Synthesis, spectroscopic and magnetostructural evidence for the formation of Cu(II) complexes of pyridyl-2-carboxylate (2-pca) and quinolyl-2-carboxylate (2-qca) as a result of a novel oxidative P-dealkylation reaction of diethyl 2-pyridylmethylphosphonate (2-pmpe) and diethyl 2-quinolylmethylphosphonate (2-qmpe) ligands

Article abstract of DOI:10.1016/j.ica.2003.06.017

The interaction of diethyl 2-pyridylmethylphosphonate (2-pmpe) ligand with CuX₂ salts unexpectedly leads to the formation of compounds of the formula Cu(2-pca)₂ [X=Cl^- (1), CH₃COO ^- (3)], and Cu(2-pca)Cl [X=Cl^- (2)] (2-pca=pyridine-2- carboxylate ion). The diethyl 2-quinolylmethylphosphonate ligand (2-qmpe) reacts with CuX₂ salts to similarly yield compounds of stoichiometry Cu(2-qca)₂·H₂O (X=ClO₄^- (4)], and for X=Cl Cu(2-qca)₂·H₂O (5) and Cu(2-qca)Cl (6), (2-qca=quinoline-2-carboxylate ion). These compounds are products of a novel oxidative P-dealkylation reaction, which takes place on 2-pmpe and 2-qmpe ligands under the used conditions. The compounds were characterized by infrared, ligand field, EPR spectroscopy and magnetic studies. Cu(2-pca)₂ exists in two crystalline forms, a blue form (1) and a violet form (3). For 3 the single-crystal structure was determined. The copper atom is four-coordinated in a square-planar geometry. The stack between related (and hence parallel) pca moieties involves interatomic distances of 3.27 A?. Cu(2-qca)₂·H₂O also exists in two forms, a green (4) and a blue-green (5). Both these complexes are five coordinated, involve the same CuN₂O₃ chromophore and are examples of the distortion isomers. Variable-temperature magnetic susceptibility measurements (1.9-300 K) have shown that the antiferromagnetic coupling observed is much stronger in 6 than in 1, 3, 4 and 5. For 2 a ferromagnetic exchange occurs.

Full text of DOI:10.1016/j.ica.2003.06.017

Inorganica Chimica Acta 357 (2004) 755–763
www.elsevier.com/locate/ica
Synthesis, spectroscopic and magnetostructural evidence for
the formation of Cu(II) complexes of pyridyl-2-carboxylate
(
2-pca) and quinolyl-2-carboxylate (2-qca) as a result of a
novel oxidative P-dealkylation reaction of diethyl
2
-pyridylmethylphosphonate (2-pmpe) and diethyl
-quinolylmethylphosphonate (2-qmpe) ligands
2
Bogumiła Zurowska , Justyn Ochocki , Jerzy Mrozi nꢀ ski a,*,
a,
b
_
*
a
Zbigniew Ciunik , Jan Reedijk
c
a
Faculty of Chemistry, University of Wroclaw, 14F. Joliot-Curie St., Wroclaw 50-383, Poland
Department of Bioinorganic Chemistry, Faculty of Pharmacy, Medical University, Lodz 90-151, Poland
b
c
Leiden Institute of Chemistry, Gorlaeus Laboratories, Leiden University, Leiden 2300 RA, The Netherlands
Received 9 May 2003; accepted 16 June 2003
Abstract
The interaction of diethyl 2-pyridylmethylphosphonate (2-pmpe) ligand with CuX2 salts unexpectedly leads to the formation of
compounds of the formula Cu(2-pca)2 [X ¼ Clꢀ (1), CH3COO (3)], and Cu(2-pca)Cl [X ¼ Clꢀ (2)] (2-pca ¼ pyridine-2-carboxylate
ion). The diethyl 2-quinolylmethylphosphonate ligand (2-qmpe) reacts with CuX2 salts to similarly yield compounds of stoichi-
ꢀ
ꢀ
ometry Cu(2-qca)2 ꢁ H2O (X ¼ ClO (4)], and for X ¼ Cl Cu(2-qca)2 ꢁ H2O (5) and Cu(2-qca)Cl (6), (2-qca ¼ quinoline-2-carbox-
4
ylate ion). These compounds are products of a novel oxidative P-dealkylation reaction, which takes place on 2-pmpe and 2-qmpe
ligands under the used conditions. The compounds were characterized by infrared, ligand field, EPR spectroscopy and magnetic
studies. Cu(2-pca)2 exists in two crystalline forms, a blue form (1) and a violet form (3). For 3 the single-crystal structure was
determined. The copper atom is four-coordinated in a square-planar geometry. The stack between related (and hence parallel) pca
ꢁ
moieties involves interatomic distances of 3.27 A. Cu(2-qca)2 ꢁ H2O also exists in two forms, a green (4) and a blue–green (5). Both
these complexes are five coordinated, involve the same CuN2O3 chromophore and are examples of the distortion isomers. Variable-
temperature magnetic susceptibility measurements (1.9–300 K) have shown that the antiferromagnetic coupling observed is much
stronger in 6 than in 1, 3, 4 and 5. For 2 a ferromagnetic exchange occurs.
Ó 2003 Elsevier B.V. All rights reserved.
Keywords: Copper(II) complexes; N,O ligands; Magnetic properties; Crystal structures; Oxidation
1
. Introduction
gands [1,2], containing two potential donor atoms, the
heterocyclic nitrogen and phosphoryl oxygen atoms,
towards transition-metal ions. Crystal structures of two
representative compounds, i.e. [Co(4-pmpe) (H O) ]
Recently, we have reported the synthesis and coor-
dination properties of 2-pmpe, 4-pmpe and 2-qmpe li-
2
2
2
[
(ClO ) ] ꢁ 2H O and [M(2-qmpe) (H O) ][(ClO ) ],
4
2
2
4
2
2
4 2
where M ¼ Ni, Mn were determined. The results of
structural and physicochemical studies of these com-
plexes, as has been shown earlier [1,2], are interesting
from the point of view of inorganic and bioinorganic
*
Corresponding authors. Tel.: +48-71-204307; fax: +48-71-204420
_
(
J. Mrozi nꢀ ski), +48-71-328-2348 (B. Zurowska).
E-mail addresses: zurowska@wchuwr.chem.uni.wroc.pl (B. Zuro-
wska), jmroz@wchuwr.chem.uni.wroc.pl (J. Mrozi nꢀ ski).
