Countercurrent multistage fluidized bed reactor for immobilized biocatalysts: III. Hydrodynamic aspects

In Parts I and II of this series we described the modelling, design, and operation of a multistage fluidized bed reactor (MFBR) for immobilized biocatalysts. This article deals with those aspects of the MFBR which are different from single-stage fluidized beds which are operated in batch mode with respect to the solids. The semicontinuous transport of the particles requires perfect mixing of the particles in the reactor compartments, because particles are mainly transported from the bottom of these compartments. A large spread in the physical properties of the biocatalyst particles, especially of both size and density, may cause the particles to segregate into layers with different diameter and/or density. This affects the efficient use of the biocatalyst. The properties of the particles are dependent on the immobilization method. The suitability of different methods for possible future application in the MFBR is therefore compared. Because of segregation, successful use of a biofilm catalyst with a nonuniform thickness of the biofilm is doubtful. Experiments in a small scale reactor (+/- 0.1 m diameter) demonstrated that perfect particle mixing is possible using commercially available biocatalyst particles of uniform density. Co-immobilization of the biocatalyst with glass powder in a gel is a simple and effective method of increasing gel density. High density particles allow high liquid flow rates, and thus an improved external mass transfer can be achieved.The distributor plates, which separate the reactor compartments, must allow unhindered transport of particles. Therefore, the holes in these plates must have a diameter of at least 4.5 times that of the largest particles which are present in the particle mixture used. Furthermore, the plates must be designed such that, when scaling-up the reactor, a uniform liquid distribution over the cross-sectional area of the reactor occurs. Large-scale experiments were not carried out, but published correlations, indicate that particle mixing and a uniform liquid distribution can be accomplished in a large-scale reactor under similar flow conditions.     

Countercurrent multistage fluidized bed reactor for immobilized biocatalysts: II. Operation of a laboratory-scale reactor

In Part I of this series,(1) we derived a model and made simulations for a multistage fluidized bed reactor (MFBR). It was concluded that the MFBR can be an attractive alternative for a fixed bed reactor when operated with a deactivating biocatalyst. In Part II of this series, the design of a laboratory-scale MFBR and its evaluation to investigate the practical feasibility of this reactor type, will be described. Experiments with a duration as long as 10 days were carried out successfully using immobilized glucose isomerase as a model reaction system. The results predicted by the model are in good agreement with the measured glucose concentration and biocatalyst activity gradients, indicating perfect mixing of the particles in the reactor compartments.The diameters of the biocatalyst particles used in the experiments showed a large spread, with the largest being 1.7 times the smallest. Therefore, an additional check was carried out, to make sure that the particles were not segregating according to size. Particles withdrawn from the reactor compartments were investigated using an image analyzer. Histograms of particle size distribution do not indicate segregation and it is concluded that the particles used have been mixed completely within the compartments. As a result, transport of biocatalyst is nearly plug flow.     

Isopropanol biodegradation by immobilized Paracoccus denitrificans in a three-phase fluidized bed reactor

A three-phase bed bioreactor including a mix of immobilized microbes was used to degrade isopropanol (IPA). The immobilization method was studied and cells immobilized with calcium alginate, polyvinyl alcohol, activated carbon, and SiO₂ were demonstrated to be the best immobilization method for the degradation of 90% of 2?g/L IPA in just 4 days, 1 day earlier than with free cells. Acetone was monitored as an indicator of microbial IPA utilization as the major intermediate of aerobic IPA biodegradation. The bioreactor was operated at hydraulic retention time (HRT) values of 32, 24, 16, 12, and 10?hr, which correspond to membrane fluxes of 0.03, 0.04, 0.06, 0.08, and 0.10?L/m²/hr, respectively. The chemical oxygen demand (COD) removal efficiencies were maintained at 98.0, 97.8, 89.1, 80.6, and 71.1% at a HRT of 32, 24, 16, 12, and 10?hr, respectively, while the IPA degradations were 98.6, 98.3, 90.3, 81.6, and 73.3%, respectively. With a comprehensive consideration of COD removal and economy, the optimal HRT was 24?hr. The results demonstrate the potential of immobilized mixed bacterial consortium in a three-phase fluidized bed reactor system for the aerobic treatment of wastewater containing IPA.     

Citric acid production by immobilized Aspergillus niger in a fluidized bed reactor

Summary Citric acid production by immobilized of Aspergillus niger in a fluidized bed reactor was performed, evaluating the productivity and the stability of the process when pulsing device was used. The application of a pulsing flow to fluidized bed reactor and the feed nitrogen limited allow to control of bioparticles morphology avoiding bed compactation. When operated at optimum pulsation frequency (0.3 s^–1) the stability of the bioreactor was maintained for more than 30 days, increasing the citric acid production in more than 52.2%.     

