Electrochemical Concentration of Lanthanide and Actinide Elements

Abstract

A carbon paste electrode is modified with a chemical agent that is selective for a plurality of lanthanides and actinides (f-series) elements. The modified carbon paste electrode selectively has different voltages applied thereto where a first voltage is used to cause the deposition of one or more lanthanides or actinides from an industrial or environmental sample onto the electrode, and, subsequent to removal of the electrode from the sample and insertion into a second sample where concentration of lanthanides or actinides is preferred, a second voltage is used to cause the deposited lanthanides and/or actinides to be discharged from the electrode for concentration into the second sample.

Description

BACKGROUND

It has been known since the 1950's that separation, pre-concentration, and purification of f-elements (those elements in the lanthanide and actinide series of the periodic table) could be achieved through formation of mercury amalgams.^1,2 Early techniques used mercury pools as the cathode and more recent work has focused on thin mercury films to detect or pre-concentrate f-elements.³ Due to the inherent toxicity of mercury, a stronger focus has been on developing mercury-free techniques.

Since its first reporting in 1958 by Ralph N. Adams⁴, carbon paste (CP) electrodes have become a widely used electrode material in electrochemical research⁵. Many factors have lead to their popularity: low ohmic resistance⁶, large potential window⁵, and ease of modification⁷, to name a few. A quick review of recent literature indicates that CP or modified carbon paste (MCP) electrodes are applicable to aqueous^8,9 and non-aqueous^10,11 matrices in determination of organic^12,13 and inorganic^14,15 elements and compounds. Examples of manufacturing CP and MCP electrodes used as electrochemical sensors and other applications can be found in U.S. Pat. No. 7,968,191 to Hampden-Smith, U.S. Pat. No. 7,901,555 to Jiang, and U.S. Pat. No. 6,828,358 to Morrison, each of which is herein incorporated by reference.

While applications of CP and MCP electrodes are numerous, relatively little is reported in the area of f-elements. Li et al.¹⁶ developed a novel MCP electrode with alizarin used as the complexant modifier. The MCP electrode responded well for the middle to heavy lanthanides with a limit of detection (LOD) of 10⁻¹⁰ M for Ho³⁺ in an acetate buffer. Linear sweep voltammetry (−0.2V to 0.8V vs. SCE) was applied after a 60-120 second pre-concentration period at −0.2 V. A linear, concentration-dependent signal was obtained for the range 10⁻¹⁰ to 10⁻⁷ M, and concentrations of the heavier lanthanides in a dissolved, cast iron sample were quantitatively determined using this electrode. Li's subsequent work¹⁷ focused on Ce³⁺ using the same electrode and similar solution conditions described previously. After optimization, the alizarin MCP electrode exhibited a LOD for Ce³⁺ of 10⁻⁹M and a linear response range of 10⁻⁹ to 10⁻⁷ M.

Ganjali et al.¹⁸ developed an ion selective electrode (ISE) for Ho³⁺ utilizing a MCP electrode containing multi-walled carbon nanotubes, nanosilica and the ionophore N′-(2-hydroxybenzylidene)furan-2-carbohydrazide in addition to graphite. The MCP electrode had a detection limit of 10⁻⁸ M for Ho³⁺ and a linear response range from 10⁻⁷ to 10⁻² M. Additionally, a single conditioned electrode showed a reproducible stable response to standard solutions for up to two months. Continuing the work of Ganjali et al., Norouzi et al.¹⁹ developed an Er³⁺ ISE with the same basic components in the MCP electrode except the ionophore was changed to N′-(2-hydroxy-1,2-diphenylethylidene)benzohydrazide. Response of the electrode to Er³⁺ was similar to that demonstrated by Ganjali et al. with a LOD of 10⁻⁸ M and a linear range of 10⁻⁷ to 10⁻²M.

A critical limitation of these systems is their lack of general applicability to the f-elements as a group of cations. There is a need in the art for systems and methods capable of simultaneously concentrating the full range of f-elements.

SUMMARY

In an embodiment of the invention, MCP electrodes are made operable to selectively concentrate f-elements from a dilute solution and subsequently release the concentrated f-elements for follow-on separation and detection. This embodiment is a mercury free process, and, although not previously recognized in the art, CP and MCP electrodes are particularly suited to perform this type of concentration of f-elements.^20,21 The terms CP and MCP electrodes are at times used interchangeably in this description, however, it will be understood that the invention is focused on a carbon paste electrode, sometimes referred to as a conducting ink/paste electrode which can be formed from carbon black, graphite, carbon powder, carbon flake, carbon nanotubes, etc., where the carbon is modified to permit selective binding of f-series elements (as opposed to an unmodified carbon paste electrode of a carbon paste electrode which cannot selectively bind f-series elements).

