Natural occurrence of ochratoxin A in dried figs

Ochratoxin A (OTA) contamination in dried figs was investigated using high performance liquid chromatography (HPLC) with fluorescence detection after extraction with methanol and orthophosphoric acid and clean up by an immunoaffinity column. The limit of detection for OTA was 0.12 microg kg(-1). One hundred and fifteen samples were taken during the drying stage from 7 different districts in the Aegean Region in 2003 and 2004. Fifty-five (47.2%) of the 115 samples were found to contain detectable levels of ochratoxin A, ranging from 0.12 to 15.31 microg kg(-1). However, the OTA level for a majority of the samples was low, with only 4 samples containing OTA exceeding 1 microg kg(-1). The calculated overall median for the OTA level was below the limit of detection and the overall mean was estimated as 0.52 microg kg(-1). Frequency of ochratoxin A contamination in dried figs harvested in 2003 and 2004 are 47 and 50%, respectively. Highest contamination ratio was determined in dried figs from Erbeyli (60%), followed by Selcuk (56%), and Ortaklar (50%).     

Determination of Cd and Pb in food slurries by GFAAS using cryogenic grinding for sample preparation

A simple method combining slurry sampling after cryogenic grinding and the use of a permanent modification of the integrated platform inside the transversely heated graphite atomizer (THGA) was proposed for the determination of Cd and Pb in foods. Potentialities of the cryogenic grinding were evaluated for grinding different materials of difficult homogenization such as high fat and high fiber tissues. Animal and vegetal samples were cut into small pieces and ground in liquid nitrogen for 2 min. Slurries were prepared directly in the autosampler cup after cryogenic grinding by transferring an exact amount of homogeneous powdered material (5-20 mg) to the cup, followed by 1.00 mL of 0.2% (v/v) HNO3 containing 0.04% (v/v) Triton X-100 and sonication for 30 s, before transferring into the platform previously coated with 250 microg W and 200 microg Rh. Use of a tungsten carbide-rhodium permanent modifier combined with NH4H2PO4 conventional modifier improves tube lifetime and increases the pyrolysis temperature for Cd. Homogeneity tests, carried out by comparing the between- and within-batch precision for each kind of sample, showed no significant differences at the 95% confidence level, indicating good homogeneity for 5-20 mg masses. Detection limits were 3.3 ng g(-1) Cd and 75 ng g(-1) Pb for 1% m/v slurries. Results for determination of Cd and Pb in foods slurries were in agreement with those obtained with digested samples, since no statistical differences were found by the paired t-test at the 95% level.     

Determination of eight sulfonamides in bovine kidney by liquid chromatography/tandem mass spectrometry with on-line extraction and sample clean-up

A sensitive, high performance liquid chromatography/tandem mass spectrometric (i.e. mass spectrometry/mass spectrometry; LC/MS/MS) method with on-line extraction and sample clean-up for the screening and confirmation of residues of sulfonamides in kidney is described. The sulfonamides are extracted from homogenized kidney with methanol. After centrifugation of the extract, an aliquot of the extract is directly injected on the LC/MS/MS system with further extraction and clean-up of the sample on-line. Detection of the analytes was achieved by positive electrospray ionization (ESI) followed by multiple reaction monitoring. For each sulfonamide the collisional decomposition of the protonated molecule to a common, abundant fragment ion was monitored. The method has been validated for sulfadimethoxine, sulfaquinoxaline, sulfamethazine, sulfamerazine, sulfathiazole, sulfamethoxazole, sulfadiazine and sulfapyridine. Calibration curves resulting from spiked blank kidney samples at the 10-200 microg/kg level showed good linear correlation. At the level of 50, 100 and 200 microg/kg both within- and between-day precision, as measured by relative standard deviation (RSD), were less than 16%. The limits of detection (LODs) ranged from 5 to 13.5 microg/kg. The recoveries ranged from 78 to 82%. The procedure provides a rapid, reliable and sensitive method for the determination of residues of sulfonamides in bovine kidney. The advantage of this method over existing methods is its decreased sample preparation and analysis time, which makes the method more suitable for routine analysis.     

