Structural aspects of heteropolyacid microemulsions.

PCCP View Article Online

Published on 18 September 2014. Downloaded by University of California - Santa Cruz on 26/10/2014 15:47:51.

PAPER

Cite this: Phys. Chem. Chem. Phys., 2014, 16, 22566

View Journal | View Issue

Structural aspects of heteropolyacid microemulsions† Mrinal K. Bera,a Ross J. Ellis,a Benjamin P. Burton-Pyeb and Mark R. Antonio*a Metrical insights from X-ray scattering studies of dense fluid phases (known as ‘‘third’’ phases) in the Keggin heteropolyacid–tri-n-butyl phosphate (TBP)–n-alkane system are provided. Small-angle X-ray scattering (SAXS) experiments reveal inter-acid correlation peaks corresponding to average centre-of-mass to centre-of-mass separations of 18–23 Å between P P, Si Si, and Al Al of H3PW12O40, H4SiW12O40, and H5AlW12O40, respectively, consistent with the presence of TBP solvates that form by hydrogen bonding between the acids and the phosphoryl group of TBP. The Baxter sticky sphere model analyses of the SAXS data reveal identical structures for all the dense phases with inter-cluster interaction energies of B5kBT. We demonstrate that the sticky sphere model is an essential paradigm for interpreting SAXS and predicting mesoscale assembly in heteropolyacid microemulsions. The model parameters for the ternary polyoxometalate–amphiphile–oil systems reveal, in rigorous clarity, how the interactions between heteropolyacid solvates underpin their condensation to produce the observed scattering data. Aside from aiding researchers in predicting the physical origins of SAXS in strongly-interacting micellar systems found in

Received 9th July 2014, Accepted 8th September 2014

natural and engineered settings, such as chemical separations, our study provides mesostructural information

DOI: 10.1039/c4cp03014a

extraction involving the contact of aqueous electrolytes of dodecatungsto-phosphoric, -silicic, and -aluminic

that complements previously observed electrochemical behaviours for third phases formed by solvent acids with organic solutions (e.g. n-dodecane and n-octane) of TBP, and by simple dissolution of the acid

www.rsc.org/pccp

salts of the polyoxometalate hydrates in the same organic solutions.

Introduction The field of polyoxometalate (POM) chemistry is gargantuan in scope, breadth, and depth with myriad practical and proposed uses in a number of disparate fields, including soft matter sciences.1–4 In the latter pursuit, contemporary interest in the so-called ‘‘soft-oxometalates’’5 and the class of deliberatelyformed, surfactant-encapsulated POMs6–9 is undergoing rapid growth, especially with regard to their catalytic activity.10–13 In this, the acid salts of POM anions—the so-called heteropolyacids— are renowned for their combined acidity and redox activity.14–17 These behaviours have been exploited for sensor18–21 and, very recently, energy storage applications,22–26 wherein hybrid surfactantencapsulated POM materials are, in general, more durable than the unencapsulated heteropolyacids. The improved stabilities of the amalgamated, inorganic–organic core–shell systems1,11,13,27 a

Chemical Sciences and Engineering Division, Argonne National Laboratory, Argonne, IL 60439, USA. E-mail: [email protected], [email protected], [email protected]; Tel: +1 630 252 3653, +1 630 252 3647, +1 630 252 9267 b CUNY Hunter College, Department of Chemistry, New York, NY 10065, USA. E-mail: [email protected] † Electronic supplementary information (ESI) available: Guinier plots for the acidic aqueous electrolytes of H3PT, H4SiT, and H5AlT; SAXS data and distance distribution functions for the light organic phases. See DOI: 10.1039/c4cp03014a

22566 | Phys. Chem. Chem. Phys., 2014, 16, 22566--22574

result from the organization of organic surfactants around POMs that alters their physical and chemical properties. Despite the surge of research in the area, details of the colligative properties and morphological relationships between heteropolyanions and cationic surfactants as well as between heteropolyacids and neutral surfactants remain to be understood and, ultimately, controlled in order to advance applications in old and new disciplines alike, such as colloid chemistry and mesoscale science of soft-oxometalates. Recently, for example, dense fluid phases of a-Keggin heteropolyacids (see Fig. 1(a)) have been prepared by use of the liquid–liquid extraction process,28,29 which is also known as solvent extraction (SX), using the neutral organic extractant trin-butyl phosphate (TBP, see Fig. 1(b)). Complex fluids obtained in this manner are widespread in the microemulsion literature21,30–34 where they are known as Winsor type III systems. In the separation science literature, they are referred to as third phases,35–39 and result from the transition of an immiscible biphasic liquid– liquid system (e.g., aqueous|organic) to a triphasic one (e.g., aqueous|heavy-organic|light-organic or heavy-organic|aqueous|lightorganic) as a result of either chemical loading and/or environmental effects. The systems-wide generality of the micellar model for the third phase formation phenomenon,40 wherein the inter-solute attraction energy—U(r)—approaches the value

This journal is © the Owner Societies 2014

View Article Online


Paper

Fig. 1 (a) Stick model of the [Xn+W12O40]n8 anion (for X Al3+, Si4+, P5+) showing W (grey), O (red), and Al/Si/P (blue) based upon the atomic coordinates from the single-crystal X-ray and neutron diffraction data for the [PW12O40]3 anion.49 The dashed circle defines the outer circumference of the [Xn+W12O40]n8 anion with a charge-independent X–Ot radius of 5.2 Å, corresponding to the radius of the circle. (b) Stick model of one of the typical conformers of the TBP molecule with all transconformations of the n-butyl groups.50 The phosphoryl (PQO) group hydrogen bonds with the H atoms of the heteropolyacids to form oilsoluble inorganic–organic (POM–TBP) solvates.