In another embodiment of the invention, an MCP electrode is provided wherein the modifying agent is a chemical agent with a selective binding affinity for f-elements. Exemplary modifying agents for the MCP electrode which permit selective binding of f-elements are selected from at least one of 1,2-dihydroxybenzene-3,5-disulfonic acid, 2-hydroxyisobutyric acid, trimetaphosphoric acid, trans-1,2-cyclohexylenedinitrilotetraacetic acid, 2-hydroxy-2-methylpropanoic acid, iminodiacetic acid, nitrilotriacetic acid, ethylenedinitrilotetraacetic acid, diethylenetriamine-pentacetic acid, 2,2′,2″, 2″-(1,4,7,10tetraazacyclododecane-1,4,7,10-tetrayl)tetraacetic acid, N-(2-carboxyphenyl)iminodiacetic acid, dihydroxycyclobutenedione, 3,6,9,12-tetrakis(carboxymethyl)-3,6,9,12-tetraazatetradecane-1,14-dioic acid and derivatives of these compounds. Derivatives comprise compounds with identical f-element ligation sites and a similar structural motif (e.g. 2-hydroxyisobutyric acid and 2-hydroxybuteric acid).

In yet another embodiment of the invention, a method for concentrating f-elements is provided. This method includes contacting a modified carbon paste (MCP) electrode, wherein the modifying agent is a chemical agent with a selective binding affinity for f-elements, with an environmental or industrial sample containing at least one f-element; applying a voltage to the electrode suitable to deposit f-elements onto the electrode; moving the electrode to a second sample of a volume less than the first sample and; applying a second voltage suitable to release f-elements from the electrode thereby concentrating the f-elements in the second volume (because the second volume is smaller than the first, the f-elements are more concentrated in the second volume; however, it will be recognized that this method could also be used to simply transfer f-elements from one volume to another without limiting the size of the second volume). In variations on this method, the electrode can be cycled between environmental or industrial samples and a second sample applying the first voltage within the environmental or industrial sample and the second voltage within the second sample.

The MCP electrodes comprise a conductive wire, well or surface with a pre-mixed suspension of a conductive graphite and chemical modifying agent with a binding liquid are disposed thereon. See [Joseph Wang, Balashaheb K. Deshmukh, Mojtaba Bonakdar, Solvent extraction studies with carbon paste electrodes, Journal of Electroanalytical Chemistry and Interfacial Electrochemistry, Volume 194, Issue 2, 25 Oct. 1985, Pages 339-353.] for examples of different graphite-binding liquid combinations for CP electrodes which may be employed in the practice of this invention.

Ease of modification is one of the most valuable features of CP electrodes.⁵ This is due to the well-developed surface of CP which has a high adsorptivity.²² Modification of CP electrodes can be achieved through a multitude of methods. A few of these methods are: chemical pre-treatment where the carbon is soaked in the modifier and then evaporated to dryness before being prepared as an electrode;²³ in situ modification where the modifier adsorbs to the surface of plain CP electrode thus allowing for determination of analyte in the solution;²⁴ dissolution in the binding liquid which is typically achieved through the use of an ion-exchange resin;²⁵ or direct mixing of dry modifiers into the paste through mechanical means which is believed to be the most frequently used method.²²

DESCRIPTION OF THE DRAWING FIGURES

FIG. 1: Panel A is a CV scan showing the electrochemical response of a CP electrode in 0.1 M LiCl. Panel B is a HIBA-CP, solid line, electrode response in the same 0.1 M LiCl solution. For comparison, the CP electrode response is plotted in Panel B as the dashed line. Scan rate was 100 mV/s, pH 3.5 and potential reported vs. Ag/AgCl. Scan begins and ends at 0.8 V for both voltammograms.

FIG. 2: A family of scans for a single HIBA-CP electrode in 0.1M LiCl, pH 3.5. Scan rate was varied from 1 to 500 mV/s. The ratio of i_c/i_a was 1.12±0.03 suggesting chemical reversibility. General wave shape did not follow the Nernst equation indicating electrochemical irreversibility.

FIG. 3: Background subtracted CV for 0.1 M HIBA on a 3 mm Pt working electrode. Background electrolyte was 0.1 M LiCl, pH 3.5, scan rate 100 mV/s. Potential is reported vs. Ag/AgCl. Nernstian and faradaic comparison to 0.1 mM K₃[FeCN₆] on the same electrode indicates an electron transfer of 1.2±0.2 electrons.

FIG. 4: Anson Plot for CP and HIBA-CP electrode response to La³⁺. Solid line represents the response of HIBA-CP to a mixture containing 0.1M LiCl and 1×10⁻⁵ M La³⁺. Dashed line is CP response to an identical mixture of LiCl and La³⁺, The CP line intersects on the x-axis indicating no sorption phenomenon. The HIBA-CP intersection above the x-axis is indicative of a sorption phenomenon.

FIG. 5: Results of 1 second DPSC experiments with a HIBA-CP electrode. Signal intensity was measured at 500 msec and background subtracted. The concentration range for La³⁺ covered from 10⁻⁷ M to 10⁻³ M, n=3 for each datum point. Errors bars are ±1 σSD. The shape of concentration dependent response is indicative of a sorption isotherm.

FIG. 6: ICPMS results for pre-concentration of select f-elements. LOD was determined from 7 blank runs +3 σSD. Closed circles represent the results from calibration standards (1 ppb-1 ppt). Open triangles represent the pre-concentration results of 5 ppq with the HIBA-CP electrode, n=5. Open diamonds represent pre-concentration results of 5 ppq with a CP electrode. Open squares are the results of 5 ppq solutions with no pre-concentration. Errors bars are 1 σSD most of which are within the dimensions of the symbols. Y-axis for all graphs are expressed in terms of counts per second.

