Zinc complex chemistry of N,N,O ligands providing a hydrophobic cavity

Article abstract of DOI:10.1021/ic050042u

Three new highly substituted bis(2-picolyl)(2-hydroxybenzyl)amine ligands were synthesized, and their biomimetic zinc complex chemistry was explored. They have tert-butyl substituents at the 3-and 5-positions of their phenyl rings, and they bear one phenyl group (HL²), two methyl groups (HL ³), or two phenyl groups (HL⁴) at the 6-positions of their pyridyl rings. Their reactions with hydrated zinc perchlorate yield three distinctively different complex types. L² forms a trigonal-bipyramidal aqua complex, and L³, a square-pyramidal aqua complex. The substituents on L⁴ leave no room for a water ligand, and the resulting zinc complex is trigonal-monopyramidal with a vacant coordination site. The water ligands on the L²Zn and L³Zn units can be replaced by anionic halide, thiocyanate, p-nitrophenolate, benzoate, and organophosphate as well as uncharged pyridine ligands. The L⁴Zn unit forms labile halide, p-nitrophenolate, and pyridine complexes. Triethylamine converts the aqua complexes to the labile hydroxides L²Zn-OH and L³Zn-OH, and in polar media [L³Zn-OH₂] ⁺ seems to be in equilibrium with L³Zn-OH. The hydroxides, but not the water complexes, effect the hydrolytic cleavage of tris(p-nitrophenyl) phosphate to bis(p-nitrophenyl) phosphate. The kinetic investigation of the cleavage reactions has shown them to be second-order reactions, thereby supporting the proposed four-center mechanism.

Full text of DOI:10.1021/ic050042u

Inorg. Chem. 2005, 44, 3321−3329
Zinc Complex Chemistry of N,N,O Ligands Providing a Hydrophobic
Cavity
Florian Gross and Heinrich Vahrenkamp*
Institut fu¨r Anorganische und Analytische Chemie der UniVersita¨t Freiburg, Albertstrasse 21,
D-79104 Freiburg, Germany
Received January 12, 2005
Three new highly substituted bis(2-picolyl)(2-hydroxybenzyl)amine ligands were synthesized, and their biomimetic
zinc complex chemistry was explored. They have tert-butyl substituents at the 3-and 5-positions of their phenyl
rings, and they bear one phenyl group (HL2), two methyl groups (HL3), or two phenyl groups (HL4) at the 6-positions
of their pyridyl rings. Their reactions with hydrated zinc perchlorate yield three distinctively different complex types.
L2 forms a trigonal-bipyramidal aqua complex, and L3, a square-pyramidal aqua complex. The substituents on L4
leave no room for a water ligand, and the resulting zinc complex is trigonal-monopyramidal with a vacant coordination
site. The water ligands on the L2Zn and L3Zn units can be replaced by anionic halide, thiocyanate, p-nitrophenolate,
benzoate, and organophosphate as well as uncharged pyridine ligands. The L4Zn unit forms labile halide,
p-nitrophenolate, and pyridine complexes. Triethylamine converts the aqua complexes to the labile hydroxides
L2Zn OH and L3Zn OH, and in polar media [L3Zn OH2]+ seems to be in equilibrium with L3Zn
− − − −OH. The hydroxides,
but not the water complexes, effect the hydrolytic cleavage of tris(p-nitrophenyl) phosphate to bis(p-nitrophenyl)
phosphate. The kinetic investigation of the cleavage reactions has shown them to be second-order reactions,
thereby supporting the proposed four-center mechanism.
Introduction
boron anchors, the groups of Parkin,6,7 Carrano,8,9 and
Burzlaff10,11 have provided tridentate ligands for tetrahedral
zinc model complexes. Using weakly coordinating nitrogen
atoms as the centers of ligand construction, the groups of
Fenton12-15 and Ichikawa16 synthesized various tripodal
ligands which form trigonal-bipyramidal zinc complexes with
pseutotetrahedral coordination geometries. Although con-
vincing structural enzyme models and in some cases catalytic
hydrolytic activity could be achieved with these complexes,5
the key model species in the form of a (N,N,O)Zn-OH
complex or a truly tetrahedral (N,N,O)Zn‚OH2 complex have
eluded researchers until now.
The N,N,O donor motif provided by histidine and glutamate
or aspartate is a very common one in hydrolytic zinc
enzymes,1,2 including their most prominent representative,
carboxypeptidase.3 Accordingly, many polydentate ligands
containing N and O donors have been prepared in attempts
to model the zinc chemistry associated with them. Early
attempts in this field4 were not very successful because
neither the coordination number of zinc could be kept low
nor a hydrophobic encapsulation of the metal could be
provided, which are essential prerequisites for the formation
of the Zn-OH nucleophile and for catalytic activity of the
complexes.
(6) Dowling, C.; Parkin, G. Polyhedron 1996, 15, 2463.
(7) Ghosh, P.; Parkin, G. J. Chem. Soc., Dalton Trans. 1998, 2281.
(8) Higgs, T. C.; Spartalian, K.; O’Connor, C. J.; Matzanke, B. F.; Carrano,
C. J. Inorg. Chem. 1998, 37, 2263.
(9) Hammes, B. S.; Carrano, C. J. Inorg. Chem. 1999, 38, 4593.
(10) Beck, A.; Weibert, B.; Burzlaff, N. Eur. J. Inorg. Chem. 2001, 521.
(11) Hegelmann, I.; Beck, A.; Eichhorn, C.; Weibert, B.; Burzlaff, N. Eur.
J. Inorg. Chem. 2003, 339.
In recent years zinc enzyme modeling has evolved rather
quickly to a higher level of sophistication.5 This holds
particularly for the area of tripodal ligands with N,N,O donor
sets. Using ligand construction around central carbon or
(12) Adams, H.; Bailey, N. A.; Campbell, I. K.; Fenton, D. E.; He, Q.-Y.
J. Chem. Soc., Dalton Trans. 1996, 2233.
(13) Adams, H.; Bailey, N. A.; Rodriguez de Barbarin, C. O.; Fenton, D.
E.; He, Q.-Y. J. Chem. Soc., Dalton Trans. 1995, 2323.
(14) Adams, H.; Bailey, N. A.; Fenton, D. E.; He, Q.-Y. J. Chem. Soc.,
Dalton Trans. 1996, 2857.
(15) Rodriguez de Barbarin, C. O.; Bailey, N. A.; Fenton, D. E.; He, Q.-
Y. J. Chem. Soc., Dalton Trans. 1997, 161.
(16) Ogawa, K.; Nakata, K.; Ichikowa, K. Chem. Lett. 1998, 797.
* Author to whom correspondence should be addressed. E-mail:
vahrenka@uni-freiburg.de.
(1) Zinc Enzymes; Spiro, T. G., Ed.; Wiley: New York, 1983.
(2) Metal Ions in Biological Systems, Vol. 15: Zinc and its Role in Biology
and Nutrition; Sigel, H., Ed.; Marcel Dekker: New York, 1983.
(3) Christianson, D. W.; Lipscomb, W. N. Acc. Chem. Res. 1989, 22, 62.
(4) Brown, R. S.; Huguet, J.; Curtis, N. J. In ref 2, pp 55-99.
(5) Parkin, G. Chem. ReV. 2004, 104, 699.
10.1021/ic050042u CCC: $30.25
Published on Web 04/05/2005
© 2005 American Chemical Society
Inorganic Chemistry, Vol. 44, No. 9, 2005 3321

Gross and Vahrenkamp
Table 1. Bond Lengths (Å) and Angles (deg) in the Hydrated L‚Zn
For our investigations in this field we have used ligand
construction around central nitrogen atoms to obtain tripods
of general type A. Our initial attempts with the ligands
dipicolylglycine17 and dipicolyl-â-alanine18 also suffered from
limited control of the coordination number of zinc due to
aggregation or coligand association. The situation improved
when we turned to the phenolate-containing ligands derived
from B.18 Particularly the bis(tert-butyl)-substituted system
L1 allowed us the isolation of a zinc-aqua complex, proof
of its conversion to a zinc-hydroxide complex, and the use
of the latter for the hydrolytic cleavage of tris(p-nitrophenyl)
phosphate with second-order kinetics.19
Perchlorates
L1Zn(OH2)+ L2Zn(OH2)+ L3Zn(OH2)+
L4Zn+
(3)
19
(1)
(2)
Zn-O(OH2)
Zn-O1
2.059(6)
1.945(6)
2.099(7)
2.061(7)
2.178(7)
2.025(7)
1.892(7)
2.180(8)
2.114(8)
2.162(7)
2.011(2)
2.025(2)
2.204(3)
2.064(3)
2.145(3)
1.877(2)
2.034(2)
2.031(2)
2.076(2)
Zn-N1
Zn-N2
Zn-N3
N3-Zn-O(OH2) 165.0(3)
169.4(1)
90.9(2)
75.9(3)
145.2(1)
N3-Zn-O1
N3-Zn-N1
N3-Zn-N2
N2-Zn-O1
N1-Zn-O1
N1-Zn-N2
94.7(2)
78.8(3)
89.4(1)
76.6(1)
85.0(1)
155.1(1)
98.5(1)
100.6(1)
100.0(1)
85.3(1)
82.4(1)
131.5(1)
114.3(1)
114.2(1)
79.4(3)
80.6(3)
112.1(2)
114.6(2)
129.6(3)
120.5(2)
115.5(3)
118.9(2)
mixtures. Crystallization rendered complexes 1-3 in yields
of 50-75%. All three show the ligand attachment by the
typical high-energy shift of the ν(pyridine) vibration bands
to 1607, 1609, and 1608 cm-1, respectively, and the presence
of anionic perchlorate by its intense ν(ClO) vibration band
near 1100 cm-1. A typical feature in the 1H NMR spectra is
the splitting of the pyridylmethylene resonances into doublets
with geminal coupling upon complexation of the HL ligands,
as observed before19 and for all ZnL complexes here.
