Biodegradation of high phenol concentration by ...

Biochemical Engineering Journal 30 (2006) 174–183

Biodegradation of high phenol concentration by activated sludge in an immersed membrane bioreactor B. Marrot ∗ , A. Barrios-Martinez, P. Moulin, N. Roche Laboratoire de Procédés Propres et Environnement UMR-CNRS 6181, Université Paul Cézanne d’Aix-Marseille III, Europôle de l’Arbois, BP 80, 13545 Aix en Provence Cedex 04, France Received 29 June 2005; received in revised form 17 January 2006; accepted 28 March 2006

Abstract The effect of adaptation of mixed culture in the phenol biodegradation has been studied. The degradation experiments have been conducted at different phenol concentrations from 0.5 to 3 g L−1 . Biological treatment has been shown to be economical, practical and it leads to a complete removal of phenol. High concentrations of phenol are inhibitory for growth; so it is for the rates of substrates utilization that are greater at low initial concentrations. Haldane kinetics model for single substrate was used to obtain maximum specific growth rates (µm = 0.438 h−1 ), half saturation (Ks = 29.54 mg L−1 ) and substrate inhibition constant (Ki = 72.45 mg L−1 ). Although the concentration in phenol is significant, these results are in agreement with those reported in the literature for phenol removal abilities in different systems and the Haldane model is still acceptable. © 2006 Elsevier B.V. All rights reserved. Keywords: Activated sludge; Phenol; Biodegradation; Kinetic model; Haldane; Immersed membrane bioreactor

1. Introduction Phenols, i.e. hydroxy compounds of aromatic hydrocarbons, and its derivatives are widely used as raw materials in many petrochemical industries and petroleum refineries (washing and conditioning of the alkaline or acid products), chemical and pharmaceutical industries (dyes, pesticides, drugs, . . .) and others industries like pulp and paper mills, coking operations, coal refining, tannery and foundries (washing of the gas effluents). Thus, the presence of phenols in water generally comes from this industrial pollution. The increasing presence of phenols represents a significant environmental toxicity hazard; therefore, the development of methods for the removal of phenols from industrial wastewater has generated significant interest. As the toxicity of phenolic compounds is important, their concentration (up to several grams per liter) unfortunately inhibits or even reduces microorganisms in municipal biological wastewater treatment plant [1]. The presence of phenols strongly reduces the biological biodegradation of the other components; what makes the process of degradation of phenols so difficult. Accord-

∗

Abbreviations: MBR, membrane bioreactor; RCF, relative centrifugal force Corresponding author. Tel.: +33 442 908 511; fax: +33 491 289 407. E-mail address: [email protected] (B. Marrot).

1369-703X/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.bej.2006.03.006

ing to the literature, several processes are used to remove phenolic compounds like granular or biological activated carbon, H2 O2 /UV processes, O3 /UV processes, Fenton processes (Fe2+ /H2 O2 ) solvent extraction, membrane processes. Conventional processes have been mostly physicochemical processes but since they cause secondary problems in the effluents (for example, phenol becomes chlorophenols if chlorination is used), biological treatments are preferred for large-scale removal of this type of pollutants. It is one of the reasons why activated sludge reactors have been widely used for phenol removal from industrial wastewater. 2. Biological treatment 2.1. Acclimated activated sludge Biological treatment is a practical and not very expensive solution to treat this kind of effluents compared to chemical one (not need to add chemicals); because various populations of microorganisms in the activated sludge are able to degrade organic compounds and most of effluents can be biological degraded. The species most often present are pseudomonas, flavobacterium, achromobacter, rhomobacterium, azobacter, micrococcus, bacillus alkaligenes, arthrobacter, ycobacterium, aeromonas, nocardia and lophomonas. Their respective propor-

B. Marrot et al. / Biochemical Engineering Journal 30 (2006) 174–183

Nomenclature kd Ki Ks MLSS S t X

decay coefficient (h−1 ) substrate inhibition coefficient (mg L−1 ) half saturation coefficient (mg L−1 ) mixed liquor suspended concentration (g L−1 ) concentration of substrate (mg L−1 ) time (h) concentration of biomass (mg L−1 )

Greek letters µ specific growth rate (h−1 ) maximum specific growth rate (h−1 ) µm Subscript 0 initial value tion will depend mainly on the specific substrate concentration and the potentiality of growth. When microorganisms are cultivated, they consume substrates for their growth and for their energy. Nevertheless, biological treatment of the phenolic compounds is not easy because of the proper toxicity of phenol for the microorganisms. Even with weak concentrations (lower than 200 mg L−1 ) the phenolic compounds can cause the inhibition of the microbial growth. The limits of the biological process are related to the acclimation of the biomass to degrade phenol and, of course, to the variability of the wastewater composition. During this phase there is a selection and a multiplication of specialized microorganisms. Buitron et al. [2] isolated and identified the microorganisms responsible for the phenol degradation. Four types of Gram-negative unicellular bacteria were obtained from the acclimated consortium: Aeromonas sp., Pseudomonas sp., Flavomonas oryzihabitans and Chryseomonas luteola. There are other microorganisms able to degrade phenol like: Alacaligenes sp., Sarcinas, Desulfovibrio sp., Bacillus alkaligenes, Acinetobacter and more [3]. Alexievaa et al. [4] demonstrated that T. cutaneum R57 has all the properties of an efficient phenoldegrading microorganism. Two groups of degrading bacteria are mainly specified and used: the Rhodococci (Rhodococci shows considerable morphological) like Rhodococcus spp. [5–8] and Pseudomonads like Pseudomonas putida. P. putida seem to have the highest degradative potential. That is why a great number of studies upon the degradation of phenols by these bacteria has been done [9–14]. But opinions are divided, it seems that acclimated activated sludge degrades the phenolic compounds more efficiently than the pure strains by one to two orders of magnitude faster and Annadurai et al. [15] showed that the mixed liquors had a best ability for phenol degradation than pure activated sludge and P. putida. To obtain a specified biomass from activated sludge, all these microorganisms are generally used in two different ways, either the cells are immobilized within calcium alginate gel beads for the fluidized-bed bioreactor or cells grow as a suspended culture (free cells) for the bioreactor. In a lot of case, the continuous process is considered when the acclimation step in stirred

