Electrochemical Determination of Low Levels of Uranyl by a Vibrating

Dec 1, 2014 - Phone: +972 2 6585831. Fax: +972 2 6585319. Cite this:Anal. Chem. 87, 1, 768-776. Abstract. Abstract Image. In this work we report the s...
0 downloads 6 Views 2MB Size
Article pubs.acs.org/ac

Electrochemical Determination of Low Levels of Uranyl by a Vibrating Gold Microelectrode Y. Peled,†,‡ E. Krent,† N. Tal,† H. Tobias,‡ and D. Mandler*,† †

The Institute of Chemistry, the Hebrew University of Jerusalem, Jerusalem 9190401, Israel Chemistry Department, Nuclear Research Centre-Negev, Beer Sheva 84190, Israel



ABSTRACT: In this work we report the sensitive electroanalytical detection of uranium(VI) in aqueous solutions. Uranium commonly exists in aqueous solutions in the form of its oxo ion, uranyl (UVIO22+). The detection of uranyl has been accomplished by us through its deposition upon reduction by two electrons to the insoluble UO2 using a bare disk gold macroelectrode and anodic stripping voltammetry (ASV). This gave an unsatisfactory detection limit of ca. 1 × 10−5 M uranyl. Moreover, the evolution of hydrogen bubbles blocked the electrode surface as a result of water reduction at negative deposition potential (−0.7 V vs Ag/AgCl). The application of a 25 μm diameter Au microwire electrode on which UO2 precipitated at negative potential (−1.2 V) improved substantially the detection limit. Further improvement was accomplished by vibrating the microwire working electrode, which increased the amounts of UO2 deposition due to decreasing the diffusion layer. The effect of the vibrating amplitude and frequency on the electroanalytical response was studied and optimized. Eventually, a detection limit of ca. 1 × 10−9 M uranyl was achieved using a 5 min deposition time, −1.2 V deposition potential, and vibrating the electrode at frequency of 250 Hz and amplitude of 6 V. ranium and its compounds are used in a variety of fields, such as energy sources for nuclear power plants, military nuclear industry, and balancing weights of big civilian aircrafts (due to uranium high density of 19.2 g/cm3). All of these applications, their development, and research pose hazards to human health and are a source of environmental pollution. As such, it definitely calls for uranium levels monitoring. Trace amounts of uranium can be found in plants, animals, and even human tissues and urine (1−50 ppt is a normal human concentration range in urine). Due to its toxicity, monitoring uranium levels is of great importance to people exposed to the metal in their daily line of work.1 The radioactive and toxic uranium, both solid and dissolved, may harm body organs such as kidneys, bones, and the circulatory system. Hence, uranium concentration needs to be monitored in ecosystems, worldwide water facilities, and in the human bodyparticularly for nuclear industry workers. The most common form of dissolved uranium in aqueous solutions is the uranyl ion (UO22+), a thermodynamically stable form, with well-known crystallographic structure.2−5 The reactions of uranyl ions play an important role in extraction of uranium ore, processing of nuclear fuel, precipitate processes, and in uranium distribution in the environment. Due to quantum effect of the high atomic number uranium atom, the uranyl ion is characterized by great thermodynamic stability, and almost the entire coordination chemistry of uranyl ions relies on ligand-to-metal bonds leaving the UO bonds unreactive.6,7 Unlike most transition metal oxides, the uranyl ion has a linear structure and tends to create complexes with

U

© XXXX American Chemical Society

mostly six ligands, which adopt a planar hexagonal structure. Binding is through the nonbonding electron pairs of relatively “hard” Lewis bases, such as oxygen and nitrogen atoms. In recent years many efforts have been made to obtain efficient detection of uranium in aqueous solutions using a selection of chemical sensors and physical techniques. In some cases, as with ICPMS and polarography, low detection threshold of uranyl ions was reached.8−10 However, these methods are neither accessible nor simple for daily field work and tests; their setups are cumbersome, expensive, and as a result not portable. In this study, we aim at developing a new electrochemical protocol with high sensitivity toward uranyl ions. Among a variety of analytical techniques, electrochemical methods are very useful due to their operation simplicity, low equipment cost, selectivity, rapid analysis, and the ability to achieve extremely low detection limits. This meets the main requirements of developing a sensor for uranylsensitivity and speed. Over the past few decades, many studies have focused on the coordination chemistry of uranyl ion, both for extracting and determination of the toxic analyte in different environments. Trace levels of uranium can be determined and quantified, for example, by stripping voltammetry (with modified and nonmodified electrodes), spectrophotometry, fluorimetry, Received: October 4, 2014 Accepted: December 1, 2014

