(345b) Noncatalytic Process for Biodiesel Production: Kinetics Study

In one route proposed for non-catalytic biodiesel production, triglyceride is first hydrolyzed in high-temperature water to free fatty acids. These are then esterified by reaction with alcohol to make biodiesel in the second step. We have investigated this alternative process by studying the uncatalyzed hydrolysis of trioleate and the esterification of oleic acid with ethanol. The experiments were conducted in quartz tube batch reactors to ensure the absence of unintentional metal catalysis, which may have occurred in other studies. The samples were analyzed by high pressure liquid chromatography (HPLC). We explored the effects of temperature, phase behavior, reactant molar ratio, and water content in the feed (for the esterification reaction) and then used these results to determine the kinetics of the reactions.

First, the esterification of oleic acid was studied. The reaction can proceed at non-catalytic conditions well below the critical pressure of ethanol (critical point of ethanol is Tc = 240.9oC, Pc = 61.4 bar), which could lead to a lower cost process compared to a catalyst-free process at supercritical conditions. Three reaction conditions (250oC f = 0.05, 250oC f = 0.26, 230oC f = 0.56) were used to study the variables that have an effect on the esterification reaction. The parameter f is the fraction of the reactor volume that is occupied by the reactant phase at reaction conditions. Low values of f lead to mostly gas phase systems at the reaction temperature. Likewise, high values of f lead to liquid-like reactions. The results at five different ethanol to oleic acid feed molar ratios (1:1, 3:1, 5:1, 7:1, and 10:1) suggested that excess ethanol at 3:1 is enough to get reasonable conversions. Adding more alcohol did not have a significant impact on the conversion. This result has implications for the cost of the process because large excess amounts of alcohol would not be required. Moreover, the non-catalytic esterifcation reaction is more tolerant to water in the feed than is conventional transesterification biodiesel production. Experiments were conducted at five different water contents (1%, 3%, 5%, 10%, and 15% by volume). The conversion did not change significantly up to 3% water. The conversion decreased by 20% at maximum water content studied here. The hydrolysis reaction of trioleate was studied in the same manner to explore the effect of temperature, f, and the molar ratio of water to oil.

A simple kinetics model (first order with respect to each component) was used for the esterification reaction. Reactions were performed at five sets of conditions (150oC, 200oC f = 0.80; 230oC f = 0.56, 270oC, 290oC f = 0.26) where a single phase was expected to exist based upon thermodynamic phase equilibrium calculations. The kinetic rate constants were calculated using non-linear regression, and then the rate constants were used to find the activation energy. The kinetics parameters determined from these experiments were then tested by accurately predicting the results from an independent set of single-phase experiments. However, this kinetics model did not accurately predict results from some reactions conducted with two phases present at reaction conditions. The phase behavior of the system is very important and it must be incorporated directly into the kinetics model to develop a reliable model that can be used for process design and optimization. We will describe our work to this end for both esterification and hydrolysis.