BIOTECHNOLOGY AND BIOENGINEERING, VOL. XVIII, PAGES 145-165 (1976)
The Application of Constant Recycle Solids Concentration in Activated Sludge Process F. M. BONOTAN-DURA and P. Y. YANG, Environmental Engineering Division, Asian Institute of Technology, Bangkok, Thailand
Summary The applicability of the model derived by Ramanathan and Gaudy (Biotechnol. Bioeng., 11,207, (1969)) for completely mixed activated sludge treatment holding the recycle solids concentration as a system constant was investigated using an actual industrial organic wastewater. Short-term experiments were conducted at various dilution rates (1/8, 1/6, 1/4, 1/2, 1/1.5 hr-1) for two recycle solids concentration values (5000 and 7000 mg/liter). The influent substrate concentration was maintained at 1000 mg/liter COD and the hydraulic recycle ratio, a, was kept at 0.3. It was found that for bottling plant (Pepsi Cola) wastewaters, a steady state with respect to reactor biological solids and effluent COD, at different dilution rates, could be attained, lending experimental evidence to the assumption that a steady state could be reached in developing the model and also affecting the applicability of the model in industrial organic wastewater. The reactor biological solids and effluent COD calculated from the model closely agreed with the observed values a t dilution rates lower than 0.5 hr-]. Operation at dilution rates higher than 0.5 hr-* will washout the biological solids from the reactor and the recycle substrate concentration will be apparent if the concentration of X R were not increased.
INTRODUCTION Ramanathan and Gaudy1 found that Herbert’s classical theory of continuous culture, which defined the recycle solids concentration factor, c, as a system constant, was inadequate for the heterogeneous populations of the activated sludge process. The biological solids in the reactor considerably deviated from steady state, upon which the mathematical formulation was based. To offset this drawback, they devised another model2 in 1971, using the recycle solids concentration X R , rather than c, as a system parameter. Gaudy and Srini~asaraghavan~ conducted experimental studies on this model using glucose as the carbon source. They varied the organic feed concentration (500, 1000, and 2000 mg/liter COD) and maintained the dilution rate D at 0.125 hr-l, the hydraulic recycle 145 @ 1976 by John Wiley & Sons, Inc.
BONOTAN-DURA AND YANG
146
ratio a was maintained at 0.25, and X R was kept at 10,OOO mg/liter. Their experimental results showed that for glucose the proposed model provided steady performance in terms of reactor biological solids level and effluent COD and delivered a high degree of treatment efficiency (95% and above). Will this “steadiness” in reactor biological solids and effluent COD also prevail when an actual organic industrial wastewater is employed as the carbon source In other words, will the model be applicable to the treatment of industrial organic wastewaters, which are of complex composition compared with glucose? This is the subject of this research.
EXPERIMENTAL INVESTIGATION Materials and Methods
Treatment unit The laboratory completely mixed activated sludge (CMAS) system employed in this research consisted of a feed reservoir, an
Fig. 1. Laboratory scale activated plant for operation with constant
XR.
CONSTANT RECYCLE SOLIDS CONCENTRATION
147
aerator, a clarifier, and a homogenizing chamber. These components were fabricated from a PVC sheet, 6 mm thick. A photograph of this unit is shown in Figure 1. The flow diagram is shown in Figure 2. The aeration tank has a working capacity of 5.0 liters. Proper design measures were observed to prevent “dead” corners. Air was introduced into this tank through six stone diffusers located near its base. The air which was provided in excess, served not only as a source of oxygen but also as a means to induce complete mixing of the reactor contents. A chemical feed pump (model C-630 P, Engineered Products Mfg. Co.) continuously discharged the wastewater from the feed reservoir to the aeration tank. The reservoir, with a maximum capacity of 40 liters, was provided with a cooling unit. The cooling system was operated intermittently, especially at high detention times when the feed waste had to remain in the reservoir for more than 24 hr. I n order t o maintain continuous operation, it was necessary to employ duplicate pumps and feed lines. The feed lines were sterilized periodically by pumping 1% chlorox solution followed by the passage of tap water for several minutes. The reactor effluent flowed by gravity to the settling tank. This tank has a volume of 4.3 liters. The settled sludge was channeled at 4-8 hr intervals to the homogenizing chamber and diluted to the desired concentration with the clarifier effluent. It was observed that when the sludge was not withdrawn after 6 hr of settling, it began to rise in the clarifier. This could be traced to the development of anaerobic conditions which resulted in nitrogen
-
INFLUENT AERATION
EFFLUENT
i
XLUENT FOR SLWGE CCNSISTENCY
WASTE S L U E
Fig. 2. Flow diagram of a completely mixed activated sludge system using constant concentration of recycle solids, X E .
BONOTAN-DURA AND YANG
148
gas formation. Although this was the case, there were occasions when the sludge could not be withdrawn in less than 6 hr. There were times when the sludge in the clarifier did not settle well, resulting in a loose sludge. In order to meet the selected cell recycle concentration, the sludge was withdrawn and allowed to further settle in Imhoff cones. The supernatant was mixed with the clarifier effluent and the settled sludge was placed in the homogenizing chamber. The homogenizing chamber was fitted with a mechanical mixer and air diffusers for return sludge aeration. For the first set of experiments a t X R 5000 mg/liter, the mechanical mixer was used in daylight. The mechanical mixer was run by a motor. In the absence of an alternate motor, the air diffusers served in the night time. Due t o a failing motor when the experiments for the second set were carried out, mechanical mixing was only occasionally employed. Mechanical mixing insured uniform concentration of the return sludge in the homogenizing chamber. The recycle sludge from the homogenizing chamber was pumped continuously to the aeration tank by a variable speed peristaltic action pump (Type MARE 7, Watson-Narlow Ltd.) through a silicon rubber tubing. Prior to experimentation, the return sludge pump was calibrated using the Pepsi Plant mixed liquor.
Feed waste The Pepsi Cola combined wastewater was used in the study as the carbon source. Aziz4 detailed the process of Pepsi Cola production and the source of the combined waste. His analysis of the waste and that of Chanika5 are presented in Table I . TABLE I Pepsi Cola Waste Characteristics ~~
~
Analysis PH Settleable solids Suspended solids COD BOD DO NH3-N Org-N P0,Total phosphorus
Aziz'
Chanika5
7.4-9.6 0.2-0.5 mg/li ter 25-60 635-1685 663-1 145 5.1-6.2 1.1-1.7 1.6-2.1 6.4-10.6
7.6-8.9 40-60 825-1450
1.0-2.0 1.5-1.8 0.1-0.25
CONSTANT RECYCLE SOLIDS CONCENTRATION
149
An examination of Table I reveals that the nitrogen and phosphorus contents of the waste are very low. Deficiency in these nutrients can affect biological activity of the microbes. Since these nutrients are nonlimiting in this study, ammonium sulfate according to the COD to N ratio of 20: 1 and potassium dihydrogen phosphate were added. Potassium dihydrogen phosphate served not only as a source of phosphorus, but also to adjust the pH of the waste to neutral level. The low suspended solids content of the waste was an advantage. It indicates that the organic content is mostly in soluble form and conforms to the assumption of negligible suspended solids in the influent made in the development of the model. The feed solution was prepared daily or as often as needed. The COD of the feed was maintained at 1000 mg/liter. When the COD of the waste was above this level, it was diluted with tap water. Occasionally the COD of the waste was below the desired concentration. When this occurred, it was introduced into the treatment unit in its present state.
