Using a photochemical method and chitosan to prepare surface-enhanced Raman scattering–active silver nanoparticles

In this paper, we report a new strategy for the preparation of surface-enhanced Raman scattering (SERS)-active silver nanoparticles (Ag NPs), using a photochemical method and the presence of chitosan (Ch). First, Ag substrates were subjected to electrochemical oxidation/reduction cycles (ORCs) in deoxygenated aqueous solutions containing 0.1 M HNO₃ and 1 g L⁻¹ Ch (pH 6.9, adjusted by adding 1 M NaOH), resulting in Ag⁺–Ch complexes. These substrates were then irradiated with UV light at various wavelengths to yield the SERS-active Ag NPs. A stronger SERS effect was observed on the SERS-active Ag NPs prepared by using UV irradiation at 310 nm. The pH of the solution and the presence of Ch during the preparation process both affected the resulting SERS activities.     

Focused-ion-beam-fabricated Au nanorods coupled with Ag nanoparticles used as surface-enhanced Raman scattering-active substrate for analyzing trace melamine constituents in solution

A well-ordered Au-nanorod array with a controlled tip ring diameter (Au_NRs_d) was fabricated using the focused ion beam method. Au_NRs_d was then coupled with Ag nanoparticles (Ag NPs) to bridge the gaps among Au nanorods. The effect of surface-enhanced Raman scattering (SERS) on Au_NRs_d and Ag NPs/Au_NRs_d was particularly verified using crystal violet (CV) as the molecular probe. Raman intensity obtained from a characteristic peak of CV on Au_NRs_d was estimated by an enhancement factor of ≈10⁷ in magnitude, which increased ≈10¹² in magnitude for that on Ag NPs/Au_NRs_d. A highly SERS-active Ag NPs/Au_NRs_d was furthermore applied for the detection of melamine (MEL) at very low concentrations. Raman-active peaks of MEL (10⁻³ to 10⁻¹² M) in water or milk solution upon Au_NRs_d or Ag NPs/Au_NRs_d were well distinguished. The peaks at 680 and 702 cm⁻¹ for MEL molecules were found suitable to be used as the index for sensing low-concentration MEL in a varied solution, while that at 1051 cm⁻¹ was practical to interpret MEL molecules in water or milk solution bonded with Au (i.e., Au_NRs_d) or Ag (i.e., Ag NPs/Au_NRs_d) surface. At the interface of Ag NPs/Au_NRs_d and MEL molecules in milk solution, a laser-induced electromagnetic field or hotspot effect was produced and competent to sense low-concentration MEL molecules interacting with Ag and Au surfaces. Accordingly, Ag NPs/Au_NRs_d is very promising to be used as a fast and sensitive tool for screening MEL in complex matrices such as adulteration in e.g., food and pharmaceutical products.     

Solid-state pH nanoelectrode based on polyaniline thin film electrodeposited onto ion-beam etched carbon fiber

An ultramicro pH sensor has been constructed based on a thin polyaniline film that was electrochemically deposited onto a carbon fiber nanometer-size electrode. The substrate nanoelectrodes were fabricated using ion-beam conically etched carbon fibers with tip diameters ranging ca. from 100 to 500 nm. The polyaniline film was deposited from HCl solution containing the aniline monomer by cycling the potential between −0.2 and +1.0 V. The electromotive force (emf) signal between the pH sensitive polyaniline-coated nanoelectrode and an Ag/AgCl reference electrode was linear over the pH range of 2.0-12.5 with a slightly super-Nernstian slope of ca. −60 mV/pH unit. Response times ranged from several sec at pHs around 7 up to 2 min at pH 12.5. The proposed pH nanoelectrode displayed high ion selectivity with respect to K⁺, Na⁺, Ca²⁺, and Li⁺, with log K_H,M values around −12 and has a working lifetime of about 20 days. Key parameters important for the pH nanoelectrode performance, including polyaniline film preparation, selectivity, response time, temperature dependence, relative coating thickness, stability, and reproducibility, have been characterized and optimized. The performance of the pH nanoelectrode was examined by measuring the pH of several real samples including body fluids (serum, urine) and low ionic strength water samples (rain, deionized and tap water). The results agreed very well with those obtained by using commercial glass pH electrodes. The proposed pH nanoelectrode demonstrated attractive properties and seems particularly promising for use under physiological conditions.     