A

dx.doi.org/10.1021/ac503719r | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

electrochemical impedance spectroscopy (EIS), etc.11−15 A previous study conducted in our lab described the electrochemical detection of uranyl by an electrode modified with derivatives of calix[6]arene, also known as “super uranophile”.16 The work further investigated the electrochemical determination of UO22+ by a phosphate-based self-assembled monolayer.17 However, both preconcentration kinetics of the first step and the sensitivity are yet to be challenged. In recent years, there is a multiplicity of works with microelectrodes, both disc and cylindrical forms, which are simple to prepare, convenient to work with, and of promising preliminary results. Various metals such as As, Cu, Hg, Pb, and others have been determined using microwire electrodes by voltammetric methods.18 Microelectrodes have benefits such as high current densities, stirring-independent radial diffusion, enhanced mass transport, good signal-to-noise ratios (accordingly, high sensitivity), enabling high scan rates, apt for smallvolume electrochemical cells, and electrochemical setup simplicity.19−22 In this study, microelectrodes were prepared according to Chapman and van den Berg23 with some minor modifications. Our study is probably the first to report in situ uranyl sensing with vibrating microwire electrodes by voltammetry. We obtained low detection levels of ca. 10−9 M. A standard speaker (1 W, 8 Ω) was applied to which a Au microwire was attached. The vibrating electrode allows on-field, outdoors, and in principle also in vivo measurements. The vibrations decrease the diffusion layer and therefore increase the mass transport of ions to the electrode surface. Last not least, this combination of microwire electrode and vibrations almost completely prevents hydrogen bubbles accumulation on the electrode surface, which obstruct the negative potential preconcentration process. We found as well that vibrating the electrode stabilizes the measurement and is useful for the sensitive determination of UO22+.

(6 mm diameter) or platinum wire of 1 mm diameter (HollandMoran, 99.99%) and Ag/AgCl (3 M KCl, CH Instruments Inc., TX, U.S.A.) were used as counter and reference electrodes, respectively. The electrochemical cell was a standard borosilicate glass beaker of 30 mL. The primary experiments with the vibration microelectrode were performed with a few types of vibrators including (1) 3−4.5 V, 0.06 A micro-flat-button-type vibrator motor of cellphone, 10 mm diameter (China); (2) 3 V, 0.22 A microcell phone dc cordless vibrator motor, 4 mm diameter (China); (3) commercial tape and phone speakers. The experiments with the vibration microelectrode were performed, finally, with a standard speaker (1 W, 8 Ω) 5 cm in diameter, which was used as the vibrator and was connected to an Agilent 15 MHz function waveform generator (Agilent, 33120A) at a frequency of 250 Hz and 6 V. The microwire electrode was fixed to the vibrator using an iron connector tie (Figure 1). An optical microscope (Olympus BX60, Japan) was used to examine the microelectrode sealant and to determine its actual length.

EXPERIMENTAL SECTION Chemicals. Uranyl nitrate hexahydrate (Merck, 99%), potassium nitrate (Merck, 99% min), ethanol (J.T. Baker, absolute), and sulfuric acid (J.T. Baker, 96%) were purchased and used as received. Acetate buffer (50 mM, pH 3.0) was prepared using acetic acid (J.T. Baker, 99−100%) and anhydrous sodium acetate (J.T. Baker, 99+%). A 10 mM K3Fe(CN)6 in 0.5 M KCl aqueous solution (AR Gadot Israel) was used to determine the diffusion layer thickness of the vibrating microelectrodes. Alumina slurry (Al2O3, Buehler; IL, U.S.A., 0.05 and 1 μm) in deionized water and Microcloth (Buehler IL, U.S.A.) were used to polish the gold disk electrode. Gold wire of 25 μm diameter (Alfa Aesar, U.K.), standard electrical copper wire of 1 mm diameter, polyethylene pipet tip of 1−10 μL (Finntip, Thermo Scientific), silver epoxy kit (G3349, Agar Scientific, England), and Torr-Seal glue (Varian Inc., U.S.A.) were used to prepare the microwire electrodes. Deionized water (Barnstead Easypure UV system) was used for preparing the different solutions. Instruments. Electrochemical measurements were carried out with a CHI-630B potentiostat (CH Instruments Inc., TX, U.S.A.) using a conventional three-electrode cell. The working electrode was either a gold disk working electrode (CH Instruments Inc., TX, U.S.A.) of 2 mm diameter embedded in a Teflon cylinder or a homemade microelectrode that consisted of a polyethylene-sealed, gold microwire electrode of 25 μm diameter and a length of ca. 1 mm. A high-purity graphite rod

