Structural and spectral studies of N-alkyloxamates and their complexes: X-ray structures of MeHNCOCOOK and [Cu(EtHNCOCOO)2], and vibrational studies

Article abstract of DOI:10.1016/S0022-2860(00)00693-1

This work is devoted to structural and vibrational studies of N-alkyloxamates and their complexes. The structures of MeHNCOCOOK and [Cu(EtHNCOCOO)₂] have been determined by single-crystal X-ray crystallography. The potassium salt crystallizes in the monoclinic space group P2₁/n and the copper(II) complex in the triclinic space group P1?. The lattice constants are a = 3.882(2), b = 10.175(1), c = 13.004(1), β = 97.04(1)° for MeHNCOCOOK and a = 5.013(1), b = 6.996(1), c = 8.575(1) ?, α = 73.54(1), β = 85.66(1), γ = 71.22(1)° for [Cu(EtHNCOCOO)₂]. The salt has a trans secondary amide group and the K^- ion is surrounded by seven O atoms belonging to six different N-methyloxamate(-1) moieties. The ligand EtHNCOCOO^- behaves as a bidentate chelate in the trans square planar copper(II) complex with ligated atoms being the amide and one of the carboxylate oxygens. The copper(II) complex was characterized by its room-temperature magnetic susceptibility and solid-state d-d spectrum. A detailed comparison of the vibrational spectra (FT-IR, FT-Raman) of [Cu(EtHNCOCOO)₂] with the corresponding potassium salt, EtHNCOCOOK, is presented. The vibrational data are discussed in terms of the nature of bonding and the known structure of the copper(II) complex. Quantum-chemical calculations were performed to obtain the equilibrium geometries and torsional energies for an isolated oxamate(-1) ion (H₂NCOCOO^-) and its hypothetical square planar zinc(II) complex.

Full text of DOI:10.1016/S0022-2860(00)00693-1

Journal of Molecular Structure 559 (2001) 167–177
www.elsevier.nl/locate/molstruc
Structural and spectral studies of N-alkyloxamates and their
complexes: X-ray structures of MeHNCOCOOK and
[Cu(EtHNCOCOO)2], and vibrational studies
G.S. Papaefstathioua, A. Peetersb, A.T.H. Lenstrab, H.O. Desseync,1, S.P. Perlepesa,
*
aDepartment of Chemistry, University of Patras, 265 00 Patras, Greece
bDepartment of Chemistry, University of Antwerp (UIA), Universiteitsplein 1, 2610 Wilrijk, Belgium
cDepartment of Chemistry, University of Antwerp (RUCA), Groenenborgerlaan 171, 2020 Antwerp, Belgium
Received 10 April 2000; accepted 26 June 2000
Abstract
This work is devoted to structural and vibrational studies of N-alkyloxamates and their complexes. The structures of
MeHNCOCOOK and [Cu(EtHNCOCOO)2] have been determined by single-crystal X-ray crystallography. The potassium
ꢀ
salt crystallizes in the monoclinic space group P21/n and the copper(II) complex in the triclinic space group P1: The lattice
constants are a ˆ 3:882ꢀ2†; b ˆ 10:175ꢀ1†; c ˆ 13:004ꢀ1†; b ˆ 97:04ꢀ1†Њ for MeHNCOCOOK and a ˆ 5:013ꢀ1†; b ˆ
ꢁ
6:996ꢀ1†; c ˆ 8:575ꢀ1† A; a ˆ 73:54ꢀ1†; b ˆ 85:66ꢀ1†; g ˆ 71:22ꢀ1†Њ for [Cu(EtHNCOCOO)2]. The salt has a trans secondary
amide group and the KϪ ion is surrounded by seven O atoms belonging to six different N-methyloxamate(-1) moieties. The
ligand EtHNCOCOOϪ behaves as a bidentate chelate in the trans square planar copper(II) complex with ligated atoms being the
amide and one of the carboxylate oxygens. The copper(II) complex was characterized by its room-temperature magnetic
susceptibility and solid-state d–d spectrum. A detailed comparison of the vibrational spectra (FT-IR, FT-Raman) of [Cu(EtHN-
COCOO)2] with the corresponding potassium salt, EtHNCOCOOK, is presented. The vibrational data are discussed in terms of
the nature of bonding and the known structure of the copper(II) complex. Quantum-chemical calculations were performed to
obtain the equilibrium geometries and torsional energies for an isolated oxamate(-1) ion (H2NCOCOOϪ) and its hypothetical
square planar zinc(II) complex. ᭧ 2001 Elsevier Science B.V. All rights reserved.
Keywords: Potassium N-methyloxamate(-1); Bis(N-ethyloxamato(-1)-O,O0)copper(II); Crystal structures; IR spectroscopy; Raman
spectroscopy
1. Introduction
dination chemistry of the mono and dianion of oxamic
acid (H2NCOCOOϪ, HNCOCOO2Ϫ) and oxamate-
based ligands with substituents bearing donor groups
is a central theme in transition metal chemistry.
Research in this area ranges from the solution of
pure chemical [1–5] and spectroscopic [6,7] problems
to the renaissance in the fields of heterometallic chem-
istry [8,9], molecular magnetism [10,11] and
advanced materials [12]. The oxamate(-1) anion is
It is becoming increasingly apparent that the coor-
* Corresponding author. Tel.: ϩ30-61-997-146; fax: ϩ30-61-
997-118.
E-mail addresses: perlepes@patreas.upatras.gr (S.P. Perlepes),
jojans@ruca.ua.cc.be (H.O. Desseyn).
1
Co-corresponding author. Tel.: ϩ32-3-2180365; fax: ϩ32-3-
2180233.
