- Email: [email protected]
Fluid bed porosity mathematical model for an inverse fluidized bed bioreactor with particles growing biofilm
Journal of Environmental Management 104 (2012) 62e66
Contents lists available at SciVerse ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Fluid bed porosity mathematical model for an inverse fluidized bed bioreactor with particles growing biofilm K.E. Campos-Díaz a, *, E.R. Bandala-González b,1, R. Limas-Ballesteros a a
Laboratorio de Investigación en Ingeniería Química Ambiental, SEPI-ESIQIE, Instituto Politécnico Nacional, ESIQIE, Edif. Z, Secc. 6, 1er Piso. U.P. Adolfo López Mateos, C.P. 07738, Mexico b Departamento de Ingeniería Civil y Ambiental, Escuela de Ingeniería y Ciencias, Universidad de las Américas Puebla, Mexico
a r t i c l e i n f o
a b s t r a c t
Article history: Received 27 September 2009 Received in revised form 1 August 2011 Accepted 7 March 2012 Available online 7 April 2012
A new mathematic model to estimate bed porosity as a function of Reynolds and Archimedes numbers was developed based in experimental data. Experiments were performed using an inverse fluidized bed bioreactor filled with polypropylene particles, Lactobacillus acidophillus as the immobilized strain and fluidized with a ManeRogosaeSharpe culture medium under controlled temperature and pH conditions. Bed porosity was measured at different flow rates, starting from 0.95 to 9.5 LPM. The new model has several advantages when compared with previously reported. Among them, advantages such as standard deviation values 1% between experimental and calculated bed porosity, its applicability in traditional and inverse fluidization, wall effects do not take account, it gives excellent agreement with spherical particles with or without biofilm, and inertial drag coefficient allow extend the new model a nonspherical particles. Ó 2012 Elsevier Ltd. All rights reserved.
Keywords: Fluidized bed bioreactor Inverse fluidization Fluid bed porosity mathematical model Lactobacillus acidophilus Biofilm
1. Introduction Fluidization is an operation in which solid particles enters in contact with a fluid acquiring fluid characteristics (Mukherjee et al., 2009). Fluidized reactors used in biological applications are called fluidized bed bioreactors and they have received much attention since the 70’s (Ulson et al., 2008). This kind of reactors are usually characterized by the catalytic use of enzymes or microbial cells that are immobilized by attachment, entrapment, encapsulation or selfaggregation in solid particles used as support (Chun-Zao et al., 2009). In conventional fluidization, the solid particles have a higher density than the fluid therefore the bed solids can be fluidized by an upwards flow, in the case of a bed of particles having a density smaller than the fluid (usually liquid), the bed is fluidized by a downwards flow of the liquid and it is usually called inverse fluidized bed (Bimal et al., 2010). Among the advantages presented by the latest are high mass transfer rate, minimum carry over of coated microorganism due to less solid attrition, efficient control of biofilm thickness (Garcia et al., 1998a; Bimal et al., 2010). Nevertheless its major advantage over conventional biofilm reactors (the trickling filter and biodisc processes) it is the huge surface area for * Corresponding author. Tel.: þ52 55 57 29 60 55382. E-mail addresses: [email protected] (K.E. Campos-Díaz), udlap.mx (E.R. Bandala-González). 1 Tel.: þ222 229 20 00x2652.
erick.bandala@
0301-4797/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.jenvman.2012.03.019
biomass attachment (Garcia et al., 1998a). The downwards flow fluidization configuration enables the recovery of over coated particles at the bottom of the bed. Moreover, the liquid and the biogas flows in opposite directions, which improves bed expansion (Garcia et al., 1998b; Buffière et al., 1999). Additionally, the fluidized bed biofilm reactor is capable of achieving wastewater treatments with low retention time because of the high biomass concentration that can be achieved in the reactor (W1odzimierz and Wojciech, 2006). The biofilm growth in the particles of the bed greatly modifies its porosity due to the particle’s size and density variation. This modification in the bed porosity affects the proper operation and control of the fluid bed reactor. Therefore, it is important to have a reliable model in order to estimate bed porosity for design and scale-up of fluid bed bioreactors. Numerous models have been proposed to predict the bed porosity in fluidized beds with spherical, non-spherical particles and particles with biofilm. The major and most used until today by several authors (Yang and Renken, 2003; Akgiray and Soyer, 2006; Renganathan and Krishnaiah, 2007; Fuentes et al., 2008; Soyer and Akgiray, 2009) are: Richardson and Zaki (1954), Wen and Yu (1966), Ramamurthy and Subbaraju (1973), Riba and Couderc (1977), Fan et al. (1982) and Setiadi (1995) models. Mathematical models for prediction of traditional fluidized bed porosity can be used in inverse fluidization, nevertheless, some authors disagrees with this practice (Hyun, 2001). The aim of this work is to propose a mathematic model to estimate bed porosity in
K.E. Campos-Díaz et al. / Journal of Environmental Management 104 (2012) 62e66
63
Nomenclature Inertial drag coefficient Particle diameter Column diameter Superficial Reynolds number Arquimedes number Bed porosity
CI d D Res Ar 3
an inverse fluidized bed bioreactor with particle biofilm growth in terms of easily measurable physical properties. 2. Materials and methods Fig. 2. SEM photograph of biofilm microorganism.
