Iran. J. Environ. Health. Sci. Eng., 2011, Vol. 8, No. 1, pp.

25-34

DETERMINATION OF BIOKINETIC COEFFICIENTS FOR

ACTIVATED SLUDGE PROCESSES ON MUNICIPAL WASTEWATER
1
Sh. Mardani., 1A. Mirbagheri, *2M. M. Amin., 2M. Ghasemian
1
Department of Civil & Environmental Engineering, Khaje-Nasir University of Technology, Tehran, Iran.
2
Environment Research Center, Isfahan University of Medical Sciences, Isfahan, Iran.

Received 21 Jully 2009; revised 14 December 2010; accepted 18 October 2010

INTRODUCTION
In the past, designs of biological wastewater volume, substrate utilization, biomass growth,
treatment processes were based on the empirical and the effluent quality can be calculated from
parameters developed by experience, which those equations (Qasim, 1999).
included hydraulic loading, organic loading and Biokinetic coefficients used in the design of
retention time. Nowadays, the design utilizes activated sludge processes include specific
empirical as well as rational parameters based growth rate(µ), maximum rate of substrate
on biological kinetic equations. Theses equations utilization per unit mass of microorganisms (k:
describe growth of biological solids, substrate 2-10 1/day), half velocity constant (Ks: 10-60
utilization rates, food-to-microorganisms ratio, mg COD/L), maximum cell yield (Y: 0.3-0.6
and the mean cell residence time. Reactor mgVSS/mg COD), and endogenous decay
*Corresponding author: E-mail: [email protected] coefficient (kd: 0.06-0.15 1/day)(Metcalf and
Tel:+98-913-367-0508 Eddy, 2003).

25
 Sh. Mardani et al., dEtErMINatIoN of bIokINEtIc coEffIcIENtS...

In the different wastewater treatment processes, model, which is considered as the most commonly
biokinetic coefficients have been evaluated by and widely used for determining the biokinetic
several investigators (Henze, 1992; San, 1992; coefficients.
Karin Kovarova-Kovar, 1998). Muhammad Al- In continuous-flow and completely-mixed reactor,
Malack determined biokinetic coefficients of an determination of the biokinetic coefficients is
immersed membrane bioreactor for municipal usually achieved by collecting data from lab-scale
wastewater treatment. Biokinetic coefficients or pilot-scale experimental setups operated at
determined under different MLSS concentrations various hydraulic retention times (HRTs) and/or
and organic loading rates (Al-Malack, 2006). at various sludge retention times (SRTs) (Metcalf
Naghizadeh determined biokinetic parameters and Eddy, 2003).
in municipal wastewater treatment with The main objective of this study was the
a submerged membrane reactor by the determination of kinetic parameters Y, kd, µmax
monod equation. The results showed and Ks for activated sludge processes through
that Y, kd, Ks and µmax coefficients were three types including: conventional, extended
0.67 mgVSS/mgCOD, 0.5 1/day, 65.5 mg/L and aeration and contact stabilization activated sludge
1.86 1/day, respectively (Naghizadeh, 2008). systems on municipal wastewater in pilot-scale.
Eleni Vaiopoulou studied kinetics of The practical objective was also the efficiency
nitrification, denitrification and dephosphatation evaluation of one of the big Isfahan wastewater ,
in the aerobic, anoxic and anaerobic stages of WWTP, treatment plants (Isfahan South WWTP)
a pilot scale biological nutrient removal plant with conventional activated sludge process, which
treating municipal wastewater. The results provides services for about 1 million persons.
showed that the nitrification process follows
Monod type kinetics (Eleni Vaiopoulou and MATERIALS AND METHODS
Alexander Aivasidis, 2007). description of the systems setup
Talaie Khozani studied the determination of Fig.1 shows the diagram of activated sludge
biokinetic coefficients of crude oil biodegradation processes (conventional, extended aeration and
using Pseudomonas aeruginosa bacteria. The contact stabilization) in pilot-scale used in this
biokinetic coefficients based on modified Monod study. Each activated sludge system in pilot-scale
equation were calculated. The results showed that included aeration tank, secondary sedimentation
kd, Y, k and Ks were equal 0.107 1/day, 0.882 tank and auxilary apparatus. The effective
mg/L, 9.39 1/day and 169.3 mg/L, respectively volume of each reactor in the conventional and
(Talaie et al., 2010). Pala and Bolukbas evaluated extended aeration activated sludge processes was
the kinetic parameters for biological CNP removal 300 L, whereas effective volume of contact basin
from a municipal wastewater through batch and stabilization basin in contact stabilization
tests (Pala and Bolukbas, 2005). Yenkie et al process was 54 L and 238 L, respectively.
determined biokinetics of wastewater treatment Influent wastewater to extended aeration and
in the high performance compact reactor (HCR) contact stabilization systems was raw wastewater
using different cell residence times (Yenkie et al., that was passed of screen and grit chamber, but
1992). Joseph and Malina determined biokinetic influent wastewater to conventional system was
coefficients for hydraulic retention time of 24 h the effluent wastewater of primary sedimentation
in high-rate biological treatment of wastewater tank. For providing and maintaining aerobic
at the Pantex facility (alternative to the existing conditions in the reactor, stone air diffusers were
aerated lagoon-pond system), (Joseph and fixed in the bottom of basins. The operation of
Malina, 1999). pilot plant was programmed to switch on and off
Basic equations that describe the interaction by Programming Logic Control.
between the growth of microorganisms and A steady-state condition was assumed to be
utilization of the growth limiting substrate in reached when fairly constant biomass growth
activated sludge processes are based on Monod and permeate COD were attained. During the

