U.S.A.—U.S.S.R.
Working Group
on the Prevention of
Water Pollution
from Municipal and
Industrial Sources
Cincinnati, Ohio - U.S.A.
April 5-6, 1977
U.S EPA Heart*1'£.*"
Ma!1 co . ''"
1200Penns,.•""!< >•
Washing:^ J
vvEPA
SYMPOSIUM ON
Physical—Mechanical Treatment of Waste Waters
United States
Section
SP 600/9
77-504
LIBRARY
U.S. ENVIROSMEHTALF
EDISOH, H.J. 08817
-------
-------
USA-USSR
WORKING GROUP
on the
Prevention of
Water Pollution
from
Municipal and
Industrial Sources
Symposium on
Physical-Mechanical
Treatment of
Wastewaters
United States
Environmental Protection
Agency
Cincinnati, Ohio
April 5th and 6th, 1977
U.S EPA Headquarters Library
Matf code 3404T
1200 Pennsylvania Avenue NW
Washington. DC 20460
202-566-0556
LIBRARY
-------
Index
Preface
Opening Address •
Rhett
Mr. John T.
Welcome — Dr. David G, Stephan
Response — Yu. N. Andrianov
Papers Presented at the
USA/USSR SYMPOSIUM
Sebastian, P.P., Lachtman, D.S.,
Kroneberger, O.K.. Allen, T.D.,
(Envirotech), Pyrolysis Applications
for Industrial and Municipal
Treatment
Myasnikov, I.N., Ponomaryev, V.G.,
Ne'chaev. A.P.. and Kedrov, Yu. V.,
(VNU Vodgeo. Gosstroi USSR).
Wastewater Treatment by Physical
and Mechanical Methods
Lacy. William, (US EPA), Physical
Treatment of Oil Refinery
Wastewater
Skirdov, I.V.. (Vodgeo),
Improvement of Hydraulic
Conditions of Radial Settling Tanks
Gellman, Isaiah, (NCAS1, Pulp &
Paper Industry, New York, New
York), Current Status and Directions
of Development of Physico-
Mechanical Effluent Treatment in the
Paper Industry
Skirdov, I.V., Sidorova, I.A.,
Maksimenko, Yu. L.. (USSR)
Employment of Microstrainers in the
Wastewater Treatment Practice
23
26
31
Grutsch, J.F., (American Oil
Company, Indiana), The Control of
Refinery Mechanical Wastewater
Treatment Processes by Controlling
the Zeta Potential
Oberteuffer, J.A. and Allen, D.M.
(Sala Magnetics, Cambridge, Mass.),
Combined Storm Overflow
Treatment with Sala-HGMF® High
Gradient Magnetic Filters
Myasnikov, I.N., Gandurina, L.V.,
Butseva, L.N., (USSR), Use of
Flotation for Wastewater Treatment
FitzPatrick, J.A., and Swanson,
C.L., (Northwestern University,
Evanston, Illinois), Performance
Tests on Full-Scale Tertiary Granular
Filters
Pisanko, N.V., (Ukrvodokanalproekt
Institute, USSR) Sewage Treatment
in Mining, Metallurgical and Oil-
Chemical Industries
Fields, R., (US EPA), The Swirl
Concentrator for Treating and
Regulating Sewered (Separate and
Combined) and Unsewered Flows
Galanin, P.I., (USSR), Sewage-
Treatment of the City of Moscow
Protocol
Appendix I Participants
Appendix 11 Reports
Appendix III Future Program
Appendix IV Itinerary
Appendix V Symposium Program
44
75
91
too
115
122
133
138
139
140
141
143
143
-------
Preface
The fourth cooperative USA/USSR
symposium on the Physical-Mechanical
methods of Waste Water Treatment from
Municipal and Industrial Sources was
held in Cincinnati, Ohio at the U.S.
Environmental Protection Agency head-
. quarters on April 5th and 6th, 1977. This
symposium was conducted in accord with
the fifth session of the Joint USA/USSR
Commission held in Moscow, USSR from
November 15 through 19. 1976.
This symposium was sponsored under the
auspices of the Working Group on the
Prevention of Water Pollution from Mu-
nicipal and Industrial Sources. The co-
chairmen of the Working Group are H.P.
Cahill Jr. of the United States
Environmental Protection Agency and
S.V. Yakovlev of the Department of
Vodgeo in the Soviet Union.
The United States delegation was led by
John T. Rhett. Deputy Assistant Adminis-
trator for Water Program Operations,
U.S. Environmental Protection Agency.
The Soviet delegation was led by Yu. N.
Andrianov, Director, All-Union
Association. Gosstroy.
The thirteen papers that were presented at
the symposium (seven US and six USSR)
are reprinted in English in this volume in
accord with the protocol signed by the
delegation leaders on April 16th, 1977.
-------
Opening
Comments
John T. Rhett
Deputy Assistant
Administrator for Water
Program Operations
U.S. Environmental Protection
Agency
1 am extremely honored and pleased to
open this fourth symposium in the series
undertaken by the US/ USSR Working
Group on the Prevention of Water Pollu-
tion from Industrial and Municipal Sour-
ces.
This symposium is designed to concen-
trate on the prevention of water pollution
through the application of physical/ me-
chanical treatment methods. We have a
very comprehensive program on this sub-
ject for the next two days. The papers
from both the United States and the
Soviet participants are outstanding. The
results of the symposium will lead to
further progress in this very important
area of wastewater treatment.
The symposia that our Working Group
has undertaken have a planned pattern to
cover the most important areas for ad-
vancing pollution abatement from munici-
pal and industrial sources. Our first three
symposia have already led to useful results
from a systematic review of progress to
date and potentials for improvement in
the future. These first three symposia were:
• Treatment of Wastewater Sludge
(USSR — May 12-16. 1975)
• Physical-Chemical Treatment of Waste-
water (USA — Nov. 12-14. 1975)
• Intensification of Bio-Chemical Treat-
ment of Wastewaters (USSR — Aug. 22-
Sept. 5, 1976)
1 look forward to the Physical-Mechanical
symposium today to be as distinguished
and productive as the prior three sympo-
sia. And 1 know they will be.
I would like to make a few observations
from my participation in the symposium
and tour of Soviet treatment facilities in
August-September 1976.
I was very much impressed by the compat-
ibility and friendliness that has been gen-
erated by both the Soviet and American
counterparts. Also, 1 am impressed that
each one of these visits extends the circle
of acquaintances, professional colleagues
and friends. This compatibility has been
aided by the people on both sides having
common backgrounds and interests in the
scientific/technical area. I believe that
progress in the technical areas could not
have been so great without the friendliness
and cooperation of our two peoples. 1
look for this to continue into the future
with even greater enthusiasm.
In the United States we have been making
progress in water pollution abatement and
1 am confident we can look forward to
continued successes. There are some grim-
ly humorous stories of just how bad things
were at one time. On one river, we could
not take water samples in metal buckets,
because the "water" was so corrosive that
it ate the bottoms out of the sampling
pails. In another environmental "horror
story", Cleveland once considered declar-
ing the Cuyahoga River, which runs
through its center, a fire ha/ard after the
River itself repeatedly caught fire. 1 hope
such incidents are, for the most part, in the
past.
In case after case, in waters that had been
considered biologically dead, or nearly
uninhabitable by fish and aquatic plants
sensitive to pollution, we are seeing dra-
matic rejuvenation. Lake Erie is no longer
given up for lost. Salmon are returning to
the Connecticut River for the first time in
generations. The WilliameUe River, the
Detroit River, the Buffalo River, the
Houston Ship Channel, are seeing the
return of fish and biota that had disap-
peared for years or decades.
Objectively, the overall net improvement
in water quality has been substantial. As
measured by three primary indicators of
water quality — fecal coliform counts,
dissolved oxygen levels, and phosphorus
levels — a recent estimate of improvement
is on the order of 35 percent of waters
tested. This is especially heartening when
we consider the growth that has occurred
over the past eight or nine years. Our total
population served by sewers has grown by
12 percent and wastewater flows have
increased by 20 percent.
1 am sure I will not minimize the progress
we have made to date if! emphasize the
complexity and urgency of the problems
which still confront us. I would also like to
stress the importance of meeting these
problems agressively, in the knowledge
that we must undertake a long-term com-
mitment if we are to continue our past
successes. 1 know from first-hand expe-
rience how much easier it is to muster the
initial enthusiasm for an idea than it is to
maintain long-term support for even the
most worthy or necessary purposes. Yet, it
is exactly that long-term commitment
which we must accept.
In a sense, dealing with the belching
smokestacks and the grossly polluted wa-
ters is the easier part of our overall task.
These are obvious and dramatic symbols
of how the environment can be spoiled.
Improvements of such obvious abuses are
of course dramatic. Now we must begin to
look to the future consequences of our
decisions and actions involving the envi-
ronment. The extent to which we need to
make careful plans for the future is sober-
ing. This is why we must face up to the
complexities of what we are trying to
accomplish, and why I am pleased to be
addressing a group that understands the
meaning of long-term commitments.
By one estimate, waste treatment plants,
in the United States, now generate five
million dry tons of sludge each year. By
1990, that figure will double. As a Nation,
we now use about 400 billion gallons of
water each day, and that figure will double
by the end of the century. The demand for
drinkable water for municipalities is pro-
jected to increase from 30 billion gallons
daily to 50 billion gallons. We have no
choice — we must ensure that adequate
supplies of safe, clean water are available.
We are also fast approaching the situation
that in solving one environmental prob-
lem, we may create another. What, for
instance, do we do with our solid wastes?
Even the most carefully designed landfill
may release hazardous substances into an
-------
aquifer or stream that, in turn, supplies a
community with its drinking water. Do we
burn our solid wastes, and risk contami-
nating the air we breathe? Do we dump
our sludge at sea, and risk destroying
marine ecosystems? How will we deal with
the heavy metals and families of organic
chemicals which have a notorious tenden-
cy to accumulate in the environment?
Even such practices as routine chlorina-
tion can, in the presence of certain con-
taminants, create hazardous substances.
Our continued commitment to clean up
the environment, the protect the integrity
of our natural systems, both in the USSR
and the United States, cannot be seen as a
luxury, with benefits that are largely aes-
thetic. We are dealing now with the pro-
tection of our present and future health
and well-being.
We, in the United States, have just com-
pleted our latest municipal needs survey.
With the exceptions of control of storm-
waters, it indicates that $96 billion is
required to abate water pollution to accep-
table levels. Stormwater treatment would
require an additional $54 billion. You can
see that we have a long ways to go.
In dealing with such astronomical figures,
even small gains in effectiveness and effi-
ciency in methods of treatment can result
in substantial savings in funds that will not
have to be expended.
I am sure that the USSR is facing prob-
lems of funding pollution abatement that
are similar in magnitude to the United
States.
1 look to our US/USSR Working Group
to make substantial gains in so improving
treatment methods that benefit each
Nation's environment that economy will
result. The resulting improvements in
treatment methods can also be applied
worldwide to the benefit of all peoples.
With this as our objective, I look forward
to the results of this symposium.
Welcome
Dr. David G. Stephan
It is my great pleasure today, to welcome
you to both this joint US/ USSR Symposi-
um and to the EPA Environmental Re-
search Center here at Cincinnati, Ohio.
When the US/USSR Working Group on
the Prevention of Water Pollution from
Municipal and Industrial Sources last met
here in November, 1975, this great re-
search facility had not yet been completed.
We were able then to show you just the
empty shell of our building. Now we have
active research and development efforts
underway and we will be proud to escort
you around to see and understand our
activities in research and development.
More important than the fact that this
building exists is the place that the
Environmental Research Center holds as a
key point of international communication
in the environmental sciences.
Science knows no national boundaries.
Most progress in science is made from the
efforts of all scientists, worldwide.
cooperatively advancing the frontiers of
scientific research to achieve worthwhile
results. To achieve these results, however,
intercommunication of scientists and
technologists must be fostered on a
continuing basis. Face-to-face interchange
of ideas, as in this symposium, is most
helpful in keeping the spirit of cooperation
and communication at the highest possible
level.
Certainly, in the area of prevention of
pollution from municipal and industrial
waste waters, this US/USSR Working
Group has been a shining example of how
to make optimum progress. Concerted
effort to interchange both ideas and peo-
ple during visits to each other countries on
a regular basis has been extremely valua-
ble. The Environmental Research Center
is honored that it has been an important
part of the progress that has been made.
I would like to mention briefly some of the
work that we are pursuing that is related
to Physical/Mechanical Methods — the
theme of this Congress. In the area of
industrial processes, we have active work
underway, either here or by contract, on:
• joint sludge/refuse processing
• wet oxidation and pyrolysis pilot plant
studies
• cost-effective closed water cycle systems
in pulp mills
• combined industrial/municipal treat-
ment, with special emphasis on pre-
treatment standards
• water pollution problems associated
with the production of petroleum prod-
ucts
• ground water problems
• reducing pollution from the metal fin-
ishing industry
• removal of nitrocellulose from waste-
water by ultrafiltration
• recycling treated effluents from paper
and board productions, and
• nitrogen removal from meat packing
plant effluents
So you can see that we have a well
rounded program of research on industrial
processes.
But our own efforts are still not enough.
The papers that have been prepared by
our Soviet colleagues will add substantial-
ly to the world's working knowledge of
how to achieve more progress in water
pollution abatement. And the United
States papers are in turn, designed to help
add to the knowledge of our Soviet col-
leagues and to world scientific and techno-
logical advancement, in general.
I would like to recommend that this and
future symposiums include communica-
tion of ideas and methods for toxic pollu-
tant controls. The vast increase in the
chemical industry in response to modern
social needs has produced the potential for
fouling our waterways with dangerous and
offensive effluents from industrial pro-
cesses. Both of our nations face similar
problems in ensuring that effluents from
modern industrial advances do not harm
the very people that it is supposed to
benefit.
In recent years, the United States, and I
am sure the USSR, have become con-
cerned about the ever-increasing dangers
-------
of toxic pollutants. Recently in the United
States, for example, carbon tetrachloride
and other dangerous chemicals have been
released to the nation's waters
accidentally. We would hope that signifi-
cant knowledge on how to prevent releases
of toxic chemicals will result from our
joint efforts in the Working Group.
In conclusion, 1 would like to again com-
mend the efforts of this Working Group in
advancing the interchange of useful
knowledge between the US and the
USSR.
I believe that we already see evidence of
avoidance of duplication of efforts, while,
at the same time, we see the multiplying
effect of jointly cooperating to solve com-
mon problems.
The key factor is that the welfare of our
societies and our citizens, in the US and
USSR, will be greatly enhanced by our
joint efforts. Let us keep this as our guid-
ing light throughout the proceedings of
our two-day symposium.
Response
Yu. N. Andrianov
At the present stage of industrial and
urban development an increasingly impor-
tant meaning is assumed by the interrela-
tion of man and nature. The protection of
nature from pollution and the rational
utilization of natural resources acquire
now not only economic but social mean-
ing. This is why in a number of countries
this problem is treated on a national level.
For successfully dealing with environmen-
tal protection the efforts not only of nu-
. merous organization, but even of coun-
tries, are united and one may say that the
problem of environmental protection has
crossed the borders of many countries,
having united their strengths and means
for the preservation of adequate environ-
ment for human life.
The protection of water reservoirs for
industrial and municipal pollution occu-
pies a significant place in the environmen-
tal effort. The complexity of this problem
is determined by the continuously growing
economic activity of man, by the appear-
ance of complex waste materials, and by
an increased demand for quality goods. In
its turn, water — an integral part of the
majority of industrial processes —. de-
mands complex solutions for its regenera-
tion and retention of its natural purity.
In our country the required attention is
paid to the environment and to its rational
utilization. During recent time a whole
series of decisions and laws have been
adopted for the protection of the environ-
ment, including water resources. Special
attention is paid for the protection of
purity of rivers, lakes and seas. With this
goal in mind the Government has adopted
measures for the prevention of pollution
of such large water reservoirs as the river
Volga, Ural, Lake Baikal, the Caspian and
Baltic seas. Efforts in this area have al-
ready achieved significant results.
Large projects are scheduled for imple-
mentation in our country during the cur-
rent Five-Year Plan. Capital investments
alone are 11 billion rubles. Parallel with
this the necessary attention will be given to
scientific research efforts for environmen-
tal protection, including the protection of
water resources.
In this area a significant part of the effort
in the solution of water resource protec-
tion and the improvement of industrial
and municipal water utilization in our
country is carried out by the All-Union
Association Soyuzvodokanalproyekt,
which encompasses the All-Union Scien-
tific-Research Institute Vodgeo. This As-
sociation embraces a large number of
design and research institutes, their
branches and sub-divisions, located in
various cities and industrial regions of the
country. Such an association of scientists
and designers allows the solution of any
question related to the-organization of
water use in newly created industrial en-
terprises and its improvement in those
already in existence.
The Association carries out scientific and
design work related to their geographic
distribution and their water supply from
surface and underground sources, design
of sewage systems and the neutralization
of industrial and municipal wastes, and
the setting up of construction norms and
regulations and of national standards.
The largest volume of work of the Associ-
ation is in the sewage area. Here new
methods and facilities are designed for the
neutralization of industrial and municipal
wastes, zero-discharge and minimum-
water-consumption technology imple-
mented, re-circulation and multiple-use of
treated effluents systems are set up, includ-
ing industrial and municipal plants. Wide
application is given to the introduction of
latest scientific achievements in experi-
mental designing. In cooperation with
scientists, the designers of our Association
have already built a number of modern
facilities and waste-water treatment plants
measuring up to highest requirements. For
example, the Baikal waste-water treat-
ment plant, designed according to our
plans, treats the effluent 'of the paper-pulp
industry to such a high standard that it
can be used as drinking water.
A large volume of scientific and design
projects in industrial and municipal water
-------
use will be carried out by the Association
also in the future.
The USSR-US Agreement on cooperation
in water pollution prevention from indus-
trial and municipal sources is a significant
contribution to environmental protection.
In accordance with the outline of this
cooperation, Soviet and American special-
ists are working out methods and facilities
for the treatment of effluents in such water
intensive areas as paper-pulp, chemical,
petro-chemical and oil refining industries
and municipal use.
Contemporary research efforts pay great-
est attention to the design improvement of
settling basins, flotators (flotation sys-
tems), aeration tanks, filters and other
equipment. Cooperative efforts in this
area make it possible — on the basis of
existing experience — to design more
efficient facilities and to work out new
treatment systems.
Regular exchanges of scientific and re-
search information make possible the low-
ering of costs related to research and the
reduction of implementation time for in-
troduction of latest achievements into the
practice of waste water treatment.
.Symposia, dedicated to practical problems
of waste water and sludge treatment, be-
came a fine tradition in our cooperative
work. As a rule these symposia are at-
tended by large numbers of specialists of
our countries. Their proceedings are pub-
lished in Russian and English and are
disseminated in large number of copies
among specialists in our countries.
The present symposium is dedicated to
problems related to physical-mechanical
methods of waste water treatment. As is
known such methods are widely used for
the treatment of municipal and industrial
waste waters. In spite of the already exist-
ing experience in their use, many of these
methods require further refinement and
also the development of new technological
solutions. Exchange of opinions on the
efficiency of physical-mechanical methods
will be of scientific and practical interest.
The work of the symposium will make it
possible to determine the main directions
of research efforts in the area of physical-
mechanical methods for the present and
for the future.
In conjunction with the symposium. So-
viet and American specialists will discuss
future plans and forms of their coopera-
tion.
Joint work of Soviet and US specialists
will make it possible to accelerate signifi-
cantly the technical progress in water
resources protection from industrial and
municipal pollution and this will be a
worthy contribution to the protection of
the environment.
Pyrolysis
Applications for
Industrial and
Municipal
Treatment
Frank P. Sebastian
Senior Vice President
Envirotech Corporation
Menlo Park, California
Dennis S. Lachtman
Environmental and
Occupational Health Analyst
Envirotech Corporation
Menlo Park, California
Gerald K. Kroneberger
District Technical
Representative
Eimco BSP Division
Envirotech Corporation
Belmont, California
Terry D. Allen
Product Development
Manager
Eimco BSP Division
Envirotech Corporation
Belmont, California
Introduction
As suggested by the title, this presentation
focuses on the applications and energy
recovery aspects of pyrolysis for municipal
and industrial treatment.
Pyrolysis is the decomposition of organic
materials in an oxygen starved atmos-
phere. In the absence of oxygen, organic
material is driven from the solids in the
form of combustible pyrolysis gas. Com-
plete oxygen starvation would necessitate
firing fuel for heat, but with the multiple
hearth furnace (MHF) partial combustion
of the feed supplies the heat for the pyroly-
sis reaction. Combustion is minimized by
supplying only air to obtain optimal pro-
cessing temperatures.
For autothermic pyrolysis, feed materials
that yield a gas combustion temperature in
excess of 649°C (1200° F) are necessary.
This corresponds to a feed energy level of
-------
above 3500 BTU's per pound of feed
water. With feed levels having in excess of
6000 BTU's per pound of water, pyrolysis
becomes a virtual necessity because in an
incineration mode elevated furnace tem-
peratures would result in creating the like-
lihood of material fusion.
Pyrolysis has been utilized for processes
having high energy feed materials such as
carbon activation, carbon reactivation,
char production, and other
carbonization/ devolatilization processes.
In the pyrolysis mode, a multiple hearth
furnace can handle materials containing
more than 30,000 BTU's per pound (.45
kg) of feed water and still maintain mid-
zone temperatures below 833° C (1500°F),
and outlet temperatures less than 649° C
(1200°F). The temperature in the external
combustion or afterburner chamber could
exceed 1649°C (3000°F). The significance
of elevated afterburner temperatures is the
increasing driving force for steam produc-
tion in a heat recovery boiler.
Multiple
Hearth Furnace Description
Shown in Figure 2 is a multiple hearth
furnace cross-section showing the various
zones of operation used for the pyrolysis
method of thermal processing. The size
and number of hearths are determined by
evaluating the process time and feed mate-
rial characteristics. A minimum six-hearth
furnace is recommended as the best con-
figuration for most applications.
Furnace specifications are modified for
optimal handling of the particular feed
material. Materials requiring large com-
bustion volumes and low solids processing
residence time are designed with an en-
larged external afterburner and small
hearth areas. The one-hearth furnace de-
sign with external afterburner represents
the extreme situation of large combustion
volume with little or non-existent solids
handling processing area.
The furnace is usually loaded at the top.
Rotating arms on a vertically positioned
shaft spiral.the material in a counter-
clockwise direction across the hearths.
The hearths have alternating center and
peripheral drop holes such that material
falling through a peripheral drop hole is
pushed toward the center of the next •
hearth. On the hearth having a central
hole, material is spiraled around toward
the outside, and so on (see Figure 2).
The arms all move in the same direction
and alternate inward and outward move-
ment of material is obtained via the an-
gling of the plows (rabble teeth). This
design provides substantial opportunity
for retention time and zone control.
Retention time is controlled by the setting
of the variable speed shaft drive, which
sets the rotational speed of the arms and
the orientation of the rabble teeth. Each
hearth can be controlled for temperature
and atmosphere (degree of oxidation (or
reduction).
The hearths are refractory, the shaft is
refractory insulated cast iron, and the
arms and teeth are usually cast alloys. The
orientation of rabble teeth determines bed
depth, total surface area and retention
time. The live bed depth, the area between
the bottom of the rabble teeth upward, is
adjusted for the particular process.
Process Advantages
Essentially, there are two steps in the
pyrolysis process: the first is the conver-
sion of organics to gases; and the second is
the combustion of the pyrolysis gas to
recover heat energy.
The pyrolysis operating mode offers a
combination of advantages and benefits
compared to incineration, as follows:
1. improved process control of tempera-
ture, and minimization of excess air.
2. For material with high energy content
(above 6000 BTU/ pound (.45 kg) of feed
water) low temperature (1500° F) pyrolysis
prevents material fusion or clinker forma-
tion;
3. Pyrolysis minimizes dependence on
energy sources while offering energy re-
covery to offset the energy demand of
treatment plants;
4. Reduced furnace size per quantity of
dry solids;
5. Reduction in paniculate loading prior
to scrubbing;
6. It offers potential revenue from steam
or power generation, and
7. It affords opportunity to reclaim or
recycle waste materials and in the process
eliminates the environmental problems
associated with waste in the processing of
raw materials.
Industrial Applications
In the petroleum industry pyrolysis has
earned a proven record for effectively
handling high energy sludges. Processing
oil filtration sludges having minimal water
and filter earths is easily controlled by the
pyrolysis operating mode, which serves to
reduce hearth temperatures, prevents fu-
sion, and helps maintain longer furnace
life.
A major U.S. petrochemical plant recently
selected multiple hearth furnaces over
fluidized bed furnaces to pyrolyze wasten
aerobic sludge and still bottoms. The high
concentration of chloride salts in the
sludge and the resultant fusion effect of
such salts at high temperatures requires
the specification of pyrolysis for disposal
of the sludges. The organics from the high
energy sludge, 8000-9000 BTU/pound (.45
kg) water, will volatilize at comparative
low temperatures approximately 538° C
(1000° F). The volatile gases will exit the
furnace at a temperature of 482° C (900° F)
and be combusted in an external after-
burner at 802°C (1475° F).
The automotive industry has successfully
used pyrolysis to reclaim cast-iron borings
from engine blocks. The process reclaims
cast-iron by volatilizing cutting oils to
form gases. Combustion of a portion of
the gases provides the heat energy to
evaporate the moisture and heat the cast-
iron chips prior to briquetting at 538° C
(1000°F). In this application, the advan-
tage of pyrolysis is to prevent fusion
problems, reduce furnace wear and im-
prove control of process conditions.
Pyrolysis in a MHF has been shown as an
ideal method for reclaiming waste mate-
rials via charcoal production. In this pro-
cess organic materials are re-volatilized at
538°C (1000°F) to 863°C (1600°F). The
introduction of air is s controlled to maxi-
mize the formation of fixed carbon,or char.
-------
The C.B. Hobbs Corporation uses the
pyrolysis process to produce charcoal
from 50,000 tons (22,727,000 kg) of nut-
shell and fruit pit wastes each year. This
Elk Grove, California, corporation uses
the waste to produce a salable product —
charcoal. This process also solves the
waste disposal problems of growers,
canners and processors. The 50,000 tons
(22,727,000 kg) of waste are used to pro-
duce 35,000 tons (16,000,000 kg) of high
quality charcoal annually.1,2 The typical
mixture includes peach, apricot and olive
pits, plus walnut and almond shells. The
heat for drying the wet wood or other
waste materials can be supplied by the hot
gases from the combustion stack, which
saves energy.
Municipal Treatment
As demonstrated by a wealth of expe-
rience from municipal sewage treatment
plants, the multiple hearth furnace is one
of the more environmentally sound and
economically justifiable methods of han-
dling the escalating volumes of sewage
sludge. The need for improved sewage
sludge processing is increased because of
advanced water treatment methods, which
generates increased sludge volumes. Jones
et at.-' estimates that the total sludge solids
(dry basis) requiring disposal could exceed
23,000 tons per day (10,454,000 kg) by
1985 in the United States. Presently there
are over 300 multiple hearth sludge furna-
ces in existence. Furnaces process between
25% and 30% of all sludge in the United
States, with the multiple hearth configura-
tion representing about 75% of these ther-
mal processing units.-'
Recognizing the environmental problems
associated with the disposal of large quan-
tites of sewage sludge, the New York/
New Jersey Interstate Sanitation Com-
mission in a joint effort with the United
States Environmental Protection Agency
has evaluated the most cost effective and
environmentally sound alternative to
ocean dumping in the New York-North
New Jersey Metropolitan area. This tech-
nical report, prepared by Camp, Dresser
& McK.ee, found pyrolysis to be the best
alternative to ocean disposal.
Energy Recovery
The recent substantial increases in the
costs of energy and fuel consumption area
cause for major concern in the evaluation
of sewage disposal processes. As stated by
Shannon et al.4 of Environment Canada,
"It is a popular misconception that energy
cost increases will severely affect the eco-
nomic feasibility of sludge incineration. In
fact, recent developments in sludge de-
watering, heat recovery and reuse will
actually make sludge incineration energy
self-sufficient."
Pyrolysis represents an advanced thermal
design that progresses beyond the elimina-
tion of fuel requirements to recycle the
sludge's inherent fuel values into power.
The Cowl it/ County Municipal Waste-
water Incinerator plant, which started up
June 1, 1976, for example, has a full scale
MHF unit that has been run in a partial
pyrolysis operating mode. The furnace is a
12'9"(3.9 meters) OD by 7 hearth model
which is designed to handle 748 kg/hr
(1645 Ib/hr) of dry solids containing be-
tween 30% and 40% moisture.
Heat treatment or thermal conditioning
of the sludge allows it to be readily
dewatered to about 40% solids, which is
sufficiently dry to create a net exothermic
heat balance and could produce steam
required for the heat treatment
operations, plus additional steam for the
plant.
In their search for better operating tech-
niques, the Cowlitz County plant manage-
ment has achieved the proper parameters
to operate the multiple hearth furnace
without auxiliary fuel. At this installation
autothermic (without auxiliary fuel) pro-
cessing occurs when the sludge is de-
watered to approximately 40% dry solids.
Then the furnace temperature profile is
maintained by the controlled and selective
addition of air to limit rather than quench
the temperature rise at each hearth. This
ingenious operating technique maximizes
the operating efficiency and fuel economy
of the plant while minimi/ing the power
consumption of the off-gas system by
reducing the quantity of excess air.
The operator requirements are also mini-
mized due to the operational stability;
such services are limited primarily to peri-
odic checks and minor corrective adjust-
ments.
The technique of temperature control
through air limitation —• pyrolysis — is
responsible for the operational benefits
mentioned above while simultaneously
meeting the latest EPA state of the art
requirements of paniculate emissions and
odor. While it is important to emphasize
that the furnace operates without any
auxiliary fuel, a further improvement in
this system will be the addition of a waste
heat recovery boiler between the furnace
outlet and the scrubber system. The heat
recovered from the furnace off gases can
then be used to produce steam to satisfy or
supplement the energy requirements of the
treatment plant.
Another type of pyrolysis treatment for
sewage sludge has been performed on a
demonstration basis. Full scale tests com-
pleted July 30th at the Concord, Califor-
nia, Wastewater Treatment Plant have
shown that pyrolysis of municipal solid
waste and sewage sludge can be used as a
source of energy for wastewater treatment
plants.
The Concord project started in November,
1974, when the concept was proven at
Envirotech's Brisbane, California, test fa-
cility. Results were so promising that the
EPA gave grant authorization for full
scale modifications and tests at an availa-
ble furnace at Concord, California. This
testing started in May. 1976, at rates of 2-4
tons (1-1800 kg) an hour, and was carried
out by Envirotech Corporation under a
contract from Brown and CaJdwell. the
consulting engineers on the project. It
involved combusting and pyrolyzing mix-
tures of refuse and sludge in a multiple
hearth furnace. Sponsors of the project
were the Central Contra Costa Sanitary
District, the State of California, and the
Environmental Protection Agency. A
complete report on the project was pres-
ented in December.5
The study was aimed specifically at finding
-------
ways to cut the cost of tertiary wastewater
treatment and solids disposal at the
nearby Central Contra Costa Sanitary
District Water Reclamation Plant. The 30
mgd (113,550 m-1/ day) regional facility,
now under construction, will recycle do-
mestic sewage into high quality water for
industrial process and cooling use by mid-
1977.
To offset the potentially high energy bills
in operating the new plant which could be
incurred as a result of unprecedented price
increases over the past three years, studies
were conducted on how to utilize as an
energy source the 1000 tons (454,000 kg) a
day of municipal solid waste currently
generated in the district and disposed in a
landfill. In addition to eliminating auxil-
iary fuel, producing energy, and disposing
of wastes, pyrolysis will extend the life of
the landfill. Existing multiple hearth incin-
erators can be converted to pyrolysis fur-
naces.
The recommended process calls for two-
stage shredding, followed by metal recla-
mation and air classification of the solid
waste to produce a refuse-derived fuel
(RDF). The RDF is fed to the furnace
along with sewage sludge and pyrolyzed to
produce a combustible gas. No auxiliary
fuel is required for the process.
Results to date indicate that the fuel gas
has sufficiently high heat content to pro-
vide over 90% of the plant's energy re-
quirements. Plans are for the gas to be
used to fuel a lime recalcining furnace at
the district's water reclamation plant. Oth-
er uses will be to produce steam for steam
driven mechanical drives, for plant heat-
ing and cooling systems, and for electrical
power generation. Indicative of the poten-
tial energy available from this process,
afterburner temperatures in excess of
1316°C (2400° F) were recorded at the
Concord site.
Energy Recovery Economics
As previously mentioned, with proper
dewatering techniques, pyrolysis of sew-
age sludge should not require supplemen-
tary fuel and could be operated to recover
energy. As the price of energy continues to
escalate, the economic benefit of energy
recovery will become more significant.
Using preliminary data for equipment
systems and disregarding building and
other peripheral plant expenditures, it
appears that systems similar to the Contra
Costa example (using a mix of sewage
sludge and refuse) can recover enough
energy to pay back the capital, fuel, pow-
er, labor and chemical costs associated
with sludge handling within a short time
period. The degree of energy recovery is
dependent on the moisture content of the
sludge and refuse and also will vary ac-
cording to the content ratio of refuse to
sludge.
The economic advantages of pyrolysis
systems vary according to plant design
and feed materials. To illustrate the bene-
fits that are available from the use of
pyrolysis, a comparison was done of the
costs of three types of solids handling
systems for a 50 MGD (190,000 m'/day)
sewage treatment plant treating both
primary and secondary sludges.' The sol-
ids processing systems that will be dis-
cussed are as follows:
Case 1: a MHF that processes sludge and
scum in an incineration mode;
Case 2: a MHF that processes sludge and
scum in a pyrolysis mode;
Case 3: a MHF that processes sludge,
scum and refuse derived fuel (RDF) also
in a pyrolysis mode.
The values assumed for the costs and the
variables associated with power, fuel, and
chemicals that were used to compare the
three options are listed in Figure 3. The
solids handling flowsheets and economic
data associated with these three options
are given in Figures 4 to 6. Both pyrolysis
flowsheets include chemical conditioning,
filter pressing and energy recovery equip-
ment, while the incineration flowsheet has
no energy recovery equipment and uses
centrifugation instead of a filter press. It
should be stated that building, foundation
and other peripheral expenses not directly
related to the handling equipment were
omitted from the calculations.
A perusal of cases 1 through 3 contrasts
the economic differences between a con-
ventional incineration system and that of
pyrolysis. This comparison demonstrates
how a pyrolysis feed mixture of RDF and
sludge can actually generate a positive •
revenue (Figure 6). The energy recovery
savings in Case 2 could significantly re-
duce the cost (per ton of dry solids) of
sludge handling compared to Case I (Fig-
ure 4) which has no energy recovery bene-
fits. Case 2 is able to recover enough
energy to supply 55% of the entire annual
energy requirement for the sewage treat-
ment plant. These savings are roughly
equivalent to a third of the annualized
costs for the solids handling process.
The economic advantages of pyrolysis are
further accentuated when a mixture of
sludge and RDF are processed. In such
situations pyrolysis systems have the ad-
vantage over conventional thermal sys-
tems by producing revenue. As more RDF
is added to the sludge feed the dry solids
content of the feed is elevated, which
serves to increase the amount of recovera-
ble steam energy. Additionally, the prob-
lem of solid waste (RDF) disposal is also
achieved. For example, in Case 3 where 12
parts of RDF are added per equivalent of
sludge, enough energy can be recovered to
offset the entire annualized costs of the
solids dewatering and pyrolysis processes
with enough energy left over to equal a
revenue of $1.1 million (834,000 rubles).
Hence, pyrolysis can be used to dispose of
large quantities of RDF (solid waste) and
sludge, and can recover energy that could
be useful as a revenue source to offset the
costs of sewage treatment. The prospect of
revenue production is an opportunity that
most, if not all, other types of solids
processing systems cannot offer.
Summary
In conclusion, it should be reiterated that
pyrolysis in a multiple hearth furnace has
a number of advantages over conventional
incineration, which are as follows:
1. Improves process control;
2. Prevents clinker formation (material
fusion);
3. Offers economical energy recovery;
4. Increases furnace capacity;
5. Offers potential revenue from steam or
power;
-------
6. Reduces paniculate Loading prior to
scrubbing;
7. Offers a permanent solution to solids
(RDF) disposal, and
8. Has numerous industrial applications.
Pyrolysis in a MHF has in common with
MHF incineration a number of advan-
tages compared to other solids handling
systems, some of which are:
1. Meets U.S. EPA emission regulations;
2. Offers permanent solution to sludge
disposal;
3. Has been proven technologically on a
full scale, and
4. Avoids bacterial, viral and organic
contamination of ground water by ther-
mal destruction.
References
Turning Waste Materials into Profits,
Canner Packer, July 1971.
Nut Shells and Pits Reduced to Profit,
Actual Specifying Engineer, October
1971.
Jones, J.L., D.C. Bromberger and F.M.
Lewis, The Economics of Energy Usage
and Recovery in Sludge Disposal, 49th
Annual Conference of the Water Pollu-
tion Control Federation, Minneapolis,
Minnesota, October 6, 1976.
Shannon, E.E., D. Plummer and P.J.A.
Fowlic, Aspects of Incinerating Chemical
Sludges, Wastewater Technology Centre,
Environmental Protection Service, Envi-
ronment Canada.
Bracken, B.D., J.R. Coe and T.D. Allen,
Full Scale Testing of Energy Production
from Solid Waste.
Sahagian, J., Economics of 50 MGD
Pyrolysis Systems, Eimco BSP Division of
Envirotech Corporation, March 1977.
-------
steam
multiple hearth
furnace
Figure 1.
Sludge Pyrolysis and
Heat Recovery
1. Thickner
2. Pyrolyzer
3. 1500° F Combustion
4. Waste heat recovery boiler (turbine generator optional)
5. Cleansed air
6. High solids filter
7. Purified ash with phosphate
10
-------
dewatered cake
&.
* • ^
• •.,• •*•'«' V
. •• . • . » ' 1
»»**«.*•/•;>
?J&
>'.&•
«
t» •
• *.* ""
' t *. «
.*• ".''• » »
»
* »
.'•».'
> * * •
• t
V
•'•- •' •*,*
• •
• •:•
:>'•-•.-:
. ' . . ,,
• \
•,'•'• v • *' \ '.
•: .".':'
;t
•
•
p •
i •
•
1
sludge
drying
zone
active
pyrolysis
/one
some pyrolysis
and fixed
carbon burning
char/ash
Figure 2.
Pyrolysis
11
-------
Basis for Economic Calculations
50 MOD (190;OOOM3/day) — primary + secondard
Power
Chemicals
Fuel Oil
RDF
S0.03/KW-HR
$75/Ton
S0.40/Gallon
$10/Ton
Non heat treat sludge
Heat treat sludge
Scum
RDF
Continuous operation
Capital investment amortized over 20 years
Municipal bonding rate of 6%
No water, building or labor cost included
(2 Kopecks/KW-HR)
(25 R/1,000 KG)
(1.2R/Liter)
(3.4 R/1,000 KG)
10,000 BTU/# Volatile*
12,000 BTU/# Volatiles
16,000 BTU/# Volatiles
8,500 BTU/# Combustibles
8,760 HRS/Years
Figure 3.
Case 1: Solids Disposal Costs For a 50 MGD Plant
SLUDGE
(26,000 KG/DAY)
FUEL
THICKENED
fe.
p
SLUJGE
CHEMICAL
CONDITIONING
-+
CENTRI-
FUGATION
58 TPD D.S.
t-
23% SOLIDS
61% V.S.
INCINERATOR
4
L SCUM
Capital Costs & Energy Requirements
Power Fuel Power
Installed Capital Consumption Consumption Generation
Equipment Cost (KW.) (GPD) (KW)
(1,200 KG/
2.64 TPD 0.5
40% SOLIDS
95% V.S.
Centrifugation
Incinerator
Total
Capital
Power consumption
Chemicals
Fuel
Total cost
Power revenues
$ 250,000
$2,268,000
$2,518,000
Costs
ion
1. (S/Ton)
( 190,000 R)
( 1, 71 8,000 R)
(1, 908,000 R)
(1
154
146
300
,000'sof$)
220
79
279
116
694
0
37.4
796
796
(1,000 Rubles)
167
60
211
88
526
0
28
Figure 4.
12
-------
Case 2: Solids Disposal Costs For a SO MGD Plant
THICKENED
SLUDGE .
CHEMICAL
CONDITIONING
->
FILTER
PRESS
SLUDGE
(26,000 KG/DAY)
58 TPD D.S.
35% SOLIDS
PYROLYZER
617. V.S. A
STEAM
""POWER
Capital Costs & Energy Requirements
Power Fuel Power-
Installed Capital Consumption Consumption Generation
Equipment Cost (KW.) (GPD) (KW)
SCUM
(1,200 KG)
2.64 TPD D.S.
40% SOLIDS
957. V.S.
Filter Press
Pyrolyzer
Total
Annualized Costs
Capital
Power consumption
Chemicals
Total cost
Power revenues
Net cost
Cost per ton D.S. (S/Ton)
$ 603,000
$1,950,000
$2,553,000
( 457,000 R)
(1, 340,000 R)
(1, 797,000 R)
I)
135
146
757
757
(1,000'sofS)
266
38
279
583
199
384
20.7
(1,000 Rubles)
201
29
211
441
151
291
16
Figure S
13
-------
THICKENED
SLUDGE
Case 3: Solids Disposal Costs For a 50 MGD Plant
SLUDGE
(26,000 KG/DAY)
58 TPD D.S.
CHEMICAL
CONDITIONING
»
FILTER
PRESS
35% SOLIDS
61% V.S.
RDF
(724,000 KG/DAY)
1593 TPD D.S.
80% SOLIDS
90% COMBUSTIBLES
STEAM
PYROLYZER
POWER
Capital Costs & Energy Requirements
Power Fuel Power
Installed Capital Consumption Consumption Generation
Equipment Cost (KW.) (GPD) (KW)
SCUM
(1,200 KG/DAY)
2.64 TPD D.S.
40% SOLIDS
95% V.S.
Filter Press
Pyrolyzer
Total
S 603,000
$19,000,000
SI 9.603.000
( 457,000 R)
( 14,400,000 R)
( 14,857,000 R)
11
3,014
3,025
42,460
42,460
Annualized Costs
Capital
Power consumption
Chemicals
R.D.F.
Total cost
Power revenues
Net revenue
Revenue per ton D.S. ($/Ton)
(l,000'sof$)
1,719
795
279
7,268
10,061
11,158
1,097
59
(1,000 Rubles)
1,302
602
211
5,506
7,621
8,453
831
45
Figure 6.
14
-------
Waste Water
Treatment by
Physical and
Mechanical
Methods.
Mjasnikov I.N.,
Ponomarjev V.G.,
Nechaev A.P.,
Kedrov I.V.
VNII VODGEO, Gosstroi
USSR
Industrial development and municipal
improvements inevitably involve forma-
tion of wastes—. part of which is
discharged into water reservoirs, the same
time a successful realization of plans of
national economy is determined by
various factors including the quality of
water sources.
In this connection the modern human
activities and environmental control are
closely interrelated.
Considering the importance of this prob-
lem the majority of countries have taken a
number of measures maintaining the safe-
ty and cleanness of natural water sources.
Considerable allocations have been made
at present to solve this problem.
The difficulty of rendering safe the con-
taminants consists in the composition of
industrial and domestic effluents and their
big amount. However, the ever growing
requirements for the quality of water
treatment are not always met by existing
treatment plants. Therefore there is a
necessity of developing existing methods
and elaborating new ones with the
purpose of water sources protection. In
spite of sufficiently high level of rendering
industrial and domestic effluents much
attention is paid to these problems in the
USSR and USA. In this respect uniting
efforts and means of both countries to
solve the problems of water medium
protection will be efficient and mutually
profitable from an economical point of
view.
The most efficient measures taken at pres-
ent to protect water reservoirs against
contamination with industrial and domes-
tic wastes are waste water treatment, use
of closed water supply systems, working
out of technological processes which are
not associated with generation of waste
water, repeated use of treated industrial
and municipal effluents in technological
processes of enterprises etc.
Mechanical, physicomechanical, physico-
chemical and biochemical methods are
used to treat industrial and municipal
effluents. Depending on the composition
of effluents, local conditions and other
aspects use is made at enterprises of sys-
tems that include a combination of var-
ious methods and installations.
The methods and installations of physico-
mechanical treatment are widely used in
municipal services and industry. They
allow it to separate from waste water solid
mechanical impurities of mineral and
organic character as well as a number of
dissolved matter (for instance, surface
active matter). Depending on quality and
quantity of waste water it is treated by
various modifications of sand traps
settling tanks, hydrocyclones, centrifuges,
flotators. filters etc.
To intensify the process of treatment, use
is now made of some new methods such as
water treatment with magnetic or electric
field, vibration, ultrasound, various
reagents. In a number of cases these
measures bring about structural changes
or designing special treatment apparata.
The use of above mentioned installations
in each particular case is determined by a
number of factors and principally by wa-
ter properties, character of impurities it
includes and by required degree of treat-
ment.
These installations are designed on the
basis of kinetic curves of settling (Fig. I)
determined by experiments. These curves
characterize sedimentation properties of
suspended matter in waste water.
To model the sedimentation process a
number of relationships have been
suggested. Let us lake one suggested by
the institute under the academy of
municipal services:
T H"
T=iT <"•
where h. H — sedimentation height
accordingly in the model
and in a real installation;
t, T — corresponding sedimenta-
tion period in the model
and in a real installation;
n — agglomeration coefficient
of settling particles.
