Synthesis, characterization and dioxygen reactivity of copper(i) complexes with glycoligands

Article abstract of DOI:10.1039/b801109e

A new family of copper(i) complexes with glycoligands containing a central saccharide scaffold, with 2-picolyl ether groups or 2-picolylamine or N-imidazolylamine groups, has been prepared and characterized. For this purpose, the following tetradentate ligands have been synthesized: methyl 2,3-di-O-(2-picolyl)-α-d-lyxofuranoside (L1), 1,5-anhydro-2-deoxy- 3,4-di-O-(2-picolyl)-d-galactitol (L2), 5-(amino-N-(2-salicyl))-5-deoxy-1,2-O- isopropylidene-3-O-(2-picolyl)-α-d-xylofuranose (L3), and 5-(amino-N-(2-salicyl))-5-deoxy-1,2-O-isopropylidene-3-O-(methylimidazol-2-yl) -α-d-xylofuranose (L4). The ligands and the complexes were characterized by elemental analysis, IR, ¹H and ¹³C NMR spectroscopies, ESI mass spectrometry, and cyclic voltammetry. Collaterally with the experimental work, HF-DFT(B3LYP/6-31G*) computations were performed to obtain additional structural information. The Cu(i) complexes are found to be pentacoordinated. The redox properties and the O₂-reactivity of the Cu^ILn complexes have been studied. Reactions of Cu(i) complexes with dioxygen in ethanol yield stable Cu(ii) complexes as confirmed by UV-visible spectrophotometry and EPR spectroscopy. The Royal Society of Chemistry 2008.

Full text of DOI:10.1039/b801109e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Synthesis, characterization and dioxygen reactivity of copper(I) complexes
with glycoligands†‡
Ziad Damaj,a Federico Cisnetti,c Laurent Dupont,a Eric Henon,d Clotilde Policar*c and Emmanuel Guillon*a,b
Received 21st January 2008, Accepted 28th March 2008
First published as an Advance Article on the web 7th May 2008
DOI: 10.1039/b801109e
A new family of copper(I) complexes with “glycoligands” containing a central saccharide scaffold, with
2-picolyl ether groups or 2-picolylamine or N-imidazolylamine groups, has been prepared and
characterized. For this purpose, the following tetradentate ligands have been synthesized: methyl
2,3-di-O-(2-picolyl)-a-D-lyxofuranoside (L1), 1,5-anhydro-2-deoxy-3,4-di-O-(2-picolyl)-D-galactitol
(L2), 5-(amino-N-(2-salicyl))-5-deoxy-1,2-O-isopropylidene-3-O-(2-picolyl)-a-D-xylofuranose (L3), and
5-(amino-N-(2-salicyl))-5-deoxy-1,2-O-isopropylidene-3-O-(methylimidazol-2-yl)-a-D-xylofuranose
(L4). The ligands and the complexes were characterized by elemental analysis, IR, 1H and 13C NMR
spectroscopies, ESI mass spectrometry, and cyclic voltammetry. Collaterally with the experimental
work, HF-DFT(B3LYP/6–31G*) computations were performed to obtain additional structural
information. The Cu(I) complexes are found to be pentacoordinated. The redox properties and the
O2-reactivity of the CuILn complexes have been studied. Reactions of Cu(I) complexes with dioxygen in
ethanol yield stable Cu(II) complexes as confirmed by UV-visible spectrophotometry and EPR
spectroscopy.
refers to positive package/product/environment interaction. This
Introduction
is one of the most innovative and promising areas, in recent years,
Investigations of the reactivity of copper(I) complexes with
in packaging in general and in food packaging in particular. In
molecular dioxygen is a worthy endeavour from the point of view
active packaging the product, the package and the environment
of understanding of the active intermediate species involved in
interact in a positive way, resulting in either an extension of the
copper–protein O2 binding/activation and for “green” practical
product’s shelf-life or the attainment of some specific property
reagents or catalysts for substrate oxidations. The reactions of
that cannot be obtained by other means. Such interactions are
beneficial and are therefore sought after.8 Many products are
sensitive to oxygen and deteriorate in its presence. Modified
atmosphere packaging (MAP) of foods, developed about two
decades ago, is capable of providing, in some cases, a partial
solution to this problem. In MAP, air is replaced by a mixture
of gases (N2, CO2, and small amount of O2) at partial pressures
different from that of air. For many foods, the levels of residual
oxygen that can be achieved by regular MAP technologies are
too high for maintaining the desired quality and for achieving
the sought shelf-life. Oxygen scavengers are capable of reducing
the oxygen concentration in a package to very low levels. Most
of the commercially available oxygen scavengers contain iron as
the oxygen absorber and are marketed in the form of a sachet.
They require water to activate them.8 Recently, oxygen-scavenging
films have been developed as well. Some of these films contain iron
in their structure, while others are based on organic compounds
that absorb oxygen after being activated, normally by ultra-violet
light.9–11
Keeping this aim in mind we propose to develop a new family
of oxygen scavengers based on copper(I) complexes with ligands
derived from natural molecules. These complexes will be then
incorporated in organic polymer. Some of us have recently initiated
a novel strategy using saccharides as central scaffolds to generate
ligands for transition metal cations (named glycoligands).12–15
This “glycoligand strategy” is original in that it uses sugars as
a distribution frame to generate a polydentate chelating claw
Cu(I) complexes with O2 and the oxidative properties of the
resulting Cu/O2 complexes have attracted much interest during the
past decades because of their potential relevance to biochemical
systems1,2 and synthetic catalysis.3,4 A great deal of knowledge has
been obtained from these studies and a number of structural and
reactivity types are now well established.5–7
Another way of application of such complexes is the packaging
industry. Indeed, copper(I) complexes can be used as oxygen
scavenger. Oxygen scavenging is part of “active packaging”, which
aInstitut de Chimie Mole´culaire de Reims, Equipe Chimie de Coordi-
nation, Universite´ de Reims Champagne-Ardenne, UMR CNRS 6229,
Faculte´ des Sciences, B.P. 1039, F-51687, Reims cedex 2, France. E-mail:
emmanuel.guillon@univ-reims.fr; Fax: +33 (0)3 26 91 32 43
bESIEC (Ecole Supe´rieure d’Inge´nieurs en Emballage et Conditionnement),
Poˆle Technologique Henri Farman, B.P. 1029, F-51686, Reims cedex 2,
France
cInstitut de Chimie Mole´culaire et des Mate´riaux d’Orsay, Equipe de Chimie
Bio-organique et Bio-inorganique, Universite´ Paris-Sud 11, UMR CNRS
8182, F-91405, Orsay cedex, France. E-mail: cpolicar@icmo.u-psud.fr;
Fax: +33 (0)1 69 15 72 81
dInstitut de Chimie Mole´culaire de Reims, Equipe BSMA, Universite´ de
Reims Champagne-Ardenne, UMR CNRS 6229, Faculte´ des Sciences, B.P.
1039, F-51687, Reims cedex 2, France. E-mail: eric.henon@univ-riems.fr;
Fax: +33 (0)3 26 91 84 97
† Z. D., F. C. and C. P. for ligand design and synthesis; Z. D., L. D. and
E. G. for synthesis of the complexes and physico-chemical studies.
‡ Electronic supplementary information (ESI) available: Fig. S1, S2, S3,
ZPE corrected potential energies and Cartesian coordinates of all minima
obtained within the water PCM model. See DOI: 10.1039/b801109e
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 3235–3245 | 3235
©

View Article Online
for transition metal cations. The glycoligands show interesting
characteristics, such as a fine-tuning of the geometry at the metal
centre, which is stereogenic.13,14
In this paper, we report the synthesis, characterization, and
oxygenation reactivities of Cu(I) complexes with such glycoligands.
A computational approach helped us to examine the structure
of the complexes. The coordination structure shows two five-
membered chelate rings or one five- and one six-membered chelate
rings fused by the furanoside or the pyranoside framework. A
discussion about the influence of the various substituents on the
saccharide scaffold on the redox properties and on the dioxygen
reactivity is provided.
Results and discussion
Synthesis of the glycoligands
Four glycoligands have been synthesized (see Scheme 1). Here
are presented two glycoligands of first generation (viz. L1 and
L2), derived from monosaccharides by simple alkylation of
selected hydroxyl groups by Lewis bases such as pyridine. Efficient
alkylation phase transfer protocols derived from a benzylation
protocol,16 have been used. Two sets of conditions are described:
a liquid–liquid phase transfer toluene/water and a liquid–solid
toluene/DMSO13 phase transfer. These first-generation glycoli-
gands are characterized by a chelating unit involving 2-picolyl
ether groups.12–15 In order to improve binding strength with metal
cations, a second generation of glycoligands have been designed,
that involve amine groups instead of ether groups (see Scheme 1,
L3, L4). In order to do so, the xylo azido-alcohol E17 was alkylated
using a liquid–liquid phase transfer protocol. The azide was then
reduced. Functionalization of the amine function was achieved
by reductive amination. The synthetic pathways are described in
Scheme 1.
