WATER POLLUTION CONTROL RESEARCH SERIES
16050EMFOS/71
A Quick Biochemical
Oxygen Demand Test
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
WATER POLLUTION CONTROL RESMRQfi
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 Environmental Protection Agency, through
inhouse 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, B.C. 20^-60.
-------
A QUICK BIOCHEMICAL OXYGEN DEMAND TEST
by
Regents of the University of California, Davis
Davis, California, 95616
for the
ENVIRONMENTAL PROTECTION AGENCY
Grant #16050 EMF
June 1971
-------
EPA Review Notice
This report has been reviewed by the Water
Quality Office, EPA, 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.
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price 60 cents
ii
-------
ABSTRACT
Studies were conducted to develop a satisfactory, short term biological
oxygen demand test suitable for operational control of waste treatment
processes. The Total Biological Oxygen Demand (TbOD) test, a mass
culture technique which utilizes the change in chemical oxygen demand
as resulting from bacterial action, was chosen as the basic system.
Because the T^OD test was developed for and is conceptually limited
to soluble wastewaters, considerable modification of the basic test
was necessary.
The results of the studies show that the modified TbOD test can be
utilized for the determination of the oxygen demand of nonsoluble
wastewaters. Values were not affected by dilution as long as the
initial (time = 0) wastewater COD value was greater than 100 mg/£ .
Additionally, cell concentration does not affect T^OD values obtained.
Because the test was developed from consideration of the stoichiometry
of conversion of organic materials to cells and oxidized end products,
values obtained can be related to ultimate or theoretical biochemical
oxygen demand values. Of greater utility is the development of
COD vs TbOD correlations for a specific wastewater, however. These
correlations are limited due to their empirical nature, but can be
updated continually by running additional TbOD tests.
This report was submitted in fulfillment of Grant Number 16050EMF
under the partial sponsorship of the Water Quality Office, Environ-
mental Protection Agency.
m
-------
CONTENTS
Section Page
I Conclusions 1
II Recommendations 3
III Introduction 5
IV Experimental Procedure 17
V Experimental Results 21
VI Discussion 33
VII Acknowledgments 35
VIII References 37
IX Publications 41
X Appendices 43
-------
FIGURES
No. Page
1 THE PLATEAU BOD 9
2 THE TbOD TEST 12
3 TbOD RESULTS FOR SEWAGE 18
4 TbOD RESULTS FOR SEWAGE 24
5 TbOD RESULTS FOR SEWAGE 25
6 TEST SERIES 1 COD vs TbOD FOR SEWAGE 26
7 TEST SERIES 2 COD vs TbOD FOR SEWAGE 27
8 TOMATO WASTEWATER TbOD 30
9 WINE STILLAGE TbOD 32
A-l TfaOD RESULTS FOR SEWAGE 48
VI
-------
TABLES
No. Page
1 Winter T, OD of Davis Municipal Sewage 22
2 Summer T^OD of Davis Municipal Sewage 23
vn
-------
SECTION I
CONCLUSIONS
1. The Total Biological Oxygen Demand (T^OD) Test can be utilized for
the determination of the biological oxygen demand of nonsoluble
organic wastes.
2. The best method of endpoint determination for the TbOD test is the
attainment of a constant filtrate chemical oxygen demand value.
3. Total biological oxygen demand values are not affected by bacterial
concentration in the measurement system.
4. Because the total biological oxygen demand test is based upon
stoichiometric relationships, has a sound theoretical basis, is highly
reproducible and has a relatively short determination time, T^OD values
are preferable to standard five day biochemical oxygen demand values.
5. Useful empirical correlations can be obtained between waste COD and
for a particular wastewater.
6. Because of the reproducibility, precision and short determination
time the TbOD test is a suitable tool for operational control of waste
treatment processes.
-------
SECTION II
RECOMMENDATIONS
1. The TfoOD test should be subjected to statistical studies which will
provide information on the reproducibility of the test in a number of
laboratories.
2. Total Biological Oxygen Demand Values should be accepted by
regulatory agencies as valid estimates of ultimate BOD.
3. Yield data for biological oxidation processes should be reported
as a function of T^OD removed and T^OD remaining.
-------
SECTION III
INTRODUCTION
Operational control of biological treatment systems such as the
activated sludge process cannot be accomplished without a rapid
technique for determining the biochemical oxygen demand (BOD) of the
wastewater being treated. Currently used methods, including the five
day BOD test, the chemical oxygen demand (COD) test, and instrumental
methods of COD or total carbon determination, are unacceptable for
process control for the following reasons: The COD test measures the
amount of oxidizable biodegradable material of all kinds in a given
wastewater in terms of equivalent oxygen. The BOD test is used to
measure the amount of oxygen consumed in the biochemical oxidation of
the organic matter, but over a time period unrelated to the biochemical
events taking place in the process. Widespread use of instrumental
COD or total carbon test is limited due to the cost of the equipment.
None of these methods are based upon stoichiometric conversion (i.e.,
conservation of mass) of the organics in wastewaters, a prerequisite
for validity.
The purpose of this research was to develop a rapid and reproducible
method that has stoichiometric validity for determining the total BOD
of a wastewater.
Previous Work
When values obtained from BOD measurements are plotted against time, it
has been observed that the curve traced by these points can be approx-
imated using an equation derived by assuming the process follows first
order reaction kinetics. The most commonly used form of the equation
is (1,2):
where
t
L
k
- 10"kt)
BOD at any time, t;
time in days;
a constant = ultimate BOD;
constant.
(1)
Use of Equation (1) implies that a monomolecular relationship exists
between oxidation rate and the concentration of oxidizable material
(i.e. the concentration of material yet to be oxidized at any point on
the curve). But, in fact the heterogeneous culture of oxidizing
biological organisms, primarily bacteria and protozoa, will use
-------
different enzyme systems and different substrates. This would result
in shifting bacterial populations and oxidation rates (3). It follows
that the different food uptakes will proceed at different rates and that
the amount of food remaining to be oxidized will not control the rate
of oxidation until the process becomes food limited. The preceding
statement has been firmly established experimentally (4,5). From this
discussion it becomes apparent that the monomolecular formula and the
attendant five day BOD test is based upon an arbitrary time limit which
is unrelated to the biochemical events taking place within the bottle
or within a treatment process (6,7).
In their review, Orford and Ingram (1) state: "...biological oxidation
of sewage or other substrates involves a complex biological flora of
many different types of bacteria and protozoa, which oxidize a very
heterogeneous organic substrate of simple and complex organic compounds,
some of which require specialized bacteria for oxidation. Thus it
would not be expected that a monomolecular equation used to formulate
relatively simple chemical phenomena could also theoretically formulate
a very complex biochemical oxidation." and "Thus, k and L values,
as ordinarily calculated, are statistical constants in an equation
which follows a portion of a biological oxidation curve and are not
parameters with true physical and biological significance." In
addition, L corresponds to ultimate BOD which cannot be determined
experimentally because in theory the ultimate BOD occurs at infinite
time.
