The results of a collaborative study are reported for the determination of 3-chloro-1,2-propanediol (3-monochloropropane-1,2-diol; 3-MCPD) in a wide range of foods and food ingredients, using gas chromatography with mass spectrometric detection and incorporating the use of a deuterated internal standard. After a pretrial study, 12 laboratories (6 United Kingdom, 1 Switzerland, 1 Japan, 2 United States, 1 The Netherlands, and 1 from the European Commission) were asked to analyze 12 test materials (as known duplicates or split-level samples) by using a prescribed procedure. The test materials consisted of duplicate samples of acid-hydrolyzed vegetable protein (containing 3-MCPD at 0.029 mg/kg), malt extract (0.055 mg/kg), wholemeal bread crumbs (0.030 mg/kg), salami (0.016 mg/kg), cheese alternative (0.043 mg/kg), and soup powder (split levels at 0.045 and 0.041 mg/kg). Repeatability ranged from 0.005 to 0.013 mg/kg and reproducibility, from 0.010 to 0.027 mg/kg, for the samples tested. Precision values were well within statistically predicted levels (HORRAT values of <1 for 5 of the 6 matrixes tested) and within method criteria prepared by a joint working group composed of the United Kingdom Ministry of Agriculture, Fisheries and Food and industry representatives. The study demonstrated the satisfactory validation of the method for quantifying 3-MCPD at levels of ≥0.010 mg/kg. The limit of detection derived from separate in-house studies was estimated to be 0.005 mg/kg. The method was adopted First Action by AOAC INTERNATIONAL.

It is anticipated that high precision analyses via the ICP can be expanded to a wide range of elements and can be applied to a large variety of samples which are currently analyzed by other methods.During the one to three seconds necessary to heat commercial electrothermal atomizers to the desired atomization temperature, many reactions take place, and analyte compounds may be lost from the rapidly heating furnace at varying temperatures with varying matrices-often at sub-optimal temperatures with inadequate atomization, since residence times are short. Thus, matrix interferences are common in these pulsed-type atomizers. However, the same solutions, when atomized in a constant temperature furnace (CTF), show no significant matrix interferences. Lack of ruggedness of analytical procedures using pulsed-type atomizers seems to be an inherent limitation, whereas equipment ruggedness limitations of the CTF are amenable to elimination by appropriate attention to engineering aspects of fabrication. Difficult samples representing common matrices reveal the ease of obtaining interference-free results directly with the CTF-and the difficulty, even with pretreatments, of correcting for interferences on a routine basis in pulsed-type atomizers.

The application of electrothermal atomization techniques for atomic absorption spectroscopy has increased many fold in recent years. Along with this widespread application, matrix interferences in pulse-type electrothermal atomizers have been observed. Until recently, very few interference studies have involved constant temperature electrothermal atomizers. In this paper a comparison is made of interferences in pulse-type vs constant temperature atmoziers for various metal chlorides on the Mn atomic absorption signal. The results show no interferences in the constant temperature system while significant matrix interferences are observed in the pulse type unit. It is proposed that the observed differences are due mainly to the rate at which volatilized analyte attains atomization temperature and the residence time in the hot environment.

An interlaboratory study was performed to evaluate the effectiveness of a headspace gas chromatography (GC) method for the determination of 1,3-dichloro-propan-2-ol (1,3-DCP) in soy sauce and related products at levels above 5 ng/g. The test portion is mixed with an internal standard (d5-1,3-DCP) and ammonium sulfate in a sealed headspace vial. After achieving equilibrium, the headspace is sampled either by gas-tight syringe or solid-phase microextraction (SPME) and analyzed by GC with mass spectrometric detection. 1,3-DCP is detected in the selected-ion mode (monitoring m/z 79 and 81 for 1,3-DCP and m/z 82 for the deuterated internal standard) and quantified by measurement against standards. Test materials comprising soy, dark soy, mushroom soy, and teriyaki sauces, both spiked and naturally contaminated, were sent to 9 laboratories in Europe, Japan, and the United States; of these, 5 used SPME and 4 used syringe headspace analysis. Test portions were spiked at 5.0, 10.0, 20.0, 100.0, and 500.0 ng/g. The average recovery for spiked blank samples was 108% (ranging from 96–130%). Based on results for spiked samples (blind pairs at 5, 10, 20, 100, and 500 ng/g) as well as a naturally contaminated sample (split-level pair at 27 and 29 ng/g), the relative standard deviation for repeatability (RSDr) ranged from 2.9–23.2%. The relative standard deviation for reproducibility (RSDR) ranged from 20.9–35.3%, and HorRat values of between 1.0 and 1.6 were obtained.

Mercury present in water samples Is reduced to the metal in a bubbler and sparged from the solution with a stream of air. The gas stream is filtered through porous, gold-plated graphite furnace atomizer tubes which retain the mercury. The tubes are then placed into either a carbon rod atomizer or a Woodriff furnace for the actual atomic absorption (AA) mercury determination. Each determination takes about 3 minutes. The detection limit Is about 10 parts-per-trllllon.