Abstract

In the present study, we evaluate the behavior of real textile wastewater treatment using a system composed of two sequential pilot-scale reactors (anaerobic followed by aerobic) during 622 days. The work focused on the competition between color and sulfate removal processes, when the hydraulic retention time (HRT) was increased in the anaerobic/aerobic reactors from 16/12 hours in phase I (PI) to 4/3 days in phase II (PII). The organic matter was successfully removed in both phases through the system, and the highest efficiency (75%) was achieved in PII. The increase in the HRT did not improve azo dye degradation under anaerobic conditions. Instead, it favored sulfate reduction, which removal efficiency increased from 26% in PI to 75% in PII. Aromatic amines were detected in the anaerobic reactor effluent and removed in the aerobic reactor.

INTRODUCTION

The treatment of textile effluents is still a challenge worldwide due to its composition, which usually includes chemical compounds such as dyes, sulfate, surfactants, oxidants (mainly hydrogen peroxide and potassium permanganate), salts and organic matter. In general, dyes are hydrophilic molecules linked to complex aromatic compounds, based on a chromophore group (O'Neill et al. 1999). For this reason, dye removal represents a bottleneck in the textile effluent treatment systems. Those containing one or more azo functional groups (–N=N–) represent approximately 80% of the dyes used in the textile industry (Singh et al. 2015). Azo dyes are known for their chemical stability and resistance to aerobic degradation (Pandey et al. 2007). Their degradation has been more successful under anaerobic conditions.

Depending on the structure of the azo dye, its anaerobic degradation releases aromatic amines, which can range from simple anilines to benzidines. Anilines are considered toxic to aquatic life, and possibly carcinogenic, while benzidines are known for their toxicity to humans and their carcinogenic potential (Pinheiro et al. 2004). Aromatic amines are well degraded through aerobic processes (Stolz 2001), and the application of small amounts of oxygen has stimulated the breakdown of aromatic amines in anaerobic systems (Gavazza et al. 2015). The aerobic phase is required for the conversion of aromatic amines to CO2 and H2O (Tan et al. 2000), thus removing the effluent's toxicity.

Azo dyes are electron acceptors in the anaerobic microbial electron transportation chain, while the organic matter, usually available in textile wastewaters as starch or short chain fatty acids, serves as an electron donor. The presence of competitive electron acceptors in the anaerobic stage, such as nitrate and sulfate, can divert the reducing equivalents and result in low color removal under anaerobic conditions (Amaral et al. 2014). Sulfate is normally found in textile-processing wastewater and generally comes from the salts added to dye baths for ionic strength adjustment (van der Zee et al. 2003). When it is reduced under anaerobic conditions by sulfate-reducing bacteria, the sulfide produced can in turn serve as an electron donor for azo dye removal (Albuquerque et al. 2005). Cirik et al. (2013) studied the competitive biochemical reactions between sulfate and the azo dye Remanzol Brilliant Violet 5R, using glucose as an electron donor in the synthetic textile effluent. The authors reported that the sulfate and azo dye reductions took place simultaneously in all operational conditions. They also reported that increasing the sulfate concentration generally stimulated the azo dye reduction.

Although both sulfate and azo dye reduction have been reported in similar conditions of oxidation–reduction potential (ORP), most of these studies were conducted using synthetic textile wastewaters, which excludes the interference of many chemical compounds found in real wastewater. Therefore, the aim of the present study was to evaluate the influence of the hydraulic retention time (HRT) on the treatability of real textile wastewater through a combination of anaerobic and aerobic processes, focusing on the competition between azo dye and sulfate reductions.

METHOD

Industrial effluent characterization

The present study was conducted in a medium-sized textile factory located in the city of Caruaru (geographical location: 08°17′00″ south and 35°58′34″ west) in the state of Pernambuco (Brazil) over a period of 622 days. All chemical products used by the factory were registered monthly. The dyes used by the factory did not have their chemical structure on their labels. Only the commercial name of the dye and the application method was supplied on the label. The dyes were identified using X-ray fluorescence and diffraction (Spectrum 100FT-IR and Brunker DPX-300), followed by mass spectrophotometry (QP 5000/DI-50). The industrial effluent was characterized during one week by sampling twice a day, at 10 am and 3 pm, when the highest effluent production occurs. The characterization parameters were temperature, pH, conductivity, salinity, total and partial alkalinity, total and filtered chemical oxygen demand (COD), sulfate, sulfide, color, and ORP. All parameters were analyzed in accordance with Standard Methods for the Examination of Water and Wastewater (APHA et al. 2005).

