Journal of Bioscience and Bioengineering

VOL. xx No. xx, 1e6, 2017

www.elsevier.com/locate/jbiosc

Decolorization of reactive azo dyes using a sequential chemical and activated

sludge treatment

Ken Meerbergen,1 Sam Crauwels,1 Kris A. Willems,1 Raf Dewil,2 Jan Van Impe,3 Lise Appels,2 and
Bart Lievens1, *

Laboratory for Process Microbial Ecology and Bioinspirational Management (PME&BIM), Department of Microbial and Molecular Systems, KU Leuven, Campus De Nayer,
Fortsesteenweg 30A, B-2860 Sint-Katelijne-Waver, Belgium,1 Process and Environmental Technology Lab (PETLab), Department of Chemical Engineering, KU Leuven, Campus De
Nayer, B-2860 Sint-Katelijne-Waver, Belgium,2 and Chemical and Biochemical Process Technology and Control (BioTeC), Department of Chemical Engineering, KU Leuven, Technology
Campus Ghent, B-9000 Ghent, Belgium3

Received 13 February 2017; accepted 7 July 2017

Available online xxx
Textile wastewater contains high concentrations of organic substances derived from diverse dyes and auxiliary
chemicals, some of which are non-biodegradable and/or toxic. Therefore, it is essential that textile wastewater is treated
and that these substances are removed before being discharged into the environment. A combination of advanced
oxidation processes (AOPs) to obtain partial dye degradation followed by a biological treatment has been suggested as a
promising method for cost-effective decolorization of wastewater. The aim of this study was to develop and evaluate a
combined method of partial Fenton’s oxidation and biological treatment using activated sludge for decolorization of azo
dyes, which represent an important group of recalcitrant, toxic textile dyes. Using Reactive Violet 5 (RV5) as a model dye,
color removal was significantly higher when the combined Fenton treatment/activated sludge method was used, as
opposed to separate application of these treatments. More specifically, pretreatment with Fenton’s reagent removed
52.9, 83.9 and 91.3 % of color from a 500 mg lL1 RV5 aqueous solution within 60 min when H2O2 concentrations of 1.0,
1.5, and 2.0 mM were used, respectively. Subsequent biological treatment was found to significantly enhance the
chemical treatment, with microbial decolorization removing 70.2 % of the remaining RV5 concentration, on average.
Molecular analysis of the microbial community within the activated sludge revealed that exposure to RV5 shifted the
community composition from diverse towards a highly-specialized community harboring taxa with azo dye degrading
activity, including Trichosporon, Aspergillus and Clostridium species.
Ó 2017, The Society for Biotechnology, Japan. All rights reserved.

[Key words: 454-Pyrosequencing; Activated sludge; Advanced oxidation process; Decolorization; Fenton’s oxidation; Reactive Violet 5]

Textile wastewater is typically intensely colored and contains wastewaters using appropriate and effective methods prior to
high concentrations of organic substances derived from diverse their discharge into the environment (5,8).
dyes and auxiliary chemicals (1,2). More than 10,000 different Many approaches have been proposed to remove dyes from
dyes are used in the textile industry, of which approximately textile wastewaters, including chemical coagulation/precipitation,
280,000 tons are discharged worldwide every year (3). About 70 % physical adsorption, electrochemical oxidation, chemical oxidation,
of these dyes are synthetic azo dyes, representing a highly diverse and biological anaerobic/aerobic degradation and/or conversion
group of dyes characterized by nitrogen to nitrogen double bonds (6,9e12). Recently, advanced oxidation processes (AOPs) have been
(so-called azo bonds; eN]Ne) (4). Many azo dyes and their proposed as a promising technique for wastewater treatment as
metabolites are recalcitrant in nature and can be highly toxic to AOPs are able to oxidise a wide range of compounds that are
both terrestrial and aquatic life (5,6). Furthermore, disposal of azo otherwise difficult to degrade (13). Among AOPs, oxidation using
dyes into surface water strongly affects not only its aesthetic Fenton’s reagent is an attractive treatment for decolorization and
qualities, but also its transparency, by which photosynthesis in degradation of dyes because it uses effective, easy to handle, non-
aquatic plants is hampered (7). Therefore, it is essential that these toxic reagents (i.e., Fe2þ and H2O2) (14,15). Many studies have
dyes and auxiliary chemicals are removed from textile been performed regarding the decolorization of dyes using Fenton’s
oxidation (15,16). Additionally, its capacity to improve organic
biodegradability of toxic or non-biodegradable wastewaters has
* Corresponding author. Tel.: þ32 15 305590; fax: þ32 15 305599. been described (17). Despite the apparent potential AOPs show for
E-mail addresses: [email protected] (K. Meerbergen), the degradation and removal of recalcitrant dyes, disadvantages of
[email protected] (S. Crauwels), [email protected] these processes include high reagent costs, input of energy and
(K.A. Willems), [email protected] (R. Dewil), [email protected]
production of iron sludge waste in the Fenton process, which re-
(J. Van Impe), [email protected] (L. Appels), [email protected]
(B. Lievens). quires management and safe disposal (18).

