See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/321626618

Evaluation of anionic surfactant removal by anaerobic degradation of

commercial laundry wastewater and domestic sewage

Article  in  Environmental Technology · December 2017

DOI: 10.1080/09593330.2017.1414317

CITATIONS READS

7 113

4 authors:

Clara Faria Tiago Delforno

Federal University of Minas Gerais University of Campinas
6 PUBLICATIONS   18 CITATIONS    110 PUBLICATIONS   339 CITATIONS   

SEE PROFILE SEE PROFILE

Dagoberto Yukio Okada Maria Bernadete A. Varesche

University of Campinas 38 PUBLICATIONS   71 CITATIONS   
159 PUBLICATIONS   398 CITATIONS   
SEE PROFILE
SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Remoção simultânea de matéria orgânica e nitrogênio em um reator de leito estruturado submetido à aeração intermitente e recirculação do efluente: efeito do
diâmetro do material suporte e da razão de recirculação do efluente View project

All content following this page was uploaded by Tiago Delforno on 21 August 2018.

The user has requested enhancement of the downloaded file.

 Environmental Technology

ISSN: 0959-3330 (Print) 1479-487X (Online) Journal homepage: http://www.tandfonline.com/loi/tent20

Evaluation of anionic surfactant removal by

anaerobic degradation of commercial laundry
wastewater and domestic sewage

Clara Vieira de Faria, Tiago Palladino Delforno, Dagoberto Yukio Okada &
Maria Bernadete Amâncio Varesche

To cite this article: Clara Vieira de Faria, Tiago Palladino Delforno, Dagoberto Yukio Okada &
Maria Bernadete Amâncio Varesche (2017): Evaluation of anionic surfactant removal by anaerobic
degradation of commercial laundry wastewater and domestic sewage, Environmental Technology,
DOI: 10.1080/09593330.2017.1414317

To link to this article: https://doi.org/10.1080/09593330.2017.1414317

View supplementary material

Accepted author version posted online: 06

Dec 2017.
Published online: 13 Dec 2017.

Submit your article to this journal

Article views: 3

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=tent20

Download by: [Irstea] Date: 13 December 2017, At: 02:03

 ENVIRONMENTAL TECHNOLOGY, 2017
https://doi.org/10.1080/09593330.2017.1414317

Evaluation of anionic surfactant removal by anaerobic degradation of commercial

laundry wastewater and domestic sewage
Clara Vieira de Fariaa, Tiago Palladino Delforno b
, Dagoberto Yukio Okada c
and Maria Bernadete
Amâncio Vareschea
a
Laboratory of Biological Processes, Department of Hydraulics and Sanitation, Engineering School of São Carlos - University of São Paulo (EESC -
USP) Campus II, São Carlos, Brazil; bMicrobial Resources Division, Research Center for Chemistry, Biology and Agriculture (CPQBA), Campinas
University - UNICAMP, Campinas, Brazil; cUniversity of Campinas (UNICAMP), School of Technology, Division of Technology in Environment
Sanitation, Limeira, Brazil

ABSTRACT ARTICLE HISTORY

An expanded granular sludge bed reactor was evaluated for the anaerobic digestion of commercial Received 3 March 2017
laundry wastewater and domestic sewage focused on the removal of linear alkylbenzene sulfonate Accepted 1 December 2017
(LAS). The reactor was operated in three stages, all under mesophilic conditions and with a hydraulic
KEYWORDS
retention time of 36 h. At stage I, the laundry wastewater was diluted with tap water (influent: 15.3 Expanded granular sludge
± 4.9 mg LAS/L); at stage II, 50% of the feed volume was domestic sewage and 50% was a mixture of bed; linear alkylbenzene
tap water and laundry wastewater (influent: 15.8 ± 4.9 mg LAS/L); and at stage III, only domestic
Downloaded by [Irstea] at 02:03 13 December 2017

sulfonate; 16S rRNA gene;

sewage was used as a diluent of the laundry wastewater (influent: 24.1 ± 4.1 mg LAS/L). Due to nonparametric statistical
the addition of domestic sewage the organic compounds content and LAS in the influent tests; domestic sewage
increased. Under such conditions, it was observed that LAS removal rate decreased from 77.2 ±
14.9% (stage I) to 55.3 ± 18.4% (stage III). Statistical tests indicated that the decrease of the LAS
removal rate was significant and indicated a correlation between the removal of LAS and specific
organic loading rate. The analysis of 16S rRNA gene sequencing revealed genera similar
to Geobacter, Desulfovibrio, Syntrophomonas, Syntrophus, Desulfobulbus, Desulfomonile, and
Desulfomicrobium, which were related to the degradation of LAS.

