Absence of a magnetic interaction in a dinuclear copper complex? The case of a crossed axial-equatorial oxalate coordination mode

Article abstract of DOI:10.1016/j.poly.2013.06.054

A simple and easy preparation of the hydrogen oxalate ligand (HOOCCOO), with only one deprotonated carboxylic function, is described. ¹³C NMR spectroscopy furnishes a straightforward characterization of this interesting ligand. This ligand coordinates in its monodeprotonated form to a Cu ^II complex with a free coordination position, yielding a complex that presents a very weak antiferromagnetic interaction. In a subsequent step, the second deprotonation allows the isolation of an oxalato-bridged dinuclear Cu^II complex, which is characterized by structural determination, the copper ions being pentacoordinate in an equivalent centrosymmetry-related environment. DFT calculations show that the magnetic interaction through the oxalato bridge is strictly equal to 0 cm¹, demonstrating that the singlet and triplet states of this oxalato-bridged dinuclear Cu^II complex have the same energy. This dinuclear complex presents a rhombic EPR spectrum, with three well-defined g values, at room temperature as well as at 100 K.

Full text of DOI:10.1016/j.poly.2013.06.054

Polyhedron 63 (2013) 127–132
Contents lists available at SciVerse ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Absence of a magnetic interaction in a dinuclear copper complex?
The case of a crossed axial–equatorial oxalate coordination mode
a
b,c,
b,c
Antonio J. Mota a, , Enrique Colacio , Jean-Pierre Costes
, Françoise Dahan
⇑
⇑
a Departamento de Quimica Inorganica, Facultad de Ciencias, Universidad de Granada, Av. Fuentenueva s/n, 18002-Granada, Spain
b Laboratoire de Chimie de Coordination du CNRS, 205, route de Narbonne, BP 44099, F-31077 Toulouse Cedex 4, France
c Université de Toulouse;UPS, INPT, F-31077 Toulouse Cedex 4, France
a r t i c l e i n f o
a b s t r a c t
A simple and easy preparation of the hydrogen oxalate ligand (HOOCCOO)ꢀ, with only one deprotonated
carboxylic function, is described. 13C NMR spectroscopy furnishes a straightforward characterization of
this interesting ligand. This ligand coordinates in its monodeprotonated form to a CuII complex with a
free coordination position, yielding a complex that presents a very weak antiferromagnetic interaction.
In a subsequent step, the second deprotonation allows the isolation of an oxalato-bridged dinuclear CuII
complex, which is characterized by structural determination, the copper ions being pentacoordinate in an
equivalent centrosymmetry-related environment. DFT calculations show that the magnetic interaction
through the oxalato bridge is strictly equal to 0 cmꢀ1, demonstrating that the singlet and triplet states
of this oxalato-bridged dinuclear CuII complex have the same energy. This dinuclear complex presents
a rhombic EPR spectrum, with three well-defined g values, at room temperature as well as at 100 K.
Ó 2013 Elsevier Ltd. All rights reserved.
Article history:
Received 19 June 2013
Accepted 28 June 2013
Available online 16 July 2013
Keywords:
Oxalic acid
Copper complexes
Structural determination
Magnetism
DFT calculations
1. Introduction
2. Experimental
Oxalic acid has been used so extensively for the preparation of
di or polynuclear complexes of 3d ions that it seems impossible to
give an exhaustive list of references dealing with that subject [1–
5]. Indeed the resulting deprotonated oxalate ligand is a good che-
lating agent of 3d ions and also a good transmitter of magnetic
interactions. The different topologies obtained for the coordination
geometries of oxalate with transition metal ions and lanthanides
have been reported [6]. At variance, the hydrogen oxalate ligand,
in which only one carboxylic function is deprotonated, has only
been reported in two papers dealing with the structural determi-
nation of CrIII and CuII compounds [7,8]. In both cases, the ligand
was prepared by serendipity, protonation of the starting oxalato li-
gand or complexation after degradation of squaric acid. However,
this ligand can be isolated in large amounts and then used in the
synthesis of a copper complex without further deprotonation.
Afterwards, deprotonation of this entity yields a dinuclear oxala-
to-bridged copper compound. In this contribution, we focus on
the formation of this interesting ligand, its characterization in solu-
tion by 13C NMR spectroscopy and its use as a ligand towards CuII
ions.