Biochemical interference also plays an important role in discrediting
the monomolecular formula in that it may go undetected in the test and
will depress the BOD curve resulting in lower values than actually
exist. Biochemical interference can occur in two ways, internally and
externally.
An example of internal interference is multiphase substrate uptake due
to catabolity repression or inhibition. Catabolite repression is the
phenomenon that occurs when a substrate related catabolite, present in
the culture, interferes with any further synthesis of an enzyme
necessary for the utilization of another substrate. Catabolite inhibi-
tion occurs when a catabolite inactivates a catabolic enzyme already
produced (8). Both regulatory mechanisms result in two phase (diauxic)
growth and sequential uptake. Inhibition appears immediately while
repression shows up after an initial lag period. The phenomenon of
sequential uptake of multicomponent substrates by both pure and hetero-
geneous bacterial cultures has been studied extensively (9,10,11,12).
It is difficult to determine whether or not sequential or concurrent
substrate uptake is occurring in the five day BOD test. If sequential
uptake is occurring, the BOD curve will be depressed resulting in
erroneous five day BOD values. Sequential uptake cannot be represented
-------
with the monomolecular formula.
Nonvolatile phenol is an example of an external interference that may
be found in wastewater. For even at low concentrations (5 mg/l or
above), it is toxic to protozoa and will result in a high BOD concentra-
tion in the effluent (13). The disappearance of protozoa in a hetero-
geneous culture will also depress the BOD curve and result in lower
five day values.
The five day BOD test is not consistently reproducible by different
laboratories because of the arbitrary test period and biochemical
interference of trace contaminants. It is unrelated to biochemical
occurrences in the system being tested. Consequently the five day
BOD test does not provide reliable stoichiometric measurement of the
bio-oxidizable constituent oxygen demand in an aerobic treatment
process.
The COD test is rapid and inexpensive. Results can be obtained in
three hours, which is suitable for operational control of activated
sludge systems. Unfortunately COD values represent oxygen equivalents
in chemical oxidation not biochemical oxidation and therefore are not
indicative of the biological characteristics of a wastewater. For
example a toxic wastewater may have a high COD and no BOD. Values
obtained from the COD test are definitely more reproducible than those
from the five day BOD test. This fact is why the COD test, with
certain modifications to be discussed later, has stoichometric validity.
Instrumental methods for determining the COD are extremely rapid (about
15 minutes). Unfortunately, instrumental methods suffer from the same
problems that are encountered in the COD test. That is, the total
carbon (as oxygen equivalents) is measured as the sum of both the
biodegradable and the nonbiodegradable organic carbon. Another factor
against the use of instruments is that they are expensive thus limiting
their use to large sanitary districts and other large agencies or
industries. From the preceding paragraphs, suitable criteria for the
BOD determination to have applicability are 1) short duration, lending
itself to regular monitoring of treatment processes and streams;
2) reproducibility, i.e. stoichiometric validity; 3) simple and
inexpensive so as to utilize equipment common to any laboratory; and
4) of reasonable precision, i.e., the range of sensitivity should be
well established. All of the above criteria are met by the Total
Biochemical Oxygen Demand (TbOD) test developed by Hiser and Busch (14).
Total Biochemical Oxygen Demand In 1951 work by Busch and Sawyer with
a soluble synthetic sewage substrate demonstrated that one day BOD
values were extremely reproducible and thus were a very reliable
parameter (6). The "plateau theory" was then offered as an explanation
-------
for the observed phenomena.
"...the reproducibility of the first day values
was due to the nature of the substrate - in that
the food was soluble, readily available and was
practically all consumed after 24 hours of incu-
bation. It was further theorized that if
practically all of the available food had been
converted into cell material after 24 hours, then
any further oxygen utilization, in excess of the
endogenous rate, was due to the activity of
protozoa and other higher organisms feeding on
the bacteria. This would explain the eratic BOD
values obtained for incubation times greater than
24 hours, oxygen consumption being a function of
the predator population." (6)
The plateau observed (Figure 1) corresponded to the point where all of
the food was consumed represented in the BOD curve by a flat region
between the initial rapid bacterial BOD uptake region and the somewhat
slower BOD uptake region consisting of combined endogenous respiration
and predator metabolism of the bacteria.
Much research has been performed to confirm or disprove the plateau
but the occurrence has been well established when the predator to
bacteria ratio has been reduced to prevent obscuring of the plateau
(2,15 through 23).
Initial BOD exertion by bacteria consists of a rapid substrate uptake
resulting in the conversion of organic carbon to C02 and cell tissue.
At the point where all of the available biodegradable organic carbon
has been converted, the BOD exertion rate flattens out to the endogenous
respiration rate representing a food limited respiration process (5).
In a heterogeneous population protozoa became the predominate predators
of bacteria and cause an increased BOD exertion rate in excess of the
bacterial endogeneous respiration rate. However, being of a higher
life form than bacteria, they respire at a lower reproduction efficiency
and rate (23). This is reflected in a slower BOD exertion rate as
shown in Figure 1. The occurrence and the importance of predator
activity has been well documented (2,13,16,23,24). Thus the BOD curve
for a heterogeneous culture is comprised of three distinct sequential
stages: 1) conversion of the biodegradable substrate by the bacteria
to carbon dioxide and cell tissue, 2) endogeneous bacterial respiration,
and 3) protozoan predatory metabolism of bacteria tapering off into
protozoan endogenous respiration (2,25).
The importance of the reproducibility of the plateau cannot be over-
emphasized from the standpoint of stoichiometry. For a given set of
-------
DQ
txo
O
DQ
ce
-------
environmental conditions, bacterial growth and substrate oxidation
should be a stoichiometric process, i.e., the coefficients of
Equation (2) should be reproducible and constant (18).
n, substrate + n2 nutrients + n3 02 -»• n4 cells + n5 C02 + n6 H20 (2)
where
n- = number of moles of each material, and
cells would be in terms of an empirical formulation (26)
such as C5HJ)2N
The BOD bottle provides a nearly constant physical and chemical
environment and as such the value of each stoichiometric number
should be constant. The sequential stages of bacterial growth in
going from the lag phase through the log phase and into the stationary
phase (endogenous respiration) would cause physiological changes
within the cell that should affect the stoichiometric numbers. This
problem could be solved by taking data up to the point where all of
the biodegradable organic carbon in the substrate has been converted
to cellular material and C02 . This would correspond to the endpoint
of Equation (2) and is represented by the plateau in the BOD curve.
