A CO2/N2-Responsive Pickering Emulsion Stabilized by Novel Switchable Surface-Active Alumina Nanoparticles

An Chen, Duo Wang, Jingsi Chen, Jianhong Xu*, Hongo Zeng,*

a State Key Laboratory of Chemical Engineering, Department of Chemical Engineering, Tsinghua University, Beijing 100084, China

b Department of Chemical and Materials Engineering, University of Alberta, Edmonton, AB T6G 1H9, Canada

Keywords:CO2/N2 responsive Pickering emulsion Switchable surfactant Alumina nanoparticles Superamphiphile

ABSTRACT This article reports the development of a novel switchable Pickering emulsion with rapid CO2/N2 responsiveness,which is stabilized using alumina nanoparticles hydrophobized in situ with a trace amount of a switchable superamphiphile via electrostatic interactions. With the introduction of CO2 for 30 s, the Pickering emulsion can be spontaneously demulsified with complete phase separation; the emulsion can then be reconstructed in response to N2 purging for 10 min followed by homogenization.Moreover,the stable Pickering emulsion can be stored for more than 60 days at room temperature without any visible change. The CO2/N2-responsive behavior of the switchable Pickering emulsion is attributed to the reversible desorption/adsorption of the switchable surfactants on the surfaces of the alumina nanoparticles upon the alternative bubbling of CO2 or N2.Thanks to the simple fabrication of the surfactant and the hydrophobization of the alumina nanoparticles, this research has developed an extremely facile and cost-efficient method for preparing a rapidly CO2/N2-responsive switchable Pickering emulsion.The dosage of the switchable surfactants has been significantly reduced by nearly 1500 times(from 150 to 0.1 mmol·Lˉ1) as compared with the dosage used in previous studies. Moreover, the as-prepared CO2/N2-responsive switchable Pickering emulsion is environmentally friendly, mild, and nontoxic; thus,it holds great potential for practical applications with considerable economic and environmental benefits,such as oil transport, fossil fuel production, environmental gases detection, and the encapsulation and release of active ingredients.? 2022 THE AUTHORS. Published by Elsevier LTD on behalf of Chinese Academy of Engineering and Higher Education Press Limited Company. This is an open access article under the CC BY-NC-ND license(http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction

Switchable emulsions have been widely applied in a myriad of fields, such as drug delivery, oil transport, and fossil fuel production [1-3]. The preparation of a switchable surfactant is the key to switchable or stimuli-responsive emulsions,which can be reversibly transformed from surface-active forms to surface-inactive forms by means of certain triggers. To date, a variety of triggers have been employed for the fabrication of switchable emulsions,and their systems are mainly classified into two categories. Many systems can be switched by a single trigger, including pH [4-6],temperature[7,8],carbon dioxide/nitrogen(CO2/N2)[2,9-13],light irradiation [14,15], a magnetic field [16,17], and redox [18,19].Other systems are dependent on multiple triggers, which can consist of magnetic field-temperature [20], pH-temperature[21,22], light-temperature [23,24], CO2-light [25], and CO2-redox[26] combinations. Compared with triggers based on the addition of acids, bases, oxidants, and reductants, which present inevitable economic and environmental issues, the use of gases such as CO2/N2as triggers has attracted tremendous research interest over the last few decades because they are mild,nontoxic,cost-efficient,environmentally friendly, readily removable, and recyclable.

A CO2/N2-responsive switchable surfactant was first proposed in 2006 [9], where a long-chain alkyl amidine was synthesized to achieve a reversible transformation between emulsification and demulsification by means of a CO2/N2trigger.The long-chain alkyl amidine was protonated by the addition of CO2to become charged amidinium in order to stabilize the emulsion.By purging N2into the emulsion to remove the CO2, the charged amidinium was altered into the neutral form,resulting in a destabilized emulsion.In other works, a rapidly responsive superamphiphile was successfully applied to oil-in-water (O/W) emulsions to achieve CO2/N2-triggered switchable behaviors [11,13]. The superamphiphile was assembled by simply mixing two eco-friendly compounds through noncovalent electrostatic interactions—unlike the amidine in the earlier work,which was synthesized via covalent bonds.

