Triethylenetetramine penta- and hexa-acetamide ligands and their ytterbium complexes as paraCEST contrast agents for MRI

Article abstract of DOI:10.1039/b803557a

The ligand triethylenetetramine-N,N,N′,N″,N′″, N′″-hexaacetamide (ttham) was synthesized with the aim of forming lanthanide complexes suitable as contrast agents for magnetic resonance imaging applications utilizing the chemical exchange-dependent saturation transfer (CEST) effect. It was designed to exclude water molecules from the first coordination sphere and provide a high number of CEST active amide protons per lanthanide ion. The ligand was characterized by its protonation behavior and its complexation properties with ytterbium ions in aqueous solution. The basicity of the ttham backbone amine protons decreases in the order N _central(1) > N_terminal(1) > N_terminal(2) > N_central(2), as deduced from NMR titration experiments and from a comparison of its protonation constants with those of two ttham derivatives, in which either a terminal (N-benzyl-triethylenetetramine-N,N′,N″, N′″,N′″-pentaacetamide, 1bttpam) or a central acetamide group (N′-benzyl-triethylenetetramine-N,N,N″,N′″, N′″-pentaacetamide, 4bttpam) is substituted with a benzyl group. This protonation sequence results from the combined influence of inductive effects, the intramolecular hydrogen bonding network, and the Coulomb repulsion between protonated ammonium groups. The ytterbium complex of ttham, [Yb(ttham)]Cl₃, is coordinatively frustrated. Due to steric constraints, in addition to the four backbone nitrogen atoms, only three of the four symmetry-equivalent terminal acetamide donors can coordinate simultaneously to the ytterbium ion, and the dangling fourth one exchanges quickly with the other three. The ytterbium complexes of a total of five ligands (ttham, 1bttpam, 4bttpam, 2,2′,2″-triaminotriethylaminehexaacetamide (ttaham), and diethylenetriamine-N,N,N′,N″,N″-pentaacetamide (dtpam)) were studied with respect to their CEST properties. In solution, all of these complexes have a low symmetry. The presence of multiple magnetically different amide groups in each complex prevents the realization of very high CEST effects. These results nevertheless form an excellent basis for a further optimization of this class of ligands.

Full text of DOI:10.1039/b803557a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Triethylenetetramine penta- and hexa-acetamide ligands and their ytterbium
complexes as paraCEST contrast agents for MRI†
Dirk Burdinski,*a Johan Lub,a Jeroen A. Pikkemaat,b Diana Moreno Jalo´n,a Sophie Martiala and
Carolina Del Pozo Ochoaa
Received 29th February 2008, Accepted 16th April 2008
First published as an Advance Article on the web 12th June 2008
DOI: 10.1039/b803557a
The ligand triethylenetetramine-N,N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀ-hexaacetamide (ttham) was synthesized with the
aim of forming lanthanide complexes suitable as contrast agents for magnetic resonance imaging
applications utilizing the chemical exchange-dependent saturation transfer (CEST) effect. It was
designed to exclude water molecules from the first coordination sphere and provide a high number of
CEST active amide protons per lanthanide ion. The ligand was characterized by its protonation
behavior and its complexation properties with ytterbium ions in aqueous solution. The basicity of the
ttham backbone amine protons decreases in the order Ncentral(1) > Nterminal(1) > Nterminal(2) > Ncentral(2), as
deduced from NMR titration experiments and from a comparison of its protonation constants with
those of two ttham derivatives, in which either a terminal
(N-benzyl-triethylenetetramine-N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀ-pentaacetamide, 1bttpam) or a central acetamide
group (Nꢀ-benzyl-triethylenetetramine-N,N,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀ-pentaacetamide, 4bttpam) is substituted with a
benzyl group. This protonation sequence results from the combined influence of inductive effects, the
intramolecular hydrogen bonding network, and the Coulomb repulsion between protonated
ammonium groups. The ytterbium complex of ttham, [Yb(ttham)]Cl3, is coordinatively frustrated. Due
to steric constraints, in addition to the four backbone nitrogen atoms, only three of the four
symmetry-equivalent terminal acetamide donors can coordinate simultaneously to the ytterbium ion,
and the dangling fourth one exchanges quickly with the other three. The ytterbium complexes of a total
of five ligands (ttham, 1bttpam, 4bttpam, 2,2ꢀ,2ꢀꢀ-triaminotriethylaminehexaacetamide (ttaham), and
diethylenetriamine-N,N,Nꢀ,Nꢀꢀ,Nꢀꢀ-pentaacetamide (dtpam)) were studied with respect to their CEST
properties. In solution, all of these complexes have a low symmetry. The presence of multiple
magnetically different amide groups in each complex prevents the realization of very high CEST effects.
These results nevertheless form an excellent basis for a further optimization of this class of ligands.
%CEST = (M0 − MS)/M∞ × 100%
(1)
Introduction
MS is the intensity of the magnetic resonance (MR) signal of
the bulk water taken immediately after an RF saturation pulse has
been applied at the resonance frequency of the CA’s exchangeable
protons. M0 is the intensity of the reference signal that is recorded
after the RF saturation pulse has been applied symmetrically on
the opposite side of the bulk water signal to correct for non-
selective saturation, most prominently direct water saturation,
and M∞ is the intensity of a second reference MR signal that is
recorded at a significantly different frequency that is not affected
by magnetization transfer effects.
The majority of current magnetic resonance imaging (MRI)
contrast agents (CAs) are based on paramagnetic gadolinium
complexes. These complexes shorten the relaxation time of bulk
water protons in tissue by utilizing water molecules that are weakly
coordinated and exchange rapidly with those of the bulk water
pool.1–3 In recent years, a conceptually new class of MRI CAs
that is based on chemical exchange-dependent saturation transfer
(CEST) has attracted much attention.4–6 In the presence of a CEST
agent the signal intensity of the bulk water protons can be altered
by selectively saturating, with a strong radio frequency (RF) pulse,
the resonance frequency of mildly acidic protons of the CEST
agent, which are in an intermediately fast exchange equilibrium
with bulk water. The magnitude of the magnetization transfer, the
CEST effect, is usually calculated according to eqn (1).5
In a two-site exchange model, at complete saturation of the
exchangeable proton resonance, and in the absence of direct satu-
ration of the bulk water resonance and background magnetization
transfer, MS/M∞ is given by eqn (2a) and eqn (2b).7,8
MS/M∞ = 1/(1 + kexT1sat
)
(2a)
(2b)
aDepartment of Bio-Molecular Engineering, Philips Research, High Tech
Campus, 5656 AE, Eindhoven, The Netherlands. E-mail: dirk.burdinski@
philips.com
kex = n[CA]/2[H2O]sM
bDepartment of Materials Analysis, Philips Research, High Tech Campus,
5656 AE, Eindhoven, The Netherlands
T1sat is the spin–lattice relaxation time of the bulk water
protons during saturation of the exchanging protons and kex is
the pseudo-first-order exchange rate constant, where [H2O] is
† Electronic supplementary information (ESI) available: Fig. S1–S3. See
DOI: 10.1039/b803557a
4138 | Dalton Trans., 2008, 4138–4151
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
the concentration of the bulk water, [CA] is the concentration
of the CEST CA, n is the number and sM is the life time of
the exchangeable protons. The coordination of suitable CEST-
active structures to a paramagnetic lanthanide ion in so called
paraCEST agents increases the sensitivity of low molecular weight
CEST agents by allowing higher exchange rates (kex) of the
CEST-active protons with paramagnetically shifted resonance
frequencies without exceeding the optimum proton exchange
range (|kex| < |x(Hex) − x(H2O)|).9,10
complex bears a total of eight CEST-active amide protons on
the four equivalent acetamide pendent arms. Dotam and most of
its numerous and intensively studied derivatives are octadentate
ligands and allow for the further coordination of at least one aqua
ligand to the encapsulated lanthanide ion, which is a rudiment of
the need for an exchangeable water molecule in the gadolinium
complex of the parent ligand 1,4,7,10-tetraazacyclododecane-
1,4,7,10-tetra(acetic acid) (H4dota), a clinically used T1 CA
(DotaremTM, Guerbet). In contrast to conventional gadolinium-
based T1 CAs, such amide-based paraCEST agents do not require
a coordination site for water exchange. Its presence is even
unfavorable, due to lanthanide-induced relaxation mediated by the
weakly coordinated water molecule.22 Such T1 relaxation reduces
the magnetic saturation that is built up with the preparative RF
pulse and thereby limits the achievable CEST effect [eqn (2a)].
It was therefore an aim of this study to develop a new
ligand system for paraCEST agents that would exclude water
molecules from the first coordination sphere, at the same time
increase the number n of CEST active amide protons per
lanthanide ion [eqn (2b)], and allow for the introduction of
linker groups for a straightforward coupling to targeting bio-
active ligands. We considered the hexaamide derivative, ttham
(Chart 2), of the known decadentate ligand triethylenetetramine-
N,N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀ-hexa(acetic acid) (H6ttha, Chart 1) to be a
good candidate.23,24 The H6ttha ligand forms nonacoordinate
complexes with the late lanthanide ions, in which it typically
saturates the first coordination sphere by binding to the metal via
the four backbone amino groups as well as the oxygen atoms of five
of the six carboxylate groups.25–28 A similar mode of coordination
to ytterbium ions was therefore expected for the ttham ligand.
