Determination of fifteen priority phenolic compounds in environmental samples from Andalusia (Spain) by liquid chromatography–mass spectrometry

This work describes the optimisation of a method to determinate fifteen phenolic compounds in waters, sediments and biota (green marine algae) by liquid chromatography coupled to mass spectrometry (LC-MS) with atmospheric pressure chemical ionisation (APCI) in the negative mode. The LC separations of the studied compounds and their MS parameters were optimised in order to improve selectivity and sensitivity. Separation was carried out with a C₁₈ column using methanol and 0.005% acid acetic as mobile phase in gradient mode. The molecular ion was selected for the quantitation in selective ion monitoring (SIM) mode. A solid-phase extraction (SPE) method was applied in order to preconcentrate the target analytes from water samples. However, extraction of the compounds from sediment and biota samples was carried out by liquid–solid extraction with methanol/water after studying the influence of other organic solvents. In addition, a clean-up step by SPE with HLB Oasis cartridges was necessary for sediments and biota. The proposed analytical methodology was validated in the target environmental matrices by the analysis of spiked blank matrix samples. Detection limits were 10–50 ng L^–1 for water, 1–5 g kg^–1 for sediments and 2.5–5 g kg^–1 for biota samples. Good recoveries and precision values were obtained for all matrices. This methodology has been successfully applied to the analysis of incurred water, sediment and biota samples from Andalusia (Spain).     

Integrated procedure for determination of endocrine-disrupting activity in surface waters and sediments by use of the biological technique recombinant yeast assay and chemical analysis by LC–ESI-MS

An integrated procedure using mass spectrometry and molecular biology for determination of estrogenicity in natural waters and sediments is reported. Solid-phase extraction (SPE) and pressurized-liquid extraction (PLE), respectively, were used for isolation of endocrine-disrupting compounds (EDC) from surface waters and sediments, followed by liquid chromatography–mass spectrometry using an electrospray interface (LC–ESI-MS). Twenty seven EDC were determined: non-ionic surfactants (nonylphenol ethoxylate), alkylphenols (e.g. nonylphenol and octylphenol), bisphenol A, phthalates, and natural and synthetic steroid sex hormones. Limits of detection varied from 0.02 to 0.22 g L^–1 and from 1 to 10 g kg^–1 in water and sediments, respectively. Recoveries ranged from 65 to 125% and 73 to 97% for waters and sediments, respectively. In addition to LC–ESI-MS determination, extracts obtained by SPE and PLE were analyzed by the recombinant yeast assay (RYA) to assess total estrogenic activity. This bioassay detects natural estrogens and xenoestrogens, producing a quantitative measurement of EDC irrespective of the identity of the chemical responsible for the activity. As a novelty, a relative estrogenicity factor was determined for 19 analytes with EC ₅₀ values ranging from 10^–10 to 10^–9 mol L^–1 for synthetic estrogens, from 10^–7 to 10^–5 mol L^–1 for alkylphenol derivatives, and from 10^–5 to 10^–4 mol L^–1 for phthalates and benzothiazoles. By use of this integrated chemical–ecotoxicological approach good correlation was usually established between chemical composition and estrogenic effects for surface water and sediment samples from Portugal. Estrogenic activity observed was mainly attributed to the presence of nonylphenolic compounds (with concentrations of NP ranging from 0.1 up to 44 g L^–1 in waters and up to 1172 g kg^–1 in sediments), and to the sporadic presence of estrogens, detected at ng L^–1 levels.     

Fluorimetric determination of bromate by ion-exchange separation and post-column derivatization

For the determination of bromate in drinking water a stopped-flow post-column reaction was developed following the separation of bromate from the matrix by an anion-exchange column. In the post-column reaction the analyte was used to oxidize the azo dye sulfonaphtholazoresorcinol, SNAR, and the residual amount was converted into a fluorescent binuclear complex by an excess of gallium ions. The fluorescence was monitored at 585 nm, with a maximum excitation wavelength at 521 nm. The determination of bromate is based on the decrease of the fluorescence intensity with increasing bromate concentration. The given hydrodynamic parameters and the condition of equal flow rates of the two branch streams at each T-piece have to be considered as an important criterion for the experimental set-up. The volume flows and the concentrations required for the reagent solutions in the influent of each T-piece were determined as a result of batch experiments and theoretical considerations. The limit of detection was 0.28 g L^–1bromate for the flow method, which shows linearity up to 15 gL^–1 bromate.     

