Determination of pesticides in drinking water by micellar electrokinetic capillary chromatography

A new analytical procedure using a two-step sample preconcentration (solid-phase extraction (SPE) and field-amplified sample stacking) prior to separation by micellar electrokinetic capillary chromatography was developed for the determination of 14 pesticides such as aldicarb, carbofuran, isoproturon, chlorotoluron, metolachlor, mecoprop, dichlorprop, MCPA, 2,4-D, methoxychlor, TDE, DDT, dieldrin, and DDE in drinking water. Good recoveries of pesticides were obtained using SPE with sample pH adjusted to 2-3. Field-amplified sample stacking was found to give enrichment factors up to 30-fold preconcentration of various pesticides under reversed polarity at -2 kV for 50 s. The optimized background electrolyte (BGE) consisted of 50 mM sodium dodecyl sulfate (SDS), 10 mM borate buffer, 15 mM beta-cyclodextrin (beta-CD), and 22% acetonitrile at pH 9.6, running was under 25 kV and detection at 202 nm. Good linearity was obtained for all pesticides with detection limits down to 0.04-0.46 ng/mL and a working range of 0.1-40 ng/mL. The repeatabilities of migration time and peak area were satisfactory with relative standard deviations (RSDs) between 0.66 and 13.6% and 4.1 and 28%, respectively. All pesticides except dieldrin were found to be detected at concentrations at least tenfold lower than the World Health Organization (WHO) guideline values. The analytical procedure developed offers an economic method for fast screening of multiple pesticide residues in drinking water for health protection. It had been applied to determine carbofuran and MCPA in agricultural run-off water samples, giving satisfactory repeatabilities of 10 and 12%, respectively, with n=5 for the determination of pesticides in contaminated water samples.     

Determination of pesticides in wine using micellar electrokinetic chromatography with UV detection and sample stacking

In this work, the analysis of a group of four fungicides (pyrimethanil, nuarimol, procymidone and cyprodinil) and one insecticide (pirimicarb) by micellar electrokinetic chromatography (MEKC) with UV detection using the on-line preconcentration strategy called reversed electrode polarity stacking mode (REPSM) is proposed. After optimisation, an adequate separation electrolyte for the separation and stacking of these pesticides was obtained which consisted of 100 mM borate, 60 mM sodium dodecyl sulphate (SDS), at pH 9.0 and 2% 2-propanol. The use of this running buffer together with the REPSM preconcentration method provided limits of detection (LODs) between 38.3 and 241 microg/L. In order to apply the developed methodology for the analysis of these pesticides in wine samples, several off-line preconcentration strategies (mainly, solid-phase extraction, SPE, and solid-phase microextraction, SPME) were tested. Although the use of a SPE procedure, optimized in this work for water samples, using Oasis HLB cartridges, provided mean recovery values between 79 and 100% for spiked water samples, it could not be applied to the extraction of these pesticides from wine samples due to high interference from the sample matrix. However, the use of a SPME procedure using polydimethylsiloxane/divynilbenzene (PDMS/DVB) fibers allowed the selective extraction of four of the five pesticides which could be perfectly determined. The final combination of the off-line SPME and on-line REPSM preconcentration strategies allowed obtaining LODs between 17.6 and 32.3 microg/L.     

Multiresidue method for the simultaneous determination of four groups of pesticides in ground and drinking waters,using solid-phase microextraction-gas chromatography with electron-capture and thermionic specific detection

A common sample preparation procedure capable of efficiently concentrating various groups of pesticides, taking advantage of universal detectors like the mass spectrometer or combined techniques of group selective detectors like gas chromatography-electron capture detection (ECD)/thermionic specific detection (TSD), is desirable in environmental analysis. Six solid-phase microextraction fibres available for analysis of semi-volatiles (7, 30 and 100 microm poly(dimethylsiloxane) (PDMS), 85 microm polyacrylate, 60 microm PDMS-divinylbenzene (PDMS-DVB) and 65 microm Carbowax-DVB) were evaluated and the 60 microm PDMS-DVB was selected for the simultaneous extraction of 34 compounds, included in the organochlorine (OCPs), organophosphorous (OPPs), pyrethroid and triazine pesticide groups. All parameters affecting the extraction efficiency from water samples, namely fibre coating, sample agitation, pH and ionic strength, extraction temperature and time, were optimised. The analytical procedure involves solid-phase microextraction extraction, gas chromatographic separation and subsequent ECD and TSD via a post-column splitter adjusted to a split ratio of 1:10, respectively. Detection limits in the range of 1-10 ng l(-1) for OCPs, 1-30 ng l(-1) for OPPs, 20-30 ng l(-1) for pyrethroids and 8-50 ng l(-1) for triazines are easily attainable with the optimised procedure. The method validated for ground and drinking waters has low cost of implementation and operation although it requires careful maintenance.     

