Talanta 185 (2018) 586–591

Contents lists available at ScienceDirect

Talanta
journal homepage: www.elsevier.com/locate/talanta

Separation and preconcentration of actinides from concentrated nitric acid T

by extraction chromatography in microsystems
Marion Losnoa, Julien Pelléa, Mylène Mariea, Ivan Ferranteb, René Brennetota,
⁎
Stéphanie Descroixb, Clarisse Marieta,
a
Den - Service d′Etudes Analytiques et de Réactivité des Surfaces (SEARS), CEA, Université Paris-Saclay, F-91191 Gif sur Yvette, France
b
MMBM Group, Institut Curie Research Center, CNRS UMR 168, Paris, France

A R T I C LE I N FO A B S T R A C T

Keywords: An original method of monolith impregnation in microsystem for the analysis of radionuclides in nitric acid is
Microfluidics reported. Three microcolumns made of monolith poly(AMA-co-EDMA) were impregnated in COC microsystems.
Radiochemistry The robustness of the microsystems in nitric acid media until 8 M was demonstrated. High exchange capacity
Monolithic stationary phase and affinity for tetravalent and hexavalent actinides in concentrated nitric media were obtained. The retention
TBP
characteristics of the microcolumns impregnated by TBP, TBP-CMPO and DAAP were compared with those of
DAAP
Nuclear spent fuel
the equivalent commercial particulate resins TBP™, TRU™ and UTEVA™ respectively. The separation of U, Th
and Eu was validated in a classical microsystem and a procedure is proposed in a centrifugal microsystem.

1. Introduction An attractive option consists in taking advantages of all the

Increased public attention to radioactive waste and potential public
health effects of release of radioactive materials to the environment has
made accurate and reliable methods for actinide determinations in
various environmental and biological samples increasingly important.
Some methods of preconcentrating the actinides and of isolating them
from both the large quantities of inactive substances present and po-
tential interfering species must usually precede the determination.
Extraction chromatography plays a prominent role in radiochemical
analysis by providing an easy way to preconcentrate and even separate
a large number of radionuclides, particularly actinides and selected
fission products, from a wide variety of sample matrices [1–3], notably
UTEVA™, TRU™ and TBP™ columns containing impregnated particulate
resins [4,5] (Table 1).
The miniaturization of the analytical procedures in the nuclear field
is a way forward to reduce both risk and nuisances related to radio-
active samples, which are direct functions of involved quantities of
samples [10]. The major advantage of the miniaturization is the sig-
nificant reduction of volume (or quantity) of samples, reagents and
waste as well as time of analysis. Microsystems are strong parallel
processing tools and present a noteworthy ability to automation. The
transfer of liquid chromatography onto microfluidic format can be ad-
dressed as a simple miniaturization challenge. There are very few
chromatographic microsystems for analysis of radionuclides in nuclear
samples.
knowledge on particulate resins by packing of conventional particles in
a trapping zone in microchannel using weirs [11–14]. Packing has been
achieved at low pressure in PDMS chips [11,15–17], at high pressure in
the HPLC-Chip [17], using electroosmotic flow [12,13] or even with
centrifugal force [18,19]. However, two major issues may arise from
the packing technology: reproducibility of the packing process and
particle immobilization [20,21]. Gao et al. [22] filled the channel of a
microsystem by a commercial resin. The open microchannel was filled
with particulate phases that were held in place thanks to sintered at
both ends of the microchannel. The chip has two complementary col-
umns; an anion exchange column allows to isolate Th from a nitric
solution containing traces of transition metals while the chromato-
graphic extraction column allows to isolate U.
A second option is based on the use of monoliths as stationary
phases in miniaturized format. Their synthesis, based on the photo-
polymerization, can easily be conducted in-situ and the resulting con-
tinuous bed exhibits a bimodal structure with large through-pores for
flow and small pores for retention [23]. Many reviews are available,
focusing on the study of monoliths in capillary format [24–27]. The first
syntheses of organic monoliths were carried out in glass chips [28–30].
However, chips fabricated in synthetic polymers reflects the require-
ment of cheap, disposable and versatile devices. Bruchet et al. [31]
have developed a centrifugal microsystem for U(VI)/Eu (III) separation
in [HCl] = 9.5 M. The stationary phase was a methacrylate monolith
functionalized with an anion exchanger integrated in a cyclic olefin

⁎
Corresponding author.
E-mail address: clarisse.mariet@cea.fr (C. Mariet).