175

batch reactor is finished. In comparison of these two processes [9,11,13,14] it seems that better phenol degradation efficiency is obtained with a fluidized-bed bioreactor rather than with a stirred tank. It shall be explained by the substrate inhibition effect; for example in the study of Chung et al. [14], free cells can degrade phenol only up to about 600 mg L−1 whereas this level becomes up to 1000 mg L−1 for immobilized cells. But it could be observed the formation of intermediate catechol for immobilized cells; phenomenon has not been detected in the case of free cells. So, it seems that the use of immobilized cells with a medium diffusion resistance could be a mean for detection of intermediates during substrate degradation. Moreover, phenol degradation reaction is slower in the immobilized systems. 2.2. Effect of temperature and pH on phenol degradation Phenol degradation seems to be determined by some environment factors such as temperature and pH [9,10,12–15]. Regarding the temperature effect, authors are almost unanimous and find a higher phenol removal efficiency near 30 ◦ C. However the rate and the extent of degradation is relatively sensitive to deviations outside the optimal range [9]. A variation of 5 ◦ C may cause a decrease in phenol degradation rate of at least 50% at the lower end and almost 100% at the higher end. The difference between phenol removal efficiency at 30 ◦ C is probably due to the higher production of metabolites at this temperature [12]. Moreover, at this temperature, the degradation rate seems better for free than immobilized cell system (1.45 times higher). Chung et al. [14] found an optimal temperature of 30 ◦ C for the two processes but different optimal pH values: 6.8 for immobilized cells and 8.0 for free cells. This difference is related to the carboxylic parts of alginate that attract H+ around them. Hence, the pH of buffer solution has to be lowered to provide optimal concentration of H+ for P. Putida. The follow-up of the medium pH can be an indicator of the phenol degradation and one of the factors significant in the success of the biological treatment. A slight reduction is observed as biomass grows and pH variation increases when the initial phenol concentration increases [5,10]. The decrease in pH suggests that biological degradation of phenol occurs and with a stable pH of about 7 (and a sufficient oxygen supply) phenol was successfully degraded. The pH significantly affects the biochemical reactions required for phenol degradation; tests with pure P. putida could not efficiently resist pH change [15]. pH medium affects the substrate decomposition rate and phenol decomposition leads to a considerable decrease in pH. Consequently, phenol degradation is deteriorated as the medium pH deviates from neutral condition. For Aksu and Gönen [16] pH affects the surface charge of the cells of the activated sludge biomass. The surface charge of biomass is predominantly negative over the pH range of 3–10. Phenol could be expected to become negatively charged in phenoxide ion above a pH of 9. Below a pH of 3, the overall surface charge on cells becomes positive due to isoelectric point of activated sludge so the electrostatic attraction between phenol and activated sludge biomass will be insignificant [16]. In conclusion, it seems that these studies indicate a best pH range [6.5; 7.5] for the phenol degradation from effluents.

176


Table 1 Composition of growth medium Acclimation Culture

Process

T (◦ C)

pH

Nutrient medium

[Phenol] (mg L−1 )

Phenol loading rate (g L−1 d−1 )

Growth medium (mg L−1 )

Pseudomonas putida ATCC 17484 Biotype B

Batch

30

6.6

5–1000

0.5–4

Pseudomonas putida

Batch

–

–

Beef extract, 1 g L−1 ; yeast extract, 2 g L−1 ; peptone, 5 g L−1 ; agar, 15 g L−1 ; NaCl, 5 g L−1 ; K2 HPO4 , 2.39 g L−1 Glucose, 2 g L−1

