&EPA
United States
Environmental Protection
Agency
Office of Research and
Development
Washington, DC 20460
EPA/600/9-90/036
August 1990
Proceedings -Twelfth
United States-Japan
Conference on Sewage
Treatment Technology
October 12-13, 1989
Cincinnati, Ohio
-------
-------
EPA/600/9-90/036
August 1990
PROCEEDINGS
TWELFTH UNITED STATES-JAPAN CONFERENCE
ON
SEWAGE TREATMENT TECHNOLOGY
OCTOBER 12-13, 1989
CINCINNATI, OHIO
RISK REDUCTION ENGINEERING LABORATORY
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
CINCINNATI, OHIO 45268
-------
DISCLAIMER
These Proceedings have been reviewed by the U.S. Environmental Protection
Agency and approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the U.S. Environmental
Protection Agency, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
-------
FOREWORD
The maintenance of clean water supplies and the
management of municipal and industrial wastes are vital
elements in the protection of the environment.
The participants in the United States-Japan cooperative
project on sewage treatment technology have completed their
Twelfth Conference. These conferences, held at 24-month
intervals, have given the scientists and engineers of the
cooperating agencies an opportunity to study and compare the
latest practices and developments in the United States and
Japan. These Proceedings of the Twelfth Conference comprise
a useful body of knowledge on sewage treatment, which will be
available not only to Japan and the United States but also to
all nations of the world who desire it.
William K. Rei
Administrator
Washington, D.C.
-------
-------
TABLE OF CONTENTS
Foreword. •
List of Japanese Delegates and Presentation Topics
List of United. States Delegates and Presentation Topics
Joi nt Communique
Japanese Papers f
United States Papers
VI
v i i i
1
3
5
-------
JAPANESE DELEGATION
Delegates
Dr. Ken Murakami (Head of Delegation)
Director, Regional Sewerage Division
Sewerage & Sewage Purification Dept.
City Bureau
Ministry of Construction
Mr. Taigo Matsui
Director, Water Quality Control Dept.
Public Works Research Institute
Ministry of Construction
Mr. Shunsoku Kyosai
Research Coordinator for
Wastewater Treatment
Water Quality Control Dept.
Public Works Research .Institute
Ministry of Construction
Mr. Eiichi Nakamura
Chief, Water Quality Section
Water Quality Control Dept.
Public Works Research Institute
Ministry of Construction
Dr. Kazuhiro Tanaka
Director, Research & Technology
Development Division
Japan Sewage Works Agency
Mr. Masanobu Aoki
Deputy Director
Research & Technology Development Division
Japan Sewage Works Agency
Mr. Kazuo Hoya
Director, Engineering Division
Regional Sewerage Center
Sewerage Bureau
Tokyo Metropolitan Government
Presentation Topics
New Movement in Sewerage
Construction Projects in Japan
Renovation of an Extended
Aeration Treatment Plant for
Biological Nutrient Removal
Process
Innovative Wastewater Treatment
Processes - Interim Results of
Biofocus WT Project
Toxicity Monitoring Method by
Biosensor
Innovative Sludge Handling System
Using Pelletization for
Simultaneous Thickening and
Conditioning
Development of a High Efficiency
Nitrogen Removal System Using
Immobilized Nitrifiers in
Synthetic Resin
Fundamental Study on Application
of High Gradient Magnetic
Separation to Municipal
Wastewater Treatment System
VI
-------
JAPANESE DELEGATION (continued)
Mr. Senji Kaneko
Chief, Research & Development Division
Sewage Works Bureau £
Yokohama City
Mr. Hirdshi Oniki
Director, Planning Department
Sewage Works Bureau
Fukupka City
Operation of Centralized Sludge
"Treatment Plant
Development of Sludge Treatment
Technology in Biological
Phosphorus Removal Process
Dr. Takeshi Kubo (Advisor)
Counselor, Japan Sewage Works Agency
VI 1
-------
UNITED STATES DELEGATION
Delegates
Mr. John J. Convery (Head of Delegation)
Deputy Director
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Mr. Dolloff F. Bishop
Chief, Treatment Assessment Branch
Water & Hazardous Waste Treatment Research Division
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency .>
Presentation Topics
Dr. William C. Boyle
Dept. of Civil & Environmental Engineering
University of Wisconsin-Madison
Mr. William F. Brandes
Permits Division
Office of Water Enforcement & Permits
U.S. Environmental Protection Agency
Performance of Fine Pore
Aeration Systems in Process
Water
Current Status and Future
Direction of Organic Toxics
Control in the Water
Environment - Discharge
Regulations and Stream
Standards
Mr. Richard C. Brenner
Environmental Engineer
Treatment Assessment Branch
Water & Hazardous Waste Treatment Research Division
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Mr. Donald S. Brown
Environmental Engineer
Municipal Wastewater Branch *
Water & Hazardous Waste Treatment Research Division
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
In-Vessel Composting of
Municipal Wastewater Sludge
Mr. Randall J. F. Bruins
Environmental Scientist.
Environmental Criteria & Assessment Office
U.S. Environmental Protection Agency
The Development of Risk
Assessment Methodologies
for Use in Regulation of
Sewage Sludge Disposal
-------
UNITED STATES DELEGATION (continued)
Dr. Carl A. Brunner.
Acting Chief, Hazardous Waste Treatment Branch
Water & Hazardous Waste Treatment Research Division
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Dr. Arthur J. Condren
James M. Montgomery Consulting Engineers, Inc.
Dr. 'Richard A. Dobbs
Research Chemist
Treatment Assessment Branch
Water & Hazardous Waste Treatment
Research Division
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Dr. Rahesh Govind
Dept. of Chemical and Nuclear Engineering
University of Cincinnati
A Preliminary Assessment of
High Biomass Systems
Partitioning of Toxic
Organic Compounds Between
Wastewater and Wastewater
Solids
Prediction of Biodegradation
Kinetics of Toxic Organic
Compounds
Dr. James A. Heidman
Sanitary Engineer
Municipal Wastewater Branch
Water & Hazardous Waste Treatment Research Division
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Wastewater Branch
Treatment
Mr. James F. Kreissl
Acting Chief, Municipal
Water & Hazardous Waste
Research Division
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Dr. Vincent P. Olivieri
Department of Geography and
Environmental Engineering
The Johns Hopkins University
Dr. 0. Karl Scheible
Hydro-Qual, Inc.
Experience with Intrachannel
Clarifiers in the U.S.A.
Levels and Removal of
Selected Indicator and
Pathogenic Microorganisms
During Conventional -
Anaerobic Sludge Digestion
UV Disinfection
-------
UNITED STATES DELEGATION (continued)
Mr. Henry H. Tabak
Research Chemist
Treatment Assessment Branch
Water & Hazardous Waste Treatment
Research Division
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Biodegradability and
Biodegradation Kinetics
Assessment for Toxic
Organic Compounds Using
Oxygen Uptake Measurements
Dr. Albert D. Venosa
Microbiologist
Municipal Wastewater Branch
Water & Hazardous Waste Treatment Research Division
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
-------
UNITED STATES AND JAPANESE DELEGATES TO THE
TWELFTH CONFERENCE
DR. KEN MURAKAMI, HEAD OF THE JAPANESE DELEGATION, AND
MR. JOHN CONVERY, HEAD OF THE U.S. DELEGATION, MAKING
OPENING REMARKS AT THE CONFERENCE
XI
-------
r
DELEGATES OBSERVING PILOT ROTATING BIOLOGICAL CONTACTOR STUDIES AT
THE USEPA TEST AND EVALUATION FACILITY, CINCINNATI, OHIO
DELEGATES OBSERVING MECHANICAL COMPOSTING OF SLUDGE AT HAMILTON, OHIO
xn
-------
WASTEWATER REUSE PLANT OPERATION AT
DENVER, COLORADO, BEING DESCRIBED TO DELEGATES
DELEGATES OBSERVING SEWER RELINING AT LOS ANGELES COUNTY, CALIFORNIA
-------
DELEGATES OBSERVING THE
COLLECTION AND EVALUATION
OF MARINE LIFE NEAR SEWAGE
OUTFALLS, LOS ANGELES
COUNTY, CALIFORNIA
xiv
-------
JOINT COMMUNIQUE
Twelfth United States/Japan Conference on
Sewage Treatment Technology
Friday, October 13, 1989
1. The Twelfth United States/Japan Conference on Sewage Treatment Technology
was held in Cincinnati, Ohio, October 12 and 13, 1989.
2. The US Delegation headed by Mr. John J. Convery, Deputy Director, Risk
Reduction Engineering Laboratory, U.S. Environmental Protection Agency
(USEPA), Cincinnati, Ohio, was composed of twelve representatives from the
USEPA, two representatives from consulting engineering companies, and three
representatives from the universities.
3. Dr. Ken Murakami, Director, Regional Sewerage Division, Sewerage and
Sewage Purification Department, Ministry of Construction was the Head of the
Japanese Delegation, which consisted of four National Government
representatives, three Japan Sewage Works Agency representatives, and three
representatives from'the Tokyo Metropolitan Government, Yokohama City and
Fukuoka City.
4. During the Conference, papers related to toxics control in municipal
wastewater, and sludge treatment and disposal technologies.were presented by
both sides. Additional presentations were made by the Japanese side
concerning sewage construction projects and innovative wastewater treatment
processes including nutrient removal. The US side, too, presented additional
papers related to aeration methods, pathogens and disinfection, and innovative
treatment. Data and findings were useful to the development of sewage
treatment technology for each country. .
5. Field visits in Chicago, IL, -Denver, CO, and Los Angeles, CA are planned
to inspect sewers and wastewater treatment facilities.
6. A proceedings of the Twelfth Conference shall be printed in English and
Japanese.
7. Regarding the operation of the Sewage Treatment Technology Project under
the United States/Japan Environmental Cooperation Agreement (hereinafter
referred to as the Project), both sides agreed that the Project be amended
following .this conference to have a workshop for technology exchanges and
visits in either country by sending one .or two representatives to the host
country. Each side would host the workshop at least once every two years
alternatively..
-------
JOINT COMMUNIQUE (continued)
8. Both sides agreed that engineer exchanges and the cooperative research
program be continued as before.
9. Summaries of the research projects concerning sewage treatment technology
of each country shall be exchanged every year.
10. A progress report shall be prepared each year by both the US and Japan to
summarize briefly the activities of the Project. The progress reports' shall
be exchanged at the time of the workshops discussed in paragraph 7.
11. The Co-chairmen'of the Project shall be the Branch Chief responsible for
municipal wastewater research, Water and Hazardous Waste Treatment Division,
Risk Reduction Engineering Laboratory, USEPA for the US side and Director,
Sewerage and Sewage Purification Department, Ministry of Construction for the
Japanese side. . ,
12. It was proposed by the Japanese side that the forthcoming workshop based
on paragraph 7 of this Joint Communique shall be held in Tsukuba around May
1990. - v
-------
JAPANESE PAPERS
INNOVATIVE WASTEWATER TREATMENT PROCESSES - INTERIM RESULTS OF
BIOFOCUS WT PROJECT
Shunsoku Kyosai, Research Coordinator for Wastewater Treatment,
Yutaka Suzuki, Masahiro Takahashi, Katsumi Moriyama, Masashi
- Ogoshi, Toshiyuki Nakajima, Takashi Masuda, Suuichi Ochi, and
Hiroaki Tanaka, Water Quality Control Department, Public Works
Research Institute, Ministry of Construction
FUNDAMENTAL STUDY ON APPLICATION OF HIGH GRADIENT MAGNETIC
SEPARATION TO MUNICIPAL WASTEWATER TREATMENT SYSTEM
, Kazuo Hoya, Director, Engineering Division, Regional Sewerage
Center, Yasuo Tanaka, Ph.D., Researcher of Research and
Technological Development Section, Planning Division, and Khouichi
Ishida, Asst. Head of Research and Technology Development Section,
Planning Division, Sewerage Bureau, Tokyo Metropolitan Government
59
NEW MOVEMENT IN SEWERAGE CONSTRUCTION PROJECTS IN JAPAN
Dr. Ken Murakami, Director, Regional Sewerage Division,
Sewerage and Sewage Purification Department, City Bureau,
Ministry of Construction
INNOVATIVE SLUDGE HANDLING SYSTEM USING SLUDGE PELLETIZATION
FOR SIMULTANEOUS THICKENING AND CONDITIONING
Dr. Kazuhiro Tanaka, Director, Shuzo Koike, Research Engineer,
Research and Technology Development Division, Japan Sewage Works
Agency, and Naohiro Taniguchi, Director, General Planning Section
of General Planning Division, City Planning Bureau, Tokyo
Metropolitan Government
,101
,119
OPERATION OF CENTRALIZED SLUDGE TREATMENT PLANT
Senji Kaneko, Chief, Research and Development Division,
Sewage Works Bureau, Yokohama City
DEVELOPMENT OF SLUDGE TREATMENT TECHNOLOGY IN BIOLOGICAL PHOSPHORUS
REMOVAL PROCESS
Hiroshi Oniki, Director, Planning Department, Sewage Works
Bureau, Fukuoka.City
,157
,215.
TOXICITY MONITORING METHOD BY BIOSENSOR
Eiichi Nakamura, Chief, and Hiroaki Tanaka, Senior Research
Engineer, Water Quality Section, Public Works Research Institute,
Ministry of Construction
.239-
-------
JAPANESE PAPERS (continued)
RENOVATION OF AN EXTENDED AERATION TREATMENT PLANT FOR BIOLOGICAL
NUTRIENT REMOVAL PROCESS i.
Taigo Matsui, Director, K. Sato, K. Moriyama, Water Quality
Control Department, Public Works Research Institute, Ministry
of Construction, M. Imamichi and Y. Yarada, Sewage Works
Division, Hamamatsu City . «.
DEVELOPMENT OF A HIGH EFFICIENCY NITROGEN REMOVAL SYSTEM USING
IMMOBILIZED NITRIFIERS IN SYNTHETIC RESIN
Masanobu Aoki, Deputy Director,-Minoru Tada, Deputy Director,
Takashi Kimata, Shouji Harada, and Yuhko Fujii, Research and
Technology Development Division, Japan Sewage Works Agency
,253
.265
-------
UNITED STATES PAPERS
CURRENT STATUS AND FUTURE DIRECTION OF ORGANIC'TOXICS CONTROL IN
THE WATER ENVIRONMENT - DISCHARGE REGULATIONS AND STREAM STANDARDS 295
William F. Brandes, and Catherine Crane, Permits Division,
Office of Water Enforcement and Permits, U.S. EPA, Washington, DC
PERFORMANCE OF FINE PORE AERATION'SYSTEMS IN PROCESS WATER - 303
Dr. William C. Boyle, Professor, Department of Civil and
• Environmental Engineering, University of Wisconsin, Madison,
' Wisconsin; Richard C. Brenner, Environmental Engineer, Water
and Hazardous Waste Treatment Research Division, Risk Reduction
Engineering Laboratory, U.S. EPA, Cincinnati, Ohio; James J. Marx,
Consulting Engineer, Donohue and Associates, Milwaukee, Wisconsin;
and Thomas C. Rooney, Consulting Engineer, Waukesha, Wisconsin
BIODEGRADABILITY AND BIODEGRADATION KINETICS ASSESSMENT FOR TOXIC
ORGANIC COMPOUNDS USING OXYGEN UPTAKE MEASUREMENTS 325
Henry H. Tabak, Research Chemist, Water and Hazardous Waste
Treatment Research Division, Risk Reduction Engineering
Laboratory, U.S. EPA, Cincinnati, Ohio; Senjay Desai, Chemical
Engineer, and Dr. Rakesh Govind, Professor, Department of Chemical
and Nuclear Engineering, University of Cincinnati, Cincinnati,
Ohio „• '
PREDICTION OF BIODEGRADATION KINETICS OF TOXIC ORGANIC COMPOUNDS I..383
Sanjay Desai, Chemical Engineer, Dr. Chao Gao, Visiting Chinese
Scholar, Dr. Rakesh Govind, Professor, Department of Chemistry
and Nuclear Engineering, University of Cincinnati, Cincinnati,
Ohio; and Henry H. Tabak, Research Chemist, Water and Hazardous
Waste Treatment Research Division, Risk Reduction Engineering
Laboratory, U.S. EPA, Cincinnati, Ohio
PARTITIONING OF TOXIC ORGANIC COMPOUNDS BETWEEN WASTEWATER AND
WASTEWATER SOLIDS ' 401
Dr. Richard A. Dobbs, Research Chemist, Water and
Hazardous Waste Treatment Research Division, Risk Reduction
Engineering Laboratory, U.S. EPA, Cincinnati, Ohio
-------
UNITED STATES PAPERS (continued)
LEVELS AND REMOVAL OF SELECTED INDICATOR AND PATHOGENIC MICROORGANISMS
DURING CONVENTIONAL ANAEROBIC SLUDGE DIGESTION : 425
Dr. Vincent P. Olivieri, Professor, Lynne Cox, and Mohammed Sarai,
Department of Geography and Environmental Engineering, The Johns
Hopkins University, Baltimore, Maryland; Dr. Jan L. Sykora,
Professor, and Patrick Gavagahn, University of Pittsburgh,
Pittsburgh, Pennsylvania
IN-VESSEL COMPOSTING OF MUNICIPAL WASTEWATER SLUDGE .. 441
Donald S. Brown, Environmental Engineer, Water and Hazardous
•Waste Treatment Research Division, Risk Reduction Engineering
Laboratory, U.S. EPA, Cincinnati, Ohio; John Johnston, Associate
Professor, California State University-Fresno, Fresno, California;
and Leslie Bayer, Eastern Research Group, Arlington, Massachusetts
THE DEVELOPMENT OF RISK ASSESSMENT METHO*DOLOGIES FOR USE IN
REGULATION OF SEWAGE SLUDGE DISPOSAL 475
Randall J. F. Bruins, Physical Scientist, Dr. Norman E. Kowal,
Research Medical Officer, Cynlthia Sonich-Mullin, Environmental
Health Scientist, and Steven D. Lutkenhoff, Physiologist,
Environmental Criteria and Assessment Office, U.S. EPA, Cincinnati,
Ohio
UV DISINFECTION 1
Dr. 0. Karl Scheible, HydroQual, Inc., Mahwah, New Jersey
.485
,515
A PRELIMINARY ASSESSMENT OF HIGH BIOMASS SYSTEMS
Dr. Arthur J. Condren, James M. Montgomery Consulting
Engineers, Inc., Pasadena, California; Dr. Bjorn Rusten,
Aquateam Norwegian Water Technology Centre AS, Oslo, Norway;
and Dr. James A. Heidman, Sanitary Engineer, Water and Hazardous
Waste Treatment Research Division, Risk Reduction Engineering
Laboratory, U.S. EPA, Cincinnati, Ohio
EXPERIENCE WITH INTRACHANNEL CLARIFIERS IN THE U.S.A...: 531
James F. Kreissl, Environmental Engineer, Water and Hazardous
Waste Treatment Research Division, Risk Reduction Engineering
Laboratory, U.S. EPA, Cincinnati, Ohio; Dr. A. T. Wallace,
Professor, College of Engineering, University of Idaho, Moscow,
Idaho; and Jon H. Bender, Environmental Engineer, Technical Support
Division, Office of Drinking Water, Office of Water, U.S. EPA,
Cincinnati, Ohio
-------
INNOVATIVE WASTEWATER TREATMENT PROCESSES
- INTERIM RESULTS OF BJOFOCUS WT PROJECT -
by
Shunsoku Kyosai, Yutaka Suzuki, Masahiro Takahashi,
Katsumi Moriyama, Masashf Ogoshi, Toshiyuki Nakajima,
Takashi Masuda, Shuuichi Ochi, Hiroaki Tanaka
Water Quality Control Department
Public Works Research Institute
Ministry of Construction
Asahi, Tsukubashi, Ibarakiken 305 Japan
The work described in this paper was not funded by
the U.S. Environmental Protection Agency. The contents
do not necessarily reflect the views, of the Agency and
no official endorsement should be inferred.
Prepared for Presentation at:
12th United States/Japan Conference
on
Sewage Treatment Technology
October 1989
Cincinnati, Ohio
-------
1. PREFACE
The Ministry of Construction of the Government of Japan started a five-
year research project, "The Development of New Wastewater Treatment Systems
Employing Biotechnology", which is commonly called Biofocus WT, in ffscal year
1985 as one of the comprehensive research and development projects of the
Ministry.
By broadly applying biotechnology, which is a technology that utilizes
the life maintenance functions of biological organisms, to wastewater treat-
ment, Biofocus WT intends to solve the present and future problems in waste-
water treatment listed below.
a. lower energy requirement and lower 0 & M costs
b. smaller land requirement for wastewater treatment facilities
c. better effluent quality
d. recovery of valuable resources from wastewater treatment
e. easier operation & maintenance of wastewater treatment facilities
f. discovery of useful microorganisms for wastewater treatment
The Public Works Research Institute and the Building Research Institute
of the Ministry of Construction -are the'main organizations carrying out Biofo-
cus WT. The former is mainly in charge of matters related to publicly owned
wastewater treatment plants (POWTP), while the latter oversees matters related
to household wastewater treatment tanks.
Table 1.1 shows the subjects of research and development in Biofocus WT
related to POWTP. This, paper presents the interim results of Biofocus WT
being carried out'by the Public Works Research Institute. :
Table 1.1 Subjects of Research and Development in Biofocus WT
a. Study on Microorganism Bank for Wastewater Treatment
b. Study on Application of Genetic Engineering to.Microorganisms for
Wastewater Treatment
*c. Study on Immobilization Methods of Microorganisms for Wastewater
Treatment
d. Development of Bioreactors for Wastewater Treatment
e. Development of Bioreactors for Sludge Treatment
f. Development of Solid-Liquid Separators for Raw Wastewater
g. Development of Biosensors for Water Quality Measurement
h. Development of New Wastewater Treatment Systems
-------
2. STUDY ON MICROORGANISM BANK FOR WASTEWATER TREATMENT '
2-1 OUTLINE OF MICROORGANISM BANK
Recently, environmntal microbiology has achieved great progress. Though
there is much microbiological information useful for wastewater systems, most
sanitary engineers can not fully utilize such information. There may be
several reasons for this situation: (1) The results of microbiology research
are too difficult for sanitary engineers to utilize. (2) The esults are too
specialized in a certain field. (3) Sanitary engineers can not obtain the
information easily.
A microorganism bank for wastewater treatment has been investigated in
or'der to-apply the results of microbiology research to wastewater engineering.
There are several key points for establishing such a bank. A technical opera-
tion strategy is one of the most important issues as well as a financial
operation strategy. Development of the technical operation strategy would be
the result of this research.
The bank deals with the microorganisms which are principally related to
wastewater systems. A unique feature of the bank is that it consists of two
p'arts, a data bank and a preservation bank. The data bank accumulates data
sheets in which microbiological, and ecological data on microorganisms related
to wastewater systems are summarized. The preservation bank is an ordinary
one. It wi.ll not collect every strain listed in the data bank, but only those
which are not kept at other places. The preservation places where the strains
.are kept are listed in the data bank. •
2-2 DATA BANK
Most microbiological data on any individual strain are obtained from
pure cultures. However, most microbiological phenomena in biological waste-
water treatment processes are observed in mixed cultures of bacteria, protozoa
and other microorganisms. The results of the pure cultures are usually dif-
ferent from the phenomena in the mixed cultures. This understanding impedes
of microbiological information for wastewater treatment sys-
bank would reconcile this situation by providing the ecologi-
of each strain. It also employs special classification of
All strains are classified by their relationship to waste-
A tentative classification is shown in Table 2.1.
the utilization
terns. "The data
cal properties
mi croorgan i sms.
water systems.
Table 2.2 is the data sheet from the data bank.
information as well as ordinal microbiological data.
water treatment systems are also summarized.
It contains ecological
Data concerning waste-
Table 2.3 shows the list of bacteria species which have 'been registered
in the bank. These species were collected from papers, or were obtained
directly from researchers.
-------
Table 2.1 Tentative Classification of Microorganisms
in Relation to Wastewater Treatment
WASTEWATER TREATMENT
•Removal ,of Biodegradable Organics
Removal of Refractory Organics
Removal of Synthetic Hazardous Organics
Nitrification
Denitrification
Phosphorus Removal
Iron, Manganate Removal
Removal of Trace Heavy Metals
Removal of. Soluble Inorganic Materials
Oxidation of Sulfur Compounds
Reduction of Sulfur Compounds
Sedimentation of Suspended Solids
Flotation of Suspended Solids
Liquefaction of Suspended Solids
Odor Control
Color Removal
Epidemic Risk Control
SLUDGE TREATMENT AND DISPOSAL
Liquefaction of Sludge
Volatilization of Sludge
Heavy Metal Removal from Sludge
Composting
Resource Recovery from Sludge
Table 2.3 List of Strains
Strains
Characteristics
Nitrosomonas europaea
Nitrobacter sp.
Thiothrix sp.
Beggiatoa sp.
Thlosheara pantotropha GB17
Rhodopseudomonas sp.
Rhodospirill urn rubrum
Flavobacterium sp., or
Pseudomonas sp.
Alcallgenes eutrophus H850
Aclntobacter sp. GJ70
Rhodococcus sp. TTD-1
Acintobacter calocaeticus
SH-1
Pseudomonas sp. K172
Pseudomonas sp. S100
Nitrification
Nitrification
Sulfur Oxidation, Bulking
Sulfur Oxidation, Bulking
Sulfur Oxidation
Removal of Organic Material
Removal of Organic Material
Removal of Synthetic Hazardous Organic Material
Removal of Synthetic Hazardous Organic Material
Removal of Synthetic Hazardous Organic Material
Removal of Synthetic Hazardous Organic Material,
Odor Control, Sulfur Oxidation .
Removal of Synthetic Hazardous Organic Material
Removal of Synthetic Hazardous Organic Material,
Denitrification
Removal of Synthetic Hazardous Organic Material,
Denitrification
10
-------
Table 2.2 Data Sheet of Data Bank
1. CODE
(1) Bank Code
(2) Scientific Classification: 1) Family 2) Species- 3) Common Name
2. RELATION TO WASTEWATER TREATMENT
(1) Functional Name
(2) Relationship to Wastewater Treatment
3. DATA SHEET WRITER
(1) Name
(2) Affiliation
(3) Date of Register
4. RESEARCHER
(1) Name ' " .
(2) Affiliation
5. PRESERVATION
(1) -.Organization
(2) Code Number in the Organization
6. ORIGIN OF .STRAIN
(1) Purification from Natural Environment
a) Purification Data: Source of Strain, Date of Purification,
Researchers, Affiliation
b) Identification Data: Researchers, Affiliation
(2) Genetic Engineering . .
' a) Applied Genetic Engineering Method
b) Conditions: Host, Vector, Other
- c) Characteristics of Strain
(3) Other
a) Obtained from Other Researcher
b) Characteristics of Strain
•7. CHARACTERISTICS OF STRAIN
(1) Physiological Characteristics: Growth Rate, Substrate, Products,
Genetic Data, so forth
(2) Ecological Characteristics: Environmental Condition for Growth,
Competition, Symbiosis, so forth
8. NOTES FOR EXPERIMENT
Epidemic Notes, Hazardous Products, Patents, so forth
9. CONDITION FOR PURIFICATION .
(1) Cumulative Cultivation: Temperature, Media, Method
(2) Purification: Temperature, Media, Method
10. PRESERVATION OF VIABLE CELL: Temperature, Media, Subculturing Method
1.1. PRESERVATION AND REACTIVATION OF FREEZING CELL
(1) Preservation: Temperature, Media, Period of Preservation
(2) Reactivation: Temperature, Media, Reactivation of Specified Nature
12. RELATED LITERATURE '
11
-------
3. STUDY ON APPLICATION OF GENETIC ENGINEERING TO MICROORGANISMS FOR
WASTEWATER TREATMENT
3-1 ECOLOGICAL AND EVOLUTIONARY STABILITY OF.ENGINEERED MICROORGANISMS
3-1-1 INTRODUCTION
ft
There are various technical problems in the application of engineered
microorganisms to wastewater treatment systems. From the view point of sta-
bility of useful genetic traits or genetic information of engineered microor-
ganisms, the problems can be divided into three as to stability at each level
of the biological system. • *
* Firstly, the useful genes must be stably retained in the cell. This
stability includes synchronous replications of plasmids with replications of
the host cell. This stability is called physiological stability here.
Secondly, a sufficient population of the cells with physiologically
stable genetic information is'required to survive in a microbial community
within a reactor under ecological interactions such as interspecific competi-
tion, predator-prey interactions and so forth. .This stability is called
ecological stability of genetic information here.
Thirdly, various mutants may arise from an ecologically stable popula-.
tion with useful traits after a long period. Then, intraspecific competition
between the cells of original genotype with useful traits and mutants without
useful traits would occur. If the fitness of the original phenotypic cells
are lower than that of mutants, the population of the original genotype would
decrease and disappear from the reactor. Thus, engineered organisms are
required to have higher fitness in a given reactor than any other mutants
which might arise from,themselves. This stability, called evolutionary sta-
bility of the genotype of a microbial organism here, may be a kind of ecologi-
cal stability, because evolutionarily stable bacteria with useful traits are
able to exist under intraspecific competition with the mutants. The notion of
evolutionary stability, however, can be useful to differentiate the ecological
stability of a population in evolutionary (i.e., long) time scale where the
effects of mutants can not be ignored from that in ecological (i.e., short)
time scale where there is no significant effect of mutants on the population
dynamics.
In this study, from the view point of the ecological and evolutionary
stability of bacterial populations, a basic approach to the application of
genetically engineered microorganisms to wastewater treatment technology was
done using long term experimental cultures of enteric bacteria.
3-1-2 MATERIALS AND METHODS
Bacteria used : Escherichia vulneris JCM1688, Escherichia blattae
OCM1650, Enterobacter cloacae JCM1232, Enterobacter aerogenes JCM1235, Citro-
bacter freundii IF012681, Citrobacter freundii IF013539, and Citrobacter
freundii IF013545. Long term cultures of bacteria were conducted by a serial
transfer culture illustrated in Figure 3.1 and changes in ecological traits
related to their fitness in the culture were examined as .to specific growth
rate, cell dispersiveness, biochemical characteristics, etc.
12
-------
• same
3-1-3 RESULTS .
The experimental results are shown in Figures 3.2 and 3.3, and summa-
rized 'as follows: • •
(1) The fitness of bacterial populations, which is expressed by the specific
growth rate in the culture, increased with the number of transfers. This
is considered to be caused by the appearance and increase of mutants with
higher fitness in the culture.
(2) The fitness of a bacterial population eventually came to a plateau
level.
(3) The plateau values of the fitness of bacteria were not changed by the
treatment of a chemical mutagen with i\ITG.
(4) The plateau values were different in species (see Figure 3.2), and in
case of C._ freundi i the values varied in strains.
(5) Genetically identical bacterial strains which "were derived from the
clone have the same value for the plateau level.
(6) From the experimental results of mixed cultures consisting of two kinds
of bacteria, both of which had reached each plateau level,' it was revealed
that the population with the lower plateau level decreased and became
extinct in the mixed cultures, while that with the higher plateau level
remained (see Figure 3.3).
3-1-4 DISCUSSION " ....''
It was revealed that ecological traits of bacteria can genetically
change through intraspeci.fic competition. In addition, the changes in their
growth rates were limited to certain values which- were depending on species.
From the view point of the application of engineered microorganisms to waste-
water treatment, it is important to use the bacterial species or strains with
higher plateau levels for treatment, whether they are selected from natural
microbial communities or genetically modified in laboratories.
3-2 GENETIC MANIPULATION OF ECOLOGICAL TRAITS OF MICROORGANISMS
RELATED TO THEIR FITNESS
3-2-1 INTRODUCTION
Many genes with useful traits, which, are integrated into particular
plasmids, have been transferred into bacterial cells in order to genetically
modify the bacterial ability for some industrial demands. This method, howev-
er, can not manipulate a number of genetic traits which would be related to
ecological fitness of the bacteria in some way.. It commonly deals only with
genes related to some biochemical or physiological traits which are considered
to be useful for such demands. For wastewater treatment, engineered microor-
ganisms should grow under various kinds'of ecological interactions with other
species in a given reactor. Then, physiological and. ecological traits of
microorganisms required for the stable existence in a reactor should be
to genetic modification, if it becomes necessary.
open
In this study, focusing on how we can genetically manipulate such quan-
titative phenotypes as ecological traits which are considered to be expressed
by many genes on the chromosomes, for the first step, we investigated the
genetic changes of ecological traits of bacteria by genetic recombination at a
genomic level.
13
-------
Hfr strains of Escherlchia coll K12 are known as conjugational donors of
the chromosome DNA into F~ strains at high frequency. When an Hfr cell joins
to a F~ cell, conjugation-induced replication of F~ factor which is integrated
into the chromosome begins, and transfer of the chromosome DNA follows.
Complete transfer of the chromosome requires about 100 minutes at 37°C. By
breaking apart the male (Hfr) and female (F~) conjugant unions by violent
agitation at different times, we can get various kinds of recombinant strains
which vary in the degree of recombination between parental strains. Using
these strains, we investigated the genetic changes of ecological traits of
bacteria by genetic reco.mbination at a genomic level.
•a
3-2-2 MATERIALS AND METHODS
E. coli K12 JE5879 (sue, man) and E. coli K12 ME7767 (ara, mal, xyl,
mtl, lac, sue, str, gal) were used as an Hfr strain and an F~ strain, respec-
tively.
Conjugation between OE5879 and ME7767 and isolation of recombinant
strains between the two were carried out by the method illustrated in Figure
3.4.
Each strain was cultured in an L-broth medium. 6 ml of JE5879 and 3 ml
of ME7767 suspensions at a stationary phase of their growth were mixed togeth-
er in a 300 ml flask. The mixture was statically incubated for 180 minutes at
37°C.
After 30, 60, and 100 minutes of incubation, 1 ml of the bacterial
suspension were taken from the flask. The samples were agitated' with a vio-
lent blender for 30 seconds, and plated onto three kinds of EMB agar plates
containing maltose and streptomycin, xylose and streptomycin, and galactose
and streptomycin. After incubation for 1-2 days, dark-violet bacterial colo-
nies, which indicated acid forming ability from the sugars, were randomly
isolated from the plates with a loop. The ten strains isolated from the
plates that were incubated for 30 minutes are here referred to as hy30-l
through hy30-10. Each isolated strain was purified two times on the same kind
of EMB agar plate as the plate for the isolation. These isolates were trans-
planted to L-broth slant agar, and stored at 4°C. To obtain bacterial growth
curves, optical densities of bacteria were monitored with a spectrophotometer
in their cultures.
3-2-3 RESULTS
The changes of viable cell counts on the three kinds of EMB agar plates
during the conjugation experiment are shown in Figure 3.5. The growth curves
of ME7767, hylOO-7, hylOO-8, and hylOO-9 are shown in Figure 3.6. Various
types of growth curves which varied in the growth rate and maximum density at
a stationary phase of the population growth were observed. While both the
growth rate and the maximum density of hylOO-7, for example, were higher than
those of parental strains, the growth rate was lower than that of hy100-9.
The growth rate of JE5879 was much lower than that of ME7767.
3-2-4 DISCUSSION
Although the ecological traits of hy-30 and hy-60 are not examined at
14
-------
this time, from the result of hy-100 it is revealed that ecological traits
such as growth rate, maximum density at a stationary phase can .be changed by
genetic recombination at a'genomic DNA level.
15
-------
4. STUDY ON IMMOBILIZATION METHODS OF MICROORGANISMS FOR WASTEWATER
TREATMENT
The immobilization of microorganisms in bioreactors for wastewater
treatment and sludge treatment is useful to increase the concentration of
microorganisms in the reactor, and to maintain a large amount of microorgan-
isms for specifically removing particular pollutants. Three immobilization
methods will be discussed: binding method, entrapping method and self-
immobilization method.
Microorganisms themselves attach to the surface of media in the binding
method. Microorganisms are immobilized by chemicals in the entrapping method.
Immobilization is made by microorganisms themselves with no other materials in
the self-immobilization method.
4-1 BINDING METHOD
Suitable shape and materials of attached media depend on the features of
reactors in which the media are used. The reactors can be classified into
fixed bed type or fluidized bed type and aerobic or anaerobic.
4-1-1 ATTACHED MEDIA FOR FIXED BED TYPE REACTOR
(1)*For Aerobic Treatment
Attached biomass was compared among eight different media under condi-
tions of moderate aeration and synthetic substrate addition. Figure 4.1 shows
the results. Though attached biomass was different between media, sufficient
biomass grew after 3 months on all media tested.
Table 4.1 is an example of physical characteristics of attached media.
The media with more specific surface area accumulate more biomass but require
regular washing due to clogging. The media with diameter less than 1 cm
remove SS, then additional SS removal is unnecessary, though washing is re-
quired. On the other hand, washing is not necessary for media with bigger.
diameters but solids separation from the effluent is essential.
Table 4.1 Examples of Physical'Characteristics of Attached Media
Media
Ceramics
Ceramics
Polypropylene
Polypropylene
Shape
Saddle
Ball
Cyl inder
Ring
Void
Ratio
77
40
89
89
Size
(cm)
2
» 1
5
5
Specific
Surface
Area
(g/cm3)
335
400
76
185
Specific
Gravity
2.64
1.63
0.95
1.45
(2) For Anaerobic Treatment
A comparison test of attached media for anaerobic treatment revealed
that porous media produce gas more than media with smooth surfaces, showing
that microorganisms attach to the porous media more than the smooth.media. As
no aeration results in insufficient mixing in anaerobic treatment compared
with aerobic treatment, it was necessary to select and arrange media so that
16
-------
adequate influent distribution was attained.
4-1-2 ATTACHED MEDIA FOR FLUIDIZED BED TYPE REACTOR
Attached biomass was compared among different media under aerobic and
anaerobic conditions. Table 4.2 shows the results.
Table 4.2 Results of Comparison Tests of Attached Media . .
for Fluidized Bed Type Reactor
Media
Silica
Sil ica
Silica
Zeol ite
Sand
Sand
Sand
A
B
C
Polypropylene
Plastic
Foam
Size
(mm)
0.2-0.3
.074-0.15
0.09-0.25
0.35
2-5
1-2
Specific Aerobic
Gravity or
(g/cm3) Aerobic
2.
2.
2.
1.
1.
1.
65
65
63
98
03
3
Aerobic
Aerobic
Anaerobic
Anaerobic
Aerobic
Aerobic
Attached
Biomass
(mg*VSS/
g%edia
40
1,200
3.5
10-30
26
95-150
)
4
1
0
'1
3
4
Biomass
in
Reactor
(g/i)
.4
2 '
.8
0-20
.8
. 3-6.
7
Attached biomass and total biomass in reactors was more for sand B than
A, which has a smaller diameter under aerobic conditions. Comparison of sand
B and C shows that aerobic microorganisms tend to attach to media more than
anaerobic microorganisms. Zeolite had more biomass than sand in an anaerobic
condition. As polypropylene pellets have.the . least specific gravity, they
were fluidized in a reactor with the least mixing. As about 90 % of the
volume of plastic foam has pores with diameters 0.03 to 1 mm, it accumulates,
many microorganisms.
Granular activated carbon and porous ceramics had problems with abra-
siveness. .
4-2 ENTRAPPING METHOD ' -= .
>
Polyvinyl alcohol (PVA), acrylic amide (ACAM) and polyethylene glycol
(PEG) were compared as chemicals to entrap microorganisms.