Metal–organic frameworks of zeolitic imidazolate framework-7 and zeolitic imidazolate framework-60 for fast mercury and methylmercury speciation analysis

A fluorescence sensing platform based on metal–organic frameworks (MOFs) nanoparticles (NPs) of both zeolitic imidazolate framework-7 (ZIF-7) and zeolitic imidazolate framework-60 (ZIF-60) was developed for speciation analysis of inorganic Hg [Hg(II)] and methylmercury (MeHg⁺). Microwave-ultrasound assisted synthesis was employed for the preparation of ZIF-7 and ZIF-60 NPs, with short reaction time, easy procedure, and small particle size obtained. Based on strict cavity confinement of the ZIF-7 and ZIF-60 structures, the proposed method exhibited excellent selectivity for both Hg(II) and MeHg⁺, even in the presence of the other Hg species or various cations or anions with the concentration of 50 times high. Effect of pH and ionic strength on sensing behaviour of the ZIF MOF was studied as well. The calculated detection limit is 3 ng mL⁻¹ and 6 ng mL⁻¹ for Hg(II) and MeHg⁺, respectively. Furthermore, the application of the developed method to the analysis of local drinking water was demonstrated to be feasible, and the obtained recovery was 102% and 96.2% for Hg(II) and MeHg⁺, respectively.     

Study of the interaction of Azur B with DNA and the determination of DNA based on resonance light scattering measurements

Based on the measurements of molecular absorption and resonance light scattering (RLS), the aggregation of Azur B (AB) was in a medium of pH ranging from 1.98 to 2.56 and ionic strength <0.12 M. The presence of double stranded DNA prompts the aggregation, resulting in enhanced RLS signals. Linear relationships were achieved between the enhanced RLS intensity at 359.7 nm and DNA concentration in the range of 0-4.5 μg ml⁻¹ for both calf thymus DNA (ctDNA) and fish sperm DNA (fsDNA) if 3.0×10⁻⁵ M AB was employed. The 3σ limits of detection were 9.3 and 8.9 ng ml⁻¹ for ctDNA and fsDNA, respectively. Five synthetic samples were analysed satisfactorily.     

The defluoridation of drinking water using multi-walled carbon nanotubes

In this study, the potential sorption capacity of multi-walled carbon nanotubes (MWCNTs) was investigated as a means of removing fluoride from the drinking water of a number of regions in Iran and from experimental solutions. The test was conducted in both batch and continuous operation modes. Batch mode experiments were used to study the effect of parameters such as pH, contact time, ionic strength, adsorbent dose, adsorbent capacity, and the presence of foreign anions on the efficiency of fluoride removal. The results showed that the highest level of sorption occurs at pH 5 (about 94% at 18 min). The ionic strength of the solutions and the presence of co-anions such as chloride, nitrate, sulfate, hydrogen carbonate, perchlorate had a negligible effect on the sorption of F⁻ onto MWCNTs. Sorption capacity measurements revealed that MWCNTs have a saturation capacity of 3.5 mg of F⁻ per gram. Sorption data were best fitted with the Fruendlich sorption isotherm equation, which indicates that F⁻ tends to be adsorbed on MWCNTs in a multilayered manner. Experiments using Kohbanan city drinking water, which contains the highest level of F⁻ among the drinking water samples studied, showed that MWCNTs can remove over 85% of fluoride content.     

A rapid quantitative method for humic substances determination in natural waters

A simple and rapid method was proposed for humic substances (HS) determination at microgram levels in natural waters. This assay method is based on the binding of a dye, Toluidine Blue (TB), to HS molecules to produce a dye-HS complex, which causes a decrease in absorbance at 630 nm. This method was calibrated with HS samples with up to a concentration of 40 mg L⁻¹, which covered the range of dissolved HS concentrations present in natural waters. The detection limit was 0.8 mg L⁻¹ of HS, and the relative standard deviation of 10 replicate measurements for a 20-mg L⁻¹ standard sample was 3.5%. From the Langmuir adsorption isotherm theory, the binding equilibrium constant and total number of binding sites at neutral pH were calculated to be (8.17 ± 0.42) × 10⁵ L mol⁻¹ and N of 1.45 ± 0.04 mmol g⁻¹ HS, respectively. The determination results with five water samples from lake, river and pond were consistent with those measured with the reference methods, demonstrating that this quantification method for HS determination was rapid, sensitive and feasible.     

