. f". “WW-w v. mflwfl-owqmfl'... _- I ~ RELATIONSHSP BETWEEN‘RErENTIONL . - , TEME AND. SPECIFIC GROWTH RATE m A MUNICIPALWASTEWATER UNDER - commuous now coumnous - - Thesis for the Degree of M. S. MICHIGAN STATE UNIVERSETY JGHN K. NELSON 1973 ,3 M? LIB I“ ,1 RY Michigan State University nmome av“ HMS & SUNS 800K BIND‘RY INBS. LSBRARY B DER 'nmnw' OT .I. u' ABSTRACT RELATIONSHIP BETWEEN RETENTION TIME AND SPECIFIC GROWTH RATE IN A MUNICIPAL WASTEWATER UNDER CONTINUOUS FLOW CONDITIONS BY John K. Nelson A relationship between bacterial specific growth rate (kl) and bacterial specific respiration rate (kr) was de- veloped for a municipal wastewater. Determination of the specific respiration rate was accomplished by use of a con- tinuous flow respirometer connected to a mixed bacterial culture grown under continuous flow conditions with munici— pal wastewater as a feed solution, using the equation: kr - b k1 = d where k1 = specific growth rate (hr-l) d = mg of oxygen consumed per gram of volatile suspended solids (VSS) produced k = specific reSpiration rate r (mg Oz/hr/g VSS) U‘ I! endogenous reSpiration rate (mg OZ/hr/g VSS) The data resulted in an endogenous respiration rate of b = 27 mg Oz/hr/g VSS under steady state conditions. John K. Nelson In addition, the value of d was found to be 540 mg 02 con- sumed per gram of volatile suspended solids produced. A linear relationship between effluent soluble sub- strate concentration and specific growth rate k was de- l veloped using the following equations: COD 20 + 240 k s 1 TOC = 9 + 45 k s l where COD = dissolved chemical oxygen demand 5 (mg/l) TOC = dissolved total organic carbon 5 (mg/l) - These equations were valid up to k = 0.167/hr. l The linear relationship between specific growth rate and both dissolved CODS substrate concentration and specific . . . . -, -hr respiration rate became discontinuous arter k1 = 0.167 . This discontinuity was attributed to a substrate limitation. RELATIONSHIP BETWEEN RETENTION TIME AND SPECIFIC GROWTH RATE IN A MUNICIPAL WASTEWATER UNDER CONTINUOUS FLOW CONDITIONS BY John K. Nelson A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Fisheries and Wildlife 1973 cums; To Jan ii ACKNOWLEDGEMENTS The writer wishes to express his appreciation to Dr. Karl L. Schulze, Dr. Eugene W. Roelofs and Dr. Frank M. D'Itri for their assistance and guidance during the perform— ance of this research. Appreciation is also extended to the staff of the East Lansing Wastewater Treatment Plant whose support and encouragement allowed this research to be conducted at a practical level. iii INTRODUCTION. TABLE METHODS AND PROCEDURES. . . DATA AND RESULTS. DISCUSSION. . CONCLUSIONS . BIBLIOGRAPHY. OF CONTENTS iv Page 19 39 46 48 LIST OF TABLES TABLE Page I. Average D-values, Retention Times and Feed Rates. 0 O O O O O I O O O O O O O O O O O O 0 17 II. Substrate and Respiration Data (tr = 48 hr. D = 0.0208). 0 O O O I O O O O O O O O O O O O 20 III. Substrate and Respiration Data (t = 24 hr. D = 0.0417). . . . . . . . . . . F . . . . . . 21 IV. Substrate and Respiration Data (t = 12 hr. D = 0.0833). . . . . . . . . . . f . . . . . . 22 V. Substrate and Respiration Data (tr = 6 hr. D = 0.167) O O C O O O O O O O O O O O O O O O 23 VI. Substrate and Respiration Data (tr = 3 hr. D = 0.333) I I O O I O O O O O O O O O I C O O 24 VII. Average Soluble Substrate and Respiration Data 25 VIII. Comparison of Experimental Specific Growth Rates (kl) . O O O O O O O O O O C O O O O O O 44 FIGURE 1. 2. 10. ll. 12. l3. 14. 15. LIST OF FIGURES Chemostat and respirometer. . . . . . . . . . . Continuous flow culture apparatus showing mixer and respirometer cell . . . . . . . . . . . . . Flow through respirometer attached to continu- ous flow culture apparatus. . . . . . . . . . . Flow through respirometer showing respiration cell, variable speed peristalic pump and DO meters. . . . . . . . . . . . . . . . . . . . . Respiration cell. . . . . . . . . . . . . . . . Variation of overall respiration rate (rr) for a series of D-values--no temperature correction Variation of overall respiration rate (rr) for a series of D-values--corrected to 20°C . . . . Variation of influent and effluent dissolved CODS for a series of D-values . . . . . . . . . Variation of influent and effluent dissolved TOCS for a. series Of D-Values o o o o o o o o 0 Variation of temperature (°C) for a series of D values 0 C O O O O I O O O O I O I O O O O O 0 Variation of specific respiration rate (kr) for a series of D-values--corrected to 20°C . . . Plot of specific respiration rate (kr) at 20°C vs D. C C O O O O O O O O O O O O O O O O O O 0 Plot of effluent dissolved CODS vs D. . . . . Plot of effluent dissolved TOCS vs D. . . . . . Plot of S/kl vs S . . . . . . . . . . . . . . . vi Page 11 12 13 14 15 26 27 28 29 3O 31 33 35 36 38 BOD mg/l m1 TOC s VSS SYMBOLS Endogenous respiration rate (mg 02/9 VSS/hr). Biochemical oxygen demand (mg/l). Concentration of dissolved oxygen (mg/l). Teissier equation constant. Degrees centigrade. Dissolved chemical oxygen demand (mg/l). Dilution rate = Q/V. Weight of oxygen consumed per gram of volatile suspended solids produced. Dissolved oxygen (mg/l). Grams. Specific growth rate (grams per gram cell weight per hour). Maximum specific growth rate Specific respiration rate (mg 02/9 VSS/hr)- Milligrams per liter. Milliliters. Feed