500

–

Pseudomonas putida DSM 548

Batch

26

6.8

Agar

1–100

–

25

6.6

Agar, 11 g L−1

500

–

KH2 PO4 , 420; K2 HPO4 , 375; (NH4 )2 SO4 , 244; NaCl, 30; CaCl2 ; MgSO4 ·7H2 O, 61.4; FeCl2 ·4H2 O, 4.7 CaCl2 ·4H2 O, 69.9; NaCl, 8; KNO3 , 103; NaNO3 , 698; MgSO4 ·7H2 O, 100; NTA, 100; FeSO4 ·7H2 O, 2; ZnSO4 ·7H2 O, 0.1; MnSO4 ·5H2 O; 0.043; H3 BO3 , 0.3; CoSO4 ·7H2 O, 0.24; CuSO4 ·5H2 O, 0.01; NiSO4 ·7H2 O, 0.02; NaMoO4 ·2H2 O, 0.03; Ca(OH)2 ,0.5; EDTA, 5; KH2 PO4 , 544.4; Na2 HPO4 , 2148.9; (NH4 )2 SO4 , 30 KH2 PO4 , 420; K2 HPO4 , 375; (NH4 )2 SO4 , 244; NaCl, 15; CaCl2 ·2H2 O, 15; MgSO4 ·7H2 O, 50; FeCl3 ·6H2 O, 5.4 KH2 PO4 , 840; K2 HPO4 , 750; (NH4 )2 SO4 , 488; NaCl, 60; CaCl2 , 60; MgSO4 , 60; FeCl3 , 60 KH2 PO4 , 420; K2 HPO4 , 375; (NH4 )2 SO4 , 244; NaCl, 15; CaCl2 ·2H2 O, 15; MgSO4 ·7H2 O, 50; FeCl3 ·6H2 O, 54 –

Pseudomonas putida

Pseudomonas putida CCRC14365

Batch

30

7

Beef extract, 3 g L−1 ; peptone, 5 g L−1 ; mineral salt

100

–

Pseudomonas putida

Batch

30

7

–

–

Pseudomonas putida ATCC 31800 and activated sludge

Batch

30–36

7 to 9

Beef extract, 1 g L−1 ; yeast extract, 2 g L−1 ; peptone, 5 g L−1 ; NaCl, 5 g L−1 ; agar, 15 g L−1 Beef extract, 1 g L−1 ; yeast extract, 2 g L−1 ; peptone, 5 g L−1 ; NaCl, 1 g L−1 ; agar, 20 g L−1 ; glucose, 0.5, 0.6, 0.7, 0.8 g L−1

–

0.25

Activated sludge (2500 mg L−1 )

Batch

25

–

–

300

–

(NH4 )2 SO4 , 500–800 and medium mineral. **pH 6.25 the Pseudomonas putida could not resist pH change t = 48 h **phenol loading of 0.25 g L−1 day (NH4 )2 SO4 , 240; K2 HPO4 , 45; NaOH, 120; MgCl2 ·6H2 O, 15; CaCl2 , 4; FeCl3 ·6H2 O, 0.6

Biodegradation of phenol Process

Reference

Batch reactor fluidized-bed reactor

[11]

Trickling bed reactor

[12]

Batch reactor

[10]

Immobilized beads

[9])

Free suspension Immobilized Ca-alginate beads

[14]

Immobilized on calcium alginate

[17]

Batch reactor

[15]

Activated sludge

[18]


177

Table 1 (Continued ) Acclimation

Biodegradation of phenol Process

Reference

Culture

Process

T (◦ C)

pH

Nutrient medium

[Phenol] (mg L−1 )

Phenol loading rate (g L−1 d−1 )

Growth medium (mg L−1 )

Aeromonas sp.; Pseudomonas sp.; Flavomonas oryzihabitans; Chryseomonas luteola and activated sludge Pseudomonas putida

Batch

15

–

N, P and trace of elements

40 Phenol 2,4 DCP; 2,4,6 TCP

0.03

30 of P; 30 of 2,4 DCP; 30 of 2,4,6 TCP; nutrients

Batch reactor

[2]

Immobilized beads

30

6.6

Agar, 11 g L−1

2.6–10.6

0.25–0.5

Batch reactor

[19]

Trichosporon cutaneum R 57 Rohodococcus sp. DCB-p0610

Inmovilized cells Inmovilized on GAC; aginate beads by entrapment

28

6

Beer agar and 0.67% yeast nitrogen bases NaHPO4 , 10 g; KH2 PO4 , 1 g; (NH4 )2 SO4 , 3 g; FeCl3 , 0.018; NaCl, 0.2 g; CaCl2 , 0.08; MgSO4 ·7H2 O, 0.2 g; yeast extract, 0.5 g L−1 ; phenol, 10 g

0.5 Isopropylbenzene 2000

0.25–0.5

(NH4 ) NO3 , 1; (NH4)2 SO4 , 0.5; NaCl, 0.5; MgSO4 ·7H2 O 0.5; KH2 PO4 , 1.5; K2 HPO4 , 0.5; CaCl2 , 0.01; FeSO4 ·7H2 O, 0.01 in 1 L of solution –

Immobilized cells Continuous reactor

[20]

Activated sludge

Batch

30

–

Phenol variable (750–4500)

Bioreactor

[21]

6 to 8; 4 to 7

7.2

From the literature (Table 1) we know that optimum conditions (medium, temperature, pH) to acclimatize bacteria to this particular substrate which is the phenol. It is also noticed that few reports on phenol biodegradation using real effluents has been done; that is why we have worked with activated sludge from a wastewater treatment plant. The aim of this study was double: (i) to acclimatize bacteria to phenol under experimental conditions easier to implement but less favorable for the growth of these microorganisms (limitation of the substrate, pH and temperature different of the ideality). (ii) To investigate the possibility of phenol biodegradation at high initial concentrations and to study the microorganism growth kinetics using Andrews–Haldane model during biodegradation of phenol for single substrate. This study is carried out with an immersed membrane bioreactor.

