Bioethanol production from agave juice continuous fermentation process is shown below. In continuous fermentation process, the effects of dilution rate, nitrogen and phosphorus source addition and micro-aeration on growth, and synthesis of ethanol of two native Saccharomyces cerevisiae S1 and S2 strains were studied.

Continuous cultures were carried out in a 3 L bioreactor (Applikon, The Netherlands) with a 2 L working volume. Cultures were started in a batch mode, by inoculating fermentation medium with 3.5 x 106 cells/mL (97±2 %. initial viability) and incubating at 30 °C and 250 rpm for 12 h. Afterwards, the culture was fed with fermentation medium (12 °Brix = 95 ± 5 g/L reducing sugar and 1 g/L of ammonium sulfate). Culture media were sterilized at 121 °C for 15 min.

To reach the steady state in each studied condition, the culture was maintained during five residence times and samples were taken every 6 h. A steady state was reached, when the variation in the concentrations of biomass, residual sugars and ethanol were less than 5%. Data presented on tables and figures are the mean ± standard deviation of three assays at the steady state.

Effect of the dilution rate on S. cerevisiae strains fermentative capability in continuous cultures

Both yeast strains (S1 and S2) were used and fermentation medium was fed at different D (0.04, 0.08, 0.12 and 0.16 h-1) for studying the effect of dilution rate (D) on the kinetic parameters and concentrations of biomass, residual reducing sugar and ethanol at a steady state of agave juice continuous fermentation process (Table 5 and Figure 5).

Concentrations of biomass and ethanol decreased as D increased for both strains cultures while residual reducing sugars increased parallel with the increase of D (Figure 5).

Although, S. cerevisiae S2 consumed more reducing sugars than S1 for each D, ethanol yields reached by S1 were higher than those obtained by S2, which were near the theoretical value (0.51) with no significant differences among the different D tested (p>0.05) (Table 5).

At D = 0.04 h-1, S1 and S2 strains reached the highest ethanol productions (43.92 and 38.71 g/L, respectively) and sugar consumptions (96.06 and 94.07 g/L, respectively) which were similar to those obtained using batch fermentations (see Batch fermentation process section). The low fermentative capacities displayed by both strains at higher D than 0.04 h-1 could be due to a low content of nutrients and/or toxic compounds in agave juice cooked [15].

Both strain cultures reached maximal ethanol production rates at 0.12 h-1 (2.37 and 2.53 g/L-h, respectively for S1 and S2), maximal growth rates were achieved at 0.16 h-1 (0.44 and 0.38 g/L-h, respectively for S1 and S2) and maximal sugar consumption rates were obtained at 0.08 h-1 (5.08 g/L-h) for S1 and at 0.12 h-1 (9.96 g/L-h) for S2 (Table 5 and Figure 6).

Effect of the pH value on the fermentative capacity of S1 and S2 strains — The effect of pH was observed, switching from a controlled pH (at 4) to an uncontrolled pH (naturally set at 2.5±0.3). Figure 7 shows biomass and ethanol productions for strain S1, in non-aerated or aerated (0.01 vvm) systems fed with sterilized medium. Results did not show significant differences on the biomass or ethanol productions (P > 0.05) between the fermentations with control (4) and with no control (2.5) of pH. Conversely, biomass and ethanol productions increased on aerated culture compared to that non aerated, for both pH levels studied. These results agreed with those reported by Diaz-Montano et al. [20]. These results are important, since the operation of a continuous culture naturally adjusted to a low pH would limit the growth of other yeasts [21, 22] or bacteria [23, 24], indicating the feasibility of working with non-sterilized media on an industrial scale. Another advantage of not controlling the pH is that instrumentation for this operation is not required, thus removing it from the initial investment [25].

Table 5. Kinetic parameters at the steady state of continuous cultures of two strains of S. cerevisiae (S1 and S2) fed with agave juice at different dilution rates (D).

phosphorus supplementation on S. cerevisiae S1 sugar consumption

Since both S. cerevisiae strains were unable to consume sugars efficiently in cultures fed at D higher than 0.04 h-1, a nutritional limitation and/or some inhibitory substances formed in the agave cooking step (Maillard compounds), which can act on S. cerevisiae strain activity. In fact, Agave tequilana juice is deficient in nitrogen sources (Table 3). Amino acids are the most important nitrogen source in agave juice; however, their natural concentrations (0.02 mg N/L) are not enough to support balanced yeast growth and the complete fermentation of sugars [26]. Therefore, agave juice supplemented with ammonium sulfate at 1 g/L could be insufficient. Several authors point out the importance of nitrogen sources (type and
concentration) for achieving a complete fermentation, since they improve cell viability, yeast growth rate, sugar consumption and ethanol production (11; 20). It is worth noting that ammonium phosphate (AP) was chosen as a nitrogen source, since the two macronutrientes frequently implied in the causes of stuck fermentation when present in small quantities are nitrogen and phosphate (see the reviews by Bisson [11]).

