TMCnet News
The Oxygen Reduction Reaction in Ferrofluids: Towards Membrane-less and Spill-less Gas Sensors [ChemPlusChem](ChemPlusChem Via Acquire Media NewsEdge) In this study, a "membrane-less and spill-less" gas-sensing device has been evaluated for the electrochemical detection of oxygen. Iron oxide magnetic nanoparticles were prepared by chemical co-precipitation and used to prepare an aqueous ferrofluid. The iron oxide nanoparticles were subsequently stabilised and passivated with a cationic polymer, namely, poly(diallyldimethyl ammonium chloride). The resulting ferrofluid was evaluated as an electrolyte for the analytical quantification of oxygen on screen-printed carbon electrodes. An applied magnetic field immobilised the ferrofluid electrolyte in place to result in a "membrane-less and spill-less" ferrofluid-based gas sensor. The polymer poly(diallyldimethyl ammonium chloride) was found to result in an apparent enhancement in the electrocatalysis of the system towards the oxygen reduction reaction. Furthermore, as the strength of the applied magnetic field was increased, the oxygen reduction current also increased owing to the response of the polymer-coated nanoparticles. The oxygen reduction current was linear from 0 to 100 % oxygen content. Keywords : electroanalysis * iron * magnetic properties * nanoparticles * sensors Introduction Iron oxide nanoparticles, including maghemite (g-Fe2O3) and magnetite (Fe3O4), are an important class of magnetic materials that have long been a topic of interest owing to their low toxicity, easy synthesis, biocompatibility and unique superparamagnetic properties.[1-6] Typically, iron oxide based nanoparticles have been increasingly used in a broad range of applica- tions, for example, as contrast agents in magnetic resonance imaging (MRI),[6, 7] targeted drug delivery,[6] electrochemical and biosensors[8-10] and as electrode materials for semiconductor gas sensors.[11-18] In the latter application, iron oxides (g-Fe2O3 and a-Fe2O3) with n-type semiconducting properties have been employed as thin-film or thick-film active sensing layers in gas sensors for the detection of H2,[11, 12] CO,[13] NO2,[14] H2S,[15] O2[16] and ethanol.[17,18] Iron oxide (para)magnetic nanoparticles are also widely used to prepare ferrofluids.[5, 19-21] Ferrofluids, also known as magnet- ic fluids, are stable colloidal dispersions of small, single-domain magnetic nanosized particles (typically with an average diame- ter of ca. 10-20 nm) suspended in either polar or non-polar liquid carriers, for example, water, benzene and toluene.[1, 2, 20-22] Characteristically, ferrofluids possess both magnetic and fluid properties, and their rheological behaviours can be magnetical- ly controlled by means of the magnetic field.[2, 21, 23] Therefore, ferrofluids can change almost instantaneously from a liquid to a solid-like state and vice versa in the presence and absence of an applied magnetic field, respectively.[2, 21] Several synthetic methods have been developed and used to prepare highly stable and monodisperse iron oxide magnet- ic nanoparticles ; however, the chemical co-precipitation of iron(II) and iron(III) salts generally appears to be the simplest and most efficient technique to synthesise iron oxide nanopar- ticles (typically on a large scale) with reported diameters from 2to17nm.[1-3, 21, 24] The surface properties of the iron oxide magnetic nanoparticles are the key factors that determine the colloidal stability of ferrofluids.[1, 3] Stabilisation of magnetic nanoparticles can be simply achieved by coating the surface of nanoparticles with polymers or surfactants, which provide re- pulsive forces between particles, keeping them in a stable col- loidal state. The oxygen reduction reaction is of extensive importance, especially with regard to electrochemical energy systems.[25] It is also commonly investigated to evaluate novel electrochemi- cal gas-sensor systems. Electrochemical gas sensors make up the majority of the gas-sensor market,[26] with development over the last few decades largely based upon the classic Clark electrode design,[27] whereby a working electrode placed in contact with a layer of electrolyte (usually aqueous) is covered by a gas-permeable membrane. This membrane acts as a barri- er, preventing the liquid electrolyte from spilling out of the sensor, partially reduces electrolyte evaporation and partially impedes gas molecules from reaching the underlying electrode surface.[28, 29] Once the target gas undergoes electrolysis at the sensing electrode, it generates a current that is proportional to the gas concentration.[29] These membrane-based gas sensors can be used effectively (within a narrow range of ambient con- ditions), although they have some drawbacks inherently caused by the gas-permeable membrane.[28, 29] Membrane-free amperometric gas sensors have been report- ed that utilise microlitre volumes of ionic liquid as electro- lyte.