Response of nitrifying bacterial communities to the increased thiocyanate concentration in pre-denitrification process.

Abstract

Changes in process performance and the nitrifying bacterial community associated with an increase of thiocyanate (SCN-) loading were investigated in a pre-denitrification process treating industrial wastewater. The increased SCN- loading led to the concentration of total nitrogen (TN) in the final effluent, but increasing the internal recycling ratio as an operation parameter from 2 to 5 resulted in a 21% increase in TN removal efficiency. In the aerobic reactor, we found that the Nitrosomonas europaea lineage was the predominant ammonia oxidizing bacteria (AOB) and the percentages of the AOB population within the total bacteria increased from about 4.0% to 17% with increased SCN- concentration. The increase of nitrite loading seemed to change the balance between Nitrospira and Nitrobacter, resulting in the high dominance of Nitrospira over Nitrobacter. Meanwhile, a Thiobacillus thioparus was suggested to be the main microorganism responsible for the SCN- biodegradation observed in the system.

Full text

Page 1

Response of nitrifying bacterial communities to the increased thiocyanate
concentration in pre-denitrification process
Young Mo Kima, Hyun Uk Chob, Dae Sung Leec, Chul Parka, Donghee Parkc,⇑, Jong Moon Parkb,⇑⇑
aDepartment of Civil and Environmental Engineering, University of Massachusetts, Amherst, MA 01003, USA
bAdvanced Environmental Biotechnology Research Center, Department of Chemical Engineering, School of Environmental Science and Engineering, Pohang
University of Science and Technology, San 31, Hyoja-dong, Pohang 790-784, Republic of Korea
cDepartment of Environmental Engineering, Kyungpook National University, Daegu 702-701, Republic of Korea
a r t i c l ei n f o
Article history:
Received 8 June 2010
Received in revised form 8 September 2010
Accepted 9 September 2010
Available online 17 September 2010
Keywords:
Thiocyanate
Pre-denitrification process
Internal recycling ratio
Nitrifying bacteria
Total nitrogen
a b s t r a c t
Changes in process performance and the nitrifying bacterial community associated with an increase of
thiocyanate (SCN?) loading were investigated in a pre-denitrification process treating industrial waste-
water. The increased SCN?loading led to the concentration of total nitrogen (TN) in the final effluent,
but increasing the internal recycling ratio as an operation parameter from 2 to 5 resulted in a 21%
increase in TN removal efficiency. In the aerobic reactor, we found that the Nitrosomonas europaea lineage
was the predominant ammonia oxidizing bacteria (AOB) and the percentages of the AOB population
within the total bacteria increased from about 4.0% to 17% with increased SCN?concentration. The
increase of nitrite loading seemed to change the balance between Nitrospira and Nitrobacter, resulting
in the high dominance of Nitrospira over Nitrobacter. Meanwhile, a Thiobacillus thioparus was suggested
to be the main microorganism responsible for the SCN?biodegradation observed in the system.
? 2010 Elsevier Ltd. All rights reserved.
1. Introduction
Thiocyanate (SCN?) is generated from a range of industrial
sources such as plastic production, metal finishing, gold mining,
and insecticide production (Sorokin et al., 2001). However, one of
its major sources is the carbonization of coal to produce coke. In
particular, cokes wastewater generated through coal coking in
the iron and steel manufacturing industries typically has high con-
centrations of SCN?(Kim et al., 2008b). This wastewater often con-
tains more nitrogen in the form of SCN?than ammonia, for
example, in the range of 200 to 800 mg/L SCN?(Kim et al., 2009).
Among various biological processes, a pre-denitrification pro-
cess, which is a combined process for simultaneous carbon and
nitrogen removal, has generally been employed due to its effi-
ciency and cost effectiveness (Kim et al., 2009). The pre-denitrifica-
tion process takes place in single-sludge systems recycling
nitrified-effluent and consists of the anoxic and aerobic stage of
recycling. In the anoxic stage, heterotrophic microorganisms re-
move nitrogen compounds through denitrification using organic
matter as carbon and energy sources (Kim et al., 2008b). In the aer-
obic stage, ammonia is nitrified into nitrite or nitrate ions by auto-
trophic nitrifying bacteria and any remaining organic matter is
degraded by heterotrophic bacteria. Thiocyanate is aerobically de-
graded to a combination of ammonia, carbonate, and sulfate by
several species of chemolithotrophic bacteria, Thiobacillus sp.,
which uses SCN?as a carbon and nitrogen source. Thiocyanate bio-
degradation by autotrophic bacteria results in the accumulation of
ammonia and sulfate. The ammonia that is produced is further oxi-
dized into nitrite or nitrate via nitrifying bacteria. However, it has
been reported that SCN?degrading bacteria compete with nitrifi-
ers for inorganic carbon or other essential nutrients (Kim et al.,
2008a), and the high concentration of SCN?can inhibit the nitrifi-
cation process (Kim and Kim, 2003). Actually, the full-scale pre-
denitrification process has occasionally failed to maintain the legal
discharge level of total nitrogen in the final effluent and field oper-
ators have had trouble with its instability and the restoration to its
original state, when the raw wastewater contains a high concen-
tration of SCN?. Additionally, nitrogen balances for an activated
sludge system at a coke plant revealed that in the absence of sub-
stantial nitrification, the effluent concentrations of ammonia were
significantly greater than influent values, primarily due to the bio-
degradation of SCN?. Although significant information has been
published about the effects of biological treatment using SCN?
(Staib and Lant, 2007; Lay-Son and Drakides, 2008), no study of
the effect of a SCN?shock loading occurrence on the process per-
formance and microbial population has yet been conducted. There-
fore, it is desirable to investigate changes in microbial population
in relation to changes in process performance in order to better
understand system instability. Many studies have been conducted
0960-8524/$ - see front matter ? 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2010.09.032
⇑Corresponding author. Tel.: +82 53 950 7566; fax: +82 53 950 7879.
⇑⇑Corresponding author. Tel.: +82 54 279 2275; fax: +82 54 279 2699.
E-mail addresses: dhpark@knu.ac.kr (D. Park), jmpk777@paran.com, jmpark@
postech.ac.kr (J.M. Park).
Bioresource Technology 102 (2011) 913–922
Contents lists available at ScienceDirect
Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Page 2

