Luminescence and Relaxometric Properties of Heteropolymetallic Metallostar Complexes with Selectively Incorporated Lanthanide(III) Ions

Article abstract of DOI:10.1002/ejic.201500648

The synthesis and characterization of two diethylenetriaminepentaacetic acid (DTPA) based heteropolymetallic metallostar lanthanide complexes with the general formulas (GdL¹)₃Ln and (GdL²)₃Ln are described. The

Full text of DOI:10.1002/ejic.201500648

FULL PAPER
DOI:10.1002/ejic.201500648
Luminescence and Relaxometric Properties of
Heteropolymetallic Metallostar Complexes with Selectively
Incorporated Lanthanide(III) Ions
Matthias Ceulemans,[a] Elke Debroye,[a] Luce Vander Elst,[b]
Wim De Borggraeve,[a] and Tatjana N. Parac-Vogt*[a]
Keywords: Metallostars / Imaging agents / Luminescence / Gadolinium / Lanthanides
The synthesis and characterization of two diethylenetri-
aminepentaacetic acid (DTPA) based heteropolymetallic
metallostar lanthanide complexes with the general formulas
(GdL1)3Ln and (GdL2)3Ln are described. The synthesis uses
a synthetic approach recently developed in our group for the
selective complexation of gadolinium(III) and luminescent
for all (GdL1)3Ln complexes than for (GdL2)3Ln. A large in-
crease of the quantum yield from 1.5 to 9.8% was observed
for the (GdL1)3Eu complex compared with (GdL2)3Eu,
whereas the (GdL1)3Tb complex exhibited a quantum yield
(QY) of 30.9% compared with 15.3% for (GdL2)3Tb. A slight
increase of the QY from 0.8 to 1.2% was observed for the
lanthanide ions with a ditopic ligand to form highly paramag- Dy(III) complex when switching from ligand L2 to L1. The
netic and luminescent metallostar complexes. The lumines-
cence data and relaxometric studies suggest the potential ap-
plicability of the complexes as bimodal contrast agents for
magnetic resonance and optical imaging. Owing to the
higher excited state of L1, better sensitization was observed
nuclear magnetic relaxation dispersion (NMRD) measure-
ments of the (GdL2)3Ln complexes (Ln = EuIII, DyIII, TbIII)
showed respective longitudinal relaxivity (r1) values of 24.27,
22.80 and 21.72 s–1 mmol–1 per metallostar complex at 310 K
and 20 MHz.
22.5 and 25.3 for DTPA and DOTA, respectively.[8–10]
Furthermore, the eightfold coordination allows the binding
of one water molecule to the metal centre, and the residence
time of this water molecule (τm) is one of the four main
parameters that govern the relaxivity r1. The relaxivity de-
scribes the effectiveness of a contrast agent to induce relax-
ation and is defined by the increase of the longitudinal re-
laxation rate in s–1 measured in a 1 mm gadolinium(III)
solution. In addition to the water residence time (τm), the
other parameters that influence relaxivity are the relaxation
behaviour of the electron spin of the gadolinium(III) ions
(τS1,2), the rotational correlation time of the complex in
solution (τR) and the amount of water molecules directly
bound to the gadolinium(III) ion (q).[9] Although the gado-
linium(III) complexes currently used clinically have relaxivi-
ties of ca. 4–5 s–1 mm–1, an increase of up to 100 s–1 mm–1
could be achieved through the optimization of these dif-
ferent parameters according to the Solomon–Bloembergen–
Morgan theory.[9,11] A frequently used approach towards
increasing the relaxivity focuses on the increase of the rota-
tional correlation time, for example, by conjugation of GdIII
chelates into linear polymers or dendrimers.[10,12–14] An-
other approach includes noncovalent interactions, such as
the binding of chelates to human serum albumin, the most
abundant protein in human plasma.[4,15,16] The formation
of supramolecular structures such as micelles or liposomes
from amphiphilic gadolinium(III) complexes[17–21] results in
a high density of paramagnetic species and, simultaneously,
Introduction
The interest in contrast agents as efficient, responsive
and tissue-specific markers has grown tremendously since
the introduction of magnetic resonance imaging (MRI) as
a diagnostic tool. Gadolinium(III) chelates have been used
widely as MRI contrast agents.[1–5] Gadolinium(III) ions
can efficiently induce relaxation of water molecules owing
to their seven unpaired electrons, which produce a large
8
magnetic moment (7.94 μB), and a symmetric S7/2 ground
state, which provides relatively long electron relaxation
times.[1] Owing to the toxicity of free gadolinium(III) ions
(LD50 = 0.2 mmolkg–1 in mice) and the relatively high
doses of contrast agent needed (0.1–0.3 mmol per kg of
body weight), the use of strong chelates is necessary.[6,7] Two
of the most commonly used ligands for gadolinium(III) in
modern molecular imaging techniques are the acyclic dieth-
ylenetriaminepentaacetic acid (DTPA; Magnevist®, Bayer
Shering Pharma AG) and the cyclic 1,4,7,10-tetraazacyclo-
dodecane-1,4,7,10-tetraacetic acid (DOTA; Dotarem®,
Guerbet).[2] The eightfold coordination of these ligands to
gadolinium(III) ions ensures stable complexes with logK =
[a] Department of Chemistry, KU Leuven,
Celestijnenlaan 200F, 3001 Leuven, Belgium
E-mail: Tatjana.Vogt@chem.kuleuven.be
http://chem.kuleuven.be/lbc/
[b] Department of General, Organic and Biomedical Chemistry,
University of Mons,
Place du Parc 23, 7000 Mons, Belgium
Eur. J. Inorg. Chem. 2015, 4207–4216
4207
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
an increase in the rotational correlation time.[22] Although this new ligand have been synthesized and characterized.
these approaches effectively increase the overall relaxivity The magnetic and luminescence properties of the complexes
of the contrast agent, the high theoretical maximum effi- have been studied, and their potential as bimodal contrast
ciency has unfortunately not yet been reached.
agents for MRI and optical imaging has been evaluated.
The MRI technique excels in its high spatial resolution
and tissue penetration but suffers from a low sensitivity.