[28-31] These systems have also been expanded to droplets on screen-printed electrodes (SPEs).[32, 33] SPEs have been widely used in electrochemical sensing applications owing to their low cost, disposable sensing surface and their regular combination with simple and portable electrochemical equip- ment. However, at all stages, the above membrane-less sensors must be kept perfectly level and protected from rapid gas flows or the liquid droplet will roll off the sensing surface. Ad- ditionally, Compton et al.[29] reported that some ionic liquid layers were unstable, as they changed shape or split into multi- ple droplets owing to the uptake of atmospheric moisture. In this study, we have developed a "membrane-less and spill-less" electrochemical gas sensor. Aqueous colloidal disper- sions of iron oxide magnetic nanoparticles were prepared by a co-precipitation method to form an aqueous ferrofluid. The analytical performance and utility of this ferrofluid (containing 0.1 m HCl as electrolyte) for oxygen gas sensing was then stud- ied on screen-printed electrodes (SPEs). The cationic polymer poly(diallyldimethyl ammonium chloride) (or poly(ac)) was added, whereby it played a threefold role : 1) further stabilising the iron oxide nanoparticle colloid, 2) passivating the iron oxide nanoparticles to prevent unwanted electrochemical sig- nals and 3) interacting with the SPE electrode surface to result in an apparent enhancement in the electrocatalytic reduction of oxygen. An applied magnetic field resulted in immobilisa- tion of the ferrofluid droplet, resulting in a spill-less ferrofluid that could still be used as a gas sensor for oxygen. Interesting- ly, the applied magnetic field actually resulted in an enhance- ment in the oxygen reduction current, likely owing to the poly(ac)-coated iron oxide nanoparticles increasing the amount of poly(ac) near the electrode surface and thus enhancing the apparent electrocatalysis of the system with respect to oxygen reduction. Result and Discussion Characterisation of the iron oxide nanoparticles The size and the morphology the iron oxide nanoparticles (Fe2O3 NPs) synthesised and used in this study were character- ised by transmission electron microscopy (TEM). The aqueous ferrofluid dispersions were synthesised as described in the Ex- perimental Section, diluted to a concentration of 0.1 % w/w in water, deposited on a TEM grid and dried under ambient con- ditions. Figure 1a displays a typical TEM image of the as-syn- thesised Fe2O3 NPs. The results demonstrate that the particles are spherical in shape and moderately polydisperse, which is not unusual for nanoparticles prepared by the chemical co-pre- cipitation method, as previously reported in the literature.[2, 21] The primary particle sizes measured from the image were found to range from 8 to 20 nm, consistent with the observa- tions of others.[2,21] Stability of the ferrofluid in an applied magnetic field The stability of an aqueous suspension of Fe2O3 NPs (the ferro- fluid) in an applied magnetic field was first investigated by cyclic voltammetry (CV). In all cases the ferrofluid was com- posed of 10 % w/w Fe2O3 NPs in water, which was spiked so as to contain 0.1 m HCl to act as electrolyte. Figure 2 shows the cyclic voltammograms of the acidified ferrofluid (50 mL) drop cast on a platinum screen-printed electrode (PtSPE). The CVs were recorded in the potential range of about + 0.1 and + 0.7 V versus Ag in the absence and presence of the applied magnetic field. Also shown are images that highlight the be- haviour of the ferrofluid in the absence and presence of the external magnetic field. In both cases, the reduction and oxidation peaks of the FeII/ FeIII redox couple could be clearly observed at about + 0.31 and + 0.45 V versus Ag, respectively. The direct electrochemical interrogation of nanoparticles in solution is an area of increas- ing significance with regard to fundamental investigation, envi- ronmental monitoring and so forth.[34, 35] The electrochemical response of the Fe2O3 NPs were then monitored as a function of time for up to 2 h in the absence and presence of the exter- nal magnetic field. Figure 2a shows that when there was no applied magnetic field, the reduction and oxidation peaks that correspond to the FeII/FeIII redox process remained largely un- changed (no significant shift in either peak potentials or peak currents observed) for at least 2 h. In addition, we could ob- serve that with no magnetic field applied, the water-based Fe2O3 NPs indeed remained extremely stable under acidic con- ditions (at pH [asymptotically =] 2.75) for at least several months (Figure 2c). Generally, water- and oil-based ferrofluids are reported to be stable at pH values below 5 (acidic ferrofluids) or above 8 (al- kaline ferrofluids).