FIGS. 7A-C: Schematic representations of a modified carbon paste electrode suitable for use in device operable to electrochemically transfer f-series elements from a first medium to a second medium.

FIG. 8: Process overview for a device operable to electrochemically transfer f-series elements from a first medium to a second medium.

DETAILED DESCRIPTION

The description below shows the fabrication and testing of certain exemplary electrodes and methods according to the invention. It will be recognized by those skilled in the art that the electrodes, the materials used for their fabrication, and methods of use can be varied within the spirit and scope of the appended claims.

Methods and systems for the electrochemical transfer of f-series element constituents present in a first medium to a second medium are described herein. A modified paste electrode can be utilized to accumulate f-series elements within the paste when the voltage applied to the electrode is held at a first voltage and subsequently released from the paste when the voltage applied to the electrode is held at a second voltage. The ability to accumulate f-series elements enables a range of applications wherein said elements can be transferred between media. In particular embodiments the transfer of said elements can serve to concentrate the elements (e.g. when the volume of the second medium is less than the volume of the first medium or where the elements from plurality of first media are deposited in a common second medium).

An exemplary modified carbon paste electrode is schematically depicted in FIG. 7AC. In reference to FIG. 7A a typical implementation for a modified carbon paste electrode comprises a paste 703 on the surface (portion of the component in diffusive communication with the media) of a conductive element 702 encased within an insulating housing 701, and a connector 704 operable to connect the conductive element 702 to an external voltage source. The implementation depicted in FIG. 7A provides an example wherein the surface or tip of the conductive element 702 is coplanar with the tip of the insulating housing 701. Other embodiments may comprise configurations where the tip of conductive element 702 and paste 703 are not coplanar. In particular embodiments the paste 703 may also be present in a void in the insulating housing 701 as depicted in FIG. 7B. In such implementations the surface of the conductive element may not extend to the tip of the insulating housing. FIG. 7B may also comprise an embodiment wherein the conductive element 702 is a porous material and the paste has been distributed into the pores of the conductive element 702. In yet further embodiments, the tip of the conductive element 702 may extend further from the tip of the insulating housing 701 as depicted in FIG. 7C. Common to all configurations is that the paste 703 is applied to the portion of the conductive element 702 that is in diffusive communication with the external medium (not shown).

The insulating housing 701 may comprise any material that does not conduct electrical current. Examples include, but are not limited to Glasses, Ceramics, polyethylene, crosslinked polyethylene (either through electron beam processing or chemical crosslinking), polyvinyl chloride (PVC), Kapton, rubber-like polymers, oil impregnated paper, Teflon, silicone, and/or modified ethylene tetrafluoroethylene (ETFE). The conductive element 702 may comprise a material that can conduct electrical current. Examples include, but are not limited to platinum, gold, silver, glassy carbon, brass, copper, graphite, porous graphite, and/or molybdenum, and combinations thereof.

The paste 703 applied to the portion of the conductive element 702 that is in diffusive communication with the external medium is composed of three general components: a binder, a conductive component, and a modifier. The binder serves to adhere the paste 703 to the surface of the conductive element 702 and provide a fluid like medium to uniformly disperse the conductive component and modifier within the binder. Traditional binders comprise organic liquids which link mechanically the conductive component and modifier. However, besides this main function, the binder as the second main moiety of carbon paste co-determines its properties. Typical parameters required for binders are: i) chemical inertness and electroinactivity, ii) high viscosity and low volatility, iii) minimal solubility in aqueous solutions, and iv) immiscibility with organic solvents. Example binding agents (binders) used for preparation of carbon pastes include, but are not limited to; mineral (paraffin) oils; namely, i) Nujol or a similar trade-mark product and solvent for spectroscopy ii) Uvasol iii) aliphatic and aromatic hydrocarbons, including their iv) halogenated derivatives, as well as v) silicone oils and greases, or nearly solid silicone rubbers.

The conductive component within the paste typically comprises a carbonaceous material. In particular embodiments powdered carbon (graphite) as the conductive component within the paste provides for proper function of an electrode or a sensor in electrochemical measurements. Suitable carbonaceous materials should obey the following criteria: i) particle size in micrometers, ii) uniform distribution of the particles, iii) high chemical purity, and iv) low adsorption capabilities. Naturally, the type and quality of graphite used, as well as its overall amount in the carbon paste mixture, are reflected in all typical properties of the respective mixture. A typical carbon powder comprises spectroscopic graphite with particles in the low micrometric scale (typically, 5-20 mm). Alternatives to graphite include but are not limited to i) soot and charcoal, ii) acetylene black, iii) glassy carbon powders with globular particles, iv) pulverized diamond of both natural and synthetic origin, v) template carbon, vi) porous carbon foam, and vii) carbon microspheres viii) fullerenes, ix) carbon nanofibers or various types of x) carbon nanotubes. In general the conductive component is present in a concentration of between 5 g/ml binder and 0.2 g/ml binder within the paste.