Therefore, to remedy the low sensitivity of MRI, bimodal
imaging has gained importance in recent years. Clinically,
bimodal imaging combining positron emission tomography
(PET) and MRI is already available.[23] This technique is
Results and Discussion
Ligands and Complexes
The general ligand design is based on the attachment of
able to provide high-quality three-dimensional images of a DTPA scaffold through an amide linkage to para-substi-
soft tissue combined with the high sensitivity of PET im- tuted dipicolinic acid (DPA), which has previously been
aging. On the other hand, optical imaging (OI) is also a demonstrated to form tris complexes with lanthanide(III)
very sensitive technique for which no ionizing radiation is ions.[30] This approach allows the incorporation of different
necessary. The combination of both MRI and OI into one antennae between the two coordinating moieties to sensitize
probe would allow images to reveal more details than those the luminescent lanthanide ions. The first method has been
obtained from both techniques separately. The use of one reported previously by our group and resulted in the syn-
contrast agent for both techniques would ensure the same thesis of a EuIII metallostar complex with the structure
distribution and reduce the stress imposed on the body by shown in Scheme 1.[30] A Suzuki–Miyaura coupling[32] was
two different probes. The bimodal contrast agents for MRI/ used to obtain a fully protected dipicolinate derivative,
OI with organic fluorophores have several shortcomings in- which allowed to selectively incorporate gadolinium(III)
cluding small Stokes shifts, short luminescence lifetimes and ions into the DTPA unit. The subsequent removal of the
photobleaching. Therefore, increasing attention has been protecting groups and coordination to the luminescent eu-
focused on luminescent lanthanide-based systems, which ropium(III) ions yielded the desired metallostar complex.
can emit light in the visible[24] and near-infrared regions.[25] In this work, the complexation to other luminescent lantha-
However, tissue penetration is a very important factor when nide ions (LnIII = TbIII, DyIII, SmIII, HoIII, TmIII and
deciding which lanthanide to choose for in vivo probes. A YbIII) has been performed by the same approach. This re-
good trade-off between the image resolution and penetra- sulted in a range of metallostar complexes that all contain
tion depth can be made in the wavelength region 665– three gadolinium(III)–DTPA units but differ in the nature
900 nm.[26] Mixed-complex systems have been reported in of the central lanthanide(III) ion bound to the dipicolinate
which the generation of mixed micelles produces probes units (Scheme 1).
with high relaxivities and luminescence. Alternatively, the
The synthesis of the new ligand L2 makes use of Sonoga-
use of heteropolymetallic complexes has been investi- shira cross-couplings and employs a slightly altered syn-
gated.[15,27–29] In addition to gadolinium(III) chelating moi- thetic pathway than the one previously reported by Bünzli
eties, these complexes must contain antenna links surround- et al.[33] The first step is the protection of 4-iodobenzyl-
ing an emissive metal ion. However, the control of the site- amine with t-boc-anhydride (boc = butoxycarbonyl) to
selective incorporation of lanthanide ions within one ligand form 1 (Scheme 2). Compound 1 is subsequently coupled
scaffold is a challenging task because of the similar com- to trimethylsilylacetylene through a Sonogashira cross-cou-
plexation abilities across the lanthanide series.[30]
pling to form 2, and subsequent deprotection of the tri-
In a previous paper, we reported the successful synthesis methylsilyl group yields 3. The NMR spectrum clearly
of the heterotetrametallic complex (GdL1)3Eu, in which two shows the disappearance of the signals of the highly shi-
different lanthanide(III) ions were selectively incorporated elded trimethylsilyl protons at δ = 0.24 ppm and the appear-
into one ligand, L1. The resulting complex contained three ance of a single peak of the ethynyl group at δ = 3.06 ppm.
Gd-DTPA moieties linked to a europium(III) chelate con- Compound 3 is able to undergo another Sonogashira cross-
sisting of a para-substituted dipicolinic acid (DPA) and ex- coupling with dimethyl 4-bromo-2,6-pyridinedicarboxylate
hibited both high relaxivity, owing to the presence of three under the same conditions to form 4.
gadolinium(III) ions, and favourable europium(III) lumi-
The t-Boc group is selectively deprotected by using tri-
nescence.[30] Encouraged by these results, in this work, we fluoroacetic acid in dichloromethane (50:50), as confirmed
further extend this strategy toward the creation of potential by the disappearance of the tert-butyl peaks at δ = 1.45 ppm
bimodal contrast agents with selectively incorporated lan- and the persistence of the methyl peaks at δ = 4.06 ppm in
thanide ions. A series of (GdL1)3Ln complexes, in which Ln the 1H NMR spectrum. This approach results in a free
represents different luminescent lanthanide(III) ions, have amine (5), which can be coupled to the one free acid of the
been prepared. In addition, the new ligand L2, which con- DTPA tert-butyl ester through amide synthesis to afford a
tains an ethynyl linker between the two aromatic centres, fully protected ligand (6). Finally, the DTPA tert-butyl es-
has been synthesized with the aim to decrease the excited ters are selectively cleaved with a 6 m HCl solution to afford
triplet state of the ligand to provide better energy transfer 7 (Scheme 3) The acidic groups of DTPA can then be fur-
to ytterbium(III) ions.[31] By the same selective synthetic ap- ther coordinated to gadolinium(III) ions. The complexation
proach, a series of heterotetrametallic complexes based on was performed with GdCl3·6H2O in pyridine, and the ab-
Eur. J. Inorg. Chem. 2015, 4207–4216
4208
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Scheme 1. Schematic representation of the complex GdL1 and the heterotetrametallic metallostar (GdL1)3Ln (Ln = EuIII[30] DyIII, TbIII
,
HoIII, NdIII, SmIII, TmIII and YbIII).
of the methyl ester protecting groups from the dipicolinate
moiety was performed under alkaline conditions, and the
final ligand was mixed with a luminescent lanthanide
(LnCl3·xH2O) salt (LnIII = EuIII, TbIII, DyIII, SmIII, HoIII
,
TmIII, YbIII). By this approach, the desired metallostar tris
complexes (GdL2)3Ln were generated, as is schematically
shown in Scheme 4.
Scheme 2. Synthetic pathway to L2: (a) boc-anhydride; (b) ethynyl-
trimethylsilane, PdCl2(PPh3)2; (c) 1 m tetra-n-butylammonium
fluoride (TBAF); (d) dimethyl 4-bromo-2,6-pyridinedicarboxylate,
PdCl2(PPh3)2.
Photophysical Properties of the (GdL1)3Ln and (GdL2)3Ln
Complexes
The absorption spectrum of the GdL1 complex has a
sence of free lanthanide ions was verified by the addition well-defined maximum at λ ≈ 282 nm (ε = 4250 cm–1 m–1)
of an arsenazo indicator solution.[34] The selective removal caused by the πǞπ* transition of DPA.[35] The excitation
Scheme 3. Synthetic pathway to L2: (a) DTPA tert-butyl ester, O-(benzotriazol-1-yl)-N,N,NЈ,NЈ-tetramethyluronium tetrafluoroborate
(TBTU); (b) 6 m HCl.
Eur. J. Inorg. Chem. 2015, 4207–4216
4209
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Scheme 4. Schematic representation of the complexes GdL2 and the heterotetrametallic metallostars (GdL2)3Ln (Ln = EuIII, DyIII, TbIII
,
NdIII, SmIII, TmIII, and YbIII).
spectrum shows a broad band between λ = 250 and 300 nm
with a maximum at 295 nm. The absorption spectra of
GdL2 and the tris complex (GdL2)3Ln both show an ab-
sorption maximum at λ = 315 nm. The molar extinction
coefficient of the free complex GdL2 at this maximum is
ε315 = 6692 cm–1 m–1, whereas the tris complex has ε315
=
11000 cm–1 m–1. In the emission spectra of the (GdL1)3Eu
and (GdL2)3Eu metallostar complexes, the splitting of the
5D0Ǟ7F1 transition into three bands suggests a slightly de-
formed D3 symmetry around the central ion (Fig-
ure 1).[30,36] This is further supported by the ratio of the
5
5D0Ǟ7F2 and D0Ǟ7F1 transitions, which is ca. 5–6.