26
 Iran. J. Environ. Health. Sci. Eng., 2011, Vol. 8, No. 1, pp. 25-34

Conventional activated sludge

Feed tank
Primary Secondary
sedimentation Aeration tank sedimentation

Return activated sludge

PLC
Extended aeration process

Sewage
Aeration tank
Raw wastewater

Return activated sludge

Contact stabilization process

PLC

Contact
Stabilization

tank
Air pipe tank
Sludge return
Return activated sludge
process
Slide valve

Feed pump

Air blower

Angle valve
Fig.1: Schematic diagram of three activated sludge processes used in this study

first stage of the investigation, the mixed liquor analytical methods

suspended solids concentration, in the aeration Samples from the influent wastewater, reactor
tank was kept constant. The first steady-state and effluent were analyzed for various physical
condition was achieved after 25 days from the start and chemical parameters in accordance with
of the unit operation. The steady-state condition the Standard Methods (APHA, 2005). The TSS,
for each SRT during first stage in each process VSS and sCOD parameters were experimented
was maintained for about 10 days, and during the in influent and effluent wastewater every day
second stage, the steady-state condition in each but BOD5, sBOD5 and COD parameters were
process was maintained for about 5 days. tested 3 times per week. Also temperature in
the bioreactors was held about 20 °C and DO
Wastewater characteristics (dissolve oxygen) was 2 mg/L in the biological
Raw and effluent wastewaters of the primary reactors. After sampling, the data were analyzed
sedimentation tank from Isfahan South WWTP by Excel and SPSS softwares.
were used for feeding the reactors in pilot plants.
The general characteristics of the raw wastewater Experimental procedure
used in the study are shown in Table 1. Initial seeding of each bioreactor (conventional
and extended aeration) was accomplished by
Table 1: Characteristics of the
raw wastewater
inoculating the bio-reactor with 100 L, in contact
stabilization process: the contact basin with 54 L
Parameter Value (mg/L)
and the stabilization basin with 70 L of returned
BOD5 240
COD 575 activated sludge, respectively. During the first 25
sBODave 85.3 days of operation, the biomass was allowed to
sCODave 180.25 acclimatize to the influent substrate.
TSSin 226 The investigation was carried out at different
pHin 7.64
MLSS concentrations in each process, as:
sBOD/sCOD 0.47
VSS/TSS 0.81 -Conventional process (2000 and 3000 mg/L

27
 Sh. Mardani et al., dEtErMINatIoN of bIokINEtIc coEffIcIENtS...

-Extended aeration process (4000 and 5000 S: Effluent substrate concentration, mg sCOD/L
mg/L) X: Biomass concentration, mg VSS/L
-Contact stabilization process (2000 and 3000 Ө: Hydraulic retention time, d
mg/L) Ks: Half-velocity constant, mg sCOD/L
Each MLSS value was subjected to five different k: Maximum rate of substrate utilization, mg
sludge retention times (SRT). Sludge was sCOD/mg VSS.d
withdrawn from the sedimentation tank once a
day to keep the SRT at the designated values. RESULTS
Biokinetic coefficients were determined by conventional activated sludge process
collecting data from pilot-scale experiments During the first stage of the investigation in
and under equations. MLSS concentration was conventional activated sludge process, the mixed
assumed steady in each process and section; liquor suspended solids concentration in the
then flowrate and SRT process computation were aeration tank was kept at about 2000 mg/L.Table
changed, but the tank volume was constant in 2 shows the steady-state data obtained in aeration
each process. tank, while Figs. 2 and 3 show the determination
of the coefficients. The biokinetic coefficients
1 S0 -S
= Y U - Kd = Y - Kd (1) were found to be as follows: Y=0.4872 mg
SRT θX VSS/mg sCOD, kd=0.0258 1/day, KS=52.1 mg
sCOD/L and k=1.95 1/day. The effect of F/M on
θX Ks 1 1 1 the removal efficiency of COD was significant,
= + = (2) where COD removal was found to range between
S0 -S K S K U
83% and 92.5%.
Results of the steady-state conditions of the
where: conventional process operating at MLSS
SRT: Solids retention time, d concentration of 3000 mg/L, are presented in
Y: Biomass yield, mg VSS/mg sCOD Table 3. In order to determine the biokinetic
U: Substrate utilization rate, mg sCOD/mg VSS.d coefficients, Figs. 4 and 5 are plotted. The values
kd:Endogenous decay coefficient, 1/d of the biokinetic coefficients were found to be as
S0:Influent substrate concentration, mg sCOD/L follows: Y=0.804 mg VSS/mg sCOD, kd=0.0189

