The Effect of Surface Confined Gold Nanoparticles in Blocking ... - MDPI

membranes Article

The Effect of Surface Confined Gold Nanoparticles in Blocking the Extraction of Nitrate by PVC-Based Polymer Inclusion Membranes Containing Aliquat 336 as the Carrier Ya Ya N. Bonggotgetsakul, Robert W. Cattrall and Spas D. Kolev *

ID

School of Chemistry, The University of Melbourne, Victoria 3010, Australia; [email protected] (Y.Y.N.B.); [email protected] (R.W.C.) * Correspondence: [email protected]; Tel.: +61-3-8344-7931 Received: 1 January 2018; Accepted: 22 January 2018; Published: 25 January 2018

Abstract: Clusters of gold nanoparticles (AuNPs) formed on the surface of PVC-based polymer inclusion membranes (PIMs) with a liquid phase containing Aliquat 336 as the carrier and in some cases 1-dodecanol or 2-nitrophenol octyl ether as plasticizers were found to inhibit the extraction of nitrate by the PIMs. This observation was based on gradually increasing the mass of AuNPs on the membrane surface and testing the ability of the membrane to extract nitrate after each increase. In this way, it was possible to determine the so-called “critical AuNP masses” at which the studied membranes ceased to extract nitrate. On the basis of these results, it can be hypothesized that the surfaces of these PIMs are not homogeneous with respect to the distribution of their membrane liquid phases, which are present only at certain sites. Extraction takes place only at these sites, and at the “critical AuNP mass” of a PIM, all these extraction sites are blocked and the membrane loses its ability to extract. Keywords: polymer inclusion membrane (PIM); gold nanoparticles (AuNPs); surface morphology; Aliquat 336; nitrate extraction

1. Introduction Polymer inclusion membranes (PIMs) are a type of liquid membranes composed of a base-polymer and a membrane liquid phase consisting of an extractant (often referred to as the carrier) and in some cases a plasticizer or modifier [1–3]. The most frequently used base-polymers are poly(vinyl chloride) (PVC) and cellulose triacetate (CTA), but other base-polymers such as poly(vinylidene-fluoride-co-hexafluoropropylene) (PVDF-HFP) and semi-interpenetrating crosslinked PVDF-HFP poly(ethylene glycol) dimethacrylate (PVDF-HFP/PEG-DMA) have recently been found to provide a better performance—particularly in terms of their long-term stability [4–8]. The majority of the published research on PIMs has been focused on the extraction and transport of metallic cations and anions and small organic molecules (e.g., [9–11]). However, recent studies have demonstrated that PIMs are having an increasing role in chemical analysis techniques involving separation and sensing [12] and in the manufacturing of layers of metallic nanoparticles on membrane surfaces [13–17]. Successful PIMs are described as those that show good compatibility between the membrane components and the extracted complex or ion-pair of the target chemical species. Such membranes appear transparent and homogeneous to the naked eye, and are mechanically strong [18]. This evaluation of PIMs is useful in their practical application, but does not provide an insight into their morphology at micrometer- and nanometer-size scales. A commonly held view is that the liquid phase in a PIM is located between the entangled polymer chains in a network of nanometer-size

Membranes 2018, 8, 6; doi:10.3390/membranes8010006

www.mdpi.com/journal/membranes

Membranes 2018, 8, 6

2 of 12

channels, whereas the liquid phase in a supported liquid membrane (SLM)—the most popular type of liquid membranes at present—is held only by capillary forces within the micrometer-size pores of an inert polymeric membrane. It is suggested that this is the reason why PIMs are generally more stable than SLMs. Several advanced material characterization techniques (e.g., scanning electron microscopy (SEM), atomic force microscopy (AFM), X-ray diffraction (XRD), and Fourier transform infrared spectroscopy (FTIR)) were used in studies aimed at clarifying the morphology of PIMs [1,3]. Arous et al. [19] reported that the SEM images of a pure CTA-based membrane showed a highly porous matrix, but the pores vanished and a dense membrane was formed when 2-nitrophenol octyl ether (NPOE) was added as a plasticizer to the membrane composition. On the other hand, Xu et al. [20] reported that dense films were formed with no apparent porosity in PVC-based PIMs with a low concentration of Aliquat 336 as the carrier. As the concentration of Aliquat 336 was increased, the authors reported that a porous membrane structure with irregularly-shaped pores and pore sizes was observed. A more recent study by St John et al. [21] using synchrotron-based FTIR microspectrometry showed that PVC-based PIMs containing Aliquat 336 were chemically homogeneous at the micrometer-size scale. The transport mechanism in PIMs is also still uncertain, and both the “mobile carrier” and the “fixed site” models have been proposed [1–3]. The “mobile carrier” model suggests that PIMs consist of liquid-filled nanometer-size channels, which are connected to extraction sites on the membrane surface, and that the extracted species diffuse through these surface extraction sites into the bulk of the PIM. However, no direct experimental evidence has yet been provided to support the existence of such extraction sites. In a previous study, we reported on the use of a PVC-based PIM containing Aliquat 336 as a template for the preparation of surface-confined gold nanoparticles (AuNPs) [15]. The preparation process involved the extraction of Au(III) from an HCl solution and the subsequent reduction of the extracted Au(III) at the membrane surface with a solution of disodium ethylenediaminetetraacetate (Na2 EDTA). When prepared under the appropriate conditions, the AuNPs provided maximum coverage of the PIM surface and the membrane became incapable of extracting chemical species. However, this phenomenon was not studied in detail in the previous work [15]. This raised the interesting question of whether the AuNPs were formed at extraction sites on the PIM surface and the PIM became extraction inactive due to the blocking of these sites. The present study was aimed at confirming the existence of such extraction sites and determining the “critical AuNP mass” at which all surface extraction sites were blocked by AuNPs. 2. Materials and Methods 2.1. Chemicals All chemicals were used as received. Deionized water (18 MΩ cm, Millipore, Synergy 185, Molsheim, France) was used in the preparation of all aqueous solutions. Aliquat 336 (Aldrich, a mixture of quaternary alkylammonium chlorides), high molecular weight powdered PVC (Fluka), 1-dodecanol (DD) (Aldrich), NPOE (Fluka), and tetrahydrofuran (THF) (Chem-supply) were used as received. Au(III) calibration standards were made from a 1000 mol·L−1 Au(III) stock solution (BDH Spectrosol). Au(III) solutions for membrane extraction were prepared from HAuCl4 (Aldrich) dissolved in 2.5 M HCl (Chem-supply) solution. EDTA solutions of concentration 0.10 mol·L−1 were prepared from Na2 EDTA (Fison), and the pH was adjusted to pH 6.0 with 1.0 M NaOH solution (Chem-supply) [15]. Nitrate solutions (100 mg·L−1 ) were prepared by dissolving KNO3 (Asia Pacific Specialty Chemicals) in deionized water.


3 of 12

2.2. Instrumentation The concentration of Au(III) was determined by atomic absorption spectrometry (AAS) (Z-2000 Series Polarized Zeeman spectrometer, Hitachi, Tokyo, Japan). The extraction experiments were carried out by continuously shaking the nitrate solutions with a PIM immersed in each one of them. A platform orbital shaker (OM6, Ratek, Melbourne, Australia) was used in these experiments. The concentration of NO3 − was determined by ion chromatography (IC) (DX-120 ion chromatograph, Dionex, Sunnyvale, CA, USA) with the following experimental conditions: eluent—4.8 mM Na2 CO3 , 0.6 mM NaHCO3 ; flow rate—1.2 mL min−1 ; and sample loop size—25 µL. A scanning electron microscope (Quanta 200 F, FEI, Zurich, Switzerland) was used for membrane imaging. Measurements were carried out at 20 kV in high vacuum. The resolution of this instrument as stated by the manufacturer is between 1.2 nm to 3.0 nm at 30 kV and 1 kV, respectively. 2.3. Membrane Preparation A mixture of Aliquat 336, PVC, and in some cases a plasticizer (NPOE or DD) with a total mass of 400 mg was dissolved in a small volume of THF (8–10 mL), and the mixture was poured into a glass ring (diameter—7 cm) positioned on a flat glass plate. The mixture was covered with a filter paper and a watch glass to allow slow evaporation of the THF over 24 h to give a visually transparent, homogeneous, and flexible circular membrane. The membrane was then removed from the glass plate and cut with a metal cutter (diameter 3.5 mm). The cut edge was trimmed to give an average mass and thickness of 60 ± 3 mg and 50 ± 5 µm, respectively. 2.4. Preparation of AuNPs on the Surface of PIMs The AuNP-coated PIMs were prepared by firstly immersing the membranes in individual flasks containing 100 mL of 100 mg·L−1 Au(III) (present as [AuCl4 ]− ) and 2.5 M HCl and shaking them on a platform orbital shaker (150 rpm) for a predetermined period of time. Samples of the Au(III) solution (0.20 mL) were removed at the start and the end of the extraction period. The samples were diluted to 4 mL with deionized water, and the Au(III) concentration was determined by AAS. The Au(III)-loaded membranes were then rinsed with 5 mL of deionized water and dried before immersing them into 100 mL of 0.10 M EDTA solution at pH 6.0. The solutions were shaken on the platform orbital shaker for 24 h to reduce Au(III) to AuNPs on the surface of the PIMs. 2.5. Extraction of Nitrate The AuNP-coated membranes and PIMs without AuNPs were immersed individually in 100 mL of 100 mol·L−1 NO3 − solutions in flasks which were shaken (150 rpm) on a platform orbital shaker. Samples of the NO3 − solution (1 mL) were removed at predetermined times throughout the course of the extraction experiment. The samples were diluted to 6 mL with deionized water, and the NO3 − concentration was determined by IC. 2.6. Recovery of Au from the AuNP-Coated PIMs A centrifuge tube was dried in an oven at 35 ◦ C overnight, and its mass was recorded. An AuNP-coated PIM was placed in it, and this was followed by the addition of 1.5 mL of THF. The tube was shaken and sonicated for 10 min in an ultrasonic bath until the membrane had dissolved, and the tube was centrifuged for 5 min at 8000 rpm to ensure that all AuNPs settled at the bottom of the tube. THF was then removed from the centrifuge tube with a pipette, and the precipitate was rinsed twice with 1 mL of THF to remove any traces of membrane material. The precipitate in the centrifuge tube was dried in an oven at 35 ◦ C overnight, and the tube was reweighed to obtain the mass of metallic gold.


