WATER POLLUTION CONTROL RESEARCH SERIES 17050 EOY 01/72
Biomass Determination - A New Technique
for Activated Sludge Control
ENVIRONMENTAL PROTECTION AGENCY
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes the
results and progress in the control and abatement of pollution
in our Nation's waters. They provide a central source of
information on the research, development, and demonstration
activities in the water research program of the Environmental
Protection Agency, through in-house research and grants and
contracts with Federal, state, and local agencies, research
institutions, and industrial organizations.
Inquiries pertaining to Water Pollution Control Research Reports
should be directed to the Chief, Publications Branch (Water),
Research Information Division, R&M, Environmental Protection
Agency, Washington, D. C. 20460
-------
BIOMASS DETERMINATION - A
NEW TECHNIQUE FOR
ACTIVATED SLUDGE CONTROL
by
BIOSFHERICS INCORPORATED
4928 Wyaconda Road
Rockville, Maryland 20852
for the
Office of Research and Monitoring
ENVIRONMENTAL PROTECTION AGENCY
PROJECT NO. 17050 EOY
CONTRACT NO. 14-12-419
CONTRACT NO. 14-12-871
January 1972
For sale by tlio Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $1.25
-------
EPA REVIEW NOTICE
This report has been reviewed by the Environmental
Protection Agency, and approved for publication.
Approval does not signify that the contents
necessarily reflect the views and policies of the
Environmental Protection Agency, nor does mention
of trade names or commercial products constitute
endorsement or recommendation for use.
11
-------
ABSTRACT
Research was conducted to determine the feasibility of using ATP as a
measure of viable biomass in activated sludge. Methods were developed
for the extraction of ATP from sludge and mixed liquor, and for the
determination of ATP using the firefly bioluminescent procedure.
Measurements of ATP were conducted on various pure cultures, pilot
plant and full-scale activated sludge treatment plants. Additional
parameters including BOD, TOG, oxygen uptake rate, and suspended
solids were measured to provide comparative and supportive information.
Preliminary tests in which ATP measurements of biomass were used to
control the percent sludge return were conducted at two full-scale
municipal sewage treatment plants. Lowered return sludge rates were
found to produce effective treatment and increase the biological
activity of the sludge. Changes in the rate of return sludge re-
sulted in changes in ATP concentration of mixed liquor which preceded
changes in suspended solids by as much as 24 hours. The assay was
found to be reproducible and rapid. Results can be obtained within
approximately ten minutes. This report was submitted in fulfillment
of Project Number 17090 EEM, Contract Number 14-12-419, and Project
Number 17050 EOY, Contract Number 14-12-871, under the sponsorship of
the Environmental Protection Agency.
111
-------
CONTENTS
SECTION
I CONCLUSIONS
II RECOMMENDATIONS
III INTRODUCTION
IV PRELIMINARY STUDIES
V TECHNIQUE IMPROVEMENT
VI QUANTITATION
VII CORRELATION BETWEEN ATP AND OTHER PARAMETERS
VIII PILOT PLANT STUDIES
IX OPERATION OF THE BALTIMORE CITY BACK RIVER
ACTIVATED SLUDGE SEWAGE TREATMENT PLANT
X OPERATION OF THE ARLINGTON COUNTY ACTIVATED
SLUDGE SEWAGE TREATMENT PLANT
XI SUMMARY
XII ACKNOWLEDGEMENTS
XIII REFERENCES
XIV ABBREVIATIONS USED IN TEXT AND FIGURES
XV APPENDIX
PAGE
1
3
5
7
19
23
37
51
61
81
99
103
105
107
109
v
-------
FIGURES
'.Co. Page
1 ATP Assay Instrument 7
2 Cuvette Turette 8
3 Photomultiplier Mount and Reaction Chamber 9
4 Block Diagram of Electronics Circuitry 10
5 Relationship of Measured ATP Concentration to
Bacterial Cell Concentrations of Pure Cultures 11
6 Relationship of Measured ATP Concentration to
Mixed Liquor Concentrations 12
7 Relationship Between Measured ATP Concentration
and Number of Bacteria in Mixed Liquor 13
8 Concentration of Intracellular and Extracellular ATP 14
9 Relationship of ATP to SVS in Return Sludge 15
10 ATP, BOD (settled) and SVS at Various Points in the
Aeration Basin at Baltimore Back River Treatment 16
Plant
11 BOD5 Reduction vs. ATP Concentration 17
12 Chromatogram of ATP Standard 24
13 Instrument Response as Function of ATP Concentration 27
14 Efficiency of Extraction by Variation of Boiling
Time in Tris Buffer 34
15 Apparatus Used for Pure Culture Studies 38
16 ATP, Tyrosine, Oxygen Uptake Rate, Total Count and
Turbidity of a Growing E_. coli Culture 40
17 ATP, Tyrosine, Oxygen Uptake Rate, Total Count and
Turbidity of a Growing "L_. ramigera Culture 41
18 ATP, Tyrosine, Total Microscopic Counts and
Turbidity of a Growing Bacillus sp. Culture 42
vi
-------
FIGURES
No.
19 Changes in Population (count/ml), Dry Weight (g/ml),
Cell Weight (g/cell), RNA/g of Cells and DNA/g of
Cells in an Idealized Bacterial Culture Following
Inoculation with Cells in Lag Phase (Mallette 1970)
(10) 46
20 Growth of Aerobacter aerogenes in Continuous Culture
(Herbert 1961) (7) 47
21 Diagram of Identical Pilot Plants Which Were Used
for Test and Control Experimentation 52
22 Concentration of ATP (yg/ml) in the Mixed Liquor
of the First Aeration Cylinder 55
23 Oxygen Uptake Rate of Mixed Liquor in the First
Aeration Cylinder 55
24 Suspended Solids of Mixed Liquor as Determined by
Optical Density in the First Aeration Cylinder 56
25 Suspended Solids of Mixed Liquor as Determined by
Dry Weight (105°C) in the First Aeration Cylinder 57
26 Tyrosine Content of Mixed Liquor in the First
Aeration Cylinder 58
27 Efficiency of Treatment Expressed as Percent
Reduction in Biochemical Oxygen Demand (BODr) and
Total Organic Carbon (TOC) 59
28 Schedule of Return Sludge and Excess Waste Pumping 60
29 Simplified Diagram of the Activated Sludge Portion
of the Baltimore Back River Sewage Treatment Plant 62
30 ATP Levels in Mixed Liquor (2/10/71 - 2/11/71) at
the Baltimore Back River Activated Sludge Plant
Prior to ATP Regulation 64
31 Fluctuation in the Concentration of TOC, Ortho-
phosphate (PO^-P) and Total Phosphate (TP) of
Primary Effluent with Daily Variation in Raw Waste
Flow at the Baltimore Back River Sewage Treatment
Plant 66
vii
-------
FIGURES
32 Oxygen Uptake Rate in Mixed Liquor and Return
Sludge (2/10/71 - 2/11/71) at the Baltimore Back
River Activated Sludge Plant Prior to ATP
Regulation 67
33 Profile of 02 Uptake Rate Throughout the Aeration
Basin at the Baltimore Back River Sewage Treatment
Activated Sludge Plant on 2/12/71 Prior to ATP
Regulation 68
34 Profile of 02 Uptake Rate and Dissolved Oxygen (DO)
in the Aeration Basin at the Baltimore Back River
Sewage Treatment Activated Sludge Plant During ATP
Regulation Period 3/10/71 69
35 Profile of 02 Uptake Rate and Soluble TOC in the
Aeration Basin at the Baltimore Back River Sewage
Treatment Activated Sludge Plant on 2/12/71 Prior
to ATP Regulation 70
36 Parameters Used for Control of the Baltimore Back
River Activated Sludge Plant from 2/8/71 to 3/11/71 71
37 Parameters of Plant Effectiveness at Baltimore Back
River Activated Sludge Plant from 2/8/71 to 3/11/71 73
38 Phosphate Removal During ATP Control On Activated
Sludge 74
39 Correlation Between 02 Uptake Rate/mg MLSS at the
Head of the Aeration Basin and TOC of the Primary
Effluent at the Baltimore Back River Activated
Sludge Plant 75
40 ATP in Mixed Liquor and Return Sludge at the
Baltimore Back River Activated Sludge Plant During
the ATP Regulation Period 3/10/71 - 3/11/71) 77
41 Simplified Diagram of the Activated Sludge Portion
of the Arlington County Secondary Treatment Plant 82
42 Daily Fluctuations at the Arlington Sewage Treatment
Plant 83
viii
-------
FIGURES
No. Page
43 Diagram of Aeration Tanks Showing Sampling Points
(Circled Numbers) Initially Used for Study 84
44 Concentration of ATP in the Aeration Tank at the
Arlington Sewage Treatment Plant on Several Days 85
45 Oxygen Uptake. Rate in the Aeration Tank at the
Arlington Sewage Treatment Plant 86
46 Diagram of Aeration Tanks Showing Sampling Points
Used for Study After 3/29/71 87
47 ATP Concentration of Mixed Liquor From the Aeration
Tank and Flow Rate of Primary Effluent at the
Arlington Secondary Treatment Plant 88
48 Oxygen Uptake Rate of Mixed Liquor From the
Aeration Tank and Flow Rate of Primary Effluent at
Arlington Secondary Treatment Plant 89
49 Concentration of ATP Found in the Aeration Tank of
the Arlington Sewage Treatment Plant Between 3/18/71
and 4/23/71 91
50 Suspended Solids Found in the Aeration Tank of the
Arlington Sewage Treatment Plant Between 3/18/71 and
4/23/71 92
51 Sludge Density Index (SDI) and 02 Uptake Rates
Measured in the Aeration Tank of the Arlington
Sewage Treatment Plant 94
52 Ammonia Nitrogen (NH3~N), Nitrite Nitrogen (N02~N)
and Nitrate Nitrogen (N03~N) Measured in the Primary
and Secondary Effluent at the Arlington Treatment
Plant 95
53 BOD, TOG and SS of Primary and Secondary Effluent
(Composite Samples) at the Arlington Sewage
Treatment Plant 96
IX
-------
TABLES
No. Page
1 Comparison of BOD Concentration, ATP Concentration,
!4C02 Evolution Rate and 02 Uptake Rate at Various
Intervals During the Aeration of Mixed Liquor from
the Baltimore and Washington, D. C. Sewage Treatment
Plants 18
2 Analysis of ATP Standard Solutions 25
3 Replicate Assays of ATP Solutions 26
4 Stability of ATP Solutions - Four-Day Storage 28
5 Comparison of Assay Values With One-Day and 29
Four-Day Old Enzyme Solutions
6 Effects of Sample Size on Boiling Tris Buffer
Extraction 30
7 Effect of Sludge Concentration on the Efficiency
of Boiling Tris Buffer Extraction 31
8 Comparison of Tris Buffer and Perchloric
Acid Extractions 33
9 Comparison of Tris Buffer and Trichloroacetic
Acid Extractions 33
10 Effect of Blending on the Extraction of ATP from
Sludge 35
11 Recovery of ATP in the Tris Buffer, Perchloric, and
Trichloroacetic Acid Extraction Procedures 35
12 Concentrations of ATP Per Cell, Based on Total Count
and Amount Per ug of Tyrosine Found During the Growth
of the _E. coli Culture 43
13 Concentration of ATP Per Cell, Based on Total Count
and Amount Per yg of Tyrosine Found During the Growth
of the Z_. ramigera Culture 43
14 Concentration of ATP Per Cell, Based on Total Count
and Amount Per ug of Tyrosine Found During the Growth
of the Bacillus sp. Culture 44
-------
No. Page
15 Concentration of ATP Per mg Dry Weight of _E. coli 45
16 Range of Values for ATP/mg Dry Weight Determined by
This and Other Studies 49
17 Composition of Synthetic Sewage (16) 53
18 Tests and Sample Points Employed During the Pilot
Plant Study 54
19 ATP Concentrations in Mixed Liquor and Return Sludge
(2/10/71 to 2/11/71) at the Baltimore Back River
Activated Sludge Plant Prior to Onset of ATP
Regulation 65
20 ATP Concentrations in Mixed Liquor and Return Sludge
at Baltimore Back River Activated Sludge Plant During
ATP Regulation Period (3/10/71 - 3/11/71) 78
21 Phosphate Analysis Made at the Arlington County Sewage
Treatment Plant Prior to ATP-Based Control 90
22 Effects of Using ATP Based Control on Full-Scale
Treatment 97
XI
-------
SECTION I
CONCLUSIONS
1. Adenosine trlphosphate can be extracted from sludge and mixed
liquor using boiling Tris buffer, perchloric acid or trichloroacetic
acid; however, the use of boiling Tris buffer had advantages over
the other methods.
2. Methods based upon the firefly bioluminescent reaction were
developed for the rapid quantitative assay of solutions containing
ATP.
3. Pilot plant studies were conducted to test the use of ATP for
control of return sludge rates. A concentration of 2 yg/ml ATP in
the first aeration tank was found to be most effective for treat-
ment of waste of 240 mg/1 BOD5-
4. Preliminary studies using ATP measurements for control of
aeration sewage treatment have been performed in full-scale treat-
ment plants. The assay was performed without difficulty and gave
indications of possible utility.
-------
SECTION II
RECOMMENDATIONS
Studies in both the Baltimore and Arlington plants were designed as
preliminary investigations of the feasibility of operational control
by means of ATP assay of biomass. These initial goals have been
reached. Studies should now be conducted for longer periods of time
throughout seasonal changes. If possible, the plants under study
should be divided into test and control sections.
In addition to the influence of weather and climate conditions, in-
plant ATP measurements should be made during bulking, shock loading,
severe hydraulic overloading, introduction of toxic materials, heavy
silting, and foam formation.
Pilot plant studies should be continued to support full-scale plant
operations when these types of problems arise. Many of the opera-
tional stresses mentioned above can be duplicated in the pilot plant.
Information gained by pilot plant operation can be rapidly converted
into useful action in a full-scale plant.
Further studies to determine the optimum concentration of ATP should
be conducted. The influence of sewage strength, temperature, contact
time, and sludge activity should all be determined and integrated
into a workable operational procedure.
A manual for the plant operator should be prepared providing informa-
tion on methods and procedures to be followed in actual plant
operation. This should apply to several types of activated sludge
plants and contain precise information to permit effective control
under all types of foreseeable conditions.
-------
SECTION III
INTRODUCTION
The operation of an activated sludge sewage treatment plant depends
upon microorganisms in the return sludge. The basic assumption is
that the recycled sludge is composed of living cells which absorb
and metabolize components of the incoming waste. The effectiveness
of the return sludge is related directly to the number and physio-
logical state of these cells. The parameter currently used for
return sludge control is total suspended solids. McKinney (1) points
out that living microorganisms may constitute as little as 25% of the
suspended volatile solids. Patterson, et_ alL. (2) came to a similar
conclusion and state that "a significant portion of the suspended
volatile solids is nonviable organic material not associated with
the oxidative degradation of the substrate."
Since the functional portions of any biological treatment plant are
the living microorganisms which comprise the sludge, a means of
measuring this active fraction is highly desirable. Conventional
methods of microbial enumeration are difficult to apply because of
the mixed and clumped nature of the flora present, and the time
required for culturing techniques. Conventional plating methods
do not produce results for at least one day - too late to be of use
for control of the plant.
Adenosine triphosphate (ATP) is universally present in living
microorganisms and its measurement by the firefly bioluminescent
reaction is rapid. This study was therefore undertaken to determine
the feasibility of using the measurement of ATP in sludge as a
workable parameter for control of aeration sewage treatment. The
task involved the fabrication of an instrument suitable for measure-
ment of light emitted in the firefly bioluminescent reaction,
establishment of suitable controls and standards, selection of an
extraction procedure, establishment of a methodology which would yield
useable levels of precision and accuracy, correlation of ATP measure-
ments with other parameters of cellular activity and plant operation,
and a preliminary test of the basic principle in full-scale plant
operations.
-------
SECTION IV
PRELIMINARY STUDIES
The application of the firefly bioluminescent assay for ATP to the
control of operation of sewage treatment plants was begun by
Biospherics Incorporated in July 1968. At that time there was no
assurance that a commercial bioluminescence instrument would become
available soon enough for us. Accordingly, a portion of the contract
called for the design and fabrication of an ATP instrument. The ATP
assay instrument fabricated under the contract is shown in Figure 1.
It integrates a photomultiplier light sensing device with a bio-
chemical reaction chemaber as a composite structure. In the detection
of very low levels of light, it is necessary to place the light source
as close as possible to the detector, and ambient light from an
external source must be excluded from contact with the sensor.
FIGURE 1
ATP Assay Instrument
7
-------
Figures 2 and 3 provide details on the design of the reaction drum.
The drum, with a milled recess for the test cuvette, rotates from a.
front loading position to that directly in front of the sensor window.
Optimum coupling is achieved, and extraneous light eliminated by the
long, blackened path around the drum. Injection is through a replace-
able rubber seal directly over the cuvette. Figure 4 shows a block
diagram of the complete instrument.
Procedures for the quantitative extraction of the ATP from biological
materials were examined. Two extractants, butanol and dimethylsulf-
oxide (DMSO) were used under various conditions. While difficulty
in achieving quantitative extraction was encountered with all
extractants, a procedure employing 100% butanol as the extractant,
coupled with sonication for ten minutes was used.
Opening for Loading
4. 0
Photomultiplie r
Opening - 2.0"
6mm Cuvette
Holder
Rotating
Drum
Ball Bearing Pivot
FIGURE 2
Cuvette Turette
-------
Handle
Mirror Surface R 0.75
Drum Cleararu-e .005 .010'
FIGURE 3
Photomultiplier Mount and Reaction Chamber
9
-------
High Voltage
Regulator
24 V
High Voltage
Power Supply
Photomultiplier
(Centronic vmp13/44kb)
Low Voltage
Power Supply
Operational
Amp!i fier
(Philbrick;
SPA-2
Low Voltage
Supply/Regulator
±15 V
24 V
117 VAC
X-Y
Recorder
(Houston
Omni-
graphic)
FIGURE 4
Block Diagram of Electronics Circuitry
This extraction technique was used to determine the quantitative
relationship between ATP and numbers of cells. Measurement of an
unidentified pure culture isolated from mixed liquor, as shown in
Figure 5, verified the linear relationship between ATP and cell
numbers as determined by bacterial plate counts. This, and several
similar experiments, also showed that ATP measurements could be
performed on pure bacterial cultures with an average coefficient
of variation of 10%
The extraction of mixed liquor from the District of Columbia Sewage
Treatment Plant was also investigated to determine if a linear
relationship between measured ATP and concentration of mixed liquor
could be obtained. Representative results are shown in Figure 6.