Mass culture The original mass culture was obtained from the Pepsi Cola Waste Treatment Plant. Two liters of the Pepsi mixed liquor were added to 3 liters of the nutrient-supplemented feed waste in an 8 liter vessel and were well aerated. The pH was adjusted to a neutral level with the phosphate buffer. The culture was refreshed twice a day.unti1 the concentration reached about 5000 mg/liter. One liter of the culture was then transferred to the aeration tank to serve as inoculum for the continuous flow experiment ; the remaining portion was placed in the homogenizing chamber to supply the recycle requirement. Analytical Procedures The biological solids concentration in the reactor, clarifier effluent, and homogenizing chamber was determined by the membrane filter technique. Millipore membrane filters with a pore size of 0.8 p were used. The chemical oxygen demand of the waste, effluent filtrate, and mixed liquor (Set A experiments) was measured in accordance with Standard Methods.6 A spot COD test wai also made on the recycle filtrate. The sludge volume index of the mixed liquor was determined according to Standard Methods.6
150
BONOTAN-DURA AND YANG
Operating Procedures To start a continuous culture experiment, a batch culture was initiated by inoculating the aeration tank with 20y0 by volume mass culture. The pH of the culture was controlled a t 7.0 f 0.3 by the phosphate buffer system. After approximately 20 hr of batch cultivation, continuous flow studies a t a selected dilution rate were started. The mixed liquor temperature was measured. The treatment unit was operated as a once-through system for several hours prior t o operation with cell recycle. The system was allowed to reach the steady state. The system was considered t o have approached this condition when the substrate concentration in the effluent had practically become constant. Samples of the mixed liquor, clarifier effluent, and return sludge were collected at average intervals of 5 hr, until a sufficient amount of data had been collected. The influent flow rate was checked frequently to ensure that the desired rate was maintained.
Parameters and Variables I n Table 11, two sets of continuous flow experiments a t dilution rates of 1/8, 1/6, 1/4, 1/2, and 1/1.5 each were performed. For set A, the recycle cell concentration was maintained at about 5000 mg/liter and for set B, the recycle cell concentration was controlled a t approximately 7000 mg/liter level. I n all cases, the recycle volumetric flow rate was 30% of the wastewater inflow (a = 0.3) and the feed concentration was maintained a t 1000 mg/liter COD. The pH of the reactor contents was held at 7.0 f 0.3. A mixed liquor temperature of 27 f 3°C was observed during the duration of the study.
-
Substrate concentration (Si)
TABLE I1 Summary of Parameters and Variables Dilution rat.e
(D) 1/8 1/6 1/4
COD = lOOOmg/ liter
112 111.5
Recycle cell concentration (XR)
Hydraulic recycle ratio
5000 mg/liter 0
7000 mg-liter
=
0.3
CONSTANT RECYCLE SOLIDS CONCENTRATION
151
The experimental investigation was initiated at the dilution rate of 1/2 hr-I for set A. The dilution rate was then adjusted to 1/4, 1/6, and 1/8 hr-l. To have adequate data points, studies at the dilution rate of 1/1.5 hr-' were also conducted. For set B, studies were conducted according to the increasing magnitude of the dilution rate, that is, 1/8, 1/6, 1/4, 1/2, and 1/1.5 hr-l.
RESULTS AND DISCUSSION Performance Characteristics of a CM AS System Operated at Constant Recycle Solids Concentration For CMAS fed with bottling plant (Pepsi Cola) waste of 1000 mg/liter COD, the behavior of the operational parameters, namely, the reactor biological solids concentration, effluent COD, and effluent suspended solids at various dilution rates when the recycle solids was set at about 5000 mg/liter are shown in Figures 3 to 7. It is seen that at each dilution rate, the reactor biological solids and effluent soluble COD did approach a steady-state condition and the recycle solids concentration did remain almost constant. The 2000
I
I
I800 -
- 8IOCOGICAL
I
I
SOLIDS
(7)
CELL RECYCLE CONC
10000 -9ooo
( X,
W
600
-
400
-
-
J
-3000 EFFLUENT
COD ( 3 )
V W
LT
EFFLUENT SOLIDS ( X, )
-2000
-A
J W V
0
20
40
60
80
100
TIME, HOURS
Fig. 3. Steady-state parameters a t D
= 1/2 hr-1, X z ==. 5000 mg/liter (feed COD = 1000 mg/liter).
BONOTAN-DURA AND YANG
152
J
0
-3000
5w
a
z
4
EFFLUENT SOLIDS
200
0
-2000
( Xc )
(5)
EFFLUENT COD
20
40
60
1000
J J W
0
80
TIME, HOURS
Fig. 4. Steady-state parameters at D = 1/4 hr-1, X R = 5000 mg/liter (feed COD = 1000 mg/liter).
20001
1800(
I
BIOLOGICAL SOLIDS
(
I
1)
1 - 9000
> z ga -I
\ 0
-8000
1600-
1400
E
-
-7ooo CELL RECYCLE CONC. (X,)
v)
1200IW
5
m 1
-d
(L a
-_
5000
800-
Fig. 5.
Steady-state parameters at D
-6000
5
: 8 Y
=
1/6 hr-1, X R = 5000 mg/liter.
CONSTANT RECYCLE SOLIDS CONCENTRATION
153
\ 3
4
F
\
z
ga
W
E
t-
v)
z
a
6000
W V
I
So00
0
W IW
z
a
a
W
2
4000
0 W
3000
0
-J
V
>
t-
a 0
0
400
z
-7
0
(( X Xcc ))
m
EFFLUENT
(s)
- 11000000
60 40 60 TIME, HOURS
20
J J
EFFLUENT EFFLUENT SOLIDS SOLIDS COD
uwa W V
0 inn 1 00
80
Fig. 6. Steady-state parameters at D = 1/8 hr-1, X R = 5000 mg/liter (feed COD = 1000 mg/liter). 1600
I
I
I
-8000
1
-J \
W
1400
-
FFLUENT COD
(5)
EFFLUENT SOLIDS [ X,)
-2000
y
a
TIME, HOURS
Fig. 7. Steady-state parameters at D = 1/1.5 hr-1, X R
2:
5000 mg/liter.
154
BONOTAN-DURA AND YANG
effluent solids level was generally low signifying good settleability in the clarifier. The mean steady-state values of the operational parameters mentioned at each dilution rate studied are given in Table 111. Also shown in this table are the loading factors, specific substrate removal rates, COD removal efficiencies, and cell output rates. The mean steady-state reactor biological solids concentration and effluent soluble COD corresponding to each dilution rate shown in Table I11 are plotted in Figure 8 (solid lines). It is seen that the CMAS system remained stable up to a dilution rate of 0.35 hr-l. Beyond this dilution rate, the effluent soluble COD started to rise and the biological solids began to be diluted out from the reactor. The variations in reactor biological solids, effluent soluble COD, effluent suspended solids, and COD of inflowing Pepsi Cola wastewater at dilution rates considered for the CMAS operated at “constant” recycle solids concentration of 7000 mg/liter are depicted in Figures 9 to 13. These figures also show that a steady-state condition with regard to reactor biological solids and effluent soluble COD has been attained.
0
g e
PREDICATED BIOLOGICAL SOLIDS CONC .T
I
, 2 0 0 ~ OBSERVED BIOLOGICAL SOLIDS CONCENTRATION
I
0.2
I
0.3 0.4 DILUTION RATE
,
,
1
0.5
0.6
1 0.1
HOUR-’
and s’ and the predicted values from the Gaudy model with X R (5000 mg/liter) as system constant.
Fig. 8. Observed levels of
a
0.125 0.167 0.25 0.5 0.67
D X 24.
1000 982 1000 1000 882
Sp. substrate utilization rate = D(Si
X
4X
AS.
6.2 4.6 3.1 1.5 1.2
Loading factor =
1/8 1/6 1/4 1/2 1/1.5
Dilution rate (hr-1) 1623 1632 1877 1.587 1074
- s)/x.
