WO2012071605A1 - Process for preparing gold nanoparticles - Google Patents

Abstract

The present invention generally relates to the field of nanoparticle formation and in particular to processes which generate gold nanoparticles on the surface of a polymer substrate. The invention particularly relates to processes for preparing monolayers of gold nanoparticles on the surface of polymer inclusion membranes (PIMs), and PIMs with gold nanoparticles deposited on the surface thereof.

Description

PROCESS FOR PREPARING GOLD NANOP ARTICLES

FIELD OF THE INVENTION

The present invention generally relates to the field of nanoparticle formation and in particular to processes which generate gold nanopaiticles on the surface of a polymer substrate. The invention particularly relates to processes for preparing monolayers of gold nanopaiticles on the surface of polymer inclusion membranes (PIMs), and PIMs with gold nanopaiticles deposited on the surface thereof.

BACKGROUND OF THE INVENTION

During the past two decades, there has been a rapid growth in nanoscience with many important applications emerging. Gold nanopaiticles (Au nps), in particular, have become a popular topic for research due to their unique photonic, electronic, magnetic and catalytic properties. Different sizes and structures for Au nps can be prepared using a wide variety of methods which result in changes in their appearance and properties. This has led to many important applications (e.g. novel biomedical imaging techniques and therapies; sensor development for detecting biological molecules and heavy metals ions; catalytic surfaces for various reactions). In general, there are two main processes for Au nps synthesis, namely, linear-template methods and template-free self-assembly methods. Linear-template methods involve the use of templates to grow Au nps in the presence of a suitable reducing reagent. The templates used range from linear polymers, biomoJecules (e.g. DNA and proteins), inorganic nanowires and nanotubes,. and step edges on solid substrates (e.g. electrodepositing of nps on solid substrates). The preparation of Au nps on linear polymers such as Nafion, po!y(vinyl chloride) (PVC) and cellulose triacetate (CTA) has been studied. In recent years, membrane-based processes for selective separation of heavy metals have attracted considerable attention, They have many advantages over other processes as they use lower amounts of solvents than solvent extraction and the membrane systems are usually compact and robust, enabling them to be easily implemented in existing industrial processes. Polymer inclusion membranes are a relatively new type of liquid membranes formed by casting a solution containing an extractant, . a plasticizer and/or modifier and a base polymer dissolved in a volatile solvent and allowing the solvent to evaporate. The resulting membrane is transparent and visually homogeneous and has a number of advantages over the other liquid membrane types (e.g. supported liquid membranes, bulk liquid membranes and emulsio liquid membranes), particularly in terms of stability. The extractant, often referred to as the carrier, is usually a complexing agent or an ion-exchanger and is an essential component for metal extraction. A plasticizer (e.g. 2-nitrophenyloctyl ether (2- NPOE)) and/or modifier (e.g. 1-dodecanol) is often added to the membrane composition' to increase the membrane softness, flexibility and compatibility of the membrane components and also to increase the diffusive flux within the polymer matrix.

Kumar et. al., (J. Colloid Interface Sci., 337 (2009) 523-530) recently published a paper in which they described the synthesis of Au nps using a PIM containing Aliquat 336 immobilized in a CTA polymer matrix with dioctyl phthalate (DOP) as plasticizer. Firstly, gold(III) was extracted into the membrane from chloride solutions and was then reacted with a NaBHL_t solution. This resulted in the formation of metallic Au nps with an average size of 10 nm distributed throughout the bulk of the membrane.

The present invention utilizes similar PIM based methodology with the distinct advantage of being able to prepare monolayers of gold nanoparticles predominantly on the surface of the membrane.

SUMMARY OF THE INVENTION The present invention provides a process for preparing monolayers of gold nanoparticles on a PIM surface in an efficient and high yielding manner. . In one aspect the present invention provides a method of preparing gold nanoparticles on the surface of a polymer inclusion membrane (PIM) including the steps of:

i. mixing a quaternary ammonium salt, plasticizer and/or modifier, polymer, and organic solvent for a time and under conditions suitable for preparing a homogenous solution and then casting a PIM from said solution;

ii. contacting the PIM with an acid solution comprising a Au(III) species and extracting at least a portion of the AuQII) species from the acid solution into the PIM to form a PIM loaded with said Αυ(ΙΠ) species; and iii. treating the loaded PIM from step (ii) with a Au(III) reducing agent characterised in mat the agent cannot readily reduce the Au(IIl) species within the PIM, leading to the formation of gold nanoparticles on the surface of the PIM,

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1. The extraction of Au(III) from 2.5 M HC1 solutions with CTA (M) or PVC

(♦) based PIMs (Experimental conditions: solution volume and composition; 100 mL, 100 mg L"¹ Au(III), 2.5 M HC1; membrane mass and composition: 60±3 mg, 20% m m Aliquat 336, 10% m/m n-dodecanol (DD), and 70% m m PVC or CTA; shaking rate: 150 rpm).

Figure 2 (a)-(d). Photographs of PVC/Aliquat 336/DD PIMs loaded with Au(M) after 24 h exposure to solutions of 1.00 mol L^"1 of (a) L-ascorbic acid, (b) NaBHj, (c) tri-sodium citrate, and (d) di-sodium ethylenediaminetetraacetate NazEDTA.

Figure 3 (a)-(h). SEM images of the surface and cross-section of PVC/Aliquat 336/DD

PIMs loaded with Au(III) after 24 h exposure to solutions of 1 mol L^"1 of (a,e) L-ascorbic acid, (b,f) NaBH , (c,g) tri-sodium citrate, and (d,h) Na₂EDTA. Figure 4. The IR spectrum of PIMs before (dotted line) and after (solid line) extraction of 1.00 10^'3 M ethylenediaminetetraacetic acid (EDTA) solutions at pH 4.5 (Experimental conditions: solution volume: 100 mL; membrane mass and composition: 250±3 mg, 20% m/m Aliquat 336, 10% m/m DD, and 70% m m PVC; shaking rate: 150 rpm).

Figure 5. The extraction of 1.00 I0^*3 M EDTA solutions at pH 4.0 (♦), 4.5(B), 6.0

(A) and 8.0 (X) (Experimental conditions: solution volume: 100 mL; pH adjusted with 0.1 M HO or NaOH solutions; membrane mass and composition: 250*3 mg, 20% ra/m Aliquat 336, 10% m/m DD, and 70% m m PVC; shaking rate: 150 rpm).

