Gel-surface enhanced fluorescence sensing system coupled to a continuous-flow assembly for simultaneous monitoring of benomyl and carbendazim

A novel, versatile and sensitive continuous-flow on-line solid phase fluorescence based system is proposed for the simultaneous determination of benomyl and carbendazim. The continuous-flow system is based on the on-line preconcentration and resolution of the pesticides on a solid sensing zone, followed by the sequential measure of their native fluorescence, monitored at 235/306 and 293/398 nm (λ_exc/λ_em for carbendazim and benomyl, respectively), and later desorption of these analytes (from the flow-through cell filled with C₁₈ silica gel) using aqueous methanol mixtures as carrier and eluent solutions.A double discrimination is used for the simultaneous monitoring of these analytes: (1) the usage of two pair of excitation/emision wavelengths, performed by the use of a multiwavelength fluorescence detection mode and (2) a temporary sequentiation in the arrival of the analytes to the sensing system by on-line separation due to the different kinetics showed by the analytes in the sorption-desorption process performed just in the solid support placed in the flow-through cell. Carbendazim is determined the first, because it shows a weaker retention in the C₁₈ bonded phase silica beads, while benomyl is strongly fixed. Then, benomyl is conveniently eluted from the flow-through sensing zone and its native fluorescence signal is measured (at 398 nm). The sensor was calibrated for two different injection volumes: 400 and 2000 μl. Using a 2000 μl sample volume, the analytical signal showed linearity in the range 0.050-1.0 and 0.020-0.50 μg ml⁻¹ with detection limits of 3.0 and 7.5 ng ml⁻¹ for carbendazim and benomyl, respectively, and R.S.D. values smaller than 2% for both analytes. A recovery study was performed on four different spiked environmental water samples at concentration levels from 0.05 to 0.35 μg ml⁻¹. The recovery percentage ranged from 97 to 104%, and from 98 to 104%, for benomyl and carbendazim, respectively.     

Determination of clarithromycin in rat plasma by HPLC-UV method with pre-column derivatization

A novel high-performance liquid chromatographic (HPLC) method using pre-column derivatization and UV detection at 275 nm for the determination of clarithromycin in rat plasma has been validated. Clarithromycin was extracted from plasma sample spiked with internal standard (erythromycin) under alkaline condition with ethyl ether and derivatizated with trimethylbromosilane. The analyses were run on a C₁₈ column, maintained at 40 °C during elution, using a mobile phase comprised of potassium dihydrogen phosphate (50 mM, pH 6.8, contained 0.7% triethylamine), acetonitrile, and methanol (30:45:25, v/v/v). The standard calibration curve for clarithromycin was linear (r² = 0.9998) over the concentration range of 0.1-10 μg ml⁻¹ in rat plasma. The limit of detection (LOD) and limit of quantitation (LOQ) was 30 ng ml⁻¹ and 0.1 μg ml⁻¹ respectively. The intra- and inter-day assay variability range was 2.6-7.4% and 3.3-8.5%, respectively. This method has been successfully applied to a pharmacokinetic study of clarithromycin in rats.     

Simultaneous determination of sulfamonomethoxine, sulfadimethoxine, and their hydroxy/N(4)-acetyl metabolites with gradient liquid chromatography in chicken plasma, tissues, and eggs

A simultaneous determination of sulfamonomethoxine, sulfadimethoxine, and their hydroxy/N⁴-acetyl metabolites in chicken plasma, muscle, liver, and eggs using gradient high-performance liquid chromatography (HPLC) with a photo-diode array detector is developed. All the compounds are extracted by a handheld ultrasonic homogenizer with ethanol followed by centrifugation. The separation is performed by a reversed-phase C4 column with a gradient elution (ethanol:1% (v/v) acetic acid, v/v; 10:90 → 20:80). Average recoveries from samples spiked at 0.1-1.0 μg g⁻¹ or μg ml⁻¹ for each drug were >90% with relative standard deviations within 4%. The limits of quantitation were <30 ng g⁻¹ or ng ml⁻¹.     

