[TIIII(dota)]: An extraordinarily robust macrocyclic complex

Article abstract of DOI:10.1021/acs.inorgchem.5b00458

The X-ray structure of {C(NH₂)₃}[Tl(dota)]·H₂O shows that the Tl³⁺ ion is deeply buried in the macrocyclic cavity of the dota^4- ligand (1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetate) with average Tl-N and Tl-O distances of 2.464 and 2.365 ?, respectively. The metal ion is directly coordinated to the eight donor atoms of the ligand, which results in a twisted square antiprismatic (TSAP′) coordination around Tl³⁺. A multinuclear ¹H, ¹³C, and ²⁰⁵Tl NMR study combined with DFT calculations confirmed the TSAP′ structure of the complex in aqueous solution, which exists as the ??(????)/Δ(????) enantiomeric pair. ²⁰⁵Tl NMR spectroscopy allowed the protonation constant associated with the protonation of the complex according to [Tl(dota)]^- + H⁺ a?? [Tl(Hdota)] to be determined, which turned out to be pK^H_Tl(dota) = 1.4 ± 0.1. [Tl(dota)]^- does not react with Br^-, even when using an excess of the anion, but it forms a weak mixed complex with cyanide, [Tl(dota)]^- + CN^- a?? [Tl(dota)(CN)]^2-, with an equilibrium constant of K_mix = 6.0 ± 0.8. The dissociation of the [Tl(dota)]^- complex was determined by UV-vis spectrophotometry under acidic conditions using a large excess of Br^-, and it was found to follow proton-assisted kinetics and to take place very slowly (～10 days), even in 1 M HClO₄, with the estimated half-life of the process being in the 10⁹ h range at neutral pH. The solution dynamics of [Tl(dota)]^- were investigated using ¹³C NMR spectroscopy and DFT calculations. The ¹³C NMR spectra recorded at low temperature (272 K) point to C₄ symmetry of the complex in solution, which averages to C_4v as the temperature increases. This dynamic behavior was attributed to the A → δ(σσσσ) enantiomerization process, which involves both the inversion of the macrocyclic unit and the rotation of the pendant arms. According to our calculations, the arm-rotation process limits the δ → δ(σσσσ) interconversion.

Full text of DOI:10.1021/acs.inorgchem.5b00458

Article
pubs.acs.org/IC
[TlIII(dota)]−: An Extraordinarily Robust Macrocyclic Complex
Tamas Fodor,† Istvan Banyai,‡ Attila Benyei,§ Carlos Platas-Iglesias,∥ Mihaly Purgel,⊥ Gabor L. Horvath,†,#
́
́
́
́
́
́
́
Laszlo Zekany,† Gyula Tircso,† and Imre Toth*,†
́
́
́
́
́
́
†Department of Inorganic and Analytical Chemistry, ‡Department of Colloid and Environmental Chemistry, §Department of Physical
Chemistry, University of Debrecen, H-4032 Debrecen, Egyetem ter 1, Hungary
∥Departamento de Química Fundamental, Facultade de Ciencias, Universidade da Coruna, Campus da Zapateira-Rua da Fraga 10,
́
́
̃
15008 A Coruna, Spain
̃
⊥MTA-DE Homogeneous Catalysis and Reaction Mechanisms Research Group, H-4032 Debrecen, Egyetem ter
́
1, Hungary
#Institute of Innate Immunity, University Hospital, University of Bonn, 53127 Bonn, Germany
S
* Supporting Information
ABSTRACT: The X-ray structure of {C(NH2)3}[Tl(dota)]·H2O shows that the
Tl3+ ion is deeply buried in the macrocyclic cavity of the dota4− ligand (1,4,7,10-
tetraazacyclododecane-1,4,7,10-tetraacetate) with average Tl−N and Tl−O distances
of 2.464 and 2.365 Å, respectively. The metal ion is directly coordinated to the eight
donor atoms of the ligand, which results in a twisted square antiprismatic (TSAP′)
1
coordination around Tl3+. A multinuclear H, 13C, and 205Tl NMR study combined
with DFT calculations confirmed the TSAP′ structure of the complex in aqueous
solution, which exists as the Λ(λλλλ)/Δ(δδδδ) enantiomeric pair. 205Tl NMR
spectroscopy allowed the protonation constant associated with the protonation of
the complex according to [Tl(dota)]− + H+ ⇆ [Tl(Hdota)] to be determined,
which turned out to be pKHTl(dota) = 1.4 0.1. [Tl(dota)]− does not react with Br−,
even when using an excess of the anion, but it forms a weak mixed complex with
cyanide, [Tl(dota)]− + CN− ⇆ [Tl(dota)(CN)]2−, with an equilibrium constant of
Kmix = 6.0 0.8. The dissociation of the [Tl(dota)]− complex was determined by UV−vis spectrophotometry under acidic
conditions using a large excess of Br−, and it was found to follow proton-assisted kinetics and to take place very slowly (∼10
days), even in 1 M HClO4, with the estimated half-life of the process being in the 109 h range at neutral pH. The solution
dynamics of [Tl(dota)]− were investigated using 13C NMR spectroscopy and DFT calculations. The 13C NMR spectra recorded
at low temperature (272 K) point to C4 symmetry of the complex in solution, which averages to C4v as the temperature increases.
This dynamic behavior was attributed to the Λ(λλλλ) ↔ Δ(δδδδ) enantiomerization process, which involves both the inversion
of the macrocyclic unit and the rotation of the pendant arms. According to our calculations, the arm-rotation process limits the
Λ(λλλλ) ↔ Δ(δδδδ) interconversion.
complexes.14 While this is hardly a technical problem for stable
INTRODUCTION
■
isotopes, the long formulation time could be a crucial point for
The macrocyclic ligand 1,4,7,10-tetraazacyclododecane-
1,4,7,10-tetraacetate (dota4−) and its derivatives are known to
be the most versatile ligands for medical imaging applications.1
With a suitably sized cavity and four potentially coordinating
pendant arms, this macrocyclic ligand forms complexes with
many cations (i.e., Ln3+ ions,2 Sc3+,2 Y3+,3 Ga3+,4 Bi3+,5 Ca2+,6
short-lived radioisotopes.15
Very intensive research during the last 2 to 3 decades has
produced a trove of chemical knowledge about dota complexes,
i.e., the stability constants,12,14 formation and dissociation
kinetics,12−15 and structure in both the solid state2,10 and
solution5,16 have been explored for many metal ions. A
significant portion of the information that has allowed general
conclusions to be established regarding dota complexes has
been collected in lanthanide(III)−dota systems,16 particularly
the Gd3+ complex. This is related to the fact that [Gd(dota)]−
is a very good and safe general (nonselective) magnetic
resonance imaging (MRI) contrast agent (CA) widely used in
clinical practice under the trade name Dotarem.
8
Sr2+,7 Zn2+, and transition metal ions such as Fe3+,9 Co2+,4
Ni2+,10 and Cu2+,10). The complexes are usually thermodynami-
cally very stable11 and kinetically quite inert.12,13 These
behaviors together make dota a very safe metal-binder because
it can carry the metal component in biological fluids without
risk of decomplexation, which would result in uncontrolled
distribution of the components (i.e., metal ion and the ligand)
in the biological system and may cause toxic effects or a
decrease in the signal (e.g., relaxivity, luminescence, radio-
activity). The advantages of providing high stability and
inertness (often called kinetic stability in medical papers) are
partly counterbalanced by the slow formation of metal−dota
Received: February 26, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.5b00458
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
brought to a low boil, and SO2 was bubbled through for ∼30 min to
reduce all thallium to Tl+. (Excess SO2 was removed by further boiling
for ∼10 min.) The process for Tl+ determination was repeated with
The solution chemistry of the heaviest Group 13 element
thallium is not very vital nowadays. Indeed, thallium was
referred to as the forgotten element of the periodic table some
time ago in a Ph.D. thesis.17 Working with Tl3+ in aqueous
solution is not a trivial task, as the trivalent ion has a strong
tendency toward hydrolysis, so precipitation of Tl2O3 can occur
even in acidic solutions (at pH 2.5). Tl3+ is also a strong
oxidant, and organic ligands may reduce it to Tl+.18 The
trivalent ion Tl3+ forms very stable complexes with amino-
polycarboxylates such as edta (ethylendiaminetetraacetate),
with the stability constant of log K = 3719−22 being exceeded
only by the constant of [CoIII(edta)]− (log K = 40.7).23
The coordination sites of Tl3+ are not completely saturated
by the six donor atoms of the edta ligand and therefore quite
0.100 M KBrO3.28 This method yields the density, cTl , cTl(III), and cH
+
+
data for the stock solution. The data for the stock solution used for the
studies described herein were ρ = 1.72 g/cm3, cTl = 9.25 mM, cTl(III)
1.192 M, and cH = 3.96 M.
=
+
+
pH Measurements. pH values were measured by a Metrohm
6.0234.100 combined glass electrode connected to a Delta Ohm HD
8705 pH meter. The bridge electrolyte (3 M KCl) in the electrode was
replaced with 1 M NaCl to avoid KClO4 precipitation in the
membrane. The pH meter was calibrated with KH-phthalate (pH
4.005) and KH2PO4−Na2HPO4 (pH 6.865) buffers, and the H+
concentrations were calculated from the measured pH values by
applying the method proposed by Irving et al.29
Spectrophotometry. UV−vis spectra were recorded in quartz
cuvettes with a 1.0 mm path length using a Varian Cary 1E UV−visible
spectrophotometer at 25 °C in the wavelength range 350−200 nm,
with data points every 0.5 nm and a scan rate of 100 nm/min, unless
otherwise stated. Short-term dissociation kinetics were followed by
repeated recordings of spectra, with delay times of 1 h. The UV light
was cut off from the sample during the delays to avoid photochemical
decomposition. Long-term kinetics were followed by separate
recordings, with delay times of 1 day or 1 week.