0022-2860/01/$ - see front matter ᭧ 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-2860(00)00693-1

168
G.S. Papaefstathiou et al. / Journal of Molecular Structure 559 (2001) 167–177
also a flexible building block capable of generating
extended hydrogen-bonded motifs for crystal engi-
neering [13].
attention to vibrational spectra. Here we report on
the X-ray structures of the potassium salt of N-
methyloxamic acid (MeHNCOCOOK) and complex
bis(N-ethyloxamato(-1)-O,O0)copper(II), [Cu(EtHN-
COCOO)2], together with the spectroscopic character-
ization of the later. The vibrational analysis of
MeHNCOCOOK has been given [30]. We also
include in the present paper quantum-chemical calcu-
lations in order to obtain equilibrium geometries and
torsional energies for an isolated H2NCOCOOϪ ion
and its 1:2 zinc(II) complex.
Single-crystal X-ray crystallography has revealed
that the oxamic acid normally coordinates as a mono-
anion, only after the ionization of the carboxylic
hydrogen, through the amide and the one carboxylate
oxygen atom forming five-membered chelate rings
[1,14–17]. It also coordinates as a monodentate O-
carboxylato ligand [1] or as a tridentate bridging
ligand, using all of its oxygen atoms [18]. Oxamic
acid also can coordinate as dianion since both the
carboxylic and amidic hydrogens can be removed,
and the oxamato(-2) complexes be prepared. In the
only crystallographically characterized example,
complex [Pr(HNCOCOO)1.5(H2O)3]n·2.75nH2O [19],
the oxamate(-2) anion behaves as a bis-bidentate
O(carboxylate), O(amide)/N, O(carboxylate) bridging
ligand. Other coordination modes of H2NCOCOOϪ
[20,21] and HNCOCOO2Ϫ [7,22–28] have also been
proposed on the basis of vibrational spectroscopy.
For many years, our groups have been engaged
in the development of chemistry, thermal properties
and vibrational spectroscopy of transition metal
complexes with oxamate anions as ligands
[5–7,21,26–28]. In contrast to the obtained large
body of knowledge, almost nothing is known about
the corresponding complexes of alkyl-substituted
oxamic acids (RHNCOCOOH, R2NCOCOOH),
although this class of compounds could offer inter-
esting features in terms of structural and spectroscopic
properties. The presence of an alkyl group instead of a
hydrogen atom in the amide functional group is
expected — due to different steric, electronic and
hydrogen-bonding effects — to alter the coordinating
behaviour of these ligands in comparison with the
behaviour of simple oxamate anions and, hence,
differences might be expected. From the vibrational
spectroscopy point of view, the different spectral char-
acteristics of the primary (–NH2CO–) and secondary
(–RHNCO–) or tertiary (–R2NCO–) amide groups
[29] may give rise to interesting phenomena; the thor-
ough study of the vibrational fundamentals of these
functional groups should give valuable information
regarding their coordination modes.
2. Experimental
2.1. General
All manipulations were performed under aerobic
conditions using materials (analytical grade) as-
received; H2O was distilled in-house. Microanalyses
for C, H and N were performed by the University of
Ioannina (Greece) Microanalytical Laboratory.
Copper analysis was carried out by EDTA titration.
2.2. Compound preparation
The salts RHNCOCOOK ꢀR ˆ Me; Et† were
synthesized by reactions 1 and 2 as described
elsewhere [30]. Their purity was checked by micro-
analyses and IR spectra.
EtOH
ꢀCOOEt†2 ϩ RNH2 ! RHNCOCOOEt ϩ EtOH ꢀ1†
0ЊC
RHNCOCOOEt
EtOH=H2O
ϩKOH
!
RHNCOCOOK ϩ EtOH
ꢀ2†
For the preparation of the complex [Cu(EtHN-
COCOO)2], to a solution of EtHNCOCOOK (3.10 g,
20.0 mmol) in H2O (20 ml) was added a solution of
CuCl2·2H2O (1.70 g, 10.0 mmol) in the same solvent
(10 ml). A blue microcrystalline material formed
within 1–2 min. The reaction mixture was stirred at
room temperature for about 15 min after the precipi-
tation was observed. The solid was collected by filtra-
tion, washed with H2O (2 × 3 ml), EtOH ꢀ2 × 5 ml†
and copious amounts of Et2O, and dried in vacuo
over P4O10. Yield: 2.10 g (71%). Anal. found (calcd.
With the above considerations in mind we have
begun a systematic study of the coordination chem-
istry of alkyl-substituted oxamic acids, with special

G.S. Papaefstathiou et al. / Journal of Molecular Structure 559 (2001) 167–177
169
Table 1
Summary of crystal data, data collection and structure refinement for X-ray diffraction study of MeHNCOCOOK and [Cu(EtHNCOCOO)2]
Crystal data
Compound
MeHNCOCOOK
C3H4NO3K
141.16
Colourless prisms
0.25 × 0.30 × 0.15
Monoclinic
P21/n
3.882(2)
10.175(1)
13.004(1)
[Cu(EtHNCOCOO)2]
C8H12N2O6Cu
295.73
Chemical formula
Formula weight
Colour, habit
Blue prisms
Crystal size (mm)
Crystal system
Space group
0.20 × 0.25 × 0.15
Triclinic
ꢀ
P1
˚
a (A)
5.013(1)
6.996(1)
8.575(1)
73.54(1)
85.66(1)
71.22(1)
273.0(1)
1
1.799
2.02
MoKa ꢀl ˆ 0:71073†
˚
b (A)
˚
c (A)
a (Њ)
b (Њ)
g (Њ)
97.04(1)
3
˚
V (A )
509.8(4)
4
1.683
0.93
Z
Dcalc (g cmϪ3
)
m (mmϪ1
)
˚
Radiation (A)
MoKa ꢀl ˆ 0:71073†
Data collection
Temperature (K)
Monochromator
Collection method
Absorption
293
Graphite
v/2u scan
Numerical
293
Graphite
v/2u scan
Numerical
umax (Њ)
27
27
Measured reflections
Observed reflections [I Ͼ 3s(I)]
Index ranges
1097
918
1194
1178
h, 0–4; k, 0–13; l, –16–16
h, 0–6; k, –8–8; l, –10–10
Refinement on ͉F͉
Variables
85
97
[D/s]max
0.05
0.6
0.1
0.3
Ϫ3
˚
͉Dr͉ (e A
)
max
R
Rw
GoF
0.039
0.047
2.17
0.026
0.039
2.24
for C8H12N2O6Cu): C, 32.5 (32.3); H, 4.1 (4.1); N, 9.5
(9.6); Cu, 21.5 (22.1)%.
