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

net/publication/282983748

An alternative anaerobic treatment process for treatment of heavy oil refinery

wastewater containing polar organics

Article  in  Biochemical Engineering Journal · January 2016

DOI: 10.1016/j.bej.2015.08.012

CITATIONS READS

22 445

7 authors, including:

Qinghong Wang Yingnan Yang

China University of Petroleum University of Tsukuba
50 PUBLICATIONS   494 CITATIONS    104 PUBLICATIONS   2,089 CITATIONS   

SEE PROFILE SEE PROFILE

Shaohui Guo

81 PUBLICATIONS   554 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Qinghong Wang on 02 February 2018.

The user has requested enhancement of the downloaded file.

 Biochemical Engineering Journal 105 (2016) 44–51

Contents lists available at ScienceDirect

Biochemical Engineering Journal

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

Regular article

An alternative anaerobic treatment process for treatment of heavy oil

refinery wastewater containing polar organics
Yu Wang a , Qinghong Wang a,1 , Min Li a , Yingnan Yang b , Wei He a , Guangxu Yan a ,
Shaohui Guo a,∗
a
State Key Laboratory of Heavy Oil Processing, Beijing Key Laboratory of Oil & Gas Pollution Control, China University of Petroleum, 18 Fuxue Road,
Changping District, Beijing 102249, PR China
b
Graduate School of Life and Environmental Sciences, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-8572, Japan

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

Article history: Heavy oil is an important part of energy sources, but the refining wastewater is difficult to treat by the
Received 13 May 2015 conventional anaerobic process, which has low efficiency and poor ability to resist impact load. In this
Received in revised form 10 August 2015 study, an up-flow anaerobic sludge bed (UASB) reactor was applied to treat heavy oil refinery wastewa-
Accepted 24 August 2015
ter containing large amounts of polar organics. Through a progressive increase of hydraulic conditions,
Available online 28 August 2015
the average removal efficiencies of COD and total oil reached 70% and 72%, respectively, at an organic
loading rate (OLR) of 3.44 kg COD/m3 d. GC–MS analysis revealed that more biodegradable organic acids
Keywords:
and alcohols were generated and macromolecular polar organics were degraded into small molecular
Anaerobic processes
Heavy oil refinery wastewater
intermediates after UASB treatment. The morphology observation of the sludge demonstrated that gran-
Polar organics ular sludge with an average particle size of 1 mm was formed. Moreover, the predominant species and
Bioreactors microbial community shift could reflect the performance of the reactor. The long-term operation of UASB
Biodegradation exhibited excellent polar organic removal efficiency. The study demonstrated the potential of UASB as
Waste-water treatment an alternative for high-efficiency anaerobic treatment of heavy oil refinery wastewater.
© 2015 Elsevier B.V. All rights reserved.

1. Introduction ical treatment. Currently, most heavy oil refinery wastewater is

With the increase in energy demand, the world oil consump-
tion increased from 88.3 million barrels per day in 2010 to 91.2
million barrels per day in 2013 [1]. Driven by petroleum and chem-
ical product consumption of emerging markets, such as China and
India, the global oil requirement is projected to rise by 20–25% over
the next decade [2]. Heavy oil, as an important part of the energy
supply, is being increasingly exploited. Accordingly, huge amounts
of heavy oil refinery wastewater are generated from crude oil pro-
cessing. Compared with light oil refinery wastewater, heavy oil
refinery wastewater contains more polar organics and complicated
dissolved recalcitrant compounds, such as naphthenic acid, hete-
rocyclic compounds, surfactants and heavy mineral oil [3]. Most of
those contaminants are toxic and macromolecular, which causes
poor biodegradability and low solubility of the wastewater [4].
Furthermore, heavy oil refinery wastewater strongly fluctuates in
terms of quantity and quality, increasing the difficulty of biolog-
∗ Corresponding author.
E-mail address: cupgsh@163.com (S. Guo).
1
These authors contributed equally to this work.
treated by conventional anaerobic-aerobic treatment processes. In
an anaerobic unit, only an organic loading rate (OLR) of 0.3–0.6 kg
COD/m3 d was obtained [5]. It is possible that the poor ability of
sludge sedimentation leads to microbe washout and poor resis-
tance to toxic compounds, causing biological inhibition.
In recent years, researchers have attempted to develop some
innovative pretreatment methods to remove biorefractories and
improve biodegradability, such as electrolysis [6], ozonation [7] and
the electron-beam method [8]. Although these methods are well
established in lab-scale experiments, it is necessary to have a rapid
application of low cost and mature technology in the treatment
of heavy oil refinery wastewater. Highly-efficient anaerobic bio-
logical technology has been widely applied, because it has minimal
operational cost, small occupation, strong resistance to impact load
and excellent organic removal efficiency [9]. Because of the sim-
ple design, easy construction and maintenance, UASB bioreactors,
as one of the most popular highly-efficient anaerobic wastewater
treatment systems, have been used throughout the world since the
1980s [10]. It has proven effective for the treatment of medium- and
high-strength wastewater, such as printing and dyeing wastewa-
ter [11], slaughter wastewater [12], paper-making wastewater [13],
brewery wastewater [14] and food processing wastewater [15]. In