On-line ion-pair solid-phase extraction-liquid chromatography-mass spectrometry for the analysis of quaternary ammonium herbicides

An ion-pair on-line solid-phase extraction procedure using C8 extraction disks, suitable for liquid chromatography-mass spectrometry analysis is developed to determine quaternary ammonium herbicides (quats) in water samples. The separation of these compounds was performed using ion-pair chromatography with heptafluorobutyric acid (15 mM, pH 3.3) and acetonitrile gradient elution. Detection was carried out using a quadrupole mass spectrometer. Water sample volumes up to 50 ml can be preconcentrated with recoveries higher than 70%. Good precision and accuracy (day-to-day and run-to-run) were obtained and the detection limits ranged from 6 to 85 ng l(-1). The proposed on-line ion-pair solid-phase method enables compliance with European Community directives for drinking waters (100 ng l(-1)).     

Simultaneous quantification of neutral and acidic pharmaceuticals and pesticides at the low-ng/l level in surface and waste water

A new analytical method is presented that allows simultaneous determination of neutral and acidic pharmaceuticals and pesticides in natural waters. The compounds investigated include frequently used pharmaceuticals, i.e., the anti-epileptic carbamazepine, four analgesic/anti-flammatory drugs (ibuprofen, diclofenac, ketoprofen and naproxen) and the lipid regulator clofibric acid and important pesticides including triazines, acetamides and phenoxy acids. Sample enrichment was achieved in one step with a newly developed solid-phase extraction procedure using the Waters Oasis HLB sorbent. The neutral compounds were analyzed by GC-MS in a first step, and then the acidic compounds after derivatization with diazomethane. Relative recoveries using isotope labeled internal standards were between 71 and 118% and the detection limits were in the range of 1 to 10 ng/l in drinking water, surface water and waste water treatment plant effluents (precision: 1-15%). The developed analytical method proved to be very durable during a 3-month field study and the target analytes were detected in concentrations of 5-3,500 ng/l in waste water treatment plant effluents, river water and lake water.     

Quantitative analysis of pesticides by capillary column high performance liquid chromatography combined with solid-phase extraction

High performance liquid chromatography (HPLC) combined with solid-phase extraction was reported on, for simultaneous analysis of pesticides in this work. The separation of 12 pesticides was achieved on a C(18) capillary column with gradient elution. Sub-microlitre injection volume of the samples and a U-shaped 35 nl flow cell were used to improve the separation and detection. In addition, the method used C(18) solid-phase extraction disks to allow a 250-fold enrichment of the pesticides from fortified water and apple samples. The calculated detection limits range was 0.15-0.8 mug/l. Under the optimal extraction conditions, recoveries of 85-107% for most of the pesticides at 1.0-10.0 mug/l level, were obtained.     

Determination of pesticides in waters by automatic on-line solid-phase extraction-capillary electrophoresis

The separation of seven pesticides by micellar electrokinetic capillary chromatography in spiked water samples is described, allowing the analysis of pesticides mixtures down to a concentration of 50 microg l(-1) in less than 13 min. Calibration, pre-concentration, elution and injection into the sample vial was carried out automatically by a continuous flow system (CFS) coupled to a capillary electrophoresis system via a programmable arm. The whole system was electronically coupled by a micro-processor and completely controlled by a computer. A C18 solid-phase mini-column was used for the pre-concentration, allowing a 12-fold enrichment (as an average value) of the pesticides from fortified water samples. Under the optimal extraction conditions, recoveries between 90 and 114% for most of the pesticides were obtained.     

