Modelling of a fluidized bed bioreactor for immobilized enzymes (Part 2)

In the first part of this contribution, a mathematical model was presented for a liquid fluidized bed using immobilized enzymes, with reversible Michaelis-Menten kinetics. This part is focused on the experimental results. The reaction kinetics of native and immobilized enzymes was determined in continuous stirred tank reactors under comparable conditions. The influence of external mass transfer was investigated in a fixed bed reactor column. The extend of pore diffusional resistance was examined in a continuous stirred tank reactor and with a numerical simulation. Hydrodynamics was measured in different reactor columns (diameter d_t = 0.052 ? 0.225 m; length L: 1.0–2.0m) and with a static mixer. Further, the concentration profile was determined in a fluidized bed reactor with side stream analysis for different biocatalyst samples, fluid velocities and bed heights. The simulation of experimental results indicates that they are well described by the developed model. Furthermore, the model is well suited to predicting the influence of specific parameters on the effective kinetics of the biocatalyst and the expansion of the fluidized bed.     

Modelling of fluidized bed reactors—VI(a): An isothermal bed with stochastic bubbles

A two-phase stochastic isothermal fluidized bed reactor model with first order reaction in the dense phase is developed to investigate the significance of the fluctuating nature of fluidized beds on reactor performance. Several stochastic processes are employed as the overall mass transfer coefficient between phases. Analytical moment solutions are obtained for white noise coefficients while hybrid computer simulation was used for correlated stochastic coefficients. Results indicate that a gamma distributed coefficient is preferred over white noise and Gaussian correlated coefficients. When compared with the deterministic model, randomness in the mass transfer coefficient is seen to lead to a decrease in reactor performance. Deviation from the deterministic model increases with increasing variance and decreasing fluctuation frequency of the correlated stochastic coefficients.     

Modelling coal char gasification in a fluidized bed

Predictions obtained by using the accepted reaction model together with three different reactor models — plug-flow, complete-mixing and bubble-assemblage — are compared. In general, the three reactor models gave significantly different results. The bubble-assemblage model, therefore, represents a valid alternative when modelling fluidized-bed gasifiers. For small inverse space velocities, meaning high steam feed rates and shallow beds, the bubble-assemblage model predicts the lowest (among the three) steam conversions. For deep beds and low steam feed rates, the conversion predicted by the bubble-assemblage model becomes greater than that predicted by the completely mixed model and approaches the values for the plug-flow model. However, this occurs only at very large height/diameter ratios which are not generally used in operating fluidized beds. Increase in the reaction rate constant associated with the partial pressure of water, k₁, caused a marked increase in conversion. The effect of k₂,-associated with the partial pressure of hydrogen, was not so great, conversion decreasing as k₂ was increased. Inclusion of the water-gas shift reaction when calculating the gas composition in the kinetic rate expression may be of significance in predicting steam conversion by reactor model 3. However, varying the value of K_eq did not have a significant effect.     

The use of peat granules in a fluidized bed bioreactor

This study describes the particle characteristics and fluidized hydrodynamics of peat granules. Peat granules, moistened with water, are a potential packing material in a gas–solid fluidized bed bioreactor used for treating air pollution. Information on the fluidization of wet peat granules is lacking. In order to advance this new type of bioreactor and to scale up its design for industrial use, fluidization studies of suitable packing material are required. Using abiotic experiments, three sizes of peat granules have been fluidized with air and fluidization characteristics were observed at different superficial gas velocities. Relative to other biomass particles, peat granules have a high particle density and sphericity, which contributes to favourable fluidization behaviour, without gas channelling. Fluidization experiments demonstrate that as the mean size of peat particles increased, minimum fluidization velocity increased. Increasing the moisture content of the peat granules resulted in a transition from bubbling bed fluidization to poor fluidization behaviour. Other types of moist biomass particles such as sawdust are difficult to fluidize and typically exhibit Geldart group C behaviour. In contrast, it was observed that wet peat granules could be fluidized in a bubbling bed regime, typical of group B particles.     

Modelling of circulating fluidized bed combustion of a char

The combustion of a char in the 41 mm ID riser of a laboratory circulating fluidized bed combustor has been investigated at different air excesses and rates of solids (char and sand) circulating in the loop. Riser performance was characterized by an axial oxygen concentration profile as well as by the overall carbon content and particle size distribution. The proposed model accounts for carbon surface reaction, intraparticle and external diffusion, and attrition. External diffusion effects were relevant in the riser dense region where char was potentially entrapped in large clusters of inert solids. Experimental data and results of the model calculations are in satisfactory agreement.     