[1,20,24,36] Once the external magnetic field was present, these magnet- ic Fe2O3 NPs were strongly attracted towards the applied mag- netic field and could be easily magnetically manipulated (Fig- ure 2d). After an hour under the applied magnetic field, it could be clearly observed that the magnetic nanoparticles started to settle and became increasingly unstable (Figure 2e). The voltammetric behaviour of the FeII/FeIII redox couple was monitored periodically when the external magnetic field was applied continuously for 2 h, and the resulting CVs are shown in Figure 2b. As the length of time of exposure to the magnet- ic field increased, both cathodic and anodic peak currents at- tributed to the FeII/FeIII redox couple increased constantly (ca. 27 mA per hour). This is likely due to the fact that more and more Fe2O3 NPs aggregated and precipitated over time onto the screen-printed electrode where the magnetic field exist- ed,[1, 3] resulting in the increase in the peak currents of the FeII/ FeIII redox couple. Alternatively, an increase in the local concen- tration of Fe2O3 NPs in the vicinity of the electrode would in- crease the probability of nanoparticles interacting with the electrode surface. A plot of the oxidative peak current of the FeII/FeIII redox couple against the time of the magnetic field ex- posure also shows a good linear response (shown as an inset in Figure 2b), indicating the moderate instability of these water-based Fe2O3 NPs in the presence of an external magnetic field. Stabilisation of the Fe2O3 NP colloid with polymer coating The stability of colloidal dispersions (against aggregation) is significantly influenced by the equilibrium between the attrac- tive and repulsive forces that exist between particles.[1, 3,5, 20,37, 38] In general, the repulsive interactions (both electrostatic and steric repulsion) between particles play an important role in improving the stability of the colloidal ferrofluids.[1, 3] Stabilisa- tion of these magnetic nanoparticles could therefore be ach- ieved by either controlling the surface charge of the parti- cles[1, 24] or coating the surface of the nanoparticles with poly- mers and surfactants.[1,3,4,19,39] Recently, several coating methods have been developed (in- cluding in situ coatings and post-synthesis coatings) and vari- ous coating materials have been employed to coat Fe2O3 mag- netic nanoparticles to yield stable ferrofluids.[1-4, 19, 21, 39] General- ly, polymers and surfactants could be either chemically teth- ered to or physically adsorbed onto the magnetic nanoparti- cles, forming a single or double layer in which the repulsive forces (with steric repulsion being predominant) were generat- ed to balance the magnetic and van der Waals attractive forces that acted on the nanoparticles, maintaining the stability of these nanoparticles under the influence of gravity and an ex- ternal magnetic field.[1, 40] It has also been reported in the litera- ture that the suitable polymers used for coating usually con- tain hydroxyl, amino, phosphate, sulfate and carboxylic acid functional groups that can actively bind to the surface of iron oxide magnetic nanoparticles.[1, 4] In this study, the synthesised Fe2O3 NPs were deliberately coated with a range of polymeric stabilisers, including anionic (poly(ss) and poly(ss-Na)) and cationic (poly(ac)) polymers (Figure 3, with average molecular weights of about 75 000- 200000 gmol-1 to obtain the colloidal dispersion of magnetic nanoparticles that are highly stable against the applied mag- netic field. For the modification, a 20 % w/w solution of the polymers in water was added to a 20 % w/w aqueous suspen- sion of Fe2O3 NPs (to result in 10 % w/w of each). The mixture was then sonicated for 30 min at room temperature to obtain the polymer-coated ferrofluid, which appeared more viscous relative to the uncoated one. The stability of the resulting polymer-coated Fe2O3 NPs under the applied magnetic field was investigated next. When the external magnetic field was applied, only the aqueous dispersion of Fe2O3 NPs coated with cationic polymer (poly(ac)) was observed to be very stable for at least 24 h, demonstrating that the nanoparticles could be favourably sta- bilised by this positively charged polymer through the electro- statically charged layer between particles and the steric repul- sion between the polymer chains, as previously described.