Migration of fluorochemical paper additives from food-contact paper into foods and food simulants.

Fluorochemical-treated paper was tested to determine the amount of migration that occurs into foods and food-simulating liquids and the characteristics of the migration. Migration characteristics of fluorochemicals from paper were examined in Miglyol, butter, water, vinegar, water-ethanol solutions, emulsions and pure oil containing small amounts of emulsifiers. Additionally, microwave popcorn and chocolate spread were used to investigate migration. Results indicate that fluorochemicals paper additives do migrate to food during actual package use. For example, we found that microwave popcorn contained 3.2 fluorochemical mg kg(-1) popcorn after popping and butter contained 0.1 mg kg(-1) after 40 days at 4 degrees C. Tests also indicate that common food-simulating liquids for migration testing and package material evaluation might not provide an accurate indication of the amount of fluorochemical that actually migrates to food. Tests show that oil containing small amounts of an emulsifier can significantly enhance migration of a fluorochemical from paper.

Rapid quantitative and qualitative confirmatory method for the determination of monofluoroacetic acid in foods by liquid chromatography-mass spectrometry.

A rapid quantitative method and a qualitative confirmatory method for the determination of monofluoroacetic acid (MFA) in complex food matrices are presented. The quantitative method utilizes a water extraction, solid phase extraction clean-up and liquid chromatography-mass spectrometry (LC-MS) for determination of MFA. This method showed a high degree of specificity, detecting MFA in all of the spiked samples, while none of the unfortified samples tested positive for MFA. Spike recoveries were high in all matrices analyzed, varying from 85 to 110%, and comparable at low (2mg/L) and high (20mg/L) spiking levels. Repeatability tests at the low spiking levels yielded RSDs of less than 5% for all matrices analyzed. The qualitative confirmatory method developed is conceptually different from the quantitative method, ensuring that both methods would not be subject to the same interferences. The method uses the formation of the hydrazide of MFA through derivatization with 2-nitrophenylhydrazine. This derivatization is well established for the determination of carboxylic acids, but this is the first application to the determination of MFA. The derivatization yield was matrix dependent, however the limit of detection (LOD) (0.8microg/L) was sufficient to confirm the presence of MFA in all spiked matrices. Repeatability tests at the low spiking levels yielded RSDs of approximately 7% for all matrices analyzed.

LC-UV and LC-MS analysis of food and drink products containing kava.

A method for the determination of six kava lactones, methysticin, dihydromethysticin, kawain, dihydrokawain, yangonin and desmethoxyyangonin, in solid foods and beverages has been developed. Solid samples were prepared using methanol extraction, while beverages were extracted using a separate solid phase extraction (SPE) method. After sample preparation, the extracts were analysed using LC-UV or atmospheric pressure photoionization (APPI) LC-MS in the positive mode. Using the method, 10 beverage products, two chocolate products, three unbrewed tea products, three dietary supplements and a drink mix product were analysed. The results obtained using the LC-UV were comparable to those obtained using APPI-LC-MS for most products. Using the SPE method in conjunction with LC-MS, individual kava lactones were detected in drink products at ppb concentrations. Concentrations of total kava lactones ranged between 135-0.035 mg per serving in the food and beverage products tested and between 40-61 mg per serving for the dietary supplement products tested. Results of these analyses as well as extraction efficiency and reproducibility data are reported.

Determination of 1,3-dichloropropanol in soy and related sauces by using gas chromatography/mass spectrometry.

A gas chromatography/mass spectrometry method for 3-MCPD in foods and food ingredients was modified for the determination of 1,3-DCP in soy and related sauces. The method was validated by using a blank soy sauce. The detection limit, quantitation limit and recoveries were determined, and identities were confirmed by mass spectrometry on the basis of analyses of test portions spiked with 1,3-DCP at 10, 25, 50 and 100 ng g(-1). The spiked test portions were quantitated by using an internal standard calibration curve. For the spiked test portions, the mean internal standard-corrected recovery for 1,3-DCP was 100% with a relative standard deviation of 1.32%. The limits of detection and quantitation were determined as 0.055 and 0.185 ng g(-1), respectively. The method also was compared with a headspace GC/MS method recently developed by the UK's Central Science Laboratory. Results from the method comparison showed that the recoveries for the spiked test portions, as well as the amounts of 1,3-DCP found in the retail products, were comparable.