rate (ml/min). Overall respiration rate (mg/l Oz/hr). Substrate concentration (mg/l). Substrate concentration at which k1 = km/2. Temperature (°C). Retention time = V/Q (hours). Dissolved total organic carbon (mg/l). Volume (liter). Volatile suspended solids (mg/l). Concentration of volatile suspended solids (grams/1). vii INTRODUCTION The design, operation and performance of a biological wastewater treatment process is largely dictated by the rate at which the bacteria are capable of assimilating or oxidiz- ing carbonaceous matter. Assimilation of organic matter is accompanied by an increase in the number of bacteria. This increase in numbers may be defined by the specific growth rate constant (kl) which is part of an equation used to char- acterize the exponential multiplication of bacteria in a culture: kl = ln :X/XO) (I) k1 = specific growth rate x = cell concentration at time t x0 = initial cell concentration t = time elapsed Knowing the maximum rate of growth at which bacteria are capable of multiplying in a given wastewater is of im- portance in relation to design and Operational parameters for wastewater treatment. An increase in microbial mass can be determined directly by measurement of volatile suspended solids, analysis of total organic (cell) nitrogen, assessment of DNA content or application of the firefly luminescent test for measurement of cellular ATP (1). Common indirect measurements have re- lated bacterial density to turbidity or to the metabolic activity of cells. Oxygen uptake has been used for aerobic organisms because the rate of oxygen uptake is proportional to cell mass under certain conditions. Experimentation with continuous flow cultures of bac- teria has shown that a direct pr0portionality exists between the specific respiration rate (kr) and the specific growth rate (kl) (21): l k = — r = b + dk (II) r x r l k = specific respiration rate (mg 0 r per gram of cell weight per houg) x = cell concentration (grams/1) r = overall respiration rate (mg/l 02 r per hour) b = endogenous respiration rate d = mg 0 utilized per gram of cell weigfit produced k1 = specific growth rate (grams per gram cell weight per hour) Equation II states that by measuring the overall reSpira— tion rate (rr) and dividing this value by the cell concen- tration (x) a specific respiration rate (kr) may be found for a bacterial culture. Equation II also ShOWS that plotting kr versus k should produce a straight line with the y 1 intercept equal to the endogenous respiration rate (b) and the slope (d) equal to the mg 0 consumed per gram of cell 2 weight produced. Thus determination of the specific respiration rate (kr) would provide a means to obtain the specific growth rate (kl), if (b) and (d) are known. A variety of techniques are available for the measure— ment of overall respiration rate (rr) (10). These tech- niques rely primarily on the somewhat involved manometric (Warburg) principles or the less involved oxygen electrode systems. Ideally, the closer one can approach actual steady state conditions and at the same time measure bacterial respiration lfl.§iEE the better can laboratory results be re- lated to full scale treatment plant conditions. Several tech- niques are available for determining overall respiration rate (rr) of biological organisms under flow-through conditions (6-8,13-15,17,18,20,24). A unique flow—through respirometer using polarographic dissolved oxygen probes has been speci- fically adapted to wastewater treatment (8,17). This method uses a dissolved oxygen electrode immersed in a bacterial growth chamber with a second electrode immersed in a separate respiration cell through which the bacterial culture from the growth chamber is pumped at a pre-determined rate of flow. The bacterial cells in the respiration cell and in the pump line respire at the same rate as the bacterial cells in the growth chamber from which the culture is continuously drawn. The bacterial culture flowing through the respiration cell is closed off from the atmosphere and therefore the dis- solved oxygen level in the respiration chamber falls to a lower level than that existing in the aerated growth chamber. The oxygen level in the growth chamber and the oxygen level in the respiration cell are monitored separately with the difference between the two levels being proportional to the retention time of the bacterial suspension in the respira- tion cell. Measurement of the overall respiration rate (rr) of the bacterial culture in mg/l O utilized per hour 2 can thus be obtained by the following expression: (C - C )Q _ 1 2 rr — V .x 60 (III) r = overall respiration rate (mg/l r 02 per hour) C1 = mg/l of O2 in the aerated growth chamber C = mg/l of O in the respiration 2 2 cell Q = rate of flow in ml per minute through the respiration cell V = volume of the respiration cell and tubing (ml) The specific respiration rate (kr) may then be found as a result of a linear relationship which exists between the bacterial mass (x) measured as volatile suspended solids (VSS) and the overall respiration rate (rr) as shown in r . = 1: equation II. kr 3?. To provide steady-state conditions under which the respiration rate could be measured, a continuous flow cultur— ing technique was used. The Chemostat has provided a method by which a bacterial culture can be studied under continuous flow and