From structural properties of the EuIII complex with ethylenediaminetetra(methylenephosphonic acid) (H8EDTMP) towards biomedical applications

Article abstract of DOI:10.1039/b606420e

Crystals of Eu^III with ethylenediaminetetra(methylenephosphonic acid) (H₈EDTMP) and with ethylenediaminetetraacetic acid (H ₄EDTA) have been synthesized in the same experimental conditions and their X-ray analyses have been performed. The EDTMP ligand wraps the Eu ^III ion in a fashion similar to its carboxylic analogue, EDTA, i.e. coordinating through two nitrogen atoms and four oxygen atoms in such a way that only one oxygen atom from each phosphonate group is bonded to the central ion. The coordination sphere is completed by two oxygen atoms of the bidentate carbonate anion in the case of the Eu^III-EDTMP complex, whereas the inner sphere of the Eu^III-EDTA crystal is completed by three water molecules. Spectroscopic studies (UV-Vis and ³¹P NMR spectra) of Eu^III-EDTMP solutions at controlled pH showed that the replacement of inner sphere water molecules and/or OH hydroxy groups by a carbonate anion in the Eu^III-EDTMP complex at physiological pH results in the formation of [Eu(EDTMP)(CO₃)]^7- species which is thermodynamically stable and kinetically inert. The affinity of the carbonate anion towards the Eu^III-EDTMP species was studied by analysis of f-f intensities and luminescence decay rates. The dissociation constant of the Eu ^III-EDTMP-carbonate complex was found to be approximately 43 mM. The presented results may be helpful in understanding the role played by the ¹⁵³Sm^III-EDTMP complex known as Quadramet in the seeking of metastatic tissue in bones as well as possibly giving some premises for future ligand design of these types of complexes with lanthanide radionuclides. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b606420e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
From structural properties of the EuIII complex with
ethylenediaminetetra(methylenephosphonic acid) (H8EDTMP)
towards biomedical applications†
Anna Mondry* and Rafał Janicki
Received 5th May 2006, Accepted 19th July 2006
First published as an Advance Article on the web 10th August 2006
DOI: 10.1039/b606420e
Crystals of EuIII with ethylenediaminetetra(methylenephosphonic acid) (H8EDTMP) and with
ethylenediaminetetraacetic acid (H4EDTA) have been synthesized in the same experimental conditions
and their X-ray analyses have been performed. The EDTMP ligand wraps the EuIII ion in a fashion
similar to its carboxylic analogue, EDTA, i.e. coordinating through two nitrogen atoms and four
oxygen atoms in such a way that only one oxygen atom from each phosphonate group is bonded to the
central ion. The coordination sphere is completed by two oxygen atoms of the bidentate carbonate
anion in the case of the EuIII–EDTMP complex, whereas the inner sphere of the EuIII–EDTA crystal is
completed by three water molecules. Spectroscopic studies (UV-Vis and 31P NMR spectra) of
EuIII–EDTMP solutions at controlled pH showed that the replacement of inner sphere water molecules
and/or OH hydroxy groups by a carbonate anion in the EuIII–EDTMP complex at physiological pH
results in the formation of [Eu(EDTMP)(CO3)]7− species which is thermodynamically stable and
kinetically inert. The affinity of the carbonate anion towards the EuIII–EDTMP species was studied by
analysis of f–f intensities and luminescence decay rates. The dissociation constant of the
EuIII–EDTMP–carbonate complex was found to be approximately 43 mM. The presented results may
R
be helpful in understanding the role played by the 153SmIII–EDTMP complex known as Quadrametꢀ in
the seeking of metastatic tissue in bones as well as possibly giving some premises for future ligand
design of these types of complexes with lanthanide radionuclides.
EDTMP complexes are kinetically inert, whereas the depro-
Introduction
tonation of the ligand leads to higher lability of species as
found from the kinetic studies of the EuIII–EDTMP system.9
The latter studies also proved that the kinetic behaviour of
the LnIII–EDTMP species is in contrast to that observed for
the LnIII–EDTA complexes.9 Some insight in the formation of
lanthanide acyclic polyaminopolyphosphonate species in aqueous
solution was gained previously by the luminescence spectroscopy
studies.7,8 These studies performed on the 1 : 1 EuIII–EDTMP
complex have revealed that species which appear in aqueous
solutions at pH above 6 are monomeric and contain, besides the
EDTMP ligand, two water molecules in the inner sphere of the
EuIII ion.7
Previously, by applying the method of crystallization for an-
ionic lanthanide complexes with polyaminopolycarboxylic acids
described by Ruloff et al.,13 we demonstrated the usefulness of the
guanidinium countercation for obtaining crystals with different
coordination geometry.14 For this reason we have decided to use
the guanidinium cation in the form of guanidinium carbonate for
crystallization of highly negative anionic lanthanide complexes
with polyphosphonic acids.15 Following this approach we have
been able to obtain in the same experimental conditions crystals
of the EuIII–EDTMP complex and of its carboxylic analogue,
EuIII–EDTA. In this work we present their X-ray analyses and
the spectroscopic results for both complexes in aqueous solutions
and in the form of single crystals.
Recently a great interest in studying lanthanide complexes
with polyamino polyphosphonic ligands has been observed.
One of the reasons for which these chelate complexes have
been extensively investigated is their potential use in biomed-
ical diagnosis1 and in cancer therapy.2,3 Contrary to the lan-
thanide complexes with 1,4,7,10-tetraazacyclododecane-1,4,7,10-
tetrakis(methylenephosphonic acid) (H8DOTMP) or related cyclic
ligands, which have been widely studied in solutions and
in single crystals,1,4 knowledge of the possible structure of
LnIII species with acyclic polyaminopolyphosphonates is still
fragmentary.5–9 In spite of this, the 153SmIII complex with ethylene-
diaminetetrakis(methylenephosphonic acid) (H8EDTMP), known
R
as Quadrametꢀ, is commonly used for the palliation of pain from
metastatic bone cancer.2,3 Difficulties with estimating stability
constants by means of the potentiometric method for lanthanide
complexes with the EDTMP ligand are known.10–12 The stability
constants determined for the LnIII–EDTMP system differ up
to ten orders of magnitude depending on the experimental
conditions.12 It was shown that protonated species of the LnIII–
Faculty of Chemistry, University of Wrocław, F. Joliot-Curie 14, 50-383,
Wrocław, Poland. E-mail: anm@wchuwr.chem.uni.wroc.pl
† Electronic supplementary information (ESI) available: Fig. S1:
View of the unit cells of [C(NH2)3]7[Eu(EDTMP)(CO3)]·10H2O and
[C(NH2)3][Eu(EDTA)(H2O)3]. See DOI: 10.1039/b606420e
4702 | Dalton Trans., 2006, 4702–4710
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Results and discussion
Crystal structures
The crystal structures of the EuIII compounds with EDTMP
and
EDTA,
[C(NH2)3]7[Eu(EDTMP)(CO3)]·10H2O
and
[C(NH2)3][Eu(EDTA)(H2O)3] respectively, were determined.
