Relationship between the ligand structure and the luminescent properties of water-soluble lanthanide complexes containing bis(bipyridine) anionic arms

Article abstract of DOI:10.1002/chem.200600915

A series of six new ligands (L¹-L⁶) suitable for the formation of luminescent lanthanide complexes in water is described. Ligands L¹-L⁴ are constructed from two 6′-carboxy-6- methylene-2,2′-bipyridine chromophoric arms bonded to the amino function of a 2-aminomethylene-6-carboxy-pyridine (L¹), an N,N-diacetate-ethylene diamine (L²), a serine (L3), or an aminomalonic acid (L⁴). For ligands L⁵ and L⁶, the linking amino function is provided by a glutamic acid, and the anionic functions at the 6′-position of the bipyridyl arms are made of the sodium salts of monoethylphosphonic ester (L⁵) and phosphonic acid (L6). The synthesis and characterisation of the ligands are described, together with the study of the formation of lanthanide complexes with europium and terbium. In the case of L3, the europium complex obtained in acidic conditions was crystallised and the X-ray crystal structure is depicted. Photophysical properties of the complexes were studied by means of UV-visible absorption, and steady-state and time-resolved luminescence spectroscopy. Excited-state luminescence lifetimes of the complexes were determined in water and deuterated water to gain insight into the number of water molecules directly coordinated in the first coordination sphere of the complexes. The coordination behaviour of the series of ligands is questioned in the light of the spectroscopic data and discussed in terms of protection of the cation towards water molecules and their impact on the luminescence efficiency.

Full text of DOI:10.1002/chem.200600915

DOI: 10.1002/chem.200600915
Relationship Between the Ligand Structure and the Luminescent Properties
of Water-Soluble Lanthanide Complexes Containing Bis(bipyridine) Anionic
Arms
Loïc J. Charbonni›re,*[a] Nicolas Weibel,[a] Pascal Retailleau,[b] and Raymond Ziessel*[a]
Abstract: A series of six new ligands
(L1–L6) suitable for the formation of
luminescent lanthanide complexes in
water is described. Ligands L1–L4 are
constructed from two 6’-carboxy-6-
methylene-2,2’-bipyridine chromophor-
ic arms bonded to the amino function
of a 2-aminomethylene-6-carboxy-pyri-
dine (L1), an N,N-diacetate-ethylene di-
amine (L2), a serine (L3), or an amino-
malonic acid (L4). For ligands L5 and
L6, the linking amino function is pro-
vided by a glutamic acid, and the
anionic functions at the 6’-position of
the bipyridyl arms are made of the
sodium salts of monoethylphosphonic
ester (L5) and phosphonic acid (L6).
The synthesis and characterisation of
the ligands are described, together with
the study of the formation of lantha-
nide complexes with europium and
terbium. In the case of L3, the europi-
um complex obtained in acidic condi-
tions was crystallised and the X-ray
crystal structure is depicted. Photo-
physical properties of the complexes
were studied by means of UV-visible
absorption, and steady-state and time-
resolved luminescence spectroscopy.
Excited-state luminescence lifetimes of
the complexes were determined in
water and deuterated water to gain in-
sight into the number of water mole-
cules directly coordinated in the first
coordination sphere of the complexes.
The coordination behaviour of the
series of ligands is questioned in the
light of the spectroscopic data and dis-
cussed in terms of protection of the
cation towards water molecules and
their impact on the luminescence effi-
ciency.
Keywords: europium · lanthanides ·
ligand design
terbium
·
luminescence
·
Introduction
nance imaging,[1] as magnetic probes for structural determi-
nation of proteins[2] and biological environments.[3] The de-
velopment of lanthanide-based probes for luminescence ap-
plications has attracted much interest owing to their excep-
tional photophysical properties.[4,5] The strongly forbidden
character of f–f electronic transitions has resulted in very
long luminescence excited-state lifetimes, which in some
cases can reach the millisecond range for europium or terbi-
um complexes. However, this property results in a major
photophysical drawback in these compounds. Molar absorp-
tion coefficients of the f–f transitions of lanthanides rarely
exceed unity and their direct excitation appears difficult ne-
cessitating laser excitation. This inconvenience has neverthe-
less been bypassed by the introduction of the so-called an-
tenna effect,[6] which consists of embedding the lanthanide
cations into an organic framework containing chromophoric
units. The role of the ligand is first to absorb light and to
transfer this excess energy to the lanthanide excited states,
but also to complete the first coordination sphere of the
metal in order to prevent luminescence quenching due to vi-
Lanthanide complexes are becoming ever more important
in the field of analytical biochemistry, in which they are
used in prominent applications in nuclear magnetic reso-
[a] Dr. L. J. Charbonni›re, Dr. N. Weibel, Dr. R. Ziessel
Laboratoire de Chimie MolØculaire
UMR 7509-CNRS Ecole de Chimie, Polym›res et MatØriaux
25 rue Becquerel, 67087 Strasbourg Cedex 02 (France)
Fax : (+33)3-90-24-26-35
E-mail: charbonn@chimie.u-strasbg.fr
ziessel@chimie.u-strasbg.fr
[b] Dr. P. Retailleau
ICSN-CNRS, 1 avenue de la Terrasse
91198 Gif sur Yvette (France)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the authors. Full experimen-
tal description of the intermediate compounds 5, 6, 8, 9, 10, 13, 14,
16, 17, and 20, and their characterisation, full description of the crys-
tal structure determination of [Eu
(L3H2)Cl]Cl·3H2O, and table of the
7
H
5
7
ꢀ
ꢀ
ꢀ
relative intensities of the D0! F2/5D0! F1 transitions of the europi-
brational deactivation through O H, N H and C H oscilla-
um complexes (17 pages).
tors of the solvent molecules.[7,8] The overall luminescence
346
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 346 – 358

FULL PAPER
efficiency of a lanthanide complex (Fov) is the product of
the efficiency of the ligand to sensitise the lanthanide (hsens
nated by the two N,N,O-tridentate bipyridyl carboxylate
arms (dark-grey strands in Figure 1), the central Nglu atom
of the glutamate and its geminal carboxylate function coor-
dinated by Oglu forming the third coordinating arm (light-
grey strands in Figure 1). In coordination mode A, the bipyr-
idyl arms (bipy) occupy equatorial positions in the complex.
The water molecule is located at one pole of the molecule,
opposite to the Oglu atom. In modes B and C, the bipy arms
occupy all the lower sites of the complex and the water mol-
ecule is displaced at the upper part, next to the glutamate
carboxylate strand. The difference between the two modes
relies on the position of the water molecule relative to the
glutamate. In mode B, H2O is located in the plane defined
by the lanthanide, Nglu and Oglu, and the water molecule and
Nglu are placed symmetrically relative to the Ln–Oglu axis.
The plane containing Nglu, Oglu, Ln, and H2O represents a
pseudosymmetry plane in the complex. For mode C, the
water molecule is nearly equidistant from Nglu and Oglu, dis-
rupting the planar symmetry observed in mode B.
)
and the efficiency of the lanthanide to emit photons once in
its excited state (FLn). The exact parameters governing the
sensitisation process are still unclear, despite comprehensive
works on the matter, which have led to general rules on the
optimisation of the ligand-centred energy levels.[9] The cen-
tral question arises in the exact mechanism of sensitisation,
and whether excitation is shuttled through the ligand-cen-
tred triplet excited state or if direct excitation of the lantha-
nide is possible by the singlet excited state of the ligand.[10]
Concerning the shielding of the metal cation by the ligand,
coordination numbers of eight
to ten are commonly observed
for lanthanide complexes, and
nona- or decadentate ligands
are particularly targeted.
In prior work in the field, we
disclosed[11] that the Eu and Tb
complexes formed with the oc-
The synthetic strategies were then fitted to these three
possible coordination modes in view of removing the water
molecule. Mode B was the simplest case, as it consisted of
elongating the glutamate strand by providing a ninth coordi-
nation site. Ligands L1 and L2 were designed, in which the
carboxylate function was replaced by a 2-aminomethylene-
6-carboxy-pyridine or by an amino diacetate function. For
mode C, the additional coordination site is provided by the
introduction of a new coordination entity on the methylene
bridging Nglu and Oglu species. This was achieved with ligands
L3 and L4 in which the new coordination sites are a neutral
hydroxymethylene or a negatively charged carboxylate func-
tion (at neutral pH in water), respectively. The hardest case
to envisage was an A mode of coordination. The replace-
tadentate ligand
L are very
good probes for luminescence
microscopy[12] and time-re-
solved fluoroimmunoassays.[13]
Based on a glutamate skeleton
functionalised on the amino
function by two 6-carboxy-bi-
pyridine arms, this ligand
formed complexes with the generic formula [Ln(L)(H2O)]
G
upon coordination with lanthanide cations in water. The re-
sulting Eu and Tb complexes are highly luminescent with
long-lived luminescence lifetimes (t) despite the presence of
a single water molecule in the first coordination sphere of
the cations. The removal of this water molecule would un-
doubtedly be beneficial to the photophysical parameters
such as Fov, FLn, and t. Up to now, all our efforts to crystal-
lise such complexes failed. Gaining insight into the molecu-
lar structure would allow for the localisation of the water
molecule, for a better understanding of its interactions with
the surroundings and would provide new ideas for removal
of the water molecule. It is surmised that the utilisation of
an additional chelating donor atom is an attractive approach
to increase the coordination number, providing that it is
well placed in the ligand architecture.
Figure 1 summarises the different coordination modes en-
visaged for [Ln(L)(H2O)] assuming that the ligand is coordi-
G
Figure 1. Three different coordination modes envisaged for [Ln(L)-
(H2O)].
ACHTREUNG
Chem. Eur. J. 2007, 13, 346 – 358
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
347