1. Introduction The removal of LAS from laundry wastewater has

Since 1970, much attention has been given to monitor-
ing contaminants in the environment. These contami-
nants comprise several classes of drugs (analgesics,
antibiotics, and anti-inflammatory), synthetic hormones,
substances used in household cleaning and personal
hygiene products, and compounds applied in the pro-
duction of resins and plastics [1].
Among the substances used in cleaning and
personal hygiene products, anionic surfactants are
noteworthy due to their high utilization. The extensive
use of linear alkylbenzene sulfonate (LAS) in the
formulation of household and industrial detergents
makes this contaminant dominant in the wastewater.
Although the levels of LAS in domestic sewage are
not extremely high (1–18 mg/L) [2], effluents from
commercial laundry and industries may contain con-
centrations greater than 300 mg/L. Braga and Varesche
[3] analyzed samples of commercial laundry effluents
and reported levels ranging from 12.2 to 1023.7 mg
LAS/L.
been studied using anaerobic reactors such as the
expanded granular sludge bed (EGSB) [4], upflow anaero-
bic sludge blanket (UASB) [5], and fluidized bed [6,7]
reactors. In these works, LAS applications ranged from
9.5 to 20 mg/L, since LAS level above 50 mg/L is detri-
mental to the anaerobic process [8].
In the EGSB reactor, the high upflow velocity
improves the sludge–substrate contact, the dilution of
the influent with the recycling may even allow the
treatment of toxic compounds [9]. LAS removals
between 78% and 93% were observed in the EGSB
reactor with 11.5 and 12.0 mg LAS/L [4,10]. Moreover,
Okada et al. [11] reported a lower concentration of vola-
tile fatty acids (VFAs) in the effluent of an EGSB reactor
than in a UASB reactor, which favors the removal of
LAS [12].
Since high LAS levels could be harmful to the anaero-
bic processes, the laundry wastewater must be diluted to
decrease the surfactant level and most of the previous
studies that evaluated the removal of LAS of the

CONTACT Maria Bernadete Amâncio Varesche varesche@sc.usp.br Laboratory of Biological Processes, Department of Hydraulics and Sanitation,
Engineering School of São Carlos - University of São Paulo (EESC - USP) Campus II, São Carlos, SP CEP 13563-120, Brazil
Supplemental data for this article can be accessed at doi:10.1080/09593330.2017.1414317
© 2017 Informa UK Limited, trading as Taylor & Francis Group
 2 C. V. FARIA ET AL.
laundry wastewater used tap water for the dilution.
However, a suitable alternative is to dilute it with dom-
estic sewage, which is an easily obtained wastewater.
Furthermore, the addition of metabolic co-substrates
and buffering agents in the influent could be avoided
with diluted domestic sewage. Nevertheless, the
addition of domestic sewage may result in an increase
in the organic load in the reactor, which can impair the
removal of LAS. Okada et al. [13] reported the increase
rate of LAS removal, from 37% to 76%, when the specific
organic loading rate (SOLR) decreased from 0.18 to
0.03 g chemical oxygen demand (COD)/g total volatile
solids (TVS).d, in UASB reactors with a hydraulic retention
time (HRT) of 35 h.
Thus, this study evaluated the feasibility of LAS
removal in the treatment of laundry wastewater diluted
with domestic sewage, using an EGSB reactor. Domestic
sewage was gradually added to the reactor feed until the
laundry wastewater was diluted only with sanitary
Downloaded by [Irstea] at 02:03 13 December 2017
sewage (stage III). Additionally, samples from the
microbial community at the end of each stage were ana-