Survey of chloropropanols in soy sauces and related products.

A survey of soy sauces and related products available in the USA was conducted to determine the levels of 3-monochloropropane-1,2-diol (3-MCPD) and 1,3-dichloro-2-propanol (1,3-DCP) in these products. Fifty-five retail samples were purchased and analysed for 3-MCPD. 3-MCPD determinations were made according to a gas chromatography/mass spectrometry method validated by a collaborative trial. Eighty-five per cent of the samples analysed contained greater than the detection limit of 0.005 ppm (microg g(-1)) for 3-MCPD. Thirty-three per cent contained greater than 1 ppm; the highest level was 876 ppm 3-MCPD. Thirty-nine of the samples analysed for 3-MCPD also were analysed for 1,3-DCP by using a modified method developed and validated in-house. Fifty-six per cent of the samples analysed for 1,3-DCP contained greater than the detection limit of 0.055 ppb (ng g(-1)) for 1,3-DCP; the highest level was 9.8 ppm 1,3-DCP. Products manufactured in Asia contained the highest chloropropanol levels.

Ethyl carbamate levels resulting from azodicarbonamide use in bread.

Azodicarbonamide (ADA), a dough conditioner, is an additive approved in the US up to a maximum of 45 mg/kg in flour. The addition of 45 mg/kg of ADA was investigated and found to increase the ethyl carbamate (EC) content of commercially prepared breads by 1-3 micrograms/kg. A similar increase in EC was observed in breads baked in the laboratory with a bread machine. The increase in EC levels appears to depend on a variety of factors, most notably the concentration of ADA added and the time of fermentation. The addition of 20 mg/kg ADA caused only a slight increase, if any, in commercial products but a 2.3 micrograms/kg increase of EC in breads baked with a bread machine. When 100 mg/kg of ascorbic acid was added along with ADA, smaller EC increases were observed. Addition of urea was also found to enhance the EC content of the bread. Toasting, which was previously shown to increase EC levels, caused even larger increases when ADA or urea had been added.

Screening method for the gas chromatographic/mass spectrometric determination of microgram/litre levels of bromate in bottled water.

Bromate can be formed as a by-product of ozone treatment that is sometimes used for the disinfection of municipal water supplies and bottled waters. The US Environmental Protection Agency has proposed a maximum contaminant level (MCL) of 10 micrograms/l for bromate in public drinking water. Should the proposed MCL for bromate become final, it may then be considered for adoption as a bottled water quality standard by the US Food and Drug Administration. This paper reports the development of a gas chromatographic/ mass spectrometric (GC/MS) method for the determination of parts-per-billion (microgram/l) levels of bromate (BrO3-) in bottled water. The GC/MS method was validated by using distilled and deionized Milli-Q water; detection limits, quantitation limits, and recoveries were determined and identities were confirmed by MS on the basis of analyses of test portions fortified with BrO3- at 0.8, 3.8, 7.7, 15, and 46 micrograms/l. The method also was evaluated on the basis of recoveries determined for two commercial brands of bottled water fortified with BrO3- at 3.8 and 7.7 micrograms/l and two commercial brands fortified at 0.8, 3.8, and 7.7 micrograms/l. For the Milli-Q water, recoveries ranged from 100 to 121%; for the fortified commercial products, recoveries ranged from 87 to 115%. The limits of detection and quantitation were determined to be 0.4 and 0.7 microgram/l, respectively. Several commercial brands of bottled water were analysed, and BrO3- was found in these products at levels ranging from none to 38 micrograms/l.

Measurement of bromate in bottled water by high-performance liquid chromatography with post-column flow reactor detection.