The luminescence decays of the metallostar complexes
(GdL1)3Eu and (GdL2)3Eu have been measured in water
and D2O, and the results are shown in Table 1. The shorter
luminescence lifetime in water than that in D2O is due to
the presence of inner-sphere high-energy O–H vibrations
and can be used to determine the number of coordinated
water molecules. The best fit was observed by applying a
biexponential decay, which suggests the presence of two dif-
ferent species in solution. The phenomenological equation
for LnIII–polyaminocarboxylate systems has been employed
to determine the hydration number q with an accuracy of
Ϯ0.1; see Equation (1).[37,38]
Figure 1. Corrected and normalized luminescence spectra of
(GdL1)3Eu (λexc = 293 nm, 298 K) and (GdL2)3Eu, (λexc = 315 nm,
298 K).
pressed in ms–1 for EuIII. The qX values represent the
number of OH, NH or CONH groups bound directly to
the lanthanide centre. For these calculations, only the con-
tribution of the amide groups has been considered, that is,
qCONH = 1. The results indicate that an equilibrium has
been set between the bis and tris complexes at the low
qEu(H O) = 1.11(Δkobs – 0.31 + 0.44qOH + 0.99qNH + 0.075qCONH
)
2
(1)
In Equation (1), Δkobs represents the difference of the de- concentrations used for luminescence measurements
cay rate constants [kH O = 1/τH O and kD O = 1/τD O] ex- (2.0ϫ10–5 m). These findings are consistent with those pre-
2
2
2
2
Eur. J. Inorg. Chem. 2015, 4207–4216
4210
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Table 1. Luminescence lifetimes of the different (GdL)3Ln com-
plexes (Ln = EuIII, DyIII, TbIII), average number of calculated water
molecules in the first coordination sphere of the Ln ion (q) and the
ratio of tris/bis complexes under the conditions of the measure-
ments (2.0ϫ10–5 m).
(3)
(4)
H2O
D2O
q
Tris/bis ration
(GdL1)3Eu[a] 0.23 ms
1.12 ms
39 μs
2.42 ms
1.11 ms
14 μs
0.6
1.2
1.3
0.2
0.3
0.8
80:20
85:15
55:45
96:4
90:10
73:27
(GdL1)3Dy
(GdL1)3Tb
(GdL2)3Eu
(GdL2)3Dy
(GdL2)3Tb
9 μs
of the magnetic dipole (MD). The intrinsic quantum yield
1.42 ms
0.20 ms
9 μs
Eu
values Q for both GdL1 and GdL2 are in the range 22–
Eu
24%. Finally, we can acquire the ratio between the quan-
tum yield under ligand excitation and the intrinsic quantum
yield to obtain the sensitization efficiency (ηsens) of the li-
gand; see Equation (5).
2.20 ms
2.40 ms
[a] From ref.[30]
viously reported for (DPA)3Eu complexes, for which a sig-
nificant amount of the bis complex was detected in micro-
molar concentrations.[39] As no water is present in the first
coordination sphere of the central ion in the tris complex,
whereas the bis complex should have three water molecules
bound, this allows the determination of the ratio between
the tris and bis complexes. An increase of the ratio of tris
to bis complex from 80:20 for (GdL1)3Eu to 95:5 for
(GdL2)3Eu is seen upon the introduction of an ethynyl
linker to the ligand structure. The luminescence quantum
yields QLLN were determined upon ligand excitation by a
comparative method with a solution of rhodamine 101 in
ethanol (Q = 100%) as the standard. The quantum yield
was determined according to Equation (2):
(5)
This results in sensitization efficiencies of 47% for
(GdL1)3Eu and only 6.3% for (GdL2)3Eu. These values are
lower than those for the parent (DPA)3Eu complex, which
has QEu = 24% and ηsens = 61%.[35]
Dysprosium(III) complexes with GdL1 and GdL2
showed yellow emission owing to F9/2Ǟ6HJ (J = 15/2, 13/
4
2 and 11/2) transitions. However, owing to the poor sensiti-
zation of the ligands towards dysprosium(III) ions, some
ligand emission is also observed (Figure 2). (GdL1)3Dy
clearly shows the three transitions of the Dy(III) centre with
some ligand emission, whereas the (GdL2)3Dy emission
spectrum is dominated by ligand emission. The 4F9/2Ǟ6H15/2
transition, which is typically observed at λ = 485 nm, has
completely disappeared under the ligand emission tail.
However, the 4F9/2Ǟ6H13/2 transition, which is located
at λ ≈ 575 nm is clearly visible together with the weak
4F9/2Ǟ6H11/2 transition at λ = 665 nm. The quantum yields
for the dysprosium(III) samples have been measured with a
solution of rhodamine 101 in ethanol (Q = 100%) as a stan-
dard. From Equation (2), low values of 1.2% for
(GdL1)3Dy and 0.8% for (GdL2)3Dy were obtained
(2)
In this equation, s and x refer to the standard and the
unknown sample, respectively, I represents the corrected to-
tal integrated emission intensity, A is the absorbance at the
excitation wavelength, and η is the refractive index of the
solution (ηwater = 1.33 and ηethanol = 1.36). The quantum
yields are summarized in Table 2. Owing to the lower π–π*
energy level of GdL2, the quantum yield of the complex
drops from 9.8% for (GdL1)3Eu to 1.5% for (GdL2)3Eu;
therefore, L2 is a less efficient sensitizer of europium emis-
sion. The direct excitation of the lanthanide ions is possible
but is very inefficient because the f–f transitions are Laporte
LN
forbidden. However, the intrinsic quantum yields Q
Ln
(LnIII = EuIII, TbIII) can be estimated from Equations (3)
and (4) for the ratio between the observed (τobs) and radia-
tive (τrad) lifetimes.
Table 2. Calculated quantum yields of (GdL1)3Ln (λexc = 293 nm,
298 K) and (GdL2)3Ln (λexc = 315 nm, 298 K).
Ln
H2O
D2O
Eu
Dy
Tb
9.8%
1.2%
30.9%
1.5%
0.8%
15.3%
The Einstein coefficient AMD,0 equals 14.65 s–1, n is the
refractive index set to nH O = 1.34, which is equal to that
of the neat solvent, and (It2ot/IMD) represents the ratio of the
Figure 2. Corrected and normalized luminescence spectra of
(GdL1)3Dy (λexc = 293 nm, 298 K) and (GdL2)3Dy (λexc = 315 nm,
total integrated intensity of the transitions to the transition 298 K).
