Bioresource Technology 159 (2014) 272–279
Contents lists available at ScienceDirect
Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech
A soil infiltration system incorporated with sulfur-utilizing autotrophic
denitrification (SISSAD) for domestic wastewater treatment
Zhe Kong a,b, Chuanping Feng a,b,⇑, Nan Chen b, Shuang Tong a,b, Baogang Zhang b, Chunbo Hao b,
Kun Chen b
a
Key Laboratory of Groundwater Circulation and Evolution, China University of Geosciences (Beijing), Ministry of Education, No. 29 Xueyuan Road, Haidian District, Beijing
100083, China
b
School of Water Resources and Environment, China University of Geosciences (Beijing), No. 29 Xueyuan Road, Haidian District, Beijing 100083, China
h i g h l i g h t s
A SISSAD for domestic wastewater treatment is successfully developed.
CO2 from heterotrophic aerobic respiration in AES could be used for autotrophic bacteria in ANS.
þ
COD and NH4 -N are removed in AES, NO3 -N is reduced in ANS, TP is removed by adsorption.
a r t i c l e
i n f o
Article history:
Received 30 December 2013
Received in revised form 16 February 2014
Accepted 22 February 2014
Available online 3 March 2014
Keywords:
Soil infiltration
Wastewater
Sulfur
Autotrophic denitrification
a b s t r a c t
To enhance the denitrification performance of soil infiltration, a soil infiltration system incorporated with
sulfur-utilizing autotrophic denitrification (SISSAD) for domestic wastewater treatment was developed,
and the SISSAD performance was evaluated using synthetic domestic wastewater in this study. The aerobic respiration and nitrification were mainly taken place in the upper aerobic stage (AES), removed
88.44% COD and 89.99% NHþ
4 -N. Moreover, autotrophic denitrification occurred in the bottom anaerobic
stage (ANS), using the CO2 produced from AES as inorganic carbon source. Results demonstrated that the
SISSAD showed a remarkable performance on COD removal efficiency of 95.09%, 84.86% for NO
3 -N,
95.25% for NHþ
4 -N and 93.15% for TP. This research revealed the developed system exhibits a promising
application prospect for domestic wastewater in the future.
Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction
In China, the total amount of discharged wastewater in 2012 is
68.48 billion tons, in which domestic wastewater accounts for
67.6% (SEPA of China, 2012). It is of increasing concern for both
researchers and government officials to prevent pollution of aquatic environments on domestic wastewater treatment (Wu et al.,
2011). Owing to the excellent performance, the conventional activated sludge process has become the most widely used alternative
for the treatment of domestic wastewater. However, the large land
occupation and the high energy consumption make a wastewater
treatment plant barely built in small towns or rural areas, and even
more difficulties will be produced on handling wastewater in rural
areas than that in cities (Ying et al., 2011). Fortunately, constructed
⇑ Corresponding author at: School of Water Resources and Environment, China
University of Geosciences (Beijing), No. 29 Xueyuan Road, Haidian District, Beijing
100083, China. Tel.: +86 10 82322281; fax: +86 10 82321081.
E-mail address: [email protected] (C. Feng).
http://dx.doi.org/10.1016/j.biortech.2014.02.079
0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.
wetland (CW), benefiting from its low cost, energy saving, simple
maintenance, and higher effluent quality, has been widely applied
to domestic wastewater in some rural areas (Brunch et al., 2011).
However, as a natural system, the performance of CW always relies
on seasonal changes and vegetations (Kouki et al., 2009).
On the other hand, subsurface wastewater infiltration treatment (SWI) is somehow superior to CW because of their higher removal efficiency and better controllability as well as space
occupation (Li et al., 2013b). However, this method need to overcome the hurdle that using natural soil as an filter media will eventually cause clogging (Pozo-Morales et al., 2013). The multi-soillayering system (MSL), filling the void with some large particle filter material such as zeolite and sand around the soil to create a filter layer, was invented to maintain relatively high removal
efficiency and avoid clogging (Watatsuki et al., 1993). Compared
with traditional soil infiltration methods, MSL remarkably lengthened the service life and performed the removal efficiencies of
75% of TN, 80% of TP and 95% of BOD for treating restaurant wastewater (Luanmanee et al., 2001). However, the nitrate removal
Z. Kong et al. / Bioresource Technology 159 (2014) 272–279
efficiency was relatively low in the MSL, mainly due to the inhibition on denitrification by aeration and lack of electron donor
caused by low C/N (Masunaga et al., 2007). It has been reported
that artificial aeration could improve nitrogen removal efficiency
(Butterworth et al., 2013). However, an excessive aeration will consequently bring about the decreasing of both denitrification efficiency and overall nitrogen removal efficiency.
Thereby, the majority of previous soil treatment researches
mainly focused on individually promoting and optimizing the heterotrophic denitrification process by artificial aeration controlling
or organic carbon source addition such as wood chips and saw dust
(Liu et al., 2013). In comparison with heterotrophic denitrification,
anoxic autotrophic denitrification using sulfur as an electron donor
exhibited its higher nitrate removal and less cell yield (Sierra-Alvarez et al., 2007), and some studies related to sulfur/limestone autotrophic denitrification (SLAD) system have been reported (Ergas
et al., 2010; Read-daily et al., 2011). The sulfur/limestone autotrophic process corresponded to the principle described as following equation:
55S þ 20CO2 þ 50NO3 þ 38H2 O þ 4NHþ4
Thiobacillus denitrificans
2
ƒƒƒƒƒƒƒƒƒƒƒ
ƒ! 4C5 H7 O2 N þ 25N2 þ 55SO4 þ 64Hþ
ð1Þ
Although there are some attempts of combining aerobic stage
with anaerobic one to establish a two-stage type (Chinalia et al.,
2012), few researches related to incorporating sulfur-based denitrification with soil-based nitrification had been carried out.
