WATER POLLUTION CONTROL RESEARCH SERIES
12050 DSH 03/71
   The Impact of Oily Materials
   on  Activated Sludge Systems
ENVIRONMENTAL, PROTECTION AGENCY • WATER QUALITY OFFICE

-------
         WATER POLLUTION CONTROL RESEARCH SERIES

The Water Pollution Control Research Series describes the
results and progress in the control and abatement of pollu-
tion of our Nation's waters.  They provide a central source
of information on the research, development, and demonstration
activities of the Water Quality Office, Environmental Protection
Agency, through inhouse research and grants and contracts
with Federal, State, and local agencies, research institutions,
and industrial organizations.

Inquiries pertaining to the Water Pollution Control Research
Reports should be directed to the Head, Project Reports
System, Office of Research and Development, Water Quality
Office, Environmental Protection Agency, Washington, D.C. 20242.

-------
                    THE  IMPACT OF
                    OILY MATERIALS
                           ON
              ACTIVATED SLUDGE  SYSTEMS
                     Submitted to:

          The American  Petroleum Institute
     Committee  for Air and Water Conservation

                          and

          Industrial Pollution Control Branch
     Environmental Protection Administration
                          by:

                  Hydroscience,  Inc.
                  363 Old Hook Road,
                 Westwood, New Jersey
                         07675
                 Project No. 12050 DSH


                      March 1971
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, B.C. 20402 - Price $1.25
                      Stock Number 5501-0088

-------
                     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 recom-
mendation for use.
                             11

-------
                          ABSTRACT
Small scale continuous activated sludge systems were exposed
to a variety of oily compounds at several loading levels and
performance observed.  Batch studies to determine biodegrad-
ability and the effect of emulsification and temperature on
the rate of biological reaction were also conducted.

Oils introduced into an activated sludge system are absorbed
on the floe and slowly degrade.  If the loading rate is higher
than the degradation rate and the rate of wastage, the oil
accumulates on the sludge.  This accumulation causes a loss
of density and a loss of acceptable settling characteristics.
The biological system fails due to the loss of sludge but the
ability of the microbial system to remove other substrates is
not inhibited.

The continuous feed level of oils to activated sludge should
not exceed 0.10 pounds per day per pound of sludge under aera-
tion.  Shock loads should not exceed 5% of the weight of the
sludge under aeration.

The study also considered physical separation of oils before
biological treatment, and various chemical methods of separat-
ing complex emulsions.

This report is submitted in fulfillment of Grant Project No.
12050DSH between the Environmental Protection Agency and the
American Petroleum Institute.
                             111

-------
                     TABLE OF CONTENTS


CONCLUSIONS                                       1

RECOMMENDATIONS                                   3

INTRODUCTION                                      4

PLAN OF OPERATION                                 5
          Report by Others                        6

BIOLOGICAL STUDIES                                g
          Overall Study Discussion

EXPERIMENTAL STUDIES                             1Q
          Characteristics of Test Compounds
          Study Systems                          ,,
            a)  Activated Sludge
            b)  Respirometer System              14
            c)  Difficulties in Experimental     ,_
                  Studies

INITIAL STUDIES                                  2Q
          Study Purpose
            a)  Activated Sludge Studies         20
          Sludge Settling Studies                23
            b)  Respirometer Studies
            c)  Experimental Results             25
          Effect of Emulsifying Agents and       „,
           Temperature
          Benzene Derivatives Study              27
          Activated Sludge Performance Under     „„
           Moderate Loading Conditions
          Reexamination of Crankcase and         __
           Return Waste Oil
          Additional Vegetable Oil Studies       37
          Ancillary Studies                         42
          Detention Time Studies                 42
          Acclimatization to Crude Oil           49
          Shock Loading Studies                  49
          Industrial Waste Studies               54
                              v

-------
                     TABLE OF CONTENTS
                         (continued)
GENERAL DISCUSSION OF TEST DATA                    55
          Biological Considerations
          Oxygen Transfer Studies                  55
          Sludge Settling Characteristics          56
            a)  Study Measurements                 57
          Efficiency of Removal of Hexane          ^
           Soluble Materials

SUMMARY                                            60
                      STUDY SECTION TWO
    REMOVAL OF OIL COMPOUNDS BEFORE BIOLOGICAL TREATMENT

          Coagulation and Adsorption              g,
            a)  Theory and Practice
                1.  Coagulation                   61
                2.  Adsorption                    62
            b)  Experimental Results              64
          Flotation                               65
            a)  Theory                            67
            b)  Experimental Results              68
            c)  Filtration Study                  70
          Discussion of Results                   72
          Study Implications with Respect to
           Conventional Activated Sludge Treat-   73
           ment

ACKNOWLEDGMENT                                    75

REFERENCES                                        76

GLOSSARY                                          78

                          APPENDIX

  1.  Theory of Biological Treatment              80
  2.  Gas Chromatography Study                    89
  3.  Rapid Chemical Oxygen Demand Test          100
  4.  Dissolved Oxygen Uptake Rate               101
  5.  Sedimentation                              102
  6.  Flotation Procedure                        105
  7.  Standard Jar Test Procedure                108

REFERENCES                                       109
                             VI

-------
                        LIST OF FIGURES

Figure
Number
Page
   1      SCHEMATIC  OF BIOLOGICAL UNIT             12

   2      DIFFERENTIAL RESPIROMETER SYSTEM WITH    , ,-
          SINGLE  REFERENCE FLASK

   3      WARBURG APPARTUS                         16

   4      COMPARISON OF SETTLING CHARACTERISTICS   22

   ,-      SCHEMATIC  REPRESENTATION OF WARBURG      24
          RESPIROMETER
   ,      TEMPORAL HEXANE PLOT - CRANKCASE         -,,
   b       (12/24  - 2/3)

   7      TEMPORAL SUSPENDED SOLIDS PLOT -         -, -
          CRANKCASE  (12/24 - 2/3)
   „      TEMPORAL HEXANE PLOT - WASTE OIL         ->-.
   b       (1/17 - 2/3)
          TEMPORAL HEXANE PLOT - CRUDE OIL         ,.
   y       (3/5 -  4/16)
          TEMPORAL SUSPENDED SOLIDS PLOT - CRUDE   -. ,.
          OIL AND CONTROL UNIT (2/25 - 4/16)
          TEMPORAL HEXANE PLOT - VEGETABLE OIL     .,,
  1J-       (12/24  - 2/3)                            Jb
          TEMPORAL HEXANE PLOTS - CRANKCASE        _
  12       (2/25 - 4/16)
          TEMPORAL SUSPENDED SOLIDS PLOT -         _.
  1J      CRANKCASE  OIL (2/25-4/16)               y
          TEMPORAL HEXANE PLOT - WASTE OIL         .
  14       (2/25 - 4/16)                            4l)
          TEMPORAL SUSPENDED SOLIDS PLOT -
  15      VEGETABLE  AND WASTE  (12/25 - 4/16)

          TEMPORAL HEXANE PLOT - CRANKCASE         ..
  16      NO. 4  (4/25 - 6/6)                       44
          TEMPORAL HEXANE PLOT - CRANKCASE
  17      NO. 2  (4/25 0 6/6)                       45
                              VII

-------
                       LIST OF FIGURES
                         (continued)
         TEMPORAL SUSPENDED SOLIDS  PLOT -        4fi
 18      CRANKCASE,  NO.  2 & 4  (4/25  -  6/6)
         TEMPORAL HEXANE PLOT - CRANKCASE        47
 19      NO.  5  (4/25 - 6/6)
         TEMPORAL SUSPENDED SOLIDS  - CRANKCASE   .g
 20      NO.  5  (4/25 - 6/6)
         TEMPORAL COD AND SUSPENDED  SOLIDS  PLOT  5Q
         OF WASTE OIL SLUG, SHOCK NO.  1
         TEMPORAL SUSPENDED SOLIDS  PLOT,  WASTE   5,
         OIL  SHOCK STUDY
         TEMPORAL BOD, COD PLOTS, WASTE OIL      52
         SHOCK  STUDY
 9       TEMPORAL HEXANE PLOT, WASTE OIL         53
         SHOCK  STUDY
 oc      SETTLING VELOCITY - OIL VS. HEXANE      -o
 25      LEVEL                                     58
 „,      EFFLUENT SUSPENDED SOLIDS  LOSS          sq
         VS.  OIL  LOADING
 ?7      CHEMICAL STUDIES FOR CRUDE  OIL AND      ,,
         CRANKCASE OIL
 28      FLOTATION STUDIES FOR CRANKCASE OIL     69
 29      FLOTATION STUDIES FOR WASTE OIL         71
A_,      SCHEMATIC REPRESENTATION OF BIOLOGICAL
         TREATMENT PROCESS
A-2      CARBON NUMBER VS. RETENTION TIME (SEC)  91
A_3      RESOLUTION  OF CRUDE AND CRANKCASE  OILS
         WITH AND WITHOUT USE OF A  SOLVENT
A-4      TYPICAL  SETTLING CURVE                  104
A-5      SCHEMATIC OF FLOTATION APPARTUS        106
                            Vlll

-------
                        LIST OF  TABLES
TABLE 1

TABLE 2


TABLE 3

TABLE 4

TABLE 5

TABLE 6


TABLE 7


TABLE 8


TABLE 9


TABLE 10
CHARACTERISTICS OF TEST  SUBSTRATES

SYNTHETIC DOMESTIC WASTE COMPOSITION
FOR A BOD OF 100,000 MG/L

INITIAL ACTIVATED SLUDGE STUDIES

LONG TERM WARBURG STUDIES ON  THE
INSOLUBLE TEST MATERIALS

EMULSIFICATION AND TEMPERATURE  STUDY
OXYGEN UTILIZATION OF  BENZENE DERI-
VATIVES - GILSON STUDY (2/3-2/6/67)

CONTINUOUS ACTIVATED SLUDGE SYSTEM
PERFORMANCE
ACTIVATED SLUDGE PERFORMANCE  DATA
MODERATE OIL LOADS
EFFECT OF DETENTION TIME ON SYSTEM
PERFORMANCE - CRANKCASE  OIL

DENSITIES OF OIL - ACTIVATED  SLUDGE
MIXTURE
10

13

21

26

28

29

29

37

43

56
                              IX

-------
               CONCLUSIONS AND RECOMMENDATIONS


1.  Oily materials introduced into an activated sludge system
are almost immediately absorbed on the biological floe.  (Page 57)


2.  Excessive amounts of oily materials interfere with the
performance of an activated sludge system by lowering the den-
sity of the floe to the level where the sludge settling pro-
perties are destroyed.  (Pages 55-57)


3.  Within the concentration limits explored in this study,
oily material in a system does not interfere with the ability
of the microbes to efficiently oxidize other substrates in
the system.


4.  Soluble hydrocarbons do not appear to interfere with acti-
vated sludge treatment.   (Pages 27-29)

5.  An activated sludge system will perform satisfactorily
with a continuous loading of hexane extractables of 0.1 pounds
per pound mixed liquor suspended solids.  Questionable stab-
ility develops at a loading of 0.15 pounds HE per pound mixed
liquor suspended solids.  For conventional plant operations,
the influent to the biological system should contain less than
75 mg/1 hexane extractable materials and preferably less than
50 mg/1.

6.  The impact of a particular shock load of oily material on
an activated sludge may be estimated by calculating the re-
sulting density of the system.  In general, a load equivalent
to 5% of the sludge weight under aeration is acceptable, while
a 10% load will cause significant upset in the system. (Pages
49-54)

7.  Each of the oily materials studied could be biologically
degraded to some degree.  The oxidation rate as compared to a
sewage-like substrate was very slow and even after extensive
contact, a significant residue of oily substrate remained.
(Pages 25-26)

8.  Increasing the solids under aeration will improve the
ability of an activated sludge system to withstand a high
loading of oily compounds but wherever possible, the level of
hexane soluble material entering such systems should be mini-
mized.  (Pages 73-74)

9.  This study has shown that in most cases, all necessary oil
separation can be achieved by simple aeration followed by
gravity separation.  These features can be incorporated in the

-------
grit chamber-settling tank designs of most plants.


10.  For highly emulsified systems pretreatment by air float-
ation, with chemical addition in some cases, provided a sat-
isfactory primary effluent for activated sludge feed.

-------
              RECOMMENDATIONS FOR FURTHER STUDY
1.  An engineering study of the modification required to max-
imize oil removal in an aerated grit chamber-settling tank
combination should be considered.


2.  In view of the problems encountered in the analytical
measurements and systems operations aspects of this project,
considerable efforts to develop better methodology should be
the starting point of any further study of this phenomena.


3.  The fate of an oil laden floe escaping to the receiving
stream should be investigated.


4.  Methods of preventing solids loss should be investigated.
Final filtration appears to be the most promising method in
this regard.


5.  The impact of oil materials on the sludge handling portion
of the plant should be studied.

-------
                        INTRODUCTION
This project was undertaken to investigate the capability of
biological systems to assimilate various oils.  It has been
reported that when an activated sludge system becomes over-
loaded with oily materials, a deterioration in treatment oc-
curs .   The conditions under which such adverse behavior occur
are poorly defined.  As a result of real or anticipated prob-
lems,  most municipalities have enacted strict ordinances con-
trolling the discharge of oils and greases.  Typical ordinances
call for the total elimination of any free and/or floating
oils,  while the concentration of hexane soluble materials in
tne flow is usually restricted to 50 mg/1 or less.  There is
little information available upon which to base an evaluation
of the validity and necessity of such restrictions.  Further-
more,  there is little information available on how a treatment
plant will be affected by a loading of oily materials, or how
to operate such a facility to minimize the inherent problems.

The manner in which oily materials cause system failure is of
considerable concern.  The system failure can be either bio-
chemical or physical; that is, the inhibition may occur either
within the microbial population itself, or in the environment
necessary for its performance and life.  If oils are removed
by a biological system, the mechanism by which the removal is
accomplished must be defined to understand the reactions of
the biological system.  Definition of the nature of any inter-
ference is essential.  The effect of oil loadings on oxygen
transfer ability and the adsorptive ability of the sludge must
be considered.  In addition, studies must be conducted to
evaluate sludge coagulation and settling to determine if sys-
tem deterioration occurs in these areas.  Once the nature of
the interference is understood, techniques can be developed
to overcome the problem to the greatest possible extent.  Al-
so, improved methods of pretreatment are considered essential
in that a more efficient removal of the oily materials before
they enter secondary treatment will aid in preserving the
treatment efficiency of the system.


Project Objectives


The present study was undertaken in an attempt to generate
useful data on the operation and improvement of biological
systems subject to continuous or intermittent discharges of a
variety of oily materials.  The project objectives were:

          1.     To determine the tolerance of municipal
               and industrial waste treatment systems to
               various oily materials.

-------
2.      To determine the nature of the inter-
     ference of oily materials with biological
     treatment.


3.      To determine the fate of oil materials
     in biological systems above and below
     the tolerance limits of the systems.


4.      To study methods of operation which
     would improve the ability of biological
     systems to assimilate oils.


5.      To examine methods of improving pre-
     treatment of oily materials in municipal
     and industrial waste systems.

-------
                      PLAN OF OPERATION
Four oily materials, spent crankcase oil, vegetable oil, re-
finery waste oil, and crude oil were chosen as the primary
test substrates.  Spent crankcase oil and vegetable oil find
their way to most sewage treatment plants as the result of a
variety of municipal operations, while waste oil or crude oil
would only reach a municipal plant as the result of an indus-
trial accident or system failure.  Some effort was directed
toward identifying any potential treatment plant problems as-
sociated with benzene or its derivatives.

The main course of the study involved the operation of small
scale activated sludge units at conditions similar to normal
domestic plants and measuring and observing the impact of var-
ious concentrations of the test substrates on this performance,
In addition to such studies, a variety of batch and semi-con-
tinuous systems studies were conducted to develop supporting
information.

It became obvious through the biological studies that sieable
slugs of oily materials should be precluded from entering bio-
logical treatment.  The second phase of the study concerned
itself with the possible pretreatment method for oil removal.
Both physical and chemical methods of agglomeration and separ-
ation were conducted.

Reports by Others


The literature concerning the fate of hydrocarbons in biologi-
cal systems is sketchy, and in some cases, contradictory.
Certainly a good part of the confusion results from the diffi-
culty in measuring the presence and persistence of the hydro-
carbons by the tests conventionally employed in sanitary en-
gineering studies.  Although such investigators as Ludzack,
       (18 19 )
et.al.,   '    have observed degradation of hydrocarbons rang-
ing from between 50 and 80 percent in aerobic systems, it is
questionable whether the investigator observed complete oxi-
dation of mateirals, or only a partial oxidation of the study

substrates.  Benger    reports that a large percent of the ob-
served change does not result in ultimate degradation of ma-
terials but simply in modification.  In all cases, it is ap-
parent however, that complete degradation of materials is not
obtained in conventional systems.

"Oil wastes which escape treatment enter the stream primarily
as emulsified materials.  The oil globules are then trapped in
the flocculent biological solids, conglomerate and settle.
The sedimentation results in the formation of partially decom-
posed beds beneath the water surfaces.  As warm weather occurs

-------
and the beds become anaerobic, gasification of the sludge
tends to lift the oil and solids back into the flowing water.
This action results in a general lowering of water quality and
in some cases, in the collection of these materials at shore-
The principal concern with the collection of significant in-
formation on the fate of hydrocarbons is the lack of suitable
methodologies to determine the concentration and type of hy-
drocarbons present either in the waste or in the sludge.  Con-
ventional BOD and COD techniques have been employed by several
investigators.  The results obtained have been at best, par-
tially indicative of the concentration of materials; at worst,
the materials have shown little response to the BOD or COD
tests.  Techniques such as adsorption on an activated carbon
with subsequent extraction or collection on inorganic solids
have proved more successful, but have limited applications in
cases concerned with activated sludge.  Spectrophotometric
techniques have been used successfully in some cases.

Studies by McKinney     have shown that specific oily sub-
strates will undergo oxidation to a rather high degree, if the
proper conditions are maintained.  These conditions are how-
ever, significantly different from those normally encountered
in a municipal sewage treatment plant.