to link the microbial community to the functional attributes of
wastewater treatment using molecular techniques. However, few
have focused on changes in the nitrifying bacteria population
dynamics in relation to increased SCN?concentration.
In this study, a lab-scale pre-denitrification process reactor was
operated using actual cokes wastewater, in a continuous mode,
under the operating conditions of a full-scale one. The effect of
SCN?shock loading on the performance of the pre-denitrification
process was investigated. The study also aimed to improve the
efficiency of total nitrogen removal under SCN?shock loading.
Different internal recycling ratios were tested to evaluate the per-
formance of the pre-denitrification process reactor in treating
industrial wastewater. Meanwhile, nitrifying bacteria communities
were examined in relation to changes in process performance dur-
ing SCN?shock loading were examined using terminal restriction
fragment length (T-RFLP) and quantitative real-time PCR (qPCR).
2. Methods
2.1. Set-up and operation of the lab-scale reactor
Actual cokes wastewater was collected from the full-scale
wastewater treatment plant (WWTP) of a cokes manufacturing
plant in Pohang, Korea. During the operation of our reactor, the
concentrations of pollutants in the raw wastewater were as
follows: 1875–2675 mg/L of chemical oxygen demand (COD),
575–720 mg/L of total organic carbon (TOC), 226–267 mg/L of phe-
nol, 179–360 mg-N/L of total nitrogen (TN), 106–118 mg-N/L of
ammonia, 206–811 mg-SCN?/L of thiocyanate and 12.1–16.5 mg-
CN?/L of total cyanides. The pH was in the range of 8.2 to 8.9. To
mimic the full-scale pre-denitrification process as closely as possi-
ble, no supplementary nutrients were added into the raw waste-
water. Active sludge from the anoxic and aerobic tanks of the
full-scale process was sampled and used as microbial inoculums
for the anoxic and aerobic reactors of the lab-scale process.
A lab-scale pre-denitrification process reactor, consisting of an
anoxic reactor, an aerobic reactor, and a settler, was designed to
mimic a full-scale reactor. The working volumes of the anoxic and
aerobic tanks were 8 and 16 L, respectively. The reactor was oper-
ated in down flow mode. In the anoxic reactor, a variable-speed
stirrer was used to maintain biomass while operation was sus-
pended. In the aerobic reactor, compressed air was supplied for
mixing and aeration. Effluent from the aerobic reactor was allowed
to flow into a settler for the settling of sludge, which was then recy-
cled to the anoxic reactor. The concentration of suspended solids in
each reactor was maintained at 3500–4000 mg/L, similar to that in
the full-scale process. A temperature controller was used to keep
each reactor at 33–34 ?C and the pH of the aerobic reactor was
controlled at 7.3–7.5 by adding a 1 N NaOH solution. The dissolved
oxygen (DO) concentration in the aerobic reactor was more
than 4.0 mg/L, while the DO level of the anoxic reactor was main-
tained below 0.3 mg/L. The total hydraulic retention time (HRT) of
the reactor was 16.7 h and the sludge retention time (SRT) was
24 days.
2.2. Experimental design
These experiments were designed to evaluate nitrogen and car-
bon removal with different SCN?loads and using different internal
recycling ratios. The influent SCN?concentration and internal recy-
cling ratio were the two variables modified throughout the exper-
iments. Operation of the reactor began using normal cokes
wastewater with a mean COD of 1900 mg/L, thiocyanate of
200 mg/L and TN of 180 mg/L. The reactor operation was initiated
at a feed flow rate of 5.76 L/d and an internal recycling ratio of 2.
First, the performance of the reactor was monitored under three
SCN?shock loadings, at a fixed feed flow rate and recycling ratio of
2. The shock loadings were given continuously for about 12 days by
increasing the influent SCN?concentration to 400, 600, and
800 mg/L. Extra SCN?was added into the feed as potassium thiocy-
anate (Junsei), while the other influent components were main-
tained equal to those of normal cokes wastewater. With regard
to the internal recycling ratio, four different ratios corresponding
to the high concentration of SCN?(800 mg/L) were tested. The
reactor operated for about 12 days at each of four recycling ratios:
2–5. In the last operation, internal recycling ratios of 2, 3, and 5
were each applied to the reactor with different influent SCN?con-
centrations of 400, 600, and 800 mg/L. These experiments were
carried out to evaluate the best compromise between removal effi-
ciency and internal recycling ratio.
2.3. DNA extraction
One milliliter of the sample was centrifuged at 16,000g for
5 min and the supernatant was decanted. The pellet was washed
with 1 mL of deionized and distilled water (DDW) and centrifuged
again in the same manner to ensure a maximal removal of residual
medium. The supernatant was carefully removed, and the pellet
was resuspended in 100 lL of DDW. Total DNA in the suspension
was immediately extracted using an automated nucleic acid
extractor (Magtration System 6GC, PSS, Chiba, Japan). Purified
DNA was eluted with 100 lL of Tris–HCl buffer (pH 8.0) and stored
at ?20 ?C for further analyses.
2.4. Denaturing gradient gel electrophoresis (DGGE) analysis
PCR-DGGE was performed as described previously (Chen and
LaPara, 2008). Briefly, the variable V3 region of the 16S rDNA gene
from members of the domain Bacteria was amplified by PCR using
primers 338f and 518r (Muyzer et al., 1993), with a GC-clamp at-
tached to the forward primer. DGGE was performed using a D-code
system (BioRad, Hercules, CA, USA). Samples containing equal
amounts of PCR amplicons were loaded on an 8% (w/v) acrylamide
gel containing 30–70% denaturant gradient, where 100% was de-
fined as 7 M urea with 40% (v/v) formamide. Following electropho-
resis, the gel was stained with ethidium bromide and then scanned
under UV illumination. DNA bands were excised directly from the
gel and eluted in 40 lL of sterilized DW. The eluted solution was
amplified by using 338f and 518r primers without attaching a
GC-clamp. These PCR products were purified using the PCR purifi-
cation kit (Bioneer, Korea) prior to nucleotide sequence determina-
tion. The sequencing results, determined fully in both directions
for each purified PCR-DGGE band using 338f and 518r as sequenc-
ing primer, were compared with the reference sequences in the
GenBank database.
2.5. T-RFLP analysis
T-RFLP was used to analyze the nitrifying bacteria community
in the pre-denitrification process reactor based on the known
16S rDNA genes of AOB and NOB as described in a previous study
(Siripong and Rittmann, 2007). Considering the low concentration
of DNA from the nitrifiers, we amplified DNA from the 16S rDNA
gene of AOB and NOB through nested PCR, using the universal
primers 11f and 1492r (Table 1) to achieve an initial increase in
template concentration, followed by the specific amplification of
nitrifier genes (Nitrifier-specific reverse primer: Nso1225r, NIT3r,
Ntspa685r, Forward primer: Eub338f included phosphoramidite
dye 6-FAM (Table 1)) (Siripong and Rittmann, 2007).
After the first round of universal amplification of the 16S rDNA
gene, we purified PCR products using the PCR purification kit
914
Y.M. Kim et al./Bioresource Technology 102 (2011) 913–922