Determination of carbendazim, fuberidazole and thiabendazole by three-dimensional excitation-emission matrix fluorescence and parallel factor analysis

We simultaneously determined carbendazim, fuberidazole and thiabendazole by excitation-emission matrix (EEM) fluorescence in combination with parallel factor analysis (PARAFAC). Three-way deconvolution provided the pure analyte spectra from which we estimated the selectivity and sensitivity of the pesticides, and the relative concentration in the mixtures from which we established a linear calibration. Special attention was given calculating such figures of merit as precision, sensitivity and limit of detection (LOD), derived from the univariate calibration curve. The method, which had a relative precision of around 2-3% for the three pesticides, provided limits of detection of 20 ng ml⁻¹ for carbendazim, 4.7 ng ml⁻¹ for thiabendazole and 0.15 ng ml⁻¹ for fuberidazole. The accuracy of the method, evaluated through the root mean square error of prediction (RMSEP), was 27.5, 1.4, and 0.03 ng ml⁻¹, respectively, for each of the pesticides.     

Implementation of multicommutation principle with flow-through multioptosensors

For the first time, the combination of multicommutation concept with flow-through multioptosensors is exploited and a biparameter multicommutated UV sensor is developed. A very easy-operated and automated sensing device is proposed here and demonstrated to be useful in the routine analysis by applying it to the determination of two widely used pharmaceuticals chosen as model analytes: salicylamide and caffeine. The particulated solid phase used performs two functions: (a) states selectivity conditions by itself in the flow cell and (b) provides appropriate separation of the analytes in the on-line precolumn, thus making possible the sequential arrival of the analytes to the detection zone. Both carrier and eluting solutions chosen allow (with a very simple solenoid valve configuration and operation) the transitory signals to be developed in a completely automated device. Drastic reduction of carrier and eluting solutions used are achieved as compared to typically FIA optosensor configuration. The sensor responds linearly in the range 2-30 and 1-14 μg ml⁻¹ for SLC and CF, with detection limits of 0.33 and 0.15 μg ml⁻¹, respectively. Solution savings between 50 and 85% were obtained when comparing to typical flow injection analysis (FIA).     

A environmentally friendly reversed-phase liquid chromatography method for phthalates determination in nail cosmetics

A simple and rapid analytical method was developed for the determination of phthalates, usually employed in nail cosmetic products. The method is based on an ultrasonic extraction of the sample with ethanol-water (90:10, v/v) followed by HPLC separation and quantitation. HPLC was carried out using a C18 column and spectrophotometric detection at 254 nm. A linear gradient elution was performed with ethanol-water starting from 50 to 95% ethanol in 30 min. Standard calibration curves were linear for all the analytes over the concentration range 5-200 μg ml⁻¹ with LOD values of about 0.5 μg ml⁻¹.The proposed green analytical method has been successfully applied for the analysis of commercial samples in order to check the presence of phthalates and to determine their concentration.     

Ferric nanoparticle-based resonance light scattering determination of DNA at nanogram levels

A novel assay of DNA has been proposed by using ferric nanoparticles as probes coupled with resonance light scattering (RLS) detection. At pH 7.40, the RLS intensity of ferric nanoparticles can be greatly enhanced by the aggregation of positively charged ferric nanoparticles through electrostatic interaction with negatively charged DNA. The enhanced intensity of RLS at 452 nm is proportional to the concentration of DNA in the range of 0.01-0.8 μg ml⁻¹ for calf thymus and salmon sperm DNA and in the range of 0.005-0.3 μg ml⁻¹ for E. coli K12 genomic DNA. Detection limits are 3.6 ng ml⁻¹ for calf thymus DNA, 4.4 ng ml⁻¹ for salmon sperm DNA, and 1.9 ng ml⁻¹ for E. coli K12 genomic DNA, respectively. Compared with the chromophores previously used in RLS assay, the ferric nanoparticles have offered several advantages in easy preparation, good photostability and high sensitivity without being modified or functionalized.     