Emulsions stabilized by colloid particles (so-called Pickering emulsions) usually have better stability than emulsions stabilized by surfactants[27-31],as a dense particle film covers the oil/water interface to prevent the coalescence of droplets.The steric barriers between Pickering emulsion droplets are relatively difficult to destroy due to the high desorption energy of particles from the oil/water interface, making it a huge challenge to induce the demulsification of Pickering emulsions. However, in some industrial processes, the stability of an emulsion is only required temporarily and is followed by a controlled demulsification process,which has attracted great interest. Numerous attempts have been made to develop stimuli-responsive or switchable Pickering emulsions in recent years [6,10,21,32-38]. In general, the initially charged colloid particles are usually hydrophilic and surfaceinactive in order to stabilize the emulsion; however, after the adsorption of oppositely charged switchable surfactants, the particles can be hydrophobized in situ to induce the formation of a Pickering emulsion, and the destabilization of the system can be realized by certain triggers. Jiang et al. [27] reported that negatively charged silica nanoparticles could be hydrophobized in situ through the adsorption of long-chain alkyl amidines via electrostatic interactions, resulting in the successful generation of a CO2/N2-responsive switchable Pickering emulsion. Nevertheless,the synthesis of long-chain alkyl amidines is relatively complicated,and the switching time is relatively long in comparison with some other CO2/N2-responsive surfactants.

In this work,we have developed a rapidly switchable Pickering emulsion stabilized by modified surface-active alumina nanoparticles, which display rapid responsiveness to CO2within 30 s to demulsify the stable Pickering emulsion; the emulsion can then be recovered back to its initial state through purging with N2. The switchable surface-active alumina nanoparticles are developed by in situ hydrophobization through the adsorption of a trace amount of superamphiphile at the surfaces of the alumina nanoparticles,and the superamphiphile is prepared by a facile mixing of anionic fatty acid (oleic acid) and cationic amine (Jeffamine D-230) at a 1:1 molecular ratio. The switchable Pickering emulsion prepared using the hydrophobized alumina nanoparticles responds to gases rapidly and efficiently,as the stable Pickering emulsion undergoes a complete phase separation upon having CO2bubbled through it for 30 s,while the oil/water mixture can be re-emulsified to a stable Pickering emulsion by purging it with N2for 10 min.The adsorption behavior of the superamphiphile at the alumina nanoparticle surfaces has been well confirmed by measurements of zeta potential and interfacial tension. Furthermore, we have determined the switchable mechanism of the developed Pickering emulsions,which mainly occurs through the reversible desorption/adsorption of fatty acid molecules on the alumina nanoparticles upon alternatively bubbling CO2or N2. Compared with the previously reported CO2/N2-responsive emulsions stabilized by superamphiphiles[11,12],the dosage of the switchable surfactant in this work is significantly reduced by nearly 1500 times(from 150 to 0.1 mmol·Lˉ1).The developed switchable Pickering emulsion is environmentally friendly,with useful implications for practical applications such as oil transport,fossil fuel production,environmental gases detection,and the encapsulation and release of active ingredients.

2. Materials and methods

2.1. Materials

Hydrophilic alumina nanoparticles (Al2O3, ＞99.8%) with a primary particle diameter of 13 nm and a specific surface area of 85-115 m2·gˉ1(Brunauer-Emmett-Teller, BET) were purchased from Sigma-Aldrich, USA. Poly(propylene glycol) bis(2-aminopropyl ether) with an average molecular weight (Mw) of about 230 (Jeffamine D-230), which is a primary diamine with active amine groups, was purchased from Sigma-Aldrich, USA.Oleic acid (technical grade) was also purchased from Sigma-Aldrich,USA.Toluene(＞98%)was purchased from Fisher Scientific,Canada, and was used without purification as the oil phase for all emulsions.All the reagents were used as received.Deionized water(resistance ≥18.2 MΩ·cm) was obtained from a Milli-Q water purification system (Thermo Scientific, USA) and used in all the experiments.Unless specified otherwise,all experiments were carried out at room temperature.

