See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/231742158

Curium(III) citrate speciation in biological systems: A europium(III) assisted

spectroscopic and quantum chemical study

Article  in  Dalton Transactions · October 2012

DOI: 10.1039/c2dt31480k · Source: PubMed

CITATIONS READS

26 809

6 authors, including:

Anne Heller Astrid Barkleit

Technische Universität Dresden Helmholtz-Zentrum Dresden-Rossendorf
25 PUBLICATIONS   90 CITATIONS    64 PUBLICATIONS   613 CITATIONS   

SEE PROFILE SEE PROFILE

Harald Foerstendorf Satoru Tsushima

Helmholtz-Zentrum Dresden-Rossendorf Helmholtz-Zentrum Dresden-Rossendorf
77 PUBLICATIONS   1,172 CITATIONS    100 PUBLICATIONS   1,718 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Sorption of actinides and fission product in brine View project

Speciation-dependent cellular uptake and cytotoxicity of tri- and tetravalent lanthanides and actinides in/onto mammalian kidney cells (DFG) View project

All content following this page was uploaded by Satoru Tsushima on 17 May 2014.

The user has requested enhancement of the downloaded file.

Dalton Dynamic Article Links

Transactions
Cite this: Dalton Trans., 2012, 41, 13969
www.rsc.org/dalton PAPER
Curium(III) citrate speciation in biological systems: a europium(III) assisted
spectroscopic and quantum chemical study†
Anne Heller,* Astrid Barkleit, Harald Foerstendorf, Satoru Tsushima, Karsten Heim and Gert Bernhard
Received 6th July 2012, Accepted 3rd September 2012
DOI: 10.1039/c2dt31480k

Citrate complexes are the dominant binding form of trivalent actinides and lanthanides in human urine at
pH < 6. Hence, an accurate prediction of the speciation of these elements in the presence of citrate is
crucial for the understanding of their impact on the metabolism of the human organism and the
corresponding health risks. We studied the complexation of Cm(III) and Eu(III), as representatives of
trivalent actinides and lanthanides, respectively, in aqueous citrate solution over a wide pH range using
time-resolved laser-induced fluorescence spectroscopy. Four distinct citrate complexes were identified and
their stability constants were determined, which are MHCit0, M(HCitH)HCit2−, M(HCit)23−, and
M(Cit)25− (M = Cm, Eu). Additionally, there were also indications for the formation of MCit− complexes.
Structural details on the EuHCit0 and EuCit− complexes were obtained with FT-IR spectroscopy in
combination with density functional theory calculations. IR spectroscopic evidence for the deprotonation
of the hydroxyl group of the citrate ion in the EuCit− complex is presented, which also revealed that the
complexation of the Eu3+ ion takes place not only through the carboxylate groups, like in EuHCit0, but
additionally via the hydroxylate group. In both EuHCit0 and EuCit− the carboxylate binding mode is
mono-dentate. Under a very low metal : citrate ratio that is typical for human body fluids, the Cm(III) and
Eu(III) speciation was found to be strongly pH-dependent. The Cm(III) and Eu(III) citrate complexes
dominant in human urine at pH < 6 were identified to be Cm(HCitH)HCit2− and a mixture of
Eu(HCitH)HCit2− and EuHCit0. The results specify our previous in vitro study using natural human urine
samples (Heller et al., Chem. Res. Toxicol., 2011, 24, 193–203).

Introduction last few decades.2 Consequently, there are several possibilities

Heavy metals, particularly radionuclides, represent a serious
health risk to humans in the case of incorporation. For the under-
standing of their (radio-) toxicity, distribution, deposition and
elimination, it is crucial to investigate their aqueous speciation
and molecular transport mechanisms in biosystems. Unfortu-
nately, only little is known about the behavior of trivalent acti-
nides (An(III)) and lanthanides (Ln(III)) in the human organism.
An(III) are artificial, highly radioactive elements that are mainly
produced within the nuclear fuel cycle in nuclear power plants
and, therefore, are contained in nuclear waste. Due to nuclear
incidents, natural disasters, or non-professional storage of radio-
active waste, An(III) can be released into the environment and the
biosphere. In contrast to this, Ln(III) are naturally occurring, non-
radioactive elements with a variety of applications in technology
and medicine.1 Especially the use of Ln(III) chelate complexes as
contrast enhancing agents in magnetic resonance imaging and as
molecular sensors and probes has grown enormously during the
Helmholtz-Zentrum Dresden-Rossendorf, Institute of Resource Ecology,
P.O. Box 510119, 01314 Dresden, Germany. E-mail: a.heller@hzdr.de;
Fax: +49 351 260 3553; Tel: +49 351 260 2251
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c2dt31480k
how humans potentially may get in contact with both An(III) and
Ln(III).
Actinide and lanthanide elements are hazardous to health due
to their radiological and/or chemical toxicity. Irrespective of the
uptake pathway, i.e., inhalation, ingestion or cutaneous absorp-
tion, the heavy metal ions are resorbed and transported by the
bloodstream prior to deposition in target organs or tissue. For all
An(III) and Ln(III) these are, in particular, the bones and liver,
respectively.3–5 Excretion of these elements is very low and
occurs mainly via the kidneys and, therefore, with urine.3–5 In
general, trivalent actinides and lanthanides exhibit analog chemi-
cal properties due to their similar ionic radii and oxidation states.
This, in turn, results also in a very similar behavior of An(III) and
Ln(III) in the human organism.
Apart from the well-studied effects of distribution, accumu-
lation and elimination, only a few investigations on the molecu-
lar chemical binding form of these elements in body fluids and
tissues have been performed. The speciation of An(III) in human
blood was studied by some authors, who showed that these
metal ions mainly bind to the plasma proteins transferrin and
serum albumin.5,6 Additionally, a significant part of the heavy
metal ions was found to bind to low molecular weight ligands,
such as phosphate and citrate.5,6 Preliminary investigations on
An(III) in urine suggested citrate complexation of the investigated

