Water-soluble cyclopalladated aryl oxime: A potent 'green' catalyst

Article abstract of DOI:10.1016/S0022-328X(00)00825-1

Orthopalladated aryl oxime with a 15-crown-5 motif is prepared in a 72% yield by the exchange of cyclopalladated ligands from oxime of 4′-acetylbenzo-15-crown-5 and orthometalated N,N-dimethylbenzylamine, [Pd(o-C₆H₄CH₂NMe₂)Cl]₂. Its solubility in water is more than ten times higher than that of the related complex without the crown fragment and increases further in the presence of magnesium(II) salts. Potential applications of the newly synthesized palladacycle in catalysis is demonstrated by the example of biomimetic hydrolysis of 4-nitrophenyl 2,3-dihydroxybenzoate. It has been demonstrated that a 120-fold rate increase is achieved by realization of the intramolecular general base mechanism.

Full text of DOI:10.1016/S0022-328X(00)00825-1

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 622 (2001) 38–42
Water-soluble cyclopalladated aryl oxime: a potent ‘green’ catalyst
Ekaterina Yu. Bezsoudnova, Alexander D. Ryabov *
Department of Chemistry, Moscow State Uni6ersity, 119899, Moscow, Russia
Received 13 September 2000; received in revised form 15 September 2000
Abstract
Orthopalladated aryl oxime with a 15-crown-5 motif is prepared in a 72% yield by the exchange of cyclopalladated ligands from
oxime of 4%-acetylbenzo-15-crown-5 and orthometalated N,N-dimethylbenzylamine, [Pd(o-C6H4CH2NMe2)Cl]2. Its solubility in
water is more than ten times higher than that of the related complex without the crown fragment and increases further in the
presence of magnesium(II) salts. Potential applications of the newly synthesized palladacycle in catalysis is demonstrated by the
example of biomimetic hydrolysis of 4-nitrophenyl 2,3-dihydroxybenzoate. It has been demonstrated that a 120-fold rate increase
is achieved by realization of the intramolecular general base mechanism. © 2001 Elsevier Science B.V. All rights reserved.
Keywords: Palladium; Cyclometalation; Ligand exchange; Crown ethers; Ester; Hydrolysis; Kinetics
1. Introduction
essential not only for bioinorganic chemistry, but for
industrial catalysis as well because of severe environmen-
tal regulations [13]. We are trying to increase the solubil-
ity of cyclometalated compounds in water since 1992
[14,15]. It was first decided to use pyridine-3-sulfonic acid
as an extra ligand, the charge of which was thought to
increase the solubility. However, there was no improve-
ment, since the coordination of the py-3-SO−3 ligand
generated a zwitter-ionic species insufficiently soluble as
well. Addition of charges cannot thus be necessarily
successful, and uncharged hydrophilic groups should
alternatively be used to make a metallacycle water-solu-
ble. In this work, we demonstrate that the solubility in
water is increased by attaching a 5-crown-15 motif to the
orthopalladated aryl oxime and compound 1 becomes
sufficiently soluble and catalyzes, for example, the hy-
drolysis of 4-nitrophenyl 2,3-dihydroxybenzoate, the
Chemistry of cyclopalladated complexes has recently
got a new impulse. In addition to traditional applications
of palladacycles as starting materials in fine organic
synthesis [1–3], the compounds are nowadays frequently
used as catalysts. Impressive progresses have been
achieved in the Heck type reactions in which palladacy-
cles proved to be extremely promising [4,5]. Very re-
cently, the oxime palladacyles have been reported to be
particularly efficient as catalysts of the Heck, Stille,
Suzuki, and Ullmann reactions [6]. Our interests in
catalytic applications of oxime pallada- and platinacycles
are focused on mimicking of proteolytic [7–11] and redox
enzymes [12]. However, low solubility of the metallacy-
cles in aqueous media has hampered the progress in this
field. Solubility in ‘green’ solvents such as water is
(1)
* Corresponding author. Fax: +7-095-9395417.
E-mail address: ryabov@enz.chem.msu.ru (A.D. Ryabov).
0022-328X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(00)00825-1

E.Y. Bezsoudno6a, A.D. Ryabo6 / Journal of Organometallic Chemistry 622 (2001) 38–42
39
analogs of which are used for the modeling of catalysis
by serine proteases [16].