Anaerobic treatment has been applied in various  studies, with
little or no degradation of hydrocarbons being observed.  In-
vestigators disagree on the nature and extent of interferences
caused with anaerobic digestion.  Here again, it is probable
that partial degradation of the materials occurs; whether or
not significant reduction in overall hydrocarbon concentration
is obtained is highly questionable.

There is significant evidence that emulsions holding the oil
within the waste are broken during biological treatment.  Ex-
perience by several investigators has shown that the inclusion
of secondary skimming devices in industrial plants can be of
major aid in controlling the discharge of oils to the receiv-
ing stream.  It has also been found that oils and similar ma-
terials accumulate in the biological sludges that are returned
to the aeration tanks.  Most investigators report difficulty
in obtaining accurate measurements of oily materials bound
with the sludge.  This is principally due to the difficulty
in separating the oils from the sludge floe.  Special sampling
techniques are required if proper, proportional, samples are
to be obtained.

A large amount of information is available concerning physical
and chemical treatment for the removal of oils.  By far, the
most popular method employed today is the use of an API separ-
ator; design criteria for these units have been widely published,

-------
The units are, in general, quite effective and succeed in re-
moving floating oil with the efficiency dependent on the type
of material encountered.

Use of cnemicals to break oil emulsions is widespread.  Signi-
ficant information on the efficiency of alum and lime in aid-
ing such processes, the use of polyelectrolytes, absorbent
clays, demulsifiers,  etc., is available in the industrial
journals.  Air flotation is used in many cases to achieve
higher efficiencies.   In this type of treatment, difficulties
are continually encountered in obtaining proper measurements
of the efficiency of  the treatment, due to the nature of the
material being tested.

In summary then, the  literature presents significant informa-
tion on the physical  and chemical treatment of oily materials,
but little data is available on the ultimate fate of such ma-
terials in biological systems.   There are indications, however,
that aerobic treatment may be successfully employed to at least
partially degrade these materials.

-------
                BIOLOGICAL TREATMENT STUDIES
Overall Study Discussion


The majority of the biological experiments involved the study
of continuous, laboratory scale treatment systems simulating
conventional activated sludge treatment plants.  Normal sys-
tems analysis was supplemented with off stream analyses, such
as oxygen utilization rate measurements, settling studies,
and respirometer investigations.

After an initial study of the test compounds and the applica-
bility of methods of analyses, a group of experiments were
conducted to determine the nature of the reaction between oily
materials and activated sludge.  The resulting data were sup-
plemented with information describing the biodegradability of
the study compounds and the effects of temperature and emulsi-
fication on system performance.

Having derived from the first phase of the studies an under-
standing of how systems will behave, detailed investigations
to analyze the influence of oil loading rate, detention time,
acclimatization, and waste composition were conducted under
varying system conditions.

The final set of experiments considered the behavior of the
systems under shock loading conditions and the comparison of
performance of an industrial waste treatment system to domes-
tic performance.  The study also considered the possible de-
velopment of a culture that would treat the oily materials
alone.

-------
                    EXPERIMENTAL STUDIES
Characteristics Of Test Compounds


The froup of test compounds included a crude oil, a spent
crank case oil, a refinery waste oil, and a vegetable oil.
These oil compounds were chosen because they are relatively
insoluble materials that are considered to be extremely dif-
ficult to remove in biological systems, and are suspected of
causing adverse effects on treatment plant performance.  In
addition, certain benzene derivative compounds were investi-
gated on a preliminary basis.  These materials were selected
since they are found in the raw waste streams entering treat-
ment facilities.

Initially, these test materials were examined for their phys-
ical and biochemical characteristics.  The analyses included
chemical oxygen demand, biochemical oxygen demand, solvent ex-
tractability,  total oxygen demand,  density, and viscosity -

The chemical oxygen demand (COD) and biochemical oxygen demand
(BOD) were carried out employing the methods described in
"Standard Methods for the Examination of_ Water and Wastewater"
--12th Edition.The total oxygen  demand associated with the
test materials was measured employing respirometer system
studies and involved monitoring, for at least one month, the
oxygen utilization of a known weight of oily materials.  A de-
tailed description of this study is presented in a subsequent
section of the report.  The density of the oily material was
determined by  weighing out a known  volume of material at a
given temperature.  A Cannon-Fenske viscometer was used to
determine the  viscosity of the oils.  Table 1 summarizes the
results from these analyses.
                          TABLE 1

             CHARACTERISTICS OF TEST SUBSTRATES
Test Materials
BOD
                    1
COD
         2
           3
Viscosity   Density
Total" Percent
Oxygen Hexane
Demand Soluble  (centistokes) (gms/ml)


Synthetic Sewage .58
Crank case oil
Crude Oil
Refinery Waste
Vegetable oil
Benzene
.39
.45
.27
.48
.60

0
1
2
0
2


.70
.98
.62
.98
.37
(6)

0.
1.
1.
1.
2.
1.

72
16
03
21
05
77
(4)

82
71
50
89

(5)

88
83
61
96



102
27
5
76



.6
.6
.0
.0



0.
0.
0.
0.



899
846
840
909

                             10

-------
                           TABLE 1
                          (continued)
Test Materials
                 BOD;
COD
                          2
                                 T
Viscosity   Density
Total" Percent
Oxygen Hexane
Demand Soluble (centistokes)(gms/ml)

Toluene
Toluene
Benzoic Acid
Benzaldehyde
Nitrobenzene
* ' /T -*- -^ -m <-. r>/^r^

1.00
1.00
1.38
2.45
0.41


1.40
1.40
1.96
1.45
0.32

(4) (5)
1.80
1.80
1.67
2.08
1.20

, 4v
, ,-»
             ^                .
    grams COD/gram substrate.
    grams total oxygen/gram substrate
    recovered from water
    recovered directly
    erratic data.
Vegetable oil had the highest density and was most readily de-
graded by hexane extraction.  The refinery waste oil had the
lowest density, was the least viscous, and was the most dif-
ficult to extract in hexane.  The other tests showed that the
vegetable oil had the highest five-day BOD, while the crude
oil had the highest COD.  In addition, the long term Warburg
experiments demonstrated that metabolizing the vegetable oil
required more oxygen than the other compounds.  The COD re-
sults for benzene were very erratic, demonstrating the limi-
tation of this test as a method for monitoring benzene.
Study Systems
                      Activated Sludge
The majority of the experiments involved the comparison of
performance in related small scale activated sludge systems.
A typical system is shown schematically in*Figure 1, and con-
sists of a 6.8 liter cylindrical aeration tank followed by a
solids clarification unit.  Diffused air was pumped continu-
ously into the base of the reactor to provide both turbulent
conditions and a source of oxygen.  The test oil material and
sewage were fed to the aerated vessel at the top of the tank.
The effluent line from the reactor was placed in the middle of
the reactor, four inches from the bottom, and served to mini-
mize any short circuiting in the system.  The biochemical pro-
ducts leaving the reactor flowed into the clarification unit
                             11

-------
         INFLUENT
N
        FEED
      RESERVOIR
   (SYNTHETIC MIXTURE)
                         AERATION
                         CHAMBER
                              \
                               '•A Iff SUPPLY
                                              OIL RESERVOIR
                                                       EFFLUENT
                                                    ^SETTLING
                                                    CHAMBER
                                     \_ SLUDGE
                                       HE CYCLE
                                                                   EFFLUENT
                                                                   RESERVOIR
               SCHEMATIC  OF BIOLOGICAL  UNIT

                            FIGURE  I

-------
where the heavier microbial floe settled out.  The microorgan-
isms were recycled to the reactor continuously, and could be
wasted from the system, if desired.  The clarified effluent
was collected in a graduated reservoir to check the volumetric
flow rate and to examine the effluent characteristics.  In
addition to the usual measurements to evaluate performance in
these systems, special attention was directed toward the oxy-
gen utilization rate and settling properties of the sludge.
The method of analysis employed in these investigations is
presented in the Appendix to the report.

Instead of using an "actual" domestic waste as the primary
food source, a synthetic mixture was prepared daily.  This
mixture was selected in order to minimize the problems of
concentration variance and the possibility of objectionable
compounds interfering directly with the system performance.
The inconsistent composition of a domestic waste due to un-
expected disturbances, such as shock loads and normal varia-
tions in the domestic load, could dampen out oil effects on a
biological system.  The composition of the synthetic mixture
is given in Table 2, and includes sources of protein, carbo-
hydrates, amino acid, ammonia, phosphate, and other basic
nutrients:
                          TABLE 2

            SYNTHETIC DOMESTIC WASTE COMPOSITION
                  FOR A BOD OF 100,000 MG/L*
                  Ingredient
grams/liter
skim milk
peptone
gelatin
soluble starch
urea
disodium hydrogen phosphate
KC1
CaCl2
MgSO^
Fe (§0 )
NH^Cl
48
48
16
32
8
8
1.12
1.12
0.80
0.80
8
         *The characteristics of synthetic sewage
         have been presented in Table 1.
Samples were diluted to the strength shown in the various
studies.
                             13

-------
                    Respirometer Systems


In order to explore certain phenomena that could not easily be
investigated on a continuous basis, respirometer studies were
performed.  The respirometer measurement systems utilized in
the project included the Gilson system and the Warburg system.

The Gilson respirometer shown in Figure 2 is described by its
manufacturer as "a differential respirometer, in which envi-
ronmental fluctuations can be minimized by the manifold con-
necting the reference flask and the sample flasks".  The sys-
tem is designed to eliminate the variations in readings due to
temperature and atmospheric pressure changes.  Pressure diff-
erences from oxygen utilization in the sample flask are moni-
tored using an internal manometer connected to a micrometer.
This device is capable of measuring a pressure difference of
one microliter.  A valve on the manifold and one connecting
the two' legs of the manometer regulate the resupply of oxygen
to the sample flasks.  The manometer is constructed of plastic
and uses a mixture of kerosene and Sudan Red III as the mano-
meter fluid.  A system of capillary tubing connects the mano-
meter to the sample flask.  The flask is controlled from var-
iations in atmospheric temperature by a thermally controlled
water bath.

The Warburg system shown in Figure 3 is the standard manomet-
ric apparatus used to measure oxygen utilization. The pressure
difference or oxygen utilization is measured directly from the
graduated manometer which uses "Brodie" solution as the mano-
meter fluid.  A pressure difference of 1 mm can be measured in
this system.  The flask is refurnished with oxygen by opening
the valve at the top of the manometer unit.  A water bath sys-
tem controls the temperature within +_ 0.1° C.

The basic difference between the two respirometer systems is
that for the Warburg apparatus, the sample flasks are indepen-
dent from the other flasks in the system.  Control flasks con-
taining only distilled water with potassium hydroxide in the
center well are used to determine the changes in environmental
conditions.  These observations are then subtracted from the
sample flask readings in order to yield the true biochemical
uptake.

Calculation of the oxygen uptake rate and accumulation was
carried out using an IBM 1130 computer.  Programs were devel-
oped to compute both the rate of oxygen utilization for an
average period of time and the total oxygen demand.  For a
small increment of time, the rate of oxygen utilization was
determined from the slope of a straight line least square an-
alysis of the pressure differences observed for this time
period.  These rates were then plotted as the dependent var-
iable in an X-Y coordinate system with the corresponding


                              14

-------
                                                             ATMOSPHERE CR GAS
                                                         WATER L^VEL
                         FIGURE 2
DIFFERENTIAL RESP1ROMETER SYSTEM VSITH SINGLE REFERENCE FLASK

-------
                  CONTROLLED  TER-'.PERATURE
                   WATER BATH
                                      GROUND
                                      GLASS
                                      FITTINGS
                                        CENTER
                                        WELL
                                        (KOH)
MANOMETER
FLASK 8
SUPPORT  IN
PLACE
        SHAKER MECHANISM WITH
        ATTACHMENTS  FOR
        MANOMETER SUPPORTS
                                                          GAS/VENT
                                                    ij \*?  PORT
                                                          GRADUATIONS
• BROOIES
 FLUID
                                                         ADJUSTING
                                                         SCREW

                                                         RESERVOIR
                                              MANOMETER
                                                  a
                                            SUPPORT ASSEMBLY
                           WARBURG APPARATUS
                                    FiGUSc  3
                                     16

-------
cumulative time period as the independent variable.  The area
under this curve, computed using the trapazoidal rule, was the
cumulative oxygen demand.


          Difficulties In The Experimental Studies


The principle reason so little information is available on the
behavior of oily materials in biological systems is the extreme
difficulties involved in working with these compounds and in
measuring them.  In order to understand the method of procedure
employed in the studies and to evaluate the validity and limit-
ations of the experimental results, it is necessary to fully
appreciate the experimental problems associated with the test
materials.

The principal study difficulties were encountered in two areas.
The first area was the method of simulating the introduction
of low concentrations of the high molecular weight compounds
into the test systems.  Since the materials were insoluble,
they could not be mixed with the basic sewage in the receiving
well.  If this were done, the materials would tend to collect
on the walls of the reservoir, or at the liquid surface, and
would not enter the system under any known or definable flow
rate.  To overcome this problem, it was decided that the test
oils would be introduced into the system separately from the
sewage.  The method that was chosen was to drip these mater-
ials directly into the aeration chamber.  Investigations
showed that, if the flow rates could be controlled at a low
enough level, the materials would become involved in the mixed
contents of the tank.  The principal concern, however, was the
feed rate.  For example, considering the volume of the aera-
tion tank as six liters, one would calculate that the average
sewage flow for a six hour detention time is 1.0 liters/hour.
Under these conditions, an oil feed rate of 1 ml/hour is
equal to approximately 900 mg/1.  Feed rates below 1 ml per
hour were extremely difficult to achieve with any degree of
accuracy or control.  Study of this problem resulted in the
development of a system where the flow rates were controlled
by discharge to the system through capillary tubing.  In order
to overcome the difficulty of the long lengths of capillary
tubing involved, a "spaghetti" type tubing  (Teflon — TFE)
Fluorocarbon Resin, a registered trademark of E.I. duPont de
Nemours and Company) was obtained.  Employing this type of
material in coils it was possible to obtain capillary lengths
as long as twenty to thirty feet.  The oil could then be pre-
pared in a burete and allowed to flow through this capillary
tubing.  Flow rates in the order of less than 0.05 ml/hour
could be measured accurately in such a system.  Operating at
these flow rates it was then possible to simulate low concen-
tration systems.
                             17

-------
Tne volumetric flow rates ranged from 0.7 ml/day to 32 ml/day
and depended upon the diameter (0.25 to 0.5 mm) and length  (6
inches to 20 feet)  of the tubing.  These feed systems were
caecked frequently in order to maintain the designed flow rate
and were constructed in duplicate in order to provide standby
systems in case of failure.

Tne second, and by far the more serious, problem concerned  the
measurement and identification of the hydrocarbons.  Although
the petrochemical industry has developed many analytical
methods, virtually none are applicable to the conditions en-
countered in biological systems.   The method of analysis stan-
dardly employed in "sanitary analysis", the extraction in hex-
ane or a similar solvent with subsequent gravimetric determin-
ation, is inaccurate and provides neither insight into the  na-
ture of change of compounds nor the ability to separate various
compounds one from the other.  Tests such as BOD, COD, and  the
like are similarly of limited value in providing any substan-
tial information on change of the primary study substrates.

A detailed study, described in the Appendix to the report,  in-
vestigated the use of a gas/liquid chromatography as a more
subtle method of analysis.  Although the procedures developed
were quite applicable to low molecular weight compounds, it
was not possible to develop a satisfactory method of separation
and identification of heterogeneous high molecular weight ma-
terials.  The study did develop methods of sample collection
and extraction which can form the basis for a continued inves-
tigation of the phenomena.

Presented with the difficulty of having no direct way to mea-
sure the test compounds, the study relied primarily on indi-
rect measurements related to the biological population.  The
most important indicators were the oxygen utilization rate  of
tne biological mass, and the sludge settling properties.  In
addition, microscopic examination of the microbial population
and measurement of the ability of the biological system to
utilize other test substrates, such as the synthetic sewage,
were employed to describe system conditions.

Finally, an observation must.be presented on the ability to
perform a reasonable mass balance of any of the systems under
study.  The nature of the primary test compounds is such that
any contact with a surface results in adsorbtion of material
ana a subsequent loss from the system.  In systems as complex
as the continuous flow treatment system, normal study proce-
dures resulted in apparent losses of materials as high as 40%.
Even under conditions where extreme care was exercised, as
much as 20% of the test material would be unaccounted for in
the studv.
                             18

-------
The impact of the problems outlined above limits the precision
of measurement with respect to the primary compounds.  It was
possible, however, through repeated experimentation and care-
ful use of preferential experiments to accurately define the
influence of the material on the physical and biological sys-
tems investigated.
                             19

-------
                        INITIAL  STUDIES
 Study  Purpose


 The  goals of the  initial  studies were  to  determine  how signif-
 ciantly  the test  compounds  interferred with  activated sludge;
 to examine how this  interference was caused,  and  if it could
 be eliminated by  system operation.  Questions relating to the
 biodegradability  of  the compounds and  the effect  of emulsifi-
 cation agents and temperature on biodegradability were also
 explored in this  study sequence.


                  Activated Sludge Studies


 Six  continuous systems were set up and acclimated to synthetic
 sewage for two weeks.  After the systems  had  come to balance
 and  were all achieving in excess of 90% BOD reduction,  the
 test substrates were introduced into the  systems.   Vegetable
 oil, crank case oil, and benzene were  studied  and the perti-
 nent data is presented in Table 3.

 In the benzene study, since the substrate was  soluble,  it was
 mixed  with the sewage in the receiving well.   The influent
 strength reflects both the sewage and the test substrate.
 The  results indicated that after a short  acclimatization  per-
 iod, the system performed essentially the same as the control
 unit,  with the exception that the uptake rate was lower.   The
 lower  oxygen use was traced, in part to the stripping of  the
 benzene.

