Simultaneous biotreatment of Polycyclic Aromatic Hydrocarbons and dyes in a one-step bioreaction by an acclimated Pseudomonas strain

Bioresource Technology 198 (2015) 181–188

Contents lists available at ScienceDirect

Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

Simultaneous biotreatment of Polycyclic Aromatic Hydrocarbons and dyes in a one-step bioreaction by an acclimated Pseudomonas strain María S. Álvarez, Ana Rodríguez, Ma Ángeles Sanromán, Francisco J. Deive ⇑ Department of Chemical Engineering, University of Vigo, 36310 Vigo, Spain

h i g h l i g h t s
Acclimated Pseudomonas stutzeri efficiently remediates PAH and dye-polluted effluents.
Viable biotreatment medium and optimum operating conditions were determined.
Kinetics of the biotreatment and its economic advantages were ascertained.

a r t i c l e

i n f o

Article history: Received 31 July 2015 Received in revised form 26 August 2015 Accepted 27 August 2015 Available online 1 September 2015 Keywords: Polycyclic aromatics Azo dyes Biodegradation Biosorption Process simulation

a b s t r a c t A Pseudomonas stutzeri strain acclimated to the presence of neoteric contaminants has been proposed for simultaneously remediating an effluent polluted with Polycyclic Aromatic Hydrocarbons and a diazo dye. The pollutants chemical nature imposed a strict control of both the medium composition and the operating conditions. pH, temperature and agitation rates of 7.0, 37.5 and 146 rpm, respectively, led to optimum levels of contaminant removal (higher than 60%) after RSM optimization. The validity of these conditions was checked at flask and bioreactor scale and the kinetics of the biotreatment was elucidated. The simulation of this one-step process applied at larger scale for the remediation of a 200,000 m3/year-effluent from a leather factory was compared with a conventional two-steps option. Great reductions in treatment times and in investment and manufacturing costs were concluded, proving the promising potential of the proposed process. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Globally, during the last decades economic development has marched hand in hand with an environmental collapse due to the thoughtless introduction of polluted-industrial effluents. More and more regulations prompt the academic and industrial community to come forward with competitive and environmentally friendly solutions. One of the sectors causing great environmental concerns is the leather and textile industry, since they generate a variety of pollutants ranging from surfactants, heavy metals, sulfides, acids, alkalis, and dyes to Polycyclic Aromatic Hydrocarbons (PAHs) (Li et al., 2010). The importance of the latter two kinds of contaminants has been underscored by current international environmental legislation (USEPA, 2008; EU-EEB, 2005). The health and environmental risk of these aromatic compounds has been well documented, as they involve carcinogenic, mutagenic and toxic effects, and are considered to bear a great recalcitrance (Simarro et al., 2011; Haritash and Kaushik, 2009; Bae and Freeman, 2007; ⇑ Corresponding author. Tel.: +34 986818723. E-mail address: [email protected] (F.J. Deive). http://dx.doi.org/10.1016/j.biortech.2015.08.125 0960-8524/Ó 2015 Elsevier Ltd. All rights reserved.

Zaharia and Suteu, 2013). These concerns have urged the search of treatment technologies to remove them from the environment, and a number of physico-chemical alternatives have been successfully proposed like adsorption, ozonation, electrochemistry and flocculation (Vecino et al., 2013; Sancar and Balci, 2013; Iglesias et al., 2013; Devesa-Rey et al., 2012). However, economic and operational inconveniences have favored the application of biotechnological tools to remediate PAH- and dye-polluted effluents, since they usually involve lower cost and improved social perception (Deive et al., 2010; Moscoso et al., 2012a, 2013a). Hitherto, research works have mainly focused on the treatment of a mixture of PAHs or dyes independently (Moscoso et al., 2012b; Álvarez et al., 2013), while a lack of knowledge is detected in the finding of suitable strategies to remediate all the contaminants when present together in the same effluent. A successful outcome should satisfy three main requirements: (i) the chemical structure of the contaminants, (ii) the selected microbial agent, and (iii) the operating conditions of the process (Haritash and Kaushik, 2009; Moscoso et al., 2012a). Attending to these demands, the aspect related to the chemical nature should be firstly addressed to ensure that the contaminant