2.2. Preparation of CO2/N2-responsive Pickering emulsion

Oleic acid and Jeffamine D-230 were mixed at a molecular ratio of 1:1, allowing the formation of a CO2/N2-responsive switchable superamphiphile—that is, a CO2/N2-responsive gemini-type amphiphile surfactant—through noncovalent electrostatic interactions(Fig.1)[11].The structure of the assembled superamphiphile was characterized by Fourier-transform infrared spectroscopy(FTIR) (Nicolet Is50, Thermo Scientific) and proton nuclear magnetic resonance spectroscopy (1H NMR, 400 MHz, CDCl3); the results are shown in Figs. S1 and S2 in Appendix A. One of the amino groups of Jeffamine D-230 received a proton from the carboxyl group of oleic acid to transform into-NH3+,while the neutral carboxyl group was transformed into the ionized form (-COO-),resulting in the bending vibration peak of -NH2shifting from 1586 to 1625 cmˉ1, as shown in Fig. S1. The other amino group in the molecule was unreacted and existed at the other end of the superamphiphile; this was well confirmed by the vibrational peaks at 3362 and 3269 cmˉ1in the high wavenumber regime,which were similar to those of the unprotonated amino groups in Jeffamine D-230. However, when different molar ratios of oleic acid and Jeffamine D-230 were mixed, the assembled superamphiphiles had different chemical structures. More details are provided in Refs. [11,12,39], and the FTIR and1H NMR spectra are provided in Figs. S1 and S2.

Fig. 1. Chemical structure and formation route of the CO2/N2-responsive superamphiphile.

Superamphiphile solutions with different concentrations (0.1,0.2, 0.3, 0.6, and 1.0 mmol·Lˉ1) were prepared by dissolving the surfactant in deionized water(resistance ≥18.2 MΩ·cm).A certain amount of alumina nanoparticles was weighed into a glass vessel;then, deionized water was added to the vessel to prepare the alumina nanoparticle dispersions (0.1 wt%, 0.5 wt%, and 1.0 wt%).Solutions containing the superamphiphile and alumina nanoparticles were prepared by adding different volumes of surfactant solution (0.1, 0.2, 0.3, 0.6, and 1.0 mmol·Lˉ1) to the 0.5 wt% alumina nanoparticle suspensions, which were used as the aqueous phase for the preparation of Pickering emulsions. The superamphiphile and the alumina nanoparticles were dispersed using an ultrasonic probe(2510R-DTH,Marshall Scientific,USA)at an output of 100 W for 40 s.Three types of O/W emulsions in the presence of bare alumina nanoparticles alone, superamphiphile alone, and hydrophobized alumina nanoparticles with a trace amount of superamphiphile, were prepared by mixing 5.0 mL of the desired aqueous phase and 5.0 mL of toluene through homogenization in a 20 mL glass vial at room temperature (25 °C).

2.3. Interfacial tension measurement

The interfacial tension (IFT) between toluene and the aqueous superamphiphile solution in the absence or presence of alumina nanoparticles was characterized by means of the pendant drop method using a tensiometer (ramé-hart instrument co., USA) at room temperature. The aqueous solution, as the heavy phase,was loaded into a syringe with a straight needle, which was inserted in a sealed cell filled with toluene. A pendant droplet was generated at the tip of the needle by a syringe pump, with the volume remaining constant in each interfacial tension measurement. The droplet profile was captured by a high-speed charge-coupled device (CCD) camera every 1 s for 600 s. All the experiments were performed on an anti-vibration table.The aqueous phase used for the measurements included pure superamphiphile solutions with surfactant concentrations of 0, 0.1, 0.2,0.3, 0.6, 1.0, 3.0, 6.0, 10.0, and 20.0 mmol·Lˉ1, as well as solutions containing 0.5 wt% alumina nanoparticles and superamphiphiles at concentrations of 0.1, 0.2, 0.3, 0.6, and 1.0 mmol·Lˉ1.

