Whole-cell hollow-fiber reactor: Effectiveness factors

Analytical expressions, which allow the generation of effectiveness factor graphs for a reactor system employing immobilized whole cells a biocatalyst, are presented. In particular hollow-fiber devices (such as dialysis or ultrafiltration units) are considered. Such devices are analogs to a shell-and-tube heat exchanger. Whole cells are entrapped on the shell side: a nutrient solution is circulated through the tubes, substrate diffuses from the tube side, across the fiber, and into the cell mass on the shell side, where it irreversibly reacts to form product. The product back-diffuses into the circulating nutrient solution. The overall substrate mass-transfer process is hypothesized to be either diffusion limited in the hollow-fiber tube wall and/or the shell-side cell suspension and/or reaction limited at the enzyme sites within the whole cells. The first- and zero-order limits of the Michaelis-Menten rate law are used in generating effectiveness factor expressions. The effectiveness factor is a function of reaction order, Thiele modulus, diffusion coefficient ratio (defined as the effective substrate diffusivity in the hollow-fiber membrane wall divided by the effective substrate diffusivity in the cell suspension), partition coefficient, volume of the cell suspension, and hollow-fiber width. Equations for the effectiveness factor are also detailed when the hollow-fiber mass-transfer resistance is far greater than the cell suspension mass-transfer resistance. An effectiveness factor chart is presented specifically for the commercially available C-DAK 4 dialyzer (Cordis Dow Co., Miami, Florida). In general terms the effectiveness factor expressions are applicable for characterizing diffusion and reaction within a catalytically active cylindrical annulus, Whose inner surface offers a diffusional resistance and whose outer surface is impermeable to reactants. Some generalization of the Thiele modulus is undertaken which serves to draw the asymptotes on the effectiveness factor charts together. Comment is made on the variation of the slope of the effectiveness factor graph and its relation to the change in the observed reaction activation energy. Possible application of the model to the catalytic tube wall reactor is discussed.     

Kinetic study of a hollow-fiber enzyme reactor

Enzymes, such as urease and uricase, were entrapped in three kinds of hollow fibers. The apparent Michaelis–Menten constants K_m(app) obtained for these enzyme reactors were always larger than K_m of free enzyme because of the permeation resistance of substrate across the hollow-fiber membrane. K_m(app) increased with increasing degree of permeation resistance across the membrane by the increase in enzyme concentration. The half-life of the entrapped urease in the continuous reaction system was 60–80% of that of free enzyme. Activation energies of hollow-fiber enzyme reactors were always smaller than that of the free enzyme, because the activation energy of permeation was smaller than that of the enzyme reaction.     

Theoretical analysis of G6P production and simultaneous ATP regeneration by conjugated enzymes in an ultrafiltration hollow-fiber reactor

The performance of an ultrafiltration hollow-fiber reactor, in which the enzymatic synthesis of glucose 6-phosphate from glucose and cofactor ATP and the enzymatic regeneration of ATP from ADP and acetyl phosphate are performed simultaneously, was analyzed theoretically. A simple analytical model in which the liquid flowing in the fiber tubes is assumed to be plug flow, and the radial concentration gradients in the tube and shell sides are both neglected, could simulate the reactor performance with satisfactory accuracy. The simulation elucidated the effects of the reactor configurations and various operational conditions on glucose conversion, ATP recycle number, and space-time yield. If the fiber tubes, through which the permeability of the relevant components such as substrates is high, were packed as much as possible in the reactor, good reactor performance could be expected. Furthermore, with a sufficiently high enzyme concentration, low ATP concentration in the feed solution, and appropriate space velocity, good space-time yield with high glucose conversion and with very high ATP recycle number is theoretically possible.     

Experimental investigation of G6P production and simultaneous ATP regeneration by conjugated enzymes in an ultrafiltration hollow-fiber reactor

Enzymatic synthesis of glucose 6-phosphate from glucose and ATP catalyzed by glucokinase from B. stearothermophilus and enzymatic regeneration of ATP from ADP and acetyl phosphate catalyzed by acetatekinase from B. stearothermophilus were simultaneously performed in an Ultrafiltration hollow-fiber reactor of the multitubular heat-exchanger type. Experimental results of space-time yield, the conversion, and ATP recycle number were in good agreement with the theoretical predictions based on the simple analytical model developed in the preceding article. The best results for space-time yield and conversion were Y(s) = 1.97 mol/m(3) h and X = 92.8%, respectively, under the same conditions, and the best result for the ATP recycle number was N(R) = 2130 under conditions different from those above. However, Y(s) = 1.72 mol/m(3) h, X = 81.2%, and N(R) = 1620 were the results when the space-time yield, conversion, and recycle number were at the highest in combination under the same conditions. Results of long-term operation showed that the apparent remaining activity of the enzyme system was ca. 55% after continuous operation for 16 days, the decrease in the enzyme activity being faster than that expected from their half-life times determined individually in the homogeneous system.     