1389-1723/$ e see front matter Ó 2017, The Society for Biotechnology, Japan. All rights reserved.
http://dx.doi.org/10.1016/j.jbiosc.2017.07.005

Please cite this article in press as: Meerbergen, K., et al., Decolorization of reactive azo dyes using a sequential chemical and activated sludge
treatment, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.07.005
2 MEERBERGEN ET AL. J. BIOSCI. BIOENG.,

To circumvent these limitations, various methods have been Microbial decolorization Microbial decolorization was assessed using fresh
investigated, including the use of photo-Fenton techniques (19), activated sludge from a well-operating municipal wastewater treatment plant
located in Flanders, Belgium. Besides municipal wastewater, the plant also treats
electro-Fenton techniques (20), sono-Fenton techniques (21) and septic material from households, but little to no industrial wastewater. Following
Fenton-like oxidation techniques (22). Additionally, a combination aeration of 1 h, the sludge was centrifuged and added (10 g l1) into autoclaved
of AOPs to obtain partial dye degradation followed by a biological minimal medium (0.5 l), of which 10 ml was added to 90 ml RV5 containing
treatment has been shown to have potential to achieve effective demineralized water. The final test medium consisted of 5.0 g l1 glucose,
1.0 g l1 (NH4)2SO4, 1.0 g l1 KH2PO4, 0.5 g l1 MgSO4.7H2O, 0.1 g l1 yeast extract
decolorization and mineralization of azo dyes (13). Thus far, most
and 0.1 g l1 CaCl2.2H2O (23), and dye concentrations ranged between 100 and
studies have focused on the combination of a chemical pretreat- 500 mg l1. The experiment was performed in duplicate. Treatments without
ment followed by a biological treatment with a specific microbial sludge were included as a control and showed that no spontaneous color
strain or microbial assemblages such as biofilms which are able to degradation nor dye absorption occurred throughout the duration of the
complete the dye’s degradation (23e25). Surprisingly, a combina- experiment (data not shown). Incubations were carried out at 20 C on an orbital
shaker at 150 rpm for a total of 168 h. Samples were gathered periodically after 0,
tion of AOP and activated sludge processes has remained under- 24, 48, 72, 96, 120 and 168 h of incubation and were centrifuged to remove solid
explored, despite some studies suggesting that a Fenton treatment particles prior to absorbance measurement. Absorbance readings and calculations
combined with a (aerobic) biological treatment could be an inter- of decolorization capacity were performed as described above.
esting option for the treatment of recalcitrant compounds, Sequential Fenton’s oxidation and microbial decolorization In a final
including dyes (26,27). Furthermore, only little is known about the experiment, the efficacy of a combination of Fenton’s oxidation (1.0, 1.5 or 2.0 mM)
and activated sludge was evaluated for a dye amount of 500 mg l1. After 60 min of
impact of azo dyes on the microbial populations in activated sludge
Fenton’s oxidation, 90 ml of the Fenton treated RV5 water was combined with 10 ml
processes after dye exposure. The main objective of this study was of the activated sludge minimal medium solution, followed by incubation on an
to develop and evaluate a combined method of Fenton’s oxidation orbital shaker as described above. As a comparison, treatments with the chemical
and a biological treatment using activated sludge to achieve reagent or the activated sludge alone were included. Samples were gathered at the
decolorization and enhance mineralization of azo dyes using start of the experiment and 1, 3, 5, 7.5, 12.5, 20, 30, 45 and 60 min after Fenton
treatment, and after 24, 48, 72, 96, 120 and 168 h of incubation and were analyzed as
Reactive Violet 5 (RV 5) as a model dye. To this end, our first goal described above. The experiment was repeated twice.
was to evaluate and optimize Fenton’s oxidation process to achieve Microbial community characterization For a number of treatments
partial degradation of azo dyes in order to make them easily described above, molecular microbial community analyses were performed at the
biodegradable. Secondly, we evaluated the potential of activated end of the experiment (i.e., after 168 h of incubation). Furthermore, analyses were
sludge to perform dye removal, and finally the overall performance performed on the activated sludge before addition of the sludge to the dye medium.
Genomic DNA was extracted from 0.15 g precipitated, homogenized sludge using the
of a combination of both methods was evaluated. In order to
Power Soil DNA isolation kit (MoBio Laboratories Inc., Solana Beach, CA, USA) ac-
identify key microbes in the biological treatment, activated sludge cording to the manufacturer’s instructions. Subsequently, two 454 pyrosequencing
microbial communities were characterized using 454 amplicon amplicon libraries were built as described previously, including one for bacteria
pyrosequencing. (primers used: S-D-Bact-0341-b-S-17/S-D-Bact-0785-a-A-21) and one for fungi
(primers used: ITS86F/ITS4) (Tables S1 and S2) (two PCR replicates per DNA
extract) (31,32). Next, amplicon libraries were sequenced, and the sequencing
results were analyzed as described previously (32), using the UPARSE v8.0
MATERIALS AND METHODS standard pipeline (http://drive5.com/uparse/). Sequences from both PCR replicates
per sample were combined and grouped into operational taxonomic units (OTUs)
based on a 3 % sequence dissimilarity cut-off. Due to uneven sequencing depth
Reagents Experiments were performed using the reactive (mono)azo dye and correlation between number of sequence reads and number of OTUs per
Reactive Violet 5 (RV5; Fig. S1) as model dye. RV5 is widely used in textile and dyeing sample (data not shown), the number of sequences was rarefied to 700 sequences
industries and is typically found in high concentrations over other reactive dyes in
per sample for bacteria and 2200 for fungi, while excluding initial global
dyebath effluents (28). Additionally, it is, for example, also used in antifreeze (29).
singletons (i.e., OTUs represented by a single sequence in the unrarefied data)
Both RV5 and all other required chemicals were obtained from SigmaeAldrich
(31). Bacterial OTUs were assigned taxonomic identities by using the ‘classify.seqs’
(Saint Louis, MO, USA), unless mentioned otherwise.
command in Mothur (v. 1.36.1) (https://www.mothur.org/) and the Silva taxonomy
Fenton’s oxidation Fenton’s oxidation was performed in 250 ml Erlen- database v. 119 (https://www.arb-silva.de/), which we manually curated to include
meyers filled with 200 ml demineralized water containing RV5. Erlenmeyers were additional microbes previously observed in activated sludge (using the Midas