The modifier in the paste generally comprises an organic compound that contains a (or a plurality of) functional group that demonstrates a preference for ligating to f-series elements. In the preferred embodiment the organic compound has an affinity for a range of f-series elements. In yet further embodiments a plurality of organic compounds may be incorporated as modifiers wherein each organic compound provides a preference for binding a distinct subset of elements within the to f-series elements. Examples of organic compounds include but are not limited to, 1,2-dihydroxybenzene-3,5-disulfonic acid, 2-hydroxyisobutyric acid, trimetaphosphoric acid, trans-1,2-cyclohexylenedinitrilotetraacetic acid, 2-hydroxy-2-methylpropanoic acid, iminodiacetic acid, nitrilotriacetic acid, ethylenedinitrilotetraacetic acid, diethylenetriamine-pentacetic acid, 2,2′,2″,2′″-(1,4,7,10tetraazacyclododecane-1,4,7,10-tetrayl)tetraacetic acid, N-(2-carboxyphenyl)iminodiacetic acid, dihydroxycyclobutenedione, 3,6,9,12-tetrakis(carboxymethyl)-3,6,9,12-tetraazatetradecane-1,14-dioic acid and derivatives of these compounds. Derivatives comprise compounds with identical f-element ligation sites and a similar structural motif (e.g. 2-hydroxyisobutyric acid and 2-hydroxybuteric acid). In general the modifier (or modifiers) is present in a concentration of between about 0.1 mM and 20 mM within the binder. The concentration of modifier can be tailored to the f-element content within the intended operational media.

FIG. 8 provides an operational process for accumulating f-elements present in a medium and transferring the f-elements to a second medium. A modified paste electrode can be utilized to accumulate f-series elements within the paste when the voltage applied to the electrode is held at a first voltage within a first medium. Implementations wherein the first medium is an aqueous medium the applied voltage required is about −0.05 to −0.05 V (vs. Ag/AgCl). The duration of time necessary for accumulation will vary with the specific implementation. Once the f-elements are accumulated, the electrode can be transferred from the first medium to a second medium. An optional cleaning step may be added during this transfer; here the cleaning may involve a physical or chemical cleansing of the body of the electrode. Once the electrode is placed in the second medium, a second voltage is applied to release the f-elements accumulated in the paste. Implementations wherein the second medium is an aqueous medium the applied voltage required is about 0.1 to 0.09 V (vs. Ag/AgCl). The specific voltage applied should be dependent upon both the operational media and the organic compounds present within the paste. More specifically, the accumulation voltage should be a voltage wherein the organic group is reduced and the release or stripping voltage should be at a voltage wherein the organic group is in it's electrochemical ground state (e.g. not oxidized or reduced). The process of accumulation and stripping may be repeated or cycled between media.

In a specific particular embodiment, Lanthanide cations in solution can be rapidly sequestered onto and subsequently removed from a modified carbon paste electrode by cycling the voltage of the electrode. The electrode comprises a paste produced by mechanically mixing 5 grams of Acheson 38 carbon with 3 milliliters of paraffin oil. Prior to forming the paste with the carbon, the paraffin oil is modified by mixing approximately 5 millimoles (300 milligrams) of alpha-hydroxyisobutyric acid (HIBA) into the 3 milliliters of paraffin oil. The required range of concentration of HIBA in paraffin oil is 0.1-20 mol/L. This paste is then applied to the end of a Teflon electrode body. As a group of metal cations, the lanthanides accumulate from a solution of 0.1 M LiCl onto the carbon paste surface within 30 second when a voltage of −0.4 V (vs. Ag/AgCl) is applied. The sorbed lanthanide cations can then be quantitatively stripped off the electrode surface into a different solution by applying an oxidative step of +0.8 V (vs. Ag/AgCl).

EXAMPLES

Materials: Reagent grade graphite, LiCl, paraffin oil and 2-hydroxyisobutyric acid (HIBA) were used as received, from Fisher Scientific, (Waltham Mass. USA, www.fishersci.com). For the multi-element analysis, a stock solution containing 10 ppm of analytes (Sc, Y, La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu and Th) in 2% HNO₃ was purchased from High-Purity Standards, (Charleston S.C. USA, www.highpuritystandards.com). This solution was used as received.

Electrode Fabrication: Procedures for preparation of the CP electrode were modified from Adams.²⁶ In brief, 5 grams of Acheson 38 grade graphite was mechanically mixed with 3 mL of paraffin oil using a glass mortar and pestle to produce a thick, uniform paste. To prepare the HIBA-CP electrode, 5 mmoles of solid HIBA were added to 3 mL of paraffin oil and mixed until the slurry was homogenous. Then five grams of carbon were added and mixed to form the paste. Procedures for smoothing and renewing the electrode followed those of Adams, which is herein incorporated by reference.²⁶

Electrochemical Procedures: Electrochemical measurements were made using a Model 100B potentiostat, 15 mL Teflon™ electrochemical cell, and 3 mm CP Teflon™ electrode body purchased from Bioanalytical Systems Inc. (West Lafayette Indiana USA, www.basinc.com). The 15 mL Teflon™ electrochemical cell was constructed and setup similarly to Schumacher et al.³ The reference and auxiliary electrodes were Ag/AgCl and Pt wire, respectively. Solutions within the electrochemical cell were purged with purified Ar prior to conducting an experiment. Cyclic voltammetric (CV) experiments were typically scanned at a rate of 100 mV/s starting at +0.8 V to −0.4 V. Double potential step chronoamperometric (DPSC) and chronocoulometric (CC) experiments were stepped from +0.8 V to −0.4 V and back to +0.8 V vs. Ag/AgCl. The background electrolyte was 0.1 M LiCl that was pH adjusted using 2% HNO₃. Pre-concentration and stripping experiments followed the procedures outlined in Wang^27,28 with modifications. After a deposition step of 30 sec, the CP or HIBA-CP electrode was removed from the cell, wiped with a Kim-Wipe on the insulating shroud and transferred to a separate vial containing 2 mL of 2% HNO₃. A stripping step from −0.4V to +0.8V vs. Ag/AgCl for 30 sec, was performed and the solution analyzed by ICPMS. A typical experiment involving conditioning the CP or HIBA-CP electrode, pre-concentration of the trivalent f-element and stripping into 2% HNO₃ required approximately five minutes. The cell was cleaned between experiments following the procedures of Schumacher et al.³