Complexes 1 and 2 showed a 1H NMR signal for zinc-bound
water in the 2.9 ppm range. This signal was missing in the
1H NMR spectrum of 3, which instead showed a signal for
noncoordinated water at 1.53 ppm. This was the basis for
the given formulations of 1-3, which were subsequently
confirmed by the structure determinations.
We have now extended this work by attaching further
substituents at the ligands of type B, with the purpose of
improving the encapsulation of zinc in the ligand pocket.
Methyl and phenyl groups were placed at the 6-postions of
the pyridyl rings, yielding ligands L2, L3, and L4. This paper
reports the consequences of this in terms of structures and
reactivities of the zinc complexes derived from the new
ligands.
[L2Zn(OH2)]ClO4 [L3Zn(OH2)]ClO4 [L4Zn]ClO4
1
2
3
The structures of all four L‚Zn perchlorate compounds
are closely related, including the nonhydrated complex 3;
see Table 1. Yet there are significant differences among
details. One of them concerns the coordination geometry:
while 1 is close to ideally trigonal-bipyramidal (84% TBP
according to the Holmes method20), 2 is almost perfectly
square-pyramidal (73% SP20). The situation in L1Zn(OH2)+19
closely resembles that in L2Zn(OH2)+. In a comparison of
complexes 1 and 3, it becomes evident that the latter can be
described as trigonal-monopyramidal; i.e., it is characterized
by a vacant coordination site. The bond lengths show a
considerable spread among the four complexes: both the 0.15
Å variation of the Zn-O(phenolate) bond lengths and the
fact that the typical elongation of the axial bonds does not
occur in 1 and 3 are remarkable. Part of this may be related
to the fact that hydrogen-bonding interactions exist between
the perchlorate ions and the oxygen donors of 1 and 2.
The structures of the complex cations of 1-3 are displayed
in Figures 1-3. It is evident that the zinc ions (and in case
of 1 the aqua ligand as well) are encapsulated between the
tert-butyl and phenyl substituents of ligands L2 and L4. This
is not the case for ligand L3, whose methyl substituents are
much less space-filling. The consequences of this for the
aqua ligands in 1 and 2 are comparable. While in 1 there is
Results and Discussion
Ligand Syntheses. The preparation of HL2, HL3, and HL4
followed the scheme developed by us for the reference
compound HL1.19 First the substituted dipicolylamines were
synthesized by condensation between the corresponding
pyridine-2-aldehyde and the corresponding 2-(aminomethyl)-
pyridine and subsequent reduction with sodium borohydride.
Then the dipicolylamines were combined with 2,4-di-tert-
butyl-6-(chloromethyl)phenol. HL2, HL3, and HL4, which
are colorless solids, had to be purified by chromatography.
They are characterized by intense ν(pyridine) vibration bands
at 1591, 1593, and 1588 cm-1, respectively. Their methylene
groups give rise to singlet 1H NMR resonances in the range
3.75-4.00 ppm.
Complexation of Zinc Perchlorate. All three protonated
ligands HL were treated with Zn(ClO4)2‚6H2O and an
equimolar amount of KOH in methanol/dichloromethane
(17) Abufarag, A.; Vahrenkamp, H. Inorg. Chem. 1995, 34, 2207.
(18) Tro¨sch, A.; Vahrenkamp, H. Eur. J. Inorg. Chem. 1998, 827.
(19) Tro¨sch, A.; Vahrenkamp, H. Inorg. Chem. 2001, 40, 2305.
(20) Holmes, R. R. Acc. Chem. Res. 1979, 12, 257.
3322 Inorganic Chemistry, Vol. 44, No. 9, 2005

Zinc Complex Chemistry of N,N,O Ligands
trigonal axis of the coordination polyhedron, such that agostic
interactions can be ruled out. It is noteworthy that there is a
water molecule in the environment of the complex, but it is
located on the far side of the tripodal ligand and hydrogen-
bonded to the perchlorate anion.
Trigonal-monopyramidal coordination is not unprec-
edented in zinc chemistry. Borovik reported the structures
of two (tris(carbamoylmethyl)aminato)zinc complexes.21,22
In addition a copper complex of tris(6-phenylpicolyl)amine
has this coordination.23 In both cases small rodlike coligands
such as NO or CH3CN could be incorporated in the ligand
cavity, demonstrating that despite the limited access the metal
center may be available for catalytic activity.
Figure 1. Structure of the complex cation of 1.
Attachment of Anionic Ligands. The water ligand in its
hydrophobic environment could be expected to be a good
leaving group. This was easily verified for complexes 1 and
2. Reactions with the corresponding potassium salts yielded
4a-c as well as 5a,b. Likewise p-nitrophenolate, p-nitroben-
zoate, and bis(p-nitrophenyl) phosphate were attached in
complexes 4d-f as well as 5d-f. Complexes a-c can be
called simple models for the inhibition of zinc enzymes by
simple anions. Complexes d-f were prepared as reference
compounds for the hydrolytic cleavage reactions described
below.
Figure 2. Structure of the complex cation of 2.
L2Zn-X L3Zn-X
4a-f
5a,b,d-f
(a) X ) Cl, (b) X ) I, (c) X ) SCN,
(d) X ) p-nitrophenolate, (e) X ) p-nitrobenzoate,
(f) X ) bis(p-nitrophenyl) phosphate
Complexes 4 and 5 are uncharged molecular species with
five-coordinate zinc. Their IR and 1H NMR spectra confirm
(by comparison with 1 and 2) the tripodal attachment of L2
or L3. For d-f they show the bands for the aromatic
coligands. It is to be assumed that, like in complexes 1-3,
the coordination geometries in complexes 4 and 5 represent
various stages of the transition between trigonal-bipyramidal
and square-pyramidal. For 4d, the structure of which was
determined (see Supporting Information), it is about halfway.
The two independent molecules of 4d in the asymmetric unit
show geometries which according to the Holmes method20
calculate as 60% and 66% square-pyramidal.
Figure 3. Structure of the complex cation of 3.
an open site for access to the zinc-bound water, due to the
missing substituent on the second pyridine ring, in 2 the tert-
butyl group, the two methyl groups, and the zinc-bound water
lie roughly in one plane, thereby making the aqua ligand
accessible from the open side of this plane.
Anionic coligands could also be attached to complex 3.
Reactions with the corresponding salts and quick precipita-
tion yielded the products 6a-c:
In complex 3 the sterical demands of the phenyl substit-
uents obviously are so large that the situation becomes
unfavorable for the binding of a water molecule at the zinc
ion. As a result, the coordination geometry is truly trigonal-
monopyramidal, and the zinc ion is actually placed outside
the plane defined by its three equatorial donors by 0.02 Å.
Due to the lower coordination number, all zinc-ligand bond
lengths are shorter than their counterparts in the other three
complexes of Table 1. The two phenyl substituents are
positioned face-to-face, approximating a stacking interaction
with distances of 3.41 Å between their para-carbons and
4.28 Å between their ipso-carbons. Both have one hydrogen
within contact range of the zinc ion (2.65 and 3.06 Å), but
both these hydrogens are displaced considerably from the
L4ZnX
6a-c
(a) X ) Br, (b) X ) I, (c) X ) p-nitrophenolate
While the compositions of 6a-c are evident from their
spectra and elemental analyses, their structures could not be
(21) Hammes, B. S.; Ramos-Maldonado, D.; Yap, G. P. A.; Liable-Sands,
L.; Rheingold, A. L.; Young, V. G.; Borovik, A. S. Inorg. Chem. 1997,
36, 3210.
(22) Ray, M.; Hammes, B. S.; Yap, G. P. A.; Rheingold, A. L.; Borovik,
A. S. Inorg. Chem. 1998, 37, 1527.
(23) Chuang, C.-L.; Lim, K.; Chen, Q.; Zubieta, J.; Canary, J. W. Inorg.