Busch (6) reported results with sewage, glucose, and glutamic acid as
substrates. The sewage used for substrate or as a source of bacterial
seed was filtered through a number 2 Whatman filter to reduce the
predator population thus making the plateau more distinct. In each
experiment a definite plateau was exhibited in the BOD exertion curve
which was preceded by the sigmoidal pattern characteristic of classical
bacterial growth. Busch also noted that the plateau BOD value was a
linear function of initial substrate concentration, a relationship
which has not been demonstrated with the five day test.
Experiments with unfiltered sewage and sewage seed gave less distinct
plateaus. In view of this Busch concluded, as mentioned earlier, that
the increased uptake past the plateau was due to protozoan activity.
Subsequent experiments by Busch (6) utilizing the Warburg Respirometer
demonstrated that homogenized domestic sewage gave a distinct plateau
thus lending credence to his theory that protozoan activity was
responsible for the shape of the BOD curve.
The same conclusion was reached by Bhatla and Gaudy (16). Homogeni-
zation by a Waring blender, which creates high local shear, destroys
the protozoa because they lack a cell wall to provide structural
strength. Bacteria being much smaller and having a cell wall can
withstand much higher local shears.
10
-------
Independent work by Lee and Oswald (3) on the effect of seeding
demonstrated that the ultimate oxygen demand of a waste should not
be influenced by initial seed concentration alone if a variety of
species, i.e. a heterogeneous culture, is present. Hence, the BOD
exertion curve should fit a stoichiometric equation.
Test Hiser and Busch (14) recognized that to use the BOD plateau
to determine the total biochemical oxygen demand would require at least
24 hours using the conventional dilution method or a respirometric
method. Because hydraulic residence times in waste treatment processes
are about six hours or less, continuous monitoring for process control
demands a BOD test of considerable shorter determination time.
Consequently, the plateau method is not of significantly greater utility
for process control than the five day BOD test, even though the former
is stoichiometrically valid. To overcome this, Hiser and Busch
developed a simple yet accurate total biochemical oxygen demand (T^OD)
measurement for soluble substrates using the mass culture aeration
technique combined with the COD test. Mass culture is defined as a
bacterial culture concentration of several hundred to several thousand
milligrams per liter. Mass culture concentrations are then much greater
than the concentrations normally utilized in the BOD test.
The mechanics of the TfoOD test consists of first separately measuring
the COD of the soluble substrate and a washed mass culture of aerobic
microbial organisms suspended in tap water. Known volumes of the
soluble substrate and the mass culture are then mixed together and
aerated in a batch operation. The utilization of the substrate can
then be traced by sampling at incremental time periods and running the
COD test on samples filtered through a 0.45 micron filter. Because
the mixture is homogeneous sampling has no effect on the concentration
of the mixture. At the point where all of the biodegradable substrate
organic carbon has been converted to C02 and new cells the filtrate
COD will become constant. The point at which the filtrate COD becomes
constant is referred to as the endpoint filtrate COD. The T^OD is
then determined by the following equation:
TbOD = D.F. x ACOD (mg/A) (3)
where
volume of mixture
D.F. = dilution factor =
volume of substrate
ACOD = the difference between initial COD and
endpoint filtrate COD
A typical curve is shown in Figure 2. Analysis of the curve shows
that the proper substrate to culture ratio (i.e., loading) will result
11
-------
TIME (hours)
COD of Cells
Initially Present
COD
m
BO
E
Q
O
O
A COD
COD
COD of Cells
Grown
t
TIME (hours)
Figure 2
THE TbOD TEST
12
-------
in a reasonable short completion time for the test.
Hiser and Busch contend that the biodegradable soluble organic material
in an acclimated mass culture is completely absorbed and metabolized
by the culture in a batch aeration system. The remaining soluble
organic material in the substrate will be relatively stable and will
result in a constant COD value with respect to time. This residual
COD consists of nonbiodegradable soluble organics, metabolic end-
products, and probably some lysed organisms.
Because the mass culture TbOD test is used to measure the disappear-
ance of food from the biological system rather than measure the amount
of oxygen utilized in the disappearance of the food, cell synthesis is
not relevant. That is, the amount of soluble carbon remaining in the
system with respect to time is being measured not the oxygen uptake.
Consequently the ratio of food that is converted to cellular material
to that which is converted to C02 is not involved. This ratio
greatly affects the five day BOD test.
By defining soluble as that material which will pass through a 0.45
micron filter, the possibility of confusing the organic carbon
originally in the substrate with the new cells produced is eliminated.
Bacteria are usually greater than 1.0 micron at their smallest
dimension and, therefore, do not pass a 0.45 micron filter. This
distinction is made because use of the COD test does not allow
differentiation between substrate organic carbon and cellular organic
carbon. By filtering the culture-substrate mixture, interference
by the cellular organic carbon in measuring the disappearance of the
substrate organic carbon is eliminated. It should be noted that the
organic carbon measured is only that which can be oxidized by the
dichromate. Measurement is made only of the total amount of soluble
organic carbon that is biodegradable to the particular bacterial
culture. Consequently, the TbOD is a measure of the ultimate oxygen
demand of the substrate whereas the five day BOD is a measurement of
the oxygen uptake at an arbitrary point in time that bears little
relationship to the conversion of substrate to C02 .
Hiser and Busch (14) reported that the measured T(jOD of pure substrates
as a percent of the theoretical value varied from 84 to 103 percent.
When the measured value was shown as percent of substrate COD, the
variation was 89 to 99 percent and demonstrates dependence upon the
recovery efficiency of the COD test. Industrial wastes were reported
in percents of the average measured TkOD since the theoretical amounts
were unknown and the variance was 90 to 106 percent. Hiser and Busch
noted that Gaudy (10,27) used the COD test to measure the sequential
disappearance of glucose and sorbitol from a mixed culture. From
Gaudy's work Hiser and Busch calculated a value which was 96.5 percent
of the theoretical removal of glucose-sorbital substrate in an activated
13
-------
sludge system. Hence, the work by Hiser and Busch has shown that the
T^jOD test is accurate within plus or minus 10 percent. This is-
acceptable for process control purposes and is more accurate than the
five day BOD test.
In view of the fact that the previous work by Hiser and Busch (14) and
others was performed on soluble wastes only, their test is limited in
application. Specifically, as developed by Hiser and Busch the TbOD
test cannot be used for domestic wastes which contain varying components
of soluble and insoluble biodegradable organic carbon. Consequently,
modification of the test is necessary. The purpose of this research
is to provide the necessary modifications to make the test more
applicable.
Modified TbOD Test
At this point several definitions need to be introduced for clarifi-
cation. CODjn is defined as the mixed liquor value of COD at any time,
t, of the unfiltered aerated mixture of substrate and bacteria in a
graduated cylinder batch test. The substrate plus new cells value CODS
is defined as the CODm value at any time minus the initial bacterial
cell COD measured prior to mixing. Because this difference is constant
during the test the two curves will be parallel. In this sense, the
CODS curve will be artificial because it is generated mathematically
rather than by laboratory analysis. However, the change in the lower
curve represents the conversion of the biodegradable portion of the
substrate to C02 with respect to time. The value of the material
which passes through a 0.45 micron filter is defined as CODf. The
three symbols defined above will be used extensively throughout the
remainder of this discussion.