A new analytical application of nylon-induced room-temperature phosphorescence: determination of thiabendazole in water samples

This paper discusses the first analytical determination of the widely used fungicide thiabendazole by nylon-induced phosphorimetry. Nylon was investigated as a novel solid-matrix for inducing room-temperature phosphorescence of thiabendazole, which was enhanced under the effect of external heavy-atom salts. Among the investigated salts, lead(II) acetate was the most effective in yielding a high phosphorescence signal. An additional enhancement of the phosphorescence emission was attained when the measurements were carried out under a nitrogen atmosphere. There was only a moderate increase in the presence of cyclodextrins. The room-temperature phosphorescence lifetimes of the adsorbed thiabendazole were measured under different working conditions and, in all cases, two decaying components were detected. On the basis of the obtained results, a very simple and sensitive phosphorimetric method for the determination of thiabendazole was established. The analytical figures of merit obtained under the best experimental conditions were: linear calibration range from 0.031 to 0.26 μg ml⁻¹ (the lowest value corresponds to the quantitation limit), relative standard deviation, 2.4% (n = 5) at a level of 0.096 μg ml⁻¹, and limit of detection calculated according to 1995 IUPAC Recommendations equal to 0.010 μg ml⁻¹ (0.03 ng/spot). The potential interference from common agrochemicals was also studied. The feasibility of determining thiabendazole in real samples was successfully evaluated through the analysis of spiked river, tap and mineral water samples.     

Simultaneous determination of 11 drugs belonging to four different groups in human urine samples by reversed-phase high-performance liquid chromatography method

A new, accurate, sensitive and fast reversed-phase high-performance liquid chromatography (RP-HPLC) as an analytical method for the quantitative determination of 11 drugs in human urine was worked out, optimized and validated. The objects of analysis were imipenem (IMP), paracetamol (PAR), dipyrone (DPR), vancomycin (VCM), amikacin (AMK), fluconazole (FZ), cefazolin (CFZ), prednisolone (PRE), dexamethasone (DEX), furosemide (FUR) and ketoprofen (KET) belonging to four different groups (antibiotics, analgesic, demulcent and diuretic). For HPLC analysis, diode array (DAD) and fluorescence (FL) detectors were used. The separation of analyzed compounds was conducted by means of a LiChroCART^® Purospher^® C₁₈e (125 mm × 3 mm, particle size 5 μm) analytical column with LiChroCART^® LiChrospher^® C₁₈ (4 mm × 4 mm, particle size 5 μm) pre-column with gradient elution. Analyzed drugs were determined within 20 min. The mobile phase was comprised of various proportions of methanol, acetonitrile and 0.05% trifluoroacetic acid in water. AMK was separated and determined from human urine using ortho-phthaldialdehyde-3-mercaptopropionic acid (OPA-3-MPA) as a fluorescent reagent by RP-HPLC-FL. The following retention times for drugs IMP, PAR, DPR, VCM, AMK, FZ, CFZ, PRE, DEX, FUR and KET in human urine were found: 4.01 min, 4.86 min, 6.71 min, 8.14 min, 9.46 min, 10.01 min, 10.90 min, 13.34 min, 14.06 min, 16.03 min and 18.98 min, respectively. Excellent linearity was obtained for compounds in the range of concentration: 0.35-42 μg ml⁻¹, 0.5-45 μg ml⁻¹, 4.5-38 μg ml⁻¹, 0.25-25 μg ml⁻¹, 0.5-35 μg ml⁻¹, 0.25-22 μg ml⁻¹, 0.03-52 μg ml⁻¹, 0.15-25 μg ml⁻¹, 0.25-28 μg ml⁻¹, 0.05-18 μg ml⁻¹ and 0.15-35 μg ml⁻¹ for IMP, PAR, DPR, VCM, AMK, FZ, CFZ, PRE, DEX, FUR and KET, respectively. The limits of detection (LOD) and limits of quantification (LOQ) for analyzed drugs were calculated in all cases and recovery studies were also performed. Ten human urine samples obtained from patients treated in hospital have been tested. In analyzed samples, one or more drugs from the 11 examined drugs were detected. The concentrations of examined drugs in urine samples ranged between: 1.5-12 μg ml⁻¹ of PAR, 5.2-11.5 μg ml⁻¹ of DPR, 0.13-9.5 μg ml⁻¹ of CFZ and 0.1-8 μg ml⁻¹ of FUR. This method can be successfully applied to routine determination of all these drugs in human urine samples.     