stable [Tl(edta)X]2− ternary complexes with halide and
2−
pseudohalide anions have been detected (log Kmix[Tl(edta)X]
=
2.30, 3.50, 2.70, and 8.72, where X = Cl−, Br−, SCN−, and CN−,
respectively).24 The [TlIII(edta)]− complex is light-sensitive, as
UV light induces oxidative decarboxylation of the ligand and
formation of formaldehyde, CO2, and Tl(I).25 Recently,
[
201TlIII(dota)]− has been tested as a potential SPECT (single
photon emission computed tomography) agent.26 However,
the structure of the [TlIII(dota)]− complex and its thermody-
namic stability has not been reported so far. Our present article
summarizes new results on the equilibrium, kinetics, and
structure of the TlIII−dota system obtained with UV−vis and
1H, 13C, and 205Tl NMR spectroscopy, single-crystal X-ray
diffraction measurements, and DFT (density function theory)
methods.
1
NMR Measurements. H and 13C spectra were recorded with a
Bruker AM 400 spectrometer, at 400.1 and 100.6 MHz, respectively.
In spite of lacking matching NMR probes capable of 205Tl
measurement, thallium spectra were obtained with a Bruker AM 360
spectrometer by inserting a 500 MHz BB probe and tuning the x-
channel to the frequency of 205Tl (207.8 MHz). This setup does not
1
2
1
interact with H or H and does not enable locking or H-decoupling
to be used. Probe temperatures were kept at 25 ( 0.1) °C if not stated
otherwise. Calibration was performed using the signal of D2O (4.79
ppm) for H spectra, tetramethylsilane (0 ppm) for 13C spectra, and
1
50 mM Tl(I)- and Tl(III)-perchlorate solutions (Tl+: −4.72 ppm,
Tl3+: 2039 ppm) for 205Tl spectra. Line-shape analysis was carried out
with a homemade program written in MatLab, using the algorithm
developed by Reeves and Shaw.30 This program fits a signal-shape
function simulating the change of coupling-free NMR signals caused
by chemical exchange using modified Bloch equations (the coupling-
free approximation implies that the same exchange rate is applied to
both sets of signals). The relative positions of the two doublet signals
used for the analysis were fixed, and the integrals (populations) and
line widths within the doublets were assumed to be identical. Five
parameters were estimated: two to define the baseline, one for the
intensity, the frequency specific to the position of the signal, and the
exchange rate constant. The calculated rate constants were then fitted
EXPERIMENTAL SECTION
■
Materials and Preparation of Solutions. The stock solution of
Tl(ClO4)3 was prepared by anodic oxidation of TlClO4. The
electrolytic cell (see Supporting Information, Scheme S1) was filled
with 6 M perchloric acid, and bubbles were removed mechanically
from the cylinders and by opening the stopcock on the bridge to let
any air out. A portion of solid TlClO4 and a stirrer were added to the
anode compartment. Electrolysis took place between a fine platinum
mesh anode and a platinum plate cathode with an initial current
strength of 120 mA. This gradually dropped to ∼80 mA during the
process, which took a total of ∼51 h. The remaining portions of
TlClO4 were added as the anode compartment was slowly cleared of
solids. In our case, 21 g of TlClO4 was used, split into 7 g portions.
Solutions of [Tl(dota)]− were prepared by (a) adding small portions
of the Tl(ClO4)3 stock solution to a mildly basic (pH ∼ 8) solution of
dota and dropwise addition of ∼20% NaOH to neutralize the acid
(high acidity causes [TlH(dota)] to precipitate), (b) adding the
required amount (1 equiv) of dota in one portion to a solution of
Tl(OOCCH3)3 (prepared by mixing a portion of the Tl(ClO4)3 stock
solution (1 equiv) and a 10-fold excess of acetic acid/sodium acetate
buffer), and (c) following method (a) but using guanidine carbonate
instead of NaOH. The pH of the resulting solutions was set to ∼4 to
ensure that only [Tl(dota)]− was present (any excess of Tl(III)
precipitates as Tl2O3). Photochemical decomposition was avoided by
keeping the solutions in the dark.27
to the Eyring−Polan
́
yi equation [ln(k/T) = −ΔH⧧/RT + ln(kB/h) +
ΔS⧧/R, where k is the rate constant, T is the absolute temperature, kB
is the Boltzmann constant, R is the ideal gas constant, h is the Planck
constant, and ΔH⧧ and ΔS⧧ are the activation enthalpy and activation
entropy, respectively].
Crystal Structure Analysis. The sodium salt of [Tl(dota)]− could
not be crystallized in spite of repeated trials. Thus, we used the
guanidine cation, which is known to aid the preparation of single
crystals due to its strong ability to form hydrogen bonds.31 Single
crystals of {C(NH2)3}[Tl(dota)]·H2O could be grown at the interface
between 96% ethanol and a solution prepared by method (c)
described above.
Diffraction intensity data collection was carried out at 293(2) K on
a Bruker-Nonius MACH3 diffractometer equipped with a point
detector using graphite-monochromated Mo Kα radiation (λ =
0.71073 Å). The structure was solved with the SIR-92 program32
and refined by the full-matrix least-squares method on F2. All non-
hydrogen atoms were refined with anisotropic thermal parameters
using the SHELXL-97 package.33 Hydrogen atoms were located
geometrically and refined in the rigid mode, except for hydrogen
atoms of water molecules, which could be found in the difference
electron density map. The remaining electron density peaks were
found close to the Tl atom. Crystal data and structure refinement
Analysis. Analysis of the Tl(ClO4)3 stock solution was performed
as follows: A carefully measured volume of the stock solution (300 μL)
was pipetted into a previously weighed titration flask and weighed. An
excess of sodium chloride was added to avoid the hydrolysis of Tl(III),
the sample was diluted with 5 mL of distilled water, and the acid
content was determined by titration with 0.200 M NaOH, using
methyl red as an indicator. Two milliliters of concentrated hydro-
chloric acid was added to re-acidify the solution, and the concentration
of Tl+ was determined by titration with 0.0100 M KBrO3 at ∼95 °C. In
the absence of Tl+, methyl red is oxidized, turning the solution
colorless, which signals the end point of the titration. The solution was
B
DOI: 10.1021/acs.inorgchem.5b00458
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
details are as follows: formula: C17H32N7O9Tl; MW: 682.87; crystal
system: triclinic; space group: P1; a = 7.498(10) Å; b = 12.406(5) Å; c
̅
= 12.800(4) Å; α = 91.01(1)°; β = 90.86(4)°; γ = 107.31(6)°; V =
1136.3(16) Å3; Z = 2; Dcalc = 1.996 g cm−3; F(000) = 672; μ = 7.17
mm−1; reflections collected: 4909; unique reflections with I > 2σ(I):
4117; parameters refined: 331; GOF on F2 = 1.09; R[F2 > 2σ(F2)] =
0.057; Rint = 0.046; wR(F2) = 0.156; Δρmax and Δρmin = 3.58 and
−4.05 e Å−3.
DFT Calculations. All calculations were performed in aqueous
solution employing DFT within the hybrid GGA approximation with
the B3LYP exchange-correlation functional34 and the Gaussian 09
package.35 Full geometry optimizations of the [Tl(dota)]− system
were performed in aqueous solution without symmetry constraints. In
these calculations, we used the relativistic effective core potential
(RECP) of Ross et al. for Tl (CRENBL), which includes 68 electrons
in the core, with the valence space (5d, 6s, and 6p) represented by an
uncontracted (3s3p4d) basis set.36 The ligand atoms were described
using the standard 6-311+G(d,p) basis set. The stationary points
found on the potential energy surfaces as a result of geometry
optimizations were tested to represent energy minima rather than
saddle points via frequency analysis.
The relative free energies of the different conformations of
[Tl(dota)]− complexes were calculated in aqueous solution at the
B3LYP/CRENBL/6-311+G(d,p) level, and they include non-
potential-energy contributions (zero-point energies and thermal
terms) obtained through frequency analysis. The arm-rotation and
ring-inversion processes of [Tl(dota)]− were investigated by means of
the synchronous transit-guided quasi-Newton method at the B3LYP/
CRENBL/6-31+G(d,p) level.37 The nature of the saddle points (one
imaginary frequency) was characterized by frequency analysis. The free
energy barriers calculated include non-potential-energy contributions
obtained by frequency analysis.