https://doi.org/10.1016/j.talanta.2018.04.036
Received 11 January 2018; Received in revised form 28 March 2018; Accepted 8 April 2018
Available online 10 April 2018
0039-9140/ © 2018 Published by Elsevier B.V.
M. Losno et al.
Table 1
Characteristics of three commercial impregnated columns. (Triskem resins).
Resin TRUTM (TRansUraniens) [6,7]
Substrate Amberchrom CG-71
Impregnated extractant
76% TBP and
N,N-Diisobutyl Carbamoyl Methyl Phosphine Oxide (CMPO)
Exchange capacity 7 mg N d/g of resin
copolymer (COC) chip. However, this functionalized monolith does not
withstand nitric acid.
To our knowledge, these two studies are the only examples con-
cerning the chromatographic miniaturization for the selective pur-
ification of actinide of nuclear samples in concentrated acidic medium.
Bruchet et al. [31] chose to make a centrifugal microsystem integrating
a specific stationary phase in order to increase the analytical
throughput using parallelization proposed by Madou et al. [32]. They
are similar but present different technological choices, while Gao et al.
[22] have preferred the coupling of two versatile micro-columns with
the detection system.
Along this study, we describe the development of an original
method for monolith impregnation in microsystems for the analysis of
radionuclides in nitric acid. The studied extractants are similar than
those of commercial columns presented Table 1 in order to properly
compare the performance of the microcolumns. Retention properties of
the monolith impregnated for actinides and lanthanides have been
studied for three different extractants in microsystem. The influence of
the microcolumn dimensions has been highlighted using two different
geometries.
2. Experimental
2.1. Chemicals and reagents
Ethylene glycol dimethacrylate (97%, EDMA), allyl methacrylate
(98%, AMA), 2,2-dimethoxy-2-phenylacetophenone (> 99%, DMPA),
benzophenone (> 99%, BP), 1,4-butanediol (99%), 1-propanol
(99,7%), tri-n-butyl phosphate (98%, TBP), diamyl Amyl phosphonate,
(> 99%, DAAP), methanol (HPLC-grade), ethanol (HPLC-grade), H2O2
30 wt% and HNO3 65 wt% and Red dye Azorubin corresponding to
E122 dye were purchased from Sigma Aldrich (Isle-D′Abeau, France).
Uranyl nitrate hexahydrate (analytical grade) was purchased from
Merck (Fontenay sous bois, France). All aqueous solutions were pre-
pared using > 18 MΩ DI water (Direct-Q UV3, Millipore). Microsystem
was built with a 381 µm thick film of COC 6013 and COC 6013 pellets
purchased from Topas (Polyplastics Co, Germany). Certified stock
multi-elemental 1000 mg g−1 solutions SPEX, (Jobin Yvon, Palaiseau,
France) of Eu, Nd, Th and U were used. The samples were daily pre-
pared by diluting each analytes solutions into concentrated HNO3 so-
lutions.
Pu certified solution consists of 97.7% of 239Pu, 2.2% of 240Pu with
traces of other isotopes in nitric acid at (5337 ± 10) µg g−1. A valence
cycle on was performed on a solution of [Pu] = 44.5 µg g−1 to bring all
the Pu under the oxidation state IV [33–35]. This cycle was made with a
Talanta 185 (2018) 586–591
UTEVATM [7,8] TBP [9]
Amberchrom CG-71 Amberlite XAD-7
Diamyl amyl
24% octyl(phenyl)-
phosphate (DAAP) Tributylphosphate (TBP)
100 mg U/g of resin 75 mg U/g of resin
Fig. 1. (a) Picture and (b) scheme of P1 mi-
crosystem with Vfree = (12.2 ± 0.7) µL.
solution of H2O2 to 0.3% diluted in [HNO3] = 6 M. After heat to dry the
sample of Pu, the residue is picked up by 250 µL of the solution of H2O2
[36]. After 5 min, 250 µL were heat to dryness. Then, the residue was
dissolved in a proper volume of nitric acid to get a Pu (IV) standard
solution at 4 µg g−1. 200 µL of this solution were used for each de-
termination of the distribution coefficient in microsystem which re-
presents approximately 1 µg of Pu.