4-2-1 POLYVINYL ALCOHOL (PVA)
(1) PVA-Boric Acid Method
In this method, a mixture of PVA and activated sludge is dropped into
saturated boric acid solution, then activated sludge is entrapped in lattices
of pellets made of PVA and boric acid. Appropriate conditions of the method
were as follows:
-PVA concentration in the mixture of PVA & activated sludge: 7.5-10%
-Activated sludge concentration: 80-90 g/1
-Contact time of the mixture'with the saturated boric acid solution: 24
hours
*
Though the respiration rate of the entrapped activated sludge was low at
the beginning of cultivation, the rate increased rapidly to 3 - 4 times the
initial value after 10 days, as shown in Figure 4.2.
17
-------
(2) PVA-Refrigeration Method
The mixture of PVA and activated sludge is refrigerated and becomes a
gel. The activity of microorganisms is only slightly reduced after the immo-
bilization by this method compared with the method using boric acid, which is
harmful to microorganisms.
The appropriate refrigeration temperature was minus 20°C. The strength
of the gel did not increase at temperatures lower than the value. The PVA
concentration of 10 % was not enough to make pellets with enough strength..
Pellets of 22.5 % showed a lower respiration rate than that of 15 % as shown
in Figure 4.3, thus PVA concentration, of about 15 % is suitable.
4-2-2 ACRYLIC AMIDE (ACAM)
ACAM is relatively cheap and a typical polymer for entrapping microor-
ganisms. However, it has the ability to impair the activity of microorganisms
due to a harmful.monomer present during the immobilization.
(1) Agar-ACAM Monomer Method
Microorganisms are first immobilized by agar, then an ACAM monomer is
added to the immobilized pellets to polymerize and harden them. The more ACAM
monomer added resulted in less activity of microorganisms. But less additio"
of monomers leads to weak gels. The proper concentration of ACAM was about
7.5 % as shown in Figure 4.4.
(2) ACAM Monomer Method
A cross-linking agent and a polymerization agent are added to a mixture
of ACAM monomer and activated sludge to make gels. Figure 4.5 shows the test
results. Proper concentration of ACAM monomer was about 15 % in this method.
Variation of concentration of a cross-linking agent had no effect on the
activity of microorganisms.
4_2-3 POLYETHYLENE 6LYCOL (PEG)
An immobilization method by PEG was developed as a method to immobilize
nitrifiers. Activity of nitrifiefs in gels by PEG was higher by 5 times than
those by ACAM, melamine and urethane.
A respiration rate of gels prepared by PEG after sufficient cultivation
was 1,100 mg*02/l*hr which was higher.by 2 times than those prepared by other
materials. This was mainly because the penetration rate of ammonia from water
to gels was the highest in PEG gels among the chemicals tested.
4-3 SELF-IMMOBILIZATION METHOD
Immobilization is made by microorganisms themselves with no other mate-
rials in this method. Dense granules made of microorganisms are naturally
formed without sludge recirculatidn in upflow sludge blanket type bioreactors.
UASB (upflow anaerobic sludge blanket) process applied for treatment of waste-
water with high organic concentration has been well known as a process utiliz-
ing this method. In Biofocus WT, four processes using self-immobilization
methods have been investigated. There are two UASB processes, an AUSB (aero-
bic upflow sludge blanket) process and MRB (multi-stage reversing-flow biore-
actor).
•18
-------
4-3-1 UASB PROCESSES
.(1) For Municipal Wastewater Treatment
. Anaerobic sludge blanket of 10 g*SS/.l was obtained under
temperature TO to 25°C, linear velocity (LV) 5 to 6 cm/min
retention time 6 to 9 hours. Biological granules
acclimation. Though they are small in diameter (1
same as biological granules generally found in
wastewater with high-organic concentration.
conditions of
and hydraulic
have formed after 6 months
mm), they are basically the
the UASB process treating
(2) For Treatment of Supernatant from Heat Treatment of Sludge
Anaerobic sludge blanket shown in Figure 4.6 was obtained.
influent was 5,000 mg/1 and temperature in the reactor was 35°C.
granules were 0.5 to 2 mm in diameter and had 1 to 5 cm/sec of
velocity. Methanothrix seemed to dominate in the granules.
CODcr in
Biological
a settling
4-3-2 AUSB PROCESS
Aerobic .granules were developed in this process. DO in influent to the
reactor is as high as 50 to 70 mg*02/l for keeping the reactor aerobic. This
high DO concentration is achieved by contact of wastewater to pressurized pure
oxygen as shown in Figure 4.7. Granules of several milimeters in' diameter
have been observed. The mechanism of the aerobic granulation is under inves-
tigation. :,
4-3-3 MRB
Multi-stage Reversing-flow Bioreactor (MRB) which utilizes new biologi-
cal interaction for removal of organic substrate in wastewater has been de-
veloped in Biofocus WT.
(1) Principle of MRB
Figure 4.8 is a flowsheet of the MRB. Influent is introduced to the
first downflow aeration vessel (AV). No suspended solids remain here. The
next vessel is a .biological reaction vessel (BV) of upflow without aeration.
Suspended solids with a higher settling velocity than the upflow velocity
remain here. The following vessels have the same sequences. The AV is merely
for oxygen supply^ . In the BV, suspended solids and biomass are gently mixed
by the upflow, resulting in formation of self-granulated sludge (SGS) with
diameters of 2 to 10 mm. The primary effluent of a municipal wastewater
treatment plant was "influent to the MRB. In the initial period of the opera-
tionj anaerobic sludge was accumulated in the first BV and then the sludge was
carried over from'the first BV to the second and third BVs, the SGS was formed
in due course as shown in Figure 4.9. ' Once the sludge blankets, of SGS were
built up in these BVs, the fourth and fifth BVs receiving partially treated
wastewater became aerobic and developed an aerobic sludge blanket. The aero-
bic sludge blankets did not have the firm SGS, but could be kept in these BVs
against 144 m/day upflow velocity. Sludge returning, liquid circulation and a
final sedimentation tank were not required.,
(2) Mechanism of Self-Granulation of Sludge in MRB
One of the interesting topics of the MRB is the formation of SGS. The
SGS is firm enough to be picked up by tweezers, but is crushed by strong shear
stress, such as when it is pumped out. Most of the SGS is covered with a thin
white film as shown in Figure 4.9. The film consists of filamentous bacteria
19
-------
as shown in Figure 4.10. The bacteria, having small sulfur granules in their
cells and showing active gliding mobility, are identified as Beggiatoa. The
inside of the S6S is black and anaerobic. The plate count method for sulfate
reducing bacteria indicated that there were 10° N/g-SS of sulfate reducing
bacteria (SRB). In the MRB, oxygen is supplied at the AV where few microor-
ganisms are present. DO in the AV is limited to the saturation concentration
for wastewater at a given temperature. On the other hand, a significant
amount of microorganisms accumulate in the BV, and the DO supplied in the AV
is consumed rapidly if organic substrate in the wastewater is high. Conse-
quently, near-anaerobic conditions allowing the growth of the SRB are de-
veloped in the BV. From the above-mentioned observations, the mechanism of
self-granulation of sludge in the BVs is considered as follows. Figure 4.11
explains the interaction among the SRB, Beggiatoa as sulfide oxidizing bacte-
ria and other anaerobic bacteria. Organic substrate diffusing into the SGS is
hydrolyzed to organic acids by anaerobic bacteria, then utilized by the SRB
resulting in sulfate reduction.
*
Sulfide produced from this reaction diffuses into bulk liquid through
the SGS surface. Though the oxygen supply is limited in the BV, there is the
chance of the microorganisms contacting oxygen on the SGS surface. Because
an oxygen consumption rate of bacteria oxidizing sulfide is much higher than
that of organic substrate oxidizing bacteria, most the oxygen is utilized by
the sulfide oxidizing bacteria which are microaerophil ic. Beggiatoa, which is
one of the most common bacteria existing in such an environment, increases on
the SGS surface in a thin film.
20
-------
5. DEVELOPMENT OF BIOREACTORS FOR WASTEWATER TREATMENT
Many municipal sewage systems are being designed or constructed in rural
areas in Japan. Even in rural areas, the land for wastewater treatment plants
is limited in many cases. In addition, small municipalities tend not to be
able to employ enough skilled technicians.
Many local governments will have to reconstruct or to improve their old
wastewater treatment plants in the future. An increase in treatment capacity,
and/or upgrading removal efficiency of .-nutrients or refractory organic materi-
als might be .required in such cases.. ,
Development of bioreactors for wastewater treatment aims to answer these
needs mainly by introducing immobilization of microorganisms. There are three
targets of the development, in addition to develop bioreactors with features
of easier 0 & M, phosphorus removal or refractory organic removal.
(1) To reduce the retention time in reactors to 1/2 to 1/10 compared with
the conventional processes.
(2) To reduce the energy requirement and/or sludge production to 1/2 com-
pared with conventional processes.
(3) To reduce the retention time in reactors for biological nitrogen removal
to 6 to 12 hours so that existing plants-can be modified to remove nitro-
gen.
The first target would be accomplished mainly by the development of
aerobic bioreactors. Anaerobic bioreactors are for the second target. The
third target would .be concluded by immobilization of microorganisms related to
nitrogen removal.
In total, 23 bioreactors are under development. AIT of the bioreactors
have been developed as joint researches between the Public Works Research
Institute and private companies. Twenty-three pilot plants are operated at
present.
5-1 AEROBIC BIOREACTOR
Thirteen aerobic bioreactors have been developed. They are divided into
three groups: fixed, bed type, fluidized bed type and sludge blanket type.
5-1-1 FIXED BED TYPE
Figure 5.1 shows an example of fixed bed type aerobic bioreactors which.
have been developed. The aerobic biofilter has several experimental and full
scale trials in Japan and abroad. Influent is introduced to the top of the
fflter. Air is dispersed from the bottom of the filter. Wastewater is treat-
ed by the biofilm on the surface of filter media. Suspended solids are re-
moved by a filtration mechanism.
Suspended solids in influent is better to be low in order not to cause
clogging in the filter. The filter is washed by backwashing. The operational
conditions and the performance are summarized in Table 5.1.
21
-------
Table 5.1 Operational Conditions and Performance
of Fixed Bed Aerobic Bioreactor
CONDITIONS
Linear Velocity, m/day
Hydraulic. Retention Time, hr
BOD Volumetric Load, kg-BOD/m3/day
31
1.65
1.1
' 50
1.01
1.5
70
. 0.7
2.3
PERFORMANCE
ss
BOD
T-K-N
T-P
mg/1
mg/1 -
mg/1
mg/1
Inf.
42
72
28.8
2.8
Eff.
4
6
4.1
1.9
Inf.
39
62
27.1
2.2
Eff.
4
9
5.7
1.5
Inf.
35
69
24.1
2.1
Eff.
4'
11
11.1
1,5
5-1-2 FLUIDIZED BED TYPE
Figure 5.2 shows an example of fluidized bed type aerobic bioreactors
which have been developed. Air diffusers are installed in the reactor. Fine
sand, fine slag, granulated activated carbon, plastic media, and chemical
media which are made of polyacryl amide and poly-ethylenglycol for immobiliz-
ing microorganisms are used as the fluidizing media in these developments.
Media separation from liquid is one of the engineering problems to be
solved. When effluent from bioreactors has little SS, final solid-liquid
separation might be accomplished not by sedimentation but by filtration be-
cause settlability of 'the SS is poor. How to fluidize the media is another
hurdle. Lighter and finer media are better for fluidizing, but such media are
difficult to separate from liquid.
Table 5.2 Operational Conditions and Performance
of Fluidizing Aerobic Bioreactor
RUN
1
CONDITIONS
Hydraulic Retention Time of Reactor, hr 2.0
Linear Velocity of PF,
Linear Velocity of SF,
PERFORMANCE
T-BOD mg/1
S-BOD mg/1
SS g/1
m/day
m/day
Influent
PF Effluent
DF Effluent
Influent
PF Effluent
DF Effluent
Influent
PF Effluent
DF Effluent
80
161
138
20.6
9.4
44.3
'10.9
8.0
109
10.1
3.2
2.0
80
161
91.9
• 16.0
11.1
47.0
9.5
8.0
• 40.1
7.5
2.6
4.0
53 r"
115
132
17
9.1
44.7
8.5
6.3
102
•6.7
2.2
PF: Plastic Media Filter
DF: Dual Media Filter
22
-------
A plastic medium of 3 to 5 mm in diameter and about 1.0 in density has
been employed in a development. It can be separated by a screen. The opera-
tional conditions and the performance of the plastic media bioreactor followed
by a two stage filtration process are summarized in Table 5.2.
5-1-3 SLUDGE BLANKET TYPE
A flow diagram of an aerobic sludge blanket reactor developed in" this
project is shown in Figure 5.3. Influent flows into a pressurized oxygen
tank. Effluent from the tank'with .50 to 70 mg/1 of DO is introduced to the
bottom of the.upflow reactor. At the beginning of the operation, activated
sludge was seeded in the reactor. Within several weeks, aerobic biological
granules formed in the reactor. The operational conditions and the perform-
ance are summarized in Table 5.3. Though the mechanism of self-granulation
has not been clear, sludge that, settles easily has been maintained for one
year. _
Table 5.3 Operational Conditions and Performance of
Sludge Blanket Type Aerobic Bioreactor
CONDITIONS
Hydraulic Retention Time, hr ,1.72
Recirculation Ratio of Effluent 1.57
Up Flow Velocity in Reactor, m/day 125
Influent DO mg/1 63
Effluent DO mg/1 4.8
PERFORMANCE
T-BOD
S-BOD
SS
T-N
T-P
mg/1
mg/1
mg/1
mg/1
mg/1
Influent
112.6
42.4
91.3
30.9
2.7
Effluent
19.9
10.2
9.6
24.6
1.8
5-2 ANAEROBIC BIOREACTOR
Six anaerobic bioreactors are under development. They are also divided
into 3 groups: fixed bed type, fluidized bed type and sludge blanket type:
Anaerobic wastewater treatment requires the post-treatment due^to the follow-
ing two reasons. Effluent from anaerobic treatment can. hardly satisfy the BOD
standard of 20 mg/1 for secondary treatment. It also contains sulfide which
causes corrosion and odor problems.
Most post-treatment processes under development are aerobic biological
treatments which require additional energy and facilities. A pilot plant has
installed a modified trickling .filter for post-treatment.. The filter media
are cubic sponges which' catch up fine suspended solids. This system can
eliminate further SS removal facilities and aeration devices for post-treat-
ment.
5-2-1 FIXED BED TYPE
Non-woven fabrics, saddle-shaped ceramics and plastic braids are used as
23
-------
media for fixed beds.
Figure 5.4 shows an example of a fixed bed type anaerobic bioreactor
under development. Concentrated anaerobic sludge can shorten the hydraulic
retention time to less than 8 hours for primary effluent. There is no recir-
culation of effluent.
5-2-2 FLUIDIZED BED'TYPE
One fluidizing anaerobic bioreactor with fine zeolite of 0.5mm in diame-
ter as media has been developed. The flow diagram is shown in Figure 5.5.
Recirculation of effluent is required to fluidize the media and the recircula-
tion ratio depends on the fluidized velocity of the media. The retention time
is less than 3 hours, but this reactor requires energy for recirculation..
5-2-3 SLUDGE BLANKET TYPE '
An upflow anaerobic sludge blanket (UASB) reactor shown in Figure 5.6
has been applied for domestic wastewater treatment. Pellets of anaerobic
microorganisms formed 6 months after seeding digested sludge to the reactor.
The effluent from the reactor has been good as the influent to the modified
trickling filter mentioned before. The operational conditions and the per-
formance are summarized in Table 5.4. The system has neither liquid recircula-
tion, nor aeration devices. If there is enough water head difference avail-
able between influent and effluent in a plant, this system can be an energy
saving system.
Table 5.4 Operational Conditions and Performance of
Sludge Blanket Type Anaerobic Bioreactor
CONDITIONS
Hydraulic Retention
Hydraulic Retention
Water Temperature,
PERFORMANCE
BOD mg/1
SS mg/1
Sulfide as
S mg/1
Time
Time
C°
of
of
Inf.
71-111
45-99
Reactor
MTF, hr
, hr
E'ff.(UASB)
24-40
8-15
6-8
8
2
13-1.6
Eff
7-1
2-7
0
•
7
(MTF)
MTF: Modified Trickling Filter for post-treatment
5-3 MULTI-STAGE REVERSING-FLOW BIOREACTOR (MRB).
The Multi-stage Reversing-flow Bioreactor (MRB) which utilizes new
biological interaction for organic substrate removal in wastewater has been
developed. As shown in Figure 4.8, MRB has several stages, one stage con-
sists of a downflow aeration vessel (AV) and an upflow biological reaction
vessel (BV). Successful accumulation of self-granulated sludge (SGS) can be
achieved in BV. As described in 4-3-3, the biological interaction between
sulfate reducing bacteria and sulfide oxidizing bacteria (Beggiatoa) .is con-
sidered to contribute to form SGS with 2 to 10 mm in diameter. SGS can be
maintained in BV with linear velocity of 144 m/day. A prototype plant study
was conducted to demonstrate the performance of the MRB.
24
-------
'The design parameters, the operational conditions and the performance of
the prototype MRB are summarized in Tables 5.5 and 5.6
Table 5.5 Design Parameters of Prototype MRB
Diameter: AV
Lower part of BV
Upper part of BV
Depth Average
Total volume
Hydraulic retention time
Up-flow velocity at lower part of BV
10 cm
10 cm
18 cm
220 cm
213 1
3-4.5 hour
144-216 m/day
Table 5.6 Operational Conditions of Prototype MRB
Influent
BOD volumetric load for total volume
for BV volume
Hydraulic retention time
Dissolved oxygen concentration in AV
Settled municipal
0.42 kg/m3.day
0.73 kg/m3.day
4.5 hour
5-9 mg/,1
wastewater
Table 5.7 Typical performance and Sludge Concentration in MRB
No. of BV
Total Solids (mg/1)
BOD (mg/1)
D-BOD (mg/1)
SS (mg/1)
No.l
21,000
Influent
80
50
30
No. 2
14,600
No. 3
8,900
No. 4 No. 5
5,900 14,000
Effluent
14
9
4
Though the influent BOD was sightly low, the effluent quality was as
good as that of a conventional activated sludge process. The influent was
municipal primary effluent. The total hydraulic retention time was 4.5 hours
including the final solid-liquid separation in the last BV. The MRB has
several advantages over conventional biological wastewater treatment, such as
lower energy requirements, less sludge production, and no necessity for a
final settling tank.
5-4 BIOREACTORS FOR NITROGEN REMOVAL
This study has been conducted to develop biological nitrogen removal
processes utilizing immobilization methods of microorganisms related to bio-
logical nitrogen removal in pilot plants. The immobilization methods used are
a binding method and an entrapping method.
5-4-1 BIOREACTOR FOR NITROGEN REMOVAL WITH BINDING METHOD
(1) Experimental Conditions of Pilot Plant Study
The flowsheet of the pilot plant is shown in Figure 5.7. t The size of
the cross section of the reactor is 700 mm by 700 mm and 2,800 mm in height.
25
-------
The volume is 1.37 m3. Sands with diameters of 0.2 mm to 0.3mm constitutes 8
% of the reacting volume in the reactor. The specific surface area of the
sand is 1.5 m^/g. Sand was selected as attaching-media after a comparison
test among clistobalite, sand and two kinds of light weight aggregate.
Influent flow was 11.5 m3/day and the retention time in the reacting
volume in the reactor was 2.9 hours. BOD, total nitrogen and ammonia nitrogen
in influent were 110 to 160 mg/1, 40 to 60 mg/1 and 30 to 40 mg/1, respective-
ly. The water temperature was 28 to 29°C. .
The operation mode of the reactor was intermittent aeration with contin-
uous inflow. Both aeration time and anaeration time were 15 minutes. The
volumetric loadings of BOD and total .nitrogen in the reactor were-0.9 to 1.3
kg/m3/day and 0.3 to 0.5 kg/m3/day, respectively. The DO level was set to 2
to 3 mg/1 during aeration period.
(2) Results of Pilot Plant Study and Discussion.
. Total nitrogen in the effluent was 8 to 15 mg/1 as shown in Figure 5.8.
The fact that ammonia nitrogen and nitrate nitrogen in the effluent were about
2 mg/1 and 8 to 10 mg/1, respectively, shows that it is necessary to enhance
denitrification for upgrading nitrogen removal.
MLVSS in the reactor was 7,000 to 9,000 mg/1.. About 90 % of the MLVSS
was attached to sands, which demonstrates the effectiveness of sand to in-
crease microorganisms in the reactor.
The following are necessary to study in order to improve this biological
nitrogen removal process.
a. optimum combination of aeration time and anaeration time
b. DO level during aeration time
c. air flow rate to fluidize sand and to supply DO
d. solid-liquid separation'of effluent from the reactor
5-4-2 BIOREACTOR FOR NITROGEN REMOVAL WITH ENTRAPPING METHOD
(1) Experimental Conditions of Pilot Plant Study
Figure 5.9 is the flowsheet of the pilot plant. The volume of the
nitrification tank is 4 m3 and that of the denitrification tank is 6m3. 7.5
% of the nitrification tank volume is filled by pellets in which activated
sludge taken from a biological nitrification-denitrification plant are en-
trapped. Polyethylene glycol was selected as a suitable chemical to make the
pellets with activated sludge containing nitrifiers based on a selection test
among 5 chemicals. As it was revealed that chemical entrapping of denitrifi-
ers for denitrification was not effective, microorganisms including denitrifi-
ers were suspended in the reactor. Experimental conditions are summarized in
Table 5.8.
(2) Results of Pilot Plant Study and Discussion
Figure 5.10 is the performance of nitrogen removal in the pilot plant.
BOD, TN and SS in effluent were 10 mg/1, 7 mg/1 and 7 mg/1, respectively in
RUN 1 in which the total retention time in the reactor was 8 hours. This
performance was satisfactory, considering the water temperature was between 10.
to 12°C. The total retention time was reduced to 6 hours in the following RUN
2 in which BOD and TN in the effluent were less than 20 mg/l and 10 mg/1.
26
-------
Table 5.8 Experimental Conditions
RUN-1 RUN-2
Inflow (m3/day) , . 31.5 42
Retention Time in Biological Reactor (hr) 86
Retention Time in Nitrification Tank (hr) 4.8 3.6
Retention Time in Denitrification Tank (hr) 3.2 2.4
Percent of Pellets Filled into l\l. Tank by Volume (%) 7.5 7.5
Air Supplied (Nm3/day) 200 230
Recirculation Ratio of Mixed Liquor (%) 300 300
Sludge Return Ratio (%) 40 40
This process can be applied to existing activated sludge plants for
upgrading Tl\l removal. The following should be studied in order to make this
biological nitrogen removal process feasible.
a. 1ife of pel lets
b. a production method of huge amounts of pellets for full-scale plant
application - .
c. cost of pellets, including disposal
27
-------
6. DEVELOPMENT OF BIOREACTORS FOR SLUDGE TREATMENT
6-1 ANAEROBIC SLUDGE DIGESTION
There are two objectives in this development. One is to shorten the
period of sludge digestion to less than 10 days, and the other is to increase
the percentage of solid decomposition for recovering more digested gas by 1.5
times that of the conventional process. Two techniques have been investigated
in order to attain the objectives. They are liquefaction of feed sludge and
immobilization of bacteria related to anaerobic sludge digestion by the bind-
ing method.
It is well known that liquefaction of "feed sludge, especially excess
biological sludge, is the most limiting factor in the digestion. As attempts
were unsuccessful to find bacteria and enzymes which accelerate the liquefac-
tion of sludge, the direction has been changed to use physical-chemical meth-
ods. The following three methods have been investigated:
(1) Crushing excess sludge by a mill with NaOH of 0.05 N addition An
example of the experiment is shown in Table 6.1.
(2) Heat treatment of excess sludge with conditions of 60°C for 1 hour
(3) Heat treatment of excess sludge with conditions of 90°C for 1 hour and
protease addition to the heat treated sludge flowing to ah acidification
process
Table 6.1 Liquefaction of Excess Sludge by Crushing with NaOH
Items
(mg/1)
TS
VS
VSS
PH
Sol. Sugars
Sol. Proteins
VFA
formic
acetic
propionic
isobutyric
butyric
isovaleric
valeric
CRUSHING • —
Control
24,100
19,700
18,400
6.0
. 199
250
'1,836
-
580
425
84
401
' 157
189
Crushed
23,900
18,800
12,500
6.2
. 458
2,420
3,070
114
2,082
-
348
-
526
—
CRUSHING WITH
Control
24,800
19,500
18,100
6.1
128
205
1,361
—
460
346
250
327
119
109
NaOH
Addi ti on
25,000 '
19,300
16,400
7.1
232
1 , 060
3,222
28
1 , 824
350
99
239
169
513
NAOH
Crushed
with
NaOH
24,800
18,400
8,100
7.0
554
3,030
3,847
51
3,162
_
__
— .
_
168
Attached media for bacteria in sludge digestion are different from those
for bacteria in wastewater treatment, because sludge is much more viscous and,
of course, contains more SS than wastewater. Cement balls as attached media
for a fluidized bed type reactor and ceramics as attached media for a fixed
bed type reactor have been investigated.
Three pilot plants are under operation. Their flowsheets are in Figure
28
-------
6.1. In addition to their digestion performance, dewaterability of the di-
gested . sludge, biodegradability of the filtrate and heat balance of each
system are scheduled to Be examined.
6-2 BACTERIA LEACHING OF HEAVY METALS FROM SLUDGE
The bacteria leaching technique, which is practiced in recovering metals
from ores by oxidizing sulfur and lowering pH by sulfur oxidizing bacteria and
iron oxidizing bacteria, was applied to sewage sludge in order to remove heavy
metals from it.
6-2-1 GROWTH-OF BACTERIA IN SLUDGE :
Thiobacillus thiooxidans and Thiobacillus ferrooxidans were selected as
sulfur oxidizing bacteria and iron oxidizing bacteria. The conditions for the
two kinds of bacteria to grow in sludge were examined. Results of the study
are summarized as follows,:
>
(1) There .was no difference in the growth of the bacteria in primary sludge,
excess sludge, mixed sludge or digested sludge.
(2) Sterilization of sludge was not necessary for the growth of the bacte-
ria.
(3) Though it was indicated that Thiobacillus thiooxidans and Thiobacillus
ferrooxidans or bacteria similar to those existed in sludge, addition -of
Thiobacillus thiooxidans and Thiobacillus ferrooxidans accelerated the
decrease of pH in sludge.
(4) Initial pH of sludge had no influence to the growth of the bacteria if
the pH was between 4 to 7.
(5) Addition .of powdered sulfur or ferric sulfate
sary for the growth of bacteria.
as a substrate was neces-
6-2-2 HEAVY METAL REMOVAL FROM SLUDGE BY BACTERIA LEACHING
Bench scale tests on the bacteria leaching of mixed sludge and digested
sludge were conducted. Figure 6.2 is the procedure. The results were com-
pared with those attained by sulfuric acid addition to sludge. Results of• the
•study are summarized as follows.
(1) Mixed sludge and digested sludge
after the start of the incubation. The pH values were 1.0 and.1.8 for
mixed sludge and digested sludge, respectively.
(2) Leaching ratios depended on heavy metals. Cd, Zn and Mg were
more than 80 %. Little Hg was removed from the sludge. Figure 6
changes of leaching ratios of heavy metals.
(3) There was no significant difference in the bacteria leaching and "chemi-
cal leaching1 as to the leaching ratios of heavy metals.
came to pH equilibrium within 7 days
leached
3 shows
29
-------
7. DEVELOPMENT OF SOLID-LIQUID SEPARATORS FOR RAW WASTEWATER
Almost all bioreactors for wastewater treatment being developed in
Biofocus WT incorporate immobilization methods in the reactors in order to
increase concentration of microorganisms. A problem may exist in reactor
clogging due to SS in influent for such reactors, especially fixed bed type
reactors. In addition, as dissolved organics are decomposed by microorganisms
faster than particulate organics, removal of the bigger particles from raw
wastewater makes bioreactors more efficient.
Solid-liquid separators for raw wastewater, of which SS removal rate is
higher than that of a conventional primary sedimentation tank, have been
developed in this subject.
7-1 SIZE DISTRIBUTION OF SUSPENDED SOLIDS IN RAW WASTEWATER
Filtrate were prepared by sieves of 2 and 74 mm and filter papers of 8
and 1 microns. SS, BOD, COD, TN and TP in the filtrates were -analyzed.
Examples of the study are shown in Figures 7.1 and 7.2, and summarized as
follows:
CD Percentage of particulate BOD in raw wastewater was higher than those of
particulate nitrogen and phosphorus. Then" C/N and C/P ratios in influent
to bioreactors become small, if the removal rate of SS in raw wastewater
is high (see Figure 7.1). . ...
(2) Size distribution of SS in raw wastewater is different during high
influent loading and during low influent loading. Percentage of SS more
than 8 microns is higher during high influent loading than during low
influent loading (see Figure 7.1). .
(3) There was a possibility that influent to small plants with shorter time
of flow contained soluble BOD more than that of large plants with longer
time of flow (see Figure 7.2).
7-2 BIODEGRADABILITY OF PARTICULATE ORGANICS
Biodegradability of particulate TOC was examined under aerobic and
anaerobic conditions. Examples of the study are shown in Figures 7.3 and 7.4.
Initial biodegradation rate of particulate TOC in particle size more than 1
micron meter was much lower than that of smaller than 1 micron meter in both
aerobic and anaerobic conditions.
7-3 SOLID-LIQUID SEPARATORS FOR RAW WASTEWATER ^
Five solid-liqui'd separation processes have been developed in this
subject.
(1) Upflow sedimentation process
(2) Floating-bed filtration process
(3) Filtration process with tube-shaped filtration media
(4) High rate bioflocculation process with lamella settler
(5) Two stage filtration process with upflow moving bed and downflow
fixed bed
An upflow sedimentation process has been investigated using an experi-
mental column of 280 mm in diameter and 1,500 mm in effective height and
mixing paddles. Figure 7.5 is 'the result.
30
-------
Removal rates of the processes of (2),. (3) and (4) are summarized in
Figure 7.6. Removal.rates of SS larger than 8 microns in raw wastewater were
more than 80 % in these three processes. On the other hand, SS smaller than 8
microns were difficult to remove by these processes.
31
-------
8. DEVELOPMENT OF BIOSENSORS FOR WATER QUALITY MEASUREMENT
Biosensors, which use biomaterials such as enzymes and microorganisms as
detectors to measure objective matter, have been developed as for BOD, ammo-
nia-nitrogen and organic acids.
8-1 PRINCIPLE OF WATER QUALITY MEASUREMENT BY BIOSENSOR
Aerobic microorganisms are applied to these biosensors. Microorganisms
are immobilized in a thin biofi1m which is attached to an oxygen electrode.
When the microbial electrode is immersed into a substrate solution, the sub-
strate begins to diffuse into the-biofi1m. If microorganisms in the biofi1m
metabolize the substrate, they consume DO in the biofilm resulting in decline
of oxygen electrode current. In a few minutes the oxygen consumption rate
becomes equivalent to the oxygen diffusion rate towards the biofilm. Conse-
quently the current of the electrode becomes steady (peak current).
When the microbial electrode .is removed from the substrate solution and
put into a solution without substrate, oxygen consumption rate decreases the
level of endogenous metabolism (base current). An oxygen consumption rate,
which is expressed as the difference between the peak current and base cur-
rent, is in proportion to substrate concentration within a certain range.
The flow chart of a system is shown in Figure 8.1. The microbial elec-
trode attached to the flow cell sends the current to a microcomputer (CPU) by
way of an amplifier and an analog digital (A/D) transfer. A standard solution
or a sample is injected into the flow cell for calibration or measurement. DO
in the flow cell is almost completely eaturated. The flow cell is .set in an
isothermal water bath to maintain a constant temperature.
8-2 BOD SENSOR
8-2-1 SELECTION OF BIOFILM FOR BOD SENSOR
Microorganisms and immobilization methods to make biofilm for BOD sensor
were experimentally compared among the following candidates.
»
Microorganisms:
Sacharomyces cerevisiae(yeast)
Sacharomyces uvarum(yeast)
Schizosacharomyces pombe(yeast)
Trichosporon cutuneum(yeast)
Candida lypolylica(yeast)
Candida tropicalis(yeast)
soil microorganisms
wastewater microorganisms
activated sludge
Immobilization methods:
entrapping method by glutaraldehyde treatment
entrapping method by PVA-boric acid
immobilization by acetylcellulose dialysis membrane absorption method
Three combinations of microorganisms and an immobilization method were
selected for further BOD sensor development based on performance of the dif-
ference between the peak current and the base current, the transient time to
32
-------
the peak current and the recovery time to the base current. They were;
a) Trichosporon cutuneum + gelatin-glutaraldehyde film,
b) Trichosporon cutuneum + acetylcellulose film, and
c) soil microorganisms + PVA "film
8-2-2 PERFORMANCE OF BOD BIOSENSOR
The three selected biofi1ms . were evaluated. Glucose-glutamic acid
(GGA) solution and synthetic wastewater were used as samples. The results are
shown in Figure 8.2. Each biofilm had a'linear relationship between BOD
concentration and the current difference. The most sensible,biofi1m was of
soil microorganisms immobilized by PVA in both GGA solution and 'synthetic
wastewater. The current difference of a GGA solution shows larger than the
artificial sewage in each biofilm. ,
A biofilm of soil microorganisms immobilized by PVA was used for further
study in which BOD in several kinds of raw wastewater, secondary effluent and
polluted urban river water was measured by the BOD sensor and by a manual
method. Calibration curves were prepared with GGA solution and synthetic
wastewater. Figure 8.3 shows the relationship between the sensor BOD and
manual 6005.
In the case of raw" wastewater, the sensor BOD showed enough output
current. The sensor BOD calibrated by the synthetic wastewater was more
agreeable with the manual BOD on the whole than that by GGA.
In cases of secondary effluent and river water, however, the current
difference was so small that the sensor BOD showed much lower than the manual
BOD. The difference between the two calibration methods seemed very small.
Samples which have been biologically treated seem difficult to detect by this
BOD sensor in a short peripd. We have been trying to improve the detection
ability for such refractory organic solution.
8-3 AMMONIA.SENSOR
8-3-1 PERFORMANCE OF AMMONIA BIOSENSOR ,.
Nitrosomonas europaea was immobilized between two porous acetylcellulose
membranes to make a biofilm for an ammonia biosensor. The relationship be-
tween ammonia nitrogen concentration below 10 mg/1 and a current difference
was linear as shown in Figure 8.4. According to the biomass increase, the
upper sensible concentration became lower, but the resolution„became better.
It takes 7 to 8 minutes for the sensoring system to reach the peak current
after sample injection. It takes 7 to 12 minutes to return to the base cur-
rent after stopping the sample injection.
Environmental samples .such- as river water and effluent from* wastewater
treatment .plants were measured by the sensor system in batch bases. The
sensor ammonia agreed well with the manual ammonia. The sensor system was
tested as a monitor for effluent from a wastewater treatment plant. The
monitoring of ammonia in the effluent was conducted every 15 minutes. A
stable agreement between sensor ammonia and manual ammonia was observed during
more than 3 weeks. Ammonia nitrogen standard solution with a glucose glutamic
acid solution (100mg/l) was .sometimes injected into the sensor system to check
the interference of other microorganisms. But only the response to the ammo-
33
-------
nia nitrogen solution was recognized. Therefore, the interference of other
microorganisms seemed negligible for this sensor.
8-3-2 APPLICATION OF AMMONIA BIOSENSOR TO TOXIC SENSOR
As a nitrification bacterium is said to be very sensitive to toxic
substances, application of the developed ammonia biosensor to a toxic sensor
has been expected. Ortho-chlorophenol (OCR) was selected as a .toxic material.
It was solved in the ammonia nitrogen standard solution (1 mg/1). The re-
sponse of sensor output was observed when the concentration of OCR was added.
The results are shown in Figure 8.5. When the OCR concentration was below 0.5
mg/1, any decrease .of the current difference was not observed. The inhibitory
effect of OCR to the ammonia sensor was recognized over 0.5 mg/1. The
current difference of the electrode decreased with increasing OCR concentra-
tion. When the OCR concentration was over 4 mg/1, the current difference
became almost zero. After the OCR solution was removed, the current differ-
ence recovered to the initial current difference.
This experiment shows tha.t this ammonia sensor has the ability to detect
toxic materials. Additional experiments have been carried out to study the
detective concentration of other chlorinated organics.
8-4 ORGANIC ACID BIOSENSOR
A yeast of Trichosporon brassicae is used as a detector. The yeast is
immobilized between two porous acetylcellulose membrane filters. The biofilm
was sandwiched between an oxygen electrode and a gas permeable teflon membrane
filter. The yeast utilizes gasified organic acid and consumes oxygen in the
biofilm when the microbial electrode is immersed in^to an organic acid solu-
tion.
A linear relationship was observed between an organic acid concentration
below 600 mg/1 and current difference as shown in Figure' 8.6. The current
difference decreased with the ascent of pH from 3 in the sample and fell to
zero at a pH of 7 or more due to ionization of acetic acid. It shows pH in a
sample must be kept lower than 3.
The sensitivities of other organic acid samples were tested. , The
sensor could detect formic acid, n-butyric acid, propionic acid as well as
acetic acid. However lactic acid, tartaric acid, succinic acid could not be
detected. That shows it is difficult to convert the senson value to the
actual acetic acid equivalent value in solutions containing a mixture of
organic acids.
#
A field test in an actual digestion tank has. been carried out for the
check of maintenance . -
34
-------
9. DEVELOPMENT OF NEW WASTEWATER TREATMENT SYSTEMS
Bioreactors for wastewater and sludge treatment, liquid-solid separators
and biosensors are being developed in addition to the basic studies on the
microorganism bank and genetic engineering in Biofocus WT, as mentioned above.
Systematizing the individual development to^propose new wastewater treatment
systems is being investigated. Table 9.1 summarizes new wastewater treatment
systems to be proposed based on results of pilot plant studies. Table 9.2 is
the list of pilot plants in Biofocus WT.
Table 9.1 New Wastewater Treatment Systems to be Proposed in Biofocus W.T.
a. Energy-saving Type Wastewater Treatment System
b. Area-saving Type Wastewater Treatment System
c. Nitrogen Removal Type Wastewater Treatment System
d. Simultaneous Nitrogen & Phosphorus Removal Type
Wastewater Treatment System
e. Wastewater Treatment System for Small Flow
A self-sustaining energy system has been proposed a.s an ultimate system
of an energy-saving type wastewater treatment system. The system is illus-
trated in Figure 9.1. Most SS in raw wastewater are removed by a high-effi-
ciency liquid solid separator. Energy consumption in the secondary treatment
can be greatly reduced by two means. One has influent to the secondary treat-
ment containing not only less organics, but less particle organics which are
more difficult to decompose than soluble ones. The other is adopting an
energy-saving type bioreactor. ,
In sludge digestion, the feed sludge has a large amount of primary
sludge which has higher potential for generating digestion gas than that of
excess sludge. Also, ah energy-creating type bioreactor for sludge digestion
is employed, so that a greater amount of energy can be recovered. A key point
of the system development is how energy consumption of a high-efficiency
liquid solid separator can be reduced.
10. POSTSCRIPT
As Biofocus W.T. is a research project which intends to broadly apply
biotechnology to biological wastewater treatment for its improvement and
greater understanding, it includes various research subjects from very basic
ones to practical ones, as mentioned above. Therefore, results of Biofocus
W.T. will be varied, from those of identifying further research needs to
practically feasible technologies which may break through present technolo-
gies.