Table 2: Operational parameters at MLSS=2000 mg/L in aeration tank

Steady-state Q HRT MLVSS sCODin sCODout SRT F/M

period (L/h) (h) (mg/L) (mg/L) (mg/L) (day) (1/day)
1-12 38 7.89 1505 127.66 9.2 9.21 0.261
18-31 45 6.67 1316 145.48 13.18 7.097 0.403
36-50 56 5.36 1440 138.9 18.63 4.237 0.44
57-68 66 4.55 1442 128.43 21.64 3.15 0.479

4.5
0.35 y = 26.64x + 0.5118
4.0
y = 0.4872x - 0.0258 R 2 = 0.9021
1/U sCOD)

0.30 R2 = 0.9526 3.5

1/SRT, (1/day)

0.25 k=1.95
1/d

1/U (mg VSS.d/mg

kd=0.0258 3.0
kks=52.1
s
=52.1
Y=0.4872
0.20 2.5
0.15 2.0

0.10 1.5

0.05 1.0
0.20 0.30 0.40 0.50 0.60 0.70 0.02 0.04 0.06 0.08 0.10 0.12 0.14
1/S,
1/S,(1/mg sCOD)
U, (mg sCOD/mg VSS.d
mg sCOD/mg VSS.day) mg s/COD

Fig. 2: D etermination of Y a nd kd Fig. 3: Determination of Ks and k

28
 Iran. J. Environ. Health. Sci. Eng., 2011, Vol. 8, No. 1, pp. 25-34

Table 3. Steady-state data in MLSS concentration of 3000 mg/L

Steady-state Q HRT MLVSS sCODin sCODout SRT F/M

period (L/h) (h) (mg/L) (mg/L) (mg/L) (day) (1/day)
78-86 38 7.89 2127 110.2 8 12.25 0.158
93-101 42 7.14 2164 103.4 9.3 11.25 0.161
107-115 47 6.38 2057 106 13.4 9.23 0.195
120-129 54 5.56 2172 115.8 16 6.17 0.231
135-143 66 4.55 2232 128 19.9 5.04 0.303

0.23
y = 0.0042x + 0.0523
0.20 2 10
R = 0.8805
y = 58.509x + 0.8226
9 2
R = 0.9249
1/SRT,(1/day)

0.17 kd=0.0189
8
1/d

Y=0.8035 K=1.22
ks= 71.12
KS=71.12
0.14 7

1/U
6
0.11
5
0.08
4

0.05 3
0.10 5.10 10.10 15.10 20.10 25.10 30.10 35.10 40.10 0.04 0.06 0.08 0.10 0.12 0.14
U, (mg sCOD/mgVSS.d
mg sCOD/mg VSS.day) 1/S, mg s/COD
(mg VSS.d/mg sCOD)

Fig. 4: Determination of Y and kd Fig. 5: Determination of Ks and k

1/day, KS=71.12 mg sCOD/L and k=1.22 1/day. During the second phase, the MLSS
The effect of F/M on the removal efficiency of concentration was increased to 5000 mg/L.
COD was significant, where COD removal was Table 5 shows data obtained at steady-
found to range between 84.4% and 92.7%. state conditions and Figs. 8 and 9 show the
determination of biokinetic coefficients. The
Extended aeration activated sludge process values of the biokinetic coefficients obtained at
During the first stage of the investigation in MLSS of 5000 mg/L were as follows: Y= 1.25 mg
exetended aeration process, the mixed liquor VSS/mg sCOD, k d=0.0198 1/day, K S=508 mg
suspended solids concentration in the aeration sCOD/L and k=2.53 1/day. The effect of F/M
tank was kept at about 4000 mg/L. Table 4 on the removal efficiency of COD was not
shows the steady-state data obtained at MLSS significant, where COD removal was found
concentration of 4000 mg/L, while Figs. 6 and to range between 91.3% and 93.76%.
7 show the determination of the coefficients.
The biokinetic coefficients were found to be as contact sabilization activated sludge process
follows: Y=0.6182 mg VSS/mg sCOD, kd=0.0308 During the first stage of the investigation in
1/day, KS=311.7 mg sCOD/L and k=3.17 1/day. contact stabilization process, MLSS in the contact
The effect of F/M on the removal efficiency of and stabilization tanks, were kept at about 2000
COD was not significant, where COD removal and 5000mg/L, respectively. Table 6 shows the
was found to range between 88% and 92.5%.

Table 4: Steady-state data at MLSS=4000 mg/L in the aeration tank

Steady-state Q HRT MLVSS sCODin sCODout SRT F/M

period (L/h) (h) (mg/L) (mg/L) (mg/L) (day) (1/day)
1-12 11 27.27 1795 175 13.09 40.18 0.087
18-31 12 25 2167 216.32 16.77 35.36 0.101
36-53 13 23.08 2298 223.9 19.2 25.26 0.101
59-72 14 21.43 1866 214.57 25.17 23.11 0.129
78-89 15 20 1864 171.63 16.95 20.37 0.111

29
 Sh. Mardani et al., dEtErMINatIoN of bIokINEtIc coEffIcIENtS...

0.055 12
y = 0.0006x + 0.0217 y = 98.333x + 0.3156
0.050

sCOD)
2 2
R = 0.9141 11 R = 0.9749
(1/day)