Figure 1. View of the structure of the [Tl(dota)]− complex present in
crystals of {C(NH2)3}[Tl(dota)]·H2O. The ORTEP plot is at the 30%
probability level. Bond distances (Å): Tl01−N1, 2.454(8); Tl01−N4,
2.497(9); Tl1−N7, 2.443(8); Tl01−N10, 2.458(9); Tl01−O1,
2.322(8); Tl01−O11, 2.438(8); Tl01−O41, 2.383(8); and Tl01−
O71, 2.316(7).
of the structure of {C(NH2)3}[Tl(dota)]·H2O shows that
crystals contain the Λ(λλλλ)/Δ(δδδδ) enantiomeric pair, with
the two enantiomers being centro-symmetrically related in
accordance with the centro-symmetric character of space group
The NMR shielding tensors of the [Tl(dota)]− system were
calculated in aqueous solution using the B3LYP functional and the
GIAO method.38 In these calculations, the 6-311+G(d,p) basis set was
used for the ligand atoms. For 13C NMR chemical shift calculation
purposes, the NMR shielding tensors of TMS were calculated at the
same level.
P1. Both the guanidinium cations and the water molecules are
̅
involved in hydrogen-bonding interactions with the oxygen
atoms of the acetate groups of the ligand.
Dota-like complexes with the light Ln3+ ions are usually nine-
coordinate, where the ligand acts as eight-dentate and one
water molecule is in the inner coordination sphere. Eight
coordination with the absence of a coordinated water molecule
is typical for heavier Ln3+ ions such as Tm3+. In both subgroups,
SAP (Ln = Pr, Nd, Sm, Dy) and TSAP (Ln = Ce, Tm)
geometries are found. Eight-coordinate complexes that lack the
apical water molecule are often labeled as SAP′ and TSAP′ to
differentiate them from the nine-coordinate SAP and TSAP
complexes.2a [Sc(dota)]− adopts SAP′ geometry,2a in contrast
with the TSAP′ coordination observed for [Ca(dota)]2−,
[Bi(dota)]−, and [Tm(dota)]−.2a,5,6 In [Tl(dota)]−, the
coordination geometry around Tl3+ is TSAP′. The distinctive
structural feature that discriminates SAP and SAP′ versus TSAP
and TSAP′ structures is represented by the twist angle between
the two square planes of the antiprism, the basal plane
comprised of the four nitrogen atoms of the macrocycle, and
the upper plane containing the four coordinated oxygen atoms
of the pendant arms. The average twist angle observed for
[Tl(dota)]− (25.4°) is close to that expected for a regular
twisted square antitiprism (22.5°) and also similar to those
found for other complexes having TSAP or TSAP′ structures
(Table 1).
Throughout this work, solvent effects were included by using the
polarizable continuum model (PCM), in which the solute cavity is
built as an envelope of spheres centered on atoms or atomic groups
with appropriate radii. In particular, we used the integral equation
formalism (IEFPCM) variant, as implemented in Gaussian 09.39 The
default values for the integration grid (75 radial shells and 302 angular
points) and the SCF energy convergence criteria (10−8) were used in
all calculations.
RESULTS AND DISCUSSION
■
X-ray Crystal Structure. Crystals of {C(NH2)3}[Tl-
(dota)]·H2O contain the [Tl(dota)]− complex, a guanidinium
cation, and a water molecule. A view of the structure of the
complex and bond distances of the metal coordination
environment is provided in Figure 1. The Tl3+ ion is directly
coordinated to the eight donor atoms of the ligand, with the
metal coordination environment being best described as twisted
square antiprismatic (TSAP).
The stereochemistry of eight- and nine-coordinate metal
complexes of dota is well-documented; the coordination of the
ligand to the metal ion introduces two sources of chirality, one
associated with the conformation of the cyclen moiety [(δδδδ)
or (λλλλ)] and another related to the layout of the four acetate
pendant arms [represented as Δ or Λ]. The combination of
these two sources of helicity gives rise to four possible
stereoisomers existing as two diastereoisomeric pairs of
enantiomers. These enantiomeric pairs provide either a square
antiprismatic [SAP, Δ(λλλλ)/Λ(δδδδ)] or a TSAP [Λ(λλλλ)/
Δ(δδδδ)] coordination around the metal ion.16e,40 Inspection
Table 1 provides a comparison of structural parameters
reported in the literature for [M(dota)]− complexes (M =
5
Sc3+,2a Tm3+,2a and Bi3+, ) with those of the Tl3+ analogue.
Additionally, we have also included the corresponding data for
selected [M(dota)(H2O)]− complexes, with M = La42 or Gd9
as representative members of the earlier and middle members
of the lanthanide series, respectively. The average M−O bond
C
DOI: 10.1021/acs.inorgchem.5b00458
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Structural Parameters Observed for Selected [M(dota)(H2O)]− and [M(dota)]− Complexes in the Solid State
M
La
Gd
Bi
Sc
Tm
Tl
a
ionic radius/Å
1.216
2.467
2.770
0.89
1.107
2.368
2.663
0.89
1.17
0.87
0.994
2.279
2.529
0.90
0.98
b
M−O/Å
M−N/Å
2.538
2.526
1.00
2.150
2.441
0.88
2.365
2.464
0.96
M−O/M−N
c
M−QO/Å
M−QN/Å
0.728
1.810
2.54
0.715
1.632
2.35
1.112
1.434
2.55
1.007
1.327
2.33
1.064
1.466
2.53
1.252
1.324
2.58
d
QO−QN
e
twist angle/deg
23.2
38.5
25.9
41.1
24.4
25.4
isomer
TSAP
SAP
TSAP′
SAP′
TSAP′
TSAP′
a
b
Ionic radii for coordination number eight (Bi, Sc, Tm, and Tl) or nine (La and Gd) taken from ref 41. Average bond distances between the metal
c
ion and the O atoms of the acetate arms. Distance between the metal ion and the least-squares plane defined by the coordinated O atoms of acetate
d
e
groups. Distance between the metal ion and the least-squares plane defined by the N atoms of the macrocycle. Calculated as the average value of
the four torsion angles N−CN−CO−O, where N and O are the donor atoms defining a five-membered chelate ring and CN and CO represent the
centroids of the QN and QO planes, respectively.
lengths are in the range of 2.54−2.15 Å, whereas the M−N
distances range between 2.44 and 2.77 Å. The distances
between the virtual planes defined by the four N atoms of the
macrocycle (QN) and the four donor oxygen atoms of the
pendant arms (QO) are ca. 2.53−2.58 Å for the complexes with
TSAP and TSAP′ coordination and 2.33−2.35 Å for those with
SAP and SAP′ geometries. Thus, the QO−QN distances for a
given coordination polyhedron remain nearly unaffected by the
nature of the metal ion or the presence of a coordinated water
molecule,43 whereas the position of the metal ion with respect
to the QO and QN planes varies significantly. As expected, the
M−QO and M−QN distances are substantially different for
complexes having different coordination numbers (see, for
instance, the values for La, Gd, and Tm in Table 1).43 However,
Figure 2. Variation of the M−N and M−O bond distances with the
different M−QO and M−QN distances appear to be also related
to important changes in the ratio of the M−O and M−N
distances. The M−O/M−N ratio is very similar for complexes
with hard metal ions La3+, Gd3+, Tm3+, and Sc3+ (0.88−0.90),
and it increases to 0.96 and 1.00 for the Tl3+ and Bi3+
complexes, respectively. For the latter complex, the M−O
and M−N distances are nearly identical, whereas for the
complexes with La3+, Gd3+, Tm3+, and Sc3+, the M−N bond
distances are clearly longer than the M−O ones, and the
[Tl(dota)]− complex presents an intermediate situation. A plot
of the M−donor distances versus the ionic radii of the metal
ions for the [M(dota)]− (M = Tm, Sc, Bi, or Tl) and
[Ca(dota)]2− complexes (Figure 2) shows that both the M−N
and M−O distances vary linearly with the ionic radius for the
set of complexes with the hard metal ions Tm3+, Sc3+, and Ca2+,
whereas the Bi−O and Tl−O distances are longer, and the Bi−
N and Tl−N distances are shorter, to the values expected from
the linear trends. Thus, the soft nature of Tl3+ and Bi3+
according to the Pearson classification44 provokes a strengthen-
ing of the coordination bonds with the softer N donor atoms,
which results in a deeper penetration of the metal ion into the
macrocyclic cavity. This effect becomes obvious when
comparing the structural data reported in Table 1 for the
complexes with Tm3+ and Tl3+, which possess nearly identical
ionic radii.
ionic radius in [M(dota)]− (M = Tm, Sc, Bi or Tl) and [Ca(dota)]2−
complexes. The solid lines are linear fits of the data reported for the
Ca2+, Sc3+, and Tm3+ complexes (see text).
spectrum of [Tl(dota)]− shows a single signal at δ = 2218 ppm
with a chemical shift very similar to that measured earlier for
[Tl(edta)]− (δ = 2301 ppm).24 The positions of these signals
are within the range expected for a Tl(III) compound, whereas
the lack of a Tl(I) signal at around δ = 0 ppm indicates that
there was no reduction of Tl(III) to Tl(I) during the
preparation of [Tl(dota)]−.47 Furthermore, the absence of a
signal due to Tl3+ at δ = 2039 ppm confirms the full
aq
complexation of the metal ion by the ligand. The broad 205Tl
NMR signal observed at 298 K (w1/2 = 2200 Hz) did not show
multiplicity caused by 1H spin−spin coupling to the protons of
the dota ligand, which is likely related to the fluxional behavior
of the complex (see below). However, the multiplicity due to
1H−205Tl coupling could be observed in the spectrum recorded
at 289 MHz and 268 K, which is observed as a pseudononuplet
with an apparent 1H−205Tl coupling constant of ∼500 Hz. This
1
suggests that the H−205Tl coupling pattern in [Tl(dota)]− is
dominated by the coupling to eight equatorial ring protons of
the ligand.
