Base


Effective and Reversible Switching of Emulsions by an Acid/Base...

3 downloads 186 Views 4MB Size

Article pubs.acs.org/Langmuir

Effective and Reversible Switching of Emulsions by an Acid/BaseMediated Redox Reaction Yuandi Zhang,† Hui Chen,† Xuefeng Liu,*,† Yongmin Zhang,† Yun Fang,† and Zhirong Qin‡ †

School of Chemical & Materials Engineering, Key Laboratory of Food Colloids and Biotechnology Ministry of Education, Jiangnan University, Wuxi 214122, PR China ‡ Zhejiang Zanyu Technology Co. Ltd., Hangzhou 310009, PR China S Supporting Information *

ABSTRACT: To develop a fast, effective, and reversible strategy for phase separation and re-emulsification of the surfactant-based emulsions, a strategy for using acid/basemediated redox reactions was established to switch the emulsions formed from a redox-responsive anionic surfactant of potassium dodecyl seleninate (C12SeO2K). Upon acidification, C12SeO2K was reduced by KI to give didodecyl diselenide (C12Se)2, a state of almost no surface or interfacial activity; upon basification, (C12Se)2 was oxidized by I2 to give C12SeO2K again. The fractional conversion of C12SeO2K in the reversible switching processes was close to 100%. Consequently, an unusually large change in interfacial tension (ΔIFT) as high as ∼27.1 mN m−1 was obtained at a wider concentration range starting from the critical micelle concentration of C12SeO2K; the highest IFT at the oil−water interface was obtained after an almost complete switch-off, giving an oil−aqueous solution interface very similar to that without any emulsifiers, which leads to the effective and fast phase separation of the C12SeO2K-based switchable emulsions.



INTRODUCTION Surfactant-stabilized emulsions have important roles in cleaning, manufacturing, oil recovery, and other industrial processes. However, in many practical applications, such as in “emulsion-liquid-membrane extraction”,1,2 the emulsion is useful only during one specific step of the process, after which it is broken into separate phases.3,4 The fact that the “broken” emulsions can be recovered and reused afterward1−4 is of equal importance. Thus, efficient methods for the reversible switching of emulsions, that is, phase separation (switch-off) and re-emulsification (switch-on), are desirable. It is believed that switchable surfactants3−7 have the potential to solve this problem. The switchable surfactants can undergo reversible conversions between active and inactive forms under particular stimuli, including light, 5−7 pH, 8−12 temperature, 8,13 CO2,3,4,14−20 host−guest interactions,21 magnetism,22,23 and redox reactions.24−32 Generally, emulsions emulsified by switchable surfactants are always switchable under the stimuli.3,4,6,7 In most cases, the reversibility of these switchable emulsions has been well-documented. However, from the view point of practical applications, the effectiveness and efficiency of switching processes should not be ignored.4 “Effectiveness” means the maximum change in the properties of switchable emulsions that can be reversibly obtained; “efficiency” means the shortest time-cost in the switching process and the widest range of emulsifier concentrations that give the largest © XXXX American Chemical Society

effectiveness. Reasonably, the larger the effectiveness and efficiency, the more effective the switching of the emulsions. The reduction in the interfacial tension (IFT) and the mechanical, steric, and/or electrical barriers formed by the surfactant at the oil−water interface is thermodynamically and kinetically key factors in the surfactant-stabilized emulsions. It has been reported that the change in IFT (ΔIFT) and then the demulsification of switchable emulsions occur upon stimulation.6,7 Thus, from the view point of thermodynamics, the ΔIFT could be used to partly characterize the effectiveness of a stimulus or as a tuning strategy for the switchable emulsions. To date, the largest ΔIFT of ∼20 mN m−1 was obtained by a light-responsive azo-based surfactant (C4AzoTAB) with the help of sodium dodecyl sulfate (SDS).7 It is noteworthy that the aforementioned ΔIFT could be achieved only in a limited concentration range near the critical micelle concentration (CMC) of C4AzoTAB.7 Thus, to get a larger ΔIFT in a wider concentration range of surfactants, it is necessary to develop more effective strategies for tuning the IFT at the oil−water interface in switchable emulsions. On the other hand, under the premise of the largest ΔIFT, the time needed to complete the switching of switchable emulsions is of equal significance. The time-cost of Received: October 5, 2016 Revised: November 24, 2016 Published: November 29, 2016 A

DOI: 10.1021/acs.langmuir.6b03645 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

the aforementioned emulsion-switching process can be reversibly and effectively achieved at a wider concentration range of the surfactant.

demulsification using a photoswitchable azo-based Gemini surfactant was ∼6 h, and ΔIFT was ∼0.6 mN m−1.6 With the help of SDS, the time required for the complete phase separation of the SDS−C4AzoTAB (1:9, mol/mol)-emulsified emulsion was reduced to 1 h at a surfactant concentration, c, of 7 mmol L−1 and to 1.5 h at c = 10 mmol L−1; the ΔIFT values were ∼16.6 and ∼10.2 mN m−1.7 The time-cost of 1.5 h at a surfactant concentration of 10 mmol L−1 was drastically reduced to 3.5 min when the same volume of emulsions was thinned using a special microreactor.7 Thus, to obtain larger ΔIFT across a wider range of surfactant concentrations, more effective and faster strategies for switching the emulsions are urgently desired. In this study, we used the acid/base-mediated redox chemistry of diselenides33−36 to reversibly and effectively switch emulsions emulsified by the redox-responsive surfactant potassium dodecyl seleninate (C12SeO2K, Scheme 1). To