In this study, a lab-scale SISSAD reactor was developed by integrating aerobic stage with anaerobic stage, which allows secondary
effluent directly flow into sulfur-based denitrification stage without pumping, consequently reducing space occupation and energy
cost. To evaluate the performance of SISSAD, the removals of COD,
NHþ
4 -N, NO3 -N, and TP in both aerobic and anaerobic stage were
investigated using synthetic domestic wastewater, and the denitrification mechanism was analyzed by stoichiometric calculation for
nitrogen, carbon source and sulfur in the SISSAD.
2. Methods
2.1. Wastewater
In order to effectively control the experiment conditions, a synthetic domestic wastewater was used, which was prepared by dissolving ammonium chloride (NH4Cl), glucose (C6H12O6),
monobasic potassium phosphate (KH2PO4) (AR, Sinopharm Chemical Reagent Co., China) in tap water (22.6–24.3 °C). The concentrations of COD, NHþ
4 -N and TP in the synthetic wastewater were
prepared to 462.75–546.21 mg L1, 4.42–72.56 mg L1 and 8.65–
11.12 mg L1, respectively. The pH in the wastewater was kept at
7.02–7.36.
2.2. Experimental apparatus
As shown in Fig. 1, the SISSAD system mainly consisted of a
reactor, an aeration pump and a wastewater reservoir.
A polyvinyl chloride reactor sizing of 30 cm (L) 20 cm
(W) 50 cm (H) was separated into aerobic stage (AES) for heterotrophic nitrification and anaerobic stage (ANS) for sulfur-utilizing
autotrophic denitrification.
The AES, containing quartz sands (grain diameter 0.5–1 mm and
2–4 mm), volcanic (2–4 mm), zeolites (2–4 mm) and activated carbon (2–4 mm, all from Shengquan Water Clean Materials Co., China), has an oxygen-enriched environment formed by both surface
natural re-oxygenating and internal mechanical aeration with an
air pump (At-A7500, Atman, Chuangxing Co., China). The aeration
flow rate was 1 L min1 and the aeration time was 6 h every day.
273
A spongy porous membrane (30 cm 20 cm 1 cm) was covered
upon the surface of AES to uniformly distribute influent. A rubber
pipe (length 50 cm, diameter 1 cm), on which holes (diameter
5 mm) were drilled every 2 cm, was bended into a crescent shape,
and then buried at depth of 2 cm from the bottom of AES to assure
enough aeration. Especially, the holes were set downside to avoid
clogging resulted from filter materials.
In the anaerobic stage (ANS), the filter materials (Table 1) and
mixture of sulfur granules (grain diameter 0.5–1 mm, CR, Guangfu
Research Institute, China) and crushed oyster shells (S/C = 3:1 (v/
v)) were filled. There were more quartz sands with larger size
and volcanic filled in AES, aiming to create a relatively higher
porosity in AES than in ANS. The oyster shells were obtained from
local restaurant and crushed into particles (diameter 2–4 mm) by a
small high-speed pulverizer (ZN-04, KINGSLH Co., China) after
washing and drying. AES and ANS were separated by another
spongy porous membrane (30 cm 20 cm 1 cm) to ensure the
distribution uniformity for the effluent from AES.
A glass funnel (diameter 90 mm) covered with gauze (1 mm
thickness) was planted between the two stages to exactly collect
the effluent from AES. The height of AES was 4 cm longer than that
of ANS in order to ensure the growth of nitrifying bacteria.
2.3. Experimental procedure
2.3.1. Acclimation for nitrifying and denitrifying bacteria
For acclimation of nitrifying bacteria, an activated sludge collected from secondary sedimentation tank (Qinghe Wastewater
Plant, Beijing) was cultured in a 1 L beaker fed with 5 g C6H12O6,
2 g NH4Cl and 0.5 g KH2PO4 (AR, Sinopharm Chemical Reagent
Co., China), and aerated with air pump for 6 h every day, while another activated sludge from anaerobic tank (Qinghe Wastewater
Plant, Beijing, China) for acclimation of denitrifying bacteria was
cultivated in a sealed 500 mL conical flask containing 200 mL medium, which was prepared by dissolving 1 g Na2S2O35H2O, 0.4 g
KNO3, 0.2 g NaHCO3, 0.4 g KH2PO4, 0.1 g NH4Cl, 0.1 g MgCl26H2O
and 0.002 g FeSO47H2O (AR, Sinopharm Chemical Reagent Co.,China) in distilled 200 mL water, and then incubated in a DDHZ300 Shaker (Peiying, Taicang Experiment Equipment Co., China)
at 30 °C. The medium was renewed every day to provide a proper
condition for the anaerobic denitrifying bacteria growth. After
2 weeks aqueous culturing, the sediment in the conical flask was
transferred into another 250 mL conical flasks filled with 10 g solid
sulfur granules and 100 mL liquid medium only containing 0.05 g
KNO3, 0.004 g KH2PO4 and 0.2 g NaHCO3 (AR, Sinopharm Chemical
Reagent Co., China). After 10-day culturing, aqueous samples collected from the supernatant were filtered through 20 lm membrane filter (diameter 18 cm, Nova, Hangzhou Special Paper
Industry Co., China) to determine the nitrate concentration in the
liquid medium. The NO
3 -N removal efficiency was almost 100%
during the culturing period (data not shown), demonstrated that
the sulfur-utilizing autotrophic bacteria had been successfully
acclimated.
