Article pubs.acs.org/Langmuir
Mesoscopic Aspects of Phase Transitions in a Solvent Extraction System Ross J. Ellis,* Matthieu Audras, and Mark R. Antonio Chemical Sciences & Engineering Division, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois 60439, United States S Supporting Information *
ABSTRACT: In liquid−liquid extraction, organic phase splitting arises when high concentrations of polar solutes (acids/metal ions) are extracted. Herein, we investigate the mesoscopic roots that underpin phase splitting in alkane phases containing mixed amphiphiles, of contemporary interest in solvent extraction separation systems, by extracting various oxoacids. The oxoacids exhibited individual macroscopic (extractive and physical) behaviors, inducing phase splitting into heavy and light domains under markedly different conditions. Using small-angle X-ray scattering (SAXS) data analyzed using the generalized indirect Fourier transform (GIFT) method, we showed that, in all cases, acid extraction drove the self-assembly of reverse micelles into rods. These grew with increased acid extraction until reaching a critical length of 20 nm, at which point interactions produced interconnected cylinders or lamellar sheets that prelude phase splitting into heavy and light domains. In all cases, the heavy phase contained the same surfactant ratioTBP (tri-n-butyl phosphate) and CMPO (octyl(phenyl)-N,N-diisobutylcarbamoylmethylphosphine oxide)even though the concentrations of acid, water, and amphiphiles were markedly different. The remarkable similarities in structure and amphiphile stoichiometries underpinning phase splitting across the macroscopically different acid extraction series allude to the mesoscopic roots of organic phase behavior in solvent extraction. Our studies show that the structures underpinning phase splitting in solvent extraction systems are more complex than previously thought and are reminiscent of phase transitions in soft matter.
■
separations science researchers Chiarizia et al.10−12 and Erlinger et al.13,14 proposed that RMs play a pivotal role in the splitting of solvent extraction organic phases. This detrimental phenomenon involves the collapse of the monophasic organic media into a biphasic system under high solute loadings, limiting the maximum loading capacity (and therefore the economic viability) of solvent extraction systems. Their research shows that RMs behave according to Baxter’s model for hard sticky spheres15 so that small spherical aggregates become increasingly “sticky” as polar solute is incorporated into the core, increasing the interactions until they spontaneously condense, splitting the phase. This interpretation is a simple example of mesoscopic phenomena, where underlying fundamental mechanisms behind macroscopic behaviors are related to interactions between nanoscale particles. Although useful in understanding the energetics of phase transition, these studies did not tackle the importance of higher-ordered architectures that may assemble as a result of interacting reverse micelles. In the disparate field of soft matter science, the physical manifestation of the phase transition in complex fluids
INTRODUCTION Reverse micelle (RM) architectures and their transitions into liquid crystals (LCs) are subjects of fascination, not just for the myriad applications discovered for LCs1 and RMs2−5 but also for the fundamental curiosity of the self-assembling properties of amphiphilic molecules that occur throughout nature, not the least of which is the chemistry that underpins life itself.6 Amphiphiles also play a key role in engineering processes, such as solvent extraction. This important separations process employs amphiphilic “extractant” molecules dissolved in a hydrocarbon oil that, upon contact with an aqueous phase containing polar solutes (metal ions, acids, etc.), binds the target polar entity, drawing it into the organic phase and thus removing it from the other aqueous solutes. The amphiphilic nature of extractants makes them surface-active so that, when dissolved in a nonpolar diluent, they represent a surfactant-inoil system. Surfactant-in-oil systems are an important area of research in soft matter chemistry where solution physical properties are linked to the morphology of RM and LC architectures.7 Solvent extraction organic phases are also known to contain RMs;8 however, the system complexity makes them difficult to analyze on a supramolecular structural level so that fundamental research has traditionally taken a molecularfocused approach, relating extractive properties to coordination complexes and hydrogen bonding.9 More recently, however, © 2012 American Chemical Society
Received: August 29, 2012 Revised: October 11, 2012 Published: October 12, 2012 15498
dx.doi.org/10.1021/la3034879 | Langmuir 2012, 28, 15498−15504
Langmuir
Article