2.2. Experimental set up
2.2.1. Designs of the microsystems
Two prototypes of microsystems have been designed: the P1 mi-
crosystem and a centrifugal one. They are presented with their di-
mensions Figs. 1 and 2, respectively. The P1 microsystem, with capil-
lary connectors, allows easier development and characterizations
during the development stage. The injection of liquids and flow control
are made with syringe-pumps (KD Scientific, KDS-100). The centrifugal
microsystem was developed to test the feasibility of syntheses, func-
tions and separations in this geometry. Microfluidic channels and re-
servoirs were designed to enable an easy integration of the separation
column (synthesis and functionalization of the monolith (step 1)) and a
reproducible and accurate flow handling through centrifugation (step
2). Fig. 2 shows a detailed view of the microchip design where E, C and
R stand respectively for access ports, channels and reservoirs. The di-
mensions of R1 reservoir and B channel were depend of the targeted
ion-exchange separation. C channel is a restriction channel expected to
reinforce the mechanical stability of the column. D and E are venting
channels (used only during centrifuge experiments). Dimensions of R2
reservoir were designed to fit with standard chromatographic connec-
tions (Upchurch Scientific).
2.2.2. Microfabrication
Microchips were fabricated in COC thermoplastic because of its high
chemical resistance (to organic and aqueous solutions even for highly
acid solutions and irradiation), its high mechanical stability (the ma-
terial withstands high temperature and pressure) and its good optical
transmittance at 365 nm [37] used for synthesis of the monolithic
column. The COC microdevices fabrication process by thermoforming
was described previously [31,38]. Around 10 g of COC were used to
fabricate P1 or centrifugal microsystem.
2.2.3. Integration of the functionalized column
The microchannel was first functionalized by benzophenone (BP)
method described in literature [39,40]. Then, the monolith was syn-
thetized in the microsystem using a polymerization mixture (1 wt%
587
M. Losno et al.
Fig. 2. Scheme of the centrifugal system with Vfree = (0.27 ± 0.03) µL and dimensions ((A) Injection hole (R1) Injection reservoir, (B) Monolithic column, (C)
Restriction channel, (R2) Collection reservoir, (D) and (E) blowholes) Ø is diameter.
water, 24 wt% 1,4-butanediol, 35 wt% 1-propanol, 24 wt% EDMA, 16
wt% AMA, 1 wt% DMPA) according to the protocol previously de-
scribed [38]. The microchannel was flushed with the polymerization
mixture at 1.2 mL h−1 for 15 min before UV exposition. Then, the
stationary phase in the microchannel was achieved by photo-induced
free radical polymerization at 365 nm of a monolithic copolymer of
EDMA and AMA. Then the microcolumn was flushed at 0.6 mL h−1
ethanol during 5 min.
2.2.4. Functionalization of the monolith by impregnation in microsystem
Impregnation of the poly(AMA-co-EDMA) monolith has been ap-
plied with the TBP extractant in P1 microsystem, and then in the cen-
trifugal microsystem. After optimization of the procedure, it has been
applied for the sorption of the DAAP extractant and a CMPO/TBP (24/
76% v/v) mixture. For impregnation of the monolith in P1 micro-
system, the extractant was injected at 5 µL min−1 during 2 h and then
rinsed 7 min with water at 5 µL min−1. In centrifugal microsystem, the
protocol was the same as for the P1 microsystem using the centrifugal
platform at a speed of 1200 rpm for the introduction and the flow of the
liquids from the injection reservoir to the collection one.
2.2.5. Characterizations
The infrared spectrophotometer used for the characterization by
Fourier transform infrared spectroscopy with attenuated total re-
flectance (FTIR-ATR) of functionalization is a Bruker Equinox 55 S with
a Golden Gate™ Specac and a deuterated triglycerin sulfate detector.
Spectrum was acquired between 5800 and 550 cm−1 with a 2 cm−1
resolution thanks 32 scans through OPUS software, by placing a frag-
ment of functionalized monolith on the FTIR-ATR germanium crystal.
No other sample preparation is required. After cleaning the germanium
crystal from any residues, air spectrum is acquired as background
signal.
Exchange capacity measurements were performed directly in the
microsystems. The flow rate was set at 2 µL min−1 for all the experi-
ments. First, the monolith was conditioned with a 3 M nitric acid so-
lution during 15 min. Then, a solution of 15 000 µg g−1 of uranium in
[HNO3] = 3 M was introduced during 2 h. The column was rinsed 30
min by [HNO3] = 3 M. Lastly, the elution of uranium was performed by
[HNO3] = 0.1 M acid during 2 h. All the fractions were analyzed via
ICP-MS measurement (7700x Agilent Technologies).