L. J. Charbonni›re, R. Ziessel et al.
ment of the water molecule would require the introduction
of a new strand on the ligand, necessitating total remodel-
ling. Our approach was to increase the steric hindrance pro-
vided by the bipy arms. It has been shown previously that
the replacement of carboxylate functions by phosphonate
functions can increase steric hindrance in lanthanide com-
plexes,[14] thus resulting in a decrease of the coordination
number and of the number of water molecules coordinated
in the first sphere of the cation.[15] Switching the bipyridyl
carboxylate functions by phosphonate ones should lead to
such an increased congestion and potentially to the removal
of the water molecule (L5 and L6).
Results and Discussion
Synthesis of the ligands: Synthesis of ligands L1–L6 relies on
the key intermediate 6-bromo-6’-bromomethyl-2,2’-bipyri-
dine (1; Scheme 1),[16] which allows for the anchoring of the
bipy arms on preorganised platforms through nucleophilic
substitution at the bromomethylene position, while carboal-
koxylation[17] or phosphorylation[18] reactions on the bromo-
pyridine orthogonally permit the introduction of carboxylic
or phosphonic acid functionalities.
Scheme 1 depicts the protocol for the synthesis of L1.
Starting from diethyl ester of dipicolinic acid (2), reduction
with substoichiometric amounts of NaBH4 affords the 6-hy-
droxymethyl-ethyl picolinic ester in 49% yield,[19,20] which
was transformed into the chlorinated compound 3 in 96%
yield by treatment with SOCl2.[21] Amine 4 was obtained by
a Gabriel synthesis by treating 3 with sodium phthalimidate
in DMF at room temperature (92%) followed by treatment
with hydrazine in refluxing ethanol (quantitative).[21] Nucle-
ophilic substitution of 4 with 1 afforded the intermediate 5
in 67% yield, from which a carboalkoxylation reaction[17]
gave the ester precursor 6 in 45% yield. Saponification of 6
with NaOH followed by neutralisation with hydrochloric
acid gave ligand L1H3 in 42% yield as its tetrahydrochloride
salt.
The ligands L2–L4 were obtained by using similar proto-
cols (Schemes 2 and 3). The ligand L2H4 was obtained from
the monoprotected ethylene diamine 7,[22] which gave the
tertiary amine intermediate 8 in 47% yield upon condensa-
tion with 1. Deprotection of the Boc group with TFA (92%
yield) followed by reaction with ethylbromoacetate afforded
the diester derivative 10 (68% yield). The sequence of car-
boalkoxylation, saponification, and neutralisation gave L2H4
as its tetrahydrochloride salt with 59% yield for the three
steps. Intermediate 11 could hardly be isolated in its pure
form as a result of the presence of triphenylphosphine oxide
This work presents the synthesis of ligands L1–L6, together
with the formation of their lanthanide complexes with Eu
and Tb and studies of their photophysical properties in
water. A critical discussion of the relationship between the
coordination ability of the ligands and potential removal of
the water molecule is also presented.
Scheme 1. i) NaBH4 (0.6 equiv), EtOH, reflux (49%); ii) SOCl2, 08C (96%); iii) Na phthalimidate, DMF, RT(92%); iv) H 2NNH2·H2O (3.5 equiv),
EtOH, reflux (quant); v) 1 (2.2 equiv), K2CO3, CH3CN, 808C (67%); vi) [Pd
H2O, 60 8C; viii) 2n HCl (42% for the two steps).
A
348
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 346 – 358

Lanthanide Complexes
FULL PAPER
(Scheme 3). Carboalkoxylation
of 13 and 16 gave the diesters
14 (61%) and 17 (68%), re-
spectively, and saponification of
the ester functions of 14 and 17
followed by acidification led to
L3H3 in 82% yield and L4H4 in
53% yield, respectively, for the
two steps).
Finally, as described during
the synthesis of ligand L,[11] the
N-alkylation of the dimethyl
ester glutamate hydrochloride
salt 18 with slight excess of 1
led to compound 19 with a
67% yield (Scheme 4). A palla-
dium(0)-catalysed step[18] was
used to introduce
a diethyl
phosphonate moiety in each
2,2’-bipyridine arm at the 6’-po-
sitions. The reaction was con-
ducted in hot toluene contain-
Scheme 2. i) Compound 1 (2.2 equiv), K2CO3, CH3CN, 808C (47%); ii) TFA, CH2Cl2 (92%); iii) ethyl bromo-
A
N
v) NaOH, MeOH/H2O, 60 8C; vi) 2n HCl (59% for the three steps).
ing
diethylphosphite
and
Hünigꢁs base to offer 20 in a
31% yield after purification.
formed during the carboalkoxylation reaction. Nevertheless,
the lipophilic phosphine oxide can be easily removed from
the reaction mixture after saponification.
Adapting the previous synthetic methodology, ligands
L3H3 and L4H4 were obtained from the reaction of 1 and the
hydrochloride salt of the methyl ester of serine (12) or the
diethylester of aminomalonic acid (15) respectively, leading
to compounds 13 (83%) and 16 in (56%), respectively
Two distinct hydrolysis routes were considered to obtain
either ligand L5 or L6. First, a saponification with sodium hy-
droxide in refluxing water led to the selective cleavage of
one ethyl ester group of each phosphonate and to the hy-
drolysis of the carboxylic ester groups. Ligand L5Na4 was ul-
timately isolated with a good yield (79%) after precipitation
in a water/THF mixture. Secondly, the hydrolysis of both
phosphonic diester functions was possible due to trimethyl-
Scheme 3. i) K2CO3, CH3CN, 808C (83% for 12, 56% for 15); ii) [Pd
A
ACHTREUNG
Chem. Eur. J. 2007, 13, 346 – 358
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
349

L. J. Charbonni›re, R. Ziessel et al.
the UV-visible absorption spec-
tra of a 5.6”10ꢀ5 m solution of
L3 (0.01m Tris/HCl buffer,
pH 7.0) upon addition of
EuCl3·6H2O. In water at neu-
tral pH, the free ligand dis-
played two strong absorption
bands at 238 and 288 nm, typi-
cal of the p!p* transitions
centred on the bipyridyl units
in the trans conformation
[18b,24]
(Table 1).
Upon excitation
Scheme 4. i) 1 (2.2 equiv), CH3Cn, K2CO3, 808C, (67%);[11] ii) [Pd
(iPr)2Et, 1008C, (31%); iii) 0.05m aq NaOH, 1008C (79%); iv) TMSBr (29 equiv), CH2Cl2/CHCl3, reflux;
NaOH, H2O, 80 8C (71% for the two steps).
A
G
in this absorption band, the
emission spectra of the free
A
A
silyl bromide.[18a,23] After evaporation of the solvent, a final
saponification was performed to ensure that all the methyl
esters were hydrolysed. Ligand L6Na6 was precipitated from
a methanol/water mixture by adding THF, and obtained as
its dihydrate with a 71% yield.
Table 1. UV/Vis absorption and emission data of the free ligands in
water (0.01m Tris/HCl buffer, pH 7.0).
Absorption
lmax [nm] (e
Emission
[cmꢀ1
[mꢀ1 cmꢀ1])
E
G
]
L1
240 (18700), 287 (22300)
236 (14750), 288 (18400)
238 (20700), 288 (24800)
239 (20300), 288 (24800)
237 (11400), 289 (14700)
239 (20200), 288 (22300)
25800
24650
24750
25800
24600
–
L2
L3
Photophysical studies of the ligands and complexes: For the
purpose of clarity, the photophysical studies of the ligands
and of their lanthanide complexes will be presented by
order of increasing complexity rather than by numerical
order.
L4
L6
L[11]
ligand in neutral buffered water displayed an intense emis-
sion band with a maximum at 404 nm (24750 cmꢀ1), which
disappeared as soon as a delay time was enforced, pointing
Ligand L5H4: The case of L5 is clearly the simplest one as it
was observed that it did not form stable complexes with the
lanthanide cations in water solutions (0.01m Tris/HCl,
pH 7.0). This result is in line with prior results on the com-
plexation of monoalkyl phosphonic esters functionalised at
the 6-position of 2,2’-bipyridines,[18b] which showed such tri-
dentate units to be very weakly coordinating, in contrast to
the corresponding phosphonic acid. For this reason, the pho-
tophysical properties of L5 were not further investigated.
1
to a pp* state. Upon addition of up to one equivalent of
the EuIII ion (Figure 2), these transitions were bathochromi-
cally shifted to 251 and 308 nm, respectively, as a result of
the trans to cis isomerisation of the bipyridyl fragments con-
secutive of the coordination of the cation. As expected, UV-
visible spectrophotometric titration using TbCl3·6H2O dis-
played the same tendency, pointing to a 1:1 metal/ligand sto-
ichiometry for the formed complexes.
Ligand L3H3: The behaviour of ligand L3H3 was easily un-
derstood as it is perfectly in line with previous results ob-
tained with ligand LH4. Figure 2 displays the evolution of
The corresponding europium complex was obtained from
a water/MeOH solvent mixture starting from equimolar
amounts of the ligand and europium chloride salts. Interest-
ingly, as long as the solution remained acidic, as a result of
the introduction of L3 in its protonated hydrochloric salt
form, all species remain in solution, but the increase of the
pH with diluted NaOH resulted in the formation of a pre-
cipitate. It is surmised that the different protonated forms of
the complex are related to different coordination modes, as
verified by the X-ray crystal structure of the complex ob-
tained at low pH (see below). The elemental analysis and
mass spectrum of the complex isolated at neutral pH point-
ed to a [Eu(L3)] formula. In dilute solution, the excitation of
the complex into the bipyridyl-centred absorption bands led
to the characteristic emission spectrum of europium, dis-
7
playing the typical 5D0! FJ (J=0 to 4) electronic transi-
tions. Calculation of the relative intensities of the 5D0!
7
7F2/5D0! F1 transitions indicated a low symmetry in the
Figure 2. UV/Vis absorption spectra of a 5.6”10ꢀ5 m solution of L3 in
water (0.01m Tris/HCl buffer, pH 7.0) titrated by EuCl ·6H2O.
complex (as well as for all other Eu complexes described
3
350
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 346 – 358