attainable in the organic phase without phase splitting), 2 mL of each phase was contacted. After equilibration, the phases were separated by centrifugation (5 min at 3300 rpm). Aliquots of the various phases were withdrawn for analysis. For the determination of the critical acid concentration, 1.5 mL of each phase was contacted under conditions where the phase split into two. At this point, very small volumes (10 to 20 μL) of fresh organic phase and water were added until the organic phase remerged as one as indicated by the absence of turbidity after further equilibration and centrifugation. Beyond the critical concentration, larger volumes (5 mL of each phase) were contacted to get a sufficiently large volume of heavy and light phases for further analyses. In addition to the critical concentration of acid, three precritical and three postcritical concentrations were also explored. The acid concentration in the organic phases following equilibration was determined after stripping with three volumes of water in a 1:2 organic to aqueous phase ratio and analyses of the aqueous strip solutions. (Complete stripping was confirmed by mass balance with concentrations in the initial aqueous phase minus raffinate.) The heavy phases were stripped as described above after appropriate dilution in ndodecane. Acid concentrations were determined by potentiometric titration with 0.1 or 1 M NaOH solutions. Water in the organic phase was determined by Karl Fischer titration using a Metrohm 756/831 KF coulometer (Riverview, FL). The concentrations of TBP and CMPO in the stripped organic phases were determined using a 31P NMR technique developed in-house. This involved making 10 calibration solutions containing 0−0.4 M CMPO and 0−2.4 M TBP in n-dodecane and taking their 31P NMR spectrum on a Bruker Ultrashield 500 Plus spectrometer. Integrals for the CMPO and TBP peaks were measured and plotted against concentration to give straight lines with R = 0.999 and 0.998 for the TBP and CMPO calibration plots, respectively (10 points in each plot). The CMPO and TBP concentrations in the stripped organic phases were measured by comparing the 31P NMR peak integrals with these calibration functions. Small-Angle X-ray Scattering (SAXS) Data Collection. For each TRUEX−acid system (HNO3, HClO4, H2SO4, and H3PO4), representative organic phase samples were prepared at precritical, near-critical, and postcritical concentrations for SAXS analysis in order to investigate acid-dependent changes in the aggregated structures. This was achieved by contacting the TRUEX organic phase with aqueous acidic phases given in Table 1. Eight organic phases were
is correlated to mesoscopic transitions in solution structures involving a variety of aggregate shapes and sizes.16 The soft matter research approach to understanding phase transitions traditionally involves carefully and independently manipulating the mole fractions of constituents in surfactant− oil−water systems, measuring changes in the physical properties, and correlating to structural studies.17 However, solvent extraction systems are more complex because they involve not just water but also other polar solutes. Moreover, the mole ratios of polar solutes are intrinsically linked because solvent extraction involves two phases (aqueous and organic) or even three after phase splitting. These phases are all at equilibrium so that the researcher has limited control over constituent ratios in each domain. In addition to this, solvent extraction systems often involve mixtures of surfactants so that modeling experimental data becomes difficult if not impossible. We believe that fundamental mesoscopic mechanisms, understood as soft matter science, underpin macroscopic phenomena in solvent extraction systems. In the present study, we sought to investigate the metamorphosis of higherordered architectures in a practical solvent extraction system and show how these connect to phase transitions. The subject of this investigation was the TRUEX (TRansUranic EXtraction) solvent system, which is important in the separation of trivalent actinides and lanthanides from spent nuclear fuel.18−20 The organic solvent comprises a 6:1 molar mixture of two extractants (Figure 1), tri-n-butyl phosphate (TBP) and
Figure 1. TRUEX solvent composed of extractants 0.2 M octyl(phenyl)-N,N-diisobutylcarbamoylmethylphosphine oxide (CMPO, left) and 1.2 M tri-n-butyl phosphate (TBP, right) in n-dodecane, providing a TBP/CMPO molar extractant ratio of 6:1.
Table 1. Aqueous Acid Concentrations (M) Used to Prepare Representative SAXS Samples
octyl(phenyl)-N,N-diisobutylcarbamoylmethylphosphine oxide (CMPO) in n-dodecane. By extracting various inorganic acids and water, we were able to study the stability of the TRUEX solvent with respect to a range of polar solutes. Our goal was to investigate how acid extraction from aqueous solutions affects the topology of nanoscale architectures in the organic phase through phase splitting,21−23 thus bringing a soft-matter perspective to complicated phase behaviors in a practical separations system.