Chem. 1995, 34, 2562.
Inorganic Chemistry, Vol. 44, No. 9, 2005 3323

Gross and Vahrenkamp
determined. After precipitation the compounds dissolve only
in very polar solvents in which they slowly decompose and
from which they cannot be recovered in a pure or crystalline
form. This may indicate that 6a-c are salts [L4Zn]X with
the same L4Zn cation as 3. However, the solid-state IR
spectra of the complexes show vibration bands for both
coordinated pyridine rings around 1605 cm-1 and uncoor-
1
dinated pyridine rings around 1590 cm-1. Likewise the H
NMR spectra of 6a,b contain singlet signals for the picolyl
methylene groups in addition to the doublets, while that of
6c is dominated by decomposition products. As discussed
above, the doublet signals indicate zinc coordination of the
pyridine rings, while the singlets are observed for the
uncoordinated ligands L. The intensity ratio of the singlets
and doublets, however, is not 1:1. This may again indicate
decomposition of the complexes with release of the free
ligands. Yet alternatively it may point to another structure
(possibly of an equilibrium constituent) in which the zinc
ion is only four-coordinate due to the noncoordination of
one of the pyridyl rings, depicted as
Figure 4. Structure of the complex cation of 7b. Relevant bond lengths
(Å) and angles (deg): Zn-O1 1.927(6), Zn-N1 2.142(7), Zn-N2 2.217-
(6), Zn-N3 2.113(7), Zn-N4 2.093(7); O1-Zn-N1 105.5(2), O1-Zn-
N2 107.7(2), O1-Zn-N3 96.1(2), O1-Zn-N4 105.4(3), O1-Zn-O2
175.0(3), N1-Zn-N2 140.9(3), N3-Zn-N4 158.3(3).
expected when the carbonyl oxygen is attached to a metal.
Yet the lowering of the wavenumbers is not in the 30-45
cm-1 range observed for confirmed cases of N,O-chelating
2-acetylpyridine.27
The attachment of the pyridine donors was confirmed by
structure determinations of 7a,b. The zinc coordination in
7a (see Supporting Information) is halfway between trigonal-
bipyramidal and square-pyramidal (55% TBP according to
the Holmes method20) with a Zn-N(pyridine) bond length
of 2.10 Å. The structure of 7b is displayed in Figure 4. The
five donor atoms around zinc define a square-pyramidal
coordination to a good approximation (80% SP according
to the Holmes method20) with O1 at the apex. However, the
interaction between zinc and the carbonyl oxygen (Zn‚‚‚O2
) 2.56 Å) cannot be ignored, giving the zinc ion a seriously
distorted octahedral coordination. This Zn‚‚‚O2 interaction
is nevertheless a rather weak one when compared to similar
ones in zinc acetyl- and benzoylpyridine complexes,24,26,29
where Zn-O bond distances around 2.1 Å were observed.
One reason for the orientation of the zinc-bound pyridine
ligand in 7a,b is π-stacking with the phenyl substituent of
the tripod ligand L2, with distances of 3.4-3.8 Å between
opposing carbon atoms. This also explains the location of
the acetyl group (which is in the plane of the pyridine ring)
without invoking attractive interactions between zinc and
oxygen. π-Stacking may also be the reason the pyridine
adducts 9a,b are more stable than their counterparts 6a-c
with anionic coligands. While the steric stress due to the
two phenyl substituents on L4 may labilize the Zn-anion
interactions in 6a-c, the presence of π-stacking beetween
the pyridine ligand and both phenyl groups in 9a,b may
outweigh this labilization.
Attachment of Pyridine Donors. The water ligands in
complexes 1 and 2 were also replaced easily by pyridine
ligands, and likewise could pyridine ligands be added to
complex 3. Pyridine and 2-acetylpyridine were used, the latter
with the purpose of finding out whether the formation of a
favorable five-membered chelate ring and of a octahedrally
coordinated zinc ion would induce coordination of the
carbonyl oxygen, a very weak donor which was found by
us both to coordinate24-27 and not to coordinate25,28 to zinc.
Combination of the reagents in methanol or dichloromethane
yielded complexes 7-9.
[L2Zn‚Py′]ClO4 [L3Zn‚Py′]ClO4 [L4Zn‚Py′]ClO4
7a,b
8a
9a,b
(a) Py′ ) pyridine, (b) Py′ ) 2-acetylpyridine
These complexes are stable in the solid state and in
solution. This includes 9a,b, in contrast to the lability of the
L4-derived complexes 6. 7-9 indicate the zinc coordination
of all pyridine donors present by vibration bands at or above
1600 cm-1. The IR data are inconclusive with respect to a
coordination of the carbonyl function in 7b and 9b. The
corresponding vibration bands (1690 and 1675 cm-1) are
lower than that of free 2-acetylpyridine at 1700 cm-1, as
Hydrolytic Activity. We had already found that the aqua
complex of L1 can be deprotonated to form the labile
hydroxide complex.19 The latter in turn was found to be the
active species in the hydrolytic cleavage of tris(p-nitrophenyl)
phosphate (TNP). One reason for the synthesis of L2, L3,
and L4 was the hope that these ligands, due to the better
(24) Mu¨ller, B.; Vahrenkamp, H. Eur. J. Inorg. Chem. 1999, 137.
(25) Mu¨ller, B.; Schneider, A.; Tesmer, M.; Vahrenkamp, H. Inorg. Chem.
1999, 38, 1900.
(26) Sudbrake, C.; Vahrenkamp, H. Inorg. Chim. Acta 2001, 318, 23.
(27) Schneider, A.; Vahrenkamp, H. Z. Anorg. Allg. Chem. 2003, 629, 2122.
(28) Tesmer, M.; Shu, M.; Vahrenkamp, H. Inorg. Chem. 2001, 40, 4022.
(29) Sheldrick, G. M.; Stotter, D. A. J. Chem. Soc., Dalton Trans. 1975,
666-668.
3324 Inorganic Chemistry, Vol. 44, No. 9, 2005

Zinc Complex Chemistry of N,N,O Ligands
encapsulation of a zinc-bound water molecule, would
facilitate its deprotonation and stabilize the resulting L‚Zn-
OH complexes. The accessibility of the water complexes 1
and 2 allowed us to test this for L2 and L3.
with formation of 4d and 5d (see above), which are also
yellow with UV band maxima at 392 and 380 nm. Perform-
ing the hydrolyses in this way did not lead to free bis(p-
nitrophenyl) phosphoric acid, as evident by the absence of
its 31P NMR resonance at -10.2 ppm.
Treatment of 1 or 2 with equimolar amounts of triethyl-
amine in chloroform or acetonitrile resulted in small, but
These quantitative interconversions enabled a kinetic study
which was performed with the purpose of comparing the
hydrolytic strength of complexes 10 and 11 with that of other
Zn-OH2 or Zn-OH complexes investigated by other
groups14,30-33 and ourselves.19,34-36 Pseudo-first-order kinetic
data were obtained by treating TNP with large excesses of
10 or 11 in acetonitrile. As observed above, under these
conditions both cleavage products p-nitrophenolate and bis-
(p-nitrophenyl) phosphate become zinc-bound in complexes
4d/4f and 5d/5f, respectively. Control experiments also
confirmed that p-nitrophenolate fixation to 4d and 5d is much
faster than the formation of the phosphate complexes 4f and
5f. The latter, however, according to the four-center mech-
anism established for such hydrolytic cleavages,36 are the
primary cleavage products resulting without liberation of
anionic bis(p-nitrophenyl) phosphate. This was favorable for
obtaining the kinetic data which were gathered by recording
the intensities of the UV-vis absorption bands of 4d and
5d, the only colored species in the reaction mixtures. The
rate of formation of 4d and 5d is thus identical with the rate
of disappearance of TNP, and the pseudo-first-order rate
constants are obtained using the equation
1
significant, shifts of the H NMR resonances (see Experi-
mental Section), allowing the conclusion that deprotonation
with formation of 10 and 11 has occurred. But this conclusion
could not be supported by a characteristic sharp ν(OH)
vibration band in the IR spectra near 3600 cm-1, nor were
10 or 11 isolable by precipitation. Attempts to separate the
hydroxide complexes from the triethylammonium perchlorate
byproduct yielded cloudy precipitates presumed to contain
Zn(OH)2 and solutions in which the 1H NMR signals of the
uncoordinated ligands L were observed. Thus, as observed
for L1Zn-OH before,19 complexes 10 and 11 evade isolation
by decomposition, and their presumed identity rests solely
on their reactivity.