10
-------
M
c
o
etf
ti
4J
§
U
C
o
0.7
0.6
0.5
0.4
0.3
0.1
V
Butanol sonication extraction of
a diluted cell suspension.
Individual points are the average
of three replicates.
1 x 108 2 x 108 3 x 108
Bacterial Plate Counts (cells/ml)
4 x 10
8
FIGURE 5
Relationship of Measured ATP Concentration
to Bacterial Cell Concentrations of Pure Cultures
The precision in performing the ATP assays on mixed liquor was
achieved with a coefficient of variation of approximately 10%.
Measurements of ATP and bacterial plate counts of mixed liquor
were also compared. As shown in Figure 7, a linear relationship
with some scatter was found.
11
-------
0.003
0.002
e
00
PU
H
0.001
Butanol sonication extraction of
diluted mixed liquor. Individual
points are the average of three
replicates.
0.20 0.40 0.60 0.80
Percent Mixed Liquor
1.00
FIGURE 6
Relationship of Measured ATP Concentration
to Mixed Liquor Concentrations
12
-------
20 r
X
a:
16
12
U
W
U
Butanol sonication extraction
Range and average of three
replicates are shown.
0.2 0.4
0.6
0.8
ATP Concentration (pg/ml)
FIGURE 7
Relationship Between Measured ATP
Concentration and Number of Bacteria
in Mixed Liquor
The question concerning the presence of extracellular ATP (that not
contained in microorganism which could enter mixed liquor with
primary effluent or result from death and lysis of sludge micro-
organisms) was raised. A series of experiments was, therefore,
conducted on mixed liquor and mixed liquor filtrate to determine
the percentage of ATP which might be expected to occur outside
viable organisms. Figure 8 shows the results of a typical experi-
ment which was conducted. Extracellular concentrations of ATP were
found to be relatively insignificant in proportion to the levels
present in the particulate fraction of sludge.
13
-------
Intrjcellular
- 0.1
P.
H
0.01
0.001
Butanol sonication extraction
of diluted sludge.
Extracellular
20
40 60
Percent Sludge Mixture
100
FIGURE 8
Concentration of Intracellular and Extracellular ATP
Correlation between ATP concentration and suspended solids in return
sludge was investigated in the activated sludge component of the
Baltimore Back River Sewage Treatment Plant. Measurements of ATP,
SVS and SS levels in return sludge were made on separate days. As
shown in Figure 9, there was no correlation between ATP and SVS.
However, many of the changes in ATP level could be explained in
terms of operational changes made at the plant.
A number of analyses were also performed on samples which were
withdrawn from various stages in the aeration basin. Figure 10 shows
the typical correlation between SVS, ATP, and BOD which were found.
As shown, an increase in ATP throughout the aeration basin appeared
to correlate with decreases in BOD. The SVS and SS (not shown)
remained relatively constant.
14
-------
00
3.
12
10
Samples were taken on different
days at the Baltimore Back River
Activated Sludge Plant.
2,000
4,000
SVS
6,000
8,000
FIGURE 9
Relationship of ATP to SVS in Return Sludge
15
-------
3.0
2.0
1.0
ATP
T 200
SVS
BOD
_i_
100
Head
of
Basin
1/6
1/3
1/2
2/3 5/6
Effluent
from
Basin
FIGURE 10
ATI, BOD (settled) and SVS at Various
Points in the Aeration Basin at Baltimore
Back River Treatment Plant
A series of flask studies were performed to determine the effect of
ATP concentration on BOD removal. Utilizing return sludge and
primary effluent from the Back River Plant in Baltimore, nine mixed
liquors were prepared for each of a series of laboratory experiments.
Return sludge for the nine mixed liquors ranged from zero percent to
40%, in increments of 5%, as measured by volume. The initial ATP
concentration of each mixed liquor was prepared by calculation based
on ATP measurements of the return sludge and primary effluent.
Flasks containing these mixed liquors were then aerated for eight
hours. Typical results are shown in Figure 11. An ATP concentration
of 1-2 pg/ml was found to be the minimum concentration which could
effect 95% BOD reduction during the eight-hour aeration period.
16
-------
X \
Supernatant BOD (mg/1
80
70
60
50
40
30
20
10
Eight-hour flask
aeration with
various ratios of sludge
and primary effluent
Initial Values
o
I Return Sludge
\ ATP = 11.56 mg/1
\ SS = 9150 mg/1
I Primary Effluent
\ ATP = 0.06 mg/1
\ SS 210 mg/1
- \ BOD5 = 295 mg/1
\ P04-P = 8.15 mg/1
\x 95% Removed
. ,. ^-» Q~_^ ^Q ... _ ...__ __. __
1 1 1 1 1
1.0 2.0 3.0 4.0
Initial ATP Concentration (yg/ml)
5.0
FIGURE 11
BOD5 Reduction vs. ATP Concentration
Increased initial ATP concentration did not significantly increase
BOD reduction. However, no experiments with Baltimore sludge were
performed using shorter aeration periods to determine whether the
higher ATP concentrations might shorten the time required for 95%
BOD removal.
Similar experiments were run on sludge and primary effluent from a
Washington, D. C. plant. A two and one-half hour aeration period
was used. Concentrations of ATP (below 1 ug/ml) appeared to be
most effective, again indicating that heavier than needed (on the
basis of BOD5 reduction) MLSS were being carried at the plant.
Flask studies were undertaken to measure the metabolic activity of
mixed liquor in relation to the measured biomass level. Mixed
17
-------
liquor from the Baltimore Back River Treatment Plant was compared
with mixed liquor from the Washington, D. C. plant. Oxygen uptake
rate and evolution of -^CC^ from media containing -^C-glucose by
sludge were the methods of determining metabolic activity. As
shown in Table I, the ATP concentration of both systems was similar;
however, the Washington sludge was far more active than the Baltimore
sludge when placed in either Washington or Baltimore primary effluent,
The difference was thought to be caused by the large BOD concentra-
tion introduced by the Washington sludge as the result of digester
supernatant being introduced into the return sludge line.
TABLE 1
14,
Comparison of BOD Concentration, ATP Concentration, -L4C02 Evolution
Rate and 02 Uptake Rate at Various Intervals During the Aeration
of Mixed liquor from the Baltimore and Washington, D. C.
Sewage Treatment Plants
Primary Effluent
Return Sludge
Aero Time Aeration
0.5 Hrs. Aeration
1.0
2.0
3.0
4.0
5.0
Final Effluent
WASHINGTON
02
Uptake
Rate
BOD5 ppm/
mg/1 min.
209
10.1
323
1.90
1.90
1.37
1.70
1.38
1.11
108
14C02
Evolu-
tion
CPM
-
5,762
800
-
547
1,583
1,057
1,063
-
-
ATP
mg/1
-
7.75
1.94
2.29
2.62
2.83
2.79
3.19
2.38
-
BALTIMORE
02
Uptake
Rate
BOD5 ppm/
mg/1 min.
185
2.5
143
0.88
0.89
0.56
0.49
0.37
-
41
I4c()2
Evolu-
tion
CPM
-
1,652
200
-
156
548
466
393
-
-
ATP
mg/1
-
7.74
2.00
2.29
2.61
2.99
2.79
2.99
2.43
-
18
-------
SECTION V
TECHNIQUE IMPROVEMENT
The ATP assay procedure was further developed to provide greater
accuracy and precision. The earlier technique had been to dispense
0.1 ml of enzyme solution into a cuvette, place the cuvette in the
bioluminescence instrument and inject 0.01 ml of ATP solution into
the assay cuvette. The ATP solution was quantitatively measured
with a 0.05 ml Hamilton syringe. A rubber septum in the top of
the instrument was used as the injector port.
A careful step by step analysis of this procedure revealed several
sources of error which in toto reduced both the accuracy and precision
of the assay.
ATP was found to adhere to the inside walls of the Hamilton syringe.
A "memory" of prior solutions was thus carried over to subsequent
solutions. Even with extensive rinsing procedures using several
different wash solutions, contamination with ATP could not be over-
come. The problem was most critical when a solution containing a low
level of ATP was assayed after assay of a solution which had rela-
tively high levels of ATP- The injection of the ATP solution was
also critical. A rapid, smooth injection directly into the reaction
mixture was necessary; however, the injection itself was not visible;
therefore, one could not be sure that the entire sample had quanti-
tatively entered the reaction mixture. Aerosol formation and splash
during injection could easily result' in an undetermined amount of
sample not reaching the bottom of the cuvette. Careful scrutiny of
the injection needle after injection, but prior to withdrawal from
the injection port, did reveal the presence of minute droplets and
a liquid film. It was subsequently found that the septum wiped
the needle during withdrawal, and continuous injections caused a
large droplet to form on the bottom of the septum. This meant that
the needle could easily become contaminated during insertion.
Several changes in the procedure were made to minimize these diffi-
culties. The injection port was changed in such a way as to eliminate
the septum. An adaptor containing 0-ring seals was fixed to the
barrel of the injection syringe. This adapted syringe therefore
formed the light seal when fitted into the recess in the top of
the instrument. The injection procedure was also reversed. Instead
of ATP solution being injected into reaction mixture, the reaction
mixture was injected into the ATP solution. This eliminated the
necessity of rinsing the injection syringe since that syringe was
used only for reaction mixture and therefore did not become con-
taminated. The volume of reaction mixture was also found to be
noncritical, therefore, a precise quantitative delivery into the
19
-------
cuvette was not necessary. Losses due to spray or splash were
essentially not important. It was also determined that a plateau
in response occurred with an injection of 0.04 ml or more of reaction
mixture. A 0.05 ml injection volume was therefore adopted. This
procedural change represented a saving since costs for the reagents
were greatly reduced.
Solutions containing ATP were placed in cuvettes by the following
procedure: 26 gauge teflon spaghetti tubing was cut into sections
approximately 10 cm in length. A section of this tubing was then
placed over the tip of a 50 yl Hamilton syringe, and a 10 yl sample
drawn up into the tubing and measured according to calibrations on
the syringe. However, no sample was allowed to come in contact with
the syringe needle or body. Excess sample was tapped or touched-off
from the tip of the tubing and the measured quantity extruded slowly
and carefully into the bottom of the cuvette. Since the aqueous
sample does not wet the teflon, the entire sample is quantitatively
removed from the tubing with no remaining droplet. Placement of the
entire 0.01 ml sample in the bottom of the cuvette may be easily
observed. The teflon tubing is disposable and was discarded after
each sample; hence, the problem of carryover was completely eliminated.
In attempting to simplify the extraction procedure, a number of
extractants were tried, including dimethylsulfoxide (DMSO), boiling
arsenate buffer, boiling Tris buffer, perchloric and trichloroacetic
acids.
A number of experiments were run using DMSO at concentrations varying
from 20 to 90% on sludge to extract ATP and also on standard ATP
solutions. Conditions included the use of hot DMSO solutions and
solutions chilled in an ice bath and sonication. When compared with
the butyl alcohol procedure, the DMSO procedures were all unsatisfac-
tory. While high concentrations of DMSO appeared to extract ATP from
sludge, this result was offset by the fact that similar concentrations
of DMSO inhibited the reaction of the ATP standards.
Knowles and Smith (3) used perchloric acid extraction to measure the
ATP content of Azotobactef vinelandii. This procedure was adapted
for the extraction of ATP 'from sludge. Perchloric acid inhibits the
action of the enzyme luciferase on ATP at the concentrations required
for the extraction of ATP from cells. Consequently, two steps must
be taken to eliminate the inhibition. The first step is to use a
concentrated potassium hydroxide solution to precipitate potassium
perchlorate, which has a low solubility, and then to filter off the
insoluble salt. The second step requires further dilution to remove
completely the inhibitory effect of the remaining perchlorate ions.
The neutralization of perchloric acid generates considerable heat,
and the mixture must be chilled at this step. The subsequent filtra-
tion and dilution provide a solution that gives satisfactory values
for both the sludge and the standards. Objections to this procedure
are the time consuming manipulations involved and the large dilution
factor imposed.
20
-------
Patterson, et al. (2) used boiling Tris buffer to extract ATP from
sludge. Holm-Hansen and Booth (4) had also used this extractant in
the assay of ATP in ocean water. The Tris buffer method has a number
of advantages. The use of a buffer at the same pH as the enzyme
reaction mixture subsequently employed in the assay eliminates
neutralization and filtration. The final dilution can be controlled
to provide an ATP concentration range that is suitable for the ATP
assay. The contact time of the diluted sludge with the boiling Tris
buffer is short, and the whole extraction procedure can be performed
speedily by an operator, or it could be readily automated. The
'dilution factor is also small in comparison with the other procedures.
The use of boiling arsenate buffer for extraction was also investi-
gated since enzyme preparations used during the early portion of the
study (Contract No. 14-12-419) were made up in this buffer (0.02 M,
pH 7.4). Arsenate buffer had also been used for preparation of ATP
standard solutions. During the extraction studies ATP standards
were found to be deactivated by freezing in arsenate buffer, but
at least partially reactivated by boiling.
This unusual finding prompted a study on quantitation of standard
ATP solutions, which is presented in Section VI of this report. As
a result, the use of arsenate buffer was eliminated from all phases
of the extraction and assay procedures.
The butanol sonication extraction procedure, employed the use of
arsenate buffer. After the problems with this buffer were discovered,
attempts were made to substitute Tris buffer. These attempts were
not successful. Since an alternate procedure to butanol sonication
was being sought, emphasis was placed on other extraction techniques
and the use of butanol sonication was abandoned.
The use of trichloroacetic acid in relatively dilute solutions, 5 to
10%, for the extraction of cell components and the precipitation of
protein has been a standard method for many years. The extraction
and precipitation can be done without the use of heat and, because
of the low concentration of extractant, the pH adjustment of this
solution to that required for the ATP assay could be readily accom-
plished by dilution with Tris buffer. The method is simple to use
and can be accomplished rapidly, either manually or by automation.
A series of comparisons were made using Tris buffer or perchloric or
trichloroacetic acids as the extractants of ATP from sludge. The
validity of these methods was also established on ATP standard
solutions of known concentration. The determination of the purity
of the ATP employed, the replication of standard solution assays,
and the extraction of ATP by the three methods are given in detail
in order to establish the validity of the assays subsequently
presented.
21
-------
SECTION VI
QUANTITATION
The ATP used was adenosine-5-triphosphate disodium salt 4.5 1^0 ,
molecular weight 632.2, purchased from Calbiochem, Los Angeles,
California. Two methods were used to determine purity. The first
was based upon the ultraviolet absorption measured at wave length
257 nm. The second was based upon paper chromatrography and visuali
zation of the spot under ultraviolet light.
For the absorption method, 15 mg of the ATP were dissolved in 500 ml
H^O. Absorbance average of five measurements was 761. Molecular
weight was calculated as follows :
In this equation, 15.4 is the molar extinction coefficient as given
in the literature, 30 is the concentration of ATP in mg/1 and 761 is
the observed absorbance. The experimental molecular weight was 607.9,
the calculated molecular weight is 632.2; the sample is, therefore,
96.2% of the theoretical value.
Paper chromatography was done using the following solvent system:
isobutyric acid, ammonium hydroxide, water, ratio 66:1:33. The spot
can be seen only under ultraviolet light. The four different spots
in the chromatograph (Figure 12) represent areas of four different
concentrations circled with a pencil. Under ultraviolet light, only
one migration spot is visible for each concentration. If ADP were
present a spot would have shown on a different site. None was
observed. It is apparent that the purchased sample has a high degree
of purity.
Five separate samples of ATP were weighed and serially diluted to
provide concentrations of 1.0 x 10~3, 1.0 x 10~^, and 1.0 x 10~5 mg/ml.
Each dilution was assayed in quadruplicate. The assay was conducted
on 10 yl of the standard solution contained in disposable cuvettes.
The enzyme reaction mixture, 50 yl, was injected into the ATP solution.
The average of the responses for the three dilutions was 569 mv/yg/ml
with a range of ±2%.
The precision of the four replicates is illustrated by the typical
examples in Table 3. It is apparent that the replication range is not
more than 3% and the average responses over a dilution range of one
hundred fold is of the same magnitude.
23
-------
5 10 15
Added Concentration in Micrograms
FIGURE 12
Chromatogram of ATP Standard
A series of assays on ATP dilutions ranging from 1 x 10 0 yg/ml through
1 x 10~^ yg/ml using 10 yl assay solution was run. Figure 13 is a
logarithmic plot of the mv responses for each sample. A straight
line is achieved between 1.0 x 10~0 yg/ml and 1.0 x 10~4 mg/ml. At
the highest dilution (10-5 yg/ml), there is deviation.
Five separate stock solutions of ATP were made and dilutions of 1.0 x
10~3 mg/ml, 1.0 x 10~^ mg/ml, and 1.0 x 10~5 mg/ml were prepared from
each. These were assayed using freshly prepared enzyme solutions
and the assays were repeated four days later with fresh enzyme solu-
tions prepared from the same batch. The ATP solutions were stored
in the refrigerator during the interval. The results are given in
Table 4.
24
-------
Sample No.
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
TABLE 2
Analysis of ATP Standard Solutions
ATP (me/ml) Millivolts Millivolts*/(ug/ml)
1.0 x 10
1.0 x 10
1.0 x 10
1.0 x 10
1.0 x 10
-3
-3
-3
1.0 x 10
1.