Determination of biogenic amines in wine after clean-up by solid-phase extraction

Summary A suitable method for the determination of 16 biogenic amines in wine has been developed. The method involves clean-up of wine samples using ion-exchange cartridges and a preconcentration step, under controlled vacuum, before derivatization of the amines by treatment with phthalaldehyde (PA) and reversedphase HPLC with gradient elution and fluorimetric detection. Linearity of response was obtained for all the biogenic amines from 100 g L^–1 to mg L^–1. Limits of detection for the amines were similar for all PA-derivatives (25–50 g L^–1) and the quantitation limits were about 0.1 mg L^–1. After clean-up and preconcentration, the concentration levels increased 10-fold for all amines except putrescine and cadaverine, which gave poor recovery by this method unlike the rest which gave recoveries of almost 90%. The overall process was successfully applied to identify and quantify biogenic amines in several red wines from the Tarragona region.     

Determination of Triphenyltin and Its Metabolite Diphenyltin in Culture Medium by High-Performance Liquid Chromatography with UV detection

A method for the determination of triphenyltin and diphenyltin was developed by reversed-phase high-performance liquid chromatography with UV detection. Triphenyltin and diphenyltin were separated using a reversed-phase Symmetry C₁₈ column (150 × 3.9 mm, 5 m) with tetrahydrofuran-water-acetonitrile-glacial acetic acid (13:25:5:7, v/v) containing 0.05% triethylamine and 1.0% sodium acetate as mobile phase at 0.50 mL min^–1 and detection at 257 nm. The calibration curves were linear from 0.26 mol L^–1 to 1100 mol L^–1 for triphenyltin with a correlation coefficient of 0.9999 (n=12) and from 0.60 mol L^–1 to 1200 mol L^–1 for diphenyltin with a correlation coefficient of 0.9991 (n=12), respectively. The detection limits of triphenyltin and diphenyltin were 0.2 mol L^–1 and 0.4 mol L^–1, respectively. The method was successfully applied to the determination of triphenyltin and its metabolite diphenyltin in culture medium. The recoveries of triphenyltin and diphenyltin were in the ranges of 97.7% to 103.3% and 85.5% to 91.6%, respectively.     

Trace analysis of bromate and bromide with ion chromatography on coated reversed phase materials

 The ion-chromatographic method for trace analysis of bromate and bromide presented in this paper is based on coating of reversed phase (RP) material with an ionogenic agent, tetrakisdecylammonium bromide, to obtain a pseudo ion-exchange column. The analysis is carried out with usual HPLC pump and UV-detection near 200 nm. Some commercially available RP materials were tested for the coating procedure. The differences between the reversed phases are not significant. All HETP values are calculated between 0.02 and 0.14 mm. The calibration, the sensitivity of the method and the long-time stability of the coated column were tested with one selected RP material. It is shown that the simultaneous trace analysis of bromate and bromide in surface and drinking waters with chloride concentrations up to 50 mg/L is possible without any clean-up on Ag precolumns. A comparison of performance data with a determination method for bromate and bromide employing a commercially available equipment demonstrates the efficiency of the new technique. Received: 23 February 1996/Received: 10 May 1996/Accepted: 14 May 1996     

Trace analysis of bromate and bromide with ion chromatography on coated reversed phase materials

 The ion-chromatographic method for trace analysis of bromate and bromide presented in this paper is based on coating of reversed phase (RP) material with an ionogenic agent, tetrakisdecylammonium bromide, to obtain a pseudo ion-exchange column. The analysis is carried out with usual HPLC pump and UV-detection near 200 nm. Some commercially available RP materials were tested for the coating procedure. The differences between the reversed phases are not significant. All HETP values are calculated between 0.02 and 0.14 mm. The calibration, the sensitivity of the method and the long-time stability of the coated column were tested with one selected RP material. It is shown that the simultaneous trace analysis of bromate and bromide in surface and drinking waters with chloride concentrations up to 50 mg/L is possible without any clean-up on Ag precolumns. A comparison of performance data with a determination method for bromate and bromide employing a commercially available equipment demonstrates the efficiency of the new technique. Received: 23 February 1996/Received: 10 May 1996/Accepted: 14 May 1996     