Alpha-galactosidase production and use in a hollow-fiber reactor.

Soybean milk serves as a base for a variety of beverages designed for consumption in developing countries. Soybean flour contains raffinose and stachyose considered to be responsible for flatulence often associated with these products (J.J. Rackis, D.H. Honig, D.J. Sessa, and F.R. Steggerda, 1970). alpha-Galactosidase, produced on wheat bran, hydrolyzes the galactooligosaccharides of soybean milk.     

A hollow-fiber reactor design for NMR studies of microbial cells

A hollow-fiber membrane reactor was designed and constructed to allow perfusion of entrapped, dense Escherichia coli cells with nutrient medium during examination of cell metabolism using nuclear magnetic resonance (NMR) spectroscopy. Phosphorus-31 NMR spectra of the perfused cells included peaks for nucleoside di- and triphosphates, sugar phosphates, and pH-sensitive peaks for inorganic phosphate. The observed intensity of the lumenal inorganic phosphate peak was found to depend on flow rate, ruling out the use of this peak as a concentration reference. Absolute intracellular pH values obtained from NMR measurements were found to be accurate to 0.2 pH units due to uncertainties in intracellular ionic concentrations. Relative pH values, however, were found to be sensitive to cell energetic status. The response of E. coli intracellular pH following a shift to carbon starvation medium was monitored with a resolution of 3 min. Use of a hollow-fiber reactor for cell containment and perfusion during NMR spectroscopy enables metabolic experiments of longer duration and of greater variety than is possible using standard, nonperfused sample tubes.     

Controlling fermentation of lignocellulose hydrolysates in a continuous hollow-fiber reactor using biosensors

Fermentation of lignocellulose hydrolysates as spent sulfite liquor or as hydrolysate from sulfur-dioxide-treated wood to ethanol has been controlled by using biosensors for glucose and ethanol. Yield and productivity were studied with respect to concentration level of the metabolites in a continuous hollow-fiber reactor. High constant yield was achieved by controlling the glucose to low concentration levels. Reduced productivity were obtained when fermenting at high ethanol concentrations as an effect of inhibition of the yeast cells. The observations emphasize the advantage of controlling the process to favorable concentrations of monosaccharides and ethanol.     

Air-bubbling, hollow-fiber reactor with cell bleeding and cross-flow filtration

Continuous asymmetric reduction of dyhydrooxoisophorone (DOIP) to 4-hydroxy-2,2,6-trimethylcyclo-hexanone (4-HTMCH) was achieved by a thermophilic bacterium Bacillus stearothermophilus NK86-0151. Three reactors were used: an air-bubbling hollow-fiber reactor with cell bleeding and cross-flow filtration, an air-lift reactor, and a CSTR with PAA immobilized cells. The maximum cell concentration of 11.1 g dry wt L(-1) was obtained in an air-bubbling hollow-fiber reactor, while in the other reactors the cell densities were between 3.5 and 4.1 g dry wt L(-1) The optimum bleed ratio was 0.1 at the dilution rate 0.3 h(-1) in the hollow-fiber reactor. The highest viable cell concentration was maintained in the dilution range of 0.4-0.7 h(-1) by a combination of proper cell bleeding and cross-flow filtration. The maximum volumetric productivity of 4-HTMCH reached 826 mg L(-1) h(-1) at the dilution rate 0.54 h(-1). This value was 4 and 2 times higher than those in the air-lift reactor and CSTR, respectively. The increasing viable cell concentration increased the volumetric productivity of 4-HTMCH. A cell free product solution was continuously obtained by cross-flow filtration.     

Comparison of the performance of a hollow-fiber microbe reactor with a reactor containing alginate entrapped cells

Summary Alginate entrapped Pseudomonas denitrificans have been compared with cells confined in the outer space of a hollow-fiber membrane unit with respect to continuous denitrification of water. The hollow-fiber unit had a higher productivity as well as a stability similar to that of the alginate unit. A reduction of cell-leakage in the eluate was found in the hollow-fiber unit. The nitrogen gas produced could be removed by circulating the cell containing fluid over a hydrophobic membrane.     