2.9 g L−1 day from GAC 2.1 g L−1 day from aginate beads

1.44, 2.40 and 4.32

NaHPO4 , 10 g; KH2 PO4 , 1 g; (NH4 )2 SO4 , 3 g; FeCl3 , 0.018; NaCl, 0.2 g; CaCl2 , 0.08; MgSO4 ·7H2 O, 0.2 g; yeast extract, 0.5 g L−1 ; phenol, 10 g MgSO4 ·7H2 O, 9.4; CaSO4 ·2H2 O, 4.7; Na2 HPO4 ·2H2 O, 752; KH2 PO4 , 63.92; NH4 Cl, 18.8 and trace minerals solution 0.47 ml: Na2 ·EDTA, 2500; ZnSO4 ·7H2 O, 100; MnCl2 ·6H2 O, 30; H3 BO3 , 300; CoCl2 ·6H2 O, 200; CuCl2 ·2H2 O, 10; NiCl2 ·2H2 O, 20; Na2 Mo4 ·2H2 O, 900; Na2 SeO3 , 20 and FeSO4 ·7H2 O, 1000

[5]

3. Materials and methods 3.1. Apparatus The main advantage resulting from the microbial consortium formed by acclimated activated sludge is the interaction between all the species present in flocs. But, it is necessary among other things to pay attention to the phenomenon of flocculation since, on the basis of mass transfer, flocs would be less efficient than free cells in phenol degradation. That is the reason why the acclimatization of microorganisms was carried out in a continuous bioreactor. The substrate is fed uninterrupted at 12 L day−1 with a peristaltic pump. Ultrafiltration hollow fibre membranes (0.01 ␮m) immersed in the bioreactor make possible to preserve a constant volume and

178


to make no attempt to separate SS into active and inactive portions. 3.3. Synthetic effluent

Fig. 1. Laboratory scale bioreactor. (A) Bioreactor (90 L); (B) membrane (hollow fibre module); (C) air flowrate (0–10 L min−1 ); (D) peristaltic pump (cole parmer instrument model 7454-95); (E) peristaltic pump (cole parmer instrument model 7453-77); (F) feed medium (30 L); (G) permeate (20 L).

to preserve the microorganisms themselves in the bioreactor. An immersed membrane bioreactor (A) was employed for the cultivation of organisms (Fig. 1). The reactor had a total volume of 90 L (with active volume of 60 L). The mixed liquor (10 g L−1 ) was only agitated by aeration, from a compressed air source through a diffuser (H) in the bioreactor’s base at a flow rate of 10 L min−1 (C). The aeration rate was sufficient to maintain the dissolved oxygen concentration near the saturated level (4.5 mgO2 L−1 ) during cultivation, so growth is assumed not to be limited by oxygen. For the determination of the kinetics of phenol degradation we have used batch reactors (3 L) in which we put the activated sludge coming from reactor A. 3.2. Analytical methods For the determination of phenol, samples from the aeration tank are filtered through filters having a pore size of 0.45 ␮m. The phenol is analysed by gas chromatography (Chrompack CP 9001, Middelburg, Netherlands) with a capillary column [i.d. (32 mm), length (15 m), film thickness (0.25 ␮m), SGE, Courtaboeuf, France] and detected by a flame ionization detector (FID). Phenol concentration is determined with a calibration curve made from known phenol standard. Each experiment is at least duplicated under identical conditions. Reproducibility of the concentration measurements remains within 5%. The dissolved oxygen concentration in the aeration tank is directly measured with a specific probe (Consort 932, Fisher Bioblock, Germany). The pH is measured with pH meter Hanna Instrument. pH adjustment (pH 6.5) of the reactor content was performed with sodium hydroxide. Suspended solids (SS) are measured by centrifuging at 3900 rpm (1900 RCF) for 15 min a sample of 25 mL of sludge, and by drying the deposit at 105 ◦ C for 24 h. The concentrations of the activated sludge are quantified as dry weight of the suspended solid. So, the determined mass of microorganisms contains active cells but also dead or inactive cells, which do not show biological activity. However, because of the great difficulty of identifying this inactive portion of SS, we have decided

Considering that the diversity of bacteria in activated sludge makes them able to degrade most of the compounds, we have collected waste activated sludge from urban wastewater treatment plant of Aix-en-Provence (France—175,000 inhabitant equivalent). They are used as the source of microorganisms for phenol degradation in this study. The initial concentration near 3 g L−1 has been increased to 10 g L−1 by membrane filtration (hollow fibre module), under aeration. The nutrient medium was carried out according to the bibliography (Table 1). As seen before, there are various mediums according to the type of microorganisms to be developed, but, in almost all cases there are a source of carbon, a source of nitrogen, a source of phosphorus, and oligoelements such as: Mg, Zn, Ca, Fe, Cu, Ni, K, Na and Co. However, phenol degradation became less efficient by increasing the concentrations of carbon (glucose) and nitrogen ((NH4 )2 SO4 ) sources (and with increasing temperature) [15]. In order to prevent any deficiency we have chosen a weight ratio of 100:5:1 for phenol:nitrogen:phosphorus (C:N:P). Effectively, as phenolics compounds are rich in carbon content, we have used phenol as a sole carbon source [18,22]. The salt medium was only composed by the following mineral salts: KH2 PO4 , K2 HPO4 , NH4 Cl, CaCl2 , MgSO4 (Table 2). 4. Results and discussion 4.1. Phenol acclimation and degradation Mixed culture (activated sludge) was grown in the presence of glucose and then adapted to increasing concentrations of phenol from an initial minimum inhibition concentration of 0.5 gphenol L−1 + 0.4 gglucose L−1 to the highest 3.0 gphenol L−1 . The first 2 days the source of substrate is a mixture of urban wastewater and glucose (0.25 gglucose L−1 ); from the third day and during 3 days we mixed phenol (0.46 gCOD L−1 ) and glucose (0.5 gCOD L−1 ). From the seventh day, the activated sludge was grown in the presence of phenol as the sole carbon source and then adapted to increasing concentrations of phenol over a period of 4 months; the reactor was continuously operated during this period. The sludge was supposed to be acclimated to the system when phenol was completely degraded in repeated uses Table 2 Mineral salt concentrations in the activated sludge bioreactor at the beginning and at the end of the acclimation step Parameters