Application of single-drop microextraction and comparison with solid-phase microextraction and solid-phase extraction for the determination of alpha- and beta-endosulfan in water samples by gas chromatography-electron-capture detection

Water contamination due to the wide variety of pesticides used in agriculture practices is a global environmental pollution problem. The 98/83 European Directive requires the measurement of pesticides residues at a target concentration of 1.0 microg/l in surface water and 0.1 microg/l in drinking water. In order to reach the level of detection required, efficient extraction techniques are necessary. The application of a new extraction technique: single-drop microextraction (SDME), followed by gas chromatography with electron-capture detection, was assessed for determining alpha-endosulfan and beta-endosulfan in water samples. Experimental parameters which control the performance of SDME, such as selection of microextraction solvent and internal standard, optimization of organic drop volume, effects of sample stirring, temperature and salt addition, and sorption time profiles were studied. Once SDME was optimized, analytical parameters such as linearity, precision, detection and quantitation limits, plus matrix effects were evaluated. The SDME method was compared with solid-phase microextraction and solid-phase extraction with the aim of selecting the most appropriate method for a certain application.     

Optimisation and validation of a solid-phase microextraction method for simultaneous determination of different types of pesticides in water by gas chromatography-mass spectrometry

A solid-phase microextraction (SPME) method for the simultaneous determination of a large number of pesticides (46) with a wide range of polarities and chemical structures (organochlorine, organophosphorous, triazines, pyrethroids and others) in water samples by GC-MS has been developed. Three different fibres and parameters that influence the extraction and desorption efficiency were studied. The selected conditions were: a 60 microm polydimethylsiloxane/divinylbenzene (PDMS/DVB) fibre, 45 min of extraction time, sample agitation and temperature control at 60 degrees C; neither pH adjustment nor ionic strength correction were applied. Good detection limits, linearity and repeatability were obtained with this method for the 46 pesticides studied. The method was validated for 29 pesticides following the recommendations of the international norm ISO/IEC 17025 including the calculation of the uncertainties. The detection limits ranged from 4 to 17 ng l(-1). Furthermore, repeatability (6.9-20.5%) and intermediate precision (4.5-19.7%) were shown to be satisfactory. To validate matrix effects for drinking and surface water analytical recoveries were calculated for these matrices. The accuracy of the method was also evaluated by participating in a proficiency inter-laboratory test.     

Determination of polar pesticides with atmospheric pressure chemical ionisation mass spectrometry-mass spectrometry using methanol and/or acetonitrile for solid-phase desorption and gradient liquid chromatography

Thirty-seven polar pesticides, mainly triazines, phenylurea herbicides and phenoxy acids, were determined by LC-atmospheric pressure chemical ionisation MS-MS with methanol and acetonitrile as the organic modifiers. For most pesticides, detection limits were the same irrespective of the modifier. However, for the phenylurea herbicides, propachlor, carbetamide, triadimefon, triadimenol, triethylcitrate, benzothiazole and metazachlor, the results were much poorer in the presence of acetonitrile; in several cases, no meaningful results were obtained at all. When carrying out trace enrichment of 100 ml water samples on a 10x2 mm I.D. solid-phase extraction precolumn containing a polymeric sorbent, rapid desorption with a small volume of pure organic solvent and the introduction of a T-piece in between the solid-phase extraction precolumn and the analytical column was necessary. Aliquots of 300 microl of acetonitrile were optimal for the complete desorption of all analytes from the sorbent. With methanol as the modifier and when using an identification criterion of three ions, the detection limits for most analytes, in the full-scan mode, were 10-100 ng/l. The linearity of the procedure, which was tested at the 0.1 and 1 microg/l levels, was satisfactory in the positive, but not in the negative ionisation mode. The procedures were used to analyse surface water samples.     

Investigation of preconcentration strategies for the trace analysis of multi-residue pesticides in real samples by capillary electrophoresis

In this work, on-line preconcentration strategies were investigated for the multi-residue analysis of pesticides in drinking water and vegetables using micellar electrokinetic chromatography. Among the on-line strategies, sweeping and stacking with reverse migration of micelles (SRMM), with and without the insertion of a plug of water before sample injection, were contrasted. A new version of SRMM was also introduced. The modification consisted of momentarily applying a positive voltage at the inlet vial right after sample has been injected, increasing the efficiency by which the analytes are captured. Nine pesticides from different classes, carbendazim (benzimidazole), simazine, atrazine, propazine and ametryn (triazine), diuron and linuron (urea), carbaryl and propoxur (carbamate), were baseline separated in less than 6 min with a electrolyte composed of 20 mmol l(-1) phosphate buffer at pH 2.5, containing 25 mmol l(-1) sodium dodecyl sulfate and 10% methanol. Limits of detection (LODs) in the order of 2-46 microg l(-1) for the pesticides under investigation were obtained solely using the on-line strategies. Enrichment factors of 3-18-fold were obtained. These factors were computed as the improvement of the concentration LODs with respect to the reference condition (injection of 10 s at 2.5 kPa pressure). The proposed methodologies were applied to the analysis of pesticides in complex matrices such as carrot extracts where the detection of 2.5 microg l(-1) was illustrated. By combining off-line solid-phase extraction and the proposed on-line strategies, the detection of pesticides in drinking water at the 0.1 microg l(-1) level was conceived.     

