초록
<P>The development of rapid, accurate, and cost-effective technologies for process monitoring is highly desirable in the biofuels sector. Here, a technique based on the combination of Fourier transform mid-infrared (FT-MIR) spectroscopy and partial least-squares (PLS) regression was evaluated as a tool for real-time monitoring of the process for bioethanol production from sucrose by <I>Saccharomyces cerevisiae</I>. Industrial musts composed of juice and molasses of sugar cane and sweet sorghum juice were used in the fermentations. Considering the chemical complexity of the substrates, the PLS models presented excellent external validation results, with root-mean-square error of prediction (RMSEP) values of 0.54, 1.02, 0.66, 1.93, 0.50, and 1.64 g L<SUP>-1</SUP> for sucrose, glucose, fructose, ethanol, glycerol, and cells, respectively. The findings demonstrated that FT-MIR spectroscopy can be applied for real-time monitoring of the process, with potential to reduce costs and provide accurate information for more efficient control of the process.</P><BR>[FIG OMISSION]</BR>