Procedures. Macroelectrodes Pretreatment. The gold macroelectrodes (2 mm diameter) were polished first with emery paper followed by alumina slurry (0.1 and 0.05 μm) on a Microcloth fabric. Then the electrodes were washed with ethanol, deionized water, and sonicated in an ultrasonic bath, first for 5 min in deionized water and then in 0.1 M H2SO4 for 10 min. Finally, it was electrocycled in 0.1 M H2SO4 between the oxidation and reduction of water until a reproducible CV was obtained. Preparation of Microwire Electrodes. The microwire electrodes were prepared similarly to the method reported by Nyholm and Wikmark.22 Briefly, a gold wire of 25 μm diameter and 0.5−1 cm long was connected to a copper wire of 1 mm diameter and ∼10 cm length using conductive glue (silver epoxy kit, G3349, Agar scientific, England) and then dried at 100 °C for 10 min. The wire was inserted in a disposable 1−10 μL polyethylene pipet tip (Finntip, Thermo Scientific) with only the gold end emerging from the narrow orifice of the pipet tip. The emerging gold wire was then sealed by melting the polypropylene orifice tip around the gold at >300 °C for 6−10 s using a micropipette puller (PP-830, Narishige group, Tokyo, Japan). The upper side of the pipet tip was sealed with TorrSeal glue (Varian Inc., U.S.A.). After drying, the accurate length (varying between 0.5 and 2 mm) of the gold microwire was determined by optical microscopy (Olympus BX60, Japan). Then the electrode was washed and electrocycled in 0.1 M H2SO4 at the same manner as the gold disk macroelectrode.

Figure 1. Photo of the vibrating microwire electrode.



B

dx.doi.org/10.1021/ac503719r | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

Determination of Uranyl in Tap Water. The determination of uranyl in tap water was performed using the standard addition calibration method. The zero addition consisted of 10 mL of tap water sample in acetate buffer 0.05 M and KNO3 0.1 M. The oxidation current of UO2 to UO22+ was measured using anodic stripping voltammetry (ASV), while the deposition potential was −1.2 V and deposition time of 2 min, and the vibrations at 250 Hz frequency and 6 V amplitude were operated only during the deposition time. The first, second, and third additions consisted of 200 μL of 1 × 10−6 M uranyl solution, to final concentrations of 1.96 × 10−8, 3.85 × 10−8, and 5.66 × 10−8 M, respectively.



RESULTS AND DISCUSSION The first step in developing an electroanalytical method is selecting the working electrode. In this work, we chose to work with a gold electrode, which showed superior performance and is utilized frequently for sensitive electrochemical measurements of minor concentrations of uranyl ions in aqueous solutions.16,17,21,24 In the past few years the application of gold electrodes has increased because of its special properties, such as inertness and resistance in different environmental conditions, its processability as a screen-printed electrode, relatively low-cost, and high sensitivity toward metallic and organic contaminants in water, including uranyl ions. Another important decision to take was how to choose the experimental conditions and in particular the electrolyte and pH. Previous studies, using gold electrodes, have shown that the electrolyte composition and pH have a significant effect on the redox mechanism of uranyl.25,26 According to these works, which were supported by our results as well,16 we conducted our measurements at pH > 2.5 to ensure that the limiting current of U6+ reduction was exclusively dictated by diffusion. We found that the best medium for the detection of low concentrations of uranyl in water was 0.05 M acetate buffer (pH 3−3.8) with 0.1 M KNO3 (supporting electrolyte) which served as the background solution as well. Electrochemical Experiments with a Bare Gold Macroelectrode. Electrochemistry of Uranyl. The treated gold macroelectrode was immersed in an electrochemical cell consisting of 50 mM acetate buffer (pH 3.0) and 0.1 M KNO3 as supporting electrolyte. Cyclic voltammetry (CV) was performed using the bare macroelectrode, as a working electrode, for initial electrochemical studies of the system and for acquiring qualitative information about the electrochemical reactions of uranyl. Figure 2A shows the CV of 0.1 mM UO22+ using a bare gold macroelectrode and a scan rate of 100 mV/s. It can be seen that a reduction wave appears at E = −0.162 V and an oxidation wave at −0.119 V versus Ag/AgCl, which are associated with the one-electron reduction and oxidation of UO22+ (eq 1).27 This gives a standard redox potential, E0′ = −0.141 V, which is in fair agreement with the literature value of 0.163 V versus NHE. Moreover, the peak potential difference (ΔEpk = 43 mV) is smaller than expected for one-electron transfer, which can be explained by the known disproportionation of the UO2+ into UO22+ and UO2.28 The second reduction wave, associated with the reduction of UO2+ to the insoluble species UO2 (eq 2), is masked by the larger reduction wave of water to hydrogen (eq 4) which commences at ca. −0.4 V. A second oxidation wave peaks at 0.115 V and is associated with the direct two-electron oxidation of UO2 to the species UO22+ (eq 3). However, upon reduction of the UO22+ species by two electrons, neutral UO2 is formed, which is water