2.1. Support Polypropylene spherical particles average diameter 3.36 mm and density of 808 kg m3 were used as support.
NaOH (0.05 N) was kept in a storage tank and dosified to the culture media storage tank by means of a peristaltic pump.
2.2. Microorganism, culture media and inoculum preparation 2.4. Start-up Lactobacillus acidophilus NCFM LYO 10 DCU, obtained from Danisco Cultures was used in all the experiments of fluidization. L. acidophilus was growth in MRS (ManeRogosaeSharpe) culture medium. The MRS culture medium was composed of sucrose as carbon source (20 g L1), bactopeptone (10 g L1), bactoyeast extract (5 g L1), bactomalt extract (10 g L1), dipotassium phosphate (2 g L1), sodium acetate (5 g L1), magnesium sulphate (0.1 g L1), manganese sulphate (0.05 g L1) and lactose 22 (g L1) and initial pH ¼ 7. Inoculum preparation was made with the transfer of L. acidophilus lyophilized with 10 mL of liquid MRS medium and incubated at 37 C 1 C for 24 h. After this time 1 mL of first inoculum was transferred to a 9 mL of new MRS medium and incubated under the same conditions and finally, after 24 h, 1 mL of second inoculums was transferred to a 9 mL of new medium and incubated again. 2.3. Installation A schematic diagram of the fluidized bed reactor used in this work is shown in Fig. 1. As shown, the experimental setup included a 4 cm I.D. and 50 cm high cylindrical glass column. The culture medium was stored in a tank, and pumped into the column by a 1/8 HP little giant centrifugal pump, model 583002. The flow rate in the reactor was measured with a rotameter and fixed in the range between 0.95 and 9.5 LPM. To maintain the reactor pH, a solution of
H2O BP DP
C El
BST R TCI
LI
FT CP
SYMBOLOGY BP = Bed of solid particles R = Rotameter C = Column LI = Level indicator TCI=Temperature controller indicator FT =Feed tank CP =Centrifugal Pump El = Electrode BST=Buffer solution tank CP =Centrifugal Pump DP =Dosifier Pump H2O Media Buffer solution
Fig. 1. Schematic diagram of the inverse fluidization installation.
L. acidophilus was immobilized in the polypropylene particles support. The pH within the reactor was regulated at 5.5 0.5 with a NaOH (0.05 N) solution during the start-up period. The reactor temperature was kept constant at 37 C 1 C with a heated circulating bath. The sucrose (110 g L1) consumed by the microorganisms was determined by the DNS colorimetric technique for determination of reducing sugars (Miller, 1959) and it was compensated by a culture medium refilling every 12 h without stopping fluidization experimental process. The height of the bed was measured after 168 h of continuous fluidization process at different flow rates, to determine the experimental bed porosity with biofilm.
2.5. Verification of Lactobacillus presence in biofilm In order to verify the presence of the Lactobacillus biofilm in the polypropylene particles, a small particle sample was taken at the end of the fluidization process and submitted to gram staining (Fig. 2).
2.6. Biofilm thickness measurement The average biofilm thickness was measured experimentally in a particle sample at the conclusion of the inverse fluidization by cutting the particle by the half with a scalpel. Photographs were taken in an optical microscope coupled to the computer as shown in Fig. 3 and biofilm thickness was measured with a reticule calibrated in mm. The biofilm thickness measured was verified using SEM micrographs of a sample particle. The samples for SEM were fixed in 2.5% glutaraldehyde in phosphate buffer solution (PBS) 0.1 M, pH 6.8 for 1 h. The fixed samples were washed three times with PBS and dehydrated with increasing ethanol concentrations (25, 50, 70, 80, 90, 95, 100% vol vol1) for 10 min each one. Finally, samples were coated with a gold layer in Argon atmosphere in a vacuum evaporator Bal-Tec, prior to examination with a scanning electron microscope (JEOL 2000, FX), operated at a 15 kV acceleration voltage.
64
K.E. Campos-Díaz et al. / Journal of Environmental Management 104 (2012) 62e66
Fig. 3. Photograph of biofilm thickness.
1:753*3 3:807 ¼ Ar=0:75* 24*Res þ CI *Re2s
3. Mathematical model development The model proposed in this work was developed following the approach for fluidization of spherical and non-spherical particles, which use the typical balance of forces that exerts on an isolated particle (gravity, buoyancy, drag) proposed by Wen and Yu (1966) and extending it to the particles bed. Limas et al. (1982) suggested the use of the inertial drag coefficient (CI) proposed by Becker (1959) for different values of Reynolds number for spherical and non-spherical particles. This mathematical model calculates the bed porosity as a function of Reynolds and Archimedes numbers. The inertial drag coefficient (CI) is related with the shape, orientation of the particle and Reynolds number. Becker (1959) developed a set of equations for the inertial drag coefficient for isolated regular and irregular particles in a wide range of values of Reynolds number (from 0.1 to 200,000), and showed that the inertial drag coefficient is a surface function of the particle. In this work Inertial drag coefficient was calculated using the equation proposed by Becker (1959) for values of the Reynolds number in the range 5.5 Res 200.
