Comparison of a batch, fed-batch and continuously operated stirred-tank reactor for the enzymatic synthesis of (R)-mandelonitrile by using a process model

Abstract

The suitability of a batch, fed-batch and continuously operated stirred-tank reactor for the enzymatic production of (R)-mandelonitrile in an aqueous-organic biphasic system was investigated by using a process model. The considered biphasic system is 10±50% (v/v) 100 mM sodium citrate buffer of pH 5.5 dispersed in methyl tert-butyl ether. The constraints were that 750 moles of benzaldehyde per cubic meter should react towards (R)-mandelonitrile with an enantiomeric excess of 99% and a conversion of 98%. A continuously operated stirred-tank reactor could not meet the constraints, but the production in a batch or fed-batch reactor was feasible. The choice for a batch or fed-batch reactor is dependent on the in¯uence of the costs for reactor operation and for the enzyme on the product costs. The choice for operating at a small or large aqueous-phase volume fraction is dependent on the costs and reusability of the enzyme. The volumetric productivity is maximal when operating as batch. The enzymatic productivity and turnover are maximal when operating as fed batch. In the fed-batch mode, the enzymatic productivity increased by 24±37%, the turnover increased by 50±60% and the volumetric productivity decreased by 33±71% as compared to a batch reactor. By enhancement of mass transfer both the volumetric and enzymatic productivity can be increased considerably, while the turnover is only slightly decreased. List of symbolsC aq,j concentration in aqueous phase [mol.m aq ±3 ] C aq,E enzyme concentration [g.m aq ±3 ] C org,j concentration in organic phase [mol.m org ±3 ] C in org;j concentration in entering organic¯ow [mol.m org ±3 ] ee enantiomeric excess of (R)-mandelonitrile [%] F cont org;j¯o w rate continuous reactor [m org 3 .s ±1 ] k f cat forward catalytic constant [s ±1 ] k r cat reverse catalytic constant [s ±1 ] k non-enz forward chemical reaction rate constant [m aq 3 .mol ±1 .s ±1 ] k L,j a lumped mass transfer coef®cient [s ±1 ] K R eq ; K S eq enantiomeric equilibrium constant [m aq 3 .mol ±1 ] K RS eq racemic equilibrium constant [m aq 3 .mol ±1 ] K mj Michaelis constant for compound j [mol.m aq ±3 ] K ij inhibition constant for compound j [mol.m aq ±3 ] m j partition coef®cient [m aq 3 .m org ±3 ] n j total amount of compound j [mol] n 0 j initial total amount of compound j [mol] N fb j feed rate of compound j [mol.s ±1 ] r j reaction rate for compound j [mol.m aq ±3 .s ±1 ] t time [s] t conv time to obtain 98% conversion [s] V aq volume aqueous phase [m 3 aq ] V org volume organic phase [m 3 org ] V tot total reaction volume [m 3 ] g volumetric volumetric productivity [mol P .m ±3 .h ±1 ] g enzymatic enzymatic productivity [mol P .g E ±1 .h ±1 ] / j mass transfer rate [mol.m aq ±3 .s ±1 ] n,n eq extent of conversion of aldehyde [%]

Analysis of Two-Liquid-Phase Multistep Biooxidation Based on a Process Model: Indications for Biological Energy Shortage

“…Gerrits et al exploited mass transfer limitation in a similar system with methyl tert -butyl ether as organic solvent to enhance the enantiomeric excess in the hydroxynitrile lyase-based synthesis of chiral cyanohydrins. , However, such a process for the production of ( R )-mandelonitrile from benzaldehyde and hydrogen cyanide was consistently modeled without considering enzyme activity at the interface but assuming the reaction to occur in the bulk of the aqueous phase and by modeling substrate mass transfer from the organic to the aqueous phase. , A comparative evaluation of the mass transfer model and the adsorbed enzyme model showed that the two mechanisms have qualitatively similar consequences and suggested that both models may be valid simultaneously . Further developed versions of the mass transfer model, which is similar to the model developed in this study, facilitated reaction temperature optimization 38 and allowed comparisons of batch, fed-batch, and continuous modes of operation . Modeling indicated that continuous operation is not feasible and that the choice between batch and fed-batch operation depends on reactor and enzyme costs.…”

“…73 Further developed versions of the mass transfer model, which is similar to the model developed in this study, facilitated reaction temperature optimization 38 and allowed comparisons of batch, fed-batch, and continuous modes of operation. 74 Modeling indicated that continuous operation is not feasible and that the choice between batch and fed-batch operation depends on reactor and enzyme costs. It also facilitated the development of a process for the production of (R)-4-hydroxymandelonitrile from 4-hydroxybenzaldehyde and hydrogen cyanide.…”

Application of modeling and simulation tools for the evaluation of biocatalytic processes: A future perspective

“…For example, the studies of Berendsen et al 25 and Chen et al 26 have performed process analysis using model simulations and suggest the model can be used for process design purposes. Similarly the studies of Illanes et al, 27 Schroen et al, 22 and Willeman et al, 23 use the process models (previously developed based on experimental work) to evaluate different process configurations as well as performance limitations by carrying out simulations. These studies are valuable as they demonstrate how process model simulations can be useful for evaluating process performance with different reactor choices or configurations.…”

Model description of bacterial 3-methylcatechol production in one- and two-phase systems

“…A reliable description of growth of P. putida MC2 and production of 3-methylcatechol is required. Modelling of bioproduction processes in organic/aqueous two-phase systems has been described before, for example for the enzymatic synthesis of (R)-mandelonitrile [7,8] and the enzymatic hydroxylation of phenol [9]. Cesa´rio et al [10] described modelling and validation of a dispersed organic solvent in a three-phase system (organic/aqueous/gaseous) for the biological treatment of waste gases.…”