■
precritical near-critical postcritical
HNO3
HClO4
H2SO4
H3PO4
3.3 10.5 15
0.05 0.15 0.4
1.3 4 7
1.3 4.5 8
obtained at near-critical and precritical conditions containing various concentrations of acid and water but virtually constant concentrations of extractant amphiphilesTBP and CMPOand dodecane oil. In the four postcritical systems, the heavy phase that contained varying concentrations of all components was isolated. Altogether, these 12 organic samples gave snap shots of the effect of increasing acid concentration for each system and were analyzed using SAXS. The exact concentrations of the various organic phase components are given in Supporting Information Table S1. SAXS measurements were made at the Advanced Photon Source (Argonne National Laboratory) using beamline 12-ID-B at which the 12 keV incident X-ray beam is monochromatized with a Si⟨220⟩ double-crystal monochromator and double focused by flat horizontal and vertical Pd-coated mirrors down to a spot size of 200 (H) × 50 (V) μm2. All solution samples were contained in 2-mm-diameter quartz capillary tubes (Charles Supper Co., 20-QZ) and held under ambient conditions. At the incident photon energy, the 1/e attenuation length of the pure diluent, n-dodecane, is 12 mm, which is 6 times longer than the path length of the capillary tubes. The 2-D scattering
EXPERIMENTAL SECTION
Solvent Extraction Procedure. CMPO came from existing stocks of which the preparation and recrystallization have been previously described.24 The organic alkane diluent n-dodecane (99%) was obtained from Sigma-Aldrich (Milwaukee, WI) as was TBP, which was purified using vacuum distillation before use. The acids were optima grade from Fisher Scientific (Pittsburgh, PA). All other reagents were analytical grade and used as received. Aqueous solutions of acid were prepared using ultra-high-purity water (18.2 MΩ cm). Aliquots of aqueous phases containing varying acid concentrations for each acid (HNO3, HClO4, H2SO4, and H3PO4) were equilibrated with equal aliquots of organic phases (0.2 M CMPO and 1.2 M TBP in n-dodecane) at 23 ± 0.5 °C by vortex mixing in screw cap glass test tubes for 30 min. For aqueous acid concentrations lower than that corresponding to the critical concentration (the highest concentration 15499
dx.doi.org/10.1021/la3034879 | Langmuir 2012, 28, 15498−15504
Langmuir
Article
profiles were acquired from 1−5 s exposures with a Pilates 2 M counting detector that has a pixel size of 0.172 mm and a dynamic range of 2.20 The sample-to-detector distance was 2200 mm, providing a range for momentum transfer of 0.01 ≤ q (Å−1) ≤ 0.7. The scattering vector, q, was calibrated using a silver behenate standard.25 The 2-D scattering images were radially averaged to produce plots of scattering intensity, I(q) versus q, where q (Å−1) = (4π/λ)sin θ in which 2θ is the scattering angle and λ is the wavelength of the X-rays, following standard procedures.26 The I(q) data were put on an absolute scale (cm−1) by calibration with water (18.2 MΩ cm) scattering. The background subtraction was facilitated using a single capillary tube to obtain responses that were free of capillary-tocapillary variation, and each of the diluent volume subtractions was evaluated to reveal shot-to-shot variations of I(q) with less than ±2% differences over all q values for each solution. Both the capillary and the dodecane solvent were taken into account for the background subtraction so that the SAXS waves presented in this article are produced mainly from the aggregates and not the surrounding medium. Application of the Generalized Indirect Fourier Transformation (GIFT) Technique. The GIFT method27−29 was used to obtain pair−distance distribution functions, PDDFs (structured information in real space), from scattering data in q space in the same way as described previously.30 GIFT interprets the globular interacting particle system, I(q) = nP(q) S(q), where P(q) is the average form factor (corresponding to the shape and size of the scattering particles), S(q) is the average structure factor (from interparticle interactions), and n is the number of particles per unit volume. P(q) is the Fourier transformation of its real space counterpart, p(r), according to
P(q) = 4π
∫0
∞
p(r )
Figure 2. Organic water (blue lines/square markers) and acid (red lines/diamond markers) equilibrium concentrations (y axes) vs aqueous equilibrium acid concentrations (x axes) for (a) HNO3, (b) HClO4, (c) H2SO4, and (d) H3PO4.
acids differ markedly in both concentration and ratio of water to acid in the heavy phase. In line with the partitioning of polar solutes, extractants TBP and CMPO also split into concentrated and dilute domains after the critical concentrations of acids are reached. The partitioning is unique for each acid, with different concentrations in the heavy and light phases. However, remarkable consistency is observed in the TBP/CMPO ratio in the heavy domain, which remains constant at 5:1, independent of both acid type and concentration and different from the initial 6:1 ratio (Table 1). This indicates that the two extractants are stoichiometrically linked and suggests a fundamental underlying mechanism for phase splitting that is common to all acids. Chiarizia et al.10−12 proposed that the concentrated phase forms as a result of increased interactions between RMs containing polar solutes so that the micelles are said to condense. The consistent 5:1 stoichiometry of the two extractants would result from the liquid “precipitation” of RMs of fixed extractant ratio. Soft matter chemistry studies lend support to this hypothesis, where mixed micelle systems commonly undergo RM to liquid-crystalline phase transitions without any change in the composition of the RM unit.17 Mesoscale System Behavior. SAXS provides a window into the aggregated solute structures,37 and we used this to understand how extracting the various inorganic acids affects the macromolecular assembly in the TRUEX system through phase splitting. Measurements were made on the organic TRUEX solvent after placing it in contact with aqueous phases containing increasing concentrations of acids. In all systems, increasing acid concentration had the same general effect on the SAXS data (Figure a−d), suggesting that similar evolutions in solution structure were taking place. The TRUEX system without acid (i.e., that in contact with water) gave a scattering pattern that reached a plateau in the low-q region (red markers in Figure 3a−d), showing the q0 behavior that is indicative of globular micellar particles with little extended structure. When the TRUEX solvent was in contact with increasing concentrations of acid (precritical concentrations shown by blue markers, near-critical concentrations by green), the scattering in