For determination of the distribution coefficients, the microsystem
was first conditioned 15 min to 5 µL min−1 by a solution of nitric acid
to the desired acidity. 200 µL of multi-elementary solution containing
each element U (VI), Th (IV), Eu (III) and Nd (III) at a concentration of
588
Talanta 185 (2018) 586–591
10 µg g−1 was injected into the microsystem at 1.67 µL min−1. Tests
were first performed at batch scale to control the behavior of the
monolith before establishing the final protocol in microsystem.
Experimental protocol was determined to keep the same time and the
same volume of liquid phase over solid phase weight ratio as in a classic
batch study. The eluate was then analyzed by ICP-MS for each acidity
for several elution times. After 2 h, monolithic column in microsystem
was at steady state. The same protocol was then directed to a not
functionalized monolith in order to measure a blank. In the case of Pu
(IV), the experiment was conducted in the same way in glove box for
the microsystems impregnated by the TBP with [Pu (IV)] = 4 µg g−1.
Distribution coefficients were measured 3 times for each extractant at
each acidity.
2.3. Separation procedures
2.3.1. Separation procedure in P1 microsystem
First, the separation U(VI)/Th(IV)/Eu(III) was made on the TBP
functionalized monolith in P1 microsystem prototype before im-
plementation in centrifugal microsystem. The different elution media
were determined through the distribution coefficients obtained for
these elements. The column was conditioned with 30 µL of HNO3 5 M to
prepare the medium to the sample. The sample was diluted in 5 M nitric
acid to set U(VI) and Th(IV) on the column and eluted Eu(III). Then a
solution [HNO3] = 2 M allowed the elution of Th(IV) followed by the
elution of U(VI) with 0.1 M nitric acid. In order to compare the results
obtained in P1 microsystem and in centrifugal one, separations were
followed in number of free column volumes of mobile phase through
the column. The free column volume of each synthesized in P1 column
was determined thanks to porosity measurement [38] to average
(3.6 ± 0.5) µL. This value is then added the extra-column volume to get
the total free column volume of the P1 system equal to (12.2 ± 0.7) µL.
Each 2 total void volume fractions were analyzed by ICP-MS.
2.3.2. Separation procedure using the centrifugal microfluidic platform
The separation procedure is detailed in Table 2. The column was
conditioned with 2 µL of HNO3 5 M to prepare the medium to the
sample. The injection volume was defined at 5 µL.
2.4. Detection of europium, thorium, plutonium and uranium with ICP-MS
Inductively coupled plasma mass spectrometry (ICP-MS) measure-
ments have been performed using a quadrupole ICP-MS spectrometer
7700x (Agilent Technologies) equipped with a concentric nebulizer.
M. Losno et al.
Table 2
Procedure for extraction chromatographic separation of europium, uranium
using the centrifugal microfluidic platform.
Fraction ω (rpm) t (min) [HNO3] (M) Vfraction (µL)
Washing HNO3 0.1 M 1400 5 0.1 2 Exchange capacity mg
Conditioning HNO3 5 M 1400 10 5 2
Loading Sample 1400 27 5 5.5
Eluting 1 (Eu) 1450 21 5 6.25
2 (Th) 1400 24 2 7.5
3 (U) 1400 20 0.1 7.5
Instrumental parameters were fixed as indicated Table 2. For ICP-MS
method, calibration solutions were prepared from certified stock multi-
elemental 1000 mg g−1 SPEX solutions, (Horiba Jobin Yvon, France).
Analytical calibration standards were prepared daily over the range of
0–20 ng g−1 for Eu, Th, Pu and U by suitable serial dilutions of multi-
element stock solution in 2% (v/v) HNO3. 158Gd and 209Bi were used as
internal standards at the concentration of 2 ng g−1. The internal stan-
dards were diluted from 1000 mg g−1 stock solution. Sensitivity max-
imization and short-term stability test were performed on a daily basis
using a 1 ng g−1 solution of U and In.
3. Results and discussion
Our objective was to develop extraction chromatographic micro-
columns in microsystems for radionuclides separation in nitric acid
media. The impregnation was chosen as a versatile functionalization
method because of the selectivity provided by the extractant. After a
first validation of the impregnation of the monolith in the microsystem
P1, the versatility of the method has been illustrated for two others
extractants (the DAAP and the 76% TBP/24% CMPO mixture). Hence,
three microcolumns were developed and characterized. The affinity of
the microcolumns for the elements U(VI), Pu(IV), Th(IV), Eu(III) and
Nd(III) was measured in order to compare the performance of the mi-
crocolumns with those of the three corresponding commercial resins
presented Table 1.
