ARTICLE pubs.acs.org/Langmuir
Effect of Metal Ion Hydration on the Interaction between Sodium Carboxylates and Aluminum(III) or Chromium(III) Ions in Aqueous Solution Rui F. P. Pereira,† Maria J. Tapia,‡ Artur J. M. Valente,† and Hugh D. Burrows*,† † ‡
Department of Chemistry, University of Coimbra, 3004-535 Coimbra, Portugal Department of Chemistry, University of Burgos, Plaza Misael Ba~nuelos s/n, Burgos 09001, Spain
bS Supporting Information ABSTRACT: The interaction between sodium octanoate, decanoate, and dodecanoate and aluminum(III) and chromium(III) has been studied in water at natural pH values, starting well below the surfactant critical micelle concentration, using electrical conductivity, turbidity, and potentiometric measurements. With decanoate or dodecanoate, maximum interaction occurs at 3:1 stoichiometry, corresponding to charge neutralization. Although the solutions become turbid with both metal ions, indicating phase separation, differences are observed and attributed to the fact that aluminum(III) is relatively labile to substitution and rapidly replaces its water ligands, whereas chromium(III) is substitution inert. This shows up in well-defined floc formation with Al3+, whereas Cr3+ suspensions do not precipitate, probably because that replacement of coordinated water by carboxylate ligands is impeded. This can be overcome by increasing temperature, and differences in the thermal behavior with Al3+ and Cr3+ are suggested to be due to increased involvement of substitution reactions in the latter case. The effect of octanoate on the trivalent metal ions is less clear, and with Cr3+ interaction only occurs when the carboxylate is in excess. Hydrophobic interactions between alkyl chains play a major role in driving phase separation. At high surfactant concentrations, the solid phases do not dissolve, in contrast to what is observed with the corresponding alkylsulfates. This has implications for use of these systems in metal separation through froth flotation. The concentration of metal ions in supernatant solution has been determined for sodium dodecanoate and sodium dodecylsulfate with Al3+ and Cr3+ over the whole surfactant concentration range by inductively coupled plasma-mass spectrometry (ICP-MS). From this, association constants have been determined and are found to be larger for the carboxylate than the alkylsulfate, in agreement with the greater Lewis basicity of the CO2 group.
’ INTRODUCTION Colloidal systems involving trivalent metal ions have applications in detergency,1 wastewater treatment,2,3 gel formation and development of thickeners and dispersants,4,5 catalysis,6,7 analytical chemistry,8 froth flotation for metal ion recovery,911 and contrast agents for magnetic resonance imaging.12 These systems also have considerable potential in materials science, including templated synthesis of mesoporous materials,13,14 preparation of nanoparticles,15 and formation of metalorganic frameworks (MOFs).16 Interactions between trivalent metal ions and anionic surfactants in aqueous solutions can lead to solid or floc formation at a stoichiometry corresponding to electrical neutrality in a process which depends on surfactant alkyl chain length and concentration. Solids formed from interaction between metal ions and long chain surfactants frequently have low melting points and, because of their limited volatility, have the potential for applications as ionic liquids.17 In the case of alkylsulfates, the solids show a lamellar structure and the redissolution starts in the presence of an excess of surfactant at concentrations below the critical micelle concentration (cmc) of the pure surfactant. r 2011 American Chemical Society
For aqueous solutions of long chain sodium alkyl carboxylates, addition of salts of trivalent ions almost invariably leads to precipitation due to formation of the trivalent metal carboxylate (the so-called, metal soap). As discussed elsewhere18,19 the stoichiometry of these systems depends on both the nature of the metal ion and the precipitation conditions (pH, temperature, etc.), and there are possibilities of forming neutral ((RCO2)3M), acid ((RCO2)3M) 3 xRCO2H), or basic ((RCO2)x(OH)3xM) soaps. With the important aluminum(III) system, precipitation from aqueous solutions of long chain sodium or potassium alkyl carboxylates involves a complex pH-dependent process, which forms mainly a basic aluminum(III) dicarboxylate, with differing degrees of polymerization.2024 These produce gels in organic solvents, one of the best known being Napalm.25 Although it has generally been believed that gelation involves formation of a Received: August 31, 2011 Revised: November 17, 2011 Published: November 22, 2011 168
dx.doi.org/10.1021/la2034164 | Langmuir 2012, 28, 168–177
Langmuir
ARTICLE
polymeric, chain-like structure through the hydroxyl groups,26 there are recent indications that formation of networks of nanosized micelles might be involved instead.27 Given the importance of this area for development of novel alumina-based ceramics,28 there is a clear need for further research on these systems. The precipitates formed by interaction of chromium(III) with alkali metal carboxylates in aqueous solution generally also involve basic soaps, possibly with an excess of acid.18,29 The two cations Al3+ and Cr3+ have been chosen to encompass an important property of hydrated trivalent metal ions, the rate of substitution of ligands within the primary coordination sphere of the metal ion. These ions are present in water as the hexaaquo form, [M(H2O)6]3+, or some hydrolyzed species derived from it and have the same general physical properties, including ionic radii,30 primary3134 and secondary3234 hydration numbers, hydration enthalpies,35 free energies,36 and acid dissociation constants of coordinated water.37 However, they differ kinetically in their tendency to lose water in their primary hydration sphere, and the exchange of water in the primary hydration sphere of Cr3+ is at least 6 orders of magnitude slower than with Al3+.38 This allows the assessment of the effect of hydration on possible interactions and on the structure of any resulting aggregate. The interaction of aluminum and chromium with sodium carboxylates of different chain lengths (sodium octanoate, sodium decanoate, and sodium dodecanoate) has been studied using electrical conductometry, turbidity, potentiometry, and inductively coupled plasma-mass spectrometry. These were studied with constant metal ion concentration (1 mM), starting with surfactant concentrations 23 orders of magnitude below their cmc’s to avoid competing effects resulting from direct interaction between metal ions and surfactant micelles. This allows us to have a better understanding of the role of the metal ion hydration and surfactant hydrophobicity on the mechanism of trivalent metal ionalkanoate interaction. Since electrostatic or other headgroup effects are expected to be similar with all three systems, this implies a major role for hydrophobic interactions between the alkyl chains in driving precipitation. An interaction mechanism will be proposed and association constants estimated. In addition, sodium carboxylates are salts of weak acids,39 and experimental data for these metal ions will be compared with those obtained for the strong acid salt sodium dodecyl sulfate (SDS), in order to shed light on the effect of headgroup. Theoretical studies show significant differences in charge distribution40 and cation binding41 between carboxylate and sulfate groups.