205Tl NMR shifts are very sensitive to small changes in the
chemical environment around the Tl center and therefore can
be used to follow the protonation/deprotonation of the
[Tl(dota)]− complex by measuring the 205Tl chemical shift at
different pH values. Adding a strong acid to a solution of
[Tl(dota)]− indeed causes significant changes in the observed
205Tl NMR shift (Figure 3).
205Tl NMR Spectroscopy. The 205Tl nucleus has a nuclear
1
spin of I = /2 and very good sensitivity; therefore, 205Tl NMR
is quite useful to follow the formation of metal complexes in
aqueous solution.45 The 205Tl NMR chemical shift range is
huge and very sensitive to the oxidation state of the Tl ion in
the particular compound.46 The (proton coupled) 205Tl NMR
D
DOI: 10.1021/acs.inorgchem.5b00458
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(δ = 2218 ppm) was observed, whereas neither signals due to
any mixed complex nor the signal of [TlBr4]− at 1310 ppm that
would indicate complex dissociation could be detected.49
The formation of a mixed [Tl(dota)(CN)]2− complex could
be detected, however, using 205Tl NMR experiments on the
[Tl(dota)]−−CN− system. Addition of CN− to a solution of
[Tl(dota)]− resulted in the emergence of a new signal at δ =
2463 ppm (w1/2 ∼ 2500 Hz) whose chemical shift was almost
identical to that of the [Tl(edta)(CN)]2− complex (δ = 2460
ppm).24 The intensity ratio of the signals assigned to
[Tl(dota)]− and [Tl(dota)(CN)]2− allowed us to estimate a
stability constant of ∼10 for the mixed complex. However, in
order to obtain a more reliable equilibrium constant, pH-
dependent 205Tl NMR spectra were recorded. The pH
dependence of the intensity ratio of the two signals indicated
that the formation of the mixed complex was depressed by the
protonation of cyanide to form HCN (Figure 4). The following
simultaneous equilibria were considered:
Figure 3. 205Tl NMR shift of [Tl(dota)]− vs −log[H+]; measured data
points (red) and fitted curve (green).
Plotting the chemical shift values vs pH shows a typical S-
shaped curve with an inflection point at pH 1.4, which can be
attributed to the protonation of [Tl(dota)]− according to
[Tl(dota)]− + CN− ⇌ [Tl(dota)(CN)]2−
(3)
Kmix = [Tl(dota)CN2−]/[CN−][Tl(dota)−]
(4)
[Tl(dota)]− + H+ ⇌ [HTl(dota)]
(1)
H+ + CN− ⇌ HCN
(5)
KHTl(dota) = [HTl(dota)]/[H+][Tl(dota)−]
(2)
H
+
The analysis of the δ − cH curve provides pK Tl(dota) = 1.4
0.1 (δ[HTl(dota)] = 2405
3 ppm and δ[Tl(dota)] = 2216
1
ppm). The protonation process most likely involves the
protonation of one of the carboxylate arms. However, the
accuracy of this constant was limited by several conditions: (1)
it was not possible to keep the ionic medium constant, i.e., the
most acidic sample contained 3 M perchloric acid; (2) the
samples were not homogeneous. Formation of the charge-
neutral [Tl(Hdota)] species at 1.0 and 0.1 M acid
concentrations might be responsible for the precipitation of a
white solid from the solution of [Tl(dota)]− under acidic
conditions (see Experimental Section). Even so, the pK value
could be calculated from the δ205Tl vs pcH curve without
knowing the total concentration of the [Tl(dota)]− species. (3)
At higher acid concentrations (3 M) the solution became
homogeneous, which might be attributed to the formation of a
charged diprotonated [Tl(H2dota)]+ species, with the two
protonation steps being strongly overlapped. In any case, the
evaluated protonation constant can be considered to be a
reasonable estimate.
Figure 4. 207.1 MHz 205Tl NMR spectra of samples with cTl(dota) = 25
mM and cCN = 100 mM at different pH values (from top to bottom):
10.62, 10.09, 9.76, 9.58, 9.40, 9.21, and 9.00.
−
Using the mass balance equations for the components of the
equilibrium system and KHCN = 109.09 (I = 1 M),50 we obtained
Kmix = 6.0 0.8, a value that is 8 orders of magnitude lower
than that of [Tl(edta)(CN)2−] (Kmix = 108.7). This difference is
in line with a very tight wrapping of Tl3+ by the dota4− ligand. It
is worth mentioning that our attempts to measure the
The 205Tl NMR chemical shift of [Tl(dota)]− was constant
in the pH range of 4−11, indicating the lack of formation of the
[Tl(dota)(OH)]2− mixed complex. Considering the strong
tendency of Tl(III) to undergo hydrolysis, this can be taken as a
strong indication of a very tight coordination of the octadentate
dota4− ligand to the central Tl3+ ion. For instance, in the case of
[Tl(edta)]−, the formation of a ternary hydroxo complex
characterized by a pK = 6.0 has been reported.20,24,48 The
decreased reactivity of [Tl(dota)]− compared to that of
[Tl(edta)]− is also evidenced by the very limited tendency of
the former to form mixed complexes with bromide and cyanide
ions. In comparison, [Tl(edta)]− forms relatively stable mixed
205Tl−13C coupling constant (expected to be similarly as large
1
205Tl−13
as the J
= 10.5 kHz for [Tl(edta)(CN)]2−)24 by 205Tl
NMR using 13CC-enriched NaCN failed, which is likely related
to the fast cyanide exchange at the alkaline pH values required
for the formation of the mixed complex.
Equilibrium and Kinetic Study by UV−Vis Spectros-
copy. The stability constant of [Tl(edta)]− has been
determined by a competition method using a large excess of
both halide ligands and H+.20 In the case of [Tl(dota)]−, an
even greater stability is expected; furthermore, the macrocyclic
complex forms very slowly in acidic solutions, which prevents
the use of direct potentiometry for stability constant
determination. Instead, we have used a competition method
in which both a halide, namely bromide, and H+ compete for
complexes with the soft halide ions (i.e., KTl(edta)Br
=
[Tl(edta)Br2−]/[Tledta−] [Br−] = 103.5),24 whereas similar
mixed complexes could not be detected for the [Tl(dota)]−−
Br− system, even when using a large excess of halide over
[Tl(dota)]− (cTl(dota) = 5 mM, I = 1 M NaClO4, pH ∼ 3, cBr
=
0−154 mM). Only the signal of [Tl(dota)]− at a constant shift
E
DOI: 10.1021/acs.inorgchem.5b00458
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Tl3+ and dota4−, respectively, according to the following
other dota complexes). We invested a high amount of time and
energy trying to avoid this complication (exclusion of air from
samples with argon, increasing temperature to shorten the
equilibration time by conventional means and by using a
microwave reactor) without any success in measuring the
equilibrium constant defined by eq 7.
equilibrium characterized by the equilibrium constant, Kcomp
:
(n − 1)H+ + [HTl(dota)] + 4Br−
(4−n)−
⇄ TlBr4− + Hndota
(6)
Finally, we decided to only use spectra recorded at low
conversion to evaluate the kinetics of our Tl−macrocyclic
complex by means of the initial rate method. Assuming that
[Tl(dota)]− dissociates similarly to [Gd(dota)]−,12c the
following equations may be used to evaluate the kinetic data
(4−n)−
[TlBr4−][Hndota
]
Kcomp
=
[HTl(dota)][H+]n−1[Br−]4
(7)
Tl3+ behaves as a soft acid according to the classification of
Pearson, as the stability of halide complexes increases from the
light halide ions to the heavy ones. Therefore, iodide would be
the strongest competing agent for Tl3+. However, one has to
consider redox reactions, i.e., Tl3+ can oxidize iodide to iodine
while Tl+ is formed (TlI4− is the only TlIn+3−n species detected
in solution at high I−/ Tl3+ ratios).45 Preliminary experiments
have confirmed the immediate formation of iodine and
therefore we have selected Br− for the competition reactions.
Knowing the kinetic inertness of [M(dota)] complexes,
suitable thermodynamic data can be extracted only from
properly equilibrated competition systems. Thus, batch samples
containing 0.5 mM [Tl(dota)]− and 0.5 M NaBr were
prepared, varying the concentration of the acid in the range
of 0.1, 0.3, 0.5, 0.8, and 1 M HClO4 and keeping the total
−d[Tl(dota)]t /dt = kobs[HTl(dota)]
kobs = k0 + k1[H+] + k2[H+]2
(8)
(9)
t
where kobs is the observed pseudo-first-order reaction rate
constant, k0 is the rate constant of spontaneous (or water-
assisted) dissociation of the monoprotonated complex, and k1
and k2 are the constants of proton-assisted dissociation. The
experimental results are shown in Figure 6. Curve fitting gives
perchlorate concentration constant (at 1 M with (Na+
+
H+)ClO4 ). The dissociation of the complex, i.e., the formation
of [TlBr4]− was followed over time, recording spectra in the
200−350 nm range until the expected equilibration of the
system (Figure 5).