Reactor configuration

The two pilot-scale reactors (Figure 1) were installed in the factory, with an average water consumption of 1,800 m3/month. The main industrial activity was jeans dyeing. The effluent from the stages of degumming, dyeing and softening was equalized and then pumped into the reactors’ feeding reservoir (500 L). The two reactors were sequentially fed by gravidity (Figure 1).

Figure 1

Anaerobic reactor followed by aerobic reactor used in the experiment: (a) schematic illustration of the system and (b) photograph of the reactors installed in the textile factory.

Figure 1

Anaerobic reactor followed by aerobic reactor used in the experiment: (a) schematic illustration of the system and (b) photograph of the reactors installed in the textile factory.

The first reactor was an upflow anaerobic sludge blanket (UASB) reactor type with a 250 L working volume, and the second reactor was a submerged aerated biofilter (SAB) with a 187 L working volume (Figure 1(a)). The SAB was radially aerated through a perforated PVC tube and by a compressor (Schulz, CSA 8.2/25), regulated by a valve (Ball valves ½″ NPT) to maintain the dissolved oxygen at 3 ± 0.2 mg O2/L at the top of SAB reactor. The SAB was filled with encapsulated polyurethane foam (BioBob®).

In phase I (PI), the HRT applied was 16 and 12 hours for the UASB and SAB reactors, respectively. In phase II (PII), the corresponding HRTs were 4 and 3 days, respectively. The UASB was inoculated with anaerobic sludge from a previous operation from the same reactor in another small factory, as described by Ferraz et al. (2011). The SAB was not inoculated. It is worth mentioning here that when the reactors were performing PI, dyeing raw white cotton clothes predominated in the factory over the conventional degumming processes, which became dominant in PII.

Reactor monitoring

The reactors were monitored by analyzing the following parameters: flow rate and pH (daily); color, COD, sulfate and phosphate (twice weekly). The analytical procedures followed Standard Methods for the Examination of Water and Wastewater (APHA et al. 2005). Aromatic amine formation was qualitatively assessed by scanning the filtered (0.22 μm fiberglass membrane) samples in the UV-VIS range from 200 to 600 nm. According to Pinheiro et al. (2004) aromatic amines absorb light in the wavelength range of 288–300 nm, which are usually free of influence from anaerobic digestion metabolites.

RESULTS AND DISCUSSION

Chemical products used by the jeans dyeing factory

The dyes used by the factory are listed in Table 1. In total, 28 different kinds of dyes were used during the experimental period. The chemical structure of the dyes was not specified on the product labels nor made available by the manufacturer. The only dye that was properly identified was Reactive Red 241, a mono-azo dye (Figure 2). It was not possible to accurately identify the other dyes due to the low purity level. However, azo bonds were detected in the chemical structure of Diresul® Brown FSB, Diresul® Denim Blue GB, Diresul® Olive RDT-T, and Diresul® Black RDT-VLS 200. Thus, azo dyes represented at least 65% of the dyes used by the factory. Dye consumption in this medium-sized factory (913 kg/d) was higher than that in the small-sized factory (219 kg/d), where the reactors were previously installed (Amaral et al. 2014).