Millipore-Q water as solvent. The presence of residual CO2 in solutions does not affect the solution behavior, as checked by electrical conductivity measurements. No control was made on the pH, which was the natural value for each solution. Conductance Measurements. Electrical conductance measurements were carried out with a Wayne-Kerr model 4265 automatic LCR meter at 1 kHz, through the recording of solution electrical resistances, measured by a conductivity cell with a constant of 0.1178 cm1, uncertainty 0.02%.43 The cell constant was determined from electrical resistance measurements with KCl (reagent grade, recrystallized, and dried) using the procedure and data of Barthel et al.44 Measurements were taken at various temperatures ((0.02 °C) in a Thermo Scientific Phoenix II B5 thermostat bath. In a typical experiment, 20 mL of metal ion solution 1 mM was placed in the conductivity cell; then aliquots of the surfactant solution were added in a stepwise manner using a Methrom 765 Dosimate micropipet. The specific conductance of the solution was measured after each addition and corresponds to the average of three ionic conductance measurements (uncertainty less than 0.2%), determined using homemade software. The specific electrical conductance of the solutions, k, is calculated from the experimental specific conductance, kexp, corrected for the specific conductance of water, k0: k = kexp k0. Turbidity Measurements. Turbidity was measured by optical transmittance at 550 nm using a Shimadzu UVvisible 2450 spectrophotometer. In a typical experiment, different amounts of surfactant were added to a 1 mM metal ion solution with continuous stirring for about 5 min before each transmittance measurement. pH Measurements. Potentiometric measurements were carried out with a pH Radiometer PHM 240. The pH was measured on fresh solutions with an Ingold U457-K7 pH conjugated electrode and calibrated immediately before each experimental set of solutions using IUPAC-recommended pH 4 and 7 buffers. In a typical experiment, using a Methrom 765 Dosimate micropipet, aliquots of the surfactant solution were added to 20 mL of metal ion solution. All measurements were carried out at 25.00 °C ((0.02 °C), and the electrode potential was recorded after signal stabilization. Inductively Coupled Plasma-Mass Spectrometry. In a typical experiment, different amounts of sodium dodecyl sulfate or sodium dodecanoate (from 0 to 15 mM) were added to a 1 mM metal ion solution. At least 20 samples were prepared in the surfactant concentration range of interest. All samples were continuously stirred for 6 h and then left to equilibrate for 12 h (in the samples where flocculation occurs, a clear twophase separation can be observed). Aliquots for ICP-MS analysis were taken from the supernatant after samples were centrifuged at 3500 rpm for 15 min. Solutions were pipetted into a polypropylene tube and thoroughly mixed with an yttrium internal standard solution. They were then diluted with 2% (w/w) aqueous nitric acid (HNO3, purum p.a. from Fluka) solution to obtain the necessary final analytical metal concentration (between 10 and 30 ppb). Before each set of measurements, calibration was carried out using metalcontaining standard solutions; an aqueous solution of 2% (w/w) HNO3 was used both as a blank and for rinsing the instrument after the highest concentration standard solution measurement. Each sample was injected in triplicate, and the reported value corresponds to an average value of those three measurements. ICP-MS analysis was performed using an Agilent 7500i instrument. Thermogravimetric Analysis. Samples for thermogravimetric analysis were transferred to open platinum crucibles and analyzed using a TA Q50 Thermogravimetric Analyzer at a heating rate of 10 °C min1 using dried nitrogen as purge gas (40 and 60 mL min1 in the balance and sample chamber, respectively).
’ MATERIALS AND METHODS Reagents and Sample Preparation. Aluminum(III) nitrate nonahydrate (98%) and chromium(III) nitrate nonahydrate (99%) were purchased from Fluka and Aldrich, respectively. Sodium octanoate (99%), sodium decanoate (98%), and sodium dodecanoate (99100%) were supplied by Sigma. For the sake of simplicity, these surfactants will be labeled C7COONa, C9COONa, and C11COONa. All experiments have been carried out at concentrations below the surfactant critical micelle concentration: 340, 94, and 24 mM, respectively.42 Sodium dodecyl sulfate, SDS (g98%), was purchased from Sigma. Purity was confirmed by measurement of its cmc (8.34((0.03) mM), which was identical to literature values. These reagents were used as received; all solutions were prepared immediately before the experiments using
’ RESULTS AND DISCUSSION Electrical Conductivity. The study and interpretation of electrical conductivity of aqueous solutions of high-valent metal 169
dx.doi.org/10.1021/la2034164 |Langmuir 2012, 28, 168–177
Langmuir
ARTICLE
trivalent metal ion salt, kM, in aqueous solutions. In addition, the total experimental conductance will depend upon a third contribution, ΔkSM, which describes species resulting from metal ionsurfactant interactions, such that k = kS + kM + ΔkSM. Figure 1 shows the effect of metal ion and alkyl chain length on the electrical conductance interaction (metal ionsodium carboxylate) contribution (ΔkSM) for Al3+ and Cr3+ solutions. Various effects are observed. First, addition of n-alkylcarboxylates to Al3+ or Cr3+ solutions leads to a decrease in conductivity (i.e., negative ΔkSM values), suggesting either formation of larger and less mobile structures or a reduction of the ionic strength. For C9COONa and C11COONa systems with both metal ions, the decrease of ΔkSM occurs up to the molar ratio of 3. At higher molar ratios, ΔkSM reaches a plateau, meaning that the specific electrical conductances of the mixed solutions are similar to those observed with surfactant solutions in the absence of metal ions. These results clearly imply that carboxylatemetal ion interactions are controlled by charge neutralization50 and that the interaction involves removal of free metal ion from solution, similar to what was observed with trivalent metal ions and decyl and dodecyl sulfates.51,52 Completely different behavior is observed with sodium octanoate-containing systems, where there is a continuous decrease of ΔkSM as a function of [CnCOONa] over the whole range of molar ratios (from 0 to 24). A shoulder is observed at a molar ratio around 1, which is less pronounced in the case of Al3+. Further analysis of Figure 1 shows that the effect of alkyl carboxylates on the Cr3+ solution is higher than in the case of aqueous Al3+ systems, leading to more negative ΔkSM. This will be discussed later. Turbidity. To obtain further information on the mechanism behind interactions between sodium alkylcarboxylates and Cr3+ and Al3+, turbidity measurements have been used to test for formation and growth of aggregates. It is found that adding CnCOONa to Al3+ for [CnCOONa] < 3[Al3+] leads to formation of well-defined flocs, which sediment after a certain period of time, leaving a completely transparent supernatant solution (the optical transmittance at 550 nm is around 100% at [CnCOONa]/ [Al3+] = 3). Upon increasing sodium n-alkylcarboxylate concentration ([CnCOONa] > 3[Al3+]), the solution becomes milky. This process is independent of the carboxylate alkyl chain length. The effect of addition of alkylcarboxylates on the turbidity of 1 mM chromium(III) solution is rather different (Figure 2); the turbidity for CnCOONa/Cr3+ is dependent on both carboxylate hydrophobicity and the [CnCOONa]/[Cr3+] molar ratio. The dependence of optical transmittance on the molar ratio follows the order C11 COONa > C 9 COONa > C 7 COONa. The turbidity for C11COONa and C9COONa reaches a maximum at [CnCOONa] = 3[Cr3+] and for molar ratios greater than 3 remains constant. This behavior is consistent with the electrical conductance data. However, for C7COONa the solution only becomes milky for [CnCOONa] > 3[Cr3+]. It should be stressed that, in contrast to the Al3+ systems, no well-defined flocs are formed by Cr3+. Instead, these mixed solutions show a milky color. This indicates different interactions between alkyl carboxylates and Cr3+ than those involving Al3+, possibly due to the high kinetic stability of hydrated Cr3+ species. From electrical conductivity and turbidity measurements we can conclude that (a) interaction between decanoate and dodecanoate and trivalent metal ions shows a 3:1 stoichiometry, with Cr3+ systems being more affected by the hydrophobicity of the surfactant alkyl chain than Al3+ ones and (b) the effect of
Figure 1. Dependence of the interaction parameter contribution of the electrical conductance of mixed solutions upon the sodium alkylcarboxylate-trivalent metal ion molar ratio at 25 °C: (a) [Al3+] = 1 mM and (b) [Cr3+] = 1 mM; (0) n = 7, (O) n = 9, and (Δ) n = 11.