Initial (mg L−1 )

Final (mg L−1 )

C6 H5 OH NH4 Cl KH2 PO4 K2 HPO4 CaCl2 Mg SO4

1060 200 23 29 7 13

3030 570 69 85 21 39


179

Fig. 2. Phenol degradation and variation of mixed liquor suspended solid in batch reactor—initial phenol concentration: (a) 0.1 g L−1 , (b) 0.2 g L−1 , (c) 0.5 g L−1 , (d) 0.7 g L−1 , (e) 1.0 g L−1 and (f) 1.5 g L−1 .

in fixed time intervals. At the end, the culture was acclimated and the phenol concentration was maintained at a concentration of 3 g L−1 . The kinetics of phenol degradation by the sludge was assessed in a batch reactor using different concentrations of phenol (Fig. 2). Sludge acclimated from the continuous process was taken and phenol was added at different concentration (between 0.5 and 3 g L−1 ) at the start of the experiment. The working volume of the reactor was maintained at 3 L, overhead stirrer was used to keep the organism in suspension and to ensure the mixing of air bubbles. Control experiments, in the same conditions with water containing phenol and substrate, were done to evaluate the possible degree of phenol removal by gas transfer. It was observed that the phenol concentration remained constant during the experiment, so we considered that aeration and mixing do not involve volatilization of phenol. Fig. 2 shows for the same initial cell concentration, that the higher the concentration of phenol is the more time it takes to be consumed. The phenol concentration in the culture medium decreased clearly when the microorganisms started to grow. The acceleration of the growth rate when phenol concentration decreased was characteristic

of a substrate inhibition phenomenon. The extent of phenol degradation and the time required for phenol degradation varied as a function of the initial phenol concentration in the medium. Until a concentration of 2.5 g L−1 , the experimental results show that the mixed culture has a potential for the removal of phenol from wastewaters. However, the activated sludge process could not cope with phenol at loading rate in excess; the inhibition effects of phenol as substrate have become predominant above the concentration of 3.5 g L−1 . This behavior is characteristic of a toxic substrate metabolism. This kind of limit has been reported by Watanabe et al. [18] and Kibret et al. [21] when the inlet phenol loading was increased until 2.0 and 2.5 g L−1 , respectively. For the same initial biomass concentration, the extent of phenol degradation and the time required for phenol degradation varied as a function of the initial phenol concentration in the culture medium. Degradation of more than 80% of phenol during all experiments occurred in less than 6 h. If one wants completely to degrade phenol, the period should be prolonged, for example up to 54 h for an initial phenol concentration of 3 g L−1 .

180


The effect of a toxic compound on a treatment process is quantified in terms of the inhibition coefficient, Ki . It should be noted that when Ki is very large the Haldane equation simplifies to the Monod equation (implies that the culture is less sensitive to substrate inhibition). So, low values of Ki show that the inhibition effect can be observed at low phenol concentration. The apparent KS value is of practical importance because a bacterium expressing activity with a lower apparent KS value can efficiently remove the pollutant down to lower concentration. At higher substrate concentrations, S KS , the Haldane equation becomes: µ=

µm 1 + (S/Ki )

(2)

Graphical determination of Ki is obtained by linearization of Eq. (2): Fig. 3. Effect of increased concentrations of phenol for the same degree of biodegradation (80%) vs. the time taken for degradation.

The results shown in Fig. 3 indicate that an increase of phenol concentration affected the time of phenol removal (80%) it seems that we have an adequate estimation by the use of a exponential law. This curve shows the inhibiting character of this molecule and its potential biodegradadability. The more significant the initial phenol concentration is, the longer it will take time for the microorganisms to degrade 80% of this phenol. 4.2. Determination of kinetic parameters—substrate inhibition model

µm S KS + S + (S 2 /Ki )

(1)

with µm the maximum specific growth rate (h−1 ), S the concentration of substrate (mg L−1 ), Ks the half-saturation coefficient or the substrate affinity constant (mg L−1 ) (the affinity of a bacterium for a substrate) this constant is defined as the substrate concentration at which µ is equal to half µmax and Ki is the substrate inhibition coefficient (mg L−1 ).