CI ¼ 2:25*ð5:5=Res Þ0:34
(1)
where CI is the inertial drag coefficient and Res is the superficial Reynolds number. The experimental data used for estimation of the drag coefficient by Equation (1) were the published by Wilhelm and Kwauk (1948), and the experimental data obtained in this work. Equation (2) was used to calculate the bed porosity for spherical particles for the range of Reynolds number and porosities spanned by the experimental mentioned data (5.5 Res 200 and 0.375 3 0.928):
Fig. 4. Biofilm thickness.
where Ar is Arquimedes number and
3
(2)
is the bed porosity.
4. Results and discussion 4.1. Microorganisms adhesion on the support and biofilm development Microorganism growth was analyzed after 168 h of inverse fluidization. Fig. 3 shows a photograph of this sampled which was obtain by scanning electron microscopic. It can be seen the presence of a gram positive microorganism with the typical Lactobacillus shape. As shown in Figs. 4 and 5 the microorganisms biofilm is thin, but the approximation allowed measure the biofilm thickness. Average biofilm thickness was 0.077 mm. Fig. 6 shows that the biofilm developed on surface particle was attached by clusters of microorganism agglomerates and for small individual rods with bacillus morphologically. The micrographs with an electron microscope allowed confirming the presence of microorganisms inside the particle’s pores (Fig. 7) which are associates with bacillus chains observing that the microorganism morphology and grouping is not the same at different sections of the support (Fig. 8). 4.2. Mathematical model proposed and comparison with other models The comparison of the results was made by plotting the porosity function obtained from values proposed by Richardson and Zaki
Fig. 5. Approach biofilm thickness.
K.E. Campos-Díaz et al. / Journal of Environmental Management 104 (2012) 62e66
65
Fig. 6. Biofilm on surface particle.
Fig. 7. Porous particle.
(1954), Wen and Yu (1966), Ramamurthy and Subbaraju (1973), Riba and Couderc, 1977, Setiadi (1995) and this work (2007), versus experimental bed porosity data. Wen and Yu (1966) model fits the trend of experimental data in general. The plot shows a slight experimental underestimation for experimental data between 0.6 and 0.73 of bed porosity. The standard deviation between experimental bed porosity data and the data obtained by Wen and Yu (1966) model was 4.2%. Fig. 9 shows that the mathematical model of Riba and Couderc (1977) has a good correlation of bed porosity data; the standard deviation was 1.3%. Ramamurthy and Subbaraju, 1973 model fits well some of the experimental bed porosity data. This deviation can be attributed to the principal assumption by these authors that the solid particles are represented as an ensemble moving about points considered as nodes of an imaginary lattice through whose free volume the fluid is flowing. It is important to notice that this model was developed for annular fluidized beds. Richardson and Zaki (1954) porosity function fits the experimental data tendency, but do not represent the experimental bed porosity. The deviation in fitting data is attributed to the diameter relationship particle diameter/column diameter (d/D) parameter used by Richardson and Zaki (1954). The model proposed by Setiadi (1995) showed that the calculate values crosses the trend of the experimental bed porosity data, which indicates that the function is not suitable. The mathematical model proposed in this work (Equation (2)) fits well the trend of the experimental bed porosity data for particles with biofilm which was well represented by this function in the interval of porosity from 0.47 to 0.73.
Standard deviation obtained by the models published in literature are: Wen and Yu (1966) 4.2%, Riba and Couderc, 1977 1.3%, Ramamurthy and Subbaraju, 1973 6.1%, Setiadi (1995) 8.3% and the standard deviation between experimental and calculated bed porosity values was of 1% with the model proposed in this work. There are several important motivations for the new model. First, although some of the existing correlations seem to be accurate enough (giving mean errors in porosity from 2% to 9%), further improvement in accuracy will not be superfluous. This is because relatively small errors in porosity may lead to large errors in fluidized bed height, which is frequently the quantity of practical interest. Second, the only practically useful correlations that can be applied to non-spherical media are those presented. The new model has several advantages because Wen and Yu (1966), Riba and Couderc, 1977, and Ramamurthy and Subbaraju, 1973, restrict their model to traditional fluidization (upward flow) and spherical particles. On the other hand, Setiadi (1995) model works with spherical and non-spherical particles with biofilm growing in the traditional fluidized bed. Furthermore the new model is considerably simply in form, it can be used for spherical particles, but includes a term of inertial drag coefficient proposed by Becker (1959) which includes a shape factor to non-spherical particles, also applies to particles with and without biofilm, very small deviation can be used in a simple manner, the wall effects do not take account as a Richardson and Zaki (1954) and Ramamurthy and Subbaraju, 1973 model.
Fig. 8. Biofilm inside particle.
66
K.E. Campos-Díaz et al. / Journal of Environmental Management 104 (2012) 62e66
Fig. 9. Porosity function f(3 ) versus bed porosity (3 ) used experimental data to biofilm.