Figure 6 (a)-(d). Photographs of PVC/AIiquat 3367DD PIMs loaded with Au(lll) after 24 h exposure to solutions of 1.00 M EDTA at pH (a) 4.0, (b) 4.5, (c) 6.0, and (d) 8.0.

Figure 7 (a)-(h). SEM images of the surface and cross-section of PVC AIiquat 336/DD

PIMs loaded with Au(III) after 24 h exposure to solutions of 1.00 M EDTA at pH (a,e) 4.0, (b,f) 4.5. (c.g) 6.0. arid (d.h) 8.0.

Figure 8. Size distribution histogram for Au nps on the surface of a PIM conditioned with 0.10 mol L^"1 EDTA at pH 6.

Figure 9. Reflectance visible spectrum of the surface of a PIM conditioned with 0.10 mol L^"1 EDTA at pH 6.

Figure 10 (a)-(d). Photographs of PVC/AIiquat 336 DD PIMs loaded with Au(HI) after

24 h exposure to solution at pH 7 containing different concentrations of EDTA: (a) 5.00 10^"2 mol L^'1, (b) 0.100 mol L^"1, (c) 0.150 mol I/¹, and (d) 0.200 mol L ¹.

Figure 11 (a)-(h). SEM images of the surface and cross-section of PVC/Aliquat 336/DD

PIMs loaded with Au(III) after 24 h exposure to solutions at pH 7 containing different concentrations of EDTA: (a,e) 5.00 10'² mol L"¹, (b,f) 0.100 mol L^"1, (c,g) 0.150 mol L^"\ and (d,h) 0.200 mol L^"1.

Figure 12 (a)-(e). Photographs of PVC/Aliquat 336/DD PIMs with different Au(III) extraction time and exposed to 0.100 M EDTA solutions at pH 6 for 24 h

(a) 0.5 h, (b) 3 h, (c) 6 h, (d 9 h, and (e) 12 h.

Figure 13 (a)-(j). SEM images of the surface and cross-section of PVC/Aliquat 336/DD

PIMs with different Au(III) extraction time and exposed to 0.100 M EDTA solutions at pH 6 for 24 h (a,f) 0.5 h, (b,g 3 h, (c,h) 6 h, (d,i) 9 h, and (e j)

12 h.

Figure 14 (a)-(g). Photographs of PVC/Aliquat 336 DD PIMs loaded with Au(III) after different exposure periods to 0,100 M EDTA solutions at pH 6 (a) 1 h, (b) 2 h, (c) 3 h, (d) 4 h, (e) 5 h_i (f) 10 h, and (g) 24 h.

Figure IS (a)-(n). SEM images of the surface and cross-section of PVC/Aliquat 336/DD

PIMs loaded with Au(III) after different exposure periods to 0.100 M EDTA solutions at different pH values: 6 (a,h) 1 h, (b,i) 2 h, (cj) 3 h, (djc) 4 h, (e,l) 5 h, (f,m) I0 h, and (g-n) 24 h.

Figure 16 (e)-(c). SEM images of the surface of PVC/Aliquat 336/DD PIMs loaded with

Au(IH) after exposure to 0.100 M EDTA solution at pH 6 and different temperatures: (a) 10 °C, (b) 20 °C and (c) 30 °C. Figure 17 (a)-(c). SEM images of the surface of PVC Aliquat 336VDD PI s loaded with Au(III) after exposure to 0.100 M EDTA solutions at pH 6 and different shaking rates: (a) 75 rpm and lower, (b) 100 rpm and (c) 150 rpm. Figure 18. The average size (diameter) of Au nps formed on the surface of

PVC/Aliquat 3367DD PIMs loaded with Au(IH) after exposure to 0.100 M EDTA at pH 6 and different shaking rates: (a) 75 rpm and lower, (b) 100 rpm and (c) 150 rpm. Figure 19. Graphs depicting the relationship between the amount of Hg accumulated onto the Au-PI based on (a) the concentration of Hg in the ambient air for different adsorption periods (2-7 days) or (b) the duration of adsorption for different concentrations of Hg in the ambient air.

DESCRIPTION OF THE INVENTION

Throughout this specification and the claims which follow, unless the context requires otherwise, the word "comprise", and variations such as "comprises" and "comprising", will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

The reference in this specification to any prior publication (or information derived from it), or to any matter which is known, is not, and should not be taken as an acknowledgment or admission or any form of suggestion that prior publication (or information derived from it) or known matter forms part of the common general knowledge in the field of endeavour to which this specification relates.

The present invention is predicated on the discovery that a monolayer of gold nanoparticles can be prepared on the surface of a PIM through the reduction of Au(III) species by a reducing agent which cannot readily reduce the Au(HI) species in the bulk environment of the PIM and as such the reduction occurs almost exclusively at the PIM surface. Nature and Preparation of the PIM

Polymer inclusion membranes are generally known in the art, and may also be referred to as "polymer liquids", "gelled liquids", "polymeric plasticized", "fixed-site carriers" or "solvent polymeric membranes". The main advantage of PI s over, for instance, supported liquid membranes (SLMs) is their stability. Also, unlike bulk liquid membranes (BL s), PIMs are generally not characterised as having low interfacial surface areas and mass transport rates. PIMs also do not suffer the problem of emulsion breakage which tends to plague emulsion liquid membranes (ELMs).

PIMs according to the present invention are generally formed by mixing (casting) a solution which contains a Au(III) species extractant (ie the quaternary ammonium salt), a plasticizer/modifier and a base polymer. The casting process is typically facilitated with the use of organic solvents (such as ethers (eg THF, dicthylether) and chlorinated solvents (eg dichloromethane)), which are typically removed during membrane formation (eg by air drying or in vacuo).

It will be appreciated that the "extraction" process referred to herein involves the controlled transport of Au(III) species into the membrane. Such a process is facilitated by a carrier (referred to herein as the "extractant") that is essentially a Au(IH) complexing agent or an ion-exchanger. In respect of the PIMs of the present invention this extractant is the immobilised quaternary ammonium salt. a) Quaternary ammonium salt

In an embodiment the quaternary ammonium salt is represented by formula (1):

(1)

where R'-R⁴ are independent alkyl chains and X© is an anion.