Synthesis and characterization of Cu(I) complexes
The copper(I) complexes were readily prepared by mixing a
copper(I) salt (Cu(CH3CN)4)(PF6)) methanolic solution with
an equimolar amount of the glycoligand in the same solvent
under anaerobic conditions. These Cu(I) complexes are very air-
sensitive, and in ethanolic solution they are rapidly oxidized to
Cu(II) complexes upon exposure to air at room temperature. The
complexes were isolated as colourless microcrystalline solids for
CuIL1 and CuIL2, and as oils for CuIL3 and CuIL4 in good yield
(higher than 90%). As the complexes were extremely sensitive
to air, it was necessary to handle them in a glove box. The
Cu(I) complexes were soluble in polar organic solvents such as
ethanol or acetonitrile. Each complex has been characterized
by elemental analysis, IR spectroscopy, mass spectrometry, and
cyclic voltammetry. Unfortunately, all attempts to obtain well-
formed crystals suitable for X-ray determination have failed. The
IR spectra of the complexes are consistent with the presence of
Scheme 1 Synthetic pathways for Ln. Reagents and conditions: L1:
(i) Picolylation protocol 1: 2-picolyl chloride hydrochloride,
toluene/NaOHaq 50%, NBu4HSO4, tert-amyl alcohol, overnight;
(ii) HClaq, EtOH, overnight. L2: (iii) TrCl, py, DMAP, 50 ◦C, overnight;
(iv) picolylation protocol 2: 2-picolyl chloride, DMSO–toluene, tert-amyl
alcohol, NaOH/K2CO3, overnight. L3/L4: (v) alkylation protocol
1: 2-picolyl chloride hydrochloride or methyl(N-methylimidazol-2-yl)
chloride hydrochloride, toluene/NaOHaq 50%, NBu4HSO4, tert-amyl
alcohol, overnight; (vi) MeOH, Pd/C, H2 (5 bar), 3h (vii) salicylaldehyde,
EtOH, NaBH4, 0 ◦C, 30 min.
allowed us to propose a five-coordinated copper(I) ion with one
oxygen atom from a water molecule and the nitrogen and oxygen
atoms of the two picolyl moieties for CuIL1 and CuIL2. For
CuIL3 and CuIL4, the phenolic oxygen atom, maybe involved
in the coordination, is protonated as underlined on the UV-
visible spectra (Fig. 1), which do not present any phenolate
specific transition around 400–450 nm. The deprotonation of the
phenolic group was underlined by the addition of an equimolar
of strong base on the two complexes CuIL3 and CuIL4, which
leads to the formation of CuILH−1 species. The p→p* transition
coordinated picolyl (mC= at 1615 cm−1).12 Satisfactory elemental
N
analysis and mass spectra were obtained for all the complexes,
which confirm the formation of 1 : 1 complexes (metal to ligand
ratio) consistent with monocationic [CuILn(H2O)]+,PF6 species
−
(Ln corresponds to the ligand in its neutral form, i.e. with the
phenol moiety protonated in the case of L3 and L4). These data
3236 | Dalton Trans., 2008, 3235–3245
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Fig. 2 Pentacoordinate distorted square pyramidal structures for CuIL1
and CuIL2; HF-DFT (B3LYP/6–31G*/PCM) optimized geometries.
˚
to range from 2.112 to 2.511 A. In both cases, the two distances
from the copper(I) ion to the O atoms of the glycoligand are
I
I
˚
substantially different (by 0.09–0.16 A for Cu L1 and Cu L2,
respectively) reflecting an asymmetry of the copper bonding
environment around the sugar. The main difference between these
two copper(I) complexes is that they exhibit this asymmetry with
an opposite sign. Due to the conformational freedom of the
hydroxyl group a short hydrogen-bond (H · · · O distance 1.767–
I
I
˚
1.789 A for Cu L1 and Cu L2, respectively) is formed with the
water molecule. Non hydrogen-bridged conformers have also been
found, with slightly higher energies.
Concerning the CuIL3 and CuIL4 complexes, in the absence of
X-ray determination, the question was whether the hydroxyl of
the phenol moiety could interact with the metal ion. That is the
reason why the oxygen atom was placed around the copper ion
before starting the geometry optimizations. The resulting bond
distances and angles about the copper center are consistent with
a pentacoordinate distorted square-pyramidal structure (Fig. 3)
involving the protonated phenolic oxygen atom. However, the Cu–
Fig. 1 UV-visible spectra in aqueous solution of CuIL4 and L4, 3.70 mM,
underlining the absorption transition of the phenolate moiety in presence
of an equivalent of strong base.
corresponding to the phenolate moiety was detected at 404 nm
(Fig. 1, top). This phenomenon is also confirmed by the UV-
visible spectra of the ligand alone (Fig. 1, bottom). Indeed, the
ligand spectrum shows the appearance of the p→p* phenolate
transition at 403 nm in presence of an equivalent of strong base.
O bond distance (involving the phenolic oxygen atom) in CuIL3
I
˚
˚
(2.875 A) significantly differs from that found in Cu L4 (2.567 A).
Internal rotations of the phenol ring were then performed to
move the hydroxyl group away from the metal center in the
CuIL3 complex. Another minimum, of slightly higher energy (by
about 0.5 kcal mol−1), was then found, which underlines that the
interaction between the phenolic oxygen atom and the copper(I)
DFT calculations
In parallel with the experimental work, we performed DFT
molecular orbital computations to obtain further information
about the structure of the complexes. The details of the quantum
mechanical calculations are reported in the Experimental section.
Starting from a pentacoordinate environment (suggested by our
experimental results) with a water molecule occupying an apical
position, each geometry was first optimized within the gas-
phase model. Solvent effects on the structures were subsequently
assessed by the water solvent polarized continuum model18–20
(PCM) (see computational details in the Experimental section).
Comparison with the gas-phase results indicates that the non-
specific interaction of the solvent exerts no substantial effect on
the geometrical parameters of all complexes. The PCM results are
presented.
ion is relatively weak in the present CuIL3 complex. The Cu–
N distances (1.896–2.019 A) in Cu L3 are very similar to those
in Cu L4 complex (1.893–2.041 A), while the distances between
I
˚
I
˚
the copper(I) ion and the sugar oxygen atom are significantly
I
I
˚
˚
different for Cu L3 (2.216 A) and Cu L4 (2.384 A). It is to be
I
˚
noticed that hydrogen bonding (1.888 and 1.860 A for Cu L3
and CuIL4, respectively) is predicted between the water molecule
The solvent-optimized geometries of CuIL1 and CuIL2 are
shown in Fig. 2. For these two complexes, the computational study
aims to get an insight into the structures involving a five-coordinate
copper atom. Both complexes are found to adopt a very similar
distorted square pyramidal geometry. The Cu–N distances are in
Fig. 3 Pentacoordinate distorted square pyramidal structures for CuIL3
and CuIL4; HF-DFT (B3LYP/6-31G*/PCM) optimized geometries.
˚
the range 1.920–1.938 A while the Cu–O distances are predicted
This journal is The Royal Society of Chemistry 2008
Dalton Trans., 2008, 3235–3245 | 3237
©

View Article Online
and the ring oxygen atom as depicted Fig. 3. The distance
couples. The oxidation potential values, which follow the order:
CuIL4 < CuIL3 < CuIL1 < CuIL2, are in agreement with the
increasing electron-withdrawing effect of the ligands. Indeed, the
more electron rich copper(I) complexes with better donor ligands
are easier to oxidize than those with electron-deficient ligands.22
Turning to using the results of our theoretical study, we consider
the highest occupied molecular orbital (HOMO) energy, since it
is clear that this property should be physically correlated with
oxidation potential.
While Koopmans theorem (the first ionization energy of a
molecule is equal to the energy of the HOMO) does not apply
to DFT HOMO eigenvalues or to orbital eigenvalues from a
solvated wave function, they can still often be correlated with
ionization potentials.23 The negative HOMO energies (eV) found
at the B3LYP/6-31G*/PCM level of theory for L4, L3, L1 and
L2 are 4.60, 5.02, 5.40 and 5.39, respectively. This relative order
is preserved when doing a single point energy calculation at the
B3LYP/6-311G** level of theory. According to this reasonable
qualitative theoretical approach, the ionization potentials of L3
and L4 should be the smallest ones, supporting the experimental
findings. Further analysis would require more accurate predicting
methods of redox potential (accounting for the thermal and the
aqueous free energy ofsolvation contributions). Moreover, such an
electrochemical behaviour (electrochemically irreversible waves)
is difficult to interpret and hinders a more detailed comparison
between the electronic properties of the complexes.
between the copper(I) ion and the apical water oxygen atom
I
˚
is very similar (about 2.09 A) than that found in Cu L1 and
I
˚
Cu L2 complexes (about 2.12 A). All these results favourably
agree with the experimentally suggested five-coordinate copper
center. However, other coordination arrangements cannot be
rigorously ruled out. As above-mentioned, the starting geometries
used for the optimization of CuIL3 and CuIL4 were chosen
within a pentacoordinate copper surrounding, which involves the
phenolic oxygen atom. Supplementary calculations using other
initial guess geometries suggest that copper might tend to adopt
a distorted tetrahedral geometry of higher energy (as shown in
Fig. S1, ESI†). But there is only small energetic differences between
these geometries and the presented pentacoordinate structures
(< 0.5 kcal mol−1 at the B3LYP/6-31G* level of theory including
zero-point vibrational energy correction). On the other hand,
optimizations starting from the above presented five-coordinated
geometries (Fig. 3) and interchanging the moiety phenol with a
water molecule at the apical position (this can be achieved because
the phenol chelating group possesses internal rotational degrees of
freedom) lead to new minima about 6 kcal mol−1 higher in energy
(see Fig. S2, ESI†).