(3)

But this linearized Haldane’s equation could not represent the growth kinetics [23]. Microorganism growth rate in a batch reactor may be modelled by the following equation: dX µm S = X − kd X dt KS + S + (S 2 /Ki )

(4)

with X the concentration of biomass (mg L−1 ). Moreover, during exponential phase endogenous coefficient kd may be neglected. Eq. (4) therefore reduces to the following equation: dX = µX dt

The kinetics of phenol biodegradation by microbial populations has been largely studied [10,14,23–26]. The mathematical model that has been found in order to get the statistically best description of the growth kinetics is the Haldane model (1965). This model describes relatively well the microbial growth in the presence of a substrate which is at the same time an inhibitor of the metabolism of this microbial population that is pure or heterogeneous (mixed). Even at low concentrations, phenol had a substantial inhibitory effect on the specific growth rate (µ). The specific growth rate tends to increase with the substrate (Monod type relationship), but µ rises to a peak and finally decreases due to the inhibitory effect of S as its concentration is increased. The Haldane model that has frequently been used to describe this inhibition is: µ=

S 1 1 + = µm µm Ki µ

(5)

In the exponential phase, the specific growth rate is obtained from the end of exponential growth phase and is calculated by: µ=

ln(X2 /X1 ) t2 − t 1

(6)

In the literature we found a lot of data concerning the degradation of phenol by P. putida and comparatively few data with mixed culture or acclimated activated sludge. Kinetic constants obtained by the Haldane model for phenol biodegradation are given in Table 3. Since phenol is an inhibitory substrate for most species, the Haldane equation has been frequently used to model phenol degradation, and, compared to the Monod equation, has often provided a better representation of the observed data. But, many authors do not find the model suitable to the strong phenol concentration. Various kinetic relationships have been suggested attempting to describe the joint dependence of µ on S as substrate and S as inhibitor, for higher phenol concentration [24,26,27]. It has been determined that during phenol degradation various metabolic intermediates are produced and accumulated. Unfortunately, the Haldane model does not take into account the effect of metabolic intermediates on phenol degradation (model too complicated).


181

Table 3 Kinetic constants for phenol biodegradation in batch reactor (Haldane equation) Culture

T (◦ C)

pH

[Phenol]max a (mg L−1 )

µm (h−1 )

Ks (mg L−1 )

Ki (mg L−1 )

Reference

P. putida DSM 548 P. putida CCRC14365 P. Putida P. putida ATCC17514 P. putida ATCC700007 P. putida Q5 P. putida MTCC1194 P. fluoroescens Acinetobacter + Pseudomonas Acinetobacter Trichosporon cutaneum R57 Candida tropicalis Mixed culture Mixed culture Mixed culture Mixed culture Mixed culture Mixed culture Mixed culture

26 30 – – 30 10 29.9 – 30 30 – – 25 15 28 20 – – Ambient

6.8 6.8 – – 7 7 7.1 – 6.8 – – – – – 6.6 6.8 – – 6.5

100 610 800 – 200 200 – – – 350 – – 1450 40 900 – – – 2500

0.436 0.33 0.90 0.897 0.051 0.119 0.305 0.823 0.418 0.83 0.42 0.48 0.143 0.258 0.260 0.326 0.746 0.131 0.438

6.19 13.9 6.93 12.204 18 5.27 36.33 71.4 29.37 1.5 110 11.7 87.44 3.9 25.4 19.2 53.9 5.0 29.5

54.1 669 284.3 203.678 430 377 129.79 241 370 250 380 207.9 107.06 217 173 229.3 516 142 72.4

[10] [14] [24] [28] [29] [30] [23] [25] [31] [27] [4] [32] [26] [2] [33] [34] [25] [35] This study

a

Represent the maximum concentration being able to be degraded.

Although some authors do not consider the Haldane model adapted, we have calculated the specific growth rate using Eq. (1) whereas the dependence of activated sludge specific growth rate on the concentration of phenol is shown in Fig. 4. The specific growth rate, µ, for each value of the initial phenol concentration, S0 , has been determined in the exponential growth phase. The phenol degradation rate increases with phenol concentration and declines with further increases in phenol concentration as substrate inhibition effects became important. The three-parameter model of Haldane was fitted to the experimental data using Statistica 6.0 software. We have worked on the Levenberg–Marquardt algorithm using 95% confidence intervals for minimizing the sum of square of residuals. It is well known that the nonlinear optimization procedure is strongly sensitive to the initial values and the variation intervals of the model parameters [4]. For this reason, the search for the values of the kinetic constants was constrained within boundaries predetermined on the basis of the process knowledge and experimental data.

Fig. 4. Specific growth rate as a function of the initial phenol concentration. Experiment data () and Haldane model (—).