In an embodiment R¹ is a C1-C4 alkyl chain, and R²-R⁴ are independently C8-C30 alkyl chains.

In a further embodiment R¹ is a CI-CJ alkyl chain, and R^?-R⁴ are independently Ce-Cio alkyl chain, and more preferably Cg-Cio alkyl chain. In a further embodiment X© is an anion such as chloride, nitrate or bromide! hi a further embodiment the immobilised quaternary ammonium salt is Aliquat 336^® (Cognis Corp.). Aliquat 336 is a mixture of compounds of formula (1) where Ri is methyl, R2-R4 are mixtures of C» (octyl) and CH> (capryl) chains (predominantly Cg), and X© is chloride.

In an embodiment the quaternary ammonium salt constitutes from 5 - 40% m/m of the PIM, preferably from about 10% - 30% m/m and more preferably from about 15 - 25% m/m of the P1M. b) Polymer

The PIM according to the present invention may be formed from any suitable base polymer which provides mechanical strength to the membrane.

In one embodiment the polymer is selected from polyvinyl chloride) (PVC), cellulose triacetate (CTA), and cellulose tributyrate (CTB), or suitable derivatives thereof.

In an embodiment the polymer is PVC, CTA or a derivative thereof.

In an embodiment the polymer is PVC. - Gr _¬

in an embodiment the polymer constitutes from about 40-80% m/m of the PIM, preferably from about 50-75% m m, more preferably from about 55-75% m m, and even more preferably about 70% m m. c Plasticizer/modifier

The PIM preferably also comprises a plasticizer or modifier component. T e role of the plasticizer is to penetrate between polymer molecules and to "neutralize" the polar groups of the polymer with its own polar groups or to merely increase the distance between the polymer molecules and hence reduce the strength of the intermolecular forces. Accordingly, the plasticizer may be any suitable organic compound which is able to function as described above. Suitable organic compounds include those containing a hydrophobic alkyl backbone with one or several highly solvating polar groups. The role of the modifier is to increase the solubility of the extracted chemical species in the membrane liquid phase.

In an embodiment the plasticizer/modifier is selected from the group consisting of 2- nitrophenyl octyl ether (2-NPOE), dibutyl butyl phosphorate (DBBP), 1-hexanol, 1- heptanol, 1-octanol, 1-nonanol, 1-decanol, l-dodecanol, 1-tctradecanol, o- ratrophenylpentyl ether (oNPPE), tributylphosphate (TBP), dioctylphthalate (DOP), bis(2- ethylhexy terephthalate (DDTP), dioctylsebacate (DOS) and t -(2- emylliexyl)phosphate(T2EHP). In an embodiment the plasticizer is selected from TBP, 2-NPOE, 1-tetradecanol and 1- dodccanol.

In an embodiment the plasticizer is l-dodecanol. In an embodiment the plasticizer/modifier constitutes from about S-40% m/m of the PIM, preferably about 5 - 30% m/m and more preferably from about 5-15% m/m. In an embodiment the ratio (based on % m/m) of polymer : quaternary ammonium salt : pIastici2fir/modifier is 5:2:3 - 7:1:2. In a further embodiment the ratio range (based on % m/m) of polymer : quaternary ammonium salt : plasticizer/modifier is about 7:2:1 to 16:3:1.

In a further embodiment the ratio is about 7:2:1, for instance, a preferred composition is 70% m m PVC, 20% m/m Aliquat 336 and 10% m/m 1-dodecanol. d) Optional other components

The skilled person would appreciate that the PIMs of the present invention may also include additional components to aid in the extraction process. For instance, the PIMs may include other quaternary ammonium salts, plasticizcrs/modifiers and base polymers, antimicrobial agents (for instance, to inhibit membrane fouling), antioxidants (for increased stability), porosity agents (porogens), ferromagnetic particles, and residual amounts of casting solvents. Extraction of Au(III) species

The extraction process according to the present invention comprises the step of contacting an aqueous acid solution (containing a negatively charged Au(III) complex) with the PIM (as characterised above). The term "contacting" as used herein (and with reference to the processes of the present invention) includes any means by which the acid solution comes into physical contact with the PIM such that the AuflH) species from the solution can be extracted by the PIM. It would be appreciated that the extraction process involves ion- exchange chemistry between the anion of the extractant (i.e., the chloride anion of a quaternary ammonium chloride) and the negatively charged Au(III) complex. Th quaternary ammonium AuQII) ion-pair thus formed on the surface f the membrane will diffuse into the bulk membrane. In this respect the Au(III) can be thought of as essentially being accumulated in the bulk of the membrane and at least partially on its surface.

Contacting in the above manner includes adding or physically immersing the PIM with or into the acid solution (for instance, in a batcbwise extraction process) or allowing a flow of the Au(III) solution to come into contact with a surface of the PIM (such as in a continuous extraction process). Possible configurations for batchwise and continuous extraction/stripping (regenerating) processes would be known by those of ordinary skill in the art and are discussed in more detail below. The Au(III)species may be any species which is able to perform ion-exchange chemistry with the extractant of the PIM under the conditions set out herein. The Au(III) species must be negatively charged (e.g. AuCl- to be able to exchange with the chloride anion of Aliquat 336 or it can be a neutral species capable of forming a complex with the chloride ion of Aliquat 336 (e.g. AuC¾).

The usual way of extracting Au(M) int the membrane mentioned above is from acid solutions, for instance, HC1 solutions, where suitable negatively charged or neutral Au(III) complexes are formed. In an embodiment the Au(III) species is present in an acid solution with a concentration of 1 -5M, such as 2-4M, for instance, about 2.5, 3, 3.5 or 4 M.

In an embodiment the Au(Hl) species can be supplied commercially as HAuCl* (from Aldrich Chemical Company) dissolved in about 2.5M HC1.

The acid solution of Au(III) is preferably contacted with PIM so as to provide the maximum surface area of PIM to absorb (and hence extract) the Au(III). Preferably the treatment process involves immersing the PIM in the aqueous solution. Immersion of the PIM may be achieved by any convenient means and will depend on the form of the PIM (i.e., beads, hollow fibres, sheets, plates, etc). For instance if the PIM is in the form of beads, the PIM beads may be immersed in the solution and dispersed by mechanical agitation such as stirrers and the like or with the use of mixing pumps immersed in the aqueous solution, or by the use of gas (eg air) bubbled through the aqueous solution. Sufficient shear forces will need to be imparted on the solution to optimise dispersion of PIM beads.