2.3.2. Inoculation
According to different components for each stage as given in
Table 1, the filter materials were adequately mixed by a shovel
in a 25 L plastic bucket after washing. Firstly, the mixture of filter
materials for ANS was packaged into the reactor; S/C layer (thickness 5 cm) filled in the middle of ANS, and 100 mL of the acclimated sulfur/liquid solution was inoculated onto the S/C layer.
After the glass funnel was buried steadily, the mixture of filter
materials was compacted firmly by shovel to reduce the porosity
for creating an anaerobic environment. On the other hand,
100 mL of cultured nitrifying bacteria solution was taken, and then
directly poured into a bucket, stirred by the shovel to fully mix
274
Z. Kong et al. / Bioresource Technology 159 (2014) 272–279
Fig. 1. A schematic diagram of the lab-scale SISSAD reactor.
Table 1
Component of filter materials.
Stage
Quartz sand
(S) (%, v/v)
Quartz sand
(L) (%, v/v)
Volcanic
(%, v/v)
Zeolite
(%, v/v)
Activated carbon
(%, v/v)
AES
ANS
30
50
35
15
20
10
10
10
5
5
with filter materials, and finally the mixture was filled into the
upper part of the reactor to construct the AES.
2.3.3. Experimental procedure
The synthetic wastewater was stored in a 25 L plastic tank as
reservoir and then pumped onto the surface sponge via a sprinkler
at the flow rate of 20 mL min1 by a peristaltic pump (Lead 15–24
Longer Precision Pump Co., China), and then flowed through AES
and ANS by gravity, finally discharged from the bottom outlet.
The hydraulic retention time (HRT) was 6 h. The SISSAD operated
for 63 d at the room temperature (20–25 °C). Samples were collected everyday from the influent, Effluent A and Effluent B.
2.4. Analytical methods
In this study, analytical reagent chemicals were used to prepare
standards and stock solution (Sinopharm Chemical Reagent Co.,
China).
COD was digested by potassium dichromate using a 5B-1F
Speed Digester (Lianhua Tech Co., China) at 150 °C for 10 min
and determined by ultraviolet spectrophotometry at 610 nm using
a spectrophotometer (Hach DR6000, USA). NHþ
4 -N was measured
by Nessler’s reagent colorimetry and determined using Hach
DR6000 at 420 nm. TP was measured by molybdate colorimetry
2
using Hach DR6000 at 700 nm, while NO
3 -N, NO2 -N and SO4
were analyzed using a Dionex ICS-2500 ion chromatograph (Dionex Corporation, Sunnyvale, USA). Water temperature and pH were
monitored using a FE30 Benchtop Meter (Mettler Toledo. Co., USA).
3. Results and discussion
3.1. COD
It can be seen from Fig. 2, SISSAD showed an effective removal
ability for COD, even on the first day, the COD concentration in
Effluent A from AES sharply reduced from 462.75 to 229.76 mg L1,
by removal efficiency of 50.68%, and then COD gradually decreased. On Day 17, the COD concentration dropped to
40.82 mg L1 with the removal efficiency of 91.95%, after then,
the COD concentration varied at the range between 53.42 and
20.35 mg L1, average COD removal efficiency was 88.44%. Therefore, it could be considered that the biofilming for AES was completed on Day 17. In comparison with former similar studies of
Lei et al. (2013), whose biofilming period was 4 months, SISSAD
shortened the biofilming period to nearly half a month, mainly
due to the acclimation for nitrifying bacteria in this study. On the
other hand, although COD was mostly consumed in AES, the COD
concentration in Effluent B continued decreasing. Similar with
Effluent A, on Day 1, the COD concentration in Effluent B decreased
to 108.53 mg L1 by the removal efficiency of 77.05%, on Day 15,
reduced to the lowest concentration of 15.63 mg L1, after that,
the COD concentration was almost kept at a constant. It was
showed that ANS also had the capacity for removing the residual
COD from AES, the average COD concentration in Effluent B was
27.30 mg L1. Therefore, SISSAD exhibited a high performance for
removing COD, the average removal efficiency was 95.09%. In contrast, other soil treatment systems such as a SWI, the COD removal
efficiency was 87.2% (Li et al., 2012) and a vertical flow CW was
88% (Abou-Elela and Hellal, 2012). The explanation for COD removal in this study was due to that (1) COD was consumed as
nutrient for the growing of aerobes in AES; (2) COD was used as organic electron for heterotrophic denitrification in AES; (3) the activated carbon granules in the mixture maybe adsorbed a portion of
COD (Zhao et al., 2013b) in both AES and ANS, especially in the initial period of the experiment.