35
-------
.
e> *tt <— • c
J5 G o _o
o ** "-3
*"* "
I
S"
i
§
P
eo
CO
«ji cj « e
- - — «a o •-* *~N m je us -w o N
t»e *•» s«s eo co cd co 00 -C* cd
CU co CU *O "O ie *o *O *O C
fc >4 [—* L. L. _rt Id La la •-•
-
co ca
*-. U3
ae •<:
O*->
ca ca
csw
S S S OS 2
ro
> u a
g;
AH1ZA
£3
s s
°3
U3
(SI
S
ggj
ss^
fed
^
:3 ..S | S
S 2 o If c 5
Q CJ U3 O E _
to £. .- e= . —• .
•u M :
S.SS
L C C -J .
^. — »- — — ___i bu cd W ••*
.2 '5C ^ -5 S "= g tS « 1 S §
"£,5.5 =>S"| " JS J ^ " ?
^3 c t- > j= .c a. w cd
.o ca(ar5c:ci.je—*-> cd w
2513 §'
•a § "^3 -p "2
£ E"g § S £ £
••• O >• *O O)
SET O C 5 O C
Si "a> o — < — » a>
u w t> w bo
1 1 o: B S -
>•« >-< j^ ;>-" >-" cd
CO IO W W I
111 § i -I
3 11 f !° I
CQ CQ CO CO CO f2
as -^
ua »— t
-< ce
•— • o
> O t->
U3 O
i CD b4 ^ O I
o =e PS «-* .
« S S S
•-• cu co ce :
•
S o 3
: ce e?
iSSS'
, ts3 IK tx3 C_3 ta3
I •<; rO i—' CQ
r-> OS *— CO
•-3 eti co ca to ca
f—i •—I OS U3
> CJ L? J3 t»3 X
. •—> -< Kj •< I—I
^- o CQ ua t*. 55 u.
> •" SC 01
t, «C CO
. cd
't. «,
SE
«*
'O *-< •""
«j cd *>
;S=F e
O cd Q
s .! §•
.
*O O *-i
coo
-
t cd c^ cd cd I
JO *->
- «
"
—
= e c
*~ *~ S:
3SS
:-t- o a -->
• o o e
*" 0 — —
t< Jtf (0 CU
a » o cd
'S* O >£3 *J
o >» o —
DO O tA
S •§ 0
e 5 *>
0) t-
Z 0 i
£--5
I Si
So 5
OS F—• H^
CO CO ^
{— b3 O
sate
~ -J U3
£ S
3 .
'"- jc to td
J-t_ —• t- t-
o se o o
•*H —' n
'^ g J2 fe
.2'^g"
.O to C cd
I i « S
.
CO J3
,.5-g
.
d) C Oi "^
I- — C t-
o g ° a
• -
g
s
<§
S'S. S
in E «—«
^^ Cu ^"
to os w
I-I
CO tS *CQ
>-« cc ca
a M w
CU .—3 CQ
:= p* ea
s"l
S 5 «<
as o
£
ri B -
S~3 b=
U. C?
ca o •«:
u ca co
u o '' =
I s^~.
-•<-. -r • H
•"^ C O T3
B5 g.5
—3 =3 bO rt
"x.. W CD a
.•i 1 i s
•5 s s s
1 is
•^'S
II
O 0)
s
EV
fe
) c
d tj)
-------
1/100
30C
24h
+
sot;
24h
lOal
Oppo glucose
Figure 3.1 Diagram of'Serial Transfer Culture
as
cs
u
uu
u
ULJ
o.
CO
1- blattae
C- freundij IFO 13545
10 20 30 40
NO. OF TRANSFERS
50
'2 ,Chanfes of specific growth rates of E. blattae and C. freundii
isolated from the serial transfer cultures (i.e., from their single species
cultures, not a mixed culture)
-10'
01
u
bQ
DB E. cloacae i.
1-70
10 20
NO. OF TRANSFERS
Figure 3.3 Changes in population densities of E. cloacae and E. blattae in their
mixed culture. Both were isolated from the serial transfer cultures of 70 times
transfer
37
-------
ME77B7 JE5879
;• i I Hfr
01 [01 10
hy30 hyBO hylOO
Figure 3.4 Schematic Diagram of the- Relationship between Parental Strains and
Recombinant Strains
-10'
OJ
u
2
uu
en
totalCstt-1
0 30 CO 100 180
TIME [minutes]
Figure 3.5 Changes in Density of Recombinant Cells in Mixed Cultures
Q
hylDO-7
05 10 15 20
TIME Chours]
Figure 3.6 Growth Curves of ME7767, hylOO-7, -8, and hylOO-9
38
-------
200 r
o
(3
T3
-------
0)
cs
c
o
2
0}
ec
0.5
J_
_L
_1
0 10 20
ACM Concentration (%)
Figure 4.4 Effect .of ACAM Concentration on Respiration Rate
Microorganisms Immobilized by Agar-ACAM Monomer Method
of
a
ca
e
o
w
0)
ca
R)
•-H
CD
ACAM Concentration
Figure 4.5 Effect of ACAM Concentration on Respiration Rate of
Microorganisms Immobilized by ACAM Monomer Method
337days
141days
(m)
Height from the Bottom
of the Reactor
t •
Figure 4.6 Distribution of Sludge Concentration in UASB Reactor
for Treatment of Supernatant from Heat Treatment of Sludge
40
-------
Liquid Oxygen
Gas Recirculation
Pump Gas
Discharge
Bioreactor
CMl
Effluent
Oxygen
Absorption
Tank
Feed Pump
Figure 4.7 Flowsheet of AUSB Process
Influent
3
•
Air
diffuser
\ '
i:
X
/ Biological reaction vessel
/
f
•
4
s.
Xi
. . Sludge Blanket
1\
/
/
/
/
/
/
/
/
/
/
/
s
/
/
/
/
•
.
•.
i
t
0
^
* X
k
J
A
/
/
1
/
>
^
/
/
/
Z
X
-.
-.
.
•^
6
N
U
^
^
^
^
/
',
/
/
/
&
r-
'
'
O
^N
L
rr
7
^
^
^
/
',
/
/
/
^i
i*.
s
•;•
•
o
\
y
^^
VI
N;
/
/
/
/
y
/
'
^
^
h
Effluent
Figure 4.8 Flowsheet of MRB
41
Sludge Blunket
-------
Figure 4.9 Self-Granulated Sludge (SGS)
Figure 4.10 Microscopic View of Film
42
-------
Organic
Substrate
CO,
Anaerobic Zone
Sulfate Reducing Bacteria
Anaerobic Hydrolysis
\ /
Organic Acids
Micro Aerobic Zone
, V
Sulfide Oxdizing Bacteria
S
Beggiatoa
Figure 4.11 Mechanism of Self-Granulation
43
-------
Washed Sludge
Bioreactor
Solid-Liquid Separator
Influent
Back Washing Water
Storage Tank for Back Wash
Sludge Thickener
Figure 5.1 Flow Diagram of Fixed Bed Type Aerobic Bioreactor
Influent
Air
Screen
ra
•^H
TJ
Fluidizing
Polypropylene Me
w
'!'!'!'ii!i!'!i!'!^
:X*4ftX;;*S55tt
i;4i^iji'.i;j|frW
t
V
•^-
A
Storage Tank
for Back Washing
r= 1 Effluent
->-r-
Back Washing Water
Bioreactor
Plastic Media Filter Sand Filter
Figure 5.2 Flow Diagram of Fluidized Bed Type Aerobic Bioreactor
44
-------
Recirculation of Effluent
Screen
Influent
Oxygen
Effluent
•a
CD
«
•—I
c
CO
0
«H
r—I
CD
w
U <"
" &
Jd T3
O D
L, "^
CD W
<
Pressurized Oxygen Tank
Excess Sludge Tank
Bioreactor
Figure 5.3 Flow Diagram of Aerobic Sludge Blanket Bioreactor
Influent
Gas
******* ******
*************
»*****«******
******* ******
******* ******
»»****< ******
»*****« **«*««
>*****« ******
»****+< ******
»«****<«****«
*****< ******
Effluent
Plastic Braids Media/
Air
Final Sedimentation Tank
Bioreactor Contact Aeration Tank
Figure 5.4 Flow Diagram of Fixed Bed Type Anaerobic Bioreactor
• • " 45
-------
+J
0)
i—I
«H
f*
<»-4
O
O
I
I
TJ
0>
Influent
Sulfide Removal
Gas „
Bioreactor
Sedimentation Tank
Aeration Tank
""] Aerobic Bio-Filtration
Vacuum Tank
V
r
*,
ir
s
^
0)
4->
a)
v
^7
Ay
.^
>
II
' P
00
-»H
in
ra
CQ
Effluent
Storage Tank
for Back Washing
Figure 5.5 Flow Diagram of Fluidized Bed Type Aerobic Bioreactor
Bioreactor Modified Trickling Filter
ra
Effluent
Storage Tank for Back Washing
Excess Sludge Storage Tank
Figure 5.6 Flow, Diagram of UASB with Modified Trickeling Filter
46
-------
Influent
Sedimentation
Tank Effluent
Exess Sludge
'Sand Separation Part
Reaction Part
Reactor
:-Air
Figure 5.7 Flowsheet of Pilot Plant
c
CD
bfl
o
100;
80
60
40
20
0
40
30
20
10
0
40
30
20
Eff. fronT Reactqr
Influent
Final Eff.
-N in Influent
100
110
120
130' Elapsed Days
Figure 5.8 Nitrogen Removal
47 .
-------
Recirculation
Influent
Denitrifi-
cation
Tank
V
Sedimentation
Tank EffIuent
Nit'rifi-'.
cat i on
Tank' •'..'
"Pellets
'—Air
Return Sludge
"Exess Sludge
Figure 5.9. Flowsheet of Pilot Plant
40
"S 30
2
i
X 20
Z
10
H-
RUN-1
Influent NH*
Effluent T-N
RUN-2
: Iuent T-N
Effluent NH4-N
Feb.l 10 20 Mar.l 10
DATE
Figure 5.10 Performance of Nitrogen Removal
48
-------
[Pretreatment]
No.l "a? — =>| HT
3 60°C-11
F-H
w
•a
No. 2 •£ — > HT
£ 90°C-1
S Milling
"-»• £-» 4Mn.
w
T
h
i
[Ac
Pr
»v
:id Fermentation
Process]
CMR 35° C
CMR 70 °C
^
otease
CMR
\
37° C
[Methane Fermentation
Process]
•v pi qrr o
1 FX 55°
• SI S
0)
'— »«•
w
•a .
01
«
C >§,
•^c
w-»
7°C
(Digested Sludge}
(Digested Supernatant}
]iJ:Heat Treatment
£MB.:Complete Mixing Reactor
E^iFluidized Bed Reactor (Cement Ball)
SLS:Solid Liquid Separator
EXrFixed Bed Reactor (Ceramic)
Figure 6.1 Flowsheets of Pilot Plants on Anaerobic Sludge Digestion
Supernatant
Sludge
4
Centrifuging
i Deionized water (make 100 ml sample)
I Sulfur Ig
Sulfur bacteria
Iron bacteria
Shaking culture
I
30 °C for 0. 3. 7, 14 days
I
Centrifuging
I
Sludge
I
LiquoKContaining leached metals)
Measurement
Figure 6.2 Procedure of Bacteria Leaching of Sludge
49
-------
14 (day) oi 3
100
14 (day) oi 3
100
a
14 (day)
013
14 (day)
14 (day)
01 3
bacteria not added, mixed sludge
bacteria not added, digested sludge
bacteria added, mixed sludge
bacteria added, digested sludge
. ,O --- O.pH = 4
chemical /n --- Q pH = 3
leaching I A^ — A pH = 2
PH = 1
14 (day)
Figure 6.3 Changes of Leaching Ratios of Heavy Metals
50
-------
A nim>
Concentration(mg/l)
300 250 200 150 100
50
B MWTP
Concent ration(mg/l)
50 100 150 200
50 40. 30- 20 10
6 5.4 3 2 1
BOD
week max.flow
day min.flow
Holi max.flow
-day min.flow
COD
week max.flow
day min.flow
Holi max.flow
-day min.flow
SS
week max.flow
day min.flow
Holi max.flow
-day min.flow
T-N
week max.flow
day min.flow
Holi max.flow
-day min.flow
T-P
week max.flow
day min.flow
Holi max.flow
-day min.flow
Figure 7.1 Distribution of Particulate Contaminants in Raw Wastewater
10 20 30 40
1 2 34
BOD
s
L
i M
\ L
: L
5 L
S
S
S
composition rateCO
SO 80 TOO
\~
S
uj L
5 M
:-L
I L
- L
S
S
s
Pl^xsgro^ggpiafl
g C '
composition rate(!0
40 60 80 100
D i
1~!0« 10-74 u 74 «<
particle size,
D i
l~10tf 1D~7<« 74 u<
particle sire
Figure 7.2 Comparison of Distribution of SS and Particulate BOD
in Raw Wastewater by Plant Size
51
-------
• 2mm-74>m
M 74jum- 8>mi
A 8/mi- IJJITI
ODissolved
Figure 7.3 Difference of Biodegradation Rates of TOC due to
Particle Size of Contaminant in Aerobic Condition
• 2mm-74>m
• 74jum- 8}im
A 8/rni- Ijurn
ODissolved
TIME(h)
Figure 7.4 Difference of Biodegradation Rates of TOC due to
Particle Size of Contaminant in Anaerobic Condition
52
-------
sa-
ri-
a-
n-
a-
•SB-
f f:fV
4
k Y
I 4
i
upflow rate(cm/min.)
(1) SS removal rate and upflow rate
«•
4-
i-
l-
1
1 i
I
T
>i j
i
™ • * >1 • 13
upflow rate(cm/min.)
(2) SS concentration in sludge blanket
and upflow rate
Figure 7.5 Performance of Upflow Sedimentation Column
100
90
8 80
o>. '70
2 60
~a
§ 10
§ 30
20
10
8
-10
-
•
-
^
1
$
[;'
5
•5;
1
i*
£
':'•!
%
Fl
v
&'
i=j
1
{*i
lj
V*
S
V
c:
M
!T
particle size
C3
D 8-71
& 1-8
i
(2) (3) (4)
Separators in deveroping .
Figure 7.6 SS Removal by Three Solid-Liquid Separation Processes
~ 53
-------
CPU
amplifier
A/D transfer
Microbial electrode
Isothermal water bath
Drainage
Standard solution Buffer solution
Figure 8.1 Flow Chart of Biosensoring System
54
-------
200
150
•*•>
C
(U
50
O
u
a
o
CQ o
GGA standard solution
0.5 1.0 1.5 2.0
current difference (uA)
2.5
20i
150
100
synthetic wastewater
- 2
50
0 0-5 1.0 1.5 2.0 2.5
current difference (uA)
Figure 8.2 Performance of Three Selected Biofilms
• river water
A secondary effluent
150
200
10 20 30 40 50
sensor BOD (mg/1) sensor BOD (mg/1)
calibrated with GGA solution calibrated with artificial sewage
Figure 8.3 Relationship between Sensor BOD and Manual BOD
55
-------
0)
TJ-
O
O
•— i
13
•P
O
0)
Q)
J-,
1.0
OCP concentration (mg/1)
Figure 8.5 OCP concentration and Current Difference of Ammonia Sensor
56 '
-------
500T
400
4)
O
£ 300
~tion
more
>s
jy
combustible
gas
Generation
of
electricity
uti nzation or
electricity
Utilization o
waste heat
Sludge
Utilization
Figure 9.1 Self-energy Sustaining Wastewater Treatment System
57
-------
-------
FUNDAMENTAL STUDY ON APPLICATION OF HIGH GRADIENT MAGNETIC SEPARATION
TO MUNICIPAL WASTEWATER TREATMENT SYSTEM
Kazuo HOYA, Director of Engineering Division,
Regional Sewerage Center
Yasuo TANAKA, Ph.D., Researcher of Research and
Technological Development Section,
Planning Division
Khouichi ISHIDA, Assist. Head of Research and
Technological Development Section,
Planning Division
Sewerage Bureau
Tokyo Metropolitan Government
The work described in this paper was not funded by
the U.S. Environmental Protection Agency. The contents
do not necessarily reflect the views of the Agency and
no official endorsement should be inferred.
Prepared for Presentation at:
12th United States/Japan Conference
on
Sewage Treatment Technology
October 1989
Cincinnati, • Ohio
59
-------
Abstract c „
A research was carried out on the development of small land
required treatment process which is applicable for both combined
sewer overflows treatment in wet weather and tertiary treatment of
secondary effluent in dry weather using high gradient magnetic
separator. Experiments with the bench scale and pilot scale
apparatuses of high gradient magnetic separation process exhibited
that the pollutants were removed sufficiently for both combined
overflow treatment and tertiary treatment. The results of the
experiments suggested that an optimum flow velocity of the magnetic
separator was about 2000 to 3000 m/day and appropriate magnetite
concentration was about 25 mg/1 for combined sewer overflow
treatment and about 5 mg/1 for tertiary treatment. Preseparation of
large floes by high rate sedimentation prior to magnetic separation
was considered to be effective measure to lengthen the filter run of
the magnetic separator for combined sewer overflows treatment.
Optimum overflow rate of preseparation was about 800 to 900; mVm2
• day. Sludge produced from the treatment process could be thickened
and dewatered easily. Preliminary design was carried out for a
demonstration size plant with a 10,000 m 3 /day treatment flow and
full scale plant with a 100,000 m3 /day treatment flow.
60
-------
1. INTRODUCTION
In recent years, - several kinds of technological improvements
has been required for wastewater treatment plant in Tokyo
Metropolis. One is the improvement of effluent water quality at
storm weather. Collection system. in the ward area of Tokyo
Metropolis is combined type, hence large volume of wastewater is
discharged from treatment plants at storm weather after only
primary treatment. This causes the pollution of public water body
at storm weather. The other is the introduction of advanced
treatments' far the prevention of eutrophication and for the reuse" of
effluent for aesthetic, recreastional and subpotable water supply .
purposes. Removal of phosphorus, BOD, COD, turbidity, chromaticity
and coliform bacteria is especially important for these purposes.
However, treatment plants in Tokyo do not .have enough space to
install many new facilities, hence developement of a treatment
technology which is small land required and applicable to above all
purposes is important nowadays.
High gradient magnetic separation using seeding material seems
suitable technique for satisfying above technological requirements
because of its high processing rate, high effluent water quality and
adaptability for wide range of influent water quality. The theory of
"High Gradient Magnetic Separation" was established by Oberteuffer
(Oberteuffer, 1973). Currently, a high gradient magnetic . separator
(HGMS) based on this theory is applied for removing magnetic
suspended solids in discharge from steel industry and for eliminating
collosion products in the feed water of power plant. High gradient
magnetic separation is also applicable to municipal and industrial
wastewater treatments for the removal of non-magnetic contaminants
using magnetic seeding method. In this method, non-magnetic
pollutants in water are coagulated with magnetic seeding material by
coagulant and subsequently removed by HGMS. Magnetite (Fe3 O 4 )
has been usually used as seeding material in the seeding method
because of its large magnetic moment. Chemical coagulation of the
magnetite and pollutants is made by alum and polyelectrolyte.
Principal advantage of the seeding method is its 5high processing
rate and high removal rate of suspended and colloidal matters.
Seeding method has been investigated by many researchers. Allen
et al. (1977) and Allen (1978) investigated about application of
the seeding method for combined sewer overflows treatment. According
to them, the seeding method was effective for reducing most forms of
pollutants present in a combined sewer overflows. They emphasized
that the treatment system has several additional benefits, namely,
high, processing rate, small land requirements, and lower chlorine
demand. Several studies have been also carried out on the ^removal of
phytoplankton in. eutrophic seawater or freshwater (Kurinobu &
Uchiyama, 1982; Kaneko et al., 1981; Kaneko, 1982). These studies
61 - ' ' .
-------
confirmed the excellent efficiency of the seeding method for
removing phytoplankton.
The seeding method, however, lias not received practical
application because of following two problems. First is high cost of
chemicals, especially for magnetite. Reducing the concentration of
magnetite or using low price seeding material instead of magnetite;
are important to reduce the chemical cost. In addition to this,
selection of the most suitable polyelectrolyte may be important for
reducing the required amount of seeding material because
characteristics of polyelectrolyte strongly affect the mechanical
strength of coagulated floe. Second is that filter run of the
magnetic separator are impractically short when the feed water
contains high concentration of aggregates. Preseparation of large
aggregates before magnetic separation or increase in the matrix
thickness should be required for lengthen the filter run.
Tokyo Sewerage Bureau has been carrying out the research about
the seeding method since three years ago. The main objectives of the
research are determination of optimum treatment conditions at low
magnetite concentration and establishing the treatment flow
applicable to both combined sewer overflows treatment and tertially
treatment. The present paper outlines the results obtained by this
research.
62
-------
O
4—
ro
ro
Q_
-------
2. COMBINED SEWER OVERFLOWS TREATMENT
2-1. Bench Tests
Scope of testing
Bench tests were carried out to obtain basic data "to about an
optimum dosing rate and treatment conditions.
Bench test apparatus
Fig.2 shows the appearance of a batch type apparatus used for
t,he bench test. The flowsheet of the apparatus is indicated by
Fig.3. Flocculatiori tank was 5 Jitter in capacity. Size of mixing
blade in the flocculation tank was 15 cm in length and 6 cm in
width. The matrix for the high gradient magnetic separator consists
of 100 pieces of SUS 430 expanded metal disc which were 10 mm in
diameter (Fig.4). The matrix is 80 mm in thickness and 80% jn void
ratio. Electromagnets are ori both sides of the matrix.
Test procedures are described as follows. Magnetite powder (1
to 4 urn in diameter) were added to 5 litter of raw water in the
flocculation , tank and mixed by tho mixer at 150 rpm for one minute.
Then, alum was added and flush mixed at 150 rpm fro 15 seconds
followed by gentle mixing at 50 rpm for 1 minute. Finally,
polyelectrolyte was added and stirred at 150 rpm for 15 seconds and
50 rpm for 2 minutes successively. The flocculated slurry was pumped
to. the magnetic separator by a filter pump which was .located
downstream from the magnetic separator to avoid the disruption of
the floes. Effluent from magnetic separator was sampled continuously
for chemical analysis untill the breakthrough. After the
breakthrough, matrix was backflushed with tap water.
Alum concentration
Ordinary jar, test was performed prior to the experiments with
the test apparatus to clarify the relationship between the alum
requirement and SS value of raw sewage. Fig.5 shows one of the
results of the jar test. In this case, optimum concentration was
estimated for about 5mg/l because the transparency of supernatant
reached to a plateau at -tliis concentration. Relationship between SS
and optimum alum concentration is shown by Fig.6. Optimum alum
concentration fluctuated in narrow range (3 to 7 mg Al/1) while the
SS values fluctuated considerably (100 to 600 mg/1).
Magnetite concentration
Fig.7 shows relationship between the magnetite concentration
and the removal rate of totaj organic carbon (TOG). Magnetite
concentration had almost no effect on the removal rate and also on
the filter run in the range between 50 to 400 mg/1, hence magnetite
requirements should be less than or equal to 50 mg/1. Further tests
showed that magnetite concentration can be reduced to about 25 mg/1.
This optimum value is considerably lower than tliat recontended. by
Allen (1978) (200mg/l). Lower flow velocity (2000m/day) comparing
with that of Allen (5000 m/day) should be cause of low magnetite
64
-------
demand in the present study. In Japan, price of magnetite is
considerably higher than that in the United States, hence, it may be
unavoidable to reduce the flow velocity. Reduction of magnetite
concentration is also important from the viewpoint of cutting down
the amount of sludge generated from the process.
Polyelectrolyte concentration
To select suitable sort of polyelectrolyte, ' filter run and the
transparency of effluent were compared for many kinds of
polyelectrolyte. As . shown in Table. 1, anionic polyelectrolytes were
generally superior to amphoteric and cationic ones. Among the
anionic polyelectrolytes, P-720 had especially excellenct
performance, hence it was used for all of remaining tests.
Fig. 8 shows the effects of P-720 concentration on the filter
run and effluent water quality. Polyelectrolyte' concentration had
almost no effect on the effluent. water quality within the range of
0.5 to 3 mg/1. The filter run, however, decreased markedly under the
concentration of 2 mg/1. It is concluded that the optimum
concentration was 2 mg/1.
Flow velocity of high gradient magnetic separator
Fig. 9 shows the effects of the flow velocity on the filter run
and effluent water quality. There was no observable effect of flow
velocity on the effluent water quality, except a slight increase of
total phosphorus over 1500 m/day. On the other hand, the filter run
markedly dropped at the velocity over 1500 m/day.
Magnetic field strength
As shown in Fig. 10, magnetic field strength has little effect
on the effluent water quality, while filter run increased markedly
with the strength. Electrical power consumption 'increases with the
magnetic field strength, hence optimum magnetic field strength must
be determined by the consideration of total running cost.
2-2. Pilot Plant Tests
Scope of testing
To obtain the fundamental data for the design of full scale
plant, a continuous flow pilot plant was operated on optimum
conditions which had been determined by the bench test.
Pilot plant facility
Fig. 11 and Fig. 12 s"hows a flowsheet and side view of the pilot
plant. Raw sewage taken by submerged pump was applied to automatic
flushing strainer with 0.5 mm opening J,o remove rubbish, and then
pumped to flush mixing unit. After addition of magnetite and alum,
the mixture was stirred for 3 minutes. Coagulated slurry was fed to
a floe preseparation tank (preseparator) after addition of the
polyelectrolyte (P-720). Overflow rate of the preseparator was about
65
-------
800 m3 /ra2 • day. The preseparator consisted of three layers;
settling layer in which large sized floes settle down, helically
flowing flocculation layer in which feed slurry flew into and
flocculation proceed, and sludge blanket layer by which fine floes
was removed, in order of height. Remaining floes in the effluent
from the preseparator were removed completely by the magnetic
separator. Matrix of the magnetic separator was 150 mm in diameter
and 150 mm in thickness. SUS 430 expanded metal plates which mesh
size was 13 mm in length and 5.1 mm in width were used for the
matrix. Operating conditions of the magnetic separator were 2000
m/day for flow velocity, 3 KG for magnetic field strength and 1 hour
for filter run. Backflush of the magnetic separator was made with
mixture of pressurized water and air for 10 seconds with flow
velocity of 2000 to 3000 m/hr.
Optimal parameter runs
Table 2 shows the results from an optimized pilot plant run of
6 hours continuous operation using raw sewage at dry weather as
influent. The results show that high removal rates (more than 90%)
of SS, turbidity, phosphorus, coliform bacteria can be achieved by
the seeding method. Removal rates of BOD and COD were lower than
those values. However, it was confirmed by bench test that higher
removal rates of BOD and COD could be obtained in wet weather,
probably because the ratio of soluble COD and BOD to suspended ones
decreased markedly in wet weather. Fig. 13 shows the counts of
coliform bacteria in effluent for 19 runs operation. The counts were
much lower than 3000 no./ml, which is the regulation number in
Japan, hence addition of. chlorine to the effluent should be
avoidable.
The approximate amount of sludge generated from the process was
estimated for about 140 to 310 g-dry weight per 1m3 of the
influent. Seventy eight .percent of the total sludge was generated
from the preseparator and remainings from the magnetic separator. As
shown in Table 3 and. Fig. 14, the generated sludge had excellent
thickenability and dewaterability.
It was confirmed that about 99% of trapped floes in the matrix
were washed out by the backflushing which was made with the mixture
of effluent water and two times volume of pressurized air. Little
amount of cellulose fiber however remained in the matrix after the
backflushing. It has not been, clarified yet whether these fibrous
materials causes the clogging of matrix after long time continuous
operation. Ball shaped matrix seems -to be more convenient than the
expanded metal matrix to prevent clogging because tangle of
cellulose fiber should not occur. Test of ball matrix is earring out
now. cc
oo
-------
Fig. 2. Bench test apparatus.
67
-------
D,
c5
H
1
(
'
:j
[
C
s
a
f.
1 1
•3
S
{'
i,
t
I
— i
=&<
/<
-I
vj
1
-
Ij
r"
'<
^*^i
\.
^Xe
?Y
/
S^.
fl
^"
r
4-
(
't
S
1
(
--.
XX
/>
>
)
H ,
L,
. t
1
J •
H
5
-{
-i
4
r
a
i
A
J ^5
+j
^H
s-
' [|III|I"|1II|IU| |^
1
•J !
ib ,.,
X
s
1 '
txi=
CD
•P
ca
s.
a
CO
g
+J
•p
I
0
CO
5
CJ
o
cd
68
-------
Fig. 4. Matrix (16 mm0 x 80 mm H) of high gradient magnetic
separator used for bench test.
69
-------
>100
CX
0
0
8
10
Alum Concentration (mg-Al/1)
Fig.5. Effect of alum concentration on transparency and
SS of supernatant in jar test of raw sewage.
(magnetite: 25 mg/1, polyelectrolyte: 2 mg/1)
70
-------
10
fi 8
O
cd
£
§
s
i—i
<
•H
I
•o o
• o
I
I
o
O I dry weather
© :wet weather(first flash)
a ;wet weather(after first flash)
I . I . I . I
100 200
300
400
500
600
SS (mg/1)
Fig. 6. Relationship between SS concentration of raw sewage
and optimum alum concentration.
71
-------
Removal Rate (%)
80
60
40
20
0
1 1 II
0 100 200 300 400
Magnetite Concentration (mg/1)
Fig. 7. Effect of magnetite concentration on removaJ. rate
of total organic carbon (TOG).
Test conditions
Magnetic Field Strength: 5KG
Flow Velocity: 2000 m/day
Alum Cone.: 8.6 mg/1
Polyelectolyte Conc.(C-599-lP): 1 mg/1
Influent TOG: 73 mg/1
72
-------
Table 1. Polyelectrolyte comparison.
Test conditions
Magnetic Field Strength : 5KG
Flow Velocity : 1400 m/day
Magnetite Cone.: 25 mg/l
Ferrous Alum Cone.: Al-8mg/l,Fe-3mg/l
Polyelectrolyte : 2 mg/l
Table l-l. Anionic Polyelectrolyte
Name
I' - 700
P - 710
P - 713
P - 720
P - 730
P - 790
©
®
©
®
© :
©
®
®
©
©
©
®
©
®
©
©
©
Characteristics
Chemical composition (mole%)
DMC
80
80
~0~~
40
DM -Ac
'
80
80.
40
40
-
A A ii
55
55
55
55
60
60
80
90
90
80
95
AAc
20
20
5
5
20
20
5*
5
OAC
40
40
20
io
10
20
5
others
M.VV.
Mt*)
300
500
300
500
300
500
300
400
21)0
400
500
300
600
700
600
Filter run
(relative
value )
34
34
34
29
- 47-
.28
44
55
.46
66
-9
84
136
29
19
'rarspaierty
relative
value)
-
—
-
-
85
—
—
83
75
89
75
85
60
—
-
73
-------
Table 1-3. Cationic Polyelectrolyte
Name
c - 100
c - 200
c - 201
C - 202
C - 252
C - 270
C - 300
C - 301
C - 340
C - 341
C - 343
C - 360
C - 373
C - 380
C - -155
C - 500
CA
DC-802P
UC-753P
1A1-883P
DM -254 P
®
©
(D
©___
"C~5U9-1P
CP-G58
Characteristics
Chemical composition (mo(e%)
DMC
100
100
70
100
40
20
60
50
~80
100'
100
100
95
•~95
DM- Ac
100
100
40
20
20
5
5
A A in
50
30
60
60
60
40
30
20
50
40
'
AAc
DAC
50
others
S3 100
*3 50
*4 100
$5 100
•KS 100
*6 60
*7 100
M.W.
Mb*)
100
200
200
200
200
200
300
300
300
300
300
300
300
300
400
500
1
1
8
7
25
500
20
200
400
1000
Filter run
(relative
value )
19
21
23
28
26
32
23
33
60
48
45
50
43
36
71
28
19
13
14
17
25
76
18
12
17
27
73
rrarspaimy
(relative
value)
—
—
—
-
-
-
-
72
—
-
-
-
—
83
-
—
—
-
—
-
68
' -
—
-
68
64
74
-------
G 20
S
10
80
0
fi 60
0)
fc
cd
ft 40
CO
G
03
ti 20
H
30
20
10
0.3
bo
cd
4->
0.2
2
Turbidity
ss
__ 0
^ . 0 1 2 3
Polyelectrolyte Cone. (mg/1)
Fig. 8. Effects of polyelectrolyte concentration on filter
run and effluent water quality.
Test conditions
= Magnetic Field Strength:5KG
Flow Velocity : 1400m/day
Magnetite Cone. : 25mg/l
Alum Cone. : 8mg/l
Polyelectrolyte : P-720
75
InfluentCSettled Sewage)
Transparency : 5.5
Turbidity : 95
SS : 34 mg/1
TOG : 77 mg/1
Total P : 5.3 mg-P/1
-------
+3 20
S 10
fc,
1000
1500
2000
C.100
g 50
2 ol—
E-, 1000
14 -
12
10
1500
2000
W/i
TOC
Turbidity
o o-
.55
1000
1500
2000
CU -N o.2
1000 1500 2000
Flow Velocity (ra/day)
Fig. 9.. Effects of flow velocity on filter run of high
gradient magnetic separator and effluent water
quality.
Test Conditions fj
Magnetic .Field Strength : 5KG
Magnetite Cone. : 25mg/l
Alum Cone. : 8mg/l
Polyelectrolyte Cone. (P-720) : 2mg/l
Influent (Settled Sewage)
Transparency : 5.5, Turbidity : 93, SS : 62rag/l
TOC : 52mg/l, Total P : 3.8mg/l
76
-------
30
C ^
<§ "$ 20
I | 10
£ S
M .... . . L _J
^ 0 1 23 4 5
0
§150
CB 10u
W 50
c
ca °
.
*
i i i . i r
n i 9 Q .1 c
30 r
20
10
•TOC
Turbidity
0.3
0.2
0.1
012345
Magnetic Field Strength
(KG)
Fig.10. Effects of magnetic field strength on filter run of
high gradient magnetic separator and effluent water
quality.
Test Conditions
Magnetite Cone. : 25mg/l
Alum Cone. : 8mg/l
Polyelectrolyte Cone. . (P-720) : 2mg/l
Influent (Settled Sewage)
Transparency : 6, Turbidity : 97, SS : 41mg/l
TOC : 68rag/l, Total P : 5.9mg/l
77
-------
CD
CO
CD
C
•H
fi
§
0
cd
a
a)
en
I
cd
-H CD
fd
Z
bi
E
-P
§
78
-------
TOf
-©
i
0)
w
O
3
I
_ra
a +J
, o>
«M
o
II
W
w ••
E
79
-------
Table'2. Results of continuous operation of pilot
plant for combined sewer overflows treatment.
Influent
PH
Transparency
Turbidity
Chromatic! ty
SS (mg/1)
COD (mg/1)
BOD (mg/1)
Total P (mg-P/1)
Coliform bacteria
(No. /ml)
7.3 -
r\
o -
81 -
30 -
108 -
74 -
106 -
25 -
2xl05 -
7.8
5
202
50
190
109
177
4.2
3xlOE
Effluent
6.7
more
2
18
1 ess
21
21
0.1
13
- 7.0
than .150
- 7
- 24
than 5
- 50
- 51
- 0.2
- 16
(Operating conditions)
Influent : Raw ,se*wage at dry weather
Magnetic Field Strength :• 3KG
2000* m/day
6 mg-Al/1
: 26 mg/1
Cone. : 1.8 mg/1
Flow Veloci ty
Alum Cone. : 7
Magnetite Cone.
Polyelectrolyte
Backflush
Backflush
cycl e
T i me :
1
10
hour
seconds
80
-------
8
T: AOOO
$
0
ca
E
O
2000
rv
0.
Ln H
'88-
Regulation Number (3000 cells/ml)
n .n ,m n : n
-'88. '89-
-fi n n'n • n.
£
Experimental Day
Fig. 13.
Number of coliform bacteria in effluent.
(Influent : raw sewage at dry and wet weather)
81
-------
Table 3. Characteristics of sludge generated from pilot
plant for combined sewer overflows treatment.
Raw sludge
S3
'2000
mg/1
Thickened sludge
pH
TS
SS
VSS
Coarse suspended
sol ids
Calorific value
6.5
28600 mg/1
26600 mg/1
17290 mg/1
24%
3690 Cal/g-dry weight
(gravity thickening for 12 hours)
82
-------
I
3
IQ
•p.
i
%
«H
0
£
(!)
-
-------
3. TERTIARY TREATMENT OF ACTIVATED SLUDGE PROCESS EFFLUENT
3-1. Bench Tests
Scope of testing
The bench test was carried out to obtain the basic data
about the best dosing rate and treatment conditions.
Bench test apparatus
The bench test apparatus described above was used for also the
experiment of tertiary treatment.
Alum concentration
Fig. 15 shows the relationship between alum dosing .rate and
removal rate of total phosphorus. It is obvious' that the total
phosphorus concentration became lower than 0.2 -mg/1 when the
mole-ratio of alum to total phosphorus is higher than 2.5.
Magnetite and polyelectrolyte concentrations
Fig. 16 and 17 shows the effects of the magnetite and
polyelectrolyte (C-599-1P) concentrations on the removal rate of TOC
and total phosphorus. In the case of poor water quality influent
(Fig. 16), removal rate increased markedly with concentrations of
polyelectrolyte and magnetite. On the other hand, in the case of
improved water quality influent (Fig. 17), high removal rate -was
obtained regardless of the concentrations of polyelectrolyte and
magnetite. Optimum magnetite concentration should be ranges between
5 to 10 mg/1.
Fig. 18
shows
the
effect
of
polyelectrolyte
(P-720)
concentration on the filter run and effluent water quality. Effluent
water quality improved with polyelectrolyte concentration in the
ranges between 0 and 0.5 mg/1, and became stable in the ranges from
0.5 to 1 mg/1. On the other hand, filter run increased steadily with
polyelectrolyte concentration in the ranges between 0 and 1 mg/1.
These results suggest that appropriate concentration is 0.5 to 1
mg/1.
Flow velocity of high gradient magnetic separator
As. shown. in Fig. 19, flow velocity had almost no effects on the
effluent water quality. The filter run however dropped markedly with
increase of the -flow velocity. It is estimated that filter run will
be about 30 minutes at flow velocity of 2000 m/day when using the
matrix of 15 cm thickness, which is standard value of full scale
magnetic separator. This filter run should be practical for the
operation of full scale magnetic separator, hence it is assumed that
optimum flow velocity is 2000 m/day.
Magnetic field strength >
Fig. 20 shows the effects of the magnetic field strength on the
84
-------
effluent water quality and filter run. Though the effluent * water
quality was not affected by magnetic field strength, filter run '
increased" considerably with magnetic field strength. Filter run
became rather stable over 3 KG, hence this value should be
appropriate for practical operation.
3-2. Pilot Plant Tests
«
Scope of testing °
A continuous flow pilot plant was operated with the optimum
conditions determined by the bench tests to obtain a design
parameters of full-scale plant.
Pilot plant facility
Fig. 21 shows the flowsheet of the pilot plant. Magnetite and
alum were dosed successively and mixed by a static mixer. The
mixture was gently stirred by helical flow in flocculation tank for
4 minutes. The flocculated slurry was fed to the magnetic
separator after addition of polyelectrolyte. Magnetic separator was
operated with, the flow velocity of 2000 m/day and the 'magnetic field
strength of 3 KG. The matrix of magnetic separator was 10 cm in
diameter and 15 cm in thickness. When the turbidity of the effluent
increased markedly, backflushing was carried out with 40 litter tap
water and 120 litter of pressurized air for ten seconds. Fig. 22
shows the side view of the pilot plant.