lyzed by high-throughput sequencing, using the Illumina
Platform.
2. Materials and methods
2.1. EGSB reactor
The EGSB reactor was made of acrylic with working
volume of 1.80 L, height of 1.0 m, six sampling points,
and an internal diameter of 0.04 m. The operation
was carried out at an HRT of 36 h and at mesophilic
conditions (30°C). The choice of the HRT of 36 h was
based on preliminary studies where it was observed
that the removal of LAS is superior when an HRT of
32 h is applied to the EGSB reactor than when an
HRT of 26 h is applied [14]. The upflow velocity was 4
m/h [14], and to maintain these conditions the feed
flow was 8.55 L/h and the recirculation flow was
8.50 L/h.
A siphon and a water seal were installed in the upper
part of the reactor to ensure anaerobic conditions (Sup-
plementary Figure 1).
2.2. Operation stages
The operation was divided into three stages. The identi-
cal operating conditions were kept at all stages, each one
with a different feed composition. At stage I, the laundry
wastewater was diluted with tap water and the concen-
tration of LAS influent was 15.3 ± 4.9 mg LAS/L; at stage
II, 50% of the volume of the feed was domestic sewage
and 50% was a mixture of tap water and laundry
Table 1. Operation stages of the EGSB reactor.
Stages Feed condition
Stage I Laundry wastewater, tap water, and
(15.3 ± 4.9 mg/L of LAS) sodium bicarbonate (400 mg L−1)
Stage II 50% of domestic sewage and 50% of
(15.8 ± 4.9 mg/L of LAS) laundry wastewater diluted in tap water
Stage III Laundry wastewater diluted with domestic
(24.1 ± 4.1 mg/L of LAS) sewage
wastewater, and the concentration of LAS influent was
15.8 ± 4.9 mg LAS/L; and at stage III, only domestic
sewage was used as a diluent of the laundry wastewater
and the concentration of LAS influent was 24.1 ± 4.1 mg
LAS/L (Table 1).
The wastewater from the first rinse was collected
periodically from a commercial laundry in the city of
São Carlos (SP, Brazil). This source of wastewater was
used in previous studies [4,5,10]. The laundry waste-
water used in our study presented a LAS level
ranging from 94.44 to 919.93 mg/L, with a mean con-
centration of 324.22 ± 325.14 mg/L. The cleaning pro-
ducts employed in the laundry consisted of
formulations with LAS, nonionic surfactants, alkalizing
agents and a neutralizer–acidulant (composed of
sodium sulfate, sodium metabisulfite, and an alkaline
vehicle). Because of this formulation, the laundry
wastewater is alkaline (pH was 9.1 ± 1.2 and the total
alkalinity was 359.5 ± 195.3 mg CaCO3/L) and sulfate
concentration was 762.9 ± 531.2 mg/L (Supplementary
Table 1).
The domestic sewage used to dilute the laundry
wastewater presented COD concentration of 471 ±
173 mg/L and LAS level of 8.72 ± 1.38 mg/L.
As already mentioned, in this study, the concen-
trations of LAS applied in the reactor were maintained
to reach a maximum of 20 mg/L, so the dilution factors
of the laundry wastewater had an average from 1:5 to
1:45.
2.3. Inoculation
Inoculation was carried out with granular sludge from a
UASB reactor used in the treatment of wastewater from
a poultry slaughterhouse (Avícola Dacar S/A, Tietê/SP,
Brazil). A total of 200 mL of sludge were added in the
EGSB reactor, resulting in a solids level of 5.79 g/L TS
(Total solids) and 4.52 g/L TVS.
After inoculation, in order to activate the microorgan-
isms, the reactor was fed with mineral medium for
approximately 40 days [15], consisting of a MgCl2.6H2O
concentration of 36 mg/L, vitamins solution [16],
sodium bicarbonate (400 mg/L), and co-substrates
(270 mg/L of yeast extract, 0.23 mL/L methanol, and
0.16 mL/L ethanol).

