Enzymatic determination of sulfite in foods: NMKL interlaboratory study

• Published in: Journal of AOAC International Vol. 76; no. 1; p. 53
• Main Author: Edberg, U. (Swedish National Food Administration, Uppsala, Sweden)
• Format: Journal Article
• Language: English
• Published: England 01.01.1993
• Subjects: AGENT DE CONSERVATION; ALIMENTOS; ENZIMAS; ENZYME; ENZYMES; Food Analysis - methods; FOOD PRESERVATIVES; FOODS; MEASUREMENT; MEDICION; MESURE; NAD; Oxidoreductases Acting on Sulfur Group Donors; Peroxidases; PRESERVADORES; PRESERVATIVES; PRODUIT ALIMENTAIRE; Reproducibility of Results; Scandinavian and Nordic Countries; SULFITE; Sulfites - analysis; SULFITOS; SULPHITES; Scandinavian and Nordic Countries
• ISSN: 1060-3271; 1944-7922
• DOI: 10.1093/jaoac/76.1.53

An enzymatic method for the determination of sulfite in foods was collaboratively studied in Nordic industry and government laboratories. The sulfite in liquid foods or extracts of solid foods is analyzed according to the following principle: Sulfite ions are oxidized to sulfate ions by oxygen in the presence of sulfite oxidase, thereby forming hydrogen peroxide. Hydrogen peroxide is transformed to water by reduced nicotinamide adenine dinucleotide (NADH) in the presence of NADH peroxidase. In this reaction, NAD+ is formed (and NADH is consumed) in amounts proportional to the sulfite concentration. Consumption of NADH can be measured spectrophotometrically at 340 nm. The method was collaboratively tested in 2 separate studies with high and low levels of sulfite tested. Results of both studies are reported here. The study samples consisted of potato flakes, wine, juice, and dried apples containing between 0 and about 960 mg SO2/kg. Eleven laboratories participated in the full study and analyzed 12 samples. Six laboratories analyzed 8 samples in the complementary study. Before statistical evaluation of the collaborative study data, results were adjusted for the time-dependent decrease of sulfite in the case of materials with high sulfite content (dried apples and wine). For 2 blind duplicate samples of wine containing 75 mg SO2/kg, the relative standard deviation for repeatability (RSDr, within-laboratory variation) was 3.9%. Relative standard deviation for reproducibility (RSDR, between-laboratory variation) was 7.6%. For 2 samples of dried apples containing 800 and 960 mg SO2/kg, an RSDr value of 13.3% and an RSDR value of 13.9% were calculated.