Science of the Total Environment 573 (2016) 585–593

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

Sequential anaerobic-aerobic decolourization of a real textile wastewater

in a two-phase partitioning bioreactor
M. Concetta Tomei a,⁎, Domenica Mosca Angelucci a, Andrew J. Daugulis b
a
Water Research Institute, C.N.R., Via Salaria km 29.300, CP 10, 00015, Monterotondo Stazione, Rome, Italy
b
Department of Chemical Engineering, Queen's University, Kingston, Ontario K7L 3N6, Canada

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• A two-phase partitioning bioreactor is

applied to treat real textile wastewater.
• DuPont polymer Hytrel 8206 is effective
in colour absorption reaching removal
of 84%.
• Low pHs favour the sorption process
with equilibrium times ≤24 h.
• The polymer improves biological colour
removal in the two-phase system.

a r t i c l e i n f o a b s t r a c t

Article history: This work describes the application of a solid-liquid two-phase partitioning bioreactor (TPPB) for the removal of
Received 16 June 2016 colour from a real textile wastewater containing reactive azo-dyes. Four polymers were tested over the pH range
Received in revised form 2 August 2016 of 4–9 to select the most effective absorbant to be used as the partitioning phase in the TPPB. The best results
Accepted 19 August 2016
were obtained with Hytrel 8206 at pH 4 achieving ~70% colour removal, based on the dominant wavelength,
Available online 28 August 2016
in the first 5 h of contact time, and 84% after 24 h. Wastewater treatment was undertaken in a solid-liquid
Editor name: D. Barcelo TPPB operated with Hytrel 8206 in sequential anaerobic-aerobic configuration. The reaction time of 23 h was
equally distributed between the anaerobic and aerobic phases and, to favour colour uptake, the pH was con-
Keywords: trolled at 4.5 in the first 4 h of the anaerobic phase, and then increased to 7.5. Colour removal (for the dominant
Textile effluents wavelength, 536 nm) increased from 70 to 85% by modifying the bioreactor operation from single-phase to TPPB
Reactive dyes mode. Based on COD measurements nearly complete biodegradation of the intermediates produced in the anaer-
Sequential anaerobic-aerobic decolourization obic phase was obtained, both in the single-phase and two-phase mode, with better performance of the TPPB sys-
Solid-liquid TPPB bioreactor tem reaching 75% CODDye removal.
Polymer Hytrel
© 2016 Elsevier B.V. All rights reserved.

1. Introduction generate among the largest volumes and pollution levels of discharged
water (Sen and Demirer, 2003). Because dyes are designed to provide
The textile industry is one of the largest producers of aqueous indus- a high degree of chemical and photolytic stability to resist breakdown
trial emissions since dyeing, and associated finishing operations, over time, microbial attack, and the action of water and soap, they are
recalcitrant and resistant to biodegradation (Solis et al., 2012). Azo
⁎ Corresponding author. dyes are the most common synthetic dyes used in industry, and thus
E-mail address: [email protected] (M.C. Tomei). the most commonly released into the environment and are used in

http://dx.doi.org/10.1016/j.scitotenv.2016.08.140
0048-9697/© 2016 Elsevier B.V. All rights reserved.
586 M.C. Tomei et al. / Science of the Total Environment 573 (2016) 585–593

the textile industry, paper printing (Chang et al., 2004) and in plastics, single-phase mode and in a TPPB system for the treatment of an indus-
leather, cosmetics and food industries (Telke et al., 2008). Release of un- trial effluent containing Remazol Black 5, Remazol Yellow RR and
treated or poorly treated effluents containing azo dyes is harmful to the Remazol Brilliant Red 21. In addition, COD levels during the treatment
aquatic environment, causing inhibition of photosynthesis and growth steps were followed to have a cumulative estimate of the fate of the in-
of aquatic biota, reduction of dissolved oxygen, potential toxicity to termediates during the biodegradation process. Finally, according to
humans, flora and fauna (Saratale et al., 2011), and also for aesthetic Muda et al. (2011) and Ricco et al. (2004), on line measurement of
reasons as colour in wastewater can be detected even at very low levels the specific oxygen uptake rate (SOUR) was used to assess the toxicity
(b1 mg/L) (Pereira and Alves, 2012). of dyes/by-products on the aerobic biomass in the different contacting
Conventional wastewater treatment, such as activated sludge, can configurations.
be ineffective for decontaminating dye effluents because of the physi-
cal-chemical stability and poor biodegradability of these pollutants, as 2. Materials and methods
noted above (Forgacs et al., 2004), but, it is worth noting that relative
to chemical and/or physical processes, biological methods have the ad- 2.1. Textile wastewater
vantages of being cost-competitive and more environmentally friendly.
Moreover, if operated with effective technologies, are often able to Textile wastewater was obtained from the dyeing bath of a factory
achieve complete mineralization of the pollutants at ambient condi- located in the textile district of the Como area, in the North of Italy.
tions. Several research and review articles have focused on the biologi- The contaminant load is mainly comprised of a mixture of mono and
cal decolourization of textile wastewater, especially on synthetic azo di-azo reactive dyes (commercially named Remazol Black 5, Remazol
dye solutions summarizing the use of different microorganisms, reac- Yellow RR and Remazol Brilliant Red 21), accounting for ~ 90% of the
tion environments and technologies (Sponza and Isik, 2002; Pandey et pollution load, and chemical additives used for industrial processing.
al., 2007; Saratale et al., 2011; Solis et al., 2012). Wastewater characterization is reported in Table 1.
The biodegradation of reactive azo dyes is difficult since their com-
plex structure and synthetic nature often require a variety of bacteria, 2.2. Polymers
both aerobic and anaerobic, able to achieve the reductive cleavage of
the \\N_N\\ bond, resulting in the formation of generally colourless, A first selection of the polymers has been performed on the basis of
but potentially hazardous, aromatic amine by-products. These aromatic their previous applications in TPPBs and on their predicted affinity for
intermediates could in principle be degraded aerobically or anaerobical- the dyes under investigation (Bacon et al., 2014).
ly (Joshi et al., 2004), however these metabolites can be more toxic than Four commercial polymers were tested in this study, and their
the parent dyes (Solis et al., 2012), and their biodegradation is generally source, properties and composition are shown in Table 2. Before use,
more effective under aerobic conditions (van der Zee and Villaverde, the polymers were pre-treated in order to remove any impurities aris-
2005). This has led to the use of sequential anaerobic-aerobic processes ing from the manufacturing process by adding a known mass of poly-
aimed at more effective removal of azo dyes from wastewater, recently mer (~100 g) to an equal volume of methanol (Fluka, Italy) in a flask
reviewed by Popli and Patel (2015), who have considered biological an- and mixing it vigorously for at least 20 min. The polymers were then re-
aerobic-aerobic decolourization of azo dyes for various types of reactors, peatedly washed with distilled water until a clean (i.e. non-turbid)
under different operating conditions. They concluded that the anaerobic wash water was obtained. Each step lasted 20 min and, generally, five
stage of the sequential process can be negatively affected by the high washing steps were sufficient, at the end of which the polymers were
initial concentration and complex structure of dyes, and confirmed the dried by air exposure for several days.
possibility of achieving mineralization of the intermediates under aero-
bic conditions. 2.3. Sorption and desorption tests
Here, we propose the application of two-phase partitioning bioreac-
tors (TPPBs), which have been demonstrated to be advantageous for xe- Sorption tests of the wastewater colour by polymers were per-
nobiotic removal (Tomei et al., 2011b). Employment of a TPPB in this formed in batch mode in flasks containing a fixed volume (in the
case can be effective in reducing the exposure of biomass to toxic sub- range of 100–200 mL) of wastewater diluted 1:10 with tap water. Mag-
strate concentrations, with the additional feature, in contrast to the netic stirrers provided the mixing, and the temperature was controlled
use of activated carbon proposed by Kuai et al. (1998) for the same pur- at 25 ± 0.5 °C. The duration of the tests varied in the range of 24–72 h,
pose, of providing gradual substrate release and subsequent biodegra- i.e. a time estimated to be sufficient to reach equilibrium. Liquid samples
dation in the aqueous phase. The uptake of toxic substrates by were taken from the flasks at time intervals of 1 h, during the first 8 h,
amorphous polymers in TPPB systems occurs via absorption rather then after 24 h and then at subsequent intervals of 24 h. The pH was reg-
than adsorption: i.e. when biodegradation occurs in the cell-containing ularly monitored and adjusted by acid/base addition in order to main-
aqueous phase the thermodynamic equilibrium of the system is re- tain the desired pH values, which were pH 4 and 7.5 at a fixed
stored by the release of the absorbed compound driven by the cellular polymer/solution ratio of 5% (v/v).
metabolic demand. This feature maintains sub-inhibitory concentration
levels in the liquid phase to allow the biomass to achieve high biodegra-
dation rates of the compound, which is consumed, and not simply Table 1
adsorbed. The mechanism of uptake-release also provides for polymer Textile wastewater characteristics: average values and standard deviation (SD).