182

M.S. Álvarez et al. / Bioresource Technology 198 (2015) 181–188

is susceptible to be bioremediated in the aqueous effluent. In this sense, PAHs are thermodynamically stable molecules, with elevated hydrophobicity, so they should be solubilized by adding surfactants in order to make them bioavailable (Yang et al., 2015). On the other hand, dyes are usually hydrophilic and possess complex aromatic molecular structures that are classified on the basis of the chromophore group (Robinson et al., 2001). In this sense, azo dyes make up the most common group of direct dyes, since about 60–70% of the produced dyes belong to this category (Bae and Freeman, 2007). In relation to the bioremediation agent, different microbial strains have been proposed as suitable candidates to yield high levels of PAHs (Ghosh et al., 2014; Peng et al., 2013) or dye removal (Liu et al., 2014; Manenti et al., 2014). However, a lack of studies is detected on the finding of microorganisms able to concomitantly biotreat both kinds of contaminants. In previous investigations, we have underscored the potential of a Pseudomonas stutzeri strain for the degradation of PAHs (Moscoso et al., 2012a,b,c, 2013a, 2015), metal working fluids (Moscoso et al., 2012d), or pesticides (Moscoso et al., 2013b), and its capacity to be adapted to neoteric solvents like ionic liquids (Álvarez et al., 2015). This fact was explained in terms of a genetic alteration, as the acclimated strain throve under pollutant concentrations up to 10 times higher by means of the synthesis of an exopolysaccharide (Álvarez et al., 2015). Therefore, this flexible nature has encouraged us to apply it for the combined bioremediation of both kinds of pollutants, which is the main aim of this work. Special heed must be paid to the operating conditions selected to develop the bioprocess once the biotreatment medium was designed. Factors like pH, temperature, and agitation should be optimized prior to sketch the bioremediation process at real scale. Valuable means to reach this target are computational tools like simulation software (SuperPro Designer v8.5) and experimental designs (Design Expert 7.0), saving time and money to reach the optimum process. In summary, considering the pollutant charge of textile and leather waste effluents, three model PAHs of low (phenanthrene, PHE) and high molecular weight (pyrene, PYR, and benzo[a]anthracene, BaA) and a common azo dye (Reactive Black 5) have been selected. This scenario raises problems related to the different nature of the pollutants such as the degree of hydrophobicity and the carbon source, which will compel us to optimize the biotreatment medium and propound the ideal range of operation. Additionally, the bioprocess will be kinetically characterized both at flask and bioreactor scale by fitting to known models and these data will be employed to simulate the process and will pay off in a onestep biotreatment strategy. 2. Methods 2.1. Chemicals The pollutants Reactive Black 5 (RB5), phenanthrene (PHE), pyrene (PYR) and benzo[a]anthracene (BaA) (structures shown in Fig. S1) were acquired from Sigma–Aldrich, with purities higher than 99%. The same supplier provided the non-ionic surfactant Tween 80, benzyl benzoate, salts of the medium and chloroform. Glucose was purchased from Scharlau, and HCl and hexane were supplied by Prolabo. 2.2. Microorganism The bacterium P. stutzeri CECT 930 was acquired from the Spanish Type Culture Collection (ATCC 17588). This bacterium was acclimatized for two months in a lab-scale bioreactor in the

presence of C2C1imC2SO4 (0.2 mM) under controlled agitation, aeration and temperature as previously reported (Álvarez et al., 2015). 2.3. Bioremediation medium Minimal medium (MM) was used, composed of (g/L in distilled water): Na2HPO42H2O 8.5, KH2PO4 3.0, NaCl 0.5, NH4Cl 1.0, MgSO47H2O 0.5, CaCl2 0.0147. MM also contained trace elements as follows (mg/L in distilled water): CuSO4 0.4, KI 1.0, MnSO4H2O 4.0, ZnSO47H2O 4.0, H3BO3 5.0, FeCl36H2O 2.0. Different concentrations of glucose and Tween 80 were also included in the culture medium as carbon source and solubilizing agent, respectively. 2.4. Biotreatment at flask scale It was carried out in 250 mL-Erlenmeyer flasks containing 50 mL of MM. The pH was initially adjusted to 7.0 and the MM was autoclaved at 120 °C for 20 min. The dye (0.04 g/L) and PAHs (100 lM each) were sterilized by filtration through a 20 lm filter prior to the addition to the autoclaved medium in order to avoid any possible alteration of the chemical structure of the pollutants. The flasks were inoculated (3% v/v) with previously obtained cell pellets, which were them incubated in an orbital shaker (Thermo Fisher Scientific 496) at 37 °C and 146 rpm. 2.5. Biotreatment at bioreactor scale The scaling up of the process was carried out by operating in a 2-L bioreactor (model BIOSTATÒB-MO), filled with 1.5 L of medium. The temperature and initial pH was fixed at the optimum operating conditions. It was inoculated with actively growing cells (3% v/v) and air was sparged at a continuous rate of 0.17 vvm (volumes per minute, which involves the use of an air flowrate of 0.25 L/ min). 2.6. Analytical methods 2.6.1. Biomass determination Cells were harvested by centrifugation (10 min, 9300 g, and 4 °C), and the supernatant was reserved for pollutants analysis. Biomass concentration was measured by turbidimetry at 600 nm in a UV–vis spectrophotometer (UV-630 Jasco), and the values were converted to grams of cell dry weight per liter using a calibration curve. (Biomass (g/L) = 0.5663Absorbance – 0.0401, R2 = 0.996). 2.6.2. Adsorption test PAHs biosorption over the biomass was determined as follows. 50 mL of culture medium were taken and centrifuged for 10 min at 5.900 g and 4 °C. The supernatant was withdrawn and biomass was freeze-dried during 4 h at 40 °C and 7.9  105 atm using a TelStarCryodes. Afterwards, 10 mL of hexane were added and ultrasounds were applied (Bransonic 3510) for 30 min. Again, the sample was centrifuged for 10 min and 100 lL of supernatant were taken into a vial, where 10 lL of Internal Standard (IS) were added. Samples were analyzed by GC–MS as explained later on. 2.6.3. Dye decolourisation Dye concentration in the culture media was analyzed by UV–vis spectrophotometry taking into account the maxima wavelength recorded for RB5 dye (597 nm). Each decolourisation value was the mean of two parallel experiments. Abiotic controls (without microorganisms) were always included. Decolourisation (D) was expressed in terms of percentage units by using the expression:

183

M.S. Álvarez et al. / Bioresource Technology 198 (2015) 181–188

ðIi  If Þ  100 Ii

Inc., Minneapolis, USA). A second order polynomial equation was applied to correlate the dependent and independent variables:

being Ii and If are the abiotic control and culture concentration of the dye, respectively. The assays were done in duplicate, and the experimental error was less than 15%. 2.6.4. PAHs and intermediates determination Aliquots (1 mL) of supernatant were added over 0.8 g of MgSO47H2O following 0.1 mL of HCl 1 M and 1 mL of hexane. They were shaken for 1 h and an aliquot of 100 lL was collected from the organic phase and 10 lL of internal standard (benzyl benzoate) were added. PAHs concentration in supernatant was analyzed using an Agilent GC 6850 gas chromatograph equipped with a HP-5MS column (30 m  0.25 mm; 0.25 lm, Agilent), operating with hydrogen as carrier gas, and coupled to an Agilent MSD 5975C mass spectrometer. Injections (1 lL) of samples were made up in split mode (10:1) split relation; GC oven was programmed under the following conditions: 50 °C for 4 min and 10 °C/min to 280 °C for 10 min. The mass spectrometer was operated in SIM mode. Intermediates were detected by adding 25 ml of chloroform to 250 mL of supernatant and the pH was adjusted to 2 to favor the extraction of intermediates formed during the biodegradation process. The water content in the organic phase was removed by addition of anhydrous sodium sulfate and subsequently filtered. The sample was then introduced in a rotatory vacuum concentrator (RVC 2-25 CCHRIST/CHRIST CF04-50 SR), and the residue was dissolved in chloroform. The same gas chromatograph equipment served our goal to detect the intermediate metabolites, and 1 lL-injections of the samples were made up in split mode (2:1 split relation); GC oven was programmed under the following conditions: 50 °C for 5 min, then 5 °C/min to 280 °C for 5 min. The mass spectrometer was operated in SCAN mode.

Dye removal (%)

2.6.5. Statistical design The statistical design was analyzed through the ANalysis Of VAriance (ANOVA) using Design ExpertÒ 9.0.0 software (Stat-Ease

Y i ¼ x0 þ x1 T þ x2 pH þ x3 agitation þ x4 T  pH þ x5 T  agitation þ x6 pH  agitation þ x7 T 2 þ x8 pH2 þ x9 agitation

3. Results and discussion The proposal of an efficient bioremediation process for effluents containing PAHs and azo dyes requires the finding of a suitable medium and operating conditions. Therefore, prior to approach the operation at bioreactor scale and simulating the treatment of a real-scale effluent, the optimization at small scale is sketched. 3.1. Optimization of medium and treatment conditions The first aim is to find a suitable biotreatment medium allowing the solubilization of the hydrophobic contaminants (PAHs) and without negatively interfering in the bioremediation of the hydrophilic pollutants (azo dye RB5). As previously demonstrated, the acclimation of a P. stutzeri strain to the presence of neoteric contaminants like imidazolium-based ionic liquids triggered a permanent alteration at a gene level that led to the synthesis of an exopolysaccharide (Álvarez et al., 2015). This modification widened the proved versatility of this microbial strain for the remediation of different kinds of pollutants (Moscoso et al., 2012b,d, 2013b), as this biopolymer can help to increase its potential for the biotreatment of dyes, by promoting dye adsorption phenomena. Therefore, the addition of a non-ionic surfactant to the biotreatment medium is critical, as it assists in increasing hydrophobic contaminants bioavailability but may solubilize the synthesized exopolysaccharide, thus hindering the dye removal.