Optimal parameter runs
Table 4 shows one of the results of the water quality analyses.
High removal rate was obtained for SS, BOD, total phosphorus and
coliform bacteria. Chromaticity and COD also removed for about 40%.
These results prove that the seeding method has high performance of
tertiary- treatment.
The amount of sludge generated was determined about 30 to 50
g-dry weight per 1 m3 of influent. This production rate is
one-fifth of that for combined sewer overflow treatment. As shown in
Table 5, the sludge was easy for thickening and dewatering. The
caloric value of the dewatered sludge was almost half of that from
the combined sewer overflows treatment. Incinerated ash contained
ferromagnetic material for about 1%, hence it seems possible to reuse
the ash as seeding material after ore dressing by magnetic "separation.
The matrix was washed almost compfetely by the backf lushing
described above. In contrast to the combined sewer overflows
treatment, fibrous material did not accumurate in the matrix.
85
-------
3
o
100
• 90
80
70
I
60
50
0 10' 20
Alum concentration
(mg-Al/1) .
Fig. 15. Effect of alum concentration on removal rate of
total phosphorus.
Test Conditions
Magnetic Field Strength : 5KG
Flow Velocity : 1000 m/day
Magnetite Cone. : 200 mg/1
Polyelectrolyte(C-599-lP) : lrag/1
Influent (activated sludge process effluent)
Total P : 4.6 mg/1
86
-------
o
CO
I
0)
PS
100
80
60
40
20
Polyelectr olyte:0. 6rag/l
0 10 20 30 40 50
a.
"ca
•p
5
0
100
80
"60
40
f> 20
0
S o
PS °
Polyeleotrolyte:0. 6mg/l
0.3n»/t
0 10 20 30 40 50
Magnetite concentration (rag/1)
Fig. 16. Effect of magnetite and polyelectrolyte
concentrations on removal rates of -TOG and total
phosphorus.
(The case of poor water quality influent.)
Test Conditions
Magnetic Field Strength : 5KG
Flow Velocity : 1000 m/day
Alum Cone. : 8mg/l
Polyelectrolyte : C-599-1P
Influent C Activated sludge process effluent)
TOG : 28mg/l
Total P : 4.7 mg-P/1
87
-------
8
H
I
60
40
20
10
o Polyeleetrolyte:0.1mg/l
• 0.3mg/i
* Q.Gng/l
20
30
03
•P
O
I
I
100
80
60
40
20
0 10 20 30
Magnetite concentration (rag/1)
Fig. 17. Effect of magnetite and polyelectrolyte *
concentrations on removal rates of TOG and total
phosphorus.
(The case of the influent of high grade water quality)
Test Conditions
Magnetic 'Field Strength : 5KG.
Flow Velocity : 1000 m/day
Alum Cone. : 8mg/l
Polyelectrolyte : C-599-1P
Influent (Activated sludge process effluent)
TOC : 13mg/l
Total P : 0.57 mg-P/1
88
-------
80
60
40
20
Transparency
150
100
50
0
10
0 0.2
[ t T
0 0.2
0.4 0.6 0.8
t 1
0.4 0.6 0.8 •
^ TOG
1.0
f
1.0
60 4
2
0
0.15
1 ,-H.
^ 0.10
1? 0.05
0.2
0.4 0.6
0.8
1.0
0 0.2 0.4. OS 0.8 1.0
Polyelectrolyte Conc.(mg/l)
F i g. 18. Effects of polyelectrolyte concentration on filter
run of high gradient magnetic separator and effluent
water quality.
Test Conditions
Magnetic Field Strength : 5KG
Flow Velocity : 1400m/day
Magnetite Cone. : 10mg/l
Alum Cone. : 4mg/l
Polyelectrolyte : P-720
Influent (Activated Sludge Process Effluent)
Transparency : 65, Turbidity : 8.7, SS : 4.3mg/l
TOG : 13rag/l, Total P : Q.54mg-P/l
89
-------
s
s
20
1000
2000
3000
Transparency
150
100
50
0
1
t- r T
DOO 2000
T t
3000
txQ 4
S
2
0
fri ""* .
C^ 6.1
s*
P I1 o
TOG
Turbidity
1000
2000
3000
1000 2000 . 3000 .
Flow Velocity (ra/day)
Fig. 19. Effects of flow velocity on filter run of high
gradient magnetic separator and effluent, water
qual i ty.
Test Conditions
Magnetic Field Strength : 5KG
Magnetite Cone. : 10mg/l
Alum Cone. : 4mg/l
Polyelectrolyte Conc.(P-720) : lmg/1
Influent (Activated Sludge Process Effluent)
Transparency : 36, Turbidity : 11.2, SS : 9.8mg/l
TOC : 18.4mg/l, Total P : 1.3mg-P/l
90
-------
9
+•> e
i— i ~~'
•H
40
20
Transparency
u
150
100
50
0
10
R
0 1
t-
0 1
2 3 ,4
t
2 3.4
TOC
.
5
1
5
__— -•
G
O. "^
-V.
la ^
O £
4
2
Turbidity
..ss-
0 1.2 3 4 5
0.04
0.02
n
•
• — •
012345
Magnetic Field Strength
, . (KG) ''
F i g. 2 0. Effects of magnetic field strength on filter run of
high gradient magnetic separator and effluent water
quality.
Test Conditions
Flow Velocity : 2200m/day
Magnetite Cone, : lOmg/1
Alum Cone. : 4mg/l
- Polyelectrolyte Conc.(P-720) : lmg/1
Influent (Activated Sludge Process Effluent)
Transparency : 42, Turbidity : 9, SS : 9mg/l
TOG : 13.5mg/l, Total P : 0.67rag-P/l
91
-------
Air tank
£5"
Magnetite
(M)
Alum
Influent _/g>
7
Flocculation
tank
-KZS —
Static mixer
Effluent
Sludge
\_y
Polyelectrolyte
Fig. 21. Flow sheet of the pilot plant for tertiary
treatment.
(HGMS : High Gradient Magnetic Separator)
92
-------
CO
p
J-l
-p
ca
5
CO
93
-------
Table 4. Results of continuous^ operation of pilot plant for
tertiary treatment.
PH
Transparency
Chromatici ty
SS (mg/1)
CODMn (mg/D
BODS (mg/1)
Coliform bacteria
(No. /ml)
Total phosphorus
(mg-P/ml )
In.f luent
7.8
34
29
- 12
19
7.5
340
1.7
Effluent Removal rate
7.4
more than 150
17
3.6
11
2.3
14'
0.33
-
—
41%
70%
42%
'69%
96%
81%
(Operating conditions)
Magnetic Field Strength : 3KG
Flow Velocity : 2000 m/day
Magnetite Cone. : 10 mg/1
Alum Cone. : 4 mg-Al/1
Polyelectrolyte Cone. : 1.5 mg/1
94 .
-------
Table 5.
Characteristics of sludge generated from tertiary
treatment process. " '
Raw sludge
PH
SS
VSS
SVI
6.9
440 mg/1
200 mg'/l
102
Thickened sludge
SS
9800 mg/1
(gravity thickening for 30 minutes)
Dewatered sludge
Water content
Calorific value
85%
2380 cal/g-dry weight
(Buchner funnel test with 760 mmHg-for 15
minutes, no coagulant addition)
95
-------
4. PRELIMINARY DESIGN OF DEMONSTRATION SIZE AND FULL SCALE PLANTS
Based on the results of the present study, preliminary design
was carried out for a demonstration plant and a full scale plant
whjLch are used for combined sewer overflows treatment in wet
weather and tertiary treatment of activated sludge process effluent
in dry weather.
Fig. 23 shows the flowsheet of the demonstration plant with
capacity of 10000 m 3 /day. The plant consists the apparatus for
dosing and flocculation, two preseparators of 2.75 m in diameter, and
high gradient magnetic separator of 2.5 m in diameter. Preseparators
are used only for treatment of wet weather flow. In dry weather,
flocculated slurry is fed directly to the magnetic separator through
by-path. Fig. 24 shows a layout of the demonstration plant. Land
required of this plant is 165 m2. Fig.25 shows a layout plan of the
full scale plant with treatment capacity of 100000 m 3 /day. The
required area of this plant is estimated about 860 ma. The
preseparators occupied large area in the plants, hence omit of the
preseparator is desirable from the view point for reduction of land
required. Thickness of the matrix , however, must be increased
considerably to omitt the preseparator, which increases electrical
power consumption of the magnetic separator. Furhter careful
consideration should be given to determine wheather adoption of sthe
preseparator is profitable or not.
5. SEEKING FOR LOW PRICE SEEDING MATERIAL
Seeking for low price seeding material will be important from
the view point of running cost reduction. In the present
investigation, magnetite recovered from coal ash was tested as one
of the promising low price seeding material. In power plants in
Japan, coal was adopted as energy source in addition to oil and
nucleous. Ash derived from some kinds of coal used in those power
plants contains magnetite particles. Recovering technique of these
magnetite particles by magnetic separator has §t>een almost
established. Preliminary pilot plant test showed that the coal ash
magnetite was suitable for seeding material. In addition to the coal
ash magnetite, industrial wastes from magnetite production plant and
iron mill are seemed to be hopeful for utilizing as seeding material.
96
-------
•*« c
£1 O
E E
O -P
O (0
> o
+J «H H-5
C CO ?!
- «
= !
1
II
0) O
c in
CO
•s
97
-------
Preseparator
22,000
HGMS
Fig-24. Arrangement plan of demonstration scale plant,.
(10000 m3 /day)
98
-------
DDDDDDQD 4.
DD D 0 D D D
' 4-> ,
0)
r^
CO
0
w
c
ft
|§
B *-l
r
s
to
cq
to
.E
99
-------
6. CONCLUSIONS
The following conclusions were obtained from the bench test and
pilot plant test of seeding method on the treatments of combined
sewer overflows and secondaly effluent.
(1) Seeding method was effective for both combined sewer overflows
treatment and tertially treatment.
(2) Optimum operating conditions of high gradient magnetic separator
were 2000 m/day for flow velocity and 3 KG for magnetic field
strength.
(3) Optimum concentrations of chemicals for combined sewer overflows
treatment were 25 mg/1 for magnetite, 3 to 7 mg-Al/1 for alum and 1
to 2 mg/1 for anionic polyelectrolyte.
(4) Optimum concentrations of chemicals for tertially treatment were
5 mg/1 for magnetite, 2 to 4 mg-Al/1 for alum and 0.5 to 1 mg/1 for
anionic polyelectrolyte.
(5) Sludge produced from the seeding process was easy for thickening
and dewatering.
(6) Magnetite which was recovered from coal ash was possible to
utilize- as seeding material.
7. ACKNOWLEDGEMENTS
The authors wish to thanks Mr. Masao Kobayashi, Mr. Suguru
Katura and Mr. Hideaki Hakamada for their helpful assistance in the
bench test. The authors would also like to acknowledge the
co-operation of Hitachi Plant Engineering & Construction and Daido
Steel Co. Ltd. for the pilot plant test. ,
REFERENCES
Allen, D. M., Sargent, R.L. & Oberteuffer, J. A. (1977) Treatment
of combined sewer overflows by high gradient magnetic
separation. EPA-600/2-77-015 (PB-264935).
Allen, D. M. (1978) Treatment of combined sewer overflows by high
gradient magnetic ' separation. EPA-600/2-78-209.
Kurinobu, S. & Uchiyama, S. (1982) Recovery of plankton from red
tide by HGMS. IEEE Trans. Magn., MAG-18; 1526-1528.
Oberteuffer, J. A. (1973) High gradient magnetic separation. IEEE
Trans., Magn., MAG-9; 303-306.
Kaneko, M., Kunikane, S., Maehara, R., Ishibashi, T., & Kimura, M.
(1981) Journal of Water and Waste, Vol.23; 1053-1062. (In
Japanese)
Kaneko, M. (1982) Zousui Gijutu, Vol.8, No.3; 49-53. (In Japanese)
100
-------
NEW MOVEMENT IN SEWERAGE CONSTRUCTION PROJECTS IN JAPAN
by
Ken Murakami
Regional Sewerage Devision
Dept. of Sewerage and Sewage Purification
Ministry of Construction
. Kasumigaseki, Tokyo 100, Japan
The work described in this paper was not funded by
the U.S. Environmental Protection Agency. The contents
do not necessarily reflect the views of the Agency and
no official endorsement should be inferred.
Prepared for Presentation at:
12th United States/Japan Conference
on
Sewage Treatment Technology
October 1989
Cincinnati, Ohio
101
-------
ABSTRACT
Huge investment in construction of public sewerage systems has been
made in Japan based on the five year programs. Overall percentage of the
sewered population still remains about 40%, but that in major cities has
reached almost 90%. Under this situation, the scope of sewage works in
Japan has been expanding to put more emphasis on such problems as more
safety from inundation in urbanized aeras, combined sewer overflow problem,
advanced wastewater treatment, wastewater reclamation and reuse, beneficial
utilization of sewage sludge, multi-purpose utilization of sewers and
treatment plant sites and so, forth. This paper summarizes these movements
in Japan.
102
-------
INTRODUCTION
. Human wastewater in Japan is basically treated by three different
systems. These are public sewerage, septink tanks and collective human
excreta treatment plants which treat human excreta collected by tank
trucks. Public sewerage systems currently covers about 40% of the total
population. Human wastewater from approximately 30% of the population is
treated by septic tank systems. The remaining 30% is treated at collective
excreta treatment plants. Most of the existing septic tank systems, as well
as collective excreta treatment plants, treat only human excreta. Grey
water from households served by these systems is discharged without any
treatment. ,
Therefore, necessity of public sewerage systems is still very high.
The sixth five year program for construction of sewerage systems plans to
invest 12.2 trillion yen "in construction of public sewerage systems during
the period of five years from 1986 to 1990. In FY1989, the investment is to
be approximately 2.3 trillion yen. About one third of the total investment,
800 billion yen is appropriated as a government subsidy.
In big ten Cities in Japan, percentage of sewered population has
reached almost 90% on the average, although there still remain problems to
be solved. One of the problems is storm water discharge including the
combined sewer overflow problem. In many places, the capacity for storm
water discharge is not sufficient, and safety from, inundation needs to be
improved. Various measures to cope with this problem are beginning to be
undertaken, in big cities where sewage works have rather long history,
combined sewer systems are generally adopted. The combined sewer overflow
is another problem which needs to be improved.
As sewerage systems with a secondary treatment level are installed, the
next target of providing advanced wastewater treatment has become a
practical one in many areas, particularly in the highly loaded areas or in
the river basins where wastewater is finally discharged into lakes or other
stagnant bodies of water. People's wishes to have better water environment
and more water space amenity have been stronger than ever.
It has also, been required to seek for additional functions in sewerage
systems, such as wastewater reclamation and reuse, beneficial utilization of
sewage sludge and multi-purpose uses of sewers and treatment plant sites.
Interesting progresses have been made in these fields, and the movement
toward this direction will be promoted further in the future.
Wastewater facilities in small communities are generally very poor. A »
number of small scale facilities will have to be constructed to improve the
situation. Therefore, one of the current major concerns is to develop
suitable technology for constructing economical and appropriate facilities
in small communities.
103
-------
1. STORMWATER DISCHARGE
1.1 More Safety from Inundation
Measures to prevent stormwater inundation in cities are generally based
on improving sewer pipes and pumping stations to discharge stormwater to
nearby rivers and the sea as quickly as possible. However, when such
methods cannot be adopted because of the poor capacity of the river
receiving the discharge and so forth, the following methods have to be
applied.
(1) Bypath sewer
A bypath sewer line functions to divert stormwater flow to a river with
a greater capacity, or directly to the sea, to cut the storm discharge in
the upperstream region. This occurs when urbanization in the upperstream
region has led to the occurence of inundation, but improvement in the sewer
pipes and in the lower reaches of the river is difficult. One example is
the Hirao-Takamiya region in Fukuoka City, where rapid urbanization caused
frequent inundation in the lowlands, and it was difficult to improve the
storm trunk sewers and rivers in the downstream region. However, a portion
equivalent to two-thirds of the basin area, or 212 ha (34 m3/sec), which
equals the deficient capacity, will be cut. The stormwater is to be
discharged directly to a river through the Hirao-Takamiya Trunk Sewer (3 to
4 m diameter, 2.5 km length) constructed along, an abandoned railway line.
The sewer is currently under construction. •
(2) Stormwater retention tank
A stormwater retention tank functions to cut the peak discharge by
holding the stormwater temporarily, and it may be constructed either on the
ground or under the ground. Stormwater retention reservoirs have been
constructed in many housing complexes in order to control the increase in
storm runoff. Similar facilities have been built in sewer systems, too.
The Jiyugaoka retention reservoir built in a residential area east of Nagoya
City is a good example. The reservoir has a capacity of 10,000 m3, and
its top is used as a playing ground for a primary school. In urban areas
where land use is very intensive, public spaces (such as roads and parks)
are used to construct underground retention tanks. Good examples are the
Imazu retention tank (26,000 m3 capacity) in Osaka City, which was built
under a baseball ground, and the Takatsuji retention tank (30,000 m3
capacity) in Nagoya City, which was built under a tall residential building
and a park. In addition to such large facilities, many small retention
tanks have been constructed to cope with localized inundation disasters in
urban areas.
104
-------
In an area of Nagoya City, the citizens filed a suit against the city,
claiming that the city did not respond properly. The court ^case lasted for
9 years. When a retention tank of 2,000 m capacity was constructed under "
this street and inundation was successfully prevented, the citizens withdrew
the suit.
A stormwater retention tank is an extremely effective means of
preventing inundation. However, there remains the question of whether space
for a stormwater retention tank can be ensured in congested urban areas. In
the future, it will be necessary to examine the more effective utilization
of public spaces, such as parks and schools.
(3) Large-scale storm trunk sewer
Where inundation is highly likely to occur in catchment basins, and the
capacity of the receiving rivers is critically short, large-scale storm
trunk sewers can be built to both bypass and hold stormwater. These* may
therefore be called "second rivers".
In Osaka City, such a facility, called the "Tennoji-Benten Trunk Sewer"
(80,000 in3 capacity), is already in service. In addition, the
"Hirano-Suminoe Trunk Sewer" is under construction. Tokyo Prefecture is
planning to construct a large-scale storm trunk sewer under the Loop 7
highway, and is expected to prevent inundation in the Kanda River and
Shakujii River basins.
(4) Infiltration of stormwater
Along with the above measures to reserve and divert .stormwater, various
attempts to reduce stormwater discharge by the infiltration of stormwater
have been made in many municipalities. This method maximizes infiltration
into the ground by providing infiltration in yards, porous pavements,
infiltration curbs and infiltration inlets. Furthermore, the runoff of
stormwater flowing into pipes is delayed through pipe storage and pipe route
bypassing.
1.2 Combined Sewer Overflow Problem
Combined sewer overflow should be basically controled by ensuring an
appropriate intercept volume to a treatment plant. When this is difficult,
measures based mainly on storage have been adopted. In addition, since
storage and infiltration as inundation prevention measures are also
effective in reducing the overflow volume, qualitative as well as
quantitative measures are under examination.
105
-------
(1) Sedimentation/storage of stormwater
H
The first facility for stormwater sedimentation/storage to reduce the
overflow load was constructed at the Nakanoshima Pumping Station in Osaka
City in 1974. Since then, many such facilities have been constructed. Most
of them have the capacity to store a 5 mm rainfall, on average, within a
drain basin.
(2) Treatment of wet weather sewage
The water volume and water quality of wet weather sewage fluctuates
enormously and requires intermittent treatment. Consequently, the
application of a biological treatment process is difficult. Thus, currently
planned methods are based on physical-chemical treatment processes.
One of them is a swirl concentrator, which is characterized by having
no moving parts and by ease of maintenance. Tokyo Prefecture and Matsuyama
City are currently constructing swirl concentrators.
As for wet weather sewage treatment, plate settlement using chemical
additives, and a super high-rate filtration, are under technological
development.
2. ADVANCED WASTEWATER TREATMENT
2.1 Necessity of Advanced Wastewater Treatment
The sewage system plays an important role in preventing water pollution
in public waters, and in preserving precious water resources. Depending on
the conditions of the basin secondary,treatment is not sufficient to protect
water environment, making advanced treatment necessary. Currently, advance
wastewater treatment is undertaken mainly in the basins of the rivers which
are used as important sources of water supply, and lakes, in order to meet
the water quality standards as quickly as possible. Typical lakes in Japan
(including Lake Biwa and Lake Kasumigaura), have become considerably
eutrophicated, so that nutrient removal is required.
2.2 Present state of Advanced Wastewater Treatment
The advanced treatment is under way at 22 sewage treatment plants
throughout Japan. Each of them aims to meet or keep meeting the water
quality standards. In some plants discharging their effluent into Lakes,
biological denitrification and chemical addition into aeration tanks are
practiced to remove nitrogen and phosphorus. As for advanced treatment
processes in plants discharging their effluent into rivers, rapid filtration
and chemical clarification are used to get higher removal of biochemical
oxygen demand (BOD) and suspended solids (SS).
106
-------
2.3 Future Direction
Advanced wastewater treatment has been practiced in areas where such
treatment is in great demand. People's wishes, on the other hand, to have a
better water environment and more water space amenity have become stronger
and stronger. It is considered necessary, therefore, to take a policy to
adopt advanced wastewater treatment more actively. There will be an
increasing demand for advanced wastewater treatment, particularly at the
following areas: *
(1) Improvement of water environment in big cities and their
surrounding areas
The sewage service has made great progress in big cities and their
surrounding areas. However, due to too much concentration of pollutant
loading, some rivers in the urbanized areas cannot be fully restored in
water quality by secondary treatment. This may necessiate the employment of
advanced treatment in the catchment areas of these rivers. In addition, in
some big cities where water shortage is always a potential problem, the
treated sewage has become precious water resources. The necessity of
advanced treatment will further increase in relation to reuse of treated
water. For example, many urban streams have experienced a reduced flow with
the progress of urbanization. The. restoration of stream flow to these
rivers is greatly desired. For this purpose, the utilization of sewage
water treated by the advanced treatment process is being planned in many
municipalities.
(2) Protection of water resources for drinking water supply
In planning a sewerage project and examining the effect of effluent
discharge on the water uses in the downstream reach, special attention has
been paid to water intakes for drinking water supply. When the receiving
river is used as a source for drinking water supply at the downstream,
advanced treatment of sewage has been practised depending on the necessity.
Recently, people are demanding more safety in drinking water, and
calling for better "taste" of drinking water. In this regard, necessity of
advanced treatment is becoming higher.than before.
(3) Prevention of eutrophication of lakes and reservoirs
Preventing eutrophication of lakes and reservoirs has become a
significant problem these days. Effluent standards regarding phosphorus are
established at 1066 lakes, and those regarding nitrogen at 78 lakes. Most
of these standards do not necessarily require higher level of treatment than
secondary treatment for domestic sewage. However, to restore a beautiful
water environment and maintain'good water quality for the lakes and
reservoirs advanced treatment of sewage is the most effective measure to be
taken, and it will be more significant in the future.
107
-------
(4) Control of eutrophication of bays and the inland sea
Many big cities in Japan including Tokyo, Osaka and Nagoya are situated
on bays and the Seto inland sea where exchange of water is little.
Eutrophication of these seas has already intensified. To restore the water
quality, nutrient loading should be minimized. At present, no quality
standards concerning nutrients in the seas have yet been set forth. And,
therefore, actually no advanced wastewater treatment is practised to remove
nutrients. >
However, more and more people are considering it quite necessary to
restore water quality in the seas near big cities. In these areas, too, the
necessity of advanced treatment is advocated.
3. MULTI-PURPOSE UTILIZATION
3.1 Reclamation and Reuse of Sewage
(1) Reclamed sewage as precious water resources
The annual demand for domestic water in Japan has been increasing by an
average of 3%. In order to meet the increasing water demand, continual
effort to develop new water resources should be made. At the same time,
quality of water resources must be ensured.
The total volume of the effluent from about 700 sewage treatment plants
in Japan reached approximately 8 billion cubic meters in FY1987. This
volume corresponds to approximately one-fourth of the municipal water supply
(the total of water for domestic and industrial water). Most of the
effluent is discharged to the rivers and seas, although some is reused for
various purposes.
With the expansion of industrial activities and the concentration of
population into big citites, the municipal water demand is still increasing
these days. Consequently, a system to reuse reclaimed sewage is being
introduced as a means to cope with an intensive water demand in urban areas,
and to achieve a stable water supply. Sewage can be a stable and abundant
water resource obtained in a city, and its volume is expected to increase
with the expansion of the sewerage systems.
108
-------
(2) Model project for sewage reclamation and reuse
The Ministry of Construction has launched a model project for sewage"
reclamation and reuse since FY1979 in order to facilitate reuse of treated
sewage. This project aims to supply reclaimed sewage to corporate buildings
in a certain district as toilet flush water after implementing a necessary
treatment of the reclaimed sewage. Currently, the projects are under way in
eight districts throughout Japan. These include the Tenjin district of
Fukuoka city, where the citizens suffered severe water shortage for over 287
days in 1978, and the subcenter of Shinjuku in the metropolis of Tokyo,
where skyscrapers are abundant. Among them, five projects are already'under
operation. In FY1989, the Tokyo Metropolitan government is planning to
start the project in a coastal subcenter.
(3) Flow augmentation, and ornamental streams and lakes
Many of the urban streams in large cites in Japan lose their flow in
dry weather due to an increase in the impermeable area within their basins
as the result of urbanization. On the other hand, people in cities have
come :to put more and more emphasis on a better environment,, particularly an
environment with clean water. Reclaimed water is a stable water resource
within cities, and thus, there have been many projects to utilize reclaimed
water for restoring live streams and creating clean water bodies within
cities.
Restoring live streams utilizing reclamed water was practised at 6
locations in 1986. Among these 6 locations, reclaimed water was discharged
year-round at 4 locations. The outline of these 4 projects is shown in
Table 1. :
Table 1. .Typical examples of restration of live streams
using reclaimed water
Name of city Name of STP
Reclaimed
Water flow
(m'/day)
Name of
stream
Reclamation
process
Tachikawa Nishikimachi
212
Nekawa
Sand filtration
Metro. Tokyo Tamagawa-Jouryu 20,720
Nobidome and Sand filtration
Tamagawa
Osaka
Nakahama
3,000 Outer Moat of Sand filtration
Osaka Castle
Osaka
Hirano
7/074
•Imagawa and
Komagawa
Sand filtration
109
-------
In addition, a model project named "restoration of live streams in
stormwater drains" was initiated in 1985. Most of the open storm sewers
used to be natural storm drains. This project is to restore live streams in
these storm sewers utilizing reclaimed water. Currently 8 projects of this
type are being carried out as shown in Table 2.
Table 2. Effluent reuse model projects "Restoration of
live streams in storm water drains"
Year of
project
start
1985
1985
1986
1986
1987
1987
1988
1988
Name of
prefecture
Kyoto
Oita
kanagawa
Osaka '
Kanagawa
Aichi
Aichi
Hyogo
Name of city
Nagaokakyo
Oita
Yokohama
Toyonaka
Yokahama
Nagoya
Nagoya
Nishinomiya
Name of storm
drains
Shoryu j i
Nakajima No. 1
Saedo
Houno-Nanbu
Terao
Arakogawa
Shirako-Sakae
Edakawa No. 2
Design flow
(m3/S)
0.06
More than 0.1
0.12
0-1
0.12
0.23
0.002
Furthermore, reclaimed water from sewage treatment plants is used at
many locations for ornamental lakes and streams within treatment plants and
in parks nearby.
(4) Snow disposal
In order to remove and dispose of snow in urban areas including private
lots, it is necessary to provide suitable disposal sites. Open channels,
including open storm sewers,.are often used for this purpose. A problem
usually encountered is the lack of sufficient flow to flush away the.
disposed snow.
A '
The effluent from sewage treatment plants has good characteristics for
this use, such as being stable in flow, being available within a city, and
having a relatively high temperature in. winter.
A model project named "The rapid removal of snow in urban areas using
reclaimed water" was initiated in 1985. Currently nine projects are being
carried out in cities in heavy snow areas, as shown in Table 3.
110 .
-------
Table 3. Model effluent reuse projects "Rapid removal of
snow in urban areas .using reclaimed water"
Year of
project
start
1985
1985
1986
1986
1987
1987
1987
1987
1988
Name of
prefecture
Hokkaido
Aomori
Akita
Niigata
Hokkaido
Niigata
Ishikawa
Fukui
Ishikawa
Name of
city
Ebetsu
Aomori
Akita
Tpkaichi
Sapporo
Yuzawa
Kaga
Tsuruga
kanazawa
Name of drains Design reclaimed
storm utilized water flow (m3/S)
Ebetsu No. 32
Namiuchi
Kusouzugawa-
Sagan No. 1
Inaricho and
Honmachi
Yasuharu
Nunoba No. 1
Jougan j i
Maizaki-Shimizu
Ekiura
0.15
0.12
0.12
0.04
0.5
0.238
0.03
Another project is being planned in Sapporo City to utilize a
stormwater reservoir in winter time as a disposal site for the removed
snow. The disposed snow is melted away by introducing effluent from a
sewage treatment plant into the reservoir.
Snow removal in snowy area is equivalent to stormwater drainage in
other areas. More contributions to sewage works from this field are
expected.
Ill
-------
3.2 Beneficial Utilization of Sewage Sludge
(1) Necessity of effective utilization of sewage sludge
Wastewater treatment is accompanied by the generation of sewage
sludge. As the sewered population increases, the volume of sewage sludge
generated increases. Advanced treatment may increase sludge production^in
the future. In 1987, when the percentage of sewered population was 39%, the
sludge production in dry solids (DS) was approximately 1.25 million tons,
and the annual volume of sludge disposed of was 2.1 million m3. When the
percentage of sewered population reaches 50%, it is estimated, that the
sludge production will be approximately 1.7 million tons in DS, and that the
annual volume to be disposed of will be 3.7 million m3.
On the other hand, the sites for landfill are limited. As urbanization
progresses, the acquisition of landi.fill sites is getting more difficult.
And the associated costs are increasing. Therefore, further efforts must be
taken to reduce the volume"of sludge, and to facilitate beneficial .
utilization of sludge as resources.
Table 4. Status of sewage sludge disposal. (1987)
(Units: 1000 m )
Disposal
method
Treatment
Landfill
coastal
Beneficial
reclamation utilization
Others Total (%)
Dewatered 762
cake
Incinerated 156
ash
Dried sludge 17
Digested and 7
thickened
sludge
306
105
4
0
274
43
126
114
19
191.
1,456 (68)
310 (15)
166 ( 8)
198 ( 9)
Total (%)
942
(44)
415
(19)
443
(21)
330
(16)
2,130 ( 9)
(100)
112
-------
Table 5. Status the beneficial utilization of sewage sludge (1987).
Treatment Dewa- 'incin-
tered erated
Classification cake ash
Green Committed by 117
farm- the municipality
land
Assigned to a 157
, fertilizer company
0
(Units: 1000 m3)
Com-
SlUd96 SSgf Slud9e
15
80
24
0
0
ouuLutaj. 274 7 22 104 0
Use as a
material
Total
^ — . ^_^_
construction
0 36 0 0 0
274 43 . 22 104 0
Total
204
203
407
36
443
(2) Current status, of beneficial utilization
spacefor . of ^Plications, such as for farmlands and green
spaces, and for construction materials. Recovery of energy is another
example of beneficial utilization. 9Y another
T5eqam°"nt °f Slud9e utilized for farmlands and green spaces in the
L>^TT^i8lUdOB CakS' C°mP°sted Slud9e- and dried sLdge, Somes to
11.5% of the total amount of sludge generated or 140,000 tons (DS
year The amount of sludge (mainly as incinerated ash) utilized as
construction materials is estimated to be 6.5% or 85,000 tons.
(i) Utilization for farmlands and green spaces
An-7 nnlf f™^ of sewage sludge utilized for farmlands and green spaces was
407,000 m /year, in 1987. The amount has been fairly constant in t2 last
flLllnZ' Ef °f ^ SlUd9S Utll±Zed f°r this P^P°se is applied to
farmlands Only a sma.ll amount of sewage sludge is used for green spaces
Sewage sludge is mostly used in the state of dewatered sludge cake
However more and more composted sludge is being utilized recently* Sewaae
6ffiCientl* ««"** 'or farmlands and green- spaces in Europe
113
-------
Currently, 21 municipalities are running sewage sludge composting
plants. The product is sold in bulk, or 20 kg package. The retail price
varies greatly, depending on the districts. The heavy metal standards for
the sludge utilized for farmlands are that the maximum concentrations of
mercury arsenic, and cadmium shall be 2,. 50, and 5 mg/kg, respectively. In
addition, a tentative guideline has been worked out, specifying .that sludge
application is allowed in the areas where the zinc content of the soil is
less than 120 mg per kg dried soil. These values are rather strict compared
with those stipulated in European nations. In particular, the guideline for
zinc in soil is very strict in terms of the zinc concentration in sludge or
soil. This is an obstacle, in a sense, to distributing composted sludge in
Japan. ,
(ii) Utilization for construction materials
' "-v
There are two utilization methods: One is the direct utilization of
incinerated ash or melted slag, and the other is to use incinerated ash or
melted slag as a part of the raw materials to produce building materials.
Examples of the former include roadbed material for pavement, back-filling
material for retaining walls, and intermediate cover for sanitary
landfills. Recently, incinerated ash has been used to improve soil
characteristics of the excavated material from sewer construction sites. In
order to use the excavated material for refilling, it is often required to
increase the kinetic strength by mixing an additive. Incinerated ash has
good characteristics as an additive for this purpose. This is a newly
* development in the utilization of. incinerated ash.
In the latter utilization of incinerated ash or melted slag, these
materials can be used as raw materials for producing such products as -
ceramic pipes, light-weight aggregates, bricks, tiles, and permeable
blocks. Some of these raw materials are manufactured on a commercial
basis. The amount of sludge used for construction materials is limited at
present, but is expected to increase with the encouraged employment of the
melting process.
-------
3.3 Multi-purpose Utilization of Sewers and Treatment Plant Sites
(1) Utilization of treatment plant sites
an ov^rnfT^ Plant S±teS and P^P1^ stations are precious open spaces in
rf ?he Snf /in7' MVltlple utili^tion of the area permits effective v '
of the land and the realization of pleasant urban scenery, thus providing
n]Ln?n? ^rl^0d Wlth a PlaCS °f recreation and relaxation. This kind of
planning is becoming popular these days.
baseblr ^ Sporting ^cilities including tennis courts and
baseball grounds, as well as parks and green zones are being constructed in
th ' s
the covered top of the facilities are used as a refuge at the time
of disaster, in addition,, buildings such as music halls and mf rtial art
^S° ^ COnStructed above Plant, sites. This ne^ tre"nd of
is becoming noticeable.
(2) Effective utilization of sewers
carried out at
due
Under9rouna se»ers
sews win °Ver a
sewers will become more intensive in the future.
not been used efficiently
n^rstS.
- this utilization of
4. SMALL-SCALE SEWERAGE SYSTEMS
50
the
arS 2817 communities with ^ Population. of less than
percentage of sewered population is as little as 6% on
APProxim^^ly 2200 of these communities have not started
6 S Y6t- men Plannin9 Sewera9e system« ^ these
n sometimes encountered with selecting a plant site
because the community is located in a mountainous area, or on the sea or
"11 ° ^^ & C°nSenSUS °f ^^ In -ch a ^ase, a
Sewa9e treatment plant may be an amicable
• ° deveiop
Sections wil1 introduce prefabricated treatment plants
treatment Plants- tunnel-type treatment plants, and
exampl's of
115
-------
4.1 Prefabricated Treatment Plants
The Japan Sewage Work Agency has studied cost-effective methods of
constructing small-scale treatment plants since 1982. Based on the studies,
a prefabricated treatment plant has been developed and standardized.
For the prefabricated plant, an oxidation ditch process is adopted.
This is because it can withstand inflow load variations, is easy to
maintain, and generates less sludge.
The prefabricated treatment plant is a circular structure that a
conventional ditch is laid out as an outer circle, with a secondary
sedimentation tank at its center. As for the structure pre-stressed
concrete panels manufactured at the factory are assembled at the site. For
mechanical and electrical facilities, standardized equipments are used to
reduce the costs. To rationalize the design, standardized designs are
prepared for the plants with capacities from 300 to 1,200 m /day with an
interval of 100 mf/day. Thus, when a certain number of plants? have been
constructed, the costs of construction as well as design can be considerably
reduced. Furthermore, the construction period can be shortened, and the
construction work can be easily performed at the site. The first
prefabricated treatment plant was completed in Nakanojo, Gunma Prefecture
in 1988. There are currently nine plants in Japan which are under planning
or construction.
4.2 Floating-type Treatment Plants (Plants Floating on the Sea)
A number of small communities are located on the hilly sea shores,
where it is sometimes quite difficult to acquire suitable sites for
constructing sewage treatment plants. In these cases, it has been a common
practise to create the land by coastal reclamation. This is sometimes very
costly and requires the long period of construction.
The Japan Sewage Works Agency has developed a floating-type treatment
plant as an alternative technology, and the first, plant of this type was
constructed in Atami City, Shizuoka Prefecture in 1985.
Steel made, ark-structure plant constructed at a dockyard is taken in
tow to the sea site. The plant is installed at a specified position,
connected with the sewer, and then put into operation. The floating plant
can utilize either the conventional activated sludge process, oxidation
ditch process, or sequencing batch reactor process.
There are two installation methods: Grounding at the bottom of the
sea, and moorage at a pier with an anchor. Compared with the method of
constructing a plant after reclaiming the sea, the advantages of the
floating method are that the construction period can be shortened and the
construction costs could be reduced, particularly for smaller scale plants.
116
-------
4.3 Tunnel-type Treatment Plants
In areas where it is difficult to find a suitable plant site, one
alternative is a tunnel-type treatment plant. In Sweden and other
Scandinavian nations, the rock bed is tunneled so that a so-called "tunnel
treatment plant" can be constructed inside, in Japan, too, feasibility
Since a treatment plant is built in a tunnel, the plant has a long
' ^ ^ difference comPared to a plant built above the
pi™ PrOCeSS' n° Si^f icant -strictions are
rrcn treatmenj Plant is operated in a tunnel, no. considerations are
required regarding the foundation and covers. In heavy snow areas where
coverage of the facilities is required for maintenance or to prevent tSe
water temperature from dropping, the tunnel method will prove effective.
4.4 Alternative Collection Systems
The cost for sewers occupies 70% of the total sewage construction costs
of a sewerage system. Therefore, to. reduce the construction costs, sewer
construction costs must be minimized.
Conventionally, most Japanese sewer systems are of the gravity type
Now that sewerage systems are constructed in small communities, the gravity
sewers are not necessarily economical due to geographical conditions.
Consequently, the employment of pressure or vapuum sewers has been activelv
investigated. A pilot project is underway to acquire data concerning the
costs and design. Based on the results, the pressure or vacuum sewers may
largely be substituted for gravity sewers in the future.
117
-------
-------
Innovative Sludge Handling System Using Pelletization
for
Simultaneous Thickening and Conditioning
by
Kazuhiro TANAKA, Dr. Eng., Director
Shuzo KOIKE, Research Engineer
Research and Technology Development Division
Japan Sewage Works Agency
Naohiro TANIGUCHI, Director
General Planning Section of General Planning Division
City Planning Bureau
Tokyo Metropolitan Government
The work described in this paper was not funded by
the U.S. Environmental Protection Agency. The contents
do not necessarily reflect the views of the Agency and
no official endorsement should be inferred.
