ARTICLE IN PRESS

Water Research 39 (2005) 11991204

www.elsevier.com/locate/watres

Operational strategies for an activated sludge process in

conjunction with ozone oxidation for zero excess sludge
production during winter season
J.W. Leea,, H.-Y. Chab, K.Y. Parkc, K.-G. Songd, K.-H. Ahnd
a
Department of Civil and Environmental Engineering, University of Delaware, Newark, DE 19716, USA
Envirotech Co. Ltd., New-Seoul Bldg., Nonhyeon-dong 90-1, Gangnam-gu, Seoul 135-010, Republic of Korea
c
Department of Civil and Environmental System Engineering, College of Life and Environmental Sciences, Konkuk University,
1 Hwayang-dong, Kwangjin-gu, Seoul 143-701, Republic of Korea
d
Future Technology Research Division, Korea Institute of Science and Technology P.O. Box 131, Cheongryang, Seoul, 136-791,
Republic of Korea
b

Received 10 May 2004; received in revised form 25 August 2004; accepted 4 October 2004
Available online 27 March 2005

Abstract
A pilot-scale activated sludge system coupled with sludge ozonation process was operated for 112 days of a winter
season without excess sludge wasting. The concept of this process is that the excess sludge produced is rst disintegrated
by ozone oxidation and then recirculated to a bioreactor in order to mineralize the particulate and soluble organic
compounds. The basis of operation was to determine either the optimal amount of sludge in kg SS ozonated each day
(SO) or the optimal ozonation frequency under the variable inuent chemical oxygen demand (COD) loading and
temperature conditions, since the ozone supply consumes costly energy. The optimal SO was obtained using the
theoretically estimated sludge production rate (SP) and experimentally obtained ozonation frequency (n). While the SP
was mainly subject to the COD loadings, sludge concentration was affected by the temperature changes in winter
season. The optimal n was observed between 2.5 and 2.7 at around 15 1C, but it was doubled at 10 1C. Mixed liquor
suspended solids (MLSS) concentration was leveled off at around 5000 mg/L in bioreactor at 15 1C, but the volatile
fraction of MLSS was xed around 0.7 indicating that there was no signicant inorganic accumulation. Suspended
solids (SS) and soluble COD in efuents kept always a satisfactory level of 10 and 15 mg/L with sufcient
biodegradation. It was recommended to apply a dynamic SO under variable inuent COD loadings and temperature
conditions to the activated sludge system without excess sludge production for saving energy as well as system
stabilization.
r 2004 Elsevier Ltd. All rights reserved.
Keywords: Activated sludge system; Biodegradability; Excess sludge; Ozonation frequency; Sludge ozonation rate

1. Introduction
Corresponding author. Tel.: +1 302 831 1588;
fax: +1 302 831 3640.
E-mail address: jaewoos@udel.edu (J.W. Lee).

Treatment and disposal of excess sludge in an

activated sludge system requires tremendous costs,
amounting to approximately half the entire operationing

0043-1354/$ - see front matter r 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.watres.2004.10.004

ARTICLE IN PRESS
J.W. Lee et al. / Water Research 39 (2005) 11991204

1200

Nomenclature
n
ntheo
Qin
S0

daily ozonation frequency;

theoretically obtained daily ozonation frequency;
wastewater inow rate (m3/day);
inuent COD concentration (kg COD/m3);