This journal is © The Royal Society of Chemistry 2012 Dalton Trans., 2012, 41, 13969–13983 | 13969
elements.7 Recently, we reported on the in vitro speciation of
curium(III) and europium(III), as representatives of An(III) and
Ln(III), respectively, in human urine.8 We demonstrated that, in
dependence on the pH, two different dominant species are
formed with each metal ion, out of which the one at slightly
acidic pH ≤ 5.7 is a curium(III) and europium(III) citrate
complex, respectively.8 Hence, in all studies on the speciation of
An(III) and Ln(III) in human body fluids published until now,
citrate is found to be at least partly involved in binding of the
heavy metal ions. Furthermore, it is known that the presence of
citrate also has a large influence on the solubility and, therefore,
on the mobility of trivalent actinides in aqueous solutions and
liquid nuclear waste.9
Citrate is a ubiquitous bioligand occurring in each body fluid,
not only in blood and urine, but also in sweat, liquor, and
saliva.10 Each citrate molecule has three carboxylic functional-
ities as well as one hydroxyl group. Hence, several species exist
in dependence on the pH. The deprotonation of citric acid is pre-
sented in Fig. 1a along with the respective pKa values.11,12 The
resulting speciation diagram is shown in Fig. 1b. In accordance
with Eberle and Moattar13 and Ohyoshi and Ohyoshi,14 citric
acid is regarded as a polybasic acid of the general formula
HxCitHy (x = 0–1, y = 0–3) within this paper. The first proton
reflects that of the hydroxyl group while the latter three represent
those of the carboxyl groups. Therefore, the five distinct ligand
Fig. 1 Stepwise deprotonation of citric acid (HCitH3) with the respective
pKa values for I = 0.1 M (a) and the resulting speciation diagram of 0.1 M
citric acid in aqueous solutions at I = 0.1 M and room temperature (b).
13970 | Dalton Trans., 2012, 41, 13969–13983
species are (i) citric acid HCitH3, (ii) dihydrogen citrate
HCitH2−, (iii) hydrogen citrate HCitH2−, (iv) citrate HCit3− and
(v) citrate with the deprotonated hydroxyl group Cit4−
(cf. Fig. 1a).
Literature data on the complexation of trivalent actinides and
lanthanides with citrate are quite numerous.9,13–27,28,29 While
europium(III) is well studied,9,15,19–22,24,25 less work was con-
ducted with curium(III).9,15,18,24,26 Luminescence spectroscopic
data of the complexes formed were reported once for europium(III)9
but to the best of our knowledge not for curium(III). Further-
more, the published literature data are inconsistent (see ESI,
Fig. S1 and Table S1†). For instance, the stability constant
reported for a certain complex species varies considerably
among the literature. In the case of 1 : 1 and 1 : 2 complexes of
europium(III) with the citrate anion HCit3− at I = 0.1 M and
room temperature, the reported complex formation constants
vary in the range of log K11 = 6.725–8.020 and log K12 = 9.821–
12.8,20 respectively. Moreover, there is also no agreement to
what extent the hydroxyl group of the citrate ligand is involved
in the heavy metal complexation and, thus, various structural
models have been published for An(III) and Ln(III) citrate
complexes.13,27–29
In the present study, we re-investigated the citrate complexes
of curium(III) and europium(III) as representatives of An(III) and
Ln(III), respectively, within a wide pH and ligand concentration
range. Due to the above-mentioned chemical similarities, it is
commonly assumed that Ln(III) mimic An(III) and can be used as
suitable nonradioactive chemical analogs.3,30 Because both
elements exhibit unique luminescence properties, their speciation
was studied with time-resolved laser-induced fluorescence spectro-
scopy (TRLFS). This technique is highly suitable to determine
the binding form of a luminescent metal ion at trace
concentrations.8,31–34 From both single and time-resolved experi-
ments, valuable parameters, such as emission wavelengths and
luminescence lifetimes, are obtained. These parameters are sensi-
tive to the chemical environment of the heavy metal ions and
vary in dependence on the ligand. Furthermore, the number of
water molecules in the first coordination shell of the heavy metal
ions can be estimated from the emission lifetime.35 Structural
details on the formed europium(III) citrate complexes were
obtained by Fourier-transform infrared (FT-IR) spectroscopy as
well as quantum mechanical calculations using the density func-
tional theory (DFT). This combined approach is highly suitable
to determine the functional group(s) of the ligand actually
binding the Eu3+ ion and the solution structure of the complex
species. Furthermore, the question whether the hydroxyl group
participates in the Eu3+ binding is addressed. The presented data
provide a comprehensive image of this complex ligand system
and the various species formed with curium(III) and europium(III)
in aqueous solutions with implications for their biological
relevance.
Experimental section
Solutions and reagents
Water free citric acid (≥99.5%, p.a., ACS) was purchased from
Roth, EuCl3·6H2O (99.999) from Sigma, NaClO4 ( p.a.),
This journal is © The Royal Society of Chemistry 2012
NaCl ( p.a.), HCl ( p.a.) and NaOH ( p.a.) from Merck. Stock
solutions of citric acid were freshly prepared for each series of
measurement by dissolving in water and europium(III) stock solu-
tions were prepared by dissolving EuCl3 in water. In the case of
curium(III) a 2.36 × 10−4 M stock solution of the long-lived
248
Cm isotope (half-life = 3.4 × 105 years) in 1 M HClO4
was used. This radionuclide was supplied by the U.S.
Department of Energy, Office of Basic Energy Sciences, via the
transplutonium element production facilities at Oak Ridge
National Laboratory. A stock solution of 3 M NaClO4 (TRLFS
measurements) or 3 M NaCl (IR measurements) was used as the
background electrolyte to maintain constant ionic strength in the
experiments.
All stock solutions and samples were prepared in threefold
distilled and carbonate free water. The pH of the samples was
measured with glass electrodes (InLab 427 combination pH
puncture electrode from Mettler-Toledo and BlueLine 16 pH
electrode from Schott, respectively) and corrected with
NaOH and HClO4 (TRLFS measurements) or HCl (IR
measurements).
TRLFS measurements
Time-resolved laser-induced fluorescence spectroscopy was per-
formed using a pulsed flash lamp pumped Nd:YAG-OPO laser
system (Powerlite Precision II 9020 laser equipped with a Green
PANTHER EX OPO) from Continuum. Details on the experi-
mental setup are given elsewhere.36 The laser pulse energy was
monitored using a photodiode and varied between 2 and 3 mJ.
The emission spectra were detected by an optical multichannel
analyzer consisting of a monochromator (Oriel MS 257), a spec-
trograph with different gratings (300 and 1200 lines per mm,
respectively), and an ICCD camera (Andor iStar); all com-
ponents were purchased from the Lot-Oriel Group. A constant
time window of 1 ms and an excitation wavelength of 395 nm
were applied for all measurements. Curium(III) and europium(III)
steady-state and time-dependent luminescence spectra were
recorded in the 570–640 nm (grating = 1200 lines per mm with
0.2 nm resolution) and 450–750 nm (grating = 300 lines per mm
with 0.7 nm resolution) ranges, respectively. For time-resolved
measurements a total of 35 spectra with delay steps of 10–100 μs
were recorded per sample.
In the case of curium(III) all experiments were run in a glove
box under a nitrogen atmosphere, whereas europium(III)
measurements were carried out under nitrogen as well as normal
atmosphere. All experiments were performed at 24 ± 1 °C and a
constant ionic strength of 0.1 M (NaClO4). The complexation
reaction was studied at a fixed heavy metal concentration of
3 × 10−5 M europium(III) and 3 × 10−7 M curium(III), respecti-
vely, by varying the citrate concentration from 10−5 to 0.1 M
and the pH from 2 to 13. All samples were prepared directly in
the cuvette and the measurements were performed as spectro-
photometric titrations.
ATR FT-IR measurements
Attenuated total reflection Fourier-transform infrared spec-
troscopy was performed using a Vertex 80/v vacuum
This journal is © The Royal Society of Chemistry 2012
spectrometer from Bruker Optics, equipped with a mercury
cadmium telluride detector. Details on the experimental setup are
given elsewhere.37 The used ATR unit (DURA SamplIR II from
Smiths Inc.) is a horizontal diamond with nine internal reflec-
tions on the upper surface and an incidence angle of 45 °. An
ATR flow cell with a volume of 200 μL was used to ensure
adequate background subtraction without external thermal inter-
ference. Spectra were recorded in the range of 1800–800 cm−1 at
a spectral resolution of 4 cm−1. All experiments were performed
under normal atmosphere at 24 ± 1 °C and a constant ionic
strength of 0.1 M (NaCl). Due to the high specific radioactivity
of curium(III), IR measurements were only performed with
europium(III).
As a first step, the spectral properties of the pure ligand were
studied at pH 1 to 13.5. The complexation with europium(III)
was then investigated at a fixed metal to ligand ratio of 1 : 1
(0.01 M) in the same pH range. In the case of the absorption
spectra, a spectrum of pure water was used for background cor-
rection. Difference spectra were calculated from single beam
spectra of the solutions containing either complexes or the pure
citric acid. Positive and negative bands in the difference spectra
represent vibrational modes of the complexes and the pure
ligand, respectively.
It should be noted that within IR measurements a colorless,
amorphous precipitation occurred in europium(III) citrate
solutions at pH 4 to 6. Using various techniques for elemental
and thermal analysis, this precipitate was identified to be
most probably of the empirical formulae EuHCit0·H2O.
Nevertheless, only IR spectra of the liquid phases are presented.
Data analysis
All absorption (IR) and emission (TRLFS) spectra were analyzed
with the Origin 7.5G software38 to obtain peak positions.
TRLFS spectra were baseline corrected and the steady-state
spectra were normalized. Since curium(III) exhibits a degenerated
ground state,39 this was applied to the intensity ( peak area) of
the whole spectrum. In the case of europium(III) normalization
was applied only to the peak area of the 5D0 → 7F1 band
because the luminescence to this ground state is considered to be
independent from the chemical environment of the metal
ion.32,40 Additionally, in the case of europium(III), the intensity
ratio (RE/M) between the electric dipole transition 5D0 → 7F2 and
the magnetic dipole transition 5D0 → 7F1 was determined as
follows:
RE=M ¼ Ið7 F2 Þ=Ið7 F1 Þ; ð1Þ
with I being the luminescence intensity ( peak area) of the
respective transition. Furthermore, Origin was also used to deter-
mine the luminescence lifetime of emitting species according to
the equation of exponential decay:
IðtÞ ¼ Σi I i expðt=τ i Þ: ð2Þ
In eqn (2), I is the total luminescence intensity at the time t, Ii
the luminescence intensity of species i at t = 0, and τi the corre-
sponding lifetime. With this lifetime, the number of water mole-
cules in the first coordination sphere of the heavy metal ions was
Dalton Trans., 2012, 41, 13969–13983 | 13971
calculated using the following equations given by Kimura
et al.35 in which the lifetime is inserted in ms:
nH2 O + 0:5 ¼ 0:65=τ  0:88 for curiumðiiiÞ ð3Þ
nH2 O + 0:5 ¼ 1:07=τ  0:62 for europiumðiiiÞ: ð4Þ
Complex formation constants and emission spectra of the
single species were determined independently using Specfit41
and HypSpec.42 These factor analysis programs decompose
spectra of mixtures into their components based on the spectro-
scopic properties of each species, which vary in dependence on
the pH or the ligand concentration. Furthermore, they calculate
reasonable complex stability constants. The successful appli-
cation of this software was demonstrated by earlier work on the
complexation of curium(III), europium(III), and americium(III)
with various ligands.31,33,36,43 Input parameters for the data fitting
were the total concentrations of the heavy metals and citrate, the
stability constants of the heavy metal hydroxides,44 the protonation
constants of the ligand species,11,12 the pH, and the measured
sample spectra. For all constants, the values valid for I = 0.1 M
have been used and, where necessary, recalculated from I = 0 M
using the Davies equation. The measured luminescence spectra of
the Cm3+ and Eu3+ aqua ions were used as fixed normative spectra.
Based on the identified species, the formation constants for
the direct reactions, log Kpqr ( p = number of M3+ ions, q =
number of ligand molecules, r = number of protons), were calcu-
lated according to the mass action law for the following
equations:
M3þ þ HCit3 ! MHCit0 for log K 111 ð5Þ
M3þ þ HCitH2 þ HCit3 ! MðHCitHÞHCit2 for log K 123
ð6Þ
M3þ þ 2HCit3 ! MðHCitÞ2 3 for log K 122 ð7Þ
M3þ þ 2Cit4 ! MðCitÞ2 5 for log K 120 ¼ log β120 : ð8Þ
Extrapolation of the complex formation constants for zero
ionic strength, log K0, was calculated with ionic strength
corrections using Specific Interaction Theory (SIT).45 The SIT
parameters of Eu3+ and citrate were taken from Vercouter et al.46
and De Stefano et al.,47 respectively. In the case of Cm3+,
no SIT parameter exists so far. Nevertheless, the SIT parameter
of Nd3+, which is similar to almost all Ln(III), is recommended
to be also used for Am3+.48 Therefore, this value was taken for
Cm3+. The log βpqr values for the overall reactions were calcu-
lated according to the mass action law for the following
equations:
M3þ þ Cit4 þ Hþ ! MHCit0 for log β111
ð9Þ
M3þ þ 2Cit4 þ 3Hþ ! MðHCitHÞHCit2 for log β123 ð10Þ
M3þ þ 2Cit4 þ 2Hþ ! MðHCitÞ2 3 for log β122 : ð11Þ
13972 | Dalton Trans., 2012, 41, 13969–13983
Additionally, the Gibbs energy of the reaction according to
eqn (5)–(8) was then determined as follows:
ΔR G 0 ¼ RT ln K 0 : ð12Þ
Speciation calculations were carried out with Hyss200649
using the same parameters as in Specfit and HypSpec as well as
the calculated stability constants (log β) of the heavy metal
citrate complexes.
DFT calculations
DFT calculations were performed using Gaussian 03.50
Geometries were optimized in the aqueous phase at the B3LYP
level using the CPCM solvation model51 with UAHF radii.52
The large core effective core potential as well as the corres-
ponding basis set suggested by Dolg et al.53 was used on
europium(III). For C, O, and H, the all-electron valence triple-ζ
basis set plus polarization and diffuse functions have been
used.54 More details of the DFT calculations have been described
in a previous publication.55
Results and discussion
TRLFS spectra of europium(III) citrate complexes
The citrate ligand shows no luminescence in the investigated
wavelength range (570–640 nm), whereas the Eu3+ ion exhibits a
characteristic emission spectrum. Hence, there is no interference
with the ligand and the detected luminescence results solely
from the trivalent europium. To enable the complexation by
different citrate species (cf. Fig. 1a), europium(III) luminescence
spectra were measured at a constant metal concentration of
3 × 10−5 M and pH 2.0, 4.0, 5.5, 8.5, and 13.0 with varying
ligand concentrations of 10−5–10−1 M (Fig. 2).
The luminescence spectrum of the Eu3+ aqua ion is character-
ized by emission bands at 585–600 nm and 610–625 nm repre-
senting the magnetic dipole transition 5D0 → 7F1 and the
hypersensitive electric dipole transition 5D0 → 7F2, respecti-
vely.8,32,56,57 The intensity ratio (RE/M) according to eqn (1) is
0.25–0.5 and the luminescence lifetime is 110 ± 10 μs corre-
sponding to 9 water molecules in the first hydration sphere of the
Eu3+ ion.8,32,56,57 Spectroscopic changes indicating the com-
plexation of trivalent europium were observed at each pH. In
general, luminescence wavelengths remain nearly unaltered upon
complexation and a shift of the emission maxima was observed
only for exceptional cases. The luminescence spectra of all
europium(III) citrate complexes exhibit an additional emission
band at around 578 nm corresponding to the Laporte forbidden
transition 5D0 → 7F0, which occurs only upon deformation of
the spherical symmetry of the Eu3+ aqua ion. Furthermore, the
7
F1 and 7F2 bands are split and the RE/M is significantly enhanced.
Finally, also the emission lifetime prolongs upon complexation.
In each sample, we observed mono-exponential luminescence
decay. All spectroscopic changes are summarized in Table 1.
In the acidic region at pH 2.0 (cf. Fig. 2a), the luminescence
spectrum of europium(III) remains unaltered up to a ligand con-
centration of 2.5 × 10−4 M and all parameters correspond to
those of the uncomplexed Eu3+ aqua ion. At higher citrate
This journal is © The Royal Society of Chemistry 2012
Fig. 2 Normalized steady-state luminescence spectra of 3 × 10−5 M europium(III) + citrate at pH 2.0 (a), 4.0 (b), 5.5 (c), 8.5 (d), and 13.0 (e), I = 0.1 M,
and room temperature in dependence on the ligand concentration (inset: enlargement of the 5D0 → 7F0 transition band) as well as the deconvoluted
emission spectra (f ) of the single europium(III) citrate complexes (Eu3+ without citrate = grey line).