−
Figure 6. Plot of kobs vs acid concentration in the Tl(III)−dota4−−
H+−Br− system: measured data points (blue) and fitted curve (green).
k0 = 0 (fixed to 0 as the fitting of the dissociation kinetic data in
a trial returned a small negative value with large error), k1 = (9
7) × 10−7 s−1 M−1, and k2 = (5 0.8) × 10−6 s−1 M−2 values.
The data shows quadratic dependence of kobs on acidity with no
detectable spontaneous dissociation, which is not surprising
because spontaneous dissociation of the monoprotonated
complex is expected to occur slower by several orders of
magnitude than the acid-catalyzed dissociation (which is the
dominant reaction pathway in the acid concentration range
applied in the current study). The k1 value also has a large
uncertainty and may be a calculation artifact caused by the
limited number of data points and their inaccuracy in the lower
Figure 5. Time dependence of UV spectra attributed to the ligand-
exchange reaction between [HTl(dota)] and Br− in 1 M HClO4:
−
[HTl(dota)] (red, recorded at t0) is slowly converted to [Tl(Br)4 ]
(blue, recorded at 450 h; the absorbance value and the shape of the
spectrum are in accordance with separate measurements of [TlBr4]−).
Intermediate spectra were recorded at 1 h intervals for 19 h.
Conversion is ∼99% at 240 h (magenta).
+
cH range related to the above-mentioned side reactions during
the very slow competition reaction. This kind of rate equation
is in accordance with the related literature of dota complexes,
i.e., the dissociation is a dominantly proton-assisted process (we
have found some minor spectral changes that might be related
to some interaction between [HTl(dota)] and the Br− ion, but
we could not quantitatively measure this probably small
The spectra shown in Figure 5 clearly show isosbestic
point(s), which is in agreement with the presence of two
species ([HTl(dota)] and [Tl(Br)4 ]) in equilibrium with
−
measurable absorbance during the 240 h experimental time in
the system, in accordance with eq 6. However, the reaction is
substantially slower in the less acidic samples, where the
isosbestic points were not observed. This indicates that some
kind of parallel side reaction is occurring under these
conditions. This may be a slow redox reaction involving
bromide ions and the Tl3+ ion in an out-of-cage intermediate in
which the metal ion is coordinated only by carboxylate groups
(as in the formation/dissociation intermediate observed for
contribution to the dissociation kinetics). Using the kobs
=
k1[H+] + k2[H+]2 rate equation, one can calculate the half-life of
the complex at different pH values. Half-life of [Tl(dota)]− is
32 h in 1 M HClO4 and approximated to 5 × 109 h at pH 7.4.
The latter value indicates that the dissociation of [Tl(dota)]− at
physiological pH is negligible. To our surprise, however, the
[
201Tl(dota)]− complex was recently found to decompose in
F
DOI: 10.1021/acs.inorgchem.5b00458
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 7. 100 MHz 13C NMR spectra of a 0.1 M Na[Tl(dota)] solution in D2O recorded at different temperatures in D2O (pH 4).
vivo.26 Our data suggests that this phenomenon is more likely
to be caused by the bioreduction of Tl(III) in the samples
rather than the dissociation of the [TlIII(dota)]− complex. Tl+
(showing similarity to K+ both in size and complexation
properties) likely forms a weak and labile complex with dota,
making the dissociation of the reduced radioisotope from the
complex easy.
1
1H and 13C NMR Spectroscopy. The H NMR spectrum
of [Tl(dota)]− presents six broad signals at room temperature:
one doublet can be assigned to the protons of the acetate
methylene groups and two doublets to those of the ring
3
1
205
( J H− Tl scalar coupling splits these signals into doublets). The
broadness of these signals is related to exchange processes
occurring in solution. On the basis of the composition of the
complex and that of the solution, the exchange is attributed to
the intramolecular rearrangement of the complex. At higher
temperatures, the signals sharpen and their structure becomes
visible as a consequence of the acceleration of the exchange
processes. This is also supported by 13C NMR spectra, which at
high temperatures show two signals in the aliphatic region: a
singlet assignable to the acetate methylene carbons and a 205Tl
coupled doublet of the ring carbons (Figure 7). The carbon
nuclei of the methylene groups of the pendant arms do not
show 205Tl coupling. This is likely caused by the combined
effect of a two-bond coupling (through the nitrogen) and a
three-bond coupling (through the carboxylate oxygen). These
couplings may possess differing signs that presumably cancel
each other.
1
Figure 8. 400 MHz H−1H COSY spectrum of 0.1 M [Tl(dota)]−
recorded in D2O solution (pDreading = 4) at 272.5 K (see Chart 1 for
labeling).
Chart 1. Ligand dota4− and the Numbering Scheme Used for
NMR Spectral Assignment
1
The H and 13C NMR spectra recorded at low temperature
(272 K) are relatively well-resolved and could be fully assigned
with the aid of the homonuclear 1H−1H COSY and
1
heteronuclear H−13C HSQC spectra (Figure 8; see Chart 1
for labeling). At this temperature, the ring carbon doublet splits
into two, indicating the presence of two different chemical
G
DOI: 10.1021/acs.inorgchem.5b00458
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 2. NMR Spectral Data of [Tl(dota)]−
b
c
3
1H
δ/ppm
J H− Tl/Hz
13C
δ/ppm
2J
Tl/Hz
δcald
δcald
1
205
13C−205
TSAP
SAP
178.2
66.7
H2ax
H2eq
H3ax
H3eq
H4ax
H4eq
3.19/3.16
11
C1
C2
C3
C4
179.7
58.8
53.6
49.2
64
179.7
59.2
51.8
48.4
d
a
3.68 /2.24
574
12
2.55/2.52
161
156
56.9
e
3.32/2.23
441
<5
55.2
2.91
f
3.24/2.00
499
a
b
c
Not observed. Chemical shifts of the TSAP′ isomer calculated at the B3LYP/CRENBL/6-311+G(d,p) level using the GIAO method. Chemical
d
e
f
2
2
shifts of the SAP′ isomer calculated at the B3LYP/CRENBL/6-311+G(d,p) level using the GIAO method. J2eq−2ax = 13.5 Hz. J3eq−3ax = 14.5 Hz
2J4eq−4ax = 14.0 Hz.
environments for the carbon nuclei of the macrocyclic moiety.
The spectrum shows four signals: the 205Tl-coupled doublet
assignments given in Table 2. An inspection of the X-ray
structure of [Tl(dota)]− shows that the equatorial protons
provide average Tl−N−C−Heq dihedral angles of 160.4−
162.9°, whereas the corresponding values for axial protons Tl−
N−C−Hax take values of ca. 80−82°. These results indicate that
(2J
C = 64 Hz) of the carboxylate (−CO) groups at δ =
205Tl−13
179.7 ppm (C1), a singlet due to the −CH2− groups of the
acetate pendant arms at 58.8 ppm (C2), and two 205Tl-coupled
doublets of the nonequivalent carbon atoms of the macrocycle
3
1
205Tl
the J H−
coupling constants follow the empirical relation-
ship of Karplus (3J = A − B cos ϕ + C cos 2ϕ, where ϕ is the
dihedral angle and A, B, and C are empirical constants). A
similar Karplus-like dependence was also observed for the spin
density distributions in the paramagnetic [Gd(dota)(H2O)]−
complex and related systems.53
Solution Structure and Dynamics. The multinuclear
NMR study described in the previous sections indicates an
octadentate binding of the ligand to the Tl3+ ion in
[Tl(dota)]−. Furthermore, the X-ray structure suggests that
this complex might adopt a TSAP′ structure in solution. In
order to confirm this hypothesis, we performed theoretical
calculations in aqueous solution at the B3LYP/CRENBL/6-
311+G(d,p) level. Our calculations provide two energy minima
that correspond to the expected SAP′ and TSAP′ isomers of
the complex (Figure 9). A comparison of the experimental (X-
ring centered at δ = 49.2 (2J
= 156 Hz) and 53.6
205Tl−13
C
(2J
C = 161 Hz) ppm. This pattern points to C4 symmetry
205Tl−13
of the [Tl(dota)]− complex at low temperatures, whereas the
spectra at temperatures higher than 320 K can be interpreted as
an effective C4v symmetry. The 1H−13C HSQC spectrum
(Figure S2, Supporting Information) shows that each of the 3
carbon nuclei in the aliphatic region is coupled with 4 proton
signals, with two of them showing very small chemical shift
differences. This is clearly observed for the proton nuclei of the
−CH2− groups of the acetate arms (H2). For these protons,
2
one expects an AB spin system with a Jax−eq of ∼16 Hz.
However, the coordination of the ligand to the Tl3+ ion creates
an AB−X system, which splits each of the components of the
3
1
205Tl
AB spin systems into doublets due to the J H−
coupling.
1
The H−1H COSY spectrum relates the signals at 3.19 and
3.68 ppm and those at 3.16 and 2.24 ppm, while no cross-peak
is observed relating the signals at 3.68 and 2.24 ppm. All four of
these signals are, however, correlated to the carbon atom of the
acetate arms in the HSQC spectrum. Thus, the signals at 3.68
and 2.24 ppm are attributed to a single proton nucleus whose
resonance is split by the coupling to 205Tl. Analogous reasoning
allowed us to assign the proton signals of the macrocyclic
fragment (Table 2). Furthermore, the spin pattern of the
signals allowed us to discriminate between axial (ax) and
equatorial (eq) proton nuclei. Indeed, the different H−C−C−
H dihedral angles relating axial and equatorial protons with the
protons at a three-bond distance result in different coupling
patterns, as expected by the Karplus relationship.51,52 As a
result, the equatorial protons provide small 3J coupling
constants, so their coupling patterns are dominated by the
strong 2Jeq−ax coupling (∼16 Hz). This is, for instance, the case
of the signal at 2.00 ppm, which is observed as a pseudodoublet.