EXPERIMENTAL SECTION

Materials. C12SeO2K was obtained by the neutralization reaction of dodecyl seleninic acid (C12SeO2H) with KOH. The preparation of C12SeO2H was according to the reported HNO3 method (see Supporting Information).34 Surface and IFT Measurements. The surface tension (γ, mN m−1) of the aqueous samples was measured using the drop volume method.37 The IFT (mN m−1) of a petroleum ether (PE)−aqueous surfactant solution interface was measured using the spin drop method with a spin rate of 5000−6000 rpm. All measurements were carried out and repeated at least three times at 25 °C; the error ranges of γ and IFT were less than ±0.1 mN m−1. Emulsion Preparation. To produce the emulsion, the mixture of PE and aqueous surfactant solution was homogenized for 0.5 min at 21 000 rpm at room temperature. The volume ratio of PE and the aqueous surfactant solution was 1:1, and the molar ratio of KI to C12SeO2K was 3:1. Emulsion Characterization. The type of emulsions was determined by the drop test.7 Emulsion droplets were visualized using a VHX-1000 microscope system (Keyence, USA). The relative stability of the emulsion was measured as the time needed to separate 1 mL of H2O from 6 mL of the emulsion (t1mL, s) at 25 °C. Emulsion Switching. To switch-off the emulsion, concentrated HCl was added to the emulsion to a final pH of ∼1. The mixture was then homogenized for 5 s. To switch-on the emulsion, KOH was added to the emulsion to a final pH of ∼13. The mixture was then homogenized for 10 s. To examine the possible aggregates or droplets upon phase separation, samples of the aqueous and oil phases were measured using dynamic light scattering (DLS) on an ALV/DLS/SLS5022F spectrometer (HOSIC, Germany) with a 90° back-scattering angle and a He−Ne laser (λ = 633 nm). The molecular evidence of the transformation of C12SeO2K was obtained by 1H, 77Se NMR, and electrospray ionization-mass spectrometry (ESI-MS) using an Avance III 400 MHz nuclear magnetic resonance (NMR) spectrometer (Bruker), an Agilent DD2 600 NMR spectrometer (Agilent), and a Maldi Synapt Q-TOF mass spectrometer (Waters), respectively. Tetramethylsilane and dimethyl selenide were used as the references in the 1H and 77Se NMR measurements, respectively. The fractional conversion of C12SeO2K was calculated by the concentration variation in C12SeO2K, which was monitored using the methylene blue method38 at room temperature. Measurements of the contact angle and foam ability can be seen in the Supporting Information.

Scheme 1. Reversible Switching between C12SeO2K and (C12Se)2

switch-off the emulsions, C12SeO2K can be almost completely reduced by acidic iodide to give didodecyl diselenide [(C12Se)2],33 a hydrophobic compound with almost no interfacial activity. Consequently, the IFT at the initial oil− water interface will increase with that of an oil−water interface without any surfactants, and the mechanical, steric, and electrical barriers formed by C12SeO2K at the oil−water interface will vanish, which results in complete phase separation. To switch-on the emulsion, (C12Se)2 can be oxidized by basic iodine to give C12SeO2K again.33 Therefore, the IFT at the oil−water interface can be lowered and restored to that of the initial emulsion. Thus, the reversible switching of emulsions can be enabled by an acid/base-mediated redox reaction strategy. Compared with the reported switchable surfactant-based emulsions,6,7,21 an additional advantage is that



RESULTS AND DISCUSSION Maximum Differences in Surface Activities. As shown in Scheme 1, C12SeO2K and (C12Se)2 are the boundary states of the switching process. To characterize the surface activities of

Figure 1. Plots of (A) γ−log c and (B) IFT−log c for C12SeO2K and (C12Se)2. The insets are snapshots of (A) foam and (B) emulsion of C12SeO2K (A1 and B1) and (C12Se)2 (A2 and B2), where the concentration of C12SeO2K and (C12Se)2 was fixed at 5 mmol L−1. The volume ratio of PE to aqueous solution was 3:3 mL. B

DOI: 10.1021/acs.langmuir.6b03645 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 2. Snapshots of (A) oil droplets and (B) their size distribution in an emulsion of PE and 7 mmol L−1 of the aqueous C12SeO2K solution (v/v = 1:1) with 21 mmol L−1 of KI. (C) Dependence of the relative stability of the emulsions (t1mL, s) on the surfactant concentration. The emulsion was diluted by (D) 0.1 mol L−1 KOH and (E) PE. The emulsions in (D) and (E) were colored by trace amounts of methylene blue. The scale bar in A is 25 μm.

