Experimental Investigation of Poly-β-Hydroxybutyrate Production by Azohydromonas lata: Kinetics and Artificial Neural Network Modeling

Batch culture of Azohydromonas lata was investigated for the production of intracellular poly-β-hydroxybutyrate (PHB). In order to determine the C:N value of the culture media for maximizing the microbial productivity of PHB, different concentrations of glucose and ammonium chloride were used as carbon and nitrogen sources, respectively. The optimal temperature and shaking rate was obtained at 30°C and 180 rpm, respectively. The maximum intracellular PHB concentration obtained was 5.09 g/l, which was 20% (w/w) of the cell dry weight (CDW) after 72 h. Also, the synthesis of PHB was growth associated with the C:N ratio of 153.71. The maximum calculated Yp/s was 0.212 (gr/gr) and the specific production rate value after 12 h was 0.264 g/l/h, with 40 and 50 g/l of glucose concentrations, respectively, with 0.5 g/l ammonium chloride kept constant. The chemical composition of the resulting PHB was analyzed by Fourier transform infrared spectroscopy and proton nuclear magnetic resonance spectroscopy. The Leudeking–Piret model was used for kinetic analysis of the PHB production, the statistical analysis of which was modeled by response surface methodology. An artificial neural network technique was applied for modeling of the microbial production of PHB by A. lata as a function of the glucose concentration and CDW, where the minimum mean square error of the model was 0.0012 and 0.0038 for glucose concentrations of 50 g/l and 40 g/l, respectively, when 0.5 g/l ammonium chloride was kept constant.