Both crystals were obtained by the same method of crystallization
with the pH of the natant solution 10.7 in each case (see
Experimental section for further details). The crystals are
composed of the respective complex anions, guanidinium cations
and water molecules. Both the complex anions are presented in
Fig. 1, while the cell packing views are given in Fig. S1 (see ESI†).
Selected bond lengths of both compounds are given in Table 1. The
presence of the guanidinium cation in the EuIII–EDTA crystal does
not bring about a change of the composition of the coordination
environment of Eu, which is the same (four O and two N atoms
from the EDTA ligand and three water molecules) as in the earlier
reported various salts of EuIII–EDTA complexes.16 Since each
of the counterions, namely Na+, K+, NH4 and a guanidinium
+
cation, interacts with the anionic [Eu(EDTA)(H2O)3]− complex
through different hydrogen-bonding networks, small differences
in the EuIII–ligand distances are observed in crystals of these
salts.16 One may notice, for instance, a small decrease of the
EuIII–O(COO−) distances and elongation of the EuIII–N as well as
the EuIII–O(H2O) distances in the guanidinium crystal (this work)
as compared with the sodium one.16a
Although the EDTMP ligand has twice as many oxygen donor
atoms as the EDTA ligand, it wraps the EuIII ion in the same
fashion as its carboxylic analogue, i.e. by two nitrogen atoms (N1,
N2) and four oxygen atoms (O1, O6, O8, O10) in such a way that
only one oxygen atom from each phosphonic group is coordinated
to the central ion and the Eu–O(PO32−) bond distances are similar
(Table 1). All bond distances (C–C, C–P, C–N and P–O) within the
ligand backbone are very similar to those determined for the free
ligand (H8EDTMP).17 The P–O bond distances when the O atoms
are coordinated to the EuIII ion are generally slightly elongated
and resemble the P–O(H) distances in the free ligand.17
Due to the presence of bulkier, tetrahedral phosphonic groups
and longer C–P bonds in the EuIII–EDTMP crystal as compared
with the planar carboxylic groups and shorter C–COO distances
in the EuIII–EDTA complex, free space available for other ligands
(usually solvent molecules) around the central ion is more limited
Fig. 1 Molecular structure of (A) [Eu(EDTMP)(CO3)]7− and (B)
[Eu(EDTA)(H2O)3]− anions together with the atom numbering schemes.
in the phosphonate complex than in its carboxylate analogue. For
instance, the coordination number of the lanthanide ion, eight, is
one less in the LnIII crystals with DOTMP4c than that in DOTA
(nine),18 the difference being brought about by coordination of one
more water molecule in the case of the carboxylate complexes. The
change of the coordination number of the EuIII ion in its complexes
with EDTMP and EDTA from eight to nine is also observed in the
crystal structures presented in this work. Three water molecules
which complete the first coordination sphere of the metal ion in
the EuIII–EDTA complex are replaced by two oxygen atoms (O22,
O23) from the bidendate carbonate anion in the case of the EuIII–
EDTMP crystal. This is one of the reasons that the symmetry of
the phosphonate complex is C2, while that of the carboxylate one
is C1.
The coordination of the carbonate anion to the lanthanide ion
in the present crystal, is however quite unexpected since up to date
there are only few crystal structures known where the binding
of carbonate to the lanthanide ion has been achieved at ambient
conditions (temperature and pressure) of synthesis.19
It is perhaps of note that the X-ray analysis of the
[C(NH2)3]7[Eu(EDTMP)(CO3)]·10H2O crystal has revealed, con-
trary to the X-ray results of the H8EDTMP ligand,17 a slight
nonequivalence of the phosphonate groups. This point may be
illustrated by the values of angles between the vectors Eu–P and
Eu–O21 (the latter determines the approximate twofold symmetry
axis C2 of the complex anion). Thus the angles are slightly greater
in the case of P2 and P3 than those involving P1 and P4. They
amount to 117.8◦ (O21–Eu–P2), 109.9◦ (O21–Eu–P3), 102.0◦
˚
Table 1 Selected bond lengths (A) for crystals of [C(NH2)3]7-
[Eu(EDTMP)(CO3)]·10H2O and [C(NH2)3][Eu(EDTA)(H2O)3]
EuIII–EDTMP
EuIII–EDTA
Eu–O6
Eu–O1
Eu–O8
Eu–O10
Eu–Oav
Eu–O23
Eu–O22
Eu–Oav
2.409(2)
2.361(2)
2.343(2)
2.376(2)
2.372(28)
2.419(2)
2.446(2)
2.432(19)
Eu–O1
Eu–O3
Eu–O5
Eu–O7
Eu–Oav
Eu–OW1
Eu–OW2
Eu–OW3
Eu–Oav
Eu–N1
Eu–N2
Eu–Nav
Eu–Eu
2.426(2)
2.458(2)
2.392(2)
2.383(2)
2.415(34)
2.464(2)
2.435(2)
2.528(2)
2.476(48)
2.727(2)
2.622(2)
2.674(74)
7.814(3)
Eu–N1
Eu–N2
Eu–Nav
Eu–Eu
2.703(3)
2.734(3)
2.718(22)
9.998(3)
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 4702–4710 | 4703
©

View Article Online
(O21–Eu–P1) and 105.5◦ (O21–Eu–P4). On the other hand O22
and O23 atoms from the carbonate anion are nearer to the P2 and
P3 phosphonate groups than to the P1 and P4 ones. In this way
the carbonate anion is located in a quasi-saddle, with P1 and P4
phosphonate groups supporting the anion from both sides, and
2−
the CO3 plane is almost parallel to the P2–P3–Eu plane and
perpendicular to the P1–P4–Eu plane. Furthermore, the lateral
support of the anion is strengthened by two guanidinium cations
which bridge the phosphonate groups with the coordinated O
carbonate atoms by relevant H-bonds. Namely one of the cations
is H-bonded to O10 (of P4 group) and to O22, whereas the other
2−
links O6 (of P1) with O23. Such a system brings about the CO3
coordination rigidity.