Confirmation of synthetic glucocorticoids with liquid chromatography/mass spectrometry: organization and results of an international interlaboratory comparison test

Within the framework of a European Union (EU) research project entitled "Food Safety Screening: Synthetic Glucocorticoids (QLK1-1999-00122)," an international interlaboratory ring test was organized to compare and evaluate different liquid chromatography/mass spectrometry (LC/MS) confirmatory methods that are applied in European monitoring programs for detecting the use of synthetic glucocorticoids. Liver and urine samples of bovines treated with synthetic glucocorticoids were collected and sent to the participants of the study for analysis. Participants received 3 liver and 3 urine samples and were free to use either their own LC/MS method or an LC/MS-based method developed during the EU research project. The residue concentrations in the samples were calculated as the mean of the concentrations reported by each laboratory. The mean dexamethasone concentration of liver sample L1 was calculated as 2.27 microg/kg [relative standard deviation (RSD) 43%, n = 9], which exceeds the maximum residue level (MRL) of 2 microg/kg. Three of the 9 laboratories (33%) reported concentration levels less than 2 microg/kg, resulting in obviously false compliant results. The overall mean concentration of flumethasone in liver sample L2 was calculated as 3.27 microg/kg (RSD 33%, n = 8). Applying a comparable limit for flumethasone of 2 microg/kg, 8 of the 9 laboratories would have obtained a correct noncompliant result. As for the blank liver sample, 1 participant found a false noncompliant result. The urine sample U1 contained prednisolone residues at a mean concentration of 1.58 microg/kg (RSD 43%, n = 9). Four out of 9 results were less than a theoretical minimum required performance level (MRPL) of 2 microg/kg. The calculated concentration of dexamethasone in urine sample U3 was 5.21 microg/kg (RSD 62%, n = 9). One of the 9 results was lower than 2 microg/kg. Urine sample U2 was correctly reported as blank by all participants.     

Residue analysis for halofuginone in sturgeon muscle by immunoaffinity cleanup and liquid chromatography

A high-performance liquid chromatography (LC) method was developed for the determination of halofuginone (HFG) in sturgeon muscle. The extracted samples were cleaned up by an immunoaffinity chromatography column that was prepared by covalently coupling polyclonal antibodies against HFG to cyanogen bromide (CNBr) activated Sepharose 4B. The eluate was evaporated to dryness, and residues were determined by LC with absorbance detection at 243 nm. Recoveries of HFG from samples fortified at 20-200 microg/kg levels ranged 74.6-81.1%, with coefficients of variation of 0.7-8.6%. The detection limit was estimated to be 10 microg/kg in a 2 g sample.     

Dropping mode chemiluminescence system with a poly(dimethylsiloxane) cell for applications to semiconductor processes.

A lab-made chemiluminescence system with a polymer cell for the dropping mode was used to determine ultra-trace metal ions in hydrofluoric acid (HF) and the standard cleaning solution-1 (SC-1) used in semiconductor manufacturing processes. The cell was made of poly(dimethylsiloxane) (PDMS) with dimensions of about 10 mm i.d. and 8 mm in height, was cheap, disposable, chemically inert to alkalis and acids, especially HF, and was optically transparent in the visible region. A dropping method for sample injection was adopted to minimize pulsation and the dramatic pH change of the luminol-H2O2 reagent when adding the sample. The average sample weight of a single drop was 7.17 microg with a remarkable reproducibility of +/-0.37% relative standard deviation (RSD). This very small sample volume compared to the reagent volume made it possible to avoid any precipitation being formed when HF was added. For an application, Fe was determined in deionized (d.i.) water, sulfuric acid, SC-1, and a diluted HF (1:200 DHF) solution, which have been commonly used in semiconductor manufacturing processes. The limits of detections for Fe2+ in those solutions were found to be in the range of 42 to 62 pg ml(-1). Based on the analytical results, this chemiluminescence system with the PDMS cell was reproducible, resistant to HF, had less sample consumption and waste generation, and was sensitive enough to apply to the semiconductor industry as an on-line monitoring sensor. Although this chemiluminescence system does not have selectivity for each specific metal ion, it can be used as an on-line sensor to monitor the metal contamination level of Fe, Cu, Co, etc., which are major elements of concern in the semiconductor manufacturing process.     