2.1. Materials
Oxalic acid, piperidine and LiOHꢁH2O (Aldrich) were used as
purchased. High-grade solvents (acetone, diethyl ether and meth-
anol) were used for the syntheses of the ligands and the complexes.
The copper complex [9], bis(l-acetato)bis(7-amino-4-methyl-5-
aza-3-hepten-2-onato(1ꢀ))dicopper(II), [(LCu(CH3COO)]2, and the
dideprotonated oxalate (pipH)2+(Ox)2ꢀ [10] were prepared as de-
scribed previously.
2.2. Ligand
(pipH)+(HOOCCOO)ꢀ. To oxalic acid (3 g, 33.3 mmol) dissolved in
acetone (100 mL) was added an acetone solution (20 mL) contain-
ing piperidine (2.83 g, 33.3 mmol). The white precipitate which ap-
peared immediately, was stirred at room temperature for 30 min
and filtered off, washed with acetone and diethyl ether, and then
dried. Yield: 5 g (86%). Anal. Calc. for C7H13NO4 (175.0): C, 48.0;
H, 7.5; N, 8.0. Found: C, 47.8; H, 7.8; N, 8.2%. 13C NMR (DMSO) d
(ppm): 168.1 (HOOCCOO), 46.9 (N–Cortho), 24.6 (Cmeta), 23.8 (Cpara).
2.3. Complexes
⇑
Corresponding authors. Address: Laboratoire de Chimie de Coordination du
CNRS, 205, route de Narbonne, BP 44099, F-31077 Toulouse Cedex 4, France. Tel.:
+33 561333154; fax: +33 561553003 (J.P. Costes).
LCu(HOOCCOO)(H2O) 1. To [(LCu(CH3COO)]2 (0.5 g, 1.9 mmol),
E-mail address: jean-pierre.costes@lcc-toulouse.fr (J.-P. Costes).
dissolved in distilled water (15 mL) was added (pipH)+(OxH)ꢀ
0277-5387/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.poly.2013.06.054

128
A.J. Mota et al. / Polyhedron 63 (2013) 127–132
(0.33 g, 1.9 mmol) as a solid. A few minutes later a blue precipitate
appeared in the blue solution, which was stirred for 30 min at
room temperature. The precipitate was filtered off by suction,
washed with acetone and diethyl ether, and then dried. Yield:
0.2 g (34%). Anal. Calc. for C9H16CuN2O6 (311.8): C, 34.7; H, 5.2;
N, 9.0. Found: C, 34.3; H, 5.0; N, 8.9%. IR (ATR, cmꢀ1): 3379w,
3315m, 3238m, 3153m, 2951w, 1661m, 1616m, 1586s, 1509s,
1461m, 1443w, 1403s, 1355w, 1337m, 1310m, 1275m, 1218w,
1136m, 1092m, 1073w, 1015m, 947w, 790m, 753m, 700w, 647w.
[LCu(Ox)CuL]ꢁ2H2O 2. As the yield of the previous reaction is
rather low, the filtrate was kept and LiOHꢁH2O (0.04 g, 1 mmol)
was added to the filtrate. The resulting solution was kept undis-
turbed for three days, yielding blue crystals. Yield: 0.15 g (30%).
Anal. Calc. for C16H30Cu2N4O8 (533.5): C, 36.0; H, 5.7; N, 10.5.
Found: C, 35.8; H, 5.4; N, 10.2%. IR (ATR, cmꢀ1): 3451m, 3393m,
3324m, 3266m, 2929w, 1663s, 1596m, 1586m, 1504s, 1462m,
1437m, 1414s, 1396m, 1339m, 1314w, 1281w, 1261w, 1216w,
1137w, 1123w, 1087m, 1010m, 941m, 915w, 796m, 741m, 668w.
This compound was also prepared in a more classical way by
the addition of (pipH)2+(Ox)2ꢀ (0.26 g, 1 mmol) to a water solution
of [(LCu(CH3COO)]2 (0.26 g, 1 mmol). The precipitate that appeared
quickly in 80% yield gave the same infrared and analytical data.
and diamagnetic corrections were applied by using Pascal’s con-
stants [17]. Isothermal magnetization measurements were per-
formed up to 5 T at 2 K. The magnetic susceptibilities have been
computed by exact calculations of the energy levels associated
with the spin Hamiltonian through diagonalization of the full ma-
trix with a general program for axial symmetry [18], and with the
MAGPACK program package [19] in the case of magnetization. Least-
squares fittings were accomplished with an adapted version of the
function-minimization program MINUIT [20].
