Rapid, low level determination of silver(I) in drinking water by colorimetric-solid-phase extraction

A rapid, highly sensitive two-step procedure for the trace analysis of silver(I) is described. The method is based on: (1) the solid-phase extraction (SPE) of silver(I) from a water sample onto a disk impregnated with a silver-selective colorimetric reagent, and (2) the determination of the amount of complexed analyte extracted by the disk by diffuse reflectance spectroscopy (DRS). This method, called colorimetric-solid-phase extraction (C-SPE), was recently shown effective in determining low concentrations (0.1-5.0 mg/ml) of iodine and iodide in drinking water. This report extends C-SPE to the trace (∼4 μg/l) level monitoring of silver(I) which is a biocide used on the International Space Station (ISS). The determination relies on the manually driven passage of a water sample through a polystyrene-divinylbenzene disk that has been impregnated with the colorimetric reagent 5-(p-dimethylaminobenzylidene) rhodanine (DMABR) and with an additive such as a semi-volatile alcohol (1,2-decanediol) or nonionic surfactant (Brij 30). The amount of concentrated silver(I) is then determined in a few seconds by using a hand-held diffuse reflectance spectrometer, with a total sample workup and readout time of ∼60 s. Importantly, the additive induces the uptake of water by the disk, which creates a local environment conducive to silver(I) complexation at an extremely high concentration factor (∼800). There is no detectable reaction between silver(I) and impregnated DMABR in the absence of the additive. This strategy represents an intriguing new dimension for C-SPE in which additives, directly loaded in the disk material, provide a means to manipulate the reactivity of the impregnated reagent.     

Rapid determination of ions by combined solid-phase extraction--diffuse reflectance spectroscopy

We introduce colorimetric solid-phase extraction (C-SPE) for the rapid determination of selected ions. This new technique links the exhaustive concentration of an analyte by SPE onto a membrane disk surface for quantitative measurement with a hand-held diffuse reflectance spectrometer. The concentration/measurement procedure is complete in approximately 1 min and can be performed almost anywhere. This method has been used to monitor iodine and iodide in spacecraft water in the 0.1-5.0 ppm range and silver(I) in the range of 5.0-1000 microg/l. Applications to the trace analysis of copper(II), nickel(II), iron(III) and chromium(VI) are described. Studies on the mechanism of extraction showed that impregnation of the disk with a surfactant as well as a complexing reagent results in uptake of additional water, which markedly improves the extraction efficiency.     

A rapid, simple method for determining formaldehyde in drinking water using colorimetric-solid phase extraction

Formaldehyde has been detected in drinking water supplies across the globe and on board NASA spacecraft. A rapid, simple, microgravity-compatible technique for measuring this contaminant in water supplies using colorimetric-solid phase extraction (C-SPE) is described. This method involves collecting a water sample into a syringe by passage through a cartridge that contains sodium hydroxide, to adjust pH, and Purpald, which is a well-established colorimetric reagent for aldehydes. After completing the reaction in the syringe by agitating for 2 min on a shaker at 400 rpm, the 1.0-mL alkaline sample is passed through an extraction disk that retains the purple product. The amount of concentrated product is then measured on-disk using diffuse reflectance spectroscopy, and compared to a calibration plot generated from Kubelka-Munk transformations of the reflectance data at 700 nm to determine the formaldehyde concentration. This method is capable of determining formaldehyde concentrations from 0.08 to 20 ppm with a total work-up time of less than 3 min using only 1-mL samples.     

Determination of colloidal and dissolved silver in water samples using colorimetric solid-phase extraction

The increase in bacterial resistance to antibiotics has led to resurgence in the use of silver as a biocidal agent in applications ranging from washing machine additives to the drinking water treatment system on the International Space Station (ISS). However, growing concerns about the possible toxicity of colloidal silver to bacteria, aquatic organisms and humans have led to recently issued regulations by the US EPA and FDA regarding the usage of silver. As part of an ongoing project, we have developed a rapid, simple method for determining total silver, both ionic (silver(I)) and colloidal, in 0.1-1 mg/L aqueous samples, which spans the ISS potable water target of 0.3-0.5 mg/L (total silver) and meets the US EPA limit of 0.1 mg/L in drinking water. The method is based on colorimetric solid-phase extraction (C-SPE) and involves the extraction of silver(I) from water samples by passage through a solid-phase membrane impregnated with the colorimetric reagent DMABR (5-[4-(dimethylamino)benzylidene]rhodanine). Silver(I) exhaustively reacts with impregnated DMABR to form a colored compound, which is quantified using a handheld diffuse reflectance spectrophotometer. Total silver is determined by first passing the sample through a cartridge containing Oxone, which exhaustively oxidizes colloidal silver to dissolved silver(I). The method, which takes less than 2 min to complete and requires only ∼1 mL of sample, has been validated through a series of tests, including a comparison with the ICP-MS analysis of a water sample from ISS that contained both silver(I) and colloidal silver. Potential earth-bound applications are also briefly discussed.     