Improving the analytical sensitivity of alpha-emitting impurities in grinding high purity silica

The source of contaminating high purity silica during grinding process and analytical sensitivity of alpha-emitting impurities have been investigated. To improve the analytical sensitivity for alpha-emitting impurities in grinding high purity silica, a new procedure was studied by employing an irradiation facility in the hydraulic transfer system (HTS) and the nitrogen flushing system. The source of silica contamination was found to be attributed mostly to alumina balls. The detection limit of U and Th in 3N-grade silica samples by neutron activity analysis (NAA) could be improved to a sub-ng/g level. The contamination rates of Th, U, Cl, Fe and Na from the alumina balls were calculated to be 95%, 15%, 15% and 3.5%, respectively.     

Analysis of fumonisins B(1), B(2) and B(3) in corn-based baby food by pressurized liquid extraction and liquid chromatography/tandem mass spectrometry

A sensitive and reliable method using pressurized liquid extraction (PLE) and liquid chromatography (LC)/electrospray ionization (ESI) tandem mass spectrometry with a triple quadrupole (QqQ) analyzer has been developed for the analysis of fumonisin B(1) (FB(1)), fumonisin B(2) (FB(2)) and fumonisin B(3) (FB(3)) in corn-based baby foods. Influence of several extraction parameters that affect PLE efficiency such as temperature, pressure, solvent extraction, number of cycles and dispersant/clean-up agents were studied. The selected PLE operating method was: 3g of sample was packed into 11 ml stainless-steel cell and fumonisins were extracted with methanol at 40 degrees C, 34 atm in one cycle of 5 min at 60% flush. The analytes were ionized in ESI operating with positive ion mode and identified by selecting two monitoring transitions, permitting quantification and confirmation in a single injection. Recoveries ranged from 68% to 83% at fortification levels of 200 microg kg(-1) with relative standard deviation (RSD) from 4% to 12%. The limits of quantification were from 2 microg kg(-1) for FB(1) and FB(2), and 5 microg kg(-1) for FB(3), which are below the maximum residue level established by the European Union legislation in infant formulas. The proposed method was successfully applied to the analysis of twenty seven samples of baby food products collected from different markets, and one positive sample with a content of 15.9 microg kg(-1) for FB(1), 9.2 microg kg(-1)for FB(2) and 5.8 microg kg(-1) for FB(3) was obtained. Given the simplicity and potential of the proposed procedure, its application for safety control is recommended.     

A new method for determination of hydroperoxides in cellulose

A new chromatographic method for determination of hydroperoxides in cellulose is described, whereby the sample is dispersed in phosphate buffer solution (pH 7) of FeCl(3), EDTA and the hydroxyl radical scavenger, N,N'-(5-nitro,1,3-phenylene)bisglutaramide. The reaction time, concentration of reagents and chromatographic conditions for subsequent separation and quantification of hydroxylated products are optimised. The limit of detection is 0.14 micro mol L(-1) H(2)O(2) which corresponds to 2.1 micro mol hydroperoxides per kg of cellulose (mass of sample: 0.4 g). It was shown that in the concentration range of H(2)O(2) up to 15 micro mol L(-1) the method gives linear response and that adsorption of reaction products on fibres is negligible. For cellulose samples, this corresponds to a linear range of 2.1-225 micro mol of hydroperoxides per kg. Due to possible side reactions of hydroxyl radicals in the suspension of analyte, it is recommended that the standard addition technique is used. In purified cotton cellulose, the steady state concentration of hydroperoxides at room conditions was found to be 15.1+/-1.5 micro mol kg(-1).     

Effect of grinding on thermal reactivity of ceramic clay minerals

The effect of grinding on thermal behavior of pyrophyllite and talc as commonly used ceramic clay minerals was investigated by DTA, TG, emanation thermal analysis (ETA), B.E.T. surface area (s.a.) measurements, X-ray diffraction (XRD) and scanning electron microscopy (SEM). A vibratory mill was used in this study, grinding time was 5 min. It was found that the grinding caused an increase in surface area and a grain size reduction of the samples. From TG and DTA results it followed that grinding caused a decrease of the temperature at which the structure bound OH groups released. The formation of high temperature phases was enhanced with the ground samples. For the ground talc sample the crystallization of non-crystalline phase into orthorhombic enstatite was observed in the range of 800°C. For ground pyrophyllite a certain agglomeration of grains was observed in the range above 950°C. Moreover, for both clays the ETA characterized a closing up of subsurface irregularities caused by grinding as a decrease of the emanation rate in the range 250–400°C. The comparison of thermal analysis results with the results of other methods made it possible to better understand the effect of grinding on the ceramic clays.     