Spectral results
The f–f transition spectra of the EuIII ion are one of the most
convenient among the spectra of all lanthanide ions due to the fact
that the ground 7F0 and the emitting 5D0 states of this ion are non-
degenerate.20 For this reason the number of components observed
Fig. 2 Emission spectra of EuIII complexes with EDTMP and EDTA in
the form of single crystals and in solutions.
5
in the 7F0 ↔ D0 transitions directly renders the number of chem-
ically distinct environments of the EuIII ion. These transitions are
forbidden by the symmetry selection rule when the local symmetry
of the EuIII complex is high or when an inversion centre in the EuIII
solution was adjusted by guanidinium carbonate or NaOH. A
7
5
distinct decrease of the F0 → D0 intensity (for both ligands)
in the solutions containing a carbonate anion as compared to the
solutions with the pH fixed by NaOH results from the substitution
of some water molecules and/or OH− groups by the CO32− anion.
Contrary to the EuIII–EDTA solutions of the sodium salt, for
which no considerable changes of the shape of f–f bands are
observed in the pH range between 5.4 and 10.7, a strong influence
of pH changes on the EuIII–EDTMP spectra is clearly seen.
To obtain additional information about the studied systems,
particularly those with the EDTMP ligand, the absorption spectra
of the EuIII–EDTMP and EuIII–EDTA complexes were measured.
5
species is present. The position of the 7F0 → D0 transition is also
used to study the nephelauxetic effect in europium compounds.21
This effect is interpreted as a covalent contribution to the bonding
between the metal ion and the ligands and is probably connected
with metal–ligand distances, coordination numbers and the total
charge of the first coordination sphere. The 7F0
→
5D1 and
7
5D0 → F1 transitions are magnetically allowed and therefore their
intensities are essentially independent of the EuIII environment.
5
7
This feature of the D0 → F1 transition may be exploited to
calibrate the intensities of other EuIII f–f transitions in the emission
5
5
5
The spectra of the 7F0 → D0, 7F0 → D1 and 7F0 → D2 transitions
5
7
spectra. The 7F0 → D2 and 5D0 → F2 transitions are known as
hypersensitive transitions which means that their intensities are
strongly influenced by the changes of the nearest surrounding of
the EuIII ion. The intensity of these transitions depend mainly on
the symmetry of the EuIII ion and/or the ligand polarizability.22,23
On account of these characteristic f–f transition features it is of
interest to compare spectral properties of both studied crystals as
well as the relevant complexes in aqueous solutions, particularly
the properties of the EuIII–EDTA complex in aqueous solution
which are rather well recognized.24,25
together with the peak maxima are presented in Fig. 3. As can
be seen in Fig. 3(A), in the 7F0
→
5D0 spectra of the EuIII
complexes with EDTMP and EDTA in solutions of which pH
was adjusted by NaOH, two well separated peaks occur indicating
two different coordination arrangements of the EuIII ion in both
systems. Introduction of the carbonate anion to EuIII complexes
with EDTMP and EDTA at pH 10.7 brings about changes in
7
7
the inner sphere of the metal ion. In the F0 → D0 spectrum of
the EuIII–EDTMP complex a single peak with the same energy at
maximum (17261 cm−1) as that of the crystal appears. Moreover,
it may be supposed that among the two peaks found in the 7F0 →
7D0 spectra of EuIII–EDTMP solutions adjusted by NaOH, the
red-shifted peak with energy similar to that of the crystal, may
belong to species in which two oxygen atoms from water molecules
and/or hydroxy groups are located in similar positions as the O
atoms from the carbonate anion in the crystal. Since the peak
Spectra of the f–f transitions. The corrected emission spectra
of single crystals and of their natant solutions, i.e. solutions with
1 : 1 EuIII to ligand ratio basified with guanidinium carbonate
to pH 10.7 as well as of solutions with pH adjusted by NaOH
are presented in Fig. 2. As can be seen, the intensity of the
5
7F0 → D0 transition is much weaker in the crystal of the EuIII–
EDTMP complex than that observed in the EuIII–EDTA crystal.
This fact is in accordance with a higher symmetry of the EuIII
surrounding in the former crystal, as has been disclosed by the
diffraction study presented in the preceding section. One may
also notice a high correspondence between the spectra of the
EuIII–EDTMP–carbonate complex in solution and in the crystal
whereas the corresponding spectra of the EuIII–EDTA system
differ considerably. The spectra of EuIII solutions with both ligands
at pH 10.7 vary strongly depending on whether the pH of the
positions of the remaining 7F0 → D1 and 7F0 → D2 transitions
in spectra of the EuIII–EDTMP–carbonate complex in the solution
and in the crystal are similar, one may presume the presence of the
[Eu(EDTMP)(CO3)]7− entity, found from the X-ray analysis, also
in solution.
7
7
The appropriate spectra of the EuIII–EDTA complex of the
crystal and of the solution from which the crystals were obtained
differ considerably. As was found previously, in the case of the
EuIII–EDTA–carbonate solution besides species with inner sphere
4704 | Dalton Trans., 2006, 4702–4710
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
water molecules replaced by a carbonate anion also a carbonate
complex with the EuIII ion is formed.26
Intensities of the f–f transitions. The oscillator strength values
5
(P) of the 7F0 → DJ transitions of the EuIII ion in the studied sys-
tems were also determined and are given in Table 2. The P values
5
of the 7F0 → D1 transition calculated for all complexes are nearly
constant. The high correspondence between the respective 7F0 →
5DJ oscillator strength values of the EuIII–EDTMP–carbonate
crystal and the EuIII–EDTMP–carbonate solution is in accordance
with the resemblance of the absorption and emission spectra
for both these systems. The high P values of the hypersensitive
7F0
→
5D2 transition of both the crystal and the solution in
comparison with a much lower P value of the NaOH-adjusted
solution at pH 10.7 may be related to the influence of the carbonate
polarizability on the EuIII ion.