Carbon paste electrode modified with silver thimerosal for the potentiometric flow injection analysis of silver(I)

The utility of carbon paste electrode modified with silver ethylmercurythiosalicylate (silver thimerosal) in both static mode and flow injection analysis (FIA) is demonstrated. The electrode was fully characterized in terms of composition, response time, thermal stability, usable pH and ionic strength ranges. It has been shown that diisononyl phthalate (DINP) acts as more suitable solvent mediator for preparation of the electrode, which exhibits linear response range to Ag(I) extending from 5.0 × 10⁻⁷ to 1.0 × 10⁻³ M with detection limit of 2.5 × 10⁻⁷ M and Nernstian slope of 59.3 ± 1.0 mV/decade. The proposed chemically modified carbon paste electrode shows a very good selectivity for Ag(I) over a wide variety of metal ions and successfully used for the determination of the silver content of silver sulphadiazine (burning cream) and developed radiological films. The electrode was also used as an indicator electrode in the potentiometric titration of thiopental and thimerosal with AgNO₃.     

Evaluation of a cloud point extraction approach for the preconcentration and quantification of trace CuO nanoparticles in environmental waters

The cloud point extraction (CPE) of commercial copper(II) oxide nanoparticles (CuO NPs, mean diameter of 28 nm) in water samples was fully investigated. Factors such as Triton X-114 (TX-114) concentration, pH, incubation temperature and time, were optimized. The effects of CuO NP behavior like agglomeration, dissolution, and surface adsorption of natural organic matter, Cu²⁺, and coating chemicals, on its recovery were studied. The results indicated that all the CPE factors had significant effects on the extraction efficiency. An enrichment factor of ∼89 was obtained under optimum CPE conditions. The hydrodynamic diameter of CuO NPs increased to 4–5 μm upon agglomeration of NP-micelle assemblies, and decreased at pH >10.0 at which the extraction efficiency was also lowered. The solubility and therefore, the loss of NPs were greatly enhanced at pH <8.5 and in the first 60 min of incubation, whereas it declined at elevated incubation temperatures. Our results showed that the dissolved organic carbon (DOC) >5 mg C L⁻¹ and Cu²⁺ >2 times that of CuO NPs, lowered and enhanced the extraction efficiency, respectively. Pre-treatment of samples with 3% w v⁻¹ of hydrogen peroxide and 10 mM of ethylenediaminetetraacetic acid minimized the interferences posed by DOC and Cu²⁺, respectively. The decrease in CPE efficiency was also evident for ligands like poly(ethylene glycol). The TX-114-rich phase could be determined with either inductively coupled plasma mass spectrometry following microwave digestion, or graphite furnace atomic absorption spectrometry. The detection limits for CuO NPs were 0.02 and 0.06 μg L⁻¹ using these techniques, respectively. The optimum sample pre-treatment and CPE conditions were successfully applied to the river and wastewater samples. The relative recoveries of CuO NPs spiked at 5–100 μg L⁻¹ (as Cu) in these samples were in the range of between 59.2 and 108.2%. The approach demonstrates a robust analytical method for detecting trace levels of CuO NPs at their original states and assessing their exposure risks in real aquatic environments.     

A functionalized gold nanoparticles and Rhodamine 6G based fluorescent sensor for high sensitive and selective detection of mercury(II) in environmental water samples

A gold-nanoparticles (Au NPs)-Rhodamine 6G (Rh6G) based fluorescent sensor for detecting Hg (II) in aqueous solution has been developed. Water-soluble and monodisperse gold nanoparticles (Au NPs) has been prepared facilely and further modified with thioglycolic acid (TGA). Free Rh6G dye was strongly fluorescent in bulk solution. The sensor system composing of Rh6G and Au NPs fluoresce weakly as result of fluorescence resonance energy transfer (FRET) and collision. The fluorescence of Rh6G and Au NPs based sensor was gradually recovered due to Rh6G units departed from the surface of functionalized Au NPs in the presence of Hg(II). Based on the modulation of fluorescence quenching efficiency of Rh6G-Au NPs by Hg(II) at pH 9.0 of teraborate buffer solution, a simple, rapid, reliable and specific turn-on fluorescent assay for Hg(II) was proposed. Under the optimum conditions, the fluorescence intensity of sensor is proportional to the concentration of Hg(II). The calibration graphs are linear over the range of 5.0 × 10⁻¹⁰ to 3.55 × 10⁻⁸ mol L⁻¹, and the corresponding limit of detection (LOD) is low as 6.0 × 10⁻¹¹ mol L⁻¹. The relative standard deviation of 10 replicate measurements is 1.5% for 2.0 × 10⁻⁹ mol L⁻¹ Hg(II). In comparison with conventional fluorimetric methods for detection of mercury ion, the present nanosensor endowed with higher sensitivity and selectivity for Hg(II) in aqueous solution. Mercury(II) of real environmental water samples was determined by our proposed method with satisfactory results that were obtained by atomic absorption spectroscopy (AAS).     