The comparison between experimentally obtained specific growth rate µ and the one that predicted by the model shows that the growth kinetics of phenol could be represented by Haldane’s growth kinetics model very well. The values of kinetics constants, µm , Ks , Ki , obtained in this work as 0.438 h−1 , 29.54 mg L−1 and 72.45 mg L−1 , respectively, are compared with other published data in Table 3. As we have used activated sludge, which is a mixture of many microorganisms, we expected to obtain a possible competition for the common substrate resulting a lower growth rate, compared to a pure culture. However, we notice that we have a specific growth rate of the same order of magnitude as the one obtained for a pure culture. The small magnitude of KS values indicates that for microbial species utilizing phenol, the maximum growth rate could be reached quickly, if substrate inhibition has not been a factor. Compared to others culture studies (pure and mixed), the value of Ki indicates that the inhibition effect can be observed in a mild concentration range. Mixed culture had a good resistance to substrate inhibition. Thus, although we have a strong biomass concentration (10 g L−1 ) and that we inject into our batch reactor a concentration in phenol much more significant than those in literature, ours kinetics values are in the range of literature values. It is often difficult to determine the limits of a model; we guess that the Haldane model remains acceptable even with the strong phenol concentrations. But, at substrate concentrations between 0.5 and 1.0 g L−1 , the inadequacy of the Haldane model seems to be apparent. The Haldane model predicted shorter completes degradation times than these measured. The observations are substantiated by other, like Wang and Loh (P. Putida) [24], Hao et al. (Acinetobacter) [27] and Nuhoglu and Yalcin (Mixed culture) [26] who have pointed out an inadequacy of the Haldane substrate inhibition kinetics at higher substrate concentrations. However in the case of a mixed culture and to our knowledge this inadequacy seems to be reached for phenol concentrations weaker than in this study (for example 100 mg L−1 for Ref. [26]).

182


Unfortunately, in the literature most of the studies do not give any information on microorganisms concentration in their reactor. 5. Conclusion A mixed culture (activated sludge from WWTP) was grown in continuous culture using phenol as the limiting substrate. The experimental results show that it is possible to treat effluents containing high phenol concentration (up to 1.0 g L−1 ) by activated sludge at typical biomass concentrations of a membranes bioreactor (around 10 g L−1 ). We know the importance of the acclimatization plant for supporting the microorganisms which have the enzymatic material necessary to the degradation of phenol and revealing a new population which is adapted to this toxic agent and which is able to consume it like substrate. But, this acclimatization is relatively easy to realize by using a reduced salt medium and without inevitably being under conditions of temperature and pH the most favorable for the development of these microorganisms, subject to respect steps of increasing phenol concentration. So, the biodegradation of phenol has become an alternative to the traditional physical and chemical methods that can be costly and dangerous to handle (H2 O2 , O3 ). From batch experiments carried out at different phenol concentration with acclimated biomass, the kinetic constants of the Haldane equation have been determined: specific growth rates (µm = 0.438 h−1 ), half saturation (KS = 29.54 mg L−1 ) and substrate inhibition constant (Ki = 72.45 mg L−1 ). There is a good fit between the degradation data and the calculated kinetics parameters; Haldane model remains substantiate even with the strong phenol concentrations. The comparison with other studies using of the mixed or pure cultures shows a certain disparity in the kinetic parameters; the need for better understanding of substrate inhibition kinetics models is still apparent. In order to validate the model parameters, experiments will be carried out with an external MBR with initially the same synthetic substrate then a real effluent from an oil industry. Acknowledgments Mexican Institute of Petroleum contributes to this study. This work was supported by Sfere, Conacyt and IFR 112. References [1] W. Gernjak, T. Krutzler, A.G.S. Malato, J. Caceres, R. Bauer, A.R. Fernández-Alba, Photo-Fenton treatment of water containing natural phenolic pollutants, Chemosphere 50 (2003) 71–78. [2] G. Buitron, A. Gonzalez, L.M. Lopez-Marin, Biodegradation of phenolic compounds by an acclimated activated sludge and isolated bacteria, Water Sci. Technol. 37 (4–5) (1998) 371–378. [3] A. Lante, A. Crapisi, A. Krastanov, P. Spettoli, Biodegradation of phenols by laccase immobilised in a membrane reactor, Process Biochem. 36 (1–2) (2000) 51–58. [4] Z. Alexievaa, M. Gerginova, P. Zlateva, N. Peneva, Comparison of growth kinetics and phenol metabolizing enzymes of Trichosporon cutaneum R57 and mutants with modified degradation abilities, Enzyme Microb. Technol. 34 (3–4) (2004) 242–247.