ions is not a simple task due to the presence of a variety of species as a consequence of hydrolysis and other interactions.45 This is further complicated in the presence of ionic surfactants. In particular, micellization at higher concentrations leads to changes in the ionic strength of solution, with consequent changes in the degree of counterion dissociation.46 However, electrical conductivity is still a valuable tool for assessment of the effect of ionic47 or nonionic surfactants48 on the structure of ionic solutions, since it allows observation of complex behavior of electrolyte solutions, resulting from changes in the size and shape of moving particles and/or effective particle changes.49 Consequently, electrical conductance measurements were carried out to evaluate the overall effect on the aqueous solutions of aluminum(III) or chromium(III) nitrates of addition of sodium alkyl carboxylates. For the reasons described above, a complete analysis of all the different contributions to the experimental electrical conductance is outside the scope of this paper. However, it can be established that the experimental electrical conductance, after correction for the solvent contribution, k, is markedly dependent upon the concentrations and nature of sodium alkyl carboxylate, kS, and 170
dx.doi.org/10.1021/la2034164 |Langmuir 2012, 28, 168–177
Langmuir
ARTICLE
Figure 2. Turbidity of mixed CnCOONa/Cr(NO3)3 solutions at different molar ratios [CnCOONa]/[Cr3+] with [Cr3+] = 1 mM: (0) n = 7, (O) n = 9, and (Δ)) n = 11.
octanoate on the trivalent metal ions is not yet well understood but data on Cr3+ system suggests that the interaction only occurs when the carboxylate is in large excess. Probably hydrophobic interaction between alkylcarboxylate chains is the major driving force in phase separation. This will involve mainly van der Waal’s forces, which increase with alkyl chain length.53 Cr3+ is kinetically inert (hydrated species have lifetimes of several days) and will be present in water as [Cr(H2O)6]3+ or a hydrolyzed species derived from it, while Al3+ is kinetically labile and the water molecules in the primary hydration sphere can readily be substituted by carboxylate ligands.32,54 This difference in behavior can explain the differences between Cr3+ and Al3+ in their interactions with the carboxylates. Comparison of the data shown in Figures 1 and 2 with those describing interactions between these metal ions and the analogous alkylsulfate surfactants55,56 indicates two major differences: the interaction stoichiometry and absence of redissolution of the formed precipitate at higher surfactant concentrations in the case of alkylcarboxylates. With the carboxylates the stoichiometry of, e.g., dodecanoate and trivalent metal ion interaction perfectly matches the charge neutralization, while for the analogous alkylsulfates the interaction occurs at 4:1 stoichiometry. This has been explained in terms of formation of lamellar aggregates, which are stabilized by interaction with excess surfactant.6,57 Several authors have reported that formation of trivalent metal ion (lanthanide) carboxylates also form lamellar structures.16,58 Our data hint that the packing is somewhat different with the carboxylate soaps, which can be explained by shorter interaction distances between oxygen atoms from the carboxylate group and metal ion with subsequent greater stability compared with the alkylsulfate.41 This is probably due to the carboxylate group being a stronger Lewis base than the sulfate one.41 Both these facts also explain why formation of Al3+ or Cr3+ alkylcarboxylates is not followed by a redissolution of the aggregate when the surfactant is in stoichiometric excess, in complete contrast to what happens with the corresponding alkylsulfate-based aggregates, even for a surfactant concentration below the critical micelle concentration.56 Potentiometric Titration. Further information was obtained from potentiometric titrations. Figure 3 shows the pH titration
Figure 3. Effect of surfactant chain length on the pH of 1 mM aqueous solutions of Cr3+ (a) and Al3+ (b) at 25 °C: (0) n = 7, (O) n = 9, and (Δ) n = 11.
curves for 1 mM chromium(III) nitrate (Figure 3a) and aluminum(III) nitrate (Figure 3b) with sodium carboxylates. The initial pH of an aqueous solution of 1 mM chromium(III) nitrate is 3.4, indicating that the main Cr3+ species present in solution is [Cr(H2O)6]3+.37 Hexahydrated chromium(III) will undergo ionization according to ½CrðH2 OÞ6 3þ þ H2 O a ½CrðH2 OÞ5 ðOHÞ2þ þ H3 Oþ
ð1Þ
which explains the acidic pH of the metal ion solution. This reaction is kinetically favorable (1.4 105 s1) when compared with elimination of a water molecule from the coordination sphere (k ≈ ca. 105106 s1).37 Further, the carboxylates are strong Lewis bases but can also act as Br€onsted bases, and consequently, in aqueous solution the following equilibrium takes place Cn COO þ H2 O a Cn COOH þ OH 171
ð2Þ
dx.doi.org/10.1021/la2034164 |Langmuir 2012, 28, 168–177
Langmuir
ARTICLE
Since the titration curves (Figure 3a) show an increase of pH upon addition of carboxylates, a plausible mechanism corresponds to the acidbase neutralization taking place
following reactions ½AlðH2 OÞ6 3þ þ Cn COO a ½AlðH2 OÞ5 ðCn COOÞ2þ þ H2 O
ð4Þ ½CrðH2 OÞ6 3þ þ Cn COO a ½CrðH2 OÞ5 ðOHÞ2þ þ Cn COOH
½AlðH2 OÞ5 ðCn COOÞ2þ þ Cn COO
ð3Þ
a ½AlðH2 OÞ4 ðCn COOÞ2 þ þ H2 O
This process occurs until complete charge neutralization of Cr3+. Probably the hydroxychromium species will then undergo further hydrolysis to form oligomers Crx(OH)y(xy). This explains the decrease in both electrical conductivity as well as the turbidity measurements. The neutralization will become stronger when the hydrophobicity of carboxylate is higher (and hydrolysis becomes more relevant). This is also supported by measurements at the stoichiometric excess of carboxylate, relative to Cr3+ concentration. The pH of the resulting mixed solutions is in agreement with carboxylate pKb (9.11, 9.10, and 8.70 for C7COO, C9COO, and C11COO, respectively).59 Thus, for the carboxylate:metal ratio r > 3, the behavior is similar for all carboxylates: pH increases with addition of surfactant, following a behavior similar to what occurs in the absence of trivalent ions (see Figure 1 of the Supporting Information60,61), and hydrolysis of alkanoates increases with increasing alkyl chain length. If we combine these results with the electrical conductance behavior for C7COONa/Cr3+, a possible explanation can now be attempted for the continuous decrease of electrical conductance upon increasing surfactant concentration. Since the interaction is very weak, which is supported by the fact that the variations of pH with C7COONa concentration in the presence and absence of Cr3+ are quite similar, the electrical conductance measurements probably just reflect the pure effect of the ionic strength on the hydrolysis constant.62 The same arguments can be also considered for the numerically larger ΔkSM values observed for C9COONa:Cr3+ and C11COONa:Cr3+ systems when compared with the corresponding Al3+ ones. It is also worthy of note that different clear steps of neutralization can be observed upon increasing the carboxylate hydrophobicity. For example, for C11COONa two “equivalence” points can be observed (see arrows at Figure 3a). Although we cannot exclude the idea that some neutralization takes place upon floc formation, it is noteworthy that these points occur at molar ratios r = 1 and 3, suggesting that at these stages the main processes are a first neutralization at stoichiometric ratio 1:1, while the second and third neutralizations are occurring simultaneously. A final remark on the interpretation of these results is that although there is very slow exchange of water molecules from the metal ion primary hydration shell, possible formation of chromium(III) carboxylates or basic carboxylates cannot be ruled out. These have been identified in the solid state.28 A more complex interaction mechanism is needed to analyze the results of titration of Al3+ with carboxylates. The pH of 1 mM aqueous solutions of aluminum(III) nitrate is 4.2, below the first hydrolysis constant (log Q 4.97), indicating that the main species present in solution is the hexahydrated aluminum(III). The interaction mechanism between carboxylates and Al3+ must be different from that with Cr3+ since addition of carboxylates leads to a decrease of pH up to a molar ratio r = 3. For Al3+ the water exchange rate at 298 K is 1.29 s1 (much higher than that for Cr3+, 2.4 106 s1),38 and it can be anticipated that Al3+ is likely to undergo substitution of water by carboxylate. That can be represented by the
