WO2002001214A1

WO2002001214A1 - Sensor for the selective detection of metal cations, method for the identification and quantitative determination of metal cations in a composition and method for the removal of metal cations from a composition

Info

Publication number: WO2002001214A1
Application number: PCT/GB2001/002805
Authority: WO
Inventors: Colin Boxall; Simon Nigel Port; Stephen Giles David Shackleford
Original assignee: British Nuclear Fuels Plc
Priority date: 2000-06-24
Filing date: 2001-06-22
Publication date: 2002-01-03
Also published as: GB0015406D0; AU6615801A

Abstract

The invention provides a sensor, adapted for the selective detection of cations of a specific valence state, derived from metals capable of existing in more than one valence state, said detector comprising a quartz crystal microbalance which comprises a quartz crystal sandwiched between two electrodes used to generate an alternating electric field across the crystal, the surface of said quartz crystal microbalance being coated with a polymer impregnated with ligands having a high degree of sequestration specificity towards cations of a specific valence state, and the sensor being adapted to provide a variable potential such that changes in the valence state of the sequestered cations may be induced. The invention also provides a method for the identification can quantitative determination of cations in a composition, said cations having a specific valence state and being derived from metals capable of existing in more than one valence state; additionally, the invention provides methods for removing cations from such a composition and for resolubilising the sequestered cations.

Description

SENSOR FOR THE SELECTIVE DETECTION OF METAL CATIONS, METHOD FOR THE IDENTIFICATION AND QUANTITATIVE DETERMINATION OF METAL CATIONS IN A COMPOSITION AND METHOD FOR THE REMOVAL OF METAL CATIONS FROM A COMPOSITION

This invention relates to quartz crystal microbalance (QCM)-based sensors and their application for the analysis of cationic materials. More specifically, the invention is 5 concerned with the use of QCM sensors in the identification and quantitative determination of cationic analytes, typically actinide metal ions, which are of particular interest to the nuclear industry.

The use of QCM resonators for the in-situ measurement of changes in mass on 10 electrode surfaces is now well established. Consequently, these devices have found widespread utility in a variety of analytical and electroanalytical applications, wherein the determination of very small changes in mass has facilitated the investigation of mass transfer processes at both gas/solid and gas/liquid/solid interfaces. Typically, a QCM resonator comprises a thin quartz crystal sandwiched 15 between two electrodes which are used to generate an alternating electrical field across the crystal. This field produces a mechanical oscillation in the crystal at its resonance frequency which may then be directly related to changes in crystal mass.

More recently, the electrochemical quartz crystal microbalance (EQCM), in which

20 one crystal face is exposed to an electrolyte solution and serves as a working electrode, has been used to study mass changes at electrified solid/solution interfaces.

However, in situ microweighing with an EQCM resonator in the course of an electrochemical experiment provides not only data relating to mass changes, which may then be correlated to electrochemical processes, but also facilitates the

25 measurement of electrochemical variables. The combination of electrochemical and mass measurements which is available via EQCM provides an extremely useful and versatile tool for the study of physicochemical phenomena which occur at the electrode surface. Thus, studies of electrocrystallisation, electrodeposition, electrodissolution and corrosion, electrochromism, gas adsorption, polymer

30 permeation, electropolymerisation and redox polymers have all benefited from the use of EQCM. Of particular interest in the present context are EQCM systems which include a surface-immobilised film on the surface of an electrode, and which may be used to selectively sequester species from solution. In such cases, however, difficulties may be experienced in distinguishing mass changes from contributions from parameters such as film density, viscosity and elasticity and, in these circumstances, the use of additional analytical techniques in association with EQCM analysis has proved to be especially effective. Most particularly, it has been found that spectroscopic techniques are of significant value, notably in those cases wherein the species which are selectively adsorbed from solution contain or form a coloured and/or luminescent entity upon complexation with the species-selective surface film.

For present purposes, it is desirable to employ a system wherein the electrode surface film should be capable of selectively sequestering from solution those ions which are of particular concern in the nuclear industry, typically actinide metal cations.

It is well known to provide chemically modified electrodes for metal ion determinations, such that selective accumulation of the analyte of interest may be achieved by immobilisation of metal ion binding reagents at or on an electrode surface, prior to analysis of the material by volta metric techniques. For such purposes, various ligands have found applicability including, for example, ethylenediamine, phenanthroline, dimethylglyoxime and dithiocarbamate drivatives. It has been known for some time that hydroxyamic acid (HA) ligands have an important role in microbial iron mechanism and, more recently, it has been shown that such derivatives may be incorporated by ion-exchange into glassy carbon electrodes coated with Nafion (a perfluorosulphonate cation exchange polymer). Thus, modified electrodes have been prepared in which desferrioxamine (a trihydroxamic acid supplied as the methane sulphonate salt) and glycine hydroxamic acid have been used as sources of these ligands. The immobilised HA ligands may then be used as sequestration sites for iron(III) ions, which may subsequently be detected within the film by electrochemical means and also form highly coloured species which may be detected spectroscopically. The ability of HA-containing Nafion coated electrodes to take up iron(III) ions from dilute solution makes them highly attractive as sensor devices for this species.