http://dx.doi.org/10.1016/j.bej.2015.08.012
1369-703X/© 2015 Elsevier B.V. All rights reserved.
 Y. Wang et al. / Biochemical Engineering Journal 105 (2016) 44–51
these wastewater treatment systems, UASB was utilized as an effec-
tive anaerobic treatment unit replacing the conventional anaerobic
digestion tank. High-density biomass with high activity and excel-
lent settling properties satisfy greater efficiency and better effluent
quality. Recently, interest in utilizing UASB reactors for the treat-
ment of petrochemical wastewater has grown [16]. For example,
Gasim et al. [17] reported a UASB reactor was successfully operated
to treat petroleum refinery wastewater. Additionally, Nasirpour
et al. [18] examined the treatment of petroleum refinery effluents
in a combined UASB and an aerobic packed-bed biofilm reactor.
However, these reports are mostly limited to petroleum refinery
wastewater treatment. There are few applications of UASB to heavy
oil refinery wastewater. Such wastewater is enriched with dis-
solved recalcitrant organic compounds and higher levels of sulfur,
metals and salts [3]. In case of overloading, a serious influence on
biochemical systems can occur [19]. Until now, there are no ref-
erences of whether UASB proceeds smoothly during the long-term
operation of heavy oil refinery wastewater. Further investigation in
assessing the continuous stability of UASB is still urgently required
to drive industrial application.
Therefore, in this study, a lab-scale UASB reactor was con-
structed. The objective of the research is to investigate an
alternative anaerobic treatment process for the treatment of heavy
oil refinery wastewater that contains polar organics. By choosing
this strategy, the problems of low OLR and poor biodegradability
encountered in current anaerobic units can be overcome.
2. Materials and methods
2.1. Heavy oil refinery wastewater and seed sludge
Heavy oil refinery wastewater was collected in a sour water
stripping unit from the Liaohe Petrochemical Branch Company
(Liaoning province, China), which is a main heavy oil refinery in
China, with a capacity of over 3 Mt/a. The characteristics of the
wastewater are shown in Table S-1 (Supplementary material).
Seed sludge was obtained from a hydrolysis acidification tank
of the refinery wastewater treatment plant of the Liaohe Petro-
chemical Company (Liaoning province, China). It showed good
settleability and a dark brown color, with 17.23 g/L of mixed liquor
suspended solids (MLSS) and 8.74 g/L of mixed liquor volatile sus-
pended solids (MLVSS), respectively.
2.2. UASB configuration
The laboratory-scale UASB bioreactor consisted of a 5 mm-thick
Plexiglas plate, a total volume of 4.45 L, with effective volume of
3.38 L, an internal diameter of 7 cm and a height of 55 cm. The
reactor was composed of a three-phase separator with a volume
of 0.27 L to separate gas, water and sludge. The UASB reactor con-
figuration is illustrated in Fig. 1. The technological process can be
found in the supplementary material section. The UASB reactor
ran for 285 days, and the steady state performance was studied
under four influent flows (1.81 L/d, 2.72 L/d 3.63 L/d and 4.54 L/d)
and the corresponding HRTs (45 h, 30 h, 22 h and 18 h). The pH,
biogas production and composition, VFA, ORP, COD and oil con-
centration were determined during the experiment. Samples were
analyzed in duplicate and average values were reported.
2.3. Anaerobic biodegradability evaluation experiment
A batch-scale experiment was conducted to evaluate the
biodegradability of the heavy oil refinery wastewater. Detailed
information can be found in the supplementary material section.
According to stoichiometric calculation, methane conversion
(M%), acidification rate (A%), biodegradable COD (CODBD ) and
45
biodegradability (BD%) were obtained by the following equations
[20]:
CODCH4
Methaneconversion(M%) = × 100 (1)
COD0
CODacid
Acidifiationrate(A%) = × 100% (2)
COD0
CODBD
Biodegradability(BD%) = × 100% = E% + VFA% (3)
COD0
in which COD0 = initial COD, CODCH4 = conversion of CH4 to COD,
1 g COD is equal to 350 mL CH4 under the standard condition,
CODacid = acidified COD = CODCH4 + CODvfa , CODvfa = conversion of
VFA to COD, 1 g VFA in terms of acetic acid is equal to 1.067 mg
COD
COD, E% = the removal rate of COD, and VFA% = CODvfa × 100%.
0
2.4. Microbial structure analysis
2.4.1. DNA extraction, PCR amplification and high-throughput
sequencing
The bacterial communities in seed and acclimated sludge were
investigated by Illumina high-throughput sequencing. DNA was
extracted from sludge samples with a PowerSoil DNA Isolation kit
(Mobio Laboratories Inc., Carlsbad, CA, USA) following the man-
ufacturer’s protocol. The homogenization step was performed at
5000 rpm for 20 s using Precellys® 24 (Bertin Technologies, France).
The extracted DNA was stored at −20 ◦ C prior to further analyses.
The V3 and V4 regions of the 16S rDNA gene were selected for
PCR. The primers are 340F (5 -TCCTACGGGAGGCAGCAGT-3 ) and
805R (5 -GGACTACCAGGGTATCTA ATCCTGTT-3 ) [21]. All PCR reac-
tions were conducted in 50 ␮L reactions with 38.8 ␮L of ddH2 O,
5 ␮L of 10× Buffer A (KAPA BIOSYSTEM, UAS), 1 ␮L 10 mM dNTPs,
2 ␮L reverse primers, 0.2 ␮L of KAPA Taq DNA Polymerase (KAPA
BIOSYSTEM, UAS) and 1 ␮L template DNA. PCR amplification was
performed as follows: 95 ◦ C for 3 min, followed by 30 cycles of
denaturation at 95 ◦ C for 30 s, annealing at 50 ◦ C for 30 s, elonga-
tion at 72 ◦ C for 60 s and final extension at 72 ◦ C for 7 min. After
PCR amplification, 5 ␮L PCR products were mixed with the same
volume of 1X loading buffer (containing SYBR green) and elec-
trophoresis was conducted on 1.5% agarose gel to detect positive
amplified bands. Samples with the bright main strip between 400
and 450 bp were chosen for further analyses. Then, PCR products
were purified with a gel extraction kit (MinElute Gel Extraction
Kit, QIAGEN, Germany). Amplicons from different sludge samples
were sent out for pyrosequencing on the Illumina MiSeq platform
at SinoGenoMax Co., Ltd (Beijing, China).
2.4.2. Pyrosequencing data analysis for microbial communities
For quality control, the reads that contained one or more
ambiguous bases (“N”) were first removed. Illumina sequencing
generated a pair of reads from the two ends (paired-end reads) for
one DNA fragment. Paired-end reads from the original DNA frag-
ments were merged by software (SOAPdenovo) to avoid overlaps.
Then, the tag sequences were sorted into different individual files
according to the barcodes of all samples. Sequence data were pro-
cessed by read trimming and identification of V3–V4 sequences,
followed by filtering and assigning the operational taxonomic units
(OTUs). OTUs were identified with a cutoff of 97 % identity. The
reads from filtered OTUs are processed using the Quantitative
Insights into Microbial Ecology (QIIME) program to construct a rep-
resentative sequence for each OUT. Representative sequences from
each OUT were selected to annotate taxonomic information.
46 Y. Wang et al. / Biochemical Engineering Journal 105 (2016) 44–51
Fig. 1. Schematic diagram of the UASB process: (1) influent tank; (2) magnetic stirrer; (3) peristaltic pump; (4) UASB reactor; (5) safety bottle; (6) wet type gas flowmeter;
(7) thermostatic waterbath; (8) effluent tank.
2.5. Analytical methods
The total COD (CODt ), BOD5 , NH3 -N, TSS, SS, MLSS, and MLVSS