The H/D substituted complex was prepared on a
smaller synthetic scale as described for the normal
complex using D2O as the solvent; the solid product
was washed only with MeOD.
Pascal’s constants. The solid-state (diffuse reflec-
tance, 850–350 nm) electronic spectrum of the
copper(II) complex was recorded on a Varian 634
instrument. IR spectra (4000–500 cmϪ1
)
were
recorded and averaged on a Bruker IFS 113v
Fourier-transform (FT) spectrometer, using a liquid
nitrogen-cooled MCT detector, with samples prepared
as KBr pellets. FT far-IR spectra (500–50 cmϪ1) were
recorded on a Bruker IFS 113v spectrometer with a
DTGS detector using polyethylene pellets. FT Raman
data have been collected on a Bruker IFS 66v inter-
ferometer equipped with a FRA 106 Raman acces-
sory, a CW Nd-YAG laser source and a liquid
nitrogen-cooled Ge detector.
2.3. Physical measurements
The magnetic susceptibility was measured at room
temperature by the Faraday method with a Cahn–
Ventron RM-2 balance
standardized
with
HgCo(SCN)4. The molar susceptibility was corrected
for the diamagnetism of the constituent atoms using

170
G.S. Papaefstathiou et al. / Journal of Molecular Structure 559 (2001) 167–177
Table 2
˚
Interatomic distances, angles and dimensions of the hydrogen bonds (distances in A and angles in Њ) for MeHNCOCOOK and [Cu(EtHN-
COCOO)2]
The potassium coordination in MeHNCOCOOKa
K–O1I
2.855(2)
2.713(2)
2.996(2)
2.810(2)
2.821(2)
2.765(2)
2.962(2)
84.4(1)
O1ii–K–O2iii
O1ii–K–O5ii
O1ii–K–O5iv
O1ii–K–O5v
O2–K–O2iii
O2–K–O5ii
O2–K–O5iv
O2–K–O5v
O2iii–K–O5ii
O2iii–K–O5iv
O2iii–K–O5v
O5ii–K–O5iv
O5ii–K–O5v
O5iv–K–O5v
71.4(1)
59.1(1)
144.5(1)
82.6(1)
83.9(1)
138.9(1)
82.6(1)
142.2(1)
127.5(1)
139.5(1)
82.5(1)
85.5(1)
75.0(1)
85.3(1)
K–O1ii
K–O2
K–O2iii
K–O5ii
K–O5iv
K–O5v
O1i–K–O1ii
O1i–K–O2
O1i–K–O2iii
O1i–K–O5ii
O1i–K–O5iv
O1i–K–O5v
O1ii–K–O2
66.9(1)
124.7(1)
72.8(1)
83.7(1)
146.6(1)
125.4(1)
The copper(II) coordination in [Cu(EtHNCOCOO)2]
Cu–O1
Cu–O5
1.942(1)
1.952(1)
O1–Cu–O5
84.7(1)
The RHNCOCOOϪ dimensions and geometry (R ˆ Me, Et)
MeHNCOCOOK
[Cu(EtHNCOCOO)2]
1.269(2)
1.228(3)
1.544(3)
1.261(2)
1.299(2)
1.460(3)
1.503(3)
127.6(2)
113.0(1)
119.4(1)
117.5(1)
117.7(1)
121.8(2)
124.6(1)
111.4(2)
1.0(2)
O1–C3
O2–C3
C3–C4
C4–O5
C4–N6
N6–C7
C7–C8
O1–C3–O2
O1–C3–C4
O2–C3–C4
C3–C4–O5
C3–C4–N6
O5–C4–N6
C4–N6–C7
N6–C7–C8
O1–C3–C4–O5
O1–C3–C4–N6
C3–C4–N6–C7
C4–N6–C7–C8
Hydrogen bondsa
Compound
1.248(3)
1.245(3)
1.560(3)
1.237(3)
1.317(3)
1.447(3)
128.9(2)
114.5(2)
116.6(2)
121.6(2)
115.0(2)
123.3(2)
122.6(2)
8.5(3)
Ϫ171.3(2)
179.2(2)
Ϫ179.5(2)
Ϫ174.2(2)
114.2(2)
Db–H···Ac
N6–HN6···O2
N6–HN6···O1I
N6–HN6···O2
Db···Ac
2.681(3) ϩ
2.810(3)
2.752(3)
H···Ac
2.42(3)
1.94(3)
2.39(3)
ϽDbHAc
97(2)
166(3)
117(3)
MeHNCOCOOK
MeHNCOCOOK
[Cu(EtHNCOCOO)2]
a
Symmetry transformations: (i) Ϫ0:5 Ϫ x; 0:5 ϩ y; 1:5 Ϫ z; (ii) 0:5 Ϫ x; 0:5 ϩ y; 1:5 Ϫ z; (iii) 1 ϩ x; y; z; (iv) Ϫ0:5 ϩ x; 0:5 Ϫ y; Ϫ0:5 ϩ
z; (v) 0:5 ϩ x; 0:5 Ϫ y; Ϫ0:5 ϩ z:
b
D ˆ donor atom.
c
A ˆ acceptor atom.

G.S. Papaefstathiou et al. / Journal of Molecular Structure 559 (2001) 167–177
171
techniques. For both structures, all non-H atoms were
readily located and refined with anisotropic thermal
parameters. H atoms were clearly visible in difference
electron density maps. Minimum and maximum trans-
missions were calculated using an absorption correc-
tion based on the work by De Meulenaer and Tompa
[31]. The values of 0.72 and 0.78 were found for the
potassium salt, and 0.64 and 0.74 for the copper(II)
complex. In view of the size of the absorption errors,
both measurements were corrected for X-ray absorp-
tion. Least-squares calculations on ͉F͉ were used to
optimize the model parameters r and b. Reflections
were weighted following their statistical errors. The
temperature factors of the H atoms were kept fixed at
4.0 A . We stopped the refinement when the largest
Fig. 1. PLATON view of the anion MeHNCOCOOϪ showing the
atom labeling scheme.