Since the direction of P changes observed for the EuIII–
EDTMP complex in sodium solutions seems to be associated
with the protonation of EuIII species, therefore intensities of
absorption and emission transitions as the functions of pH of
EuIII–EDTMP solutions were plotted. Plots of these relations
together with percentage contents of analogous SmIII–EDTMP
species calculated from the stability constants given by de Witt
et al.12 are shown in Fig. 4. As may be seen in this Figure there is
no pH region in which the deprotonated [Eu(EDTMP)]5− species
would exist without significant admixture of other species.
Fig. 4 Absorption and emission intensities of EuIII–EDTMP solutions
as a function of pH (kexc = 394.8 nm in emission spectra). The included
percentage contents of SmIII–EDTMP species[a] were calculated from
ref. 12.
Fig. 3 Absorption spectra of (A) 7F0
→
5D0, (B) 7F0
→
5D1 and (C)
5
7F0 → D2 transitions of EuIII complexes with EDTMP and EDTA (cEu
=
40 mM, d = 5 cm for EuIII–EDTMP solutions; cEu = 20 mM, d = 5 cm for
EuIII–EDTA solutions; cEu = 1.368 M, d = 0.063 cm for EuIII–EDTMP
crystal; cEu = 3.642 M, d = 0.067 cm for EuIII–EDTA crystal); instead of
Binding of H2O and/or OH−. The luminescence lifetime
measurements of the EuIII–EDTMP complex in the crystal and
solutions at pH 10.7 and 7.5 have been measured. It is worth
noting that the estimation of the inner sphere water molecules
5
the 7F0 → D0 absorption spectrum which could not be recorded for the
EuIII–EDTMP crystal due to its low intensity, the luminescence excitation
spectrum (kem = 612 nm) of this transition is displayed.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 4702–4710 | 4705
©

View Article Online
Table 2 Oscillator strength values (P) of EuIII complexes with EDTMP and EDTA
EuIII–EDTMP EuIII–EDTA
5
5
5
5
5
5
7F0 → D0
7F0 → D1
7F0 → D2
7F0 → D0
7F0 → D1
7F0 → D2
pH
5.40
5.98
7.18
8.10
9.15
10.70
10.70a
Crystal
108P
0.24
0.51
0.77
0.79
0.86
0.88
0.22
0.12b
108P
1.65
1.68
1.64
1.64
1.64
1.63
1.48
1.60
108P
10.40
8.13
6.44
6.06
6.00
6.24
19.62
21.01
pH
4.50
108P
108P
108P
0.59
1.43
8.59
7.50
0.63
1.45
8.72
10.70
0.61
0.48
0.65
1.49
1.53
1.45
10.49
19.42
4.95
10.70a
Crystal
pH of solutions adjusted by NaOH a pH of solutions adjusted by (gua)2CO3. b P value estimated from the excitation spectrum.
Table 3 Emission lifetimes (s) and hydration numbers (q) of the EuIII
EDTMP complex
–
Added compound
pH
sH O/ls
sD O/ls
q
2
2
Crystal (none)
(gua)2CO3
Na2CO3
NaOH
(gua)2CO3
Na2CO3
—
1036
1022
1038
422
826
864
—
—
10.70
10.70
10.70
7.50
7.50
7.50
2107
2047
2412
2066
2016
2193
0.22
0.18
1.83
0.46
0.39
1.79
NaOH
420
in this complex is difficult due to the fact that the excited state
of the EuIII ion will be also quenched by the hydroxy groups.
The hydration numbers (q), or more strictly, the numbers of OH
oscillators (2q) were determined from eqn (1):27
−1
−1
O
q = 1.11(sH2
− sD2
− 0.31)
(1)
O
where sH −1 and sD −1 are the luminescence decay rates measured
O
O
2
2
in H2O and D2O solutions. This equation complies with the
deactivation equal to 0.31 ms−1 by water molecules of the second
coordination sphere. The results are presented in Table 3.
The close similarity between luminescence lifetimes determined
for the crystal and both carbonate solutions at pH 10.7 of the
EuIII–EDTMP complex in H2O clearly indicates the substitution
of all OH oscillators originating from water molecules and/or
hydroxy groups by a carbonate anion. As may be suggested by
the number of OH oscillators calculated for the solution of which
pH 10.7 was adjusted by NaOH, the dominant species in that
solution contains one inner sphere water molecule and one inner
sphere hydroxy group.
The q numbers calculated for the solution adjusted by NaOH to
pH 7.5 (at this pH the distribution of SmIII species is 30.8% [SmL]5−
and 68.4% [SmH−1L]6− 12) are very similar to those at pH 10.7. The
relatively low precision of the method does not allow detecting the
shifts of the equilibrium between the species with two coordinated
water molecules and that with one H2O and one OH− group.
Fig. 5 31P NMR spectra of EuIII–EDTMP solutions.
the solution at pH 5.4, similarly as the peak of the free ligand
(12.6 ppm, pH = 5.4) but its width is 302 Hz as compared to the
6 Hz of the free ligand peak. The large width of this peak indicates
that this resonance originates from the complexed ligand. Apart
from this peak, another weak peak at 51.6 ppm with 340 Hz width
is also observed at pH 5.4. As the pH is increased to 7.2, the
peak at 12.7 ppm disappears and that at 51.6 gains in intensity.
This suggests that the former peak corresponds to the protonated
species [EuHL]4− and [EuH2L]3−, which vanish at higher pH, as the
speciation curves of analogous SmIII–EDTMP complex indicate.12
The peak at 51.6 ppm in the spectrum at pH 5.4 corresponds then
to the [EuL]5− species.
31P NMR spectra. Further information about the solution
structure of the EuIII–EDTMP complex has been provided by
NMR spectroscopy. The 31P NMR spectra of EuIII–EDTMP
solutions for which the pH was adjusted by NaOH, Na2CO3
and [C(NH2)3]2CO3 are presented in Fig. 5. In the spectra of the
NaOH adjusted solutions of the complex one strong phosphorus
resonance appears. It is located at 12.7 ppm in the spectrum of
As long as the concentration of [EuL]5− species is considerable,
the intramolecular exchange is fast due to the fact that four
phosphorus atoms are averaged. A significant broadening of the
peak which is associated with the nearly complete formation of
the hydroxy species of the complex at pH 9.2 may then reflect that
the ligand exchange rate slows down. The appearance of a broad
band divided into two observed resonances at pH 10.7 may mean
4706 | Dalton Trans., 2006, 4702–4710
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Table 4 The dissociation (KD) and association (KA) constants determined
from different optical measurements for the EuIII–EDTMP–carbonate
complex
that the structure of the complex becomes more rigid, thus the
formation of double hydroxy species cannot be excluded.