2.4. Zeta potential and particle size measurements

Alumina nanoparticles of 0.5 wt%were blended with superamphiphile solutions at different concentrations(0.1,0.2,0.3,0.6,and 1.0 mmol·Lˉ1). The zeta potentials of the blended solutions, with or without introducing CO2,were measured using a Zetasizer Nano ZSP (Malvern Instruments, UK). Dynamic light scattering was employed to monitor the size of the alumina nanoparticles in the solutions using the same instrument (Zetasizer Nano ZSP).

2.5. Switchable behavior of the Pickering O/W emulsion

The switchable behaviors of the prepared Pickering O/W emulsions were evaluated by alternatively introducing CO2or N2.To be specific,the Pickering O/W emulsion was constantly bubbled with CO2at a flow rate of 300 mL·minˉ1for 30 s, allowing the occurrence of demulsification. To re-stabilize the emulsion, the O/W mixture was stored in a water bath at 60 °C and purged with N2at a flow rate of 300 mL·minˉ1for 10 min, followed by homogenization.Here,a faster flow rate resulted in a higher switching efficiency. However, a faster flow rate also had a certain negative impact on the actual operation, such as liquid splashing, energy consumption issues, and so on. Therefore, the specific operating conditions described in this work are the most suitable results,based on our continuous optimization. During the cycles of bubbling CO2and N2, the pH values of the blended aqueous solutions were monitored. The stability of the Pickering emulsion was evaluated by visual observation and optical microscopy after being stored at room temperature for more than 60 days.

3. Results and discussion

3.1. Stabilization of Pickering emulsion by alumina nanoparticles and superamphiphile

Alumina nanoparticles are highly hydrophilic in deionized water and carry a positive charge, as is evident from the zeta potential measurement ((37.8±0.5) mV) shown in Fig. 2(a). It is extremely difficult for bare alumina nanoparticles to completely disperse as single particles in an aqueous solution due to their high surface energy; thus, they usually form aggregates in aqueous solutions. In this work, the 0.5 wt% alumina nanoparticles usually formed aggregates with an average diameter of 182.3 nm in aqueous solutions (i.e., in deionized water at 25 °C), even though the suspension was sonicated using an ultrasonic probe, which is demonstrated by the size distribution shown in Fig.S3 in Appendix A. The toluene-in-water emulsions were prepared at different concentrations of alumina nanoparticles (0, 0.1 wt%, 0.5 wt%, and 1.0 wt%) followed by homogenization. However, separated phases were observed in all samples, implicating that the bare alumina nanoparticles alone were too hydrophilic to stabilize the O/W emulsion, as shown in Fig. S4 in Appendix A. Complete phase separation occurred rapidly,and most of the alumina nanoparticles sank into the aqueous phase due to their high density(3.95 g·mˉ3)and hydrophilicity.

Toluene-in-water emulsions were also prepared using pure superamphiphile solutions at concentrations of 0.1, 0.2, 0.3, 0.6,and 1.0 mmol·Lˉ1, followed by homogenization. The O/W emulsions were stable for 24 h in the presence of a superamphiphile with a concentration greater than 0.3 mmol·Lˉ1, whereas separated layers were observed with a relatively long aging time(10 days) in all cases, as clearly shown in Fig. S5 in Appendix A.As shown in Fig. 2(b), even when the concentration was as high as 1.0 mmol·Lˉ1,the interfacial tension of the toluene/water interface still remained relatively high (～11.6 mN·mˉ1), which suggested that it was not possible to form a long-term stable emulsion with a small amount of superamphiphile alone. This is why only high concentrations of superamphiphiles were used in previous studies to stabilize emulsions [11,12].

In contrast, the synergistic utilization of alumina nanoparticles and a superamphiphile dramatically promoted the stability of the O/W emulsions(Figs.2(c)-(e)).After the O/W emulsions had been stored for 24 h,10 days,and 60 days at room temperature,no obvious coalescence of oil droplets was observed, even though the superamphiphile concentration was down to 0.1 mmol·Lˉ1. This result indicated the excellent stability of the emulsions prepared with the combination of alumina nanoparticles and superamphiphile. As the bare alumina nanoparticles are hydrophilic and the pure superamphiphile cannot effectively reduce the interfacial tension, the enhanced stability can be attributed to the in situ hydrophobization of the alumina nanoparticles by the adsorption of the superamphiphile through electrostatic interactions, which was indicated by the change in the zeta potential of the alumina nanoparticles (Fig. 2(a)). The surfaces of the modified alumina nanoparticles were partially hydrophobic due to the formation of a surfactant monolayer, which stabilized the Pickering emulsion for a long time.