_
0
020-1693/$ - see front matter Ó 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2003.06.017

_
B. Zurowska et al. / Inorganica Chimica Acta 357 (2004) 755–763
756
Standard Schlenk techniques were used un the experi-
mental performed under a dinitrogen atmosphere. The
solvents and solutions were deoxygenated by bubbling
with dinitrogen. For IR and EPR measurements the air-
sensitive solutions of compounds were prepared in a
drybox under an inert atmosphere of argon or dinitrogen.
Metal content was determined using a Carl Zeiss
Jena atomic absorption spectrophotometer and a ARL
Model 3410 ICP spectrometer. Elemental analyses were
carried out using a Perkin–Elmer elemental analyzer
Fig. 1. A schematic drawing of the starting 2-pmpe and 2-qmpe
ligands.
2
400CHN.
Infrared spectra (100–4000 cm 1) were recorded on a
Bruker IFS 113v spectrophotometer using KBr pellets,
ꢀ
chemistry. Thus, we performed a further study on the
coordination properties of some copper(II) salts with
and solid-state electronic spectra (28 000–4000 cm 1
ꢀ
)
2
-pmpe and 2-qmpe (Fig. 1) ligands [3].
The results indicate that the interaction of Cu(II) with
-pmpe and 2-qmpe unexpectedly produces the com-
were performed on a Cary 500 spectrophotometer.
2
EPR spectra were recorded at X-band at room tem-
perature, 77 K on an ESP 300E Bruker spectrometer
with an ER 035 M Bruker NMR gaussmeter and a
Hewlett-Packard microwave frequency counter.
plexes of pyridine-2-carboxylate (2-pca) and quinoline-
-carboxylate (2-qca) ions, respectively. From this re-
2
actions the compounds of formula: Cu(2-pca)2 (blue and
violet form), Cu(2-pca)Cl and Cu(2-qca)2 ꢁ H2O (blue–
green and green form), and Cu(2-qca)Cl were isolated.
The object of this investigations has been synthesis,
properties and the study of the conditions of formation
of the isolated compounds. The mechanism of this re-
action will be the object of further investigation. In the
present paper, we report the synthesis, spectral and
magnetic properties of copper(II) compounds of 2-pca
and 2-qca obtained in an unusual reaction.
For the violet form of Cu(2-pca)2, the X-ray crystal
structure study was performed. It is emphasized that the
described compound is a novel square-planar complex
of copper(II) with pyridine-2-carboxylate. Until now
only a few structural data for such adducts have been
reported [4–6]. It is interesting from synthetic and
structural points of view that the violet form of Cu(2-
pca)2 cannot be obtained directly from picolinic acid,
which was checked experimentally. The crystal structure
data of Cu(2-pca)Cl obtained by some of us in an oxi-
dative reaction agreed with the data of the same com-
pound prepared directly from picolinic acid [7] and is
not repeated here. The compound crystallizes in the te-
tragonal system and its lattice constants agree with the
values reported earlier [7]. It is interesting to underline
that the Cu(2-pca)Cl is a two-dimensional (2D) polymer
in which ferromagnetic exchange via dichloro and syn–
anti carboxylate bridges occurs. Some of us elsewhere [8]
discussed the magnitude of its ferromagnetic interaction
on the basis of the structural data.
Magnetic susceptibility measurements were carried
out with a Quantum Design SQUID magnetometer
(type MPMSXL-5). Measurements were recorded at a
magnetic field of 5 kG in the temperature range 1.9–300
K. A calibrated gallium arsenide diode was used for the
temperature measurements. The corrections for dia-
magnetism of the constituent atoms were made by use of
Pascal constants [9] and the effective magnetic moments
1
=2
were calculated from leff ¼ 2:83ðv TÞ using temper-
M
ꢀ
6
cm3
ature-independent paramagnetism of 60 ꢂ 10
ꢀ
1
mol for Cu(II).
2.2. X-ray crystallography
Crystal data of Cu(2-pca)2 (3) are given in Table 1,
together with refinement details. All measurements of
crystal were performed on a Kuma KM4CCD j-axis
Table 1
Crystal data and structure refinement for Cu(2-pca)2 (3)
Empirical formula
Formula weight
Space group
Crystal system
T (K)
12 8 2 4
C H N O Cu
307.74
P2 =c
monoclinic
1
100(2)
ꢁ
k (A)
0.71073
3.6975(7)
11.9890(19)
11.8886(19)
91.108(14)
526.92(15)
2
ꢁ
a (A)
ꢁ
b (A)
ꢁ
c (A)
b (°)
V (A )
ꢁ
3
Z
3
D
calc (Mg m )
1.940
2.085
2
. Experimental
l (mm)
F ð000Þ
310
0, 2.38, 3
1.296
2
.1. General comments and physical measurements
a
Weighting scheme: a; b; c
2
Goodness-of-fit (F )
Most of the starting materials and solvents for syn-
thesis were obtained commercially and used as received.
Final R
1
=wR
2
indices (I > 2r
I
)
0.0629/0.1068
a
2
2
o
2
2
o
2
c
w¼1=½r ðF ÞþðaꢂPÞ þbꢂPꢃ, where P ¼½Max:ðF ;0Þþ2ꢂF ꢃ=c.

_
B. Zurowska et al. / Inorganica Chimica Acta 357 (2004) 755–763
757
diffractometer with graphite-monochromated MoKa
radiation. The crystal was positioned at 65 mm from the
KM4CCD camera. Six hundred and twelve frames were
measured at 0.75° intervals with a counting time of 45 s.