Contamination of Rock Samples by Laboratory Grinding Mills

We have measured the contamination of samples of granite, quartz porphyry, greenstone and dolerite after grinding them in steel and cemented carbide grinding mills. The steel mill contaminated the granite with Fe. The cemented carbide mill contaminated all samples with W and Co, and the granite and quartz porphyry with Ta. Contamination during grinding is proportional to the free quartz content of the rock and to the grinding time.     

Determination of synthetic chelating agents in surface and waste water by ion chromatography-mass spectrometry

Coupling of ion chromatography with electrospray mass spectrometry (IC-MS) is a simple, sensitive and quick method for the determination of polar organic traces in water samples without derivatization. Analysis of the chelating agents ethylenediamino tetraacetate (EDTA) and diethylenetriamino pentaacetate (DTPA) in aqueous samples was done by IC-MS on an anion exchange column after simple sample preparation steps. Quantification down to a concentration level of 1 microg L(-1) even in wastewater influents and effluents was achieved utilizing 13C marked internal standards and measuring the individual [M - H+]- and stable [M - 4H+ + Fe3+]- cluster ions. The method was validated against certified, but more time consuming routine methods. Applying this method a series of several European water samples were analyzed for EDTA and DTPA indicating their nature as polar persistent pollutants.     

Development and validation of a multiresidue method for the analysis of 151 pesticide residues in strawberry by gas chromatography coupled to a triple quadrupole mass analyzer

A new method was developed and validated for the simultaneous determination of 151 pesticide residues in strawberry by gas chromatography coupled to a triple quadrupole mass analyzer (GC/QqQ-MS/MS), mainly using the selected reaction monitoring (SRM) mode. The list of target compounds included various classes of pesticides such as organochlorine (OCPs), organophosphorus (OPPs), carbamates, pyrethroids, triazoles and dicarboximides. A single extraction of 10 g of sample with acetonitrile followed by liquid-liquid partition formed by the addition of 4 g of MgSO4 and 1 g of NaCl was applied in sample preparation. Cleanup of the extracts was carried out by applying dispersive solid-phase extraction (D-SPE) with primary secondary amine (PSA). The analysis time was 21 min. The method was subjected to a thorough validation procedure. The recovery data were obtained by spiking blank samples at two concentration levels (11.5 and 50 microg/kg), yielding recoveries in the range 70-110%. Precision values expressed as relative standard deviation (RSD) were lower than 18% and 22% for the intraday and interday precision, respectively. Linearity was studied in the range 10-200 microg/kg and determination coefficients (R(2)) were higher than 0.98% for all compounds. Limits of detection (LODs) and limits of quantification (LOQs) were established as 4 and 10 microg/kg, respectively. The overall uncertainty of the method was estimated at two different concentrations (11.5 and 50 microg/kg), being lower than 25% in both cases. According to the validation data and performance characteristics as well as the high sample throughput and low cost, the proposed method is suitable for routine application.     

Application of hollow fiber supported liquid membrane extraction to the simultaneous determination of pesticide residues in vegetables by liquid chromatography/mass spectrometry

A new analytical procedure using a hollow fiber supported liquid membrane (HFSLM) has been developed for the simultaneous determination of pesticide residues in vegetables by liquid chromatography (LC) coupled with electrospray mass spectrometry (MS). The extraction technique requires minimal sample preparation and solvent consumption. Optimum extraction conditions have been evaluated with respect to sample pH, ionic strength, liquid membrane composition, extraction time, stirring rate and acceptor composition. The extraction method has been validated for matrices such as cucumber, tomato and pepper, indicating that cucumber can be selected as representative matrix for routine analysis of these food commodities. Linear ranges of pesticides in vegetable samples were 10 to 200 microg/kg, and the repeatability of the method was less than 20% for the lowest calibration point. The limits of detection ranged from 0.06 to 2.7 microg/kg and the limits of quantification from 0.2 to 9.0 microg/kg, which were low enough to determine the pesticide residues at concentrations below or equal to the maximum residue levels (MRLs) specified by European Union. The method was finally applied to the determination of more than 20 pesticides in market vegetable samples and the concentrations found in these samples were always lower than the MRLs. This new approach can be considered as a powerful alternative to the traditional extraction techniques.     