3.2. NHþ
4 -N
As shown in Fig. 3, on the first day of operating, the NHþ
4 -N concentration in Effluent A from AES reduced to 22.08 mg L1 with removal efficiency of 64.33%, and then the concentration continued
to decrease. On Day 16, the NHþ
4 -N concentration decreased to
4.14 mg L1 by the removal efficiency of 92.39%. After then, the
1
NHþ
, expect for
4 -N concentration was kept at less than 6 mg L
some of them were over 6 mg L1 due to higher influent NHþ
4 -N
concentration. Furthermore, the variation of NHþ
4 -N concentration
from Day 1 to Day 15 was almost consistent with COD variation in
biofilming period mentioned above. However, the NHþ
4 -N concen-
Z. Kong et al. / Bioresource Technology 159 (2014) 272–279
275
Fig. 2. COD concentration and removal efficiency.
tration reached the inflection point two days earlier than COD,
mainly due to that the ammonium adsorption by zeolite in the filter materials occurred. AES obtained a well ammonium removal
capacity with the average NHþ
4 -N removal efficiency of 89.99% in
Effluent A.
On the other hand, the NHþ
4 -N concentration in Effluent B from
ANS on Day 1 declined to 5.52 mg L1 with the removal efficiency
of 91.08%, the concentration continued to drop to 0.39 mg L1 on
Day 12, obtaining the removal efficiency of 99.32%. After Day 12,
1
the concentration of NHþ
, the
4 -N mostly remained below 5 mg L
þ
1
average NH4 -N concentration in Effluent B was 2.86 mg L with
the removal efficiency of 95.25%, meant that the residual NHþ
4 -N
from AES were used in ANS. The NHþ
-N
decrease
in
ANS
was
4
attributed to (1) adsorption by the filter materials; (2) utilization
by sulfur-utilizing autotrophic denitrification according to Eq. (1).
Seen from Eq. (1), 4 mol NHþ
4 -N were needed for reducing
50 mol NO
3 -N.
Therefore, the NHþ
4 -N concentration requirement by autotrophic denitrification could be calculated as follow:
CANS ðNHþ4 -NÞ ¼
4CA ðNO3 -NÞ EANS ðNO3 -NÞ
50
ð2Þ
Furthermore, the practical NHþ
4 -N concentration removed by
ANS is:
CANS:pra ðNHþ4 -NÞ ¼ CA ðNHþ4 -NÞ CB ðNHþ4 -NÞ
ð3Þ
þ
þ
Where, CA(NHþ
4 -N) and CB(NH4 -N) are the average NH4 -N concentrations of Effluent A and B (mg L1), CA(NO
-N)
is
the
average
3
1
NO
), EANS(NO
3 -N concentration of Effluent A (mg L
3 -N) is the
average removal efficiency of ANS (%). Substituted 45.31 for
þ
þ
CA(NO
3 -N), 6.02 for CA(NH4 -N), 2.86 for CB(NH4 -N) and 84.86 for
þ
EANS(NO3 -N), then CANS(NH4 -N) = 3.07 and CANS.pra(NHþ
4 -N) = 3.16
were obtained. The practical NHþ
4 -N removal was higher than the
requirement by autotrophic denitrification, demonstrating that
the additional adsorption process was involved.
Both AES and ANS had shown their high removal performance
on NHþ
4 -N, even on the first day, obtaining 66.33% and 91.08% removal efficiencies respectively. Nevertheless, the nitrate concentration in Effluent A (Fig. 4) was less than 40 mg L1 before Day
19, and reached 60 mg L1 until Day 24. In other word, the
NO
3 -N generation rate was unable to keep abreast with the
þ
NHþ
4 -N removal rate. Therefore, the NH4 -N removal at the initial
period (until Day 24) was mainly resulted from the adsorption
by the mixed filter materials. It has been demonstrated that zeolite
had an excellent adsorbing effect on NHþ
4 -N and such adsorption
follows the Freundlich model (Zhao et al., 2013a). In this study, a
series of batch tests were also conducted to estimate the adsorption capacities for different materials, suggesting that zeolite obtained 33.10 mg g1 saturated adsorption capacity for 48 h,
showed the best performance among sand, volcanic and activated
carbon (data not shown). From Table 1, zeolite accounts for 10%
(v/v) of the mixed filter materials filled in the reactor, weighing
Fig. 3. NHþ
4 -N concentration and removal efficiency.
276
Z. Kong et al. / Bioresource Technology 159 (2014) 272–279
5250 g. As known, average influent NHþ
4 -N concentration was
60.04 mg L1, and 64.33% NHþ
4 -N removal in the reactor on Day 1
was hypothesized as the adsorption efficiency, 772.47 mg NHþ
4 -N
could be adsorbed by the zeolite in the reactor every day (20 L synthetic wastewater). However, 5250 g zeolite could adsorb
21721.88 mg NHþ
4 -N calculated using the HRT 6 h, then divided
by 772.47 mg d1, figuring out 28 days as the theoretical saturated
adsorption period for NHþ
4 -N, and it was close to 24 days mentioned above, after which the ammonia removal was considered
resulting from microbial activities. Nevertheless, owing to the oxidation of NHþ
4 -N, zeolite might never achieve a practical saturated
þ
adsorption for NHþ
4 -N. Thus, the NH4 -N removal was due to both
adsorption and nitrification.