2.6. Computational details
All theoretical calculations were carried out at the density func-
tional theory (DFT) level using the hybrid B3LYP exchange-correla-
tion functional [21], as implemented in the GAUSSIAN 09 program
[22]. A quadratic convergence method was employed in the self-
consistent-field process [23]. The triple-f quality basis set pro-
posed by Ahlrichs and co-workers has been used for all atoms
[24]. Calculations were performed on complexes built from exper-
imental geometries as well as on model complexes. The electronic
configurations used as starting points were created using Jaguar
7.9 software [25].
2.4. Crystal structure
3. Results
Crystals of 2 suitable for X-ray analyses were obtained by slow
evaporation of the corresponding water solution. The selected
crystal of 2 (yellow parallelepiped, 0.50 ꢂ 0.20 ꢂ 0.10 mm3) was
mounted on an Enraf-Nonius CAD4 diffractometer using graph-
Addition of piperidine to a DMF solution of oxalic acid yields a
white precipitate that corresponds to the dideprotonated oxalate
salt along with two piperidinium cations. When carried out in ace-
tone, the same reaction yields only the monodeprotonated oxalic
acid with a piperidinium countercation. This monodeprotonation
is clearly shown by 13C NMR spectroscopy. Oxalic acid gives a sin-
gle signal at 160.9 ppm for the COOH carbon atom, while the
deprotonated COOꢀ of the (pipH)2+(Ox)2ꢀ species appears at
175.7 ppm. In our (pipH)+(HOOCCOO)ꢀ compound, we could ex-
pect two signals corresponding to the COOH and COOꢀ carbon
atoms, but we observed a unique signal at 168.1 ppm. This is due
to the presence of a fast proton exchange between the carboxylate
and carboxylic functions, with an intermediately located signal, in
between those corresponding to the non-deprotonated and dide-
protonated forms. However, the signals of the piperidinium cations
are quite similar in the two different species, (pipH)+(HOOCCOO)ꢀ
and (pipH)2+(Ox)2ꢀ. Three signals are observed at 46.9, 24.6 and
23.8 ppm in the two samples, for the carbon atoms in ortho, meta
and para positions of the nitrogen atom, respectively. (pipH)+(-
HOOCCOO)ꢀ reacts with [(LCu(CH3COO)]2 in a 1:1 ratio to give
the neutral LCu(HOOCCOO)(H2O) blue complex for which we have
not been able to isolate crystals suitable for X-ray structure deter-
mination. Addition of a base, lithium hydroxide, to the blue filtrate
of the reaction yielded nice blue crystals corresponding to a dinu-
clear copper complex in which the two copper ions are linked by
the dideprotonated oxalate ligand. Infrared spectra of complexes
1 and 2 differ in the 3500–3000 and 1670–1560 cmꢀ1 regions. In
the first region, there is an additional stretching OH vibration,
not easy to assign among the bands originating from the stretching
NH2 and water vibrations. In the other area, a band at 1663 cmꢀ1
and two bands of equal intensity appear at 1596 and 1586 cmꢀ1
ite-monochromated Mo K
collected at 293 K up to 27° in the
with the MOLEN package [11]. Absorption corrections [12] from
a
radiation (k = 0.71073 Å). Data were
x
ꢀ2h scan mode and reduced
w
scans were applied (Tmin ꢀ max = 0.7831 ꢀ 0.9972).
The structure was solved using SHELXS-97 [13] and refined on F2
by full-matrix least-squares using SHELXL-97 [14] with anisotropic
displacement parameters for all non-hydrogen atoms. H atoms
were introduced in the calculations using the riding model. Isotro-
pic UH factors were 1.1 times higher than those of the atom to
which they were bonded. The atomic scattering factors and
anomalous dispersion terms were taken from a standard compila-
tion [15]. The maximum and minimum peaks on the final differ-
ence Fourier map were 0.633 and ꢀ0.310 e Åꢀ3 for 2. A drawing
of the molecule was performed with the program ZORTEP [16]. Crys-
tal data collection and refinement parameters are given below, and
selected bond distances and angles are gathered in the figure
caption.