0.045 k=3.17
kd=0.0308
1/SRT, 1/d

ks=311.7
Ks=311.7

1/U (mg VSS.d/mg

10
0.040 Y=0.6182

1/U
0.035 9
0.030
8
0.025
0.020 7
0.08 10.08 20.08 30.08 40.08 50.08 60.08 0.07 0.08 0.09 0.10 0.11 0.12
1/S,
1/S(1/mg sCOD)
,mg s/COD
U, (mg sCOD/mgVSS.d
mg sCOD/mg VSS.day)

F ig. 6: D etermination of Y a nd kd Fig. 7: Determination of Ks and k

Table 5. Operational parameters in MLSS=5000 mg/L aeration tank

Steady-state Q HRT MLVSS sCODin sCODout SRT F/M
period (L/h) (h) (mg/L) (mg/L) (mg/L) (day) (1/day)
95-101 11 27.27 2886 123.76 7.66 40.48 0.038
107-113 12 25 2820 134.2 9.23 35.9 0.046
117-123 13 23.08 2980 135.25 10.89 25.4 0.047
129-135 14 21.43 3053 147.07 12.9 22.47 0.054
138-143 16 18.75 3131 150.88 11.9 20.38 0.062

0.055

0.050 y = 1.2512x - 0.0198

2 31
R = 0.9086
y = 200.77x + 0.3957
1/SRT, (1/day)

0.045 29
1/U sCOD)

2
kd=0.0198 R = 0.9399
1/d

27
0.040 Y=1.2512 k=2.53
25 kKs=508
=508
1/U (mg VSS.d/mg

s
0.035 23

0.030 21
19
0.025
17
0.020 15
0.030 0.035 0.040 0.045 0.050 0.055 0.060 0.08 0.09 0.10 0.11 0.12 0.13 0.14 0.15

U, (mg sCOD/mgVSS.d
mg sCOD/mg VSS.day) 1/S,1/S,
(1/mg
mg sCOD)
s/COD

Fig. 8: Determination of Y and kd Fig. 9: Determination of Ks and k

steady-state data obtained at MLSS of 2000 mg/L coefficients obtained at an MLSS of 3000 mg/L
in the contact tank, while Figs. 10 and 11 show the were as follows: Y=0.63 mg VSS/mg sCOD,
determination of the coefficients. The biokinetic kd=0.0387 1/day, KS=13.8 mg sCOD/L and
coefficients were found to be as follows: Y=0.713 k=0.366 1/day. The effect of F/M on the removal
mg VSS/mg sCOD, kd=0.0172 1/day, KS=50.8 mg efficiency of COD was significant, where COD
sCOD/L and k=0.586 1/day. The effect of F/M removal was found to range between 77.86% and
on the removal efficiency of COD was significant, 92%.
where COD removal was found to range between
77% and 91.7%. DISCUSSION
During the second phase, MLSS in the contact Based on the biokinetic results that are shown in
and stabilization tanks were increased to 3000 Figs. (2-13), Table 8 presents a summary of the
and 6000 mg/L, respectively. Table 7 shows data biokinetic coefficients obtained for three activated
obtained at steady-state conditions and Figs. sludge processes and for two MLSS values in
12 and 13 show the determination of biokinetic each process. The Table clearly shows that values
coefficients. The values of the biokinetic of the biokinetic coefficients vary significantly

30
 Iran. J. Environ. Health. Sci. Eng., 2011, Vol. 8, No. 1, pp. 25-34

0.25
0.23 y = 0.713x - 0.0172 10
2
0.21 R = 0.9805 9 y = 86.748x + 1.7073

sCOD)
1/SRT, (1/day)

2
0.19 R = 0.8691
kd=0.0172 8
1/d

0.17 k=0.586

1/U (mg VSS.d/mg

Y=0.713 7
0.15 ks=50.8
Ks=50.8
6

1/U
0.13
5
0.11
0.09 4
0.07 3
0.05 2
0.10 0.15 0.20 0.25 0.30 0.35 0.015 0.035 0.055 0.075 0.095
U, (mg sCOD/mg VSS.d
mg sCOD/mg VSS.day) 1/S,
1/S,(1/mg sCOD)
mg s/COD

Fig. 10. Determination of Y and kd Fig. 11. Determination of Ks and k

Table 6. Steady-state data in MLSS of 2000 mg/L in contact tank

Steady-state Q HRT MLSS MLVSS sCODin sCODout SRT F/M

period (L/h) (h) (mg/L) (mg/L) (mg/L) (mg/L) (day) (1/day)
1-12 53 1.015 1832 1199 175 14.7 15.44 0.241
18-31 63 0.85 1906 1249 216.32 31.25 10.6 0.355
36-53 75 0.717 1935 1275 223.9 34.93 7.1 0.42
59-72 88 0.609 1773 1181 214.57 43.99 6.26 0.546
78-89 103 0.522 1794 1172 171.63 39.68 4.486 0.526