 In the case of the oils,  it was not possible  to add  the sub-
 strate to the sewage feed, so they were introduced  from a  cap-
 illiary drip system.  The influent strength measurements  do
 not  reflect the test substrate effect.

 The  introduction of the oils had an immediate  impact  on the
 systems.  This impact was most noticeable in the  settling
 properties  of the sludges.  Figure 4 shows the comparative
 settling data for the control and the vegetable oil units.
 Similar results were obtained for the crank case  oil.

The  loss of settling properties resulted in a very high efflu-
ent  suspended solids from the system,  and complete system
failure.   To maintain treatment,  solids were externally separ-
ated and returned to the  system.   Under such conditions,  the
systems could continue to operate and be observed.  Although
the data is quite difficult to interpret,  it is clear that the
bacterial system did not  lose its ability to assimilate or-
ganics, since the filtered effluents from the oil units are
                             20

-------
                                TABLE 3
                   INITIAL ACTIVATED SLUDGE STUDIES
Control
Unit Number

Sewage BOD mg/1
COD mg/1
Substrate (mg/1)
Concentration
Hexane Solubles
(mg/1)
Detention Time
(hours)
MLSS 2,
Oxygen uptake
rate mg/l/hr
Effluent BODJ mg/1
COD mg/1
1
308
452


6.2
313
50
10
33
Crank
2
308
452
711
585
6.2
2,524
46
14
55
Vegetable
3
308
452
1,100
980
6.2
2,692
65
10
45
Vegetable
Oil
4
308
452
793
710
6.2
2,621
45
61
90
Benzene
5
262
350
50

6.2
1,959
34
13
(2)
Benzene
6
313
346
500

6.2
2,300
37
17
(2)
*  'Filtered samples

(2)
  'Erratic  data.
                             21

-------
NJ
          ki
IOOO

 900

 COO

 700

 600

 50O

    * • /*)
                                                                Z^r  %O   S-i>   Ov
                          COMPARISON OF SETTLING  CHARACTERISTICS
                                            FIGURE 4

-------
quite similar to the control.   (This is not true in the #4
unit, where some interference was observed.)  It is also not-
able that little increase in oxygen use is noted in the sys-
tems.  The test compounds seem to seriously affect the physi-
cal properties of the sludges, but otherwise have a limited
effect.

After approximately four  (4) weeks of operation, these systems
were discontinued in favor of lower loading studies where more
detailed analyses could be performed.


Sludge Settling Studies

Detailed settling studies were conducted on all test units.
The results of these studies as well as all other settling
data collected in the study is considered in a separate sec-
tion appearing later in the report.

                    Respirometer Studies

In addition to the continuous systems studies, batch studies
employing the Gilson and Warburg respirometer were conducted.
The studies included investigation of the ultimate oxygen de-
mand of each test compound; studies into the effect of tem-
perature and emulsifying agent on biological activity; and
examinations of the biodegradability of a select group of
benzene derivatives.  Similar to the BOD technique, a respir-
ometer may be used as a method of determining the overall bio-
chemical strength of a waste sample.  The biochemical pheno-
mena observed in the respirometer flask are explored in the
Theory section of the Appendix.  This discussion emphasizes
the basic mass transport occurring in this type of system.

The continuously mixed biological unit shown in Figure 5 typi-
fies the respirometer flask.  The gas phase contains the res-
ervoir of oxygen needed to metabolize the substrate.  The mic-
robial population, nutrients including oxygen and ammonia, and
substrate are the principal components of the liquid phase.
The cylindrical center well is a reservoir- for potassium hy-
droxide, which is used to absorb the carbon dioxide produced
by the biochemical reaction.  As the bacteria use the test
substrate, oxygen is utilized creating a concentration grad-
ient between the gas and  liquid phase.  This driving force in
the direction of the liquid phase causes a decrease in the
amount of oxygen in the flask.  This change in pressure is di-
rectly related to oxygen utilization.  The rate of oxygen
utilization can be continuously monitored and determined.  The
cumulative utilization of oxygen can also be determined from a
continuous monitoring of pressure differences in a manometric
flask.
                              23

-------
      GAS PHASE
LIQUID PHASE
                                            CENTER V/ELL
                                               (KOH)
                           FLASK
                         FIGURE 5
      SCHEMATIC REPRESENTATION OF WARBURG RESF1ROMETER
                            24

-------
                    Experimental Results


The,experiment involved the examination of:


               1.   the quantity of oxygen utilized
                    per gram of test material added;

               2.   the effect of emulsifying agents
                    on the biodegradation of hydro-
                    carbons ;

               3.   the effect of temperature on re-
                    moval of test material;

               4.   biodegradability of some simple
                    benzene derivatives.

The first series of experiments using the Warburg respirometer
were conducted to determine the ultimate oxygen demand of the
test materials.  The experiment was run for approximately one
month and examined the test materials as sole sources of or-
ganic material for the microorganisms.  In all, three respir-
ometer experiments were conducted for initial concentrations
that ranged as follows:


               1.   crank case oil: 325 to 450 mg/1

               2.   crude oil: 302 to 550 mg/1

               3.   waste oil: 420 to 450 mg/1

               4.   vegetable oil: 265 to 720 mg/1


For each respirometer study, seven flasks were used and in-
cluded a control flask, a seed flask, and five test flasks.
In each flask, a known mass of test material was mixed with a
seed acclimated in the continuously fed oil unit and was sus-
pended with dilution water which contained the basic nutrients,
An emulsifying agent was not provided in this phase of the
study.  The average oxygen utilization is presented in Table
4.  The vegetable oil seemed to be the easiest to degrade;
the ratio of grams of oxygen utilized to grams of oil added
being 2.06.  All the substrates exhibited oxygen utilization
indicating that a fraction of the hydrocarbon was metabolized
in each case; however, the time required for oxidation was
very long.  As a consequence of the limitations described
elsewhere in this report, studies using the gas chromatograph
could not be conducted to determine the degree of biodegrad-
ability of the test material.  Sample sizes of 40 ml limited
                             25

-------
trie use of the hexane extraction procedure as a means of mon-
itoring tae removal of organic materials.
                           TABLE 4

   LONG TERM WARBURG STUDIES ON THE INSOLUBLE TEST MATERIALS
Substrate
Crank Case Oil
Crude Oil
Refinery Waste
Vegetable Oil
Total
Oxygen
Utilized
(mg/1)
466
397
522
913
Substrate
(mg/1)
400
385
430
443
grams
oxygen
used
grams
substrate
added
1.16
1.03
1.21
2.06
Average
Rate
(ppm/hr)
1.34
1.10
1.80
2.52
Maximum
Rate
(ppm/hr)
7.87
4.28
2.33
10.73
Data presented are average values.
Effect Of Emulsifying Agents And Temperature


Considering the insoluble nature of the test materials and the
limited biodegradation observed in the long term Warburg
studies, it was suggested that the addition of an emulsifying
agent may provide a means of dispersing the hydrocarbon
throughout the liauid phase and increase microbial activity.
The effect of temperature was also explored with the idea that
increasing the temperature might increase the rate of biochem-
ical activity significantly.

The respirometer experiments were run using the Gilson respir-
ometer at temperatures of 20, 25, and 30°C; the test substrate
was crank case oil and the emulsifying agents were anionic,
cationic, and non-ionic surfactants.  Each experiment was
studied for approximately four days, and the primary goal of
these experiments was to observe any change in the oxygen
utilization rate due to the emulsifying agent and/or the
temperature.

A summary of the three experiments is presented in Table 5.
These data are average values and are based on total oxygen
utilized in four days.  The results show that for a specified
temperature,  the emulsifying agents had only a minor effect on
the metabolism of the test material.  The results also seem to
indicate that the microorganiss preferentially selected the
                             26

-------
emulsifier as the primary source of carbon instead of the hy-
drocarbon.  This data interpretation is supported by the ex-
periments carried out at 25 and 30°C.

An increase in the biological activity was observed with in-
creasing temperature as demonstrated by the increase in grams
of oxygen utilized per gram of material added  (see column 8,
Table 5).  Not only was there an increase in the system with
the crank case oil by itself, but similar observations were
monitored for the emulsifiers alone and in combination with
the crank case oil.  However, the overall increase in the me-
tabolism of the hydrocarbon with the addition of an emulsify-
ing agent was approximately that associated with the normal
increase of biological activity with temperature.

No attempt was made to use non-biodegradable emulsifiers as
addition of these materials could have an adverse effect on
the biochemical reaction.  To reduce the inhibition, long term
acclimation and toxicity studies would have had to be conduc-
ted, and such experiments did not seem warranted at that time.


Benzene Derivatives Study


The final series of respirometer studies were conducted in
conjunction with long term biochemical oxygen demand experi-
ments to determine the biodegradability of several benzene
derivatives.  This step was an initial phase of a possible
study to investigate the problem in a continuous flow system
of benzene based compounds that were observed to be difficult
to metabolize.  Experiments were set up studying the biode-
gradability of benzene, toluene, ethylbenzene, and n-propyl
benzene.  The results presented in Table 6 are typical of the
respirometer experiments conducted.  All compounds were read-
ily biodegradable, especially after seed acclimation.  The
biodegradability and oxygen utilization studies of benzoic
acid, nitrobenzene, and benzyaldehyde were conducted using the
BOD bottle and the results have previously been reported in
Table 1.  Since all the compounds were readily biodegradable,
this series of experiments was terminated in favor of a more
detailed examination of the removal of the insoluble hydrocar-
bons in biological systems.
                             27

-------
                                               TABLE 5

                                EMULSIFICATION AND TEMPERATURE STUDY
                                           (Average Values)
Date Temp Substrate
Initial
Concentration
Oil Emul.
mg/1 mg/1
5/13/69 20 Nonionic
6/03/69





5/27/69
6/03/69





6/25/69
7/02/69





Nonionic + Crank
Anionic
Anionic + Crank
Crank
Cationic
Cationic + Crank
Nonionic
Nonionic + Crank
Anionic
Anionic + Crank
Crank
Cationic
Cationic + Crank
Nonionic
Nonionic + Crank
Anionic
Anionic + Crank
Crank
Cationic
Cationic + Crank
285

256
450

233

263

254
252

256

245

288
320

220
62.5
62.5
62.5
62.5

62.5
62.5
31.3
31.3
31.3
31.3
31.3
31.3
31.3
31.3
31.3
31.3
31.3
31.3
31.3
31.3
Total*
Oxygen
Utilized
mg/1
96
82
54
55
198
52
68
57
88
50
101
(1)
61
67
151
109
56
137
170
79
84
Maximum
Rate
(mg/l/hr)
.92
,67
.71
.69
1.05
.66
.61
.94
1.31
.98
.65
(1)
.64
.66
3.39
,91
.67
1.03
1.15
.87
.76
Oxygen Req'd Gr. Oxygen
For Oil
Removal
mg

(1)

1
198

16

31

51
(1)

6

(1)

81
170

5
Gr,
Oil



.004
.440

.07

.108

.200
(1)

.002



.355
.532

.006
Gr . Oxygen
Gr.
Emulsif ier
1.53
1.31
.87
.88

.83
1.09
1.82
2.82
1.60
3.22

1.95
2.14
4.82
3.42
1.79
4.38

2.52
2.68
(1)  Erratic  data.
 *   Four-day values.

-------
                          TABLE 6

          OXYGEN UTILIZATION OF BENZENE DERIVATIVES
                 GILSON STUDY  (2/3 - 2/6/67)
Substrate „
(
Benzene
Benzene
Toluene
Toluene
Ethylbenzene
n-propylbenzene
Temperature = 20 °C
nitial
entration
mg/1)
125
125
100
100
105
37.5

Total
Oxygen grams oxygen utilized
Utilized grams substrate added
(mg/1)
180
185
53
65
130
23
Total time
1.44
1.48
0.53
0.65
1.70
0.67
(minutes) = 4,300.
Activated Sludge Performance Under Moderate Loading Conditions

The next series of continuous system studies considered the
loading range where oil concentrations were thought to be cri-
tical.  The activated sludge systems were acclimated for sev-
eral weeks, then fed each of the test substrates, spent crank
case oil, vegetable oil, waste refinery oil, and a crude oil.
The systems were adjusted so that each was at the same mixed
liquor suspended solids level and was producing the same fil-
tered effluent.  The overall system data is presented in Table
7, but as in the previous studies, reflects only in a limited
way, the performance of the systems.
                           TABLE 7
        CONTINUOUS ACTIVATED SLUDGE SYSTEM PERFORMANCE

                              Crank
                     Control
                       Unit
            vegetable  Refinery  Crude
Sewage COD *            353
       BOD              323

Concentration of
Substrate added

Material recovered *
as Hexane Solubles

Detention time  (hours)  6.4

MLSS *                2,233
Oxygen Uptake Rate
(mg/1/hour)
50
  328
  301

  226


  185

  6.8

2,397

   46
  3.36
  316

  148


  132

  6.1

2,175

   65
  371
  327

   88


   44

  4.3

2,200

   72
  350
  330

   82


   38

  7.8

2,320

   50
                             29

-------
                           TABLE 7

                         (continued)
                     Control  CJ;aJk  Vegetable  Refinery  Crude
                       Unit       e     Oil       Oil      Oil
Effluent BOD *           20      20       18        20       16
         COD             57      91       57        59       52

mg/1/hour oxygen          Q     25     16 0         0      7.5
for oil utilized
mg/l/hr oxygen used
   mg/l/hr H.E.0.075     0.68        —     0.71
	applied	_____^__	

* measured as mg/1


The influent strength reflects only the strength of the basic
sewage.  Assuming that the uptake rate associated with degra-
dation of the sewage is constant, one can calculate the oxy-
gen use associated with the oil compounds.  This calculation
has been performed and the results are presented in the pen-
ultimate row of Table 7.   In the last row, the oxygen use is
related to the quantity of test substrate added to the system.
Little or no increased oxygen use is observed in the case of
crank case or waste refinery oils.

The behavior of the crank case system can be more thoroughly
understood by examining the oil content of the mixed liquor
and the effluent suspended solids.  Figure 6 shows the results
of a hexane extraction from the mixed liquor and shows an
average content of 220 mg/1 with substantial variation.  The
effluent suspended solids shown in Figure 7 increased contin-
ually, in spite of all efforts to control them and the system
is, for practical purposes, in failure during most periods.

In a similar fashion, the level of oil in the sludge of the
refinery waste oil system rose to a very high level, (Figure
8), and the system failed because of high effluent suspended
solids.

The crude oil system is more interesting.  As seen in Figure
9, the mixed liquor oil level initially reached a very high
level and thereafter, decreased sharply to about 100 mg/1.
Figure 10 shows the impact of this change on effluent suspen-
ded solids.  The solids initially very high, reduce to an
easily manageable level,  as the oil content decreases.

The vegetable oil results are somewhat erratic, the effluent
suspended solids vary randomly about 80 mg/1 while the mixed
liquor oil level, shown in Figure 11, seems to trace some
                             30

-------

§
520

480

440

440

36O

320

28O

24O

2OO
   I2O

    80

    40
                            CRANKCASE OIL (SEE TABLE 7)
                             ? = 6.8HRS
                                 MIXED LIQUOR

                                 EFFLUENT
      24  28
      DEC
            I   5
            JAN
13  17  21  25
29  2  6
    FEB
    TEMPORAL HEXANE PLOT-CRANKCASE (12/24-2/3)
                      FIGURE 6
                         31

-------





¥
§
§
8
rLUENT SUSPEI\
k.


300
280
260
24O
220
2OO
ISO
160
140
120
100
60
40

20
0
-
CRANKCASE OIL
(SEE TABLE 7)



/
/
* f
1 . i
i f i
!i |i i
1 1
^ i! '! '
4 j\ i i
M ' \ 1 ' 1 ' '
, ' | N ' 1 ' 1 1
' 1 1 \ 1 '/\ 1 ' A '
/ , 1 \ I A \ A » ' W '
li 1 \i |A|1''
i ' V \ i **
i ' A \ I
^ M
N 1 y
\ '
J 1 1 1 1 1 1 1 1 1
  24  2
13  17  21  25  29  2
TEMPORAL SUSPENDED SOLIDS PLOT-CRANKCASE (12/24-2/5)

                   FIGURE 7
                       32

-------
00
CO
         fc>
         £
         f°
         ki

         §
         s
260

240

220

200

180

160
 120

 100

 80

 60

 40

 2O

   O
                     I
                24  28
                DEC
I
                                   O
             WASTE OIL
             4, J HR$
                                                   TA&LE
                                         MIXED LIQUOR
                                         EFFLUENT
I
             /    5
             JAN
            13   17  21  25  29
                             2   6
                             FEB
    _X
IO  14  18
                         FIGURE 8  TEMPORAL HEXANE PLOT-WASTE OIL (1/17-2/3)

-------
§
 450

 420

 390


 360

 330

 3OO

 27O

 24O

:, 2/0
j

i  180
\
\  150
   60
                                         CRUDE OIL (SE€ TABLE?)
                                              = 7.9 MRS
                                                 MIXED LIQUOR
                                                 EFFLUENT
      I   5   9
      MARCH
                      17  21   25   29
 2   6
APRIL
IO  14   18  22
           TEMPORAL HEXANE PLOT-CRUDE (3/5-4/16)

                             FIGURE 9
                                34

-------
U)
LH
/c?u
140
120
IOO
eo
^ 6O
vS 40
^ 20
Co /?
^^^
EFFLUENT SUSPEND
r\> ^ o^ 03 c
:y o c> Q o <:

J CRUDE
,' . (SEE TABLE 7)
\ t^
i' » !^
Jin i
' i r,
1 \ n !
I'M
I'll
i ' i i
i ' Jl i i r^
' / » ' ' 1 ' I
/ » / \ "«' V M A n
/ \v \i\l \ \ i\ .
- ' \ y " ^ / u '\
i * W . i v
i i i i i i i i i ii i i
/ 5 9 13 17 21 25 29 2 6 IO 14 18 22 2t
MARCH APRIL
J\ CONTROL
/ \ fl '(SEE TABLE 7)
' \ 'l
^ /' \ !
/i ; \ i
/ ' ' \ 'v ; ^
X ' ' > * M R /' «. / N
/ \ f « / » M n i y \ ,\ / v- »
/V/ I/" I,"'"* "">/• 1, ;\
/ ''< /^-\
/ 1 ^ \l V fc
1*1 I i i i i i i 1^1 i (
/ 5 9 13 17 21 25 29 2 6 IO 14 18 22 2i
MARCH APRIL

•»

t*i
                      FIGURE IO- TEMPORAL SUSPENDED  SOLIDS PLOT-CRL^QCONTRO1'
                                         (2/25-4/16)

-------
Ul
cr>
                                           VEGETABLE OIL (SEE TABLE 7)
                                             T* 6.1 MRS
                                              •A— MIXED LI QUO ft
                                                 EFFLUENT
                                        13   17  21   2^   29
24   28   I
DEC      JAN
10  14   10
                             FIGURE II   TEMPORAL HEX ANE PLOT-VEG,O!L( 12/24-2/3)

-------
erratic pattern.  This phenomenon was"studied in  some  detail  in
the next  study  section.