In such an arrangement the sixth, dangling pendent arm would
then be available for further chemical modification of the probe
with little impact on the CEST properties of the core complex.
To test this property, two ttham derivatives were synthesized,
in each of which one of the two different acetamide groups
was substituted with a non-coordinating benzyl group, rendering
the ligand potentially nonadentate. The ytterbium complexes of
ttham, of its two benzyl derivatives, and of two related new
reference ligands (Chart 2) were synthesized and studied with
respect to their NMR and CEST properties.
A general advantage of CEST-based CAs over paramagnetic
relaxation agents is that the image contrast can be switched on
and off at will, which is particularly interesting for molecular
imaging applications. The information obtained with this specific
contrast-enhancement technique can be immediately correlated
with an unenhanced reference image, which can be measured
without time delay. In this context, strategies have been devised
to modulate the paraCEST effect in response to a physiologically
meaningful parameter, such as pH,11–13 temperature,14 metabolite
concentration,15 or enzyme activity.16
A number of approaches have been described to increase the
sensitivity of known paraCEST agents into the range that makes
them attractive for molecular imaging applications. A very promis-
ing strategy is to increase the number n of CEST-active protons
[eqn (2b)] per agent by utilizing polymeric or dendritic agents
or by preparing nanoparticles that are decorated with a large
number of small molecule paraCEST agents.13,17–19 Poly(propylene
imine) dendrimers, covalently functionalized with up to 16 CEST-
active Yb complexes, were reported to show a decrease of the
lowest detectable concentration that was proportional to the
number of Yb complexes per molecule.13 The therein employed
Yb complex is derived from [Yb(dotam)(H2O)]3+ (Chart 1), the
archetypal small molecule paraCEST agent.20,21 Of the lanthanide
ions, ytterbium allows for the highest practical CEST effect, in part
due to its comparably small magnetic moment and the related long
T1 values in aqueous solution.22 The unsubstituted Yb–dotam
Experimental
All solvents were obtained from Merck. 4-Morpholine-
propanesulfonic acid (MOPS, 99.5%) and all other commercial
chemicals were obtained from Sigma-Aldrich in the highest
available quality. Di-tert-butyl 2-bromoethylamine-N,N-diacetate
31
(12),29 K3[Yb(ttha)],25 dotam,30 and [Yb(dotam)(H2O)]Cl3 were
prepared following published procedures.
pD values in D2O solutions were calculated (pD = pHD + 0.4)
from operational pH values (pHD) measured with a WTW SenTix
Mic glass electrode linked to a WTW pH 315i pH meter.32,33
NMR spectra for structure verification were recorded using a
Bruker DPX300 spectrometer of solutions in deuterated solvents
(dichloromethane, chloroform, or dimethyl sulfoxide). FT-IR
spectra were recorded on an ATI Mattson Genesis II spectrometer.
Reported absorption bands were sharp and exhibited medium to
strong intensity, unless noted otherwise (w = weak, br = broad).
MALDI-TOF mass spectra were recorded on a Voyager-DE-Pro
Chart 1
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 4138–4151 | 4139
©

View Article Online
128 ◦C for 2 h. Then a small amount of 70% perchloric acid
was added and the temperature was raised to 180 ◦C for 1 d.
The obtained solutions were cooled and diluted with water to a
known volume. C, H, N analyses were performed by means of
conductivity measurement after combustion of the material with
a plug of oxygen and separation of the combustion gasses.
Potentiometry
Titrations were carried out in a water jacket 10 cm3 titration vessel
maintained at 25.0 0.1 ◦C, using a 716 DMS Titrino automatic
titrator (Metrohm), which was coupled to a PC controlled by
the VESUV 3.0 titration software (Metrohm). The “Dynamic
Equivalent point Titration” (DET) method was chosen optimized
to achieve the maximum sensitivity for weak equivalent points.
Titrations were carried out with 0.1002 M HCl in 0.1 M NaNO3
solution under Ar to exclude atmospheric CO2. During titrations
the pH was recorded with a Metrohm “LL combined pH glass
electrode” (type 6.0233.100), which was calibrated with a series of
9 pH buffers (Merck), ranging from pH 1.00–9.00. The accuracy
of the pH measurements was 0.01 pH. Equivalence points in
the titration curve were evaluated based on the zero crossing
of the second derivative (after correction for curve distortion)
using the VESUV titration software package. The calculation of
the log K values by the software is based on the Henderson–
Hasselbalch relation between activities of a conjugated acid–base
pair in aqueous solution.
CEST NMR spectroscopy
Samples were prepared by dissolving the appropriate amount of
ytterbium complex in 20 mM MOPS buffer (pH 7.40). If required,
the pH values of the solutions were adjusted by adding small
aliquots of either 0.1 M NaOH or 0.1 M HCl stock solutions.
To enable frequency locking, a coaxial glass capillary insert
filled with deuterated tetrachloroethane was used. NMR spectra
were recorded on a Bruker Avance NMR spectrometer equipped
1
with an Oxford wide-bore 7 T superconducting magnet. 1D H
NMR measurements were performed without water suppression.
Complex data points (128k) were acquired with a dwell time
of 200 ls. Prior to Fourier transform, the data were apodized
with an exponential filter (line broadening = 5 Hz). All spectra
were calibrated relative to the H2O frequency. CEST spectra were
recorded using standard continuous-wave irradiation (2 s pulse
duration; 22.1 lT pulse amplitude) for selective presaturation
of the exchangeable-proton resonances. Of the Yb complexes
Chart 2
(PerSeptive Biosystems, Perkin-Elmer) machine using a-cyano-4-
hydroxycinnamic acid as a matrix and a positive reflector mode
detection method. ESI mass spectra were recorded on an LCQ
Deca XP Max (Thermo Finnigan) machine using an H2O–CH3CN
(1 : 1) mixture as the solvent and an injection rate of 0.05 cm3 min−1.
1
of ttham, 1bttpam, and 4bttpam, 511 individual 1D H NMR
spectra were acquired at different values of the presaturation
offset frequency (500 Hz intervals up to 127.5 kHz) centered
around the water resonance frequency and stored in a single 2D
NMR data set. Of the Yb complexes of dtpam and ttaham, 149
individual 1D 1H NMR spectra were acquired at different values
of the presaturation offset frequency (200 Hz intervals up to
10 kHz, then 1000 Hz intervals up to 100 kHz). To reconstruct
the CEST spectrum, the water signal of each individual spectrum
in the 2D data set was integrated and plotted as a function of the
presaturation offset frequency. From the resulting spectrum the
CEST effect was determined using eqn (1).
Elemental analysis
The chloride content was determined by ion chromatography of
aqueous solutions against calibration standards. The ytterbium
content was determined by inductively coupled plasma-atomic
emission spectroscopy (ICP-AES). In preparation for ICP-AES
analysis, material samples were heated with hydrochloric acid at
4140 | Dalton Trans., 2008, 4138–4151
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Syntheses
(s, 12H), 2.81 (t, J = 6.3 Hz, 6H), 2.62 (t, J = 6.3 Hz, 6H), 1.27 (t,
J = 7.1 Hz, 18H).
Hexaethyl triethylenetetramine-N,N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀ-hexaace-
tate (2). A mixture of triethylenetetramine-N,N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀ-
hexaacetic acid (1) (20.0 g, 40.4 mmol), concentrated sulfuric
acid (9.0 cm3, 0.17 mol), and ethanol (400 cm3) was refluxed
in a nitrogen atmosphere for 20 h. The reaction mixture was
concentrated under reduced pressure at 45 ◦C and diluted with
dichloromethane (150 cm3). The obtained solution was extracted
with 1.5 N NaOH solution (250 cm3) and brine (100 cm3). Drying
of the organic layer over magnesium sulfate, filtration over a short
neutral alumina column, and evaporation of the solvent yielded 2
as a yellow oil (22.4 g, 84%). 1H NMR (CDCl3, d/ppm): 4.16 (q,
J = 7.1 Hz, 8H), 4.14 (q, J = 7.1 Hz, 4H), 3.57 (s, 8H), 3.44 (s,
4H), 2.85 (t, J = 6.0 Hz, 4H), 2.78 (t, J = 6.0 Hz, 4H), 2.74 (s,
4H), 1.27 (t, J = 7.1 Hz, 12H), 1.26 (t, J = 7.1 Hz, 6H).
Triethylenetetramine-N,N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀ-hexaacetamide (tth-
am). A solution of 2 (22.4 g, 33.8 mmol) in 7 N ammonia in
methanol (300 cm3) was stirred at RT for 9 d. A precipitate was
formed, which was collected by filtration, washed with methanol
(200 cm3), and dried over silica. The product was obtained
as a white powder (5.4 g, 33%). FT-IR (ATR, cm−1): 3387
2,2ꢀ,2ꢀꢀ-Triaminotriethylaminehexaacetamide (ttaham). A so-
lution of 6 (1.82 g, 2.75 mmol) in 7 N ammonia in methanol
(14 cm3) was stirred at RT for 9 d. Evaporation of the volatiles
yielded a light yellow solid (1.24 g, 93%). The product was very
hygroscopic, but could be stored in a desiccator over silica. FT-
−1
=
IR (ATR, cm ): 3327 (NH2), 1656 (C O). MS (MALDI-TOF,
+
1
m/z): Calcd for C18H36N10O6 : 488.28. Found: 488.36. H NMR
(DMSO-d6, d/ppm): 7.62 (s, 6H), 7.14 (s, 6H), 3.04 (s, 12H), 2.53
(s, 12H). 13C NMR (DMSO-d6, d/ppm): 173.3, 59.1, 53.1, 52.5.
