Biodegradation of phenolic industrial wastewater in a uidized bed bioreactor with immobilized cells of Pseudomonas putida

Transcription

1 Bioresource Technology ) 137±142 Biodegradation of phenolic industrial wastewater in a uidized bed bioreactor with immobilized cells of Pseudomonas putida G. Gonzalez *, G. Herrera, Ma.T. Garcõa, M. Pe~na Department of Chemical Engineering, University of Valladolid, Paseo Prado de la Magdalena s/n, Valladolid 47011, Spain Received 21 August 2000; received in revised form 3 April 2001; accepted 9 April 2001 Abstract The paper presents the main results obtained from the study of the biodegradation of phenolic industrial wastewaters by a pure culture of immobilized cells of Pseudomonas putida ATCC The experiments were carried out in batch and continuous mode. The maximum degradation capacity and the in uence of the adaptation of the microorganism to the substrate were studied in batch mode. Industrial wastewater with a phenol concentration of 1000 mg/l was degraded when the microorganism was adapted to the toxic chemical. The presence in the wastewater of compounds other than phenol was noted and it was found that Pseudomonas putida was able to degrade these compounds. In continuous mode, a uidized-bed bioreactor was operated and the in uence of the organic loading rate on the removal e ciency of phenol was studied. The bioreactor showed phenol degradation e ciencies higher than 90%, even for a phenol loading rate of 0.5 g phenol/l d corresponding to 0.54 g TOC/l d). Ó 2001 Elsevier Science Ltd. All rights reserved. Keywords: Phenol degradation; Pseudomonas putida; Immobilized cells; Fluidized bed 1. Introduction Some industrial wastewaters, especially those coming from the production processes of phenolic resins, contain high concentrations >10 g/l) of phenolic compounds Patterson, 1985). Several physico-chemical and biological treatments have been suggested in the last 20 years to remove e ciently these compounds: adsorption with bone char or zeolites, stripping with air or steam Zilli et al., 1993), wet air oxidation Lin and Chuang, 1994) or biological treatments with pure or mixed cultures of microorganisms Lakhwala et al., 1992; Buitron et al., 1998; Kapoor et al., 1998; Loh et al., 2000) have been used. Several works appeared in the literature concerning the biodegradation of phenol especially from model solutions, by Pseudomonas putida Zilli et al., 1993; Hannaford and Kuek, 1999; Mordocco et al., 1999), with high removal e ciencies. On the other hand, the biodegradation of industrial wastewaters can be improved if the microorganism is previously adapted to the toxic chemical Zilli et al., 1993), especially when high phenol concentrations are present. Moreover, other compounds di erent from * Corresponding author. Tel.: ; fax: address: gerardo@siq.iq.cie.uva.es G. Gonzalez). phenol, also present in the industrial wastewater, can a ect the biodegradation process. The biodegradation of phenol from model solutions has been studied by the authors and reported in previous papers Gonzalez and Herrera, 1995, 2001). These experiments were carried out with free and immobilized cells of P. putida in batch and continuous mode and best results were obtained when a continuous uidized bed bioreactor with immobilized cells was operated. This work presents the results obtained for the biodegradation of high phenol concentrations from industrial wastewaters, by cells of P. putida immobilized in calcium-alginate gel beads. Batch experiments were made in order to obtain the maximum phenol degradation capacity, analyzing the in uence of the adaptation of the microorganism to the medium. Then, continuous experiments in a uidized-bed bioreactor, in order to determine the maximum phenol loading rate to be treated, were carried out. 2. Methods 2.1. Raw wastewater The raw wastewater came from the industrial production of phenolic resins. Some of the characteristics of /01/$ - see front matter Ó 2001 Elsevier Science Ltd. All rights reserved. PII: S )

2 138 G. Gonzalez et al. / Bioresource Technology ) 137±142 this wastewater were ph, 4.5; phenol, 42,000 mg/l; total organic carbon TOC), 45,000 mg C/l and chemical oxygen demand COD), 124,000 mg O 2 =l. The theoretical TOC and COD corresponding to the analytical phenol concentration were calculated. Results obtained were 32,000 mg C/l and 100,100 mg O 2 =l for TOC phenol and COD phenol, respectively. The di erence between experimental and calculated values indicated the presence of other compounds, di erent from phenol, that contributed to the organic load of the industrial wastewater. It was intended to identify these other compounds by HPLC. While some peaks appeared in the analysis, it was not possible to identify any compound and in the paper these are named ``non-phenolic compounds''. Due to the high phenol concentration, it was necessary to dilute the raw wastewater before each experiment Culture and growth