2.4. Physical–chemical and chromatographic
analysis
The laundry wastewater, influent, and effluent samples of
the EGSB reactor were analyzed periodically regarding
the COD, pH, volatile suspended solids, and dissolved
total sulfide according to the Standard Methods for
Examination of Water and Wastewater [17].
The alkalinity was determined by the titration method
of Dillalo and Albertson [18] modified by Ripley [19]. The
VFAs were quantified by high-performance liquid chrom-
atography (HPLC) using a Shimadzu system (controller
SCL10AVP, LC-10ADVP pump, CTO-20A Oven, and UV
detector SCL10AVP) with an Aminex HPX-87H column
(Biorad) [20]. The LAS was quantified by HPLC on a Shi-
madzu system (SCL10AVP, LC-10ADVP, CTO-10A, and
RF-10AXL) with a reverse phase C8 column (Supelco)
and a fluorescence detector [21].
The mass balance of LAS was determined by the sur-
factant levels in the influent and effluent, and adsorbed
Downloaded by [Irstea] at 02:03 13 December 2017
in the biomass of the reactor. The extraction of adsorbed
LAS in the sludge granules was performed in accordance
with the method described by Duarte et al. [22], where
the LAS was extracted from the biomass with methanol
in an ultrasound bath for 30 min. Then, the methanol
samples were filtered in ionic exchange columns (SAX)
and C-18 and analyzed by HPLC [21].
2.5. Nonparametric statistical tests
Differences between the concentrations of LAS effluent
at the three stages were tested using Kruskal–Wallis non-
parametric statistical tests followed by the Multiple Com-
parisons by the level of significance (α) of 5%. The
correlation between the concentration of LAS in the
effluent and other physical–chemical parameters was
analyzed by determining the Spearman correlation coef-
ficient. All statistical analyses were performed using Sta-
tistica 10.0 software.
2.6. Massive sequencing of the 16s rRNA gene
At the end of each stage, samples were collected for the
16S rRNA gene sequencing. The analyses were carried
out at the Analytical Center located at the Institute of
Chemistry of the State University of Campinas in Campi-
nas, SP. The DNA was extracted according to the method
described by Griffiths et al. [23]. The integrity of the DNA
was verified by electrophoresis on 0.8% agarose gel fol-
lowed by purification using Illustra GFX PCR DNA and
Gel Band Purification Kit (GE Healthcare). Quantification
and analysis of purity were determined using a Nano-
Drop 2000 spectrophotometer. The extracted DNA
ENVIRONMENTAL TECHNOLOGY 3
purified with a minimum concentration of 50 ng/μL
and 260/280 ratio 1.8–2.0 were sent to the Animal Bio-
technology Laboratory, Department of Animal Science
(ESALQ/USP Piracicaba, Brazil), for the sequencing on Illu-
mina MiSeq platform (2 × 250 bp).
The primers used were CCTACGGGNGGCWGCAG 16S
(S-D-Bact-0341-b-S-17) and GACTACHVGGGTATCTAATCC
(S-D-Bact-0785-a-A-21). For the preprocessing and
assembly, the CASAVA 1.8.2 software was used, provided
by Illumina. This software manages the base call on the
raw data and turns them into reads in fastq format
accompanied by phred quality scores.
The filtering of reads was performed using the
Seqclean software (https://bitbucket.org/izhbannikov/
seqyclean). The minimum quality was 24 (QPhred) and
the minimum length of sequences was 65 bp. Removal
of vectors was performed using the contaminants base
of Univec (http://www.ncbi.nlm.nih.gov/VecScreen/
UniVec.html).
The alignment of the sequences was performed
according to Pynast [24]. Operational taxonomic units
(OTUs) were determined at the similarity level of 97% [25].
The Ribosomal Database Project (RDP) was used for
the taxonomic classification of representative sequences
of each OTU. The threshold confidence applied was 50%.
Alfa (Chao1 and Shannon) and Beta (Jaccard similarity
coefficient) diversity were quantified using Past software
[26]. The phylogenetic tree, using representative
sequences of predominant OTUs, was carried out using
tools from the RDP [27] (such as Tree Builder: http://
www.rdp.cme.msu.edu/) based on the Kimura 2p
model and neighbor-joining method. Moreover, the
circos diagram [28] was used to compare the genus
found in the samples.
The sequences were submitted to the European
Nucleotide Archive (http://www.ebi.ac.uk) under the
accession number ERS1294201 (Stage I), ERS1294202
(Stage II), and ERS1294203 (Stage III). The project acces-
sion number is PRJEB15156.
3. Results and discussion
3.1. EGSB reactor operation
The LAS removal rate decreased as a portion of domestic
sewage used for dilution increased (Table 2). Two
hypotheses have been raised to explain this: because
of the SOLR growth due to the diluted domestic
sewage, or due to the increasing concentration of LAS
in the influent, particularly at stage III with a concen-
tration of 24.1 ± 4.1 mg LAS/L (Table 2).
At stage I, the SOLR and the surfactant removal rate
were 28 mg COD/g TVS.d and 77%, respectively;
 4 C. V. FARIA ET AL.

