Coordination chemistry of unsymmetrical tripodal ligands with an NNO2 donor set

Article abstract of DOI:10.1002/1099-0682(200109)2001:9<2427::AID-EJIC2427>3.0.CO;2-7

The synthesis of two new tripodal ligands N(CH₂CH₂NH₂)(CH₂CH₂CH ₂OH)₂ (H₄-1) and N[2,3,5-C₆H₂(OH)(SCH₃)(CH₃)](CH ₂

Full text of DOI:10.1002/1099-0682(200109)2001:9<2427::AID-EJIC2427>3.0.CO;2-7

FULL PAPER
Coordination Chemistry of Unsymmetrical Tripodal Ligands with an NNO2
Donor Set
Christian Ochs,[a] F. Ekkehardt Hahn,*[b] and Roland Fröhlich[c]
Keywords: Tripodal ligands / Copper / Radicals / N ligands / O ligands
The
synthesis
of
two
new
tripodal
ligands
molecular structures of 3 and 4·2CH3CN have been deter-
mined by X-ray diffraction. Complex 3 contains two slightly
distorted square-pyramidal (τ = 0.181) copper atoms, with O-
N(CH2CH2NH2)(CH2CH2CH2OH)2 (H4-1) and N[2,3,5-C6H2-
(OH)(SCH3)(CH3)](CH2CH2CH2NH2)(CH2CH2CH2OH) (H4-
2) is reported. Both tetradentate ligands contain a central ter- donors in the apical positions. The dinuclear complex 4,
tiary nitrogen atom, as well as two OH and one NH2 func- which was synthesized to model the copper site in galactose
tionalized ligand arm. The tripods do not only exhibit an un- oxidase, also shows a distorted square-pyramidal coordina-
symmetrical N2O2 donor set, but also possess two C3 and one tion geometry (τ = 0.205 and τ = 0.101) around both copper(II)
C2 chains between the central nitrogen atom and the ter- atoms. Complex 4 contains two uncoordinated primary alco-
minal donors. On coordination of the central tertiary nitrogen
atom, the ligands are capable of forming both six- and five-
membered chelate rings in their metal complexes. Both li-
hol functionalities of the ligands. In addition, both the ligand
H4-2 and its dinuclear copper complex [(η3-H3-2)Cu(µ-
OAc)2(η3-H3-2)]·2CH3CN can easily be oxidized to yield free
gands react with [Cu(OAc)2·H2O]2 to give the dinuclear cop- or coordinated phenoxyl radicals, which are stable on the
per(II) complexes [Cu(η4-µ-O-H3-1)2Cu](PF6)2 (3) and [(η3-
H3-2)Cu(µ-OAc)2Cu(η3-H3-2)]·2CH3CN (4·2CH3CN). The
time scale of cyclic voltammetry.
Introduction
position is occupied by the oxygen atom of another tyrosin-
ate ligand (Tyr-495). A unique characteristic of the copper
site in galactose oxidase is that the equatorial tyrosinate
residue Tyr-272 is covalently linked to a sulfur atom of the
neighboring cysteine Cys-228. This leads to a lowering of the
oxidation potential and significantly facilitates the formation
of a tyrosyl radical from Tyr-272.[3] Furthermore, in the oxid-
ized form of GO, antiferromagnetic coupling of the single
Redox reactions involving a transition metal ion and a
redox-active amino acid side chain, e.g. tyrosine, play a key
role in many biochemical processes.[1] Some years ago, Ito
et al.[2] have shown that the reactive site in galactose oxidase
(E.C. 1.1.3.9) GO, a copper enzyme that was isolated from
the parasitic fungus Fusarium dendroides, contains a tyrosyl
radical which is coordinated to a copper(II) center. Galac-
tose oxidase catalyzes the stereospecific oxidation of -gal-
actose and various primary alcohols to the corresponding
aldehydes, coupled to the simultaneous reduction of O2 to
H2O2 [Equation (1)].
2
2
electron in the copper(II) dx Ϫy orbital with the tyrosyl
radical is observed, giving a diamagnetic and EPR silent
(S ϭ 0) ground state of oxidized, active galactose oxidase.[4]
RCH2OH ϩ O2 Ǟ RCHϭO ϩ H2O2
(1)
The X-ray crystal structure analysis of the inactive form
˚
of galactose oxidase at a 1.7 A resolution showed a mono-
nuclear copper(II) center that is surrounded by a square-
pyramidal N2O3 coordination environment.[2] As depicted
in Figure 1, two histidine residues (His-496, His-581), one
tyrosinate (Tyr-272) and one water or acetate molecule
reside in the equatorial plane, while the remaining apical
Figure 1. The active site in galactose oxidase
[a]
Institut für Anorganische und Analytische Chemie, Freie
Universität Berlin,
Several attempts have been made in the last years to
mimic the geometric, spectroscopic and catalytic properties
of native galactose oxidase. Functional model complexes
have been presented by Tolman[5] and by Stack and co-
workers.[6] Recently, Wieghardt et al.[7] described a func-
Fabeckstraße 34Ϫ36, 14195 Berlin, Germany
[b]
Anorganisch-Chemisches Institut, Westfälische Wilhelms-
Universität Münster,
Wilhelm-Klemm-Straße 8, 48149 Münster, Germany
[c]
Organisch-Chemisches Institut, Westfälische Wilhelms-
Universität Münster,
Corrensstraße 40, 48149 Münster, Germany
tional model of galactose oxidase derived from a
Eur. J. Inorg. Chem. 2001, 2427Ϫ2436
WILEY-VCH Verlag GmbH, 69451 Weinheim, 2001
1434Ϫ1948/01/0909Ϫ2427 $ 17.50ϩ.50/0
2427

C. Ochs, F. E. Hahn, R. Fröhlich
FULL PAPER
CuIIϪiminosemiquinone
with
a
square-planar
[Cu(OϪNϪO)(NEt3)] (OϪNϪO ϭ dianionic ligand with
one imine and two two phenolate donors) reaction center,
which showed a high catalytic activity despite the lack of
most structural features found at the active center of native
galactose oxidase.
Various CuII complexes have been prepared in order to
model the structural features of the reactive center in galac-
tose oxidase. Mostly tripodal tetradentate ligands that con-
tain one or two phenol and pyrazole or pyridine donor
groups were used, providing an N3O[3a,8,9] or N2O2[10,11] co-
ordination environment for the copper center. Pentacoord-
ination is normally achieved by coordination of a coun-
terion or a solvent molecule. Unfortunately, all of the model
compounds lack significant features found in the reactive
site of native galactose oxidase. These features include a
CuII atom coordinated in a square-pyramidal fashion which
exhibits a thioether-functionalized phenoxyl donor in the
equatorial plane and another O-donor in the apical posi-
tion. Two more N-donors and one labile (possibly solvent
molecule) O-donor in the equatorial plane should complete
the overall N2O3 coordination environment.