No absorption correction was applied. Data reduction
and analysis were carried out with the Kuma diffraction
Found: C, 32.88; H, 1.84; N, 6.55; Cl, 16.41; Cu,
28.88%.
2
.3.2. Cu(2-pca)
2
(3)
3
This compound was prepared by adding 35 cm of a
methanol solution of copper(II) acetate monohydrate (1
mmol) to 15 cm of an ethanol solution of the 2-pmpe
(
methods (program SHELXS97 [10]) and refined by the
Wrocław) programs. The structure was solved by direct
3
ligand (2 mmol). From the blue reaction mixture after
evaporation of the solvent green crystals of [Cu(CH3
COO)2(2-pmpe)] (3a) crystallize. Dissolution of this
compound in ethanol and repeated evaporation of the
solvent lead to a dark-green solution from which the
violet single crystals of Cu(2-pca)2, suitable for X-ray
analysis, slowly precipitated. The violet product was
filtered off, washed carefully with ethanol and ether.
Anal. Calc. for C H CuN O (3): C, 46.54; H, 2.62; N,
2
full-matrix least-squares method on all F data using the
SHELXL97 [11] programs. Non-hydrogen atoms were
refined with anisotropic thermal parameters; hydrogen
atoms were included from Dq maps and were refined
with isotropic thermal parameters. Measurements of
the unit cell parameters of a single crystal of Cu(2-
pca)Cl (2) were performed on a Kuma KM4CCD j-axis
diffractometer with graphite-monochromated MoKa
radiation.
1
2
8
2
4
9
2
.10; Cu 20.65, Found: C, 46.32; H, 2.58; N, 8.95; Cu,
0.79%. Anal. Calc. for C14H22CuNPO7 (3a): C, 40.92;
2
.3. Synthesis of ligands and complexes
H, 5.41; N, 3.41; Cu, 15.47, Found: C, 41.23; H, 5.29; N,
3.62; Cu, 15.16%.
The ligand diethyl 2-pyridylmethylphosphonate
2-pmpe) and diethyl 2-quinolilmethylphosphonate (2-
(
2
.3.3. Cu(2-qca) ꢁ H O (4)
2
2
qmpe) were prepared according to the procedures de-
scribed in details elsewhere [2,12,13]. The coordination
compounds were prepared by dissolving the appropriate
This compound was obtained by recrystallization of
the crude reaction product precipitated in the reaction
of Cu(ClO4)2 ꢁ 6H2O with 2-qmpe ligand from metha-
nol solution. This reaction was performed according
to the general procedure described above. Anal. Calc.
for C20H14CuN2O5 (4): C, 56.40; H, 3.32; N, 6.58;
Cu,14.92, Found: C, 56.75; H, 3.24; N, 6.43; Cu,
3
hydrate copper(II) salts (1 mmol) in ethanol (10 cm )
and dropwise adding to the stirred solution of the ligand
3
(
2 mmol) in ethanol (15 cm ). The resulting solution was
filtered off, to remove any impurities. Details for each
case are given below.
1
4.68%.
2
.3.1. Cu(2-pca)
2
(1) and Cu(2-pca)Cl (2)
2
.3.4. Cu(2-qca) ꢁ H O (5)
2
2
The reaction of CuCl2 ꢁ 2H2O with 2-pmpe per-
The reaction of CuCl2 ꢁ 2H2O with 2-qmpe was per-
formed according to the procedure described above,
results in a dark-green solution. Evaporation of sol-
vent, leads to the precipitation of green viscous oily
product, which, crystallizes very slowly (over a few
weeks). This product of crystallization was not ob-
tained in a pure form. A mixture of two compounds
blue and blue–green was obtained using this method.
The blue compound 1 in a pure form was precipitated
from the oil mixture (after evaporation of part of the
solvent) using acetone. Subsequently, with from the
green residue, the blue–green compound 2 in a pure
form was precipitated using ether. Blue–green single
crystals of 2 suitable for X-ray identification were ob-
tained with from the reaction mixture by slow evapo-
ration from the corresponding reaction mixture (over a
few weeks) of the solvent. The attempts of the prepa-
ration of blue form suitable for X-ray identification
were unsuccessful.
formed according to the general procedure described
above. The solid product was precipitated during slow
evaporation of the solvent, and then washed with cold
ethanol and ether. Anal. Calc. for C20H14CuN2O5 (5): C,
5
3
6.40; H, 3.32; N, 6.58; Cu,14.92, Found: C, 56.95; H,
.31; N, 6.23;Cu,14.92%.
2.3.5. Cu(2-qca)Cl (6)
Unexpectedly in the reaction of CuCl with 2-qmpe
also another green product of the formula Cu(2-qca)Cl
2
was isolated. The reaction was carried out under similar
conditions to these used for the preparation of Cu(2-
qca) ꢁ H O (4). The reason is probably a delicate kinetic
2
2
balance between a few possible courses of the reaction.
Anal. Cal. for C H CuNO Cl (6); C, 44.29; H, 2.23; N,
1
0
6
2
5.17; Cl, 13.07; Cu, 23.43, Found: C, 44.61; H, 2.14; N,
5.26; Cl, 13.66; Cu, 23.68%.
Anal. Calc. for C H CuN O (1): C, 46.84; H,
8
Caution! Although no problem was encountered in
this work, perchlorate salts containing organic ligands
are potentially explosive. They should be prepared in
small quantities and handled with care.
