Determination of sulfonylurea herbicides in soil extracts by solid-phase extraction and capillary zone electrophoresis

Summary Sulfonylurea herbicides in soil extracts were concentrated using off-line solid-phase extraction (SPE), and determined by capillary zone electrophoresis (CZE) and UV detection. The method involves extraction of soils with 0.1 M NaHCO₃ solution and subsequent preconcentration by using C₁₈ cartridges prior to separation of the pesticide using CZE. The results show that a C₁₈ cartridge is suitable for the purification of sulfonylurea herbicides in soil extracts with the recoveries ranging from 65–103%. The separation conditions affecting the resolution and detection sensitivity was systematically investigated. The sulfonylureas were resolved well using 30 mM sodium acetate (NaAc)/acetic acid (HAc)+10% acetonitrile (ACN) buffer at pH 4.80. The calibration plots for the test solutes in the concentration of 0.2–50 mg L⁻¹ were linear with detection limits in the range of 0.05–0.10 mgL⁻¹. The proposed method has been successfully demonstrated for the determination of sulfonylurea herbicides in soil samples.     

Borate complexation-assisted field-enhanced sample injection for on-line preconcentration of cis-diol-containing compounds in capillary electrophoresis

A new on-line preconcentration technique called borate complexation-assisted field-enhanced sample injection (BCA-FESI) was proposed for preconcentrating cis-diol-containing compounds (CDCCs) in capillary electrophoresis (CE). The principle relies on amplification of the difference in the electrophoretic mobilities of CDCC in sample matrix and background electrolyte (BGE) through complexation of CDCC with borate in a sample matrix of basic pH and dissociation of the complex in a BGE of acidic pH. Meanwhile, CDCC and borate ions electro-injected into the capillary are finally in neutral state, which maintains the pre-filled low conductivity zone and thus allows for longer injection time. With catechol as a test compound, the principle and effectiveness of BCA-FESI was verified. As compared to conventional sample injection, BCA-FESI allowed for sensitivity enhancement of 1850-fold. The established method was further evaluated with three catechins, including (−)-epicatechin gallate (ECG), (−)-gallocatechin gallate (GCG), and (−)-epigallocatechin (EGC), in a standard mixture of trace content. The limit of detection (LOD) was found to be 1.4, 3.8, 17.5 nM (S/N = 3) for ECG, GCG, EGC, respectively. Finally, the BCA-FESI method was applied to a real sample of diluted tea beverage, in which the three catechins were detected.     

Determination of oxytetracycline and some impurities in plasma by non-aqueous capillary electrophoresis using solid-phase extraction

Summary The potential of a non-aqueous, capillary electrophoresis (NACE) system for separating oxytetracycline from three of its impurities—tetracycline, 4-epioxytetracycline and 4-epitetracycline—using UV detection has been studied. The running buffer was: 25 mM sodium acetate, 1 mM EDTA, methanesulfonic acid, pH 4, dissolved in MeOH-ACN (50∶50,v/v). The method was also used to determine these compounds in pig plasma. A solid-phase extraction (SPE) procedure as a clean-up step has also developed. For this we tested Sep Pak C₁₈, LiChrolut EN and OASIS cartridges. OASIS cartridges were best. Recoveries were 90–100% for all compounds except EOTC which had a recovery of 74%.     

Solvent-bar microextraction of herbicides combined with non-aqueous field-amplified sample injection capillary electrophoresis

Solvent-bar microextraction (SBME) based on two-phase (water-to-organic) extraction was for the first time used as the sample pretreatment method for the non-aqueous capillary electrophoresis (NACE) of herbicides of environmental concern. Due to the compatibility of the extractant organic solvent and the NACE separation system, the extract could be introduced directly to the CE system after SBME. Through investigations of the effect of sample pH, extraction time, agitation speed and salt addition on extraction efficiency, the most suitable extraction conditions were determined: sample solution at a pH of 1, without added salt, and stirring at 700 revolutions per minute for 30 min. SBME as applied here was also compared with single-drop microextraction and hollow fiber-protected liquid-phase microextraction. SBME showed the highest extraction efficiency. In addition, field-amplified sample injection with pre-introduced organic solvent plug removal using the electroosmotic flow as a pump (FAEP) was used to enhance the sensitivity further in NACE. Based on studies of the effect of different organic solvents, different lengths of the organic plugs and different volumes of sample injection on stacking efficiency under the most suitable separation conditions, methanol was found to be the most efficient solvent for on-line preconcentration. Combined with SBME, FAEP-NACE achieved limits of detection of between 0.08 ng/mL and 0.14 ng/mL for the studied analytes. This preconcentration approach for NACE was demonstrated to be amenable to aqueous environmental samples by applying it to spiked river water.     