Prepared for Presentation at:
12th United States/Japan Conference,
on
Sewage Treatment Technology
October 1989
Cincinnati, Ohio~
119
-------
Innovative Sludge Handling System Using Pelletization
for
Simultaneous Thickening and Conditioning
Kazuhiro TANAKA, Dr. Eng., Director
Shuzo KOIKE, Research Engineer
Research and Technology Development Division
Japan Sewage Works Agency
Naohiro TANIGUCHI, Director
General Planning Section of General Planning Division
City Planning Bureau
Tokyo Metropolitan Government
ABSTRACT
Many sewage treatment plants in Japan have suffered from difficulties
in sludge handling processes, especially in sludge thickening/dewatering
processes, because of the changes in sludge characteristics in the past one
or two decades.
*
This paper introduces an innovative sludge handling system, developed
by the Japan Sewage Works Agency, to solve those difficulties.
The newly developed sludge handling system employs a metal coagulant
aid and an amphoteric polymer that has the characteristics of flocculants
that are both cationic and anionic. The .system consists of a sludge-charge
neutralization reactor and a sludge pelletizing thickener. The sludge from
wastewater treatment process is directly fed to the system. In the
sludge-charge neutralization reactor, the fed sludge is dosed with the metal
coagulant aid to neutralize anionic sludge charge* In the pelletizing
thickener, the amphoteric polymer is fed into the charge-neutralized sludge
to pelletize and thicken it. The small sludge particles form spherical
sludge masses of 1 to 2 cm in diameter, which have excellent mechanical
strength and low water content. •
The function of this system as a sludge thickener has eliminated the
necessity of the conventional sludge thickening process. An SS recovery
efficiency of 90%, or higher, is provided by this system. Compared with
conventional sludge handling systems, this system achieves a higher
filtration rate of 200 kg-DS/m/h and lower dewatered sludge cake moisture
content of 75%, by using a belt-press filter.
Phosphate is also physicochemically fixed to the sludge through use of
a metal coagulant aid. Hence, this system can be also characterized as a
sludge handling system that is suitable for biological phosphorus removal
processes. •
120
-------
Innovative Sludge Handling System Using Pelletization
for
Simultaneous Thickening and Conditioning
Kazuhiro TANAKA, Dr. Eng., Director
Shuzo KOIKE, Research Engineer
Research and Technology Development Division
Japan Sewage Works Agency
Naohiro TANIGUCHI, Director
General Planning Section of General Planning Division
City Planning Bureau
Tokyo Metropolitan Government
1. Introduction
Sludge handling systems have the following objectives:
1) to reduce the volume of sludge to economize land space for disposal
and costs of sludge disposal,
2) to improve the ease of handling during disposal, and
3) to change the characteristics of the sludge to a stable form.
In genreal, the sludge handling systems consist of thickening,
digesting, dewatering, and incinerating processes. In the past one or two
decades, the characteristics of sewage sludges have changed in Japan.
Especially, the VSS/SS ratio of sludge has increased significantly. The
causes of these changes,in the sludge characteristics are due to the
transition from combined sewer collection systems to separate sewer
collection systems, and the changes in our way of life.
*
As a result of the changes in sludge characteristics, many sewage
treatment plants have suffered from difficulties in sludge handling,
especially, in sludge thickening and dewatering processes.
This paper introduces an innovative sludge handling process, developed
by the Japan Sewage Works Agency, to solve those problems in sludge handling
systems.
2. Sludge Handling Difficulties in Japanese Sewage Treatment Plants
1) Problems Related to Gravitational Thickeners
The status quo of the sludge handling difficulties are well pointed out
in a research report prepared by the Tokyo metropolitan government.1'
121
-------
The chronological changes in the Volatile Suspended Solid to Suspended
Solids (VSS/SS) ratio of excess sludges are shown in Figure 1. Figure 1
reveals that the VSS/SS ratio of excess sludges in 1965 was around 60%, and
increased to around 80% in 1985. The chronological changes in the solid
content of thickened sludges are shown in Figure 2, which reveals that the
sludge was thickened to about 5% of the solid content in 1965, and decreased
to 3%, or less, of the solid content in 1985. The gravitational
thickenability of sewage sludge has decreased as the VSS/SS ratio increased.
The increase in the volume of thickened sludge caused by the decrease
in solid content means an increase in the volume of the sludge to be
supplied to the dewatering machines. This is one of the reasons for the
recent deterioration of the dewatering efficiency.
The chronological changes in-the dewatering rate and in the coagulants
(FeCl3 and CaCO3) dosing ratio (% to DS), for a vacuum-type dewatering
machine, which had been widely employed in Japan are shown in Figure 3. As
shown in the figure, the dewatering rate has been decreasing, and the dosing
ratio has been increasing.
The changes in the characteristics of sewage sludges such as those in
Tokyo, have been observed in almost all Japanese sewage treatment plants.
These changes include the gravitational thickenability of sewage sludge, the
dosing ratio of chemicals, the dewatering capacity of the dewatering
processes, and the moisture content of dewatered sludge cakes.
co
co
en
CO
CO
8
o
X
CD
o
•H
•p
CO
CO
en
en
90
80
70
60
50
40
I
I
I
1965
1970
1975
Year
1980
1985
Figure. 1 Chronological change in the VSS/SS ratio of
excess sludges. (Tokyo)
122
-------
6
o
O
C/3
7.
6
5
4
3
2
1
0
1965
Plant B
Plant C
_L
j_
_L
1970
1975
Year
1980
1985=
Figure 2. Chronological change in the solid content of
graviat.ionally thickened sludges. (Tokyo)
CO
Q
0)
•p
CO
i-l
0)
•p
§
0)
Q
,25
20
15
10
5
0 '
Coagulant
dosing
ratio
Dewatering rate
_L
50 *>
40
30
20
10
1965,
1970 . 1975
Year
1980
o
•H
4->
•2
(7)
•H
c
(0
Figure 3. Chronological changes in the dewatering rate
and coagulant dosing ratio (with vacuum-type
dewatering machine). (Tokyo)
123
-------
The increase of the moisture content of dewatered sludge cakes results
in an increase in fuel consumption in incineration systems. Thus, to '
improve the performance of the total sludge handling system, it is very
important to improve the performance of the sludge thickening and dewatering
processes. Therefore, municipalities having difficulties with the sludge
thickening processes have started to employ mechanical thickeners, such as
centrifugal thickeners, instead of the gravitational thickeners.
2) Problems Related to Biological Phosphorus Removal
t
It is known that phosphate is released from sludges when a
gravitational thickener is employed to handle the excess sludge in a
biological phosphorus removal process. This results in a poor total
phosphorus (T-P) removal efficiency.
An experimental study cm the biological phosphorus removal efficiency
was conducted in the S treatment plant. A schematic flow diagram of the
plant is shown in Figure 4. The results obtained can well explain the
relationship between the phosphate release in a sludge handling process and
the effluent T-P. The operational conditions of the plant during the
experimental study are as follows.
Back washings
Return flow
*
Raw influent
Primary
settling
tank
372 m3
Biological reactor
2160 or 4320 ra3
Final effluent
Septic tank
sludge
Deatered
cake
Rapid sand filter
Figure 4. Schematic flow diagram of the S treatment plant
Influent Flow Rate
Hydraulic Retention Time
of the Biological Reactor
(1/4 was operated as anaerobic)
MLSS
: 8,000 - 9,500 m3/day
: 5.5 to 6.5 hours
: 2,400 to 3,500 mg/L
124
-------
The chronological changes in the sidestream phosphorus load ratio*
and the effluent T-P concentration are shown in Figure 5.
The sidestream phosphorus load ratio:
Sidestream phosphorus load
Influent phosphorus load x 100% (1)
150
o
•H
4->
(0
U
m
u
o
•a
•a
100 -
- 1
a?
CM
EH
•P
G
t-i
t-i
W
Figure 5.
SidestrearaPhosphorus load ratio
£
t! rati° resulted in an overloaig of the T-P load
the biologxcal reactor, which caused the increase in thl effluent ?-P
etticiency of biological phosphorus removal, it is-necessarv to
*ich handles'sludge
125
-------
3. Targets in This Study ,
The Japan Sewage Works Agency decided to develop an innovative sludge
handling system to solve the sludge handling difficulties. The following
are its targets.
(1) Elimination of conventional sludge thickening systems
The difficulties in sludge handling systems come from conventional
sludge thickening processes. One of the targets was the elimination of
conventional sludge thickening processes to achieve a high efficiency of
sludge handling.
(2)' Reduction of the moisture content of the dewatered sludge cake with
belt-press filter dewatering machines
An 80% moisture content of a dewatered sludge cake is common when the
sewage sludge is dewatered with a belt-press filter. The target was to
reduce the sludge cake moisture content of a dewatered sludge .cake using a
belt-press filter, from 80% to less than 75%.
(3) Reduction of phosphate release from excess sludges
The reduction of phosphate release from the excess sludge to attain an
efficient biological phosphorus removal was another of the targets.
4. Theory of the Innovative Sludge Handling System
As sewage sludge particles are volatile-fraction rich, they form
hydrophilic'colloids because of the anionic charge. It is difficult to
remove water from sludge particles under such conditions. In addition, as
the particles are very small, there is high electric repulsion among the
particles, and they are poorly agglutinative. Furthermore, small particles
easily cause the filter cloth to be clogged.
To dewater sludges, usually, cationic polymer flocculants (cationic
polymer/polymers) are dosed to the sludge to neutralize the anionic charge
of the sludge, and thus the sludge particles are flocculated. However, as
the flocculated sludge floe is still small and its mechanical strength is
still -weak, the moisture content of the dewatered sludge cake will be
decreased up to 80% when dewatered with a belt-press filter. Also, poor
separation of dewatered sludge cakes from the filter cloth is often observed.
The newly developed innovative sludge handling system employs a metal
coagulant aid and a functional .polymer, which has characteristics of
flocculants that are both cationic and anionic (the amphoteric polymer). A
schematic flow diagram of the system is shown in Figure 6.
126
-------
Separated water -•
Sludge charge
neutralizing reactor
Fed sludge
Sludge pelletizing
thickener
To
the dewatering
machine
Charge-
neutralized
Q sludge
Metal coagulant aid
container
Amphoteric polymer
solution tank
Figure 6. Schematic flow diagram of the system.
First, raw sludges withdrawn from sedimentation tanks are fed to the
sludge-charge neutralizing reactor. Then, the metal coagulant aid is dosed
to the sludge.
Next, the charge-neutralized sludge is fed to the pelletizing thickener
with an amphoteric polymer. In the reactor, the sludge will be pelietized
to spheres of 1 to 2 cm in diameter (the sludge pellets).
A conceptual model of the pelletization process is shown in Figure 7.
127
-------
Cationic
conpound
®£
Nucleus of floe
Metamorphic
agglutinative
layer
Anionic
conpound
(inactive)
Sludge particles after
dosing polyaluminum chloride
J~
o
\
STEP I
STEP II
STEP III
Figure 7. Conceptual model of the palletization process.
128
-------
STEP 2 :
STEP 3 :
The pelletization is carried out by the following 3 steps.
STEP 1 : The amphoteric polymer is fed to the charge-neutralized sludge.
- The cationic compounds of the amphoteric polymer adhere to the
sludge particles. At the same time, inavtive anionic compounds of
the amphoteric polymer will be charged due to charge in pH.
The anionic compounds of the amphoteric polymer combine the
cationic compounds which adhere to the sludge particles. The
bridged sludge particles will start to form larger sludge floes.
The amphoteric polymers which bridge the sludge particles contract
as the reaction continues, and the sludge pellets grow and become
thicker and stronger. The growth of the sludge pellets will be
accelerated by contact with growing sludge pellets and sludge
particles.
An increase in the filtration rate, and a decrease in the moisture
content of the dewatered sludge cake are expected, as a consequence of
strong and thick sludge pellets being formed with this system. '
From these system characteristics, the system can be explained as a
sludge handling system which makes sludge particles into Ball-shaped
sludge pellets and a very Effective Sludge Thickening system. Using
these initial letters, the system has been named the BEST System.
-t
5. Development of the Pelletizing Thickener
The development of the BEST system was originally begun to develop a
sludge handling system suitable for biological phosphorus removal
processes. At that time, a sludge handling system which combined metal
coagulant aid dosing with polymer flocculant dosing was evaluated for the
physicochemical fixation of phosphorus. A schematic flow diagram of the
prototype system is shown in Figure 8.
129
-------
Primary
sedimentat ion
tank
Excess sludge
Aeration tank
V
Secondary
sedimentation
tank
Gravitational
thickener
Sludge
charge
neatral-
ization
reacter
Belt press filter
Palletizing
reacter
Metal coagulant, Amphoteric polymer
aid storage tank storage tank
Figure 8. Schematic flow diagram of prototype system.
As can be seen in Figure 8, the prototype system-consisted of a
sludge-charge neutralization reactor.and a pelletization reactor. The
thickened/not thickened primary sludge and the gravitationally thickened
excess sludge were fed to the system. The mixed sludge was
charge-neutralized by the metal coagulant aid dosing, and pelletized by the
amphoteric polymer dosing. The reason the amphoteric polymer was used is as
follows. The sludge fed to the pelletization reactor was conditioned with a
metal coagulant aid which is cationic, and the electrical charge condition
changes to cationic or anionic, depending on the dosage of the coagulant
aid. So, the amphoteric polymers were selected to be able to pelletize the
conditioned sludge, regardless of whether it was charged as cationic or
anionic. In the prototype system, the sludge thickening process was
indispensable to making the following facilities as compact as possible.
130
-------
Through the experiments, the development target to reduce phosphate
release waa achieved. In addition to this success, the observation of quite
fine separation of liquid and solids in the palletizing reactor suggested
another possibility to develop an innovative sludge thickening process.
This led the development of a sludge thickening/conditioning system, by
•adding a separated water withdrawer to the sludge pelletization reactor.
The structure of the separated water withdrawer was examined, using the
following type of screens, which is installed at the bottom..of the separated
water withdrawer.
*
1) Metal nets,
2) Punching plates, and
3) . Plates with concentric circular slits.
. These screens are shown in Figure 9. ,
Separated water
withdrawer
Metal net
Punching plate
Plate with
concentric
curcler slits
Figure ,9. Examined screens.
131
-------
The performance of each screen was examined in terms of the SS recovery
efficiencies, clogging occurrences, and available flow rates through the
screens. It was found that the plate with concentric circular slits was the
most suitable for the separated water withdrawer, as it showed the best
performance on the SS recovery efficiency, and no clogging troubles were
observed. The clearance of the slit was 1.5 to 2 mm wide.
After gathering all the information about the favorable performances
for the sludge pelletization thickener, the sludge pelletizing thickener
shown in Figure 10 was developed.
Separated
water
Sludge
pellets
TT~D
Amphoteric Charge-neutralized
polymer sludge
Figure 10. Structure of sludge pelletizing thickener.
132
-------
The sludge from'the sludge charge neutralization reactor is fed to the
reactor at the bottom of the reactor, with an amphoteric polymer. There are
two stirring paddles in the sludge pelletizing thickener. The fed sludge is
mixed by the lower paddle with the amphoteric polymer. The lower paddle
also establishes an internal recirculation of growing sludge pellets and fed
sludge particles contact each other. The upper paddle propels the growing
sludge pellets along the internal wall of the reactor. Then, the sludge
pellets are consolidated and the water contained in the sludge pellets is
removed from the pellets. Thus, the strength of the pellets increases.
This is a benefit not only for improving dewatering performance, but also
for improving the SS recovery efficiency of the system. As the sludge
pellets contain a small amount of water, sludge pellets with a sufficiently
high solid content will be supplied to dewatering machines.
The development of this sludge pelletizing thickener has made it
unnecessary to employ conventional thickening processes. This elimination
has made stable sludge dewatering possible, even when sludges were fed which
showed poor thickenability or poor dewaterability to conventional sludge
thickening, conditioning, and dewatering processes.
This system just requires tanks, stirring equipment, and pumps to feed'
sludges and coagulants. As all of them are very simple and compact, the
construction cost of this system-can be reduced compared with other
conventional sludge thickening systems.
6. Experimental Procedures
The following four kinds of sludge were examined for the development of
the BEST system.
(1) Mixed Raw Sludge of the Conventional Activated Sludge Process (the
Mixed Raw Sludge),
(2) Excess Sludge of the Conventional Activated Sludge Process,
(3) Excess Sludge of the Anaerobic/Oxic Process, and
(4) Excess Sludge of the Oxidation Ditch Process
(Sludges described in (2) to (4) will be referred to as the Excess
Sludges).
The characteristics of the above sludges are listed in Table 1. In
general, the Mixed Raw Sludge is less difficult to thicken and to dewater,
compared with other sludges. It is said that a moisture content of 80%
would be the lowest, when the Mixed Raw Sludge is dewatered by a belt-press
filter with the available maximum filtration rate of 130 kg-DS/m/h after
single polymer conditioning. For the Excess Sludges, it is said to be very
difficult to reduce the moisture content of dewatered sludge cake to less
than 85% with an available maximum filtration rate 'of 50 kg-DS/m/h, when
they are conditioned and dewatered in the same way as the Mixed Raw Sludge.
133
-------
Table 1. Characteristics of four kinds of sludges
Kind of sludge
Characteristics
Mixed raw sludge
of conventional
activated sludge
process
Excess sludge of
anaerobic/aerobic
process
Excess sludge of
convent ional
activated sludge
process
The fresher sludge has more fibers and shows better
dewatering efficiency. However, this kind of sludge
is generally corruptible. Once corruption
progresses, the thickneability and dewatering
efficiency deteriorate.
Compared with the excess sludge of a conventional
activated sludge process, this sludge shows a
slightly better dewatering efficiency and higher
phosphorus content. However, when this sludge is
put into the gravitational thickener, or mixed with
primarily sludge, phosphorus release occurs.
In general, this kind of sludge shows poor dewatering
efficiency. It is said to be very difficult to
treat this sludge, even if the belt-press filtration
is employed. The moisture content is as high as 85%.
Excess sludge of
the oxidation
ditch process
This sludge is considered to be poorly flocculated by
the single polymer conditioning.
The characteristics of the sludges used in the experiments- were
measured on the items listed in Table 2.
134
-------
Table 2. Measurement items and methods for the characteristics of sludges
Item
. Unit Notation
Method
PH
PH
JIS (Note 1)
Electrical
conductivity
us/cm Cond JIS
Methyl orange mg/£
alkalinity
MA Jis, Japanese standard method for the
examination of wastewater quarity.
TS
TS Jis, Japanese standard method for the
examination of wastewater quarity-.
VTS/TS
VTS/TS Japanese standard method for the
examination of wastewater. quarity.
SS
SS JIS, Japanese standard method for the
examination of wastewater quarity.
VSS/SS
VSS/SS Japanese standard method for the
examination of wastewater quarity.
Fiber/SS
% Fiber/SS
CST
sec
CST
Toriton's CST measuring instrument and
Wattman filter are employed.
Total phosphorus mg-p/2 T-P (Mixed reagent method) (Note 2)
Orthophosphoric
acid
mg-P/£ o-P (Mixed reagent method) (Note 2)
Note 1:
Note 2:
Japanese Industrial Standards
Methods for chemical analysis of water and and wastes (1971)
of the Environmental Protection Agency.
Lab tests and pilot scale experiments were conducted to examine the
basic performances of the BEST system with regard to pelletization and
dewatering. The tests studied the relationship between the moisture
contents of the sludge pellets after gravitational drainage, the achievable
moisture contents of dewatered sludge cakes*5 (AMC), and the operational
conditions, such as chemical dosing ratios, etc.
*) The moisture content of a dewatered sludge cake decreases and the
clogging level increases, as the dewatering pressure of the belt-press
filter increases. Thus, the most preferable operation of a belt-press
filter is done with the highest dewatering pressure, and the least filter
cloth clogging. The achievable moisture content of the dewatered sludge
cake is defined as the minimum moisture content of a dewatered sludae
cake with less than 10% of the clogging area on the filter cloth in batch
dewatering tests with sludge- compression.
135
-------
The tests were done to examine the dewatering efficiencies in the
gravitational drainage zone, the primary pressure zone, the secondary
pressure zone, and the final dewatering zone with the shearing force of a
belt-press filter dewatering machine.
The dosage of the metal coagulant aid was examined by the capillary
suction time technique, searching for the inflection point of the CST versus
chemical dosage curve (the CST procedure).
The dosage of the amphoteric polymer was decided by examining the SS
recovery efficiency for the sludge pelletization thickener and the
mechanical strength of sludge pellets under various operational conditions.
7. Results and Discussions
7-1. Sludge Pelletization/Thickening Efficiency
The dosage of the metal coagulant aid, the dosage of the amphoteric
polymer, the condensing ratip in the sludge pelletizing thickener, and the
system's solid retention time in the pelletizing thickener are the
operational conditions of the BEST system. The influence of these
operational conditions on pelletizing and dewatering performances were
examined.
1) Dosage of Metal Coagulant Aid (Polyaluminum chloride)
A metal coagulant aid is dosed to the fed sludge, with the objective of
neutralizing the anionic sludge charge. The dosage of the metal coagulant
aid is decided by the CST procedure.
The relationships between the CST value and the dosage of the metal
coagulant aid for the mixed raw sludge and the excess sludges of a
conventional activated sludge process are shown in Figure "11 and Figure 12,
respectively.
Both Figures 11 and 12 show that the inflection points lay in almost
the same dosage range. This fact suggests that proper dosage of the metal
coagulant aid will range from 0.6 to 0.8% as A£ to SS.
136
-------
o
0)
W"
£3
o
500
400 -
300 -
200 -
100 -
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
Dosage of metal coagulant aid (% as A£ to SS)
Figure 11. Relationship between the dosage of metal coagulant aid
and the CST value. (Mixed raw sludge of a conventional
activated sludge process.)
200 -
8
I
100 -
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Dosage of metal coagulant aid (% as AS. to SS)
Figure 12. Relationship between the dosage of metal coagulant aid
and the CST value. (Excess sludge of a conventional
activated sludge process.)
137 .
-------
The relationship between the AMC and the dosage of the metal coagulant
aid is shown in Figure 13. As shown in the figure, the smallest AMC value
is about 70%, with the metal coagulant aid dosage of 0.6 to 0.8% as A£ to
SS. This means that the actual dewatered sludge cake separation will be
satisfactory, when 75% of the moisture content of the cake is achieved,
which was one of the targets.
75
70
65
SS = 1.06%
Amphoteric polymer = 1.38% to SS
SSRT = 12 min
Condensing ratio = double
100
90
80
0.2
0.4 0.6 0.8 1.0 1.2
Dosage of metal coagulant aid (% as A2 to SS)
•H
O
•H
-------
thP ™nh relationship between the SS recovery efficiency and the dosage of
the amphotenc polymer for the Mixed Raw Sludge is shown in Figure 14 ?he
fU lin^Cate that thS mC Sh°Wed far less than the target vSue of "75%
when the dosage was 0.6% to SS, or more. And an SS recovery Ifficiencv of
95% or more was achieved with the dosage of 0 7% to SS e"iciency of
70
o
65
Figure 14.
SS = 1.76%
C?agulant aid: °'45% as
= 12 mm
Condensing ratio: double
to
100
95
90
0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2
Dosage of amphoteric polymer (% to SS)
B
O
•H
_
a)
s
o
-------
85 r
80
75
SS = 1.04%, CST = 170 sec
Metal coagulant, aid: 0.76% as A£ to SS
SSRT - 12 min
Condensing ratio: double
i L u :
0.5 1.0 1.5
Dosage of amphoteric polymer (% to SS)
100
95
90
2.0
H
CD
•H
O
•rl
0)
§
8
1-1
CO
Figure 15. Relationship between the dosage of amphoteric polymer,
and the mechanical strength of sludge pellet and the SS
recovery efficiency. (Excess sludge of a conventional
activated sludge process.)
From these results, the proper dosage of the amphoteric polymer is
revealed to be 1% to SS, or less.
3) Condensing Ratio and SSRT of Sludge Pelletizing Thickener
As one of the functions of the BEST system is a sludge thickening
process, the solid content of the sludge to be supplied to a dewatering
machine from this system is very important. Therefore,, to estimate the
solid content of the sludge from this system, the condensing ratio and the
SS recovery efficiency are defined as follows.
CR
SSRE
Q
in
CA
where CR :
SSRE:
Qm :
Qt :
x 100%
the Condensing Ratio (-)
the SS Recovery Efficiency (%)
Sludge feeding Rate (m3/min)
Separated Water withdrawal Rate at the Sludge
Pelletizing Thickener (m /min) ' ' -
140 • ' .
(2)
(3)
-------
Cm-
CA
Ct
Solid Content of Supplied Sludge (mg/jg)
Solid Content in the Reactor (mg/£)
Solid Content of the Separated Water (mg/£)
.By reforming Equation 3, the following equation is obtained.
CA =
x
Cin x Oin = SSRE
(4)
According to Equation 4, to increase the solid content of sludge fed to
o 3 maCh^6' the Censing ratio should be kept high while keeping
of t rec°verj efflciency hi9h- To examine the extent of the solid coJten?
of the sludge from this system, the relationship between the SS recovery
^ M16n!;Vndo?he S°lid content in the sludge pelletizing thickener when
the Mixed Raw Sludge was fed to the system was examined, and is shown in
Figure 16. it is suggested that the SS recovery efficiency tends to
decrease when the solid content in the reactor exceeds 3 or 4% This
implies that the opportunity for the sludge pellets to clash increased (and
then small sludge particles were produced) as the solid content in the -
8
CO
•H
O
•H
Q)
8
0)
u
CO
CO
100
90
80
Figure 16.
_L
J.
1 2 3 4
Solid content in the sludge pelletizing thickener (%)
Relationship between the solid content in the sludge
pelletizing thickener and the SS recovery efficiency.
(Mixed raw sludge of a conventional activated sludge process.)
141
-------
The same investigations were conducted for the Excess Sludges, and
almost the same trend was observed. However, the critical solxd content
claXing was slightly smaller than that for the Mixed Raw Sludge. It was
found to be about 3%, or less, of solid content.
From these results, it was found that the solid content in the sludge
pelletizing thickener should be operated with 3 to 4% when the Mixed Raw
Sludge is fed, and with 2 to 3% when the Excess sludges are fed.
Next the relationship between SSRT*> and the moisture content of the
dewaterS'sludge cake was examined. The relationship for the Mixed Raw
Sludge is shown in Figure 17.
dp
0)
A:
-------
Figure 17 implies that the mechanical strength of the pellets will be
higher, as the moisture content of a dewatered sludge cake is lowered, when
the PT value is increased, and that a fine separation will be expected
under these conditions. , .
Figure 17 also implies that the mechanical strength of the sludge
pellets increased as the SSRT increased. However, the sludge cake
separation was sufficient when the SSRT was 6 minutes, or longer. Thus, the
SSRT in the sludge pelletizing thickener should be kept for a minimum of 6
minutes when the Mixed Raw Sludge is fed. • ..
Also, it^ was revealed that the SSRT in the sludge pelletizing thickener
should be kept for about 12 minutes for the Excess Sludges.
7-2. Dewatering Performance with a Batch Tester
Dewatering examinations were made employing a belt-press filter batch
tester. Figure 18 describes the tester. The dewatering performances were
evaluated with the procedure shown in Figure 19.
Side roll
Filtered water
receiver
Press dewatering roll
Filter cloth
'Air cylinder / Filter cloth
tensidning roll
Filter cloth
driving motor
Compressor
Figure 18. Belt press filter batch tester.
143
-------
Feeding pelletized sludge
Conditions:
Gravitational drainage
time: 1 min.
Compressing time:
1 min.
Dewatering^pressure:
0.03kg/cm2
Evaluation:
Pellet diameter
Filtration
test
Times of roll compres-
sions: 3 to 15 times _
Roll pressure: 0.5kg/cmz
"Filter cloth driving
rate: Im/min
Roll commpres-
sion test
Gravitational drainage
efficiency: *
Filtered water volume,
Cake volume
Preliminary press-
dewatering property:
Side leak,
Filtered water volume,
Cake volume
Moisture content of cake,
Cake thickness,
Cake expansion,
Cake separation,
SS of filtered water
Analysis
Filtration rate,
Moisture content of
cake,
Others
Figure 19. Procedure of batch dewatering tests.
The dewatering performances as a control were also examined with single
polymer conditioning after centrifugal sludge thickening, to evaluate the
dewatering performance by the BEST system.
The characteristics of the applied sludges are listed in Table 3.
144
-------
•a
§ •
CO
- cu
cu cn
•a1?
D rH
rH W
CO
•a
3 0)
u
o
•H >.
4-> rH
CO rH
•H Id'
ij Cn
CD S
4-* *4-l
O -H
J x^
^ -H CO OP
u
-J8 ~
•H CO <#>
tK CO »-'
^x.
co '-•
co co *>
> CO v
* /"t
CO dp
CO —
co ~
P-I U5 c#p
^ E^ ^-*
u> 5?
FH ^-^
•a >«s.
c to ~
5 a E
O ~ 0
E T
a ^
0)
>J
4->
a)
O)
r0
rH
CO
CO '•-*
o in
^* c^*
rH CO
vo co
t-^ in
CO 00
*— '
cn CO
O *d*
CM CM
• •
0 0
vo CO
cn vo
rx tx
'—'
VO ^
CO CO
CO 00
rH CM
"•^
^^
co in
CO ^J*
VO CO
CO rH
•* VO
^5* ^
• •*-*
CM .7
CO C*1*
(-1 (H
s«."
^— s
ir>
• *
CO t-l
,-^
t>- CO
CM CM
VO VO
*— ^
^^
rx r>-
vo cn
CO rH
*— '
o o
rH O
CM M<
CO CM
VO O
in vo
* —
Q)
cn
t3
CO
S
<0
u
1
is
tx <-*
•<* in
rH rH
^^ -
vo O
CD r-i
O^ C7^
N— '
oo
CD CU O
O CO U
X G ®
W CO (0
CO ^
rH CM
rH VO
CO CM
cn cn
>— '
cn [x
^d* CO
CM CO
• » "-"
O O
VO ^
tx 0
co rx
i*— '
0 0
CO cn
cn rH
CO CM
\^r
^^
•^ CO
o in
cn rH
cn co
t^ rH
in vo
cn TJI
rH CO
^.^
CO CM
cn o
VO tx
s-^
vo Cx
O *^
• •
• co o
»^-v
CM cn
o in
Cx VO
•~*t
^^
O CM
CM in
CO O
*^f
o o
rH CO
tx CO
t-l tH
in vo
VO VO
* — '
4-1
O
CO
0) CO
en I-H cu
'O (0 O
^3 C O
rH O lJ
CO -rH Q,
CO C 0)
co a) cn
a) > -a
g c 3
X O rH
wow
.CO --<
rH ^<
^— ^
CO CD*
rH CM
cn cn
cn CM
CM CO
CM CM
CD CD
rH CO
CO rH
tx in
V— '
o in
rH cn
CO rH
cn cn
rH ^~*
^
O CO
tx ^
O"v !>•
00 C^-
VD KD
IO 1%
00 OJ
t-l r-1
^_^
T^< *^f
, .
^s, Q
C^ CO
,~-+,
vo O
cn CM
CO CM
f^~^
*^* r~\
C^ CT^
r^" r^>
***r
^_^
O vo
-rH 00
^Ji CS
^-^
8§
VO ^J1
rH rH
cn in
in vo
~-^
t-l
o &
r)
CU l->
Cn -H
T3 'O
— J
rH C
co o
•H CO
CO 4-> CO
w (0 cu
CD -a o
x "x 8
W O Pj
•
. Cn
c
•H
s
.s
o
•H
x:
rH
(0
cn
D
u
*J
a)
0
£
o
-------
r
The relationship between the,filtration rates and the moisture contents
of dewatered sludge cakes are shown for both the Mixed Raw Sludge and the
Excess Sludges are shown in Figure 20 and Figure 21, respectively.
0)
0)
O)
•a
3
r-t
CO
Q)
J-i
0)
Q)
•P
C
0)
4->
s
O
a)
u
-M
CO
•rt
80
75
70
65
O
D/0
BEST system
Centrifugal thickening
+ Single polymer
conditioning
Filtration limit
for the control
0-O
I
I
100 200
Filtration rate (kg-ds/m«hr)
300
Figure 20. Relationship between the filtration rates and the moisture
content of dewatered sludge cakes. (Mixed raw sludge of a
conventional activated sludge process.)
146
-------
85
(0
o
o
4->
O
0)
(-1
3
•H
i
80
75
Arrowed data showed side-leaks.
0 50 100 150
Filtration rate (kg-DS/m/h)
BEST : •Conventional, A A/O, • OD
Contl: O Conventional, A A/O, D OD
200
Figure 21. Relationship between the filtration rates and the moisture
content of dewatered sludge cakes.
(Excess sludges)
According to the results in Figure 20 for the Mixed Raw Sludge, the
achievable maximum filtration rate for the control was found to be less than
130 kg-DS/m/h, as a side leak of sludge was observed when dewatered with
this rate. Then, the moisture content of the dewatered sludge cake was 71%.
On the other hand, when the BEST system was employed, it was possible -
to dewater the Mixed Raw Sludge with a filtration rate of 220 kg-DS/m/h
without any failure in dewatering. Then, the moisture content of the
dewatered sludge cake was 71% in the examination. For the dewatering
performance of a belt-press filter in Japan, it is generally said that 78%
of the moisture content will be obtained with a filtration rate of 130
kg-DS/m/h when a gravitationally thickened mixed raw sludge is fed. The
moisture content attained with ±he BEST system was lower than the
conventional value, and the achieved filtration rate was almost twice that
of the conventional performance. r „
147
-------
The results for the Excess Sludges in Figure 21, also show that the
BEST system could attain lower moisture content .of the cake and larger
filtration rate compared with the control system. For example, a sludge
side leak was observed when sludges were dewatered with a filtration rate of
about 50 kg-DS/m/h with the control method, and more than twice the
filtration rate was achieved when the BEST system was used to dewater the
excess sludge of a conventional activated sludge process. In addition, the
moisture content for the BEST system was 2 to 5% lower than that for the
control method.
7-3. Full Scale Evaluation of Dewatering Performance
Full scale run of the BEST system was tested in the T treatment plant
with the operational conditions listed in Table 4. The BEST system was
incorporated between the sludge storage tank and the belt-press filter as
shown in the schematic flow diagram of Figure 22.
Primary
sedimentation tank
Aeration tank
Secondary
sedimentation tank
Mixed raw sludge
Belt press filter
Gravitational thickener
Sludge storage tank
External disposal
Figure 22. Schematic flow diagram of the T plant.
148
-------
Table 4. Operational Conditions of the T Plant.
Wastewater Collection System
Separate Collection System
Wastewater Treatment Process
Conventional Activated Sludge
Process
Influent Rate
3,600 to 6,400 m3/day
BOD-SS Loading
0.64 kg-BOD/kg-SS/day
Sludge Handling System
Gravitational Thickening
==> Dewatering
(Belt-Press Filter)
==> Disposal
Solid Content of Sludge
Mixed Raw Sludge
= 0.8 to 1.6% of SS
Thickened Sludge
= 0.98 to 1.63% of SS
7? ?ef rUn WaS Performed for one month, and the characteristics of
betwpS *iUdf*dUling thiS P6"^ are shown in Table 5- The relationship
between the filtration rates and the moisture content of the dewatered
sludge cakes are shown in Figure 23. in the figure, the annual average
values of the moisture content of the dewatered cake and the filtration rate
are also shown.
Table 5. Characteristics of sludges during the experiment.
PH
Cond
(uS/cm)
SS
VSS/SS
TS
VTS/TS
Fiber/TS
5.6
to 6.3
1,560
to 1,900
0.98
-to 1.63
86.2
to 88.3
1.11
to 2.30
82.3
to 86.4
15.5
to 29.3
149
-------
0)
o
0)
en
to
0)
0)
•o
0)
•p
o
•P
CO
•H
74
72
70
68
A
A :
O :
Annual average with single
polymer conditioning
Condensing ratio = 0
Condensing ratio
1.5
O
_L
BEST system
0 100
200 300
Filtration rate (kg-ds/m-hr)
400
Figure 23. Relationship between the filtration rate and the moisture
content of dewatered sludge cakes. (At the D plant; Mixed raw
sludge of a conventional activated sludge process.)
As shown in Figure 23, the average filtration rate and the average
moisture content of the cake were 90 kg-DS/m/h and 75%, respectively when
the single polymer conditioning had been employed. Improvements in the
molsturJ content decreaseding to 72%, and in the filtration rate increasing
to 150 to 160 kg-DS/m/h are shown in the figure, even when the BEST system
was employed wi?h a condensing ratio of zero. Also, the further improvement
in the moisture content to 70% decreasing with an increased filtration of
220 kg-DS/m/h are shown when the BEST system was employed with a condensing
ratio of 1.5. :
The further improvements in the moisture content and the filtration
rate are considered to be due to the following reason. The mechanical
SrLgth o? the sludge pellets increased because of the elongated, SSRT in
the sludge pelletizing thickener brought about by the increase, of solid
content in Se reactor as the result of withdrawing separated water Thus
withdrawal of separated water from the sludge pelletizing thickener is found
to be very effective in improving the dewatering efficiency of the system.
150
-------
The conventional value of dewatering performance for a belt-press
filter is described (in the specification code of the Japan Sewage Works
Agency) as being able to make the moisture content of the dewatered sludge
cake 75%, or less, of the solid content, with a filtration rate of
130kg-DS/m/h, when a mixed raw sludge is fed with 3% solid content after
gravitational thickening. However, it is said that to attain these values
have become extremely difficult in many municipalities in recent years.
From these facts, the dewatering performance of the BEST system can be
described as being very good.
7-4. Phosphate Fixation ;
The relationship between the ortho-phosphate concentration (O-P) and
the dosage of metal coagulant aid, when a fresh biological phosphorus
removal excess sludge was fed to the system is shown in Figure 24.
O-i
o
CO
8
10 -
- 10
0
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Dosage of metal coagulant aid (% as A2 to SS)
Figure 24. Relationship between the orthophosphate concentration and the
dosage of metal coagulant aid. (Fresh excess sludge of an
anaerobic/oxic process.)
The results in the figure indicate that O-P was almost completely fixed
within the sludge with a metal coagulant aid dosage of 0.5% as A£ to SS. As
discussed in Section 7-1, a metal coagulant aid dosage of about 1% as A£ to
SS is required for charge neutralization, so the O-P in the sludge liquid
will be almost completely fixed within the solids with this dosage.
151
-------
Figure 25 shows the relationship between the same items shown in
Figure 24, using excess sludge from a biological phosphorus removal
process. The sludge was stored 5 days under anaerobic condition prior to
the test. The results shown in the figure indicate that a dosage of 1.5% as
A£ to SS was required to reduce O-P to the level attained with 0.5% dosage
when the fed sludge was fresh. From the economic point of view, the sludge
should be processed while it is fresh to fix O-P within the sludge.
1135
0.
6
20 -
10 -
- 400
- 300
- 200
- 100.
o
-------
Table 6. Conditions of cost estimation
Kind of Sludge
BEST system
Moisture '
content of _ .
dewatered Dosa?e of Dosa9e of
cake (%) metal • amphoteric
coagulant aid polymer
(% as A£ to SS) (% to SS)
Convent ional
system
Dosage of
cationic
polymer
(% to SS)
The Mixed Raw
Sludge
70
0.46
0.78
0.60
Excess Sludge of
convent ional
activated sludge
process
82
0.89
0.86
1.31
Excess Sludge of
anaerobic/oxic
process
82
0.74
1.18
0.63
Excess Sludge of.
oxidation ditch
process
82
0.60
0.83
0.33
The estimates for the BEST system include costs for the metal coagulant
aid and the amphoteric polymer. The estimates for the conventional system
include costs for the cationic polymer and electricity for the centrifuge.