On the contrary, axial protons give two strong couplings (2Jax−eq
Figure 9. Geometries of the SAP′ and TSAP′ isomers of [Tl(dota)]−
obtained from geometry optimizations in aqueous solution at the
B3LYP/CRENBL/6-311+G(d,p) level. Hydrogen atoms are omitted
for simplicity. Bond distances: Tl−N = 2.537 (TSAP′) and 2.528 Å
(SAP′); Tl−O: 2.354 (TSAP′) and 2.341 Å (SAP′).
3
and Jax−ax) and a weak one (3Jax−eq), and they are observed as
pseudotriplets (see, for instance, the signal at 2.91 ppm, Figure
8).
ray) and calculated structure of the TSAP′ isomer indicates that
our calculations provide molecular geometries in good
agreement with the reference crystallographic data (Table 1).
The main discrepancy between the experimental and calculated
structures is the overestimation of the Tl−N distance by ca.
0.08 Å in the latter.
According to our calculations, the TSAP′ isomer is more
stable than the SAP′ one by 15.2 kJ mol−1, which is in line with
the structure observed in the solid state. In order to confirm
that [Tl(dota)]− adopts a TSAP′ structure in solution, we
3
1
205Tl
The J H−
coupling constants obtained from the analysis
of the NMR spectra (Table 2) indicate that the equatorial
protons provide coupling constants of ca. 440−575 Hz,
whereas the coupling constants involving axial protons are
rather small (<15 Hz). These coupling constants are very
similar to the average coupling constant estimated from the
splitting of the 205Tl NMR spectrum recorded at low
temperature (∼500 Hz, see above), which confirms the
H
DOI: 10.1021/acs.inorgchem.5b00458
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
calculated the 13C NMR chemical shifts of the SAP′ and TSAP′
isomers at the B3LYP/CRENBL/6-311+G(d,p) level. It has
been shown that the magnetic shieldings calculated with the use
of ECPs are not gauge invariant, which means that magnetic
shieldings calculated with different origins are different.
However, the errors introduced by this violation of the gauge
invariance have been shown to amount to a few parts per
million and therefore 13C NMR shifts can be calculated by this
method with reasonably good accuracy.54,55 The 13C NMR
shifts calculated for the TSAP′ isomer are in very good
agreement with the experimental data, with absolute deviations
< 1.8 ppm, whereas much larger deviations between the
experimental and calculated data are obtained for the SAP′
isomer (up to 6.8 ppm, see Table 2). Thus, our calculations
confirm that the [Tl(dota)]− presents a TSAP′ structure in
aqueous solution.
arms [Λ ↔ Δ], as described previously in detail for Ln3+ dota-
like complexes.56 Since the temperature dependence of the
signals at 53.6 and 49.2 ppm covers both the fast and slow
exchange regimes, we used these resonances to perform a
quantitative band-shape analysis of the exchange process, which
provides the exchange rates kex at every temperature. The
agreement between the experimental and simulated spectra is
excellent (Figure 10).
The temperature dependence of the evaluated time constants
was subsequently used to obtain the activation parameters for
the exchange process using the Eyring−Polanyi equation
́
(Figure S3, Supporting Information). The results are compared
to those obtained for related dota4− complexes in Table 3.
The activation enthalpy (ΔH⧧) determined for [Tl(dota)]−
falls between the values reported for [Bi(dota)]− and
[Lu(dota)]−. However, the different signs of the activation
entropy obtained for the latter two complexes and the small
positive ΔS⧧ value determined for [Tl(dota)]− result in
relatively similar activation free energies at 298 K (ΔG⧧298).
The [Lu(dota)]− complex exists in solution as a mixture of the
eight-coordinate TSAP′ isomer and the nine-coordinate SAP
form, and activation parameters for both the ring-inversion and
arm-rotation processes could be obtained.2b Thus, the large
positive activation energy obtained for this complex might be
related to the expulsion of the coordinated water molecule in
the transition state responsible for the Λ(λλλλ) ↔ Δ(δδδδ)
interconversion. For [Tl(dota)]−, ΔS⧧ takes a relatively small
positive value, which is likely related to the rearrangement of
the hydration shell during the interconversion process.
The increase of the symmetry observed in the 13C NMR
spectra upon increasing temperature can be explained by the
higher rate of the chemical exchange between the two different
chemical environments of the ring carbon atoms. Indeed, the
doublet signals due to the carbon nuclei of the macrocyclic
fragment are clearly separated at 272 K, indicating a slow
exchange regime (Figure 7). The signals gradually broaden as
the temperature increases, and they are hardly visible in the
spectra recorded in the temperature range 300−320 K. The
temperature of coalescence is ca. 315 K. The time-averaged
signal (a Tl-coupled doublet) sharpens above 325 K, as
expected for a two-site exchange system being in the fast
exchange regime. It is worth noting, however, that the other
two signals, i.e., the signals of the acetate arms, behave
differently from the ring carbon signals. The half-widths for
both signals are changing with temperature. However, the
signal of aliphatic carbon atoms of the acetate arms becomes
sharper with increasing temperature, which is typical of systems
in fast exchange on the actual NMR time scale. One would
expect splitting of the acetate peak into two signals below 272
K due to a slow isomerization process between two species with
lower symmetry, as observed earlier for LnDOTA-like
complexes. However, we could not cool the sample without
freezing in order to find experimental evidence for the
hypothesis. Thus, we could not measure the chemical shift
difference for the two sites, and the actual NMR time scale for
those nuclei is not known. The doublet signal of the carbonyl
carbon atoms changes similarly to the signal of the aliphatic
carbon atoms at low temperatures (i.e., sharpening in the range
of 272−300 K), but above 300 K, substantial broadening can be
observed. The actual NMR time scales of the signals due to the
To obtain a more detailed information on the mechanism
responsible for the enantiomerization process in [Tl(dota)]−,
we performed a computational investigation of the ring-
inversion and arm-rotation processes. According to our
calculations performed in aqueous solution at the B3LYP/
CRENBL/6-31+G(d,p) level, the inversion of the macrocyclic
ring is a four-step process involving the stepwise inversion of
each of the four five-membered chelate rings formed by the
coordination of the macrocyclic moiety. In each of these steps,
one of the chelate rings changes its configuration from δ to λ
(or vice versa) through a transition state in which the H−C−
C−H units adopt a nearly eclipsed conformation. Similar four-
step processes have been obtained previously using HF and
DFT calculations for different Ln3+ complexes with cyclen-
based ligands.57 Due to the symmetry properties of the
[Tl(dota)]− complex (C4 point group), two different routes are
possible for the inversion of the second and third chelate rings
(Figure 11), as once the first chelate is inverted the second step
may proceed via inversion of a chelate ring either in opposite
(TS2a and TS3a) or adjacent (TS2b and TS3b) position at the
macrocyclic ring with respect to the former. According to our
calculations, the lowest-energy pathway corresponds to that
proceeding through TS3b, which provides an activation Gibbs
free energy for the ring-inversion process of ΔG⧧ = 60.8 kJ
mol−1 at 298 K.
pendant arms are likely defined by the spin−spin coupling
3
1
205
constants J H− Tl. The signal due to the carbonyl groups is in
the slow exchange regime at low temperatures and thus it
broadens as the temperature increases. The signal due to the
−CH2− carbon nuclei of the pendant arms is, however, in the
fast exchange regime at low temperatures, presumably due to a
3
1
205
Tl
smaller J H−
coupling.
The [Tl(dota)]− complex exists in solution as a racemic
mixture of the TSAP′ Λ(λλλλ)/Δ(δδδδ) enantiomeric pair,
whereas no significant populations of the SAP′ isomers could
be detected by NMR spectroscopy. Thus, the exchange process
revealed by the 13C NMR data can be attributed to a Λ(λλλλ)
↔ Δ(δδδδ) interconversion, which represents a simple case of
a two-site exchange between equally populated sites. This
process involves both the inversion of the macrocyclic unit
[(λλλλ) ↔ (δδδδ)] and the rotation of the four acetate pendant
Contrary to the ring-inversion process, the arm-rotation
pathway is a single-step process involving the simultaneous
rotation of the four pendant arms of the ligand. The structure
of the TS (TS5, Figure 11) is quite distorted, with four different
Tl−N distances in the range 2.50−2.75 Å and four Tl−O
distances ranging between 2.30 and 2.37 Å. The activation free
energy for the arm-rotation process amounts to ΔG⧧ = 67.5
298
kJ mol−1 and thus the arm-rotation process likely represents the
rate-determining step for the Λ(λλλλ) ↔ Δ(δδδδ) inter-
I
DOI: 10.1021/acs.inorgchem.5b00458
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 3. Activation Parameters Obtained for the Λ(λλλλ) ↔
Δ(δδδδ) Interconversion in [Tl(dota)]− and Related
Systems
ΔH⧧/
ΔS⧧/
ΔG⧧
/
298
kJ mol−1
J K−1 mol−1
kJ mol−1
[Tl(dota)]−
exp
66
2
+23
6
59
2
a
calcd
exp
73.2
40
+19.2
67.5
63
b
[Bi(dota)]−
−76 15
+116
c
[Lu(dota)]−
exp
101
66
a
Values are obtained for the arm-rotation process with DFT
calculations in aqueous solution at the B3LYP/CRENBL/6-31+G-
b
c
(d,p) level. From ref 5. From ref 2b.