After Day 15, the biofilm of nitrifying bacteria was gradually
formed, the conversion from ammonium to nitrate occurred by
nitrifying bacteria became a dominant role instead of adsorption.
Overall, the average NHþ
4 -N removal efficiency of the whole system was 95.25%, and the highest removal efficiency was 99.32%,
exhibiting that the SISSAD had a high performance on NHþ
4 -N removal, also evidenced that artificial aeration could enhance ammonia removal in soil infiltration system.
3.3. NO
3 -N and NO2 -N
As shown in Fig. 4, the concentration of NO
3 -N in Effluent A was
3.57 mg L1 on Day 1, and then increased rapidly. After 7 days, the
NO
3 -N concentration in Effluent A had already increased to
43.67 mg L1, and then continued to grow up to the highest concentration of 62.55 mg L1 until Day 25. Since Day 26, the concentration had been slightly floating for 22 days. Adversely, the NO
3 -N
concentration trended to decrease slowly after Day 49, finally ending with 45.87 mg L1.
1
The NO
on Day
2 -N concentration in Effluent A was 0.09 mg L
1
1, and smoothly increased to 0.81 mg L on Day 17 with a small
fluctuation, then rapidly growing to 2.50 mg L1 on Day 26. After
then, the concentration began to slightly decrease and float by a
tiny range between 1.30 and 2.49 mg L1, owning an average concentration of 1.40 mg L1.
As discussed above, the average NHþ
4 -N removal efficiency in
AES was 89.99%, it was hypothesized that the 89.99% of NHþ
4 -N
were all converted into NO
3 -N, a theoretical average NO3 -N concentration of 60.04 mg L1 would be obtained from AES. However,
the average NO
3 -N concentration detected in Effluent A was
45.31 mg L1. The disagreement between theoretical and practical
concentration was probably due to: (1) a portion of NHþ
4 -N was
used as nutrient substance for the growth of microorganisms to
produce biomass such as cell wall; (2) denitrification might be taken place in some anoxic parts of AES, consequently leads to the
formation of NO
2 -N. Yang et al. (2013b) also reported that the
reduction of NO
3 -N in anoxic denitrification would bring NO2 -N
into existence. On the other hand, NO
-N
was
found
to
be
an
obli2
gate intermediate in nitrification (Morkved et al., 2007), and previous researches on CW indicated that most conventional subsurface
flow constructed wetlands were unable to fulfill the process of
NHþ
4 -N converting into NO3 -N on occasion of lack of oxygen,
resulting in nitrite generation (Fan et al., 2013). Therefore, it was
indicated that some portions in AES remained in an anoxic condition due to that non-uniform oxygen supplying resulted from a single pipe in this study.
The NO
3 -N concentration in Effluent B was kept at the range
from 0.40 mg L1 to 9.33 mg L1. The average NO
3 -N concentration
in ANS was 4.35 mg L1 with the removal efficiency of 84.86%,
demonstrating a high denitrification performance in the ANS.
Seen from Eq. (1), autotrophic denitrification needed inorganic
carbon (CO2 or HCO
3 ), however, ANS still gave a well denitrification performance even though inorganic carbon was not added in
the synthetic wastewater. A previous study has revealed a mixotrophic process combining simultaneous heterotrophic denitrification with autotrophic denitrification in sulfur-based column
bioreactor (Sahinkaya and Dursun, 2012). The explanation in this
study might be:
(1) CO2 generated from aerobic respiration by heterotrophic
aerobes, following Eq. (4):
C6 H12 O6 þ 6O2 ! 6CO2 þ 6H2 O
ð4Þ
A stoichiometric calculation based on aerobic respiration for glucose indicated that consuming 88.44% of COD (average influent concentration 500.32 mg L1) in AES would produce 608.41 mg L1
CO2;
(2) CO2 generated from the denitrification in some anoxic parts
in AES, following Eq. (5):
C6 H12 O6 þ 12NO3 ! 6CO2 þ 6H2 O þ 12NO2
Therefore, 0.73 mg L1 CO2 and 1.40 mg
generated.
ð5Þ
L1 NO
2 -N
would be
(3) CO2 from the reaction of SO2
4 with CaCO3. In fact, this portion of CO2 might not be the main source of inorganic carbon, but CO2 from acidized CaCO3 also played a role in
supplying additional inorganic carbon source.
Fig. 4. NO
3 -N and NO2 -N concentration.
Z. Kong et al. / Bioresource Technology 159 (2014) 272–279
However, according to Eq. (1), to reduce 45.31 mg L1 NO
3 -N
from Effluent A, only 56.96 mg L1 CO2 was needed, indicated that
CO2 from aerobic respiration by heterotrophic bacteria was enough
to supply inorganic carbon source for sulfur-utilizing autotrophic
denitrification.
On the other hand, NO
2 -N occurred in AES was mostly removed
by ANS with sulfur/limestone system, obtaining an average NO
2 -N
concentration of 0.18 mg L1 and an average removal efficiency of
87.41% in ANS, because sulfur-based autotrophic denitrification
could reduce nitrite to N2 as well as nitrate (Qambrani et al., 2013).