bioregeneration, and the consequent restoring of its sorption capacity, Parameter Unit Average value SD
which is not exhausted. As a final distinction between absorptive amor- pH – 9 0.50
phous polymers and adsorptive activated carbon, rational selection of COD mg/L 1117 58.00
such polymers allows the use of materials, which are specifically TSS g/L 0.54 0.10
targeted to the substrate in question, unlike activated carbon which is VSS g/L 0.23 0.05
TC mg/L 471.1 35.2
generally non-selective to organic molecules.
TOC mg/L 158.0 9.80
To date, most studies and applications of this powerful technology N-NH4 mg/L 40.0 4.50
were related to synthetic solutions, and in this study we have investi- Chlorides g/L 38.6 3.10
gated for the first time the application of TPPB systems to a real waste- Nitrate g/L 3.8 0.35
water. The main objective of this study was to evaluate colour removal Phosphates g/L 3.2 0.33
Sulphates g/L 4.5 0.38
for a sequential anaerobic-aerobic process operated in a conventional
 M.C. Tomei et al. / Science of the Total Environment 573 (2016) 585–593 587

Table 2
Suppliers, properties and compositions of commercial polymers used in this study.

Commercial name Grade Supplier Density (g/cm3) Size Tga Composition

(mm) (°C)

Hytrel 8206 DuPont 1.170 1.5-5b -59 poly(ether-block-butylene terephthalate)

Tone P787 Dow 1.145 4c -69 poly(caprolactone)
Elvax 770 DuPont 0.928 3.5–5.5b -100 poly(ethylene-co-vinyl acetate)
Pebax 2533 Arkema 1.010 3-5b -65 poly(tetramethylene ether-block-amide 12)
a
Glass Transition Temperature.
b
Minimum – maximum dimension.
c
Diameter.