Electrochemistry
The redox potential is a key point in the electronic transfer
processes. It notably depends on the electronic properties of the
ligands. Thus, the electrochemical behaviours of the complexes
were studied by cyclic voltammetry (Fig. 4). All potentials
were measured in acetonitrile with tetrabutylammonium hex-
afluorophosphate as the supporting electrolyte and internally
referenced to the ferrocenium/ferrocene redox couple. Potentials
were reported vs. the Fc+/Fc couple.21 The cyclic voltammogram
revealed an electrochemically irreversible oxidation of the Cu(I)
complexes (Table 1), which could be assigned to the CuIILn–CuILn
Dioxygen reactivity
We have investigated the reactivity of the copper(I) complexes
towards dioxygen in some detail. These experiments were per-
formed in ethanolic solution and were monitored using electronic
absorption and EPR spectroscopies, and ESI-mass spectrometry.
The presence of dioxygen in solution resulted in an apparent colour
change from light yellow to a more or less dark green, which
indicates that dioxygen reacts with copper(I) species to lead to a
copper(II) species. The absorption spectral change in the reaction
of the Cu(I) complexes with dioxygen is shown in Fig. 5. The
oxidation process leads to a hyperchromic phenomenon with an
increase of the absorbance. These dark green Cu(II) species showed
a single intense broad d–d band in the visible region close to
720 nm (717 (e = 126), 704 (e = 149), 709 (e = 264), 742 nm (e =
176 mol−1 dm3 cm−1), corresponding to CuIIL1, CuIIL2, CuIIL3
Fig. 4 Cyclic voltammogram of complex CuIL3, 10−3 mol L−1, in CH3CN
containing 0.10 mol L−1 [nBu4N][PF6] at a scan rate of 100 mV s−1
.
Table 1 Results from cyclic voltammetry experiments on solutions of
Cu(I) complexes
CuL1
−0.40
CuL2
−0.18
CuL3
−0.57
CuL4
−0.64
Epaa/V (vs. Fc+/0
)
Fig. 5 Development of optical absorption spectra of CuIL3, 3.41 mM,
upon oxygenation in ethanolic solution at room temperature.
a According to Fc+/0 = 0.40 V vs. SCE.18
3238 | Dalton Trans., 2008, 3235–3245
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
and CuIIL4, respectively). In the case of CuIIL3 and CuIIL4 a band
around 460–470 nm appears, which is assigned to a phenolate to
Cu(II) ligand-to-metal charge transfer (LMCT).24 The transition
around 720 nm (e.g. ∼14 000 cm−1), that is unsymmetrical, is
consistent with a tetragonal distorted octahedral geometry with
a CuN2O4 chromophore25 and confirms the formation of Cu(II)
species.
Moreover, the increase of the transition at 468 nm (Fig. 5),
assigned to the LMCT transition of the Cu(II)–phenolate chro-
mophore, indicates that the oxidation process leads to the depro-
tonation of the phenol moiety through an electron and proton
transfer according to the Scheme 2.
Fig. 6 Evolution of the maximal absorbance of CuILn, 3.50 mM,
complexes in contact with dioxygen at kmax = 717, 704, 709 and 742 nm for
CuL1, CuL2, CuL3 and CuL4, respectively, as a function of time at room
temperature in ethanolic solution.
Scheme 2 Proposed oxidation process (Ln = L3 or L4).
EPR spectroscopy
In order to identify the chemical species present in solutions of
the oxygenated copper(I) complexes, EPR spectra of the solutions
were recorded. Fig. 7 shows as example the 77 K EPR spectrum
of the green solution recorded after the reaction of O2 with CuIL2
in ethanol, and the spectra of the other complexes are similar. The
EPR spectra display a typical anisotropic copper(II) signal with
four lines in parallel region arising from the hyperfine coupling of
the S = 1/2 electron spin of Cu(II) with its nuclear spin I = 3/2.
These spectra reveal that the Cu(II) center is axial (gꢀ > g⊥ > 2.0),
with simulated spin-Hamiltonian parameters reported in Table 2.
The coordination of the Cu(II) metallic cation by the phenolate
moiety is confirmed by the disappearance of the p →p* transition
(404 nm) of the phenolate ion on behalf of a LMCT transition
(468 nm) of the Cu(II)–phenolate chromophore (Fig. 1 top).
In order to identify the chemical species present in solutions
of the oxygenated copper complex species, ESI-MS spectra
were recorded. The spectra revealed positive ions at m/z values
corresponding to [CuIILn] species. This is consistent with the fact
that the dioxygen was not fixed on the copper(II) complexes,
which is in contradiction with most of the publications on this
topic.6 Nevertheless, we can not exclude the possible formation
of intermediate species during the redox process, in which the
dioxygen is coordinated to the complex.
Bubbling nitrogen into an oxygenated solution of copper
complexes, or purging the oxygenated solution with a stream
of argon gas resulted in no noticeable loss and/or change in
absorption intensity of the UV-Vis spectra of the solutions.
Thus, the oxygenation of the complexes was physico-chemically
irreversible. Nevertheless, the addition of a small amount of
sodium borohydride (NaBH4) leads to the discoloration of the
solution, which indicates that the copper(II) complexes may be
reduced to the corresponding copper(I) complexes through NaBH4
reduction. The regenerated copper(I) complex is then active
towards oxygen.
The dioxygen reactivity with the Cu(I) complexes was also
followed by tracing the maximum intensity of the d–d transition
(extracted from Fig. 5) vs time (Fig. 6) This figure underlines two
types of behaviour. CuIL3 and CuIL4 rapidly react with dioxygen
to reach equilibrium after 1 and 4 h, respectively, while CuIL1
and CuIL2 exhibit a slower oxidation kinetics. Indeed, the curves
present two stages. (1) a relatively fast kinetics in which more than
50% of the copper(I) was converted into copper(II) and (2) a slower
kinetics leading to a plateau (corresponding to 100% of Cu(I)
conversion) or a pseudo-plateau after 6 h. These curve shapes are
in agreement with the potential values of the Cu(II)/Cu(I) systems.
Indeed, the fact that the redox potential of CuIL3 and CuIL4 are
lower than those of CuIL1 and CuIL2 (see Table 1) suggested a
higher reactivity of the former towards dioxygen.5
Fig. 7 77 K EPR spectrum of CuIIL2, 4.24 mM, complex generated by
oxygenation of CuIL2 in ethanol at 293 K.
These values lie in the middle of the Peisach and Blumberg
maps26 and most closely correlate with copper in the +2 oxidation
state in tetragonal CuN2O4 structures (gꢀ ∼2.28, Aꢀ ∼(150–180) ×
10−4 cm−1).27 The spin-Hamiltonian parameters are in agreement
with a tetragonal elongation of the octahedron.28,29 Indeed, if
2
2
an axial coordination around copper(II) ion exists, the dx
–y
orbital receives an increase in destabilization (i.e. DE(x2−y2 → xy)
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 3235–3245 | 3239
©

View Article Online
Table 2 EPR parameters of copper(II) complexes in ethanolic solution
wave function displays some contributions of the ring oxygen
atom lone pair orbital to molecular orbitals involving copper
3d functions. Thus, hexacoordinate complexes with a slightly
distorted octahedral geometry are also predicted for CuIIL3 and
CuIIL4. Comparison with the gas-phase results indicates that the
non-specific interaction of the solvent exerts no substantial effect
on the geometrical parameters of all complexes.
gꢀ
g⊥
10−4Aꢀ/cm−1
CuIIL1a
CuIIL2a
CuIIL3a
CuIIL4a
2.297
2.305
2.273
2.268
2.075
2.066
2.080
2.071
178.0
179.7
147.5
173.0
a Values obtained by simulation.