medium A strain of P. putida ATCC 17484, biotype B, from the Laboratory of Microbiology Voor Gante, Belgium) was used as pure culture. The maintenance medium was reported in a previous work Gonzalez and Herrera, 1995). The cultures were grown aerobically in 250 ml asks lled to 150 cm 3 with growth medium and stirred in a orbital shaker at 250 rpm at a constant temperature of 30 C andphˆ 6.6. The growth medium was also reported in the previous work with the only di erence in the substrate used for this work: a mixture of the industrial wastewater and commercial phenol was employed, with a nal phenol concentration of 75 mg/l Immobilization protocol The microorganisms were immobilized by entrapment in calcium-alginate gel beads hardened with Al 3. Bioparticles were formed by mixing a suspension of P. putida with a solution of sodium alginate 3% w/w; Protonal LF 10/60), according to Bravo and Gonzalez 1991). These bioparticles spherical gel beads) obtained had an average diameter of 1±2 mm. The concentration in the bioparticles was mg microorganism per litre of alginate Batch experiments Volumes of 250 ml of the wastewater, with bioparticles and a varying phenol concentration in the range 200±1000 mg phenol/l, were disposed in a 500 ml asks. Initially the ph was adjusted to 6.6 with sodium hydroxide in order to attain the ph value necessary to carry out the biodegradation process. The ask was continuously stirred in an orbital shaker at 250 rpm and the temperature was maintained at 30 C. Periodical samples were taken in order to analyze the operating parameters ph, phenol, COD and TOC concentration Fluidized-bed bioreactor The continuous biodegradation of phenol was carried out with a uidized-bed bioreactor FBB). The reactor body consisted of a jacketed cylinder, made of methyl methacrylate 420 mm height and 140 mm internal diameter). An enlargement at the top 170 mm height and 250 mm internal diameter) was provided, to ensure the degassi cation of the liquid and to avoid the loss of bioparticles. The working volume was 3 l. The reactor was thermostated. Sterile air was supplied from the bottom of the column at a ow rate of 85 l/h 43 vvm) through a porous glass distributor average pore diameter 20±40 lm). The bioreactor worked with immobilized cells of the microorganism and bioparticles were suspended in the column by air up ow. Several openings at the top of the bioreactor allowed for the insertion of di erent probes ph, DO), the addition of chemicals nutrients, acid/base and antifoam agents), and liquid sampling. The reactor was fed using peristaltic pumps of variable ow rates. A software developed in a previous work Vallejo, 1994) was used for the implementation of control closed loops in the bioreactor. The structure of the control can be considered as a feed-forward system with a feedback loop cascade. This software allowed for the acquisition and the registration of the main control variables dissolved oxygen, ph and temperature). The following experimental conditions were maintained in the bioreactor: ph 6.6, temperature 30 C and air ow, 43 vvm. In order to attain the initial ph necessary to carry out the biodegradation, the addition of sodium hydroxide was necessary. Initially the reactor was operated in batch mode: the bioreactor was initially lled up with a solution containing the bioparticles necessaries to attain a concentration level of 5 mg cells/l solution, plus 250 mg/l of phenol from diluted industrial wastewater. No nutrients were added during the biodegradation process and air was supplied to provide the dissolved oxygen necessary for the bacteria 2±4.5 mg/l). When almost complete phenol degradation was attained, an amount of the industrial wastewater necessary to obtain again a phenol concentration of 250 mg/l in the bioreactor was added, in order to obtain a better adaptation of the bioparticles. When no phenol was detected in the e uent, the reactor was switched to continuous ow conditions, feeding the reactor with a diluted solution of the industrial wastewater 250 mg phenol/l). Initially, a HRT of 4 days was xed in the bioreactor. When a steady-state was reached no phenol or a very low phenol concentration in the e uent was

3 observed), the in uent loading rate was increased, raising the phenol concentration in the feed from 250 to 500 mg phenol/l, until the steady-state was again obtained. Then, phenol in the in uent was still increased and the phenol concentration range studied was 250±2500 mg/l. Periodical samples of the mixed liquor were taken throughout all the experiments in order to measure the concentrations of phenol, COD and TOC Analytical methods G. Gonzalez et al. / Bioresource Technology ) 137± Phenol was determined by HPLC, using a Nucleosil 120 C-18 column mm inner diameter; 5 lm particle size) in combination with a Waters LC-Spectrophotometer. The mobile phase was acetonitrile: water 65:35 v/v) pumped at 0.6 ml/min. COD was measured according to standard methods APHA, 1993). TOC was determined by a Shimadzu Analyzer, TOC-5050 model. A selective electrode Ingold U 457-S7) connected to a ph-meter Aqua Lytic ph 21) was used for ph determinations. 