Difficulties arise when the TbOD test, as developed by Hiser and
Busch, is applied to a substrate consisting of both soluble and non-
soluble biodegradable organic carbon. Initially, the CODS will
represent the total organic carbon including both the soluble and
nonsoluble components. However, the initial CODf value will include
only the amount of organic carbon that is soluble. For this reason,
the two initial values will not be equal and cannot coincide as found
by Hiser and Busch. A plot of MLSS concentration versus time will
not yield any information about the cell growth as the MLSS values
will include both the cells and the nonsoluble organics. This
inclusion renders the MLSS data meaningless except at the endpoint
where all of the nonsoluble organics have been converted to COg and
cell tissue. The CODf values will also be meaningless until the
endpoint because of the exclusion of the nonsoluble organics prior to
the endpoint. The filtrate endpoint is reached when all of the
nonsoluble organics have hydrolyzed and become converted to C02 and
cell tissue. This point represents completion of the bio-oxidation
14
-------
process. Beyond the endpoint the COD* value ceases to change with
time. The similarity between the T^OD test for soluble wastes and
the modified test ends here because the difference between initial and
endpoint COD* does not represent the ThOD but represents the value of
the nonsoluble organics and the new cells grown due to cell synthesis.
The TbOD of the waste is found by formula (3) where COD is the initial
CODS minus the endpoint COD*.
15
-------
SECTION IV
EXPERIMENTAL PROCEDURE
The experimental procedure developed differed slightly from that of
Hiser and Busch (14), and is described below.
Chemical oxygen demand was measured according to the procedure given
in Standard Methods (28). Prior to starting an experiment 20 mi-
samples were taken of the sewage and bacterial culture. The COD of
the samples was measured, allowing the calculation of initial CODm
for comparison with mixed liquor samples taken at time equal to zero.
Wastewater was then mixed with the cell suspension (the suspension was
maintained in the laboratory) in a predetermined ratio (usually 9:1).
The mixture was aerated in a two liter graduated cylinder throughout
the experiment. Aeration was stopped briefly to take samples. All
samples were taken at the midpoint of the quiescent mixed liquor
column to insure a representative mixed liquor concentration. Mixed
liquor samples (CODm) were taken every half hour for three hours and
every hour thereafter for a total of eight hours. Periodically an
experiment was extended to twenty-four hours to check the validity of
the endpoint determination. Filtrate samples (CODf) were taken at
one hour intervals. Approximately 50 milliliters of mixed liquor was
removed from the graduated cylinder and filtered through a 0.45 micron
filter (Selas Flotronics FM 47). Filtrate was collected in screw top
culture tubes and placed in a freezer until CODf determinations could
be made.
In conjunction with the above TbOD test procedure additional data were
gathered on MLSS vs time. For practical applications, the MLSS need
not be incorporated into the TbOD test. At the same time the COD
samples were taken, two additional twenty-milliliter samples were
taken to determine the MLSS concentration. The MLSS was collected on
pre-weighed 0.45 micron filters (Selas Flotronics FM 47). The filters
and MLSS were dried for two hours at 105° Centigrade, cooled for two
hours in an airtight desiccator and then weighed on a Metier model H-16
balance to determine the MLSS concentration at the particular point in
time. These concentration data were then plotted vs time.
The precision of the determinations can be estimated by comparing the
calculated and measured initial CODm values. Throughout the experi-
ments, the average difference between the theoretical value and the
measured value was 3.8% and the maximum observed difference was 14.2%.
TbOD test results were plotted as CODm vs time, CODS vs time and
CODf vs time as shown in Figure 3. The CODS is parallel to the
17
-------
"IIIII
Sewage Source: Davis, Calif. Treatment Plant
Data Taken: 2/17/70
Sewage T OD = 1/0.9 (254) = 282 mg/L
Q
O
O
300
200
100
t
ACOD = 287 - 33
-254 mg/L
8 12 16
TIME (hours)
Figure 3
TbOD RESULTS FOR SEWAGE
20
24
18
-------
curve and has an initial (t = 0) value equal to the measured sewage
COD multiplied by the inverse of the dilution factor. As with the
CODm curve, change in COD value with time is due to conversion of
organics to carbon dioxide.
When bio-oxidation is complete, all three curves should become constant
with time. This is the endpoint of the test. The difference between
the CODS and the endpoint CODf is the COD of the cells grown in the
experiment and is practically constant until endogenous respiration
and predator activity becomes significant following the stationary
period. Therefore, the TkOD of the nonsoluble waste is simply the
initial CODS value minus the endpoint CODf value multiplied by the
dilution factor. All other intermediate points are not used in this
test.
19
-------
SECTION V
EXPERIMENTAL RESULTS
Municipal Sewage
Typical data resulting from the TjjOD test with sewage are presented
in Tables 1 and 2, and shown in Figures 3, 4 and 5. Column 2, the
theoretical initial CODra, is the sum of the initial bacterial COD and
sewage COD (which were determined prior to mixing), multiplied by the
appropriate dilution factors (one minus inverse column 8 and inverse
column 8a respectively). Column 3, the mixture initial CODm, is the
measured COD of the mixture at time equal to zero and is equal in value
to the theoretical initial CODm. Although the difference between
theoretical and measured initial COD is well within allowable experi-
mental error it has been determined that calculation of the TbOD should
be based upon the theoretical initial CODm. There are two reasons for
this. One, the mixture initial CODm is less reliable because a short
time period during which bacterial activity is proceeding must be
allowed for complete mixing upon initiation of the test. This mixing
period results in a slightly lower COD value than actually existed at
time equal zero. Two, the utility of the TbOD test is the generation
of the empirical linear relationship between the substrate COD value
and the TbOD as depicted in Figures 6 and 7. A maximum of three hours
are required to find the TbOD of the substrate by measuring the sewage
COD strength. Consequently, column 5, the initial CODS, is equal to
the measured (prior to mixing) sewage COD divided by the dilution
factor in column 8. The TbOD value (columns 8a and 9) is then equal
to ACOD (column 7 = column 5 - column 6) multiplied by the dilution
factor (column 7 x column 8).
In Figure 3, the CODm and its parallel CODS curve represent the typical
biological activity of a heterogeneous bacterial culture exposed to an
organic enriched environment. Initially, there is very rapid growth
represented in the CODm curve by a rapid decrease in sewage COD. The
next portion of the CODm curve consists of a "flat" region corresponding
to the classical stationary phase of bacterial activity. The change in
growth rate is due to complete conversion of the biodegradable portion
of the substrate initially present to new cells and carbon dioxide.