incubated at 24 C and continuously stirred (300 rpm) using a Cimarec i Poly 15 database, http://www.midasfieldguide.org/). Taxonomic assignments were
stirrer (Thermo Fisher Scientific, Waltham, MA, USA) to prevent oxygen depletion. considered reliable when 0.80 score value was found. Fungal OTUs were
Experiments were carried out at an initial pH of 5.0, which is known to decrease identified by querying a representative sequence (selected by UPARSE) against
during the experiment to a pH around 3e4 (30), as was also observed in our GenBank (http://www.ncbi.nlm.nih.gov/genbank/index.html), excluding
experiments. In a first experiment, Fenton’s reagent decolorization capacity (i.e., uncultured/environmental entries. Rarefaction curves were generated for each
the ability to remove color) was evaluated using different dye concentrations sample using the Vegan package (v. 2.4e1) for R (Fig. S2) (http://www.R-project.
ranging from 100 to 900 mg l1. For this experiment, a H2O2 concentration of org/; https://cran.r-project.org/web/packages/vegan/index.html). Nonmetric
1.5 mM and a Fe2þ concentration of 0.15 mM was used as proposed in Lucas et al. multidimensional scaling (NMDS) and Chao1 and Ace coverage calculations were
(23). In a second experiment, the decolorization capacity of different H2O2 performed using the R-package Vegan (v. 2.4e1) and the ‘summary.single’
concentrations (ranging from 0.0 to 2.5 mM, while maintaining the molar ratio command in Mothur (v. 1.36.1). Sequence data obtained in this study have been
between H2O2 and Fe2þ at 10:1 (23)) was evaluated for a dye concentration of deposited in the Sequence Read Archive under BioProject accession PRJNA355983.
500 mg l1. For all experiments, Fe2þ, H2O2 and dye solutions were freshly
prepared from FeSO4.7H2O, H2O2 and RV5 stock solutions, respectively. The
required amounts of Fe2þ and H2O2 were added simultaneously to the dye
solution. Dye decolorization capacity (%) was determined by measuring the RESULTS AND DISCUSSION
solution’s absorbance at the start of the experiment and after 1, 2, 3, 4, 5, 7.5, 12.5,
20, 30, 45 and 60 min, and was calculated as:
Chemical decolorization of Reactive Violet 5 As pollutant
concentration is an important parameter in textile wastewater
Initial absorbanceeFinal absorbance treatment, first Fenton’s reagent decolorization capacity was eval-
Decolorization capacity ð%Þ ¼ x100 (1)
Initial absorbance uated using different RV5 dye concentrations. The concentrations
used in this study were 100, 200, 300, 500, 700 and 900 mg l1.
Further, a H2O2 and Fe2þ concentration of 1.5 mM and 0.15 mM was
Measurements were performed immediately after sampling to avoid further used based on previous research (23). After 60 min of Fenton’s
decolorization. Absorbance readings were performed using a Multiskan GO Micro- reaction, dye removal decreased from 98.4 to 60.8 % when
plate Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) at visual
maximum peak wavelength (530 nm), and demineralized water without azo dye
increasing the dye concentration from 100 to 900 mg l1 (Fig. 1).
was used as a blank solution to calibrate the spectrophotometer. All experiments Most activity was seen during the first 45 min of the reaction,
were performed twice. after which no substantial further decolorization was observed

Please cite this article in press as: Meerbergen, K., et al., Decolorization of reactive azo dyes using a sequential chemical and activated sludge
treatment, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.07.005
VOL. xx, 2017 DECOLORIZATION OF REACTIVE AZO DYES 3