It has been observed that the activity of these electrodes is quite specific towards iron(III) ions in this context, and the same effect is not observed with iron(II) cations. This, of course, provides a potentially useful means for distinguishing between the different oxidation states of iron; indeed there is the possibility of extending the concept to other metals which exist in different oxidation states, the most suitable example for present purposes being the actinides.

Furthermore, the differing affinities of the iron(II) and iron(III) ions for the Nafion coated electrode may be utilised to provide for desorption of material from the electrode surface by inducing reduction of the bound iron(III) to iron(II), which may be achieved electrochemically by changing the applied voltage. Thus, it is possible to promote desorption of the material back into solution.

These observations again have particular relevance to other metals which may exist in more than one oxidation state; again, therefore, the concepts find ready application in the nuclear industry, wherein actinide ions, such as uranium or plutonium, may be present in more than one valence state.

Consequently, it is an objective of the present invention to provide a sensor device which is capable of detecting cationic analytes. More particularly, it is an objective to provide a device which is capable of selectively detecting such analytes when they comprise metal ions which are capable of existing in more than one valence state. Most specifically, it is preferred that such cations should comprise actinide (4+) metal ions, which are of interest to the nuclear industry.

It is a further objective of the invention to provide such a sensor device which allows for the identification and quantitative determination of the said cations, thereby facilitating improved process control and providing a convenient means for the monitoring of pollutant levels in effluent streams discharged to the aquatic environment. Specifically, in the case of radioactive materials, it provides a means for measuring isotope levels in the said effluent streams.

It is a still further objective of the present invention to provide a sensor device which is capable of selectively removing specific metal cations from solution in order to facilitate qualitative and quantitative determination, but which can then be adapted to promote subsequent dissolution of the metal cations by inducing a change in the valence state of the metal. Such a device would find particular application in the case of the actinides, since prior art devices which are utilised for removing these materials from solution for analytical purposes suffer from the serious drawback that the subsequent resolubilisation of the ions by desorption from the surface of the device is only achieved by eluting with copious volumes of solvents, thereby leading to significant problems of effluent disposal.

The present inventors have found that a sensor which is sequestration specific to the cations of interest, namely those which are capable of existing in more than one oxidation state, may be obtained by providing a QCM having an ion exchange polymer, in combination with a suitable ligand such as HA, coated on its surface. Thus, it has been possible to selectively remove iron from solution as iron(III) ions, rather than iron(II) ions. Similar considerations are found to apply in the case of the actinide(4+) cations.

According to a first aspect of the present invention, there is provided a sensor, adapted for the selective detection of cations of a specific valence state, derived from metals capable of existing in more than one valence state, said detector comprising a quartz crystal microbalance which comprises a quartz crystal sandwiched between two electrodes used to generate an alternating electric field across the crystal, the surface of said quartz crystal microbalance being coated with a polymer impregnated with ligands having a high degree of sequestration specificity towards cations of a specific valence state, and the sensor being adapted to provide a variable potential such that changes in the valence state of the sequestered cations may be induced.

Said device finds particular application with iron(III) ions, which may be selectively removed from solutions which also include iron(II)ions. Subsequently, resolubilisation of the iron removed from solution may be achieved by promoting reduction of the iron(III) ions to iron(II) ions. Similar considerations apply in the case of actinide(4+) cations, which may be removed from solution and resolubilised in analogous fashion.

The QCM provided for present purposes is of the type well known to those skilled in the art. The surface is coated with a composite layer comprising an ion exchange polymer and suitable ligands. Said polymer may be radiation-hardened, and preferably comprises a perfluorosulphonate cation exchange resin; said ligands preferably comprise at least one of hydroxamic acid and its derivatives, such as formohydroxamic acid or acetohydroxamic acid. It is known, however, that certain hydroxamic acid derivatives, including those mentioned, are susceptible to irreversible oxidation in the presence of plutonium(IV), oxyplutonium or oxyneptunium ions, all of which are frequently encountered in certain of the applications in which the sensor of the present invention may be deployed. Consequently, in such cases, it is necessary to identify HA derivatives which are not susceptible in this way, or to restrict the use of the device to those situations in which oxidising ions do not occur. Studies of the redox potentials of these derivatives in nitrate media have been carried out by means of cyclic voltarnmetry measurements, and have provided a useful means of determining the degree of susceptibility of the compounds to such oxidative decomposition. It has thus been possible to identify suitable derivatives, and to avoid the use of systems in which such problems may occur.