[1, 3] The Fe2O3 NPs coated with anionic polymers (both poly(ss) and poly(ss-Na)) ; however, settled very quickly within a few mi- nutes in the presence of the applied magnetic field. This is due at least in part to the fact that when the anionic polymers were added to the inherently acidic ferrofluid (with pH [asymptotically =] 2.75), protonation of the negatively charged sulfonate group on these anionic polymers would have occurred extensively in the system, resulting in neutral polymeric molecules that could no longer electrostatically stabilise or balance the magnetic attrac- tive force applied to the ferrofluids. Steric repulsion appears to play a negligible role in these systems. The resulting poly(ac)-coated Fe2O3 NPs were subsequently characterised by means of TEM and CV. Figure 1b displays a typical TEM image of about 0.1 % w/w Fe2O3 NPs coated with about 0.1 % w/w poly(ac). In comparison with Figure 1a, no significant change in the primary particle size or shape could be observed for the polymer-coated Fe2O3 NPs, suggesting that the morphology of Fe2O3 NPs remained largely unaffected by the polymeric stabilisation. However, aggregation of the polymer-coated Fe2O3 NPs could be seen on the TEM grid, which was not significant in the actual ferrofluid (see the elec- trochemical characterisation of the system below), and could have resulted from capillary action during the drying process on the grid.[2] Next, the electrochemical behaviour of the poly(ac)-contain- ing ferrofluid (10% w/w Fe2O3 NPs, 10 % w/w poly(ac), 0.1 m HCl) was examined by CV. Figure 4 shows the overlaid CVs ob- tained from bare Fe2O3 NPs in the acidified ferrofluid (dotted line) and poly(ac)-coated Fe2O3 NPs in the acidified ferrofluid (solid line). As seen in Figure 4, no reduction or oxidation fea- tures could be observed for the poly(ac)-coated Fe2O3 NPs, whereas for the uncoated Fe2O3 NPs both cathodic and anodic peaks that correspond to the FeII/FeIII redox couple could be clearly seen. Further voltammetric measurements of the bare and the poly(ac)-coated Fe2O3 NPs (in the ferrofluid) were also conducted over a wider electrochemical potential window and the resulting CVs are shown as an inset in Figure 4. The ab- sence of electrochemical activity over the whole electrochemi- cal window demonstrates that the polymer-coated Fe2O3 NPs were no longer electrochemically active as a result of surface passivation by the coating process. The oxygen reduction reaction should be clearly observed at platinum surfaces under acidic conditions.[32] However, in this study only small and ill-defined oxygen reduction features could be observed at the PtSPEs, regardless of the presence or absence of both polymer and ferrofluid. In addition, these re- duction features disappeared in consecutive scans, demon- strating further electrode passivation. Silvester et al.[32] investi- gated oxygen reduction using DropSens PtSPEs in ionic liquids and found that oxygen reduction features on the PtSPEs were significantly different from what was observed on the conven- tional Pt macroelectrodes. This was attributed to reaction of the electrogenerated superoxide with the compounds in the PtSPE ink/paste, resulting in a change in the PtSPE surface.[32] Similar experiments were also performed at DropSens carbon SPEs pre-modified with bismuth oxide and nickel oxide to evaluate their suitability as an electrode substrate. Both dis- played stable oxygen reduction features and identical results were observed at both SPE materials, indicating that oxygen reduction was likely occurring at the carbon electrode surface ; the incorporated metals result in significantly lowered back- ground currents relative to unmodified carbon SPEs.[33] The nickel oxide modified (NiO-modified) electrodes were used throughout, and the resulting cyclic voltammograms are shown in Figure 5. In comparison to the voltammetric re- sponse of ferrofluid (dashed line), no comparable voltammetric signal could be observed for the poly(ac)-containing ferrofluid (solid line), highlighting the electrochemically inactive behav- iour of the poly(ac)-coated Fe2O3 NPs, consistent with that ob- served earlier at the PtSPE. However, a newly formed irreversi- ble reduction peak with a peak potential at about -0.5 V versus Ag (solid line) could be clearly seen for the polymer- coated Fe2O3 NPs at the NiO-modified SPE that corresponded to oxygen reduction (see below). The oxygen reduction feature was observed at about -0.5 V versus Ag in both 10 wt % poly(ac)-containing acidified ferro- fluid (inset in Figure 5, solid line) and also for 0.1 m HCl that contained 10 wt % poly(ac) (blue dotted line in inset). As near identical electrochemical behaviour could be observed for both the 10 % w/w poly(ac)-containing acidified ferrofluid and just 10 % w/w poly(ac) in 0.1 m HCl (no Fe2O3 NPs), only data for one system is presented here. Figure 6 displays the cyclic voltammograms of 10 % w/w poly(ac) solution in 0.1 m HCl under an ambient (black dotted line), an oxygen-free (blue solid line) and an oxygen-saturated (red solid line) environ- ment. These experiments demonstrate that the large irreversi- ble reduction peak at about -0.5 V versus Ag corresponds to the electrochemical reduction of oxygen, likely to H2O2 owing to the non-platinum electrode.