MODEL 1: FFA< 2% Alcoholysis TG X DGNaEster−−−−−−−−−−−−−−−eq1 k1f k1r DGNaMethanol DG x−−−−−−−−−−−−−−− eq 2 k2f k2r DG X MGNaEster−−−−−−−−−−−−−−−−−eq 3 k3f k3r MGNaMethanol MG X −−−−−−−−−−−−−−−−−eq 4 k4f k4r MG X GlyNa Ester−−−−−−−−−−−−−−−−−− eq 5 k5f k5r GlyNaMethanol =Glycerol Ester−−−−−−−−−−−−− eq 6 k6f k6r Saponification EsterOH Methanol Salt of Fatty acid −−−−−−−−−−eq 7 k7 TGOH DGSalt of Fatty acid−−−−−−−−−−−−−eq 8 k8 DGOH MGSalt of Fatty acid −−−−−−−−−−−−−eq 9 k9 MGOH GSalt of Fatty acid −−−−−−−−−−−−−eq 10 k10 kif = forward reaction rate constant kir =reverse reaction rate constant Assumption: formation of MG, DG &Glycerol is much faster than MGNa, DGNa, GlyNa and much faster than irreversible saponification. k2f , k2r ≫k1f , k1r , k4f , k4r ≫ k3f , k3r , k6f , k6r ≫k5f , k5r , k2 , k4 , k6 k7 , k8 , k9 , k10 --eq 11 dTG =−k1f ∗TG∗X k1r∗DG∗Ester−k8∗TG∗OH --------eq 12 dt
dDGNa k2r∗DG = ∗X -------eq 13 dt k2∗Methanol dMGNa k4r∗MG∗X = --------eq 14 dt k4f ∗M dGlyNa k6r∗Gly∗X = -------------eq 15 dt K6f ∗Methanol
Normalising the whole equation with the concentration of the alcohol ----------eq 16 dTG 1 =−k1f ∗TG∗X ∗methanol ∗DG∗Ester∗X −k8∗TG∗OH dt K1 dDG =−k3f −k2r ∗DG∗ X ∗Methanol k3r ∗MG∗Ester− k2f ∗MG∗methanol∗ X −k9∗DGk8∗TG∗OH ∗methanol dt K4 K2
------------eq 17
dMG k5r K4f =−k5f −k4r∗MG∗X ∗methanol ∗Gly∗Ester ∗MG∗methanol∗X k10∗MG−k9∗DG∗OH ∗methanol dt K6 K4
--------------------eq 18 dG =−K2−k3∗DG−k4r−k5f ∗MGk6r∗Gly∗X ∗methanol ---------------eq 19 dt dM dE k3r k5f k6f = =k2r∗DG∗methanol k4r∗MG∗methanol ∗MG∗Ester ∗Gly∗Ester ∗Gly∗methanol dt dt K4 K6 K6
---------------------eq 20 dOH dA = =−k7∗Esterk8∗TGk9∗DG k10MG∗OH ∗methanol ------------------eq 21 dt dt NB: this is the first part of the model. The second part will involve Eq 16-21 to be solved analytically and the initial values of the reactants will be used to plot the reaction concentration over a period, t. Literature values of the various reaction rate constants, k will be used. GEPASI and Berkeley Madonna software for Chemical Kinetics will be used to simulate the reaction trend. This trend observed from this simulation of will be compared with the result from literature results of Komers 2002, Noureddiini and Zhu, 1997 and Devender et al 2005.