Mathematical model for the fluidized bed biofilm reactor

A mathematical model is proposed for the fluidized bed biofilm reactor (FBBR). For individual biofilm-covered particles (bioparticles) within the reactor, an analysis of intrabiofilm mass transfer and simultaneous intrinsic zero order reaction yields an effectiveness factor expression which is a function of the modified, zero order Thiele modulus, Φ_0,m. This expression is linked to a one-dimensional reactor flow model and a fluidization model to yield an overall reactor model describing convective transport and simultaneous biochemical conversion of substrate within a FBBR. For Φ_0,m<1.15, FBBR is mass transfer limited and 0.45 order kinetics are observed. For Φ_0,m<1.15, mass transfer limitations are insignificant and intrinsic zero order kinetics are observed. A sensitivity analysis using the proposed mathematical model indicates that biofilm thickness and media size are the two most important operating parameters. These two parameters can be optimized simultaneously for a specific application. The proposed model provides a rational approach for FBBR design.     

Degradation kinetics of phenol by immobilized cells of Candida tropicalis in a fluidized bed reactor

Degradation kinetics of phenol by free and agar-entrapped cells of Candida tropicalis was studied in batch cultures. The initial phenol degradation rate achieved with free cells was higher than that obtained with immobilized cells, when phenol concentrations up to 1000 mg l^–1 were used. However, at higher phenol concentrations, the behaviour was quite different. The initial degradation rate of the immobilized yeast cells was about 10 times higher than that of the free cells, at a phenol concentration of 3500 mg l^–1. The semicontinuous and continuous degradation of phenol by immobilized yeast cells was also investigated in a multi-stage fluidized bed reactor. The highest phenol removal efficiencies and degradation rates as well as the lowest values of residual phenol and chemical oxygen demand were obtained in the semicontinuous culture when phenol concentrations up to 1560 mg l^–1 were used.     

Hydrogen sulfide removal by immobilized Thiobacillus novellas on SiO2 in a fluidized bed reactor

The removal of hydrogen sulfide (H2S) from aqueous media was investigated using Thiobacillus novellas cells immobilized on a SiO2 carrier (biosand). The optimal growth conditions for the bacterial strain were 30 degrees C and initial pH of 7.0. The main product of hydrogen sulfide oxidation by T. novellus was identified as the sulfate ion. A removal efficiency of 98% was maintained in the three-phase fluidized-bed reactor, whereas the efficiency was reduced to 90% for the two-phase fluidized-bed reactor and 68% for the two-phase reactor without cells. The maximum gas removal capacity for the system was 254 g H2S/m3/h when the inlet H2S loading was 300 g/m3/h (1,500 ppm). Stable operation of the immobilized reactor was possible for 20 days with the inlet H2S concentration held to 1,100 ppm. The fluidized bed bioreactor appeared to be an effective means for controlling hydrogen sulfide emissions.     

Ethanol fermentation in a magnetically fluidized bed reactor with immobilized Saccharomyces cerevisiae in magnetic particles

Ethanol fermentation by immobilized Saccharomyces cerevisiae cells in magnetic particles was successfully carried out in a magnetically stabilized fluidized bed reactor (MSFBR). These immobilized magnetic particles solidified in a 2 % CaCl(2) solution were stable and had high ethanol fermentation activity. The performance of ethanol fermentation of glucose in the MSFBR was affected by initial particle loading rate, feed sugar concentration and dilution rate. The ethanol theoretical yield, productivity and concentration reached 95.3%, 26.7 g/L h and 66 g/L, respectively, at a particle loading rate of 41% and a feed dilution rate of 0.4 h(-1) with a glucose concentration of 150 g/L when the magnetic field intensity was kept in the range of 85-120 Oe. In order to use this developed MSFBR system for ethanol production from cheap raw materials, cane molasses was used as the main fermentation substrate for continuous ethanol fermentation with the immobilized S. cerevisiae cells in the reactor system. Molasses gave comparative ethanol productivity in comparison with glucose in the MSFBR, and the higher ethanol production was observed in the MSFBR than in a fluidized bed reactor (FBR) without a magnetic field.     