complete mixing conditions (4,9,11-13,l9,21,24). The Chemostat is a device which consists of a feeding system that admits a nutrient liquid at a controlled rate to a con- stant volume reactor chamber. The flow of the nutrient liquid into the reactor and the resulting loss of bacterial biomass from the reactor is at a constant rate. The equation for the growth of bacteria in a batch culture can be written as (11): dx _ 5E “ kiX (Iv) k1 = specific growth rate (growth per unit cell weight per unit time) x = microbial cell concentration (grams/1) t = time (hr) The growth rate in a continuous flow culture then becomes: 5? = klx - Dx ‘ (V) D = dilution rate or flow (Q) divided by volume (V) i l/tr where tr = retention time Equation V shows that if D is greater than kl the culture will be diluted out more rapidly than it can grow and will eventually be completely washed out of the growth chamber. If D is less than kl the bacteria are growing more rapidly than they are being washed out and the density of the bac— teria in the reactor will continuously increase. Steady state operation is possible only when the specific growth rate (kl) is equal to the dilution rate (D). Equation V ex- presses this mathematically. For steady state, i.e., g% _ O _ I we obtain k = D (VI) Since the volume of the Chemostat is constant, the di- lution rate is determined by the feed rate and can be varied simply by varying the feed rate. This provided a means for establishing the relationship between specific respiration rate and specific growth rate. Decreasing the feed rate results in a prOportional decrease in substrate concentration due to the increased contact time between cells and substrate and vice versa. Corresponding to a given D-value, as long as it remains below the maximum growth rate km, the culture will automatically establish a definite substrate concentration in the reactor and in the reactor effluent (11). The system can thus be operated at a series of different growth rates and for each value of k the corresponding sub- 1 strate concentration (S) can be experimentally determined. Several mathematical models have been proposed for the rela- tionship between kl and S. The Monod equation is most often referred to in the literature (4,9,12,13,16,19,23,27) and is expressed as: S k1 - km(§—;§) (VII) n where k1 = specific growth rate km = maximum growth rate 3 = substrate concentration S = substrate concentration at n I = which kl km/Z A second model uses the Teissier equation (19,21) and may be expressed as: dkl The integrated forms of equation VII is similar to a first order reaction equation: kl = km(l - e'CS) (IX) From equation VIII it can be seen that the growth rate in- creases with the increasing substrate concentration propor- tionally to the difference between the existing growth rate (kl) and a maximum growth rate (km)’ In a third model the relationship between kl and S has been expressed as a two phase discontinuous relationship (3,4), where the rate of growth is directly proportional to the substrate conCentration up to a critical point (km). Above this point of discontinuity (km), the growth rate is constant and independent of substrate concentration. The experimental data obtained were then used in an attempt to establish relationships between specific growth rate (k1)' specific respiration rate (kr) and substrate con- centration (S) under actual field conditions using municipal wastewater as a feed solution. METHODS AND PROCEDURES The continuous flow culture apparatus is shown in Figures 1, 2, and 3. It was constructed from a 190 liter (55 gal.) barrel lined by a polyethylene container of equal size. The actual volume of the complete mix reactor was 160 liters (30 gal.). Complete mixing was maintained by a 1/3 horsepower Model S—7 Lightning Mixer running at a constant speed of 1725 rpm. Compressed air was passed through an air filter to a pressure regulator set at 6 psi. The air was then passed through an Ace Glass #lA-lS-l flow meter and dispersed from a modified Nalgene plastic aspirator for the purpose of supplying large bubble aeration. The mixing supplied by the Lightning Mixer served to break up the large air bubbles into a finer size. The dissolved oxygen range maintained during the experiment was 2.0 to 6.0 mg/l. Primary effluent from the East Lansing Wastewater Treatment Plant was obtained by tapping into a sampling pump which transferred a portion of the flow to a small stilling well that was continuously flushed by the incoming flow. From the stilling well, the primary effluent was pumped at a steady rate into the reactor using a Sigmamotor peristalic pump Model T65. The feed rate was determined by using a graduated cylinder and stopwatch. Effluent from the reactor was returned to the implant waste. 