Lanthanide Complexes
FULL PAPER
herein, see Table T8, Supporting Information) and to the ab-
sence of an inversion centre in the coordination sphere. A
weak fluorescence signal was observed around 400 nm, at-
bathochromic shift of the p!p* transitions in the absorp-
tion spectra, found at 252 and 307 nm for Eu, and 252 and
306 nm for Tb. Upon excitation in the main absorption band
(307 nm), both complexes exhibited emission spectra charac-
teristic of their respective lanthanide cation, as a result of
the ligand–metal energy transfer.[6] The metal-centred lumi-
1
tributed to residual pp* emission and suggesting a partial
energy transfer from the ligand to the metal. Photophysical
data of the europium and terbium complexes (obtained
from an equimolar solution of the ligand and Tb in the
same solvent) are collected in Table 2.
7
nescence was evidenced by the characteristic 5DI! FJ transi-
tions around 575, 595, 615, 655 and 690 nm for Eu (I=0, J=
0 to 4) and around 495, 545,
585 and 620 nm for Tb (I=4,
Table 2. Photophysical data of the complexes in water (0.01m Tris/HCl buffer, pH 7.0).
lmax [nm] (e tH O (tD O)[ms]
[mꢀ1 cmꢀ1]) FH O (FD O)[%] q[a]
J=6 to 3). Interestingly, no re-
sidual fluorescence could be ob-
served, indicating a very effi-
cient ligand–metal energy trans-
fer. Recording the excitation
spectra with emission moni-
tored on the most intense emis-
sion band of the lanthanide cat-
ACHTREUNG
2
2
2
2
L1
L2
L3
L4
L6
Eu 273 (9500), 297 (9200), 321 (5100) 0.32 (2.18)
Tb 272 (8500), 297 (9000), 318 (5200) 0.83 (1.78)
2.4
1.4
9.5 (26)
19 (32)
4.4 (15)
32 (51)
U
2.8 (2.9)
2.7 (2.9)
0.1 (0.2)
0.1 (0.1)
1.2 (1.4)
1.3 (1.2)
1.2 (1.3)
0.8 (0.7)
1.2 (1.4)
1.9 (2.0)
1.1
Eu 247 (15850), 306 (18600)
Tb 244 (13660), 305 (16000)
Eu 251 (19000), 308 (24390)
Tb 251 (19200), 307 (25500)
Eu 252 (20700), 307 (23400)
Tb 252 (20500), 306 (21000)
Eu 250 (11600), 306 (14900)
Tb 251 (12100), 305 (15400)
1.12 (2.02)
1.81 (1.92)
0.57 (2.86)
1.24 (1.98)
0.57 (2.65)
1.36 (1.87)
4.4
12 (35)
(17.2)
(5.5)
7
ions (5D0! F2 at 617 nm for Eu
0.58 (93%); 1.52 (7%), (2.89) 3.3
5
7
and D4! F5 at 546 nm for Tb)
afforded a perfect match with
the corresponding absorption
spectra, thereby unambiguously
confirming the antenna effect.
Luminescence lifetime mea-
1.03 (1.97)
0.62 (2.48)
1.48 (2.53)
3.5
L[11] Eu 253 (14400), 308 (19700)
8 (35)
31 (53)
Tb 253 (15100), 308 (20800)
1.1
[a] Values in brackets correspond to those obtained according to Parker and co-workers,[7b] see the Experimen-
tal Section for more details.
A
Luminescence decay lifetimes of the complexes in water
and deuterated water were similar to complexes of L, al-
though systematically smaller. From these data, it was possi-
ble to calculate the hydration number (q) corresponding to
the number of water molecules directly coordinated into the
first coordination sphere of the metal. Similarly to L, a hy-
dration number of one was obtained for both complexes. Fi-
nally, the notable difference observed between the com-
plexes of L and L3 concerns the luminescence quantum
yield of the europium complex, which appeared to be about
twice as strong with L than with L3 (Table 2). This behaviour
(Table 2) allowed for the calculation of the hydration
number, q=1.2[7c] or 1.3,[7b] in line with the coordination of
a single carboxylate function of the malonic moiety. Com-
parison of the luminescence lifetimes for L4 and L (Table 2)
pointed to a more pronounced deactivating effect due to
solvation for L4. This is translated into a slightly higher hy-
dration number for L4 compared to L (q=1.1) that may be
attributed to second-sphere interactions possibly arising
from the replacement of the alkyl chain in L with a more
hydrophilic carboxylate function in L4. The higher activity
ꢀ
of deactivating O H oscillators is also reflected into smaller
can be explained by the proximity of the O H oscillator
luminescence quantum yields for L4. Finally, the introduc-
tion of a geminal carboxylate function on the amine did not
displace the water molecule from the first coordination
sphere, which suggests that coordination mode C is not ef-
fective in this case.
ꢀ
provided by the serine alcohol function that brings addition-
al deactivation pathways particularly sensitive in the case of
europium. From these data, it can be estimated that the co-
ordination geometry of L3 must be rather similar to that of
L and the introduction of a geminal hydroxymethyl function
did not severely perturb the geometry of the coordination
pocket at neutral pH.
Ligand L1H3: The UV-visible absorption spectrum of L1H3
(Figure 3, Table 1) in buffered water at pH 7.0 (0.01m Tris/
HCl) is slightly different from those observed for L3 and L4.
An intense absorption band is still present with maximum at
287 nm (Table 1), but it is far broader, particularly in the
270–280 nm region, compared to ligands lacking the pyridine
carboxylate moiety (Figure 2). This enlarged absorption is
ascribed to the combination of the p!p* absorption band
of the bipyridyl moieties at 288 nm, together with the
weaker p!p* transition centred on the pyridyl carboxylate
subunit, with the maximum peaking at 268 nm as observed
in compounds containing the pyridine–carboxylate chromo-
phore.[26] Upon excitation in the UV domain, the emission
spectrum displayed a broad emission band with maximum at
Ligand L4H4: The absorption spectrum of L4H4 in buffered
water at pH 7.0 displayed the typical features of the bipyri-
dinecarboxylate framework, as previously observed for L3
and L, with the two intense p!p* transitions, at 239 and
288 nm (Table 1). Upon excitation into these transitions the
emission spectra displayed a broad emission band with max-
imum at 387 nm (25800 cmꢀ1). The use of a 50 ms delay be-
tween excitation and integration of the emitted signal result-
ed in the loss of this band, which was attributed to the sin-
glet pp* state centred on the bipyridyl unit. The coordina-
tion of europium or terbium cations led to the characteristic
Chem. Eur. J. 2007, 13, 346 – 358
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
351

L. J. Charbonni›re, R. Ziessel et al.
The calculations of the hydration numbers[7] appeared par-
ticularly informative. For both Eu and Tb, a hydration
number close to three was obtained, in complete contradic-
tion with the increased denticity afforded by the ligand. On
the basis of an N-podand-type structure, a nonacoordination
was expected for L1, which was supposed to remove all the
water molecules from the first coordination sphere. Howev-
er, addition of the pyridyl carboxylate functions resulted in
a total change of the coordination mode, as preliminarily
suggested by the absorption data. Prior experiments made
on a ligand composed of an N-butyl amine functionalised by
two bipyridyl carboxylate arms showed that the emergent
lanthanide complexes contained two water molecules in the
first coordination sphere.[27] The presence of three water
molecules can hardly be explained, but could arise from the
decoordination of one bipyridyl arm concomitantly to the
coordination of the pyridyl one. The coordination of the
Figure 3. UV/Vis absorption spectra of L1 (normal line) and its europium
complex (bold line) in water (0.01m Tris/HCl buffer, pH 7.0).
1
387 nm (25800 cmꢀ1), which is attributed to the pp* state
of the bipyridyl units, in agreement with what was previous-
ly observed for L3, L4, and L. No additional emission arising
from the pyridine ring could be observed.
ꢀ
central nitrogen atom, the N O bidentate fragment of the
The absorption spectra of the Eu and Tb complexes of
ligand L1 (Figure 3) are very different from those previously
observed (see, for example, Figure 2). They both display
three main bands at 321, 297 and 273 nm for Eu and 318,
297 and 272 nm for Tb. Upon coordination of the lanthanide
cations, the p!p* transitions are not as significantly dis-
placed toward lower energy as previously observed (down
to 308 nm for example for L and L3, Figure 2). The spectra
better reflect the superimposition of two distinct bands.
While the former band, displaying a shoulder at low energy
(320 nm), could be attributed to a coordinated bipyridine–
carboxylate arm, the latter band (at 297 nm) reflected an in-
termediate situation, in which the bipyridine is neither di-
rectly coordinated (lmax =308 nm) nor free (lmax =288 nm).
The bathochromic shift observed in comparison to the free
ligand points to a conformational change in which the trans
conformation, observed for the free ligand and due to the
nitrogen lone-pair repulsion,[18b,24] is replaced by a conforma-
tion having a pronounced cis character. Similar behaviour
has already been observed in bipyridine containing lantha-
nide complexes[25] and was understood to be second-sphere
interactions mediated by bridging water molecules. A large
hypochromic effect is concomitantly observed upon com-
plexation, in agreement with a spreading of the absorption
bands. Upon excitation in the ligand absorption bands, the
complexes displayed emission spectra typical of their corre-
sponding lanthanide cations, with no residual fluorescence
from the ligand, pointing to an efficient ligand–metal
energy-transfer process. The corresponding excitation spec-
tra are in excellent agreement with the absorption spectra,
which indicates that even if one bipyridyl arm is not coordi-
nated, it efficiently transfers its energy to the cation. Fur-
thermore, this almost perfect superimposition also points to
the efficient sensitisation from the pyridyl carboxylate subu-
nit of the ligand, as previously suggested.[26] Surprisingly, the
measured luminescence quantum yields were particularly
weak (Table 2). In parallel, the excited-state lifetimes mea-
pyridyl carboxylate function and of the N-N-O tridentate
fragment of a bipyridyl carboxylate would lead to a hexa-
denticity that should explain the presence of three water
molecules coordinated to the cation. This should also ex-
plain the observed results of absorption experiments that
pointed to one coordinated bipyridyl, the second one being
relegated to second-sphere interactions. Different reasons
for the decomplexation of one bipyridyl strand can be postu-
lated. The first is that the wrapping of the three heterocyclic
moieties around the cation is disfavored by steric congestion
at the central nitrogen atom. Prior results on the complexa-
tion of lanthanide cations with podand-type ligands such as
a,a’,a’’-nitrilo-(6-methyl-2-pyridinecarboxylic acid) (tpaaH3)
showed that the wrapping is perfectly feasible with short co-
ordination arms such as pyridyl carboxylate.[28] More proba-
bly, the introduction of one supplementary pyridine group
on two coordinating arms led to steric hindrance opposite to
the nitrogen node, resulting in the preferred decoordination
of one bipyridyl arm in favour of the pyridine tether.
Ligand L2H4: The UV-visible absorption spectrum of ligand
L2H4 in neutral buffered water (Figure 4) displayed absorp-
tion bands at 288 and 236 nm (e=18400 and
14750mꢀ1 cmꢀ1, respectively), attributed to the p!p* transi-
tions centred on the bipyridyl units. Excitation into this tran-
sition (288 nm) led to a large emission band with a maxi-
1
mum at 406 nm (24650 cmꢀ1), attributed to the pp* emis-
sion, on the basis of its short-lived character.
In contrast to what was previously observed for L3
(Figure 2) or L, the UV-visible titration of L2 by TbCl3 in
water clearly showed the absence of isosbestic points, point-
ing to the presence of at least two new absorbing species
formed during the titration. Careful examination of the evo-
lution of the absorbances at fixed wavelengths suggested the
successive formation of species with [M(L2)2] and [M(L2)]
formulations.
Eu and Tb complexes were isolated from methanol/water
mixtures, each containing equimolar amounts of the ligands
in their acidic forms and the corresponding lanthanide chlo-
ACHTREUNG
352
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 346 – 358