Silica gel-polyethylene glycol as a new adsorbent for solid phase extraction of cobalt and nickel and determination by flame atomic absorption spectrometry

In this paper a novel solid phase extraction method to determine Co(II) and Ni(II) using silica gel-polyethylene glycol (Silica-PEG) as a new adsorbent is described. The method is based on the adsorption of cobalt and nickel ions in alkaline media on polyethylene glycol-silica gel in a mini-column, elution with nitric acid and determination by flame atomic absorption spectrometry. The adsorption conditions such as NaOH concentration, sample volume and amount of adsorbent were optimized in order to achieve highest sensitivity. The calibration graph was linear in the range of 0.5-200.0 ng mL⁻¹ for Co(II) and 2.0-100.0 ng mL⁻¹ for Ni(II) in the initial solution. The limit of detection based on 3S_b was 0.37 ng mL⁻¹ for Co(II) and 0.71 ng mL⁻¹ for Ni(II). The relative standard deviations (R.S.D.) for ten replicate measurements of 40 ng mL⁻¹ of Co(II), and Ni(II) were 3.24 and 3.13%, respectively. The method was applied to determine Co(II) and Ni(II) in black tea, rice flour, sesame seeds, tap water and river water samples.     

Rapid determination of aluminum by UV-vis diffuse reflectance spectroscopy with application of suitable adsorbents

Diffuse reflectance spectroscopy (DRS) can be used as a rapid and sensitive method for the quantitative determination of low amounts of aluminum. In this analytical technique, the analyte in samples are extracted onto a solid sorbent matrix loaded with a colorimetric reagent and then quantified directly on the adsorbent surface. Alternatively, colored aluminum complexes formed in solution can also be immobilized onto adsorbent surface and be measured by DRS technique. Octadecyl silica disk, methyltrioctylammonium chloride–naphthalene and MCM-41 were examined as adsorbents. Eriochrome cyanine R and quinalizarin were used as coloring reagents. Optimal sorption conditions were found for each system of analyte–reagent–adsorbent. The concentration of analyte is determined using the appropriate form of the Kubelka–Munk function. We obtained for each of the aluminium–reagent–adsorbent system a calibration curve by plotting the absorbance versus the log 10²[Al³⁺] μg ml⁻¹. The linear dynamic range extends over two orders of magnitude within 0.01–15 μg ml⁻¹ with little differences in the range and in the correlation coefficients among the adsorbents. We consider that for a rapid determination of aluminum a spot-test-DRS combination with a detection limit of 1.0 × 10⁻² μg ml⁻¹ is the more facile and preferred technique.     

Interfacing in-line gas-diffusion separation with optrode sorptive preconcentration exploiting multisyringe flow injection analysis

An automatic multisyringe flow injection analysis (MSFIA) system coupling a flow-through optical fiber diffuse reflectance sensor with in-line gas-diffusion (GD) separation is proposed for the isolation, preconcentration and determination of traces of volatile and gas-evolving compounds in samples containing suspended solids, with no need for any preliminary batch sample treatment. The flowing methodology overcomes the lost of sensitivity of the in-line separation technique, when performed in a uni-directional continuous-flow mode, through the implementation of disk-based solid-phase extraction schemes. The high selectivity and sensitivity, the low reagent consumption and the miniaturization of the whole assembly are the outstanding features of the automated set-up. The proposed combination of techniques for separation, flow analysis, preconcentration and detection was applied satisfactorily to sulfide determination in environmental complex matrixes. The method based on multicommutation flow analysis involves the stripping of the analyte as hydrogen sulfide from the donor channel of the GD-module into an alkaline receiver segment, whereupon the enriched plug merges with well-defined zones of the chormogenic reagents (viz., N,N-dimethyl-p-phenylenediamine (DMPD) and Fe(III)). The in-line generated methylene blue dye is subsequently delivered downstream to the dedicated optrode cell furnished with a C₁₈ disk, while recording continuously the diffuse reflectance spectrum of the pre-concentrated compound. This procedure provides a linear working range of 20-500 μg l⁻¹ sulfide with a relative standard deviation of 2.2% (n = 10) at the 200 μg l⁻¹ level, and a detection limit of 1.3 μg l⁻¹.     