As proved by investigations, the "n" value
depends on many parameters of sedi-
mentation process including the origin of
suspended matter and may vary within a
considerable range from 0.2 to I.
Determination of "n" value (for recalcula-
tion of experimental characteristics for
actual installation) is done on the basis of
two sedimentation curves obtained in a
laboratory for different depths (ni and n2)
of sedimentation (floating) and assigned
^treatment efficiency (Fig. I)
n =
At'
/eh- — /?h<
As it has been already noted use is made of
various types of settling tanks. They can
be rectangular, round, cylindrical etc.
Depending on working flow direction
settling tanks are subdivided into hori-
zontal, radial and vertical (Figs. 2, 3, 4).
One of the widely used methods of water
treatment is sedimentation. This method is
used due to simplicity of installations, low
operational costs. These installations
permit removal of up to 40-60% of
mechanical impurities from water. 90%
and over oil products and to reduce BOD
and COD values.
The main disadvantage that reduces the
efficiency of employed settling tanks is a
considerable flow movement due to
imperfection of water collecting and water
distributing equipment as well as due to
the effect of convective flows as a result of
irregular temperatures in the installation.
That is why the volume of the settling tank
is not entirely used.
U.S EPA Heac
15
PA Headousters Library
Mail coor '"••' "
1200 Pennsylva-
Wasrriftgt6i1 ;
202-56fw
-------
' Investigations show that the volume
utilization factor (efficiency) of settling
tanks is, as a rule 50-60%, of radial—50-
60%. of vertical—40-50%.
The required period of working flow stay-
ing in the installation to provide necessary
degree of treatment is determined on the
base of sedimentation curves according to
the formula:
= const
(3)
When specifying dimensions of a settling
tank in accordance with the amount of
waste water one should consider the
volume utilization coefficient.
Due to a low efficiency of settling tanks
they require considerable installation
areas which is sometimes the reason for
their limited use.
One of the ways to increase the efficiency
of settling tanks is improving water
distributing and water collecting equip-
ment. An example of such improvement is
attachment of the perforated partition
installed at the inlet of a settling tank and
increasing the length of spillway edge.
The total area of holes in the partition is
taken 6-8% of the settling tank section. To
increase volume utilization coefficient of
vertical settling tank provision is made for
arranging distribution grates and
changing the inlet system (Fig. 5).
It has been determined that in installa-
tions with small settling depth the influ-
ence of density and convective flows on
the velocity of working flow of liquid is
miserable and its distribution provided at
the inlet remains practically unchanged
along the entire length of the settling tank.
This principle is used in structures of shelf-
type and tubular settling tanks.
A variety of thin-layer settling tanks is
known at present. In accordance with the
scheme of water flow and sediment move-
ment, all thin-layer settling tanks can be
subdivided into three groups.
The first group includes settling tanks
which operate according to cross scheme
when the sedi'ment creeps across the flow
movement; the second group — counter-
flow scheme, when the sediment creeps
against the flow movement and the third
group — settling tanks operating accord-
ing to straight-flow scheme when the
direction of sediment movement and that
of water flow coincide.
Hydraulic investigations of thin-layer set-
tling tanks that'operate according to cross
scheme arid is provided
distribution devices have been carried out
by VODGEO. The investigations have
shown that maximum efficiency of vol-
ume utilization up to 80-90% is achieved
in case of using proportional water dis-
tribution devices designed by VODGEO
(Fig. 7). Testing as thin-layer settling tank
in real conditions at one of the oil-refining
plants has proved the high efficiency of
this installation for oil removal. Under a
hydraulic load which is five times greater,
this installation provides the same
efficiency as an oil catcher which is 36 m
long. 6 m high and 2 m deep.
Investigations which are being carried out
at present on thin-layer settling tanks aim
at widening the field of their use.
The process of separating mechanical im-
purities from water is intensified in the
field of centrifugal forces. The degree of
intensification is characterized by the
separation factor.
Fr =
8.R
where V — linear speed of rotation.
m/sec.,
q — acceleration of gravity force,
m/sec/,
R — radius of rotation, m.
To clean waste water from roughly
dispersed impurities use is made of pres-
sure hydrocyclones (Fig. 8) where centri-
fugal field is maintained owing to tan-
gential inlet of feed water.
The industry of the USSR produces pres-
sure hydrocyclones from 9 to 500 mm dia.
Greater treatment efficiency is achieved in
hydrocyclones of small diameters. These
cyclones are assembled into blocks.
Hydrodynamic investigations on pressure
hydrocyclones show that their separation
factor can reach Fr = 5000 which indicated
a greater separating ability of these appa-
rata. However due to a short period of
flow staying in hydrocyclone and due to
turbulence caused by zones of flow
circulation, the provided treatment effi-
ciency is much less than theoretical one.
Another disadvantage of pressure hydro-
cyclones is the wearing of walls in case of
treating waste water contaminated with
abrasive suspended matter.
This feature confines the field of using
three apparata.
Pressure hydrocyclones are used as a rule,
for local treatment of small quantities of
industrial waste water contaminated by
roughly dispersed impurities.
They are used in chemical industry, build-
ing materials industry, at concentration
plants, for treatment of mine waste water
etc.
Of great use can be exposed hydro-
cyclones which according to the character
of separation process hold intermediate
position between settling tanks and hydro-
cyclones.
A relatively small speed of liquid.entering
this apparata results in a small hydraulic
resistance which does not exceed 0.5 m w.c
and in much greater efficiency as
compared to that of pressure
hydrocyclones.
Rotation movement in exposed hydro-
cyclone enables agglomeration of sus-
pended matter which increases the effi-
ciency of the sedimentation process.
Therefore hydraulic unit loads (m-Vm2hr)
on these installations are greater than
those on se'ttling tanks.
Several structures of exposed hydro-
cyclone have been designed by VODGEO
research institute. Of great interest is the
hydrocyclone with internal cylinder and
tapered membrane located in the upper
16
-------
part (Fig. 9) as well as multistage hydrocy-
clone.
In the first structure water is fed to the
space confined by the internal cylinder.
This cylinder enables it to create a closed
water flow-which delivers suspended
matter separated from the rising flow to
the tapered part. The membrane located in
the upper part of the apparata as well as
the end wall at the end of the horizontal
settling tank prevents suspended matter
which is moving together with a rising
flow from discharge.
Industrial tests of this structure on waste
water from rolling mills have shown that
the same treatment efficiency can be
achieved under the load which is 2.5 times
greater than that on settling installations.
The principle of thin-layer sedimentation
realized owing to subdivision of structure
volume into separate tiers by means of
tapered membranes is used in multistage
hydrocyclone.
The rotation movement of working flow
achieved by tangencial inlet of initial wa-
ter through three vertical slots which are
common for all tiers contributes to
agglomeration of suspended matter and to
better volume utilization of tiers (70%).
Testing multistage hydrocyclone for treat-
ment of waste water from rolling mills has
shown that the required treatment
efficiency is obtained with hydraulic loads
several times greater than those on ordi-
nary settling tanks.
Unlike the pressure hydrocyclones the
exposed ones can be used for separation of
coagulated suspended matter. Rotation
movement intensifies coagulation and
flocculation processes.
Of interest are also other apparata of
physicomechanical treatment where con-
taminants are separated under the action
of centrifugal forces. These are bowl cen-
trifuges.
The main advantage of bowl centrifuges as
compared to settling installations is the
compactness of installations owing to
great hydraulic loads with higher clarifica-
tion efficiency as well as the possibility of
simultaneous sediment treatment.
This advantage plays an important part
when chosing the proper method of waste
water treatment for enterprises with limit-
ed areas.
Very advantageous types of centrifuges for
industrial sewage treatment are pendulum
(basket-type) centrifuges of interrupted
action and scroll centrifuges of con-
tinuous action.
Pendulum centrifuges (Fig. 10) afford to
achieve complete clarification of waste
water and sufficient dehydration of sedi-
ment. However, the intervals in operation
due to the necessity of removing the sedi-
ment make the field of its use limited.
Investigation carried out by VODGEO
have shown that treatment of waste water
with contaminant concentration over
3 g/1 is not reasonable.
Particularly efficient is the use of pen-
dulum centrifuges for local treatment of
waste water generated at enterprises with
interrupted technological process.
Scroll centrifuges (Fig. 2) arc much more
efficient as compared to pendulum ones.
However, as shown by investigations, the
efficiency of waste water clarification in
these apparata depends on the properties
of waste water and generated sediment
and in some cases is lower than in pen-
dulum centrifuges.
Investigations carried out by VODGEO
have shown it reasonable to use low-speed
centrifuges with a separation factor up to
Fr = 800. In this case it is possible to design
high capacity centrifuges. That is why fur-
ther investigations carried out at present
are aimed at studying their structural
pecularities with the purpose of designing
special high efficient installations for waste
water clarification.
One of the widely spread methods of waste
water treatment and sediment com-
paction is flotation. This method allows
removal of mechanical impurities from
waste water as well as fine-dispersed
matter, dissolved and surface active
matter and in some cases even ions and
molecules.
This method can be successfully used to
remove substances with water-repellent
properties from waste water.
The possibility of getting air or gas bub-
bles having the same sizes as those of re-
moved particles ensures high efficiency of
treatment and affords to get the required
water quality since flotation plants can be
automatized.
Flotation is used in home practice for
industrial waste water treatment in such
branches of industry as chemical, oil-
producing, oil-processing, metallurgical,
machine building, food, cellulose-paper,
textile etc. as well as for treatment of
municipal effluents.
The capacity of flotat: jn plants varies
within a considerable range and reaches
the load of 1500 m'/hr and over per
frother cell. The diameter of such installa-
tion can be up to 30 m.
In most cases it is possible using flotation
to remove the basic amount of suspended
matter, surface active matter, oil prod-
ucts, fats, oils etc. The efficiency of con-
taminant removal reaches 60-80% and in
case of reagents use — up to 90-95% and
over.
Flotation is also used of late for final
purification of biologically treated waste
water.
In this case suspended matter and some-
times hardly oxidized matter which have
passed aeration tanks are removed from
water. In case of pressure flotation, for in-
stance, a considerable amount of oxygen
is dissolved. Oxygen content in treated
water exceeds by 30-50% its concen-
tration.
Various schemes of flotation plants and
frother cell designs as well as combined
installations are used in practice: hydro-
cyclones-flotators, flotators-settlers, fil-
ters-flotators, flotators combined with rea-
gent mixing and flocculation chambersetc.
17
-------
Available experience and investigations of
flotation show that this method of treat-
ment-compression (pressure) flotation,
impeller, frothy, electroflotation etc. will
be still wider used for waste water
purification and sediment treatment.
Installations with filtering charge are
widely used in systems of physico-
mechanical treatment. As the working
medium, use is made of various natural
materials such as for instance adequate
fractions of sand, hard coal, rock as well
as specially prepared charges of slag.
ceramics, polymer materials. Metal nets
and various filtering arrangements are
used for this purpose. Working medium of
the filter can he arranged from one
material (two-three layer) which in turn
determines technical and economic
characteristics of installation operation.
In most of the cases filters are used for
considerable water cleaning from sus-
pended matter and are most frequently
used in treatment systems of industrial
plants and municipal services after aera-
tion facilities. In home practice, for in-
stance sand filters are used at treatment
plants of oil-refineries and some chemical
industries before aeration tanks. The ca-
pacity of filters varies within a consid-
erable range and depends on a number
of conditions. The rate of filtration for
single-layer and double-layer sand filters
used in home practice can be as a rule,
within 3-20 m/ hr.
For instance the rate of filter used for final
purification of biologically treated waste
water at one of the treatment plants near
Moscow reaches 9 m/hr with the effi-
ciency of suspended matter removal about
90% and reduction of BOD5 value over
70%. As working medium in this case use
is made of 1-2 mm sand fractions with the
height of layer 1.2 m and of 0.8-2.5 mm
coal fractions, with the height of layer
0.6m.
Utilization of specially prepared charge
makes it possible to increase filter capacity
and to achieve considerable treatment effi-
ciency. For instance, investigations carried
out by VODGEO have shown that the use
of filter charged with polyurethane ma-
terial (the size of particle about I x I x 1
cm, the height of the layer 1-1.2 m) pro-
vides the removal of oil products from a
number of waste water categories by 90%
with the rate of filtering being up to
20 m/hr.
Filtration of waste water as has been
already mentioned can be in some cases
combined with settling and flotation. The
use of such installations is particularly
efficient in systems of local installations
and treatment of small amounts of waste
water and provides reduction of required
areas and reduces operational costs.
The processes and apparata touched upon
in this report do not cover all types and
structures of physicomechanical installa-
tions used for treatment of industrial and
municipal waste water.
In spite of the fact that they are most fre-
quently used in schemes of treatment
plants and that considerable operation
experience is available these installations
still need further development both in
respect of technological process improve-
ment and design appearance.
In this connection attention should be
paid in future to the following aspects:
• improving the existing installations for
physical and chemical treatment, increas-
ing their capacity and operational effi-
ciency;
• creating combined settling installations
which include filtration, flocculation and
flotation zones;
• investigating new kinds of filtering
charge for filters;
• designing new structures of settlers,
flotators, centrifuges, hydrocyclones, fil-
ters to be used for intensification of proc-
ess of magnetic and electric field, vibra-
tion, ultrasound etc.
18
-------
Settling Duration
Figure 1.
Kinetics of Insoluble Solid Settling
i-o.od
Figure 2.
Horizontal Settling Tank
1. Raw water inlet
2. Clarified water outlet
3. Scraping mechanism
4. Sludge drain
5. Grease Collected trough
19
-------
Figure 3.
Circular Settling Tank
1. Raw water inlet
2. Clarified water outlet
3. Scraping mechanism
4. Sludge drain
Figure 4.
Vertical Settling Tank
1. Influent trough
2. Effluent trough
3. Sludge drain
4. Floating bodies outlet
Figure 5.
Vertical Settling Tank
with Peripheral Inlet
1. Raw water inlet
2. Clarified water outlet
3. Sludge drain
20
-------
Figure 6.
Scheme of Operation of Thin Bed Settling Tank
1. Crossflow scheme
2. Uniflow scheme
3. Counterflow scheme
Direction of Water Movement
Direction of Sludge Movement
Figure 7.
Thin Bed Settling Tank of Crossflow Type
I. Raw water inlet
2. Water proportioned distributor
3. Clarified water outlet
4. Scraping mechanism
5. Sludge Drain
6. Grease collected trough
21
-------
raw water
thickened product
Figure 8.
Pressure Hydrocyclone
1. Cylindrical part
2. Conical part
3. Discharge nozzle
4. Slime nozzle
5. Overflow nozzle
diaphragm
pseudowall
Figure 9.
Open Hydrocyclone
effluent
suspended
solid feed
Figure 10.
Sedimentation Centrifuging with Continuous
Suspended Solid Feed and Periodical Sludge Discharge
22
-------
Physical Treatment
of Oil Refinery
Wastewater*
William J. Lacy, P.E.**
Introduction
The treatment of oil refinery process
waste-waters usually involves a series of
process steps as shown in Table I.
This paper will be divided into two parts
the first part defines physical treatment
processes, i.e., gravity separation,
stripping, solvent extraction, adsorption.
combustion, and filtration. The second
part of the paper describes an
Environmental Protection Agency
demonstration project on a actual refinery
effluent water treatment plant using
activated carbon.
As background physical treatment data on
refinery wastewater, let us compare several
types of units under various loadings.
Typical efficiencies of these oil separation
units are shown in table II.
Physical Treatment Processes
Physical treatment processes commonly
used in treatment of refinery wastes
include gravity separation, flotation,
stripping processes, adsorption,
extraction, and combustion. The waste
from a refinery plant may require a
combination of these processes.
Gravity separation includes the removal of
materials less dense than water such as oils
and air-entrained particulates by flotation
and the removal of suspended materials
which are more dense than water by
sedimentation. Sedimentation and
flotation techniques commonly use
chemicals to enhance the separation
process. Wastewaters often contain
*To be presented at the Joint US-USSR
Symposium on Physical Treatment held in
Cincinnati, Ohio, April 5-6, 1977.
** Principal Engineering Science Adviser,
Office of Research and Development, U.S.
Environmental Protection Agency,
Washington. D.C. 20460
quantities of free and emulsified oil which
must be removed prior to subsequent
treatment. Free oils are much easier lo
remove if their concentration is high. Slop
oils that are recovered by the separation
process can be cleaned and reused.
The most commonly employed separator
design is that presented by the American
Petroleum Institute. Although some
reduction in chemical oxygen demand
(COD) can be expected due to removal of
oils and tars, little or no biochemical
oxygen demand (BOD) removal will be
prevalent.
Oil emulsions are the major problem of
oil-water separation. Emulsifying agents
prevent the oils from coalescing and
separating from the water phase. .
Emulsifying agents are surface-active
agents and include catalysts, the sulfonic
acids, naphthenic acids, and fatty acids, as
well as their sodium and potassium salts.
In an alkaline medium, calcium and
magnesium salts form finely divided
suspended solids which stabilize the
emulsions.
To separate the emulsified oils from the
wastewater, the emulsion must be broken.
The application of heat and pressure is the
more effective method used in de-
ernulsification of a waste. Distillation
methods, are also effective in breaking
emulsions as are filtraton, acidification,
and electrical methods.
Sedimentation in the pre- or primary
treatment of refinery wastes with high
suspended solids concentrations, in
secondary clarification, and for sludge
thickening. Wastewaters high in collidal
material must be chemically treated before
adequate separation by sedimentation can
be obtained. The removal of solids and
oils from Wastewaters and the concentra-
tion of sludges can be accomplished using
air flotation.
Gravity oil separators usually precede
flotation units in most industrial
applications. One advantage of flotation
over sedimentation is the shorter
detention time required to clarify a waste
by flotation, resulting in a smaller unit.
Stripping processes are used to remove
volatile materials from liquid streams.
These methods are employed generally to
remove relatively small quantities of
volatile pollutants from large volumes of
wastewater. Stripping is a low-
temperature distillation process whereby
reduction of effective vapor pressure by
the introduction of the stripping medium
replaces the high temperature
requirement. The two types of stripping
agents commonly used are steam and inert
gas.
The stripping of hydrogen sulfide and
ammonia from sour water is the most
common use of stripping. Stripping agents
used to remove these contaminants are
steam, natural gas, and flue gas. Phenols
can be removed from aqueous waste
streams by steam stripping which is
applicable when a wastewater is subject to
short variations in termperature, specific
gravity, phenol concentrations, and
suspended solids.
The stripping rate is a function of
temperature, the stripping gas flow rate,
and tank geometry. Laboratory testing
has indicated that most of the BOD
removal during the stripping of
biodegradable volatile organic
compounds was the result of biological
action rather than physical stripping.
Solvent extraction methods utilize the
preferential solubility of materials in a
selected solvent as a separation technique.
The criteria for effective use of a solvent in
wastewater treatment include (a) low
water solubility, (b) density differential
greater than 0.02 between solvent and
wastewater, (c) high distribution
coefficient for waste component being
extracted, (d) low volatility and resistance
to degradation by heat if distillation is
used for regeneration or low solubility in
liquid regenerants, and (e) economical to
use. Equipment used for extraction of
wastewater include counter current
towers, mixer-settler units, centrifugal
extractors, and miscellaneous equipment
of special design.
Adsorption is the process whereby
substances are attached to the surface of a
-------
solid by electrical, physical, or chemical
phenomenon. A carbon medium has been
the most successful adsorbent in removing
certain refractory chemicals from
wastewaters. Phenols, nitriles, and
substituted organics are also adsorbed by
carbon when present in low
concentrations.
Combustion processes are feasible for
disposal of some refinery wastes which
may be too concentrated, too toxic, or
otherwise unsuitable for other methods of
disposal. Combustion may be either direct
or catalytic, depending on the waste to be
oxidized.
Submerged combustion has been used
successfully in the total or partial
evaporation of waste streams as well as
concentrating dissolved solids. This
method produces an effluent which either
has reuse value or which is easier to
dispose of than large volumes of the liquid
waste. Incineration is the commonly used
combustion process for refinery wastes.
Fluidized bed incinerator can be used for
burning oily sludges. The fluidized bed
incinerator provides better controlled
combustion with lower requirements for
excess oxygen than conventional
incinerators lor oily sludges. Incineration
occasionally converts a water pollution
problem into an air pollution problem.
Filtration processes are used to remove
and concentrate solids on oily materials
from a waste stream. A filter can be
specifically designed to remove small
quantities of these materials as a final step
in wastewater treatment, or it may be used
to concentrate a waste so that further
treatability of wastewater will be
enhanced. A polishing filter employing
sand filtration can be used to remove
additional suspended material.
EPA Demonstration Project
Petroleum refineries are faced with the
problem of disposing of hugh volumes of
wastewater. These large amounts of water
come from a wide variety of sources.
Sources include process water used for
heat transfer, wash water, and rain runoff
which collects oils and chemicals from
within the refinery.
Separation of the visible, floatable oils
was considered satisfactory to allow
discharge into adjacent waterways. To
protect our waterways from these harmful
discharges, new and improved technology
is needed.
One major pollutant existing in refinery
discharge waters is the oxygen demanding
material. Oxygen demand, by chemicals
and oils lowers the water's available
dissolved oxygen content, a vital need for
marine life.
The US EPA awarded a $274,719 grant to
Atlantic Richfield Company as the
government's share of a $1,159,584 project
(28% of cost) at the Watson Refinery in
Los Angeles. Atlantic Richfield
Company's Watson Refinery is one of
about 16 industrial facilities discharging
water into the Dominguez Channel. The
Water Refinery was limited to 1000 kg per
day of COD in its discharge water to the
Channel. Meeting this requirement meant
reducing the COD in its discharge waters
by 95% if the water was discharged to the
Channel.
The Watson Refinery made an agreement
with the Los Angeles County Sewer
District to put its process wastewater
through the County's primary treatment
unit. Due to hydrolic limitations in the
unit, the County was unable to handle rain
water runoff.
During periods of rainfall, the process
wastewater and rain water mixture which
were interconnected could not be sent to
the sewer district facilities due to the
presence of rain water, nor could it be sent
to the Dominguez Channel due to high
COD from the process wastewater. A
system was needed which would treat all
the process water plus rain water as it
was produced, or an impounding plus
processing system which would allow
large volume impounding during the rain
followed by low volume processing. A
system was needed which could be started
up easily when rain fell and shut down
when not required. Biological units
require continuous feed, therefore,
conventional technology was not ,
satisfactory. It was decided to use
impounding followed by activated carbon
treatment to adsorb the COD material
from the impounded rain diluted process
water.
Construction of the first commercial sized
carbon adsorption plant for treatment of
petroleum refinery wastewater was
completed in 1971 and contained over
one-half million pounds of activated
carbon. Activated carbon was used during
the two-year project to adsorb organic
chemical oxygen demand materials from
the impounded rain water and process
water mixture. The adsorbent carbon used
is a granular activated carbon of 8 to 30
mesh made from bituminous coal. It has
high hardness standards to minimize
attrition loss in handling, regeneration,
and hydraulic transport. It has a broad
spectrum of pore sizes to meet adsorption
requirements for a broad range of organic
molecule sizes.
Thermal Regeneration
The carbon is regenerated by selective
oxidation of the organic impurities in the
pores, at high temperatures (900° C -
970° C) and with a controlled low oxygen
atmosphere in a multiple hearth furnace.
As the carbon is heated, the more volatile
organic compounds are vaporized. With
further heating, additional organics are
pyrolysed. The remaining organics are
then oxidized selectively by addition of
air. The carbon is then quenched in water.
Time, temperature, and atmosphere are
the controllable parameters for
regeneration. Free oxygen must be
carefully controlled in the lower hearths of
the furnace to avoid burning up the
regenerated carbon.
Chlorination
Chlorine can be added to the effluent
stream to permit operating the carbon
beds with some breakthrough of organics
because chlorine will reduce organic COD
level as well as inorganic. This provides
greater flexibility in loading the carbon or
in handling an unusual COD load.
24
-------
Description of Activated Carbon Plant
The water treatment facility is made up of
four main systems plus an impounding
reservoir. Figure 1 describes this system.
Reservoir
The reservoir is a 1.2 million barrel
holding basin which impounds all refinery
wastewater and rain water runoff when the
Los Angeles County Sewer District
cannot accept it.
Water Treatment Carbon Adsorption
Unit
The water treatment unit is shown in
Figure 1. This unit reduces the organic
COD of water impounded in the refinery
prior to discharge to the channel. The
water treatment unit consists of twelve
identical adsorber cells. (V-l through
V-12) each 3.7m x 3.7m metric 1 x 7m
deep. Each cell originally contained a 4m
deep bed of carbon having a dry weight of
approximately 22,100 kg. Supporting the
carbon bed is a one-foot layer of gravel on
top of a Leopold underdrain system. The
depth of the carbon was altered in 1972 for
reasons discussed later in this report.
Impounded water is delivered to the
influent water distribution trough through
a 35cm line from the impounding
reservoir. The water is distributed to any
or all of the twelve adsorber cells (V-l
through V-12) by slide gates. Flow to each
cell is regulated by a handwheel operated
slide gate.
The regeneration furnace is a 140cm l.D.j
six hearth multiple hearth unit. A diagram
of the furnace is shown in Figure 2. It is
gas-fired on two hearths. A center shaft
rotates arms with teeth which move the
carbon across the hearths and downward
through the furnace. The burners on the
furnace automatically control furnace
temperatures at the desired levels via
thermocouple element and controllers.
Steam and air addition rates are manually
set.
The afterburner section is separately gas-
fired to raise off-gas temperatures to about
800° C. This is required to combust
organic vapors in the furnace exit gas.
The hot gas from the after burner goes
through a quencher and is cooled by water
injection; is pulled through induced draft
fan (K-4) again with water injection for
, scrubbing, and exits through an
entrainment separator where entrained
water and particulate matter scrubbed out
of the off-gasses are removed. The clean
flue gases are exhausted via stack to the
atmosphere. Hot air from the center shaft
is added to the stack to reduce the
humidity and minimize the vapor plume.
Regenerated carbon discharges down a
chute from the furnace in periodic slugs as
the rabble arms pass over the drop hole on
the bottom hearth. This chute has two
legs. Normal carbon flow is vertically into
the quench tank (V-19). Water level in the
quench tank is kept above the bottom of
the Chute to prevent air from being drawn
into the furnace. A trash screen is
provided in the quench tank to protect the
eductor.
The 45° legal on the furnace discharge
chute is used to bypass the quench tank in
case transfer problems are encountered.
Opening the dump gate permits hot
carbon to drop directly into drums and
allows continued furnace operation.
Water sprays are provided to quench the
hot carbon to eliminate sudden evolution
of steam which would upset furnace
pressures. Quenched carbon is stored
(V-l7) until needed for refilling an
adsorber cell.
Design Basis
The volume of carbon in the adsorbers
and its exhaustion rate are set by the
volume of wastewater to be treated, the
concentration and types of adsorbable
material in the influent, and the
permissible COD concentration in the
effluent water. Design of the full scale unit
was based on pilot tests performed on
diluted refinery wastewaler.
The carbon exhaustion rate was difficult
to establish from the pilot plant tests
because the influent COD concentration
varied considerably, both above and
below the limits expected under actual
operating conditions. The actual design of
the unit was based on the following
criteria:
30 days per year of rain (maximum)
300,000 barrels water per rainy day
(maximum)
9,000,000 barrels of water per year
(maximum)
250 ppm COD average influent
concentration
37 ppm COD average effluent
concentration
500 grams of carbon exhausted per 1000
gallons water treated
The unit was designed to handle 100,000
barrels of water per day. Thus, the unit
would run 90 days if the maximum rains
were received without having to replace or
regenerate any carbon. Based on this
operation premise, regeneration of all
carbon would be done during the summer,
non-rainy season. The regenerator furnace
was designed to regenerate 3,900 kg
pounds of carbon per day. This is
equivalent to 11.3 GPM of slurry from
V-l 8. The approximate utility consump-
tion, based on design, was as follows:
Steam - I kg of steam per kg of carbon =
320kg/hr.
Refinery fuel gas (including after burner) =
3,000 SCFH
Quantities and Costs Based on Conditions
During the Project
During the two-year project, 172,040,000
gallons of water was processed to load
747,000 kg of carbon with 165,000 kg of
COD. This resulted in an average carbon
loading of 100 grams of COD per 400
grams of carbon. The Carbon was used at
the rate of 4.5 kg per 1000 gallons of water
treated. The average feed COD was 249
ppm and the average effluent was 50 ppm.
Averaged data for each of the three
periods of operation is given in Table 9.
The data in Table 111 shows that the
second rains single stage operation had a
slightly higher loading than the first rains
single stage operation, and a lower
effluent COD.
The overall average cost to operate the
25
-------
plant was 49C per 1000 gallons of water
treated, or 524 kg of COD removed from
the water.
With the improved operation of the
second year, the plant demonstrated the
ability to operate at 40c per 1000 gallons
of water treated, or 40i per kg of COD
removed.1
It might be noted that oil refining is the
most capital-intensive industry in the U.S.
with an investment of $108,000 per
employee, according to the Conference
Board, a private group. Least capitol-
intensive is the apparel industry which has
about $5,000 invested per employee.
In closing, let me point out that
environmental expenditures are on the
increase in all nations of the world. In
table IV, it is shown that the cost for
environmental control for all U.S.
industry was reported by Chemical Week
magazine. May 1976 issue, to be 17.1
billion dollars to meet the EPA 1976
regulations. The cost to the U.S.
petroleum industry is 2.1 billion dollars.
Data on the USSR, Japan, and Sweden
are also included in the table for
comparison.
There is little doubt that environmental
protection will cost both in money and
energy; but the price will be paid either by
money or by degradation of our
environment, and even out health. A
balanced approach by all nations is the
desired goal. This symposium is one of the
ways of achieving this goal.
1 Additional detailed data on this project can be
found in an EPA Report $66012-75-020 entitled
"Refinery Effluent Water Treatment Plant
Using Activated Carbon," June 1975
Improvement of
Hydraulic
Conditions of Radial
Settling Tanks
Skirdov I.V.
Ail-Union Scientific-Research
Institute "VODGEO"
Radial settling tanks have found a wide
application in the waste water treatment
practice. This type of installation is
successfully used for purification of
municipal and industrial wastes, which
nature, concentration and granulometric
composition of suspended solids (SS) vary
within a wide range.
Radial settling tanks are mainly used for
clarification of recirculating waters of
blast furnace gas scrubbing, conveyer
wastes of sugar mills characterized by a
high content of SS. But they are practised
on the largest scale in the field of
municipal sewage treatment at treatment
facilities of medium and high capacity.
In spite of the fact that the field of applica-
tion of rectangular settling tanks extends
due to the tendency to combine installa-
tions in one unit and thus to save required
spaces, radial settling tanks still remain
the most advantageous and profitable for
use at large treatment facilities.
Though the hydraulic conditions of rec-
tangular settling tanks are somewhat
better than those of radial ones, the latter
have indisputable advantages over rec-
tangular settling tanks, as the design of
scraping devices of radial settling tanks is
more simple and reliable in operation than
any of known modifications used in rec-
tangular settling tanks.
Besides, the circular configuration of the
basin provides greater strength of outside
walls and considerable saving of building
materials.
The most serious of them is a relatively
low efficiency of the volume usage.
According to experimental data obtained
by different authors this value ranges from
30 to 60 per cent and averages 45 per cent
(I).
When influent enters the central part of
radial settling tanks and moves towards
the periphery the expansion of the work-
ing jet takes place causing its hydro-
dynamic instability. Under these condi-
tions the working jet is subjected to dense
circulation and turbulent pulsations
preventing the volume of the tank from
being used effectively.
Various modifications of distributing
devices located in the central part of the
tank cannot significantly improve
hydraulic conditions of radial settling
tanks. In most cases the expansion of the
ireas required by collecting devices has
ippeared to be even less effective as the
obtained result is not worth the complica-
tion of the design.
Inlet of the influent from the periphery
:urned out to be rather effective (2). Rapid
reduction of the feeding velocity followed
by the slow speeding-up of the flow which
are characteristic of these conditions, have
increased the stability of the working jet.
The investigations have shown that the
advantages of this way of the influent
distribution diminish as the dimensions of
the installation increase.
At present radial settling tanks with
peripheral inlet are being tested in opera-
tion.
Large sizes of collecting devices as well as
their complexity are among serious short-
comings of radial settling tanks. For
example in large installations (40 m in
diameter) the length of water collecting
troughs and weirs which manufacturing is
very labour-consuming, amounts to
hundreds linear meters.
In settling tanks with peripheral inlet the
same difficulties take place during installa-
tion and operation of water distributing
devices.
A new type of the radial settling tank has
been recently used in the Soviet Union.
The water distributing device of the tank is
26
-------
Section A-A
Figure 1.
Settling tank with rotating distributing arm.
27
-------
continuously rotating together with the
scraping arm. The design of the tank was
developed at VODGEO Institute (3), the
project was madeout at Sojuzvodokanal-
project Institute.
General view of the settling tank is
represented in Fig. 1.
The basin of the settling tank has circular
configuration similar to that of radial
tanks. Influent pipe (2) passing to the
central part of the tank is conjugated with
the central part (4) of the distributing
device with the help of air valves (3). This
device is connected with the sludge scrap-
ing arm (5) and rotates around the central
axis. It consists of distributing trough (6)
furnished with the slotted bottom (7), a
baffle (8) and a series of jet guiding blades
(9). The distributing trough is conjugated
with the collecting trough (10), compris-
ing the weir (11) and the solid bottom (12).
In front of the weir (II) there is the half-
submerged baffle (13) for collection of the
floated matter.
The scrapers (14) are hung up to the
bottom part of the distributing trough by
means of hinges. A chamber with sub-
merging funnel (15) for removal of floated
matter is envisaged at the side wall of the
settling tank's basin.
In the centra! part of the settling tank
there is a sludge collecting hopper (16).
Clarified effluent and sludge are removed
through pipings (17) and (18).
The settling tank with the rotatory dis-
tributing device operates in the following
manner: The influent enters through the
piping (2) to the central part of the settling
tank (4) and then to the distributing
trough (6). By means of the baffle (8) and
blades (9) water is directed to the spaces
between the blades and comes out to the
settling tank. Distribution is carried on in
such a way that detention times for each
separate jet are equal, that is the flow from
the trough to the settling tank is increas-
ing from the centre to the periphery. The
collection of the clarified effluent is carried
on the same way.
The slots in the bottom (7) of the dis-
tributing trough prevent the sand and
other heavy solids, dragged along the
bottom from being collected in the trough.
When the distributing device moves in the
direction opposite to the direction of the
water, flowing out of it, reduction of the
flow rate in the settling tank occurs. If
velocities of the water exhaust and the
distributing device rotation are equal the
state close to the static is observed in the
settling tank.
The attention should be paid to the fact
that by changing rotation speed of the dis-
tributing device one can regulate the
velocity of the water movement in the
settling zone irrespectively of the wastes'
flow and thus to create optimum condi-
tions with due regard for settling prop-
erties of SS.
In the outlet zone a slight turbulence of
the flow is observed promoting the
flocculation of suspended solids, but soon
it fades and sedimentation of the suspen-
sion occurs. The layer of clarified effluent
is collected by the moving water-collect-
ing device. Having flown over the weir
edge (10) the clarified effluent flows along
the trough (9) to the central part and then
is withdrawn through the pipe (17).
Air valves (3) provide conjugation of
rotating and fixed units of the central part
of the distributing device and safely isolate
the clarified effluent from feeding and de-
tained influent.
Settling tanks with rotatory distributing
devices operate at a number of industrial
and municipal treatment facilities (Fig. 3).
The experience of their exploitation per-
mitted to make some general conclusions
and to compare their operation with that
of conventional radial settling tanks.
Graphical form of the relationship be-
tween the hydraulic particle size of
suspended solids and the hydraulic load is
given in Fig. 2. This relationship allows to
compare the efficiency of facilities operat-
ing under different conditions.
It follows from Fig. 2 that with.equal
efficiencies of clarification, that is with
equal values of hydraulic particle size of
captured suspended solids, settling tanks
with the rotatory distributing devices pro-
vide higher hydraulic load, 1.4 — 1.6 times
exceeding the load on radial settling tanks.
At technological calculation of settling
tanks one should also take into account
hydraulic particle size of settled suspended
solids. Besides, the degree of the hydraulic
perfection of the installation should be
also taken into consideration. It may be
characterized by the coefficient of the
volume usage.
The following relationship is valid for
radial settling tanks:
R =
/ 2
V3.6irl
6irKU,
-M ,
(I)
where Q — calculated water flow,
m3/hour
K — coefficient of the volume
usage
for radial settling tanks K =
.045
for settling tanks with
rotatory distributing device
k = 0.85
U0 — hydraulic particle size of
retained suspended solids,
mm/c
\n its turn U0 is characterized by settling
properties of SS and by the depth of the
continuous-flow part of the settling tank
— H:
t =
lOOOKUo
(2)
where t — is the detention time for the
influent in the cylinder no
less than 160 mm in diam-
eter, with the water layer h,
corresponding to the given
clarification efficiency. This
parameter is determined
experimentally for a con-
crete waste water;
n —-a factor characterizing
gravity coagulation of SS.
-------
S
n
H.o
I 2
Hydraulic load,
0
Fig- it 2.
Dependence of the CDM hydryulic particle size on hydraulic load
for different types of settling tanks:
I. radial settling tanks, D = 33 m;
2, radial settling tanks of 40 m diameter;
3. settling tanks with rotating distributing devices. D = 33 m;
4. settling tanks of the same type, D = 20 m.
The value of n is deter-
mined experimentally at
varying the water layer in
the cylinder — h.
For municipal wastes n =
0.25:
For heavy mineral particles
of SS n = 0.4 — 0.6.
H — calculated depth of the con-
tinuous-flow part of the
settling tank: with the
rotatory distributing devices
H = 1 — 1.5 m: for radial
settling tanks it is assumed
to be equal to 3 — 3.5 m.
For settling tanks with the rotatory dis-
tributing devices the period of rotation of
the sludge scraper and distributing devices
must be connected with the output:
T = 0.667
HD2
where T — the period of rotation of a
distributing device, hours;
H — the depth of the settling
zone, m;
D — diameter of the settling tank,
m;
Q — waste water flow, m/hour
It should be noted that normal hydraulic
conditions in these installations cannot be
observed if the difference between the
density of the influent and the density of
the clarified effluent is great, so the
application of settling tanks with the
rotatory distributing devices is limited by
the SS concentration of the influent up to
500mg/l.
At present modifications of the settling
tanks with the rotatory devices, per-
mitting to eliminate this kind of limita-
tions are being tested in the full-scale
operation.
Due to the effective reduction of the influ-
ent velocity the depth of the protective
zone in the settling tanks with the rotatory
distributing devices can be reduced to 0.7
m and thus the total depth of the installa-
tion can be reduced to 2-2.5 m, that is half
as many as the depth of conventional
radial settling tanks.
Conclusions
1. Radial settling tanks have significant
advantages over other types of settling
tanks and are widely used at large treat-
ment plants.
2. Hydraulic conditions of conventional
radial settling tanks do not ensure the
effective use of the volume of these
installations.
3. The application of the rotatory dis-
tributing devices allows to improve hy-
draulic conditions, to increase the load on
the installation about 1.5 times and sig-
nificantly reduce the depth of the tank.
4. Rotatory distributing devices allow to
control the hydraulic conditions of settling
tanks and their development and applica-
tion for the mechanical waste water treat-
ment is very promising.
29
-------
Figure 3.
Functioning settling tank with rotating distributing arm. The view
on the side of distributing grid.
30
-------
Current Status and
Directions of
Development of
Physico-mechanical
Effluent Treatment
in the Paper Industry
Dr. Isaiah Gellman,
Technical Director
National Council of the Paper
Industry for Air and Stream
Improvement, Inc.
Introduction
At the previous conference in this series I
reviewed the then current areas of
application of physico-chemical effluent
treatment in the paper industry. These
included (a) chemical assistance to
physical separation processes, (b) use of
chemical reactants to remove undesired
effluent constituents, and (c) physical
treatment to remove specific chemical
constituents as in steam- or air-stripping
of pulping liquor evaporation
condensates. The impetus for further
development work and technology
application in these areas was considered
and a number of areas were proposed for
possible cooperative study (1).
The objective of the current paper is to
review this same general area, omitting,
however, those developments and
technologies that are dependent on
chemical addition and readily identifiable
chemical reactions as the basis for their
effectiveness. In a sense then, physico-
mechanical treatment is viewed as a sub-
set of physico-chemical treatment,
generally stemming from the same
objectives but omitting the use of chem-
ical additives or processes. The review is
general in character, leaving both detailed
examination of the reports cited and
further cooperative activity in specific
areas of mutual interest as the means for
deeper exploration of specific findings.
Three areas of physico-mechanical
treatment are addressed below and are
identified as follows:
1 clarification of untreated or biotreated
mill effluents dy gravity separation or
media filtration,
2 in-process use of physico-mechanical
treatment for improved water economy
and organic load reduction,
3 dewatering and disposal of effluent
treatment sludges.
The objectives of such treatment are
numerous and include the following:
(a) effluent clarification in preparation for
discharge or secondary treatment, or
following secondary treatment,
(b) improved sludge dewatering and
further disposal, (c) enhanced
opportunities for wastewater reuse in
papermaking through removal of
suspended matter and objectionable
organic constituents,
(d) relief of overload conditions at
biotreatment facilities, (e) reduced
possibilities for odor emission at
biotreatment facilities, and (f) economic
and technical optimization of wastewater
management systems.
Clarification of Untreated or
Biotreated Effluents by Gravity
Separation or Media Filtration
Within this general category the objectives
differ somewhat, hence each area of
clarification is considered separately.
Clarification of Untreated Effluents
Clarification of mill effluents without
decollation chemical assistance is widely
practiced as a preparatory step for
biotreatment. The range of clarifier
volumetric loading rates and removal of
either total suspended or settleable solids
for a variety of mill categories is reviewed
in an earlier National Council report (2).
An analysis of seven production unit
categories indicated that average total
suspended solids removal ranged from 70
to 93 percent. Settleable solids removal
invariably exceeded 90 percent and
commonly reached 95 percent. Reductions
in BOD accompanying separation of the
settleable solids depended on the relative
contribution of those solids to the overall
load, exceeding 80 percent for a group of
non-integrated tissue mills and averaging
20 percent for integrated kraft packaging
or newsprint mills. The survey at that time
showed a median volumetric surface
loading rate of 600 gallons/sq. ft./day, the
level at which 95 percent settleable solids
removal was considered achievable.
Clarification of Biotreated Effluents
The two most commonly employed
biotreatment processes either (a) have
inherent clarification capability, as in the
relatively quiescent zones between
aerators in the longer retention aerated
stabilization basins, or (b) provide such
clarification through final clarifiers
following activated sludge aeration or
shorter retention aerated stabilization
basins.
/ Quiescent zone sedimentation in aerated
stabilization basin systems - A recent
National Council report (3) provides the
results of a detailed study of the mixing
characteristics of a number of such basins.
It defines the extent of those basin areas
where mixer-induced velocities are lower
than those needed to maintain biological
floe particles in suspension. Such data
provides a basis for determining when
separate sedimentation basins or special
exit area quiescent zones are needed for
clarification of aerated stabilization basin
effluents.
2 Secondary sedimentation for activated
sludge treatment systems - Operational
experience with such systems in the paper
industry was recently examined at a
National Council conference (4). While
the norms of secondary clarifier surface
loading design have been viewed as 700 to
1000 gal/ft2/day, experience has shown
that loadings exceeding 700 are incapable
of adequately dealing with bulky sludges.
There appears to be a growing view that
loadings in the 400 to 500 range are better
capable of accommodating process upsets.
This subject is receiving further attention
in attempting to deal with two separate
but related questions, namely (a) how to
maintain process integrity in the face of as
yet incompletely controlled process load
variability, and (b) how to achieve
excessively restrictive governmental
discharge permit requirements on effluent
residual total suspended solids.
-------
3 Secondary flotation fo aerated
stabilization basin effluent clarification -
One recent paper industry installation
relies on air pressurized flotation to clarify
the effluent from a six day aeration basin
at a waste paperboard mill. While
successful operation has been found
dependent on addition of substantial alum
coagulant to produce large size floe
particles, the process is referenced here
because it provided a more effective
clarification alternative than use of a
solids contact reactor. At temperatures
above 60° F a 1 MGD flow from a 100
TPD mill receiving 140 ppm alum is
clarified to an average residua! TSS level
of 24 ppm using a flotation tank unit
loading of 3000 gal/ft'/day at 80 psi
pressurization (5). Performance is,
however, substantially lower during colder
temperature periods. At temperatures
averaging 51sF for the winter period the
flotation-clarified effluent averages 65 ppm
TSS.
4 Media filtration of biologically treated
effluents - A number of recent inudstry
studies have explored the capability of
media filtration to clarify such effluents
both with and without coagulant addition.
One National Council study (6) conducted
with bleached kraft mill effluent showed
minimal clarfication capability unless
coagulants were employed. Final TSS
levels before and after coagulant addition
were 116 and 42 ppm respectively. At the
Great Northern unbleached kraft mill (7)
granular media filtration was found
capable of reducing TSS levels from 65
ppm (effluent from 14 days aeration) to 42
ppm without coagulants, and 36 ppm
using 30 ppm alum. A recent study at the
Buckeye Cellulose kraft dissolving pulp
mill (8) showed that a biotreated effluent
containing 32 ppm could be clarified to
only 24 ppm without coagulants and 18
ppm using cat ionic polyelectrolytes at a
filtration rate of 2 to 4 gal/ft2-min. In
summary, it appears that granular media
filtration is ineffective in clarifying aerated
stabilization basin system effluents
without coagulant addition, and only
modestly so following such addition.