The stabilization of the hydroxy EuIII–EDTMP species increases
remarkably when the H2O and OH− groups are replaced by a
carbonate anion as is well demonstrated by two well resolved
phosphorus peaks in spectra of the [Eu(EDTMP)(CO3)]7− com-
plex. The appearance of two P resonances separated by 44.7
and 52.7 ppm in spectra of sodium and guanidinium solutions
respectively, allows to observe the nonequivalence of phosphonate
groups in the [Eu(EDTMP)(CO3)]7− species as may be predicted
from the X-ray analysis, as well as the influence of countercations
on the signal positions in the 31P NMR spectra.
Method of KD and KA determinations
Intensities
Absorption
Emission
Luminescence lifetimes
KD/mM
KA/M−1
40.3
24.8
14.8
67.6
46.1
21.7
The determined KD and KA constants from different mea-
surements for the EuIII–EDTMP–carbonate complex are given
in Table 4. While the data obtained from absorption intensities
and luminescence lifetime measurements are in good agreement,
the results estimated from emission intensities differ from them
about three. The only probable reason of this disagreement was
the timescale. The absorption spectra and luminescence lifetime
measurements were taken after one day of solution preparation,
whereas the emission spectra of those solutions were recorded
a week later. A very slow establishing of the equilibrium was
probably a reason of the lower value of the dissociation constant
obtained from the emission intensities measurements.
Comparing the obtained values of the dissociation constants
with the K values available from the literature26,29 we see that
the discussed system is an order of magnitude more stable than
the EuIII–EDTA–carbonate complex (KD = 298 mM),26 while
similar or higher affinity of the carbonate anion towards europium
chelates was estimated for the neutral and positively charged (+3)
EuIII complexes, as for example [Eu(DO3A)] (KD = 35.3 mM,
where H3DO3A is 1,4,7,10-tetraazacyclododecane-1,4,7-triacetic
acid)26 and its alkylphenatridinium derivatives (KD = 2.5 and
0.2 mM).29
Binding of carbonate. Since the electronic and 31P NMR
spectra of EuIII–EDTMP solutions at pH around 7.5 adjusted
by Na2CO3 or [C(NH2)3]2CO3 are very similar to those recorded
for alkaline solutions with carbonates at pH 10.7, we deemed
it advisable to undertake studies of the EuIII–EDTMP complex
interaction with the carbonate ion at physiological pH. Such
studies with physiologically available anions have been recently
carried on LnIII complexes with polyaminopolycarboxylates and
with cyclen derivatives.26,28–31
To the EuIII–EDTMP complex adjusted by NaOH up to pH 7.5
in a HEPES buffer different amounts of NaHCO3 were added.
Binding of the carbonate anion towards the EuIII–EDTMP com-
plex was monitored by the intensity changes of the hypersensitive
transitions in absorption and emission, and also by luminescence
lifetime measurements. The intensity changes were followed by
7
the determination of oscillator strength values (P) of the F0 →
7
5D2 transition whereas the intensities of the 5D0 → F2 transition
5
7
were calibrated to the D0 → F1 ones. Plots of P and k (where
k is a decay constant rate of the excited state) changes as a
function of added NaHCO3 are shown in Fig. 6. Fitting of the
three curves to equations which combine the measured quantities
−
with concentrations of EuL and A (where A is HCO3 anion) as
well as the dissociation KD and association KA constants of the
EuLA complex were performed. The KD and KA constants of the
equilibrium given by eqn (2):
Conclusions
The X-ray analysis of the EDTMP crystal with the EuIII ion
has shown a rare example of lanthanide ion coordination with
a bifunctional carbonate anion. Therefore the inner sphere of the
EuIII ion in the [C(NH2)3]7[Eu(EDTMP)(CO3)]·10H2O crystal is
filled by donor atoms (four O and two N) of the EDTMP ligand
in a similar way as was revealed in the crystal structure of its
EuLA ⇔ EuL + A
(2)
are expressed as:
KD = [EuL][A]/[EuLA], KA = 1/KD
(3)
Fig. 6 Oscillator strengths and luminescence decay rates as function of added NaHCO3 to the EuIII–EDTMP complex at pH 7.5.
This journal is The Royal Society of Chemistry 2006 Dalton Trans., 2006, 4702–4710 | 4707
©

View Article Online
carboxylate analogue, EDTA, and additionally, by two oxygen
atoms from the carbonate anion.
solution of Eu(ClO4)3 (1 and 2 mmol), respectively. The resulting
solutions were basified with guanidinium carbonate to pH 10.7.
Colourless crystals of [C(NH2)3]7[Eu(EDTMP)(CO3)]·10H2O (I)
and [C(NH2)3]7[Eu(EDTA)(H2O)3] (II) were formed during slow
evaporation.
The comparison between spectral results of the crystal and
EuIII–EDTMP solutions of which the pH was adjusted by
NaOH, Na2CO3 and [C(NH2)3]2CO3 allowed us to show that the
formed anionic complex of [Eu(EDTMP)(CO3)]7−, though highly
negative, is thermodynamically stable. A very good fitting of a
carbonate anion into the coordination space vacated by water
molecules and/or hydroxy groups is a reason that the kinetically
labile complex becomes inert. The high affinity of the carbonate
anion towards the EuIII–EDTMP complex was confirmed by the
determined low dissociation constant.
X-Ray crystal analysis
An appropriate crystal was cut from a larger one and mounted
on a Kuma KM4 diffractometer equipped with a CCD counter.