were determined according to the American Public Health Associ-
ation method [22]. For soluble COD (CODsol ) and volatile fatty acid
(VFA) analyses, the samples were first filtered through a 0.45 ␮m
polyethersulfone membrane. VFA were determined using an Agi-
lent 6890 gas chromatograph equipped with a flame ionization
detector (FID) using helium as the carrier gas. The pH and oxidation-
reduction potential (ORP) values were measured using a pH/ORP
meter (FE20, METTLER, TOLEDO, Switzerland). Total oil concen-
tration in the wastewater was determined by a national standard
method [23]. It was composed of polar organics and petroleum
hydrocarbons. Polar organics were the parts of the total oil adsorbed
by magnesium silicate, also determined according to the national
standard method. The total biogas production was measured by
a wet type gas flowmeter (SHAKIC LMF-1-0.2, Changchun). Bio-
gas compositions were identified with a gas chromatograph (GC,
Agilent 6890) equipped with a TCD detector and a Porapakq pack-
ing column. The morphology of the anaerobic sludge in the reactor
was observed with a Scanning Electron Microscope (S-3000N SEM,
Hitachi, Japan) on day 0 and 285, respectively. Particle size mea-
surement of the seed and acclimated anaerobic sludge in the UASB
reactor was performed using a LS 200 laser diffraction particle size
analyzer (Beckman Coulter).
The organic chemical composition was characterized using a
gas chromatograph-mass spectrometer (GC–MS). The water sam-
ple pretreatment method and GC–MS detection condition are
described in the supplementary material.
3. Results and discussion
3.1. Anaerobic biodegradability of heavy oil refinery wastewater
Biodegradability in terms of BD% is an integrate index, which can
reflect the degree of biodegradation of the wastewater, preferably
under anaerobic conditions. A batch experiment was performed to
detect anaerobic biodegradability of heavy oil refinery wastewa-
ter. The experiment was undertaken for 661 h. At the end of the
experiment, cumulative methane production, residual VFA and
soluble COD concentration were determined. According to the
above-mentioned equations, methane conversion (M%), acidifica-
tion rate (A%) and biodegradability were calculated, and the results
are shown in Table 1. The results indicated that such wastewater
can be treated by the anaerobic treatment process because more
than half of organic compounds can be biodegraded into biogas.
However, considerable time was spent on removing over 90 % of
the organic pollutants when the initial COD concentration was
approximately 2000 mg/L. This result demonstrated that the refin-
ery wastewater contains lots of macromolecular and complicated
organic compounds. Therefore, the UASB reactor should be oper-
ated under low OLR and long HRT when treating heavy oil refinery
wastewater.
3.2. Long-term performance of UASB reactor
HRT and OLR are important factors that reflect a supply-demand
relationship between microorganisms and the substrate. It should
be step-feed to keep balance between VFA accumulation and
methane production, especially for the treatment of wastewater
containing toxic and refractory compounds. If the HRT is too short
or the OLR is too high, this may cause serious problems of sludge
washout, superacidulation and low COD removal efficiency [24].
In this study, HRT shortened stepwise from 45 to 18 h with OLR
increasing from 0.6 to 3.44 kg COD/m3 · d in fluctuation. Fig. 2 shows
the process performance of the UASB reactor in terms of COD, pH,
ORP, total oil concentration, polar organics, biogas production rate
and composition during the experiment.
As shown in Fig. 2(a), the influent COD concentration fluctuated
greatly during the entire operation period. Effluent COD concen-
tration was stable after a start-up period of 75 days. The CODt
removal rate was steadily maintained in the range of 61–86 %
from day 75 to day 285. More specifically, when HRT gradually
reduced from 45 h to 22 h, both CODt and CODsol removal effi-
ciencies increased approximately 10% on average. With further
reduction of HRT to 18 h, CODt and CODsol removal efficiencies
decreased by 8% and 6%, respectively. However, during the last
 Y. Wang et al. / Biochemical Engineering Journal 105 (2016) 44–51
Table 1
Results of anaerobic biodegradability of heavy oil refinery wastewater.
Cumulative esidual VFA Residual
methane concentration soluble COD
production (mg/L) concentration
(mL) (mg/L)
115.0 6.2 114.7
Fig. 2. Operational performance of the UASB reactor at different HRTs and OLRs.
ten days of the experiment, 78% of influent CODt was removed on
average. This indicates that after a long time in the adaptive phase,
stable organics removal performance can be obtained at an OLR
of 3.44 kg COD/m3 d. This performance was acceptable considering
such wastewater with characteristics of refractory and complicated
organics. In the literature, 76% COD removal efficiency at OLR 0.4 kg
/m3 d was reported when treating petroleum refinery effluent in a
UASB reactor [16]. Another study showed that 83 % COD removal
efficiency was obtained at an optimum OLR of 1.21 kg/m3 d [17].
Comparison with the literatures for similar wastewater indicated
that this study made progress in higher sustainable organic load-
ing rates and stable performance of refractory organics treatment.
Furthermore, CODsol removal was higher than CODt because the
insoluble COD (CODins ) removal was low (less than 60 %), which
was calculated by CODt minus CODsol . The macromolecular organic
compounds, such as colloid and asphaltene in CODins , affected the
47
Methane Acidification Biodegradability (%)
conversion rate rate
(%) (%)
63.3 63.6 94.3
substrate transportation and subsequent biological degradation.
CODins removal is attributed to adsorption/filtration of sludge and
hydrolysis of microbial communities [25]. The tendency of CODins
removal efficiencies was initially decreased and then increased (the
average values at the four stages were 51%, 47%, 58% and 56%,
respectively). This result demonstrated that anaerobic granular
sludge enhanced the physical entrapment, filtration and biological
degradation.
The pH variations have significant influence on microbial
metabolism including utilization of carbon and energy sources, effi-
ciency of substrate degradation, cell morphology and structure [26].
If pH were severely fluctuated, it would lead to a massive die-off of
microorganisms. In Fig. 2(b), the influent pH was changed between
6.76 and 8.73, and the effluent pH remained balanced at approxi-
mately 7.50. This indicates that the system was well self-buffered
and the operation was stable.
48 Y. Wang et al. / Biochemical Engineering Journal 105 (2016) 44–51
It is reported that lower ORP is instrumental in the survival and
growth of obligate anaerobes, especially methanogens [27]. In an
anaerobic system, ORP reflects the dissolved oxygen concentration,
organic substrate and organism activity [28]. It can serve as an indi-
cator to monitor the reactor condition. It was reported that ORP has
been widely used as a consistent parameter for process control [29].
In Fig. 2 (c), the influent ORP ranged from −85 mV to −281 mV,
and the effluent ORP fluctuated at approximately −350 mV. This
result showed that the reactor was maintained under strict anaer-
obic conditions. According to Hirasawa et al. [30], a favorable ORP