3.1. Validation and versatility of the impregnation in microsystem
In literature, two methods are used for impregnation of particulate
stationary phases: the dry and the wet one. Although the dry way is the
more widely used, particularly in the case of the columns developed by
Horwitz et al. [2,3,7], this technique is hardly transposed to micro-
system in which drying could generate a degradation of the monolith.
Then, the wet method was used to impregnate the poly(AMA-co-EDMA)
monolith by the TBP extractant in P1 microsystem. The efficiency of
impregnation was demonstrated by FTIR-ATR characterization (see
Supplementary material) and by the measure of the exchange capacity
of the impregnated monolith thanks to the Eq. (1):
mU
CE =
mmonolith
where CE is the exchange capacity (mg U/g of monolith), mU is the mass
of uranium(VI) recovered from the elution step (mg) and mmonolith is the
monolith weight (g). For a non-impregnated monolith, a measure of the
CE was realized to obtain a blank of experiment.
An exchange capacity of (153 ± 18) mg U/g of monolith was ob-
tained validating the wet method of impregnation in microsystem. This
value is about two times higher than that of the commercial TBP™
column [9] equal to 75 mg U/g (Table 1). To investigate the versatility
of this approach, the same protocol of impregnation was applied to the
monolith for the extractants CMPO/TBP and DAAP and their exchange
capacity values were also evaluated (Table 3) and compared with those
of the corresponding commercial resins (Table 1). Once again, the ex-
change capacities of the impregnated monoliths surpass the equivalent
Talanta 185 (2018) 586–591
Table 3
Exchange capacity values in P1 microsystem for the three tested extractants.
Non impregnated CMPO/TBP DAAP TBP
monolith
– 149 ± 18 280 ± 33 153 ± 18
U/g of resin
Permeability (10−14 m2) 8±5 – 8±3 7±4
commercial particulate column. Compare to particulate column the
increase of the exchange capacity observed could be explained by the
larger specific surface area of monoliths [41] which extends the contact
surface during both the impregnation and the adsorption step. In ad-
dition, the specific high permeability of the monoliths compared to
particulate in the order of 10−15 m2 columns [37] allows easy move-
ment of fluids and so better contact with the surface of the impregnated
monolith.
3.2. Determination of the distribution coefficients of the monoliths
impregnated in P1 microsystems
To establish separation protocols of radionuclides by solid phase
extraction in the microsystems, affinity of the microcolumn for radio-
nuclides have to be determine for each acidity. Then, knowing the
adsorption behavior of radionuclide in nitric acid, the choice of the
conditioning, loading and eluting phases can be achieved. Respect to
achieve a total adsorption or a total desorption by putting the stationary
phase with a given volume of acid at a well-chosen concentration, we
have characterized the adsorption by the coefficient of distribution. The
distribution coefficients of the three microcolumns were determined for
the U, Pu, Th Eu and Nd as a function of nitric acid concentrations. In
order to compare the results easily, distribution coefficients have been
defined with the nomenclature used by Horwitz [1] and calculated by
the relationship (2).
C 0 − Cs V
Dw = *
Cs mmonolith (2)
with Dw the distribution coefficient of the element (mL g−1), C0 the
element concentration in the volume V (mL) of the initial solution in
contact with the monolith (µg g−1), Cs the element concentration in the
volume V of the initial solution after contact with the monolith (µg g−1)
and mmonolith the mass of the monolith in the microchannel (g). The
experiments showed that for a Dw < 100 value for which elements are
not or weakly retained in the column on the contrary, a Dw > 100 value
demonstrates a retention of the analyte.
The retention properties of the microcolumns were represented on
graphs for each extractant and compared with the equivalent graph for
commercial resins. Thus, distribution coefficients of the monolith im-
pregnated by TBP are of the same order of magnitude than the
(1) equivalent commercial resin (Fig. 3), with an order of elution Th < Pu
(IV) < U(VI). The distribution coefficients of lanthanides are 1–2 orders
of magnitude lower than the one for thorium with a maximum value at
1 mL g−1, which indicates that these compounds are virtually not held
by the column in the acidity range corresponding of this study. The
distribution coefficients of the monoliths impregnated by the TBP/
CMPO mixture and by the DAAP are presented in Supplementary ma-
terial.
Three microcolumns were developed thanks to the impregnation of
the monolith. They show high exchange capacity and affinity for tet-
ravalent and hexavalent actinides in nitric media until 8 M. These
characteristics highlighted the robustness of the chromatographic mi-
crosystems in corrosive media and can be used for the separation of
radionuclides.