Figure 3. Snapshots of (A) the emulsion of PE and 7 mmol L−1 of aqueous C12SeO2K solution (v/v = 1:1) with 21 mmol L−1 of KI, (B) phase separation after the addition of HCl with a final pH of ∼1, and (C) size distribution of the droplets in the recovered emulsion.

C12SeO2K and (C12Se)2, the curves of γ−log c (c, concentration of surfactant, mol L−1) and IFT−log c were investigated (Figure 1), respectively. During the procedure of surface and IFT measurements, a sufficient aging time was necessary for the surface or the interface to reach an equilibrium state (Figure S1). C12SeO2H is a weak acid with pKa = 6.98 ± 0.06 (Figure S2); a value that is larger than that of areneseleninic acids (pKa ≈ 5).39,40 Thus, to avoid the hydrolysis of C12SeO2K, a 0.1 mol L−1 of KOH aqueous solution was used as a solvent unless otherwise stated. The γ and IFT values decreased gradually with increasing C12SeO2K concentration, with sharp break points (CMC) at ∼4.8 × 10−3 and ∼5.0 × 10−3 mol L−1, respectively, whereas the γ and IFT values of (C12Se)2 remained unchanged, regardless of the concentration of (C12Se)2. The values were close to those of the solvent (∼72.3 mN m−1) and the PE− aqueous solution interface without any surfactants (∼35.8 mN m−1). Furthermore, both the foam and emulsification abilities of C12SeO2K aqueous solution (A1 and B1, Figure 1) were remarkably better than those of (C12Se)2 (A2 and B2, Figure 1). The above-mentioned results of γ, IFT, foam, and

emulsification confirmed that C12SeO2K was a surfactant, whereas (C12Se)2 had almost no surface or interfacial activity at all. Hence, the maximum differences in γ and IFT between C12SeO2K and (C12Se)2 were as high as ∼32.5 and ∼27.3 mN m−1, respectively. Reversible and Fast Switching of Emulsions. Emulsions consisting of mixtures of PE and 7 mmol L−1 of aqueous C12SeO2K solution (v/v = 1:1) were observed using a light microscope (Figure 2A). Figure 2B shows the number distribution of the diameters of the oil droplets (PE phase) dispersed in an aqueous solution. The measurement was taken by observing 500 oil droplets through the light microscope. The number-averaged diameter was 7.8 ± 4.3 μm. In the concentration range of 1.0 × 10−5 to 1.0 × 10−2 mol L−1, the relative stability of the emulsions increased with an increase in the surfactant concentration (Figure 2C); the emulsions with t1mL ≥ 1 h are considered arbitrarily as stable emulsions in this study. To determine the type of the emulsions, 2 mL of the emulsion was colored by methylene blue and diluted by PE and aqueous KOH solution, respectively. The emulsion was easily diluted by the aqueous KOH solution (Figure 2D), but it could C

DOI: 10.1021/acs.langmuir.6b03645 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 4. Reversible responsiveness of (A) IFT and (B) γ for C12SeO2K under the stimuli of HCl−KI and KOH−I2; the initial sample comprised a C12SeO2K (10 mmol L−1) aqueous solution in the presence of KI (30 mmol L−1); the volume ratio of PE to the aqueous surfactant solution is 3:3 mL. The insets are snapshots of the (A) emulsification and (B) foam of C12SeO2K before (A1 and B1) and after (A2 and B2) switching.

Figure 5. Evidence at the molecular level for the redox-triggered switching between C12SeO2K and (C12Se)2: (A) 1H NMR and (B) 77Se NMR spectra of C12SeO2H and (C12Se)2 using CDCl3 as the solvent.

respectively, are quite extreme. As demonstrated in Figure S4, t1mL ≈ 111 min when the pH decreases from 13 to 8 and less than 8 min when the pH increases from 1 to 7, which indicates that the pH values of 8 and 7 are sufficient for the ON and OFF states of the emulsions, respectively. Thus, the emulsion could be cycled at least 20 times when the pH varied between 8 and 7 (Figure S5). Unusually Large ΔIFT in the Process of Switching. To evaluate the ΔIFT in the emulsion switching, the variation in IFT (Figure 4A) owing to the addition of HCl and/or KOH was determined. The initial IFT at the interface of a PE−10 mmol L−1 of aqueous C12SeO2K solution (with KI 30 mmol L−1) was ∼8.7 mN m−1. After being acidized by HCl to a final pH of ∼1, the IFT changed remarkably to ∼35.8 mN m−1, a value that is close to that of the PE−aqueous solution without any emulsifiers (Figure 1B), and the IFT was completely restored to the initial value (∼8.7 mN m−1) after alkalization by KOH to a final pH of ∼13. Thus, an unusually large ΔIFT = ∼27.1 mN m−1 at the PE−aqueous solution interface was reversibly obtained. Such an unusually large ΔIFT at the PE− aqueous solution interface is comparable to that of the pHswitchable, structurally related traditional surfactant potassium laurate (∼29.9 mN m−1). In addition to IFT, γ (Figure 4B) and contact angle (θ/°, see Figure S6) were also reversibly tuned under the redox-stimuli of HCl−KI and KOH−I2. Unusually large Δγ and Δθ values of ∼32.2 mN m−1 and 29.2 ± 1.1° were obtained, respectively. Obviously, ΔIFT for the redox-responsive C12SeO2K was obviously larger than that for C4AzoTAB-SDS (ΔIFT = ∼20 mN m−1);7 Δγ for the redox-responsive C12SeO2K was larger than that for a ferrocenyl surfactant ((11-ferrocenylundecyl)trimethylammonium bromide (FTMA), Δγ = ∼23 mN

