Anaerobic degradation of BTEX in a packed-bed reactor

Water Science and Technology Vol 45 No 10 pp 175–180 © IWA Publishing 2002

I.R. de Nardi, M. B. A. Varesche, M. Zaiat, E. Foresti
Laboratório de Processos Biológicos, Departamento de Hidráulica e Saneamento – Escola de Engenharia
de São Carlos – Universidade de São Paulo, Av. Trabalhador São-carlense, 400, CEP: 13566–590, São
Carlos, SP, Brazil

Abstract A bench-scale horizontal-flow anaerobic immobilized biomass (HAIB) reactor was assayed aiming
to verify its performance in degrading benzene, toluene, ethylbenzene and xylene (BTEX). A 138 ml HAIB
reactor filled with polyurethane foam matrices containing immobilized anaerobic biomass was initially fed
with synthetic substrate containing protein, carbohydrates and lipids. Thereafter, BTEX degradation was
evaluated in the presence of the co-solvents ethanol and lineal alkylbenzene sulphonate (LAS), in two
sequential experiments. The inlet BTEX concentration ranged from 1.3 to 27.0 mg/L of each compound and
outlet concentrations were lower than 0.1 mg BTEX/L for both the experiments with ethanol and LAS. An
active enriched microbial consortium was observed in the reactor, containing BTEX-degraders, and also
acetogenic, acetotrophic and hydrogenotrophic microorganisms. The results from the most probable
number (MPN) tests indicated a decrease in the number of methanogenic archae, while the number of
anaerobic microorganisms in the biofilm was maintained during the experimental period. Methanogenic
archae were found to represent less than 0.5 % of the total anaerobic organisms in the biomass inside the
reactor.
Keywords Anaerobic process; BTEX; fixed-bed reactor; HAIB reactor; immobilized biomass; polyurethane
foam

Introduction
Benzene, toluene, ethylbenzene, and xylenes, collectively known as BTEX, are widely
used as industrial solvents for organic synthesis and equipment cleansing. As the major
aromatic components in many petroleum products, they are often found in groundwater due
to leaks in underground storage tanks and pipelines, improper waste disposal practices,
inadvertent spills, and leaching from landfills (Shim and Yang, 1999). According to
Corseuil et al. (1998), migration of BTEX in groundwater is enhanced when ethanol is used
as a gasoline additive at much higher concentration than BTEX, as it happens in Brazil. In
this case, ethanol acts also as a BTEX solvent.
BTEX degradation by bacterial consortia from sewage, indigenous soil, groundwater
microorganisms or pure cultures, either in batch microcosms or in continuous-flow reac-
tors has been investigated. However, the results are not conclusive and have varied with the
experimental conditions (Shim and Yang, 1999).
According to Zaiat et al. (1997), the horizontal-flow anaerobic immobilized biomass
(HAIB) reactor can be applied to treat recalcitrant compounds, since the predominant plug-
flow regime provides suitable conditions for such a specific purpose. Moreover, as the
reactor is filled with polyurethane foam matrices, another important condition is satisfied,
for this support offers excellent conditions for anaerobic biomass growth and retention
(Varesche et al., 1997). It is also confirmed that biofilm-associated cells are more resistant
to many toxic substances (Watrick and Kotter, 2000). Therefore, HAIB reactors attain the
most important conditions for establishing a proper environment for recalcitrant compound
degradation.
This paper reports on the evaluation of BTEX degradation in a HAIB reactor fed with a
synthetic wastewater containing these compounds in the presence of ethanol and linear 175
 alkylbenzene sulphonate (LAS). Both of these compounds can be found in BTEX contain-
ing wastewaters in Brazil.

Methods
The assays were performed at 30 ± 3°C in a 138 ml horizontal-flow anaerobic immobilized
biomass (HAIB) reactor (Figure 1). The reactor was filled with polyurethane foam cubic
matrices (23 kg/m3), previously inoculated with a mixture of sludges taken from up-flow
I.R. de Nardi et al.

anaerobic sludge blanket (UASB) reactors treating recycled paper industry wastewater,
domestic sewage and poultry slaughterhouse wastewater.