Column system using diaion HP-2MG for determination of some metal ions by flame atomic absorption spectrometry

A column solid-phase extraction method for the preconcentration and determination of cadmium(II), copper(II), cobalt(II), iron(III), lead(II), nickel(II) and zinc(II) dithizone chelates by atomic absorption spectrometry has been described. Diaion HP-2MG was used as adsorbent for column studies. The influences of the various analytical parameters including pH of the aqueous solutions, amounts of ligand and resin were investigated for the retentions of the analyte ions. The recovery values are ranged from 95 to 102%. The influences of alkaline and earth alkaline ions were also discussed. The preconcentration factor was 375, when the sample volume and final volume are 750 and 2 ml, respectively. The detection limits of the analyte ions (k=3, N=21) were varying 0.08 μg/l for cadmium to 0.25 μg/l for lead. The relative standard deviations of the determinations at the concentration range of 1.8×10⁻⁴ to 4.5×10⁻⁵ mmol for the investigated elements were found to be lower than 9%. The proposed solid-phase extraction procedure were applied to the flame atomic absorption spectrometric determinations of analyte ions in natural waters (sea, tap, river), microwave digested samples (milk, red wine and rice) and two different reference standard materials (SRM1515 apple leaves and NRCC-SLRS-4 riverine water).     

A novel antibody immobilization and its application in immunoaffinity chromatography

A novel antibody immobilization and its application in immunoaffinity chromatography (IAC) were presented. Using acrylamide (AM) as monomer, ethylene glycoldimethacrylate (EGDMA) as cross-linker and bulk polymerization as synthetic method, we prepared a polymer in which the Cu(II) was embedded. The Cu(II)-embedded polymer was tested for its binding with protein. It was found that Cu(II)-embedded polymer displayed a strong binding with bovine serum albumin (BSA). At 80% of methanol, no BSA was released from Cu(II)-embedded polymer. The Cu(II)-embedded polymer was then used as a novel solid support for antibody immobilization. IAC column was prepared by immobilizing polyclonal antibody (pAb) against clenbuterol (CL) on Cu(II)-embedded polymer and packing the Cu(II)-embedded polymer-pAb into a common solid phase extraction (SPE) cartridge. Under optimal extraction conditions, the IAC column was characterized in terms of maximum binding capacity for target analyte, extraction efficiency and reusability. It was revealed that, for IAC column packed with 0.1 g of solid support immobilized with antibody, the maximum capacity for CL was 616 ng; the extraction recoveries of the column for CL from three spiked food samples were 84.4-95.2% with relative standard deviation (RSD) of 9.3-15.5%; after more than 30 times repeated usage, there was not significant loss of specific recognition. The results demonstrated the feasibility of the prepared IAC column for CL extraction. The proposed antibody immobilization method exhibiting the properties of simplicity, low cost, strong binding for target analyte, no leaching of antibody, etc., would be a very useful tool applied in the field of IAC.     

Rapid detection of nitroaromatic and nitramine explosives on chromatographic paper and their reflectometric sensing on PVC tablets

Rapid and inexpensive sensing of explosive traces in soil and post-blast debris for environmental and criminological purposes with optical sensors has recently gained importance. The developed sensing method for nitro-aromatic and nitramine-based explosives is based on dropping an acetone solution of the analyte to an adsorbent surface, letting the solvent to dry, spraying an analytical reagent to produce a persistent spot, and indirectly measuring its reflectance by means of a miniature spectrometer. This method proved to be useful for on-site determination of nitro-aromatics (trinitrotoluene (TNT), 2,4,6-trinitrophenylmethylnitramine (tetryl) and dinitrotoluene (DNT)) and nitramines (1,3,5-trinitro-1,3,5-triazacyclohexane (RDX) and octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX)) pre-adsorbed on a poly vinyl chloride (PVC) surface, with the use of different spray reagents for each group of explosives producing different colors. The calibration equations of the tested compounds as reflectance vs. concentration showed excellent linearity (correlation coefficient: 0.998-0.999). The linear quantification interval in terms of absolute quantity of analyte was 0.1-0.5 μg. The developed method was successfully tested for the analysis of military explosives Comp B and Octol, and was validated against high performance liquid chromatography (HPLC). The reflectometric sensing method could also be used for qualitative identification of the nitrated explosives on a chromatographic paper. The reagent-impregnated paper could also serve as sensor, enabling semi-quantitative determinations of TNT and tetryl.     