Page 3

(Bioneer, Korea) to eliminate the universal primers so that they
would not interfere with the next round of amplification. We used
2 lL of template DNA for the universal amplification step and 1 lL
of universal amplification product as the template for the nitrifier-
specific amplification. The thermal profile used for the universal
amplification was 5 min at 95 ?C; 35 cycles of 30 s at 95 ?C, 30 s
at 55 ?C, and 45 s at 72 ?C; and a final elongation for 10 min at
72 ?C. The thermal profile used for the nitrifier-specific amplifica-
tion was 5 min at 95 ?C; 35 cycles of 90 s at 95 ?C, 30 s at 60, and
90 s at 72 ?C; and a final elongation for 10 min at 72 ?C (Siripong
and Rittmann, 2007). Finally, we purified the PCR products again
and digested 16S rDNA-gene amplicons with MspI restriction
endonuclease at 37 ?C for 3 h. The restriction digestion mixture
contained 9 lL of purified PCR product, 1 lL of enzyme buffer,
and 1 lL of 10U of restriction endonuclease (Siripong and
Rittmann, 2007). Digested PCR products were run through an ABI
3130XL Genetic Analyzer (Applied Biosystems, Forster City, CA)
at the SolGent Company (Korea). The peak results were analyzed
using the Peak Scanner software v 1.0 (Applied Biosystems, Foster
City, CA).
2.6. qPCR analysis
To investigate the changes in the nitrifying bacteria population
according to variation of the process performance, four indepen-
dent qPCR assays were conducted by quantifying total bacterial
16S rDNA, ammonia oxidizing bacterial 16S rDNA, Nitrospira spp.
16S rDNA, and Nitrobacter spp. 16S rDNA (Table 1). All qPCR assays
were performed using a 7300 Real-Time PCR system (Applied Bio-
systems, CA) in 25 lL reaction capillary tubes. Each capillary tube
was separately loaded with 2 lL of template DNA, followed by 1
lL of the forward and reverse primers, together with 0.5 lL of
the TaqMan probe corresponding to each primer and probe set,
12.5 lL of TaqMan Universal PCR Master Mix (PE Applied Biosys-
tems), and PCR-grade sterile water to a final volume of 25 lL.
The amount of total bacterial 16S rDNA was amplified using pri-
mer 1055f and 1392r (Ferris et al., 1996). The TaqMan probe
16STaq1115 was modified by the 1114f primer (Harms et al.,
2003). The PCR program was 2 min at 50 ?C, 10 min at 95 ?C; 45 cy-
cles of 30 s at 95 ?C, 60 s at 50 ?C, and 40 s at 72 ?C. To determined
the amount of AOB 16S rDNA genes, two forward primers CTO
189A/B and CTO 189C, one reverse RT1r, and the TaqMan probe
TMP1 were used as described previously by Hermansson and
Lindgren (2001). The PCR program for AOB 16S rDNA quantifica-
tion included 2 min at 50 ?C, 10 min at 95 ?C; 40 cycles of 30 s at
95 ?C, 60 s at 60 ?C. The Nitrospira spp. 16S rDNA primers NSR
1113f and NSR 1264r (Dionisi et al., 2002) were tested using geno-
mic DNA extracted from the activated sludge as templates. The
TaqMan probe NSR 1143Taq was derived from a conserved se-
quence region between the primers NSR 1113f and NSR 1264r
(Harms et al., 2003). PCR amplification consisted of 2 min at
50 ?C, 10 min at 95 ?C; 50 cycles of 30 s at 95 ?C, 60 s at 60 ?C).
Lastly, the amount of Nitrobacter spp. from Graham et al. (2007)
was amplified using primer Nitro 1198f /Nitro 1423r and TaqMan
probe Nitro 1374Taq. The program used for amplification was
2 min at 50 ?C, 10 min at 95 ?C; 50 cycles of 20 s at 94 ?C, 60 s at
58 ?C, and 40 s at 72 ?C.
All experiments were performed in duplicate per sample and all
PCR runs included control reactions without the template. The
gene copy numbers were calculated through a comparison of
threshold cycles obtained in each PCR run with those of known
standard DNA concentrations.
2.7. Analytical methods
The collected samples were centrifuged at 3500 rpm for 3 min
(MF550, HANIL), and then the supernatants were used for the fol-
lowing analyses. According to the standard method (APHA, 1998),
COD was analyzed using the colorimetric method; ammonia with
the phenate method; phenol with the chloroform extraction meth-
od; and SCN?with reaction with ferric nitrate using a spectropho-
tometer (GENESYS TM 5, Spectronic Inc.). The total cyanide
concentration was determined by the pyridine-pyrazolone method
after distillation. Nitrite and nitrate concentrations were measured
Table 1
Primers and probes used in this study.
Target Primer/probeSequence (50–30) References
For DGGE
Bacterial 16S rDNA338f
518r
50-(GC clamp)-ACTCCTACGGGAGGCAGCAG-30
50-ATTACCGCGGCTGCTGG-30
Muyzer et al. (1993)
For T-RFLP
Bacterial I6S rDNA11f
1492r
Eub 338f
Nso 1225r
NIT 3r
Ntspa 685r
50-GTTTGATCCTGGCTCAG-30
50-TACCTTGTTACGACTT-30
50-(6-FAM)-ACTCCTACGGGAGGCAGC-30
50-CGCCATTGTATTACGTGTGA-30
50-CCTGTGCTCCATGCTCCG-30
J0-CGGGAATTCCGCGCTC-30
Kane et al. (1993)
Lin and Stahl (1995)
Amann et al. (1990)
Mobarry et al. (1996)
Wagner et al. (1995)
Regan et al. (2002)
Bacterial 16S rDNA
AOB 16S rDNA
Nitrobacter 16S rDNA
Nitrospira 16S rDNA
For qPCR
Bacterial 16S rDNA1055f
1392r
16STaql 115
50-ATGGCTGTCGTCAGCT-30
50-ACGGGCGGTGTGTAC-30
50-(6-FAM)-CAACGAGCGCAACCC-(TAMRA)-30
Ferris et al. (1996)
Harms et al. (2003)
AOB 16S rDNA CTO 189fA/Ba
CTO 1891Ca
RTlr
TMP1
50-GGAGRAAAGCAGGGGATCG-30
50-GGAGGAAAGTAGGGGATCG-30
50-CGTCCTCTCAGACCARCTACTG-30
50-(6-FAM)-CAACTAGCTAATCAGRCATCRGCCGCT-(TAMRA)-30
Hermansson and Lindgren (2001)
Nitrospira spp. 16S rDNANSR 1113f
NSR 1264r
NSR 1143Taq
50-CCTGCTTTCAGTTGCTACCG-3,
50-GTTTGCAGCGCTTTGTACCG-30
50-(6-FAM)-AGCACTCTGAAAGGACTGCCCAGG-(TAMRA)-30
Dionisi et al. (2002)
Harms et al. 2003
Nitrobacter spp. 16S rDNANitro 1198f
Nitro 1423r
Nitro 1374Taq
50-ACCCCTAGCAAATCTCAAAAAACCG-30
50-CTTCACCCCAGTCGCTGACC-30
50-(6-FAM)-AACCCGCAAGGAGGCAGCCGACC-(TAMRA)-30
Graham et al. (2007)
aA mixture of CTO 189fA/B and CTO 189fC at the weight ratio of 2:1 was used as the forward primer.
Y.M. Kim et al./Bioresource Technology 102 (2011) 913–922
915