In this region the production and destruction of cell mass are occurring
at approximately equal rates. Additionally, the stationary phase
corresponds to the endpoint of the CODf and the maximum cell mass. The
region of the CODm curve following the stationary phase, indicated by
a slow decrease in CODm with time, represents increased predator
activity (protozoa), bacterial autolysis and oxidation of particulate
organics which may be adsorbed on floe surfaces. The growth rate of
predators is much slower than the bacterial growth rate and, therefore,
conversion is less than fifty percent efficient. Hence, the total COD
21
-------
ro
ro
1
Test
Date
2/17
2/24
3/5
3/25
3/26
4/2
2
Theoretical
Initial
CODm
585
618
791
372
534
658
3
Mixture
Initial
CODm
584
598
746
363
524
626
4
Measured
Cell
COD
293
367
681
188
296
330
5
Initial
CODS
287
251
110
184
238
328
6
Endpoint
CODf
36
33
62
49
60
38
7
ACOD
251
218
48
133
178
290
8
Sewage
Dilution
Factor
1
0.9"'
•t
0.918"'
On™ 1
.9
9
TbOD
279
242
53
145
198
322
10
Measured
Sewage
COD
318
279
122
200
264
364
Table 1
WINTER TbOD OF DAVIS MUNICIPAL SEWAGE (Test Series 1)
-------
co
1
Test
Date
6/26
6/28
7/2
7/9
7/16
1
2
1
2
1
2
1
2
1
2
2
Theoretical
Initial
CODm
333
461
-
527
813
256
399
353
505
3
Mixture
Initial
CODm
316
437
417
703
520
817
258
358
357
433
4
Measured
Cell
COD
144
272
292
564
291
570
141
285
147
299
5
Initial
CODS
189
189
125
139
236
243
115
114
206
206
6
Endpoint
CODf
34
32
23
30
27
28
20
20
37
37
7
ACOD
155
157
102
109
209
215
95
94
169
169
8
Sewage
Dilution
Factor
0.9"1
8a
TbOD
172
174
113
121
232
239
105
105
188
188
9
Average
TbOD
173
117
235
105
188
10
Average
Measured
Sewage COD
210
147
266
128
229
Table 2
SUMMER TbOD OF DAVIS MUNICIPAL SEWAGE (Test Series 2)
-------
700
600
500
400
1 I I I I
Sewage Source: Davis, Calif. Treatment Plant
Data Taken: 2/24/70
Sewage TbOD = 1/0.9 (218)
COD
m
= 242 mg/L
O
O
O
300
200
f
COD = 251 - 33
= 218 mg/L
8 12 16
TIME (hours)
Figure 4
TbOD RESULTS FOR SEWAGE
20
24
24
-------
T
T
T
ro
en O
O
Data Taken: 3/25/70
Sewage TbOD = (133) = L45 mg L
0.918
Sewage Source: Davis, Calif. Treatment Plant
ACOD = 182-49 = 133 mg/L
2 3
TIME (hours)
Figure 5
TbOD RESULTS FOR SEWAGE
-------
O
O
O
O
LU
CE
Samples Collected From
2/17/70 to 4/2/70
150 -
100 -
50 -
150 200
TbOD (mg/L)
Figure 6
TEST SERIES 1 COD vs TbOD FOR SEWAGE
26
-------
txo
O
O
O
O3
C/0
O
LU
o:
400
350
300
250
200
150
100
50
'b 7/2
7/16
6/26
6/28
7/9
Samples Collected From —
6/26/70 to 7/16/70
I
I
1
I
50 100 150 200
TbOD (mg/L)
250
300
Figure 7
TEST SERIES 2 COD vs. TbOD FOR SEWAGE
27
-------
of the mixed liquor will continue to decline at a slow rate. The CODf
curve then declines rapidly to a constant value corresponding to the
amount of soluble nonbiodegradable material that was originally in the
sewage. The quantity of residual COD will remain constant as shown
in Figures 3 and 4 because the biodegradable fraction initially present
in the sewage has been consumed by the time the endpoint is reached.
The only other biodegradable organic materials in the mixture are the
floe particles which consist primarily of cells. These particles are
not in the filtrate and hence they are not measured in the CODf
determination.
Although Figure 4 is similar to Figure 3 with respect to sewage
strength, filtrate COD was significantly more erratic in the experi-
mental results reported in Figure 3. As stated previously, the
erratic CODf data points prior to the endpoint do not affect the
value and, therefore, are not significant in this test.
Data for sewage having a very low initial soluble COD are reported
in Figure 5. The CODf might be expected to increase during the early
phase of the test as the initially insoluble COD becomes soluble.
This is born out by the data.
Data from TbOD tests on sewage were found to be consistently similar
to that shown in Figures 3, 4 and 5. Use of the filtrate COD curve
for endpoint determination has proved to be very satisfactory and
is supported by Gifford's (29) studies at Texas A & M University on
soluble carbon in sewage. A complete sample calculation of TbOD is
included in Appendix A.
Direct utilization of the TbOD test as reported for process control is
difficult because approximately eight to ten hours is necessary to
obtain data. Because most waste treatment processes receive a rela-
tively consistent waste with respect to the fractional organic makeup
for short periods of time, a correlation between waste COD and waste
TbOD should be obtainable and the relationship should be linear. A
correlation for settled sewage entering the Davis, California sewage
treatment plant over a forty-five day period and a twenty-one day
period are depicted in Figures 6 and 7 respectively. (Test Series 1
and Test Series 2). The sewage COD values varied for Test Series 1 by
a factor of three (122 to 364 mg/2.) and for Test Series 2 by a factor
of two (129 to 268 mg/Jl), which are approximately the ranges that
would be expected in most sewage treatment plants. Changes in the
sewage characteristics (such as caused by the July through October
canning season in Davis) would necessitate development of a new curve
(see Figures 6 and 7). Figures 6 and 7 were developed using the data
reported in Tables 1 and 2 (columns 9 and 10) respectively.
28
-------
Industrial Wastewaters
Wastewaters from the Hunt-Wesson Cannery of Davis, California and the
Bear Creek Winery of Guild, Inc in Lodi, California, were chosen as
industrial waste sources. The cannery wastewater was studied during
the tomato processing season, and was fairly typical of tomato wastes
discussed in the literature. Wastewater COD values were approximately
1000 mg/£ with more than ninety percent of the COD passing a 0.45
micron filter. These COD values are for settled samples and thus do
not include easily settled or skimmed particles of the fruit.
Results of the studies on tomato processing Wastewaters were similar
to those for municipal sewage, even though the wastewater was consider-
ably higher in COD. Cell concentrations used were of the order of
400 mg/£ (dry weight) in all cases. Reaction time necessary to remove
the degradable organic material was approximately eight hours. This
time probably could be shortened considerably by increasing the cell
concentration. Based upon the studies with municipal sewage, test
precision would probably remain about the same for cell concentrations
up to 1500 mg/£ .