Table 1

Dye consumption during the 622 days of the experimental period

Dye Total (kg) kg/d 
Diresul® Yellow gold 48 0.07 
Diresul® Bordeaux RDT 165 0.25 
Diresul® Brown FSB 2,225 3.37 
Diresul® Denin Blue GB 195 0.30 
Diresul® Olive RDT-T 885 1.34 
Diresul® Black RDT-VLS 200 755 1.14 
Diresul® Green 20 0.05 
Drimaren® Yellow CL-5B 418 0.63 
Drimaren® Blue CL-R 80 0.12 
Drimaren® Orange CL-3R 79 0.12 
Drimaren® Turquoise CL-B 173 0.26 
Reactive Red 241 447 0.68 
Black Direct D Jag-S 100 0.15 
Brown Biosulphur CS 100 0.15 
Indosol® Navy blue 20 0.02 
Solar black 40 0.06 
Olive green 13 0.02 
Pigment paste yellow gem 70 0.10 
Pigment paste pink fluor 25 0.04 
Pigment paste green chlorine 25 0.04 
Pigment paste pink pink 15 0.04 
Pigment paste blood orange 15 0.02 
Pigment paste lux silver 0.02 
Pigment paste blue fluorine 35 0.01 
Pigment paste violet fluorine 0.05 
Pigment paste green fluorine 20 0.01 
Pigment paste red BSB 10 0.03 
Pigment paste white 65 0.02 
Total 6,024 9.13 
Dye Total (kg) kg/d 
Diresul® Yellow gold 48 0.07 
Diresul® Bordeaux RDT 165 0.25 
Diresul® Brown FSB 2,225 3.37 
Diresul® Denin Blue GB 195 0.30 
Diresul® Olive RDT-T 885 1.34 
Diresul® Black RDT-VLS 200 755 1.14 
Diresul® Green 20 0.05 
Drimaren® Yellow CL-5B 418 0.63 
Drimaren® Blue CL-R 80 0.12 
Drimaren® Orange CL-3R 79 0.12 
Drimaren® Turquoise CL-B 173 0.26 
Reactive Red 241 447 0.68 
Black Direct D Jag-S 100 0.15 
Brown Biosulphur CS 100 0.15 
Indosol® Navy blue 20 0.02 
Solar black 40 0.06 
Olive green 13 0.02 
Pigment paste yellow gem 70 0.10 
Pigment paste pink fluor 25 0.04 
Pigment paste green chlorine 25 0.04 
Pigment paste pink pink 15 0.04 
Pigment paste blood orange 15 0.02 
Pigment paste lux silver 0.02 
Pigment paste blue fluorine 35 0.01 
Pigment paste violet fluorine 0.05 
Pigment paste green fluorine 20 0.01 
Pigment paste red BSB 10 0.03 
Pigment paste white 65 0.02 
Total 6,024 9.13 
Figure 2

Molecular structure of the azo dye Reactive Red 241 (CAS 89157-03-9; C.I. 18220).

Figure 2

Molecular structure of the azo dye Reactive Red 241 (CAS 89157-03-9; C.I. 18220).

The other chemical products used by the factory during the experimental period are shown in Table 2. It is important to highlight the amount of sodium metabisulfite (8,925 kg), resulting in a consumption of 1,352 kg/d, which is twice as high as that in the small factory (Amaral et al. 2014). Sodium metabisulfite was used as a bleaching agent and was the main source of sulfate in the effluent.

Table 2

Chemicals and quantities used during the experimental period of 622 days

Product Total (kg) 
Oxalic acid 25 
Acetic acid 1,790 
Chlorine 930 
Sodium hydroxide 1,095 
Hydrogen peroxide 845 
Sodium metasilicate 1,725 
Potassium permanganate 1,020 
Sodium metabisulfite 8,925 
Soda sodium carbonate 2,525 
Brancal 500 
Oxidant Diresul 600 
Sodyefide B Liq 4,250 
Trisize AB 3000 400 
Ceramim HB 140 
Biofix 350 
Trifloc 200 
Product Total (kg) 
Oxalic acid 25 
Acetic acid 1,790 
Chlorine 930 
Sodium hydroxide 1,095 
Hydrogen peroxide 845 
Sodium metasilicate 1,725 
Potassium permanganate 1,020 
Sodium metabisulfite 8,925 
Soda sodium carbonate 2,525 
Brancal 500 
Oxidant Diresul 600 
Sodyefide B Liq 4,250 
Trisize AB 3000 400 
Ceramim HB 140 
Biofix 350 
Trifloc 200 

Textile effluent characteristics

The results of the effluent characterization are shown in Table 3. The significant variability in the composition is a consequence of the different jeans-dyeing processes used. The pH was suitable for anaerobic digestion without any correction. The salinity level was similar to that previously found for textile effluent (Correia et al. 1994) and it is not within the anaerobic toxicity range for the decolorization process (Vyrides et al. 2014). The ORP (−162 mV) indicates that anaerobic conditions had started in the equalization tank.