3. Results and discussion 3.1. Batch experiments: in uence of adaptation on the degradation capacity Fig. 1. In uence of adaptation on the phenol degradation capacity when immobilized cells and an initial phenol concentration of 1000 mg/l were used for the biodegradation process. }) without any adaptation; M) after one adaptation; ) after two adaptations. Several experiments were carried out varying the initial phenol concentration of the wastewater in the range 200±1000 mg phenol/l, to obtain the maximum concentration of phenol that could be degraded by P. putida. The results showing a degradation capacity of 500 mg phenol/l and 25 h gave the maximum degradation. Then new batch experiments were programmed in order to increase both the capacity and rate of phenol degradation. The culture was subjected to successive adaptation tests Zilli et al., 1993), from low to high concentrations of phenol, using as inoculum the e uent from the preceding run. The method was described in a previous work Gonzalez and Herrera, 1995). The results showed an increase in the phenol degradation capacity from 500 to 1000 mg/l when acclimated bioparticles were employed and the greater the number of adaptations the lesser was the time necessary to carry out the biodegradation. When a phenol concentration of 1000 mg/l was used, the degradation time decreased from 340 h one adaptation) to 260 h for the second adaptation and also the lag phase decreased, as shown in Fig. 1. Similar results were obtained when phenol concentrations of 200 and 500 mg/l were tested. P. putida was able to degrade the compounds, other than phenol, present in the industrial wastewater. The change in TOC and phenol concentrations along with time was followed for every experiment and the contribution of phenol to the TOC and the TOC corresponding to the other compounds determined. The results obtained for TOC total TOC phenol and TOC non-phenol when immobilized cells of P. putida and an initial phenol concentration of 1000 mg/l were used, are shown in Fig. 2. It was observed that an increase in the number of adaptations led to an appreciable decrease in the lag phase during the degradation of phenol, but the degradation of the other compounds did not start until a percentage 33% approximately) of the phenol had been removed. The phenol degradation was complete in every experiment. However, only the 75% of the non-phenolic compounds present in the industrial wastewater had been removed. The maximum phenol concentration degraded, 1000 mg phenol/l was lower than that obtained with P. putida Fig. 2. Experimental values for TOC total N), TOC phenol r) and TOC non-phenolic j and Paris predicted ± ), when an initial phenol concentration of 1000 mg/l, and adapted immobilized cells were used.

4 140 G. Gonzalez et al. / Bioresource Technology ) 137±142 and model solutions of phenol Gonzalez et al., 2001): 2000 mg phenol/l. This result could indicate the negative in uence of the non-phenolic compounds on the biodegradation process Kinetic model The Haldane equation has been frequently used to describe the phenol biodegradation process by pure or mixed cultures Allsop et al. 1993), Buitron et al. 1998), and Gonzalez and Herrera 1995). Usually, the parameters from the Haldane equation have been obtained from batch experiments. Some authors Hutchinson and Robinson, 1988; Wang et al., 1996; Magbanua et al., 1994) have introduced in the model equation the e ect of other substrates, di erent from phenol. In addition, some authors Wang et al., 1996) have proposed a competitive cross-inhibitory kinetics while others Hutchinson and Robinson, 1988) proposed a dual-substrate growth model, including the inhibition of both phenol and non-phenolic compounds in the kinetic model. However, when acclimated microorganisms were used, the Monod model appeared to be suitable to describe the biodegradation process, since the inhibition due to the substrate could be neglected. Nevertheless, this model could not take into account the inhibition due to the presence of refractory compounds in the wastewater. On the other hand, when relatively small concentrations of phenol are tested, the model of Paris is frequently used Paris et al., 1982; Gonzalez and Herrera, 1995). The Paris and Monod models were selected to t the experimental data obtained in our batch experiments with immobilized cells of P. putida, due to the simplicity of both models. The Paris model considers a second-order kinetics order one for both substrate and microorganisms) and it was applied successfully in a wide range of phenol concentrations 200±1000 mg/l). Fig. 2 shows data obtained when an industrial wastewater with 1000 mg phenol/l TOC total ˆ1000mgC=l;TOC phenol ˆ750mgC=l; TOC non-phenolic ˆ 250 mgc=l