of 2kBT (where kB is the Boltzmann constant and T is temperature) at the critical point of phase splitting, is explained by van der Waals interactions between polar cores of reverse micelles. In the organic phases of SX systems, heteropolyacids are encapsulated by a shell of neutral extractant molecules like TBP, which has both amphiphilic and surface-active characteristics, through hydrogen bonding interactions. The resulting inorganic-acid–organic-extractant solvate is thus solubilized in nonpolar paraffinic solvents, like n-octane, where the heteropolyacid alone would be otherwise insoluble. Unlike the dissociated heteropolyacids in aqueous electrolytes consisting of hydrated heteropolyanions and separately-hydrated protons, heteropolyacids in third phases are associated neutral entities in nano-size pools of water (reverse micelles) that may contain other inorganic solutes, such as mineral acids, and the polar portion (i.e., the PO4 group) of the TBP extractant molecules. In recent articles, redox behaviours of micelle-confined a-Keggin heteropolyacids solvated by TBP28 in water-in-oil microemulsions and immobilized in Triton X-405/silica gel films41 were studied. The results showed substantial shifts in the electrode potentials for the redox couples of the heteropolyacid–organic solvates in the micellar environments compared to the heteropolyanionhydrates in the aqueous electrolytes. For the TBP systems, the redox activities of the heteropolyacids in the third phases were shown to involve the synchronized transfer of protons across the aqueous-electrolyte–third-phase-heteropolyacid–TBP solvate interface. This provided an example of another novel aspect `-vis third phase physical properties and the motivation vis-a to carry out the structural characterization of heteropolyacid microemulsions reported here. In spite of long-standing efforts taking advantage of third phases for POM syntheses,42 a number of fundamental structural aspects of the fluid phases remain to be understood. In particular, under conditions exactly comparable to those used for the liquid–liquid extraction of mineral acids like HClO4,


PCCP

the SX of 12-phosphotungstic acid (H3PW12O40, abbreviated as H3PT hereafter) reveals an extremely low (0.0011 M) limiting organic concentration condition,29 far lower than the 0.116 M concentration for the most effective third-phase-forming inorganic acid, HClO4.36 Contrary to the micellar model of third phase formation, the easy collapse of the organic phase upon extraction of H3PT was attributed to the limited solubility of the heavy and highly-polar H3PW12O40TBP3 solvate in the nonpolar media (n-octane), wherein solute correlation effects—attraction and repulsion—facilitate electrostatic interactions and H-bonding that lead to the precipitation of a fluid. In this, the solute attraction energy of 3.2kBT well exceeds the 2kBT level that was found to be associated with phase splitting in all previous studies on TBP third phase formation.19,36–40,43–48 Although these interaction energies are small compared to those of thousands of kBT for ionic crystals, the structural correlations in fluids are more easily influenced than in solid salts, wherein the motions of ions and dipoles are constrained. Through use of small-angle X-ray scattering (SAXS), we have elucidated the solute energetics and correlations in third phases formed by the Keggin systems of H3PT, 12-silicotungstic acid, H4SiW12O40 (abbreviated H4SiT), and H5AlW12O40 (abbreviated H4AlT). The results provide a new perspective into the meso- and macro-molecular scales of liquid structuring in POM-containing microemulsions and a quantitative measure of the energetics that dictate aggregation along with their physical and chemical properties. In particular, we demonstrate that the Baxter modelling routinely employed for sticky hard-sphere colloidal particles also applies to POM systems in third phases. Finally we provide metrical comparisons with results from PMF (potential of mean force) simulations for the P P and Si Si contact distances between H3PT and H4SiT clusters, respectively,51–53 in aqueous acidic media, and for the same contacts reported in crystal structures of the solid salts.49,54

Experiments General solvent extraction procedure Through use of SX exactly as described elsewhere,28,29 we have prepared a series of denser-than-water third phases with Keggin heteropolyacids H3PT, H4SiT, and H5AlT. In an alternative preparative manner as also described elsewhere,28 third phases were formed by the simple dissolution of the H3PT, H4SiT, and H5AlT hydrates in organic solutions. All reagents, except H5AlT, were obtained from commercial sources. Dodecatungstoaluminic acid was prepared as described beforehand.55 Small-angle X-ray scattering SAXS data were collected at beam line 12-ID-C of the Advanced Photon Source (APS) at Argonne National Laboratory with incident photon energy of 28 keV, which was chosen to provide good transmittance through the solutions and, at the same time, to be as high above the L-edge energies of W in order to minimize the fluorescence background signal. Solutions (ca. 20 mL) were contained in 2.0 mm outer diameter quartz capillary tubes (Cat. # 15-QZ, Charles Supper Company). The sample-to-detector

Phys. Chem. Chem. Phys., 2014, 16, 22566--22574 | 22567

View Article Online


PCCP

distance was adjusted to provide a detecting range for momentum transfer of 0.02 r q r 1.2 Å1, where q was calibrated using a silver behenate standard. After correction for spatial distortion and detector sensitivity, the 2-D scattering images were azimuthally averaged to produce log–log plots of scattered intensity, I(q) vs. q = (4p/l)sin(y/2) (Å1), in which y is the scattering angle and l is the X-ray wavelength, following standard procedures.56–58 The background response was removed in identical fashion, involving the subtraction of the empty cell scattering and the scattering arising from the aqueous electrolytes, n-octane, and n-dodecane diluents as appropriate. SAXS data analysis A third phase is a Winsor III30,34 type dense microemulsion phase that is immiscible with the bulk aqueous and the bulk oil phases. As per Winsor III type microemulsion classifications, the structure of our heteropolyacid third phases could be either a bicontinuous or a percolating cluster system. A bicontinuous structure generally forms under the condition of equal volumes of water and oil phases in the presence of low surfactant concentration.59 It is known that Keggin heteropolyacid third phases form under very low limiting organic concentration conditions as compared to the third phases of inorganic acids, like HClO4, with high (0.73 M) extractant concentrations. The extraction of heteropolyacids into n-octane is essentially quantitative and although the H3PT third phase contains approx. 2 M H2O,29 this concentration is about one-half of that (3.9 M)36 for the corresponding HClO4–TBP third phase. The smaller concentration of water contained in the H3PT third phase is attributed, in part, to the larger size of H3PT as compared to HClO4. Due to the comparatively low water transfer to the third phases in the extraction of heteropolyacids, we attribute their formation to a percolating POM–TBP cluster solvate structure rather than to a bicontinuous structure. The SAXS scattering intensity can be written as a function of wave transfer vector, q, as I(q) = CF(q)S(q), where C is a constant that depends upon the total number of microemulsion clusters in the scattering volume, and F(q) is the form factor of an individual cluster; considering it as a sphere of radius Rc and electron density contrast, Drc = rc rs, F(q) can be written as F(q,Rc) = Drc2[(sin(qRc) qRc cos(qRc))/q3Rc3]2, where rc and rs

