Mass transfer of terpenes through a silicone rubber membrane in a liquid-liquid contacting system.

Terpenoids are important compounds for the fragrance industry, and recently, biocatalytic methods have been developed to produce them from cheap monoterpenes, such as alpha-pinene oxide. The biotransformation of alpha-pinene oxide using resting cells of Pseudomonas fluorescens NCIMB 11671 produces isonovalal (cis-2-methyl-5-isopropylhexa-2,5-dienal), which is a fragrance. However, this biotransformation has technical problems including the following: alpha-pinene oxide undergoes autoxidation in water and light; it is hydrophobic and relatively toxic to the biocatalyst; and it suffers from product inhibition. Therefore, removal of isonovalal as it is formed should reduce its toxicity and increase volumetric productivity and production yield. Aqueous-organic two-phase (AOTP) systems can be used in the biotransformation of hydrophobic substrates and can protect biocatalysts from toxic substrates and products. However, the formation of stable emulsions makes further downstream processing and continuous operation difficult. One solution to these problems is to use a solid-phase membrane between the aqueous and organic phases in a Membrane Bioreactor for Biotransformation (MBB). Since there are no data in the literature on the behavior of terpenes and terpenoids with solid-phase membranes, or their mass transfer behavior, the objective of this work was to measure these parameters for a wide range of compounds so as to be able to design an MBB, and to gain a greater understanding of their behavior in these types of systems. Organic/aqueous (P (org) (aq)) and membrane/aqueous (P (mem) (aq)) partition coefficients were measured first, and subsequently used to quantify the overall mass transfer coefficients (k(ov)). The overall mass transfer coefficient (k(ov)) of alpha-pinene oxide through the membrane was found to be 2.5 x 10(-5) m x s(-1) using thicknesses of both 250 and 1,000 microm. Extraction kinetics were successfully described using a resistance-in-series model and were controlled by the aqueous boundary layer and/or membrane resistance (k(mem)), while organic film resistance played an insignificant role. Aqueous film resistance (k(aq)) was found to be the limiting step for mono- and diterpenes, and the effect of the hydrodynamics on k(ov) was successfully predicted using a Wilson plot. However, the extraction kinetics of larger terpenes, such as steroids, were influenced solely by the k(mem), suggesting that membrane diffusivity also depended on the size of the permeating molecule. Finally, the influence of other terpene byproducts on the flux of alpha-pinene oxide was investigated and found to decrease the flux into the organic phase by up to 10%.