Table 3

Results of textile effluent characterization parameters

Parameter Unit Mean value 
Temperature (°C) 27.0 ± 1.2 
pH – 7.5 ± 0.7 
Electrical conductivity (mS/cm) 4.8 ± 1.3 
Salinity ‰ 3.2 ± 1.3 
Oxidation-reduction potential (ORP) (mV) −162.0 ± 215.0 
Total alkalinity (mg CaCO3/L) 510.0 ± 1.7 
Partial alkalinity (mg CaCO3/L) 390.0 ± 65.0 
Apparent color (Pt/Co) 604.0 ± 456.0 
True color (Pt/Co) 288.0 ± 119.0 
Total COD (mg O2/L) 692.0 ± 459.0 
Filtered COD (mg O2/L) 401.0 ± 170.0 
Sulfate  334.1 ± 40 
Sulfide (mg HS/L) 20.6 ± 15.0 
Parameter Unit Mean value 
Temperature (°C) 27.0 ± 1.2 
pH – 7.5 ± 0.7 
Electrical conductivity (mS/cm) 4.8 ± 1.3 
Salinity ‰ 3.2 ± 1.3 
Oxidation-reduction potential (ORP) (mV) −162.0 ± 215.0 
Total alkalinity (mg CaCO3/L) 510.0 ± 1.7 
Partial alkalinity (mg CaCO3/L) 390.0 ± 65.0 
Apparent color (Pt/Co) 604.0 ± 456.0 
True color (Pt/Co) 288.0 ± 119.0 
Total COD (mg O2/L) 692.0 ± 459.0 
Filtered COD (mg O2/L) 401.0 ± 170.0 
Sulfate  334.1 ± 40 
Sulfide (mg HS/L) 20.6 ± 15.0 

The apparent color (604 mg/L Pt/Co), in which the color is measured without removing turbidity, was twice as high as the true color, and indicates high turbidity level in the textile effluent. Low levels of true color (288 mg/L Pt/Co) and filtered COD (401 mg O2/L) were detected. In the case that azo dyes contributed 65% of the true color, it means that they generated 187 mg/L Pt/Co. Amorim et al. (2013) indicated a correlation of 0.06 mM (65 mg/L) of the tetra-azo dye DB22 corresponding to a true color of 400 mg/L Pt/Co. By applying that correlation in the present study, it would result in 30.4 mg/L of azo dye; this would require 0.04 × 10−2 electrons to break all azo bonds (the worst-case scenario would be four azo bonds). The complete oxidation of the filtered COD (401 mg O2/L) detected in the present study would release four electrons at most from the complete oxidation of CnHaObNc (Rabaey et al. 2010). If glucose were the source of COD (worse scenario), considering that each carbon atom loses four electrons during the oxidation process, 24 electrons are oxidized for 1 mol of glucose (180 g). Thus, the COD oxidation (376 mg/L of glucose from 401 mg O2/L of COD) in the present work would release a total of 0.05 electrons, thereby, indicating no limitation of electrons from the carbon source for the decolorization process.

Reactor performance

The ORP in the effluent of the UASB reactor, in both operational phases, was between −282 mV and −253 mV, which is closer to the sulfate reduction range (−220 mV) than to methanogenic conditions (−600 mV to −350 mV) (Table 4). This ORP is also compatible with the sulfate level detected in the influent (323 ± 35 mg in PI and 268 ± 68 mg in PII) (Table 5). An aerobic ORP range was detected in the effluent of the SAB reactor (+21 mV to +65 mV). The mean sodium and chloride content in the reactors' influent were 784 ± 66 mg Na+/L and 1,193 ± 170 mg Cl/L, respectively, in both phases. These figures did not change during the treatment process. The values found for sodium are between the ranges reported for stimulation (100 to 200 mg Na+/L) and a moderate inhibitory effect (3,500 mg Na+/L) (McCarty 1964). The differences found in the textile wastewater characteristics during the characterization (Table 3) and the reactor monitoring (Table 4) times, especially for alkalinity, reflect the variability of this kind of effluent in practice.