The objective of this work was to develop a reliable, rugged high-performance liquid chromatographic (HPLC) method for determination of trace levels of bromate (< 10 micrograms/l) in bottled water. HPLC separation was achieved by ion interaction chromatography using a C-18 reversed-phase column and a mobile phase consisting of methanol/water (20:80, v/v) with tetrabutylammonium acetate as the ion interaction reagent. A post-column reaction based on oxidation of o-dianisidine in acidic solution to a product detected at 500 nm provided selective measurement of the oxidants. The limit of detection and the limit of quantitation were 1 and 3 micrograms/l, respectively. Iodate, chlorite, and nitrite were chromatographically separated from bromate and measured by monitoring the post-column reaction. Chloride and chlorate at levels that might be found in bottled water did not interfere with the determination of bromate. Bromate was detected in bottled waters at concentrations up to 40 micrograms/l.

Survey of trihalomethanes and other volatile chemical contaminants in processed foods by purge-and-trap capillary gas chromatography with mass selective detection.

A limited number of soft drinks, juices, beers, and waters from processed vegetables were analyzed for trihalomethanes (THMs), benzene, and toluene by a modified Environmental Protection Agency (EPA) Method 524.2. The THMs, which include chloroform, bromodichloromethane, dibromochloromethane, and bromoform, are reaction by-products of water disinfection by chlorination. EPA Method 524.2 is a purge-and-trap capillary gas chromatographic method based on mass spectrometric detection which identifies and simultaneously measures purgeable volatile organic compounds in drinking water. Chloroform was present at concentrations ranging from none detected to 94 ng/g in the 44 foods analyzed. Bromoform was not found in any of the products at a detection limit of 0.1 ng/g. Residue levels of the other THMs ranged from none detected to highs of 12 and 2 ng/g for bromodichloromethane and dibromochloromethane, respectively. Benzene residues were typically < 5 ng/g, except for 7 and 9 ng/g in 2 foods. Toluene residues were typically < or = 3 ng/g except for 23, 29, and 75 ng/g in 3 canned foods.

Determination of free glutamic acid in a variety of foods by high-performance liquid chromatography.

A survey of free glutamic acid levels in a variety of foods was conducted. The foods were analysed by high-performance liquid chromatography (HPLC). Free glutamic acid was extracted from the food with 0.02 M potassium phosphate with subsequent precipitation of other food components with acetone. Impurities were removed by passing the extract through a C-8 or C-18 reversed-phase solid-phase extraction column. The free glutamic acid was derivatized with phenylisothiocyanate, and the derivative was separated by HPLC with detection at 254 nm. Free glutamic acid levels ranged from not found (< 20 ppm) in some spices to as high as 89% in another spice.

Determination of ethyl carbamate in alcoholic beverages and soy sauce by gas chromatography with mass selective detection: collaborative study.

A method using gas chromatography with mass selective detection for the determination of ethyl carbamate (EC; also known as urethane) in alcoholic beverages and soy sauce was collaboratively studied by 17 laboratories including authors' laboratories. The method uses prepacked columns for extraction of liquids with methylene chloride, and n-propyl carbamate as the internal standard. A practice sample and 6 samples of distilled spirits, fortified wines, table wines, and soy sauces were analyzed by each collaborator. Each matrix included blind duplicates of incurred and fortified EC at 3 levels. Distilled spirits contained 50-330 ng EC/g (ppb), fortified wine 40-160 ppb, table wine 10-50 ppb, and soy sauce 15-70 ppb. The ranges of the repeatability relative standard deviations, excluding outliers, were 4.03-6.63% for distilled spirits, 4.01-5.05% for fortified wine, 3.94-6.73% for table wine, and 4.70-8.49% for soy sauce. The ranges of the reproducibility relative standard deviations, excluding outliers, were 8.53-9.49% for distilled spirits, 6.84-12.02% for fortified wine, 8.86-18.47% for table wine, and 13.87-27.37% for soy sauce. Recoveries of added EC ranged from 87 to 93%. Recoveries relative to reference values, labeled as the internal standard, obtained by using gas chromatography/tandem mass spectrometry with a triple quadrupole mass spectrometer ranged from 89 to 100%.

Heterocyclic aromatic amine content of selected beef flavors.