Table 1 Luminescence spectroscopic parameters of the determined europium(III) species

λ/nma

pH [citrate]/M 7
F0 7
F1 7
F2 RE/M b τ/μs nH2Oc Assigned species Reference

1–6 0 — 591.7 616.4 0.5 110 ± 10 9.1 Eu3+ aqua ion Present work
2.0 >5 × 10−4 578.7 590.5/594.0 614.1/616.7 1.6 >145 <6.8 EuHCit0 Present work
4.0 ≤10−3 578.7 590.5/593.8 614.0/616.8 1.5 168 ± 4 5.8 EuHCit0 Present work
5.5 <4 × 10−4 578.7 590.5/593.9 614.0/616.8 1.5 162 ± 10 6.0 EuHCit0 Present work
4.0 >10−3 578.9 590.6/595.1 614.1/617.2 2.2 >245 <3.8 Eu(HCitH)HCit2− Present work
5.5 ≥4 × 10−4 578.9 590.5/595.0 614.0/617.3 2.0 250 ± 5 3.7 Eu(HCitH)HCit2− Present work
8.5 >5 × 10−4 579.3 589.5/595.3/589.9 614.8/619.7 2.4 490 ± 20d 1.6d Eu(HCit)23− Present work
13.0 ≥10−3 578.6 589.1/598.5 612.5/620.2 1.8 676 ± 5 1.0 EuCit25− Present work
≤5.7 — 578.7 590.2/594.5 613.9/616.7 2.2 153 ± 19 6.4 Eu(III) in human urine 8
a
Wavelengths of luminescence in the specified ground states with a standard deviation of ±0.1 nm. b Intensity ratio of the 7F2 (electric dipole
transition) over the 7F1 (magnetic dipole transition) luminescence band with a standard deviation of ±0.2 nm. c Number of water molecules calculated
according to Kimura et al.35 with a standard deviation of ±0.5. d Luminescence spectra show that Eu(HCit)23− and a hydrolyzed europium(III) species
co-exist at this pH. Therefore, both species contribute to the measured lifetime.

This journal is © The Royal Society of Chemistry 2012 Dalton Trans., 2012, 41, 13969–13983 | 13973
concentrations, the 7F0 peak evolves at 578.7 ± 0.1 nm. Further-
more, at ≥5 × 10−3 M citrate, the RE/M is enhanced up to 1.6 ±
0.1 and the luminescence lifetime is prolonged to >145 μs indi-
cating the formation of a Eu(III) citrate species 1.
At pH 4.0 and 5.5 (cf. Fig. 2b and 2c), the same spectral
changes were observed even with the smallest ligand addition. In
the range of 10−5–10−3 M citrate at pH 4.0 and 10−5–4 × 10−4 M
at pH 5.5, all luminescence spectra are very similar with an RE/M
of 1.5 ± 0.1 and a luminescence lifetime of 165 ± 10 μs. Accord-
ing to eqn (4), this corresponds to a number of 5.9 ± 0.5 water
molecules in the first hydration shell of the Eu3+ ion. All spectro-
scopic parameters equal those of the Eu(III) citrate species 1
determined at pH 2.0. Citrate concentrations of >10−3 M at pH
4.0 and >4 × 10−4 M at pH 5.5 result in a further enhancement
of the RE/M to 2.1 ± 0.1 and a prolonged luminescence lifetime
of 250 ± 5 μs corresponding to 3.7 ± 0.5 water molecules in the
first coordination sphere of the trivalent europium. This indicates
the formation of a new Eu(III) citrate species 2 at both pH values
and a large excess of citrate. Furthermore, both complex species
differ in the shape of the 7F0 peak. In the case of the Eu(III)
citrate species 1, the sharp and quite intense peak occurs at
578.7 nm. In contrast, for the Eu(III) citrate species 2, it occurs at
578.9 nm and is of much less intensity as well as broader
(cf. insets of Fig. 2b and 2c).
At pH 8.5 (cf. Fig. 2d), due to metal ion hydrolysis,56–58 spec-
tral changes occur already in the europium(III) solution without
citrate. Ligand addition induces further alterations, such as the
fine structures of the 7F1 and 7F2 bands being substantially
different from those of the hydrolyzed Eu3+ ion or those of the
two complex species determined at lower pH. This indicates
the formation of another Eu(III) citrate species 3. Strikingly, the
splitting of the 7F1 band is enhanced to four single peaks upon
complexation. Since the 7F1 ground state of the Eu3+ ion can
only be split into three Stark levels at the most,59 four single
peaks point to the simultaneous existence of two different euro-
pium(III) species. Since one of the split peaks (592.5 nm) signifi-
cantly decreases in intensity with increasing ligand concentration
and nearly disappears at the highest concentration of 10−3 M
citrate, it does not belong to the spectrum of the Eu(III) citrate
species 3 but to that of the hydrolyzed europium(III) species.
This demonstrates the competition of citrate complexation and
metal hydrolysis at this pH value. Furthermore, the measured
emission lifetime is initially prolonged with increasing citrate
concentrations of ≤10−4 M (hydrolysis + citrate complexation)
but shortened at higher ligand concentrations ( predominantly
citrate complexation). At the highest ligand concentrations, the
lifetime is 490 ± 20 μs corresponding to 1.6 ± 0.5 water mole-
cules in the first hydration shell of the Eu3+ ion.
At pH 13.0 (cf. Fig. 2e), no luminescence signal was detected
in pure europium(III) solution and after the addition of ≤2.5 ×
10−4 M citrate. This is due to the formation of the poorly water
soluble Eu(OH)3, which dominates the europium(III) speciation
at high pH.57 At ligand concentrations exceeding 5 × 10−4 M
citrate, steady-state and time-resolved spectra are evaluable indi-
cating the partial re-solution of Eu3+ ions due to complexation
with citrate. In contrast to the spectra at all other pH values, the
luminescence intensity of the 7F2 band decreases with increasing
ligand concentrations. At ≥10−2 M citrate, the RE/M is 1.8 ± 0.2
and the lifetime is prolonged to 676 ± 5 μs corresponding to
13974 | Dalton Trans., 2012, 41, 13969–13983
1.0 ± 0.5 water molecules. This indicates the formation of
another Eu(III) citrate species 4 at this pH.
In summary, four different europium(III) citrate species were
determined and characterized by varying the citrate concentration
at pH 2.0–13.0. The estimation of the number of water molecules
left in the first coordination shell of the Eu3+ ion and literature
data on several europium(III) citrate complexes9,15,19–22,24,25
allow a tentative assignment of the four europium(III) citrate
species.
The Eu(III) citrate species 1 is identified as the charge neutral
1 : 1 complex EuHCit0 with the citrate anion HCit3−. The lumi-
nescence lifetime of 165 ± 10 μs equals 3.2 ± 0.5 water mole-
cules that have been replaced by the ligand upon complexation.
The Eu(III) citrate species 2 is found to be the 1 : 2 complex
Eu(HCitH)HCit2−. In this species, one hydrogen citrate HCitH2−
and one citrate anion HCit3− bind to the Eu3+ ion. The lumines-
cence lifetime of 250 ± 5 μs indicates that 5.3 ± 0.5 water mole-
cules have been displaced by the ligands. The Eu(III) citrate
species 3 is assigned to the 1 : 2 complex Eu(HCit)23−. The
citrate anion HCit3− is the exclusive ligand species at neutral to
alkaline pH and protonation upon complexation is very unlikely.
Furthermore, the luminescence lifetime of 490 ± 20 μs corre-
sponds to a replacement of 7.5 ± 0.5 water molecules and, there-
fore, also supports the binding of two ligand molecules, which
each displace at least three water molecules. Finally, Eu(III)
citrate species 4 is found to be the 1 : 2 complex EuCit25−. This
results, first of all, from the fact that the complexation occurs
only at high citrate excess indicating the binding of more than
one ligand molecule. Moreover, the luminescence lifetime of
676 ± 5 μs is very long and equals the displacement of 8.1 ± 0.5
water molecules by the ligand. However, also the hydrolysis of
water molecules from the first coordination sphere of the Eu3+
ion can not be excluded at such alkaline pH. Up to now, there
are no reference data on europium(III) citrate species at pH > 10.
Finally, two series were measured at a constant metal to ligand
ratio (equimolar and with ligand excess) by varying the pH from
2.0 to 12.0 (see ESI, Fig. S2†). In the series of ligand excess, all
four complex species were identified and there were no indi-
cations for metal ion hydrolysis. At an equimolar metal to ligand
ratio, only the EuHCit0 complex and an additional, yet uniden-
tified, species were detected. In addition, metal ion hydrolysis
occurred at very alkaline pH. Taking the equimolar metal to
ligand ratio into account, the unknown species is most probably
the 1 : 1 complex EuCit−, where the fully deprotonated citrate
anion Cit4− binds to the Eu3+ ion. Analog complexes are
reported for curium(III), americium(III), gadolinium(III), and neo-
dymium(III).13,17,23,26,29 All pH-dependent measurements
demonstrate that the determined complexes subsequently equili-
brate with each other. In conclusion, the An(III) and Ln(III) citrate
species predominantly reported in the literature at the several pH
ranges and metal to ligand ratios were identified for europium(III)
and the respective TRLFS spectra were assigned (see Table 1).
The single component spectra are presented in Fig. 2f.
To date, only one luminescence spectroscopic work on the
complexation of europium(III) with citrate has been published.
Mathur et al. reported 7F0 → 5D0 excitation spectra for 10−4 M
europium(III) + 0.003 and 0.008 M citrate, respectively, at pH 3.6
to 9.0 and a very high ionic strength of 6.6 M NaClO4.9 They
detected a slight red shift of the excitation band from 579.1 nm
This journal is © The Royal Society of Chemistry 2012
at pH 3.6 to 579.9 nm at pH 9.09 comparable to that of the 7F0
peak reported in this paper (cf. Table 1). Emission spectra poten-
tially serving as references for this work were not published. In
general, the lifetimes reported by Mathur et al. prolong with
increasing pH, which is in agreement with our findings (cf.
Table 1). The lifetime of EuHCit0 determined in the present
work is in very good agreement with the value reported by
Mathur et al.9 Regarding the identification of the other europium(III)
citrate species, discrepancies occur upon interpretation of the
data. Whereas in our work three complexes were determined in
the range of pH 3 to 9, Mathur et al. solely account for com-
plexation by the citrate anion HCit3–9 and, therefore, only for the
formation of EuHCit0 and Eu(HCit)23− but not for that of the
mixed Eu(HCitH)HCit2−. Moreover, the reported lifetimes for
Eu(HCit)23− vary significantly between 213 and 640 μs, which
was explained by varying numbers of water molecules in the
first hydration shell of the Eu3+ ion.9 A comparison of the litera-
ture lifetimes with those determined in this study is presented in
Table 2. On the basis of this comparison, it is reasonable to
suppose that not only the EuHCit0 and Eu(HCit)23− species were
formed within the study of Mathur et al.9 but also the mixed
Eu(HCitH)HCit2− and possibly also the EuCit25− complex.
different pH values ranging from 2 to 13.5 (Fig. 3). These
spectra serve as a reference for the interpretation of the spectra
of the europium(III) complexes. In general, the IR spectra of the
ligand species are in good agreement with published literature
data,60 and the tentative assignments of the vibrational bands are
given in Table 3. All spectra are dominated by bands represent-
ing the carboxyl groups, observed at 1722 and 1226 cm−1 in the
spectra recorded at pH ≤ 4, and the carboxylate groups, observed
around 1568 and 1390 cm−1 at higher pH. At pH ≥ 6, no bands
representing carboxyl groups are observed anymore. The spec-
trum recorded at pH 13.5 shows an additional band at
1016 cm−1, which is due to the deprotonation of the hydroxyl
group ( pKa = 13.511). To the best of our knowledge, an IR
absorption spectrum, which evidences the deprotonation of the
hydroxyl group in citric acid, has not been published before.
IR spectra and DFT calculations of europium(III) citrate
complexes
The IR spectra of equimolar aqueous solutions of europium(III)
and citrate recorded in the range of pH 2–12 are shown in