2.4.1. Crystal data for 2
C16H30Cu2N4O8, M = 533.52, monoclinic, P21/c (N°14), Z = 2,
a = 14.499(2), b = 6.0623(16), c = 12.8240(18) Å, b = 91.899(12)°,
V = 1126.6(4) Å3, 2554 collected reflections, 2455 unique reflec-
tions (Rint = 0.0214), R = 0.0336, Rw = 0.0623 for 1788 contributing
reflections [I > 2r(I)].
2.5. Physical measurements
C, H and N elemental analyses were carried out at the Labora-
toire de Chimie de Coordination Microanalytical Laboratory in Tou-
louse, France. IR spectra were recorded with a Perkin-Elmer
Spectrum 100FTIR using the ATR mode. 1D 13C spectra using 1H
broadband decoupling {1H}13C were performed with a Bruker
WM250 apparatus working at 62.89 MHz using (CD3)2SO as sol-
vent. Chemical shifts are given in ppm versus TMS. Magnetic data
were obtained with a Quantum Design MPMS SQUID susceptome-
ter. Magnetic susceptibility measurements were performed in the
2–300 K temperature range under a 0.1 T applied magnetic field,
for complex 2, while two intense bands at 1661 and 1586 cmꢀ1
,
along with a band of lower intensity at 1616 cmꢀ1 are present
for complex 1. The bands around 1600 cmꢀ1 are attributable to
the C@O and C@N stretching vibrations originating from the li-
gand, by comparison with other LCuX complexes (X = N3, SCN)
[26] and the band at 1663–1661 cmꢀ1 corresponds to the masym
COO vibration of the deprotonated oxalato function. We have then
concluded that the band at 1586 cmꢀ1 originates from the carbox-
ylic acid of the hydrogen oxalato HOOCCOOꢀ ligand.

A.J. Mota et al. / Polyhedron 63 (2013) 127–132
129
Solution visible spectra (DMSO) of complexes 1 and 2 are char-
acterized by very similar d–d transitions at 640 nm for 1 and
636 nm for 2. These spectra are consistent with at least a pentaco-
ordination of the Cu ions.
3.1. Structural determination
A view of the dinuclear complex 2 is shown in Fig. 1 while se-
lected bond lengths and angles are listed in the figure caption.
Two LCu cationic units are linked by the deprotonated oxalate, thus
yielding a neutral dinuclear complex. There is an inversion center
located in the middle of the C–C oxalate bond. The copper ion is
pentacoordinate, in a distorted (4 + 1) geometry. The four equato-
rial bond lengths span the range 1.923(2)–2.001(2) Å; they involve
the three Cu–O and the Cu–N bonds to the ligand and one of the
Cu–O oxalato bonds. We note the large difference between the
two Cu–O oxalate bond lengths, 2.001(2) for Cu–O2 and
2.287(2) Å for Cu–O3. The plane defined by the three ligand donor
atoms and the copper ion makes an angle of 86.9(1)° with the oxa-
late plane, and the two planes containing the three ligand donor
atoms and the copper ion are strictly parallel, the distance between
the two planes being equal to the Cuꢁ ꢁ ꢁCu distance of 5.541(3) Å.
The non-coordinated water molecules are involved in hydrogen
bonds between two symmetry-related oxalato oxygen atoms and
the keto oxygen atoms of the tridentate ligand, which results in
the formation of 1D chains that are linked together through other
hydrogen bonds involving the other two symmetry-related oxalato
oxygen atoms and the coordinated primary amine functions, thus
yielding a 2D lattice.
Fig. 2. Temperature dependence of the
corresponds to the best data fit, see text.
v
MT product for complex 2. The solid line
slow decrease of v
MT down to 2 K (0.38 cm3 molꢀ1 K). This behav-
ior indicates that a weak antiferromagnetic interaction operates at
low temperature. A qualitative analysis has been performed with a
simple isotropic Hamiltonian, H = ꢀ2J(SCu1.SCu2), while intermolec-
ular interactions through the hydrogen bonding network were ta-
ken into account by the mean-field Hamiltonian [27]. The resulting
interaction parameter is very weak, JCuCu = ꢀ 1.15 cmꢀ1, zj = ꢀ
0.22 cmꢀ1
,
with g = 2.13 and
a
nice agreement factor
P
2
R ¼ ½ð
v
MTÞobs ꢀ ð
v
MTÞcalcꢃ equal to 6 ꢂ 10ꢀ5. As the structure of
complex 1 is unknown, it was first analyzed as a mononuclear
complex, but the
vMT decrease at low temperature confirms pres-
3.2. Magnetic properties
ence of a higher nuclearity. By analogy with the starting dinuclear
acetate complex, a dinuclear model was been investigated. The
The magnetic susceptibility of complexes 1 and 2 has been mea-
sured in the 2–300 K temperature range under an applied magnetic
corresponding
v
MT product is constant (0.75 cm3 molꢀ1 K) from
300 to 50 K, and then decreases to 0.03 cm3 molꢀ1 K at 2 K
(Fig. 3). This larger decrease means that a slightly larger magnetic
interaction occurs in this expected dinuclear complex. Applying
field of 0.1 T. The thermal variation of the
played in Fig. 2, M being the molar magnetic susceptibility of the
dinuclear species corrected for the diamagnetism of the ligands.
