Modelling of combined mass transfer-kinetic effects in an enzyme membrane reactor system with a direct approach to the identification of intrinsic rate parameters

A methodology for identifying non-linear parameters in a generalised Briggs-Haldane rate expression for immobilised enzyme membrane flow reactors has been developed for the intermediate case where mutually rate-controlling processes of mass transfer and enzymatic reaction prevail. The technological significance of the estimation procedure lies in the ability of the algorithm to handle parameter identification in non-linear, two-point boundary value problems which are prevalent in both immobilised enzyme and non-enzymic heterogeneous catalysed reactor systems. The methodology was successfully implemented on an experimental level using quasi-steady-state conversion data for the model system of urea/collagen-urease in a spiral-wound, biocatalytic reactor module. By using a novel procedure of pepsin pretreatment of the collagen host membrane before enzyme impregnation, followed by glutaraldehyde crosslinking, it was possible to enhance the activity and stability of the immobilised enzyme complex to relatively high levels where interphase and intraphase diffusional limitations mutually controlled the rate of reaction. Information on the coefficient of mass transfer at the interface between solid and fluid phases was also obtained as part of the estimation procedure.     

Factors governing the activity of lyophilised and immobilised lipase preparations in organic solvents

Active site titration and activity measurements were performed in hexane on lyophilised lipase preparations containing different amounts of phosphate buffer and lipase immobilised on porous polypropylene. Lyophilisation of Thermomyces lanuginosus lipase with large quantities of phosphate salts (200 mM) increased the specific activity fourfold, and the number of rapidly titratable active sites increased to 50 % from the 13 % observed when smaller amounts of phosphate buffer were used (20 mM) during lyophilisation. The phosphate buffer worked as an immobilisation matrix for the lipase, and the increase in specific activity was at least partly due to decreased mass transfer limitations. When lipase was immobilised on porous polypropylene, the specific activity was 770 times higher than that of the best freeze-dried preparation. At optimal enzyme loading, 93 % of the enzyme molecules were titrated at a high rate; this indicates that this adsorption on a hydrophobic surface was a very efficient means of reducing mass transfer limitations and of immobilising the enzyme in its active conformation for use in organic solvents. The variation in specific activity with water activity was found to correlate very well with the variation in titratable active sites when lipases from Burkholderia cepacia and Thermomyces lanuginosus were immobilised on porous polypropylene. The catalytic activity per competent active site was thus constant over the whole range of water activities.     

Enzyme immobilisation within insolubilised gelatin

A method originally developed for entrapping invertase-active yeast cells (Saccharomyces cerevisiae) in gelatin has been extended to enzymes. Six enzymes belonging to three different classes were immobilised in gelatin made insoluble by formaldehyde. The activity yield of the resulting immobilised enzyme preparations ranged from 9.6 to 48.0%. These values, however, can be roughly doubled by crushing the immobilised enzyme particles into very small fragments, thus indicating that independently of enzyme sensitivity to formaldehyde or of enzyme interaction with gelatin, the limiting factor is mass transfer resistance to diffusion of substrate into the insolubilised gelatin matrix.     

Electron Transfer and Stability of a Quinohaemoprotein Alcohol Dehydrogenase Electrode

Quinohaemoprotein alcohol dehydrogenase from Comamonas testosteroni (QH-EDH) was immobilised in a redox polymer network of a polyvinylpyridine, partially N-complexed with osmiumbis(bipyridine)chloride. Substrate-dependent electron transfer occurred, indicating that the enzyme was active and that effective electron transport was achieved. It was shown that the enzyme molecular weight is of importance with respect to the enzyme electrode stability. Long term stability and current density of the QH-EDH electrodes were highest when the enzyme was immobilised at pH 10·0 and 4°C, followed by an additional cross-linking step with glutaraldehyde (1%) at pH 7·0. With such an electrode current densities of 40 μA cm⁻² were obtained for several primary alcohols. The affinity of the immobilised enzyme for these substrates (K_m(app) values) was similar to that of the enzyme in solution. The half-life time of the electrodes was between 50 h and 200 h depending on the time the enzyme was in contact with the substrate. When the immobilised enzyme electrode was operated at temperatures above 37°C the stability decreased. © 1997 SCI.     

Design of immobilised glucoamylase reactors using a simple kinetic model for the hydrolysis of starch

A comparative study is reported of the three main types of continuous reactors, continuous feed stirred tank reactor (CFSTR), packed bed reactor (PBR) and fluidised bed reactor (FBR), for the conversion at 50°C of a 30% (w/w) corn syrup 40 DE using glucoamylase (exo-1,4-D-glucosidase, EC 3.2.1.3) immobilised on carbonyl derivatives of titanium(IV)-activated porous silica by a method developed previously.^1–3 The hydrolysis of starch in these reactors is described by a simple kinetic model⁴ which involves the intrinsic kinetic constants as well as mass transfer and dispersion effects, and allows the computation of enzyme activity values under continuous operation. FBR appears to be more effective for the hydrolysis of starch. This reactor also confers a better operational stability (t_1/2=775h) on the immobilised enzyme than the other immobilised glucoamylase reactors (PBR, t_1/2=629h; CFSTR, t_1/2=239h).     