10 Due to the tendency for bacterial floc to adhere to the Side walls of the aeration chamber, the walls were brushed down daily. The respiration cell as illustrated in Figures 1, 2, 3, 4 and 5 was designed by R. Knop and has been described in his thesis on the determination of the overall oxygen trans- fer coefficient in an operating sewage plant (6). The device was constructed of galvanized sheet metal with a top made of 0.6 cm (1/4 in.) lucite attached to the base with eight screws to permit removal for cleaning. The lucite top was slanted to allow immediate passage of air bubbles through the effluent line. An inlet piece of 1.25 cm ID (1/2 in.) lucite tubing entered to within 1.9 cm (3/4 in.) of the bottom of the cell. The outlet of the cell was constructed of 1.9 cm (3/4 in.) lucite tubing to avoid possible clogging and provide fast entrapment of rising air bubbles. Stirring within the cell was provided by a variable speed Cenco Labora- tory Stirrer #18802 equipped with a chain stirrer attached to the shaft. The bacterial culture was pumped from the reactor to the reSpiration cell via a Cole Palmer Model 7017 peris- talic pump using 0.6 cm (1/4 in.) Tygon tubing. Monitoring of the dissolved oxygen levels in the reactor and the respiration cell was done with two dissolved oxygen meters employing polarographic dissolved oxygen probes (Model 54--Yellow Springs Instrument Co.). ll Figure l. Chemostat and Respirometer. l. 10. 11. Chemostat--complete mix reactor operated at continuous flow. Influent, wastewater feed. Chemostat effluent. Sparger and mixer blade. Variable speed peristalic pump drawing bacterial culture from Chemostat to respiration cell. Polarographic dissolved oxygen probe recording concen- tration of D0 in Chemostat. Respiration cell. Variable speed chain stirrer. Polarographic dissolved oxygen probe recording concentra- tion of D0 in respiration cell. Effluent from respiration cell. Air supply line. 12 Figure 2. Continuous flow culture apparatus show— ing mixer and respirometer cell. 13 Figure 3. Flow through respirometer-attached to continuous flow culture apparatus. 14 Figure 4. Flow through respirometer showing respiration cell, variable speed peristalic pump and DO meters. 15 Figure 5. Respiration cell. Polarographic dissolved oxygen probe Variable speed chain stirrer Influent from Chemostat Effluent from respiration cell 16 Using primary effluent from the East Lansing Wastewater Treatment Plant as a feed solution, a series of overall re- spiration rates (rr) were measured for a pre-determined D-value. D-values (D l/tr) or retention times were closely controlled by measuring feed rates every 4 hours and dividing these values into the volume of the reactor (tr = V/Q). The average D-values and retention times obtained in this manner are shown in Table I. For each series of D-values, the reactor was allowed to acclimate for a one-week period. This period of acclimation was then followed by a series of ten daily respiration rate and effluent substrate concentration measurements. Substrate concentration was determined as dissolved chemical oxygen demand (CODS) and dissolved total organic carbon (TOCS). Prior to the measurement of each overall respiration rate, the dissolved oxygen meters were calibrated against an air-saturated sample of tap water. Volatile suSpended solids (VSS) were analyzed by filtra- tion through a glass fiber mat and ignited at 550°C (25). Concentration of the volatile suspended solids was then designated as X and used to determine the specific respira- tion rate (kr) from the overall respiration rate (rr) as described in equation II. Total dissolved organic carbon (TOCS) and dissolved chemical oxygen demand (CODS) were determined for the filtrate 17 TABLE I AVERAGE D-VALUES, RETENTION TIMES AND FEED RATES D t Q .r_ r -1 . hr hr ml/min Y a + — .0207 2.8 1.9 48.3 55 .0420 2.2 2.2 23.8 112 .0833 2.8 2.2 12.0 223 .167 1.8 2.7 6.0 445 .333 2.4 1.5 3.0 889 4 Y = Average D value for period of experiment Average variation of % of D for period of experiment 18 obtained from the suspended solids analysis. The filtrate sample was acidified to pH 1 with sulfuric acid, sparged with nitrogen and analyzed for total organic carbon with a Beckman organic carbon analyzer (25). Chemical oxygen demand was measured by using an acidified solution of 0.01 N potas— sium dichromate and refluxing for two hours (25). DATA AND RE SULTS The oxygen utilization rates (kr and rr) obtained from the flow through respirometer and the dissolved substrate data (CODS and TOCS) for a series of D values are listed in Tables II t1) VII. Figures 6 to 11 represent the variation in the data for each D value. As can be seen from these graphs and the tabulated standard deviations, the variation in rr, kr, effluent CODs and effluent TOCS increased as the values of D increased. When influent CODS values are tabu- lated and graphed (Tables II to VII, and Figure 8), it can be seen that as the influent CODS increases and then declines with the last series of D values, so do the rr, kr and the effluent CODS values. Figure 9, showing incoming TOCS values, tends to follow those patterns as shown by the CODS values although not as pronounced as with the influent CODS values. As can be seen from Figure 10 and Tables II to VII, the temperature fluctuations during this experiment were small, with a range of 24°C-29°C. In order to compensate for any variation in rr due to these temperature changes, corrections were made by using the following formula (28): 19 20 coau6H>wU cumwcwum u o msam> mmmum>< n m Ilk. o.m H.m m.H m.m w.a moo.o m.o m.o o.a *6 OH mm mm mm mm mmo.o m.m H.v mm «m n I um I ow mwo.o n.m m.v mm m I am I mm mmo.o m.m H.v mm m mm mm mm mm mmo.o o.m m.m mm Ha I am I mm vmo.o o.m m.m mm HH Hm mm mm He mmo.o m.m m.v mm Ha mm mm mm ov mmo.o m.m H.v em Ha mm mm Hm mm mmo.o H.m m.m em Ha em mm mm mm mmo.o o.m m.m mm NH I em I mm omo.o m.m