Table 2. Physical–chemical parameters of EGSB reactor at Table 3. Mass balance of LAS.

different operation. mg of LAS %
Parameters Stage I Stage II Stage III Added mass 5555.75 –
COD Mass in the content of the effluent 1747.16 31
Influent (mg/L) 176 ± 59 175 ± 65 324 ± 47 Removed mass 3808.60 69
Effluent (mg/L) 68 ± 17 70 ± 40 113 ± 32 Adsorbed mass 112.21 2
Removal (%) 57 ± 16 60 ± 26 65 ± 10 Degraded mass 3741.17 67
Specific organic load rate (mg 28 ± 9 26 ± 11 51 ± 9
COD/g TVS.d)
LAS
Influent (mg/L) 15.3 ± 4,9 15.8 ± 4.9 24.1 ± 4.1 (approximately 24–29 mg/L) and HRT (35–39 h). Conver-
Effluent (mg/L) 3.2 ± 1.7 4.0 ± 2.3 11.1 ± 5.5 sely, the SOLR at stage III was higher than that reported
Removal (%) 77.2 ± 14.9 74.4 ± 13.4 55.3 ± 18.4
Specific load (mg/g TVS.d) 2.4 ± 1.1 2.6 ± 1.0 3.5 ± 0.7 by Delforno et al. [10]: in our work, the SOLR was 51 mg
Specific removal (mg/g TVS.d) 1.9 ± 1.2 1.9 ± 0.8 1.8 ± 0.8 COD/g TVS.d while the aforementioned study reported
Partial Alkalinity (mg CaCO3/L)
Influent 218 ± 62 82 ± 19 130 ± 16 21 mg COD/g TVS.d. Therefore, it can be inferred that the
Effluent 251 ± 80 132 ± 23 174 ± 35 dilution of laundry wastewater with domestic sewage did
Total Alkalinity (mg CaCO3/L)
Influent 290 ± 81 129 ± 27 205 ± 23
not particularly damage the removal of surfactant, even
Effluent 331 ± 99 175 ± 29 231 ± 41 at higher SOLR.
pH Conversely, the LAS removal rate decreased slightly
Influent 8.0 ± 0.2 7.4 ± 0.4 7.4 ± 0.1
Effluent 8.0 ± 0.2 7.9 ± 0.3 7.9 ± 0.1 from stage I to II (Table 2). Despite the dilution of the
VFA (mg HAc/L) laundry wastewater with domestic sewage at a pro-
Influent 42.6 ± 47.7 41.1 ± 14.5 82.3 ± 20.0
portion of 50%, the SOLR and LAS level in the influent
Downloaded by [Irstea] at 02:03 13 December 2017