Paper

are the average electron densities of the particle clusters and solvent molecules respectively. S(q) is the structure factor due to correlations between heteropolyacid clusters. In dilute solutions, the clusters do not interact with each other and, in this absence of correlations, S(q) = 1. But in condensed solutions, like the third phases, the clusters come in close proximity of one another resulting in inter-cluster correlations (S(q) a 1). In the third phases studied here, the heteropolyacid clusters are charge neutralized by protonation and solvated by TBP through hydrogen bonding. As illustrated in Fig. 2(a), the acid solvates are core–shell spheres with core radii, Rc, surrounded by spherical coronas of radii of Rhs. Accordingly, the electron density contrast of the spherical cores and the spherical shells is Drc = rc rl rs and Drl = rl rs, respectively, where rl is the average electron density of the ligand shells. The form factor of the cluster follows as F(q,Rc) = [Drc((sin(qRc) qRc cos(qRc))/q3Rc3) + Drl((sin(qRhs) qRsh cos(qRhs))/q3Rhs3)]2. Typically, electron densities of oil phases, like n-dodecane, are close to those of the alkyl tail groups (here, n-butyl) on the extractants due to the elemental equivalence of their –CH2– chains. By comparison, the inner cores consist of POMs with electron-rich high-Z tungsten frameworks (Fig. 1(a)) that provide a large electron density contrast compared to the oil phase. For these reasons, X-rays are not sensitive to the tails of the extractants in the corona regions but, rather, are only sensitive to the cores of the scattering particles, which consist of heteropolyacids. Thus, Drl = 0 is a good approximation. The intercluster correlations in third phases are the manifestations of short-range attractive interactions between the charge-neutral heteropolyacid–TBP solvates.29,36,37 In order to interpret the SAXS data, we assumed the simplest attractive interaction model, i.e., the ‘‘sticky sphere’’ or, properly, the Baxter model.43,45 As shown in Fig. 2(a), the interaction energy, U(r), between the scattering clusters is written as follows: U(r)/kBT = N for

r r Rhs

UðrÞ 12td ¼ ln for Rhs o r Rhs þ d kB T Rhs U(r)/kBT = 0

for r 4 Rhs + d

Fig. 2 (a) Baxter sticky sphere model for the SAXS data analysis illustrating the principal metrical parameters Rc and Rhs as well as the square well width (d) and interaction energy, U(r). (b) Calculated scattered intensities from the Baxter sticky sphere model as discussed in the text with different stickiness parameters (t) and a constant volume fraction, fp = 0.3, and (c) with different fp and a constant t = 0.5. The POM cluster radius (Rc) and the radius of the hard-core potential (Rhs) were fixed at the values 5 and 10 Å, respectively (based upon the approximately equivalent physical radii for the heteropolyacid clusters (5.2 Å) and TBP (4.7 Å)).



View Article Online


Paper

PCCP

Here, d is the width of a square well of infinitesimal width; t is referred to as the stickiness parameter and its value is indicative of the strength of the attractive interaction between the clusters. The structure factor S(q,t,Rhs,Z) can then be analytically calculated, in the limit of d/Rhs - 0, by solving the Ornstein– Zernike equation with the Percus–Yevick closure relation46,48 using various volume fractions, fp, for the acid cores of the microemulsion scatterers. Fig. 2(b) and (c) shows calculated scattering profiles from spherical particles of radius Rc = 5 Å that are interacting via Baxter’s sticky sphere model with an average interparticle separation Rhs = 10 Å. It is clear from these modelderived scattering profiles that the intensity in the low q region is sensitive to the values of t, such that a decrease in t increases the low q intensity and, simultaneously, shifts the correlation peak to higher q values. Conversely, increasing fp values sharpen the correlation peak. These calculations show that both fp and t play important roles in shaping the correlation peak and the low q response attributable to the structure factor.

Results and discussions Keggin anions in acidic aqueous electrolytes The 12-heteropoly acid salts of the 53-atom, a-Keggin anions (e.g., [PM12O40]3, [SiM12O40]4, and [AlM12O40]5) are prototypes of a large family of heteropolyacids60 with X–M–O framework architectures (for X B3+, Al3+, Co3+, Si4+, P5+, etc.; M Mo6+, W6+). Despite the difference in charge, the Keggin anions have a spherical diameter of approximately 10.4–10.5 Å, as measured by the twelve terminal O atoms (Ot) at the outermost circumference of the molecular anion surface (refer to the dashed circle around the stick model shown in Fig. 1(a)). The P5+, Si4+, and Al3+ ions are at the geometric centres of the [Xn+W12O40]8n anions, providing a handy and easily visualized reference point for the cluster centre-of-masses. The distances from these central cations to the Ot atoms correspond to the cluster radii of approx. 5.2 Å. The tungsten–oxygen shell consists of twelve crystallographically-equivalent WO3 units that incorporate oxygen atoms that bridge between W atoms (Ob) and another type that bridges between W and X (Ox).49 The SAXS data for dilute heteropolyacid (0.0004 M) solutions in aqueous HCl electrolytes (pH = 1) provides a point of reference with regard to particle scattering by the monodisperse H3PT, H4SiT, and H5AlT entities in solution. Even at 0.0004 M concentrations, the large contrast between the high-Z, electron-rich (W) atoms in the X–W–O frameworks and the low-Z atoms (H, O, Cl) of the H2O electrolyte provides the strong X-ray scattering responses shown in Fig. 3(a) (colored symbols). The shapes of the log–log plots of Fig. 3(a) are typical of particle scattering from dilute solutions of spherical solutes. In fact, the I(q) response calculated (red dashed line) from the atomic coordinates of [PW12O40]3 is in excellent agreement with the experimental data, for which the fitted form factor responses (black solid lines) provide equivalent Rc values (4.8 Å); these are consistent with the known spherical radii for the ‘‘nude’’ molecular anions of PW12O403, SiW12O404, and AlW12O405.