IR spectra of citrate

Prior to the measurement of the europium(III) citrate complexes,

the vibrational spectra of the pure ligand were recorded at

Table 2 Comparison of the luminescence lifetime of europium(III)

citrate species reported in the literature with those determined within this
study

Species determined in the

Species reported by Mathur et al.9 present work

Formulaea Lifetime/μs Lifetime/μs Formulae

EuHCit(H2O)60 180 ± 6 165 ± 10 EuHCit0

Eu(HCit)2(H2O)33− 213 ± 8 250 ± 5 Eu(HCitH)HCit2−
Eu(HCit)2(H2O)3− pH 7 470 ± 18 490 ± 20 Eu(HCit)23−
Eu(HCit)2(H2O)3− pH 9 636 ± 10 676 ± 5 EuCit25−
a
Renamed according to the ligand species used in the present work. Fig. 3 IR absorption spectrum of 10−2 M citrate at I = 0.1 M (NaCl)
and room temperature in dependence on the pH.

Table 3 Tentative assignment of the IR absorption bands of the ligand species

Wavenumber/cm−1 a

HCitH3 (pH 1–2) HCitH2−/HCitH2− (pH 4–6) HCit3− (pH 8–12) HCit3−/Cit4− (pH 13.5) Tentative assignmentb

1722 (s) 1722 (m) — — νs(HOCvO)

1633 (w) — — — δ(H–O–H)
— 1576 (s) 1568 (s) 1568 (m) νas(COO−)
1399 (w) 1396 (s) 1390 (s) 1390 (m) νs(COO−)
— — 1292 (w) — δ(OvC–O−) or δ(H–C–H) + δ(C–O–H)
— 1280 (w) 1280 (m) — ν(C–OH) + δ(C–O–H)
— — 1258 (w) — δ(OvC–O−) or δ(H–C–H) + δ(C–O–H)
1226 (s) 1226 (m) — — ν(OC–OH) + δ(C–O–H)
— 1106 (w) 1095 (w) — ν(C–OH)
— — — 1016 (s) ν(C–O−)
a
s = strong, m = medium, w = weak. b ν = stretching vibration, νs/νas = symmetric/antisymmetric stretching vibration, δ = deformation vibration.

This journal is © The Royal Society of Chemistry 2012 Dalton Trans., 2012, 41, 13969–13983 | 13975
Fig. 4a. In general, the absorption spectra potentially show con-
tributions from uncomplexed citric acid. Therefore, difference
spectra were calculated from the respective single beam spectra
of the solutions containing europium(III) citrate complexes and
the pure ligand (Fig. 4b). In these spectra, only modes of the
ligand undergoing alterations upon complexation are detected.
Hence, negative bands in the difference spectra correspond to
vibrational modes of the pure ligand while positive ones re-
present the respective complex species. Table 4 summarizes the
tentative assignments of the difference bands to the functional
groups.
At pH ≤ 4, the presence of carboxyl groups is evidenced by
the absorption and difference spectra showing positive and nega-
tive bands respectively, at 1722 and 1226 cm−1, indicating con-
tributions of uncomplexed citric acid molecules. The shape of
the difference spectra changes up to pH 6, whereas at pH ≥ 8 no
significant spectral alterations are observed either in the absorp-
tion or the difference spectra. Hence, from both series of spectra,
two different europium(III) citrate complexes can be derived. The
first species dominates at low pH, while the second one is predo-
minantly formed at higher pH. The spectra recorded at pH 6
obviously show contributions from both complex types dominat-
ing the speciation at lower and higher pH values.
In the acidic region, at pH 2 and 4, the predominant species is
designated as the EuHCit0 species according to the TRLFS
results. IR spectra of this complex are characterized by two
strong bands representing vibrational modes of carboxylate
groups observed at 1600 and 1570 cm−1 for the ν3,as(COO)
mode and at 1417 and 1397 cm−1 for the ν3,s(COO) mode. The
splitting of these modes in the spectra of the europium(III)

Fig. 4 IR absorption (a) and difference (b) spectra of 10−2 M citrate + 10−2 M europium(III) at I = 0.1 M (NaCl) and room temperature in depen-
dence on the pH.

Table 4 Tentative assignment of the ATR-FT-IR absorption and difference bands of europium(III) citrate complex species (all data as wavenumber/
cm−1)

EuHCit0 (pH 2–4) Mixture (pH 6) EuCit− (pH 8–12)

Absa Diffb Absa Diffb Absa Diffb Tentative assignmentc Ligand/complex

1719 (m) 1720 (−) — 1711 (−) — — νs(HOCvO) Ligand

1600 (sh) 1600 (+) — 1604 (+) — 1579 (−) νas(COO−) Complex binding
1570 (s) 1568 (+) 1565 (s) 1550 (+) 1564 (s) 1550 (+) νas(COO−) Complex non-binding
1417 (s) 1419 (+) 1430 (sh) 1427 (+) 1425 (sh) 1427 (+) νs(COO−) Complex binding
1397 (s) 1397 (+) — — 1405 (s) 1406 (+) νs(COO−) Complex non-binding
— — 1400 (s) 1380 (−) 1385 (sh) 1385 (−) νs(COO−) Ligand
1305 (w) 1300 (+) 1297 (w) 1303 (+) 1293 (w) 1310 (+) δ(OvC–O−) or δ(H–C–H) + δ(C–O–H) Complex non-binding
1282 (m) 1276 (+) — 1279 (−) — 1278 (−) ν(C–OH) + δ(C–O–H) Ligand
1254 (w) 1254 (+) 1252 (w) 1252 (+) 1248 (w) 1246 (+) δ(OvC–O−) or δ(H–C–H) + δ(C–O–H) Complex non-binding
1237 (m) 1222 (−) — 1209 (−) — 1207 (−) ν(OC–OH) + δ(C–O–H) Ligand
1078 (m) 1078 (+) 1078 (w) 1074 (+) — — ν(C–OH) Complex
a
s = strong, m = medium, w = weak, sh = shoulder. b + = positive band/maximum, − = negative band/minimum. c ν = stretching vibration, νs/νas =
symmetric/antisymmetric stretching vibration, δ = deformation vibration.