From 300 to 15 K,
MT is equal to 0.81 cm3 molꢀ1 K, which corre-
vMT product for 2 is dis-
v
the above Hamiltonian yields JCuCu = ꢀ 3.9 cmꢀ1, zj = ꢀ 0.68 cmꢀ1
,
v
with g = 2.11 and R = 2 ꢂ 10ꢀ5. It should be noted that a fitting of
these data with a chain model does not give satisfactory results.
EPR spectra are interesting because they allow a straightfor-
ward characterization of complexes 1 and 2. In the solid state, a
spectrum of axial symmetry, with gpar = 2.225 and gper = 2.084, is
associated with complex 1 (Fig. 4). A very weak half-field transition
can be shown with a 1600 amplification gain, in agreement with
the presence of weak Cu–Cu magnetic interactions (Fig. S3). A
sponds to the value expected for two uncoupled copper ions with
g = 2 (0.75 cm3 molꢀ1 K). Lowering the temperature results in a
Fig. 1. Molecular structure of the oxalate-bridged dinuclear [LCu(Ox)CuL].2H2O
complex 2 with the partial atom numbering scheme. H atoms, except those of water
molecules and primary amine functions, have been omitted for clarity. Selected
bond lengths [Å] and angles [°]: Cu–O1 1.923(2), Cu–O2 2.0005(18), Cu–O3
2.2869(18), Cu–N1 1.940(2), Cu–N2 1.983(3), O1–Cu-N1 93.86(9), N1–Cu–N2
85.10(10), O2–Cu–O3 77.78(6), O1–Cu–O3 91.30(8).
Fig. 3. Temperature dependence of the
corresponds to the best data fit, see text.
vMT product for complex 1. The solid line

130
A.J. Mota et al. / Polyhedron 63 (2013) 127–132
produces, if any, negligible magnetic interactions regardless of
the structural parameters of the bridge [28]. This is because the
magnetic orbital on the CuII is located in the equatorial plane
(the dx2ꢀy2 magnetic orbital is directed toward the nitrogen and
oxygen atoms of the L ligand on each copper moiety) and, there-
fore, the spin density on its axial position is expected to be almost
zero, as it appears in the spin density map depicted in Fig. 6. We
have also collected some representative spin density values in
Table 1.
2 600
3 000
3 400
3 800
Although the crystal structure of complex 1 is still unknown,
DFT calculations have been carried out, starting with the model
crystal structure of the [LCu(CH3COO)]2 complex previously pub-
lished [9], in which the mono-O-acetato bridging ligand is replaced
by a monodeprotonated oxalato ligand acting in the same bridging
mode. We have indeed calculated two different structures called
1a and 1b, in which all the atoms (including the oxygen bridging
atom) have the same positions as found in the crystal structure, ex-
cept for the C2O3H fragment which should be properly accommo-
dated. The structure 1a has the same CuObridgedC angle as the
acetato ligand in the parent compound, whereas 1b has been mod-
ified by an in plane displacement of the oxalato moieties by 10°
away from the corresponding neighboring ligands in order to avoid
some steric hindrance occurring between the oxalato group and
the close ligand, see Fig. 7. For these structures, we have found cou-
pling constant (J) values of ꢀ0.02 and +0.09 cmꢀ1, respectively,
confirming the small exchange coupling exhibited by these
systems.
Fig. 4. X-band EPR spectrum of complex 1.