2
˚
shift over the error ratio dropped below 0.05.
2.4. X-ray crystallographic studies
3. Results and discussion
Suitable crystals of MeHNCOCOOK were
prepared by recrystallization from H2O. Crystals of
[Cu(EtHNCOCOO)2] suitable for diffractometry
were obtained by layering of acetone into a dilute
aqueous solution of the microcrystalline solid. Crystal
data and full details of the data collection and data
processing are listed in Table 1. The crystals were
mounted in air on an Enraf–Nonius CAD-4 diffracto-
meter. Unit cell dimensions were determined by using
the angular settings of 25 reflections in the ranges
4Њ Ͻ u Ͻ 12Њ for MeHNCOCOOK and 6Њ Ͻ u Ͻ 16Њ
for the copper(II) complex. Three intensity and orien-
tation control reflections were measured every hour;
they did not show significant fluctuations indicative of
crystal decay or diffractometer instability.
3.1. Description of structures
Interatomic distances, angles and details of the H-
bonds for MeHNCOCOOK and [Cu(EtHNCOCOO)2]
are given in Table 2. Figs. 1 and 2 show perspective
views of MeHNCOCOOϪ and the copper(II)
complex, respectively.
The Kϩ ion is surrounded by seven O atoms (ca.
˚
2.71–3.00 A). The environment of the potassium
includes oxygens from six N-methyloxamate(-1)
moieties, because O5ii and O1ii (for symmetry codes
see Table 2) are part of the same anionic moiety. The
ligands are in fact three translational pairs, in which
the individual units are separated by the short a-axis
˚
of 3.882(2) A.
The structures were solved by Patterson and Fourier
Fig. 2. PLATON view of complex [Cu(EtHNCOCOO)2] showing the atom labeling scheme. Atoms generated by the inversion centre have not
been labelled.

172
G.S. Papaefstathiou et al. / Journal of Molecular Structure 559 (2001) 167–177
also consequences in the dimensions of the secondary
amide group. Although strictly speaking we compare
different “free” and coordinated N-alkyloxamates
(MeHNCOCOOϪ vs. EtHNCOCOOϪ), the compar-
ison between the MeHNCOCOOK and [Cu(EtHN-
Scheme 1.
The K–O distances may be compared with the
corresponding distances in the seven-coordinate salt
COCOO)2] structures shows that there is
a
˚
lengthening of the C4–O5 [1.237(3) vs. 1.261(2) A]
and a shortening of the C4–N6 bond [1.317(3) vs.
˚
H2NCOCOOK (2.74–2.84 A) [13] and the five-coor-
˚
˚
dinate salt H2NCSCOOK (2.64–2.83 A) [32].
1.299(2) A] in the copper(II) complex. This can be
The asymmetric unit of the copper(II) complex
contains half of the [Cu(EtHNCOCOO)2] molecule;
the CuII atom sits on an inversion centre. The mono-
anionic ligand is coordinated to the copper(II) centre
as bidentate chelate via the amide oxygen (O5) and
one of the carboxylate oxygen (O1) atoms. Thus, the
CuII atom shows a square planar coordination with a
Ci symmetry. The Cu–O bond distances of 1.942(1)
explained by the fact that the partial double-bond
character of the C–O and C–N bonds of the amide
group, due to the resonance configurations shown in
Scheme 1, is influenced by the copper(II) binding of
the group. Upon coordination via oxygen, the doubly
positively charged copper(II) ion stabilizes the nega-
tive charge on the oxygen atom. The amide function
now occurs in its polar resonance form and, thus, the
double-bond character of the C–N bond increases,
while the double-bond character of the C–O bond
decreases. Complex [Cu(EtHNCOCOO)2] is not
extreme in the deformation of the N-alkyloxamate(-1)
moiety. Its flexibility due to coordination effects
can be demonstrated by tetrachloro(N-methyl-
oxamato(-2)-O,O0)antimony(V), [SbCl4(MeHNCOC-
OO)], which shows [35] a large difference between
˚
and 1.952(1) A are very close to the corresponding
distances [1.934(1) and 1.966(1), respectively] in the
complex trans-[Cu(LH)2(H2O)2], where LHϪ is the
malonamate(-1) ion (H2NCOCH2COOϪ), which
contains the coordinated amide and carboxylate
oxygens at the equatorial plane of the trans-elongated
octahedron [33]. The formation of the five-membered
chelate ring causes a 5Њ decrease in the O1–Cu–O5
angle [84.7(1)Њ]. Such a ring is also present in MeHN-
COCOOK, but the corresponding angle (O1ii–K–
O5ii) is reduced to 59.1(1)Њ. This great decrease is a
consequence of the elongation in the metal–oxygen
˚
the carboxylate C–O distances [1.31,1.22 A]; the
carbon–nitrogen and carbon–oxygen bond lengths
of the secondary amide function are 1.27 and
˚
1.29 A, respectively, with the former being shorter
˚
distances from ca. 1.95 A in the copper(II) complex to
than the latter!
˚
ca. 2.75 A in the potassium salt. For a fixed O1···O5
As mentioned above, the O1–C3–C4–O5 torsion
angle is 8.5(3)Њ in MeHNCOCOOK and 1.0(2)Њ in
[Cu(EtHNCOCOO)2] indicating that ‘free’ and coor-
dinated RHNCOCOOϪ ions are slightly non-planar
and planar, respectively. The CuII atom is situated
distance, the O–K–O angle should be equal to 55Њ;
the slight increase to 59Њ is brought about by a
increase in the O1···O5 distance caused by a change
in the O1–C3–C4–O5 torsion angle from 1Њ in the
complex to 8.5Њ in the salt.