Hytrel showed the best results in the sorption tests so it was selected cycle duration a minimal biomass increase was expected, as reported
as the partitioning phase for the TPPB and its uptake-release capacity in Tomei et al. (2016), and the biomass wastage associated with the ef-
was more thoroughly investigated through sorption and desorption fluent discharge balanced the biomass growth. The reactor exchange
tests under different operating conditions according to the test plan ratio (added volume/total volume) was controlled at 0.3 in all tests in
shown in Table 3. order to achieve data suitable for comparison. The feed contained the
In the desorption tests the colour-loaded polymer, resulting from real textile wastewater and sodium acetate with a dye-acetate ratio of
the sorption tests, was put in contact with fresh tap water under 1:20, in terms of COD, diluted with tap water to reach the desired
mixed conditions. The same experimental apparatus, sampling and an- load, and supplemented with a mineral medium (Williams and Unz,
alytical procedures of the sorption tests were utilized. Desorption tests 1989) to ensure an adequate supply of nutrients for microbial growth.
were performed at pH = 7.5, which is the typical value of the biological The sequential anaerobic-aerobic process was applied in three series
processes. of tests and a microbial consortium, previously acclimated to the same
Parallel control tests for sorption and desorption were performed textile wastewater, was employed as inoculum (Tomei et al., 2016). In
under the same operating conditions with tap water and polymers, the first series (I) the bioreactor was operated in conventional (single-
but without wastewater, to take into account the possible presence of phase) mode, with the first part of the experiment dedicated to micro-
released substances from the polymers, which could affect the measure- bial acclimatization achieved by progressively increasing the dye load-
ments. The absorbance readings of the control samples were subtracted ing (in the range of 0.005–0075 kgCOD/(m3 d), followed by reactor
from the corresponding test sample readings. operation for 45 days at a fixed feed of 0.01 kgCOD/(m3 d). The same
Sorption tests have been also executed to verify that Hytrel does not load was applied in the second series (II), while in the third one (III)
absorb acetate. Initial concentration was 100 mg/L and applied modali- the dye load was increased to 0.013 kgCOD/(m3 d).
ties are the same above reported. The tests were conducted for 24 h in The three series of tests were performed at different pH conditions.
two operating conditions: at pH 7.5 and with the pH pattern applied The first one (I) was conducted at pH 7.5, while in the second (II) and
in the bioreactor, i.e. pH 4.5 in the first 4 h and pH 7.5 for the subsequent the third (III) ones, the pH was controlled at 4.5 during the first 4 h of
20 h. the anaerobic period, then increased to 7.5 to restore suitable conditions
for microbial growth during the residual anaerobic phase (440 min) and
2.4. Bioreactors: type and operation modes the following aerobic period.
For the TPPB operation mode the polymer Hytrel 8206 was added to
Experiments were conducted in lab scale sequencing batch reactors the bioreactor at a polymer/aqueous phase ratio of 10% (v/v) at the be-
(SBRs) (volume 0.5 and 1.2 L) operating at 27 ± 1 °C, mixed with a mag- ginning of experimental phase III and kept in the system until the end of
netic stirrer, and equipped with peristaltic pumps (Cellai) for feeding the experiment. This value has been chosen on the basis of previous
and discharge of the effluent. Additional information about timing and studies on TPPBs demonstrating their effective performance with poly-
control devices and management are given elsewhere (Tomei et al., mer/water ratio ≤ 10% (Amsden et al., 2003; Tomei et al., 2009; Tomei et
2004). al., 2011b).
Each SBR working cycle, consisting of fill (20 min), reaction Treatment performance was evaluated by daily analysis of the colour
(1360 min), settle (40 min) and draw (20 min) periods, lasted one content and COD in the influent and the effluent, and kinetic tests were
day, with the reaction time being equally distributed between the an- carried out in the three different operating modes described above.
aerobic and aerobic periods during each 24 h cycle. The time distribu- Table 4 shows an overview of the operating conditions for the three se-
tion of the SBR was established on the basis of previous studies on ries of tests.
xenobiotic removal in SBRs (Tomei and Daugulis, 2013; Mosca
Angelucci and Tomei, 2015). Furthermore, no wastage of biomass was 2.5. Kinetic tests
utilized during the entire experiment because due to the high reaction
Colour removal in the three experimental series of tests was charac-
Table 3 terized by kinetic tests carried out in the bioreactors under the operat-
Experimental plan and operating conditions for sorption and desorption tests with Hytrel ing conditions reported in Table 4. Kinetic tests allow to follow the
8206. colour pattern during the subsequent biological phases. Tests were per-
Test Polymer/water ratio Sorption Desorption
formed during the reaction phase of the SBR working cycles, by taking
% (v/v) samples from the bioreactors at time intervals of 1–2 h. Samples were
pH time (h) pH time (h)
centrifuged (10–20 min, 13,000 rpm), and spectrophotometric analysis
H1 2.5 7.5 72 7.5 96 was performed on the supernatant. COD and VSS concentrations were
H2 5 7.5 72 7.5 96
also monitored at the start (after feeding the bioreactor) and at the
H3 5 4 72 7.5 96
H4 10 7.5 72 7.5 96 end of the test. Additionally, during the aerobic period, the dissolved ox-
H5 10 4 24 7.5 96 ygen (DO) concentration was monitored and controlled (in the range of
H6 10 5 24 7.5 96 3–4 mg/L) through an on–off aeration strategy, and DO data were
H7 10 6 24 7.5 96 employed to calculate the specific oxygen uptake rate of the biomass
H8 10 9 72 7.5 96
(SOUR) according to the procedure reported in Tomei et al. (2004).
588 M.C. Tomei et al. / Science of the Total Environment 573 (2016) 585–593

Table 4
Operating conditions in the three series of tests.

Test Series Operation mode Working volume Acclimation Stable operation

(L)
Duration Dye load kgCOD/(m3 d) Duration Dye load kgCOD/(m3 d)
(d) (d)

I Conventional 1 22 0.005–0.0075 45 0.01

II Conventional pH modifieda 0.3 7 0.0075 21 0.01
III TPPBb 0.3 7 0.01 21 0.01–0.013
pH modifieda
a
pH controlled at 4.5 in the first 4 h of the anaerobic phase.
b
Polymer to water ratio 10% (v/v).

In order to verify data reproducibility, all kinetic tests were conduct- spectrophotometer (PerkinElmer, Lambda 25), in the 200–900 nm
ed in two replicates under the same operating conditions. spectral band. From a spectrum analysis three wavelengths, 396 nm,
492 nm and 603 nm, were identified as being representative of the dif-
2.6. Analytical methods ferent intervals of colour distribution in the sample on the basis of the
positions of the peaks. The dominant wavelength was 536 nm and can
COD Cell Tests (MERCK-referring to EPA 410.4 method) based on po- be considered to be the most representative of the entire spectrum.
tassium dichromate oxidation and spectrophotometric determination Decolourization performance was evaluated as percent colour removal
(Spectroquant Nova30) were employed to measure the COD concentra- from the absorbance readings at the three representative wavelengths
tion. Samples were appropriately diluted to eliminate the effects of and at the dominant wavelength. The absorbance of the liquid in the
chloride interferences on the measurements. bioreactor before feeding was accounted for the measurement and the
TOC, employed for wastewater characterization, was analyzed with evaluation of colour removal.
a total organic carbon analyser (TOC-V CSN SHIMADZU), which deter-
mines the TOC concentration by the difference between the amount of 3. Results
total and inorganic carbon in the sample.
Ion concentrations were measured with an ionic chromatograph 3.1. Polymer screening: sorption and desorption tests
(DIONEX, DX 100) equipped with a Dionex AS-14 column.
Biomass concentration in the reactor was determined as VSS The first step to evaluate the potential applicability of a two-phase
concentration. bioreactor is polymer screening to determine the best absorptive mate-
VSS concentrations and pH were determined according to standard rial in terms of uptake-release. Previous studies have generally been
procedures (APHA, 1998). performed on single target dyes, however in this case a real wastewater
Acetate was analyzed by using a PerkinElmer Auto System gas chro- comprised of a mixture of dyes was investigated, making it more diffi-
matograph equipped with 2 m × 2 mm stainless steel column packed cult to find a polymer suitable for all the components in the mixture.
with 60/80 mesh Carbopak B-DA 80–120 4% CW 20 M Supelco. Other Four polymers were selected for testing based on their previous use
operating conditions were: N2 carrier gas 20 mL/min, oven temperature for solute uptake, and an overview of the screening results is reported in
120 °C, injector temperature 200 °C, flame ionization detector (FID) at Fig. 1, expressed as percent colour removal at the different wavelengths
temperature 200 °C. for 24 h of contact time. A noticeable pH effect is evident with the best
Colour content was determined according to Standard Methods performance given by Hytrel 8206 at pH 4 for all wavelengths, reaching
(APHA, 1998) from the UV–visible absorbance scans recorded with a colour removal in the range of 83–87%.