12
2
4
2
9
(
.62; N, 9.10; Cu, 20.65, Found: C, 46.58; H, 2.86; N,
.34; Cu, 20.88%. Anal. Calc. for C6H4CuNO2Cl
2): C, 32.60; H, 1.83, N, 6.34, Cl, 16.03; Cu, 28.74,

_
B. Zurowska et al. / Inorganica Chimica Acta 357 (2004) 755–763
758
3
. Results and discussion
2-pca ligand acts as a chelating didentate ligand coor-
dinating through a pyridine nitrogen and carboxylate
oxygen atoms. The second carboxylate oxygen atom is
not involved in the coordination to the metal ion. The
copper atom has a trans square-planar configuration,
which is clearly demonstrated by the angle N(1)–Cu(1)–
O(1): 83.92(15)°. The Cu–N and Cu–O distances of
3
.1. Crystal structure of 3
The crystallographic parameters are summarized in
Table 1. The bond distances and angles are listed in
Table 2.
The molecular structure with crystallographic num-
bering of the compound is depicted in Fig. 2. Stacking of
molecules Cu(2-pca)2 (2-pca ¼ pyridine-2-carboxylate)
in the crystal is depicted in Fig. 3.
ꢁ
1.958(4) and 1.934(3) A, respectively, agree with the
values reported for other copper-2-picolinate complexes
[4–6]. The average C–C and C–N lengths in the pyridine
ꢁ
ring of 1.380 and 1.345 A, respectively, are typical. The
crystal packing (Fig. 3) is mainly due to short stacking
The structure consists of mononuclear units in which
the copper atom lies at the centre of symmetry. The
ꢁ
(3.27 A interactions between pyridine rings belonging to
different molecules giving rise to a mono-dimensional
polymeric network arrangement, resulting a Cu–Cu
Table 2
ꢁ
contacts of 3.6975(7) A.
ꢁ
Selected bond lengths (A) and angles (°) for Cu(2-pca)2 (3)
Cu(1)–O(1)
Cu(1)–N(1)
O(1)–C(1)
O(2)–C(1)
N(1)–C(6)
N(1)–C(2)
C(1)–C(2)
1.934(3)
1.958(4)
1.287(6)
1.229(6)
1.340(6)
1.351(6)
1.506(7)
O(1)–Cu(1)–N(1)
C(1)–O(1)–Cu(1)
C(6)–N(1)–C(2)
C(6)–N(1)–Cu(1)
C(2)–N(1)–Cu(1)
O(2)–C(1)–O(1)
O(2)–C(1)–C(2)
O(1)–C(1)–C(2)
N(1)–C(2)–C(1)
C(3)–C(2)–C(1)
83.92(15)
114.4(3)
119.4(4)
129.1(3)
111.4(3)
125.6(5)
119.6(5)
114.7(4)
114.3(4)
124.1(4)
3
.2. Infrared spectra
The colors and the most relevant IR absorption
bands of the copper(II) compounds are listed in Table 3.
All the compounds show the strong carboxylate
stretching frequencies m (COO) and m (COO) in the
range characteristic of copper(II) carboxylate com-
pounds [14,15]. For compounds 1, 3, 4 and 5 the Dm
separation values ca. 260 or 300 cm of m(COO ) bands
indicate that the coordination of the carboxylate group
is monodentate [14,15]. However, for compounds 2 and
as
s
ꢀ
1
ꢀ
ꢀ
1
6
the observed Dm separation value 186 cm is in the
range expected for a bridging carboxylate group [14,16].
In the complexes 4 and 5, coordinating water is present.
The observed a strong bands in the m(OH) stretching
region are in good agreement with the polymeric inter-
molecular O–H ꢁ ꢁ ꢁ O association [17]. In the far-IR re-
gion, metal-nitrogen stretching vibrations are observed
(Table 3). For 3 of the known structures (including the
present one), where the CuN2O2 chromophore has a
trans square-planar geometry and for 1 of identical
stoichiometry, only a single absorption band corre-
sponding to Cu–N and Cu–O stretches is observed. This
is consistent with the criteria published by Herlinger
Fig. 2. Molecular structure with crystallographic numbering of
Cu(2-pca)2 (3). Atoms are drawn at the 30% probability level.
[
18]. For the cis-isomer, two absorption bands would be
expected [18], so we assume that the blue form of 1 also
has a trans-configuration. According to the literature
[
19,20] the positions of the bands assignable to Cu–Cl
vibrations for 2 and 6 (Table 3), suggest that the chlo-
ride anions act as bridged between copper ions. For 2
this is consistent with the reported crystal structure [7,8].
3
.3. Ligand-field and EPR spectra
The X-band EPR spectra of the polycrystalline sam-
ples of 1 and 3 of formula Cu(2-pca)2 are of axial type
(g ¼ 2:237, g? ¼ 2:066 at r.t. and g ¼ 2:238, g? ¼ 2:068
Fig. 3. Stacking of molecules of Cu(2-pca)2 (3) in the crystal. Per-
ꢁ
pendicular distances between molecules are 3.27 A. Copper–copper
ꢁ
contacts are 3.6975(7) A.
k
k
at 77 K for 1, and g ¼ 2:219, g? ¼ 2:065 at r.t. and
k

_
B. Zurowska et al. / Inorganica Chimica Acta 357 (2004) 755–763
759
g ¼ 2:218, g? ¼ 2:075 at 77 K for 3). This was expected
k
for copper(II) ions with d
2
2
(or dxy) ground state [21].