sin qr dr qr
This means that to deduce p(r) the inverse Fourier transformation (IFT) of an experimental P(q) must be calculated. In concentrated interacting systems, such as those involved in solvent extraction, the structure factor, S(q), must be modeled and subtracted from the scattering data, I(q), to give P(q) from which p(r) is derived. The selection of the appropriate structure factor model is key to achieving coherent PDDFs from GIFT, and the model selected in the present study was the Percus−Yevick (PY) closure relation41 that has been shown to approximate the interaction effects of micelles in the Baxter model studies closely (see above). More precedence for the applicability of the PY closure relation structure factor model to the present study is given by a string of recent publications by Glatter and Shrestha et al., who used GIFT in combination with the PY closure relation to interpret SAXS data from interacting nonionic surfactant reverse micelle systems in nonpolar media (similar to solvent-extracted organic phases), where they observed the self-assembly of reverse micelles into higher-ordered architectures.31−36
■
RESULTS AND DISCUSSION Macroscopic System Behavior. The initial part of the investigation involved characterizing the macroscopic (extractive and physical) properties of the system by tracking the distribution of solutes between the organic phases at various aqueous acid concentrations (Figure 2). The macroscopic system properties are profoundly influenced by the type of acid, which in turn drives the correlated extraction of H2O in a chemical pas de deux. Nitric acid, in particular, differs in that the cA (the critical concentration of acid in the organic phase at which phase splitting occurs) is much higher (i.e., 2 M compared to 0.4, 0.2 and 0.1 M for phosphoric, sulfuric, and perchloric acids, respectively). Also, increasing nitric acid concentration caused decreasing organic water content, whereas the opposite was true for the other acids. Beyond the cA, the 15500
dx.doi.org/10.1021/la3034879 | Langmuir 2012, 28, 15498−15504
Langmuir
Article
for the acid-free water contact shows an aggregate that is 30 Å in length and 6 Å in diameter (all errors on metric parameters are ±2%), supporting the conclusion drawn from the SAXS data that the TRUEX system without acid forms globules with little extended structure. Upon contact with the precritical acid concentrations, the corresponding PDDFs in Figure 3 (blue curves) indicate the 1-D growth of aggregates with an increased rmax in all cases. The extension of the tail of this PDDF function is nonlinear, which is an effect of the polydisperse rod length and is also why no q−1 dependence is observed in the corresponding SAXS data indicated by blue markers in Figure 3a.29 (A q−1 dependence is often observed for monodisperse linear aggregates.) The positions of the second inflection points show that the cross-sectional diameters of the rods also increase as acid is introduced, betraying a swelling of the RM cores to 7, 8, 9, and 16 Å for perchloric, nitric, sulfuric, and phosphoric acids, respectively. A second interaction peak also develops as acid is increased near to the critical concentration, indicating increasing interactions between the rods.29,32,33,35,42 At this concentration, the position of the second inflection is completely obscured by the interaction peak for all but the phosphoric acid system. The results obtained from fits to the I(q) data with cylindrical and spherical form factors (Supporting Information, Figure S2) provide lengths and radii that agree with the PDDF analysis, showing a transition from globular/spherical shape to cylinders of increasing length up to 200 Å and approximately constant radii as acid is extracted toward the critical point. All acids induced significant increases in the lengths of rodlike aggregates with increased acid extraction and the development of an interaction peak approaching the critical concentration. These system-independent mesostructural transformations caused by acid extraction suggest that, although phase splitting occurs under different conditions because of the individual macroscopic behaviors of the acids in the TRUEX solvent, there is a common underlying mechanism dominating the phase stability. To pursue this notion, we compare the SAXS data for all four acids at the near-critical acid concentrations (Figure 4a). The waves are essentially superimposable, indicating similar mesoscopic structures near the critical concentrations despite the significantly different macroscopic conditions (Figure 2). This is remarkable considering the different inorganic acid scattering-length densities, which depend on the atomic numbers of the atoms, suggesting that the scattering profile is dominated by the arrangement of the polar head groups of the TBP and CMPO molecules (Figure 1). The corresponding PDDFs of these waves (Figure 4b) show cylindrical aggregates of polydisperse length, with a maximum length (rmax) of approximately 200 Å. All p(r) data show the evolution of an interaction peak, which dominates near the critical concentrations so that only one broad peak is observed for all acids except phosphoric. Glatter et al. have shown in analogous soft-matter studies of cloud-point behavior that the broadening of the interaction peak is diagnostic of many cylindrical aggregates interacting,29 ultimately leading to the evolution of a lamellar micelle (a 2-D micellar sheet).43 Solvent-extracted heavy phases are suggested to have lamellar liquid-crystalline structures38 that form from the stacking of lamellar micelles. Alternatively, the growing interaction peak and the apparent 200 Å length may show an increasing interconnectivity/coalescence of cylindrical micelles as described by Ninham et al. in the phase behavior of a quaternary ammonium−water−oil system.44 This would lead to a
Table 2. Concentrations of Polar Solutes (Acid and Water) and Extractants (TBP and CMPO) in the Heavy Phases Formed after Splitting in Each Systema acid
[acid]org
[water]org
[TBP]
[CMPO]
TBP/CMPO
HNO3
2.91 3.47 3.94 0.18 0.28 0.47 0.76 1.27 1.91 0.76 0.89 1.45
0.73 0.87 1.04 1.75 2.31 3.11 3.01 3.94 4.13 1.22 1.34 1.67
1.61 1.74 1.85 1.57 1.78 1.97 2.13 2.25 2.52 1.50 1.57 1.77
0.33 0.36 0.38 0.32 0.39 0.40 0.43 0.45 0.48 0.29 0.32 0.35
4.9 4.8 4.9 4.9 4.5 4.9 4.9 4.9 5.2 5.1 4.9 5.1
HClO4
H2SO4
H3PO4
a
The TBP/CMPO ratio was consistent at 5:1 in each case.