The collected data were corrected for polarization, Lorentz and
absorption, the latter calculated from the crystal habits captured
from photo scans. The positions of Eu were found from Patterson
maps, and the remainder of non-H atoms from difference Fourier
maps. The positions of the C-bonded hydrogen atoms were
calculated geometrically. In I 16 H atoms from 8 water molecules
were located and refined freely (except HW91 which had its
temperature factor fixed). In II all 6 water H atoms were found
and also refined freely. The refinement was full-matrix with all
non-H atoms anisotropic. All computations were performed using
SHELXS97 and SHELXL97 programs.32 The molecular graphics
was prepared with DIAMOND.33
The presented results may be helpful in understanding of the
uptake mechanism of the 153SmIII–EDTMP complex by metastatic
bones and may provide some indications for future ligand design
for therapeutic complexes with lanthanide radionuclides.
Experimental
Materials and samples preparation
Stock solutions of europium chloride and perchlorate were
prepared from Eu2O3 (99.99% Stanford Materials) by dissolving
it under heating in 2 mol dm−3 hydrochloric and perchloric
acids, respectively. The EuIII concentrations were determined
complexometrically using xylenol orange as an indicator. For
spectral measurements of EuIII : L = 1 : 1.1 solutions (cEu = 4.0 ×
10−2 M for EuIII–EDTMP and cEu = 2.0 × 10−2 M for EuIII–
EDTA) of which pH was adjusted by NaOH, the stock solutions
of EDTMP (ABCR) and EDTA (Merck) adjusted to pH around
7 with NaOH were used. Stock solutions of EDTMP and EDTA
with carbonate were prepared by neutralization of the ligand by
Na2CO3 or guanidinium carbonate to pH 7.0. These solutions
were added to EuCl3 solutions in the ratio of EuIII : L equal to 1 :
Selected crystal data for I. C14H74EuN23O25P4, M = 1240.80,
monoclinic, space group P21/n, Z = 4, a = 12.407(4), b =
◦
3
˚
˚
16.606(6), c = 24.071(9) A, b = 101.85(3) , V = 4854(3) A , l =
1.525 mm−1, Dc = 1.698 g cm−3, F(000) = 2568, crystal size =
0.4 × 0.25 × 0.20 mm, h = 3–28.5◦, index ranges: −16 ≤ h ≤
16, −21 ≤ k ≤ 17, −31 ≤ l ≤ 31, reflections collected/unique =
31261/11207 (Rint = 0.0415). Final R indices [I > 2r(I)] R(F) =
0.0336, Rw(F2) = 0.0617 and R(F) = 0.0618, Rw(F2) = 0.0660 (all
data). Data completeness to 2h = 28.5◦, 91.2%. Largest diff. peak
−3
˚
and hole 1.021 and −0.590 e A .
1.1 (cEu = 4.0 × 10−2 M for EuIII–EDTMP and cEu = 2.0 × 10−2
M
Selected crystal data for II. C11H24EuN5O11, M = 554.31,
monoclinic, space group P21/c, Z = 4, a = 11.269(5), b = 8.646(4),
for EuIII–EDTA) and then the pH of the solutions was adjusted
by Na2CO3 or guanidinium carbonate, respectively.
◦
3
−1
˚
˚
c = 18.718(9) A, b = 90.07(4) , V = 1823.7(15) A , l = 3.509 mm ,
Dc = 2.019 g cm−3, F(000) = 1104, crystal size = 0.4 × 0.17 ×
0.1 mm, h = 3–28.5◦, index ranges: −14 ≤ h ≤ 14, −11 ≤ k ≤ 10,
The buffered solutions of the EuIII–EDTMP complex with
−
different concentrations of HCO3 ligand were prepared as
follows. The stock solutions of the EuIII–EDTMP complex and
of HEPES were adjusted with NaOH to pH 7.5 and then mixed
together. To a fixed amount of the latter solution the calculated
amount of NaHCO3 was added and the solutions were diluted to
10 cm3. The final concentrations of particular components in such
−23 ≤ l ≤ 24, reflections collected/unique = 11182/4233 (Rint
=
0.0369). Final R indices [I > 2r(I)] R(F) = 0.0211, Rw(F2) = 0.0552
and R(F) = 0.0234, Rw(F2) = 0.0565 (all data). Data completeness
to 2h = 28.5◦, 92.0%. Largest diff. peak and hole 0.823 and
−3
˚
−0.696 e A .
prepared solutions were: cEu = 4.0 × 10−2 M, cHCO = 0.90 M,
−
3
cHEPES = 0.50 M.
CCDC reference numbers 297437 and 297438.
For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b606420e
For NMR and luminescence lifetimes measurements in D2O
each of the solutions prepared earlier in water was evaporated to
dryness and then dissolved in D2O. This procedure was repeated
three times.
All measurements of solutions were performed after one day of
equilibration.
UV-Vis measurements
All electronic absorption spectra were recorded on a Cary 500
UV-VIS-near-IR spectrophotometer. Corrected emission spectra
were taken on a SLM Aminco 500 spectrofluorometer. The
luminescence of the EuIII ion was excited with a wavelength
of 394 nm. The luminescence decay curves were detected on
Edinburgh Instruments FLS 920 Spectrometer with the monitored
emission at 614 nm.
Crystal synthesis
Crystals of EuIII–EDTMP and EuIII–EDTA complexes were
prepared as follows. The ligand solutions of EDTMP (0.48 g,
1.1 mmol) and EDTA (0.64 g, 2.2 mmol) were neutralized
with [C(NH2)3]2CO3 to pH around 7 and added to an aqueous
4708 | Dalton Trans., 2006, 4702–4710
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
NMR measurements
935–940; (c) I. G. Finlay, M. D. Mason and M. Shelley, Lancet Oncol.,
2005, 6, 392–400.
The 31P NMR spectra were recorded on a Bruker AMX 500
spectrometer at 298 K. 5% H3PO4 in D2O solution was used as an
external standard.
4 (a) I. Lukesˇ, J. Kotek, P. Vojt´ısˇek and P. Hermann, Coord. Chem. Rev.,
2001, 216–217, 287–312; (b) R. L. Luck, C. L. Maupin, D. Parker, J. P.
Riehl and J. A. G. Williams, Inorg. Chim. Acta, 2001, 317, 331–337; (c) F.
Avecilla, J. A. Peters and C. F. G. C. Geraldes, Eur. J. Inorg. Chem.,
2003, 4179–4186; (d) J. Rudovsky´, P. C´ıgler, J. Kotek, P. Hermann, P.
Vojt´ısˇek, I. Lukesˇ, J. A. Peters, L. V. Elst and R. N. Muller, Chem.