Simultaneous determination of DTPA,EDTA, and NTA by capillary electrophoresis after complexation with copper

We present a method for simultaneous determination of the aminopolycarboxylic acids DTPA, EDTA and NTA in dishwashing detergents, paper mill waters, and natural waters by capillary electrophoresis (CE). The complexing agents were examined as their copper(II) complexes and separated by conventional CE with reversed polarity of the applied voltage. The optimum separation conditions were established by varying the pH and phosphate and tetradecyltrimethylammonium bromide (TTAB) concentrations in the run buffer. The separations were carried out in a fused-silica capillary (61 cm×75 m i.d.) filled with phosphate buffer (80 mmol L^–1, TTAB concentration 0.5 mmol L^–1, pH 7.1, voltage –20 kV) using direct UV detection at 191 and 254 nm. With this CE method all the peaks in the electropherograms were properly separated, the calibration plots gave good correlation coefficients and all three complexing agents could be detected in less than 4 min. Linear calibration plots were obtained for CuDTPA, CuEDTA and CuNTA; limits of detection were 0.03 mmol L^–1 for all complexing agents and recoveries for all tested samples were within the range 104±7%. Results obtained from dishwashing detergent samples were found to be reliable and comparable with those from HPLC (R²=0.989) and UV–Vis (R²=0.985) methods.     

Simultaneous determination of DTPA,EDTA, and NTA by UV–visible spectrometry and HPLC

In this study, UV–visible spectrophotometry (UV–Vis) and high-performance liquid chromatography (HPLC) were used for simultaneous analysis of chelating agents diethylenetriamine pentaacetic acid (DTPA), ethylenediamine tetraacetic acid (EDTA), and nitrilotriacetic acid (NTA), as their metal chelates in dishwashing detergents, natural waters, and pulp mill water. The total amounts of the chelating agents in dishwashing detergents were verified by potentiometric titration with Fe(III) solution. Nickel(II) chelates were determined by UV–Vis and iron(III)chelates by HPLC and titration. Recoveries of DTPA, EDTA, and NTA from a standard mixture of analytes by UV–Vis were 107±7, 101±12 and 94±13%, respectively, and the recovery of the total amount of complexing agents was 99±4%. The limits of detection for DTPA, EDTA, and NTA were 667, 324, and 739 mol L^–1, respectively. In HPLC measurements the optimized mobile phase contained 0.03 mol L^–1 sodium acetate, 0.002 mol L^–1 tetrabutylammonium bromide, and 5% methanol at pH 3.15 and the detection was by UV–Vis detection at 254 nm. All three complexing agents could be separated from each other in a simultaneous analysis in less than 5 min. The limits of detection were 0.34, 0.27, and 0.62 mol L^–1 for DTPA, EDTA, and NTA, respectively. The total amounts of the analytes measured in the dishwashing detergents by the three techniques were found to be highly comparable (ANOVA: F=0.04, P=0.96). R² values were 0.99 for EDTA, 0.99 for NTA, and 0.99 for all the results when UV–Vis and HPLC determinations were compared using regression lines. The UV–Vis and HPLC methods were proved to be viable also for analyses of natural and pulp mill waters. The absence of matrix interferences was verified by the standard addition technique.     

Adsorptive stripping voltammetric determination of cobalt in the presence of dimethylglyoxime and cetyltrimethylammonium bromide

A sensitive procedure for determination of micro-traces of Co(II) by adsorptive stripping voltammetry is proposed. The procedure exploits the enhancement of the cobalt peak obtained by use of the system Co(II)–dimethylglyoxime–piperazine-1,4-bis(2-ethanesulfonic acid)–cetyltrimethylammonium bromide. Using the optimized conditions, a detection limit (based on the 3 criterion) for Co(II) of 1.2×10^–11 mol L^–1 (0.7 ng L^–1) was achieved. The calibration plot for an accumulation time of 30 s was linear from 5×10^–11 to 4×10^–9 mol L^–1. The procedure was validated by analysis of certified reference materials and natural water samples.     