Costs for electricity required to operate pumps and stirring equipment were
not included in the estimates, as they were thought to be in significant "in
relation to the total costs. The results are shown in Figure 26. In the
figure, operation hours are shown as well.
153
-------
Mixed raw sludge of a
conventional activated
sludge process
Excess sludge of a
conventional activated
sludge process
Excess sludge of an
anaerobic/aerobic
process
Excess sludge of an
oxidation ditch process
O CO
OP
en 4->
83
•H ^x
30,000
20,000
10,000
10
12
Metal coagulant aid
Amphoteric polymer.
Electricity for centrifugal
sludge thickening
Cationic polymer
Note 1: The left bar represents the BEST system, and the right bar, the
single polymer conditioning after centrifugal thickening.
Note 2: The operation hours for the conventional system indicate the
hours required to dewater the same amount of sludge that the
BEST system dewater within 7' hours.
Figure 26. Comparison of operational cost.
154
-------
According to the results, the estimates for the BEST system are a
little bit high compared with those for the conventional system when
handling the Mixed Raw Sludge. The moisture contents of the dewatered
sludge cake dewatered with both systems are the same, however, though the
achievable filtration rates differ significantly between the two. When
7-hour operation of a belt-press filter is required to dewater the Mixed Raw
Sludge with the BEST system, 11.4-hour operation of the same belt-press
filter will be necessary to dewater the same amount of sludge processed with
the control system. '
9. Conclusion
1)
The following is obtained in this study of the BEST system.
The BEST system makes it possible to dewater any of the mixed raw sludge
of the conventional activated sludge process, the excess sludge of the
conventional activated sludge process, the excess sludge of an
anaerobic/aerobic process, and the excess sludge of an oxidation ditch
process sludges, without any of conventional sludge thickening processes.
2) The required dosage of the metal coagulant aid for the BEST system
ranges between 0.6 and 0.8% as A£ to SS of any sludges examined in this
study*;
3) The dosage of the amphoteric polymer for the BEST system is about 1% to
SS for any of the fed sludges. The required dosage of the amphoteric
polymer is almost the same, or slightly larger, compared with that of
the cationic polymer dosage in a single polymer conditioning system.
. 4) The sludge pelletizing thickener should be operated with 3 to 4% of the
solid content, and with 6 minutes, or .more, of SSRT when the mixed raw
sludge is fed into the system. When excess sludges are fed into the
system, the sludge pelletizing thickener should be operated with 2 to 3%
of the solid content, and for 12 minutes, or more, of SSRT. The system
is very compact compared with a gravitational thickener.
5) The dewatering performance of the BEST system revealed that the moisture
content of dewatered sludge cake can attain the target volume of 75% or
less, with a filtration rate of 200 kg-DS/m/h, when a mixed raw sludge
is fed to the system and dewatered with a belt-press filter, when
excess sludges are fed to the BEST system, the dewatering performance
was improved compared with conventional system.
6) The running costs for this system are-estimated to be almost the same
or slightly more compared with that for the mechanical thickening
dewatering system, which employs a centrifuge for thickening, and the
single polymer dosing to condition the sludge.
7)
The release of phosphate in the system is almost completely cut off by
dosing the metal coagulant aid.
155
-------
For these characteristics of the BEST system, the following applications
are now under consideration.
(1) Application to sewage treatment plants, where the gravitational sludge
thickeners do not work efficiently.
(2) Application to small size sewage treatment plants, as an itinerant
sludge handling system by trucking a belt-press filter combined with
this system.
(3) Application to the sewage treatment plants where biological phosphorus
removal processes are employed.
Reference ,
1. Annual Technical Report, Sewage Works Bureau of Tokyo Metropolitan
Government, March, 1989.
156
-------
Operation of Centralized Sludge Treatment Plant
by
Senji Kaneko
Chief, Research & Development Division
Sewage Works Bureau
City of Yokohama
The work described in this paper was not funded by
the U.S. Environmental Protection Agency. The contents
do not necessarily reflect the views of the Agency and
no official endorsement should be inferred
Prepared for Presentation at:
12th United States/Japan Conference
on
Sewage Treatment Technology
October 1989
Cincinnati, Ohio
157
-------
Abstract
Th'e construction of sewage works in the city of Yokohama has progressed
at a rapid pace ever since the first wastewater treatment plant went into
operation in 1962. As of 1984, all eleven of the planned wastewater treatment
plants were in operation. In March of 1989, the percent of sewered population
reached'80 percent, or 2.54 million of the city's total population of 3.15 million.
The quantity of sludge that has to be, treated and disposed of has, in the
meantime, increased along with this construction. Recognizing the crucial
importance of constantly keeping abreast of the ever-increasing quantity of
sludge, the city has been pursuing technological-development aimed at estab-
lishing more effective and efficient methods of sludge treatment and disposal.
In view of this city's rapid urbanization and the rising citizen concern for
improving and preserving the environment, the city.decided to modify the
sewerage plan, shifting from the former system of decentralized sludge
treatment at each of .the eleven wastewater treatment plants to a centralized
treatment system at two sludge treatment centers constructed within two
waterfront area wastewater treatment plants.
One of'these sludge treatment centers (Hokubu) began operating in
September 1987. The center contains egg-shaped digesters, the first of their
kind in Japan. To insure the efficient operation of these digesters, centrifugal
thickeners are employed during pretreat'ment, enabling high-concentration
digestion. In addition, the digestion gas energy is recovered as electrical power
owing to a gas power generation system that transforms the energy. Further-
more, before incineration, the sludge cake is dried using the heat released from
the incineration system, thus enhancing the efficiency of incinerator opera-
tions. These and similar measures demonstrate the city's efforts to incorporate
technological development into treatment facilities in order to optimize the
total system.
Full-scale operation records for approximately one year indicated that
power generated from digestion gas supplied 64 percent of the center's total
power needs. This figure amounts to an annual savings of ¥ 200 million for the
present volume of sludge treatment, which is one-fifth of the planned volume.
The record also confirms the advantages of centralized treatment; it was found
that sludge treatment cost at Hokubu was less than half the sludge treatment
cost for the same quantity of treatment processed at the individual wastewater
treatment plants.
This report presents the center's operational record for the year in
question, the actual status of energy recovery and recycling, at the center, and
current modes for effective utilization of incinerated sludge ash.
158
if?
-------
1. Foreword
The quantity of sewage sludge generated in cities has been steadily increasing
with the expansion of sewerage systems. The increase is particularly eminent in
major cities, which not only shoulder the enormous burden of costs required for
treatment and disposal, but also are finding it increasingly difficult to acquire the
needed disposal sites.
In the city of Yokohama, the construction of the sewerage system has
progressed at a rapid pace ever since the first wastewater treatment plant went into
operation in 1962. As of 1984, all eleven of the planned wastewater treatment plants
were in operation. As of March 1989, the sewerage service rate reached 80 percent,
or 2.54 million residents of the entire population 3.15 million.
The quantity of sludge generated in the city has been expanding each year with
this progress. Efficient and effective treatment and disposal of sludge are becoming
increasingly important issues. Moreover, the city experienced rapid urbanization
during Japan's recent period of high-level economic growth. Subsequently, the
improvement and preservation of the environment have become major concerns of
city residents. In view of these factors, the city decided to shift from its former
system of decentralized sludge treatment at each wastewater plant to a new system
of centralized treatment at two sludge treatment centers constructed at two
wastewater treatment plants located in the waterfront area.
One of these sludge treatment centers (Hokubu) went into operation in
September 1987. This report presents the operational record of this facility and the
effects of centralized sludge treatment.
159
-------
2. Current status of sewerage system construction in Yokohama
Full-scale-construction of sewerage system including waste water treatment
plants in Yokohama dates from 1957. This construction is being promoted in a series
of five-year plans, the sixth of which is now underway, in a program called the Five-
year Program for Sewerage Construction. In the process, the city's sewerage system
has greatly expanded. Initially, the construction was confined to the oldest parts of
the city in the waterfront area; it now covers all areas of the city. At present, the
main priority is an increase in the sewered population in the city as a whole, and
particularly in suburban areas.
In the sewerage system plan, the city is divided into nine treatment districts
and includes 11 wastewater treatment plants, as shown in Figure 3-1 in the
following section. With the initiation of the newest plant in 1984, all eleven of the
planned wastewater treatment plants were in operation 22 years after the start-up of
the first plant in 1962.
The original plan called for each wastewater treatment plant to carry out its
own sludge treatment. Subsequently, however, it was decided to construct sludge
treatment centers at two specific plants in the water front zone, where centralized
treatment of the sludge generated at each of these plants could be done. This report
concerns the operation of the Hokubu Sludge Treatment Center, which began
operation in September 1987. The other center-the Nambu Sludge Treatment
Center—is scheduled to go into operation during fiscal year 1989. Table 2-1 shows
the current status of the city's sewerage system construction.
Table 2-1. Status of Sewerage System Construction (as of March 1989)
City area (ha)
Urbanization promotion area (ha)
Treatment district area (ha)
Total city population (thousands of persons)
Sewered population (thousands of persons)
Percent of sewered population (%)
Extended sewer length (km)
Number of WTPs (in operation)
Treatment capacity (103 m3)
Number of pumping stations (in operation) .
Number of stormwater retention reservoirs (in operation)
43,082
32,473
22,136
3,158
2,542
80
7,906
11
1,844
22
3
160
-------
3. 'Centralized sludge treatment plans and an outline of the facilities
3-1. Plans and outline ....
3-1-1. Centralized sludge treatment plans
The city has formulated plans to construct two centralized sludge treatment
centers in the waterfront area~the Hokubu and Nambu sludge treatment centers.
As shown in Figure 3-1, the plans call for the treatment of sludge generated from the
waste water treatment plants in the northern section of the city, namely, the Midori,
Kokubu, Hokubu No. 1, Hokubu No. 2 and Kanagawa plants, at the Hokubu Sludge
Treatment Center, and for the treatment of sludge.generated from those plants in
the southern part, namely, the Chubu, Nambu, Kanazawa, Sakae No. 1, Sakae No.
2, and Seibu plants, at the Nambu Sludge Treatment Center. Table 3-1 shows the
outline of the city's wastewater treatment plants.
As part of these plans, sludge is transported to the centers by means of
pressurized pipelines. The sludge is not routed directly from the wastewater
treatment plants to the centers, however. Wherever possible, it is routed through
one or more wastewater treatment plants served by the same center.
Table 3-2 shows an outline of the sludge treatment center plans. Several
treatment processes are planned for both centers in order to provide flexibility in
sludge treatment and accommodate future technological innovation.
3-1-2. Current construction status of the centralized sludge treatment plant
As of March 1989, virtually all of the sludge pipelines had been completed.
Routes 11 and 12 are still under construction, but routes 1 through 6 are already in
operation, and routes 7 through 10 are scheduled for operation during fiscal year
1989.
Table 3-3 shows the construction status of the sludge treatment facilities at the
Hokubu Sludge Treatment Center. In September 1987, the Center began full-scale
operation based on a sequence process of thickening, digestion, dewatering, and
incineration. The Nambu Sludge Treatment Center is scheduled'for operation
during fiscal year 1989.
161
-------
Fig. 3-1. Map of Sludge Pipelines
\ ,-KanagawaWTP
HokubuNo. 1WTP
\
HokobuNo.2WTP
Hokubu Sludge
Treatment Center
Isogo Pumping Station'
\
.
SakaeNo.2WTP I/
ChubuWTP
NambuWTP
Nambu Sludge]
Treatment Center
SeibuWTP
KanazawaWTP
Legend:
» In operation
===£> Under construction
162
-------
Table 3-1. Sewage Treatment Plants
(as of March 1986)
Treatment plant
1. Hokubul
2. Hokubu n
3. Kanagawa
4. Chubu
5. Nambu
6. Kanazawa
7. Kohoku
8. Midori
9. Sakael
lO.SakaeH
ll.Seibu
Design treatment capacity
(m3/day) (population) .
196,000 (377,000)
86,400 (166,000)
543,200 (1,074,000)
96,300 (185,000)
255,000 (433,000)
'345,000 (663,000)
439,000 (844,000)
433,000 (833,000)
124,000 (238,000)
206,000 (396,000)
191,000 (367;000)
Current treatment
capacity
(m3/day)
196,000
86,400
380,200 •
96,300
225,000
230,000
172,500
126,000 .
62,000
206,000
63,600
Operation
started
July '68
Aug. '84
Mar. '78
July '62
July '65
Oct. '79
Dec. '72
May '77
Dec. '84
Oct. '72
Mar. '83
Table 3-2. Outline of Centralized Sludge Treatment Plans
Center
Hokubu
Nambu
Wastewater
treatment
plants covered
Midori
Kohoku
Hokubu No.l
Hokubu No.2
Kanagawa
Chubu
Nambu '
Kanazawa
SakaeNo.l
Sakae No.2
Seibu
Capacity
(TS: 1%)
44,100
m3/day
30,000
m3/day
Sludge treatment processes
Thickening-* anaerobic digestion—* dewateringr-* incineration
(gravity-type/centrifugal) (centrifugal) ^drying
Thickening—" wet oxidation—" dewatering
(centrifugal) (filter press)
Thickening-^ anaerobic digestion— *de watering^- incineration
(centrifugal) (belt press)
Thickening-^- wet oxidation— * dewatering
(centrifugal) (filter press)
163
-------
Table 3-3. Construction Status of Hokubu Sludge Treatment Center
Facilities - , „„„„,
(as of March 1989)
Facility
Sludge intake tanks
Thickening tanks
Centrifugal thickeners
Digesters
Centrifugal dehydrators
Incinerators
Gas engine power generators
Gas holders
In operation
1,500 m3X 2
1,260 m3X 8
100m3/hrX3
6,800 m3x 6
50m3/hrX3
lOOwt/dayXl
150wt/dayX2
920kWX3
8,000 m3xl
Under
construction
—
—
—
6,800 m3X 6
—
—
150wt/dayXl
920kWXl
8,000 m3 XI
Note: The abbreviation wt refers to wet ton.
3-2. Outline of centralized sludge treatment (Hokubu Area)
3-2-1. Sludge transportation system
The sludge generated from the primary and final sedimentation basins at
wastewater treatment plants is transferred, in mixed form, to concentration
conditioning tanks, where the TS concentration is adjusted to about 1 percent. The
sludge is then stored in a holding tank and transported through pipelines
pressurized by pumps to the Hokubu Sludge Treatment Center.
The sludge generated at the Midori plant is first transported via the Kohoku
plant to the Hokubu No. 1 plant. At each plant, the sludge is mixed with that
generated from the wastewater treatment plant, and then transported to the Center.
The sludge generated at the Kanagawa plant is transported directly to the
Center after its TS concentration has been adjusted. The sludge generated at the
Hokubu No. 2 plant is also transported directly to the Center, in the form of primary
sludge, i.e., sludge drawn from the primary sedimentation tank.
Ideally, the operation and management of the sludge transportation and intake
system should be integrated with that of sludge treatment, the two forming a single
system. Taking this into account, the city has adopted a system of centralized
supervision and decentralized control.
164
-------
3-2-2. Hokubu Sludge Treatment Center
As shown in Table 3-2, the Center has a daily treatment capacity of 44,100
cubic meters of sludge (TS: 1%) through a combination of three processes: anaerobic
digestion/dewatering/incineration, digestion/dewatering/drying, and wet air
oxidation /dewatering. At present, only the first process (digestion /dewatering/
incineration) is in operation. This section presents an account of the facilities
involved in this process. Figures 3-2 and 3-3 present the layout of the Center and the
flow of the digestion/dewatering/incineration process, respectively.
(a) Thickening process
The sludge transported from the wastewater treatment plants first enters the
sludge intake tanks and is then transported to thickening tanks. In these tanks the
sludge concentration is adjusted to about 2 percent and then increased to about 5
percent, by means of a centrifugal thickeners. In recent years, it is presumed that
the thickening properties of sludge have deteriorated as a result of the rise of sludge
VTS and decomposition during long-term transportation. Consequently, gravity
thickening is supplemented with mechanical thickening to provide for the stability
of, and alleviate the burden on, subsequent processes. The Center has adopted a
mode of centrifugal thickening for this purpose, seeing that maintenance is
relatively simple and the equipment's compact size facilitates installation.
(b) Digestion process
s
The Center has adopted an anaerobic digestion process in the interests of
enhancing the overall flexibility of the facilities and making effective use of the gas
generated in digestion. The egg-shaped tanks used for this digestion are the first of
their kind in Japan and offer various advantages; they generate an extremely high
agitation efficiency, ensure a minimal generation of scum, and facilitate the easy
removal of sediment. In addition, the construction cost per unit of treated solids is
virtually the same as for conventional tanks. There are six such tanks, each with a
capacity of 6,800 m.3. .
(c) Power generation process using digestion gas
The heat and power, derived as a result of the power generation using digestion
gas, are effectively utilized for the purposes of heating the digesters and supplying
power for various facilities within the Center. The system now consists of three
generators in the 920-kilowatt class.
165
-------
(d) Dewatering process *
The Center uses centrifugal dehydrators to dewater the sludge after digestion.
These centrifugal units are easy to maintain and offer a high solids capture rate-and
contribute to efficient operations. There are now three such units in operation, each
with a capacity of 50 cubic meters per hour.
(e) Incineration process
To reduce the quantity of sludge for disposal, fluidized bed incinerators are
used to incinerate the entire quantity of sludge cake. The temperature of the
incinerator exhaust gas is about 800°C, and the exhaust gas is used for drying sludge
cake before incineration; thus increasing the efficiency of the operation. There are
now three such incinerators operating in the Center, one with a daily capacity of 100
wt and two with a capacity of 150 wt per day.
(f) Sidestreams treatment process
Various types of sidestreams are separated out in the sludge treatment
processes (e.g.,thickener effluent, supernatant from digestion tanks, filtrate from
dehydrators, and scrubber drain). There are plans to first pretreat these sidestreams
at facilities constructed for this purpose within the Center, and then undergo
secondary treatment process.
Fig. 3-2. Layout of the Hokubu Sludge Treatment Center
^^.
HokubuNo.2WTP
Hokubu Sludge
Treatment Center
Facilities for
effective use of
incinerated ash
washing facilities p
Power generation
"Digesters
Centrifugal thickeners
166
-------
Fig. 3-3 Flow of Anaerobic Digestion/Dewatering/Incineration Process Flow Chart
Gasholder Desuifurizati0n equipment
Gas engine power generator
h
Sludge intake tank Centrifugal
Thickening tank thickener
Digestier Centrifugal
dehydrator
Incinerator
Incinerated ash
iDigestion/drying process
I Wet air oxidation process -
"*•{ Sidestreams 1
.^treatment process J~
Wastewater treatment system
Legend
» Flow of sludge
*• Flow of digestion gas
*• Flowofsidestreams
L~~~\ Planned
At present, these plans are at the preliminary stages of process selection. In
the meantime, the sidestreams generated at the Center are being sent back for
treatment at three wastewater treatment plants (Hokubu No.-l, Hokubu No. 2, and
Kanagawa), distributed according to the load produced by such treatment.
3-2-3. Instrumentation
(1) Outline
The instrumentation systems of the sludge transportation facilities at each
wastewater treatment plant are linked to the centralized supervision and
decentralized control system at the* Center. This facilitates the integrated
management of the sludge intake facilities, thickening and digestion facilities, and
other sludge treatment facilities.
167
-------
The main computer system, which is at the core of this instrumentation
system, is equipped with a backup system for reliability and safety. The backup
system assures operational continuity in the event of an unplanned shutdown or
planned shutdown in accordance with the maintenance schedule of the other system.
Even if both systems are shut down, the facilities can be monitored and operated
through the monitor panel located within the central supervision and control room.
At the same time, each of the treatment facilities within the Center is equipped
with a local station, microcontroller, and sequence controller. 'These devices can
automatically control the facilities and exchange information with the main
computer. Utilizing optical fiber cable, information is transmitted between the'main
computer and local stations through a high-speed dataway.
Figure 3-4 shows the Composition of Instrumentation at the Hokubu Sludge
Treatment Center.
(2) Sludge transportation facility instrumentation *>
The function of controlling sludge transportation facilities is to assure the
smooth transportation of sludge generated at each of the wastewater treatment
plants to the Center. Sludge from numerous plants is transported through a single
pipeline in a cascade. For this reason, coordinating the transportation time.sequence
at each transportation facility is a major part of the operation. To effect such
coordination, the transportation facilities ar-e controlled by a daily sludge
transportation schedule. "
The schedule is developed in accordance with the Center's general operational
plans, which are based on estimates of the quantity of sludge generated at the
wastewater treatment plants, and is formulated by main computer operators. The
basic goals behind creating the plans are to maximize the total quantity of sludge
transported per day, assure that the minimum required flow, speed within the
pipeline is maintained, and that a proper balance between the quantity of sludge
entering the holding tanks and that of the transported sludge is upheld.
Sludge transportation is automatically initiated when the time specified in the
transportation schedule is reached or when the level of sludge in the holding tank in
the transportation facilities reaches its upper limit. Transportation is automatically
terminated when the quantity specified in the schedule' is reached (as calculated
from the quantity and frequency of sludge transportation for the day in question) or
when the level of sludge in the holding tank reaches its lower limit.
The sludge transportation facilities are also equipped with a timer control,
which has a sequence controller as a backup system in the event of microcontroller
failure. In this mode of control, the start-up and shutdown of the sludge
transportation pump are based on the time of transportation initiation and the
duration of pump operation in each case. .
168
-------
Fig. 3-4. Composition of Instrumentation at the Hokubu Sludge Treatment
Center
, • Ho^ubu Sludge Treatment Center-Areas of Control
Computer
Computer
Data way
(Optical fiber cable)
(STF)
IOC
CTR
SQC]
Sludge _p
I JTM/I1VJUL/ l°lugge p
VtD|MIiTC'ntake facilities^
RS -
CTR
SQC
RS
h
CTR
SQC
Digesters*
.... RS -.
CTR
SQC
[Thickeners Dehydrators Power generators
|_ Public telephone circuits
• •
Sludge transportation
facilities
Direct
transmission
Legend
FD: •
CB:
LP:
IRB:
TW:
MS:
CRT: ,
RS:
SYC:
Off premises of Hokubu Center
i i
On the premises of
Hokubu Center
Flexible disk drive
Central monitor board
Line printer IOC:
Information retrieval board CTR:
Output typewriter SQC:
Master station MD:
CRT display IB:
Remote station TM/TC:
System console STF:
F.low of sludge
'- Direct transmission
"1U1"" Public telephone circuits
(_ J Data way
Input/output controller
Microcontroller
Sequence controller
Modem
Instrumentation board
Telemeter/telecontrol
Sludge treatment facility
169
-------
(3) Detection of pipeline abnormalities
Blockage: Sludge transportation pipelines are considered blocked when both
the pump discharge pressure is higher than the set value and the quantity of sludge
flow is lower than the set value or has rapidly decreased. When blockage is detected,
the pump is immediately shut down and a blockage alarm is transmitted.
Leakage: Sludge transportation pipelines are deemed leaking when both the
pump discharge pressure is lower than the set value and the quantity; of sludge flow
is higher than the set value or has rapidly increased. When leakage is detected, the
pump is immediately shut down and a leakage alarm is transmitted.
(4) Instrumentation of sludge intake facilities
Because sludge residue within pipelines can cause blockage or generate gas,
they must be cleaned after transportation. As a general rule, the pipelines are
cleaned once every day. The cleaning is performed by flushing the pipelines with
pressurized water, using the sludge transportation pump. After the sludge
transportation for the day has been completed, the operator in the central
supervision and control room in the Center orders the commencement of the
cleaning.
On the intake side, separate tanks are used to receive the sludge and cleaning
water. A near infrared-type sludge -densitometer is attached to the pipeline
on this side to constantly measure the concentration. When the concentration of
fluids transported through the pipeline falls below the set level, the intake is
automatically switched to the cleaning water tank. The operation makes use of the
comparatively well-defined interface between sludge and cleaning water within the
pipeline, and of the sharp drop in concentration when sludge is replaced by cleaning
water.
170
-------
4. Operation of Hokubu sludge treatment center facilities
4-1. Operation of sludge transportation system
The Hokubu Sludge Treatment Center treats sludge transported by pipeline
from five wastewater treatment plants.
The sludge transportation system consist of transportation facilities (sludge
holding tanks and sludge pumps), sludge pipelines, and sludge intake facilities
(screenings removal units and sludge holding tanks).
Table 4-1 presents the quantities and the TS and VTS concentrations of sludge
generated at each of the five wastewater treatment plants during fiscal year 1988.
Table 4-1. Sludge Quantities and TS and VTS Concentrations
STP
Item
Quantity of sludge
generated (m3/day)
TS(%) '
VTS (%)
Midori
540-870
(680)
1.0-3.4
(2.4)
49-88
(80)
Kohoku
250-600
(360)
1.9-4.0
f (3.4)
51-79
(69)
Hokubu No. 1
630-1,000
(760)
1.5-5.5
(3.3)
52-84 .
(67)
Hokubu No. 2
160-660
(480)
0.8-2.2
(1.4)
. 66-83
(75)
Kanagawa
1,750-2,770
(2,120)
1.4-3.2
(2.0)
62-85
(78)
valuesoveraone-yearperi^F^^
The sludge is transported from the Midori plant to the Kohoku plant and
further to the Hokubu No. 1 plant with treated wastewater through-a pipeline In
order to separate .the sludge from the water, a near infrared reflective-type sludge
densitometer (TS meter) is attached to the pipeline in front of the sludge
intake tank at the latter two plants. The line of separation is a TS concentration of
0.1 percent; the pipeline is emptied into the sludge intake tank if the concentration is
over 0.1 percent and into the treatment plant grit chamber if it is under 0.1 percent
Change m TS concentration within the pipeline is well defined, and the separation of
sludge and treated wastewater could be effected even at a higher TS concentration
setting on the TS meter. However, a setting of 0.1 percent was decided upon to
reduce the burden on the wastewater treatment plants.
Only sludge is transported in the pipelines leading to the Center from the
Hokubu No. 1, Kanagawa, and Hokubu No. 2 plants. These pipelines are cleaned as
required.
171
-------
Sludge transportation management is based at the Center, and is a system of
centralized supervision and decentralized control.
4-1-1. Operation and control of sludge transportation facilities
The time required for the movement of sludge and the sludge transportation
time schedule are shown in Figures 4-1 and 4-2, respectively.
Maintenance and inspection of the various transportation facilities is the
responsibility of the Center.
Sludge transportation is carried out three times per day. As shown in Figure 4-
2, the transportation is initiated at prescribed times and terminated when the
prescribed quantity has been transported or when the level of sludge in the holding
tank reaches the prescribed quantity. In the event of an increase of sludge due to
storms or other developments, agreements are reached with personnel at the various
wastewater plants to adjust the aforementioned upper limits and increase the
quantity of sludge transported.
Fig. 4-1. Movement of Sludge from Transportation Facilities to the Center
tun j - ;
Hokubu
No.l
6,010m 7,680m
2 hr 40 min 2 hr 32 min
540-870 m3/day 990-1,280 m3/day
„
4,9E
23 1
1,7(
>0m
ir 27 min
)0~2,170
m3/day
Hokubu
, No. 2
1
160-660 m3/day
On-premise pipes
1 r
5,030m
9 hr 3 min
1,750-2,770 m3/day
\
Kanagawa
Legend
First line: section distance
Second line: transportation time
Third line: transportation quantity
Hokubu Sludge Treatment
Center intake facilities
172
-------
Fig/4-2. Sludge Transportation Time Schedule
Send from
Midori
W£
Intake at
Kohoku
Send from
Kohoku
\wm
Intake at
HokubuNo. 1
Send from
Hokubu No. 1
Intake at
Center
Send from
Kanagawa
Legend •• Sludge transportation
Cleaning
4-1-2. Management of sludge pipelines
The sludge transportation pipeline is basically composed of ductile castiron
pipe (DCIP) in sections that are buried and of fiber glass reinforced plastic mortar
(FRPM) pipe in sections that are installed in public utility ducts. As part of the
maintenance work, the operation of the pipeline's air valves, sludge discharge
valves, and other equipment is checked periodically four times a year.
However slight, pipeline leaks must be immediately detected and repaired
since the effluent is sludge. Consequently, all sections are equipped with flow
meters on both the sending and receiving sides. Leakage is detected by monitoring
and checking for variation between these meters. Similarly, the correlation between
the quantity sent and the discharge pressure is used to check for leakage and
blockage. At hourly intervals, the quantities sent and received are integrated to
check for leaks and to check that operation is proper. The Center also maintains a
store of materials for repairing leaks (e.g., various types of short pipe sections,
connecting bolts,.packing materials, and coupling rings).
173
-------
4-1-3. Operation of sludge intake facilities
During fiscal'year 1988, the quantity of sludge received at the Center ranged
from 4,110 to 5,290 m3 per day. The corresponding ranges for TS and VTS
concentration were 1.4-3.0 percent and 62-84 percent, respectively. The trends of
sludge intake quantity and TS and VTS concentrations are shown in Figure 4-3.
Fig. 4-3. Monthly Trend of Sludge Intake Quantity, and TS and VTS
Concentrations
I
Q)
ho
CO
100
90
80
70
60
50
40
30
20
10
0
4 5'6 7 8 91011121 2 3
Month
O tUD
I
55
Legend
Intake quantity
Sludge VTS
Sludge TS
As shown in Figure 4-3, the intake quantity gradually increases, and is in
inverse proportion to that of the TS concentration. This is presumably due to the fact
that the quantity transported is increased, in order to maintain the quality of the
treated wastewater at the plants when the sludge thickening properties deteriorate
due to meteorological changes. It can also be seen that the VTS concentration is
lowest during the summer (around August) and highest during the winter (around
January). This trend is thought to stem from sludge decay that accelerates with the
rise in water temperature.
4-1-4. Changes in sludge properties * •
As shown in Figure 4-1, the duration of time from the start of sludge
transportation to the start of treatment at the Center is longer than one day for the
Midori plant, which is the remotest plant. Table 4-2 indicates the change in sludge
properties during this time.
-------
Table 4-2. Changes in Sludge Properties durftig Transportati
on
TS
VTS
SS
DS
Total nitrogen
Ammonia nitrogen
Total phosphorus
Total soluble phosphorus
Volatile organic acid
Total quantity at the five
wastewater plants (t/day)
®
92.2
69.5
89.3
2,90
6.26
0.47
1.23
0.15
3.20
Intake quantity at Hokubu
Sludge Treatment Center (t/day)
©
92.0
67.2
73.6
18.4
5.34
0.63
1.16
. 0.28
5.34
Rate of change (%)
(®-®)/®X100
A 0.2
A3.3
A 27.6
534
A 14.7 , '
' 33.8
A 5. 7
84.0
66.9
The
In the course of transportation, the sludge decays and solids within it are
increasingly dissolved and changed into solubles. As a result, the suspended solids'
concentration declines and the dissolved solids concentration rises. Naturally, there
is also a rise in the amount of organic acid, an indicator of the degree of decay
Nitrogen and phosphorus also dissolve, leading to a rise in the concentration of
ammonia nitrogen and total soluble phosphorus.
With this progressive decay of the sludge over time, the thickening property
and characteristics of the. sludge deteriorate. For this reason, great care must be
taken in the operation and management of the centrifugal thickeners and other
equipment used in centralized sludge treatment.
4-2. Operation record of the Hokubu Sludge Treatment Center
4-2-1. Outline
Table 4-3 presents the operation record of the Hokubu Sludge Treatment
Center during fiscal year 1988. The entire quantity of sludge intake was subjected to
the sequential process of centrifugal thickening, digestion, dewatering and
incineration. The uses of the gas generated during digestion greatly depend on the
volume of gas generated and the operational status of the incineration facilities
During fiscal year 1988, about 30 percent of this gas was used as auxiliary fuel for
the incinerator; the remainder was used as fuel for the gas engine of the electrical
power generator.
175 .
-------
Table 4-3. Hokubu Sludge Treatment Center Operation Record (FY1988)
•
Sludge intake
Centrifugal thickeners
Digesters
Digested sludge
Sludge cake
Incineration
Incinerated ash
Digestion gas generated
Electrical power generated
Quantity treated "
1,606 X 103 m3/year
4,110-5,290 m3/D
(4,400)
1,793 X 103 m3/year
4,220-5,560 m3/D
(4,900)
560X103m3/year
1,100-5,650 m3/D
(1,530)
589Xl03m3/year
1,100-2,000 m3/D
(1,560)
76,826 tyear
130-270 t/D
(210)
83,280 t/year
170-360 t/D
(230)
15,270 t/year'
24-55 t/D
(41.8)
10,500 X103 Nm3/year
21400-37000 Nm3/D
(28,800)
13,362 kWh/year
23000-51800 kWh/D
(36,600)
TS (%)
1.4-3.0
(2.0)
1.6-2.3
(1.9)
4.2-6.2
(5.2)
2.6-4.0
(3.2)
18-24
(22)
."
.,"
VTS(%)
62-84
. (74)
64-83
(73)
63-84
(72)
60-65
(56)
50-67
(57)
"
Note:* The difference between the quantity of sludge cake and the quantity
incinerated reflects the transport of sludge cake from other
wastewater treatment plants for incineration. Figures in parentheses
indicate averages.
4-2-2. Operation of thickening facilities
As noted in section 4-1-4, the sludge received ^at the Center is in a state of
advanced decay, which would seriously lower the effectiveness of gravity-type
thickening. As a result, thickening is carried out employing a horizontal-type
continuous centrifugal thickener, after the sludge has been passed through a drum-
screen-type racks. The sludge TS concentration was in the range of 1.6-2.3 percent
before the thickening process and 4.2-6.2 percent after it. the screenings separated
out through the mechanically cleaned bar racks (amounting to 860 tons for the year)
were first,dewatered using a screw press and,then transported by pneumatic
conveyors to the incinerator for incineration.
Table 4-4 shows the quantity and TS and VTS concentrations of sludge treated
during fiscal year 1988. Figure 4-4 presents monthly changes in centrifugal
thickenr operations. " .
, 176
-------
Table 4-4. Quantity Treated and TS and VTS Concentrations (FY1988)
Intake sludge
Thickened sludge
Quantity treated per day
(m3)
4,220-5,560
(4,910)
1,100-1,900
(1,530)
TS
(%)
1.6-2.3
(1.9)
4.2-6.2
(5.2)
• VTS
,(%)
64-83
(73)
63-84
(72)
Fig. 4-4. Monthly Changes in Centrifugal Thickener Operations
o
iH
X
o
'55
0)
13
g? 2
4 5
8 9 10 11 12 1
Month
2 3
100
80
60
40
£
§•
u
to
3
'o
CO
Legend
Intake sludge TS
Thickened sludge TS
Intake sludge VTS
Thickened sludge VTS
Solid capture rate
Rate of dosing
; fC^ gal thickneners recorded an operation rate in the range of 64-84
percent of ful capacity. The centrifugal thickeners were operated mainly through
dosing (chemical feeding), and the target capture rate for SS was 90 percent In
plantrtneS1raCte ^*™ ****** ™ sent back to the wastewater treatment
plants the rate of dosing with polymer coagulants was controlled appropriately in
order to improve the properties of the effluent. The centrifugal effect" weir radius,
and differential speed were adjusted in accordance with the TS concentration of the
intake sludge The average values for centrifugal effect and differential speed were
i,Ulb (j and 17.8 rpm, respectively.
tfc • i am°Unt °f intake slud^e' but the quantity of
thickened sludge remained fairly constant. As shown in Figure 4-4, there was 1 ttle
rrduor^nth Ts;°ncrratir °f the intake siudge' but the Ws —^1
rose during the winter. Even when the rate of dosing with polymer coagulants was
increased, it was impossible to obtain a stable retrieval of solids during the winter
Ihe average solids capture rate for the year was about 83.3 percent. .
177
-------
4-2-3. Operation of digestion facilities
The Center contains egg-shaped digestion tanks, each with a capacity of 6,800
cubicmeters. The tanks are organized in series of three tanks^each D™****
year 1988 two such series (six tanks) were used for anaerobic digestion. All tanks
function as primary tanks. The operation specifications were as follows: a digestion
temperature of 35 degrees centigrade, an average holding time of 27 days, and an
a^mge ntakeTS concentration of 5.1 percent. For the operation method employed
^ere was high concentration and mesophilic digestion. The agitation was mecham-
X performed using a draft tube, and formed a downward flow within the tube. An
outline of the digesters and their operational conditions are shown in Table 4-5.
Table 4-5. Outline of Digesters' Operation Condition
Quantity of sludge
withdrawn
Quantity of sludge
recycled
Heating
Agitator
Conditions
45
2m3/min
Recycled sludge heated by
means of warm water; the
warm water circulation
pump is started at 34.5°C
and shut down at 36°C.
400 rpm
Comments
Automatic control within the
scope of liquid level change
)ontinuous 24-hour operations
External heating using a heat
exchanger (heat source: waste
heat from gas engine)
Continuous agitation in form of
a downward flow within the
tube; agitating intensity is
3,000 m3/hr (10 times tank
volume per day)
All tanks were operated as primary tanks because of the poor solids-liquid
separa on in tL secondary digesters. This is probably due to the fact that
settiab ity of digested sludge was low and transferring the sludge from the primary
Tthe secondary tanks, and with drawing the sludge out of the secondary tanks
Its so frequently thai there is not enough time for the sludge to « stil . And
also the difference L temperatures of the sludge in the primary tanks and that of the
secondary tanks is such that heat convection occurs.
The intake pump was used to inject the thickened sludge into the sludge
circulation pipe, where it was mixed with the recycled sludge. The average daily
178
-------
intake quantity per tank was about 256 m3, and the average quantity of sludge
withdrawn per tank was about 260 m3 per day. Scum generation has not been
detected since the start of operation in September 1987. The average pH values and
alkalinity within the digestion tanks were about 7.3 and 4,100 mg/1, respectively,
with little variation. . ' '
Figure 4-5,4-6 show the monthly trend of digester operation indicators.
Fig. 4-5. Monthly Trend of Digester Operation Indicators (I)
. 4,000 r
• 3,000
&
1
g-
0)
00
2,000
1,000
-1 1 ' I ' '
4 5 6 7 8 9 . 10 11 12 1 2 3
Month
Legend
. Intake sludge quantity
• Digested sludge quantity.
-Intake sludge TS
• Digested sludge TS
Fig. 4-6. Monthly Trend of Digester Operation Indicators (H)
100
•_g £ 80
Crf M
II
-*3 ctf
I a 60
2hfl
»
I "2
S §' 40
£> 20
O
' * . . i..
4 567 89 10 11 12 12 3
Month
40,000
30,000
I
1
10,000
0)
%
O
Legend
20,000 a
O
"43 '
Intake sludge VTS
• Digested sludge VTS
• Solids reduction rate *
Organic matter degradation rate
• Gas generation quantity
-------
As shown in Figure 4-5, 4-6, the trends of both the organic loading and the
intake sludge quantity- exhibit basically the same patterns of decrease in the
summer and increase in the winter. Virtually the same patterns characterizes the
trends of the VTS concentration of intake and withdrawn sludge. By contrast, the
pattern for the TS concentration of intake and withdrawn sludge is one of increase in
the summer and decrease in the winter—the inverse of the VTS concentration
pattern. The trends of the respective rates of solid reduction and organic matter
degradation resemble those of VTS concentration. The trend of the quantity of
digestion gas generated resembles that of organic loading, and indicates that a
sufficiently stable digestion was carried out at that load level. Since all of the tanks
were operated as primary tanks without supernatant runoff, the quantity of intake
thickened sludge was almost the same as that of withdrawn sludge.