100

100

80

80

60

60

40

40

20

20

0

0 0

2

4

6

8

10

2

0

Tween concentration (g/L) 10

8

6

2

where Y i is the response variable (contaminant remediation) x0 is a constant, x1, x2 and x3 are the regression coefficients for linear effects; x4, x5 and x6 are the regression coefficients for interaction effects, and x7, x8 and x9 are the regression coefficients for quadratic effects, and T, pH and agitation are the independent variables.

PAHs solubilisation (%)

Dð%removalÞ ¼

4

Glucose concentration (g/L) Fig. 1. PAHs solubilization (d) and RB5 removal (s) for different concentrations of glucose and Tween 80.

184

M.S. Álvarez et al. / Bioresource Technology 198 (2015) 181–188

Table 1 Values of the coefficients for the equation of effects in the remediation of RB5, PHE, PYR and BaA. Linear effects

Interaction effects

Quadratic effects

Pollutant

x0

x1

x2

x3

x4

x5

x6

x7

x8

x9

RB5 PHE PYR BaA

381.4 52.6 381.0 131.4

8.04 10.37 6.38 3.50

186.6 101.7 211.3 31.28

0.055 1.69 1.26 0.595

3.28 0.307 0.374 1.111

0.012 0.019 0.008 0.006

0.026 0.118 0.126 0.027

0.131 0.125 0.106 0.081

23.40 8.11 16.77 0.830

0.002 0.005 0.005 0.003

Parameters in bold are significant (P < 0.05) (c.f. Supplementary Material).

Fig. 2. Effect of pH and temperature in the biotreatment of dyes and PAHs at the optimum agitation (150 rpm).

As Tween 80 and glucose may act as carbon source in cultures of P. stutzeri, as previously reported (Moscoso et al., 2012a; Álvarez et al., 2015), the combination of different concentrations of both compounds may be crucial to reach a compromise between PAH bioavailability and exopolysaccharide solubilization. Since 10 g/L is the carbon source concentration leading to the highest levels of biomass, declining concentrations of Tween 80 were combined with growing compositions of glucose ([Glucose], [Tween 80] in g/L = (0.0, 10), (2.5, 7.5), (5.0, 5.0), (7.5, 2.5), (9.0, 1.0) and (9.9, 0.1)), and the data are presented in Fig. 1. These data evidence the existence of an optimum ratio (9.0, 1.0), as the PHE, PYR and BaA are completely solubilized while the decolorisation of the azo dye RB5 overtook 60%. Once this ratio was chosen, Response Surface Methodology (RSM) based on a central composite face-centred design was

applied to optimize the contaminants remediation when using temperature, pH and agitation as independent variables. The operation range was defined after a previous screening, and the designed experimental 34 runs (including five replicates of the central point to evaluate the reliability of the data) are presented in the supplementary material (Table S1) together with the bioremediation percentages. The analysis of the statistical parameters shown in Table S2 demonstrates that a quadratic model is significant (P < 0.0001) for a suitable description of the 4 responses under study (RB5, PHE, PYR and BaA removal). Hence, the coefficients for defining the equation of effects are shown in Table 1. In a visual inspection of the data compiled in this table, it becomes patent that the influence of pH, agitation and temperature is significant for almost all the contaminants, while the interaction and quadratic effects seem to be more dependent on the

185

M.S. Álvarez et al. / Bioresource Technology 198 (2015) 181–188

led to average contaminants removal levels higher than 60% for PHE, PYR, BaA and RB5, respectively.