Lanthanide Complexes
FULL PAPER
L3 and L, the intensities ratio at 308 and 320 nm was very
close to unity (1.1), whereas in the case of L2, this ratio ap-
proached a value of 2.0. The shape of the band suggests a
cis configuration of the bipyridyl units, but a weakened coor-
dination of these arms. Upon excitation in the p!p* ab-
sorption bands, the typical emission of Eu is observed, with
no remaining emission from the ligand, and the excitation
spectrum obtained upon monitoring the europium emission
is almost perfectly superimposable to the absorption spec-
trum, suggesting an efficient energy-transfer process.
The photophysical data of the Eu and Tb complexes are
gathered in Table 2. They showed the two complexes to be
highly luminescent in water, the europium complex of L2
being even more luminescent than the complex of L (9.5
and 8% overall quantum yield, respectively). In all cases,
the metal-centred luminescence decays were perfectly mon-
oexponential, indicative of the presence of single-emitting
species in solution. A striking point concerns the calculation
of the hydration number. For both complexes, luminescence
lifetime values obtained in H2O and D2O showed that no
water molecules were bonded to the cations. The perfect fill-
ing of the first coordination sphere is clearly in favour of a
coordination mode of type B (Figure 1), in which the length-
ening of the upper coordinating arm allows for displacement
of the water molecule. Nevertheless, the energy and shape
of the absorption band of the complex suggest a weakened
coordination of the bipyridyl arms, which may result from
the introduction of the fourth carboxylate function. This ad-
ditional coordinating arm may induce a large change in the
coordination mode of the ligand, possibly acting as a deca-
dentate ligand, strengthened by electrostatic interactions
with the triethylammonium cation[26a] or by a so-called “clip-
ping effect”.[29]
Figure 4. Evolution of the UV/Vis absorption spectra of a 1.05”10ꢀ4
m
solution of L2H4 in water (0.01m Tris/HCl buffer, pH 7.0) upon addition
of 0.11, 0.22, 0.33, 0.44, 0.55, 0.66, 0.77, 0.88, 1.0, and 1.22 equiv of
TbCl3·6H2O.
ride salts. After heating to 608C for a few hours, allowing
for formation of the thermodynamically stable species, the
solutions were cooled and the pH raised to 8–9 with Et3N to
neutralise the excess of protons and to generate a strong
complexation with the carboxylate functions. Concentration
of the solutions followed by addition of Et2O resulted in the
precipitation of white powders, displaying the typical lumi-
nescence of the corresponding lanthanide ions upon UV ir-
radiation. Elemental analysis of the powders pointed to a
general formulation with a 1:1 metal/ligand ratio, together
with the presence of
a
triethylammonium cation,
(Et3NH)[Ln(L2)]. Finally, mass spectrometry of the europi-
um complex by using FAB in the negative detection mode
showed the main peak as being due to a [Eu(L2)]ꢀ species.
Figure 5 displays the absorption, excitation, and emission
spectra of the europium complex. As expected on the basis
of the titration, the p!p* absorption bands centred on the
bipyridyl units were bathochromically shifted upon complex-
ation, but in contrast to what was observed for L3 (Figure 2)
or L, the absorption band is less shifted (306 nm compared
to 308 nm for L3 and L) and the band appears thinner. For
Ligand L6H6: The UV-visible absorption spectrum of ligand
L6 in water (0.01m Tris/HCl buffer, pH 7.0) is presented in
Figure 6, and displayed the characteristic p!p* transitions
centred on the bipyridyl units at 237 and 289 nm (Table 1).
In addition, a pronounced shoulder can be observed around
315 nm. The pK value for the second deprotonation of bi-
pyridyl-functionalised phosphonic acid was determined to
be 7.1,[18b] and at pH 7.0 the mono- and fully deprotonated
forms of phosphonic acid coexist. In the monodeprotonated
form, the p!p* transitions have been shown to be batho-
Figure 5. UV/Vis absorption (thin line), excitation (bold line), and emis-
sion (dotted line) spectra of (Et3NH)[Eu(L2)] in water (0.01m Tris/HCl
buffer, pH 7.0).
Figure 6. UV/Vis absorption spectra of L6H6 (normal line) and its Eu
complex (bold line) in water (0.01m Tris/HCl buffer, pH 7.0).
Chem. Eur. J. 2007, 13, 346 – 358
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
353

L. J. Charbonni›re, R. Ziessel et al.
chromically shifted,[18b] and are probably the cause of the
320 nm shoulder. Upon excitation in the maximum of the
absorption band, a large emission band can be observed
with maximum at 406 nm (24600 cmꢀ1), attributed to the
1pp* on the basis of its short-lived character.
The absorption spectra of the Eu (Figure 6) and Tb com-
plexes (Table 2) showed the typical bathochromic shift of
the p!p* absorption bands observed upon complexation of
the lanthanides. Excitation in this absorption band (304 nm)
revealed emission features characteristic of the Eu and Tb
cations and evidenced an effective ligand-to-metal sensitisa-
tion process. Surprisingly, the overall quantum yield meas-
ured for metal-centred luminescence upon ligand excitation
appeared very low for both complexes, when compared to
the reference complexes of L. Finally, the measurements of
luminescence decay lifetimes revealed differences between
europium and terbium. For Eu, a biexponential decay was
observed in water with a main component (93%) for the
shortest-lived species, with a luminescence decay lifetime of
0.58 ms, similar to those measured for L, L3 and L4. A
longer component (t=1.52 ms) was observed amounting to
7% of the total luminescence. In deuterated water, the lumi-
nescence decay was perfectly monoexponential and by using
this value it was possible to calculate that the short-lived
species has an average hydration number of 1.5, whereas the
long-lived one contains no coordinated water molecules.
Among possible explanations for the presence of this latter
species, one arose from the possibility of intermolecular in-
teractions in which phosphonate functions bridge two lan-
thanide cations, as was observed for Eu and Yb complexes
with the anion of 6-phosphonic-2,2’-bipyridine acid in the
solid state.[30] For Tb, the calculation led to a hydration
number of two. Such elevated values of q suggest that L6 is
acting as a heptadentate ligand. This statement agrees with
the fact that the formal “ꢀ4” charge brought by the phos-
phonate functions largely counterbalances the positive
charge of the lanthanide cations and this polarisation desta-
bilises the interaction with the carboxylate function, possibly
disrupting it. A rapid kinetic equilibrium between species
with a bonded or a free carboxylate would result in average
hydration numbers of between one and two.
Figure 7. ORTEP view of complex [Eu
(L3H2)Cl]+ (top). Representation
of the europium coordination sphere showing the helical wrapping of the
ligand (bottom left) and the coordination polyhedron around europium
(bottom right).
A
ing the coordination of the four nitrogen atoms of the bipyr-
idyl strands and of the oxygen atoms of the carboxylic acid
functions. The bipyridyl–carboxylate frames are almost
ꢀ
planar with dihedral angles of 8.6 and 4.68 around the C C
bond that links the pyridyl rings, and 16.4 and 4.68 for the
ꢀ
dihedral angles around the C C bond that links the external
pyridyl ring and the carboxylic functions. The helical wrap-
ping of this heptadentate part results in more than one turn,
so that the carboxylic acid functions are placed one above
the other. As a result of the acidic conditions of crystallisa-
tion, only a single bipyridyl carboxylic function was deproto-
X-ray crystal structure of [Eu
A
crystals of [Eu
ACHTREUNG
concentration of a solution containing equimolar amounts of
L3H3·4HCl and EuCl3·6H2O in water. The crystal structure
of the complex (Figure 7) shows it to be composed of a eu-
ꢀ
nated, which resulted in two distinct Eu O distances of
2.390 and 2.520 Š, respectively, for the deprotonated and
ꢀ
ropium cation coordinated by a monodeprotonated L3H2
protonated oxygen atoms. The Eu N distances of the pyrid-
ꢀ
ligand and a chloride anion. The charge of the resulting [Eu-
yl rings are, on average, 2.56 Š, corresponding to typical
values for this type of ligand.[32,33] The remaining two coordi-
nation sites of the nonacoordinated europium atom are posi-
tioned on opposite sites from the plane perpendicular to the
helical axis. They are occupied by a chloride anion and by
the alkoxy residue of the serine moiety. Surprisingly, the im-
posed acidic conditions led to the protonated carboxylic
acid function and, in that case, a coordination of the alcohol
was preferred to that of the protonated acid.
A
chloride anion. Three water molecules of solvation complete
the content of the crystallographic asymmetric unit.
In the complex, the ligand reveals very interesting fea-
ꢀ
tures. In particular, L3H2 is coordinated to the europium
ion through the central nitrogen atom of the serine moiety
and the bipyridyl arms are wrapped around the cation in a
single-stranded mononuclear helical arrangement,[31] allow-
354
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 346 – 358