Studies of L-phenylalanine production by immobilised aminoacylase in stabilized calcium alginate beads

Aminoacylase I (EC 3.5. 1.14) was immobilised by entrapment in uncoated calcium alginate beads and calcium alginate beads coated with chitosan, polyethyleneimine and polyethyleneimine-glutaraldehyde for the production of L-phenylalanine by the hydrolysis of a racemic mixture of N-acetyl-DL-phenylalanine. The operational stability, thermal stability, effects of pH and temperature and kinetic constants, K_m and V_max values of free and immobilised enzymes were studied. Scanning electron micrographs revealed differences in the structures of the surfaces of coated and uncoated beads. Chitosan-coated calcium alginate beads was found to be the best among the immobilised systems studied. The activity and the optimum temperature of immobilised aminoacylase were higher than those of the free enzyme. In addition, the beads showed stable activity under operational conditions. The immobilised aminoacylase lost about 20% of its initial activity after ten cycles of reuse. Polyethyleneimine and polyethyleneimine-glutaraldehyde treatments were also found to enhance the operational stability of the enzyme but its activity was greatly reduced.     

Modelling of phenol biodegradation by Candida tropicalis immobilised in alginate gel beads

Phenol‐degrading yeast Candida tropicalis were immobilised in alginate gel beads and photographed by scanning electron microscopy. Batch phenol biodegradation experiments were done in shaking flasks under varying conditions such as initial phenol concentrations and bead loadings. A mathematical model was proposed to simulate the batch phenol biodegradation process in the immobilised system, which took into account the internal and external mass transfer resistances of phenol and oxygen and the double‐substrate phenol–oxygen intrinsic kinetics. The validation of this model was done by the comparison between the model simulations and the experimental measurements of phenol concentration profiles in the main liquid phase. Moreover, the time and radius courses of phenol, oxygen, and cell concentration profiles within the alginate gel beads were reasonably predicted by the proposed model.     

Properties of native and immobilised preparations of β-D-glucosidase from Aspergillus niger

The enzyme β-D-glucosidase from Aspergillus niger has been immobilised through its carbohydrate moiety on concanavalin A-Sepharose and on cyanogen bromide-activated Sepharose after aminoalkylation of the carbohydrate side chains of the enzyme. For comparison, the enzyme was also immobilised on microcrystalline cellulose through its protein moiety. High retention of activity and a decrease in K_m and V_max. were observed when β-D-glucosidase was immobilised by these methods. An increase in the thermal stability of the immobilised β-D-glucosidase preparations over the soluble enzyme was achieved if it was treated with glutaraldehyde before its adsorption on concanavalin A-Sepharose or if the enzyme immobilised on cyanogen bromide-activated Sepharose was subsequently treated with glutaraldehyde. Treatment of β-D-glucosidase immobilised on microcrystalline cellulose with glutaraldehyde hardly increased its thermal stability over the soluble enzyme.     

Invertase immobilised on activated clay: Properties and kinetic study

Invertase was immobilised on activated ball clay. The immobilised enzyme showed maximum activity at pH 4.8–6. The thermal deactivation was significant at temperatures above 30°C. The kinetics of sucrose hydrolysis with free invertase showed substrate and product inhibition. Immobilised enzyme kinetics followed the same mechanisms. Modelling curves were plotted to explain the experimental results. The thermal deactivation followed first order kinetics. The characteristics of thermal deactivation of immobilised enzyme were studied in the range 25–50°C.     

Mass transport characterization of a novel gas sparged photoelectrochemical reactor

The photoelectrochemical treatment of waste water using immobilised TiO₂ electrodes has been demonstrated to be an attractive alternative to TiO₂ slurry reactors; however, it is generally believed that the diffusion of species to the surface of the catalyst imposes severe mass transfer limitations and hence is a disadvantage of the photoelectrochemical approach. To challenge this view, this paper reports the characterization of the mass transport properties of a novel gas sparged photoelectrochemical reactor. It is shown that passing a constant stream of nitrogen gas through the reactor increased the mass transfer coefficient by an order of magnitude above that in the absence of gas and this was attributed to the turbulent flow regime imposed by rising gas bubbles. It is also demonstrated that the gas–liquid transfer coefficient was greater than that for the rate of liquid diffusion and this has important implications for heterogeneous processes.     