Hydrolysis of menhaden oil by a Candida cylindracea lipase immobilized in a hollow-fiber reactor

A lipase from Candida cylindracea immobilized by adsorption on microporous polypropylene fibers was used to selectively hydrolyze the saturated and monounsaturated fatty acid residues of menhaden oil at 40 degrees C and pH 7.0. At a space time of 3.5 h, the shell and tube reactor containing these hollow fibers gives a fractional release of each of the saturated and monounsaturated fatty acid residues (i.e., C14, C16, C16:1, C18:1) of ca. 88% of the corresponding possible asymptotic value. The corresponding coproduct glycerides retained over 90% of the initial residues of both eicosapentaenoic (EPA; C20:5) and docosahexaenoic (DHA; C22:6) acids. The half-life of the immobilized lipase was 170 h when the reactor was operated at the indicated (optimum) conditions. Rate expressions associated with a generic ping-pong bi-bi mechanism were used to fit the experimental data for the lipase catalyzed reaction. Both uni- and multiresponse nonlinear regression methods were employed to determine the kinetic parameters associated with these rate expressions. The best statistical fit of the uniresponse data was obtained for a rate expression, which is formally equivalent to a general Michaelis-Menten mechanism. After reparameterization, this rate expression reduced to a pseudo-first-order model. For the multiresponse analysis, a model that employed a normal distribution of the ratio of Vmax/Km with respect to the chain length of the fatty acid residues provided the best statistical fit of the experimental data.     

Transient response method for evaluating the rate constants of reactions over immobilized enzymes

The transient response method was utilized to evaluate the rate constants of reaction over immobilized enzyme. Glucose oxidation catalyzed by the immobilized glucose oxidase in a fixed-bed reactor was selected as an example. A theoretical model including the effects of axial dispersion, film diffusion, and intraparticle diffusion was established for the reactor. The individual rate constant of each elementary step of this enzymatic reaction was determined through direct fitting of the experimental response data to the model.     

Modeling of hollow-fiber capillary reactor for the production of L-Alanine with coenzyme regeneration

Continuous production of L-alanine with conjugated enzyme systems of alanine dehydrogenase (AlaDH) and lactate dehydrogenase (LDH) or alcohol dehydrogenase (ADH) was carried out with NAD regeneration in an ultrafiltration hollow-fiber capillary reactor (HFCR) which was proposed as a test bioreactor with very small scale. In the AlaDH/LDH system, pyruvate is the intermediate product for L-alanine so that an optimal point existed in pyruvate concentration for the production rate of L-alanine. NAD cycling number of 4850 and L-alanine productivity of 61.7 mmol/L h were obtained at the best condition. In the AlaDH/ADH system, however, the substrate inhibition in the AlaDH reaction by pyruvate should be considered and the best results of NAD cycling number and (L)-alanine productivity were 2700 and 13.5 mmol/L h, respectively. In consideration of concentration distribution and mixing in the axial direction on an HFCR, performance of the reactor was theoretically analyzed with a multistage stirred tank reactor model combined with the kinetic model based on all the elementary reactions involved. Although quantitative discrepancy existed in some cases, the present theoretical model could explain experimental results and is expected to be generally applicable to standard hollow fiber reactors.     

Bioreduction of para-chloronitrobenzene in a hydrogen-based hollow-fiber membrane biofilm reactor: effects of nitrate and sulfate

A continuous-stirred, hydrogen-based, hollow-fiber membrane biofilm reactor (HFMBfR) that was active in nitrate and sulfate reductions was shown to be effective for degradation or detoxification of para-chloronitrobenzene (p-CNB) in water by biotransforming it first to para-chloroaniline (nitro-reduction) and then to aniline (reductive dechlorination) with hydrogen (H₂) as an electron donor. A series of short-term experiments examined the effects of nitrate and sulfate on p-CNB bioreduction. The results obtained showed both higher nitrate and sulfate concentration declined the p-CNB bioreduction in the biofilm, and this suggests the competition for H₂ caused less H₂ available for the p-CNB bioreduction when the H₂ demand for the reductions was larger. Denitrification and sulfate reduction intermediates were thought to be potential factors inhibiting the p-CNB bioreduction. Analysis of electron-equivalent fluxes and reaction orders in the biofilm further demonstrated both denitrification and sulfate reduction competed more strongly for H₂ availability than p-CNB bioreduction. These findings have significant implications for the HFMBfR used for degrading p-CNB under denitrifying and/or sulfate reducing conditions.     