Conclusions
increase), which causes an increase in the g value as can be seen
from eqn (1) and (2):
We have synthesized four copper(I) complexes with tetradentate
glycoligands. These complexes have been isolated and fully charac-
terized. Thus, based on the chemical and spectroscopic evidence,
the structures of CuLn are best described as pentacoordinate,
probably square-pyramidal, and six-coordinate tetragonal Cu(I)
and Cu(II) complexes, respectively. The results of the computa-
tional investigation give theoretical support to the existence of
these structures. In all these species, the Cu(I) ion is coordinated
to two nitrogen and two oxygen atoms from the glycoligands, the
coordination sphere being completed with one water molecule.
In solution, these four compounds behave quite differently.
CuIL1 and CuIL2, which have higher redox potential values, react
more slowly with dioxygen. On the other hand, CuIL3 and CuIL4
react rapidly with dioxygen in accordance with their lowest redox
potential values. These results of the reactivity towards dioxygen
of the Cu(I) complexes are promising for future application in the
packaging field.
Future studies of incorporation of these complexes into organic
polymers and on their reactivity with O2 may further contribute to
our understanding of oxygen scavenger processes. Indeed, data on
absorption capacity of oxygen scavengers and the absorption rate
are required for designing an optimal and cost-effective package
using a scavenger. The capacity and rate of scavenging of copper(I)-
based oxygen scavenger depend also on the environment inside the
package, especially if carbon dioxide is present. The information
provided by oxygen scavenger manufacturers does not always
provide all the information needed for designing a good package.
Much more insight into the action of oxygen scavengers in different
environments is required.
(1)
(2)
where k represents the spin–orbit coupling constant, DE is the
difference in the corresponding state energies, and ge the g factor
of the free electron.
In addition, the value of the ratio gꢀ/Aꢀ, which is approximately
equal to 130 cm (except for CuIIL3, gꢀ/Aꢀ = 154 cm), suggests an
absence of significant dihedral angle distortion in the xy-plane.26
This conclusion is supported by the EPR parameters of tetrahedral
complexes, for which gꢀ/Aꢀ is greater than 180 cm.30
We have also examined the deprotonated CuIILn structures
using DFT molecular orbital computations in order to confirm the
proposed geometries. First, five-coordinate structures analogous
to that found for copper(I) complexes were obtained (see Fig. S3,
ESI†). In these molecules, the phenolate oxygen atom is now
˚
strongly coordinated to the metal ion (about 1.87 A). However,
additional geometries have been found with structural features
more consistent with a six-coordinated tetragonal environment
(Fig. 8) involving the sugar ring oxygen atom. These structures
have nearly the same energy (less stable by only 1 kcal mol−1)
than their pentacoordinate counterpart. They are the analogous
of those obtained for copper(I) by interchanging the phenol moiety
with a water molecule at the apical position (see Fig. S2, ESI†). In
the absence of hydrogen bonding between the ring oxygen atom
and the phenolic hydroxyl hydrogen, the ring oxygen atom gets
Experimental
closer to the metal ion. The distance between copper and this
Materials
II
˚
oxygen atom is not so large (2.642 and 2.542 A for Cu L3 and
CuIIL4, respectively) to rule out possible interactions. Indeed, the
All the solvents were purified by conventional procedures, dis-
tilled and deaerated prior to use. Solutions of the copper(I)
salt (Cu(CH3CN)4)(PF6)) and complexes were handled under
inert atmosphere in a glove box. All the chemicals which were
commercially available (Aldrich or Acros) were used as supplied.
Analytical procedures
Elemental analyses (C, H and N) were carried out on a Perkin-
Elmer 2400 C, H, and N elemental analyzer. The copper analyses
were performed on an ICP-AES Liberty Series II Varian appara-
tus. The IR spectra were recorded from KBr pellets on a Nicolet
Avatar 320. 1H and 13C{ H} NMR spectra were recorded at room
1
temperature on a Bruker AV360 spectrometer in CDCl3. ESI-MS
were recorded on a Finnigan Mat 95S in a BE configuration at
low resolution out on a hybrid tandem quadrupole/time-of-flight
Fig. 8 Six-coordination tetragonal (distorted octahedral) structures
for CuIIL3 and CuIIL4; HF-DFT (B3LYP/6-31G*/PCM) optimized
geometries.
3240 | Dalton Trans., 2008, 3235–3245
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
(Q-TOF) instrument, equipped with a pneumatically assisted
electrospray (Z-spray) ion source (Micromass, Manchester, UK)
operated in positive mode. The electrospray potential was set to
3 kV in positive ion mode and the extraction cone voltage was
usually varied between 30 and 90 V in order to obtain optimized
mass spectra. The solutions of complexes were introduced using
a syringe pump with a flow rate of 5 ll min−1. The peaks
related to the complex species were identified by analyzing their
specific isotopic profile due the different Cu isotopes (e.g. 63Cu
and 65Cu). The stoichiometries of the molecular associations were
determined in accordance with the higher intensity isotopic peak
(e.g. 63Cu). The difference between experimental and calculated
m/z values was less than or equal to 0.1 unit. Complementary
high resolution mass spectra (HRMS) were obtained on a Q-
TOF Micromass positive ESI (CV = 30 V). The UV-Vis spectra
in ethanolic solution were recorded on a Shimadzu UV-2401-
PC spectrophotometer equipped with a temperature-controlled
cell holder TCC-240A. The reactivity of the Cu(I) complexes
towards dioxygen, followed by UV-Vis spectroscopy, was studied
at room temperature in ethanolic solutions in air condition.
The dioxygen comes from the naturally oxygen dissolved at
atmospheric pressure.
resulting spin contamination of the HF-DFT wave functions was
very small (S2 = 0.7525). For reasons of computational cost, a
diffuse function could not be added for these radical species to
describe the excess electron on the deprotonated oxygen atom.
The minima were first optimized in the gas-phase model. As the
synthesis and characterizations of the ligands and complexes were
carried out using a variety of solvents (water, methanol, ethanol,
and acetonitrile with a supporting electrolyte) we have chosen to
estimate solvent effects on structures (full optimization) within
the polarized continuum model (PCM) in the field of water (the
used solvent having the largest dielectric constant, thus the largest
expected effect on the structures within the PCM model). The
solvent accessible surface (SAS) was employed. Comparison with
the gas-phase results indicates that there is no significant solvent
effect on the structures. Vibrational frequencies were determined
within the harmonic approximation both in the gas-phase model
and in the PCM. All minima were characterized by no imaginary
frequencies. All the figures were drawn using Molden39 and VMD40
(http:ꢀwww.ks.uiuc.edu/Research/vmd/) packages.
Synthesis
The fixation of the 2-picolyl chloride or of the 2-imidazolyl
chloride was achieved using protocols derived from that published
for benzylation by Szeja et al.16 Two general alkylation protocols
were used (conditions 1 and 2 described below).
EPR spectroscopy
The anisotropic X-band (9.44 GHz) EPR spectra of frozen
ethanolic solutions were recorded at 77 K using a Bruker ESP
300 spectrophotometer. The EPR spectra were referenced to
2,2-diphenyl-1-picrylhydrazyl (DPPH) (g = 2.0037). All spectra
were recorded using 100 kHz modulation frequency, 0.596 G
modulation amplitude and microwave power of 6 mW. Simulations
of EPR spectrum were performed using the WINSYMPHONIA
Bruker software package.
Condition
1
(solid–liquid)13. The alcohol derivative (ca.
5 mmol, 1 eq.) was dissolved in DMSO (2 mL). The organic
chloride (2-picolyl chloride hydrochloride or 2-imidazolyl chloride
hydrochloride) (1.2 eq. per function to be functionalized) was
suspended in toluene (10 mL). Neutralisation and concomitant
extraction in the toluene phase of the organic chloride were per-
formed by pouring sat. Na2CO3 on this suspension until gaseous
evolution ceased. The aqueous phase was decanted and the organic
phase was added to the alcohol–DMSO solution. 0.1 eq. of
(NBu4)+(HSO4)−, 0.2 mL of tert-amyl alcohol, and a 4 : 1 ground
mixture of K2CO3/NaOH containing a large excess of NaOH per
alcoholic group to be functionalized (ca. 2 to 5 eq.) were added
to the reaction mixture which was vigorously stirred overnight.
TLC analysis (AcOEt–MeOH 9 : 1) showed in all cases the
complete disappearance of the starting alcohol compound. The
product was recovered by extraction (dichloromethane/water).
The organic phase was dried on Na2SO4 and evaporated. The
product was purified by column chromatography (SiO2, elution
gradient: AcOEt to AcOEt–MeOH 9 : 1).
Electrochemistry
Cyclic voltammetry was carried out using an Autolab with PG-
STAT12 potentiostat (ECO Chemie). The three-electrode electro-
chemical cell used in our studies consisted of a glassy carbon disk
working electrode, a silver electrode (separated from the complex
solution) reference electrode, and a platinum plate auxiliary elec-
trode. All sample solutions (around 10−3 mol L−1) were prepared
in acetonitrile with (nBu4N)(PF6) (0.10 mol L−1) as the supporting
electrolyte and were deaerated during 10 min. Chemical potentials
+
were internally referenced to the FeCp2 /FeCp2 redox couple.