2 800
3 000
3 200
3 400
3 600
4. Discussion
Potassium and sodium oxalate ligands have been used for a long
time to yield oxalato bridged transition metal complexes. There are
so many complexes that we cannot furnish an exhaustive list. Or-
ganic bases such as piperidine have been used as countercations to
provide more soluble oxalate salts [29]. Following this route, it is
possible to isolate the monodeprotonated form of oxalic acid
HOOCCOOꢀ in large amounts, by using acetone as the solvent.
The resulting hydrogen oxalate ligand is nicely characterized by
its 13C NMR spectrum. This ligand can also be linked to a copper
complex in this form, without deprotonation of the second carbox-
ylic function, using distilled water as the solvent. In methanol, a
mixture of complexes 1 and 2 was obtained, complex 2 being the
most abundant. Unfortunately, the structure of compound 1 re-
mains unknown. This ancillary COOHCOOꢀ ligand can coordinate
to the copper ion as a chelate or only through the deprotonated
oxygen atom, as does the acetato ligand in a complex involving
the same LCu unit in the [LCu(CH3COO)]2 dinuclear complex [9].
These two possibilities result at least in a pentacoordination of
Fig. 5. X-band EPR spectrum of complex 2.
rhombic spectrum with three g values, g1 = 2.166, g2 = 2.120,
g3 = 2.057, is seen for complex 2 (Fig. 5). This observation along
with the absence of half-field transition should be in relation to
the weak magnetic interaction found for this complex.
In order to confirm the very weak magnetic coupling observed
in complex 2, we have performed DFT calculations following the
Broken-Symmetry (BS) approach from the model crystal structure
described above. The exactly zero value found for the exchange
coupling constant (J) means an equal energy value of the two pos-
sible magnetic states, namely the parallel-spin (triplet) and the
opposite-spin (singlet) states. This behavior should be ascribed to
the relative orientation of the bridge between neighboring CuII
ions, suggesting the existence of a unique operative magnetic path
connecting the equatorial and axial positions, thus leading to
non-coupled magnetic centers. This exchange pathway always
Fig. 6. Calculated spin densities for complex 2 in the triplet (left) and singlet (right) states. The represented isodensity surfaces correspond to a cut-off value of
0.0015 e Bohrꢀ3. Grey and blue colors correspond to positive and negative values, respectively. (Colour online).

A.J. Mota et al. / Polyhedron 63 (2013) 127–132
131
Table 1
of the C–C oxalato bond constrains the two copper magnetic orbi-
tals to be in a parallel orientation, which results in the perfect ab-
sence of a magnetic interaction, as estimated in a previous work
[2]. The observation of a nice rhombic EPR spectrum without any
half-field transition does confirm that there is no magnetic interac-
tion between the two equivalent CuII ions. It has also been shown
by DFT calculations that the triplet and singlet states have the
same energy, in agreement with the presence of non-coupled mag-
netic CuII centers.
Spin density values (in eꢀ) on selected atoms for complex 2.
Atomsa
Parallel-spin
(triplet) state
Opposite-spin
(singlet) state
Cu1/Cu2
N1amine/N2amine
N1imine/N2imine
O1enolate/O2enolate
O1oxalate-eq/O2oxalate-eq
O1oxalate-ax/O2oxalate-ax
+0.6219/+0.6261
+0.0945/+0.0908
+0.1076/+0.1065
+0.1013/+0.0929
+0.0724/+0.0792
+0.0069/+0.0056
ꢀ0.6216/+0.6258
ꢀ0.0943/+0.0907
ꢀ0.1075/+0.1065
ꢀ0.1012/+0.0929
ꢀ0.0707/+0.0783
+0.0016/ꢀ0.0001
a
Indexes 1 and 2 refers to the Cu1 and Cu2 moieties, respectively.
5. Conclusion
the CuII ion, in agreement with the d-d transition at 640 nm. In a
recent work [8], the structural determination of a copper complex
of this hydrogen oxalate ligand, originating from the decomposi-
tion of squaric acid, indicates that COOHCOOꢀ is chelated through
two oxygen atoms to the CuII ion and that the outer oxygen atom is
able to enter the coordination sphere of the neighbouring CuII ion
in the apical position to yield a 1D chain compound. Such a coordi-
nation in our case would yield again a 1D chain, with a hexacoor-
dinated CuII ion, which it is not in accordance with the observation
of a visible spectrum similar to that of complex 2, along with the
impossibility to obtain an acceptable fit of the magnetic data to a
1D chain model.