Table 2 Parameters of the logistic model to characterize the kinetic growth and pollutant remediation by the adapted P. stutzeri at small and bioreactor scale. Xmax (g L1)

lmax (h1)

R2

Flask scale Bioreactor scale

0.44 0.01

3.55 6.27

0.22 0.97

0.91 0.98

Contaminant

D0 (%)

Dmax (%)

lD (h1)

R2

Flask scale RB5 PHE PYR BaA

0.1 8.6 6.2 0.5

73.7 70.6 56.1 64.5

0.31 0.15 0.11 0.52

0.98 0.93 0.97 0.93

Bioreactor scale RB5 PHE PYR BaA

0.1 7.9 7.7 3.8

78.1 83.6 68.5 77.2

0.59 0.13 0.11 0.29

0.98 0.95 0.93 0.94

3.2. Scaling-up, modeling and simulation of the process at real scale After the operating conditions and biotreatment medium were selected, the scaling-up of the process was approached. Then, a bench-scale bioreactor will provide valuable data prior to simulate the process at real scale. Therefore, the first step was carrying out the biological reaction at the optimum conditions, going from flask to bioreactor scale. A kinetic model widely applied in the characterization of bioremediation processes allowed describing two important variables of the process, biomass concentration and pollutant removal (Deive et al., 2010):

X¼

X h
max  ln

1þe

Pollutant Removal (%)

Biomass concentration (g/L)

contaminant under study. In this context, the graphical representation of the response surfaces for each contaminant at optimum agitation rates (150 rpm) is shown in Fig. 2. The visualization of the data licenses to draw a distinction between azo dye and PAHs, as a result of their completely different chemical nature, even though both of them share the presence of condensed aromatic rings. On the one hand, maximum dye removal levels can be attained at pH values lower than 6.5 and temperatures higher than 32.5 °C. On the other hand, PAHs removal is only feasible for pH values higher than 6.5 for all the temperatures under study. The numerical optimization carried out by using the software Design ExpertÒ 9.0.0 led to the conclusion that pH 7.0, T = 37.5 °C and agitation rates of 146 rpm

D¼

X max 1 X0

lm t

D h
max  1þe

ln

Dmax 1 D0

lD t

i

i

where X and D are the biomass (g/L) and contaminant removal (%) at an specific moment of the biotreatment (t), X0 and D0 are the initial biomass and removal, Xmax and Dmax are the maximum biomass and pollutant removal, and lm and lD are the maximum specific growth rate and maximum specific remediation rate (h1). The values of the regression coefficients R2 listed in Table 2 (always higher than 0.9) evidence the suitability of the proposed models to get a deep insight in the kinetic characteristics of the process carried out at flask and bioreactor scale at the optimum conditions obtained previously. The data presented in Fig. 3 also

10.0

10.0

7.5

7.5

5.0

5.0

2.5

2.5

0.0

0.0

75

75

50

50

25

25

Biomass concentration (g/L)

X0 (g L1)

Pollutant removal (%)

Scale

0

0 0

20

40

Time (h)

60

0

20

40

60

80

Time (h)

Fig. 3. Biomass concentration (s) and removal of RB5 (4), PHE (5), PYR (h) and BaA ( ) in the biotreatment processes carried out at flask (black) and bioreactor scale (blue). Dots represent the experimental data and solid lines are employed for the modelled data. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

186

M.S. Álvarez et al. / Bioresource Technology 198 (2015) 181–188

Table 3 Parameters of the model from Marqués et al. for the remediation process at bioreactor scale.

D0 (%) m (%/g) n (%/g/h) R2

RB5

PHE

PYR

BaA

0 4.6 0.1 0.90

0 8.7 0.0 0.89

0 6.2 0.0 0.90

0.0 9.4 0.0 0.96

makes it evident this adequate description for both the biomass and contaminants remediation. A conscious analysis of the biomass parameters points to the benefits of operating at bioreactor scale, as both the maximum biomass concentration and specific growth rate are enhanced by about 2 and 4 times, respectively. These results are coincident with previous studies tackling the scaling-up of dye-remediation processes from flask to benchscale bioreactors (Deive et al., 2010). These ameliorations are also reflected in the maximum levels of pollutant removal recorded, as an average increase of about 12% and 5% is recorded for the PAHs and RB5, respectively, when going from flask to bioreactor scale. The reason for this boosted behavior can be attributed to the inherent benefits of operating in this kind of stirred tank bioreactor, like the greater mass transfer of contaminants and oxygen promoted by the Rushton impeller. In this line, it has already been well documented the superior performance of this turbine for improving oxygen mass transfer coefficients (Moucha et al., 2003). This is crucial for an efficient biodegradation process because aerobic biodegradation mechanisms demand the existence of molecular oxygen as electron acceptor, thus easing the activation of the substrate through oxygenation reactions biocatalyzed by mono or dioxygenases (Cao et al., 2009). In this vein, GC–MS analysis confirms this hypothesis, since the three PAHs seem to follow the same metabolic route and, after a double hydroxylation of one aromatic ring, its cleavage is eased. Then, depending on the PAH, the stages are repeated up to diethylphthalate and phthalic acid are obtained (c.f. Fig. S2 in Supplementary