In-process Use of Physico-
mechanical Treatment for
Water Economy and Organic
Load Reduction
A major focus of the paper industry's
current wastewater management
technology development efforts involves
in-process load reduction. Three such
areas of activity are based on physico-
mechanical approaches of widely
divergent character, namely separation of
fines and control of impurity buildup in
Whitewater system streams, air- and
stream-stripping of condensates, and
activated carbon adsorption of
miscellaneous organic constituents. All
are designed, however, to improve
wastewater management practices by
either volumetric or organic load
reduction.
White Water System Reuse Studies
The National Council has recently
completed three studies (9) (10) (11) whose
objective is to further the reuse of process
wastewater in non-integrated paper mills.
The program has proceeded in three
phases as follows:
a the collection of information related to
water reuse stressing operational limits to
reuse related to changes in water quality,
b development of a computerized data
retrieval system capability for entering
new, and assessing existing information
relating water quality and reuse potential,
c evaluation of the water renovation
capability of new and proposed process
equipment adapted to inclusion within the
white water system.
The program has examined water reuse in
the waste paperboard, fine papers and
tissue/towelling segments of the industry
with the following results:
/ Combination waste paperboard
operations - Studies at thirteen mills
indicated that closure to 3000 gallons/ton
could be readily achieved without
encountering serious water quality-related
problems. Beyond that point piping
system revisions and material changes
were indicated to reduce corrosion and
abrasion problems. Plugging at such
points as pump packing gland seals, -
cylinder showers, bearing cooling jackets
and machine felts was also encountered
and attributed to fibre buildup in the
recirculated water. The remedial approach
has involved use of savealls, microstrain-
ers, and fibre fractionators and
concentrators. In-line slotted strainers are
also gaining use.
I
Two recent reports document the utility of
a 'float-wash' fractionator for assisting in
achieving board mill water-economy.
According to Godin (12), water use and
TSS losses at a board mill were
substantially reduced by installation of a
float-wash fractionator to allow recycle of
board machine white-water to cylinder
and felt showers. Holmrich (13) described
such a proprietary device. Fractionatiqn
occurs by directing a stream against a
screen and separating the long fibres from
the fines with recycyle of the separated
streams for various applications.
2 Fine paper manufacture - Twelve mills
were examined in this study, all
discharging less than 10,000 gallons/ton
effluent. Plugging problems at wire and
head box showers, bearing cooling lines
and on the machine felts have been dealt
with by removal of long fibre material as
in the board mills cited above. Felt hair
removal is accomplished using horizontal
rotating center-fed cylindrical screens of
either nylon or stainless steel with a 120 to
140 mesh medium (14) (15).
In all but one of the mills visited fresh
water was used for felt washing. This use
remains as a large user of fresh water on
the modern paper machine. While in the
future it may be possible to reuse
Whitewater clarified by gravel filtration or
other means for felt cleaning, tests carried
out in mills forced to take this measure in
times of extreme water economy indicate
that premature deterioration of felts takes
place (16). In the one mill visited which
was using other than fresh water it was a
50 percent mixture of fresh water and dual
media filtered clarifier water and they
reported no loss of felt life using this
.12
-------
mixture. In the use of stronger needled
felts which are better able to stand up to
hightpressure water jets, the tendency
today is to use high-pressure, low volume
oscillating needled jet showers for
intermittent felt cleaning. This
significantly reduces water use.
3 Tissue and towelling manufacture -
Twelve tissue and towelling mills were
studied to determine current reuse
capability, problem areas and
opportunities for further water use
economy. Of these, eight used less than
10,000 gal/ton while two fell below 2000
gal/ton. The problems retarding further
use were similar to those reported
previously; namely (a) increased line,
nozzle, wire, and felt plugging, (b) scale
and corrosion, (c) product color, and (d)
slime accumulation.
The corrective measures of a physico-
mechanical nature have emphasized
improvement in in-plant flotation saveall
operation, use of in-line strainers, and
cascading of water use on vacuum pump
systems to permit further use of clarified
Whitewater. The mill currently achieving
the highest degree of closure, Kimberly-
Clark at Fullerton, California (17)
accomplishes this through use of bentonite
flocculation- or lime softening-assisted
diatomacious earth filtration, with the
latter approach preferred.
Condensate Stripping
Progress has been recorded recently in this
area for both kraft and sulfite liquor
condensates as follows:
/ Kraft condensate processing - An early
approach to condensate stripping for
BOD load reduction was reported by
Estridge (18) in which evaporator
< condensates, barometric condenser water
and turpentine decanter underflow at an
850 ton per day linerboard mill were
recycled to a cooling tower. Aeration of
these process streams in this manner
achieved a load reduction of 10,000 Ibs
BOD. per day due almost exclusively to air
stripping of-methanol and other volatiles.
An inherent drawback lay, however, in the
transfer of odorous volatiles such as
terpenes and organic sulfides to the
atmosphere.
This deficiency has been met by systems
employing either air or stream stripping
with the stripped material being burned at
a subsequent combustion unit. An exam-
ple of such a system is that in operation at
the Mead Corporation mill at Escanaba,
Michigan. Initially this 800 ton bleached
kraft mill installed a steam stripper
handling hot water accumulator overflow,
turpentine decanter overflow, evaporator
condensates and miscellaneous hot
odorous streams. The initial objective was
to reduce odor release at the biotreatment
system. This was accomplished using a
fractional distillation column 53 feet high
containing 24 trays, and steamed at a 2.5
percent rate, with non-condensibles and
collected foul air burned in the lime kiln
(19).
In 1973 a program was begun to determine
whether the mill BOD load could be
significantly reduced as well. Analysis
showed that the stripper bottoms
contained 15 to 18,000 Ibs BOD daily and
that 90 percent of this was accounted for
by the methanol present in the condens-
ates. Modifications to the system enable
the steam feed to be increased to 8 percent
so that the residual BOD was reduced to 4
to 5,000 Ibs daily for a net reduction of 11
to 13,000 Ibs daily. Total steam and power
requirements are reported as 8000 Ibs/hr
of 60 psi steam and 65 HP for pumping
condensates.
Kraft mill condensate stripping is now in
use at seven United States mills. A recent
report on this subject (20) reviews this
technology and reports capital costs for
systems ranging from air stripping
through partial- to complete steam
stripping at from 1000 to 5000 dollars per
ton daily capacity, with comparable
operating costs of 1.5 to 7 dollars per ton.
2 Sulfite Condensate Processing -
Currently underway at the Flambeau
Paper Company, Park Falls, Wisconsin
sulfite mill is a project directed toward
demonstrating the possibilities for
recovering methanol, furfural and acetic
acid as ethyl acetate from sulfite liquor
evaporation condensates. The process
being investigated involves steam
stripping and activated carbon adsorption
to achieve removal and separation of these
components. This investigation expands
on a project previously conducted at the
Institute of Paper Chemistry (21).
Another treatment approach pertinent to
sulfite liquor condensates involves the
upward adjustment of spent liquor pH
prior to evaporation so as to retain the
volatile acids in the liquor concentrate.
permitting their destruction in the liquor
furnace rather than allowing their entry
into the condensates. Previous studies
have shown that upward adjustment of
liquor pH from 2.5 to 4.0 could result in
condensate BOD load reduction from ISO
to 200 Ibs per ton to 50 Ibs per ton (22).
Removal of Organic Materials from Paper
Mill Effluents Using Activated Carbon
Explorations of activated carbon
treatment have been undertaken at a
number of non-integrated paper mills
involving in several instances combined
treatment with municipal sewage.
Included among these are (a) St. Regis,
Bucksport, Maine, (b) Fitchburg Papers,
Fitchburg, Massachusetts, (c) Mohawk
Papers, Waterford, New York and (d)
Neenah, Menasha, Wisconsin. Details of
several such projects which follow indicate
that chemical coagulation will generally
represent an integral step in activated
carbon treatment, if only to protect the
carbon columns from plugging problems,
but also to minimize the residual organic
loading.
/ Fitchburg, Mass. - Pilot studies for a full
scale project involving 14 MOD from two
paper mills and 1.25 MOD municipal
effluent were conducted using a process
consisting of chemical coagulation with
alum, sedimentation and activated carbon
column treatment. Process results
indicated an anticipated 50 percent BOD
reduction through coagulation, followed
by 82 and 86 percent reduction in COD
and BOD passing through four carbon
columns at 3.4 gal/sq. ft./min. The
activated carbon process design criteria
selected were 7.2 Ibs carbon/Ib BOD at
33
-------
exhaustion, 30 minutes contact time, and
reactivation in a multiple hearth furnace.
Costs were estimated at $2.1 million for
construction of the carbon system in 1970,
and $140,000 for annual operation, of
which 70 percent is carbon makeup. The
facility has been constructed, but
operating results are not yet available (23).
2 Neenah - Menasha, Wisconsin - A pilot
plant study was performed using
municipal sewage containing 80 percent
paper mill effluent. The process studied
involved chemical coagulation with ferric
chloride, sedimentation, high rate
filtration at 12 gal/sq. ft/min. to remove
fibrous material followed by a second
filtration stage at 9 gal/sq. ft/min through
PVC media, and upflow expanded bed
filtration through 14-40 mesh carbon at 3
to 5 gal/sq. ft./min. Results obtained by
the 1PC study group indicated removals of
BOD, COD and suspended solids prior to
carbon treatment of 76, 83 and 98 percent
respectively. Following carbon treatment.
overall reductions were increased to 93.95
and 99.5 percent respectively. Project
costs were estimated in 1973 at 17 cents
per 1000 gal capital and 35 cents per 1000
gal operating for a total of 52 cents per
1000 gal for a 10 MGD plant (24).
Sludge Dewatering and
Disposal
Sludge dewatering constitutes one area of
extensive physico-mechanical treatment
process involvement. This takes the form
of the dewatering process itself, the impact
of hydrous sludge presence on sludge
dewatering, use of physical dewatering
aids designed to dilute the presence of
hydrous sludges, the potential use of
thermal treatment to enhance dewatering,
and the possibilities for physical
separation of reusable materials from
sludges. These subjects are considered
below.
Current Sludge Dewatering Practices
A recent National Council study (25)
examined the extent of, and principal
methods employed for mechanically
dewatering treatment plant sludges. The
study encompassed 15 percent of the mills
operating in the United States and
producing 35 percent of the industry's
tonnage. Findings of the study can be
summarized as follows:
1 The mechanical dewatering alternatives
currently used in the pulp and paper
industry (in the order of decreasing
number of installations) include vacuum
filters, centrifuges, V-presses, plate and
frame presses, and screw presses. In the
dewatering of sludges with ash contents of
less than 40 percent vacuum filters
outnumber centrifuges by a two-to-one
ratio. On the other hand, at mills
dewatering sludges with ash contents in
excess of 40 percent centrifuges are
favored by a ratio of two to one.
V-presses are utilized exclusively as
second-stage dewatering devices while
screw presses are being used for both first
or second-stage dewatering. Second-stage
dewatering is practiced by three-quarters
of those mills incinerating sludge.
Combined sludges containing up to 30
percent biological solids are being
dewatered on vacuum filters, centrifuges
and V-presses but at lower capacities, cake
consistencies, solids recoveries, and higher
conditioning costs than commonly
encountered in primary sludge dewater-
ing. Mills requiring cake consistencies of
35 percent or greater from combined
sludges or 20 percent or greater from
secondary sludges are frequently consid-
ering plate and frame pressure filtration.
2 Centrifuge capacities associated with
pulp and paper industry primary sludge
dewatering applications are hydraulically
limited by acceptable solids recovery.
3 Scroll abrasion is the most commonly
cited constraint to the continued
application or extension of centrifugation
for pulp and paper industry sludge
dewatering. The only production
categories where scroll lives between
stellite resurfacing were reported to exceed
nine months continuous operation were
drinking mills manufacturing tissue and
tissue mills generating sludges without
pulp mill or flyash solids. In the absence of
effective grit removal, scroll life for the
balance has been limited to an average of 6
months.
4 Current advances in centrifuge
technology have resulted in development
of replacable conveyor tip wear plates
which have exhibited the potential for . .
superior wear characteristics, as well as
reduced time requirements for resurfacing.
The newer plates are in service at several
mills. Their long term performance
warrants attention of the industry to
reevaluate the acceptability of
centrifugation for sludge dewatering
where circumstances and sludge
characteristics suggest it to be a viable
alternative.
5 Cake consistencies attainable by
centrifugation or vacuum filitration of
primary or combined sludges appear
related to sludge ash content, cake
consistency increasing with ash content
among other variables.
6 Assuming the presence of adequate
fiber, vacuum filter loading rates for
primary sludges typically range from 5 to
10 Ibs/sq ft/hr. Lower rates are common
among nonintegrated mills and some
groundwood operations. In those
instances where chemical conditioning
was encountered, it was practiced for
improvement in solids loading rates,
though increased solids recovery was an
associated benefit.
7 Because of otherwise unacceptable cake
discharge characteristics, the applicability
of vacuum filters appears to be limited to
those situations where 100-mesh fiber
exceeds 10 to 20 percent of the sludge
mass. A higher threshold is likely for
combined sludge dewatering.
Deteriorating filter performance can be
anticipated as greater fiber recovery
results in sludge fiber contents
approaching that threshold.
8 Vacuum filter primary sludge solids
recovery is influenced by sludge ash
content and any chemical conditioning
that might be practiced. In the absence of
conditioning, increases in ash content are
often reflected in less solids recovery.
9 Fabric media vacuum filters are often
applied in situations where the fiber
content is low, the ash content is high, or
-------
the solids are otherwise difficult or
impossible to dewater on a coil filter.
10 If adequate quantities of primary solids
are available for admixing, secondary
solids are usually dewatered as combined
sludges.
1! Combined sludges with as high as 30
percent biological solids are being
dewatered on vacuum filters, centrifuges
and V-presses at conditioning require-
ments of 2 to 7 pounds polymer/ton.
Machine capacities can be expected to be
lower in combined sludge dewatering than
primary sludge dewatering.
12 High cake consistencies and the
necessity for handling difficult-to-dewater
sludges have prompted recent interest in
plate and frame pressure filtration.
Industry pilot investigations suggest that
combined sludge cake consistencies of 30
to 45 percent and secondary sludge cakes
of 30 to 40 percent may be anticipated
with a variety of conditioning techniques.
The relative merits of high and low
pressure operation and precoat utilization,
and comparison of anticipated and actual
performance remain unresolved.
Performance in full scale installation will
warrant continued industry attention.
13 Moving belt presses and recent
generation centrifuges offer new
alaternatives for sludge dewatering,
especially in hydrous sludge applications.
14 The pulp and paper industry disposes
of most of its waste treatment solids on the
land, incineration being a common
alternative with sludges of less than 10
percent ash content. Definition of
potential for wider application of sludge
incineration requires further investigation.
Most recently the P. H. Glatfelter mill at
Spring Grove, Pa. (26) reported successful
startup and operation of a twin belt press
for dewatering primary sludge. Clarifier
underflow is first thickened to 4-6 percent
solids in a 60 ft (18m) gravity thickener
before pressing to 35-40 percent solids at a
rate of 50-60 tpd on the 80 in (200 cm)
width press which is operated
intermittently depending on the amount of
sludge needing to be dewatered. Uses and
applications for the dewatered sludge,
which is presently disposed of on land, are
being investigated.
Special Physico-Mechanical Approaches
to the Dewatering of Waste Activated
Sludges
As noted above, some success has been
achieved in dewatering mixtures of
primary and secondary sludge without
chemical coagulant addition where the
long fibred sludge content is adequate to
overcome the difficulties introduced by
hydrous activated sludge. Where the long
fibre content in inadequate, several
remedial approaches have been employed
or considered. Those that can be
categorized as physico-mechanical
treatment include (a) thermal conditioning
of sludges, (b) specialized dewatering of
the segregated activated sludge or other
hydrous sludges, and (c) addition of waste
bark as a filtration aid for conventional
filtration. Developments in these areas are
reviewed below.
1 Thermal Conditionong of Hydrous
Sludges - A recent National Council study
(27) has explored this subject in depth.
Study findings can be summaried as
follows:
a Hydrous groundwood fines and waste
activated sludges associated with the
treatment of wastewaters from pulp and
paper manufacturing, inclusive of sludges
resulting from chemical treatment of
biologically treated effluents, were all
demonstrated to be highly responsive to
thermal conditioning. That was not the
case for alum based water treatment
sludge. For waste activated sludge, acid-
assisted oxidative conditioning offered
only a very slight, if any, advantage
beyond that achieved with only a non-
oxygen limiting envornment.
b Improvement in sludge filterability is
related to the solubilization of sludge
volatile constituents up to some optimum
solubilization level between 40 and 50
percent. Beyond that degree of
solubilization, a capability limited to
oxidative conditioning, solids
dewaterability exhibited a classical
reversion.
c Oxidative conditioning in a non-oxygen
limiting environment poses no distinct
conditioning advantage over rion-
oxidative processes. In fact, in the case of
alum-coagulated biological solids, filter
leaf tests confirmed results with
conventional filtration parameters
showing that oxidative conditioning was
less effective than conditioning in the
absence of oxygen. However, the presence
of oxygen accelerates the rate and extent
of volatile solids solubilization.
d Regardless of the mode of conditioning
or acid addition, solubilization of
activated sludge volatile constituents
• resulted in an associated supernatant
COD increase equivalent to
approximately 50 percent of the mass of
volatile solids hydrolyzed. The
corresponding ratio for the alum-
coagulated solids was less, suggesting the <
possible importance of aeration system
sludge age to the character of the
supernatant. The ratio of filtrate BOD to
COD for groundwood fines and waste
activated sludge including when acidified
was 0.5 to 0.6 in comparison to 0.3 for the
solids separated with chemical
coagulation of biologically treated
effluents.
e Heat treatment represents a viable
means of conditioning the most difficult of
sludges. Though not within the scope of
. this study, technical personnel at
. installations contemplating its use should
be aware of experience cited within the
literature outlining its implications for
significantly increased raw waste and color
load and (b) such reported operational
problems as corrosion, scale, plugging and **
equipment maintenance.
2 Specialized dewatering ofwaxte
activated sludge - The National Council
recently completed a pilot-scale evaluation
of seven alternative dewatering
approaches for waste activated sludge
(28). These included (a) pressure filtration,
(b) precoat vacuum filtration, (c) Tail-
Andritz filter belt press, (d) Permutit dual
cell gravity-multiple press roll filter.
-------
(e) capillary suction sludge dewatering,
(f) Sharpies horizontal bowl decanter
centrifugation and (g) ultra-filtration. The
test sludge was that generated in an
integrated bleached kraft pulp and paper
mill activated sludge treatment system.
Results of the study were summarized as
follows:
a All of the units investigated were capable
of dewatering waste activated sludge.
Chemical conditioning, however, was
mandatory in all cases except for
centrifugation and ultrafiltration.
4 b The pressure filter and precoat vacuum
filter generated the driest cakes (25 to 40
percent solids) making them likely
alternatives where incineration, or dry
cake for landfilling or hauling are involved
in final disposal. The possible effects of
handling and disposing of the nonsludge
fraction of these cakes require
consideration.
c Pressure filter performance on this
sludge was not enhanced by higher
operating pressures (13 atmospheres vs. 7
atmospheres). Additional industry
experience suggests lower operating
pressures to be adequate for other types of
sludges as well.
d The filter belt presses generated cake
consistencies approaching 20 percent
solids. Relatively low power and
maintenance costs are likely to be
associated with these units. The filter belt
presses generated the driest cakes of the
units not utilizing inorganic conditioning,
suggesting them as a likely alternative
where inorganic contamination of the
sludge is not desired.
e The gravity filter and centrifuge gen-
erated cakes at 8 to 10 percent maximum
consistency, indicating probably
applications in prethickening for pressure
filtration, digestion or heat treatment, or
where land disposal of semi-fluid sludge is
envisioned. The gravity filter offers very
low power costs while the centrifuge offers
no sludge conditioning costs.
f The ultrafilter was capable of thickening
sludge to 7 percent consistency. However,
the high power costs associated with
overcoming the pressure drop through the
system suggest that a different membrane
configuration would be required to make
the process feasible.
g There were several units that con-
sistently operated at solids recoveries
of 99 percent or better suggesting their
applicability where the liquid fraction
solids levels had to be low. These units
were the pressure filter, the precoat
vacuum filter, the gravity filter and the
ultrafilter.
h Artificially increasing the specific
resistance of the unconditioned sludge
from a range of 50 to 200 x 107 sec2/gm to
a range of 150 to 400 x !07 sec2/gm had a
substantial detrimental impact on the
performance of those units which handled
the sludge. Caution is, therefore, indicated
in applying the data in this study directly
to other sludges particularly where
knowledge of the specific resistance of the
sludge in question is lacking.
i Most of the units indicated a sensitivity
to feed concentration, and achieved or
indicated significantly superior
performance at sludge feed consistencies
of 2 percent solids as compared to 1
percent solids.
j Additional pulp and paper industry
experience has shown both pressure
filtration and filter belt pressing to be
applicable to primary, secondary, and
combined sludge dewatering.
Performance levels are determined by the-
equipment configuration, the nature of the
solids, the type and amoung of sludge
conditioning, and the feed consistency.
k Some mills have encountered substantial
startup difficulties with pressure filters
including plate breakage, media tearing,
and poor sludge distribution in the
chambers. These problems have generally
been overcome, some, however only with
intense effort.
3 Use of waste bark as a filtration additive
- Waste bark is normally available at all
integrated pulp and papermills except
where the wood supply is derived'from
sawmill chips. Encouraged by success "•
reported by Bishop (29) at a bleached
kraft newsprint-mill; the National
Council's continuing research program
pertinent to the dewatering and disposal
of hydrous sludges was extended to bark
conditioning of biological sludges.
Variables investigated included bark type,
particle size, size distribution, and
addition level. Factors of significance to
filtration established on the basis of
filtration theory and empirical studies
cited in the literature were reviewed and
observed to apply to bark-conditioned
slurries (30).
An optimum bark particle size in the
range of 20 to 40 mesh was suggested at~
minimum addition levels of 3-parts bark .
to l-part secondary sludge. Solids
retentions greater than 85 percent were.-,
attained only where particle sizes of less
than 40 mesh constituted at least 10
percent of the bark supplement.
Associated sludge solids loading rates
were less than.2 Ibs/sf/hr. Nevertheless, -
application of bark for the conditioning of
secondary sludges will require further
study to evaluate bark size reduction
processes and assess the conditioning ,
properties of resulting size distributions.
Based on the results in this investigation,
the following conclusions were reached:
1 Successful use of bark in the size range
greater than 8 mesh for conditioning of
combined primary-secondary sludges
cannot be achieved without the presence
of fibre in the sludge.
2 Optimum conditioning of biological
sludges requires size range selection of* '
bark in the 20 to 40 mesh range. Resulting
net solids loading rates increase linearly
with bark addition.
3 Solids retention on the filter medium
exceeding 85 percent is attained only
where the bark size distribution is'such
that at least 10 percent passes a 40-mesh
sieve. Associated net solids loading rates
range from 1 to 2 Ibs/ sf/ hr, with the major
values occurring with more uniform
particle size distributions.
-------
4 Solids retention increases as the
uniformity of particle sizes constituting
the bark supplement is decreased.
5 Cake handling properties require a
minimum addition of three parts bark to
one part secondary sludge.
6 Within the scope of this study,
hardwood bark is unsuitable for the
conditioning of biological sludges, and
exerts a detrimental effect on the superior
softwood bark where the two are
combined.
Separation of High Ash Component from
High Ash Fine Papermill Sludges
The manufacture of filled and coated
grades of paper frequently gives rise to '
sludges rich in ash components such as
clay and TiO2. Their high ash content (a)
can have a negative bearing on their
subsequent disposal, particularly by
incineration, and (b) raises the possibility
of their subsequent processing for
recovery and reuse of the ash components.
and to facilitate disposal of the low-ash
component. A study of this possibility was
recently completed by the National
Council.
The summary report (31) presents the
results of (a) semi-pilot plant scale study
of several means for separating the ash
component from the fibrous organic
portion of the papermill primary sludge,
and (b) pilot plant paper machine trials of
reuse of this material as either filler or
coating pigment for filled and coated
paper grades. The recovery methods
studied were screen-separation and wet
oxidation, and were selected for more
detailed examination in a previous
laboratory investigation of a larger
number of alternatives. The possibility of
brightening such screen-recovered
material using several bleaching sequences
was also explored.
Results of the study offer considerable
encouragement regarding the use of screen
separation to produce reusable filler
material. While some reduction in sheet
brightness was noted, the possibility of
remedying this drawback by oxidative
bleaching appears encouraging. Wet
oxidation was also found capable, of
producing reusable filler material, and
with lesser adverse effects on product
brightness. Preliminary evaluation
indicated, however, that its potential for
producing reusable coating material is not
too promising.
The findings open up a number of
possibilities for mill-specific consideration
.of such processes for recovery and
processing of the high ash component and
its subsequent reuse. They also identify a
number of areas for further study that
should expand the technological/ process
economics base for this sludge
disposal/materials recovery approach and
identify the limits for its practical,
continuous application at individual mills.
A recent application of these laboratory
and mill-scale trial results is reported by
Flynn (32) for the recovery of filler clay
and pigment from paper mill sludge by
wet air oxidation. The mill employs two-
stage activated sludge for effluent
treatment and all sludge was lagooned.
Because conventional high temperature
incineration of the sludge increased
abrasive characteristics of the filler, wet air
oxidation was chosen in order to recover
the filler in a reusable form. The process
was described in which the sludge at 8
percent solids is reacted at 2500 psig and
600°F(17.24 MN/tn2. 315°C)for90
percent destruction of organic material.
Summary - Developmental
Trends and Possible Areas for
Cooperative Study
As noted earlier in our paper on physico-
chemical treatment, the grouping of
treatment methods under the heading of
'physico-mechanical'covers a multiplicity
of approaches, objectives, technologies
and practical possibilities. In some
respects the approaches represent a more
elementary level of treatment, to the
degree th'at they are unaided by chemical
addition. In other respects, however, they
can be viewed as representing an attempt
toward economic optimization of
processes so as to avoid the use of
chemicals and substitute additional power
use or equipment cost to achieve that goal.
The developmental trends evident can be
summarized as follows:
1 It is likely that total mill effluent
clarification at the primary level will
continue to be characterized by physico-
mechanical since the necessity of
unusually high clarity is not critical at this
point. One notable exception is evident
with regard to lime decolorization which
in a number of instances will be combined
with primary sedimentation.
2 At the secondary clarification level,
more frequent use of coagulant aids is
anticipated both to meet agreed on water
quality protection objectives as well as the
frequently arbitrarily restrictive effluent
limitations stemming from governmental
effluent limitations guidelines. Similarly,
flotation and mechanical screening
options will receive greater attention in
efforts to exceed the present capability of
gravity sedimentation clarification.
3 Concerted efforts to reduce mill effluent
flow through greater internal water reuse
will lead to more extensive screening of
white water streams to aid in closure of
white water systems.
4 Stripping of condensates from chemical
pulp mill liquor evaporation and heat
recovery systems will become more
extensive as part of an optimization
approach to unifying internal and external
control measures.
5 Modest beginnings have been made in
the area of activated carbon renovation of
paper mill white water for reuse. It
remains to be seen how extensive this
practice will become in contrast to
screening, filtration and gravity separation
of fines from white water.
6 Sludge dewatering and clarficalion will
continue to represent a major area of
study and application of physico-
mechanical treatment approaches.
Some areas that suggest themselves for
-------
possible cooperative study include the
following:
1 Determination of water quality needs for
both pulp and paper manufacture as a
guide to development of specific effluent
treatment objectives, particularly for in-
process effluent streams which are
candidates for further reuse.
2 Examination of the economic and
treatment optimization aspects of varying
degrees of condensate stripping, stressing
the most economic level of use and source
of stripping steam and the incremental
savings achieved in use of existing
biotreatment systems, or their possible
future avoidance or minimization for a
new generation of more highly closed pulp
mills.
3 Evaluation of new sludge dewatering
and conditioning approaches for sludges
rich in hydrous components or from
which long-fibred material has been
deliberately separated. Coupled with this
work should be a common examination of
the possibilities for segregation of high ash
material for industrial reuse.
Literature References
Gellman, I., Current Status and
Directions of Development of Physical-
Chemical Affluent Treatment in the Paper
Industry. Proceedings US-USSR
Conference on Physical-Chemical
Treatment of Wastewater from Municipal
and Industrial Sources, November 12-14,
1975, Cincinnati. Ohio
Follett, R. and H.W. Gehtn, Manual of
Practice for Sludge Handling in the Pulp
ami Paper Industry, NCASI Stream
Improvement Technical Bulletin No. 190
{June 1966)
McKeown. J. and D. Buckley, A Study of
Mixing Characteristics of Aerated
Stabilization Basins, NCASI Stream
Improvement Technical Bulletin No. 245
(June 1971)
NCASI Technical Conference, Detection,
Prevention and Solution of Activated
Sludge Treatment System Operational
Problems in the Paper Industry -
Summary Report (January 1977)
Greenhouse, H., Flotation Process
Performance for the Separation of
Biological Solids Following ASB
Treatment, in NCASI Special Report,
Proceedings of 1976 NCASI Central-Lake
States Regional Meeting, No. 76-11,
(December 1976)
Whittemore, R. and J. J. McKeown, Pilot
Plant Studies of Turbidity and Residual
Cell Material Removal from Mill
Effluents by Granular Media Filtration,
NCASI Stream Improvement Technical
Bulletin No. 266 (April 1973)
Smith, O.D., Stein, R.M. and Adams,
C.E., Analysis of Alternatives for
Removal of Suspended Solids in Pulp and
Papermill Effluents, Tappi 58(10) 73
(1975)
E. C. Jordan Co. Report to U.S.
Environmental Protection Agency, Study
of Filterabilily of Aerated Lagoon
Effluent from a Pulp Mill in Foley,
Florida (1976)
Marshall, D. and A. Springer, The
Relation Between Process Water Quality
Characteristics and Its Reuse Potential in
Combination Board Mills, NCASI
Stream Improvement Technical Bulletin .
No. 282 (October 1975)
Marshall, D. and A. Springer, The
Relation Between Process Water Quality
Characteristics and Its Reuse Potential in
Fine Paper Mills, NCASI Stream
Improvement Technical Bulletin No. 287
(August 1976)
Marshall, D. and A. Springer, The
Relation Between Process Water Quality
Characteristics and Its Reuse Potential in
the Non-Integrated Manufacture of Tissue
and Toweling, Stream Improvement
Technical Bulletin No. 289 (November
1976)
Godin, K., Float Wash Fractionator
Saves Fibre and Water at Grand Mere,
Pulp Paper Mag. Can. 76 (6) 81 (1975)
Holmrich, M., Float Wash - A Closing-
Up System for the Pulp and Paper
Industry. Paper 182 (8) 500 (1974); Abs.
Bui. Inst. Pap, Chem. 45(9)9483'(I975)
Rakosh, L., A 6000 Gallon/Ton Fine
Paper Machine Water System, Pulp and
Paper Mag. Can. 75 (3) T69 (March 1974)
Water and Slock Conservation on
Fourdrinier Machines, Summary of
Annual Meeting Panel Presentation,
Tappi 54 (10) (Oct. 1971)
Wilkinson, J.J.,.A Practical Approach to
Water Conservation in a Paper Mill, Pulp
and Paper International 59-62 (May 1973)
Le Compte, A.R., Water Reclamation by
Excess Lime Treatment, Tappi 49 No. 12
121A-124A (December 1966)
Estridge, R.B. etal, Treatment of Selected
Kraft Mill Wastes in a Cooling Tower,
Proc. 7th TAPPI Water and Air
Conference (June 1970)
Ayers, K.C. and G.L. Butryn, Mead
Experience in Steam Stripping Kraft Mill
Condensates, Tappi 58 (10) 78 (October
1975)
Hough, G. and Sallee, T.W., Treatment of
Contaminated Condesates, Tappi 60 (2)
83 (February 1977)
Baierl, K.W. etal.. Treatment ofSulfite
Evaporator Condensates for Recovery of
Volatile Components, EPA-660/2-73-030
(December 1973)
Blosser, R.O. and Gellman, I.,
Characterization ofSulfite Pulping
Effluents and Available Alternative
Methods, Tappi 56 (9) 46 (September
1973)
Camp, Dresser and McKee, Report on
Comprehensive Plan for Domestic and
Industrial Wastewater Disposal
Supplement "C" (August 1970)
McCuaig, W.B. etal. Physical-Chemical
Treatment of Combined Municipal, Pulp
and Paper Wastes, Proc. TAPPI
38
-------
Environmental Conference (April 1974)
Miner, R.A. and D.W. Marshall, Sludge
Dewatering Practice in the Pulp and
Paper, Industry, NCASl Stream
Improvement Technical Bulletin No. 286
(June 1976)
GlatfeUer, P.H.G., New Sludge
Dewatering Press Makes Its Debut, Putp
Paper 50 (14) 146 (December 1976)
Marshall, D. and Fiery, F., A Laboratory
Investigation of Heat Treatment for Pulp
and Paper Mill Sludge Conditioning,
NCASl Stream Improvement Technical
Bulletin (in press)
Miner, R. A. and Marshall D., A Pilot
Plant Study of Mechanical Dewatering
Devices Operated on Waste Activated
Sludge. NCASl Stream Improvement
Technical Bulletin No. 288 (November
1976)
Bishop, F.W. and Drew, A.E., Disposal of
Hydrous Sludges from a Paper Mill,
Proceedings of TAPP1 8th Water and Air
Conference, Boston, Mass. (1971)
Marshall, D., Effect of Bark Addition on
the Dewatering Properties of Biological
Solids, Stream Improvement Technical
Bulletin No. 261 (December 1972)
Marshall, D. and A. Springer. An
Investigation of the Separability and
Reuse Potential of the Ash Component of
High Ash Content Paper Mill Sludges,
NCASl Stream Improvement Technical
Bulletin No. 285 (June 1976)
Flynn, B.L., Jr., Swiss Mill is Trying Wet
Air Oxidation to Get Rid of Sludge and
Recover Filler, Paper Trade Jour. 160 (9)
23 (May t, 1976)
Employment of
Microstrainers in the
Waste Water
Treatment Practice
Skirdov I.V.
Sidorova LA.
Maximenko Yu. L.
At present microstrainers are mainly used
in water supply for plankton removal
from natural water. They alleviate the
sand filters operation and allow to
increase their output.
However, microstrainers can also be used
for waste water treatment. They can serve
as possible substitutes for primary
clarificrs in the technological scheme of
treatment. The experience has shown that
microstrainers ensure final treatment of
effluents after secondary settling tanks.
Microstraining is very promising because
microstrainers are compact ana
economical, simple in operation and
permit the process to be automated.
In the Soviet Union a number of scientific
researches were carried out evaluating the
efficiency of the micros!raining process in
application to concrete conditions.
The gained experience allows them to
make some general conclusions
concerning the methods of experiment
and the ability of expansion of
microstrainer's field of application.
The characteristic feature of
microstraining as the method of waste
water treatment from suspended solids
consists in the fact, that a very thin layer
formed by SS particles retained on the
screen with mesh sizes of 40-90 microns.
In the Soviet Union as well as abroad
drum microstrainers are preferably used
(Fig. 1)
14 15. /A 1?, /A
.:„"-.-:. ;r~--- l^.._-;; .-!
Figure 1.
Scheme of a screen drum
1. Drum
2. Transverse frames
3. Stringers
4. Rigid ribs
5. Drain pipes
6. Influent channel
7. Front frame
8. Inlet pipe
9. Cast-in fitting pipe
10. Pin wheel
II. Wash water draw-off pipe
filter- '2- Front bearing
uuei. ,3 Electric motor
14. Reducer
15. Gear
16. Bunker
17. Wash water feed pipe
18. Sprinkler
19. Bactericidal lamps
20. Weir
21. Filtrate channel
22. Back frame
23. Back bearing.
-------
Microstraining is carried on at continuous
rotation and regeneration of the screen.
Microstraining efficiency depends on a
number of factors and. in the first place,
on the formation of a thin sludge bed on
the inner surface of the microscreen.
Formed filtering sludge bed is thickening
negligibly and can be easily washed off by
washing water. So the control of washing
speed is of great importance for creation
of optimum conditions under which the
sludge layer is formed and for ensuring
high efficiency of waste water treatment.
Washing water flow should be regulated in
such a manner as to preserve a required
sludge layer on the screen after washing
and to maintain the head loss within the
certain limits.
Optimum parameters of microstraining
defy theoretical calculation and are
usually defined experimentally.
Great variety of types and compositions of
industrial wastes demand special tests to
be carried out on wastes of each branch of
industry and sometimes on wastes of each
enterprise. The need for the large number
of tests to be carried out and labour-
consuming character of these tests caused
the development of a simple method of
modelling for tentative evaluation of main
relationships, permitting to give an
approximate characteristics of
microstraining under concrete conditions.
A filtering column is used lor this
purpose. In the lower part of the column
there is fixed filtering screen. Despite the
fact that filtering is carried on in static
conditions, this method of modelling
allows to objectively evaluate the ability of
microstraining employment for given
conditions. The influence of such factors
as the type and the mesh size of the micro-
screen, the losses of head, filtering
velocities on the efficiency of treatment is
determined at the first approximation.
After preliminary investigations in static
conditions, the main operation factors
may be corrected at pilot plants.
It should be noted, however, that existing
models of drum microfilters ensure only
tentative study of microfiltering, as the
investigation of back washing of the
micro-screen which in many respects
determines the structure of filtering sludge
bed, is impossible. Non-observation of
adequate speeds of the screen rotation
under the nozzle of the washing device in
the model and in the full-scale installation
is among the serious shortcomings of the
known technique of modelling. When
modelling the working parameters of the
model one should proceed from the
equality of the back wash duration at
equal periods of rotation.
In the laboratory-scale model of the
microstrainer, developed at VODGEO
Institute, full-scale velocities of the screen
rotation were ensured by a special
mechanism, speeding up the screen
movement when it moved under washing
jet. (Fig. 2)
The results of investigations carried out on
the laboratory model were verified at pilot
plant.
A pilot microstrainer is a cylindrical
screen drum of 300 mm in diameter and
200 mm length, located in metal reservoir.
It rotates with the help of electric motor.
The number of revolutions of the drum is
changing step by step within the range
from 5 to 20 rev/min.
The drum was equipped by the brass
screens, two of which are supporting. The
working screen is located between the
supporting ones. The mesh size of the
working screen ranged from 40 to 90
microns. The mesh size of supporting
screens percepting the loads formed by the
pressure overfall during filtering and
washing was equal to 1.25 mm. The
filtering sludge bed and required overfall
of the pressure on the screen were
maintained by adjacement of the wash
water flow and speed.
Washing of the screen was carried on with
the help of the travelling nozzle situated
under the drum and moving along the
generating line of the drum. The
employment of such a way of screen
washing brought the washing conditions
to the fuli-scale operation conditions.
The following parameters of the process
were registered during investigation:
duration of the filtering cycle, the rate of
filtering, waste water treatment efficiency.
On the basis of the investigations there
were determined relationships between the
efficiency of the microstraining and the
filtering rate, initial concentration of
suspended solids in the influent and the
number of the drum revolutions.
In order to choose the optimum
parameters of the process, obtained
relationships are developed quickly to be
presented in the form of summary graph,
where the treatment efficiency is
represented as a function of the filtering
rate and drum rotation rate (Fig 3,4'
Such parameters as the head on the
screen, the mesh size, and wash-water flow
are changed within a narrow range and
were equal respectively:
• head losses 150-200 mm of water
column;
• wash-water flow 6% (for raw waste) and
3% (for the effluent after secondary
treatment).
On the basis of the data given in the
summary graphs the following parameters
of microstraining may be recommended.
For clarification of raw waste (instead of
primary clarifiers) at initial concentration
of suspended solids 150-200 mg/1.
• clarification efficiency - 45%
• filtering rate - 30 m/ hour
• filtering cycle - 9 sec.
• mesh size of the screen - 90 microns
For final waste water treatment (after
secondary settling tanks) at initial
concentration of suspended solids 25-60
mg/1.
• clarification efficiency - 65%
• filtering rate - 45 m/ hour
• filtering cycle - 6-8 sec.
• mesh size of screen - 40 microns.
40
-------
Section
2 ,
mesh
lever of
the reducer
1 2 3
Scheme of drive
mechanism's kinematics
5 5 7 8 9 10 II 12
-4-
0.5
rotation period
1,0
Figure 2.
Scheme of a microstrainer's model
1, A cell with a micromesh
2. Feeding of influent
3. Wash-water collection trough
4. Wash water drain
5. Washing jet
6. Wash-water receiving funnel
7. Micromesh drive reducer.
41
-------
Figure 3,
Dependence of the process efficiency on the
filtering rate, initial concentration of
suspended solids, revolutions of the drum per
minute (after primary treatment).
rt "•-
—
•H
O
•H
I
n = 20 rev/nir.
C0 =. 65 ms/1
30 40
Kigure 4.
Dependence of the process efficiency on
filtering rate, initial concentration of
suspended solids and revolutions of the drum
per minute (after secondary treatment).
n s 20 rev/min
2.O
-fo
C0 = 60 ms/1
50
42
-------
In case of preliminary treatment BODs
removal up to 25% was observed. After
filtering of biologically treated water
BOD5 removal was up to 40%.
Parameters taken from summary graphs
can be used for approximate design of
microstrainers operating under similar
conditions. More general dependences
may be obtained on the basis of
theoretical premises and generalization of
experimental data known.
The microstraining process can be
described with the help of semi-empirical
dependences based on the following
assumptions:
1 a sublayer of certain mean thickness
"Mo" is formed on the rotation drum
microstrainer.
Thickness of this sublayer depends on the
initial suspension flow, the time of filtering
and the difference between the initial and
resulting concentration of suspended
solids:
_ (C.-CJTW
where : Mo = thickness of the sublayer,
mm;
Co — initial concentration of
suspended solids, mg/ 1 ;
C' — resulting concentration of
suspended solids, mg/ 1 :
T — filtering cycle, sec.;
W — filtering rate, cm/ sec;
y — specific weight of parti-
cles, forming the sludge
bed. g/ cm3, y - 0.8 — 1
g/cm'.
2 The sublayer can be described as the
additional filtering sludge bed with orifice
diameter "d<>", which is equal to the
minimum diameter of detained particles.
Diameter of the particles forming the
sublayer can be determined by the Stokes
equation:
particle diameter, cm;
water viscosity, 17.= 0.01
g/ cm/sec;
specific weight of particles,
g/cm3;
p1 = 1.6 g/cm3;
specific weight of water,
g/cm3;
2 = 1
do —
77 —
p ' —
2 —
3 Relationships between the head loss and
the flow in the sublayer resembling a very
thin gau?.e can be described by well-known
hydraulic formulae for outflow through
submerged orifice.
On the basis of experimental data
obtained for microstraining and
sedimentation of suspended solids, the
following formula can be used for
practical calculations:
P' -p2)-cm/sec-
where Do — the rate of the particles
sedimentation, cm/sec;
W -
0.33 — 10 U"
cm/sec
Besides filtering rate, relationships
between filtering cycle and the given
efficiency of suspended solids, capture and
construct dimensions of the filter and
determine the microstrainer's design:
T = f(Q, C...CT, D, n)
The relationship between the filtering
cycle duration and the filter dimensions is
determined by the following ratio:
T = _§_ .. sec.
V
where S - the length of the submerged
part of the drum perimeter, m.;
V - linear rotation speed of the
drum generating, m/scc.
The efficiency of microstraining is
characterized by the specific load
expressed by the amount of pollutants
detained by one working unit of
microscreen area versus unit of time.
G = JL (Co - Cr), kg/m'/hour
The dependence of the filtering cycle on
the specific load is described by the
empiric equation:
T = 0.0017 G"1-93. sec.
The technical-economic evaluation of the
microstrainers' employment has revealed
their considerable advantages over
standard schemes of treatment. Technical-
economic evaluation was carried on for
pre-treatment and final treatment of
biologically treated effluents. Both
schemes were evaluated at the treatment
capacities of plants equal to 50,000;
100,000 and 200,000 m3/day. Two
variants were compared in each scheme.
In the first variant waste waters were pre-
treated on screens, sand traps, in primary
clarifiers, aeration tanks, secondary
settling tanks with chlorination before
discharge into receiving waters.
In the secondary variant primary clarifiers
were replaced by microstrainers in order
to reduce capital expenditures. The other
treatment plants remained unchangeable.
The use of microstrainers instead of
primary clarifiers at treatment plants with
output 50,000. 100,000 and 200,000
m'/day gives the possibility to reduce
capital expenditures by 75-90%.
For final treatment of effluents treated in
the secondary settling tanks the following
two alternatives were compared:
1 - final treatment of the secondary
settling tank effluent in biological
ponds:
11 - tertiary treatment of the secondary
settling tanks effluent on
microstrainers followed by final
treatment in biological ponds.
Employment of microstrainers for final
treatment of biologically treated effluents
permits the area of biological ponds to be
half reduced. In its turn this reduces
capital expenditures for treatment plants
with output of 50.000-200,OOOmVday by
40-60%.
Conclusions
I. Microstraining has been employed
recently both as an independent method of
waste water treatment and as a part of
technological scheme at different waste
water treatment plants.
-------
2. Observation of full-scale conditions of
washing is necessary when modelling
microstrainers. The equipment developed
gives us the possibility to determine the
main parameters of microstrainer
operation and the screen regeneration in
laboratory and full-scale conditions.
3. Microst rain ing provides BODs
removal up to 25-30% and suspended
solids removal by 45% (after pre-
treatment) and suspended solids removal
by 50-60% and 30-40% removal of BODs
(for final treatment).
