Modelling and Optimizing of Mashing Enzymes - Effect on Yield of Filtrate of Unmalted Sorghum by Use of Response Surface Methodology

ABSTRACT The effect of commercial enzymes on liquefaction of starch from unmalted sorghum was studied. The effects which these enzymes had on rates of filtration were evaluated. Models were developed, validated and optimized to establish the actions of enzymes, either alone or in combination. Preliminary studies on the sorghum cultivars Safrari, Madjeru and S.35 showed that α‐amylase was the backbone enzyme for starch liquefaction among the enzymes used (α‐amylase, Filtrase, protease and β‐amylase). Models confirmed this observation as α‐amylase individually in its first order (X1) contributed 25, 11 and 17%, and in its sum of first and second orders (X1+X12) contributed a 29, 31 and 36% yield of filtrate for Safrari, Madjeru and S.35 respectively. The ease of starch liquefaction, assessed by summing the first and second orders of individual intervention of all enzymes, was found to be in the order of Madjeru, S.35 and Safrari (79, 70 and 56% of yield of filtrate respectively). The importance of the enzyme combination in starch liquefaction in Safrari, S.35 and Madjeru was shown to be 44, 30 and 21% respectively. Enzyme combinations giving maximal starch liquefaction, as identified from a Doehlert experimental matrix, displayed a similar yield of filtrate (Safrari: 85 mL, Madjeru: 84 mL and S.35: 81 mL) after filtration of a 130 mL mash during 1 h. Validation of the models revealed the model developed for Madjeru was the most reliable (R2 = 0.994), while those developed for Safrari (R2 = 0.987) and S.35 (R2 = 0.976) were slightly less reliable. Model optimization gave theoretical enzyme (Brewers Amyliq TS, Filtrase NLC, Brewers Protease and β‐amylase) combinations of 25 mg, 5.68 mg, 100 mg and 67.4 U for Safrari, 15.06 mg, 0.51 mg, 24.32 mg and 53.8U for Madjeru and 19.01 mg, 6.36 mg, 58.76 mg and 43.48 U for S.35, with a resulting yield of filtrate of 94, 87.7 and 83.8 mL respectively.