Microbial decolorization of Reactive Violet 5 using activated

sludge After 168 h of incubation, batch culture tests containing
activated sludge decolorized 14.2e43.9 % of the RV5 containing
solution, depending on the amount of RV5 added (500 to
100 mg l1) (Fig. 3). Further, our results suggest that color removal
will continue to occur after 7 days of incubation, as the
decolorization did not reach saturation (Fig. 3), especially for the
test concentration <500 mg l1. As can be observed from Fig. 3,
the activated sludge samples first entered an adaptation period of
approximately 24 h, after which an increase in decolorization rate
is seen. However, it should be noted that the test medium
contained a relatively high glucose concentration (5.0 g), which
may have influenced the degradation of RV5 during activated
sludge treatment, wherein glucose has the potential to act as an
electron donor, carbon source and/or energy source. Further
investigation is required to pinpoint what influence glucose may
have had (34).

Combination of Fenton’s reagent and activated sludge

performs better than a single chemical and single activated
sludge treatment for decolorization of Reactive Violet
FIG. 1. Effect of different Reactive Violet 5 (RV5) concentrations on the decolorization 5 Although H2O2 concentration of 2.5 mM could almost
ability of Fenton’s reagent. Experimental conditions: [RV5] ¼ 100e900 mg l1;
completely decolorize an initial RV5 concentration of 500 mg l1, a
[H2O2] ¼ 1.5 mM; [Fe2þ] ¼ 0.15 mM; temperature 24  C; initial pH 5.0. Data represent
the mean of two experiments. Standard error of mean is plotted but not visible due to major drawback of such approach is the cost involved in using
low error values. (For interpretation of the references to colour in this figure legend, higher peroxide concentrations. Therefore, to reduce the operating
the reader is referred to the web version of this article.) costs inherent to an AOP treatment, the next step of this study was
to investigate the capacity of a combination of Fenton’s reagent
with an activated sludge treatment to efficiently decolorize the RV5
(Fig. 1). Almost complete decolorization was obtained for dye
dye at a concentration of 500 mg l1 (Fig. 4). First, Fenton’s reagent
concentrations up to 300 mg l1 (Fig. 1). In contrast, Fenton’s
was used to oxidize RV5 during 60 min at different H2O2
reagent was unable to completely decolorize RV5 solutions at dye
concentrations (molar ratio H2O2/Fe2þ ¼ 10:1). Afterwards, each
concentrations of 500 mg l1 (Fig. 1).
Fenton-treated solution was inoculated with activated sludge for
As dye concentrations in textile wastewaters may often reach or
further RV5 color removal. The combined Fenton e activated
exceed 500 mg l1 (33), in a second experiment color removal at an
sludge treatment performed significantly better than the single
initial RV5 concentration of 500 mg l1 was assessed using higher
chemical (pairwise comparisons; P ¼ 6.03E-03, P ¼ 1.65E-03, and
Fenton’s reagent dosages (Fig. 2). For a H2O2 concentration of
P ¼ 1.69E-03 for a H2O2 concentration of 1.0, 1.5 and 2.0 mM,
0.5 mM, 24.1 % color was removed after 60 min of reaction.
respectively) and single activated sludge treatment (pairwise
Increasing the H2O2 concentration to 1.5 mM led to a color removal
comparisons; P ¼ 7.22E-03, P ¼ 4.73E-03, and P ¼ 4.47E-03 for a
of 84.7 %, confirming the results of the first experiment (decolor-
H2O2 concentration of 1.0, 1.5 and 2.0 mM, respectively). After
ization capacity of 87.4 % for a dye concentration of 500 mg l1;
60 min, the Fenton treatment removed 52.9, 83.9 and 91.3 % of
Fig. 1). A further increase of the peroxide dosage increased decol-
orization up to 93.7 % for a H2O2 concentration of 2.5 mM (Fig. 2).

FIG. 2. Decolorization of a Reactive Violet 5 (RV5) containing aqueous solution by

Fenton’s reagent at different H2O2 concentrations. Experimental conditions: FIG. 3. Decolorization of a Reactive Violet 5 (RV5) aqueous solution by activated sludge
[RV5] ¼ 500 mg l1; [H2O2] ¼ 0e2.5 mM and [Fe2þ] ¼ 0e0.25 mM (molar ratio be- during a period of seven days. Experimental conditions: [RV5] ¼ 100e500 mg l1;
tween H2O2 and Fe2þ was kept at 10:1); temperature 24  C; initial pH 5.0. Data temperature 20  C; incubation at 150 rpm; activated sludge end
represent the mean of two experiments. Standard error of mean is plotted but not concentration ¼ 1 g l1. Data represent the mean of two experiments. Error bars
visible due to low error values. (For interpretation of the references to colour in this represent standard error of mean. (For interpretation of the references to colour in this
figure legend, the reader is referred to the web version of this article.) figure legend, the reader is referred to the web version of this article.)

