(101a) Proposal and Validation of A Colombian Biodiesel Plant Simulation, Operative and Economic Optimization BASED ON Numeric Modelling

This work validates the numeric simulation carried out of typical Colombian Biodiesel production plants. Based on this validation achieved through its comparison with real operative data of national plants, an optimization of operative conditions is developed using two objective functions: an economic one that seeks to maximize economic yields, and another one in which increasing biodiesel productivity is intended. Finally, an economic analysis is done for the operation and construction of the optimized plant

The simulation involves rigorous mathematical models, considering transesterification reaction kinetics approximated to the kind of oil used in the country [1], and it is based on the plant configuration proposed by Zhang et al [2], taking into account as raw materials waste cooking oil and vegetal oil, and employing acid and alkali catalysts for biodiesel production. From this study, Velosa [3] developed and optimized a simulation at the conditions and characteristics proposed from palm oil. The results were employed as base for the adjustment of the simulation comparing it with an industrial level Colombian process. For its development, data supplied by CORPODIB® (Corporación para el Desarrollo Industrial de la Biotecnología y Producción Limpia) [4] ?institution that counts with a biofuels production pilot plant in Bogotá- was used.

After comparing the theoretical process with the industrial, the main differences obtained revolved around the reactors operative conditions and glycerol purification section, subproduct of the reaction. In the adjusted simulation the reactors operate at atmospheric conditions (industrial conditions) instead of the pressures reported by the literature, and the glycerol purification configuration, which consisted of a distillation tower, was replaced by two evaporators. Initially an assessment of the raw materials required was made based on previous research that assures the project viability within the industrial context of the country and the technical specifications of the product. Hence, it was established that methanol and NaOH 1%wt (relative to the oil quantity) must be employed for the transesterification reaction, and HCl must be used in order to remove catalyst from the process.

Once the adjustement of real operative conditions was made, the next step was to carry out the optimization of the working conditions of the plant, trying to maximize two separate objective functions: i) biodiesel production and ii) economic profit taking into account both the capital and manufacture cost. In both cases the configuration Reactor-Separator proposed by CORPODIB was assessed for the transesterification section. This arrangement consists of implementing a decanter after each reactor in order to achieve the separation of biodiesel from glycerol, instead of the original proposal that establishes the Reactor-Reactor sequence without intermediate separation. When the Reactor-Separator simulation was done it was found that it is necessary to add more raw material into the process, and even so a considerable diminishment of FAME (Fatty Acid Methyl Esters) and glycerol production appeared. Due to the previous reasons and as suggestion obtained from the optimization, the original Reactor-Reactor configuration was selected since it grants greater production flows.

The optimization comprehends the three fundamental elements of this type of problem: objective function, constraints and decision variables vector. For the FAME production objective function, the equality constraints are constituted of the fundamental equipment equations: material and energy balances, thermodynamic relations, sum of composition equations and kinetic expressions. The inequality constraints are composed of biodiesel purity specifications according to ASTM D6751, and glycerol concentration minimum specifications required. The decision variables vector is conformed of all of those taking part of the model and are not fixed and those that work as freedom degrees such as the water flow of the washing tower, the molar reflux ratio and the distillate rate of the biodiesel purification tower. On the other hand, the economic profit objective function comprehends two elements: i) investment capital costs, and ii) manufacture costs [5]. Equality and inequality constraints are the same of those in the operative function, but the variables that work as freedom degrees are the reaction operative pressure and the use of ethanol as raw material.

As resolution strategy for the operative optimization problem, the optimization rutine of Aspen Plus® modeling program was used, which uses Successive Quadratic Programming (SQP) Algorithm to solve the problem, sequentially looking for local optimal of equipment until it reaches a previously determined general convergence criteria [6].

The importance of this work is essentially its direct contribution to process engineering taking in notice that conditions and processes of biodiesel production are widely reported in literature, but there is no actual comparison between theoretical data and industrial reality. The final result of this study consists of establishing general guidelines to producers and future investors in this field, besides of a detailed economic evaluation of the optimized plant. As a general result, a validated simulation under Colombian industrial conditions, the optimal operative conditions and associated costs to the optimal configuration is presented. This conclusion pretends to collaborate and to continue with current researches and investigation made in the country referring to biodiesel production from palm oil. The impact of this work is ratified considering that Colombia is the fifth worldwide oil producer and the first in Latin America [7].

References

[1] Darnoko, D. & Cheryan, M (2000). Kinetics of Palm Oil Transesterification in a Batch Reactor. Journal of the American Oil Chemists' Society, 77, (12), 1263-1267. Recuperado el 21 de agosto de 2008 de la base de datos SpringerLink.

si3ea/documentos/documentacion/Biodiesel/Produccion_Biodiesel.pdf

[2] Zhang, Y., Dubé, McLean, D., & Kates, M (2003). Biodiesel production from waste cooking oil :1. Process design and technological assessment. Bioresource Technology, 89, (1), 1-16. Recuperado el 21 de agosto de 2008 de la base de datos ScienceDirect.

[3] Velosa, Federico. Simulación y optimización del proceso utilizado en Colombia para la producción de biodiesel a partir de aceite de palma. Bogotá, 2007, pp. 1-47. Tesis (Ingeniería Químico). Universidad de los Andes. Facultad de Ingeniería. Departamento de Ingeniería Química.

[4] CORPODIB: Corporación para el Desarrollo Industrial de la Biotecnología y Producción Limpia Recuperado el 17 de Octubre de 2008 de http://www.corpodib.com.co

[5] Turton, Richard. Analysis, Synthesis and Design of Chemical Process, 2ed. Prince Hall International, 2003.

[6] Fletcher, R. Practical Methods of Optimization, 1987. John Wiley & Sons, 2da edición, 436 pág.

[7] Federación nacional de productores de palma. Colombia. Recuperado el 13 de Octubre de 2008 de http://www.fedepalma.org.