5. Conclusions The mathematical model obtained in this study represent the bed porosity in an inverse fluidized bed bioreactor for spherical particles with biofilm with a value of standard deviation smaller than those obtained using Richardson and Zaki (1954), Wen and Yu (1966), Ramamurthy and Subbaraju, 1973, Riba and Couderc (1977) and Setiadi (1995) models published in the literature. The results demonstrate that the models published in literature used to predict the bed porosity in traditional fluidized bioreactor (upwards flow) can also be used to estimate the bed porosity in inverse fluidization (downwards flow). The inverse fluidization bioreactor do not present accumulation or obstruction in the flow caused by presence of biomass at the bottom of the bioreactor, which represents an advantage in terms of death times reduction for maintenance. Acknowledgments The authors wish to express their gratitude to the National Polytechnic Institute of Mexico (IPN), CINVESTAV-IPN, UDLAP and CONACYT. References Akgiray, O., Soyer, E., 2006. An evaluation of expansion equation for fluidized solidliquid systems. Journal of Water Supply: Research and Technology-AQUA 55, 517e526. Becker, H.A., 1959. The effects of shape and Reynolds number on drag in the motion of a freely oriented body in an infinite fluid. The Canadian Journal of Chemical Engineering 37, 85e91. Bimal, D., Uma, P.G., Sudip, P.G., 2010. Inverse fluidization using non-newtonian liquids. Chemical Engineering and Processing: Process Intensification 49, 1169e1175. Buffière, P., Bergeon, J.P., Moletta, R., 1999. The inverse turbulent bed: a novel bioreactor for anaerobic treatment. Biotechnology and Bioengineering 8, 673e677. Chun-Zao, L., Feng, W., Fan, O.Y., 2009. Ethanol fermentation in a magnetically fluidized bed reactor with immobilized Saccharomyces Cerevisiae in magnetic particles. Bioresource Technology 100, 878e882.
Fan, L.S., Muroyama, K., Chern, S.H., 1982. Hydrodynamic characteristics of inverse fluidization in liquid-solid and gas-liquid-solid systems. The Chemical Engineering Journal 24, 143e150. Fuentes, M., Scena, J.N., Aguirre, A.P., Mussati, M.C., 2008. Hydrodynamic aspects in anaerobic fluidized bed reactor modeling. Chemical Enginnering and Processing 47, 1530e1540. Garcia, B.D., Buffière, P., Elmaleh, S., Moletta, R., 1998a. Application of the downflow fluidized bed to the anaerobic treatment of wine distillery wastewater. Water Science and Technology 38, 393e399. Garcia, C.D., Buffiere, P., Moletta, R., Elmaleh, S., 1998b. Anaerobic digestion of wine distillery wastewater in down-flow fluidized bed. Water Research 38, 3593e3600. Hyun, L.D., 2001. Transition velocity and bed expansion of two-phase (liquid-solid) fluidization system. Korean Journal of Chemical Engineering 18, 347e350. Limas, B.R., Riba, J.P., Courderc, J.P., 1982. Expansion de couches de particules non sphériques fluidisées par un liquide. Entropie 106, 37e45. Miller, L.G., 1959. Use dinitrosalicylic acid reagent for determination of reducing sugar. Analytical Chemistry 31, 426e428. Mukherjee, A.K., Mishra, B.K., Ran, V.K., 2009. Application of liquid/solid fluidization technique in beneficiation of fines. International Journal of Mineral Processing 92 (1e2), 67e73. Ramamurthy, K., Subbaraju, K., 1973. Bed expansion characteristics of annular liquid-fluidized beds. Industrial and Engineering Chemistry Process and Design Development 12, 184e189. Renganathan, T., Krishnaiah, K., 2007. Spatio-temporal evolution of void fraction in liquid-solid inverse fluidized bed. Chemical Engineering Science 62 (1e2), 328e338. Riba, J.P., Couderc, J.P., 1977. Expansion de couches fluidisées par des liquides. The Canadian Journal of Chemical Engineering 55, 118e121. Richardson, J.F., Zaki, W.N., 1954. Sedimentation and fluidization. Transactions of the Institution of Chemical Engineers 32, 35e53. Setiadi, T., 1995. Predicting the bed of an anaerobic fluidized bed bioreactor. Water Science and Technology 31, 181e191. Soyer, E., Akgiray, O., 2009. A new simple equation for the prediction of filter expansion during backwashing. Journal of Water Supply: Research and Technology-AQUA 58, 336e345. Ulson, D.S.M.A., De Lima, B.H., Ulson, D.S.A.A., 2008. Modelling of liquid pollutant biodegradation process in a fluidized bed reactor with biofilm. Separation and Purification Technology 60, 162e173. Wen, C.Y., Yu, Y.H., 1966. Mechanics of fluidization. Chemical Engineering Progress Symposium Series 62, 100e111. Wilhelm, H.R., Kwauk, M., 1948. Fluidization of solid particles. Chemical Engineering Progress 44, 201e218. W1odzimierz, S., Wojciech, K., 2006. Aerobic treatment of wastewater in the inverse fluidized bed biofilm reactor. Chemical Engineering Journal 118, 199e205. Yang, J., Renken, A., 2003. A generalized correlation for equilibrium of forces in liquid-solid fluidized bed. Chemical Engineering Journal 92, 7e14.