Determination of a flame retardant hydrolysis product in human urine by SPE and LC–MS. Comparison of molecularly imprinted solid-phase extraction with a mixed-mode anion exchanger

Diphenyl phosphate is a hydrolysis product and possible metabolite of the flame retardant and plasticiser additive triphenyl phosphate. A molecularly imprinted polymer solid-phase extraction (MISPE) method for extracting diphenyl phosphate from aqueous solutions has been developed and compared with SPE using a commercially available mixed-mode anion exchanger. The imprinted polymer was prepared using 2-vinylpyridine (2-Vpy) as the functional monomer, ethylene glycol dimethacrylate (EGDMA) as the cross-linker, and a structural analogue of the analyte as the template molecule. The imprinted polymer was evaluated for use as a SPE sorbent, in tests with both aqueous standards and spiked urine samples, by comparing recovery and breakthrough data obtained using the imprinted form of the polymer and a non-imprinted form (NIP). Extraction from aqueous solutions resulted in more than 80% recovery. Adsorption by the molecularly imprinted polymer (MIP) was non-selective , but selectivity was achieved by selective desorption in the wash steps. Diphenyl phosphate could also be selectively extracted from urine samples, although the urine matrix reduced the capacity of the MISPE cartridges. Recoveries from urine extraction were higher than 70%. It was important to control pH during sample loading. The MISPE method was found to yield a less complex LC–ESI–MS chromatogram of the urine extracts compared with the mixed-mode anion-exchanger method. An LC–ESI–MS method using a Hypercarb LC column with a graphitised carbon stationary phase was also evaluated for organophosphate diesters. LC–ESI–MS using negative-ion detection in selected ion monitoring (SIM) mode was shown to be linear for diphenyl phosphate in the range 0.08–20 ng L^–1.     

Determination of levodopa methyl ester and its metabolites in rat serum by CZE with amperometric detection

A reliable and reproducible method, capillary zone electrophoresis with amperometric detection (CZE–AD), has been developed for separation and quantification of levodopa methyl ester (LDME) and its biotransformation products levodopa (L-DOPA) and dopamine (DA) in rat serum. A carbon-disk electrode was used as working electrode. The optimum conditions for CZE detection were 50 mmol L^–1 phosphate solution at pH 7.0 as running buffer, 17 kV as separation voltage, 1.0 V (vs Ag/AgCl, 3.0 mol L^–1) as detection potential, and sample injection for 8 s at 17 kV. The linear ranges were from 2.4×10^–2 to 2.2 g mL^–1 for LDME, 2.9×10^–1 to 49.5 g mL^–1 for L-DOPA, and 1.4×10^–2 to 1.5 g mL^–1 for DA with correlation coefficients of 0.9997, 0.9994, and 0.9999, respectively. The detection limits for LDME, L-DOPA, and DA were 14.6, 98.0, and 9.7 ng mL^–1, respectively. Recoveries were 80.3% for LDME, 93.5% for L-DOPA, and 86.5% for DA. This method was applied to serum samples after intravenous injection of LDME and L-DOPA to rats.     

Spectrophotometric Determination of Polyoxyethylene Nonionic Surfactants in Ground Waters Using Triple-Stage Solid-Phase Extraction Followed by an Improved Ferric Thiocyanate Complexation Method

A practical spectrophotometric determination method for polyoxyethylene nonionic surfactants in ground waters was established, which is based on a ferric thiocyanate complexation colorimetric method preceded by a triple-stage solid-phase extraction technique using SCX, SAX, and C18 cartridges interconnected. Cationic and anionic surfactants interfere with the determination and were therefore effectively trapped and isolated by the SCX and SAX solid phases, respectively. Nonionic surfactants (NSs) were finally introduced and concentrated in the C18 cartridge. The analyte was quantitatively eluted from the C18 sorbent, and the residue was subjected to the colorimetric determination. The calibration line was linear (r² = 0.9997) up to 200µgL^–1 of heptaoxyethylenedodecylether when analyzing sample sizes of 100mL. Overall recoveries were 95–97% with an RSD of less than 3%. The method was applied to the analysis of river water, and 6.4µgL^–1 of NSs as heptaoxyethylenedodecylether was found by means of the standard addition method. The proposed method is very practical and features minimum consumption of chemicals.     