405 nm [19]. The reactions were run at pH 8.0 (0.01 M
phosphate buffer) and 25°C. Stock solutions of the PdII
complexes and 4 were prepared in acetonitrile as sol-
vent. The concentration of MeCN was maintained at a
6% level. Pseudo-first-order rate constants (kobs) were
calculated from the absorbance versus time (t) traces by
plotting log (Aꢀ/(Aꢀ−A)) against t. Each kobs value is
a mean value of at least three measurements.
2. Experimental
2.1. General
The equilibrium constants Kc for binding of catechol
(L) with complexes 1 and 3 were obtained by measuring
the absorbance changes at 308–320 nm at comparable
concentrations of catechol and PdII. The profiles ab-
sorbance (A) versus catechol concentration were ana-
lyzed by fitting the experimental to the square equation
(A−mPd[PdII])2−(A−mPd[PdII])([PdII]+[L]+Kc)Dm
+[PdII][L]Dm2=0
Spectrophotometric and kinetic measurements were
carried on a Hitachi 150-20 instrument with a ther-
mostated by circulating water cell compartment. 1H-
NMR spectra were recorded on a CXP-200 Bruker
instrument with a residual solvent signal as internal
standard. Oxime of 4%-acetylbenzo-15-crown-5 was pre-
pared as described elsewhere [17].
2.2. Synthesis of complex 1
where Dm=mPdL−mPd. All calculations were carried out
using a SigmaPlot 4.00 package.
Complex 3 (0.138 g, 0.25 mmol) [18] was dissolved in
7 ml of CHCl3 and 10 ml of HOAc and the oxime
solution (0.162 g, 0.5 mmol) in 3 ml of CHCl3 was
added. The mixture was kept at 50°C for 48 h. Grey–
yellow precipitate formed was filtered off, the filtrate
concentrated to 5 ml using a rotary evaporator, com-
plex 1 precipitated by slow addition of n-hexane, then
3. Results and discussion
3.1. Synthesis of complex 1
Equation 1 shows the synthetic approach to complex
1. The target compound 1 could not be obtained by
reacting the oxime with K2PdCl4 in aqueous methanol.
Poorly soluble grey–yellow material precipitated imme-
diately at room temperature, presumably an N-coordi-
nated species, since it dissolved only in excess of
pyridine to give [PdCl2(py)2] in accord with its 1H-
NMR spectrum. Other synthetic routines based on
palladation of the CꢀH bond [20] were also not promis-
ing. Therefore, the exchange of cyclopalladated ligands
was applied [21–23], which, as shown by us [24] and
other workers [25], is very efficient for the synthesis of
orthopalladated oximes. Dinuclear chloro-bridged
dimer 1 was isolated in a 72% yield and characterized
1
filtered, washed with hexane, and air dried. H-NMR
(200 MHz, l, CDCl3, J in Hz): oxime of 4%-acetylbenzo-
15-crown-5: 2.24(s, 3H, CH3), 3.77(s, 8H, 8,9,11,12CH2),
3.91(m, 4H, 6,14CH2), 4.15(m, 4H, 5,15CH2), 6.83(d, 1H,
J 8, 5%H), 7.11(dd, 1H, J 8 and 2, 4%H), 7.23(d, 1H, J 2,
2%H), 8.75(bs, 1H, NOH). 1: 1.88(s, 3H, CH3), 3.79(s,
8H, 8,9,11,12CH2), 3.95(m, 4H, 6,14CH2), 4.20(m, 4H,
5,15CH2), 6.23(s, 1H, 2%H), 7.05(s, 1H, 5%H), 8.45(s, 1H,
NOH). Anal. Found: C, 40.6; N, 2.2; H, 4.9. Calc. for
[C19H31NO6ClPd]2: C, 41.2; N, 3.0; H, 4.7%.
2.3. Kinetic and equilibrium measurements
The rate of hydrolysis of 4 was studied spectrophoto-
metrically by monitoring the 4-nitrophenolate release at
1
by the analytical data and the H-NMR spectrum. The
cyclopalladation of the CꢀH bond is proved by the
observation of the two singlets in the aromatic region.
