Rice samples cultivated in the USA, Europe and Basmati regions have been analysed using Isotope Ratio Mass Spectrometry (IRMS) and Inductively Coupled Plasma Mass Spectrometry (ICP-MS). Nine key variables (carbon-13, oxygen-18, boron, holmium, gadolinium, magnesium, rubidium, selenium and tungsten) were identified by canonical discriminant analysis as providing the maximum discrimination between rice samples from these regions. High levels of boron (>2500 ppb) were associated with rice samples from America and notably high levels of holmium were found in rice samples from the state of Arkansas. European rice samples generally contained relatively high levels of magnesium and Indian/Pakistani samples were characterised by relatively low oxygen-18 (18O) abundance.

Intestinal nitrosation produces ATNCs (Apparent Total N-nitroso Compounds) and these have been linked with an increased risk of colon cancer from eating red meat. Modern LC-MS instrumentation makes direct detection of ATNC components in faecal water a possibility. The difficulty is in determining which of the many compounds present are N-nitrosamines before embarking on efforts to characterise them. We have assumed that any in vivo nitrosation of alimentary tract contents will be non-specific and depend on the amount and basicity of amine present, with concentration of nitrosating agent being the limiting factor. By further nitrosating faecal waters (and ileostomy fluids) we can increase the amount of ATNC and readily access these compounds. The amount and the number of nitrosamines generated depend on the concentration of individual amines present. By derivatisation separately using both 14N and 15N labelled nitrite, we demonstrated that inspecting chromatograms in parallel for a unit mass difference provides a novel and practicable means for identifying unknown ATNC in faecal and ileostomy samples. MS procedures were linked with the traditional approaches of thermal energy analyser (TEA), preparative HPLC, visualisation of nitrosamines with Griess reagent and degradation of N-nitroso compounds by UV irradiation. We have demonstrated that this approach is repeatable and have used it to identify 30 putative N-nitroso compounds (as protonated parent masses [M + H]+ at: 242, 258, 312, 313, 333, 348, 365, 377, 382, 386, 392, 412, 414, 421, 434, 442, 466, 467, 483, 493, 572, 582, 625, 636, 637, 656, 662, 752, 808, and 870.

An improved procedure for determining (13)C and (2)H isotope ratios, using gas chromatography-isotope ratio mass spectrometry (GC-IRMS), has been developed for identifying the addition of low cost commercial sugar syrups to apple juices and related products. Isotopic techniques are commonly used to identify the addition of low cost sugars to fruit juices and are difficult to circumvent as it is not economically viable to change the isotopic ratios of the sugars. The procedure utilizes the derivative hexamethylenetetramine, which is produced through chemical transformation of a sugar degradation product and provides position-specific (13)C and (2)H ratios that relate to the parent sugar molecule. The new procedure has advantages over methods using nitro-sugar derivatives in terms of analysis time and sensitivity. The differences between the delta(2)H per thousand and delta(13)C per thousand values of the 100 authentic apple juices and beet and cane commercial sugar syrups permit their addition to be reliably detected.

Lutein, zeaxanthin, β-cryptoxanthin, α-carotene and β-carotene have been determined in samples of Brazilian orange juice (Citrus sinensis). The concentrations found in factory-produced concentrates have been compared with those obtained on the Brazilian retail market and with authentic hand-squeezed juices. The analyses of the latter enabled a comparison of varieties to be made. A concentration range of 0.11–1.21 mg litre−1 was determined for total carotenoids with β-carotene, the most important source of Vitamin A, being found in the highest concentration in the Pera variety followed by Valência, Natal, Lima and Baı́a varieties. The total carotenoids present in samples of frozen concentrated orange juice (FCOJ) obtained from factories ranged from 0.26–0.48 mg litre−1, while retail samples of this product contained slightly more (0.46–0.81 mg litre−1). Frozen concentrated orange pulp-wash presented much lower concentrations, ranging from 0.04 to 0.08 mg litre−1 of total carotenoids. Fourteen samples of retail freshly-squeezed orange juice contained carotenoids ranging from 0.04 to 0.55 mg litre−1, with only one sample out of the range found for authentic samples. This could be due to the addition of pulp-wash to this sample, in which undeclared sorbic acid was also detected.

Authentic samples of oranges, frozen concentrated orange juice and pulp-wash, and retail samples of freshly squeezed orange juice and frozen concentrated orange juice have been collected in Brazil and analysed for the flavanone glycosides (FG) narirutin and hesperidin by reversed phase HPLC with UV detection at 280nm. The juice from hand-squeezed fruit gave narirutin and hesperidin concentrations of 16–142mgl−1 and 104–537mgl−1, respectively. The ratio of hesperidin to narirutin showed varietal difference with Pera having the highest ratio (mean 8.4) and Baı́a the lowest (3.6). Frozen concentrated orange juice contained higher quantities of FG with narirutin ranging from 62 to 84mgl−1 and hesperidin from 531 to 690mgl−1 (after dilution to 12°Brix). In frozen concentrated orange juice pulp-wash, the narirutin level ranged from 155 to 239mgl−1 and hesperidin from 1089 to 1200mgl−1. The analysis of 23 samples of freshly squeezed juice from the Brazilian market place showed that the FG content of most samples (9.1 to 94.8 and 105.8 to 586.6mgl−1, respectively, for narirutin and hesperidin) was similar to those found for authentic ones, indicating that these orange juices were not adulterated.