Eur. J. Inorg. Chem. 2015, 4207–4216
4211
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
(Table 2). The poor efficiency of the ligands to sensitize dys- 2F5/2 Ǟ2F7/2 transition could be observed (Figure 4),
prosium(III) ions could be observed in the luminescence whereas emission was absent for (GdL1)3Yb. Luminescence
spectra. The amounts of water molecules were calculated in the IR region was not seen for aqueous solutions of
by using the phenomenological Equation (6).[19,40,41] The either (GdL1)3Yb or (GdL2)3Yb, most likely because of the
values of 1.2 and 0.3 obtained for (GdL1)3Dy and quenching of the radiative emission by water.
(GdL2)3Dy result in ratios of the tris complex to bis com-
plex of 85:15 and 90:10, respectively.
qDy(H O) = 21.1ϫΔkobs – 0.60
(6)
2
The characteristic green emission was observed for the
terbium(III) complexes GdL1 and GdL2 with the ligand ex-
cited at λ = 295 and 315 nm respectively, owing to the TbIII
5D4Ǟ7FJ (J = 6–0) transitions (Figure 3). The quantum
yields for the terbium(III) complexes are relatively high and
were measured to be 30% for (GdL1)3Tb and 15% for
(GdL2)3Tb by the comparative rhodamine 101 method. The
amount of water molecules bound to the luminescent centre
can also be determined from lifetime measurements in H2O
and D2O by using Equation (7),[37,42] and the results are
shown in (Table 2).
qTb(H O) = 5ϫ(Δkobs – 0.06)
(7)
2
Figure 4. Corrected and normalized solid-state luminescence spec-
tra of (GdL2)3Yb, λexc = 315 nm, 298 K.
Relaxometric Studies
The efficiency of a 1 mm solution of a gadolinium(III)
agent to shorten the longitudinal relaxation time (T1) can
be derived from proton nuclear magnetic relaxation disper-
sion (NMRD) profiles by measuring the water proton re-
laxivity (r1) as a function of the magnetic field strength.
The relaxation rate is enhanced by the dipolar interaction
between the water molecules and the paramagnetic gadolin-
ium(III) centre. In addition to inner-sphere contri-
butions,[11,42] which result from the water molecules directly
bound to the paramagnetic centre and exchanging with the
bulk, outer-sphere[43] interactions of water have to be taken
into account; in some cases, second-sphere interactions[44,45]
can also have significant effects. Several parameters are de-
fined for the inner-sphere water molecules. Although water
molecules directly bound to the luminescent centre have a
negative effect on the luminescence, a high relaxivity can be
Figure 3. Corrected and normalized luminescence spectra of
(GdL1)3Tb (λexc = 293 nm, 298 K) and (GdL2)3Tb (λexc = 315 nm,
298 K).
The overall ratios of the tris to bis complexes for obtained with a higher amount of water molecules directly
(GdL1)3Tb and (GdL2)3Tb are 55:45 and 73:27, respec- bound to the paramagnetic centre (q). Other parameters are
tively. These ratios are lower than those for the other lan- the distance between the gadolinium(III) centre and the
thanides, but very high quantum yields can still be ob- water molecules (r), the water residence time (τM), the rota-
tained, apparently because the energy from the excited state tional correlation time of the paramagnetic centre (τR), the
of terbium(III) is poorly quenched by water.
electronic relaxation time of gadolinium(III) at zero field
The tris complexes of all other luminescent lanthanide (τS0) and the correlation time that modulates the electronic
ions such as neodymium(III), holmium(III), samarium(III) relaxation (τv).
1
or thulium(III) unfortunately only give rise to ligand emis-
The H NMRD profiles of GdL1 and (GdL1)3Eu have
sion, and no distinct peaks for lanthanide emission could been reported in our recent publication.[30] The profile of
be observed. The low energy of the π–π* state apparently (GdL1)3Eu displays a characteristic hump between 20 and
makes both L1 and L2 poor sensitizers for these lantha- 100 MHz, which is assigned to the formation of a supra-
nides. However, for the ytterbium(III) complex (GdL2)3Yb, molecular structure. At 20 MHz and 310 K, the relaxivity
solid-state emission at λ = 980 nm corresponding to the is enhanced to 8.3 s–1 mm–1 for GdL1 and to 9.6 s–1 mm–1
Eur. J. Inorg. Chem. 2015, 4207–4216
4212
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Table 3. Parameters obtained by fitting the 1H NMRD data in water at pH 7.4 and 310 K. Fixed values: q = 1, r = 0.31 nm, d = 0.36 nm
and D = 3.0ϫ10–9 m2 s–1.
Gd-DTPA[a]
GdL2[b]
(GdL2)3Eu[b]
(GdL2)3Dy[b]
(GdL2)3Tb[b]
τR [ps]
τS0 [ps]
τV [ps]
54Ϯ1
87Ϯ3
25Ϯ3
809Ϯ48
83Ϯ1
20Ϯ1
1000Ϯ3
64Ϯ1
21Ϯ1
818Ϯ35
70Ϯ1
18Ϯ1
916Ϯ41
66Ϯ1
17Ϯ1
[a] From ref.[46] [b] τM was fixed to 1500 ns.
for (GdL1)3Eu compared with 3.8 s–1 mm–1 for Gd-DTPA, temperatures of 25–45 °C increased by less than 13%, which
as could be expected given the higher molecular weight of indicates a slow water exchange. Thus, during the fitting
the synthesized chelates. With the presence of three gadolin- procedure, τM was fixed to 1500 ns for all of the metallo-
ium ions per metallostar compound considered, a longitu- stars as well as for the GdL2 complex. At 20 MHz and
dinal relaxation rate of 28.8 s–1 mm–1 per (GdL1)3Eu mo- 310 K, the relaxivity of the complexes (GdL2)3Ln with Ln
lecule is obtained at 20 MHz and 310 K.
= EuIII, DyIII and TbIII were determined to be very similar
In this work, the relaxometric properties of metallostars (8.09, 7.60 and 7.24 s–1 mmol–1 respectively). Assuming the
based on the L2 ligand have been examined (Table 3). Simi- presence of three Gd-DTPA moieties, the expected r1 relax-
larly to that of (GdL1)3Eu, each of the proton NMRD pro- ivities of the molecules would be 24.27, 22.8 and
files of the (GdL2)3Ln metallostars (measured in water at 21.72 s–1 mmol–1 per metallostar complex; these values are
pH 7.4 and 310 K) show a hump between 20 and 100 MHz, slightly lower than the value of 28.8 s–1 mm–1 observed for
characteristic of supramolecular structures in solution (Fig- the (GdL1)3Eu metallostar.[30]
ure 5). The similarity of the NMRD profile of GdL2 to
those of the metallostar complexes suggests a self-aggrega-
tion of the monomer to form dimeric, trimeric or
multimeric species.