These difficulties may, in any event, be overcome by electrochemical control of the solution environment adjacent the QCM resonator surface, which is possible within an EQCM configuration, wherein one face of the quartz crystal is exposed to the solution environment and serves as a working electrode; by correct biasing of said electrode, it is possible to ensure that any species capable of oxidising the HA will itself be reduced by this electrode. Such a configuration, therefore, represents a preferred embodiment of the first aspect of the present invention

According to a second aspect of the present invention, there is provided a method for the identification and quantitative determination of cations in a composition, said cations having a specific valence state and being derived from metals capable of existing in more than one valence state, said method comprising:

(a) providing a sensor adapted for the selective detection of cations of a specific valence state, derived from metals capable of existing in more than one valence state, said detector comprising a quartz crystal microbalance which comprises a quartz crystal sandwiched between two electrodes used to generate an alternating electric field across the crystal, the surface of said quartz crystal microbalance being coated with a polymer impregnated with ligands having a high degree of sequestration specificity towards cations of a specific valence state, and the sensor being adapted to provide a variable potential such that changes in the valence state of the sequestered cations may be induced; and

(b) utilising said sensor for QCM analysis of said composition.

Provided that the HA derivatives are rigidly coupled to the QCM resonator surface, said method allows for measurement of the total mass of ions adsorbed in a mixture of such cations. Selective sequestration of the individual cations by the HA within the polymer layer may be detected by, for example, microgravimetric QCM, EQCM or, preferably, spectroscopically coupled QCM. The use of coupled spectroscopic and QCM analysis on post-complexed surface adsorbed species containing a chromophore allows for the ready determination of the mass of individual materials, thereby facilitating the identification and relative quantification of each individual cation in a composition. Additionally, data relating to the viscoelastic properties of the film may be acquired from simple conductance-frequency spectra.

The method according to the second aspect of the present invention provides a convenient means for the monitoring of effluent streams and affords a reliable process for the maintenance of strict environmental controls.

According to a third aspect of the present invention, there is provided a convenient method for the removal of cations from a composition, said cations having a specific valence state and being derived from metals capable of existing in more than one valence state, said method comprising:

(a) providing a sensor as hereinbefore described;

(b) utilising said sensor for QCM analysis of said composition; and

(c) adjusting the potential across said sensor such that changes in the valence state of the sequestered cations may be induced, thereby facilitating solubilisation of the said cations.

The method according to the third aspect of the invention thereby provides a significantly more efficient means of promoting desorption of the cation material than is known from the prior art, wherein removal of material from such devices is only achieved with great difficulty by means of protracted washing procedures which produce large volumes of effluent subsequently requiring disposal.

The invention will now be illustrated, though without limitation, by way of the following experimental details: EXPERIMENTAL

A CALIBRATIONOFTHE QCM

(I) Procedure for Calibration

0.0427g of AgNO₃ (1.01 mM) were mixed with 2.11 ml HClO₄ (0.1 M) in a 250ml volumetric flask. The solution was then bubbled through with N₂ for 10 min to remove all O₂. The QCM was adjusted to its resonant frequency ₀ and this frequency was recorded. The QCM was placed into the AgNO₃ / HClO solution under N₂ and simultaneous cyclic voltammetric and EQCM frequency responses recorded. The results are illustrated in Figure 1.

The increase in the cathodic current is due to silver electrodeposition and is accompanied by a frequency decrease due to the mass gain; the converse is true for the anodic current.

(II) Results and Discussion

The values for frequency shift derived from the electrodeposition of Ag (see Figure 1) were used in conjunction with charged passed to obtain a value of S_ex, the mass sensitivity of the device.

Charge passed during the Ag electrodeposition and stripping processes was determined by a simple Simpsons rule integration of the current passed as a function of time.

A linear correlation of Δf versus Charge (Q) is observed for both electrodeposition (see Figure 2) and electrodissolution (Figure 3). The measured mass change (Δm) may be obtained from the charge (Q) by using Faraday's Law.

where

Δm is the mass change (g),

Q is the charge passed during electro deposition / stripping (C), F is faradays constant (96484.6 C/mol), _wthe molecular weight of the deposit (g), and z is the number of electrons involved during the deposition / stripping phases.