Removal of hydrogen sulfide in a three phase fluidized bed bioreactor

To remove hydrogen sulfide biologically, a three phase fluidized bed bioreactor was used in whichThiobacillus sp.IW was immobilized on biosands. The optimum operating condition of the bioreactor was found to be 30 ‡C, pH 7, bed height of 0.85–1.0 m and aspect ratio of 1.0. At these conditions, the bioreactor removed more than 88 % of the hydrogen sulfide for an inlet concentration of 30–160 ppm and a gas flow rate of 2–5 Z/m. The maximum removal rate obtained was about 1,000 mg H₂S/min. In continuous operation of the bioreactor, the removal efficiency remained at 99 % for up to 16 hours and decreased to 91 % at 52 hours.     

Modelling of fluidized bed reactors—VI(b): The nonisothermal bed with stochastic bubbles

Two stochastic nonisothermal fluidized bed reactor models are developed to investigate the significance of the fluctuating nature of fluidized beds on reactor performance. Fluctuating bubble size distributions within the bed are simulated by stochastic mass and heat transfer coefficients. Results of hybrid computer simulations indicate that randomness can enhance or inhibit reactor performance depending on the operating parameters of the nonisothermal model. Bubble and dense phase concentration statistics are fairly similar to those of corresponding isothermal models because dense phase temperatures are relatively insensitive to transfer coefficient fluctuations due to the high dense phase beat capacity. However, the corresponding stochastic isothermal models predict decreases in conversion with increasing variance in the transfer coefficients for all operating conditions. Results indicate that a deterministic system with two stable steady states may have fewer stable random stationary solutions. The existence of the stationary states is dependent on fluctuation frequency and variance of the transfer coefficients.     

Modelling fast pyrolysis of biomass in a fluidized bed reactor

A compartmental one-dimensional model of a fluidized bed pyrolytic converter of biomass is presented. Reference conditions are those of non-catalytic fast pyrolysis of biomass in a shallow fluidized bed with external regeneration of the bed material. The fate of biomass and of the resulting char has been modelled by considering elutriation of biomass and char particles, char attrition as well as bed drain/regeneration. The course of primary and secondary pyrolitic reactions is modelled according to a semi-lumped reaction network using well-established kinetic parameters taken from the literature. A specific focus of the present study is the role of the heterogeneous volatile–char secondary reactions, whose rate has been modelled by borrowing a kinetic expression from the neighbouring area of tar adsorption/decomposition over char. The results of computations highlight the relevance of heterogeneous volatile–char secondary reactions and of the closely associated control of char loading in the bed. The sensitivity of the reactor performance to char elutriation and attrition, to proper management of bed drain/regeneration, and to control of gas phase backmixing is demonstrated. Model results provide useful guidelines for optimal design and control of fluidized bed pyrolyzers and pinpoint future research priorities.     

Neural network prediction of fluidized bed bioreactor performance for sulfide oxidation

Sulfide oxidation rate of a fluidized bed bioreactor was predicted using ANN, with upflow velocity, hydraulic retention time, reactor operation time and pH given as input. The reactor was fed with 100mg/L synthetic sulfide wastewater after biofilm formation on nylon support particles. Feedforward neural network model was prepared using 81 data sets, of which 63 were used for training and 18 for testing in a three-way cross validation. Prediction performance of the network was evaluated by calculating the percent error of each data set and mean square error for test data set in three partitions. The mean square error for test data set was 5.55, 4.08 and 2.30 for partition 1, partition 2 and partition 3, respectively. The predicted sulfide oxidation values correlated with the experimental values and a correlation coefficient of 0.96, 0.97 and 0.98 was obtained for partition 1, partition 2 and partition 3, respectively.     

生物流化床处理废水的研究进展

生物流化床技术因具有处理效率高、容积负荷大、传质速度快、应用范围广等优点受到了广泛的关注。介绍了近年来国内外生物流化床技术在废水处理中的应用研究情况,并指出了生物流化床技术目前存在的主要问题及今后的发展方向。     

Modelling of dense gas-particle flow in a circulating fluidized bed by Distinct Cluster Method (DCM)

