Cyanide Formation in Freezer Stored Foods: Freezing of a Glycine and Nitrite Mixture.

Freezing is not always the best way to keep foods safely. Some reactions are known to be accelerated in ice. Furthermore, some other reactions that are not observed in solution are also promoted in ice. We found that the formation of nitrosamines through the reaction of an amine with a nitrite is accelerated in ice. Surprisingly, cyanide is formed through the reaction of glycine with nitrite in ice but not in solution. Amines are present in many kinds of foods. Nitrite is present in vegetables and is used as a food coloring agent and to inhibit the reproduction of Clostridium botulinum. The maximum amount of cyanide formed reaches a dangerous level, and the intake of this formed cyanide in a few tens of cubic centimeters causes some people to get headaches. These facts suggest that hazardous compounds could be generated in frozen processed foods. We report here the formation of cyanide and its possible formation pathway in ice. Finally, we propose a way to prevent cyanide formation in food under frozen conditions.

Acceleration and Reaction Mechanism of the N-Nitrosation Reaction of Dimethylamine with Nitrite in Ice.

Some reactions (e.g., oxidation of nitrite, denitrification of ammonium) are accelerated in freeze-concentrated solution (FCS) compared to those in aqueous solution. Ice is highly intolerant to impurities, and the ice excludes those that would accelerate reactions. Here we show the acceleration of the N-nitrosation reaction of dimethylamine (DMA) with nitrite to produce N-nitrosodimethylamine (NDMA) in FCS. NDMA is a carcinogenic compound, and this reaction is potentially accelerated in frozen fish/meat. The eaction rate of the N-nitrosation reaction becomes fastest at specific pH. This means that it is a third-order reaction. Theoretical pH values of the peak in the third-order reaction are higher than the experimental one. Freeze-concentration of acidic solution causes pH decrement; however, the freeze-concentration alone could not explain the difference of pH values. The theoretical value was obtained under the assumption that no solute took part in ice. However, solutes are incorporated in ice with a small distribution coefficient of solutes into ice. This small incorporation enhanced the decrement of pH values. Using the distribution coefficient of chloride and sodium ion and assuming those of nitrite and DMA to explain the enhancement, we succeeded in estimating the distribution coefficients of nitrite: 2 × 10-3 and DMA: 3 × 10-2.