Determination of nonsteroidal anti-inflammatory drugs in biological fluids by automatic on-line integration of solid-phase extraction and capillary electrophoresis

A new, automatic method for the clean-up, preconcentration, separation, and quantitation of nonsteroidal anti-inflammatory drugs (NSAIDs) in biological samples (human urine and serum) using solid-phase extraction coupled on-line to capillary electrophoresis is proposed. Automatic pretreatment is carried out by using a continuous flow system operating simultaneously with the capillary electrophoresis equipment, to which it is linked via a laboratory-made mechanical arm. This integrated system is controlled by an electronic interface governed via a program developed in GWBasic. Capillary electrophoresis is conducted by using a separation buffer consisting of 20 mM NaHPO4, 20 mM beta-cyclodextrin and 50 mM SDS at pH 9.0, an applied potential of 20 kV and a temperature of 20 degrees C. The analysis time is 10 min and the detection limits were between 0.88 and 1.71 microg mL(-1). Automatic clean-up and preconcentration is accomplished by using a C-18 minicolumn and 75% methanol as eluent. The limit of detection of NSAIDs can be up to 400-fold improved when using sample clean-up. The extraction efficiency for these compounds is between 71.1 and 109.7 microg mL(-1) (RSD 2.0-7.7%) for urine samples and from 77.2 to 107.1 microg mL(-1) (RSD 3.5-7.1%) for serum samples.     

Determination of chlorophenoxy acid herbicides by capillary electrophoresis with integrated potential gradient detection

This report describes separation and detection of chlorophenoxy acid herbicides spiked in drinking water by the technique combining solid-phase extraction, field-amplified sample stacking, capillary electrophoresis, and potential gradient detection. The herbicide solution (400 mL) was concentrated to 0.1 mL by the solid-phase extraction procedure. The buffer containing 3 mM ammonia and 0.3 mM hydroxypropyl-beta-cyclodextrin was adjusted to pH 9.0 with ammonia. The sample solution was injected into the capillary to 30% of the whole length, and -9 kV and 9 kV were employed for field-amplified sample stacking and separation, respectively. The herbicides were baseline separated and the detection limits with the above combined techniques were in the range of 1-4 x 10(-2) ng/mL.     

Fractionation and separation of human salivary proteins by pH-gradient ion exchange and reversed phase chromatography coupled to mass spectrometry

In the present work, a 2-D capillary liquid chromatography method for fractionation and separation of human salivary proteins is demonstrated. Fractionation of proteins according to their pI values was performed in the 1-D employing a strong anion exchange (SAX) column subjected to a wide-range descending pH gradient. Polystyrene-divinylbenzene (PS-DVB) RP columns were used for focusing and subsequent separation of the proteins in the 2-D. The SAX column was presaturated with a high pH buffer (A) consisting of 10 mM amine buffering species, pH 9.0, and elution was performed with a low pH elution buffer (B) having the same buffer composition and concentration as buffer A, but pH 3.5. Isoelectric point fractions eluting from the 1-D column were trapped on PS-DVB trap columns prior to back-flushed elution onto the PS-DVB analytical column for separation of the proteins. The 1-D fraction eluting at pH 9.0-8.7 was chosen for further analysis. After separation on the RP analytical column, nine RP protein fractions were collected and tryptic digested for subsequent analyses by MALDI TOF MS and column switching capillary LC coupled to ESI TOF MS and ESI QTOF MS. Eight proteins and two peptides were identified in the pH 9.0-8.7 fraction using peptide mass fingerprinting and uninterpreted MS/MS data.     