Adding a base to a solution of complex 1 allows complex 2 to be
generated, in which the oxalato dianion bridges two cationic LCu
units, thus yielding a neutral compound. The structural determina-
tion shows that the copper ions are in a pentacoordinated (4 + 1)
coordination mode, with four short bond lengths and a fifth longer
bond. The SHAPE program [30] allows the confirmation that the CuII
ion can be equally considered in a vacant octahedron (SOh = 2.76)
or a square pyramidal environment (SSP = 2.66). Importantly, the
mean coordination planes involving the donor atoms of the triden-
tate ligand and the copper ions are strictly parallel, separated by a
distance of 2.019(2) Å. Indeed, when two CuII ions are chelated to
an oxalato ligand, the two copper magnetic orbitals can be copla-
nar, perpendicular, parallel or in a trigonal-bipyramidal topology
[31]. These four orientations have been experimentally found. In
the parallel case, the magnetic interaction is weak, varying from
ꢀ37 to +1.2 cmꢀ1. In a more recent example, a ferromagnetic inter-
action of +5.4 cmꢀ1 has been reported [32]. These examples involve
mainly complexes made with polyamine ligands. In comparison
with these polyamine ligands, a partial electronic delocalization
occurs in the tridentate ligand coordinated to copper in complex
2, thus defining a planar equatorial plane around the copper ion.
The presence of a crystallographic inversion center in the middle
We have described a synthetic pathway able to yield large
amounts of the hydrogen oxalate HOOCCOOꢀ ligand. This ligand
can be coordinated to a copper complex possessing a free equato-
rial position without further deprotonation. In spite of the lack of
structural determination, the magnetic study is in agreement with
a dinuclear compound including a five coordinate CuII ion. Such a
result can be reached in two ways, the hydrogen oxalato HOOC-
COOꢀ ligand bridging two copper ions through a single oxygen
atom, as observed in the [LCu(CH3COO)]2 complex [9] or through
a more classical two oxygen atoms bridge, as in the copper acetate
complex. In view of a recent X-ray study of a CuII compound
including the sole hydrogen oxalate ligand [8], the second possibil-
ity should be considered. Complete deprotonation by addition of
base to a water solution of the above complex yields a dinuclear
complex in which the two symmetry-related CuII ions are pentaco-
ordinate and linked by the oxalato ligand. The absence of a mag-
netic interaction through the oxalato bridge is confirmed by the
constant
vMT product from 300 to 15 K, the absence of half-field
transition in the rhombic EPR spectrum and by DFT calculations
evidencing that the singlet and triplet states have the same energy.
The low vMT product in the 2–10 K temperature range could result
from the hydrogen bonding network through the water molecules.
We note that deprotonation of the LCu(HOOCCOO)(H2O) complex
should yield polynuclear Cu-3d or Cu-4f complexes. Unfortunately,
the expected complexes have not been characterized to date.
Appendix A. Supplementary data
CCDC 945567 contains the supplementary crystallographic data
for complex 2. These data can be obtained free of charge via http://
www.ccdc.cam.ac.uk/conts/retrieving.html, or from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ
UK; fax: (+44) 1223-336-033; or e-mail: deposit@ccdc.cam.ac.uk.
The 13C NMR spectrum of the monodeprotonated oxalic acid and
Fig. 7. DFT calculated structures 1a (left) and 1b (right). In 1b, the oxalate moieties were symmetrically rotated by 10° away the neighboring ligands.

132
A.J. Mota et al. / Polyhedron 63 (2013) 127–132
[19] (a) J.J. Borrás-Almenar, J.M. Clemente-Juan, E. Coronado, B.S. Tsukerblat, Inorg.
Chem. 38 (1999) 6081;
EPR spectra of the complexes are given in the Supplementary
Material. Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/
j.poly.2013.06.054.
(b) J.J. Borrás-Almenar, J.M. Clemente-Juan, E. Coronado, B.S. Tsukerblat, J.
Comput. Chem. 22 (2001) 985.
[20] F. James, M. Roos, MINUIT program, a system for function minimization and
analysis of the parameters errors and correlations, Comput. Phys. Commun. 10
(1975) 343.
References
[21] (a) A.D. Becke, Phys. Rev. A 38 (1988) 3098;
(b) C.T. Lee, W.T. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785;
(c) A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
[1] T.R. Felthouse, E.J. Laskowski, D.N. Hendrickson, Inorg. Chem. 16 (1977) 1077.