m.v mm m mm am mm mm ooo.o m.m v.v am Lem m aH Lem cH ocean H\m ooom u .mno o. 009 000 x mm> H @809 u mome.o n a .Hn me n D ¢B . .H came eHso.o u o .ua Hm I us dfida ZOHBdMHmmmm Q24 mfidemmDm HHH mqmflfi 22 0.0 0.0 0.0 0H 0.0 000.0 0.0 0.0 0.0 0 0H 00 00 00H 00 000.0 0.0 0.0 00 m 0H 00 00 00H 00 000.0 0.0 0.0 00 0H 00 00 00H 00 000.0 0.0 0.0 00 0H H0 00 00H 00 000.0 0.0 0.0 00 0H 00 00 00H 00 000.0 0.0 0.0 00 0H 00 00 00 00 000.0 0.0 0.0 00 0 00 H0 00 00 000.0 0.0 H.0 00 0 00 00 00 H0 000.0 0.0 0.0 00 0H 00 00 00H H0 000.0 0.0 0.0 00 0H 00 00 00H 00 000.0 0.0 0.0 00 HH H0 00 0HH 00 000.0 0.0 H.0 00 00m 0 0H 000 0H 0.000..H H\0 0.00 .000 0. cos goo 00> Esme 0000.0 n o .0: 0H u up flefla ZOHBdMHmmmm DZ¢ mBH mdmflfi 23 0.0 0.0 0H 00 00 H00.0 0.0 0.H 0.H 0 0H 00 H0 00H 00H 000.0 HH 0H 00 m 00 00 00 00H 00H 00H.0 0H 00 00 0H 00 00 00H 00 0HH.0 0.0 0H 00 HH I 00 I 00H 000.0 0H 0H 00 «H I H0 I 00 00H.0 0H 0H 00 00 00 00 00H 00H 0HH.0 0H 0H 00 0H I 00 I 00H 00H.0 0H 0H 00 00 00 00 00H 00H 000.0 0.0 HH 00 0H 00 00 00H 0HH 000.0 0.0 HH 00 0H 00 00 00 00H 000.0 0.0 0H 00 00 00 00 0HH 00 0HH.0 0.0 0H 00 000 m 0H 000 0H 0.00..H H\0 0.00 .000 0. 000 000 00> .0000 00H.0 o 0 n 0 0000 on000H0000 020 000000000 > mam/NH. 24 0.0 0.0 0H 00 0H 000.0 0.0 0.0 0.0 0 00 H0 00 00 00H H00.0 HH HH 0H m 00 00 00 H0 00H 000.0 0H 0H 0H 00 00 00 00 00H H00.0 0H 0H 0H 00 00 00 00 00H 000.0 0.0 0.0 0H 00 00 00 00H 00H 000.0 H.0 H.0 00 00 00 00 H0 00H 000.0 HH 0H 0H 00 00 00 00H 00H 000.0 0H 0H 0H H0 00 00 0HH 00H 00H.0 0H 0H 0H 00 00 00 00H 00H 000.0 0.0 0.0 00 00 00 00 00H 0HH H00.0 0.0 0.0 0H 00 00 00 00H 00H HH0.0 0H 0H H0 000 0 :0 000 0H 0.000 H\0 0.00 .000 0. 000 000 0 00> .0000 .H 000.0 I a .00 0 u 0 «Ema one mqmde 25 TABLE VII AVERAGE SOLUBLE SUBSTRATE AND RESPIRATION DATA D t r k COD TOC r r r S S hr'l hr Obs. 20°C 20°C In Eff. In Eff. 0.0208 48 4.1 2.3 38 68 26 27 10 0.0417 24 5.4 3.4 48 65 33 3o 12 0.0833 12 8.7 5.7 74 100 38 38 13 0.167 6 14 11 120 120 61 38 17 0.333 3 ll 11 140 98 65 41 29 26 D-Values 0.0208 A 0.0417 0 0.0833 0 0.0167 V 20... 0.3.33 V 18—. V 16—- v E a 14.1 . ‘3’ E V V P‘ m E 12—- V a H W V \N m o 10—- . . <> 0 g g e O 8&6 m a II ._ 0 O a HH 8 (v) 0 00 6_‘ ..22__U,__. . 90 m A ”AAA-2f AVAP 2—-I 1 l I I 1 0.0208 0.0417 0.0833 0.167 0.333 D(hr—l) Figure 6. Variation of overall respiration rate (r ) for a series of D-values--no temperature correction. 27 l a 6-i D-Values 15__‘ 0.0208A a ‘ 0.0417 0 l4_1 0.0833 0 a 8 0.167 V V 13_4 0.333 G V 12_‘ V V 11.... a 10.. V’ a 9...) V a a ’II 8 V? ,1: 'fi \\ H \N o 7.. I? VH 5— e 000 “ -0——v—- 9 0 5—4 0 a 4-H 90 L— 3—4 00° A. e H A A0 2-) AA AA ]-_4 _ 1 1 0.0208 0.0416 0.0833 1 0.167 0.333 I D(hr-l) Figure 7. Variation of overall respiration rate (rr ) for a series of D—values--corrected to 20° C. 28 D-Values* 0.0208.A 0.0417 0 0.0833 0 0.167 V 200 0.333 a . *1 *Inf. values shaded lBQ—q ‘ .‘II v IJ 160.1 E. v v 140_ I I H «u . 0 ‘5 120. V__—_—__— I I Q0 9 o 100“ ' vv --——'— U . .. .—-—‘-‘.‘—-. 8L ‘ ‘ .. ... V D a a H—' .0 v V V a '3 600 A v ‘7 ‘7.g ‘ 9 V’ ‘V a F 0 0 40H 0 '0 A. 9 0"? ‘7 a P—ew—v-V lrfifié A A 0 Po 909 20-0 0.0208 ‘ 0.0416 I 0.0833 I 0.167 I 0.333 I D(hr-l) Figure 8. Variation of influent and effluent dissolved CODS for a series of D-values. TOC (mg/l) 29 60—~ D-Values* 0.0208 A 1 0.0417 0 0.0833 6 ' 500 0.167 V I 0.333 m . *Inf. values shaded VV ' ' 9 .0 40““ O 6 . I II ¢ F E 1 00 V w [3 I S . .A 0 °. 660 30" s A . In I I. O. . ‘ s V I v v E' 20- ‘7 V 0 V ‘ 09 G} V WV (29-00903— 00 v MMA 008 v 10-4 63 00 A A I I I F I 0.0208 0.0417 0.0833 0.167 0.333 D(hr_l) Figure 9. Variation of influent and effluent dissolved TOCS for a series of D—values. 30 3o-—000 AAA 4%. AA 0 9 00 ‘UtriL-1mhv®wmmflmwF—— oeefiiyy: VV 20‘“I W '95 m —0*nau—uuu f\ 0.0208 A P 0.0417 6 0 0.0833 0 H 3 10.1 0.167 v 3 0.333 n (D Q. s (D E-I I I I I 0.0208 0.0417 0.0833 0.167 0.333 D(hr—l) Figure 10. Variation of temperature (°C) for a series of D values. k (mg Oz/hr/g VSS) r 31 160-. D-Values a a 0.0208 A 150-— 0.0417 G 8 0.0833 0 .3, _____IEI__fl. 0.333 a 130-— V V :1 I51 E! 120—. 09 110—3 V’ m ‘V .90-« 0 V v 80— 6 9 _$______49 70— 9 ° (9 0 O 60-: 00 50- o I 0 AA 40'1AAhér—AAA9 A. 30 I I I I I 0.0208 0.0417, 0.0833 0.167 0.333 D(hr_l) Figure 11. Variation of specific respiration rate (kr ) for a series of D-values--corrected to 20°C. 32 r20 = 10(rt - 0.0315AT) (X) r20 = rr at 20 C rt = the observed value of rr at temperature T AT = difference in temperature between observed value and 20°C Plotting the average corrected values of kr against the associated D values produced a curve as shown in Figure 12. It becomes evident that there exists a direct relation- ship between D and kr as shown by the straight line. Thus only one point, that obtained for D = 0.33 clearly does not fit the straight line relationship. Equations V and VI have shown that for steady state conditions k1 = D. The data presented in Figures 8, 9 and 11 suggest that steady state existed up to D = 0.167 i.e., up to a retention time of tr = 6 hr. At D = 0.33, tr = 3 hr., the performance of the reactor became very irregular. It is therefore assumed that up to a value of 0.167/hr. D = kl so that up to D = 0.167 the D values represent the specific growth rate at which the continuous flow mixed cul- ture was operating. The linear portion of the curve shown in Figure 12 follows the equation kr = b + dkl as previously discussed (Equation II). 