Fig. 1. Sorption tests: % colour removal at different pH values for the different wavelengths (nm).
 M.C. Tomei et al. / Science of the Total Environment 573 (2016) 585–593 589

a) 50 b) 100

40 80

Removal (%)

Removal (%)
30 60
396
20 396 40 492
492
536
10 536 20
603
603
0 0
0 5 10 15 2 4 6 8 10 12
% pol/wat ratio (v/v) pH

c) 1 d) 0.3
H4
0.25 H5
0.8 H6
0.2 H7
0.6
A/A0

H8
0.15

A
0.4 pH4
pH5 0.1
0.2 pH6
pH7.5
0.05
0 pH9 0
0 20 40 60 80 0 20 40 60 80 100
time (h) time (h)

Fig. 2. Overview of the sorption tests results, polymer Hytrel. a) Sorption colour removal vs. polymer/water ratio for the different wavelengths (nm); contact time 72 h, pH = 7.5 (Test H1,
H2, H4); b) Sorption colour removal vs. pH for the different wavelengths (nm); polymer/water ratio 10%, contact time 24 h (Test H5, H6, H7) and 72 h (Test H4, H8); c) Sorption kinetics
(H4–8) at different pH: absorbance profile vs. time for the dominant wavelength i.e. 536 nm, polymer/water ratio 10%; A = absorbance at time t; A0 = absorbance at time 0; d) Desorption
kinetics (H4–8): absorbance profile vs. time for the dominant wavelength i.e. 536 nm, polymer/water ratio 10%; pH = 7.5.

On the basis of the obtained results, Hytrel was chosen as the components during the treatment process. Finally, Fig. 3c shows the
partitioning phase in the TPPB and investigated more thoroughly for percent colour removal for the sequential process at the same selected
sorption and desorption. wavelengths, and the total colour removal (TCR) calculated as the dif-
Fig. 2a displays the % colour removal vs. the polymer/water ratio for ference between the area of the two surfaces bounded by the spectrum
Hytrel and demonstrates enhanced colour uptake with increasing poly- curves at the start and at the end of the anaerobic and aerobic phases.
mer fraction utilized. The test, conducted at pH = 7.5, reached colour The colour appears to be mainly removed during the anaerobic phase
uptake in the range of 30–37% even at the highest polymer fraction (~50%) with additional removal in the aerobic phase of ~18%. The per-
(10%) and 72 h of contact time, so the test was repeated with a fixed cent removal increases with the wavelength, as also reported in Table
polymer/water ratio of 10% and pH in the range of 4–9 with the results 5. The first series is the reference condition for comparison of the subse-
reported in Fig. 2b. Increased colour removal can be seen for decreasing quent operation modifications applied in the 2nd and 3rd series of tests.
pH with the maximum value occurring at pH 4. Lower pH values were Since colour uptake by Hytrel is greater at low pH values (see Fig.
not explored due to their likely negative effect on biomass activity. No 2b), which can be detrimental to biomass activity, the second series of
appreciable improvement was observed for pH ≥ 7.5. tests examined the possibility of operating at low pHs for a reduced por-
The sorption and desorption kinetic results are reported in Figs. 2c tion of the biological reaction time. The results of a typical kinetic test
and d respectively and provide an estimation of the time required for performed during a work cycle of series II are shown in Fig. 4. No signif-
the physical processes of absorption/desorption to occur, which is one icant differences are observed in comparison with the first series of tests
of the important parameters to be considered in practice. The absor- (see Fig. 3): removal efficiencies at λ = 536 nm are 52% and 19% for the
bance profile displayed in Fig. 2c shows that at lower pH values (4–5), anaerobic and aerobic phases, respectively. For the other wavelengths
~6 h of contact time are sufficient to reach maximum dye removal. De- removal efficiencies are comparable and even higher with a slight de-
sorption data reported in Fig. 2d do not show significant differences in crease only for the lowest wavelength, λ = 396 nm, which contributes
the desorbed amounts reached after contact times ≤24 h, which is the minimally to the colour content. The positive results of this second se-
duration for the work cycle of the bioreactor. ries confirmed the possibility of operating at lower pH in the TPPB.
In the third series of tests the bioreactor was operated in TPPB mode
3.2. Biotic tests: colour removal at increasing dye loading rates, and the results of the kinetic tests are