Table 7: Steady-state data at MLSS of 3000 mg/L in contact tank

Steady-state Q HRT MLVSS sCODin sCODout SRT F/M

period (L/h) (h) (mg/L) (mg/L) (mg/L) (day) (1/day)
95-101 56 0.96 2077 123.76 9.83 18.3 0.16
107-113 62 0.87 1994 134.18 13.36 13.5 0.189
117-123 74 0.73 2068 135.25 19.6 10.2 0.212
129-135 85 0.63 2071 147.07 26.9 9.35 0.294
138-143 100 0.54 2223 150.88 33.42 7.39 0.453

0.14

y = 0.6322x - 0.0387
R2 = 0.9649 7
y = 37.682x + 2.7286
sCOD)
(1/day)

2
0.11 kd=0.0387 R = 0.9366
6
1/SRT, 1/d

Y=0.6322
k=0.366
1/U (mg VSS.d/mg

kKs=13.8
s
=13.8
1/U

5
0.08

0.05 3
0.14 0.16 0.18 0.20 0.22 0.24 0.26 0.28
0.02 0.04 0.06 0.08 0.10 0.12
U,(mg sCOD/mg
mg sCOD/mg VSS.day)
VSS.d 1/S,
1/S,(1/mg sCOD)
mg s/COD
Fig. 12. Determination of Y and kd Fig. 13. Determination of Ks and k

with the change in MLSS concentration in each Generally, values of the biokinetic coefficients
process. Hovever, this variability does not follow presented in Table 8 for conventional activated
any definite pattern. This could be attributed to sludge process (Figs. 2-5) are within the normal
the nature of the process itself, as it could be a range reported for the conventional activated
selective one and the biokinetic coefficients sludge processes.
obtained may represent different species .

31
 Sh. Mardani et al., dEtErMINatIoN of bIokINEtIc coEffIcIENtS...

Table 9 summarizes some of the biokinetic indicate that the amount of sludge produced at
coefficients obtained from different sources. higher MLSS values is not reduced. Increasing the
Other values for the biokinetic coefficients can MLSS concentration was also found to increase
be found in Benefield and Randall (Benefield and the maximum rate of growth of the biomass and
Randall, 1980) and Grady and Lim (Grady and saturation constant, which could be also attributed
Lim, 1980). to the same reason given before.
Although, Y, kd, Ks and µmax coefficients for Values of the biokinetic coefficients presented
conventional process are within the reported values in Table 8 for extended aeration process, except
for conventional activated sludge processes, for Ks and Y in MLSS=5000 mg/L, are within
they differ quite significantly. The values of Y the normal range reported for the conventional
in conventional process were increasing with activated sludge processes. Values of Ks, especially
the increase in MLSS concentrations, since they at MLSS of 5000 mg/L, were much higher than
represent all the amount of biomass produced by those reported in the literature (Table 9). This
the growth during the removal of the substrate. is attributed to the fact that the determination
This clearly shows that the type of substrate of the Ks is affected by estimation of the decay
and bacterial can have a significant effect on the rate, kd, thus any uncertainty in estimating kd will
determination of the biokinetic coefficients. On be reflected on the Ks value. In this system also
the other hand, kd was found to decrease with the values of Y were increasing with the increase
increase in the MLSS concentration, which could in MLSS concentrations. Values of Y at MLSS

Table 8: Monod kinetic coefficients for three activated sludge processes at different MLSS concentrations

Process
MLSS Y Kd Ks k µ max
µ
(mg/L) (mg/mg) (1/day) (mg /L) (1/day) (1/day)
2000 0.4872 0.0258 52.1 1.95 0.95
Conventional
3000 0.8035 0.0189 71.12 1.22 0.98
Extended
Extended 4000 0.6174 0.0309 311.7 3.17 1.96
aeration
aeration 5000 1.2512 0.0198 508 2.53 3.17
Contact
Contact 2000 0.713 0.0172 50.8 0.586 0.42
stabilization
stabilization 3000 0.6322 0.0387 13.8 0.366 0.23

Table 9: Kinetic coefficients obtained from different sources at COD

kd µµ max
Substrate Y (mg/mg) Ks (mg /L) Source
(1/day) (1/day)
Domestic 0.31-0.35 0.016-0.068 1.7 43-223 Pala, 2005
Domestic 0.67 0.07 3.75 22 Lawrence, 1970
Glucose
Glucose 0.5-0.62 0.025-0.48 7.4-18.5 11-181 Al-Malack, 2006
Synthetic
Synthetic 0.42-0.53 0.05-0.19 0.8-6.3 83-646 Al-Malack, 2006
MWW
MWW 0.46-0.6 0.05-0.16 5.6-8.1 250-3720 Al-Malack, 2006
MWW
MWW 0.4-0.8 0.025-0.075 2-10 15-70 Metcalf&Eddy, 1991
Domestic
Domestic 0.4-0.67 0.07-0.09 3.2-3.75 22-60 Pala, 2005
Synthetic
Synthetic 0.49-0.58 0.037-0.151 1.28-6.46 289-2933 Al-Malack, 2006
MWW
MWW 1.78 0.12 0.28 36.6 Joseph, 1999
MWW
MWW 0.67 0.5 1.86 65.5 Naghizadeh, 2008
MWW
MWW 0.49-0.804 0.019-0.026 0.95-0.98 52-71.12 This study (conventional)
MWW
MWW 0.62-1.25 0.02-0.031 1.96-3.17 311.7-508 This study (extended aeration)
MWW
MWW 0.63-0.713 0.017-0.039 0.23-0.42 13.8-50.8 This study (contact stabilization)