4 of 12

Membranes 2018, 8, x FOR PEER REVIEW

4 of 12

2.7. Initial Flux Calculation 2.7. Initial Flux Calculation The calculation of the initial flux (J0 , mol·m−2 ·s−1 ) was made according to Fick’s first law by The calculation of the initial flux (J0, mol·m−2·s−1) was made according to Fick’s first−law by fitting fitting the transient concentration with an exponential decay function (C = a1 + a2 e a3 t ), the first the transient concentration with an exponential decay function (𝐶̅ = 𝑎1 + 𝑎2 𝑒 −𝑎3𝑡 ), the first derivative derivative of which ((dC/dt)t=0 ) was used to calculate J0 according to Equation (1) [22]. of which ((𝑑𝐶̅ ⁄𝑑𝑡)𝑡=0 ) was used to calculate J0 according to Equation (1) [22]. V dC J0𝐽 == 𝑉·∙ [𝑑𝐶 ] (1) (1) 0 S𝑆 dt 𝑑𝑡 t=0 𝑡=0

where V V is volume (m (m33),), SS is where is the the solution solution volume is the the PIM PIM surface surface area area (m (m22),), and and tt is is time time (s). (s). 3. Results and Discussion 3.1. Formation of of AuNPs AuNPs on on the the PIM PIM Surface Surface 3.1. Formation In aa previous In previous study study [15], [15], we we demonstrated demonstrated that that surface-confined surface-confined AuNPs AuNPs on on aa PVC-based PVC-based PIM PIM could be produced by aa straightforward procedure. In In it, it, aa PIM PIM (20 (20 wt wt % % Aliquat Aliquat 336, 336, 10 10 wt wt % % DD, DD, could be produced by straightforward procedure. −1 was firstly immersed in a 2.5 M −1 and 70 wt % PVC) with an ion-exchange capacity of 0.40 meq g and 70 wt % PVC) with an ion-exchange capacity of 0.40 meq g was firstly immersed in a 2.5 M HCl HCl solution of Au(III). Au(III) extracted into thePIM PIMasas[AuCl [AuCl4]4−]−and andthe theamount amountextracted extracted was was solution of Au(III). Au(III) waswas extracted into the determined by the the experimental conditions; however, however, complete of the determined by experimental conditions; complete loading loading of the PIM PIM with with Au(III) Au(III) can can be easily obtained (Figure S1, Supplementary Material). The Au(III)-loaded PIM was then immersed be easily obtained (Figure S1, Supplementary Material). The Au(III)-loaded PIM was then immersed in in 1.0 1.0 M M EDTA EDTA solution solution at at pH pH 66 for for 24 24 h, h, producing producing the the surface-confined surface-confined AuNPs. AuNPs. The The SEM SEM image image (Figure 1a)of ofthe thesurface surface AuNP-coated maximum Au(III) loading and prepared (Figure 1a) of of an an AuNP-coated PIMPIM withwith maximum Au(III) loading and prepared under under the optimum conditions [15] shows that AuNPs were present on the surface of the PIM, where the optimum conditions [15] shows that AuNPs were present on the surface of the PIM, where depending ontheir theirsurface surfacedensity densitythey they can aggregate to form clusters. The SEM the depending on can aggregate to form clusters. The SEM imageimage of the of crosscross-section (Figure 1b) revealed the absence of AuNPs within the bulk of the membrane. section (Figure 1b) revealed the absence of AuNPs within the bulk of the membrane.

(a)

(b)

Figure 1. SEM SEM images images showing showing (a) (a) surface-confined surface-confined gold gold nanoparticle nanoparticle (AuNP) (AuNP) clusters clusters at at maximum maximum Figure 1. Au(III) loading and (b) the cross-section of a polymer inclusion membrane (PIM) coated with Au(III) loading and (b) the cross-section of a polymer inclusion membrane (PIM) coated with AuNPs AuNPs under the same same conditions conditions which which is is virtually virtually free free of of AuNPs. AuNPs. The The white white bars bars are are equal equal to to 10 10 µm. μm. under the

It was expected that the surface-confined AuNPs and clusters of those were likely to affect the It was expected that the surface-confined AuNPs and clusters of those were likely to affect the extraction properties of the corresponding PIMs. This effect was experimentally studied using the extraction properties of the corresponding PIMs. This effect was experimentally studied using the nitrate ion as the extracted chemical species because Aliquat 336 had been shown to exhibit a nitrate ion as the extracted chemical species because Aliquat 336 had been shown to exhibit a relatively relatively high affinity for this ion [23,24]. high affinity for this ion [23,24]. 3.2. Extraction of NO3− Using an AuNP-Coated PIM An AuNP coated PIM which was fully loaded with Au(III) prior to its reduction with EDTA was found to be incapable of extracting NO3−, while a PIM with the same composition but without AuNPs extracted NO3− as expected (Figure 2).

In order to explain this phenomenon, it is proposed that the PIM surface was not homogeneous and contained membrane liquid phase located at extraction sites, while the remainder of the PIM surface was free of the liquid phase. The inability of the PIM coated with AuNPs to extract NO3− is due to blocking of these extraction sites by the AuNPs or their clusters, thus preventing the transfer of NO3− ions to the bulk of the membrane. If this were the case, then the extent of coverage of the PIM Membranes 2018, 8, 6 of 12 surface by the AuNPs is crucial, and by lowering the extent of coverage, some of the extraction5 sites would be open and extraction of NO3− would take place. This line of reasoning suggests that there should be a minimal surface (which can Usingcoverage an AuNP-Coated PIMbe referred to as the “critical AuNP mass”), at which 3.2. Extraction of NO3 − the membrane ceases to extract because of complete blockage of its surface extraction sites. Thus, a Anwas AuNP coated PIM which was fully loadedAuNP with Au(III) priortotoinvestigate its reduction EDTA was study conducted to determine this “critical mass” and itswith dependence on − , while a PIM with the same composition but without AuNPs found to be incapable of extracting NO 3 the membrane composition. extracted NO3 − as expected (Figure 2).