Effluent 49.0 ± 93.1 25.7 ± 9.3 49.8 ± 14.1

Biomass (g/L) Initial
Biomass
TS 5.79
Total fixed solids (TFS) 1.27 – 2.72
TVS 4.52
Operation time (days) 84 72
whereas at stage III, 51 mg COD/g TVS.d and 55% was
observed, respectively (Table 2). Duarte et al. [29] and
Okada et al. [13] also observed the relationship
between SOLR and LAS removal rate. Okada et al. [13]
reported that in an UASB reactor, a lower SOLR (0.03 g
COD/g TVS.d) led to a greater removal of LAS (76%).
Duarte et al. [29] observed that an increase from 183 to
771 mg COD/L resulted in a decreasing LAS removal
rate from 37% to 24%, in an anaerobic sequencing
batch reactor. According to the authors, microorganisms
probably preferred metabolizing the co-substrates (yeast
extract, starch, and sucrose) than the LAS, a xenobiotic
compound.
At stage III, the LAS concentration in the influent
increased from approximately 15.3 ± 4.9 to 24.1 ± 4.1 mg/
L, which led to a decrease in the LAS removal rate from
77.2 ± 14.9% to 55.3 ± 18.4% (Table 2). This increment
was due to the LAS level in the domestic sewage (8.72 ±
1.38 mg/L) which was the sole diluent of laundry waste-
water at stage III. Moreover, Delforno et al. [10] reported
a decrease in the LAS removal rate from 92.9 ± 10.3% to
58.6 ± 25.8% when the LAS influent level increased from
12.0 ± 3.2 to 28.8 ± 6.4 mg/L, in an EGSB reactor treating
laundry wastewater diluted with tap water. In fact, the
LAS removal rate at stage III (55.3%) was similar to the
rate (58.6%) observed by Delforno et al. [10] which was
under similar conditions as the LAS influent concentration
Final were approximately 30 mg COD/g TVS.d and 15 mg
Biomass
9.50 LAS/L, respectively. Because of the aforementioned par-
ameters, the conditions at stages I and II were very
6.77
84
similar, which led to slight differences in the LAS and
COD removal rates (Table 2).
Low concentrations of VFA were observed in the effluent.
The mean concentration of VFA was below 50 mg HAc/L
at all stages (Table 2). The major acids observed were
valeric acid (8.5 mg/L) at stage I and caproic acid at stages
II and III (7.9 and 11.6 mg/L, respectively).
According to the TVS concentration, the biomass
increased by approximately 50%. The TVS concentration
increased from 4.52 to 6.77 g/L.
According to the mass balance of LAS, 2% of added
surfactant was adsorbed into the biomass and 67%
was removed by degradation (Table 3).
3.2. Nonparametric statistical tests
The Kruskal–Wallis test was applied to comparatively
examine LAS concentrations in the effluents from the
three stages. The nonparametric tests of multiple compari-
sons indicated similarities for the median concentrations of
LAS in the effluent at stages I and II, while at stage III, the
median concentrations differed from the others stages, at
a significance level of 5%. In fact, the LAS removal rate pre-
sented a significant decrease at stage III, while the rate
slightly decreased from stage I to II (Figure 1).
Analyzing the parameters related to the LAS removal,
the highest Spearman correlation coefficient was −0.53,
between the LAS removal rate and the SOLR (Figure 2
(a)). An inverse correlation between the LAS removal
and SOLR was observed, indicating that higher LAS

Figure 1. Box plot of LAS removal rate at the three operation
stages.
removal rates occurred at lower SOLR. Consequently,
lower levels of LAS in the effluent were found at lower
SOLR (Figure 2(b)), presenting a Spearman correlation
coefficient of 0.66.
3.3. Molecular biology analysis
The good estimator was above 97%, indicating a suit-
able coverage by the sequencing of all samples (Table
4). The average length of the sequences was 385–410 bp.
Downloaded by [Irstea] at 02:03 13 December 2017
Figure 2. LAS removal rate and SOLR (A); and LAS concentration in the effluent and SOLR (B) observed in all operation stages of the
EGSB reactor, with the Spearman correlation coefficient.
ENVIRONMENTAL TECHNOLOGY 5
The diversity and richness of the microbial commu-
nity in the EGSB reactor increased with the addition
of domestic sewage. The highest richness indices
were observed at stage III (laundry wastewater only
diluted with domestic sewage): Chao1 index was
6800 ± 184, and the number of OTUS was 3903. There
was a slight difference between the richness index of
the samples of stages I and II (Table 4), indicating
that the addition of domestic sewage at the proportion
of 50% resulted in a small shift in the microbial commu-
nity of the EGSB reactor. In fact, the Jaccard index
between the samples from stages I and II (56%) was
higher than between the samples from stages I and III
(32%), indicating a greater shift between samples
from stages I and III (Supplementary Figure 2). The rich-
ness and similarity indices indicated a remarkable
change in the microbial community at stage III, which
used only domestic sewage to dilute the laundry waste-
water. The main modifications at stage III were the
 6 C. V. FARIA ET AL.

Table 4. Results of 16S rRNA sequencing. the dilution of laundry wastewater with domestic
Parameters Stage I Stage II Stage III sewage at the proportion of 50% resulted in a smaller
16S rRNA data shift of the microbial community (similarity and richness
Sequences Count 27,344 47,682 91,041
OTU count 1,327 1,757 3,903
indices) in addition to the insignificant difference
Singletons 642 763 1829 between the LAS removal rates at stages I and II. The
Mean Sequence Length (bp) 410 ± 89 385 ± 96 390 ± 95 SOLR and LAS level in the influent with very similar
Richness Estimation
Chao1 2,477 ± 110 2,994 ± 121 6,800 ± 184 values at stages I and II were related to the smaller
Rarefaction 1,327 ± 1 1,757 ± 3 3,903 ± 4 shift in the microbial community, which most likely led
Diversity index
Shannon (H) 4.6 ± 0.03 4.6 ± 0.03 5.3 ± 0.02 to the obtained surfactant removal rate. At stage III, the
Coverage higher SOLR and LAS level resulted in a greater shift
Good’s estimator 97.7% 98.4% 98.0%
in the microbial community, which led to a lower surfac-
tant removal rate. Therefore, the SOLR and LAS concen-
substrate concentrations due to the addition of sewage, tration in the influent are most likely related to the
and the higher level of LAS. It is likely that the main microbial community and, consequently, to the surfac-
feature related to the shift of the microbial community tant removal.
at stage III was the substrate concentrations provided From the 14 bacterial phyla identified, 5 phyla pre-
by the domestic sewage. Delforno et al. [10] reported sented the highest relative abundances: Synergistetes
the reduction of richness due to an increase in LAS con- (21–32%), Proteobacteria (17–24%), Firmicutes (6–15%),
centration in the influent. Consequently, the increasing Chloroflexi (7–12%), and Bacteroidetes (11–18%)
Downloaded by [Irstea] at 02:03 13 December 2017