Please cite this article in press as: Meerbergen, K., et al., Decolorization of reactive azo dyes using a sequential chemical and activated sludge
treatment, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.07.005
4 MEERBERGEN ET AL. J. BIOSCI. BIOENG.,

FIG. 4. Decolorization of a Reactive Violet (RV5) aqueous solution with a combined Fenton’s reagent (reaction time of 60 min) e activated sludge (AS) process (incubation of 7 days)
(solid lines). As a reference results of the corresponding chemical and biological treatments alone are presented (dashed lines). Experimental conditions: [RV5] ¼ 500 mg l1;
[H2O2] ¼ 1.0e2.0 mM and [Fe2þ] ¼ 0.10e0.20 mM (molar ratio between H2O2 and Fe2þ was kept at 10:1); temperature 20  C; sludge incubation at 150 rpm; activated sludge end
concentration ¼ 1 g l1. Data represent the mean of two experiments. Error bars represent standard error of mean but are not visible for all points due to low error values. The inset
shows the results from the Fenton pretreatment in detail. At the end of the treatment, all treatments differed significantly (ANOVA; P < 0.05). (For interpretation of the references to
colour in this figure legend, the reader is referred to the web version of this article.)

the color when a H2O2 concentration of 1.0, 1.5, and 2.0 mM was Compared to the experiments where only activated sludge was
used, respectively (Fig. 4, inset). Whereas the curves for the 1.5 used, no apparent lag phase was detected when the chemical and
and 2.0 mM H2O2 treatment were almost identical to the ones biological method were combined (Figs. 3 and 4), suggesting that
obtained in the previous experiment, curve shapes were different the dye compounds have been partially degraded to compounds
for the 1.0 mM treatment across both experiments (Fig. 2 vs. readily usable by the sludge microbes. Previous research has shown
Fig. 4), most probably due to different experimental conditions that Fenton treatment is able to dearomatize azo dyes (removal of
(e.g., different reagent solutions were used). No residual H2O2 double NeN bond), making them more amenable for biodegrada-
was found at the end of the Fenton reaction (determined using tion (23,26,27). Indeed, for example, when applying a H2O2 con-
MachereyeNagel H2O2 Quantofix test trips and a Quantofix Relax centration of 1.0 mM, the RV5 dye concentration dropped to
semi-automated reader (Düren, NRW, Germany)) to adversely 235.3 mg l1 after 60 min. Further activated sludge decolorization
affect the biological treatment and formation of ferric-based reduced the dye concentration to 69.5 mg l1, resulting in a total
sludge was almost negligible. Through subsequent activated decolorization of 86.1 %. In contrast, when activated sludge was
sludge treatment, microbial decolorization removed 70.2 % (0.5 directly subjected to a similar dye concentration of 200 mg l1, only
(standard error of mean)) of the remaining RV5 concentration, on 33.3 % was removed, leaving a residual dye concentration of
average (across the different treatments) (Fig. 4). Compared to 133.3 mg l1 (Fig. 3). Furthermore, it is reasonable to assume that
application of the Fenton’s reagent alone, the combination of the Fenton treatment has increased the O2 concentration in the test
methods enabled 33.2, 11.1 and 6.2 % more dye decolorization at medium (35), which may also have contributed to the change in lag
the end of the experiment at a H2O2 concentration of 1.0, 1.5, and phase.
2.0 mM, respectively (Fig. 4). When compared to the biological
treatment alone (Fig. 4), 71.9, 80.8 and 83.3 % more Microbial communities of activated sludge turn into highly
decolorization was obtained using the combination of methods, specialized communities when exposed to Reactive Violet
respectively. The total amount of dye removed after the 5 As our experiments were performed with activated sludge
combined treatment was the highest for the 2.0 mM H2O2 from a wastewater treatment plant treating municipal wastewater
treatment (97.5 %), as opposed to 95.0 and 86.1 % for the 1.5 and (mostly from domestic activities) rather than textile industry
1.0 mM H2O2 treatment (Fig. 4). It can be expected that the wastewater, it is reasonable to predict that the activated sludge
differences in remaining dye concentrations at the end of the microbial community would change after exposure to toxic com-
experiment are caused by the different H2O2 concentrations that pounds like azo dyes (32). To test this hypothesis, both bacterial and
have been used during the chemical pretreatment. As can be seen fungal community compositions were assessed for a number of
from Fig. 2 and the inset of Fig. 4, different H2O2 concentrations samples, including the original activated sludge samples used in
in the pretreatment will lead to different dye concentrations at this study, two sludge samples taken after 7 days of incubation in
the end of the pretreatment that subsequently can be used in the RV5 solution (100 mg l1 and 500 mg l1), and three samples
biological treatment. Furthermore, different start concentrations taken following a combination of a Fenton (H2O2 concentration of
in the biological treatment lead to different dye concentrations at 1.0, 1.5, and 2.0 mM) and an activated sludge treatment. In total,
the end of the experiment (Fig. 3), explaining the observations 237 bacterial OTUs and 195 fungal OTUs were detected in the
seen in the combined treatment. samples studied, global singletons and aspecific fragments

Please cite this article in press as: Meerbergen, K., et al., Decolorization of reactive azo dyes using a sequential chemical and activated sludge
treatment, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.07.005
VOL. xx, 2017 DECOLORIZATION OF REACTIVE AZO DYES 5