1,
0 x 10
0 x 10
1.0 x 10
1.0 x 10
-4
-4
-4
-4
-4
1.0 x 10
1.0 x 10
1.0 x 10
1.0 x 10
1.0 x 10
-5
-5
-5
-5
-5
557
546
581
554
563
Avg. 560
56.5
58.6
56.4
55.1
56.2
Avg. 57.4
5.76
6.05
5.80
5.76
5.68
Avg. 5.81
557
546
581
554
563
Avg. 560 range ±3%
565
586
564
551
562
Avg. 574 range ±2%
576
605
580
576
568
Avg. 581 range ±4%
Average of all values and standard error 569 ±4%
*Sample Nos. 1-5 indicate separate weighings of ATP standard.
The responses after the four-day storage were slightly higher, 8.4,
6.1, and 0.8%, respectively, for the three dilutions. These small
differences could be due to small differences in the activities of
the freshly prepared enzyme solutions.
Table 2 and Table 4 give the results of the assays on ten separate
weighings of ATP. There were five weighings in each table, and the
preparation of 15 separate dilutions for each group produced a total
of 30 solutions that were assayed in quadruplicate. The standard
error for the means of all values was calculated in Table 2, and the
average and its standard error is 569 ±4%; in Table 4, the proper
comparison is column 1 with an average of 590 ± 20.5%. The effect
of storage is shown in column 2 of Table 4, and the average is 618
± 7.4%. Comparison of these three means by student's t test showed
that there were no significant differences between the means. The
25
-------
TABLE 3
Replicate Assays of ATP Solutions
ATP(mg/l) Millivolts*
1.0 x 10~3 545
554
579
550
Avg. 557 range ±3%
1.0 x 10~3 569
559
538
550
Avg. 554 range ±3%
1.0 x 10~4 57.0
55.8
55.7
56.9
Avg. 56.4 range ±1%
1.0 x 10~4 56.0
56.9
55.0
57.0
Avg. 56.2 range ±2%
1.0 x 10"5 5.79
5.80
5.80
5.82
Avg. 5.80 range ±1%
1.0 x 10"5 5.89
5.64
5.83
5.68
Avg. 5.76 range ±2%
*Four responses are result of four assays of single solution.
26
-------
looo.oo r
100.00
d
o
o.
(O
&
10.00
1.00
.10
.01
10
-5
10
10
-3
10
-2
10
-1
10
-0
Concentration of ATP Solution (pg/ml)
(sample size 10 pi)
FIGURE 13
Instrument Response as Function of ATP Concentration
27
-------
TABLE 4
Stability of ATP Solutions - Four-Day Storage
Sample No,
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
ATP(mg/l)
-3
0 x 10
,0 x 10
,0 x 10
,0 x 10
1.0 x 10
-3
-3
-3
-3
1.0 x 10
1.0 x 10
1.0 x 10
1.0 x 10
1.0 x 10
-4
-4
-4
-4
-4
1.0 x 10
1.0 x 10
1.0 x 10
1.0 x 10
1.0 x 10
-5
-5
-5
-5
-5
Initial Response
Millivolts*
555
540
568
525
480
Avg. 534
553
570
581
556
531
Avg. 558
650
606
767
759
612
Avg. 679
Four-Day Response
Millivolts*
567
576
600
579
571
Avg. 579
588
594
616
584
578
Avg. 592
620
606
787
782
623
Avg. 684
Average of all values and standard
error
590 ±20.5
618 ±7.4
^Response expressed as mv per 1 ug ATP in 1 ml.
value for the difference of 21 between 659 of Table 2 and 590, first
column, Table 4, was 1.0; the t value for the difference of 28
between the means in Table 4 was 1.3; and both values were well below
the t value of 1.7 for significance at the 10% level. It is apparent
that the storage for four days had not significantly affected the
assay results.
In order to determine the stability of enzyme solutions, the standard
solutions shown in Table 2 were assayed using enzyme solutions pre-
pared one and four days previously and stored in the refrigerator.
The results are shown in Table 5.
The assays with the day-old enzyme solution showed a slightly higher
response, but it was well within the variation of the method. The
assays with the four-day-old enzyme solution showed an average
decrease in activity of 13%.
28
-------
TABLE 5
Comparison of Assay Values With One-Day
and Four-Day Old Enzyme Solutions
Millivolts*
Original Value
from Table 1
One-Day-Old
Enzyme
Four-Day-Old
Enzyme
ATP (MR/I)
1.0 x 10~^ 557 594 516
1.0 x 10_ 581 587 503
1.0 x 10 563 575 493
Avg. 567 Avg. 585 +3% Avg. 504 -11%
1.0 x 10~^ 565 581 497
1.0 x W_ 564 588 495
1.0 x 10 562 574 489
Avg. 564 Avg. 581 +3% Avg. 494 -12%
1.0 x 10~^ 576 566 486
1.0 x 10_ 580 599 493
1.0 x 10 568 568 490
Avg. 575 Avg. 578 +.5% Avg. 490 -15%
^Response expressed as mv per 1 yg ATP in 1 ml.
The concentration of ATP in mixed liquor and in the return sludge is
such that the samples must be diluted in order to reach a final
concentration range (after extraction) between 2 x 10~^ and 2 x 10~°
mg/ml, which produces maximum extraction and falls within the range
of linear responses.
Table 6 shows the effect of sample size on the extraction efficiency
of the boiling Tris buffer extraction. A 3 ml sample gave results
that were 25% lower than when a 1.5 ml sample was used. Note also
that the precision of assay was reduced considerably by the larger
sample. An experiment in which sludge was diluted in increasing
amounts prior to extraction with boiling Tris buffer is shown in
Table 7. The higher concentrations of sludge which were extracted
yielded lower results than those which were diluted further prior
to extraction. An average of 4.4 yg/ml was found in dilutions 2 to
10. Assuming this value is correct, the sludge which was not
diluted prior to extraction showed an assay value which was 40% low.
It was decided that sludge with an ATP concentration greater than
1 yg/ml (1 x 10~3 mg/ml) should be diluted prior to extraction.
29
-------
TABLE 6
Effect of Sample Size on Boiling Tris Buffer Extraction
Boiling Tris Buffer Extraction and Assay of Five
1.5 ml Mixed Liquor Aliquots
0.99 yg ATP/ml sludge
1.05 yg ATP/ml sludge
1.00 yg ATP/ml sludge
1.00 yg ATP/ml sludge
1.04 yg ATP/ml sludge
Avg. = 1.02 yg ATP/ml sludge
Coefficient of Variation 2.75%
Boiling Tris Buffer Extraction and Assay of Five
3 ml Mixed Liquor Aliquots
0.78 yg ATP/ml sludge
0.87 yg ATP/ml sludge
0.72 yg ATP/ml sludge
0.70 yg ATP/ml sludge
0.85 yg ATP/ml sludge
Avg. = 0.78 yg ATP/ml sludge
Coefficient of Variation 9.7%
On the other hand, the procedure should not produce a dilution so
great that the multiplication of the response times the dilution
results in a large magnification of the error in precision. The
perchloric acid procedures applied to return sludge produce a high
dilution factor; the trichloroacetic acid and Tris buffer procedures
give lower dilution factors.
The three procedures are described in detail so that the times
required to perform each can be compared and also the dilution
factors.
Perchloric Acid Procedure
Dilute 1 ml of sludge to 20 ml with distilled water. Add 1 ml of
the diluted sludge to 4 ml chilled water in an ice bath, then add
1 ml of 30% perchloric acid solution. Neutralize carefully with
1.3 ml of a solution prepared by mixing 67.5 ml of saturated
potassium hydroxide solution with 180 ml of Tris buffer pH 7.75.
Filter and dilute 2 ml of the filtrate to 50 ml with Tris buffer
pH 7.75. This solution, which represents a dilution of 3650, is
30
-------
TABLE 7
Effect of Sludge Concentration on the Efficiency
Boiling Tris Buffer Extraction
of
Dilution Dilution
Factor Factor
Before Due to
Extraction Extraction
1
2
4
6
8
10
50
50
50
50
50
50
Calculated*
Concentration Concentration
of ATP (ug/ml) of ATP Actually
of Nondiluted Measured
Sludge (ma/ml)
Calculated*
Concentration
of ATP Prior to
Extraction
(rag/ml)
2.59
3.93
4.44
4.05
4.36
4.74
8.8 x 10
4.4 x 10
2.2 x 10
1.5 x 10
1.1 x 10
8.8 x 10
-5
4.4 x 10
-3
-5
-5
-5
-6
2.2 x 10
-3
1.1 x 10
-3
7.5 x 10
-4
5.5 x 10
-4
4.4 x 10
-4
*Based upon average value obtained in dilutions 2 to 10.
assayed. The elapsed time for preparation and assay is approximately
20 minutes.
Tris Buffer Procedure
Dilute 1 ml of sludge to 10 ml with distilled water. To approximately
35 ml of Tris buffer pH 7.75 in a 50-ml volumetric flask that has
reached a temperature of at least 98°C in a boiling water bath add
1 ml of the diluted sludge. Mix thoroughly and immediately chill in
an ice bath, dilute to 50 ml with Tris buffer, filter, and assay.
The dilution factor is 500. The elapsed time for preparation and
assay is approximately ten minutes.
Trichloroacetic Acid Procedure
Dilute 1 ml of sludge to 10 ml with distilled water. To 4 ml of
5% trichloroacetic acid solution, chilled in an ice bath, add 1 ml
of the diluted sludge. Mix thoroughly, then add 1 ml of the mixture
to approximately 35 ml of chilled Tris buffer pH 7.75 in a 50-ml
volumetric flask. Make to volume with Tris buffer and assay.
31
-------
The dilution factor is 2500. The trichloroacetic acid method can be
used at a dilution factor of 1250 by bringing 1 ml of the mixture to
25 ml with Tris buffer. The values reported hence were obtained
with the 2500 dilution factor. The elapsed time for preparation
and assay is approximately five minutes.
Tables 8 and 9 compare results obtained by the three procedures
using the Tris buffer as the standard. As shown in Table 8, the Tris
buffer method gave a higher ATP concentration than the perchloric acid
extraction. The perchloric acid procedure requires more manipulation
than does either the Tris buffer or the trichloroacetic acid procedure,
and the high dilution which results is a source of error that must be
considered. The Tris buffer and trichloroacetic acid procedures,
Table 9, gave close results. Either of these methods will provide
reproducible results.
The effect of boiling time on ATP extraction from sludge in the Tris
buffer procedure was examined. As shown in Figure 14, maximum
extraction occurs very soon after immersion of the sample into
boiling Tris buffer. A gradual decrease in activity resulted as
boiling time was increased. Maximum extraction was achieved within
one minute and this time was selected for the procedure.
The use of a blending technique to aid sampling was investigated.
Table 10 shows that a lower ATP value was always obtained for
blended samples. It is likely that blending resulted in the rupture
of larger cells (protozoa, algae, and fungi) and that the released
ATP was rapidly metabolized by other organisms. Differences in the
percent decrease due to blending may be due to differences in the
populations present. Although precision of sampling was improved
by blending the procedure was considered detrimental; investigation
into the lysis of various microbial forms by blending might provide
useful information.
Recovery experiments of ATP added to sludge presented a problem.
When ATP was added directly to sludge, the ATP began to disappear
immediately as the viable organisms incorporated it into their
metabolic cycle. To stop the action of the organisms in the
instances of perchloric acid and trichloroacetic acid extractions,
the ATP solution was added immediately after the mixture of the diluted
sludge and the acid. In the boiling Tris buffer procedures, the ATP
solution was added immediately after the diluted sludge was intro-
duced into the boiling Tris buffer. These procedures effectively
stopped the action of the viable cells, and the recoveries, as
shown in Table 11, were considered satisfactory. Each value shown
in the table is the average of three individual determinations.
32
-------
TABLE 8
Comparison of Tris Buffer and
Perchloric Acid Extractions
ATP (yg/ml sludge)
Sample
H 1
H 2 duplicate
E 1
E 2 duplicate
M 1
M 2 duplicate
Sample
a
b
Tris Buffer
4.03
4.03
Avg. 4.03
3.71
3.77
Avg. 3.74
3.18
3.82
Avg. 3.50
Perchloric Perchloric/Tris
Acid %
3.04
3.42
Avg. 3.22 80
3.23
3.42
Avg. 3.33 88
2.70
2.66
Avg. 2.68 77
TABLE 9
Comparison of Tris Buffer and
Trichloroacetic Acid Extractions
ATP
Tris Buffer
3.39
3.88
3.20
3.28
3.08
Avg. 3.39
(yg/ml sludge)
Trichloroacetic TCA/
Acid (TCA) Tris %
3.56 100
3.26 84
3.34 104
3.16 97
3.42 111
Avg. 3.35 99.2
33
-------
50
40
30
20
10
4 6
Boiling Time (minutes)
10
FIGURE 14
Efficiency of Extraction by Variation of
Boiling Time in Tris Buffer
Inspection of the results shows that, essentially, the recovery of
ATP added during the extraction procedure is the same for all three
methods. This type of experiment provides no absolute information
relative to the efficiency with which ATP is extracted from the cells.
If one procedure is significantly more effective than another in the
extraction of ATP, then higher results would be obtained by that
procedure. The values by the Tris buffer procedure and by extraction
with trichloroacetic acid are practically identical, Table 9, whereas
the perchloric acid procedure gave extractions ranging from 10 to 25%
less, in the three cases listed in Table 8.
34'
-------
**Sludge Samples
1
TABLE 10
Effect of Blending on the
Extraction of ATP from Sludge
Blended*
2.00
1.98
1.62
1.92
1.31
1.20
0.76
0.80
Non-Blended
2.70
2.85
2.56
2.14
1.41
1.36
1.05
0.67
Decrease Due to
Blending
28%
25%
10%
10%
* Blending was conducted for two minutes using the high speed setting
on a Sears Eight Speed Blender.
**Samples were obtained on separate days from Biospherics' Pilot Plant.
TABLE 11
Recovery of ATP in the Tris Buffer, Perchloric,
and Trichloroacetic Acid Extraction Procedures
Tris Buffer
Perchloric Acid
ATP
in Sludge
(yg/ml)
3.63
2.93
3.05
4.70
4.02
4.30
2.15
2.15
2.30
ATP
Added
(yg/ml)
2.0
2.0
2.0
5.0
5.0
5.0
2.0
2.0
2.0
Total ATP
Measured
(yg/ml)
5.63
4.93
5.05
9.70
9.02
9.30
4.15
4.15
4.30
Added ATP
Recovery
(pg/ml)
2.08
1.78
1.82
Avg. 1.89 (94.5%)
5.75
4.80
5.22
Avg. 5.25 (105%)
1.82
2.03
1.84
Trichloroacetic Acid
Avg. 1.90 (95%)
N.B. These recoveries were performed on different sludge samples.
35
-------
SECTION VII
CORRELATION BETWEEN ATP AND OTHER PARAMETERS
In the experiments on pure cultures, the ATP content of the micro-
organisms was determined and the following parameters were also
measured: the tyrosine content of the cells (to provide an identica-
tion of total mass based upon protein concentration), the oxygen
uptake rate (to measure the metabolic state of the organisms),
turbidity, and microscopic cell counts.
Oxygen uptake rates were determined using a Beckman Model 77 oxygen
analyzer and a polarographic sensor. Samples were placed in a
stirred 125-ml Erlenmeyer flask. The 02 uptake rate was measured
either immediately or after a one-minute aeration period to provide
an initial Q£ concentration. Calibration of the instrument and 02
uptake determinations were performed according to instructions
provided with the instrument.
Tyrosine determination was made according to the Folin-Ciocalteu
method (5). One ml of test solution was placed in a 50-ml volumetric
flask, 2 ml of 5N NaOH were added, and the mixture was heated in a
boiling water bath for five minutes. The flask was then removed
from the bath, 25 ml of 1^0 were added, and the solution was cooled
on ice. Next, 3 ml of phenol reagent were added, and the solution
was brought to the mark with water. After a five-minute color
development, the absorbance was measured in a B & L Spectronic 20
spectrophotometer at 640 nm. Tyrosine concentration was determined
from a standard curve prepared from standard solutions of tyrosine.
In the case of pure cultures, a sample of culture was filtered through
a 2 cm glass fiber filter, washed with 0.25 M of pH 7.75 Tris buffer,
and the entire filter was subjected to treatment as described above.
Microscopic counts were made in a Levy ultra-plane counting chamber,
using a Zeiss phase contrast microscope. Cell suspensions were
diluted 1:10 immediately after withdrawal from the culture flask with
0.1 N KIo. This solution stopped all further cell replication and
imparted a light stain on the cells.
Pure cultures of several typical organisms found in wastewater were
grown in a 2-liter flask equipped as shown in Figure 15. Oxygen
uptake rate was monitored periodically with a Beckman Oxygen Analyzer
connected to a Hewlett-Packard recorder. Both influent and effluent
air passed through a sterile glass wool filter. An air flow rate of
approximately 150 cc/min. was used to maintain a disolved oxygen
level of at least 4 mg/1 in the growing culture. Oxygen uptake
readings were made by stopping the inflowing air and recording the
decrease in oxygen concentration at a known chart speed. Air was
37
-------
Sampling Port
u>
OO
Sensor
Magnetic Stirrer
Air Filter (Exit)
Sterile Air (Entrance)
2 Liter Flasks
Sintered Disc Sparger
FIGURE 15
Apparatus Used for Pure Culture Studies
-------
shut off during these readings for a maximum of four minutes.
Sampling of the culture was performed by clamping the exit air
tube and collecting the sample which was pressure siphoned up and
out of the sampling port. Samples were analyzed immediately after
collection.
Figure 16 illustrates the results obtained upon a culture of JE. coli.
Oxygen uptake rate and tyrosine and ATP contents reached a maximum
at six hours of growth and then declined, whereas turbidity remained
at a high level and total count declined slightly. Similar results,
shown in Figure 17, were obtained with a culture of Z^. ramigera; in
the case of the latter organism, the total count declined rapidly-
Since total microscopic count measures both viable and nonviable
organisms, it is possible that death rate and cellular decay are
greater for _Z_. ramigera than they are for _E. coli. It is difficult
to estimate the total counts for Z_. ramigera because the organism
forms clumps. Assays were also performed on a gram-positive spore
forming rod designated Bacillus sp. A typical run with this organism
is shown in Figure 18. Concentration of ATP per unit volume of
culture reached a maximum before either turbidity or total micro-
scopic counts attained^maximum values, then decreased rapidly.