Fig. 3 (a) Log–log plots of the experimental SAXS data for the aqueous solutions of H3PT, H4SiT, and H5AlT (red circles, green squares, and blue triangles, respectively) in acidic (pH = 1) electrolytes. The scattering response is typical of monodisperse spherical cluster scattering, demonstrating that the structures of the plenary a-Keggin anions are equivalent in the aqueous electrolyte solutions. The calculated SAXS curve from the atomic coordinates of H3PT is shown as the red dashed line and shifted vertically for clarity. The spherical form factor F(q,Rc) fits are shown as solid black lines. There is no indication for anion–anion oligomerization (e.g., dimer formation) under the dilute (0.4 mM) conditions employed here. Both data and fits are shifted by factors of 2 and 4 for H4SiT and H5AlT for clarity. (b) The distance distribution functions, p(r) vs. r, obtained by indirect Fourier transformations61 of the primary data in (a). The symmetry of the peaks at 5 Å is a typical of that seen for solid sphere structures, and the maximum linear extent of 10 Å is consistent with the maximum linear dimension of the Keggin anions as indicated by the dashed circle of Fig. 1(a).

The match between experiment and calculation as well as experiment and fit demonstrates that interparticle (anion–anion) interactions do not contribute to the observed response from the free and unassociated acids. If protons and hydronium ions were associated with the anion surfaces, the contrast is too low to be detected against the H2O solvent. A further quantitative measure of the cluster sizes, in terms of their radii of gyration (Rg), is obtained by the analysis of the low-q data by using the Guinier approximation. The Guinier plots (ESI,† Fig. S1) show linear and parallel responses, indicating that the cluster size is independent of charge. The equivalent Rg values (3.8 Å) can be used to calculate the average physical radius (Rs) of the Keggin acid anions. In view of their spherical morphologies, the relationship between Rg and Rs is Rs = O(5/3)Rg. The Rs value (4.9 Å) is somewhat lower than but consistent with the 5.2 Å radii determined from single-crystal X-ray diffraction measurements as the average distance between the X ion and the twelve outermost Ot atoms, see Fig. 1(a), whose contribution to the total I(q) response is small because the contrast is not resolved from the H2O solvent. The metrical results (Rg, Rs)


View Article Online


PCCP

Paper

are in agreement with the sphere radii (Rc) obtained from form factor fits to the I(q) data of Fig. 3(a) (solid lines). Moreover, the bell-shaped distance distribution functions, p(r) vs. r (Fig. 3(b)), for the three anions are typical for solid spheres. The peaks at 5 Å and the maximum linear dimensions of 10 Å confirm the equivalent structures. Each of these metrics and X-ray responses serve as a handy gauge for comparisons of the concentrated third phase solution systems discussed below. SAXS of third phases formed by SX The high heteropolyacid concentrations (380–400 times that of the aqueous solution, refer Table 1) as well as the contrast between the W atoms in the acid anions and the C, H atoms of the solvents and TBP provides an intense response in I(q), see coloured lines in Fig. 4(a) and (b). Unlike the response for the dilute aqueous solutions of Fig. 3, there are strong, broad peaks at q = 0.30–0.35 Å1 in the data for the third phases. These peaks result from myriad unresolved inter-acid (HnXT HnXT) correlations, corresponding to centre-of-mass to centre-of-mass distances of 18–23 Å, that are modulated by intervening TBP, nitric acid, and water molecules. Their presence indicates correlations between the TBP-solvated acids in the third phases—as depicted in Fig. 4(c)—that are diagnostic of structured fluids, similar to what is observed in the SAXS of ionic liquids,62,63 in aliphatic solutions with high concentrations of malonamide extractants,64,65 and in Winsor III microemulsions.31 As such, the correlation peaks for the H3PT, H4SiT, and H5AlT third phases are typical for concentrated solutions of electron-rich (i.e., W) scattering clusters. In view of the 5.2 Å acid anion radii, the space between two anions is estimated at 7.6–10.6 Å (2p/qmax 2 5.2). This is sufficient space to accommodate the TBP molecules that solvate the heteropolyacid. In fact,

even with its many (five–six) low-energy conformers,50,66 the average radius of a TBP molecule has been calculated to be 4.65 Å (ref. 67) and, hence, its 9.3 Å diameter is readily accommodated within the 10.6 Å space between HnXT entities. Without intervening molecules, like TBP, results from PMF simulations reveal that the H3PT, H4SiT, and H5AlT clusters have P P, Si Si, and Al Al contact distances of approximately 11.0 Å,51–53 exactly as found in crystal structures.49,54 These distances (which are essentially equivalent to twice the anion radius) indicate contact ion pairing in aqueous acidic solutions and solid salts. Such close contacts are not possible with the third phase system systems because of the large size of the HnXT(TBP)3 solvates. The experimental SAXS data shown in Fig. 4(a) and (b) for the H3PT and H4SiT systems were fit with the Baxter sticky sphere model by refining five parameters, Rc, Rhs, t, Z, and C. The best fit parameters are shown in Table 1. The fitted curves, also shown in Fig. 4(a) and (b) as black lines, are in excellent agreement with the experimental data. In order to calculate the interaction energy, U(r), the value of the ratio, d/Rhs, was taken to be 0.001 exactly as for the numerical simulations of Fig. 2, which predict the correlation peaks found in the experimental data (cf. Fig. 2(b), (c), 4(a), (b) and (d)–(f)). The values (5.0 to 5.8kBT) indicate that the heteropolyacid solvates in third phases obtained from SX interact strongly. The magnitude of the attraction in terms of U(r) is more than twice the value of 2kBT associated with phase splitting and third phase formation in the micellar model for the extraction of simple mineral acids.40 The metrical information from the Baxter model of SAXS analysis (Table 1) indicates that there are no appreciable differences in the sizes of the H3PT and H4SiT acid cores. A system-by-system comparison shows that the 2Rhs values are

Table 1 The parameters obtained by fitting the SAXS data of Fig. 4(a), (b) and (d)–(f) with the Baxter sticky sphere model illustrated in Fig. 2(a) and as discussed in the text. The estimated standard deviations at the 3-sigma level are shown (in parentheses) for the last significant figure