[5] S.L. Pai, Y.-L. Hsu, N.-M. Chong, C.-S. Sheu, C.-H. Chen, Continuous degradation of phenol by Rhodococcus sp. immobilized on granular activated carbon and in calcium alginate, Bioresour. Technol. 51 (1) (1995) 37–42. [6] A. Hidalgo, A. Jaureguibeitia, M.B. Prieto, C. Rodr´ıguez-Fernández, Biological treatment of phenolic industrial wastewaters by Rhodococcus erythropolis UPV-1, Enzyme Microb. Technol. 31 (3) (2002) 221–226. [7] M.B. Prieto, A. Hidalgo, J.L. Serra, M.J. Llama, Degradation of phenol by Rhodococcus erythropolis UPV-1 immobilized on Biolite® in a packed-bed reactor, J. Biotechnol. 97 (2002) 1–11. [8] R. Margesin, P.A. Fonteyne, B. Redl, Low-temperature biodegradation of high amounts of phenol by Rhodococcus spp. and basidiomycetous yeasts, Res. Microbiol. 156 (1) (2005) 68–75. [9] A. Mordocco, C. Kuek, R. Jenkins, Continuous degradation of phenol at low concentration using immobilized Pseudomonas putida, Enzyme Microb. Technol. 25 (6) (1999) 530–536. [10] A.M.G. Monteiro, R.A.R. Boaventura, A.E. Rodrigues, Phenol biodegradation by Pseudomonas putida DSM 548 in a batch reactor, Biochem. Eng. J. 6 (1) (2000) 45–49. [11] G. González, M.G. Herrera, M.T. Garc´ıa, M.M. Peña, Biodegradation of phenol in a continuous process: comparative study of stirred tank and fluidized-bed bioreactors, Bioresour. Technol. 76 (3) (2001) 245–251. [12] C.S.A. Sá, R.A.R. Boaventura, Biodegradation of phenol by Pseudomonas putida DSM 548 in a trickling bed reactor, Biochem. Eng. J. 9 (3) (2001) 211–219. [13] N.K. Pazarlio˘glu, A. Telefoncu, Biodegradation of phenol by Pseudomonas putida immobilized on activated pumice particles, Process Biochem. 40 (2005) 1807–1814. [14] T.P. Chung, H.Y. Tseng, R.S. Juang, Mass transfer effect and intermediate detection for phenol degradation in immobilized Pseudomonas putida systems, Process Biochem. 38 (10) (2003) 1497–1507. [15] G. Annadurai, R.S. Juang, D.J. Lee, Microbiological degradation of phenol using mixed liquors of Pseudomonas putida and activated sludge, Waste Manage. 22 (7) (2002) 703–710. [16] Z. Aksu, F. Gönen, Biosorption of phenol by immobilized activated sludge in a continuous packed bed: prediction of breakthrough curves, Process Biochem. 39 (5) (2004) 599–613. [17] K. Bandhyopadhyay, D. Das, P. Bhattacharyya, B.R. Maiti, Reaction engineering studies on biodegradation of phenol by Pseudomonas putida MTCC 1194 immobilized on calcium alginate, Biochem. Eng. J. 8 (3) (2001) 179–186. [18] K. Watanabe, S. Hino, N. Takahashi, Responses of activated sludge to an increase in phenol loading, J. Ferment. Bioeng. 82 (5) (1996) 522–524. [19] G. Banerjee, Treatment of phenolic wastewater in RBC reactor, Water Res. 31 (4) (1997) 705–714. [20] T. Godjevargova, D. Ivanova, Z. Alexieva, N. Dimova, Biodegradation of toxic organic components from industrial phenol production waste waters by free and immobilized Trichosporon cutaneum R57, Process Biochem. 38 (6) (2003) 915–920. [21] M. Kibret, W. Somitsch, K.H. Robra, Characterization of a phenol degrading mixed population by enzyme assay, Water Res. 34 (4) (2000) 1127–1134. [22] S. Timur, N. Pazarlioglu, R. Pilloton, A. Telefoncu, Detection of phenolic compounds by thick film sensors based on Pseudomonas putida, Talanta 61 (2) (2003) 87–93. [23] A. Kumar, S. Kumar, S. Kumar, Biodegradation kinetics of phenol and catechol using Pseudomonas putida MTCC 1194, Biochem. Eng. J. 22 (2004) 151–159. [24] S.J. Wang, K.C. Loh, Modeling the role of metabolic intermediates in kinetics of phenol biodegradation, Enzyme Microb. Technol. 25 (1999) 177–184. [25] P. Kumaran, Y.L. Paruchuri, Kinetics of phenol biotransformation, Water Res. 31 (1) (1997) 11–22. [26] A. Nuhoglu, B. Yalcin, Modelling of phenol removal in a batch reactor, Process Biochem. 40 (3–4) (2005) 1233–1239. [27] O.J. Hao, M.H. Kim, E.A. Seagren, H. Kim, Kinetics of phenol and chlorophenol utilization by Acinetobacter species, Chemosphere 46 (6) (2002) 797–807.

B. Marrot et al. / Biochemical Engineering Journal 30 (2006) 174–183 [28] S. Dikshitulu, B.C. Baltzis, G.A. Lewandowski, Competition between two microbial populations in a sequencing fed-batch reactor: theory, experimental verification, and implications for waste treatment applications, Biotechnol. Bioeng. 42 (1993) 643–656. ¨ Mehmetoglu, Kinetics [29] T. Abuhamed, E. Bayraktar, T. Mehmetoglu, U. model for growth of Pseudomonas putida F1 during benzene, toluene and phenol biodegradation, Process Biochem. 39 (2004) 983–988. [30] G. Kotturi, C.W. Robinson, W.E. Inniss, Phenol degradation by a psychrotrophic strain of Pseudomonas putida, Appl. Microbiol. Biotechnol. 34 (4) (1991) 539–543. [31] A.G. Livingston, H.A. Chase, Modeling phenol degradation in a fluidised bed bioreactor, AICHE J. 35 (1989) 1980–1992.

183

[32] J. Yan, W. Jianping, L. Hongmei, Y. Suliang, H. Zongding, The biodegradation of phenol at high initial concentration by the yeast Candida tropicalis, Biochem. Eng. J. 24 (3) (2005) 243–247. [33] U. Pawlowsky, J.A. Howell, Mixed culture biooxidation of phenol. Determination of kinetic parameters, Biotechnol. Bioeng. 15 (1973) 889–896. [34] R.W. Szetela, T.Z. Winnicki, A novel method for determining the parameters of microbial kinetics, Biotechnol. Bioeng. 23 (1981) 1485– 1490. [35] P.D.D. Adamo, A.F. Rozich, A.F. Daudy, Analysis of growth data with inhibitory carbon sources, Biotechnol. Bioeng. 26 (1984) 397– 402.