33 .Q m> 0.0m um A000 0000 COHHMHHQmmH OHMHommm 00 uon .NH wusmwm 0800 GOHHvamH\H n 9 00.00.00. _ .0 _ NW «N; NN.ON. ma. ma. VH.NH. 0H. mo. wo.vo. No. _-_.H0u____ + _ _ M ovm _ H _ 100 I00 I00 I03 I0HH T 00H 1.00H IIOVH 11X (SSA f>/~Iq/ZO 5m) 9198 Uoraelrdsea orgroads 34 Using the least squares method, values of b = 27 mg OZ/hr/gram VSS and d = 540 mg Oz/gram VSS produced were ob— tained from the experimental data. Thus, kr = 27 + 540 kl (XI) kr - 27 k1 _ 54o (XII) In general, the data demonstrate that at 20°C, 0.540 “fi" grams 0f oxygen were consumed per gram of volatile solids ”j produced and that the endogenous respiration rate of the bacterial cells was 27 mg 0 per gram of volatile suspended 2 solids per hour when the specific growth rate (k1) equals zero. For kl values up to 0.167 the curve shows that the specific respiration rates (kr) were directly proportional to the specific growth rate (kl). A plot showing the relationship between D and dissolved substrate concentration remaining (CODS and TOCS) in the reactor indicated that the substrate concentration increased with increasing values of D (Table VII and Figures 13 and 14). In Figure 13 the dissolved CODS concentration increases in direct proportion to D up to a D-value of 0.167. Just as in Figure 12 the CODS concentration for D = 0.33 does not follow the straight line relationship. In fact, CODs at D = 0.33 is much lower than would be expected by extending the straight line. The data support the previously made assumption that 35 .Q 0> 900 Um>Hommwp usmsHmmm 00 uon @800 Hafiusmumu\a n 0 .MH musmflm G 00. 00. 00.00. 00. 00. 00.00. 0H. 0H. 0H. 0H. 0H.. 00. 00. 00. 00. H... rr________-_w___r-_._ m A 8 10H 6. D O H I m 00 e m. H row T. 00 + H 0:000 u 000 .w .100 ) m 16% / o _ u 100 I00 36 0m. mm. .0 0> 000 00>H000H0 0:0:H000 00 Hon 0500 COHusmpmu\H no .0H 005000 om.wm. om. 0N.NN. om. ma..mH. vH.NH. OH. mo. mo.¢o. No. _ _ I H I _ _ _ _ _ _ _ _ _ _ _ _ 0 + HHxv 00 u oo0 1.0m (I/bm) Suturemag DOI paAIossrq 37 steady state existed up to D = 0.167, so that D = kl up to D = 0.167. The TOCS data in Figure 14 show a similar relationship to D, except that in this case the TOCS value for D = 0.33 is somewhat higher than the straight line extension would indicate. As mentioned before, the relationship between kl and substrate concentration remaining has been expressed either as a direct prOportionality up to a maximum value or as a hyperbola by the Monod and Teissier equations. The applic- ability of these concepts to the experimental data was tested. Treating the relationship between specific growth rate and remaining substrate concentration as linear gave the equation COD = 20 + 240 (k ) s l as shown in Figure 13 for k values up to 0.167. 1 For the TOCS data shown in Figure 14, the following relationship was observed: TOCS = 9 + 45 (kl). The Monod equation in rectified form is: S/kl = sn/kM + (l/km)S (XIII) Plotting S/kl against S produced a curve as shown in Figure 15. According to Equation XII, a plot of S/kl versus 5 should 1300—« 1200i 1100.. 1000_. 900.4 400—3 300—4 200—0 100—4 Figure 38 I 30 CODS Plot of S/k A If 40 (mg/l) 1 VS S. 50 39 produce a straight line where the y-intercept would be equal to Sn/km and the slope would be equal to 1/km. Figure 15 shows that the experimental data do not agree with Equation XII. In applying the Teissier equation it is necessary to obtain the two constants km and c from the data. In an attempt to use the Slope Method (26) for this purpose, it was found that the four sets of experimental data for kl and S were insufficient. The data would thus indicate that using East Lansing wastewater, steady state conditions existed up to a retention time of 6.0 hours corresponding to a specific growth rate of k1 = 0.167.hr and a specific respiration rate of 120 mg 02/ hr/g VSS at an effluent substrate concentration of 61 mg/l as dissolved CODS and 17 mg/l as TOCS. DISCUSSION Experimentation has shown that under field conditions and constant flow, relationships between specific growth rate, specific respiration rate and substrate concentration existed for a domestic wastewater. Measurement of the soluble sub- strate (TOCs and CODS) and specific respiration rate (kr) of a continuous flow mixed bacterial culture provided a linear relationship with the specific growth rate (kl) under steady state conditions. When overall variations of influent CODs and TOCS values are compared with the overall variations of rr, kr and effluent CODS and TOCS (Figures 6, 7, 8 and 9), it would appear that there is a direct relationship between these parameters, e.g., as the incoming CODS and TOCs increased so did rr, kr and effluent CODS and TOCS values. However, when the variation for an individual series of influent CODs and TOCS values are compared with those of rr, kr and effluent CODS and TOCS it can be seen that for the four series of D values (0.0208, 0.0416, 0.0833 and 0.0167) no such direct relationship can be found. Large fluctuations of incoming CODS and TOCS values could not be correlated with large fluctuations in values for rr, kr' and effluent CODS and TOCS. Such were also the 40 41 findings of a recent study (13) where a threefold increase of influent CODS did not increase the effluent CODS of an aerated system using artificial substrate and mixed culture operating at a 6-hour mean residence time (k = 0.167). 