Efficient hydrophobization and solvent microextraction for determination of trace nano-sized silver and titanium dioxide in natural waters

Hydrophobic silver and titanium (IV) oxide nanoparticles (commercial Ag and TiO₂ NPs with average particle sizes of 17 and 19 nm, respectively) were quantitatively transferred into organic phase in natural water samples. Five NP surface modification and solvent extraction agents (reagents) types, mercaptocarboxylic acid, alkylamine, mediator solvent, extraction solvent, and surfactant, were investigated and optimized with three-level orthogonal array design (OAD), an OA₂₇ (3¹³) matrix. The most favorable reagents and experimental conditions were then examined. The best extraction efficiencies of 78.6 and 73.7% were obtained for 1 mg L⁻¹ citrate-stabilized Ag and TiO₂ NPs, respectively, with 0.5 mM of 11-mercaptoundecanoic acid, 1.5 mM of octadecylamine, 1 mL of methanol, 150 μL of cyclohexane, 0.05 mM of tetra-n-octylammonium bromide, pH = 8.0, adsorption time of 2 h, sonication time of 3 min, and centrifugation time of 10 min. Enrichment factors were 97 and 83, for Ag and TiO₂ NPs, respectively. The optimum extraction conditions were successfully applied to genuine water samples at spiking levels of 2–100 μg L⁻¹ of Ag and TiO₂ NPs. The relative recoveries of (69.0–85.1)% and (61.5–78.5)% were obtained for Ag and TiO₂ NPs, respectively. The extracted surface-modified NPs were characterized with transmission electron microscopy, selected area electron diffraction, energy-dispersive X-ray, ultraviolet–visible, and Fourier transform infrared spectroscopic techniques. Based on the results, efficient ligand exchange and acid–base pair formation were observed on the NP surface without significant change in its original properties. The organic phase was microwave digested, and analyzed with inductively coupled plasma (ICP) optical emission spectroscopy and ICP mass spectrometry (ICP-MS). Detection limits of ICP-MS analyses of Ag and TiO₂ NPs were 0.02 and 0.07 μg L⁻¹, respectively.     

The petentiometric determination of peroxide hydrogen and glucose on the glassy electrode modified by the calix[4]arene

A new petentiometric method to determine peroxide hydrogen and glucose had been studied. This method had been applied on the petentiometric determination of peroxide hydrogen and glucose in the total ionic strength adjustment buffer (TISAB) (pH 7.5) solution with the glassy electrode modified by the calix[4]arene. The glassy carbon electrode covered with the calix[4]arene depended on the H₂O₂ concentration in the range of log[H₂O₂] from −3.3 to −1.2 in the solution of TISAB (pH 7.5) with nearly Nernstian slope of about 65.6 ± 3 mV and the detection limit of peroxide hydrogen was 4.0 × 10⁻⁵ mol L⁻¹. The glassy carbon electrode covered with the calix[4]arene depended on the glucose concentration in the range of log[glucose] from −3.6 to −2.8 in the solution of TISAB (pH 7.5) with nearly Nernstian slope of about 50.2 ± 2 mV and the detection limit of glucose was 2.0 × 10⁻⁵ mol L⁻¹. The electrode had the good selectivity, sensitivity, stability and repeatability.     

Preparation of a new solid state fluoride ion selective electrode and application

A new solid state fluoride ion selective electrode composed of 70% Ag₂S, 10% Cu₂S and 20% CaF₂ has been developed. An analytically useful potential change occurred, from 1 × 10⁻⁶ to 1 × 10⁻¹ M fluoride ion. The slope of the linear portion (1 × 10⁻¹-1 × 10⁻⁵ M) was about 26 ± 2 mV/10-fold concentration changes in fluoride. It was found that pH change between 1 and 8 had no effect on the potential of the electrode. There was no interference of most common cations such as K^+, Na⁺, Ca²⁺ and Mg²⁺ and anions such as Cl⁻, NO₃⁻, SO₄²⁻ and PO₄³⁻. The lifetime of the electrode was more than 2 years, when used at least 4-5 times a day, and the response time was about 60 s.The measurements were made at constant ionic strength (0.1 M NaNO₃) and at room temperature. This electrode has been used for the determination of fluoride ion in Ankara city tap water and in bottled spring water using standard addition method. The validation of the electrode has been made with a commercial fluoride ion selective electrode (Orion) and high consistency was obtained.     