˚
ϳ0.1 A outside the least-squares plane through the
The impact of complexation on the RHNCOCOOϪ
dimensions and geometry is clearly visible. The
carboxylate O1–C3 and O2–C3 distances [1.248(3),
ligand fragment O1O2C3C4O5N6, the dihedral
angle between this plane and the plane defined by
atoms Cu, O1, O5 being 4.1(5)Њ. The virtual planarity
of the RHNCOCOOϪ ions in the potassium salt and
the copper(II) complex was rather unexpected,
because, in theory, rotation can take place around
the central C–C bond. This planarity of the anions
can partly be attributed to the presence of an intramol-
ecular hydrogen bond, N6···O2. Due to the geometry
of anions, the NH···O angle is very steep (97,117Њ).
This intramolecular hydrogen bond is also an invari-
able characteristic feature in the crystal structures of
˚
1.245(3) A] are almost identical in MeHNCOCOOK.
The observed equivalence is typical for salts and has
also been seen in ammonium oxamate [34]. The
monodentate coordination of the carboxylate group
results in a significant difference between the length
of the two carboxylate CO bonds, the O1–C3 distance
˚
[1.269(2) A] involving the coordinated oxygen being
˚
longer than the O2–C3 distance [1.228(3) A]. The
formation of a strong bond between O5 and Cu has

G.S. Papaefstathiou et al. / Journal of Molecular Structure 559 (2001) 167–177
173
Table 3
Most characteristic and diagnostic vibrational fundamentals (cmϪ1) of EtHNCOCOOK and [Cu(EtHNCOCOO)2] (b ˆ broad; m ˆ medium;
s ˆ strong; sh ˆ shoulder; v ˆ very)
EtHNCOCOOK
[Cu(EtHNCOCOO)2]
Assignments
IR
Raman
IR
Raman
3290 m
1669 vs
1630 vs
1510 s
1398 s
1215 m
1163 m
850 m
3290 [4]
1661 [6]
1633 [3]
1514 [4]
1403 [10]
1216 [6]
1158 [4]
850 [4]
3288 s
1667 sh
1638 vsb
1548 m
1374 s
1259 s
1156 s
835 m
791 mb
407 m
349 s
3293 [2]
1673 [2]
1649 [5]
1544 [10]
1372 [8]
1258 [5]
1153 [4]
835 [4]
790 [1]
385 [3]
265 [1]
237 [10]
n(NH)
n(CO) [Amide I]
nas(CO2)
d(NH) ϩ n(CN) [Amide II]
ns(CO2)
n(CN) ϩ d(NH) [Amide III]
n(NEt)
n(CC)
p(NH) [Amide V]
d(NEt)
n(CuOcarboxylate
n(CuOamide
803 mb
334 m
803 [2]
337 [4]
)
308 m
)
2
˚
six salts containing the unsubstituted oxamate anion
(H2NCOCOOϪ) and different organic or inorganic
cations [13].
The crystal structure of MeHNCOCOOK is stabi-
lized by the intermolecular N6–H···O1 ꢀϪ0:5 Ϫ
x; 0:5 ϩ y; 1:5 Ϫ z† hydrogen bond, which creates
an infinite low-dimensional motif.
to B(C3) and B(C4) [1.79(3) and 1.81(3) A , respec-
2
˚
tively], but also to B(Cu) [1.87(1) A ]. In our opinion
this points out a strong correlation in the dynamic
behaviour of the CuII atom and the ligand entity.
This equality in behaviour is absent in the potassium
salt with its difference of 30% between Bligand (transla-
tion) and B(Kϩ).
Up to this point we have directly analysed the
geometrical evidence of the structures. Their dynamic
behaviour, observed via the atomic displacement
parameters, is also interesting. Hirshfeld’s rigid
bond test [36] shows that the anion can be regarded
as a ‘rigid body’ in the salt as well as in the complex.
This makes a TLS-analysis [37] worthwhile. The
analysis of atomic b’s in the potassium salt
Librations also show a strong interaction between
copper(II) and ligand. Mean square librations for the
ligand in the copper(II) complex are 10.7, 8.3 and
7.5 degrees2. The decrease in L11 from ϳ52 to
ϳ11 degrees2 shows a drastic decrease in the partial
rotation around the C3–C4 bond due to extra bonding,
which is absent in the salt.
reveals rms-translational values of 0.16, 0.13 and
2
˚
˚
0.11 A. This represents a trace T of 0.0526 A ,
3.2. Magnetic moment and visible spectrum of
[Cu(EtHNCOCOO)2]
2
˚
which suggests a Beq value of 1.4 A . So the N-
methyloxamate translation is practically dictating
the B values for C3 and C4 [1.70(4) and 1.60(4)
A , respectively]. The adp’s of the other atoms
The
room-temperature
effective
magnetic
2
˚
moment(meff) of [Cu(EtHNCOCOO)2] is 1.99 B.M.,
indicating that the complex is magnetically dilute
[38].
The solid-state (diffuse reflectance) d–d spectrum
of the copper(II) complex exhibits two broad struc-
tured bands at about 15,000 and 17,800 cmϪ1, which
involve the three expected transitions to dx2Ϫy2 from
dz2 ; dxy and dxz, yz [39]. The d–d frequencies are fairly
typical of square planar stereochemistries with a
CuIIO4 chromophore [39].
deviate from this reference value via molecular
libration. We calculated mean square librations
of 51.8, 9.3 and 4.4 degrees2. Here L11 of
52 degrees2 is a partial rotation around the inertia
axis that practically coincides with C3–C4.
A similar analysis was made for the copper(II)
complex. Here the rms-values for translation are 0.19,
2
˚
˚
0.14 and 0.10 A. With a trace T of 0.064 A we arrive at
2
˚
a Beq value of 1.68 A . This value is not only very close

174
G.S. Papaefstathiou et al. / Journal of Molecular Structure 559 (2001) 167–177
are weaker in [Cu(EtHNCOCOO)2] than in EtHNCO-
COOK.