Reexamination of Crankcase and Return Waste Oil


To determine if it was in fact possible  to treat  the crankcase
and waste oils, a special set of experiments  were conducted.
The experimental conditions were essentially  the  same  as  in
the previous studies.  The influent and  effluent  conditions
are presented in Table 8, while Figures  12 through 15  present
graphic demonstrations of the initial buildup and the  reduction
of mixed  liquor oil concentration with the resulting stabiliza-
tion of effluent suspended solids.  The  uptake rate data  sup-
ports the contention that some reduction in oil content is oc-
curring through biological oxidation.
                           TABLE  8

              ACTIVATED SLUDGE PERFORMANCE DATA
                     MODERATE OIL LOADS
               Control   rp  Refinery  Veget.  Veget.  Veget.
               Control   Case    oil      Qil     oil
Sewage BOD *
       COD

Substrate
Concentration

Hexane solubles *

Detention time

MLSS *


s *

2,
349


7.8
259
353
40
33
7.3
2,278
351
58
29
7.6
2,307
351
33
30
7.5
2,495
295
36
32
6.5
2,165
382
22
20
4.0
1,465
 /,-,/T,   *   Rate 48       52       55      67      57      43
 (mg/1/hour)

Effluent BOD *                                      20      13
         COD       67       52       5.8.      46.      5.6      46

* measured as mg/1.


Additional Vegetable Oil Studies


The data collected during the final tests in this series are
also reported in Table 8.   The study considered a variety of
systems treating vegetable  oil at various solids levels and
detention times.  The study was intended to examine the vari-
ability of response observed with vegetable oil in earlier


                             37

-------
00
           kj
460



420



380



340



3OO



26O



220



 160



MO



IOO



 60 V



 40
                O
                 159
                 MAFtCH
                 13   17  21  25  29
                                                            CRANKCASE OIL (ser TABLE a)
                                                                 -A— MIXED LIQUOR

                                                                 -&— £FFLUENT
2   6   IO
APRIL
1	J

/ /">  f}
ib
        FIGURE 12-  TEMPORAL HEXANE PLOTS- CRANKCASE
                                                                            6)

-------
00



v_

^
~J
o
*O
8
Ci
^
CO
^
CO
.
1
.-J
t
^4^^
\t I

f- IXIX
ISO
I6O
I4O
I2O


IOO

60


6O

40

2O

/o
u
1
/!

-
CRANKCASE OIL
(SEE TABLE @)
J
l\
1 \
1 \ ,
/ n

i
- ' ! \ '"\ 1 r
\ ' + i * , "
. i * i it /i
v ii', i ii
V ', ' "
V III1 i ii
\ i 1 ' \ ! i ' '
^\ l\ * L ' , ' ' ^ ' |
. / 1 ' T ' . 1 1 f i • '\ ' .
\ ' l ' \ ! i ^ I I ' * ' ^
1 ^ . ' ^ . / \ / V ' '
V ^ \ i / ^ 1 y i i \
\ t k , ' a i ^
• V-* *• / A ' •
i 	 i i ii i i ^T i r-8" ^i i i
' • - *• " ...
5 9 13 17 21 25 29 2 6 10 14 12 22
/I ARCH APRIL






















                    TEMPORAL  SUSPENDED  SOLIDS PLOT-CRANKCASE OIL (2/25-4/16)
                                          FIGURE 13

-------
§
520


48O


440


400


360


320 -


280


240


200
   120


    80


    40
                I
                                  WASTE OIL ($K TABLE 8)
                                  ? -  7.6HRS
                              -^r-^-£r-  MIXED LIQUOR

                              -Q-Q-0-  EFFLUENT
       /   5   9
       MARCH
                 13   17  21   25  29
6   JO
APRIL
14
      TEMPORAL HEX ANE PLOT-WASTE CIL (2/25-4/16)

                         FIGURE 14.
                             40

-------

-X
%
^
v—
CN
V»J
• r^
ID
~O ^J
6g
§
<0


120

IOO


80


60

40
2O

n
U
\
<
1
' |
/ 1
,'
f H
f \\
1 l\
/ / \
( 1 1
f 1 1

\ / |A ' ^ '* ' \ /
^ . i/ y \i \ J ^ '
i r -*•*" i i i \ ^ \ \( -i
1 5 9 13 17 21 25 29 %

Y/A$T£ OIL
(SEE TAPLe &J






i
• tJ\
\ ' 1
N ' 1 / \ ^

6 10 14 18 22
MARCH
                                APRIL
TEMPORAL SUSPENDED SOLIDS PLOT~V£6.aWA$T£(g/25-4/I®
                    FIGURE: is

-------
studies.  The systems in the early stages showed the typical
buildup and the decline of mixed liquor oil content and then
settled down to low levels.  The pulsation of effluent suspen-
ded solids continued throughout the study.  The reason for the
seemingly erratic behavior is unknown, but a possible reason
is suggested in the section dealing with sludge settling pro-
perties.


Ancillary Studies

The final series of continuous flow experiments were designed
to explore other phenomena which could provide additional in-
sight into the performance of biological systems.  The exper-
iments involved the following:

          1.   An investigation of the effect of
               detention time on the performance
               of a biological system which is
               fed oil;

          2.   An attempt to acclimate a bio-
               logical population to crude oil
               as the primary substrate;

          3.   A study of the response of a bio-
               logical system to shock loads of
               crank case and refinery waste oils;

          4.   The behavior of a system with a pri-
               mary substrate other then sewage.


Acclimation and sample analysis were conducted employing tech-
niques similar to those previously defined in the other studies.


Detention Time Studies


In the detention time studies, four units and a control were
set up and operated treating crankcase oil.  Three of the units
operated at the same oil loading rates but at varying detention
times, while the fourth unit operated at a higher loading level.
The operating conditions and performance data from the study is
presented in Table 9.  The three units operating at the same
loading rate show substantially the same behavior.  Typical
mixed liquor hexane content data and effluent suspended solids
information is presented in Figures 16, 17, and 18.  After an
initial unsettled period, the systems settled down and provided
an effluent of comparable quality to the control.  Data showing
the variation in the system at higher oil loading is presented
in Figures 19 and 20.  In this case, the system does not adjust
                             42

-------
TABLE 9
EFFECT OF DETENTION TIME ON SYSTEM
CRANKCASE OIL
Detention
Time
(hours)
Control
7.3
7.4
10.4
15.0
6.4
BOD
in
(mg/1)
296
296
296
296
296
COD
in
(mg/1)
315
315
315
315
315
Oil
Added
(mg/1

34.9
53.5
65.9
60.5
Hexane
Solubles
Recovered
(mg/1)

28.6
44.0
54.0
49.5
PERFORMANCE
MLSS
(mg/1)
2,152
1,928
2,093
1,826
2,068
COD
out
(mg/1)
57
69
44
64
150
Effluent
Suspended
Solids
(mg/1)
30
40
31
38
95

Loading
Rate
mg Oil
mg MLSS

0.06
0.061
0.058
0.11

-------

§
52O

480

440

4OO

36O

32O

280

240

200
   120
    80
    40
     0
                              CRANKCASE OIL (SEE TABLE 9)
                              1 = IO.6  HRS
                            A-A-air- MLS'S
                            0-0-0— EFFLUENT
      25  29   3   7   II  15  19  23   27  31   4   8
      APRIL   MAY                               JUNE
      TEMPORAL HEXANE PLOT-CRANKCASE,NO.4 (4/25-6/6)

                       FIGURE 16
                           44

-------

§
52O

480

440

400

36O

320

280

240

200
 120

  80

  40
                               CRANKCASE OIL (SEE TABLE 9)
                                  - 7.4 MRS
                                   MIXED LIQUOR
                                   EFFLUENT
    25  29
    APRIL
                3    7
                MAY
II   15  19  23   27  31
4   8
JUNE
   TEMPORAL HEXANE PLOT-CRANKCASE     (4/25.-6/S)
                     FIGURE 17

-------

120

100

 80

 60
 20
    9   13
    APRIL
            17   21   25
I   5
MAY
13  17   21
IOO

 DO

 CO

 < 0

 20
9 13
APRIL
17
21
25
1
/If.
1Y
5
9
13
17
21
                          25  29
                                                              (SEE TABLE 9)
                                                            CRANKCASE OIL
                                                            H=7.4HRS
10  15   20
                                                               (SEE TABLES^
                                                             CRANKCASE  OIL
                                                             1= I0.6HRS
                      2   6
                       JUN"
                                                                     10
                                                                          \
                                                                         /u
    FIGURE 18-  TEMPORAL SUSPENDED  SOL/PS  PLOT-CRANKCAS!.

-------
s!
1
450

42 O


390

36O

330

3OO


27O

240

2IO

180

150

120


 9O

 60

 3O
  (SEE TABLE 9)
CRANKCASE OIL
     * 6.4 HRS
        MIXED LIQUOR
        EFFLUENT
      '25  29   3   7   II   15  19  23  27   31   4    8
       APRIL   MAY                             JUNE
        TEMPORAL HEXANE PLOT - CRANKCASE
                        FIGURE 19
   (4/25-6/6)
                            47

-------
£*•
00
                                                   CRA NKCA SE OIL (SEE TABU 9)
                                                     ? = 6.4HRS
            FIGURE 2O - TEMPORAL SUSPENDED SOLIDS- CRANKCAS

-------
but continues at an unacceptable level of solids loss even af-
ter acclimization.  This study was continued significantly be-
yond the period of data collection to determine if later ac-
climization occurred, but it did not.

Perusing the test data, one may conclude that in the experi-
ments, oil loading rate more directly controls system perfor-
mance than does detention time.  It should be pointed out,
however, that this observation may have limitations.  In prac-
tice, as the detention time increases, the new sludge cells
produced per day decrease, and this phenomenon, not explored
herein, may have some influence on the efficacy of the system
with respect to oil removal.


Acclimatization To Crude Oil


An attempt was made to develop a culture that would utilize
crude oil as its primary substrate.  The system was stabilized
as a mixture of synthetic sewage and oil for several weeks.
After thorough acclimatization the sewage was gradually re-
moved from the system.  Despite all attempts to control losses
the system solids deteriorated to a low level and a totally
dispersed floe developed.  All similarity to activated sludge
was lost.

It was possible by filtering the organisms from the effluent
to maintain a culture that did have an uptake rate and was
capable of stabilizing the crude oil to some degree.  Unfor-
tunately, the analytical tools to measure these changes were
not available.  The experiment did demonstrate the possibility
of developing a biological culture capable of utilizing crude
oil as a primary substrate but the conditions for maintenance
of optimum life appear to be substantially different from ac-
tivated sludge.

Shock Loading Studies

To investigate system response to shock loading, standard sys-
tems were set up and operated on a synthetic sewage feed.  At
intervals, a shock load of the study substrate was introduced
into the system.  The results for the refinery waste oil typi-
fy the data obtained.  Figure 21 shows the system response
after a shock load equivalent to 0.1  pound of oil/pound mixed
liquor suspended solids.  As seen in Figure 21 immediately af-
ter the load was introduced, the suspended solids and COD of
the effluent rose sharply, but the filtered COD shows little
change.  Within 24 hours the system has returned to normal
operation.  The system was rested 5 days, then was shocked
again and then again after 10 days.  The longer term data is
presented in Figures 22, 23, and 24.  In the first case, the
                             49

-------
140
120
   17
   9 A.M.
                   (5/5-6/6/69)
          UNIT 7- SHOCK SYSTEM (WASTE OIL)
                     SUSPENDED SOLIDS
                                                 SHOCK LOAD O.I LBS. OIL/LBS. MLSS
            e - o - •
           -«--•*•--«-- COD (FILTERED)
           A — A — A— COD (TOTAL)
18
19
  21         22
MONTH OF  MAY
23
24
                                                                                 250

                                                                                 200 _
                                                                                    H
                                                                                 150 LiJ
                                                                                    u.'.
                                                                                 100 '
                                                                                  50 S
25
26
                 TEMPORAL COD a SS  PLOT OF  WASTE OIL  SLUG, SHOCK NO. I
                                          FIGURE 21

-------
   o>
   CO
   Q
   _J
   O
   CO
   LU
   Q
   2
   LJ
   Q_
   CO
   UJ
   Z)
240


220


200


180


160


140


120


100


 80


 60


 40


 20
                  SHOCK SYSTEM (WASTE OIL)
                      5/1-6/6/69
                       t=6.7 HRS
               5
                 13   17
21
25  29
10
           MAY
                                       JUNE
TEMPORAL SUSPENDED SOLIDS  PLOT, WASTE OIL SHOCK STUDY


                        FIGURE 22

                           51

-------
(Jl
NJ
CP
JE


Q
o
o


0
O
CO

h-
2
UJ
13
-J
LJL
U.
UJ
                   200



                   I 50



                   100



                    50



                      0



                   200
               7-SHOCK SYSTEM
                 5/1 - 6/6/69

                  t= 6.7 MRS
                                                           —*—*—*— EFF BOD
d

*
o
o

CO
                                                       EFF COD
          to
o

§
CO
     - + -4-- EFF BOD

     —e—t^- EFF COD
                      TEMPORAL  BOD, COD PLOTS, WASTE  OIL  SHOCK  STUDY


                                           FIGURE 23

-------
    480


    450


    420


    390


    360


_  330


.§  300

LJ
m  27°
D
O  240
         _7-SHOCK SYSTEM.
           (WASTE OIL)
           5/5-6/6/69
LU
   210
X
UJ
3=   180
    150


    120


    90


    60


    30 h
     0
          1	I
         13
        MAY
                    •MIXED, LIQUOR
                     EFFLUENT
17    21   25
                            29    2   .6
                                 JUNE'
10
 TEMPORAL HEXAISJE PLOT, WASTE OIL  SHOCK  STUDY

                     FIGURE 24
                          53

-------
system absorbs virtually all of the test substrate,  (refer to
Figure 24), with almost none appearing in the effluent, except
that associated with the lost suspended solids.  In the second
shock, the load was 0.5 pounds of oil per pound of mixed li-
quor suspended solids.  In this case, the biological system
was unable to adsorb the substrate and it is washed out in the
effluent over the next 26 hours.  The sludge increased to a
maximum in hexane soluble content from which it gradually pur-
ged itself.  The upset in the effluent suspended solids and
COD is apparent for several days but the system does recover.

The third shock is equivalent to 0.10 pounds oil per pound
mixed liquor as in the first case, but is introduced over 28
hours rather than as a point injection. In this case, the re-
turn of the sludge hexane soluble level is somewhat quicker
than before, but all other responses are about the same as in
the previous cases.

Examination of other study substrates produces essentially the
same data.  Although a detailed quantative evaluation is im-
possible, the qualitative assessment of the observations is
consistent with the results obtained for the continuously fed
system.  The response is principally physical with the entrap-
ped oil compound being very slowly degraded.  Shock load, even
very high loads, do not destroy the system but rather disrupt
it for a short while until the density conditions necessary
for proper balance return.


Industrial Waste Studies


A brief investigation was conducted to determine if different
reactions could be expected with a diverse basis substrate,
mainly an industrial waste rather than sewage.  Relatively lit-
tle quantative data was obtained and the study was discontin-
ued when it became apparent that the response of the system
was quite similar to that recounted for the sewage system.
                             54

-------
               GENERAL DISCUSSION OF TEST DATA
The data and observations gathered during the study defined a
most confusing situation.  The ability to clearly handle the
information is further restricted by the limited validity of
certain of the measurements and the inability to define the
nature of changes occuring in the test substrates.  It is
possible, however, to draw certain conclusions from the data
and observations.


Biological Considerations


The oil compounds do not exhibit toxic or inhibiting effects
on the biological culture.  This observation is confirmed by
continued high levels of organic removal of the systems even
until conditions of obvious physical failure, and from de-
tailed microscopic studies conducted continuously throughout
the studies.  The presence in virtually all units of free
swimming and stalked ciliates, as well as a significant roti-
fer population, indicated healthy microbial conditions existed
in each of the systems.

Correlary with the lack of biological inhibition is the sur-
mise that the oil compounds collect in the floe and outside
the cell wall.

The oily materials are each biodegraded to some extent.  Ob-
servations indicate however, that even after substantial times,
significant residue persists.  Unfortunately, it was not poss-
ible to determine the nature of the residue in the present ex-
periments .

Evidence is available indicating that activated sludge will
very effectively adsorb the oily compounds with which it comes
in contact.  The sludge will then proceed to oxidize the com-
pound but at a very slow rate.  The period of acclimation of
the sludge to the test compounds was quite long, the average
period being about two weeks.


Oxygen Transfer Studies

Studies were initiated to determine if the test compounds in-
fluenced the transfer of oxygen to the system.  No noticeable
effect was identified.  This observation is consistent with
the previously stated conclusion that the substrates are ad-
sorbed onto the biological floe very quickly.  The material
thus adsorbed is not free to concentrate at an oxygen transfer
interface and cause degradation of system efficiency.
                             55

-------
Sludge Settling Characteristics
The studies have clearly shown that inoperability of a conven-
tional activated sludge system occurred as a result of a des-
truction of the settling properties of the floe.  Under such
conditions the loss of solids from the system was greater than
the buildup and continued deterioration occurred.