Figure 2.2. The The extraction extraction of of NO NO33– using an AuNP-coated PIM, fully fully loaded loaded with with Au(III) Au(III) prior prior to to Figure reduction ((●), and using using aaPIM PIMofofthe thesame samecomposition compositionbut butwithout withoutAuNPs AuNPs(#). (○). Experimental Experimental reduction ), and conditions:solution solutionvolume volumeand andcomposition: composition:100 100mL, mL,100 100 mg· NO33−−;; PIM composition: conditions: mg ·L−L1−1NO PIM mass mass and composition: 60 60 ± ± 3 mg, 20 wt % Aliquat 336, 10 wt wt % % 1-dodecanol 1-dodecanol (DD) (DD) and and 70 70 wt wt % % poly(vinyl poly(vinyl chloride) chloride) (PVC); (PVC); shaking rate: 150 rpm. Data points are the average of three extraction experiments with an average shaking rate: 150 rpm. Data points are the average of three extraction experiments with an average standard ·LL−−11.. standarddeviation deviation(SD) (SD)of of0.77 0.77mg mg·

In order ordertotoexplain determine the “critical AuNP mass” ofthat a PIM composition, PIMs the same In this phenomenon, it is proposed the PIM surface was not with homogeneous composition were loaded with different amounts of Au(III) before reduction with EDTA. was and contained membrane liquid phase located at extraction sites, while the remainder of This the PIM carried was out free by varying the time during which PIMs were in AuNPs the corresponding Au(III) − surface of the liquid phase. The inability of the PIMimmersed coated with to extract NO 3 is solutions. The membranes were then tested for their ability to extract NO 3– and the results are shown due to blocking of these extraction sites by the AuNPs or their clusters, thus preventing the transfer in NO Figure − 3. The mass of the AuNPs collected on the surface of the membrane was calculated by the of 3 ions to the bulk of the membrane. If this were the case, then the extent of coverage of the method described Section The and PIM by which was not AuNPs extracted an extraction amount of PIM surface by the in AuNPs is 2.6. crucial, lowering thecoated extentwith of coverage, some of the −1. However, as the amount of AuNPs on NO 3− equivalent to its ion-exchange capacity of 0.40 meq g − sites would be open and extraction of NO3 would take place. This line of reasoning suggests that the membrane increased, amount(which of NO3−can extracted into the PIMs until the PIM there should besurface a minimal surfacethe coverage be referred to as the decreased “critical AuNP mass”), became unable to extract nitrate. The critical mass of AuNPs to completely block the PIM surface was at which the membrane ceases to extract because of complete blockage of its surface extraction sites. found to be 4.54 mg, which occurred at Au(III) extraction (loading) times equal to or greater than Thus, a study was conducted to determine this “critical AuNP mass” and to investigate its dependence 12 the h. membrane composition. on In order to determine the “critical AuNP mass” of a PIM composition, PIMs with the same composition were loaded with different amounts of Au(III) before reduction with EDTA. This was carried out by varying the time during which PIMs were immersed in the corresponding Au(III) solutions. The membranes were then tested for their ability to extract NO3 – and the results are shown in Figure 3. The mass of the AuNPs collected on the surface of the membrane was calculated by the method described in Section 2.6. The PIM which was not coated with AuNPs extracted an amount of NO3 − equivalent to its ion-exchange capacity of 0.40 meq g−1 . However, as the amount of AuNPs on the membrane surface increased, the amount of NO3 − extracted into the PIMs decreased until the PIM became unable to extract nitrate. The critical mass of AuNPs to completely block the PIM surface was found to be 4.54 mg, which occurred at Au(III) extraction (loading) times equal to or greater than 12 h.

Membranes 2018, 8, 6x FOR PEER REVIEW Membranes 2018, 8, 8, Membranes 2018, x FOR PEER REVIEW

of 12 666of of 12 12

6 of 12

Figure 3. The extraction of NO3−− by PIM with different AuNP masses and extraction times (∎, 0.00 − the of of NO 3 by AuNPAuNP massesmasses and extraction times (∎, 0.00 Figure 3. The Theextraction extraction NO byPIM the with PIM different with different and extraction times 3 the mg/0 h; , 3.57 mg/4 h; ●, 4.19 mg/8 h; □, 4.54 mg/12 h; △, 4.54/12.5 h; ○, 4.54 mg/79 h). Experimental sses and extraction times ((∎, mg/0 h; mg/0 , 3.57h;mg/4 h; ●, 4.19 h; mg/8 h; □,mg/8 4.54 mg/12 h; △,mg/12 4.54/12.5 h;, ○, 4.54 mg/79 h).4.54 Experimental , 0.00 N, 3.57 mg/4 , 4.19 h; , 4.54 h; 4 4.54/12.5 h; #, mg/79 h). conditions: solution volume and composition, 100 mL, 100 mol·L−1−1 NO3−−; PIM mass and − composition, h; ○, 4.54 mg/79 h). Experimental conditions: solution volume and composition, mL, 100 mol· L mL, NO100 3 ; PIM and3composition, Experimental conditions: solution volume and100 composition, 100 molmass ·L−1 NO ; PIM mass 60 ± 3 mg, 20 wt % Aliquat 336, 10 wt % DD, and 70 wt % PVC; shaking rate: 150 rpm. Data points are NO3−; PIM mass and composition, 60 ± composition, 3 mg, 20 wt %60Aliquat 10 % wtAliquat % DD, 336, and 10 70 wt wt % % DD, PVC;and shaking 150shaking rpm. Data are and ± 3 mg,336, 20 wt 70 wt rate: % PVC; rate:points 150 rpm. −1 the average of three extraction experiments with an average SD of 0.69 mg·L−1 . ng rate: 150 rpm. Data points are Data points are the average of three extraction experiments an0.69 average the average of three extraction experiments with an averagewith SD of mg·LSD. of 0.69 mg·L−1 . 69 mg·L−1.

The SEM SEM images images of of the the PIM PIM surface surface for for 8, 8, 12, 12, 12.5, 12.5, and and 79 79 hh of of extraction extraction (loading) (loading) are are shown shown in in The The SEM images of the PIM surface for 8, 12, 12.5, and 79 h of extraction (loading) are shown in Figure 4. extraction (loading) are 4. shown in Figure 4. Figure

Figure 4. SEM images of the surface of the PIM corresponding to Au(III) extraction times of (a) 8 h; Figure 4. SEM images of the surface of the PIM corresponding to Au(III) extraction times of (a) 8−h; Figure 4. SEM images of the thePIM PIMloaded corresponding to Au(III) times (a)·L 8 h;1 (b) 12 h; (c) 12.5 h; and (d)surface 79 h. of The with Au(III) was extraction exposed to 0.10ofmol (b) 12 h; (c) 12.5 h; and (d) 79 h. The PIM loaded with Au(III) was exposed to 0.10 mol·L−1−1 u(III) extraction times of ethylenediaminetetraacetate (a) 812 h; h; (b) (c) 12.5 h; and (d) 79 h. The PIM loaded with Au(III) was exposed to 0.10 mol· L (EDTA) at pH 6.0 for 24 h. The white bars are equal to 10 µm. ethylenediaminetetraacetate (EDTA) at pH 6.0 for 24 h. The white bars are equal to 10 μm. −1 I) was exposed to 0.10 ethylenediaminetetraacetate mol·L (EDTA) at pH 6.0 for 24 h. The white bars are equal to 10 μm. ars are equal to 10 μm.

The surface coverage of the PIM surface for 8 hfor (Figure(Figure 4a) is clearly Thelower lowerAuNP AuNP surface coverage of the the PIM PIM surface 4a) is isevident, clearly whereas evident, The lower AuNP surface coverage of surface for 88 hh (Figure 4a) clearly evident, for the other times the extent of coverage appears to be approximately the same and the mass of the whereas for the the other other times times the the extent extent of of coverage coverage appears appears to to be be approximately approximately the the same same and and the 8 h (Figure 4a) is whereas clearly evident, for the AuNPs was the same (4.54 mg; i.e., “critical AuNP mass” for this PIM composition). mass of the AuNPs was the same (4.54 mg; i.e., “critical AuNP mass” for this PIM composition). approximately the same andAuNPs the mass of the was the same (4.54 mg; i.e., “critical AuNP mass” for this PIM composition). ss” for this PIM composition). 3.3. Quantitative Production of AuNPs on the PIM Surface 3.3. Quantitative Quantitative Production Production of of AuNPs AuNPs on on the the PIM PIM Surface Surface 3.3. It was of interest to determine if the extracted Au(III) was quantitatively converted into AuNPs It was was of of interest interest to to determine determine if if the the extracted extracted Au(III) Au(III) was was quantitatively quantitatively converted into into AuNPs at theItPIM surface as a result of the EDTA-based reduction process. This wasconverted carried out byAuNPs using at the the PIM surface as as aa result result of of the the EDTA-based EDTA-based reduction reduction process. process. This This was was carried carried out out by by using using the the antitatively converted into AuNPs at PIM surface the procedure described in Section 2.6. The results were then compared with the amount of Au(III) procedure described in Section 2.6. The results were then compared with the amount of Au(III) . This was carried extracted out by using the procedure described Section The results were then compared withbythe amount of Au(III) into the PIMinduring the2.6. Au(III) extraction process as determined AAS. The possibility extracted into the PIM during the Au(III) extraction process as determined by AAS. The possibility pared with the amount of Au(III) extracted into theof PIM duringAu(III) the Au(III) extraction process as determined AAS. The possibility of the reduction residual in the membrane by THF itself was by checked by dissolving ofThe thepossibility reduction of of residual residual Au(III) Au(III) in in the the membrane membrane by by THF THF itself itself was was checked checked by by dissolving dissolving Au(III)Au(III)termined by AAS.Au(III)-loaded of the reduction PIMs in THF. It was found that no metallic gold was produced in this way. loaded PIMs in THF. It was found that no metallic gold was produced in this way. was checked by dissolving Au(III)loaded in THF. It of was foundextracted that no metallic was PIMs produced thisSD way. ThePIMs average mass Au(III) into 10 gold identical (4.54inmg, of 0.15 mg, and 95% The average mass of Au(III) extracted into 10 identical PIMs (4.54 mg, SD of 0.15 0.15 mg, mg, and and 95% 95% uced in this way. confidence The average mass of Au(III) extracted into 10 identical PIMs (4.54 mg, SD of interval of 4.43–4.64 mg) agreed very closely with the average mass of metallic Au collected confidence interval of 4.43–4.64 mg) agreed very closely with the average mass of metallic Au (4.54 mg, SD of 0.15 mg, and interval 95% confidence of 4.43–4.64 mg) agreed closely with theand average mass of metallic by dissolving the same 10 PIMs in THF (4.52 very mg, SD of 0.18 mg, 95% confidence intervalAu of collected by dissolving the same 10 PIMs in THF (4.52 mg, SD of 0.18 mg, and 95% confidence interval the average mass4.39–4.65 of metallic AuThere was collected by dissolving the no same 10 PIMs insignificant THF (4.52 difference mg, SD of 0.18 mg, and interval mg). statistically between the 95% two confidence average masses at of 4.39–4.65 4.39–4.65 mg). There There was was no no statistically statistically significant significant difference difference between between the the two two average average masses masses at at 8 mg, and 95% confidence interval of mg). the 95% confidence level. These results confirmed that under the experimental conditions used, the 95% confidence level. These results confirmed that under the experimental conditions used, all etween the two average masses at the 95% confidence level. These results confirmed that under the experimental conditions used, all