ð5Þ
½AlðH2 OÞ4 ðCn COOÞ2 þ þ Cn COO a ½AlðH2 OÞ3 ðCn COOÞ3 þ H2 O
ð6Þ
Equations 46 are written ignoring formation of oligomers and on the assumption of monodentate ligand binding, which may not be the case. However, the above hypothesis is supported by similar pH behavior observed for the interaction between a less strong base (dodecyl sulfate) and Al3+ leading to formation of aluminum tridodecylsufate.55 Such behavior also justifies the decrease in the electrical conductance as well as the strong formation of flocs observed directly and by turbidity. It is expected that those reactions occur in parallel with normal hydrolysis of Al3+ ½AlðH2 OÞ6 3þ þ H2 O a ½AlðH2 OÞ5 ðOHÞ2þ þ H3 Oþ ð7Þ ½AlðH2 OÞ5 ðCn COOÞ2þ þ H2 O a ½AlðH2 OÞ4 ðOHÞðCn COOÞþ þ H3 Oþ
ð8Þ
and hydrolysis of CnCOO (eq 2). Consequently, we can suggest that the observed initial decrease in the pH can be due to an increase of pKa of eq 8 when compared with that in eq 7. Although we have not been able to find data on how carboxylate affects the pKa of the coordinated water with Al3+, in support of the above hypothesis, it is suggested from theoretical studies that for Zn2+ 63 and Mg2+ 64 coordinated carboxylate ligands increase the pKa relative to the hexaaquo ion. Although there may be differences in the way carboxylate binds to Al3+ compared with Mg2+ 65 and there is the possibility of forming oligomers, we believe that this provides a reasonable justification for the observed pH behavior of Al3+ solution after addition of carboxylates. An alternative explanation for the pH decrease in this zone comes from the breakup of dimers/oligomers of carboxylates following the interaction with trivalent metals, with the consequent release of hydronium ions.61 Finally, we consider the case of aqueous solutions of sodium octanoate and Al3+. From the electrical conductivity study it appears that octanoate has a low ability to interact with trivalent metal ions, and even for Al3+ it is necessary to reach a molar ratio of 1 for the interaction to take place. It is worth noting that below this molar ratio the pH increases slightly, suggesting that an acidbase neutralization is occurring just before formation of aluminum(III) octanoates. A likely explanation is that in all cases the interactions between Al(III) and the carboxylate groups are relatively weak and, as suggested earlier, precipitation is driven by hydrophobic interactions between that alkyl chains. Thermogravimetric Analysis of the Solid Phase. The above results all indicate different mechanisms for the interaction of long chain carboxylates with hydrated Cr(III) and Al(III) ions in water. This is expected to lead to different solid products. 172
dx.doi.org/10.1021/la2034164 |Langmuir 2012, 28, 168–177
Langmuir
ARTICLE
The solid precipitates obtained from addition of sodium dodecanoate to aqueous solutions of aluminum(III) and chromium(III) nitrate (1 mM) were filtered off, dried, and analyzed by thermogravimetry (TG) over the range from room temperature to 900 °C (Figure S3, Supporting Information). Thermal degradation of trivalent metal long chain carboxylates normally leads to formation of carbon dioxide, dialkylketone, and the metal oxide (M2O3).19 The experimental and calculated weight percent of the solid residue at 900 °C based on the reaction 2ðRCO2 Þ3 M f 3R 2 CO þ 3CO2 þ M2 O3
ð9Þ
are 10.5 (exp.), 11.7 (calcd) for chromium and 5.4 (exp.), 8.2 (calcd) for aluminum. These indicate the same 3:1 stoichiometry for the solid products as detected in solution. However, as discussed elsewhere,19 this does not necessarily indicate formation of the pure metal tricarboxylates. It is generally accepted that reaction of sodium alkylcarboxylates with chromium(III) or aluminum(III) in aqueous solutions leads to mixtures of basic hydroxycarboxylates (M(OH)x(RCO2)y) with the corresponding carboxylic acids, and this can lead to the same stoichiometric ratio. Comparison of the TG thermograms for the chromium and aluminum derivatives shows marked differences, indicating different structures for the two solid products. In both cases, an initial very small weight loss was observed around 100 °C, which probably corresponds to absorbed water. However, with the chromium(III) derivative, four weight loss steps were observed between 200 and 500 °C, whereas with the aluminum(III) compound there were only three steps. Detailed studies of the thermal degradation of chromium(III) soaps have previously been reported.29,66,67 It was suggested that the main species precipitating from aqueous solutions of the dodecanoate is the basic Cr(OH)(OOCC11)2, which decomposes thermally in two steps.67 This is expected to be present with an equimolar quantity of dodecanoic acid to give the 3:1 stoichiometry observed in our system. In addition, at least one chromium(III) coordination water must be present to maintain the hexacoordinate octahedral geometry of this ion.68 Thus, the extra decomposition steps observed are likely to correspond to processes related to the loss of an extra carboxylic acid and coordinated water. Pure aluminum(III) tridodecanoate shows a single decomposition step above 200 °C.69 This is seen in our thermogram, in addition to two weight loss steps at higher temperature. Although detailed determination of the structures of the products from reaction of sodium carboxylates with chromium(III) and aluminum(III) is outside the scope of our study, the TG results indicate the precipitated solids are different in the two cases. Temperature Effect on Al3+ and Cr3+:C11COO Interactions. From the above, we can see that the mechanism of interaction between alkylcarboxylates and Al3+ or Cr3+ is markedly dependent on the hydration properties of these metal ions, and in addition, it is found that the interaction is increased upon increasing the pKa and hydrophobicity of the carboxylate. To obtain further information, the effect of temperature has been studied on the electrical conductivity of solutions containing sodium dodecanoate and trivalent metal ions. Figure 4 shows the effect of temperature on addition of sodium dodecanoate to Al3+ and Cr3+ solutions. With the specific conductance at the maximum interaction (which occurs at r = 3) (Figure 5) it can be seen that in the case of Al3+ there is a monotonic increase in this value as a function of temperature. This behavior is similar to that occurring in aqueous solutions,62
Figure 4. Temperature effect (25, 30, 40, 50, and 60 °C) in Al3+ (a) and Cr3+ (b) interaction with sodium dodecanoate, seen by electrical conductivity measurements.