Extraction of cobalt (II) and nickel (II) by a solvent impregnated resin containing bis(2,4,4-trimethylpentyl)monothiophosphinic acid

In the present work, studies have been conducted on the equilibrium distribution of cobalt (II) and nickel (II) between aqueous hydrochloric solution and macromolecular resin impregnated with bis(2,4,4-trimethylpentyl)monothiophosphinic acid (Cyanex302, HL). Effects of extraction time, pH values, metal ion concentration, and temperature were investigated. Analysis of the results shows that the extraction of the two metal ions can be explained assuming the formation of metal complexes in the resin phase with a general composition ML ₂. An extraction reaction is proposed and the equilibrium constants of the complexes were determined. The Freundlich isotherm and thermodynamic quantities, i.e., ΔG, ΔH and ΔS were also obtained. Both of the extraction reactions of cobalt (II) and nickel (II) are endothermic ones. The efficiency of the resin in the separation of cobalt (II) and nickel (II) is provided according to the separation factors. Under the experimental conditions employed, pH₅₀ values for cobalt (II) and nickel (II) are 3.76, 5.01, respectively. The logarithmic value of separation factor was calculated as 2.50.     

A novel multi-element coprecipitation technique for separation and enrichment of metal ions in environmental samples

A multi-element preconcentration-separation technique for heavy metal ions in environmental samples has been established. The procedure is based on coprecipitation of gold(III), bismuth(III), cobalt(II), chromium(III), iron(III), manganese(II), nickel(II), lead(II), thorium(IV) and uranium(VI) ions by the aid of Cu(II)-9-phenyl-3-fluorone precipitate. The Cu(II)-9-phenyl-3-fluorone precipitate was dissolved by the addition 1.0 mL of concentrated HNO₃ and then the solution was completed to 5 mL with distilled water. Iron, lead, cobalt, chromium, manganese and nickel levels in the final solution were determined by flame atomic absorption spectrometer, while gold, bismuth, uranium and thorium were determined by inductively coupled plasma mass spectrometer. The optimal conditions are pH 7, amounts of 9-phenyl-3-fluorone: 5 mg and amounts of Cu(II): 1 mg. The effects of concomitant ions as matrix were also examined. The preconcentration factor was 30. Gold(III), bismuth(III), chromium(III), iron(III), lead(II) and thorium(IV) were quantitatively recovered from the real samples. The detection limits for the analyte elements based on 3 sigma (n = 15) were in the range of 0.05-12.9 μg L⁻¹. The validation of the presented procedure was checked by the analysis of two certified reference materials (Montana I Soil (NIST-SRM 2710) and Lake Sediment (IAEA-SL-1)). The procedure was successfully applied to some environmental samples including water and sediments.     

Simultaneous extraction of trace amounts of cobalt, nickel and copper ions using magnetic iron oxide nanoparticles without chelating agent

The present study investigates the application of Fe₃O₄ nanoparticles as an adsorbent for solid phase extraction and their subsequent determination of trace amounts of cobalt, nickel and copper from environmental water samples using flame atomic absorption spectrometry. The analyte ions were adsorbed on magnetic nanoparticles in the pH range of 10–12 and then, Fe₃O₄ nanoparticles were easily separated from the aqueous solution by applying an external magnetic field and decantation. Hence, no filtration or centrifugation was needed. After extraction and collection of magnetic nanoparticles, the analyte ions were desorbed using 1.0 M HNO₃. Several factors that may affect the preconcentration and extraction process, including pH, type and volume of eluent, sample volume, salt effect and matrix effect were optimized. Under the optimized conditions, linearity was maintained from 0.005–3.0 μg/mL for cobalt and nickel and 0.001 to 1.25 μg/mL for copper in the initial solution. The detection limits of this method for cobalt, nickel and copper ions were 0.9, 0.7 and 0.3 ng/mL, respectively. Finally, the method was successfully applied to the extraction and determination of the analyte ions in natural waters and reference plant samples.     