The higher-frequency shift of the nas(CO2) mode
and the lower-frequency shift of the ns(CO2) mode
in the copper(II) complex are indicative of the mono-
dentate carboxylate coordination [44]. The monoden-
tate coordination removes the equivalence of the two
CO bonds (present in the potassium salt) and a
pseudo-ester configuration can be considered for the
complex [29,44]. The spectroscopic behaviour of the
carboxylate group is then no longer described by
the nas(CO2) and ns(CO2) modes, but by an ‘ester’
nꢀC ˆ Ouncoordinated
†
mode situated at a higher
frequency than nas(CO2) and a nꢀC ˆ Ocoordinated
†
mode situated at a lower frequency than ns(CO2).
Considering the trans secondary amide function,
the Amide II and III bands — which involve the
n(CN) mode — are situated at higher frequencies in
the spectra of [Cu(EtHNCOCOO)2] than for EtHN-
COCOOK, whereas the Amide I band [n(CO)]
shows a frequency decrease. These shifts are consis-
tent with amide-O coordination [42,43].
Fig. 3. The ‘pile-up’ effect for the coordinated carboxylate (a) and
secondary amide (b) groups in [Cu(EtHNCOCOO)2].
3.3. Vibrational spectra
The shifts of the carboxylate and amide fundamen-
tals described above can very well be explained by
considering the ligand–metal bonding as a donor–
acceptor complex [45] as shown in Fig. 3. The ‘pile-
up’ effect indicates a shortening of the O2–C3 and a
lengthening of the O1–C3 bond for the coordinated
carboxylate group, and a lengthening of the C4–O5,
N6–H, N6–C7 bonds and a shortening of the C4–N6
bond for the coordinated amide group. This model
also explains the lower frequency of n(NEt) in the
copper(II) complex. According to this model we
should also observe lower and higher frequencies
for the n(NH) and p(NH) modes, respectively,
in the complex. As we do not observe the
expected shifts, we can only conclude that the
difference in hydrogen-bonding strength influences
these fundamentals much more than the effect of
coordination.
Table 3 gives diagnostic IR and Raman bands for
the anion EtHNCOCOOϪ (present in its potassium
salt) and its copper(II) complex. The full vibrational
analysis of MeHNCOCOOK, considering a planar Cs
configuration, has been published [30]. Assignments
in Table 3 have been given in comparison with the
data obtained for MeHNCOCOOK [30], for ligands
containing carboxylate groups and their complexes
[6,21,29,32,33,40,41] and for ligands with trans
secondary amide groups and their complexes
[29,42,43], and have been assisted by H/D isotopic
substitution. The spectroscopic discussion will be
confined only to the most important features in rela-
tion to the coordination mode of EtHNCOCOOϪ and
the structure of its copper(II) complex.
The frequency difference between some IR and
Raman bands in [Cu(EtHNCOCOO)2], especially in
the n(CuO) region (350–230 cmϪ1), clearly indicates
the presence of an inversion centre.
The Amide V mode has a pure p(NH) character
and appears as a broad band of medium intensity in
the IR spectra and as a weak Raman band in the
800 cmϪ1 region. This mode is very sensitive to
hydrogen bonding and its lower frequency in the
copper(II) complex indicates that hydrogen bonds
The fundamental with a high n(CC) character is
observed at a lower frequency in the copper(II)
complex due to the ring system and the mass effect.
The complicated character of n(CC), i.e. coupling
with other vibrations in the ring, also influences this
fundamental.
The presence of one n(CuOcarboxylate) and one
n(CuOamide) vibration in the low-frequency IR

G.S. Papaefstathiou et al. / Journal of Molecular Structure 559 (2001) 167–177
175
Fig. 4. The energy profile (Kcal/mol) for a torsion (Њ) around C3–C4. The energy scale on the left-hand side applies to the torsion in
[Zn(H2NCOCOO)2] (solid line). The scale on the right-hand side corresponds to the isolated H2NCOCOOϪ ion (dashed line).
(Bu) and Raman (Ag) spectra of [Cu(EtHN-
COCOO)2] reflects its trans square planar
structure [46] established by single-crystal X-ray
crystallography.
3.4. Quantum-chemical calculations
We decided to look at the molecular geometries and
the variation in L11-libration in more detail. Quantum-
chemical calculations were performed at the RHF/
6-31Gءء level using the ab initio program
BRABO [47]. We obtained equilibrium geometries
and torsional energies for an isolated oxamate(-1)
Table 4
Calculated bond distances, bond angles and torsional angles
˚
(distances in A and angles in Њ) for the oxamate(-1) ion and its
ion (H2NCOCOOϪ) and for
a
hypothetical
square planar zinc(II) complex (The atomic labeling scheme for
the oxamate(-1) is similar to that used in the structures of MeHN-
COCOOK and [Cu(EtHNCOCOO)2] (see Figs. 1 and 2).)
square planar Zn(II)/oxamate(-1) complex ([Zn
(H2NCOCOO)2]). We selected the zinc(II)
complex in order to avoid complications caused
by open shell systems. The results of the calcula-
tions are presented in Fig. 4.
The equilibrium geometries of the oxamate(-1)
moiety and its geometry in the Zn(II) complex are
included in Table 4. The difference of 0.02 A between
the O1–C3 and O2–C3 bond distances in the structure
of the free ion reflects the intramolecular N–H···O2
interaction. A very small barrier of 0.250 kcal/mol at
t ˆ 0Њ separates the two energy minima at t ˆ ^16Њ:
This leads at room temperature to an rms-torsional
H2NCOCOOϪ
[Zn(H2NCOCOO)2]
O1–C3
O2–C3
C3–C4
C4–O5
1.217
1.237
1.572
1.202
1.366
131.3
115.7
113.0
125.6
112.3
122.1
16.8
1.263
1.196
1.550
1.227
1.311
130.6
111.8
117.6
119.1
116.3
124.6
0.1
C4–N6
˚
O1–C3–O2
O1–C3–C4
O2–C3–C4
C3–C4–O5
C3–C4–N6
O5–C4–N6
O1–C3–C4–O5
O1–C3–C4–N6
value of 25Њ. The crystalline packing clearly reduces
this torsional variation to a maximum of 7Њ, i.e. L
Ϫ162.8
Ϫ179.9
1=2
11
:

176
G.S. Papaefstathiou et al. / Journal of Molecular Structure 559 (2001) 167–177
At the same time the molecular torsion of 16Њ for a
free moiety shifts to 8Њ in the solid state.
complexes and comparable metal–ligand bond
strengths (as deduced from far-IR spectroscopy).