Table 5 shows a summary of the removal efficiencies achieved in the Table 5

biotic tests. Colour removal is expressed as average values ± (SD) for Removal efficiencies in the different experimental series: for each series, the first line gives
the different wavelengths in the three series (i.e. calculated by taking the average percent removal (±SD) referred to the sequential anaerobic-aerobic process
into account all work cycles performed in the series) in order to provide while in the second line the removal in the anaerobic phase is reported. Bold characters
refer to the dominant wavelenght.
an exhaustive representation of the removal efficiencies achieved dur-
ing the different experimental campaigns. For each series, the total per- Series Dye removal (%)
cent removal (i.e. for the complete sequential anaerobic-aerobic λ = 396 nm λ = 492 nm λ = 536 nm λ = 603 nm
process) and the removal only in the anaerobic phase are reported.
I - Total 53.7 ± 11.4 61.3 ± 10.4 68.2 ± 8.1 90.5 ± 7.4
The results of a typical kinetic test performed during a work cycle of Anaerobic 38.2 ± 6.8 47.5 ± 9.8 49.9 ± 3.9 66.0 ± 11.1
the first series are shown in Fig. 3. Fig. 3a shows the absorbance profile II - Total 48.0 ± 14.4 63.8 ± 14.0 67.1 ± 12.9 83.7 ± 12.7
at the dominant wavelength, which is the most representative of colour Anaerobic 35.2 ± 5.6 54.5 ± 10.4 55.5 ± 4.5 69.4 ± 4.9
removal during the test, while Fig. 3b shows the absorbance spectra vs. III - Total 64.4 ± 16.9 79.9 ± 11.7 84.5 ± 8.9 92.0 ± 4.9
Anaerobic 42.4 ± 6.1 79.4 ± 7.6 82.0 ± 5.9 84.3 ± 4.4
time (in the range 300–800 nm), which gives an overview of the colour
590 M.C. Tomei et al. / Science of the Total Environment 573 (2016) 585–593

a) 0.1
b) 0.2
0
c) 100

0.18 90
2
0.08
0.16 80
4
0.14 8 70
Absorbance

Removal (%)
Absorbance
0.06
0.12 14 TCR 60

0.1 18 50

0.04 24 40
0.08

0.06 30

0.02 0.04 20

0.02 10

0 0 0
0 5 10 15 20 25 300 400 500 600 700 800
time(h) wavelenght(nm) (nm)

Fig. 3. Results of the kinetic tests: series I: a)Absorbance pattern for the dominant wavelength; b)Evolution of the colour spectrum vs. time; c)Colour removal at the different wavelengths
(evaluated on the test replicates).

shown in Figs. 5 and 6 for work cycles at the two loading conditions. In that its biodegradation takes place in a short time in the very first part of
Fig. 5a, corresponding to the same operating conditions of the two pre- the anaerobic phase, and the residual COD at the end of the anaerobic
vious series, an improvement in colour removal for the dominant wave- and aerobic phase is attributable to the colour component (CODDye).
length is observed with a value of 84%. There is also a different evolution Total COD removal efficiencies of ≥96% and ≥ 98% after the anaerobic
of the colour spectrum (Fig. 5b), in comparison to the previous tests of and aerobic phase respectively, have been achieved in the three series,
series I and II. In the conventional biological systems, even with the while a different trend is observed for CODDye: a low anaerobic removal
modified pH, a progressive and regular lowering of the spectrum vs. for CODDye (11–14%) is observed in Series I and II, which increases up to
time is observed, while in the TPPB, a distinct decrease of the spectrum 41% in Series III. At end of the aerobic phase CODDye percent removal of
is observed after 4 h followed by a slight increase, more obvious for the ~59 and 75% are achieved in Series I - II, and III, respectively.
higher loading. It is also observed that good performance of the TPPB A first quantification of the toxicity effect on the biomass may be
system is maintained for a colour load increase of 30%, with colour re- evaluated by SOUR data continuously recorded during the kinetic
moval at λ = 536 nm N 80% (see Fig. 6 and Table 5). tests. In Fig. 7 the SOUR profiles of kinetic tests of Series I and III show
a distinct difference between the maximum detected values, i.e. 8 and
3.3. COD and SOUR pattern 28 mgO2/(gVSS L), respectively. The higher specific oxygen consumption
rates reached in Series III successfully demonstrated the increased activ-
COD is a parameter commonly used in wastewater treatment plants ity of the biomass operating in TPPB mode, as already pointed out with
to evaluate the process efficiency and the effluent quality when specific xenobiotic compounds such as 2,4-dichlorophenol (Tomei et al., 2014).
characterization of a wastewater component is not required, or is diffi-
cult. COD concentrations in the feed and at the end of the anaerobic and 4. Discussion
aerobic phases and the related colour removal for the kinetic tests per-
formed at 0.01 kgCOD/(m3 d) load in the three series, are reported in The results of the abiotic sorption tests showing the best perfor-
Table 6. The two COD components, i.e. biogenic (acetate) and dye mix- mance for Hytrel, can be explained considering that the dye molecules
ture, have been differentiated in the mass balance. In the last two col- are very polar and have a strong affinity for water. As such, polar poly-
umns, the TCR (evaluated from the absorbance spectra) is reported for mers, such as Hytrel can potentially work well. As for the effect of pH,
comparison. Concerning the acetate fraction, the control tests excluded the positive effect of low pH could be justified considering that a
its absorption on the polymer in all the investigated conditions (varia- lower pH provides conditions that would protonate the sulphate groups
tion coefficient ≤ 3% for both neutral and acid-neutral conditions). Ace- and allow for enhanced absorption into the polymer. Sorption charac-
tate is a readily biodegradable substrate both under anaerobic and teristic times (~6 h) observed for Hytrel at lower pH values (4–5) are
aerobic conditions, thus, if it not absorbed, it can be reasonably assumed similar in magnitude to the kinetics of many biological processes, thus

a) 0.1 b) 0.2 c) 100

0.18 90
0
0.08 0.16 80
4
0.14 70
Removal (%)

8
Absorbance
Absorbance

TCR
0.06 0.12 60
14
0.1 50
18
0.04 0.08 40
24
0.06 30

0.02 0.04 20

0.02 10

0 0 0
0 5 10 15 20 25 300 400 500 600 700 800
time (h) wavelenght (nm) (nm)

Fig. 4. Results of the kinetic tests: series II: a)Absorbance pattern for the dominant wavelength; b)Evolution of the colour spectrum vs. time; c)Colour removal at the different wavelengths
(evaluated on the test replicates).
 M.C. Tomei et al. / Science of the Total Environment 573 (2016) 585–593 591

a) 0.1 b) 0.2 c) 100

0.18 0 90

0.08 0.16 4 80

0.14 8 70

Removal (%)
Absorbance
Absorbance

0.06 0.12 14 TCR 60

0.1 18 50

0.04 0.08 24 40

0.06 30

0.02 0.04 20

0.02 10

0 0 0
0 5 10 15 20 25 300 400 500 600 700 800
time (h) wavelenght (nm) (nm)