N-Benzyl-triethylenetetramine (8). A mixture of benzaldehyde
(48.8 cm3, 480 mmol) and triethylenetetramine (7) (78.0 g,
533 mmol) in EtOH–water (1 : 1, 425 cm3) was stirred for
15 minutes at RT. After cooling in an ice–water bath, sodium
borohydride (12.1 g, 320 mmol) was added in small portions
while maintaining the temperature below 15 ◦C. The mixture
was stirred overnight at RT, then for 3 h at reflux temperature.
After evaporation of the volatiles, water (300 cm3) was added
and the suspension was extracted with n-butanol–toluene (1 : 1,
2 × 300 cm3). The solvents were removed by evaporation and the
remains were fractionated with the aid of a Kugelrohr apparatus.
Fractions containing more than 90% of 8 (as detected by NMR
analysis) were combined and dissolved in EtOH. The solution was
cooled in an ice–water bath and concentrated hydrochloric acid
was added until no further precipitation occurred. A solid (22.4 g)
was obtained after collection by filtration and drying. This raw
product was used in the next step.
=
(NH2), 1679 and 1635 (C O). MS (MALDI-TOF, m/z): Calcd
+
1
for C18H36N10O6 : 488.28. Found: 488.19. H NMR (DMSO-d6,
d/ppm): 7.60 (s, 4H), 7.28 (s, 2H), 7.10 (s, 4H), 7.03 (s, 2H), 3.03 (s,
8H), 2.96 (s, 4H), 2.56 (s, 8H), 2.54 (s, 4H). 13C NMR (DMSO-d6,
d/ppm): 173.4, 173.2, 59.2, 58.4, 53.3, 53.2, 53.1.
Pentaethyl
(4). This compound was obtained as
diethylenetriamine-N,N,Nꢀ,Nꢀꢀ,Nꢀꢀ-pentaacetate
yellow oil in
a
76% yield following the synthesis protocol of 2, using
diethylenetriamine-N,N,Nꢀ,Nꢀꢀ,Nꢀꢀ-pentaacetic acid (3) instead of
triethylenetetramine-N,N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀ-hexaacetic acid (1). 1H
NMR (CDCl3, d/ppm): 4.16 (q, J = 7.1 Hz, 8H), 4.14 (q, J =
7.1 Hz, 2H), 3.57 (s, 8H), 3.47 (s, 2H), 2.85 (t, J = 5.7 Hz, 4H),
2.79 (t, J = 5.7 Hz, 4H), 1.27 (t, J = 7.1 Hz, 12H), 1.26 (t, J =
7.1 Hz, 3H).
Pentaethyl
N-benzyl-triethylenetetramine-N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀ-
pentaacetate (9). Triethylamine (36.2 cm3, 257 mmol) was added
to a vigorously stirred suspension of the fine-mashed product
of the previous reaction (8) (6.27 g) in toluene (250 cm3). Ethyl
bromoacetate (25.5 cm3, 230 mmol) was added dropwise. After
stirring for 6 d at RT and evaporation of the volatiles, chloroform
(250 cm3) was added. The resulting solution was extracted with
water (2 × 100 cm3), dried over magnesium sulfate, and the
solvent of the organic phase was evaporated. The excess of ethyl
bromoacetate was removed with the aid of a Kugelrohr apparatus.
The product was obtained as a yellow oil (1.88 g, 2% overall yield
over two reaction steps) after column chromatography (silica,
solvent: ethyl acetate–heptane (4 : 1)). 1H NMR (CDCl3, d/ppm):
7.18 (m, 5H), 4.16 (q, J = 7.1 Hz, 6H), 4.14 (q, J = 7.1 Hz, 4H),
3.79 (s, 2H), 3.57 (s, 4H), 3.46 (s, 2H), 3.45 (s, 2H), 3.38 (s, 2H),
2.79 (t, J = 6.3 Hz, 2H), 2.76 (s, 4H), 2.69 (t, J = 6.3 Hz, 2H),
2.68 (s, 4H), 1.27 (t, J = 7.1 Hz, 15H).
Diethylenetriamine-N,N,Nꢀ,Nꢀꢀ,Nꢀꢀ-pentaacetamide
(dtpam).
This compound was obtained as a white powder in 53% yield
following the synthesis protocol of ttham, using 4 instead of 2.
−1
=
FT-IR (ATR, cm ): 3337 (NH2), 1658 (C O). MS (MALDI-
+
1
TOF, m/z): Calcd for C14H28N8O5 : 388.22. Found: 388.15. H
NMR (DMSO-d6, d/ppm): 7.60 (s, 4H), 7.28 (s, 1H), 7.10 (s, 4H),
7.04 (s, 1H), 3.05 (s, 8H), 2.93 (s, 2H), 2.56 (s, 8H). 13C NMR
(DMSO-d6, d/ppm): 173.4, 173.2, 59.2, 58.4, 53.2, 53.1.
Hexaethyl 2,2ꢀ,2ꢀꢀ-triaminotriethylaminehexaacetate (6). 2,2ꢀ,2ꢀꢀ-
Triaminotriethylamine (5) (3.08 g, 21.1 mmol) was added dropwise
to a vigorously stirred mixture of ethyl bromoacetate (21.1 g,
126 mmol), potassium bicarbonate (24.8 g, 248 mmol), and DMF
(90 cm3), which was cooled in an ice–water bath. Stirring was
continued for 1 h in an ice–water bath and subsequently for 48 h
at RT. Following evaporation of the DMF (60 ◦C, 7 mbar), toluene
(90 cm3) and water (150 cm3) were added. After separation, the
organic layer was extracted with water (2 × 100 cm3) and brine
(100 cm3). Drying of the organic layer over magnesium sulfate and
evaporation of the solvents afforded pure 6 as a yellow oil (9.15 g,
65%). 1H NMR (CDCl3, d/ppm): 4.14 (q, J = 7.1 Hz, 12H), 3.58
N-Benzyl-triethylenetetramine-N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀ-pentaacetamide
(1bttpam). This compound was obtained as a white powder
in 58% yield following the synthesis protocol of ttham, using 9
instead of 2. FT-IR (ATR, cm−1): 3374 (NH2), 1688 and 1630
+
=
(C O). MS (MALDI-TOF, m/z): Calcd for C23H39N9O5 : 521.31.
Found: 521.23 1H NMR (DMSO-d6, d/ppm): 7.61 (s, 2H), 7.4–7.6
(broad, 5H), 7.25–7.15 (broad, 3H), 7.13 (s, 2H), 7.10 (s, 1H),
7.06 (s, 2H), 3.63 (s, 2H), 3.02 (s, 4H), 2.94 (s, 6H), 2.7–2.4 (broad,
12H). 13C NMR (DMSO-d6, d/ppm): 173.4, 173.2, 139.0, 129.3,
128.6, 127.4, 59.2, 58.9, 58.4, 57.5, 53.3, 53.2, 53.1, 52.4.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 4138–4151 | 4141
©

View Article Online
tert-Butyl N-benzyl-ethylenediamine-Nꢀ-acetate (11). A solu-
tion of tert-butyl 2-bromoacetate (1.64 cm3, 11.1 mmol) in DMF
(15 cm3) was added dropwise over a period of 1.5 h to a stirred
solution of N-benzylethylenediamine (5.00 g, 33.3 mmol) in DMF
(100 cm3), which was cooled in an ice–water bath. Stirring was
continued for 1 d. The reaction mixture was poured into water
(200 cm3), followed by extraction with toluene (200 cm3). The
organic layer was extracted with water (2 × 100 cm3), dried over
magnesium sulfate, and the solvents were removed by evaporation
to afford 11 as a viscous, light yellow liquid (2.8 g, 95%). 1H NMR
(CDCl3, d/ppm): 7.18 (m, 5H), 3.79 (s, 2H), 3.24 (s, 2H), 2.75 (s,
4H), 1.72 (br., 2H), 1.44 (s, 9H).
suspension was heated under reflux for 0.5 h, whereupon a clear
solution was obtained (within a few minutes). At RT, propionitrile
(50 cm3) was added and the solution was filtered into a covered
crystallization dish. Overnight needle-like crystals were formed,
which were collected by filtration, washed with acetone, and
dried. The product was recrystallized from methanol–propionitrile
(1 : 2, 60 cm3) to obtain a white, crystalline product, which was
dried and stored over silica (0.60 g, 75%). Anal. (%) Calcd for
C18H36N10O6Yb1Cl3·2H2O: C, 26.89; H 5.01; N 17.42; Yb, 21.52;
Cl, 13.23. Found: C, 27.00; H, 5.00; N, 17.10; Yb, 21.58; Cl,
12.53. FT-IR (ATR, cm−1): 3549 (w, br), 3250 (br), 3150 (br),
=
=
=
1676 (C O), 1649 (C O), 1604 (C O), 1461, 1430, 1412, 1322,
1285, 1092, 920, 958, 896, 659, 429. MS (MALDI-TOF, m/z):
Calcd for [Yb(ttham) − 2H]+: 660.21. Found: 660.17. MS (ESI,
m/z): Calcd for [Yb(ttham) − 2H]+: 660.21. Found: 660.10.