Obviously, the replacement of the primary amide
function by the secondary amide group has little struc-
tural effect.
A final point of interest is the fact that even though
the syntheses of RHNCOCOOK ꢀR ˆ Me; Et† and
[Cu(EtHNCOCOO)2] took place in an aqueous envir-
onment, these compounds are anhydrous. On the
contrary, H2NCOCOOK·H2O [13] and [Cu(H2NCO-
COO)2]·H2O [6] incorporate a solvent molecule into
their lattices as revealed by X-ray single-crystal crys-
tallography and analytical/vibrational data, respec-
tively. An obvious reason for the anhydrous nature
of the compounds described in this work is the hydro-
Complexation has
a direct impact on the
oxamate(-1) geometry. The changes in the bond
lengths and valence angles that were observed in
the analysis of MeHNCOCOOK and [Cu(EtHN-
COCOO)2] are matched by the calculated shifts. In
the planar zinc(II) complex with the ZnII atom at the
inversion centre, we arrived at Zn–O distances of
˚
1.933 and 2.087 A with an O–Zn–O angle of 81.1Њ.
We calculated the torsional energy profile C3–C4 in
[Zn(H2NCOCOO)2] while preserving the inversion
centre at the ZnII position. The torsional energy
changes practically as a parabolic function with the
deformation in t, i.e. we observe an harmonic poten-
2
¨
tial. With DE ˆ fꢀt Ϫ ͗tꢀ͘† we find ͗t͘ ˆ 0Њ and f is
phobic character of the alkyl groups. Aakeroy and
7 cal/mol degree2. Since U ˆ RT=2f we expect an rms
torsional angle of 6.5Њ, which is four times smaller
than the rms-value calculated for the free oxamate(-1)
coworkers [13] have explained the presence of water
in H2NCOCOOK·H2O in terms of the hydrogen-bond
donor/acceptor ratio. A crystal structure of anhydrous
potassium oxamate would have a surplus of hydrogen-
bond acceptors (three acceptor atoms, two donor
atoms). One way in which potassium oxamate could
achieve a more desirable ratio would be to include a
water molecule, since potassium oxamate monohy-
drate is well balanced with four acceptor atoms and
four donor atoms. This, in combination with the fact
that the small Kϩ ion is capable of interacting with
(coordinating) H2O molecules, is the likely explana-
tion for the appearance of one solvent molecule in
H2NCOCOOK·H2O. The hydrophobic nature of the
ethyl group makes this explanation not valid here,
even though the crystal structure of anhydrous
MeHCOCOOK has a clear surplus of hydrogen-
bond acceptors.
ion. The rms torsional angle for an isolated zinc(II)
1=2
11
complex is about twice the value of L observed in
the solid state.
4. Summary, conclusions and perspectives
The presence study, along with recent results from
our groups and literature reports [35], shows that: (i)
the monoanions of alkyl-substituted oxamic acids
behave as Oamide
, Ocarboxylate-bidentate chelating
ligands; and (ii) vibrational spectroscopy can safely
be used to predict this coordination mode. We believe
that the RHNCOCOOϪ ions are also capable of acting
as bridging ligands and much work still remains to be
done in exploring their structural chemistry. More
recent work has unearthed novel lanthanide(III) and
yttrium(III) complexes in which RHNCOCOOϪ
behave as bridging ligands. Work is also in progress
for the preparation and characterization of N-alkylox-
amato(-2) complexes. The dianions RNCOCOO2Ϫ are
proven to be reactive and excellent springboards into
new cluster and polymer chemistry; our studies,
already well advanced, will be reported soon.
5. Supplementary material
Atomic coordinates, anisotropic thermal para-
meters, and observed and calculated structure factors
for both X-ray structures included in this work are
available from S.P.P. upon request.
The comparison between the coordinating abilities
of EtHNCOCOOϪ (present work) and the simple
oxamate anion, H2NCOCOOϪ [6], towards copper(II)
reveals the same coordination mode for both ligands,
a similar stereochemistry (square planar) in the
Acknowledgements
GSP, HOD and SPP thank NFWO (Belgium) for
financial support. Funding for the purchase of IR and

G.S. Papaefstathiou et al. / Journal of Molecular Structure 559 (2001) 167–177
177
[22] P. Arrizabalaga, P. Castan, J.-P. Laurent, Transition Met.
Chem. 5 (1980) 204.
[23] A.G. Galinos, J.M. Tsangaris, J.K. Kouinis, Z. Naturforsch B
32 (1977) 645.
Raman equipment from NFWO is also gratefully
acknowledged.
[24] J.K. Kouinis, J.M. Tsangaris, A.G. Galinos, Z. Naturforsch B
33 (1978) 987.
References
[25] F. Wallace, E. Wagner, Spectrochim. Acta A 34 (1978) 589.
[26] S.P. Perlepes, Th.F. Zafiropoulos, J.K. Kouinis, A.G. Galinos,
Inorg. Nucl. Chem. Lett. 16 (1980) 475.
[27] S.P. Perlepes, Th.F. Zafiropoulos, J.K. Kouinis, A.G. Galinos,
Z. Naturforsch B 36 (1981) 697.
[28] H. Desseyn, G. Schoeters, Bull. Soc. Chim. Belgium 95
(1986) 13.
[29] H.O. Desseyn, in: S. Patai (Ed.), The Chemistry of Acid Deri-
vatives, vol. 2, Wiley, New York, 1992, pp. 271–303 (chap. 7).
[30] H.O. Desseyn, B.J. Van der Veken, A.J. Aarts, M.A. Herman,
J. Mol. Struct. 63 (1980) 13.
[31] J. De Meulenaer, H. Tompa, Acta Crystallogr. 19 (1965) 1014.