Contents lists available at SciVerse ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Fluid bed porosity mathematical model for an inverse fluidized bed bioreactor with particles growing biofilm K.E. Campos-Díaz a, *, E.R. Bandala-González b,1, R. Limas-Ballesteros a a
Laboratorio de Investigación en Ingeniería Química Ambiental, SEPI-ESIQIE, Instituto Politécnico Nacional, ESIQIE, Edif. Z, Secc. 6, 1er Piso. U.P. Adolfo López Mateos, C.P. 07738, Mexico b Departamento de Ingeniería Civil y Ambiental, Escuela de Ingeniería y Ciencias, Universidad de las Américas Puebla, Mexico
a r t i c l e i n f o
a b s t r a c t
Article history: Received 27 September 2009 Received in revised form 1 August 2011 Accepted 7 March 2012 Available online 7 April 2012
A new mathematic model to estimate bed porosity as a function of Reynolds and Archimedes numbers was developed based in experimental data. Experiments were performed using an inverse fluidized bed bioreactor filled with polypropylene particles, Lactobacillus acidophillus as the immobilized strain and fluidized with a ManeRogosaeSharpe culture medium under controlled temperature and pH conditions. Bed porosity was measured at different flow rates, starting from 0.95 to 9.5 LPM. The new model has several advantages when compared with previously reported. Among them, advantages such as standard deviation values 1% between experimental and calculated bed porosity, its applicability in traditional and inverse fluidization, wall effects do not take account, it gives excellent agreement with spherical particles with or without biofilm, and inertial drag coefficient allow extend the new model a nonspherical particles. Ó 2012 Elsevier Ltd. All rights reserved.
Keywords: Fluidized bed bioreactor Inverse fluidization Fluid bed porosity mathematical model Lactobacillus acidophilus Biofilm
1. Introduction Fluidization is an operation in which solid particles enters in contact with a fluid acquiring fluid characteristics (Mukherjee et al., 2009). Fluidized reactors used in biological applications are called fluidized bed bioreactors and they have received much attention since the 70’s (Ulson et al., 2008). This kind of reactors are usually characterized by the catalytic use of enzymes or microbial cells that are immobilized by attachment, entrapment, encapsulation or selfaggregation in solid particles used as support (Chun-Zao et al., 2009). In conventional fluidization, the solid particles have a higher density than the fluid therefore the bed solids can be fluidized by an upwards flow, in the case of a bed of particles having a density smaller than the fluid (usually liquid), the bed is fluidized by a downwards flow of the liquid and it is usually called inverse fluidized bed (Bimal et al., 2010). Among the advantages presented by the latest are high mass transfer rate, minimum carry over of coated microorganism due to less solid attrition, efficient control of biofilm thickness (Garcia et al., 1998a; Bimal et al., 2010). Nevertheless its major advantage over conventional biofilm reactors (the trickling filter and biodisc processes) it is the huge surface area for * Corresponding author. Tel.: þ52 55 57 29 60 55382. E-mail addresses: [email protected] (K.E. Campos-Díaz), udlap.mx (E.R. Bandala-González). 1 Tel.: þ222 229 20 00x2652.
erick.bandala@
0301-4797/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.jenvman.2012.03.019
biomass attachment (Garcia et al., 1998a). The downwards flow fluidization configuration enables the recovery of over coated particles at the bottom of the bed. Moreover, the liquid and the biogas flows in opposite directions, which improves bed expansion (Garcia et al., 1998b; Buffière et al., 1999). Additionally, the fluidized bed biofilm reactor is capable of achieving wastewater treatments with low retention time because of the high biomass concentration that can be achieved in the reactor (W1odzimierz and Wojciech, 2006). The biofilm growth in the particles of the bed greatly modifies its porosity due to the particle’s size and density variation. This modification in the bed porosity affects the proper operation and control of the fluid bed reactor. Therefore, it is important to have a reliable model in order to estimate bed porosity for design and scale-up of fluid bed bioreactors. Numerous models have been proposed to predict the bed porosity in fluidized beds with spherical, non-spherical particles and particles with biofilm. The major and most used until today by several authors (Yang and Renken, 2003; Akgiray and Soyer, 2006; Renganathan and Krishnaiah, 2007; Fuentes et al., 2008; Soyer and Akgiray, 2009) are: Richardson and Zaki (1954), Wen and Yu (1966), Ramamurthy and Subbaraju (1973), Riba and Couderc (1977), Fan et al. (1982) and Setiadi (1995) models. Mathematical models for prediction of traditional fluidized bed porosity can be used in inverse fluidization, nevertheless, some authors disagrees with this practice (Hyun, 2001). The aim of this work is to propose a mathematic model to estimate bed porosity in
K.E. Campos-Díaz et al. / Journal of Environmental Management 104 (2012) 62e66
63
Nomenclature Inertial drag coefficient Particle diameter Column diameter Superficial Reynolds number Arquimedes number Bed porosity
CI d D Res Ar 3
an inverse fluidized bed bioreactor with particle biofilm growth in terms of easily measurable physical properties. 2. Materials and methods Fig. 2. SEM photograph of biofilm microorganism.