Computational details
Calculations were performed using the Gaussian 03 package.31
Density functional theory (DFT) has become a popular choice
for quantum chemical calculations on transition metal complexes.
The structure of copper complexes has suitably been obtained
in previous studies32–34 using the methodology HF-DFT with the
B3LYP hybrid functional and the double-zeta 6-31G basis set
family. In our study, the structures were fully optimized at the HF-
DFT (B3LYP)35–38 level of theory using the analytical gradients
and the all-electrons 6-31G* basis set (thus, adding one set of d
polarization functions on the carbon, oxygen and nitrogen atoms,
and one set of f polarization functions on copper atom). The
unrestricted approach was used for the Cu(II) complexes. The
Condition 2 (liquid–liquid). The alcohol derivative (ca. 5 mmol,
1 eq.) was dissolved in toluene (14 mL). 0.35 mL of tert-
amyl alcohol, NaOHaq 50% (10 mL), (NBu4)+(HSO4)− (0.1 eq.),
and the organic halide (2-picolyl chloride hydrochloride or 2-
imidazolyl chloride hydrochloride) (1.2 to 1.5 eq. per function to be
functionalized) were added. The resulting mixture was vigorously
stirred overnight, after which TLC analysis (AcOEt–MeOH 95 : 5)
showed the complete disappearance of the starting alcohol. The
products were recovered by extracted (dichloromethane/water).
The combined organic phases were dried on Na2SO4 and
evaporated to dryness. The product was purified by column
chromatography (typically, SiO2, AcOEt–MeOH 95 : 5).
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 3235–3245 | 3241
©

View Article Online
Methyl 2,3-di-O-(2-picolyl)-(6-O-trityl)-a-D-lyxofuranoside (B).
Methyl (6-O-trityl)-a-D-lyxofuranoside (A) was synthesized as
previously published41 and then picolylated using an adaptation
of a benzylation protocol by Szeja et al. (conditions 2, see
above).16 A (2.6 g, 6.4 mmol) was picolylated in toluene/NaOHaq
50% (14 mL/10 mL) using 0.35 mL of tert-amyl alcohol,
(NBu4)+(HSO4)− (0.2 g, 0.1 eq.) and 2-picolyl chloride hydrochlo-
ride (2.5 g, 2.4 eq.). The product was purified by column
chromatography (SiO2, AcOEt–MeOH 95 : 5). Yield: 1.7 g (45%).
1H NMR (360 MHz, CDCl3): d (ppm, see Scheme 3) 8.52 (d, 1H,
J = 5.0 Hz, H6py), 8.46 (d, 1H, J = 4.8 Hz, H6py), 7.61 (td, 1H,
J1 = 7,6, J2 = 1,7 Hz, H4pya), 7,47 (m, 7H, H4pyb, 6 × Htr), 7.34 (d,
1H, J = 7.8 Hz, H3pya), 7.19 (m, 12H, H3pyb, 2 × H5py, 9 × HTr ),
5.07 (d, 1H, J = 3.5 Hz, H1), 4.83 and 4.65 (2 × d, 2 × 1H, J =
13.6 Hz, 2 × OCH2Py), 4.46 (m, 1H, H4), 4.36 (quasi-t, J1 ≈ J2 =
4 Hz, H3), 4.07 (m, 1H, H2), 3.57 (dd, J1 = 9.3 Hz, J2 = 6.4 Hz,
H5a), 3.44 (s, 3H, Me), 3.38 (dd, J1 = 9.3 Hz, J2 = 6.1 Hz, H5b).
13C NMR (90 MHz, CDCl3): d (ppm) (158.5, 158.1) C2py, (148.8,
148.6) (CquatTr), 144.0 C6py, (136.6, 136.5) C4py, (128.7, 127.7 and
127.9, Ctr), {122.3 (2C), 122.1, 121.1} (C3py, C5py), 106.5 C1, 84.3
C2, {(78.7 and 78.6) (C3, C4)}, (74.2, 73.4) CH2-C2py, 62.7 C5,
56.0 Me.
1.2 eq.) in pyridine (25 mL) at 50 ◦C with DMAP in catalytic
amount. After one night, TLC analysis showed the disappearance
of C. After addition of 2.5 mL water, the solvent was evaporated.
CH2Cl2 (50 mL) was added and washed with water. The organic
phases were dried on MgSO4 anhydrous and evaporated to dry-
ness. The 1,5-anhydro-2-deoxy-6-O-trityl-D-galactitol was purified
by chromatography (SiO2, from CH2Cl2 to CH2Cl2–MeOH, 95 : 5)
and was used as such in the next step. It was then picolylated using
an adaptation of a benzylation protocol by Szeja et al.16 (condition
1, see above) previously published.13 The picolylation was realized
in toluene–DMSO (10 mL/2 mL) using 0.2 mL of tert-amyl
alcohol, ground NaOH/K2CO3 (1.6 g/6.4 g), (NBu4)+(HSO4)−
(0.17 g, 0.1 eq.), and the solution of neutralized 2-picolyl chloride
hydrochloride (2.1 g, 2.4 eq./1,5-anhydro-2-deoxy-6-O-trityl-D-
galactitol). D was purified by column chromatography (SiO2,
elution gradient: AcOEt to AcOEt–MeOH 95 : 5). Yield: 1.73 g
(56%). 1H NMR (360 MHz, CDCl3): d (ppm, see Scheme 4) 8.45
(d, 1H, J = 4.3 Hz, H6pya), 8.34 (d, 1H, J = 4.2 Hz, H6pyb), 7.70
(td, 1H, J1 = J2 = 4.0 Hz, H4pya), 7,47 (m, 7H, H4pyb, 6 × Htr),
7.33 (d, 1H, J = 12.6 Hz, H3pya), 7.18 (m, 12H, H3pyb, 2 × H5py,
9 × HTr ), {5.06 (d, 1H, J = 13.7 Hz), 4.92 (d, 1H, J = 13.7 Hz)
OCH2py}, {4.76 (d, 1H, J = 13.4 Hz), 4.72 (d, 1H, J = 13.4 Hz)
OCH2py}, 4.12 (m, 1H, H4), 4.08 (ddd, 1H, J1 = 1.5 Hz, J2 =
4.6 Hz, J3 = 11.4 Hz, H1a), 3.92 (d, 2H, J = 7.2 Hz, H6), 3.72
(ddd, J1 = 2.5 Hz, J2 = 4.3 Hz, J3 = 11.7 Hz, H3), {3.47 (m, 2H)
H1b and H5}, 2.12 (qd, 1H, J1 ≈ J2 ≈ J3 ≈ 12 Hz, J4 = 4.6 Hz,
H2a), 1.95 (ddd, J1 = 1.61 Hz, J2 = 4.2 Hz, J3 = 12.5 Hz, H2b)
13C NMR (90 MHz, CDCl3): d (ppm) (158.6, 158.4) C2py, (148.7,
148.5) C6py, 145.3 (CquatTr), (137.5, 137.1) C4py, (129.4, 128.3, 128.6)
Ctr, (122.9, 122.7, 122.4, 121.5) (C3py, C5py), 79.8 C3, 78.9 C5, 76.2
C4, (75.2, 71.5) CH2C2py, 66.4 C1, 60.9 C6, 27.6 C2.
Methyl 2,3-di-O-(2-picolyl)-a-D-lyxofuranoside (L1). B (1.7 g,
2.9 mmol) was dissolved in EtOH (95%) (50 mL). HCl aqueous
solution (1M, 10 mL) was added and the resulting solution was
stirred at ambient temperature overnight. Ethanol was evaporated.
Water was added (ca. 70 mL) and the precipitate of triphenyl-
carbinol was filtered off. 60 mL of CHCl3 was added to the aqueous
solution that was then basified up to pH 11–12 by Na2CO3 and
extracted twice further. The joined organic phases were dried
on Na2SO4 and evaporated to provide an oil that was purified
by column chromatography (SiO2, AcOEt–MeOH, 9 : 1). Yield:
0.45 g (72%). Microanalysis (calc./found) C18H22N2O5·H2O: C
59.33/59.25, H 6.08/6.12, N 7.69/7.71. 1H NMR (360 MHz,
CDCl3): d (ppm, see Scheme 3) 8.56 (d, 2H, J = 4.8 Hz, 2 ×
H6py), 7.70 (m, 2H, 2 × H4py), {7.50 (d, 1H, J = 7.8 Hz) and 7.30
(d, 1H, J = 7.9 Hz) 2 × H3py}, 7.23 (m, 2H, 2 × H5py), 5.10 (d,
1H, J = 2.6 Hz, H1), {4.94 (d, 1H, J = 14.0 Hz) and 4.8 (m, 3H),
2 × OCH2Py}, 4.44 (quasi-t, J1 ≈ J2 ≈ 5 Hz, H3), 4.36 (m, 1H,
H4), 4.07 (dd, 1H, J1 = 4.6 Hz, J2 = 2.6 Hz, H2), 4.00 (dd, J1 =
11.4 Hz, J2 = 7.3 Hz, H5a), (dd, J1 = 11.4 Hz, J2 = 4.61 Hz, H5b),
3,41 (s, 3H, Me}. 13C NMR (90 MHz, CDCl3): d (ppm) = (158.6,
158.0) C2py, (148.6, 148.5) C6py, (137.0, 136.7) C4py, {122.6, 122.3,
121.6 (2C) (C3py, C5py)}, 99.5 C1, 83.1 C2, 79.2 C3, 78.7 C4, (73.9,
73.2) CH2C2py, 60.5 C5, 55.6 Me. MS-ES : 369.2 (100%) [M + Na].