The settling phenomena associated with activated sludge is
somewhat poorly defined.  Experiments have indicated that most
well settling sludges have a density of about 1.016.  If the
density of the floe is reduced to about 1.008 or less, random
turbulence is usually sufficient to maintain the cells in sus-
pension and the settling is greviously impaired.

In the case where activated sludge is contacted with an oil
compound that is adsorbed but not biologically assimilated,
the resulting density for any combination of sludge and oil
can be calculated.  This calculation has been performed and is
presented in Table 10.  It will be noted from the Table that
for any system, undesirable settling properties may be expected
to develop when the oil content approaches 5% of the mixed li-
quor suspended solids in the system.   (Based on the 2,000 to
3,000 mg/1 study range this is a sludge oil content of 100 to
150 mg/1.)  This calculation relates only to the unacclimated
system, such as the batch case.  In the continuous system, a
certain fraction of the oil material is metabolized by the
bacterial system and although complete oxidation does not oc-
cur, some change resulting in an apparent increase in density
occurs.  The continuous, acclimated system is therefore able
to operate satisfactorily at somewhat higher mixed liquor oil
                          TABLE 10

    CALCULATED DENSITIES OF OIL-ACTIVATED SLUDGE MIXTURE
     Sys tern
             Density at indicated percent Hexane Extractables
— — — — - — — U J.UU X.^_l *. . -1 _» . U -LA . _> ^.->.U

A
B
C
D
E
Activated
Sludge
Crankcase
Crude Oil
Waste Oil
Vegetable

1.016
1.00
1.00
1.00
1.00


0
0
0
0


.899
.846
.840
.909


1
1
1
1


.013
.0125
.0120
.013


1.
1.
1.
1.


0116.
010
010
0116


1.010
1.007
1.007
1.010


1.003
.996
.995
1.003


.986
.974
.972
.989
Example:
At 5.0% under
vated sludge
concentration
Note:
Unacceptable settling properties develop below 1.008.
 Crankcase oil (B)  implies a system with acti-
at a concentration of 2,000 mg/1 and a crankcase
 of 25 mg/1.
                             56

-------
                     Study Measurements


Related efforts to develop a suitable method for measurement
of sludge density were not successful, and it was therefore
necessary to use sludge settling rate as a primary measurement.
Each of the study systems were investigated twice a week for
settling rate and sludge oil content.  A typical correlation
of the data is presented in Figure 25.  Although a variety of
correlation techniques have been employed, it is not possible
to relate the settling velocity to the sludge oil content.  A
trend of decreasing velocity with increasing oil content is
generally observed but the data is not sufficiently defined to
quantatively identify the tendency.  For crankcase oil and
vegetable oil, it has been possible to relate a portion of the
effluent suspended solids loss data to oil loading.  This cor-
relation is shown in Figure 26.  A similar correlation cannot
be drawn for the other test substrates due to the limited range
of loading data.  Examination of all available data indicates
that a loading range of from 0.075 to 0.10 pounds oil/pounds
sludge is the range of loading in which settling problems begin
to develop.

Efficiency of Removal of Hexane Soluble Materials

Although not of prime concern, it is germane to consider the
efficiency of the biological system with respect to oil re-
moval.  A detailed examination of the data indicates that all
measurable hexane soluble materials in the effluent were asso-
ciated with suspended solids carryover.  If the suspended so-
lids were filtered out, the hexane solubles remaining in the
effluent approached zero.

It should be noted that the test systems had a submerged dis-
charge from the aeration tank and that in a system where this
is not the case some floating hexane soluble materials may en-
ter the final clarifier or receiving water.
                             57

-------






,-»,
cc.
1C
1 1
L-L_
T
H
O
O
UJ

O

I]
J-
H-
LU
V)





OU


45

40



35
30


25

20


15


10

5
0
— _, _-. .. -
O-II/I2-II/26
A -12/20-2/3
• - 3/11 -4/17
0- 4/25-6/6 -7 MR
+ - 4/25-6/6 -I5HR
X- 4/25-6/6 -IOHR

Q- 4/25-6/6
—
_

Q

e
0 x
f^ CM
+ 0
*
i * A Q
i ~y~
A 4-
X O
—
1 1 1 1 !

0 100 200 300 400 500 6C
MIXED LIQUOR HEXANE CONCENTRATION- (MG/L)
SETTLING VELOCITY VS. MIXED LIQUOR HEXANE CONCENTRATION-(MG/L)
                          FIGURE 25

-------
    1,4

    •1.3-

    1.2 h
GO
GO
O
    1.0
   0.9
   0.8
    0.7
o  0.6
§  0.5
O  0-4

   0.3

   0.2

   0.!

     0
O CRANKCASE OIL
X VEGETABLE OIL
O"AVERAGE CONTROL
                50        100       150       200       250
                   EFFLUENT SUSPENDED SOLIDS  (mg/l)
                            FIGURE 26

                                59

-------
                           SUMMARY
The failure of activated sludge system due to oily materials
is physical rather than biochemical.  The oily compounds are
adsorbed in the sludge floe but are only slowly assimilated.
The results of accumulation of oil in the floe is a lowering
of density which results in a loss of culture settling pro-
perties.  Once the culture settling properties have been im-
paired to the point where the solids loss is greater than the
rate of sludge buildup, a system failure occurs.
                            60

-------
    STUDY SECTION TWO






REMOVAL OF OILY COMPOUNDS




          BEFORE




   BIOLOGICAL TREATMENT

-------
In view of the difficulties associated with biological treat-
ment of various insoluble hydrocarbons it appears germane to
explore pretreatment steps that might b© employed to prohibit
such; materials from entering the biological system in the
first place.  Consistent with the project objectives, those
processes and unit operations compatible with conventional
biological treatment have been given most significant atten-
tion.  Both physical and chemical systems have been studied
and included coagulation, adsorption, flocculation, and fil-
tration.

Coagulation And Adsorption


                     Theory And Practice


Coagulation

Coagulation is employed for the removal of waste materials in
suspended or colloidal form.  Colloids are represented by par-
                              — 5         —7
tides over a siz,e range of 10   cm to 10   cm.  These parti-
cles do not settle out on standing and cannot be removed by
conventional physical treatment processes.  Mixtures involving
the dispersion of one liquid in another are known as emulsions.

Colloidal systems possess electrical properties which create a
repelling force and prevent agglomeration and settling.  Coll-
oids can be either hydrophobic or hydrophilic.  The hydrophobic
colloids possess no affinity for the liquid medium and lack
stability in the presence of electrolytes.  They are readily
susceptible to coagulation.  Hydrophilic colloids exhibit a
marked affinity for water.  The absorbed water retards floccu-
lation and frequently requires special treatment to achieve
effective coagulation.

The stability of a colloid is due primarily to electrostatic
forces and neutralization of this charge is necessary to induce
flocculation and precipitation.  The vast majority of colloids
in wastes possess a negative charge, so coagulation can usually
be induced by the addition of high valence cations.  Coagula-
tion results from two basic mechanisms: perikinetic or electro-
kinetic coagulation,  in which the electrostatic forces are re-
duced by ions or colloids of opposite charge to a level below
the van der Wells attractive force; and orthokinetic coagula-
tion, in which the micelles aggregate and form clumps which ag-
glomerate the colloidal particles.

The addition of high valence cations disperses the particle
charge.  As the coagulant dissolves, the cations serve to neu-
tralize the negative  charge on the colloids.  Microflocs are
formed which retain a positive charge in the acid range because


                             61

-------
of the adsorption of hydrogen ions.  These microflocs also
serve to neutralize and coat the colloidal particles.  Floc-
culation agglomerates the colloids.  In this phase, surface
adsorption is also active.  Colloids not initially adsorbed
are removed by enmeshment in the floe.

A polyelectrolyte is generally regarded as a coagulant aid,
but can serve as a coagulant itself by reducing the effective
charge on a colloid.  Polyelectrolytes are high molecular
weight polymers which contain adsorbable groups and form
bridges between particles or charged floes.  There are three
(3) types of polyelectrolytes: a cationic, which adsorbs on a
negative colloid or floe particle; an anionic, which replaces
the anionic groups on a colloidal particle and permits hydro-
gen bonding between the colloid and the polymer; and a non-
ionic, which adsorbs the flocculates by hydrogen bonding be-
tween the solid surfaces and the polar groups in  the polymer.

This study examined two of the most common coagulants in waste
treatment application, aluminum sulfate or alum, and ferric
chloride.  Aluminum hydroxide is amphoteric in that it can act
as either an acid or a base.  Under acidic conditions it has
the following properties:

                  (Al+3)(6H~3) = 1.9 x 10~3

(pH = 4, 51.3 mg/1 Al   is in solution.)   And under alkaline
conditions:

                    (AlCf)(H+) = 4 x 10~13

(pH = 9, 10.8 mg/1 of aluminum is in solution.)  The ferric
ion has the following properties:

                      (Fe+3)(OH)3 = 10~36

(Insoluble hydrous ferric oxide is produced for a pH range of
3.0 to 13.0.)


Adsorption


Organic material in a waste can frequently be removed by ad-
sorption on an active solid surface.  A solution in contact
with a solid surface has the tendency to condense upon that
surface.  This phenomenon is defined as adsorption.

Removal of pollutants from a solution in an adsorption pro-
cess is dependent on both chemical and physical properties.
Chemical adsorption results in the formation of a mono-molecu-
lar layer of the pollutant on the surface of the adsorbent,
due to forces of residual valence of the surface molecule.
                             62

-------
Physical adsorption results from molecular condensation in the
capillaries of the porous adsorbent.  In general, substances
of highest molecular weight are most easily adsorbed. Chemical
adsorption is rapid in that an equilibrium concentration of
the pollutant forms on the interfacial surface of the adsor-
bent, while diffusion into the adsorbent particles is slow.
The overall rate of adsorption is controlled by the rate of
diffusion of the solute molecules within the capillary pores.

The adsorptive capacity of an adsorbent is dependent on the
characteristics of both the adsorbent and the waste.  The
rate of adsorption varies inversely with the particle diameter
of the adsorbent, increases with increasing concentration of
solute, increases with increasing temperature and increases
with decreasing pH due to the surface charges of the adsorbent.
In addition, the rate of adsorption varies with the time of
contact with the waste.

.activated carbon is the most commonly used adsorbent. In addi-
tion to activated carbon, an evaluation of bentonite and^dia-
toraaceous earth as adsorbents for use in removing the hexane
soluble oils was performed.

In these studies only fresh powder carbon was used and no at-
tempt was made to determine the regenerative capacity of the
carbon.  The large quantity of carbon required to effect the
necessary removals of the hexane solubles in the waste econo-
mically requires the reuse of the carbon for this to be a
feasible treatment alternative.  Further studies would be re-
quired to determine (a) the loss of carbon from a treatment
system, and  (b) the effect the waste characteristics would
have on the adsorptive capacity of the carbon regenerated for
reuse.

Activated carbon can be reactivated for reuse by several
methods.  Some of these methods are:

          1.   controlled incineration of the spent
               carbon;

          2.   caustic or acid wash of the spent
               materials;

          3.   solvent extraction to remove the
               adsorbed material from the carbon;

          4.   use of superheated steam.


The most common method of carbon regeneration, however, is
thermal reactivation.  The spent carbon is dewatered and fed
to a furnace where the adsorbed organic contaminants are
burned from the carbon.

                             63

-------
Bentonite is composed mainly of the clay material moritmor.ill-
onite.  Pure clay is a mass of mineral fragments, strongly
bound together when dry, but separable, — when dispersed in
water, — into myriads of particles, many of which are micro-
scopic in size.

The montmorillonite or bentonite clays may be divided into two
general classes: (1) those that absorb large quantities of wa-
ter, "swelling" enormously in the process, and that have the
property of remaining in suspension in their water disper-
sions; (2) those that absorb only slightly more water than
ordinary plastic clays or fullerr's earths, do not swell not-
iceably, and settle rapidly in this water dispersion.  Type 1
is known as "sodium bentonite", and type 2 is "calcium benton-
ite" .  The dominance of either lime or soda is an important
factor in causing swelling or non-swelling.

The two properties of adsorption and electrokinetic behavior
are the major mechanisms which enable montmorillonite clays to
effect an emulsion break and clarification in treating process
wastes.

The procedure followed in the adsorption studies was as follows;

               1.   the waste samples were mixed
                    with various dosages of
                    adsorbent;

               2.   the adsorbent was filtered from
                    the mixture and the hexane soluble
                    concentration remaining in solution
                    was measured.

Experimental Results


A series of experiments were set up to examine the removal of
oil materials from a domestic influent to which chemical and
adsorptive materials were added to an oil-sewage mixture.  The
concentration of oil studied ranged from 100 to 215 mg/1 in
the mixed system.  The overall results from the experiments
demonstrated that dosages in the range of 1 gram chemical per
gram oil present reduced the effluent hexane extractables
level to less than 30 mg/1.  Based on the results from the
biological units, oil entering a biological system at this
level would have a minor effect on the performance of the
plant.  Two typical experiments are presentd in which crude
oil and crankcase oil were the test materials.

Crankcase oil and crude oil were added to separate units con-
taining a raw waste from a local treatment plant.  This mix-
ture was supplemented with an anionic emulsifier in order to


                             64

-------
provide an emulsified system.  Before settling, the hexane
level of crankcas.e oil and crude oil was 105 and 211 mg/1,
respectively.  The mixture was then settled for two hours and
the supernatant examined for oil remaining.  A settling period
was imposed to simulate treatment in a primary clarification
unit or separate pretreatment device.  The additives studied
were alum, ferric chloride, activated carbon, bentonite, and
diatomaceous earth.  The jar test procedure was used as the
testing method, and is described in the Appendix.  The super-
natent from the adsorption study was filtered through a coarse
media before analyzing for hexane soluble materials.

The results for the two test materials are presented on Figure
27.  In either case system, any of the additives or dosages of
100 mg/1 reduced the oil level to less than 15 mg/1.  Dosages
of only 50 mg/1 were required to have effluent oil level less
than 30 mg/1.  The results ob-tained with the study of vege-
table oil: and crude oil are even, more drastic: these materials
reduced below problem\levels with 25 to 50 mg/1 of any addi-
tion.  Further studies showed that for the range of emulsi-
fants usually encountered in sewage that the chemical dosage
and effluent concentration of oils is independent of initial
oil concentration.  The process appears to be two phased, in-
itial removal ;down to some level occuring easily and with
minimal aid, while the second phase of removal requires chem-
ical assistance.

In the case Qf any chemical addition, considerable sludge
problems are created, the magnitude of the problem varying
with the volume of material being handled.

The coagulation and adsorption studies demonstrated that this
pretreatment operation could be used to eliminate major upsets
in biological :systems.  In application, careful consideration
should be given to the type of pretreatment system based on
efficiency and economics of the operation and the sludge re-
moval and. handling ease.

                          Flotation


The previous experiments suggest that in many cases, a simpler
method of separation might be adequate.  The least complicated
method of separating an oil-water solution is providing a sys-
tem where separation:.,occurs , through the action of gravity
forces.  The oil, being lighter than the water, will rise to
the surface forming a film or layer over the surface.  The
separation can be achieved by gravity alone, if sufficient
time is provided and if the concentration of emulsifying
agents is sufficiently low.  Since the time required for grav-
ity separation is usually too long for conventional applica-
tion, the process may be aided by aeration or dissolved air
flotation.


                             65

-------
                                                            ALUM
                                                            CARBON
                                                           :i OEUTONIfE
                                                            FECL3
                                                            EARTH
     o ALUM
     A CAfWON
    -n BENTONITE
      FECL3
      EARTH
                                                    CRANKCASE OIL
CRUDE OIL
5O      IOO      ISO
    DOSAGE (mg/l)
              2OO
50       100       150
   DOSAGE (mg/l)
200
CHEMICAL STUDIES FOR  CRUDE OIL AND  CRANKCASE  OIL
                         FIGURE 27

-------
Theory


Flotation is used for the removal of suspended solids, oils,
and greases from wastes as well as for the separation and
concentration of sludges.  In the process waste flow a portion
of clarified effluent is pressurized to 40 to 60 psi in the
presence of sufficient air to approach saturation.  When the
pressurized air-liquid mixture is reduced to atmospheric pres-
sure in the flotation unit, minute air bubbles are released
from solution.  The oil and suspended solids are floated by
these minute bubbles, which attach themselves to and become
enmeshed in the floe particles.  The air-solids mixture rises
to the surface, where it is skimmed off.  The clarified liquid
is removed from the bottom of the flotation unit or through an
underflow weir.  When flocculent sludges are to be clarified,
pressurized recycle will usually yield a superior effluent
quality since the floes are not subjected to shearing stresses
through the pumps and pressurizing system, while primary sys-
tem pressurization of the influent is most common.

The performance of a flotation system depends upon having suf-
ficient air bubbles present to float substantially all of the
test materials.  An insufficient quantity of air will result
in only partial flotation of the oils, and excessive air will
yield no improvement.  The performance of a flotation unit in
terms of effluent quality and solids concentration in the
float can be related to an air-solids ratio, which is usually
defined as pounds of air released per pound of solids in the
influent waste.  Since these studies were concerned primarily
with the removal of oil materials, the basic flotation rela-
tionship (given in the Appendix) was modified such that the
effluent oil level leaving the flotation unit was a function
of the air to influent oil concentration.  The relationship
used was:

                  A    l-3SaR(fP - 1)
                  H.        H.


          where:

               H.   =    influent oil concentration  (mg/1)

               A    =    air available for reaction  (.mg/1)

               S    =    saturation value of air in water
                a          3
                         cm /I

               R    =    pressurized volume liters

               P    =    absolute pressure  (psi)

               f    =    saturation achieved
                             67

-------
The primary variables for flotation design are pressure, recy-
cle ratio, influent oil concentration, retention period, and
rise rate.  The effluent oil level decreases and the concen-
tration in the float increases with increasing retention per-
iod.  When the flotation process is used primarily for clari-
fication, a detention period of 20 to 30 minutes is adequate
for separation and concentration.  Rise rates of 1.5 to 4.0
gpm/square foot are commonly employed.  When the process is
employed for thickening, longer retention periods are neces-
sary to permit the sludge to compact.