Fig. 5. Results of the kinetic tests: series III; Dye load = 0.01 kgCOD/(m3 d): a) Absorbance pattern for the dominant wavelength; b) Evolution of the colour spectrum vs. time; c) Colour
removal at the different wavelengths (evaluated on the test replicates).

suitable for TPPB application. Higher pH requires longer contact times, The good performance of the biological system in Series II, in spite of
which could be limiting when coupled with a biological process, and the low pH applied in the first part of the anaerobic period, can be ex-
this is why low pH values have been applied in TPPB operation here (Se- plained considering that the bacterial strains catalysing the first steps
ries II and III). of the anaerobic reaction chain (i.e. the Acidogens and the Acetogens),
Biotic tests were performed in sequential anaerobic-aerobic mode, in contrast to the Methanogens, are not affected by lower pH (Kim et
because, as already mentioned, the two reaction environments act in a al., 2004; van Lier et al., 2008; Chen et al., 2008). In this study, the objec-
synergic mode with the aerobic stage completing the biodegradation tive was not to maximize methane production but to have colour re-
process of compounds and intermediates not biodegraded in the anaer- moval so, in principle, it is not detrimental to operate a portion of the
obic period. This approach has been successfully applied for treating anaerobic reaction time at low pH, thus providing conditions for en-
synthetic textile wastewater (Kapdan et al., 2003; Supaka et al., 2004; hancing colour uptake by the polymer. Moreover, in two-phase anaero-
Lourenco et al., 2006), and in other biological systems, i.e. anaerobic bic digestion (Acidogenic phase followed by the Methanogenic one)
followed by post-aerobic treatment of urban wastewater (Chan et al., applied pH values are in the range 4.3–6 (Demirel and Yenigün,
2009), and a sequential anaerobic-aerobic process for sludge stabiliza- 2002), and in specific applications, such as in the production of hydro-
tion (Tomei et al., 2011a). gen, the pH is controlled around 4 (Chen et al., 2008; Fang and Liu,
In Series I, i.e. the conventional single-phase system, greater 2002). According to Firmino et al. (2010), due to the competition of
decolourization with increasing wavelengths is observed. A possible ex- dye reduction with methanogenesis for the electrons generated upon
planation of this finding is given by the different biodegradability of the electron donor oxidation, the two-stage anaerobic system could be an
single dyes in the mixture. According to Shah et al. (2014) and Sanmuga interesting option to enhance colour removal. In our case, the sequential
Priya et al. (2015), Remazol Red (peak 580–590 nm) is characterized by system can be considered as a two-stage anaerobic system where the
higher biodegradability, followed by Remazol Yellow (peak 398– two phases are undertaken in the same unit on a time basis rather
480 nm), while Remazol Black (peak 595 nm) is the most difficult to than in two separate units, thus the low pH in the first part of the reac-
biodegrade. Remazol Red and Remazol Black are characterized by absor- tion is compatible with the acidogenic step. In addition, the high pH tol-
bance peaks at λ ≥ 580 nm that is in the higher wavelength region. Thus, erance of decolorizing bacteria reported in Saratale et al. (2011) can also
considering the higher biodegradability of Remazol Red (with respect to explain why pH change had no effect on the performance of the biore-
the Black) it may be plausible to assume a higher contribution of actor in series II with respect to Series I. In any case, the inhibitory effect
Remazol Red to the colour removal. on methanogenesis can also be mitigated by appropriate technologies:

a) 0.1 b) 0.2 c) 100

0.18 0
0.09 90
0.16 4
0.08 80
8
0.07 0.14 70
Removal (%)
Absorbance
Absorbance

14 TCR
0.06 0.12 60
18
0.05 0.1 50
24
0.04 0.08 40

0.03 0.06 30

0.02 0.04 20

0.01 0.02 10

0 0 0
0 5 10 15 20 25 300 400 500 600 700 800
(nm)
Time (h) wavelenght (nm)

Fig. 6. Results of the kinetic tests: series III; Dye load = 0.013 kgCOD/(m3 d): a) Absorbance pattern for the dominant wavelength; b) Evolution of the colour spectrum vs. time; c) Colour
removal at the different wavelengths (evaluated on the test replicates).
592 M.C. Tomei et al. / Science of the Total Environment 573 (2016) 585–593

Table 6 first part of the anaerobic phase followed by dye release. In other
COD data and % removal. AN = at the end of the anaerobic phase; AE = at the end of the words, the maximum decolourization is achieved just after the first 4–
aerobic phase (average values of the two replicates of the kinetic tests in the three Series).
5 h due to sorption into the polymer, but this does not imply actual deg-
Series COD (mg/L) Removal (%) radation of dyes, which is achieved only after their release into the
CODTot CODDye TCR aqueous phase. The mechanism of uptake-release in the TPPBs is
highlighted by the stable low colour level observed during the aerobic
Feed Feed AN AE AN AE AN AE AN AE
Tot (dye)a period: this pattern can be explained with the near-complete removal
of the colour released from the polymer. Performance of TPPB system
I 425 21.2 18.2 8.8 95.7 97.9 14.4 58.6 50.3 58.9
II 443 22.2 19.6 9 95.6 98.0 11.5 59.4 55.7 62.2
was followed for more than 20 days: it is worth noting that no satura-
III 485 24.3 14.4 6 97.0 98.8 40.6 75.3 70.6 74.9 tion of the polymer and/or loss in the overall efficiency of the process
a were observed during this period, in spite of the increased load. This
Estimated taking into account that the total COD of feed solution is constituted by a
1:20 acetate:dye ratio (on COD basis). confirms the robustness of the uptake/release mechanism of amor-
phous polymer and the added advantage of sub-inhibitory and optimal
conditions for the biomass.
The COD trend in the three series gives indirect indications on the
25 fate of intermediates and on the role of the polymer. The low anaerobic
Conventional
SOUR (mgO2/gVSS L)