Table 4

Mean values of physical and chemical parameters

Parameter Unit System influent UASB effluent SAB effluent System influent UASB effluent SAB effluent 
Phase I (PI) Phase II (PII) 
Temperature °C 25.0 ± 3.0 25.0 ± 2.0 25.0 ± 1.0 24.0 ± 3.4 24.0 ± 2.2 24.0 ± 2.6 
pH – 7.9 ± 0.3 7.6 ± 0.8 8.2 ± 0.6 7.9 ± 0.5 7.7 ± 0.6 8.0 ± 0.4 
ORP mV −148.0 ± −156.5 −253.0 ± −107.8 21.0 ± 97.2 −155.2 ± −123.1 −282.4 ± −143.6 65.2 ± 41.4 
Electrical conductivity mS/cm 4.6 ± 0.6 4.9 ± 0.8 4.9 ± 0.5 4.4 ± 0.6 4.6 ± 0.7 4.6 ± 0.6 
Salinity ‰ 2.4 ± 0.3 2.6 ± 0.5 2.6 ± 0.3 2.3 ± 0.4 2.5 ± 0.5 2.4 ± 0.3 
Partial alkalinity mg CaCO3/L 336.0 ± 81.0 279.0 ± 107.0 260.0 ± 113.0 344.0 ± 123.0 310.0 ± 144.0 237.0 ± 142.0 
Total alkalinity mg CaCO3/L 310.0 ± 104.0 359.0 ± 131.0 303.0 ± 125.0 420.0 ± 154.0 378.0 ± 171.0 283.0 ± 155.0 
Parameter Unit System influent UASB effluent SAB effluent System influent UASB effluent SAB effluent 
Phase I (PI) Phase II (PII) 
Temperature °C 25.0 ± 3.0 25.0 ± 2.0 25.0 ± 1.0 24.0 ± 3.4 24.0 ± 2.2 24.0 ± 2.6 
pH – 7.9 ± 0.3 7.6 ± 0.8 8.2 ± 0.6 7.9 ± 0.5 7.7 ± 0.6 8.0 ± 0.4 
ORP mV −148.0 ± −156.5 −253.0 ± −107.8 21.0 ± 97.2 −155.2 ± −123.1 −282.4 ± −143.6 65.2 ± 41.4 
Electrical conductivity mS/cm 4.6 ± 0.6 4.9 ± 0.8 4.9 ± 0.5 4.4 ± 0.6 4.6 ± 0.7 4.6 ± 0.6 
Salinity ‰ 2.4 ± 0.3 2.6 ± 0.5 2.6 ± 0.3 2.3 ± 0.4 2.5 ± 0.5 2.4 ± 0.3 
Partial alkalinity mg CaCO3/L 336.0 ± 81.0 279.0 ± 107.0 260.0 ± 113.0 344.0 ± 123.0 310.0 ± 144.0 237.0 ± 142.0 
Total alkalinity mg CaCO3/L 310.0 ± 104.0 359.0 ± 131.0 303.0 ± 125.0 420.0 ± 154.0 378.0 ± 171.0 283.0 ± 155.0 
Table 5

Mean values of apparent color, true color and sulfate in the treatment system

Parameter System influent UASB effluent SAB effluent 
Apparent color in PI (mg/L Pt/Co) 1,828 ± 1,253 601 ± 501 316 ± 221 
Apparent color in PII (mg/L Pt/Co) 2,624 ± 1,520 1,181 ± 824 408 ± 335 
True color in PI (mg/L Pt/Co) 536 ± 459 282 ± 321 169 ± 161 
True color in PII (mg/L Pt/Co) 753 ± 672 325 ± 279 211 ± 186 
Sulfate in PI  323 ± 35 238 ± 58 204 ± 58 
Sulfate in PII  268 ± 63 61 ± 72 176 ± 76 
Parameter System influent UASB effluent SAB effluent 
Apparent color in PI (mg/L Pt/Co) 1,828 ± 1,253 601 ± 501 316 ± 221 
Apparent color in PII (mg/L Pt/Co) 2,624 ± 1,520 1,181 ± 824 408 ± 335 
True color in PI (mg/L Pt/Co) 536 ± 459 282 ± 321 169 ± 161 
True color in PII (mg/L Pt/Co) 753 ± 672 325 ± 279 211 ± 186 
Sulfate in PI  323 ± 35 238 ± 58 204 ± 58 
Sulfate in PII  268 ± 63 61 ± 72 176 ± 76 

COD removal

The mean COD (Figure 3(a)) in the reactors' influent in PI (664 ± 227 mg O2/L) was lower than in PII (926 ± 448 O2/L). During PI, the predominant process was the dyeing of raw white cotton clothes using different kinds of paste (Table 1). Degumming was not frequently performed in PI, causing the low COD level. However, in PII degumming processes dominated, causing the higher COD. Due to the lower influent flow increasing the HRT in PII, the resulting organic loading rate (OLR) applied to the UASB reactor was 1.77 and 1.41 kg COD/m3·d for PI and PII, respectively. The corresponding values for the aerobic reactor were 1.33 and 0.73 kg COD/m3·d, respectively. Therefore decreasing the OLR from PI to PII and due to the high degradability of the organic matter from the degumming processes, the COD removal efficiency increased from 44% to 62% for the UASB reactor, and from 64% to 75% for the system. The differences are statistically significant (ρ < 0.01). These results are close to those found by Firmino et al. (2010) for COD removal in real or synthetic textile wastewater. Those authors reported that the efficiency ranged from 60 to 65% in a UASB reactor fed with real textile wastewater (HRT of 20 hours and influent COD of 1.0 g O2/L). Additionally, higher COD removal efficiency rates have been reported when textile effluent was mixed with another kind of wastewater; for instance, 88% removal efficiency was obtained by Senthilkumar et al. (2011) for a 30:70 mixture ratio of textile:sago wastewater.