LAS concentration in the influent could not be related (Figure 3). Microorganisms affiliated to the phyla Actino-
to the richness indices observed in our study. Conversely, bacteria, Chloroflexi, Bacteroidetes, Firmicutes,

Figure 3. Phyla of bacteria observed at stages I, II, and III.

Figure 4. Phylogenetic tree of two main OTUs found in the EGSB reactor. Bootstrap values (100 replicate runs, shown as %) greater than
50% are listed. GenBank accession numbers are listed after species identification. Aeromonas punctata (X74674.1) was used as the
outgroup.

Proteobacteria, and Synergistetes were identified in the
anaerobic reactors used for the removal of LAS [4,6,7,
10,14,22,30,31]. Members of the phylum Synergistetes
were characterized as anaerobic mesophilic, which can
use amino acids, provide short fatty acids to methano-
genic archaea and provide sulfate to sulfate-reducing
bacteria [32]. The remarkable sulfate concentrations in
laundry wastewater probably contributed to maintaining
the bacteria related to the reduction of this anion.
Members of the phylum Chloroflexi, in particular
Downloaded by [Irstea] at 02:03 13 December 2017
Figure 5. Circos ideogram of the genera found in the three operation stages. The highlighted blue genera are related to LAS or degra-
dation of aromatic compounds.
ENVIRONMENTAL TECHNOLOGY 7
Chloroflexi subphylum I, were related to the granule for-
mation in the UASB reactor [33,34].
Bacteria assigned to genus E6 (phylum Synergistetes)
were identified at all stages of the reactor operation with
relative abundance between 9% and 16%. At stages II
and III (addition of domestic sewage in the laundry waste-
water), bacteria assigned to Blvii28 (phylum Bacteroidetes)
and vadinCA02 (phylum Synergistetes) genera were
found, whose relative abundance ranged from 8% to
11% and from 5% to 9%, respectively.
 8 C. V. FARIA ET AL.
The OTU CF3_1412 was assigned to the E6 genus
(Figure 4), and it was similar to the bacteria found in
environments with recalcitrant compounds, such as pet-
roleum [35] (accession number: JN698228.1). The relative
abundance of this OTU (CF3_1412) was 10.1%, 11.7%,
and 7.0% in the samples from stages I, II, and III,
respectively.
The OTU CF3_416 was assigned to Blvii28 genus, and it
was similar to bacteria related to the degradation of
chlorinated compounds such as perchloroethylene and
chlorophenol [36] (accession number: AY780554.1),
and chlorinated ethene [37] (accession number:
AY780554.1). The reads affiliated to this OTU presented
a relative abundance of 10% and 7.3% in the samples
from stages II and III, respectively. Duarte et al. [22]
found clones similar to unidentified Eubacterium clone
vadinCA02 in a reactor fed with LAS. The reads affiliated
to Blvii28 and vadinCA02 genera may be related to the fer-
mentative metabolism due to the increase of its relative
Downloaded by [Irstea] at 02:03 13 December 2017
abundance in the presence of domestic sewage (Vadin
is an acronym for Vinasses Anaerobic Digester of Narbonne)
[38]. However, E6 genus can be related to the degradation
of LAS, since some members of the phylum Synergistetes
can carry out the β-oxidation [39,40], one of the LAS
degradation steps [41]. Nonetheless, the remarkable pres-
ence of reads affiliated to E6, Blvii28, and vadinCA02
genera indicate these taxonomies may resist the presence
of LAS molecule, a recalcitrant compound.
Additionally, of the 46 genera identified in the three
stages, 7 (Geobacter, Desulfovibrio, Syntrophomonas, Syn-
trophus, Desulfobulbus, Desulfomonile, and Desulfomicro-
bium) (Figure 5) were related to the degradation of the
LAS and aromatic compounds [4–6,10,14]. Geobacter, Syn-
trophus, Desulfobulbus, Desulfovibrio, Desulfomonile, and
Desulfomicrobium genera were related to the degradation
of aromatic compounds [42]. Furthermore, the genera
Geobacter and Syntrophomonas can carry out the β-oxi-
dation, which is a step in the catabolism of LAS [42,43].
4. Conclusion
In an EGSB reactor, the removal of LAS from laundry
wastewater was not significantly affected when domestic
sewage was used for dilution, provided that the SOLR
applied to the reactor did not exceed values close to
26 ± 11 mg COD/g TVS.d. Additionally, sequencing of
the samples from the EGSB reactor indicated that SOLR
could be related to the structure of the microbial com-
munity and consequently to the LAS removal obtained.
Among the taxonomies identified in samples from
the EGSB reactor, the E6 genus was remarkable in all oper-
ation stages. Moreover, the most found in the stages were
the genera Blvii28 and vadinCA02, with the addition of
domestic sewage. These three genera may be associated
with a greater tolerance in the presence of the surfactant.
Furthermore, the genera related to the degradation of LAS
and aromatic compounds were identified (Geobacter,
Desulfovibrio, Syntrophomonas, Syntrophus, Desulfobulbus,
Desulfomonile, and Desulfomicrobium).
Disclosure statement
No potential conflict of interest was reported by the authors.
Funding
This study was funded by Coordenação de Aperfeiçoamento de
Pessoal de Nível Superior (CAPES).
ORCID
Tiago Palladino Delforno http://orcid.org/0000-0002-1705-
0763
Dagoberto Yukio Okada http://orcid.org/0000-0003-1859-
9851
References
[1] Aquino SF, Brandt EMF, Chernicharo CAL. [Removal of
drugs and endocrine disrupters in sewage treatment
plants: literature review]. Eng Sanit Ambient. 2013;18
(3):187–204. Portuguese.
[2] Mungray AK, Kumar P. Fate of linear alkylbenzene sulfo-
nates in the environment: a review. Int Biodeter Biodegr.
2009;63:981–987.
[3] Braga JK, Varesche MBA. Commercial laundry water
characterisation. Am J Analyt Chem. 2014;5:8–16.
[4] Delforno TP, Moura AGL, Okada DY, et al. Effect of biomass
adaptation to the degradation of anionic surfactants in
laundry wastewater using EGSB reactors. Bioresource
Technol. 2014;154:114–121.
[5] Okada DY, Delforno TP, Etchebehere C, et al. Evaluation of
the microbial community of upflow anaerobic sludge
blanket reactors used for the removal and degradation
of linear alkylbenzene sulfonate by pyrosequencing. Int
Biodeter Biodegr. 2014;96:63–70.
[6] Braga JK, Motteran F, Macedo TZ, et al. Biodegradation of
linear alkylbenzene sulfonate in commercial laundry
wastewater by an anaerobic fluidized bed reactor. J
Environ Sci Heal A. 2015;50:946–957.
[7] Macedo TZ, Okada DY, Delforno TP, et al. The comparative
advantages of ethanol and sucrose as co-substrates in the
degradation of an anionic surfactant: microbial commu-
nity selection. Bioproc Biosyst Eng. 2015;38(10):1835–
1844.
[8] Angelidaki I, Torang L, Waul CM, et al. Anaerobic biopro-
cessing of sewage sludge, focusing on degradation of
linear alkylbenzene sulfonates (LAS). Water Sci Technol.
2004;49(10):115–122.
[9] Seghezzo L, Zeeman G, Van Lier JB, et al. A review: the
anaerobic treatment of sewage in UASB and EGSB reac-
tors. Bioresource Technol. 1998;65:175–190.