Eur. J., 2005, 11, 2373–2384.
5 I. N. Marov, L. W. Ruzaykina, W. A. Ryabuchin, P. A. Korovaykov and
N. M. Dyatlova, Koord. Khim., 1977, 3, 1333–1339.
6 E. N. Rizkalla and G. R. Choppin, Inorg. Chem., 1983, 22, 1478–1482.
7 R. C. Holz, G. E. Meister and W. DeW. Horrocks, Jr., Inorg. Chem.,
1990, 29, 5183–5189.
Calculations. The luminescence decay lifetimes of the emitting
EuIII level were obtained by fitting the decay curves to a single
exponential function. Emission and absorption intensities of
the f–f transitions were calculated by integrating their bands.
Absorption intensities were determined as oscillator strengths (P)
by using eqn (4):
8 S. T. Frey and W. DeW. Horrocks, Jr., Inorg. Chem., 1991, 30, 1073–
1079.
ꢀ
P = 4.32 × 10−9(cd)−1 A(r) dr
where: c is the concentration of the EuIII ion in M, d is the length
of the optical pathlength in cm and A(r) is the absorbance as the
function of the wavenumber in cm−1.
(4)
9 S. Anis and G. R. Choppin, Inorg. Chim. Acta, 1998, 278, 185–189.
10 (a) M. I. Kabachnik, N. M. Dyatlova, T. T. Medved, Yu. F. Begulin and
V. V. Sidolenko, Dokl. Acad. Nauk SSSR, 1967, 175, 351–353; (b) K.
Sawada, M. Kuribayashi, T. Suzuki and H. Miyamoto, J. Solution
Chem., 1991, 20, 829–839.
11 N. V. Jarvis, J. M. Wagener and G. E. Jackson, J. Chem. Soc., Dalton
Trans., 1995, 1411–1415.
The following eqn (5) was used for the dissociation constant KD
calculation:
12 G. C. de Witt, P. M. May, J. Webb and G. Hefter, BioMetals, 1996, 9,
351–361.
X − XEuL
2[EuL]
= [EuL] + [A] + KD
13 (a) R. Ruloff, P. Prokop, J. Sieler, E. Hoyer and L. Beyer, Z. Natur-
forsch., Teil B, 1996, 51, 963–968; (b) R. Ruloff, T. Gelbrich, J. Sieler
and E. Hoyer, Z. Naturforsch., Teil B, 1997, 52, 805–809.
14 (a) A. Mondry and P. Starynowicz, Polyhedron, 2000, 19, 771–777;
(b) A. Mondry and P. Starynowicz, New J. Chem., 2000, 24, 603–607;
(c) R. Janicki, A. Mondry and P. Starynowicz, Z. Anorg. Allg. Chem.,
2005, 631, 2475–2477; (d) A. Mondry and P. Starynowicz, Eur. J. Inorg.
Chem., 2006, 1859–1867.
15 J. Gałęzowska, R. Janicki, A. Mondry, R. Burgada, T. Bailly, M. Lecou-
vey and H. Kozłowski, Dalton Trans., 2006, DOI: 10.1039/b601941b.
16 (a) K. Nakamura, T. Kurisaki, H. Wakita and T. Yamaguchi, Acta
Crystallogr., Sect. C, 1995, 51, 1559–1563; (b) J. Wang, X. D. Zhang,
W. G. Jia, Y. Zhang and Z. R. Liu, Koord. Khim., 2004, 30, 141–147;
(c) J. Wang, X. D. Zhang, Y. Zhang, Y. Wang, X. Y. Liu and Z. R. Liu,
Koord. Khim., 2004, 30, 901–909.
XEuLA − XEuL
ꢁ
−
([EuL] + [A] + KD)2 − 4[EuL][A]
(5)
−
where: [EuL] is the complex concentration, [A] is the HCO3
concentration, X is the measured quantity: oscillator strength
7
5
5
7
value of the F0 → D2 transition, intensity of the D0 → F2
transition scaled to the 5D0 → F1 intensity, k is the luminescence
7
decay rate in ms−1 equal to s−1 and s is the luminescence lifetime
in ms. The XEuLA values were determined for the EuL solution
saturated by NaHCO3, while the XEuL values were determined for
the solution of which the pH adjusted by NaOH. The curves X =
f ([NaHCO3]) were fitted by an iterative least-squares procedure
in the Microsoft Excel programme. All data were well fitted to
the linear dependence with the coefficients of determination (R2 ≥
0.996).
17 G. V. Polyanchuk, L. M. Shkolnikova, M. V. Rudomino, N. M.
Dyatlova and S. S. Makarevich, Zh. Strukt. Khim., 1985, 26, 109–
118.
18 (a) M. R. Spirlet, J. Rebizant, J. F. Desreux and M. F. Loncin, Inorg.
Chem., 1984, 23, 359–363; (b) C. A. Chang, L. C. Francesconi, M. F.
Malley, K. Kumar, J. Z. Gougoutas, M. F. Tweedle, D. W. Lee and L. J.
Wilson, Inorg. Chem., 1993, 32, 3501–3508; (c) S. Aime, A. Barge, M.
Botta, M. Fasano, J. D. Ayala and G. Bombieri, Inorg. Chim. Acta,
1996, 246, 423–429; (d) S. Aime, A. Barge, F. Benetollo, G. Bombieri,
M. Botta and F. Uggeri, Inorg. Chem., 1997, 36, 4287–4289; (e) F.
Benetollo, G. Bombieri, S. Aime and M. Botta, Acta Crystallogr., Sect.
C, 1999, 55, 353–356; (f) F. Benetollo, G. Bombieri, L. Calabi, S. Aime
and M. Botta, Inorg. Chem., 2003, 42, 148–157.
19 (a) G. Bombieri, F. Benetollo, A. Polo, K. K. Fonda and L. M.
Vallarino, Polyhedron, 1991, 10, 1385–1394; (b) C. A. Chang, L. C.
Francesconi, M. F. Malley, K. Kumar, J. Z. Gougoutas, M. F. Tweedle,
D. W. Lee and L. J. Wilson, Inorg. Chem., 1993, 32, 3501–3508; (c) S.
Fleming, C. D. Gutsche, J. M. Harrowfield, M. I. Ogden, B. W.