Non-aqueous capillary electrophoresis of biological samples after at-line solid-phase extraction

Summary Solid-phase extraction (SPE) was coupled at-line to capillary electrophoresis (CE) for the determination of a series of basic test compounds (i. e. tricyclic antidepressants). The analysis was performed using a non-aqueous CE buffer, which resulted in baseline separation of all test compounds. This is in marked contrast with CE using aqueous buffers where hardly any separation was obtained either with or without micelles. The SPE procedure was used to remove simultaneously most of the water from the sample, because no direct analysis of aqueous samples is possible when a non-aqueous CE buffer is used. With the present method the antidepressants can be determined in both urine and serum. Analyte detectability is increased up to 10-fold due to trace enrichment during the extraction process; the limits of detection (LODs; UV 214 nm) are 30–300 ng mL⁻¹ in urine and 300–1000 ng mL⁻¹ in serum. TheRSD values (n=5) of the within-day and between-day precision are below 9% and 11% respectively. Therefore, the present procedure can be used for drug monitoring.     

Comparison of off- and in-line solid-phase extraction for enhancing sensitivity in capillary electrophoresis using ochratoxin as a model compound

This paper proposes and compares two approaches based on off- and in-line solid-phase extraction (SPE), intended to enhance sensitivity in capillary electrophoresis with ultraviolet detection (CE-UV) using as a model the determination of ochratoxin A (OA) in river water samples. In the off-line SPE mode, the reversed-phase sorbent (octadecilsylane, C₁₈) selectively retains the target analyte (OA) and allows large volumes of the sample (70 mL) to be introduced and subsequently eluted in a small volume (0.1 mL) of an appropriate solution. In the in-line SPE mode, a custom-made microcartridge is inserted near the inlet of the capillary, which is filled with the same C₁₈ sorbent. This solid phase selectively retains OA present in a sample volume as low as approximately 640 μL for subsequent elution with ca. 135 nL of an appropriate eluent. The limit of detection (LOD) obtained with the in-line SPE method was 1 ng L^-1, which is 3 orders of magnitude lower than that obtained with CE-UV and roughly 1 order lower than that provided by the off-line SPE-CE-UV method.     

Isoelectric focusing sample injection for capillary electrophoresis of proteins

Carrier ampholyte-free isoelectric focusing (IEF) sample injection (concentration) for capillary electrophoresis (CE) is realized in a single capillary. A short section of porous capillary wall was made near the injection end of a capillary by HF etching. In the etching process, an electric voltage was applied across the etching capillary wall and electric current was monitored. When an electric current through the etching capillary was observed, the capillary wall became porous. The etched part was fixed in a vial, where NaOH solution with a certain concentration was added during the sample injection. The whole capillary was filled with pH 3.0 running buffer. The inlet end vial was filled with protein sample dissolved in the running buffer. An electric voltage was applied across the inlet end vial and etched porous wall. A neutralization reaction occurs at the boundary (interface) of the fronts of H+ and OH-. A pH step or sharp pH gradient exists across the boundary. When positive protein ions electromigrate to the boundary from the sample vial, they are isoelectricelly focused at points corresponding to their pH. After a certain period of concentration, a high voltage is applied across the whole capillary and a conventional CE is followed. An over 100-fold concentration factor has been easily obtained for three model proteins (bovine serum albumin, lysozyme, ribonuclease A). Furthermore, the IEF sample concentration and its dynamics have been visually observed with the whole-column imaging technique. Its merits and remaining problem have been discussed, too.     

Analysis of urinary neurotransmitters by capillary electrophoresis: sensitivity enhancement using field-amplified sample injection and molecular imprinted polymer solid phase extraction