2.1. Support Polypropylene spherical particles average diameter 3.36 mm and density of 808 kg m3 were used as support.
NaOH (0.05 N) was kept in a storage tank and dosified to the culture media storage tank by means of a peristaltic pump.
2.2. Microorganism, culture media and inoculum preparation 2.4. Start-up Lactobacillus acidophilus NCFM LYO 10 DCU, obtained from Danisco Cultures was used in all the experiments of fluidization. L. acidophilus was growth in MRS (ManeRogosaeSharpe) culture medium. The MRS culture medium was composed of sucrose as carbon source (20 g L1), bactopeptone (10 g L1), bactoyeast extract (5 g L1), bactomalt extract (10 g L1), dipotassium phosphate (2 g L1), sodium acetate (5 g L1), magnesium sulphate (0.1 g L1), manganese sulphate (0.05 g L1) and lactose 22 (g L1) and initial pH ¼ 7. Inoculum preparation was made with the transfer of L. acidophilus lyophilized with 10 mL of liquid MRS medium and incubated at 37 C 1 C for 24 h. After this time 1 mL of first inoculum was transferred to a 9 mL of new MRS medium and incubated under the same conditions and finally, after 24 h, 1 mL of second inoculums was transferred to a 9 mL of new medium and incubated again. 2.3. Installation A schematic diagram of the fluidized bed reactor used in this work is shown in Fig. 1. As shown, the experimental setup included a 4 cm I.D. and 50 cm high cylindrical glass column. The culture medium was stored in a tank, and pumped into the column by a 1/8 HP little giant centrifugal pump, model 583002. The flow rate in the reactor was measured with a rotameter and fixed in the range between 0.95 and 9.5 LPM. To maintain the reactor pH, a solution of
H2O BP DP
C El
BST R TCI
LI
FT CP
SYMBOLOGY BP = Bed of solid particles R = Rotameter C = Column LI = Level indicator TCI=Temperature controller indicator FT =Feed tank CP =Centrifugal Pump El = Electrode BST=Buffer solution tank CP =Centrifugal Pump DP =Dosifier Pump H2O Media Buffer solution
Fig. 1. Schematic diagram of the inverse fluidization installation.
L. acidophilus was immobilized in the polypropylene particles support. The pH within the reactor was regulated at 5.5 0.5 with a NaOH (0.05 N) solution during the start-up period. The reactor temperature was kept constant at 37 C 1 C with a heated circulating bath. The sucrose (110 g L1) consumed by the microorganisms was determined by the DNS colorimetric technique for determination of reducing sugars (Miller, 1959) and it was compensated by a culture medium refilling every 12 h without stopping fluidization experimental process. The height of the bed was measured after 168 h of continuous fluidization process at different flow rates, to determine the experimental bed porosity with biofilm.
2.5. Verification of Lactobacillus presence in biofilm In order to verify the presence of the Lactobacillus biofilm in the polypropylene particles, a small particle sample was taken at the end of the fluidization process and submitted to gram staining (Fig. 2).
2.6. Biofilm thickness measurement The average biofilm thickness was measured experimentally in a particle sample at the conclusion of the inverse fluidization by cutting the particle by the half with a scalpel. Photographs were taken in an optical microscope coupled to the computer as shown in Fig. 3 and biofilm thickness was measured with a reticule calibrated in mm. The biofilm thickness measured was verified using SEM micrographs of a sample particle. The samples for SEM were fixed in 2.5% glutaraldehyde in phosphate buffer solution (PBS) 0.1 M, pH 6.8 for 1 h. The fixed samples were washed three times with PBS and dehydrated with increasing ethanol concentrations (25, 50, 70, 80, 90, 95, 100% vol vol1) for 10 min each one. Finally, samples were coated with a gold layer in Argon atmosphere in a vacuum evaporator Bal-Tec, prior to examination with a scanning electron microscope (JEOL 2000, FX), operated at a 15 kV acceleration voltage.
64
K.E. Campos-Díaz et al. / Journal of Environmental Management 104 (2012) 62e66
Fig. 3. Photograph of biofilm thickness.
1:753*3 3:807 ¼ Ar=0:75* 24*Res þ CI *Re2s
3. Mathematical model development The model proposed in this work was developed following the approach for fluidization of spherical and non-spherical particles, which use the typical balance of forces that exerts on an isolated particle (gravity, buoyancy, drag) proposed by Wen and Yu (1966) and extending it to the particles bed. Limas et al. (1982) suggested the use of the inertial drag coefficient (CI) proposed by Becker (1959) for different values of Reynolds number for spherical and non-spherical particles. This mathematical model calculates the bed porosity as a function of Reynolds and Archimedes numbers. The inertial drag coefficient (CI) is related with the shape, orientation of the particle and Reynolds number. Becker (1959) developed a set of equations for the inertial drag coefficient for isolated regular and irregular particles in a wide range of values of Reynolds number (from 0.1 to 200,000), and showed that the inertial drag coefficient is a surface function of the particle. In this work Inertial drag coefficient was calculated using the equation proposed by Becker (1959) for values of the Reynolds number in the range 5.5 Res 200.