experimental conditions used, all

Membranes 2018, 8, x FOR PEER REVIEW Membranes 2018, 8, 6

7 of 12 7 of 12

extracted Au(III) was reduced to AuNPs on the membrane surface. This result indicated that all Aliquat 336 ion-pairs containing 4]− as on thethe anion in the bulk of the PIM prior to the reduction all extracted Au(III) was reduced[AuCl to AuNPs membrane surface. This result indicated that all − step were 336 converted back to their original form,inand therefore Aliquat ion-pairs containing [AuCl4 ] chloride as the anion the were bulk of the PIMpotentially prior to theavailable reductionfor extracting 3− [23,24]. step wereNO converted back to their original chloride form, and were therefore potentially available for extracting NO3 − [23,24].

3.4. Effect of the Aliquat 336 Concentration on the “Critical AuNP Mass” 3.4. Effect of the Aliquat 336 Concentration on the “Critical AuNP Mass”

PIMs with 20, 25, 30, and 35 wt % Aliquat 336 but without an additional plasticizer were studied PIMs with 25, 30, wt % Aliquat 336 but without an additional plasticizer wereAs studied to to determine the20, effect of and the 35 Aliquat 336 concentration on the “critical AuNP mass”. expected determine the effect of the Aliquat 336 concentration on the “critical AuNP mass”. As expected [15,25–27], [15,25–27], the extent and rate of Au(III) extraction increased with increasing the concentration of the extent rate of Au(III) increased with increasing the concentration of Aliquat 336 in Aliquat 336 and in the PIM (Figureextraction S2, Supplementary Material). the PIM (Figure S2, Supplementary Material). After extraction, each PIM was immersed in an EDTA solution to produce AuNPs and then After extraction, each PIM was immersed in an EDTA solution to produce AuNPs and then tested tested for its ability to extract NO 3−. The NO3− extraction curves which also show the mass of metallic − for its ability to extract NO3 . The NO3 − extraction curves which also show the mass of metallic gold recovered from the membranes after its dissolution in THF are presented in the Supplementary gold recovered from the membranes after its dissolution in THF are presented in the Supplementary Material (Figures S3–S6). The “critical AuNP mass” with the corresponding initial Au(III) flux value Material (Figures S3–S6). The “critical AuNP mass” with the corresponding initial Au(III) flux value and “critical” Au(III) extraction (loading) time for each membrane composition are summarized in and “critical” Au(III) extraction (loading) time for each membrane composition are summarized in − Table 1.1.InIneach “critical AuNP AuNPmass” mass”was wasreached. reached. Table Table eachcase, case,extraction extraction of of NO NO33 −ceased ceasedonce once the the “critical Table 1 1 shows very clearly of Aliquat Aliquat336 336ininthe thePIM PIMled ledtotoanan increase shows very clearlythat thatan anincrease increasein in the the amount amount of increase in in thethe “critical AuNP mass”. “critical AuNP mass”. Table “CriticalAuNP AuNPmasses”, masses”,initial initial Au(III) Au(III) flux flux values, (loading) Table 1.1.“Critical values,and and“critical” “critical”Au(III) Au(III)extraction extraction (loading) times PIMsfor fordifferent differentconcentrations concentrations of times ofof PIMs of Aliquat Aliquat336. 336.

Aliquat Aliquat 336 Concentration (wt %) Concentration %)

Initial for Au(III) Initial FluxFlux for Au(III) Extraction −2 ·s−1−2) −1 (J0 ) (mol Extraction (J0)·m (mol ·m ·s )

Critical Critical Au(III) Au(III) Extraction Time Time (h) Extraction (h)

CriticalAuNP AuNP Critical Mass Mass(mg) (mg)

20 20 25 25 30 30 35 35

3.12 × × 10 10−8−8 1.56 × 10 × 10−7−7 5.20 × 10−7−7 5.20 × 10 2.60 × 10−6−6 2.60 × 10

2.5 2.5 3.5 3.5 6.5 6.5 7.5 7.5

0.16 0.16 0.45 0.45 3.45 3.45 5.44 5.44

The SEMimages imagesofofthe thesurfaces surfacesof of the the PIMs PIMs listed 5. 5. The increase The SEM listed in in Table Table11are areshown shownininFigure Figure The increase numberofofAuNP AuNPclusters clusterswith with the the increase increase in 336 is evident. in in thethe number inthe theconcentration concentrationofofAliquat Aliquat 336 is evident.

Figure with aa “critical “criticalAuNP AuNPmass” mass”forfor following Figure5.5.SEM SEMimages imagesofofthe thesurfaces surfaces of of the the PIMs PIMs with thethe following concentrations 30; and and (d) (d)35 35wt wt%. %.The Thewhite whitebars barsare are equal μm. concentrationsofofAliquat Aliquat336: 336:(a) (a)20; 20; (b) (b) 25; 25; (c) (c) 30; equal to to 10 10 µm.

It It can the “critical “criticalAuNPs AuNPsmass” mass” is related to the number of extraction sites canbebeexpected expectedthat, that, if if the is related to the number of extraction sites on onthe thePIM PIMsurface, surface, then PIM compositions produce a higher “critical mass” willfaster exhibit then PIM compositions that that produce a higher “critical AuNP AuNP mass” will exhibit faster extraction of Au(III); i.e., higher flux values. This is certainly befor thethe case extraction of Au(III); i.e., higher initial initial Au(III)Au(III) flux values. This is certainly found to found be the to case forfour the membrane four membrane compositions studied, as shown in Table 1. compositions studied, as shown in Table 1. 3.5. Effect DDand andNPOE NPOE 3.5. Effect of of DD ourprevious previousstudies studies[15,27], [15,27], we we optimized optimized the thethe most InInour the PIM PIMcomposition compositionand andfound foundthat that most efficient membranefor forthe theextraction extractionof of Au(III) Au(III) was was obtained efficient membrane obtainedby byadding addingDD DDtotothe thePIM PIMcomposition. composition. The research described in Sections 3.1–3.3 involved the use of this PIM. Another common plasticizer used in PIM compositions is NPOE [1–3], and it was of interest to examine the effect of the concentrations of these two plasticizers on the “critical AuNP mass”. Plasticizers are generally


8 of 12

The research described in Sections 3.1–3.3 involved the use of this PIM. Another common plasticizer used in PIM compositions is NPOE [1–3], and it was of interest to examine the effect of the concentrations of these two plasticizers on the “critical AuNP mass”. Plasticizers are generally employed to improve the compatibility of the membrane components and to improve the extraction rate and extraction efficiency [1–3]. In this study, the concentration of Aliquat 336 in the PIMs was kept constant at 20 wt % to be consistent with the previous studies [15,27], while the concentrations of DD or NPOE were varied to provide PIMs containing 0, 5, 10, and 15 wt % of plasticizer with the concentration of PVC being varied accordingly. The PIMs were first loaded with Au(III) as described before. As reported previously [15], the extraction was faster for the PIMs containing DD, and these PIMs reached equilibrium with the solution after several hours. Additionally, there was little difference in the initial flux for 10 and 15 wt % DD. The extraction curves are presented in the Supplementary Material (Figures S7 and S8). After Au(III) loading, the PIMs were treated with an EDTA solution to produce surface-confined AuNPs. The AuNP coated PIMs were then used in NO3 – extraction experiments (Figures S9–S13, Supplementary Material) to determine the “critical” Au(III) loading time and the corresponding “critical AuNP mass”. The data obtained are presented in Table 2. Table 2. “Critical AuNP mass”, initial Au(III) flux, and “critical” Au(III) extraction (loading) time for PIMs containing 20 wt % Aliquat 336 and different concentrations of DD or NPOE. DD/NPOE Concentration (wt %)