Lanthanide Complexes
FULL PAPER
The three-dimensional framework crystal structure results
from a combination of a cooperative hydrogen-bond net-
work (Table T7, in the Supporting Information) andp stack-
ing interactions (Figure 8). The three water molecules link
firmed by FTIR studies in the solid state displaying a sharp
band at 1737 cmꢀ1 in the spectrum for the protonated car-
boxylic acid of [Eu
(L3H2)Cl]Cl, which disappeared in the
A
spectrum of [Eu(L3)
(H2O)]. Finally, deprotonation of the
G
second bipyridyl arm, with a pK value in the 4–5 range,[18b]
will compensate the overall charge of the cation, weakening
ꢀ
the Eu Cl bond, and resulting in its facile decoordination
and replacement by a water molecule. According to this X-
ray crystal structure, a coordination mode of type A is rea-
sonably envisaged.
Conclusion
On the basis of simple convergent synthetic strategies, we
have developed a family of multitopic ligands for the coordi-
nation of lanthanide cations. All these ligands carry two bi-
pyridine subunits functionalised with carboxylate or phos-
phate anionic functions. The corresponding europium and
terbium complexes are all luminescent in water at neutral
pH and their photophysical characteristics can be finely in-
terpreted within the frame of a structure–property relation-
ship.
The various changes made in comparison with the refer-
ence complexes of L (glutamate) allowed us to draw the fol-
lowing conclusions. From the complexation behaviour of
ligand L3 (serine) and L4 (aminomalonate), it clearly ap-
peared that the coordination mode C can be disregarded. In
both cases, hydration numbers and absorption data suggest
complexation trends similar to that of L and the additional
geminal coordinating arms notably influence the relative ef-
ficiencies of luminescence, providing supplementary radia-
tive deactivation pathways. Replacing bipyridyl carboxylic
acid functions by phosphonic functions had the opposite
effect to the one expected. While it was hoped that the bulk-
ier phosphonic functions should sterically hinder the com-
plexation of the coordinated water molecules,[14,15] it resulted
in a weakened coordination of the carboxylate moiety open-
ing the coordination shell to even more water molecules. Fi-
nally, the approach consisting of the elongation of the gluta-
mic coordinating arm showed contrasting results. Elongation
through the addition of a supplementary pyridyl ring led to
the decoordination of one bipyridyl strand, allowing the
access of three water molecules. The elongation through ad-
dition of an aminodiacetate function was finally successful,
leading to the displacement of the water molecule and the
perfect filling of the coordination sphere. It is nevertheless
not clear whether this is the proof of a coordination mode
of type B, or, as suggested by the absorption data, it arose
from a change in the coordination mode with a potential
decadentate coordination mode. Last but not least, a better
insight into the coordination mode was probably gained
through the methodology that first failed and engendered
this study. X-ray diffraction analysis on single crystals of the
europium complex of L3 (serine) revealed a wrapping of the
two bipyridylcarboxylate arms that did not leave much
room for another coordination mode other than the A type.
Figure 8. Three-dimensional packing of [Eu
(L3H2)Cl]Cl·3H2O.
A
the protonated O5 atom from the carboxylate function of a
[Eu
(L3H2)Cl]+ molecule at (x, y, z) to the unprotonated O7
ACHTREUNG
atom of a molecule at (1ꢀx, 1ꢀy, ꢀz) in a cooperative hy-
drogen-bond network to form a layer structure parallel to
the ꢀab plane (Figure S1, in the Supporting Information).
The hydrogen-bond ring pattern formed is of graph set
R45(8).[34] The stabilisation is increased by a single p–p stack-
ing interaction. The N5 pyridyl rings in the molecule at (x, y,
z) and (ꢀx, 1ꢀy, ꢀz) are strictly parallel, with an interplanar
spacing of 3.443(3) Š; the ring-centroid separation is
3.665(3) Š, corresponding to a ring offset of 1.103(3) Š.
Among the three water molecules, OW2 and OW3 are in-
volved with the chloride anion Cl2 in a hydrogen-bond pat-
tern of graph set R46(12) that runs along the b-axis direction,
ꢀ
alternately with the N1 N2 bipyridyl groups in the molecule
at (x, y, z) and (1ꢀx, 1ꢀy, 1ꢀz) in weak p–p stacking inter-
action (minimum centroid separation 4.013(3) Š; minimum
interplanar spacing of 3.225(3) Š).
It is worth noting that if the pH of the mother solution
was raised above 4 by addition of diluted NaOH, a white
precipitate formed affording the [Eu(L3)
(H2O)] neutral
R
complex. The deprotonation of the carboxylic acid functions
resulting from the pH increase is suspected to generate a re-
arrangement in the coordination mode. A simple rotation
ꢀ
around the C N bond of the serine core will lead to a pre-
ferred coordination of the carboxylate function. This is con-
Chem. Eur. J. 2007, 13, 346 – 358
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
355