The principal components of a flotation system are a pressur-
izing pump, air injection facilities, a retention tank, a back
pressure regulating device, and a flotation unit.  The pres-
surizing pump creates an elevated pressure to increase the
solubility of air.  Air is usually added through an injector
on the suction side of the pump, or under pressure on the
downstream side.  Of the total air induced, 30 to 45 percent
will usually be dissolved.

The air and liquid are mixed under pressure in a retention
tank with a detention time of one to three minutes.  A back
pressure regulating device maintains a constant head on the
pressurizing pump.  Various types of valves are used for this
purpose.  The flotation unit may be either circular or rectan-
gluar with a skimming device to remove the thickened, floated
sludge.


Experimental Results


The initial phase of the flotation experiments consisted of
the removal achieved by employing only diffused air for twenty
minutes simulating the behavior of an aerated grit chamber.
The experiments indicated that such a procedure would work
acceptably and reduce the hexane soluble material to less than
30 mg/1, in a system containing no emulsifiers.  In cases
where emulsifiers were present more elaborate systems would be
required.

In the main experiments the test materials were crankcase oil
and refinery waste oil, mixed in a domestic influent that had
been supplemented with an emulsifying agent.  The emulsifiers
added were an anionic compound, a cationic material, and a
nonionic compound.  The mixture, stirred overnight, was then
tested using the flotation procedure given in the Appendix.  A
schematic of the apparatus is also presented in the Appendix.
Basically, the study involved measuring the effluent oil con-
tent for different values of the air to hexane ratio.  A ty-
pical output for crankcase oil is shown in Figure 28.  The
initial oil concentration (H.) for the three emulsifier exper-
iments are given on this Figure.
                            68

-------

-------
Data is also presented for the concentrations of substrate re-
maining after diffused aeration, representing the maximum
gravity separation under test conditions.  Study results indi-
cate that satisfactory results are obtained over the range of
0.2 to 0.5 pounds of air per pound hexane soluble materials
applied.  The anionic emulsificant is the most effective agent
in maintaining the oil in solution.

Similar experiments were conducted on refinery waste oils with
similar results  (see Figure 29).  In this case, the anionic
emulsifier agent inhibited treatment to a very significant ex-
tent and supplemental chemical addition would be required to
achieve a process effluent of less than 50 mg/1.

As in the case of chemical treatment, the effluent concentra-
tion was generally independent of initial concentration and
was principally related to the state of emulsificaton and air
to substrate ratio.


Filtration Study

Although the previous techniques demonstrated excellent remov-
al of the oil material, the unit operations would require the
possible construction of additional units to the existing
treatment facility.  However, it may be impractical to con-
sturct such treatment units and other methods would have to be
employed to remove the oil materials.  One method that could
be used would be a continuously or intermittently revolving
screening device on which a finely meshed synthetic fiber was
constructed.  One would place the system at the primary efflu-
ent so that clogging due to suspended solids would be mini-
mized.  By having the unit operate mechanically, provisions to
clean or change the filter could be developed.

A laboratory study was designed to explore the possible use of
synthetic fibers as a method for removal of oily materials.
The synthetic materials tested included nylon, Dacron,  (a
tradename of E.I. duPont de Nemours and Company), polypropy-
lene, and acrylic fibers of different mesh sizes.  Having
eliminated a number of fibers by investigating the materials
permeability to water, continuous flow bench scale studies
were conducted examining the passage of a synthetic sewage-oil
mixture and a primary effluent-oil mixture through the fibers.
The test oil was crankcase oil and the synthetic materials
tested included an acrylic, nylon, and Dacron* fiber.  The
plastic tank was divided into two sections by two plastic
baffles.  Liquid passed from one section to the other through
a six-inch hole in the center of the baffles.  The baffles
were constructed so that the test fiber could easily be placed
across this hole and sealed into place using the other baffle.
For either mixture, the oil was fed separately into the unit
                             70

-------
1.0

 .9

 .8

 .7

 .6

 .5

 A

 .J

 .2

 ./

  O
                                  Q 19 PPM AN IONIC HI-223 Ho'154
                                  A 30 PPM CATIONIQ Hi - 229 Ho = 63
                                  E3 30PPMNON IONIC Hi-^2SHo:l03
                                  Hi ^INITIAL OIL, CONCENTRATION
                                  Ho-EFFLUENT OIL CONCENTRATION
                                    AFTER  DIFFUSED AERATION
   a
20
                     40       60       80       100      120
                     , EFFLUENT HEXANE  SOLUBLES  (mg/l)
140
           FLOTATION STUDIES FOR WASTE OIL
                       FIGURE 29

-------
using the Teflon* tubing arrangement previously described in
the biological study section.  The hydraulic loading was ap-
proximately 15,000 gpd/square foot.  Mixing was provided using
a diffused air system.  Although the oil separation was highly
successful the filter medium soon plugged.  Attempts were made
to reduce the solids buildup by continuously blowing diffused
air near the fiber and although this procedure improved condi-
tions, the end effect was the same.

A large amount of oil was eventually adsorbed on the fiber,
causing clogging of the test material, and the passage of oil
into the effluent.  Washing the fiber did not completely re-
move the oil and passage of the feed solution through the
washed fabric was more difficult than in the original system.
This condition was observed for both the acrylic and nylon
materials.  The washing procedure involved cleaning the fil-
ter with a mixture of hot water and soap and,was the limiting
restriction on the use of these materials.

Due to the limited time remaining on the project, further
studies were not conducted.  These preliminary studies do in-
dicate the possible use of finely meshed fibers to remove oil
materials in wastewaters.  Extensive laboratory studies invol-
ving examining different flow rates, solids levels, fiber mesh
sizes, cleaning techniques, fiber surface area, and other
characteristics for this type of system would be required to
develop such a system.  Preliminary experiments were also con-
ducted to determine if this type separation could be employed
to capture biological floe escaping secondary clarification
during failure conditions.  The results were highly successful,
capturing essentially all lost floe; however, the clogging
problems caused considerable difficulties.  It is believed,
however, that a successful system can be developed.


Discussion Of Results


The investigations conducted in the first section of this study
indicate that limitation of oily compounds entering an acti-
vated sludge system to less than 50 mg/1 is desirable.  The
experiments conducted in this study section indicate that in
systems containing a low level of emulsificants, normal sewage,
that simple aeration followed by gravity separation will pro-
vide the proper separation.  A combination of an aeration grit
chamber and primary tank with properly designed underflow weirs
would provide such a system.

In the case where significant emulsification occurs, dissolved
air flotation may be employed for separation of oils.  This
process, operating in the range of 0.2 to 0.5 pounds of air
per pound of hexane soluble material applied will provide an
acceptable effluent in all but the most highly emulsified case.
                             72

-------
In; the extreme case, the addition of,any of the chemicals
tested in the.study result in generation of a satisfactory
effluent.

The selection of the final unit operation for any particular
case should also be based on cost considerations, including
evaluation of the additional cost of sludge handling which
results from the new process.

Study Implications With Respect To Conventional Activated
Sludge Treatment


Although the solvent extraction generally employed to describe
the oil content of a sewage has limited validity and may in-
clude a variety of materials other than heavy oils, this mea-
sure will have to be employed until some simple and more ac-
curate method is developed.  The study has shown that the re-
sistance of a sludge to upset by loading with an oily compound
is a function of the quantity of sludge available to adsorb
and' assimilate the substrate and the acclimitization of the
sludge to the substrate.  It is also obvious that the density
and settling properties of the sludge before contact with the
substrate will have a significant influence on the system
performance.

In the case of a shock load the system is unacclimated and the
density conditions control the allowable loading.  An increase
of the sludge hexane level to 5% of the total mixed liquor
weight appears to be the maximum safe loading.  For a one
million gallon flow with usual treatment conditions (2,000
mg/1 MLSS and 6 hours detention time)  a level of 200 pounds is
acceptable while double that level would most certainly cause
problems.  It appears that whether the load is introduced
quickly or over many hours has little effect on the system
behavior.  If successive slugs occur before the system has re-
turned to normal, either through oxidation of the substrate
or by sludge wasting, this effect may be expected to be cumu-
lative.

A system continuously exposed to an oily substrate will be
able to tolerate a loading of 0.1 pounds hexane extractable
per pound mixed liquor suspended solids.  This loading corres-
ponds to a concentration of 50 mg/1 in a 2,000 mg/1 system, or
75 rng/1 at 3,000 mg/1.

In the case of a system that will normally receive higher oil
loads settling will fail to be effective but if a more suit-
able separation system is employed the bacteria may be expec-
ted to perform satisfactorily at much higher oil loadings.

The most satisfactory method of protecting a biological system
against upset by oily compounds is to exclude these materials,


                             73

-------
at concentrations higher than desirable, from the system.  The
methods of primary oil separation are outlined in the report
and present little technical problem.  Within the biological
system, about the only useful modification is the maintenence
of more biological solids under aeration.  The solids level
must, however, be maintained at a level where the sludge set-
tling properties are good.

In extreme cases, supplemental solids separation, combined with
or following secondary settling, can be employed.
                            74

-------
                           ACKNOWLEDGEMENT


This project was conducted by Hydroscience, Inc., under the direction of
Edwin L. Barnhart.  Dr. George J. Kehrberger was project engineer and
Meyer Deitch, Chief Chemist.

The participation of the American Petroleum Institute (API) Committee on
Air and Water Conservation and the Industrial Pollution Control Branch of
the Environmental Protection Agency are gratefully acknowledged.

The investigators particularly appreciated the assistance of Messrs. William
Lacy, Edward Dulaney, and Paul Lefcourt of the Water Quality Office, Mr. Art
Rescorla of the API staff, and Mr. Robert Simonsen of the Task Force of the
API Committee on Air and Water Conservation concerned with the project.

The report is submitted in fulfillment of commitments incurred under a
federal grant - Project #12050DSH.
                                  75

-------
                         REFERENCES



1.     Beychok, M.R. "Aqueous Wastes from Petroleum and Petro-
       chemical Plants", John Wiley and Sons, London, 1967^.


2.     "Biological Treatment of Petroleum Refinery Wastes",
       American Petroleum Institute, New York.


3.     Gloyna and Malina, "Petrochemical Wastes Effect On
       Water", Industrial Water and Wastes, Part I, September-
       October, 1962, Part II, November-December, 1962, Part
       III, January-February, 1963, Part IV, March-April 1963.


4.     Weston, Merman,  and DeMann, "Waste Disposal Problems of
       the Petroleum Industry", in Industrial Wastes,(ed.
       Rudolfs) American Chemical Society Monograph 118,
       Reinhold Publishing Corporation, New York, 1953.


5.     Lamkin and Sorg,  "American Oil Cleans Up Wastes in
       Aerated Lagoons", Petroleum Refinery, May, 1964.


6.     Roth, Helwig and  Hall, "Atlantic Selects Air Flotation
       for Refinery Waste Treatment", American Petroleum In-
       stitute 25th Midyear Meeting, May, 1960.


7.     Benger, "The Disposal of Liquid Effluents From Oil Re-
       fineries", Fluids Handling,June,1964.


8.     Weston and Merman, "Chemical Flocculation Of A Refinery
       Waste", American  Petroleum Institute 19th Midyear Meet-
       ing, May, 1954.

9.     Weston, "Separation Of Oil Refinery Wastes" Industrial
       Engineering Chemical, April,1950.


10.    Ingersoll, "Fundamentals And Performance Of Gravity
       Separation", Petrol Refinery, June, 1951.


11.    McKee, J.E., "Report on Oily Substances and Their Ef-
       fects on the Beneficial Uses of Water", State Water
       Pollution Control Board,  Sacramento, California,  June,
       1956.


12.    Ettinger, M.B.,  "Evaluation of Methods Currently Used to
       Determine BOD, COD and Odor for Industrial Wastes", Sew-
       age And Industrial Wastes, Volume 28, No. 9, Page 11T6T
       September, 1956.


                             76

-------
13.     Moore, W.A., and Ruchhoft, C.C., "Chemical Oxygen Con-
        sumed and Its Relationship to BOD", Sewage And Indus-
        trial Wastes, Volume 28, No. 6, Page 705, June, 1951.

14.     Elkin, H.F., "Activated Sludge Process Applications to
        Refinery Effluent Waters", Sewage and Industrial Wastes,
        Volume 28, No. 12, Page 1112-1116, September, 1956.

15.     Elkin, H.F., Mohler, E.F., Jr., and Kumnick, L., "Bio-
        logical Oxidation of Oil Refinery Wastes in Cooling
        Tower Systems", Sewage and Industrial Wastes, Volume
        28, No. 12, Page 1475, December, 1956.


16.     Melpolder, F.W., Warfield, C.W., and Headington, C.E.,
        "Mass Spectrometer Determination of Volatile Contami-
        nants in Water", Analytical Chemistry, Volume 25, Page
        1453, 1953.


17.     Ettinger, M.B., Moore, W.A., and Lishka, R.J., "Anaero-
        bic Persistence of Phenol and o-Cresol", Industrial
        and Engineering Chemistry, Volume 43, Page 1132, 1951.

18.     Ludzack, F.J., and Kinkead, D., "Persistence of Oily
        Wastes in Polluted Water Under Aerobic Conditions",
        Industrial and Engineering Chemistry, Volume 48, Page
        263, 1956.

19.     Ludzack, F.J., Ingram W.M., and Ettinger, M.B., "Char-
        acteristics of a Stream Composed of Refinery and Acti-
        vated Sludge Effluents" Sewage and Industrial Wastes,
        Volume 29, No.10, Page 1177, October, 1957.

20.     McKinney, R.C., "Microbiology For Sanitary Engineers",
        McGraw-Hill, New York, 1962.
                             77

-------
                          GLOSSARY
BOD - Biochemical Oxygen Demand.  Standard five day test per-
      formed with acclimated seed.

UOD - Ultimate Oxygen Demand.  Estimated from long term dilu-
      tion data.  Includes both carbonaceous and nitrogenous
      reactants unless specifically noted separately.

Mixed Liquor Suspended Solids - The total suspended solids con-
      centration in the aeration tank.  Includes both inert
      and active fractions but is assumed proportional to the
      microbial population.

floe - The individual colonies or groupings within the mixed
      liquor.  Includes microbes, habitat, and detritus.

Hexane Extractable Materials - Those materials which are ex-
      tractable from a substrate by hexane extractions under
      defined experimental conditions.  The test results will
      include a wide variety of materials but is usually
      throught of as describing the concentration of oily ma-
      terials .

Oily Materials - Insoluble or minimally soluble petroleum com-
      ponents, vegetable oils, animal fats and the like.  These
      materials are usually defined by the hexane extractables
      test but the efficiency of recovery will vary significantly
      depending on the nature of the materials.
                             78

-------
                           APPENDIX
1.  Theory of Biological Treatment






2.  Chromatography






3.  Rapid Chemical Oxygen Demand Test






4.  Dissolved Oxygen Uptake Rate






5.  Sedimentation






6.  Flotation Procedure






7.  Standard Jar Test Procedure

-------
               THEORY OF BIOLOGICAL TREATMENT
Biological treatment is used to reduce the organic concentra-
tion in a waste.  It is a process wherein active bacteria mixed
with a waste under suitable environmental conditions reduce the
waste to a more stable form.  When the reaction proceeds in the
presence of sufficient dissolved oxygen, the system is aerobic
and the final decomposition products are carbon dioxide and
water.  Two basic phenomena occur when organic matter is re-
moved by microorganisms.   Oxygen is consumed by the organisms
for erespiration and new cell mass is synthesized.  The organ-
isms also undergo progressive auto-oxidation of their cellular
mass.

Wastes can contain suspended, colloidal, and dissolved organics.
The organic matter is measured by the Biochemical Oxygen Demand
(BOD) or by the Chemical Oxygen Demand (COD).   The BOD may be
defined as the amount of oxygen required by suitable organisms
in the stabilization of a given quantity of organic matter.
Theoretically, an infinite time is required for complete bio-
logical oxidation of organic matter, but for practical pur-
poses, the reaction may be considered complete in twenty days.
The conventional BOD test is a measure of the quantity of oxy-
gen utilized in the first five days of oxidation, under stan-
dard conditions, and is designated as BOD-.  The quantity of
oxygen required to satisfy the twenty day demand is usually
referred to as ultimate BOD .  In the COD determination, or-
ganic matter is converted to carbon dioxide and water regard-
less of the biological assimilability of the substances.  In
the analysis of data, it must be remembered that some materials
which are chemically oxidized will not be biologically oxidized.

Biological waste treatment, then, essentially consists of con-
trolling environmental factors to enable a mixed microbial cul-
ture to use the organic matter as a food source for reproduc-
tion  (synthesis) and energy (assimilation).  In aerobic treat-
ment systems, organisms are suspended in a liquid medium with
the waste to be treated.   Dissolved oxygen is  required by the
culture and sufficient time is allowed for the organisms to
utilize the organics as a food source.  Figure A-l graphically
represents the occurring biochemical phenomena.  The suspended
and colloidal organic matter measured as BOD undergoes an ini-
tial reduction by adsorption to the organisms.

In general, the system will approach some minimum BOD value
rather than zero concentration, due to an equilibrium estab-
lished between the bacteria and their liquor.   The magnitude
of the initial removal is a function primarily of the micro-
bial concentration, acclimatization, and waste composition.
The rate of reaction is a function of temperature, nutrient
level, concentration of both the waste and microbial population.
                             80

-------
                      b
00
                      UJ
                      O
                      O
                      U.I

                      fee
                      cr
                      Q

                      I
                      ID
                      o

                      O
                      UJ
                      K

                      O
                      O
X'iTIAL DOD RE;;10\'AL
                                                                CO
                                                                UJ
                                                                o
                                                                Q
                                                                ID

                                                                GO
                             I LAG [    LOG G,~OV/TH    | D£CL!.\iNG  GROV/TH  |
                            A   6
                                     D
                                                        TIME

                                                      FIGURE  A-I
                              SCHLV.'iATiC RCP.inSft ITATION OF BIOLOGICAL 1REATMLNT fr.OCh'S?