3.2. Switchable behavior of the CO2/N2-responsive Pickering emulsion

To demonstrate the switching property of the Pickering emulsion, a toluene-in-water (5.0 mL/5.0 mL) emulsion that was stabilized using 0.5 wt%alumina nanoparticles and a 0.3 mmol·Lˉ1CO2/N2-responsive superamphiphile was first prepared in a 20 mL glass bottle,followed by homogenization.The switchable behavior

Fig.2. (a)Zeta potentials of alumina nanoparticles dispersed in the surfactant solutions of different concentrations.The dotted line represents the boundary at which the zeta potential=0. (b) Interfacial tension of the toluene/aqueous solutions as a function of surfactant concentration. The dashed lines represent the boundaries of surfactant concentrations of 0.1 and 1.0 mmol·Lˉ1. (c-e) Digital photographs of Pickering toluene/water emulsions (5.0 mL/5.0 mL) stabilized by alumina nanoparticles (0.5 wt%) and different concentrations of superamphiphile for an aging time of(c)24 h,(d)10 days,and(e)60 days.The superamphiphile concentrations from(1)to(5)are 0.1,0.2,0.3,0.6,and 1.0 mmol·Lˉ1, respectively.of the resultant Pickering emulsion was investigated by alternatively bubbling CO2or N2for several cycles; the results are shown in Fig. 3. Without any treatment, the CO2/N2-responsive Pickering emulsion had a homogeneous appearance, and no visible signs of coalescence, demulsification, or phase separation occurred after aging the emulsion for 60 days, which demonstrated the superior stability of the Pickering emulsion(Fig.3(f)).With the introduction of CO2at a flow rate of 300 mL·minˉ1for 30 s at room temperature, the Pickering emulsion spontaneously demulsified and completely separated (Fig. 3(b)); reconstruction of the emulsion then occurred in response to N2purging at a flow rate of 300 mL·minˉ1at 60 °C for 10 min, followed by homogenization (Fig. 3(c)). The results indicated that the switchable behavior (i.e., emulsification and demulsification) of the CO2/N2-responsive Pickering emulsion was highly sensitive and reversible upon alternately bubbling CO2or N2. The reformed Pickering emulsion displayed similar droplet sizes to that before demulsification, as shown in Fig. 3(c′). Moreover, Figs. 3(d) and (e) highlight this switchable behavior in the second cycle,demonstrating the excellent repeatability of the proposed Pickering emulsion. If the surfactant could be properly replenished (as multiple introductions of gas consume the surfactant), then the maximum number of cycles could be higher.

During three cycles of alternatively bubbling CO2or N2into the Pickering emulsion,the pH value of the aqueous solution was monitored, as shown in Fig. 4. The original pH of the aqueous solution was 7.95, due to the adsorption of the ionized -COOH functional group (i.e., -COO-, the functional group of the superamphiphile)on the surface of the positively charged alumina nanoparticles.During the process of bubbling CO2/N2, the fluctuation of the peak of the pH value was extremely low (7.95, 8.11, 7.98, and 8.06 for each peak), which may have been caused by an instrument measurement error. As CO2at a flow rate of 300 mL·minˉ1for 30 s at room temperature was introduced into the Pickering emulsion,the dissolution of CO2resulted in a decrease in the pH to 4.83 and the protonation of carbonic acid. Moreover, the pH returned to the original value after bubbling with N2for 10 min at 60 °C at a flow rate of 300 mL·minˉ1to remove CO2. This finding indicated that the neutral superamphiphile transformed back to its active form for re-adsorption onto the surfaces of the alumina nanoparticles. These results fully demonstrated that the rapid responsiveness of the Pickering emulsion was provided by the switchable hydrophobization of the alumina nanoparticles, which could be modulated by the CO2trigger.