the medium-to-low q regions increased significantly, indicating increased interactions between RMs and assembly into higherordered architectures. The SAXS data for the heavy phases are shown by black markers. For all acids, the heavy phase showed a relative decrease in scattering intensity in the medium-to-low q region because of the strong excluded volume effect as well as an increase in the scattering intensity at high q that was due to the emergence of a correlation peak from the structured arrangements of solute moieties.38 Going across the tetra-acid series, from monoprotic HClO4 to diprotic H2SO4 and then triprotic H3PO4, the intensity of the correlation peak in Figure 3a−d increases and the tail of a second peak at very low q develops. This indicates that the structure of the heavy phase becomes more resolved as the number of H-bond-donor acidic O−H groups increases. The morphologies of architectures formed in interacting solution systems such as solvent extraction can be challenging to discern from SAXS data. This is because random interactions between RM units influence the scattering in the low-tomedium q region, which is also the region that holds information on the extended structure of aggregates.39 Therefore, to understand the supramolecular structures of the higher-ordered architectures, the scattering contribution from the interactions must be separated from the scattering produced by aggregates. This was achieved using the GIFT method, which uses a model to approximate the interactions between micellar cores and simultaneously converts the scattering contribution from the structure of the mesoscopic architectures into real-space functions known as PDDFs (pair distance distribution functions, Figure 3a′−d′). The PDDF describes the scattering entity by a distribution of distances between points within the assembly, with r being the distance and p(r) being the relative number of particular distances that occur. In Figure 3a′−d′, PDDFs for the TRUEX system are shown, with the acid-free contact in red, the precritical acid in blue, and the near-critical acid concentration in green. These show pronounced peaks in the low-r region followed by extended tails, indicating cylindrical or rod-shaped aggregates.40 The maximum extent of the tail (rmax) provides a measure of the maximum extent of the aggregate in terms of the length of the rod.41 The plots clearly show that rmax increases with acid concentration. The inflection point after the initial peak on the PDDF can be calculated from the second differential and gives the diameter of the cylinder (rc).29 The red curve of the PDDF 15501
dx.doi.org/10.1021/la3034879 | Langmuir 2012, 28, 15498−15504
Langmuir
Article
Figure 3. Small-angle X-ray scattering data for the TRUEX extraction systems with nitric (a, a′), perchloric (b, b′), sulfuric (c, c′), and phosphoric (d, d′) acids. The red curves (circular markers) are the organic phases after contacting water (no acid), the blue curves (square markers) are for the precritical concentrations of acids, the green curves (diamond markers) are for the near-critical concentrations of acids, and the black curves (triangular markers) are for the postcritical concentrations of acids. (a−d) Background-subtracted normalized response. (a′−d′) Fourier-transformed PDDF plots using the GIFT technique. Initial aqueous phase acid contacts: (precritical) 3.3 M nitric acid, 0.05 M perchloric acid, 1.3 M sulfuric acid, and 1.3 M phosphoric acid; (critical) 10.5 M nitric acid, 0.15 M perchloric acid, 4 M sulfuric acid, 4.5 M phosphoric acid; and (postcritical) 12 M nitric acid, 0.4 M perchloric acid, 6 M sulfuric acid, and 7 M phosphoric acid.