Initial Flux for Au(III) Extraction (J0 ) (mol·m−2 ·s−1 )

Critical Au(III) Extraction Time (h)

Critical AuNP Mass (mg)

0

3.12 × 10−8

2.5

0.16

DD 5 10 15

4.68 × 10−7 2.08 × 10−6 2.08 × 10−6

8 12 12

3.05 4.54 4.57

NPOE 5 10 15

2.60 × 10−7 3.64 × 10−7 5.20 × 10−7

4.5 4.5 4.5

0.88 2.02 2.84

It can be seen that the “critical AuNP mass”, initial Au(III) flux, and “critical” Au(III) extraction time increased with increasing the concentrations of DD and NPOE, with DD producing a considerably higher “critical AuNP mass” value than NPOE. Additionally, the “critical AuNP masses” for both plasticizers was considerably higher than that for an un-plasticized PIM, thus suggesting the presence of more available surface extraction sites, which is consistent with the faster extraction of Au(III) by plasticized PIMs compared to un-plasticized PIMs [27]. Membranes containing concentrations of DD or NPOE higher than 15 wt % became mechanically weak and unstable, and often had an oily surface. The SEM surface images in Figure 6 of the PIMs listed in Table 2 clearly show these “critical AuNP mass” trends.


9 of 12


9 of 12

Figure 6. SEM SEM surface surfaceimages imagesofofthe thePIMs PIMs listed Table 2. Aliquat 20%; wtDD, %; (a) DD, (b) 10; 5; Figure 6. listed in in Table 2. Aliquat 336,336, 20 wt 0; (a) (b) 0; 5; (c) (c) 10; (d) 15 wt %; 2-nitrophenol octyl ether (NPOE); (e) 0; (f) 5; (g) 10; and (h) 15 wt %. The white (d) 15 wt %; 2-nitrophenol octyl ether (NPOE); (e) 0; (f) 5; (g) 10; and (h) 15 wt %. The white bars are bars equalare to equal 10 µm.to 10 μm.

4. Conclusions 4. Conclusions The research reported in this paper has produced a number of interesting observations which The research reported in this paper has produced a number of interesting observations which are consistent with the surface of PIMs studied incorporating an array of sites where extraction can are consistent with the surface of PIMs studied incorporating an array of sites where extraction can only occur. only occur. The experimental observations made in this research are the following: The experimental observations made in this research are the following:  Individual AuNPs that aggregate into clusters are formed on the surface of the PIMs after the • Individual AuNPs that aggregate into clusters are formed on the surface of the PIMs after the extraction of Au(III) and its subsequent reduction with EDTA. extraction of Au(III) and its subsequent reduction with EDTA.  At a critical surface mass of the AuNPs, the PIM loses its ability to extract NO3−−, which is • consistent At a critical surface mass the AuNPs, PIM losesblocking its ability extract NO , which is with AuNPs andof clusters of thosethe completely the to extraction sites3 on the PIM consistent with AuNPs and clusters of those completely blocking the extraction sites on the PIM surface. At AuNP masses lower than the corresponding critical values, some sites are still surface. Atfor AuNP lowerofthan corresponding critical values, some sites are still available available the masses extraction NOthe 3−, but in such cases, the rate of extraction is reduced − for the extraction of NO3 , but in such cases, the rate of extraction is reduced accordingly. accordingly. The mass mass of of AuNPs AuNPscollected collectedfrom fromPIMs PIMsafter afterdissolution dissolutioninin THF equates exactly mass •  The THF equates exactly to to thethe mass of of Au(III) originally extracted. This demonstrates that all Au(III) extracted has been reduced to Au(III) originally extracted. This demonstrates that all Au(III) extracted has been reduced to AuNPs on the PIM surface and the bulk of the PIM contains free Aliquat 336. AuNPs on the PIM surface and the bulk of the PIM contains free Aliquat 336. The “critical “critical AuNP AuNP mass”, mass”, and and hence hence the the population population of of extraction extraction sites, sites, is is directly directly related related to to the the • The PIM composition. Higher concentrations of Aliquat 336 result in higher Au(III) fluxes during PIM composition. Higher concentrations of Aliquat 336 result in higher Au(III) fluxes during Au(III) extraction higher “critical AuNP mass”mass” values.values. Additionally, the addition increasing Au(III) extractionand and higher “critical AuNP Additionally, the of addition of concentrations of DD or NPOE to the PIM formulation produces higher Au(III) fluxes and “critical increasing concentrations of DD or NPOE to the PIM formulation produces higher Au(III) fluxes AuNP mass”AuNP values.mass” values. and “critical PIMs appear appear to to be homogeneous homogeneous [21]. [21]. As mentioned mentioned in the On the micrometer-size scale, PIMs introduction, aanumber numberofofpapers papershave have suggested that a “pore” structure exists in PIMs; however, introduction, suggested that a “pore” structure exists in PIMs; however, the the resolution associated with most of the instrumental methods employed in studying PIM resolution associated with most of the instrumental methods employed in studying PIM morphology morphology is not to high enough to define a “pore” structure in the nanometer-size It could be is not high enough define a “pore” structure in the nanometer-size range. It couldrange. be hypothesized hypothesized that a PIM isby characterized by a tortuous channel-like structure and that end theseinchannels that a PIM is characterized a tortuous channel-like structure and that these channels “pores” end “pores” surface ofHowever, the membrane. withoutthe direct evidence, the the hypothesis at theinsurface ofat thethe membrane. withoutHowever, direct evidence, hypothesis about internal about the of internal PIMs is highly though the paper research this paper on structure PIMs structure is highly of speculative, evenspeculative, though the even research in this oninthe membrane the membrane surface morphology lend some surface morphology does lend somedoes credence to it. credence to it. aboveconclusions conclusions observations also consistent with the following The above and and observations are alsoare consistent with the following experimental experimental evidence: evidence:

(1) PIMs are generally more resistant to leaching of the liquid membrane phase to the aqueous phase they are in contact with.



10 of 12

10 of 12

(1) PIMs are generally more resistant to leaching of the liquid membrane phase to the aqueous phase they are(2) in PIMs contact with. have lower rates of extraction and transport than SLMs, since the pores in SLMs are (2) PIMs havein lower rates of extraction and transport than SLMs, since the pores in SLMs are generally the micrometer-size range. generally in the micrometer-size range. Research is continuing on this approach with a view to refining the preparation of the surface Research is continuing on this approach with a view to refining the preparation of the surface confined AuNPs and to obtain a higher resolution examination of both the PIM surface and its confined AuNPs and to obtain a higher resolution examination of both the PIM surface and its internal internal structure using SEM, AFM, XRD, and synchrotron-based techniques. The ultimate aim is to structure using SEM, AFM, XRD, and synchrotron-based techniques. The ultimate aim is to elucidate elucidate the true nature of the PIM surface and its internal structure. the true nature of the PIM surface and its internal structure.