Table 4-6 presents additional data for the operational record during fiscal year
1988.
Table 4-6. Digesters' Operation Record for FY1988
Item
Quantity of thickened sludge intake (m3/day)
TS (%)
VTS(%)
Quantity of digested sludge (m3/day)
TS(%)
VTS(%)
Digestion period (number of days)
Digestion temperature (°C)
Amount of heat used for heating (1,000 kcaVmS)
Volumetric loading (kgVTS/m3/day)
Solids reduction rate (%)
Digestion ratio (%)
Digestion gas generation quantity (Nm3/day)
Digestion gas generation rate (Nm3/kg-VTS)
Multiplier for digestion gas generation (Nm3/m3-day)
Operation record
1,100-1,900
(1,530)
4.2-6.2
(5.2)
63-84
(72)
1,100-2,000
(1,560)
2.6-4.0
- (3.2)
50-65
(56)
21-36
(27)
35.1-35.3
(35.1)
9.3-28.7
(18.2)
0.96-1.9
(1.4)
28-45 '
(37)
39-61
(51)
21,400-37,000
(28,800)
0.3-0.6
(0.5)
16.7-21.0
(18.7)
Note: Figures in parentheses indicate yearly averages.
180
-------
Table 4-7 shows the composition of the digestion gas. Hydrogen sulfide
.contained m the digestion gas was removed using a wet-type desulfurization unit in
order to enable using it as a source of heat for the incinerator and a source of power
tor the gas engine-driven power, generator. The hydrogen sulfide concentration
averaged 148 (30 - 600) ppm at the unit inlet and 0.9 (0 -12.5) ppm at the unit outlet.
Table 4-7. Concentration of Digestion Gas Components
Constituent components
Methane
Carbon dioxide •
Hydrogen sulfide
Moisture
Concentration (v/v%)
59-62
34-37
30-600 ppm
2.3-3.2
4-2-4. Operation of gas engine power generation facilities
After the digestion gas was removed from the gas holder, a gas booster was
used to raise its pressure to about 1 kilogram-force per square centimeter. In-this
form, the gas was used as fuel for the gas engines driving power generators. The
engines are water-cooled, 12-cylinder, boiling coolant gas models with a horsepower
of 1,350 ps. There are directly connected to the generators, which are triphase
current synchronized models (920 kW, 50 Hz, 6.3 kV). The generators constantly
operated parallel with commercial power sources, and supplied a portion of the
Center's power needs. The number of generators in operation were regulated in
accordance with the quantity of digestion gas in the holding tank. The quantity of
digestion gas used for power generation and of the generated power are shown in
Figure 4-7.
Fig. 4-7. Monthly Trend of the Quantity of Digestion Gas Used for Power
Generation and of the Generated Power
Legend
§j|^ Generated electric power
E_J Purchased electric power
50,000
1
§ 40,000
Quantity of Electric Power (kV
1- • 10 CO
p p o
o "o "o
_ o o o
o o o o
/
[
i
nn
tS
'•
%
s
•
f
y-f
-,:
T"
!
••'
* & 6 7 8 9 10 11
>
12
:".
v
1
•m
I
^
2
7"
*
^
3
Month
181
-------
The daily average for the quantity of digestion gas used for power generation
was 21 300 Nm3, of the power so generated, the daily average was 43,000 kWh, and
of the power used for the auxiliary generator unit, this was 6,400 kWh. The effective
quantity of power generated (excluding the power required for operation of the
auxiliary'unit) averaged 36,600 kWh per day. This figure represents about 64
percent of the power used within the Center during the year in question.
The lower calorific value of the digestion gas was 5,500 kcal/Nm3. The rate of
power generated was 31.5 percent of the amount of heat supplied to the gas engine
(26.8 percent when the power consumed by the auxiliary unit is deducted). The
generated power per digestion gas unit was 2.02 kWh/Nm3.
The waste heat from the generator (contained in the cooling water for the
engine and in exhaust gas) was retrieved in the form of pressurized steam using a
heat exchanger (the steam temperature averaged 121 degrees centigrade). The
quantity of heat retrieved was equivalent to about 35 percent of that supplied to the
gas engine. The pressurized steam was used as a source of heat for warming the
digesters and for air conditioning. The quantity of heat used for. heating the
digesters ranged from 12,000 to 51,000 Meal (averaged 28,000 Meal) per day; the
average utility rate was equivalent to 23.9 percent of the amount of calories supplied
to the gas engine. The calorific value for heating the digesters per unit volume of
intake sludge, varied greatly depending on the season. The respective maximum
and minimum values per cubic meter were 28.7 Meal (during March) and 9.3 Meal
(during August); the average value was 17.7 Meal.
4-2-5. Operation of dewatering facilities
After digestion, the sludge was dewatered using a horizontal-type continuous
centrifugal dehydrator unit. The TS concentration of the digested sludge before
dewatering ranged from 2.5 to 4.2 percent. The moisture content of the sludge cake
ranged from 76 to 82 percent. Table 4-8 shows 'the operational record of the
dewatering process and Figure 4-8, 4-9 shows the monthly trend of dewatering
operation indicators.
Table 4-8. Operational Record of Dewatering Processes
Digested sludge
Sludge cake
Daily handling
quantity
1,100-2,000 m3/day
(1,600)
130-270 t/day
(.210)
TS/
moisture content*
(%)
2.6-4.0
(3.2)
76-82*
(78)
VTS
(%)
50-65
(56) .
50-67
(57)
Note- Figures in parentheses indicate yearly averages.
182
-------
Fig. 4-8. Monthly Changes in Dewatering Facilities Operations (I)
o
1-1
X
'£»
11
a
p.-
I
'43 -w
rt rt
I!
0> O
. bo 4)
T3 >-<
r2 -3
M 03
'
Fig. 4-8.
200
150
100
50
0' ' 1
Legend
bo
-s
-
o
0)
O
4 5 6 7 8 9 10 11 12 1 2 3 °
Month
Digested sludge quantity
Sludge cake quantity
Sludge cake moisture content
Solids capture rate
Rate of dosing
Fig. 4-9. Monthly Changes in Dewatering Facilities Operations (II)
3
s<
•3
o
02
100
90
80
70
60
50
40
30
20
10
-i 1—__i 1 i i
4 5 6 7 8 9 10 11 12 1 2 3
Month
Legend
s>
3 -a
•3
Sw
O
$'
03
2 tf
B
Digested sludge TS
Digested sludge VTS
Sludge cake VTS
Solids capture rate
Rate of dosing
183
-------
As shown in Figure 4-8, 4-9, there is a basic correlation between the trend of
the intake quantity of digested sludge and that of the sludge cake quantity; both
exhibit a decline in summer. By contrast, the trend of TS concentration of digested
sludge declines in winter. However, the trends of the VTS concentration of digested
sludge and sludge cake both show an increase in winter.
This decline in TS concentration and rise in VTS concentration during winter
is presumed to be the cause of the rise in the sludge cake moisture content and
decline in the solids retrieval rate during the same season. There is a close correla-
tion between the trend of the sludge cake moisture content and that of the. digested
sludge VTS concentration. When the VTS concentration was high, it proved difficult
to maintain the moisture contents on a constant level even when the dosage was
increased.
The annual average values for the solids capture rate and of dosage were about
90 percent and 1.06 percent per solid, respectively.
4-2-6. Operation of incineration facilities
The sludge cake generated at the Center was incinerated along with the sludge
cake transported from the city's wastewater treatment plants. Incineration facilities
consist of the incinerator proper (an erect hollow cylindrical structure), a sludge cake
feeder, cake drier, heat recovery unit, and an exhaust gas treatment unit. The
incinerators proper-fluidized bed furnaces-are made of fire-resistant brick and
castable materials, which are lined on the outside with a steel sheet shell.
The ash resulting from incineration is composed of small particles with a
diameter ranging from about 50 to 300 microns. The ash was removed from the
incinerator along with the combustion gas and collected by cyclone separators and
electrostatic precipitators. The collected ash was moisturized and then disposed of in
industrial waste disposal sites within the city.
The quantity of sludge cake fed into the incinerator was maintained ori a
constant level. The incinerator was supplied with additional heat using auxiliary
fuels in order to assure a temperature of about 800°C on top of the fluidized bed. The
incinerator construction proceeded in a sequence (Nos. 1 through 3), and the thermal
efficiency rises in correspondance with the newness of the incinerator in question,
due to the advanced heat recovery. Table 4-9 shows the respective modes of incinera-
. tion, heat recovery, and exhaust gas treatment adopted for each.
184
-------
Table 4-9. Modes of Incineration, Heat Recovery, and Exhaust Gas Treatment for
Each Incinerator
Mode of
incineration
Auxiliary fuel
Heat recovery
Mode of exhaust
gas treatment
SO* treatment
NO* treatment
Smoke and dust
reatment
Fluidized bed incinerator
Pulverized coal, kerosene
Heating of air for fluidization,
waste heat boiler
3aCO3-based dry desulfurization
iontrol of internal temperature
and combustion air ratio
Cyclone separator, electrostatic
>recipitator
Fluidized bed incinerator
"" —
Pulverized coal, kerosene,
digestion gas
Heating of air for fluidization,
waste heat boiler
Heating of air for fluidization,
heating of heat medium oil used
1aCO3-based dry desulfurization
Control of internal temperature
and combustion air ratio
lyclone separator, electrostatic
>recipitator
Fluidized bed incinerator (with
drier)
Digestion gas, city gas
NaOH-based wet desulfurization
:ontrol of internal temperature
and .combustion air ratio
Cyclone separator, electrostatic
irecipitator
1 «n f mTTIT1 °peration centered around the No. 3 incinerator (capacity of
150 tons per day) offers the best thermal efficiency. Table 4-10 and Figure 410 show
the record of operation and monthly trend of the quantity of sludge cake incinerated,
^
Table 4-10. Incinerator Operation Record (FY1988)
—
No. 1 incinerator
No. 2 incinerator
No. 3 incinerator
Total
Number of
days in
operation
151
259
297
— ._
Annual quantity of
sludge cake
incinerated
14,510 t/year
24,850 t/year
43,820 t/year
—
83,180 t/year
Annual quantity of
incinerated ash (with
30% moisturization)
3,430 t/year
5,110 t/year
6,730 t/year
15,270 t/year
185
-------
Fig. 4-10. Monthly Trend of Quantity of Sludge Cake Incinerated
1
I
I
faD
Legend .
No. 1 incinerator
HI No. 2 incinerator
No. 3 incinerator
Month
The incinerators were operated in accordance with quarterly plans formulated
on the basis of estimates of the sludge cake quantity to be generated during the
quarter in question. This enabled each incinerator to operate continuously for at
least one month. Repair work was scheduled to coincide with the legally stipulated
boiler inspection (once annually) and conducted intensively along with preventive
maintenance work. The repair periods, each lasting about one month, were May-
June for the No. 1 incinerator, February for the No,. 2 incinerator, and November-
December for the No. 3 incinerator.
4-3. Treatment of sidestreams from the sludge treatment facilities
4-3-4. Properties of sidestreams
Figure 4-11 shows the sources of sidestreams out in the sludge treatment
process, and Table 4-11, the related quantities and quality.
186
-------
Fig. 4-11 Sources of Sidestreams
Intake
sludge
J fn
'hickening Centrifugal
4-n-nl* I ^1.1- • ^ —:
tank
- KanagawaWTP
I
I
Digester
Centrifugal Incinerator
Thickner effluent
Sidestreams receiving tank
(4)
denydrat
1
D
j
Dr"
Filtrate ©
? tank
,— *
\
Scrul
[ Hokubu No. 2 WTP
Table 4-11. Quantities and Quality of Sidestreams by Source (actual, FY1988)
OD-Mn (mg/1)
••
BOD (mg/1)
Note: The quantity for number three Was calculated as ® - (® +©).
187
-------
The sidestream sources are centrifugal thickeners, centrifugal dehydrators,
scrubber drains and others. During fiscal year 1988, the first three sources
accounted for 91 percent of the entire quantity of sidestreams. The single greatest
source was thickener effluent, which accounted for about 45 P^* a*^?£X the
can also be noted that the quantity of the central** relatively .high compared to the
conventional-type sludge treatment system. This is due to the fact that the Center
does not remove supernatant during digestion, meaning that the entire quantity
undergoes the dewatering process.
As shown in Table 4-11, the concentrations of the thickener effluent are
extremely high in virtually every category of water quality Regging ritrogen,
however, the centrate concentration (SBOmg/1) is higher than that of me tmcKener
effluent (360 mg/1). This is because the nitrogen compound is decomposed during
digestion and ammonia is present in the centrate.
' Table 4-12 presents the calculation results (based on the above data) of the
sidestreams loading for each category. In virtually all categories, the percentage of
the total load occupied by thickener effluent is extremely high, from 75 to 90 percent.
As these percentages suggest, this is the source of the bulk of the sidestreams
load. An exception is in relation to nitrogen, where the centrate accounts fl>r just over
50 percent of the entire load. Since the thickener effluent accounts for another 40
percent these two taken together account for over 90 percent of the total load m this
category. By contrast, scrubber drains account for only a few percentage points of
the entire load in every category, with the exception of TS.
Table 4-12. Composition of Sidestreams' Loads
Classification
Sidestreams
TS
VTS
SS
COD-Mn
BOD
Nitrogen
Phospho
rus
Thickened effluent
(t/day)
13.5
9.6
6.9
4.1
9.0
1.2
0.26
(%)
75
91
87
85
84
• 40
76
Centrate
(t/day)
1.8
0.63
0.12
0.24
0.11
1.7
0.07
(%)
10
6
2
5
1
53
21
Other processes
(t/day)
0.2
0.01
0.9
0.4
1.6
0.1
0.01
(%)
2
1
11
9
15
3
3
Scrubber drains
(t/day)
2.4
0.23
0.01
0.06
0.03
0.13
0
(%)'
13
2
0
1
0
4
0
Total
(t/day)
17.9
10.5
7.9
4.8
10.7
3.0
0.34
(%)
100
100
100
100
100
100
100
For the time being, the Center has been sending the sidestreams (thickener
effluent filtrate, and miscellaneous drainage) to three wastewater plants (Hokubu
No.l, Hokubu No. 2, and Kanagawa), where they are mixed with wastewater
188
-------
influent and treated. Since scrubber drains account for only a small load (as noted
above) they are not distributed among these three plants. Instead, it is sent
separately from the other fluids to the Hokubu No. 2 plant (see Figure 4-11). The
distribution ratio of the sidestreams among the three plants in the aforementioned
order was 3:1:6.
Although the system is only temporary, the Center is currently sending side-
streams out in the course of sludge treatment to the wastewater treatment plants. It
strictly controls the quality of the separated fluid, analyzing it at least threejimes
per week. A major focus in water quality management is proper operation of the
centrifugal thickener, since thickener effluent accounts for the bulk of the
sidestream load. \
4-3-2. Impact of sidestreams on wastewater treatment
(1) Treatment properties of sidestreams
In order to ascertain the treatment properties of sidestreams, the Center
conducted gravitational sedimentation tests and biodegradability tests on it.
Concerning SS, COD, and BOD, the removal rate by gravitational sedimentation
was low (below 35 percent). Since the removal rate of SS and BOD in primary
treatment of ordinary influent is generally in excess of 50 percent, this suggests that
the sidestreams contained a large quantity of substances (e.g., coloids) that have low
settleability.
Similarly, the biodegradability of sidestreams was also found to be low. The
degree of biodegrada'tion over a 24-hour period in the case of ordinary municipal
wastewater was in excess of 90 percent for BOD and 70 percent for COD. In the case
of sidestreams, however, the respective figures were 61 percent for BOD and 43
percent for COD after four hours, and 81 and 43 percent .after twenty-four hours.
These figures also indicate the difficulty of treating sidestreams.
(2) Impact of sidestreams on waste water treatment
Table 4-13 shows the share of the total wastewater inflow for the three
wastewater treatment plants receiving sidestreams from the Center (actual shares
for fiscal year 1988). . .
Although sidestreams account for only 1.3 percent of the total quantity of water
treated at the plants, they are the source of an extremely high share of the SS loads
(13 percent), BOD (18 percent), and nitrogen (24 percent). Regarding quality,
sidestreams contain many ingredients that cannot easily be removed through
sedimentation or activated sludge treatment.
189
-------
Table 4-13. Sidestreams' Share of Total Loads
Water quantity
(m3/day)
SS(t/day)
BOD (t/day)
COD (t/day)
Nitrogen (t/day)
Phosphorus (t/day)
Sidestreams
®
5,730
7.5
11
4.3
2.9
0.43
Total influent at
theSWTPs©
442,000
57
61
34
12
1.6
Share
(®-e-@)X100
1.3
13
18 •
13
24
27
Table 4-14 supplies estimates of the extent to which water quality
improvement would be made at the three plants now receiving the sidestreams if
they were no longer receiving this.
In terms of the average value for the three plants, sidestreams account for an
estimated 42 percent of the BOD, 42 percent of the COD, and 34 percent of the
nitrogen of the respective secondary effluent load. When this amount is added, the
effect on BOD is even greater. Furthermore, the figures are estimates based on
annual averages; the sidestreams' impact would presumably increase if the
treatability declines due to the decreased water temperature during winter or the
deterioration state of activated sludge.
At present, no particular problems are experienced at the plants, in. part,
because the pumping time and distribution ratio among the three plants are
adjusted in accordance with the treatment situation at each plant.
««
Table 4-14. Impact of Sidestreams on Waste water Treatment Plants
Item
BOD
COD (Mn)
Nitrogen
WTP
Hokubu No, 1
Hokubu No. 2
Kanagawa
Average
Hokubu No. 1
Hokubu No. 2
Kanagawa
Average
Hokubu No. 1
Hokubu No. 2
Kanagawa
Average
Treated water
• quality
®
12
12
19
16
11
11
12
12
15
16
17
16
Sidestreams' shares of
treated water
©
7.4
5.0
6.7
6.7
5.6
3.8
5.1
5.1
6.1
4.1
5.5
5.5
Separated fluids
. impact (%)
@/®XlOO
62 /
42
35
42
51
34 ,
42
42
41
26
32
34
Note: (1) = actual FY1988 figures: average figures are weighted averages.
190
-------
4-4. Environmental pollution control
4-4-1. Background and regulations pertaining to pollution control
Japan's fast-paced economic development during the postwar period was
accompanied by an ever worsening environmental pollution resulting from the smoke
and wastewater discharged by factories and other sources. The pollution eventually
began to affect local communities, stimulating campaigns aimed at winning
compensation through litigation for those whose health had been affected by it.
Structuring a regulatory framework to prevent such health problems began
with the enactment of the Basdic Law for Environmental Pollution Control in 1967.
This law was followed in 1968 by the Air Pollution Control Law (amended most
recently in 1974) and by regulations pertaining to the enforcement of the same law,
and then in 1978 by pollution control bylaws enacted by Kanagawa Prefecture. For
its part, in 1975, the City of Yokohama devised general plans for administrative
guideance to curtail the release of SOX, soot and dust and in 1977 of NOX.
Several laws related to aid for pollution victims were also enacted. The
Pollution-related Health Problem Compensation Law, which was enacted in October
1973, has played a major role in providing aid for those suffering with health
problems stemming from pollution. Revised in September 1987, when it was
retitled: Law Concerning Compensation and Other Matters Pertaining to Pollution-
related Health Problems. This law provides for the cancellation of so-called first-
class regions in the event of violation and allows for projects to be carried out
pertaining to the prevention of health hazards, funded by endowments created for
that purpose. In accordance with this law, the Center pays pollution load charges as
well, commensurate with its total quantity of SO* discharged.
In all, discharge of soot and dust, SO* , and NO* are regulated by the Air
Pollution Control Law and pollution control bylaws. The Air Pollution Control Law
contains specific regulations concerning the permissible concentrations of SO* and
NO*. The total quantity of such pollutants is regulated by this law and by pollution
prevention bylaws and regulations.
5 "
Industrial waste is regulated by the Waste Disposal and Public Cleansing Law.
The main sources of industrial waste generated by the Center are incinerated ash,
ash removed from the cyclone separators and electrostatic precipitators.
The Center's modes of treatment for the substances covered by antipollution
regulations are as follows: >. • '
Soot and dust: Removal by cyclone separators and electrostatic
precipitators
191
-------
SO*:
Desulfurization in exhaust gas treatment units (removal of
S0;e by showering with a caustic soda solvent)
NO*:
Curtailment of generation through low-oxygen combustion
Incinerated ash: Disposal in inland landfill sites after boosting moisture
content to 30 percent, to prevent scattering
Subrubber drain: Treatment with influent wastewater at wastewater
treatment plants
Figure 4-12 shows the flow sheet of Exhaust Gas Treatment Facilities.
Fig. 4-12. Outline of the Flow of Exhaust Gas Treatment Facilities
Incinerator
re
Waste heat | _^
cdvery facilities
Curtailment of
Cyclone
separator
Desooting
_>Electrostatic_>| Smoke | ^
precipitator treatment unit
•Chimney
Removal of SO*
generation
4-4-2. Discharge of SO*, NO*, and soot and dust
Table 4-15 shows the pollutant concentrations and exhaust gas quantities
generated by each incinerator during fiscal year 1988.
Table 4-15. Concentration of Pollutants Within Incinerator Exhaust Gas
Place of measurement
EP outlet of No. 1
incinerator
EP outlet No.2
incinerator
Treatment unit outlet
of No. 3 incinerator
SO.x concentration
(ppm)
19-39
<5~41
<15~19
NOX concentration
, (pPm)
16-35
34-56
<10~15
Soot concentration
(g/Nm3)
0.018-0.039
0.018-0.039
0.001-0.002
Exhaust gas quantity
(1000 Nm3/hr)
15-17.7
9.2-15.7
10.5-12.5
Note: The 02 conversion factor for the SO* and NO* concentration was 12 percent. "
As noted in the previous section, SOX and NOX are regulated in terms of the
entire quantity, and soot and dust are regulated in terms of the discharge
concentration. The regulation standards and the maximum discharge levels
recorded during the year in question are shown in Table 4-16.
192
-------
Table 4-16 Pollutant Regulation Standards and Center Discharge Levels
Maximum quantity of SOX
No. 1 incinerator
No. 2 incinerator
No. 3 incinerator
Total.
discharge (Nm3/hr)
Note: Soot and dust discharge is regulated with reference to concentration.
Maximum quantity of
NO* discharge (Nm3/hr)
Soot and dust
concentration
(g/Nm3)
4-4-3. Odor countermeasures
Regulatory standards concerning offensive odors as contained in the national
Offensive Odor Control Law and in the City of Yokohama's regulation '
permissible concentrations (Ppm) for eight types of malodorous Lbrt
are also odor concentration standards contained in the city's provisional
tive guidelines concerning malodor countermeasures. The "odor c^nt s
based on the odor perceptible .by the human olfacotry sense. It is expressed as a
" to a
Table 4-17 shows the regulatory standards for the eight types of malodorous
substances on the boundary line of the Center, which is located in lT±tu2ve
dTff " ^ Pr°SPeCtS ^ 10WerIng the o^r eoncentrat"rugh
1 r°? T uncertain> the actu^l deodorization unit design aims for a
at the odor discharge outlet that is lower than the standards at the
boundary, for both material and odor concentration. ^ne.
The main malodorous substances of concern at the Center are four
related-substances: hydrogen sulfide, methyl mercaptan, methyl suS
"tf yTr s seh Tht re:rctive;oncentrations vary ^^ ™**^™
site This is shown by the readings for one of the most typical malodorous
substances-hydrogen sulfide. The readings (material and odor coc
respectively) were 3 ppm and 10,000- 40,000 at the sludge intake a
thickening tank, and 40 ppm and about 100,000 at the centrifugal thickener
and
193
-------
Table 4-17. Malodorous Substance Standards and Actual Data
Malodorous substances
Hydrogen sulfide (ppm)
Methyl mercaptan
(ppm)
Methyl sulfide (ppm)
Methyl disulfide (ppm)
Ammonia (ppm)
Trimethylamine (ppm)
Acetaldehyde (ppm)
Styrene (ppm)
Odor concentration
Note: Figures with an as
to nlrs and other de
Regulatory
standards (at
premise boundary)
0.02
0.002
0:01
0.009
1
0.005
0.05
0.4
32
Base odor level*
2.06
2.554
0.145
0.068
0.4
— — •
23700
Odor level at outlet of
deodorization ,
facilities* .
" 0.0003
0.0049
0.0341
0.025
0
208
terisk indicate an average of four readings taken at sludge thicfcenmg
odorization facilities during fiscal years 1986 and 198 /.
Efforts are made to completely seal the sources generating offensive odors and
steps are taken to reduce the deodorization airflow. In addition leakage of odors
from such facilities is prevented by absorbtion. of odors and by maintaining a
negative pressure within them! The deodorization airflow varies depending on the
location; facilities are ventilated anywhere from once to ten times per hour.
The centralized treatment of sludge has enabled resolution of the problem of
odor transmission to communities in the vicinity of wastewater treatment plants It
has also enabled centralized treatment of the generated odors The Center s mam
mode of deodorization is soil deodorization, which also provides an effetive use of
natural tracts. It is incorporating countermeasures for odors with a high concentra-
tion (50,000 or more) in its plans for pretreatment facilities.
194
-------
5. Effects of centralized treatment
5-1. Effective use of energy
There are various methods of sludge treatment, ea.ch with their own
advantages and disadvantages. The type of process adopted will depend on the
disposal method and the purpose. This section presents an account of the effects of
the sludge treatment process adopted at the Center, based on actual operations for
fiscal year 1988, focusing on the use of energy, the energy balance, and on
comparisons with other plants.
5-1-1. Energy use '
One of the major objectives of tne process adopted by the Center involved both
the recovery and effective use of the latent energy harbored by sludge The unit
processes which make up centralized treatment are organically integrated during
implementation. The overall flow of energy use and the proportions occupied by each
process are shown in Figure 5-1 and Table 5-1, respectively. The processes in
question are all those intermediate to and included in the sludge intake and
' incineration processes.
As shown in Table 5-1, the use of energy by process varies considerably
depending on whether the focus is electrical power or fuel. As such, figures for
electrical power were converted into calories to indicate the total quantity of energy
use. It can be seen that sludge transportation, centrifugal thickening and
dewatermg each account for less than 2 percent of this total. By contrast the
respective percentages occupied by digestion and incineration are roughly 5- and 30-
fold that level. Incineration fuel accounts for about 50 percent of the total quantity
of energy use.
Table 5-1 also takes into account the energy recovered as digestion gas in the
centrifugal thickening, digestion, and gas engine generation processes, which form
the nucleus of the entire treatment process. These three processes can basically be
.regarded as yielding a positive net energy. Collectively, these three processes
generated about 59,000 Meal per day of surplus energy, which can be effectively
utilized for other uses. ' . . *
195
-------
Fig. 5-1. Overall Flow of Energy Use
, Sludge-—,
transportation
I Sludge sending t
Power supply
I
Sludge intake
-.Centrifugal-,
thickening 1
^
Thickene
sludge
Dewateringunit|
Commercial power source
Digestion gas.
• Sludge
> Digestion gas
• Electric power
and other
Gas engine
power generation
Purchased fuel
Sludge cake
Table 5-1. Allotment of Energy Use by Process
Sludge
transportation
Centrifugal
thickening
Digestion
Gas engine *
power generation
Dewatering
Incineration
Total
Electrical power (kWh/d)
©Quantity
used
6,503
(9.8%)
8,432
(12.7%)
7,063
(10.6%)
6',352
(9.6%)
7,855
(11.8%)
30,207
(45.5%)
66,412
(100%)
©Quantity used
—quantity generated!)
6,503
8,432,
7,063
-36,632
, 7,855
30,207
23,428
Fuel (Mcal/d)
©Quantity
used
0
(0%)
0
(0%)
28,079
(7.9%)
117,282
(33.2%)
0
(0%)
207,992
(58.9%)
353,353
(100%)
©Quantity used
—quantity generated!)
0
0
-130,123
89,203
0
207,992
167,072
Quantity of energy use (Mcal/d)
® +®
5,593
(1.4%)
7,252 .
(1.8%)
34,153
(8.3%)
122,745
(29.9%)
6,755
(1.6%)
233,970
(57.0%)
410,468
(100%)
©+® •
5,593
7,252
-124,049
57,699
6,755
233,970
187,220
tity generated indicate the difference after the deduction of electrical
5 generated by each process from the amount of electrical and thermal Uuel)
energy (or equivalent) which each used.
2. Figures in parentheses indicate percentages of total.
196
-------
5-1-2.
Unit process energy balance and unit energy consumption
„ K^^vtgt, it is possible to reduce the
^s:i:sj»~~?=H?F=d"-"
Table 5-2. Units of Treatment Energy Use in Each Process
I • —-
[Sludge transportation
f "
Centrifugal thickening
Digestii
Gas engine power
I generation 2)
~ .
Dewatering
— ——
Incineration (No. 3) 3)
Treated sludge
quantity
DSt/d
Quantity of electrical
power use
kWh/d
Quantity of fuel
use
Mcal/d
Consumption
units of electrical
power
kWh/DSt
Consumption
units of thermal
power
Mcal/DSt
92^0 KK/VJ 1 ' — I" 1— .
92.0
78.0
(21,324)
50.0
33.3
8,432
7,063
(36,632)
7,855
15,967
28,079 1)
Total: 24,591
Gas: 21,283
70.7
91.7
90.6
(1.72)
157
479
360
1: • 738
3as: 639
qu""ly °f
-------
"owmraton fedlities Therefore, these three processes must always funcbon
as a unit.
The energy balances of the digesters, gas engine generation facilities, and
incineration facilities are described below.
(1) Energy balance of digesters
Figure 5-2 shows the energy balance of the digesters.
Fig. 5-2. Energy Balance of Digesters
Quantity of heat used for warming .
28,079 (Mcayd) Electrical power use quantity
(8.6%)
7,063 (kWh/d)
6,074 (Mcal/d)
(1.9%)
1,
290,500(Mcayd)(89.5%)
. _ .V4.. v
(Digestion tank facilities)
Digestion gas generated
158,S
(48.7%)
(35°C)
161 ,025 (Mcayd)
(49.6%)
Sludge : 5, 000 Meal /VS- ton
Digestion gas : 5,500 kcayNmS
Loss 5j426 (Mcal/d) (1.7%)
shown in Figure 5-
half of the intake sludge calories are
used for warming).
(2) Energy balance of gas engine power generation facilities
power generation facilities.
r '
198
-------
Table 5-3. Energy Use in Power Generation Facilities
Year/
month
'88/4
'88/5
'88/6
'88/7
'88/8
'88/9
'88/10
'88/11
'88/12
'89/1
'89/2
'89/3
Average
Note: Ilnrf
Quantity of
digestion gas
generated
Nm3/d
32,565
31,810
28,013
25,262
23,352
23,313
24,923
29,440
31,140
31,510
32,210
31,530
28,764
(100%)
Digestion gas use quantity
Incinerator
Nm3/d
9,919
11,090
7,675
3,464
4,021
9,265
7,426
7,480
7,720
10,060
3,180
7,980
7,440
<24.9%)
Gas engine generation
facilities
Nm3/d.
22,647
20,720
20,338
21,798
19,331
14,049
17,497
21,960
23,420
21,450
29,030
23,650
.21,324
(74.1%)
Mcal/d
124,557
113,960
111,861
119,889
106,320
77,268
96,234
120,780
128,810
117,975
159,665
130,075
117,283
Gas engine generation facilities
® Quantity
of power
generation
kWh/d
45,694
42,059
41,124
44,175
37,809
28,000
34,619
43,920
47,630
43,610
59,370
47,790
42,984
© Quantity
of power used
kWh/d
6,343
6,093
6,287
6,583
6,871
6,070
5,835
6,410
6,580
5,980
7,560
6,610
6,352
®-©l)
kWh/d
39,351
35,966
34,837
37,592
30,938
2,930
28,784
37,510'
41,050
37,630
51,810
41,180
36,632
Quantity of
heat used for
warming
digesters
Mcal/d
31,187
22,794
17,765
16,488
12,035
17,264
19,591
30,210
36,450
42,885
39,696
50,583
28,079
A major factor in the fluctuation of the digestion gas quantity generated
reflects the change in the quantity of organic substances contained in the thickened
sludge. For the year in question, this latter quantity ranged from a maximum of
about 70 tons (in February) to a minimum of about 50 tons (in September) and
averaged about 58 tons. There was a close correlation between this quantity and
that of the dige'stion gas.
Figure 5-3 shows the balance of energy in the gas engine power generation
facilities.
Fig. 5-3. Energy Balance of Gas Engine Power Generation Facilities
Effective quantity of power generated
Power consumed by auxiliary equipment ^--~\ 36,632 (kWh/day)
Krt(iW^ay!, ^-. ^1 r 31>504 (Meal/day) (26.9%)
5,463(Meal/day) (4 &%\/^_
D
__ Total power output
Quantity of digestion gas use 42,983 (kWh/day)
(100%)
(Gas engine generation facilities).
Heating of digesters
28,079 (Mcal/d) (23.9%)
Air-conditioning and heating
12,916 (Mcal/d) (11.0%)
?.
[uantity not retrieved
4,128 (Mcal/d) (12.0%)
Loss 30,652(Mcal/d) (25.9%)
199
-------
The generative efficiency obtained from the effective quantity of power
generated amounted to 26.9 percent, but the efficiency of the generator itself was
31 5 percent. The digester's heating needs were met using about one-fourth of the
digestion gas supplied to the generation facilities. A certain quantity of energy
(expressed in terms of calories in the figure) was not retrieved, only because an
effective use for this energy was lacking. However, this quantity can be effectively
utilized.
(3) Energy balance of incineration facilities
As noted in previous sections, the Center operates three incinerators. The first
two (No. 1 and No. 2) began operation before the digesters did, The No. 2 incinerator
can use both digestion gas and other types of .auxiliary fuel, but the No. 1
incinerator, which is the oldest, cannot use digestion gas. By contrast, the No. 3
incinerator, which is the newest, was basically designed to use digestion gas.
Table 5-4 shows the specific differences among the incinerators regarding
energy use. „
Table 5-4. Energy Balance of Incinerators
Incinerator
(No.)
No.l
No. 2
No. 3
Sludge incinerated
quantity ®
(DSt/d)
22.7
21.3
33.3
Incinerated
ash quantity
(DSt/d)
17.5
15.2 -
17.4
Quantity of energy use
Electrical
power
(kWh/d)
6,191
8,048
15,967
Auxiliary fuel (Mcal/d)
Digestion
gas
0
28,986
21,283
Other auxiliary
fuels
— . — -
62,480
. 49,496
3,308
Subtotal
62,480
78,482
24,591
© Total
(Mcal/d)
67,804
85,403
38,322
®/®
(Mcal/DSt)
2,987
4,010
1,151
As shown in Table 5-4, the No. 3 incinerator has the lowest energy
consumption units per DS tons, and digestion gas can be used to meet about, 87
percent of its auxiliary fuel needs. In view of this advantage, which suggest that
incinerators of the No. 3 type will constitute the main mode of incineration in future
centralized sludge treatment, it was decided to base the consumption units for
incineration on the values for the No. 3 incinerator for the following discussion.
5-1-3. Energy balance comparison with other processes
The aforementioned findings provide the basis for Figure 5-4, which shows the
energy balance per unit of solids received for treatment at the Center. For the
purposes of comparison, calculations were also made of the energy balance in the
conventional digestion/incineration process in which a boiler is used to warm the
tank and effect separation of supernatant and calculations were also made in raw
sludge dewatering/incineration processes. In these calculation, the values applied
for the various units are based on the actual figures obtained in the Center s
operation, with the exception of the units for digesters in the conventional
digestion/incineration process, which are nationwide averages for fiscal year 1986.
200
-------
*"
the three above-'
Fig. 5-4. Energy Balance per Unit of Treated Solids at the Center
(No. 3 Incinerator)
Fuel-
.50Mcal
Commercial electric power
134 kWh
(24%)
•* Fuel and other
< • Sludge
-3-=- Electric power
•1 Digestion gas
Air-conditioning and
beatingsystems
562 Meal
Loss
Table 5-5. Energy Balance Comparison
U) Digestion/incineration
© Conventional
process in Yokohama
digestion/incineration
dewatering/incinera
(Centralized treatment)
Treated sludge quantity
Total energy use
Electrical power
Fuel
kWh
Meal
Total (in terms of calories) |Mcal
Outside energy introduced
Electrical power
Fuel
Total (in terms of calories)
,
201
-------
The figures for total energy use reveal that processes (1) and (2) require about
the sa^ne amount of energy for treatment (expressed in terms of calones),but ti£
nrocess (3) requires only about one-fourth the energy that processes (1) and (2) do. In
Sis respect process (3) is the most advantageous. (The calculate for process (3)
SsuSeralgeneouscombustionbacauseitexcludedthedigestaonprocess.)
TTnwever process (1) is the most advantageous in respect to outside energy
introotTwherTiSQuantities for processes (2) and (3) are virtually the same. The
ippufddtcttyIn'order to heat the tanks, which rule out the install of power
generation facilities fueled by this gas.
Process (1) is clearly superior, in view of the fact that it offers energy retrieval
in the form of both electrical power through power generation using the digestion
gas aJidTermal power for heating the digesters using the generaUon umt exhaust
heat.
5-2. Construction costs
5-2-1. Hokubu Sludge Treatment Center construction costs
Table 5-6 shows the composite construction costs for the Hokubu Sludge
Treatment Center by unit process.
Table 5-6. Hokubu Sludge Treatment Center Construction Costs
Facility
Sludge intake
Gravity thickening
Centrifugal thickening
Digestion
Dewatering
Digestion gas power generation
Incineration
Other facilities
Construction costs
(millions of yen)
1,330
1,770
2,590
10,490
2,440
3,270
12,990
7,340
42,220
202
-------
5-2-2. Comparison: of the construction cost
It is not possible to make a fair comparison of the construction costs for
centralized treatment and the conventional method because of the differences in the
processes themselves and the level of technological development involved.
Nonetheless, the following may be presented as relevant data calculated on the basis
of the next three points:
1) The evaluation is based on .a comparison with the Kohoku Wastewater
Treatment Plant. This comparison was used because the sludge treatment
process varies from plant to plant, and the main process of the Kohoku plant
closely resembles the process at the Center. - .
2) The process sequence that is the subject of comparison is thickening, digestion,
dewatering and gas power generation; incineration is not included.
3) The cost of land acquisition was not taken into consideration.
Following is the evaluation method that was employed. First, a calculation
was made of the construction costs per unit of treatment capacity in each unit
.process, at the Kohoku WTP and the Center. These figures were then totaled. Next,
the cost of the facilities at Kohoku WTP (conventional decentralized treatment) was
adjusted to reflect the rise in prices. Also, the construction cost of facilities
associated with sludge transportation were added into the total for the. Center
(centralized treatment facilities). The two figures were then compared. The
procedures and results are shown in Table 5-7.
Table 5-7. Comparison of Construction Costs of Sludge Treatment Facilities
Unit construction costs
Thickening
Digestion
Dewatering
Other.processes
Gas power generation
Subtotal
Deflator
Sludge transportation
facilities
Total construction costs
Conventional treatment
system
6,100
51,810
91,430
7,270
156,610
1.440
225,520
Centralized treatment
system
39,660
78,280
34,130
17,630 .
42,280
212.-520
rT ^
26,530
239,050
Note: The deflator is the price increase ratio between 1975-1985.