Immobilisation of β-galactosidase onto gelatin by glutaraldehyde and chromium(III) acetate

β-Galactosidase (EC 3.2.1.23) was immobilised onto a gelatin carrier system by crosslinking. Glutaraldehyde and chromium(III) acetate were used as crosslinking agents. The effect of pH on enzyme activity, kinetic parameters and reusability of the immobilised enzyme were investigated. To obtain a better relative activity, a method was developed and further modified to increase the surface area of the support system by using polyacrylamide. Relative activity enhancement of 68% was achieved in the gelatin–glutaraldehyde system.     

Characterization and optimization of an oscillatory baffled reactor (OBR) for ozone-water mass transfer

Ozone-water mass transfer was investigated using an oscillatory baffled reactor (OBR) operated as a semi-batch and as a co-current up flow continuous reactor. The effects of input ozone concentration, input gas and water flow rates, and oscillation conditions on gas hold up, volumetric mass transfer coefficient and mass transfer efficiency were determined. The same reactor was operated as a baffled column (without oscillation) and as a bubble column to assess the effect of the reactor arrangement on the mass transfer. The results show that the OBR was 5 and 3 times more efficient for ozone-water mass transfer than the baffled and bubble columns, respectively. The enhancement obtained with OBR over the baffled column reactor was found to decrease with gas flow rate due to changes in bubble flow pattern from homogenous to heterogeneous. Under continuous flow conditions, the performance of the baffled reactor and the OBR were found to be twice efficient for ozone-water mass transfer than when operating under semi-batch conditions. The mass transfer effeciency (MTE) was found to increase from 57% using the baffled reactor to 92% with OBR under continuous flow at water and gas superficial velocities of 0.3 and 3.4 cm s⁻¹, respectively.     

Physiology of amidase production by M. methylotrophus: Isolation of hyperactive strains using continuous culture

Penicillium chrysogenum spores were immobilised in K-carrageenan. The effect of the number of viable spores immobilised per bead on the rate of germination and degree of subsequent mycelial growth was investigated. The distribution of active mycelium throughout the bead was determined. At a high spore loading (10³?10⁴ viable spores per bead) the biomass concentration was low and the majority of the actively respiring biomass was located at the bead periphery. Reducing the spore loading (to 50 viable spores per bead) resulted in a fourfold increase in immobilised biomass concentration. At very low spore loadings (5 and 10 viable spores per bead) the concentration of biomass decreased, but mass transfer throughout the bead improved and the uniformity of active immobilised biomass increased. The spore loading also affected the morphology of the growing hyphae and the extent of free cell growth.     

Production of specific-structured lipids by enzymatic interesterification in a pilot continuous enzyme bed reactor

Production of specific-structured lipids (interesterified lipids with a specific structure) by enzymatic interesterification was carried out in a continuous enzyme bed pilot scale reactor. Commercial immobilized lipase (Lipozyme IM) was used and investigations of acyl migration, pressure drop, water dependence, production efficiency, and other basic features of the process were performed. The extent of acyl migration (defined as a side reaction) occurring in the present enzyme bed reactor was compared to that in a pilot batch reactor. The continuous enzyme bed reactor was better than the batch reactor in minimizing acyl migration. Generally the former produced about one-fourth the acyl migration produced by the latter at a similar extent of incorporation. Pressure drop and production efficiency were evaluated in order to obtain a suitable yield in one reaction step. High incorporation was favored by high substrate ratios between acyl donors and oils, requiring long reaction times on the enzyme bed. Under these conditions, the pressure drop of the reactor was modeled statistically and theoretically. Residence time, water content, and effects of mass transfers were also investigated. Incorporation of medium-chain fatty acids increased with increased residence time. Approximately 40% of lipase activity was lost after a 4-wk run. External mass transfer was not a major problem in the linear flow range, but internal mass transfer did impose some transfer limitations.     

Invertase immobilised on montmorillonite: reusability enhancement and reduction in leaching

Invertase was immobilised on microporous montmorillonite K-10 via adsorption and covalent binding. The immobilised enzymes were tested for sucrose hydrolysis activity in a batch reactor. K_m for immobilised systems was greater than free enzyme. The immobilised forms could be reused for 15 continuous cycles without any loss in activity. After 25 cycles, 85% initial activity was retained. A study on leaching of enzymes showed that 100% enzyme was retained even after 15 cycles of reuse. Leaching increased with reaction temperature. Covalent binding resisted leaching even at temperatures of 70 °C.     