Optical fibre sensor for hydrogen sulphide monitoring in mouth air

A reversible optical fibre chemical sensor for hydrogen sulphide monitoring in mouth air based on reflectance measurements has been developed. The active sensing phase has been prepared by immobilising the colorimetric reagent 2,6-dichlorophenolindophenol (DCPI) in a silica gel support. The principle of the determination is based on the increase of reflectance of such solid sensing phase when hydrogen sulphide reduces the colorimetric reagent with the subsequent decolouration process. The addition of 1.26 μg of Cu(II) per gram of solid support improved the response time and reversibility of the sensing phase.The detection limit is 10 ppb (v/v) of hydrogen sulphide. The linear range using the Kubelka-Munk function extends at least up to 1000 ppb (v/v). The sensor exhibits a response time of less than 2 min for hydrogen sulphide concentrations in the linear range and the signal is reversible.The optical sensor has been successfully tested for human malodour monitoring and the results validated by comparison with those obtained for the same individuals using a commercially available electrochemical instrument.     

Investigation of the iodine–poly(vinylpyrrolidone) interaction employed in the determination of biocidal iodine by colorimetric solid-phase extraction

Colorimetric solid-phase extraction (C-SPE) has been previously explored as a means to monitor the iodine-based disinfectant used in the water systems on board the space shuttle. This same disinfectant is baselined for eventual deployment in the US water recovery system planned for node 3 of the International Space Station (ISS). With C-SPE, the I₂ concentration is determined from the diffuse reflectance spectrum (DRS) of the yellow iodine–poly(vinylpyrrolidone) (PVP) complex using the Kubelka–Munk function. However, the solution chemistry of iodine is very complex and results in a variety of inorganic species (e.g., I⁻, I₂, I₃⁻, HOI) that have very different biocidal capabilities. Thus, the nature of the interaction of iodine with PVP, and more specifically, the identity of the iodine species involved in the interaction, requires more elucidation. This paper reports the findings from a series of detailed experiments conducted to elicit a more complete understanding of the iodine–PVP system employed in C-SPE. The results indicate that I₂, one of the two dominant biocidal forms of iodine, is the species responsible for the analytical signal in our C-SPE platform. These findings lay the ground work for the planned development of a multiplexed iodine determination and speciation platform for in-flight analysis of spacecraft water samples.     

Vortex-assisted magnetic solid phase extraction of Cd(II), Cu(II) and Pb(II) on the Nitroso–R salt impregnated magnetic Ambersorb 563 for their separation,preconcentration and determination by FAAS

A novel vortex-assisted magnetic solid phase extraction method followed by flame atomic absorption spectrometry was improved to separate, preconcentrate and determine the lead, copper and cadmium ions by using 1-Nitro-2-naphthol-3, 6-disulfonic acid disodium salt (Nitroso-R salt) impregnated magnetic Ambersorb-563 resin. The adsorbent was characterised by Fourier transform infrared spectra, BET surface analyser and scanning electron microscopy. The influences of various analytical parameters, such as pH value, type and volume of the eluent, sample volume, were optimised and the effects of potentially interfering ions were investigated. The analyte ions were quantitatively recovered at pH 7.0 on magnetic adsorbent and desorbed with a 2.0 M HNO₃ in 10% acetone as eluent. The detection limits were 1.4, 5.8 and 1.5 ng mL^?1 for Cd(II), Cu(II) and Pb(II), respectively. The method has been validated with analytically by the analysis certified reference materials and standard additions prior to application to determine metal ions in water samples.     

Spectrophotometric flow-injection determination of copper and nickel in plant digests exploiting differential kinetic analysis and multi-site detection

A novel strategy for implementing differential reaction-rate methods in flow-injection analysis is proposed and applied to the determination of copper and nickel in plant digests using 2-(5-brom-2-pyridylazo)-5-(diethylamino)-phenol (Br-PADAP) as the color-forming reagent. Multi-site detection is involved, therefore the flow cell is displaced between two monitoring sites, and the analytical signals refer to different conditions of sample dispersion, reaction development and timing.The system handles 20 samples h⁻¹ and requires 0.32 mg Br-PADAP per determination. Signal additivity was evaluated within 98 and 102%, and linear responses (r > 0.999; n = 6) were verified for both copper and nickel up to 0.80 mg l⁻¹. Detection limits of 0.01 and 0.04 mg l⁻¹ Cu and Ni were estimated by considering the highest concentration of the counter analyte. Results are precise (R.S.D. < 2%) and in agreement with ICP-OES (95% confidence level). Potentialities and limitations of the approach are discussed.     