Penta-tert-butyl Nꢀ-benzyl-triethylenetetramine-N,N,Nꢀꢀ,Nꢀꢀꢀ,
Nꢀꢀꢀ-pentaacetate (13). A solution of 11 (6.65 g, 25.2 mmol)
in DMF (10 cm3) was added dropwise to a stirred mixture of
di-tert-butyl 2-bromoethylamine-N,N-diacetate (12) (17.7 g,
50.3 mmol), sodium bicarbonate (16.9 g, 202 mmol), and DMF
(250 cm3), which was cooled in an ice–water bath. The mixture
was stirred for 1 h in the ice–water bath, then for 5 d at 50 ◦C. It
was poured into water (500 cm3), followed by extraction with
toluene (2 × 150 cm3). The combined organic layers were washed
with water (2 × 100 cm3) and dried over magnesium sulfate.
Column chromatography (silica, solvent: ethyl acetate–heptane
[Yb(1bttpam)]Cl3. The compound was prepared following
the synthesis protocol of [Yb(ttham)]Cl3 starting from 1bttpam
(0.31 g, 0.59 mmol) and YbCl3·6H2O (0.23 g, 0.59 mmol). The
product was obtained as white crystals (0.49 g, 83%). Anal. (%)
Calcd for C23H39N9O5Yb1Cl3·3H2O: C, 32.31; H, 5.30; N, 14.74;
Yb, 20.24; Cl, 12.44. Found: C, 32.50; H, 5.44; N, 14.10; Yb, 20.52;
−1
=
Cl, 12.29. FT-IR (ATR, cm ): 3305 (br), 3123 (br), 1666 (C O),
=
1599 (C O), 1452, 1420, 1330, 1092, 917, 890, 652, 451, 431. MS
(MALDI-TOF, m/z): Calcd for [Yb(1bttpam) − 2H]+: 693.23.
Found: 693.09. MS (ESI, m/z): Calcd for [Yb(1bttpam) − 2H]+:
693.23. Found: 693.09.
1
(2 : 1)) afforded pure 13 as a yellow oil (3.79 g, 18%). H NMR
(CDCl3, d/ppm): 7.18 (m, 5H), 3.65 (s, 2H), 3.41 (s, 4H), 3.40 (s,
4H), 3.25 (s, 2H), 2.6–2.9 (m, 12H), 1.44 (s, 45H).
Nꢀ - Benzyl - triethylenetetramine - N,N,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀ - pentaacetic
acid (14). A mixture of 13 (20.8 g, 25.6 mmol), phosphoric acid
(85%) (45 cm3, 0.66 mol), and toluene (250 cm3) was stirred at RT
in a nitrogen atmosphere for 48 h, then the toluene was decanted.
The residual oil was stirred successively with dichloromethane
(100 cm3), ethyl acetate (100 cm3), and ethanol (100 cm3), and
each of the solvents was separated thereafter from the solidifying
residue. Drying of the obtained solid at 100 ◦C under high vacuum
yielded pure 14 (14 g, 100%).1H NMR (D2O, d/ppm): 7.20 (m,
5H), 4.31 (s, 2H), 3.90 (s, 4H), 3.51 (s, 4H), 3.37 (s, 2H), 3.2–3.4
(m, 8H), 3.02 (t, J = 6.0 Hz, 2H), 2.95 (t, J = 6.0 Hz, 2H).
[Yb(4bttpam)]Cl3. The compound was prepared following
the synthesis protocol of [Yb(ttham)]Cl3 starting from 4bttpam
(0.50 g, 0.96 mmol) and YbCl3·6H2O (0.37 g, 0.96 mmol). The
product was obtained as light yellow, needle-like crystals (0.65 g,
83%). Anal. (%) Calcd for C23H39N9O5Yb1Cl3·3H2O: C, 32.31; H,
5.30; N, 14.74; Yb, 20.24; Cl, 12.44. Found: C, 32.30; H, 5.40; N,
14.30; Yb, 20.98; Cl, 11.98. FT-IR (ATR, cm−1): 3073 (br), 1660
=
=
(C O), 1599 (C O), 1453, 1430, 1323, 1119, 1098, 704, 648, 596,
457. MS (MALDI-TOF, m/z): Calcd for [Yb(4bttpam) − 2H]+:
693.23. Found: 693.16.
Ytterbium chloride complexes of dtpam and ttaham. We were
unsuccessful in crystallizing these very hygroscopic compounds.
Samples of these complexes for NMR experiments were prepared
in situ by stirring a buffered aqueous solution (20 mM MOPS,
pH 7.40) containing the respective ligand (22 mM) and YbCl3
(20 mM) at 80 ◦C for 2 h and subsequently at RT overnight. The
pH was adjusted to 7.40. Dtpam complex: MS (HPLC-ESI): Calcd
for [Yb(dtpam) − 2H]+: 560.14. Found: 560.14. Ttaham complex:
a large number of HPLC peaks was detected. The related MS
spectra showed peaks at high m/z values, which are indicative of
the formation of polymeric complexes in solution, but they could
not be assigned unambiguously to particular multinuclear species.
Pentaethyl
Nꢀ-benzyl-triethylenetetramine-N,N,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀ-
pentaacetate (15). This compound was obtained as a yellow
oil in 61% yield following the synthesis protocol of 2, using 14
1
instead of 1. H NMR (CDCl3, d/ppm): 7.18 (m, 5H), 4.15 (q,
J = 7.1 Hz, 8H), 4.13 (q, J = 7.1 Hz, 2H), 3.57 (s, 2H), 3.51 (s,
4H), 3.50 (s, 4H), 3.39 (s, 2H), 2.7–2.9 (m, 8H), 2.5–2.6 (m, 4H),
1.25 (t, J = 7.1 Hz, 15H).
Nꢀ-Benzyl-triethylenetetramine-N,N,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀ-pentaacetamide
(4bttpam). This compound was obtained as a cream colored
solid in 86% yield following the synthesis protocol of ttham,
using 15 instead of 2. FT-IR (ATR, cm−1): 3339 (NH2), 1680 and
+
=
1647 (C O). MS (MALDI-TOF, m/z): Calcd for C23H39N9O5 :
Results and discussion
1
521.31. Found: 521.27. H NMR (DMSO-d6, d/ppm): 7.60 (s,
Synthesis
2H), 7.58 (s, 2H), 7.2–7.4 (broad, 5H), 7.25 (s, 1H), 7.12 (s, 4H),
7.07 (s, 1H), 3.55 (s, 2H), 3.00 (s, 8H), 2.92 (s, 2H), 2.7–2.4 (broad,
12H). 13C NMR (DMSO-d6, d/ppm): 173.5, 173.2, 139.5, 129.1,
128.5, 127.2, 59.1, 58.8, 58.6, 53.3, 53.1, 53.0, 52.1, 52.0.
The ligands ttham, dtpam, and ttaham were prepared by reaction
of ammonia with their corresponding ethyl esters 2, 4, and 6,
respectively, in methanol following an earlier described procedure
for the synthesis of N,N,Nꢀ,Nꢀ-ethylenediaminetetraacetamide
(Scheme 1).34 The ethyl esters 2 and 4 were obtained in high
[Yb(ttham)]Cl3. To a suspension of ttham (0.50 g, 1.0 mmol) in
methanol (30 cm3) was added YbCl3·6H2O (0.39 g, 1.0 mmol). The
4142 | Dalton Trans., 2008, 4138–4151
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
yield by esterification of the corresponding commercially available
acids 1 and 3, respectively, under acidic conditions. Since the
corresponding acid of hexaethyl ester 6 was not commercially
available, an alternative route was followed, in which tetraamine
5 was per-alkylated with ethyl bromoacetate. The ligands ttham
and dtpam precipitated from the methanolic reaction mixture and
could easily be isolated by filtration. The hexaamide ttaham did
not precipitate, but was obtained by evaporation of the alcohols
and ammonia from the reaction mixture after quantitative conver-
sion of 6, as probed by NMR analysis. This ligand was found to
be very hygroscopic, since the initially obtained solid turned sticky
within seconds, which made this material more difficult to handle
compared to the other two ligands.
Scheme 2
formation requires complex synthetic routes that did not promise
a significant advantage over the extensive purification procedure
of 9. For similar reasons, no attempts were made to improve the
synthesis of 8, which was isolated as its hydrochloride salt after
reductive amination of 7 in a yield of approximately 15%.
The ligand 4bttpam, the 4-benzyl isomer of 1bttpam, was
prepared according to Scheme 3. The two final steps are equal
to those of the synthesis of ttham and dtpam. In this case
the triethylenetetramine skeleton was, however, derived from the
coupling of ethylenediamine derivative 11 with two equivalents of
monoamino derivative 12. Conversion of the obtained penta-tert-
butyl ester 13 with ammonia to form 4bttpam was unsuccessful
due to the low reactivity of the sterically hindered esters. Therefore
the penta-tert-butyl ester was converted to the more reactive
pentaethyl ester 15 via the free acid 14. In an attempt to avoid
this indirect route to 15, we tried its direct synthesis from the
ethyl analogues of the tert-butyl esters 11 and 12. Formation of
the latter compound from N-benzylethyleneamine 10 and ethyl
bromoacetate was, however, not met with success, in part due to a
quick lactamization of the product forming 1-benzylpiperazin-2-
one. This side reaction is similar to the above discussed cyclization
that limited the yield of 9 (Scheme 2).
Scheme 1
The ligand 1bttpam, the 1-benzyl derivative of ttham, was
synthesized by reaction of pentaethyl ester 9 with a methanolic
ammonia solution similar to the formation of ttham (Scheme 2).