Application of 2-mercaptobenzothiazole self-assembled monolayer on polycrystalline gold electrode as a nanosensor for determination of Ag(I)

Fabrication and application of a voltammetric sensor based on gold 2-mercaptobenzothiazole self-assembled monolayer (Au-MBT SAM) for determination of silver ion is described. Preliminary experiments were performed to characterize the monolayer. The surface pK_a determined for the MBT monolayer is 7.0. This value was obtained by impedimetric titration of the monolayer in the presence of Fe(CN)₆^3−/4− as a redox probe. The extent of surface coverage was evaluated as 1.52 × 10⁻⁹ mol cm⁻² based on charged consumed for reductive desorption of the monolayer in the 0.50 M NaOH solution. Then the sensor was used for determination of Ag(I) by square wave voltammetry. The parameters affecting the sensor response, such as pH and supporting electrolyte, were optimized. A dynamic calibration curve with two linear parts was obtained in the concentration ranges of 5 × 10⁻⁸-8 × 10⁻⁷ and 1 × 10⁻⁶-1 × 10⁻⁵ M of Ag(I). The detection limit adopted from cathodic striping square wave voltammetry was as 1 × 10⁻⁸ M for n = 7. Furthermore, the effect of potential interfering ions on the determination of Ag(I) was studied, and an appropriate method was used for the elimination of this effect.     

A combined osmotic pressure and cake filtration model for crossflow nanofiltration of natural organic matter

A combined osmotic pressure and cake filtration model for crossflow nanofiltration of natural organic matter (NOM) was developed and successfully used to determine model parameters (i.e. permeability reduction factor (η) and specific cake resistance (α_cake)) for salt concentrations, NOM concentrations, and ionic strength of salt species (Na⁺ and Ca⁺⁺). In the absence of NOM, with increasing salt concentration from 0.004 to 0.1 M, permeability reduction factor (η)) decreased from 0.99 to 0.72 and 0.94 to 0.44 for monovalent cation (Na⁺) and divalent cation (Ca⁺⁺), respectively. This reduced membrane permeability was due to salt concentrations and salt species. In the presence of NOM, specific cake resistance tended to increase with increasing NOM concentration and ionic strength in the range of 0.85 × 10¹⁵–3.66 × 10¹⁵ m kg⁻¹. Solutions containing divalent cation exhibited higher normalized flux decline (J_v/J_vo = 0.685–0.632) and specific cake resistance (α_cake = 2.89 × 10¹⁵–6.24 × 10¹⁵ m kg⁻¹) than those containing monovalent cation, indicating a highly compacted NOM accumulation, thus increased permeate flow resistance during NF filtration experiments. After membrane cleaning, divalent cation exhibited lower water flux recovery than monovalent cation, suggesting higher non-recoverable (R_non-rec) resistance than monovalent cation.     

Quantitative parameters for the sequestering capacity of polyacrylates towards alkaline earth metal ions

The complex formation constants of polyacrylic (PAA) ligands (1.4≤log N≤2.4, N=number of monomer units) with calcium and magnesium ions were determined in different ionic media at different ionic strengths, 0≤I≤1 mol l⁻¹, at t=25 °C. Experimental pH-metric data in the presence of Ca²⁺ or Mg²⁺ were firstly analysed in terms of apparent protonation constants, log K^H*, using the “three parameter model” proposed by Högfeldt; differences in log K^H*, determined in different ionic media, were interpreted in terms of complex species formation. The only species present in the system M-PAA (M=Ca²⁺ or Mg²⁺) is ML₂: attempts to find species of different stoichiometry were unsuccessful. The stability dependence of this species on ionic strength, on the degree of neutralisation (α) and on PAA molecular weight is discussed using empirical equations. The formation constant, log β₂, is significantly higher for Ca²⁺ than for Mg²⁺: at I=0.1 mol l⁻¹ (NaCl), log N=1.8 and α=0.5, log β₂^Ca=4.43 and log β₂^Mg=4.24. The formation of polyacrylate-alkaline earth metal complexes is discussed in the light of sequestering effects in natural waters.     