4. Analysis of the physical nature of
microstraining has shown that the process
can be described with regard for
granulometric composition of suspended
solids (SS).
5. Technical-economic calculations have
shown that microstrainers' employment is
profitable in case of prc-treatment of
waste water (instead of primary clarifiers).
Capital expenditures are reduced by 70-
90%.
Employment of microstrainers for final
treatment of biologically treated effluents
twice reduces the are of biological ponds.
ponds.
6. Microstraining is widely used for final
treatment of biologically treated effluents
of articifical fiber plants, pulp and paper
complexes, chemical enterprises and
municipal sewages. In case of
pretreatment the use of microstraining is
limited. It finds application mainly for
pulp and paper and chemical effluents'
pretreatment.
The Control of
Refinery Mechanical
Waste Water
Treatment Processes
by Controlling The
Zeta Potential
James F. Grutsch
Coordinator of Environmental
Projects.
Standard Oil Co. (Indiana)
Abstract
Granular media filtration and dissolved
air flotation play a key role in optimizing a
water management program for refineries.
Proper utilization of these processes can
achieve major capital and energy savings
as well as provide for enhanced
performance of biological treatment
systems.
An intrinsic property of solids in the
presence of water is an electrical surface
charge. Maximizing the efficiency of
filtration and dissolved air flotation
processes require that these coulombic
repulsive forces be controlled by
controlling the physicochcmical properties
of the dispersed solids. Of the four
theoretical colloid destabili7.ation
mechanisms, charge neutralization and
bridging are the key mechanisms to
optimize these unit operations whereas
colloid entrapment and double layer
repression are less attractive.
Zeta potential titration curves are used to
quantify proper destabili/.aiion chemical
treatment, compare the effectiveness of
chemicals, and measure synergisms and
antagonisms of various components in the
waste stream, thereby determining the
most cost effective chemical treatment.
Hydraulic loading limitations of granular
media filters relate chiefly to I) the nature
of the destabilized colloidal aggregates.
and 2) water temperature during filtration.
Biocolloids and coke fines represent
extremes of the former, and the latter
limitation is viscosity related, which
contributes to dispersion forces on the
captured aggregates.
For air flotation, the design of the recycle
air system and the surface loading of the
flotation zone are the critical design
elements.
Introduction
Granular media filtration and/or
dissolved air Rotation (DAF) units play a
key role in optimizing a water
management program for refineries.
Proper utilization of these process
elements can achieve major capital and
energy savings as well as provide for
enhanced performance of biological
treatment systems.
An intrinsic property of solids in the
presence ofwater is an electrical surface
charge. When colloids are being
considered, the electrical charge is called
zeta potential (ZP). Almost all matter
dispersed in spent process water such as
oil particles, silt, biocolloids. inorganic
colloids, etc., has a negative ZP and is
repelled by the negative electrical surface
charge of the granular media as flotation
bubbles. Mother Nature is always "left-
handed" and you must control Mother
Nature if granular media filtration and
DAF are to achieve maximum solids
removal effectiveness. Maximizing solids
removal efficiency requires that these
coulombic repulsive forces be controlled
by controlling the physicochemical
properties of the dispersed solids.
This article will:
1 Define and describe what is meant by
granular media filtrations and dissolved
air flotation (DAF);
2 Discuss briefly filtration and flotation
mechanisms;
3 Cite refinery examples where
application of granular media filtration
and DAF make a major savings possible
in an overall water management program;
44
-------
4 Describe and discuss the a) electrical
properties of waterborne suspended
matter, b) colloid destabili/ation
mechanisms, and c) the chemical
properties of destabilization chemicals
used to control these electrical properties;
5 Describe the approach to determining
optimal chemical treatment preparatory
to filtration and dissolved air flotation;
and
6 Present typical chemical pretreatment
and hydraulic guidelines for various
refinery filtration and dissolved air
flotation applications.
Granular media filtration
Granular media filtration may be defined
as clarification of a suspension of
dispersed material by passage through a
bed of porous media that separates, and
retains within the media, the solids
constituting the suspension.
Granular media filtration in the petroleum
industry usually means "direct filtration."
Direct filtration involves injection of
required chemicals into the water and
immediate transfer of the treated water to
the filter, i.e., there are no flash mix,
flocculation, or clarification process steps
prior to filtration.
The normal operation of granular filters
involves downward flow through the
media until pressure drops due to
clogging, or breakthrough of suspended
matter, increases to a predetermined limit.
The filter is then cleaned by reversed How
lluidi/ation after pretreatmcnl by air
scrubbing or a hydraulic surface wash.
Filter media include beds composed of
sand; sand and coal; sand, coal and
garnet; and other minerals and synthetic
materials. For best results, the suspension
being filtered should pass through a bed of
increasingly smaller pore si/.e.
A more recently developed bed
comprising four media that achieves this
objective is shown in Figure 1(1).
Granular media filters are frequently
referred to as gravity or pressure filters.
Since granular media filters are only a
small part of the spectrum in filtration art,
the meaning associated with these
descriptions may be at variance with other
filter technology. In the simplest terms, a
gravity filter is a downflow design in which
the water standing above the filter media is
under atmospheric pressure. A gravity
unit may be operated as a constant or
declining rate unit, i.e., as the filter media
clogs and the pressure drop increases, the
rate may be maintained by increasing the
head of water above the media, or allowed
to decrease by maintaining a constant
head. The pressure drop across a freshly
regenerated unit is about 1 foot of water
and the pressure drop at the end of the
filter cycle may be as little as 5 feet or as
much as 10-12 feet of water. A pressure
filter of the granular media type is simply
the same system in an enclosed vessel; i.e.,
the operating pressure drops across the
media arc about the same. In contrast.
pressure filters in filtration systems other
than the granular media type may have
pressure drops orders of magnitude
higher. Within the framework of these
descriptions it is obvious that a variety of
engineering and hydraulic designs are
possible. Historically, granular media
filtration was viewed as a polishing step
following a clarifier. More recently, direct
filtration of highly contaminated waters
has been widely demonstrated, with large
savings in capital, chemical, and sludge
treating costs.
Filtration Mechanisms
Mechanisms for retention of solids within
the pores of granular media arc separated
into two principal processes; a transport
step and an attachment step. The
transport step involves movement of the
dispersed phase material to the vicinity of
the granular media surface and the
attachment step involves attachment of
the particles on the media surface. The
transport mechanisms may involve
diffusion, interception, sedimentation and
hydrodynamic actions. The attachment
mechanisms may involve van der Waal
forces, electrical double layer interaction.
mutual absorption or hydrogen bonding
(2).
Experimental comparisons of the
filtration of colloidal ferric hydroxide
suspensions with the filtration of well-
flocculated suspensions of ferric
hydroxide, both performed under
identical conditions of filtration,
demonstrate a much higher filtration
efficiency for the floe. Studies (3) have
shown that removal of colloidal
suspension by filtration appears to be
possible only when the colloidal particles
to be filtered carry' an electrokinetic charge
opposite to the charge of the filter media
used. The removal of the colloid is by
electrokinetic sorption of the colloidal
particles on the surface of the filter media.
This kind of "filtration" phenomena has
limited application to the filtration of
industrial waste water because of
complications encount"red in application,
the complexity of which are beyond the
scope of this article.
Dissolved Air Flotation
Dissolved air flotation (DAF) may be
defined as clarification of a suspension of
flocculated material by contact with
minute bubbles that attach to the solids
constituting the suspension, causing the
air/solids mass to be buoyed to the surface
leaving a clarified water.
This definition puts proper emphasis on
the requirement that the material being
separated should not be colloidal, with the
inherent high negative /.eta potential
required to maintain the colloidal state,
but a flocculated suspension. A
flocculated suspension implies a proper
chemical pretreatment consistent with the
needs of the colloidal system. A
flocculated suspension requirement also
determines the design basis of the DAF
unit; it shall have provisions for mixing
chemicals, flocculating the destabilized
suspension, a flotation /one for phase
separation, and shall be the recycle air
pressurization design. The seven near
boxes in the API separator shown in
Figure 2 were converted to a DAF unit. A
profile view is shown in Figures 3 and 4.
45
-------
Figure 5 illustrates a critical design
problem. The vena contracta of the
pressure reducing valve creates a very
strong vacuum and will cause the
dissolved air to disengage as very large air
bubbles unsuitable for air flotation if the
valve is more than about 1 meter from the
distribution header. To keep the pressure
reducing valve out of the water, orifice
plates in the distribution header are used
to provide the required back pressure.
With a properly designed recycle air
pressurization system, when the pressure
is reduced the dissolved air separates as
finely divided bubbles ranging in size from
30 to 120 microns with most of them in the
60 - 100 micron range and ready for
attachment to the dispersed solids.
According to Vrablik d.'i the air can
become attached three different ways:
1. By adhesion of an air bubble to a
particle either by collision or as
condensation sites as the air precipitates
out of solution.
2. By entrapment of the air bubble in the
irregularities of flocculated particles.
3. By adsorption of the air bubble in the
floe structure as the floe is formed.
Provided the particles are adequately
destabilized, and the flotation zone is
relatively free from turbulence, the
particles tend to behave in accordance
with Stokes' equation:
_ g (dw — da) D2
(Equa. I)
Vi = terminal velocity of air-solid
agglomerate
dw = density of water
da = density of air-solids agglomerate
D = diameter of air-solids agglomerate
u = viscosity of water
u = gravity constant
For the air flotation process to work, the
value of the term (dw-da) must be greater
than zero. The rate of rise varies directly
with the square of the diameter of the air-
solid agglomerate and inversely with its
density. Density can be changed only by
making more flotation air adhere to the
particles. Diameter can be easily increased
by substantial amounts by means of
chemical coagulation and flocculation. In
addition, the flocculated particles will trap
and retain more air than individual
particles.
The key to optimizing the efficiency of
filtration and DAF operations, therefore,
is to recognize that essentially all colloidal
systems encountered in nature have a
negative electrical surface charge.
Considering that, in the case of filtration,
granular media (coal, sand, garnet, etc.)
and in the case of DAF, the flotation
bubbles, have a negative electrical surface
charge also, it is not unexpected that
colloidal solids rapidly break through
granular media due to repulsion of like
charges when not properly chemically
pretreated, or have a poor capture
efficiency in the case of DAF. The optimal
application of these processes to waters
relates primarily to recognizing and
responding to the required water
chemistry for destabilizing the colloidal
material into agglomerates tough enough
to resist redispersive hydraulic forces.
Refinery Examples Where
Application of Granular Media
Filtration and DAF Results in
Major Savings
The elements of inplant and end-of-pipe
treatment which must be addressed at
each refinery are outlined in the
"Development Document for Effluent
Limitations Guidelines and New Source
Performance Standards for the Petroleum
Refining Point Source Category." The
objectives and some process options for
each of these elements are outlined in
Figure 6. Filtration is used for intake
water clarification, and the filtration of
primary and secondary effluents in the
end-of-pipe treatment sequence. DAF
treatment of primary effluents is an
option.
Filtration or DAF Treatment of API
Separator Effluents Results in ASU
Savings
The filtration or DAF of primary effluents
can have a major impact on the
purification effectiveness of secondary
treatment, and the amount and
characteristics of waste activated sludge
produced. As in municipal wastes, a
refinery's raw waste load (RWL is defined
as contaminants in primary effluent)
BOD/COD is characterized as being
about 50% soluble and 50% insoluble, on
the average. But the NPDES Permit
program requirements are restrictive
enough that the designer of process units
must use about a 98% probability basis for
RWL; and here is where the use of
filtration or DAF results in a major capital
and operating savings.
The impact of granular media filtration or
DAF on the design of a following
activated sludge unit (ASU) can be
visualized by making probability plots of
refinery RWL data (Figures 7, 8, and 9).
In large and complex refineries the oil and
grease in the raw waste load can vary over
a year's time as shown in Figure 7.
Whereas the 50% probability (median) is
only 70 mg/1 oil, about 5% of the data
may be in the 600+ mg/1 range due to
causes previously outlined. Optimized
ASU operation at very high sludge ages
has been achieved with excellent sludge
properties when all discontinuous phase
oil and solids are removed. When
intermediate treatment includes chemical
destabilization and filtration, results for
oil and grease as shown in Figure 7 can be
achieved.
Removing oil from the process water
effluent prior to biological treatment
recovers the oil in a sludge much easier
and less costly to handle that if the oil is
commingled in a waste activated sludge.
Waste activated sludge is costly enough to
dewater but when commingled with oil,
the oil blinds precoat filters, increases
significantly the volume of waste sludge to;
be dewatered, and makes the final
disposition of dewatered oily sludge more
troublesome.
Equalizing feed quality in terms of organic
loading also can be achieved by filtration
or DAF as shown for BOD/COD data in
Figures 8 and 9. Colloidal and suspended
46
-------
matter contributes significantly to the
total BOD/COD and periods of storm
flow which flushes sludge and silt from
sewers, high pH, loss of emulsifiers, coke
fines, clay fines, etc., typically give a highly
variable raw waste load which is
frequently underestimated because of
sampling and testing difficulties. Removal
of the discontinuous phase contaminants
can have a large impact on reducing the
total organic load and the variability in
quality of water entering secondary
treatment.
Equalizing the organic loading results in
substantial savings in capital and
operating costs, simpler operations, and
better effluent quality with less variability
in quality of water leaving the secondary
treatment facilities. For example, data
from Figures 7, 8, and 9 are as follows:
% Probability
Less Than
Indicated Value
Contaminant
Oil and Grease,
mg/1
BOD, mg/ 1
COD, mg/ 1
Treatment
After
Primary
Intermediate
Primary
Intermediate
Primary
Intermediate
50 95
70 600
4 10
185 380
80 96
400 980
220 310
98
800
12
400
105
1,400
350
An engineer faced with design of an ASU
to operate in the nitrification mode at 2
mg DO/1 with uniform effluent quality
faces an almost insurmountable challenge,
both economically and performance-wise,
if he has to deal with the variability of the
raw waste load. Considering that data in
the preceding table represent a year of
operations, they also reflect (1) seasonal,
(2) weekly, (3) daily, and (4) hourly cycles.
The final design of an ASU based on
primary effluent would be such a
compromise that year-round operation
would be unsatisfactory, yet the unit
would be expensively over-designed. If,
for example, a guideline of 1 Ib. 02/lb.
COD is used and the 98% COD value is
selected as design basis, the raw waste
would require four times the aeration
horsepower the intermediate treated waste
does. Actually, it would require
substantially more, because the alpha
(oxygen transfer) characteristics of the
raw waste are only about one-half as good
as the intermediate treated waste.
Filtration or DAF of API Separator
Effluents Simplifies ASU Process Control
The amount of colloidally dispersed inert
solids and oil entering the aeration tank
affects the degree of process control
available to the operator. Colloidal and
suspended matter in the influent to the
ASU at quite low concentration levels can
make the process nonoperable at the
optimum conditions.
The ASU process control is via the food-
to-microorganism (F/ M) ratio, or perhaps
more practically the sludge age (SA). The
sludge age and F/ M ratio are related as
follows:
1 _ a (AF/AT)
SA " M
(Equa. 2)
where, a = the sludge yield
coefficient,
b = the endogenous rate
coefficient,
M = the mass (Ibs) of
microorganisms in
the system,
SA = the sludge age (days),
and
AF/AT = the mass (Ibs) of food
(BOD or COD)
supplied per day.
The F/ M process control involves meas-
urement of the BOD or COD load
(AF/ AT) per day and adjustment of the
sludge inventory (M) to maintain a desired
ratio. The sludge age method of process
control can provide a simple hydraulic
means to achieve the same end by sludge
wastage. The sludge wastage rate to main-
tain an indicated sludge age is calculated
from the equation:
Wastage (MGD) =
I. (YA + Vc) XA
X-mr f, . ~" IJ^\C
r - AC I>A
(Equa. 3)
where, Xr = sludge concentration in
recycle
Xe = sludge concentration is
effluent
XA = MLSS
VA = volume of aeration tank,
MMGD
Vc = volume of clarifier, MMG
Q = feed volume, MMGD
The sludge age method of process control
has many advantages which are discussed
in detail by Walker (4).
The accumulation of negatively charged
colloids, inerts and oil in the activated
sludge mass is hypothesized to be a
principal cause of poor sludge setting
properties (SVI) and dispersed solids,
thereby limiting the effective SA or F/ M
operating range. The sensitivity of the
ASU to influent inert solids can be readily
demonstrated. For example, if undesirable
materials such as heavy oils, catalyst fines,
clay, coke fines, metallic sulfides, etc.,
escape preceding treatment at the rate of
30 mg/1. become flocculated by activated
sludge and accumulate, their effect can be
predicted. Assuming for simplicity of
calculation that the inerts can escape only
via sludge wastage and not via the clarifier
overflow, the equilibrium character of the
mixed liquor solids can be estimated using
equations (2) and (3). These data in Figure
10 show that a fixed amount of entering
inert solids accumulate in the total solids
mass at an ever increasing rate with
increasing sludge age. Inspection of the
curve for biological solids shows that
operation at high sludge age will result in
minimal waste sludge and process control
Jests and responses,
A more comprehensive discussion of the
advantages for complete phase separation
is available (5,6).
Filtration of Raw Intake Water Markedly
Reduces Oily Primary Sludges
In a refinery case history involving
filtration of raw intake water, removal of
intake solids markedly reduced and
simplified solid waste problems (6).
Solids in intake water that contacts, or
commingles with, plant process water can
have a major impact on waste sludge
generation. In a refinery, for example, a
47
-------
make-up water containing about 120
mg/1 suspended solids contributed 25,000
IDS/ D solids to the plant system. A solids
material balance around primary
treatment comparing the cases of intake
solids removal with no removal is included
in Figure 11, In this example, removal of
intake solids by filtration reduced the
amount of wet. oily primary sludges
generated by more than 85%.
Solids removed from intake water by
filtration don't, of course, just disappear.
Direct filtration of raw water using only
polyelectrolytes for colloid destabilization
yields filter backwash solids so receptive to
modest chemical treatment that further
mechanical dewatering cannot be justified.
Combining these backwash solids with
lime softening sludge yields a significant
chemical and dewatering synergism such
that by combining sludges at least 25% less
thickened sludge volume is produced. The
flocculation and dewatering action of
combined sludges is virtually
instantaneous (6).
Overall Impact of Filtration on a Refinery
Water Management Program
The overall impact of filters on a refinery
water management program can be
illustrated by comparing the amount, and
kind, of sludges generated in two refinery
cases; one involving filtration of activated
sludge plant effluent only, and the second
involving filtration of intake water,
primary effluent, and activated sludge
plant effluent.
The amounts of sludges generated in the
two cases are shown in Table I for a case
history. The use of three filtration steps
reduced the amount of sludges generated
by more than one-half that of the single
filter case. Removing intake solids at the
source resulted in a major, net reduction
of primary and intake sludges, and
removing primary solids resulted in a
major, net reduction of waste activated
and primary sludges.
Not only is the amount of sludge
generated significantly reduced by the
three stages of filtration, but even more
important from a handling and cost basis
is the fact that the sludge properties are
much improved as outlined in Table 2. As
shown, the generation of the large amount
of oily, secondary sludges with the
noxious handling problems previously
described can be avoided. For a major,
integrated water management program,
therefore, the use of three stages of
filtration can result in very large capital
and operating cost savings of ancillary
facilities for sludge thickening,
dewatering, and disposition. Additionally,
savings accrue from the more modest
secondary facility requirement because of
the major reduction in raw waste load
achieved by the filtration of primary
effluent, and the more favorable mode of
activated sludge unit operation. (5)
Bioenhancemenl
With reference to Figure 6, effluent
standards for 1983 (best available
technology economically achievable,
BATEA), are predicted upon addition of a
stage of granular carbon treatment to the
1 977 best practicable technology currently
achievable (BPTCA) treatment sequence.
Addition of granular carbon to the
BPTCA sequence will typically double the
capital investment and more than double
the operating costs of the effluent
treatment sequence. In response to this
incentive, research by the petroleum
industry has recently demonstrated that
the BPTCA sequence can be optimized.
and the ASU process "enhanced" such
that the effluent water quality is equivalent
to that treated by a stage of granular
carbon.
Part of this research achievement lies in
simply responding to the biological kinetic
expression of Adams et al ( 1 1 ):
= S«(kF-i + 1)
(Equa. 5)
„ •
Se=
S<>
kMt
where.
Se = soluble organics in effluent,
mg/l
S<> = organics in influent, mg/ 1
k = kinetic constant
M = biomass, mg/ 1
t - aeration time
Since S»/Mt is the F/M (food/ micro-
organism) ratio, letting F - F/ M equation
4 becomes:
For best effluent quality from the ASU,
equations 4 and 5 say that So and'F should
be minimized. Or, from equation 2, the
sludge age (S A) should be maximized. The
most important means for minimizing So
is to remove essentially all colloidal matter
remaining after primary treatment
(Figures 7, 8, and 9). Once this is achieved
F can be minimized (SA maximized)
because the biological floe properties
(Sludge Volume Index) do not deteriorate
with increasing sludge age. This is
illustrated in Figure 12 where the SV1
properties of activated sludge from a unit
with prefiltered feed is compared to
literature data for unfiltered feed (12).
Some recent data from two refineries are
available treating refinery effluents
according to the BPCTCA sequence
(Figure 5) in which the performance of a
unit passing colloidal solids to the ASU
was compared to the performance of a
parallel treatment in which the colloids
where removed:
Refinery A
Parameter
Soluble organic
carbon
Soluble chemical
oxygen demand
Suspended solids
Phenolics
Refinery B
Soluble organic
carbon
Soluble chemical
oxygen demand
% Improvement
with Colloid
Removal
20.5
31.1
81.7
59.8
26
33.7
These data demonstrate that removing
colloids essentially completely prior to
activated sludge treatment of the residual
solubles yields a significant improvement
in effluent quality. The principle subject of
this article, therefore, is to address the
chemistry of these colloids to illustrate
how their removal can be optimized- by
granular media filters or DAF units. •
48
-------
Properties of Suspended Solids
Refinery effluents from API separators,
aerated lagoons, fire and cooling water
ponds, etc., are similar to surface waters in
that the suspended materials usually are
predominantly colloidal, or a combination
of colloidal and very slightly flocculated
suspensions. The stability of these
colloidal systems relates to the fact that
the individual particles carry like electrical
charges, causing their mutual repulsion.
Except for some very isolated examples,
the charge on organic, inorganic, and
biocolloids is negative when suspended in
water. Colloidal destabilization by
chemical treatment has the objective of
neutralizing, or reducing, the electrical
charge so that mutual repulsion is reduced
to the extend that individual particles can
approach each other close enough for van
der Waals and/or chemical forces to
become effective. The attractive van der
Waals forces cause the particles to
aggregate into agglomerates, which
facilitates their removal by sedimentation,
DAF, or nitration processes. The surface
charge on colloidal particles may be
estimated by electrophoretic,
electroosmotic, streaming and
sedimentation potential techniques.
We have found that the electrophoretic
procedures and equipment of Riddick (7)
permits the rapid determination of
colloidal charge to be made and all our
investigations involved use of the Zeta
Meter. Accordingly, electrokinetic values
reported herein are zeta potentials (ZP). •
Colloid Destabilization
Mechanisms
Destabilization of the waterborne
suspended solids may involve four
mechanisms: (1) colloid entrapment or
removal via the-sweep floe mechanism, (2)
reduction in surface charge by double
layer repression, (3) bridging by polymers,
and (4) charge neutralization by
adsorption.
Colloid Entrapment
Colloid entrapment involves chemical
treatment with comparatively massive
amounts of primary coagulants; the
amount of coagulant used is typically so
great in relation to the amount of colloidal
matter that the nature of the colloidal
material is not relevant. The amount of
primary coagulant used may be 5 to 40
times as much as is used for charge
neutralization by adsorption. The rate at
which the primary coagulants form
hydrous metal oxide polymers (Figure 12)
is relatively slow and depends chiefly upon
water temperature and pH. Coupled with
the high conentration used, all negatively
charged colloidal material is initially
exposed to charge neutralization by the
transient cationic species. The polymer
matrix is 3-dimensional and voluminous,
as illustrated by Figure 14, providing for
entrapment of solids. As the polymer con-
tracts, freeing solvent water molecules,
and settles, the suspended solids remain
enmeshed in the settling floe and appear to
be swept from the water, hence the
description of the process as a "sweep
floe" mechanism. This destabilization
mechanism results in the generation of
large amounts of wet alum (or iron)
sludges, which are difficult and costly to
dewater. Even though it is by far the most
widely used mechanism for water clarifica-
tion it is to be avoided in granular media
filtration and DAF because of the sludge
problem and because the use of other
mechanisms result in significantly lower
operating and capital costs. In the case of
direct filtration, the comparatively
massive chemical treatment used in this
destabilization mechanisms rapidly blinds
the filter, causing very short runs.
Double Layer Repression
Reduction in surface charge by double
layer repression is caused by the presence
of an indifferent electrolyte, which in
refineries is chiefly sodium chloride from
brackish water usage or salt water ballast.
For water and monovalent electrolytes,
the thickness of the double layer is
approximately 10 Angstroms (Ao) for
O.IM, IDOAoforO.OOlM, and 1000 Ao for
0.00001M. For double layer repression of
colloid surface charge in brackish waters,
the sodium ions of the indifferent
electrolyte, which surrounds the colloid
particles in order to electrically balance
their negatively charged surfaces, have less
tendency to diffuse away from the colloid
surface as the salinity increases. Some salt
concentration may eventually be reached
such that the thickness of the double layer
may be small enough that two colloids
approach each other closely enough that
van der Waals forces cause aggregation.
An important aspect of double layer
repression is that the quantity of colloidal
charge is not significantly reduced, but just
the extent to which it extends out from the
colloid surface. This relates to the nature
of the destabilizing chemical (salt) and its
mode of action; i.e., the sodium ions
remain free in the solvent and cause rapid
dissipation of the charge as the distance
from the colloid surface increases. Double
layer repression can improve solids
removal by direct filtration, but this
mechanism does not achieve the best
results and can conceal definition of
optimal chemical pretreatment lo achieve
best filtration results if the interference of
this destabilization mechanism is not
' recognized. Our refinery experience
indicates that the colloidal aggregates
destabilized by double layer repression are
readily redispersed by hydraulic forces as
if the net binding forces are very weak.
Bridging
Bridging by organic and inorganic
polymers describes the destabilization
mechanism where the molecules of the
added chemical attach onto two or more
colloids causing aggregation. Cationic
polyelectrolytes up to about 10.000,000
molecular weight are available for this
service. Weakly anionic organic polymers
are negatively charged; however, they are
useful for aggregating and binding
together some moderately negatively
charged aggregates into agglomerates that
resist redispersion. Thus, in this latter
instance, attractive forces of a chemical
nature overcome moderate electrostatic
repulsion forces due to like charges.
Bridging by polymers proved to be an
important destabilization mechanism for
application to direct filtration and DAF.
Charge Neutralization
Charge neutralization bv adsorption of
the destabilizing chemical to the colloid is
49
-------
a key mechanism for optimizing removal
ofwaterborne solids from waters by direct
filtration. Perhaps adsorption is a poor
word choice here, since the mechanism
may not be different from bridging
previously discussed. While the
mechanism may be the same, the end
results are different in that the colloidal
charge may not only be reduced to zero,
but beyond zero, i.e., reversed. Charge
neutralization by adsorption infers that
the colloid-water interface is changed and
thereby its physicochemical properties.
Destabilizing Chemicals
Primary Coagulants
Efficient destabilization of colloidal
suspensions using salts of iron and
aluminum as primary coagulants must
recognize the properties of these primary
coagulants. The chief properties of
concern are the ZP-pH relationships and
hydrolytic reactions.
Stumm and O'Melia (8) describe the
equilibrium composition of solutions in
contact with precipitated primary
coagulants in the interesting manner
shown in Figures 15 and 16. These
diagrams are calculated using constants
for solubility and hydrolysis equilibria.
The shaded areas A and B we have added
in each figure are approximate operating
regions for air flotation and clarifiers by
colloid entrapment (region A) and direct
filtration by charge neutralization (region
B). Both regions are assumed to cover a
pH range of 6.0 to 8.5. The coagulant
dosage ranges from 33 to 200 mg/1 in
region A and 3.3 to 20 mg/1 in region B.
These figures are useful in the
interpretation of some of our filtration
results.
With reference to Figure 16, the isoelectric
point for ferric hydroxide coincides with
the region of minimum solubility, and the
operating regions for water treating
(destabilization) yield a hydrolyzed
primary coagulant with a desirable
positive zeta potential.
In many refinery situations, however, it is
difficult to use this attractive condition
because the presence of sulfides and
strongly reducing conditions cause the
reduction of ferric to ferrous iron and the
formation of mixed iron sulfides with no
coagulation powers. In fact, in some
refinery waters the use of iron coagulants
at modest dosages may contribute to
stabilizing solids rather than destabilizing
them.
While alum has no redox or sulfide
chemistry comparable to iron, it's
amphoterism and solubility pose definite '
limitations on alum usage. With reference
to Figure 15, a substantial portion of
operating region B lies in the area where
alum is soluble and the predominant
equilibrium specie is negative, Al (OH)4.
In the more acidic part of region B,
however, the concentrtion of equilibrium
ionic species is very much lower, and
much less negative. Considering these
data, it is not unexpected that
investigators consistently report optimal
coagulation/ flocculation results with alum
at a pH of 5-6.
With inspection of Figure 15, one may
question why alum is effective at all for
neutralizing negatively charged colloids in
the indicated operating regions. One
approach to explaining observed
performance requires understanding that
the data are equilibrium data; but before
equilibrium is reached, substantially
different conditions exist.
Alum very readily hydrolyzes to form
polymers in a complex manner not well
defined. The hydrolytic pathway and
reaction rates are affected by pH,
temperature, other ions, etc. One
hypothesized route which includes
different aluminum hydrolysis products
which are known to exist is outlined in
Figure 13. When alum is added to water in
amounts which exceed the solubility
limits, sequential kinetic reactions occur
until the ultimate precipitate is formed
and the ionic species appropriate to the
pH equilibrate with the precipitate. The
hydrolytic reactions are not
instantaneous, and as they proceed
positively charged hydroxo polymers are
formed which are available for colloid
adsorption. The hydrolyzed species have
enhanced adsorption capabilities, possibly
due to larger size and less hydration and
the presence of coordinated hydroxide
groups (8). In solutions more alkaline than
the isoelectric point the positively charged
polymers are transient and, at ..
equilibrium, anionic polymers prevail.
in modestly alkaline solutions the
transient positively charged polymers
appear to contribute to destabilization of
colloids. On the other hand, in solutions •
more acidic than the isoelectric point the
positively charged polymers prevail at
equilibrium and destabilization of colloids
may be achieved at significantly lower
coagulant treatment levels.
In Figure 17 the zeta potential of colloidal
iron hydroxide fresh water solutions is
plotted as a function of pH. The zeta
potential decreases in positive charge as
the pH increases until the isoelectric point
is reached at a pH of 8.3 at which the
charge reverses. In the vicinity of the
isoelectric point the charge may vary as
indicated. Alum has a similar fresh water
zeta potential -pH relationship as shown
in Figure 18. The zeta potential may be
negative or positive over the pH range of
7.0 to 7.8.
Surfactants
Certain substances, even when present in
very low concentrations, possess the
unique property of altering the surface
energy of their solvents to an extreme
degree. Almost always, a lowering rather
than an increase of the surface energy is
affected. Substances or solutes possessing
such properties are known as surface-
active agents or surfactants and their
unique effect is known as surface activity.
By broad definition then, surface-active
chemicals are soluble substances whose
presence in solution markedly changes the
properties of the solvent and the surfaces
they contact. They are categorized
according to the manner in which they
dissociate or ionize the water and are
characterized, structurally, by possessing a
molecular balance of a long lipophilic,
hydrocarbon "tail" and a polar,
50
-------
hydrophilic "head."
Surfactants owe their physicochemical
behavior to their property of being
adsorbed at the interface between liquids
and gases (where they contribute to the
electrical charge on the DAF bubble), or
liquid and solid phases (where they may
contribute to the zeta potential).
Surfactants tend to concentrate in an
oriented manner, at the interface, in such a
ways that, almost entirely, they turn a
majority of their hydrophilic groups
toward the more polar phase and a
majority of their lipophilic groups away
from the more polar phase and perhaps
even into a nonpolar medium. The surface
active molecule or ion, in a sense, acts as
sort of a bridge between two phases and
makes any transition between them less
abrupt.
There are three types of chemical surface-
active agents which are classified
according to their dissociation
characteristics in water. These are:
1. Anionic Surfactants—Where the
electrovalent and polar hydrocarbon
group is part of the negatively charged
ion, when the compound ionizes:
ANIONIC
CH3(CH2)i6COO"Na+
2. Nonionic Surfactants—Where the
hydrophilic group is covalent and polar
and which dissolves without ionization:
NONIONIC
CHs(CH2)i6 COO(CH2CH2O)H
3. Cationic Surfactants—Where the
electrovalent and polar hydrocarbon
group is part of the positively charged ion
when the compound ionizes:
CATIONIC
CH3(CH2)i7NH3~tCL~
Surfactants are powerful charge
neutralizers (and charge reversers). In the
petroleum industry anionic surfactants are
used as emulsifiers for asphalt by
imparting a zeta potential on asphalt
particles ranging from -30 to -80mV.
Cationic types impart a zeta potential
ranging from + 18 to 128 mV. Each
surfactant possesses a distinct
characteristic capability of imparting
quantitatively to asphalt during
emulsification a specific /.eta potential.
Surfactants have not found wide use for
destabilizing colloidal systems. In fact,
they are an important cause for the
existence of colloidal systems, particularly
in primary municipal effluents. The
principal organic colloidal destabilizing
chemicals are polyelectrolytes.
Polyelectrolytes
Polyelectrolytes used as water treating
chemicals are macromolecules having
many charged groups and may be
classified as cationic, anionic, and
nonionic depending upon the residual
charge on the polymer in solution.
Examples of the structural types are
shown in Table 3.
In solution the polyelectrolytes are
dissociated into polyvalent macroionsand
a large number of small ions of opposite
charge (counter ions). The macroion is
highly charged, which is the cause for the
characteristic properties of the
polyelectrolytes. Most of the macroions
are long, flexible chains, their size and
shape depending on the macroion charge
and interaction with counter ions. With
increasing charge the macroion extends;
with decreasing charge the macroion
assumes a contracted random coil. The
source of the charge is illustrated by the
polyacrylates, a widely used polymer. In
distilled water polyacrylic acid's
carboxylic functional group is only
slightly dissociated. The addition of
NaOH reacts with the carboxylic acid
groups causing them to dissociate leaving
a charge on the macroion and producing
sodium counter ions as shown in Figure
19.
The number of monomers in a
polyelectrolyte may range from about 102
to 10s. In the case of polyacrylic acid the
length of each monomer along the chain is
2.5A° (Angstroms). Since the radius of
each atom is about 1 A°, a polymer
macroion is envisioned as a flexible
cylinder like a garden hose. The flexibility
of the macroion exists because bonds in
the main chain can rotate around the
neighboring bond keeping the bond angle
constant, the macroion has. therefore, a
very large number of conformations
classified as cither random coil or-helix
(extended). In the random coil there is no
long range regularity in the bond angles.
In the helix the chain bond angles have a
long range regularity.
The Dimensions Involved
The dimensions of the various
components involved in colloid
destabilization vary a million-fold, from a
few A° to more than 10* A° as shown in
Table 4.
Where color is not a significant factor, the
problem is usually one of causing colloidal
particles down to about 1000 A° in
diameter to aggregate. When a clarifier or
DAF is used for phase separation, it is
desirable to build aggregates to fairly large
size, say greater than 106A°. On the other
hand, when filters are used simply
destabilizing the colloidal particles is
sufficient, because the destabilized
particles will build aggregates in the filter
bed as the destabilized suspension passes
through the media and the particles
impinge and adhere to the media or
trapped suspended matter.
Systematic Approach to
Determining Chemical
Treatment Requirement
Broad experience in refinery effluent
treatment led to outlining the condition-
response schematic for chemical treatment
of waterborne colloids shown in Figure
20. In phase removal by filtration, or even
DAF, we are not concerned with, and
indeed it is desirable to avoid, the use of
(1) the "sweep floe" or colloid entrapment.
and (2) the double layer repression
mechanisms for colloid destabilization.
Destabilization efforts must focus on the
charge neutralization and bridging ,.<•"''
-------
mechanisms. Charge neutralization
correlated with plant performance as the
optimum destabilization mechanism. For
plant control of direct filtration, charge
neutralization has been the key test
parameter correlating with performance
of refinery filters. Brackish water required
that charge neutralization be measured
after dilution with distilled water to
separate the effects of double layer
repression and charge neutralization; i.e.,
under plant conditions of high salinity, the
addition of destabilization chemicals
could reduce the ZP to approximately
zero by a range of chemical treatments:
however, when double layer repression
was the cause of reduced ZP, reduced
filter run lengths and performance were
observed. Reducing the ZP to
approximately zero, as measured by
means responsive to charge neutralization,
point out more definitely the required
destabilization chemical treatment and
resulted m optimum filter performance.
Waterborne colloids subject to chemical
destabilization and phase separation fall
into two general categories: relatively
inert substances such as clays, sand, and
organic materials; and microorganisms or
biocolloids. Both categories of colloidal
matter may be stabilized because they are
charged and/or arc highly hydrophilic.
Bolh categories of colloidal matter also
may vary in response to treatment by
destabilization chemicals, and within each
category the state of subdivision seems to
require additional consideration; i.e.,
extremely small colloidal particles are
sometimes more difficult to aggregate for
removal by filtration. Typically,
destabilization of biocolloids. such as are
in aerated lagoon effluents, is a more
demanding problem.
In the case of polyelectrolytes, some
counter ions at high concentrations screen
the charged functional groups with an
ionic cloud as previously described.
Salinity, hydroxide, phenolics, sulfides,
etc., are examples of the kind of counter
ions found to affect various cationics.
Each waste water application of cationics
must address the contaminants present if
">t£e most cost-effective polyelectrolyte is to
Titration Curves
A comparison of polyelectrolyte
performance and determination of
antagonisms and synergism is
conveniently and quantitatively
determined by use of ZP titration curves.
Because the stability of a suspension is
determined by the balance between the
short range (van der Waals) attractive
forces and the repulsive coulombic forces
between the particles, the objective of ZP-
cationic polyelectrolyte titration curves is
to quantify the amount of polyelectrolyte
needed to reduce the repulsive coulombic
forces to levels that permit total
destabilization by attractive forces (10).
ZP titration curves were determined using-
the equipment of Riddick (7). Increments
of regent are added to the water being
investigated, mixed and recirculated
through the electrophoretic mobility cell
for two minutes, and then the ZP
determined. Plotting the data points yield
ZP titration curves as in Figure 21 which
show the effect of pH on the charge
reversal properties of two good cationics.
In this example, when compared to pH 8,
the cationic 1190 is synergized and C-31 is
antagonized at pH 10.
The sensitivities of the two other good
cationics to pH and phenol are shown in
Figures 22 and 23. These data show that
431 is not significantly affected by phenol;
however, phenol not only antagonizes 581
at a pH of 7.8, but overcomes a pH
synergism and antagonizes 581 at a pH of
10.
Biocolloids, being quite hydrophilic, can
be more challenging a problem if tough,
stable aggregates are required (I).
Example ZP titrations for biocolloids in
aerated lagoon effluent are shown in
Figure 24.
The cationics in Table 5 are listed in order
of the amount required to reach zero zeta
potential. This "end-point" is somewhat
arbitrary since many colloid systems are
destabilized adequately for phase removal
when the zeta potential is only reduced to -
5 or -3 rnV.
Comparative rankings at these other "end-
points" are also shown; A summary of
polyelectrolyte synergisms and
antagonisms to salinity, pH and sulfides is
shown in Table 6 (9).
While the most cost-effective chemical
treatment would be to use just enough to
reduce the ZP to permit van der Waals
forces to predominant, in some cases at
least it is not necessary to "titrate" the
chemical addition this carefully to achieve
excellent results; i.e., a minimum amount
of chemical usage may be determined but
adding too much is not harmful.
Consider the refinery problem of the
aggregation of coke fines in hydraulic
decoking water so they are readily
removable. As shown in Figure 25, coke
fines are originally stabilized by a negative
ZP, but the negative charge is readily
neutralized and reversed by cationic
polyelectrolytes. One might think that too
much cationic .would simply restabilize the
system as a positive colloid. However,
using a cationic that has good bridging
properties in addition to charge
neutralization causes most of the solids to
be enmeshed by the polymer. The larger
size aggregates are more readily separable
even though comparatively highly
charged. As shown in Figure 25. addition
of a small amount of high molecular
weight weakly anionic polyelectrolyte to
the positively charged cationic treated
coke particles will once again reverse the
charge. Rather than redisperse the
particles, however, the weakly anionic
polymer efficiently "collects" the positively
charged particles into massive aggregates
easy to separate; i.e., once the high
molecular weight weakly anionic polymer
establishes bonds with the solids and
forms aggregates, the aggregates are
bound together with strong enough forces
to resist redispersion by hydraulic forces
in a clarifier, filter or DAF unit (9).
Summary and
R ecommendations
Chemical Destabilization
API separator effluents and other refinery
waste streams contain the kinds of
suspended matter that respond to rather
simple chemical treatment.
52
-------
Destabilization is achieved by using the
proper amount of an effective cationic
polyelectrolyte to achieve charge
reduction. The destabilized aggregates are
then agglomerated by using a weakly
anionic polyelectrolyte such that the
agglomerates are not redispersed by
hydraulic forces in the filters used for
phase separation; 0.02 to 0.05 mg/1
anionic polyelectrolyte is the expected
concentration range for anionics. For
DAF units the cationically destabilized
aggregates are agglomerated into tougher
floes by using about 0.5 mg/1 of a weakly
anionic polymer. The most widely
applicable weakly anionic polymer is a
polyacrylamide polymer of about
15,000,000 molecular weight with from 5
to 10% of the amide groups hydrolyzed.
Granular Media Filtration and DAF
If the chemical destabilization
recommendations are followed, the
hydraulic loading limitations of most
commercially available granular media
filters relate chiefly to I) the nature of the
destabilized aggregates, and 2) water
temperature during filtration.
Two extremes pointing out the
importance of the nature of the
destabilized suspended matter are the data
for coke fines (Figure 24) and biocolloids.
Coke fines form large, nonblinding, tough
agglomerates that resist redispersal with
proper chemical treatment. Very high
hydraulic rates are expected with
agglomerates with these characteristics.
On the other hand, an intrinsic property of
biocolloids is their hydrophilicity which
makes their chemical destabilization more
challenging. Even with optimized
chemical pretreatment, destabilized
aggregates of hydrophilic solids are
especially sensitive to the temperature of
filtration because the chemical forces
binding the aggregates are not as strong as
for hydrophobic colloids. The hydraulic
loading-temperature guidelines
recommended for this most challenging
case arc shown in Figure 26.
For DAF units a flash mix (1-2 minutes)
and floccuiation zones (10-15 minutes)
using two chemicals (cationic for
destabilization and anionic for
floccuiation) are recommended
pretreatment. A flotation zone loading of
1.5 gpm/ft2 und 30% recycle at 50 psi air
pressurization are recommended for
optimal phase separation.
Bibliography
Grutsch, J. F., Mallatt. R. C., and Peters,
A. W., Chemical Coagulation/ Mixed-
Media Filtration of Aerated l,Mgoon
Effluent, EPA-660/2-75-025, June 1975.
Ives, K. J., Filtration and Separation,
Nov./Dec., p. 700(1970).
Heertjes, P. M., and Lerk, C. F., Tran.i
fnstn. Chem. Engrs., p. Tl39. (1967).
Walker, L. F., Jour. WPCF. 43, No. I. 30
(1971).
Grutsch, J. F.. and Mallatt, R. C.,
Hydrocarbon Processing 55. No. 4. 21.1
(1976).
Grutsch, J. F., and Mallatl, R. C.,
AWWA 95th Annual Conference
Proceedings, Paper 15-1 (1975).
Riddick, T. M., Control of Colloid
Stability Tlwough Zera Potential.
Livingston Publ. Co.. (1968).
Stumm, W., and O'Mclia, C. R., Jour.
AWWA, 60. 514(1968).
Grutsch, J.R., and Mallatt, R.C.,
Hvtirocarbon Processing. 55. No. 6, 115
(1976).
Grutsch, J. R., and Mallatt, R. C.,
Hydrocarbon Processing, 55. No. 5, 221
(1976).
Adams, C. F.., et al.. Water Research. 9, 37
(1975).
Eckenfelder, W. W., Manual of Treatment
Processes. V. 1. Chap. 4, p. 7.
Environmental Science Services Corp.
Vroblik, F,. R.,Proreecii>jgx 14th Industrial
Waste Conference. Purdue Univ., 1959,
pp. 743-779.
-------
Source of Sludge
Table t
Sludge Balance From Water Treating Operations
Sludges, Tons/Day
Filtration of Solids From
Filtration of Activated Intake, API Separator, and
Sludge Effluent Only Activated Sludge Effluents
Treatment of Make-up Water
Intake Water, Solids
Boiler Water Treatment
Combined, after Thickening1
Primary Treatment of Effluent
API Separator Sludge1
Intermediate Treatment of Effluent
Filter Backwash Sludge1
'Secondary and Tertiary Treatment
Waste Activated Sludge2
Totals
1 Thickened to 10%, solids.
- Thickened to 4% solids.