POMs (TBP%)

Rc a (Å)

Rhs b (Å)

tc

fp d (%)

Ue (kBT)

Cf

Co g (M)

Molecular solutions in TBP H5AlT (100%) H4SiT (100%) H3PT (100%)

4.92(5) 4.91(5) 5.0(1)

10.4(2) 10.8(2) 10(1)

0.35(2) 0.36(2) 0.134(7)

14.0(5) 14.2(5) 10(1)

5.47(5) 5.44(6) 6.43(5)

0.226(4) 0.216(4) 0.196(6)

0.06 0.06 0.06

10.0(1) 10.53(8) 10.0(1) 11.3(2)

0.38(1) 0.28(1) 0.58(3) 0.24(1)

33.8(4) 25.3(4) 32.9(8) 20.8(8)

5.40(4) 5.70(4) 4.97(6) 5.84(5)

2.80(5) 2.22(4) 2.78(5) 1.84(5)

0.16 0.16 0.15 0.15

Third phases from dissolution in TBP–n-C12H26 H5AlT (20%) 4.79(6) 9.73(7) 4.79(6) 9.7(2) H5AlT (30%) 4.74(5) 9.61(7) H4SiT (20%) 4.81(5) 10.01(8) H4SiT (30%) 4.66(4) 9.07(7) H3PT (20%) 4.66(5) 9.37(7) H3PT (30%)

0.36(2) 0.32(3) 0.89(9) 0.56(4) 0.94(9) 0.57(4)

41.5(5) 37(1) 38.9(9) 38.1(8) 44(2) 44(1)

5.45(6) 5.58(9) 4.5(1) 5.01(6) 4.5(1) 4.99(7)

0.86(2) 0.77(3) 0.81(1) 0.70(1) 0.90(1) 0.83(1)

0.21 0.21 0.18 0.18 0.27 0.27

Third phases from SX with TBP–n-alkanes 4.56(5) H3PT (20% n-C8H18) 4.64(5) H3PT (30% n-C8H18) 4.66(5) H4SiT(20% n-C12H26) 4.73(8) H4SiT(30% n-C12H26)

a The range of values (4.6–5.0 Å) are less than the X–O radius of 5.2 Å (for X Si, P) determined by single-crystal X-ray diffraction because the X-ray scattering response from 12 terminal O atoms (Ot in Fig. 1(a)) provides little contrast against the solvent (water-in-oil microemulsions) background. b The range of values (4.4–6.6 Å) for the effective radial dimension of TBP obtained by (Rhs Rc) are comparable with the average radius of 4.65 Å calculated for a population of TBP conformers. c The dimensionless stickiness parameter. d The cluster volume fraction. e The interaction energy or depth of the potential well. f A measure of total number of clusters in the scattering volume and the electron density contrast of the clusters against the solvent background. g POM concentrations in the organic solutions and the dense phases formed by SX.



View Article Online


Paper

PCCP

Fig. 4 (a) and (b) SAXS data (coloured thick solid lines) and Baxter model fits (solid black lines) for H3PT third phases (red) obtained by SX with 20% (a) and 30% (b) TBP concentrations in n-octane and for H4SiT third phases (green) obtained with SX with 20% (a) and 30% (b) TBP concentrations in n-dodecane. (c) Model of fluid structures in the third phases showing interactions between the individual H3PT3TBP and H4SiT3TBP solvates.68 SAXS data (coloured thick solid lines) and Baxter model fits (solid black lines) for samples with H3PT (red), H4SiT (green), and H5AlT (blue) formed by direct dissolution into 20% (d) and 30% (e) TBP solutions in n-dodecane, and in neat TBP (f).

larger for the third phases obtained with 30% TBP–n-alkane solutions than for 20% TBP–n-alkane ones. This is consistent with the fact that the higher concentration of TBP extracts more water and nitric acid than the lower concentration. The additional solutes occupy space between the heteropolyacids, causing the small 0.5–1.3 Å elongation of the inter-cluster distances in the H3PT and H4SiT systems, attributable to relaxation of TBP conformers or a change in solvation number (from 3, ref. 29 and 68). Consistent with this is the significantly higher volume fractions (Z) for the acid-solvate clusters in the 20% TBP systems where they are more condensed (32–34%) compared with the values for the 30% TBP systems where they are more disperse (20–25%). In line with previous voltammetric measurements,28 all of the SAXS data of Fig. 4(a) and (b) are consistent with the presence of protonated heteropolyacids in the third phases. The neutral entities reside in essentially equivalent TBP-solvated environments. Because the heteropolyacid solutes report to the lower (heavy) third phases in quantitative manner, the middle (aqueous) phases as well as the upper (light) organic phases are devoid of heteropolyacids. The SAXS data for the light organic phases (shown as ESI,† Fig. S2) reveal contrast from particle scattering (without correlation peaks) that is independent of the rudimentary


heteropolyacid response observed in I(q) and p(r) of Fig. 3(a). Although interesting in their own right as water-in-oil microemulsions with TBP and nitric acid in the n-alkane solutions, detailed analyses of the data for the light organic phases is outside the scope of this investigation. SAXS of third phases formed by dissolution in TBP–n-dodecane To compare with SX systems, third phases were prepared by the direct dissolution of H3PT, H4SiT, H5AlT into 20%- and 30%TBP–n-dodecane solutions. As shown in Fig. 4(d) and (e), the I(q) responses, especially in terms of the correlation peaks, are similar to those of the third phases prepared by SX. The average HnXT HnXT distances (18–20 Å) obtained from the positions of correlation peaks show that the cluster solvates in 20% and 30% n-dodecane solutions are about 2–3 Å closer than those in the SX systems. This suggests that the supramolecular structures of the heteropolyacidTBP solvates are influenced by environmental effects of which the principal differences are the high concentrations of nitric acid and water, which are absent and very low, respectively, in the third phases obtained by dissolution. The metrical information from the SAXS data analyses using the Baxter model (Table 1) is presented in Fig. 5 to facilitate