1,5-Anhydro-2-deoxy-3,4-di-O-(2-picolyl)-D-galactitol (L2).
D
(1.57 g, 2.7 mmol) was dissolved in EtOH (95%) (50 mL). HCl
aqueous solution (1 M, 10 mL) was added and the resulting
solution was stirred at ambient temperature overnight. Ethanol
was evaporated. Water was added (60 mL) and the remaining
precipitate of triphenylcarbinol was filtered off. 100 mL of CHCl3
was added to the aqueous phase that was then basified up to
pH 11–12 by Na2CO3 and extracted twice further. The joined
organic phases were dried over Na2SO4 and evaporated to provide
L2 as an oil. Yield 0.68 g (77%). Microanalysis (calc./found)
C18H22N2O4·H2O: C 62.05/61.91, H 6.36/6.46, N 8.04/7.98%. 1H
NMR (360 MHz, CDCl3): d (ppm, see Scheme 4) 8.57 (m, 2H,
H6py), 7.70 (2H, m, H4py), 7.51 (d, J = 7.8 Hz, 1H, H3pya), {7.23
(m, 3H, J = 7.9 Hz) (H3pyb, H5py)}, {5.11 (d, 1H, J = 13.8 Hz),
4.93 (d, 1H, J = 13.8 Hz) OCH2py}, {4.81 (d, 1H, J = 13.5 Hz),
4.75 (d, 1H, J = 13.5 Hz) OCH2py}, 4.30 (m, 1H, H4), 4.09 (ddd,
1H, J1 = 1.6 Hz, J2 = 4.8 Hz, J3 = 11.5 Hz, H1a), 3.76 (d, 2H,
Scheme 3 Proton assignments.
1,5-Anhydro-2-deoxy-3,4-di-O-(2-picolyl)-6-O-trityl-D-galacti-
tol (D). C (1,5-anhydro-2-deoxy-D-galactitol) was synthesized as
previously published.17 C (1.5 g, 10.1 mmol) was tritylated on the
6-position by a conventional protocol using trityl chloride (3.7 g,
Scheme 4 Proton assignments.
3242 | Dalton Trans., 2008, 3235–3245
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
J = 7.1 Hz, H6), 3.70 (ddd, J1 = 2.6 Hz, J2 = 4.5 Hz, J3 = 11.8 Hz,
H3), 3.47 (m, 2H, H1b and H5), 2.22 (qd, 1H, J1 ≈ J2 ≈ J3 ≈
12 Hz, J4 = 4,8 Hz, H2a), 1.88 (ddd, J1 = 1.6 Hz, J2 = 4.3 Hz,
J3 = 12.5 Hz, H2b). 13C NMR (92 MHz, CDCl3): d (ppm) (158.7,
158.6) C2py, (149.1, 149.0) C6py, (137.2, 136.7) C4py, (122.8, 122.5,
122.3, 121.2) (C3py, C5py), 79.5 C3, 78.6 C5, 75.8 C4, (75.0, 71.3)
CH2C2py, 66.2 C1, 60.4 C6, 27.2 C2. MS-ES: 331.2 (4%) [M + H];
353.2 (100%) [M + Na].
(OCH2py), 40.0 (C5), {26.7 and 26.2 (MeIp)}. MS-ES: 281.2 (5%)
[M + H] ; 303.1 (100%) [M + Na].
5-(Amino-N-(2-salicyl))-5-deoxy-1,2-O-isopropylidene-3-O-(2-
picolyl)-a-D-xylofuranose (L3). G (500 mg, 1,78 mmol, 1 eq.) was
dissolved in 15 mL absolute ethanol. Salicylaldehyde (217 mg,
1.78 mmol, 1 eq.) was added, which caused the immediate appear-
ance of a yellow colour. The solution was cooled (water-ice bath),
NaBH4 (67 mg, 1.77 mmol, 1 eq.) was added inducing the disap-
pearance of the yellow coloration. The reaction mixture was stirred
more than 30 minutes at RT. Excess NaBH4 was destroyed with
acetic acid. The solvent was evaporated and L3 was purified by
chromatography (AcOEt–MeOH/NH3, v/v/v = 95/5/2). Yield:
420 mg (61%). Microanalysis (calc./found) C21H26N2O5·0.4H2O :
C 64.07/64.12, H 6.86/6.81, N 7.12/6.35%. 1H NMR (300 MHz,
CDCl3): d (ppm, see Scheme 5) 8.30 (1H, d, J = 4.3 Hz, H6py),
7.61 (1H, td, J1 = 7.7 Hz, J2 = 1.6 Hz, H4py), 7.24 (1H, d, J =
7.7 Hz, H3py), {7.12 (m, 2H), 6.96 (d, 1H, J = 6.6 Hz), 6.77 (m,
2H), (HPhOH s, H5py)}, 5.92 (d, 1H, J = 3.8 Hz, H1), {4.76 and 4.58
(2 × d, 2 × 1H, J = 13.4 Hz, OCH2py)}, 4.65 (d, 1H, J = 3.8 Hz,
H2), 4.37 (td, J1 = 7.0 Hz, J2 = 3.3 Hz, H4), {4.05 and 3,92 (2 ×
d, 2 × 1H, J = 13.9 Hz, NHCH2PhOH}, 3.9 (d, J = 3.3 Hz, H3),
5-Azido-5-deoxy-1,2-O-isopropylidene-3-O-(2-picolyl)-a-D-xylo-
furanose (F). 5-Azido-5-deoxy-1,2-O-isopropylidene-a-D-xylo-
furanose (E) was synthesized as previously published.17 The
condition 2 were applied. E (1.29 g, 6 mmol) was picolylated
in toluene/(NaOHaq 50%) (14/10 mL) using 2-picolyl chloride
hydrochloride (1.48 g, 9 mmol, 1.5 eq./E), 0.2 mL of tert-amyl
alcohol, and NBu4HSO4 (0.19 g, 0.6 mmol, 0.1 eq.). After
treatment (see above), F was purified by chromatography
(SiO2, from CH2Cl2 to CH2Cl2–MeOH, 9 : 1). Yield: 1.22 g
(66%). Microanalysis: calc./found for C14H18N4O4·1/3H2O: C
54.10/54.06, H 6.00/5.78, N 17.94/17.69%. 1H NMR (250 MHz,
CDCl3): d (ppm, see Scheme 5) 8.53 (d, 1H, J = 4.8 Hz, H6py),
7.69 (td, 1H, J1 = 7.7 Hz, J2 = 1.6 Hz, H4py), 7.37 (d, 1H, J =
7.7 Hz, H3py), 7.19 (m, 1H, H5py), 5.92 (d, 1H, J = 3.7 Hz, H1),
{4.78, (d, 1H, J = 12,9 Hz), 4,65 (m, 2H) OCH2py, H2}, 4.32 (m,
1H, H4), 4.02 (d, J = 3.1 Hz, H3), 3.57 (m, 2H, H5a and H5b),
{1.48 and 1.30 (2 × s, 2 × 3H, MeiP)}. 13C NMR (62.5 MHz,
CDCl3): d (ppm) 157.3 (C2py), 149.2 (C6py), 136.8 (C4py), {122.8,
121.5} (C3py, C5py), 111.9 (CquatIp), 105.1 (C1), {82.2, 81.2} (C2,
C4), 78.6 (C3), 72.6 (OCH2py), 49.2 (C5), {26.8, 26.2} (MeIp).
MS-ES: 329.1 (100%) [M + Na].
2.98 (m, 2H, H5a, H5b), {1.47 and 1.30 (2 × s, 2 × 3H, MeIp)}. 13
C
NMR (75 MHz, CDCl3): d (ppm) {158.2 and 157.1 (C1PhOH and
C2py)}, 149.1 (C6py), 136.8 (C4py), {128.7, 128.5, 122.8, and 121.6
(C3py, C5py, C3PhOH, C5PhOH)}, 126.1 (C2PhOH), {119.0 and 116.3,
(C6PhOH, C4PhOH)}, 111.7 (CquatiP), 104.9 (C1), {82.4 and 82.1 (C2,
C4)}, 78.8 (C3), 71.8 (OCH2Py), 52.6 (NHCH2PhOH), 48.6 (C5),
{26.7 and 26.2 (MeIp)}. MS-ES: 409.2 (100%) [M + Na].