(2)

(5)

(4)
(3)
(1)

Figure 1 Scheme of the bench-scale HAIB reactor (1) affluent reservoir, (2) gas bag with N2, (3) peristaltic
pump, (4) reactor, (5) gas outlet

Initially, the reactor was operated for 6 months at the hydraulic detention time (HDT) of
5 hours, fed with a synthetic substrate containing protein (50%), carbohydrates (40%) and
lipids (10%), in order to promote suitable biomass development and adhesion in its packed
bed. After the reactor had achieved stable operating conditions, the HDT was increased to
13.5 hours and BTEX compounds and co-solvents were added to the synthetic substrate. In
a first phase, the reactor was operated during 75 days, and ethanol was used as co-solvent
(848 mg/L) and, consequently, as co-substrate. In the second phase lasting 72 days, ethanol
addition was suppressed and LAS (0.70g of commercial detergent/L) was used to enhance
the BTEX dissolution. Table 1 presents the mean concentrations of BTEX, the mean chem-
ical oxygen demand of non-filtered samples (COD) and the pH of the synthetic wastewater
along the two experimental phases.
COD, pH, bicarbonate alkalinity (BA), total volatile acids (TVA), total solids (TS) and
volatile solids (VS) were monitored according to the Standard Methods for Examination of
Water and Wastewater (1995). The BTEX concentrations were determined using a gas
chromatograph HP 6869 with HP-1 column – 30 m × 0.53 mm (internal diameter) × 2.65
µm (film thickness). The concentrations of specific volatile acids were determined using a
gas chromatograph HP 6869 with HP Innovax column – 30 m × 0.25 mm (internal diame-
ter) × 0.25 µm (film thickness). Microbial morphology characterization was performed
using Olympus microscope BX60-FLA and Zeiss digital scanning microscope DSM-960.
Adsorption assays were carried out based on Enkiri et al (1995). Matrices of
polyurethane foam were taken from the reactor without receiving BTEX. They were dried
at 55.2°C for 21 hours and inactivated by UV radiation for 2 hours. BTEX adsorption pro-
files were obtained from batch assays, in which the inactivated bioparticles were in contact
with the synthetic wastewater containing all the compounds used. The profiles obtained
permitted to determine the maximum adsorption capacity of the bed and the adsorption
time.
The number of microorganisms was estimated in the acclimatization phase and at the
176 end of the second phase using the most probable number (MPN) technique under an N2
atmosphere (100 %), as described by Vazoller (1995). The quantification was performed
by turbidity, methane production and fluorescence microscopy. MAN Table was used to
determine MPN values with confidence band of 95 %.
Table 1 BTEX composition, COD and pH in the synthetic wastewater

Parameter First phase with ethanol Second phase with LAS

Mean concentration Min. Max. Mean concentration Min. Max.

I.R. de Nardi et al.

Benzene (mg/L) 9.7 + 3.2 5.3 13.00 11.8 + 5.4 3.6 27.0
Toluene (mg/L) 7.1 + 1.3 5.9 9.0 8.7 + 4.6 3.5 22.3
Ethylbenzene (mg/L) 5.3 + 0.7 4.6 6.3 7.0 + 4.5 1.3 19.3
(m+p)-xylenes (mg/L) 9.9 + 1.4 8.7 12.00 14.3 + 8.7 2.6 37.7
o-xylene (mg/L) 5.6 + 1.0 4.7 7.0 8.0 + 4.6 1.9 20.5
COD (mg/L) 1968 + 407 1025 2494 797 + 54 719 911
pH 7.1 8.6 6.5 7.6

Results and discussion

Operating data
During the acclimatization phase without BTEX addition, the synthetic wastewater pre-
sented COD of 492 + 77 mg/L, TVA of 47 + 11 mg HAc/L, BA of 119 + 31 mg CaCO3/L
and pH ranging from 6.6 to 7.7. The mean values of the monitoring parameters in the efflu-
ent were: COD of 95 + 18 mg/L, TVA of 18 + 4 mg HAc/L, BA of 174 + 37 mg CaCO3/L
and pH ranging from 7.5 to 8.2.
In the first phase with ethanol as co-solvent, BTEX concentrations in the effluent stream
were lower than 0.1 mg/L throughout the phase. Therefore, almost complete anaerobic
biodegradation was achieved in the packed-bed reactor operated at HDT of 13.5 hours.
During this phase, the main operating problem occurred due to the organic overload
imposed to the reactor in the 34th day of operation. In order to enhance the BTEX dissolu-
tion, the ethanol concentration was increased, thus elevating COD from 492 mg/L to 1968
mg/L. Initially, the pH decreased to 4.9 as a consequence of the high TVA production
resulting the maximum concentrations of 1036 mg/L of acetic acid, 32 mg/L of propionic
acid and 27 mg/L of butyric acid, besides 15 mg/L of valeric, iso-valeric and caproic acids.
Sodium bicarbonate was added to the substrate and the reactor recovered the initial per-
formance even at high influent ethanol concentration after 30 days. Effluent COD dropped
to 131 + 21 mg/L and pH increased to 8.2. It is worth to mention that the BTEX degradation
persisted even during the organic overloading period.
The addition of linear alkylbenzene sulphonate (LAS) in the second experimental phase
enhanced the BTEX dissolution, thus resulting in influent concentrations higher than those
obtained in the first phase (Table 1). After 19 days of operation at HDT of 13.5 hours, the
reactor achieved stable performance with effluent BTEX concentrations lower than 0.3
mg/L, COD of 161 + 4 mg/L, TVA close to 27 mg HAc/L and pH ranging from 7.7 to 8.0.
After stabilization, the reactor was operated for 53 days at HDT of 5.3 hours. Under these
conditions, the effluent COD increased to 242 + 29 mg/L. The pH values ranged from 7.6 to
8.0 and TVA concentration remained as low as 20 mg HAc/L, with 11 + 8 mg/L of acetic
acid and 5 + 3 mg/L of propionic acid. Effluent BTEX concentration increased as compared
with the operation at HDT of 13.5 hours. Even so, the maximum concentration observed
was of 300 µg/L, approximately. Figure 2 summarizes the results obtained during HAIB
reactor operation at HDT of 5.3 hours. TS and VS contents in the polyurethane foam matri-
ces were, respectively, 0.82 g/g foam and 0.46 g/g foam.
The adsorption tests permitted to verify that the adsorption time is almost the same for
all the compounds assayed (results omitted in this text). For benzene, both the time of 177
 adsorption and adsorption capacity were lower than the values obtained for the other com-
pounds. Adsorption capacity varied from 12.81 + 1.21 µmol/g particle (benzene) to 19.10 +
0.80 µmol/g particle (ethylbenzene). The low values obtained for adsorption capacity indi-
cated that biodegradation was the main phenomenon responsible for the good performance
observed in the experiments in HAIB reactor.