[10] Delforno TP, Moura AGL, Okada DY, et al. Microbial diver-
sity and the implications of sulfide levels in an anaerobic
reactor used to remove an anionic surfactant from laundry
wastewater. Bioresource Technol. 2015;192:37–45.
[11] Okada DY, Delforno TP, Esteves AS, et al. Influence of vola-
tile fatty acid concentration stability on anaerobic degra-
dation of linear alkylbenzene sulfonate. J Environ Manage.
2013;128:169–172.
[12] Lobner T, Torang L, Batstone DJ, et al. Effects of process
stability on anaerobic biodegradation of LAS in UASB
reactors. Biotechnol Bioeng. 2005;89:759–765.
[13] Okada DY, Delforno TP, Esteves AS, et al. Optimization of
linear alkylbenzene sulfonate (LAS) degradation in UASB
reactors by varying bioavailability of LAS, hydraulic reten-
tion time and specific organic load rate. Bioresource
Technol. 2013;128:125–133.
[14] Delforno TP, Okada DY, Polizel J, et al. Microbial character-
ization and removal of anionic surfactant in an expanded
granular sludge bed reactor. Bioresource Technol.
2012;107:103–109.
[15] Angelidaki I, Petersen SP, Ahring BK. Effects of lipids on
thermophilic anaerobic-digestion and reduction of lipid
Downloaded by [Irstea] at 02:03 13 December 2017
inhibition upon addition of bentonite. Appl Microbiol
Biot. 1990;33(4):469–472.
[16] Touzel JP, Albagnac G. Isolation and characterization of
Methanococcus mazei strain MC3. Fems Microbiol Lett.
1983;16:241–245.
[17] APHA-AWWA-WPCF. Standard methods for the examin-
ation of water and wastewater. 20th ed.Washington (DC):
American Public Health Association/American Water
Works Association/Water Environment Federation; 2005.
[18] Dillalo R, Albertson OE. Volatile acids by direct tritation. J
Water Pollut Con F. 1961;33:356–365.
[19] Ripley LE, Boyle WC, Converse LC. Improved alkalimetric
monitoring for anaerobic digestion of high-strength
wastes. Water Pollut Control. 1986;58:106–111.
[20] Penteado ED, Lazaro CZ, Sakamoto IK, et al. Influence of
seed sludge and pretreatment method on hydrogen pro-
duction in packed-bed anaerobic reactors. Int J Hydrogen
Energ. 2013;38:6137–6145.
[21] Duarte ICS, Oliveira LL, Buzzini AP, et al. Development of a
method by HPLC to determine las and its application
in anaerobic reactors. J Brazil Chem Soc. 2006;17
(7):1360–1367.
[22] Duarte ICS, Oliveira LL, Saavedra NK, et al. Evaluation of
the microbial diversity in a horizontal-flow anaerobic
immobilized biomass reactor treating linear alkylbenzene
sulfonate. Biodegradation. 2008;19:357–385.
[23] Griffiths RI, Whiteley AS, O’Donnell AG, et al. Rapid
method for coextraction of DNA and RNA from natural
environments for analysis of ribosomal DNA- and rRNA-
based microbial community composition. Appl Environ
Microb. 2000;66(12):5488–5491.
[24] Caporaso JG, Bittinger KD, Bushman F, et al. PyNAST: a
flexible tool for aligning sequences to a template align-
ment. Bioinformatics. 2010;26(2):266–267.
[25] Edgar RC. Search and clustering orders of magnitude
faster than BLAST. Bioinformatics. 2010;26(19):2460–2461.
[26] Hammer Ø, Harper DAT, Ryan PD. PAST: paleontological
statistics software package for education and data analy-
sis. Palaeontol Electron. 2001;4:1–9.
View publication stats
ENVIRONMENTAL TECHNOLOGY 9
[27] Cole JR, Wang Q, Cardenas E, et al. The ribosomal data-
base project: improved alignments and new tools for
rRNA analysis. Nucleic Acids Res. 2009;37:D141–D145.
[28] Krzywinski M, Schein J, Birol I, et al. Circos: an information
aesthetic for comparative genomics. Genome Res.
2009;19:1639–1645.
[29] Duarte ICS, Oliveira LL, Okada DY, et al. Degradation
of detergent (linear alkylbenzene sulfonate) in an anaero-
bic stirred sequencing-batch reactor containing granular
biomass. Int Biodeter Biodegr. 2010;64(2):129–134.
[30] Oliveira LL, Costa RB, Duarte ICS, et al. Anaerobic degra-
dation of linear alkylbenzene sulfonate in fluidized bed
reactor. Braz J Chem Eng. 2010;27(4):539–543.
[31] Carosia MF, Okada DY, Sakamoto IK, et al. Microbial
characterization and degradation of linear alkylbenzene
sulfonate in an anaerobic reactor treating wastewater
containing soap powder. Bioresource Technol.
2014;167:316–323.
[32] Vartoukian SR, Palmer RM, Wade WG. The division ‘syner-
gistes’. Anaerobe. 2007;13:99–106.
[33] Bjornsson L, Hugenholtz P, Tyson GW, et al.
Filamentous Chloroflexi (green non-sulfur bacteria) are
abundant in wastewater treatment process with
biological nutrient removal. Microbiology. 2002;148:
2309–2318.
[34] Yamada T, Sekiguchi Y, Imachi H, et al. Diversity, localiz-
ation and physiological properties of filamentous
microbes belongs to Chloroflexi subphylum I in mesophilic
and thermophilic methanogenic sludge granules. Appl
Environ Microb. 2005;71:7493–7503.
[35] Scherr KE, Lundaa T, Klose V, et al. Changes in bacterial
communities from anaerobic digesters during petroleum
hydrocarbon degradation. J Biotechnol. 2012;157(4):564–
572.
[36] Li H, Shen TT, Wang XL, et al. Biodegradation of perchlor-
oethylene and chlorophenol co-contamination and toxic
effect on activated sludge performance. Bioresource
Technol. 2013;137:286–293.
[37] Duhamel M, Edwards EA. Microbial composition of chlori-
nated ethene-degrading cultures dominated by
Dehalococcoides. Microb Ecol. 2006;58(3):538–549.
[38] Godon JJ, Zumstein E, Dabert P, et al. Molecular microbial
diversity of an anaerobic digestor as determined by small-
subunit rDNA sequence analysis. Appl Environ Microb.
1997;63(7):2802–2813.
[39] Allison MJ, Mayberry WR, Mcsweeney CS, et al. Synergistes
jonesii, gen. nov., sp.nov.: A rumen bacterium that
degrades toxic Pyridinediols. Syst Appl Microbiol.
1992;15(4):522–529.
[40] Kumar AG, Kirubagaran R, Nagesh N, et al. Influence of fer-
mentation metabolites on redox potential in anaerobic
digestion of proteinaceous solid wastes by Synergistes
sp. Eng Life Sci. 2010;10(4):368–373.
[41] Schörberl P. Basics principles of LAS biodegradation.
Tenside Surfact Det. 1989;26:86–94.
[42] Brenner DJ, Krieg NR, Staley JT. Bergey’s Manual of
Systematic Bacteriology. 2nd ed.New York (NY):
Springer; 2005.
[43] Vos PD, Garrity G, Jones DR, et al. Bergey’s Manual of
Systematic Bacteriology, v. 3 (The Firmicutes). 2nd ed.
New York (NY): Springer; 2009.