589
M. Losno et al.
Fig. 3. Comparison of TBP impregnated monolith (solid lines) and the corre-
sponding TBP™ resin [9] (dashed lines) distribution coefficients for U(VI) (blue
circle), Pu(IV) (green dash), Th(IV) (black diamond), Eu(III) (red triangle) and
Nd(III) (purple square) (For interpretation of the references to color in this
figure legend, the reader is referred to the web version of this article.).
3.3. Separation of U/Th and Eu
One of the applications covered by microsystems robust in nitric
media are the separation of the radionuclides contained in samples of
fuel cycle. Therefore, separation procedures have been studied in mi-
crosystem P1 and then in the centrifugal microsystem to study the in-
fluence of the geometry of the microcolumn on retention. The feasi-
bility of separation was conducted on a synthetic sample containing U
(VI), Th(IV) and Eu(III) at concentrations equal to 10, 8.89 and 2.5 µg
g−1 in HNO3 5 M, respectively.
3.3.1. Separation in microsystem P1
The separation of U(VI), Th(IV) and Eu(III) was carried out with a
sample of 72.8 µL with the procedure presented Fig. 4. As shown on the
chromatogram presented Fig. 5, the resolution of 4.5 was achieved for
Eu(III)/Th(III) and of 1.2 for the Th(IV)/U(VI) separation. It demon-
strates that the separation is effective using only nitric acid as mobile
phase like imposed in nuclear industry. However, the rectangular sec-
tion of the P1 prototype could be an obstacle to the peaks resolution
due to a larger anisotropy of flow that could lead to a greater diffusion
of the elements [42]. We expect that the use of the centrifugal device
will improve this aspect, the device aspect ratio being of 2.5 instead of
11 for the microsystem P1.
Fig. 4. Scheme of the U(VI)/Th(IV)/Eu(III) separation procedure with TBP
monolith in P1 microsystem.
590
Talanta 185 (2018) 586–591
Fig. 5. Separation of Eu(III) (solid line), Th(IV) (dashed line) and U(VI) (doted
line) in TBP impregnated P1 microsystem (Vfree= (12.2 ± 0.7) µL). Vertical
dashed lines show a change of nitric acid concentration of the mobile phase.
3.3.2. Separation in the centrifugal microsystem
The separation evolved in P1 microsystem was transferred in cen-
trifugal microsystem. For this microsystem flow control depends on the
rotation speed of the platform, the study of the flow rates was needed.
This part is described in Supplementary files.
Separation in centrifugal microsystem was carried out by injecting
5.5 µL of a solution containing U(VI), Eu(III) and Th(IV) in HNO3 5 M.
Three fractions have been collected corresponding to the three peaks of
Fig. 5. Fig. 6 shows the composition of the obtained fractions. Eu(III) is
correctly separated from Th(IV) and U(VI) and totally recovered in the
sample fraction and fraction 1 as it was expected considering the se-
paration carried out in P1 (Fig. 5). Th(IV) is present in fraction 2 as
expected but a large amount is recovered in fraction 3 while U(VI) is
totally recovered in fraction 3 as expected. This difference could be
explained by an increase of the distribution coefficient of Th(IV) in the
TBP impregnated monolith in centrifugal microsystem compared to the
P1 microsystem.
Distribution coefficients of Th(IV) are then determined for the TBP
impregnated centrifugal monolith to know the impact of the column
size on the distribution coefficient of Th(IV). A difference of 3 orders of
magnitude is observed between the distribution coefficients of Th(IV)
obtained in the two microsystems (see Supplementary material). For
example, values of Dw are equal to (5 ± 0.5)·10−1 and (5.7 ± 0.7)·102
in [HNO3] = 0.1 M in P1 microsystem and in centrifugal one, respec-
tively. These higher values of the distribution coefficients of Th(IV) in
the case of centrifugal monolithic column explains the elution of Th(IV)
in fraction 3 in this type of system. Moreover, it is consistent with the
higher exchange capacity determined in the case of centrifugal TBP
impregnated monolithic column ((222 ± 26) mg U/g of monolith)
Fig. 6. Amount of Eu(III) (grey), Th(IV) (squared) and U(VI) (white) recovered
in each fraction of the separation (Vfree = 0.27 ± 0.3 µL)).
M. Losno et al.
Table 4
Comparison of exchange capacity values obtained in P1 microsystem with the
values of the corresponding commercial resins for the three tested extractants.