L2Zn-OH
L3Zn-OH
10
11
The reactivity studies, however, revealed a distinction
between 1 and 2, the precursors of 10 and 11. When 1 was
treated with TNP in CD3CN, no interconversion could be
1
detected by H or 31P NMR. Equimolar mixtures of 2 and
TNP, in contrast, were interconverted to new compounds
within a few hours. The initial 31P NMR signal of TNP at
-18.7 ppm vanished and was replaced by the signal of bis-
(p-nitrophenyl) phosphoric acid at -10.2 ppm. Addition of
triethylamine to the resulting colorless solution effected an
immediate color change to yellow. This color change did
not occur when bis(p-nitrophenyl) phosphoric acid alone was
treated with triethylamine. We interpret these observations
by the presence of 11 as an equilibrium species in solutions
of 2. As observed for all our (tripod)Zn-OH complexes,5,19
11 then is the nucleophilic agent cleaving TNP and producing
bis(p-nitrophenyl) phosphate and p-nitrophenol. The former
is protonated by residual water in the solution, which also
regenerates complex 2. The latter is colorless and turns
yellow when deprotonated by triethylamine. The conclusion
that 2 is more acidic than 1 is in agreement with their
substitution patterns, the two methyl substituents in 11
offering a better stabilization of the resulting hydroxide
complex by hydrophobic encapsulation than the single phenyl
substituent in 10.
The hydroxide complexes 10 and 11, generated in situ,
proved their nucleophilic strengh by the cleavage of TNP.
Stoichiometric mixtures of 1 or 2, triethylamine, and TNP
in acetonitrile underwent quantitative conversions with
formation of the bis(p-nitrophenyl) phosphate complexes 4f
and 5f (see above), which were indentified by their 31P NMR
resonances at -12.6 and -11.5 ppm. The byproduct of these
hydrolyses is p-nitrophenol. In the presence of excess
triethylamine, this becomes p-nitrophenolate with its char-
acteristic yellow color and its UV band maximum at 402
nm. When both 1 or 2 and triethylamine are present in excess,
the liberated p-nitrophenolate is also fixed by complexation
ln[1 - (It/I∞)] ) -kobs
t
The values of kobs for both cleavage reactions were
obtained for five different concentrations each of 10 and 11
in the presence of 0.025 mM of TMP. The resulting plots
are given in the Supporting Information. They define clean
second-order reactions. The least-squares lines show a small,
but not insignificant, intercept of the ordinate, indicating that
hydrolysis reactions occur also in the absence of 10 or 11,
presumably by traces of water. The dominating reactions,
the TNP cleavages by the LZn-OH complexes, are char-
acterized by the second-order rate constants k2 of 0.34 M-1
s-1 for 10 and 0.67 M-1 s-1 for 11.
These two k2 values indicate that 10 and 11 are indeed
strong nucleophiles for TNP hydrolysis, taking into account
that both substrate and nucleophile as well as the solvent of
the reactions were nonpolar. Alkaline hydrolysis of nitro-
phenyl phosphates in solvent mixtures containing water was
found to be faster but only around 1 order of magnitude.30,31,37
When zinc complexes of other polydentate ligands were used,
whose hydrolytically active constituent is assumed to be a
(30) Gellmann, S. H.; Petter, R.; Breslow, R. J. Am. Chem. Soc. 1986,
108, 2388.
(31) Koike, T., Kimura, E. J. Am. Chem. Soc. 1991, 113, 8935.
(32) Ito, T.; Fujii, Y.; Tada, T.; Yoshikawa, Y.; Hisada, H. Bull. Chem.
Soc. Jpn. 1996, 69, 1265.
(33) Jurek, P.; Martell, A. E. Inorg. Chim. Acta 1999, 287, 47.
(34) Weis, K.; Rombach, M.; Ruf, M.; Vahrenkamp, H. Eur. J. Inorg.
Chem. 1998, 263.
(35) Weis, K.; Vahrenkamp, H. Eur. J. Inorg. Chem. 1998, 271.
(36) Rombach, M.; Maurer, C.; Weis, K.; Keller, E.; Vahrenkamp, H.
Chem.sEur. J. 1999, 5, 1013.
(37) Cox, J. R.; Ramsay, O. B. Chem. ReV. 1964, 64, 317.
Inorganic Chemistry, Vol. 44, No. 9, 2005 3325

Gross and Vahrenkamp
(ligand)Zn-OH species, the reaction rates for the hydrolyses
in water/organic solvent mixtures were found to be 5-8
times faster than here, and it was observed that solvent
polarity has a strong influence on the reaction rate.14,30-32
Our own investigations on the cleavage of TNP by the
pyrazolylborate complex TpPh,MeZn-OH in chloroform yielded
a k2 value of 1.55 M-1 s-1.36 The fact that this cleavage is
3-5 times faster than that by 10 or 11 can be explained by
the fact that the TpZn-OH complexes exist as such at neutral
pH in contrast to the LZn-OH complexes applied here
which in the absence of an additional base are LZn(OH2)+.
The closest relative of L2Zn-OH (10) and L3Zn-OH (11)
is L1Zn-OH, which has yielded a k2 value for TNP
hydrolylsis in chloroform of 0.27 M-1 s-1.19 Considering that
such reactions in acetonitrile are 3-5 times faster than in
chloroform,36 the “corrected” k2 value for L1Zn-OH should
be around 1.0 M-1 s-1. We then have an influence of L on
the rate of the TNP hydrolysis by LZn-OH in the order
L1 > L3 > L2. This sequence reflects both steric and
hydrophobic effects. The accessibility of the Zn-OH center
decreases in the order L1 > L2 > L3, while the hydrophobic
encapsulation increases the nucleophilicity of the zinc-bound
hydroxide in the reverse order, i.e., L1 < L2 < L3. The
combination of both effects makes L2Zn-OH the slowest
reagent and leads to comparable reactivities of L1Zn-OH
and L3Zn-OH.
Experimental Section
General Information. The general working and measuring
techniques were as described previously.38 Starting materials were
obtained commercially; 2,4-di-tert-butyl-6-(chloromethyl)phenol,39
6-phenylpyridine-2-aldehyde,40 6-phenyl-2-(aminomethyl)pyri-
dine,22 and 6-methyl-2-(aminomethyl)pyridine21,41 were prepared
according to the published procedures.
The 1H NMR data for ligands L2, L3, and L4 are virtually
identical for all their complexes. Therefore, full spectra are reported
only for two cases each (1/4a, 2/5a, 3/6a), and for the other
1
complexes only the H NMR data for the coligands are given. A
frequent problem with the elemental analyses of this kind of
complexes is that the carbon values are outside the accepted range.
When this was the case here, at least one additional analysis value
(N or S) was determined.
Ligands HL. HL2. A solution of 6-phenylpyridine-2-aldehyde
(1.33 g, 7.26 mmol) in ethanol (35 mL) was cooled with ice. A
solution of 2-(aminomethyl)pyridine (743 µL, 785 mg, 7.26 mmol)
in ethanol (35 mL) was added dropwise with stirring. The mixture
was stirred 30 min at 0 °C and then 2 h at room temperature. Then
it was cooled to 0 °C again and NaBH4 (549 mg, 14.5 mmol) was
added in small portions. After the mixture was stirred overnight at
room temperature, 20 mL of 5 M hydrochloric acid were slowly
added and the mixture stirred for 1 h at room temperature. Then 2
M sodum hydroxide was added until the solution was strongly
alkaline. This mixture was extracted three times with 25 mL of
dichloromethane each. The organic phase was dried over sodium
sulfate, filtered, and evaporated to dryness. The raw (6-phenyl-2-
pyridylmethyl)(2-pyridylmethyl)amine was disolved in dioxane (40
mL) and treated with triethylamine (3.57 mL, 2.57 g, 25.4 mmol).
After the solution was cooled to 0 °C, a solution of 2,4-di-tert-
butyl-6-(chloromethyl)phenol (1.85 g, 7.26 mmol) in dioxane (40
mL) was added dropwise with stirring, upon which a colorless
precipitate formed. After the mixture was stirred for 15 h at room
temperature and filtration, the solvent was evaporated in vacuo.
The resulting yellow oil was chromatographed over a 2 × 10 cm
silica gel column with cyclohexane/ethyl acetate/triethylamine (3:
1:1). HL2 was eluted with a Rf value of 0.59. Removal of the solvent
in vacuo left behind 2.37 g (66%) of HL2 as a colorless solid, mp
Conclusions
This work has brought the zinc complex chemistry of the
tetradentate tripodal N,N,O-ligands L to a stage that is
comparble with that for the 3-substituted tris(pyrazolyl)-
borates. While in the complexes of the unsubstituted ligand
the zinc ion is exposed to the environment, the step-by-step
attachment of organic substituents around it in ligands L1-
L4 results in a progressive encapsulation. In the extreme case
of L4 the encapsulation has become so severe that complexes
L4Zn-X bind the coligands X only weakly, and the most
stable species is the cation [L4Zn]+ with a vacant coordina-
tion site.