3.4. TP
As shown in Fig. 5, total phosphorus (TP) in Effluent A decreased
from 10.11 mg L1 to 5.32 mg L1 on the first day, and then fluctuated from 2.01 mg L1 to 6.22 mg L1 until Day 18. After Day 19,
the TP concentration began to stabilize and remained stable
around 3 mg L1.The average TP concentration from AES was
3.13 mg L1 with the removal efficiency of 68.39%, mainly due to
the adsorption and consumption as nutrient by microorganisms.
Flowing through ANS, TP in Effluent B gradually decreased from
2.02 mg L1 on Day 1 to 0.92 mg L1 on Day 20, and then the TP
concentration had never been up to 1 mg L1. The average TP concentration in ANS was 0.67 mg L1 and the total removal efficiency
in the reactor was 93.15%, which was superior to 60% of bioretention systems (Ergas et al., 2010). In this study, it was considered
that the volcanics in the mixed-filter-materials played a significant
role in adsorbing phosphorus. Yang et al. (2013a) have reported
that volcanics such as Kunama mud exhibited great phosphate
adsorption capacity. Furthermore, phosphate was necessary for
the process of autotrophic denitrification by denitrifying bacteria
because phosphorus was used by denitrifying bacteria for producing ADP sulfurylase (Bruser et al., 2000), consequently phosphate
could be removed.
3.5. SO2
4 and alkalinity
As shown in Fig. 6, the growing of SO2
4 indicated that denitrifying bacteria for autotrophic denitrification used sulfur granules in
the ANS. The SO2
concentration on the first day was only
4
28.46 mg L1, and then smoothly rose up to 166.93 mg L1 by
Day 49. However, after then, the SO2
concentration rapidly in4
creased, reaching 396.22 mg L1 on the last day.
During the experiment period, the pH value in Effluent A floated
from 7.36 to 6.37 due to the oxidation of NHþ
4 -N and the generation
277
of nitrate and nitrite, which accorded with Yuan’s report on soil
wastewater infiltration system (Yuan et al., 2013). Nevertheless,
pH in Effluent B decreased from 6.8 to 5.32, much lower than Effluent A, suggested that sulfur-utilizing autotrophic denitrification in
ANS had more hydrogen ions generated than that in AES. After Day
52, the pH in Effluent B sharply dropped to 5.24, at which the internal environment was no longer suitable for autotrophic denitrifying bacteria. It has been reported that sulfur-utilizing autotrophic
denitrification occurred between pH 6 and 9 (Oh et al., 2001),
and SO2
showed severe inhibition on denitrification process at
4
pH lower than 5.5 (Liu and Koeing, 2002). Unfortunately, pH continued to lower down, finally a small amount of solid particle
impurities with dark color was observed and sour smelled in the
final effluent. The impurity was speculated to be a mixture containing dead bodies of autotrophic anaerobes and the acidized
limestone particles. The SISSAD was sacrificed immediately on
Day 63 when pH tragically dropped to 3.84. Compared with Ergas’
76 cm-high biofilter for 3 months (Ergas et al., 2010), the practical
operating period of SISSAD seemed rather short, which yet does
not mean a vertical infiltration system is not as superior as those
up-flow sulfur packed-bed reactors (USPR) or a horizontal-flow
bioreactor (Read-daily et al., 2011) since the excess and uncontrollable consumption of alkalinity was not estimated. The alkalinity
reducing caused by a growing sulfate concentration was inevitable
in sulfur/limestone autotrophic denitrification (SLAD) (Moon et al.,
2006; Zhou et al., 2011), however, in contrast to USPRs, SISSAD had
more alkalinity consumption mainly due to the restriction of vertical water flow. Former issues on CW suggested that vertical-flow
had greater oxygen transfer capacity than others (Cui et al.,
2012). Similarly, in this study, owing to the vertical-flow infiltration design, a non-uniform water distribution caused the increasing of DO and much quicker neutralizing reaction occurred in
some parts in the mixed-filter-materials, resulting more alkalinity
consuming. Moreover, low hydraulic loading rate (HLR) and long
artificial aeration time could bring about an excellent nitrogen removal, while also led the sulfur oxidized rapidly. To avoid the risk
of harming aquatic environment, SO2
4 in the final effluent would
be further treated by some approaches such as CaCO3 filter process
in the practical application.
A stoichiometric calculation was performed based on Eq. (1) to
explain the fatal problem involved with alkalinity. We hypothesized that all NHþ
4 -N from the synthetic wastewater were converted into NO
3 -N and then reduced into N2 by sulfur-utilizing
denitrifying bacteria, means that removing an average
1
60.04 mg L1 NO
SO2
and
3 -N will generate 457.87 mg L
4
Fig. 5. TP concentration and removal efficiency.
278
Z. Kong et al. / Bioresource Technology 159 (2014) 272–279
Fig. 6. SO2
4 concentration and pH variation.
consume 274.47 mg L1 CaCO3 as alkalinity. However, the average
1
SO2
and highest
4 concentration in effluent was only 149.59 mg L
1
SO2
concentration
was
396.22
mg
L
,
demonstrated
that crushed
4
oyster shells were effective on removing most of SO2
4 and neutralizing H+. The rapid increasing of SO2
4 after Day 56 (Fig. 6) indicated
that oyster shells were no longer able to maintain the pH over 6 for
an adequate autotrophic denitrification. According to stoichiometric calculation, the total alkalinity consumption for the whole
experiment period was 354.83 g CaCO3 (20 L synthetic wastewater
per day), however, 513.42 g of the prepared oyster shell only survived for 2 months. This might be due to that (1) the large cross
sectional area for the SISSAD reactor under low HLR may cause a
non-uniform water distribution, consequently resulted in lower
utilization of oyster shells; (2) large particle size of crush oyster
shells could also results in low utilization. It was suggested dispersing the S/C particles instead of gathering them in a layer, and
reducing the particle size of oyster shells would be necessary in
the future study. On the other hand, a volume ratio of 3:1 for S/C
was considered no longer fit for the sulfur-utilizing autotrophic
denitrification in a vertical infiltration system. Another low cost
alternative electron donor source such as iron pyrite (FeS2) could
be utilized as well as element sulfur in autotrophic denitrification
but gaining fewer sulfates (Li et al., 2013a).