Capillary electrophoresis (CE) has been investigated for the analysis of some neurotransmitters, dopamine (DA), 3-methoxytyramine (3-MT) and serotonin (5-hydroxytryptamine, 5-HT) at nanomolar concentrations in urine. Field-amplified sample injection (FASI) has been used to improve the sensitivity through the online pre-concentration samples. The cationic analytes were stacked at the capillary inlet between a zone of low conductivity - sample and pre-injection plug - and a zone of high conductivity - running buffer. Several FASI parameters have been optimized (ionic strength of the running buffer, concentration of the sample protonation agent, composition of the sample solvent and nature of the pre-injection plug). Best results were obtained using H₃PO₄–LiOH (pH 4, ionic strength of 80 mmol L⁻¹) as running buffer, 100 μmol L⁻¹ of H₃PO₄ in methanol–water 90/10 (v/v) as sample solvent and 100 μmol L⁻¹ of H₃PO₄ in water for the pre-injection plug.In these conditions, the linearity was verified in the 50–300 nmol L⁻¹ concentration range for DA, 3-MT and 5-HT with a determination coefficient (r²) higher than 0.99. The limits of quantification (10 nmol L⁻¹ for DA and 3-MT, 5.9 nmol L⁻¹ for 5-HT) were 500 times lower than those obtained with hydrodynamic injection. However, if this method is applied to the analysis of neurotransmitters in urine, the presence of salts in the matrix greatly reduces the sensitivity of the FASI/CE–UV method.Therefore, a solid phase extraction (SPE) on a dedicated imprinted polymer (MIP) was developed to extract specific neurotransmitters, catecholamines, metanephrines and indolamines, from urine. Matrix salts were thus discarded after sample extraction on AFFINIMIP™ Catecholamine & Metanephrine (100 mg) cartridge.Therefore, lower limits of quantification were determined in artificial urine (46 nmol L⁻¹ for DA, 11 nmol L⁻¹ for 3-MT and 6 nmol L⁻¹ for 5-HT).The application of this protocol MIP-SPE/FASI–CE–UV analysis of neurotransmitters in human urine gave rise to electropherograms with a very good base line and signal to noise ratios above 15.     

Determination of cimetidine in human plasma by use of coupled-flow injection, solid-phase extraction, and capillary zone electrophoresis

Summary An on-line flow injection-solid-phase extraction-capillary zone electrophoresis (FI-SPE-CZE) method has been developed for determination of cimetidine in human plasma. Sodium dodecylsulfate (SDS) was used as dynamic chemical modifier for elimination of capillary contamination by biological macromolecules. FI on-line preconcentration and cleaning of the analyte by means of a C₁₈ microcolumn was performed automatically and CZE separation was performed consecutively without interruption of the applied voltage and between-run-washing of the capillary. A detection limit of 8 μgL⁻¹ (3×σ) was achieved at a sample throughput of 12h⁻¹. The approach was successfully used for a pharmacokinetic study of cimetidine.     

Combination of cationic surfactant-assisted solid-phase extraction with field-amplified sample stacking for highly sensitive analysis of chlorinated acid herbicides by capillary zone electrophoresis

This report describes a novel online field-amplified sample stacking (FASS) procedure to analyze 16 chlorinated acid herbicides. By using a poly(vinyl alcohol) (PVA)-coated capillary to reduce electroosmotic flow and introducing a methanol-water plug before sample loading, the sample injection time could be very long without loss of sample and separation efficiency. Under the optimized condition, the FASS procedure could provide great sensitivity enhancement (5000-10 000-fold) and satisfactory reproducibility (relative standard deviations of migration times less than 2.4%, relative standard deviations of peak areas less than 8.0%). Combined with cationic surfactant-assisted solid-phase extraction (CSA-SPE), the limit of detection of the herbicides ranged from 0.269 to 20.3 ppt, which are two orders lower than those of the US Environmental Protection Agency standard method 515.1. The CSA-SPE-FASS-CE method was successfully applied to the analysis of local pond water.     

固相萃取-毛细管气相色谱法测定中草药及其土壤中多种有机氯农药残留量

建立了中草药及其土壤中多种有机氯农药残留量的固相萃取-毛细管气相色谱(SPE-CGC)分析方法,并对7种中草药及其土壤中多种有机氯农药残留量的相关性进行了初步研究。样品以正己烷-丙酮用超声波提取,Florisil(1 g)固相萃取小柱快速净化提取物。采用SPB-5弹性石英毛细管柱分离样品,GC-ECD检测7种中草药及其土壤中的13种有机氯农药的残留量。方法的线性范围为1.26×10-10~2.24×10-7g/mL;检出限为6.4×10-11~6.1×10-10g/mL;加样平均回收率为87.3%~104.4%(RSD为1.1%~7.0%)。     

Separation of antidepressants by capillary electrophoresis with in-line solid-phase extraction using a novel monolithic adsorbent