Application of a new potentiometric method for determination of phosphate based on a surfactant-modified zeolite carbon-paste electrode (SMZ-CPE)

A phosphate-selective electrode based on surfactant-modified zeolite (SMZ) particles into carbon-paste has been proposed (SMZ-CPE). The electrode was fully characterized in terms of composition, response time, ionic strength, thermal stability and usable pH range. The electrode containing 20% SMZ exhibited linear response range to phosphate species in the range of 1.58 × 10⁻⁵ to 1.00 × 10⁻² M with a detection limit of 1.28 × 10⁻⁵ M and a Nernstian slope of 29.9 ± 0.9 mV per decade of phosphate concentration. The electrode response to phosphate remains constant in the pH range of 4-12 and in the presence of 1 × 10⁻⁴ to 4 × 10⁻³ M NaNO₃. The response of the electrode reaches equilibrium within several seconds after immersing the electrode in phosphate solution. Common anions such as Cl⁻, Br⁻, I⁻, NO₃⁻, SO₄²⁻ and Cr₂O₇²⁻ have little effect on the determination of phosphate but AsO₄³⁻ shows some interference. A successful application of the electrode for determination of phosphate in a fertilizer, using direct potentiometry, is presented. The electrode was also used for the potentiometric titration of phosphate. The validation of the obtained results in each case was proved by statistical methods.     

Voltammetric determination of hafnium with calconcarboxylic acid

The interaction of Hf(IV) with calconcarboxylic acid (1-(2-hydroxy-4-sulfo-1-naphtylazo)-3-naphtolic acid) was investigated by cyclic voltammetry varying various factors (pH, ionic strength, contents of ethanol and scan rate). Optimal conditions of Hf(IV) determination in the presence of calconcarboxylic acid were found: acetic buffer solution pH 2.6, scan rate 0.5 Vs⁻¹. The detection limit of Hf(IV) concentration was 2.46 × 10⁻⁷ mol L⁻¹. The influence of foreign ions on the Hf(IV) determination was studied. It was established that some metals like cadmium, nickel, zinc, copper and titanium could be determined with Hf(IV) simultaneously. The method of voltammetric determination of hafnium was tested on model solutions and used in the determination of Hf(IV) in terbium-base alloy.     

Ionic liquid salt bridge based on tributyl(2-methoxyethyl)phosphonium bis(pentafluoroethanesulfonyl)amide for stable liquid junction potentials in highly diluted aqueous electrolyte solutions

A moderately hydrophobic ionic liquid, tributyl(2-methoxyethyl)phosphonium bis(pentafluoroethanesulfonyl)amide ([TBMOEP⁺][C₂C₂N⁻]), shows a very stable liquid junction potential upon contact with an aqueous solution whose ionic strength is as low as 1 μmol dm⁻³. The stability with the maximum excursion of the potential within ±0.5 mV for 30 min is very promising for accurate determination of pH and other single ion activities potentiometrically.     

Factorial design for optimization of experimental variables in preconcentration of copper by a chromotropic acid loaded Q-Sepharose adsorbent

An agarose-based anion exchanger (Q-Sepharose) was loaded with chromotropic acid (CTA) and used for column preconcentration and determination of copper by flame AAS. Preliminary experiments indicated that a sample pH of 5.7-6.5 is best suited for accumulation of copper and a 2.5 ml portion of a 0.02 mol l⁻¹ HCl solution can efficiently desorb the analyte from the column. An incomplete factorial design was used for optimization of five different variables that affect recovery of copper. The results indicated that ionic strength, pH and sample volume variables are the most important effects, respectively. Hence, these variables and their possible interactions were studied more carefully. In optimized conditions, the column could tolerate up to 0.18 mol l⁻¹ sodium nitrate in the matrix. A 5 ml portion of a 0.02 mol l⁻¹ CTA was sufficient for loading of a 0.5 ml column prior to preconcentration of copper from a 150 ml sample solution. Matrix ions of Ca²⁺, Mg²⁺, Na⁺ and K⁺ and potentially interfering ions of Pb²⁺, Ni²⁺, Cd²⁺, Co²⁺, Zn²⁺ and Mn²⁺ with relatively high concentrations did not have any significant effect on the recovery of the analyte. A preconcentration factor of 60 and a detection limit of 1.0 μg l⁻¹ was obtained for the determination of copper by the flame AAS method. A precision better than 2.5%, expressed as R.S.D., was also achieved. Application of the method to tap water and two different river water samples resulted in values well confirmed by direct determinations with ET-AAS.