3.3. Switchable mechanism of the novel Pickering emulsion

Fig. 5 clearly shows that, without the addition of CO2, the zeta potential of the prepared Pickering emulsion displayed a significant decrease with increasing superamphiphile concentration. In the initial state, the zeta potential of the positively charged alumina nanoparticles dispersed in pure water was around (37.8±0.5) mV. It then decreased gradually as the surfactant concentration increased, indicating that the amount of negatively charged superamphiphile absorbed onto the surfaces of the alumina nanoparticles was increased.It was notable that the zeta potential value became negative when the superamphiphile concentration was greater than ～0.7 mmol·Lˉ1,indicating that the modified alumina nanoparticles were negatively charged due to the adsorption of a large amount of active superamphiphile molecules.In contrast,when CO2was bubbled into the Pickering emulsion, the zeta potential remained positive at a constant value over the studied range of the superamphiphile concentration, indicating that the adsorption of neutral superamphiphile onto the surfaces of alumina nanoparticles was extremely low.

Fig. 3. (a-e) Appearance of the switchable Pickering emulsion stabilized by using 0.5 wt% alumina nanoparticles and 0.3 mmol·Lˉ1 superamphiphile upon alternatively bubbling(b,d)CO2 or(c,e)N2 followed by homogenization.(f)Appearance of the Pickering emulsion after being stored for 60 days without any treatment.(a′,c′,e′,f′)Optical micrographs of droplet sizes of Pickering emulsion in (a), (c), (e), and (f), respectively.

Fig. 4. Transient pH of the switchable Pickering emulsion for three cycles of alternatively bubbling CO2 or N2.

In order to investigate the CO2/N2-responsive behavior of the Pickering emulsions,samples were prepared with 0.5 wt%alumina nanoparticles and 0.3 mmol·Lˉ1superamphiphile;in this case,the zeta potential was determined to be (5.9±0.8) mV (point a in Fig. 5) before CO2treatment and (36.6±0.7) mV (point b) in the presence of CO2. Furthermore, similar values (i.e., the overlapping points of a and c and of b and d in Fig.5)were obtained in the second cycle of alternatively bubbling N2and CO2, which indicated that the absorption/desorption of the switchable surfactant from the solid/water interface took place following the alternate bubbling of N2or CO2through the solution, leading to the emulsification/demulsification of the system.

Fig. 5. Zeta potentials of alumina nanoparticles (0.5 wt%) dispersed in aqueous solutions of switchable surfactants, in both the active(ionic)and inactive (neutral)forms, as a function of surfactant concentration. The zeta potential alternation during cycles of bubbling CO2/N2 through the dispersion with 0.3 mmol·Lˉ1 surfactant was as follows: point a: (5.9±0.8) mV before CO2 treatment; point b:(36.6±0.7) mV in the presence of CO2; point c: (5.9±0.8) mV after N2 treatment;point d: (36.6±0.7) mV in the presence of CO2 during the second cycle. These repeating values indicate that the absorption/desorption of the switchable surfactant from the solid/water interface took place following the alternative bubbling of N2 or CO2 through the solution, leading to reversible emulsification and demulsification.

The interfacial tensions of the as-prepared Pickering emulsions were investigated in order to unravel the emulsification and demulsification mechanisms, as displayed in Fig. 6. It is well accepted that the formation of surfactant-stabilized emulsions occurs due to a decrease in interfacial tension [40], which contributes to a lowered interfacial energy and enhanced stability of the O/W interface. In the presence of 1.0 mmol·Lˉ1superamphiphile, the interfacial tension decreased from 32.5 mN·mˉ1(without superamphiphile) to about 11.6 mN·mˉ1. A negligible decrease of the interfacial tension was observed when the superamphiphile concentration was lower than 0.1 mmol·Lˉ1.This slight decrement of the interfacial tension is highly consistent with the results of the emulsification experiment(Fig.S5),in which a trace amount of superamphiphile could not stabilize the emulsions. On the other hand, Pickering emulsions with or without bubbling CO2exhibited profound stability in terms of interfacial intension with a superamphiphile concentration ranging from 0.1 to 1.0 mmol·Lˉ1(Fig. 6(b)); this was mainly due to the massive adsorption of active superamphiphile onto the surface of the alumina nanoparticles(without CO2)and to the low interfacial activity and water solubility of the inactive superamphiphile [41] (with CO2),respectively.These results reveal that the alumina nanoparticles were partially hydrophobized by the superamphiphile and were then adsorbed at the oil/water interface,allowing the formation of a stable Pickering emulsion.

