Intraparticle diffusion limitations in the hydrogenation of monounsaturated edible oils and their fatty acid methyl esters

Intraparticle diffusion limitation in the hydrogenation and isomerization of fatty acid methyl esters (FAMEs) and edible oils (triacylglycerol, TAG) in porous nickel catalyst was investigated both under reactive and under inert conditions. Under reactive conditions, the diffusion coefficients were determined from the best fits of the model simulations applying the intrinsic reacting kinetics of monounsaturated FAME hydrogenation to experiments under diffusion limited conditions. For 0.02 < P-H2 < 0.50 MPa, an effective hydrogen diffusion coefficient of D-e,D-H2 = (1.6 +/- 0.7) x 10(-10) m(2)/s (T = 443 K) was obtained in Pricat 9910 (sample 1992). TAG hydrogenations, at 373 < T < 443 K and 0.30 < P-H2 < 0.50 MPa, appear to be limited by diffusion limitation of TAG, yielding D-e,D-TAG = (4.5 +/- 2.2) x 10(-13) (T = 373 K), (2.0 +/- 0.9) x 10(-12) (T = 403 K), and D-e,D-TAG = (3.3 +/- 1.1) x 10(-12) m(2)/s (T = 443 K). The temperature dependency of D-e follows the relation D(e)eta/T = 2 x 10(-7) (m(2) Pa)/K (eta is the fluid viscosity, Pa s) for 373 < T < 443 K. Independent determination of D-2 in porous nickel catalyst (Pricat 9933, sample 1992: average particle diameter, d(p) = 10 mu m) was performed with tracer pulse column experiments using the HPLC thigh-performance liquid chromatography) technique, in the absence of reaction. The packed columns (length 0.05 and 0.10 m; diameter 4.6 x 10(-3) m) were eluted with n-octane, 3 x 10(-10) < Phi(v) < 80 x 10(-10) m(3)/s (eta = 0.5 mPa s), and MCT oil (medium C-10-C-12 chain triglyceride, eta =17 mPa s) in the temperature range 313 < T < 353 K. The response curves indicated a regular packing of the column. Intraparticle effective diffusivities were measured with more viscous MCT oil as an eluent: 1.4 x 10(-11) < D-e < 2.7 x 10(-11) m(2)/s for methyl palmitate and 4 x 10(-12), D-e < 7 x 10(-12) m(2)/s for trioleate and triacyl octadecenoic ester (95% confidence limit, T = 313 K). Due to the absence of reaction (hydrogenation of double bonds), the obtained D-e with the HPLC technique is volume averaged and thereby determined by the larger intercrystalline pores (<30% of the total pore volume) only. Moreover, D-e measured under reaction conditions reflected the influence of the micropores, resulting in a 10-fold lower value.