A plot of measured mass change versus frequency gives an experimental mass sensitivity. The mass sensitivity for electrodeposition of silver upon a gold electrode, determined from Figure 4, is 2.376 ng Hz^"1 conversely the mass sensitivity for electrostripping of silver, determined from Figure 5, is 2.078 ng Hz^"1

B ELECTROCHEMICAL STUDIES

(I) Experimental Procedures

(a) ELECTROCHEMISTRY OF ACETOHYDROXAMIC ACID

(i) Studies of Susceptibility to Oxidative Decomposition

A preliminary investigation into the electrochemistry of acetohydroxamic acid has been conducted by means of cyclic voltammetry, performed using a PGSTAT 10 electrochemical work station (Windsor Scientific) in conjunction with a conventional three electrode cell (saturated calomel (SCE) as reference, platinum as auxiliary and a gold disc microelectrode (125 μm radius) as a working electrode). All experiments were made with a starting potential of 0 V vs SCE from which the applied potential was initially swept in the anodic direction at a sweep rate of 1 mV/s. Anodic and cathodic sweep limits were usually +1 V and -1 V vs SCE respectively. A solution of acetohydroxamic acid was freshly prepared in doubly distilled, deionised water immediately prior to measurement to avoid problems associated with hydrolysis of the acid. As a precaution against thermal decomposition, the solid form of the acid was stored in a refrigerator.

The results obtained are illustrated in Figure 6 and show good agreement with those obtained during earlier studies, which are illustrated in Figure 8. The effect of using a gold disk microelectrode having twice the radius of that described above is shown in Figure 7.

Observation of Figures 6, 7, and 8, shows a wave plateauing at approximately 0.2V due to the formation of premonolayer hydrous gold oxide. The wave with an onset of approximately 0.4V is due to the oxidation of acetohydroxamic acid via the cleavage of the C-N bond, yielding a carboxylic acid and various nitrogen species.

The post wave peak at 0.6V (Figure 6), 0.55V (Figure 7) and 0.75V (Figure 8) is due to the oxidation of surface adsorbed acetohydroxamic acid.

Initial results taken for acetohydroxamic acid (Figures 6 & 7) show a reasonable correlation to Figure 8; however, it is apparent that the adsorption peaks in both Figures 6 and 7 have moved to a more negative potential. Also Figure 8 displays a more distinguishable adsorption postwave than can be seen in Figures 6 & 7. This is probably due to differing surface conditions between Figures 6 & 7 and 8.

(ii) Data Acquisition during Electrochemical Experiments

In order to assess the thermodynamic stability of various hydroxamic acids, cyclic voltammetric (CV) experiments were performed using a conventional three-electrode cell. A gold microdisc electrode (as described in the section on gold microdisc electrode fabrication below) served as a working electrode, a platinum wire coil (250μm diameter, Advent Research Materials Ltd. UK) served as the counter electrode, and a saturated calomel electrode (SCE) (Russel, type CRL/s7, UK) was used as reference electrode in all electrochemical studies of hydroxamic acid (HA) stability. All CVs were recorded using a commercially available specialist low current potentiostat (Autolab PGSTAT 10, Windsor Scientific Ltd. UK).

(Hi) Data Acquisition during Electrochemical Quartz Crystal Microbalance Experiments

The electrochemical quartz crystal microbalance (10 MHz, Windsor Scientific Ltd. UK) used in these experiments incorporated a mass sensitive 14mm diameter, planar- planar, AT-cut quartz crystal oscillator (Institute of Physical Chemistry, Warsaw, Poland) with a resonant

of 10 MHz. A 6mm diameter gold "key hole" pattern was deposited upon both oscillator faces. The gold layer facing the solution was connected to the working electrode input of the potentiostat.

The EQCM experiments were performed with a conventional three electrode cell. The solution side face of the EQCM oscillator served as the working electrode, a platinum wire coil served as the counter electrode and a saturated calomel electrode was used as the reference electrode. Simultaneous QCM and CV experiments were performed using a general purpose electrochemical system (GPES) consisting of a PGSTAT 10 potentiostat driven by the appropriate proprietary software.

The frequency was measured with a Philips PM6680 High Resolution Programmable Timer/Counter (J.Fluke, Windsor Scientific Ltd. UK). Simultaneous CV and Δf versus E curves were recorded by interfacing the output of the frequency counter to a second channel input of the GPES via an ADC to the computer software GPES. The calibrated sensitivity of the EQCM system was calculated to be 2.078ng Hz^"1. Nitrogen (Whitespot, BOC, UK) was purged through the solution to remove any oxygen prior to EQCM experiments.

(iv) Preparation of Acetohydroxamic Acid Solution

All solutions are prepared in doubly distilled water that was further purified by an E pure Deionisation System (Barnstead model 04642). 0.075g of AHA (AnalaR, Sigma- Aldrich C°. Ltd. UK) were mixed with 500cm³ of 0.1M HNO₃ (dilution of cone. HNO₃, AnalaR, BDH Chemicals Ltd. UK) in a 500ml volumetric flask to give a 2mM solution of AHA. 30cm³ of this solution ware placed into a Pyrex beaker where N₂ was purged through for lOmin to remove all O₂. All measurements were carried out under a continuous blanket of N .