32
 Iran. J. Environ. Health. Sci. Eng., 2011, Vol. 8, No. 1, pp. 25-34

of 5000 mg/L were much higher than those were decreasing with the increase in MLSS
reported in the literature (Table 9). This clearly concentrations, because the amount of biomass
shows that the type of substrate and bacterial can produced by the growth during the removal of
have a significant effect on the determination the substrate in stabilization tank decreases with
of the biokinetic coefficients. Sludge retention increasing in MLSS concentration. On the other
time was high in extended aeration system, that hand, kd was found to increase with the increase
cause reduction in substrate. The presence of in the MLSS concentration, which could indicate
many vorticella coloni, rotifer and nematode in that the amount of sludge produced at higher
this SRT in substrate at MLSS of 5000 mg/L, MLSS values is reduced. Increasing the MLSS
increases the yield coefficient (Y). Increasing the concentration was also found to decrease the
MLSS concentration was also found to increase maximum rate of growth of the biomass and
the maximum rate of growth of the biomass and saturation constant, which could be also attributed
saturation constant, which could be also attributed to the same reason given before.
to the same reason given before. In order to determine which biokinetic parameter
Values of the biokinetic coefficients presented in has the greatest influence on the effluent
Table 8 for contact stabilization process, except substrate concentration, a sensitivity analysis was
that for µmax, are within the normal range reported performed. The values of each of the kd, µmax, and
for the conventional activated sludge processes. Ks were individually varied by ± 50%, while the
The values of Y in contact stabilization process other parameters were kept constant. The sludge

200
200
150
effluent COD (%)
effluent COD (%)

150

100 100

50 50
Variation in Effluent
Variation in Effluent

0 0
-50 -50
-100 -100
Ks kd µ
umax
max µmax
-150 -150 Ks kd umax

-200 -200
-50 0 50 -50 0 50
biokinetic coefficients
Variation of Biokinetic Coefficients (%)
Variation of biokinetic
Biokinetic coefficients
Coefficients (%)

Fig. 14. S ensitivity a t M LSS of 2000m g/L Fig. 16. S ensitivity a t M LSS of 4000m g/L

200
200
150
effluent COD (%)

150
effluent COD (%)

100 100

50
Variation in Effluent

50
Variation in Effluent

0 0

-50 -50

-100 -100
Ks kd µmax
umax
-150 -150 Ks kd µ
umax
max

-200 -200
-50 0 50 -50 0 50
Variation of biokinetic
Biokinetic coefficients
Coefficients (%) biokinetic coefficients
Variation of Biokinetic Coefficients (%)

Fig. 15. Sensitivity at MLSS of 3000mg/L Fig. 17. Sensitivity at MLSS of 5000mg/L

33
 Sh. Mardani et al., dEtErMINatIoN of bIokINEtIc coEffIcIENtS...

200 1400
Effluent COD (%)

Effluent COD (%)

150 1200
100 1000
Ks kd µ
umax
max
50
Variation in effluent

Variation in effluent
800
0 600
-50 400
-100 200
-150 Ks kd µumax
max 0
-200 -200
-50 -30 0 30 50 -50 -30 0 30 50

Variation of biokinetic coefficients(%)

Biokinetic Coefficients Variation of biokinetic coefficients(%)
Biokinetic Coefficients

Fig. 18. S ensitivity a t M LSS of 2000m g/L Fig. 19. Sensitivity at MLSS of 3000mg/L

retention time in conventional activated sludge,

extended aeration and contact stabilization
systems were kept at 10, 25 and 15 days,
respectively, during the sensitivity analysis. The
sensitivity of the various biokinetic coefficients
was studied by simulating the effluent COD.
The result of the sensitivity analysis are shown
in Figs. (14-19): Figs. (14,15) for conventional
process, Figs. (16,17) for extended aeration
process and Figs. (18,19) for contact stabilization
process. It can be clearly seen that kd and Ks are
directly proportional to the effluent substrate
concentration, while µmax is inversely proportional
to the effluent substrate concentration. Regardless
of the MLSS concentration, the effluent substrate
concentration was found to be more sensitive to
µmax when compared to kd and Ks.

ACKNOWLEDGEMENTS
The authors are grateful for the financial support
of the Research Council of Isfahan Water and
Wastewater Corporation, research project # 4824-
146/2, performed at the Isfahan South Wastewater
Treatment Plant.