Material) which are easily mineralized. The proposed route is in agreement with previous results of our group and other researchers (Moscoso et al., 2012a, 2015; Khanna et al., 2011), which confirms that the acclimated P. stutzeri is able to follow the same metabolic strategies to degrade the contaminants. A deeper insight into the nature of the remediation process can be achieved by applying the model reported by Marqués et al. (1986), and subsequently adapted by Deive et al. (2010), where the remediation is presented as a function of the growth rate and the biomass as can be seen in the following equation:

8 <
D ¼ D0 þ mX 0 h : 1; 0 

9 = elt  i  1; 0 X0 ; ð1; 0  elt Þ X max       X max X0 ln 1; 0  ð1; 0  elt Þ þn l X max

This algorithm relates the degradation efficiency with the growth rate (m = 0), the biomass (n = 0) or both parameters (m – 0 and n – 0). The data obtained are presented in Table 3. The values of the parameters reflect that the remediation of all the contaminants displays a greater dependence on the biomass production, since in all cases m is, at least, more than one order of magnitude higher than n. This behavior is in agreement with the results reported for the remediation of this kind of contaminants independently (Deive et al., 2010; Moscoso et al., 2012a). This higher relationship with the biomass production may be related to the nature of the remediation process, as usually, two subsequent stages are underlying the contaminant removal: biosorption and metabolisation. In this sense, it has been observed that PAHs and di-azo dye RB5 behave differently, and this behavior is confirmed both at flask and bioreactor scale. Thus, while levels of biosorption lower than 35% are recorded for the PAHs (with just 7% for the low molecular weight PAH, PHE), 60% of the RB5 is adsorbed on the bacterial biomass. The reason for the higher affinity of RB5 dye in relation to the PAHs lies again in the different chemical

Fig. 4. Two-step (PFD 1) vs. one-step (PFD 2) process flowsheet diagrams for the industrial biotreatment of PAHs and RB5-polluted effluents as obtained with the software SuperProDesigner v8.5.

M.S. Álvarez et al. / Bioresource Technology 198 (2015) 181–188 Table 4 Treatment capacity, remediation efficiency and scheduling summary for the twostages and one-stage biotreatment processes.

Batch time (h) Batch number per year Total RB5 removal (%) Total PHE removal (%) Total PYR removal (%) Total BaA removal (%)

2-Steps biotreatment

1-Step biotreatment

220.8 309 78.0 95.6 95.6 95.6

52.5 51 79.5 85.0 70.9 78.6

nature of these contaminants. Thus, the ionic character of the dye will ease the establishment of electrostatic interactions with the protonated nitrogen-containing functional groups in microbial cells and proteins, as a consequence of the existence of slightly acidic conditions (Bidisha et al., 2006). This fact also explains the improved results of dye removal previously observed at acid pHs. All in all, the optimized conditions allowed accomplishing high levels of remediation of dyes and PAHs simultaneously. In order to further quantify the advancements, this one-step biotreatment process will be likened with a traditional option including a twostages process: a P. stutzeri mesophilic step to treat the PAHs (with a duration of 150 h) followed by a thermophilic step employing Anoxybacillus flavithermus to decolorize RB5 (with a total time of 12 h), in line with prior investigations (Moscoso et al., 2015; Deive et al., 2010). Both processes are presented in Fig. 4, with a view to ease the analysis between the two options, and they were simulated to remediate a 200,000 m3/year polluted effluent (with 300 mM PAHs and 0.04 mg/L of azo dye) from a leather industry. The software tool employed was SuperProDesigner v8.5 (Intelligen Inc.), as it is a simple way to interactively analyze on a consistent basis the viability of both remediation alternatives at large scale. One of the advantages provided by this program is that it enables to easily peruse the throughput capacity and time utilization of each operation unit. Hence, on the basis of the technical needs indicated above, time requirements, remediation yields, and biomass production, both alternatives were simulated, and the main results for each of them are compiled in Table 4. It becomes patent that the one-stage biotreatment involves a drastic cycle time reduction from 221 h/batch to 53 h/batch, thus allowing the performance of up to 309 batches per annum, while maintaining high levels of pollutants remediation. Additionally, this greater throughput capacity parallels a reduction in fixed capital investment and manufacturing cost up to about 40%. When all this information is taken together, the total costs of effluent treatment are reduced by ten times, which makes it patent the aptness of the proposed process. Apart from that, and given the versatility of the proposed strain in terms of substrates utilization (complex media, metal working fluids, etc.) as already demonstrated in previous research works (Moscoso et al., 2012d), the future search of cheaper nutrients and other operation modes, as well as strategies including biomass recycling will allow decreasing the total costs of the process. 4. Conclusion The present manuscript has demonstrated the technical and economic superiority of a one-step bioremediation process based on an acclimated P. stutzeri strain, to treat an effluent polluted with 3 model PAHs and a model diazo dye. An optimum medium and operating conditions were selected prior to demonstrate the viability of the strategy at bioreactor scale. The kinetic parameters of the process were obtained in order to license its simulation with the software SuperPro Designer, recording enhancements of treatment throughput near to 6 times while reducing the total cost in one order of magnitude.