If a GO model is to be built from a tetradentate tripodal
ligand, certain restrictions apply. Such ligands, together
with a monodentate coligand, form trigonal-bipyramidal or
square-pyramidal complexes with CuII. For example
[Cu(tren)(NH3)]2ϩ in the solid state is an almost perfect tri-
gonal bipyramid.[12] A systematic study with higher homo-
logues of tren, e.g. with ligands where methylene groups
were added stepwise to the three ligand arms, revealed that
the tripod with one ethyl and two propyl ligand arms
N(CH2CH2CH2NH2)2(CH2CH2NH2) forms a square-pyr-
amidal CuII complex where the apical position is occupied
by a donor linked to the central nitrogen atom by a propyl
chain.[13] Thus, the topology of the tetradentate tripod de-
termines both the overall geometry of the CuII complex
(sqp vs. tbp), as well as the position of the donors in the
complex (donor group at the end of a propyl chain in the
apical position). Studies with unsymmetrical tetradentate
tripods providing an N3O donor set for CuII [8,9] gave sim-
ilar results. Again, it was found that the formation of
square-pyramidal CuII complexes is preferred when the li-
gand is capable of forming a 5,6,6-membered chelate ring
sequence.[8]
Figure 2. Topology of the tripodal N2O2 ligands H4-1 and H4-2
The second ligand H4-2 (Figure 2), providing an N2O2
donor set, was designed for a more accurate structural
modeling of the reactive site in GO. This ligand is expected
to form a square-pyramidal CuII complex with a 5,6,6-
membered chelate ring sequence. The phenolate oxygen
donor, linked by a C2 chain to the central nitrogen atom,
is expected to occupy an equatorial coordination site. In
addition, the aromatic ring is functionalized with an o-
(methylthio) group which in native GO is essential for the
formation of a tyrosyl radical on oxidation. The mode of
coordination of the two donors linked by C3 chains to the
central nitrogen atom can be elucidated by X-ray crystallo-
graphy. One of them is expected to occupy an apical coor-
dination site in a CuII complex. Overall, ligand H2-2 pos-
sesses an N2O2 donor set. It differs from N2O2 ligands pre-
viously reported[10,11] by possessing aliphatic amine and al-
cohol donor groups in the presence of an o-(methylthio)-
functionalized phenol.
Results and Discussion
Ligand Syntheses
Our strategy for the preparation of the unsymmetrical
N2O2 ligand H4-1 involves three reaction steps as depicted
in Scheme 1. For the condensation of the C3 chains to
These observations can be used to design ligands for the
preparation of GO model compounds. Any tetradentate tri-
pod with a central nitrogen atom capable of forming a
5,6,6-membered chelate ring sequence should give a more
or less perfectly coordinated square-pyramidal CuII com-
plex with a donor at the end of a propyl chain in the ap-
ical position.
The aliphatic tripodal ligand H4-1 (Figure 2) was synthe-
sized to investigate the extent to which the above mentioned
results for N4 and N3O ligands (control of complex geo-
metry vs. ligand topology) are transferable to a N2O2 donor
set. Ligand H4-1 should form a square-pyramidal CuII
complex with one of the oxygen donors in the apical and
the other in an equatorial position.
Scheme 1. Preparation of ligand H4-1
Eur. J. Inorg. Chem. 2001, 2427Ϫ2436
2428

Coordination Chemistry of Unsymmetrical Tripodal Ligands with an NNO2 Donor Set
FULL PAPER
ammonia, the Michael addition of ethyl acrylate had previ-
ously proved useful.[14] However, this reaction leads to a
mixture of the tertiary (A) and secondary amines (B), which
can be separated by distillation.[14a] Cyanomethylation of
the secondary amine B by the Strecker synthesis according
to the modification of Knoevenagel,[15] yielded the α-nitril-
amine C. Finally, all protected donor groups were simultan-
eously liberated by reduction with AlH3, which led directly
to the free ligand H4-1.
Ligand H4-1 contains one NH2- and two OH-substituted
ligand arms in addition to the central tertiary nitrogen
atom. Furthermore, the ligand possesses an unsymmetrical
topology with regard to the three alkyl chains, containing
one C2 and two C3 arms. This allows for the formation of
both five- and six-membered chelate rings yielding a 5,6,6
ring sequence on complex formation. Due to the flexibility
and lengths of the alkyl arms, the ligand should be capable
of providing an N2O2 coordination environment for a cop-
per(II) ion without any steric strain.
Ligand H4-2 (Figure 2) was developed on the basis of
simple force field calculations and molecular modeling ex-
periments. It was assumed that a tripod, which could form a
5,6,6-chelate ring sequence, would form a square-pyramidal
CuII complex with one of the C3-linked donors in the apical
position. Linking the phenol group directly to the central
nitrogen atom should force the oxygen atom of this group
in the equatorial position of the expected square-pyramidal
complex. Taking into account the significance of the equa-
torial (RS)-substituted Tyr-272 for galactose oxidase activ-
Scheme 2. Preparation of ligand H4-2
ity, an o-(methylthio)phenol group was selected. In addi-
tion, the para position relative to the phenolic OH group
was blocked by a methyl group in order to prevent dimeriz- cies that was not isolated.[18] This ortho-lithiated interme-
ation when a radical intermediate is formed after one-elec- diate was quenched with dimethyl disulfide.[19] The o-
tron oxidation. A similar N2O2 ligand capable of forming a (methylthio)-substituted compound F was isolated in good
5,6,6-chelate ring sequence and possessing an o-(methyl- yield. According to a literature procedure,[20] the nitro
thio)phenol group was presented by Whittaker.[10] However, group in F can be reduced quantitatively to the amine by
this ligand like all other tripods providing an N2O2 donor an excess of hydrazine in the presence of Raney nickel to
set and a 5,6,6-chelate ring sequence,[11] contains a phenol yield the aniline derivative G. Subsequently, a β-cyanoethyl
group connected to the central nitrogen atom in a way that group which will serve as the synthon for the 3-aminopro-
the phenolic oxygen atom will be part of a six-membered pyl arm, was introduced by a copper(II) acetate catalyzed
chelate ring.
Michael addition of acrylonitrile to G. This resulted in the
For the synthesis of H4-2, o-nitro-p-cresol was selected as formation of compound H.[21] Finally, a C3 chain (the pre-
a suitable starting material. This molecule contains, at least cursor for the 3-hydroxypropyl arm) was added to the cent-
in the protected form, all three of the desired functional ral nitrogen atom by acylation of H with β-acetoxypro-
groups for the construction of the aromatic part of H4-2. pionyl chloride, which was synthesized as described.[22]
The complete reaction sequence for the preparation of H4- After chromatographic purification, amide I was obtained
2 is depicted in Scheme 2.
as a crystalline white solid. Reduction of I with AlH3
In a first reaction, the aromatic core of o-nitro-p-cresol yielded the ligand H4-2. Under these conditions even the
was brominated by treatment with bromine in the presence MOM-protected phenolic OH group is liberated. This is
of the Lewis acid catalyst AlCl3.[16] The position ortho to quite surprising since MOM ethers are known to be stable
the OH group is exclusively brominated, yielding com- in basic media and are normally cleaved under acidic condi-
pound D, due to the ortho- and meta-directing effect of the tions. Ligand H4-2 is a sticky white solid which is slightly
OH and the NO2 substituents, respectively. Subsequently, D air-sensitive.
was deprotonated with NaOH, and the resulting phenolate
Ligand H4-2, like H4-1, possesses an N2O2 donor set and
was treated with (chloromethoxy)methane to give the an unsymmetrical topology. It is capable of forming a 5,6,6-
MOM-protected derivative E.[17] Halogen/metal exchange membered chelate ring sequence and should be suitable for
of the bromine substituent, by use of phenyllithium at Ϫ105 the preparation of square-pyramidal CuII complexes. H4-2
°C resulted in the formation of a reactive aryllithium spe- contains an o-(methylthio)-substituted phenol which is
Eur. J. Inorg. Chem. 2001, 2427Ϫ2436
2429

C. Ochs, F. E. Hahn, R. Fröhlich
FULL PAPER
directly connected to the central nitrogen atom of the ligand
backbone. This is in contrast with tripodal N2O2 ligands
published to date with regard to the modeling of the copper
site in galactose oxidase. The electron-donating o-(methyl-
thio) and the p-methyl substituents and the direct linkage
of the aromatic ring to the central nitrogen atom should
significantly lower the redox potential for the oxidation of
this phenol group, while the substituents in the ortho and
para positions relative to the OH group will prevent the
formed phenoxyl radical from dimerizing.
Preparation of Copper Complexes
Freshly prepared Cu(OH)2, suspended in water, reacted
at 40 °C with a stoichiometric amount of ligand H4-1 to
give a blue suspension which contained unidentified copper
species (probably hydroxides). Addition of 2 equiv. of
NH4PF6 to such a suspension yielded a deep blue solution.