x ꢀy
X-ray analysis of 3 demonstrated, that CuN2O2 chro-
mophore is rigorously square-planar. The shape (the
broad flap band) of d–d band of 3 with maximum at
ꢀ
1
1
7 750 and 16 750 cm is also typical for the square-
planar coordination of the copper(II) ion
21] For 1 the maximum at 16 600 cm is observed. The
ꢀ
1
[
difference between g and d–d band values for 1 with
k
respect to 3 is in full agreement with their different
copper environment (very elongated tetragonal octahe-
dral and square-planar, respectively) [21]. The ligand
field spectra of 2 and 5 exhibit similar poorly resolved
broad-band envelopes with two maxima of almost
comparable intensity. These maxima occurring at 13 550
ꢀ
1
ꢀ1
and 9650 cm for 2, and 13 950 and 9550 cm for 5 are
characteristic for Cu(II) chromophores with intermedi-
ate stereochemistries between those of the tetragonal
pyramid and trigonal bipyramid [21]. In contrast, 4 and
6
show strong unresolved bands at 11 950 and 11 850
cm 1, respectively, with a poorly resolved high-energy
shoulder, suggesting a trigonal-bipyramidal coordina-
tion [21]. For the powdered samples of 2 and 6 the X-
band EPR spectra show single broad absorption only
ꢀ
(
giso ¼ 2:17 for 2, g ¼ 2:14 for 6) at r.t. and 77 K due
iso
to exchange coupling. The X-band EPR spectra of the
polycrystalline solids of Cu(2-qca)2 ꢁ H2O (4) and (5)
compounds at room temperature and 77 K exhibit only
one isotropic line with a small deformation with g value
approximately at 2.13 and 2.09, respectively, and does
not allow conclusions about the geometry. Based on the
ligand-field spectral properties 4 and 5 complexes may
be considered as the examples of the distortion isomers
differing by the degree of distortion of the CuN2O2O
chromophore.
3.4. Magnetic properties
Variable-temperature (1.9–300 K) magnetic suscepti-
bility data were collected for solid samples of 1 and 3.
The magnetic susceptibility data of both complexes obey
the Curie–Weiss law between 25 and 300 K
corr
v
¼ ½C=ðT ꢀ HÞꢃ;
M
2
2
where the Curie constant C ¼ Ng b SðS þ 1Þ=3K and
S ¼ 1=2. The temperature variation of magnetic
susceptibility and v T of 1 and 3 are shown in Figs. 4
M
and 5, respectively.
The value of v T decreases in the low-temperature
M
3
region with decreasing temperature. They are 0.373 cm
ꢀ
1
3
K mol (leff ¼ 1:73l ) at 300 K and 0.259 cm
K
B
ꢀ
ꢀ
1
3
mol (leff ¼ 1:44l ) at 1.9 K for 1, and 0.423 cm K
B
1
3
ꢀ1
mol (leff ¼ 1:84l ) at 300 K and 0.262 cm K mol
B
(
leff ¼ 1:45l ) at 1.9 K for 3. The best values of the
B
Curie (C) and Weiss (h) constants for 1 are C ¼ 0:375
3
ꢀ1
cm K mol and h ¼ ꢀ0:63 K. For 3 these values are

_
B. Zurowska et al. / Inorganica Chimica Acta 357 (2004) 755–763
760
Ising chain model, but also by interaction of the mag-
netic center of Cu(II), Eq. (1) in an molecular field, Eq.
(2) [24]:
2
2
Nb g
vCu
¼
ð1Þ
ð2Þ
4
kT
v
corr
Cu
Cu
2zJ0
v
¼ 1
;
ꢀ
vCu
2
2
Nb g
where vCu is the equation for the magnetic susceptibility
corr
Cu
of paramagnetic center, v
is the measured experi-
0
mental susceptibility, zJ is intermolecular exchange
parameter, z is the number of nearest neighbors. Least-
squares fitting of the experimental susceptibility data
in this equation gives zJ ¼ ꢀ1:43 cm (z ¼ 2) and
Fig. 4. Plot of vCu (ꢅ) and vCuT (ꢆ) vs. T for Cu(2-pca)2 (1). The solid
line is the results of the best fit obtained with the parameters reported
in the text.
0
ꢀ1
ꢀ
5
0
g ¼ 2:09 (R ¼ 6:32 ꢂ 10 ). As expected this J value is
of the same order as that obtained from the Ising model.
Investigation of the magnetic susceptibility of 2
shows an increase of the magnetic moment suggesting
ferromagnetically coupled copper(II) ions similar to the
compound of the same composition and structure, pre-
viously prepared directly from picolinic acid [7,8]. The
interpretation of the magnetic behavior has been pre-
sented by some of us elsewhere [8].
Variable-temperature (4.2–300 K) magnetic suscepti-
bility data were collected for 4 and 5. The values of vMT
are slightly dependent on the temperature. They are
3
ꢀ1
0
.432 cm K mol (leff ¼ 1:86l ) at 300 K and 0.404
B
3
ꢀ1
cm K mol (leff ¼ 1:80l ) at 4.2 K for 4, and 0.413
B
3
ꢀ1
3
Fig. 5. Plot of vCu (ꢅ) and vCuT (ꢆ) vs. T for Cu(2-pca)2 (3). The solid
line is the results of the best fit obtained with the parameters reported
in the text.
cm K mol (leff ¼ 1:82l ) at 300 K and 0.384 cm K
B
ꢀ
1
mol
susceptibility data of both complexes obey the Curie–
Weiss law within the measured temperature region with
(leff ¼ 1:75l ) at 4.2 K for 5. The magnetic
B
3
ꢀ1
3
ꢀ1
C ¼ 0:416 cm K mol and h ¼ ꢀ0:24 K for 4, and
C ¼ 0:375 cm K mol and h ¼ ꢀ2:82 K. A negative
Weiss constant value for both complexes can suggests a
very weak intermolecular antiferromagnetic interaction
occurring in the crystal lattice of 1 and 3 in which un-
coordinated oxygen atoms of carboxylate groups and
stacking interaction can participate, respectively. On the
basis of the spectral and magnetic properties a polymeric
structure of 1 is proposed with pseudo-octahedrally
coordinated copper(II) ion linked by carboxylate oxygen
donors of the picolinate group. For the complexes 1 and
3
ꢀ1
C ¼ 0:397 cm K mol and h ¼ ꢀ0:25 K for 5. The
small negative value of h obtained for both complexes
may suggests the existence of a very weak intermolecular
antiferromagnetic exchange between copper ions within
crystal lattice via the hydrogen bond system and J must
be considered as less than ꢄ)0.5 cmꢀ1.
The temperature-dependent magnetic susceptibility
studies of Cu(2-qca)Cl (6) presented below revealed
antiferromagnetic coupling. This result contrasts mark-
edly with magnetic behavior of Cu(2-pca)Cl (2) in which
the dichloro bridge and the syn–anti carboxylate bridge
conformation produces ferromagnetic coupling (J ¼ 15
3
, data were analyzed using an Ising chain model [22,23].