microemulsion of connected cylinders that show a similar broad correlation peak in the SAXS data as was observed in the heavy phase (black markers in Figure 3). Thus, we draw similarities in the mesostructural evolutions that underpin phase transitions in both soft matter systems and solvent-extracted organic phases. In this way, the phase transition that results in the splitting of the organic phase is rooted in the mesoscopic assembly of structurally complex aggregates, not the spontaneous condensation of simple spheres as previously thought.15
Our study of the TRUEX−acid systems using the GIFT method shows globule-to-rod transitions driven by acid extraction to underpin the phase behavior. In solvent extraction, this perspective contrasts with the accepted Baxter model studies. In soft matter science, however, the basic equation linking the chemical potential to the curvature variation was established in 1975 by Israelachvili et al.,45 predicting sphere-to-rod transitions as a function of amphiphile concentration in aqueous solutions. In 1989, Bellini et al.46 linked critical phenomena in an ionic amphiphile solution to 15502
dx.doi.org/10.1021/la3034879 | Langmuir 2012, 28, 15498−15504
Langmuir
Article
Figure 4. (Left) Small-angle X-ray scattering data and (right) PDDFs of the TRUEX system at critical concentrations of HNO3 (green line/ markers), HClO4 (blue line/markers), H2SO4 (red line/markers), and H3PO4 (purple line/markers).
and form factors. This material is available free of charge via the Internet at http://pubs.acs.org.
the 1D growth of micellar rods, induced by increasing concentrations of amphiphile. At the critical point, the rods become flexible and entangled, leading to a phase transition. Since then, a number of soft matter studies have linked changes in curvature to phase boundaries,47 with some involving the addition of a polar solute to amphiphile systems,48 although none yet involve the extraction of inorganic acids from aqueous solvents into organic solvents. In the present study, we show that the solvent extraction of acids by the TRUEX system produces the same changes in curvature as observed in recent analogous soft matter research on surfactant-in-oil systems that shows the assembly of reverse micelles comprising nonionic surfactants into cylindrical chains in organic media, with the length controlled by the temperature or surfactant concentration.31−33,35,36,42 We show, therefore, that the extraction of acid into organic phases of practical relevance to separation systems behaves like soft matter, with phase splitting traceable to mesoscopic roots.
■
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work and the use of the Advanced Photon Source are supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, Division of Chemical Sciences, Biosciences and Geosciences under contract no DE-AC0206CH11357.
■
■
REFERENCES
(1) Chigrinov, V. G. Liquid Crystal Devices: Physics and Applications; Artech House: London, 1999. (2) Hua, D.; Jiang, J.; Kuang, L.; Jiang, J.; Zheng, W.; Liang, H. Smart chitosan-based stimuli-responsive nanocarriers for the controlled delivery of hydrophobic pharmaceuticals. Macromolecules 2011, 44, 1298. (3) Ganguli, A. K.; Ahmad, T.; Vaidya, S.; Ahmed, J. Microemulsion route to the synthesis of nanoparticles. Pure Appl. Chem. 2008, 80, 2451. (4) Haeger, M.; Currie, F.; Holmberg, K. Organic reactions in microemulsions. Top. Curr. Chem. 2003, 227, 53. (5) Luisi, P. L.; Giomini, M.; Pileni, M. P.; Robinson, B. H. Reverse micelles as hosts for proteins and small molecules. Biochim. Biophys. Acta, Rev. Biomembr. 1988, 947, 209. (6) Rudolph-Bohner, S.; Quarzago, D.; Czisch, M.; Ragnarsson, U.; Moroder, L. Conformational preferences of Leu-enkephalin in reverse micelles as membrane-mimicking environment. Biopolymers 1997, 41, 591. (7) Piazza, R. Soft Matter: The Stuff That Dreams Are Made of; Copernicus: Milan, 2010. (8) Osseo-Asare, K. Aggregation, reversed micelles, and microemulsions in liquid-liquid-extraction: the tri-normal-butyl phosphatediluent-water-electrolyte system. Adv. Colloid Interface Sci. 1991, 37, 123. (9) Tasker, P. A.; Plieger, P. G.; West, L. C. In Comprehensive Coordination Chemistry II; Ward, M. D., Ed.; Elsevier: New York, 2004; Vol. 9, p 759. (10) Chiarizia, R.; Jensen, M. P.; Borkowski, M.; Ferraro, J. R.; Thiyagarajan, P.; Littrell, K. C. Third phase formation revisited: the U(VI), HNO3-TBP, n-dodecane system. Solvent Extr. Ion Exch. 2003, 21, 1.
CONCLUSIONS By attacking an old problem in applied separations with fundamental soft matter science, we have uncovered the mesoscopic roots of phase splitting in an important solvent extraction system. The results indicate that phase splitting in the TRUEX system caused by the extraction of a series of inorganic acids involves the mesoscopic assembly of linear aggregates toward a critical length. This length is common to all acids tested in this study, so the critical length is a fundamental property of the organic solvent matrix and not the polar solute. The different critical acid concentrations observed on the macroscopic scale are then due to the effectiveness of the acids at inducing self-assembly into the critical mesostructural configuration. The assembly of nanoscale aggregates toward phase splitting accounts for the ordered structure of the heavy phase and is contrasted with the current understanding of the spontaneous condensation of RM spheres in terms of the oftenutilized Baxter model. We suggest that similar mesoscopic structural evolutions, understood in terms of soft matter science, may underpin a range of physical phenomena in solvent-extracted organic phases.