clearly suggesting that the mechanism does not change with temperature. However, with the Cr3+-based systems the behavior is rather different and the specific conductance at r = 3 remains constant for temperatures ranging from 25 to 60 °C. A simple way to explain this phenomenon is to consider that the charge neutralization becomes more effective upon increasing temperature. The substitution characteristic of Cr3+ ions is due to the slow exchange rate of water molecule in the hydration shell;38 this process is activated by temperature.70 Consequently, it can be expected that upon increasing temperature the exchange of water is easier and, consequently, the interaction with dodecanoate can now occur in a similar fashion to Al3+, leading to more pronounced charge neutralization. Inductively Coupled Plasma-Mass Spectrometry: Quantitative Analysis. Inductively coupled plasma-mass spectrometry allows quantification of metal ion concentration in the supernatant as a function of surfactant-to-metal ion concentration ratio. ICP-MS experiments were made to determine the concentration of Al3+ and Cr3+ species in the supernatant in the presence of surfactants. 173
dx.doi.org/10.1021/la2034164 |Langmuir 2012, 28, 168–177
Langmuir
ARTICLE
Figure 5. Temperature effect on the specific conductance of Al3+: C11COONa (0) and Cr3+:C11COO (Δ) mixed solutions at maximum interaction (r = 3).
Figure 6. Dependence of concentration of supernatant Al3+ on the C11COONa (O) and SDS (0) concentration. Initial Al3+ concentrations are 1.09 and 1.14 mM, respectively. Solid lines are obtained by fitting eqs 13 and 14 to the experimental data.
Association with Al3+. From the above results we know that
the turbidity of a 1 mM Al3+ solution increases upon addition of sodium dodecanoate or dodecyl sulfate as a consequence of formation of solid white flocs. ICP-MS data (shown in Figure 6) not only confirms the decrease of (free) aluminum concentration in solution but also supports the stoichiometry of interaction of 3:1 between dodecanoate and Al3+, respectively. This contrasts with the case of SDS:Al3+, where the maximum interaction occurs at r = 4 as a consequence of formation of lamellar aggregates.57 With dodecanoate the aggregates are that formed as a direct consequence of charge neutralization, and the higher interaction efficiency is supported by the increased maximum aluminum removal: 98.7% for C11COONa compared with 90.5% for SDS. Another relevant observation occurs for r > 4. It is found55,71 that aluminum dodecyl sulfate is redissolved in excess dodecyl sulfate, even at concentrations below the cmc, with the concentration of aluminum, as seen by ICP-MS, being similar to the total concentration in solution. However, this does not happen with aggregates of aluminum tridodecanoate. Although Figure 6 shows an increase of Al3+ in solution, these data have been obtained without formation of a two-phase solution, and consequently, aliquots for ICP-MS include the floc from a whitish solution. In order to overcome this difficulty and to check the reliability of data, C11COONa:Al3+ has been prepared at 65 °C; although some decrease in the turbidity has been found, no phase separation has been observed, and the ICP-MS results were similar to those reported in the Figure 6. This clearly shows that, in contrast to aluminum dodecyl sulfate aggregates, those formed by dodecanoate do not redissolve, which can be directly related with their higher stability. An assessment on this can be made taking into account the behavior of the mixed systems at molar ratios below 34. Assuming that the main aluminum species present in solution is the hexaaquoaluminium(III), as seen by potentiometry, and the stoichiometry of interaction is 3:1 (surfactant:Al3+), as indicated by the different techniques, the interaction equilibria can be described by eqs 46, with the corresponding association constants K1, K2, and K3, respectively, given by K1 ¼
½MðSÞ2þ ½M3þ ½S
K2 ¼
½MðSÞþ 2 ½MðSÞ2þ ½S
ð11Þ
K3 ¼
½MðSÞ3 ½MðSÞþ 2 ½S
ð12Þ
where M3+ represents the trivalent metal and S the anionic surfactant. Taking into account the mass balances57 and eqs 1012 the concentration of Al3+, [Al3+], and anionic surfactant, [S], are dependent on the association constants through the following equations ½S total ¼ ðK1 ½M3þ þ 1Þ½S þ 2K2 K1 ½M3þ ½S 2 þ 3K3 K2 K1 ½M3þ ½S 3
ð13Þ
and ½M3þ total ¼ K1 ½M3þ ½S þ K2 K1 ½M3þ ½S 2 þ K3 K2 K1 ½M3þ ½S 3 þ ½M3þ
ð14Þ
The free surfactant concentration can be estimated through an analytical solution of the real solution of a third-degree equation using the CardinTartaglia formulas72 and using experimental data of the concentration of metal ion species in the supernatant. Finally, the association constants, K1, K2, and K3, are obtained from a least-squares fit (solid line in Figure 6) of eqs 13 and 14 to the ICP-MS experimental data (Figure 6). Despite the possibility of multicollinearity73 occurring as a consequence of strong dependence of terms in eqs 13 and 14, analysis of the root mean squared errors (RMSE) between the calculated and the observed values of free metal concentration provides a reliable indication of the goodness of fit (see Table S1 and Figures S4 and S5 in the Supporting Information) and, thus, on the association constants, Ki(=1,2,3). Table 1 summarizes the set of association constants for association between sodium dodecyl sulfate and sodium dodecanoate and Al3+. It is evident
ð10Þ 174
dx.doi.org/10.1021/la2034164 |Langmuir 2012, 28, 168–177
Langmuir
ARTICLE
Table 1. Association Constants for Interactions between Al3+ and Dodecyl Sulfate and Dodecanoate at 25 °C K1/M1 K2/M1 K3/M1 Al3+:DS 3+
Al :C11COO
RMSE/M
Ko/M3a
198.7
749.9
5800.0
2.0 105 8.6 108
3162.1
10032.0
9912.1
1.2 105 3.1 1011
a
Ko is the overall equilibrium constant (=K1 3 K2 3 K3) and RMSE =[∑i=1n](yi y^i )2/n]1/2