Supplementary Materials: The following are available online at www.mdpi.com/s1: Figures S1–S13. Figure S1: Supplementary Materials: The following are available online at www.mdpi.com/2077-0375/8/1/6/s1: Figures −1 HCl. Experimental conditions: solution volume and composition, 100 The extraction of Au(III) 2.5from mol·L S1–S13. Figure S1: The extraction of from Au(III) 2.5 mol ·L−1 HCl. Experimental conditions: solution volume and −1 −1 mL, 100 mol·L Au(III), 2.5 mol·L HCl; PIM 60 ± 3 mg, 20% % Aliquat − 1 − 1 composition, 100 mL, 100 mol·L Au(III), 2.5 mol·L mass HCl; and PIM composition, mass and composition, 60 ±wt 3 mg, 20% wt336, % 10 wt % 1-dodecanol, 70 wt % PVC; shaking 150 rpm. points arepoints the average 3 extraction experiments with Aliquat 336, 10 wt % 1-dodecanol, 70 wt % PVC;rate, shaking rate, Data 150 rpm. Data are the of average of 3 extraction . Figure TheS2: extraction of Au(III) from 2.5 mol·L−1 HCl an with average standard deviation (SD) of(SD) 0.77 of mol·L experiments an average standard deviation 0.77−1mol ·L−1 . S2: Figure The extraction of Au(III) from − 1 into PIMs containing Aliquat 336336 in inconcentrations: 25((■); 2.5 mol·L solutions HCl solutions into PIMs containing Aliquat concentrations:20 20 (); ( ); 25 ); 3030 (N(▲); ); andand 35 wt35%wt % (●) −mol·L 1 Au(III), −1 HCl; −1 Au(III), −1 HCl; PIM (Experimental conditions: solution volume and composition; 100 mL, 100 2.5 mol·L ( ) (Experimental conditions: solution volume and composition; 100 mL, 100 mol · L 2.5 mol · L x FOR PEER REVIEW 10 of 12 PIM mass:mass: 60 ±603 ±mg; shaking rate: 150 rpm). of 33 extraction extractionexperiments experimentswith withan average 3 mg; shaking rate: 150 rpm).Data Datapoints pointsare are the the average average of 1 . Figure S3: The extraction of −NO − with PIMs of different AuNP loadings and an averageSD SD 0.79 mol−1 ·L. − 3 PIMs of different AuNP loadings and extraction times. ofof 0.79 mol·L Figure S3: The since extraction NO3 inwith ave lower rates of extraction and transport than SLMs, the of pores SLMs are − extractionExperimental times. Experimental conditions: and composition; 100100 mL,mol·L 100 mol ·L−3−1; NO −1 NO 3 ;mass and PIM conditions: solutionsolution volume volume and composition; 100 mL, micrometer-size range. PIM mass and composition: 60 ± 3 mg, 20 wt % Aliquat 336 and 80 wt % PVC; shaking rate: 150 rpm. Data points composition: 60 ± 3 mg, 20 10 wtof%12Aliquat 336 and 80 wt % PVC; shaking rate: 150 rpm. Data points are the average are the average of 3 extraction experiments with an average SD of 0.60−1mol·L−1 . Figure S4: The extraction of s continuing on this approach with a view to refining the of0.60 themol·L surface of 3 extraction experiments with anpreparation average SD of . Figure S4: The extraction of NO3− with PIMs of NO3 − with PIMs of different AuNP loadings and extraction times.Experimental conditions: solution volume different AuNP loadings extraction solution volume and composition; 100 n SLMs, since the pores SLMs −of 1 are − the times.Experimental Psand andtransport to obtainthan aand higher resolution examination both PIM surface and conditions: its composition; 100 mL, 100inmol ·Land NO 3 ; PIM mass and composition: 60 ± 3 mg, 25 wt % Aliquat 336 and −1 NO3−; PIM mass and composition: 60 ± 3 mg, 25 wt % Aliquat 336 and 75 wt % PVC; shaking mL, 100 mol·L 75 wt % PVC; rate: 150 rpm.techniques. Data points The are the average of 3isextraction experiments with an average re using SEM, AFM, XRD, andshaking synchrotron-based ultimate aim to −1rpm. 150 DataS5: points the average of 33−extraction an loadings average SD 0.62 mol·L−1. Figure SD surface of 0.62rate: mol ·Lits .internal Figure The are extraction of NO with PIMsexperiments of different with AuNP andofextraction ue nature of the PIM and structure. h with a view to refining the preparation of the surface −with PIMs of different AuNP loadings and extraction − 1 − S5: The extraction of NO 3 times. Experimental times. Experimental conditions: solution volume and composition; 100 mL, 100 mol·L NO3 ; PIM mass andconditions: −1 NO esolution examination of both its 336 composition: 60the ±volume 3PIM mg, surface 30 wtcomposition; % and Aliquat and 70100 wt mol·L % PVC; shaking rate: 150 rpm. Data points 3−; PIM mass and composition: 60 are ± 3 the mg, 30 wt % solution 100 mL, Materials: The following are available online atand www.mdpi.com/s1: Figures S1–S13. Figure S1: −1 . Figure S6: The extraction of NO − average of 3 extraction experiments with an average SD of 0.62 mol · L nd synchrotron-based techniques. The ultimate aim is to Aliquat 336 and 70 wt % PVC; shaking rate: 150 rpm. Data points are the average of 3 extraction 3experiments −1 Au(III) from 2.5 mol·L HCl. Experimental conditions: solution volume and composition, 100 with PIMswith of different AuNP loadings and times. Experimental conditions: solution volume and −1. extraction an average SD of 0.62 mol·L Figure S6: The extraction of NO 3− with PIMs of different AuNP loadings and and its internal structure. −1 Au(III), 2.5 mol·L HCl; PIM mass100 andmL, composition, 3 mg, % Aliquat 336, 10 wt %60 ± 3 mg, 35 wt % Aliquat 336 and −1 ±NO − ; 20% composition; 100 mol ·L60 PIMwt mass and composition: 3 conditions: extraction times. Experimental solution volume and composition; 100 mL, 100 mol·L−1 NO3−; PIM wt % PVC; shaking rate, rpm. Data points are therpm). average ofpoints 3 extraction 65 wt150 % PVC; shaking rate: 150 Data are theexperiments average of 3with extraction experiments with an average mass 60 ± 3 mg, wt % Aliquat 336 and−165HCl wt % PVC; shaking rate: 150 rpm). Data points are ailable online at(SD) www.mdpi.com/s1: Figures S1–S13. Figure S1:of35Au(III) −1.and −1 .composition: S2: The extraction fromfrom 2.5 mol·L ard deviation of 0.77 mol·L SD of 0.62 mol ·LFigure Figure S7: The extraction of Au(III) 2.5 mol ·L−1 HCl solutions into PIMs containing the average of 3 extraction experiments with an average SD of 0.62 mol·L−1. Figure S7: The extraction of Au(III) xperimental conditions: solution volume and composition, 100 IMs containing Aliquat in concentrations: 25( (■); 30 15 (▲); wt % (●) 20 wt 336 % Aliquat 336 and 0 (),205 ((); ), 10 ), and wt and % (N)351-dodecanol. Experimental conditions: solution −1 HCl solutions into PIMs containing 20 wt % Aliquat 336 and 0 (■), 5 (), 10 (●), and 15 wt % from 2.5 −1 Au(III), −1 ·HCl; ss and composition, 60 ± 3 and mg, 20% wtmol·L % Aliquat 336, 10 wtmol % −1 nditions: solution volume composition; 100 mL, 100 mol·L 2.5 2.5 mol·L PIM PIM mass: 60 ± 3 mg; shaking rate: volume and composition; 100 mL, 100 ·LAu(III), mol L−1 HCl; Au(III), 2.5 (▲) 1-dodecanol. Experimental conditions: solution volume and composition; 100mol mL, mol·L−1S8: 1 . Figure m. Data points are the average of 3 extraction experiments with 150Data rpm.points Data points the average of 3 extraction experiments with an average SD of 0.73 ·L−100 shaking rate: 150 rpm). are theare average of 3 extraction experiments with an average −1 −1. Figure S2: The extractionmol·L −1 − 1 HCl; PIM mass: 60 ± 3 mg; shaking rate: 150 rpm. Data points are the average of 3 extraction experiments of Au(III) from 2.5 mol·L HCl 1L − The extraction Au(III) 2.5 molAuNP ·L HCl solutions into PIMs times. containing 20 wt % Aliquat 336 and 0 (), . Figure S3: The extraction of NO3 of with PIMsfrom of different loadings and extraction −1. Figure S8: The extraction of Au(III) from 2.5 mol·L−1 HCl solutions into PIMs an of 0.73 mol·L nnditions: concentrations: 25and (■); 30 average (▲); wt (●) 5 ((); ), 10 (with ), composition; and 15 wtand %SD (100 N35 ) NPOE. Experimental solution 3−; PIM mass and volume and composition; 100 mL, solution20volume mL,% 100 mol·L−1 NOconditions: − 1−1Au(III), −−11 HCl;336 containing 20 wt % Aliquat and 0 (■), 5 (), 10 (●), and 15 wt (▲) NPOE. conditions: omposition; 100 mL, 100 mol·L Au(III), 2.5 mol·L 100 mol · L 2.5 mol · L PIM mass: 60 ± 3 mg; shaking rate: rpm. DataExperimental points are the average solution 3 mg, 20 wt % Aliquat 336 and 80 wt % PVC; shaking rate: 150 rpm. Data points are the average%150 −1 Au(III), −1 HCl; PIM mass: 60 ± 3−mg; shaking rate: 150 −1 . 2.5 volume and composition; 100 mL, 100 mol·L mol·L of 3 extraction experiments with an average SD of 0.70 mol · L Figure S9: The extraction of NO with PIMs −1 − nts are the average of 3 extraction experiments with an average 3 periments with an average SD of 0.60 mol·L . Figure S4: The extraction of NO3 with PIMs of −1. Figure S9: The − with PIMs of different of different AuNP loadings andthe extraction times. Experimental conditions: solution volume and composition; rpm. Data points are average of 3 extraction experiments with an average SD of 0.70 mol·L O 3 AuNP loadings and extraction times. oadings and extraction times.Experimental conditions: solution volume and composition; 100 − ;− PIM mass 100 mL,100 100 mol·L−1−1NO NO and composition: 60 ± 3 mg,and 20 wt % Aliquat 336, Experimental 5 wt % 1-dodecanol and solution extraction NO 3PIM with PIMs of different extraction times. conditions: 3−3; 25 mass and d mL, mol·L NOcomposition; 3−; PIM mass 100 and composition: 60 ± 3ofmg, wt % Aliquat 336 andAuNP 75 wtloadings % PVC; shaking 75 wt % PVC; shaking rate: 150 rpm. Data points are −1 the average of 3 extraction experiments with an average NO 3−; PIM mass and composition: 60 ± 3 mg, 20 wt % Aliquat 336, volume and composition; 100 mL, 100 mol·L −1 0 wt % PVC; shaking rate: 150 rpm. Data points are the average . Figure ta points are the average 3 extraction withextraction an averageofSD of − 0.62 mol·L −1 .experiments SD ofof0.69 mol ·L Figure S10: The NO PIMs of rpm. different AuNP loadings and extraction 3 with −1. Figure S4: The5extraction − with wt % 1-dodecanol and 75 wt % PVC; shaking rate: 150 Data points are the average of 3 extraction of 0.60 mol·L of NO 3 PIMs of − n of NO3 with PIMs times. of different AuNP loadings and extraction times. Experimental conditions: − Experimental conditions: solution volume and composition; 100 mL, 100 mol·L−1 NO 3 ; PIM mass and −1. Figure S10: experiments with an average SD of 0.69 mol·L The extraction of NO 3− with PIMs of different AuNP xperimental conditions: solution volume and composition; 100 −1 − ; PIM mass and 336, composition: 60 ± 3 mg, and 30 wt and composition; 100 mL, 100 mol·L composition: 60 ± 3NO mg,3 20 wt % Aliquat 15 wt % 1-dodecanol 55% wt % PVC; shaking rate: 150 rpm. Data loadings and extraction Experimental conditions: solution volume and composition; 100 mL, 100 mol·L−1 on: 60 ±%3PVC; mg, 25 wtpoints % Aliquat 336 and 75ofwt % PVC; shaking 70 wt shaking rate: rpm. Data points are times. the average of with 3 extraction experiments are150 the average 3 extraction experiments an average SD of 0.67 mol·L−1 . Figure S11: The extraction −1. Figure − average NOextraction 3−; PIM mass and composition: 60and ± 3 extraction mg, 20 wt % Aliquat % 1-dodecanol 55 wt % PVC; raction experiments an SDof of 0.62 mol·L ofwith NOS6: with PIMs different loadings times. Experimental conditions: solutionand volume SD of 0.62 mol·L−1. Figure of NO 3− AuNP with PIMs of different AuNP loadings and 336, 15 wt 3 The − 1 − shaking rate: 150 rpm. Data points are the average of 3 extraction experiments with an average SD of 0.67 mol·L−1. −1 NO AuNP loadingsconditions: andand extraction times. Experimental conditions: composition; 100 mL, mol ·L NO ; PIM mass and composition: 60 ± 3 mg, 20 wt % Aliquat 336, 3−; PIM Experimental solution volume and100 composition; 100 100 mol·L 3 mL, − with PIMs of different AuNP loadings and extraction times. Experimental −1 − 3 Figure S11: The extraction of NO 5 wt % NPOE and 75 wt % PVC; shaking rate: 150 rpm. Data points are the average of 3 extraction experiments ; PIM composition: 3 mg, wt % rate: 150 rpm). Data points are mol·L sition: 60NO ± 33 mg, 35mass wt % and Aliquat 336 and 6560wt± % PVC;30shaking − −1 NO with anaverage average of 0.68 molvolume ·L−1 . −1 Figure S12: NO PIMs of different AuNP loadings60 ± 3 mg, solution composition; 100 mL,ofof 100 mol·L 3−; PIM mass and composition: rpm. Dataexperiments points are the ofSD 3 extraction experiments 3 with extraction with anconditions: average SD of 0.62 mol·L .and Figure S7:The Theextraction extraction Au(III) −1 NO − ; − and extraction times. Experimental conditions: solution volume and composition; 100 mL, 100 mol · L 20 wt % Aliquat 336, 5 wt % NPOE and 75 wt % PVC; shaking rate: 150 rpm. Data points are the 3average of 3 eHCl extraction of NO with containing PIMs of different AuNP loadings and0 (■), 5 (), 10 (●), and 15 wt % solutions into3 PIMs 20 wt % Aliquat 336 and PIM mass and composition: 60 ± 3 mg, 20 wt % Aliquat 336, 10 wt NPOE and 70 wt % PVC; shaking −1% −1 − . Figure S12: The extraction of NO 3− rate: with PIMs of extraction experiments with an average SD of 0.68 mol·L NO3 ; PIM100 mL, 100 mol·L−1 Au(III), 2.5 ion volume andconditions: composition; 100 mL, 100 mol·L Experimental solution volume and composition; 150 rpm. Data points are the averageand of 3extraction extractiontimes. experiments with anconditions: average SDsolution of 0.65 mol · L−1 .and Figure S13: different AuNP loadings Experimental volume composition; 100 tmass: 336 and PVC; shaking rate: 150Data rpm). Dataare points are − with 60 ±653 wt mg;%shaking rate: 150 rpm. points the average of 3 extraction experiments The extraction of NO PIMs of different AuNP loadings and extraction times. Experimental conditions: 3 −1 −; PIM −1 mL, 100 mol·L NO 3 mass and composition: 60 ± 3 mg, 20 wt % Aliquat 336, 10 wt % NPOE and 70 wt % −1. Figure −1 average SD of 0.62 mol·L S8: . Figure S7: The extraction Au(III) The extraction of Au(III)of from 2.5 mol·L intomass PIMsand composition: 60 ± 3 mg, 20 wt % SD of 0.73 mol·L solution volume and composition; 100 mL, 100 molHCl ·L−1solutions NO3 − ; PIM PVC; shaking rate: 150 rpm. Data points are the average of 3 extraction experiments with an average SD of 0.65 ning 20 wt336 % Aliquat 336 and 0 (■), 5and (), and 15 % % Aliquat and 0 Aliquat (■), 5 (), 10 (●),wt 1510 wt(●), %and (▲) Experimental solution 336, 15 % NPOE 65NPOE. wtwt % PVC; shaking conditions: rate: 150 rpm. Data points are the average of 3 extraction −1. Figure S13: −1 The extraction 3− with PIMs of different AuNP loadings and extraction times. 1of Au(III), 2.5 ion volume 100 mL, 100 mol·L −1 SD experiments with an average of PIM 0.67 mol ·L−60 . ± NO position; 100and mL,composition; 100 mol·L−1 mol·L Au(III), 2.5 mol·L HCl; mass: 3 mg; shaking rate: 150 Experimental conditions: solution and composition; 100 mL, 100 mol·L−1 NO3−; PIM mass and −1. Figure 50 points are the average of 3 extraction experiments S9: TheUniversity arerpm. the Data average of 3Acknowledgments: extraction experiments with an average SD of volume 0.70 mol·L Ya Ya Nutchapurida Bonggotgetsakul is grateful to the Melbourne forrate: receiving −1 60 ± 3 mg, 20 PIMs wt Experimental % Aliquat 336, 15 wt % NPOE and 65 wt %ofPVC; shaking 150 rpm. Data e− with extraction Au(III) 2.5composition: mol·L HCl solutions into PIMs of of different AuNP loadings and extraction times. thefrom Albert Shimmins Postgraduate Writing-up Award. conditions: solution −1. points are the average of 3 extraction experiments with an average SD of 0.67 mol·L −1 − 0position; (●), and 100 15 wt % (▲) NPOE. Experimental conditions: solution mL, 100 mol·L NO3 ; PIM mass and composition: 60 ± 3 mg, 20 wt % Aliquat 336,