for hydrolysis acidification varies from +100 to −350 mV for differ-
ent fermentation types. Methanogenesis occurs at ORP values lower
than −300 mV [31]. Therefore, observed ORP values were favorable
for both hydrolysis acidification and methanogenesis in the UASB
reactor. In addition, ORP was inversely related to pH in most cases.
There was an obvious increase of effluent ORP from day 170 to day
190, accompanying low influent pH. It could be that low influent pH
provides a favorable environment for the rapid growth of the acido-
genic fermentation bacteria. Afterwards, CH4 production decreased
and ORP increased gradually.
High oil concentration is one of the characteristics of heavy oil
refinery wastewater. In this study, influent total oil concentration
was lower than 200 mg/L in the first 75 days and then increased
gradually, varying between 200 mg/L and 400 mg/L. The concen-
tration of the effluent showed a gradual decreasing tendency.
Especially, when HRT was 18 h, the total effluent oil concentration
was below 50 mg/L and the oil removal efficiency was approach-
ing 80%. Petroleum hydrocarbons were not significantly degraded,
which were detected by subtracting the amount of polar organics
from the total oil. The results also can be found in the supplemen-
tary material section (Fig. S-1). Comparing with Fig. 2(d) and (e), it
was found that almost 4/5 total oils came from polar organics. The
concentration of polar organics in effluent was much lower, and
its removal efficiency was much higher. The UASB reactor showed
excellent oil removal performance especially in terms of polar
organics. Section 3.3 shows detailed information on the organic
composition and its variation in the reactor.
Biogas production rates (BPR) and composition are displayed in
Fig. 2(f). As OLR increased, the production rate increased gradu-
ally during the experiment, which is consistent with Hu’s research
[32]. Methanogens could utilize acetic acid or CO2 /H2 in bio-
chemical reactions to produce CH4 . Only a few substrates can
be utilized by methanogens during the startup period, and as a
consequence, BPR was low (only 0.13 L/L d on average). After the
startup operation, the macromolecular organics were effectively
degraded into micromolecular organics through hydrolysis acid-
ification. These micromolecular organics including formic acid,
acetic acid, methanol and methylamine promoted methanogen-
esis. Therefore, both BPR and the methane concentration were
increased gradually. As shown in Fig. 2(f), BPR increased from
0.03 L/L d to 0.77 L/L d and the methane concentration increased
from 45% to 86%. However, BPR was sharply decreased when
the influent pH or ORP was not favorable for methanogenesis
because methanogens are restricted by environmental conditions.
Because of the enhancement of hydrolytic acidification and hys-
teretic reflection of methane production, several days later, BPR
increased again. From stage 2 to stage 4, methane remained the
principal component in the biogas, mostly in the range of 66–86%.
In most anaerobic digestion systems, the methane concentration
was between 65% and 75% [33]. The higher amount of methane
compared with CO2 demonstrates the presence and growth of
H2 -consuming and CO2 -consuming bacteria in the UASB reactor.
Because of the low incoming COD, bacteria will consume the pro-
duced CO2 to generate CH4 . This phenomenon was also observed
in other works at low influent COD levels [16,34].
3.3. Characterization of organic compositions
The organic components in the influent and effluent in the
UASB are subjected to GC–MS, and the results are shown in
Fig. 3. 109 different chemical compounds were detected, belong-
ing to 12 major groups in the influent, including phenols, organic
acids, ketones, alcohols, esters, ethers, alkanes, alkenes, amides,
amines, nitriles and heterocyclic macromolecular organics. 23 com-
ponents of abundances of these groups were above 60,000,000
(Fig. 3(A)) and the highest peak response value occurred at
6.72 min (1270,000,000), corresponding to o-cresol. In addition,
phenols (phenol, xylenol), organic acids (pentenoic acid, hexade-
canoic acid and octadecanoic acid) and N-containing organics
(docosenamide, quinoline, and octadecenamide) accounted for
more than 60% of the total content. This result confirmed that
polar organics are primary pollutants. The influent also contained
13 types of alkanes ranging from tetradecane (C14 H30 ) to hexa-
cosane (C26 H54 ), as shown in Fig. S-1. The influent esters mainly
consisted of dimethyl phthalate, di-n-butyl phthalate, di-n-octyl
phthalate, bis(2-ethylhexyl) phthalate, 5-ethyl-4-phenyloxazole,
benzyl thiocyanate, and 2,4-bis(ethylamino)-6-(methylthio)-1,3,5-
triazine. All of these organics were large-molecular, toxic and
refractory. Most of them were classified as priority pollutants both
by the U.S. Environmental Protection Agency (EPA) and in China
[35,36].
In the effluent, there were 115 different species also belonging to
the above 12 major groups. The highest peak was also o-cresol, but
the abundance decreased to 932,000,000. A comparison of Fig. 3
(a) and (b) showed that the GC–MS abundance and area values
for each of the substances in the effluent decreased considerably
from the corresponding value in the influent, especially for macro-
molecular N-containing compounds and heteroatomic compounds.
C18 -amides, C16 -nitriles and chlorooctadecane even disappeared in
the effluent (Fig. S-2 and Fig. S-3). The UASB is confirmed as capable
of decomposing organic substances and macromolecular organics
into small pieces that are more easily broken down. It is reported
that phenols are toxic to anaerobic granular sludge to some extent
[37]. In this study, phenols were degraded and not inhibitive to the
activity of microbes, as shown in Fig. S-4. Phthalate compounds are
known as fuel additives in the petroleum industry (such as peak
11 dimethyl phthalate, peak 15 diisobutyl phthalate and peak 22
diisooctyl phthalate), in Fig. 3. Biodegradation of these compounds
was not examined in this study. Benzene dicarboxylic acid with
two side chains could cause structural restrictions to biodegrada-
tion. The steric effect of phthalates side ester chains, which hinders
the hydrolytic enzymes from binding to the phthalates, thereby
inhibits hydrolysis[38].
Some new peaks are shown for substances (a–m) in Fig. 3
(B), and these new organic compounds included organic acids
and alcohols (in Fig. S-2 and Fig. S-2 ). These substances could
be metabolic products or hydrolytic intermediates under anaer-
obic conditions. Microbial catabolic and anabolic processes are
always directed to benefit their own growth. Although refrac-
tory and toxic organics such as phenol, cresol, free benzene
molecule additives (propanoic acid, 2-methyl-,2,2-dimethyl-1-(1-
methylethyl)-1,3-propanediyl ester and bis(2-ethylhexyl) adipate,
triethyl citrate) can not be decomposed directly, they can be
hydrolyzed into more readily degradable organic acids or alcohols
for subsequent utilization. Therefore, the UASB is effective for the
degradation of refractory organic compounds, which is attractive
for heavy oil refinery wastewater treatment.
3.4. Microbiology morphology
Morphological characteristics of seed sludge and acclimated
anaerobic sludge from UASB after 285 days of operation were
 Y. Wang et al. / Biochemical Engineering Journal 105 (2016) 44–51 49

