Design and optimization of poly(hydroxyalkanoate)s production plants using alternative substrates.

In this work, we propose a Mixed Integer Nonlinear Programming (MINLP) model to determine the optimal design of a poly(hydroxyalkanoate)s (PHAs) production plant configuration. The superstructure based optimization model considers different carbon sources as raw material: glycerol (crude and purified), corn starch, cassava starch, sugarcane sucrose and sugarcane molasses. The PHA extraction section includes four alternatives: the use of enzyme, solvent, surfactant-NaOCl or surfactant-chelate. Model constraints include detailed capital cost for equipment, mass and energy balances, product specifications and operating bounds on process units. The resulting MINLP model maximizes the project net present value (NPV) as objective function and it is implemented in an equation oriented environment. Optimization results show the sugarcane-enzyme option as the most promising alternative (NPV = 75.01 million USD) for PHAs production with an energy consumption of 22.56 MJ/kg PHA and a production cost of 3.02 USD/kg PHA. Furthermore, an economic sensitivity analysis is performed.

Modeling the bioconversion of starch to P(HB-co-HV) optimized by experimental design using Bacillus megaterium BBST4 strain.

Poly(hydroxybutyrate-co-hydroxyvalerate) (P(HB-co-HV)) is a prominent biopolymer as a potential candidate for use in the biomedical area. Several Bacillus spp. strains show promising characteristics in the use of several carbon sources and are an interesting alternative for the production of P(HB-co-HV). Sewage from the agricultural and food processing industries can be used to obtain abundantly starch as a carbon source for PHA production. The aim of the present study was to optimize by response surface methodology and desirability, the production of PHA by a Bacillus megaterium strain using starch as the sole carbon source. Two optimal conditions were determined without sporulation and were used to perform new experiments to calibrate and validate a mechanistic model, developed to simulate the dynamics of PHA and biomass production. The developed model successfully represents the kinetics of the microorganism. Employing different characterization techniques, it was determined that the PHA produced by the strain is a copolymer composed of different HB:HV proportions. Using starch as the sole carbon source in a minimal salt medium, this work shows the first reports in the literature of: 1) a mathematical model for predicting growth kinetic and PHA production for B. megaterium strain and 2) a Bacillus spp. producing P(HB-co-HV) copolymer.