The separation of three selective serotonin reuptake inhibitors (SSRIs) by capillary electrophoresis (CE) with fully integrated solid-phase extraction (SPE) is described. Polymeric monolithic SPE modules were prepared in situ within a fused silica capillary from either butyl methacrylate-co-ethylene dimethacrylate or 3-sulfopropyl methacrylate-co-butyl methacrylate-co-ethylene dimethacrylate. Using a 1 cm SPE module placed at the inlet of the capillary, a mixture of sertraline, fluoxetine and fluvoxamine was extracted from aqueous solution by applying a simple pressure rinse. Under pressure-driven conditions, efficient elution was possible from both SPE materials investigated using 50 mM phosphate buffer, pH 3.5 in acetonitrile (20/80, v/v). Two different strategies were investigated for the efficient elution and subsequent CE separation. Injection of an aqueous sample plug directly into the non-aqueous elution/separation buffer was found to be unsuitable with poor elution profiles observed in the electrodriven mode. Alternatively, a sample plug equivalent to several capillary volumes could be injected by pressure followed by filling the capillary with the non-aqueous elution/separation buffer from the outlet end using a combination of pressure and electrodriven flow. Using a neutral monolith, efficient elution/separation was not possible due to an unstable electroosmotic flow (EOF), however, by adding the ionisable monomer, 3-sulfopropyl methacrylate to the SPE module to increase and stabilise the EOF, it was possible to achieve efficient elution from the SPE module, followed by baseline separation by CE using a 200 mM acetate buffer, pH 3.5 in acetonitrile (10/90, v/v). With enrichment factors of over 500 achieved for each of the analytes this demonstrates the potential of in-line SPE-CE for the sensitive analysis of these drugs.     

Application of extraction disks in dissolution tests of clenbuterol and levothyroxine tablets by capillary electrophoresis

Sample preparation procedures using octadecyl (C₁₈) extraction disks were developed to obtain accurate and reproducible results for determinations of clenbuterol (20 μg per dose) and levothyroxine (100 μg per dose) in dissolution media of solid oral dosage forms. Preconcentration of samples allowed final concentrations of 1.1 μg/ml of clenbuterol and 4.0 μg/ml of levothyroxine to be reached prior to CE analysis. The results obtained by CE were in good agreement with those of HPLC. The precision of the migration time, peak area, peak height and accuracy were determined in both intea-day (n = 6) and inter-day (n =18) assays. Linearity was demonstrated over the ranges 0.5–80.0 μg/ml of clenbuterol and 1.0–30.0 μg/ml of levothyroxine. The mean recoveries were higher than 94.0%, ranging from 50 to 125% levels with respect to dose potencies. The proposed methodology may be generally applied to determine drugs at ng/ml concentrations.     

Determination of quinolones in milk samples using a combination of magnetic solid-phase extraction and capillary electrophoresis

A magnetic solid-phase extraction (MSPE) method combined with capillary electrophoresis for the simultaneous determination of seven quinolones (QNs) (danofloxacin, ciprofloxacin, marbofloxacin, enrofloxacin, difloxacin, oxolinic acid, and flumequine), using (S)-(+)-6-methoxy-α-methyl-2-naphthaleneacetic acid as internal standard, in milk samples was developed. The variables involved in the preconcentration magnetic procedure were: the composition of the magnetic support composition, the sample pH, and the weight of magnetic adsorbent used. The variables were optimized using a simplex-lattice design. Different magnetite covered with octyl-phenyl silica adsorbents were synthesized by varying the molar ratio of phenyltrimethylsilane and octyltrimethoxysilane; the solids were evaluated for QN preconcentration. Under optimal conditions, a linear range was obtained from 27 to 1000 μg L(-1) with limits of detection ranging from 9 to 12 μg L(-1) for the seven QNs. The absolute recoveries of the seven QNs at three different spiked levels (40, 150, and 400 μg L(-1) ) ranged from 74% to 98% with a relative standard deviation less than 10% in all cases. The proposed method was applied to analyze 20 whole milk samples of different brands. All samples were positive for the presence of QN residues; in some cases, extract dilution was required. The concentrations found are in the range from 31.1 to 5047.3 μg L(-1) . Marbofloxacin was the most frequently found. The method proposed offers advantages in terms of simplicity, sensitivity, efficiency, cost, and analysis time making it an alternative for the analysis of QNs in whole milk samples.     