Modeling of experiments on biofilm penetration effects in a fluidized bed nitrification reactor

A four-component, diffusion-reaction model with double Michaelis-Menten kinetics was used to describe the experimental data obtained from a laboratory biofilm, fluidized-bed nitrification reactor. Theory and experiment demonstrated that the stoichiometric ratio (3.5 mg O(2)/mg NH(4) (+)-N) can be employed as a criterion to determine whether the limiting substrate is oxygen or ammonia. For the present work, in the range of concentrations where limitation occurred, 4 mg/L NH(4) (+)-N and 14 mg/L O(2), the ratio of oxygen to ammonia in the bulk liquid determined which substrate was penetration-limiting-O(2) if <3.5 and NH(4) (+) if > 3.5. Halforder kinetics with respect to the limiting substrate described the apparent overall rates. Simulations provided biofilm concentration profiles which demonstrated the role of the oxygen-ammonia ratio. Experiments indicated that, generally, high NO(2) (-) concentrations can be expected. These depend on the residence time, biofilm area, and oxygen concentration. This dependency was investigated with the model, as was the parametric sensitivity with respect to the saturation constants. Particularly important for the NO(2) (-) levels were the ratios of the saturation constants for oxygen.     

Jute stick pyrolysis for bio-oil production in fluidized bed reactor

Pyrolysis of jute stick for bio-oil production has been investigated in a continuous feeding fluidized bed reactor at different temperatures ranging from 300 degrees C to 600 degrees C. At 500 degrees C, the yields of bio-oil, char and non-condensable gas were 66.70 wt%, 22.60 wt% and 10.70 wt%, respectively based on jute stick. The carbon based non-condensable gas was the mixture of carbon monoxide, carbon dioxide, methane, ethane, ethene, propane and propene. The density and viscosity of bio-oil were found to be 1.11 g/mL and 2.34 cP, respectively. The lower heating value (LHV) of bio-oil was found to be 18.2 5 MJ/kg. Since bio-oil contains some organic acids such as formic acid, acetic acid, etc., the pH and acid value of the bio-oil were found to be around 4 and 135 mg KOH/g, respectively. The water, lignin, solid and ash contents of bio-oil were determined and found to be around 15 wt%, 4.90 wt%, 0.02 wt% and 0.10 wt%, respectively.     

Removal of aqueous phenol using immobilized enzymes in a bench scale and pilot scale three-phase fluidized bed reactor

The main objective of this work was to investigate the removal of aqueous phenol using immobilized enzymes in both bench scale and pilot scale three-phase fluidized bed reactors. The enzyme used in this application was a fungal tyrosinase [E.C. 1.14.18.1] immobilized in a system of chitosan and alginate. The immobilization matrix consisted of a chitosan matrix cross-linked with glutaraldehyde with an aliginate-filled pore space. This support matrix showed superior mechanical properties along with retaining the unique adsorptive characteristics of the chitosan. Adsorption of the o-quinone product by the chitosan reduced tyrosinase inactivation that is normally observed for this enzyme under these conditions. This approach allowed reuse of the enzyme in repeated batch applications. For the bench scale reactor (1.2-l capacity) more than 92% of the phenol could be removed from the feed water using an immobilized enzyme volume of 18.5% and a residence time of the liquid phase of 150 min. Removal rates decreased with subsequent batch runs. For the pilot scale fluidized bed (60 l), 60% phenol removal was observed with an immobilized enzyme volume of 5% and a residence time of the liquid phase of 7 h. Removal decreased to 45% with a repeat batch run with the same immobilized enzyme.     

Biodiesel production in a magnetically-stabilized, fluidized bed reactor with an immobilized lipase in magnetic chitosan microspheres

Biodiesel production by immobilized Rhizopus oryzae lipase in magnetic chitosan microspheres (MCMs) was carried out using soybean oil and methanol in a magnetically-stabilized, fluidized bed reactor (MSFBR). The maximum content of methyl ester in the reaction mixture reached 91.3 (w/v) at a fluid flow rate of 25 ml/min and a magnetic field intensity of 150 Oe. In addition, the MCMs-immobilized lipase in the reactor showed excellent reusability, retaining 82 % productivity even after six batches, which was much better than that in a conventional fluidized bed reactor. These results suggested that a MSFRB using MCMs-immobilized lipase is a promising method for biodiesel production.     

Continuous production of chiral 1,3-butanediol using immobilized biocatalysts in a packed bed reactor: promising biocatalysis method with an asymmetric hydrogen-transfer bioreduction

An asymmetric hydrogen-transfer biocatalyst consisting of mutated Rhodococcus phenylacetaldehyde reductase (PAR) or Leifsonia alcohol dehydrogenase (LSADH) was applied for some water-soluble ketone substrates. Among them, 4-hydroxy-2-butanone was reduced to (S)/(R)-1,3-butanediol, a useful intermediate for pharmaceuticals, with a high yield and stereoselectivity. Intact Escherichia coli cells overexpressing mutated PAR (Sar268) or LSADH were directly immobilized with polyethyleneimine or 1,6-diaminehexane and glutaraldehyde and evaluated in a batch reaction. This system produced (S)-1,3-butanediol [87% enantiomeric excess (e.e.)] with a space time yield (STY) of 12.5 mg h⁻¹ ml⁻¹ catalyst or (R)-1,3-butanediol (99% e.e.) with an STY of 60.3 mg h⁻¹ ml⁻¹ catalyst, respectively. The immobilized cells in a packed bed reactor continuously produced (R)-1,3-butanediol with a yield of 99% (about 49.5 g/l) from 5% (w/v) 4-hydroxy-2-butanoate over 500 h.     