Computational Fluid Dynamics (CFD) is a powerful tool to study the dense gas-solid flow in a circulating fluidized bed. Most of the existing methods focus on the microscopic properties of individual particle. Therefore, the simulation scale is significantly limited by the huge number of individual particles, and so far the numbers of particles in most of the reported simulations are less than 10⁵. The hydrodynamics behaviour of particle clustering in a dense gas-solid two-phase flow has been verified by several experimental results. The Distinct Cluster Method (DCM) was proposed in this paper by studying the macroscopic particle clustering behaviour, and comprehensive models for cluster motion, collision, break-up, and coalescence have been well developed. We model the dense two-phase flow field as gas-rich lean phase and solid-rich cluster phase. The particle cluster is directly treated as one discrete phase. The gas turbulent flow is calculated by Eulerian approach, and the particle behaviour is studied by Lagrangian approach. Using the proposed method, a three-dimension dense gas-particle two-phase flow field in a circulating fluidized bed with square-cross-section, with particle number up to 7.162 × 10⁷ are able to be numerically studied, on which few results have been reported. Details on instantaneous and time-averaged distributions are obtained. Developing process of non-uniform particle distribution is visualized. These results are in agreements with experimental observations, which justified the feasibility of using the DCM method to model and simulate dense gas-solid flow in a circulating fluidized bed with large number of particle numbers.     

Modelling fluidized bed combustion of high-volatile solid fuels

A model of an atmospheric bubbling fluidized bed combustor operated with high-volatile solid fuel feedings is presented. It aims at the assessment of axial burning profiles along the reactor and of the associated temperature profiles, relevant to combustor performance and operability. The combustor is divided into three sections: the dense bed, the splashing region and the freeboard. Three combustible phases are considered: volatile matter, relatively large non-elutriable char particles and fine char particles of elutriable size. The model takes into account phenomena that assume particular importance with high-volatile solid fuels, namely fuel particle fragmentation and attrition in the bed and volatile matter segregation and postcombustion above the bed. An energy balance on the splashing zone is set up, taking into account volatile matter and elutriated fines postcombustion and radiative and convective heat fluxes to the bed and the freeboard.Results from calculations with a high-volatile biomass fuel indicate that combustion occurs to comparable extents in the bed and in the splashing region of the combustor. Due to volatile matter segregation with respect to the bed, a significant fraction of the heat is released into the splashing region of the combustor and this results in an increase of the temperature in this region. Extensive bed solids recirculation associated to solids ejection/falling back due to bubbles bursting at bed surface promotes thermal feedback from this region to the bed of as much as 80-90% of the heat released by afterburning of volatile matter and elutriated fines. Depending on the operating conditions a significant fraction of the volatile matter may burn in the freeboard or in the cyclone.     

Continuous ethanol fermentation using immobilized yeast in a fluidized bed reactor

Continuous ethanol fermentation of glucose using fluidized bed technology was studied. Saccharomyces cerevisiae were immobilized and retained on porous microcarriers. Over two-thirds of the total reactor yeast cell mass was immobilized. Ethanol productivity was examined as dilution rate was varied, keeping all other experimental parameters constant. Ethanol yield remained high at an average of 0.36 g ethanol g^?1 glucose (71% of theoretical yield) as the dilution rate was increased stepwise from 0.04 h^?1 to 0.14 h^?1. At a dilution rate of 0.15 h^?1, the ethanol yield steeply declined to 0.22 g ethanol g^?1 glucose (44% of theoretical yield). The low maximum percentage of theoretical yield is primarily due to an extended mean cell residence time, and possibly due to the inhibitory effect of a high dissolved carbon dioxide concentration, enhanced by the probable intermittent levels of low pH in the reactor. Constant ethanol production was possible at a high glucose loading rate of 840 g dm^?3 day^?1 (attained at a dilution rate of 0.14 h^?1). Although the highest average ethanol concentration (97.14 g dm^?3) occurred at the initial dilution rate of 0.04 h^?1, the peak average ethanol production rate (2.87 g (g yeast)^?1 day^?1) was reached at a greater dilution rate of 0.11 ^h-1. Thus, the optimal dilution rate was determined to be between 0.11 h^?1 and 0.14 h^?1. Ethanol inhibition on yeast cells was absent in the reactor at average bulk-liquid ethanol concentrations as high as 97.14 g dm^?3. In addition, zero-order kinetics on ethanol production and glucose utilization was evident.     

Modelling and simulation of an oxychlorination reactor in a fluidized bed

Taking 1,2‐dichloroethane from the oxychlorination reaction is a commercially very important process due to the large application of the 1,2‐dichloroethane in the chemical industry of PVC production. This work presents the modeling and simulation of an oxychlorination reactor with a fluidized bed. The pseudo‐homogeneous model with one‐dimensional flow in steady state was applied based on the theory of fluidized bed in two phases. It allows the sensitivity analysis of the operational and project parameters of the reactor. The ordinary differential equations system that represents the mathematical model of the reactor was solved through the application of the numerical method of Newton–Raphson's. The results obtained have proved that the developed model represents the system suitably, in spite of the one‐dimensional model. The effect of different parameters was investigated through the sensitivity analysis, and the results show that the parameters that have the largest influence on the reactor performances are: fluidized bed height, bubble diameter, residence time, cupric chloride weight in the catalyst, and emulsion phase temperature.     