Comparison of Partial Least Squares Regression and H-Point Standard Addition Method for Simultaneous Spectrophotometric Determination of Zinc, Cobalt and Nickel by 1-(2-Pyridylazo)2-Naphthol in Micellar Media

Simultaneous spectrophotometric methods are described for the determination of Zn²⁺, Co²⁺ and Ni²⁺ by 1-(2-pyridylazo)2-naphthol (PAN) in micellar media, using absorbance correction-H-point standard addition method (HPSAM) and partial least squares (PLS) regression. The ligand and its metal complexes, i.e. Zn²⁺-PAN, Co²⁺-PAN and Ni²⁺-PAN, were made water-soluble by the neutral surfactant Triton X-100, and therefore extraction with organic solvents was no longer required. Formation of all of these complexes was complete within 10min at pH 9.2. The linear range was 0.1–1.5mgL^–1 for Zn²⁺, 0.1–2.0mgL^–1 for Co²⁺ and 0.1–2.0mgL^–1 for Ni²⁺. The relative standard deviation (RSD) for the simultaneous determination of 0.50mgL^–1 each of Zn²⁺, Ni²⁺ and Co²⁺ by applying the H-point standard addition method was 2.55%, 2.04% and 3.70%, respectively. The total relative standard error for applying the PLS method to 9 synthetic samples in the linear ranges of these metals was 1.8%. Interference effects of common anions and cations were studied, and both methods were applied to the simultaneous determination of Zn²⁺, Co²⁺ and Ni²⁺ in alloy samples.     

Solid-Phase Extraction and LC with Fluorescence Detection for Analysis of PAHs in Rainwater

The efficiency of extraction of polycyclic aromatic hydrocarbons (PAHs) from rainwater by solid-phase extraction (SPE) with three different types of cartridge, and analysis by high-performance liquid chromatography with fluorescence detection, are discussed in this paper. Three cartridges were investigated but only one was suitable. After equilibration in a desiccator for 65–80 h or in ambient air for 90–100 h the SPE cartridges were activated with 5 mL dichloromethane then 5 mL 2-propanol. The volume of sample passed through the cartridges was 50 mL; after loading of the sample the cartridges were dried under vacuum for approximately 20 min by application of a pressure of 15 mbar to the SPE manifold. The PAHs were eluted with 5 mL dichloromethane–hexane, 50:50 (v/v). The flow rate used for conditioning, sample loading, and elution was 2.5 mL min⁻¹, achieved by application of a pressure of 6 mbar. For analysis of PAHs in rainwater, recovery was between 67 and 99%, the relative standard deviation varied between 2 and 5%, and the detection limits of the method were less than 16.9 ng L⁻¹ for several PAHs. These optimum conditions were used for analysis of rainwater collected between June 2002 and May 2003 at two sites in Alsace (eastern France) and 17 PAHs were quantitatively determined. Concentrations varied between 1.6 and 968.1 ng L⁻¹.     

Thick-film electrodes for measurement of superoxide and hydrogen peroxide based on direct protein–electrode contacts

Cytochrome c was immobilized on screen-printed thick-film gold electrodes by a self-assembly approach using mixed monolayers of mercaptoundecanoic acid and mercaptoundecanol. Cyclic voltammetry revealed quasi-reversible electrochemical behavior of the covalently fixed protein with a formal potential of +10 mV vs. Ag/AgCl. Polarized at +150 mV vs. Ag/AgCl the electrode was found to be sensitive to superoxide radicals in the range 300–1200 nmol L^–1. Compared with metal needle electrodes sensitivity and reproducibility could be improved and combined with the easiness of preparation. This allows the fabrication of disposable sensors for nanomolar superoxide concentrations. By changing the electrode potential the sensor can be switched from response to superoxide radicals to hydrogen peroxide—another reactive oxygen species. H₂O₂ sensitivity can be provided in the range 10–1000 mol L^–1 which makes the electrode suitable for oxidative stress studies.     