60 50 73 192 539
x
4826 4871 5267 4870 4601
53 143 55 154 193
9.2 12.0 16.0 37.8 66.0
Loading Reactor detention factor" (mg Si 9 XR X. COD/mg time (hr) (mg/liter) (mg/liter) (mg/liter) (mg/liter) (mg/liter) SS/day) 0.074 0.144 0.161 0.331 0.27
Sp. substrate utilizationb rate (mg COD/mg SS/hr)
94.0 95.0 92.7 80.8 39.0
4 24 16 77 129
COD Cell removal output efficiency mg/liter/hr (%I (DXe)
TABLE 111 Mean Operational Parameters a t Different Dilution Rates with Pepsi Cola Wastewater as Substrate (a = 0.3, X R = 5000 Nominally)
BONOTAN-DURA AND YANG
156 2200-
-10000
-I
\
-8ooo
-x"F
-7000
z 0
- 9 m
J \
-
-6000 -5000
ta 5 W V 0' V
-4000 EFFLUENT SOLIDS
( X,)
-3000
-2000
9 8 a j
w
- 1000
V
looo
TIME, HOURS
Fig. 9. Steady-state parameters at D = 1/8 hr-1, X R = 7000 mg/liter (feed COD controlled at 1000 mg/liter).
2
W V 2
0 FEED
COD
V
(Si)
Y
it
-'-
,.4-.,.
LA..
_I
100 A I
10
A
-J w -J
u
-.".-.d.''
EFFLUENT COD ( A
8 a
/e.,,
a , . .
W'
3)
A
w
V >.
EFFLUENT SOLIDS ( X c )
A
A
A
A
I
I
I
I
20
30
40
50
TIME, HOURS
4
6b0
Fig. 10. Steady-state parameters at D = 1/6 hr-1, Xg = 7000 mg/liter.
CONSTANT RECYCLE SOLIDS CONCENTRATION
157
2200t BIOLOGICAL SOLIDS (
)
2000
K
' -I
z
v)
II:
w iw 1200 -
I-
-600
z
a CL
1000
r
0
800
-
2 w
I-
a 0
0
z
FEED COD
0
0
P
P
J -I
-400
--
0
g W
-I
200 100
5z
./
EFFLUENT
SOLIDS ( Xc )
EFFLUENT
COD
I
I
u > 0 -2O0
( $ )
\ 1
-.d I
.' 1
E
-.
-
2
Fig. 11. Steady-state parameters at D = 1/4 hr-1, X R = 7000 mg/liter.
Fig. 12. Steady-state parameters at D = 1/2 hr-1, X R = 7000 mg/liter (feed COD controlled at 1000 mg-liter).
BONOTAN-DURA AND YANG
158
-czs H
1 mFEED
COO
( Si)
Kxx)
0 800-
-4Ooo -I J
W V
600-
W V W P
_t EFFLUENT SOLIDS ( X e )
200,_ , _ _
I
5
.o-’I .-.o-.-.-.o--.---.e.-.’.-o._.-. -0 I 10
I5 TIME, HOURS
20
25
3oo
Fig. 13. Steady-state parameters at D = 1/1.5 hr-l, X R = 7000 mg/liter.
The average steady-state values of the operational parameters, mentioned above, at each dilution rate studied are presented in Table IV. In this table, the loading factors, specific substrate removal rates, COD removal efficiencies, cell output rates, and substrate concentration in the recycle sludge are also given. The mean steady-state values of the reactor biological solids and effluent soluble COD are plotted in Figure 14. This figure shows that the effluent soluble COD began to rise at the dilution rate of 0.33 hr-*. The reactor biological solids remained highly stable until the dilution rate of 0.5 hr-* was reached. These results show that for an input of about 1000 mg/liter COD of bottling plant wastewater, a CMAS process with constant recycle solids concentration of 5000 or 7000 mg/liter could be operated up to a dilution rate of 0.5 hr-1 without appreciable dilute-out of reactor biological solids and considerable substrate leakage in the effluent. However, it is noticed that the substrate concentration in the recycle sludge, S,, are lower than 8 in Table IV. Such a reduction can be related to the mean hydraulic retention time in the sludge settling and reaeration tanks. The rather high increase of S , after the dilution rate of 1/2 hr-1 can be related to the lower hydrauxc retention
1/8 1/6 1/4 1-2 1/1.5
0.125 0.167 0.25 0.50 0.67
Dilution rate (hr-1)
6.2 4.6 3.1 1.5 1.2
1000 969 982 lo00 lo00
Reactor detention Si time (mg/ (hr) liter)
34 32 63 158 266
10 8 36 132 269
1844 1892 1907 1898 1746
Substrate concentraB tion in 8 (mg/ recyclesludge (mg/ liter) (&, mg/liter) liter)
7098 7224 7356 7350 7639
120 150 69 116 68
8.1 10.3 15.4 31.6 46.0
0.084 0.11 0.16 0.30 0.24
96.6 96.7 93.6 84.2 63.4
15 25 17 58 45
Sp. substrate Loading utilization COD Cell factor (mg rate (mg removal output XR X, COD/mg COD/mg efficiency mg/liter/hr (mg/liter) (mg/liter) SS/day) SS/hr) (%) (DXJ
TABLE IV Mean Operational Parameters a t Different Dilution Rates with Pepsi Cola Wastewater as Substrate ( a = 0.3, X R = 7000 Nominally)
Ei
3
Q 0
m
U
Y
g ik
m
d
*
3 EQ
b-
BONOTAN-DURA AND YANG
160 2ooQ
A
1800-
\
1600-
LEGEND v)
a W
1400-
+ W
..
- OBSERVED BIOLOGICAL SOL1DS CONCENTRATION - PREDICTED
2 1200U
DILUTION RATE
SOLIDS
,
BDLOGICAL CONCENTRATlON
HOUR-'
s
Fig. 14. Observed levels (solid line curve) of t, and and the predicted values (broken line curve) from the Gaudy model with X e (7000 mg/liter) as a system constant.
time in the reaeration tank because a higher recycle flow rate was operated to maintain a = 0.3. Conventionally, the volume of the sludge or solids after 30 min settling of the mixed liquor in a 1 liter cylinder is combined with the reactor suspended solids to form a sludge volume index (SVI) which can be used as an indication of the settling characteristics of the sludge. These data have been collected and are shown in Table V. TABLE V Sludge Volume Index at Different Reactor Detention Times ~
SVI
Reactor detention time (hr)
X R = 5000 mg/li ter
X R = 7000 mg/liter
1.2 1.5 3.1 4.6 6.2
41.4 148.0 227.0 354.0
71.9 76.1 70.5 213.0 207.0
~
~~
CONSTANT RECYCLE SOLIDS CONCENTRATION
161
It can be seen that the application of higher constant recycle concentration of sludge do decrease the values of SVI with higher reactor retention time. The SVI values can be decreased from 35 to 150, which are reported by Metcalf and Eddy Inc.’ as the typical SVI values for good settling sludges when the constant sludge concentration are increased to the values higher than 7000 mg/liter. Therefore, the model incorporated with constant recycle sludge concentration does not have an inverse effect on the process. Predictability of the C M A S Model for Operation at Constant Recycle Solids Concentration The values of the kinetic constants, namely, maximum specific growth rate (pm), saturation constant ( K 8 ) ,and sludge yield (U), need to be known in order to be able to use the CMAS model of Ramanathan and Gaudy.* Since cell maintenance energy could have a significant role in CMAS systems with solids recycle, the true sludge yield ( Y , ) and the cell maintenance coefficient ( K d ) were evaluated using the straight-line approach of Marr et a1.* as shown in Figure 15. The maximum specific growth rate and saturation constant were calculated and given in Table VI. Employing the kinetic constants in Table VI, computer programs were designed to obtain the reactor biological solids concentration
-.-
IA
0 U
Y
U
A
0-A,X,
= 5000 m p / l
a--B,x,
= 7000 mg/l
_I
a
0
I -
0
Y+
0
=
I/intercept
k d = Yt x slope
LT
a
0 W
Lz
0
~~
RECIPROCAL
OF SPECIFIC GROWTH RATE
~~
( I /U
Fig. 15. Determination of true yield ( Y J and maintenance coefficient according to the method of Marr et a1.8
(Kd)
162
BONOTAN-DURA AND YANG
TABLE VI Kinetic Parameters Determined in the Present Study for Bottling Plant (Pepsi Cola) Wastewater Values Kinetic parameters Maximum specific growth rate (s,), hr-' Saturation constant ( K 8 ) ,mg/liter True yield (Yt) Cell maintenance coefficient (Kd),day-]
XR
X R = 5000 mg/liter
=
7000 mg/liter
0.45
0.15
221 0.76
200 0.33 0.072
0.14
and effluent soluble COD predicted by the proposed model for two cases : 1) when the cell maintenance coefficient K d is neglected (refer to Table VII for the equations) ; 2) when the cell maintenance coefficient Kd is considered (refer
also to Table VII for the equations). TABLE VII Steady-State Equations for CMAS Model Employing Constant X R Ramanathan and Gaudy' without K d
s =- b
f d b l - 4 aC
2a
(2)
u=pm-(l+a)D b = D [S;
(1
-
(1
+ a) k,]
Srinivasaraghavan and Gaudy9 with Kd
s
=
-b f d b a
- (1 + a ) D -k
u =
p,
b
D [S; - (1
=
+ a)
- 4 aC
(5)
2a
(1
kd
+ a) k.]