−1 u(III), 2.5 75 mol·L mass: 60 ± 3150 mg; shaking 150are the average anol and wt %HCl; PVC;PIM shaking rate: rpm. Data points of 3 extraction Acknowledgments: Ya rate: Ya Nutchapurida Bonggotgetsakul is grateful to the University of Melbourne for −1. Figure S9: The − experiments with an average SD of 0.70 mol·L −1 an average SD of 0.69 mol·L receiving . Figure S10: extraction of NO 3 with PIMs of differentAward. AuNP theThe Albert Shimmins Postgraduate Writing-up oadings and extraction times.conditions: Experimental conditions: solution action times. Experimental solution volume and composition; 100 mL, 100 mol·L−1 Oand 3−; PIM mass and composition: 60 ± 3 mg, 20 wt % Aliquat composition: 60 ± 3 mg, 20 wt % Aliquat 336, 15 wt 336, % 1-dodecanol and 55 wt % PVC; g rate: 150points rpm. are Data areofthe average experiments of 3 extraction rpm. Data thepoints average 3 extraction with an average SD of 0.67 mol·L−1. −


11 of 12

Author Contributions: Y.Y.N.B. conducted the experimental work as a Ph.D. student under the supervision of S.C.K. and R.W.C. All three authors contributed to the writing of the manuscript. Conflicts of Interest: The authors declare no conflict of interest.