Conclusions
In this work, the selective incorporation of several lan-
thanides into two ditopic ligands has been accomplished,
and the resultant metallostar complexes exhibited favour-
able luminescence and relaxometric properties for potential
use as bimodal MRI/OI agents. The incorporation of EuIII
,
DyIII and TbIII ions into the complexes resulted in emission
in the visible region upon excitation into the ligand levels.
Quantum yields of up to 10% for (GdL1)3Eu were achieved
with a sensitization efficiency (ηsens) of 47%. The introduc-
tion of a ligand with an ethynyl group (GdL2) lowered the
energy of the π–π* excited state, which resulted in a de-
crease of the quantum yield of the (GdL2)3Eu complex to
1.5% and a decrease of the sensitization efficiency to only
6.3%. The same effect has been observed for the dysprosi-
um(III) and terbium(III) complexes. At the concentrations
used in the luminescence measurements, the tris complexes
partially converted into bis complexes; however, the pres-
ence of an extra linker between the aromatic rings in GdL2
allowed for a larger tris to bis complex ratio compared with
that for GdL1. The lower energy resulting from this linker
also lowered the π–π* excited state of GdL2 and allowed it
to sensitize ytterbium(III) ions in the solid state. The
NMRD profiles of (GdL2)3Ln complexes displayed charac-
teristic humps between 20 and 100 MHz owing to the for-
mation of supramolecular structures with enhanced longi-
tudinal relaxivities of up to 25 s–1 mm–1 per metallostar as-
sembly at 20 MHz and 310 K.
Figure 5. NMRD profiles of GdL2 and (GdL2)3Ln (Ln = Eu, Dy
and Tb) compared with that of Gd-DTPA in water at 310 K.
The NMRD data shown in Figure 5 were fitted to the
Solomon–Bloembergen–Morgan equation, which indicated
significant increases of τR compared with that of Gd-
DTPA, as expected. However, owing to the possible equilib-
ria between the tris and bis complexes, the precise τR values
are difficult to determine. It should be noted that the equi-
libria between the tris and bis complexes observed in the
luminescence measurements might be absent or present to
a lesser extent owing to the higher concentrations used in
the NMRD measurements. Also, from the stability con-
stants of the Ln-DTPA and Ln(DPA)3 complexes, it is
plausible that the redistribution of GdIII and EuIII ions be-
tween both the ligands may occur. However, this is unlikely
as the emission spectra of (GdL1)3Eu and (GdL2)3Eu do
not show the splitting of the 5D0Ǟ7F1 transition, which
would be expected to occur if the Eu3+ ions were complexed
Experimental Section
Materials, Reagents and Solvents: Materials, reagents and solvents
were obtained from Sigma–Aldrich (Bornem, Belgium), Acros Or-
ganics (Geel, Belgium), ChemLab (Zedelgem, Belgium), ABCR
in the DTPA-part. The relaxivities measured at 20 MHz at (Karlsruhe, Germany), Iris Biotech GmbH (Marktredwitz, Ger-
Eur. J. Inorg. Chem. 2015, 4207–4216
4213
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
many) and BDH Prolabo (Leuven, Belgium) and were used without
further purification. The gadolinium(III) salt was obtained from
Alfa Aesar (Ward Hill, USA), and europium(III) chloride hexa-
hydrate and lanthanum(III) chloride heptahydrate were obtained
from Acros Organics (Geel, Belgium).
General Synthesis of (GdL1)3Ln: All complexes were synthesized by
applying the previously reported procedure[30] and by replacing
EuCl3 with the appropriate LnCl3·xH2O salt. The complexes were
characterized by total reflection X-ray fluorescence (TXRF), IR
and optical spectroscopy. The Gd/Ln ratios obtained by TXRF
spectroscopy were 2.6 (Gd/Tb), 2.8 (Gd/Dy), 3.0 (Gd/Ho), 3.0 (Gd/
Sm), 3.1 (Gd/Nd) and 3.0 (Gd/Yb).
bis(triphenylphosphine)palladium(II) chloride (5 mol-%, 52.7 mg,
0.075 mmol) were added to dry THF (10 mL). Triethylamine
(2 equiv. 0.304 g, 3.00 mmol) was added, and the mixture was
stirred for 15 min. Compound 3 (1.1 equiv., 0.382 g, 1.65 mmol)
was added, and the mixture was stirred for 4 h at 40 °C. The THF
was evaporated, CH2Cl2 was added, and the suspension was
washed with water. The crude product was dissolved in methanol,
and the solution was stirred for 30 min. The precipitate was col-
lected by filtration and dried to afford the desired product (0.535 g,
84%). ESI-MS (MeOH): m/z = 447.7 [M + Na]+. 1H NMR
(300 MHz, CDCl3, 25 °C): δ = 1.47 (s, 9 H, tBu), 4.04 (s, 6 H,
3
OMe), 4.36 (d, JH,H = 6.1 Hz, 2 H, NHCH2C), 4.92 (br., 1 H,
3
NHCH2C), 7.31 (d, JH,H = 8.1 Hz, 2 H, CCH2CH2C), 7.50 (d,
Compound 1: 4-Iodobenzylamine (1 equiv., 0.680 g, 2.92 mmol) and
triethylamine (1.5 equiv., 4.38 mmol, 0.443 g) were dissolved in
tetrahydrofuran (THF; 20 mL). The solution was cooled in an ice
bath, and boc-anhydride (1.5 equiv., 4.38 mmol, 0.955 g) was added
slowly. The mixture was stirred overnight at room temperature, and
the solvent was evaporated. Water was added (20 mL), and the
product was extracted with EtOAc (3ϫ20 mL). The combined or-
ganic layer was washed with brine and dried with MgSO4. The
solvent was evaporated to yield the desired product (0.966 g, 99%).
ESI-MS (MeOH): m/z = 356.6 [M + Na]+. 1H NMR (300 MHz,
3JH,H = 8.1 Hz, 2 H, CCH2CH2C), 8.36 (s, 2 H, CCH2CC) ppm.
13C NMR (75 MHz, CDCl3, 25 °C): δ = 28.39, 44.36, 53.33, 79.83,
85.41, 96.79, 120.24, 127.54, 129.65, 132.36, 132.73, 141.15, 148.44,
155.88, 164.75 ppm.
Compound 5: Compound 4 (1 equiv. 0.212 g, 0.5 mmol) was dis-
solved in a 50:50 dichloromethane/trifluoroacetic acid (DCM/TFA)
mixture (10 mL), and the solution was stirred for 12 h at room
temperature. The solvents were evaporated, DCM/MeOH (6:4,
10 mL) was added, and the solvents were evaporated again. The
crude product was dissolved in DCM, washed with a saturated
aqueous NaHCO3 solution and brine and then dried with MgSO4,
yield 93%. ESI-MS (MeOH): m/z = 347.7 [M + Na]+. 1H NMR
(300 MHz, [D5]pyridine, 25 °C): δ = 3.91 (s, 6 H, OMe), 4.68 (s, 2
3
CDCl3, 25 °C): δ = 1.45 (s, 9 H, tBu), 4.23 (d, JH,H = 5.6 Hz, 2
3
H, NHCH2C), 4.87 (br., 1 H, NHCH2C), 7.01 (d, JH,H = 8.4 Hz,
3
2 H, ICCH2CH2C), 7.65 (d, JH,H = 8.4 Hz, 2 H, ICCH2CH2C)
ppm. 13C NMR (75 MHz, CDCl3, 25 °C): δ = 28.38, 44.10, 79.71,
3
92.59, 129.36, 137.62, 138.73, 155.83 ppm.