5-Azido-5-deoxy-1,2-O-isopropylidene-3-O-(N-methylimidazol-
2-yl)-a-D-xylofuranose (H). H was synthesized using the same
protocol as F using 2-imidazolyl chloride hydrochloride42 instead
of 2-picolyl chloride hydrochloride and E. E (1.84 g, 8.55 mmol)
was imidazolated in toluene/(NaOHaq 50%) (14/10 mL) using 2-
imidazolyl chloride hydrochloride (1.71 g, 10.26 mmol, 1.2 eq./
E), 0.35 mL of tert-amyl alcohol and NBu4HSO4 (0.28 g,
0.85 mmol, 0.1 eq.). After treatment (see above), H was purified
by chromatography (SiO2, elution gradient: AcOEt to AcOEt–
MeOH 95 : 5). Yield: 1.50 g (57%). 1H NMR (250 MHz, CDCl3):
d (ppm, see Scheme 6) 6.73 (1H, d, J = 6.9 Hz, H3Im), 6.70 (1H, d,
J = 6.9 Hz, H4Im), 5.64 (1H, d, J = 3.6 Hz, H1), {4.49 (d, 1H, J =
12.4 Hz), 4.46 (m, 2H) (OCH2Im, H2)}, 4.33 (m, 1H, H4), 4.06 (d,
J = 2.9 Hz, H3), 3.46 (s, 3H, MeIm), 3.26 (dd, J1 = 12.0 Hz, J2 =
7.4 Hz, H5a), 3.11 (dd, J1 = 12.0 Hz, J2 = 4.8 Hz, H5b), {1.25
and 1.08 (2 × s, 2 × 3H, MeiP)}. 13C NMR (90 MHz, CDCl3):
d (ppm) 143.4 (C2Im), 127.5 (C3Im), 122.2, (C4Im), 111.7 (Cquat iP),
104.8 (C1), {(81.6, 81.1, 78.3) (C2, C3, C4)}, 63.2 (OCH2Im), 48.9
(C5), 32.6 (MeIm), (26.5, 26.0) (OCMe2).
Scheme 5 Proton assignments.
5-Amino-5-deoxy-1,2-O-isopropylidene-3-O-(2-picolyl)-a-D-xylo-
furanose (G). F (1 g, 3 mmol) was dissolved in MeOH (10 mL)
and hydrogenated under pressure (P(H2) = 5 bar) using Pd/C as a
catalyst. After 3 h, the catalyst was removed by filtration on Celite
which was washed with MeOH. The solvent was evaporated and G
was obtained quantitatively as an oil. Microanalysis: calc./found
for C14H20N2O4·2/3H2O:
C
57.52/58.04,
H
7.36/7.05,
N
5-Amino-5-deoxy-1,2-O-isopropylidene-3-O-(N-methylimidazol-
2-yl)-a-D-xylofuranose (I). I was synthesized quantitatively
using the same protocol as G. H (0.88 g, 2.84 mmol) was
dissolved in MeOH (10 mL) and hydrogenated under pressure
(P(H2) = 5 bar) using Pd/C as a catalyst. After one night, the
catalyst was removed by filtration on Celite which was washed
with MeOH. The solvent was evaporated and I was obtained
quantitatively as an oil. 1H NMR (250 MHz, CDCl3): d (ppm, see
Scheme 6) 6.83 (1H, d, J = 7.0 Hz, H3Im), 6.78 (1H, d, J = 7.1 Hz,
H4Im), 5.85 (1H, d, J = 3.8 Hz, H1), {4.62 (d, 1H, J = 12.6 Hz),
9.12/9.85%. 1H NMR (250 MHz, CDCl3): d (ppm, see Scheme 5)
8.51 (m, 1H, H6py), 7.65 (td, 1H, J1 = 7.7 Hz, J2 = 1.7 Hz, H4py),
7.31 (d, 1H, J = 7.7 Hz, H3py), 7.18 (m, 1H, H5py), 5.90 (d, 1H,
J = 3.7 Hz, H1), {4.77 and 4.59 (2 × d, 2 × 1H, J = 13.3 Hz,
OCH2py)}, 4.63 (d, J = 3.7 Hz, H2), 4.24 (m, 1H, H4); 4,00 (d,
J = 3.2 Hz, H3), 3.07 (m, 2H, H5a and H5b), {1.44 and 1.26 (2 ×
s, 2 × 3H, MeiP)}. 13C NMR (62.5 MHz, CDCl3): d (ppm) 157.0
(C2py), 149.0 (C6py), 137.1 (C4py), {122.9, 121.5} (C3py, C5py),
111.7 (CquatIp), 104.9 (C1), {82.8, 82.1} (C2, C4), 80.3 (C3), 72.0
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 3235–3245 | 3243
©

View Article Online
Found: C, 37.65; H, 4.09; N, 5.06; Cu, 11.19%. HR-MS: m/z (%):
409.0823 (100) (calc m/z for C18H22N2O5Cu 409.0819 [CuIL]+).
Characterization of CuIL2. Yield: 22 mg (0.056 mmol, 92%).
−1
=
=
IR (KBr): m(cm ) 1610 (mC N), 1572 (mC C). Anal. Calc. for
C18H22N2O4CuPF6·H2O: C, 38.82; H, 4.34; N, 5.04; Cu, 11.41.
Found: C, 38.61; H, 4.35; N, 5.24; Cu, 11.48%. HR-MS: m/z (%):
393.0875 (100) (calc. m/z for C18H22N2O4Cu 393.0825 [CuIL]+).
Characterization of CuIL3. Yield: 23 mg (0.051 mmol, 91%).
Scheme 6 Proton assignments.
−1
=
=
IR (KBr): m(cm ) 1615 (mC N), 1568 (mC C). Anal. Calc. for
C21H26N2O5CuPF6·H2O: C, 41.14; H, 4.24; N, 4.57; Cu, 10.37.
Found: C, 40.98; H, 4.04 N, 4.73; Cu, 9.66%. HR-MS: m/z (%):
448.0957 (100) (calc. m/z for C21H25N2O5Cu 448.1123 [CuIIL−]+).
4.58 (m, 2H) (OCH2Im, H2)}, 4.29 (m, 1H, H4), 3.94 (d, J =
2.9 Hz, H3), 3.53 (s, 3H, MeIm), 2.84 (dd, J1 = 12.1 Hz, J2
=
7.5 Hz, H5a), 2.69 (dd, J1 = 12.1 Hz, J2 = 4.9 Hz, H5b), {1.44
and 1.25 (2 × s, 2 × 3H, MeiP)}. 13C NMR (90 MHz, CDCl3):
d (ppm) 143.6 (C2Im), 127.9 (C3Im), 122.7, (C4Im), 111.7 (Cquat iP),
104.9 (C1), (82.6, 81.9, 78.9) (C2, C3, C4), 63.1 (OCH2Im), 52.4
(C5), 32.7 (MeIm), (26.8, 26.1) (OCMe2).
Characterization of CuIL4. Yield: 21 mg (0.046 mmol, 91%).
−1
=
=
IR (KBr): m(cm ) 1614 (mC N), 1572 (mC C). Anal. Calc. for
C20H27N3O5CuPF6·H2O: C, 38.99; H, 4.39; N, 6.82; Cu, 10.32.
Found: C, 39.38; H, 4.68; N, 6.77; Cu, 10.79%. HR-MS: m/z (%):
451.1160 (100) (calc. m/z for C20H26N3O5Cu 451.1163 [CuIIL−]+).
The copper(I) complexes (CuIL3 and CuIL4) are oxidized in
the spray and detected as adducts of copper(II) complexes, which
differ by one mass units only, due to the deprotonation of the
phenolic moiety coordinated to the metallic centre induced by its
oxidation.