90 °C. IR (KBr; cm-1): 1591 (s) (CdN). H NMR (CDCl3): δ
1
1.24 (s, 9H, t-Bu), 1.43 (s, 9H, t-Bu), 3.86 (s, 2H, CH2-phenol),
3.89 (s, 2H, CH2-py), 3.93 (s, 2H, CH2-py), 6.85-7.68 (m, 11H,
Ar), 8.08 [dd, J ) 8.0/1.5 Hz, 2H, Ph(2,6)], 8.50 (d, J ) 5.0 Hz,
1H, HR), 10.76 (s, 1H, OH). 13C NMR (CDCl3): δ 30.0 [Me(t-
Bu)], 32.1 [Me(t-Bu)], 34.5 [C(t-Bu)], 35.4 [C(t-Bu)], 59.0 (CH2),
59.7 (CH2), 60.2 (CH2), 121.9, 122.3, 122.7, 123.6, 124.4, 124.7
(phenol), 127.6, 129.1, 129.3 (phenyl), 136.2, 137.2, 137.6, 139.7,
140.9, 148.9, 154.3, 157.5, 157.9, 156.2 (pyridyl).
The main purpose of the enhanced encapsulation, the
protection and activation of a zinc-bound water molecule,
was achieved. There seems to be an increasing ease of
deprotonation of the [LZn-OH2]+ complexes in going from
L1 to L3. Although we still failed to determine the corre-
sponding pKa values or to isolate one of the resulting LZn-
OH complexes, there is evidence that L3Zn-OH exists in
equilibrium with its parent complex [L3Zn-OH2]+.
The results of this work again underline the observation
that the zinc-hydroxide complexes are the active species in
the hydrolytic cleavage of biologically relevant substrates.
As shown here for the hydrolysis of tris(p-nitrophenyl)
phosphate, the ease of formation of the LZn-OH species
corresponds with the ease of substrate cleavage, and as the
kinetic data have shown, the hydrolytic activity can rise
despite reduced access to the LZn-OH center due to steric
hindrance.
Anal. Calcd for C33H39N3O (Mr ) 493.69): C, 80.25; H, 7.96;
N, 8.51. Found: C, 79.97; H, 8.05; N, 8.34.
HL3. This was synthesized like HL2 from 6-methylpyridine-2-
aldehyde (1.43 g, 11.8 mmol), 6-methyl-2-(aminomethyl)pyridine
(1.44 g, 11.8 mmol), NaBH4 (1.32 g, 35.0 mmol), triethylamine
(5.79 mL, 4.17 g, 41.2 mmol), and 2,4-di-tert-butyl-6-(chloro-
methyl)phenol (3.00 g, 11.8 mmol). Chromatography (Rf ) 0.39)
yielded 4.2 g (81%) of HL3 as a colorless solid, mp 84 °C. IR
(KBr; cm-1): 1593 (vs) (CdN). 1H NMR (CDCl3): δ 1.26 (s, 9H,
(38) Fo¨rster, M.; Burth, R.; Powell, A. K.; Eiche, T.; Vahrenkamp, H. Chem.
Ber. 1993, 126, 2643.
(39) Abufarag, A.; Vahrenkamp, H. Inorg. Chem. 1995, 34, 3279.
(40) Parks, J. E.; Wagner, B. E.; Holm, R. H. Inorg. Chem. 1971, 10, 2472.
(41) Fuentes, O.; Paudler, W. W. J. Org. Chem. 1975, 40, 1210.
3326 Inorganic Chemistry, Vol. 44, No. 9, 2005

Zinc Complex Chemistry of N,N,O Ligands
t-Bu), 1.46 (s, 9H, t-Bu), 2.56 (s, 6H, CH3), 3.76 (s, 2H, CH2-
phenol), 3.84 (s, 4H, CH2-py), 6.88 (d, J ) 2.3 Hz, 1H, HR), 6.96-
7.20 (m, 5H, Ar), 7.49 (t, J ) 7.7 Hz, 2H, Hγ), 10.65 (s, 1H, OH).
13C NMR (CDCl3): δ 24.5 [Me(py)], 30.0 [Me(t-Bu)], 32.1 [Me-
(t-Bu)], 34.5 [C(t-Bu)], 35.4 [C(t-Bu)], 58.3 (CH2), 59.9 (CH2),
120.6, 122.0, 122.4, 123.4, 125.2 (phenol), 135.9, 137.3, 140.5,
154.3, 158.1, 158.2 (pyridyl).
Complex 3. This was synthesized like 1 from Zn(ClO4)2‚6H2O
(248 mg, 0.67 mmol), HL4 (380 mg, 0.67 mmol), and KOH (37.4
mg, 0.67 mmol). After removal of the KClO4 precipitate by
filtration, the filtrate was evaporated to 5 mL, upon which a
colorless powder precipitated. This was filtered off, washed with a
small amount of ice-cold methanol, dissolved in dichloromethane
(10 mL), and treated with n-heptane until the first traces of
precipitation showed up. Slow evaporation of the solvent from the
open flask resulted in the precipitation of 245 mg (49%) of 3 as
colorless crystals, mp 280 °C (dec). IR (KBr; cm-1): 1608 (s) (Cd
N), 1096 (vs) (ClO4). 1H NMR (CDCl3): δ 1.30 (s, 9H, t-Bu), 1.52
(s, 2H, H2O), 1.58 (s, 9H, t-Bu), 4.14 (s, 2H, CH2-phenol), 4.50
(d, J ) 17.3 Hz, 2H, CH2-py), 4.73 (d, J ) 17.3 Hz, 2H, CH2-py),
7.00 (d, J ) 2.5 Hz, 1H, HR), 7.22-7.68 (m, 15H, Ar), 8.01 (t,
J ) 7.8 Hz, 2H, Hγ).
Anal. Calcd for C39H42ClN3O5Zn‚H2O (Mr ) 733.62 + 18.02):
C, 62.32; H, 5.90; N, 5.59. Found: C, 62.22; H, 5.89; N, 5.52.
Attachment of Anionic Ligands. Complex 4a. A solution of 1
(37.0 mg, 0.056 mmol) in methanol (10 mL) was added with stirring
to a solution of KCl (4.64 mg, 0.062 mmol) in methanol (10 mL).
After the mixture was stirred for 16 h, the fine precipitate of KClO4
was filtered off and the filtrate evaporated to dryness. The colorless
residue was dissolved in dichloromethane (2 mL), and n-heptane
(5 mL) was added. Slow evaporation in the open flask resulted in
the precipitation of 18 mg (54%) of 4a as colorless crystals, mp
150 °C (dec). IR (KBr; cm-1): 1604 (s) (CdN). 1H NMR
(CDCl3): δ 1.20 (s, 9H, t-Bu), 1.43 (s, 9H, t-Bu), 3.17-4.51 (m,
6H, CH2), 5.29 (s, 2H, CH2Cl2), 6.76 (d, J ) 2.6 Hz, 1H, HR),
6.87 (d, J ) 7.8 Hz, 1H, Hδ), 7.02 (d, J ) 2.6 Hz, 1H, Hγ), 7.11
(m, 1H, Hâ), 7.25 (d, J ) 7.8 Hz, 1H, Hδ), 7.54 (m, 4H, Ar), 7.65
(d, J ) 7.8 Hz, 1H, Hâ), 7.88 (t, J ) 7.8 Hz, 1H, Hγ), 8.28 (m, 2H,
Ar), 8.77 (d, J ) 5.3 Hz, 1H, HR).
Anal. Calcd for C29H39N3O (Mr ) 445.65): C, 78.16; H, 8.82;
N, 9.43. Found: C, 77.94; H, 8.89; N, 9.25.
HL4. This was synthesized like HL2 from 6-phenylpyridine-2-
aldehyde (3.06 g, 16.4 mmol), 6-phenyl-2-(aminomethyl)pyridine
(3.02 g, 16.4 mmol), NaBH4 (1.86 g, 49.1 mmol), triethylamine
(8.05 mL, 5.80 g, 57.3 mmol), and 2,4-di-tert-butyl-6-(chloro-
methyl)phenol (4.17 g, 16.4 mmol). Chromatography (4:1 cyclo-
hexane/triethylamine 4:1; Rf ) 0.53) yielded 8.12 g (87%) of HL4
as a colorless solid, mp 64 °C. IR (KBr; cm-1): 1588 (s) (CdN).
1H NMR (CDCl3): δ 1.27 (s, 9H, t-Bu), 1.47 (s, 9H, t-Bu), 3.96
(s, 2H, CH2-phenol), 4.01 (s, 4H, CH2-py), 6.89 (d, J ) 2.3 Hz,
1H, HR), 7.21 (d, J ) 2.3 Hz, 1H, Hγ), 7.26-7.69 (m, 12H, Ar),
8.10 [dd, J ) 7.9/1.6 Hz, 2H, Ph(2,6)], 11.00 (s, 1H, OH). 13C
NMR (CDCl3): δ 29.6 [Me(t-Bu)], 31.6 [Me(t-Bu)], 34.1 [C(t-
Bu)], 34.9 [C(t-Bu)], 58.5 (CH2), 59.3 (CH2), 119.0, 121.3, 122.2,
123.3, 124.2 (phenol), 127.1, 128.7, 128.9 (phenyl), 137.2, 139.2,
154.1, 157.0 (pyridyl).