Figure 2. (A) CV of the background (black) and 0.1 mM (red) of UO22+ in 50 mM acetate buffer and 0.1 M KNO3 (pH 3) recorded with a gold macroelectrode. (B) CV under the same conditions as in panel A but different negative potentials: −0.4 V (black), −0.5 V (red), −0.55 V (green), −0.60 V (blue), −0.65 V (cyan), −0.70 V (magenta).

insoluble and readily precipitates. This can be exploited to an efficient quantitative collection of dissolved uranium by electrochemical means and as an analytical tool to determine uranium in aqueous solutions. The first oxidation wave is not always observed, as some of the measurement techniques are insufficiently sensitive and due to the instability of U5+ species, which is oxidized or disproportionates shortly to the more stable U6+ species. The background was measured under exactly the same conditions but without the uranyl. UO2 2 +(aq) + e− ↔ UO2+(aq)

(1)

UO2+(aq) + e− + 4H+(aq) ↔ U 4 +(aq) + 2H 2O(l)

(2)

U 4 +(aq) + 2H 2O(l) → UO2 2 +(aq) + 2e− + 4H+(aq) (3) −



2H 2O(l) + 2e → H 2(g) + 2OH (aq)

(4)

CV measurements were performed also to study the effect of the negative potential on the oxidation wave (U4+ to U6+). Figure 2B shows the CV of 0.1 mM UO22+ using a gold macroelectrode scanned to different negative potentials. It is clearly seen that scanning to more negative potentials than −0.4 V caused an increase in the oxidation current wave (U4+ to U6+) at 0.115 V. The reason is that, when we scanned to more C

dx.doi.org/10.1021/ac503719r | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

found that the direct oxidation wave at the macroelectrode is undetectable below 0.01 mM UO22+. Electrochemical Experiments with a Gold Microelectrode. The experiments with the bare gold macroelectrode were interfered with hydrogen bubbles evolution during the deposition stage, which partially blocked the electrode surface. The insufficient sensitivity, due to limited diffusion currents, along with the cumbersome electrode pretreatment process were two additional disadvantages of our gold macroelectrode. In order to overcome these problems, we decided to use microwire electrodes as working electrodes. The major advantage of microelectrodes is their radial diffusion due to their cylindrical geometry that enhances mass transport of the species to the surface electrode. Enhanced rates of mass transport of electroactive species accrue from the radial (cylindrical) diffusion to the microelectrodes as well as the spherical diffusion to the edges. Such “edge effects” contribute significantly to the overall diffusion current. The rate of mass transport to and from the electrode (and hence the current density) increases as the electrode size decreases. As a consequence of the increase in mass-transport rates and the reduced charging currents, microelectrodes exhibit excellent signal-to-background characteristics in comparison to their larger counterparts.29 Effect of Deposition Potential and Deposition Time. Using ASV in 0.01 mM UO22+ solution consisting of 50 mM acetate buffer (pH 3.0) and 0.1 M KNO3 as supporting electrolyte, the microelectrode was first tested to lower the deposition potential at