that the aggregation is a cooperative process in both cases, with the aluminum tridodecanote being 3 orders of magnitude more stable than the homologous dodecyl sulfate. This is in complete agreement with the theoretical calculations, which show stronger cation binding by carboxylate than sulfate groups, probably due to their greater Lewis basicity.41 Association with Cr3+. Figure 7 shows the effect of initial concentration of surfactant (sodium dodecyl sulfate and sodium dodecanoate) on the concentration of Cr3+ species in the supernatant in mixed surfactantCr3+ solutions at 25 °C. In general, the effect of surfactant on Cr3+ is similar to that occurring with Al3+, that is, addition of surfactant at Cr3+ solutions leads to a decrease in the concentration of metal ion in solutions as a consequence of formation of violet aggregates. It can also be observed that C11COONa is more efficient for Cr3+ removal than SDS, with a maximum metal removal efficiency (of 88.7% and 92.3%) at r = 4 and 3 for SDS and C11COONa, respectively. This behavior is similar to that seen with aluminum (as with Al3+ the stoichiometry with carboxylate is 1:3 and with SDS there is a minimum at r = 4, which we explain in terms of formation of lamellar aggregates). In addition, as with Al3+-containing systems, SDS has the capacity to redissolve the Cr3+DS aggregates and C11COONa does not. However, the mechanism proposed for Al3+ fails to describe the decrease in the concentration of metal ion (as measured by ICP-MS) versus r. A more detailed analysis (see dashed lines in the Figure 7 , inset) shows that the variation of metal ion concentration as a function of r has two different regimes for r < 34. From r = 0 to r around 1 the dependence of [Cr3+] = f(r) is small when compared with that at r > 1. That dependence is more persistent in the case of C11COONa, suggesting that a conjugated base of a strong acid easily promotes substitution by water exchange. Analysis of potentiometric results helps understand the interaction mechanism involving surfactants and Cr3+. In contrast to what happens with Al3+, in the presence of Cr3+ there is an increase of pH by adding dodecanoate, while with dodecyl sulfate the pH remains constant up to r = 1.1 and then start to increase up to pH ca. 5.0 at r = 18.0 (see Figure S2 in the Supporting Information). According to Baes and Mesmer the first hydrolysis of hexaaquo chromium(III) occurs at pQ between 3.66 and 4.01.37 The pH of C11COONaCr3+ solution is 4.39 for r = 1, well above the pQ, and corresponds to the first inflection point observed in Figure 7b. Consequently, it is reasonable to consider that at r values higher than 1.0 the main chromium species in the solution is Cr(H2O)5(OH)2+ and not Cr(H2O)63+; since the stoichiometry of interaction is 1:3 it is also likely that in both regimes (below and above r = 1) the metal ion interacts with 3 dodecanoates, and consequently, the first association equilibrium with Cr(H2O)5(OH)2+ occurs via exchange with OH. This also justifies the continuous increase of pH up to r around 3. We are aware of the complexity of mechanisms involving Cr3+ species, especially imposed by kinetic restraints.
Figure 7. Dependence of concentration of supernatant Cr3+ on the SDS (a) and C11COONa (b) concentration. Initial Cr3+ concentrations are 1.13 and 1.10 mM, respectively. Lines are obtained by fitting eqs 13 and 14: dashed lines and solid lines are considering two different mechanisms (see text).
However, it is known that substitution reactions of hydroxycomplexes of chromium(III) are several orders of magnitude faster than those of the corresponding aquo species.74 This will greatly favor incorporation of carboxylate within the coordination sphere. Taking into account our observations and the above assumption, the following mechanism is proposed: from r = 0 to r = 1 the main reactions are similar to those described in eqs 46, which correspond to the following association constants K1,1, K2, and K3, respectively; at r greater than 1, the initial reaction is considered to be ½CrðH2 OÞ5 ðOHÞ2þ þ Cn COO a ½CrðH2 OÞ5 ðCn COOÞ2þ þ OH
ð16Þ
or the equivalent two-step process ½CrðH2 OÞ5 ðOHÞ2þ þ Cn COO a ½CrðH2 OÞ4 ðOHÞðCn COOÞþ þ H2 O 175
dx.doi.org/10.1021/la2034164 |Langmuir 2012, 28, 168–177
Langmuir
ARTICLE
Table 2. Association Constants for Interactions between Cr3+ and Dodecyl Sulfate and Dodecanoate at 25 °C Cr3+:DS 3+
Cr :C11COO
K1,1/M1, r < 1
K1,2 / M1, r > 1
K2/M1
K3/M1
Ko,1/M3, r < 1
Ko,2/M3, r > 1
RMSE/M
1.1
2.5 103
9.9 103
2.5 104
2.6 108
6.2 1011
2.0 105
0.4
2.0 10
1.6 10
1.5 10
1.1 10
4.8 10
1.5 105
4
4
½CrðH2 OÞ4 ðOHÞðCn COOÞþ þ H2 O a ½CrðH2 OÞ5 ðCn COOÞ2þ þ OH
8
12
neutral, acid, and basic soaps, with differing degrees of hydration, may be formed. To help understand what is happening, we compared the behavior of the soaps of two metal ions, Al(III) and Cr(III), having very different hydration behavior. In both cases, phase separation is seen in the region of charge neutrality (3 [carboxylate] = [M3+]). However, using a variety of techniques, we have shown that with aluminum(III) there is a direct interaction with the carboxylate ion, leading to exchange of water within the primary hydration sphere, whereas with Cr(III) exchange of the water molecules is inhibited at room temperature due to the slow substitution rate. It can, however, occur on increasing temperatures. Studies of the effect of alkyl chain length show that interactions between the trivalent metal ions and octanoate are much weaker than those with the longer chain length carboxylates. This demonstrates the importance of hydrophobic interactions in the precipitation reactions. Finally, studies of the trivalent metal ions interacting with long chain alkyl carboxylates and sulfates show that with the sulfate, redissolution of flocs occurs at higher surfactant concentrations, while the same is not true with the carboxylates. This is likely to be related to the greater Lewis basicity of the carboxylates. In summary, precipitation of trivalent metal carboxylates is seen to involve a combination of contributions from cation hydration, interchain hydrophobic (largely van der Waal’s) interactions, and specific headgroup binding.