In all cultures studied, the ATP concentration showed a peak which
preceded stationary growth as determined by turbidity. As the
cultures aged, the ATP concentration fell to a low level. Of all
parameters measured, the ATP levels followed oxygen uptake rates
most consistently.
Tables 12, 13, and 14 show the. concentration of ATP per cell, based
upon the total microscopic count and the ratio of the ATP concentra-
tion to tyrosine (ATP yg:tyrosine yg) . The Bacillus sp. cultures
provided the most reliable ATP/cell data because these large rods
could be counted more accurately than the smaller Zooglea and
12. coli cells * Data from three separate Bacillus sp. cultures
yielded an average value of 7.4 x 10-9 yg ATP/cell with a coefficient
of variation of 15%. Values of the same order of magnitude had been
reported by Chappelle and Levin (6). Several cultures of Sphaerotilus
natans were assayed. The results could not be evaluated in terms of
total counts since the organisms formed trichom.es. The ratio of
ATP/tyrosine for the large cells of Bacillus sp. and of Sphaerotilus
fell within the same range of values that were obtained for the smaller
cells of 12. coli and Z_. ramigera.
Some cultures-1 showed a slightly elevated ATP/cell or ATP/tyrosine
concentration'during logarithmic growth. However, a generalization
could not be made. _Z_. ramigera (Table 13) showed the highest levels
during the lag phase of growth. Some cultures after reaching sta-
tionary growth showed a decrease in ATP/cell or ATP/tyrosine con-
centration, but others failed to show a decrease after more than
39
-------
1.2
20r 1.1-
16
12
0.8 -
0.2
A = O - Total microscopic counts (xlO )
A- Tyrosine (ug/ml)
B = O - Oxygen uptake rate (mg/l/min)
« - Turbidity (optical density)
ATP extracted with perchloric acid (yg/ml)
Nutrient broth, 25°C
FIGURE 16
ATP, Tyrosine, Oxygen Uptake Rate,
Total Count and Turbidity of a Growing _E. coli Culture
40
-------
40 48
Time (hr.)
A
B
Q - Total microscopic bacterial counts (xlO )
A - Tyrosine (yg/ml)
O - Oxygen uptake rate (mg/1/tnin)
O - Turbidity (optical density)
- ATP extracted with perchloric acid (
FIGURE 17
ATP, Tyrosine, Oxygen Uptake Rate,
Total Count and Turbidity of a Growing Z^ ramigera Culture
41
-------
110
100
90
80
70
60
50
40
30
20
10
C D
1.8
1.6
O
11
10
9
8
7
6
5
4
3
2
1
D
1.4
1.2
1.0
. 0.8
0.6
0.4
0.2
2.6
2.4
2.2
2.0
l.E
1.6
1.4
1.2
1.0
0.8
- 0.6
0.4
0.2
-
D -28
A = A - Tyrosine (ug/ml)
B = D - Total microscopic counts (x 10 cell/ml)
C = - ATP extracted with perchloric acid (pg/ml)
D = O Turbidity (optical density)
FIGURE 18
ATP, Tyrosine, Total Microscopic Counts
and Turbidity of a Growing Bacillus sp. Culture
42
-------
TABLE 12
Concentrations of ATP Per Cell, Based on Total Count
and Amount Per yg of Tyrosine Found During
the Growth of the E_. coli Culture
Ratio
Time ATP (yg) to
(hr.) ATP (yg/cell) Tyrosine (yg)
0 3.9 x 10~10 0.09
2 6.3 x 10~10 0.12
4 17.0 x 10~10 0.13
6 6.3 x 10~10 0.10
8 6.3 x 10~10 0.11
TABLE 13
Concentration of ATP Per Cell, Based on Total Count
and Amount Per yg of Tyrosine Found During
the Growth of Z_. ramigera Culture
Incubation Ratio
Time ATP (yg) to
(hr.) ATP (ug/cell) Tyrosine (yg)
5.5 30.8 x 10-10 0.56
25.5 3.1 x 10~10 0.54
29.5 5.0 x KT10 0.45
50.0 0.17
43
-------
TABLE 14
Concentration of ATP Per Cell, Based on Total Count
and Amount Per yg of Tyrosine Found During
the Growth of the Bacillus sp. Culture
Incubation Ratio
Time ATP (yg) to
(hr.) ATP (yg/celll Tyrosine (ug)
3 7.2 x l(f9 0.19
4.5 5.3 x 10"9 0.10
6 9.7 x 10"9 0.32
7 7.6 x 10~9 0.17
21 7.5 x 10~9 0.23
24 7.1 x 10~9 0.18
12 hours of stationary growth. Analysis of all pure culture data
failed to show a definite pattern in maximal or minimal ATP concen-
trations relative to growth phase. Very old cultures would very
likely show decreased ATP levels; however, they were not studied.
Differences in ATP concentration per cell, due either to cell size
or growth conditions may not be important if the ratio of ATP to
dry weight of viable biomass is constant. Attempts to measure the
ATP to viable biomass ratio are difficult.
Viable plate counts do not give cell mass; microscopic counts do not
give viability or cell mass; tyrosine and dry weight determinations
do not distinguish viable cells. The use of a chemostat to achieve
steady state culture conditions would probably provide the best source
of material from which to study the ATP to dry weight ratio. However,
this latter approach was not undertaken in the current study.
A number of dry weight determinations were made on the _E. coli culture.
The relationship of ATP content to dry weight is shown in Table 15.
The range of values, 0.82 to 2.00 yg ATP/mg dry weight, is in agree-
ment with the average value of 2.0 yg ATP/mg dry weight cited by
Patterson, et_ al_. (2).
44
-------
TABLE 15
Concentration of ATP Per mg Dry Weight of E^. coli
Time yg ATP/
(hr.) mg dry wt.
0 1.0
2 .82
3 1.76
5 2.00
6 1.64
8 .85
This study showed apparent differences in the concentration of ATP
per cell of approximately ten fold depending upon the growth phase
of the culture. However, these differences do not necessarily
demonstrate a difference in the ATP/viable biomass ratio. Cell
size, and viability must be considered. Bacterial cultures show a
variation in cell size which occurs during the various growth
phases. Figure 19 shows the generalized variation in cell mass
(g/cell) which occurs during the normal growth curve. Cells are
found to increase in size during logarithmic growth and to decrease
as the culture becomes older. Herbert (7) has found that the average
cell mass of Aerobacter aerogenes may vary by as much as four fold,
depending upon the growth rate (See Figure 20).
On the other hand, as shown in Figure 20, the content of protein per
unit of dry weight varies only slightly over the range of growth
rates studied. The percent tyrosine in proteins from various sources,
as shown below, varies only slightly. The selection of the tyrosine
assay as a determinant of cellular protein was, therefore, made to
provide the same type information as dry weight.
Tyrosine Content
Moles Percent Reference
E. coli 2.1 (8) p. 20
Mycobacteria (11 strains) 2.0 - 2.5 (8) p. 20
Plant Proteins 3.2-6.3 (9) p. 131
Blood Proteins 3.0-6.8 (9) p. 132
Egg Albumin 3.7 (9) p. 132
45
-------
e
too
O
iH
e
O
a
toO
O
60
Q
f\
too
0)
O
bO
Time
FIGURE 19
Changes in Population (count/ml), Dry Weight (g/ml),
Cell Weight (g/cell), RNA/g of Cells and DNA/g of Cells in an
Idealized Bacterial Culture Following Inoculation with Cells
in Lag Phase (Mallette 1970) (10)
46
-------
0.8
60
O
o
c
n)
d)
0.4
0.2
C
H
0)
O
£
I
Q
O
<
80
60
40
20
20r
15
1C.
0.2
0.4 0.6 0.8 1.0
Exponential Growth Rate y(hr )
Content of proteins (O), and nucleic acids
RNA (O) and DNA (A) as % of dry weight and
mean cell mass () (dry weight/ml divided by
total cell count/ml) as a function of growth
rate y. Cultivation mediuffl: glycerol -
NH^ - salts; glycerol is the limiting factor.
FIGURE 20
Growth of Aerobacter aerogenes in Continuous Culture
(Herbert 1961) (7)
47
-------
Tvrosine would be present in dead cell structure, however, unless
the dead cell lysed, it would contribute falsely to a tyrosine based
determination of biomass.
Nonetheless, since the concentration of tyrosine would be greater for
larger cells, the measurement of ATP biomass based upon tyrosine was
expected to be more constant than ATP biomass based upon cell numbers.
Table 12 showed that, while the concentration of ATP/cell varied four
fold during the various growth phases, the concentration of ATP per
unit tyrosine varied only 18%. As shown in Table 13, the ATP/cell
of Z_. ramigera was ten fold greater for 5.5 hours than it was after
22.5 hours. However, the ATP/unit tyrosine showed less than 2%
variation during this time. This latter culture did show a three
fold decrease in 50 hours which was probably indicative of a loss
in viability in which cells remained intact thereby maintaining the
tyrosine level. Table 14 showed three fold differences in concen-
tration of ATP/unit tyrosine; however, these values and those from
all other cultures fell within a range of approximately 0.06 to 0.5 Ug
ATP/pg tyrosine. This means that the variation in ATP/biomass on
the basis of all cultures measured did not exceed ten fold regardless
of genus or physiological state.
It can be assumed that a certain percentage of tyrosine measured at
some growth phases was due to non-viable cells; therefore, the range
in ATP/biomass variation was probably less than ten fold. The range
in ATP/dry weight which was found in this study falls within the
same range of values determined by other workers (see Table 16) .
Work concerning growth measurements and calculated energy yields
supports the theory of a constant ATP pool. It has been found,
Gunsalas and Shuster, that the dry weight of bacterial cells
produced by the utilization of a given substrate is directly pro-
portional to the moles of ATP produced during the metabolism of
that substrate by a known pathway (11) . A remarkably constant value
of 10.5 g dry weight of cells per mole of ATP produced has been
found for many different organisms and substrates. For example, if
Streptococcus faecalis is fed one mole of glucose (which it utilizes
via the Embden-Meyerhof pathway, producing 2 moles of ATP), the yield
in dry weight of cells is 21 g.
This same organism degrades arginine via citrulline for a yield of one
mole of ATP per mole of arginine. The molar growth yield for
arginine grown S_. f aecalis has been found to be 10.5 g dry weight
of cells per mole of arginine utilized. Many organisms including
yeasts have been shown to yield an approximate value of 10.5,
although more recent studies involving continuous culture have shown
a variation in yield which is dependent upon growth rate. Decreased
yields are explained by Tempest in terms of energy of maintenance
requirements (12).
48
-------
TABLE 16
Range of Values for ATP/mg Dry Weight Determined
by This and Other Studies are as Follows:
Organism
Based on Tyrosine -
All Growth Phases
Escherichia coli
Zooglea ramigera
Bacillus sp.
Mixed E. coli &
Bacillus sp.
Based on Dry Weight
yg ATP/mg
Dry Wt. Cells
Reference
cerevxsiae
E.. coli
Saccharomyces
Streptococcus faecalis
Pseudomonas sp.
Mammalian
Cells
0.92
1.1
0.6
1.3
6.0
3.3
This Study
This Study
This Study
0.6 - 1.3 This Study
1.0 Chappelle and Levin (1968) (6)
1.1 Chappelle and Levin (1968) (6)
2 - 12 Patterson (1970) (2)
0.7 - 2.2 Patterson (1970) (2)
3.7 West & Todd (1962) (13)
This evidence, although not concerned directly with an ATP pool as
measured during this study, shows that many different organisms
utilize different substrates and conduct a flow of energy in the
form of ATP from the degradation of substrates to the construction
of cellular structures. If large variations in an ATP pool existed,
a constant relationship between ATP and cell production would not
be expected. Energy in the form of ATP is rapidly utilized in energy
requiring synthesis. No storage of ATP occurs and the means by
which cells uncouple energy production and growth in a non-energy
limited medium have been studied. Gunsalus and Shuster (11) have
discussed three mechanisms of energy dissimilation other than the
production of protoplasm which occur: (1) accumulation of polymeric
products, either in storage form or as unusual waste; (2) dissipation
as heat by ATP use mechanisms, and (3) activation of shunt mechanisms
bypassing energy yielding reactions or requiring a greater expenditure
of energy for priming.
The location and state of transfer of ATP which is extracted from
cells and measured by the bioluminescent reaction is not known.
However, the close coupling of ATP production and utilization argues
for the constancy of its concentration in living cells.
A ten-fold variation in a method of microbial quantitation of mixed
populations is not great. Patterson (2) has calculated that only
49
-------
15% to 20% of the MLVSS at the University of Florida plant was viable
biomass. McKinney (1) has estimated that the ratio may range from
25 - 50%. Lesperance (14) suggests that 80% of the MLVSS in normal
sludge is biological material, but cautions that this figure does not
hold true for certain industrial wastes. These estimates indicate
that the range in percent viable biomass of normal sludge may be in
the neighborhood of five fold. Toxic materials could easily reduce
the viable biomass.
Thus, the variation in biomass determination by the ATP method is
about as great as that of the more conventional techniques. However,
it is quite possible that additional work will narrow the ATP range.
It might be shown, for example, that treatment correlates more closely
with the mass of living cells than with the numbers of cells. ATP
appears more closely tied to mass than cell numbers which relationship,
if further verified, will reduce the range of ATP determination as
applied to sewage treatment. Furthermore, the variations in ATP
associated with various physiological stages of the cell may be
effectively reduced. The extremes of this range are between young
logarithmic cells and very old stationary cells. For the various
respective sampling points in a sewage treatment plant, at known
levels of loading and sludge age, the physiological state of the cells
will be sufficiently well known to reduce the possible range of ATP/
unit biomass. Further, since ATP comparison for treatment control
purposes will be made at respective sampling points at which the
physiological state of the cells would be the same from sampling
time to sampling time (except perhaps for periods of upset detected
by ATP assays at other plants), the measurements should serve adequately.
The possibility that organism differences caused by specific wastes
and geographical locations may give rise to slightly different ATP/
biomass ratios has not been excluded. An empirical determination
of the optimum biomass concentration may be necessary for individual
treatment plants.
In further work aimed at this problem, a prime difficulty exists. This
is the possibility that the ATP assay may be more precise a measure
of biomass than any of the techniques against which the researchers
may attempt to calibrate it. Colony forming units may be the result
of single cells of clumps. Many living cells do not survive to produce
colonies because of culturing deficiencies. Direct microscopic exami-
nations cannot distinguish living from dead cells. Fluorescent stain-
ing suffers from these problems plus one of background discrimination.
DNA, protein, or enzyme determinations vary with cell types and
physiological state.
Jannasch and Jones found discrepancies as high as 13 to 9,700 fold
among standard culturing and enumeration procedures (15) .
50
-------
SECTION VIII
PILOT PLANT STUDIES
The Biospherics Incorporated pilot sewage treatment plant was operated
for a month to test the feasibility of the control of plant operation
through the determination of ATP at various selected points in the
plant (Figure 21).
The plant has a 500-gallon capacity reservoir tank to contain the
synthetic sewage. Table 17 gives the composition of the synthetic
sewage, as recommended by Eberhardt and Nesbitt (16). The mixture
was prepared daily in 360-gallon batches. The synthetic waste flowed
into two identical treatment systems. Each system was composed of
three 15-gallon aeration tanks, connected in series. The flow rate
was 168 gallons per day into each system; retention time in the aera-
tion tanks was 6.5 hoars. The effluent from the third aeration tank
entered a final clarifier, volume 30 gallons, from which the final
effluent was sampled. The conical shape of the final clarifier
permitted the sludge to settle and be returned to the first aeration
tank or be wasted. The entire system was controlled by valves and
pumps, which were activated by timers programmed for a desired return
sludge flow and wasting schedule. Aeration which was at the rate of
3 1/min. produced a level of 4 mg/1 dissolved oxygen in the first
aeration tank and 8 mg/1 in the third aeration tank.
The systems were primed with several gallons of sludge from a local
sewage treatment plant; both systems were run in the same manner for
one week. The following measurements were then made on each system:
ATP, oxygen uptake rate, tyrosine, optical density, suspended solids,
BOD, and TOG. Table 18 lists the tests employed and the sampling
points. The numbers of the sampling points refer to the numbers on
Figure 21.
The BOD of the synthetic sewage was determined daily and had an
average value of 244 mg/1 with a standard deviation of 18, and the
TOG had an average value of 151 mg/1 with a standard deviation of
3.4. The ratio of TOG to BOD had an average value of 1.63, with a
standard deviation of 0.15 and a coefficient of variation of 2.1%.
After one week of operation, when both sections of the pilot plant
showed similar values for the parameters measured, one section was
continued in the same manner and the second section was altered to
provide a test section. Prior to changes in the test section, the
ATP content of the first aeration tank was 4 yg/ml (18 January 1971).
Based upon previous bench studies in this laboratory, it was decided
to reduce the ATP content of the first aeration tank to 2 yg/ml.
Accordingly, on 15 January and again on 18 January, the flow of
51
-------
TEST SECTION
CONTROL SECTION
LEGEND
A » Aeration Cylinders
P = Pump
C Clarifier
S - Solenoid Valve
Excess
Circled Numbers Indicate Sample Points
FIGURE 21
Diagram of Identical Pilot Plants Which Were Used
for Test and Control Experimentation
52
-------
TABLE 17
Composition of Synthetic Sewage (16)
Nutrient Broth 496 g
Urea 49 g
KH2PO, 39 g
add 360 gallons water
BOD 244 ±18 mg/1
TOG 151 ±3.4 mg/1
return sludge in the test system was reduced to achieve a return
flow of one-fourth that of the control system. The ATP content of
the first aeration tank responded to the reduction by declining 50%
to an average value of 2 yg/ml over the seven-day test period,
18 January to 25 January 1971. The BOD 5 reduction averaged almost
95% and did not suffer with respect to the control despite the drastic
reduction in the amount of sludge returned. Figures 22 through 26
provide information on the parameters determined and their relation
to the first aeration tank. These charts were selected from all those
prepared on other sampling points because they represent most clearly
the differences between the test and control systems. Figure 27
shows that the reduction in BOD 5 and TOG, as determined by comparison
of the analyses of the primary effluent and the final effluent.