Field‐amplified sample injection combined with capillary electrophoresis for the simultaneous determination of five chlorophenols in water samples

A sensitive method of CZE‐ultraviolet (UV) detection based on the on‐line preconcentration strategy of field‐amplified sample injection (FASI) was developed for the simultaneous determination of five kinds of chlorophenols (CPs) namely 4‐chlorophenol (4‐CP), 2‐chlorophenol (2‐CP), 2,4‐dichlorophenol (2,4‐DCP), 2,4,6‐trichlorophenol (2,4,6‐TCP), and 2,6‐dichlorophenol (2,6‐DCP) in water samples. Several parameters affecting CZE and FASI conditions were systematically investigated. Under the optimal conditions, sensitivity enhancement factors for 4‐CP, 2‐CP, 2,4‐DCP, 2,4,6‐TCP, and 2,6‐DCP were 9, 27, 35, 43, and 43 folds, respectively, compared with the direct CZE, and the baseline separation was achieved within 5 min. Then, the developed FASI‐CZE‐UV method was applied to tap and lake water samples for the five CPs determination. The LODs (S/N = 3) were 0.0018–0.019 µg/mL and 0.0089–0.029 µg/mL in tap water and lake water, respectively. The values of LOQs in tap water (0.006–0.0074 µg/mL) were much lower than the maximum permissible concentrations of 2,4,6‐TCP, 2,4‐DCP, and 2‐CP in drinking water stipulated by World Health Organization (WHO) namely 0.3, 0.04, and 0.01 µg/mL, respectively, and thereby the method was suitable to detect the CPs according to WHO guidelines. Furthermore, the method attained high recoveries in the range of 83.0–119.0% at three spiking levels of five CPs in the two types of water samples, with relative standard deviations of 0.37–8.58%. The developed method was proved to be a simple, sensitive, highly automated, and efficient alternative to CPs determination in real water samples.     

Solid-phase microextraction coupled with capillary electrophoresis to determine ephedrine derivatives in water and urine using a sol-gel derived butyl methacrylate/silicone fiber

A sensitive method for determination of ephedrine derivatives using headspace solid-phase microextraction (SPME) with a novel fiber followed by capillary electrophoresis has been developed. The co-poly(butyl methacrylate/hydroxy-terminated silicone oil) (BMA/OH-TSO) was used as stationary phases with the aid of γ-methacryloxypropyltrimethoxysilane (KH-570) as bridge in SPME using sol-gel-coating method and cross-linking technology. It has high extraction efficiency for ephedrine derivatives in comparison with commercial poly(dimethylsiloxane) and poly(acrylate)-coated fiber. The coating exhibits good thermal and solvent stability as well as long lifetime. A simple and flexible device for desorption of analytes after headspace SPME was constructed. The effect of various experimental parameters for SPME (temperature, time, pH, ionic strength, desorption solvent, etc.) were discussed. Field amplified sample injection (FASI) was applied for on-line sample concentration and a sensitivity enhancement of two orders of magnitude was achieved. Linear ranges were found to be 20-5000 ng/ml. The detection limits for (1R,2S)-ephedrine, (1R,2R)-pseudoephedrine and (1S,2S)-pseudoephedrine were 3, 5 and 5 ng/ml, respectively. Relative standard deviation (n = 6) was found to be 4.96-7.57%. The method was successfully applied to the analysis of ephedrine derivatives in human urine.     

Determination of trace bisphenol A in complex samples using selective molecularly imprinted solid-phase extraction coupled with capillary electrophoresis

The highly selective, fast and effective sample pretreatment technique molecularly imprinted solid-phase extraction (MISPE) can overcome the low sensitivity of the highly efficient capillary electrophoresis-UV method (CE-UV). In this work, narrowly dispersible bisphenol A (BPA)-imprinted polymeric microspheres with a high capacity factor of k′ = 6.8 and an imprinted factor of I = 6.53 were investigated as selective solid-phase extraction (SPE) sorbents for use in extraction of BPA from different sample matrices (tap water, wastewater, Yangtze River water, soil from the Yangtze River, shrimp and human urine). Washing and eluting protocols of MISPE were optimized. Under optimal conditions, recoveries of MISPE were investigated. Recoveries were basically constant and the relative standard deviation (RSD) was lower than 5.8% when loading volumes changed from 1 to 50 mL. Recoveries ranged from 71.20% to 86.23% for different sample matrices. Compared with C18 SPE, MISPE had higher selectivity and recovery for BPA. BPA was determined with good accuracy and precision in different complex samples using CE-UV coupled with MISPE. Spiked recoveries ranged from 95.20% to 105.40%, and the RSD was less than 7.2%. Because a large loading volume was achieved, the enrichment efficiency of pretreatment and the sensitivity of this method were improved. The limits of detection of this MISPE-CE-UV method for BPA in tap water, wastewater, Yangtze River water, soil from the Yangtze River, shrimp and human urine were 3.0 μg L^− 1, 5.4 μg L^− 1, 6.9 μg L^− 1, 2.1 μg L^− 1, 1.8 μg L^− 1 and 84 μg L^− 1, respectively.     

