Fumaric acid production from hydrolysates of starch-based substrates.

Fumaric acid production by Rhizopus arrhizus from commercial hydrolysates of corn starch (i.e. glucose molasses) was studied at different initial concentrations of glucose (S) and C:N ratios (R) by performing a 3(2) factorial experiment. By using the response surface methodology and statistical analysis, fumaric acid (YF) and mycelial biomass (YX) yields, as referred to the initial concentration of glucose and fumaric acid productivity (PF), were fitted to the only significant first-order effects of S and R with mean percentage errors ranging from 11 to 15%. The resulting empiric models were used to determine the optimal values of S (100-130 g dm-3) and R (150-210 g-atom C per g-atom N) associated with YF and PF varying in the ranges 40-49% and 7-8.5 g dm-3 day-1, respectively. After establishing the validity of these data at the 95% confidence level, an optimal operating condition (S = 120 g dm-3 and R = 150) was further tested using other substrates (i.e. glucose and acid or enzymatic hydrolysates of cassava, corn and potato flours). Statistically significant improvements in the fumaric acid yield and productivity were determined with respect to the predicted values. Since the highest values of YF and PF were obtained from the acid hydrolysates of the starch-based materials and such values were also found to be insensitive to the substrate used (at a probability level of 0.05), the above operating condition might be further employed to minimise fumaric acid production costs as a function of the feedstock used.