not be diluted by PE (Figure 2E). Therefore, the emulsions were O/W-type. Figure 3 shows the reversible switching of the emulsion under redox-stimuli of HCl−KI and KOH−I2. On a microscopic scale, the oil droplets disappeared within 0.5 min after HCl addition (see Movie S1). On a macroscopic scale, 6 mL of emulsions in the C12SeO2K concentration range of 1.0 × 10−5 to 1.0 × 10−2 mol L−1 caused complete phase separation within 0.5 min after HCl addition with agitation (Figure 3B), and a wine-red color from I233 was obtained. The DLS results (Figure S3) indicated that there were no detectable aggregates or droplets in the separated oil and aqueous phases. The two separated oil and aqueous phases were restored to an emulsion within 0.5 min after KOH addition with agitation, and the number-averaged diameter of the droplets (6.8 ± 4.2 μm, Figure 3C) in the recovered emulsion was close to the initial number-averaged diameter (7.8 ± 4.3 μm, Figure 2B). In addition to KOH, the aforementioned emulsion was recovered by the addition of NaOH, NH3·H2O, and K2CO3 (see Movies S2−S4). Obviously, the rate of complete phase separation for the redox-responsive emulsions is greater than that of photoresponsive emulsions with comparable concentrations of switchable surfactants and a volume scale of emulsions.6,7,41 Furthermore, when the concentration of C12SeO2K and the volume scale of the emulsion increased to 30 times that of the CMC (Figure 1B) and 100 mL, respectively, the time-cost for complete phase separation and re-emulsification was less than 1.0 min. Hence, the strategy of using an acid/base-mediated redox reaction to switch C12SeO2K-based emulsions across a wider range of surfactant concentrations and the volume scale is effective, fast, and reversible. It is noteworthy that the pH values of 13 and 1 for the ON and OFF states of the emulsion, D

DOI: 10.1021/acs.langmuir.6b03645 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir m−1)29,42 and a Gemini surfactant containing a disulfide bond in the spacer (sodium didecamino cystine (SDDC), Δγ = ∼25 mN m−1).28 Therefore, it is thought that the larger ΔIFT accounts for the effective and fast reversible switching of the C12SeO2K-based emulsions. The aforementioned remarkable changes in the surface activities of C12SeO2K can be ascribed to its unusual switching process (Scheme 1), which differs from those of other switchable surfactants such as SDDC,28 FTMA,29 and C4AzoTAB.7 These switchable surfactants shuttle back and forth between two types of surfactants, one with a relatively high surface activity (active form) and the other with a relatively low surface activity (inactive form). However, C12SeO2K switches between a surfactant and (C12Se)2, a drastically different state with almost no surface activity (Figure 1). On the other hand, the surface activity may also be attributed to the completely reversible redox reaction between C12SeO2K and iodide mediated by the acid/base.33 To confirm the above two hypotheses, evidence of the molecule structural transition of C12SeO2K in the switching process was monitored using 1H NMR (Figure 5A), 77Se NMR (Figure 5B), Fourier transform infrared (FT-IR) (Figure S7), and ESI-MS (Figure S8). Both C12SeO2H and (C12Se)2 are soluble in CDCl3, and thus, C12SeO2K was changed into C12SeO2H in monitoring the molecular structure transformation. After the switch-off, the 1H NMR signals of two methylene groups (CH and DH, Figure 5A), which were close to selenium in C12SeO2H, shifted to a high field (Δδ = 0.1−0.15 ppm)43−45 because of the reduction in the initial C12SeO2H, which is denoted as C12SeO2H(HNO3), by HCl−KI to give (C12Se)2;33 after the switch-on, (C12Se)2 was oxidized by KOH−I2 to give C12SeO2H, which is denoted as C12SeO2H(I2).33 No obvious difference between C12SeO2H (I2) and C12SeO2H (HNO3) was observed in the 1H NMR spectra (Figure 5A). However, the 77 Se NMR signal of C12SeO2H shifted drastically from 1214.7 to 306 ppm43−45 (Figure 5B) after the switch-off, which further confirmed the change from C12SeO2H into (C12Se)2.33 The results of FT-IR and ESI-MS (Figures S7 and S8) also confirmed the mutual transformation between C12SeO2H and (C12Se)2 under the redox-stimuli of HCl−KI and KOH−I2. Consequently, the rufous of I2 will appear and disappear alternatively during redox-responsive switching for C12SeO2Kbased emulsions, as exhibited in Figure 3 and Movies S2−S4. Evidently, the unusual switching between a surfactant of C12SeO2K and the drastically different state of (C12Se)2 with almost no surface activity accounts for the remarkable ΔIFT, Δγ, and Δθ values of C12SeO2K under the redox-stimuli. To obtain deep insights into the quantitative relationship between the redox reaction and the efficient switching for C12SeO2K-based emulsions, the fractional conversion of C12SeO2K was determined using the methylene blue method38 in the switching process (Figure 6). The calculation of the fractional conversion can be seen in Figure S9. As exhibited in Figure 6, the fractional conversion of C12SeO2K was 99.5 ± 0.4%, which was reversibly achieved during the redoxresponsive switching, when either (C12Se)2 was oxidized to give C12SeO2K or C12SeO2K was reduced to give (C12Se)2.33 The almost complete redox reactions of (C12Se)2−I2 and C12SeO2K−KI were enabled reversibly under the alternating stimuli of KOH and HCl, which accounts for not only the reversible responsiveness of γ, IFT, and the unusually large Δγ and ΔIFT for C12SeO2K but also an oil−aqueous solution interface with almost no emulsifiers after switch-off. Therefore,

