| Milk coagulation is fundamental to cheese manufacturing and is initiated by the addition of a proteolytic enzyme to milk, which causes the hydrolysation of {dollar}kappa{dollar}-casein and produces para-{dollar}kappa{dollar}-casein and macropeptide.; A technique for monitoring and predicting enzymatic hydrolysis of milk using diffuse reflectance of near infrared (NIR) radiation at the wavelength of 820, 940, 1250, and 1650 nm was developed and tested.; Macropeptide concentration which is traditionally referred to as non-protein nitrogen (NPN) was used as an indicator of enzymatic reaction. NPN data were fitted to a first order diffusion model to determine the reaction kinetic constant. It was found that protein concentration, temperature, and enzyme concentration had a significant effect on the diffusion model kinetic constant, k{dollar}sb{lcub}rm f{rcub}{dollar}. The activation energy, E{dollar}sb{lcub}rm a{rcub}{dollar}, for {dollar}kappa{dollar}-casein hydrolysis using the enzyme, chymosin derived from fermentation (CDF), was 49.96 kJ/mol which corresponded to values reported in the literature. The kinetic constant k{dollar}sb{lcub}rm f{rcub}{dollar} was found to be inversely proportional to protein concentration.; The NIR reflectance response was used to develop a model to predict enzymatic hydrolysis. Normalized NPN (NNPN) was found to correlate with several combinations of reflectance ratios. The best fitting model among the combinations which used four wavelengths, had an R{dollar}sp2{dollar} of 83.9%, and a standard error of estimate of 0.08. The best model which used only one wavelength (1250 nm) had an R{dollar}sp2{dollar} of 79.2% and a standard error of estimate of 0.09. Empirical relationships were developed for predicting enzymatic hydrolysis, for the same combinations of reflectance ratios, that improved the prediction of NNPN. The relationship using four wavelengths improved R{dollar}sp2{dollar} to 84.8% with standard error of estimation of 0.08 and the relationship using one wavelength improved R{dollar}sp2{dollar} to 83.1% with a standard error of estimation of 0.08.; A linear regression between the measured enzymatic hydrolysis time constant, {dollar}tau{dollar}, and the induction period measured from reflectance curves, had a correlation of 95.7, 97.0, 98.1, and 97.0% for the wavelengths of 820, 940, 1250, and 1650 nm, respectively. A non-linear regression for predicting the NNPN by substituting the time constant with induction period had a correlation R{dollar}sp2{dollar} of 94.2, 94.81, 95.1, and 94.8% with standard errors of 0.07, 0.07, 0.06, and 0.07, for respective wavelengths.; It was concluded that given the protein concentration, temperature, and enzyme concentration, NIR diffuse reflectance at the wavelengths of 820, 940, 1250, and 1650 nm shows a potential for monitoring the enzymatic hydrolysis of milk. The induction period measured from the diffuse reflectance profile had a strong correlation with the measured enzymatic hydrolysis time constant, {dollar}tau{dollar}. The enzymatic hydrolysis of milk can be predicted using the diffusion model and the induction period obtained from the diffuse reflectance profile of 820, 940, 1250, or 1650 nm wavelength. |