Pentaethyl ester 9 in turn was prepared by alkylation of tetraamine
8 with ethyl bromoacetate in analogy with the synthesis of 6.
However, while the yield of 6 was reasonably high (66%), 9 was
obtained in a much lower yield (2% overall, starting from 7)
after extensive column chromatography. During the alkylation
of 8, intramolecular attack of primary and secondary amines of
reaction intermediates on the ethyl ester bonds leads to cyclization
with the formation of piperazin-2-one derivatives. This process is
not possible during the alkylation of 5, since the central amino
group is a less reactive tertiary amine. Avoiding this lactam
Ytterbium complexes of the ligands ttham, 1bttpam, and
4bttpam were prepared by reaction of stoichiometric amounts
of the respective ligand with ytterbium trichloride in methanol.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 4138–4151 | 4143
©

View Article Online
complexes of the ligands dtpam and ttaham was however unsuc-
cessful. From initially crystalline precipitates, only oily materials
were obtained, which indicates that these materials are very
hygroscopic. Samples of these metal complexes for NMR studies
were therefore prepared in situ from stoichiometric amounts of
the respective ligand and ytterbium trichloride in MOPS-buffered
aqueous solution.
Protonation constants
Table 1 summarizes the protonation constants (all in log values)
determined for the five ligands of this study and compares these
values with literature data for related acetamide- and methyl-
substituted reference amines. The protonation constants are
defined in eqn (3a) and (3b).
+ H+ ꢀ BHn
(3a)
(3b)
(n − 1)+
n+
BH(n − 1)
Kn = [BHnn+]/[BH(n − 1)(n − 1)+][H+]
With the exception of dtpam, all new ligands bear four tertiary
amine groups. For each of those with a backbone derived from the
linear triethylenetetramine (ttham, 1bttpam, 4bttpam) two proto-
nation constants were determined, whereas only one protonation
constant could be determined for the branched ttaham derivative
as well as for the potentially tribasic dtpam ligand.
Based on the two protonation constants of the hexaacetamide
ttham, which were determined to be 6.32 and 3.93, an equilibrium
of three species is expected in the pH ranges between about 2 and
8. Fig. 1a shows the proton spectra of the ligand in D2O solution
in the pD range between 9.1 and 5.1. Above pD 8 (up to about pD
11, see supporting information, Fig. S1) four singlet peaks were
observed with an intensity ratio of 8 : 4 : 8 : 4 (left to right). The
two peaks at 3.30 and 3.21 ppm are assigned to methylene protons
of the eight terminal (protons a in Scheme 4) and four central
(b) acetamide groups, respectively. Likewise the second pair of
resonances at 2.72 and 2.68 ppm is assigned to the two terminal
(c,d) and one central (e) ethylene groups, respectively.
Table 1 Protonation (log Ki) constants of acetamide- and methyl-
substituted amine ligands (25 ◦C, 0.1 M NaNO3)
Liganda
ntam
log K1
log K2
log K3
log K4
Reference
2.6(1)
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
2.08(1)
5.26(3)
—
—
—
—
—
—
—
—
—
—
—
—
—
—
2.24(3)
44
51
52
51
2.47(2)b
4.36(2)
4.57(1)b
6.02(3)
6.32(3)
6.67(3)
7.54(3)
7.21(3)
9.06
edtam
dtpam
ttham
1bttpam
4bttpam
ttaham
BzMe2N
Me3N
—
This work
This work
This work
This work
3.93(3)
5.08(3)
4.01(3)
—
—
—
This work
53
54
55
56
57
58
9.79(1)
9.80
9.281(1)
9.42(1)
9.11(2)
—
Scheme 3
Me4en
Me5dien
Me6trien
6.130(2)
8.53(1)
8.35(1)
a BzMe2N = benzyldimethylamine; Me3N = trimethylamine; Me4en =
The mononuclear complexes [Yb(L)]Cl3 (L = ttham, 1bttpam,
4bttpam) were found to be highly water soluble and moderately
hygroscopic as a solid. They could, nevertheless, be recrystallized
from methanol–propionitrile. The crystallization of the ytterbium
N,N,Nꢀ,Nꢀ-tetra(methyl)ethylenediamine; Me5dien
=
N,N,Nꢀ,Nꢀꢀ,Nꢀꢀ-
penta(methyl)diethylenetetramine; Me6trien
=
N,N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀ-
hexa(methyl)triethylenepentamine. b 20 ◦C, 0.25 M KCl.
4144 | Dalton Trans., 2008, 4138–4151
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Fig. 1 1H NMR spectra of ttham solutions (30 mM) in D2O at various pD values (DNO3, KOD, no buffer, 25 ◦C). Chemical shifts (d) are relative to the
signal of sodium 3-(trimethylsilyl)-1-propanesulfonate (DSS, d = 0 ppm) as the internal standard (c(DSS) = 3 mM).
timescale the monoprotonated ttham thus appears to show a two-
fold symmetry. This suggests a fast equilibrium between mainly
two tautomeric forms of Httham+ bearing the proton at either one
of the two “central” amines, N4 or N7, while N1 and N10 remain
unprotonated.36 In other words, on the NMR time scale N4 and
N7 share a single proton.
In the second protonation step of ttham the resonance of the
four terminal acetamide methylene groups (a) became highly
shifted from 3.42 ppm at pD 5.1 to about 3.75 ppm at pD 3.5
Scheme 4
(Fig. 1b). Of the ethylene groups, now the central one (e) was
hardly affected by the protonation, but the triplets of the two
terminal groups (c,d) became shifted by some +0.2 ppm to reform
a singlet, so that at pD 3.5 a spectrum comprising four singlet
peaks was observed. It thus resembled the spectrum of the fully
deprotonated ttham with the difference that now the peaks of the
terminal methylene (a) and ethylene (c,d) groups had a smaller
chemical shift than those of the respective central groups. The
apparent symmetry of the doubly protonated ttham indicates that
on the NMR timescale all four of the amino groups share equally
a total of two protons, such that at any given time predominantly
either N1 and N7, or N4 and N10 are each singly protonated.
In fact, the second protonation constant of ttham is only slightly
smaller (Dlog K = 0.5–0.6) than the first protonation constant of
In the first protonation step a strong change of the chemical
shift of the central acetamide methylene groups (b) was observed,
which increased from 3.30 to 3.74 ppm, whereas the resonance of
theterminalmethylenegroups (a)was hardlyaffected. Protonation
is known to cause deshielding of the protons that are close to the
protonation sites, resulting in a downfield shift.35 We therefore
conclude that the first protonation is located predominantly at
the central amine groups (N4 and N7). In accordance with this,
upon protonation, the resonance of one ethylene group (assigned
to the central ethylene group (e)) shifted significantly to 3.21 ppm
and remained a singlet, whereas the remaining two groups now
appeared as a set of two less shifted triplets. We assign these triplet
peaks to the two terminal ethylene groups (c,d). On the NMR
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 4138–4151 | 4145
©

View Article Online
edtam (Table 1, Chart 3), which, considering the positive charge
of the Httham+, is consistent with the assignment of the second
protonation to occur at a terminal amine group.
the first protonation step was much less affected by the substitution
on N1 (Dlog K1 ≈ 0.35) than the second proton addition (Dlog
K2 ≈ 1.15), which is consistent with a localization of this second
protonation step on a terminal backbone nitrogen. The opposite
was found for the N4 benzyl derivative 4bttpam in comparison
to ttham. The first protonation was facilitated (Dlog K1 ≈ 1.22),
whereas the addition of the second proton occurred at almost the
same pH (Dlog K2 < 0.1). Both results are consistent with the
proposed ttham protonation sequence of N7, N1, N10 (or N4,
N10, N1, considering the symmetry of ttham).
In the case of dtpa-bis(amide)s, it had been shown that the
unprotonated ligand is capable of forming two hydrogen bond ring
systems that involve the amide hydrogens, the terminal nitrogens
and the terminal carboxylate anion.39,42,43 A very similar hydrogen
bond network is expected to be formed at the terminal nitrogens
of the per(acetamide)-substituted oligoamines, which involves
both terminal acetamide groups (Scheme 5). The preference for
the initial protonation of a central amino group can thus be
explained considering that the addition of a proton to a terminal
backbone nitrogen causes a destruction of this hydrogen bond
system. A second consideration is the electron-withdrawing power
of the acetamide substituent, which decreases the basicity of
the amino group, making each acetamide-substituted ligand of
Table 1 less strong a base compared to the respective permethylated
ethyleneamines.44 This includes the ttaham ligand, which does not
bear an acetamide group directly linked to the central basic nitro-
gen. The direct and indirect impact of the acetamide substituent
on the amine basicity are clearly illustrated by the decrease of log
K1 in the series ttaham, dtpam, edtam, and ntam (Chart 3), in
which zero, one, two, or three (H2NCOCH2)2NCH2CH2 groups,
respectively, of the basic amine are replaced with H2NCOCH2
groups. Each substitution results in a decrease of log K1 of about
1.2–2.0.
Chart 3
The spectra in Fig. 1b further suggest a third protonation of
ttham below pD 2.7, which again is associated with a large shift of
the aliphatic protons close to the terminal nitrogen atoms (a,c,d)
accompanied by the reformation of a triplet signal of the terminal
ethylene groups (c,d). It can therefore be concluded that the
protonation of ttham follows the order N7, N1, N10, although the
protonation constant of this third step has not been determined.