Figure 6. Fractional conversion and its reversibility for C12SeO2K in the switching process starting from (C12Se)2 under the redox-stimuli of KOH−I2 and HCl−KI, respectively, where the initial amounts of (C12Se)2 and I2 are 1.76 × 10−4 and 5.28 × 10−4 mol, respectively, in 100 mL of H2O.

the fast and effective switching of the emulsions between phase separation and re-emulsification could be achieved across a wider range of switchable surfactant concentrations and the volume scale of emulsion. Tuning the IFT Step by Step. We wondered whether the IFT at the PE−aqueous C12SeO2K solution interface could be finely tuned by controlling the addition of the acid (or base) step by step. For this purpose, the dependences of IFT on the amount of HCl or KOH added were measured starting from the PE−aqueous C12SeO2K solution interface (Figure 7A) and PE-dissolved (C12Se)2−aqueous solution interface (Figure 7B), respectively. As shown in Figure 7A, the IFT at the initial interface was ∼8.7 mN m−1, and it increased gradually to a maximum of ∼35.6 mN m−1 with the addition of HCl (the final pH was ∼1) step by step. With the maximum amount of HCl, the largest IFT was ∼35.6 mN m−1, which is close to that of the interface of the PE−aqueous solution without any emulsifiers, as shown in Figure 1B. Therefore, the process of switch-off was completed, and the interface of the PE−10 mmol L−1 of C12SeO2K aqueous solution changed into that of the 5 mmol L−1 of (C12Se)2 PE solution−aqueous solution (∼35.5 mN m−1, Figure 7B) to give an unusually large ΔIFT of ∼26.9 mN m−1; upon the gradual addition of KOH (the final pH was ∼13), the IFT decreased step by step from ∼35.6 mN m−1 to the initial value of ∼8.7 mN m−1. The opposite behavior was observed when the switching process started from the PEdissolved (C12Se)2−aqueous solution interface (Figure 7B). Thus, the IFT at the PE−aqueous C12SeO2K solution interface could be finely and reversibly tuned step by step. In addition to the IFT, the γ (Figure S10A), foaming ability (Figure S10B,C), and wetting ability (Figure S10D) also could be tuned step by step under the acid/base-mediated redox reaction strategy. Possible Mechanism of the Emulsion Switching. We discuss here why the O/W-type C12SeO2K-based emulsions effectively and reversibly switched between phase separation and re-emulsification under the redox-stimuli of KOH−I2 and HCl−KI, respectively. Considering the results of ΔIFT, the change in the molecular structures, and the fractional conversion of C12SeO2K in the redox reaction, we suggest the following mechanism for the redox-induced switching between phase separation and re-emulsification for O/W-type C12SeO2K-based emulsions. As illustrated in Scheme 2A, C12SeO2K molecules are adsorbed on the entire interface of the PE−aqueous surfactant solution in the O/W-type emulsions. Upon the addition of HCl, C12SeO2K is reduced by KI to give (C12Se)2. As shown in Figure 1, (C12Se)2 has E

DOI: 10.1021/acs.langmuir.6b03645 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 7. Dependence of IFT at the PE−aqueous solution interface on the amount of HCl or KOH starting from the interfaces of (A) PE−10 mmol L−1 of C12SeO2K−30 mmol L−1 of KI aqueous solution and (B) PE−5 mmol L−1 of (C12Se)2−15 mmol L−1 of I2 aqueous solution, respectively.

Scheme 2. Possible Mechanism of Reversible C12SeO2K-Based Emulsion Switching between Phase Separation and Emulsification Enabled by the Acid/Base-Mediated Redox Reaction

by C12SeO2K at the oil−water interface will vanish after switchoff, and then, the PE drops readily coalesce, decreasing the interfacial area in the emulsion. This is followed by a decrease in the free energy until the PE, and water phases are fully separated. Light microscopy observations confirmed the abovementioned mechanism (see Movie S1). Upon basification, (C12Se)2 was oxidized by I2 to give the C12SeO2K surfactant again, the IFT at the interface of the PE−water phase was decreased, and the mechanical, steric, and electrical barriers formed by C12SeO2K at the oil−water interface were recovered (Figures 4A and 7), which lead to re-emulsification (see Movies S2−S4). More specifically, the redox-responsive C12SeO2K has unusual switching between a surfactant and a state of almost no surface or interfacial activity, as seen in Scheme 1 and Figure 1, and the switching process showed nearly 100% fractional conversion for C12SeO2K (Figure 6). Thus, the IFT at the