(b) GOLD MICRO-DISC ELECTRODE FABRICATION

(i) Electrode Fabrication

Borosilicate glass Pasture pipettes (John Poulton Ltd., UK) were trimmed so as to remove tapered ends. The pipette bodies and gold wire (250, 125, 50, and 25μm diameter, 99.99+%, pure, Advent Research Chemicals Ltd., UK) were washed with sonication, for 15min, in chloroform, acetone, ethanol (AnalaR grade, BDH Chemicals Ltd. UK) and doubly distilled E pure water. After each wash, excess solvent was removed by a lens cloth. The glass tubes and gold wire were then dried for 15min at 383K. The gold wires and glass tubes were then washed in 5M HNO₃ for lhr, finishing with a wash in doubly distilled E pure water. The gold wires and glass tubes were then oven dried at 383K for 30min.

Approximately 20mm lengths of the washed gold wires were then sealed into the glass tubes by use of a bunsen burner, ensuring that ~5mm of gold wire protruded from either side of the glass seal. Smaller electrodes, not manufactured in house, were purchased from commercial suppliers. These include the following a 0.6μm diameter platinum microdisc electrode and a 5μm diameter gold microdisc electrode (both from BAS Technicol Ltd., UK).

(ii) Fabrication of Electrical Contact

For electrodes prepared in house electrical contact with the interior exposed gold wire was established by the following method: Woods metal (Bi50/Cdl2.5/Pb25/ Snl2.5, Advent Research Materials Ltd. UK) was cut into small 2.5mm pieces and dropped into the sealed glass tubes and melted by immersing the tubes in boiling water. An electrical contact was then made by placing a length of Cu wire (l-3mm diameter, 99.99+% pure, Advent Research Materials Ltd. UK) into the molten woods metal, which was cooled to form a set. The sealed glass tube was given additional strength by injecting the interior of the glass tube with Araldite CY1300 and 1301 (Ciba-Geigy Plastics, UK). The excess gold wire protruding from the distal end of the sealed glass tube was trimmed and a gold, disc shaped surface exposed with light sanding.

(c) GOLD MICRODISC ELECTRODE POLISHING PROCEDURE

The above fabricated gold microdisc electrodes were polished using diamond pastes of a decreasing granular size; 6μm, 3μm, and lμm (marcon, UK) upon a synthetic micro polishing cloth (Buehler, USA) and were washed with sonication in ethanol for 15min between each polish. Finally, the gold microdisc electrodes were polished with a slurry of lOnm Al O₃ (AnalaR, BDH Chemicals Ltd. UK) in a lOmM solution of NaCN and sonicated in doubly distilled E pure water for 15min, so removing any embedded alumina particles which could potentially act in an electrocatalytic fashion. (d) GOLD MICRODISC ELECTRODE ELECTROCHEMICAL CLEANING PROCEDURE

Finally, the gold disc microelectrodes were polished electrochemically. This was accomplished by immersing the electrode in question in a solution of appropriate buffer (0.1M HNO₃) and sweeping 5 times between the solvent limits at a sweep rate of 90 mV/s. The electrochemical polishing procedure always terminated at the end of an anodic going sweep.

(e) NAFION /HYDROXAMIC ACID COMPOSITE LAYER PREPARATION

The 14mm diameter, planar-planar, Au coated AT-cut quartz crystals were used as received. They were coated with 5ml of 5% Nafion solution in lower aliphatic alcohols and 10% water (950 Equi. Wt. Polymer, Electrosynthesis C°. Inc., USA.). This volume covered the total area of the gold electrode flags including that presented to the solution. The solvent was allowed to evaporate at room temperature giving a Nafion coated gold/quartz electrode.

The Nafion coated gold/quartz crystal was then immersed in an aqueous solution of the desired ligand for 18 hours. Ligands used were acetohydroxamic acid, desferrioxamine (DFA⁺) and glycine hydroxamic acid (GHA⁺) (AnalaR, Sigma- Aldrich C°. Ltd., UK). Each ligand was separately dissolved in doubly distilled E- pure water to give a ImM solution. The hydroxamic ligand partitioned into the Nafion film producing a Nafion/hydroxamic acid composite modified gold/quartz crystal.

IRON (III) DETERMINATION USING QCM/EQCM EXPERIMENTS

The iron(III) content of a solution under investigation can be obtained by EQCM via the medium exchange technique. This involves first impregnating the Nafion/AHA composite layer overlying the Au/quartz oscillator with Fe(III). The amount of Fe(III) adsorbed by the Nafion/AHA layer will then be proportional to the Fe(III) concentration in the solution in question. Fe(III) uptake in the layer is then assessed by CV and QCM measurements. The former is accomplished by immersing the oscillator composite in the solution under interrogation for 5min. The latter is accomplished by extracting the oscillator from the solution under investigation, rinsing and drying it and then immersing it an electrochemical cell containing 0.5M HNO3.