5-(Amino-N -(2-salicyl))-5-deoxy-1,2-O-isopropylidene-3-O-
(methylimidazol-2-yl)-a-D-xylofuranose (L4). L4 was synthe-
sized using the same protocol as L3. I (600 mg, 2.12 mmol,
1 eq.) was dissolved in 15 mL absolute ethanol. Salicylaldehyde
(694 mg, 5.68 mmol, 2 eq.) was added which caused the immediate
appearance of a yellow colour. The solution was cooled (water-ice
bath). NaBH4 (61.1 mg, 2.12 mmol, 1 eq.) was added inducing a
disappearance of the yellow coloration. The reaction mixture was
then stirred for 30 min at RT. Excess NaBH4 was destroyed with
acetic acid. The solvent was evaporated and L4 was purified by
chromatography (AcOEt–MeOH–NH3, v/v/v = 95/5/2). Yield:
560 mg (68%). Microanalysis (calc./found) C20H27N3O5·H2O: C
58.96/59.01, H 6.68/6.72, N 10.31/10.29%. 1H NMR (360 MHz,
CDCl3): d (ppm, see Scheme 6) 7.07 (1H, t, J = 7.6 Hz, H3phOH),
{6.90 (1H, d, J = 7.2 Hz), 6.7 (2H, m) (H2PhOH, H4PhOH, H5PhOH)},
6.79 (1H, slarge, H3Im), 6.74 (1H, slarge, H4Im), 5.80 (1H, d, J = 3.7 Hz,
H1), {4.60 (d, 1H, J = 12.8 Hz), 4.5 (m, 2H) (OCH2Im, H2)} 4.24
(m, 1H, H4), 3.95 (d, 1H, J = 13.8 Hz, NHCH2aPhOH), 3.90 (d,
J = 3.0 Hz, H3), 3.81 (d, 1H, J = 13.8 Hz, NHCH2bPhOH), 3.49
(s, 3H, MeIm), 2.82 (dd, J1= 12.3 Hz, J2 = 7.6 Hz, H5a), 2.67
(dd, J1= 12.3 Hz, J2 = 5.0 Hz, H5b), {1.40 and 1.23 (2 × s, 2 ×
3H, MeiP)}. 13C NMR (90 MHz, CDCl3): 143.6 (C2Im), {(128.6,
128.5, 127.5) (C3PhOH, C5PhOH, C3Im)}, {(122.2, 118.9, 116.2) (C4Im,
C2PhOH, C4PhOH)}, 111.7 (Cquat iP), 104.5 (C1), {(82.1, 81.8, 78.7)
(C2, C3, C4)}, 63.2 (OCH2Im), 52.4 (NH-CH2-PhOH), 46.4 (C5),
32.7 (MeIm), (26.7, 26.1) (OCMe2). MS-ES: 389.2 (23%) [M]; 388.2
(100%) [M − H].
Acknowledgements
We thank the Re´gion Champagne-Ardenne for a grant to Z. D.
and for its financial support, and the French ministry government
for financial support (ACI-jeune chercheurs JC4044) for C. P. The
C.R.I.H.A.N computing center (http://www.crihan.fr) and the
computational center of the Universite´ de Reims Champagne-
Ardenne (ROMEO: http://www.romeo2.univ-reims.fr) are ac-
knowledged for the CPU time donated.
References
1 J. P. Klinman, Chem. Rev., 1996, 96, 2541.
2 E. I. Solomon, U. M. Sundaram and T. E. Machonkin, Chem. Rev.,
1996, 96, 2563.
3 N. Kitajima and Y. Moro-Oka, Chem. Rev., 1994, 94, 737.
4 K. D. Karlin and Z. Tyeklar, Bioinorganic Chemistry of Copper,
Chapman & Hall, New-York, 1993.
5 D. H. Lee, L. Q. Hatcher, M. A. Vance, R. Sarangi, A. E. Milligan,
A. A. Narducci, Sarjeant, C. D. Incarvito, A. L. Rheingold, K. O.
Hodgson, B. Hedman, E. I. Solomon and K. D. Karlin, Inorg. Chem.,
2007, 46, 6056.
6 L. M. Mirica, X. Ottenwaelder and T. D. P. Stack, Chem. Rev., 2004,
104, 1013.
7 E. A. Lewis and W. B. Tolman, Chem. Rev., 2004, 104, 1047.
8 J. Miltz and M. Perry, Packag. Technol. Sci., 2005, 18, 21.
9 P. Cook, Stealth scavenging systems: the scavenger in the polymer, in
International Food Technologists Annual Meeting Book of Abstracts,
IFT, Atlanta, GA, 1998.
10 M. L. Rooney, Overview of oxygen-scavenging technologies, in Inter-
national Food Technologists Annual Meeting Book of Abstracts, IFT,
New Orleans, LA, 2001.
Copper complexes [CuI(Ln)]
The protocol was the same for all the complexes. The synthesis
of CuIL1 is detailed here. Degassed methanol solutions (1 mL)
of methyl-2,3-di-O-(2-picolyl)-a-D-lyxofuranoside (L1, 20 mg,
0.0577 mmol) and [Cu(CH3CN)4][PF6] (21.5 mg, 0.0577 mmol)
were separately prepared under Ar in a glove box. The copper(I)
salt solution was added dropwise to the ligand solution, and the
resulting bright yellow solution was stirred during a few minutes.
Slow evaporation led to a colourless solid.
11 G. Tewari, D. S. Jayas, L. E. Jeremiah and R. A. Holley, Int. J. Food
Sci. Technol., 2002, 37, 209.
12 F. Bellot, R. Hardre´, G. Pelosi, M. The´risod and C. Policar, Chem.
Commun., 2005, 5414.
13 F. Cisnetti, R. Guillot, M. Desmadril, G. Pelosi and C. Policar, Dalton
Trans., 2007, 1473.
14 G. Charron, F. Bellot, F. Cisnetti, G. Pelosi, J.-N. Rebilly, E. Rivie`re,
A.-L. Barra, T. Mallah and C. Policar, Chem.–Eur. J., 2007, 13, 2774.
Characterization of CuIL1. Yield: 22 mg (0.054 mmol, 93%).
−1
=
=
IR (KBr): m(cm ) 1615 (mC N), 1574 (mC C). Anal. Calc. for
C18H22N2O5CuPF6·H2O: C, 37.73; H, 4.19; N, 4.89; Cu, 11.09.
3244 | Dalton Trans., 2008, 3235–3245
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
15 F. Cisnetti, R. Guillot, M. The´risod and C. Policar, Acta Crystallogr.,
Sect. C: Cryst. Struct. Commun., 2007, 63, m201.
16 W. Szeja, I. Fokt and G. Grynkiewicz, Recl. Trav. Chim. Pays-Bas, 1989,
108, 224.
T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E.
Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. 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, K.
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. G. Johnson,
W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, GAUSSIAN 03
(Revision D.02), Gaussian, Inc., Wallingford, CT, 2004.
32 I. Morao, J. P. McNamara and I. H. Hillier, J. Am. Chem. Soc., 2003,
125, 628.
17 S. Hotha, R. I. Anegundi and A. A. Natu, Tetrahedron Lett., 2005, 46,
4585.
18 S. Miertus, E. Scrocco and J. Tomasi, Chem. Phys., 1981, 55, 117.
19 M. Cossi, G. Scalmani, N. Rega and V. Barone, J. Chem. Phys., 2002,
117, 43.
20 R. Cammi, B. Mennucci and J. Tomasi, J. Phys. Chem. A, 2000, 104,
5631.
21 N. G. Connelly and W. E. Geiger, Chem. Rev., 1996, 96, 877.
22 L. Q. Hatcher, M. A. Vance, A. A. Narducci, Sarjeant, E. I. Solomon
and K. D. Karlin, Inorg. Chem., 2006, 45, 3004.
23 P. Winget, E. J. Weber, C. J. Cramer and D. G. Truhlar, Phys. Chem.
Chem. Phys., 2000, 2, 1231.
24 H. Kunkely and A. Vogler, Inorg. Chim. Acta, 2004, 357, 888.
25 A. B. P. Lever, Inorganic Electronic Spectroscopy, Elsevier, New York,
1984.
33 M. Itagaki, K. Masumoto, K. Suenobu and Y. Yamamoto, Org. Process
Res. Dev., 2006, 10, 245.
34 J. Cody, J. Dennisson, J. Gilmore, D. G. VanDerveer, M. M. Henary,
A. Gabrielli, C. D. Sherrill, Y. Zhang, C.-P. Pan, C. Burda and C. J.
Fahrni, Inorg. Chem., 2003, 42, 4918.
26 J. Peisach and W. E. Blumberg, Arch. Biochem. Biophys., 1974, 165, 691.
27 P. J. Benites, D. S. Rawat and J. M. Zaleski, J. Am. Chem. Soc., 2000,
122, 7208.
28 B. J. Hathaway and D. E. Billing, Coord. Chem. Rev., 1970, 5, 143.
29 I. Bertini, D. Gatteschi and A. Scozzafava, Inorg. Chem., 1977, 16,
1973.
35 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter Mater.
Phys., 1988, 37, 785.
36 B. Miehlich, A. Savin, H. Stoll and H. Preuss, Chem. Phys. Lett., 1989,
157, 200.
37 S. H. Vosko, L. Wilk and M. Nusair, Can. J. Phys., 1980, 58, 1200.
38 A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
30 G. F. Kokoszka, C. W. Reimann and H. C. Allen, J. Phys. Chem., 1967,
39 G. Schaftenaar and J. H. Noordik, J. Comput. Aided Mol. Des., 2000,
14, 123.
71, 121.
31 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,
40 W. Humphrey, A. Dalke and K. Schulten, J. Mol. Graphics, 1996, 14,
33.
41 W. G. Overend, F. Shafizadeh and M. Stacey, J. Chem. Soc., 1950, 671.
42 P. Gamez, C. Simons, G. Arom`ı, W. L. Driessen, G. Challa and J.
Reedjik, J. Appl. Catal. A: General, 2001, 214, 187.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 3235–3245 | 3245
©