Figure 3

Behavior through the experimental time for (a) COD during the operational period in the () system influent, () UASB reactor effluent, and () system effluent; and for (b) color removal efficiency found for () UASB reactor, () SAB reactor, and () system.

Figure 3

Behavior through the experimental time for (a) COD during the operational period in the () system influent, () UASB reactor effluent, and () system effluent; and for (b) color removal efficiency found for () UASB reactor, () SAB reactor, and () system.

Color and sulfate removal

Figures 3(b) and 4 display the results for color and sulfate throughout the experimental period; and Table 5 displays the corresponding mean values. Increasing the HRT from 16 hours to 4 days did not increase the color removal efficiency in the UASB reactor; on the contrary, it was 67% in PI and 55% in PII. However, the sulfate removal efficiency increased from 26% in PI to 75% in PII, thereby indicating an electron transfer limitation for dyes in comparison to sulfate, particularly in PII. It has been widely reported that thermodynamic conditions can lead to the predominance of sulfate or dye reduction, depending on the kind of dyes in the wastewater. When synthetic wastewater was used, most of the published studies have reported that sulfate reduction occurs simultaneously with the dye removal (Albuquerque et al. 2005; Prato-Garcia et al. 2013). However, in those studies they dealt with mono-azo dye compounds, which are less complex, less stable and require fewer electrons during the reduction process. In the case of real wastewater, there is a significant lack of reports on this competition in practice. The reactors used in the present study were previously installed in another factory with similar sulfate levels in the wastewater, although with much lower diversity of dyes (seven types) (Amaral et al. 2014). A predominance of sulfate reduction processes in relation to color removal was observed by those authors.

Figure 4

Average and standard deviation for sulfate concentration found in the () influent, () UASB reactor effluent, and () SAB effluent for phases PI and PII.

Figure 4

Average and standard deviation for sulfate concentration found in the () influent, () UASB reactor effluent, and () SAB effluent for phases PI and PII.

In the present study, the influent apparent color (worse case) was 1,828 mg/L Pt/Co in PI and 2,624 mg/L Pt/Co in PII (Table 5), requiring 0.39 × 10−2 and 0.56 × 10−2 electrons to reduce the dyes, respectively. By using glucose, the COD required for dye reduction in PI and PII would be 31 and 45 and mg O2/L, respectively. Since the mean sulfate concentration in the UASB influent was 323 and 268 in PI and PII, respectively, the COD required for sulfate reduction was 216 and 180 mg O2/L, respectively (stoichiometric requirement of 0.67 for ratio). The influent COD content in both phases indicates that possibly there were no COD limitations for the simultaneous occurrence of sulfate and color removal processes. Biodegradable compounds were probably available as electron donors, such as the acetic acid used by the factory (Table 2) and starch from the degumming process applied in PII. Additionally, the COD:N:P ratios found in the UASB influent (500:16:10 in PI and 500:6:5 in PII) indicate no nutrient limitation.

On the other hand, there is also no evidence that the thermodynamic conditions controlled the process. ORP values of −253 and −282 mV (Table 4) were found in the UASB effluent in PI and PII, respectively. The ORP of −220 mV has been cited as standard for sulfate reduction through H2S formation (Dos Santos et al. 2007). Connell & Patrick Jr. (1968) reported no H2S formation from sulfate reduction at above −150 mV. However, no standard ORP range has been established for azo dye reduction. Ozdemir et al. (2013) analyzed an anaerobic baffled reactor that was fed with synthetic textile wastewater containing 2,000 mg/L of sulfate, ethanol as the carbon source (1,340 mg COD/L), and a mono-azo dye (20 to 200 mg/L). The authors found that −280 mV was the lowest ORP at which the reductions of sulfate and azo dye were concomitant. In addition, Bromley-Challenor et al. (2000) used synthetic wastewater and reported azo dye reduction under an ORP as low as −50 mV. In the present study, the reduced environment (Table 4) and the COD available in the UASB reactor (664 mg O2/L in PI and 926 mg O2/L in PII) were favorable for the simultaneous occurrence of sulfate (required COD of 216 and 180 mg O2/L in PI and PII, respectively) and dye (required COD of 31 and 45 mg O2/L in PI and PII, respectively) reduction processes. Therefore, considering the huge diversity of chemical compounds used by the factory (Tables 1 and 2), other compounds may have played a more significant role in the anaerobic process of color removal for the real textile wastewater used in the present study.