L. J. Charbonni›re, R. Ziessel et al.
Data reduction and cell-dimension post-refinement were performed by
using the HKL2000 package.[41] Correction for absorption mainly due to
Eu atoms was used as implemented within SCALEPACK.[41] The struc-
ture was solved by using the PATTERSON method (SHELXS-97)[42] and
all non-hydrogen atoms were refined with anisotropic displacement pa-
rameters by using SHELXL-97[42] by full-matrix least-squares on F2
values. All hydrogen atoms were located in difference Fourier maps, but
a riding model was used with Uiso(H) set at 1.2Ueq of the attached atom
for all them except the hydrogen atoms of the water molecules for which
the positional parameters were freely refined, but their Uiso(H) values set
at 1.5Ueq(O).
This last result unfortunately points to great difficulties in
trying to remove the coordinated water molecule by a judi-
cious ligand design. Even when the coordination sphere
could be completed, the photophysical improvements
brought about by the removal of the water molecule re-
mained largely modest compared to the synthetic efforts de-
voted to their preparation.
Experimental Section
Crystal data for [Eu
(L3H2)Cl]Cl·3H2O: [C27H28Cl2EuN5O10], Mr =
¯
group P1, a=10.002(1), b=12.793(1), c=13.012 (1) Š, a=99.960(6), b=
Materials and methods: Column chromatography and flash column chro-
matography were performed by using silica (0.063–0.200 mm, Merck) or
silica gel (40–63 mm, Merck) or on standardised aluminium oxide (Merck,
Activity II-III). Acetonitrile was filtered over aluminium oxide and distil-
led over P2O5, DMF was distilled under reduced pressure and
96.173(6), g=106.178(3)8, V=1553.5(2) Š3, Z=2, l=0.7107 Š, F
(000)=
A
804, m=2.254 mmꢀ1, T=293(2) K, 1calcd =1.722 mgmꢀ3, 11750 data meas-
ured of which 5981 unique (Rint =0.0324) and 5979 observed with I>
2s(I), Completeness to qmax =98.0%, 426 parameters with no restraints,
wR2 =0.1025, S=1.051, R1 =0.0468 (all data), final difference max peak/
diisopropylethylamine was refluxed over KOH and distilled prior to use.
G
ꢀ3
hole=1.144/ꢀ1.824 eŠ
.
Other solvents were used as purchased. Melting points were measured by
using a Büchi Melting Point 535 apparatus and are uncorrected. 1H and
13C NMR spectra were recorded on Bruker AC 200, Avance 300 and
Avance 400 spectrometers working at 200, 300 or 400 MHz, respectively.
31P (162 MHz) spectra were obtained by using the Bruker Avance 400
spectrometer, implemented with internal calibration mode. Chemical
shifts are given in ppm, relative to the residual protonated solvent.[35] IR
spectra were recorded by using a Nicolet 210 spectrometer as KBr pel-
lets. Compounds 1,[16] 3,[20] 4[21] and 7[22] were obtained according to litera-
ture procedures and full experimental procedures for the synthesis of in-
termediate compounds 5, 6, 8, 9, 10, 13, 14, 16, 17, and 20 are reported in
the Supporting Information.
CCDC-611675 contains the supplementary crystallographic data for this
paper, these data can be obtained free of charge from the Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Crystal data and selected experimental details, list of atomic coordinates,
bond lengths and angles, anisotropic thermal parameters of non-hydrogen
atoms, atomic coordinates of hydrogen atoms are reported in the Sup-
porting Information.
Preparation of ligand L1H3·4HCl: A solution of 6 (182 mg, 0.28 mmol)
and NaOH (66 mg, 1.65 mmol) in a mixture of methanol (18 mL) and
water (5 mL) was heated to 608C for 17 h. The solution was evaporated
to dryness and the resulting solid was dissolved in water (10 mL). Aque-
ous HCl (2n) was slowly added until a precipitate appeared. After centri-
fugation, the product was dissolved in methanol, filtered over Celite and
evaporated. L1H3·4HCl (83 mg, 42%) was isolated as a white powder.
1H NMR (D2O/tBuOH, 200 MHz): d=3.63 (s, 4H), 3.83 (s, 2H), 7.16 (d,
3J=7.5 Hz, 2H), 7.41–7.53 (m, 3H), 7.64–7.72 (m, 8H), 7.79–7.83 ppm
(m, 2H); IR (KBr): n˜ =3420 (s), 3085 (m), 2921 (m), 2846 (m), 1723 (m,
C=O), 1620 (m, COOasym), 1584 (m, COOasym), 1444 (m), 1384 (m,
COOsym), 1345 (m), 1302 (m), 1258 (m), 1076 cmꢀ1 (w); UV/Vis (0.01m
Tris/HCl buffer, pH 7.0): lmax (emax)=240 (18700), 287 nm (22300); MS
(FAB+): m/z (%): 577.3 (30) [L1H3+H]+; elemental analysis calcd (%)
for C31H24N6O6·4HCl: C 51.54, H 3.91, N 11.63; found: C 51.46, H 4.28,
N 11.49.
Absorption and emission spectroscopy: UV/Vis absorption spectra were
recorded on a Uvikon 933 spectrometer. Emission and excitation spectra
were recorded on a Perkin–Elmer LS50B (working in the phosphores-
cence mode) or a PTI Quantamaster spectrometer. When necessary (Eu
complexes) a Hamamatsu R928 photomultiplier was used. Luminescence
decays were obtained on the PTI Quantamaster instrument over tempo-
ral windows covering at least five decay times. Luminescence quantum
yields were measured according to conventional procedures,[36] with dilut-
ed solutions (optical density<0.05), by using [Ru(bipy)3]Cl2 in nonde-
T
gassed water (F=2.8%)[37] or rhodamine 6G in ethanol (F=88%)[38] as
references, with the necessary correction for refractive index of the
media used.[39] Estimated errors are Æ15%. Photophysical measurements
were obtained on isolated complexes for ligands L1 and L2, and on solu-
tions containing equimolar amounts of ligand and lanthanide chloride in
the other cases. Spectrophotometric titrations were performed according
to a previously published procedure.[27b] Hydration numbers (q) were ob-
tained by using Equation (1), in which tH O and tD O refers to the mea-
Preparation of ligand L2H4·4HCl·2H2O: Compound 10 (210 mg,
0.29 mmol) was heated at 808C in a solution of [Pd(PPh3)2Cl2] (20 mg,
R
29 mmol) in Et3N/EtOH (10 mL each) as catalyst under a continuous flow
of CO gas. After 16 h, a second portion of catalyst (20 mg, 29 mmol) was
added and the heating and bubbling continued for another 16 h. After
cooling, the solution was evaporated to dryness and the residue was dis-
solved in CH2Cl2, filtered over cellulose and evaporated again to dryness.
The remaining residue was purified by column chromatography (SiO2,
CH2Cl2/MeOH, 100:0 to 96:4) to afford the tetraester 11 containing trie-
thylammonium bromide and Ph3PO as impurities (Rf =0.60, SiO2,
CH2Cl2/MeOH, 80:20). The solid was dissolved in MeOH (10 mL)) and
NaOH (214 mg, 5.3 mmol) in H2O (3 mL) was added. The solution was
heated for 3 h at 608C. The pH of the solution was raised to 2 with concd
HCl, the solution was evaporated to dryness, and the residue was dis-
solved in MeOH. Addition of THF resulted in precipitation of a volumi-
nous white solid (NaCl), which was filtered. After evaporation of to dry-
ness, L2H4 (59 mg, 59%) was isolated as a beige solid as its tetrahydro-
chloride salt. 1H NMR (CD3OD, 200 MHz): d=3.6–3.9 (brm, 4H), 4.34
(s, 4H), 4.73 (s, 4H), 8.20 (d, 3J=7.6 Hz, 2H), 8.29–8.43 (m, 4H), 8.72–
8.80 (m, 4H), 8.89 ppm (d, 3J=7.6 Hz, 2H); 13C NMR (CD3OD,
50 MHz): d=51.6, 55.2, 56.4, 57.7, 125.2, 126.2, 127.5, 129.1, 129.3, 142.4,
148.3, 148.6, 149.0, 155.1, 166.6, 168.7 ppm; MS (FAB+): m/z (%): 601.2
2
2
A
buffer, pH 7.0) or deuterated water (after evaporation of water, drying
under vacuum, and dissolution in D2O), respectively, by using A=1.11
and B=0.31,[7c] or A=1.2 and B=0.25,[7b] respectively, for Eu, and A=5
and B=0.06,[7b] or A=4.2 and B=0,[7a] respectively, for Tb (estimated
errorÆ0.2 water molecules).
q ¼ A Â ð1=tH Oꢀ1=tD OꢀBÞ
ð1Þ
2
2
X-ray crystallography: Room-temperature data collection was from a
thick plate-type, air-sensitive crystal mounted inside a sealed Lindemann
capillary on an Enraf–Nonius Kappa-CCD diffractometer by using graph-
ite-monochromated MoKa (l=0.71073 Š) radiation. A preliminary orien-
tation matrix and unit cell parameters were determined from a f scan of
108 with 20 sdegꢀ1 of oscillation, followed by spot integration and least-
squares refinement.[40,41] The strategy[40,41] for complete data collection in
the triclinic system suggested a f scan of 1838 range with a 1.88 oscilla-
tion, followed by three w scans of successive lengths: 47.3, 57.8 and 60.18.
“Dezingering” was accomplished by measuring each frame twice with
45 s exposure per degree. The detector was put at the position of 40 mm.
(20)
[L2H4+H]+;
elemental
analysis
calcd
(%)
for
C30H28N6O8·4HCl·2H2O: C 46.04, H 4.65, N 10.74; found: C 45.88, H
4.51, N 10.49.
356
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 346 – 358