4. Conclusion
The lab-scale SISSAD, combining soil infiltration with SLAD had
been constructed. Except for its uncontrollable sulfate generation,
the system had shown a remarkable performance on treating synthetic domestic wastewater. The majority of COD and NHþ
4 -N were
removed in AES by aerobic respiration and nitrification respectively; phosphorus was mostly removed by adsorption, while nitrate and nitrite were reduced by autotrophic denitrification in
ANS. The integration design brought about a compact construction
that allows CO2 generated from AES to be directly used in ANS,
consequently lowering cost and space occupation. Further study
will be focused on oxygen supply and alkalinity consumption.
Acknowledgements
This research work was supported by the National Natural
Science Foundation of China (NSFC) as a young scholar project
(No. 21307117) and the National Natural Science Foundation
(No. 31140082), the Fundamental Research Funds for the Central
Universities (No. 2652013058).
References
Abou-Elela, S.I., Hellal, M.S., 2012. Municipal wastewater treatment using vertical
flow constructed wetlands planted with Canna, Phragmites and Cyprus. Ecol.
Eng. 47, 209–213.
Brunch, I., Fritsche, J., Bänninger, D., Alewella, U., Sendelov, M., Hürlimann, H.,
Hasselbach, R., Alewell, C., 2011. Improving the treatment efficiency of
constructed wetlands with zeolite-containing filter sands. Bioresour. Technol.
102, 937–941.
Bruser, T., Selmer, T., Dahl, C., 2000. ‘‘ADP sulfurylase’’ from Thiobacillus denitrificans
is an adenylylsulfate: phosphate adenylyltransferase and belongs to a new
family of nucleotidyltransferases. J. Biol. Chem., 1691–1698.
Butterworth, E., Dotro, G., Jones, M., Richards, A., Onunkwo, P., Narroway, Y.,
Jefferson, B., 2013. Effect of artificial aeration on tertiary nitrification in a fullscale subsurface horizontal flow constructed wetland. Ecol. Eng. 54, 236–244.
Chinalia, F.A., Garbossa, H.P., Rodriguez, J.A., Lapa, K.R., Foresti, E., 2012. A two-stage
aerobic/anaerobic denitrifying horizontal bioreactor designed for treating
ammonium and H2S simultaneously. Appl. Biochem. Biotechnol. 168, 1643–
1654.
Cui, L.H., Feng, J.K., Ouyang, Y., Deng, P.W., 2012. Removal of nutrients from septic
effluent with re-circulated hybrid tidal flow constructed wetland. Ecol. Eng. 46,
112–115.
Ergas, S.J., Sengupta, S., Siegel, R., Pandit, A., Yao, Y.F., Yuan, Xin., 2010. Performance
of nitrogen-removing bioretention systems for control of agricultural runoff. J.
Environ. Eng. 136, 1105–1112.
Fan, J.L., Zhang, B., Zhang, J., Ngo, H.H., Guo, W.S., Liu, F.F., Guo, Y.Y., Wu, H.M., 2013.
Intermittent aeration strategy to enhance organics and nitrogen removal in
subsurface flow constructed wetlands. Bioresour. Technol. 141, 117–122.
Kouki, S., M’hiri, F., Saidi, N., Belaïd, S., Hassen, A., 2009. Performances of a
constructed wetland treating domestic wastewaters during a macrophytes life
cycle. Desalination 246, 452–467.
Lei, Z.F., Wu, T., Zhang, Y., Liu, X., Wan, C.L., 2013. Two-stage soil infiltration
treatment system for treating ammonium wastewaters of low COD/TN ratios.
Bioresour. Technol. 128, 774–778.
Li, R.H., Niu, J.M., Zhan, X.M., 2013a. Simultaneous removal of nitrogen and
phosphorus from wastewater by means of FeS-based autotrophic
denitrification. Water Sci. Technol. 67 (12), 2761–2767.
Li, Y.H., Li, H.B., Pan, J., Wang, X., Sun, T.H., 2012. Performance evaluation of
subsurface wastewater infiltration system in treating domestic sewage. Water
Sci. Technol. 65 (4), 713–720.
Li, Y.H., Li, H.B., Wang, H., Wang, X., Zou, Y., Sun, T.H., 2013b. Comparison of the
treatment performance of bio-substrate based and meadow brown soil based
subsurface infiltration systems for domestic wastewater treatment. Water Sci.
Technol. 67 (3), 506–513.
Liu, L., Zhao, X.H., Zhao, N., Shen, Z., Wang, M., Guo, Y.Z., Xu, Y.B., 2013. Effect of
aeration modes and influent COD/N ratios on the nitrogen removal performance
of vertical flow constructed wetland. Ecol. Eng. 57, 10–16.