13976 | Dalton Trans., 2012, 41, 13969–13983 This journal is © The Royal Society of Chemistry 2012
complex indicates the presence of coordinated and uncomplexed
carboxylate groups.
In contrast, at alkaline pH 8–12, predominantly the negatively
charged EuCit− complex with the completely deprotonated
citrate anion Cit4− is formed. IR spectra of this complex show
strong absorption bands at 1564 and 1405 cm−1 representing the
ν3,as(COO) and ν3,s(COO) mode, respectively. Because of the
broad shape of these bands, a more accurate interpretation can be
derived from the difference spectra (Fig. 4b). The ν3,as(COO)
mode is represented by a difference band at 1578(−)/1550(+) cm−1,
whereas the ν3,s(COO) mode is observed at 1427(+)/1385(−) cm−1
showing an interfering small band at 1406 cm−1. The difference
bands, again, clearly evidence the coordination of the carboxy-
late groups to the Eu3+ ion.
The type of coordination of carboxylate groups to heavy metal
ions can be derived from vibrational spectra by the degree of the
spectral splitting of the ν3(COO) modes (Δν3).61 The Δν3 value
of the uncomplexed ligand serves as a reference and is
∼180 cm−1 throughout the pH range (cf. Fig. 4). In general, a bi-
dentate coordination of the carboxylate groups to the Eu3+ metal
ion is expected to show a significantly lower spectral splitting
than a mono-dentate binding.61,62 For citrate, Δν3 values of
164–228 cm−1 were reported for mono-dentate complexes,
whereas Δν3 values ranging from 42 to 77 cm−1 were found for
a bi-dentate binding.61,62
In the spectrum of the EuHCit0 complex, the bands of the car-
boxylate groups coordinated to the heavy metal ion are observed
at 1600 and 1417 cm−1 resulting in a Δν3 value of ∼185 cm−1.
This value is very close to that of the pure ligand and substanti-
ates a mono-dentate coordination of the carboxylate groups to
the Eu3+ ion. In the respective spectrum of the EuCit− complex,
the maxima observed at 1550 and 1427 cm−1 reflect the differ-
ence bands of the ν3(COO) modes coordinated to the metal ion.
The resulting Δν3 value of ∼125 cm−1 is smaller than the range
reported for a mono-dentate complex. However, it is larger than
those reported for a bi-dentate coordination.61,62 Hence, contri-
butions of both binding modes have to be considered for the
EuCit− complex.
In the spectral region between 1300 and 1200 cm−1, all
spectra exhibit characteristic features. In the absorption spectra,
three bands at 1305, 1282, and 1254 cm−1 are observed at pH ≤
4, which are partially superimposed by the strong band at
1226 cm−1 in the spectrum recorded at pH 2 (cf. Fig. 4a). At
higher pH, only two bands at 1293 and 1248 cm−1 are observed
and the corresponding difference spectra show a difference band
Fig. 5 Schematic structure models of reported MHCit0 (a: models 1–3) and MCit− (b: models 4–6) complexes of several heavy metal ions (M =
UO22+, Gd3+, Dy3+, Cr3+, Al3+, Fe3+, Ga3+, Pr3+ or Am3+).
This journal is © The Royal Society of Chemistry 2012
at 1278(−)/1246(+) cm−1. With respect to the spectra of the pure
ligand (Fig. 3), the band at 1278 cm−1 can be attributed to the
hydroxylate group because it is clearly absent in the spectrum
recorded at pH 13.5, i.e., the Cit4− starts to predominate. There-
fore, from the spectral features observed in this frequency range,
a contribution of the hydroxylate group to the binding of citrate
to the Eu3+ ion can be concluded.
In summary, the IR spectra suggest a predominant mono-
dentate binding of the Eu3+ ion via carboxylate groups in both
complex species EuHCit0 and EuCit−. Additionally, the hydro-
xylate group is contributing to the complexation of the Eu3+ ion
in the EuCit− complex.
To check this assumption and to deduce the actual solution
structure of EuHCit0 and EuCit−, different complex structure
models were calculated with DFT and the resulting IR absorption
spectra were compared to the measured ones. For europium(III),
no solution structure for MHCit0 and MCit− has been reported
until now, but for other metal ions, such as Gd(III),27 Dy(III),27
Pr(III),28 Am(III),13 U(VI),63 Cr(III),64 Fe(III),64,65 Al(III)66,67 and
Ga(III),67 several structures have been published in the literature.
Fig. 5 shows the structure models reported for such MHCit0
(models 1–3) and MCit− (models 4–6) complexes, which vary
with regard to the number and the nature of the functional
groups actually binding to the respective heavy metal ion. Con-
sequently, all these structure models were taken into account as
possible EuHCit0 and EuCit− structures and carefully checked
with DFT calculations.
In the case of EuHCit0, model 2 can be excluded, since the
DFT calculated absorption spectrum of this complex species
exhibits a strong peak at 1502 cm−1 representing a vibrational
mode of the co-ordinated hydroxyl group that is not observed in
the measured IR spectra (see ESI, Fig. S4†). Likewise, model 3
can also be ruled out, because no bands were observed in the IR
absorption spectra that can be assigned to the presence of car-
boxyl groups. Hence, out of the published structure models for
MHCit0, only model 1 remains. The calculated IR spectrum for
this complex structure is in very good agreement with the
measured spectrum at pH 2 and 4 (Fig. 6a). Both νas,s(COO)
modes and the characteristic bands in the 1300–1200 cm−1
region occur and are comparable to the measured IR spectrum.
This confirms that the Eu3+ ion is complexed according to model
1 via one terminal and the central carboxylate group in the
EuHCit0 species. Nevertheless, TRLFS results show that up to
three water molecules are replaced in the first coordination shell
of the Eu3+ ion by the citrate molecule HCit3−due to sterical
Dalton Trans., 2012, 41, 13969–13983 | 13977
 Fig. 6 Comparison of the measured (dotted line) and the calculated (continuous line) IR spectra of EuHCit0 (a) and EuCit− (b).
Fig. 7 DFT calculated solution structures of the EuHCit0 (a) and EuCit− (b) complexes.
reasons. Another possibility is a change of the total coordination
number from 9 to 8 upon complexation with citrate.
In the case of EuCit−, both models 5 and 6 can be excluded
since DFT calculations starting with these structures always con-
verge to other structure models (see ESI, Fig. S4†). Furthermore,
with respect to model 6, complexation of the Eu3+ ion by all
four functional groups is unlikely due to sterical reasons. There-
fore, out of the published structures for MCit−, only model 4
remains. The calculated absorption spectrum of this structure is
depicted in Fig. 6b and reveals very good agreement with the
measured spectrum at pH 8–12. Compared to the EuHCit0
species, the signal of the νs(COO) mode is less split in both the
calculated and the measured IR spectrum. Furthermore, in the
1300–1200 cm−1 region, the absorption band at 1280 cm−1 is
present neither in the calculated nor in the measured IR
spectrum. This verifies the postulated deprotonation of the
hydroxyl group upon complexation and the direct binding of
this functionality to the Eu3+ ion. Hence, in the EuCit−
species, the Eu3+ ion is complexed according to model 4 via
one terminal and the central carboxylate as well as the hydroxy-
late group.
Regarding the binding mode of the carboxylate groups, all
reported structure models have in common that the respective
heavy metal ion is bound in a mono-dentate fashion
(cf. Fig. 5).13,27,28,63–68 To the best of the authors’ knowledge, a
bi-dentate coordination of citrate has not yet been published.
In conclusion, IR measurements and DFT calculations provide
new aspects of the solution structure of the EuHCit0 and EuCit−
complexes formed at equimolar metal to ligand ratio. For the
EuHCit0 species, a mono-dentate binding of one terminal and
13978 | Dalton Trans., 2012, 41, 13969–13983
the central carboxylate group was found, whereas the hydroxyl
group is not coordinated to the heavy metal ion in this species.
In contrast, an additional coordination of the hydroxylate group
to the Eu3+ ion was demonstrated for the EuCit− species. The
calculated solution structures of both the EuCitH0 and EuCit−
complexes are depicted in Fig. 7.
TRLFS spectra of curium(III) citrate complexes
In analogy to the europium(III) experiments, curium(III) spectra
were recorded at a constant metal concentration of 3 × 10−7 M
and pH 2.4, 4.0, 5.5, 8.1, and 12.5 with varying ligand concen-
trations of 10−5–10−3 M (Fig. 8).
The luminescence spectrum of the Cm3+ aqua ion is character-
ized by a single emission band at 593.5 ± 0.2 nm representing
the 6D7/2 → 8S7/2 transition and a luminescence lifetime of 67 ±
1 μs equalling 9 water molecules in the first hydration
sphere.8,33,39,69 Spectroscopic alterations that indicate the com-
plexation of the Cm3+ ion were observed at each pH. In general,
the emission band is shifted to longer wavelengths. A splitting
of the luminescence band or the appearance of additional peaks
is not observed. Furthermore, the luminescence lifetime is pro-
longed compared to that of the Cm3+ aqua ion. In each sample,
we observed mono-exponential luminescence decay. All spectro-
scopic changes are summarized in Table 5.
Curium(III) spectra develop in striking analogy to those of
europium(III). Therefore, only a brief discussion is presented. At
pH 2.4 (cf. Fig. 8a), all spectra with ligand concentrations ≤7.5
× 10−5 M equal that of the uncomplexed Cm3+ aqua ion. At
This journal is © The Royal Society of Chemistry 2012
Fig. 8 Normalized steady-state luminescence spectra of 3 × 10−7 M curium(III) + citrate at pH 2.4 (a), 4.0 (b), 5.5 (c), 8.1 (d), and 12.5 (e), I = 0.1 M
(NaClO4) and room temperature in dependence on the ligand concentration as well as the deconvoluted emission spectra (f) of the single curium(III)
citrate complexes (Cm3+ without citrate = grey line).