Improved extraction and clean-up of imidazolinone herbicides from soil solutions using different solid-phase sorbents

Extraction and quantification of herbicide residues from soil are important in understanding the behaviour of persistent herbicides. This research investigated extraction and clean-up methods for imidazolinone herbicides from soil and soil amended with organic material. A series of solvent mixes, pH conditions and sorbents was tested. Across three imidazolinone herbicides: imazapyr, imazethapyr and imazaquin, 0.5 M NaOH extraction gave greater than 90% recovery from soil samples; however, 0.5 M NaOH:MeOH (80:20) resulted in higher recovery for imazaquin, but not for the other two herbicides. Of the sorbents tested, the use of chromatographic mode sequencing using C₁₈ and SCX sorbents provided consistent high (>85%) recovery of all three herbicides from soil and separation of the herbicides from other soil components by high performance liquid chromatography (HPLC). These two methods will allow high recovery of these imidazolinone herbicides from soil and have the ability to detect these herbicides without interference from other soil components.     

Oxidized multi-walled carbon nanotubes for the dispersive solid-phase extraction of quinolone antibiotics from water samples using capillary electrophoresis and large volume sample stacking with polarity switching

In this work, a new method for the determination of eleven quinolone antibiotics (moxifloxacin, lomefloxacin, danofloxacin, ciprofloxacin, levofloxacin, marbofloxacin, enrofloxacin, difloxacin, pefloxacin, oxolinic acid and flumequine) in different water samples using dispersive solid-phase extraction (dSPE) and capillary zone electrophoresis with diode-array detection was developed. Oxidized multi-walled carbon nanotubes (o-MWCNTs) were used for the first time as stationary phases for the off-line preconcentration by dSPE of the antibiotics. A 65 mM phosphate buffer at pH 8.5 was found adequate for analyte separation while large volume sample stacking with polarity switching of the analytes dissolved in water containing 10% (v/v) of acetonitrile was carried out in order to improve the sensitivity. dSPE parameters, such as sample volume and pH, o-MWCNT amount, volume and type of eluent in dSPE were optimized. Application of the developed method to the analysis of spiked Milli-Q, mineral, tap, and wastewater samples resulted in good recoveries values ranging from 62.3 to 116% with relative standard deviation values lower than 7.7% in all cases. Limits of detection were in the range of 28-94 ng/L. The proposed method is very fast, simple, repeatable, accurate and highly selective.     

Gentamicin assay in human serum by solid-phase extraction and capillary electrophoresis

We describe the development of a capillary electrophoresis method for the determination of gentamicin C1, C1a, C2a, and C2 components in human serum. Using a weak cation-exchanger with 20 mM phosphate buffer, pH 7.4, 200 mM borate buffer, pH 9.0, and ammonia/methanol, solid-phase extraction (SPE) of gentamicin components from the human sera was performed. The extract was derivatized with 1,2-phthalic dicarboxaldehyde/mercaptoacetic acid reagent. The derivatives were separated with a background electrolyte comprising 60 mM 2-(N-cyclohexylamino)ethanesulfonic acid (CHES) buffer at pH 9.5 containing 31.6% m/v methanol, and quantified with UV-light absorption detection at 230 nm. The identity of the gentamicin components was confirmed by mass spectrometry. The SPE recovery of the gentamicin ranged from 78% to 93%. The calibration curves were linear from the concentration limit of quantitation (LOQ) to 30 mg/L for the gentamicin mixture. The LOQ for gentamicin C1 was 0.33 mg/L, for C2a 0.23 mg/L, C2 0.25 mg/L, C1a 0.27 mg/L and the concentration limit of detection (LOD) for C1 was 0.15 mg/L, C2a 0.11 mg/L, C2 0.12 mg/L, C1a 0.13 mg/L. Intra-assay relative standard deviation (RSD) values were for C1 (5%), C1a (7%), C2 (6.5%) and C2a (9%); inter-assay RSD values were for C1 (11%), C1a (13.3%), C2 (15%) and C2a (14%). The Pearson's correlation between capillary electrophoresis and immunoassay revealed a linear relationship between these two techniques with r = 0.9. This method for determination of gentamicin C1, C1a, C2a, and C2 in human serum can thus be used in the entire therapeutic concentrations range of gentamicin.