From this solution complex 3 crystallized over a period of
6 h (Scheme 3). The complex is air-stable and can be recrys-
tallized from water as deep blue prisms. The reaction of
[HNEt3]ϩ[H3-2]Ϫ with an equimolar amount of
[Cu(OAc)2·H2O]2 in methanol at room temperature yielded
a dark green solution. On addition of a mixture of
acetonitrile and tert-butyl methyl ether, and on cooling to
Ϫ30 °C, air-sensitive blue plates crystallized, which were
identified as 4·2CH3CN by X-ray crystallography
(Scheme 3).
Molecular Structures of 3
The X-ray crystal structure analysis shows the formation
of the dinuclear complex [Cu(η4-µ-O-H3-1)2Cu](PF6)2 (3)
from Cu(OH)2 and the ligand H4-1. The dication of com-
plex 3 resides on a crystallographic inversion center in the
unit cell. Each copper(II) ion is coordinated by two nitro-
gen atoms, one terminal propyloxy group (O1) and two µ2-
propyloxo groups (O2) (Figure 3). The bridging oxygen
atom O2 is deprotonated while the monodentate oxygen
atom O1 remains protonated in the complex. The charge
distribution was assigned based on the presence of only two
Scheme 3. Preparation of complexes 3 and 4
Ϫ
PF6 anions which make the complex a dication with two
monodeprotonated ligands H3-1Ϫ. The most acidic func-
tions of the ligand are the alcohol groups. We concluded
that O2 is the deprotonated alcohol group, based on the
˚
difference in the CuϪO distances [CuϪO1 2.339(3) A,
˚
CuϪO2 1.920(3) A].
The coordination geometry around the copper centers is
best described by the use of the τ-criterion,[23] which indic-
ates that the coordination geometry in 3 is only slightly dis-
torted (τ ϭ 0.181) from square-pyramidal. Atom O1 occu-
pies the apical position. The Cu2O2 core shows a significant
rhombic distortion. The angle O2ϪCuϪO2* is 78.10(13)°,
Figure 3. ORTEP plot of one complex dication of 3 (hexafluoro-
˚
phosphate anions are not shown); selected bond lengths [A] and
angles [°]: Cu···Cu 3.001(1), CuϪO1 2.339(3), CuϪO2 1.920(3),
CuϪO2* 1.944(3), CuϪN1 2.033(3), CuϪN2 1.992(4);
O1ϪCuϪO2 90.91(14), O1ϪCuϪO2* 91.77(13), O1ϪCuϪN1
94.74(14), O1ϪCuϪN2 106.3(2), O2ϪCuϪO2* 78.10(13),
O2ϪCuϪN1 98.66(13), O2ϪCuϪN2 161.9(2), O2*ϪCuϪN1
172.79(14), O2*ϪCuϪN2 95.43(15), N1ϪCuϪN2 85.7(2),
while for CuϪO2ϪCu* an angle of 101.90(13)° is observed.
˚
The Cu and Cu* distance is 3.001(1) A. The CuϪN and CuϪO2ϪCu* 101.90(13)
CuϪOeq bond lengths are only slightly different [CuϪN1
2.033(3), CuϪN2 1.992(4) and CuϪO2 1.920(3), CuϪO2*
1.944(3) A], leading to an almost symmetrical bridge be- of 3. This should and does lead to a square-pyramidal coor-
A 5,6,6-chelate ring sequence is observed in the dication
˚
tween the copper atoms.
dination geometry, with a donor atom of a six-membered
2430
Eur. J. Inorg. Chem. 2001, 2427Ϫ2436

Coordination Chemistry of Unsymmetrical Tripodal Ligands with an NNO2 Donor Set
FULL PAPER
chelate ring (O1) in the apical position. In this regard, the
In complex 4 both copper(II) ions reside in a square-pyr-
molecular structure of 3 meets our expectations and experi- amidal (τ ϭ 0.205 and τ ϭ 0.101) N2O3 coordination envir-
ences gained from related aliphatic tripodal N4 ligand sys- onment, but in contrast to the situation in 3, the µ2-oxy
tems.[13] However, complex 3 cannot serve as a GO model bridges in 4 result from the bridging coordination of acetate
compound since GO contains a mononuclear square-pyr- oxygen atoms. The two acetate oxygen atoms occupy equat-
amidal copper center and not a binuclear one.
orial (O3) and apical (O3*) positions at the copper atom.
To the best of our knowledge 3 is one of two examples Each CuII ion is coordinated by two nitrogen and three oxy-
of a clearly square-pyramidal copper complex with an ap- gen donors (one phenolate and two acetate oxygen atoms),
ical O-donor derived from a tripodal N2O2 ligand. Previ- while the hydroxypropyl arm (O2) of the ligand remains
ously described tripodal N2O2 ligands with pyridyl and uncoordinated. Like in 3, both copper atoms are coordin-
phenolate donors gave dinuclear copper complexes which ated in a square-pyramidal fashion, but again a dinuclear
deviate significantly from the square-pyramidal geometry, complex was obtained and hence it cannot act as an accept-
towards a trigonal-bipyramidal geometry.[10][11b] However, able GO model compound.
one tripodal N2O2 ligand with bulky substituents at its pyri-
The apical CuϪO bond is not exactly perpendicular to
dyl and phenolate donors[11c] forms a distorted square-pyr- the equatorial plane which also leads to a rhombic distor-
amidal mononuclear copper complex (τ ϭ 0.273) with a tion of the Cu2O2 core [O3ϪCuϪO3* 83.47(12)° and
5,6,6-chelate ring sequence and with an oxygen donor at 89.36(12)°, CuϪO3ϪCu* 96.53(12)° and 90.64(12)°]. For
the end of a C3 chain in the apical position. These results the same reason all O3*ϪCuϪLeq angles significantly devi-
together with the molecular structure of 3 indicate again, ate from the ideal value of 90° expected for a perfect tetra-
that the coordination geometry of copper complexes with gonal pyramid.
tripodal ligands can at least partly be controlled and manip-
ulated by the ligand topology.
The CuϪN bond lengths are almost equal, as found in 3
˚
˚
˚
[Cu1ϪN1 2.056(3) A and 2.056(4) A, Cu1ϪN2 1.980(4) A
˚
and 1.970(3) A]. The CuϪOeq distances [CuϪO1 1.916(3)
˚
˚
˚
˚
A and 1.909(3) A, CuϪO3 1.999(3) A and 1.993(3) A] are
similar. However, the separation CuϪOapical is much more
Molecular Structures of 4·2CH3CN
˚
˚
pronounced [CuϪO3* 2.374(3) A and 2.529(3) A], a be-
havior observed in all square-pyramidal copper complexes
with an O-donor in the apical position.
The X-ray structure analysis of 4·2CH3CN revealed that
the oxygen-bridged, inversion-symmetric, dinuclear com-
plex [(η3-H3-2)Cu(µ-OAc)2Cu(η3-H3-2)]·2CH3CN had
formed (Scheme 3, Figure 4). The asymmetric unit contains
two halves of the complex, which are not related by crystal-
lographic symmetry.
Due to reasons we do not exactly know yet, and in con-
trast with the coordinated ligand (H3-1)Ϫ in 3, each (H3-
2)Ϫ ligand anion in complex 4 uses only three of the four
donor groups for binding to the copper atom, while the
hydroxypropyl arm is not coordinated and remains pro-
tonated. We believe that this is not exclusively due to the
steric situation in 4, but also to the low stability of the six-
membered chelate ring which would have formed with the
hydroxypropyl ligand arm. A similar behavior of related ali-
phatic tripodal ligands has been reported recently.[24] In ad-
dition it has been shown that the stability of the copper
complex with the ligand trpn, which exclusively forms six-
membered chelate rings is about 105 times lower than the
stability of the corresponding tren complex that contains
only five-membered chelate rings.[13b] The acetate oxygen
atoms are apparently better donors than the hydroxypropyl
group when competing for the coordination site at the cop-
per center. Similar behavior has been reported for copper
complexes with a tripodal ligand capable of forming a
5,6,6-chelate ring sequence, where the presence of chloride
ions prevents the formation of the chelate complex, and a
Figure 4. SCHAKAL plot of one molecule of complex 4 (the asym-
metric unit contains two nearly identical halves of a molecule); se- copper complex with chloride ligands was obtained.[11c]
˚
lected bond lengths [A] and angles [°] for molecule A [molecule
Table 1 lists some structural parameters for 3 and 4.