FIG. 5. Non-metric multidimensional scaling (NMDS) ordination of the bacterial (A) and fungal (B) communities, together with the relative distribution of their operational
taxonomic units (OTUs) (C, D) (OTUs with a total relative abundance of less then 1 % are classified as others) in activated sludge samples taken before and after exposure to a
Reactive Violet (RV5) aqueous solution. OTUs were assigned taxonomic identities to the highest accurate taxonomic rank possible. Samples represent activated sludge communities
which have been exposed to RV5 containing water (500 mg l1; 7 days of incubation) that has been treated with Fenton’s reagent ([H2O2] ¼ 1.0, 1.5 or 2.0 mM and [Fe2þ] ¼ 0.1, 0.15
or 0.2 mM (molar ratio between H2O2 and Fe2þ was kept at 10:1); 60 min), communities that have been exposed to RV5 containing water (100 mg l1or 500 mg l1) which has not
been chemically treated and communities that were present in the activated sludge before the start of the experiment. Sample identifiers X_y_z contain information about their
origin: X: C, combination of chemical and microbial treatment; M, microbial treatment only; y, [RV5] (mg l1); and z, [H2O2] (mM). nMDS constructs a ’map’ on which ’similar’
samples cluster closely together and ’dissimilar’ samples plot far apart: the greater the distance between two data points, the more dissimilar the samples (here the microbial
community composition of the activated sludge samples). Stress values of the NMDS analyses were low due to the low number of samples analyzed. (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of this article.)

excluded (Tables S3 and S4). NMDS ordination of the microbial identified as azo dye reducing yeasts, most probably performed by
community composition revealed that the community reductive reactions through an NADH-dependent reductase and
composition changed after 7 days of exposure to RV5 (Fig. 5). azoreductase (36). Furthermore, the fungal genus Aspergillus, rep-
While the bacterial communities appeared scattered on the plot, resenting another abundant genus in our data set (Fig. 5 and
fungal communities diverged into two separate groups, each Table S4), contains different species known for their decolorizing
represented by a different experiment (activated sludge ability (37,38). When zooming in at the bacterial communities,
treatment alone versus combination of chemical and activated several of the most abundant bacteria, such as Acidocella, Acid-
sludge treatment) (Fig. 5). Likewise, exposure to RV5 dramatically ithiobacillus and Streptococcus have been shown to have azo dye
reduced the number of OTUs recovered from the samples reducing abilities (39,40). Further, Clostridium represented an
(Table S5). Indeed, when zooming in at the taxonomic important OTU in the colored water (Fig. 5 and Table S3). Several
composition of the communities and the relative abundance of Clostridium species have been identified as degraders of toxic
their most abundant members, it is clear that the communities compounds, such as chlorinated aliphatic compounds, ethylene
changed from diverse communities into less diverse communities and toluene compounds, as well as some herbicides. Additionally,
harboring highly adapted members (Fig. 5), which may also azo dye reducing abilities have been identified in Clostridium
explain the observed lag-phase before actually starting with dye (41,42). Altogether, it is clear from our results that the microbial
removal (Fig. 3). communities in the activated sludge samples shifted from diverse,
Further investigation of the most abundant OTUs revealed that general communities (32) towards more specialized communities,
some of them are known for their decolorizing ability. For example, which were able to withstand and degrade the tested azo dye.
Trichosporon species, which were abundantly present in the RV5 Further research could aim at the isolation and characterization of
solutions investigated in this study (Fig. 5 and Table S4), have been these key players in the community that may be exploited for

Please cite this article in press as: Meerbergen, K., et al., Decolorization of reactive azo dyes using a sequential chemical and activated sludge
treatment, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.07.005
6 MEERBERGEN ET AL. J. BIOSCI. BIOENG.,