Sinensetin has been quantified in authentic samples of Brazilian orange juice. In addition, six further polymethoxylated flavones (PMFs) have been determined in terms of their relative amounts. The PMFs were extracted into toluene and analysed using reversed phase HPLC with detection at 340 nm. Peak identification was based on the UV-visible spectra and the elution order described in the literature. Hand-squeezed orange juices contained a mean of 0.10 (SD 0.04) mg(-1) sinensetin with the highest concentrations found in Pera and Natal varieties. Commercial samples of frozen concentrated orange juice (FCOJ), frozen concentrated pulp-wash (FCOPW), retail FCOJ and retail freshly squeezed orange juice (FSOJ) typically contained at least ten times more sinensetin than those found for samples squeezed by hand. The PMFs peak area ratios for these sample classes were examined further using canonical discriminant analysis. This procedure could distinguish the hand-squeezed juices of Pera and Hamlin varieties from those of Natal and Valencia. Similarly, hand-squeezed juices could be readily distinguished from the commercial samples of FCOJ, FCOPW, retail FCOJ and retail FSOJ. (C) 1998 Elsevier Science Ltd. All rights reserved.

Isotopic analysis was used to characterize authentic samples of orange juice (Citrus sinensis) from Brazil. Site specific natural isotopic fractionation nuclear magnetic resonance (SNIF-NMR) was used to determine deuterium/hydrogen ratios at the methyl [(D/H)(I)] and methylene [(D/H)(II)] sites of ethanol produced by fermentation of orange juice. Stable isotope ratio mass spectrometry (SIRMS) was used to determine the ratio of carbon isotopes (C-13/C-12) in the same ethanol and the ratio of oxygen isotopes (O-18/O-16) in the citrus juice water. The mean ratios found for these parameters in authentic hand-squeezed orange juice were as follows: (D/H)(I), 102.3 ppm (SD = 1.7); (D/H)(II), 126.5 ppm (SD = 1.8); C-13/C-12, delta(13)C = -26.6 parts per thousand PDB (SD = 0.9); and O-18/O-16, delta(18)O = +2.27 parts per thousand SMOW (SD = 2.48). Retail samples taken from the Brazilian market place were evaluated by comparison against these data. No evidence was found for the addition of sugar to orange juice or for the dilution with tap water of samples labeled as freshly squeezed.

Multi-element analysis of 112 Spanish and English wines by inductively coupled plasma mass spectrometry (ICP-MS) was undertaken to ascertain whether or not this method could provide data for determining the region of origin of wine. Flow injection analysis proved superior to continuous nebulisation. Quality assurance of these measurements proved entirely satisfactory and was typical of that previously established for this technique. An intercomparison exercise with another expert laboratory showed satisfactory agreement.The data was examined using the statistical technique of discriminant analysis. This was able to unequivocally identify the region of origin of Spanish wines from three different regions. It was also possible to completely differentiate English and Spanish white wines. If red and rose wines were included in the Spanish set, the English and Spanish populations could be distinguished with 95% accuracy.This preliminary study has indicated the power which multi-element analysis can bring to determining the region of origin of wines. At present, no other technique has the ability to perform this categorisation. (C) 1997 Published by Elsevier Science Ltd.

The authenticity of single seed vegetable oils which utilise the C3 photosynthetic pathway was investigated using gas chromatography-combustion-stable isotope ratio mass spectrometry (GC-C-SIRMS). Samples of authentic groundnut, palm, rapeseed and sunflower oils were derivatised to form fatty acid methyl esters (FAMEs) and their carbon isotope ratios (13C12C) determined. In-house reference materials (IHRMs) and internal standards were used routinely to monitor the extraction procedure and SIRMS measurement. These materials demonstrated the consistent performance of the technique. δ13C%. data for the authentic vegetable oil fatty acids fell into the narrow range of −27.6%. to −32.1%. However, the values within the oil varieties considered were significantly different. The data from sunflower oils were such that they could be separated from the other varieties by canonical discriminant analysis. The determination of fatty acid carbon isotope ratios may therefore provide an additional indication of the varietal authenticity of oils which use the C3 photosynthetic pathway.

The new N-nitroso compound identified in smoked bacon is 2-(hydroxymethyl)-3-nitrosothiazolidine-4-carboxylic acid. Identification was based on comparison of data from liquid chromatography, capillary gas chromatography and mass spectrometry with those for the authentic compound, the synthesis of which is described.

Bacon has been cooked in air and in nitrogen atmospheres and the effect on the formation of N-nitrosodimethylamine and N-nitrosopyrrolidine has been studied. Under nitrogen, reductions of 50%–90% were obtained in the concentrations of both nitrosamines in the cooking vapour. The results are discussed in terms of the likely role of nitric oxide in nitrosamine formation in cooking bacon.

AbstractThe reactions of sodium nitrite, S‐nitrosocysteine hydrochloride, S‐nitrosoglutathione and a ‘protein‐bound nitrite’ model system with N‐methylaniline at pH 5.5 and 37°C have been compared. The results show that the rate of nitrosamine formation from sodium nitrite and S‐nitrosocysteine is greater than that from the model system although the latter depends upon the local concentration of nitrosothiol groups on the matrix. Transnitrosation by S‐nitrosoglutathione is considerably slower than the other reactions. The implications of the results with respect to nitrosamine formation in cured meat are discussed.