We did not attempt to quantify the fractions of protonation
of the four backbone nitrogens.37 Nevertheless, the above inter-
pretation of the NMR data is consistent with NMR titration
data reported for various terminal dtpa- and ttha-bis(amide)
derivatives.38–41 All of those are firstly protonated at the central
backbone nitrogen, unless the amide nitrogens are bis(alkylated),
which prevents the formation of a strong hydrogen bond system
at the unprotonated terminal nitrogens. It was therefore not
unexpected to find the first protonation constant of dtpam to
be only 0.3 log K units smaller than that of ttham and not to be
able to observe a second protonation step in the titration range
of pH 1–8 (Table 1). The initial protonation at a central amine
of both molecules makes the second protonation at a neighboring
backbone nitrogen more difficult, which in the case of dtpam
prevents a second protonation at a moderate pH, whereas in
the case of ttham the second next nitrogen neighbor becomes
protonated instead. Consequently the third proton is not placed
between the first two protonation sites of ttham, but on the second
terminal amine group to minimize the Coulomb repulsion between
the positively charged ammonium groups.
NMR and CEST experiments
1
Water-suppressed H NMR spectra of all ytterbium complexes
were measured in aqueous buffer solution at pH 2.7 (red lines)
and 7.4 (blue lines). Fig. 2 shows those of the ttham, 1btpam, and
4btpam complexes at 310 K. Since the exchange reaction of all
amide protons with solvent water is fast under these conditions,
their NMR resonances are very broad at pH 7.4.45 They become
detectable, however, at pH 2.7, where the base-catalyzed exchange
reaction is sufficiently slow.
The ttham complex did not show any sharp resonances at pH 7.4
(Fig. 2a), but multiple broad resonances were observed between
about −40 and +110 ppm. The spectrum at pH 2.7 revealed in
addition two signals of exchangeable protons, which appeared
as intensive, relatively sharp peaks at −14 and −27 ppm. They
The above protonation sequence of ttham is further supported
by the protonation constants determined for its two benzyl
derivatives 1bttpam and 4bttpam. The benzyl group itself is
expected to cause a smaller decrease in basicity (compare Me3N
vs. BzMe2N in Table 1) compared to the acetamide substituent.
Therefore a relative increase in basicity upon substitution of an
acetamide group with a benzyl group is indicative of a respective
protonation site in the parent ttham ligand. In the case of 1bttpam,
Scheme 5
4146 | Dalton Trans., 2008, 4138–4151
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
sixth acetamide donor allows for the formation of coordination
isomers in [Yb(ttham)]3+. Taking into account the broad NMR
resonances, these isomers are considered to be in fast equilibrium
exchange with each other. This situation is well known for the
lanthanide complexes of the respective hexaacetic acid derivative
H6ttha.46,47 In this case, on average only five of the six possible
acetate donors are bound to the late lanthanide ions, resulting
in a nonadentate coordination mode of the deprotonated H6ttha
ligand. Since one of the four terminal acetate groups remains
uncoordinated, this opens the possibility of its quick exchange
with any of the three coordinated and, based on the structure of
the free ligand, symmetry-equivalent acetate groups.
To understand better, if [Yb(ttham)]3+ behaves similarly, the
NMR spectra of its 4bttpam derivative were recorded (Fig. 2c). In
this case, one of the central acetamide ligands is substituted with a
non-coordinating benzyl group, and a nonadentate coordination
could only be achieved if, in addition to the four backbone
amines and the central acetamide, all four terminal acetamides
were involved. A large number of sharp NMR resonances were
observed between −50 and +80 ppm. Judging from their number
and their intensity distribution we suspect the presence of at least
two conformational isomers in solution. Some 12 resonances of
exchangeable protons could be identified, but we did not attempt
an assignment of any of these peaks.
An overall comparison of the above NMR spectra, nevertheless,
reveals a much higher similarity of the peak pattern found for
[Yb(ttham)]3+ (a) and [Yb(1bttpam)]3+ (b) than for any other
combination. Furthermore, in both of these spectra, peaks or
clusters of peaks of exchangeable protons were found at about
−14 and −27 ppm. We interpret this correspondence in terms
of a high structural similarity of these two molecules, which
in turn is in agreement with the above assumption that in the
ttham complex on average one of the terminal acetamide donor
groups remains uncoordinated. The appearance of only two amide
peaks in the [Yb(ttham)]3+ spectrum can be explained in terms of
a spectral overlap of the six sharp individual amide peaks that
were closely clustered around −14 and −27 ppm in the spectrum
of the more rigid [Yb(1bttpam)]3+. Such a spectral overlap does
not occur in case of the more isolated amide peaks that were
observed between +20 and +40 ppm in the spectrum of the
more rigid [Yb(1bttpam)]3+. The relatively high intensity of the
negatively shifted amide peaks may be caused by a combination
of this spectral overlap and the above discussed coalescence due
to conformational exchange.
If this model is a good description of the situation in solution,
a strong temperature dependence of the [Yb(ttham)]3+ NMR
spectrum is to be expected. NMR spectra of this complex
between 293 and 353 K (Fig. 3) demonstrate the expected
coalescence and subsequent sharpening of all resonances at higher
temperatures, which is a good indication that all of them indeed
stem from a set of isomers that exchange quickly with each
other.
The differences in the NMR spectra of these three complexes are
reflected in their CEST properties. A comparison of the respective
Z spectra is shown in Fig. 4a. In these Z spectra the relative inten-
sity of the bulk water signal is plotted as a function of the frequency
offset of the applied presaturation RF pulse. Reference points are
the resonance frequency of the solvent water on the frequency
offset axis (Dm = 0 Hz) and the intensity of this signal, after
Fig. 2 1H NMR spectra of [Yb(ttham)]3+ (a), [Yb(1bttpam)]3+ (b),
and [Yb(4bttpam)]3+ (c) in aqueous solution (c(complex) = 20 mM,
c(MOPS) = 10 mM, T = 310 K) at two pH values (7.40 (blue spectra),
2.70 (red spectra)). The water signal was suppressed by application of a
continuous-wave presaturation pulse (1.5 s, 10 lT). Chemical shifts (d) are
relative to the resonance frequency of TMS (d = 0 ppm), approximated
by setting the solvent water protons at 4.75 ppm. The resonances of
exchangeable protons are indicated by arrows.
are assigned to the protons of ytterbium-bound amide groups.
Unconstrained aminoacetamides coordinate to lanthanide ions
via their oxygen atom such that the NH2 group points away from
the metal ion. This results typically in a lower lanthanide induced
shift of these protons compared to those of the backbone ethylene
groups, which can show paramagnetic resonance shifts of some
hundred ppm. The observed poorly resolved NMR resonances
of the aliphatic protons are indicative of a flexible coordination
geometry, possibly involving various structural isomers in an
intermediate exchange regime. In such a situation the exchangeable
resonances of the less chemically shifted amide protons can be in a
fast exchange regime showing relatively sharp coalescence signals
whereas the aliphatic protons remain in an intermediate exchange
regime, which results in broad resonances.
Comparably well-defined were the resonances of the
[Yb(1bttpam)]Cl3 complex shown in Fig. 2b. At pH 7.4 the
expected number of 29 C-H protons was matched by the observed
number of sharp resonances between −40 and +110 ppm. From
the spectrum at pH 2.7, after juxtaposition, the resonances of 10
exchangeable protons could be identified (indicated by arrows).
The presence of individual resonances for all C–H protons is
indicative of the absence of any apparent symmetry of this
complex in solution, as expected based on the low symmetry
of the 1bttpam ligand itself. The NMR spectrum is consistent
with the coordination of all nine donor groups to the lanthanide
ion. Directly comparing the 1bttpam and the ttham complexes,
it appears that the substitution of the 1-benzyl group with a
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 4138–4151 | 4147
©

View Article Online
Fig. 3 1H NMR spectra of [Yb(ttham)]3+ (20 mM) in aqueous solution
(c(MOPS) = 10 mM, pH 7.40) at four temperatures (293, 310, 331, and
353 K). The water signal was suppressed by application of a continu-
ous-wave presaturation pulse (1.5 s, 10 lT). Chemical shifts (d) are relative
to the resonance frequency of TMS (d = 0 ppm), approximated by setting
the solvent water protons at 4.75 ppm.
a presaturation pulse with an “infinitely” large frequency offset
(M∞, Dm = 200 kHz) had been applied, on the intensity axis.
Both complexes, [Yb(ttham)]3+ (black squares) and
[Yb(1bttpam)]3+ (red circles), exhibited peaks in their
Z
Fig. 4 Proton Z spectra (a) and thereof calculated CEST spectra (b) of
[Yb(ttham)]3+, [Yb(1bttpam)]3+, and [Yb(4bttpam)]3+ in aqueous solution
(c(complex) = 20 mM, c(MOPS) = 10 mM, T = 310 K, pH 7.40). Chemical
shifts (d) are relative to the resonance frequency of TMS (d = 0 ppm),
approximated by setting the solvent water protons at 4.75 ppm.
spectra that are indicative of saturation transfer at about −15
and −30 ppm, which is in good agreement with the above
discussed exchangeable proton resonances found around −14 and
−27 ppm. The Z-spectrum of [Yb(1bttpam)]3+ further showed
two relatively sharp signals between +20 and +35 ppm, which
correspond nicely with at least two exchangeable-proton signals
found in this region (Fig. 2b). This feature was also present
in the [Yb(ttham)]3+ Z spectrum (Fig. 4a), although no clear
peak structure was resolved here. Taking into consideration that
no sharp NMR resonances of exchangeable protons could be
observed in this region (Fig. 2a), this suggests the presence of a
dynamic process associated with these protons. These protons are
apparently heavily affected by the fast conformational exchange
reaction occurring in the dissolved [Yb(ttham)]3+ complex. In
support of the above discussed coordination model for the ttham
complex, the Z spectrum of the [Yb(4bttpam)]3+ complex (Fig. 4a)
showed no clear similarities with that of [Yb(ttham)]3+.