was tested. As can be observed, there is a good agreement between experimental and tted values. The kinetic parameters were k x 2: l=mgh and Y xs 0.15 mg microorg/mg phenol). The Monod model was also applied when experiments with high phenol concentration and adapted microorganisms were made. Results with the industrial wastewater, containing 1000 mg phenol/l. TOC total ˆ 1000 mgc=l; TOC phenol ˆ 750 mgc=l; TOC non-phenolic ˆ 250 mgc=l are shown in Fig. 3 and the kinetic parameters were l max 0:03 h 1 ; K s 280 mg=l and Y xs 0.15 mg/mg). The comparison of these results with those obtained when data from batch experiments with model solutions Fig. 3. Experimental values for TOC total N ; TOC phenol r and TOC non-phenolic j and Monod predicted ± ), when an initial phenol concentration of 1000 mg/l, and adapted immobilized cells were used. were tested Gonzalez and Herrera, 1995) shows the lower values of kinetic parameters when the industrial wastewater was used in the biodegradation process. This fact indicates less degradation capacity and a slower kinetic, probably due to the non-phenolic compounds present in the industrial wastewater Continuous biodegradation process Continuous essays were planned to study the biodegradation of phenolic compounds from industrial wastewaters. The methodology used in the tests was the same as described when model solutions were treated Gonzalez et al., 2001). With the aim of attaining a good adaptation of the bioparticles, the bioreactor was started in the batch mode, introducing a solution of the industrial wastewater containing 250 mg/l of phenol and the amount of bioparticles required to attain a nal concentration of 5 mg microorg/l solution. The initial ph value was xed at 6.6 and no reagents were needed during the biodegradation process to maintain this ph value. Complete phenol degradation was observed after 6 days and the bioreactor was again operated in batch mode to attain a better adaptation of the bioparticles, introducing the amount of industrial wastewater to attain initially 250 mg/l. After 10 days, the biodegradation was completed. The operation was then switched to the continuous mode and the reactor was fed with a diluted industrial wastewater containing a concentration of 250 mg phenol/l, and a ph value of 4.5. In continuous mode, the reactor was operated with 4 days HRT phenol loading rate of 62.5 mg/l d) and no phenol was detected in the e uent. Then, the phenol concentration in the in uent was increased progressively from 250 to 2500 mg/l of phenol phenol loading rates from 62.5 to 625 mg/l d), working with a constant HRT of 4 d.

5 G. Gonzalez et al. / Bioresource Technology ) 137± Fig. 4. In uent and e uent concentrations of phenol and TOC in the FBB operated under di erent phenol concentrations and a HRT of 4 d: phenol concentration in the in uent ) and e uent and ); TOC in the in uent M) and e uent N. Fig. 4 shows the behavior of the system operated with di erent in uent phenol concentrations, in terms of in uent and e uent concentration values of phenol and TOC. Within the range of concentrations between 250 and 2000 mg/l, the system operation was very stable and the phenol and TOC in the e uent were always lower than 1 mg phenol/l and 15 mg C/l, respectively. Similar results were obtained when the COD content was followed: e uent lower than 120 mg O 2 =l and removal e ciencies higher than 95%. The removal of nonphenolic compounds, evaluated as TOC non-phenolic, was always higher than 95%. At higher concentrations 2500 mg phenol/l), the operation in the bioreactor became unstable and the phenol and TOC concentration in the e uent increased progressively. After this time, the bioreactor was fed with industrial wastewater containing 2000 mg phenol/l, and an e uent free of phenol was obtained after a 10- days operation. With these conditions, the bioreactor showed the same performance for more than 60 days data not shown). The phenol loading rate susceptible to degradation 0.5 g phenol/l d) was lower than that obtained in a previous paper Gonzalez et al., 2001) when model solutions were treated 4 g phenol/l d). However, the biodegradation e ciencies are comparable >98% for phenol, >95% for COD and TOC) and the phenol concentration in the e uent was lower than 1 mg/l. 4. Conclusions The experimental results showed that it was possible to treat industrial e uents containing high phenol concentrations. When it was operated in batch mode, phenol concentrations up to 1000 mg/l were degraded and high removal e ciencies >90%) were attained for both phenol and non-phenolic compounds. The operation of an FBB bioreactor, with a phenol loading rate of 500 mg/l d COD: 1500 mg/l d; TOC: 525 mg/l d), was proven to be e cient and e uent phenol concentrations were lower than the highest discharge limit permitted for the Spanish legislation 1 mg/l). The COD and TOC contents in the e uent were always lower than 120 and 40 mg/l, respectively.