Determination of haloacetic acids in aqueous environments by solid-phase extraction followed by ion-pair liquid chromatography-electrospray ionization mass spectrometric detection.

Haloacetic acids (HAAs) were determined in different water samples by a new, fast and simple analysis method based on enrichment of 50-ml water samples at pH 1.8 by solid-phase extraction (SPE) followed by liquid chromatography (LC) separation and electrospray ionization mass spectrometric detection in the negative ionization mode. Deprotonated (M-H)-haloacetates and decarboxylated (M-COOH)- ions were detected. Different polymeric SPE sorbents were tested, and LiChrolut EN was found to be the best material for the extraction. Complete LC separation of all compounds could only be achieved by ion-pair chromatography using triethylamine as volatile ion-pairing reagent. The detection limits were in the low microg/l range. High microg/l concentration levels for the chlorinated and brominated haloacetates were found in drinking water from a drinking water treatment plant in Barcelona, and the corresponding tap water. In swimming pool water samples from Catalonia mg/l levels and in surface river water from Portugal microg/l values were detected. These results confirm other recent reports on the ubiquitous occurrence of HAAs in aqueous environments.     

Determination of triazine herbicides in aqueous samples by dispersive liquid-liquid microextraction with gas chromatography-ion trap mass spectrometry

A simple and rapid new dispersive liquid-liquid microextraction technique (DLLME) coupled with gas chromatography-ion trap mass spectrometric detection (GC-MS) was developed for the extraction and analysis of triazine herbicides from water samples. In this method, a mixture of 12.0 microL chlorobenzene (extraction solvent) and 1.00 mL acetone (disperser solvent) is rapidly injected by syringe into the 5.00 mL water sample containing 4% (w/v) sodium chloride. In this process, triazines in the water sample are extracted into the fine droplets of chlorobenzene. After centrifuging for 5 min at 6000 rpm, the fine droplets of chlorobenzene are sedimented in the bottom of the conical test tube (8.0+/-0.3 microL). The settled phase (2.0 microL) is collected and injected into the GC-MS for separation and determination of triazines. Some important parameters, viz, type of extraction solvent, identity and volume of disperser solvent, extraction time, and salt effect, which affect on DLLME were studied. Under optimum conditions the enrichment factors and extraction recoveries were high and ranged between 151-722 and 24.2-115.6%, respectively. The linear range was wide (0.2-200 microg L(-1)) and the limits of detection were between 0.021 and 0.12 microg L(-1) for most of the analytes. The relative standard deviations (RSDs) for 5.00 microg L(-1) of triazines in water were in the range of 1.36-8.67%. The performance of the method was checked by analysis of river and tap water samples, and the relative recoveries of triazines from river and tap water at a spiking level of 5.0 microg L(-1) were 85.2-114.5% and 87.8-119.4%, respectively. This method was also compared with solid-phase microextraction (SPME) and hollow fiber protected liquid-phase microextraction (HFP-LPME) methods. DLLME is a very simple and rapid method, requiring less than 3 min. It also has high enrichment factors and recoveries for the extraction of triazines from water.     

Determination of regularly distributed plant protectants in raw and drinking waters, using a multiresidue method with cyclodextrin-modified micellar electrokinetic chromatography

Eighteen plant protectant compounds were separated and determined by cyclodextrin-modified micellar electrokinetic chromatography (MEKC) in a multiclass/multiresidue method. The pesticides included are those dispersed in the greatest amounts today over agricultural acreage, and they represent 8 different classes of compounds (azoles, benzoic acids, chloroacetanilides, phenoxy acids, phenylureas, sulfonylureas, thiocarbamates, and triazines) covering a wide range of chemical reactivities and physicochemical properties. A 500 mL sample of tap water is preconcentrated by solid-phase extraction (SPE) with 300 mg combined polystyrene-divinylbenzene and methacrylate macroporous resins. Trapped analytes are eluted collectively with diethyl ether. Concentration and solvent change yield 250 microL of an acetone "concentrate," which is further worked up and concentrated 1:10 to produce the MEKC injection solution containing 10 mmol/L sodium dodecyl sulfate (SDS) surfactant. For MEKC, 2 phosphate/SDS buffer systems were designed, each allowing complete separation of all pesticides in a single run. Sensitivity was enhanced by a self-etched bubble cell and an injection procedure which employs stacking at reversed polarity. The ability of MEKC to determine plant protectants in raw and drinking waters at the 0.1 microgram/L level, as demanded by the guidelines of the European Union, was demonstrated with spiked tap waters. Recoveries were between 75 and 110%, and limits of quantification, evaluated as method detection limits according to guidelines of the U.S. Environmental Protection Agency, ranged between 0.03 and 0.10 microgram/L. The precisions of the relative migration times were all below 0.5%.     