Once used in an extraction process the PIM is said to be loaded with Au(III) (i.e., loaded PIM).

In a further embodiment, the process of the present invention may typically involve an additional step of physically separating the loaded PIM from the aqueous acid solution prior to the reduction step.

Physical separation may be achieved by allowing the loaded PIM to settle or by simply filtering through a mesh of appropriate porosity. Other means for separation and collection of the loaded PIM include (for instance, on an industrial scale) the use of vacuum collectors, magnetic transport (for instance where the PIM comprises magnetic particles), belts, pipes, disks, drums, auger screws, etc. Whatever the means it is preferred that the separation and collection process does not (to any great extent) cause mechanical wear which may lead to attrition of the PIM.

Reduction of Au(III) species

An important feature of the present invention is the treatment of the loaded PIM with a reducing agent. The reducing agent is said to be characterised in that it is not capable of readily reducing the Au(III) species within the bulk of the PIM thus leading to the formation of a gold nanoparticles on the surface of the PIM. One notable reducing agent for the present process is EDTA (ethylenediaminetetraacetic acid). The reduction process is said to be completed once >90% of the PIM surface is covered with Au nps. In an embodiment this coverage can be achieved by exposing the PIM to an EDTA solution for about 24 h. Exposure as used herein refers to contacting the EDTA solution with the spent OT loaded PI , for instance, by immersion, mixing or shaking.

In an embodiment the reducing agent is selected from a compound of formula (II):

wherein:

m is an integer from 1 to 6 (preferably 2); and

n to q are independently selected from an integer from 1 to 4 (preferably n, o, p, and q are all 1),

or a salt thereof.

In an embodiment the reducing agent is a citric acid salt, for instance, trisodium citrate.

In an embodiment the average size of the Au nanoparticles on the PIM surface is from about 10-30 nm.

In another embodiment the average size of the Au nanoparticles is from about 15-25 nm.

In a further embodiment the average size of the Au nanoparticles is about 20 nm.

In an embodiment, where the reducing agent is EDTA, the reduction step is conducted at a pH of 5,0-6.8, and more preferably in a range of 5.5-6.5, and preferably at about 6.

Without wanting to be bound by any particular theory it is believed that the reducing agent of present invention, such as, EDTA (and polyamino carboxylic acid reducing agents like EDTA) requires water to reduce Au(IIi) to metallic gold and that water is not available within the membrane (i.e., hydrophobic membrane bulk). Reducing agents used in prior art methods such as NaBRi do not need water to facilitate the reduction process and as such these agents can operate inside the membrane bulk. Accordingly, it is postulated that for this reason (at least in part), the reduction process occurs almost exclusively at the membrane surface.

From this one can observe (based on the present methodology) at least 80 % of metallic gold forming on the surface (relative to the t% of the extracted Au(III) species), and preferably >90% and more preferably > 95%, such as about 96%, 97%, 98%, 99% or 100%.

Accordingly, in another aspect the invention provides a PIM comprising gold in the form of Au(III) and metallic gold nanoparticles wherein at least 80 % of the gold is in the form of metallic gold nanoparticles which are present on the surface of said PIM. In this aspect it will be appreciated that the remaining 20% of the gold (either as metallic gold or Au(lII)) will reside in the bulk of the PIM.

In an embodiment more than 90% of the gold nanoparticles are present on the surface of satd PIM.

In an embodiment more than 95% of the gold nanoparticles are present on the surface of said PIM. In the above aspect and embodiments the PIM is preferably a PVC polymer based PIM.

In the above aspect and embodiments the PIM is preferably a PVC polymer based PIM, and the gold nanoparticles are spherical with an average size of about 1 -30 run. The gold nanoparticles at the surface of the membrane as produced by the present invention may be used to react directly with suitable reactive species (e.g., molecules and ions, and metals such as mercury vapour) in a variety of applications due to their photonic, electronic, magnetic, and catalytic properties.

As a further advantage, SE analysis of the surface of the PIMs of the present invention revealed that the Au nps are formed predominantly (and in embodiments, almost exclusively) as a monolayer which will enhance the functioning of the Au nps through increased surface area.

The PIMs of the present invention are thus of interest in optical sensing, passive sampling and catalytic applications.

In an embodiment PIMs of the present invention which are characterized with a surface of Au nps (or Au-PIM) may be employed in the passive sampling of mercury in air. Passive sampling, which is the sampling over a long period of time (e.g., from days to many months), is a simple and low cost technology that allows the sampling of a huge array of chemicals at numerous locations. It also allows the determination of the average concentration of bio-available analytes over time, which is not possible with spot sampling due to their ever-fluctuating concentrations in nature and/or episodic contamination. This approach also allows the study of the uptake and accumulation of chemicals, e.g. metals, and organic pollutants, in organisms.

Accordingly, in a further aspect the invention provides for the incorporation of the Au- PIMs of the present invention into a passive sampling device.

The present inventors have also found that the Au-PIMs of the present invention have a strong adsorption ability and high capacity for metallic mercury (Hg°) accumulated from the air by forming an amalgam. Thus in an embodiment the passive sampling device incorporating the Au-PIMs of the present invention is used for the monitoring of mercury vapour in air. Accordingly, the invention also provides a method of monitoring mercury in an air sample, said method comprising the step of contacting an air sample with a passive sampling device whioh comprises a Au-PIM of the present invention^•wherein the air sample is exposed to the Au-PIM.

The monitoring process described above also typically involves a detection step. This detection step may involve any one which is currently employed to detect the production of the Au-Hg amalgam. For instance, such a reaction may lead to a colour change in the Au nps which can be detected using standard colorimetric techniques, for instance, a localized surface plasmon resonance (LSPR) - based colorimetric detection method.

Alternatively, mercury in the amalgam on the surface of the Au nps can be dissolved by . immersing the PIM into a suitable acidic solution where it can be determined by a suitable analytical method (eg atomic fluorescence spectrometry).

The invention will now be further described with reference to the figures and the following non-limiting examples. However, it is to be understood that the particularity of the following description of the invention is not to supersede the generality of the preceding description of the invention.