B]: Cu1···Cu1 3.2729(11) [3.2375(12)], Cu1ϪO1 1.916(3) [1.909(3)],
Cu1ϪO3 1.999(3) [1.993(3)], Cu1ϪO3* 2.374(3) [2.529(3)],
Cyclic Voltammetry
Cu1ϪN1 2.056(3) [2.056(4)], Cu1ϪN2 1.980(4) [1.970(3)];
O1ϪCu1ϪO3 92.15(12) [92.21(12)], O1ϪCu1ϪO3* 88.18(12)
[85.33(12)], O1ϪCu1ϪN1 86.09(13) [86.21(14)], O1ϪCu1ϪN2
171.36(14) [169.21(14)], O3ϪCu1ϪO3* 83.47(12) [89.36(12)],
O3ϪCu1ϪN1 159.04(14) [163.13(14)], O3ϪCu1ϪN2 88.95(13)
[88.74(14)], O3*ϪCu1ϪN1 117.30(12) [107.23(12)], O3*ϪCu1ϪN2
83.42(13) [83.93(13)], N1ϪCu1ϪN2 95.91(15) [95.92(15)],
Cu1ϪO3ϪCu1* 96.53(12) [90.64(12)]
The cyclic voltammogram of complex 3 recorded in an
acetonitrile solution exhibits an irreversible oxidation po-
tential at Epa ϭ ϩ1375 mV vs. Ag/AgCl. Complex 3 is quite
stable relative to to the corresponding CuIII species which
is highly reactive and rapidly decomposes to non-identified
Eur. J. Inorg. Chem. 2001, 2427Ϫ2436
2431

C. Ochs, F. E. Hahn, R. Fröhlich
FULL PAPER
Table 1. Comparison of selected structural details in complexes 3 and 4·2CH3CN
˚
˚
˚
CuϪOap [A]
Complex
CuϪNeq [A]
CuϪOeq [A]
τ
3
1.992(4), 2.033(3)
1.980(4), 2.056(3)
1.920(3), 1.944(3)
1.916(3), 1.999(3)
2.339(3)
2.374(3)
0.179
0.205, 0.101
4·2CH3CN
products. Upon application of a negative potential an irre- have also been shown to be essential for enzymes other than
versible reduction is observed at Epc ϭ Ϫ725 mV vs. Ag/ galactose oxidase, e.g. glyoxal oxidase[25] and ribonucleo-
AgCl in acetonitrile. This corresponds to the process CuII tide reductase[1b]
.
ϩ eϪ Ǟ CuI, which is followed by the dissociation of the
CuI complex and disproportionation of CuI to CuII and plex 4 takes place at a somewhat higher potential than the
Cu0.
protonated ligand H4-2, 4 is also easily oxidized. A broad
Although the oxidation of the blue dimeric copper com-
The results of the cyclic voltammetric analysis of the peak, which seems to be the result of the overlapping of
redox potentials of ligand H4-2 and complex 4 are shown two independent redox processes, is observed between 0 and
in Figure 5. In the cathodic region, the free ligand H4-2 ϩ1000 mV vs. Ag/AgCl in acetonitrile (Figure 5, bottom).
displays one quasireversible oxidation peak at E ϭ ϩ130 The oxidation of the phenolate takes place at a potential of
mV vs. Ag/AgCl in acetonitrile (top, solid line).
E1 ϭ ϩ430 mV vs. Ag/AgCl, while a second oxidation pro-
cess occurs at E2 ϭ ϩ870 mV vs. Ag/AgCl. The superim-
posing cyclic voltammetric responses suggest a quasirevers-
ible electron transition, which indicates that in 4 a stable
phenoxyl radical is formed on oxidation. In addition, the
ligand stabilizes the copper(II) state, since reduction of the
complex is observed at a very low potential (E ϭ Ϫ1320
mV vs. Ag/AgCl in acetonitrile; not shown in Figure 5).
This reduction process is irreversible and leads to a deposit
of copper(0) on the electrode surface, as concluded from a
sharp anodic oxidation peak at Epc ϭ Ϫ60 mV during the
reverse scan with the typical features of a redissolution pro-
cess.[10]
Conclusion
We have prepared two novel nonsymmetric tripodal li-
gands which provide an N2O2 donor set. Both ligands are
capable of coordinating to CuII to give a 5,6,6-chelate ring
sequence. For this chelate ring sequence, square-pyramidal
complexes with a donor linked by a C3 chain in the apical
position are expected. Ligand H4-1 indeed forms such a
complex. However, this complex is dinuclear. With ligand
H2-2, we present a tripod with an o-amino-o-(methylthio)-
substituted phenolate donor. This ligand also forms a
square-pyramidal CuII complex, which is also dinuclear.
While the dinuclear nature of complexes 3 and 4·2CH3CN
Figure 5. Cyclic voltammograms of H4-2 (top, solid line), (H3-
2)Ϫ (top, dashed line) and of complex 4 (bottom); E vs. Ag/AgCl
in acetonitrile
This process corresponds to the oxidation of the phenolic excludes them from the acceptable mononuclear GO
moiety, leading to a phenoxyl radical which is prevented models, they still provide information about the ligand de-
from dimerizing. The same process is observed when the sign for the preparation of structural GO models. Among
phenolic OH group is deprotonated in the presence of the dinuclear complexes with an N2O2 donor set, 3 and
(nBu)4NϩOHϪ. The oxidation potential of the phenol 4·2CH3CN contain copper atoms which deviate least from
group is then shifted towards a more negative potential by the required square-pyramidal coordination geometry, as
about 440 mV than reported previously,[10,11] and is now judged by the τ values of 0.181 for 3 and 0.205 and 0.101
observed as a quasireversible voltammetric response at E ϭ for 4·2CH3CN. Thus, the 5,6,6-chelate ring sequence indeed
Ϫ310 mV vs. Ag/AgCl in acetonitrile (top, dashed line). leads to the desired square-pyramidal complexes. Finally,
This low-potential redox step arises from the relatively fa- the unique o-amino-o-(methylthio)-substitution of the
cile oxidation of the electron-rich thioether and amino-sub- phenol donor in ligand H4-2 leads to a significant reduction
stituted phenol. Recently, such stable, free phenoxyl radicals of the redox potential for the oxidation of the free or coor-
2432
Eur. J. Inorg. Chem. 2001, 2427Ϫ2436

Coordination Chemistry of Unsymmetrical Tripodal Ligands with an NNO2 Donor Set
FULL PAPER
Compound C (5.07 g, 19.8 mmol), dissolved in 30 mL of THF, was
added dropwise to the vigorously stirred AlH3 suspension, at room
temperature. The reaction mixture was stirred for 24 h at ambient
temperature, and residual AlH3 was carefully hydrolyzed with water
(8.2 mL, 0.46 mol). All solids were removed by filtration and re-
peatedly washed with methanol (6 ϫ 50 mL). The organic layers
were combined, all solvents removed in vacuo, and the residue was
distilled at 135 °C (2·10Ϫ3 mbar). Ligand H4-1 was obtained as a
colorless oil. Yield 2.234 g (64% relative to C). Ϫ 1H NMR
(250 MHz, CDCl3): δ ϭ 3.67 (s ϩ t, 6 H, OH ϩ CH2O), 2.80 (t, 2
H, NCH2CH2NH2), 2.58 (t, 4 H, NCH2CH2CH2), 2.50 (t, 2 H,
CH2NH2), 1.70 (m ϩ s, 6 H, CH2CH2CH2 ϩ NH2). Ϫ 13C{1H}
dinated phenolate group to the corresponding phenoxyl
radical, which is stabilized against dimerization by p-methyl
substitution of the aromatic ring. Based on these observa-
tions we propose that substitution of the aliphatic OH and
NH2 donors in H4-2 by aromatic ones (phenol and pyri-
dine) and sterically more bulky substituents at the phenol
groups could lead to a square-planar, mononuclear CuII
complex, which would serve as a satisfactory structural
GO model.