Lanthanide Complexes
FULL PAPER
Preparation of ligand L3H3·3HCl: Compound 14 (142 mg, 0.24 mmol)
was dissolved in MeOH (20 mL) and water (20 mL) containing NaOH
(33 mg, 0.83 mmol) was added. The solution was heated at 608C for 5 h,
cooled, concentrated to approximately 10 mL under reduced pressure,
and washed with CH2Cl2 (20 mL). The pH of the aqueous layer was ad-
justed to 1 with concd HCl and the solvent was evaporated. The solid
was triturated with MeOH (3 mL), the insoluble white solid was filtered
and addition of THF resulted in the formation of a yellowish precipitate
which was isolated by centrifugation and dried under reduced pressure to
(w); UV/Vis (0.01m Tris/HCl buffer, pH 7.0): lmax (emax)=237 (11400),
289 nm
(14700);
elemental
analysis
calcd
(%)
for
C27H21N5Na6O10P2·2H2O·NaBr: C 35.47, H 2.76, N 7.66; found: C 35.75,
H 3.03, N 7.62.
Preparation of compound [Eu(L1)]·6H2O: Ligand L1H3·4HCl (22 mg,
30 mmol) was dissolved in a mixture of methanol (10 mL) and water
(10 mL), and EuCl3·6H2O (12 mg, 33 mmol) in methanol (3 mL) and
water (3 mL) was added. The solution wad heated to 708C for 1 h. After
cooling to RT, the pH was adjusted to 7.0 with a 0.5% NaOH solution in
water, and the solution was evaporated to dryness. The residue was dis-
solved in methanol and filtered over Celite, and the product was precipi-
tated with Et2O. Upon centrifugation, [Eu(L1)]·6H2O (24 mg, 95%) was
isolated as a white powder. IR (KBr): n˜ =3436 (s), 1636 (m, COOasym),
1432 (w), 1384 (m, COOsym), 1258 (w), 1086 (w), 1016 cmꢀ1 (w); UV/Vis
(0.01m Tris/HCl buffer, pH 7.0): lmax (emax)=273 (9500), 297 (9200),
321 nm (5100); MS (FAB+): m/z (%): 725.2 (85) [Eu(L1)+H]+, 727.1
yield L3H3·3HCl (126 mg, 82%) as
a
pale yellow solid. 1H NMR
(CD3OD, 200 MHz): d=4.18–4.42 (m, 3H), 4.92 (s, 4H), 8.02–8.12 (m,
4H), 8.22 (d, 3J=7.8 Hz, 2H), 8.46 (t, 3J=7.6 Hz, 2H), 8.56 (d, 3J=
7.8 Hz, 2H), 8.66 ppm (d, 3J=8.0 Hz, 2H); IR (KBr): n˜ =2974 (w), 2901
(w), 1719 (s), 1629 (s), 1262 (m), 1049 cmꢀ1 (m); UV/Vis (0.01m Tris/HCl
buffer, pH 7.0): lmax (emax)=239 (20300), 288 nm (24800); MS (FAB+):
m/z (%): 530.2 (20) [L3H3 +H]+; elemental analysis calcd (%) for
C27H23N5O7·3HCl: C 50.75, H 4.11, N 10.96; found: C 50.63, H 4.38, N
10.82.
(100)
[Eu(L1)+H]+;
elemental
analysis
calcd
(%)
for
C31H21EuN6O6·6H2O: C 44.67, H 3.99, N 10.08; found: C 44.59, H 3.56, N
Preparation of ligand L4H3·3HCl·3H2O: A solution of compound 17
(103 mg, 0.16 mmol) and NaOH (50 mg, 1.25 mmol) in a mixture of
methanol (10 mL) and water (5 mL) was heated to 708C for 5 h. The so-
9.91.
Preparation of compound [Tb(L1)]·5H2O: Ligand L1H3·4HCl (20 mg,
28 mmol) was dissolved in a mixture of methanol (10 mL) and water
(10 mL), and TbCl3·6H2O (11 mg, 30 mmol) in methanol (3 mL) and
water (3 mL) was added. The solution wad heated to 708C for 1 h. After
cooling to RT, the pH was adjusted to 7.1 with a 0.5% NaOH solution in
water, and the solution was evaporated to dryness. The residue was dis-
solved in methanol and filtered over Celite, and the product was precipi-
tated with Et2O. Upon centrifugation, [Tb(L1)]·5H2O (19 mg, 84%) was
isolated as a white powder. IR (KBr): n˜ =3434 (s), 1636 (m, COOasym),
1587 (m, COOasym), 1474 (w), 1384 (m, COOsym), 1262 (w), 1169 (w),
1083 (w), 1023 cmꢀ1 (w); UV/Vis (0.01m Tris/HCl buffer, pH 7.0): lmax
(emax)=272 (8500), 297 (9000), 318 nm (5200); MS (FAB+): m/z (%):
733.4 (80) [Tb(L1)+H]+; elemental analysis calcd (%) for
C31H21N6O6Tb·5H2O: C 45.27, H 3.80, N 10.22; found: C 44.97, H 3.25, N
10.02.
lution was evaporated to dryness and the resulting solid was dissolved in
water (8 mL). At 08C, aqueous HCl (1n) was slowly added until a pre-
cipitate appeared (pH 4–5). Upon centrifugation, L4H3·3HCl·3H2O
(59 mg, 53%) was isolated as a white powder. 1H NMR (NaOD/tBuOH,
300 MHz): d=3.75 (s, 4H), 4.04 (s, 1H), 6.84 (d, 3J=7.5 Hz, 2H), 7.15–
3
3
7.26 (m, 4H), 7.32 (d, J=7.5 Hz, 2H), 7.42 (t, J=7.5 Hz, 2H), 7.56 ppm
(d, 3J=7.5 Hz, 2H); 13C NMR (NaOD/tBuOH, 75 MHz): d=60.3, 79.4,
119.9, 122.9, 124.1, 124.4, 138.2, 138.6, 152.8, 153.7, 154.0, 158.7, 172.3,
177.3 PPM; IR (KBr): n˜ =3443 (s), 2913 (w), 2846 (w), 1720 (m, C=O),
1622 (m, COOasym), 1448 (w), 1384 (m), 1356 (m, COOsym), 1249 (w),
1169 cmꢀ1 (w); UV/Vis (0.01m Tris/HCl buffer, pH 7.0): lmax (emax)=239
(20300), 288 nm (24800); MS (FAB+): m/z (%): 544.2 (20) [L4H3+H]+;
elemental analysis calcd (%) for C27H21N5O8·3HCl·3H2O: C 45.87, H
4.28, N 9.91; found: C 45.75, H 4.09, N 9.78.
Preparation of compound (Et3NH)[Eu(L2)]·4H2O: In
a solution of
Preparation of ligand L5Na4·5H2O: Compound 20 (51 mg, 65 mmol) was
dissolved in aqueous NaOH (6 mL, 0.05n). The solution was heated to
1008C for 19 h. After cooling to RT, the aqueous layer was extracted
four times with CH2Cl2 (5 mL) and evaporated to dryness. Upon precipi-
tation in a H2O/THF mixture, L5Na4·5H2O (45 mg, 79%) was isolated as
MeOH (20 mL) and water (5 mL) were dissolved ligand L2H4
·4HCl·2H2O (25 mg, 35 mmol) and EuCl3·6H2O (14 mg, 38 mmol). The
solution was heated to 608C for 3 h, cooled to RT, and concentrated to
5 mL. Et3N was added to bring the pH to 8–9 and the solution was
evaporated to dryness. The residue was dissolved in H2O (10 mL), and
the aqueous phase was washed twice with CH2Cl2 (15 mL), and evaporat-
ed to dryness. The solid was dissolved in a minimum amount of MeOH
and addition of Et2O resulted in the formation of a white precipitate,
which was collected by centrifugation and dried under vacuum to afford
(Et3NH)[Eu(L2)]·4H2O (17.7 mg, 51%). IR (KBr): n˜ =3420 (s), 1592 (s,
COOasym), 1460 (m), 1420 (m), 1384 (m, COOsym), 1254 (w), 1090 (w),
1013 (w), 776 cmꢀ1 (w); MS (FABꢀ): m/z (%): 749.3 (100) [Eu(L2)]ꢀ; ele-
mental analysis calcd (%) for C36H40EuN7O8·4H2O: C 46.85, H 5.25, N
10.63; found: C 46.78, H 5.17, N 10.42.
Preparation of compound (Et3NH)[Tb(L2)]·3H2O: The compound was
obtained similarly to the europium one starting from L2H4·4HCl·2H2O
(20 mg, 28 mmol) and TbCl3·6H2O (10.5 mg, 28 mmol) to give
(Et3N)[Tb(L2)]·4H2O (21.3 mg, 78%). Elemental analysis calcd (%) for
C36H40TbN7O8·3H2O: C 47.42, H 5.09, N 10.75; found: C 47.32, H 4.89, N
10.58.
a
beige powder. 1H NMR (300 MHz, D2O/tBuOH): d=1.18 (t, 3J=
7.0 Hz, 6H), 2.06–2.27 (m, 2H), 2.37–2.58 (m, 2H), 3.50 (t, 3J=7.5 Hz,
1H), 3.86–3.99 (m, 4H), 4.02–4.24 (m, 4H), 7.48 (d, 3J=7.0 Hz, 2H),
7.59–7.81 ppm (m, 10H); 13C NMR (75 MHz, D2O/tBuOH): d=16.4,
16.5, 27.8, 35.6, 59.8, 62.4, 62.5, 71.6, 121.2, 124.0, 124.1, 125.7, 127.1,
127.4, 138.0, 138.2, 138.5, 154.6, 155.0, 156.3, 156.6, 157.8, 160.6, 181.1,
183.6 ppm; 31P NMR (162 MHz, D2O): d=10.17 ppm; IR (KBr): n˜ =3450
(s), 1637 (m, COOasym), 1384 (m, COOsym), 1215 (w, P=O), 1149 (w), 1063
(w), 1030 cmꢀ1 (w); MS (FAB+): m/z (%): 764.2 (10) [Na4L5ꢀNa]+; ele-
mental analysis calcd (%) for C31H31N5Na4O10P2·5H2O: C 42.43, H 4.71,
N 7.98; found: C 42.35, H 4.55, N 7.78.
Preparation of compound L6Na6·2H2O·NaBr: Compound 20 (65 mg,
83 mmol) and trimethylsilyl bromide (165 mL, 1.2 mmol) were dissolved in
CH2Cl2 (10 mL). The solution was stirred at RT under argon for 17 h,
and was evaporated to dryness. Trimethylsilyl bromide (165 mL,
1.2 mmol) and CHCl3 (10 mL) were added and the resulting solution was
refluxed for 3 h. After evaporation of the solvent, NaOH (20 mg,
0.5 mmol) and water (5 mL) were added and the solution was heated to
808C for 31 h. The solution was evaporated to dryness, the residue was
solubilised in a mixture of methanol and water, and the product was pre-
cipitated by addition of THF. Upon centrifugation, L6Na6·2H2O·NaBr
Preparation of compound [Eu
(L3H2)Cl]Cl·3H2O: Ligand L3H3·3HCl
R
(41.6 mg, 65 mmol) and EuCl3·6H2O (24 mg, 65 mmol) were dissolved in
MeOH (25 mL) and the solution was refluxed for 3 h. The solvent was
removed under reduced pressure, the residue was redissolved in MeOH
(3 mL) and THF (12 mL) was added, resulting in the formation of a volu-
minous white precipitate. The solution was filtered over Celite and was
allowed to concentrate by slow evaporation of the solvent. White crystals
formed and were recovered by removal of the solvent. The crystals were
1
(54 mg, 71%) was isolated as a yellow powder. H NMR (300 MHz, D2O/
tBuOH): d=2.06–2.18 (m, 2H), 2.36–2.48 (m, 2H), 3.45 (t, 3J=7.0 Hz,
1H), 4.03–4.21 (m, 4H), 7.51 (d, 3J=7.5 Hz, 2H), 7.67–7.87 ppm (m,
10H); 13C NMR (75 MHz, D2O/tBuOH): d=27.5, 35.6, 59.1, 70.5, 121.5,
122.5, 124.9, 126.1, 126.4, 137.6, 137.7, 138.7, 155.4, 155.7, 155.8, 160.4,
163.4, 164.7, 181.2, 183.6 ppm; 31P NMR (162 MHz, D2O): d=7.90 ppm;
IR (KBr): n˜ =3451 (s), 1637 (m, COOasym), 1384 (m, COOsym), 1076 cmꢀ1
dried under reduced pressure to afford [Eu
(L3H2)Cl]Cl·3H2O (24 mg,
C
51%). IR (KBr): n˜ =3347 (s, br), 3070 (w), 1737 (w, COOHasym), 1625,
1593 (m, COOasym), 1572 (w) 1439 (w), 1415 (w), 1378 (m, COOsym), 1273
(w), 1252 (w), 1187 (w), 1012 (w), 773 cmꢀ1 (m); MS (FAB+): m/z (%):
714.2 (100) [Eu
N
N
Chem. Eur. J. 2007, 13, 346 – 358
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
357