almost no surface or interfacial activity, which leads to a remarkable increase in IFT (Figure 4A) at the PE−aqueous surfactant solution interface. The exposure of the PE−water interface causes the coalescence of PE droplets in the O/Wtype emulsion (Scheme 2B), reducing the interfacial Gibbs free energy, followed by the complete phase separation to give a state of minimum interfacial area and Gibbs free energy (Scheme 2C). For the emulsion of PE and 7 mmol L−1 of aqueous C12SeO2K solution, the number-averaged diameter of PE droplets was ∼7.8 μm (Figure 2B). There were therefore ∼2.0 × 1010 PE droplets in 10 mL of the emulsion, and thus, the emulsion had an interfacial area of ∼3.8 m2. The unusual increase in the IFT by ∼27 mN m−1 as a result of the redox reaction (Figures 4A and 7) corresponds to an increase in the interfacial Gibbs free energy in the emulsion of ∼102.6 mJ; meanwhile, the mechanical, steric, and electrical barriers formed F

DOI: 10.1021/acs.langmuir.6b03645 Langmuir XXXX, XXX, XXX−XXX

Langmuir



interfaces of the initial emulsions could be increased to a maximum upon completion of the switch-off process and the achievement of the oil−water interface without any emulsifiers. Therefore, the complete phase separation is a reasonable result for the oil−water system in the absence of surfactants. Hence, both the unusual switching between a surfactant and a state of almost no surface or interfacial activity and the nearly 100% fractional conversion for C12SeO2K during the acid/basemediated redox switching process account for the effective and fast phase separation and re-emulsification of the C12SeO2Kbased emulsions.

AUTHOR INFORMATION

Corresponding Author

*E-mail: xfl[email protected]. ORCID

Xuefeng Liu: 0000-0002-9506-7798 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was financially supported by the National Natural Science Foundation of China (grant no. 21503094, 21673103), the Qinglan Project of Jiangsu Province, and the Zhejiang Zanyu Technology Co. Ltd., P. R. China. We are also grateful to Prof. Huping Zhu (Shanghai Institute of Organic Chemistry, Chinese Academy of Science) and Prof. Guanjun Tao (State Key Laboratory of Food Science & Technology, Jiangnan University) for the 77Se NMR and ESI-MS measurements and discussions, respectively.



CONCLUSIONS In summary, a fast, effective, and reversible strategy for switching O/W-type C12SeO2K-based emulsions from phase separation to re-emulsification and vice versa has been achieved by an acid/base-mediated redox reaction. One of the advantages of the present acid/base-mediated redox reaction strategy is the unusual and almost 100% successful switching of C12SeO2K from a surfactant to a state of almost no surface or interfacial activity [(C12Se)2]. Consequently, unusually large Δγ, ΔIFT, and Δθ were obtained over a wider concentration range starting from the CMC of C12SeO2K. The other advantage is that an oil−aqueous solution interface nearly without any emulsifiers can be reversibly obtained after switchoff. Both the unusual switching between a surfactant and a state of almost no surface or interfacial activity and the nearly 100% fractional conversion for C12SeO2K during the acid/basemediated redox switching process account for the effective and fast phase separation and re-emulsification of the C12SeO2Kbased emulsions. For the present strategy, if the addition of chemicals, the inorganic byproducts, the probable toxicity, and the economic cost of C12SeO2K or C12SeO2H are not problematic, we think that this effective, rapid, and wellcontrolled tuning strategy for redox-responsive anionic surfactant-based emulsions under mild conditions could be used in fields where effective, rapid, and well-controlled demulsification and re-emulsification strategies are needed.