0
(200)
200
144
0
562
906
(3900)
(200)
250
19
28
140
437
Kind
Inert, Non-contaminated Sludges
Oily, Primary Sludges
Non-Oily, Secondary Sludges
Oily, Secondary Sludges
Totals
Table 2
Kinds and Amounts of Sludge Generated
Sludges, Tons/Day
Filtration of Solids From
Filtration of Activated Intake, API Separator,'and
Sludge Effluent Only Activated Sludge Effluents
200
144
0
562
906
250
47
140
0
437
54
-------
Table 3
Examples of Cationic, Nonionic and Anionic Polyelectrolytes
Structural Tyyc
Anlnei
Functional Croup
B
I
— H— R
I ©
H
~f — cas — CHS—na2 — F
Konionlc
Folycaldc
I
— »—R
I®
D
-(— CE2— CB— K
CKO
I
ehlorlte)
Poljnlcohol
a0
— C — 0
OB
»(— tXg— C— )"
Sttlfonlc
©
— CB— )
i©
Table 4
Dimensions Involved in Colloid Destabflization
A. Some Colloidal Systems Diameter. Angstroms
Color Bodies 50 — 1000
Inert Colloids (Clay. silt.
inorganic salts, etc.) 1.000 — 30.000
Emulsions 2.000 — 100.000
Bacteria 5,000 — 100,000
Algae 50,000 — 8.000,000
B. Cations
Na+ 1.9
Mg-n- 1.3
1
AH++
C. Polyelectrolytes
Potential Tunnel
Chain Length, 100,000 —
15,000.000 M.W.
D. Electrical Double Layer
Range of Expected Values
Expected Typical in Refinery
E. Solvent
H2O
7—11
250,000 — 40.000.000
5—100
30
55
-------
Table 6
The Comparative Effectiveness of Cationic Polyelectrolytes For Charge
Neutralization of Suspended Matter in API Separator Effluents
as Determined by Zeta Potential Titration Curves
Conditions Pub-electrolyte (a.b) mg/1 to Achieve Indicated ZP Endpoint
Rank at Indicated ZP Endpoint
A. pH = 8;
Specific Cond.
= 560. and
Zero Suliides.
B. pH - 9.8:
Specific Cond.
= 680: and
Zero SuKkics.
C. ph =10:
Specific Cond.
= 4.100; and
Zero Sulfides.
-5
.25
.5
.5
.25
1.5
2.5
2.25
.75
2.5
2.25
2.25
2.5
5.25
6.75
.75
2.25
1.75
3
2.5
4.25
5
4
3
2
3
2.5
4
5
5
6.5
7
4
-3
.5
1 ,
1.75
1.75
2
1.5
2.75
2.75
1
3
3
1
3.25
7
7.25
1.25
1.25
1.5
1.25
2.25
2.5
2.25
3.5
3.5
4.5
5.5
4.75
4.5
2.75
3.75
3.5
4.75
6
6
7.25
8.5
(c)
0
.5
1.25
2.5
2.5
2.5
3
3
3.25
3.25
3.75
4
4
4
10
(c)
1.5
1.5
2
2
3
3
3.5
4+
4.5
4.75
6
6
6.5
4
4.5
5
5.75
6.5
7.25
to
to
(c)
+3
1
3.75
3
3.25
3
6
3.25
3.75
6.75
4.5
4.5
5
5
(c)
(c)
2
2
2.75
3.5
3.5
3.5
4.75
4.75
5.25
5.25
6.5
7.25
(c)
5.25
5
6.75
6.75
7.5
9
(c)
(c)
(c)
-5
-3
0
+3
C-31
581
43!
7132
I1KO
1190
2860
2870
863
7134
2640
751
FA
864
860
1180
1190
581
7132
2870
2860
863
431
75!
C-31
2460
7134
1-A
2870
431
751
581
2860
7! 32
2640
863
C-31
(a) C-31 (Dow): 581 (Cyamimid); 431 (Dearborn); 7132 (Nalco): 1180. 1190 (Bet?.): 2860. 2870. 2640 (Calgon): 860. 863. 864
(Hercules); 75! (Marer). and FA (BASF Wyanuotte).
(h) Arranged in order of performance using y.cro ZP as cndpoint.
(cj Indicated ZP not achieved.
1
2
6
5
6
4
8
7
3
8
7
7
8
9
10
1
1
1
1
2
3
2
5
4
7
8
6
5
t
3
2
4
5
5
6
7
4
1
2
4
4
5
3
6
6
2
7
7
7
8
9
10
I
1
3
2
3
4
3
5
5
6
8
7
6
!
3
2
4
5
5
6
7
8
1
2
3
3
3
4
4
5
5
6
7
7
7
8
9
1
1
2
2
3
3
4
5
6
7
8
8
9
1
2
3
4
5
6
7
7
8
1
5
2
3
2
8
4
5
9
6
6
7
7
10
10
1
1
2
3
3
3
4
4
5
5
6
7
8
2
i
3
3
4
5
6
6
6
56
-------
Table 5
Polyelectrolyte Synergisms and Antagonisms to Salinity, pH
and Sulfides in API Separator Effluents
mg/1 of Polyelectrolyte Required to Reach
Zero Zela Potential at Indicated Specific Conductance
-------
figure 1.
Filter Media
Garnet 4.2Sp. Qr
58
-------
Figure 3.
Flocculator Cell
« ' ' ' " " ' • • >•
__ Polyelectrolyte
-^Outlet baffle Peripheral speed 1 to 3 FPS f ^
^
•*
/
3
[ Water depth
I 9'0" — - -•---"> V
I _ •* —
J ^ = ^—
Each cell is 20' -0" wide
t
HB
^3
I-
Influent
chamber
Figure 4.
Flotation Chamber
Channel
effluent
t
o
55' 0"
46'-3" to (£. of skimmer sprockets
Skimmer travel 3-10 FPM
Outlet Secondary dual header
baffle (one box only)
*
. .
1
ib,
Each chamber is 20' 0" wide
Soum __
trough
Flow-directing
r^
\i-
Recycle header
L_
59
U.S EPA Headquarter* l\
Mai! code 34041
1200 Pennsylvania Avenue i
Vvasniogfon, DC 20460
2U2-566-0556
-------
Figure 5.
Recycle System
f & &
I Row meter Air saturation drum
2 1
T 1 Pressur
A fc^H reducin
,—1 J ll V*. ir*.iiN.irNL*l^ilvtt
Recycle
pump
Contacting
zone 7
Flotation chamber &
_T\ i
~ ^ \ i
i
Figure 6.
Optional Refinery Treatment Sequence
TrPatmPnt p'« •>' Inplant Primary Intermediate Secondary
treatment Treatme£ Treatment Treatment Treatment
nhiprtiup* Phenolics. S=. NHa, Free Oil and Emulsified Oil Dissolved
UDjecilVBS RgH F_ Acid S)ud« Suspended Solids Suspended and Organ.cs
Oil Etc., Removal & Removal Colloidal Solids Removal
Water Reuse or Removal
Waste Equalization
Processes Unit s«p*«ofs
1 Steam Stripping
Fuel Gas Stripping
Air Oxidation
Neutralisation 1
Surge Ponds
Chem Coagulation Trickling
API Separators & Ajr F|omion Filter
^ „ . Chem. Coagulation Activated
CPI.PPI Separators & Filtration Sludge
Oxidation
pH Control pond
Immediate Oxygen Aerated
Demand Reduction Lagoon
1 Equalization
of Wastes
Sludges Sludges Sludges
ii i i
e
g valve
1^ ...
— i
1 1
>|
• •* r-— -
J r~
Tertiary
Treatment
Variable
Objectives
1 Chem. Coagulation j
I & Air Flotation ]
Chem. Coagulation 1
& Filtration |
Activated I
Carbon 1
Sludges
! !
-------
1000
800
600
500
400
300
200
100
_, 80
CD 60
^ 50
1 40
30
o3
6 20
10
8
6
5
4
3
Figure 7.
Probability Plots of Oil and Grease Data Before
and After Intermediate Treatment
Raw Waste Load from
API Separator
O
O
Waste Load after
Intermediate Treatment
I i I I I
0.5 1 2 5 10 20 40 60 80 9095 989999.5
% Probability less than Indicated Value
-------
400
300
J 200
s
i
n
Q
g mo
| 90
80
70
60
O
1
£
o
£
50
40
30
20
Figure 8.
Probability Plots of BOD Data Before and After
Intermediate Treatment
O Raw Waste Load from API Separator
• Waste Load after Intermediate Treatment
I I
I I
I I I I
I 1
0.5 1 2 5 10 20 30 40 SO 60 70
% Probability less than Indicated Value
80
90
95 98 99-99.5
2000
Figure 9.
Probability Plots of COD Data Before and After
Intermediate Treatment
400
200
100
ORaw Waste Load API Separator
• Waste Load after Intermediate Treatment
I
I i J i L
I
I
J I
0.5 1
10
20 30 40 50 60 70 80 90 95 98 99 99.5
62
-------
Figure 10.
Character of Mixed Liquor Solids with Sludge
Age
6000
4000
2000
Mixed Liquor Suspended Solids (MG/L)
Flow,20 MGD
Retention Time,12Hrs.
a = 0.4 - b = 0.05
COD Rmvd.,25000 Ibs/D
Inert Solids in, ^^
so MG/L J-**'/ Inert Solids
Total Solids*
Biological Solids'
1 'I I I
6000
4000
2000
» 10 15 20 25 30 35 40 45
Sludge Age (days)
Figure 11.
Effect of Intake Water Solids on Primary Treatment
Oil
Process Water
Contaminants. Lbs/D
A. From Operations
Oi! 650,000
TSS 10.000
B. In Intake Water
TSS, Avg., 25,000
Max., 150,000
I
f *n Mn/1 TCC {*? pjffi | n,/r)
75 MG/L Oil (9,375 Lbs/D)
»*• API *5an->i"it«r
1
'
Sludge, Tons/D
Sludge Intake Clarification
Contaminant % With Without
Oil 4
Solids 4
Water 92
Total 100 47 359
r
^H— Thickening
\
Oil 6
Solids 10
Water 84
Total 100 19 144
r
^— - — Centrifuging
Oil
Solids
Water
Total
10
50
40
100
3.75
29
-------
100 r~
Figure 12.
SVI Of Activated Sludge In Unit Treating
Refmerv Effluent Pretreated by Filtration
300
BOD
Removed
(from Eckenfelder)
.2 .3 .4 .5 .6 .8 1.0
Loading, Lbs. BOD/D/LB MLVSS
~J- J—L
35 25 15 10 5 3
Sludge Age, Days (a = ,6 (3 =• .05)
13.
|AL!H,0)tr
Sequential Formation of Hydrous Aluminum
Oxide Polymers
AI AHOHKHjOlj
OH / OH X
[(H,0).(OHI AI [AI JAIIOHI
AUOHI
HH,OKIOHI,AI
OH \ OH
-OH>
, AI ' I fll .
^OB'
[Fo? solid phase see isometric in following ligute]
1IHZOI3(OHIAI AHOH)2|HZOS,I
OH
-------
Figure 14.
Example of Complex which may exist in
Precipitated Hydrous Aluminum Oxide
Polymers
o
OH2
Figure IS.
Equilibrium Compositions of Solutions in
Contact with A1(OH)3
0
-2
-4
-6
-8
-10
-12
Operating Region for Air
Flotation and Clarifiers"
B • Operating Region for
Direct Filtration
pH
10
12
-------
Figure 16.
Equilibrium Compositions of Solutions in
Contact with Fe(OH)a
0
-2
-6
-8
-10
-12
^
A • Operating Region for Air Flotation and Qanfiers
B • Operating Region for Direct Filtration
C Fe2
E - Fe(OH)2*
F • Fe
G
4 6 pH 8 K)
Figure 17.
Zeta Potential of Colloidal Iron Hydroxide
Solutions Plotted As A Function of pH
12
+40 =1
+30
+20
j= +10
g o
+*
o
CL.
» -10
-------
Figure 18.
Zeta Potential — pH Plot for Aluminum
Hydroxide
+10
-10
I I
I I I
6 7
PH
8
9
10
Figure 19.
Dissociation Of Polyacrylic Acid By NaOH
H H H H
I I I I
•C—C — C—C —
I I I I
C C H C
/\ /\
O O O O
H
H
+ NaOH'
C^ ^^^^ ^ ^^^_ ^ ^^^M
•^^ ,V ^^^ W ^^^ w ^^^
-------
Figure 20.
Condition — Response Flow Schematic for
Chemical Treatment of Waterborne Colloids
Waterborne
Suspended Solids
1 1} Charge
Neutralization
2) Bridging
1) "Sweep Floe"
Mechanism
2) Double Layer
Repression
Inert
Solids
Biological
Cell Material
I
Cold Water
Warm Water
Fresh Water
Brackish Water
Fresh Water
Brackish Water
-------
Figure 21.
Zeta Potential — Cationic Polyelectrolyte
Titration Curves of API Separator Effluent
+10
IZeta Potential, mV
I
-10
-15
-20
API Separator Effluent
A= Betz 1190, pH=~8
B = Betz 1190, pH = 10
C = DowC-31, pH = 8
D= DowC-31, pH = 10
•Specific Conductivity in 570-770 Range.
I I J L I J
12345
Polyelectrolyte, MG/L
8
9
69
-------
Figure 22.
Sensitivity of Dearborn 431 to pH and Phenol
API Separator Effluent
A = As Sampled, pH - 7.8
B = 20 MG/L Phenol Added to "A"
C = pH of "A" Adjusted to 10
D = 20 MG/L Phenol Added to "C"
E - 40 MG/L Phenol Added to "C"
I
5 10
Polyelectroiyte, MG/L
15
70
-------
Figure 23.
Sensitivity of Cyanamid 581 to pH and Phenol
+10
, D&E
API Separator Effluent
A = As Sampled, pH = 7.8
B = pH of "A" Adjusted to 10
C = 20 MG/L Phenol Added to "A"
D = 40 MG/L Phenol Added to "A"
E = 20 MG/L Phenol Added to "B"
5 10
Polyelectrolyte, MG/L
15
20
71
-------
Figure 24.
Titration of Freshwater Aerated Lagoon Effluent
with Cationics Zeta Potential, mV
(_) Amer. Cyan. 573
Tretolite311
-10 —
1 2
Cationic, mg/L
16
72
-------
Figure 25.
Zeta Potential — Cationic Polyelectrolyte
Titration Curves of Coke Fines in Hydraulic
Decoking Waters (pH = 7.6)
+15
Zeta Potential, mV.
Nalco 603
+10
+5
-5
-10
-15
-20
Cyanamid 517
0.32 MG/L
Wt. 2700
0.5 MG/L
1820
1.0 MG/L
1820
5 10
Polyelectrolyte, MG/L
15
20
7.1
-------
Figure 26.
Recommended Hydraulic Loading Rates as a
Function of Water Temperatures and Showing
the Correlation with Water Viscosity
Hydraulic Loading,
m/hr
(gpm/sq ft)
14.7 (6)
12.2(5)
9.8 (4)
7.3(3)
4.9(2)
Kinomatic Viscosity,
sq m/s©c
(sq ft/sec)
Recommended Water
"Temperature My-
draulic Loading
Envelop©
/Temperature-Viscosity
Relationship for Water
I
.07 (.8
.1 (1.10
.13(1.40)
.16(1.70
10(50) 21(70) 32(90}
Temperature, °C (°F)
74
-------
Combined Storm
Overflow Treatment
With SALA-
HGMF® High
Gradient Magnetic
Filters
D. M. Alien and
J. A. Oberteuffer
Sala Magnetics, Inc.
This work reported herein was performed
in pan under USEPA Contract Number
68-03-2218
Summary
Magnetite-seeded high gradient magnetic
filtration is an exciting new technology .
which permits highly effective removal of
BODs, COD, bacteria, color, turbidity
and viruses from many types of polluted
or contaminated waters. The theory and
design of high gradient magnetic filters is
presented and the application to two
groups of polluted waters—combined
storm overflow and raw sewage—is
discussed in detail. The design of a large
system is presented and compared with
conventional methods in terms of
efficiency and capital and operating costs,
showing that this technology offers many
advantages over conventional treatment
methods.
Introduction
High gradient magnetic filtration is a new
physical treatment technique. High
gradient magnetic filters use a matrix filter
bed developed by Kolm1 and a magnetic
separator originally developed by
Marston2 for brightening kaolin clay.
High gradient magnetic filtration has
broad application in both mineral
beneficiation and water treatment.
Numerous metal ores may be beneficiated
and industrial minerals purified by this
technique. Water treatment applications
are numerous and include the filtration of
waters contaminated with nonmagnetic
pollutants as well as the more obvious
filtration of wastewaters containing
magnetically susceptible particulates.
Water treatment applications are
categorized as "direct filtration" and
"indirect filtration" applications. In the
former, the particulates to be removed are
paramagnetic or ferromagnetic and are
directly removed without additional •
chemical treatment. In direct filtration, the
particulates to be removed are
nonmagnetic and must be pretreated to
render then filtrable by magnetic means.
Applications of direct filtration are in the
treatment of steel making waters, metal
finishing waters, boiler feed waters, and
condensate polishing. These applications
are reported in papers by Harland,3
Oberteuffer,4 and Marston5.
Indirect filtration has been applied to
industrial effluents such as those from
paper and pulp mills, to sewage, to
polluted river waters, to storm sewer
overflow and drainage waters, and to the
removal or concentration of viruses. As
the particulates to be removed arc
nonmagnetic, it is necessary to render
them magnetic so that they can be trapped
on the magnetized filaments of the filter
matrix. This is achieved by flocculating
the contaminants around finely divided
magnetic seed (magnetite) by the use of
alum or other flocculating agents.
Polyelectrolytes may be added to
strengthen the floes against physical
disruption, allowing for rapid flow flux
rates, and to increase filtration efficiency
by increasing the size of the agglomerates
and thus the chance that each particle will
be bound to a seed particle, as well.
Theory
Background
The theory and design of high gradient
magnetic filters is presented in review by
Oberteuffer6 and more specifically by
Watson,7 Luborsky* and others.«-i i
However, the fundamentals of operation
and design may be understood from
reviewing Figure 2-1, a schematic of a
cyclic filter which also indicates the forces
interacting to affect filtration effectiveness.
The solenoid electromagnet is surrounded
by an iron return frame which serves to
limit fringe fields and also adds to the
uniformity of the magnetic field in the
working volume of the filter. The canister,
placed within the solenoid electromagnet,
contains a filamentary ferromagnetic
matrix or filter bed material which, when
magnetized, forms the large number of
very high gradients necessary for trapping
and holding fine and even extremely
weakly magnetic contaminants until the
magnetic de-energized. The magnetic
force (Fm) which attract these particles
may be determined from the general
equation
Fm = V M grad H
(2)
where V is the particle volume, M is the
particle magnetization (a function of
magnetic susceptibility of the particle and
magnetic field) and grad H, the magnetic
field gradient. The magnetic field gradients
developed in a high gradient magnetic
filter range from 1000 to 10,000 kG/cm, or
one to two orders of magnitude greater
than those in conventional magnetic
devices. These gradient forces are strong
enough to pull particles to the matrix
filaments against the competing forces of
hydrodynamic drag and gravity which
tend to push particles through the filter.
The hydrodynamic drag force (Fc) is
approximated by the expression
Fc = 3 TT TI b
(2)
where 17 is the fluid viscosity, b is the
particle diameter and v is the fluid velocity.
Magnetite-Seeded Water Treatment
The use of these magnetic filters is
obvious when the particles to be removed
have even low magnetic susceptibilities.
However, their use in removing
nonmagnetic contaminants is less obvious.
In the treatment of storm sewer overflow,
the contaminants to be removed are
essentially nonmagnetic and it is
necessary, therefore, to render them
magnetically susceptible. This is achieved
75
-------
Figure 2-1. '
Schematic High Gradient Magnetic Filter and Interactive Forces
Influencing Filtration Efficiency
RETURN
FRAME
CANISTER
MAGNETIC FORCE
MAGNET
'COIL
FM VMgradH
L magnetic field
gradient
-particle magnetization
—particle volume
COMPETING FORCE
hydrodynamic drag
I Lfluid velocity
^particle diameter
- fluid viscosity
-------
Figure 3-1.
FLOW SHEET
Seeded Water Treatment Pilot Plant
(M UkJU*TtMHT
ri«CCUI.MT I
CLIM •»?!•
CzAr> «T3C=> <—^ C7^T»
i MUCHito
•MTt*
•ALA-M0B8-
by mixing finely ground magnetite into the
fluid to be filtered and then adding
coagulating and flocculating agents such
as alum, iron sulfate or similar
compounds to cause the precipitation or
coagulation of contaminants around the
magnetite seed. The chemical mechanisms
of coagulation have been studied and
described by deLatour12 using artificially
contaminated aqueous systems. Often a
polyelectrolyte, a polar organic
compound, is added to strengthen the
floes and prevent their disruption during
the high velocity filtration process.
SALA-HGMF® Magnetite
Seeded Magnetic Filtration
Pilot Plant
SALA has designed and constructed a 4-
liter/min. continuous pilot plant1-' for
studies of seed, flocculant and
polyelectrolyte concentrations and mixing
times as well as the standard high gradient
magnetic separation parameters of flow
velocity, field strength, matrix loading.
and matrix structure. The pilot plant flow
sheet is shown in Figure 3-1. This
automatically operated pilot plant permits
investigation of these filtration parameters
and also is capable of producing sludge for
preliminary studies on seed recovery and
regeneration.
The applicability of the indirect filtration
process for treating combined sewer
overflow (CSO), storm water, secondary
effluent, and dry weather sewage flow is
now being investigated. The interest in this
study is part of a broad EPA concern with
controlling pollution of municipal waters
and in treating those heavily contaminated
waters feeding into natural water bodies.
SALA is also involved in determining
possible seeded water treatment
applications in the paper mill industry in
cooperation with NCASI.
Tests were carried out in the 4-liter/min.
continuous pilot plant shown in Figure 3-
2. As indicated in the flow sheet (Figure 3-
I), the pilot plant contains a flocculation
train in which all necessary chemical
additions are made, a SALA-HGMF®
magnetic filter and power supply, and a
surge tank and thickener to handle the
backflushed floes that were trapped in the
matrix bed. An automatic controller
sequences the addition and mixing of
chemicals, fluid movement from tank-to-
tank, and the actual filtration and
backwash cycles through the magnet.
Untreated feed is pumped from the
storage tank to the first residence tank
where the feed is mixed with alum. The
magnetite seeding material is added in a
second, flash-mixing tank. Finally a
polyelectrolyte is flash mixed into the
stream to induce bonding among floe
particles, bridging them together into
larger agglomerates. The pH may be
metered and controlled automatically.
77
-------
Figure 3-2.
Four liter per minute Pilot
Plant
78
-------
The treated feed water is next drawn
through the SALA-HGMF® high
gradient magnetic filter by a filter pump
located downstream from the magnet to
avoid floe disruption. Once the filter cycle
is completed, the filter pump and the
magnet are shut off and the matrix is
backflushed with high pressure water. The
back flush water containing the floes
washed from the matrix filter bed enters a
surge tank from which the integrated flow
is fed to the thickener. Periodic chemical
or detergent rinse cycles may also be
necessary to ensure that no residual sludge
buildup occurs on the matrix fibers.
. The pilot plant system is so synchronized
that concentrations of chemicals remain
constant throughout the residence chain.
The system is adjusted easily, moreover, to
permit a wide range of individual chemical
concentrations. As operation is
continuous and automatic, matrix loading
is much more simply tested than in the
bench test system by merely lengthening
the filtration time preceding backflush.
Performance reliability may also be
assessed by taking consecutive samples
over many cycles run under the same set of
operating conditions.
This 4-liter/min. pilot plant has provided
all results presented and discussed herein.
At present, this sytem is being modified,
redesigned and installed into a mobile
trailer unit under a USEPA contract so
that tests may be performed on-site for the
treatment of CSO and other polluted
waters. The trailer unit will have a
capacity of more than 20-liters/min with
two magnetic filters in parallel to simulate
a full-scale system's anticipated ability to
phase in and out with multiple separators
following surges in flow typically found in
storm water applications. A more
sophisticated process flow control system
will be used and a chemical rinse system .
has been added that will prevent sludge
buildups on the matrix fibers. Among the
tests to be performed are those aimed at
determining best flocculation techniques
and system operational limits. Sludge
characteristics will be evaluated for both
disposal and alum/seed reuse possibilities.
Matrix cleaning requirements will be
determined as well as the most effective
matrix fiber configuration and packing
scheme. On-site testing will also attempt
to demonstrate seeded water treatment
adaptability to variability in flow volume-
in a "real" situation as the pilot plant will
be run for the duration of at least one
complete storm water activation.
Pilot Plant Results
Background
Initial operation of the 4-liter/min. pilot
plant was based on an extensive series of
parametric studies conducted at the Sala
Magnetics laboratories. These studies
established ranges of working conditions,
including magnetic field strength, ratio of
input contaminant levels to magnetite seed
and alum concentrations, flow velocities
through the matrix, mixing rates and
residence times to form floes of sufficient
cohesiveness to remain intact when
trapped in the filter.
Pilot Tests
Pilot plant tests were conducted on several
large samples of combined storm overflow
(from Cottage Farm Chlorination and
Detention Facility influent, Cambridge,
Mass.), on raw sewage (from Deer Island
Sewage Treatment Plant influent, Boston,
Mass.), and on secondary effluent from a
final clarifier overflow using the activated
sludge process (Brockton, Mass. Sewage
Treatment Plant). A bench type test wtfs
also performed on a sample of paper mill
effluent (from International Paper Co.,
Ticonderoga, N.Y.). Detailed descriptions
of these tests and the results obtained are
presented below.
Experiments carried out on the pilot plant
were designed to test independently the
following parameters and their
interactions: alum, polyelectrolyte and
magnetite concentrations; pH; magnetic
field strength; flow velocity; residence
time; and matrix loading. Evaluation of
results was based chiefly on the analyses
done on feed and treated samples for
suspended solids, apparent color and
turbidity. Fecal and total coliform counts
were also performed on many of the
samples. Selected treated and untreated
samples were analyzed for COD, BOD5
and trace metals.
Since the character of the waste water used
during this study was very different for
each storm or sewage sample collected, a
wide range of "type" of waste waters was
encountered. CSO, of course, differs from
raw sewage, secondary effluent, or. paper
mill effluent. But even CSO as observed at
Cottage Farm fluctuates tremendously in
solids loading, etc. The time of year, time of
the day, the amount of recent rainfall, the
source of water (rain, snow, snow melt) and
the duration of the storm, all affect the
composition of the CSO. Likewise, raw
sewage effluent varies a great deal.
Table 4-1 compares overall results obtained
with CSO and raw sewage. These data
should approximate the actual
performance to be expected from an on-
stream installation as a result of the wide
range of CSO influent characteristics to be
encountered and because of the practical
reality that optimal running conditions
cannot always be maintained in such an
installation. It should be noted that even
with such experimental fluctuations, the
results were excellent.
Table 4-11 contains summarized data
collected in continuous pilot plant
treatment of CSO over a six hour span.
Included among these data are results from
many tests run at non-optimal operating
conditions as the goal of this study was to
establish optimal parameter values. The
curves shown in the section below are
derived from the treatment of both CSO
and raw sewage. Although chemical
concentrations necessary to achieve
quantitatively equal levels of purity or
separation efficiency may differ from
effluent to effluent, the relationships shown
by these curves have been shown to be
relevant for all waters polluted with
suspended solids and treatable by the
seeded water technique.
Magnetic Parameters and their
Interactions
Magnetite Seed
Magnetite seed is necessary to impart a
magnetic property to the flocculated solids
in the water, thus allowing for floe
retention inside the magnetized matrix of
SALA-HGMF® magnetic filters,
minimum concentration of magnetite
79
-------
Table 4-1
Summary of Percent Removals for All Tests
(Bench and Pilot Plant)
cso
Average of all
Samples Tested
('.'('• Removals)
Range Percent
(•'/ of Tests)
Raw Sewage
Average of all
Samples Tested
(','( Removals)
Range Percent
(^ of Tests)
CSO Collected On
3/17/76
Keed Average
{it of Tests)
Range
Treated CSO
Average
(Si of Tests)
Range
% Reductions
Average
Range
Test Conditions
Magnetic Field:
Flow Velocity:
Alum Cone.:
Polvclectrolvte Colic.:
Suspended Apparent Turbidity Fecal Total
Solids Color Coliform Coliform
95 87 93 99.2 99.3
83-99.1 55-98 74-99 95-99.96 97-99.89
(85)- (78) (77) (8) (10)
91 82 88 99.4
70-93 74-94 81-91 98.7-99.6
(30) (30) (30) (7)
Table 4-11
Continuous Pilot Plant (6 Hour Run)
Suspended Apparent Turbidity Fecal Total
Solids Color (FTD) Coliform Coliform BOD5
(niR/Jl ) (PCH) (cells/100 ml) (cells/ 100 ml) (mg/K.)
460 650 230 3.6 x 10' 6.3 x 10' >79
(3) (3) (3) (4) (4) (2)
400-520 600-800 200-250 2-5 x I07 5,l-7xl07 >75-83
28 85 19 5.3 x I04 1.1 x 10' 6.0
(42) (42) (42) (6) (6) (4)
4.1-185 41-210 8-65 1.5-13 x I04 0.70-2.2 % UP 5.2-7.0
945;, X7f;;. 92% 99.85% 99.83% >92%
60-99% 68-94% 72-97% 99. 64-99. 96% 99.65-99.88% >91-93%
0 I; 0.5: 1.0: 1.9 kG Maunetitc Cone.: 200: 420; 500 mg
56;225m/hr pll: natural 7.3
50: 70: 100: ISO: 200 mg/?. Residence Time: 3; 12 minutes
0:0.1:0.5; 1: 2:2.5 mail
BOD5
>92
>91-93
(4)
COD
(nig/i )
410
(2)
395-125
106
(5)
93-138
74%
66-77%
It
-------
(about 100 mg/1 for example) is needed for
good removal. Above this value additional
magnetite does not improve filtration
efficiency significantly. For concentrations
below about 100 mg/1, however removal
percentages decrease indicating that
insufficient magnetite seed is available as
nuclei around which floes may form.
Matrix Loading and Seed to Solids Ratio
An important parameter to consider when
designing a system is matrix loading (the
ratio of the weight of sludge held on the
matrix in a single cycle to the weight of
matrix material). The size of the magnet
and the related system depends directly on
the matrix volume needed to handle the
expected flow at reasonable flow velocities
with adequate filter cycle length. Thus that
point in a prolonged cycle at which
filtration efficiency drops below acceptable
levels should be determined before a
system is designed. This drop in efficiency
of filtration usually occurs suddenly in the
filter cycle and is called "breakthrough".
The point at which breakthrough occurs in
the filter cycle is dependent upon a number
of factors, including total solids loading,
flow rate, poly electrolyte concentration,
and seed-to-sol ids ratio.
Matrix loading tests performed by varying
seed-to-solids ratios are of value in
determining how much magnetite will be
needed to attain an adequate duty cycle for
practical application. As delay and flush
cycles occupy only about 5 seconds for this
size apparatus, a duty cycle of well over
99% can be achieved at a seed-to-solids
ratio of about 1.5:1. The duty cycle can be
extended even further for high seed-to-
solids ratios. Thus, by varying the amount
of magnetite put into the system, duty cycle
length can be manipulated to a substantial
extent.
When a closer look is taken at the physics
and physical phenomena occurring at the
site of capture (matrix disk surface) the
explanation of the above described
relationship becomes clear.
Were the materials being separated 100%
highly magnetic in nature, the matrix
volume (% 93% void volume) would
continue to load to capacity (clogging)
with very little decrease in effluent quality.
When the particles being trapped by the
magnetic matrix have equal magnetic
susceptibility, the net gradient present on
the surface formed by their layering on the
matrix decreases only very-gradually with
layer thickness, because each new layer of
magnetic material concentrates the
magnetic flux nearly as efficiently as the
matrix fibers. In this case, the formation
of additional layers is limited only by
shearing forces created by the flow and
irregularities in the layers themselves. This
is not the case, however, when
nonmagnetic suspended solids are a major
portion of the solids held to the matrix.
Here, the net force acting on a volume of
solid particles (floe) is much lower because
the net susceptibility of the agglomerate to
the field is due only to the magnetite
portion of the total mass. This has two
effects on the magnetic separation;
decreased filtration efficiency (some floes
may not have had enough magnetite
incorporated within.to be held by the
magnetic gradient) and decreased capacity
of the matrix before breakthrough (the
point at which shear forces equal or
exceed magnetic forces). Thus, the thicker
the solids layer or the less (relatively) the
magnetite concentration in the floes, the
shorter will be the effective filter cycle
length.
Magnetic Field Strength
Like magnetite concentration, magnetic
field strength has a rather straightforward
relationship to filtration efficiency (Figure
4-1). For the high matrix loadings of the
CSO of 3-17-76, a field of at least 0.5 kG
was necessary to hold the seeded floes.
Almost total breakthrough occurred at 0.1
kG, indicating the all-or-none type
response typically obtained with magnetic
seeding techniques. Curves given show
this relationship for suspended solids,-
turbidity and apparent color. Above 0.5
kG there appears to be little change in %
removal with increased magnetic field
strength at this flow rate (224 m/hr).
During surge flows, increased flow
velocities through the SALA-HGMF®
magnetic filter are necessary to
accomodate the large volumes of water if
reserve units are not available. Increased
flow rate means larger shearing forces will
be present within the system, especially
within the matrix where drag forces are
maximal. To overcome this added stress,
the magnetic field must be increased to
hold the magnetic ftocs securely. In such
cases, polyelectrolyte concentration may
also have to be increased so that floe
breakup will not occur. Seed-to-solids
ratios used during surge flows must be
considered because a higher proportion of
magnetite may be required to hold the
floes in the magnetic matrix.
Flow Velocity and Residence Time
Flow velocity and residence time are also
directly related to surge flow. As
mentioned above, flow velocity through
the magnetic matrix will increase with
increasing influent surges (unless reserve
systems are adequate to handle the
increased flow). Also, if the flocculation
mixing tanks are of fixed capacity, a large
surge will result in a faster turnover of
water and therefore a shorter mixing time.
In the present study it was found that a
residence time of only three minutes was
adequate for complete flocculation.
Longer mixing times had no adverse
effects on separation efficiency. Thus, this
parameter is relatively insensitive to
change and need not be controlled
precisely.
Figures 4-2 and 4-3 sho%v results from a
series of tests on CSO to determine the
maximum practical flow flux rate possible
with a particular feed and seed-to-solids
ratio. Other parameters were unchanged.
These curves clearly show that as flow rate
is increased, efficiency is decreased (slope
of curves) and the effective cycle time is
shortened (divergence of curves).
Depending upon the water quality
requirements and economics of the
individual situation, different flow flux
rates may be used. It is apparent from
these curves that as the flow increases
above 200 gpm/ft2, the stronger shear
forces disrupt the magnetically seeded
floes releasing some of the suspended
solids from the magnetite.
m
-------
100-i
95-
90-
ui
DC
85-
80-
Figure 4-1.
Percent removal as a function of Magnetic Field Strength
. n-
"X '
•a
-X
KEY
SUSPENDED SOLIDS
TURBIDITY
APPARENT COLOR
OPERATING CONDITIONS; PILOT PLANT, CSO OF 3/17/76
MAGNETIC FIELD STRENGTH: Oj0.1;0.5;l.Q;1.6jl,9 KG
FLOW VELOCITY: 224 M/HR
MAGNETITE CONC: 500 MG/£
ALUM CONC: 100 MG/£
POLYELECTROLYTE CONC: 2.5 MG/£
RESIDENCE TIME: 3 MIN
PH: NATURAL 7.3
- SAMPLES TAKEN BETWEEN 2 AND 1 MIN IN 4 MIN CYCLES
.1
.5 1.0
MAGNETIC FIELD
1.5
2.0
KG!
Figures 4-2 and 4-3 indicate that a
reasonable flow flux rate for standard
operation (86 gpm/ftj) during test
programs has been chosen. Any lower
flow flux rate would increase treatment
costs without providing a significant
increase in efficiency. Faster flow flux rates
show a fairly rapid decrease in separation
efficiency for this set of conditions. Some
improvement in efficiency of separation
can probably be achieved for the higher
flow flux rates by strengthening the floes.
This involves the use of additional
polyelectrolyte and thus higher
operational cost per gallon, but such costs
might in some instances be offset by a
reduction in capital expenditures for
equipment necessary to handle the given
maximum flow.
Interacting Chemical Parameters
Alum
Alum (A12SO4)3 x 18 H2O) was used as the
principal coagulant in all tests performed.
Both a reagent and commercial grade have
been tried, but careful correlations have
not yet been made. Alum concentration is
a very important parameter for effective
filtration, as without good floes
nonmagnetic particles will not be removed
by the magnet. The key to efficient seeded
high gradient magnetic filtration is. above
all. to coagulate a high percentage of the
solids around a magnetite nucleus. Once
this is achieved, the floes can be
strengthened easily, if necessary, by an
organic polyelectrolyte and removed by
the high gradient magnetic filter.
Optimal alum coagulation is directly
dependent upon a number of interrelated
factors including pH, solids loading and
character, and relative amounts and kinds
of other flocculating agents used
(polyelectrolyte in the present case).
Optimum alum coagulation occurs at a
slightly acid pH while polyelectrolyte
flocculates best at around pH 8 for raw
sewage or CSO. Thus, where tests have
shown that both a coagulant (e.g. alum)
and a floe-strengthening agent (e.g.
polyelectrolyte) are necessary, a
compromise must be made between the
two pH extremes. Since results show that
this compromise is not detrimental to
filtration efficiency, it may in fact prove to
be advantageous, since the pH of CSO
and sewage is usually close to 7 and
amenable to this combination of chemical
flocculants. It is thought that pH
adjustment should be minimal or might be
eliminated altogether when these
flocculahts are used in the proper
proportions. Except for certain tests
devised specifically to determine pH
effects, all tests in this study were made at
natural pH. The excellent results obtained
show that alum-polyelectrolyte
flocculation works well at normally
encountered waste water pH values. Feed
water character also affects optimal alum
concentration. Tests conducted without
alum (and with polyelectrolyte) showed
essentially no separation as floes were very
stringy and did not entrap significant
amounts of the solids in the water. Tests
with paper mill effluent have shown that
true color removal with magnetic '
separation techniques is due entirely to
alum precipitation of that color.
Polyelectrolyte
Filtration efficiency seems to reach an
optimal plateau above a concentration of
-------
Figure 4-2.
Flow Flux Rate versus Suspended Solids
100
90
80
S 70
"8 ««>
-
40
3°
20
10
t 2 minutes
FIGURE 4-2: Flow Flux Rate
versus
Suspended Solida
100
200
300
100
90
80
70
£• 60
3
I 5°
IM
0 40
•H
-------
about 0.5 mg/1 polyelectrolyte. Below this
minimum value, separation drops off
rapidly. This is explained by
hydrodynamic shearing forces, which
break up the floes inside the matrix when
too little polyelectrolyte has been added.
Earlier bench tests showed that at rather
low flow velocities, less than 60 m/h, the
curve for polyelectrolyte was essentially a
straight line (i.e., polyelectrolyte was not
needed at this flow rate as the shearing
forces were below the threshold of floe
disruption). Likewise higher
concentrations of polyelectrolyte may be
necessary at flow rates above the 224 m/h
maximum flow velocity tested.
pH
As discussed above, pH plays an
important role in the filtration
effectiveness achieved using alum and
polyclectrolyte as flocculants. As is
expected, when there is relatively more
polyelectrolyte present, optimal pH for
filtration efficiency moves in the basic
direction (towards pH 8) white, when
relatively more alum is present, optimal
pH moves in the opposite direction
(towards pH 6). Other chemical factors
present in the waste may also play a part,
but the flocculant concentration ratios
seem to be of primary importance in
determining the preferred pH. While good
flocculation was always achieved without
pH adjustment, somewhat better filtration
might have resulted from slight pH
adjustments. However, the acceptable
range of pH for good separation is wide
enough to accomodate normal waste
water fluctuations making continuous pH
monitoring and adjustment unnecessary.
Tests with Secondary Effluent
A secondary effluent taken from the final
clarifier of an activated sludge plant was
treated using SALA's continuous pilot
plant. Pilot plant operating conditions
were based on the results,of previous
testing. Using these operating parameters,
an average reduction of 88% was observed
for suspended solids; an actual reduction
from approximately 40 mg/1 to 4-5 mg/1
was achieved. In a second sample, solids
were reduced from 26 mg/1 to 1.6 mg/1, a
93% reduction. Apparent color and
turbidity were reduced 71% and 85%,
respectively, while fecal coliform levels
were decreased by between 95 and 99%.
Results for this polishing application
would be improved by optimizing
operating parameters.
Paper Mill Effluent
Bench tests were conducted on paper mill
effluent to indicate the effectiveness of
treating this water with a SALA-HGMF®
magnetic nitration seeded water treatment
system. The sample collected was aeration
lagoon effluent from Ticonderoga, New
York. Magnetite, alum and
polyelectrolyte concentrations, pH and
flow rates were varied over appropriate
ranges in a parametric test. Although in
absolute values, the purity of the water
observed both before and after treatment
was substantially lower than for CSO or
raw sewage ( M 5 times the solids loading)
or for secondary effluent ('x. 30 times the
solids loading) relative relationships
between effluent quality and required
chemical concentrations remained about
the same. Optimal pH was near neutral as
for CSO and raw sewage treatment and
the practical limit for flow rate was ^ 86
gpm/ft2. as has been for other polluted
waters. Summarized results for this paper
mill effluent are given in Table 4-1II below.
Analysis
HOD, (ing/1)
Tola! Suspended Solids (mg/1)
Volatile Suspended Solids(mg/1)
True Color
-------
Table 5-1
Magnetic field strength
Flow velocity in matrix
Maximum flow capacity @
100 gpm/ft2 flow velocity
Nominal capacity
Backflush flow rate
Flow rate/ magnetic filter @
maximum capacity, 100 gpm/ft2
flow velocity, and 90% duty cycle
Average influent suspended solids
Magnetite concentration
Alum concentration
Polyelectrotyte concentration
Pumping head for filter pumps
"G" factor for flash mixer
"G" factor for flocculators
Design Values
1.5 kG
245m/hr(IOOgpm/ftJ)
l,18mVsec(25mgd)
0.60^sec (12.5 mgd)
0.95 m'/sec
l2.7mVmin(3.4x IO'gal/MlN)
300 mg/1
300 mg/ a
100 mg/£
I.Omg/£
7.6 m (25 ft)
300/sec
mo/sec
Figure 5-1.
25 MGD SALA-HGMF® Integrated Wet and Dry Weather
Combined Sewer Treatment Facility
FLOCCULATION SYSTEM
PRESCREENIN3
SYSTEM
I 1 [J-l-l
ONiaot | r—, T^
"f rr. irM ' i I /I /I »/l
"~ :~~i / ^' )|'~"'!t;~^~f "I i i
_
j . —- --
W JESS; 7., ti i 1 "- ,,.„„ L;„'.
««"«» ""-1:1 '«m:.
DENATURING
SYSTEM
.
rt-
_]-_
.I»M]
iTCfl^CC
1
1
1
1
DISINFECTION SYSTEM
Costs of a pH adjustment system are
included, although tests performed so far
indicate there is little need for such a
system. Also included are costs for a
conventional chlorination facility. It is
assumed that sludge generated in this
facility would be shipped to incineration,
landfill, or disposal facilities at a larger
plant. Evaluation of sludge characteristics
would indicate the best disposal means.
On-site pilot tests may indicate
modification of the sludge disposal system
and provide the basic information for
design of seed recovery and reuse systems.
Installation costs were based on estimates
from both equipment suppliers and from
the Chemical Engineer's Handbook.14
Operation and Maintenance
Estimated operation and maintenance
costs presented in Table 5-1II are based on
the operating conditions listed in Table 5-
I. The maintenance costs are based on
annual operation at 50% capacity.
Chemical costs for pH control are not
included, since pH adjustment has proven
unnecessary thus far. Similarly, the
operating costs for the final chlorination
step were not included due to the lack of
effluent chlorine demand data.
Operator labor is based on 24-hour-a-day
monitoring of the facility. In addition, an
8 hour shift is included for routine labor
such as lubrication, cleaning bar and
screening equipment, etc. The operator
labor figure is considered conservative as
the plant is designed for automatic
operation.
The electrical costs breakdown (Table 5-
IV) shows that the power consumption by
the electromagnetic coils in SALA-
HGMF® magnetic filters is relatively
small in comparison with the power
consumed by other process equipment.
For operation at a 1.5 kG magnetic field,
the magnetic filters consume only 16% of
the total power required for the system.
An evaluation of the sludge handling costs
is difficult due to the many considerations
necessary. These considerations are
economic, ecological and legal.
85
-------
Table 5-II
25 MGD Integrated Wet and Dry Weather Flow Treatment
Facility*
System
Prescreening
Floe Train and Chemical Feeding and Storage
Thickening and Dewatering Equipment
Backflush System
High Gradient Magnetic Filters (5 @ $165,600 ea.)