1 The observed data support the theory of steady state effluent y CODS and TOCs values, regardless of incoming CODS and TOCS :l values up to D values of 0.167 (tr = 6.0). I a As mentioned before, Figure 12 shows that specific 3,0- respiration rate is directly proportional to specific growth rate up to the values of k1 = 0.167 and kr = 120. At D = 0.33 the value of kr = 140 is much lower than would be ex- pected by extending the straight line. In fact, according to equation XI the specific consumption rate0should have been kr = 27 + 540 x 0.33 = 205 mg 02/g VSS/hr at D = k1 = 0.33. Apparently the mixed culture in the aera- tion tank was not capable of multiplying at a rate of k1 = 0.33 per hour and therefore steady state operation was not possible at D = 0.33. The explanation may be found by going back to Figure 13 where the average substrate concentration in the reactor effluent was 65 mg/l measured as CODS. Accord- ing to equation XIII, at D = k1 = 0.33 CODS should be equal to 20 + 240 x 0.33 = 100 mg/l. Table VII indicates that at D = 0.33 even the incoming average CODS reached only 98 mg/l. Thus it appears that in this case the substrate concentration in the incoming feed solution (primary effluent) was too low 42 to sustain a specific growth rate of k = 0.33 per hour or a 1 specific oxygen consumption rate of kr = 205 per hour. The attainment of steady state conditions at D = 0.33 was there- fore prevented by the existing limitation in the substrate concentration. It would appear that the observed values of k1 = 0.167/ hr at a specific respiration rate of 120 mg OZ/hr/g VSS and a reactor substrate concentration of CODS = 61 mg/l represent maximum values for East Lansing Wastewater. The relationship between specific growth rate and dis- solved TOC substrate concentration in Figure 14 was linear up to a D-value of 0.167 yet did not indicate a discontinuity beyond this point. The decline of the incoming cons concentration as shown in Figure 8 at k = 0.33 was probably due to the decline of l the population served by the East Lansing Wastewater Treatment Plant. The final phase of this experiment was conducted at a time when the student body of Michigan State University had left the community for Christmas recess. The problem of limited substrate measured as CODS in sewage has been documented in an experiment where it was neces- sary to concentrate a domestic sewage prior to developing a curve showing the relationship between k and S (5). l The fact that domestic wastewater is a relatively poor growth medium in comparison to prepared nutrient solutions became evident under field conditions where the highest 43 specific growth rate was k1 = 0.167 at a specific respiration rate of 120 mg Oz/hr/g VSS and an effluent dissolved CODS substrate concentration of 61 mg/l. Data previously obtained from East Lansing primary effluent by manometric measurement of the overall respiration rate (rr) found the highest specific growth rate to fall in the range of 0.28 to 0.33/hr. (22). The experimentally ob— served value of k1 = 0.167 falls within the range of values used by Wuhrmann (27) and Pearson (16). These authors also attributed their lower kl values to the poor growth charac- teristics of sewage (Table VIII). Information pertaining to specific growth rates in an actual domestic wastewater is limited (Table VIII). To make a comparison between the values obtained from the data in this paper and those discussed in Table VIII may not be war- ranted in light of the varying techniques and composition of nutrient substrates used. The data gained from this experi- mental work represent specific growth rates obtained under continuous monitoring of the overall respiration rate using a continuous flow respirometer and directly relating these values to specific respiration rates while other techniques relied on batch processes, manometric measurements or theo- retical applications of reaction kinetics. A review of the literature has failed to provide additional data relating in situ respiration values to specific growth rates for 44 mopsHm ©000>0uow 000 .Uaoo msflpmummo pufismmd 0mm3mm I mmH.o Anmv :GMEHSSB HmpoE onw msHm 000 @009 COHHQEsmmd I I om.o Ammv £0050 000000000H 000300 00 00.0I00.0 H000 00H0000 muoumnonmq .mauud om om.o H00 0000000 0000000000 .0H000 00 0.0I0.0 H00 00000 UGESmmd I I om.o Amy msflskoa mmcsHm pmum>fluom How .psoo mcflumummo pmasmmm 0m030m I mmH.o HmHv cowummm Mpsum mo_0m>H .00mwnmasw o. HIM: mosmnmmmm .mfima Hx can Honusd HHAV 00000 m0so00 onHommm H00zmsHmmmxm 0o zo0Hm0mso0 HHH> mqmdfi 45 municipal wastewaters. It would appear that additional stud- ies are needed to develop more reliable estimates of the kinetic characteristics of an actual wastewater source so that it will be possible to predict performance and offer guidelines to process design on a rational basis. CONCLUSIONS 1. £n_§itu continuous, polarographic measurement of the over- all respiration rate (rr) of a municipal wastewater with conversion to specific respiration rate (kr) provided a rela- tionship with the specific growth rate (kl) under steady state conditions as described by the formula: k - b _r 13‘“??— Where b is equal to the endogenous respiration rate and d equals the amount of oxygen utilized per unit of volatile suspended solids produced. 