■
AUTHOR INFORMATION
ASSOCIATED CONTENT
S Supporting Information *
Further details including tabulated concentrations of organic phase components in samples that were analyzed using SAXS, long-cylinder model fits to SAXS data, and the GIFT structure 15503
dx.doi.org/10.1021/la3034879 | Langmuir 2012, 28, 15498−15504
Langmuir
Article
surfactants and structural transition induced by oils. J. Phys. Chem. B 2009, 113, 1615. (32) Shrestha, L. K.; Shrestha, R. G.; Aramaki, K. Intrinsic parameters for the structure control of nonionic reverse micelles in styrene: SAXS and rheometry studies. Langmuir 2011, 27, 5862. (33) Shrestha, L. K.; Yamamoto, M.; Arima, S.; Aramaki, K. Chargefree reverse wormlike micelles in nonaqueous media. Langmuir 2011, 27, 2340. (34) Shrestha, R. G.; Shrestha, L. K.; Aramaki, K. Rheology of wormlike micelles in aqueous systems of a mixed amino acid-based anionic surfactant and cationic surfactant. Colloid Polym. Sci. 2009, 287, 1305. (35) Shrestha, L. K.; Sato, T.; Dulle, M.; Glatter, O.; Aramaki, K. Effect of lipophilic tail architecture and solvent engineering on the structure of trehalose-based nonionic surfactant reverse micelles. J. Phys. Chem. B 2010, 114, 12008. (36) Shrestha, L. K.; Shrestha, R. G.; Oyama, K.; Matsuzawa, M.; Aramaki, K. Structure of diglycerol polyisostearate nonionic surfactant micelles in nonpolar oil hexadecane: a SAXS study. J. Oleo Sci. 2010, 59, 339. (37) Putnam, C. D.; Hammel, M.; Hura, G. L.; Tainer, J. A. X-ray solution scattering (SAXS) combined with crystallography and computation: defining accurate macromolecular structures, conformations and assemblies in solution. Q. Rev. Biophys. 2007, 40, 191. (38) Testard, F.; Bauduin, P.; Martinet, L.; Abecassis, B.; Berthon, L.; Madic, C.; Zemb, T. Self-assembling properties of malonamide extractants used in separation processes. Radiochim. Acta 2008, 96, 265. (39) Glatter, O. The interpretation of real-space information from small-angle scattering experiments. J. Appl. Crystallogr. 1979, 12, 166. (40) Glatter, O. Evaluation of small-angle scattering data from lamellar and cylindrical particles by the indirect transformation method. J. Appl. Crystallogr. 1980, 13, 577. (41) Glatter, O. In Neutrons, X-rays and Light; Lindner, P., Zemb, T., Eds.; Elsevier Science B.V.: Amsterdam, 2002; p 73. (42) Shrestha, L. K.; Sato, T.-A.; Shrestha, R. G.; Hill, J.; Ariga, K.; Aramaki, K. Structure and rheology of reverse micelles in dipentaerythrityl tri-(12-hydroxystearate)/oil systems. Phys. Chem. Chem. Phys. 2011, 13, 4911. (43) Shrestha, L. K.; Sharma, S. C.; Sato, T.; Glatter, O.; Aramaki, K. Small-angle X-ray scattering (SAXS) study on nonionic fluorinated micelles in aqueous system. J. Colloid Interface Sci. 2007, 316, 815. (44) Ninham, B. W.; Barnes, I. S.; Hyde, S. T.; Derian, P. J.; Zemb, T. N. Random connected cylinders: a new structure in three-component microemulsions. Europhys. Lett. 1987, 4, 561. (45) Israelachvilli, J. N.; Mitchell, D. J.; Ninham, B. W. Theory of self-assembly of hydrocarbon amphiphiles into micelles and bilayers. J. Chem. Soc., Faraday Trans. 2 1976, 72, 1525. (46) Bellini, T.; Mantegazza, F.; Piazza, R.; Degiorgio, V. Stretchedexponential relaxation of electric birefringence in a polydisperse colloidal solution. Europhys. Lett. 1989, 10, 499. (47) Chevalier, Y.; Zemb, T. The structure of micelles and microemulsions. Rep. Prog. Phys. 1990, 53, 279. (48) Testard, F.; Zemb, T. Solute effect on connectivity of water-inoil microemulsions. Langmuir 2000, 16, 322.