Pumps, Filter
Chlorination System
Process Control
Miscellaneous
Physical Plant
Installation Costs not Included in Above
Construction Contingency 10%
Engineering and Administration 109?;
*ENR = 2300
Total Capital Cost
Capital Cost
$ 132,000
293.900
218,250
169.500
828.000
47,300
42,250
108,000
30,000
110,000
229,600
$2.208,800
220.800
$2,429,600
243.000
$2,672,600
86
-------
Table S-1II
Operation and Maintenance Costs
Chemicals
Alum @ $0.l32/kg
100 mg/ £
Magnetite @ $0.022/kg
300 mg/ H (does not include
freight charges)
Polyelectrolyte $3.IO/kg
s/yr
VI000 gal (atl2.5mgd)
Total Chemical Costs
0.050
0.025
0.012
0.087
$ 228,000
114.000
55,000
S 397.000
Operator Labor
32 Man-hours/day @ $IO/hr
0.026
119.000
Maintenance
Mechanical Equipment and physical
plant (3% of equipment costs)
Electrical, Instrumentation and
Piping (2% of equipment costs)
0.012
0.002
55,000
9.000
Total.Labor and
Maintenance
0.040
S 183.000
Electrical
@$0.020/kWh
SAI.A-HGMF® Magnetic filters
1.5 kG require 85 kWh; other
equipment 440 kWh
Total Operations and
Maintenance Costs
0.010 46.000
0.137 S 626,000
87
-------
Table 5-IV
Power Consumption
SAl.A-HGMF*1 Magnetic Filters 1.5 kG
Flash Mixer Units (3)
Flocculators
Air Compressor @ 70% duty cycle
Vacuum Station
Filter Pump
Miscellaneous Consumption
Total
K5 kWh
135
52
35
48
130
.40
525 kWh
SALA-HCMF® magnetic filters use approximately 16% of total power consumed.
For many applications, especially those
involving low solids loading, the
magnetite seed costs are small compared
with other processing costs and thus a
recycle step may not be economically
justified. Ecological consideration such as
improved land-fill qualities resulting from
very dense sludge may favor the seed
discard option. Because of these
considerations, seed discard may not
always be the most expensive choice. Seed
recycle might he advantageous for some
large scale, continuous-flow treatment
facilities. In such cases, additional capital
expenditures might be justified in order to
minimi/e operational costs.
Determination of the characteristics of the
sludges generated by seeded high gradient
magnetic filtration will indicate whether
recycle of seed or seed-discard is most
economical in a specific case.
Comparison of High Gradient
Magnetic Filter Systems to
Conventional Treatment
Methods
The comparison of high gradient magnetic
filtration to conventional processes has
been divided into separate comparisons
for CSO and for sewage treatment
systems. This has been done because most
conventional systems do not perform as
well in treating CSO as they do in treating
dry weather flows. The extreme variations
in flow and rapid change in influent
character make conventional treatment
processes inadequate for CSO treatment.
Performance Advantages
The performance of high gradient
magnetic filters in removal of suspended
solids, COD, BOD5. and coliform bacteria
is excellent in comparison with other
processes. Tables 6-1 and 6-11 compare
high gradient magnetic filtration to the
most commonly used conventional
treatment systems.
High gradient magnetic filter performance
generally exceeds that of all other systems.
Single stage treatment not only attains
removal efficiencies exceeding those of
conventional treatment, but, in addition,
significantly reduces the levels of
phosphates, heavy metals (Table 6-1II)
and virus in waste waters.
High gradient magnetic filtration for CSO
treatment compares favorably with some
tertiary level physical-cheminal
treatments. Tables 6-1 and 6-II also
compare costs of high gradient magnetic
filtration with costs for several alternate
processes under evaluation of CSO and
sewage treatment. The capital cost for a
high gradient magnetic filtration
integrated wet and dry weather flow
facility is $107,000/MGD as compared to
$ 168.000/MOD for a comparable
physical-chemical treatment facility, and
S73.000/ MGD for a dual media filtration
plant using polyelectrolyte.
Operation and maintenance costs are
$0.137/1000 gallons for high gradient
magnetic filtration, $0.187; 1000 gallons
for comparable physical-chemical
treatment, and $0.169/1000 gallons for
dual media filtration with polyelectrolyte.
Table 6-II compares high gradient
magnetic filtration with several
conventional sewage treatment processes.
The cost of high gradient magnetic
filtration also compares favorably with the
cost of conventional raw sewage
treatment.
Other Advantages
The economic value of reserve capacity is
rather difficult to estimate. Systems
employing high gradient magnetic filters
can increase capacity simultaneously in
two ways. One method is to increase flow
velocity while simultaneously increasing
magnetic field, and the other is to bring
additional magnetic filters into operation.
Since conventional systems do not have
this flexibility, this consideration in
particular makes high gradient magnetic
filtration, where an array of filters are used
in parallel, an attractive process for CSO
treatment where handling of surge flow
has been a problem.
Other advantages may occur due to the
nature of the sludge produced. Because of
the magnetite seed in the floes, the sludge
generated is much denser (approximately
20-30% solids by weight) than that
produced in conventional waste water
treatment systems. The mass of the
relatively heavy magnetite acts to reduce
the sludge water content.
Cost savings may be realized from reduced
thickener and dewatering equipment size
and better land-fill properties.
88
-------
Table 6-1
Comparison of CSO Treatment Processes*
Treatment Process
% Removal
Total
Suspended Solids BOIV
Operation and
Coliform Capital Costs** Maintenance**
Bacteria 25 MGI> S/1000 Ral
92-98
90-98
99-99.99
2.672,600 S (!.!??
High Gradient Magnetic
Filtration
Physical Chemical
Treatment***
Dual Media Fill ration
with Polyclectrolyle
Rotating Biological
Contactor
Contact Stnbili/ation
High Hate Trickling
Filtration
High Rate Trickling
Filtration
Dissolved Air Flotation
with Fine Screening
Micros! rainers
* Reference ft
** Since operating casts from "Urban Stormwater Management and Technology; An Assessment" are expressed in 1974, January S. these
figures have been adjusted bv +209?! to effect a fair comparison to high gradient magnetic Ttltraiion. Capital costs are adjusted to K\K -
2300.
*** Albany. N. Y, Pilot Plans (Ref. 3)
99
36-92
70
92
65
65
56-77
70
94 99
66-79
54
83 83
65 83
65 83
41-57 99
50
4.190.000
1.817.000
862.500
2.251.100
2.251.100
2.275.600
968.300
325.500
0.1X7
0.169
0.053
0.058
0.058
0.073
0.079
0.0023
Tank- 6-11
Comparison Between High Gradient Magnetic Filtration and ,Other Sewage Treatment
Processes*
Treatment Process
%Rcmoval Operation and
Suspended Cnliform Capital Costs** Maintenance**
Solids COD Bacteria 25 MOD S/IOIMIgal
High Gradient Magnetic
Filtration
Chemical Clarification
Activated Sludge
Treatment
Physical Chemical
Treatment
88-95
60
55-95
99
60-75
70-80
80
90-98
99+
5 2.672.600 S 0.137
1.522.500
11.132.000
4.190.000
O.OH5
0.187
* Fair. C.M.. (iever, J.C. andOkun, O.A.. Water Purification and Water Treatment and Disposal. Vol.2. John Wiler&Sons. New York.
1968. pp. 21-72
** Since operating costs from "Urban Stormwater Management ami Technology: An Assessment" arc expressed in 1974. Januarys, these
figures have been adjustedhv *2ffK: to effect a fair comparison to high gradient magnetic filtration. Capital costs arc adjusted to I~\K -
'2300.
-------
Table 6-III
Heavy Metals Removal
Metal
Cd
Cr
Cu
Pb
Hg
Ni
Zn
Average
% Reduction
(Range)
> 43%
> 41%
53%
(0-67%)
> 71%
(0-67%)
84%
A further advantage of high gradient
magnetic filtration system is its
comparatively small land use when
building a new system and its
configurationa! flexibility when updating
an existing facility. Small land use is
achievable because of the very high rates
of filtration. The 25 MOD system
described above would, for example.
occupy only 0.35 acres (0.14 hectare). In
regions where land is scarce and real estate
values are high, this could result in
considerable savings. In any situation the
ecological advantages of a small system
are numerous.
In existing plants, the magnetic filters can
be arranged in such a way that space is
used efficiently, perhaps preventing the
need for additional land acquisition and
facility construction.
Conclusions
The preceding sections have detailed the
theory of high gradient magnetic filtration
and discussed its application to one group
of filtration problems—the treatment of
combined storm overflow and of sewage
using magnetite seeding.
Besides the excellent removal of total and
volatile suspended solids, fecal and total
coliform bacteria, true and apparent
color, turbidity and BOD5, the SALA
seeded water treatment technique has been
shown to remove nearly 100% of all
viruses,15 and substantial amounts of
COD (% 75%), trace metals16 such as zinc,
chromium, lead and copper, phosphates,l7
and algae.1"
The seeded water filtration process is a
simple one involving well-known
flocculation procedures and a simple
highly effective easily maintained filter to
remove contaminants from the effluent
stream. No pH alteration seems necessary,
although incoming solids loadings and
flow rates should be monitored
continuously to maintain optimal
chemical concentrations. The system
handles high flow rates and requires
comparatively little land use for this
reason. These factors, combined with
excellent performance and competitive
costs make magnetite-seeded high gradient
magnetic filtration an exciting new
technology for the treatment of many types
of polluted or contaminated waters.
References
Kolm, H.H., Magnetic Device, U.S.
Patent No. 3,567,026, assigned to M.I.T.,
1971.
Marston, P.G., et al, Magnetic Separator
and Magnetic Separation Method, U.S.
Patent No. 3,627,678, assigned to ME A,
now Sala Magnetics, Inc., 1971.
Harland, J.R., et al, Pilot Scale High
Gradient Magnetic Filtration of Steel Mill
Wastewater, IEEE Trans. Mag., Vol.
MAG-IO(6), 1976.
Harland, J.R., el al, High Gradient
Magnetic Filters for Polishing Steam
Condensates and Other Thermal Power
Plant Waters, Paper IWC-76-19.
presented at the 37th Annual Meeting,
International Water Conference,
Pittsburgh, PA, 1976.
Oberteuffer, J.A., et al. High Gradient
Magnetic Filtration of Steel Mill Process
and Waste Water, IEEE Trans. Mag., Vol.
MAG-11(5), 1975.
Marston, P.G., et al. The Application of
High Gradient Magnetic Separation to the
Treatment of Steel Industry Waste
Waters, 2nd Intern. Congress on Waste
Water and Wastes, Stockholm, Sweden,
1975.
Oberteuffer, J.A., High Gradient
Magnetic Separation: A review of
principles, devices and applications, IEEE
Trans. Mag., MAG-10(2), 223. 1974.
Watson, J.H.P., Magnetic Filtration, J,
Appl. Phys., 44(9), 1973.
Luborsky, F.E., and B.J. Drummond,
High Gradient Magnetic Separation:
Theory versus Experiment, IEEE Trans.
Mag., MAG-ll(6), 1696-1700, 1975.
Cowen, Carl, et al. High Gradient
Magnetic Field Particle Capture on a
Single Wire, IEEE Trans. Mag., MAG-
11(5), 1600-1602, 1975.
Himmelblau, D., Observation and
Modeling of Parametric Trapping in a
Magnetic Field, M.I.T., M.S. thesis, June,
1973.
Dolby, G., and J.A. Finch, Capture of
Mineral Particles in a High Gradient
Magnetic Field, presented at First Intern.
Powder and Bulk Solids Handling and
Processing Conf., May, 1976.
deLatour, C., Magnetic Fields in Aqueous
Systems, M.I.T., Ph.D. thesis, August.
1974
Harland, J.R., et al, High Gradient
Magnetic Filtration Pilot Plant, Paper
presented at the 80th Nat. Mtg. AIChE,
Boston, September, 1975.
Chemical Engineer's Handbook, Perry,
J.G. (Ed.), McGraw-Hill Co., N.Y., 1969.
Bitton, G. and Mitchell, R., Phosphate
Removal by Magnetic Filtration, Water
Research, 8. 1974.
90
-------
Warren, J. and A. L. Neal, Concentration
and Purification of Viruses from
Paniculate Magnetic Iron Oxide— Virus
Complexes, U.S. Patent 3,470,676,
September 30, 1969.
Okuda, T., Sugano, I., and T. Tsuji,
Removal of Heavy Metals from
Wastewater by Ferrite Co~precipilation.
Filtration and Separation,
September/October, 1975.
Bitlon, G. and Mitchell, R., Phosphate
Removal by Magnetic Filtration, Water
Research,*, 1974.
Bitton, C., et al. Removal of Algae from
Florida iMkes by Magnetic Filtration,
Applied Microbiology. JO305. 1976.
Use of Floatation for
Waste Water
Treatment
Mjasnikov I.N.
Gandurina L.V.
Butseva L.N.
The problem of water basins protection
against contamination with industrial and
municipal wastes has acquired primary
importance. The existing methods and
traditional schemes of waste water
treatment do not always provide the
required extent of contaminant removal.
In this connection investigations are
conducted on elaborating new methods
and facilities for treatment of various
kinds of waste water and improving the
existing ones.
It is floatation that is considered to be one
of the perspective methods used to remove
suspended matter from water, including
finely dispersed impurities. This method of
treatment, based on separation of
contaminants from liquid by means of
floating up air or gas bubbles, is widely
used in industry and municipal services.
Floatation as a method of contaminant
removal is used at the following stages of
waste liquid treatment:
• local treatment of wastes at enterprises
and utilization of valuable products;
• water preparation for biological
treatment;
• separation of sludge mixtures in
aeration tanks, consolidation of excessive
sludge and waste water sediments;
• final treatment of biochemically treated
waste water.
It is quite reasonable to use floatation
methods for treatment of waste water
contaminated with oil wastes, oil-
processing products, fats, resins, latexes,
organic synthesis products, surface-active
agents and dyestuffs. As compared to
waste water treatment by way of settling,
the floatation methods are more efficient
especially in cases when the rate of
particles settling is less than O.I m/sec, or
when it is necessary to remove suspended
matter from liquid containing
considerable amount of dissolved gases
etc.
The process of waste water treatment by
floatation method is 4-6 times faster than
that of settling, the efficiency of
contaminant removal being the same.
The possibility of removing finely
dispersed suspended matter from liquid by
way of floatation allows the latter to
compete with sand filters.
In case the hardly oxidized substances that
retard biochemical processes or cause
frothing are present in liquid the
floatation may be one of the main
methods of decontaminating these
substances.
The characteristic feature of floatation.
The method is the possibility of using the
latter for treatment of large as well as
small amounts of waste liquid. Floatation
plants are distinguished by compactness
which makes it possible to locate them
right at places of waste water generation.
The operation of floatation plants can be
automatized in a simple way which in turn
ensures secure operation of these plants
and affords to gain for high efficiency of
treatment.
The following kinds of floatation are being
used at present for treatment of natural
and waste water as well as for treatment of
sediments and activated sludges:
compression (pressure), impeller,
vacuum, biological chemical, electric and
ionic floatation, froth separation.
In all these cases the removal of
contaminants from waste water as well as
consolidation of sediments takes place
owing to availability of gas bubbles on the
surface which is the area where the
substances to be removed are
concentrated. At the same time various
-------
kinds of floatation differ by a number of
features: method of gas bubble
generation, kinds of components to be
removed (dissolved, colloidal substances,
ions, molecules) design of frother cells etc.
Structural pecularities of installations
depend on the field of their use.
Combination of processes in one
installation is widely used at present for
waste water treatment, for instance,
floatation with settling, floatation with
filtration, etc. It should be noted that
perspectivcness of floatation increases
with the development of reagent
production, creation of combined schemes
for physicochemical as well as biochemical
treatment, necessity of surface-active
agent removal, etc.
The efficiency of floatation for waste water
and sediment treatment has been proved
by considerable experience of using these
installations in industry and municipal
services. Such floatation methods as
compression and impeller one are used in
practice for a long time. This made it
possible to design serial (standard)
installations ensuring high efficiency of
contaminant removal.
Compression floatation method is widely
used for industrial and municipal waste
water treatment. This method is used to
treat oil-containing waste water generated
at enterprises or oil-producing, oil-
processing and petrochemical industries,
at storage terminals, ports, automobile
plants as well as for treatment of effluents
of cellulose-paper, chemical, power
engineering, machine building and food
industries, in treatment plants of
municipal services.
Sufficient experience has been gained in
these branches of national economy on
maintenance of treatment plants, new
frother cells have been designed with up to
80-98% efficiency of water cleaning from
suspended matter and oil products.
Considerable reduction of COD (chemical
oxygen demand) and BOD (biological
oxygen demand) values of liquid has been
achieved.
Several modifications of floatation plant
systems are used for waste water
treatment. These plants include:
liquid supply pump, ejector or compressor
for air delivery into system, saturator
(capacity for air dissolving under pressure
of 3-5 atm), frother cell. The plants can
have special inlet devices for reagents and
as a rule are provided with automatic
control and regulation system.
Schematic diagrams of some of these
systems are shown in fig. I.
Depending on the amount of liquid
supplied to saturator these diagrams can
be classified as follows:
• straight-flow, when the total amount of
treated liquid is supplied to saturator
(diagram a);
• rccirculation, when 20% of clarified
liquid and more than that is supplied to
saturator (diagram 5 );
• partially-straight flow, when about 30-
70% of non-treated liquid is supplied to
saturator and the rest of it is sent directly
to frother cell.
These diagrams can be completed by a
number of facilities, such as mixing and
flocculation chamber in case of
reagentizing, devices for air and water-air
mix dispersion etc.
The field of using these systems is
determined by physicochemical properties
of treated liquid and its components, by
requirements to quality of treatment and
by local conditions.
Straight-flow system is used in industry to
remove the major part of contaminants
from waste water when it is necessary to
inject considerable amount of air and gain
high efficiency of treatment.
Straight-flow system ensures sticking of
generated micro-bubbles of air right to
contaminant particles in waste water and
thus provides for a high efficiency of
treatment.
Floatation with recirculation is used to
treat waste water containing contaminants
that can be easily dispersed for instance, in
the process of water pumping, during air
dissolving or in case of liquid treatment in
saturator.
These are flocculant suspended matter,
activated sludge and some metal
hydroxides that can be easily subjected
to such changes. In this case treated waste
water is used as working fluid to involve
air in floatation process. Such water is also
used in floatation plant.
Mixing of contaminated liquid with air-
water mixture takes place in a piping
located in front of frother cell or right in
the installation.
The process of generating and fixing
microbubbles on contaminant particles in
this system will have some pecularities as
compared with straight-flow floatation. In
case of recirculation the microbubbles are
formed, as a rule,' in a clarified liquid layer
up to the point where it is mixed with
waste water as it is subjected to a lesser
pressure. That is why the possible contact
between generated bubbles and
contaminant particles depends on mixing
conditions and affects floatation
efficiency. This system is used in case of
activated sludge treatment, final
purification of biologically treated waste
water, when making use of reagents and
also when it is necessary to reduce power
consumption, etc.
Partially straight-flow floatation system is
used when it is necessary to remove the
bulk quantity of contaminants from waste
water in case of local treatment of
industrial waste water and gaining
valuable components before biological
treatment plants and final treatment.
Plant operation according to this scheme
provides high efficiency of treatment when
using reagents. As compared to the
previous schemes this one provides
considerable reduction of power
consumption.
To maintain the process of water
saturation with air or gas, provision is
92
-------
made in pressure floatation plants for
saturators working at a pressure of 3-5
atm. The period of liquid staying in
saturators is from 0.5 to 2 minutes.
Depending on the quality of treated water
use is made of saturators (absorbers) of
bubbling, dispersion and cap type.
Schematic diagrams showing some of
these saturators are given in fig. 2.
The required period of liquid staying in
the saturators is determined by calculation
and depends on phase contact surface,
mass transfer coefficient and impellent
force of the process.
Bubbling type saturators (a) are used to
treat liquid containing considerable
amount of suspended matter, part of it
having adhesive properties. Phase contact
surface is formed by gas or air bubbles
which bubble the liquid.
Dispersion type saturators can be
successfully employed when using well
dissolved gases for waste water saturation.
Various spraying devices are used in these
installations.
Combined type, saturators (B, ) provide
better saturation of water with air
especially in case of using capping.
Bubbling type saturators are most
frequently used for waste water treatment
and sediment processing.
Such apparata are also used for the
majority of floatation plants designed for
oil products removal.
More efficient are saturators equipped
with capping. Plates of different shape,
grates, rings, nets etc. can be used as
capping. Arrangement of these elements in
the saturator affords to increase greatly
phase contact surface.
Investigations carried out by VODGEO in
industrial conditions have shown that the
use of saturator with a capping for
treatment of oil-containing waste water
allows to reduce the period of liquid
staying in the apparata and to increase the
amount of dissolved air as compared with
conventional installations having
bubbling layer of liquid.
Some researchers have suggested at
present the schemes of presure floatation
provided with saturators have centrifugal
nozzles, Venturi tubes and other devices
which makes it possible to reduce the
period of liquid staying in the apparata up
to 30-60 sec.
Pressure floatation plants are available in
our domestic industry which do not
contain saturators, and the process of .air
solution takes place in pressure piping
which supplies waste water to frother cell.
Separation of contaminants from waste
water by means of floating up gas and air
bubbles is performed in a frother cell.
Horizontal, radial and vertical type cells
provided with devices for froth skimming
and bottom sediment removal are used for
waste water treatment by pressure
floatation method. Considering that waste
water includes hydrophilic particles and
other admixtures which cannot be
separated with the help of floating up
bubbles. Cells are also designed for liquid
settling (flotators-sett'.ers) or include
filtering bed (filter-flotator).
Key diagram of some frother cells of more
than 300 m3/hr capacity are shown in fig.
3.
Frother cells (fig. 3a) are used to treat
waste water containing substances with
hydrophobic properties as well as oils, oil
products, fats and surface-active matter.
These installations are provided with a
device for continuous froth removal.
Bottom deposit formed in a small quantity
is periodically removed. The operating
capacity of the plant reaches m3/hr. The
efficiency of treatment, for instance, oil-
containing waste water (straight-flow
system) is 50-60% and in case of reagent
use up to 70-85%.
These installations were employed to treat
industrial waste water before biological
installations or for additional water
cleaning from mechanical impurities.
' Flotators-settlers (fig. 3b) are more widely
used for treatment of industrial waste
water and in some cases for final treatment
of biologically cleaned waste water.
These installations up to 1200 m3/hr
capacity and over are equipped with froth
skimming device.
The sediment formed in the installation is
periodically removed thus confining the
field of using these installations for
treatment of waste water as the latter
mainly contains easily flotable matter.
These installations operate with a higher
treatment efficiency as compared to above
mentioned frother cells. They are used in
some systems for treatment oil-containing
waste water as well as in chemical
industry.
More efficient are flotators-settlers
provided with scraper device for
continuous sediment removal.
New installation have been designed with
the capacity from 300 up to 900mVhr. Use
of these installations for treatment of oil-
containing waste water provides oil
products removal to their residual
concentration of 20-30 mg/1. Use of
flocculants in these conditions makes it
possible to remove oil products to residual
concentration of 10-15 mg/l.
Water treatment efficiency by means of
flotation depends also on hydrodynamic
conditions in frother cell. Availability of
stagnant zones in installations and created
convective flows of liquid cause intensive
movement of water in various directions
and make floatation process worse.
To improve hydrodynamic conditions of
the process provision is made for new
designs of floatation installations
containing various types of headpieces.
Frother cell with a cylindrical headpiece
(fig 3, ) made of polymers or metal have
been suggested for treatment and final
purification of biologically treated waste
water and compaction of activated
sludges.
-------
Installations of 250 m3/hr capacity have
been designed with scraper devices for
froth skimming and sediment removal.
These installations were also used in
systems of additional reagent treatment of
waste water.
Horizontal frother cells with units of flat
parallel plates (fig. 3 -2) and 1-1.5 m
depth of liquid working layer are also used
to treat industrial waste water. A
proportional water distribution device is
provided at installation inlet to maintain
adequate water distribution. Such
apparatus of up to 80 m3/ hr capacity
makes it possible to remove oil products
to residual concentration of 5-10 mg/1 in
case of reagent usage.
Wide use is made of horizontal frother
cells {fig. 3 ) to treat small quantities of
industrial waste water.
Installations of 5-20 m3/hr capacity
comprising saturator equipped with water
delivery pump and air injection device as
well as rectangular frother cell are
produced by our industry.
This brief list of described frother cells
does not cover the variety of possible
structures.
As it has been already mentioned use is
made in waste water treatment systems for
combined structures filters-flotators. Such
apparatus devised in food industry is
shown in fig. 4.
Saturated water and air mix is supplied to
floatation zone located in the upper part
of the apparatus. The filtration zone is
formed of sand bed of about 1 m thick.
This apparatus operating on waste water
of one of the mills of butter and fat
industry provides for high water treatment
efficiency at 3 m-Vm2 hr load and 25-50
mg/1 flocculant doze, including that for
fat matter from 50 mg/1 to residual
concentration of 1 mg/1. This filter-
flotator is recommended to treat waste
water of varnish and paint industries
where it was shown by investigations
the efficiency of cleaning from suspended
matter reaches 96%.
Other structures of filters-flotators have
been also designed which are
recommended to treat various kinds of
industrial waste water.
The authors consider that the apparatus
shown in fig. 4 can be used to treat
natural and waste water.
Floatation method of activated sludge
compaction is maintained in some of
presented structures, in apparatus shown
in fig. 4 . This apparatus of up to 100
m-Vhr capacity operates according to
recirculation scheme. The period of liquid
staying in it is about 40-60 minutes, the
height of working zone is up to 3 m, the
moisture content of compacted deposit is
about 95%, the content of suspended
matter in treated water is 30-50 mg/1.
Along with compression floatation it is
recommended to use frother separation
for final treatment of biologically purified
waste water. The scheme of apparatus
designed to clean waste water of textile
and tannery plants from synthetic surface
active matter is shown in fig. 5. Air of
about 12 m3/m2 hr intensity is delivered
into this apparatus through porous
material.
With aeration period of about one hour
the efficiency of surface-active matter
removal is up to 60%, the reduction of
COD and BOD of waste water is up to 25-
40%. Some other designs of apparata for
surface-active matter removal by bubbling
method have been also suggested (fig. 5b).
One of the floatation methods that special
attention is being paid to is
electrofloatation. Electrofloatation gives
the possibility of getting microbubbles of
gas which are uniformly distributed in the
treated liquid.
Use of electrofloatation for treatment of
oil-containing waste water makes it
possible to reduce the content of oil
products up to 1-10 mg/ 1 with original
concentration being 200 mg/1. Power
consumption in this case is 0.28-0.55 Kw
per m3 of treated water.
Electrofloatation is also efficient for final
treatment of biochemically purified waste
water of oil refineries.
With current density of 17 ma/cm2 and
treatment period of 4 minutes the content
of suspended matter in waste water is
reduced from 15 mg/1 to 2 mg/1, the
COD value is reduced from 175 mg/1 to
14 mg/1 and the content of ester matter
from II mg/1 to 4 mg/1.
Electrofloatation plants of up to 10-15
m3/hr capacity are usually provided with
one cell and for greater efficiency—with
two cells. The apparata is usually
comprised of electrode compartment and
settling part. Waste liquid enters the
calming basin separated from electrode
compartment with a grate. Floating up of
particles take place in the settling zone.
The floated sludge is removed by scraper
devices.
Use is made of either insoluble electrodes
(in case of small content of contaminants)
or of soluble electrodes (in case of
aggregate-resistant contaminants with
high concentration). The gap between the
electrodes is usually 15-30 mm.
As it has been already mentioned it is not
possible to present here the whole variety
of designed flotator structures with their
schemes, but it should be pointed out that
this method of treatment will be widely
used for waste water treatment.
For the sake of further development of
methods and structures of flotation plants
for waste water purification and sediment
treatment attention should be paid to the
following aspects:
• increasing operating efficiency of
devices for dissolving air and gas in liquid;
• providing optimum conditions of
flotation separation of gas and liquid
medium;
• studying hydrodynamic conditions of
frother cells (velocity of flow and bubble
movement as well as generation and
movement conditions of aeration
floccules;
94
-------
• studying physical and chemical
properties of water-air medium;
• creating reagents for flotation treatment
of waste water and compaction of
sediments and activated sludges;
•studying flotation process accompanied
by sorption, adsorption, oxidation, etc:
• designing combined installations
(coagulator-flotator, filter-flotator,
flotato r-degasator, flotator-settler);
* studying the peculiarities of flotation
treatment of waste water and final
treatment of biologically purified water
with activated sludge compaction;
• designing devices for froth removal, its
processing and gaining valuable
components;
* designing automatized flotation plants;
• generalizing the operational results of
flotation plants and their technical and
economic evaluation.
References
AIR
Q
Q
AIE
A.I. Matsnev. Waste water treatment by
flotation, Kiev, Publ. house "Budivelnik",
1976.
V.S. Nadysev. Waste water treatment at
plants of oil and fat industry, M, Food
industry, 1976.
B. M. Matov. Flotation in food industry,
M. Food industry, 1976.
A.M. Koganovsky, N.A. Climenko.
Physical and chemical methods of
cleaning industrial waste water from
surface active matter, Publ. house
"Naukova dumka", Kiev, 1974.
N.A. Lukinyh and others. Methods of
final waste water treatment, M. Stroiizdat,
1974.
Q
AIR
Figure 1.
Principal Scheme of Flotation Unit
A. Concurrent type
B. Recycle type
C. Partly concurrent type
1. Pump for influent
2. Saturator
3. Flotation cell
95
-------
» t f
r
4
JLJ.
11
Figure 1.
Principal Scheme of Saturators
A. Barbotage type
B. Hollow spray apparatus
C. Combined type
D. Nozzle type
-------
S-l)
Kigurc 3.
Principal Scheme of Flotation Cells
A. Flotation unit
B. Combined flotation-sedimentation tank
I. Flotation-sedimentation tank with mechanisms for foam and sludge disposal
-------
r*fe
e
T n *
u
•••rs
r
^
a-
Figure 3.1.
Principal Scheme of Flotation Cells
1. Sectioned flotation cell
"2. Flotation cell with no/.7.1e of I/annular type and 2/rectangular type
98
-------
Figure 4.
Principle Scheme of Flotation-filtration Unit (a) and Sludge
Thickener (b)
99
-------
COi-rO.'.iToAl'E
i
LL
WAvllR
AIR
EFFLUENT
AIR
Figure 5.
Scheme of Foam Separating
Cells
I. Barbotage one-sectioned cell
2. Barbotage two-sectioned cell
Performance Tests
on Full-Scale
Tertiary Granular
Media Filters
Joseph A. FitzPatrick
Charles L. Swanson
Department of Civil
Engineering
Northwestern University
Introduction
Filtration is considered to be the most
important tertiary process in the
implementation of the Federal Water
Pollution Control Act Amendments of
1972. Based on tertiary filtration needs
reported by Lykins and Smith{ 1), over
1500 plants will be required to achieve
current water quality standards.
Approximately 94 percent of plants will be
smaller than 5 mgd and 80 percent will be
smaller than 1 mgd. An equivalent
number of plants will be required to meet
anticipated future standards by 1985.
This paper summarizes the results of a
full-scale evaluation of 8 small-scale (0.8-
2.5 mgd) tertiary granular media filtration
plants in the Chicago metropolitan area.
The objective of this portion of the study
was to characterize filtration plant
performance as a function of design and
operational variables.
Data Collection
All analyses were made on flow-
proportioned composite samples of
secondary effluent (filter influent) and
filter effluent. Most sampling periods were
24 hours, with occasional periods of 4, 8,
and 16 hours. Samples were analyzed for
turbidity, suspended solids, and 5-day
BOD on a routine basis. Flow rate was
obtained from plant meters and run
lengths were determined from backwash
frequency counters. Headless at the end of
runs was determined if filters were
equipped with altitude gauges.
Filtration Plant Design
A summary of filtration plant design data
is shown in Table 1. Chemical treatment
of secondary effluent prior to filtration,
except for chlorination, was riot practiced
at any of the plants. Five plants have filter
configurations shown in Figure 1 with a
backwash storage tank located above the
sand and anthracite filter media, and flow-
splitter box for proportioning flow to
individual filters (Smith & Loveless,
General Filter, and Eimco designs).
Adjustments were made to weir lengths in
the splitter boxes for the Addison South,
Lake Zurich, Lisle, and Marionbrook
Plants to vary the flow from
approximately 0.5 to 2 times average for 3
filters. The Environmental Elements
(Hardinge) and Hydro-Clear designs
utilize single medium sand filters, and the
Neptune Microfloc design utilizes tri-
media of anthracite, garnet, and silica
sand.
Geometric mean media sizes (dso) varied
from 1.4 to 2.7 mm for anthracite and 0.6
to 1.4 mm for sand. Design depths ranged
from 12 to 24 in. for anthracite and 10 to
24 in. for sand. Actual depths when
sampling was conducted were
considerably less for many plants because
of media loss during backwashing since
the plant was constructed or.the media last
replaced.
Average Performance Data
A summary of average flow and
suspended solids data is shown in Table 2.
Coefficients of variation for influent
suspended solids ranged from .33 to .60
while for effluent suspended solids the
range was .25 to .80, indicating a wider
range in effluent than influent
characteristics during the sampling
periods. The average ratio effluent to
influent suspended solids (C/G>) varied
from 0.17 to 0.53 for the 8 plants.
Plots oft/Co vs. average flow rate (Q)
and average influent solids loading (CoQ)
are shown in Figures 2A and 2B,
respectively, for plants where flow was
varied to individual filters. Although the
100
-------
independent variables Co and Q varied
widely as did C/G>, the average
performance (C/ G>) and effluent
suspended solids (C) appear linear with
flow rate. For individual data it is not
possible for both relationships to be linear
unless Co is constant and Co is constant
for each data point for Lisle and Addison
South plants. For those plants data was
obtained for low, medium and high
filtration rate simultaneously. The more
nonlinear dependence apparent for the
Lake Zurich (LZ)and Marionbrook (MB)
plants may result from the fact that the
three filters (low, medium and high rate)
were not always in operation at parallel
times and thus Co is not constant for each.
Figures 3A and 3B show average effluent
suspended solids (C) as a function of flow
and solids loading for the Addison South
and Lisle plants. Data for the Lake Zurich
and Marionbrook Plants are excluded
from this plot because Co was not constant
as already mentioned. As with average
data for C/G>, a linear dependence of C
with Q and C with G>Q is apparent.
Linear correlation coefficients for all pairs
of data in each of the data sets is shown on
Figures 2 and 3. The Addison plant (AV)
shows highest (r=0.70) correlation of C
with CoQ. Remaining data sets on Figure
3 shown marginal correlation (r < 0.5)
and some data sets in Figure 2 shown no
correlation individually, e.g.. Lake Zurich
and Lisle on Figure 2B.
Particle properties, particularly suspended
particle size and suspended particle size
distribution are expected to play
important roles in granular bed tertiary
filtration (2, 5, 6). In this phase of the
work, no routine quantitative
measurements were made to determine
particle size distributions. However, it was
observed that the suspended solids for the
Lake Zurich and Marionbrook plants
secondary effluents were more finely
divided compared to the other plants. This
is not surprising since both plants have
ponds ahead of tertiary filtration. The
secondary processes for the Lake Zurich
and Addison South plant consist of
activated sludge and trickling filters in
parallel. It would be expected that
suspended solids in trickling filter plants
effluents would be less flocculent
compared to activated sludge plant
effluents. The role of suspension
characteristics are currently under
investigation in this study on filtration
plant performance.
Although interpretation is not possible on
theoretical grounds one can linearly
extrapolate the lines of best fit in Figures 2
and 3 back to very small flow rates or
solids loadings. Residual values of C/G>
and C at zero flow rates or solids loadings
suggests that a portion of the secondary
effluent suspended solids cannot be
removed by granular media filtration even
at very low flow rates without alteration of
suspension characteristics, e.g. particle
size and charge by chemical addition.
Graphs of average C/Co as functions of
average tdmperature, total media depth,
and equivalent media diameter (de) are
shown in Figure 4,5A and 5B respectively,
where de is the equivalent media diameter
(dso) that would give the same clean bed
head loss as the sand and anthracite media
sizes at the same total depth. Values of d?
were calculated form Kozeny's equation
for flow through porous media. For plants
where flow was varied, only data for the
medium flow rates are shown.
If one excludes the single medium sand
filters at the DesPlaines River (DP) and
Romeoville (RO) plants, Figure 4 shows a
strong dependence of filter performance
on temperature (broken line) over the
relatively narrow temperature range of 55
to 67° F. It is believed that this is an
artifact resulting from "chance" in the
selection of plants and times of the year
for sampling. Alternatively, if one realizes
that the Lake Zurich and Addison North
plwnts had the finest and coarsest particle
size respectively in the filter influent, then
their average C/Co ordinate is fixed
basically by this factor. Eliminating those
plants and realizing that other plants had
comparable filter influent particle size,
C/Co is independent of water
temperature.
If suspension properties are constant, an
inverse dependence of performance on
media depth would be expected if in
depth, rather than surface filtration, is the
dominant particle collection mode.
Surface cake and/or sieving filtration are
believed to be the dominant modes for the
sand filters at the DesPlaines River and
Romeoville plants. If suspension
properties are constant, higher removal
efficiency should be obtained for finer
media whether straining or depth
filtration is operative. However, if the bulk
of the removal is of coarse solids, little
dependence of removal efficiency on
media size or depth will be apparent, and
residual suspended solids will be a
relatively constant fraction of influent
solids. With the information that the Lake
Zurich and Addison North filter influent
suspensions were coarsest and finest
respectively for any plant tested then
Figures 5A and 5B show that average
removal of the remaining plants (with
comparable influent suspensions) is
independent of media depth and
equivalent media diameter, the latter of
which is not independent of the former.
Thus, some of the filters are overdesigned
with respect to media depth and size to
remove the bulk of the coarse suspended
solids.
Bivariate Correlation
Coefficients
Bivariate correlation coefficients for plant
performance and selected operating
parameters are shown in Table 3.
Although performance showed a strong
correlation with flow and solids loading
for average data where flow to individual
filters was varied, much poorer
correlations were obtained for individual
data.
Correlation coefficients for C to Co varied
from 0.18 to 0.71. Figures 6A and 6B show
C as a function of Co for the Addison
South (Filter No. 2) and the Romeoville
plants, respectively.
If solid breakthrough occurs before the
headloss criterion is exceeded or before
filters are backwashed manually or by
timers, it would be expected that
performance would decrease with
increasing run length. Except for the Lake
Zurich Plant, very poor correlations were
101
-------
obtained for C and C/G> with run length
(RL). This is partly explained by the fact
that the Lake Zurich filter influent had
very fine suspended solids and fully
developed depth filtration took place.
Thus the active deposition zone (clogging
front) advances through the filter with
time giving rise to a breakthrough
phenomena. Furthermore, run length was
varied over a wider range for this plant
compared to other plants where filters
were manually backwashed at the start of
a run.
Except for plants where flow to individual
filters was varied, correlation coefficients
for C and C/Co to Q were less than 0.3.
This suggests that, although flow is an
important design parameter, day to day
changes are too small compared to other
variables, e.g., suspension properties, to
result in measureable changes in filter
performance. For plants where flow was
varied to individual filters, correlations for
flow increased significantly for most data
sets.
Multiple Linear Correlations
If we hypothesize that C or C/Co is
linearly related to variables such as solids
• loading (C«Q), flow rate (Q), run length
(RL) etc. this may be tested by performing
multiple linear regression on the data sets.
Regression coefficients for the assumed
relationship were generated from a data
set. Correlation coefficients for selected
multiple linear regression equations are
shown in Table 4. In the notation used
here, the functional form of C = f(Q, CoQ)
isC = a + bQ + cG>Q, where'a, b, and c are
regression coefficients (constants) for data
from individual plants. The highest
multiple correlationicoefficients obtained
were 0.70 for C and 0.60 for C/ Co).
As shown in Table 3 and Table 4,
correlation coefficients for all of the
Marionbrook data were less than 0.6.
These data were obtained during the
period from January to May, 1976, and
during August, 1976. Data have been
analyzed separately for both dry and wet
weather periods. Prior to melting snow
and heavy rains in mid-February, headloss
buildup was less than 2 ft/day and in-
depth filtration was believed to be the
dominant removal mechanism. For this
period, C calculated from the multiple
regression equation C = f(G>, CoQ) is
plotted versus measured C in Figure 7. A
greatly improved multiple correlation
coefficient of 0.84 was obtained for this
plot.
Data for the Lake Zurich plant have been
analyzed separately excluding the data for
the medium flow rate for a period when
flow was not varied. The regression
equation used was C/Co = f (Q, CoQ,
QRL). Calculated C for this equation is
compared to measured C in Figure 8
(multiple r = 0.73). Similar analyses over
shorter time frames have been made for
other data sets.
Discussion
Performance of tertiary filters is the result
of complex interactions between a number
of variables. The relatively poor
correlations-between filter performance
and operating parameters obtained in this
study can be attributed in part to
variations in the characteristics of
secondary effluent suspended solids. Such
changes can'occur on a seasonal or daily
basis, or over a shorter time frame during
filter runs. Causes include changes in
secondary process biota, changes in
process operating modes, process upsets,
and flow surges causing temporary loss of
floe from secondary sedimentation tanks.
Parameters-of importance'that were not •
measured routinely in this study include
particle size characteristics, floe strength,
and particle surface charge properties.
Even so, it was found that if filtration
process data are analyzed over shorter
time frames where suspended solids
character is more constant, improved
correlations can be obtained.
Daily variations in 24-hour composite
samples obtained during May to August,
1975, for influent and effluent suspended
solids for the Addisori South iplant (Filter
No. 2) are shown in Figure 9. Periods of
high filter influent suspended solids for
this plant are normally caused by high
sludge blanket levels in activated sludge
process clarifiers, resulting in part from
bulking sludge arid mechanical-problems
with return sludge-pumps. Figure 9 shows
that process upsets had only slight effect
on filter effluent quality. Large activated
sludge floe is readily removed at the
surface and in the upper layers of
anthracite filter media by straining
mechanisms. Except for periods of
secondary process upsets, filter influent
and effluent suspended'solids appear to
vary on a random basis. The overall
bivariate correlation coefficient for C with
C is a rather low 0.38 (Figure 6A).
In pilot-scale studies on filtration of
activated sludge process effluents,
Tchobanoglous (2) reported that media
grain size had a significant effect on
removal efficiency for single medium sand
and anthracite filters at depths from 18 to
30 in. Influent suspended solids levels for
his studies varied from approximately 14
to 24 mgjjL . Small removals of suspended
solids occurred below depths of 16 to 20
in. Baumann and Huang (3) reported that
variations in performance over a wide
range fo media sizes were not significant
for dual media pilot-scale filtration of
trickling filter process effluents: Trickling
filter effluent suspended solids varied from
15 to 50 mg/j£ . It was reported that'no
significant removal occurred below depths
of 12 in. for both the sand'and anthracite
media. In pilot-scale studies using
unstratified bed filters, Dahab and Young
(4) concluded that removal efficiency is
not reduced greatly by increasing the
effective size (dio) of media from 1 to 2 mm
at flow rates of 2 to 4 gpm/sf. Influent
suspended solids levels fortheir studies
ranged from approximately 20 to 50
mg/Ji .
In the studies reported by Tchobanoglous
(2), influent solids levels were:rnuch lower
than the averages of 28 to 62 mg/S,
observed for the plants in this full-scale
investigation. At lower solids levels, it
would be expected that mean particle size
would be smaller and media grain size
would have a greater influence on removal
efficiency than in our study.
No firm conclusions can be reached on the
102
-------
effect of media grain size and depth on
filter performance for the full-scale plants
in this study. The differences in
characteristics of suspensions are most
likely the major reason for the variations
in performance between plants. However,
based on reported pilot studies at
equivalent and tower influent suspended
solids levels (3,4) it is probable that media
size and depth do not have a large effect
,on mass removal efficiency over the ranges
encountered in this study. However,
media size in particular may have an effect
on clogging rate.
Conclusions
1 For plants where flow to individual
filters was varied, average clarification
performance decreased linearly with
average flow rate.
2 Correlation coefficients between filter
solids removal efficiency and operating
parameters such as flow, solids loading
and run length are poor for data obtained
over long time frames. Improved
correlations were obtained for data
analyzed over shorter time periods which
minimized the effect of seasonal and other
variations in secondary plant effluent
characteristics.
3 Run length did not have a significant
effect on plant performance, except for
one plant where backwash frequency was
varied over a wide range.
4 Differences in characteristics of
secondary effluent suspended solids rather
than media grain size and depth are most
likely the major reason for variations in
clarification efficiency between plants.
REFERENCES
Lykins, B.W. and Smith, J.M., Interim
Report on the Impact of Public Law 92-
500 on Municipal Pollution Control
Technology, Environmental Protection
Agency, EPA-600/2-76-018. January
1976.
Tchobanogtous, George, "Filtration
Techniques in Tertiary Treatment," J,
Water Pollution Control Federation, Vol.
42, No. 4, April 1970.
Baununn, E.R. and Huang, J.Y.C.,
Granular Filters for Tertiary Wastewater
Treatment, J. Water Pollution Control
Federation, Vol. 46, No. 8, August 1974.
Dahab, M.F. and Young, J.C.,
Unsiratifted Bed Filtration of Wastewater,
J. Environmental Engineering Division,
American Society of Civil Engineers,
February 1977.
Feuerstein, D.L., In-Depth Filtration for
Wastewater Treatment, Contract No. 14-
12-852, by Engineering-Science, Inc., for
U.S. Environmental Protection Agency,
Feb., 1976.
Ghosh, M.M., Jordan, T.A. and Porter,
R.L., Physicochemical Approach to
Water and Wastewater Filtration, J,
Environmental Div., ASCE101, No. EE1,
p 71-86, 1975.