Because the cannery wastewater contained such a high soluble fraction
the results are more typical of those for soluble wastes and as such
require fewer assumptions and approximations in analyzing the data.
Typical results are shown in Figure 8. An approximation of cell
production can be made as shown in Figure 2. Assuming that the
empirical cell formulation C5Hy02N is valid, cell yield would be 0.45
mg cells per mg T^OD, and the oxygen required for treatment would be
251 mg Q£ per liter of wastewater.
For the purpose of continuous process control a COD vs T^OD correlation
curve could be constructed in the same manner as shown in Figures 6 and
7. As in the case of municipal sewage the correlation curve is empir-
ical and should be checked at intervals of approximately one week.
Studies of winery wastewater were considerably more difficult than
either those on municipal sewage or tomato processing wastewater.
Winery wastewater results from production of beverage brandy and
fortifying spirits used in production of sweet wines. The wastewater
(known as stillage or still slops) consists of the "bottoms" coming
off the still. Characteristics of the stillage vary considerably with
the type of material being distilled. Stilling material can typically
be divided into wine, lees (the residue after wine is decanted from
secondary fermentation vats) and pomace washings (fermented material
resulting from crushed stems, grapes, skins, etc). Because pomace
still age has proven to be the strongest and the most difficult treat-
ment problem all of the TbOD studies were made on pomace still age
samples.
29
-------
1200
Wastewater TROD = (816/0.9) = 907 mg/L
tta
o
o
o
4 6
TIME, (hours)
TOMATO WASTEWATER TbOD
Figure 8
30
-------
The stillage studied had a COD of approximately 17,000 mg/£ and
suspended solids were approximately 15,000 mg/l . Considerable
reduction in strength of the wastewater could be obtained by settling.
Supernatant liquid COD values were approximately 13,000 mg/Jl after
settling for one hour. Stillage of all types has high acidity (1000
to 3000 mg/Si as CaCOa) and pH values vary between 3.0 and 6.5 .
Because of the high acidity all samples were neutralized (to pH 7.0)
with NaOH prior to making T^OD determinations.
Typical TfcOD data for pomace stillage is shown in Figure 9. The
stillage was diluted by a factor of 40 in order to facilitate COD
determination. No problems occurred in the bio-oxidation process and
the results are very similar to those for municipal sewage. Data
shown in Figure 9 was taken over an unusually long period (71 hours)
for the purpose of observing the stability of the minimum CODf value.
A problem related to assuming that the nonsoluble organics initially
present are degradable is that degradation produces soluble COD. As
Figure 9 demonstrates, this is not a problem in determining the
endpoint.
31
-------
tao
O
O
O
CO
900
800
700
600
500
400
300
200
100
0
10
20
COD
m
Wine Stillage Dilution = 1:40
TbOD -(1/0.9) (256) (40) = 11,400 mg/L
Stillage COD = 13,300 mg/L
30 40 50
TIME (hr.)
Figure 9
WINE STILLAGE TbOD
60
-------
SECTION VI
DISCUSSION
The comparison of COD values refluxed for shorter periods than two
hours may result in a determination time of less than the standard
two hours. The rapid COD test developed by Jeris (30) would be useful
when applicable. Once the initial COD vs TbOD curve is generated,
continuous monitoring of influent COD can be used to effectively main-
tain effluent quality. For example, adjustments in the MLSS concen-
tration could be made in response to varying Tj-,00 loadings. The
correlation curve should be checked (about once per week) by running
a TbOD test to defect changes in the influent wastewater characteristics.
The main limitation of this test is that for a wastewater with a COD
of less than 100 mg/£ the test becomes erratic. There will
be no limit on higher values as the two COD values (initial COD and
endpoint CODf) used to calculate the TbOD of the substrate will be
directly proportional to the dilution used to bring the sewage COD
strength down within the test sensitivity range for titration.
The MLSS data generally conformed to the predicted behavior, in that
the maximum cell mass occurred approximately at the point in time
where the biodegradable portion of the organic carbon was completely
converted to cellular material and C02 . However, a comparison
between the computed mass of the new cells determined from the MLSS
data and from the TbOD data was inconclusive. The computed cell mass,
AMLSS, is equal to the difference between the maximum and initial cell
mass. The new cell mass computed from the TbOD data is equal to the
difference between the value of CODS in the flat region and the CODf
endpoint value multiplied by the dilution factor, and divided by the
oxygen equivalent factor 1.414 (based on the cell formula CsHyOpN).
The difference between the two computed new cell mass values ranged
from 0% to 57%. Differences were not consistent except that
all values were greater than predicted from T^OD data. Because pre-
diction of cell yield was not one of the original objectives of the
study time and funds necessary to improve the experimental methods were
not available.
One possible source of error in the determination of cell growth is
related to the nonsoluble material present in the wastewater. If the
material differed significantly in composition from the cells or was
not biodegradable valid growth predictions from gravimetric data
would be impossible.
33
-------
Several other sources of possible error were analyzed during the period
of experimentation. Evaporation of water in the aerated graduated
cylinder could result in higher concentrations of the mixed liquor thus
giving higher COD values than actually exist. Consequently, an equal
volume of tap water was aerated in a graduated cylinder and the decrease
in volume was monitored simultaneously with an actual TbOD test run.
The amount of water lost was negligible thus eliminating the necessity
of adding water to the aerated cylinder during the course of a test run.
The reliability of the TsOD test was evaluated by running simultaneous
Tt>OD tests using two different cell concentrations. This procedure
was repeated for five different sewage samples taken from the same
source (see Table 2).
The difference between the values of the replicate samples (column 8a)
and their average ranged from 0 to 3.2 percent.
34
-------
SECTION VII
ACKNOWLEDGMENTS
Work reported was done in the Environment Engineering Laboratories of
the Department of Civil Engineering of the University of California,
Davis. Contributors were M. K. Mull is, J. H. Sherrard, T. M. Chadwick,
J. C. Dodd and S. Lee. E. D. Schroeder, principal investigator,
compiled the final report.
The support of the project by the Water Quality Office, Environmental
Protection Agency and the interest of the Project Officer, Dr. Walter M.
Sanders is acknowledged with sincere thanks.
35
-------
SECTION VIII
REFERENCES
1. Orford, H. E. and Ingram, W. T., "Deoxygenation of Sewage, I,
Critical Review of the Monomolecular Formula," Sewage and
Industrial Wastes, 25, 4, 419 (1953).
2. Sawyer, C. N. and McCarty, P. L., Chemistry for Sanitary
Engineers. 2nd Ed., McGraw-Hill, New York, 39T~( 1967).