Modeling and simulation of liquid–solid circulating fluidized bed ion exchange system for continuous protein recovery

Liquid–solid circulating fluidized bed (LSCFB) is an integrated two‐column (downcomer and riser) system which can accommodate two separate processes (adsorption and desorption) in the same unit with continuous circulation of the solid particles between the two columns. In this study, a mathematical model based on the assumption of homogeneous fluidization was developed considering hydrodynamics, adsorption‐desorption kinetics and liquid–solid mass transfer. The simulation results showed good agreement with the available experimental results for continuous protein recovery. A parametric sensitivity study was performed to better understand the influence of different operating parameters on the BSA adsorption and desorption capacity of the system. The model developed can easily be extended to other applications of LSCFB. Biotechnol. Bioeng. 2009; 104: 111–126 © 2009 Wiley Periodicals, Inc.     

Mathematical model for a three-phase fluidized bed biofilm reactor in wastewater treatment

A mathematical model for a three phase fluidized bed bioreactor (TFBBR) was proposed to describe oxygen utilization rate, biomass concentration and the removal efficiency of Chemical Oxygen Demand (COD) in wastewater treatment. The model consisted of the biofilm model to describe the oxygen uptake rate and the hydraulic model to describe flow characteristics to cause the oxygen distribution in the reactor. The biofilm model represented the oxygen uptake rate by individual bioparticle and the hydrodynamics of fluids presented an axial dispersion flow with back mixing in the liquid phase and a plug flow in the gas phase. The difference of settling velocity along the column height due to the distributions of size and number of bioparticle was considered. The proposed model was able to predict the biomass concentration and the dissolved oxygen concentration along the column height. The removal efficiency of COD was calculated based on the oxygen consumption amounts that were obtained from the dissolved oxygen concentration. The predicted oxygen concentration by the proposed model agreed reasonably well with experimental measurement in a TFBBR. The effects of various operating parameters on the oxygen concentration were simulated based on the proposed model. The media size and media density affected the performance of a TFBBR. The dissolved oxygen concentration was significantly affected by the superficial liquid velocity but the removal efficiency of COD was significantly affected by the superficial gas velocity. An erratum to this article can be found online at .     

Selection of Type I and Type II methanotrophic proteobacteria in a fluidized bed reactor under non-sterile conditions

Type II methanotrophs produce polyhydroxybutyrate (PHB), while Type I methanotrophs do not. A laboratory-scale fluidized bed reactor was initially inoculated with a Type II Methylocystis-like dominated culture. At elevated levels of dissolved oxygen (DO, 9 mg/L), pH of 6.2–6.5 with nitrate as the N-source, a Methylobacter-like Type I methanotroph became dominant within the biofilms which did not produce PHB. A shift to biofilms capable of PHB production was achieved by re-inoculating with Type II Methylosinus culture, providing dissolved N₂ as the N-source, and maintaining a low influent DO (2.0 mg/L). The resulting biofilms contained both Types I and II methanotrophs. Batch tests indicated that biofilm samples grown with N₂ became dominated by Type II methanotrophs and produced PHB. Enrichments with nitrate or ammonium were dominated by Type I methanotrophs without PHB production capability. The key selection factors favoring Type II were N₂ as N-source and low DO.     

Evaluation of inverse anaerobic fluidized bed reactor for treating high strength organic wastewater

The aim of this work is to evaluate the feasibility of an inverse fluidized bed reactor for the anaerobic digestion of distillery effluent, with a carrier material that allows low energy requirements for fluidization, providing also a good surface for biomass attachment and development. Inverse fluidization particles having specific gravity less than one are carried out in the reactor. The carrier particles chosen for this study was perlite having specific surface area of 7.010 m2/g and low energy requirements for fluidization. Before starting up the reactor, physical properties of the carrier material were determined. One millimeter diameter perlite particle is found to have a wet specific density of 295 kg/m3. It was used for the treatment of distillery waste and performance studies were carried out for 65 days. Once the down flow anaerobic fluidized bed system reached the steady state, the organic load was increased step wise by reducing hydraulic retention time (HRT) from 2 days to 0.19 day, while maintaining the constant feed of chemical oxygen demand (COD) concentration. Most particles have been covered with a thin biofilm of uniform thickness. This system achieved 84% COD removal at an organic loading rate (OLR) of 35 kg COD/m3/d.