Liu, L.H., Koeing, A., 2002. Use of limestone for pH control in autotrophic
denitrification: batch experiments. Process Biochem. 37 (8), 885–893.
Luanmanee, S., Attanandana, T., Masunaga, T., 2001. The efficiency of a multi-soillayering system on domestic waste-water treatment during the ninth and tenth
years of operation. Ecol. Eng. 18 (2), 185–199.
Masunaga, T., Sato, K., 2007. Characteristics of wastewater treatment using a multisoil-layering system in relation to wastewater contamination levels and
hydraulic loading rates. Soil Sci. Plant Nutr. 53 (2), 215–223.
Moon, H.S., Nam, K., Kim, J.Y., 2006. Initial alkalinity requirement and effect of
alkalinity sources in sulfur-based autotrophic denitrification barrier system. J.
Environ. Sci. 132, 971–975.
Z. Kong et al. / Bioresource Technology 159 (2014) 272–279
Morkved, P.T., Dorsch, P., Bakken, L.R., 2007. The N2O product ratio of nitrification
and its dependence on long-term changes in soil pH. Soil Biol. Biochem. 39,
2048–2057.
Oh, S.E., Yoo, Y.B., Young, J.C., Kim, I.S., 2001. Effect of organics on sulfur-utilizing
autotrophic denitrification under mixotrophic conditions. J. Biotechnol. 92, 1–8.
Pozo-Morales, L., Franco, M., Garvi, D., Lebrato, J., 2013. Influence of the stone
organization to avoid clogging in horizontal subsurface-flow treatment
wetlands. Ecol. Eng. 54, 136–144.
Qambrani, N.A., Jung, S.K., Ok, Y.S., Kim, Y.S., Oh, S.E., 2013. Nitrate-contaminated
groundwater remediation by combined autotrophic and heterotrophic
denitrification for sulfate and pH control: batch tests. Environ. Sci. Pollut. Res.
20, 9084–9091.
Read-Daily, B., Tank, J., Nerenberg, R., 2011. Stimulating denitrification in a stream
mesocosm with elemental sulfur as an electron donor. Ecol. Eng. 37, 580–588.
Sahinkaya, E., Dursun, N., 2012. Sulfur-oxidizing autotrophic and mixotrophic
denitrification processes for drinking water treatment: elimination of excess
sulfate production and alkalinity requirement. Chemosphere 89, 144–149.
SEPA of China, 2012. The national environment statistical bulletin of 2012
(in Chinese). <http://zls.mep.gov.cn/hjtj/qghjtjgb/201311/t20131104_262805.
htm>.
Sierra-Alvarez, R., Beristain-Cardoso, R., Salazar, M., Gomez, J., Razo-Flores, E., Field,
J.A., 2007. Chemolithotrophic denitrification with elemental sulfur for
groundwater treatment. Water Res. 41, 1253–1262.
Watatsuki, T., Esumi, H., Omura, S., 1993. High-performance and N-removable and
P-removable on-site domestic waste-water treatment system by multi-soillayering method. Water Sci. Technol. 27 (1), 31–40.
279
Wu, S.B., Astin, D., Liu, L., Dong, R.J., 2011. Performance of integrated household
constructed wetland for domestic wastewater treatment in rural areas. Ecol.
Eng. 37, 948–954.
Yang, S.J., Ding, D.H., Zhao, Y.X., Huang, W.L., Zhang, Z.Y., Lei, Z.F., Yang, Y.N., 2013a.
Investigation of phosphate adsorption from aqueous solution using Kanuma
mud: behaviors and mechanisms. J. Environ. Chem. Eng. 1, 355–362.
Yang, X.N., Ye, J.X., Lyu, L.M., Wu, Q.H., Zhang, R.D., 2013b. Anaerobic
biodegradation of pyrene by Paracoccus denitrificans under various nitrate/
nitrite-reducing conditions. Water Air Soil Pollut. 224, 1578.
Ying, H.L., Li, H.B., Sun, T.H., Wang, X., 2011. Effects of hydraulic loading rate on
pollutants removal by a deep subsurface wastewater infiltration system. Ecol.
Eng. 37, 1425–1429.
Yuan, H.P., Nie, J.Y., Zhu, N.W., Miao, C., Lu, N., 2013. Effect of temperature on the
wastewater treatment of a novel anti-clogging soil infiltration system. Ecol. Eng.
57, 375–379.
Zhao, C., Zheng, Z.Y., Zhang, J., Wen, D.H., Tang, X.Y., 2013a. Adsorption
characteristics of ammonium exchange by zeolite and the optimal application
in the tertiary treatment of coking wastewater using response surface
methodology. Water Sci. Technol. 67 (3), 619–627.
Zhao, Q., Han, H.J., Xu, C.Y., Zhuang, H.F., Fang, F., Zhang, L.H., 2013b. Effect of
powdered activated carbon technology on short-cut nitrogen removal for coal
gasification wastewater. Bioresour. Technol. 142, 179–185.
Zhou, W.L., Sun, Y.J., Wu, B.T., 2011. Autotrophic denitrification for nitrate and
nitrite removal using sulfur–limestone. J. Environ. Sci. 23 (11), 1761–1769.