cost for domestic wastewater treatment plants. An

innovative concept of an activated sludge system
coupled with disintegration in biomass recycle system
has been developed to achieve zero discharge of excess
sludge by promoting cryptic growth (Deleris et al., 2000;
Huysmans et al., 2001; Yasui et al., 1996). Several
methods such as mills, ultrasounds, heat, alkali, and
ozone treatment have been developed for wastewater
sludge disintegration (Muller et al., 1998; Scheminski
et al., 2000; Weemaes et al., 2000; Yasui and Shibata,
1994). Sludge ozonation was referred as one of the most
cost effective technologies with the highest disintegration capability (Muller, 2000; Park et al., 2003).
Furthermore, ozonated sludge could be effectively
utilized as an additional carbon source in a biological
nitrogen removal process saving a great deal of cost for
external carbon source (Ahn et al., 2002a, b). Sludge
disintegration by ozone was well described with the
sequential decomposition processes of oc disintegration, solubilization and mineralization (Ahn et al.,
2002a, b). The oc disintegration and solubilization
generate a large amount of micro-solids and soluble
organic matter that can be easily decomposed by
microorganisms. Improvement of the biodegradability
of the ozonated sludge has also been conrmed by
several researchers (Scheminski et al., 2000; Weemaes
et al., 2000; Yeom et al., 2002).
The feasibility of the activated sludge system coupled
with ozonation process was veried through the fullscale plant operations without excess sludge production
(Yasui and Shibata, 1994; Yasui et al., 1996). Minimization of excess sludge production could be achieved
by recirculating the ozone-treated sludge into a bioreactor. Of many operational parameters, the sludge
ozonation rate (SO) can be dealt with the most
important one in situ. According to Yasui and Shibata
(1994), sludge recirculation rate into ozonation process
should be at least three times that of a control bioreactor
uncoupled with ozonation unit in order to achieve zero
excess sludge discharge. Nonetheless, the strategy based
on the xed SO may not be exible enough for either
daily or seasonal uctuations of sludge production. It is
true that higher ozone dose produces less excess sludge.
However, overdosing ozone only increases the operational cost and the residual ozone may inhibit biomass
production that plays a role in organic degradation.
Namely, inappropriate design parameters regarding

Se
SO
SP
Yobs

efuent COD concentration (kg COD/m3);

sludge ozonation rate (kg SS/day);
sludge production rate (kg SS/day);
observed sludge yield coefcient (kg SS/kg
COD)

ozonation degree possibly result in decreased system

performance costing a great deal. Therefore, consideration should be placed on the dynamic sludge production
rate (SP) subjected to the inuent chemical oxygen
demand (COD) loading and temperature variations
when determining operation parameters.
In this study, the SP is hypothesized as follows if COD
consumption and yield coefcient are known.
SP Qin  S0  S e  Y obs .

(1)

The above equation implies that the dynamic pattern

of SP is described in terms of the dynamic COD
consumption. Thereafter, SO is simply dened as
follows:
SO SP  n.

(2)

The purpose of this study was to theoretically estimate

the optimal SO and ozonation frequency (n) to
completely mineralize the excess sludge. A pilot study
was implemented to validate the feasibility of those
parameters for pilot-scale domestic wastewater treatment under dynamic inuent COD loading and temperature conditions.

2. Materials and methods

2.1. Pilot-scale experiments
A pilot plant was installed at the Jungrang Wastewater Treatment Plant located in Seoul, South Korea.
The pilot-scale activated sludge system consisted of a
bioreactor for wastewater treatment and a sludge
ozonation unit as shown in Fig. 1. The biological
treatment facility was fed with efuent from the fullscale primary sedimentation tank. The pilot system was
designed with a treatment capacity of 10 m3/day and a
hydraulic retention time (HRT) of 8 h. An average COD
loading and food to microorganism (F/M) ratio was
1.4 kg COD/day and 0.11 kg COD/kg MLSS/day,
respectively. No sludge wasting was conducted during
operation period.
The ozonation process mainly consisted of an ozone
generator and a 350-L cylindrical shape ozone contactor. Ozone gas was delivered into the ozone contactor
and distributed through an ejector. The ozone ow rate
was set at 30 ml/day. The ozone gas concentration was

ARTICLE IN PRESS
J.W. Lee et al. / Water Research 39 (2005) 11991204

1201

Fig. 1. Schematics of the pilot-scale activated sludge system coupled with ozonation process.