3.2. Solubility of complex 1 in water
The solubility of complex 1 in water was tested by
adding stock solutions of 1 and its crown ether-free
analogue [Pd(o-C6H4C(Me)ꢁNOH)Cl]2 (2) in MeCN to
a buffered aqueous solution (pH 8.0, 0.01 M phos-
phate) with and without 0.01 M MgSO4 (Fig. 1). The
final concentration of MeCN in water did not exceed
6%. Maximal solubility of 2 at neutral pH in phosphate
buffer was around 5×10−4 M. The radius of the inner
,
cavity of 15-crown-5 equals 0.85 A and should be
compared with the ionic radii of Na+, Mg2+ and Li+
Fig. 1. Diagram showing solubilities of 1 and 2 in phosphate buffer
,
pH 8.0 in the absence and in the presence of 0.01 M MgSO4.
(1.02, 0.78 and 0.74 A, respectively) [26]. The best

40
E.Y. Bezsoudno6a, A.D. Ryabo6 / Journal of Organometallic Chemistry 622 (2001) 38–42
makes 4 a functioning mimetic for the simulation of
fine regulatory and/or allosteric effects encountered in
the enzymatic systems [29,30]. On the other hand,
metallacycles as C,N-chelates have a proper geometry
for complexation with 1,2-diols. They exist in a
monomeric form in MeCN and water and, hence, two
cis coordinative sites are available for the complexation
with the hydroxy groups of 4. The binding of palladiu-
m(II) species with aromatic 1,2-diols is of precedent
[31–33]. Therefore, the rate of hydrolysis of 4 (Eq. 2)
has been investigated as a function of concentration of
complexes 1–3.
(2)
Fig. 2. Dependence of the observed pseudo-first-order rate constants
for hydrolysis of 4 (2×10−5 M) against concentrations of 1–3 at pH
8.0 (phosphate buffer, 25°C).
The dependencies of kobs for reaction 2 against con-
centrations of palladacycles 1–3 are shown in Fig. 2.
As seen, oxime complexes 1 and 2 accelerate the hy-
drolysis of 4, whereas tertiary amine palladacycle 3
retards the rate. Remarkably, the kobs values display the
Michaelis type dependence and level off at higher con-
centrations of PdII indicative of the formation of reac-
tive (1 and 2) and inactive (3) intermediates. The kinetic
data is interpreted in terms of a general mechanism
presented in Scheme 1 where 4 is shown as a semi-anion
(pKa=8.8) [19]. Two types of oxime-diol palladium(II)
complexes are depicted in Scheme 1. Obviously, only B
is reactive, since the ester and oxime functionalities in
complex A are too much separated to affect the ester
cleavage.
On assumption that fast reversible formation of A
and B is characterized by similar equilibrium constants
K (Scheme 1), one arrives to the following rate expres-
Scheme 1. Proposed mechanism of hydrolysis of 4-nitrophenyl 2,3-di-
hydroxybenzoate catalyzed be 1.
sion for kobs
:
correspondence between the cavity size and the ionic
radius of Mg2+ and Li+ implies that 1 should be better
soluble in the presence of these ions. In fact, the
concentration of 1.1×10−3 M is easily achieved in 0.01
M MgSO4. For comparison, the concentration of 2 is
not higher 8×10−5 M under the same conditions.
ko+nk1K[PdII]
1+2K[PdII]
kobs
=
(3)
Here n=1 for 1 and 2 for 3. The data in Fig. 2 was
fitted to Eq. (3) and the best-fit values of equilibrium
and kinetic constants are summarized in Table 1.
As it was stated above, the hydrolysis of free 4 occurs
via the intramolecular general base mechanism
[19,28,34]. The complex formation between 4 and com-
plex 3, viz. palladacycle without nucleophile within the
C,N-chelate, leads to diminishing the nucleophilic
power of the semi-deprotonated ‘intramolecular’ base
and the activation of water is no more feasible. As a
result, there is the rate retardation by complex 3. Com-
plexes 1 and 2 accelerate the rate of hydrolysis of 4. In
contrast to 3, these do have a strong extra nucleophile,
viz. oximate ion, since the pKa of the cyclopalladated
3.3. Application of complex 1 in catalysis.
To illustrate advantages of using complex 1 as a
catalyst, we have investigated palladacycles 1–3 as ef-
fectors of the biomimetic hydrolysis of 4-nitrophenyl
2,3-dihydroxybenzoate (4). Its vicinal hydroxy groups
may serve as a trap for certain molecules which, on
coordination, are capable of both increasing or decreas-
ing the reaction rate and alter the intramolecular gen-
eral base mechanism of hydrolysis of 4 [27,28]. Such a
regulation of the reactivity by binding extra molecules

E.Y. Bezsoudno6a, A.D. Ryabo6 / Journal of Organometallic Chemistry 622 (2001) 38–42
41
Table 1
Kinetic and equilibrium constants for hydrolysis of 4 in the presence of complexes 1 and 3 at pH 8 and 25°C.