Simultaneous determination of seventeen glucocorticoids residues in milk and eggs by ultra-performance liquid chromatography/electrospray tandem mass spectrometry

A comprehensive analytical method has been developed and validated for the simultaneous determination of seventeen glucocorticoid residues in eggs and milk. The mass spectrometer parameters, the composition of the mobile phase and the sample preparation method were firstly optimized to obtain maximum sensitivity. The samples were deconjugated with beta-glucuronidase/arylsulfatase enzyme and concentrated using an Oasis HLB solid-phase extraction cartridge, followed by cleanup with a dual Sep-pak silica and aminopropyl cartridge. The analytes were quantified by ultra-performance liquid chromatography (using a C18 column)/electrospray ionization tandem mass spectrometry (UPLC/ESI-MS/MS) operating in the negative ion mode. The assay for the 17 glucocorticoids was linear over the range of 1-200 microg/L for milk and egg samples with a high correlation coefficient (>0.99). The limits of quantification (LOQs) for the target analytes were 0.04-1.27 microg/kg for the egg samples and 0.03-0.73 microg/kg for the milk samples. The average extraction recoveries of the glucocorticoids from eggs and milk at two concentration levels (spiked at 0.40 and 2.00 microg/kg) were 65.6-118.7% and 61.5-119.6%, respectively, with relative standard deviations between 1.8-17.0% and 2.4-18.4%, respectively. Because of its high sensitivity, good precision and specificity, the method was found to be suitable for trace analysis of synthetic and natural glucocorticoids in complex biosamples such as eggs and milk.     

Preparation and certification of zearalenone mass concentration of two low-level maize reference materials

The contamination of maize by fungi, especially by Fusarium species, is a worldwide problem. One of the most prevalent Fusarium mycotoxins frequently found on European maize is zearalenone (ZON), which has been implicated in a range of human and animal diseases. It shows remarkable estrogenic properties and can cause severe infertility problems in farm animals. Currently, 9 countries have set maximum tolerable levels for ZON in food, ranging from 0 to 1000 microg/kg. This paper describes the preparation of 2 maize reference materials (BCR-716 very low level ZON and BCR-717 low level ZON) and the certification of their individual ZON contents (mass concentration and mass fraction). Uncertainties were calculated in compliance with the Guide to the Expression of Uncertainty in Measurement and include uncertainties that are due to possible inhomogeneity and instability. Finally, BCR-716 was certified at a level of <5 microg/kg and BCR-717 at a level of 83 microg/kg with an expanded uncertainty (k = 2) of 9 microg/kg.     

Automated headspace-solid-phase micro extraction-retention time locked-isotope dilution gas chromatography-mass spectrometry for the analysis of organotin compounds in water and sediment samples

An automated method for the simultaneous determination of six important organotin compounds namely monobutyltin (MBT), dibutyltin (DBT), tributyltin (TBT), monophenyltin (MPhT), diphenyltin (DPhT) and triphenyltin (TPhT) in water and sediment samples is described. The method is based on derivatization with sodium tetraethylborate followed by automated headspace-solid-phase micro extraction (SPME) combined with GC-MS under retention time locked (RTL) conditions. Home-synthesized deuterated organotin analogues were used as internal standards. Two high abundant fragment ions corresponding to the main tin isotopes Sn118 and Sn120 were chosen; one for quantification and one as qualifier ion. The method was validated and excellent figures of merit were obtained. Limits of quantification (LOQs) are from 1.3 to 15 ng l(-1) (ppt) for water samples and from 1.0 to 6.3 microg kg(-1) (ppb) for sediment samples. Accuracy for sediment samples was tested on spiked real-life sediment samples and on a reference PACS-2 marine harbor sediment. The developed method was used in a case-study at the harbor of Antwerp where sediment samples in different areas were taken and subsequently screened for TBT contamination. Concentrations ranged from 15 microg kg(-1) in the port of Antwerp up to 43 mg kg(-1) near a ship repair unit.     