ð17Þ
described by an association constant K1,2; the following two steps will occur in the same way as with Al3+ (eqs 5 and 6). Consequently, two different set of association constants must be computed, with K2 and K3 being constants for both processes. A further question is should we neglect the interaction with hexaquochromium(III) when the main species present in solution is [Cr(H2O)5(OH)]2+. In practice, we do not neglect it since we assume that its contribution is constant and equal to sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 3þ total ½Cr3þ 3 ½Cr ½S ¼ ð16Þ ½Cr3þ Ko, 1 where Ko,1 is the overall constant obtained for the first regime. It is interesting to note that the effect of addition of SDS to Cr3+ solution is slightly different from that occurring with dodecanoate; at 0 < r e 1.1 the pH remains constant and equal to 3.58; at r values above 1.1 the pH starts to increase (being 3.6 at r = 1.6), suggesting a slight release of hydroxyl ions. Although the pH at this molar ratio is slightly smaller that pQ, ICP-MS results show that the mechanism should be similar to that occurring with the dodecanoate, especially to fit the experimental data at r below 1.5. Table 2 shows the association constants that best fit the experimental data shown in Figure 7 (solid lines) using eqs 1014 and eq 16 for r greater than ca. 1.0. For the two systems, K1,1 values are less than K2 and K3, reflecting the instability of the complex resulting from the first mechanistic step, which is closely related with a very low rate constant for the loss of one water molecule from the Cr3+ hydration sphere. The overall association constants for r < 1, Ko,1, are very similar (i.e., the same order of magnitude) for both surfactant systems, reflecting that other factors than the type of surfactant are controlling the aggregation. As suggested earlier, a major driving force may be the van der Waal’s interactions between the alkyl chains. However, for the second association regime, r > 1, the first association constant (K1,2) is much higher than for r < 1, with similar values to those obtained for K2 and K3, justifying the assumption that at this concentration zone the predominant Cr3+ species is different. It is also important to note that the first association constant (K1,2) and the overall association constant (Ko,2) for Cr3+:C11COO is an order of magnitude greater than for Cr3+:DS, which is also in agreement with the a strong binding and higher metal removal efficiency when C11COONa is used.
4
’ ASSOCIATED CONTENT
bS
Supporting Information. The dependence of pH upon sodium carboxylate concentration in water, the effect of sodium dodecylsulfate on the pH of a 1 mM Cr(III) solution, thermograms of solid aluminum(III) and chromium(III) dodecanoates and effects of initial guess values of association constants residual mean square error parameters on fitting of experimental data for Al(III) with SDS. This material is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION Corresponding Author
*Phone: +351 239 854 482. Fax: +351 239 827 703. E-mail:
[email protected].
’ ACKNOWLEDGMENT MEC and FEDER are thanked for financial support through the project MAT2008-06079/MAT and the University of Burgos for financial support of the ICP-MS measurements. R.F.P.P. thanks FCT for a Ph.D. grant (SFRH/BD/38696/2007). ’ REFERENCES
’ CONCLUSIONS Although the water-insoluble, long chain carboxylates of highvalent metal ions (metal soaps) have long been known and find a wide variety of technical applications, details of their precipitation from aqueous solution have remained unclear. Depending on the reaction conditions, a variety of products, including
(1) Holmberg, K.; J€onsson, B.; Kronberg, B.; Lindman, B. Surfactants and Polymers in Aqueous Solution, 2nd ed.; Wiley: Chichester, 2003; pp 193214. (2) Bou-Maroun, E.; Goetz-Grandmont, G. J.; Boos, A. Sep. Sci. Technol. 2006, 41, 2933–2946. (3) Paton-Morales, P.; Talens-Alesson, F. Langmuir 2001, 17, 6059– 6064. 176
dx.doi.org/10.1021/la2034164 |Langmuir 2012, 28, 168–177
Langmuir
ARTICLE
(44) Barthel, J.; Feuerlein, F.; Neueder, R.; Wachter, R. J. Solution Chem. 1980, 9, 209–219. (45) Vila, J.; Rilo, E.; Segade, L.; Cabeza, O.; Varela, L. M. Phys. Rev. E 2005, 71, 031201. (46) Chatterjee, A.; Moulik, S. P.; Sanyal, S. K.; Mishra, B. K.; Puri, P. M. J. Phys. Chem. B 2001, 105, 12823–12831. (47) Valente, A. J. M.; Burrows, H. D.; Pereira, R. F.; Ribeiro, A. C. F.; Pereira, J.; Lobo, V. M. M. Langmuir 2006, 22, 5625–5629. (48) Burrows, H. D.; Tapia, M. J.; Fonseca, S. M.; Valente, A. J. M.; Lobo, V. M. M.; Justino, L. L. G.; Qiu, S.; Pradhan, S.; Scherf, U.; Chattopadhyay, N.; Knaapila, M.; Garamus, V. M. ACS Appl. Mater. Interfaces 2009, 1, 864–874. (49) Turq, P.; Barthel, J.; Chemla, M. Transport, Relaxation and Kinetic Processes in Electrolyte Solutions; Springer-Verlag: Berlin, 1992. (50) Nilsson, M.; Cabaleiro-Lago, C.; Valente, A. J. M.; Soderman, O. Langmuir 2006, 22, 8663–8669. (51) Neves, A. C. S.; Valente, A. J. M.; Burrows, H. D.; Ribeiro, A. C. F.; Lobo, V. M. M. J. Colloid Interface Sci. 2007, 306, 166–174. (52) Pereira, R. F. P.; Cerqueira, D. A.; Valente, A. J. M.; Polishchuk, A. Y.; Burrows, H. D.; Lobo, V. M. M. J. Appl. Polym. Sci. 2009, 111, 1947–1953. (53) Bazuin, C. G.; Guillon, D.; Skoulios, A.; Amorim da Costa, M. A.; Burrows, H. D.; Geraldes, C. F. G. C.; Teixeira-Dias, J. J. C.; Blackmore, E.; Tiddy, G. J. T Liq. Cryst. 1988, 3, 1655–1670. (54) Burrows, H. D.; Chimamkpam, T. O.; Encarnac-~ao, T.; Fonseca, S. M.; Pereira, R. F. P.; Ramos, M. L.; Valente, A. J. M. J. Surf. Sci. Technol. 2010, 26, 197–212. (55) Pereira, R. F. P.; Valente, A. J. M.; Burrows, H. D.; Ramos, M. L.; Ribeiro, A. C. F.; Lobo, V. M. M. Acta Chim. Slov. 2009, 56, 45–52. (56) Pereira, R. F. P.; Valente, A. J. M.; Burrows, H. D. J. Mol. Liq. 2010, 156, 109–114. (57) Pereira, R. F. P.; Tapia, M. J.; Valente, A. J. M.; Evans, R. C.; Burrows, H. D.; Carvalho, R. A. J. Colloid Interface Sci. 2011, 354, 670–676. (58) Marques, E. F.; Burrows, H. D.; Miguel, M. D. J. Chem. Soc., Faraday Trans. 1998, 94, 1729–1736. (59) Albert, A.; Serjeant, E. P. The Determination of Ionization Constants: A Laboratory Manual; Champman Hall, Taylor & Francis Group: London, 1971. (60) Lucassen, J. J. Phys. Chem. 1966, 70, 1824–1830. (61) Kralchevsky, P. A.; Danov, K. D.; Pishmanova, C. I.; Kralchevska, S. D.; Christov, N. C.; Ananthapadmanabhan, K. P.; Lips, A. Langmuir 2007, 23, 3538–3553. (62) Robinson, R. A.; Stokes, R. H., Electrolyte solutions, 2nd ed.; Dover Publications, Inc.: New York, 2002. (63) Bertini, I.; Luchinat, C.; Rosi, M.; Sgamellotti, A.; Tarantelli, F. Inorg. Chem. 1990, 29, 1460–1463. (64) Katz, A. K.; Glusker, J. P.; Markham, G. D.; Bock, C. W. J. Phys. Chem. B 1998, 102, 6342–6350. (65) Rezabal, E.; Mercero, J. M.; Lopez, X.; Ugalde, J. M. J. Inorg. Biochem. 2007, 101, 1192–1200. (66) Rai, A. K.; Parashar, G. K. Thermochim. Acta 1979, 29, 175–179. (67) Seddon, A. B.; Wood, J. A. Thermochim. Acta 1986, 106, 341–354. (68) Cotton, F. A.; Wilkinson, G.; Murillo, C. A.; Bochmann, M. Advanced Inorganic Chemistry, 6th ed.; Wiley: New York, 1999; p744. (69) Rai., A. K.; Mehrotra, R. C. J. Indian Chem. Soc. 1963, 40, 359–364. (70) Cusanelli, A.; Frey, U.; Richens, D. T.; Merbach, A. E. J. Am. Chem. Soc. 1996, 118, 5265–5271. (71) Vasilescu, M.; Angelescu, D.; Caldararu, H.; Almgren, M.; Khan, A. Colloid Surf. A: Physicochem. Eng. Asp. 2004, 235, 57–64. (72) Valente, A. J. M.; Dinis, C. J. S.; Pereira, R. F. P.; Ribeiro, A. C. F.; Lobo, V. M. M. Port. Electrochim. Acta 2006, 24, 129–136. (73) Draper, N. R.; Smith, H. Applied Regression Analysis, 3rd ed.; Wiley-Interscience: New York, 1998. (74) Tobe, M. L. Acc. Chem. Res. 1970, 3, 377–385.