2. At 20°C an endogenous respiration rate of 27 mg 02 per hour per gram of volatile solids present was found with d = 540 mg 0 consumed per gram of volatile suspended solids 2 produced. 3. Specific growth rate (kl) was found to be directly pro- portional with specific respiration rate (kr) and effluent dissolved CODS concentration up to a value of k = 0.167--hr l at a specific respiration rate of 120 mg OZ/hr/gram volatile suspended solids and CODS = 61 mg/l. This maximum value for 46 47 k1 was attributed to a substrate limitation in the incoming feed solution. 4. Specific growth rate (k1) was found to be directly pro- portional with dissolved TOCS concentration up to a value of k = 0.333"hr l at a TOCS value of 29 mg/l. 5. Steady state conditions were in all probability estab- lished for retention times of 48 hr (k = 0.0208), 24 hr 1 (k1 = 0.0417), 12 hr (k1 = 0.0833) and 6 hr (k1 = 0.167). 6. A relationship between specific growth rate and dis- solved substrate concentration could not be established through the use of the Monod equation. N a 10. BIBLIOGRAPHY "Biomass Determination--A New Technique for Activated Sludge Control." Water Pollution Control Research Series 17050 EOY 01/72, Environmental Protection Agency (1972). Downing, A. L. and Wheatland, A. B., "Fundamental Consider- ations in Biological Treatment of Effluents." Trans. Institution of Chemical Engineers, 40:2, 91 (1962). Eckenfelder, W. W. and McCabe, J., "Process Design of Biological Waste Systems for Industrial Waste Treatment." Proc. 2nd Symposium on Treatment of Waste Waters, Pergamon Press, New York (1960). Garrett, M. T. and Sawyer, Clair N., "Kinetics of Removal of Soluble BOD by Activated Sludge." Proceedings of the Seventh Industrial Waste Conference, Purdue Univ., 79, 51 (1952). Gaudy, Anthony F. and Gaudy, Elizabeth T., "Biological Principles for Design and Operation of the Activated Sludge Process." Water Pollution Control Research Series 17090 FQJ 09/71, Environmental Protection Agency (1971). Knop, Ralph J., "Determination of the Overall Oxygen Transfer Coefficient in an Operating Sewage Plant." Michigan State University Masters Thesis (1968). Ludzack, F. J., "Laboratory Model Activated Sludge Unit." Jour. WPCF, 32, 605 (1960). Mikesell, Ritchie D., "A Continuous Bacterial Respiro- meter--A Means to the Control of Aerobic Bacterial Pro- cesses." Unpublished Data. Monod, Jacques, "The Growth of Bacterial Cultures." Annual Review of Microbiology, 3, 371 (1949). Montgomery, H. A. C., "The Determination of Biochemical Oxygen Demand by Respirometric Methods." Water Research, 1, 631 (1967). 48 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 49 Novick, Aaron and Szilard, Leo, "Description of the Chemostat." Science, 113, 715 (1950). Novick, Aaron, "Growth of Bacteria." Annual Review of Microbiology, 9, 97 (1955). O'Brien, Walter J. and Burkhead, Carl E., "Oxygen Consump— tion in Continuous Biological Culture." Water Pollution Control Research Series 17050 DJS 05/71, Environmental Protection Agency (1972). O'Brien, W. J. and Clark, J. W., "A Continuous-Flow Recording Respirometer." Process Biochemistry, Oct. 21 (1967). O’Hara, James, "A Continuously Monitored Respiration Chamber for Fish." Water Research, 5, 143 (1971). Pearson, Erman A., "Kinetics of Biological Treatment." Advances in Water Quality Improvement, Univ. of Texas Press, Austin. 381 (1968). Reynolds, L. F., "Method for Measuring Aerobic Decomposi- tion Activity of Activated Sludge in an Open System—- Discussion." Water Research, 3, 319 (1969). Rueger, M. E., Olson, T. A. and Scofield, J. I., "Oxygen Requirements of Benthic Insects as Determined by Manometric and Polarographic Techniques." Water Research, 3, 99 (1969) Sack, W. A. and Schulze, K. L., "An Experimental Analysis of a Continuous Flow Model of the Activqted Sludge Pro— cess." Proc. 15th Okla. Indust. Waste Conf. Okla. S. U., Stillwater. Paper #8 (1964). Sanders, Walter M., "Oxygen Utilization by Slime Organisms in Continuous Culture." Water Pollution Int. Jour., 10, 253 (1966). Schulze, K. L., "The Activated Sludge Process as a Con- tinuous Flow Culture." Water and Sewage Wk., 111, 526 (1964). Schulze, K. L. and Hoogerhyde, T. C., "Bio-Degradability as Measured by a Differential Manometer." Developments in Industrial Microbiology, 8, 284 (1967). 23. 24. 25. 26. 27. 28. 50 Smith, Robert and Eilers, Richard G., "A Generalized Computer Model for Steady-State Performance of the Activated Sludge Process." Water Pollution Control Research Series, Environmental Protection Agency (1969). Stack, V. T. and Conway, R. A., "Design Data for Complete- ly Mixed Activated Sludge Treatment." Sewage and Indust. Wastes, 31, 1181 (1959). Standard Methods for the Examination of Water and Waste- Water--l3th Edition. American Public Health Assoc., New York, N. Y. (1972). Thomas, Harold A., "The Slope Method of Evaluating the Constants of the First Stage Biochemical Oxygen Demand Curve." Sewage Wks. Jour., 9, 425 (1937). Wuhrmann, Karl, "Research Developments in Regard to Concept and Base Values of the Activated Sludge System." Advances in Water Quality Improvement, Univ. of Texas Press, Austin. 143 (1968). Wuhrmann, Karl, "High Rate Activated Sludge Treatment and Its Relation to Stream Sanitation." Sewage and Indust. Wastes, 26, l (1954). MMMMM ITII‘ILIIIII’IIILII(III/[10111111110S