Abstract

Undamaged, uninfected and ripe mango fruits were collected from mango trees grown in Tuxtla Gutiérrez, Chiapas, México. In a first experiment, a Plackett-Burman statistical experimental design was used to study the effect of KH₂PO₄ (0.5 to 6.0 g/ll); (NH₄)₂SO₄(0.5 to 4.0 g/l); ZnSO₄ (0.05 to 1.0 g/l); MgSO₄ (0.4 to 2 g/l); CaCl₂ (0.1 to 1.0 g/l); CoSO₄(0.1 to 1.0 g/l) and the total sugar content of mango pulp (Mangifera indica var. criollo) (80 to 120 g/l) on ethanol production using two yeasts strains (denominated TL-ITTG-01 and TL-ITTG-06) isolated from a Mexican fermented native beverage. Saccharomyces cerevisiae NRRL-Y-2034 was used as reference strain. The variables studied had no significant effect on ethanol production by the yeast strain TL-ITTG-01 and S. cerevisiaeNRRL-Y-2034. Highest ethanol productions were 24.89±0.17 for TL-ITTG-01 and 51.24±3.40 g/l for S. cerevisiae NRRL-Y-2034. Substrate concentration, that is, sugars in the mango pulp, and CoSO₄, however, had a significant effect on the ethanol production by strain TL-ITTG-06 and the highest ethanol production for this strain was 47.92±0.82 g/l. Ethanol production by the strain TL-ITTG-06 was further optimized using a factorial design 2². The maximum ethanol concentration predicted by the model was of 48.56 g/l with 210.0 g/l sugar and 0.1 g/l of CoSO₄, but a fermentation with the aforementioned concentrations resulted in an ethanol production of 52.60±0.77 g/l.

Key words: Ethanol, Plackett-Burman design, mango pulp.