Determination of proteins with Alizarin Red S by Rayleigh light scattering technique

A new protein determination method by enhanced Rayleigh light scattering (RLS) technique has been developed. In acid condition (pH=3.60), RLS of 1,2-dihydroxyanthraquinone-3-sulfonate (Alizarin Red S) can be greatly enhanced by addition of proteins, resulting in two characteristic peaks, 360 and 505 nm, respectively. The new protein assay is based on the RLS enhancement and spectrum change. The optimum condition for the reaction was investigated. The linear range is 0.20-24.9 μg ml⁻¹ for BSA and 0.20-15.5 μg ml⁻¹ for HSA. The detection limits (S/N=3) are 9.59 ng ml⁻¹ for BSA and 9.51 ng ml⁻¹ for HSA. The results of determination for human serum samples were comparable to those obtained by Bradford method. The binding stoichiometry was determined.     

Simultaneous liquid–liquid microextraction and polypropylene microporous membrane solid-phase extraction of organochlorine pesticides in water,tomato and strawberry samples

A procedure involving the simultaneous performance of liquid–liquid microextraction and polypropylene microporous membrane solid-phase extraction was carried out. The applicability of the proposed procedure was evaluated through extraction of several organochlorine pesticides from river water, tomato and strawberry samples. The parameters affecting the extraction efficiency were optimized by multivariable designs, and the analytical features were estimated. Under optimized conditions, analytes were concentrated onto 1.5 cm long microporous membranes placed directly into the sample containing 15 mL of water with 20 μL of 1-octanol. The best extraction conditions were achieved at 59 °C, with 60 min of extraction time and 2.91 g of sodium chloride. The desorption of the analytes was carried out using 30 μL of a mixture of toluene and hexane in the proportion of 60:40% (v/v) for 10 min. Detection limits in the range of 2.7–20.0 ng L⁻¹, 0.50–1.15 μg kg⁻¹, and 1.53–12.77 μg kg⁻¹ were obtained for river water, strawberry and tomato samples, respectively. Good repeatability was obtained for all three sample types. The results suggest that the proposed procedure represents a very simple and low-cost microextraction alternative rendering adequate limits of quantification for the determination of organochlorine pesticides in environmental and food samples.     

A multicommuted fluorescence-based sensing system for simultaneous determination of Vitamins B2 and B6

A multicommuted flow-through optosensor based on the direct fluorescence measurements of Vitamins B₂ and B₆ using a non-polar sorbent (C₁₈ silica gel) as solid sensing zone (to accomplish the separation and subsequent preconcentration/detection of the target analytes) have been developed. The proposed flow system was controlled by Java-written home-made software and designed using three-way solenoid valves for independent automated manipulation of sample and carrier solutions. The native fluorescence signal was simultaneously monitored at two pairs of excitation/emission wavelengths (450/519 and 294/395 for B₂ and B₆, respectively). The separation of the analytes was performed in the detection flow cell, using the differences in the sorption/elution process on the solid support between the two vitamins, due to their different polarity. Using an optimised sampling time, the analytical signal showed linearity in the range 0.01-0.4 and 0.15-3 μg ml⁻¹ with detection limits of 0.003 and 0.045 μg ml⁻¹ for B₂ and B₆, respectively, obtaining R.S.D. (%) values better than 2% for both analytes. The proposed methodology was applied to different pharmaceutical preparations, obtaining remarkably good results with recoveries ranging from 96 to 107.5%.     