Article



REFERENCES

(1) Gang, L.; QiongHua, L.; PanSheng, L. Break-down of liquid membrane emulsion under high electric field. J. Membr. Sci. 1997, 128, 1−6. (2) Sastre, A. M.; Kumar, A.; Shukla, J. P.; Singh, R. K. Improved techniques in liquid membrane separations: an overview. Sep. Purif. Rev. 1998, 27, 213−298. (3) Liu, Y.; Jessop, P. G.; Cunningham, M.; Eckert, C. A.; Liotta, C. L. Switchable surfactants. Science 2006, 313, 958−960. (4) Jiang, J.; Zhu, Y.; Cui, Z.; Binks, B. P. Switchable Pickering emulsions stabilized by silica nanoparticles hydrophobized in situ with a switchable surfactant. Angew. Chem. 2013, 125, 12599−12602. (5) Brown, P.; Butts, C. P.; Eastoe, J. Stimuli-responsive surfactants. Soft Matter 2013, 9, 2365−2374. (6) Takahashi, Y.; Fukuyasu, K.; Horiuchi, T.; Kondo, Y.; Stroeve, P. Photoinduced demulsification of emulsions using a photoresponsive gemini surfactant. Langmuir 2014, 30, 41−47. (7) Takahashi, Y.; Koizumi, N.; Kondo, Y. Active demulsification of photoresponsive emulsions using cationic−Anionic surfactant mixtures. Langmuir 2016, 32, 683−688. (8) Jia, K.; Cheng, Y.; Liu, X.; Li, X.; Dong, J. Thermal, light and pH triple stimulated changes in self-assembly of a novel small molecular weight amphiphile binary system. RSC Adv. 2015, 5, 640−642. (9) Lu, H.; Xue, M.; Wang, B.; Huang, Z. pH-regulated surface property and pH-reversible micelle transition of a tertiary amine-based gemini surfactant in aqueous solution. Soft Matter 2015, 11, 9135− 9143. (10) Zhang, Y.; An, P.; Liu, X. Bell-shaped sol−gel−sol conversions in pH-responsive worm-based nanostructured fluid. Soft Matter 2015, 11, 2080−2084. (11) Yan, Z.; Dai, C.; Zhao, M.; Zhao, G.; Li, Y.; Wu, X.; Du, M.; Liu, Y. pH-switchable wormlike micelle formation by N-alkyl-N-methylpyrrolidinium bromide-based cationic surfactant. Colloids Surf., A 2015, 482, 283−289. (12) Lv, J.; Qiao, W.; Xiong, C. Synthesis and surface properties of a pH-regulated and pH-reversible anionic gemini surfactant. Langmuir 2014, 30, 8258−8267. (13) Feng, H.; Verstappen, N. A. L.; Kuehne, A. J. C.; Sprakel, J. Well-defined temperature-sensitive surfactants for controlled emulsion coalescence. Polym. Chem. 2013, 4, 1842−1847. (14) Su, X.; Robert, T.; Mercer, S. M.; Humphries, C.; Cunningham, M. F.; Jessop, P. G. A conventional surfactant becomes CO2responsive in the presence of switchable water additives. Chem.Eur. J. 2013, 19, 5595−5601. (15) Ceschia, E.; Harjani, J. R.; Liang, C.; Ghoshouni, Z.; Andrea, T.; Brown, R. S.; Jessop, P. G. Switchable anionic surfactants for the

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.6b03645. Synthesis of (C12Se)2, C12SeO2H, and the seleninates. Additional results and discussion: aging curves, pKa, DLS results, ESI-MS, FT-IR spectra, contact angle, UV−vis absorption spectra and the working curve, the dependence of the relative stability of the emulsion on the pH and cycles, tuning IFT, γ, foam, and wetting ability step by step (PDF) Demulsification of the emulsion triggered by HCl; phase separation (triggered by HCl) (MPG) Re-emulsification of the emulsion triggered by the addition of NaOH (MPG) Re-emulsification of the emulsion triggered by the addition of NH3·H2O (MPG) Re-emulsification of the emulsion triggered by the addition of K2CO3 (MPG) G