+4
C = k. D Si (3)
C
=
k. D S;
+ (1 + 4 kd
~
*
k, . S;
CONSTANT RECYCLE SOLIDS CONCENTRATION
163
The computer results are given in Table VIII with the observed values. At the conditions (pH, temperature, influent COD, hydraulic recycle ratio, and recycle solids concentration) maintained in this study, the cell maintenance coefficient, apparently, did not have any significant effect on the predicted values of reactor biological solids and effluent COD. TABLE VIII Effect of Maintenance Coefficient on the Predicted Levels of 8 and Model”
s by Gaudy
Predicted Values by Gaudy Model Dilution rate Set (hr-l)
A
B
a
0.125 0.167 0.25 0.50 0.67 0.125 0.167 0.25 0.50 0.67
Observed values
Without K
d
With K
d
s
s
8
s
B
B
1623 1632 1877 1587 1074 1844 1892 1907 1898 1746
60 50 73 192 539 34 32 63 158 366
1675 1667 1753 1588 1420 1881 1899 1924 1895 1874
30 41 62 158 207 33 43 69 167 238
1618 1624 1722 1674 1411 1847 1873 1906 1886 1867
31 42 63 159 207 33 43 69 167 238
Predicted values were obtained with the aid of a computer.
The predicted levels of reactor biological solids and effluent COD (broken lines) are presented graphically in Figure 8 for “constant” X R of 5000 mg/liter and in Figure 14 are shown for “constant” X R of 7000 mg/liter in order to facilitate the comparison with the observed values (solid lines). The predicted and observed values closely agreed at dilution rates below 0.5 hr-l, that is, before the dilute-out zone. At dilution rates above 0.5 hr-’, the predicted and observed values tend to diverge from one other. However, in Table VIII it is understood that if the concentration of X g had been increased more than 7000 mg/liter, the present system would undoubtedly demonstrate a closer agreement between observed and predicted biological solids and effluent concentrations a t a higher dilution rate. Therefore, it can be concluded that the CMAS model proposed by Ramanathan and’Gaudy2for operation a t constant X R does offer high predictability.
BONOTAN-DURA AND YANG
164
Design Parameters for C M A S Feeding on Bottling Plant (Pepsi Cola) Wastewaters The kinetic parameters in the proposed model, which have been evaluated in the present study, are presented in Table IX. Also presented are the values from studies of a CMAS continuous flow once-through system made by Ch ar~ ik a.~ TABLE IX Design Parameters for CMAS Feeding on Bottling Plant (Pepsi Cola) Wastewaters Present work Kinetic parameters Maximum specific growth rate (PA,hr-’ Saturation constants ( K 8 ) , mg/liter True yield (Yi) Cell maintenance coefficient ( K d ) , day-1
Chanikas 0.62 209 0.57
0
X R = 5000 mg/liter 0.45 22 1
0.76 0.14
X R = 7000 mg/liter 0.15 200 0.33
0.072
The values of p , , K , , Y t , and Kd for X I Zof 5000 mg/liter are found to be higher than those found for the XR of 7000 mg/liter. Specifically, p , for “constant” X R of 5000 mg/liter is three times larger than the p m for constant X R of 7000 mg/liter. The true sludge yield and cell maintenance coefficient, too, are about twice that for XR of 7000 mg/liter. The p , values from Chanika’s studies5, however, is found t o exceed the p, values in the present work. The K , value is about the same for both types of CMAS systems. The sludge yield value, though, is lower than that for the X R of 5000 mg/liter but higher than that for X R of 7000 mg/liter. Although the difference of these values could be possibly due to some change in the predominating microbial species populating these CMAS systems, it is also quite possible that the increase of XR (higher than 7000mg/ liter) may develop a trend which changes these values. With the latter possibility in mind, the present values were compared to the extended aeration process (higher sludge retention time) by using the same wastewater.1° It was found that the values calculated from the extended aeration process were much lower than the values obtained in the present study. Therefore, it seems reasonable that
CONSTANT RECYCLE SOLIDS CONCENTRATION
165
the change in these parameters is related to the change in the concentration of X R . This work was supported by a scholarship donated by the United Kingdom through the Asian Institute of Technology.
Nomenclature recycle solids concentration factor, X R / ~ dilution rate, F / V influent flow rate of wastewater, liter/hr substrate concentration in the reactor, measured as COD, mg/liter substrate concentration in the influent wastewater, measured as COD, mg/liter biological solids concentration in the reactor, mg/liter concentration of biological solids, mg/liter biological solids concentration in the clarifier effluent, mg/liter recycle biological solids concentration, mg/liter observed sludge yield true sludge yield saturation constant, numerically, i t is equal to the substrate concentration at which the specific growth rate is half of the maximum specific growth rate, mg/liter cell maintenance coefficient, day-1 the maximum specific growth rate, hr-1 recycle flow ratio specific growth rate, hr-l recycle substrate concentration, mg/liter
References 1. M. Ramanathan and A. F. Gaudy, Jr., Biolechnol. Bioeng., 11, 207 (1969). 2. M. Ramanathan and A. F. Gaudy, Jr., Biotechnol. Bioeng., 13, 125 (1971). 3. A. F. Gaudy, Jr. and R. Srinivasaraghavan, Biotechnol. Bioeng., 16, 723 (1974). 4. J. A. Aziz, Master Thesis No. 269, AIT, Bangkok, Thailand, 1969. 5. T. Chanika, Master Thesis No. 759, AIT, Bangkok, Thailand, 1974. 6. APHA, AWWA, WPCF, Standard Methods, 13th ed., American Public Health Association, New York, 1971. 7. Metcalf and Eddy, Inc., Wastewater Engineering, McGraw-Hill, New York, 1972. 8. A. G. Marr, E. H. Nilson, and D. J. Clark, Ann. N.Y. Acad. Sci., 102,536 (1963). 9. It. Srinivasaraghavan and A. F. Gaudy, Jr., Ann. Ind. Waste Conf., Purdue University, Lafayette, Indiana, 1974. 10. Y. K. Chen, Master Thesis, AIT, Bangkok, Thailand, 1975.