187

Acknowledgements This work has been supported by the Spanish Ministry of Economy and Competitiveness and EDRF funds (Project CTM201452471-R). F.J. Deive acknowledges Spanish Ministry of Economy and Competitiveness for funding through a Ramón y Cajal contract.

Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biortech.2015.08. 125. References Álvarez, M.S., Moscoso, F., Rodríguez, A., Sanromán, M.A., Deive, F.J., 2013. Novel physico-biological treatment for the remediation of textile dyes-containing industrial effluents. Bioresour. Technol. 146, 689–695. Álvarez, M.S., Rodríguez, A., Sanromán, M.A., Deive, F.J., 2015. Microbial adaptation to ionic liquids. RSC Adv. 5, 17379–17382. Bae, J.S., Freeman, H.S., 2007. Aquatic toxicity evaluation of new direct dyes to the Daphnia magna. Dyes Pigment. 73, 81–85. Bidisha, C., Sreeranjani, R., Shaik, A., Chaudhari, S., Sumathi, S., 2006. Bioaccumulation and biosorption of drimarene red dye by Aspergillus foetidus. Int. J. Environ. Poll. 28, 517–533. Cao, B., Nagarajan, K., Loh, K.C., 2009. Biodegradation of aromatic compounds: current status and opportunities for biomolecular approaches. Appl. Microbiol. Biotechnol. 85, 207–228. Deive, F.J., Domínguez, A., Barrio, T., Moscoso, F., Morán, P., Longo, M.A., Sanromán, M.A., 2010. Decolorization of dye Reactive Black 5 by newly isolated thermophilic microorganisms from geothermal sites in Galicia (Spain). J. Hazard. Mater. 182, 735–742. Devesa-Rey, R., Bustos, G., Cruz, J.M., Moldes, A.B., 2012. Evaluation of nonconventional coagulants to remove turbidity from water. Water Air Soil Poll. 223, 591–598. EU Environmental Policy Handbook. A Critical Analysis of EU Environmental Legislation, (EEB), 2005. Ghosh, I., Jasmine, J., Mukherji, S., 2014. Biodegradation of pyrene by a Pseudomonas aeruginosa strain RS1 isolated from refinery sludge. Bioresour. Technol. 166, 548–558. Haritash, A.K., Kaushik, C.P., 2009. Biodegradation aspects of Polycyclic Aromatic Hydrocarbons (PAHs): a review. J. Hazard. Mater. 169, 1–15. Iglesias, O., de Dios, M.A., Rosales, E., Pazos, M., Sanromán, M.A., 2013. Optimization of decolorisation and degradation of reactive black 5 dye under electro-Fenton process using Fe alginate gel beads. Environ. Sci. Poll. Res. 20, 2172–2183. Khanna, P., Goyal, D., Khanna, S., 2011. Pyrene degradation by Bacillus pumilus isolated from crude oil contaminated soil. Polycycl. Aromat. Compd. 31, 1–15. Li, J., Shang, X., Zhao, Z., Tanguay, R.L., Dong, Q., Huang, C., 2010. Polycyclic aromatic hydrocarbons in water, sediment, soil, and plants of the Aojiang River waterway in Wenzhou, China. J. Hazard. Mater. 173, 75–81. Liu, J., Li, X., Luo, J., Duan, C., Huo, H., Quian, G., 2014. Enhanced decolourisation of methylene blue by LDH-bacteria aggregates with bioregeneration. Chem. Eng. J. 242, 187–194. Manenti, D.R., Modenes, A.N., Soares, P.A., Espinoza-Quinones, F.R., Boaventura, R.A. R., Bergamasco, R., Vilar, V.J.P., 2014. Assessment of a multistage system based on electrocoagulation, solar photo-Fenton and biological oxidation processes for real textile wastewater treatment. Chem. Eng. J. 252, 120–130. Marqués, A., Estañol, I., Alsina, J.M., Fusté, C., Simon-Pujol, D., Guinea, J., Congregado, F., 1986. Production and rheological properties of the extracellular polysaccharide synthesized by Pseudomonas sp strain EPS-5028. Appl. Environ. Microbiol. 52, 1221–1223. Moscoso, F., Deive, F.J., Longo, M.A., Sanromán, M.A., 2012a. Technoeconomic assessment of phenanthrene degradation by Pseudomonas stutzeri CECT 930 in a batch bioreactor. Bioresour. Technol. 104, 81–89. Moscoso, F., Deive, F.J., Longo, M.A., Sanromán, M.A., 2012b. Efficient PAHs biodegradation by a bacterial consortium at flask and bioreactor scale. Bioresour. Technol. 119, 270–276. Moscoso, F., Teijiz, I., Sanromán, M.A., Deive, F.J., 2012c. On the suitability of a bacterial consortium to implement a continuous PAHs biodegradation process in a stirred tank bioreactor. Ind. Eng. Chem. Res. 51, 15895–15900. Moscoso, F., Deive, F.J., Villar, P., Pena, R., Herrero, L., Longo, M.A., Sanromán, M.A., 2012d. Assessment of a process to degrade metal working fluids using Pseudomonas stutzeri CECT 930 and indigenous microbial consortia. Chemosphere 86, 420–426. Moscoso, F., Ferreira, L., Deive, F.J., Morán, P., Sanromán, M.A., 2013a. Viability of phenanthrene biodegradation by an isolated bacterial consortium: optimization and scale-up. Bioprocess Biosyst. Eng. 36, 133–141. Moscoso, F., Teijiz, I., Deive, F.J., Sanromán, M.A., 2013b. Approaching chlorpyrifos bioelimination at bench scale bioreactor. Bioprocess Biosyst. Eng. 36, 1303– 1309.