E ample

1.1 Chemicals

Aliquat 336 (AQ), (Aldrich, a mixture of quaternary ammonium chlorides), high molecular weight powdered PVC (Fluka), cellulose triacetate (CTA) (Aldrich), 1-dodecanol (DD) (Aldrich), dichloromethane (Lab-scan, Australia) and tetrahydrofuran (Chem-supply, Australia) were used as received. Au(III) calibration standards were made from a 1000 mg L Au(III) standard solution (BDH Spectrosol). Au(III) solutions for membrane extraction were prepared from HAuCl₄ (Aldrich) dissolved in 2,5 M HC1. Aqueous solutions (0.10 M) of L-ascorbic acid (Sigma), tri-sodium citrate (Cbera-supply, Australia) and sodium borohydride (Ajax Finechem, Australia) were used for the reduction of Au(III). BDTA solutions were prepared from the disodium salt (Fison) and the pH was adjusted with 0.10 M hydrochloric acid or sodium hydroxide solutions.

The determination of the concentration of BDTA extracted by PI s was done with Zn(PAR)2 solution (2.00 10^"3 mol L^"1) prepared by dissolving 2nCl₂ (Unilab, Australia) and 4-(2-pyridylazo) resorcinol (PAR) (Aldrich) in borate buffer. The borate buffer at pH 9.30 was prepared by mixing 5.0 10^'2 mol L^"1 solution of NaiB₄0₇ (Chem-supply, Australia) and 0.50 mol L^"1 H3BO3 solution (Ajax Finechem, Australia) in 9:1 ratio by volume. All other chemicals were of analytical reagent grade. Deionized water (18 ΜΩ cm, Millipore, Synergy 185, France) was used for the preparation of all aqueous solutions. 1.2 Instrumentation

The solution concentration of Au(III) was determined by atomic absorption spectrometry (AAS, Hitachi Z-2000 Series Polarized Zeeman atomic absorption spectrophotometer, Japan). The temperature controlled extraction experiments were conducted in conical flasks positioned on a thermostated orbital mixer incubator (Model OM11, Ratek) for temperatures of 20°C and above and on a platform orbital mixer (Model OM6, Ratek) located in a commercial refrigerator (CLEOO, temperature controller E5CN, OMRON) for temperatures below 20°C. Optical microscopy and membrane thickness measurements were conducted with a Motic SMZ-140 stereo microscope (Motic, China) with 60x magnification in combination with a MoticCam 1000 microscope camera (Motic, China). A FEI Quanta 200 F (FEI, USA) scanning electron microscope (SEM) was used for membrane imaging. The concentration of EDTA was determined by visible spectrophotometry (Libra S12, Biochrom). The surface analysis of gold coated PIMs was conducted by a quadrupole laser ablation inductively coupled plasma mass spectrometer (ICP-MS) (Model 810, Varian). The presence of EDTA in PIMs determined by an IR spectrometer (Tensor 27, Broker) equipped with OPUS software, The UV-Visible spectrum of the optimal membrane containing Au nps was obtained with a Perkin Elmer Lamda 950 UV-Vis-NIR spectrophotometer with a 150 mm integrating sphere (Lamda 950, Perkin Elmer). Reflectance mode was used as the Au nps were concentrated on the surface of the membrane which resulted in lack of transparency of the sample. 1.3 Membrane preparation

Aliquat 336 (20% m m), 1-dodecanol (10% m/m) and PVC or CTA (70% m/m) with a total mass of 400 mg were dissolved in a small volume of THF for PVC or dichloromethane for CTA. The solution (8-10 mL) was poured into a glass ring with a diameter of 7.5 cm on a glass plate. The mixture was covered with a filter paper and watch glass to allow slow evaporation of the solvent over 24 h leaving a transparent and flexible circular membrane. The membrane was removed from the glass plate and cut to size (mass 60±3 mg, diameter 3.5 cm, thickness S0±5 μπι).

1.4 Au(III) extraction

Membranes were immersed in 100 mL of 100 mg L ¹ Au(IlT) solutions containing 2.5 M HC1 in conical flasks which were shaken under controlled temperature on a platform orbital mixer at 150 rpm. Samples of the Au(III) solution (0.4 mL) were removed at predetermined time intervals throughout the course of the experiment. The samples were diluted to 8 mL with deionized water and the Au(III) concentration was determined by AAS.

1.5 Reduction of AuQIl)

Auflll) loaded membranes were immersed in conical flasks containing 100 mL solutions of 0.10 mol L^"1 of the individual reducing reagents studied. The flasks were shaken on a platform orbital mixer at 150 rpm for a period of 24 h. The membranes were then removed from the solutions, washed with deionized water and allowed to dry in air. A lxl cm square was cut from the centre of each membrane and mounted on a carbon tab (12 mm) with a pin type SEM mount (12.6 mm). The images of the surface and the cross-section of the membrane were captured using SEM. The procedure outlined above was used in studying the influence of the experimental reduction conditions (i.e., extent of membrane loading with Au(III), concentration of reducing agent, reduction time, solution pH, shaking rate and solution temperature) on the formation of Au nps when EDTA was used as the reducing agent.

1.6 EDTA extraction

The IR spectra of the membranes (250*3 mg) studied were recorded prior to being immersed in conical flasks, each containing.100 mL of 1.0 10° M EDTA solution. The flasks were shaken on a platform orbital mixer incubator for a predetermined period of time at 20°C. if not stated otherwise. Samples of the EDTA solutions (0.1 mL) were removed at different time intervals throughout the course of the experiment. The samples were mixed with 0.1 mL of 2.0 10^"3 M ZnfPARfc solution, diluted to 10 mL with deionized water, and the solution absorbance was measured at 491 nm. The membranes were then rinsed with deionized water to remove trace amounts of EDTA from the surface of the PIMs. The membranes were then allowed to dry in air for 12 h prior to recording their. IR spectra.

The IR spectrum of EDTA in nujol was also recorded on Br discs coated with a thick layer of Na₂EDTA in nujol and then compared with the IR spectra of PIMs recorded prior and after EDTA extraction.