enhanced purification of textile wastewaters, e.g., after inoculation 15. Babuponnusami, A. and Muthukumar, K.: A review on Fenton and im-
in activated sludge processes combined with a Fenton treatment, or provements to the Fenton process for wastewater treatment, J. Environ. Chem.
Eng., 2, 557e572 (2014).
in a combined treatment between a Fenton’s reagent and the 16. Nidheesh, P. V., Gandhimathi, R., and Ramesh, S. T.: Degradation of dyes from
microbe itself, as described by Lucas et al. (23). aqueous solution by Fenton processes: a review, Environ. Sci. Pollut. Res., 20,
Altogether, we have shown that RV5 color removal was signifi- 2099e2132 (2013).
cantly higher when a combination of Fenton treatment and acti- 17. Chamarro, E., Marco, A., and Esplugas, S.: Use of Fenton reagent to improve
organic chemical biodegradability, Water Res., 35, 1047e1051 (2001).
vated sludge were used, as opposed to each method applied
18. Azbar, N., Yonar, T., and Kestioglu, K.: Comparison of various advanced oxidation
separately. Pretreatment with Fenton’s reagent significantly processes and chemical treatment methods for COD and color removal from a
improved biological RV5 decolorization, and a subsequent biolog- polyester and acetate fiber dyeing effluent, Chemosphere, 55, 35e43 (2004).
ical treatment was found to significantly enhance the chemical 19. Torrades, F. and García-Montaño, J.: Using central composite experimental
treatment. However, it should be noted that these results were design to optimize the degradation of real dye wastewater by Fenton and
photo-Fenton reactions, Dyes Pigm., 100, 184e189 (2014).
obtained for simple water solutions and further research is needed 20. Panizza, M. and Cerisola, G.: Electro-Fenton degradation of synthetic dyes,
with real world textile effluents. Such effluents generally represent Water Res., 43, 339e344 (2009).
more complex solutions, due to higher organic loads and matrix 21. Cai, M., Su, J., Lian, G., Wei, X., Dong, C., Zhang, H., Jin, M., and Wei, Z.: Sono-
effects of various additives, which may impact the decolorization advanced Fenton decolorization of azo dye Orange G: analysis of synergistic
effect and mechanisms, Ultrason. Sonochem., 31, 193e200 (2016).
efficacy of the chemical and/or biological treatments applied.
22. Wang, S.: A comparative study of Fenton and Fenton-like reaction kinetics in
Furthermore, our data suggest that microbial communities that are decolourisation of wastewater, Dyes Pigm., 76, 714e720 (2008).
exposed to recalcitrant azo dyes shift from diverse towards 23. Lucas, M. S., Dias, A. A., Sampaio, A., Amaral, C., and Peres, J. A.: Degradation
specialized communities harboring taxa with azo dye degrading of a textile reactive azo dye by a combined chemical-biological process: Fen-
activity. Further research on samples from textile wastewater ton’s reagent-yeast, Water Res., 41, 1103e1109 (2007).
24. Kiran, S., Ali, S., and Asgher, M.: Degradation and mineralization of azo dye
treatments plants should confirm whether such scenario is also at reactive blue 222 by sequential photo-Fenton’s oxidation followed by aerobic
play under practical conditions. biological treatment using white rot fungi, Bull. Environ. Contam. Toxicol., 90,
Supplementary data to this article can be found online at http:// 208e215 (2013).
dx.doi.org/10.1016/j.jbiosc.2017.07.005. 25. Punzi, M., Anbalagan, A., Börner, R. A., Svensson, B. M., Jonstrup, M., and
Mattiasson, B.: Degradation of a textile azo dye using biological treatment
followed by photo-Fenton oxidation: evaluation of toxicity and microbial
ACKNOWLEDGMENTS community structure, Chem. Eng. J., 270, 290e299 (2015).
26. Tantak, N. P. and Chaudhari, S.: Degradation of azo dyes by sequential Fen-
ton’s oxidation and aerobic biological treatment, J. Hazard. Mater., 136,
This work was supported by the Industrial Research Council of 698e705 (2006).
KU Leuven (KP/10/006) and the Research Council of KU Leuven (OT/ 27. Lodha, B. and Chaudhari, S.: Optimization of Fenton-biological treatment
13/063). Further we are grateful to Michael Waud for English lan- scheme for the treatment of aqueous dye solutions, J. Hazard. Mater., 148,
guage editing and useful suggestions. 459e466 (2007).
28. Chung, Y. C. and Chen, C. Y.: Degradation of azo dye reactive violet 5 by TiO2
photocatalysis, Environ. Chem. Lett., 7, 347e352 (2009).
References 29. B. Wenderoth, L. Meszaros, U. Nitzschke and S. Dambach, Antifreeze concentrate
containing the colorant C.I. Reactive Violet 5, U.S. Patent No. 6818147 (2004).
1. Meriç, S., Kaptan, D., and Ölmez, T.: Color and COD removal from wastewater 30. Peres, J. A., Melo de Carvalho, L. H., Bonaventura, R. A., and Costa, C. A.: Char-
containing Reactive Black 5 using Fenton’s oxidation process, Chemosphere, 54, acteristics of p-hydroxybenzoic acid oxidation using Fenton’s reagent, J. Environ.
435e441 (2004). Sci. Health, Part A Toxic/Hazard Subst. Environ. Eng., 39, 2897e2913 (2004).