Performance of inverse fluidized bed bioreactor in treating starch wastewater

Aerobic digestion of starch industry waste-water was carried out in an inverse fluidized bed bioreactor using low-density (870 kg/m³) polypropylene particles. Experiments were carried out at different initial substrate concentrations of 2250, 4475, 6730, and 8910 mg COD/L and for various hydraulic retention times (HRT) of 40, 32, 24, 16, and 8 h. Degradation of organic matter was studied at different organic loading rates (OLR) by varying the HRT and the initial substrate concentration. From the results it was observed that the maximum COD removal of 95.6% occurred at an OLR of 1.35 kg COD/(m³·d) and the minimum of 51.8% at an OLR of 26.73 kg COD/(m³·d). The properties of biomass accumulation on the surface of particles were also studied. It was observed that constant biomass loading was achieved over the entire period of operation.     

Dynamic analysis of a fluidized bed reactor for sucrose inversion catalyzed by immobilized invertase

Axial mixing characteristics and performance of a liquid fluidized bed reactor for the sucrose inversion reaction, which was catalyzed by immobilized invertase was studied. The invertase enzyme was immobilized in polyacrylamide gel by the bead polymerization technique. Well-defined spherical gel particles of five different sizes (0.29–3.16 mm) were prepared. Efficiency of the immobilization technique, the optimum working conditions and the kinetic parameters were determined in a batch system. It is shown that as the particle size increases the rate of inversion first increases due to decrease of enzyme loss by leaching and then decreases because of diffusional limitations after a maixmum is reached. The performance of the fluidized bed reactor was investigated dynamically by introducing a step input of substrate at the inlet and analyzing the response curves. These experiments were performed at the optimum temperature (55°C) and using the optimum particle size (2.15 mm). The axial dispersion coefficient was found to increase from 0.45 to 1.26 cm²/s by changing the liquid velocity from 0.32 to 0.58 cm/s.     

Removal of hydrogen sulfide, benzene and toluene by a fluidized bed bioreactor

The dominant off gases from publicly owned treatment works include hydrogen sulfide, benzene, and toluene. In this research, hydrogen sulfide oxidized byBacillus cereus, and benzene with toluene were removed by VOC-degrading microbial consortium. The optimum operating condition of the fluidized bed bioreactor including both microorganisms was 30 ‡C, pH 6–8, and 150 cm of liquid bed height. The critical loading rate of hydrogen sulfide, benzene and toluene in the bioreactor was about 15 g/m³h, 10 g/m³h and 12 g/m³h, respectively. The fluidized bed bioreactor showed an excellent elimination capacity for 580 hours of continuous operation, and maintained stable removal efficiency at sudden inlet concentration changes.     

Degradation of benzene and toluene by a fluidized bed bioreactor including microbial consortium

A fluidized bed bioreactor including microbial consortium was used to remove benzene and toluene simultaneously. The microbial consortium was obtained from sewage treatment plant, and showed maximum benzene degradation rate of 45.2 mg/l·h·mg cell in 30 °C and pH 7.0, and maximum toluene degradation rate of 44.4 mg/l·h·mg cell in 30 °C and pH 8.0. The optimum operating condition of the fluidized bed bioreactor was 30 °C, pH 7.0 and 150 cm of liquid bed height. The average removal efficiency of benzene was 94% for inlet concentration of 53(±5) ppm benzene and that of toluene was 96% for an inlet concentration 48(±5) ppm toluene at 600 l/h of gas volumetric flow rate. The maximum removal capacity in the experimental condition was 22.3 g/m³·h for benzene and 16.3 g/ m³·h for toluene. This paper is dedicated to Professor Dong Sup Doh on the occasion of his retirement from Korea University.     

Modelling and simulation of a coal gasification process in pressurized fluidized bed

A coal gasification mathematical model that can predict temperature, converted fraction and particle size distribution for solids have been developed for a high pressure fluidized bed. For gases in both emulsion and bubble phase, it can predict temperature profiles, gas composition, velocities and other fluid-dynamic parameters. In the feed zone, it could be considered a Gaussian distribution or any other distribution for the solid particle size. Experimental data from literature have been used to validate the model. Finally, the model can be used to optimize the gasification process changing several parameters, such as excess of air, particle size distribution, coal type and reactor geometry.