CI ¼ 2:25*ð5:5=Res Þ0:34
(1)
where CI is the inertial drag coefficient and Res is the superficial Reynolds number. The experimental data used for estimation of the drag coefficient by Equation (1) were the published by Wilhelm and Kwauk (1948), and the experimental data obtained in this work. Equation (2) was used to calculate the bed porosity for spherical particles for the range of Reynolds number and porosities spanned by the experimental mentioned data (5.5 Res 200 and 0.375 3 0.928):
Fig. 4. Biofilm thickness.
where Ar is Arquimedes number and
3
(2)
is the bed porosity.
4. Results and discussion 4.1. Microorganisms adhesion on the support and biofilm development Microorganism growth was analyzed after 168 h of inverse fluidization. Fig. 3 shows a photograph of this sampled which was obtain by scanning electron microscopic. It can be seen the presence of a gram positive microorganism with the typical Lactobacillus shape. As shown in Figs. 4 and 5 the microorganisms biofilm is thin, but the approximation allowed measure the biofilm thickness. Average biofilm thickness was 0.077 mm. Fig. 6 shows that the biofilm developed on surface particle was attached by clusters of microorganism agglomerates and for small individual rods with bacillus morphologically. The micrographs with an electron microscope allowed confirming the presence of microorganisms inside the particle’s pores (Fig. 7) which are associates with bacillus chains observing that the microorganism morphology and grouping is not the same at different sections of the support (Fig. 8). 4.2. Mathematical model proposed and comparison with other models The comparison of the results was made by plotting the porosity function obtained from values proposed by Richardson and Zaki
Fig. 5. Approach biofilm thickness.
K.E. Campos-Díaz et al. / Journal of Environmental Management 104 (2012) 62e66
65
Fig. 6. Biofilm on surface particle.
Fig. 7. Porous particle.
(1954), Wen and Yu (1966), Ramamurthy and Subbaraju (1973), Riba and Couderc, 1977, Setiadi (1995) and this work (2007), versus experimental bed porosity data. Wen and Yu (1966) model fits the trend of experimental data in general. The plot shows a slight experimental underestimation for experimental data between 0.6 and 0.73 of bed porosity. The standard deviation between experimental bed porosity data and the data obtained by Wen and Yu (1966) model was 4.2%. Fig. 9 shows that the mathematical model of Riba and Couderc (1977) has a good correlation of bed porosity data; the standard deviation was 1.3%. Ramamurthy and Subbaraju, 1973 model fits well some of the experimental bed porosity data. This deviation can be attributed to the principal assumption by these authors that the solid particles are represented as an ensemble moving about points considered as nodes of an imaginary lattice through whose free volume the fluid is flowing. It is important to notice that this model was developed for annular fluidized beds. Richardson and Zaki (1954) porosity function fits the experimental data tendency, but do not represent the experimental bed porosity. The deviation in fitting data is attributed to the diameter relationship particle diameter/column diameter (d/D) parameter used by Richardson and Zaki (1954). The model proposed by Setiadi (1995) showed that the calculate values crosses the trend of the experimental bed porosity data, which indicates that the function is not suitable. The mathematical model proposed in this work (Equation (2)) fits well the trend of the experimental bed porosity data for particles with biofilm which was well represented by this function in the interval of porosity from 0.47 to 0.73.
Standard deviation obtained by the models published in literature are: Wen and Yu (1966) 4.2%, Riba and Couderc, 1977 1.3%, Ramamurthy and Subbaraju, 1973 6.1%, Setiadi (1995) 8.3% and the standard deviation between experimental and calculated bed porosity values was of 1% with the model proposed in this work. There are several important motivations for the new model. First, although some of the existing correlations seem to be accurate enough (giving mean errors in porosity from 2% to 9%), further improvement in accuracy will not be superfluous. This is because relatively small errors in porosity may lead to large errors in fluidized bed height, which is frequently the quantity of practical interest. Second, the only practically useful correlations that can be applied to non-spherical media are those presented. The new model has several advantages because Wen and Yu (1966), Riba and Couderc, 1977, and Ramamurthy and Subbaraju, 1973, restrict their model to traditional fluidization (upward flow) and spherical particles. On the other hand, Setiadi (1995) model works with spherical and non-spherical particles with biofilm growing in the traditional fluidized bed. Furthermore the new model is considerably simply in form, it can be used for spherical particles, but includes a term of inertial drag coefficient proposed by Becker (1959) which includes a shape factor to non-spherical particles, also applies to particles with and without biofilm, very small deviation can be used in a simple manner, the wall effects do not take account as a Richardson and Zaki (1954) and Ramamurthy and Subbaraju, 1973 model.
Fig. 8. Biofilm inside particle.
66
K.E. Campos-Díaz et al. / Journal of Environmental Management 104 (2012) 62e66
Fig. 9. Porosity function f(3 ) versus bed porosity (3 ) used experimental data to biofilm.