Least-squares fitting of the experimental susceptibility
data gives J ¼ ꢀ0:63 cm 1 and g ¼ 2:02 (R ¼
ꢀ
ꢀ
1
0
ꢀ1
ꢀ
5
ꢀ1
cm and J ¼ 0:37 cm , respectively) [8]. The differ-
ences in sign of the magnetic coupling are probably due
to two different geometric features of the carboxylate
bridge [25,26]. The precise structure of 6 is not known,
but it must necessarily involve the syn–syn bridging
carboxylate group.
3
:73 ꢂ 10 ) for 1 and J ¼ ꢀ0:60 cm and g ¼ 2:09
ꢀ
5
(
R ¼ 7:00 ꢂ 10 ) for 3.
The same fitting R value of
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
,
u
u X
t
X
1
exp
i
1
exp
2
ðvi Þ
exp
ðvi
calc
2
R ¼
ꢀ vi
Þ
2
ðv
Þ
i
i
The variable-temperature magnetic susceptibility
data (1.9–300 K) show a broad maximum in v Cu at
ꢄ90 K (Fig. 6) and the room magnetic moment of
leff ¼ 1:62lB is lower than the spin-only of 1:73lB,
was used throughout the paper. This fit is shown in Figs.
and 5. Crystallographic studies of 3 show that the
magnetic behavior may be described not only by the
4

_
B. Zurowska et al. / Inorganica Chimica Acta 357 (2004) 755–763
761
there were no infrared bands of the coordinated car-
boxylate characteristic for 2-pca or 2-qca.
IR spectra of complexes formed in the reaction of
CuCl2 with 2-pmpe recorded on solution, under a ni-
trogen atmosphere, and of CuCl2 with 2-qmpe recorded
on solid sample, in the far-IR region show very strong
bands at 171 and 184 cm 1, respectively. This may
suggest the bridging copper–chloride bond [19,20]. The
X-band EPR spectrum of the first species exhibits only a
single broad absorption at room temperature and at 77
K with g ¼ 2:14, presumably due to the exchange cou-
pling. However, for the second compound, the spectrum
ꢀ
is axial (g ¼ 2:377, g? ¼ 2:094 at room temperature)
k
Fig. 6. Plot of vCu (ꢅ) and vCuT (ꢆ) vs. T for Cu(2-qca)Cl (6).
and characteristic for a mononuclear complex of Cu(II).
The above results may suggest that these reactions lead
to the formation of a polymeric and mononuclear
compounds, respectively.
consistent with antiferromagnetic exchange coupling
between the Cu(II) ions [27,28]. Below 90 K, the sus-
ceptibility vCu (and l ) decreases markedly with de-
As previously observed [2], the reaction of Cu(ClO4)2
with 2-qmpe performed in air leads to the formation of a
stable solid species containing O-bonded 2-qmpe ligand
of the formula [Cu(2-qmpe)2(OH)(H2O)2]2(ClO4)2, for
which the hydroxo-bridged dinuclear structure was
proposed [2]. The same compound was obtained in the
inert atmosphere, as confirmed by IR spectrum. We
have now observed that this reaction in air also yields a
compound with coordinated quinoline-2-carboxylate li-
gand of the formula Cu(2-qca)2 ꢁ H2O (4). Thus, the
above results may suggest that the dihydroxo-bridged
compound of Cu(II) with 2-qmpe is air-stable in the
solution for the time, sufficient to isolate it as precipi-
tate. However when still in solution, it undergoes very
slow decomposition forming the ligand 2-qca.
B
creasing temperature to 30 K (leff ¼ 0:52l ) and then
B
increases rapidly at the lowest temperatures due to the
effect of trace quantities of a mononuclear impurity.
Unfortunately, the magnetic data cannot be satisfacto-
rily interpreted due to the lack of crystallographic data.
3
.5. Aspects of the synthesis
For the purpose of explanation of the unusual be-
havior of 2-pmpe and 2-qmpe ligands towards cop-
per(II) salts, all reactions were performed under a
nitrogen atmosphere, using degassed EtOH or MeOH.
These reactions after evaporation of the solvent lead to
the formation of viscous oily products with an exception
of the reaction of copper(II)acetate monohydrate with
As mentioned in the experimental part, the reaction
in air of Cu(CH3COO)2 ꢁ H2O with 2-pmpe primarily
leads to the formation of green compound of the for-
mula [Cu(CH3COO)2(2-pmpe)] (3a). The infrared spec-
2
-pmpe in which a blue–green product was precipitated.
The attempts to separate the solid compound in the
reaction of CuCl2 with 2-pmpe were unsuccessful.
Therefore the remaining complexes were precipitated
using degassed acetone.
ꢀ
ꢀ
trum shows mas(COO ) and ms(COO ) at 1628 and 1433
cm , respectively. The difference between the two
ꢀ
1
ꢀ
1
In order to obtain information about the interaction
of copper(II) salts with the ligands in nitrogen atmo-
sphere, the obtained species were investigated by IR
spectroscopy. In all cases, the characteristic bands of
coordinated 2-pmpe or 2-qmpe ligands were observed.
This is consistent with our earlier observation of IR
spectral features of these ligands [1,2]. These spectra
indicate that 2-pmpe and 2-qmpe ligands, containing
two potential donor atoms (the quinoline or pyridine
nitrogen, respectively and the phosphoryl oxygen, act as
O-bonded monodentate ligands, with an exception of
the product of reaction of copper(II)acetate monohy-
drate with 2-pmpe (N-bonded). In the far-IR region
bands of 195 cm indicates bridging didentate coordi-
nation [14]. The d–d spectrum shows a broad band at
ꢀ
1
ꢀ1
14 250 cm
and approximately at 26 400 cm , ob-
served as a shoulder on an intense charge transfer band.