Accepted for Publication October 14, 1975
The Application of Constant Recycle Solids Concentration in Activated Sludge Process F. M. BONOTAN-DURA and P. Y. YANG, Environmental Engineering Division, Asian Institute of Technology, Bangkok, Thailand
Summary The applicability of the model derived by Ramanathan and Gaudy (Biotechnol. Bioeng., 11,207, (1969)) for completely mixed activated sludge treatment holding the recycle solids concentration as a system constant was investigated using an actual industrial organic wastewater. Short-term experiments were conducted at various dilution rates (1/8, 1/6, 1/4, 1/2, 1/1.5 hr-1) for two recycle solids concentration values (5000 and 7000 mg/liter). The influent substrate concentration was maintained at 1000 mg/liter COD and the hydraulic recycle ratio, a, was kept at 0.3. It was found that for bottling plant (Pepsi Cola) wastewaters, a steady state with respect to reactor biological solids and effluent COD, at different dilution rates, could be attained, lending experimental evidence to the assumption that a steady state could be reached in developing the model and also affecting the applicability of the model in industrial organic wastewater. The reactor biological solids and effluent COD calculated from the model closely agreed with the observed values a t dilution rates lower than 0.5 hr-]. Operation at dilution rates higher than 0.5 hr-* will washout the biological solids from the reactor and the recycle substrate concentration will be apparent if the concentration of X R were not increased.
INTRODUCTION Ramanathan and Gaudy1 found that Herbert’s classical theory of continuous culture, which defined the recycle solids concentration factor, c, as a system constant, was inadequate for the heterogeneous populations of the activated sludge process. The biological solids in the reactor considerably deviated from steady state, upon which the mathematical formulation was based. To offset this drawback, they devised another model2 in 1971, using the recycle solids concentration X R , rather than c, as a system parameter. Gaudy and Srini~asaraghavan~ conducted experimental studies on this model using glucose as the carbon source. They varied the organic feed concentration (500, 1000, and 2000 mg/liter COD) and maintained the dilution rate D at 0.125 hr-l, the hydraulic recycle 145 @ 1976 by John Wiley & Sons, Inc.
BONOTAN-DURA AND YANG
146
ratio a was maintained at 0.25, and X R was kept at 10,OOO mg/liter. Their experimental results showed that for glucose the proposed model provided steady performance in terms of reactor biological solids level and effluent COD and delivered a high degree of treatment efficiency (95% and above). Will this “steadiness” in reactor biological solids and effluent COD also prevail when an actual organic industrial wastewater is employed as the carbon source In other words, will the model be applicable to the treatment of industrial organic wastewaters, which are of complex composition compared with glucose? This is the subject of this research.
EXPERIMENTAL INVESTIGATION Materials and Methods
Treatment unit The laboratory completely mixed activated sludge (CMAS) system employed in this research consisted of a feed reservoir, an
Fig. 1. Laboratory scale activated plant for operation with constant
XR.
CONSTANT RECYCLE SOLIDS CONCENTRATION
147
aerator, a clarifier, and a homogenizing chamber. These components were fabricated from a PVC sheet, 6 mm thick. A photograph of this unit is shown in Figure 1. The flow diagram is shown in Figure 2. The aeration tank has a working capacity of 5.0 liters. Proper design measures were observed to prevent “dead” corners. Air was introduced into this tank through six stone diffusers located near its base. The air which was provided in excess, served not only as a source of oxygen but also as a means to induce complete mixing of the reactor contents. A chemical feed pump (model C-630 P, Engineered Products Mfg. Co.) continuously discharged the wastewater from the feed reservoir to the aeration tank. The reservoir, with a maximum capacity of 40 liters, was provided with a cooling unit. The cooling system was operated intermittently, especially at high detention times when the feed waste had to remain in the reservoir for more than 24 hr. I n order t o maintain continuous operation, it was necessary to employ duplicate pumps and feed lines. The feed lines were sterilized periodically by pumping 1% chlorox solution followed by the passage of tap water for several minutes. The reactor effluent flowed by gravity to the settling tank. This tank has a volume of 4.3 liters. The settled sludge was channeled at 4-8 hr intervals to the homogenizing chamber and diluted to the desired concentration with the clarifier effluent. It was observed that when the sludge was not withdrawn after 6 hr of settling, it began to rise in the clarifier. This could be traced to the development of anaerobic conditions which resulted in nitrogen
-
INFLUENT AERATION
EFFLUENT
i
XLUENT FOR SLWGE CCNSISTENCY
WASTE S L U E
Fig. 2. Flow diagram of a completely mixed activated sludge system using constant concentration of recycle solids, X E .
BONOTAN-DURA AND YANG
148
gas formation. Although this was the case, there were occasions when the sludge could not be withdrawn in less than 6 hr. There were times when the sludge in the clarifier did not settle well, resulting in a loose sludge. In order to meet the selected cell recycle concentration, the sludge was withdrawn and allowed to further settle in Imhoff cones. The supernatant was mixed with the clarifier effluent and the settled sludge was placed in the homogenizing chamber. The homogenizing chamber was fitted with a mechanical mixer and air diffusers for return sludge aeration. For the first set of experiments a t X R 5000 mg/liter, the mechanical mixer was used in daylight. The mechanical mixer was run by a motor. In the absence of an alternate motor, the air diffusers served in the night time. Due t o a failing motor when the experiments for the second set were carried out, mechanical mixing was only occasionally employed. Mechanical mixing insured uniform concentration of the return sludge in the homogenizing chamber. The recycle sludge from the homogenizing chamber was pumped continuously to the aeration tank by a variable speed peristaltic action pump (Type MARE 7, Watson-Narlow Ltd.) through a silicon rubber tubing. Prior to experimentation, the return sludge pump was calibrated using the Pepsi Plant mixed liquor.
Feed waste The Pepsi Cola combined wastewater was used in the study as the carbon source. Aziz4 detailed the process of Pepsi Cola production and the source of the combined waste. His analysis of the waste and that of Chanika5 are presented in Table I . TABLE I Pepsi Cola Waste Characteristics ~~
~
Analysis PH Settleable solids Suspended solids COD BOD DO NH3-N Org-N P0,Total phosphorus
Aziz'
Chanika5
7.4-9.6 0.2-0.5 mg/li ter 25-60 635-1685 663-1 145 5.1-6.2 1.1-1.7 1.6-2.1 6.4-10.6
7.6-8.9 40-60 825-1450
1.0-2.0 1.5-1.8 0.1-0.25
CONSTANT RECYCLE SOLIDS CONCENTRATION
149
An examination of Table I reveals that the nitrogen and phosphorus contents of the waste are very low. Deficiency in these nutrients can affect biological activity of the microbes. Since these nutrients are nonlimiting in this study, ammonium sulfate according to the COD to N ratio of 20: 1 and potassium dihydrogen phosphate were added. Potassium dihydrogen phosphate served not only as a source of phosphorus, but also to adjust the pH of the waste to neutral level. The low suspended solids content of the waste was an advantage. It indicates that the organic content is mostly in soluble form and conforms to the assumption of negligible suspended solids in the influent made in the development of the model. The feed solution was prepared daily or as often as needed. The COD of the feed was maintained at 1000 mg/liter. When the COD of the waste was above this level, it was diluted with tap water. Occasionally the COD of the waste was below the desired concentration. When this occurred, it was introduced into the treatment unit in its present state.