50 mg O3/L. The ozone transfer efciency was computed

by in and off-gas measurements using an ozone detector
(IN-USA Co.) The efciency reached over 90%. In
order to set the ozone dose at 0.05 kg O3/kg SS, the
sludgeozone contact time was variable depending on
the sludge concentration. Sludge was intermittently
drained from the bioreactor at the designated rate and
kept in a 1-m3 reservoir prior to the ozonation process.
After ozonation, the sludge was transferred into another
1-m3 reservoir and then continuously recirculated into
the bioreactor at the equivalent drain rate.

The COD was checked for the inuent, efuent, and

mixed liquor samples.
2.3. Analytical methods
Analyses of TSS and VSS in mixed liquor samples or
efuents were performed in accordance with the
standard methods (APHA, 1995). The total and soluble
COD in liquid samples were measured using HACH
COD kits and meter (DR/3000).

3. Results and discussion

2.2. Observed sludge yield (Yobs)
In order to theoretically determine the parameters of
SO and n, it was necessary to identify the Yobs for both
the wastewater and the ozonized sludge. The Yobs values
for the wastewater and 0.05 kg O3/kg SS ozonized sludge
were estimated using the two laboratory continuous
stirred tank reactors (CSTRs) under the temperature
condition of 2071 1C. Yobs was dened as a total solid
mass increase of the sludge per unit COD removal of the
wastewater or ozonized sludge.
Two 10-L laboratory scale CSTRs were run feeding
the wastewater and 0.05 kg O3/kg SS ozonized sludge,
respectively. The wastewater was taken from the inuent
equalization basin of the pilot plant every three days and
was kept at a temperature of 4 1C in a controlled feed
reservoir in order to prevent organic matter being lost to
fermentation. Ozonized sludge was prepared to a
concentration of 0.05 kg O3/kg SS. The mixed liquor
samples were delivered from the pilot plant every 3 days
and the ozonation experiment was immediately carried
out in a lab scale 10-L cylindrical contactor. The
ozonized sludge was stored in a temperature-controlled
refrigerator at 4 1C after a 1:20 dilution with tap water.
The mixed liquor sample from the pilot plant was used
as seed sludge. The major operating parameters of both
CSTRs were identical to those of the pilot plant; an 8 h
HRT, with sludge recirculation and no sludge wasting.

3.1. Determination of the major operating parameters

(SO and n)
The relationships between the cumulative removed
COD and sludge mass are shown in Fig. 2. The two
slopes shown in Fig. 2(a) and (b) represent the Yobs
values for the wastewater and ozone-treated sludge,
respectively. There was no signicant difference between
the Yobs values; 0.40 g SS/g COD for wastewater and
0.42 g SS/g COD for ozonized sludge. Those Yobs values
could be converted into 0.48 g COD/g COD for wastewater and 0.50 g COD/g COD for ozonized sludge
assuming the conversion biomass to COD as 1.20 g
COD/g SS. Conversely, these Yobs data indicated that
the mineralization rate of the wastewater and ozonized
sludge was 52% and 50%, respectively. It was a
noticeable result that the biodegradability of raw sludge
after ozonation was enhanced to that of wastewater.
Sludge transformation after ozonation was evaluated by
means of examining three forms of CODs; total,
particulate and soluble COD. With 0.05 kg O3/kg SS
ozone contact, the raw sludge was signicantly transformed resulting in 8% mineralization, 22% solubilization, and 70% residuals based on COD.
Given the Yobs and sludge transformation data, we
could establish a mass balance for the activated sludge
system coupled with ozonation as shown in Fig. 3.

ARTICLE IN PRESS
J.W. Lee et al. / Water Research 39 (2005) 11991204

1202

Cumulative sludge increase (g)

Inevitably, different dimensions of Yobs for the wastewater (kg/kg COD) and ozonized sludge (kg COD/kg
COD) were used for the calculation. It is easy to obtain
the sludge ozonation frequency of n from the mass
balance shown in Fig. 3. The ideal system would not
35

0:50  0:92SO 0:08SO SP:

30

Then, n could be obtained using Eqs. (2) and (3) as

follows;

25

ntheo: SO=SP 1=0:50  0:92 0:08 1:85.