Fig. 3. Gas chromatography-mass spectrometry results of heavy oil refining wastewater samples in influent (A) and effluent (B): (1) phenol; (2) benzyl chloride; (3) o-cresol;
(4) m-cresol; (5) 2-ethyl phenol; (6) 2,5-xylenol; (7) 2,3-xylenol; (8) 3,4-xylenol; (9) quinoline; (10) 2,3,6-trimethyl phenol and 2,4,6-trimethyl phenol; (11) dimethyl phtha-
late; (12) N,N-dimethyl-1-dodecanamine; (13) Propanoic acid, 2-methyl-, 2,2-dimethyl-1-(1-methylethyl) -1,3-propanediyl ester; (14) N,N-dimethyl-1-tetradecanamine;
(15) diisobutyl phthalate; (16) n-Hexadecanoic acid; (17) 1-phenyl-2-benzylaminopropane; (18) stearic acid; (19) N-methyl-N -benzyltetradecanamine; (20) (z)-9-
octadecenamide; (21) bis(2-ethylhexyl) adipate; (22) diisooctyl phthalate; (23) (z)-13-docosenamide; (a) 4-methylvaleric acid; (b) hexanoic acid; (c) 2-methylhexanoic
acid; (d) heptylic acid; (e) cyclopentaneacetic acid; (f) caprylic acid; (g) 5-indanol; (h) tridecanoic acid; (i) triethyl citrate; (j) tetradecanoic acid; (k) (9E)-9-hexadecenoic
acid; (l) S8 ; (m) oleic acid; (n) dioctyl isophthalate.