View Article Online


PCCP

quantitative comparisons. The attractive interaction energies, shown in Fig. 5(a), between clusters in n-dodecane solutions for both H3PT and H4SiT are similar to the SX system energies. The acid cluster core radii of about 4.8 Å, as shown in Fig. 5(b), are also essentially equivalent. The 1.0–1.5 Å smaller ligand shell thickness compared to the SX systems, as shown in Fig. 5(c) and (d), suggest the influence of the presence/absence of nitric

Paper

acid and water on the rotational averaging and conformational variations of TBP. As with the light phases from the SX experiments (vide supra), the light organic phases from the dissolution experiments reveal contrast (see ESI,† Fig. S3 and S4) that has no connection with heteropolyacid solvate scattering because the upper phases are void of the HnXT solutes. The absence of correlation peaks in I(q) and the structures found in p(r) are accounted for by reverse micellar entities that form in n-dodecane with the amphiphilic TBP extractant and adventitious water from the HnXT salts. SAXS of molecular solutions of heteropolyacids in TBP

Fig. 5 The parameters obtained (Table 1) from the Baxter sticky sphere model illustrated in Fig. 2(a). (a) The depth of the potential well in kBT units. (b) The radii (Rc) of the inner heteropolyacid cores of the TBP solvates. (c) The radii (Rhs) of the effective hard core potentials for the TBP-solvated acids. (d) The thicknesses of the coronas formed by TBP around the acid cores. These values reflect the difference between the hard core (Rhs) and inner core radii (Rc) of the clusters. (e) The volume fractions, fp, of the clusters in the corresponding microemulsions.


In effort to probe the basic structural unit of the HnXTTBP solvates that form in the third phases obtained by SX and dissolution in 20%- and 30%-TBP–n-alkane solutions, we prepared simple homogeneous solutions in neat TBP. These solutions, which are not third phases, further serve to remove any effect of the paraffin diluents (n-octane and n-dodecane) on the organization of TBP about H3PT, H4SiT, and H5AlT. Contrary to expectations that follow from the SAXS responses obtained for the aqueous molecular solutions of Fig. 3(a), the data for the organic molecular solutions of Fig. 4(f) reveal correlation peaks between 0.3–0.4 Å1. As described in detail above for the third phase fluids, these correlation peaks are diagnostic of interacid HnXT HnXT associations with a 20–21 Å range of centre-ofmass to centre-of-mass distances. In the absence of n-alkane diluents as well as water and nitric acid, the ligand shell or TBP thickness is found to be B5 Å, which is not significantly different from the values for the third phases prepared by SX and dissolution. The sole structural difference is observed in the values of the volume fractions, fp. For the molecular solutions in neat TBP, the values are approximately 1/2 to 1/4 smaller than the values obtained for the third phases from SX and dissolution in n-alkanes (see Fig. 5(e)). This difference is attributed to the low density of HnXTTBP entities in the molecular solutions compared to their densities in the third phases obtained by SX and dissolution in TBP–n-alkane solutions. The detailed Baxter analyses of the dense third phases and the homogeneous molecular solutions show that although there are differences in the packing fractions and TBP dimensions, the interaction energies between the clusters remain constant at B5kBT. This indicates that the structures of the cluster solvates formed by each of the preparative methods are essentially identical in terms of energetics. The findings of identical core– corona cluster structures in third phases obtained by SX and dissolution and molecular solutions obtained by dissolution are in accordance with their identical electrochemical28 and SX68 behaviours. Whereas previous SX results have shown that H3PT and H4SiT are trisolvates of TBP, there is no such independent knowledge about the H5AlTTBP system other than that reported here. Despite the extra protons available for H-bonding with TBP, it might be sterically difficult to surround H5AlT with either four or five TBP molecules. In this regard, it is to be noted that there are only small differences of about 0.5–1.5 Å in the dimensions of the coronas about the HnXT acids that can easily arise from different populations of TBP conformations in the


View Article Online


Paper

HnXT3TBP solvates (for n = 3–5). Due to the very low electron density contrast between the n-butyl chains of TBP and the n-alkane diluents, we were not able to identify any structural variations that would suggest a solvation number other than 3. We suggest that different conformations of TBP are stabilized by environmental effects arising from differences in water, acid, and organic diluent content in the Winsor type III water-in-oil microemulsions containing HnXT3TBP solvates.

Conclusions Structural aspects of dense third phases of heteropolyacids formed by SX with TBP, on the one hand, and direct dissolution into paraffinic solutions of TBP, on the other, were studied using SAXS. The measurements confirmed the extraction of intact heteropolyacid entities with inter-cluster (P P and Si Si and Al Al) distances of approximately 18–23 Å. The Baxter sticky sphere model was used for detailed analyses of the SAXS data. The sticky sphere model was shown to be an essential paradigm for interpreting the mesoscale structure and energetics in heteropolyacid microemulsions. The results showed structural similarities amongst all the TBP-solvated clusters of the dense third phases obtained either through SX or direct dissolution. The analyses also showed that the inter-cluster interaction energies of the HnXT3TBP solvates are of the order of 5kBT. This value is B2.5 times larger than the interaction energies estimated earlier with SANS for third phases of inorganic acids and metal salts obtained by SX. The slight variations of 1.5–3.0 Å in the inter-cluster separations for the third phases obtained by the two methods are attributed to differences in TBP conformations as well as water and nitric acid constituents. The extraction and dissolution of acid salts of Keggin heteropolyanions by neutral organophosphorous extractants, like TBP, to form dense, self-organized fluid phases provide a parallel entry to the encapsulation of heteropolyanions by cationic surfactants, like quaternary ammonium salts, to prepare liquid crystalline3,69 and ionic liquid phases7,10 with redox activity.70

Acknowledgements We thank Dr Travis H. Bray for assistance during the initial stages ¨nke Seifert (Advanced Photon Source) of this research, and Dr So for assistance at beam line 12-ID-C with SAXS data acquisition. This material and the use of the Advanced Photon Source, a U. S. Department of Energy (DOE) Office of Science User Facility at Argonne National Laboratory, is based upon work supported by the U. S. DOE, Office of Science, Office of Basic Energy Science, Division of Chemical Sciences, Biosciences and Geosciences, under contract No DE-AC02-06CH11357.