Development of integrated chemiluminescence flow sensor for the determination of adrenaline and isoprenaline

A novel integrated chemiluminescence (CL) flow sensor for the determination of adrenaline and isoprenaline is developed based on the enhancing effect of analytes on CL emission of luminol oxidized by periodate in alkaline solution. The analytical reagents luminol and periodate are immobilized on anion exchange resins, respectively, and packed in a glass tube to construct a reagentless sensor. The proposed sensor allows the determination of adrenaline and isoprenaline over the range from 2.0×10⁻⁸ to 1.0×10⁻⁵ g ml⁻¹ and 2.0×10⁻⁷ to 5.0×10⁻⁵ g ml⁻¹, respectively. The detection limits are 7.0×10⁻⁹ g ml⁻¹ for adrenaline and 5.0×10⁻⁸ g ml⁻¹ for isoprenaline with a relative standard deviation of 1.7% for the 1.0×10⁻⁷ g ml⁻¹ adrenaline (n=11) and 2.1% for 1.0×10⁻⁶ g ml⁻¹ isoprenaline (n=11). The sample throughput was 60 samples h⁻¹. The sensor has been successfully applied to the determination of adrenaline and isoprenaline in pharmaceutical preparations.     

Development of an optical chemical sensor based on 2-(5-bromo-2-pyridylazo)-5-(diethylamino)phenol in Nafion for determination of nickel ion

An optical chemical sensor based on immobilization of 2-(5-bromo-2-pyridylazo)-5-(diethylamino)phenol (Br-PADAP) in Nafion membrane is described. The membranes were cast onto glass substrates and were used for the determination of nickel in aqueous solutions by spectrophotometry. The sensor system is highly transparent, mechanically stable and showed no evidence of reagent leaching. The influence of several parameters such as pH, ligand concentration, and type and concentration of regenerating solution were optimized. The sensor system showed good sensitivity in the range 0.5-20 μg ml⁻¹ with a detection limit of 0.3 μg ml⁻¹ Ni(II). The sensor has been incorporated into a home-made flow-through cell for determination of nickel in flowing streams with improved sensitivity, precision and detection limit. The calibration curve in the flow system was linear in the range 0.1-16 μg ml⁻¹ with a detection limit of 0.07 μg ml⁻¹. The sensor is easily regenerated by dilute nitric acid solution. The proposed method was successfully applied to the determination of nickel content in vegetable oil and chocolate samples and the results were compared with those obtained using atomic absorption spectrometry.     

Direct spectrophotometric determination of calcium in clinical samples with carboxyazo-p-CH3

A highly sensitive and relatively interference-free spectrophotometric method for determination of calcium is described. The method is based on the reaction between calcium ions and carboxyazo-p-CH₃ in aqueous citrate medium of pH 7, to form a blue complex with maximum absorption at 716 nm. The calibration is linear up to 0.12 μg ml⁻¹ calcium with a repeatability (R.S.D.) of 1.0% at a concentration of 0.04 μg ml⁻¹ (n=5). The molar absorptivity of the complex and Sandell’s sensitivity are 3.5×10⁵ l mol⁻¹ cm⁻¹ and 0.11 ng cm⁻², its 10σ limit of quantification and the 3σ limit of detection were found to be 0.3 ng ml⁻¹ and 0.09 ng ml⁻¹ respectively. The influence of reaction variables and the effect of interfering ions are studied; no interference was observed in clinical samples. The proposed method has been applied directly to the determination of calcium in clinical samples without the need for pre-concentration, masking metal ions and digesting samples.     