NMR (62.90 MHz, CDCl3):
δ
ϭ
61.27 (CH2O), 56.45
(NCH2CH2NH2), 52.27 (NCH2CH2CH2), 39.29 (CH2NH2), 28.97
(CH2CH2CH2). Ϫ MS (EI): m/z (%) ϭ 176 (2.55, [M]ϩ), 146 (100,
[M Ϫ H2O]ϩ), 102 (57.38), 58 (70.21). Ϫ IR (KBr): ν˜ ϭ 3343 (br.
s, NϪH ϩ OϪH), 2939, 2869 (s, CϪH), 1593 (m, NϪH), 1373 (m,
OϪH), 1464 (s, CϪH), 1216 (m, CϪN), 1063 (s, CϪO).
Experimental Section
General Remarks: All manipulations were carried out under argon.
Solvents were purified by standard methods, freshly distilled and
degassed prior to use. Ϫ Infrared spectra were recorded in KBr
using a PerkinϪElmer IR 983 spectrometer. Ϫ NMR spectra were
recorded with a Bruker AM 250 spectrometer. Ϫ Elemental ana-
lyses (C,H,N,S) were performed with a Vario EL Elemental Ana-
lyzer. Ϫ EI and ϩFAB mass spectra were recorded with Finnigan
MAT 112 or Finnigan MAT 711 instruments. Ϫ UV/Vis spectra
were recorded with a PerkinϪElmer Lambda 9 UV/Vis/NIR spec-
trophotometer. Ϫ Cyclic voltammetric experiments were carried
out with a Bank High Power Potentiostat Wenking HP 72 and a
Bank Scan Generator Wenking Model VSG 83, using a three-elec-
trode cell configuration (working electrode: Pt; auxiliary electrode:
Pt; reference electrode: Ag/AgCl/3  KCl; EFc/Fcϩ ϭ 435 mV). The
experiments were performed in acetonitrile with 0.1  tBu4NPF6
as the supporting electrolyte with scan rates of 100 mV/s.
D: Anhydrous AlCl3 (18.0 g, 0.12 mol) was suspended in 50 mL of
dry dichloromethane of. This mixture was treated with a solution
of o-nitro-p-cresol (15.314 g, 0.1 mol) in 20 mL of dichloromethane.
When the evolution of HCl had ceased, bromine (17.5 g, 0.11 mol)
was added dropwise and the dark green suspension was stirred
overnight at room temperature. The mixture was hydrolyzed with
10 g of ice under vigorous evolution of HCl. Subsequently, 50 mL
of water and 30 mL of dichloromethane were added and the re-
sulting suspension was mixed with 30 mL of aqueous HCl (32%).
The clear orange organic layer was separated, washed twice with
acidic water and dried with anhydrous MgSO4. The solvent was
1
stripped in vacuo to give 21.33 g (91%) of a yellow powder. Ϫ H
NMR (250 MHz, CDCl3): δ ϭ 11.0 (s, 1 H, OH), 7.92, 7.74 (s, 2
H, Ar-H), 2.38 (s, 3 H, CH3). Ϫ 13C{1H} NMR (62.90 MHz,
CDCl3): δ ϭ 150.09, 141.76, 132.70, 124.04, 118.94, 112.81
(ArϪC), 20.19 (CH3).
Ligand Syntheses
B: 3,3Ј-Iminobis(ethyl propionate) B was prepared according to a
procedure published previously.[14] Ϫ 1H NMR (250 MHz, CDCl3):
δ ϭ 4.12 (q, 4 H, ϪOCH2), 2.92 (t, 4 H, NCH2), 2.49 (t, 4 H,
CH2COO), 1.84 (s, 1 H, NH), 1.24 (t, 6 H, CH3). Ϫ 13C{1H} NMR
(62.90 MHz, CDCl3): δ ϭ 172.38 (COO), 60.15 (CH2O), 44.68
(NCH2), 34.52 (CH2COO), 13.97 (CH3).
E: The sodium salt of D was obtained by deprotonation of the
phenolic OH group with aqueous NaOH. To achieve this D (11.6 g,
0.05 mol), dissolved in 24 mL of hot (70 °C) ethanol, was added
to a warm solution of NaOH (2.04 g, 0.051mol) in a mixture of 24
mL of ethanol and 3.5 mL of water. Immediately a red precipitate
separated, which was collected by filtration and dried in vacuo for
several hours. A sample of the sodium salt of D (12.7 g, 0.05 mol)
was suspended in 70 mL of dry THF. (Chloromethoxy)methane
(MOMCl) (4.026 g, 0.05 mol) was then added dropwise to this sus-
pension. Warning: (Chloromethoxy)methane (MOMCl) is known
to be extremely carcinogenic! The mixture was refluxed for 4 h.
During this period the color of the suspension changed from red
to white. The reaction mixture was extracted three times with a
saturated K2CO3 solution, and the combined organic layers were
dried with anhydrous MgSO4. Removal of the solvent yielded
11.95 g (86.6%) of a pale yellow powder, that was purified by
recrystallization from a mixture of hexane/ether (3:1, v:v). Ϫ 1H
NMR (250 MHz, CDCl3): δ ϭ 7.66, 7.54 (s, 2 H, Ar-H), 5.12 (s, 2
H, OCH2O), 3.54 (s, 3 H, OCH3), 2.34 (s, 3 H, CH3). Ϫ 13C{1H}
NMR (62.90 MHz, CDCl3): δ ϭ 137.89, 136.05, 124.36, 119.13
(ArϪC, only 4 resonances were observed), 100.68 (OCH2O), 58.01
(OCH3), 20.36 (CH3).
C: Formaldehyde (2.027 g, 0.025 mol, 37% in water) was added to
a solution of sodium pyrosulfite Na2S2O5 (2.376 g, 0.0125 mol) in
10 mL of water. The reaction mixture was heated to 50 °C for a
period of 45 min and then cooled to 0 °C. B (5.431 g, 0.025 mol)
was added slowly while maintaining the temperature below 5 °C.
Subsequently, the flask was allowed to warm to room temperature
and the reaction mixture was stirred for 2 h. The mixture was again
cooled to 0 °C and a solution of KCN (5.431 g, 0.025 mol) in 8 mL
of water was added. The resulting solution was stirred overnight at
room temperature. During this period an oil had separated, which
was extracted with chloroform (3 ϫ 50 mL). The organic phases
were combined and dried with anhydrous MgSO4. All solvents were
removed in vacuo, and the residue was distilled at 115 °C (2·10Ϫ2
1
mbar). Yield: 5.070 g (79.1% relative to B) of a colorless oil. Ϫ H
NMR (250 MHz, CDCl3): δ ϭ 4.12 (q, 4 H, OCH2), 3.64 (s, 2 H,
CH2CN), 2.90 (t, 4 H, NCH2), 2.48 (t, 4 H, CH2COO), 1.24 (t, 6
H, CH3). Ϫ 13C{1H} NMR (62.90 MHz, CDCl3): δ ϭ 171.39
(COO), 114.84 (CN), 60.41 (OCH2), 49.37 (NCH2), 41.96
(CH2CN), 32.85 (CH2COO), 13.99 (CH3).