Previous work has shown that meat extracts contain potent mutagenic and/or carcinogenic heterocyclic aromatic amines (HAAs). Because meat extracts and some beef flavors are produced from similar precursors and processing steps, the beef flavors may also contain HAAs. This study analyzed 24 commercial beef flavors and 2 food-grade beef extracts for creatine and creatinine concentrations, mutagenic activity and HAA concentrations (IQ, MeIQ, MeIQx, DiMeIQx, Glu-P-1, Glu-P-2 and PhIP). The creatine and creatinine levels of the flavors ranged from 0 to 73 and from 0 to 21 mg/g (dry wt.), respectively. The mutagenic activities of the flavors ranged from 0 to 3200 Salmonella typhimurium TA98 revertants/g (dry wt.). No direct relationship was found between creatine and/or creatinine concentrations and mutagenic activities. However, flavors with high creatine (> 1.5 mg/g) or creatinine (> 2 mg/g) levels exhibited higher mutagenic activities than did flavors with low levels of these compounds. Flavors with high mutagenic activities (> 1500 revertants/g) contained measurable amounts of HAAs. Three flavors contained MeIQx (7.2-21.2 ng/g [dry wt.]) and one contained DiMeIQx (4.2 ng/g [dry wt.]).

Survey of benzene in foods by using headspace concentration techniques and capillary gas chromatography.

Recently, the combination of sodium or potassium benzoate with ascorbic acid was shown to produce low levels (ng/g) of benzene in fruit-flavored soft drinks. The presence of benzene also was reported in butter, eggs, meat, and certain fruits; levels of these findings ranged from 0.5 ng/g in butter to 500-1900 ng/g in eggs. Because benzoates are widely used as food preservatives, a limited survey of other foods containing added benzoate salts was conducted to investigate the potential for benzene formation. Selected foods that did not contain added benzoates but were previously reported to contain benzene were analyzed for comparison. More than 50 foods were analyzed by purge-and-trap or static headspace concentration and capillary gas chromatography. Benzene was quantitated by using the method of standard additions, and its identity was confirmed by mass selective detection. Results of this limited survey show that foods without added benzoates (including eggs) contained benzene at levels equal to or less than 2 ng/g. Slightly higher levels were present in some foods and beverages containing both ascorbic acid and sodium benzoate.

Comparison of two clean-up methodologies for the gas chromatographic/mass spectrometric determination of low nanogram/gram levels of polynuclear aromatic hydrocarbons in seafood.

The March 1989 oil spill in Alaska prompted the Food and Drug Administration (FDA) to conduct a thorough investigation of clean-up methodologies aimed at determining low ng/g (ppb) levels of polynuclear aromatic hydrocarbons (PAHs) in seafood. The clean-ups from a modified FDA method and a National Marine Fisheries Service (NMFS) method were evaluated on the basis of the determination of 18 PAHs at levels ranging from 1 to 5 ppb by gas chromatography/mass spectrometry. In the modified FDA method, seafood extracts were purified by a liquid-liquid partition followed by a three-step elution through silica, alumina, and C18 solid-phase extraction cartridges. In the NMFS method, seafood extracts were purified by column chromatography through a deactivated silica gel/alumina column and a gel permeation high performance liquid chromatography column. Both methods quantitated 18 PAHs at levels ranging from 1 to 5 ppb. With the exception of naphthalene, average recoveries based on internal deuterated standards ranged from 73 to 144% for the modified FDA method and 63 to 106% for the NMFS method.

Survey of sulphites determined in a variety of foods by the optimized Monier-Williams method.

A survey of sulphite levels in a variety of foods including dried fruit, dried mashed potatoes, shrimp, canned mushrooms and fruit juices has been conducted. The levels found ranged from 0 ppm in orange juice to 3722 ppm as sulphur dioxide in dried fruit. The foods were analysed by the optimized Monier-Williams procedure.

Determination of free and reversibly bound sulphite in foods by reverse-phase, ion-pairing high-performance liquid chromatography.