Aromatic amines

According to Pinheiro et al. (2004), aromatic amines exhibit a light intense absorption in the range of 288 to 300 nm, without the interference of contaminants usually found in textile wastewater or by anaerobic degradation by-products. Figure 5 displays a scanning range (200 to 600 nm) that is representative of PI and PII. The effluent from the UASB reactor exhibited higher absorbance values between 288 and 300 nm, when compared either with the system influent or with the effluent from the aerobic reactor in both phases. This indicates that color removal in the UASB reactor occurred through the formation of aromatic amines, but these were subsequently removed under aerobic conditions. It is important to note that there was a huge reduction in the peak of aromatic amines after the aerobic stage, particularly during PI (Figure 5(a)). This is usually correlated with a reduction in toxicity for this kind of effluent (Ferraz et al. 2011; Amaral et al. 2014).

Figure 5

Typical UV-VIS scanning performed in the range of 200–600 nm for the () influent, () UASB reactor effluent, and () SAB effluent for (a) Phase I and (b) Phase II.

Figure 5

Typical UV-VIS scanning performed in the range of 200–600 nm for the () influent, () UASB reactor effluent, and () SAB effluent for (a) Phase I and (b) Phase II.

CONCLUSIONS

In the present study, the performance of two sequential pilot-scale reactors (anaerobic and aerobic) for the treatment of real textile effluent from a medium-sized jeans dyeing factory, was evaluated. Firstly, the main characteristics of the industrial effluent were assessed. It contained: 28 kinds of dyes of which at least 65% were azo dyes; it was identified that either the chemical products or the dyes used by the factory were of poor quality. When jeans washed through degumming were compared with white cotton clothes and the OLR was reduced from 1.77 to 1.41 kg COD/m3.d, the effluent degradability was higher and the COD removal efficiency increased from 44% to 62%, in the UASB reactor, and from 64% to 75% for the whole system. Concerning the competition between azo dye and sulfate reduction, the increase in the HRT from 16 hours in phase PI to 4 days in phase PII, the sulfate reduction was favored in the UASB, which increased from 26% to 75%. The dye removal did not improve with the increased HRT, being lower in PII (55%) than in PI (65%). Considering the favorable conditions detected for the simultaneous occurrence of dye and sulfate removal (reduced environment and no nutrient or electron limitation), the complex composition of the real textile wastewater may have influenced the dye reduction process. The pathway for azo dye degradation was through the formation of aromatic amines under anaerobic conditions, followed by aerobic removal.

ACKNOWLEDGEMENTS

The authors would like to thank the CNPq for the scholarship provided to the first author (143274/2011-2) and for the financial support (CNPq 552308/2011-0), and to the jeans dyeing factory where the experiment was performed. FACEPE has also supported the present work (APQ 0603-3.07/14). We also thank the BioBob company for donating the sponges.

