After most of the available glucose was depleted and the high- est ethanol concentration was reached, which corresponded to
12 h of fermentation, a decrease in ethanol concentration was observed and this was probably due to the consumption of ethanol by the yeasts. The use of untreated elephant grass gave much lower yields compared steam-exploded substrates and the substrate
controls. Hence, this demonstrates the efficiency and importance of steam explosion for the production of ethanol from elephant grass. Furthermore, both substrate controls had fermentation profiles that were very similar to those of steam-exploded substrates. Avicel hydrolysates containing 20.12 ± 0.75 mg/mL of glucose resulted in 4.82 mg/mL ± 0.03 mg/mL of ethanol after 24 h of fer- mentation, while the 18.24 ± 0.44 mg/mL of Celuflock hydrolysates were converted 4.42 ± 1.52 mg/mL of ethanol in the same reaction time. Notably, hydrolysates from untreated elephant grass had a much lower glucose concentration at the beginning of fermentation (4.06 ± 0.18 mg/mL) and this affected the ethanol production from this substrate.