Anal. Calcd for C39H43N3O (Mr ) 569.79): C, 82.21; H, 7.61;
N, 7.37. Found: C, 81.71; H, 7.74; N, 7.23.
Zinc Perchlorate Reactions. Complex 1. A solution of Zn-
(ClO4)2‚6H2O (1.39 g, 3.72 mmol) in methanol (30 mL) was
dropped slowly with stirring into a solution of HL2 (1.84 g, 3.72
mmol) in dichloromethane (50 mL). After the solution was stirred
for 10 min, a solution of KOH (209 mg, 3.72 mmol) in methanol
(15 mL) was added, upon which a colorless precipitate formed.
After it was stirred for 3 h, the mixture was filtered and the filtrate
evaporated to dryness. The residue was recrystallized from hot
methanol (10 mL) after cooling to 0 °C. A yield of 153 g (62%) of
1 resulted as colorless crystals, mp 155 °C. IR (KBr; cm-1): 3396
(m, br) (H2O), 1607 (s) (CdN), 1104 (vs) (ClO4). 1H NMR
(CDCl3): δ 1.22 (s, 9H, t-Bu), 1.40 (s, 9H, t-Bu), 2.94 (s, 2H,
H2O), 3.35-4.55 (m, 6H, CH2), 6.82 (d, J ) 2.5 Hz, 1H, HR),
7.04 (d, J ) 7.8. Hz, 1H, Hδ), 7.09 (d, J ) 2.5 Hz, 1H, Hγ), 7.21
(m, 1H, Hâ), 7.39 (d, J ) 7.8 Hz, 1H, Hδ), 7.64 (m, 5H, Ar), 8.01
(t, J ) 7.8 Hz, 1H, Hγ), 8.21 (m, 2H, Ar), 8.41 (d, J ) 4.5 Hz, 1H,
HR).
Anal. Calcd for C33H38ClN3OZn‚CH2Cl2 (Mr ) 593.53 +
84.93): C, 60.19; H, 5.94; N, 6.19. Found: C, 59.84; H, 6.07; N,
6.06.
Complex 4b. This was synthesized like 4a from 1 (35.0 mg,
0.060 mmol) and KI (10.0 mg, 0.060 mmol). Yield: 25 mg (65%)
of 4b as colorless needles. Mp: 146 °C. IR (KBr; cm-1): 1603 (s)
(CdN).
Anal. Calcd for C33H38IN3OZn‚CH2Cl2 (Mr ) 684.98 + 84.93):
C, 53.04; H, 5.24; N, 5.46. Found: C, 52.62; H, 5.43; N, 5.41.
Complex 4c. This was synthesized like 4a from 1 (200 mg, 0.296
mmol) and KSCN (28.8 mg, 0.296 mmol). Yield: 115 mg (63%)
of 4c as a colorless powder. Mp: 210 °C. IR (KBr; cm-1): 1605
(s) (CdN).
Anal. Calcd for C34H38N4OSZn (Mr ) 616.16): C, 66.28; H,
6.22; N, 9.09; S, 5.20. Found: C, 65.56; H, 6.29; N, 9.00; S, 5.17.
Complex 4d. A solution of p-nitrophenol (30.9 mg, 0.222 mmol)
and KOH (12.6 mg, 0.222 mmol) in methanol (20 mL) was added
slowly with stirring to a solution of 1 (150 mg, 0.222 mmol) in
dichloromethane (30 mL). After 5 h of stirring, the solution was
filtered and the filtrate evaporated to dryness. The residue was
dissolved in dichloromethane (5 mL), and methanol (5 mL) was
added. Evaporation in an open flask yielded 89 mg (58%) of 4d as
yellow crystals, mp 260 °C (dec). IR (KBr; cm-1): 1603 (s) (Cd
N), 1292 (vs) NO2). 1H NMR (CDCl3): δ 6.49 [d, J ) 9.3 Hz, 2H,
Ph(2,6)], 7.05-7.91 [m, 14H, therein Ph(3,5)].
Anal. Calcd for C33H40ClN3O6Zn (Mr ) 675.54): C, 58.67; H,
5.97; N, 6.22. Found: C, 58.13; H, 6.25; N, 6.08.
Complex 2. This was synthesized like 1 from Zn(ClO4)2‚6H2O
(1.61 g, 4.32 mmol), HL3 (1.92 g, 4.32 mmol), and KOH (242
mg, 4.32 mmol). The raw product was dissolved in acetone (5 mL),
and diethyl ether was added dropwise until a fluffy precipitate
formed. This was filtered off and discarded. Further addition of
diethyl ether to the filtrate led to the slow formation of a colorless
powder. After the mixture was stirred overnight, the powder was
filtered off and dried in vacuo, leaving behind 1.32 g (74%) of 2
as a colorless powder, mp 188 °C. IR (KBr; cm-1): 3474 (m, br)
1
(H2O), 1609 (s) (CdN), 1108 (vs) (ClO4). H NMR (CDCl3 with
one drop of DMSO-d6): δ 1.09 (s, 9H, t-Bu), 1.19 (s, 9H, t-Bu),
2.66 (s, 6H, CH3), 2.85 (s, br, 4H, H2O), 3.84 (s, 2H, CH2-phenol),
4.06 (d, J ) 16.1 Hz, 2H, CH2-py), 4.18 (d, J ) 16.1 Hz, 2H,
CH2-py), 6.73 (s, br, 1H, HR), 6.85 (d, J ) 2.8 Hz, 1H, Hγ), 6.94-
7.12 (m, 4H, Ar), 7.55 (t, J ) 7.5 Hz, 2H, Hγ).
Anal. Calcd for C39H42N4O4Zn (Mr ) 696.18): C, 67.29; H, 6.08;
N, 8.05. Found: C, 67.13; H, 6.10; N, 8.02.
Complex 4e. This was synthesized like 4d from p-nitrobenzoic
acid (24.7 mg, 0.148 mmol), triethylamine (20.8 µL, 15.0 mg, 0.148
mmol), and 1 (100 mg, 0.148 mmol). Recrystallization from ethanol
at -20 °C yielded 78 mg (68%) of 4e as a yellow solid, mp 225
Anal. Calcd for C29H40ClN3O6Zn‚H2O (Mr ) 627.40 + 18.02):
C, 53.96; H, 6.56; N, 6.51. Found: C, 54.23; H, 6.47; N, 6.52.
Inorganic Chemistry, Vol. 44, No. 9, 2005 3327

Gross and Vahrenkamp
°C (dec). IR (KBr; cm-1): 1625 (m) (CdN), 1594 (m, COO), 1343
(vs, NO2). 1H NMR (CDCl3): δ 7.49 [m, 7H, therein Ph(2,6)], 8.00
[d, J ) 8.5 Hz, 2H, Ph(3,5)].
mg (52%) of 6a resulted as a colorless powder, mp 270 °C (dec).
IR (KBr; cm-1): 1602 (m) (pyridine coordinated), 1588 (m)
(pyridine uncoordinated). 1H NMR (CDCl3): δ 1.25 (s, 9H, t-Bu),
1.54 (s, 9H, t-Bu), 3.89 (s, 2H, CH2-phenol), 4.41 (d, J ) 15.8 Hz,
2H, CH2-py), 4.61 (d, J ) 15.6 Hz, 2H, CH2-py), 6.71 (d, J ) 2.5
Hz, 1H, HR), 7.27 (m, 3H, Ar), 7.57 (m, 6H, Ar), 7.73 (d, J ) 7.7
Hz, 2H, Hâ), 7.88 (d, J ) 7.7 Hz, 2H, Hγ), 8.13 (m, 4H, Ar). In
addition to these signals attributable to 6a further signals with
nonintegral intensity values are observed as described in the text.
Anal. Calcd for C39H42BrN3OZn (Mr ) 714.07): C, 65.60; H,
5.93; N, 5.88. Found: C, 65.37; H, 6.01; N, 5.83.
Anal. Calcd for C40H42N4O5Zn‚2H2O (Mr ) 724.19 + 36.03):
C, 63.20; H, 6.10; N, 7.37. Found: C, 63.25; H, 6.18; N, 6.99.
Complex 4f. This was synthesized like 4d from bis(p-nitrophen-
yl) phosphoric acid hydrate (109 mg, 0.323 mmol), KOH (18.1
mg, 0.323 mmol), and 1 (218 mg, 0.323 mmol). Recrystallization
from dichloromethane/n-heptane by evaporation yielded 138 mg
(48%) of 4f as light yellow crystals, mp 200 °C (dec). IR (KBr;
cm-1): 1608 (s) (CdN), 1345 (vs) (NO2), 1111 (vs) (PO2). 1H NMR
(CDCl3): δ 6.89-8.01 [m, 18H, therein Ph(2,6) and Ph (3,5)]. 31
NMR: δ 13.44 in CDCl3, -12.61 in CD3CN.