Fig. 4. Particle sizes, SEM images and morphology photos of seed sludge (A) and acclimated anaerobic sludge (B) in the UASB reactor.

identified. Fig. 4 shows the sludge particle size distribution and nation, the species and amount of the bacteria inside the seed
morphology using a camera and SEM. The seed sludge was floc- sludge were rare, with only filamentous bacteria and few bacilli
culent and with colored dark brown. The average particle size observed. After 285 days of operation of the UASB reactor, the
was distributed at approximately 0.12 mm. Upon SEM exami- acclimated anaerobic sludge consisted mainly of dark, small and
50 Y. Wang et al. / Biochemical Engineering Journal 105 (2016) 44–51
Fig. 5. Changes in bacterial communities (genus level) before (a) and after (b) 285 days of operation of the UASB reactor. The category “other” contains all genera with relative
abundance less than 1%. (a) Seed sludge; (b) Acclimated sludge.
compact granules with an average diameter of 1 mm. Bacteria num-
bers were markedly increased. Arthrobacter became the dominant
species instead of filamentous bacteria, and large quantities of
Methanosarcina appeared. In addition, long rod bacteria multiplied
and filamentous bacteria were surrounded by sarcina and long rod
bacteria. This indicated that the sludge system in the UASB was
integrated and rich in biological populations to degrade the refrac-
tory organics of heavy oil refinery wastewater. Such an integrated
sludge system guaranteed a stable and effective UASB reactor [39].
3.5. Microbial community structural analysis
The microbial community structures in the inoculum and
granular sludge from the UASB were investigated by 16S rDNA
gene pyrosequencing, respectively. The microbial community shift
is shown in Fig. 5. Before being inoculated to the UASB reac-
tor, the dominant bacteria in the seed sludge were Treponema
(14.5%), Sulfuricurvum (10.6%), Parvibaculum (4.2%), Blvii28 (3.6%)
and Geobacter (3.4%). The predominant groups of the archaeal com-
munity are Methanosaeta, Methanomethylovorans, Methanolinea
and Methanomassiliicoccus, which all are methanogens, accounting
for 21.3 %. Among the top 23 species, many bacteria are associated
with aerobes and facultative aerobes, such as Sulfuritalea, Acidovo-
rax, Perlucidibaca, Prosthecobacter and Aquabacterium. Furthermore,
the genus of Treponema (14.5%) is a pathogen that is harmful to
water quality [40].
When the reactor had been operated for 285 days, a signifi-
cant change of the compositions of the microbial community could
be observed. The top 11 bacterial genera are W22 (11.4%), Blvii28
(7.8%), T78 (4.3%), Treponema (3.8%), Chryseobacterium (3.4%), W5
(2.7%), Sedimentibacter (2.2%), Arcobacter (2.2%), RFN20 (1.4%),
Desulfovibrio (1.4%) and Kosmotoga (1.4%). The dominant bacterial
genera of the system were shifted to strongly anaerobic conditions
and enriched with a variety of hydrolysis fermentation bacteria, aci-
dogenic bacteria, acetogens and syntrophus. For example, Blvii28,
which is a fermentation bacteria that transforms carbohydrates to
hydrogen, accounted for 3.6% to 7.8% of the total number of clas-
sified sequences [41]. W22 (from ND to 11.4%) and W5 (from ND
to 2.7%) belong to family Cloacamonaceae, which is a family of syn-
trophic bacteria derived mostly from carbon and energy from the
fermentation of amino acids, and are identified in many anaero-
bic digesters [42–44]. T78 (from 1.8% to 4.3%) and Chryseobacterium
(from ND to 3.4 %) [45] were petroleum degrading bacteria involved
in the degradation of the long chain petroleum hydrocarbons. Sed-
imentibacter (from ND to 2.2%) is a group of acidogenic bacteria
involved in the acidogenesis of proteinaceous matter [46]. These
hydrolysis-fermentation bacteria may initiate the anaerobic degra-
dation of organics in heavy oil refinery wastewater, by which large
molecular organics were broken-down and further metabolized to
CO2 , H2 , acids and alcohols. In addition, the number of Treponema
decreased from 14.5% to 3.8%, which illustrates that the UASB reac-
tor plays a positive role in reducing pathogens. Similar research
was reported in Orzi et al. [47]. Additionally, the archaeal com-
munity was also changed. The main archaeal were composed of
Methanosaeta, Methanolinea, Methanosarcina, Methanomethylovo-
rans and Methanobacterium, which are abundant in many anaerobic
treatment processes [48–51]. The percentage of methanogens dis-
tinctly increased from 21.3% to 38.4%. The genus Methanosaeta
(from 14.8% to 23.0%) plays an important role in the onset of
granulation and the maintenance of stable granules during system
perturbations [52]. The increased populations of the methanogenic
genus partly account for the organic degradation and increased bio-
gas yield in the second period. The functional species and microbial
community shift could explain the good performance of the UASB
reactor.
4. Conclusions
This study showed that UASB was a highly efficient process to
treat heavy oil refinery wastewater after a long-term startup period
with progressively increasing hydraulic conditions. The optimum
OLR of 3.44 kg COD/m3 · d in this reactor was much higher than
current anaerobic processes used in treating heavy oil refinery
wastewater. GC–MS analysis showed that most polar organics were
degraded by the UASB process. The formed granular sludge and