[32] M.L.B.F. Hereygers, H.O. Desseyn, S.P. Perlepes, K.A.F.
Verhulst, A.T.H. Lenstra, J. Chem. Crystallogr. 25 (1995)
181.
[33] C. Vansant, H.O. Desseyn, V. Tangoulis, C.P. Raptopoulou,
A. Terzis, S.P. Perlepes, Polyhedron 14 (1995) 2115.
[34] B. Beagley, R.W.H. Small, Proc. R. Soc. London A 275
(1963) 469.
[35] M. Rossler, W. Schwarz, A. Schmidt, Z. Naturforsch B 34
(1979) 398.
[36] F.L. Hirshfeld, Acta Crystallogr. A 32 (1976) 239.
[37] V. Schomaker, K.N. Trueblood, Acta Crystallogr. B 24 (1968)
63.
[1] H. Chun, B.J. Salinas, I. Bernal, Eur. J. Inorg. Chem. (1999)
723.
[2] A. Aukauloo, X. Ottenwaelder, R. Ruiz, S. Poussereau, Y. Pei,
Y. Journaux, P. Fleurat, F. Volatron, B. Cervera, M. Carmen
Munoz, Eur. J. Inorg. Chem. (1999) 1067.
[3] A. Aukauloo, X. Ottenwaelder, R. Ruiz, Y. Journaux, Y. Pei,
´
E. Riviere, B. Cervera, M. Carmen Munoz, Eur. J. Inorg.
Chem. (1999) 209.
[4] A. Garcia, R. Costa, J. Ribas, Inorg. Chim. Acta 168 (1990)
249.
[5] P.Th. Veltsistas, S.P. Perlepes, M.I. Karayannis, J.M. Tsan-
garis, Monatsh. Chem. 121 (1990) 703.
[6] C. Vansant, H.O. Desseyn, S.P. Perlepes, Transition Met.
Chem. 20 (1995) 454.
[7] G. Schoeters, D. Deleersnijder, H.O. Desseyn, Spectrochim.
Acta A 39 (1983) 71.
[8] V. Baron, B. Gillon, J. Sletten, C. Mathoniere, E. Codjovi, O.
Kahn, Inorg. Chim. Acta 235 (1995) 69.
[9] O. Kahn, Adv. Inorg. Chem. 43 (1995) 179.
[10] H.O. Stumpf, L. Quahab, Y. Pei, P. Bergerat, O. Kahn, J. Am.
Chem. Soc. 116 (1994) 3866.
ˆ
[11] M.G.F. Vaz, L.M.M. Pinheiro, H.O. Stumpf, A.F.C. Alcan-
´
tara, S. Golhen, L. Quahab, O. Cador, C. Mathoniere, O.
Kahn, Chem. Eur. J. 5 (1999) 1486.
[38] F.A. Cotton, G. Wilkinson, C.A. Murillo, M. Bochmann,
Advanced Inorganic Chemistry, 6th ed., Wiley, New York,
1999 (p. 867).
[12] O. Kahn, Y. Pei, Y. Journaux, in: D.W. Bruce, D.O’ Hare
(Eds.), Inorganic Materials, 2nd ed., Wiley, New York,
1996, pp. 66–119 (chap. 2).
[39] A.B.P. Lever, Inorganic Electronic Spectroscopy, 2nd ed.,
Elsevier, Amsterdam, 1984 (pp. 557–559, 565, 566, 570,
571).
[40] M.L.B.F. Hereygers, H.O. Desseyn, K.A.F. Verhulst, A.T.H.
Lenstra, S.P. Perlepes, Polyhedron 15 (1996) 3437.
[41] M.L.B.F. Hereygers, H.O. Desseyn, A.T.H. Lenstra, E.
Manessi-Zoupa, S.P. Perlepes, Transition Met. Chem. 21
(1996) 273.
¨
[13] C.B. Aakeroy, D.P. Hughes, M. Nieuwenhuyzen, J. Am.
Chem. Soc. 118 (1996) 10134.
[14] M.A. Pellinghelli, A. Tiripicchio, M. Tiripicchio-Camellini,
Acta Cryst. B 28 (1972) 998.
[15] A. Braibanti, M.A. Pellinghelli, A. Tiripicchio, M. Tiripic-
chio-Camellini, Acta Crystallogr. B 27 (1971) 1240.
[16] S. Skoulika, A. Michaelides, A. Aubry, Acta Crystallogr. C 44
(1988) 931.
[42] S.P. Perlepes, P. Jacobs, H.O. Desseyn, J.M. Tsangaris, Spec-
trochim. Acta A 43 (1987) 771.
[17] C. Papadimitriou, P. Veltsistas, J. Marek, J.D. Woolins, Inorg.
Chim. Acta 267 (1998) 299.
[43] E. Manessi-Zoupa, S.P. Perlepes, A.C. Fabretti, H.O.
Desseyn, Spectr. Lett. 27 (1994) 119.
[18] S. Skoulika, A. Michaelides, A. Aubry, Acta Crystallogr. C 44
(1988) 808.
[44] G.B. Deacon, R.J. Phillips, Coord. Chem. Rev. 33 (1980) 227.
[45] V. Gutman, The Donor–Acceptor Approach to Molecular
Interactions, Plenum Press, New York, 1978.
[46] K. Nakamoto, Infrared and Raman Spectra of Inorganic and
Coordination Compounds, 4th ed., Wiley, New York, 1986
(p. 464).
[19] P.G. Veltsistas, C.D. Papadimitriou, M. Karayannis, A.M.Z.
Slawin, J.D. Woolins, D.J. Williams, Polyhedron 14 (1995)
1251.
[20] J.M. Tsangaris, P. Veltsistas, Inorg. Nucl. Chem. Lett. 17
(1981) 35.
[21] V. Lazaridou, S.P. Perlepes, J.M. Tsangaris, Th.F. Zafiro-
poulos, J. Less-Comm. Met. 158 (1990) 1.
[47] C. Van Alsenoy, A. Peeters, J. Mol. Struct. 286 (1993) 19.