F: Compound E (16.565 g, 0.06 mol) was dissolved in 250 mL of
dry THF and cooled to Ϫ105 °C with a mixture of diethyl ether/
liquid nitrogen as described previously.[18b] Over a period of 30 min
H4-1: All protected donor groups in C were liberated simul-
taneously by reduction with AlH3. The AlH3 was prepared from phenyllithium (37 mL, 0.066 mol, 1.8  solution in hexane/ether,
lithium aluminium hydride (LAH) (5.40 g, 0.143 mol) and H2SO4 70:30) was slowly added via a syringe maintaining the temperature
(6.183 g, 60 mmol, 96%) in 150 mL of dry THF as described.[13a]
below Ϫ100 °C. When all the phenyllithium had been added, the
Eur. J. Inorg. Chem. 2001, 2427Ϫ2436
2433

C. Ochs, F. E. Hahn, R. Fröhlich
FULL PAPER
red suspension was stirred for 1 h at Ϫ105 °C and subsequently
quenched with dry, degassed dimethyl disulfide (11.3 mL, 0.12
6.86 (s, 2 H, Ar-H), 5.01 (q, 2 H, OCH2O), 4.52 ϩ 4.41 (dd, 2 H,
NC(O)CH2), 4.37 ϩ 4.30 (dd, 2 H, NC(O)CH2CH2), 3.48 (s, 3 H,
mol). After 15 min, the mixture was allowed to warm to Ϫ78 °C. OCH3), 2.77 (q, 2 H, NCH2), 2.60 (t, 2 H, CH2CN), 2.46 (s, 3 H,
After stirring at this temperature for an additional 30 min, the or-
ange suspension was carefully hydrolyzed with 15 mL of degassed
water and slowly brought to 0 °C. Toluene (150 mL) was added
SCH3), 2.36 (s, 3 H, CH3), 2.02 (s, 3 H, C(O)CH3). Ϫ 13C{1H}
NMR (62.90 MHz, CDCl3): δ ϭ 170.78 [C(O)O ϩ C(O)N], 147.24,
135.69, 134.93, 134.24, 126.74, 126.70 (Ar-C), 117.89 (CN), 99.35
and the total volume of the solution was reduced to 100 mL in (OCH2O), 60.32 (OCH3), 47.51 [NC(O)CH2CH2O], 44.42
vacuo. Ether (150 mL) was then added. The resulting organic phase
was repeatedly extracted with water until the aqueous layer re-
mained colorless. After drying with anhydrous MgSO4 and removal
of the solvents in vacuo, a brown oily residue was obtained. Col-
umn-chromatographic purification (SiO2, hexane/ethyl acetate, 4:1,
[NC(O)CH2], 33.26 (NCH2), 21.00 (Ar-CH3), 20.88 [C(O)CH3],
16.41 (CH2CN), 14.70 (SCH3).
H4-2: All protected donor groups in I were liberated simultaneously
by use of AlH3. The AlH3 was prepared from lithium aluminium
hydride (LAH) (2.73 g, 72 mmol) and H2SO4 (3.065 g, 30 mmol,
96%) in 70 mL of dry THF as described.[13a] Ligand precursor I
(3.75 g, 10 mmol), dissolved in 30 mL of THF, was added dropwise
to the vigorously stirred AlH3 suspension. The reaction mixture
was stirred for 24 h at ambient temperature, residual AlH3 was
carefully hydrolyzed with water (5.4 mL, 0.3 mmol), and all solids
were removed by filtration. The solid residue was repeatedly ex-
tracted with methanol under argon. All organic phases were com-
bined and all solvents were removed in vacuo. The oily residue
obtained was dissolved in a small portion of dichloromethane and
this solution was filtered and dried with anhydrous MgSO4. Re-
moval of the solvent in vacuo yielded a yellow sticky oil that crys-
tallized at Ϫ30 °C. Alternatively, addition of diethyl ether to oily
H4-2 gave a white powder. Yield: 1.51 g (84.8%). Ϫ C14H24N2O2S
(284.42): calcd. C 59.12, H 8.51, N 9.85, S 11.27; found C 58.24,
H 8.49, N 9.58, S 11.37. Ϫ 1H NMR (250 MHz, CDCl3): δ ϭ 6.66
(s, 2 H, Ar-H), 5.64 (br. s, 4 H, Ar-OH, CH2ϪOH, NH2), 3.58 (t,
2 H, CH2OH), 2.98 (t ϩ t, 4 H, NCH2), 2.84 (t, 2 H, CH2NH2),
2.36 (s, 3 H, SCH3), 2.22 (s, 3 H, Ar-CH3), 1.60 (m ϩ m, 4 H,
CH2CH2CH2). Ϫ 13C{1H} NMR (62.90 MHz, CDCl3): δ ϭ 152.96,
136.99, 125.52, 124.34, 120.77, 118.36 (Ar-C), 60.95 (CH2OH),
52.99 (NCH2CH2CH2OH), 48.41 (NCH2CH2CH2NH2), 38.69
(CH2NH2), 29.06, 26.35 (CH2CH2CH2), 20.95 (Ar-CH3), 14.94
(SCH3). Ϫ MS (EI), m/z (%) ϭ 284 (38.79) [M]ϩ, 240 (20.22) [M
Ϫ CH2CH2NH2]ϩ, 196 (100), 182 (21.11), 134 (6.95). Ϫ IR (KBr):
1
Rf ϭ 0.29) yielded 10.05 g (72.3%) of an intense orange oil. Ϫ H
NMR (250 MHz, CDCl3): δ ϭ 7.30, 7.10 (s, 2 H, Ar-H), 5.02 (s, 2
H, OCH2O), 3.50 (s, 3 H, OCH3), 2.42 (s, 3 H, SCH3), 2.32 (s, 3 H,
CH3). Ϫ 13C{1H} NMR (62.90 MHz, CDCl3): δ ϭ 144.82, 144.18,
136.14, 135.03, 130.12, 121.11 (Ar-C), 100.06 (OCH2O), 57.88
(OCH3), 20.80 (CH3), 14.89 (SCH3).
G: A solution of F (7.298 g, 0.03 mol) in 100 mL of methanol was
cooled to 0 °C and mixed with hydrazine (6 g, 0.12 mol, 100%).
After addition of three small portions of Raney nickel, it was
stirred until the evolution of gas had ceased and the mixture decol-
orized. In order to decompose the excess hydrazine an additional
portion of Raney nickel was added and the suspension was heated
to reflux for 90 min. The solid metal was separated by filtration
under argon, and all solvents were removed in vacuo. The oily res-
idue obtained was dissolved in 50 mL of degassed dichloromethane
and the solution was dried with anhydrous MgSO4. Removal of the
1
solvent in vacuo yielded 5.77 g (90.2%) of G as a yellow oil. Ϫ H
NMR (250 MHz, CDCl3): δ ϭ 6.37, 6.34 (s, 2 H, Ar-H), 5.04 (s, 2
H, OCH2O), 3.94 (br. s, 2 H, NH2), 3.62 (s, 3 H, OCH3), 2.40 (s,
3 H, SCH3), 2.24 (s, 3 H, CH3). Ϫ 13C{1H} NMR (62.90 MHz,
CDCl3): δ ϭ 139.80, 139.51, 135.04, 132.04, 115.88, 113.95 (Ar-C),
99.11 (OCH2O), 57.58 (OCH3), 21.07 (CH3), 14.71 (SCH3).