P1 microsystem CMPO/TBP DAAP TBP
Exchange capacity mg U/g of resin 149 ± 18 280 ± 33 153 ± 18
Commercial resins TRU™ UTEVA™ TBP™
Exchange capacity mg U/g of resin 27a 100 75
a
mgA m/g of resin.
compared to the value for P1 microsystem ((153 ± 18) mg U/g of
monolith). Higher distribution coefficients and higher exchange capa-
city in centrifugal microsystem show the interest of the size calculation
of the column for separations development in impregnated micro-
columns. In this case, the centrifugal column is too narrow and a larger
column could decrease the distribution coefficient of Th(IV) and allows
the recovering of Th(IV) in the second fraction with further experi-
ments.
Finally, the separation of the three elements was carried out on a
total time of 1 h 32 min which is comparable with the protocols per-
formed on classical chromatographic columns [43,44] or by Bruchet
et al. [44] on centrifugal microsystem (65 min).
4. Conclusion
Three microcolumns were developed thanks the impregnation of the
monolith poly(AMA-co-EDMA) in COC microsystems. All the micro-
columns show high exchange capacity and affinity for tetravalent and
hexavalent actinides in concentrated nitric media. These characteristics
highlighted the robustness of the chromatographic microsystems in
HNO3 until 8 M. The retention characteristics of the microcolumns
impregnated by TBP, TBP-CMPO and DAAP were compared with those
of the equivalent commercial particulate resins TBP ™, TRU ™ and
UTEVA™ respectively. In each case, the exchange capacities were
higher in the microcolumns (Table 4). A separation of U, Th and Eu was
performed in a classical and a centrifugal microsystems that highlights
the influence of the geometry of the microchannel on the retention
properties. To our knowledge, this study constitutes an original devel-
opment of an extraction chromatographic microsystem which will
contribute to the decreased of reagents consumption and of the radio-
active quantities manipulated in radiochemical procedures.
Acknowledgement
The authors are extremely grateful to Rémy Fert, Davide Ferraro
and Bastien Venzac, of the Institut Curie Research Center, for their help
in micro-fabrication. Their implication was greatly appreciated. The
authors would like to thanks also Denis Doizi and Jean-Luc Roujou of
SECR in CEA for their help for IR measurements.
Competing interests’ statement
The authors have no competing interests to declare.
Appendix A. Supplementary material
Supplementary data associated with this article can be found in the
online version at http://dx.doi.org/10.1016/j.talanta.2018.04.036.
591
Talanta 185 (2018) 586–591
References
[1] E.P.Horwitz, International workshop on the application of extraction chromato-
graphy in radionuclide measurement, IRMM, Geel, 1998.
[2] E.P. Horwitz, M.L. Dietz, D.M. Nelson, J.J. LaRosa, W.D. Fairman, Anal. Chim. Acta
238 (0) (1990) 263–271.
[3] E.P. Horwitz, M.L. Dietz, R. Chiarizia, H. Diamond, A.M. Essling, D. Graczyk, Anal.
Chim. Acta 266 (1) (1992) 25–37.
[4] S. Aldrich, XAD resin product information, 1998.
[5] J.R. FitzPatrick, B.S. Schake, J. Murphy, K. Holmes, M.H. West, The removal of
uranium from acidic media using ion exchange and/or extraction chromatography,
in: R.G. Bautista (Ed.), Emerging Separation Technologies for Metals II, 1996, pp.
119–134.
[6] T. International, Product Sheet TRU resin, 2015.
[7] E.P. Horwitz, R. Chiarizia, M.L. Dietz, H. Diamond, M.R. Nelson, Anal. Chim. Acta
281 (1993) 361–372.
[8] T. International, Product Sheet UTEVA resin, 2016.
[9] T. International, Product Sheet TBP resin, 2014.
[10] G. Janssens-Maenhout, Nanotechnol. Percept. 3 (2007) 183–192.
[11] K.W. Ro, W.J. Chang, H. Kim, Y.M. Koo, J.H. Hahn, Electrophoresis 24 (18) (2003)
3253–3259.
[12] A.B. Jemere, R.D. Oleschuk, D.J. Harrison, Electrophoresis 24 (17) (2003)
3018–3025.
[13] R.D. Oleschuk, L.L. Shultz-Lockyear, Y. Ning, D.J. Harrison, Anal. Chem. 72 (3)
(2000) 585–590.
[14] H. Yin, K. Killeen, J. Sep. Sci. 30 (10) (2007) 1427–1434.
[15] L. Ceriotti, N.F. de Rooij, E. Verpoorte, Anal. Chem. 74 (3) (2002) 639–647.
[16] A. Gaspar, L. Hernandez, S. Stevens, F.A. Gomez, Electrophoresis 29 (8) (2008)
1638–1642.
[17] A. Gaspar, M.E. Piyasena, F.A. Gomez, Anal. Chem. 79 (20) (2007) 7906–7909.
[18] J.P. Lafleur, A.A. Rackov, S. McAuley, E.D. Salin, Talanta 81 (1–2) (2010) 722–726.
[19] J.P. Lafleur, E.D. Salin, J. Anal. At. Spectr. 24 (11) (2009) 1511–1516.
[20] W.J. Cheong, J. Sep. Sci. 37 (6) (2014) 603–617.
[21] V. Augustin, G. Proczek, J. Dugay, S. Descroix, M.C. Hennion, J. Sep. Sci. 30 (17)
(2007) 2858–2865.
[22] J. Gao, B.T. Manard, A. Castro, D.P. Montoya, N. Xu, R.M. Chamberlin, Talanta 167
(2017) 8–13.
[23] K. Faure, Electrophoresis 31 (15) (2010) 2499–2511.
[24] F. Svec, J. Sep. Sci. 28 (8) (2005) 729–745.
[25] R. Wu, L. Hu, F. Wang, M. Ye M, H. Zou, J. Chromatogr. A 1184 (1–2) (2008)
369–392.
[26] F. Svec, Y. Lv, Anal. Chem. 87 (2015) 250–273.
[27] J.C. Masini, F. Svec, Anal. Chim. Acta 964 (2017) 24–44.
[28] C. Ericson, J. Holm, T. Ericson, S. Hjertén, Anal. Chem. 72 (1) (2000) 81–87.
[29] Y. Fintschenko, W.Y. Choi, S.M. Ngola, T.J. Shepodd, Fresenius J. Anal. Chem. 371
(2) (2001) 174–181.
[30] R. Shediac, S.M. Ngola, D.J. Throckmorton, D.S. Anex, T.J. Shepodd, A.K. Singh, J.
Chromatogr. A 925 (1) (2001) 251–263.
[31] A. Bruchet, V. Taniga, S. Descroix, L. Malaquin, F. Goutelard, C. Mariet, Talanta 116
(2013) 488–494.
[32] M. Madou, J. Zoval, G. Jia, H. Kido, J. Kim, N. Kim, Annu. Rev. Biomed. Eng. 8
(2006) 601–628.
[33] G.T. Seaborg, J. Katz, U.H. Hanning, W.M. Manning (Eds.), The Transuranium
Elements: Research Papers. Part I, McGraw-Hill Book Co., Inc., New York, 1949.
[34] G.T. Seaborg, J. Katz, The Actinide Elements, National Nuclear Energy Series,
Plutonium Project Record 14A McGraw-Hill, New York, 1954.
[35] G.T. Seaborg, J. Katz, The Chemistry of the Actinide Elements, John Wiley and
Sons, Inc., New York, 1957.
[36] B. Lesigne, Etude cinétique de la réduction par le peroxyde d′hydrogène du pluto-
nium hexavalent en solution acide, Commissariat à l′Energie Atomique, Fontenay-
aux-Roses, France, 1967.
[37] K. Faure, M. Albert, V. Dugas, G. Crétier, R. Ferrigno, P. Morin, J.L. Rocca,
Electrophoresis 29 (24) (2008) 4948–4955.
[38] M. Losno, I. Ferrante, R. Brennetot, J. Varlet, C. Blanc, B. Grenut, E. Amblard,
S. Descroix, C. Mariet, Micromachines 7 (3) (2016) 45–57.
[39] T. Rohr, D.F. Ogletree, F. Svec, J.M.J. Fréchet, Adv. Funct. Mater. 13 (4) (2003)
264–270.
[40] Q. Pu, O. Oyesanya, B. Thompson, S. Liu, J.C. Alvarez, Langmuir 23 (3) (2007)
1577–1583.
[41] T.B. Stachowiak, F. Svec, J.M.J. Fréchet, J. Chromatogr. A 1044 (1–2) (2004)
97–111.
[42] M. Caude, A. Jardy, Techniques de l′ingénieur, Chromatographie et techniques
séparatives, Base documentaire: TIB385DUO (ref. article: p1455), 1994.
[43] BioRad, Strong anion exchange resin instruction manual, in 〈http://www.bio-rad.
com〉.
[44] M. Yamaura, H. Matsuda, J. Radioanal. Nucl. Chem. 241 (2) (1999) 277–280.