Mass culture The original mass culture was obtained from the Pepsi Cola Waste Treatment Plant. Two liters of the Pepsi mixed liquor were added to 3 liters of the nutrient-supplemented feed waste in an 8 liter vessel and were well aerated. The pH was adjusted to a neutral level with the phosphate buffer. The culture was refreshed twice a day.unti1 the concentration reached about 5000 mg/liter. One liter of the culture was then transferred to the aeration tank to serve as inoculum for the continuous flow experiment ; the remaining portion was placed in the homogenizing chamber to supply the recycle requirement. Analytical Procedures The biological solids concentration in the reactor, clarifier effluent, and homogenizing chamber was determined by the membrane filter technique. Millipore membrane filters with a pore size of 0.8 p were used. The chemical oxygen demand of the waste, effluent filtrate, and mixed liquor (Set A experiments) was measured in accordance with Standard Methods.6 A spot COD test wai also made on the recycle filtrate. The sludge volume index of the mixed liquor was determined according to Standard Methods.6
150
BONOTAN-DURA AND YANG
Operating Procedures To start a continuous culture experiment, a batch culture was initiated by inoculating the aeration tank with 20y0 by volume mass culture. The pH of the culture was controlled a t 7.0 f 0.3 by the phosphate buffer system. After approximately 20 hr of batch cultivation, continuous flow studies a t a selected dilution rate were started. The mixed liquor temperature was measured. The treatment unit was operated as a once-through system for several hours prior t o operation with cell recycle. The system was allowed to reach the steady state. The system was considered t o have approached this condition when the substrate concentration in the effluent had practically become constant. Samples of the mixed liquor, clarifier effluent, and return sludge were collected at average intervals of 5 hr, until a sufficient amount of data had been collected. The influent flow rate was checked frequently to ensure that the desired rate was maintained.
Parameters and Variables I n Table 11, two sets of continuous flow experiments a t dilution rates of 1/8, 1/6, 1/4, 1/2, and 1/1.5 each were performed. For set A, the recycle cell concentration was maintained at about 5000 mg/liter and for set B, the recycle cell concentration was controlled a t approximately 7000 mg/liter level. I n all cases, the recycle volumetric flow rate was 30% of the wastewater inflow (a = 0.3) and the feed concentration was maintained a t 1000 mg/liter COD. The pH of the reactor contents was held at 7.0 f 0.3. A mixed liquor temperature of 27 f 3°C was observed during the duration of the study.
-
Substrate concentration (Si)
TABLE I1 Summary of Parameters and Variables Dilution rat.e
(D) 1/8 1/6 1/4
COD = lOOOmg/ liter
112 111.5
Recycle cell concentration (XR)
Hydraulic recycle ratio
5000 mg/liter 0
7000 mg-liter
=
0.3
CONSTANT RECYCLE SOLIDS CONCENTRATION
151
The experimental investigation was initiated at the dilution rate of 1/2 hr-I for set A. The dilution rate was then adjusted to 1/4, 1/6, and 1/8 hr-l. To have adequate data points, studies at the dilution rate of 1/1.5 hr-' were also conducted. For set B, studies were conducted according to the increasing magnitude of the dilution rate, that is, 1/8, 1/6, 1/4, 1/2, and 1/1.5 hr-l.
RESULTS AND DISCUSSION Performance Characteristics of a CM AS System Operated at Constant Recycle Solids Concentration For CMAS fed with bottling plant (Pepsi Cola) waste of 1000 mg/liter COD, the behavior of the operational parameters, namely, the reactor biological solids concentration, effluent COD, and effluent suspended solids at various dilution rates when the recycle solids was set at about 5000 mg/liter are shown in Figures 3 to 7. It is seen that at each dilution rate, the reactor biological solids and effluent soluble COD did approach a steady-state condition and the recycle solids concentration did remain almost constant. The 2000
I
I
I800 -
- 8IOCOGICAL
I
I
SOLIDS
(7)
CELL RECYCLE CONC
10000 -9ooo
( X,
W
600
-
400
-
-
J
-3000 EFFLUENT
COD ( 3 )
V W
LT
EFFLUENT SOLIDS ( X, )
-2000
-A
J W V
0
20
40
60
80
100
TIME, HOURS
Fig. 3. Steady-state parameters a t D
= 1/2 hr-1, X z ==. 5000 mg/liter (feed COD = 1000 mg/liter).
BONOTAN-DURA AND YANG
152
J
0
-3000
5w
a
z
4
EFFLUENT SOLIDS
200
0
-2000
( Xc )
(5)
EFFLUENT COD
20
40
60
1000
J J W
0
80
TIME, HOURS
Fig. 4. Steady-state parameters at D = 1/4 hr-1, X R = 5000 mg/liter (feed COD = 1000 mg/liter).
20001
1800(
I
BIOLOGICAL SOLIDS
(
I
1)
1 - 9000
> z ga -I
\ 0
-8000
1600-
1400
E
-
-7ooo CELL RECYCLE CONC. (X,)
v)
1200IW
5
m 1
-d
(L a
-_
5000
800-
Fig. 5.
Steady-state parameters at D
-6000
5
: 8 Y
=
1/6 hr-1, X R = 5000 mg/liter.
CONSTANT RECYCLE SOLIDS CONCENTRATION
153
\ 3
4
F
\
z
ga
W
E
t-
v)
z
a
6000
W V
I
So00
0
W IW
z
a
a
W
2
4000
0 W
3000
0
-J
V
>
t-
a 0
0
400
z
-7
0
(( X Xcc ))
m
EFFLUENT
(s)
- 11000000
60 40 60 TIME, HOURS
20
J J
EFFLUENT EFFLUENT SOLIDS SOLIDS COD
uwa W V
0 inn 1 00
80
Fig. 6. Steady-state parameters at D = 1/8 hr-1, X R = 5000 mg/liter (feed COD = 1000 mg/liter). 1600
I
I
I
-8000
1
-J \
W
1400
-
FFLUENT COD
(5)
EFFLUENT SOLIDS [ X,)
-2000
y
a
TIME, HOURS
Fig. 7. Steady-state parameters at D = 1/1.5 hr-1, X R
2:
5000 mg/liter.
154
BONOTAN-DURA AND YANG
effluent solids level was generally low signifying good settleability in the clarifier. The mean steady-state values of the operational parameters mentioned at each dilution rate studied are given in Table 111. Also shown in this table are the loading factors, specific substrate removal rates, COD removal efficiencies, and cell output rates. The mean steady-state reactor biological solids concentration and effluent soluble COD corresponding to each dilution rate shown in Table I11 are plotted in Figure 8 (solid lines). It is seen that the CMAS system remained stable up to a dilution rate of 0.35 hr-l. Beyond this dilution rate, the effluent soluble COD started to rise and the biological solids began to be diluted out from the reactor. The variations in reactor biological solids, effluent soluble COD, effluent suspended solids, and COD of inflowing Pepsi Cola wastewater at dilution rates considered for the CMAS operated at “constant” recycle solids concentration of 7000 mg/liter are depicted in Figures 9 to 13. These figures also show that a steady-state condition with regard to reactor biological solids and effluent soluble COD has been attained.
0
g e
PREDICATED BIOLOGICAL SOLIDS CONC .T
I
, 2 0 0 ~ OBSERVED BIOLOGICAL SOLIDS CONCENTRATION
I
0.2
I
0.3 0.4 DILUTION RATE
,
,
1
0.5
0.6
1 0.1
HOUR-’
and s’ and the predicted values from the Gaudy model with X R (5000 mg/liter) as system constant.
Fig. 8. Observed levels of
a
0.125 0.167 0.25 0.5 0.67
D X 24.
1000 982 1000 1000 882
Sp. substrate utilization rate = D(Si
X
4X
AS.
6.2 4.6 3.1 1.5 1.2
Loading factor =
1/8 1/6 1/4 1/2 1/1.5
Dilution rate (hr-1) 1623 1632 1877 1.587 1074
- s)/x.