A simple microextraction and preconcentration approach based on a mixed matrix membrane

A simple adsorption/desorption procedure using a mixed matrix membrane (MMM) as extraction medium is demonstrated as a new miniaturized sample pretreatment and preconcentration technique. Reversed-phase particles namely polymeric bonded octadecyl (C₁₈) was incorporated through dispersion in a cellulose triacetate (CTA) polymer matrix to form a C₁₈-MMM. Non-steroidal anti-inflammatory drugs (NSAIDs) namely diclofenac, mefenamic acid and ibuprofen present in the environmental water samples were selected as targeted model analytes. The extraction setup is simple by dipping a small piece of C₁₈-MMM (7 mm × 7 mm) in a stirred 10 mL sample solution for analyte adsorption process. The entrapped analyte within the membrane was then desorbed into 100 μL of methanol by ultrasonication prior to high performance liquid chromatography (HPLC) analysis. Each membrane was discarded after single use to avoid any analyte carry-over effect. Several important parameters, such as effect of sample pH, salting-out effect, sample volume, extraction time, desorption solvent and desorption time were comprehensively optimized. The C₁₈-MMM demonstrated high affinity for NSAIDs spiked in tap and river water with relative recoveries ranging from 92 to 100% and good reproducibility with relative standard deviations between 1.1 and 5.5% (n = 9). The overall results obtained were found comparable against conventional solid phase extraction (SPE) using cartridge packed with identical C₁₈ adsorbent.     

Determination of triazine herbicides using membrane-protected carbon nanotubes solid phase membrane tip extraction prior to micro-liquid chromatography

A novel microextraction technique termed solid phase membrane tip extraction (SPMTE) was developed. Selected triazine herbicides were employed as model compounds to evaluate the extraction performance and multiwall carbon nanotubes (MWCNTs) were used as the adsorbent enclosed in SPMTE device. The SPMTE procedure was performed in semi-automated dynamic mode and several important extraction parameters were comprehensively optimized. Under the optimum extraction conditions, the method showed good linearity in the range of 1–100 μg/L, acceptable reproducibility (RSD 6–8%, n = 5), low limits of detection (0.2–0.5 μg/L), and satisfactory relative recoveries (95–101%). The SPMTE device could be regenerated and reused up to 15 analyses with no analyte carry-over effects observed. Comparison was made with commercially available solid phase extraction-molecular imprinted polymer cartridge (SPE-MIP) for triazine herbicides as the reference method. The new developed method showed comparable or even better results against reference method and is a simple, feasible, and cost effective microextraction technique.     

An integrated reaction-retention and spectrophotometric detection flow system for the determination of nickel

An integrated flow-through photometric sensor for the determination of nickel in real samples of various origins has been developed. The sensor is based on the reaction of Ni(II) with 1-(2-pyridylazo)-2-naphthol (PAN) immobilized on a cationic resin which was placed in a flow-cell using a spectrophotometer tuned at 566 nm as detector. The Ni(II) ion from the sample injected into the carrrier stream (pH = 5.0) of a monochannel continuous flow system reacts with the immobilized chromogenic reagent to form a red chelate which remains on the active solid support and generates the analytical signal. When this reached its maximum value the Ni(II)-PAN chelate was destroyed using 1 M H₂SO₄ as eluents, leaving the sorbed PAN untouched. The response of the sensor was linear in the three concentration ranges assayed: 0.3–4.0, 0.1–1.6 and 0.05–0.8 μg mL^–1 for sample volumes of 100, 400 and 800 μL, respectively, and the R.S.D.(%) (n = 10) were 1.80(100 μL), 3.04(400 μL) and 2.29(800 μL). The sensor showed an excellent selectivity which could also be increased with a simple on-line modification to avoid interference from copper. It was applied to a variety of real samples with very good results in all cases. Received: 15 April 1998 / Revised: 29 June 1998 / Accepted: 3 July 1998