Analysis of triazolopyrimidine herbicides in soils using field-enhanced sample injection-coelectroosmotic capillary electrophoresis combined with solid-phase extraction

In this work, a combined methodology using off-line solid-phase extraction (SPE), on-line field-enhanced sample injection (FESI) and coelectroosmotic capillary electrophoresis with UV detection (CE-UV) is developed for the trace analysis of five triazolopyrimidine sulfonanilide pesticides (i.e., flumetsulam, florasulam, cloransulam-methyl, diclosulam and metosulam). An adequate background electrolyte (BGE) was obtained for the separation of these pesticides using hexadimethrine bromide (HDB) as electroosmotic flow (EOF) modifier. This BGE consisted of 0.00042% HDB, 11 mM formic acid, 16 mM ammonium carbonate and 2.5 mM alpha-CD solution at pH 7.6. The use of this running buffer together with the FESI preconcentration method provided limits of detection (LODs) in the low microg/L range (i.e., between 13.0 and 31.5 microg/L). The optimized FESI-CE-UV method was combined with off-line SPE using C(18) cartridges and applied to the determination of the selected group of pesticides in soil samples. Recovery percentages ranged between 50 and 84% in these samples with LODs between 18 and 34 microg/kg. This work shows the great possibilities of the combined use of SPE-FESI-CE-UV to improve CE sensitivity allowing the achievement of LODs similar to other analytical techniques as GC or HPLC.     

Solid-phase microextraction for the analysis of short-chain chlorinated paraffins in water samples

A novel solid-phase microextraction (SPME) method coupled to gas chromatography with electron capture detection (GC-ECD) was developed as an alternative to liquid-liquid and solid-phase extraction for the analysis of short-chain chlorinated paraffins (SCCPs) in water samples. The extraction efficiency of five different commercially available fibres was evaluated and the 100-microm polydimethylsiloxane coating was the most suitable for the absorption of the SCCPs. Optimisation of several SPME parameters, such as extraction time and temperature, ionic strength and desorption time, was performed. Quality parameters were established using Milli-Q, tap water and river water. Linearity ranged between 0.06 and 6 microg l(-1) for spiked Milli-Q water and between 0.6 and 6 microg l(-1) for natural waters. The precision of the SPME-GC-ECD method for the three aqueous matrices was similar and gave relative standard deviations (RSD) between 12 and 14%. The limit of detection (LOD) was 0.02 microg l(-1) for Milli-Q water and 0.3 microg l(-1) for both tap water and river water. The optimised SPME-GC-ECD method was successfully applied to the determination of SCCPs in river water samples.     

Solid-phase extraction and sample stacking-capillary electrophoresis for the determination of quaternary ammonium herbicides in drinking water

Conditions for the simultaneous determination of paraquat, diquat and difenzoquat by capillary zone electrophoresis were established by combining two preconcentration procedures. Off-line solid-phase extraction was used for the isolation and preconcentration of quats in drinking water. Quats were then analysed by capillary electrophoresis using sample stacking with matrix removal as on-column preconcentration procedure. Two different porous graphitic carbon cartridges were compared. The breakthrough volumes of the three herbicides were calculated and the loading capacity of the sorbents was compared. Recoveries higher than 80% for difenzoquat and around 40% for paraquat and diquat were obtained when a sample volume of 250 ml was percolated. For the stacking-capillary electrophoresis analysis of quats, 50 mM acetic acid-ammonium acetate (pH 4.0), 0.8 mM cetyltrimethylammonium bromide with 5% (v/v) methanol as carrier electrolyte was used. Detection limits, based on a signal-to-noise ratio of 3:1, were lower than 0.3 microg l(-1) for standards in Milli-Q water, and lower than 2.2 microg l(-1) for drinking water samples. Run-to-run and day-to-day precision of the method were established. The two preconcentration procedures used together was successfully applied to the analysis of the three herbicides in spiked drinking water at concentrations below the maximum admissible US Environmental Protection Agency levels.