II Results

(a) ELECTROCHEMICAL STUDIES OF ACETOHYDROXAMIC ACID USING DISC MICROELECTRODES

In order to assess the thermodynamic and kinetic stability of the AHA ligand within an electrochemical system, we elected to study the cyclic voltammetric (CV) behaviour of AHA using a range of disc microelectrodes. The resultant voltammograms may be interpreted using the theory of Myland et al. (J. Electroanal. Chem., 270 (1989), 79) allowing calculation of the standard redox potential, E^e, and/or the standard electrochemical rate constant, k^θ, and transfer coefficient, α, for AHA oxidation.

Figures 11-14 show the cyclic voltammograms obtained from a 2 mol m^"3 solution of AHA at pH 1 using a range of gold disc microelectrodes. A sweep rate of 1 V s^"1, a starting potential of -0.3 V and cathodic and anodic sweep limits of -0.3 V and +1.5 V respectively were used for all four voltammograms.

For all electrode radii, the initial anodic-going scan appears featureless from -0.3 to +0.3 V. Between +0.3 to +0.7 V, a small oxidation peak can be observed, due to possible sub-monolayer production of an ill-characterised gold hydroxyl layer. A large oxidation wave is observed with an onset of +0.9 V, followed by a smaller peak with an onset of +1.3 V. Comparison with CVs of gold in background acid recorded by Burke et al. (Analyst, 119 (1984), 841), Johnson and LaCourse (Electroanalysis, 4 (1992), 367), Gerlache et al. (Electroanalysis, 9 (1997), 1088), Bashir (MSc Thesis, University of Central Lancashire, 1999), and in studies by the present inventors (see Figure 15) indicate that the oxidation waves at +0.9 V and +1.3 V are due to the formation of a passivating layer of gold oxide at the electrode surface.

Upon the reverse cathodic going scan, an oxidative current is observed superimposed on the reduction current of gold oxide. At pH 1, this reoxidation component is clearly present in the potential range between +1.0 V and +0.6 V. This peculiar CV behaviour of AHA on gold shows similarities with the behaviour of hydrogen peroxide, glucose and sorbitol on gold, where gold oxide formation inhibited oxidation and back reduction of gold oxide favoured H O /glucose/sorbitol oxidation. That no corresponding AHA oxidation wave was observed during the anodic going sweep is indicative of the oxide formation and stripping procedure activating the electrode surface in some way.

The form of the oxidation wave between +1.0 V and +0.6 V is strongly suggestive of diffusion controlled wave between +0.6 V and +0.8 V followed by an adsorption post wave between +0.8 V and 1.0 V. Table 1 shows how the Eιy₂, E_1/4, E_3/4 and (E_3/4- Ei_/4) values for the reoxidation wave between +0.6 V and +0.8 V of Figures 11-14 vary with electrode radii. Trends exhibited by Em and (E_3/4-E_1/4) are in keeping with those for diffusion controlled quasi-reversible/irreversible systems. Indeed, a preliminary analysis of the data of Table 1, assuming a two electron transfer for the oxidation of AHA, suggests that AHA has an E^e of +0.62 V vs SCE, a transfer coefficient, α, of 0.66 and a k^e of 3 x 10^"5 m s^"1. However, this approach is not without some risk. For instance, it is by no means certain that the wave observed between +0.6 V and +0.8 V is diffusion controlled. electrode radii / μm E1/2 V E_{1 4}/V E_3/4/V (E. i/4- E1/4) /V

12.5 0.649 0.624 0.670 0.046

25 0.652 0.626 0.675 0.049

62.5 0.645 0.622 0.663 0.041

125 0.640 0.620 0.659 0.039

Table 1 Values of Em, E , E_3/4 and (E_3/4- E_1/4) in Volts (vs SCE) extracted from the cyclic voltammograms of Figures 3-6 atpH 1

It is entirely possible that there is only one wave/peak in the potential range +1.0 to +0.6 V: the separate appearance of a wave and a peak could be due to the superimposition of the gold oxide stripping peak over a single AHA oxidation wave.

In order to deconvolute the effect of the oxide layer generation from the oxidation of AHA, the experiments of Figures 11-14 were repeated using an anodic switching potential of +0.9 V as opposed to +1.5 V. The results are shown in Figures 16-19. As can be seen from Figures 17-19, the first sweep is featureless, save for a single oxidation peak with an Ep of -0.58 V. During subsequent anodic sweeps, two further peaks appear and grow at +0.22 V and +0.4 V. Experiments on gold electrodes in background electrolyte (0.1M HNO₃) indicate that peaks are unambiguously associated with AHA electrochemistry.