REFERENCES

REFERENCES
Albuquerque
,
M. G. E.
,
Lopes
,
A. T.
&
Serralheiro
,
M. L.
2005
Biological sulphate reduction and redox mediator effects on azo dye decolourisation in anaerobic-aerobic sequencing batch reactors
.
Enzyme and Microbial Technology
36
(
5–6
),
790
799
.
Amaral
,
F. M.
,
Kato
,
M. T.
,
Florencio
,
L.
&
Gavazza
,
S.
2014
Color, organic matter and sulfate removal from textile effluents by anaerobic and aerobic processes
.
Bioresource Technology
163
,
364
369
.
Amorim
,
S. M.
,
Kato
,
M. T.
,
Florencio
,
L.
&
Gavazza
,
S.
2013
Influence of redox mediators and electron donors on the anaerobic removal of color and chemical oxygen demand from textile effluent
.
CLEAN – Soil, Air, Water
41
(
9
),
928
933
.
APHA, AWWA & WEF
2005
Standard Methods for the Examination of Water and Wastewater
,
21st edn
.
American Public Health Association
,
Washington, DC
,
USA
.
Bromley-Challenor
,
K. C. A.
,
Knapp
,
J. S.
,
Zhang
,
Z.
,
Gray
,
N. C. C.
,
Hetheridge
,
M. J.
&
Evans
,
M. R.
2000
Decolorization of an azo dye by unacclimated activated sludge under anaerobic conditions
.
Water Research
34
(
18
),
4410
4418
.
Connell
,
W. E.
&
Patrick
,
W. H.
Jr.
1968
Sulfate reduction in soil: effects of redox potential and pH
.
Science
159
(
3810
),
86
87
.
Correia
,
V. M.
,
Stephenson
,
T.
&
Judd
,
S. J.
1994
Characterization of textile wastewaters e a review
.
Environmental Technology
15
,
917
929
.
Dos Santos
,
A. B.
,
Cervantes
,
F. J.
&
Van Lier
,
J. B.
2007
Review paper on current technologies for decolourisation of textile wastewaters: perspectives for anaerobic biotechnology
.
Bioresource Technology
98
(
12
),
2369
2385
.
Ferraz
,
A. D. N.
,
Kato
,
M. T.
,
Florencio
,
L.
&
Gavazza
,
S.
2011
Textile effluent treatment in a UASB reactor followed by submerged aerated biofiltration
.
Water Science and Technology
64
(
8
),
1581
1589
.
Firmino
,
P. M.
,
Da Silva
,
M. R.
,
Cervantes
,
F. J.
&
Dos Santos
,
A. B.
2010
Color removal from synthetic and real textile wastewater in one-and two-stage anaerobic system
.
Bioresource Technology
101
(
20
),
7773
7779
.
McCarty
,
P. L.
1964
Anaerobic waste treatment fundamentals. Part Three, toxic materials and their control
.
Public Works
95
,
91
94
.
O'Neill
,
C.
,
Hawkes
,
F. R.
,
Hawkes
,
D. L.
,
Lourenço
,
N. D.
,
Pinheiro
,
H. M.
&
Deleé
,
W.
1999
Colour in textile effluents-sources, measurement, discharge consents and simulation: a review
.
Journal of Chemical Technology and Biotechnology
74
(
11
),
1009
1018
.
Ozdemir
,
S.
,
Ciriki
,
K.
,
Akman
,
D.
,
Sahinkaya
,
A.
&
Cinar
,
O.
2013
Treatment of azo dye-containing synthetic textile dye effluent using sulfidogenic anaerobic baffled reactor
.
Bioresource Technology
146
,
135
143
.
Pandey
,
A.
,
Singh
,
P.
&
Iyengar
,
L.
2007
Bacterial decolorization and degradation of azo dyes
.
International Biodeterioration & Biodegradation
59
(
2
),
73
84
.
Prato-Garcia
,
D.
,
Cervantes
,
F. J.
&
Buitrón
,
G.
2013
Azo dye decolorization assisted by chemical and biogenic sulfide
.
Journal of Hazardous Materials
250
(
15
),
462
468
.
Rabaey
,
K.
,
Angenent
,
L.
,
Schröder
,
U.
&
Keller
,
J.
2010
Bioelectrochemical Systems: From Extracellular Electron Transfer to Biotechnological Application
.
IWA Publishing
,
London
.
Senthilkumar
,
M.
,
Gnanapragasam
,
G.
,
Arutchelvana
,
V.
&
Nagarajan
,
S.
2011
Treatment of textile dyeing wastewater using two-phase pilot plant UASB reactor with sago wastewater as co-substrate
.
Chemical Engineering Journal
166
,
10
14
.
Singh
,
R. L.
,
Singh
,
P. K.
&
Singh
,
P. S.
2015
Enzymatic decolorization and degradation of azo dyes – a review
.
International Biodeterioration & Biodegradation
104
,
21
31
.
Stolz
,
A.
2001
Basic and applied aspects in the microbial degradation of azo dyes
.
Applied Microbiology and Biotechnology
56
(
1–2
),
69
80
.
Tan
,
N. C. G.
,
Borger
,
A.
,
Slender
,
P.
,
Svitelskaya
,
A. V.
,
Lettinga
,
G.
&
Field
,
J. A.
2000
Degradation of azo dye Mordant Yellow 10 in a sequential anaerobic and aerobic reactor
.
Water Science and Technology
42
(
5–6
),
337
344
.
van der Zee
,
F. P.
,
Bisschops
,
I. A. E.
,
Blanchard
,
V. G.
,
Bouwman
,
R. H. M.
,
Lettinga
,
G.
&
Field
,
J. A.
2003
The contribution of biotic and abiotic processes during azo dye reduction in anaerobic sludge
.
Water Research
37
(
13
),
3098
3109
.