Notes and references 1 S. Polarz, S. Landsmann and A. Klaiber, Angew. Chem., Int. Ed., 2014, 53, 946–954. 2 Y. Yan, B. Li, W. Li, H. L. Li and L. X. Wu, Soft Matter, 2009, 5, 4047–4053.


PCCP

3 S. Y. Yin, H. Sun, Y. Yan, H. Zhang, W. Li and L. X. Wu, J. Colloid Interface Sci., 2011, 361, 548–555. 4 B. F. Zhang, P. C. Yin, F. Haso, L. Hu and T. B. Liu, J. Cluster Sci., 2014, 25, 695–710. 5 S. Roy, CrystEngComm, 2014, 16, 4667–4676. 6 M. I. Bodnarchuk, R. Erni, F. Krumeich and M. V. Kovalenko, Nano Lett., 2013, 13, 1699–1705. 7 S. Landsmann, M. Wessig, M. Schmid, H. Colfen and S. Polarz, Angew. Chem., Int. Ed., 2012, 51, 5995–5999. 8 T. Lunkenbein, M. Kamperman, Z. H. Li, C. Bojer, M. Drechsler, S. Forster, U. Wiesner, A. H. E. Muller and J. Breu, J. Am. Chem. Soc., 2012, 134, 12685–12692. 9 Y. L. Li, X. R. Yang, F. Yang, Y. P. Wang, P. H. Zheng and X. X. Liu, Electrochim. Acta, 2012, 66, 188–192. 10 A. Nisar and X. Wang, Dalton Trans., 2012, 41, 9832–9845. 11 F. Q. Guo, Z. F. Zhang, H. F. Li, S. L. Meng and D. Q. Li, Chem. Commun., 2010, 46, 8237–8239. 12 L. Shi, Y. Z. Wang, B. Li and L. X. Wu, Dalton Trans., 2014, 43, 9177–9188. 13 Y. Yang, B. Zhang, Y. Z. Wang, L. Yue, W. Li and L. X. Wu, J. Am. Chem. Soc., 2013, 135, 14500–14503. 14 G. Marci, E. I. Garcia-Lopez and L. Palmisano, Eur. J. Inorg. Chem., 2014, 21–35. 15 A. Micek-Ilnicka, J. Mol. Catal. A: Chem., 2009, 308, 1–14. 16 I. K. Song and M. A. Barteau, J. Mol. Catal. A: Chem., 2004, 212, 229–236. 17 M. N. Timofeeva, Appl. Catal., A, 2003, 256, 19–35. 18 M. Ammam, J. Mater. Chem. A, 2013, 1, 6291–6312. 19 R. Chiarizia, D. C. Stepinski and P. Thiyagarajan, Sep. Sci. Technol., 2006, 41, 2075–2095. 20 R. Thangamuthu, Y. C. Pan and S. M. Chen, Electroanalysis, 2010, 22, 1812–1816. 21 D. P. Acharya and P. G. Hartley, Curr. Opin. Colloid Interface Sci., 2012, 17, 274–280. 22 L. Cronin, P. I. Molina and H. N. Miras, in Solar Hydrogen and Nanotechnology VI, ed. Y. Tachibana, Proceedings of SPIE, Bellingham, WA, 2011, vol. 8109, p. 8109C. 23 W. B. Kim, T. Voitl, G. J. Rodriguez-Rivera and J. A. Dumesic, Science, 2004, 305, 1280–1283. 24 M. Kourasi, R. G. A. Wills, A. A. Shah and F. C. Walsh, Electrochim. Acta, 2014, 127, 454–466. 25 H. D. Pratt, N. S. Hudak, X. K. Fang and T. M. Anderson, J. Power Sources, 2013, 236, 259–264. 26 M. D. Symes and L. Cronin, Nat. Chem., 2013, 5, 403–409. 27 H. B. Wang, Y. Yan, B. Li, L. H. Bi and L. X. Wu, Chem. – Eur. J., 2011, 17, 4273–4282. 28 M. K. Bera, T. H. Bray, R. J. Ellis and M. R. Antonio, ChemElectroChem, 2014, 1, 1173–1181. 29 M. R. Antonio, R. Chiarizia and F. Jaffrennou, Sep. Sci. Technol., 2010, 45, 1689–1698. 30 S. P. Moulik and B. K. Paul, Adv. Colloid Interface Sci., 1998, 78, 99–195. 31 M. Nilsson, P. R. Zalupski and M. R. Antonio, in Ion Exchange and Solvent Extraction. Supramolecular Aspects of Solvent Extraction, ed. B. A. Moyer, CRC Press, Boca Raton, 2014, ch. 6, vol. 21, pp. 197–268.