Skelton, D. F. Stewart and A. H. White, Dalton Trans., 2003, 3319–
3327.
20 (a) J.-C. G. Bu¨nzli, Lanthanide Probes in Life, Chemical and Earth
Sciences. Theory and Practice, ed. J.-C. G. Bu¨nzli and G. R. Choppin,
Elsevier, Amsterdam, 1989, pp. 219–290; (b) C. Go¨rller-Walrand and
K. Binnemans, Handbook of the Physics and Chemistry of Rare Earths,
ed. K. A. Gschneidner, Jr. and L. Eyring, Elsevier, Amsterdam, 1996,
vol. 23, pp. 163–164.
21 (a) M. Albin and W. DeW. Horrocks, Jr., Inorg. Chem., 1985, 24, 895–
900; (b) S. T. Frey and W. DeW. Horrocks, Jr., Inorg. Chim. Acta, 1995,
229, 383–390; (c) G. R. Choppin and Z. M. Wang, Inorg. Chem., 1997,
36, 249–252.
Acknowledgements
The authors are very grateful to Dr P. Starynowicz for helpful
discussion, the X-ray measurements and for assistance during
the structure solutions as well as to Dr M. Karbowiak for
luminescence lifetime measurements.
References
1 (a) A. D. Sherry and C. F. G. C. Geraldes, Lanthanide Probes in Life,
Chemical and Earth Sciences. Theory and Practice., ed. J.-C. G. Bu¨nzli
and G. R. Choppin, Elsevier, Amsterdam, 1989, pp. 117–123; (b) J. A.
Peters, J. Huskens and D. J. Raber, Prog. Nucl. Magn. Reson. Spectrosc.,
1996, 28, 283–350; (c) S. Aime, M. Botta, M. Fasano and E. Terreno,
Chem. Soc. Rev., 1998, 27, 19–29; (d) P. Caravan, J. J. Ellison, T. J.
McMurry and R. B. Lauffer, Chem. Rev., 1999, 99, 2293–2352; (e) D.
Parker, R. S. Dickins, H. Puschmann, C. Crossland and J. A. K.
Howard, Chem. Rev., 2002, 102, 1977–2010; (f) D. Parker, Chem. Soc.
Rev., 2004, 33, 156–165; (g) T. Gunnlaugsson and J. P. Leonard, Chem.
Commun., 2005, 3114–3131.
22 R. D. Peacock, Struct. Bonding (Berlin), 1975, 22, 83–219.
23 S. F. Mason, Struct. Bonding (Berlin), 1980, 39, 43–79.
24 (a) Yu. P. Galaktionov and K. V. Astakhov, Zh. Neorg. Khim., 1963,
8, 896–904; (b) T. F. Gritmon, M. P. Goedken and G. R. Choppin,
J. Inorg. Nucl. Chem., 1977, 39, 2021–2023.
2 W. A. Volkert and T. J. Hoffman, Chem. Rev., 1999, 99, 2269–2292.
3 (a) H. T. Gasiglia and H. Okada, J. Radioanal. Nucl. Chem. Lett., 1995,
199, 295–304; (b) W. K. A. Louw, I. C. Dormehl, A. J. van Rensburg,
N. Hugo, A. S. Alberts, O. E. Forsyth, G. Beverley, M. A. Sweetlove, J.
Marais, M. G. Lo¨tter and A. van Aswegen, Nucl. Med. Biol., 1996, 23,
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 4702–4710 | 4709
©

View Article Online
25 (a) G. Geier and C. K. Jørgensen, Chem. Phys. Lett., 1971, 9, 263–
265; (b) N. A. Kostromina and N. N. Tananaeva, Zh. Neorg. Khim.,
1971, 16, 256; (c) J. Dexpert-Ghys, J. Halwani and B. Piriou, Inorg.
Chim. Acta, 1987, 139, 303–306; (d) H. G. Brittain, G. R. Choppin and
P. P. Barthelemy, J. Coord. Chem., 1992, 26, 143–153; (e) M. Latva, J.
Kankare and K. Haapakka, J. Coord. Chem., 1996, 38, 85–99.
26 R. M. Supkowski and W. DeW. Horrocks, Jr., Inorg. Chem., 1999, 38,
5616–5619.
30 R. S. Dickins, S. Aime, A. S. Batsanov, A. Beeby, M. Botta, J. I.
Bruce, J. A. K. Howard, C. S. Love, D. Parker, R. D. Peacock and
H. Puschmann, J. Am. Chem. Soc., 2002, 124, 12697–12705.
31 (a) S. Aime, A. Barge, M. Botta, J. A. K. Howard, R. Kataky, M. P.
Lowe, J. M. Moloney, D. Parker and A. S. de Sousa, Chem. Commun.,
1999, 1047–1048; (b) K. Kimpe, W. D’Olieslager, C. Go¨rller-Walrand,
A. Figueirinha, Z. Kova´cs and C. F. G. C. Geraldes, J. Alloys Compd.,
2001, 323–324, 828–832; (c) Y. Bretonnie`re, M. J. Cann, D. Parker and
R. Slater, Chem. Commun., 2002, 1930–1931; (d) Y. Bretonnie`re, M. J.
Cann, D. Parker and R. Slater, Org. Biomol. Chem., 2004, 2, 1624–1632.
32 (a) G. M. Sheldrick, SHELXS-97, program for structure solution,
University of Go¨ttingen, 1997; (b) G. M. Sheldrick, SHELXL-97,
program for structure refinement, University of Go¨ttingen, 1997.
33 DIAMOND-Visual Crystal Structure Information System, CRYSTAL
IMPACT, Postfach 1251, D-53002 Bonn, Germany.
27 R. M. Supkowski and W. DeW. Horrocks, Jr., Inorg. Chim. Acta, 2002,
340, 44–48.
28 R. S. Dickins, T. Gunnlaugsson, D. Parker and R. D. Peacock, Chem.
Commun., 1998, 1643–1644.
29 J. I. Bruce, R. S. Dickins, L. J. Govenlock, T. Gunnlaugsson, S.
Lopinski, M. P. Lowe, D. Parker, R. D. Peacock, J. J. B. Perry, S.
Aime and M. Botta, J. Am. Chem. Soc., 2000, 122, 9674–9684.
4710 | Dalton Trans., 2006, 4702–4710
This journal is
The Royal Society of Chemistry 2006
©