Comparison of electrospray ionization and atmospheric pressure chemical ionization for multi-residue analysis of biocides,UV-filters and benzothiazoles in aqueous matrices and activated sludge by liquid chromatography–tandem mass spectrometry

This paper describes the development of a multi-residue method for the determination of 36 emerging organic pollutants (26 biocides, 5 UV-filters and 5 benzothiazoles) in raw and treated wastewater, activated sludge and surface water using liquid chromatography–tandem mass spectrometry (LC–MS/MS). The target analytes were enriched from water samples adjusted to pH 6 by solid-phase extraction (SPE) on Oasis HLB 200 mg cartridges and eluted with a mixture of methanol and acetone (60/40, v/v). Extraction of freeze-dried sludge samples was accomplished by pressurized liquid extraction (PLE) using a mixture of methanol and water (50/50, v/v) as extraction solvent followed by SPE. LC–tandem MS detection was compared using electrospray ionization (ESI) and atmospheric pressure chemical ionization (APCI) in positive and negative ionization mode. ESI exhibited strong ion suppression for most target analytes, while APCI was generally less susceptible to ion suppression but partially leading to ion enhancement of up to a factor of 10. In general, matrix effects could be compensated using stable isotope-labeled surrogate standards, indicated by relative recoveries ranging from 70% to 130%. In wastewater, activated sludge and surface water up to 33 analytes were detected. Maximum concentrations up to 5.1 and 3.9 μg L⁻¹ were found in raw wastewater for the water-soluble UV-filters benzophenone-4 (BZP-4) and phenylbenz-imidazole sulfonic acid (PBSA), respectively. For the first time, the anti-dandruff climbazole was detected in raw wastewater and in activated sludge with concentrations as high as 1.4 μg L⁻¹ and 1.2 μg g TSS⁻¹, respectively. Activated sludge is obviously a sink for four benzothiazoles and two isothiazolones, as concentrations were detected in activated sludge between 120 ng g TSS⁻¹ (2-n-octyl-4-isothiazolin-3-one, OIT) to 330 ng g TSS⁻¹ (benzothiazole-2-sulfonic acid, BTSA).     

Synchronous fluorescence determination of protein with functionalized CdS nanoparticles as a fluorescence probe

Nanometer-sized fluorescent particles have been successfully synthesized. A synchronous fluorescence method, with high sensitivity and selectivity, has been developed for rapid determination of protein with functionalized CdS as a fluorescence probe. When Δλ=260 nm, maximum synchronous fluorescence is produced at 274 nm at pH 7.0. Under optimal conditions, the calibration graphs are linear over the range 0.1-3.0 μg ml⁻¹ for bovine serum albumin (BSA), 0.1-11.0 μg ml⁻¹ for γ-globulin (γ-G) and 0.1-1.4 μg ml⁻¹ for human serum albumin (HSA), respectively. Limits of determination were 0.01 μg ml⁻¹ for BSA, 0.019 μg ml⁻¹ for γ-G and 0.021 μg ml⁻¹ for HSA, respectively. The relative standard deviations of seven replicate measurements were 1.8% for 1.0 μg ml⁻¹ BSA, 2.2% for 1.0 μg ml⁻¹ γ-G and 2.3% for 1.0 μg ml⁻¹ HSA.     

A novel flow-through fluorescence optosensor for the determination of thiabendazole

This paper presents the development of a new flow-injection system combined with solid-surface fluorescence detection for the determination of the widely used fungicide thiabendazole. Nylon powder was probed as a novel solid support for building the optosensor. The method is based on the on-line immobilization of thiabendazole onto nylon in a continuous flow system, followed by the measurement of its native fluorescence. Aqueous samples are directly injected in a water carrier, resulting in a very simple and economical method. The analytical figures of merit obtained using 1500 μL of sample and 75% methanol (v/v) as eluting solution were: linear calibration range from 8 to 120 ng mL⁻¹ (the lowest value corresponds to the quantitation limit), relative standard deviation, 0.9% (n = 5) at a level of 64 ng mL⁻¹, limit of detection calculated according to 1995 IUPAC recommendations is to 2.8 ng mL⁻¹, and sampling rate of 14 samples h⁻¹. The potential interference from other agrochemicals, metal ions and common anions, and the viability of determining thiabendazole in real water samples were also evaluated.     