View Article Online


PCCP

32 K. Osseo-Asare, in Emerging Separation Technologies for Metals II, ed. R. G. Bautista, The Minerals, Metals & Materials Society, Warrendale, PA, 1996, pp. 159–169. 33 K. Osseo-Asare, in Proceedings of the International Solvent Extraction Conference, ISEC 2002, ed. K. C. Sole, P. M. Cole, J. S. Preston and D. J. Robinson, South African Institute of Mining and Metallurgy, Johannesburg, 2002, pp. 118–124. 34 M. J. Schwuger, K. Stickdorn and R. Schomacker, Chem. Rev., 1995, 95, 849–864. 35 H. Watarai, J. Chromatogr. A, 1997, 780, 93–102. 36 R. Chiarizia and A. Briand, Solvent Extr. Ion Exch., 2007, 25, 351–371. 37 R. Chiarizia, A. Briand, M. P. Jensen and P. Thiyagarajan, Solvent Extr. Ion Exch., 2008, 26, 333–359. 38 R. Chiarizia, M. P. Jensen, M. Borkowski, J. R. Ferraro, P. Thiyagarajan and K. C. Littrell, Solvent Extr. Ion Exch., 2003, 21, 1–27. 39 R. Chiarizia, P. G. Rickert, D. Stepinski, P. Thiyagarajan and K. C. Littrell, Solvent Extr. Ion Exch., 2006, 24, 125–148. 40 F. Testard, T. Zemb, P. Bauduin and L. Berthon, in Ion Exchange and Solvent Extraction. A Series of Advances, ed. B. A. Moyer, CRC Press, Boca Raton, 2010, ch. 7, vol. 19, pp. 381–428. 41 M. Ammam, B. Keita, L. Nadjo, I. M. Mbomekalle and J. Fransaer, Electrochim. Acta, 2010, 55, 3213–3222. 42 H. Wu, J. Biol. Chem., 1920, 43, 189–220. 43 R. Chiarizia, M. P. Jensen, M. Borkowski, P. Thiyagarajan and K. C. Littrell, Solvent Extr. Ion Exch., 2004, 22, 325–351. 44 R. Chiarizia, M. P. Jensen, P. G. Rickert, Z. Kolarik, M. Borkowski and P. Thiyagarajan, Langmuir, 2004, 20, 10798–10808. 45 R. Chiarizia, K. L. Nash, M. P. Jensen, P. Thiyagarajan and K. C. Littrell, Langmuir, 2003, 19, 9592–9599. 46 R. Chiarizia, D. Stepinski and M. R. Antonio, Sep. Sci. Technol., 2010, 45, 1668–1678. 47 S. Nave, C. Mandin, L. Martinet, L. Berthon, F. Testard, C. Madic and T. Zemb, Phys. Chem. Chem. Phys., 2004, 6, 799–808. 48 J. Plaue, A. Gelis, K. Czerwinski, P. Thiyagarajan and R. Chiarizia, Solvent Extr. Ion Exch., 2006, 24, 283–298. 49 G. M. Brown, M. R. Noe-Spirlet, W. R. Busing and H. A. Levy, Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem., 1977, 33, 1038–1046.


Paper

50 P. Beudaert, V. Lamare, J. F. Dozol, L. Troxler and G. Wipff, Solvent Extr. Ion Exch., 1998, 16, 597–618. 51 A. Chaumont and G. Wipff, Phys. Chem. Chem. Phys., 2008, 10, 6940–6953. 52 A. Chaumont and G. Wipff, C. R. Chim., 2012, 15, 107–117. 53 A. Chaumont and G. Wipff, Eur. J. Inorg. Chem., 2013, 1835–1853. 54 M. T. Pope, in Applications of Coordination Chemistry, ed. M. Ward, Elsevier, Oxford, 2004, vol. 4, pp. 635–678. 55 J. J. Cowan, C. L. HiII, R. S. Reiner and I. A. Weinstock, Inorg. Synth., 2002, 33, 18–26. 56 M. R. Antonio, M. H. Chiang, S. Seifert, D. M. Tiede and P. Thiyagarajan, J. Electroanal. Chem., 2009, 626, 103–110. 57 M. R. Antonio, M. Nyman and T. M. Anderson, Angew. Chem., Int. Ed., 2009, 48, 6136–6140. 58 Small Angle X-ray Scattering, ed. O. Glatter and O. Kratky, Academic Press, London, 1982. 59 B. Lindman, K. Shinoda, U. Olsson, D. Anderson, G. Karlstrom and H. Wennerstrom, Colloids Surf., 1989, 38, 205–224. 60 U. B. Mioc, M. R. Todorovic, A. Davidovic, P. Colomban and I. Holclajtner-Antunovic, Solid State Ionics, 2005, 176, 3005–3017. 61 P. B. Moore, J. Appl. Crystallogr., 1980, 13, 168–175. 62 M. A. Firestone, J. A. Dzielawa, P. Zapol, L. A. Curtiss, S. Seifert and M. L. Dietz, Langmuir, 2002, 18, 7258–7260. 63 A. Triolo, O. Russina, H. J. Bleif and E. Di Cola, J. Phys. Chem. B, 2007, 111, 4641–4644. 64 P. Bauduin, F. Testard, L. Berthon and T. Zemb, Phys. Chem. Chem. Phys., 2007, 9, 3776–3785. 65 Y. Meridiano, L. Berthon, X. Crozes, C. Sorel, P. Dannus, M. R. Antonio, R. Chiarizia and T. Zemb, Solvent Extr. Ion Exch., 2009, 27, 607–637. 66 A. A. Varnek, A. N. Kuznetsov, O. M. Petrukhin and R. P. Ozerov, J. Struct. Chem., 1991, 32, 130–133. 67 A. V. Nikolaev, Y. A. Afanas’ev, V. B. Durasov and A. I. Ryabinin, J. Struct. Chem., 1964, 5, 449–451. 68 V. I. Lakshmanan and B. C. Haldar, J. Indian Chem. Soc., 1970, 47, 72–78. 69 X. K. Lin, W. Li, J. Zhang, H. Sun, Y. Yan and L. X. Wu, Langmuir, 2010, 26, 13201–13209. 70 J. Zhang, W. Li, C. Wu, B. Li and L. X. Wu, Chem. – Eur. J., 2013, 19, 8129–8135.


Kinetics of pressure induced structural changes in super- or near-critical CO2-microemulsions.

Structural aspects of proton-pumping ATPases.

Structural aspects of the inactive X chromosome.

Structural aspects of protein-DNA recognition.

Structural aspects of pathology in Alzheimer's disease.

Phenomenological theories of globular microemulsions.

Some structural aspects of deviant child behavior.

Structural and functional aspects of RNA pseudoknots.

Structural Aspects of GPCR-G Protein Coupling.

Structural aspects of human cervical mucus.

Pancreatic colipase. Structural and physiological aspects.

Aspects in structural studies on ribosomes.

isooctane microemulsions.

Aspects of structural landscape of human islet amyloid polypeptide.

Three-dimensional structural aspects of protein-polysaccharide interactions.

Chemical and Structural Aspects of Ebola Virus Entry Inhibitors.

Initiation of mRNA translation in bacteria: structural and dynamic aspects.

Structural and functional aspects of biological freezing techniques.

Structural aspects of recognition motifs contributing to autoimmune responses.

Structural aspects of bovine oocyte maturation in vitro.

Biochemical, structural, and molecular genetic aspects of halophilism.

Structural aspects of Plasmodium relevant to vaccination against malaria.

Structural and dynamics aspects of ASC speck assembly.

Structural and functional aspects of the thyroid follicular epithelium.