34
 w a t e r r e s e a r c h x x x ( 2 0 1 3 ) 1 e8 7

VSS degraded. However, this fcv conversion factor was not

constant for all experiments but represents directly the COD
of the degraded organic material.
The following conclusions can be drawn from the model-
ling results:

3.2.1. Degradation of XSTOR

The degradation rate of stored compounds qSTOR was
observed to fall between 0.8 and 6.6 d1 with a mean of
2.4 d1. This value matches well the magnitude of the hy-
drolysis rate parameter of activated sludge models. The
fraction fSTOR was in average 2.1% for the examined activated
sludge. One sludge sample exhibited a higher portion of 6.9%.
No correlation of qSTOR to bOHO was found. It is suggested that
the magnitude of XSTOR is not a result of the F/M-ratio of the Fig. 4 e Correlation of bOHO and spOUROHO(O).
particular sludge, but rather that of the treatment process in
the WRRF.
be derived by inserting Eq. (2) into Eq. (10) and converting
3.2.2. Initial heterotrophic OUROHO(0) hours to days as well as grams to milligrams.
Modelling the OUROHO of the first days of the batch test as
suggested makes it possible to identify the true initial OUROHO ð0Þ 24 1

$ ¼ bOHO d (18)
endogenous heterotrophic respiration rate (OUROHO(0)) of the XORG ð0Þ fDEG $1000
particular activated sludge, by excluding OURSTOR.
Then in this batch of experiments the degradable fraction of
XORG is:
3.2.3. Heterotrophic decay rate bOHO
According to Eq. (10), the heterotrophic decay rate bOHO cor- 24$100
fDEG ¼ ¼ 45; 3% (19)
responds to the ratio of the heterotrophic endogenous respi- 0; 053$1000
ration rate and the degradable fraction of the activated sludge. From the data of this investigation it can be concluded that
From the presented data this ratio is not constant between XOHO is rather constant and the decay rate varies with loading
samples and therefore bOHO is not constant. This observation conditions. This is in contrast to the recent theory that the
is different from the theory of Marais and Ekama (1976), van decay rate bOHO is constant and the active biomass in terms of
Haandel et al. (1998) and Ramdani et al. (2010). There are two XOHO varies with loading conditions of the activated sludge.
ways to explain this observation: Adding the portion of degraded stored material of in
average 2.1% a total degradable fraction of about 47% of XORG
(A) Regarding XOHO from the substrate point of view
can be calculated from the presented data. This degradable
fraction lies within the range as reported for aerobic degra-
With an decreasing ratio of OUR to active biomass (XOHO),
dation of activated sludge in Metcalf and Eddy (2004). It is
which occurs in low F/M-ratio systems, degradability charac-
assumed, that degradation rates exceeding 47% are due to
teristics of heterotrophic cell mass in activated sludge re-
further adaptation of XOHO to starvation and the activity of
duces, so that the rate of degradation (endogenous decay rate)
predators. This additional degradation takes place in the
decreases.
phase 3 of the aerobic digestion batch experiment.

(B) Regarding XOHO from the organisms point of view

Activated sludge is a highly complex ecosystem and ac- 4. Conclusion

commodates living organisms that have many strategies to
deal with changing environmental conditions. Therefore it is The objective of this study was to find out whether the anal-
rather likely that organisms from low F/M-systems are more ysis of endogenous respiration profiles lead to a constant
prepared to deal with starvation and thus die slower with a decay rate on the basis of the degradation of a homogenous
reduced endogenous decay rate. degradable activated sludge fraction. From the observed data
and the modelled results the following can be concluded:
3.2.4. Correlation of bOHO to spOUROHO(0)
From the empirical point of view the measured endogenous 1. A structural analysis of endogenous OUR profiles revealed
decay rate parameters are shown in Fig. 4 in relation to the that the endogenous decay during the first two days is
specific endogenous respiration rate of XOHO at the beginning overlain by a faster reaction that degrades most likely
of the experiment with respect to the concentration XORG. stored substrate.
Note, that the results from experiment 08A was considered an 2. In most of the experiments after 5 days of digestion the
outlier. exponential decrease of OUR slows down, does not show an
From the slope (0.053) of the line that fits the data best an exponential behaviour and can even increase temporarily.
estimate of the degradable fraction ((1  fU)*XOHO) of XORG can This behaviour is thought to be the consequence of

Please cite this article in press as: Friedrich, M., Takács, I., A new interpretation of endogenous respiration profiles for the
evaluation of the endogenous decay rate of heterotrophic biomass in activated sludge, Water Research (2013), http://dx.doi.org/
10.1016/j.watres.2013.06.043
8 w a t e r r e s e a r c h x x x ( 2 0 1 3 ) 1 e8