ICPMS Procedures: Inductively Coupled Plasma Mass Spectrometry (ICPMS) measurements were performed on an Agilent 7500 ICPMS utilizing an internal indium and rare earth standard and scanned in the positive mode. Prior to analyzing any samples, the instrument was calibrated with a set of prepared standards in 2% HNO₃ and plain 2% HNO₃ was used as the blank to correct for background.

Analyses of electroanalytical and ICPMS data were performed using GraphPad Prism version 5.02 for Windows, (GraphPad Software, San Diego Calif. USA, www.graphpad.com).

FIG. 1A shows the electrochemical response for a CP electrode to 0.1M LiCl at pH 3.5. While this voltammogram is only the response to background electrolyte, a series of scans were performed in various solutions containing HIBA and/or f-elements to determine if CP exhibited any electrochemical response. In these cases, the CP electrode showed no electrochemical response to dissolved f-elements.

FIG. 1B shows the electrochemical response for a HIBA-CP electrode to 0.1M LiCl at pH 3.5; the pKa for HIBA is 3.7.²⁹ For this voltammogram, the scan rate was 100 mV/s and started at +0.8 V. A large reduction and oxidation signal was observed for the HIBA-CP electrode. Each new HIBA-CP electrode typically required two to three conditioning scans to achieve a stable (exceeding three hours) electrochemical response. Variability of peak intensity between electrodes was less than 10%, which falls within the range expected for MCP electrodes.²⁷ Additionally, the cathodic and anodic peaks varied less than 5% between electrodes for a given scan rate. This demonstrates that once conditioned, the HIBA-CP electrode yields a stable and reproducible response.

Watanabe et al.³⁰ reported that f-elements will adsorb to carbonaceous material in acidic environments. Our result for the CP electrode is not in disagreement with Watanabe et al. as the contact time for their study was much longer (3-4 hours)³⁰ than the time period used in this study (1-2 min).

FIG. 2 shows a series of CVs for a single HIBA-CP electrode in 0.1 M LiCl where the scan rate was varied from 1 to 500 mV/s. The measured ratio of cathodic (i_c) and anodic (i_a) peak intensities was constant across scan rates at 1.12±0.03 (n=3) suggesting the electrochemical response from the HIBA-CP is chemically reversible. The plots of scan rate, u, versus i_p and v^1/2 versus i_p were inconclusive as to whether the observed results represent an adsorption or diffusion phenomenon. This result is not surprising given the range of v used. The shapes of the voltammograms did not follow the Nernst equation, suggesting that electrochemical charge transfer is irreversibile.

To further evaluate diffusion vs. adsorption, chronoamperometric analysis of the voltammetric wave shapes were conducted using the Cottrell equation:

$\begin{array}{cc}i=\frac{\mathrm{nFAC}\sqrt{D}}{\sqrt{\pi t}}& \left(1\right)\end{array}$

where i=current (amps), n=number of electrons, F=Faraday constant (96,485 C/mol), A=area of the electrode (cm²), C=initial concentration of the analyte (mol/cm³), D=diffusion coefficient for the species (cm²/s), and t=time (s) was used to evaluate the waveforms. A plot of t^−1/2 vs. i deviated from linearity based on time of exposure to the analyte, suggesting other processes were either occurring at the surface of the electrode or impeding diffusion to the electrode surface.^31,32 A series of experiments were performed at different values of pH and no pH effects were observed, suggesting that the observed phenomenon is occurring on the surface of the electrode and not a direct result of solution conditions.

To determine the number of electrons transferred per mole of HIBA, a solution containing 0.1 mM K₃[FeCN₆] and 0.1 M LiCl, which has a known 1 e− transfer, [Fe(CN)₆]³⁻+e⁻⇄[Fe(CN)₆]⁴⁻, was analyzed by CV on a 3 mm Pt electrode to determine the i_p and integrated voltammetric wave area. A separate solution containing 1 mM HIBA and 0.1 M LiCl was analyzed by CV on the same Pt electrode. (FIG. 3) Comparison of the ratios of magnitude of i_p and integrated voltammetric wave areas for HIBA to K₃[FeCN₆] indicated an electron transfer of 1.2±0.2 electrons per mole of HIBA.

Kvaratskhelia and Kvaratskhelia³³ examined the voltammetric responses of hydroxycarboxylic acids in aqueous solutions using various solid electrodes. Their E_1/2 values of the observed waves on Pt in 0.1 M NaClO₄ occurred in the range of −0.47 to −0.49 V vs. a saturated calomel electrode. Our voltammetric response for HIBA is in agreement with their observed results.