2. Muruganandham, M. and Swaminathan, M.: Decolourisation of reactive or- 31. Waud, M., Busschaert, P., Ruyters, S., Jacquemyn, H., and Lievens, B.: Impact
ange 4 by Fenton and photo-fenton oxidation technology, Dyes Pigm., 63, of primer choice on characterization of orchid mycorrhizal communities using
315e321 (2004). 454 pyrosequencing, Mol. Ecol. Resour., 14, 679e699 (2014).
3. Maas, R. and Chaudhari, S.: Adsorption and biological decolourization of azo 32. Meerbergen, K., Van Geel, M., Waud, M., Willems, K. A., Dewil, R., Van
dye Reactive Red 2 in semicontinuous anaerobic reactors, Process Biochem., 40, Impe, J., Appels, L., and Lievens, B.: Assessing the composition of microbial
699e705 (2005). communities in textile wastewater treatment plants in comparison with
4. Zollinger, H.: Colour chemistry: synthesis, properties and applications of municipal wastewater treatment plants, Microbiol. Open, 6, e00413 (2017).
organic dyes and pigments. VCH Publishers, Weinheim (1991). 33. O’Neill, C., Hawkes, F. R., Hawkes, D. L., Lourenco, N. D., Pinheiro, H. M., and
5. Hao, O. J., Kim, H., and Chiang, P. C.: Decolorization of wastewater, Crit. Rev. Delee, W.: Colour in textile effluentsesources, measurement, discharge consents
Environ. Sci. Technol., 30, 449e505 (2000). and simulation: a review, J. Chem. Technol. Biotechnol., 74, 1009e1018 (1999).
6. Saratale, R. G., Saratale, G. D., Chang, J. S., and Govindwar, S. P.: Bacterial 34. Modi, H. A., Rajput, G., and Ambasana, C.: Decolorization of water soluble azo
decolorization and degradation of azo dyes: a review, J. Taiwan Inst. Chem. dyes by bacterial cultures, isolated from dye house effluent, Bioresour. Tech-
Eng., 42, 138e157 (2011). nol., 101, 6580e6583 (2010).
7. Vandevivere, P. C., Bianchi, R., and Verstraete, W.: Review: treatment and 35. Neyens, E. and Baeyens, J.: A review of classic Fenton’s peroxidation as an
reuse of wastewater from the textile wet-processing industry: review of advanced oxidation technique, J. Hazard. Mater., 98, 33e50 (2003).
emerging technologies, J. Chem. Technol. Biotechnol., 72, 289e302 (1998). 36. Saratale, R. G., Saratale, G. D., Chang, J. S., and Govindwar, S. P.: Decolor-
8. Supaka, N., Juntongjin, K., Damronglerd, S., Delia, M. L., and Strehaiano, P.: ization and biodegradation of textile dye Navy blue HER by Trichosporon bei-
Decolorization of reactive azo dyes in a sequential anaerobiceaerobic system, gelii NCIM-3326, J. Hazard. Mater., 166, 1421e1428 (2009).
Chem. Eng. J., 99, 169e176 (2004). 37. Parshetti, G. K., Kalme, S. D., Gomare, S. S., and Govindwar, S. P.: Biodegra-
9. Kim, S., Park, C., Kim, T. H., Lee, J., and Kim, S. W.: COD reduction and dation of reactive blue-25 by Aspergillus ochraceus NCIM-1146, Bioresour.
decolorization of textile effluent using a combined process, J. Biosci. Bioeng., 95, Technol., 98, 3638e3642 (2007).
102e105 (2003). 38. Ali, N., Hameed, A., Ahmed, S., and Khan, A. G.: Decolorization of structurally
10. De Souza, S. M., Bonilla, K. A., and de Souza, A. A.: Removal of COD and color different textile dyes by Aspergillus niger SA1, World J. Microbiol. Biotechnol.,
from hydrolyzed textile azo dye by combined ozonation and biological treat- 24, 1067e1072 (2008).
ment, J. Hazard. Mater., 179, 35e42 (2010). 39. Stolz, A.: Basic and applied aspects in the microbial degradation of azo dyes,
11. Han, J. L., Ng, I. S., Wang, Y., Zheng, X., Chen, W. M., Hsueh, C. C., Liu, S., and Appl. Microbiol. Technol., 56, 69e80 (2001).
Chen, B. Y.: Exploring new strains of dye-decolorizing bacteria, J. Biosci. Bio- 40. Ito, T., Adachi, Y., Yamanashi, Y., and Shimada, Y.: Longeterm natural
eng., 113, 508e514 (2012). remediation process in textile dyeepolluted river sediment driven by bacterial
12. Solís, M., Solís, A., Pérez, H. I., Manjarrez, N., and Flores, M.: Microbial community changes, Water Res., 100, 458e465 (2016).
decolouration of azo dyes: a review, Process Biochem., 47, 1723e1748 (2012). 41. Rafii, F., Franklin, W., and Cerniglia, C. E.: Azoreductase activity of anaerobic
13. Oller, I., Malato, S., and Sánchez-Pérez, J.: Combination of advanced oxidation bacteria isolated from human intestinal microflora, Appl. Environ. Microbiol.,
processes and biological treatments for wastewater decontaminationda re- 56, 2146e2151 (1990).
view, Sci. Total Environ., 409, 4141e4166 (2011). 42. Joe, M. H., Lim, S. Y., Kim, D. H., and Lee, I. S.: Decolorization of reactive dyes
14. Kuo, W. G.: Decolorizing dye wastewater with Fenton’s reagent, Water Res., 26, by Clostridium bifermentans SL186 isolated from contaminated soil, World J.
881e886 (1992). Microbiol. Biotechnol., 24, 2221e2226 (2008).

Please cite this article in press as: Meerbergen, K., et al., Decolorization of reactive azo dyes using a sequential chemical and activated sludge
treatment, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.07.005