Figure 28 describes the changes in pumping rates of the returned
sludge for both systems and also the changes in the waste pumping.
Suspended solids during the period of 20 January to 22 January 1971
were decreased by a factor of three in the test system (Figure 24).
However, ATP and Q£ uptake rate (Figures 22 and 23) were decreased
much less, therefore, indicating an increase in viability and/or
activity of the mixed liquor. The tyrosine concentration (Figure 26)
showed that levels averaged 80% less in the test than in the control
system.
The reductions in BOD in the two systems, as shown in Figure 27, were .
similar.
MLSS of the control section were allowed to increase (no wasting as
shown in Figure 28) in an attempt to reach a food/microorganism
ratio of approximately 0.35, which is recommended by Lesperance for
conventional treatment (17) . At these higher MLSS levels sludge
tends to accumulate in the clarifier; therefore, the return sludge
pumping rate was accelerated as shown in Figure 28.
53
-------
TABLE 13
Tests and Sample Points Employed During the Pilot Plant Study
Sample*
Points
Test
1
2
3
4
5
6
7
Control
8
9
10
11
12
13
BOD
X
X
X
TOC
X
X
X
X
X
X
X
X
X
X
X
ATP
X
X
X
X
X
X
X
X
X
X
SS (OD)
X
X
X
X
X
X
X
X
X
X
X
X
0 Uptake
X
X
X
X
X
X
X
X
X
X
Tyros ine
X
X
SS
(Dry Wt.)
X
X
X
"Sample points correspond to those shown in Figure 21.
Over the weekend starting 25 January 1971, a technical difficulty
caused the settled sludge in the final clarifier of the test system
to form a rather solid mass and become anaerobic. The mass broke
apart and was partially pumped to the aeration tanks before the
problem was discovered and the remaining sludge wasted. This effect
was evidenced by an interruption in the pattern of results shown
in Figures 21-27 between 26 January and 28 January 1971. The system
54
-------
O - Control
- Test
Sample Points #3 & 9
^
DO
I
1.5r
1.0-
1/8 1/10 1/12 1/14 1/16 1/18 1/20 1/22 1/24 1/26 1/28 1/30 2/1
Date of Assay
FIGURE 22
Concentration of ATP (p,g/ml) in the Mixed
Liquor of the First Aeration Cylinder
O - Control
- Test
Sample Points (3) &
£ 0.5-
I
I
(-Test Started
,
^o-^oo o
1/8 1/10 1/12 1/14 1/16 1/18 1/20 1/22 1/24 1/26 1/28 1/30 :/l
Date of Assay
FIGURE 23
Oxygen Uptake Rate of Mixed Liquor
in the First Aeration Cylinder
55
-------
corrected itself in approximately three days and was functioning as
before.
The operation of the Biospherics Incorporated pilot sewage treatment
plant was discontinued when it was felt that the feasibility of the
ATP control system had been demonstrated. The remainder of the con-
tract period was devoted to the transfer of knowledge gained from
laboratory conduct and pilot plant studies, to the control of full-
scale municipal sewage treatment plants.
2.0
1.6
at
C
-------
2500
2000
1500
1000
500
O - Control
- Test
Sample Points (3) & (9)
Test Started
1/8 1/10 1/12 1/14 1/16 1/18 1/20 1/22 1/24 1/26 1/28 1/30 2/1
Date of Assay
FIGURE 25
Suspended Solids of Mixed Liquor as Determined
By Dry Weight (105°C) in the First Aeration Cylinder
-------
00
100 r
80
O - Control
- Test
Sampling Points
OO
1/10 1/12 1/14 1/16 1/18 1/20 1/22 1/24 1/26 1/28 1/30 2/1
Date of Assay
FIGURE 26
Tyrosine Content of Mixed Liquor
in the First Aeration Cylinder
-------
Ui
O
8
a
Q
O
ra
a)
ai
100
80
60
Test Started
O - Control
BODs
3» - Test
TOC
G- Control
A- Test
20
1/8 1/10 1/12 1/14 1/16 1/18 1/20 1/22 1/24 1/26 1/28 1/30 2/1
Date of Assay
FIGURE 27
Efficiency of Treatment Expressed as Percent
Reduction in Biochemical Oxygen Demand
(BOD5) and Total Organic Carbon (TOC)
-------
A B
600
500
400
300
2 r 200
i L 100
A Return Sludge (liter/cycle)
Test
Control
B Excess Waste (ml/cycle)
Test
Control
One cycle « ten minutes
l
1/8 1/10 1/12 1/14 1/16 1/18 1/20 1/22 1/24 1/26 1/28 1/30 2/1
Date of \diustment
FIGURE 28
Schedule of Return Sludge and Excess Waste Pumping
-------
SECTION IX
OPERATION OF THE BALTIMORE CITY BACK RIVER
ACTIVATED SLUDGE SEWAGE TREATMENT PLANT
Control by ATP Assay
The Baltimore Back River Treatment Facility maintains and operates a
conventional activated sludge system which was built with a design
capacity of 20 MGD. Although the Plant was at one time divided into
two separate sections to accommodate phosphate removal study (18),
more recent changes have integrated the two sections.
As shown in Figure 29, return sludge is collected from the two clari-
fiers and mixed by continuous pumping from a sump well to an overflow
head basin. Return flow of sludge is metered from this basin to the
heads of the two aeration basins. Mixed liquor, which moves as a
plug, flows through the aeration systems, exits into a common sluice
and goes to the secondary clarifiers.
Primary effluent at the Baltimore Plant is treated by both activated
sludge and trickling filters. Design is such that the activated sludge
receives a constant flow (20 MGD) regardless of raw sewage influent
flow variation. Normal operating procedures at the plant prescribe
a mixed liquor suspended solids level of 1500 to 2000 mg/1, which is
generally achieved with a 25% return sludge rate. Aeration time is
approximately five hours. Two pumps are used for excess sludge
wasting (capacities of 0.40 MGD + 0.96 MGD). The smaller one is
generally operated continuously, whereas the larger pump is used only
when the sludge blanket in the clarifier exceeds the 2 ft. level.
This field study was undertaken to observe the ATP content and Q£
uptake rate of sludge and of mixed liquor and to attempt to correlate
these parameters with other characteristics of plant operation. The
use of these measurements in providing more effective control was
tested by actual operation of the plant, using ATP as the criterion
of the return sludge rate. BOD and TOG reductions were used as
measures of effectiveness.
Approximately one week was spent in observation of the plant, during
which no changes were made in the method of operation. After this
baseline period of operation, sludge return was regulated on the basis
of ATP measurement. Sludge wasting was also geared to maintaining
this parameter. Based on pilot plant studies using waste of similar
BOD strength to that found in Baltimore, a concentration of approxi-
mately 2 yg/ml was selected as the test MLATP concentration. It was
also felt that this concentration would not be an overly drastic
change that could impair treatment for some time until solid levels
61
-------
Aeration ft 2
Aeration // 1
LEGEND
SE - Secondary Effluent
PE - Primary Effluent
RS - Return Sludge Pumping System
Circled letters in aeration basin indicate sampling points.
FIGURE 29
Simplified Diagram of the Activated Sludge
Portion of the Baltimore Back River
Sewage Treatment Plant
-------
were reestablished. Adjustments were made to attempt maintenance
of 2.0 i_ig ATP/ml in the mixed liquor.
The activated sludge system was initially found to be operating
excellently. Information supplied by the City of Baltimore showed
that BOD reduction during the preceding month averaged 97% and the
sludge volume index, which had declined steadily over the past months,
was approximately 100. The MLSS were being held at 1600 to 1700 mg/1.
An experiment was conducted to determine (a) the difference in ATP
concentration that might be expected in return sludge and in the mixed
liquor at the head, middle, and end of the aeration basin, and (b) the
daily ATP fluctuation in mixed liquor and return sludge. Results of
an experiment performed on 10 February 1971 through 11 February 1971,
before ATP control was initiated, are shown in Figure 30 and Table 19.
Of interest was the fact that MLATP levels were relatively stable
throughout the course of a 24-hour period. Measurements of the RSATP,
however, showed a wider variation. The reason for this variation is
not known; however, ATP assays on return sludge from the laboratory
pilot plant showed similar fluctuations. Although the flow of primary
effluent to the plant is constant, the biological loading fluctuates
widely throughout a diurnal cycle. Figure 31 shows the daily fluctua-
tions in TOG and phosphate concentrations which were routinely measured.
These fluctuations correspond quite well to raw waste flow data supplied
by the Back River Wastewater Treatment Plant. The approximately two
hour detention time required for primary treatment is probably respon-
sible for the fact that maxima and minima of PO^-P and TOC of the
primary effluent are slightly out of phase with the flow of raw waste.
In exploring the best location for sampling to control the return
sludge by ATP content, it was decided to use the head of the aera-
tion basin. The ATP fluctuations are dampened at this location with
respect to those observed in the return sludge. Inasmuch as the ATP
content of the raw sewage is relatively insignificant compared to that
in the return sludge, the former can be ignored. Therefore, percent
changes observed in mixed liquor ATP can be compensated by equal
percent changes in the rate of return sludge. This assumes no radi-
cal change in ATP content of the return sludge such as might be caused
by an overdraught on it. However, such changes in sludge composition
will be detected by the ATP-monitoring of the mixed liquor. Accordingly,
sample point (b), Figure 29, was selected for use throughout the control
period.
As shown in Table 19, the MLATP increased by 3.6% in the first half of
the aeration basin and by 7% in the second half. However, Figure 32
shows that the 62 uptake rate was much greater at the head of the
aeration basin than in the middle or end. Profiles of 02 uptake and
63
-------
(a)
12
10
8
4.2
3.8
3.4
3.0
2.6
2.2
1.8
O-
A-
O-
Return Sludge
Aeration Basin sample point
(b) Head
(f) Middle
(i) End
10 12
Time (hr )
2/10/71
Time (hr )
2/11/71
11:15 a.m.
O 2:00 p.m.
A 4:45 p.m.
O 7:30 p.m.
O10:30 p.m.
1:30 a.m.
A 4:30 a.m.
X 7:30 a.m.
2/10/71
2/10/71
2/10/71
2/10/71
2/10/71
2/11/71
2/11/71
2/11/71
RS*
Head
Middle
End
Aeration Basin
*Values were calculated by multiplying measured
RSATP times percent sludge return (.122).
FIGURE 30
ATP Levels in Mixed Liquor (2/10/71 - 2/11/71) at the
Baltimore Back River Activated Sludge Plant
Prior to ATP Regulation
64
-------
TABLE 19
ATP Concentrations in Mixed Liquor and Return Sludge
(2/10/71 to 2/11/71) at the Baltimore Back River
Activated Sludge Plant Prior to Onset of ATP Regulation
Time
2/10/71
11:15 a.m.
:00
2:
4;
7:
p .m.
45 p.m.
30 p.m.
10:30 p.m.
2/11/71
1:30 a.m.
4:30 a.m.
7:30 a.m.
Average ±s. d.
ATP(yg/ml)
Aeration Basin
Head (b)
3.37
2.85
2.66
3.07
2.84
2.68
3.16
3.39
3.00 ±.29
Middle (f)
3.10
3.03
2.90
3.26
2.87
3.09
2.93
3.70
3.11 ±.27
End (i)
2.89
3.00
4.05
3.82
3.24
3.07
3.33
3.39
3.35 +.41
Return Sludge
10.58
9.69
10.78
13.90
8.65
11.40
10.65
6.39
10.25 ±2.30
3.6%
Increase
7%
Increase
dissolved oxygen in the aeration basin are shown in Figures 33 and
34. These data suggest that although a rapid respiration rate occurred
primarily in the first one-sixth of the aeration basin, ATP production
occurred throughout the aeration basin and predominated in the second
half.
From Table 19, the measured average RSATP concentration was 10.25 yg/ml.
Since it is diluted to approximately 23% with primary effluent, the
expected average MLATP concentration would be 2.36 yg/ml. This value
is 21.4% lower than the 3.00 yg/ml which was actually measured;
therefore, an increase in biomass occurred during contact of return
sludge with primary effluent. Since the end MLATP value was 3.35 yg/ml,
the total increase in ATP during an aeration cycle was 0.99 yg/ml or
29.5%.
An experiment was performed to determine the profile of soluble TOC
in the aeration basin. Samples of mixed liquors were collected and
a portion was measured for G£ uptake rate. Results of this experi-
ment which are shown in Figure 34 show that the decrease in soluble
TOC parallels the decrease in D£ uptake rate.
65
-------
B = mg/1 of F
A - F" ow (med) and 1 IX: d"",/l)
A B
ON
ON
240
200
160
120
80
40
-
- 16
12
8
4
X°-O
o' ° \
,v \ /
r / -<( /
V * V
/\ r
/ \ / \y
/ \ / *-
/ ""* ^ "v^^A
^\ / V
V / \
D "' * r
3"*"^ L
ON
°J 00^OVOH
/" "X
* TP (mg/1)
\ A_A
\ i \
\/ 0-°N
*/°i \
/f
PO^-P (mg/1)
« Raw Waste Flow ('
/ ^^ / ^
0 °y 0-° O-o-o-o
° / °x /
\ s° O /
o' TOC (mg/1) \> ^o
*- / ^ ^«-» A-A
\ A o-^ ^-n A\ /D °
\ /° °^D. /
O O
2/25/71I2/26/71 12/27/71
*Uata supplied from treatment plant records.
V2S/71
FIGURE 31
Fluctuation in the Concentration of TOC, Orthophosphate (PO^-P)
and Total Phosphate (TP) of Primary
Effluent with Daily Variation in Raw Waste Flow
At the Baltimore Back River Sewage Treatment Plant
-------
l.Or
c
E
0.6
0.4
Return Sludge
\
12
4 6
Time (hr.)
2/10/71
10
12
Time (hr.)
2/11/71
FIGURE 32
Oxygen Uptake Rate in Mixed Liquor and Return Sludge
(2/10/71 - 2/11/71) at the Baltimore Back River
Activated Sludge Plant Prior to ATP Regulation
-------
l.Or
c
H
e
oo
E
0.8
0.6
to
OS
0.4
C.
:=>
CM
o
0.2
1/6 2/6
3/6
4/6 5/6
6/6
Sampling Points Through
Aeration Basin
FIGURE 33
Profile of 02 Uptake Rate Throughout the Aeration Basin
at the Baltimore Back River Sewage Treatment Activated
Sludge Plant on 2/12/71 Prior to ATP Regulation
As shown in Figure 36, the treatment plant was initially found to
contain a MLATP concentration greater than 3 yg/ml. This was con-
siderably more than the desired concentration. A large backlog of
sludge was also found in the clarifiers and a sludge retention of
approximately four hours was calculated. On February 14, 1971, a
large slug of sludge was wasted. This resulted in a sharp reduction
of the MLSS and MLATP, MLATP concentration was now close to the
2 yg/ml level believed desirable. Adjustments in the rate of return
sludge and wasting which were made during the following weeks, were
made in order to attempt to maintain this 2 yg/ml level. Except
where indicated on Figure 36, the small waste pump was operated
continuously. Rate of aeration throughout the experiment was con-
trolled so that the dissolved oxygen level at the end of the mixed
68
-------
1.0
0.8
0.6
0.4
0.2
A B
r 5
A = A-02 Uptake Rate (mg/l/min.)
B = «-DO Basin // 2 (mg/1)
0-DO Basin // 1
Too rapid for
measurement
1/6
2/6
3/6
4/6
5/6
6/6
Sampling Point Through
Aeration Basin
FIGURE 34
Profile of G£ Uptake Rate and Dissolved Oxygen
(DO) in the Aeration Basin at the Baltimore
Back River Sewage Treatment Activated Sludge
Plant During ATP Regulation Period 3/10/71
liquor basin never fell below 5 ppm. With 23% sludge return rate, the
MLATP level climbed back above the 2 yg/ml level; therefore, beginning
on 18 February 1971, the return sludge rate was decreased in an attempt
to reduce the MLATP to a lower stable level. However, after decreasing
the RS rate to 8%, the MLATP decreased slowly to a concentration of
1.2 g/ml. It appeared that the 8% return rate was too low to maintain
a stable level of MLATP. On 27 February 1971, the return sludge level
was increased to 12.3%. This caused an ATP increase. On 2 March 1971,
the ATP level fell sharply; however, it was observed that the percent
of solids in the sludge was also decreased, and a very low accumulation
of sludge was measured in both clarifiers. It was concluded that the
rate of wasting was in excess of sludge production; therefore, the small
waste pump was shut off for 24 hours. This caused the MLATP level to
69
-------
0.6
0.4
0.2
120
100
80
' 60
I- 40
- 20
LEGEND
A - O- oxygen uptake rate (mg/l/min.)
B - - TOC (mg/1)
1/6
3/6
4/6
5/6 5/6
Sampling Point Through
Aeration Basin
FIGURE 35
Profile of CH Uptake Rate and Soluble TOC in the Aeration Basin
at the Baltimore Back River Sewage Treatment Activated
Sludge Plant on 2/12/71 Prior to ATP Regulation
70
-------
*
s
o
0 / 0 MLATP
<> No-
-------
increase to a value near 2 yg/ml, where it remained fairly stable for
five days. On 8 March 1971, the ATP level again declined; however,
a condition similar to that experience on 2 March 1971 was found and
the waste pump was again shut off for 24 hours. Following this action,
the ATP level rose again.