Notations and Abbreviations
C Filter effluent suspended solids—
mg/l
Co Filter influent (secondary effluent)
suspended solids—mg/1
C/O> Rates of filter effluent to niter influent
suspended solids
Q Superficial filtration rate—gpm/sf
CoQ Filter solids loading—Ib/sf/day
RL Average filter run length—hrs
QRL Flow times run length
d» Geometric mean media
diameter—mm
d* Equivalent media diameter—mm
r Bivariate or multiple correlation
coefficient
Filtration Plant Codes
AN Addison North
AS Addison South (Filter No. 2)
AV Addison South (Filter Nos. 4, 5. 6, 7)—
variable flow studies
BA Harrington
DP Des Plaines River
LZ Lake Zurich
LI Lisle
MB Marionbrook
RO Romeoville
List of Figures
1. Filter Design
1. Average C/Co vs. Average Flow and Solids
Loading
3. Average C vs. Average Flow and Solids
Loading
4. Average C/Co vs. Average Temperature
5. Average C/Co vs. Media Depth and
Equivalent Media Diameter
6. C vs. Co. Addison South (#2) and
Romeoville Plants
7. Calculated vs. Measured C, Marionbrook
Plant
8. Calculated vs. Measured C. Lake Zurich
Plant
9. Daily Changes in C and Co, Addison South
Plant (#2)
List of Tables
1. Filtration Plant Design Data
2. Summary of Plant Performance Data
3. Bivariate Correlation Coefficients for Plant
Performance Data
4. Multiple Linear Correlation Coefficients for
Plant Performance Data
103
-------
FILTER
INFLUENT-
BACKWASH
DISCHARGE-
INFLUENT
SPLITTER BOX
EFFLUENT WEIR
BACKWASH
STORAGE TANK
FILTER MEDIA
UNDERDRAINS
FILTER
EFFLUENT
Figure 1.
Filter Design
104
-------
0,6
0,4
0,2
0
0
0,6
0,4
0,2
0
234
Q-GPM/SF
5
FIG, 2B
01 -
0,01
0,24
0,11
j
0123
CQQ-LB/SF/D
Figure 2.
Ave. C/Co vs. Ave. Flow and Solids Loading
4
105
-------
30
20
10
0
FIG, 3A
• AV 0,52
• LI 0,49
0
30
1234
Q - GPM/SF
20
10
0
i i i
FIG, 3R
• AV 0,70
• LI 0,54
I I I
0
CQQ - LB/SF/D
Figure 3.
Ave. C vs. Ave. Flow and Solids Loading
106
-------
0,5
0,3
0,2
0,1
0
RO
MB
LZ
\
\
\
BA
DP
AV
\ "LI
\
\
«AN\
\
50
55
60
Figure 4.
Ave. C/Co vs. Ave. Temperature
65
70
107
-------
0,6
0.4
o
0,2
I I I I T
• LZ
* BA
RO ••;_! "MB
DP
FIG, 5 A
i-AS
AV
AN
0
0
0,6
10 20 30
TOTAL MEDIA DEPTH - INCHES
50
i 4
It I I
FIG, 5B
LZ
0,4
BA
'AV
0,2
L I
DP
AN
0
I I I l I 1 I
0
0,5 1,0- 1,5
EQUIV, MEDIA DIA, - M M
2,0 2,5
Figure 5.
Ave. C/Govs. Media Depth and Equivalent Media Diameter
ION
-------
20
15
7 10
0
ADD, SOUTH PUWT
R = 0,38
FIG, 6 A
°
\*
•*••
• *
• «•
* •• ••
* *
0 20 tft 60 80 100
C - MG/L
20
0
15
10
5
0
ROMEOVILLE PLANT
= 0,65 •
FIG, 6 B
0 20
CQ - MG/L
60 80 100
Figure 6.
C vs. Co, Addison South (^2) and Romeoville Plants
109
-------
35
I I
LAKE ZURICH PLANT
R = 0,73
30
25
20
15
10
0
1
0
10 15
MEASURED C - MG/L
20
Calculated vs Measured C, Marionbrook Plant
25
no
-------
40
30
20
10
0
i I
MARION BROOK PLANT
1/18/76 - 2/U/76
R-0,84
I 1 1 '
I 1
0
10 20 30
fEASURED C - MG/L
40
Figure 8.
Calculated vs. Measured C, Lake Zurich Plant
111
-------
SUSPENDED SOLIDS-MG/L
S
02
M
1
M^>
3
ft
3
ft
o.
o.
S"
o
3
C/l
o
»
CD
ro
CD
112
-------
Table I
Filtration Plant Design Data
Plant
Addison North
Addison South*
#1, 2. 3
#4, 5, 6, 7
Harrington
Des Plaines River
Lake Zurich
Lisle
Marion Brook
Romeoville
Vendor1
S& L
S& L
S& L
Nep. Micro.2
Envir. Elem.
Gen. Fill.
Eimco
Eimco
Hydro Clear
Design
Flow
Rpm/sf
4.3
2.2
2.2
3"
1.0
2.5
2.6
2.5
2.5
mm
2.7
1.7
2.7
1.76
—
1.4
1.7
1,6
—
Anthracite
Depth
Design
24
24
24
16
—
12
24
24
—
Inches
Actual'
21
9
15
10
— ,
1
9-164
4-134
—
Sand
Depth Inches
mm
1.4
0.67
1.4
0.63
0.96
0.7!
0.58
0.66
0.72
Design
24
24
24
14
11
12
12
12
10
Actual1
24
22
23
3-
11
i!
3
10-1 14
10
Sec.
- Plant
Design
Flow-
MGD
2.0
2.1
2.1
2.0
2.0
0.8
2.5
2.5
2.0
i Abbrev.: S% L— Smith % Loveless, Nep. Micro. — Neptune Microfloc, Envir. Elem. — Environmental Elements. Gen. Fill. — General
Filter Co.
2 Tri-media, 30 in. total, anthracite, garnet and silica sand
' Estimated depth when sampling was conducted
4 Variation between individual filters
s Filters #1. 2. 3 were constructed in 1969 and #4. 5, 6, 7 in 1973
Table 2
Summary of Plant Performance Data
Plant
Addison North
Addison S. (#2)'
Addison S. (#4-#7)'
Low Flow
Medium Flow
High Flow
Harrington
Des Plaines River
Lake Zurich1 2
Low Flow
Medium Flow
High Flow
Lisle
Low Flow
Medium Flow
High Flow
Marionbrook
Low Flow
Medium Flow
High Flow
Romeoville
No.
of
Data
73
70
61
51
64
58
60
24
48
22
32
29
35
!05
100
55
36
Klow-gmp/sf
Mean'
3.8
2.1
0.87
1.8
3.4
3.6
0.76
1.3
2.3
3.9
1.4
3.0
5.3
1.2
2.4
4.4
. 1.5
CVJ
0.16
0.22
0.21
0.28
0.25
0.23
0.39
—
—
—
0.26
0.26
0.17
0.36
0.31
0.36
0.26
Suspended Solids-nig/ jj
Influent (Co)
Mean
44
38
52
49
51
37
31
28
31
29
62
59
61
38
34
32
31
. CV4
0.43
0.45
0.42
0.43
0.43
0.32
0.61
0.39
0.35
0.38
0.61
0.6!
0.61
0.53
0.50
0.59
0.55
Effluent (C)
Mean
6.6
9.0
II
14
20
12 '
5.0
9.5
14
15
9.0
12
18.0
7.1
8.4
7.2
7.4
CV
0.56
0.43
0,65
0.57
0.60
0.34
0.60
0.31
0.31
0.25
0.51
0.62
0.72
0.76
0.79
0.64
0.50
C/Co
Mean
0.17
0.26
0.24
0.30
0.40
0.32
0.2!
0.34
0.48
0.53
0.18
0.24
0.34
0.20
0.26
0.27
0.26
CV
0.65
0.46
0.63
0.53
0.38
0.34
0.76
0.26
0.33
0.21
0.50
0.54
0.59
0.65
0.65
0.67
0.38
1 Secondary process — activated sludge and trickling filters in parallel. Other plants activated sludge.
2 Filters operate at constant flow with on-off influent pump cycles.
1 Arthimetic mean
4 Coefficient of Variation = standard deviation/arithmetic mean
113
-------
Table 3
Bivariate Correlation Coefficients for Plant Performance Data
Plant
Addison North
Addison South #2
Addison S. (#4-7)—All Data
Low Flow
Medium Flow
High Flow
Harrington
Des Plaines River
Lake Zurich—All Data
Low Flow
Medium Flow
High Flow
Lisle—All Data
Low Flow
Medium Flow
High Flow
Marionbrook — All Data
Low Flow
Medium Flow
High Flow
Romeoville
C to
Co
.26
.38
.48
.27
.5!
.71
.55
.37
.53
.47
.51
.72
.32
.20
.41
.42
.40
.43
.50
.18
.66
Cto
Q
.06
.02
.52
.18
.21
.53
.17
.33
.39
—
.49
.14
.55
.24
.26
.47
.53
.50
-.02
Cto
CoQ
.25
.31
.70
.29
.48
.74
.44
.45
.62
.47
.51
.72
.54
.20
.44
.45
.49
.72
.71
.50
.68
Cto
RL
-.05
-.27
-.10
-.19
.07
-.14
.01
-.26
0
.41
.12
.09
.04
.33
-.18
.29
.10
.05
.14
.14
-.12
C/Co
toQ
-.07
.04
.40
-.09
-.14
.25
-.27
-.10
.41
—
—
—
.44
-.03
.15
.09
.41
.53
.58
.49
.25
C/Co
to CoQ
-.41
-.47
.24
-.25
-.21
.10
-.35
-.30
-.0!
-.26
-.55
-.54
.01
-.47
-.28
-.26
.11
.11
.16
,L2
.26
C/Co
to RL
-.04
.22
-.01
-.05
-.17
-.11
-.15
-.11
.18
.49
.45
.37
.02
.25
-.23
.34
-.04
-.10
-.03
.12
.15
Table 4
Multiple Linear Correlation Coefficients for Plant Performance Data
Plant
Addison North
Addison South (#2)
Addison S. (#4-7)—All Data
Harrington
Des Plaines River
Lake Zurich—All Data
Lisle —All Data
Marionbrook—All Data
Romeoville
f(Q. CoQ)
.26
.36
.70
.54
.45
.63
.57
.50
.70
C =
f(Q, CoQ, RL)
.26
.39
.70
.56
.48
.68
.60
.52
.70
f(Q, RL)
.06
.24
.41
.37
.15
.56
.46
.42
.36
C/Co =
f(Q, QRL)
.04
.26
.40
.38
.11
.60
.46
.44
.40
114
-------
Sewage Treatment in
Mining,
Metallurgical and
Oil-chemical
Industries
N.V. Pisanko
("Ukrvodokanalproekt"
Institute)
• The "Ukrvodokanalproekt" Institute is
engaged in designing the water supply,
sewer systems and hydrotechnical
structures of industrial enterprises and
populated points, giving great attention to
the most burning problem of the
present—the efficient use of water
resources, protection of the nature and
environment from pollution.
The largest part of the work is being
carried out for enterprises of the mining,
metallurgical and chemical industries.
These objects are characterized by high
water flow, the use of the recycle water
supply systems and treated sewage reuse.
The fundamental trends of the Institute
work on the development of modern water
supply systems and sewarage are the
following:
• introduction of the waterless
technological processes;
• decrease of the water consumption and
water derivation standards;
• creation of the closed water supply
cycles with the treated sewage reuse;
• introduction of the local systems of
sewage treatment, with the extraction of
the valuable components;
• thorough sewage treatment with
application of the desalting, ozonating
and sorbtioning plants;
• application of new chemicals for sewage
and nature waters treatment and use of
process wastes as chemicals too;
•treatment of a sewage sluge, its
utilisation and disposal.
A major portion of the Institute work is
concerned with the experimental
projection and construction of the new
designs, structures and flow diagrams of
water and sewage treatment.
Work is being carried out in close contact
with the scientific research institutes, the
plans of which for the nearest years and
perspective are composed without
participation.
The standard projects of sewage and water
supply structures and plants for the serial
use in various fields of industry are
developed. Many principally new
decisions of the Institute are covered by
authors' certificates and have found wide
use in designing and structural practice.
The peculiar attention is being focused on
the development of programs and
application of electronic computers, in
order to choose the optimum variants in
terms of techniques and economics, of the
individual structures, as well as
technological processes as a whole.
All the design decisions, we accept,
undergo the stage of the technical and
economical estimation.
The experience of operation of structures
and systems, built in the latest years with
new technical decisions, confirms their
efficiency and economy, when compared
to the traditional ones.
As examples, we may use some technical
decisions, developed by our Institute and
introduced into construction.
So, in mining industry, to get the
ferriferous concentrate from the natural
ore, the wet-separation method of its
concentration is widely used, following
which the large quantity of sewage in the
form of a pulp-rock, strongly diluted with
water, is left.
Weight ratio in the pulp of the solid phase
and water T:£f, ranges from 1:15, to 1:25.
The pulp is supplied to the tailstorage for
settling, whereupon the clarified water
goes back to the works.
Such a flow diagram of the pulp supply to
the tailsstorage, which is located, as a rule,
at a considerable distance from the
oreconcentration enterprises, requires
great electric energy consumption, the
construction of large pulp-pumping
stations with heavy equipment, as well as
laying the pipelines and water mains of
large length.
Our Institute developed and proposed the
flow diagram of sewage clarification after
oreconcentration, according to which, the
pulp, before supplying to the tailstorage,
is thickened up to T: JJ ratio from 1:3 to
1:1.
The clarified water goes right away to the
works, and the thickened pulp is
transported to the tailstorage.
In this case, along with the reduction of
the energy consumption, the pipelines
diameters, the number of the pumping and
energetic equipment, as well as the
expenses for the whole system operation,
are cut down.
The flow diagram of pulp thickening is
presented in Fig. I.
From the oreconcentration plant the
hydromixture by a trough is supplied to
the distribution chamber 2. At the same
chamber from the room of the building for
chemicals 3, the flocculant solution—
polyacrylamide, promoting coagulation
and settling of the fine-dispersed
suspension, arrives by the chemical
pipeline 10.
From the distribution chamber 2 the
hydromixture goes to the thickeners 4,
and in case of emergency regimes, it passes
by the trough 5 to the special tank.
The thickened hydromixture (sludge)
enters the receiving chamber of the
115
-------
pulp-pumping station 7 by the pipelines 6,
laid in tunnels, and then through pipelines
is pumped over by the dredge pumps to
the tai[storage.
The clarified water from thickeners 4 by
trough 9 goes to the pond of recycle water
supply 11, from where it is delivered to the
oreconcentration plant. The clarified
water is to be in the pond not less than 24
hours, in order to desintegrate the residual
polyacrylamide.
At one of the groups of oreconcentration
enterprises with the pulp consumption not
more than 20 cu.m./sec., the scientific-
research work was conducted, and in the
industrial conditions the new type of a
vertical thickener with the shelf-type
medium was tested. (Fig. 2).
Compared to the radial thickener, the
operation of which is based on the
principle of the gravitational suspension
settling over the whole depth of the
settling tank, the pulp clarification in the
verticle thickener with shelf-type medium
takes place in a thin layer of water.
The basic design features of the verticle
thickener lie in the special device of the
initial pulp supply to the thickener, its
uniform distribution over the thickener's
area, intensification of the water
clarification process, uniform water
collection from the whole area of the
thickener and the thickened pulp removal.
The initial pulp is delivered to the
thickener by the pipeline 1 to the central
part 2 (Fig. 2).
In order to intensify the pulp clarification
process, the thickener is provided with a
thin-layer divider—shelf type medium, in
the cells of which the laminar regime of
the flow stream is maintained.
The thin-layer divider provides the solids
particles settling in laminar stream of
small depth, thus encouraging the rise of
the thickener capacity and improvement
of the clarified water quality. The
hydraulic pulp load on the thickener
makes up 40-45 cu.m./sq. m./h;
suspended solids content with this load at
the outlet is not more than 500 mg/1,
when the initial concentration is 30000
mg/1. In this case the polyacrylamide dose
by 100% product equals 1 mg/1.
The clarified water collection is achieved
by the system of radial troughs 4, the
design of which causes the uniform flow
distribution and assures the thickener
operation with the same specific load over
the whole area.
•The clarified water is derived by trough 5,
and the thickened pulp under the
hydrostatic pressure is removed by the
pipeline 6.
Side by side with the highly efficient
thickeners in the building for chemicals,
the vortex disperser for preparing the
polyacrylamide solution and large-
capacious exchange packages are used.
The vortex disperser, being the central
chain in the unit of polyacrylamide
solution, comprises two technological
operations:
— polyacrylamide dispersion by a
turbulent fluid flow with the formation of
the polydispersed suspension of a
polymere in the solvent;
—. obtaining the homogenous solution
(without solvent water heating) with 0.05-
0.1% concentration.
The vortex disperser operates in a
continuous technological cycle, is easy to
adjust and reliable in exploitation.
The apparatus design is presented in
Fig. 3.
The high-viscous matter, passing through
the piece of pipe 2 to the feeding chamber
1 goes through the perforated partition 2
to the dispersing chamber 3, where it is
pulverized by the vortex flow, supplied
under the pressure to the chamber 3
through the tangential piece of pipe 5. In
chamber 3 the polymer dispersion in the
liquid takes place. Further the flocculant
dissolution process occurs in the vortex
flow. The ready polyacrylamide solution is
derived through the piece of pipe 4. The
proposed disperser design provides the
uniform supply of polyacrylamide over
the whole area of the partition, that is
practically impossible to be obtained in
known constructions with the vertical
arrangement of perforated surfaces.
The new flow diagram of processing the
large quantities of polyacrylamide in
vortex disperser decreases considerably
the capital investments for the
construction of the building for chemicals,
reduces the manual labour, makes the '
operation simple and excludes from the
technology the hot water use for a
flocculant dissolution.
An automatic control of the pulp
thickening process and chemicals supply is
carried out by the control of
polyacrylamide solution concentration
according to the turbidity of the clarified
water. Such a control system makes it
possible to save up to 10% of
polyacrylamide and improve the process
of pulp clarification and thickening.
One of the new trends in the field of pulp
thickening is the research on
electromagnetic flocculation of the pulp
prior to its clarification in thickeners.
Preliminary data demonstrate the high
efficiency of this method, which allows to '
reduce sharply or exclude at all the use of
polyacrylamide from the technology.
Introduction of new decisions of pulp
thickening at one of the objects made it
possible to reduce capital investments by
3400 thousand roubles and operation
expenditues—by 1830 thousand roubles.
According to the project of our Institute at
one of the metallurgical plants the
construction of facilities for cinder
extraction from sewage of the rolling mill
with the use of magnetic flocculation is
being carried out (fig. 4).
Magnetic flocculator consists of cassetes
with the magnetized ferritobarium
washers.
116
-------
In the flow diagram the magnetic
flocculator 3 is installed prior to the
secondary settling tanks 5 in supplying
trough I.
Passing through the flocculator, the
cinder, entering the composition of
sewage, is magnetized and enlarged. Flows
with the enlarged cinder floccules are
delivered from the supplying trough to the
secondary settling tanks 5 through
distribution system 4.
Secondary horizontal settling tanks are
equipped with the scraper mechanism 6
for rabling the cinder to the sump 7 and
collecting the oil, floating to the surface,
into the trough 8.
The clarified sewer liquid from the
secondary settling tanks is diverted by
trough 9 for the reuse.
Cinder from the secondary settling tanks
arrives at the sump 7, from where it is
delivered by the gantry crane, supplied
with the grab, to the bunker 11 for
dewatering. Dewatered cinder by the
railway transport is sent to the mill of
clodding and briquetting and then it is
utilized.
Oil, collected from the water surface of the
secondary settling.tanks is diverted to the
oil pumping stafion 13 by trough 8, and
after the preclarification it goes hack to
the rolling shop.
Secondary horizontal settling tanks are
composed of sections in 11 x 33 m size.
The load per section of the settling tank is
650 cu.m/h.
Application of the large-sized secondary
settling tanks with the dispersed water
inlet and preliminary magnetic
flocculation allowed us to shorten the
duration of the settling and to increase the
load for settling tanks to 30%, keeping the
clarification effect to 50-100 mg/1 of
suspended solids, with the initial cinder
content of 400-500 mg/1.
For sewage treatment from one of the
enterprises of oil-chemical industry the
diagram is developed, which provides the
preseparation of oil products and oil mud
with the initial concentration of 2r5 g/1,
and settling on oil separators and settling
ponds, followed by more thorough
treatment on quartz filters. Filters are
demonstrated in Fig. 5.
Filter's media consists of a sand layer with
2-0, 75 mm coarseness, 1,2 m height, some
layers of gravel with 2-32mm coarseness
and 1,0 m total height.
Filtration takes place from bottom to up
with 5,0 m/h rate. Due to the large
capacity of the filter, the filter-cycle
duration constitutes 2 days. The oil
products contents in sewage after
filtration decreases from 70-80 to 25-30
mg/1. Exploitation demonstrated filter's
operation efficiency.
Filters regeneration is produced by treated
sewage in two stages, applying cold and
hot water with air blowing.
Due to the complicated conditions of
regeneration, the Institute at present is
carrying out the work on introducing the
more capacious and easily regenerated
media.
In order to improve the treatment
efficiency and have the possibilities for the
following use, the sewage, coming from
the works are subdivided into separate
flows:
• process and storm flows—the main
bulk of sewage, polluted by oil products
only. After treatment these flows are
supplied back for the application in the
system of the process water line;
• salt-containing and other flows
polluted, besides oil products, by salts
(chlorides, sulphates, sulphides and
others). These flows are delivered for the
additional biological treatment, and, when
treated to the rate required by the sanitary
and fish industry standards, they are
diverted to the reservoir.
We have considered the questions of
mechanical sewage treatment of some
more watercapacious fields of industry
with their reuse in production processes,
as well as the utilization of the sludge of
metallurgical enterprises.
Alongside with flows and water treatment
our Institute is actively developing
technological processes for the purpose of
maximum reduction of pollution found in
sewage and decreasing the specific water
consumption per unit of production.
The subsequent improvement of
mechanical treatment together with other
treatment methods will make it possible to
solve important problem of the most
efficient use of water resources.
The Symposium conduction provides a
means for a wide exchange of experience
in achieving the common task to protect
the nature and environment.
Thank you for attention.
117
-------
Emergency discharge
Figure 1.
Flow Diagram of Pulp Thickening
1. Trough, supplying hydromixture
2. Distribution chamber-separator
3. Building for chemicals
4. Thickener
5. Trough of emergency discharge
6. Piping of thickened pulp
7. Pulppumping station
8. Pressure pulp pipings
9. Troughs of clarified water
10. Chemicals pipings
11. Pond of recycle water supply
Clarified water
Initial pulp
Thickened pulp
118
-------
Figure 2.
Vertical Thickener with Shelf-type Medium
1. Supplying piping
2. Central part
3. Thin-layer shelf type medium
4. Radial collecting troughs
5. Trough, diverting clarified water
6. Piping of thickened pulp derivation
119
-------
- pfclyacrylamide solution
g) |j
-------
Figure 4. 3
Secondary Horizontal Settlers with Magnetic Flocculation
Chamber for Cinder Containing Sewage Treatment
1. Delivering trough
2. Screen
3; Magnetic flocculator
4. Water distribution pipes
5. Secondary horizontal settlers
6. Scraper truck
7. Cinder sump
8. Oil diverting trough
9. Clarified water trough
10. Transborder truck
II. Sludge dewatering bunker
12. Gantry crane
13. Oil pumping station
t
n n
• — 1 r _
'•••••''••'. • .
* t. " r T 1 *
"„ "» •
4| '4* «»,. 1'
• •» 1 •-, '4 '
., P <• ! "
F
\'
17
]
^^
Filter in regeneration
process
^
n
L:
£-
v <
'
^
^ r» — _ ,, ? .
1
^
— ^»
f
L n n
• ^ *
• ^
f A * p-
^/ 6 (V
r ***
'*" 7 '
6 '
• . '. • '.
.1 ' f V
"7 5> P
Is 1 "1.
7 " *> y
Filter in filtration
process
Process-storm
flows
For reuse For biotreatment
Figure 5.
Filters for Post Treatment of Oil Containing Flows
I. Grit chamber 4, Filters 1. Sewage supply to filters 4. Wash water derivation
2. Oil separator 5. Tank and pumping 2. Filters emptying 5. Wash water supply
3. Ponds station of filtered water 3. Filtrate derivation 6. Air piping
7. Drainage
8. Gravel medium
9. Sand medium
121
-------
The Swirl
Concentrator for
Treating and
Regulating Sewered
(Separate and
Combined) and
Unsewered Flows
Richard Field,
Chief
Storm and Combined Sewer
Section
Wastewater Research Division
Municipal Environmental
Research Laboratory
Cincinnati, Ohio 45268
US Environmental Protection
Agency
February, 19
Introduction
Intensive studies to develop and
demonstrate a new device called the swirl
concentrator, for treating and regulating
sewered and unswered wastewater flows
were conducted under the general
supervision of the U.S. Environmental
Protection Agency's (EPA) Storm and
Combined Sewer Technology Program,
Municipal Environmental Research
Laboratory, Cincinnati, Ohio. As a result,
swirl devices are proving to be highly
valuable and innovative tools for the
nation's efforts to clean up pollution of its
water resources.
The swirl concentrator has been developed
following demonstration of a vortex
combined sewer overflow regulator in
Bristol, England, by Smission1 who noted
that the device permitted flow treatment
by solids separation in addition to
functioning as an overflow regulator.
Swirl concentrators achieve removals of
suspended solids by rotationally induced
forces causing inertial separation in
addition to vertical gravity sedimentation
in relatively short detention times.-Aside
from its short detention, other advantages,
such as low power and maintenance
requirements are afforded by the fact that
it is a static, non-mechanical device.
Originally developed as combined (storm
and sanitary') sewer overflow (CSO)
regulators,2.3 the concept has been refined
and extended to selective grit removal,4,s
attainment of primary removal
efficiencies,6 and erosion control.7
Untreated storm overflows from
combined sewers are a substantial water
pollution source during wet-weather
periods. There are roughly 15,000 to
18,000 CSO points in the USA that
emanate from 40% to 80% of the total
organic load in these municipalities during
wet-weather flow periods. It has been
estimated that on a national level the
expenditure for CSO pollution abatement
would be more than $10 billion.8.9.10
In considering wet-weather pollution
abatement, attention must first be directed
to control of the existing combined
sewerage system and replacement or
stricter maintenance of faulty regulators.
Consulting and municipal engineers will
agree that regulator mechanical failures
and blockages persist at the overflow or
diversion points resulting in unnecessary
by-passing, a problem also during dry
weather. Malfunctioning overflow
structures, both of the static and dynamic
varieties, are major contributors to the
overall water pollution problem.
The practice in the USA of designing
regulators exclusively for flowrate control
or diversion of combined wastewaters to
the treatment plant and overflow to
receiving waters must be reconsidered.
Sewer system management that
emphasizes the dual function of CSO
regulator facilities for improving overflow
quality by concentrating wastewater solids
to the sanitary interceptor and diverting
excess storm flow to the outfall will pay
significant dividends. A new phrase has
been coined, the "two Q's," to represent
both the quantitative and qualitative
aspects of overflow regulation. Regulators
and their appurtenant facilities should be
recognized as devices which have the
responsibility of controlling both quantity
and quality of overflow to receiving
waters, in the interest of more effective
pollution control. The swirl concentrator
should definitely be considered when
combined sewer systems are upgraded and
improved regulators are constructed to
reduce the impact of overflows on
receiving water quality.
It is for these reasons that EPA's initial
swirl research and development efforts
placed emphasis on design of the
regulator/concentrator. Subsequent
studies evaluated various swirl device
alternatives such as the degritter, primary
separator, erosion control device, etc. The
following sections describing the basic
principle, design characteristics, and
model studies involve the swirl
regulator/concentrator. It is hoped, since
the various swirl devices are basically
similar, that an appreciation for all of
them can be obtained in this manner.
Basic Principle
The swirl device differs from a sedi-
mentation tank in that it utilizes the
differences in inertia between particles
and liquid as well as gravitational
forces to effect solids-liquid .separation.
Influent flow from the combined sewer
is introduced tangentially to a circular
chamber. The kinetic energy of the
incoming sewage is guided by a vertical
deflector that makes the flow take a
"long path" around the chamber. This
is the so-called primary current and is
readily visible. In any flow around a
bend, a secondary current is intro-
duced to make the surface portion of
the liquid flow outward from the center
and the bottom portion flow toward
the center. The secondary current
that acts in the swirl device is similar
to the current observed in rivers that
makes the stream erode its outer bank
and deposit sediment on its inner bank.
Because of the effects of boundary
friction, the water near the surface
flows faster than the water near the
bottom. The secondary current carries
solids to the center, just as a river
122
-------
(a) Swirl Bagulator/CJoncentrator
(b) Swirl Priirary Separator
s sr
Figure 1.
Isometric Configurations of Swirl Device in Three Applications
123
-------
deposits sediment on the inner bank.
Thus, most of the settleable solids and
grit that are forced down (by gravity)
and outward toward the perimeter
(by inertia! force) are then swept
inward by the secondary current to
the foiil (concentrate) orifice through
which they exit for further treatment
(Figure 1)). The larger volume of
liquid, which should now be relatively
clear of solids, is relieved of its float-
able matter by a trap and passes over
the circular overflow weir and plate for
additional treatment or discharge
directly to the receiving water.
The swirl device regulates flows by a
central circular weir and bottom orifice
which is basically the same as the
common static flow regulator that
operates by the dam and orifice prin-
ciple. During low dry-weather flows,
the entire sewage flow is guided along
a gutter in the bottom and' goes
directly to the interceptor, which con-
veys it to the treatment plant.
During higher-flow storm operation,
it separates solids as described above.
The low-volume, high solids concen-
trate resulting from the swirl action is
diverted via an orifice at the end of the
bottom gutter to the interceptor, and
the clear, high-volume supernatant
overflows the central circular weir. This
liquid can be further treated and dis-
charged into the receiving waters, or,
depending on quality requirements,
directly discharged into the receiving
waters. Provisions can also be made for
storage and subsequent pumping of
this relatively clear liquid to the sani-
tary sewage treatment plant during
low-flow, dry-weather periods. This
method thus allows the maximum treat-
ment possible but still protects the
treatment plant from overloading.
Such a system makes treatment pos-
sible at all times and prevents raw
combined sewage from being passed
directly to the receiving waters during
periods of heavy flow.
Development/Verification of the
Swirl Principle
In order to develop the swirl principle
and provide a design basis, mathe-
matical and hydraulic modeling
studies 2J. *.'." were conducted. These
studies were conducted for EPA by the
American Public Works Association
(APWA).
Swirl device models were developed
using synthetic materials simulating
the particle size distributions and
specific gravities of grit and organics
found in domestic sewage, CSO's, and
erosion laden runoff. The APWA
studies resulted in a series of design
curves relating anticipated performance
to design capacity and other design
parameters.
As will be discussed in a later section,
the above models have been or are
currently being verified in pilot and
prototype facilities using actual sanitary
sewage and CSO at Syracuse, New
York13 (prototype regulator/con-
centrator), Denver, Colorado5 (pilot
degritter), Toronto, Canada" (pilot pri-
mary separator), Rochester, New
York" (pilot degritter and primary
separator), and Lancaster, Pennsyl-
vania " (prototype regulator/concen-
trator "•a in series with swirl degritter
for foul concentrate4).
Structurally, swirl regulator/concen-
trators, swirl degritters, and swirl
primary separators incorporate dis-
tinctly different features. Some of these
differences are illustrated in Figure 1.
The selected configuration for each
application was a result of considera-
tion of hydraulic principles and testing
of a variety of physical models. Dif-
ferences should be noted in weir con-
figurations, baffling and floor layouts.
The units also differ in design features
such as inlet velocities, D../D., (cham-
ber diameter/inlet dimension) ratios,
and H,/D, (weir height/inlet dimen-
sion) ratios.
Performance results were scaled from
model results to predicted prototype
results by using Froude Law scaling
relationships. Model-to-prototype con-
version used the Froude number,,
for scaling of unit dimensions, where
Nt Froude number, v=rveioc'ity, 'g=
gravity acceleration, and S=reference
length.
Since v is equivalent to Q/A, where
discharge, Q, and area, A, are a func-
tion of the square of the inlet dimen-
sion Di,
f\3
Nf=f
Froude number scaling thus employs
the relationship:
Q model
Q prototype
Di model
Di prototype
5/2
for scaling of hydraulic flows. Geo-
metric similarity must be maintained
between model and prototype. In addi-
tion, four fraction (% of flow con-
taining concentrate) must be the same
in prototype as in model.
In a similar manner,, particle settling
velocities were also scaled, using
Froude Law relationships.
Since
Nf=f v"
scaling of settling velocities employs
the relationship
v.z prototype
v.2 model
=: D' prototype
D* model =A
where D*=unit chamber diameter,
v.=particle settling velocity, and
A=scale factor
Since settling velocity is dependent on
effective particle diameter and specific
gravity, the studies employed synthetic
materials to represent settling veloci-
124
-------
ties in the model studies. These repre-
sented scaled-down settling velocities
from prototype scale for expected
particle size distributions and specific
gravities, and importantly, enabled a
constant conservative substance to
optimize hydraulic model configuration.
Swirl Design Characteristics
To be useful in coping with storm
overflow pollution, the swirl unit must
function under conditions of widely
varying flowrates and solids contents.
Thus, as implied above, the various
elements of the chamber must be care-
fully designed. The isometric view of
the swirl regulator/concentrator
[Figure 1 (a)] may help to visualize
the various special elements which are
identified by small letters.
Inlet Ramp
The inlet ramp [Figure 1 (a), a] in-
troduces the incoming flow at the
bottom of the chamber, thus allowing
the solids a greater chance of being
separated.
Inflow should be nonturbulent to pre-
vent the solids from being carried
directly to the overflow weir along
with the water. However, there should
be enough energy to force the water
around the chamber. The floor of the
inlet ramp should be v-shaped to allow
for self cleaning during periods of low
flow.
Plow Deflectors
The flow deflector [Figure 1 (a), b]
is a vertical, freestanding wall that is
an .extension of the interior wall of the
inlet ramp. It directs the flow around
the chamber, creating a longer path
and a greater chance of solids separa-
tion.
Scum Ring
The scum ring [Figure 1 (a), c] pre-
vents floating solids from overflowing.
It should extend at least 15.2 cm (6 in)
below the level of the overflow weir
crest and works by baffling floatables
and preventing them from overflowing
the weir until they reach the floatables
trap.
Overflow Weir and Downshaft
The overflow weir and weir plate
[Figure 1 (a), d] allow the liquid that
has passed through the chamber and
is now free of most of its settleable
and floatable solids to leave the cham-
ber through a central downshaft
[Figure 1 (a), i] and either be stored
or bypassed to the river.
Spoilers
The spoilers [Figure 1 (a), el are
radial flow guides that extend from the
downshaft to the scum ring and are
vertically mounted on the weir plate.
They allow efficient and controlled
operation of the swirl concentrator
under overload conditions and prevent
turbulence and vortex conditions that
impede proper functioning.
Floatables Trap
The floatables trap [Figure 1 (a), f] is
a surface flow deflector that directs
floatables under the weir plate for
storage during wet-weather operation.
When the liquid level in the chamber
decreases after the rainfall, the float-
ables exit through the foul sewer outlet.
Foul Sewer Outlet
The foul sewer outlet [Figure 1 (a), g]
located on the floor of the chamber, is
the orifice through which the dry-
weather flow and storm-flow concen-
trate exit, via the interceptor, to the
treatment plant. It is placed at the point
of maximum settleable solids concen-
tration and is designed to reduce the
clogging problems that often incapaci-
tate regulators.
Floor Gutter
The primary floor gutter [Figure 1 (a),
hi is the peak dry-weather flow chan-
nel connecting the inlet ramp to the
foul sewer outlet to avoid dry-weather
solids deposition.
Modeling: Solids Separation
Efficiencies
Actual combined sewage could not be
used in the scaled-down hydraulic
model that was constructed for initial
development of the swirl regulator/
concentration principle. Grit and settle-
able solids had to be simulated with
gilsonite, a natural hydrocarbon that
could be mathematically related to the
various solids components. Floatables
were simulated with pethrothene.
For a prototype, the predicted efficien-
cies were as follows:
For floatables with a specific gravity
range of 0.90-0.96 and particle sizes
between 5 and 50-mm, the chamber
should remove 65 to 80 percent.
For grit with a specific gravity of 2.65
and particles larger than 0.3-mm, re-
moval should be between 90 and 100
percent. Removal efficiencies decrease
to about 75 percent for 0.2-mm par-
ticles and to less than 40 percent for
0.1-mm particles.
For settleable organic solids with a
specific gravity of 1.2 and particles
larger than 1-mm, the efficiency should
range from 80 to 100 percent. This
fraction represents 65 percent of the
total amount of settleable solids in the
design solids concentration. For the
finer particles, removal efficiencies de-
crease to about 30 percent for 0.5-mm
particles and to less than 20 percent
for 0.3-mm particles.
The mathematical simulation of proto-
type efficiency indicated that the
device will work well, up to 150 per-
cent or greater of design flow.
Syracuse, New York (Prototype Swirl
Regulator/ Concentrator)
A 3.6 m (12 ft) diameter swirl CSO
regulator was installed at the West
Newell Street outfall in Syracuse, New
York13 (Figure 2). Design flood flow to
the swirl device was based on maxi-
mum carrying capacity of the .61 m
(24 in) diameter inlet combined
125
-------
sewer—33,687 ma/day (8.9 mgd)—
and a design flow for quality control,
in accordance with scale model inves-
tigation of 25,738 m'/day (6.8 mgd).
Tests indicate that the device is capable
of functioning efficiently over a wide
range (10:1) of CSO rates, and can
effectively separate suspended matter
at a small fraction of the detention
time required for conventional sedi-
mentation or flotation (seconds to
minutes as opposed to hours by con-
ventional tanks). At least 50 percent
removal of suspended solids and BODc
were obtained. Tables 1. and 2. con-
tain further treatability details on
the Syracuse prototype. The capital
cost of the 25,738 mVday (6.8 mgd)
Syracuse prototype was $55,000 or
$2/m'/day ($8,100/mgd) and
S2,470/ha ($l,000/ac) which makes
the device highly cost-effective.
Denver, Colorado (Pilot Swirl De-
gritter)
A large 1.8 m (6 ft) pilot swirl device
designed as a grit removal facility was
tested by the Metropolitan Denver
Sanitary District No. 1.* Figure 3 con-
tains a suggested swirl degritter layout
for above-the-ground installations.
It was found under testing performed
on domestic sanitary wastewater, at
times spiked with 0.25-mm dry blasting
sand to simulate swirl regulator foul
concentrate concentrations, that the
swirl unit performed well. The effi-
ciency of removing grit particles of
2.65 SG. and sizes greater than 0.2-mm
was equal to that of conventional grit
removal devices. Scaled up detention
times for full size swirl units having
volumes one tenth that of conventional
tanks, were as low as 20 seconds,
whereas detention times of one minute
and greater are normal for conven-
tional grit chambers.
The small size, high efficiency and
absence of moving parts in the basic
swirl degritter unit offers economical
and operational advantages over con-
ventional grit removal systems.
Toronto, Canada (Pilot Swirl Primary
Separator)
• - Somplmq Prtnfi
Figure 2.
Profile of Syracuse, NY Swirl Regulator/ Concentrator
Table 1.
Swirl Regulator/Concentrator: BOD s Removal
Mass Loading, kg
Storm No.
7-1974
1-1975
2-1975
Influent
277
97
175
Effluent
48
30
86
%
Rem.
82
69
51
Average BODs
per storm, mg/1
Inf.
314
165
99
Eff.
65
112
70
%
Rem.
79
32
29
Table 2.
Swirl Regulator/Concentrator: Suspended Solids Removal
Swirl Concentrator
Mass Loading Average SS
kg per storm, mg/ 1
Storm No.
02-1974
03-1974
07-1974
10-1974
14-1974
01-1975
02-1975
06-1975
12-1975
14-1975
15-1975
Inf.
374
69
93
256
99
103
463
112
250
83
117
Eff.
179
34
61
134
57
24
167
62
168
48
21
%
Rem.b
52
51
34
48
42
77
64
45
33
42
82
Inf.
535
182
110
230
159
374
342
342
291
121
115
Eff.
345
141
90
164
123
167
202
259
232
81
55
%
Rem.b
36
23
18
29
23
55
41
24
20
33
52
Conventional Regulator
Mass Loading
kg
Inf.
374
69
93
256
99
103
463
112
250
83
117
Under
flow
101
33
20
49
26
66
170
31
48
14
72
%
Rem. a
27
48
22
19
26
64
34
27
19
17
61
»For the conventional regulator removal calculation, it is assumed that the SS concentration of the
foul underflow equals the SS concentration-of the inflow.
bData reflecting negative SS removals at tail end of storms not included.
126
-------
A study6 was conducted to determine
if the swirl concentrator principle could
be used to provide primary treatment
to sanitary sewage, CSO, and storm-
water. In. comparison the swirl regula-
tor/concentrator provides a coarser
pre-treatment. Initially a 0.9 m (3 ft)
diameter hydraulic model and a mathe-
matical model were used at the
LaSalle Hydraulic Laboratory, Ltd.,
LaSalle, Quebec, Canada to arrive at
a design configuration and basis.
The hydraulic model studies were
based on synthetic solids made of
Amberlite anion exchange resin IRA-
93, which was considered to properly
simulate actual solids in sanitary
sewage flows." The design criteria
were based on Froude Law scale up.
The design was then tested on a pilot
3.7 m (12 ft) diameter installation with
real sewage at Metropolitan Toronto's
Humber wastewater treatment plant,
Toronto, Canada.
The pilot unit was tested at a design
flow of 1,137 mVday (0.3 mgd) and
at 1,700 m'/day (0.45 mgd). The
results of the tests indicated that the
unit performed as effectively (40 per-
cent suspended sofids removal) as con-
ventional settling basins at the Humber
wastewater treatment plant operating
at an overflow rate of 81.46 m*/day/
m' (2,000 gal/day/ft2) with 1.06 hours
detention time. The detention time in
the swirl separator was 0.34 hours at
an overflow rate of 108 mVday/m2
(2,650 sal/day/ft1) and 0.23 hours at
an overflow rate of 162 mVday/ms
(3,980 gal/day/ft3), respectively for
the two above mentioned capacities.
Swirl detention times are calculated to
include the sludge hopper to provide a
size comparison with conventional
tanks. However, when the size of the
unit is calculated for a full 60 percent
suspended solids removal, the size and
retention time becomes equal to that of
the conventional unit (Figure 4).
The studies provided proof of the
applicability of the swirl principle to
the function of primary clarification
and verified the design that was based
Grit
Chamber
Wash water
Overflow Weir A
Wash water
Outlet
Grit Washer
and Elevator
Section A-A
Figure 3.
Suggested Swirl Degritter Layout for Above-the-Ground
Installation with Inclined Screw Conveyor
conventional primary
Killing tatik
twirl separator
Figure 4.
60 120
Time (minutesI
Comparison of Time to Achieve Primary Treatment: Swirl vs
Conventional at Humber Wastewater Treatment Plant, Toronto,
Canada
127
-------
on hydraulic model optimization. In a
short detention period solids are de-
posited by inertial and gravity action
and agglomeration mechanisms. Figure
4 indicates that removal efficiencies
(less 50%) matched actual perform-
ance of conventional primary settling
facilities in shorter periods of time at
the Metropolitan Toronto facility.
Importantly, in storm flow treatment
application suspended solids will be
heavier due to high sewer transport
velocity and therefore will tend to
separate more readily which favors
swirl separation over sedimentation.
The basic advantages of the swirl
clarification principle are that it re-
quires: (1) less land than conventional
sedimentation, and (2) no mechanical
sludge collection equipment within the
tank for the collection of settled
solids in the sludge hopper. The latter
advantage is partially achieved by
providing a deep conical hopper over
the entire floor area, thus, imposing an
increase in costs.
The estimated construction costs of a
swirl primary separator and a con-
ventional primary settling tank both
with capacities of 1893 mVday (0.5
mgd) are $275,000 and $138,000, re-
spectively.
Annual operation and maintenance
(O&M) costs are estimated to be less
($3000 less for 0.5 mgd (1983 mV
day)) with the swirl unit. In urban areas
where land is expensive the fact that
the swirl requires one-half or less the
surface area (accordingly to overflow
rates) could make it competitive.
The engineer must consider the costs
of construction, O&M, and land in a
cost comparison figure. In the locations
where land is at a premium it is ad-
visable to compare the costs of smaller
parallel swirl units with their lower
O&M costs against those of conven-
tional tanks.
Rochester, New York (Pilot Swirl
Primary Separator and Swirl Degritter)
Pilot plant treatability studies were
undertaken to delineate the treatment
alternatives available for control of
CSO quality under the Monroe County,
Division of Pure Waters, Rochester,
New York."
A major emphasis of the study was
development of cost/benefit compari-
sons of processes that would allow
primary-level treatment efficiencies.
These processes were compared relative
to their response to treating variable
quality CSO's. Treatment of the highly
concentrated first-flush overflow was
of particular importance.
The pilot facilities included a swirl
degritter and a swirl primary separator
connected in series (Figure 5). The
swirl degritter was 0.914 m (3 ft) in
diameter and approximately 1.22m (4
ft) in total depth. During normal opera-
tions, the overflow from the degritter
became the influent for the swirl pri-
mary separator. Provisions were also
made, however, to allow the influent
to bypass the degritter and go directly
into the swirl primary separator. The
swirl primary separator was 1.8 m
(6 ft) in diameter and approximately
1.8 m (6 ft) in total depth including
hopper. The inlet pipe diameter was
10 cm (4 in) and the outlet diameter
was 7.6 cm (3 in).
SWIRL PRIMARY
SEPARATOR
Figure 5.