2. Results 2.1 Au(Iil) extraction

Prior to reduction, the PVC and CTA membranes, containing the same amount of Aliquat 336, were exposed to Au(III) solutions until equilibrium between the membrane and the solution had been attained. To ensure that equilibrium had been established the solution concentration of Au(III) was measured at regular time intervals. The corresponding extractions curves (Fig. 1) are typical for PIM-based extraction and are characterized by rapid extraction in the early stages of the process and reaching equilibrium after a few hours. The initial extraction rate is slightly higher for CTA compared to PVC but, as expected, the two membranes achieved the same Au(ffl) loading at equilibrium. The amount of Au(III extracted corresponded to complete leading of the membrane with negligible free Aliquat 336 remaining. The ton-exchange capacity of the membrane was 40 meq g^'1.

After extraction, it was observed that, although both membranes were transparent and yellow in colour, the CTA based PI had an oily surface. This was probably due to the high acidity of the aqueous phase and the fact that CTA is more amenable to decomposition in strongly acidic solutions compared to PVC. Only Au(III) loaded PVC membranes were studied further.

It should be noted that by using an aqueous phase containing a high concentration of chloride (HC1), complete loading of the membrane with AuCU^"1 can be obtained in one extraction experiment. This is in contrast to the work of Kumar et al, where several consecutive extractions were required to achieve high Au(III) loadings of their CTA-based PIMs.

2.2 Reduction of the extracted Au(III)

Au(II0 loaded PVC membranes were immersed in 0.10 mol L^'1 solutions of the following reducing agents: L-ascorbic acid, NaBttt, tri-sodium citrate and Na_∑EDTA (pH 4.5). The solutions were shaken for a period of 24 h. After rinsing with deionized water and drying in air, the membranes were examined visually (Fig. 2). It can be seen that the membranes immersed in L-ascorbic acid and NaBH₄ solutions appear black while membranes immersed in tri-sodium citrate and EDTA have a highly metallic surface appearance and are dark brown and light reddish-brown, respectively.

SEM images of the membranes and their cross sections were captured and are shown in Fig. 3. The images of membranes exposed to L-ascorbic acid, NaBH* and tri-sodium citrate show the formation of Au nanocrystallites with different and irregular shapes, and even sponge like structures in the case of tri-sodium citrate (Figs. 3a, 3b and 3c). The nanocrystalUtes vary in size from 100 nm for L-ascorbic acid to 5 nm for NaBH*. The sponge like network for tri-sodium citrate appears to have a pore size of approximately 20 nm. The SEM images of the cross sections of the membranes exposed to L-ascorbic acid and NaB¾ show that the nanocrystallites are distributed throughout the bulk of the membrane with negligible amounts on the surface (Figs. 3e and 3f). For the tri-sodium citrate treated membrane (Fig.3 (g)), it appears that an appreciable amount of Au nps has accumulated on the surface of the membrane which accounts for the metallic luster.

In addition to being a moderately strong triprotic acid with » values of 3.09, 4.75 and 5.41, citric acid has a slightly larger molar mass (192.1 Da) than the practically monoprotic L-ascorbic aid (176.1) with pK» values of 4.10 and 11.60. The borohydride anion is much smaller than the anions of both acids mentioned above. Therefore it can be expected that the citrate ion-pairs with the Aliquat 336 cations are bulkier than the borohydride and L- ascorbate ion-pairs. This explains the more pronounced formation of Au nps on the membrane surface in the case of reduction with citrate compared to exposing the membrane to the other two reducing agents (Fig. 3) whose ion-pairs can penetrate faster into the membrane. The use of EDTA as a reducing agent has produced the most interesting and unexpected on first glance effects as shown in Figs. 3d and 3h, where all of the Au nps are concentrated in a monolayer on the membrane surface and these are spherical and relatively uniform with the majority around 30 nm in size. Laser ablation I CP-MS analysis of the membrane surface revealed that the membrane was coated with a metallic gold layer which explained the highly metallic appearance of the membrane. The main reasons for this phenomenon could be: (1) the larger size (molar mass of 292.4 Da) and multiple negative charges (pK* values 2.00, 2.69, 6.13, and 10.37) of the EDTA anions which prevented their extraction into the PIM and slowed down the mass transfer of the corresponding ion-pairs within the membrane; and/or (2) the lack of water within the membrane required to facilitate the oxidation of EDTA. An oxidation mechanism of EDTA by Au(III) has been proposed which involves the participation of water, while the oxidation of the other reducing agents studied (i.e. NaBK . L-ascorbic acid, tri-sodium citrate) did not require the presence of water.

To elucidate the nature of the unique properties of EDTA when applied to the reduction of Au(III) loaded PIMs the extraction of EDTA into these PI s was studied. Preliminar extraction experiments involving the exposure of PIMs to EDTA solutions for 24 h were conducted and the membrane IR spectra were recorded. An absorbance band at 1632 cm'¹, typical for 0=0 vibration in EDTA, was observed. This band did not appear in the spectra of the fresh membranes (Fig. 4) but was present in the IR spectrum of EDTA in nujol. These preliminary experiments were followed by extraction experiments from EDTA solutions at different pH (4.0, 4.5, 6.0, and 8.0). The corresponding extraction curves (Fig. 5) confirmed the IR results indicating that EDTA could be extracted into the membrane and equilibrium was reached after 2 h. More EDTA was extracted from solutions at pH 4.0 and 4.5 than at pH 6.0 and 8.0. This can be explained by the fact that at the lower pH values EDTA is present as a mixture of mono- and divalent anions which are extracted more readily in PIMs than the trivalent anion which is the dominant EDTA species at the higher pH values.

The extraction of EDTA into PIMs suggested that the main reason for the insignificant reduction of Au(III) within the membrane and the formation of the surface Au nps layer was not due to the inability of EDTA to enter membrane but alternatively is most due to the lack of sufficient water within the membrane because of its hydrophobic nature.

The main factors influencing the reduction of Au(IH) on the surface of Aliquat 336/PVC PIMs by EDTA was studied in more detail.