Page 4

with an ion chromatograph (ICS-1000, DIONEX Co.) (Kim et al.,
2008a). TOC, inorganic carbon (IC), and TN were analyzed with a
TOC analyzer and a TN measuring unit (TOC-V csu, TNM-1,
SHIMADZU).
3. Results and discussion
3.1. Effect of increased thiocyanate concentration on the performance
of the pre-denitrification process
Industrial wastewater generated from steel or iron making fac-
tories usually contains a significant amount of thiocyanate, which
at high concentration can negatively affect the pre-denitrification
process (Kim et al., 2009). In addition, an abrupt increase in SCN?
loading sometimes causes unstable performance of the full-scale
treatment process, resulting specifically in poor removal efficiency
of TN. Therefore, the effect of shock loading SCN?on removal
behavior and on other pollutants was investigated in the lab-scale
reactor. The influent concentration of SCN?was artificially in-
creased from 200 to 800 mg/L. In spite of the diluting effect of
the pre-denitrification process due to a low internal recycling ratio
of 2, the SCN?concentration flowing into the aerobic reactor also
increased, from 62 to 259 mg/L, during the shock loading (Fig. 1).
However, the removal efficiency of SCN?was always higher than
97%, regardless of its influent concentration. Previously, SCN?
degradation was not inhibited by a high SCN?concentration of
500 mg/L when the residence time was longer than 10 h (Lay-Son
and Drakides, 2008). Meanwhile, the sulfur performed in the aero-
bic reactor with SCN?shock loading showed that DSCN-S/DSO4-S
was about 1.0. The SO2?
4
by product production and sulfur and
nitrogen balances indicated that the following reaction occurred
with the activity of chemolithotrophic organisms (Staib and Lant,
2007):
SCN?þ 2H2O þ 2O2! CO2þ SO2?
4þ NH4þ
Furthermore, this SCN?biodegradation reaction only occurred
in the aerobic condition; SCN?in the anoxic condition was not bio-
degraded by the activated sludge. This result was additionally con-
firmed by the batch assays (data not shown).
This study determined a maximum specific removal rate (VSCN)
of 0.08 g SCN/gVSS d in the aerobic reactor. The values for VSCNin-
creased from 0.02 to 0.073 with the increase of the influent con-
centration of SCN?(Table 2). However, these values were lower
than the value for VSCNof 0.15 g SCN/gVSS d reported by Staib
and Lant (2007) in their study on real cokes wastewater due to
high MLVSS concentration and long hydraulic retention time of
the aerobic reactor. Meanwhile, there was a linear relationship be-
tween the VSCNand the initial SCN?concentration. This means that
SCN?is not self-inhibitory at the concentrations investigated (Staib
and Lant, 2007).
Influent concentrations of COD and TOC increased in ratios of
about 1.4 mg-COD/mg- SCN?and 0.21 mg-TOC/mg-SCN?with
Time (day)
0 1020 3040 50 607080
Thiocyanate concentration (mg/L)
0
100
200
300
400
500
600
700
800
900
Removal efficiency (%)
0
10
20
30
40
50
60
70
80
90
100
Internal recycling ratio
0
1
2
3
4
5
Internal recycling ratio
Removal efficiency
Influent SCN-
Anoxic effluent SCN-
Aerobic effluent SCN-
Fig. 1. Influent- and effluent-concentrations and final removal efficiencies of SCN?in the lab-scale pre-denitrification process during the shock-loading of SCN?.
Table 2
Performance of the SCN?, nitrification, and denitrification obtained in the pre-denitrification process under different operating conditions.
Operating conditiona
SCN?loading rate
(g-SCN?/L-d)
Specific SCN?removal
rate (g-SCN?/g-VSS-d)
NHþ
rateb(g-N/L-d)
4-N loading
Specific nitrification
rate (g-N/g-VSS-d)
NO?
rate (g-N/L-d)
3-N loading Specific denitrification
rate (g-N/g-VSS-d)
Aerobic reactor
0.072
0.129
0.197
0.257
0.260
0.267
0.278
Aerobic reactor
0.064
0.083
0.107
0.122
0.126
0.131
0.129
Anoxic reactor
0.095
0.123
0.146
0.177
0.190
0.187
0.179
12 day: SCN 200 + 2Q
23 day: SCN 400 + 2Q
36 day: SCN 600 + 2Q
48 day: SCN 800 + 2Q
59 day: SCN 800 + 3Q
73 day: SCN 800 + 4Q
84 day: SCN 800 + 5Q
0.020
0.036
0.055
0.073
0.074
0.077
0.080
0.018
0.024
0.030
0.035
0.036
0.037
0.037
0.027
0.035
0.042
0.051
0.055
0.054
0.052
aSCN (number): Influent SCN- concentration; Q: Internal recycling ratio (L/day).
bCalculated based on influent NHþ
centration of aerobic reactor].
4-N concentration of [0.24-concentration of biodegraded SCN-(influent SCN?–effluent SCN?) in aerobic reactor + influent NHþ
4-N con-
916
Y.M. Kim et al./Bioresource Technology 102 (2011) 913–922

Page 5

increasing SCN?concentration (Fig. 2(a)). During the shock loading
of SCN?, the removal efficiencies of COD and TOC increased
slightly, from 88% to 91% and from 87% to 91%, respectively. This
was due to the almost complete biodegradation of v in the aerobic
reactor and the increased loading of nitrate into the anoxic reactor.
The removal behavior of phenol was not affected by the shock
loading of SCN?(Fig. 2(b)). Phenol in the raw wastewater ranged
between 226 and 267 mg/L, and the removal efficiency was always
maintained at higher than 98%. The aerobic degradation of phenol
is known to be fast in an activated sludge system, so its degrada-
tion may not be inhibited by SCN?itself (Staib and Lant, 2007). Fi-
nally, the discharge concentrations of COD, TOC, and phenol were
175–325, 52–95, and 1–3 mg/L, respectively. This is to say, no neg-
ative effects of SCN?shock loading on the removal of carbon pollu-
tants were observed in this study.
Although the SCN?was mostly removed in the aerobic reactor,
its shock loading caused negative side effects on the removal
behavior of nitrogen pollutants. About 55 mg-N/L of ammonia con-
sistently flowed into the aerobic reactor under SCN?shock loading;
this was almost completely nitrified to nitrate (Fig. 3). However,
the SCN?shock loading caused increased ammonia generation in
the aerobic reactor, where its biodegradation also produced 30 to
50 mg-N/L of ammonia. In spite of the increased ammonia concen-
tration in the aerobic reactor, more than 97% nitrification efficiency
was consistently achieved up to the high concentration of SCN?
(800 mg/L). Although previous studies reported that 5–25 mg/L of
sulfate inhibited a nitrification reaction in activated sludge ob-
tained from a municipal wastewater treatment and high-strength
sulfate caused toxicity problems during the anaerobic treatment
process (Kuo and Shu, 2004;), no effect of 813–2325 mg/L sulfate
formed on simultaneous nitrogen removal could be found in this
batch and continuous study (data not shown). This was because
the sampled activated sludge had been fully adapted to a high con-
centration of sulfate for a long term operation. Other research
groups also reported no inhibitory effects of sulfate on COD, SCN?,
ammonia, nitrate, and cyanide removal (Jeong and Chung, 2006).
That is, the degradation of the SCN?does not have an immediate
effect on the underlying rate of nitrification (Staib and Lant,
2007). This is supported by the rates of nitrification shown in Table
2. Throughout the operation of the system, NO?
low 0.1 mg-N/L and NO?
L to 122 mg-N/L. The denitrification reaction was never inhibited
by the shock loading of SCN?(Fig. 3(b)) and no sulfate reduction
occurred in the anoxic reactor (data not shown). However, the final
discharge concentration of TN gradually increased from 64 up to
123 mg-N/L (data not shown). This result was due to the increased
concentration of nitrate discharged from the aerobic reactor. More
than 95% of total nitrogen in the final effluent was in the form of
2was detected be-
3concentration increased from 58 mg-N/
COD concentration (mg/L)
0
500
1000
1500
2000
2500
3000
Removal efficiency (%)
0
100
10
20
30
40
50
60
70
80
90
100
Internal recycling ratio
0
1
2
3
4
5
Thiocyanate loading concentration (mg/L)
0
100
200
300
400
500
600
700
800
900
Influent COD
Anoxic effluent COD
Aerobic effluent COD
Removal efficiency
Time (day)
0 10 2030405060 7080
Phenol concentration (mg/L)
0
50
100
150
200
250
300
Removal efficiency (%)
0
10
20
30
40
50
60
70
80
90
Internal recycling ratio
0
1
2
3
4
5
Thiocyanate loading concentration (mg/L)
0
100
200
300
400
500
600
700
800
900
Internal recycling ratio
Thiocyanate loading
Influent phenol
Aerobic effluent phenol
Anoxic effluent phenol
Removal efficiency
(b)
(a)
Fig. 2. Influent- and effluent-concentrations and final removal efficiencies of (a) COD, (b) phenol, in the lab-scale pre-denitrification process during the shock loading of SCN?.
Y.M. Kim et al./Bioresource Technology 102 (2011) 913–922
917