L. J. Charbonni›re, R. Ziessel et al.
calcd (%) for C27H28EuN5O4Cl2: C 39.47, H 3.44, N 8.53; found: C 39.19,
H 3.42, N 8.37.
[13] N. Hildebrandt, L. Charbonni›re, H.-G. Lçhmannsrçben, R. Ziessel,
Angew. Chem. 2005, 117, 7784; Angew. Chem. Int. Ed. 2005, 44,
7612.
Preparation of compound [Eu(L3)(H2O)]·NaCl: Ligand L3H3·3HCl
A
[14] S. Aime, E. Gianolo, D. Corpillo, C. Cavallotti, G. Palmisano, M.
Sisti, G. B. Giovenzana, R. Pagliarin, Helv. Chim. Acta 2003, 86, 615.
[15] a) S. Aime, M. Botta, S. Geninatti Crich, G. Giovenzana, R. Pagliar-
in, M. Sisti, E. Terreno, Magn. Reson. Chem. 1998, 36, S200; b) M.
Botta, Eur. J. Inorg. Chem. 2000, 399.
[16] S. Mameri, L. J. Charbonni›re, R. Ziessel, Synthesis 2003, 17, 2713.
[17] A. El Ghayoury, R. Ziessel, J. Org. Chem. 2000, 65, 7757.
[18] a) V. Penicaud, F. Odobel, B. Bujoli, Tetrahedron Lett. 1998, 39,
3869; b) S. Comby, D. Imbert, A.-S. Chauvin, J.-C. G. Bünzli, L. J.
Charbonni›re, R. F. Ziessel, Inorg. Chem. 2004, 43, 7369.
[19] T. H. Fife, T. J. Przystas,J. Am. Chem. Soc. 1982, 104, 2251.
[20] A similar protocol was described for methoxy esters: S. Jew, B.
Park, D. Lim, M. G. Kim, I. K. Chung, C. I. Hong, J.-K. Kim, H.-J.
Park, J.-H. Lee, H. Park, Bioorg. Med. Chem. Lett. 2003, 13, 609.
[21] R. Fornassier, D. Milani, P. Scrimin, U. Tonellato, J. Chem. Soc.
Perkin Trans. 2 1986, 233.
(41.6 mg, 65 mmol) and EuCl3·6H2O (24 mg, 65 mmol) were dissolved in
25 mL MeOH and the solution was heated to reflux for 3 h. The solvent
was removed under reduced pressure, the residue redissolved in water
(10 mL), and the pH was slowly raised by addition of diluted NaOH in
water. Around pH 4–5, a voluminous precipitate formed which was col-
lected by centrifugation, was washed successively with MeOH and THF,
and was dried under reduced pressure to afford [Eu(L3)
(H2O)]·NaCl
G
(33 mg, 67%) as a white solid. IR (KBr): n˜ =3407 (s), 3264 (s), 3096 (w),
1606 (s, COOasym), 1583 (s, COOasym), 1461 (w), 1411 (m), 1371 (m,
COOsym), 1273 (w), 1010 (w), 784 cmꢀ1 (w); MS (FAB+): m/z (%): 680.2
(100) [Eu(L3)+H]+, 648.1 (30) [Eu(L3)ꢀCH2OH]+; elemental analysis
calcd (%) for C27H20N5O7Eu·NaCl: C 42.95, H 2.94, N 9.28; found: C
42.56, H 2.61, N 8.90.
[22] K. B. Jensen, T. M. Braxmeier, M. Demarcus, J. G. Frey, J. D. Kil-
burn. Chem. Eur. J. 2002, 8, 1300.
Acknowledgements
[23] M. Montalti, S. Wadhwa, W. Y. Kim, R. A. Kipp, R. H. Schmehl,
Inorg. Chem. 2000, 39, 76.
[24] a) P. Krumholz, J. Am. Chem. Soc. 1951, 73, 3487; b) K. Nakamoto,
J. Phys. Chem. 1960, 64, 1420.
The French Minist›re de la Recherche et de lꢁEnseignement SupØrieur is
gratefully acknowledged for financial support (N.W.).
[25] a) L. J. Charbonni›re, R. Ziessel, M. Montalti, L. Prodi, N. Zaccher-
oni, C. Boehme, G. Wipff, J. Am. Chem. Soc. 2002, 124, 7779; b) M.
Montalti, L. Prodi, N. Zaccheroni, L. Charbonni›re, L. Douce, R.
Ziessel, J. Am. Chem. Soc. 2001, 123, 12694.
[26] a) N. Chatterton, Y. Bretonni›re, J. PØcaut, M. Mazzanti, Angew.
Chem. 2005, 117, 7767; Angew. Chem. Int. Ed. 2005, 44, 7595; b) N.
Weibel, L. Charbonni›re, R. Ziessel, Tetrahedron Lett. 2006, 47,
1793.
[27] a) S. Mameri, L. J. Charbonni›re, R. Ziessel, Inorg. Chem. 2004, 43,
1819; b) L. J. Charbonni›re, R. Schurhammer, S. Mameri, G. Wipff,
R. F. Ziessel, Inorg. Chem. 2005, 44, 7151.
[28] Y. Bretonni›re, M. Mazzanti, J. PØcaut, F. A. Dunand, A. E. Mer-
bach, Chem. Commun. 2001, 621.
[29] F. Renaud, C. Piguet, G. Bernardinelli, J.-C. G. Bünzli, G. Hopfgart-
ner, J. Am. Chem. Soc. 1999, 121, 9326.
[30] S. Comby, R. Scopelliti, D. Imbert, L. Charbonni›re, R. Ziessel,
J.-C. G. Bünzli, Inorg. Chem. 2006, 45, 3158.
[31] C. Piguet, G. Bernardinelli, G. Hopfgartner, Chem. Rev. 1997, 97,
2005.
[1] a) E. Toth, L. Helm, A. E. Merbach in The Chemistry of Contrast
Agents in Medical Magnetic Resonance Imaging (Eds.: A. E. Mer-
bach, E. Toth), Wiley, Chichester, 2001, pp. 45–119; b) R. B. Laufer,
Chem. Rev. 1987, 87, 901; c) P. Caravan, J. J. Ellison, T. J. McMurry,
R. B. Laufer, Chem. Rev. 1999, 99, 2293.
[2] a) J. Wçhnert, K. J. Franz, M. Nitz, B. Imperiali, H. Schwalbe, J.
Am. Chem. Soc. 2003, 125, 13338; b) M. PrudÞncio, J. Rohovec,
J. A. Peters, E. Tocheva, M. J. Boulanger, M. E. P. Murphy, H. J.
Hupkes, W. Kosters, A. Impagliazzo, M. Ubbink, Chem. Eur. J.
2004, 10, 3252; c) C. Reilley, B. W. Good, Anal. Chem. 1975, 47,
2110.
[3] R. S. Prosser, H. Bryant, R. G. Bryant, R. R. Vold, J. Magn. Reson.
1999, 141, 256.
[4] a) I. Hemmila, V.-M. Mukkala, Crit. Rev. Clin. Lab. Sci. 2001, 38,
441; b) I. Hemmila, S. Webb, Drug Discovery Today 1997, 2, 373;
c) P. R. Selvin, Annu. Rev. Biophys. Biomol. Struct. 2002, 31, 275.
[5] a) J.-C. G. Bünzli, C. Piguet, Chem. Soc. Rev. 2005, 34, 1048;
b) J.-C. G. Bünzli, Acc. Chem. Res. 2006, 39, 53.
[6] a) N. Sabbatini, M. Guardigli, J.-M. Lehn, Coord. Chem. Rev. 1993,
123, 201; b) S. I. Weissmann, J. Chem. Phys. 1942, 10, 214.
[7] a) W. D. W. Horrocks, Jr., D. R. Sudnick, J. Am. Chem. Soc. 1979,
101, 334; b) 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; c) R. M. Supkowski, W. D. W.
Horrocks, Jr., Inorg. Chim. Acta 2002, 340, 44.
[32] R. Ziessel, L. J. Charbonni›re, M. Cesario, T. PrangØ, M. Guardigli,
A. Roda, A. Van Dorsselaer, H. Nierengarten, J. Supramol. Chem.
2003, 15, 277.
[33] L. J. Charbonni›re, R. Ziessel, M. Guardigli, A. Roda, N. Sabbatini,
M. Cesario, J. Am. Chem. Soc. 2001, 123, 2436.
[34] J. Bernstein, R. E. Davis, L. Shimoni, N. E. Chang, Angew. Chem.
1995, 107, 1689; Angew. Chem. Int. Ed. Engl. 1995, 34, 1555.
[35] H. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem. 1997, 62, 1997.
[36] Y. Haas, G. Stein, J. Phys. Chem. 1971, 75, 3668.
[37] K. Nakamura, Bull. Chem. Soc. Jpn. 1982, 55, 2697.
[38] J. Olmsted, J. Phys. Chem. 1979, 83, 2581.
[39] B. Valeur in Molecular Fluorescence, Wiley-VCH, Weinheim, 2002,
p. 161.
[40] B. V. Nonius, “Collect” Data Collection Software, 1999.
[41] “Macromolecular Crystallography, Part A”: Z. Otwinovski, W.
Minor, Methods Enzymol. 1997, 307.
[42] G. M. Sheldrick, SHELX97, Program for the Refinement of Crystal
Structures from Diffraction Data, University of Gçttingen, Gçttin-
gen (Germany), 1997.
[8] G. A. Hebbink, D. N. Reinhoudt, F. C. J. M. van Veggel, Eur. J.
Inorg. Chem. 2001, 4101.
[9] a) M. Latva, H. Takalo, V.-M. Mukkala, C. Matachescu, J. C. Rodri-
guer-Ubis, J. Kankare, J. Lumin. 1997, 75, 149; b) F. J. Steemers, W.
Verboom, D. N. Reinhoudt, E. B. van der Tol, J. W. Verhoeven, J.
Am. Chem. Soc. 1995, 117, 9408; c) S. Quici, G. Marzanni, M. Cav-
azzini, P. L. Anelli, M. Botta, E. Giuanolio, G. Accorsi, G. Armaroli,
N. Armaroli, F. Barigelletti, Inorg. Chem. 2002, 41, 2777; d) S. W.
Magennis, S. Parsons, Z. Pikramenou, Chem. Eur. J. 2002, 8, 5761.
[10] C. Yang, L.-M. Fu, Y. Wang, J.-P. Zhang, W.-T. Wong, X.-C. Ai, Y.-
G
F. Qiao, B.-S. Zou, L.-L. Gui, Angew. Chem. 2004, 116, 5120;
Angew. Chem. Int. Ed. 2004, 43, 5010.
[11] N. Weibel, L. Charbonni›re, M. Guardigli, A. Roda, R. Ziessel, J.
Am. Chem. Soc. 2004, 126, 4888.
[12] L. Charbonni›re, N. Weibel, C. Estournes, C. Leuvrey, R. Ziessel,
New J. Chem. 2004, 28, 777.
Received: June 27, 2006
Published online: September 29, 2006
358
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 346 – 358