(a) (b)

Concentration(og/L)
I.R. de Nardi et al.

Concentration(mg/L)

40 400
30 300
20 200
10 100
0 0
10 20 30 40 50 60 10 20 30 40 50 60
Experimental period (days) Experimental period (days)

Figure 2 BTEX concentration in the influent (a) and effluent (b) of the HAIB reactor at HDT of 5.3 h in the
presence of LAS. Benzene (), Toluene (), Ethylbenzene (), (m+p)-xylenes (x), o-xylene ()

Microbiological characterization of biomass

Table 3 and Figure 3 show the morphologies observed in the bioparticles taken from the
first (FRS) and last reactor segment (LRS). It is observed that acetotrophic archaea
Methanosarcina sp seemed to have been favored by the presence of ethanol in the first
phase, in respect to Methanosaeta sp and other hydrogenotrophic archaea fluorescent rods.
In general, the same active enriched consortium was observed after BTEX supplementa-
tion. So far, no pure culture has been shown the capability to mineralize benzene complete-
ly under anaerobic conditions (Harwood and Gibson, 1997). The presence of
Methanosarcina sp in this phase was due to the high acetate concentration. In the second
experimental phase, methanogenesis was probably inhibited due to the presence of LAS.
This hypothesis has support in the MPN assays, since no methane was produced in the
flasks. In this case, the quantification was accomplished by the visualization of fluorescent
organisms in the flask samples. It has been demonstrated that LAS is not mineralized under

Table 2 Microbial morphologies in the samples of each phase

Morphology Acclimatization First phase Second phase

phase
FRS LRS FRS* LRS FRS LRS

Domain Bacteria
Desulfococcus-like (1.5 – 1.7 µm in diameter) ++ +++ – +++ + –
Chain cocci + + – + – –
Oval to rod-sphaped – – – +++ ++ +++
Rods in pairs – + – + +++ +++
Pointed rod – ++ +++ – ++ –
Straight rod +++ + +++ ++ +++ ++
Thin-filament ++ +++ – + +++ +
Desulfovibrio-like (3.5 – 3.8 µm in length) +++ + + + +++ ++
Desulfvibrio-like (1.7 – 2.0 µm in length) ++ + + + +++ ++
Domain Archaea
Fluorescent rod ++ ++ ++ ++ – +
Methanosaeta sp. +++ ++ – + ++ +
Methanosarcina sp. – – – ++ + +
Cysts ++ – ++ – + –
+++ predominant, ++ often, + rare, – absent. *analysis of the slime formed in the first reactor segment
178 during the first phase.
anaerobic conditions and that methanogenesis is inhibited (Wagener and Schink, 1987).
The prevalence of other microorganisms and sulfate-reducing bacteria was observed along
the reactor.

I.R. de Nardi et al.

a. 2.0µ
µm b. µm
2.0µ c. 1.0 µm

d. 5.0µ
µm e. 2.0µ
µm f. 5.0µm

Figure 3 Scanning electron and phase-contrast photomicrographs of bacteria and methanogenic archaea:
(a) Methanosaeta-like in the first reactor segment during acclimatization phase; (b) cocci in the last reactor
segment during acclimatization phase; (c) rods in the first reactor segment during acclimatization phase; (d)
cocci and Methanosarcina-like in the last reactor segment during the first phase; (e) vibrios in the last reactor
segment during second phase; (f) rods in pairs MPN (10–3) in the first reactor segment during second phase.