P
Complex 6b. To a solution of 3 (135 mg, 0.179 mmol) in
methanol (20 mL) was added a solution of NaI (35.0 mg, 0.233
mmol) in methanol (10 mL), upon which a precipitate formed. After
the mixture was stirred overnight, the precipitate was filtered off,
washed with a few milliliters of ice-cold methanol, and dried in
vacuo. A yield of 100 mg (73%) of 6b remained as a colorless
powder, mp 285 °C (dec). IR (KBr; cm-1): 1605 (m) (pyridine
Anal. Calcd for C45H46N5O9PZn (Mr ) 897.25): C, 60.24; H,
5.17; N, 7.81. Found: C, 60.01; H, 5.39; N, 7.81.
Complex 5a. This was synthesized like 4a from 2 (200 g, 0.319
mmol) and KCl (26.0 mg, 0.35 mmol). Yield: 130 mg (75%) of
5a as a colorless powder. Mp: 250 °C (dec). IR (KBr; cm-1): 1604
1
1
(s) (CdN). H NMR (CDCl3): δ 1.13 (s, 9H, t-Bu), 1.37 (s, 9H,
coordinated), 1590 (m) (pyridine uncoordinated). H NMR: like
t-Bu), 2.87 (s, 6H, CH3), 3.52 (s, 2H, CH2-phenol), 4.06 (d, J )
15.1 Hz, 2H, CH2-py), 4.36 (d, J ) 15.1 Hz, 2H, CH2-py), 6.55
(d, J ) 2.8 Hz, 1H, HR), 6.68-7.06 (m, 5H, Ar), 7.50 (t, J ) 7.5
Hz, 2H, Hγ).
Anal. Calcd for C29H38ClN3OZn (Mr ) 545.48): C, 63.86; H,
7.02; N, 7.70. Found: C, 63.65; H, 7.16; N, 7.65.
Complex 5b. This was synthesized like 4a from 2 (200 mng,
0.319 mmol) and KI (80.0 mg, 0.482 mmol). Yield: 95 mg (47%)
of 5b as a colorless powder. Mp: 215 °C (dec). IR (KBr; cm-1):
1605 (vs) (CdN).
Anal. Calcd for C29H38IN3OZn (Mr ) 636.93): C, 54.49; H, 6.01;
N, 6.60. Found: C, 54.60; H, 6.09; N, 6.47.
that of 6a.
Anal. Calcd for C39H42IN3OZn (Mr ) 761.07): C, 61.55; H, 5.56;
N, 5.52. Found: C, 61.69; H, 5.73; N, 5.48.
Complex 6c. This was synthesized like 4d from p-nitrophenol
(27.8 mg, 0.200 mmol), KOH (11.2 mg, 0.280 mmol), and 3 (150
mg, 0.200 mmol). After filtration, the filtrate was evaporated to 10
mL and cooled to -20 °C. A yield of 31 mg (20%) of 6c
precipitated as yellow crystals, mp 125 °C (dec). IR (KBr; cm-1):
1
1585(s) (pyridine uncoordinated). H NMR (CDCl3): δ 6.55 [d,
J ) 8.9 Hz, 2H, Ph(2,6)], 7.0-8.0 [m, 14H, therein Ph(3,5)].
Anal. Calcd for C39H46N4O4Zn (Mr ) 772.27): C, 69.99; H, 6.00;
N, 7.25. Found: C, 69.31; H, 6.16; N, 7.13.
Complex 5d. This was synthesized like 4d from p-nitrophenol
(44.4 mg, 0.319 mmol), KOH (17.9 mg, 0.319 mmol), and 2 (200
mg, 0.319 mmol). The residue after evaporation to dryness was
suspended in diethyl ether (5 mL), stirred for 3 h, filtered, and
evaporated to dryness, leaving behind 80 mg (39%) of 5d as a
yellow powder, mp 160 °C (dec). IR (KBr; cm-1): 1605 (s) (Cd
Attachment of Pyridine Donors. Complex 7a. Pyridine (18.0
µL, 17.6 mg, 0.222 mmol) was added with vigorous stirring to a
solution of 1 (150 mg, 0.222 mmol) in methanol (35 mL). After
the solution was stirred overnight, the solvent was removed in
vacuo. The residue was dissolved in dichloromethane (5 mL) to
which was added n-heptane (5 mL). Slow evaporation from the
open flask yielded 120 mg (74%) of 7a as yellow crystals, mp 190
1
N), 1301 (vs) (NO2). H NMR (CDCl3): δ 6.54 [d, J ) 9.2 Hz,
2H, Ph(2,6)], 7.94 [d, J ) 9.2 Hz, 2H, Ph(3,5)].
1
°C (dec). IR (KBr; cm-1): 1606 (s) (CdN), 1100 (vs) (ClO4). H
Anal. Calcd for C35H42N4O4Zn‚H2O (Mr ) 648.13 + 18.02):
C, 63.11; H, 6.66; N, 8.41. Found: C, 63.74; H, 6.66; N, 8.45.
Complex 5e. This was synthesized like 4d from p-nitrobenzoic
acid (53.3 mg, 0.319 mmol), KOH (17.9 mg, 0.319 mmol), and 2
(200 mg, 0.319 mmol). Crystallization by slow evaporation from
dichloromethane/n-heptane yielded 78 mg (36%) of 5e as yellow
crystals, mp 150 °C (dec). IR (KBr; cm-1): 1625 (s) (COO), 1604
(s) (CdN), 1341 (vs) (NO2). 1H NMR (CDCl3): δ 7.59-8.15 (m,
4H).
NMR (CDCl3): δ 7.0-7.8 (m, 15H, therein pyridine-Hâ and -Hγ),
8.36 (m, 2H, pyridine-HR).
Anal. Calcd for C38H43ClN4O5Zn‚0.5CH2Cl2‚0.5H2O (Mr
)
736.62 + 9.01): C, 58.68; H, 5.76; N, 7.11. Found: C, 58.46; H,
5.86; N, 6.90.
Complex 7b. This was synthesized like 7a from 2-acetylpyridine
(35.9 mg, 0.296 mmol) and 1 (200 mg, 0.296 mmol). Yield: 122
mg (52%) of 7b as dark yellow crystals. Mp: 140 °C (dec). IR
(KBr; cm-1): 1690 (s) (CdO), 1602 (s) (CdN), 1095 (vs) (ClO4).
1H NMR (CDCl3): δ 2.54 (s, 3H, CH3CO), 6.95-7.72 (m, 13H,
therein 2py-H), 8.13 (m, 1H, Hâ), 8.35 (d, J ) 7.3 Hz, 1H, HR).
Anal. Calcd for C40H45ClN4O6Zn (Mr ) 778.66): C, 61.70; H,
5.82; N, 7.20. Found: C, 61.47; H, 6.04; N, 7.12.
Anal. Calcd for C36H42N4O5Zn (Mr ) 676.14): C, 63.95; H, 6.26;
N, 8.29. Found: C, 64.57; H, 6.77; N, 7.94.
Complex 5f. This was synthesized like 4d from bis(p-nitrophen-
yl) phosphoric acid hydrate (107 mg, 0.319 mmol), KOH (17.9
mg, 0.319 mmol), and 2 (200 mg, 0.319 mmol). Yield: 125 mg
(42%) of 5f as a yellowish powder. Mp: 190 °C (dec), which was
not analytically pure but showed only one 31P NMR signal. IR (KBr;
cm-1): 1605 (s) (CdN), 1344 (vs) (NO2), 1115 (vs) (PO2). 1H NMR
Complex 8a. This was synthesized like 7a from pyridine (25.8
µL, 25.2 mg, 0.319 mmol) and 2 (200 mg, 0.319 mmol). Yield:
36 mg (16%) of 8a as a yellow powder. Mp: 175 °C (dec). IR
(KBr; cm-1): 1608 (s) (CdN), 1097 (vs) (ClO4). 1H NMR
(CDCl3): δ 7.45 (m, 2H, Hâ), 7.82 (m, 1H, Hγ), 8.79 (m, 2H, HR).
Anal. Calcd for C34H43ClN4O5Zn (Mr ) 688.58): C, 59.31; H,
6.29; N, 8.14. Found: C, 58.76; H, 6.44; N, 7.96.
(CDCl3): δ 6.75-7.95 [m, 18H, therein Ph(2,6) and Ph(3,5)]. 31
NMR: δ 11.98 in CDCl3, -11.52 in CDCN.