Therefore, the effect of the ammonium phosphate (AP) addition on S. cerevisiae S1 sugar consumption was studied in a continuous culture (Figure 8). To study the effect of nitrogen and phosphorus source addition on the agave juice fermentation by S. cerevisiae, S1 strain was used and fermentation medium was fed at D of 0.08 h-1, while after the steady state was reached, the ammonium phosphate (AP) concentration was gradually increased, as follows: 1g/L (first addition), 2 g/L (second addition), 3 g/L (third addition) and 4 g/L (fourth addition).

The fermentation was started in batch mode using the fermentation medium. After 12 h, the culture was fed using medium supplemented with 1 g/L of AP (first addition). At the steady state, residual concentrations of sugars and ammonium nitrogen were 29.42 and 0.08 g/L, respectively. These results were not significantly different (p>0.05) from the condition previously tested for the same strain (at D = 0.08 h-1), feeding an unsupplemented fermentation medium (Figure 5).

Those residual concentrations of reducing sugars (high) and ammonium nitrogen (low) indicate the necessity of adding more AP. At the steady states of the second (2 g/L), third (3 g/L) and fourth (4 g/L) additions of AP, the residual sugars concentrations were 25.96, 21.25 and 17.60 g/L, respectively. This indicates that the residual ammonium nitrogen concentrations were 0.31, 0.36 and 1.29 g/L, respectively; indicating that the AP addition improved S. cerevisiae S1 fermentative capability, but other nutritional deficiencies still existed [27].

Effect of the micro-aeration rate on S. cerevisiae S1 fermentative capability — Lack of oxygen has proved to be a main limiting factor to fermentation [11], since yeasts require low amounts of oxygen for synthesizing some essential lipids to assure cell membrane integrity [28]. Because S. cerevisiae is Crabtree-positive, alcoholic fermentation is privileged in culture media containing high sugars concentrations, even in the presence of oxygen [29]. The effect of the micro-aeration rate (0, 0.01 and 0.02 vvm) on the fermentative capacity of S. cerevisiae S1 (at D = 0.08 h-1) was studied for investigating the yeast oxygen requirement during the continuous fermentation, using the last fermentation medium supplemented with 4 g/L of AP for feeding at D of 0.08 h-1. Biomass and ethanol concentrations increased as air flow increased, reaching at the steady state, 5.66, 7.18 and 8.04 g/L, and 40.08, 44.00 and 45.91 g/L, respectively for 0, 0.01 and 0.02 vvm (Figure 9).

Figure 9. Concentration of Residual reducing sugar, Ethanol and Biomass at the steady state of continuous culture of two strains of S. cerevisiae S1 fed with agave juice (D = 0.08 h-1) at different micro-aeration rates. Data are presented as mean ± standard deviation of four assays at the steady state.

Meanwhile, residual sugars decreased as micro-aeration increased, reaching 17.67, 10.71 and 4.48 g/L, respectively for 0, 0.01 and 0.02 vvm; showing an improvement in the fermentation process due the dissolved oxygen in the must. However, statistical differences were not found in biomass and ethanol yields at the different tested aeration rates (p>.05) (Table 6). In addition, sugars consumption rates and ethanol and biomass productions increased as micro-aeration increased, achieving a faster fermentation (Table 6). These results were in accordance to those reported by Diaz-Montano [20]. Viability of the S1 strain was 100% in aeration experiments.

Glycerol is a metabolite providing yeast metabolic activity information. In fact, yeasts produce glycerol mainly for reoxidating the NADH generated by glycolysis. Since the citric acid cycle and the respiratory chain are slightly activated by micro-aeration, NAD might be partially regenerated, and consequently, glycerol concentration decreases [30]. However, in this work, glycerol concentration increased as aeration increased (Table 6). Given that
biomass concentration and fermentation efficiency also increase as aeration increases, glycerol production could contribute to faster NAD regeneration.