The physiological capacity of sulfate-reducing bacterial consortia and even isolate individ-
uals to degrade BTEX-type compounds under mesophilic conditions has been previously
reported (Meckenstock, 1999; Robertson et al., 2001). The presence of different organisms
in the biofilm architecture is known to allow the treatment of wastewaters containing
BTEX. Apparently, such a performance was not dependent on the methanogenic activity in
the biofilm.
The cellular quantification showed that concentration of the anaerobic bacteria was
maintained in the biofilm during the acclimatization phase and in the first reactor segment
during the second phase. A decrease of biomass in the last reactor segment at the end of the
operation period was observed. In respect to the methanogenic archaea, the decrease can be
considered significant compared to the total biomass (Table 4).
Table 3 MPN during the operation time

Concentration End acclimatization phase End second phase

(MPN organisms/mL) FRS LRS FRS LRS

Anaerobic Bacteria 2.4×108 2.4×108 2.4×108 6.1×106

Methanogenic Archaea 1.3×106 1.1×106 2.4×104 9.5×103

Conclusions
The results obtained in this experiment permit us to conclude that anaerobic packed-bed
reactor filled with polyurethane foam matrices are effective for decontamination of BTEX
containing wastewaters, provided adequate environmental conditions. Almost complete
mineralization of BTEX was achieved in the presence of both the co-solvent ethanol and
LAS. While ethanol favored the maintenance of active methanogenic biomass inside the 179
 reactor, LAS was found to have inhibited the methanogenic activity. However, there were
indications that such an activity was not a determinant process for the BTEX degradation in
the presence of LAS.

Acknowledgments
The authors are grateful to FAPESP – Fundação de Amparo à Pesquisa do Estado de São
Paulo, for the financial support given to the thematic project on biological wastewater treat-
I.R. de Nardi et al.

ment under development at the Departamento de Hidráulica e Saneamento, da Escola de

Engenharia de São Carlos – Universidade de São Paulo.

References
Corseuil H. X., Hunt C. S., Santos R. C. F. and Alvarez P. J. J. (1998) The influence of the gasoline
oxygenate ethanol on aerobic and anaerobic BTX biodegradation. Wat. Res., 32, 2065–2072.
Enkiri F., Hulen C. and Legault-Demare J. (1995) Hydrophobic adsorption of aromatic compounds on
polyurethane foam as carbon source for Pseudomonas growth. Appl. Microbiol. Biotechnol., 44,
539–545.
Harwood C. S. and Gibson J. (1997) Shedding light on anaerobic benzene ring degradation: a process unique
to prokaryotes? J. Bacteriol., 179(2),301–309.
Meckenstock R. U. (1999) Fermentative toluene degradation in anaerobic defined syntrophic cocultures.
FEMS Microbiol. Let., 177, 67–73.
Robertson W. J., Bowman J. P., Franzmann P. D. and Mee B. J. (2001) Desulfosporosinus meridiei sp. Nov.,
a spore-forming sulfate-reducing bacterium isolated from gasoline-contaminated groundwater. Inter. J.
System. Evolution. Microbiol., 51,133–140.
Shim H. & Yang S. T. (1999) Biodegradation of benzene, toluene, ethylbenzene, and o-xylene by a cocul-
ture of Pseudomonas putida and Pseudomonas fluorescens immobilized in a fibrous-bed bioreactor. J.
Biotechnol., 67(2–3), 99 – 112.
Standard Methods for the Examination of Water and Wastewater (1995). 19th edn, American Public Health
Association / American Water Works Association / Water Environment Federation, Washington DC,
USA.
Varesche M. B. A., Zaiat M., Vieira L. G. T., Vazoller R. F. and Foresti E. (1997) Microbial colonization of
polyurethane foam matrices in horizontal-flow anaerobic immobilized sludge (HAIS) reactor. Appl.
Microbiol. Biotechnol., 48(4),534–538.
Vazoller R.F. (1995) Avaliação do ecossistema microbiano de um biodigestor anaeróbio de fluxo ascen-
dente e manta de lodo, operado com vinhaça sob condições termofílicas. PhD thesis, Department of
Hidraulic and Sanitation, University of São Paulo.
Wagener S. and Schink B. (1987) Anaerobic degradation of nonionic and anionic surfactants in enrichment
cultures and fixed-bed reactors. Wat. Res., 21, 615–622.
Watrick P. and Kotter R. (2000) Biofilm, City of microbes. J. Bacteriol., 182(10), 2675–2679.
Zaiat M., Vieira L. G. T. and Foresti E. (1997) Spatial and temporal variations of monitoring performance
parameters in horizontal-flow anaerobic immobilized sludge (HAIS) reactor treating synthetic
substrate. Wat. Res., 31, 1760–1766.

180