-------
Microbial growth may be considered in the following steps  Csee
Figure A-l): first, a lag period in which the culture adapts
from its previous environment to the present; second, a period
of maximum growth under conditions where unlimited food is
available; third, a period of declining growth where food
availability finally becomes a limiting condition and the or-
ganisms consume previously stored food; and finally, an endogen-
ous phase where under severely limited food conditions, cells
die and are, in turn, consumed so that mass population is re-
duced.  The final mass of organisms is always more than the
initial because certain non-biodegradable materials are gener-
ated during synthesis.  The oxygen utilization rate per unit
weight of microorganisms is low at first, quickly reaching a
maximum.  As the competition for food becomes more acute, the
rate decreases until an endogenous level of demand is reached.
A knowledge of the parameters governing these reactions is
necessary in the design of a biological system to treat any
organic waste.

The biochemical phenomena can be illustrated by the following
general relationships:

                         Synthesis;

 ..organic, . r      -, . rnutrients,  cells r      , , -, , ,__ -, , rTI -•,     ,, *
 Wter 1 +[°xy9en] H^ pQ^ ]   r^  [new cells] + [CO2] + [H2O]     (1)

                       etc.

                        Respiration:

            [cells]+[oxygen]   r2->- C02 + H2
-------
biological systems, such as activated sludge, operate under a
high microbial load such that the reaction rate is only a func-
tion of organic material present and is defined as a first or-
der rate with respect to organic material.  Therefore, the rate
of reaction for Equation (1) can be expressed:
                          rx = K-jL                               (3)
          where:
               L    =    organic material present
                         in the system  (usually ex-
                         pressed as mg/1 BOD5).

               K-   =    reaction rate constant and
                         is a function of temperature
                         only.


Applying similar analysis for Equation  (2), the rate of reac-
tion can be expressed as:

                           r2 = K2M                              (4)

          where:


               M    =    amount of organisms present
                         in the system  (usually ex-
                         pressed as mg/1 suspended
                         solids)

               Y.    =    reaction rate coefficient
                         and is a function of  temper-
                         ature only -


Therefore, the rate of formation for the organic matter, oxy-
gen, and cells can be expressed as:
Organic Matter;

                          dL
                          at = -KiL                              (5)
Oxygen;

                    dO

                             83

-------
Cells:
                                 - K2M                           (7)
          where:
               a1   =    conversion factor  (stoichio-
                         metric coefficient) relating
                         organic matter to oxygen;

               b1   =    conversion factor  (stoichio-
                         metric coefficient) relating
                         cells to oxygen;

               a    =    conversion factor  (stoichio-
                         metric coefficient) relating
                         organic matter to cells.

              minus sign (-) indicates removal


Equation  (5) describes the removal of organic material by
microorganisms.  Clearly in many systems, microorganisms
produce organic by-products that are used by other organ-
isms as food sources.  In this development, the general term,
organic matter, includes not only the initial food sources
but also organic by-products.  However, as previously men-
tioned, the experimental techniques used to measure this
term  (BOD and COD tests) determine the overall organic mater-
ial in the system and not the individual compounds.

Equation  (6) relates the amount of oxygen consumed to the oxy-
gen required for synthesis and the oxygen required for the en-
dogenous respiration of the microbial population.  The accumu-
lation of cells in a biological system [Equation  (7)], is due
to synthesis of new cells as the organic material is removed
and the subsequent respiration of these cells due to the lack
of an exogenous food source in the system.  The three equations
[Equations  (5),(6), and (7)], describe only the biochemical
phenomena occurring in a biological system where the rate of
reaction of Equation  (1) is a function of only the organic mat-
ter present and the rate of reaction for Equation  (2) is a
function of only the organisms present in the system.  For
batch systems, these relationships describe the overall activ-
ity in the unit.  However, in continuous flow units, mass bal-
ances must be carried out on the entire system and the kinetic
expressions are just one term of the total mass balance.

In analysis of biological systems, investigators are princi-
pally interested in relating the amount of organics removed to


                             84

-------
the amount of oxygen utilized and to  the  amount  of ?cells  pro-
duced.  Expressions for these parameters  are  developed  using
Equations  (5),  (6) and  (7), and the following relationships:
                         L = L± - Lr                             (8)
which relate the organic matter and oxygen remaining  (L, O-)
to the initial concentration  (L., 02.) and the amount removed
(Lr, 02r).  The rate of change of tfiese relationships is:
                            dL.   dL
                              i     r
                       dt
               and
                      do    do
Since the initial concentrations do not change with time, the
rate functions can be expressed as:
                         AT     dIV
                         ELk =	£
                         dt      dt
               and
                        dO^     d02r
                         dt   "  dt

Therefore, Equations  (5),  (6), and  (7) can be modified to re-
late these expressions to  removal as  follows:

                           dIV
                           T! = K1L                              (14)


                               dL
                         = aiK;L-gi +  bK2M,



                      dM = aK^£ -KM


These relationships are specific for  the  initial assumptions
imposed on the system for  the definitions of the rates of re-
action, namely Equations  (3) and  (4).
                             85

-------
Along with Equations  (15) and  (16), another relationship  is
used to describe the removal of organic material  in a complete-
ly mixed continuous flow reactor and includes a term for  the
biochemical kinetics.  This relationship for the  steady-state
condition is:

                     QL. - QL - VK,L = 0

               or:
                                        (17)
                        L.
    1 + TK-,
                                                                 (18)
          where:
               L.
               V
volumetric flow rate  (L /t)

inlet organic concentration  (BOD,-)

(Mass/L3)

outlet organic concentration  (BOD,-)

(Mass/L3)

volume of reactor  (L  )

reaction rate coefficient  (I/day)

V/Q hydraulic detention time  (t)
Also, the oxygen required and cells produced due to the removal
of the organics can be expressed:

       #O0 required/day = a1#600- removed/day + b'#S
         £•                      Da

          where:

               a1   =    fraction of the BOD,, removed
                         which is used to provide en-
                         ergy for growth;

               b1   =    endogenous respiration rate
                         coefficient

               S^   =    cells under aeration
                                        (19)
     # cells accumulated/day = a#BOD5 removed/day - bS.
                                        (20)
                             86

-------
          where:

               a    =    fraction of BOD- removed which
                         is synthesized  to new biologi-
                         cal cells;

               b    =    fraction of cells under aera^
                         tion which are oxidized;

               S    =    cells under aeration.
                a

This development demonstrates the continuity between batch and
continuous systems.  Although the time dependent terms are
different, the biochemical kinetic expressions are equivalent
and, therefore, batch studies serve a distinct purpose of
conveniently providing a simple system to estimate the bio-
chemical kinetics and stoichiometric coefficients.  However,
the use of a batch system requires strict experimental techni-
ques but is  (or could be) an excellent technique for determining
the biochemical kinetics.  At present, continuous information
is used to estimate the biological coefficients by employing
the relationships given by Equations  (18),  (19), and  (20).

The present discussion has been based on the premise that the
organic material was in a state which the organisms could
easily metabolize.  However, in some instances, complex organ-
ic compounds such as suspended or colloidal materials enter a
biological system and are subject to degradation.  For organ-
isms to metabolize these materials, enzymes have to be gener-
ated that are capable of attacking these complex compounds and
reducing them to a form which can be passed into the cell.
Hydrolysis of complex materials can be time-consuming and can
easily be the controlling mechanism in the biochemical pheno-
mena occurring.  Therefore, in addition to Equations  (1) and
(2) , another relationship must be used to describe the enzy-
matic process required to breakdown complex compounds.  This
reaction can be expressed by the following general relation-
ships :
                .   •  j_                organic
   jcomplexj   nutnents] + [      j   _^ enzyme] + products         (19)
   lorganicj     oxygen       Y     3  compiex
                        rf  [enzyme] +  [1 +  [products]    (20)
 complex         *

If the rate of generation of the complex material is very low
compared to the removal of organic matter  (i.e., r^ > r, and/
or r . > r, ) , the organic removal will be controlled by the
microorganisms ability to generate enzymes to hydrolyze the
                              87

-------
complex materials.  In some cases involving high weight carbon
compounds, organisms can only generate enzymes that partially
degrade these materials resulting in the formation of an or-
ganic product that is commonly labeled "non-biodegradable".
Acclimation of a microbial population to a waste results in
the formation of the required enzymes.  However, even after
a careful acclimation procedure, some organic compounds are
too complex, so that only partial reduction occurs.

This study explored both phenomena;  that is, the removal in
the same biological system of both soluble simple organic ma-
terials and high weight organic compounds.  The simple organic
material was a mixture of easily degradable organic compounds
(synthetic sewage) fed in parallel with high weight insoluble
materials.  In addition, a control unit fed only the synthetic
unit was monitored for organic removal and oxygen utilization.
By comparing the organic removals and oxygen utilization,
conclusions could be drawn regarding the assimilatory capacity
of the biological population.
                             88

-------
                  GAS CHROMATOGRAPHIC STUDY


Introduction

In an attempt to develop a more acceptable method of identify-
ing the test compounds, a study was undertaken to monitor them
using gas chromatography.  Although the extraction technique
is commonly used in wastewater analysis and yields valuable
information concerning the removal of total oil material, this
method:is too general when information is required concerning
the removal of individual compounds associated with an oil
material.  In addition, the extraction method requires a large
sample volume (at least 100 ml) and this presents a difficulty
in many studies.  A gas chromatograph employing small sample
sizes (1 yl) could be used in all applicable studies.  Infor-
mation gained from a truer knowledge of the compounds consumed
by the microbial population would aid greatly in determining
carbon balances on the system, the oxygen requirements per mass
of carbon removal, and estimating at what point (i.e., at
what carbon level) microorganisms are inhibited from synthesiz-
ing new cells.

Studies

An investigation was conducted to determine the most convenient
methods to measure the individual compounds associated with
the test materials.  A detailed literature search, and discus-
sions with chromatographic research group of several petro-
leum companies yielded two possible gas chromatographic methods.
They are:

1.  Precolumn Technique

          The precolumn would be heated to 200°C and
          a sample would be injected.  All the compounds
          present in the oil sample that vaporize at   . .
          temperatures less than or equal to 200°C would
          evaporate from the precolumn into a dual column
          an would be separated and detected.  The phys-
          ical design of the precolumn and the surround-
          ing valving /and piping would be extremely del-
          icate and would required a grea"t deal of work.
          However, a simplified but informative procedure
          was suggested relating carbon number to boiling
          points.

2.  Boiling Point Technique

          This procedure could be simply characterized
          as a gas chromatographic distillation of oil
                             89

-------
          compounds.  A sample introduced into the gas
          chromatograph is separated into individual
          compounds according to the boiling point.  The
          column temperature is raised at a known rate
          and the areas under the chromatogram and chro-
          ma togram peak heights are recorded.  A calibra-
          tion curve covering the boiling points expected
          from a test sample is constructed from a stan-
          dard mixture of low weight hydrocarbons.  The
          curve consists of carbon number plotted on the
          ordinate axis and the associated retention time
          on the abcissa axis.   Comparison of the initial
          and final chromatogram of the test sample would
          yield the fraction of oil material that has
          been metabolized.

          In view of the diverse nature of the oil sam-
          ples, a detailed analysis of the individual
          compounds would have been extremely time-con-
          suming and would digress from the purpose of
          the study.  However,  knowledge of the fraction-
          al amount of particular low weight carbon ma-
          terials that can be degraded would be very
          helpful and, therefore, the boiling point
          technique was selected.

Studies were carried out testing this technique and establish-
ing an accurate and precise sampling method.  The instrument
was a Perkin-Elmer Model 881 with flame ionization detector
and linear temperature programming.  The columns used in this
study were a 2% UC-W98 Chrom W-DMCS and a 5% UC-W98 Chrom W-
DMCS  (2 feet long).  The first investigation was to obtain
calibration curves for the two columns.  The standard mix-
ture consisted of hexane, heptane, octane, nonene, decane,
dodecane, and ercosane.  Typical outputs are shown on Figure
A-2 for the 2% and 5% columns and demonstrate the better re-
sults obtained on the 5% column.

Studies were then carried out on samples of crude and crank
case oils testing their resolution with and without the use
of a solvent.  Typical results are shown on Figure A-3  (a thru
d) for these two compounds.  The Figure shows that crude and
crankcase oil samples had good resolution of the major peaks
up to 250°C and many components were vaporized before 200°C.
The crude oil sample had a homologous series with isomers,
while the crankcase oil sample did not have any regularity
and probably consisted of more randomly cracked hydrocarbons
and oxidation products.  Comparing the two chromatograms, the
crankcase oil sample seemed to have fewer and smaller amounts
of hydrocarbons based on peak heights and the number of peaks.
No attempt was made to identify the individual components in
                             90

-------
cr
LU
CQ

z5
o
CD
cc
<
o
                                       5%UC-W98 COLUMN
                                       2%UC-W98 COLUK'iM
      0   60   120  180  240 300  360  420 480  540  600
          CA=EO,N  NUMBER VS  RITEMTIO.v


                         FIGURE A-2
                             91

-------
  CRUDE OIL
         A
184°
160
  IN CS2 SOLVENT
112°
8 8°
                       TEMPERATURE (°F)
                       TEMPERATURE (°F)
64°
                                                            X
                                                            o
                                                            UJ
                                                            UJ
                                                            Q.
                           FIGURE A-3A

           GAS CHROMATOGRAPHY RESULTS FOR CRUDE OIL
                               92

-------
U)
                       CS2- SOLVENT
                189°
I65C
141°      117°      93°




   TEMPERATURE (°F)
                                      FIGURE  A-3B
                   GAS CHROMATOGRAPHY OUTPUT FOR CARBON  DISULFIDE

-------
10
                                  I         I

                          CRANK  CASE OIL IN CS2SOLVENT
              225'
175°      151°      127°

   TEMPERATURE (°F)
103?
79e
                                                                            O
                                                                            UJ
                                                                            X
                                                                            UJ
                                                                            O_
                                         FIGURE  A-3C

                GAS CHROMATOGRAPHY RESULTS FOR CRANKCASE  OIL  IN CS2 SOLVENT

-------
ID
Ul
                 CRANKCASE OIL
            219°       195°      I7!c
  147°      123°
TEMPERATURE (°F)
99°      75°       5le
                                       FIGURE A-3D

                     GAS  CHROMATOGRAPHY RESULTS FOR CRANKCASE  OIL

-------
either crude or crankcase oil, since a chromatographic "fin-
gerprint" could serve as an indicator of the oil breakdown in
biological system.

Having tested the crude and crankcase samples, a study was set
up to develop an accurate and reproducible sampling technique.
The tests included selection of the more reliable solvent,
the type of sampler to use, the reproducibility of injection,
the reproducibility of extraction, the stability of extraction,
the effect of large concentration of test materials on the
chromatographic output, and finally, the effect of large sam-
ple volumes on the output.

The first and most crucial step in this study was to develop
an accurate and reproducibly technique to sample the oil-
water interface.  Experiments were conducted to select the
solvent that extracted the largest fraction of the test com-
pound.  Crankcase oil was selected as the test compound.   The
study is summarized as follows:

          1.   A 5 yi of oil was pipetted into 50 ml
               water and mixed.  The concentration of
               the mixture was approximately 100 mg/1.

          2.   A 1.0 ml of the test solvent was added
               to the oil-water mixture.  The test sol-
               vents were carbon disulfide, diphenylben-
               zylamine, and hexane.

          3.   A 1.0 yl sample was extracted from the
               mixture and injected into the gas chro-
               matograph.

          4.   The conclusions from this study were:

               a) Although diphenylbenzylamine has
                  a high boiling point (300°C) and
                  peaks after the compounds of in-
                  terest, it is unsatisfactory be-
                  cause of low boiling compounds
                  monitored in the solvent.  The
                  solvent seemed to contain impur-
                  ities that would have to be re-
                  ved.
               b) Hexane has a boiling point of 69 °C
                  but did not extract a large portion
                  of the components from the crank-
                  case oil when compared to the num-
                  ber of peaks observed from the car-
                  bon disulfide extraction.  In ad-
                  dition, two detectable impurities
                             96

-------
                  were present in large quantities
                  in this solvent.

               c) Carbon disulfide was the best sol-
                  vent of those tested because it ex-
                  tracted more material; had fewer
                  and smaller amounts of impurities
                  (a peak at 140°C); had a low boil-
                  ing point  (46°C) so that other low
                  boiling point components would not
                  be dampened out by the solvent peak.

Tests were also run to determine the reproducibility of dif-
ferent injections of the same extraction and the reproduci-
bility of different extractions.  In addition, studies were
conducted on the volume of solvent, the different'sample sizes,
and other pertinent variables in order to establish a reliable
sampling program.  The results from the experiments were:

          1.   Injections of the same extraction were
               reproducible in general, although some-
               times spurious peaks did appear.  It is
               advisable to inject all samples at least
               twice.  Extracts must be tightly stop-
               pered to prevent evaporation of solvent
               and components.

          2.   Extractions were reproducible but were
               very dependent upon the technique used
               to remove the oil from the water.  Also,
               the initial temperature must be control-
               led, otherwise the programmed temperature
               will be inaccurate and the apparent re-
               tention times of the components will
               change.  In general, greater recovery
               of the oil could be achieved if the oil
               slick were in a small area.

          3.   An extraction with 4 ml disulfide rather
               than 2 ml showed no significant difference
               in recovery.