In a MCP electrode study involving complexes between rare earths and alizarin, Li et al.¹⁶ reported a 2 e⁻ charge transfer irreversible process for alizarin that was not pH dependent. They point out that most electrode processes of organic compounds involve proton ion transfers thus a pH dependence is expected; however, in the case of alizarin this was not observed. Our characterization of MBA, a 1 e⁻ irreversible process with no pH dependence, agrees with the characterization reported by Li et al.¹⁶ for alizarin.

FIG. 4 shows the resultant Anson plot³⁴ for a three second chronocoulometric experiment using both a HIBA-CP and CP electrode in 0.1 M LiCl and 1×10⁻⁵ M La³⁺. The reduction lines are plotted as t^1/12 vs. Q and the oxidation lines are plotted as θ vs. Q, where θ=[τ^1/2+(t−τ)^1/2−t^1/2] as defined by Anson.³⁴ For HIBA-CP the reduction and oxidation lines exhibit different slopes with an intersection above the x-axis. The CP reduction and oxidation lines have nearly identical absolute values of their slopes and the lines intersect on the x-axis. One second and five second chronocoulometric experiments were also performed with nearly identical results to the three second experiment (data not shown). This range of timescales were chosen to minimize contributions from convective mass transport, which occurs at solid electrodes at time periods greater than 5 seconds.²⁶ According to Anson, intersection above the x-axis represents the total coulombs of adsorbed reactant because this analysis of chronocoulometric data effectively removes any contribution due to double layer charging.^34,35 Using the intersection value, 1.1±0.3×10⁻⁶ C, with Faraday's Law; the total number of moles accumulated equals 3.6±0.7×10⁻¹², n=3.

FIG. 5 is the results of chronoamperometric experiments with a HIBA-CP electrode in varying solution concentrations of La³⁺ (10⁻⁷ M to 10⁻³ M) with 0.1 M LiCl as the background electrolyte at pH 3.5. The data points were obtained by running 1 second chronoamperometric experiments (an experimentally convenient time interval) and measuring a background corrected i value at 500 ms. This time span was chosen because it excludes distortion due to charging current and minimizes contributions due to convective mass transport.³⁶ A new HIBA-CP electrode was used for each change in the concentration of La³⁺ and each point represents the average of triplicate runs. The shape of the graph shows a concentration dependent signal suggesting a sorption phenomenon.

FIG. 6 shows representative results from the pre-concentration experiments. The multi-element standard used contained all the lanthanides, minus promethium, plus scandium, yttrium, and thorium. The four graphs represent the range of responses of the lanthanides and shows that the HIBA-CP electrode will pre-concentrate above the limit of detection (LOD) for the ICPMS while the CP electrode does not pre-concentrate under the same conditions as the HIBA-CP electrode. The counts for all the lanthanides were totaled and applied to a calibration curve to determine total moles accumulated, 3.0±0.6×10⁻¹², n=3. Comparing this number to the total moles adsorbed via the Anson plot in FIG. 5, 3.6±0.7×10⁻¹² we find good agreement indicating that the HIBA-CP electrode accumulates individual lanthanides or a mixture with equal efficiency.

Interestingly, for Sc, Ce and Th the HIBA-CP electrode did not pre-concentrate above LOD. A possible explanation for the case of Sc is that while in the same group as La, Sc responds in solution more as a d-element while Y, which does pre-concentrate, responds more like an f-element.

To gain some insight into the mechanism of HIBA-CP pre-concentration, a comparison of stepwise formation constants (log K) values for HIBA in 0.1M ionic strength from Martell and Smith²⁹ and the total amount of f-element pre-concentrated by the HIBA-CP electrode was conducted. Since HIBA exhibits a systematic increase in log K values across the series of lanthanides,³⁷ one would expect that if HIBA-CP pre-concentration capability was solely a function of HIBA in the electrode, then a similar trend would be observed. While in general, heavier lanthanides pre-concentrated more readily than lighter lanthanides, no direct comparison could be made indicating that more factors are involved in HIBA-CP pre-concentration capability and further work is required to elucidate these factors.

While this work has been performed in a neat solution, interferences are expected since HIBA complexes to some extent with most metal cations present in solution. While many factors affect the strength of metal-ligand complexes, a good first approximation for determining potential interferences are thermodynamic stability constants. Nash and Jensen thoroughly discuss the solution chemistry aspects of metal-ligand complexes and provide an excellent justification for the use of stability constants for initial approximations. Surprisingly, while HIBA has been in use since its first reporting in 1956, relatively little critically reviewed thermodynamic stability constant data are available for metal cations other than the f-elements.²⁹ While no stability constant data exists for a Li⁺-HIBA complex, taking considerations of ionic charge, radius, and strength of ion-dipole interactions, we estimate that Li⁺ interactions with HIBA are minimal, resulting in little interference. In our case, Li⁺ was in 100.000-fold excess of trivalent f-elements and did not serve as a major interference.

The experiments above shows that an MCP electrode can be used to selectively bind f-elements. In the preferred embodiment, the modifying agent is a chemical agent with a selective binding affinity for f-elements. These modifying agents are included in the CP at a level of at least 0.1 mmol/L (e.g. 10 milligrams of HIBA per liter of paraffin oil) but less than or equal to 20 mmol/L (e.g. 2,000 milligrams of HIBA per liter of paraffin oil).

