Parameter Estimation of Bioethanol Production Model by a Genetic Engineered Cyanobacterium

ESTRADA , V.; VIDAL VIDAL, R.; FLORENCIO BELLIDO, F. J.; DÍAZ, M. S.

Pittsburgh

Conferencia; 2012 AIChE Annual Meeting; 2012

American Institute of Chemical Engineers

In the last years biofuels are being investigated as alternative to reduce the dependence on fossil fuels. Ethanol derived from sugar cane and corn grain is the most common renewable fuel. Now, it is possible obtain ethanol as a third generation biofuel through microalgae and cyanobacteria, nevertheless current reported ethanol yields through this process still require improvement to make this technology economically attractive. In this work we present a dynamic model for the production of ethanol by the engineered metabolic strain of the cyanobacterium Synechocystis sp. PCC 6803 (SGE9) constructed by Vidal (2009). Vidal (2009) has constructed a novel strain of the cyanobacterium Synechocystis sp. PCC 6803 able to produce ethanol from CO2 through the inclusion of the genes pdc and adhB of the ethanologenic bacterium Zymomonas mobilis under the control of the gene PetEpromoter, which is a different strategy to that used by Deng & Coleman, 1999 and Dexter & Fu, 2009. The model we propose for ethanol production from this cyanobacterium includes mass balances for biomass, ethanol, nitrate and phosphate. Biomass equation includes a growth limiting function for light and nutrients (NO3 and PO4). The biomass equation also includes a term that takes into account the kinetics of growth inhibition for ethanol toxicity. The decrease in the available light for photosynthesis associated to the increase in cell concentration is also described by the model. We formulate a parameter estimation problem with a weighted least-squares objective function subject to mass balances equations (Estrada et al., 2009). The dynamic parameter estimation problem is solved in GAMS through a simultaneous optimization approach by prior transformation it into a nonlinear programming (NLP) problem by discretizing state variables by orthogonal collocation over finite elements. The NLP is solved with an Interior Point algorithm with Successive Quadratic Programming techniques (Biegler et al., 2002). Data set for parameter estimation were obtained in an experiment performed over 73 hours for SGE9 and wild type strains of Synechocystis grown in batch liquid cultures and include profiles for biomass (Chlorophyll a and total organic carbon), ethanol, NO3 and PO4. Numerical results provide useful insights on ethanol production by the genetic modified cyanobacteria, reproducing a 0.025% (v/v) ethanol yield. References Biegler L.T., Cervantes, A. & Waechter, A. (2002). Advances in simultaneous strategies for dynamic process optimization. Chemical Engineering Science, 57: 575-593. Deng, M. D. & Coleman, J. R. (1999). Ethanol synthesis by genetic engineering in cyanobacteria. Applied Environmental Microbiology, 65: 523-528. Dexter, J. & Fu, J. (2009). Metabolic engineering of cyanobacteria for ethanol production. Energy Environmental Science, 2: 857-864. Estrada V., Parodi, E. & Diaz, M. S. (2009). Determination of biogeochemical parameters in eutrophication models as large scale dynamic parameter estimation problems, Computers and Chemical Engineering: 33, 1760-1769. Vidal, R. (2009). Producción fotosintética de etanol por la cianobacteria Synechocystis sp. PCC 6803. Tesis doctoral, Universidad Nacional de Sevilla.