          4.   Extracts were stable overnight if stored
               in tightly stoppered test tubes.

          5.   Doubling the oil concentration  (i.e.,
               20 ml instead of 10 ml) resulted in peak
               heights two to three times greater and
               therefore had more definable resolution
               of the smaller peaks which were barely
               discernable with a 10  ml of oil.
                             97

-------
Having defined the basic sampling techniaue, a study was set
up to test this procedure in a biological system and to ob-
serve the possible biochemical breakdown of the crankcase oil,
Batch biological units were set up in one liter reaction ves-
sels aerated with diffused air.  The reactors contained an ac-
climated seed, dilution water containing the basic nutrients,
and 200 mg/1 crankcase oil.  The turbulent mixing conditions
caused the oil to adhere to the walls and accurate recovery
was impossible.  Attempts were made to coat the glass walls
with a hydrophobic substance in which the oil would wet the
walls and drop back into the liquid.  Materials tested were
paraffin waxf silicone, stopcock grease, scotchguard* fabric
stain repellent, and Teflon** dry lubricant.  The results dem-
onstrated that these materials exhibit wetting properties
similar to glass.  Other studies included setting up biolog-
ical units in polyethlyene containers but, as before, the
oil adhered to the wall.

To improve the reproducibility of experiments a measurement
system employing the Gilson respirometer was set up to moni-
tor the degradation of crankcase oil.  Complete mixing in
the liquid phase was provided by mechanically shaking the
flasks at a low shake rate (60 cpm) and with this technique
it was assumed that oil adsorption on the cylinder wall would
be minimized.  For this study, the oil was initially weighed
out on glass cover slips and was placed in the respirometer
flasks.  Although some adherance to the walls was observed,
it was possible by extracting the entire flask contents, to
obtain reasonable samples.

When results were obtained, it was found that little inter-
pretation was possible.  The change in the total area was
small, relative to the mass present, and more importantly,
"cracking" appeared to be going on with the chronatograph.

Exploration of this phenomena with experts in petrochemical
analysis concluded that what was observed was quite possible
and in fact, probable.  A variety of possible solutions to
the problem were explored, but in view of the project limit-
ations, no additional studies were conducted.
                             98

-------
                         CONCLUSIONS
-1-'        The qualitative identification and quantitative
          measurement of low carbon number compounds may
          be easily accomplished using gas chromatography.

2>        The isolation of higher carbon compounds require
          special study techniques to overcome "cracking"
          within the chromatograph.

          The project was not able to extract and separate
          oily compounds from biological floe.

          Chromatography shows significant promise as a
          superior technique for such studies, but con-
          siderable further work would be required to de-
          velop suitable methods and procedures.
                             99

-------
3.  Rapid Chemical Oxygen Demand Test

REAGENTS

1.        Standard Potassium Bichromate solution 0.05 N.
          Take 2.4518 grams Y.JCr2O^ primary standard
          grade that has been previously dried at 103°C
          for two hours and dissolved in distilled water.
          Make up solution to 1,000 ml with distilled
          water.

2.        Standard Ferrous Ammonium Sulfate solution
          0.05 N.  Dissolved 19.6 grams Fe(NH4)2(SO4)2.6H2
          in distilled water.  Add 20 ml concentrated
          H2SO., cool and dilute to 1,000 ml.  This
          solution must be standardized against the
          K2Cr207 DAILY.

          Standard!zation;  Dilute 25.0 ml standard dichro
          mate solution to about 250 ml.  Add 20 ml con-
          centrated N2SO4 and allow to cool.  Titrate
          against the ferrous ammonium sulfate using 5
          6 drops of ferrion indicator.
                                             to
3.

4.
5.
6.

7.
                   N =
             ml K2Cr2O7 x 0.05

             ml Fe(NH4)2(S04)2
Sulfuric Acid concentrated.

Ferrion indicator solution:  dissolve 1.485 grams
1.19 phenanthroline monohydrate, together with
0.695 grams FeS04.7H20 in water and dilute to
100 ml.  This indicator solution may be purchased
already prepared.

Oxidizing solution:  dissolved 2.4518 grams of
K2Cr20_ in a mixture of 500 ml phosphoric acid
H3P04 concentrated and 500 ml sulfuric acid
concentrated H2S04.

Silver sulfate.

Mercuric sulfate.
PROCEDURE

A quantity of waste matter or an alicruot of a dilution there-
of according to the concentration of the waste is added to
25 ml of oxidizing solution in a 500 ml flask.  A speck of
                             100

-------
silver sulfate  (catalyst) is added, followed by 1/3 teaspoon of
mercuric sulfate  (in the presence of halogens).  This mixture
is heated to 168°C.  (Caution: this material must not be al-
lowed to heat over  (170°C).  Cool and dilute to approximately
250 ml.  Titrate against ferrous ammonium sulfate using 5 to
6 drops of ferrion  indicator.

A blank consisting  of a volume of distilled water equal to the
sample together with the reagents is run in a similar manner.

CALCULATION


                mg/1 COD =  (a"b)c x 8'000
                             ml  sample


          COD  =    Chemical Oxygen Demand

          a    =    ml Fe(NH4)2(SO4)2 used for blank

          b    =    ml Fe(NH4)2(SO4)2 used for sample

          c    =    normality Fe (NH4>2(SO4)2

Estimate of Sample Volume

Using esimate value of COD and a desired delta titration (a-b)
of 10, then the estimated sample volume would be:


             Estimated sample volume = —. . '' .  -. _ ,^
                          1             estimated COD
Using this sample volume and the procedure above to determine
the actual COD.
4.  Dissolved Oxygen Uptake Rate

          1.   Before initial use, the dissolved oxygen
               probe should be cleaned, standardized and
               allowed to stand in distilled water over-
               night.  The probe membrane should be re-
               placed if punctured or if frequent cali-
               brations or drifting of the meter is noted.
               Usually, the membranes must be replaced
               at two to three week intervals.

          2.   Fill BOD bottle to overflowing with sam-
               ple so that there are no air bubbles en-
               trapped in the liquid.
                             101

-------
          3.    Insert magnetic mixing bar, and place bot-
               tle on magnetic mixer.  Measure initial tem-
               perature of sample.

          4.    Insert dissolved oxygen probe so that no
               bubbles remain in the bottle an airtight
               seal is formed.  Be sure sufficient velocity
               (not less than 1 fps) is maintained past the
               probe head.

          5.    Record the drop in dissolved oxygen reading
               with time, time zero being the start of
               meter drop-off.  Measure the final temp-
               erature of the sample and use average temp-
               erature for determining the saturation
               value of dissolved oxygen for probe cali-
               bration.

          6.    Initial dissolved oxygen reading should be
               80 to 90 percent of oxygen saturation for
               that temperature.  Samples may have to be
               vigorously shaken in the BOD bottle, about
               three-quarters full, to achieve the desired
               dissolved oxygen before the measurement be-
               gins .

          7.    Plot the dissolved oxygen concentration
               (not meter reading) versus time on rec-
               tangular coordinate paper.
               dissolved
                  oxygen
                  (mg/1)
                                      [time]
          8.    The oxygen uptake rate is the slope of the
               straight line formed on the previous plot.
               The units should be converted to mg/1/hour.


^        .  ,     .     change in dissolved oxygen      ,,  ,,
Oxygen uptake rate = 	  change in time       = ^/I/hour


5.  Sedimentation

Solids-liquor separation is an integral part of biological treat-
ment systems.  The primary function of sedimentation is to sepa-
rate biological growths from their associated treated licruor
for return to the aeration process for subsequent disposal.
                             102

-------
In settling,"""the sludge floe agglomerates and the mass parti-
cles settles as a blanket forming a distinct interface between
the floe and the supernatant.  Often, due to solids concen-
tration and the waste characteristics, a "lag" period occurs
during which the sludge undergoes an initial period of floc-
culation.  Figure A-4 is a typical settling curve.  The sludge
settles at a uniform velocity until the settling interface
approaches an interface of critical concentration.  The mag-
nitude of this velocity is a function of the initial solids
concentration.  Free settling  (A) will not exist beyond this
critical solids concentration.  At this point, a transition
zone (B) occurs, and the settling velocity decreases due to
the increasing density and viscosity of the suspension around
the particles.  A compression zone (C) occurs when the floe
concentration becomes so great as to be mechanically sup-
ported by the layers of floe below.  The solids concentration
in the compression zone will be related to the depth of sludge
and the detention of the solids in the zone.

The settling rate is determined from a plot of the height of
the sludge interface with time.  The settling velocity is con-
verted to an Overflow Rate  (O.K.).


             O.R.(gal/day/sf) = V  (ft/hr) x 180
                                 o


The hydraulics of inlet and outlet devices are of significant
importance in the settling process.  The usual limitation,
imposed by the hydraulic consideration, is the overflow rate.
Scale up factors are required to convert laboratory scale set-
tling data to plant scale.  A factor is normally used to re-
duce the laboratory overflow rate to plant scale.
                             103

-------
12 r
          HEIGHT OF
          INTERFACE
          30    60
SO
                i !,V;Z - minutes
                   rIG'JAi  A-4
              YPICAL CITTLA'G
                      104

-------
j>.   Flotation Procedure

The  flotation characteristics  of  the  test mixture were  esti-
mated  using  the  flotation  apparatus shown in  Figure A-5.   The
procedure was as  follows:

          1.   Partially fill  the calibrated  cylinder with
               waste  or flocculated sludge mixture and  the
               pressure chamber with  clarified  effluent or
               water.

          2.   Apply  compressed air to  the pressure chamber
               to  attain the desired  pressure.

          3.   Shake  the air-liquid mixture in  the pressure
               chamber for one minute and allow to stand  for
               three  minutes to attain  saturation.  Maintain
               the pressure on the chamber for  this period.

          4.   Release a volume of pressurized  effluent to
               the cylinder and mix with the  waste or sludge.
               The volume  to be released is computed from
               the desired recycle ratio.  The  velocity of
               release through the inlet nozzle should  be
               of  such a magnitude as not to  shear the  sus-
               pended solids in the feed mixture, but to
               maintain adequate  mixing.

          5.   Measure the rise of the  liquid-sludge in-
               terface with time.

          6.   After  a detention  time to twenty minutes,
               the clarified effluent is drawn  off through
               a valve in  the  bottom  of the cylinder.

          7.   Relate the  effluent suspended  solids and
               the float solids to the  calculated air-
               solids ratio.   When pressurized  recycle  is
               used,  the air-solids ratio is  computed:

                           1.3s R(P -  1)
                     Sa       QSa

where:

               s    =    air saturation, cubic cm/liter
                a
               R    =    pressurized volume, liters

               p    =    absolute pressure, atm
                             105

-------
                           { Calibrated plailiC cylinder
                            (atmospheric chamber) •
              Cork plug —i
           Rubber seal-
         Air spargor


     prauur* chamber
i


r~

tr

u

•"» 1
^i |
-^' -. ! ^,.
1
1
1 	 1
1
Pressure |

Inlcl valvo

    Air-blocd valva

    Top plato


    Riser tuba
SCHEMATIC  OF FLOTATION  APPARATUS

                  FIGURE A-5
                       106

-------
               Q    =    waste flow, liters

               S    =    influent suspended solids,
                a        mg/liter


Note:  the foregoing relationship was modified in the report
       in order to relate the effluent oil concentration
       to the air to oil ratio.
                               107

-------
7.  Standard Jar Test Procedure

          1.   Using 200 ml of sample on a magnetic
               stirrer, add coagulant in small in-
               crements at a pH of 7.0.   After each
               addition, provide a one minute rapid
               mix followed by a three minute slow
               mix.   Continue addition until a vis-
               ible floe is formed.

          2.   Using this dosage, place 1,000 ml of
               sample in each of six beakers.

          3.   Adjust the pH to pH 4.0,  5.0, 6.0, 7.0,
               8.0,  and 9.0 with standard alkali (see
               not below).

          4.   Rapid mix each 1,000 ml sample for three
               minutes; follow this with twelve min-
               ute flocculation at slow speed.

          5.   Measure the effluent concentration of
               each settled sample.

          6.   Plot the percent removal of character-
               istic vs. pH, and select the optimum
               pH.

          7.   Using this pH, repeat steps (2), (4),
               and (5), varying the coagulant dosage.

          8.   Plot the percent removal  versus the co-
               agulant dosage and select the optimum
               dosage.

          9.   If a polyelectrolyte is used, repeat the
               procedure adding polyelectrolyte toward
               the end of the rapid mix.
Note:  Because this study was designed to simulate actual
       plant conditions,  the pH was not varied.   The chem-
       icals were added to mixture at a pH = 7.0.  Steps 1,
       2, and 3 were not followed in this study.
                             108

-------
                          APPENDIX
                         REFERENCES


Beerthuses and Kappler, "Gas Chromatographic Analysis of Higher
Fatty Acid up to and Including Caustic Acid", Nature, London,
179, 1957, 731.

Hruinah and Polo, "Separation of C9-Clfi Free Fatty Acids in
Dairy Products by Dual Column P.T.G.C7," Journal of Gas Chroma-
tography, June, 1967-

Maher, "Semi-Quantative Analysis of Hydrocarbon Mixtures by
Gas Chromatography Without Complete Separation of Components",
Journal of Gas Chromatography, October, 1965 p. 335.

McPherson, Sawicki, and Fox, "Characterization and Estimation
of n-Alkanes in Airborne Particulates", Jouranl of Gas Chroma-
tography , April, 1966, p. 156.


Byars and Jordan, "An Efficient Packed Column for Free Fatty
Acids", Journal of Gas Chromatography, September, 1964, p.
305.                           '

Albert, G.C., "Determination of C5-C,, in Paraffins and Hydro
Carbon Types of Gasoline", Analytical Chemistry, Volume 35,
No. 12, November, 1963, p. 1918.

LePlat, P., "Application of Pyrolysis-Gas Chromatography to
Study of Non-Volatile Petroleum Fraction", Vol. 5, No. 3,
March, 1967, p. 128-35.

Girand, A., Bestorrzeff, M.A., "Characterization of High Mole-
cular Weight Sulfur Compounds in Petroleum by Pyrolysis and
Gas Chromatography,  Journal of Gas Chromatography, Vol. 5, No.
9.  September, 1967, p. 464-70.

Lively, L. Rosen, A.O., Mashni, C.I.,  "Identification of Pet-
roleum Products in Water", Purdue University Engineering Ext.
Ser. 1158, 1965, p.  657-63.

Farley, R.,Valentine, F.H.H., "Coagulation as Means of Separ-
ating Oil From Effluents", American Institute of Chemical Eng-
ineers, Advances in Separation Techniques Seminar, Paper #14
London, June 13-17,  1965, p. 36-45.

Martin, R.L., Winters, J.E., "Composition of Crude Oils by Gas
Chromatography - Geological Significance of Hydro Carbon Dis-
tribution", World Petroleum Congress,  6th, - Composition, Ana-
lysis and Testing, Proceedings, Section 5, 1963, p. 231-60.
                             109

-------
Dalbke, R.G., Lenny, A.M., Baumert, H., "Problems of Detecting
and Controlling Fats, Oils and Greases at Their Source", Lub-
rication Engineering, Volume 21, No. 10, October, 1965, p. 433-38

Davies, D.S., Kaplan, R.A., "Removal of Refractory Organics
from Wastwater with Powered Activated Carbon", Journal of
Water Pollution Control Federation, Vol. 38, No. 3, March,
1966, p. 442-50.

Maehler and Greenberg, "Identification of Petroleum Industry
Wastes in Ground Waters", Journal Water Pollution Control
Federation, Vol. 34, No.  12, December, 1962, p. 1262-7.

Hoah, R.D., "Recovery and Identification of Organics in Water",
Air and Water Pollution,  Volume 6, November-December, 1962,
p. 52l=38~;

Cochran, L.G., and Bess,  F.D.,  "Waste Monitoring by Gas Chroma-
tography", Journal Water Pollution Control Federation, Vol. 38,
No. 12, December, 1966.

Hoah, R.D., Caruso, S.C., Bramer, H.C.,  "Tracing Organic Com-
pounds in Surface Streams", Air and Water Pollution, Volume 10,
No. 1, January, 1966, p.  41-8.

Dalbke, R.G., Tenney, A.M., and Baument, H., "Problems in De-
tecting and Controlling Oils,  Fats, and Greases at Their
Source", Lubrication Engineering, Volume 21, No. 10, October,
1965, p. 433-8.
                             110

-------
     Accession Number
                             Subject Field & Group
                                                SELECTED  WATER RESOURCES  ABSTRACTS
                                                      INPUT TRANSACTION FORM
     Organization


           American Petroleum Institute, Washington, D. C.
     Title
           THE IMPACT OF OILY MATERIAL ON ACTIVATED  SLUDGE SYSTEMS
 10
Author(s)
           Barnhart,  Edwin L,
                                16
Project Designation
                                                Project No.  12050 DSH
                                21
                                    Note
 22
     Citation
           Hydroscience,  Inc., Westwood, New Jersey
 23
     Descriptors (Starred First)
             il Wastes,  -^Activated Sludge, -^Sewage Treatment
 25
     Identifiers (Starred First)
           *-Bench Scale  studies,  -#Spent crankease oil, ^-Vegetable  oil,  ^Crude oil,
           ^Refinery waste  oil, -*Load tolerance, -^Continuous, #Shock
 27
Abstract Small scale continuous activated  sludge  systems were exposed to a variety
       of oily compounds at several loading  levels and performance observed.
       Batch studies to determine biodegradability and the effect of emulsifica-
       tion and temperature on the rate  of biological reaction were also conducted.

       Oils introduced into an activated sludge system are absorbed on the floe
       and slowly degrade.  If the loading rate is higher than the degradation
       rate and the rate of wastage, the oil accumulates on the sludge.  This
       accumulation causes a loss of density and a loss of acceptable settling
       characteristics.  The biological  system  fails due to the loss of sludge
       but the ability of the microbial  system  to remove other substrates is
       not inhibited.

       The continuous feed level of oils to  activated sludge should not exceed
       0.10 pounds per day per pound of  sludge  under aeration.  Shock loads
       should not exceed 5$ of the weight of the sludge under aeration.

       The study also considered physical separation of oils before biological
       treatment, and various chemical methods  of separating complex emulsions.
'46s
-------