integrated microorganism community guaranteed a stable and
effective performance of the reactor. This study demonstrated the
UASB system as a technically feasible and potential alternative to
enhance the removal efficiency of polar organics and improve the
biodegradability of heavy oil refinery wastewater.
Acknowledgments
This study was financially supported by the National Natural Sci-
ence Foundation of China (No. 21306229), the Science Foundation
of China University of Petroleum, Beijing (No. 2462014YJRC001)
and the Program of the Beijing Municipal Science & Technology
Commission (No. Z141109004414088).
 Y. Wang et al. / Biochemical Engineering Journal 105 (2016) 44–51
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bej.2015.08.012.
References
[1] E.I. Administration, International Energy Statistics in, http://www.eia.gov/
cfapps/ipdbproject/IEDIndex3.cfm?tid=5&pid=5&aid=2, 2014.
[2] S. Chopra, L.R., Lines, D.R., Schmitt, M.L. Batzle, Heavy oils: reservoir
characterization and production monitoring, Geophysical development
series. ISBN 0931830419, 9780931830419. SEG Books, 2010.
[3] K. Tong, Y. Zhang, P.K. Chu, Evaluation of calcium chloride for synergistic
demulsification of super heavy oil wastewater, Colloids Surf. A: Physicochem.
Eng. Aspects 419 (2013) 46–52.
[4] C. Chen, H. Chen, X. Guo, S. Guo, G. Yan, Advanced ozone treatment of heavy
oil refining wastewater by activated carbon supported iron oxide, J. Ind. Eng.
Chem. 20 (2014) 2782–2791.
[5] X.C. Fei, Q.Y. Wang, W.G. Zhao, W.N. Dai, J.D. Liang, Status and Development
Activated Sludge for Treatment Industrial Wastewater, Ship Ocean Eng. 39
(2010) 68–70 (in Chinese).
[6] L. Wei, S. Guo, G. Yan, C. Chen, X. Jiang, Electrochemical pretreatment of heavy
oil refinery wastewater using a three-dimensional electrode reactor,
Electrochim. Acta 55 (2010) 8615–8620.
[7] C. Chen, L. Wei, X. Guo, S. Guo, G. Yan, Investigation of heavy oil refinery
wastewater treatment by integrated ozone and activated carbon -supported
manganese oxides, Fuel Process. Technol. 124 (2014) 165–173.
[8] B. Han, J. Ko, J. Kim, Y. Kim, W. Chung, I.E. Makarov, A.V. Ponomarev, A.K.
Pikaev, Combined electron-beam and biological treatment of dyeing complex
wastewater. Pilot plant experiments, Radiat. Phys. Chem. 64 (2002) 53–59.
[9] G. Lettinga, S. Rebac, G. Zeeman, Challenge of psychrophilic anaerobic
wastewater treatment, Trends Biotechnol. 19 (2001) 363–370.
[10] T.-T. Ren, Y. Mu, B.-J. Ni, H.-Q. Yu, Hydrodynamics of upflow anaerobic sludge
blanket reactors, Aiche J. 55 (2009) 516–528.
[11] H.S. Wu, Y. Tu, M. Xu, W. Wu, X.M. Yu, F.Z. Guo, Pilot-scale study on
pretreatment of printing and dyeing wastewater by UASB, Environ Sci
Technol 33 (2010) 146–149.
[12] N.T. Manjunath, I. Mehrotra, R.P. Mathur, Treatment of wastewater from
slaughterhouse by DAF-UASB system, Water Res. 34 (2000) 1930–1936.
[13] M. Huang, J. Wan, Y. Wang, Y. Ma, H. Zhang, H. Liu, Z. Hu, C. Yoo, Modeling of a
paper-making wastewater treatment process using a fuzzy neural network,
Korean J. Chem. Eng. 29 (2012) 636–643.
[14] A. Enitan, S. Kumari, F. Swalaha, J. Adeyemo, N. Ramdhani, F. Bux, Kinetic
modelling and characterization of microbial community present in a full-scale
uasb reactor treating brewery effluent, Microbial. Ecol. 67 (2014) 358–368.
[15] F. Dong, Q.-B. Zhao, J.-B. Zhao, G.-P. Sheng, Y. Tang, Z.-H. Tong, H.-Q. Yu, Y.-Y.
Li, H. Harada, Monitoring the restart-up of an upflow anaerobic sludge
blanket (UASB) reactor for the treatment of a soybean processing wastewater,
Bioresour. Technol. 101 (2010) 1722–1726.
[16] S.O. Rastegar, S.M. Mousavi, S.A. Shojaosadati, S. Sheibani, Optimization of
petroleum refinery effluent treatment in a UASB reactor using response
surface methodology, J. Hazard Mater. 197 (2011) 26–32.
[17] H.A. Gasim, S.R.M. Kutty, M.H. Isa, M.P.M. Isa, Treatment of petroleum refinery
wastewater by using UASB reactors World Academy of Science, Eng. Technol.
6 (2012) 162–165.
[18] N. Nasirpour, S. Mousavi, S. Shojaosadati, Biodegradation potential of
hydrocarbons in petroleum refinery effluents using a continuous
anaerobic-aerobic hybrid system, Korean J. Chem. Eng. 32 (2015) 874–881.
[19] L. Chonghua, Z. Hao, J. Xiaoyan, Preprocessing technological development and
engineering practice of inferior heavy oil refining sewage, Environ. Prot. Oil
Gas Fields 23 (2013) 31–33.
[20] Y. He, Anaerobic Biological Treatment of Wastewater, China Light Industry
Press, Beijing, 1998.
[21] T. Kim, E.-H. Lee, K.-S. Cho, Effects of nonmethane volatile organic compounds
on microbial community of methanotrophic biofilter, Appl. Microbiol.
Biotechnol. 97 (2013) 6549–6559.
[22] APHA, WEF, AWWA, Standard Methods for the Examination of Water and
Wastewater, 21st Edition. American Water Works Association, Washington,
DC, 2005.
[23] National Environment Bureau Water Wastewater Monitoring Analysis
Committee, Water and wastewater monitoring analysis method., China
Environment Science Publisher, Beijing, 1989. (in Chinese).
[24] Y. Liu, H.-L. Xu, K.-Y. Show, J.-H. Tay, Anaerobic granulation technology for
wastewater treatment, World J. Microbiol. Biotechnol. 18 (2002) 99–113.
[25] R. Rajakumar, T. Meenambal, P.M. Saravanan, P. Ananthanarayanan,
Treatment of poultry slaughterhouse wastewater in hybrid upflow anaerobic
sludge blanket reactor packed with pleated poly vinyl chloride rings,