[41] Interestingly, the reduction of oxygen was independent of the presence of Fe2O3 NPs but was clearly altered by the addi- tion of the poly(ac). The reduction of oxygen in 0.1 m HCl at the NiO-modified SPE (in the absence of both Fe2O3 NPs and poly(ac)) is shown as an inset in Figure 6 (orange solid line) and compared against the same system that contained 10 % w/w poly(ac) (black dashed line). Addition of the polymer in- creased both the peak current and shifted the peak potential towards more positive values, which indicated an apparent electrocatalytic effect upon the oxygen reduction reaction. Al- though the peak shift is not absolutely quantitative owing to the use of the quasi-reference electrode of the SPE, the in- crease in peak current was reproducible and observed repeat- edly. Apparent electrocatalysis has recently been discussed in detail[25] and can occur when a surface-modifying layer such as a polymer alters the local conditions at the electrode surface (such as pH or ionic strength), and such layers can also possess oxygen solubility higher than that of the bulk solution.[42] Effect of the applied magnetic field on the reduction of oxygen in the poly(ac)-ferrofluid The effects of an applied magnetic field on various electro- chemical processes have been comprehensively studied and discussed in the literature.[43-46] Generally, the effects an exter- nal magnetic field can have on the electrochemical reactions can be divided into three categories : those that relate to elec- trolyte properties, mass transport and electrode kinetics.[43-46] In this study, the influence of the external magnetic field on the electrochemical reduction of oxygen in the poly(ac)-ferro- fluid was further investigated with regard to the strength of the magnetic field applied. Once the magnetic field was ap- plied to the ferrofluid, the Fe2O3 NPs were actively controlled, resulting in an immobilised droplet of ferrofluid, which was at- tached strongly to the surface of the SPE. Figure 7 displays (a) an image of a magnetic-field-immobilised ferrofluid droplet on a NiO-modified SPE with magnets held underneath (both drop- let and magnets immobilised purely by their interaction with each other with no other adhesives or similar in place) ; and (b) overlaid cyclic voltammograms of the poly(ac)-ferrofluid, ex- amined as a function of magnetic field strength at an NiO- modified SPE in an oxygen-saturated system. Note that the magnetic field strength was increased by varying the number of the magnets applied to the ferrofluid, with each magnet quoted to have a magnetic flux density of about 2229 gauss. When magnets were applied, measurements could be made at any inclination or angle ; without the magnets in place the electrode had to be kept level at all times or the droplet would move away from the electrodes. As seen in Figure 7b, the cathodic peak current for oxygen reduction increased with increasing magnetic field strength. This is believed to be due to the fact that when the magnetic field strength increased, more magnetic nanoparticles were lo- cated close to the surface of the electrode and therefore more poly(ac) was present, further enhancing the apparent electro- catalytic effect of poly(ac). A plot of the cathodic peak current (Ipc) versus the strength of the applied magnetic field (inset in Figure 7b) also shows a linear relationship, confirming the sig- nificant positive effect of the external magnetic field on the electrocatalytic reduction of oxygen. Magnetic nanoparticles have been dispersed to collect ana- lytes and magnetic fields have typically been used to bring them to an electrode surface.[10] Polymer, liquid and other sur- face modifiers have also been used to increase the local con- centration of analytes at the electrode surface and induce ap- parent electrocatalysis.[25, 42] This is believed to be the first known application of magnetic nanoparticles that can increase the quantity of a surface modifier proportional to an applied magnetic field, and thus induce apparent electrocatalysis. A scan-rate study in the poly(ac)-ferrofluid (10 to 200 mV s-1) in the presence of the applied magnetic field dem- onstrated that the current increased with scan rate (not shown). A plot of the log of cathodic peak current (Ipc) versus the log of scan rate also shows a linear response (R2 = 0.996) with a slope of 0.51, indicating a diffusion-controlled rather than adsorption-controlled process.[47] As the magnetic strength of the field increased, a secondary oxygen reduction peak could be increasingly observed at about -0.6 V. For the two-electron reduction of oxygen to hy- drogen peroxide, either one 2 e- or two 1e- reduction peaks can be observed, depending upon the pH, electrolyte and the relative rates of the first and second electron-transfer con- stants.