60 50 73 192 539
x
4826 4871 5267 4870 4601
53 143 55 154 193
9.2 12.0 16.0 37.8 66.0
Loading Reactor detention factor" (mg Si 9 XR X. COD/mg time (hr) (mg/liter) (mg/liter) (mg/liter) (mg/liter) (mg/liter) SS/day) 0.074 0.144 0.161 0.331 0.27
Sp. substrate utilizationb rate (mg COD/mg SS/hr)
94.0 95.0 92.7 80.8 39.0
4 24 16 77 129
COD Cell removal output efficiency mg/liter/hr (%I (DXe)
TABLE 111 Mean Operational Parameters a t Different Dilution Rates with Pepsi Cola Wastewater as Substrate (a = 0.3, X R = 5000 Nominally)
BONOTAN-DURA AND YANG
156 2200-
-10000
-I
\
-8ooo
-x"F
-7000
z 0
- 9 m
J \
-
-6000 -5000
ta 5 W V 0' V
-4000 EFFLUENT SOLIDS
( X,)
-3000
-2000
9 8 a j
w
- 1000
V
looo
TIME, HOURS
Fig. 9. Steady-state parameters at D = 1/8 hr-1, X R = 7000 mg/liter (feed COD controlled at 1000 mg/liter).
2
W V 2
0 FEED
COD
V
(Si)
Y
it
-'-
,.4-.,.
LA..
_I
100 A I
10
A
-J w -J
u
-.".-.d.''
EFFLUENT COD ( A
8 a
/e.,,
a , . .
W'
3)
A
w
V >.
EFFLUENT SOLIDS ( X c )
A
A
A
A
I
I
I
I
20
30
40
50
TIME, HOURS
4
6b0
Fig. 10. Steady-state parameters at D = 1/6 hr-1, Xg = 7000 mg/liter.
CONSTANT RECYCLE SOLIDS CONCENTRATION
157
2200t BIOLOGICAL SOLIDS (
)
2000
K
' -I
z
v)
II:
w iw 1200 -
I-
-600
z
a CL
1000
r
0
800
-
2 w
I-
a 0
0
z
FEED COD
0
0
P
P
J -I
-400
--
0
g W
-I
200 100
5z
./
EFFLUENT
SOLIDS ( Xc )
EFFLUENT
COD
I
I
u > 0 -2O0
( $ )
\ 1
-.d I
.' 1
E
-.
-
2
Fig. 11. Steady-state parameters at D = 1/4 hr-1, X R = 7000 mg/liter.
Fig. 12. Steady-state parameters at D = 1/2 hr-1, X R = 7000 mg/liter (feed COD controlled at 1000 mg-liter).
BONOTAN-DURA AND YANG
158
-czs H
1 mFEED
COO
( Si)
Kxx)
0 800-
-4Ooo -I J
W V
600-
W V W P
_t EFFLUENT SOLIDS ( X e )
200,_ , _ _
I
5
.o-’I .-.o-.-.-.o--.---.e.-.’.-o._.-. -0 I 10
I5 TIME, HOURS
20
25
3oo
Fig. 13. Steady-state parameters at D = 1/1.5 hr-l, X R = 7000 mg/liter.
The average steady-state values of the operational parameters, mentioned above, at each dilution rate studied are presented in Table IV. In this table, the loading factors, specific substrate removal rates, COD removal efficiencies, cell output rates, and substrate concentration in the recycle sludge are also given. The mean steady-state values of the reactor biological solids and effluent soluble COD are plotted in Figure 14. This figure shows that the effluent soluble COD began to rise at the dilution rate of 0.33 hr-*. The reactor biological solids remained highly stable until the dilution rate of 0.5 hr-* was reached. These results show that for an input of about 1000 mg/liter COD of bottling plant wastewater, a CMAS process with constant recycle solids concentration of 5000 or 7000 mg/liter could be operated up to a dilution rate of 0.5 hr-1 without appreciable dilute-out of reactor biological solids and considerable substrate leakage in the effluent. However, it is noticed that the substrate concentration in the recycle sludge, S,, are lower than 8 in Table IV. Such a reduction can be related to the mean hydraulic retention time in the sludge settling and reaeration tanks. The rather high increase of S , after the dilution rate of 1/2 hr-1 can be related to the lower hydrauxc retention
1/8 1/6 1/4 1-2 1/1.5
0.125 0.167 0.25 0.50 0.67
Dilution rate (hr-1)
6.2 4.6 3.1 1.5 1.2
1000 969 982 lo00 lo00
Reactor detention Si time (mg/ (hr) liter)
34 32 63 158 266
10 8 36 132 269
1844 1892 1907 1898 1746
Substrate concentraB tion in 8 (mg/ recyclesludge (mg/ liter) (&, mg/liter) liter)
7098 7224 7356 7350 7639
120 150 69 116 68
8.1 10.3 15.4 31.6 46.0
0.084 0.11 0.16 0.30 0.24
96.6 96.7 93.6 84.2 63.4
15 25 17 58 45
Sp. substrate Loading utilization COD Cell factor (mg rate (mg removal output XR X, COD/mg COD/mg efficiency mg/liter/hr (mg/liter) (mg/liter) SS/day) SS/hr) (%) (DXJ
TABLE IV Mean Operational Parameters a t Different Dilution Rates with Pepsi Cola Wastewater as Substrate ( a = 0.3, X R = 7000 Nominally)
Ei
3
Q 0
m
U
Y
g ik
m
d
*
3 EQ
b-
BONOTAN-DURA AND YANG
160 2ooQ
A
1800-
\
1600-
LEGEND v)
a W
1400-
+ W
..
- OBSERVED BIOLOGICAL SOL1DS CONCENTRATION - PREDICTED
2 1200U
DILUTION RATE
SOLIDS
,
BDLOGICAL CONCENTRATlON
HOUR-'
s
Fig. 14. Observed levels (solid line curve) of t, and and the predicted values (broken line curve) from the Gaudy model with X e (7000 mg/liter) as a system constant.
time in the reaeration tank because a higher recycle flow rate was operated to maintain a = 0.3. Conventionally, the volume of the sludge or solids after 30 min settling of the mixed liquor in a 1 liter cylinder is combined with the reactor suspended solids to form a sludge volume index (SVI) which can be used as an indication of the settling characteristics of the sludge. These data have been collected and are shown in Table V. TABLE V Sludge Volume Index at Different Reactor Detention Times ~
SVI
Reactor detention time (hr)
X R = 5000 mg/li ter
X R = 7000 mg/liter
1.2 1.5 3.1 4.6 6.2
41.4 148.0 227.0 354.0
71.9 76.1 70.5 213.0 207.0
~
~~
CONSTANT RECYCLE SOLIDS CONCENTRATION
161
It can be seen that the application of higher constant recycle concentration of sludge do decrease the values of SVI with higher reactor retention time. The SVI values can be decreased from 35 to 150, which are reported by Metcalf and Eddy Inc.’ as the typical SVI values for good settling sludges when the constant sludge concentration are increased to the values higher than 7000 mg/liter. Therefore, the model incorporated with constant recycle sludge concentration does not have an inverse effect on the process. Predictability of the C M A S Model for Operation at Constant Recycle Solids Concentration The values of the kinetic constants, namely, maximum specific growth rate (pm), saturation constant ( K 8 ) ,and sludge yield (U), need to be known in order to be able to use the CMAS model of Ramanathan and Gaudy.* Since cell maintenance energy could have a significant role in CMAS systems with solids recycle, the true sludge yield ( Y , ) and the cell maintenance coefficient ( K d ) were evaluated using the straight-line approach of Marr et a1.* as shown in Figure 15. The maximum specific growth rate and saturation constant were calculated and given in Table VI. Employing the kinetic constants in Table VI, computer programs were designed to obtain the reactor biological solids concentration
-.-
IA
0 U
Y
U
A
0-A,X,
= 5000 m p / l
a--B,x,
= 7000 mg/l
_I
a
0
I -
0
Y+
0
=
I/intercept
k d = Yt x slope
LT
a
0 W
Lz
0
~~
RECIPROCAL
OF SPECIFIC GROWTH RATE
~~
( I /U
Fig. 15. Determination of true yield ( Y J and maintenance coefficient according to the method of Marr et a1.8
(Kd)
162
BONOTAN-DURA AND YANG
TABLE VI Kinetic Parameters Determined in the Present Study for Bottling Plant (Pepsi Cola) Wastewater Values Kinetic parameters Maximum specific growth rate (s,), hr-' Saturation constant ( K 8 ) ,mg/liter True yield (Yt) Cell maintenance coefficient (Kd),day-]
XR
X R = 5000 mg/liter
=
7000 mg/liter
0.45
0.15
221 0.76
200 0.33 0.072
0.14
and effluent soluble COD predicted by the proposed model for two cases : 1) when the cell maintenance coefficient K d is neglected (refer to Table VII for the equations) ; 2) when the cell maintenance coefficient Kd is considered (refer
also to Table VII for the equations). TABLE VII Steady-State Equations for CMAS Model Employing Constant X R Ramanathan and Gaudy' without K d
s =- b
f d b l - 4 aC
2a
(2)
u=pm-(l+a)D b = D [S;
(1
-
(1
+ a) k,]
Srinivasaraghavan and Gaudy9 with Kd
s
=
-b f d b a
- (1 + a ) D -k
u =
p,
b
D [S; - (1
=
+ a)
- 4 aC
(5)
2a
(1
kd
+ a) k.]