Kinetics study of Bungarus fasciatus venom acetylcholinesterase immobilised on a Langmuir-Blodgett proteo-glycolipidic bilayer

This study deals with the kinetics properties of an enzyme immobilised in a defined orientation in a biomimetic environment. For this purpose, acetylcholinesterase (AChE) was captured at the surface of a nanostructured proteo-glycolipidic Langmuir-Blodgett film through specific recognition by a noninhibitor monoclonal antibody (IgG) inserted in a neoglycolipid bilayer. Modelling of this molecular assembly provided a plausible interpretation of the functional orientation of the enzyme. The AChE activity being stable for several weeks, the enzyme kinetics were investigated, and fitted perfectly with heterogeneous biocatalytic behaviour representative of cellular enzymatic catalysis. The AChE-IgG-glycolipid nanostructure was directly interfaced with an efficient optical device. Such an association, leading to an intimate contact between the nanostructure and the biochemical signal transducer, gives direct access to the intrinsic AChE behaviour. This study thus demonstrates the potential for direct investigation of the kinetic behaviour of an immobilised enzyme on a lipid bilayer through an efficient transduction system.     

The preparation of an immobilised glucose isomerase II. Immobilisation and properties of the immobilised enzyme

Glucose isomerase ex Lactobacillus brevis was successfully immobilised on microcrystalline cellulose, using the transition metal-link method. Immobilisation could be performed over a pH range of 5 to 9, and usually resulted in an apparent specific activity increase. The immobilised glucose isomerase generally displayed properties similar to those of the soluble enzyme, with the exception of the following differences:

• (i) a pH optimum at pH = 6, an acid shift of 0.5 units on immobilisation;
• (ii) an optimum reaction temperature at 50 °C, lower than that for the soluble enzyme;
• (iii) on incubation at 4 °C, a retention of 53% of the initial specific activity, when stored in 0.02 M, pH = 7, Tris buffer, after 8 weeks, compared with an apparent activation of the soluble enzyme after 10 and 19 weeks' storage.

Storage properties of the immobilised enzyme at 4 °C in Tris were apparently improved by the presence of Mn⁺⁺ and Co⁺⁺, although associated with some protein release. Storage at 4 °C in water alone, as opposed to Tris, resulted in a more rapid activity loss.     

析湿工况下带亲水层开缝翅片管换热器空气侧传热传质特性

对7个带亲水层开缝型翅片管换热器在析湿工况下空气侧传热传质特性进行了实验研究，分析了翅片间距和入口相对湿度对空气侧传热传质特性的影响。结果表明，随着翅片间距的增加，空气侧的传热传质特性均逐渐减弱。入口相对湿度对空气侧的传热特性几乎没有影响。随着入口相对湿度的增加，空气侧的传质特性逐渐减弱。开发了反映空气侧传热特性的Colburn传热因子Jh和反映传质特性的Colburn传质因子Jm的关联式，所开发关联式在±15%误差范围内分别能涵盖96.6%和81.4%的实验数据，其平均误差分别为5.3%和8.5%。     

Mass transfer and biochemical reaction in enzyme membrane reactor systems—II: Expanded analysis for single enzyme systems: effects of enzyme intermediates,denaturation and elution

A detailed study of the dynamics of a packed-bed reactor containing immobilized enzyme particles is presented. The analysis consists of (i) transient state behavior; (ii) models for interphase and interfacial mass transfer between fluid and solid phases and intraphase mass transfer for the solid phase; (iii) detailed reaction rate model for the Bodenstein intermediates; (iv) mass balances for substrates, Bodenstein intermediates, unoccupied enzyme active sites, and products; and (v) models for enzyme denaturation and elution. The general reactor model consists of a set of nonlinear, coupled, partial differential equations. Numerical solutions of the system equations were obtained, using the discrete-space, continuous-time method of lines and realistic parameter values. A generalized map of the range of validity of the Steady-State Hypothesis was established under conditions where multiple mass transfer gradients were present within the reactor.A detailed analysis of the computational errors was performed. It was conclusively shown that the computer simulation solutions obtained in the analyses were not disguised to any significant degree as a result of employing finite difference approximations to the spatial derivatives.It was shown that the level of “error” involved in invoking the Steady-State Hypothesis depends on the relative magnitude of the kinetic parameters and also on the level of “disturbance” at the reactor inlet (i.e. per cent change in substrate inlet concentration). The “error”, however, did appear to be strikingly insensitive to the magnitude of the resistances to mass transfer, as characterized by the Modified Sherwood Number. It was concluded that, given any complete set of kinetic parameters, a transient, heterogeneous, isothermal reactor model based on the Steady-State Hypothesis may be used for predicting time-varying concentration profiles for minor (i.e., less than 5 per cent change in substrate inlet concentration) “disturbances” at the reactor inlet. The corresponding “errors” would be at an acceptable level (i.e., less than 2 per cent in the concentration and less than 10 per cent in the time lag) under these conditions.Further, various mechanisms for enzyme denaturation and elution were incorporated in the general reactor model. Numerical solutions of the resulting system of partial differential equations were obtained, using hypothetical parameter values. Through extensive simulation research, it was shown that the loss in activity of immobilized enzyme reactors cannot be uniquely ascribed to any one particular set of mechanistic deactivation modes.