These modifying agents are included in the CP at a level of at least 0.1 mol/L but less than or equal to 20 mol/L (e.g. less than or equal to 10 mol/L).

Combination of the f-element concentration methods described herein with the lanthanide separation methods described in Clark et. al. [Journal of Radioanalytical and Nuclear Chemistry Volume 282 Issue 2 Pages 329-333 2009], which is herein incorporated by reference, provides means for concentration and separation of f-elements from industrial and/or environmental samples.

REFERENCES

• 1 Onstott, E. I. J. Am. Chem. Soc., 1956, 78, 2070-2076.
• 2 Onstott, E. I. J. Am. Chem. Soc., 1959, 81, 4451-4458.
• 3 Schumacher, P. D.; Woods, N. A.; Schenk, J. O.; Clark, S. B. Anal Chem. 2010, 82, 5663-5668
• 4 Adams, R. N. Anal. Chem., 1958, 30, 1576.
• 5 Svancara, I.; Vytras, K.; Kalcher, K.; Walcarius, A.; Wang, J. Electroanalysis, 2009, 21, 7-28.
• 6 Vytras, K.; Svancara, Metelka, R. J. Serb. Chem. Soc., 2009, 74, 1021-1033.
• 7 Arrigan, D. W. M. Analyst, 1994, 119, 1953-1966.
• 8 Munoz, C.; Zuniga, M.; Arancibia, V. J. Braz. Chem. Soc., 2010, 21, 1688-1691.
• 9 Ojani, R.; Raoof, J.; Babazadeh, R. Electroanalysis, 2010, 22, 1607-1616.
• 10 Mehrgardi, M.; Barfidokht, A. J. Electro. Chem., 2010, 644, 44-49.
• 11 Singh, A.; Singh, P.; Bhattacharjee, G. Talanta, 2009, 80, 685-693.
• 12 Wang, Z.; Sun, Z.; Zhu, H.; Gao, G.; Liu, H.; Zhao, X. Electroanalysis, 2010, 22, 1737-1742.
• 13 Luo, L.; Wang, X.; Li, Q.; Ding, Y.; Jia, J.; Deng, D. Anal. Sci., 2010, 26, 907-911.
• 14 Li, J. H.; Zhang, F. X.; Wang, J. Q.; Xu, Z. F.; Zeng, R. Y. Anal. Sci., 2009, 25, 653-657.
• 15 Cardoso, W.; Oliveira da Fonseca, T.; Marques, A.; Marques, E. J. Braz. Chem. Soc., 2010, 21, 1733-1738.
• 16 Li, J.; Liu, S.; Mao, X.; Gao, P. Yan, Z. J. Electroanal. Chem., 2004, 561, 137-142.
• 17 Li, J.; Liu, S.; Yan, Z.; Mao, X.; Gao, P. Microchim. Acta, 2006, 154, 241-246.
• 18 Ganjali, M.; Motakef-Kazemi, N.; Norouzi, P.; Khoee, S. Int. J. Electrochem. Sci., 2009, 4, 906-913.
• 19 Norouzi, P.; Rafiei-Sarmazdeh, Z.; Faribad, F.; Adibi, M.; Ganjali, M. Int. J. Electrochem. Sci., 2010, 5, 367-376.
• 20 Lee, G.; Kim, C.; Lee, M.; Rhee, C. J. Electro. Chem. Soc., 2010, 157, 241-244.
• 21 Tesarova, E.; Baldrianova, L.; Hocevar, S.; Svancara, I.; Vytras, K.; Ogorevc, B., Electrochim. Acta, 2009, 54, 1506-1510.
• 22 Stozhko, N.; Malakhova, N.; Fyodorov, M.; Brainina, K. J Sol. State Electrochem., 2008, 12, 1185-1204.
• 23 Baldwin, R.; Christensen, J.; Kryger, L. Anal Chem, 1986, 58, 1790-1798.
• 24 Raber, G.; Kaleher, K.; Neuhold, C. Talaber, C.; Kolbl, G. Electroanal., 1995, 7, 138-142.
• 25 Lupetti, K.; Vieira, I.; Vieira, H.; Fatibello, O. Analyst, 2002, 127, 525-529.
• 26 Adams, R. N. Electrochemistry at Solid Electrodes, Dekker, New York, 1969.
• 27 Wang, J in Electroanalytical Chemistry, Vol 16; Bard, A. J., Ed.; Dekker, New York, 1989; pp 1-81.
• 28 Wang, J. Stripping Analysis, VCH Publishers, Florida, 1985.
• 29 Martell, A; Smith, R. Critical Stability Constants, Vol 6; Plenum Press, New York, 1997.
• 30 Watanabe, M; Arisaka, M; Kimura, T. Abstract of Papers, 239^th ACS Nat. Mtg. 2010, 46.
• 31 Bard, A. J.; Faulkner, L. R. Electrochemical Methods, Wiley, New York, 1980
• 32 Santos, I.C; Vila-Boras, M; Piedade, M; Friere, C; Duarte, M; Castro, B. Polyhedron, 2000, 19, 655-664.
• 33 Kvaratskhelia, R.; Kvaratskhelia, E. Rus. J. Electrochem., 2008, 44, 230-233.
• 34 Anson, F. Anal. Chem. 1966, 38, 54-57.
• Christie, J; Osteryoung, R; Anson, F. J Electroanal. Chem., 1967, 13, 236-244.
• 36 Santos, D.; Sequeira, C. ECS Transactions, 2008, 16, 1855-1868.
• 37 Clark, S. B.; Friese, J. I. J. Radioanal. Nucl. Chem., 2009, 282, 329-333.