removal for CODDye (11–14%) in Series I and II, if compared to the

20 TPPB corresponding high anaerobic removal of TCR (50–56%), highlights the
formation of intermediates (presumably aromatic amines) not
15 biodegraded in the anaerobic step. According to the better aerobic bio-
degradability of aromatic amines under aerobic conditions, the removal
10 of CODDye reaches 59% for both series and is comparable with the final
TCR removal of 59–62%.
5 The beneficial effect of the polymer addition in the third series is ev-
idenced by higher anaerobic CODDye removal (41%) reaching 75% after
0
the aerobic period. Also in this case the difference between the CODDye
0 100 200 300 400 500 600 700
and TCR removal in the anaerobic step may be attributable to the inter-
time aerobic phase (min)
mediates not removed under anaerobic conditions, which are
biodegraded in the aerobic phase as demonstrated by same removal ef-
Fig. 7. SOUR time profiles during the first and the third experimental series: Dye load =
0.01 kgCOD/(m3 d); Biomass concentration = 2760 and 2560 mgVSS/L for 1st and 3rd ficiency (~75%) for CODDye and TCR.
series, respectively. In series III, the positive effect of the added polymer is reflected in
the reduction of the toxicity on the biomass after the sequential treat-
Patel and Madamwar (2000) reported high COD removal (95%) and ef- ment, as highlighted by the SOUR profiles reported in Fig. 7. This finding
ficient biomethanation for a fixed-film anaerobic bioreactor applied to highlights the increased biomass activity, attributable to the reduced
the treatment of petrochemical acidic wastewater (pH 2.5). dye concentration in the liquid phase to which the biomass is exposed.
In the third Series, a clear beneficial effect of the polymer on colour Finally, a visualization of the polymer effectiveness in the abiotic and
removal is observed, with efficiencies reaching values up to 90%. More- biotic tests is given in Fig. 8, which shows very dark colour after absorp-
over, the different evolution of the spectrum, with respect to Series I and tion, and light colouration of the beads after three weeks of use in the
II, may be due to colour removal attributable to polymer uptake in the TPPB.

Fig. 8. Colour changes in polymer beads, new beads and beads utilized in abiotic and biotic experiments.
 M.C. Tomei et al. / Science of the Total Environment 573 (2016) 585–593 593