Figure 11. Relative free energies of minima, intermediates (labeled I)
and transition states (labeled TS) involved in the Λ(λλλλ) ↔ Δ(δδδδ)
interconversion process of [Tl(dota)]− calculated in aqueous solution
at the B3LYP/CRENBL/6-31+G(d,p) level.
CONCLUSIONS
■
The X-ray and spectroscopic study presented in this work
points to a very compact structure of the [Tl(dota)]− complex,
which presents TSAP′ coordination around the metal ion. This
complex does not react with Br− or OH− anions, as observed
for the [Tl(edta)]− analogue, and forms a weak mixed complex
with cyanide. The [Tl(dota)]− complex is very inert, the
dissociation is a proton-assisted process, and the half-life of
dissociation is on the order of 109 h at neutral pH. The 1H, 13C,
205
3
1
205Tl
and Tl NMR spectra provided J H−
spin−spin coupling
constants of up to ∼575 Hz and 2J
Tl values of ca. 60−160
13C−205
Hz. The large values of these coupling constants reflect an
important degree of covalency of the Tl−ligand bonds, which is
probably responsible, at least in part, for the robustness of the
complex. A detailed study of the dynamic behavior of
[Tl(dota)]− in solution indicates that the rotation of the
pendant arms of the ligand most likely represents the rate-
determining step of the Λ(λλλλ) ↔ Δ(δδδδ) interconversion.
ASSOCIATED CONTENT
■
S
* Supporting Information
Figure S1: Scheme of the Tl(ClO4)3 electrolysis device. Figure
1
S2: 400 MHz H−13C HSQC spectrum. Figure S3: Modified
Figure 10. Experimental (left) and simulated (right) 13C NMR spectra
Eyring diagram. Figure S4: 1H NMR spectra of 0.1 M
[Tl(dota)]− at varying temperatures. Figure S5: 288.5 MHz
205Tl NMR spectrum of [Tl(dota)]− at 268 K. Figure S6:
Packing diagram of {C(NH2)3}[Tl(dota)]·H2O, view normal
to (100). Table S1: Cartesian coordinates of the intermediates
and transition states of the [Tl(dota)]− system optimized at the
B3LYP/CRENBL/6-31+G(d,p) level using the polarizable
continuum model (water solvent). Crystal structure analysis.
of [Tl(dota)]− at varying temperatures.
conversion in [Tl(dota)]−. The ΔH⧧ and ΔS⧧ values calculated
for the arm-rotation pathway (Table 3) show a very good
agreement with the experimental values, which provides
additional support for this hypothesis. In contrast, the ring-
inversion path provides negligible activation entropy.
J
DOI: 10.1021/acs.inorgchem.5b00458
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
V.; Havlickova, J.; Kotek, J.; Sabatie-Gogova, A.; Plutnar, J.; Huclier-
Markai, S.; Hermann, P. Chem.Eur. J. 2014, 20, 7944−7955.
(14) (a) Wang, X.; Jin, T.; Comblin, V.; Lopez-Mut, A.; Merciny, E.;
Desreux, J. F. Inorg. Chem. 1992, 31, 1095−1099. (b) Wu, S. L.;
Horrocks, W., Jr. Inorg. Chem. 1995, 34, 3724−3732. (c) Moreau, J.;
Guillon, E.; Pierrard, J.-C.; Rimbault, J.; Port, M.; Aplincourt, M.
Chem.Eur. J. 2004, 10, 5218−5232.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.5b00458.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: imre.toth@science.unideb.hu.
(15) Martins, A. F.; Prata, M. I. M.; Rodrigues, S. P. J.; Geraldes, C.;
Riss, P. J.; Amor-Coarasa, A.; Burchardt, C.; Kroll, C.; Rosch, F.
Contrast Media Mol. Imaging 2013, 8, 265−273.
̈
Notes
The authors declare no competing financial interest.
(16) (a) Benazeth, S.; Purans, J.; Chalbot, M.-C.; Nguyen-van-
Duong, M. K.; Nicolas, L.; Keller, F.; Gaudemer, A. Inorg. Chem. 1998,
37, 3667−3674. (b) Aime, S.; Botta, M.; Ermondi, G. Inorg. Chem.
1992, 31, 4291−4299. (c) Hoeft, S.; Roth, K. Chem. Ber. 1993, 126,
869−873. (d) Fusaro, L.; Luhmer, M. Inorg. Chem. 2014, 53, 8717−
8722. (e) Aime, S.; Botta, M.; Fasano, M.; Marques, M. P. M.;
Geraldes, C. F. G. C.; Purbanz, D.; Merbach, A. E. Inorg. Chem. 1997,
36, 2059−2068. (f) Mayer, F.; Platas-Iglesias, C.; Helm, L.; Peters, J.
A.; Djanashvili, K. Inorg. Chem. 2012, 51, 170−178. (g) Esteban-
Gomez, D.; de Blas, A.; Rodriguez-Blas, T.; Helm, L.; Platas-Iglesias,
C. ChemPhysChem 2012, 13, 3640−3650. (h) Powell, D. H.; Ni
Dhubhghaill, O. M.; Pubanz, D.; Helm, L.; Lebedev, Y. S.; Schlaepfer,
W.; Merbach, A. E. J. Am. Chem. Soc. 1996, 118, 9333−9346.
ACKNOWLEDGMENTS
The research was supported by the EU and cofinanced by the
European Social Fund under the project ENVIKUT (TAMOP-
■
́
4.2.2.A-11/1/KONV-2012-0043). The authors also acknowl-
edge the support of the Hungarian Scientific Research Fund
(OTKA K-109029 and K-84291). This work was partially
supported by the European Union and the European Social
Fund through project Supercomputer, the national virtual lab
́
(grant no. TAMOP-4.2.2.C-11/1/KONV- 2012-0010) (M.P.).
This research was performed within the framework of the EU
COST Action TD1004 “Theragnostics Imaging and Therapy:
An Action to Develop Novel Nanosized Systems for Imaging-
Guided Drug Delivery”. C.P.-I. thanks Centro de Super-
(17) Musso, S. Doctoral and Habilitation Theses, ETH Zurich,
̈
Switzerland, 1993; http://dx.doi.org/10.3929/ethz-a-000695797.
́ ̋
(18) Toth, I.; Gyori, B. Thallium: Inorganic Chemistry; In
computacion
facilities.
́
de Galicia (CESGA) for providing the computer
Encyclopedia of Inorganic Chemistry, 2nd ed.; King, R. B., Ed.; Wiley:
Hoboken, NJ, 2005; Vol. 8, pp 4824−4835.
(19) Anderegg, G. Critical Survey of Stability Constants of EDTA
Complexes, IUPAC Chemical Data Series 14; Pergamon Press: Oxford,
England, 1977.
REFERENCES
■
(1) (a) Stasiuk, G. J.; Long, N. J. Chem. Commun. 2013, 49, 2732−
2746. (b) Liu, S. Adv. Drug Delivery Rev. 2008, 60, 1347−1370.
(c) Lattuada, L.; Barge, A.; Cravotto, G.; Giovenzana, G. B.; Tei, L.
Chem. Soc. Rev. 2011, 40, 3019−3049. (d) Viswanathan, S.; Kovacs, Z.;
Green, K. N.; Ratnakar, S. J.; Sherry, A. D. Chem. Rev. 2010, 110,
2960−3018. (e) Tanaka, K.; Fukase, K. Org. Biomol. Chem. 2008, 6,
815−828.
(20) Toth, I.; Brucher, E.; Zekany, L.; Veksin, V. Polyhedron 1989, 8,
2057−2064.
(21) Bottari, E.; Anderegg, G. Helv. Chim. Acta 1967, 50, 2341−2349.
(22) Dozsa, L.; Szabo, A.; Beck, M. T. Acta Chim. Acad. Sci. Hung.
1971, 67, 189−194.
(23) (a) Reilley, C. N.; Scribner, W. G.; Temple, C. Anal. Chem.
1956, 28, 450−454. (b) Tanaka, N.; Ogino, H. Bull. Chem. Soc. Jpn.
1965, 38, 1054.
(2) (a) Benetollo, F.; Bombieri, G.; Calabi, L.; Aime, S.; Botta, M.
Inorg. Chem. 2003, 42, 148−157. (b) Aime, S.; Barge, A.; Botta, M.;
Fasano, M.; Ayala, J. D.; Bombieri, G. Inorg. Chim. Acta 1996, 246,
423−429.
(24) Blixt, J.; Glaser, J.; Solymosi, P.; Toth, I. Inorg. Chem. 1992, 31,
5288−5297.
́
(25) Toth, I.; Reiter, I.; Brucher, E.; Vince, L. In Abstracts of XIX
̈
(3) Parker, D.; Pulukkody, K.; Smith, F. C.; Batsanov, A.; Howard, J.