P
Complex 6a. To a solution of 3 (100 mg, 0.133 mmol) in
methanol was added solid KBr (19.0 mg, 0.160 mmol). After a
few minutes a precipitate formed which was filtered off, suspended
in dichloromethane (10 mL), and filtered again. To the filtrate was
slowly added n-heptane to precipitate the product. A yield of 49
Complex 9a. This was synthesized like 7a from pyridine (9.2
µL, 8.9 mg, 0.11 mmol) and 3 (85 mg, 0.11 mmol). Yield: 48 mg
(54%) of 9a as yellow crystals. Mp: 250 °C (dec). IR (KBr; cm-1):
3328 Inorganic Chemistry, Vol. 44, No. 9, 2005

Zinc Complex Chemistry of N,N,O Ligands
Table 2. Crystallographic Data
1
2
3
4d
7a
7b
formula
C33H40ClN3O6Zn‚ C29H40ClN3O6Zn C39H42ClN3O5Zn‚ C39H42N4O4Zn‚CH3OH‚ C38H43ClN4O5Zn‚
CH2Cl2 H2O
760.6 751.6
C40H45ClN4O6Zn‚
2CH2Cl2
948.5
0.5CH2Cl2‚0.5H2O
779.7
0.5CH2Cl2‚0.5H2O
788.1
Mr
627.4
space group
Z
P1h
2
P1h
2
P21/c
4
P1h
4
C2/c
8
P21/n
4
a (Å)
b (Å)
c (Å)
11.848(5)
12.462(5)
14.445(6)
96.51(1)
108.21(1)
111.18(1)
1826(1)
1.38
10.030(2)
12.151(2)
13.196(2)
85.685(3)
81.476(3)
73.702(3)
1525.6(5)
1.37
21.658(4)
10.257(2)
17.680(3)
90
112.659(4)
90
3624.5(11)
1.37
0.80
11.08(2)
15.48(3)
24.18(3)
83.08(3)
83.92(3)
79.88(4)
4036(12)
1.27
21.204(4)
18.506(3)
19.580(4)
90
95.286(7)
90
7651(2)
1.37
0.83
13.355(6)
11.687(6)
29.348(14)
90
94.385(9)
90
4567(4)
1.38
0.88
R (deg)
â (deg)
γ (deg)
V (Å3)
d(calcd) (g cm-3
)
µ(Mo KR) (mm-1
R1 (obsd reflcns)
wR2 (all reflcns)
)
0.94
0.052
0.125
0.94
0.050
0.150
0.72
0.069
0.217
0.048
0.153
0.060
0.189
0.098
0.333
1610 (m) (CdN), 1096 (vs) (ClO4). 1H NMR (CDCl3): δ 6.70-
7.80 (m, 21H, therein Hâ and Hγ), 7.87 (t, J ) 7.8 Hz, 2H, HR).
Anal. Calcd for C44H47ClN4O5Zn (Mr ) 812.72): C, 65.03; H,
5.83; N, 6.89. Found: C, 64.11; H, 5.73; N, 6.72.
Complex 9b. This was synthesized like 7a from 2-acetylpyridine
(29.8 µl, 32.2 mg, 0.266 mmol) and 3 (200 mg, 0.266 mmol).
Crystallization from dichloromethane/cyclohexane by slow evapora-
tion yielded 152 mg (61%) of 9b as orange crystals, mp 215 °C.
IR (KBr; cm-1): 1675 (s) (CdO), 1599 (s) (CdN), 1091(vs) (ClO4).
1H NMR (CD3CN, 320 K): δ 2.45 (s, 3H, CH3CO), 6.90-8.60
(m, 22H, therein HR, Hâ, Hγ).
Anal. Calcd for C46H49ClN4O6Zn‚CH2Cl2 (Mr ) 854.76 +
84.93): C, 60.07; H, 5.47; N, 5.96. Found: C, 59.79; H, 5.48; N,
5.94.
In Situ Preparation of the Hydroxide Complexes. Complex
10. 1 (10.0 mg, 0.015 mmol) and triethylamine (2.11 µL, 1.52 mg,
0.015 mmol) were added to 500 µL of CD3CN, shaken, and used
as such for the hydrolytic reactions.
(5.78 mg, 0.013 mmol, in 25 mL of CH3CN) were prepared shortly
before the experiments. The solutions of 1 (or 2) and NEt3 were
mixed in the quartz cuvettes and supplemented to a total volume
of 2.85 mL with CH3CN. Immediately prior to the measurements
150 µL of the TNP solution was added. Thus, the concentration of
TNP was 0.025 mM for all measurements.
TNP Cleavage with 10. Reagents were mixed such that there
was a 60-, 120-, 180-, 240-, and 300-fold excesses of 10. Absorption
intensities were taken for at least 5t1/2, and their final values were
taken as I∞. The measurements could be reproduced by a first-
order rate law with a correlation coefficient greater than 0.999. For
the computations the measurements up to 2t1/2 were included. The
reproducibility of the I values was within 10%.
TNP Cleavage with 11. The same procedure as for 10 was used,
again with 60-, 120-, 180-, 240-, and 300-fold excesses of 11.
Correlation coefficients throughout were greater than 0.999.
Structure Determinations. Crystals of 1-3 and 7a,b were
obtained by layering CH2Cl2 solutions with n-heptane, and those
of 4d, by slow evaporation of a CH2Cl2/MeOH solution. Diffraction
data were recorded with a Bruker AXS Smart CCD diffractometer
at -50 °C with Mo KR radiation and subjected to empirical
absorption corrections.42 The structures were solved with direct
meethods and refined anisotropically.42 Hydrogen atoms were
included with fixed distances and isotropic temperatures factors
1.5 times those of their attached atoms. Parameters were refined
against F2. Drawings were produced with SCHAKAL.43 Crystal-
lographic data are listed in Table 2.
1H NMR (CD3CN, all signals broad): δ 1.03 [t, J ) 7.0 Hz,
9H, CH3 (NEt3)], 1.18 (s, 9H, t-Bu), 1.36 (s, 9H, t-Bu), 2.61 [q,
J ) 7.0 Hz, 6H, CH2 (NEt3)], 3.50 (s, 2H, CH2-phenol), 4.08 (s,
4H, CH2-py), 6.5-8.5 (m, 14H, Ar).
Complex 11. 2 (10.0 mg, 0.016 mmol) and triethylamine (2.22
µL, 1.60 mg, 0.016 mmol) were added to 500 µL of CD3CN,
shaken, and used as such for the hydrolytic reactions.
1H NMR (CD3CN, all signals broad): δ 1.18 (s, 9H, t-Bu), 1.22
[t, J ) 7.3 Hz, 9H, CH3 (NEt3)], 1.31 (s, 9H, t-Bu), 2.67 [s, 6H,
CH3 (py)], 3.02 [q, J ) 7.3 Hz, 6H, CH2 (NEt3)], 3.65 [s, 2H,
CH2-phenol], 4.18 [s, 4H, CH2-py], 6.78 (s, 1H, HR), 6.96 (s, 1H,
Hγ), 7.03-7.40 (m, 4H, py), 7.72 (t, J ) 7.6 Hz, 2H, Hγ).
Hydrolysis of Tris(p-nitrophenyl) Phosphate (TNP). With 10.
A CD3CN solution of 10 (see above) was treated with the equimolar
amount of TNP (6.7 mg, 0.015 mmol) in a NMR tube and the
reaction followed by 31P NMR. Within 2 h the resonance of TNP
at -18.7 disappeared and was replaced by the resonance of 4f at
-12.6 ppm.
Acknowledgment. This work was supported by the Fonds
der Chemischen Industrie. We are indebted to Profs. C. Ro¨hr
and B. Kersting for their help with crystallographic problems
and Drs. H. Brombacher and C. Perez Olmo for measure-
ments.
Supporting Information Available: Fully labeled ORTEP plots
for all six structure determinations, two plots of the rate constants
for the TNP cleavages, and six crystallographic files, in CIF format.
This material is available free of charge via the Internet at http://
pubs.acs.org.
With 11. Using 11 and TNP (7.1 mg, 0.016 mmol) the resonance
of TNP was replaced by that of 5f at -11.5 ppm.
Kinetic Study. General Methods. Measurements were per-
formed on a JASCO V-570 UV-vis spectrometer by continually
recording the 392 nm absorption of 4d (the 380 nm absorption of
5d). The measuring chamber and the solutions were thermostated
to 25 °C for 30 min prior to the measurements and during the kinetic
runs. Stock solutions of 1 (203 mg, 0.300 mmol, in 10.0 mL of
CH3CN), 2 (194 mg, 0.300 mmol, in 10.0 mL of CH3CN), NEt3
(84.3 µL, 60.7 mg, 0.300 mmol, in 20 mL of CH3CN), and TNP
IC050042U
(42) Sheldrick, G. M. Bruker AXS, SHELXTL Program Package, version
5.1; Universita¨t Go¨ttingen: Go¨ttingen, Germany, 1998.
(43) Keller, E. Schakal for Windows; Universita¨t Freiburg: Freiburg,
Germany, 2001.
Inorganic Chemistry, Vol. 44, No. 9, 2005 3329