5. Conclusions The mathematical model obtained in this study represent the bed porosity in an inverse fluidized bed bioreactor for spherical particles with biofilm with a value of standard deviation smaller than those obtained using Richardson and Zaki (1954), Wen and Yu (1966), Ramamurthy and Subbaraju, 1973, Riba and Couderc (1977) and Setiadi (1995) models published in the literature. The results demonstrate that the models published in literature used to predict the bed porosity in traditional fluidized bioreactor (upwards flow) can also be used to estimate the bed porosity in inverse fluidization (downwards flow). The inverse fluidization bioreactor do not present accumulation or obstruction in the flow caused by presence of biomass at the bottom of the bioreactor, which represents an advantage in terms of death times reduction for maintenance. Acknowledgments The authors wish to express their gratitude to the National Polytechnic Institute of Mexico (IPN), CINVESTAV-IPN, UDLAP and CONACYT. References Akgiray, O., Soyer, E., 2006. An evaluation of expansion equation for fluidized solidliquid systems. Journal of Water Supply: Research and Technology-AQUA 55, 517e526. Becker, H.A., 1959. The effects of shape and Reynolds number on drag in the motion of a freely oriented body in an infinite fluid. The Canadian Journal of Chemical Engineering 37, 85e91. Bimal, D., Uma, P.G., Sudip, P.G., 2010. Inverse fluidization using non-newtonian liquids. Chemical Engineering and Processing: Process Intensification 49, 1169e1175. Buffière, P., Bergeon, J.P., Moletta, R., 1999. The inverse turbulent bed: a novel bioreactor for anaerobic treatment. Biotechnology and Bioengineering 8, 673e677. Chun-Zao, L., Feng, W., Fan, O.Y., 2009. Ethanol fermentation in a magnetically fluidized bed reactor with immobilized Saccharomyces Cerevisiae in magnetic particles. Bioresource Technology 100, 878e882.
Fan, L.S., Muroyama, K., Chern, S.H., 1982. Hydrodynamic characteristics of inverse fluidization in liquid-solid and gas-liquid-solid systems. The Chemical Engineering Journal 24, 143e150. Fuentes, M., Scena, J.N., Aguirre, A.P., Mussati, M.C., 2008. Hydrodynamic aspects in anaerobic fluidized bed reactor modeling. Chemical Enginnering and Processing 47, 1530e1540. Garcia, B.D., Buffière, P., Elmaleh, S., Moletta, R., 1998a. Application of the downflow fluidized bed to the anaerobic treatment of wine distillery wastewater. Water Science and Technology 38, 393e399. Garcia, C.D., Buffiere, P., Moletta, R., Elmaleh, S., 1998b. Anaerobic digestion of wine distillery wastewater in down-flow fluidized bed. Water Research 38, 3593e3600. Hyun, L.D., 2001. Transition velocity and bed expansion of two-phase (liquid-solid) fluidization system. Korean Journal of Chemical Engineering 18, 347e350. Limas, B.R., Riba, J.P., Courderc, J.P., 1982. Expansion de couches de particules non sphériques fluidisées par un liquide. Entropie 106, 37e45. Miller, L.G., 1959. Use dinitrosalicylic acid reagent for determination of reducing sugar. Analytical Chemistry 31, 426e428. Mukherjee, A.K., Mishra, B.K., Ran, V.K., 2009. Application of liquid/solid fluidization technique in beneficiation of fines. International Journal of Mineral Processing 92 (1e2), 67e73. Ramamurthy, K., Subbaraju, K., 1973. Bed expansion characteristics of annular liquid-fluidized beds. Industrial and Engineering Chemistry Process and Design Development 12, 184e189. Renganathan, T., Krishnaiah, K., 2007. Spatio-temporal evolution of void fraction in liquid-solid inverse fluidized bed. Chemical Engineering Science 62 (1e2), 328e338. Riba, J.P., Couderc, J.P., 1977. Expansion de couches fluidisées par des liquides. The Canadian Journal of Chemical Engineering 55, 118e121. Richardson, J.F., Zaki, W.N., 1954. Sedimentation and fluidization. Transactions of the Institution of Chemical Engineers 32, 35e53. Setiadi, T., 1995. Predicting the bed of an anaerobic fluidized bed bioreactor. Water Science and Technology 31, 181e191. Soyer, E., Akgiray, O., 2009. A new simple equation for the prediction of filter expansion during backwashing. Journal of Water Supply: Research and Technology-AQUA 58, 336e345. Ulson, D.S.M.A., De Lima, B.H., Ulson, D.S.A.A., 2008. Modelling of liquid pollutant biodegradation process in a fluidized bed reactor with biofilm. Separation and Purification Technology 60, 162e173. Wen, C.Y., Yu, Y.H., 1966. Mechanics of fluidization. Chemical Engineering Progress Symposium Series 62, 100e111. Wilhelm, H.R., Kwauk, M., 1948. Fluidization of solid particles. Chemical Engineering Progress 44, 201e218. W1odzimierz, S., Wojciech, K., 2006. Aerobic treatment of wastewater in the inverse fluidized bed biofilm reactor. Chemical Engineering Journal 118, 199e205. Yang, J., Renken, A., 2003. A generalized correlation for equilibrium of forces in liquid-solid fluidized bed. Chemical Engineering Journal 92, 7e14.