Fig.6. Interfacial tension of toluene and aqueous solutions measured at 25 °C as a function of(a)time(the unit of the numbers is mmol·Lˉ1)and(b)surfactant concentration after 30 min.

The switching mechanism of the CO2/N2-responsive Pickering emulsion is illustrated in Fig. 7. In an aqueous solution, the active superamphiphile is assembled through electrostatic interactions between the amino groups of Jeffamine D-230 (receiving a proton and transforming into-NH3+)and the carboxyl groups of oleic acid(transforming into the ionized form (-COO-)). The positively charged alumina nanoparticles are hydrophobized in situ by the adsorption of active superamphiphile through electrostatic interactions, which was confirmed by the zeta potential results of the alumina nanoparticles and the results for the interfacial tension of the toluene/water interface.The surfaces of the alumina particles are covered with a surfactant monolayer with the hydrocarbon chains protruding toward the water,rendering a considerably stable Pickering emulsion. When CO2is introduced into the Pickering emulsion, the ionized carboxyl group of the superamphiphile is protonated by H+into a neutral form that is inactive, resulting in the desorption of the superamphiphile from the surface of the alumina nanoparticles.As neither the hydrophilic alumina nanoparticles nor the inactive superamphiphile molecules are capable of stabilizing emulsions, phase separation occurs as a consequence.With the removal of CO2upon purging with inert N2, the reaction reverses back to the initial state, as the ionized carboxyl group is reformed and absorbed onto the surfaces of alumina nanoparticles,switching the separated phases to a stable Pickering emulsion.

4. Conclusions

In this work,CO2/N2-responsive Pickering emulsions were fabricated by employing alumina nanoparticles and a trace amount of synthesized superamphiphile, resulting in emulsions with rapid,highly efficient, and reversible switching characteristics. It was demonstrated that either hydrophilic alumina nanoparticles or a trace amount of superamphiphile are incapable of stabilizing emulsions on their own; however, when the alumina nanoparticles are hydrophobized with the ionized carboxyl group in the superamphiphile,they assemble into synergistic complexes and contribute to the high stability of the as-prepared Pickering emulsions. The switchable behavior of the Pickering emulsions in response to CO2/N2was investigated, and it was demonstrated that the Pickering emulsions could rapidly, reversibly, and repeatedly switch from a stable status (in the initial state and after purging with N2)to separated phases(when stimulated by CO2).The underlying mechanism was studied by measuring the zeta potential,particle size, and interfacial tension; the results suggested that the reversible desorption/adsorption of the superamphiphile on the surfaces of the alumina nanoparticles upon alternatively bubbling with CO2or N2was responsible for the responsive behavior. It should be noted that only a trace amount of surfactant (0.1 mmol·Lˉ1) is required to form a stable and switchable Pickering emulsion, as compared with the much higher dosages (e.g.,typically ～150 mmol·Lˉ1)that have been used in previous studies;thus,the proposed method greatly reduces the dosage of surfactant by nearly 1500 times. Moreover, the as-prepared CO2/N2-responsive Pickering emulsion is environmentally friendly, mild,nontoxic, and cost-efficient. These results indicate that the novel CO2/N2-responsive Pickering emulsion holds great promise for many practical applications, such as oil transport, fossil fuel production, environmental gases detection, and the encapsulation and release of active ingredients.

Fig. 7. Schematic diagram of the switching mechanism of the CO2/N2-responsive Pickering emulsion. NP: nanoparticle.

Acknowledgments

This research was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Canada Research Chairs Program (Hongbo Zeng), and the China Scholarship Council (CSC) (An Chen).

Compliance with ethics guidelines

An Chen,Duo Wang,Jingsi Chen,Jianhong Xu,and Hongbo Zeng declare that they have no conflict of interest or financial conflicts to disclose.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.eng.2020.08.031.