Claims

1. A modified carbon paste electrode, comprising: an electrically conductive substrate;a carbon paste adhered to a surface of said substrate; andone or more modifying agents dispersed within said carbon paste, wherein at least one of said modifying agents selectively binds to a plurality of lanthanide and actinide elements.

2. The modified carbon paste electrode of claim 1 wherein said one or more modifying agents include a plurality of modifying agents.

3. The modified carbon paste electrode of claim 1 wherein said carbon paste comprises a binder and a conductive component, and wherein said at least one modifying agent and said conductive component are uniformly dispersed in said binder.

4. The modified carbon paste electrode of claim 3 wherein said binder is selected from the group consisting of mineral (paraffin) oils, aliphatic and aromatic hydrocarbons, and silicone oils, greases, and rubbers.

5. The modified carbon paste electrode of claim 1 wherein said substrate includes a conductive element and an insulative housing.

6. The modified carbon paste electrode of claim 5 wherein said insulative housing is made from a material selected from glasses, ceramics, polyethylene, crosslinked polyethylene, polyvinyl chloride (PVC), Kapton, rubber-like polymers, oil impregnated paper, Teflon, silicone, and modified ethylene tetrafluoroethylene (ETFE).

7. The modified carbon paste electrode of claim 5 wherein said conductive element is made from platinum, gold, silver, glassy carbon, brass, copper, graphite, porous graphite, molybdenum, and combinations thereof.

8. The modified carbon paste electrode of claim 1 wherein the at least one modifying agent is selected from the group consisting of 1,2-dihydroxybenzene-3,5-disulfonic acid, 2-hydroxyisobutyric acid, trimetaphosphoric acid, trans-1,2-cyclohexylenedinitrilotetraacetic acid, 2-hydroxy-2-methylpropanoic acid, iminodiacetic acid, nitrilotriacetic acid, ethylenedinitrilotetraacetic acid, diethylenetriamine-pentacetic acid, 2,2′,2″, 2′″-(1,4,7,10tetraazacyclododecane-1,4,7,10-tetrayl)tetraacetic acid, N-(2-carboxyphenyl)iminodiacetic acid, dihydroxycyclobutenedione, 3,6,9,12-tetrakis(carboxymethyl)-3,6,9,12-tetraazatetradecane-1,14-dioic acid and derivatives of these compounds.

9. A retrieving and depositing method for retrieving one or more lanthanides or actinides from a first solution and depositing said one or more lanthanides or actinides in a second solution, comprising the steps of contacting a first solution which contains one or more lanthanides or actinides with a modified carbon paste electrode comprised of an electrically conductive substrate with a carbon paste adhered to a surfaces of said substrate and with one or modifying agents that selectively bind lanthanides or actinides dispersed within said carbon paste;applying a first voltage to the modified carbon paste electrode suitable to deposit said one or more lanthanides or actinides contained in said sample onto the modified carbon paste electrode;moving the modified carbon paste electrode to a second solution; andapplying a second voltage suitable to release of said one or more lanthanides or actinides into said second sample.

10. The method of claim 9 wherein said first solution is an environmental sample.

11. The method of claim 9 wherein said first solution is an industrial sample.

12. The method of claim 9 wherein said second solution is of a smaller volume than said first solution, wherein said steps of contacting, applying, moving, and applying concentrate said one or more lanthanides or actinides in said second solution.

13. The method of claim 9 wherein the modified paste electrode is repeatedly cycled between the first solution and said second solution, and said step of applying the first voltage is performed within the first solution and said step of applying the second voltage is performed within the second solution.

14. The method of claim 9 wherein said chemical agent is selected from the group consisting of 1,2-dihydroxybenzene-3,5-disulfonic acid, 2-hydroxyisobutyric acid, trimetaphosphoric acid, trans-1,2-cyclohexylenedinitrilotetraacetic acid, 2-hydroxy-2-methylpropanoic acid, iminodiacetic acid, nitrilotriacetic acid, ethylenedinitrilotetraacetic acid, diethylenetriamine-pentacetic acid, 2,2′,2″,2″-(1,4,7,10tetraazacyclododecane-1,4,7,10-tetrayl)tetraacetic acid, N-(2-carboxyphenyl)iminodiacetic acid, dihydroxycyclobutenedione, 3,6,9,12-tetrakis(carboxymethyl)-3,6,9,12-tetraazatetradecane-1,14-dioic acid and derivatives of these compounds.

15. The method of claim 14 wherein said chemical agent is reduced by said step of applying said first voltage and wherein said chemical agent is in a ground state by said step of applying said second voltage.

16. The method of claim 9 wherein the first voltage ranges from −0.05 to −0.05 V and wherein the second voltage ranges from 0.1 to 0.09 V.