2.3 Effect of pH

Depending on solution pH, EDTA can exist in several ionized species with charges from - I (acidic conditions) to -4 (alkaline conditions). The charge of the EDTA ion will determine its ability to form an ion-pair with the Aliquat 336 cation and to be transported within the membrane. Also, the actual Au(III) reduction mechanism involving EDTA could be iniluenced by solution pH. Therefore it was of interest to investigate the effect of solution pH on the formation of Au nps. In the associated experiments Au(III) loaded PIMs were immersed in 0.10 mol L^~' EDTA at pH 4.0, 4.5, 6.0, or S.O. The photographic images of the membranes, presented in Fig. 6, show that the colour of the membranes change from yellow at pH 4 to light metallic reddish-brown at pH 4.S and darker metallic brown at pH 6.0 and pH 8.0.

The SEM images of the surface and cross-section of the membranes studied are shown in Fig. 7. As the pH was increased from 4.0 to 6.0, the amount of Au nps on the membrane surface also increased and this was accompanied by a decrease in particle size from approximately 200 nm at pH 4.0 to 30 nm at pH 4.5 and 20 nm at pH 6.0 (Figs. 7a-7c). Also, by increasing pH less Au nps were formed within the bulk of the membrane, i.e. at pH 4.5 and 6.0 practically all of the Au nps were located on the membrane surface (Figs. 7f and 7g). The particle si2e further decreased to approximately 10 nm at pH 8 (Fig. 7d). However, at this pH 8 the amount of Au nps formed n the membrane surface also decreased (Fig. 7d) while that within the membrane (Fig. 7h) slightly increased in comparison with the membrane reduced at pH 4.5 and 6.0. The decreased amount of Au nps formed could be due to ion-exchange between the hydroxide anion in the solution and the AuCU^'1 anion in the membrane.

On the basis of the results outlined above the optimal pH for the formation of a surface layer of Au nps was selected as 6.0. The size distribution of the Au nps at pH 6.0 was obtained from the SEM image of the membrane and the corresponding histogram is shown in Fig. 8. It can be seen that the size range is quite narrow, with the average size being 20 nm.

A reflectance spectrum in the visible region (Fig. 9) of the surface of the pH 6.0 membrane was obtained and showed only one broad peak with X_max for the surface plasmon band (SPB) of 530 nm. This is consistent with the presence of Au nps with near-spherical morphology 2.4 Effect of EDTA concentration

The effect of the EDTA solution concentration on the formation of Au nps was studies by exposing Au(III) loaded PIMs to 0.05, 0.10, 0.15, and 0.20 M EDTA solutions at pH 6 for 24 h. The photographic images of the membranes (Fig. 10), showing similar dark metallic brown colour, suggest the formation of surface layers of Au nps at all EDTA concentrations studied.

The SEM images of the surface and cross-section of the membranes studied (Fig. 11) show insignificant amount of Au nps in the bulk of the membranes regardless of the concentration of EDTA. The coverage of the membrane surface with Au nps increased as concentration of EDTA increased from 0.05 to 0.10 mol L^'1 and the membranes appeared fully covered for concentrations higher than 0.10 mol L^"1. The average size of the Au nps decreased from 40 nm at 0.05 mol L^"1 to 20 nm at 0.10 mol L^'1 and 14 nm at 0.15 and 0.20 mol L^"1 of EDTA. Therefore, these results suggested that the size of Au nps could be controlled by altering the concentration of EDTA. >

2.5 Effect of Au(IH) extraction time

It has been reported that the size and shape of Au nps can also be controlled by varying the ratio between the concentrations of the reducing/stabilizing reagents and Au(III). Therefore PIMs with different amounts of Au(HI) were prepared by immersing the membranes in 100 mg L'¹ Au(lH) solutions for 0.5, 3, 6, 9, or 12 h which resulted in the extraction of 1.71, 3.36, 4.00, 4.07 and 4.67 mg Au, respectively. The membranes were then exposed to 0.10 M EDTA solutions at pH 6 for 24 h. The colour of the membranes changed from light metallic reddish-brown to darker metallic brown as the Au(III) extraction time increased (Fig. 12) suggesting that the amount of Au on the membrane surface depended on the extraction time.

The surface and cross-section SEM images did not show the presence of Au nps within the bulk of the membranes (Figs. 13f-13j) while the amount of surface Au nps increased with the extraction time. However, the average $i-e of Au nps (i.e. 20 nm) remained constant. 2.6 Effect of reduction time

Loaded with Au(III) membranes were exposed to 0.10 M EDTA solutions at pH 6 for different periods of time. The colour of the membranes changed from yellow to metallic dark reddish-brown colour as the time of exposure increased from 1 to 24 h (Fig. 14).

The SEM images of the surface and cross-section of the membranes studied (Fig. 1 ) showed the absence of Au nps within the bulk of the membranes and a gradual increase of the amount of surface Au nps with increasing reduction time. The average size of the Au nps in all experiments was independent of the reduction time and was close to 20 nm.

2.7 Temperature effects

The formation of Au nps was studied by immersing Au(HI) loaded PI s into 0.10 M EDTA solutions (pH 6) at different temperatures (10, 20, 30, 40 and 50 °C). As with the previous experiments, the cross-section SEM images of the membranes showed insignificant amounts of Au nps within the bulk of the membranes. All the membranes reduced at 20 °C or higher temperatures appeared to be highly metallic reddish brown in colour. The membranes reduced at 10 °C were transparent and of light yellow colour thus suggesting a rapid decrease in the formation of surface Au nps. This conclusion was confirmed by SEM surface images (Fig. 16a) which showed small amount of surface Au nps of an average size of 14 nm. The small amount and size of the Au nps was probably due to the decreased mass transfer rate of the Au(III) ion-pair from the bulk of the , membrane towards its surface. A substantially larger amount of surface Au nps were formed at 20 and 30°C with an average particle size of 20 and 30 run, respectively (Figs. 16b and 16c). At 40°C and above, the membranes became unstable that Au nps formed were not uniform both in terms of size and distribution.

2.8 Effect of shaking rate

The effect of the shaking rate on the formation of Au nps was studied by immersing Ανι(ΠΙ) loaded PIMs in 0.10 M EDTA solutions at pH 6 and varying the shaking rate from 0 to 150 rpm (0, 50, 75, 100, and 150 rpm). Irrespectively of the shaking rate all membranes appeared to be of metallic reddish brown colour and the SEM images of their cross-scctions showed insignificant Au nps formation within the bulk of the membranes. The shaking rate affected strongly the size and distribution of the Au nps on the membrane surface. The Au nps clumped together to form non-uniformly distributed Au nanocrystallites at shaking rates of 75 rpm or lower (Fig. 17a). However, at shaking rates of 100 rpm and ISO rpm, the Au nps became more evenly distributed and of more uniform size, i.e. average size of 24 and 20 nm, respectively (Figs. 17b and 17c).