Both bands are in the usual range for copper(II) com-
pounds in a square-pyramidal CuO4O environment.
[29,30]. Thus, spectroscopic properties of this compound
indicate the presence of dinuclear unit structurally sim-
ilar to those in copper(II)acetate monohydrate [31]. The
identical spectroscopic properties suggest the dinuclear
structure of the isolated compound for the above reac-
tion under inert atmosphere.
All the complexes of Cu(II) with 2-pmpe and 2-qmpe
ligands isolated in solid state are stable in air and in
solution in inert atmosphere (N2). In solution after ex-
position to air, they undergo slow decomposition.
Complexes of Cu(II) with 2-pca and 2-qca, respectively,
are formed, as was unambiguously indicated by IR
ꢀ
1
m(Cu–N) band at 262 cm is observed. The behavior of
-pmpe in the reaction with CuCl2 with is unusual. A
2
contribution of the pyridine nitrogen atom in coordi-
nation was observed in perchlorate and nitrate com-
plexes with 2-pmpe [1]. For all the products of reactions

_
B. Zurowska et al. / Inorganica Chimica Acta 357 (2004) 755–763
762
spectra. However, the isolation of the remaining organic
product of decomposition reaction in the chemically
pure form and its unambiguous identification appeared
to be difficult. Nevertheless, on the basis of the IR
spectrum and its comparison with the known spectrum
ligands and catalyze oxidation by atmospheric oxygen,
while other divalent transition metal ions do not exhibit
this activity under comparable experimental conditions
[1,2]. The processes of decomposition of primarily
formed compounds of Cu(II) with 2-pmpe and 2-qmpe
strongly depend on the nature of the ligand. The cop-
per(II) coordination compounds with monodentate co-
ordinating ligands undergo decomposition, however,
with N,O-bonded (chelating or bridging) ligands 2-pmpe
and 4-pmpe are stable in solution [1]. These complexes
are probably stabilized by didentate coordination of
the ligands to Cu(II) ion. Moreover, the 2-pca ion in the
reaction of NiCl2 with 2-pmpe (N,O-bonded) in the
presence of Cu(II) ion is not formed. In other words,
of ClP(O)(OEt) and literature data [32] we can con-
2
clude that the cleavage of the phosphonic group yields a
white solid species of the formula OHP(O)(OEt) . In
2
fact, the IR spectrum shows the characteristic bands at
035 and 1240 cm 1 which can be assigned to m(P–OC)
ꢀ
1
and m(PO), respectively. The broad bands observed in
the 2350–2080 cm and 2700–2600 cm regions can be
ꢀ
1
1
attributed to the P–OH group. The bands at 2986 and
ꢀ
1
2
900 cm indicate the presence of aliphatic CH groups.
In order to determine the role of copper(II) in decom-
these complexes do not react with O probably because
2
position reaction, the reaction of NiCl2 (90%) with
-qmpe in the presence of CuCl2 (10%) (in molar ratio
M(II): 2-qmpe ¼ 1:2) was investigated. From this reaction
in an ethanol solution, a pale green product was precipi-
tated using degassed acetone with yield ca. 86%. The
obtained compound was identified as octahedral com-
of steric and/or electronic reasons. It is interesting to
note that the rapidity of these decomposition reactions
may be associated with mono-, di- or polymeric struc-
tures of Cu(II) compounds with 2-pmpe and 2-qmpe
ligands. The mononuclear complex of CuCl2with 2-
qmpe decomposed rapidly.
In summary, we have presented a novel oxidative P-
dealkylation reaction at room temperature with oxidative
cleavage of starting ligands and the formation of 2-pca
and 2-qca ligands, respectively. The final products of this
process are complexes of Cu(II) with 2-pca or 2-qca, as
was established by spectroscopic studies and X-ray crystal
structure analysis of 3 (in this paper) and 2 [7,8].
2
3
pound of Ni(II) by its d–d spectrum [29.0, 15.5, 9.7 (ꢂ10
ꢀ
1
cm )] [33]. The IR spectrum clearly demonstrates the
presence of coordinated quinoline-2-carboxylate ligand.
The compound exhibits the bands at 1638 and 1400
cm 1with a shoulder at 1376sh cm , corresponding to
ꢀ
ꢀ1
ꢀ
ꢀ
mas(COO ) and ms(COO ), respectively. However, the
m(PO) band was not observed. It is interesting, that a non-
coordinating behavior of the 2-qmpe ligand with NiCl2
was observed. IR spectra of the isolated compounds are
The novel oxidative reaction with the aid of cop-
per(II) discovered is an important issue in contemporary
research of copper coordination chemistry and the in-
teraction of molecular oxygen with copper(II) species
and will be the subject of further studies.
characteristic for free NiCl and 2-qmpe. Steric as well as
2
electronic effects can explain this. The presented results
suggest that Cu(II) ion promotes and catalyzes conver-
sion of 2-qmpe ligand into quinoline-2-carboxylate acid
(
2-qcaH) under atmosphere dioxygen at room tempera-
4
. Supplementary material
ture. It has to be stressed, that in contrary to the above
presented reaction, the interaction of NiCl2 with 2-pmpe
in presence of a small amount of CuCl did not cause
Crystallographic data for the structure reported in
2
this paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publi-
cation no. CCDC-209000.
Copies of this data may be obtained free of charge
from The Director, CCDC, 12 Union Road, Cambridge
CB2 1EZ, UK (fax: +44-1223-336-033; e-mail: deposit@
ccdc.cam.ac.uk or www:http://www.ccdc.cam.ac.uk).
conversion of 2-pmpe to 2-pca ligand. This reaction gives
rise to the compound of Ni(II) with N,O-bonded 2-pmpe
ligand as indicated by IR spectrum (m(PO) ¼ 1210 cm 1,
ꢀ
ꢀ
1
m(Cu–N) ¼ 260 cm ) [1].