H, NHCH2C), 4.92 (br., 1 H, NHCH2C), 7.72 (d, JH,H = 7.9 Hz,
2 H, CCH2CH2C), 7.83 (d, 3JH,H = 7.9 Hz, 2 H, CCH2CH2C), 8.37
(s, 2 H, CCH2CC) ppm. 13C NMR (75 MHz, CDCl3, 25 °C): δ =
53.33, 85.24, 97.02, 119.75, 127.38, 129.63, 132.01, 132.14, 132.32,
134.59, 148.40, 164.75 ppm.
Compound 2: The synthesis involved a similar procedure to that
previously reported by Bünzli and co-workers[31] with (4-ami-
nobenzyl)acetylene and dimethyl 4-bromopyridine-2,6-carboxylate.
Compound 1 (1 equiv., 0.966 g, 2.90 mmol), bis(triphenylphos-
phine)palladium(II) chloride (5 mol-%, 0.101 mg, 0.145 mmol),
copper(II) iodide (10 mol-%, 0.055 g, 0.290 mmol), and triethyl-
amine (2 equiv., 0.587 g, 5.80 mmol) were suspended in dry THF
(10 mL), and the mixture was stirred for 15 min under an inert
atmosphere. Afterwards, (trimethylsilyl)acetylene (1.2 equiv.,
0.118 g, 1.2 mmol) was added, and the mixture was stirred over-
night. Diethyl ether (50 mL) was added, and the mixture was fil-
tered through Celite. Purification over silica with chloroform
yielded the desired product (0.704 g, 80%). ESI-MS (MeOH): m/z
Compound 6: Compound 5 (1.1 equiv., 0.28 mmol, 0.90 g) was dis-
solved in dry N,N-dimethylformamide (DMF, 10 mL), and N,N-
diisopropylethylamine (1.5 equiv., 0.38 mmol, 66 μL) was added to
the solution. The mixture was stirred for 15 min at room tempera-
ture under an inert atmosphere. At the same time, the DTPA tert-
butyl ester (1 equiv., 0.25 mmol, 0.156 g), O-(benzotriazol-1-yl)-
N,N,NЈ,NЈ-tetramethyluronium
tetrafluoroborate
(TBTU;
1.5 equiv., 0.38 mmol, 0.121 g) and N,N-diisopropylethylamine
(1 equiv.; 0.25 mmol; 44 μL) were dissolved in dry DMF (10 mL)
in a three-neck flask, and the solution was stirred for 15 min at
room temperature under an argon atmosphere. The solution from
flask 1 was added dropwise over a period of 10 min to the three-
neck flask, and the mixture was stirred at room temperature under
an argon atmosphere for 24 h. After the evaporation of the solvent,
the residue was redissolved in DCM. The suspension was washed
with a saturated aqueous NaHCO3 solution and with brine. The
organic phase was dried with magnesium sulfate and evaporated
under reduced pressure. The obtained product was further purified
through silica column chromatography (eluent: CHCl3/5% MeOH/
0.66% NH3), and the collected fractions were evaporated and dried
under vacuum at 50 °C to yield a yellow oil (0.196 mmol, 0.181 g,
70%). ESI-MS (MeOH): m/z = 947.4 [M + Na]+. 1H NMR
(300 MHz, CDCl3): δ = 1.44 (s, 36 H, tBu), 2.64 (t, 4 H,
NCH2CH2N), 2.78 (t, 4 H, NCH2CH2N), 3.24 [s, 2 H, NCH2C(O)-
NH], 3.33 [s, 8 H, NCH2C(O)O], 4.04 (s, 6 H, OMe), 4.51 (d, 3JH,H
1
= 326.5 [M + Na]+. H NMR (300 MHz, CDCl3, 25 °C): δ = 0.24
3
(s, 9 H, SiCH3), 1.45 (s, 9 H, tBu), 4.29 (d, JH,H = 5.6 Hz, 2 H,
3
NHCH2C), 4.85 (br., 1 H, NHCH2C), 7.20 (d, JH,H = 7.6 Hz, 2
3
H, CCH2CH2C), 7.42 (d, JH,H = 7.6 Hz, 2 H, CCH2CH2C) ppm.
13C NMR (75 MHz, CDCl3, 25 °C): δ = 0.37, 28.34, 44.46, 79.68,
94.18, 104.86, 122.11, 127.23, 132.20, 139.43, 155.88 ppm.
Compound 3: Compound 2 (1 equiv., 0.671 g, 2.21 mmol) was dis-
solved in THF (20 mL), and the solution was cooled to 0 °C. TBAF
(1 m, 4.02 mL, 4.42 mmol) was added, and the mixture was stirred
at 0 °C for 1 h. After the reaction, a water/dichloromethane (water/
DCM) mixture (50 mL) was added. The organic layer was collected
and dried, and the solvents were evaporated. The crude product
was purified with silica (eluent DCM) to afford the product
(0.382 g, 75%). ESI-MS (MeOH): m/z = 354.6 [M + Na]+. 1H
NMR (300 MHz, CDCl3, 25 °C): δ = 1.45 (s, 9 H, tBu), 3.1 (s, 1
3
H, CCH), 4.30 (d, JH,H = 5.9 Hz, 2 H, NHCH2C), 4.90 (br., 1 H,
3
NHCH2C), 7.23 (d, JH,H = 8.3 Hz, 2 H, CCH2CH2C), 7.44 (d,
3
= 6.2 Hz, 2 H, NHCH2C), 7.38 (d, JH,H = 8.2 Hz, 2 H,
3JH,H = 8.3 Hz, 2 H, CCH2CH2C) ppm. 13C NMR (75 MHz,
CDCl3, 25 °C): δ = 28.38, 44.37, 79.70, 83.43, 121.02, 127.1, 127.29,
132.34, 139.84, 155.87 ppm.
CCH2CH2C), 7.53 (d, 3JH,H = 8.2 Hz, 2 H, CCH2CH2C), 8.35 (s, 2
H, CCH2CC), 8.88 (br., 1 H, NHCH2C) ppm. 13C NMR (75 MHz,
CDCl3, 25 °C): δ = 28.16, 28.42, 38.61, 42.67, 52.10, 53.32, 53.76,
55.69, 81.09, 85.10, 97.31, 119.65, 128.13, 129.59, 132.13, 134.67,
Compound 4: Dimethyl 4-bromopyridine-2,6-carboxylate (1 equiv.,
0.411 g, 1.50 mmol), CuI (10 mol-%, 28.6 mg, 0.15 mmol) and 142.06, 148.43, 164.77, 170.51 ppm.