[41, 48,49] In this study, this feature is tentatively attributed to localised pH changes in the polymer layer (larger amounts of oxygen reduction in the thin polymer layer result in more significant pH changes and proton depletion). The reproducibility of the poly(ac)-ferrofluid in the presence of the magnetic field was found to be good. When the experi- ment demonstrated in Figure 7b was performed sequentially three times (on the same SPE) the relative standard deviation in the oxygen reduction peak heights was only 4.4 %. Quantitative electrochemical determination of oxygen in the poly(ac)-ferrofluid in an applied magnetic field The electrocatalytic detection of oxygen in the poly(ac)-ferro- fluid was next examined quantitatively (with and without an applied magnetic field) by varying the concentration of oxygen from 0 to 100 % (calculated on the basis of the relative flow rates of oxygen and nitrogen gases being mixed in a gas- mixing system). Although an initial response was observed within seconds (as expected for an aqueous-based sensor com- posed of a small droplet) all samples were purged for 10 min to ensure equilibration. No significant difference was observed in response and equilibration time as an effect of the magnetic field strength. Figure 8 displays CVs recorded in poly(ac)-ferrofluid drop cast on an NiO-modified SPE, which was then exposed to dif- ferent concentrations of oxygen (0, 33, 50, 67 and 100 %) before being analysed in the absence and presence of the ap- plied magnetic field with increasing strength by using 4 and 6 magnets, respectively. In all cases, as the concentration of oxygen increased, the oxygen reduction peak current increased with no shift in peak potential observed. Moreover, when the applied magnetic field strength was increased with 0, 4 and 6 magnets applied to the poly(ac)-ferrofluid (from Figure 8a, b and c, respectively), a larger cathodic peak current that corresponded to the reduc- tion of oxygen could be observed (at the same percentage concentration of oxygen), consistent with the previous investi- gations described above. In addition, plots of oxygen-reduc- tion-peak current versus concentration of oxygen in the ab- sence and presence of the applied magnetic field with increas- ing strength were found to be linear. The highest magnetic field strength applied here was relatively more sensitive to the reduction of oxygen, highlighting the ability of the applied magnetic field to not only immobilise the ferrofluid droplet onto the SPE (to create a membrane-less and spill-less gas sensor), but also enhance the (apparent) electrocatalytic reduc- tion of oxygen in the poly(ac)-ferrofluid. The system also dis- plays linearity from 0 to 100 % oxygen content. Therefore, it can be concluded that a colloid of poly(ac)-coated Fe2O3 NPs (as a poly(ac)-ferrofluid) represents a simple "membrane-less, spill-less" gas-sensor assembly, which is both achieved and en- hanced by the presence of an applied magnetic field. Conclusion An aqueous colloidal suspension of Fe2O3 nanoparticles (a fer- rofluid) was prepared by a chemical co-precipitation method and subsequently coated with poly(diallyldimethyl ammonium chloride) (poly(ac)) to enhance the stability of these nanoparti- cles against agglomeration and precipitation. The resulting poly(ac)-containing ferrofluid was drop cast as a 50 mL droplet onto a disposable NiO-modified screen-printed electrode. When used in the presence of an applied magnetic field, it acted as a "membrane-less and spill-less" electrolyte for the quantification of oxygen content. The results surprisingly re- vealed that the poly(ac) not only acted as a polymeric stabiliser to stabilise and passivate the surface of the Fe2O3 nanoparti- cles, but also resulted in an apparent increase in electrocatalyt- ic activity towards the reduction of oxygen, making the poly(ac)-ferrofluid ideal for the electrochemical detection of oxygen. Experimental Section Reagents and equipments Iron(III) chloride hexahydrate (FeCl3*6H2O, = 98%), iron(II) chlo- ride tetrahydrate (FeCl2*4H2O, = 99%) and iron(III) nitrate nona- hydrate (Fe(NO3)3*9H2O, = 98 %) were purchased from Sigma- Aldrich (Castle Hill, NSW, Australia). Ammonium hydroxide (28 % w/w NH3 in water) was supplied by Ajax Finechem (Seven Hills, NSW, Australia). Poly(diallyldimethyl ammonium chloride) solution (poly(ac); 20 wt % in water, average Mw [asymptotically =]100 000-200 000), poly(4-styrenesulfonic acid) solution (poly(ss); 18 wt % in water, average Mw [asymptotically =] 75 000), poly(sodium 4-styrenesulfonate) solution (poly(ss-Na); 30 wt % in water, average Mw [asymptotically =] 200000) and all other chemicals were purchased from Sigma-Aldrich (Castle Hill, NSW, Australia). All the re- agents were used without further purification. All solutions were prepared using deionised water of a resistivity not less than 18.2 MW cm at (298 ±2) K (Millipore Pty Ltd., North Ryde, NSW, Australia). High-purity compressed oxygen and nitrogen gases were obtained from Air Liquide (Fairfield, NSW, Australia). Neodymium-iron-boron (NdFeB) disc magnets with a size of 6 mm diameter and 1 mm height used in this study were pur- chased from Frenergy Magnets. Each disc magnet was typically magnetised through the height/thickness with a theoretical holding force (pull strength) of about 0.43 kg and a reported surface gauss (which represented the magnetic flux density