H: Acrylonitrile (3.188 g, 0.06 mol) and G (8.535 g, 0.04 mol) were
mixed with the catalyst Cu(OAc)2·H2O (0.839 g) and the poly-
merization inhibitor quinone (0.17 g). The mixture was heated to
110 °C and the reaction was monitored by chromatography (SiO2,
hexane/ethyl acetate, 4:1, RfG ϭ 0.18, RfH ϭ 0.13). The reaction
was complete after 4 h, otherwise additional portions of acrylo-
nitrile and copper catalyst had to be added. The resulting brown
residue was purified by column chromatography (SiO2, hexane/
ethyl acetate). Yield 7.55 g (70.9%) of a brown crystalline solid. Ϫ
1H NMR (250 MHz, CDCl3): δ ϭ 6.36, 6.25 (s, 2 H, Ar-H), 5.02
(s, 2 H, OCH2O), 4.88 (br. s, 1 H, NH), 3.62 (s, 3 H, OCH3), 3.50
(q, 2 H, NHCH2), 2.62 (t, 2 H, CH2CN), 2.40 (s, 3 H, SCH3), 2.24
(s, 3 H, CH3). Ϫ 13C{1H} NMR (62.90 MHz, CDCl3): δ ϭ 139.74,
135.36, 132.00, 118.05, 115.31 (Ar-C, only 5 resonances were ob-
served), 108.87 (CN), 99.32 (OCH2O), 57.71 (OCH3), 39.56
(NHCH2), 21.56 (CH3), 18.04 (CH2CN), 14.68 (SCH3).
˜
ν: 3504 (s, NϪH), 3337 (s, OϪH), 2950, 2918, 2848, 2799 (s, CϪH),
1547 (m, NϪH), 1448 (m, OϪH), 1265 (m, CϪN), 1077 (s, CϪO),
830, 871 (m, CϪH of a 1,2,3,5-substituted aromatic ring).
Preparation of Copper Complexes
3: For the preparation of 3 freshly precipitated Cu(OH)2 (195 mg,
2 mmol) was suspended in water (10 mL) and ligand H4-1 (370 mg,
2.1 mmol) was added dropwise with a syringe. The blue suspension
was stirred at 40 °C for 60 min, and then solid NH4PF6 (652 mg,
4 mmol) was added, resulting in a deep blue solution. The solution
was reduced to a volume of 5 mL and cooled to 5 °C. After 24 h,
deep blue crystals had formed. The crystals were collected by filtra-
tion and recrystallized from warm acetonitrile/diethyl ether. Yield
596 mg (68.9%). Ϫ C16H38Cu2F12N4O4P2 (767.52): calcd. C 25.04,
H 4.99, N 7.30; found C 25.09, H 5.17, N 7.46. Ϫ MS (FAB, posit-
ive ions, 3-nitrobenzyl alcohol/DMSO), m/z (%) ϭ 475 (18.95, [M
Ϫ H]ϩ), 301 (37.67, [Cu(H3-1) ϩ Cu]ϩ), 239 (33.19, [3 ϩ H]2ϩ). Ϫ
I: H (5.332 g, 20 mmol) was dissolved in a mixture of dry diethyl
ether (25 mL) and chloroform (5 mL) and treated with acetoxypro-
pionyl chloride (3.09 g, 20.5 mmol) which was synthesized accord-
ing to a published procedure.[22] On reaction the mixture slightly
warmed and became darker in color. After 30 min, triethylamine
(2.086 g, 20.6 mmol) was added, and the solution was stirred for
one additional hour at room temperature. Washing of the reaction
solution with water, drying of the organic phase with anhydrous
˜
IR (KBr): ν: 3549, 3378, 3330 (w, OϪH ϩ NϪH), 2970, 2875, 2833
(m, CϪH), 1596 (m, NH2), 1473, 1459, 1433 (m, CϪH), 1065, 1046
(s, CϪO); 833, 558 (s, PF6). Ϫ UV/Vis (CH3OH): λmax [nm] (ε) ϭ
270 (4900), 366 (2000), 585 (120).
4: H4-2 (62.6 mg, 0.22 mmol) was mixed with triethylamine
MgSO4 and removal of the solvents in vacuo yielded a brown oil (44.4 mg, 0.44 mmol). The mixture was added dropwise to a solu-
that was purified chromatographically (SiO2, hexane/ethyl acetate,
1:1, RfI ϭ 0.15). The product crystallized as colorless needles.
Yield: 4.95 g (65%). Ϫ 1H NMR (250 MHz, CDCl3): δ ϭ 6.99,
tion of Cu(OAc)2·H2O (43.9 mg, 0.22 mmol) in 10 mL of methanol,
resulting in the formation of a green solution. After stirring for 30
min, the solvent was removed in vacuo and the residue suspended
2434
Eur. J. Inorg. Chem. 2001, 2427Ϫ2436

Coordination Chemistry of Unsymmetrical Tripodal Ligands with an NNO2 Donor Set
in acetonitrile and filtered. To the clear, deep green solution was
free of charge on application to CCDC, 12 Union Road, Cam-
FULL PAPER
added tert-butyl methyl ether. Cooling the solution to Ϫ30 °C bridge CB2 1EZ, UK [Fax: (internat.) ϩ 44-1223/336-033, E-mail:
yielded blue plate-shaped crystals. Yield 8.3 mg (4.6%) of deposit@ccdc.cam.ac.uk].
4·2CH3CN. Ϫ C36H58Cu2N6O8S2 (894.08): calcd. C 47.33, H 6.45,
N 6.90, S 7.89; found C 47.48, H 6.53, N 7.09, S 8.13. Ϫ MS (FAB,
positive ions, 3-nitrobenzyl alcohol/DMSO), m/z (%) ϭ 409 (11.34)
Acknowledgments
We thank the Deutsche Forschungsgemeinschaft and the Fonds der
Chemischen Industrie for financial support.
[Cu2(H3-2)]ϩ, 346 (18.20) [Cu(H3-2)]ϩ. Ϫ UV/Vis (CH3OH): λmax
[nm] (ε) ϭ 310 nm (15360), 400 (sh, 220), 650 (140).
Crystal Structure Analyses: Crystallographic data for complexes 3
and 4·2CH3CN are presented in Table 2. Crystals of 3 are air-
stable. Diffraction data were collected at room temperature with an
EnrafϪNonius CAD-4 diffractometer. No absorption correction
was applied. All non-hydrogen atoms were refined with anisotropic
thermal parameters. Hydrogen atoms were added at geometrically
idealized positions and were refined using a riding model. Complex
4·2CH3CN rapidly degenerates in air under loss of co-crystallized
solvent. Suitable specimens were mounted at 198 K on a Nonius
Kappa CCD diffractometer with a rotating anode generator. Dif-
fraction data were collected at 198 K. An absorption correction via
SORTAV was applied to the raw data (0.85 Յ T Յ 0.99). All other
non-hydrogen atoms were refined with anisotropic thermal para-
meters. Hydrogen atoms were added at geometrically idealized po-
sitions and were refined using a riding model. The following pro-
grams were used: Data collection EXPRESS and COLLECT, data
reduction MolEN and DENZO-SMN, absorption correction for
CCD data SORTAV, structure solution SHELXS-86 and SHELXS-
97, structure refinement SHELXL-97, graphical presentation
ZORTEP and SCHAKAL-92. Crystallographic data (excluding
structure factors) for the structures reported in this paper have been
deposited with the Cambridge Crystallographic Data Centre as
supplementary publication nos. CCDC-150423 and -150424 for 3
and 4·2CH3CN, respectively. Copies of the data can be obtained
[1] [1a]
J. W. Whittaker, in: Metal Ions in Biological Systems (Eds.:
H. Sigel, A. Sigel), M. Dekker, New York 1994, vol. 30, p.
315Ϫ359. Ϫ
[1b]
J. Stubbe, W. A. van der Donk, Chem. Rev.
1998, 98, 705 and references therein.
[2] [2a]
N. Ito, S. E. V. Phillips, K. D. S. Yadav, P. F. Knowles, J.
[2b]
Mol. Biol. 1994, 238, 794. Ϫ
N. Ito, S. E. V. Phillips, C.
Stevens, Z. B. Ogel, M. J. McPherson, J. N. Keen, K. D. S.
Yadav, P. F. Knowles, Nature 1991, 350, 87.
[3]
[3a] S. Itoh, S. Takayama, R. Arakawa, A. Furuta, M. Komatsu,
A. Ishida, S. Takamuku, S. Fukuzumi, Inorg. Chem. 1997, 36,
1407. Ϫ
[3b]
M. Ruf, C. G. Pierpont, Angew. Chem. 1998, 110,
1830; Angew. Chem. Int. Ed. 1998, 37, 1736.