[2] M. Julve, M. Verdaguer, A. Gleizes, M. Philoche-Levisalles, O. Kahn, Inorg.
Chem. 23 (1984) 3808.
[3] M. Julve, J. Faus, M. Verdaguer, A. Gleizes, J. Am. Chem. Soc. 106 (1984) 8306.
[4] J. Glerup, P.A. Goodson, D.J. Hodgson, K. Michelsen, Inorg. Chem. 34 (1995)
6255.
[5] P. Roman, C. Guzman-Miralles, A. Luque, J.I. Beitia, J. Cano, F. Lloret, M. Julve, S.
Alvarez, Inorg. Chem. 35 (1996) 3741.
[6] M.H. Molina, P.A. Lorenzo-Luis, C. Ruiz-Perez, Cryst. Eng. Comm. 16 (2001) 1.
[7] X.-F. Chen, L. Liu, J.-G. Ma, L. Yi, P. Cheng, D.-Z. Liao, S.-P. Yan, Z.-H. Jiang, J. Mol.
Struct. 750 (2005) 94.
[8] O.Z. Yesilel, H. Erer, M. Odabasoglu, O. Buyukgungor, J. Inorg. Organomet.
Polym. 20 (2010) 78.
[9] J.P. Costes, F. Dahan, J.P. Laurent, Inorg. Chem. 24 (1985) 1018.
[10] M. Julve, O. Kahn, Inorg. Chim. Acta 76 (1983) L39.
[11] C.K. Fair, MOLEN: Molecular Structure Solution Procedures, Enraf-Nonius,
Delft, The Netherlands, 1990.
[12] A.C.T. North, D.C. Phillips, F.S. Mathews, Acta Crystallogr., Sect. A A24 (1968)
351.
[13] G.M. Sheldrick, SHELXS-97. Program for Crystal Structure Solution, University of
Göttingen, Göttingen, Germany, 1990.
[22] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R.
Cheeseman, J. A. Montgomery Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M.
Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li,
J. E. Knox, H. P. Hratchian, J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski,
S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, R.
Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J.
Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M.
Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez, J. A.
Pople, in GAUSSIAN 03, Revision C.02, Gaussian, Inc., Wallingford, CT, 2004.
[23] G.B. Bacskay, Chem. Phys. 61 (1981) 385.
[24] A. Schäfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 100 (1994) 5829.
[25] Jaguar, version 7.9, Schrödinger, LLC., New York, NY, 2012.
[26] J.P. Costes, F. Dahan, J. Ruiz, J.P. Laurent, Inorg. Chim. Acta 239 (1995) 53.
[27] A.P. Gingsberg, M.E. Lines, Inorg. Chem. 11 (1972) 2289.
[28] J.P. Costes, C. Duhayon, L. Vendier, E. Colacio, A.J. Mota Avila, J. Suarez Varela,
Inorg. Chem. 51 (2012) 1011.
[29] F. Lloret, M. Julve, M. Mollar, I. Castro, J. Latorre, J. Faus, X. Solans, I.
Morgenstern-Badarau, J. Chem. Soc., Dalton Trans. (1989) 729.
[30] S. Alvarez, M. Llunell, J. Chem. Soc., Dalton Trans. (2000) 3288.
[31] J. Cano, P. Alemany, S. Alvarez, M. Verdaguer, E. Ruiz, Chem. Eur. J. 4 (1998)
476.
[14] G.M. Sheldrick, SHELXL-97. Program for the refinement of crystal structures from
diffraction data, University of Göttingen, Göttingen, Germany, 1997.
[15] International Tables for Crystallography, vol. C, Kluwer Academic Publishers,
Dordrecht, The Netherlands, 1992.
[16] L. Zsolnai, H. Pritzkow, G. Huttner, ZORTEP. Ortep for PC, Program for Molecular
Graphics, University of Heidelberg, Heidelberg, Germany, 1996.
[17] P. Pascal, Ann. Chim. Phys. 19 (1910) 5.
[32] P. Gomez-Saiz, J. Garia-Tojal, M. Maestro, J. Mahia, L. Lezama, T. Rojo, Eur. J.
Inorg. Chem. (2003) 2123.
[18] A.K. Boudalis, J.M. Clemente-Juan, F. Dahan, J.-P. Tuchagues, Inorg. Chem 43
(2004) 1574.