For all three complexes, the realized CEST effect calculated ac-
cording to eqn (1) is generally smaller than 50% at all frequencies.
This is mainly a result of the appearance of CEST active proton
resonances on both sides of the resonance of the bulk water signal.
In this case, eqn (1) is not suitable to correct for the contribution
of direct water saturation, which decreases symmetrically with
increasing absolute offset frequency on both sides of this reference
signal.
The 1H NMR spectra of the ytterbium complexes of the ttaham
and dtpam ligands both showed very broad peaks between −40
and +40 or +60 ppm, respectively (Fig. 5). Both complexes
exhibited exchangeable protons on both sides of the central bulk
water resonance signal. These signals were sharp only in the case
of the dtpam complex, which is indicative of a less dynamic
solution structure. The absence of a clearly resolved pattern in
the case of the ttaham complex is consistent with observations
made for the related hexaacetic acid derivative, which despite the
three fold symmetry of the ligand itself, cannot coordinate with
all six of its acetamide donor groups to a single lanthanide ion.48,49
Hence, a situation similar to the above discussed [Yb(ttham)]3+
case is likely, in which on average only two of the three equivalent
(H2NCOCH2)2NCH2CH2 groups are coordinated to the ytterbium
ion.
The Z spectra of the ttaham and dtpam complexes depicted
in Fig. 6a both showed only broad shoulders between −20 and
+20 ppm, the chemical shift range, in which exchangeable protons
were identified in the respective NMR spectra. The distribution
of CEST-active resonances on both sides of the central water
4148 | Dalton Trans., 2008, 4138–4151
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Fig. 6 Proton Z spectra (a) and thereof calculated CEST spectra (b) of Yb
complexes of dtpam and ttaham in aqueous solution (c(YbCl3) = 20 mM,
c(ligand) = 22 mM, c(MOPS) = 10 mM, T = 310 K, pH 7.40). Chemical
shifts (d) are relative to the resonance frequency of TMS (d = 0 ppm),
approximated by setting the solvent water protons at 4.75 ppm.
Fig. 5 1H NMR spectra of Yb complexes of ttaham (a) and dtpam
(b) in aqueous solution (c(YbCl3) = 20 mM, c(ligand) = 22 mM,
c(MOPS) = 10 mM, T = 310 K) at two pH values (7.40 (blue spectra),
2.70 (red spectra)). The water signal was suppressed by application of a
continuous-wave presaturation pulse (1.5 s, 10 lT). d Values are relative
to the resonance frequency of TMS (d = 0 ppm), approximated by setting
the solvent water protons at 4.75 ppm. The resonances of exchangeable
protons are indicated by arrows.
Table 2 Proton relaxivities of Yb complexes in aqueous buffer solution
(c(MOPS) = 10 mM, pH 7.40, 310 K, 7 T)
Liganda
(T1)obs/s
r1b/10−3 mM−1 s−1
ttham
2.71
2.96
2.61
2.43
1.91
2.29
2.06
6.92
5.39
7.61
9.05
14.7
10.3
12.7
resonance results for both complexes in only small realized
CEST effects based on eqn (1), not exceeding 35% at a metal
concentration of 20 mM and at an optimized power level of the
presaturation pulse (Fig. S2 and S3).
1bttpam
4bttpam
dotam
H4dota
dtpam
The discussed structural differences are reflected in the longi-
tudinal water proton relaxivities r1 of the ytterbium complexes at
7 T as summarized in Table 2. The sterically constrained ttaham
ligand is expected to allow for the coordination of one to two water
molecules per ytterbium ion and as expected showed a relaxivity
almost as high as the respective [Yb(dota)(H2O)]− complex.50
The dotam and dtpam complexes showed a somewhat smaller
relaxivity, which is consistent with the assumption that both bear
not more than one water molecule in the first coordination sphere.
As anticipated, the ytterbium complexes of the three trien-based
acetamide ligands ttham, 1bttpam, 4bttpam showed comparably
small relaxivities in the order 4bttpam > ttham > 1bttpam. This
ttaham
a dotam
H4dota
=
1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetamide;
1,4,7,10-tetraazacyclododecane-1,4,7,10-tetra(acetic acid).
=
b Calculated as r1 = [(1/T1)obs − (1/T1)dia]/c(Yb), c(Yb) = 20 mM,
(T1)dia = 4.34 s.
order is in line with the interpretation of their NMR spectra,
from which a complete saturation of the first coordination sphere
with nine donor groups was deduced. The coordination mode is
relatively rigid in the case of the 1bttpam complex, which leaves
no opening for a close interaction with water molecules. It is more
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 4138–4151 | 4149
©

View Article Online
dynamic with respect to the terminal acetamide donors in the
case of ttham, which may allow for a transiently closer interaction
with the solvent water. Finally it involves an equilibrium with
potentially significant structural changes in the case of the 4bttpam
complex, which would allow for an even more intense interaction
of the ligand with water molecules. Overall the decrease in
relaxivity of these coordinatively saturated complexes compared
to [Yb(dotam)(H2O)]3+ is nevertheless small. This clearly shows
that the first coordination sphere contribution to the T1 relaxivity
in these positively charged amide complexes is comparably small
and second- and outer-sphere effects dominate.
Acknowledgements
We thank our colleagues Danielle Beelen and Se´verine Girard
for NMR measurements, Piet Rommers for potentiometric mea-
surements, Hugo Knobel for mass-spectrometric measurements,
Thea Haex, Jeannette Smulders, Carry Hermans, and Adrie
Jonkers for elemental analyses. We are grateful to our colleagues
Holger Gru¨ll, Sander Langereis, Rene´ Wegh, and Nico Willard
for the fruitful collaboration and their very useful comments
on the manuscript. Financial support by the Dutch government
(Project code: BSIK 03033 “Molecular Imaging of Ischemic Heart
Disease”) is gratefully acknowledged.
Conclusion
References
1 S. Aime, M. Botta and E. Terreno, Adv. Inorg. Chem., 2005, 57, 173–
The five ligands ttham, 1bttpam, 4bttpam, ttaham, and dtpam
were synthesized with the aim of forming lanthanide complexes
suitable as contrast agents for magnetic resonance imaging ap-
plications utilizing the chemical exchange-dependent saturation
transfer (CEST) effect. All ligands were characterized by their
protonation constants and their complexation properties with
ytterbium ions in aqueous solution.
In the case of the ttham ligand, three protonation steps of
the backbone amino groups were characterized by 1H NMR
spectroscopy. The basicity of these amino groups was found to
decrease in the order Ncentral(1) > Nterminal(1) > Nterminal(2) > Ncentral(2), as
further deduced from a comparison of the protonation constants
of ttham with those of the two ttham derivatives 1bttpam and
4bttpam, in which either a terminal or a central acetamide group
is substituted with a benzyl group. The sequence of protonation
is controlled by inductive effects of the substituents of the basic
amine sites, the relative stability of the affected intramolecular
hydrogen bonding networks, and the Coulomb repulsion between
the different protonated ammonium groups.
The comparison between the 1H NMR properties of the
ytterbium complexes of the ttham ligand and its two monobenzyl
derivatives 1bttpam and 4bttpam led to the identification of
a dynamic equilibrium in the ttham complex. The ytterbium
complex of ttham is coordinatively frustrated, since due to steric
constraints only three of the four symmetry-equivalent terminal
acetamide donors can coordinate simultaneously to the ytterbium
ion, and the dangling fourth one exchanges quickly with the other
three. For all five studied ytterbium complexes, including those of
the new ligands ttaham and dtpam, a number of exchangeable-
proton resonances have been identified that were distributed
over both positive and negative frequency offsets relative to the
resonance peak of the bulk water signal. As a consequence of
the low symmetry of these complexes, the thus high number of
chemically different amide groups, and the presence of different
coordination isomers that are in a fast equilibrium exchange with
each other, the overall CEST effects were smaller than expected,
when these ligands were designed to yield particularly high
CEST effects in their ytterbium complexes bearing no lanthanide-
coordinated water molecules. Our current studies are therefore
directed towards a better understanding of the dynamics observed
in these complexes and their influence on the complex stability
with the aim of their further improvement for potential MR sensor
applications. The results of these studies will be the subject of a
subsequent publication.
237.
2 P. Caravan, J. J. Ellison, T. J. McMurry and R. B. Lauffer, Chem. Rev.,
1999, 99, 2293–2352.
´
3 E. To´th, L. Helm and A. E. Merbach, Comprehensive Coordination
Chemistry II, Elsevier Pergamon, Amsterdam, 2004, vol. 9, pp. 841–
881.
4 S. Aime, S. Geninatti Crich, E. Gianolio, G. B. Giovenzana, L. Tei and
E. Terreno, Coord. Chem. Rev., 2006, 250, 1562–1579.
5 J. Zhou and P. C. M. van Zijl, Prog. Nucl. Magn. Reson. Spectrosc.,
2006, 48, 109–136.