or the magnetic induction) of about 2229 gauss. Electrochemical experiments Electrochemical measurements were recorded with a mAutolab PGSTAT 101 computer-controlled potentiostat (Ec°Chemie, Utrecht, The Netherlands) with Nova 1.10 software. CV was performed at room temperature at a scan rate of 50 mV s-1 (unless stated otherwise) with a standard three-electrode con- figuration that employed the working, counter and reference electrodes of the SPEs, which were connected with a cable connector to the potentiostat. The SPEs obtained from Drop- Sens (Oviedo, Spain) used in this study consisted of a Pt (DPR- 550) or C (DPR-110Ni) working electrode (4 mm diameter) modified with NiO, an Ag quasi-reference electrode and a Pt or C counter electrode. The sample solution (ca. 50 m L) was drop cast onto the SPE, entirely covering the three electrode surfa- ces. The SPE was then placed into an airtight cell with the inlet and outlet gas lines, and the cell (with the SPE inside) was set under a controlled atmosphere of either oxygen, nitrogen or a mixture of the two. Basically, the anhydrous gas (either oxygen or nitrogen from the compressed gas cylinders) was first humidified by passing through a water bubbler before being introduced into the cell in which the SPE with the drop- cast aqueous sample was placed. The different concentrations of oxygen were obtained by mixing oxygen and nitrogen gases together with a gas-mixing system that consisted of two flow meters with a floating ball (Cole-Parmer, Chatswood, NSW, Australia) and a glass gas-mixing segment. The relative flow rates of the two flow meters were then used to calculate the percentage concentration of oxygen being introduced into the cell. Synthesis of iron oxide magnetic nanoparticles in aqueous medium Iron oxide magnetic nanoparticles were synthesised by a chem- ical co-precipitation method, which has been previously re- ported to result in maghemite nanoparticles (g-Fe2O3) that ranged from 2 to 17 nm.[1-3,21,23,24,50] A solution of 2 m FeCl3*6H2O/1m HCl (80mL) and FeCl2*4H2O/1m HCl (40 mL) were mixed, and the mixture was diluted to 1.2 L with water. Then NH4OH (250 mL, 28 % w/w NH3 in water) was quickly added into the iron(II) and iron(III) solution, and the mixture was vigorously stirred at room temperature for 30 min. After the addition of NH4OH, the colour of the mixture changed from orange to black owing to the precipitation of magnetite (Fe3O4).[3,23,50] Next, Fe(NO3)3 was added, and the mixture (in acidic medium) was subsequently heated to 90 8C with stirring for 1 h to oxidise the precipitate Fe3O4 to maghemite ( g -Fe2O3).[3] The synthesised iron oxide magnetic nanoparticles (henceforth referred to as Fe2O3 NPs) were then magnetically decanted, washed extensively with water and redispersed in water, yield- ing a very stable colloidal dispersion with about (7.96 ± 0.02) % w/w nanoparticle concentration before being pre-concentrated to a concentration of about 10 % w/w by removing some water from this water-based ferrofluid in the oven at 60 8C. The resulting pre-concentrated suspension of Fe2O3 NPs was also found to be, to the naked eye, very stable for at least several months. The pH of this water-based ferrofluid was found to be about 2.75, indicating that the surfaces of the nanoparticles were positively charged, preventing them from agglomerating and resulting in a stable ferrofluid.[22] Characterisation of iron oxide magnetic nanoparticles The synthesised iron oxide magnetic nanoparticles were exam- ined with an FEI Tecnai G2 20 TEM operating at 2 keV and 0.2 mA. The aqueous ferrofluid dispersions were first diluted to a concentration of 0.1 % w/w in water. A single drop of the particle suspension was deposited onto a Formvar coated copper grid and dried over a period of 3 days prior to imaging. Acknowledgements The Australian Research Council (ARC DECRA DE130100770) is ac- knowledged for funding. [1] A. H. Lu, E. L. Salabas, F. Schüth, Angew. Chem. 2007, 119, 1242 - 1266; Angew. Chem. Int. Ed. 2007, 46, 1222 - 1244. [2] N. Jain, X. Zhang, B. S. Hawkett, G. G. Warr, ACS Appl. Mater. Interfaces 2011, 3, 662 -667. [3] S. Laurent, D. Forge, M. Port, A. Roch, C. Robic, L. Vander Elst, R. N. Muller, Chem. Rev. 2008, 108, 2064 - 2110. [4] C. Boyer, V. Bulmus, P. Priyanto, W. Y. Teoh, R. Amal, T. P. Davis, J. Mater. Chem. 2009, 19, 111 - 123. [5] C. Pflipsen, D. Forge, S. Benali, Y. Gossuin, J. Phys. Chem. C 2013, 117, 20919-20926. [6] S. Cheong, P. Ferguson, I. F. Hermans, G. N. L. Jameson, S. Prabakar, D. A. J. Herman, R. D. Tilley, ChemPlusChem 2012, 77, 135 - 140. [7] S. Cheong, P. Ferguson, K. W. Feindel, I. F. Hermans, P. T. Callaghan, C. Meyer, A. Slocombe, C. H. Su, F. Y. Cheng, C. S. Yeh, B. Ingham, M. F. Toney, R. D. Tilley, Angew. Chem. 2011, 123, 4292 - 4295 ; Angew. Chem. Int. Ed. 2011, 50, 4206 -4209. [8] J. B. Haun, T. J. Yoon, H. Lee, R. Weissleder, Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol. 2010, 2, 291 - 304. [9] M. Colombo, S. Carregal-Romero, M. F. Casula, L. Gutierrez, M. P. Mo- rales, I. B. Bohm, J. T. Heverhagen, D. Prosperi, W. J. Parak, Chem. Soc. Rev. 2012, 41, 4306 - 4334. [10] I. Y. Goon, L. M. H. Lai, M. Lim, R. Amal, J. J. Gooding, Chem. Commun. 