Eur. J. Inorg. Chem. 2015, 4207–4216
4214
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Compound 7: The protected ligand precursor
6
(1 equiv.,
flashlamp (μF900H) and an extended red-sensitive photomultiplier
(185–1010 nm, Hamamatsu R 2658P). All spectra were corrected
for the instrumental functions. The luminescence decays were deter-
mined under ligand excitation (293 and 315 nm), and the emission
of the most intense transitions of the luminescent lanthanide ions
was monitored. The luminescence decays were analyzed by using
Edinburgh software, and the lifetimes were averages of at least three
0.19 mmol, 0.175 g) was dissolved in a 6 m HCl solution, and the
mixture was stirred for 1 h at room temperature under an argon
atmosphere. The solvent was evaporated, water was added, and the
solution was evaporated again (2ϫ). The product was redissolved
in water, and the pH was adjusted from ca. 2 to 7 with pyridine.
The solvents were evaporated and dried under vacuum at 50 °C to
1
afford the product (0.176 mmol, 93%). H NMR (300 MHz, [D5]- measurements in water and deuterated water. The quantum yields
pyridine, 25 °C): δ = 2.64 (t, 4 H, NCH2CH2N), 2.78 (t, 4 H, were determined by a comparative method with an estimated ex-
NCH2CH2N), 3.24 [s, 2 H, NCH2C(O)NH], 3.91 (s, 6 H, OMe),
perimental error of 10% by using solutions of quinine sulfate
3
4.15 [s, 8 H, NCH2C(O)O], 4.80 (d, JH,H = 6.4 Hz, 2 H, (Fluka) in 1 n H2SO4 (Q = 54.6%) and rhodamine 101 (Sigma) in
3
NHCH2C), 7.57 (d, JH,H = 8.3 Hz, 2 H, CCH2CH2C), 7.62 (d, ethanol (Q = 100%) as standards. The solutions were diluted to
3JH,H = 8.3 Hz, 2 H, CCH2CH2C), 8.35 (s, 2 H, CCH2CC) ppm.
provide an optical density of less than 0.05 at the excitation wave-
length.
Synthesis of Methyl-Protected GdL2: The methyl-protected ligand
7 (1 equiv., 0.23 mmol, 160 mg) was dissolved in pyridine (5 mL),
and the hydrated GdCl3 salt (1.05 equiv., 0.24 mmol) in water
(0.3 mL) was added to the solution. This mixture was stirred for
3 h at 70 °C. The solvent was removed under reduced pressure, and
ethanol was added. The suspension was heated under reflux for
1 h and then filtered through a P4 glass filter. The absence of free
lanthanide ions was checked by using an arsenazo indicator.
Proton NMRD: The proton NMRD profiles were measured with a
Stelar Spinmaster FFC fast-field cycling NMR relaxometer [Stelar,
Mede (PV), Italy] over a magnetic field strength range extending
from 0.24 mT to 0.7 T. The measurements were performed at 310 K
with samples (0.6 mL) in 10 mm o.d. Pyrex tubes. Additional
relaxation rates at 20, 60, 300 and 500 MHz were obtained with
Minispec mq-20, Minispec mq-60, Bruker Avance-300 and Bruker
Avance 500 instruments (Bruker, Karlsruhe, Germany), respec-
tively.
Synthesis of GdL2: The methyl-protected GdL2 complex (1 equiv.,
0.106 g, 0.12 mmol) was dissolved in water (5 mL), and K2CO3
(2.5 equiv., 0.31 mmol, 43 mg) was added to the solution. The mix-
ture was stirred overnight at room temperature. After the comple-
tion of the reaction, a pH of ca. 9 was measured. The solvent was
evaporated, water was added, and the solution was stirred for
30 min, after which the pH changed to ca. 8. The solvent was re-
moved under reduced pressure, and orange flakes were obtained
(0.11 mmol, 95%).
Acknowledgments
CHN microanalysis was performed by Mr Dirk Henot. The ESI-
MS measurements were done by Mr Dirk Henot, and Mr Karel
Duerinckx is acknowledged for his help with the NMR measure-
ments. Mrs Corinne Piérart is acknowledged for her help in per-
forming the relaxometric measurements.
General Synthesis of (GdL2)3Ln: The deprotected GdL2 (3 equiv.,
51 mg, 0.06 mmol) was dissolved in water (3 mL). The appropriate
LnCl3·xH2O (1.1 equiv., 0.02 mmol) was added, and the reaction
was kept at 70 °C for 3 h. The solvent was removed under reduced
pressure, and ethanol was added. The suspension was heated under
reflux for 1 h and then filtered through a P4 glass filter. The IR
spectrum of the ligand shows a strong absorption in the IR region
[1] P. Hermann, J. Kotek, V. Kubícek, I. Lukes, Dalton Trans.
2008, 3027–3047.
[2] V. C. Pierre, M. J. Allen, P. Caravan, J. Biol. Inorg. Chem. 2014,
at ν = 1600 cm–1, which corresponds to the C=O bond. After com-
˜
19, 127–31.
plexation, a new peak appeared at ν ≈ 1500 cm–1. The complexes
[3] É. Tóth, L. Helm, A. Merbach, in: Contrast Agents I (Ed.: W.
Krause), Springer, Berlin, 2002, p. 61–101.
[4] P. Caravan, Acc. Chem. Res. 2009, 42, 851–62.
[5] E. Debroye, T. N. Parac-Vogt, Chem. Soc. Rev. 2014, 43, 8178–
8192.
[6] W. P. Cacheris, S. C. Quay, S. M. Rocklage, Magn. Reson. Im-
aging 1990, 8, 467–481.
[7] F. G. Shellock, E. Kanal, J. Magn. Reson. Imaging 1999, 10,
477–484.
˜
have been characterized by TXRF, IR and optical spectroscopy.
The Gd/Ln ratios obtained by TXRF spectroscopy were 3.0 (Gd/
Tb), 3.0 (Gd/Dy), 3.0 (Gd/Eu), 2.9 (Gd/Sm), 2.7 (Gd/Nd) and 3.0
(Gd/Yb).
IR and NMR Spectroscopy: The FTIR spectra were recorded with
a Bruker Vertex 70 FTIR spectrometer (Bruker, Ettlingen, Ger-
many). The 1H and 13C NMR spectra were recorded by using a
Bruker Avance 300 spectrometer (Bruker, Karlsruhe, Germany) op-
erating at 300 MHz for 1H and 75 MHz for 13C or a Bruker Avance
[8] A. Accardo, D. Tesauro, L. Aloj, C. Pedone, G. Morelli, Coord.
Chem. Rev. 2009, 253, 2193–2213.
[9] P. Caravan, J. J. Ellison, T. J. McMurry, R. B. Lauffer, Chem.
Rev. 1999, 99, 2293–2352.
1
400 spectrometer operating at 400 MHz for H and 100 MHz for
13C.