a
Complex
ko (M−1 s−1
)
k1 (M−1 s−1
)
K (M−1
)
Kc (M−1
)
1
3
(1.290.1)×10−4
(1.290.1)×10−4
(14094)×10−4
(0.3890.02)×104
(1.190.1)×104
(0.2590.12)×104
(391)×104
(0.0990.04)×10−4
a Obtained spectrophotometrically for catechol as unreactive analog of 4.
aryl oximes is around 7–8 [24]. As a result, oximate
rather than the hydroxo group functions as a general
base (the measured solvent isotope effect k(H2O)/
k(D2O) equals 2.390.2 at pD 9.0 and 1 2.5×10−4
M), and since the nucleophilicity of the metalated aryl
oximate is higher than that of the phenolic hydroxide, a
significant catalysis by cyclopalladated aryl oximes is
observed.
This study has also demonstrated a potential of pal-
ladacycles in tuning the reactivity of biochemically rele-
vant molecules and emphasized again that oximes are
very promising ligands in creation of catalytic assem-
blies [35].
Acknowledgements
The mechanism proposed involving the O,O%-bonded
intermediate is supported by the fact that hydrolysis of
4-nitrophenyl salicilate and benzoate is not catalyzed by
1 or 2 under the same conditions. Independent spec-
trophotometric measurements of the binding between
catechol as an unreactive analog of 4, on one hand, and
1 and 3, on the other (low solubility precludes the
measurements for 2 and the binding constant cannot
also be extracted from the kinetic data in Fig. 2) are
also in accord with the mechanism in Scheme 1. The
effective values of the equilibrium constants K obtained
for 1 and 3 from the kinetic data equal 0.38×104 and
1.1×104 M−1, respectively. Spectrophotometric titra-
tion of 1 and 3 by catechol provided the Kc values of
0.25×104 and 3×104 M−1, respectively. The values
obtained in the kinetic and equilibrium experiments are
quite comparable despite different molecules, i.e. 4 and
catechol, were used. The coincidence is particularly
striking for complex 1. The mechanism in Scheme 1 is
also in accord with the observations that platinum(II)
analog of 2, the complex [Pt(o-C6H4C(Me)ꢁNOH)-
Cl(py)], is catalytically inactive and does not show
evidence for binding with catechol. It is also worth
mentioning the magnitudes of both acceleration and
retardation effects achieved in the presence of com-
plexes 1 and 3, respectively. The ratio of the rate
constants k1/ko indicates that the coordinated palla-
dated oxime 1 increases the rate 116-fold, whereas the
inhibiting effect of 3 is a factor of 14. The entire range
of the reactivity change is thus a factor of ca. 1600.
Financial support from the Russian Foundation for
Basic Research (Grant No. 98-03-33023a) and INTAS
(Project 97-0166) is gratefully acknowledged. We thank
Drs I.K. Sakodinskaya and G.M. Kazankov for fruitful
discussions and Dr V.A. Polyakov for the NMR
measurements.
References
[1] A.D. Ryabov, Synthesis (1985) 233.
[2] M. Pfeffer, Recl. Trav. Chim. Pays-Bas 109 (1990) 567.
[3] J. Spencer, M. Pfeffer, in: L.S. Liebeskind (Ed.), Advances in
Metal-Organic Chemistry, vol. 6, Jai Press, Stamford, 1998, p.
103.
[4] W.A. Herrmann, V.P.W. Bo¨hm, C.-P. Reisinger, J. Organomet.
Chem. 576 (1999) 23.
[5] I.P. Beletskaya, A.V. Cheprakov, Chem. Rev. 100 (2000) 3009.
[6] D.A. Alonso, C. Najera, M.C. Pacheco, Org. Lett. 2 (2000) 1823.
[7] E.V. Krooglyak, G.M. Kazankov, S.A. Kurzeev, V.A. Polyakov,
A.N. Semenov, A.D. Ryabov, Inorg. Chem. 35 (1996) 4804.
[8] G.M. Kazankov, O.G. Dyachenko, A.V. Nemukhin, A.D.
Ryabov, Mendeleev Commun. (1997) 159.