Schematic Diagram of Swirl Pilot Facility in Rochester, NY
The swirl devices were tested at
flowrates ranging from 0.9 I/sec to
4.4 I/sec (15 to 70 gpm). Using
Froude Law scaling relationships, these
translate to flows of 757 m'/day to
3,785 mVday (0.2 to 1.0 mgd) for
a 4.6 m (15 ft) diameter swirl pri-
mary separator and nominal overflow
rates of 51 mVday/m* to
233 mVday/m8 (1245 gpd/ft8-
5705 gpd/ft"), respectively. Mathe-
matical performance models were
developed for each system relating
suspended solids removal rates to
influent flowrate (scaled' by Froude
number) and influent concentration.
These performance equations were
compared to the design curves of the
earlier development studies * *• *• ° for
swirl devices. Taking differences of
particle size distributions into account,
the Rochester data generally support
the design presented by the earlier
work.
Lancaster, Pennsylvania (Prototype
Regulator/Concentrator With Proto-
type Swirl Degritter In Series For
Foul Concentrate)
The original swirl regulator/concen-
trator and degritter hydraulic model
studies" were conducted for the
Lancaster, Pennsylvania prototypes.
The proposed prototype system being
supported by an EPA demonstration
grant,'r is comprised of a 7.3 m
(24 ft) main diameter swirl CSO regu-
lator/concentrator with a smaller
2.4 m (8 ft) diameter swirl degritter
in series to degrit the swirl regulator
foul underflow prior to its entry to
the interceptor pumping and sewer
system (Figure 6). Upstream grit
removal will prevent doKvnstream
pumping problems, sewer siltation, and
treatment problems in the Lancaster
sewerage system. The relatively clear
swirl regulator/concentrator overflow
will receive disinfection and go direc-"
tly to the Conestoga River. This
swirl system will serve a drainage
area of 55 ha (135 ac).
The dry and wet-weather underflow
128
-------
ELEVATED
GRIT
HOPPER
STEVENS AVE,
PUMPING STATION
CLEAR
CSO
INFLOW
OVERFLOW
o
OUL (CONCENTRA
INTERCEPTOR
SWIRL
DEGRITTER
SWIRL .
REGULATOR/
CONCENTRATOR
CLEAR
OVERFLOW
CONESTOGA RIVER
Figure 6.
Schematic Diagram of Swirl Prototype System at Lancaster, Pennsylvania
designed in accordance with down-
stream interceptor capacity is
12,200 mVday (3.2 mgd). The under-
flow bypasses the swirl degritter
during dry weather and is directed
through the degritter when it is con-
centrated (1,000's of mg/1 suspended
solids) during wet-weather flows.
The regulator was hydraulically de-
signed to allow 1.1x10" m'/day (291
mgd), the peak upstream inlet sewer
capacity, without flooding. At the
treatment design flow, of 97,650 mVday
(25.8 mgd) which represents a six
in one year storm frequency, 65 per-
cent suspended solids removal is
estimated*.
Hydraulic model studies" indicated that
inflows much greater than 220,060 mV
day (58 mgd) would induce too
violent a flowfield for effective
treatment. Accordingly, an overflow
relief weir is being provided along a
section of the swirl chamber periphery
to allow only that portion of the
flows greater than 220,060 mVday
(58 mgd) to bypass treatment with-
out upsetting the beneficial swirl
currents which will continuously
provide treatment to flows less than
220,060 mVday (58 mgd).
The contractor's construction cost
proposed for this system, which also
contains degritter and control housing,
a pilot microscreen, regulator roofing,
and various appurtenances and research
instrumentation is $669,000. The unit
costs are $0.61/mVday ($2,300/mgd)
or $12,l63/ha ($4,955/ac).
The Swirl Concentrator as an
Erosion Control Device
Erosion-sedimentation causes the strip-
ping of land, filling of surface waters,
and water pollution. Urbanization
causes accelerated erosion, raising
sediment yields two to three orders
of magnitude from 3.5x10' - 10s
kg/kmVyr (10* - 10' ton/miVyr) to
3.5xlOs -107 kg/kmVyr (101-10=
ton/miVyr)." At the present
national rate of urbanization, i.e.,
1,620 ha/day (4,000 ac/day), erosion/
sedimentation must be recognized
as a major environmental problem.
129
-------
LEGEND
a — Inlet
b — Flow Deflector
c -Spoilers
d — Overflow Weir
e—Weir Plate
f — Overflow (clear)
g — Underflow (solids)
h—Floor
Figure 7.
Isometric of Swirl Concentrator as an Erosion Control Device
a. inlet
b. flow deflector
c. spoilers
d. 'overflow weir
e. weir plate
f. overflow (clear)
g. underflow (solids)
.•h. 'floor
130
-------
UNDERFLOW
CONSTRUCTION
SITE
DRAINAGE
AREA
/
I
> i
\ v
SOLIDS
LAGOON
OR
FOREBAY
RETENTION POND
OR
RECEIVING WATER
OVERFLOW
SWIRL CONCENTRATOR AS
EROSION CONTROL DEVICE
PIPE
OR DITCH
Figure 8.
Swirl Erosion Control Device Schematic Diagram
TREATMENT
PLANT
\ SANITARY
N INTERCEPTOR
SMALL CONCENTRATE
TANK
SWIRL
CHAMBER
STORM DRAIN
NETWORK
Figure 9.
Swirl Urban Storm Runoff Pollution Control Device Schematic Diagram
131
-------
The APWA under an EPA contract
has 'developed a swirl erosion control
device (Figure 7) which appears to
be capable of performing an effective
job of removing erosion, particles
from stormwater' runoff at construc-
tion or other vulnerable sites.-Such a
swirl device can be rapidly and
economically installed at points of
erosion runoff by use of a conventional
cattle'watering tank having a 3.66'm
(12 ft) diameter and a 0.9 m (3 ft)
depth, fitted and equipped with a
suitable inlet line, a circular overflow
weir, a foul sewer outlet, and neces-
sary interior appurtenances (Figure 7).
This chamber could be readily
disassembled, moved to another site,
and reinstalled for the treatment of
erosion runoff flows. If a permanent
structure is desired, it can be fab-
ricated out of concrete.
The de-silted, or clarified effluent
could be discharged to drainage
facilities and disposed of into receiving
waters or other points of disposal
or use. The collected solids could be
discharged through the foul sewer
outlet and entrained or collected at
suitable points of erosion or for use
for other predetermined purposes
(Figure 8).
The Swirl Concentrator as
Urban Storm Runoff Pollution
Control Device
Swirls similar to the CSO regulator
variety can be installed on separate
storm drains before discharge and the
resultant concentrate can be stored in
relatively small tanks since concen-
trate flow is only a few percent of
total flow. Stored concentrate can
later be directed to the sanitary sewer
for subsequent treatment during low-
flow or dry-weather periods, or if
capacity is available in the sanitary
interceptor/treatment system, the
concentrate may be diverted to it
without storage (Figure 9). This
method of storm water control would
be cheaper in many instances than
building huge holding reservoirs for
untreated runoff, and offers a feasible
approach to the treatment of separate-
ly sewered urban stormwater.
Potential Uses
The swirl principle may be employed
anywhere it is desirable to remove
solid particles from liquid flows. In the
field of water pollution control this
principle could relate to the degrit-
ting of sanitary and storm flows and
to primary separation, sludge thicken-
ing, and the final clarification process.
Because the swirl creates a defined
mixing pattern it appears feasible to
apply a form of the swirl for the
simultaneous enhancement of chemical
coagulation and disinfection while
clarification of raw water for potable
usage or wastewater is taking place.
Other possible uses include various
industrial processing.
Applications to relatively steady-state,
dry-weather (municipal and industrial)
flows may involve less arduous condi-
tions of operation than does the
CSO application. Both the hydraulic
laboratory and the mathematical
model investigations have indicated
that solids separation efficiency
may increase 'if the device operates
under steady flow conditions.
Better efficiencies may also be
achieved with two half-size chambers
as opposed to one full-size unit. With
two units operating in parallel, one
chamber could be used for all flows
lower than a predetermined design
value, and the second could be
used if the storm flow exceeded that
value. This approach would provide
better separation at both higher and
lower flowrates. Another possibility
is operating the units in series to
improve solids removal by breaking a
wide range of particle characteristics
into narrower grain size and specific
gravity bands.
References
Smisson, B., Design, Construction and
Performance of Vortex Overflows.
Proc., Symp. on Storm Sewage Over-
flows, Inst. Civil Eng. (G.B.), 1967. .
American Public Works-Association.
The Swirl Concentrator,, as a Combined
Sewer Overflow Regulator Facility.
EPA, EPA-R2-72-008, NTIS No. PB
214 134, September 1972.
Sullivan, R.H., et al. Relationship Be-
tween Diameter and Height for the
Design of a Swirl Concentrator as a
Combiner Sewer Overflow Regulator.
EPA, EPA-670/2-74-039, NTIS
No. PB 234 646, July 1974.
Sullivan, R.H., et al. The Swirl Con-
centrator as a Grit'Separator Device.
EPA, EPA-670/2-74-026, NTIS No.
PB 233 964, June 1974.
Sullivan, R.H., et al. Field Prototype
Demonstration of the Swirl Degritter.
Draft copy of EPA Report, EPA Grant
No. S803157, January 1977.
Sullivan, R.H., et al. The Swirl Pri- .
mary Separator: Development and
Pilot Demonstration. Draft copy of
EPA Report, EPA Grant No. S803157,
January 1977.
Sullivan, R.H., et al. The Swirl Con-
centrator for Erosion Runoff Treat-
ment, EPA, Report at Publishers, EPA-
600/2-76-271, February 1977. .
American Public Works Association.
Problems of Combined Sewer Facilities
and Overflows 1967. EPA, 11020—
12/67, NTIS No. PB 214 469, De-
cember 1967.
Black, Crow & Eidsness, Inc. and
Jordan, Jones & Goulding, Inc., for
The National Commission on Water-
Quality. Study and Assessment of the
Capabilities and Cost of Technology
for Control of Pollutant Discharges
from Urban Runoff. Draft Report,
July 1975.
Heaney, J.F., et al. Nationwide Eval-
uation of Combined Sewer Overflows
and Urban Stormwater Discharge, Vol-
132
-------
ume 11: Cost Assessment and Impacts.
EPA, Report at Publishers, 1977.
Dalrymple, R.J., et al. Physical and
Settling Characteristics of Particulates
in Storm and Sanitary Wastewaters.
EPA, 'EPAi670/2-75-Oil, NTIS No.
PB 242 001, April 1975.
Drehwing, FJ. et al. Pilot Plant
Studies—Combined Sewer Overflow
Abatement Program (Rochester, NY).
Draft copy of EPA Report, EPA
Grant No. Y005141, November 1976.
Field, R., et al. Treatability Determina-
tions for a Prototype Swirl Combined
Sewer Overflow Regulator/Solids
Separator. Proceedings: Urban Storm-
water Management Seminars, EPA,
WPD 03-76-04, January 1976, pp.
11-99 to 11-111. Paper also presented
at the International Association on
Water Pollution Research-Workshop,
Vienna, Austria, September 1975,
and at the New York Water Pollution
Control Association Annual Meeting,
January 1976.
Sullivan, R.H., et al. Nationwide Eval-
uation of Combined Sewer Overflows
and Urban Siormwater Discharge,
Volume III: Characterization. Draft a
copy of EPA Report, EPA Contract
No. 68-03-0283, January 1977.
EPA Demonstration Grant No.
S802219 (formerly 11023 GSC),
Demonstration of a Dual Functioning
Swirl Combined Sewer Overflow
Regulator/Solids-Liquid Separator
with Swirl Degritter, City of Lancaster,
Pennsylvania. Estimated operation
start and project completion.dates,
December 1977 and December 1979,
respectively.
Other Important Swirl
Device References
White, R.A., A Small Scale Swirl Con-
centrator for Storm Flow. Thesis,
University of Wisconsin. Milwaukee,
Wisconsin.
Sullivan, R.H., et al. The Helical Bend
Combined Sewer Overflow 'Regulator.
EPA, EPA-600/2-75-062, NTIS No.
PB 250 545, December 1975.
Field, R., The Dual Functioning Swirl
Combined Sewer Overflow Regulator/
Concentrator. EPA, EPA-670-2-73-
059, NTIS No. PB 227 182, Septem-
ber 1973.
Field, R., Design of a Combined Sewer
Overflow Regulator/Concentrator.
Jour. Water Poll. Control Fed.,
46(7): 1722-1741, 1974.
Field, R., The Dual Functioning Swirl
Combined Sewer Overflow Regulator/
Concentrator. Pergamon Press (G.B.)
October 1973.
Field, R., A Swirl Device for Regulat-
ing and Concentrating of Combined
Sewer Overflows. News of Environ-
mental Research in Cincinnati, October
25, 1974.
Field, R., et al. Give Stormwater Pol-
lutants the Spin. The American City
and County, 77-78, April 1976.
Benjes, H.H., Jr. Cost Estimating
Manual-Combined Sewer Overflow
Storage Treatment. Report at Pub-
lishers, EPA, EPA-600/2-76-286,
1977.
Field, R. and E.J. Struzeski, Jr. Man-
agement and Control of Combined
Sewer Overflows. Jour. Water Poll.
Control Fed., 44(7): 1393-1415, 1972.
P. I. Galanin
Deputy Head of Water Supply
and Sewerage Administration
of the city of Moscow
Sewage Treatment
of the City of
Moscow
In view of the rapid rates of the
industrial production development, the
problem of environmental control
and environment condition improve-
ment, particularly water resources
control, is becoming more and more
acute.
In solving this problem, of greatest
importance is the act, issued by the
Central Committee of Communist
Party of the Soviet Union and
Council of Ministers of the USSR—
"On intensification of environmental
control and improvement of utiliza-
tion of natural resources," adopted in
development of the law, approved by
the Supreme Soviet of the USSR.
At present the development of
measures of long-term and annual
plans on environmental control and
rational utilization of natural resources
makes an integral part of the plans
of national economy.
The most difficult in solving the
problem of environmental control is
the problem of water resources
control in large cities with diversified
industries, high concentration of
transport facilities and housing and
public building density.
Important work • on environmental
preservation, regeneration and condi-
tion improvement is conducted in our
capital—the city of Moscow.
Preservation of rivers and water
reservoirs is an indispensable condi-
tion, ensuring sanitary welfare of such
a large city as Moscow.
In recent years large-scale work on
the construction of large sewerage
133
-------
systems and treatment plants has
been carried out and is being conducted
in the city of Moscow. Hundreds of
local industrial waste treatment plants
have been commissioned.
The central sewerage system serves
98.5% of the city population. To
the treatment plants with a total
capacity of 4,230,000 m3/day more
than 4.3 mill mVday is fed and
subjected to full biological treatment.
Industrial wastes make up about 40%
of the total municipal sewage.
The total length of channels, sewers
and mains is'more than 4500 km.
The capacity of the sewage pumping
stations is 6.5 mill mVday.
The sewage is subjected to full
biological treatment ot five activated
sludge plants of the following ca-
pacity (mVday):
Kuryanovskaya plant — 2,125,000
Ljuberetskaya plant — 1,500,000
Ljublinskaya plant — 500,000
Tushinskaya and Zele
nogradskaya plants — 105,000
Basic average figures for sewage pollu-
tion and treatment quality, mg/1
The Moskva is a low-capacity water
reservoir, the natural discharge of the
river at low-water being 13 to 15 mV
sec. For the sanitation of the river,
measures have been worked out and
effected to supply it with water from
the Moskva Canal at a rate of
31 nrVsec.
The modern satisfactory condition
of the river has been achieved both
by ceasing the discharging of raw
sewage and by a high enough level of
activated sludge plant operation.
One of the most efficient plants is
the Kuryanovskaya activated sludge
plant (Fig. I).
The plant is one of the largest in the
world. As far back as before World
War II of 1941-45, the first unit
was designed' with a capacity of
750,000 m3/day. The construction,
interrupted by the war, was completed
in 1950. In 1960 the capacity of the
the plant was increased to I mill mV
day.
In view of the rapid house building
(120,000 flats are built in Moscow
per year), systems for disposal of
sewage and treatment plants are
Plant
Kuryanovskaya
Ljuberetskaya
Ljublinskaya
Zelenogradskaya
Influent water
SS BOD
Effluent water
SS BOD
195
185
164
176
176
208
194
101
10
10
16
1.5
10.2
10.2
14;5
1.0
DO
5.8
4.2
3.5
8.7
At the large activated sludge plants
of the city of Moscow a two-stage
sewage treatment scheme has been
used: mechanical and biological. At
present, a transition is being made
to the three-stage treatment, using
filters for polishing, with different
types of granular filter beds.
High demands for better quality of
effluents, discharged to the Moskva,
have been called forth by the
necessity to maintain sanitary condi-
tions and self-purification processes
of the river.
being constructed at increasing rates.
In 1965 the Ljuberetskaya activated
sludge plant with a capacity of
1.5 mill mVday was commissioned.
In 1966 construction of the second
unit of the Kuryanovskaya plant of
1.0 mill mVday was started. The
plants were put into operation in
1971. At present the 3rd unit is
being constructed with a capacity of
1.00 mill m3/day, and 500,000 m3 will
be commissioned in 1977. Construc-
tion of the whole plant complex is to
be completed in 1978.
The total capacity of the Kury-
anovskaya plant will be 3,125,000 m3/
day, including 2 mill mVday of the
plant for polishing biologically treated
effluents. The facilities of the second
unit of the plants are more ad-
vanced in relation to .process and
design; experience of' construction
and operation of the facilities has
been widely used in many cities of
the USSR.
Sewage treatment at the activated
sludge plants is affected according
to the classical scheme:
screens -» grit chambers —» primary
settling tanks —» aeration tanks
-» secondary settling tanks
Mechanical treatment
facilities
Screens for mechanical removal and
grinding of the screenings intercepted
are provided at the plants. The
openings are 16 mm wide. 10-12.5 1
of screening are removed per 1000 m*
of sewage. Moisture content—75-
77%, ash content—6 to 9%. The
ground screenings are pumped to
the digestion tanks.
Grit chambers—vertical, horizontal,
horizontal and aerated. 20 to 25 1
of grit are removed per 1000 m8 of
sewage. Moisture content—36-44%,
ash content—72 to 80%, grit con-
tent—65-72%, grain size 0.25 mm—
75%.
Primary settling tanks of radial type,
diameters of the settling tanks—
33, 40 and 54 m. At the second and
the third units of the Kuryanovskaya
activated sludge plant the settling
tanks (diameter—54 m) • are combined
with preaerators.
Detention period—1.5 to 2.0 h, the
percentage removal of suspended
solids—45-65. BODa removal—
30-35%. Moisture content of the
sludge—93-94%.
Biological Treatment Facilities
At the Kuryanovskaya and the
134
-------
Ljuberetskaya activated sludge plants
four-corridor aeration tanks are used
with separate regeneration of acti-
vated sludge. Aeration is through
diffuser plates.
Air consumption—5.2-6.5 m' per 1
ms of sewage.. Aeration duration—
4.5-6.0 h. BOD. removal—91-92%;
mixed liquor suspended solids in
aeration tanks—1.2-1.5 g/1, in acti-
vated sludge regenerating tanks—
4.5-5.5 g/1. Dissolved oxygen—6-7
mg/1.
Secondary settling tanks of radial type,
diameters—33, 40 and 54 m. De-
tention period—2.0-2.2 h, activated
sludge removal is by sludge scrapers.
Tertiary Treatment
The solution of the problem of pre-
venting the pollution of water re-
servoirs is tightly connected with the
development of sewage treatment
technology and finding ways for re-
use and multiple use of effluent water
in the systems for industrial water
supply of industrial plants.
Three-stage and more than three-stage
treatment, or the so called tertiary
treatment, is more and more widely
used. In the system of Moscow
sewerage, tertiary treatment was
first used on an industrial scale at the
Zelenogradskaya activated sludge plant
in 1965.
The construction of the tertiary
treatment plant was necessitated by the
protection of Skhodnya River, down
the stream of which there are
several ponds, including a pond for
trout breeding; the river is being
widely used for recreational purposes
and by the population in the country
places.
Average monthly figures for final
effluent quality:
after biological
treatment
BODs (mg/ 1)
Suspended
sol ids (mg/1)
7.4
11.0
after tertiary
treatment
1.0
1.5
Figure I
Plan of the facilities of the Kuryanovskaya activated sludge plant
1.1st unit of 1.0 mill m3/day capacity
2.2nd unit of 1.0 mill m3/day capacity
3.3rd unit of 1.0 mill m3/day capacity
4.Tertiary treatment facilities of 2.0 mill m'/day capacity
S.Sludge treatment facilities
6.Experimental facilities — 125000 m3/day
/* .Moskva
Figure 2.
Flow sheet of tertiary treatment of the Kuryanovskaya activated
sludge plant:
1. feed conduit 8. chlorinator house
2. emergency water discharge pipe 9, polished water tank
3. filter and drum screen building 10. polished water pumping station
4. outlet channel 11. grit chamber
5. contact channel 12. sand drying bed
6. pumping station for filter washing 13. wash-water tank
7. sand hopper 14. wash-water pumping station.
135
-------
The experience of tertiary treatment
in the town of Zelenograd has been
used at the Kuryanovskaya activated
sludge plant, where 2 mill mVday
will be subjected to tertiary treat-
ment. (Fig. 2). At present the
polished water is being used at
several industrial plants of the town.
In the future the industrial plants will be
supplied with up to 5 mVsec of
water for industrial purposes.
It is considered whether the polished
water could be used for land irrigation
and supplying water reservoirs with
water.
The plant complex includes: rapid
filters, consisting of a drum screen
building, a filter building and a
pumping station building for filter
washing, a grit chamber, a wash-water
tank, and a wash-water pumping
station. Effluents are supplied by
gravity to the tertiary treatment
plant.
To remove coarse particles, drum
screens are used with filter cloth, mesh
size of 0.5 x 0.5 mm. Filters are of
coarse grain type with a nitration area
of 109 nr. Filtration rate is 10 to
15 m/h.
Filters are washed with filtered water
for 6 min. Wash rate is 18 1/sec.nv.
The pumping station is designed for
simultaneous washing of two filters
twice a, day. The filtered water from
the two filter sections enters the
channels, flows by gravity to the con-
tact channel, and then is discharged
to the Moskva. While flowing to
the water reservoirs the water is
oxygenated by waterfall aeration.
Wash-water from the filters passes
through the grit chamber and enters
the wash-water tank, from where it
is pumped by the pumps, installed
at the wash-water pumping station,
to the primary facilities of the
plant.
The tertiary treatment plant ensures
effluent purification to a degree higher
than that of the full biological treat-
ment, and guarantees stable
composition of the effluent, meeting
the sanitary standards for water dis-
charged to water reservoirs and for
the industial water quality.
Sewage Sludge Treatment
The activated sludge plants of the
city of Moscow produce 19,000 to
20,000 m3 of sludge per day with
moisture content 96%. The sludge
treatment is conducted by subsequent
stages: digestion in aerobic digesters
at thermophilic conditions (52°C),
washing and thickening, coagulation,
vacuum filtration and heat drying.
The charge dose of the digestion tanks
in terms of actual moisture content is
15-18%. The steam from the boiler
room enters the digestion tanks through
the injection heaters. Steam consump-
tion per 1 m3 of sludge charged is
50 to 60 kg. Gas yield per 1 kg of
volatile solids is 400 to 500 1.
The sludge is vacuum treated on drum
vacuum filters -40-3 (diameter of
the drum—3 m, filtration surface
area—40 m2). The average capacity
of vacuum filters for 1976 was 22.8
kg/mVh using chemical doses of
ferric chloride—3.7% and of lime—
10% for sludge dry residue, with
moisture content of the entering
sludge—95.9% and moisture content
of the cake—77-78%.
The sludge is heat treated in the
dryers of drum type (diameter—3.5
m, length—18 m). The design
temperature of the flue gases, produced
by the combustion of fermentation
gases of the digestion tanks in the
dryer furnaces, at the inlet of the
dryer is 800°C; the temperature of
waste gases at the outlet is 250° C.
The waste gas is used for sludge
heating in scrubbers.
Due to complex content of nitrogen,
phosphorus, potassium, lime and
microelements in the treated sludges,
it is advisable to use them as organic
and mineral fertilizers. 250,000 to
300,000 t/year of sludges (moisture
content—65 to 75%) are taken out
to the agricultural fields of the col-
lective and state farms near Moscow :
after mechanical sludge dewatering
and from the sludge drying beds.
Prospects of Sewerage
Development
In accordance with the General
Plan for development of the city of
Moscow, major trends have been de-
fined in the field of disposal and
treatment of sewage and contaminated
surface run-offs.
The capacity of the Ljuberetskaya
activated sludge plant will be increased
to 2.5 mill mVday.
In 1985 the construction of the
Pakhrinskaya plant will be started
with a capacity of 4 mill mVday.
At present technical and economical
basing is being developed for
complex facilities of sewage sludge
treatment.
By the design year (1990) the prin-
cipal approach to the sewage treatment
technology should change as to the
process, ensuring the reduction of
impurities in final effluents to a level
which would not affect the develop-
ment of natural processes in water
reservoirs.
To meet this requirement the sewage
treatment technology will involve
high rate biological treatment
processes using high MLSS.
Complex technological schemes will
be further developed as well as
methods for treatment of domestic
and industrial wastes, with improved
figures for the quality of treatment,
relating to the removal of organics
with poor biodegradation properties.
On the basis of research work being
conducted, the most economic and
efficient methods will be established
for sewage treatment (physical and
chemical, biological etc.)
136
-------
The teritary treatment must ensure
maximum reduction of all types of
impurities, including organic matter,
nutrients (nitrogen and phosphorus,
oil wastes, synthetic surfactants,
heavy metal salts etc.).
Natural ion exchange materials
(vermiculite, phosphorite, perlite,
diatomaceous earth etc.) will be
widely used for filter beds.
There will be quite a new trend in
the development of the system for
disposal of the contaminated surface
run-off. Provision is made for the
combined system, in which contami-
nated surface run-offs will be conveyed
together with domestic and
industrial wastes.
By 1980 maximum possible extension
of the traditional scheme will be
completed, and after 1980 realiza-
tion of means for the transition to the
sewerage system of deep design will
begin. Provision is made for the
construction of intercepting sewers
of a large diameter, running through
loading centres of the sewerage
areas, with pumping of sewage to
the treatment plants.
Realization of planned specific
measures for introducing the achieve-
ments of science and technology
into the practice of sewage treatment
and sludge treatment will allow to
solve great problems in the field of
control and rational utilization of
resources.
Figure 3.
Scheme for long-term development of Moscow sewerage system.
Water Supply and Sewerage
Administration of the city of
Moscow
List
of slides of sewage treatment
facilities of the city of Moscow
(to paper "Sewage treatment of
the city of Moscow" at the
symposium in the USA)
Slide
Nos.
Slide description
1. Sewerage scheme of the city of
Moscow
2. Diagram "Sewerage develop-
ment of the city of Moscow"
3. Diagram "Sewage treatment at
treatment plants"
4. Figures for degree of sewage
purification
5. Plan of facilities of the Kury-
anovskaya activated sludge
plant
6. Facilities of the I-st unit of the
Kuryanovskaya activated sludge
plant
7. Primary settling tanks of the
I-st unit
8. Digestion tanks of the I-st unit
(view of mechanical dewatering
department)
9. Secondary settling tanks
10. Distribution chamber of the
Novo-Kuryanovskaya activated
sludge plant
11. Screens of the Novo-Kuryanovs-
kaya activated sludge plant
12. Primary settling tanks of the
Novo-Kuryanovskaya activated
sludge plant
13. Aeration tanks of the Novo-
Kuryanovskaya activated sludge
plant
137
-------
14. Effluent outlet-contact channel
15. Control panel of the plant
16. Digestion tanks of the I-st unit
of the Kuryanovskaya activated
sludge plant
17. Digestion tanks of the Novo-
Kuryanovskaya activated sludge
plant
18. Vacuum filters
19. Drum dryers
20. Experimental filters for effluent
polishing
21. Experimental unit of the treatment
facilities
22. Activated sludge plant with a
capacity of 50,000 m'/day
23. Idem
24. Screen building
25. Long-term scheme of sewerage
development
26. "Crystal" plant for treatment of
waste water for automobile
parks
27. Pumping station with a capacity
of 800,000 myday
Protocol
of the fourth Meeting of the USSR
and USA delegations on the problem
of preventing Water Pollution from
Industrial and Municipal Sources
(Washington, D.C., USA, April 3-17,
1977).
In accordance with the Memorandum
of the fifth Meeting of Joint USSR-
USA Commission on Cooperation in
the field of Environmental Protection
(Moscow, November 1976), a
Meeting of the USA-USSR delega-
tions on the problem of waste water
treatment took place from April 3-17,
1977.
The Soviet delegation was led by
Yu. N. Andrianov, Director of the All-
Union Association "Soyuzvodokana-
Iniiproyekt".
The American delegation was led by
Harold P. Cahill, Jr., Director,
Municipal Construction Division, U.S.
Environmental Protection Agency.
The participants of the Symposium
were greeted by John T. Rhett,
Deputy Assistant Administrator for
Water Program Operations, U.S.
Environmental Protection Agency.
A list of participants is attached as
Appendix I.
During the meeting, the following was
accomplished:
1. The Symposium on "Physical-
Mechanical Waste Water
Treatment Facilities".
2. Activities under the 1976 program
of cooperation were discussed.
3. The Working Program for 1978
was coordinated and agreed upon.
1.
Thirteen scientific reports were pre-
sented at the Symposium: the Soviet
delegation delivered six papers; the
U.S. delegation seven.
A list of these papers is attached as
Appendix II.
Of particular interest were the Soviet
papers on waste water treatment in
notation units, centrifuges,1 settling
tanks and on waste water treatment in
the City of Moscow, and American
reports on waste water sedimentation,
filtration, and the development of
combined facilities.
The delegations have mutually
agreed that each side will publish
all the reports presented at the Sym-
posium in the necessary number of
copies in its own language prior to
February 1, 1978, and will distribute
them among interested organizations.
Both sides will exchange ten copies
each of the published Proceedings
of the Symposium.
2.
During the Meeting the specialists
actively discussed the results of current
research conducted in accordance
with the program for cooperation,
and exchanged scientific arid technical
literature.
3.
The delegations determined and
coordinated the joint cooperative work
program for 1978 (Appendix III).
Both sides have agreed that the Fifth
(5th) Symposium entitled, "Recycling
Water Supply Systems and Reuse of
Treated Water at Industrial Plants"
will be held in the USSR September
11-27, 1977.
In preparation for the forthcoming
Symposium, the following was agreed
upon:
• Each side will present 5-7 scientific
reports to the Symposium;
• Both sides will exchange these
report titles prior to July 1, 1977;
• The texts of the reports will be
exchanged in two copies in Russian
and English, prior to August 15,
1977.
138
-------
The delegations noted that it would
be expedient to carry out a long-term
exchange:
• of Soviet specialists in the USA on
the problem of waste water treat-
ment, and
• of American specialists in the USSR
on the problems of waste water
treatment and reuse of treated
water.
MSD Chicago invited two Soviet
specialists to visit for a period of 2-3
months to become familiar with waste
water treatment technology, con-
struction of the facilities and applied
instrumentation. The specialists will be
able to study scientific resarch and
the work projects of various orga-
nizations and likewise actual working
facilities located in other States. In
order to better prepare for the exchange
of specialists, both sides will exchange
proposals and requests for the long-
term exchange by June 1. The detailed
exchange program will be agreed upon
five months prior to the date of the
participants' departure. This exchange
will be carried out on the basis of
equal and "receiving-side-pays" basis.
The final dates and the length of
exchange shall be agreed upon during
this meeting of the delegations in
Moscow in September 1977.
During this visit to the USA, the
Soviet delegation visited significant
industrial and municipal waste water
treatment plants in the Cities of
Schaumberg, Chicago, Illinois; Contra
Costa, California; Richmond, Cali-
fornia; San Francisco, California;
Los Angeles, California; Pascagoula,
Mississippi; and a number of research
institutes in Cincinnati, Ohio: Los
Angeles, San Francisco, California;
and the US Environmental Protection
Agency in the City of Washington,
B.C.
Both sides expressed their satisfaction
that the meeting was conducted on a
highly scientific and technical level
in an atmosphere of friendship and in
a spirit of mutual understanding, thus
contributing to the further development
and strengthening of cooperation in
the field of environmental protection.
This protocol was signed on April 16,
1977, in two copies, in Russian and
English, both texts being equally
authentic.
From the US Side.-
Harold P. Cahill, Jr.
Chief of Delegation
From the Soviet Side:
Yu N. Andrianov
Chief of Delegation
Appendix I
List of Participants from the
Soviet Side
Yu. N. Andrianov
Director
All-Union Association,
"SOYUZVODOKANALNIIPROY-
EKT" GOSSTROY USSR
I. N. Myasnikov
Section Chief
VNII VODEGO Laboratory,
GOSSTROY USSR
I. V. Skirdov
Laboratory Chief
VNII VODGEO Laboratory
GOSSTROY USSR
Yu. L. Maksimenko
Senior Scientific Specialist
State Committee on Science and
Technology and Rational Use of
Natural Resources
P. L Galanin
Chief Engineer
Administration for Moscow Water
and Sewage Facilities
N. V. Pisanko
Chief Engineer
Ukrvodkanalproyekt Institute,
GOSSTROY USSR
List of Participants from the
American Side
John T. Rhett
Deputy Assistant Administrator
Water Programs Operations,
US Environmental Protection Agency
Washington, D.C.
Harold P. Cahill, Jr.
Director
Municipal Construction Division,
US Environmental Protection Agency
Washington, D.C.
Alan Cywin
Senior Science Advisor
US Environmental
Protection Agency
Washington, D.C.
William Lacy
Research and Development, US
Environmental Protection Agency
Washington, D.C.
Isaiah Gellman
National Council of the Paper Industry
for Air and Stream Improvement, Inc.,
Pulp and Paper Industry
New York, New York
David Allen
Project Engineer
Seeded Water Treatment,
Sala Magnetics
Cambridge, Massachusetts
Frank Sebastian
Senior Vice President
Envirotech
Menlo Park, California
Joseph FitzPatrick
Professor
Northwestern University
Evanston, Illinois
John A. Oberteuffer
Sala Magnetics
Cambridge, Massachusetts
James Grutsch
American Oil Company
Chicago, Illinois
Richard Sullivan
Associate Executive Director
American Public Works Association
Chicago, Illinois
139
-------
David G. Stephen
Senior Official
Research and Development, US
Environmental Protection Agency
Cincinnati, Ohio
Francis T. Mayo
Director
Municipal Environmental Research
Laboratory US Environmental
Protection Agency
Cincinnati, Ohio
Ralph H. Sullivan
Program Counsellor
Municipal Construction Division, US
Environmental Protection Agency
Washington, D.C.
Linda Kushner
Program Operations Assistant
Municipal Construction Division, US
Environmental Protection Agency
Washington, D.C.
Andrew Paretti
Consultant
Office of Water Program Operations
US Environmental Protection Agency
Washington, D.C.
Elaine Fitzback
Soviet Project Coordinator
Office of Research and Development
US Environmental Protection Agency
Washington, D.C.
Appendix II
List of Reports
Presented at the USSR-USA
Symposium
"Physical-Mechanical Waste-
water Treatment Facilities"
From the USSR:
Myasnikov, I. N., Ponomarev, V. G.,
Nechaev. A. P., Kedrov, Yu. V.
Wastewater Treatment by Physical and
Mechanical Methods.
Myasnikov, I. N., Gandurina, L. V.,
Butseva, L. N.
Use of Flotation for Waste Water
Treatment.
Skirdov, I. V., Sidorova, I. A.,
Makimenko, Yu, L.
Employment of Microstrainers in
Waste Water Treatment Practice.
Skirdov, I. V.
Improvement of Hydraulic Conditions
of Radial Settling Tanks.
Pisanko, N. V.
Waste Water Treatment in Mining,
Metallurgical and Petrochemical
Industries.
Galauin, P. I.
Sewage Treatment of the City of
Moscow,
From the USA:
Richard Field
Swirl Separation or Flow Regulation
and Solid Separation.
Joseph A. Fitzpatrick
Granular Media Filtration for Tertiary
Application.
John A. Oberteuffer, David M. Allen
SALA-HGMSR Magnetic Filters to
Treat Storm Overflows.
James F. Grutsch
Significant Parameters in Control of
Physical/Mechanical Treatment of
Refinery Waste Water.
Frank Sebastian
Pyrolysis Applications for Industrial
and Municipal Treatment.
Isaiah Gellman
Current Status and Directions of
Development of Physical/Mechanical
Effluent Treatment in the Paper
Industry.
William J. Lacy
Physical Treatment of Oil Refinery
. Waste Water.
140
-------
Appendix III
PROGRAM
USSR-USA Cooperation of Working Group on Prevention of Water Pollution from Industrial and Municipal Sources
NO Title
Modernization of existing and development
of new combined facilities with high
efficiency for wastewater treatment,
including hydrocyclones, multi-stage
settlers, flotators, facilities with utilization
of technical oxygen, investigations of usage
of fjoccuiants and coagulants.
— Development of hydrocyclones
and pressure flotation facilities
— Development of tubular and plate
settlers.
— Development of enriched oxygen
systems for aeration.
— Development of pure oxygen system
aeration.
— Development of filters with mixed sand
and gravel media.
— Development of multi-media filters
and equipment for continuous washing.
Form of Work
Joint development
of themes, scientific
information and
specialists
delegation exchange.
Responsible for
From the From the
USSR USA Time
VNU VODGEO
GOSSTROY
USSR
EPA 1980
Expected Results
Improvement of the efficiency
of existing and development
of new treatment facilities,
reductions of space for
location, reduction of re-
agents and cost price of waste
water treatment.
Symposium on "The
Use of Advanced
Facilities and Equip-
ment for Waste
water Treatment
(USA, Cincinnati,
Ohio,
April 2-16, 1978;
eight specialists)
Symposium on theme:
"Experiences with
Electro-Chemical
Wastewater Treat-
ment Methods and
their Further
Development."
(USSR, Moscow,
September 10-24,
1978, eight
specialists).
VNII VODGEO 1980 Recommendations on design-
ing hydrocyclones and
pressure flotation facilities.
EPA 1980 Recommendations on applying
settling tanks in waste water
treatment.
EPA 1980 Development of open
aeration tanks using technical
oxygen.
VNII VODGEO 1980 Development of covered
aeration tanks using pure
oxygen.
VNII VODGEO 1979 Recommendations on de-
signing filters for wastewater
treatment and final treatment.
EPA 1979 Recommendations on develop-
ment of multimedia filters.
141
-------
2. Advanced technological wastewater treat- Information and dele-
ment processes in petrochemical, chemical, gation exchange
petroleum refining, pulp and paper
industries.
— Advanced technological
wastewater treatment processes in
petroleum refining industries.
— Advanced technological wastewater
treatment processes in petrochemical
and pulp and paper industries.
Development of the best available
technology and facilities for removal of
nutrients and treatment of municipal
wastewaters; the industrial recycle and re-
use of the treated wastewaters.
— Development of methods and facilities for
removal of nitrogen compounds.
— Development of optimum schemes of
facilities for removal of nutrients.
Joint development of
themes, information
and delegation
exchange
4. Treatment of wastewater sludges and
residuals.
Stabilization and dewatering of
wastewater sludges and residuals.
Technology and facilities for heat treat-
ment and utilization of wastewater sludges
and residuals.
Information and
delegation exchange
VNII VODGEO EPA 1979 Upgrading wastewater treat-
GOSSTROY ment efficiency of existing
USSR treatment plants, introduction
of new treatment schemes,
maximum reuse of treated
effluents. ''
VNII VODGEO 1979 Development of'flow dia-
grams for treatment system
using mechanical, physical-
chemical and bio-chemical
methods applicable to the
petroleum refining industry.
1979 Development of flow dia-
grams for treatment systems
using chemical additions for
the pulp and paper and petro-
chemical industries.
VNII VODGEO EPA 1980 Development of new treat-
GOSSTROY ment facilities for prevention
USSR of entrophication, and de-
velopment of new advanced
treatment systems. Aimed at
closed loop systems, with by
product recovery by industry.
VNII VODGEO Development of recommenda-
tions on designing facilities.
EPA Development of recommenda-
tions on designing facilities.
VNII VODGEO EPA 1980 Reduction of cost of sludge
GOSSTROY and other residuals treatment,
USSR increasing of overall treatment
facilities efficiency.
VNII VODGEO Recommendations on design-
ing facilities for stabilization
and dewatering of waste-
water sludges and residuals.
Recommendations on design-
ing facilities for heat
treatment and utilization of
wastewater sludges and
residuals.
142
-------
Appendix IV
Itinerary
Appendix V
For the Visit of the USSR Gosstroy Delegation to the U.S.
on questions of prevention of water pollution from industrial
and municipal effluents from April 3-17, 1977.
Sunday, April 3
Monday, April 4
Tuesday, April 5
Wednesday, April 6
Thursday, April 7
Friday, April 8
Saturday, April 9
Sunday, April 10
Monday, April 11
Tuesday, April 12
Wednesday, April 13
Thursday, April 14
Friday, April 15
Saturday, April 16
Sunday, April 17
Arrive in New York City
Sightseeing in New York
Symposium
Symposium
Cincinnati Lab Tour. Leave for
Schaumberg, Illinois
Schaumberg, Illinois (H. J. Heinz
Co.). Tour of Establishment for
Treating Effluent From Food
Industry. Visit to John E. Egan
Water Reclamation Plant
San Francisco, California—Sight-
seeing
Sightseeing in San Francisco
Visit Contra Costa Sewage Treat-
ment Plant. Tour of Treatment
Plants at a Petroleum Refinery.
Leave for Los Angeles, California
Arrive in Los Angeles, California.
Visit to Jet Propulsion Laboratories
on Sediment Pyrolysis, in the
Orange County Area. Visit to
Disneyland
Travel Day. Leave for Pascagoula,
Mississippi
Arrive in Pascagoula, Mississippi.
Pet Food & Quaker Oats. Also
Pascagoula Municipal
Treatment Plant. Familiarization
with the Process of Effluent Treat-
ment in Feed Processing Enter-
prises.
Leave for Washington, D.C.
Washington, D.C. (Shopping and
Tour of D.C.). Signing the Proto-
col. Final Meeting. Leave for
New York City
New York City
Depart for USSR
Symposium Program
Tuesday, April 5
8:30 a.m. Registration
Opening Remarks—John T. Rhett
Welcome—Dr. David G. Stephan
9:00 a.m.
9:15a.m.
9:30a.m.
Sebastian, F. P., Lachtman, D. S.,
Kroneberger, G. K., Allen, T. D.,
(Envirotech), Pyrolysis Applications for
Industrial and Municipal Treatment
10:15a.m. Break
10:30 a.m. Myasnikov, I. N., Ponomaryev, V. G.,
Nechaev, A. P., and Kedrov, Yu. V., {VNII
Vodgeo, Gosstroi USSR), Wastewater
Treatment by Physical and Mechanical
Methods
11:15a.m. Lacy, William, (US EPA), Physical Treat-
ment of Oil Refinery Wastewater
12:00 noon Discussion
12:15a.m. Lunch
1:30p.m. Skirdov, 1. V., (Vodgeo), Improvement of
Hydraulic Conditions of Radial Settling Tanks
2:15 p.m. Gellman, Isaiah, (NCASI, Pulp & Paper
Industry, New York, New York), Current
Status and Directions of Development of
Physico-Mechanical Effluent Treatment in
the Paper Industry
3:00 p.m. Break
3:15 p.m. Skirdov, I. V., Sidorova, I. A., Maksimenko,
Yu. L., (USSR) Employment of Micro-
strainers in the Wastewater Treatment
Practice
4:00 p.m. Grutsch, J. F., (American Oil Company,
Indiana), The Control of Refinery Mechani-
cal Wastewater Treatment Processes by
Controlling the Zeta Potential
4:45 p.m. Discussion
5:00 p.m. Adjourn
143
-------
Wednesday, April 6
9:00 a.m. Opening Remarks—Harold Cahill
9:15 a.m. Address—Francis T. Mayo
9:30a.m. Oberteuffer, J. A. and Allen, D. M. (Sala
Magnetics, Cambridge, Mass.), Combined
Storm Overflow Treatment with
Sala-HGMF® High Gradient Magnetic
Filters
10:15 a.m. Break
10:30 a.m. Myasnikov, I. N., Gandurina, L. V., Butseva,
L. N., (USSR), Use of Flotation for
Wastewater Treatment
11:15 a.m. FitzPatrick, J. A., and Swanson, C. L.,
(Northwestern University, Evanston, Illinois),
Performance Tests on Full-Scale Tertiary
Granular Filters
12:00 noon Discussion
12:15 p.m. Lunch
1:30 p.m. Pisanko, N. V., (Ukrvodokanalproekt
Institute, USSR) Sewage Treatment in Min-
ing, Metallurgical and Oil-Chemical Industries
2:15 p.m. Fields, R., (US EPA), The Swirl Concen-
trator for Treating and Regulating Sewered
(Separate and Combined) and Unsewered
Flows
3:00p.m. Break
3:15 p.m. Galanin, P. L, (USSR), Sewage Treatment
of the City of Moscow
4:00 p.m. Discussion
4:15 p.m. Adjourn
-------
EP 600/9 EPA
77-504 Water Prog. Oper. Ofc.
AUTHOR
Symposium on Physlcal-Mechani.^
TITLE cal treatment of waste
waters
DATE DUE
QAVLORD 4tt
BORROWER'S NAME
-------
DATE DUE
BORROWER'S NAME
-------
"N-
\
DATE DUE
-------
u
LU
o
W LU
LU I-
LU O
o
<
_ < •* =
si*a
LU S 5
co 9
1
LU
H
<
P
°
^S
UJ ^
w WW ^ UJ
11 S§
m c
i£
S a:
00
Sf:
ot
<
z
LU
Q.
Z
------- |