The reaction of sulphite with formaldehyde to form hydroxymethylsulphonate (HMS), which is very stable under the controlled conditions of this assay, was used as the first step in an analytical procedure to determine foodborne sulphite. The effect of mobile-phase pH on the stability of HMS during high-performance liquid chromatography was studied. It was found that on-column HMS dissociation to formaldehyde and bisulphite increased with the pH of the mobile phase; therefore the relatively low pH 4.7, at which the dissociation of HMS was approximately 2%, was selected for the analysis. In addition, the release of sulphite from its reversibly bound forms in wine and other foods was examined as a function of the pH of the extraction medium by following the appearance of HMS formed from the reaction of the freed sulphite with formaldehyde. The rate of dissociation of the reversibly bound sulphite was relatively slow at pH 3 but very rapid at pH 7. This difference in kinetics was exploited to develop a procedure to determine free and reversibly bound sulphite in food. The method was challenged by post-reagent spiking studies, i.e. adding the sulphite spike after the food has been blended with the sulphite-protective formaldehyde solution but before proceeding with the remainder of the assay. An average recovery of 100% with a standard deviation of 5.2% (n = 45) was realized at levels of 5, 10 and 20 ppm by weight as sulphur dioxide. Recovery of the sulphite added as the bisulphite addition product of acetaldehyde, a model compound for reversibly bound sulphite, was 95%.

Ethyl carbamate levels in selected fermented foods and beverages.

Ethyl carbamate (EC), also known as urethane, is an animal carcinogen and a by-product of fermentation. Because EC has been found in distilled spirits and wines, a variety of fermented foods and beverages were analyzed to assess its occurrence in other products. Previously described methods using a gas chromatograph-thermal energy analyzer with a nitrogen converter were modified for each matrix and gave recoveries of greater than 80%, with a limit of detection in the 1-2 micrograms/kg (ppb) range. A total of 152 test samples were analyzed; EC levels ranged from none found to 3 ppb in 15 cheeses, 6 teas, 12 yogurts, and 8 ciders; from none found to 13 ppb in 30 breads and 69 malt beverages; and from none found to 84 ppb in 12 soy sauces. Gas chromatography/mass spectrometry/mass spectrometry was used to confirm EC identity and to quantitate EC in selected food extracts.

Liquid chromatographic determination of sulfite in grapes and selected grape products.

A liquid chromatographic (LC) method is described for the determination of sulfite in grapes and certain grape products. Sulfite is extracted from grapes with aqueous formaldehyde solution buffered at pH 5; free sulfite is converted to hydroxymethylsulfonate (HMS), which is extremely stable at pH 3-7. Subsequent heating to 80 degrees C for 30 min converts reversibly bound forms of sulfite to HMS. The extract is then analyzed by reverse-phase ion-pairing liquid chromatography, using a C18 column and a mobile phase of aqueous 0.005 M tetrabutylammonium ion in 0.05 M acetate, pH 4.7, and a flow rate of 1 mL/min. Aqueous KOH is added to the eluate to convert HMS to free sulfite, which is then treated with 5,5'-dithiobis[2-nitrobenzoic acid]. This reaction produces the 3-carboxy-4-nitrothiophenolate anion, which is determined by measurement of electronic absorption at 450 nm. For grapes spiked with HMS at 5-20 ppm (as SO2), recoveries ranged from 92 to 112%, with a coefficient of variation of 4.6%. The method was also used to determine sulfite in various grape products. Results were comparable to those obtained by the AOAC official Monier-Williams method.

Residues of volatile halocarbons in margarines.

Findings of residues of volatile halocarbons (VHCs) such as 1,1,1-trichloroethane, trichloroethylene and tetrachloroethylene in margarine are reported. VHCs were determined by a headspace gas chromatographic method with electron capture detection. Identity was confirmed by headspace gas chromatography-mass spectrometry for some of the higher level (100-5000 ppb) residues. A total of 70 stick, soft and diet soft margarines were purchased in the Washington, DC, metropolitan area. In addition, margarines and margarine ingredients were collected from 19 production lines. These products plus the adhesives used in the packaging were examined. Levels of VHCs ranging from 5 to 100 ppb (ng/g) were found in many of the margarines. The highest concentrations of any individual VHC (tetrachloroethylene at 1-5 ppm) were found in margarines obtained from a supermarket located next to a dry cleaner. Possible sources of VHC residues are discussed.