1000 min; error in kobs 2%
(from duplication).
a EPPS buffer with 7.5% DMSO.
b Calculated from reactions up to about 20% conversion due to
solubility problems.
c EPPS+2.5% DMSO.
d The kobs is too small for measurement.
propylene derivatives 13 to 14, which must be due to
different complex structures.7
That the metal-complex structure has a strong influence
on the catalytic activity of copper complexes is already
obvious from results with ligands such as 5; here, in
contrast to ligand 4, the pyridyl nitrogen atoms provide
binding sites for complete saturation of the Cu coordi-
nation sphere. This aspect was analyzed furthermore
with ligand 17 copper complexes; although there could
be one free coordination site for the substrate, for
intermediates or for the products we find an extremely
low catalytic activity. The same is observed with corre-
sponding tetra- and penta-amines (not shown here). It
is obvious that only with substituted ethylenediamines
the-kinetic and/or thermodynamic-stability of tetra-
coordinated Cu complexes is not so large as to inhibit
catalytic activity of the copper complexes. (It should be
noted that fully saturated CuL2 can also be formed if
one uses 1:1 ratios of ligand and Cu salts, with forma-
tion of free Cu(II) which is known to be catalytically
close to inactive). In consequence, one needs a distor-
tion of the Cu coordination enforced by a suitable
configuration of nitrogen atoms in a ligand, if one
wants to use polyamine–copper complexes for cataly-
sis.8 Rate enhancements with zinc(II) complexes with
ligands were so small that they were not analyzed
further. It should be noted that reaction with metal-free
ligands are slower by many orders of magnitudes; in
consequence, the catalyst ligands are not used up by
covalent reactions with the substrates.
Scheme 1.
most simple explanation why here the observed stack-
ing effects on the hydrolysis are significantly smaller
(Table 2). The opposite can be expected for the rate of
the electroneutral tris (p-nitrophenyl)–phosphate
(TNPP),9 which was also investigated. Sufficient solu-
bility could only be reached with TNPP by adding
DMSO (Table 3), which severely lowers the efficiency
of the catalyst (see below). If we compare, however, the
rate effects on BNPP and TNPP in the same solvent
(25% DMSO, Tables 3 and 4) it becomes apparent that
indeed the stacking effects are more pronounced with
the more lipophilic substrate TNPP.
Table 2. Hydrolysis of NPP with Cu2+ complexes
Ligand
kobs [10−5 s−1
]
krel (kobs/ko)
1
6
0.90 (ko)
1.28
5.33
2.0
0.74
1.0
1.4
5.9
2.2
0.81
2
6
3a
6
4
6
5
6
Mono (p-nitrophenyl)-phosphate (NPP) is a more
hydrophilic substrate than BNPP, which offers the
Reagents and conditions: see Table 1.
a EPPS buffer with 7.5% DMSO.

S. Negi, H.-J. Schneider / Tetrahedron Letters 43 (2002) 411–414
Table 3. Hydrolysis of TNPP with Cu2+ complexes
413
Conclusions. The introduction of cofactors in catalytic
metal complexes can in analogy to metal enzymes lead
to significantly improved performance.12 The present
results show that large rate enhancements can material-
ize by the use of stacking effects; attempts to further
increase the underlying aromatic surfaces with
anthrylderivatives yielded insoluble complexes even in
mixtures rich in DMSO. Lack of solubility prevented
also saturation kinetics which would allow to analyze
which steps of the catalytic cycle are stabilized by
stacking. Introduction of e.g. positively charged nitro-
gen centers could in the future remedy the solubility
problem and lead to additional electrostatic stabiliza-
tion. Transition metal-based catalysts like those pre-
sented here gain their activity only by suitable ligands,
but their design is limited by the need to provide free
coordination sites for the metal. Further studies are
planned to establish the nature of the underlying cata-
lytic complexes.13
Ligand
kobs [10−5 s−1
]
krel (kobs/ko)
1
6
0.95 (ko)
3.12
5.94
2.26
1.39
1.0
3.3
6.3
2.4
1.4
2
6
3
6
4
6
5
6
Reagents and conditions: see Table 1, but here 40.0°C; EPPS buffer
with 25% DMSO.
Table 4. Solvent effects on hydrolysis of BNPP with 3
6
-
Cu2+ complexes
Solvent
Sp
EN
kobs (10−5 s)
log kobs
T
Ethylene glycol
Diethylene glycol 0.244*
DMSO
Acetonitrile
Methanol
Ethanol
DMF
NMP
Propane-1-ol
Propane-2-ol
Dioxane
Trifluoro
methanol
0.376
0.790
0.713
0.444
0.460
0.762
0.654
0.404
3.28
1.49
0.97
0.57
1.43
1.22
0.80
0.75
1.12
1.23
0.38
0.91
−4.48
−4.82
−5.01
−5.24
−4.84
−4.91
−5.09
−5.13
−4.95
−4.91
−5.42
−5.04
0.227
0.217
0.120
0.144
0.138
0.122
0.108
0.099
0.079
Acknowledgements
Our work was supported by the Deutsche Forschungs-
gemeinschaft, Bonn, and the Fonds der Chemischen
Industrie, Frankfurt. Additional experiments by Laura
J. Whiting and for ligand 6 by E. Gogritchiani and T.
Liu are much appreciated.