5. Conclusions Kuai, L., De Vreese, I., Vandevivere, P., Verstraete, W., 1998. GAC-amended UASB reactor
for the stable treatment of toxic textile wastewater. Environ. Technol. 19, 1111–1117.
Lourenco, N.D., Novais, J.M., Pinheiro, H.M., 2006. Kinetic studies of reactive azo dye decol-
TPPB bioreactor improved the performance of the sequential anaer- orization in anaerobic/aerobic sequencing batch reactors. Biotechnol. Lett. 28,
obic-aerobic process applied to the removal of azo dyes from a real dye 733–739.
Mosca Angelucci, D., Tomei, M.C., 2015. Pentachlorophenol aerobic removal in a sequen-
bath wastewater, reaching 84% removal for the characteristic wave- tial reactor: start-up procedure and kinetic study. Environ. Technol. 36, 327–335.
length, and 75% for TCR in comparison to 68% and 59% observed in the Muda, K., Aris, A., Salim, M.R., Ibrahim, Z., Van Loosdrecht, M.C.M., Ahmad, A., Nawahwi,
conventional sequential process. M.Z., 2011. The effect of hydraulic retention time on granular sludge biomass in
treating textile wastewater. Water Res. 45, 4711–4721.
The COD mass balance demonstrated that for all investigated operat- Pandey, A., Singh, P., Iyengar, L., 2007. Bacterial decolorization and degradation of azo
ing conditions, the aerobic step achieved an almost complete removal of dyes. Int. Biodeterior. Biodegrad. 59, 73–84.
the intermediates produced in the first anaerobic step. Patel, H., Madamwar, D., 2000. Biomethanation of low pH petrochemical wastewater
using up-flow fixed-film anaerobic bioreactors. World J. Microbiol. Biotechnol. 16,
The study is a first demonstration of the applicability of TPPB biore-
69–75.
actors to industrial wastewater treatment, in this specific case, given the Pereira, L., Alves, M., Grohmann, E., 2012. Dyes - environmental impact and remediation.
homogeneity of the fed substrate (i.e. the same group of dyes) in the In: Malik, A. (Ed.), Environmental Protection Strategies for Sustainable Development.
segregated wastewater stream, one polymer has been effective for en- Springer Science Business, pp. 111–162.
Popli, S., Patel, U.D., 2015. Destruction of azo dyes by anaerobic–aerobic sequential biolog-
hancing colour removal and reducing the toxicity to the biomass. For ical treatment: a review. Int. J. Environ. Sci. Technol. 12, 405–420.
more complex industrial wastewater, characterized by different groups Ricco, G., Tomei, M.C., Ramadori, R., Laera, G., 2004. Toxicity assessment of common xeno-
of substrates, the flexibility of the proposed sequential TPPB system can biotic compounds on municipal activated sludge: comparison between respirometry
and Microtox®. Water Res. 38, 2103–2110.
be exploited by operating with a mixture of different polymers tailored Sanmuga Priya, E., Senthamil Selvan, P., Umayal, A.N., 2015. Biodegradation studies on
to the different substrate components, and by varying the operating dye effluent and selective Remazol dyes by indigenous bacterial species through
conditions, i.e. duration and reaction environments to optimize the per- spectral characterization. Desal. Water Treat. 55, 241–251.
Saratale, R., Saratale, G., Chang, J., Govindwar, S., 2011. Bacterial decolourization and deg-
formance of the biological process. radation of azo dyes: a review. J. Taiwan Inst. Chem. Eng. 42, 138–157.
Sen, S., Demirer, G.N., 2003. Anaerobic treatment of real textile wastewater with a fluid-
Acknowledgements ized bed reactor. Water Res. 37, 1868–1878.
Shah, M.P., Patel, K.A., Nair, S.S., Darji, A.M., 2014. Microbial degradation and decoloriza-
tion of reactive dyes by Bacillus spp. ETL-1979. Am. J. Microbiol. Res. 2, 16–23.
The Authors express their sincere acknowledgements to Eng. Solis, M., Solis, A., Perez, H., Manjarrez, N., Floresa, M., 2012. Microbial decolouration of
Giovanni Bergna of Lariana Depur S.p.a. (Como, Italy) for the textile azo dyes: a review. Process Biochem. 47, 1723–1748.
Sponza, D.T., Isik, M., 2002. Decolorization and azo dye degradation by anaerobic/aerobic
wastewater supply and for the valuable information given for its
sequential process. Enzym. Microb. Technol. 31, 102–110.
characterization. Supaka, N., Juntongjin, K., Damronglerd, S., Delia, M.L., Strehaiano, P., 2004. Microbial de-
colorization of reactive azo dyes in a sequential anaerobic–aerobic system. Chem.
References Eng. J. 99, 169–176.
Telke, A.A., Kalyani, D.C., Jadhav, J.P., Govindwar, S.P., 2008. Kinetics and mechanism of re-
Amsden, B.G., Bochanysz, J., Daugulis, A.J., 2003. Degradation of xenobiotics in a active red 141 degradation by a bacterial isolate Rhizobium radiobacter MTCC 8161.
partitioning bioreactor in which the partitioning phase is a polymer. Biotechnol. Acta Chim. Slov. 55, 320–329.
Bioeng. 84, 399–405. Tomei, M.C., Daugulis, A.J., 2013. Feasibility of operating a solid–liquid bioreactor with
APHA, AWWA, American Public Health Association and American Water Works used automobile tires as the sequestering phase for the biodegradation of inhibitory
Association, 1998n. Standards Methods for the Examination of Water and Wastewa- compounds. J. Environ. Manag. 125, 7–11.
ter. 20th ed. Water Environment Federation, Washington DC. Tomei, M.C., Annesini, M.C., Bussoletti, S., 2004. 4-Nitrophenol biodegradation in a se-
Bacon, S.L., Scott Parent, J., Daugulis, A.J., 2014. A framework to predict and experimental- quencing batch reactor: kinetic study and effect of filling time. Water Res. 38,
ly evaluate polymer-solute thermodynamic affinity for two-phase partitioning biore- 375–384.
actor (TPPB) applications. J. Chem. Technol. Biotechnol. 89, 948–956. Tomei, M.C., Annesini, M.C., Prpich, G.P., Daugulis, A.J., 2009. Biodegradation of 4-nitro-
Chan, Y.J., Chong, M.F., Law, C.L., Hassell, D.G., 2009. A review on anaerobic-aerobic treat- phenol in a two phase system operating with polymers as the partitioning phase. En-
ment of industrial and municipal wastewater. Chem. Eng. J. 155, 1–18. viron. Sci. Technol. 43, 7105–7110.
Chang, J.S., Chen, B.Y., Lin, Y.S., 2004. Stimulation of bacterial decolorization of an azo dye Tomei, M.C., Rita, S., Mininni, G., 2011a. Performance of sequential anaerobic/aerobic di-
by extracellular metabolites from Escherichia coli strain NO3 Bioresour. Technology gestion applied to municipal sewage sludge. J. Environ. Manag. 92, 1867–1873.
91, 243–248. Tomei, M.C., Rita, S., Mosca Angelucci, D., Annesini, M.C., Daugulis, A.J., 2011b. Treatment
Chen, Y., Cheng, J.J., Creamer, K.S., 2008. Inhibition of anaerobic digestion process: a re- of substituted phenol mixtures in single-phase and two-phase solid-liquid
view Bioresour. Technology 99, 4044–4064. partitioning bioreactors. J. Hazard. Mater. 191, 190–195.
Demirel, B., Yenigün, O., 2002. Two-phase anaerobic digestion processes: a review. Tomei, M.C., Mosca Angelucci, D., Daugulis, A.J., 2014. The use of used automobile tyres in
J. Chem. Technol. Biotechnol. 77, 743–755. a partitioning bioreactor for the biodegradation of xenobiotic mixtures. Environ.
Fang, H.H.P., Liu, H., 2002. Effect of pH on hydrogen production from glucose by a mixed Technol. 35, 75–81.
culture. Bioresour. Technol. 82, 87–93. Tomei, M.C., Soria Pascual, J., Mosca Angelucci, D., 2016. Analysing performance of real
Firmino, P.I.M., da Silva, M.E.R., Cervantes, F.J., dos Santos, A.B., 2010. Colour removal of textile wastewater bio-decolourization under different reaction environments.
dyes from synthetic and real textile wastewaters in one- and two-stage anaerobic J. Clean. Prod. 129, 468–477.
systems. Bioresour. Technol. 101, 7773–7779. van der Zee, F.P., Villaverde, S., 2005. Combined anaerobic–aerobic treatment of azo
Forgacs, E., Cserhati, T., Oros, G., 2004. Removal of synthetic dyes from wastewaters: a re- dyes—a short review of bioreactor studies. Water Res. 39, 1425–1440.
view. Environ. Int. 30, 953–971. van Lier, J.B., Mahmoud, N., Zeeman, G., 2008. Anaerobic wastewater treatment. In:
Joshi, M., Bansal, R., Purwar, R., 2004. Colour removal from textile effluents. Ind. J Fibre Henze, M., van Loosdrecht, M.C.M., Ekama, G.A., Brdjanovic, D. (Eds.), J.B. van Lier. Bi-
Textile Res. 29, 239–259. ological Wastewater Treatment: Principles Modelling and Design. IWA Publishing,
Kapdan, I.K., Tekol, M., Sengul, F., 2003. Decolorization of simulated textile wastewater in London, UK. ISBN: 9781843391883.
an anaerobic–aerobic sequential treatment system. Process Biochem. 38, 1031–1037. Williams, T.M., Unz, R.F., 1989. The nutrition of Thiothrix, Type 021 N. Beggiatoa and
Kim, I.S., Hwang, M.H., Jang, N.J., Hyun, S.H., Lee, S.T., 2004. Effect of low pH on the activity Leucothrix strains. Water Res. 23, 15–22.
of hydrogen utilizing methanogen in bio-hydrogen process. Int. J. Hydrog. Energy 29,
1133–1140.