6 S. Aime, D. Delli Castelli and E. Terreno, Angew. Chem., Int. Ed., 2005,
44, 5513–5515.
7 S. Forsen and R. A. Hoffman, J. Chem. Phys., 1963, 39, 2892–2901.
8 S. Forsen and R. A. Hoffman, J. Chem. Phys., 1964, 40, 1189–1196.
9 M. Woods, D. E. Woessner and A. D. Sherry, Chem. Soc. Rev., 2006,
35, 500–511.
10 S. Zhang, M. Merritt, D. E. Woessner, R. E. Lenkinski and A. D.
Sherry, Acc. Chem. Res., 2003, 36, 783–790.
11 K. M. Ward and R. S. Balaban, Magn. Reson. Med., 2000, 44, 799–802.
12 S. Aime, D. Delli Castelli and E. Terreno, Angew. Chem., Int. Ed., 2002,
41, 4334–4336.
13 J. A. Pikkemaat, R. T. Wegh, R. Lamerichs, R. A. van de Molengraaf,
S. Langereis, D. Burdinski, A. Y. F. Raymond, H. M. Janssen, B. F. M.
de Waal, N. P. Willard, E. W. Meijer and H. Gru¨ll, Contrast Media
Mol. Imaging, 2007, 2, 229–239.
14 S. Zhang, C. R. Malloy and A. D. Sherry, J. Am. Chem. Soc., 2005,
127, 17572–17573.
15 S. Aime, D. Delli Castelli, F. Fedeli and E. Terreno, J. Am. Chem. Soc.,
2002, 124, 9364–9365.
16 B. Yoo, M. S. Raam, R. M. Rosenblum and M. D. Pagel, Contrast
Media Mol. Imaging, 2007, 2, 189–198.
17 N. Goffeney, J. W. M. Bulte, J. Duyn, L. H. Bryant and P. C. M. van
Zijl, J. Am. Chem. Soc., 2001, 123, 8628–8629.
18 S. Aime, D. Delli Castelli and E. Terreno, Angew. Chem., Int. Ed., 2003,
42, 4527–4529.
19 P. M. Winter, K. Cai, J. Chen, C. R. Adair, G. E. Kiefer, P. S. Athey, P. J.
Gaffney, C. E. Buff, J. D. Robertson, S. D. Caruthers, S. A. Wickline
and G. M. Lanza, Magn. Reson. Med., 2006, 56, 1384–1388.
20 S. Zhang, L. Michaudet, S. Burgess and A. D. Sherry, Angew. Chem.,
Int. Ed., 2002, 41, 1919–1921.
21 S. Aime, A. Barge, D. Delli Castelli, F. Fedeli, A. Mortillaro, F. U.
Nielsen and E. Terreno, Magn. Reson. Med., 2002, 47, 639–648.
22 E. Terreno, D. Delli Castelli, G. Cravotto, L. Milone and S. Aime,
Invest. Radiol., 2004, 39, 235–243.
23 D. Burdinski, J. Lub, J. A. Pikkemaat, H. Gru¨ll, L. Nieto Garrido, S. G.
Girard, D. Moreno Jalo´n and T. Weyhermu¨ller, JBIC, J. Biol. Inorg.
Chem., 2007, 17, S17–S17.
24 D. Burdinski, J. Lub, J. A. Pikkemaat, R. T. Wegh, H. Gru¨ll, L. Nieto
Garrido, S. G. Girard, S. Martial, C. Del Pozo Ochoa, D. Moreno
Jalo´n and T. Weyhermu¨ller, Proc. Int. Soc. Magn. Reson. Med., 2007,
15, 1182–1182.
25 R.-Y. Wang, J.-R. Lia, T.-Z. Jina, G.-X. Xua, Z.-Y. Zhoub and X.-G.
Zhou, Polyhedron, 1997, 16, 2037–2040.
26 J. Wang, Z.-R. Liu, X.-D. Zhang, W.-G. Jia and H.-F. Li, J. Mol. Struct.,
2003, 644, 29–36.
4150 | Dalton Trans., 2008, 4138–4151
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
27 A. Mondry and P. Starynowicz, New J. Chem., 2000, 24, 603–607.
28 R. Ruloff, P. Prokop, J. Sieler, E. Hoyer and L. Beyer, Z. Naturforsch.,
B: Chem. Sci., 1996, 51, 963–968.
29 M. A. Williams and H. Rapoport, J. Org. Chem., 1993, 58, 1151–1158.
30 S. Amin, J. R. Morrow, C. H. Lake and M. R. Churchill, Angew. Chem.,
Int. Ed. Engl., 1994, 33, 773–775.
31 S. Aime, A. Barge, M. Botta, A. S. De Sousa and D. Parker, Angew.
Chem., Int. Ed., 1998, 37, 2673–2675.
32 A. K. Covington, M. Paabo, R. A. Robinson and R. G. Bates, Anal.
Chem., 1968, 40, 700–706.
44 L. A. Clapp, C. J. Siddons, D. G. VanDerveer, J. H. Reibenspies, S. B.
Jones and R. D. Hancock, Dalton Trans., 2006, 2001–2007.
45 G. P. Connelly, Y. Bai, M.-F. Jeng and S. W. Englander, Proteins: Struct.,
Funct., Genet., 1993, 17, 87–92.
46 E. Zitha-Bovens, R. N. Muller, S. Laurent, L. V. Elst, C. F. G. C.
Geraldes, H. v. Bekkum and J. A. Peters, Helv. Chim. Acta, 2005, 88,
618–632.
47 B. Achour, J. Costa, R. Delgado, E. Garrigues, C. F. G. C. Geraldes,
N. Korber, F. Nepveu and M. I. Prata, Inorg. Chem., 1998, 37, 2729–
2740.
33 R. Lumry, E. Smith and R. Glantz, J. Am. Chem. Soc., 1951, 73, 4330–
4340.
48 R. Ruloff, K. Arnold, L. Beyer, F. Eitze, W. Gru¨nder, M. Wagner and
E. Hoyer, Z. Anorg. Allg. Chem., 1995, 621, 807–811.
49 R. Ruloff, T. Gelbrich, J. Sieler, E. Hoyer and L. Beyer, Z. Naturforsch.,
B: Chem. Sci., 1997, 52, 805–809.
34 A. Badinand, A. Boucherle and C. Charbonier, Bull. Soc. Chim. Fr.,
1960, 382–383.
35 J. L. Sudmeier and C. N. Reilley, Anal. Chem., 1964, 36, 1707–1712.
36 Y. Fujiwara and C. N. Reilley, Anal. Chem., 1968, 40, 890–894.
37 J. L. Sudmeier and C. N. Reilley, Anal. Chem., 1964, 36, 1698–1706.
38 C. F. G. C. Geraldes, R. Delgado, A. M. Urbano, J. Costa, F. Jasanada
and F. Nepveu, J. Chem. Soc., Dalton Trans., 1995, 327–335.
39 C. F. G. C. Geraldes, A. M. Urbano, M. C. Alpoim, A. D. Sherry,
K.-T. Kuan, R. Rajagopalan, F. Maton and R. N. Muller, Magn. Reson.
Imaging, 1995, 13, 401–420.
50 R. Ruloff, R. N. Muller, D. Pubanz and A. E. Merbach, Inorg. Chim.
Acta, 1998, 275–276, 15–23.
51 L. Przyborowski, Rocz. Chem., 1970, 44, 1883–1894.
52 L. A. Clapp, C. J. Siddons, J. R. Whitehead, D. G. VanDerveer, R. D.
Rogers, S. T. Griffin, S. B. Jones and R. D. Hancock, Inorg. Chem.,
2005, 44, 8495–8502.
53 F. Wang and L. M. Sayre, J. Am. Chem. Soc., 1992, 114, 248–255.
54 P. Paoletti, J. H. Stern and A. Vacca, J. Phys. Chem., 1965, 69, 3759–
3762.
55 E. Grunwald and E. K. Ralph, Acc. Chem. Res., 1971, 4, 107–113.
56 P. Paoletti, R. Barbucci, A. Vacca and A. Dei, J. Chem. Soc. A, 1971,
310–313.
40 L. Sarka, I. Ba´nyai, E. Bru¨cher, R. Kira´ly, J. Platzek, B. Radu¨chel and
H. Schmitt-Willich, J. Chem. Soc., Dalton Trans., 2000, 3699–3703.
´
41 Z. Ja´szbere´nyi, E. To´th, T. Ka´lai, R. Kira´ly, L. Burai, E. Bru¨cher, A. E.
Merbach and K. Hideg, Dalton Trans., 2005, 694–701.
42 D. H. White, L. A. deLearie, T. J. Dunn, E. N. Rizkalla, H. Imura and
G. R. Choppin, Invest. Radiol., 1991, 26, S229–S231.
57 D. Kroczewska, K. Bogusz, B. Kurzak and J. Jezierska, Polyhedron,
2002, 21, 295–303.
43 H. Imura, G. R. Choppin, W. P. Cacheris, L. A. de Learie, T. J. Dunn
and D. H. White, Inorg. Chim. Acta, 1997, 258, 227–236.
58 R. Barbucci, V. Barone, M. Micheloni and L. Rusconi, J. Phys. Chem.,
1981, 85, 64–68.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 4138–4151 | 4151
©