20

10
5
0
0

Cumulative sludge increase (g)

20

40

60

80

Cumulative COD removal (g)

6
5
4
3

y = 0.4153x
R2 = 0.9589

2
1
0
0

10

15

Cumulative COD removal (g)

Fig. 2. Relations between the cumulative sludge increase and

COD removal of (a) wastewater and (b) ozone-treated sludge.

Sludge production
rate (SP, kg/day)

A pilot-scale activated sludge system was operated for

112 days in order to verify the theoretically determined
n. Experimentation began during the winter season and
the period continued until into the beginning of Koreas
spring season. Fig. 4 shows the theoretically calculated
SO, applied SO and applied n for overall period. For the
rst 50 days (phase I), n was targeted to the theoretically
obtained 1.85. However, the actual applied average n
was found to be 2.66 corresponding to 1.44 times higher
than the theoretical value. Application of a slightly
higher value of n was associated with the concerns of low
biodegradability in the pilot-scale experiment due to low
temperature and low dissolved oxygen supply.
For phase II, n was increased about double of that of
phase I because a signicant sludge accumulation
continued throughout phase I. The average applied n
Sludge ozonation
rate (SO, kg/day)
Ozone Contactor

Bioreactor

Yobs for ozonated sludge

= 0.50(kg COD/kg COD)

Mineralization
0.08 SO

Mineralization
0.50(0.92 SO)
Biomass or
Residual Particulates
0.50(0.92 SO)
Biodegradation

(4)

3.2. Pilot plant verification

Wastewater
Yobs for wastewater
= 0.40(kg/kg COD)

(3)

Eventually, at least twice the amount of SP should be

ozonized in order to accomplish a no-excess sludge
production system. This strategic approach is applicable
due to the fact that the SO is controllable depending on
daily or seasonally varied sludge production.

y = 0.3962x
R2 = 0.9499

15

(a)

(b)

produce any excess sludge if the identical amount to SP

would be mineralized either biologically or chemically.
Based on this hypothesis, if we set the SP to be equal to
the sum of the mineralization rates occurring in
ozonation and biodegradation, the following equation
could be established;

Solubilization
0.22 SO
Ozone treated
sludge recirculation
rate (kg/day)

Residual
Particulates
Ozonated
sludge
0.70 SO

Fig. 3. Sludge mass balance in activated sludge system coupled with sludge ozonation process.

ARTICLE IN PRESS
J.W. Lee et al. / Water Research 39 (2005) 11991204
5.00

Phase I

Phase II

6000

12

Phase III

1203

SOcal

5000

SOapp

6
3.00

4
2

2.00

0
1.00

Concentration (mg/l)

4.00

napp

Ozonation rate (kg/day)

10
y = 12389e-0.101x
2
R = 0.728

4000
3000
2000

y = 8891e-0.1058x
2
R = 0.7151

MLSS

0.00

1000
0

20

40

60
80
Time (days)

100

Fig. 4. Applied major operating parameters of SO and n in

pilot-scale activated system and comparison between the
applied SO and theoretically calculated SO.

Phase I

Concentration (mg/l)

7000

Phase II

Phase III

30

MLSS

6000

MLVSS

25

5000

20

4000
3000

15

2000
10

1000
0
0

20

40

60
80
Time (days)

100

10

11
12
13
Temperature (oC)

14

15

Fig. 6. Correlations of sludge concentrations and temperatures

during phase I.