The growth of the peaks at +0.22 V and +0.4 V may possibly be attributed to the slight oxidation of the gold surface occurring at +0.9 V during the anodic going sweeps of Figures 16-19. The introduction of oxide species at the gold surface would render it more like the surface of glassy carbon, which is covered with hydrophilic carboxylic acid, carbonyl and -OH groups. Such a conversion in character would be expected to be slow given the small extent of the excursion of the gold electrode potential into the region of oxide generation during each anodic-going sweep. Given the existing analogy between the electrochemistries of H₂O₂ and AHA at gold, we would then expect AHA oxidation to be catalysed by the presence of sub-monolayer oxide/hydroxide species at the surface and that this process will be controlled by AHA diffusion from solution.

It should be emphasised that all three of these peaks exhibit peak currents that are variously 3 to 10 times smaller than the plateau current associated with the wave observed between +0.6 to +0.8 V in the CVs of Figures 11-14. Further, it should be noted that peak growth slows at higher sweep numbers, probably indicating that an equilibrium oxide surface population has been established between the oxide generation step and the oxide stripping process that occurs just cathodic of +0.9 V.

(b) EQCM STUIES OF THE IRON (III)/AHA/NAFION COMPOSITE

(i) Detection of Ligands in the Nafion Film

The retention of HA ligands within the Nafion film was detected visually by placing the modified electrodes in a solution of Fe(III) (lOmM for lmin). Removal of the electrode and examination of the film for colour allowed the deep red colour characteristic of octahedral Fe(III) hydroxamate complexes to be observed. All three hydroxamate/Fe(III) complexes (AHA, GHA⁺ and DFA⁺) exhibited film retention difficulties. Leaching of the complex from the film could be visibly seen when the Nafion/ AHA/Fe(III) composite was placed in the electrolyte solution (0.1M HNO₃) ready for voltammetric investigation.

It is known that the rate of ion exchange between free solution and Nafion membranes is crucially dependent upon the "openness" of the Nafion structure. Thus, films cast from solutions of higher water content have more open structures and so allow faster diffusion of ions. Similarly, it has been found that "wet-cured" films (those dried by ethanolic evaporation under a glass cover) exhibit faster rates of ion exchange than "dry cured" films (those dryed over P₂O₅ desiccant). Thus, in order to obviate problems associated with ligand leaching, a number of Nafion preparative routes have been investigated, giving rise to a more closed layer structure, while at the same time allowing for free passage of the ion analyte into and out of the composite.

(ii) Cyclic Voltammetric Response over a Potential Range of —1.5 to 0 Volts

The visual detection of the hydroxamic ligands within the Nafion film is due to their reaction with Fe(III) upon the partitioning of Fe³⁺ into Nafion film. Iron(III) hydroxamate complexes are electroactive at the Fe(III) centre and can be detected in the film by voltammetry.

Figure 20 shows a cyclic voltammogram of a EQCM Nafion/ AHA/Fe(III) electrode after immersion for lmin in 10 mol m^" Fe(III), in 0.1M HNO₃ electrolyte. The cyclic voltammogram shown in Figure 20 is consistent with that reported for GHA⁺/Nafion composites on glassy carbon electrodes. The two reduction waves on the cathodic sweep can be interpreted as being due to the reduction of ion exchanged (-0.45 V) and complexed (-1.15 V) iron(III) within the Nafion film. On the anodic sweep an oxidation wave can be observed (with an onset of -0.5 V) which can be attributed to the oxidation of ion exchanged Fe(III) within the Nafion film.

The absence of a reoxidation peak associated with the reduced Fe(III)-AHA complex (in the range -1.2 to -0.9 V) is indicative of complex dissociation upon reduction according to:

Fe³⁺ -AHA(ads) + e^~ = Fe²⁺(aq) + AHA(ads)

and is strong evidence for the valence selectivity of AHA toward Fe³⁺ as opposed to

Fe ions. Consequently, we would expect any mass change induced in a HA based

QCM transducer in response to a solution of ferric and ferrous ions to be wholly attributable to Fe uptake. In order to investigate the feasibility of such a transducer, we decided to examine the EQCM behaviour of the AHA/Nafion composite in response to exposure to ferric ions.

(in) Simultaneous Cyclic Voltammetric and EQCM Frequency Response over a Potential Range of —1.5 to 0 Volts

As can be seen from Figure 21, the initial cathodic-going potential scan of the Fe(III)/ AHA/Nafion composite is accompanied by a gradual increase in resonator mass. This may be explained by the fact that in order to avoid problems associated with ligand/complex leaching from the composite, the AHA/Nafion layer was prepared by a loading-drying cycle, with the dry AHA/Nafion-coated QCM resonator then being immersed in a Fe³⁺ solution for 1 minute prior to being removed and placed in the electrochemical cell of the EQCM; thus, in all probability, the previously dried Nafion layer had not fully rehydrated during the Fe immersion step, and further rehydration occurred once placed in the EQCM cell. Thus, the frequency signal recorded during the initial cathodic going potential sweep from 0 to -1.5 V has superimposed on it a gradual decrease corresponding to a increase in the mass of the Nafion layer due to rehydration processes.