DOI: 10.1021/acs.langmuir.6b03645 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir remediation of oil-contaminated sand by soil washing. RSC Adv. 2014, 4, 4638−4645. (16) Zhu, Y.; Jiang, J.; Cui, Z.; Binks, B. P. Responsive aqueous foams stabilized by silica nanoparticles hydrophobised in situ with a switchable surfactant. Soft Matter 2014, 10, 9739−9745. (17) Zhang, Y.; Zhang, Y.; Wang, C.; Liu, X.; Fang, Y.; Feng, Y. CO2responsive microemulsion: Reversible switching from an apparent single phase to near-complete phase separation. Green Chem. 2016, 18, 392−396. (18) Brown, P.; Wasbrough, M. J.; Gurkan, B. E.; Hatton, T. A. CO2responsive microemulsions based on reactive ionic liquids. Langmuir 2014, 30, 4267−4272. (19) Zhang, J.; Han, B. Supercritical or compressed CO2 as a stimulus for tuning surfactant aggregations. Acc. Chem. Res. 2013, 46, 425−433. (20) Scott, L. M.; Robert, T.; Harjani, J. R.; Jessop, P. G. Designing the head group of CO2-triggered switchable surfactants. RSC Adv. 2012, 2, 4925−4931. (21) Wang, G.; Kang, Y.; Tang, B.; Zhang, X. Tuning the surface activity of Gemini amphiphile by the host−guest interaction of cucurbit[7]uril. Langmuir 2015, 31, 120−124. (22) Brown, P.; Bromberg, L.; Rial-Hermida, M. I.; Wasbrough, M.; Hatton, T. A.; Alvarez-Lorenzo, C. Magnetic surfactants and polymers with gadolinium counterions for protein separations. Langmuir 2016, 32, 699−705. (23) Brown, P.; Bushmelev, A.; Butts, C. P.; Cheng, J.; Eastoe, J.; Grillo, I.; Heenan, R. K.; Schmidt, A. M. Magnetic control over liquid surface properties with responsive surfactants. Angew. Chem., Int. Ed. 2012, 51, 2414−2416. (24) Saji, T.; Hoshino, K.; Aoyagui, S. Reversible formation and disruption of micelles by control of the redox state of the head group. J. Am. Chem. Soc. 1985, 107, 6865−6868. (25) Aydogan, N.; Abbott, N. L. Effect of electrolyte concentration on interfacial and bulk solution properties of ferrocenyl sufactants with anionic headgroups. Langmuir 2002, 18, 7826−7830. (26) Tsuchiya, K.; Orihara, Y.; Kondo, Y.; Yoshino, N.; Ohkubo, T.; Sakai, H.; Abe, M. Control of viscoelasticity using redox reaction. J. Am. Chem. Soc. 2004, 126, 12282−12283. (27) Sakai, H.; Imamura, H.; Kondo, Y.; Yoshino, N.; Abe, M. Reversible control of vesicle formation using electrochemical reaction. Colloids Surf., A 2004, 232, 221−228. (28) Fan, H.; Han, F.; Liu, Z.; Qin, L.; Li, Z.; Liang, D.; Ke, F.; Huang, J.; Fu, H. Active control of surface properties and aggregation behavior in amino acid-based Gemini surfactant systems. J. Colloid Interface Sci. 2008, 321, 227−234. (29) Rosslee, C.; Abbott, N. L. Active control of interfacial properties. Curr. Opin. Colloid Interface Sci. 2000, 5, 81−87. (30) Long, J.; Tian, S.; Li, G.; Li, L. Micellar aggregation behavior and electrochemically reversible solubilization of a redox-active nonionic surfactant. J. Solution Chem. 2015, 44, 1163−1176. (31) Zhang, Y.; Kong, W.; Wang, C.; An, P.; Fang, Y.; Feng, Y.; Qin, Z.; Liu, X. Switching wormlike micelles of selenium-containing surfactant using redox reaction. Soft Matter 2015, 11, 7469−7473. (32) Han, P.; Ma, N.; Ren, H.; Xu, H.; Li, Z.; Wang, Z.; Zhang, X. Oxidation-responsive micelles based on selenium-containing polymeric superamphiphile. Langmuir 2010, 26, 14414−14418. (33) Fredga, A. Organic selenium chemistry. Ann. N.Y. Acad. Sci. 1972, 192, 1−9. (34) Rebane, E. On the preparation and properties of long-chain seleninic acids with some remarks on their infrared spectra. Ark. Kemi 1967, 26, 345−356. (35) Wang, L.; Cao, W.; Yi, Y.; Xu, H. Dual redox responsive coassemblies of diselenide-containing block copolymers and polymer lipids. Langmuir 2014, 30, 5628−5636. (36) Ma, N.; Li, Y.; Xu, H.; Wang, Z.; Zhang, X. Dual redox responsive assemblies formed from diselenide block copolymers. J. Am. Chem. Soc. 2010, 132, 422−443. (37) Harkins, W. D.; Brown, F. E. The determination of surface tension (free surface energy), and the weight of falling drops: The

surface tension of water and benzene by the capillary height method. J. Am. Chem. Soc. 1919, 41, 499−524. (38) Koga, M.; Yamamichi, Y.; Nomoto, Y.; Irie, M.; Tanimura, T.; Yoshinaga, T. Rapid determination of anionic surfactant by improved spectrophotometric method using methylene blue. Anal. Sci. 1999, 15, 563−568. (39) Ali, S. T.; Karamat, S.; Kóňa, J.; Fabian, W. M. F. Theoretical prediction of pKa values of seleninic, selenenic, sulfinic, and carboxylic acids by quantum-chemical methods. J. Phys. Chem. A 2010, 114, 12470−12478. (40) McCullough, J. D.; Gould, E. S. The dissociation constants of some mono-substituted benzeneseleninic acids. J. Am. Chem. Soc. 1949, 71, 674−676. (41) Yang, L.; Takisawa, N.; Hayashita, T.; Shirahama, K. Colloid chemical characterization of the photosurfactant 4-ethylazobenzene 4′(oxyethyl) trimethyl ammonium bromide. J. Phys. Chem. 1995, 99, 8799−8803. (42) Gallardo, B. S.; Hwa, M. J.; Abbott, N. L. In situ and reversible control of surface activity of ferrocenyl surfactants in aqueous solutions. Langmuir 1995, 11, 4209−4212. (43) Du, P.; Viswanathan, U. M.; Xu, Z.; Ebrahimnejad, H.; Hanf, B.; Burkholz, T.; Schneider, M.; Bernhardt, I.; Kirsch, G.; Jacob, C. Synthesis of amphiphilic seleninic acid derivatives with considerable activity against cellular membranes and certain pathogenic microbes. J. Hazard. Mater. 2014, 269, 74−82. (44) Stuhr-Hansen, M.; Ebert, B.; Krogsgaard-Larsen, P.; Kehler, J. Synthesis and pharmacology of seleninic acid analogues of the inhibitory neurotransmitter γ-aminobutyric acid. Org. Lett. 2000, 2, 7− 9. (45) Back, T. G.; Moussa, Z. Diselenides and allyl selenides as glutathione peroxidase mimetics. Remarkable activity of cyclic seleninates produced in situ by the oxidation of allyl ω-hydroxyalkyl selenides. J. Am. Chem. Soc. 2003, 125, 13455−13460.

H

DOI: 10.1021/acs.langmuir.6b03645 Langmuir XXXX, XXX, XXX−XXX