Page 6

NO?
effects, the SCN?shock loading brought about an outflow of TN
higher than the legal discharge level of 60 mg-N/L.
3-N generated by nitrification. To sum up, due to negative side
3.2. Effect of internal recycling ratio on the performance of the
pre-denitrification process
The performance of the reactor under a SCN?concentration
loading of 800 mg/L was tested with an internal recycling ratio that
varied between 2 and 5 while the feed flow rate was maintained at
a constant 5.76 L/d of raw wastewater.
The effect of the recycling ratio0s variation on overall carbon re-
moval was insignificant. However, owing to the enhanced dilution
effect of increased recycling ratios, the concentration of carbon
pollutant flowing into the aerobic reactor gradually decreased
(Fig. 2). In addition, the denitrification reaction in the pre-denitri-
fication process was enhanced by the increased nitrate load (Table
2, Fig. 3(b)), resulting in an increased carbon removal rate in the
anoxic reactor. In the case of phenol biodegradation in the anoxic
reactor, the removal rate increased from 0.016 to 0.029 g/g-VSS?d.
Meanwhile, the increased recycling ratio did not improve
the total removal efficiencies of SCN?and ammonia. But the
concentrations of SCN?and ammonia in the anoxic reactor effluent
did decrease from 240 to 130 mg/L and from 55 to 30 mg-N/L,
respectively. Due to the decreased loading of the nitrogen pollu-
tant, a lower concentration of nitrate was produced in the aerobic
reactor (Fig. 3(b)). As a result, the TN removal efficiency improved
to 85% with an increase in the internal recycling ratio. In addition,
during the inflow of the highest concentration of SCN?(800 mg/L),
the pre-denitrification process with an internal recycling ratio of 5
was able to achieve a discharge level below 50 mg-N/L of TN: less
than the legal discharge level. Increasing the internal recycling ra-
tio for any SCN?concentration in the influent resulted in an in-
crease in the TN removal. Previous studies also improved the
nitrogen removal efficiency with an increase in internal recycling
ratio (Baeza et al., 2004). The maximum increment in TN removal
efficiency (65–73%) was achieved with an increase in internal recy-
cling ratio from 2 to 3. A subsequent increase in the internal recy-
cling ratio to 4 and 5 caused further increases in TN removal
efficiency by 7% and 5%, respectively. Hence, the total improve-
ment achieved when the recycling ratio was increased from 2 to
5 was 20%. This variation pattern of TN removal efficiency accord-
ing to an increase in internal recycling ratio was similar to that of
the NO?
3-N concentration in the effluent, since more than 95% of TN
Aerobic effluent thiocyanate
(a)
Ammonia concentration (mg-N/L)
0
20
40
60
80
100
120
140
Removal efficiency (%)
0
100
10
20
30
40
50
60
70
80
90
100
Internal recycling ratio
0
1
2
3
4
5
Thiocyanate loading concentration (mg/L)
0
100
200
300
400
500
600
700
800
900
Internal recycle ratio
Thiocyanate loading
Influent ammonia
Anoxic effluent ammonia
Aerobic effluent ammonia
(b)
Removal efficiency
Time (day)
010203040 50607080
Nitrogen concentration (mg-N/L)
0
20
40
60
80
100
120
140
Removal efficiency (%)
0
10
20
30
40
50
60
70
80
90
Internal recycling ratio
0
1
2
3
4
5
Thiocyanate loading concentration (mg/L)
0
100
200
300
400
500
600
700
800
900
Removal efficiency
Anoxic effluent NO3
-
Aerobic effluent NO3
-
Fig. 3. Influent- and effluent-concentrations and final removal efficiencies of (a) ammonia, (b) nitrate, in the lab-scale pre-denitrification process during the shock-loading
of SCN?.
918
Y.M. Kim et al./Bioresource Technology 102 (2011) 913–922