References 1.

2. 3. 4.

5. 6.

7.

8.

9.

10.

11. 12. 13.

14. 15.

16. 17. 18.

Nghiem, L.D.; Mornane, P.; Potter, I.D.; Perera, J.M.; Cattrall, R.W.; Kolev, S.D. Extraction and transport of metal ions and small organic compounds using polymer inclusion membranes (PIMs). J. Membr. Sci. 2006, 281, 7–41. [CrossRef] O0 Rourke, M.; Cattrall, R.W.; Kolev, S.D.; Potter, I.D. The extraction and transport of organic molecules using polymer inclusion membranes. Solv. Extr. Res. Dev. Jpn. 2009, 16, 1–12. Almeida, M.I.G.S.; Cattrall, R.W.; Kolev, S.D. Recent trends in extraction and transport of metal ions using polymer inclusion membranes (PIMs). J. Membr. Sci. 2012, 415, 9–23. [CrossRef] O’Bryan, Y.; Truong, Y.B.; Cattrall, R.W.; Kyratzis, I.L.; Kolev, S.D. A new generation of highly stable and permeable polymer inclusion membranes (PIMs) with their carrier immobilized in a crosslinked semi-interpenetrating polymer network. Application to the transport of thiocyanate. J. Membr. Sci. 2017, 529, 55–62. [CrossRef] Guo, L.; Liu, Y.H.; Zhang, C.; Chen, J. Preparation of PVDF-based polymer inclusion membrane using ionic liquid plasticizer and Cyphos IL 104 carrier for Cr(VI) transport. J. Membr. Sci. 2011, 367, 85–90. [CrossRef] Bonggotgetsakul, Y.Y.N.; Cattrall, R.W.; Kolev, S.D. Recovery of gold from aqua regia digested electronic scrap using a PVDF-HFP based polymer inclusion membrane (PIM) containing Cyphos® IL 104. J. Membr. Sci. 2016, 514, 274–281. [CrossRef] Wang, D.; Cattrall, R.W.; Li, J.; Almeida, M.I.G.S.; Stevens, G.W.; Kolev, S.D. A poly(vinylidene fluoride-co-hexafluoropropylene) (PVDF-HFP)-based polymer inclusion membrane (PIM) containing LIX84I for the extraction and transport of Cu(II) from its ammonium sulfate/ammonia solutions. J. Membr. Sci. 2017, 542, 272–279. [CrossRef] Yaftian, M.R.; Almeida, M.I.G.S.; Cattrall, R.W.; Kolev, S.D. Selective extraction of vanadium(V) from sulfate solutions into a polymer inclusion membrane composed of poly(vinylidenefluoride-co-hexafluoro-propylene) and Cyphos® IL 101. J. Membr. Sci. 2018, 545, 57–65. [CrossRef] Bonggotgetsakul, Y.Y.N.; Cattrall, R.W.; Kolev, S.D. Extraction of gold(III) from hydrochloric acid solutions with a PVC-based polymer inclusion membrane (PIM) containing Cyphos® IL 104. Membranes 2015, 5, 903–914. [CrossRef] [PubMed] Nagul, E.A.; Fontàs, C.; McKelvie, I.D.; Cattrall, R.W.; Kolev, S.D. The use of a polymer inclusion membrane for separation and preconcentration of orthophosphate in flow analysis. Anal. Chim. Acta 2013, 803, 82–90. [CrossRef] [PubMed] Pratiwi, A.I.; Sato, T.; Matsumoto, M.; Kondo, K. Permeation mechanism of succinic acid through polymer inclusion membranes with ionic liquid Aliquat 336. J. Chem. Eng. Jpn. 2014, 47, 314–318. [CrossRef] Almeida, M.I.G.S.; Cattrall, R.W.; Kolev, S.D. Polymer inclusion membranes (PIMs) in chemical analysis—A review. Anal. Chim. Acta 2017, 987, 1–14. [CrossRef] [PubMed] Kumar, R.; Pandey, A.K.; Tyagi, A.K.; Dey, G.K.; Ramagiri, S.V.; Bellare, J.R.; Goswami, A. In situ formation of stable gold nanoparticles in polymer inclusion membranes. J. Colloid Interface Sci. 2009, 337, 523–530. [CrossRef] [PubMed] Kumar, R.; Pandey, A.K.; Shukla, R.; Ramagiri, S.V.; Bellare, J.R. Plasticised polymer inclusion membrane as tunable host for stable gold nanoparticles. Int. J. Nanotechnol. 2010, 7, 953–966. [CrossRef] Bonggotgetsakul, Y.Y.N.; Cattrall, R.W.; Kolev, S.D. The preparation of a gold nanoparticle monolayer on the surface of a polymer inclusion membrane using EDTA as the reducing agent. J. Membr. Sci. 2011, 379, 322–329. [CrossRef] Bonggotgetsakul, Y.Y.N.; Cattrall, R.W.; Kolev, S.D. A method for the coating of a polymer inclusion membrane with a monolayer of silver nanoparticles. J. Membr. Sci. 2013, 428, 142–149. [CrossRef] Bonggotgetsakul, Y.Y.N.; Cattrall, R.W.; Kolev, S.D. A method for coating a polymer inclusion membrane with a monolayer of palladium nanoparticles. React. Funct. Polym. 2015, 97, 30–36. [CrossRef] Pereira, N.; St John, A.; Cattrall, R.W.; Perera, J.M.; Kolev, S.D. Influence of the composition of polymer inclusion membranes on their homogeneity and flexibility. Desalination 2009, 236, 327–333. [CrossRef]


19.

20. 21.

22. 23. 24.

25. 26. 27.

12 of 12

Arous, O.; Kerdjoudj, H.; Seta, P. Comparison of carrier-facilitated silver(I) and copper(II) ions transport mechanisms in a supported liquid membrane and in a plasticized cellulose triacetate membrane. J. Membr. Sci. 2004, 241, 177–185. [CrossRef] Xu, J.Y.; Wang, L.J.; Shen, W.; Paimin, R.; Wang, X.G. The influence of the interior structure of Aliquat 336/PVC membranes to their extraction behavior. Sep. Sci. Technol. 2004, 39, 3527–3539. [CrossRef] St John, A.M.; Best, S.P.; Wang, Y.D.; Tobin, M.J.; Puskar, L.; Siegele, R.; Cattrall, R.W.; Kolev, S.D. Micrometer-scale 2D mapping of the composition and homogeneity of polymer inclusion membranes. Aust. J. Chem. 2011, 64, 930–938. [CrossRef] St John, A.M.; Cattrall, R.W.; Kolev, S.D. Determination of the initial flux of polymer inclusion membranes. Sep. Purif. Technol. 2013, 116, 41–45. [CrossRef] Cho, Y.; Xu, C.L.; Cattrall, R.W.; Kolev, S.D. A polymer inclusion membrane for extracting thiocyanate from weakly alkaline solutions. J. Membr. Sci. 2011, 367, 85–90. [CrossRef] Cho, Y.; Cattrall, R.W.; Kolev, S.D. A novel polymer inclusion membrane based method for continuous clean-up of thiocyanate from gold mine tailings water. J. Hazard. Mater. 2017, 341, 297–303. [CrossRef] [PubMed] Argiropoulos, G.; Cattrall, R.W.; Hamilton, I.C.; Kolev, S.D.; Paimin, R. The study of a membrane for extracting gold(III) from hydrochloric acid solutions. J. Membr. Sci. 1998, 138, 279–285. [CrossRef] Kolev, S.D.; Argiropoulos, G.; Cattrall, R.W.; Hamilton, I.C.; Paimin, R. Mathematical modelling of membrane extraction of gold(III) from hydrochloric acid solutions. J. Membr. Sci. 1997, 137, 261–269. [CrossRef] Bonggotgetsakul, Y.Y.N.; Ashokkumar, M.; Cattrall, R.W.; Kolev, S.D. The use of sonication to increase extraction rate in polymer inclusion membranes. An application to the extraction of gold(III). J. Membr. Sci. 2010, 365, 242–247. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).