[10] A. J. Villaraza, A. Bumb, M. W. Brechbiel, Chem. Rev. 2010,
110, 2921–2959.
TXRF Spectroscopy: TXRF measurements were performed with a
Bruker S2 Picofox instrument by analyzing approximately 100 ppm
gadolinium solutions with respect to a Chem-Lab gallium standard
solution (500 μg/mL, 2–5% HNO3).
[11] I. Solomon, Phys. Rev. 1955, 99, 559–565.
[12] P. Lebdusˇková, J. Kotek, P. Hermann, L. Vander Elst, R. N.
Muller, I. Lukesˇ, J. A. Peters, Bioconjugate Chem. 2004, 15,
881–889.
[13] C.-H. Huang, K. Nwe, A. Al Zaki, M. W. Brechbiel, A.
Tsourkas, ACS Nano 2012, 6, 9416–24.
[14] Y. Li, M. Beija, S. Laurent, L. vander Elst, R. N. Muller,
H. T. T. Duong, A. B. Lowe, T. P. Davis, C. Boyer, Macromole-
cules 2012, 45, 4196–4204.
[15] G. Dehaen, S. V. Eliseeva, K. Kimpe, S. Laurent, L.
Vander Elst, R. N. Muller, W. Dehaen, K. Binnemans, T. N.
Parac-Vogt, Chem. Eur. J. 2012, 18, 293–302.
Optical Spectroscopy: The UV/Vis absorption spectra of freshly
prepared aqueous solutions in quartz Suprasil® cells (115F-QS)
with an optical pathlength of 0.2 cm were recorded with a Varian
Cary 5000 spectrophotometer. The excitation data, emission data
and luminescence decays were recorded with an Edinburgh Instru-
ments FS920 steady-state spectrofluorimeter. This instrument was
equipped with a 450 W xenon arc lamp, a high-energy microsecond
Eur. J. Inorg. Chem. 2015, 4207–4216
4215
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
[16] M. Zhen, J. Zheng, L. Ye, S. Li, C. Jin, K. Li, D. Qiu, H. Han,
C. Shu, Y. Yang, C. Wang, ACS Appl. Mater. Interfaces 2012,
4, 3724–3729.
[17] T. N. Parac-Vogt, K. Kimpe, S. Laurent, C. Piérart, L. V. Elst,
R. N. Muller, K. Binnemans, Eur. J. Inorg. Chem. 2004, 3538–
3543.
[30] E. Debroye, M. Ceulemans, L. Vander Elst, S. Laurent, R. N.
Muller, T. N. Parac-Vogt, Inorg. Chem. 2014, 53, 1257–1259.
[31] C. Platas-Iglesias, C. Piguet, N. André, J.-C. G. Bünzli, J.
Chem. Soc., Dalton Trans. 2001, 3084–3091.
[32] T. Ohe, N. Miyaura, A. Suzuki, J. Org. Chem. 1993, 58, 2201–
2208.
[18] F. Kielar, L. Tei, E. Terreno, M. Botta, J. Am. Chem. Soc. 2010,
[33] A. Aebischer, F. Gumy, J.-C. G. Bünzli, Phys. Chem. Chem.
Phys. 2009, 11, 1346–1353.
132, 7836–7837.
[19] E. Debroye, S. Laurent, L. Vander Elst, R. N. Muller, T. N. Pa- [34] H. Onishi, Talanta 1972, 19, 473–478.
rac-Vogt, Eur. J. Inorg. Chem. 2013, 16019–28.
[20] S. Langereis, T. Geelen, H. Grüll, G. J. Strijkers, K. Nicolay,
NMR Biomed. 2013, 26, 728–44.
[21] E. Debroye, S. V. Eliseeva, S. Laurent, L. Vander Elst, R. N.
Muller, T. N. Parac-Vogt, Dalton Trans. 2014, 43, 3589–600.
[22] S. Aime, P. Caravan, J. Magn. Reson. Im. 2009, 30, 1259–67.
[23] R. T. M. de Rosales, J. Labelled Compd. Radiopharm. 2014, 57,
298–303.
[24] M. P. Placidi, J. Engelmann, L. S. Natrajan, N. K. Logothetis,
G. Angelovski, Chem. Commun. 2011, 47, 11534–11536.
[25] G. Tallec, P. H. Fries, D. Imbert, M. Mazzanti, Inorg. Chem.
2011, 50, 7943–7945.
[26] B. P. Joshi, T. D. Wang, Cancer 2010, 2, 1251–87.
[27] G. Dehaen, P. Verwilst, S. V. Eliseeva, S. Laurent, L.
Vander Elst, R. N. Muller, W. M. De Borggraeve, K. Binne-
mans, T. N. Parac-Vogt, Inorg. Chem. 2011, 50, 10005–10014.
[28] G. Dehaen, S. V. Eliseeva, P. Verwilst, S. Laurent, L.
[35] A. L. Gassner, C. Duhot, J.-C. G. Bünzli, A. S. Chauvin, Inorg.
Chem. 2008, 47, 7802–7812.
[36] J.-C. G. Bünzli, S. V. Eliseeva, Springer Ser. Fluoresc. 2011, 1–
45.
[37] A. Beeby, I. M. Clarkson, R. S. Dickins, S. Faulkner, D. Parker,
L. Royle, A. S. de Sousa, J. A. G. Williams, M. Woods, J.
Chem. Soc. Perkin Trans. 2 1999, 493–504.
[38] R. M. Supkowski, W. D. Horrocks, Inorg. Chim. Acta 2002,
340, 44–48.
[39] E. G. Moore, Dalton Trans. 2012, 41, 5272–5279.
[40] T. Kimura, Y. Kato, J. Alloys Compd. 1998, 275–277, 806–810.
[41] W. D. Horrocks, D. R. Sudnick, J. Am. Chem. Soc. 1979, 101,
334–340.
[42] M. Harris, S. Carron, L. Vander Elst, S. Laurent, R. N. Muller,
T. N. Parac-Vogt, Chem. Commun. 2015, 51, 2984–2986.
[43] N. Bloembergen, J. Chem. Phys. 1957, 27, 572.
[44] J. Freed, J. Chem. Phys. 1978, 68, 4034–4037.
Vander Elst, R. N. Muller, W. De Borggraeve, K. Binnemans, [45] M. Botta, Eur. J. Inorg. Chem. 2000, 399–407.
T. N. Parac-Vogt, Inorg. Chem. 2012, 51, 8775–8783.
[29] E. Debroye, G. Dehaen, S. V. Eliseeva, S. Laurent, L.
Vander Elst, R. N. Muller, K. Binnemans, T. N. Parac-Vogt,
Dalton Trans. 2012, 41, 10549.
[46] S. Laurent, L. Vander Elst, R. N. Muller, Contrast Media Mol.
Imaging 2006, 1, 128–137.
Received: June 12, 2015
Published Online: August 7, 2015
Eur. J. Inorg. Chem. 2015, 4207–4216
4216
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim