Available online at www.sciencedirect.

com

Procedia Chemistry 7 (2012) 679 684

ATALANTE 2012
International Conference on Nuclear Chemistry for Sustainable Fuel Cycles

Microfluidic tools for the liquid-liquid extraction of radionuclides

in analytical procedures
Gwendolyne Hella*, Clarisse Marieta, Grard Coteb,c
a

CEA Saclay, DEN/DANS/DPC/SEARS/LANIE, 91191 Gif-sur-Yvette cedex, France

Chimie ParisTech, Laboratoire dElectrochimie, Chimie aux Interfaces et Modlisation pour lEnergie
11 rue Pierre et Marie Curie, 75005 Paris, France
c
CNRS, UMR 7575, 75005 Paris, France

Abstract
The analyses of radionuclides are in great demand and a cost effective technique for the separation of analytes is required. A
micro-scale reactor composed of microchannels fabricated in a microchip was chosen to investigate liquid-liquid extraction
reactions driven by three different families of metal extractants: neutral, acidic and ion-pair extractants. The extraction
conditions in the microfluidic device were considered. These investigations demonstrated that the conventional methodology
used for solvent extraction in macro-scale reactors is not directly transposable to micro liquid-liquid extraction systems.
However, it is expected that the understanding of the chemical and physical phenomena involved in a reference extraction
systems studied in a given selected lab-on-chip will lead us to develop and validate a methodology suitable to miniaturized
reactors.
2012
2012The
TheAuthors.
Authors.Published
Publishedby
byElsevier
ElsevierB.V.
B.V.Selection and /or peer-review under responsibility of the Chairman of the

Selection and/or
peer-review
under responsibility
the Chairman
of thelicense.
ATALANTE 2012 Program
ATALANTA
2012
Program Committee
Open accessofunder
CC BY-NC-ND
Keywords: Microchannel chip ; Solvent extraction ; Separation ; Radionuclide ; Lab-on-chip

1. Introduction
The analysis of radionuclides present in high and medium level radioactive wastes is carried out according to
operating protocols including many separation/purification steps. These steps, essential to obtain a pure fraction
containing the radionuclides of interest, are based on the chromatography techniques, liquid-liquid extraction and
precipitation/dissolution. Nevertheless, such an analytical procedure is long, complex and difficult to implement

1876-6196 2012 The Authors. Published by Elsevier B.V. Selection and /or peer-review under responsibility of the Chairman of the ATALANTA 2012
Program Committee Open access under CC BY-NC-ND license.
doi:10.1016/j.proche.2012.10.103

680

Gwendolyne Hell et al. / Procedia Chemistry 7 (2012) 679 684

in hot cells or glove boxes. Indeed these analyses entail both chemical and radiological risks so that the use of
solvents must be reduced inducing a modification of the operating procedures.
2. Miniaturization and liquid-liquid extraction
2.1. Contribution of the microfluidics
In the frame of studies aiming to improve these protocols, the microfluidic tools have their own place: they
enable to reduce both the volumes of the samples and of the solvents, the analysis time and the operators
exposure time [1]. The feature of microfluidic tools is to group together the different steps which are necessary
for the analysis of a sample or for the production of a compound since they combine the functions of injection,
purification, pre-concentration, separation, reaction and detection [2]. The integration of these different
components on a single chip gives rise to the concepts of -TAS (micro Total Analysis System) and then of labon-chip. Nevertheless, while lab-on-chips are widely used in biochemistry and medical care diagnostics, their use
in the nuclear field is still at its beginning. Recent technological breakthroughs allow to work with automated
microsystems which can be used in parallel processing to increase the throughput or in multiplexed processing of
separation/purification steps coupled to the detection system [3]. Like biochemistry and medical care diagnostics,
radioanalytical chemistry could take advantage of all these characteristics.
2.2. Interests of the -liquid-liquid extraction
Most of analytical microsystems involve chromatographic separation, only few ones concern liquid-liquid
extraction. If liquid-liquid extraction presents some drawbacks in macro-scale, the technique applied in
microsystem shows new features which make it competitive with liquid chromatography. Thanks to
miniaturization, coupling liquid-liquid extraction with detection system and with a stripping step is now allowed
[4]. On the other hand, in lab-on-chip, specific interfacial area, that is the surface-to-volume ratio, is increased,
known and reproducible since it is determined by the geometry of the extraction channel [5]. This is a great
advantage with respect to the conventional systems (ARMOLLEX cells, short-time phase contacting cells).
3. Implementation of the -LLE in the nuclear field
3.1. Objectives of the study
This study has two main objectives. The first one deals with the fundamental aspects. Indeed, the
microfluidics must allow to exacerbate the interfacial phenomena that makes it a tool of choice to study the
interfacial dynamics during the liquid-liquid extraction. The lab-on-chip used in this fundamental part and the
methodology developed will be applied to achieve the second objective. The latter consists in analyzing
radioactive solutions and thus in separating the radionuclides by liquid-liquid extraction by taking the best
advantage of the specificities related to the miniaturization (high specific area, short diffusion path, decrease of
volumes and so reduction of chemical and radiological risks).
3.2. Choice of the microsystem
The liquid-liquid extraction in microsystem with a co-current flow in the case of the analysis of solutions
coming from the nuclear field is, currently, practically exclusively studied by the Department of Applied
Chemistry of the University of Tokyo [6]. Nevertheless, their developments are about different chemical systems
studied in various designs of microsystems. In fact, the chemical systems studied are varied regarding to the

Gwendolyne Hell et al. / Procedia Chemistry 7 (2012) 679 684

different extractants (tributylphosphate (TBP), carbamoyl phosphine oxide (CMPO)), the different diluents (ndodecane, heptane, xylene) and the different analytes (rare earths, transition metals) used. In the same way,
the design of the microsystems is varied. The shape of the microsystem (H, Y-Y) varies as well as the
geometry of the channels: rectangular or cylindrical section, presence or not of a guide structure, microchannels
dimensions (length, width and depth), asymmetry or not Therefore, it is difficult to compare the performances
of different microsystems for a unique chemical extraction system and conversely, to highlight the specific
behavior of diverse chemical extraction systems in the same microsystem. Moreover, from a fundamental point
of view, only few works have been carried out, today, in order to characterize the relative contributions to the
mass transfer of the analytes of diffusion and of the chemical reactions kinetics. That is the reason why we have
chosen to study different chemical systems in a given microsystem.

(a)

(b)

Fig. 1. (a) An experimental setup for extraction experiment; (b) A schematic illustration of glass microchannel chip
with a focus on a part of the microchannel

The apparatus used is shown in Fig 1. The microsystem has a double-Y shape, is constituted of a symmetric
extraction channel and does not present a guide structure. This microchannel has an extraction length of 8 cm, a
width of 100 m and a depth of 40 m. The aqueous and organic solutions in microsyringes are pushed out at a
fixed flow rate using microsyringe pumps (KDScientific). To ensure that no leak occurs at the inlets and at the
outlets of the microchannels, we chose a tried and tested glass microsystem manufactured and commercialized by
the Institute of Microchemical Technology (IMT, Japan). This criterion is fundamental in the frame of the
analysis of radioactive solutions. The choice of the material has been realized in the same way. Indeed, this
material must be resistant to the solvents (dodecane for instance) as well as to the acids (highly concentrated
nitric and hydrochloric acid solutions) used during the radiochemical separations.
3.3. Choice of the chemical systems
In this microsystem, we propose to study liquid-liquid extraction of Eu for three chemical systems
corresponding to the three main extractants families: neutral, acidic and ion-pair extractants. In this way, it will
be possible to highlight the chemical behavior differences in the microsystem of reference and thus to exacerbate
their features.
3.3.1. Neutral extractants
Malonamides, diglycolamides and some organophosphorus extractants belong to this category. Regarding the
latter, only rare data are available because either no kinetic problems were encountered or the extractants was
developed too recently. However, the dimethyl-N1,N3-dibutyl-2-tetradecyl malonamide (DMDBTDMA) has

681

682

Gwendolyne Hell et al. / Procedia Chemistry 7 (2012) 679 684

been the subject, at the macro-scale, of numerous studies, mainly thermodynamic in nature but some kinetic
studies are also available in the literature up to date. Thus, Toulemonde [7] and Dal Don [8] have measured
kinetic constants for the extraction of U, Am, and lanthanides by the DMDBTDMA in n-dodecane with a
constant interfacial area cell (ARMOLLEX cell). Thus, the chemical system constituted by the DMDBTDMA in
n-dodecane was chosen like the reference system to validate the microfluidic tool in comparison with the results
obtained in the ARMOLLEX apparatus.
3.3.2. Acidic extractants
The di-2-ethylhexylphosphoric acid (HDEHP) is one of the most studied reagent among the acidic extractants.
The interest for this compound lies in its use in the TALSPEAK process (diluted in dodecane) for the separation
trivalent actinides/trivalent lanthanides [9]. There are some kinetic studies for the mass transfer carried out with
the HDEHP regarding the extraction of Zn, Ni, Co as well as rare earths and transplutonium elements. For these
reasons, we propose to keep the HDEHP as an acidic extractant model.
3.3.3. Ion-pair extractants
The molecules often used in hydrometallurgy for the implementation of a liquid-liquid extraction reaction by
formation of ions pairs are the protonated trioctylamine (contained in Alamine 336), the methyltrioctylammonium chloride (contained in Aliquat 336) and the 2-oxo-1-phenylhydrazinolate of ammonium more
commonly named Cupferron. Though the literature is abundant regarding the extraction of U and rare earths for
this family, we do not have found study on the extraction kinetics apart from the one of the chromium, probably
because the extraction kinetics are fast and thus do not cause any problem in analysis. So we have chosen the
system Aliquat 336 in dodecane.

3.4. Methodology of the study

Firstly, it will be necessary to develop and validate a methodology in the selected microsystem. We will
compare the results obtained for the chemical system Eu/DMDBTDMA/dodecane in macro-scale and in lab-onchip. Once the microfluidic tool is validated for this reference from the thermodynamic and kinetic points of
view, the study of the interface will be investigated via different techniques (physico-chemical characterization,
addition of surfactants, variation of the interfacial area).. Through the miniaturization, we wish to render
preponderant interface reactions with respect to volume reactions and thus to determine the precise role of the
interface in the extraction process.
The same methodology will be applied for the kinetic study involving acidic and ion-pair extractants.
3.5. Choice of the flow type
There are numerous types of flow in microsystem with different features: dispersed flow and continuous
counter-current or co-current flows (Fig 2.). In the dispersed flows, a phase is dispersed into another (organic in
aqueous or vice-versa) which leads to the formation of droplets or slugs [10]. These flows have a high specific
interfacial area favoring a faster extraction but some practical difficulties exist. The first one lies in the separation
of the two phases after extraction. Indeed, this separation necessitates an additional step, in batch, which
therefore cancels the benefit of the liquid-liquid extraction in microsystem. Secondly, the precise determination
of the specific interfacial area is difficult in the case of droplets or slugs. Another drawback of this kind of flow is
the complexity of modeling the extraction kinetics with respect to a system where the flows would be parallel and
continuous.

Gwendolyne Hell et al. / Procedia Chemistry 7 (2012) 679 684

The parallel continuous flows are obtained in the case of co-current and counter-current flows. The countercurrent flow is scarcely used in microsystem for several reasons [11]. Firstly, the interface stability is difficult to
maintain and necessitates a surface modification. This process should enable to render hydrophobic the part of
the channel dedicated to the organic phase and thus to confine the aqueous and organic phases in the hydrophilic
and hydrophobic parts of the channel, respectively. Moreover, the instability of the interface with counter-current
flow leads to a limitation of the range of flow rates being able to be used even if a surface functionalization has
been carried out. These drawbacks intrinsic to dispersed and counter-current flows lead us to choose co-current
flow for performing -LLE.

Fig. 2. Conceptual schemes of dispersed, counter-current and co-current flows in microfluidics

3.6. Choice of the hydrodynamic parameters

Although the hydrodynamic conditions influence the extraction efficiency, the reaction kinetics and the
diffusion during the mass transfer, the choice of the most relevant hydrodynamic parameters stays difficult. The
flow rates (or linear velocities), Reynolds number (often used in batch but very few in microsystems) or contact
times of the two phases are typical parameters used in the literature.
The study of the extraction efficiency, of the analyte concentration or of the distribution ratio as a function of
the flow rates of both aqueous and organic phases presents some difficulties in microsystem. Because the
viscosity is a predominant force at the micrometric scale, it is impossible to change a composition of a phase
without changing the viscosity and then the flow rate. Also, this involves a modification of the position of the
interface in the microchannel.
The second hydrodynamic parameter used, widespread, is the contact time between the two phases during the
extraction in microsystem. The simplest expression of this contact time is given by Ooe et al. and is deduced
from the total volume of the extraction channel and the faster flow rate used [12]. Nevertheless, Ban et al.
propose another method to calculate the contact time and they express it with the following relation [13]:

where wo, do, l refer to the width, the depth and the length of the extraction channel and fo denotes the flow rate
of the organic phase. This expression of the contact time takes into account the flow rates used and the specific
width of the phase considered which depends itself of the flow rates and of the viscosities of the two phases.
Thus, it seems that this calculation method of the contact time is the most adapted to the micrometric scale.
Aware of these experimental specificities of the study, we have chosen to fix the viscosities by the imposed
concentrations of the solutions whereas the flow rates will be chosen in order to keep constant the position of the
interface. For this, the flow rate of the organic phase will be adapted to obtain the height of the interface
previously chosen while the aqueous flow rate will be fixed since this phase is impoverished (during the
extraction) and its properties vary little.

683

684

Gwendolyne Hell et al. / Procedia Chemistry 7 (2012) 679 684

4. Conclusion
The nuclear field needs to miniaturize its analytical systems in order to improve their performances while
reducing the chemical and radiological risks. Among the systems to improve, the liquid-liquid extraction presents
significant advantages to be used at the miniaturized scale. Before the achievement of the optimization of the
analyses by liquid-liquid extraction in microsystem, it is necessary to understand the interfacial dynamics and the
reaction kinetics for different chemical systems in a reference microsystem. For this, the continuous co-current
flow has been chosen and the hydrodynamic parameter selected is the contact time between the two phases.
When these kinetic and interfacial studies will be carried out, the optimization of the analyses will be possible
and it will be interesting to compare the experimental results with those obtained by molecular dynamics.

References
[1] Baharudin L. Microfluidics: Fabrications and Applications. Instrumentation Science and Technology 2008;36:222-30.
[2] Arora A, Simone G, Salieb-Beugelaar GB, Kim JT, Manz A. Lastest Developments in Micro Total Analysis Systems. Analytical
Chemistry 2010;82:4830-47.
[3] Thorsen T, Maerkl SJ, Quake SR. Microfluidic Large-Scale Integration. Science 2002;298:580-6.
[4] Maruyama T, Matsuhida H, Uchida JI, Kubota F, Kamiya N, Goto M. Liquid Membrane Operations in a Microfluidic Device for
Selective Separation of Metal Ions. Analytical Chemistry 2004;76:4495-500.
[5] Ferrouillat S, Tochon P, Peerhossaini H. Micromixing enhancement by turbulence: Application to multifunctional heat exchangers.
Chemical Engineering and Processing 2006;45:633-640.
[6] Hotokezaka H, Tokeshi M, Harada M, Kitamori T, Ikeda Y. Development of the innovative nuclide separation system for high-level
radioactive waste using microchannel chip- extraction behavior of metal ions from aqueous phase to organic phase in microchannel. Progress
in Nuclear Energy 2005;47:439-47.
[7] Toulemonde V. Liquid-liquid extraction kinetics of uranyle nitrate, actinides (III) and lanthanides (III) nitrates by amide extractants.
1995, Universit de Paris 06, Paris, France.
[8] Dal Don M. Study of extraction kinetics of lanthanides(III) and actinides(III) nitrates by the molecule N,N-dimethyl-N,N-dibutyl,
tetradecylmalonamide. 1997, Universit de Paris 11, Orsay, France.
[9] Martin LR, Zalupski PR. Thermodynamics and Kinetics of Advanced Separations Systems. FY 2010 Summary Report 2010.
[10] Nichols KP, Pompano RR, Li L, Gelis AV, Ismagilov RF. Toward Mechanistic Understanding of Nuclear Reprocesing Chemistries
by Quantifying Lanthanide Solvent Extraction Kinetics via Microfluidics with Constant Interfacial Area and Rapid Mixing. Journal of the
American Chemical Society 2011;133: 15721-9.
[11] Aota A, Nonaka M, Hibara A, Kitamori T. Countercurrent Laminar Microflow for Highly Efficient Solvent Extraction. Angewandte
Chemie 2007;119:896-8.
[12] Ooe K, Tashiro Y, Saika D, Kitamoto Y, Matsuo K, Takabe T et al. Development of On-line Solvent Extraction System with
Microchips for Heavy Element Chemistry. Journal of Nuclear and Radiochemical Sciences 2007;8:59-62.
[13] Ban Y, Kikutani Y, Tokeshi M, Morita Y. Extraction of Am(III) at the Interface of Organic-Aqueous Two-Layer Flow in a
Microchannel. Journal of Nuclear Science and Technology 2011;48:1313-8.