+4
C = k. D Si (3)
C
=
k. D S;
+ (1 + 4 kd
~
*
k, . S;
CONSTANT RECYCLE SOLIDS CONCENTRATION
163
The computer results are given in Table VIII with the observed values. At the conditions (pH, temperature, influent COD, hydraulic recycle ratio, and recycle solids concentration) maintained in this study, the cell maintenance coefficient, apparently, did not have any significant effect on the predicted values of reactor biological solids and effluent COD. TABLE VIII Effect of Maintenance Coefficient on the Predicted Levels of 8 and Model”
s by Gaudy
Predicted Values by Gaudy Model Dilution rate Set (hr-l)
A
B
a
0.125 0.167 0.25 0.50 0.67 0.125 0.167 0.25 0.50 0.67
Observed values
Without K
d
With K
d
s
s
8
s
B
B
1623 1632 1877 1587 1074 1844 1892 1907 1898 1746
60 50 73 192 539 34 32 63 158 366
1675 1667 1753 1588 1420 1881 1899 1924 1895 1874
30 41 62 158 207 33 43 69 167 238
1618 1624 1722 1674 1411 1847 1873 1906 1886 1867
31 42 63 159 207 33 43 69 167 238
Predicted values were obtained with the aid of a computer.
The predicted levels of reactor biological solids and effluent COD (broken lines) are presented graphically in Figure 8 for “constant” X R of 5000 mg/liter and in Figure 14 are shown for “constant” X R of 7000 mg/liter in order to facilitate the comparison with the observed values (solid lines). The predicted and observed values closely agreed at dilution rates below 0.5 hr-l, that is, before the dilute-out zone. At dilution rates above 0.5 hr-’, the predicted and observed values tend to diverge from one other. However, in Table VIII it is understood that if the concentration of X g had been increased more than 7000 mg/liter, the present system would undoubtedly demonstrate a closer agreement between observed and predicted biological solids and effluent concentrations a t a higher dilution rate. Therefore, it can be concluded that the CMAS model proposed by Ramanathan and’Gaudy2for operation a t constant X R does offer high predictability.
BONOTAN-DURA AND YANG
164
Design Parameters for C M A S Feeding on Bottling Plant (Pepsi Cola) Wastewaters The kinetic parameters in the proposed model, which have been evaluated in the present study, are presented in Table IX. Also presented are the values from studies of a CMAS continuous flow once-through system made by Ch ar~ ik a.~ TABLE IX Design Parameters for CMAS Feeding on Bottling Plant (Pepsi Cola) Wastewaters Present work Kinetic parameters Maximum specific growth rate (PA,hr-’ Saturation constants ( K 8 ) , mg/liter True yield (Yi) Cell maintenance coefficient ( K d ) , day-1
Chanikas 0.62 209 0.57
0
X R = 5000 mg/liter 0.45 22 1
0.76 0.14
X R = 7000 mg/liter 0.15 200 0.33
0.072
The values of p , , K , , Y t , and Kd for X I Zof 5000 mg/liter are found to be higher than those found for the XR of 7000 mg/liter. Specifically, p , for “constant” X R of 5000 mg/liter is three times larger than the p m for constant X R of 7000 mg/liter. The true sludge yield and cell maintenance coefficient, too, are about twice that for XR of 7000 mg/liter. The p , values from Chanika’s studies5, however, is found t o exceed the p, values in the present work. The K , value is about the same for both types of CMAS systems. The sludge yield value, though, is lower than that for the X R of 5000 mg/liter but higher than that for X R of 7000 mg/liter. Although the difference of these values could be possibly due to some change in the predominating microbial species populating these CMAS systems, it is also quite possible that the increase of XR (higher than 7000mg/ liter) may develop a trend which changes these values. With the latter possibility in mind, the present values were compared to the extended aeration process (higher sludge retention time) by using the same wastewater.1° It was found that the values calculated from the extended aeration process were much lower than the values obtained in the present study. Therefore, it seems reasonable that
CONSTANT RECYCLE SOLIDS CONCENTRATION
165
the change in these parameters is related to the change in the concentration of X R . This work was supported by a scholarship donated by the United Kingdom through the Asian Institute of Technology.
Nomenclature recycle solids concentration factor, X R / ~ dilution rate, F / V influent flow rate of wastewater, liter/hr substrate concentration in the reactor, measured as COD, mg/liter substrate concentration in the influent wastewater, measured as COD, mg/liter biological solids concentration in the reactor, mg/liter concentration of biological solids, mg/liter biological solids concentration in the clarifier effluent, mg/liter recycle biological solids concentration, mg/liter observed sludge yield true sludge yield saturation constant, numerically, i t is equal to the substrate concentration at which the specific growth rate is half of the maximum specific growth rate, mg/liter cell maintenance coefficient, day-1 the maximum specific growth rate, hr-1 recycle flow ratio specific growth rate, hr-l recycle substrate concentration, mg/liter
References 1. M. Ramanathan and A. F. Gaudy, Jr., Biolechnol. Bioeng., 11, 207 (1969). 2. M. Ramanathan and A. F. Gaudy, Jr., Biotechnol. Bioeng., 13, 125 (1971). 3. A. F. Gaudy, Jr. and R. Srinivasaraghavan, Biotechnol. Bioeng., 16, 723 (1974). 4. J. A. Aziz, Master Thesis No. 269, AIT, Bangkok, Thailand, 1969. 5. T. Chanika, Master Thesis No. 759, AIT, Bangkok, Thailand, 1974. 6. APHA, AWWA, WPCF, Standard Methods, 13th ed., American Public Health Association, New York, 1971. 7. Metcalf and Eddy, Inc., Wastewater Engineering, McGraw-Hill, New York, 1972. 8. A. G. Marr, E. H. Nilson, and D. J. Clark, Ann. N.Y. Acad. Sci., 102,536 (1963). 9. It. Srinivasaraghavan and A. F. Gaudy, Jr., Ann. Ind. Waste Conf., Purdue University, Lafayette, Indiana, 1974. 10. Y. K. Chen, Master Thesis, AIT, Bangkok, Thailand, 1975.
Accepted for Publication October 14, 1975