188

M.S. Álvarez et al. / Bioresource Technology 198 (2015) 181–188

Moscoso, F., Deive, F.J., Longo, M.A., Sanromán, M.A., 2015. Insights into polyaromatic hydrocarbon biodegradation by Pseudomonas stutzeri CECT 930: operation at bioreactor scale and metabolic pathways. Int. J. Environ. Sci. Technol. 12, 1243–1252. Moucha, T., Linek, V., Prokopová, E., 2003. Gas hold-up, mixing time and gas–liquid volumetric mass transfer coefficient of various multiple-impeller configurations: Rushton turbine, pitched blade and techmix impeller and their combinations. Chem. Eng. Sci. 58, 1839–1846. Peng, J.J., Zhang, Y., Su, J.Q., Qiu, Q.F., Jia, Z.J., Zhu, Y.G., 2013. Bacterial communities predominant in the degradation of C-13(4)-4,5,9,10-pyrene during composting. Bioresour. Technol. 143, 608–614. Robinson, T., McMullan, G., Marchant, R., Nigam, P., 2001. Remediation of dyes in textile effluent: a critical review on current treatment technologies with a proposed alternative. Bioresour. Technol. 77, 247–255. Sancar, B., Balci, O., 2013. Decolorization of different reactive dyes wastewater by O3 and O3/ultrasound alternatives depending on different working parameters. Text. Res. J. 83, 574–590.

Simarro, R., González, N., Bautista, L.F., Sanz, R., Molina, M.C., 2011. Optimisation of key abiotic factors of PAH (Naphthalene, Phenanthrene and Anthracene) biodegradation process by a bacterial consortium. Water Air Soil Pollut. 217, 365–374. USEPA, 2008. Polycyclcic aromatic hydrocarbons (PAHs) United States Office of Solid Waste. Environmental Protection Agency, 2046-Washington, DC. Vecino, X., Devesa-Rey, R., Cruz, J.M., Moldes, A.B., 2013. Entrapped peat in alginate beads as green adsorbent for the elimination of dye compounds from vinasse. Water Air Soil Poll. 224, 1448–1458. Yang, X.J., Lu, G.N., Wang, R., Xie, Y.Y., Guo, C.L., Yi, X.Y., Dang, Z., 2015. Competitive solubilization of 4,40 -dibromodiphenyl ether, naphthalene, and pyrene mixtures in Triton X series surfactant micelles: the effect of hydrophilic chains. Chem. Eng. J. 274, 84–93. Zaharia, C., Suteu, D., 2013. Coal fly ash as adsorptive material for treatment of a real textile effluent: operating parameters and treatment efficiency. Environ. Sci. Poll. Res. 20, 2226–2235.