35

Temperature (oC)

8000

MLVSS

120

5
120

Fig. 5. Sludge concentrations in pilot-scale activated sludge

and temperatures conditions during operation period.

was found to be 5.74. This corresponded to 3.11 times of

the theoretical n.
Fig. 5 shows the variations of sludge concentrations
and temperatures during the entire operating period.
During phase I, signicant increase of sludge concentration was observed. Sludge accumulation was thought to
be due mainly to insufcient acclimation period and low
temperatures. The system might require a sufcient
acclimation period to reach a steady state for biodegradation of the ozonated sludge. Also, decrease of
temperatures increased the sludge yields. The latter part
of phase I corresponded to a severely cold winter season
in Korea. The Mixed liquor suspended solids (MLSS)
increased from 3200 mg/L up to 5300 mg/L for 42 days
with temperature decrease from 13.7 to 9.5 1C.
Dependence of sludge concentration on temperature
during phase I is demonstrated in Fig. 6. The sludge
concentration in the mixed liquor was exponentially
correlated to the temperature between 9.5 and 13.7 1C
even though it was not clear whether the temperature
solely affected the sludge concentration in this study.
However, there is no doubt that biodegradation of

particulates is strongly affected by temperature. It was

reported that volatile suspended solids (VSS) reduction
efciency is signicantly decreased proportioning to
temperature decrease (Eikum et al., 1974; Koers and
Mavinic, 1977).
Increasing n during phase II resulted in a decrease of
MLSS to 4400 mg/L for 15 days at the beginning of
phase II even though the temperature remained between
10 and 12 1C as shown in Fig. 5. Thereafter, no
signicant MLSS increase was observed. MLSS concentration was stable for the remainder of phase II.
During phase III, n returned to a value similar to that
observed during phase I because the temperature
stabilized at 15 1C. Although temperatures were below
a favorable range for biodegradation, no further
increase of the MLSS was observed and the MLSS level
was stable between 5000 and 5300 mg/L. In spite of high
sludge concentration, the accumulation of inorganic
compounds in the bioreactor was not signicant and the
volatile fraction of the MLSS was maintained between
0.65 and 0.70 during the operating period.
Conclusively, it is desirable to apply n exibly
depending on the temperature. In this study, the
available n was between 2.5 and 2.7 at a relatively low
temperature of 15 1C. However, n should be increased,
perhaps double, if the temperature decreases to 10 1C or
lower. Even though the increase of the ozonation
frequency compensated for the lowered biodegradability
at low temperature, much care should be taken in the
determination of n. Overestimation of n may lead to
diminution of the active biomass due to the increase of
sludgeozone contact frequency. Inactivation of heterotrophic organisms at different ozone concentrations was
investigated by means of a heterotrophic plate counting
method (APHA, 1995). Inactivation due to ozone was
found to be large. Ninty-seven percentage of heterotrophic organisms with 0.05 kg O3/kg SS dosage (data

ARTICLE IN PRESS
J.W. Lee et al. / Water Research 39 (2005) 11991204

1204

Effluent concentration (mg/l)

TCOD eff

100

SCODeff

Phase I

SSeff

Phase II

TCODin
Phase III

200

80
150
60
100
40
50

20
0

Influent concentration (mg/l)

250

120

0
0

20

40

60

80

100

energy consumption are considerably subject to those

operating parameters.

Acknowledgements
This study was supported by the Ministry of Science
and Technology (MOST) of Korea.

References

120

Time (days)

Fig. 7. Efuent COD and SS in the pilot-scale activated sludge

system coupled with sludge ozonation.

not shown) were inactivated. Also, the increase of n

consumes a large amount of energy; e.g., doubling n
required average 21 kWh more energy consumption in a
day, which means that the energy consumption for the
whole pilot plant facility increased by 24%.
3.3. Effluent water quality
Most of the soluble organic matter originating from
either the wastewater or the ozonized sludge was
completely degraded in the bioreactor. Fig. 7 shows the
efuent total COD (TCOD), suspended solids (SS) and
inuent TCOD concentrations. Variations of the efuent
COD and SS was not signicant regardless of the inuent
uctuations. Also, the recirculation of the ozonized
sludge did not deteriorate the efuent water quality
during the entire operating period. Even during the
highly recirculated period of phase II, the efuent SS and
SCOD remained low at 10 and 15 mg/L, respectively.