A. K. J. Chem. Soc., Dalton Trans. 1994, 689−693.
International Conference on Solution Chemistry, Lund, August 15−
18, 1988; p 109, ISBN 91-7970-299-6.
́
(4) Heppeler, A.; Andre, J. P.; Buschmann, I.; Wang, X.; Reubi, J.-C.;
(26) Hijnen, N. M.; de Vries, A.; Blange, R.; Burdinski, D.; Grull, H.
̈
Henning, M.; Kaden, T. A.; Maecke, H. R. Chem.Eur. J. 2008, 14,
Nucl. Med. Biol. 2011, 38, 585−592.
3026−3034.
(27) Horvat, H. Investigation of the decomposition of the [NaTl(edta)]-
complex. Thesis, University of Debrecen, Debrecen, Hungary, 2010;
https://dea.lib.unideb.hu/dea/handle/2437/101064.
(28) Glaser, J. Ph.D. Thesis, The Royal Institute of Technology
(KTH), Stockholm, 1981; p 15.
(5) Csajbok, E.; Baranyai, Z.; Banyai, I.; E. Brucher, E.; Kiraly, R.;
̈
Muller-Fahrnow, A.; Platzek, J.; Raduchel, B.; Schafer, M. Inorg. Chem.
̈
̈
̈
2003, 42, 2342−2349.
(6) Anderson, O. P.; Reibenspies, J. H. Acta Crystallogr. 1996, C52,
792−795.
(29) Irving, H. M.; Miles, M. G.; Pettit, L. D. Anal. Chim. Acta 1967,
38, 475−488.
́
(7) Burai, L.; Toth, E.; Moreau, G.; Sour, A.; Scopelliti, R.; Merbach,
A. E. Chem.Eur. J. 2003, 9, 1394−1404.
(30) Reeves, L. W.; Shaw, K. N. Can. J. Chem. 1970, 48, 3641−3643.
(8) Riesen, A.; Zehnder, M.; Kaden, T. A. Acta Crystallogr. 1991,
C47, 531−533.
́
́ ́ ́ ́
A.; Benyei, A. C.; Joo, F.; Sagi, M. Adv. Synth. Catal.
(31) Katho,
2002, 344, 278−282.
(9) Chang, C. A.; Francesconi, L. C.; Malley, M. F.; Kumar, K.;
Gougoutas, J. Z.; Tweedle, M. F. Inorg. Chem. 1993, 32, 3501−3508.
(10) Riesen, A.; Zehnder, M.; Kaden, T. A. Helv. Chim. Acta 1986,
69, 2067−2073.
(32) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A. J.
Appl. Crystallogr. 1993, 26, 343−350.
(33) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112−122.
(34) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652. (b) Lee,
C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789.
(11) Viola-Villegas, N.; Doyle, R. P. Coord. Chem. Rev. 2009, 253,
1906−1925.
(35) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
(12) (a) Kubicek, V.; Havlickova, J.; Kotek, J.; Tircso, G.; Hermann,
P.; Toth, E.; Lukes, I. Inorg. Chem. 2010, 49, 10960−10969.
(b) Drahos, B.; Kubicek, V.; Bonnet, C. S.; Hermann, P.; Lukes, I.;
Toth, E. Dalton Trans. 2011, 40, 1945−1951. (c) Toth, E.; Brucher, E.;
Lazar, I.; Toth, I. Inorg. Chem. 1994, 33, 4070−4076. (d) Brucher, E.;
Laurenczy, G.; Makra, Z. Inorg. Chim. Acta 1987, 139, 141−142.
(13) (a) Voracova, I.; Vanek, J.; Pasulka, J.; Strelcova, Z.; Lubal, P.;
Hermann, P. Polyhedron 2013, 61, 99−104. (b) Pniok, M.; Kubicek,
K
DOI: 10.1021/acs.inorgchem.5b00458
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
revision A.01; Gaussian, Inc.: Wallingford, CT, 2009.
(36) Ross, R. B.; Powers, J. M.; Atashroo, T.; Ermler, W. C.; LaJohn,
L. A.; Christiansen, P. A. J. Chem. Phys. 1990, 93, 6654−6670.
(37) (a) Peng, C.; Ayala, P. Y.; Schlegel, H. B.; Frisch, M. J. J.
Comput. Chem. 1996, 17, 49−56. (b) Peng, C.; Schlegel, H. B. Isr. J.
Chem. 1994, 33, 449−454.
(38) Wolinski, K.; Hilton, J. F.; Pulay, P. J. Am. Chem. Soc. 1990, 112,
8251−8260.
(39) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105,
2999−3093.
(40) Parker, D.; Dickins, R. S.; Puschmann, H.; Crossland, C.;
Howard, J. A. K. Chem. Rev. 2002, 102, 1977−2010.
(41) Shannon, R. D. Acta Crystallogr. 1976, A32, 751−767.
(42) Aime, S.; Barge, A.; Benetollo, F.; Bombieri, G.; Botta, M.;
Uggeri, F. Inorg. Chem. 1997, 36, 4287−4289.
(43) (a) Vojtisek, P.; Cigler, P.; Kotek, J.; Rudovsky, J.; Hermann, P.;
Lukes, I. Inorg. Chem. 2005, 44, 5591−5599. (b) Kotek, J.; Rudovsky,
J.; Hermann, P.; Lukes, I. Inorg. Chem. 2006, 45, 3097−3102.
(44) Parr, R. G.; Pearson, R. G. J. Am. Chem. Soc. 1983, 105, 7512−
7516.
(45) Glaser, J. In Advances in Inorganic Chemistry; Sykes, G., Ed.;
Academic Press: San Diego, CA, 1995; Vol. 43, pp 1−78.
(46) Jalilehvand, F.; Maliarik, M.; Sandstrom, M.; Mink, J.; Persson,
̈
́
I.; Persson, P.; Toth, I.; Glaser, J. Inorg. Chem. 2001, 40, 3889−3899.
(47) Blixt, J.; Dubey, R. K.; Glaser, J. Inorg. Chem. 1991, 30, 2824−
2826.
(48) Nagy, P.; Joszai, R.; Fabian, I.; Toth, I.; Glaser, J. J. Mol. Liq.
2005, 118, 195−207.
(49) Banyai, I.; Glaser, J. J. Am. Chem. Soc. 1990, 112, 4703−4710.
(50) Banyai, I.; Blixt, J.; Glaser, J.; Toth, I. Acta Chem. Scand. 1992,
46, 138−141.
(51) (a) Karplus, M. J. Am. Chem. Soc. 1963, 85, 2870−2871.
(b) Geraldes, C. F. G. C.; Sherry, A. D.; Kiefer, G. E. J. Magn. Reson.
1992, 97, 290−304.
(52) Rodríguez-Rodríguez, A.; Esteban-Gomez, D.; de Blas, A.;
Rodriguez-Blas, T.; Fekete, M.; Botta, M.; Tripier, R.; Platas-Iglesias,
C. Inorg. Chem. 2012, 51, 2509−2521.
(53) Rodríguez-Rodríguez, A.; Esteban-Gomez, D.; de Blas, A.;
Rodriguez-Blas, T.; Botta, M.; Tripier, R.; Platas-Iglesias, C. Inorg.
Chem. 2012, 51, 13419−13429.
(54) van Wullen, C. J. Chem. Phys. 2012, 136, 114110.
̈
(55) Purgel, M.; Baranyai, Z.; de Blas, A.; Rodriguez-Blas, T.; Banyai,
I.; Platas-Iglesias, C.; Toth, I. Inorg. Chem. 2010, 49, 4370−4382.
(56) Platas-Iglesias, C. Eur. J. Inorg. Chem. 2012, 2023−2033.
́
(57) (a) Regueiro-Figueroa, M.; Estabn-Gomez, D.; de Blas, A.;
Rodríguez-Blas, T.; Platas-Iglesias, C. Eur. J. Inorg. Chem. 2010, 3586−
3595. (b) Regueiro-Figueroa, M.; Bensenane, B.; Ruscsak, E.; Esteban-
Gomez, D.; Charbonniere, L. J.; Tircso, G.; Toth, I.; de Blas, A.;
́
́
̀
́
́
Rodríguez-Blas, T.; Platas-Iglesias, C. Inorg. Chem. 2011, 50, 4125−
4141. (c) Lima, L. M. P.; Lecointre, A.; Morfin, J.-F.; de Blas, A.;
̀
Visvikis, D.; Charbonniere, L. J.; Platas-Iglesias, C.; Tripier, R. Inorg.
Chem. 2011, 50, 12508−12521. (d) Cosentino, U.; Villa, A.; Pitea, D.;
Moro, G.; Barone, V.; Maiocchi, A. J. Am. Chem. Soc. 2002, 124,
4901−4909. (e) Natrajan, L. S.; Khoabane, N. M.; Dadds, B. L.;
Muryn, C. A.; Pritchard, R. G.; Heath, S. L.; Kenwright, A. M.; Kuprov,
I.; Faulkner, S. Inorg. Chem. 2010, 49, 7700−7709. (f) Di Vaira, M.;
Stoppioni, P. New J. Chem. 2002, 26, 136−144.
L
DOI: 10.1021/acs.inorgchem.5b00458
Inorg. Chem. XXXX, XXX, XXX−XXX