Sparged but Not Stirred: Enzyme-Catalyzed Deracemization of Alcohols in a Bubble Column

Enhanced selectivity and lower environmental footprint often are key benefits of enzymatically-catalyzed reactions over chemical synthesis routes. Alcohol dehydrogenases (ADHs) and analogous enzymes are highly selective enzymes for oxidizing polyol substrates (carbohydrates) and for deracemizing or stereoinverting racemic alcohols to enantiomerically pure alcohols, important life science synthons. The required oxidants, nicotinamide cofactors NAD(P)+, are advantageously regenerated to NAD(P)H via NADH oxidase-catalyzed reduction of molecular oxygen to water. We employed a bubble column sparging air into the enzyme-laden solution to transfer the co-substrate oxygen into the reaction medium. We already found that air-sparged but not stirred solutions are beneficial for sustainable reaction rates combined with low enzyme deactivation.[1] We now investigate oxygen transfer, overall reaction rate, and ADH/NADH oxidase deactivation in the deracemization of (R,S)-1-phenylethanol.[2] Four sparged bubble column conditions were tested, two spherical bubble diameters (2.6 and 2.1 mm) and two sparging rates (18 and 5 bubbles/s), and compared against quiescent and stirred solution. Calculated specific mass transfer coefficients kLa ranged from 17 to 87 h^-1. Observed enzyme deactivation rates descended from stirred and faster bubble flowrates via slower flow rates to quiescent solution. For each sparged condition, 90% conversion or greater was achieved, with optimal 99% conversion in 6 hours with sparging large air bubbles at slow flowrates. Higher bubble flowrates, i.e. more oxygen transfer per unit time, were found to increase initial enzymatic reaction rates but result in lower average rates and ultimately incomplete maximum conversion. Reaction rates under sparging were 4-8 times faster than in quiescent solution. Lastly, stirring did not significantly aid conversion (41%, 37% quiescent solution after 12h) but caused increased enzyme deactivation (44h half-life, 136h half-life quiescent solution).

References:

[1] M Dias-Gomes^¹, B Bommarius^¹, SR Anderson, et al., Adv. Synth. Cat., 2019, 361, 2574-2581

[2] BR Bommarius^¹, SR Anderson^¹, JM Woodley, AS Bommarius, submitted