Development of a miniature dielectric barrier discharge–optical emission spectrometric system for bromide and bromate screening in environmental water samples

Dielectric barrier discharge (DBD) at atmospheric pressure provides an efficient radiation source for the excitation of bromine and it is used for the first time for optical emission spectrometric (OES) detection of bromide and bromate. A portable DBD–OES system is developed for screening potential pollution from bromide and bromate in environmental waters. Bromide is on-line oxidized to bromine for in-situ generation of volatile bromine. Meanwhile, a helium stream carries bromine into the DBD micro-plasma for its excitation at a discharging voltage of 3.7 kV and optical emission spectrometric detection with a QE65000 charge-coupled device (CCD) spectrometer in the near-infrared spectral region. Similarly, the quantification of bromate is performed by its pre-reduction into bromide and then oxidized to bromine. The spectral characteristics and configuration of the DBD micro-plasma excitation source in addition to the oxidation vapor generation of bromine have been thoroughly investigated. With a sampling volume of 1 mL, a linear range of 0.05–10.0 mg L⁻¹ is obtained with a detection limit of 0.014 mg L⁻¹ by measuring the emission at 827 nm. A precision of 2.3% is achieved at 3 mg L⁻¹ bromide. The system is validated by bromine detection in certified reference material of laver (GBW10023) at mg L⁻¹ level, giving rise to satisfactory agreement. In addition, it is further demonstrated by screening trace bromide and bromate as well as spiking recoveries in a series of environmental water samples.     

Multielement determinations in ground water ultrafiltrates using inductively coupled plasma mass spectrometry and monostandard neutron activation analysis

Twelve ultrafiltrates of two ground waters rich in humic substances (up to 97.8 mg CL^–1) and in salinity (up to: cations 44.3 meq L^–1, anions 44.9 meq L^–1) were investigated with ICP-MS and with NAA in parallel. With both techniques 22 elements were analysed in a wide concentration range (mg/L to ng/L). Ultrafiltration at pore sizes from 1000 nm down to 1 nm lowers the humic colloid content as well as the concentration of the colloidborne polyvalent cations. Carbon interferences were studied in detail using artificially prepared model waters. The detection limits of ICP-MS in the ultrafiltrates (0.01 g/L–10 g/L) and in pure analyte solutions (5 ng/L–600 ng/L) are compared with those of NAA for pure water analysis (0.004 ng/L–50 ng/L).Dedicated to Professor Dr. H. Schmidbaur on the occasion of his 60th birthday     

Method Development for the Determination of Seven Organic Acids in Wines by Reversed-Phase High Performance Liquid Chromatography

A simple high performance liquid chromatographic method is described for the determination of seven organic acids usually found in wines. The acids were eluted in the isocratic mode, in less than 12 min, on a reversed-phase ODS-2 (250 × 4 mm I.D.), 5 m, column with 0.02 M potassium dihydrogen phosphate (pH 2.88), to which was added a small amount of methanol (2%) as organic modifier, and were detected by ultraviolet absorbance at 230 nm. Galacturonic, tartaric, malic (both enantiomers), lactic, acetic, citric and succinic acids were determined, using xanthine as a chromatographic internal standard. The method was applied to white and red wines of Greek origin, after sample clean-up with polyvinylpyrrolidone, followed by passage through SAX cartridges and yielded recoveries from 78.0 to 106.8%. The limits of detection ranged between 0.001–0.05 g.L^–1 and the linear ranges between 0.003–2.0 g.L^–1.     

Determination of Benzimidazole Fungicides by HPLC with Fluorescence Detection After Micellar Extraction

An analytical method was developed to determine the benzimidazole fungicides and their residues (benomyl, carbendazim, thiabendazole and fuberidazole) in real water samples. Analyses were performed by reverse phase (RP) HPLC with direct fluorescence detection with mobile phase methanol:water, 40:60 (v/v) with 0.6% (v/v) ammonia. The extraction of analytes from water samples was performed with the use of micellar systems. Specifically, oligoethylene glycol monoalkyl ether (Genapol X-080) and polyoxyethylene 10 lauryl ether (POLE) were used as extractants. The recoveries of fungicides obtained in spiked water samples ranged from 68% to 94% for Genapol and from 68% to 96% for POLE. The limit of detection (LOD) was lower than 6 g L^–1 for carbendazim, 7 g L^–1 benomyl, 0.15 g L^–1 for thiabendazole and 0.01 g L^–1 for fuberidazole in both surfactants.