Page 7

in the final effluent was in the form of NO?
loading, the 123 mg-N/L of effluent NO?
ratio of 2 decreased to 83 mg-N/L at an internal recycling ratio of 3.
It further decreased to 49 mg-N/L at an internal recycling ratio of 5.
The main reason for these results was that at a higher recycling
ratio, a greater nitrate load was supplied to the anoxic reactor for
denitrification; this reduced the nitrate ion fraction in the effluent
and enhanced TN removal in the pre-denitrification process sys-
tem. Consequently, an increased internal recycling ratio improved
the homogeneous distribution of microbial communities in the
process, increased the COD removal rate by denitrifying bacteria,
and was important in avoiding possible process inhibition by a
high concentration inflow of nitrogen and carbon pollutants (Baeza
et al., 2004).
The specific nitrification rate versus influent SCN?concentra-
tion and internal recycling ratio was investigated in this study.
These nitrification rates, based on ammonium consumed, were cal-
culated on the basis of additional ammonium generated from SCN?
biodegraded in the aerobic reactor. It can be observed that the
nitrification rate increased when the influent SCN?concentration
increased at each internal recycling ratio examined. This increase
was higher when the influent concentration rose from 400 to
600 mg/L than when it rose from 600 to 800 mg/L. This indicates
that the system was operating under conditions in which the spe-
cific nitrification rate was near maximal. A higher nitrification rate
of around 0.036–0.037 g-N/g-VSS d was observed in the case of
internal recycling ratios of 3 and 5 and an SCN?influent concentra-
tion of 800 mg/L, and was more dependent on the influent concen-
tration. An increase in nitrification rate was observed when the
internal recycling ratio increased, but it was not observed to be
an optimal value for this operational parameter. To sum up, the ef-
fect of the higher internal recycling ratio was very significant for
TN removal and helped to improve the process performance when
a high load of SCN?was applied. A further increase of internal recy-
cling ratio will continue to improve TN removal efficiency,
although this implies a higher energy cost. Finally, a higher internal
recycling ratio for a pre-denitrification process treating industrial
3-N. At 800 mg/L of SCN?
3-N at an internal recycling
wastewater when a higher load of SCN?is detected should be used
in order to meet regulations.
3.3. Response of the nitrifying bacteria community to changes in
thiocyanate concentration
The consecutive samples were analyzed at steady state using
DGGE to monitor and identify the bacteria in the system. DGGE
analysis of the PCR products amplified using the bacterial universal
primer pair (i.e., 338f and 518r) showed the presence of 16 bands
(Fig. 4). Five major bands (2, 4, 7, 8, and 15) were predominant in
all lanes. This indicates that the microorganisms corresponding to
those bands would be mainly responsible for this wastewater
treatment in all operating situations. The affiliations of the 16S
rDNA sequences were determined by comparison against the Gen-
Bank database, as shown in Fig. 4. Band 7 was most closely related
(97% similarity) to an uncultured bacterium found in a chlorophe-
nol-degrading fluidized bed bioreactor. Band 8 was most closely
related (97% similarity) to an uncultured bacterium found in petro-
leum-contaminated sediments. Therefore, these microorganisms
corresponding to band 7 and band 8 were assumed to be related
to degrading organic compounds. Band 15 showed 97% similarity
to the Thiobacillus thioparus strain. This bacterium oxidizes
SCN?to carbon dioxide, ammonia, and sulfate and assimilates this
substance as an energy source for autotrophic growth. The pres-
ence of the T. thioparus in the system indicates that this organism
had a significant role in SCN?degradation under all operation con-
ditions tested and also gives evidence of the chemolithotrophic
organism reaction referred to Section 3.1.
Among the other bands with relatively weak intensities, band 1
was closely related (97% similarity) to an uncultured Thauera strain
found in a petroleum refinery wastewater bioreactor with a high
phenolic load. The Thauera-like microorganism corresponding to
band 1 was likely responsible for degrading toluene contained in
cokes wastewater (Shinoda et al., 2004). Bands 3, 6, 10, and 14
were most closely related to uncultured bacteria participating in
ammonia oxidation. In particular, band 10 showed 97% similarity
Fig. 4. DGGE profiles of the PCR products amplified with 16S rDNA gene primers.
Y.M. Kim et al./Bioresource Technology 102 (2011) 913–922
919

Page 8

to Nitrosomonas nitrosa Nm90. This indicates that N. nitrosa Nm90
was predominant in this process treating high concentrations of
SCN?. This AOB has been found in other industrial WWTPs (Layton
et al., 2005). Therefore, the occurrence of N. nitrosa Nm90 in a pro-
cess fed with high strength industrial wastewater is consistent
with previous studies. Band 11 was most closely related to an
uncultured bacterium found in the microbial community of an A2
reactor in a lab-scale A1–A2–O fixed biofilm system for coking
wastewater treatment.
The putative functions and habitats of the microorganisms de-
duced from the DGGE bands were in agreement with our pre-deni-
trification reactor condition treating cokes wastewater containing
high concentrations of SCN?. However, a few bands were most clo-
sely related to a bacterium whose existence and roles in this pro-
cess are unclear. Among the 16 band sequences, 14 were most
closely related to the sequences of uncultured strains, and four
showed low similarities (less than 97%) to database sequences. A
previous study conducted with a SCN?degrading reactor condition
also showed a similar function and habitats of the microorganisms
deduced from the DGGE bands (Lee et al., 2008).
We determined that the nitrifying bacterial communities pres-
ent in the cokes wastewater treatment process contained high-
strength SCN?using T-RFLP specifically designed to identify AOB
and NOB with terminal fragment (TF) lengths (Regan et al., 2002).
Figure 5 shows electropherograms of AOB, Nitrobacter-specific
NOB, and Nitrospira-specific NOB, respectively, according to the
variation of SCN?concentration and the internal recycling ratio.
In Fig. 5(b), AOB-targeted T-RFLP allowed us to differentiate be-
tween AOB groups. All samples from each different operating con-
dition showed a peak at 164 bp, a signature peak of Nitrosomonas
europaea/eutropha and Nitrosomonas marina lineage. Because the
influents are from industrial wastewater, marine AOB species are
not relevant. Besides the major peak at 164 bp, we also detected a
small peak at 276 bp, which represents the potential presence of
N. europaea/eutropha, N. oligotropha, N. cryotolerans, or N. communis
lineage (Fig 5(a)).
To obtain a finer resolution in the AOB community present in
the process, AOB 16S rDNA gene based cloning and sequencing
was performed using the AOB-target primer (Nso1225r and
Eub338f) without the fluorescent dye (Table 1). Most of the AOB
clones from the reactor were closely associated with N. europaea
in the N. europaea/eutropha lineage and N. nitrosa in the N. commu-
nis lineage, but the AOB clone related to Nitrosospira lineage was
not detected.
As a result, through the 16S rDNA gene sequences, the microor-
ganisms corresponding to the peaks at 164 and 276 bp could be
identified as N. europaea and N. nitrosa, respectively (Fig. 5(a)).
Meanwhile, the TF corresponding to the peak at 92 bp was not
identified from the clones of the AOB 16S rDNA gene in this study
and does not belong to any AOB in the database. Thus, the high
peak at 164 bp implies the dominance of the N. europaea of AOB
in this wastewater treatment system, irrespective of the varia-
tion of SCN?loading rate; N. nitrosa was a minor population.
N. europaea has been widely observed in WWTPs (Lydmark et al.,
2007; Siripong and Rittmann, 2007), and N. nitrosa has previously
been detected in activated sludge treating industrial wastewater
(Layton et al., 2005). A competitive dominance between the
two AOB species was reported in previous study (Lim et al.,
2008). In an industrial wastewater treatment process controlling
high strength nitrogen, N. europaea had a specific growth rate
as high as 0.76 day?1, while N. nitrosa had a specific growth rate
of 0.34 day?1(Lim et al., 2008). Perhaps the dominance of the
N. europaea in this study is due to its high specific growth rate
compared to other AOB lineages.
Based on Nitrobacter-specific T-RFLP, Fig. 5(c) shows a promi-
nent peak at 137 bp, which belongs to Nitrobacter species. We also
Fig. 5. (a) Expected TF sizes and their corresponding AOB and NOB groups based on T-RFLP of 16S rDNA gene (Siripong and Rittmann, 2007), T-RFLP profiles of (b) AOB,
(c) Nitrobacter, (d) Nitrospira in the lab-scale pre-denitrification process during the shock loading of SCN?.
920
Y.M. Kim et al./Bioresource Technology 102 (2011) 913–922