4. Conclusions
Performance of a pilot-scale activated sludge system
pursuing zero excess sludge production can be optimized
with the operational strategy based on either sludge
ozonation rate or ozonation frequency. The sludge
ozonation rate should be dynamically manipulated for
real wastewater treatment because sludge ozonation may
be dynamic depending on the variations of inuent
organic strength. The theoretically obtained ozonation
frequency should be substantially adjusted during either
the winter season or the incipient operation since the
biodegradation of particulate organic matter was lower
with colder temperatures and insufcient acclimation to
the ozonated sludge. It is important to apply an optimal
value of sludge ozonation rate (including ozonation
frequency) for activated sludge system coupled with
ozonation because both the system performance and

Ahn, K.H., Park, K.Y., Maeng, S.K., Hwang, J.H., Lee, J.W.,
Song, K.G., Choi, S., 2002a. Ozonation of wastewater
sludge for reduction and recycling. Water Sci. Technol. 46
(10), 7177.
Ahn, K.H., Yeom, I.T., Park, K.Y., Maeng, S.K., Lee, Y.H.,
Song, K.G., Hwang, J.H., 2002b. Reduction of sludge by
ozone treatment and production of carbon source for
denitrication. Water Sci. Technol. 46 (1112), 121125.
APHA, 1995. Standard Methods for the Examination of Water
and Wastewater, 19th ed. American Public Health Association, Washington, DC.
Deleris, S., Paul, E., Audic, J.M., Roustan, M., Debellefontaine,
H., 2000. Effect of ozonation on activated sludge solubilization and mineralization. Ozone Sci. Eng. 22 (5), 473486.
Eikum, A.S., Carlson, D.A., Paulsrud, B., 1974. Aerobic
stabilization of primary and mixed primary-chemical (alum)
sludge. Water Res. 8 (11), 927935.
Huysmans, A., Weemaes, M., Fonseca, P.A., Verstraete, W.,
2001. Ozonation of activated sludge in the recycle stream.
J. Chem. Technol. Biotechnol. 76 (3), 321324.
Koers, D.A., Mavinic, D.S., 1977. Aerobic digestion of waste
activated sludge at low temperatures. J WPCF 49, 460468.
Muller, J.A., 2000. Pretreatment processes for the recycling and
reuse of sewage sludge. Water Sci. Technol. 42 (9), 167174.
Muller, J.A., Lehne, G., Schwedes, J., Battenberg, S., Naveke,
R., Kopp, J., Dicht, N., Scheminski, A., Krull, R., Hempel,
D.C., 1998. Disintegration of sewage sludges and inuence on
anaerobic digestion. Water Sci. Technol. 38 (89), 425433.
Park, K.Y., Ahn, K.H., Maeng, S.K., Hwang, J.H., Kwon,
J.H., 2003. Feasibility of sludge ozonation for stabilization
and conditioning. Ozone Sci. Eng. 25 (1), 7380.
Scheminski, A., Krull, R., Hempel, D.C., 2000. Oxidative
treatment of digested sewage sludge with ozone. Water Sci.
Technol. 42 (9), 151158.
Weemaes, M., Grootaerd, H., Simoens, F., Verstraete, W.,
2000. Anaerobic digestion of ozonized biosolids. Water Res.
34 (8), 23302336.
Yasui, H., Shibata, M., 1994. An innovative approach to reduce
excess sludge production in the activated sludge process.
Water Sci. Technol. 30 (9), 1120.
Yasui, H., Nakamura, K., Sakuma, S., Iwasaki, M., Sakai, Y.,
1996. A full-scale operation of a novel activated sludge
process without excess sludge production. Water Sci.
Technol. 34 (34), 395404.
Yeom, I.T., Lee, K.R., Lee, Y.H., Ahn, K.H., Lee, S.H., 2002.
Effects of ozone treatment on the biodegradability of sludge
for municipal wastewater treatment plants. Water Sci.
Technol. 46 (45), 421425.