3. Lee, E. W. and Oswald, W. 0., "Effects of Seeding on BOD," The
Water and Waste Treatment Journal, 400 (1959).
4. Buswell, A. M., Mueller, H. F. and Van Meter, I., "Biological
Explanation of the Rate of Oxygen Consumption in the B.O.D. Test,"
Sewage and Industrial Wastes, 26, 3, 276 (1954).
5. Hoover, S. R., Jasewicz, L. and Porges, N., "An Interpretation of
the B.O.D. Test in Terms of Endogenous Respiration of Bacteria,"
Sewage and Industrial Wastes, 25, 10, 1163 (1953).
6. Busch, A. W., "B.O.D. Progression in Soluble Substrates,"
Proceedings of the 13th Industrial Waste Conference, Purdue
University, TT958T and Sewage and Industrial Wastes, 30, 11, 1336
(1958).
7. Montgomery, H. A. C., "The Determination of Biochemical Oxygen
Demand by Respirometric Methods," Water Research, 1, 631 (1967).
8. Stumm-Zollinger, E., "Substrate Utilization in Heterogeneous
Bacterial Communities," Journal Water Pollution Control Federation,
40, 5, 2, R 213 (1968).
9. Gaudy, A. F., Jr., Gaudy, E. T. and Komolrit, K., "Multicomponent
Substrate Utilization by Natural Populations and a Pure Culture
of Escherichia Coli," Applied Microbiology, 11, 157 (1963).
10. Gaudy, A. F., Jr., Komolrit, K. and Bhatla, M. N., "Sequential
Substrate Removal in Heterogeneous Populations," Journal Water
Pollution Control Federation, 35, 7, 903 (1963).
11. Stumm-Zollinger, E., "Effects of Inhibition and Repression on the
Utilization of Substrates by Heterogeneous Bacterial Communities,"
Applied Microbiology, 14, 654 (1966).
37
-------
12. Mateles, R. I. and Chian, S. K., "Kinetics of Substrate Uptake in
Pure and Mixed Culture," Environmental Science and Technology,
3, 6, 569 (1969).
13. Curds, C. R., Cockburn, A. and Vandyke, J., "An Experimental
Study of the Role of the Ciliated Protozoa in the Activated-
Sludge Process," Water Pollution Control. Discussion, 327 (1968).
14. Hiser, L. L. and Busch, A. W., "An 8-Hour Biological Oxygen Demand
Test Using Mass Culture Aeration and COD," Journal Water Pollution
Control Federation, 36, 4, 505 (1964).
15. Pipes, W. 0., Miholits, E. M. and Boyle, 0. W., "Aerobic Cell
Yield and Theoretical Oxygen Demand," Proceedings 18th Industrial
Waste Conference, Purdue University, 418 (1963)T
16. Bhatla, M. N. and Gaudy, A. F., Jr., "Role of Protozoa in the
Diphasic Exertion of BOD," Journal Sanitary Engineering Division,
Amer. Soc. Civil Eng., 91, SA3, 63 (1965).
17. Gaudy, A. F., Jr., Bhatla, M. N., Follett, R. H. and Abu-Niaaj, F.,
"Factors Affecting the Existence of the Plateau During the Exertion
of BOD," Journal Water Pollution Control Federation. 37, 444 (1965).
18. Schroeder, E. D., "Importance of the BOD Plateau," Water Research,
2, 803 (1968).
19. Busch, A. W., Grady, L., Roa, T. S. and Swilley, E. L., "Short-Term
Total Oxygen Demand Test," Journal Hater Pollution Control
Federation, 34, 4, 354 (1962T
20. Grady, L., Jr. and Busch, A. W., "B.O.D. Progression in Soluble
Substrates - VI - Cell Recovery Techniques in the T^OD Test,"
Proceedings 18th Industrial Waste Conference. Purdue University,
194 (1963).
21. Kerhnerger, G. J., Jr., Norman, J. D., Schroeder, E. D. and
Busch, A. W., "BOD Progression in Soluble Substrates - VII -
Temperature Effects," Proceedings 19th Industrial Haste Conference,
Purdue University, 796 (1964).
22. Nakata, T. and Kumagai, Y., "Respirometric Determination of B.O.D.,"
Presented at the 5th International Conference on Water Pollution,
I.A.W.P.R., San Francisco, California, (July, 1970).
23. Fair, G. M., Geyer, J. C. and Okun, D. A., Water and Wastewater
Engineering; Water Purification and Wastewater Treatment and
Disposal. Vol. 2. John Wiley and Sons, Inc., New York, 32^T6~ (1968).
38
-------
24. Curds, C. R. and Fey, G. J., "The Effect of Ciliated Protozoa on
the Fate of Escherichia coli in the Activated-Sludge Process,"
Water Research, 3, 853 (1969).
25. Myrick, N. and Busch, A. W., "B.O.D. Progression in Soluble
Substrates - II - The Selective Stimulation of Respiration in
Mixed Cultures of Bacteria and Protozoa," Proceedings 15th
Industrial Waste Conference, Purdue University, 32 (I960).
26. Hoover, S. R. and Porges, N., "Assimulation of Dairy Wastes by
Activated Sludge," Sewage and Industrial Wastes. 24. 3, 306 (1952).
27. Gaudy, A. F., Jr. and Romanathan, M., "A Colorimetric Method for
Determining Chemical Oxygen Demand," Presented at the 3rd
Industrial Water and Wastes Conference, Rice University, Houston,
Texas, (1963~T
28. Standard Methods for the Examination of_ Water and Wastewater,
12th Ed., American PubTTc Health Association, New York, 510 (1965).
29. Gifford, Robert, M.S. Thesis, Texas A & M University, (1969).
30. Jeris, J. J., "A Rapid COD Test," Water and Wastes Engineering, 89
(1967).
31. Engelbrecht, R. S. and McKinney, R. E., "Membrane Filter Method
Applied to Activated Sludge Suspended Solids Determinations,"
Sewage and Industrial Wastes, 10, 1321 (1956).
39
-------
SECTION IX
PUBLICATIONS
1. MulUs, Michael K. and Schroeder, E. D., "A Rapid Biochemical
Oxygen Demand Test Suitable for Operational Control," Journal of
Water Pollution Control Federation. 43_, 209-215, 1971.
2. Mull is, Michael K. and Schroeder, E. D., "Use of the TbOD Test
With Colloidal Wastewaters," Presented at the 25th Industrial
Waste Conference, Purdue University, May 1970.
3. Mullis, Michael K., "Use of the TbOD Test With Colloidal
Wastewaters," Thesis submitted in partial fulfillment of the
Degree of Master of Science, University of California, Davis,
December 1970.
41
-------
SECTION X
APPENDICES
43
-------
COD DATA SHEET
Date July 9, 1970 Standardization:
Experiment ^b
OD
1.