Liquid chromatographic determination of penicillins by postcolumn alkaline degradation using a hollow-fiber membrane reactor

A high-performance liquid chromatographic method using a hollow-fiber membrane reactor is described for the determination of penicillins. This method involves separation of penicillins on a C18 column, postcolumn reaction with sodium hydroxide and mercury (II) chloride introduced into the main flow stream using sulfonated hollow-fiber membrane reactors immersed in each solution (4 M sodium hydroxide and 3 X 10(-2) M mercury (II) chloride plus 10(-2) M nitric acid), and detection at 290 nm based on the uv absorbance of the degradation products. At penicillin concentrations of 5 micrograms/ml, within- and between-run precisions (relative standard deviation) were 0.24-2.39 and 1.19-4.13%, respectively. The detection limits of the proposed method were 1-5 ng at a signal-to-noise ratio of 3. The method was applied to assays of ampicillin and its metabolites in human serum and urine.     

Production of tomato flavor volatiles from a crude enzyme preparation using a hollow-fiber reactor

In recent years there has been an increase in the interest in the production of compounds by isolation from natural sources or through processes that can be deemed "natural". This is of particular interest in the food and beverage industry for flavors and aromas. Hexanal, organoleptically known to possess "green character", is of considerable commercial interest. The objective of this study was to determine if the enzyme template known to be responsible for the synthesis of hexanal from linoleic acid (18:2) in tomato fruits could be harnessed using a hollow-fiber reactor. A hollow-fiber reactor system was set up and consisted of a XAMPLER ultrafiltration module coupled to a reservoir. The enzyme template was extracted from ripe tomato fruits and processed through an ultrafiltration unit (NMWC of 100 kDa) to produce a retentate enriched in soluble and membrane-associated lipoxygenase (LOX) and hydroperoxide lyase (HPL). This extract was recirculated through the lumen of the hollow-fiber ultrafiltration unit with the addition of substrate in the form of linoleic acid, with buffer addition to the reaction flask to maintain a constant retentate volume. Product formation was measured in the permeate using solid phase microextraction (SPME) developed for this system. At exogenous substrate concentrations of 16 mM and a transmembrane pressure of 70 kPa, hexanal production rates are in the order of 5.1 microg/min. Addition of Triton X-100 resulted in membrane fouling and reduced flux. The reactor system has been run for periods of up to 1 week and has been shown to be stable over this period.     

Lipase kinetics: hydrolysis of triacetin by lipase from Candida cylindracea in a hollow-fiber membrane reactor

The aptitude of a hollow-fiber membrane reactor to determine lipase kinetics was investigated using the hydrolysis of triacetin catalyzed by lipase from Canadida cylindracea as a model system. The binding of the lipase to the membrane appears not to be very specific (surface adsorption), and probably its conformation is hardly altered by immobilization, resulting in an activity comparable to that of the enzyme in its native form. The reaction kinetics defined on the membrane surface area were found to obey Michaelis-Menten kinetics. The specific activity of the lipase in the membrane reactor was found to be significantly higher than in an emulsion reactor. The activity and stability of the enzyme immobilized on a hydrophilic membrane surface seem not to be influenced significantly by the choice of the membrane material. The hollow-fiber membrane reactor is a suitable tool to assess lipase kinetics in a fast and convenient way.     

Reactive extraction of penicillin G in hollow-fiber and hollow-fiber fabric modules

Penicillin G (Pen G) can be rapidly extracted in hollow-fiber liquid-liquid contactors using N-lauryl-N-trialkylmethylamine (Amberlite LA-2) as the extractant. n-Butylacetate is much better than decanol as a diluent for such an extraction, although decanol can give a partition coefficient four times larger. The overall mass transfer coefficient found is a function of aqueous flow on the lumen side of the fiber, and is less dependent on shell-side flow. In backextraction, the overall mass transfer coefficient determined is only one tenth that of the forward extraction, primarily because the hydrophobic hollow fibers used have a high mass transfer resistance under these conditions. The mass transfer experiments show that hollow-fiber extraction of Pen G is competitive with centrifugal extraction. The prospects for extraction of other fermentation products with hollow fibers can be estimated based on the present study.