Determination of trace amounts of lead in mussels by flow-injection flame atomic-absorption spectrometry coupled with on-line minicolumn preconcentration

A minicolumn packed with poly(aminophosphonic acid) chelating resin incorporated in an on-line preconcentration system for flame atomic-absorption spectrometry was used to determine ultratrace amounts of lead in mussel samples at microg L(-1) level. The preconcentrated lead was eluted with hydrochloric acid and injected directly into the nebulizer for atomization in an air-acetylene flame for measurement. The performance characteristics of the determination of lead were: preconcentration factor 26.8 for 1 min preconcentration time, detection limit (3sigma) in the sample digest was 0.25 microg g(-1) (dry weight) for a sample volume of 3.5 mL and 0.2 g sample (preconcentration time 1 min), precision (RSD) 2.3% for 25 microg L(-1) and 2.0% for 50 microg L(-1). The sampling frequency was 45 h(-1). The method was highly tolerant of interferences, and the results obtained for the determination of lead in a reference material testify to the applicability of the proposed procedure to the determination of lead at ultratrace level in biological materials such as mussel samples.     

Flow injection spectrophotometric determination of ascorbic acid in soft drinks and beer

Two spectrophotometric methods, a photochemical and a non-photochemical, for the determination of ascorbic acid in soft drinks and beer using a flow-injection system are proposed. The non-photochemical method is based on the redox reaction that takes place between ascorbic acid and Fe(III), yielding dehydroascorbic acid and Fe(II). Fe(II) reacts with 1,10-phenantroline, originating the reddish orange Fe(phen)3(2+) complex (ferroin). This complex is spectrophotometrically monitored at 512 nm, and the signal is directly related to the concentration of ascorbic acid in the sample. The photochemical method has the same basis, nevertheless, uses the irradiation with visible light to enhance the redox reaction and so achieve higher sensitivities in the analysis. The non-photochemical method shows a linear range between 5 and 80 microg mL(-1), with a relative standard deviation of 1.6% (n = 11), a detection limit of 2.7 microg mL(-1) and a sample throughput of 60 samples h(-1). The photochemical method shows a linear range between 1 and 80 microg mL(-1), with a relative standard deviation of 1.0% (n = 11 ), a detection limit of 0.5 microg mL(-1) and a sample throughput of 40 samples h(-1).     

Detection of streptomycin and dihydrostreptomycin residues in milk,honey and meat samples using an optical biosensor

Immuno-biosensor inhibition assays for the detection of streptomycin and dihydrostreptomycin residues in whole cows' milk, honey, pig kidney and pig muscle are reported. The antibody showed high cross-reactivity with dihydrostreptomycin in various foodstuffs (buffer 103%, milk 96%, honey 84%, kidney extract 129% and muscle extract 98%). There was no significant cross-reaction with other aminoglycosides or commonly used antibiotics. A streptomycin derivative was used to prepare a stable, reusable sensor chip surface. The assay allowed the direct analysis of bovine whole milk (fat content approximately 3.5%). Honey samples required dilution with buffer, while kidney and muscle samples from pigs were homogenized in an aqueous extraction buffer and clarified by centrifugation. The limit of detection for each assay was determined from known streptomycin-free samples (n = 20; mean - (3 x standard deviation)) and the results were as follows: milk 30 microg kg(-1), honey 15 microg kg(-1), kidney 50 microg kg(-1) and muscle 70 microg kg(-1). Repeatability (or relative standard deviation) between runs were calculated (n = 3) at the respective Community maximum residue limits (MRL) and 0.5 x MRL with the exception of honey since no European MRL exists at present. Results were determined as 4.3% (200 microg kg(-1)) and 2.8% (100 microg kg(-1)) in milk, 13.3% (40 microg kg(-1)) and 9.5% (20 microg kg(-1)) in honey, 7.1% (1000 microg kg(-1)) and 7.6% (500 microg kg(-1)) in kidney and 7.1% (500 microg kg(-1)) and 11% (250 microg kg(-1)) in muscle.