To better understand the role of the solvent, all re-
actions were carried out either in protic (methanol,
ethanol and water) or in aprotic solvents (acetonitrile,
acetone and dichloromethane). Decomposition reac-
tions with oxidative cleavage of ligands leading to the
formation of 2-pca and 2-qca ligands occurred in these
solvents. Additional qualitative studies of O2 uptake (1
atm) revealed that dioxygen was consumed (details not
given in this paper).
Acknowledgements
The study was supported by the Polish State Com-
mittee for Scientific Research, Grant No. 4 T09 A115 23
and ESF/PESC Programme: Molecular Magnets (MM)
and by grant No. 502-13-849 of the Medical University
of Ł oꢀ d ꢀz . The authors thank M.Sc. U. Kalinowska
3
.6. Summary and conclusions
(
Ł oꢀ d ꢀz ) for skillful help in the synthesis of 2-pmpe and 2-
The results of the studies clearly demonstrate that
copper(II) ions promote/enhance decomposition of the
qmpe and Dr. J. Sobczak (Wrocław) for his help in some
experimental problems.

_
B. Zurowska et al. / Inorganica Chimica Acta 357 (2004) 755–763
763
References
[16] K. Nakamoto, J. Fujita, S. Tanaka, M. Kobayashi, J. Am. Chem.
Soc. 79 (1957) 4904.
[17] K. Nakamoto, M. Margoshes, R.E. Rundle, J. Am. Chem. Soc. 77
_
[
1] J. Ochocki, K. Kostka, B. Zurowska, J. Mrozi nꢀ ski, E. Gałdecka,
Z. Gałdecki, J. Reedijk, J. Chem. Soc., Dalton Trans. (1992)
955.
(
1985) 6480.
[
[
[
18] A.W. Herlinger, S.L. Wenhold, T.V. Long, II, J. Am. Chem. Soc.
92 (1970) 6474.
2
_
[
[
2] J. Ochocki, B. Zurowska, J. Mrozi nꢀ ski, H. Kooijman, A.L. Spek,
J. Reedijk, Eur. J. Inorg. Chem. (1998) 169.
19] R.J.H. Clark, in: V. Guttman (Ed.), Halogen Chemistry, vol. 3,
Academic Press, New York, 1967.
3] The IUPAC-recommended names for diethyl 2-pyr-
idylmethylphosphonate and diethyl 2-quinolylmethylphosphonate
are diethyl pyridyl-2-ylmethylphosphonate and diethyl quinolyl-2-
ylmethylphosphonate, respectively..
20] J. Reedijk, A.R. Siedle, R.A. Velapoldi, J.A.M. vanHest, Inorg.
Chim. Acta 74 (1983) 109, and references cited therein.
21] B.J. Hathaway, D. Billing, Coord. Chem. Rev. 5 (1970) 143.
22] J.W. Stout, R.C. Chisholm, J. Chem. Phys. 36 (1962) 979.
23] B.C. Gerstein, F.D. Gehring, R.D. Willett, J. Appl. Phys. 43
[
[
[
[
[
4] F.S. Stephens, J. Chem. Soc. A (1970) 2377.
5] A. Takenaka, H. Utsumi, T. Yamamoto, A. Furusaki, I. Nitta,
J. Chem. Soc. Jpn., Pure Chem. 91 (1970) 928.
(
1972) 1932.
[
[
24] J.S. Smart, Effective Field Theories of Magnetism, Saunders,
Philadelphia and London, 1966, p. 24.
[
[
6] R. Faure, H. Loiseleur, G. Thomas-David, Acta Crystallogr.,
Sect. B 29 (1973) 1890.
25] E. Colacio, J.P. Costes, R. Kivekas, J.P. Laurent, J. Ruiz, Inorg.
Chem. 29 (1990) 4240.
[26] R.J. Doedens, Prog. Inorg. Chem. 21 (1976) 209.
7] M.A.S. Goher, A.K. Hafez, M.A.M. Abu-Youssef, A. Popitsch,
H.P. Fritzer, F.A. Mautner, Monatsh. Chem. 125 (1994) 833.
_
8] B. Zurowska, J. Mrozi nꢀ ski, Inorg. Chim. Acta 342 (2003) 23.
[
[
[
27] B. Bleaney, K.D. Bowers, Proc. R. Soc. Lond., Ser. A 214 (1952)
9] E. K o€ nig, Magnetic Properties of Coordination and Organome-
tallic Transition Metal Compounds, Springer, Berlin, 1966.
4
28] B.N Figgis, R.L. Martin, J. Chem. Soc. (1956) 3837.
51.
[
[
[
10] G.M. Sheldrick, SHELXS97, Program for Solution of Crystal
Structures, University of G o€ ttingen, G o€ ttingen, 1997.
29] B.J. Hathaway, in: G. Wilkinson, R.D. Gillard, J.A. McCleverty
(Eds.), Comprehensive Coordination Chemistry, vol. 5, Perg-
amon, Oxford, 1987, p. 652.
[
11] G.M. Sheldrick, SHELXL97, Program for Crystal Structure
Refinement, University of G o€ ttingen, G o€ ttingen, 1997.
[
[
[
30] D.P. Graddon, J. Inorg. Nucl. Chem. 17 (1969) 222.
31] R. Tsuchida, S. Yamada, Nature 176 (1955) 1171.
32] L.J. Bellamy, The Infrared Spectra of Complex Molecules,
Chapman & Hall, London, 1975.
[
[
[
[
12] K. Kostka, J. Ochocki, Chem. Anal. (Warsaw) 28 (1983) 717.
13] K. Kostka, J. Ochocki, Pol. J. Chem. 61 (1987) 177.
14] G.B. Deacon, R.J. Phillips, Coord. Chem. Rev. 33 (1980) 227.
15] K. Nakamoto, Infrared in Raman Spectra of Inorganic and
Coordination Compounds, fourth ed., Wiley, New York,
[
33] A.B.P. Lever, Inorganic Electronic Spectroscopy, Elsevier, Am-
sterdam, 1986.
1986.