0.617
0.546
Reagents and conditions: pH 7.0; 0.01 M EPPS buffer with 25%
organic solvent; 75°C; [BNPP]=4.0×10−5 M; [metal-complex]=2.0×
10−4 M (ligand:metal=1:1); error kobs 3%.
References
1. For recent reviews see: (a) Hegg, E. L.; Burstyn, J. N.
Coord. Chem. Rev. 1998, 173, 133–165; (b) Kimura, E.;
Koike, T. Adv. Inorg. Chem. 1997, 44, 229; (c)
Komiyama, M.; Takeda, N.; Shigekawa, H. Chem. Com-
mun. 1999, 1443; (d) Cowan, J. A. Chem. Rev. 1998, 3,
1067; (e) Williams, N. H.; Tkasaki, B.; Wall, M.; Chin, J.
Acc. Chem. Res. 1999, 32, 485; (f) Kirby, A. J. Angew.
Chem., Int. Ed. Engl. 1996, 35, 707; (g) Nelson, D. J.
Coord. Chem. Rev. 1996, 149, 95; (h) Go¨bel, M. W.
Angew. Chem. 1994, 106, 1201; Angew. Chem., Int. Ed.
Engl. 1994, 33, 1141; (i) Molenveld, P.; Engbersen, J. F.
J.; Reinhoudt, D. N. Chem. Soc. Rev. 2000, 29, 75–86.
2. Recent examples: (a) Balaban, T. S.; Eichhofer, A.; Lehn,
J.-M. Chem. Eur. J. 2000, 5, 4047; (b) Munukata, M.;
Ning, G. L.; Suenaga, Y.; Kuroda-Sowa, T. Angew.
Chem., Int. Ed. 2000, 39, 4555; Angew. Chem. 2000, 112,
4729.
The influence of the solvent had to be measured with
binary water mixtures containing not more than 25% of
the organic phase, as with higher contents the reactions
became too slow. Although the high water content of
75% attenuates the expected rate variations signifi-
cantly, the observed rate constants (Table 4) differ by a
factor of almost ten, with the most hydrophilic
ethyleneglycol solution at one end, and the least polar
1,4-dioxane mixture at the other end. Correlations with
known solvent hydrophobicity or polarity parameters
like Sp10 or ET were both very poor, but the results
11
demonstrate the importance of water as a reaction
medium for biologically important catalytic processes.
In line with stacking as a major contribution, the
solvent effect increases with increasing size of the aro-
matic side group at the catalyst ligand (Table 5).
3. (a) Kool, E. T.; Morales, J. C.; Guckian, K. M. Angew.
Chem., Int. Ed. 2000, 39, 990; (b) Kool, E. T.; Morales, J.
C.; Guckian, K. M. Angew. Chem. 2000, 112, 1046; (c)
Guckian, K. M.; Schweitzer, B. A.; Ren, X. F. S.; Sheils,
C. J. J. Am. Chem. Soc. 2000, 122, 2213.
Table 5. Hydrolysis of BNPP with several Cu2+ com-
plexes in EPPS buffer containing 25% DMSO
4. Morrow, J. R.; Trogler, W. C. Inorg. Chem. 1988, 27,
3387.
5. Rammo, J.; Schneider, H.-J. Inorg. Chim. Acta 1996, 251,
125.
6. See for example: Hegg, E. L.; Burstyn, J. N. Coord.
Chem. Rev. 1998, 173, 133–165 and references cited
therein.
Ligand
(1a
kobs [10−6 s−1
]
krel (kobs/ko)
6
0.75 (ko)
6.81
9.72
1.0)
9
12.8
8.2
2
6
3
6
8
6
6.22
7. Literature binding constants of ethylene- and propylene-
diamine with copper(II) are similar and large enough to
Reagents and conditions: see Table 1.
a Measurement in pure water.

414
S. Negi, H.-J. Schneider / Tetrahedron Letters 43 (2002) 411–414
provide for full complexation under the reaction condi-
10. Abraham, M. H.; Grellier, P. L.; McGill, R. A. J. Chem.
Soc., Perkin Trans. 2 1988, 339.
11. Reichardt, C. Solvent Effects in Organic Chemistry; VCH
tions (lgK=9.7 or 10.5, respectively, for the ML com-
plex, and 16.9 or 19.6, respectively, for the ML2
complex).
Weinheim: New York, 1979.
12. See: Roigk, A.; Schneider, H.-J. Eur. J. Org. Chem. 2000,
205 and references cited therein.
8. See for example: (a) Kavan, M.; Powell, D. R.; Burstyn,
J. N. Inorg. Chim. Acta 2000, 297, 351–361; (b) Chand,
D. K.; Schneider, H.-J.; Aguilar, J. A.; Escart´ı, F.; Gar-
c´ıa-Espan˜a, E. Inorg. Chim. Acta 2001, 316, 71–78 and
references cited therein.
9. For the hydrolysis of phosphoric acid triesters see for
example: (a) Hay, R. W.; Govan, N. Polyhedron, 1997,
16, 4233–4237; (b) Hay, R. W.; Govan, N.; Norman, P.
R. Transition Metal Chem. 1998, 23; 133–138, and refer-
ences cited therein.
13. Synthetic details and analytical data of the ligands, as far
they are not commercially available, will be reported
elsewhere. The complexes were obtained by mixing ligand
(as hydrochloride) and copper(II)sulfate with conditions
as given at Table 1. Kinetic measurements were per-
formed as described before (Ref. 10) and evaluated by
non-linear least-square fitting to a first order equation,
without deviation over usual several half-life times.