AOAC Methods in Codex STAN 234 (Preliminary Methods Review)

BARBANO ET Al.: J. ASSOC. OFF. Al"AL. CHEM, (VOL. 73, NO. 6, 1990)

850

dissolved in sulfuric acid, Other reactions in the digestion involve destruction of organic compounds with the release of CO 2 and H 2 0. During the digestion, some of the sulfuric acid is c-0nvertcd to KHS04 (when H2S0 4 reacts with Ki$04). Other reaction products are H20, S02, S, and O when H 2 S0 4 reacts with sodium thiosulfate or salicylic acid ( 15). If the digest crystallizes during or after digestion, it is an indication of too little residual acid. Too little acid at the end of digestion often results in low test results because there is insufficient acid to completely retain all nitrogen. It has been reported that nitrogen loss occurs as the digest approaches the solid state (13). After complete digestion, alkali is added to release gaseous ammonia, which is volatilized by distilla– tion and trapped by absorption in an acid collection solution during distillation. The amount of ammonia collected is then quantitated by titration. A modification of the original Kjeldahl method by Wil– farth in 1885 ( 15) used a metal catalyst to speed digestion. Two popular catalysts arc mercury anti copp~r. Scbwab and Schwab-Agallidis (2) determined Kjeldahl reaction kinetics on aniline with different catalysts. They concluded that mer– cury was superior to copper in catalyzing the Kjeldahl reac– tion. Copper did not have any perceptible catalytic activity on release of nitrogen from aoiline. However, in their re· search, potassium sulfate was not used to elevate the boiling point of the digestion mixture. Mercury is thought to be the better of the 2 catalysts on hard-to-digest materials (3- 6). However, in many studies, researchers did not optimize di– gestion conditions for copper; researchers simply substituted copper for mercury. Other studies that have optimized the digestion parameters for copper catalyst have found copper to be as effective as mercury in catalyzing tbe Kjeldahl reactions (7-9). The International Dairy Federation (JDF) has published a provisional method (16) using copper cata– lyst for Kjeldahl determination of milk protein, which is based on the results of an interlaboratory study ( 17). The concentration of copper used as a catalyst is low, to avoid formation of ammonia-copper complexes that can cause un– derestimatton of nitrogen (I 7). AOAC has adopted the cop– per catalyst in Kjel

been investigated over the years by chemical analysis and theoretical calculations based on milk protein amino acid sequences. One of the more recent investigations proposed that a conversion factor of 6.355 would be more accurate for milk protein (20, 21). It is important for anaJysts to rccogniz" the sensitivity 01 test results to measurement errors within the method. Th innuence of errors in measuremeut of volume of titranl sample weight, and normality of titrant for a 3.3% protein sample are as follows: an error of ±0.05 mL m measuremcn1 of titrant volnme will cause a deviation of 0.0089%protein an error of 0.0005 g in measurement of sample weight will cause a deviation of ±0.0003% protein, and an error or 0.0005N in the normality of the titrant will cause a deviatioJ of :1:0. 0164% protein. Thus, the most critical factor is th~ titrant normality. Of course, all these examples assume tha1 the sampJe was completely digested and no nitrogen was l

Codex Trial Method Review