(11) Chiarizia, R.; Jensen, M. P.; Borkowski, M.; Thiyagarajan, P.; Littrell, K. C. Interpretation of third phase formation in the Th(IV)HNO3, TBP-n-octane system with Baxter’s “sticky spheres” model. Solvent Extr. Ion Exch. 2004, 22, 325. (12) Chiarizia, R.; Nash, K. L.; Jensen, M. P.; Thiyagarajan, P.; Littrell, K. C. Application of the Baxter model for hard spheres with surface adhesion to SANS data for the U(VI)-HNO3, TBP-n-dodecane system. Langmuir 2003, 19, 9592. (13) Erlinger, C.; Belloni, L.; Zemb, T.; Madic, C. Attractive interactions between reverse aggregates and phase separation in concentrated malonamide extractant solutions. Langmuir 1999, 15, 2290. (14) Erlinger, C.; Gazeau, D.; Zemb, T.; Madic, C.; Lefrancois, L.; Hebrant, M.; Tondre, C. Effect of nitric acid extraction on phase behavior, microstructure and interactions between primary aggregates in the system dimethyldibutyltetradecylmalonamide (DMDBTDMA)/ n-dodecane/water: a phase analysis and small angle X-ray scattering (SAXS) characterization study. Solvent Extr. Ion Exch. 1998, 16, 707. (15) Baxter, R. J. Percus-Yevick equation of hard spheres with surface adhesion. J. Chem. Phys. 1968, 49, 2770. (16) Toledano, P. Phase Transitions in Complex Fluids; World Scientific: Singapore, 1998. (17) Ekwall, P.; Stenius, P. In International Review of Science, Physical Chemistry, Series Two; Kerker, M., Ed.; Butterworths: London, 1975; Vol. 7, p 215. (18) Schulz, W. W.; Horwitz, E. P. The TRUEX Process and the Management of Liquid TRU Waste. Sep. Sci. Technol. 1988, 23, 1191. (19) Ansari, S. A.; Pathak, P.; Mohapatra, P. K.; Manchanda, V. K. Aqueous partitioning of minor actinides by different processes. Sep. Purif. Rev. 2011, 40, 43. (20) Mathur, J. N.; Murali, M. S.; Nash, K. L. Actinide partitioning a review. Solvent Extr. Ion Exch. 2001, 19, 357. (21) Nagasaki, S.; Wisnubroto, D. S.; Enokida, Y.; Suzuki, A., 3rd Phase formation in nitric acid extraction by N-octyl(phenyl)-N,Ndiisobutylcarbamoyl-methylphosphine oxide. Solvent Extr. Ion Exch. 1994, 12, 459. (22) Suresh, A.; Rao, C.; Sabharwal, K. N.; Srinivasan, T. G.; Rao, P. R. V. Third phase formation in the extraction of Nd(III) by Octyl(phenyl)-N,N-diisobutyl carbamoyl methyl phosphine oxide (OΦCMPO). Solvent Extr. Ion Exch. 1999, 17, 73. (23) Mallick, S.; Suresh, A.; Srinivasan, T. G.; Rao, P. R. V. Third phase formation in the extraction of nitric acid and metal ions by octyl(phenyl)-N,N-diisobutyl carbamoyl methyl phosphine oxide (OΦCMPO) based solvents. Desalin. Water Treat. 2011, 25, 216. (24) Gatrone, R. C.; Kaplan, L.; Horwitz, E. P. The synthesis and purification of the carbamoylmethylphosphine oxides. Solvent Extr. Ion Exch. 1987, 5, 1075. (25) Keiderling, U.; Gilles, R.; Wiedenmann, A. Application of silver behenate powder for wavelength calibration of a SANS instrument - a comprehensive study of experimental setup variations and data processing techniques. J. Appl. Crystallogr. 1999, 32, 456. (26) Guinier, A.; Fournet, G. Small-Angle Scattering of X-rays; John Wiley & Sons: New York, 1955. (27) Brunner-Popela, J.; Mittelbach, R.; Strey, R.; Schubert, K. V.; Kaler, E. W.; Glatter, O. Small-angle scattering of interacting particles. III. D2O-C12E5 mixtures and microemulsions with n-octane. J. Chem. Phys. 1999, 110, 10623. (28) Fritz, G.; Bergmann, A.; Glatter, O. Evaluation of small-angle scattering data of charged particles using the generalized indirect Fourier transformation technique. J. Chem. Phys. 2000, 113, 9733. (29) Glatter, O.; Fritz, G.; Lindner, H.; Brunner-Popela, J.; Mittelbach, R.; Strey, R.; Egelhaaf, S. U. Nonionic micelles near the critical point: micellar growth and attractive interaction. Langmuir 2000, 16, 8692. (30) Ellis, R. J.; Antonio, M. R. Coordination structures and supramolecular architectures in a cerium(III)-malonamide solvent extraction system. Langmuir 2012, 28, 5987. (31) Sharma, S. C.; Shrestha, R. G.; Shrestha, L. K.; Aramaki, K. Viscoelastic wormlike micelles in mixed nonionic fluorocarbon 15504
dx.doi.org/10.1021/la3034879 | Langmuir 2012, 28, 15498−15504