Table 5 Luminescence spectroscopic parameters of the determined curium(III) species

pH [citrate]/M λ/nm τ/μs nH2Oa Assigned species Reference

3+
1–6 0 593.5 67 ± 1 8.9 Cm aqua ion Present work
2.4 ≥5 × 10−4 597.1 >80 <7.2 CmHCit0 Present work
4.0 <5 × 10−4 597.3 88 ± 1 6.5 CmHCit0 Present work
5.5 <2.5 × 10−4 597.2 90 ± 6 6.2 CmHCit0 Present work
4.0 >5 × 10−4 600.7 100 ± 5 5.6 Cm(HCitH)HCit2− Present work
5.5 ≥2.5 × 10−4 600.4 112 ± 5 4.9 Cm(HCitH)HCit2− Present work
8.1 >2.5 × 10−5 604.2 >205b <2.3b Cm(HCit)23− Present work
12.5 ≥5 × 10−4 607.5 155 ± 20c 3.3c CmCit25− Present work
≤5.8 — 600.1 122 ± 3 4.4 Cm(III) species in human urine 8
a
Number of water molecules calculated according to Kimura et al.35 with a standard deviation of ±0.5. b Luminescence spectra show that Cm(HCit)23−
and a hydrolyzed curium(III) species co-exist at this pH. Therefore, both species contribute to the measured lifetime. c Parameter reported with
reservations since only two analyzable spectra were measured.

higher citrate concentrations, the emission maximum is shifted to Measurements at pH 4.0 and 5.5 (cf. Fig. 8b and 8c) exhibit a
597.1 ± 0.1 nm and the lifetime is prolonged to >80 μs. This two-stage development with increasing ligand concentration
indicates the formation of a Cm(III) citrate species 1. indicating the formation of two different complex species.

This journal is © The Royal Society of Chemistry 2012 Dalton Trans., 2012, 41, 13969–13983 | 13979
Luminescence spectra at lower citrate concentrations equal those
of the Cm(III) citrate species 1 determined at pH 2.4 with an
emission maximum at 597.2 ± 0.1 nm and a lifetime of 89 ±
6 μs. According to eqn (3), this corresponds to a number of
6.4 ± 0.5 water molecules in the first hydration shell of the Cm3+
ion. Higher ligand concentrations result in a further shift of the
emission maximum to 600.5 ± 0.1 nm and a prolongation of the
luminescence lifetime to 106 ± 10 μs corresponding to 5.0 ± 0.5
water molecules. Hence, another Cm(III) citrate species 2 is
formed at both pH values and a large excess of ligand.
At pH 8.1 (cf. Fig. 8d), spectral changes that result from metal
ion hydrolysis70 occur already in the curium(III) solution without
citrate. Ligand addition results in a red shift of the luminescence
peak to 604.2 ± 0.1 nm and a prolonged emission lifetime of
>205 μs, which equals <2.3 ± 0.5 water molecules left in the
first hydration shell of the Cm3+ ion. Since all spectroscopic
parameters do not correspond to one of the complex species
determined at lower pH, this points to the formation of a new
Cm(III) citrate species 3 competing with the metal ion
hydrolysis.
In the alkaline region, at pH 12.5 (cf. Fig. 8e), no lumines-
cence spectrum was obtained in pure curium(III) solution and
after the addition of <5 × 10−4 M citrate. This results from the
formation of the dominant curium(III) species at high pH, which
is the poorly water soluble Cm(OH)3.33 At ≥7.5 × 10−4 M
citrate, luminescence is detectable again but steady-state and
especially time-resolved spectra are very noisy. However, the
emission maximum is even more red-shifted to 607.5 ± 0.1 nm
than at pH 8.1 and the luminescence lifetime is about 155 ±
20 μs corresponding to 3.3 ± 0.5 water molecules. This indicates
that Cm3+ ions are partially re-solved via complexation with
citrate and a new Cm(III) citrate species 4 is formed.
In summary, like for europium(III), four different curium(III)
citrate species were characterized by varying the citrate concen-
tration at pH 2–12.5. Identification of the complex species was
carried out in the same way as for europium(III) in consideration
of the FT-IR and DFT results and on the basis of the measured
luminescence lifetimes. Literature data on curium(III) citrate
complexes9,15,19–22,24,25 were also taken into account and
support our assignments.
All curium(III) complexes are in strikingly analogy to those of
europium(III). Thus, the Cm(III) citrate species 1 likely represents
the charge neutral 1 : 1 complex CmHCit0 with the HCit3−
anion. The emission lifetime of 89 ± 6 μs equals 2.5 ± 0.5 water
molecules that have been displaced by the ligand upon com-
plexation. This fits to the IR spectroscopic results where the
citrate anion HCit3− binds to the Cm3+ ion via two carboxylate
groups in a mono-dentate way. The Cm(III) citrate species 2 turns
out to be the 1 : 2 complex Cm(HCitH)HCit2−, in which one
citrate HCit3− and one hydrogen citrate anion HCitH2− bind to
the Cm3+ ion. The luminescence lifetime of 106 ± 10 μs indi-
cates that 5.1 ± 0.5 water molecules in the first coordination shell
of the Cm3+ ion have been displaced by the ligands. Assuming a
mono-dentate complexation as in the case of the CmHCit0
complex, this indicates the replacement of about 2.5 water mole-
cules by each ligand molecule. The Cm(III) citrate species 3 is
identified as the 1 : 2 complex Cm(HCit)23−. The luminescence
lifetime of >205 μs corresponds to a replacement of at least
6.6 ± 0.5 water molecules and supports the 1 : 2 stoichiometry.
13980 | Dalton Trans., 2012, 41, 13969–13983
The Cm(III) citrate species 4 can not be clearly identified unequi-
vocally, since only two analyzable spectra were available. Never-
theless, with regard to the europium(III) measurements, it is most
probable that this species is the 1 : 2 complex CmCit25−. As
shown in the IR and DFT section, at very alkaline pH, the com-
pletely deprotonated citrate anion Cit4− exists and binds to the
heavy metal ion. The 1 : 2 stoichiometry is supported by the
luminescence lifetime of about 155 ± 20 μs, which corresponds
to 5.6 ± 0.5 water molecules that have been displaced upon com-
plexation. However, verification of this hypothesis is still due
and no reference data are reported for curium(III) citrate com-
plexes at pH > 10, yet.
Finally, one pH-dependent series was measured with ligand
excess in the range of pH 2–13 (see ESI, Fig. S3†). All four
complex species were identified and no indications for the
hydrolysis of curium(III) were detected in the whole pH range
under investigation. Strikingly, from pH 7.0 to 10.0, the time-
resolved spectra revealed a bi-exponential luminescence decay.
The shorter lifetime equals that of the Cm(HCitH)HCit2−
species, while the longer lifetime corresponds to the Cm(HCit)23−
complex. Hence, this is direct evidence that both species exist in
a mixture. The pH-dependent measurements show that all
species subsequently convert into each other. Moreover, they
clearly demonstrate that at least two different curium(III) species
simultaneously exist at each pH > 2.5.
In conclusion, the An(III) and Ln(III) complexes predominantly
reported in the literature at the respective pH values and metal to
ligand ratios were identified for curium(III) and their spectro-
scopic parameters determined (see Table 5). The single com-
ponent spectra of all curium(III) citrate species are depicted in
Fig. 8f. To the best of our knowledge, no luminescence spectro-
scopic reference data on the complexation of curium(III) with
citrate have been published until now.
Complex stability constants and speciation of curium(III) and
europium(III) citrate complexes
Based on the identified species, the formation constants for the
direct reactions, log Kpqr ( p = number of M3+ ions, q = number
of ligand molecules, r = number of protons), and for the overall
reactions, log βpqr, were calculated according to eqn (5)–(8) and
(9)–(11), respectively. The resulting complex stability constants
log K, log K0 and log β are listed in Table 6. Calculation of the
values for CmCit25− was not successful, since only two analyz-
able spectra could be measured. The Gibbs energies at zero ionic
strength are also summarized in Table 6.
A comparison of the log K values for curium(III), americium(III),
and europium(III) citrate species determined in this study with
those reported in the literature9,13–15,17–21 is given in Fig. 9. Our
complex stability constants are in very good agreement
especially with those reported by Stepanov.19 Discrepancies with
other literature data can be attributed to the different ionic
strengths (e.g. in the case of Mathur et al.9) and the complex
species that were considered to form. In particular, the existence
of the M(HCitH)HCit2− species is still controversial in the litera-
ture. Therefore, some authors account for the formation of this
complex and some do not. This might contribute to the deviating
stability constants reported in the literature (see ESI, Table S1†).
This journal is © The Royal Society of Chemistry 2012
Table 6 Thermodynamic parameters of the curium(III) and europium(III)
citrate complexes

Species log βa log Ka log K0 b ΔRG0/kJ mol−1 c

CmHCit0 21.0 ± 0.2 7.4 ± 0.2 9.3 ± 0.2 −23.1 ± 0.2

Cm(HCitH)HCit2− 43.8 ± 0.3 11.0 ± 0.3 12.9 ± 0.3 −32.0 ± 0.3
Cm(HCit)23− 38.4 ± 0.7 11.3 ± 0.7 13.2 ± 0.7 −32.7 ± 0.7
CmCit25− — — — —
EuHCit0 21.2 ± 0.2 7.5 ± 0.2 9.4 ± 0.2 −23.3 ± 0.2
Eu(HCitH)HCit2− 43.6 ± 0.5 10.8 ± 0.5 12.7 ± 0.5 −31.6 ± 0.5
Eu(HCit)23− 38.5 ± 0.4 11.4 ± 0.4 13.2 ± 0.4 −33.0 ± 0.4
EuCit25− 21.0 ± 0.2 21.0 ± 0.2 20.9 ± 0.2 −51.8 ± 0.2
a
Complex formation constants at I = 0.1 M (NaClO4). b Complex
formation constants at zero ionic strength. c Gibbs energy at zero ionic
strength.

Fig. 10 Speciation diagram of 3 × 10−5 M europium(III) + 10−3 M

citrate in aqueous solution at I = 0.1 M (NaClO4) and room temperature
in dependence on the pH.

EuCit25−. Therefore, also the speciation diagram of curium(III) in

aqueous citrate solutions should be very similar to the presented
one of europium(III). Fig. 10 shows that all complex species
determined within this study subsequently form and convert into
each other. Furthermore, it is obvious that, except for very acidic
and alkaline pH, always a mixture of three species exists. This,
in turn, reflects the difficulty in determining the spectroscopic
properties of a single complex species and accounts for the dis-
crepancies within the reported complex stability constants
(cf. Fig. 9).