(4) Pilpel, N. Chem. Rev. 1963, 63, 221–234. (5) Sanderson, G. R. Br. Polym. J. 1981, 13, 71–75. (6) Ghesti, G. F.; de Macedo, J. L.; Parente, V. C. I.; Dias, J. A.; Dias, S. C. L. Appl. Catal. A: Gen. 2009, 355, 139–147. (7) Katre, Y. R.; Joshi, G. K.; Singh, A. K. Kinet. Catal. 2009, 50, 367–376. (8) Paleologos, E. K.; Vlessidis, A. G.; Karayannis, M. I.; Evmiridis, N. P. Anal. Chim. Acta 2003, 477, 223–231. (9) Hebrant, M. Coord. Chem. Rev. 2009, 253, 2186–2192. (10) Schreithofer, N.; Wiese, J.; McFadzean, B.; Harris, P.; Heiskanen, K.; O’Connor, C. Int. J. Miner. Process. 2011, 100, 33–40. (11) Al-Otoom, A.; Allawzi, M.; Al-Omari, N.; Al-Hsienat, E. Energy 2010, 35, 4217–4225. (12) Liu, G. Z.; Conn, C. E.; Drummond, C. J. J. Phys. Chem. B 2009, 113, 15949–15959. (13) Athens, G. L.; Shayib, R. M.; Chmelka, B. F. Curr. Opin. Colloid Interface Sci. 2009, 14, 281–292. (14) Sanson, N.; Bouyer, F.; Gerardin, C.; In, M. Phys. Chem. Chem. Phys. 2004, 6, 1463–1466. (15) Wu, Z. H.; Zhang, J.; Benfield, R. E.; Ding, Y. F.; Grandjean, D.; Zhang, Z. L.; Ju, X. J. Phys. Chem. B 2002, 106, 4569–4577. (16) Corkery, R. W. Curr. Opin. Colloid Interface Sci. 2008, 13, 288–302. (17) Casado, F. J. M.; Arenas, A. S.; Perez, M. V. G.; Yelamos, M. I. R.; de Andres, S. L.; Cheda, J. A. R. J. Chem. Thermodyn. 2007, 39, 455–461. (18) Young, S. L.; Matijevic, E. J. Colloid Interface Sci. 1977, 61, 287–301. (19) Akanni, M. S.; Okoh, E. K.; Burrows, H. D.; Ellis, H. A. Thermochim. Acta 1992, 208, 1–41. (20) McBain, J. W.; McClatchie, W. L. J. Am. Chem. Soc. 1932, 54, 3266–3268. (21) Ostwald, W.; Riedel, R. Kolloid-Z 1935, 70, 67–74. (22) Lawrence, A. S. C. Trans. Faraday Soc. 1938, 34, 0660–0677. (23) Gray, V. R.; Alexander, A. E. J. Phys. Colloid Chem. 1949, 53, 23–39. (24) Lawrence, A. S. C. J. Inst. Petrol. 1945, 31, 303–314. (25) Mysels, K. J. Ind. Eng. Chem. 1949, 41, 1435–1438. (26) McRoberts, T. S.; Schulman, J. H. Nature 1948, 162, 101–102. (27) Wang, X. R.; Rackaitis, M. J. Colloid Interface Sci. 2009, 331, 335–342. (28) Narayanan, R.; Laine, R. M. J. Mater. Chem. 2000, 10, 2097– 2104. (29) Wood, J. A.; Seddon, A. B. Thermochim. Acta 1981, 45, 365–368. (30) Emsley, J. The Elements, 2nd ed.; Clarendon Press: Oxford, 1991. (31) Neilson, G. W.; Enderby, J. E. Adv. Inorg. Chem. 1989, 34, 195–218. (32) Ohtaki, H.; Radnai, T. Chem. Rev. 1993, 93, 1157–1204. (33) Lindqvist-Reis, P.; Munoz-Paez, A.; Diaz-Moreno, S.; Pattanaik, S.; Persson, I.; Sandstrom, M. Inorg. Chem. 1998, 37, 6675–6683. (34) Vinogradov, E. V.; Smirnov, P. R.; Trostin, V. N. Russ. Chem. Bull. 2003, 52, 1253–1271. (35) Marcus, Y. J. Chem. Soc., Faraday Trans. 1 1987, 83, 339–349. (36) Marcus, Y. J. Chem. Soc., Faraday Trans. 1991, 87, 2995–2999. (37) Baes, C. F.; Mesmer, R. E. The Hydrolysis of Cations; Wiley: New York, 1976. (38) Helm, L.; Merbach, A. E. Chem. Rev. 2005, 105, 1923–1959. (39) Serjeant, E. P.; Dempsey, B. Ionization Constants of Organic Acids in Aqueous Solution; Pergamon: Oxford, 1979. (40) Huibers, P. D. T. Langmuir 1999, 15, 7546–7550. (41) Remko, M.; Van Duijnen, P. T.; von der Lieth, C. W. J. Mol. Struct. (THEOCHEM) 2007, 814, 119–125. (42) Mukerjee, P.; Mysels, K. J. Critical Micelle Concentrations of Aqueous Surfactant Systems; NBS: Wasington, D.C., 1971; p 222. (43) Ribeiro, A. C. F.; Valente, A. J. M.; Lobo, V. M. M.; Azevedo, E. F. G.; Amado, A. M.; da Costa, A. M. A.; Ramos, M. L.; Burrows, H. D. J. Mol. Struct. 2004, 703, 93–101. 177
dx.doi.org/10.1021/la2034164 |Langmuir 2012, 28, 168–177