A comparison of MLATP and MLSS (Figure 36) shows that these two
parameters followed somewhat similar patterns with one significant
difference. Changes in the return sludge rate caused a change in
the MLATP level within 24 hours. However, a change in the MLSS was
not observed until the second day. The decrease in return sludge
rate which was initiated on 18 February 1971 and 19 February 1971
was followed by a MLATP decrease on 20 February 1971, but no decrease
in MLSS until 21 February 1971. On 27 February 1971, the RS rate
was increased and an increasing trend in MLATP was observed on
28 February 1971, but did not occur in the MLSS until a day later.
Prior knowledge of an impending trend in the MLSS level, hours or
days before that effect occurs, would be an important advantage for
the plant operator.
Figure 37 shows the effect of ATP-based control on the efficiency
of waste treatment. The BOD reduction of composite samples collected
and assayed by plant personnel averaged above 95%; TOG reduction
performed every four hours averaged 70%, and suspended solids
reduction calculated from plant records averaged 93% during the
period of experimentation. Thus, the high efficiency of the Baltimore
plant was maintained, despite the fact that considerably less sludge
than normal was returned. The experiment demonstrated that ATP may
be used effectively to control a full-scale treatment plant. Although
the MLSS level was changed drastically, no decrease in plant efficiency
or problems associated with settling occurred. Records showed that
the sludge volume index was lower during the period of experimentation,
including the first week of baseline observation, than for the previous
six months of operation. During the period of ATP-based control, a
dense, rapidly settling sludge was characteristic. The secondary
effluent TOC (SETOC) averaged approximately 35 mg/1 despite broadly
fluctuating concentrations in the primary effluent. The City of
Baltimore phosphate removal study (18) included much data on the
concentration of SETOC; however, the TOC level which they measured
rarely fell below 50 mg/1.
Total and orthophosphate analyses were performed during the last ten
days of the current study; however, good phosphate removal was
observed for only one brief period. As shown in Figure 38, approxi-
mately 80% phosphate removal was observed on 4 March 1971. However,
the efficiency of removal was much poorer at other times. Comparison
of Figure 38 with Figure 36 shows that the occurrence of phosphate
removal corresponded to a peak on the MLSS. It has been demonstrated (18)
72
-------
*PEBOD
(mg/1)
*SE BOD
(mg/1)
*Sludge
Volume Index
PETOC
(mg/1)
SETOC
(mg/1)
*PESS
(mg/1)
*SESS
(mg/1)
260
250
240
230
220
210
200
190
180
20
10
80
150
140
130
120
110
100
90
60
40
20
180
170
160
150
140
130
120
110
100
90
80
20
10
o-o
'"'UW
cr^VufcooO-o
12
16 20
February
24
28
4 8 12
March
*Data supplied from treatment plant records.
FIGURE 37
Parameters of Plant Effectiveness at Baltimore
Back River Activated Sludge Plant from 2/8/71 to 3/11/71
73
-------
10
00 .
e 6
o
c
o
o
2.
O Total Phosphorous SE
P04~Phosphorous SE
D Total Phosphorous PK.
PO^-Phosphorous PE
2 4 h 8 10 12
(March 1971)
Points represent the daily average of assays run
every 4 hours.
FIGURE 38
Phosphate Removal During ATP Control
On Activated Sludge
that effective phosphate removal in the Baltimore plant will occur at
low solids levels, but is seriously interrupted by a prior rapid
decrease in the suspended solids level. Although fluctuations in the
solids level did occur during the period when phosphate analysis
was made, the rate of decrease never exceeded 150 mg/I/day.
Daily oxygen rates were determined on mixed liquor samples at points
a, b, c, and i (Figure 29). The rate of 0 uptake appeared to be
due to both the activity and/or viability of the sludge organisms
and to the concentration of nutrients in the incoming waste. As
shown in Figure 39, variations of 0 uptake rate (sample point a)
correlated most clearly with PETOC. But, it appeared that a sudden
-------
LEGEND
A B
10
160
140
120
100
80
2
A = - PE TOC (mg/1)
B = O - 02 uptake rate of mixed liquor (head)
(mg/l/min.) per MLSS (mg/1)
h
Ol
*-
10 12 14 16 18 20 22 24 26 28
February 1971 |
24 6 8 10 12
March 1971
FIGURE 39
Correlation Between D£ Uptake Rate/mg MLSS at the Head of the
Aeration Basin and TOG of the Primary Effluent at the
Baltimore Back River Activated Sludge Plant
-------
drop in TOC did not result in an immediate 0^ uptake rate decrease; and
after a lag of approximately 24 hours, the 02 uptake rate would exhibit
the effects of the earlier decreased nutrient concentration and fall.
This might mean that cellular activity could remain at a potentially
high level for perhaps only one day after experiencing a sudden drop in
organic loading. This decrease in activity was not reflected by
changes in the ATP level and, therefore, cannot be interpreted as a
loss in viability.
A repeat of the 24-hour MLATP experiment, as presented in Figure 30
and Table 19, was conducted on 10 March 1971 to check the earlier
findings. Results of this experiment, given in Figure 40 and Table 20,
confirmed the earlier findings. A greater average increase in ATP
throughout the mixed liquor may have been observed in this second
experiment. A comparison of data from Table 19 and Table 20 was made
as follows:
From Table 19
2.36 yg/ml
(0.23 x 10.25)
3.00 yg/ml
3.35 yg/ml
From Table 20
1.42 yg/ml
(0.11 x 12.31 yg/ml)
2.31 yg/ml
2.73 yg/ml
Therefore:
0.64 yg/ml 0.89 yg/ml
0.35 yg/ml
0.99 yg/ml
0.42 yg/ml
1.31 yg/ml
Concentration of ATP in return
sludge relative to its dilution
in mixed liquor
Average concentration of MLATP
at the head of the aeration basin
Average concentration of MLATP
at the end of the aeration basin
ATP produced upon contact of
return sludge and primary effluent
ATP produced throughout aeration
basin
Total increase in ATP during one
aeration cycle
The BOD loading at the time of the earlier experiment (before ATP
control) was approximately 5% higher than at the time of the latter
experiment. However, as shown, the production of ATP, i.e., viable
biomass, was 23% greater after the period of ATP based control. This
may mean that after the period of ATP based control, return sludge
was better able to act upon available substrate and convert it into
viable material.
76
-------
(a)
a
(b) 3
14
12
10
8
6
4
2
- Return Sludge
Aeration Basin sample point
O- (b) Head
A- (f) Middle
D- (i) End
8 10 12"
8 10
Time (hr.)
3/10/71
Time (hr.)
3/11/71
3.0
2.6
J 2.2
1.4
2:00 p.m.
6:00 p.m.
ilO:00 p.m.
2:00 a.m.
' 6:00 a.m.
10:00 a.m.
- 3/10/71
- 3/10/71
- 3/10/71
- 3/11/71
- 3/11/71
- 3/11/71
RS*
Head
Middle
End
Aeration Basin
*Values were calculated by multiplying measured RSATP
times percent sludge return (.122).
i FIGURE 40
ATP in Mixed Liquor and Return Sludge at the
Baltimore Back River Activated Sludge Plant
During the ATP Regulation Period (3/10/71 - 3/11/71)
77
-------
TABLE 20
ATP Concentrations in Mixed Liquor and Return Sludge at
Baltimore Back River Activated Sludge Plant, During
ATP Regulation Period (3/10/71 - 3/11/71)
3/10/71
2:00 p.m.
3:30 p.m.
6:00 p.m.
10:00 p.m.
3/11/71
2:00 a.m.
6:00 a.m.
10:00 a.m.
Average ±s. d.
ATP(yg/ml)
Aeration Basin
Head (b)
2.47
2.84
2.23
2.18
2.20
2.64
2.73
2.27
2.27
1.94
1.95
2.23
2.98
2.31 ±.35
Middle (f)
2.77
2.80
2.37
2.47
2.71
2.76
2.95
2.53
2.48
2.17
2.15
2.25
2.43
2.52 ±.24
End (i)
2.82
2.69
2.32
2.72
2.97
2.95
2.95
2.96
3.03
2.62
2.51
2.46
2.49
2.73 ±.25
Return Sludge
12.5
12.6
12.19
12.35
12.00
12.02
14.10
14.10
11.2
10.2
12.3
12.3
12.31 ±.79
8.4%
Increase
7.6%
Increase
N.B. Statistical analysis of the percent deviation between duplicate
assays yielded: Std. Dev. 2.67, Std. Error of the Mean 0.39, and
Coefficient of Variation 2.67.
Also, it might be noted that approximately equal quantities of ATP
were produced in both halves of the aeration basin. In the earlier
run ATP production predominated in the second half. This would mean
that, in the second run, cells were in better condition for biomass
production earlier after the start of aeration.
Although the average values shown in Table 20 were used to compute
the percent increases shown, a statistical comparison of these
averages fails to show significance if only the mean values are
compared. A better comparison is shown in Figure 40b. For each
sampling time, the MLATP was higher at the middle than the head
78
-------
12/13 times. It was also higher at the end than at the middle 11/13
times. The probabilities that the MLATP increased between the head
and middle and middle and end of the aeration basin were, therefore,
0.92 and 0.85, respectively. The probabilities of increase during
the earlier experiment were somewhat less, being 0.75 and 0.5,
respectively.
The possibility that the composition of the incoming waste had changed
sufficiently to provide a more usable substrate during the second
experiment cannot be ruled out. No specific analyses of the primary
effluent were made. Additional experiments of this type are needed
to substantiate these preliminary findings.
Of additional interest in Table 20 is the degree of reproducibility
with which the ATP extraction and assay are routinely performed.
Each set of duplicate numbers represent two separate dilutions,
extractions, and ATP assays of nonblended return sludge or mixed
liquor sample. These data are typical of those obtained during the
one-month study at Baltimore.
79
-------
SECTION X
OPERATION OF THE ARLINGTON COUNTY
ACTIVATED SLUDGE SEWAGE TREATMENT PLANT
The Arlington County Secondary Treatment Facility is a step aeration,
activated sludge plant with a design capability of 24 MGD. As shown
in Figure 41, there are two aeration tanks, each having four passes.
Settled sludge is returned to the A pass (gates s^ and 82) and is
aerated throughout its length. A variable flow of primary effluent
enters at the ends of the first three passes (gates b, c, and d).
Mixed liquor from both aeration tanks (gates e-j_ and &^^ mixes in a
common sluice before being distributed to the three secondary clari-
fiers. Settled sludge moves by gravity to a sump well; a portion of
it is returned to the aeration basin, and the excess is wasted.
Gate heights controlling inflowing primary effluent are adjusted
according to changes in the sludge density index (SDI). In practice,
openings of the gates are regulated to shift the waste load gradually
forward in the aeration tank to improve treatment and then back to
improve settling qualities of the sludge.
Figure 42 shows the diurnal fluctuations in flow rate, waste concen-
tration measured by TOG, and DO level in the aeration tank. Although
there is some question concerning accuracy of meters, these flow data
show a daily average of approximately 22 MG of settled sewage.
A peculiarity encountered in the Arlington plant was the occurrence
of a heavy, dark foam, which, at times, covered the surface of the
aeration tanks and spread onto walkways. It accumulated most heavily
at the head of the aeration tank, but some escaped with the final
effluent, reducing its quality. Microscopic examination showed the
predominance of branching, filamentous organisms.
The BOD of primary effluent at Baltimore had averaged 215 mg/1.
Effective treatment had been maintained when the concentration of
ATP in the mixed liquor at the head of the aeration basin was main-
tained at 2 yg/ml. The BOD of Arlington's primary effluent averaged
approximately one-half the Baltimore value. Therefore, a lower level
of MLATP was expected to be effective. But contact time in the
Baltimore plant approximated five hours in contrast to that in the
Arlington aeration tank, which varied from several hours to approxi-
mately 30 minutes, depending upon sewage flow rate, return sludge
flow rate, and gate openings.
Measurement of ATP levels was made on samples collected at the points
indicated in Figure 43. Essentially, the same parameters were measured
at the Arlington plant as in the Baltimore study. In addition to the
81
-------
Aeration
Tank No. 2
Aeration
Tank No. 1
A pass
C oaas
D pass
D pass
£_
C pass
B pass
A pass
Return Sludge
-*-
PE Gates
PE Gates
-^Primary Effluent
Waste
Sludge
FIGURE 41
Simplified Diagram of the Activated Sludge
Portion of the Arlington County Secondary Treatment Plant
82
-------
LEGEND
oo
(JO
100
80
60
40.
20-
40
30
20
10
A - Primary effluent TOC (mg/1)
B - a --- O Primary effluent flow (MGD)
C - o ....... O Disolved oxygen level in the
D pass of the aeration basin
4/8/71
4/9/71
4/10/71 ' 4/11/71 4/12/71
4/13/71
FIGURE 42
Daily Fluctuations at the Arlington Sewage Treatment Plant
-------
RS
Aeration
Tank Xo. 2
A pass
B pass
C oass
D -ass
Aeration
Tank No. 1
RS
(12) D pass
C pass
B pass
A pass
gates
gates
PE
FIGURE 43
Diagram of Aeration Tanks Showing Sampling
Points (Circled Numbers) Initially Used for Study
determination of ATP, these included: oxygen uptake rate, TOG,
suspended solids (turbidity), phosphate, ammonia, nitrate, nitrite,
and total nitrogen.
When work started on 18 March 1971, a return sludge rate of 16 MGD
was in effect. The mixed liquor suspended solids level was 3350 mg/1
in the A pass and 1000 mg/1 at the end of the D pass. An average
of 82% reduction in BOD and 55% reduction of suspended solids had
been achieved during the first three weeks of March. Determination
of ATP levels were made on samples collected at the points shown in
Figure 43. The results of the determinations of ATP and of oxygen
uptake rates made on several days are shown in Figures 44 and 45.
84
-------
At the Baltimore plant, ATP levels had increased slightly and pro-
gressively throughout the plug flow aeration basin. At Arlington,
ATP levels in each of the passes were expected to reflect the dilutions
which occurred at each point of waste entrance. However, stepwise
decreasing levels of ATP were not found in the aeration basin. As
shown in Figure 44, there was a decrease, but it was not characterized
by sharp drops at the waste entrance points. Very small differences
in ATP levels of the C and D passes were observed. However, the d
gate was set to admit approximately 50% of the total flow. The ATP
values indicated that an increase in concentration occurred con-
comitant with dilution. This might result from poor mixing. The
fact that sample points near the waste inlets showed more scatter
in values than others supports this explanation. Sample points were
selected (Figure 46) that showed minimum scatter and would most
probably provide representative values for the individual passes.
LEGEND
- Aeration No. 1, 3/18; d- Aeration No. 1, 3/19; - Aeration No. 2, 3/19
O- Aeration No. 1, 3/22; A- Aeration No. 1, 3/23; O Aeration No. 1, 3/26-
V- Aeration No. 1, 3/29.
10
8
"I £
4
00
8
A
O
o
Primary Effluent Entrance Points
b gate
6 in.
r i
c gate
6 in.
d gate
10 in.
o
o
0
I
£
2 345 6 789 10
Aeration Tank Sample Points (see Figure 43)
11
i:
FIGURE 44
Concentration of ATP in the Aeration Tank at the Arlington
Sewage Treatment Plant on Several Days
85
-------
0.8 r
0.6
LEGEND
- Aeration No. 1 3/18; D- Aeration No. 1, 3/19;
O- Aeration No. 1 3/22; O- Aeration No. 1, 3/26
Primary Efflupnr F.nt-rnnce Pn-lnfg
b gate
6 in.
i 1
c gate
6 in.
i \
d gate
10 in.
t
ec
B
2 0.4
&
3
u
£ 0.2
Ol
O
0 5
° i *
D g !
.
o
! 8
'
s
90
o
e
8
o
n
,
§ §
.
o
D
34 567 89 10
Aeration Tank Sample Points (see Figure 43)
11
12
FIGURE 45
Oxygen Uptake Rate in the Aeration Tank at the
Arlington Sewage Treatment Plant
Oxygen uptake rate (Figure 45) showed results similar to those for
ATP. Contact of sludge and waste at the three feed points was
expected to produce a spike in the oxygen uptake rate, but instead,
a gradual decrease was observed throughout the aeration basin. The
individual waste gates were not metered. Since the flow rate of the
incoming waste changed constantly, it was not possible to determine
flow rates for each of the three gates. A definite pattern of
dilutions relative to the gate openings at the three feed points
could not be established.
86
-------
Return Sludge
Aeration No. 2
Aeration No. 1
Return Sludge
Mj A pass \
G» J
[^ B pass
(jj) C pass ^
if D pass
_._.c:
V
r
V!
x_x D pass A
C pass
B pass \£/ ^
@ /
A pass
^~
«-
b
c
d
d
c
b
gates
gates
Primary
Effluent
FIGURE 46
Diagram of Aeration Tanks Showing Sampling
Points Used for Study After 3/29/71
An experiment was performed to determine the fluctuations in ATP
concentration and oxygen uptake rate which might be expected in a
24-hour; period during which the waste flow rate varied widely. The
results are shown in Figures 47 and 48. As expected, the ATP con-
centration and the oxygen uptake rate vary with the flow rate.
However, these variations were not great in comparison with the
variations in flow rate.
Total phosphate analyses were run on a series of grab samples. The
phosphate reductions during a week of observations, Table 21, averaged
15%.
87
-------
40
30
LEGEND
A o O Primary Effluent Flow Rate (MGD)
B Mixed Liquor ATP (pg/ml)
- Sample Pt. 1, A pass
O- Sample Pt. 2, B pass
A- Sample Pt. 3, C pass
O- Sample Pt. 4, D pass
-O
o o
20
10
9 13
Time (hr.)
17
21
25
FIGURE 47
ATP Concentration of Mixed Liquor From the
Aeration Tank and Flow Rate of Primary Effluent
At the Arlington Secondary Treatment Plant
88
-------
40
30
20
10
0.6
0.4
0.2
LEGEND
A - e Primary Effluent Flow Rate (MGD)
B - Mixed Liquor 02 Uptake Rate (mg/l/mln.)
- Sample Ft. 1, A pass
O - Sample Ft. 2, B pass
A- Sample Ft. 3, C pass
O - Sample Ft. 4, D pass
I O""" x
9 10 17 21 25
Time (hr.)