Coupled in-tube and on-fibre solid-phase microextractions for cleanup and preconcentration of organic micropollutants from aqueous samples and analysis by gas chromatography-mass spectrometry

Matrix interference removal is an important step when large volumes of aqueous samples are required to be processed to detect trace levels of analytes. A combination of two sample extraction methods has been used in this work with the aim of cleanup and preconcentration of analytes. For first objective, mild but preferential sorption of a range of analytes has been performed with in-tube solid-phase microextraction (SPME) using polytetrafluoroethylene (PTFE) tubing, and for the second, the eluate from in-tube SPME was subjected to on-fibre SPME using DVB/Caboxen/PDMS (30/50 μm) fibre. Knitting of PTFE tubing created secondary flow pattern that enhanced radial diffusion and retention of organic analytes. Up to 2 mg L⁻¹ of a broad range of substances that are not extracted by PTFE include nitrogen containing aromatic heterocyclic compounds, anilines, phenols and certain organophosphorus pesticides, thus providing a clean extract using this method of sample preparation. The proposed combination of in-tube and on-fibre SPME produced a rectilinear calibration graph over 0.03-150 μg L⁻¹ of a range of analytes using 60 mL of aqueous sample. The overall recovery of analytes was in the range 27-78%. The detection limits were between 6.1 and 21.8 ng L⁻¹. The R.S.D. was in range 5.4-8.2% and 4.2-6.5% in the analysis of respectively 2 and 20 μg L⁻¹ of analytes.     

Determination of benzimidazolic fungicides in fruits and vegetables by supramolecular solvent-based microextraction/liquid chromatography/fluorescence detection

A supramolecular solvent consisting of vesicles, made up of equimolecular amounts of decanoic acid (DeA) and tetrabutylammonium decanoate (Bu₄NDe), dispersed in a continuous aqueous phase, is proposed for the extraction of benzimidazolic fungicides (BFs) from fruits and vegetables. Carbendazim (CB), thiabendazole (TB) and fuberidazole (FB) were extracted in a single step and no clean-up or concentration of extracts was needed. The high extraction efficiency obtained for BFs was a result of the different types of interactions provided by the supramolecular solvent (e.g. hydrophobic and hydrogen bonds) and the high number of solubilisation sites it contains. Besides simple and efficient, the proposed extraction approach was rapid, low-cost, environment friendly and it was implemented using conventional lab equipments. The target analytes were determined in the supramolecular extract by LC/fluorescence detection. They were separated in a Kromasil C₁₈ (5 μm, 150 mm × 4.6 mm) column using isocratic elution [mobile phase: 60:40 (v/v) 50 mM phosphate buffer (pH 4)/methanol] and quantified at 286/320 nm (CB) and 300/350 nm (TB and FB) excitation/emission wavelengths, respectively. Quantitation limits provided by the supramolecular solvent-based microextraction (SUSME)/LC/fluorescence detection proposed method for the determination of CB, TB and FB in fruits and vegetables were 14.0, 1.3 and 0.03 μg kg⁻¹, respectively, values far below the current maximum residue levels (MRLs) established by the European Union, i.e. 100-2000 μg kg⁻¹ for CB, 50-5000 μg kg⁻¹ for TB and 50 μg kg⁻¹ for FB. The precision of the method, expressed as relative standard deviation, for inter-day measurements (n = 13) was 3.3% for CB (50 μg kg⁻¹), 3.5% for TB (10 μg kg⁻¹) and 2.8% for FB (0.5 μg kg⁻¹) and recoveries for fruits (oranges, tangerines, lemons, limes, grapefruits, apples, pears and bananas) and vegetables (potatoes and lettuces) fortified at the μg kg⁻¹ level were in the interval 93-102%.