Final: 26.40
Initial: 0.00
A Titrant: 26.40
A Average: (26.40 + 26.
Sample:
Fe(NH4)2(S04)2
Time
Blank
Blank
Cells
Cells
Sewage
1220
1250
1320
1350
1420
1450
1520
1620
1720
1820
2020
2220
7/10/70 1220
Normal
Hour
0
0.5
1.0
1.5
2.0
2.5
3.0
4.0
5.0
6.0
8.0
10.0
24.0
ity: 1
Sample
Size, mi
20
20
2
5
20
20
20
20
20
20
20
20
20
20
20
20
20
20
0 mi x 0.
Initial
0.00
22.72
0.00
22.30
0.00
0.00
19.35
0.00
20.25
0.00
20.35
0.00
20.68
0.00
20.80
17.72
19.75
2.00
25N/26
Final
26.10
48.92
22.30
39.45
22.80
19.35
39.80
20.25
40.50
20.35
41.00
20.68
41.70
20.80
41.90
38.78
39.08
23.42
2. 3.
26.40 26.30
0.00 0.00
26.40 26130
40 + 26.30)/3 = 26.37 mi
Run #10
.37 mi = 0.
Titrant
ATitrant
26.10
26.20
22.30
17.15
22.80
19.35
20.45
20.25
20.25
20.35
20.65
20.68
21.02
20.80
21.10
21.06
21.33
21.42
0948N
Blank
26.15*
26.15
26.15
26.15
26.15
26.15
26.15
26.15
26.15
26.15
26.15
26.15
26.15
26.15
26.15
26.15
26.15
COD mg/£
1460
=1413*
1365
127
258
216
224
224
220
209
207
195
203
191
193
183
179
*Average value between two measurements
44
-------
COD DATA SHEET
Date July
Experiment
Fe(NH4)2(SO
Hour
Blank
Blank
0
1
2
3
4
5
6
8
10
24
10, 1970
TbOD
4)2 Normal
Sample
Size.mJl
20
20
20
20
20
20
20
20
20
20
20
20
Standardi
Final :
Initial :
A Titrant
A Average
zation
25
0
: 25
: (25
ity: 10 mi x
Initial
0.00
0.00
0.00
24.18
0.00
24.82
0.00
24.80
0.00
24.76
0.00
0.00
Final
25
25
24
48
24
49
24
49
24
49
25
25
.62
.50
.18
.60
.82
.80
.80
.90
.76
.90
.10
.20
1.
.79
.00
.79
.79 +
Sample
25.70
: Fil
2.
25.70
0.00
25.70
+ 25.
70J/3 =
3.
25.70
0.00
25.70
25.73 ml
trates for Run #10
0.25N/25.73 mi =
Titrant
ATitrant
25
24
24
25
24
24
25
24
25
25
25
.56*
.18
.42
.82
.98
.80
.10
.76
.14
.10
.20
0.0972N
Blank COD
25.
25.
25.
25.
24.
25.
25.
25.
25.
25.
56
56
56
56
56
56
56
56
56
56
mg/SL
54
44
29
23
30
18
31
16
18
14
*Average value between two measurements
45
-------
CALCULATIONS
Governing Equation
COD(mgA) = (B - S)8°QQ x N (standard Methods)
where B = amount of titrant for the blank, mi
S = amount of titrant for the sample, mi
N = normality of Fe(NH4)2(S04)2
T = sample size, mi
Sample COD
rnn _ (26.15 - 20.45)8000 x 0.0948
LUU0.5 " 20
CODQ 5 = 216 mg/£
Theoretical Mixture COD @ t - 0
Ratio of sewage to cells in reactor =9:1
CODT = 0.9 x 127 + 0.1 x 1413 = 256 mg/£
Measured Mixture COD @ t = 0
COOM - (26.15 - 19.35)8000 x 0.0948 . ^
Comparison of Theoretical and Measured Initial COD
Error = 258,:,.256 x 100% = 0.78%
£bo
Use CODy for plotting
Initial Cell COD = 0.1 x 1413 = 141 mg/i
Initial COD,. = 256-141 = 115 mg/A
46
-------
Calculation of
Initial CODs = 115 mg/A
Endpoint CODf = 20 mg/i
ACOD = 115-20 = 95 mg/i
= 95 x 0.9"1 = 105 mg/£
47
-------
00
OkO
o
o
o
300
<
200
100
0
(
— Sewage
Date T<
0° °°
L_T oo 0 o _ _
u ^^^^ O
V^ACOD,- 115 - 20 = 95
s>-£-^ 09^ 9 — JK —
i_ Ti T v ' • *- -i
) 2 4 6 8
— i — i — | — i — r~ « i i i * i ' i • i •
Source: Davis, Calif. Treatment Plant
aken: 7/9/70
o "^
r s /-COD,
mg/L / /
/
10 12 14 16 18 20 22 24 2
TIME (hours)
Figure A - 1
TbOD RESULTS FOR SEWAGE
-------
SELECTED WATER
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
1. Report No.
2.
4. Title
A QUICK BIOCHEMICAL OXYGEN DEMAND TEST
7. Author(s)
9. Organization
UNIVERSITY OF CALIFORNIA, DAVIS, CALIFORNIA
12. Sponsoring Organization
15. Supplementary Notes
3. Accession No.
w
5. Report Date
6.
8. Performing Organization
Report No.
10. Project No.
16050 EMF
11. Contract/Grant No.
13. Type of Report and
Period Covered
16. Abstract
Studies were conducted to develop a satisfactory, short term biological oxygen demand
test suitable for operational control of waste treatment processes„ The Total
Biological Oxygen Demand (T-^OD) test, a mass culture technique which utilizes the
change in chemical oxygen demand as resulting from bacterial action, was chosen as the
basic system. Because the T^OD test was developed for and is conceptually limited to
soluble wastewaters, considerable modification of the basic test was necessary.
The results of the studies show that the modified T^OD test can be utilized for the
determination of the oxygen demand of nonsoluble wastewaters. Values were not affected
by dilution as long, as the initial (time = 0) wastewater COD value was greater than
100 mg/1. Additionally, cell concentration does not affect T^OD values obtained„
Because the test was developed from consideration of the stoichiometry of conversion
of organic materials to cells and oxidized end products, values obtained can be related
to ultimate or theoretical biochemical oxygen demand values. Of greater utility is
the development of COD vs. T^OD correlations for a specific wastewater, however. These
correlations are limited due to their empirical nature, but can be updated continually
by running additional T^OD tests.
17a. Descriptors
17b. Identifiers
17c. COWRR Field & Group
18. Availability
19. Security Class.
(Report)
20. Security Class.
(Page)
21. No. of
Pages
22. Price
Send To:
WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON, D. C 20240
Abstractor
Institution
WRSIC 102 (REV. JUNE 1971)
GP 0 9 13.26 I
------- |