Page 9

found TF sizes at 158, 164, 208, 273, and 392 in the samples. These
unexpected peaks could be a result of an incomplete digestion,
uncharacterized Nitrobacter species, or imperfectly matched pri-
mer (Siripong and Rittmann, 2007). The results of Nitrospira-
specific T-RFLP show two dominant peaks at 272 and 334 bp in
all samples (Fig. 5(d)). The peak at 272 corresponds to several
Nitrospira clones in the database. The 334 TF belongs to one of
the Nitrospira moscoviensis strains (Siripong and Rittmann, 2007).
The Nitrospira clones corresponding to the peak at 272 bp were a
consistently dominant population under the variation of SCN?
loading, but the dominance between the two Nitrospira clusters
in the reactor shifted to N. moscoviensis when the internal recycling
ratio increased to 5 at the influent concentration of 800 mg/L SCN?.
The lower nitrite concentration in the influent or shorter residence
time might have an influence on this dominance shift.
Figure 6 shows the changes in the 16S rDNA gene cell numbers
for the total bacteria, AOB, Nitrobacter, and Nitrospira, quantified
using qPCR assays in the aerobic reactor of the pre-denitrification
process of treating industrial wastewater with a high concentra-
tion of SCN?. The number of total bacteria and nitrifying bacteria
cells per liter were calculated from copies per liter using assump-
tions. The average number of 16S rDNA copies per genome in bac-
terial cells was assumed to be 3.6 based on the average found in
cultured bacteria (Klappenbach et al., 2001). AOB, Nitrospira, and
Nitrobacter were assumed to contain 1 copy 16S rDNA per cell
based on copies found in Nitrobacter and the AOB Nitrosomonas
and Nitrosospira (Aakra et al., 1999). In all samples, the total bacte-
rial population in the aerobic reactor ranged from 4.1 ? 1012to
6.1 ? 1012cells/L and remained constant during the shock loading
of SCN?. These values are in the same order of magnitude as those
obtained from activated sludge samples from industrial WWTPs
(Layton et al., 2005).
The concentration of AOB determined using the AOB 16S rDNA
assay was gradually increased with the increase of SCN?concen-
tration in the aerobic reactor. As the ammonia nitrogen loading
into the aerobic reactor increased to 2 times, an approximately
3-fold increase was observed in the number of AOB 16S cells/L than
in the initial AOB 16S cells/L. At an internal recycling ratio between
3 and 5 with 800 mg/L SCN?loading, however, the AOB population
was stable due to the consistent ammonia nitrogen loading rate.
The percentages of the AOB population within total bacteria also
increased from about 4% to 17% in the aerobic reactor. This result
is higher than the values for activated sludge samples obtained
from systems treating industrial wastewater (0.01–9.3%) (Layton
et al., 2005). This means that about 200 mg/L of SCN?concentra-
tion inflow into the aerobic reactor did not inhibit nitrification
activity in the activated sludge system (Kim et al., 2008a), but al-
lowed the AOB population and specific nitrification rate to increase
through the additional ammonia supply by the biodegradation of
SCN?in the aerobic reactor.
We also observed coexisting Nitrospira and Nitrobacter genera
for NOB. The Nitrospira and Nitrobacter populations in the initial
operating condition were similar, at 2.7 ? 1010and 3.4 ? 1010
cells/L, respectively. However, a shift in the NOB community was
observed as the SCN?loading increased.
The 16S rDNA gene concentration of the Nitrospira increased to
from 2.7 ? 1010to 4.6 ? 1011cells/L and the percentages of the
Nitrospira population within the total bacteria also sharply in-
creased from 0.45% to 9% in the aerobic reactor. Interestingly, the
16S rDNA gene concentration of the Nitrospira bacteria present in
this study was partially higher in concentrations compared to the
AOB at an SCN?loading of between 600 and 800 mg/L (Fig. 6). Pre-
viously, many studies have found that Nitrospira populations in
nitrifying systems were equal to or even higher than the AOB pop-
ulations (Dionisi et al., 2002; Harms et al., 2003). Similarly, as the
internal recycling ratio increased under 800 mg/L SCN?loading,
the Nitrospira population no longer increased, as in the AOB popu-
lation pattern.
On the other hand, the number of Nitrobacter decreased slightly
from 3.7 ? 1010to 1.5 ? 1010cells/L during the increase of the
Nitrospira population and when the Nitrospira population was sta-
ble, the Nitrobacter population gradually increased again. However,
the percentages of the Nitrobacter population (0.33–0.89%) within
total bacteria were even smaller compared to the Nitrospira per-
centages. Although Nitrobacter have been reported to outcompete
Nitrospira at high nitrogen loading (Nogueira et al., 2002), a higher
amount of Nitrospira than Nitrobacter was observed in this aerobic
reactor. Our finding correlates with the recent expectation that
Nitrospira rather than Nitrobacter would generally dominate
WWTPs due to their better nitrite affinity (Blackburne et al., 2007).
4. Conclusions
The performance and nitrifying bacterial communities under
SCN?shock loading were characterized in pre-denitrification pro-
cess. The SCN?shock loading caused negative side effects on the
removal of TN, but the higher internal recycling ratio helped to im-
prove the process performance. During the operation, the percent-
ages of the AOB population within the total bacteria increased, and
a higher amount of Nitrospira than Nitrobacter was consistently ob-
served in the aerobic reactor. Meanwhile, a T. thioparus detected by
a DGGE band was suggested to be the main microorganism respon-
sible for SCN?biodegradation observed in the system.
Acknowledgements
This work was supported by the Korea Student Aid Foundation
(KOSAF) (S2-2009-000- 01664-1) and by Basic Science Research
Program through the National Research Foundation of Korea
(NRF) funded by the Ministry of Education, Science and Technology
(No 2010-0001437). This work was also partially supported by the
second phase of the Brain Korea 21 Program in 2010 as well as by
the Priority Research Centers Program through the National
Research Foundation of Korea (NRF) funded by the MEST (2009-
0093819). The authors thank Steven K. Lustig, Dr. Hong Soon Rhee,
MinJi Kim, and Kyung-Jin Cho for assistance during this work.
Time (day)
0 1020 3040 5060 708090
Cells/L
1010
1011
1012
1013
Total bacteria
AOB
Nitrospira (NOB-1)
Nitrobacter (NOB-2)
Fig. 6. Changes in the 16S rDNA gene cells per liter of the total bacteria, AOB,
Nitrobacter, and Nitrospira in the pre-denitrification process during the shock
loading of SCN?.
Y.M. Kim et al./Bioresource Technology 102 (2011) 913–922
921