2010, 46, 8821 - 8823. [11] J. Wang, M. Tong, X. Wang, Y. Ma, D. Liu, J. Wu, D. Gao, G. Du, Sens. Ac- tuators B 2002, 84, 95 - 97. [12] J. Chen, L. Xu, W. Li, X. Gou, Adv. Mater. 2005, 17, 582 - 586. [13] J. S. Han, D. E. Davey, D. E. Mulcahy, A. B. Yu, J. Mater. Sci. Lett. 1999, 18, 975-977. [14] G. Neri, A. Bonavita, S. Galvagno, P. Siciliano, S. Capone, Sens. Actuators B 2002, 82, 40 - 47. [15] Y. Wang, S. Wang, Y. Zhao, B. Zhu, F. Kong, D. Wang, S. Wu, W. Huang, S. Zhang, Sens. Actuators B 2007, 125, 79 - 84. [16] W. Cao, O. K. Tan, J. S. Pan, W. Zhu, C. V. Gopal Reddy, Mater. Chem. Phys. 2002, 75, 67 - 70. [17] G. Neri, A. Bonavita, S. Ipsale, G. Rizzo, C. Baratto, G. Faglia, G. Sberve- glieri, Mater. Sci. Eng. B 2007, 139, 41 - 47. [18] M. Ivanovskaya, D. Kotsikau, G. Faglia, P. Nelli, Sens. Actuators B 2003, 96, 498- 503. [19] K. Raj, R. Moskowitz, J. Magn. Magn. Mater. 1990, 85, 233 - 245. [20] C. Scherer, A. M. Figueiredo Neto, Brazilian Journal of Physics 2005, 35, 718-727. [21] N. Jain, Y. Wang, S. K. Jones, B. S. Hawkett, G. G. Warr, Langmuir 2010, 26, 4465 - 4472. [22] D. Zins, V. Cabuil, R. Massart, J. Mol. Liq. 1999, 83, 217 - 232. [23] P. Berger, N. B. Adelman, K. J. Beckman, D. J. Campbell, A. B. Ellis, G. C. Lisensky, J. Chem. Educ. 1999, 76, 943 - 948. [24] R. Massart, Magnetics IEEE Transactions 1981, 17, 1247 - 1248. [25] L. Aldous, A. Khan, Md. M. Hossain, C. Zhao, Catalysis in Ionic Liquids : From Catalyst Synthesis to Application, The Royal Society of Chemistry, London, 2014, pp. 433- 473. [26] J. R. Stetter, G. Korotcenkov, X. Zeng, Y. Tang, Y. Liu in Chemical Sensors Comprehensive Sensor Technologies, Vol. 5 (Ed. : G. Korotcenkov), Mo- mentum Press, New York, 2011, pp. 1 - 89. [27] L. C. Clark Jr., C. Lyons, Ann. N. Y. Acad. Sci. 1962, 102, 29- 45. [28] M. C. Buzzeo, C. Hardacre, R. G. Compton, Anal. Chem. 2004, 76, 4583 - 4588. [29] X. J. Huang, L. Aldous, A. M. O'Mahony, F. J. del Campo, R. G. Compton, Anal. Chem. 2010, 82, 5238 - 5245. [30] R. Toniolo, N. Dossi, A. Pizzariello, A. P. Doherty, S. Susmel, G. Bontem- pelli, J. Electroanal. Chem. 2012, 670, 23 - 29. [31] R. Toniolo, N. Dossi, A. Pizzariello, A. P. Doherty, G. Bontempelli, Electroa- nalysis 2012, 24, 865 -871. [32] J. Lee, K. Murugappan, D. W. M. Arrigan, D. S. Silvester, Electrochim. Acta 2013, 101, 158 - 168. [33] K. Murugappan, J. Lee, D. S. Silvester, Electrochem. Commun. 2011, 13, 1435-1438. [34] N. V. Rees, S. M. Matthews, K. Yunus, A. C. Fisher, R. G. Compton, Angew. Chem. 2009, 121, 2412- 2414; Angew. Chem. Int. Ed. 2009, 48, 2376- 2378. [35] Y.-G. Zhou, N. V. Rees, J. Pillay, R. Tshikhudo, S. Vilakazi, R. G. Compton, Chem. Commun. 2012, 48, 224 - 226. [36] P. C. Morais, C. B. Teixeira, K. Skeff Neto, R. B. Azevedo, Z. G. M. Lacava, L. M. Lacava, Solid State Commun. 2000, 114, 59 - 62. [37] R. A. Sperling, W. J. Parak, Philos. Trans. R. Soc. London Ser. A 2010, 368, 1333-1383. [38] Q. M. Kainz, O. Reiser, Acc. Chem. Res. 2014, 47, 667 - 677. [39] A. Wooding, M. Kilner, D. B. Lambrick, J. Colloid Interface Sci. 1992, 149, 98-104. [40] L. Shen, P. E. Laibinis, T. A. Hatton, Langmuir 1999, 15, 447 - 453. [41] A. Khan, X. Lu, L. Aldous, C. Zhao, J Phys Chem. C 2013, 117, 18334 - 18342. [42] J. Snyder, T. Fujita, M. W. Chen, J. Erlebacher, Nat. Mater. 2010, 9, 904 - 907. [43] T. Z. Fahidy, J. Appl. Electrochem. 1983, 13, 553 - 563. [44] G. Hinds, J. M. D. Coey, M. E. G. Lyons, Electrochem. Commun. 2001, 3, 215-218. [45] R. A. Tacken, L. J. J. Janssen, J. Appl. Electrochem. 1995, 25,1-5. [46] A. Bund, S. Koehler, H. H. Kuehnlein, W. Plieth, Electrochim. Acta 2003, 49, 147- 152. [47] A. J. Bard, L. R. Faulkner, Electrochemical Methods : Fundamentals and Ap- plications, Wiley, New York, 1980. [48] D. Zhang, T. Okajima, F. Matsumoto, T. Ohsaka, J. Electrochem. Soc. 2004, 151, D31 -D37. [49] E. E. Switzer, R. Zeller, Q. Chen, K. Sieradzki, D. A. Buttry, C. Friesen, J. Phys. Chem. C 2013, 117, 8683 - 8690. [50] J. B. Mamani, A. J. Costa-Filho, D. R. Cornejo, E. D. Vieira, L. F. Gamarra, Mater. Charact. 2013, 81, 28 - 36. Received : April 11, 2014 Revised : June 5, 2014 Published online on July 30, 2014 Janjira Panchompoo,[a] Mengchen Ge,[a] Chuan Zhao,[a] May Lim,[b] and Leigh Aldous*[a] [a] Dr. J. Panchompoo, M. Ge, Dr. C. Zhao, Dr. L. Aldous School of Chemistry, Faculty of Science UNSW Australia, Sydney, NSW 2052 (Australia) Fax: (+ 61) (0)-2-9385-6141 E-mail : [email protected] [b] Dr. M. Lim School of Chemical Engineering, UNSW Australia Sydney, NSW 2052 (Australia) This article is part of the "Early Career Series". To view the complete series, visit : http ://chempluschem.org/earlycareer. (c) 2014 Blackwell Publishing Ltd. |