[9] A.D. Ryabov, G.M. Kazankov, S.A. Kurzeev, P.V. Samuleev,
V.A. Polyakov, Inorg. Chim. Acta 280 (1998) 57.
[10] S.A. Kurzeev, G.M. Kazankov, A.D. Ryabov, Inorg. Chim.
Acta 305 (2000) 1.
[11] G.M. Kazankov, V.S. Sergeeva, E.N. Efremenko, L. Alexan-
drova, S.D. Varfolomeev, A.D. Ryabov, Angew. Chem. Int. Ed.
Engl. 39 (2000) 3117.
[12] L. Alexandrova, O.G. D’yachenko, G.M. Kazankov, V.A.
Polyakov, P.V. Samuleev, E. Sansores, A.D. Ryabov, J. Am.
Chem. Soc. 122 (2000) 5189.
[13] P.T. Anastas, J.C. Warner, Green Chemistry: Theory and Prac-
tice, Oxford University Press, Oxford, 1998.
[14] M. Schmu¨lling, A.D. Ryabov, R. van Eldik, J. Chem. Soc.
Chem. Commun. (1992) 1609.
3.4. Conclusion
[15] M. Schmu¨lling, A.D. Ryabov, R. van Eldik, J. Chem. Soc.
Dalton Trans. (1994) 1257.
[16] W.P. Jencks, Catalysis in Chemistry and Enzymology, Dover,
New York, 1987.
[17] M. Tada, H. Hirano, A. Suzuki, Bull. Chem. Soc. Jpn. 53 (1980)
2304.
[18] A.D. Ryabov, V.A. Polyakov, A.K. Yatsimirsky, J. Chem. Soc.
Perkin Trans. 2 (1983) 1503.
Sufficiently water soluble orthopalladated aryl oxime
1 obtained by the exchange of cyclopalladated ligands
provided a possibility for equilibrium and kinetic stud-
ies of the catalyzed hydrolysis of 4 in a wide catalyst
concentration range which cannot principally be ac-
complished for less the soluble aryl oxime complex 2.

42
E.Y. Bezsoudno6a, A.D. Ryabo6 / Journal of Organometallic Chemistry 622 (2001) 38–42
[19] A.K. Yatsimirsky, K.Y. Bezsoudnova, I.K. Sakodinskaya,
Bioorg. Med. Chem. Lett. 3 (1993) 635.
[27] M.L. Bender, F.J. Kezdy, B. Zerner, J. Am. Chem. Soc. 85
(1963) 3017.
[20] A.D. Ryabov, Chem. Rev. 90 (1990) 403.
[21] A.D. Ryabov, A.K. Yatsimirsky, Inorg. Chem. 23 (1984)
789.
[28] B. Capon, B.C. Ghosh, J. Chem. Soc. Sect. B (1966) 472.
[29] A. Fersht, Enzyme Structure and Mechanism, W.H. Freeman,
San Francisco, 1977.
[22] A.D. Ryabov, Inorg. Chem. 26 (1987) 1252.
[23] A.D. Ryabov, in: A.F. Williams, C. Floriani, A.E. Merbach
(Eds.), Perspectives in Coordination Chemistry, VCH Weinheim,
New York, 1992, p. 271.
[24] A.D. Ryabov, G.M. Kazankov, A.K. Yatsimirsky, L.G.
Kuz’mina, O.Y. Burtseva, N.V. Dvortsova, V.A. Polyakov, In-
org. Chem. 31 (1992) 3083.
[30] J. Ricard, A. Cornish-Bowden, Eur. J. Biochem. 166 (1987) 255.
[31] O. Gandolfy, J. Blum, Inorg. Chim. Acta 80 (1983) 103.
[32] G.M. Kapteijn, A. Dervis, D.M. Grove, H. Kooijman, M.T.
Lakin, A.L. Spek, G. van Koten, J. Am. Chem. Soc. 117 (1995)
10939.
[33] R. Roy, P. Chattopadhyay, C. Sinha, S. Chattopadhayay, Poly-
hedron 15 (1996) 3361.
[25] K. Selvakumar, S. Vancheesan, B. Varghese, Polyhedron 16
(1997) 2257.
[26] S.A. Oehrle, J. Chromatogr. Sect. A 745 (1996) 87.
[34] M.N. Khan, J. Mol. Catal. 40 (1987) 195.
[35] V.Y. Kukushkin, A.J.L. Pombeiro, Coord. Chem. Rev. 181
(1999) 147.
.