proliferation of higher organisms and adaptation of active Trockenrückstandes. Landesverband Bayern Friedenstraße 40,
biomass to the conditions of severe starvation. 81671 München.
3. The degradable material of the activated sludge can be Gujer, W., Henze, M., Mino, T., van Loosdrecht, 1999. Activated
sludge model No. 3. Water Sci. Technol. 39 (1), 183e193.
clearly separated into two fractions; that is easily degrad-
Haandel van, A.C., Catunda Paula, F.C., Araújo, Luiz de Souza,
able storage compounds XSTOR and slower degradable 1998. Biological sludge stabilisation Part 1: kinetics of aerobic
active heterotrophic biomass XOHO. sludge digestion. Water SA 24 (3).
4. In general the values of bOHO in this study were smaller than Hao, X., Wang, Q., Zhang, X., Cao, Y., van Loosdrecht, M.C.M.,
reported elsewhere in the literature. 2009. Experimental evaluation of decrease in bacterial activity
5. The decay rate bOHO of XOHO was not as constant between due to cell death and activity decay in activated sludge. Water
samples as regarded in the current theory of activated Res. 43, 3604e3612.
Henze, M., Grady, C.P.L., Gujer, W., Marais, G.V.R., Matsuo, T.,
sludge modelling. In this study it was found that bOHO de-
1987. Activated Sludge Model No. 1 in Scientific and Technical
creases in a strong correlation with a decreasing activity of Report No 9. IWA Publishing.
the activated sludge expressed as the ratio of the OUROHO(0) Henze, M., Gujer, W., Mino, T., Matsuo, T., Wentzel, M.C.,
to XORG(0). An important consequence of this correlation is Marais, G.V.R., 1995. Activated Sludge Model No. 2 in Scientific
that the fraction of XOHO in the examined activated sludge and Technical Report No 9. IWA Publishing.
remains rather constant over a wide range of loading Henze, M., Gujer, W., Mino, T., Matsuo, T., van
Loosdrecht, M.C.M., 2000. Activated Sludge Models ASM1,
conditions.
ASM2, ASM2d and ASM3, Scientific and Technical Report No 9.
IWA Publishing.
Lavallée, B., Lessard, P., Besser, C., 2002. Decay rate variability of
active heterotrophic biomass. Water Sci. Technol. 46 (1e2),
Acknowledgements 423e430.
Lee, B.J., Wentzel, M.C., Ekama, G.A., 2003. Batch Test for
The authors thank Britta Schmolinski for the patient assis- Measurement of Ordinary Heterotroph Organism Active Mass
tance in the laboratory work as well as the Water Companies in Activated Sludge Liquor. UCT Report W118.
(WAG Schwerin, Zweckverband Grevesmühlen, Eurawasser Liebeskind, M., Schäpers, D., Bornemann, C., Brands, E.,
Nord and Zweckverband Wasser/Abwasser Mecklenburgische Freund, M., Rolfs, T., 1996. Parameter determination and
model-fitting - two approaches for modelling processes in
Schweiz) for the supply of data and activated sludge samples.
wastewater treatment plants. Water Sci. Technol. 34 (5e6),
27e33.
Loosdrecht, van M.C.M., Henze, M., 1999. Maintenance,
references endogenous respiration, lysis, decay and predation. Water Sci.
Technol. 39 (1), 107e117.
Marais, G.V.R., Ekama, G.A., 1976. The activated sludge process
Adams Jr., C.E., Eckenfelder Jr., W.W., Stein, R.M., 1974. Part I e Steady state behaviour. Water SA 2 (4), 163e200.
Modifications to aerobic digestor design. Water Res. 8, Metcalf, Eddy, 2004. Wastewater Engineering Treatment and
213e218. Reuse. McGraw-Hill, New York, p. 1537.
Avcioglu, E., Orhon, D., Sözen, S., 1998. A new method for the Moussa, M.S., Hooijmans, C.M., Lubberding, H.J., Gijzen, H.J., van
assessment of heterotrophic endogenous respiration rate Loosdrecht, M.C.M., 2005. Modelling nitrification,
under aerobic and anoxic conditions. Water Sci. Technol. 38 heterotrophic growth and predation in activated sludge.
(8e9), 95e103. Water Res. 39, 5080e5098.
Benedek, P., Farkas, P., Literáthy, P., 1972. Kinetics of aerobic Ramdani, A., Dold, P., Deleris, S., Lamarre, D., Gadbois, A.,
sludge stabilization. Water Res. 6, 91e97. Comeau, Y., 2010. Biodegradation of the endogenous reside of
Brands, E., Liebeskind, M., Dohmann, M., 1994. Parameters for activated sludge. Water Res. 44 (7), 2179e2188.
dynamic simulation of wastewater treatment plants with Siegrist, H., Brunner, I., Koch, G., Con Phan, Linh, Le, Van Chieu,
high-rate and low-rate activated-sludge tanks. Water Sci. 1999. Reduction of biomass decay rate under anoxic
Technol. 30 (4), 211e214. conditions. Water Sci. Technol. 39 (1), 129e137.
Dold, P.L., Ekama, G.A., Marais, G.v.R., 1980. A general model for Spanjers, H., Vanrolleghem, P., 1995. Respirometry as a tool for
the activated sludge process. Prog. Water Technol. 12, 47e77. rapid characterization of wastewater and activated sludge.
DWA A 131, 2000. Bemessung von einstufigen Belebungsanlagen. Water Sci. Technol. 31 (2), 105e114.
GFA Hennef. Spanjers, H., Vanrolleghem, P., Olsson, G., Dold, P., 1996.
DWA, 2003. Leitfaden Nr. 2-8 Schlammtrocknung mit der Respirometry in control of the activated sludge process. Water
Mikrowelle zur Bestimmung der Trockensubstanz oder des Sci. Technol. 34 (3e4), 117e126.

Please cite this article in press as: Friedrich, M., Takács, I., A new interpretation of endogenous respiration profiles for the
evaluation of the endogenous decay rate of heterotrophic biomass in activated sludge, Water Research (2013), http://dx.doi.org/
10.1016/j.watres.2013.06.043