One of the main challenges in the production of second generation (2G) ethanol is related to characterization and monitoring of hydrolyzates from lignocellulosic biomass, which often contain high quantities of phenolic compounds that inhibits both enzyme and microbial activity. Rapid, efficient, and low-cost technologies for monitoring the phenolic compounds during 2G ethanol production are essential for better control of the pretreatment, hydrolysis and fermentation process. This paper reports on the viability of monitoring phenolic compounds by a spectrometric technique (region of Ultraviolet-Visible – UV-Vis) associated with Partial Least Squares (PLS) regression. For that, synthetic samples were considered in PLS-UV-Vis models (Vanillin, gallic and p-coumaric acids). The best predicting concentrations provided satisfactory accuracy by presenting PLS-UV-Vis models with root mean square error of prediction in cross validation (leave-one-out, LOO) around 8% (RMSECV). Our results indicate that the PLS-UV-Vis models were applicable for monitoring gallic acid in the samples.