Bioresour. Technol. 103 (2012) 116–122.
[26] H.Q. Yu, H.H.P. Fang, Acidogenesis of gelatin-rich wastewater in an upflow
anaerobic reactor: influence of pH and temperature, Water Res. 37 (2003)
55–66.
View publication stats
51
[27] L.W. Hulshoff Pol, S.I. de Castro Lopes, G. Lettinga, P.N.L. Lens, Anaerobic
sludge granulation, Water Res. 38 (2004) 1376–1389.
[28] P. Tanwar, T. Nandy, P. Ukey, P. Manekar, Correlating on-line monitoring
parameters, pH, DO and ORP with nutrient removal in an intermittent cyclic
process bioreactor system, Bioresour. Technol. 99 (2008) 7630–7635.
[29] S. Lackner, H. Horn, Evaluating operation strategies and process stability of a
single stage nitritation–anammox SBR by use of the oxidation–reduction
potential (ORP), Bioresour. Technol. 107 (2012) 70–77.
[30] J.S. Hirasawa, A. Sarti, N.K.S. Del Aguila, M.B.A. Varesche, Application of
molecular techniques to evaluate the methanogenic archaea and anaerobic
bacteria in the presence of oxygen with different COD:Sulfate ratios in a UASB
reactor, Anaerobe 14 (2008) 209–218.
[31] Y. Mu, H.-Q. Yu, G. Wang, Evaluation of three methods for enriching
H2-producing cultures from anaerobic sludge, Enzyme Microb. Technol. 40
(2007) 947–953.
[32] X.S.G.J.C. Hu, M.J. Zhou, et al., Theory and Technology of Anaerobic Biological
Wastewater Treatment [M], China Building Industry Press, 2003.
[33] J. Taricska, D. Long, J.P. Chen, Y.-T. Hung, S.-W. Zou, Anaerobic Digestion, in: N.
Wang (Ed.), Biosolids Treatment Processes, Humana Press, 2007, 2015, pp.
135–176.
[34] D.Z. Maat, L.H.A. Habets, The upflow anaerobic sludge blanket wastewater
treatment system: a technological review, Southam, Don Mills, ON, Canada,
1987, 88 (11), pp. 60–64.
[35] L. Keith, W. Telliard, ES&T special report: priority pollutants: I-a perspective
view, Environ. Sci. Technol. 13 (1979) 416–423.
[36] W. Zhou, D. Fu, Z. Sun, Priority controlled pollutants in water blacklist,
Environ. Monitor. China 6 (1990) 1–3.
[37] H.H.P. Fang, O.-C. Chan, Toxicity of phenol towards anaerobic biogranules,
Water Res. 31 (1997) 2229–2242.
[38] D.-W. Liang, T. Zhang, H.P. Fang, J. He, Phthalates biodegradation in the
environment, Appl. Microbiol. Biotechnol. 80 (2008) 183–198.
[39] G. -h, Z.Y. Liu, K.T. e, Y.- ong, Biotreatment of heavy oil wastewater by
combined upflow anaerobic sludge blanket and immobilized biological
aerated filter in a pilot-scale test, Biochem. Eng. J. 72 (2013) 48–53.
[40] X.M. Lu, P.Z. Lu, Characterization of bacterial communities in sediments
receiving various wastewater effluents with high-throughput sequencing
analysis, Environ. Microbiol. 67 (2014) 612–623.
[41] J. Graf, The Family Rikenellaceae, in: R. Eugene, D.E.F.L. Stephen, S. Erko, T.
Fabiano (Eds.), The Prokaryotes: Other Major Lineages of Bacteria and The
Archaea, Springer-Verlag, Berlin Heidelberg, 2014, pp. 857–859.
[42] L.-W. Meng, X. -k, K.W. Li, K.-L.M. ang, J.Z. a, Influence of the amoxicillin
concentration on organics removal and microbial community structure in an
anaerobic EGSB reactor treating with antibiotic wastewater, Chem. Eng. J. 274
(2015) 94–101.
[43] D. Riviere, V. Desvignes, E. Pelletier, S. Chaussonnerie, S. Guermazi, J.
Weissenbach, T. Li, P. Camacho, A. Sghir, Towards the definition of a core of
microorganisms involved in anaerobic digestion of sludge, ISME J. 3 (2009)
700–714.
[44] E. Pelletier, A. Kreimeyer, S. Bocs, Z. Rouy, G. Gyapay, R. Chouari, D. Riviere, A.
Ganesan, P. Daegelen, A. Sghir, G.N. Cohen, C. Medigue, J. Weissenbach, D. Le
Paslier, Candidatus Cloacamonas acidaminovorans: genome sequence
reconstruction provides a first glimpse of a new bacterial division, J Bacteriol.
190 (2008) 2572–2579.
[45] A. Saimmai, J. Kaewrueng, S. Maneerat, Used lubricating oil degradation and
biosurfactant production by SC-9 consortia obtained from oil-contaminated
soil, Ann. Microbiol. 62 (2012) 1757–1767.
[46] Y. Maspolim, Y. Zhou, C. Guo, K. Xiao, W.J. Ng, The effect of pH on
solubilization of organic matter and microbial community structures in
sludge fermentation, Bioresour. Technol. 190 (2015) 289–298.
[47] V. Orzi, B. Scaglia, S. Lonati, C. Riva, G. Boccasile, G.L. Alborali, F. Adani, The
role of biological processes in reducing both odor impact and pathogen
content during mesophilic anaerobic digestion, Sci. Total Environ. 526 (2015)
116–126.
[48] C. Rico, N. Muñoz, J. Fernández, J.L. Rico, High-load anaerobic co-digestion of
cheese whey and liquid fraction of dairy manure in a one-stage UASB
process:limits in co-substrates ratio and organic loading rate, Chem. Eng. J.
262 (2015) 794–802.
[49] A.A. Khan, I. Mehrotra, A.A. Kazmi, Sludge profiling at varied organic loadings
and performance evaluation of UASB reactor treating sewage, Biosyst. Eng.
131 (2015) 32–40.
[50] Y. Li, Y. Zhang, Z. Xu, X. Quan, S. Chen, Enhancement of sludge granulation in
anaerobic acetogenesis by addition of nitrate and microbial community
analysis, Biochem. Eng. J. 95 (2015) 104–111.
[51] P. Bhunia, M.M. Ghangrekar, Influence of biogas-induced mixing on
granulation in UASB reactors, Biochem. Eng. J. 41 (2008) 136–141.
[52] G. Collins, A. Woods, S. McHugh, M.W. Carton, V. O’Flaherty, Microbial
community structure and methanogenic activity during start-up of
psychrophilic anaerobic digesters treating synthetic industrial wastewaters,
FEMS Microbiol. Ecol. 46 (2003) 159–170.