This process is most readily seen in the stretches of the frequency signal denoted as AB and CD in Figure 21. The steeper slopes of stretches BC and DE are due to the electrochemical processes associated with the cathodic peaks observed at -0.45 V and -1.15 V i.e. the reduction of Nafion-ion exchanged and AHA-complexed Fe(III) respectively.

The Fe(III)/ AHA Nafion layer contains uncomplexed Fe(III) ions and Fe(III)/AHA complexes adsorbed at immobile ion exchange solvenate sites within the Nafion layer. Therefore, upon electroreduction of Fe(III) sites to Fe(II) electroneutrality must be maintained by cation uptake within the layer. Thus, the mass increases corresponding to the reduction processes at -0.45 V and -1.15 V are due to: (i) Additional water uptake accompanying the adsorption by the composite layer of protons necessary to maintain electroneutrality, and;

(ii) A decrease in the electrostatic crosslinking within the Nafion as Fe³⁺ is reduced to Fe²⁺ so producing a more open layer structure and facilitating further water adsorption.

This is a similar phenomenon to the mass decrease observed to accompany the replacement of protons within a Nafion membrane by large cations such as Ru(NH₃)₆ ^2/3+. As the highly solvated protons are expelled, they are replaced with weakly solvated large cations, so giving rise to a net mass decrease within the Nafion layer. If the number density of sulphonate groups within the layer is known, the number of water molecules accompanying each cation replacement reaction may be calculated, so allowing calibration of the sensor response. Such a calibration could also be achieved for both (generally) the sensor disclosed herein and (specifically) the Fe³⁺ to Fe²⁺ transformation.

Claims

1. A sensor, adapted for the selective detection of cations of a specific valence state, derived from metals capable of existing in more than one valence state, said detector comprising a quartz crystal microbalance which comprises a quartz crystal sandwiched between two electrodes used to generate an alternating electric field across the crystal, the surface of said quartz crystal microbalance being coated with a polymer impregnated with ligands having a high degree of sequestration specificity towards cations of a specific valence state, and the sensor being adapted to provide a variable potential such that changes in the valence state of the sequestered cations may be induced.

2. A sensor as defined in claim 1 wherein said polymer comprises an ion- exchange polymer.

3. A sensor as defined in claim 2 wherein said ion-exchange polymer comprises a perfluorosulphonate cation exchange resin.

4. A sensor as defined in any of claims 1 to 3 wherein said polymer comprises a radiation-hardened polymer.

5. A sensor as defined in any of claims 1 to 4 wherein said ligands comprise at least one of hydroxamic acid and its derivatives.

6. A sensor as defined in claim 5 wherein said derivative of hydroxamic acid comprises formohydroxamic acid or acetohydroxamic acid.

7. A sensor as defined in claim 5 or 6 wherein one face of said quartz crystal is exposed to the solution environment to serve as a working electrode and is biased so as to cause reduction of any species capable of oxidising said hydroxamic acid or its derivative.

8. A sensor as defined in any preceding claim wherein said cations comprise iron(III) cations.

9. A sensor as defined in any preceding claim wherein said cations comprise actinide(4+) cations.

10. A method for the identification and quantitative determination of cations in a composition, said cations having a specific valence state and being derived from metals capable of existing in more than one valence state, said method comprising :

(a) providing a sensor as defined in any preceding claim; and

(b) utilising said sensor for Quartz Crystal Microbalance analysis of said composition.

11. A method as defined in claim 10 wherein selective sequestration of individual cations within the polymer layer is determined by microgravimetric Quartz Crystal Microbalance analysis, Electrochemical Quartz Crystal Microbalance analysis or spectroscopically coupled Quartz Crystal Microbalance analysis.

12. A method as defined in claim 10 or 11 wherein said composition comprises a stream of effluent.

13. A method for the removal of cations from a composition, said cations having a specific valence state and being derived from metals capable of existing in more than one valence state, said method comprising:

(a) providing a sensor as defined in any of claims 1 to 9; (b) utilising said sensor for Quartz Crystal Microbalance analysis of said composition; and

14. A sensor as defined in claim 1 substantially as hereinbefore described and with reference to the accompanying experimental details.

15. A method as defined in claim 10 substantially as hereinbefore described and with reference to the accompanying experimental details.

16. A method as defined in claim 13 substantially as hereinbefore described and with reference to the accompanying experimental details.