Page 10

References
Aakra, A., Utaker, J.B., Nes, I.F., 1999. RFLP of rRNA genes and sequencing of the
16S–23S rDNA intergenic spacer region of ammonia-oxidizing bacteria: a
phylogenetic approach. Int. J. Syst. Bacteriol. 49, 123–130.
Amann, R.I., Binder, R.J., Olson, S., Chisholm, S.W., Devereux, R., Stahl, D.A., 1990.
Combination of 16S ribosomal RNA targeted oligonucleotide probes with flow
cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol.
56, 1919–1925.
APHA, 1998. Standard Methods for the Examination of Water and Wastewater, 20th
ed. APHA, AWWA, WPCF. American Public Health Association, Washington, DC,
USA.
Baeza, J.A., Gabriel, D., Lafuente, J., 2004. Effect of internal recycle on the nitrogen
removal efficiency of an anaerobic/anoxic/oxic wastewater treatment plant.
Process Biochem. 39, 1615–1624.
Blackburne, R., Vadivelu, V.M., Yuan, Z.G., Keller, J., 2007. Kinetic characterisation of
an enriched Nitrospira culture with comparison to Nitrobacter. Water Res. 41,
3033–3042.
Chen, R.D., LaPara, T.M., 2008. Enrichment of dense nitrifying bacterial communities
in membrane-coupled bioreactors. Process Biochem. 43, 33–41.
Dionisi, H.M., Layton, A.C., Harms, G., Gregory, I.R., Robinson, K.G., Sayler, G.S., 2002.
Quantification of Nitrosomonas oligotropha-like ammonia-oxidizing bacteria
and Nitrospira spp. from full-scale wastewater treatment plants by competitive
PCR. Appl. Environ. Microbiol. 68, 245–253.
Ferris, M.J., Muyzer, G., Ward, D.M., 1996. Denaturing gradient gel electrophoresis
profiles of 16S rRNA-defined populations inhabiting a hot spring microbial mat
community. Appl. Environ. Microbiol. 62, 340–346.
Graham, D.W., Knapp, C.W., Van Vleck, E.S., Bloor, K., Lane, T., Graham, C.E., 2007.
Experimental demonstration of chaotic instability in biological nitrification.
ISME J. 1, 385–394.
Harms, G., Layton, A.C., Dionisi, H.M., Gregory, I.R., Garrett, V.M., Hawkins, S.A.,
Robinson, K.G., Sayler, G.S., 2003. Real-time PCR quantification of nitrifying
bacteria in a municipal wastewater treatment plant. Environ. Sci. Technol. 37,
343–351.
Hermansson, A., Lindgren, P.E., 2001. Quantification of ammonia-oxidizing bacteria
in arable soil by real time PCR. Appl. Environ. Microbiol. 67, 972–976.
Jeong, Y.-S., Chung, J.S., 2006. Biodegradation of thiocyanate in biofilm reactor using
fluidized-carriers. Process Biochem. 41, 701–707.
Kane, M.D., Poulsen, L.K., Stahl, D.A., 1993. Monitoring the enrichment and isolation
of sulfate-reducing bacteria by using oligonucleotide hybridization probes
designed from environmentally derived 16s ribosomal RNA sequences. Appl.
Environ. Microbiol. 59, 682–686.
Kim, S.-S., Kim, H.-J., 2003. Impact and threshold concentration of toxic materials in
the stripped gas liquor on the nitrification. Korean J. Chem. Eng. 20, 1103–1110.
Kim, Y.M., Park, D., Lee, D.S., Park, J.M., 2008a. Inhibitory effects of toxic compounds
on nitrification process for cokes wastewater treatment. J. Hazard. Mater. 152,
915–921.
Kim, Y.M., Park, D., Jeon, C.O., Lee, D.S., Park, J.M., 2008b. Effect of HRT on the
biological pre-denitrification process for the simultaneous removal of toxic
pollutants from cokes wastewater. Bioresource Technol. 99, 8824–8832.
Kim, Y.M., Park, D., Lee, D.S., Jung, K.A., Park, J.M., 2009. Sudden failure of biological
nitrogen and carbon removal in the full-scale pre-denitrification process
treating cokes wastewater. Bioresource Technol. 100, 4340–4347.
Klappenbach, J.A., Saxma, n, P.R., Cole, J.R., Schmidt, T.M., 2001. Rrndb: the
ribosomal RNA operon copy number database. Nucleic Acids Res. 29, 181–184.
Kuo, W.C., Shu, T.Y., 2004. Biological pre-treatment of wastewater containing
sulfate using anaerobic immobilized cells. J. Hazard. Mater. 113, 147–155.
Lay-Son, M., Drakides, C., 2008. New approach to optimize operational conditions
for the biological treatment of a high-strength thiocyanate and ammonium
waste: pH as key factor. Water Res. 42, 774–780.
Layton, A.C., Dionisi, H., Kuo, H.W., Robinson, K.G., Garrett, V.M., Meyers, A., Sayler,
G.S., 2005. Emergence of competitive dominant ammonia-oxidizing bacterial
populations in a full-scale industrial wastewater treatment plant. Appl. Environ.
Microbiol. 71, 1105–1108.
Lee, C., Kim, J., Do, H., Hwang, S., 2008. Monitoring thiocyanate-degrading microbial
community in relation to changes in process performance in mixed culture
systems near washout. Water Res. 36, 1254–1262.
Lim, J., Lee, S., Hwang, S., 2008. Use of quantitative real-time PCR to monitor
population dynamics of ammonia-oxidizing bacteria in batch process. J. Ind.
Microbiol. Biotechnol. 35, 1334–1339.
Lin, C., Stahl, D.A., 1995. Comparative analyses reveal a highly conserved
endoglucanase in the cellulolytic genus Fibrobacter. J. Bacteriol. 177, 2543–
2549.
Lydmark, P., Almstrand, R., Samuelsson, K., Mattsson, A., Sörensson, F., Lindgren,
P.E., Hermansson, M., 2007. Effects of environmental condition on the nitrifying
population dynamics in a pilot wastewater treatment plant. Environ. Microbiol.
9 (9), 2220–2233.
Mobarry, B.K., Wagner, M., Urbain, V., Rittmann, B.E., Stahl, D.A., 1996. Phylogenetic
probes for analyzing abundance and spatial organization of nitrifying bacteria.
Appl. Environ. Microbiol. 62, 2156–2162.
Muyzer, G., Dewaal, E.C., Uitterlinden, A.G., 1993. Profiling of complex microbial
populations by denaturing gradient gel electrophoresis-analysis of polymerase
chain reaction-amplified genes coding for 16S rRNA. Appl. Environ. Microbiol.
59, 695–700.
Nogueira, R., Melo, L.F., Purkhold, U., Wuertz, S., Wagner, M., 2002. Nitrifying
bacteria and heterotrophic population dynamics in biofilm reactors: Effects of
hydraulic retention time and the presence of organic carbon. Water Res. 36,
469–481.
Regan, J.M., Harrington, G.W., Noguera, D.R., 2002. Ammonia- and nitrite-oxidizing
bacterial communitiesina pilot-scale
distribution system. Appl. Environ. Microbiol. 68, 73–81.
Shinoda, Y., Sakai, Y., Uenishi, H., Uchihashi, Y., Hiraishi, A., Yukawa, H., Yurimoto,
H., Nobuo, K., 2004. Aerobic and anaerobic toluene degradation by a newly
isolated denitrifying bacterium, Thauera sp. strain DNT-1. Appl. Environ.
Microbiol. 70, 1385–1392.
Siripong, S., Rittmann, B.E., 2007. Diversity study of nitrifying bacteria in full-scale
municipal wastewater treatment plants. Water Res. 41 (5), 1110–1120.
Sorokin, D.Y., Tourova, T.P., Lysenko, A.M., Kuenen, J.G., 2001. Microbial thiocyanate
utilization under highly alkaline conditions. Appl. Environ. Microbiol. 67, 528–
538.
Staib, C., Lant, P., 2007. Thiocyanate degradation during activated sludge treatment
of coke-ovens wastewater. Biochem. Eng. J. 34, 122–130.
Wagner, M., Rath, G., Amann, R., Koops, H.P., Schleifer, K.H., 1995. In situ
identification of ammonia-oxidizing bacteria. Syst. Appl. Microbiol. 18, 251–
264.
chloraminateddrinkingwater
922
Y.M. Kim et al./Bioresource Technology 102 (2011) 913–922