FIGURE 48
Oxygen Uptake Rate of Mixed Liquor From
the Aeration Tank and Flow Rate of Primary
Effluent at Arlington Secondary Treatment Plant
89
-------
Primary
Effluent
7.4
7.0
6.9
7.7
7.4
7.3
Secondary
Effluent
7.3
6.1
6.1
5.7
5.6
6.2
TABLE 21
Phosphate Analysis Made at the Arlington County Sewage
Treatment Plant Prior to ATP-Based Control
Total Phosphate, P04~P (mg/1)
Prir
Date*
3/18/71
3/19/71
3/22/71
3/23/71
3/24/71
Average
Reduction = 15%
*Six 2-hour samples were collected between 8:00 a.m. and 6:00 p.m.
on each day. Values expressed are an average of these six assays.
The most critical biomass concentration was believed to exist in the
D pass. The effect of return sludge rate change on ATP measured
biomass level in the D pass was, therefore, used for control. Control
measures reduced the return sludge rate to decrease the BOD-to-ATP
ratio and extend the reaeration period of return sludge in the A pass.
An attempt was also made to shift the waste loading from predominance
in the D pass to a slight predominance in the B pass. These changes
produced a decrease in the sludge density index. In an attempt to
counter the drop in index, changes in gate openings were made starting
on 20 April 1971 to shift the PE load back to the D pass. Figure 49
shows the levels of ATP measured in the four passes of the aeration
tank prior to and during the ATP based control period.
The ATP level in all passes rose during the period of decreased return
rate. This rise was parallel, but not proportional to, changes in
the suspended solids (Figure 50). Return sludge suspended solids
showed, at most, a doubling during the period of decreased return
rate, but ATP levels increased more than three fold. The initial
change in return rate, made on 5 April 1971, first produced a decrease
in the ATP level in the D pass. However, after two days, the ATP
concentration in this pass increased and remained at a level 25%
above that observed prior to 5 April 1971. The suspended solids
90
-------
E
60
0-
H
13.6
12.8
12.0
11.2
10.4
9.6
8.8
8.0
7.2
6.4
5.6
4.8
4.0
3.2
2.4
1.6
LEGEND
- Sample Pt. 1, A pass
O- Sample Pt. 2, B pass
A- Sample Pt. 3, C pass
D- Sample Pt. 4, D pass
.gate
height
in
16.
MGD
Primary Effluent Gates
Return Sludge Flow
7-g .
t.t .
8- 1 .
.4.4. t_
.1-1-1-
8 22 26 30
March
3 7 11 15 19 23
April
, FIGURE 49
Concentration of ATP Found in the Aeration Tank
of the Arlington Sewage Treatment Plant
Between 3/18/71 and 4/23/71
91
-------
LEGEND
.000
^
C 5000
Cf
E
g, 4-000
-»H
1 3000
u
I 2UO°
a.
0)
3
1/5 1000
gate b
height c
in. d
16
12
MGD 8
6
A pass (reaerated nondiluted return sludge)
daily composite sample
0 - Composite Average of All Four ,
Passes A
D-D pass (aeration tank effluent^ /\
daily composite sample) 1\'\ \
/ \ \ \
\«-» I ' \ A'
i v ^* '
A /
/ \A '*-!
+* ^
Ox>
-------
level in D pass did not reflect the increase in ATP. The sharp drop
in ATP in the A pass which occurred on 22 April 1971, was indicative
of cell death and it was most likely a response to the gate changes
which were made in an attempt to halt a decreasing trend in the
sludge density index (SDI). As shown in Figure 51, the SDI fell
sharply after the 12 April 1971 changes and continued to decrease
gradually thereafter. Although the elevated ATP levels were observed
during the period of decreasing SDI, the decrease in settling ability
is considered to have been associated with the points and ratio of
waste entrance into the aeration basin. Prior to the gate changes
made on 12 April 1971, the SDI had remained above 1.0. The oxygen
uptake rate, Figure 51, remained relatively constant throughout the
entire period of testing. The relatively low level of activity, in
comparison to that observed in Baltimore, may be indicative of the
lower BOD concentration of the Arlington waste.
Twenty-four hour composite samples of primary and secondary effluents
were assayed for ammonia, nitrite, and nitrate. The results are
presented in Figure 52. Ammonia was reduced by an average of 27%,
while nitrate and nitrite concentrations increased 5.9 and 7.0 times
between primary and secondary effluents. There was some tendency
for nitrate levels to decrease during the test period, but there
were no indications of major changes in ammonia or nitrite concentrations,
The efficiency of waste treatment in terms of BOD, TOC, and SS removal
during the period 18 March 1971 to 23 April 1971 is shown in Figure 53.
All indices showed improved treatment results during the test period.
However, further studies will be required to specifically determine
the factors which contributed to this improvement.
A summary of the effects of ATP based control at both the Baltimore
and ARlington plants are shown in Table 22. Data obtained during
normal plant operation immediately prior to testing and over a two-
month period prior to testing are shown.
The advantages of the ATP based biomass control are: (1) real time
control which makes possible the measurement of active biomass in
time to adjust the process to the waste being accepted. In this way,
the ATP measurement could provide information for the optimum mixing
ratio of return sludge and primary effluent. (2) better control since
the assay measures the active fraction of sludge or mixed liquor.
Factors which effect cell viability or the degree of inert accumula-
tion in sludge would not affect the control system. (3) rapidness of
response when changes are made in the system. It is possible that
changes in the population structure during the initial stages of
bulking may be observed as a change in biomass. Proper counter
measures to prevent an upset may become feasible with this early
93
-------
LEGEND
1
1
SDI*1
0
0.
- Sample Pt. 1, A pass
O- Sample Pt. 2, B pass
A- Sample Pt. 3, C pass
D - Sample Pt. 4, D pass
-------
LEGKND
40
B -
A - NO -N of PE (mg/1)
D D NO -N of SE (mg/1)
AA NH -N of PE (mg/1)
AANH -N of SE (mg/1)
NO -N of PE (mg/1)
0--0 NO^-N of SE
0.20
0.12
0.04
FIGURE 52
Ammonia Nitrogen (NH3~N), Nitrite Nitrogen (N02-N)
and Nitrate Nitrogen (NO^-N)
Measured in the Primary and Secondary Effluent
At:the Arlington Treatment Plant
95
-------
LEGEND
- Primary Effluent
- Secondary Effluent
00
E
Q
O
o
o
H
-o-o o-
00
E
18
0 T
March
26
1971
30
3 7 11
April
15
1971
19
23
FIGURE 53
BOD, TOC and SS of Primary and Secondary
Effluent (Composite Samples) at the
Arlington Sewage Treatment Plant
96
-------
TABLE 22
Effects of Using ATP Based Control on Full-Scale Treatment
Return Sludge (MGD)
BOD Reduction (%)
TOG Reduction (%)
SS Reduction (%)
Waste Sludge
Produced (Ibs./day)
Sludge Density Index
Baltimore
Prior to
Test
4.5
96 (95)
69
94 (91)
(18,900)
20,300
0.88 (0.45)
During
Test
2.6
95
71
93
18,600
0.89
Arlington
Prior to
Test
16
76 (77)
62
50 (48)
-
1.0
During
Test
8
82
68
66
-
1.2 - 0.6
Values in parentheses are the average of data obtained during prior
two months of operation. Other values obtained prior to test are the
average of the data obtained in approximately two-week period immediately
preceding the reduction in return sludge.
Sludge production at Baltimore rose to 22,300 Ibs./day during the first
week following the reduction in return sludge rate. It then fell to the
average shown during the final two weeks of testing.
warning. (4) Much can be learned about the functioning of a treatment
plant from the assay. The response of organisms to various conditions,
sites of metabolic activity, changes in the percent active fraction
of sludge and the effect of aeration and settling time in various
processes could be observed and serve as important control and engi-
neering design criteria. (5) Additional work may demonstrate that ATP
is the best parameter for measuring the living biomass responsible for
sewage treatment.
97
-------
SECTION XI
SUMMARY
The extraction procedures used included the use of butyl alcohol and
sonication, chilled perchloric acid, boiling Tris buffer, dimethyl-
sulfoxide, and trichloroacetic acid. Several of these procedures,
such as butyl alcohol extraction with sonication and chilled per-
chloric acid, gave satisfactory results; however, they were cumbersome
to perform, time-consuming, and would be difficult to automate (an
important consideration for future development). Dimethylsulfoxide
inhibited the reaction. Tris buffer and trichloroacetic acid extrac-
tion procedures were satisfactory, could be conducted rapidly, and
can readily be automated. The procedures were validated on pure
ATP solutions, pure cultures of several organisms, and return sludge.
A technique for ATP assay which eliminated syringe contamination by
ATP, allowed assays of some sludge and mixed liquor to be performed
with a coefficient of variation of approximately 3%. However,
variations in the sludge did have some effect since some sludges could
not be as reproducibly assayed. Recovery of ATP added to sludge was
found to be approximately 100% provided the sludge organisms were first
killed. Viable sludge attacks exogenous ATP at a very rapid rate.
The efficiency of extraction was found to be maximum between certain
limits of sludge concentration. Dilution of sludge prior to extraction
greatly improved the extraction efficiency. However, the dilution
needed to be held to a minimum since large dilution factors greatly
magnified errors in assay and sampling.
Methods of handling standard ATP solutions and enzyme reaction
mixtures were established. The reagents were found to be stable
and of consistent quality.
ATP assays were conducted on pure cultures of organisms to determine
the constancy of ATP concentration per cell and cell mass during
various phases of growth. Tyrosine content was measured as an
indicator of total protein of both viable and nonviable organisms.
Oxygen uptake rate of the cultures was taken as a measure of the
metabolic activity of the cells. Turbidity measurements, total
cell counts, and dry weights were used as parameters to provide
indicators related to mass. The organisms studied included
Escherichia coli, Zooglea ramigera, Sphaeratilus natans, and Bacillus
sp_.
Concentrations of ATP in some cultures were found to vary by an order
of magnitude on a per cell basis, but concentrations of ATP per yg of
tyrosine showed much less deviation. Bacillus cells which were much
larger than Z_. ramigera and E_. coli, had an average of 7 x 10~" yg
ATP/cell, while the latter organisms had an average of 5 x 10~10 yg
ATP/cell. However, concentration of ATP per yg of tyrosine for all
species tested showed similar ranges of values. The largest differences
99
-------
occurred in very young and very old cultures where cell viability may
be questioned. The data tend to indicate that the ratio of ATP to
viable cell mass is relatively constant.
Growing bacterial cultures exhibited a peak in ATP concentration which
coincided with a peak in 02 utilization. The peak occurred in the
latter portion of the logarithmic growth phase as determined by
turbidity measurements and total microscopic counts. ATP concentra-
tion and 02 uptake rate of the culture decreased rapidly during sta-
tionary growth.
The correlation of these two parameters would seem to be significant,
since they indicate the presence of living organisms undergoing
metabolic respiration - the specific basis of biological waste treat-
ment.
The concept of using the ATP content as a measurement of viable
biomass was used to control the operation of the Biospherics Incorporated
pilot sewage treatment plants. Two completely independent, but identi-
cal pilot plants were employed, one serving as control and the other,
as the test system. They were operated simultaneously on a synthetic
sewage (average BOD 244 mg/1). In addition to the determination of
ATP content, measurements were made of the oxygen uptake rate, optical
density, tyrosine, suspended solids, biochemical oxygen demand (BOD),
and total organic carbon (TOC). It was found that reductions in
sludge recycling did not produce proportionate reductions in mixed
liquor ATP levels. A 75% decrease in return rate was necessary to
effect a 50% decrease in the mixed liquor ATP level. Mixed liquor
of 2 yg ATP/ml in the test section exhibited respiration rates that
were nearly as high as mixed liquor containing 4 yg ATP/ml. Effective-
ness of treatment in terms of BOD and TOC reduction, was enhanced
slightly by the mixed liquor ATP concentration of 2 yg/ml.
The feasibility of using the ATP content of the mixed liquor as a means
of controlling full-scale plant operation was tested at the Baltimore
Back River activated sludge sewage treatment plant. This is a con-
ventional plug flow activated sludge secondary treatment plant with an
influent constant flow of 20 MGD. Average strength of primary effluent
was 240 mg/1 BOD. Control changes were initiated to reduce the ATP
content of the mixed liquor at the head of the aeration basin from
approximately 3 yg/ml to a level of 2 yg of ATP per ml, which had
been effective in pilot studies. To effect this reduction, the
return rate of the sludge was decreased to 12%. The BOD reduction,
as determined by the Baltimore plant staff, averaged more than 95%
during the baseline and test periods. The reduction in TOC also
remained constant with an average reduction of 70%. Changes in
return sludge rate caused a change in the mixed liquor ATP concen-
tration within 24 hours. However, a change in the mixed liquor
100
-------
suspended solids was not observed until the second day. The early
and prompt response of the ATP level to changes in the system is
viewed as an advantage to plant operation.
The changes in oxygen uptake rate appeared to correlate most closely
with the concentration of the nutrients in the primary effluent as
measured by its TOG. However, a sudden drop in TOG did not produce
an immediate decrease in the oxygen uptake rate; a lag of approximately
24 hours was required to notice the effect. This decrease in rate
was not reflected by changes in the ATP level and, therefore, cannot
be interpreted as a loss in viability.
In order to continue testing the feasibility of ATP control on full-
scale plant operations, the methodology was next applied to the step
aeration secondary treatment plant located in Arlington County,
Virginia. At this plant, each aeration tank consists of four passes.
Return sludge alone enters at the head of the first pass and is
aerated throughout the length of the pass. Settled waste enters at
the end of the first, second, and third passes. The volume of waste
entering each pass is under gate control. The BOD of the primary
effluent averaged 116 mg/1 daily, approximately half that of the
Baltimore plant. In contrast to the five-hour contact time in
Baltimore, contact time in the Arlington plant varied from several
hours to as little as 30 minutes.
Measurements of the"ATP content and oxygen uptake rates of the mixed
liquor at a number of points through the four passes were conducted
to determine the points at which samples would be most representative.
During the test period, the ratio of ATP to BOD was started slightly
above that found effective in Baltimore. The rate of return sludge
was reduced gradually from 16 MGD to 8 MGD in four steps.
During the test period, the ATP level rose in all passes. Return
sludge suspended solids concentration doubled, whereas ATP levels
increased three fold. The ATP level in the final pass increased,
and remained at a level 25% above that found in the pretest period
where the suspended solids level of that pass remained constant,
therefore, indicating an increase in sludge viability. The oxygen
uptake rate remained low and relatively constant throughout the entire
period of testing. This low level of activity in comparison to that
found in Baltimore may be indicative of the lower BOD concentration
of the Arlington waste.
During the test period, the reduction of BOD, TOG, and SS showed
improvement by 8, 10 and 10%, respectively. Although the factors
contributing to these improvements were not dete.mined, the effective
reduction of BOD, TOG and SS loading on the receiving stream would
have been reduced by 25, 16, and 17%, respectively.
101
-------
The results of field testing are encouraging, but not conclusive.
The test periods were relatively short and parallel controls could
not be run.
After establishment of procedural techniques in the laboratory, no
technical difficulties with the ATP assay were encountered during
the latter phase of study. The control method appears promising
and is now ready for more extended comparative, full-scale trials.
This first preliminary attempt at full-scale control was concerned
with the most obvious parameters of operation. ATP levels were.
adjusted on the basis of laboratory studies and some trial and error.
The optimum concentration of ATP in mixed liquor would depend upon
several factors. The strength of incoming waste would be of prime
importance. TOC measurements, which can be performed rapidly enough
to be of control significance, may be needed to monitor the influent
stream and so determine the rate at which sludge of a certain ATP
content should be recycled. Also, the recycling of sludge based
upon ATP content of sludge alone, may very well not be sufficient.
The state of microbial activity, in addition to viability, may be
important. Oxygen uptake rates were approached in this program
as a possible measure of this activity.
Another factor of considerable importance is the contact time between
waste and biomass. A greater concentration of viable organisms would
be expected to be necessary when contact time is decreased. It would
also be important to test processes like contact stabilization, to
determine the optimum concentration for maximum absorption.
Temperature and dissolved oxygen level are two additional parameters
which would affect the optimum biomass level. Increased temperature
and/or dissolved oxygen would increase the rate of cellular metabolism,
therefore, within a given time frame the required concentration of
viable cells should be less.
In overview of continuous monitor of TOC, DO, waste flow, temperature
and 02 uptake rate, in addition to ATP measurements would be integrated
into a signal for sludge recycling.
102
-------
SECTION XII
ACKNOWLEDGEMENTS
The cooperation of Arlington County, Virginia, and, in particular,
Mr. Harry Doe, Jr., Director of Utilities, is acknowledged with
sincere thanks. Messrs. Michael Gula, Paul Casimano, James Bailey
and Gene May of the Arlington County Sewage Treatment Facility
provided valuable assistance during the full-scale study at that
plant.
Mr. William A. Hasfurther, Chief, Division of Wastewater of
Baltimore, Maryland, was most helpful in providing cooperation for
the full-scale study at the Baltimore Back River Treatment Plant.
The assistance of Messrs, Jerald D. Wingeart, Terry Armbruster and
Hans Urtes, of that plant, is gratefully acknowledged.
The laboratory research and field studies reported herein were per-
formed by Biospherics Incorporated, Rockville, Maryland. The
research team which directed the project and prepared the report
consisted of Drs. Gilbert V. Levin, Principal Investigator, J.
Rudolph Schrot, Research Microbiologist, and Walter C. Hess, Senior
Biochemist. Mr. George Alvarez, Chemist, and Mrs. Vivian Brooks,
Technician, performed most of the analytical work. Messrs. Donald G.
Shaheen, Senior Chemist, George Topol, Chemist Engineer and Barry
Forster, Technician, assisted in various capacities. Dr. Patricia
Straat reproduced the figures included in this report.
The project was sponsored by the Environmental Protection Agency, with
Dr. Robert L. Bunch serving as Project Officer.