[4]
[5]
J. Müller, T. Weyhermüller, E. Bill, P. Hildebrandt, L. Ould-
Moussa, T. Glaser, K. Wieghardt, Angew. Chem. 1998, 110,
637; Angew. Chem. Int. Ed. 1998, 37, 616.
[5a] J. A. Halfen, V. G. Young, Jr., W. B. Tolman, Angew. Chem.
1996, 108, 1832; Angew. Chem. Int. Ed. Engl. 1996, 35, 1687.
Ϫ
Berreau, E. C. Wilkinson, L. Que, Jr., W. B. Tolman, J. Am.
[5b]
J. A. Halfen, B. A. Jazdzewski, S. Mahapatra, L. M.
[5c]
Chem. Soc. 1997, 119, 8217. Ϫ
B. A. Jazdzewski, V. G.
Young, Jr., W. B. Tolman, Chem. Commun. 1998, 2521.
[6] [6a]
Y. Wang, T. D. P. Stack, J. Am. Chem. Soc. 1996, 118,
13097. Ϫ [6b] Y. Wang, J. L. DuBois, B. Hedmann, K. O. Hodg-
son, T. D. P. Stack, Science 1998, 279, 537.
[7] [7a]
P. Chaudhuri, M. Hess, T. Weyhermüller, K. Wieghardt,
Angew. Chem. 1999, 111, 1165; Angew. Chem. Int. Ed. 1999,
Table 2. Summary of crystallographic data for 3 and 4·2CH3CN
[7b]
38, 1095. Ϫ
P. Chaudhuri, M. Hess, U. Flörke, K. Wiegh-
ardt, Angew. Chem. 1998, 110, 2340; Angew. Chem. Int. Ed.
3
4·2CH3CN
1998, 37, 2217.
[8]
[9]
H. Adams, N. A. Bailey, I. K. Campbell, D. E. Fenton, Q.-Y.
He, J. Chem. Soc., Dalton Trans. 1996, 2233 and references
therein.
M. Vaidyanathan, R. Viswanathan, M. Palaniandavar, T.
Balasubramanian, P. Prabhaharan, T. P. Muthiah, Inorg. Chem.
1998, 37, 6418.
Empirical formula
Mr
T [K]
C16H38N4CuF12P2O4
767.52
293(2)
0.31ϫ0.22ϫ0.20
9.041(6)
13.356(2)
11.990(2)
90
98.54(3)
90
1419.8(10)
2
P21/n (no. 14)
1.795
1.721
EnrafϪNonius
CAD-4
C36H58N6Cu2O8S2
894.08
198(2)
0.15ϫ0.10ϫ0.05
9.978(1)
13.433(1)
16.398(1)
97.87(1)
Crystal size [mm]
˚
a [A]
[10]
˚
b [A]
M. M. Whittaker, W. R. Duncan, J. W. Whittaker, Inorg. Chem.
1996, 35, 382 and references therein.
˚
c [A]
α [°]
β [°]
γ [°]
V [A ]
Z
[11] [11a] D. Zurita, C. Scheer, J.-L. Pierre, E. Saint-Aman, J. Chem.
[11b]
94.94(1)
90.36(1)
Soc., Dalton Trans. 1996, 4331. Ϫ
D. Zurita, I. Gauthier-
´
Luneau, S. Menage, J.-L. Pierre, E. Saint-Aman, J. Biol. Inorg.
Chem. 1997, 2, 46. Ϫ [11c] Y. Shimazaki, S. Huth, A. Odani, O.
Yamauchi, Angew. Chem. 2000, 112, 1732; Angew. Chem. Int.
Ed. 2000, 39, 1666.
3
˚
2168.7(3)
2
¯
Space group
P1 (no. 2)
[12]
ρcalcd. [g cmϪ3
]
1.369
1.130
M. Duggan, N. Ray, B. Hathaway, G. Tomlinson, P. Brint, K.
Pelin, J. Chem. Soc., Dalton Trans. 1980, 1342.
µ Mo-Kα [mmϪ1
]
Diffractometer
EnrafϪNonius
Kappa CCD
0.71073
2.05 Յ 2θ Յ 25.0
Ϯh, Ϯk, Ϯl
7597
[13] [13a] A. M. Dittler-Klingemann, F. E. Hahn, Inorg. Chem. 1996,
[13b]
35, 1996. Ϫ
A. M. Dittler-Klingemann, C. Orvig, F. E.
˚
λ [A]
0.71073
2.3 Յ 2θ Յ 26.4
h, k, Ϯl
Hahn, F. Thaler, C. D. Hubbard, R. van Eldik, S. Schindler, I.
[13c]
´
´
2θ range [°]
Index range
Unique data
Observed data
[I Ն 2σ(I)]
R
Fabian, Inorg. Chem. 1996, 35, 7798. Ϫ
F. Thaler, C. D.
´
Hubbard, F. W. Heinemann, R. van Eldik, S. Schindler, I. Fab-
ian, A. M. Dittler-Klingemann, F. E. Hahn, C. Orvig, Inorg.
´
2918
2003
5092
Chem. 1998, 37, 4022.
[14] [14a]
F. E. Hahn, S. Rupprecht, Chem. Ber. 1991, 124, 481. Ϫ
F. E. Hahn, S. Rupprecht, Chem. Ber. 1991, 124, 487. Ϫ
B. Wolff, A. Weiss, Angew. Chem. 1986, 98, 173; Angew.
[14b]
[14c]
0.0414
0.0952
257
0.0597
0.1111
735
wR
No. of variables
GOF
Chem. Int. Ed. Engl. 1986, 25, 162.
[15]
[16]
1.050
1.022
E. C. Horning (Ed.), Org. Synth., Coll. Vol. III 1955, 275.
L. C. Raiford, J. Am. Chem. Soc. 1919, 41, 2072.
Eur. J. Inorg. Chem. 2001, 2427Ϫ2436
2435

C. Ochs, F. E. Hahn, R. Fröhlich
S. A. Heininger, J. Org. Chem. 1957, 22, 1213.
T. L. Gresham, J. E. Jansen, F. W. Shaver, J. Am. Chem. Soc.
1948, 70, 1003.
A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn, G. C. Ver-
schoor, J. Chem. Soc., Dalton Trans. 1984, 1349.
C. Ochs, F. E. Hahn, R. Fröhlich, Chem. Eur. J. 2000, 6, 2193.
M. M. Whittaker, P. J. Kersten, N. Nakamura, J. Sanders-
Loehr, E. S. Schweizer, J. W. Whittaker, J. Biol. Chem. 1996,
271, 681.
FULL PAPER
[17] [17a]
[21]
[22]
K. Brand, W. Schreber, Ber. Dtsch. Chem. Ges. 1942, 75,
[17b]
156. Ϫ
M. R. Agharahimi, N. A. LeBel, J. Org. Chem.
1995, 60, 1856.
[18] [18a]
[23]
I. R. Hardcastle, P. Quayle, E. L. M. Ward, Tetrahedron
[18b]
Lett. 1994, 35, 1747. Ϫ
G. Voß, H. Gerlach, Chem. Ber.
[18c]
[24]
[25]
1989, 122, 1199. Ϫ
103, 1412.
G. Köbrich, P. Buck, Chem. Ber. 1970,
[19] [19a]
B. J. Wakefield, Organolithium Methods, Academic Press,
London, 1990. Ϫ [19b] L. Brandsma, H. Verkruijsse, Preparative
Polar Organometallic Chemistry, Springer Verlag, Berlin, 1987.
S. Cabiddu, A. Maccioni, M. Secci, V. Solinas, Gazz. Chim.
Ital. 1969, 99, 397.
Received October 2, 2000
[20]
[I00385]
2436
Eur. J. Inorg. Chem. 2001, 2427Ϫ2436