Curium(III) and europium(III) citrate speciation in biological

systems

At physiological pH 7.4, the M(HCit)23− species of curium(III)

Fig. 9 Comparison of log K values for curium(III), americium(III), and
europium(III) citrate species (filled symbol = present work, open symbol =
literature data9,13–15,17–22,24).
Within our study, we were not able to give final evidence for the
existence of the M(HCitH)HCit2− species, although this
complex was found to be the best species that fits to our experi-
mental data by now. The question, therefore, remains still open
and needs further investigation. Furthermore, a conclusion
whether curium(III) or europium(III) form stronger complexes
with citrate is not possible, since the stability constants of an
analog species are quite identical within the range of error.
Fig. 10 depicts the representative speciation diagram of
europium(III) in aqueous citrate solutions with ligand excess in
dependence on the pH, which has been calculated using the
complex stability constants determined within the present work.
Since curium(III) revealed striking analogy to europium(III)
throughout all experiments, it is reasonable to assume that the
log K120 value for CmCit25− should be comparable to those for
and europium(III) dominates the speciation to >80% (cf. Fig. 10).
Therefore, it is reasonable to assume that this complex might
also form to a certain extent in blood and other biofluids at near
neutral pH.
In the case of human urine, a previous study on the in vitro
binding form of curium(III) and europium(III) in natural samples
revealed that, at slightly acidic pH ≤ 5.7, both heavy metal ions
were, in fact, predominantly bound by citrate.8 Comparison of
the reported urinary citrate complexes with the single complex
species determined within the present work is given in Tables 1
and 5.
In the case of curium(III), it is obvious that the species formed
in human urine is the Cm(HCitH)HCit2− complex. Both the
luminescence wavelength and the emission lifetime are identical.
In contrast, for europium(III), the emission maxima and the fine
structure of the urinary spectrum fit those of the Eu(HCitH)
HCit2− complex, while the luminescence lifetime equals that of
the EuHCit0 species. Therefore, a mixture of these two com-
plexes exists in human urine in vitro at slightly acidic pH. Which
of the two species actually dominates most probably depends on
the urinary citrate concentration. Thermodynamic modeling
shows that at lower citrate concentrations, the 1 : 1 complex
EuHCit0 is formed to a higher extent, whereas at higher citrate
concentrations the 1 : 2 complex Eu(HCitH)HCit2− dominates.