103
-------
SECTION XIII
REFERENCES
1. McKinney, R.E., Microbiology for Sanitary Engineers, McGraw Hill
Book Company, Inc., New York, pp 167-169, (1962).
2. Patterson, J.W., Brezonik, P.L., and Putnam, H.D., "Measurement
and Significance of Adenosine Triphosphate in Activated Sludge,"
Environmental Science & Technology, 4_, p 569, (1970).
3. Knowles, J.C., and Smith, L., "Measurement of ATP Levels of Intact
Azotobacter vinelandii Under Different Conditions," Biochimica et
Biophysca Acta. 197, p 152, (1970).
4. Holm-Hansen, 0., and Booth, C.R., "The Measurement of Adenosine
Triphosphate in the Ocean and its Ecological Significance," Limnology
& Oceanography, 11, p 510, (1966).
5. Folin, 0., and Ciocalteu, V., "Determination of Tyrosine," Journal
Biological Chemistry, 73. p 627, (1929).
6. Chappelle, E.W., and Levin, G.V., "Use of the Firefly Biolumines-
cent Reaction for Rapid Detection and Counting of Bacteria,"
Biochemical Medicine, 2_, p. 41, (1968).
7. Herbert, D., "The Chemical Composition of Microorganisms as a
Function of Their Environment," Microbial Reaction to Environment,
(llth Symp. Soc. Gen. Microbiol., London, April 1961), Cambridge
University Press, Cambridge, p 391.
8. Luria, S.E., "Bacterial Protoplasm - Composition and Organization."
In: The Bacteria, Vol. I, by I.C. Gunsalus and R.Y. Stanier,
Academic Press, (1960).
9. Oser, Bernard L., Hawk's Physiological Chemistry, 14th ed., McGraw-
Hill, (1965).
10. Mallette, M.F., "Evaluation of Growth by Physical and Chemical
Means," In: Methods in Microbiology, Vol. I, Ed. J.R. Norris
and D.W. Ribbons, Academic Press, p 522, (1970).
11. Gunsalus, I.E., and C.W. Shuster, "Energy-Yielding Metabolism in
Bacteria," In: The Bacteria, Vol. II, Ed. I.C. Gunsalus and
Roger V- Stanier, Academic Press, (1961).
12. Tempest, D.N., "Theory of the Chemostat," In: Methods in Micro-
biology, Ed. J.R. Norris and D.W. Ribbons, Academic Press, (1970).
13. West E.S., and Todd, W.R., Textbook of Biochemistry, Third Edition,
Macmillan Col, p 1163, (1962).
105
-------
14. Lesperance, Theodore W., "A Generalized Approach to Activated
Sludge. Part I - Basic Biochemical Reaction," Water Works and
Wastes Engineering, pp 44-46, April (1965).
15. Jannasch, H.W., and G.E. Jones, "Bacterial Populations in Sea
Water as Deterr'.ined by Different Methods of Enumeration,"
Limnology & Oceanography, 4L, pp!28-139, (1959).
16. Eberhardt, W.A., and Nesbitt, J.B., "Chemical Precipitation of
Phosphorus in a High-Rate Activated Sludge System," Journal
Water Pollution Control Federation, 40, p 1239, (1968).
17. Lesperance, Theodore W., "A Generalized Approach to Activated
Sludge - Part 5 Conventional Treatment," Water Works and Wastes
Engineering, pp 31-40, October (1965).
18. Environmental Protection Agency, Program No. 17010DFV, Contract
No. 14-12-471, "Phosphate Study at the Baltimore Back River Waste
Treatment Plant," (1970).
106
-------
SECTION XIV
ABBREVIATIONS USED IN TEXT AND FIGURES
ATP Adenosine Triphosphate
BOD Biochemical Oxygen Demand
DO Dissolved Oxygen
MGD Million Gallons per Day
MLATP Mixed Liquor Adenosine Triphosphate
MLSS Mixed Liquor Suspended Solids
MLVSS Mixed Liquor Volatile Suspended Solids
PE Primary Effluent
PETOC Primary Effluent Total Organic Carbon
RSATP Return Sludge Adenosine Triphosphate
SDI Sludge Density Index
SE Secondary Effluent
SEBOD Secondary Effluent Biochemical Oxygen Demand
SESS Secondary Effluent Suspended Solids
SETOC Secondary Effluent Total Organic Carbon
SS Suspended Solids
SVS Suspended Volatile Solids
TOC Total Organic Carbon
TP Total Phosphate
107
-------
SECTION XV
APPENDIX
Description of Analytical Methods
Solutions Required for ATP Assay
Tris Buffer
6.075 g Tris dissolved in 2 1 of sterile, distilled, deionized ATP-
free water. Adjust to pH 7.75 with hydrochloric acid (equal parts
concentrated HC1 and ATP-free water). It is important that the water
is ATP free. Ordinary distilled water run through a deionizing column
should be autoclaved (two hours) to hydrolyze ATP- The buffer should
be tested for ATP activity. The response of 10 pi of buffer should be
less than 0.1 millivolts and is normally approximately 0.03 millivolts.
After preparation Tris buffer should be refrigerated.
Enzyme Solution
A commercial preparation (Dupont) was employed. The buffer is supplied
in tablet form. One tablet is dissolved in 3.0 ml of ATP-free water.
To this solution is added the entire contents of one vial of enzyme
powder. Allow the resulting solution to stand 15 to 30 minutes at room
temperature before using.
ATP Standard Solutions
Adenosine-5-phosphate, disodium salt, 4.5 t^O molecular weight 632.
62.5 mg equivalent to 50 mg of ATP, are dissolved in 50 ml of Tris
buffer (1 ing/ml concentrated). Serial dilutions using Tris buffer
are made to provide the concentrations desired to prepare a standard
curve. The dilutions should run from 1.0 x 10~2 to 1.0 x 10~6 mg/ml
ATP (10-0.001 ug/ml). Dilutions of 1, 0.1 and 0.01 yg/ml were routinely
run with each assay.
Trichloroacetic Acid
Dissolve 5 g of TCA in 100 ml of ATP-free H20.
Methods of Extraction
Place 35 ml of Tris buffer in a 50-ml volumetric flask. Bring to 100°C
in a boiling water bath and add 1 ml of the diluted solution to be
assayed. Mix rapidly and transfer the flask to an ice bath and chill.
109
-------
After the temperature has been reduced to approximately that of the
ice bath, make to volume with Tris buffer, filter, and assay. In lieu
of filtration, a brief settling period may be used.
TCA Extraction
Place 4.0 ml of the 5% TCA solution in a test tube maintained in an
ice bath, add 1 ml of the diluted solution to be assayed, and mix well.
To 35 ml of Tris buffer in a 50 -ml volumetric flask, add 1 ml of the
TCA mixture, dilute to volume with Tris buffer, and assay. If desired,
the 1 ml of TCA mixture can be diluted to 25 ml with Tris buffer.
Method of Assay
Aliquots (10 yl) of the solution to be assayed are placed in cuvettes.
The transfer is effected by placing a 10 cm length of 26 gauge Teflon
tubing on the tip of a microsyringe. The plastic tubing is replaced
for each solution to be assayed. The cuvette is inserted in the
analyzer and 50 yl of the enzyme solution contained in a microsyringe
are injected into the assay solution. Quantitation is based upon
peak height of the energy output .
Determination of NH
Place 10 ml of the solution to be assayed (filtered primary or secondary
effluent) in a 50-ml volumetric flask, add NH^-free, distilled water to
make approximately 35 ml. The flask, together with a blank solution
similarly treated, is kept in an ice bath. Add, with shaking, 1.0 ml
of Nessler's solution. Wait ten minutes after diluting to volume and
read in a colorimeter at wavelength 520 nm. The blank is set at 100%
transmittance.
Solutions Required
Nessler's Reagent
Dissolve 10 g mercuric iodide and 7 g potassium iodide in approximately
70 ml distilled water. Add the resulting solution, slowly with stirring,
into 15 ml of water containing 10 g sodium hydroxide, dilute to 100 ml.
Nitrogen Standard
Dissolve 0.3819 g of ammonium chloride in 100 ml of distilled water
(1.0 ml of this solution diluted to 100 ml contains 0.01 mg N per ml).
110
-------
Determination of NC>2
Solutions Required
Sulfanilic Acid
Dissolve 0.6 g of sulfanilic acid in 70 ml of hot H^O. Cool. Add
20 ml of concentrated HC1 and dilute to 100 ml with water.
Naphthylamine HC1
Dissolve either 16 g anhydrous sodium acetate or 27.2 g sodium acetate
trihydrate to yield 100 ml of an aqueous solution.
Sodium Nitrite Standard
Dissolve 49.3 mg sodium nitrite to yield 100 ml of an aqueous solution.
This is the stock solution. Dilute 1 ml of the stock solution to 200 ml
with water; 1.0 ml contains 0.5 yg nitrogen.
Procedure
Place 10 ml of the solution to be assayed (filtered primary or secondary
effluent) in a 50-ml volumetric flask. Add 1.0 ml sulfanilic acid solu-
tion and wait three minutes. Add 1.0 ml naphythylamine solution and 1.0 ml
sodium acetate solution. Dilute to volume, wait ten minutes, and read in a
colorimeter at 520 run. A water blank is run through the procedure and set
at 100% transmittance.
Determination of NOo
Solutions Required
Brucine-Sulfanilic Acid
Dissolve 1 g brucine sulfate and 0.1 g sulfanilic acid in approximately
70 ml of hot, distilled water. Add 3 ml concentrated HC1, cool, and
dilute to 100 ml.
Sulfuric Acid Solution
Carefully add 500 ml of a concentrated sulfuric acid to 75 ml of distilled
water. Cool to room temperature before using.
Nitrate Standard
Dissolve 72.18 mg anhydrous potassium nitrate in 100 ml water. This
solution contains 0.1 mg N per ml.
Ill
-------
Procedure
Place 2.0 ml of the solution to be assayed (filtered primary or secondary
effluent) in a 50 ml beaker. Add 1.0 ml of the brucine reagent. Into
a second 50 ml beaker, place 10 ml of the sulfuric acid reagent. Mix
the contents of the two beakers. Allow the treated sample to remain
in the dark for ten minutes. During the interval, add 10 ml distilled
water to the second beaker. After ten minutes, add the water to the
sample and mix. Allow to cool in the dark for approximately 20 minutes
and read in a colorimeter at 410 nm. A water blank run through the
procedure is used to set the instrument at 100% transmittance.
Determination of Total Nitrogen
Solutions Required
Digestion Mixture
To 30 ml phosphoric acid, add 5 ml of a 5% cupric sulfate solution and
10 ml of concentrated sulfuric acid.
Standard Solution
The same standards can be used as described under the Determination
of Ammonia.
Procedure
It should be noted that the digestion method does not reduce nitrate
to ammonia and, therefore, the nitrate-nitrogen values should be added
to the total nitrogen values that are determined by digestion. To
25 ml of the solution to be assayed (filtered primary or secondary
effluent), add 2 ml of the digestion mixture. A 50 ml Kjeldahl flask
can be used. Digestion is continued until the solution becomes color-
less. Then cool in an ice bath, add 35 ml of water and 15 ml of Nessler's
reagent, all contained in a 50-ml volumetric flask. After ten minutes,
dilute to volume and read at 520 nm. Standards and readings are the
same as for ammonia determination.
Tyrosine Determination
Solutions Required
Phenol Reagent
In a 1 1 flask equipped with a reflux condenser, place 50 g sodium
tungstate, 12.5 g sodium molybdate, 350 ml water, 25 ml, 85% phosphoric
acid, and 50 ml concentrated hydrochloric acid. Reflux gently for ten
112
-------
hours. Add 75 g lithium sulfate, 25 ml water, and several drops of
bromine. Boil the mixture for 15 minutes, without the condenser, to
remove excess bromine. Cool, dilute to 500 ml, and filter. Before
use, the reagent should be diluted with an equal volume of water.
Tyrosine Standard
Dissolve 20 mg tyrosine in 100 ml of 0.1 N hydrochloric acid; this is
the stock solution. Dilute 10 ml of the stock solution to 100 ml with
distilled water (1 ml contains 0.02 mg tyrosine/ml).
Micromethod Procedure
To 5 ml of the solution to be assayed (filtered primary or secondary
effluent) add 0.2 ml 5N sodium hydroxide. Place the test tube in a
boiling water bath for 15 minutes. Cool, add 0.2 ml of the phenol
reagent, dilute to 10 ml, and read at 690 nm. A blank is similarly
treated. Standard curve is prepared from 1.0 - 5.0 ml aliquots of the
tyrosine standard solution.
Orthophosphate Determination
Solution Required
Ammonium Molybdate Solution
Dissolve 25 g of ammonium molybdate in 175 ml distilled water. Cautiously
add 280 ml concentrated sulfuric acid to 400 ml distilled water. Cool,
add the molybdate solution, and dilute to 1 1. This is the stock solu-
tion. To 1 ml of the stock solution, add 3 ml 50% sulfuric acid and 36 ml
water. This is the reagent used for the micromethod.
Stannous Chloride Solution
Dissolve 10 mg stannous chloride in 2.5 ml 10% hydrochloric acid; add
22.5 ml distilled water.
Phosphate Standard
Dissolve 0.4393 potassium dihydrogen phosphate in 1 1 distilled water.
This is the stock solution. Dilute 10 ml to 200 ml with distilled water.
This solution contains 0.005 mg phosphorus in 1 ml (5 ppm) .
Procedure
To 5 ml of the solution to be assayed (filtered primary or secondary
effluent) add 2 ml molybdate solution and 0.3 ml stannous chloride
solution. After ten minutes, read in a colorimeter at 690 nm. A blank
and a series of phosphate standards are run at the same time.
113
-------
Total Dissolved Phosphate
Acid Hydrolyzing Solution
Slowly add 30 ml concentrated sulfuric acid to approximately 60 ml
distilled water. When cool, add 0.4 ml concentrated nitric acid and
dilute to 100 ml.
Procedure
To 5 ml of the solution to be assayed (filtered primary or secondary
effluent), add 0.1 ml of the acid hydrolyzing solution. Boil gently
for 90 minutes. Cool and neutralize to a faint pink color (phenolphtha-
lein indicator) with 5N sodium hydroxide solution. Restore volume to
5 ml and determine the orthophosphate content as described above.
Oxygen Uptake Rate Measurement
Principle
The sample (mixed liquor or return sludge) is sparged with air to
create a measurable DO level and introduced into a vessel containing
the oxygen sensor, Figure 1. The vessel is closed to exclude air,
and the rate at which organisms consume the dissolved oxygen is measured
and recorded. Rate is calculated from the slope of the recorded curve.
Apparatus
Beckman Oo Analyzer, Model No. 777 with polarographic oxygen sensor,
and a Hewlett-Packard recorder.
Procedure
A 600 ml aliquot of the sample is aerated for about one minute and
immediately introduced into the flask. The solution is allowed to over-
flow to prevent trapping of air bubbles in the neck of the flask. The
flask is stoppered and stirred with a magnetic stirrer bar. Both analyzer
and recorder are turned on and measurement is continued for three minutes.
From the resulting linear graph, the decrease in the dissolved oxygen
level is divided by the time required. Results are expressed as the
oxygen uptake rate in mg/1 of oxygen per minute.
Total Organic Carbon (TOC)
TOG measurements were performed on a Beckman Model No. 915 Total Organic-
Carbon Analyzer. Samples were acidified with concentrated HC1 to approxi-
mately pll 2 at the time of collection, and then stored under refrigeration
(no longer than three days), until assay.
114
-------
Modified 500 ml
Volumetric Flask
Rubber Stopper
Stirring Bar
Magnetic Stirrer
FIGURE 1
Glass Stoppered Vessel Used for
Measuring G£ Uptake Rates
115
-------
Biochemical Oxygen Demand (BOD)
All BOD values presented are for the five-day BOD test which is
described in Standard Methods for the Examination of Water and
Wastewater, American Public Health Association, New York, p 592
(1967).
116
GOVeRV.'C'.T PHI',Ti'.G OFFICE :",!- <" I - 1 lit; .: 75
-------
SELECTED WATER
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
/. Report No.
3. Accession No.
w
4. Title B10MASS DETERMINATION - A NEW TECHNIQUE FOR ACTIVATED 5- Report Date
SLUDGE CONTROL. «
7. Author(s)
9. Organization
Biospherics Incorporated, Rockville, Maryland
12. Sponsoring Organization
15. Supplementary Notes
116p, 53fig, 22 tab, 18 ref, 1 append.
*. Performing Organization
Report No.
10. Project No.
17050 EOY
11. Contract/Grant No.
14-12-819
13. Type of Report and
Period Covered
16. Abstract Research was conducted to determine the feasibility of using ATP as a measure
of viable biomass in activated sludge. Methods were developed for the extraction of ATP
from sludge and mixed liquor, and for the determination of ATP using the firefly bio-
luminescent procedure. Measurements of ATP were conducted on various pure cultures,
pilot plant and full-scale activated sludge treatment plants. Additional parameters in-
cluding BOD, TOC, oxygen uptake rate, and suspended solids were measured to provide com-
parative and supportive information. Preliminary tests in which ATP measurements of bio-
mass were used to control the percent sludge return were conducted at two full-scale
municipal sewage treatment plants. Lowered return sludge rates were found to produce
effective treatment and increase the biological activity of the sludge. Changes in the
rate of return sludge resulted in changes in ATP concentration of mixed liquor which
preceded changes in suspended solids by as much as 24 hours. The assay was found to be
reproducible and rapid. Results can be obtained within approximately ten minutes.
17a. Descriptors
Wastewater Treatment, Sewage, BOD Removal
17 b. Identifiers
Biomass, ATP, Activated Sludge
17c. COWRR Field & Group
18. Availability
19. Security Class.
. (Report)
20. Security Class.
(Page)
21. No. of
Pages
Send To :
22. Price
WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON, D. C. 20240
Abstractor Gilbert V. Levin
Biospherics Incorporated
WRSIC 102 (REV. JUNE 1971)
GPO 913. Ztt
------- |