6 142 G. Gonzalez et al. / Bioresource Technology ) 137±142 References APHA, Standard Methods for the Examination of Water and Wastewater, 19th edition. APHA, Washington, DC, pp. 5-12±5-16. Allsop, P.J., Chisti, Y., Moo-Yung, H., Sullivan, G.R., Dynamics of phenol degradation by Pseudomonas putida. Biotechnol. Bioeng. 41, 572±580. Bravo, P., Gonzalez, G., Continuous ethanol fermentation by immobilized yeast cells in a uidized bed reactor. J. Chem. Tech. Biotechnol. 52, 127±134. Buitron, G., Gonzalez, A., Lopez, A., Mar in, L.M., Biodegradation of phenolic compounds by an acclimated activated sludge and isolated bacteria. Water Sci. Technol. 37, 371±378. Gonzalez, G., Herrera, G., Biodegradation of phenol by free and immobilized cells of Pseudomonas putida. Acta Microbiol. Polonica 44, 285±296. Gonzalez, G., Herrera, G., Garc ia, Ma.T., Pe~na, Ma.M., Biodegradation of phenol in a continuous process: comparative study of stirred tank and uidized bed bioreactors. Bioresource Technol. 76 3), 245±251. Hannaford, A.M., Kuek, C., Aerobic batch degradation of phenol using immobilized Pseudomonas putida. J. Indust. Microbiol. Biotechnol. 22, 121±126. Hutchinson, D.H., Robinson, C.W., Kinetics of the simultaneous batch degradation of p-cresol and phenol by Pseudomonas putida. Appl. Microbiol. Biotechnol. 29, 599±604. Kapoor, A., Kumar, R., Kumar, A., Sharma, A., Prasad, S., Application of immobilized mixed bacterial culture for the degradation of phenol present in oil re nery e uent. J. Environ. Sci. Health A 33, 1009±1021. Lakhwala, F.S., Goldberg, B.S., Sofer, S.S., A comparative study of gel entrapped and membrane attached microbial reactors for biodegrading phenol. Bioprocess Eng. 8, 9±18. Lin, S.H., Chuang, T.S., Combined Treatment of phenolic wastewater by wet air oxidation and activated sludge. Technol. Environ. Chem. 44, 243±258. Loh, K.C., Chung, T.S., Ang, W.F., Immobilized-cell membrane bioreactor for high-strength phenol wastewater. J. Environ. Eng. ± ASCE 126, 75±79. Magbanua, B.S., Hoover, P.A., Campbell, P.J., Bowers, A.R., The e ect of cosubstrates on phenol degradation kinetics. Water Sci. Technol. 30, 67±77. Mordocco, A., Kuek, C., Jenkins, R., Continuous degradation of phenol at low concentration using immobilized Pseudomonas putida. Enzyme Microbial Technol. 25, 530±536. Paris, D.F., Wolfe, N.L., Steen, W.C., Structure activity relations in microbial transformation of phenols. Appl. Environ. Microbiol., 135±158. Patterson, J.W., Industrial Wastewater Treatment Technology. Butterworths, Boston. Vallejo, L., Desarrollo de un sistema de adquisicion de datos. Aplicacion al control de un proceso de fermentacion. Tesis de licenciatura, Universidad de Valladolid. Wang, K.W., Baltzis, B.C., Lewandowski, G.A., Kinetics of phenol biodegradation in the presence of glucose. Biotechnol. Bioeng. 51, 87±94. Zilli, M., Converti, A., Lodi, A., del Borghi, M., Ferraiolo, G., Phenol removal from waste gases with a biological lter by Pseudomonas putida. Biotechnol. Bioeng. 41, 693±699.