This journal is © The Royal Society of Chemistry 2012 Dalton Trans., 2012, 41, 13969–13983 | 13981
 This is the first time that the molecular binding form of a triva-
lent actinide and/or lanthanide in a biological fluid is directly
determined via experiment and demonstrates the great suitability
of TRLFS and IR spectroscopy for such in situ speciation inves-
tigations. Furthermore, knowledge of the actual An(III) and
Ln(III) citrate complexes occurring in body fluids might be a pre-
requisite to develop new and specific potential decontamination
agents based on ligands naturally produced in the human body.
Conclusion
For the first time, the complexation of curium(III) and europium(III)
with citrate was investigated over a wide pH and ligand con-
centration range using the two complementary spectroscopic
techniques TRLFS and IR as well as DFT calculations. In the
case of both elements, four complex species were identified with
TRLFS, out of which, for europium(III), two of them have been
characterized in more detail with ATR-FT-IR and DFT.
The spectral emission features of four distinct curium(III) and
europium(III) complex species were assigned to MHCit0,
M(HCitH)HCit2−, M(HCit)25− and MCit25−. We demonstrated
that, on the one hand, all species can be discriminated and ident-
ified with regard to these features but, on the other hand, that an
exact characterization is hindered by the simultaneous formation
of several complexes in solution. Furthermore, the lifetimes
measured with TRLFS give information about the number of
water molecules in the first coordination shell of the Cm3+ and
Eu3+ ion, respectively, which thereby give how many waters
have been displaced by ligand molecules upon complexation.
This allows first estimations of the structure of the respective
complex species but detailed structural information, such as the
functional groups of the ligand participating in the curium(III)
and europium(III) complexation, respectively, as well as the
binding mode, cannot be obtained with TRLFS.
IR spectroscopy combined with DFT calculation, in turn,
provide information on the mode of coordination of functional
groups to the metal ion. Thus, the two techniques are comp-
lementary to TRLFS. We demonstrated that in the EuHCit0
species, formed at acidic pH, only carboxylate groups bind the
Eu3+ ion, whereas in the EuCit− complex, formed at alkaline pH,
additionally also the hydroxylate group is involved. Furthermore,
comparing IR with TRLFS spectra we showed that, due to steri-
cal reasons, each functional group binding to the Eu3+ ion
replaces more than one water molecule from its first hydration
shell. Moreover, DFT calculations aided us to interpret the
measured IR spectra correctly and provided a basis to decide
about the actual complex solution structure. Therefore, the
results obtained with each of the three techniques are in good
agreement with and verify each other.
In summary, the present work highlights the importance of
citrate in the complexation of An(III) and Ln(III) in general and
curium(III) and europium(III) in particular. Furthermore it indi-
cates the crucial role of citrate for their speciation in human
body fluids, like urine and blood, since the dominant species at
physiological pH is the strongly negatively charged M(HCit)23−.
Further measurements of the curium(III) and europium(III) specia-
tion in natural biofluids (e.g. blood and saliva) as well as in
model solutions (e.g. for different fluids of the human
13982 | Dalton Trans., 2012, 41, 13969–13983
gastrointestinal tract) are in progress and shall emphasize the
recent results and assumptions.
Acknowledgements
This work was funded by the Deutsche Forschungsgemeinschaft
under Contract No. BE 2234/10-1/2. We are indebted to the U.S.
Department of Energy, Office of Basic Energy Sciences, for the
use of 248Cm via the transplutonium element production facilities
at Oak Ridge National Laboratory, which was made available as
part of a collaboration between the Helmholtz-Zentrum Dresden-
Rossendorf and the Lawrence Berkeley National Laboratory. We
acknowledge the Zentrum für Informationsdienste und Hochleis-
tungsrechnen, Technische Universität Dresden, Germany, for use
of their supercomputers.
Notes and references
1 J.-C. G. Bünzli, Nat. Chem., 2010, 2, 696; S. V. Eliseeva and
J.-C. G. Bünzli, Chem. Soc. Rev., 2010, 39, 189–227; H. Elsner, Com-
modity Top News, 2011, 36, 1–8; B. Karn, J. Environ. Monit., 2011, 13,
1184–1189; M. Liedtke and H. Elsner, Commodity Top News, 2010, 31,
1–6.
2 M. F. Bellin and A. J. Van Der Molen, Eur. J. Radiol., 2008, 66,
160–167; A. Thibon and V. C. Pierre, Anal. Bioanal. Chem., 2009, 394,
107–120; G. Muller, Dalton Trans., 2009, 9692–9707.
3 E. Ansoborlo, O. Prat, P. Moisy, C. Den Auwer, P. Guilbaud, M. Carriere,
B. Gouget, J. Duffield, D. Doizi, T. Vercouter, C. Moulin and V. Moulin,
Biochimie, 2006, 88, 1605–1618.
4 P. W. Durbin, in Uranium, Plutonium, Transplutonic Elements, ed.
H. C. Hodge, J. N. Stannard and J. B. Hursh, Springer Verlag, Berlin,
1973, pp. 739–896; D. M. Taylor and R. W. Leggett, Radiat. Prot.
Dosim., 2003, 105, 193–198; F. Ménétrier, D. M. Taylor and A. Comte,
Appl. Radiat. Isot., 2008, 66, 632–647.
5 J. R. Duffield, D. M. Taylor and D. R. Williams, in Handbook on the
Physics and Chemistry of Rare Earths, Vol. 18: Lanthanides/Actinides:
Chemistry, ed. K. A. Gschneidner Jr., L. Eyring, G. R. Choppin and
G. H. Lander, Elsevier Science B. V., Amsterdam, 1994, pp. 591–621.
6 G. A. Turner and D. M. Taylor, Phys. Med. Biol., 1968, 13, 535–546;
J. R. Cooper and H. S. Gowing, Int. J. Radiat. Biol., 1981, 40, 569–572;
D. M. Taylor, J. Alloys Compd., 1998, 271, 6–10; D. M. Taylor,
J. R. Duffield, D. R. Williams, L. Yule, P. W. Gaskin and P. Unalkat,
Eur. J. Solid State Inorg. Chem., 1991, 28, 271–274.
7 G. N. Stradling, D. S. Popplewell and G. J. Ham, Health Phys., 1976, 31,
517–519; D. S. Popplewell, G. N. Stradling and G. J. Ham, Radiat. Res.,
1975, 62, 513–519.
8 A. Heller, A. Barkleit and G. Bernhard, Chem. Res. Toxicol., 2011, 24,
193–203.
9 J. N. Mathur, K. Cernochova and G. R. Choppin, Inorg. Chim. Acta,
2007, 360, 1785–1791.
10 W. M. Boothby and M. Adams, Am. J. Physiol., 1934, 107, 471–479;
A. C. Kuyper and H. A. Mattill, J. Biol. Chem., 1933, 103, 51–60;
C. C. Sherman, L. B. Mendel, A. H. Smith and M. C. Toothill, J. Biol.
Chem., 1936, 113, 247–263.
11 A. M. N. Silva, X. L. Kong and R. C. Hider, Biometals, 2009, 22,
771–778.
12 W. Hummel, G. Anderegg, L. Rao, I. Puigdomènech and O. Tochiyama,
Chemical Thermodynamics, Vol. 9: Chemical Thermodynamics of Com-
pounds and Complexes of U, Np, Pu, Am, Tc, Se, Ni and Zr with Selected
Organic Ligands, Elsevier B. V., Amsterdam, 2005.
13 S. H. Eberle and F. Moattar, Inorg. Nucl. Chem. Lett., 1972, 8, 265–270.
14 E. Ohyoshi and A. Ohyoshi, J. Inorg. Nucl. Chem., 1971, 33,
4265–4273.
15 R. Guillaumont and L. Bourderie, Bull. Soc. Chim. Fr., 1971, 8,
2806–2809.
16 P. G. Daniele, A. Derobertis, C. Rigano and S. Sammartano, Ann. Chim.,
1985, 75, 115–120.
17 S. Bouhlassa and R. Guillaumont, J. Less Common Met., 1984, 99,
157–171.
This journal is © The Royal Society of Chemistry 2012
18 S. Hubert, M. Hussonno, L. Brillard, G. Goby and R. Guillaumont,
J. Inorg. Nucl. Chem., 1974, 36, 2361–2366.
19 A. V. Stepanov, Zh. Neorg. Khim., 1971, 16, 2981–2985.
20 H. Itoh, M. Fujisawa, Y. Ikegami and Y. Suzuki, Lanthanide Actinide
Res., 1985, 1, 79–88.
21 A. V. Stepanov, Radiokhimiya, 1959, 1, 668–673.
22 S. Hubert, M. Hussonnois and R. Guillaumont, J. Inorg. Nucl. Chem.,
1973, 35.
23 S. Bouhlassa, M. Petitramel and R. Guillaumont, Bull. Soc. Chim. Fr.,
1984, 5–11.
24 P. Thakur, Y. Xiong, M. Borkowski and G. R. Choppin, Radiochim. Acta,
2012, 100, 1–8.
25 B. S. Jensen and H. Jensen, Radiochim. Acta, 1988, 44/45, 45–49.
26 A. Moutte and R. Guillaumont, Rev. Chim. Miner., 1969, 6, 603–610.
27 A. P. G. Kieboom, J. M. Vandertoorn, J. A. Peters, W. Bovee,
A. Sinnema, C. A. M. Vijverberg and H. Vanbekkum, Recl. Trav. Chim.
Pays-Bas, 1978, 97, 247–248; A. P. G. Kieboom, C. A. M. Vijverberg,
J. A. Peters and H. Vanbekkum, Recl. Trav. Chim. Pays-Bas, 1977, 96,
315–316; C. A. M. Vijverberg, J. A. Peters, A. P. G. Kieboom and
H. Vanbekkum, Tetrahedron, 1986, 42, 167–174.
28 C. B. Yuan, J. Z. Liu, D. Q. Zhao, Y. J. Wu and J. Z. Ni, Polyhedron,
1995, 14, 3579–3583.
29 G. E. Jackson and J. Dutoit, J. Chem. Soc., Dalton Trans., 1991,
1463–1466.
30 E. Ansoborlo, L. Bion, D. Doizi, C. Moulin, V. Lourenco, C. Madic,
G. Cote, J. Van der Lee and V. Moulin, Radiat. Prot. Dosim., 2007, 1–6.
31 A. Barkleit, G. Geipel, M. Acker, S. Taut and G. Bernhard, Spectrochim.
Acta A Mol. Biomol. Spectros., 2011, 78, 549–552.
32 Lanthanide Probes in Life, Chemical and Earth Sciences: Theory and
Practice, ed. J.-C. G. Bünzli and G. R. Choppin, Elsevier Science B. V.,
Amsterdam, 1989.
33 H. Moll, V. Brendler and G. Bernhard, Radiochim. Acta, 2011, 99,
775–782.
34 R. N. Collins, T. Saito, N. Aoyagi, T. E. Payne, T. Kimura and
T. D. Waite, J. Environ. Qual., 2011, 40, 731–741; T. E. Keyes,
R. J. Forster and C. Blackledge, Spectrosc. Prop. Inorg. Organomet.
Compd., 2010, 41, 211–261.
35 T. Kimura and G. R. Choppin, J. Alloys Compd., 1994, 213, 313–317;
T. Kimura, G. R. Choppin, Y. Kato and Z. Yoshida, Radiochim. Acta,
1996, 72, 61–64; T. Kimura and Y. Kato, J. Alloys Compd., 1998, 277,
806–810.
36 H. Moll, A. Johnsson, M. Schäfer, K. Pedersen, H. Budzikiewicz and
G. Bernhard, Biometals, 2008, 21, 219–228.
37 K. Müller, H. Foerstendorf, S. Tsushima, V. Brendler and G. Bernhard,
J. Phys. Chem. A, 2009, 113, 6626–6632; B. Li, J. Raff, A. Barkleit,
G. Bernhard and H. Foerstendorf, J. Inorg. Biochem., 2008, 104, 718–725.
38 Origin Lab Corporation, Northhampton, 2006.
39 J. I. Kim, R. Klenze and H. Wimmer, Eur. J. Solid State Inorg. Chem.,
1991, 28, 347–356.
40 J. V. Beitz, in Handbook on the Physics and Chemistry of Rare Earths,
Vol. 18: Lanthanides/Actinides: Chemistry, ed. K. A. Gschneidner Jr,
L. Eyring, G. R. Choppin and G. H. Lander, Elsevier Science B. V.,
Amsterdam, 1994, pp. 159–196; W. D. Horrocks Jr. and M. Albin, in
Progress in Inorganic Chemistry, ed. S. J. Lippard, John Wiley Hoboken,
1984, vol. 31, pp. 1–104.
41 R. A. Binstead, A. D. Zuberbühler and B. H. Jung, Spectrum Software
Associates, 2004.
42 P. Gans, A. Sabatini and A. Vacca, Protonic Software, Leeds, 2008.
43 A. Heller, A. Barkleit, G. Bernhard and J.-U. Ackermann, Inorg. Chim.
Acta, 2009, 362, 1215–1222; A. Heller, O. Rönitz, A. Barkleit,
G. Bernhard and J.-U. Ackermann, Appl. Spectrosc., 2010, 64, 930–935;
M. Glorius, H. Moll and G. Bernhard, Polyhedron, 2008, 27, 2113–2118.
44 W. Hummel, U. Berner, E. Curti, F. J. Pearson and T. Thoenen, Nagra/PSI
Chemical Thermodynamic Data Base 01/01, Nagra, Wettingen, 2002;
R. Guillaumont, T. Fanghänel, J. Fuger, I. Grenthe, V. Neck, D. A. Palmer
and M. H. Rand, Chemical Thermodynamics, Vol. 5: Update on the
Chemical Thermodynamics of Uranium, Neptunium, Plutonium,
Americium and Technetium, Elsevier Science B. V., Amsterdam, 2003.
45 L. D. Pettit, I. Puigdomenech, H. Wanner, I. Sukhno and V. Buzko,
Academic Software http://www.acadsoft.co.uk, 2004.
This journal is © The Royal Society of Chemistry 2012
View publication stats
46 T. Vercouter, B. Amekraz, C. Moulin, E. Giffaut and P. Vitorge, Inorg.
Chem., 2005, 44, 7570–7581.
47 C. De Stefano, C. Foti, O. Giuffre and S. Sammartano, J. Chem. Eng.
Data, 2001, 46, 1417–1424.
48 R. J. Silva, G. Bidoglio, M. H. Rand, P. B. Robouch, H. Wanner and
I. Puigdomènech, Chemical Thermodynamics, Vol. 2. Chemical Thermo-
dynamics of Americium, Elsevier Science B. V., Amsterdam, 1995.
49 P. Gans, A. Sabatini and A. Vacca, Protonic Software, Leeds, 2006.
50 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin,
J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone,
B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson,
H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li,
J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo,
J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma,
G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich,
A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul,
S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-
Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill,
B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople,
Gaussian 03, Revision E.01, Gaussian, Inc., Wallingford CT, 2004..
51 V. Barone and M. Cossi, J. Phys. Chem. A, 1998, 102, 1995–2001.
52 A. Bondi, J. Phys. Chem., 1964, 68, 441–451.
53 M. Dolg, H. Stoll, A. Savin and H. Preuss, Theor. Chim. Acta, 1989, 75,
173–194.
54 R. Krishnan, J. S. Binkley, R. Seeger and J. A. Pople, J. Chem. Phys.,
1980, 72, 650–654.
55 A. Barkleit, S. Tsushima, O. Savchuk, J. Philipp, K. Heim, M. Acker,
S. Taut and K. Fahmy, Inorg. Chem., 2011, 50, 5451–5459.
56 C. Moulin, J. Wei, P. Van Iseghem, I. Laszak, G. Plancque and
V. Moulin, Anal. Chim. Acta, 1999, 396, 253–261.
57 G. Plancque, V. Moulin, P. Toulhoat and C. Moulin, Anal. Chim. Acta,
2003, 478, 11–22.
58 C. M. Andolina, R. A. Mathews and J. R. Morrow, Helv. Chim. Acta,
2009, 92, 2330–2348.
59 J.-C. G. Bünzli, in Metal Ions in Biological Systems, Vol. 42: Metal Com-
plexes in Tumor Diagnosis and as Anticancer Agents, ed. A. Sigel and
H. Sigel, Marcel Dekker Inc., New York, 2004, pp. 39–75.
60 M. Lindegren, J. S. Loring and P. Persson, Langmuir, 2009, 25,
10639–10647; S. P. Pasilis and J. E. Pemberton, Geochim. Cosmochim.
Acta, 2008, 72, 277–287; K. Lackovic, B. B. Johnson, M. J. Angove and
J. D. Wells, J. Colloid Interface Sci., 2003, 267, 49–59; S. P. Pasilis and
J. E. Pemberton, Inorg. Chem., 2003, 42, 6793–6800.
61 G. B. Deacon and R. J. Phillips, Coord. Chem. Rev., 1980, 33, 227–250;
M. Kakihana, T. Nagumo, M. Okamoto and H. Kakihana, J. Phys.
Chem., 1987, 91, 6128–6136.
62 G. Montavon, C. Hennig, P. Janvier and B. Grambow, J. Colloid Interface
Sci., 2006, 300, 482–490.
63 K. S. Rajan and A. E. Martell, Inorg. Chem., 1965, 4, 462–469.
64 Y. Z. Hamada, B. L. Carlson and J. T. Shank, Synth. React. Inorg.
Met.-Org. Chem., 2003, 33, 1425–1440.
65 A. M. N. Silva, X. Kong, M. C. Parkin, R. Cammack and R. C. Hider,
Dalton Trans., 2009, 8616–8625.
66 W. H. Kuan, M. K. Wang, P. M. Huang, C. W. Wu, C. M. Chang and
S. L. Wang, Water Res., 2005, 39, 3457–3466; R. J. Motekaitis and
A. E. Martell, Inorg. Chem., 1984, 23, 18–23.
67 M. Clausen, L. O. Ohman and P. Persson, J. Inorg. Biochem., 2005, 99,
716–726.
68 T. Toraishi, S. Nagasaki and S. Tanaka, J. Mol. Struct. (THEOCHEM),
2005, 757, 87–97.
69 N. M. Edelstein, R. Klenze, T. Fanghänel and S. Hubert, Coord. Chem.
Rev., 2006, 250, 948–973.
70 H. Wimmer, R. Klenze and J. I. Kim, Radiochim. Acta, 1992, 56, 79–83;
T. Fanghänel, J. I. Kim, P. Paviet, R. Klenze and W. Hauser, Radiochim.
Acta, 1994, 66–7, 81–87; T. Rabung, M. Altmaier, V. Neck and
T. Fanghänel, Radiochim. Acta, 2008, 96, 551–559.
Dalton Trans., 2012, 41, 13969–13983 | 13983