The effect of the shaking rate on the average size of surface Au nps is shown in Fig. 18. These results suggest that the shaking rate, which governs the overall mass transfer of EDTA from the bulk of solution towards the membrane surface, has a pronounced effect on the Au nps size. Faster mass transfer results in smaller size Au nps. The deviation from this trend for 0 rpm (i.e., no shaking) when the size of the Au nps was smaller than the average value for SO rpm was probably due to the completely different hydrodynamic conditions characterized by lack of mechanical mixing and very slow overall EDTA mass transfer governed by Fickian diffusion only. This resulted in a smaller amount of Au(III) being reduced on the membrane surface to form relatively small-size Au microcrystallites.

The results outlined above suggest that the shaking rate at which Αυ(ΙΠ) reduction takes place can be used to control the size of the Au nps in a relatively wide range.

3. Monitoring of mercury vapour in air using a passive sampler fitted with a gold nanoparticle-coated polymer-inclusion membrane

Gold nanoparticles (Au nps) were synthesized on the surface of a polymer inclusion membrane (Au-PI ) and incorporated into a passive sampler. The Au-PIM has Strong adsorption ability and high capacity for metallic mercury (Hg°) accumulation from the air by forming an amalgam. For calibration of the passive sampler, Hg° was generated on-line by reducing Hg** with NaBK,. The resultant Hg vapour was purged with air and swept through an adsorption bottle (HDPE) f 1 liter in capacity in which the passive sampler was deployed. A small electric fan was used inside the bottle to generate disturbance and improve the diffusion of Hg vapour. The Hg vapour concentration in the bottle was reasonably constant through the experimental period. After a particular period of time (2-7 days), Hg accumulated on the passive sampler was stripped with HNO3 and determined with atomic fluorescence spectrometry (AFS) and the amount of Hg accumulated was correlated with the concentration of this metal in the gaseous phase (air) in contact with the Au-PIM.

Experimental results are given in Figure 19 for the accumulation of Hg over 2-7 days. It is clear that a constant accumulation rate can be obtained The accumulation can be reduced, if necessary, by the use of an inert membrane to act as a passive barrier to slow the rate of transport of mercury from the gaseous phase to the Au nps. Figure 1 also demonstrates that the concentration of elemental mercury in air as low of 0.5 ng L can be detected.

Claims

THE CLAIMS:

1. A method of preparing gold nanoparticles on the surface of a polymer inclusion membrane (PIM) including the steps of: i. mixing a quaternary ammonium salt, plasticizer and or modifier, polymer, and organic solvent for a time and under conditions suitable for preparing a homogenous solution and then casting a PIM from said solution; ii. contacting the PIM with an acid solution comprising a Au(III) species and extracting at least a portion of the Au(HI) species from the solution into the PIM to form a PIM loaded with said Au(III) species; and

iii. treating the loaded PIM from step (ii) with a Au(III) reducing agent characterised in that the agent cannot readily reduce the Αυ(ΙΪΙ) species within the PIM, leading to the formation of gold nanoparticles on the surface of the PIM.

2. A method according to claim 1 wherein the reducing agent is selected from a compound of formula (Π): '

wherein:

m is an integer from 1 to 6 (preferably 2); and

n to q are independently selected from an integer from 1 to 4 (preferably n, o, p, and q are all l),

or a salt thereof.

3. A method according to claim 1 or claim 2 wherein the reducing agent is EDTA.

4. A method according to any one of claims 1 to 3 wherein the reducing step (iii) is conducted at a pH of 5.S-6.S.

5. A method according to any one of claims 1 to 4 wherein the solution comprising the Au(ni) species is an aqueous acid solution, which has a concentration of 2 - 4 M.

6. A method according to claim 5 wherein the aqueous acid solution is a HC1 solution.

7. A method according to any one of claims 1 to 6 wherein the Au(III) species is AuCl₄\

8. A method according to any one of claims 1 to 7 wherein the PIM is characterised with a polymer : quaternary ammonium salt : plastici2er/modifier ratio of about 7:2:1 to

16:3:1 (% m m).

9. A method according to any one of claims I to 7 wherein the PIM comprises about 70% m m PVC, about 20% m/m Aliquot 336 and about 10% m/m 1-dodecanol.

10. A method according to any one of claims 1 to 9 wherein at least 90% of the Au(III) is reduced and forms a gold nanoparticle monolayer on the surface of the PIM.

11. A gold nanoparticle monolayer obtained from a method according to any one of claims 1 to 10.

12. A PIM comprising gold in the form of Αιι(ΙΠ) and metallic gold nanoparticles wherein at least 80 % of the gold is in the form of metallic gold nanoparticles which are present on the surface of said PIM.

13. A PIM according to claim 12 wherein more than 90% of the gold nanoparticles are present on the surface of said PIM.

14. A PIM according to claim 13 wherein more than 95% of the gold nanoparticles are present on the surface of said PIM.

15. A PIM according to any one of claims 12 to 14 wherein the PIM is a PVC polymer based PIM.

16. A PIM according to any one of claims 12 to 15 wherein the gold nanoparticles are spherical with an average size of about 15 to 30 nm.

17. A PIM according to claim 16 wherein the gold nanoparticles form a monolayer on the surface of a said PIM.

18. A PIM according to any one of claims 12 to 17 wherein the PIM comprises PVC, quaternary ammonium salt, and a plasticizer and/or modifier.

1 . A PIM according to claim 18 wherein the PIM comprises the following ratios of components: PVC : quaternary ammonium salt : plasticizer/modifier from about 7:2:1 to

16:3:1.

20. A PIM according to claim 18 or claim 19 wherein the PIM comprises about 70% m/m PVC, about 20% m/m Aliquat 336 and about 10% m/m 1 -dodecanol.

21. A method of monitoring mercury in an air sample, said method comprising the step of contacting an air sample with a passive sampling device which comprises a PIM according to any one of claims 12 to 20, wherein the air sample is exposed to the PIM.

22. A passive sampling device incorporating a PIM according to any one of claims 12 to 20 for use in the monitoring of mercury in air.