203
-------
It can be seen that the unit construction costs for centralized treatment are
about 6 percent higher than those of a conventional treatment system. This
difference is due to the inclusion of gas power generation and environmental
protection facilities in the centralized treatment facilities. If comapred after
removing the gas power generation facilities' expenses to equalize conditions with
decentralized treatment, the construction costs for centralized treatment are about
13 percent lower, clearly indicating the benefits of centralized treatment.
5-3. Sludge treatment costs
5-3-1. Treatment costs at the Center
Table 5-8 shows the sludge treatment costs for the Hokubu Sludge Treatment
Center during fiscal year 1988.
Table 5-8. Treatment Costs at the Hokubu Sludge Treatment Center (FY1988)
(Unit: thousands of yen)
Personnel
Power
Chemical agents
Fuel
Repair
Other costs
Total treatment costs
Quantity of treated
solids (DSt)
Treatment costs
(thousands of yen/DSt)
Transportation
and intake
34,543
24,377
18
0
9,996
19,077
88,011
28,580
3
Digestor
dewatering
. 132,139
57,830
232,954
0
108,372
55,234
586,579
28,580
21
Incinera-
tion
132,943
46,626
39,181
118,808
77,085
144,728
559,371
28,580
20
Total
299,675
128,833
272,153
118,808
195,453
219,039
1,223,961
(28,580)
44
Breakdown of
treatment costs
(thousands of yen/DSt)
. 10 (6)
5 (3)
10 (8)
4 (0)
7* (4)
8 (3)
44 (24)
Note: Other costs include comissions and utilities (light, heat, water). Incineration costs do not include
the incineration cost of cake transported from other treatment plants. Figures in parentheses
indicate the costs which do not include incineration cost.
This table presents the breakdown of treatment cost per unit of solids in sludge
transportation and intake, digestion/dewatering, and incineration, which are the
processes involved in the comparison with the conventional treatment process
sequence in the next section. These costs per DS-tons were 3,000 yen for sludge
transportation and intake, 21,000 yen for digestion and dewatering, and 20,000 yen
for incineration, totaling 44,000 yen.
204
-------
5-3-2. Treatment costs in conventional processes
A calculation was made for the fiscal year 1988 treatment costs in the
digestion/dewatermg process sequence in the Chubu, Nambu, and Kohoku
wastewater treatment plants, in order to estimate the treatment costs in the
standard process prior to the introduction of centralized treatment. In the Kohoku
plant, this cost was obtained by adjusting the treatment costs in fiscal year 1983 bv
intervening period The finding« are presented in
Table 5-9. Treatment Costsin Conventional Processes (FY1988)
- (Unit: thousands of yen)
Quantity of treated solids
(DSt)
Treatment costs
thousands of yen/DSt))
Chubu
(WTP)
1,672
Nambu
(WTP)
7,203
Kohoku
CWTP)
2,573
Total
11,448
Breakdown of
treatment costs
(thousands of yen/DSt)
Note: Other costs include commissions and utilities (light, heat, water). Incineration costs are excluded.
As shown in Table 5-9, the total quantity of solids treated at these three plants
^solid's Treatment of this quantity cost ¥ 778 million, or ¥ 68,000 per ton
5-3-3. Comparison of sludge treatment cost
Table 5-10 compares the treatment costs of the Center with the estimated costs
in the conventional sludge treatment system.
Table 5-10. Comparison of Sludge Treatment Costs .-
Treatment costs per ton of treated solids
Treatment costs (up to and including
dewatering)
Quantity of treated solids (DS tons)
Incineration costs per ton of treated solids
Total treatment costs per ton of treated solids
Conventional
system
68
778,416
11,448 •
23
91
i yen, JJQ tons;
Centralized
system
24
674,590
28,580
20
44
Figures for incineration costs in the conventional process sequence are
based on the assumption that a power generation process is not Stflized.
205
-------
As shown in Table 5-10, the costs of treatment, up to and including dewatermg
in the conventional system per ton of treated solids, are about three times greater
than the corresponding costs of about 24,000 yen at the Center.
If the incineration process is also included, the costs per ton of treated solids in
the conventional system would increase to an estimated 91,000 yen. This figure is
almost double that of the corresponding center costs of about 44,000 yen. As such
the benefits of centralized treatment also appear in the comparison of total
treatment costs, i.e., including the incineration costs.
These findings indicate that a rise in the expanse of facilities, while involving a
rise in the total maintenance and operational costs, including power, fuel, and other
expenses, also induces a greater economy of scale including considerable labor
saving meaning a decline in the treatment costs .per ton of treated solids. 1 here is
also the benefit derived from a more effective use of energy due to the use of digestion
gas for power generation. ,
The benefits of centralized treatment also appear in regards to transportation
as well The cartage costs of transporting sludge cake by truck are about ¥ 2,300 per
ton Transportation of the entire sludge cake quantity generated at the Center
(about 77,000 tons) by truck would therefore cost about ¥ 200 million. Transporta-
tion of the sludge equivalent by pipeline costs only about ¥ 88 million.
5-4. Improvement of water quality in wastewater treatment plants
5-4-1. Improvement of water quality through sludge transportation
The shift from treating their own sludge to pumping it to the Center through
pressurized pipelines has improved the quality of treated effluent in the wastewater
treatment plants. This is due to the fact that the shutdown of the sludge treatment
facilities has eliminated the load brought about in sidestreams. However, it is
difficult to quantify this improvement, for the following reasons.
1) The wastewater treatment situation exhibits a broad fluctuation depending on
such factors as climate, the amount of rainfall, and operating conditions during
the year in question. Improved quantification would require comparing yearly
data for at least several years before and after the shift to pipeline
transportation.
2) The influent load has been fluctuating every year due to increases in the
quantity of influent wastewater and the expansion of treatment facilities. This
has made long-term comparisons difficult.
206
-------
3) The Kanagawa and Hokubu No. 1 plants both transport sludge and receive
sidestream water at the same time, making it difficult to isolate the influences
of sludge transportation.
; For these reasons, it was decided to compare the treated effluent quality for a
short period of time before and after sludge transportation at the Midori and Kohoku
plants, both of which are not subject to the influence of sidestreams loads. The period
of time in question is about one year, during which the treatment facilities were not
expanded. The results are shown in Table 5-11, just for reference.
Table 5-11. Treated Effluent Quality Before and After Sludge Transportation
WTP
te
Item
Term of data collection
Sludge transportation rate (%)
Number of facility trains
Influent wastewater quantity (m3/d)
BOD load (kg/m3day)
Transparency (cm)
SS (mg/1)
COD-Mn (mg/1)
Midori
Before sludge
transportation
Oct. 1983 -
Apr. 1984
0
2
30000
0.33
56
6
14
After sludge
transportation
Oct. 1984 -
Apr. 1985
100
2
41000
0.32
89
3
12
Kohoku
Before sludge
transportation
July 1984-
June 1985
0
3
65000
0.42
64
6
11
After sludge
transportation
July 1986 -
June 1987
21
5
62000
0.21
83
5
8.9
July 1987 -
June 1988
66
5
84000
0.24
94
4
8.6
*,^>,^. j.. .my, j..j.iuwii jjiciiinjuiuaLcuBiuuge LImmpui i/auun in vjciooer ±oot i treatment lacuities were expanded in
2. The Kohoku plant initiated sludge transportation in September 1985 (treatment facilities were
expanded in May 1985).
. The effect of transportation appears most clearly in the figures for the Midori
plant, which show major improvement in transparency, SS, and COD after the shift
to transportation, despite an increased influent. The effect cannot be clearly grasped
in the case of the Kbhoku plant, due to the fact that the transportation rate (the
share of the total sludge generated occupied by that transported to-the Center) was
increased in increments from an initial 21 percent to 100 percent over a three-year
period, during which the facilities were expanded. Nevertheless, a comparison of
water quality at the transportation rates of 21 and 66 percent, when the BOD load
was virtually the same, reveal a somewhat improved tendency in transparency and
SS.
5-4-2. Other improvements
Various factors, such as an abnormally .high influent of stormwater and
industrial drainage as well as bulking, can impair treatment in wastewater
treatment plants. Unlike the above, however, the deterioration of water quality
207
-------
associated with sludge treatment trouble is generally not transitory, but exerts a
direct impact on the activated sludge over a long period of time. For this reason, its
impact on wastewater treatment is fairly serious.
The shift to centralized treatment has resolved many such problems, owing to
the fact that the sludge is promptly sent off the plant premises for treatment. In this
area as well, centralized sludge treatment is exerting a favorable influence on
wastewater treatment.
5-5. Effective use of sludge.
In order to prolong the service life of Yokohama's existing disposal sites,
virtually the entire quantity of sludge generated in the city is incinerated at the
Hokubu and Nambu sludge treatment center. Nevertheless, the city has recognized
sewage sludge as a useful resource, and is pursuing various applications of its
effective use, as shown in Figure 5-5.
Fig. 5-5. Utilization of Sludge '
Drying Particular fertilizer (named HAMAYUKI)
_ Lime ash
Soft soil improvement material
. Pavement base material
_ Concreat material
_ Pottery material
_ Polymer ash
Agricultural material
Artifical soil
(named HAMASOIL)
. Earthen square plate
_ Earthen block
_ Cray pipe
. Interlocking block (press sintering)
. Back-filling material
208
-------
• The following is an outline of the two main types of usage that are the subject of
the city s promotional efforts: soil improvement and nursery farming.
5-5-1. Soil improvement
Incinerated ash can be divided into two categories, lime ash and polymer ash
A polymer dewatering agent has been adopted for use in both of the city's centralized
sludge treatment centers (Hokubu and Nambu). But the ash from the Hokubu
center has a higher CaO content (approx. 25%). The reason for this is that the
Hokubu center adopts a dry-type desulfurization method in which CaCCL is added to
prevent SO* generation during the incineration of the sludge cake.
_ ^ Due to its high water absorption capacity and the chemical action of CaO
incinerated ash with a CaO content in excess of 20 percent can be mixed with soil to
improve its suitability for uses relevant to civil engineering. It has been confirmed
that lime-type incinerated ash has sufficient strength for use as material for backfill
and roadbeds (especially the lower layer). In actual use, however, the powdered form
o± this ash causes handling problems.
'The disposal of surplus soil resulting from public works projects is growing into
a major problem in Japan as the number of disposal sites for it declines
Consequently, it was decided to promote the use of lime-type ash as soil improve-
ment material in order to resolve the problem of the disposal of both this ash and
surplus soil. Surplus soil generated during sewerage system construction is'disposed
oi at considerable costs when it is of a poor quality. However, improving this soil
mixing it with incinerated ash for reuse as backfill material, would in effect resolve
three problems: it would reduce the quantity of waste surplus soil and incinerated
ash, and reduce the expenditure for their disposal and backfill use on sand.
Various tests confirmed the practicality of incinerated ash and raised hopes for
the use of lime-type ash as soil improvement material. In view of these prospects
the city constructed a soil improvement plant at the Hokubu No. 2 Plant, adjacent to
the Center, for effectively utilizing incinerated ash. This plant, the first of its kind in
all ot Japan, went into operation in April 1989.
Operation of this plant at a rated output would yield a savings of about 300
million yen annually in expenditures for disposal of incinerated ash and surplus soil
and for the purchase of sand.
The plantmay be outlined as follows:
1) Capacity: 30 m3/hour (continuous operation)
209
-------
2) Rate of incinerated ash admixture: no more than 30 percent relative to
the dried weight of the base soil; quicklime and cement are used as
additional agents if necessary
5) Quality standard for improved soil: an indoor GBR of at least 10 percent
6) Quantity of incinerated ash used: 22,000 m3 per year
7) Quantity of improved soil produced: 49,000 m.3 per year
8) Construction cost: 1.08 billion yen
5-5-2. Use as nursery farming •
Using polymer ash with a low lime content, the city developed the agricultural
materials "hama soil." Production of hama soil begins with the addition and mixture -
of the incinerated ash and polyvinyl alcohol as a binder. After pelletization and
chemical treatment with an ammonium sulfate solution, the mixture is then washed
and dried. The following can be cited as the major characteristics of Hama Soil.
1) Excellent water retention, water absorption, and aeration qualities for
plant cultivation.
2) A pH of 6-7, ideal for plant cultivation
3) Durability; retains structure
4) Change of color reflecting degree of dryness/webness, which serves as a
guideline for watering
5) Lack of odor; sterile
A pilot plant is now producing about 300 kilograms of this soil per day. Efforts
are being made to reduce the production costs while promoting sample shipments to
the market.
5-6. Evaluation
The following effects were anticipated in planning the centralized sludge
treatment facilities.
(1) A more effective system of sludge treatment regarding facility construction,
maintenance and operation.
(2) Promotion of retrieval and comprehensive use of the energy generated from
sludge due to the greater quantities involved in centralized treatment.
210
-------
(3) Improvement of the secondary effluent quality at the wastewater treatment
plants due to the elimination of sidestream water from sludge treatment
facilities.
(4) Improvement in the efficiency of sidestream treatment through centralized
treatment of the same at the Center.
(5) Alleviation of the need for pollution control facilities at the individual
. wastewater treatment plants.
As has already been noted, to date, the operation of the centralized treatment
facilities has achieved the anticipated degree of success. Effects (1) through (3) have
clearly surfaced. The results regarding efficient retrieval and the comprehensive
use of energy exceeded our expectations.
This performance rests on several major factors. First, the combination of the
centrifugal thickeners and the egg-shaped digesters enable.a high-concentration
digestion that yields an increased quantity of digestion gas and enable the amount of
heat for warming the tanks, per load, to be reduced.
Second, the expanded quantity of digestion gas generated is used on a priority
basis as supplementary fuel for the incinerators, which need a considerable
proportion of the energy consumption in the centralized treatment process. Third,
any remaining digestion gas is used as a source of energy for power generation,
which supplies 64% of the Center's own electrical power needs.
Finally, energy is conserved in the operation of incinerators by effectively
using their exhaust heat to dry the sludge cake before incineration. In short, the
successful operation of centralized treatment was made possible by the fruits of
much research and extensive developments aimed at optimizing such treatment as a
total system.
Presently, the Center sends back sidestream generated in the course of sludge
treatment for treatment to certain wastewater treatment plants which send it
sludge. While this system causes no particular problems, the Center is currently
promoting detailed studies for the implementation of the original'plans, which
anticipate centralized treatment of sidestreams at the Center. This system will be
implemented in the near future.
Nevertheless, research on the following subjects is needed to augment the
effects of centralized treatment:
(1) Studies for operating applications of the centrifugal thickener during storms
and in the winter; a concentration of about 5 percent was achieved by the
thickener in mixed sludge. 911
-------
(2) More efficient operation of the egg-shaped digesters (digestion periods, organic
loading, agitation methods, gas generation quantities).
(3) Study of dewatering units and pretreatment to bring the moisture contents of
digested sludge to a level permitting autogeneous combustion, and the develop-
ment of a more efficient incinerator.
(4) Increase the Center's self-sufficiency rate in electrical power and fuel energy to
100 percent through promotion of number three above and the promotion of
energy conservation in facilities.
(5) Utilization of low-temperature energy produced through cogeneration.
(6) Facilitation of sludge transportation, since the thickening characteristics
deteriorate during transport.
212
-------
6. Afterword
The record for the first year of full-scale operation of the centralized treatment
facilities at the Hokubu Sludge Treatment Center reveals that electrical power
generated using the digestion gas supplied 64 percent of the total electrical power
consumed at the Center. This represents an annual savings of about 200 million yen
in electrical power costs, despite the fact that the Center was still operating at only
one-fifth of its planned capacity. It also reveals a decline in the amount of energy
that had to be purchased from outside sources and labor saing stemming from the
shift to centralized treatment at the Center. As a result, the Center's sludge
treatment costs were less than half those of the same sludge quantity under the
previous system of treatment at each wastewater treatment plant. These and other
effects of centralized treatment have been described in the report.
At the same time, it must be acknowledged that it'has only been a short time
since the Center went into full-scale operation. For this reason, the current report
must be termed an interim report as far as the operation and evaluation of
centralized sludge treatment facilities are concerned. It is also clear that there is
still room for improvement in several areas, including sludge transportation,
effective incinerator operation, further energy conservation, and treatment of
sidestream water. In addition, the Nambu Sludge Treatment Center will begin
operation before the end of this year. As such, we are looking forward to preparing
another report specifically on these developments and subsequent progress, at the
next available opportunity.
In closing, the author would like to thank all those who assisted him in the
preparation of this report, particularly Mr. Hachiro Shimomura and Mr. Tetsuro
Ishii, whose help was instrumental in creating the project.
### ••'••'
213
-------
-------
DEVELOPMENT OF SLUDGE TREATMENT TECHNOLOGY
IN
BIOLOGICAL PHOSPHORUS REMOVAL PROCESS
by
Hiroshi Oniki
Director
Planning Department
Sewage Works Bureau
Fukuoka City
The work described in this paper was not funded by
the U.S. Environmental Protection Agency. The contents
do not necessarily reflect the views of the Agency and
no official endorsement should be inferred.
Prepared for Presentation at:
12 th United States/Japan Conference
on
Sewage Treatment Technology
October 1989
Cincinnati, Ohio
215
-------
ABSTRACT
Fukuoka city has been developing an advanced wastewater treatment, process,
mainly based on the biological phosphorus removal, to remove large amounts of
nutrients. Biological phosphorus removal process has many advantages as
compared with the other phosphorus removal processes. However, it has one major
disadvantage that phosphorus absorbed by the activated sludge may be released
when the sludge is handled inadequately, _and this.makes it difficult to treat
sewage efficiently. All the sewage treatment plants* in Fukuoka city are
provided with mixed sludge gravity thickening process and anaerobic digestion
process. Phosphorus absorbed by the sludge may be released in these processes.
In order for the biological phosphorus removal plants to run with some stability,
it is necessary to decrease phosphorus concentration in the return water from
sludge treatment process.
For this reason, Fukuoka city has been investigating the behavior of .-
phosphorus in the sludge, and developing the sludge treatment technologies
for prevention of phosphorus release.
This paper introduces several results of our investigation and
development as follows.
1) Prevention of phosphorus release by adopting dissolved air floatation
thickening.
2) Phosphorus fixation with lime and ferric chloride in digestion tank.
3) Phosphorus fixation with lime and ferric chloride for supernatants.
4) Phosphorus fixation by crystallization of Magnesium* Ammonium Phosphate.
216
-------
1. INTRODUCTION
Recently, eutrophication or aggravation of closed water with excessive
quantities of. in-flowing nutrients, is increasingly causing social problems.
The sewage systems, which are largely dependent on the conventional
secondary treatment, cannot cope adequately with eutrophication. This is
because of the limited capacity of the conventional secondary treatment to
remove nutrients, in particular nitrogen and phosphorus. However, the
eutrophieation is steadily progressing. And it is essential to abate
nutrients flowing into closed water in order to preserve its quality.
Sewage, therefore, has been increasingly treated from the angle of
solving eutrophication problems. Several sewage treatment plants around
lakes are- attempting to remove nutrients drastically, and more advanced
wastewater treatment processes are being developed for cities facing closed
sea water.
Fukuoka City has been recently developing an advanced wastewater
treatment process, mainly based on biological phosphorus removal, .to remove
large amounts of nutrients from sewage. This paper describes some of
important developments carried out so far for the sludge treatment process •
as the core of the biological phosphorus removal process.
PHOTO 1. FUKUOKA CITY AND HAKATA BAY
- 217
-------
2. OUTLINE OF EXPERIMENTS OF ADVANCED WASTEWATER TREATMENT PROCESS
2-1.'Background of Experiments
Fukuoka City is an economic and cultural center of the Kyushu Island
located in the southwest part of Japan, and is the 8th largest city in
Japan with a population of 1.2 million. ,/.„..
The city has been populated along the Hakata Bay with the city facing
north Host of the rivers running through the city drain into the bay, and
treated water from the 6 sewage treatment systems operated by Fukuoka City
flow into the bay either directly or via rivers. The bay is fairly closed
with Unino-nakaniehi peninsula and Shika Island in the east, and Itoshima
peninsula in the west. Therefore, the sea water exchange with the open sea
(Japan Sea) requires many days. As the population and production volume
increase, the bay is increasingly becoming eutrophic, causing red tide to
occur more frequently.The sea water quality of the bay is improving
gradually according to the expanded sewage treatment facilities, but it is
generally still behind the related environmental target levels of water
quality preservation.
Fukuoka City
''JAPAN
nlanfl Sanitation
plant] district Future Pl.esent
Saitozaki
Wajiro
Tobu E=H3
Chubu I** ***•*! l-x'v-x-l
Seibu II1111II Illllllllllll
Nagao
Nanbu •
FIGURE 1. LOCATION MAP OF TREATMENT PLANTS AND SEWERAGE DISTRICTS
218
-------
TABLE 1. SERVICE OF FUKUOKA CITY
(as of March 1989)
Total area of the city
Total area of urbanizd districts
Total population
Total service area
Service area in urbanized districts
Population in sewage service area
Sewerage service ratio
Treatment capacities (total of 6 plants)
33,818
14,969
1,200,204
11,346
75.8
947,800
79.0
538,000
ha
ha
ha
%
%
m3/day •
4.0 r
TARGET LEVEL
(EASTERN AREA)
TARGET LEVEL
/CENTRAL AND \
WESTERN AREA;
79 '80 -'81 '82 '83 '84 '85 '86 '87 '88
YEAR
FIGURE 2. TREND OF WATER QUALITY IN HAKATA BAY
(COD CONCENTRATION)
219
-------
The city has decided to develop new advanced wastewater treatment
techniques because the sewer.age systems will expand to cover 79% ot the
city towards the end of 1989, and increasing quantities, of the
eutrophication causing materials cannot be treated efficiently using
secondary treatment processes.
2-2. Object of Advanced Wastewater Treatment Experiments
The object of the advanced wastewater treatment experiments is to
develop biological phosphorus removal techniques. The advanced .processes
for the prevention of eutrophication are generally designed to remove COD,
phosphorus and nitrogen. The new treatment process of the city is aimed at
removal of phosphorus. This is because the Environmental Bureau of the
city has found through many surveys that phosphorus is the limiting
material for eutrophication in Hakata Bay.
The phosphorus removal processes are three general and practical
categories:
©The physico-chemical process represented by the coagulation
-precipitation and crystallization processes; _
©The combination process in which the physico-chemical and biological
process are combined. represented by the activated sludge process combined
with alum coagulation; and * '
©The biological process represented by biological phosphorus removal
process.
Fukuoka City has selected the biological phosphorus removal approach
for the following reasons:
©The existing activated sludge plants can be easily modified into the
biological phosphorus removal plants.
©It is superior to the coagulation-precipitation process in running
C°St'®It produces less quantities of sludge than that of the coagulation
-precipitation process.
One of the major disadvantages involved in the biological phosphorus
removal process is that phosphorus absorbed by the activated s,ludge may be
released when the sludge is handled inadequately, and this makes it _
difficult to treat sewage efficiently. The sludge treatment plants in all
of the sewage treatment facilities are provided with anaerobic digestion
tanks, because the sludge is finally treated into composts for farms
Therefore, phosphorus in the return water from the sludge treatment plants
must be carefully controlled and reduced.
220
-------
3. OUTLINE OF ADVANCED WASTEWATER TREATMENT EXPERIMENTS
3-1. Outline of Experimental Plant
was mo'SmLlnto ^"^ - sludsf ^ at the Tob'u sewa^ treatment plant
was modified into the experimental plant, which has capacity of 16,700 nf/day
Tobu sewage treatment plant has total capacity of 100,000 rrf/day max with
conventional activated sludge process. A schematic diagram of experimental
Phosphorus removal process is as shown in Figure 3. The sludge treaJIent
Plants consists of the mixed sludge gravity thickening, two-stage anaerobic
digestion, elutriation, and filter press dewatering anaerobic
In the experiments, the aeration tank (herein refferd to as the'
consfstin/efaC70r> ^ ^ ^ ^ biol^ical Phosphorus removal process)
consisting of 7 compartments, devided by the partition walles, modified in
such way that the frrst 2 compartments were made anaerobic stage by
stopping aeration, the remainder left aerobic stage as before' The
anaerobic stage was mixed by the agitators and the openings in the wall '
between the anaerobic and aerobic tanks were reduced! ?o 1 mU Lf ow of
n^rf tTh* excess Activated sludge generated at the experimental line was
passed to the preaeration tank of another adjacent line, to prevent Jele«e
of phosphorus in the primary sedimentation tank. The return sludge was
'18
The target level of the phosphorus removal was 0.5 nig/ £ or less -as
he total phosphorus in the final sedimentation tank effluent The target
hB°D ''
.
. The sludge treatment experiments were carried out by bench scale fp
Plants because quantities of sludge generated from the experLerial line
were too small to be treated by the commercial line.
Biological Phosphorus Removal Process (Experimental Line)
Influent
Biological reactor
to gravity
thickener
I Primary
J sludge
Return sludge
Final
sedimentation
tank
Effluent
Influent
Standard'Activated Sludge Process (Control Line)
Aeration tank
[Excess sludge
^ to preaeration tank
of another line
V V
..V V
V v
Return sludge
Final
sedimentation
tank
Effluent
Excess sludg
Anaerobic
Aerobic
FIGURE 3, SCHEMATIC DIAGRAM OF EXPERIMENTAL
PHOSPHORUS REMOVAL PROCESS
221
-------
3-2. Results of Biological Phosphorus Removal Experiments
The experimental conditions and results are shown in Table 2 and Table
3, respectively. Total phosphorus in the treated water was in arrange of
from 0 37 to 0.44 ml & , Well below the target of 0.5 mg/ fi , and also the
BOD and SS levels were equivalent to those obtainable by the standard
activated sludge process. Temporary increases in NOx and decreases in _
phosphorus removal efficiency, caused by decreased substrates concentrations,
were observed during raining periods, but the plant conditions recovered
well soon after rain stopped. As a result, the target level was reached
for note than 90% probability through experimental period.
These favorable results, however, were with 5 to 6 rag/fi as total
phosphorus in the influent wastewater. It was anticipated by the
calculation by material balance, when all the lines were modified into the
biological phosphorus removal process, that total phosphorus in the teed
water would increase to 9 mg/fl , unless some countermeasures to prevent
release of phosphorus were taken in the sludge treatment system It was
experimental proven that increasing total phosphorus in the feed
wastewater to 8 mg/fi resulted total phosphorus in the effluent from final
sedimentation tank to above 1 mg/fi . , '
In order for the biological phosphorus removal plants to *run with some
stability, it is, therefore,.necessary to decrease phosphorus concentration
in the return water from the sludge treatment process to a level comparable
with that generally observed in the standard activated sl.udge process.
TABLE 2. EXPERIMENTAL CONDITIONS AT WASTEWATER TREATMENT PROCESS
BPRP '
1986fy 1985fy ,
Flow rate
Aeration/flow ratio
Return sludge ratio
Excess activated sludge
MLSS
SVI
C-BOD-SS loading
SRT
MLDO 3
Reactor water temp.
Anaerobic
HRT A U-
Aerobic
(mVday)
(%)
(mVday)
(mg/0
(kg/kg-day)
(day) .
(mg/.g)
(°C)
(hour)
(hour)
12,105
4.7
86
315
2570
140
. 0.16
4.9
3.6
21.5
1.3
3.1
12,139
3.9
82
313
2380
140
0.15
6.1
' 4.0
21.0
1.3
3.3
SASP 2
1986fy
12,143
4.3
77
173
2330
180
0.11
6.8
2.8
21.0 .
.4.7
Note: 1 Biological Phosuphorus Removal Process
2 Standard Activated Sludge Process
3 at the end of reactor
222
-------
CO
a>
CO
LO
!^
CO
CO
w
O
O
OS
CO
O5 4J
<—> C
CD
3
O
LO
05
OO
LO CD
CO
OO,
C
a>
3
oo
O5
w
a:
H
OS
w
00
05
oo
05
-S LO
(O
05
CO
CO
[S
CO
2:
w
I
OS
H
PH
CQ
05 -
_3
u-<
C
CO
o
CM
to
oo
CO
00
O5
CO
f-
f- O
. LO CO
LO
co
00
c
a
c
o
a
s
•
c.
CO
§
CO
w
CO
W
J
PQ
C
o
4->
«•
t.
C
0)
u
e
0
0
o
4->
a
^
a
o
0
OS
c
o
4-1
CO
(,
•4-*
c
0)
o
c
o
O
o
4->
C8
!_
rt
>
O
CU
«
c
o
•«
IS
!_
4^
C
1)
u
c
o
O
o
A
u
"ra
>
o
0)
OS
g^
o
ra
L-
Concent]
o
ra
h
Removal
n
O
ra
c
O,
H
Q
O
CQ
O
CO
CO
H
cs
u
v
OS
•B
O
O
Ul
*0
4J
c
1>
3
5n
M-*
Z
I
n
o
•o
c
a
Z
1
N
O
^ w z
V
o
Z
223
-------
4. DEVELOPMENT OF SLUDGE TREATMENT TECHNIQUES
4-1. Sludge Treatment Process
.
fro. the biological phosphorus remjval plant °°ntal phosphorus may be
than that of the standard activated sludge Plant ' ons for a long
released again into water J^le problem in the sludge treatment
havePbeennstudied in the above two stages.
4-2. Composition of Excess Activated Sludges
The co.positioas of the excess acuvate^slu^e far.hthe
,
asrs ss::1.!?: ^1^. %%. ^ f--- --.„
prl.ars sludge, aad phosphorus conteut is 2.6 *J«S than those in the
se 1:1.!-s^..r.n.i"".i:;!.:., rt..p,.r.. i. the .u,,..
TS
VS
PH
T-N
T_p
K
Ca
Mg
Cu
Zn
Fe
Cd
T-Hn
(%)
(%)
(mg/kg)
(mg/kg)
(mg/kg)
(mg/kg)
(mg/kg)
(mg/kg)
(mg/kg).
(mg/kg)
. (mg/kg)
r (me/kg)
3.67
70.8
6,53
.67,000
26,200
9,000
11,900
9,000
359
1,000
16,100
1.76
0.94
3.96
63.3
5.94
43,500
10,300
2,980
19,800
6,410 •
411
1,300
16,400
1.78
1.52
.93
112
110
110
254
302
60
140
87
77 . ' '
98
99
62
Note: Excess sludge is one thickend by dissolved air floatation
Primary sludge is one thickend by gravity thickener
224
-------
4-3. Phosphorus Release in Gravity Thickening
'"
°f
storage ?easSLf * s' o-
was collated in 24 -ho" s incre «?ni ortn^T \
liquid to 100 mg/£, acco^pan ed nL educ iLP in";
concentration in the sludge from 3 7% to 2 ?o/ ? tOtal
release in 12 hours was about 60% of th Amounts of phosphorus
From a standpoint of phosphorj ^release rh"0"11^ • "^T^ in' 24 hoii>s-
generally has a detention tile of IboJJ'^J ""'^ tMckeninS Process
gravity thickening is' not suitab ]/ "' Which indicates that
in the biological'ph s h'r r em a SarSS '" tFeating eice"
u 4 8 12 16 20
Detention time (hr)
FIGURE 4. RELEASE OF PHOSPHORUS
FROM EXCESS SLUDGE
24
225
-------
4-4. Dissolved Air Floatation
The excess sludge which is
difficult to thicken by S^^^Hf dissolved air floatation was
concentration technique The PJ^ntia o also aeroMc
studied for the reason of its shor thick JJ« Table 5. The
conditions. The operations il r .alt. .re -«»' tlon and 0>23 mg/ I as
in the separated water. ' .
TABLE 5. RESULTS OF THICKENING TEST,
_
pH
SS'
PCU-P
T-P in sludge
(mg/0
(mg/4)
(%)
6.5
4,840
1.91
3.1
6.3
33,400 .
5.46
6.5
320 '
0.23
_..
.6.2
17,300 •
9.80
'•"
6.5
•1,940
5.64
- - —
Note: 1 Thickened sludge
2 Thickener effluent
4-5. Phosphorus Release in Digestion Process
The digestion experiments were conducted, to
of phosphorus in the Digestion process where the floatatlon
phosphorus removal process, thickened by di' °^d^^ (mixing ratio: 3
r . . i j,*«4-l^TrtT»miYP. nWlTjIlbntJr'J-J-lua'J , . TII_I«COY^
are shown in Table b ana
Ce of excess sludge only the sludge Concentration^
organic -atter concentration were too high o be em Q ^ irapossible.
the 26 days detention time, and " qothep hand) solid-liquid
•»««
,
The phosphorus ^^^""^"^J/afer 2 months. The phosphorus
gradually with time, up to J13 mg/ * JJJ however, was fairly constant at
concentration in the case of ffllf J sijf ?' asing for the experimental
200 to 250 ^g/fi Witho^wae4rrenthe supernatant contains very high .
period. In any case, however, the f^e" d directly to the waste water
rr.r s°yr.! ffi;pi°.:"iUt; r.r rof 10. P^P^. ,....1.1..
drastically.
226
-------
TABLE 6. EXPERIMENTAL CONDITION OF DIGESTION TEST
Sludge volume before digestion
after digestion
Supernatant volume
Digestion temperature
Digestion period
Digestor gas generation "ratio
. Organic loading
Digestion rate
(day)
(kg/m3 • day)
5.0
3.03
1.60
37.0
26
7.46
1.05
5.0
2.97
1.72
36.1
26
.7.94
0.91
34
TABLE 7. RESULTS OF DIGESTION TEST
Excess sludee
pH (%)
TS (%)
VS ( (%)
T-P (mg/kg)
P04~P (mg/ty
T-N (mg/kg)
M-Alkalinity (mg/^)
VGA (mq/e)
Inf.1
6.5
4.14
66.6
1,130
' 101
2,610
282
12
.Dig.2
7.4
3.35
56.3
960
. 289
2; 670
3,520
112
Sup.3
; 7.4
3.63
56.7
1,040
. 274
2,760
3,520
Mixed sludge
Inf.
6.3
3.18
73.9
791
75.4
2t170,
386
135
Dig.
7.2
3.01
65.2
1,042
•245
2,550
2,950
137
Sup.
•7.4
0.78
63.0
417
239
, 1,420
2,930
Note: 1 Influent
2 Digested sludge
3 Supernatant
227
-------
500
•400
«? 300
~£b
jj
CU
*
2 200
100
Excess sludge
Mixed sludge
40
50 60 70 80 90 100 120
. Experimental period (day)
FIGURE 5. CHANGES OF PHOSPHORUS
CONCENTRATION IN SUPERNATANT
Therefore, the phosphorus blockade is necessary. In an attempt to
control phosphorus concentration in the return water from the digestion
process, the city has studied three approaches as follows.
©Addition of lime (calusium oxide) and ferric chloride to the feed
sludge for the digestion tank;
.©Addition of lime and ferric chloride to the supernatant, to fix
phosphorus present therein;
©Utilization of crystallization of Magnesium Ammonium Phosphate (MAP)
228
-------
_4-6 Phosphorus Fixation (Preventing Phosphorus Release) with Lime and
Ferric Chloride in Digestion Tank
Figur% 6 illustrates that the pilot plant with 2 tanks of 70 liter
volume for the phosphorus fixation. The sludge used in'the experiments was
the excess sludge treated by the dissolved air floatation process The
fixing chemicals were lime and ferric chloride. Four cases are.studied
where the lime addition "ratio was set at 0, 4.8, 7.4- or 9.8% of the total
solid weight, with neutralization-equivalent quantity of ferric chloride,
to follow phosphorus fixation and. digestion.
The results are shown in Figure 7 and Table 8. The extent of
Phosphorus fixation ascended in proportion to lime addition ratio, whereas
that of digestion reached a maximum at about 0.7 mol/mol as the ratio of
the fixing chemical to total phosphorus in the feed sludge, decreasing
thereafter with lime addition ratio. This could be attributed to decreased
activity of the digestive bacteria as a result of increased quantity of the
inorganic coagulant. It is supposed .that the optimum addition ratio of the
fixing chemical will be around 1.0 mol/mol (about 5% as feed lime
concentration) for the phosphorus fixation from consideration of digestion
ratio, limiting the attainable fixing ratio to around 70% or phosphate
concentration in the digested sludge to approximately 100 m/&
Measurement tank
Gas meter
> Supernatant
Digested sludge
Thickened sludge
storage tank
Digestion tank 70£X2tanks
FIGURE 6. FLOW CHART OF DIGESTION
EXPERIMENT PILOT PLANT
229
-------
TABLE 8. RESULTS OF PHOSPHORUS FIXATION IN DIGESTION TANK
Dosage Molar ratio TS
to T-P
Lime FeCls
VS T-P
(mg/kg)
P04-P
Digestion
Con.1 F.R2 ratio
Note I 1 Concentration
2 Fix ing ratio
o •
4.8
7.4
9.8
0
5.7
8.9
11.9
0
0.84
1.58
2.10
2.97
3.15
2.90
3.03
66.4
65.5
67.3
68.9
1,170
1,140
970
900
498
208
39.3
11.9
0
59 .
92
98
36
• 38
32
22
100
, 80
•2 60
a
i*
bo .
a 40
fa
20
0
150
40 S
o
30 |
o
20 'is
&
Fixing ratio
Digestion ratio
0 1.0 2.0
Molar ratio to T-P
0
FIGURE 7. PHOSPHORUS FIXATION
IN DIGESTION .TANK
230
-------
4-7. Phosphorus Fixation with Lime and Ferric chloride for Supernatant
Next, lime and ferric chloride were added to supernatant, to prevent
phosphorus release. Three cases were studied:
©lime alone
©ferric chloride alone ' ' •.
(3)a combination . .
The results are shown in Figure 8 ~ 10. Ferric chloride removed SS and
Phosphorus more efficiently than lime. The results indicated that molar
addition ratio of the fixing chemical, to keep total phosphorus removal
ratio at 90% or more, was about i mol/mol with fe'rrib chlori-de and 4 mol/ '
mol with lime. A combination of these chemical gave the most favorable
results; phosphorus fixation ratio of 90% or more was attained at a dosage'
ot about 1 mol/mol, accompanied with improved compositions of both treated
water and also precipitated sludges.
100
75
*
50
I 25
T-P
PQ4-P
2 4 6 8
Molar ratio
10
100
§^ 75
50
25
0
T-P
-* P04-P
0 2 4 6 8 10
Molar ratio
FIGURE 8. PHOSPHORUS FIXATION
WITH FERRIC CHLORIDE
FIGURE 9. PHOSPHORUS FIXATION
WITH LIME
100
75
50
"a
§
e
O 0 T-P
*—* PO4-P
24 6 8 10
" Molar ratio
FIGURE 10. PHOSPHORUS FIXATION WITH
FERRIC CHLORIDE AND LIME
231
-------
4-8. Phosphorus Fixation by Crystallization of Magnesium Ammonium
Phosphate (HAP) -Development of MAP Process
Many reports said that scales are frequently deposited on supernatant
transfer pipes of digestion tank in a sewage treatment plants near sea,
causing pipe plugging problems. J, Borgerding detailes that these scales
consist mainly of magnesium ammonium phosphate (MAP). Namely, magnesium in
sea water brought into a digestion tank reacts with phos.pha.te and ammonium
ions present in supernatant from a digestion tank to form insoluble MgNH4P04-
6H.O crystals. Therefore, our attempts were- made to fix phosphorus by
artificially forming MAP. -.
4-8-1. MAP Forming Conditions
The nesessary conditions 'for the'formation of MAP are:
© presence of magnesium . • •
® slightly alkaline condition of the system ,
Supernatant from the anaerobic digestion system contained 207 mg/fi of a
phosphate, 7.0 mg/fi of soluble magnesium, and 756 mg/fi of ammonia nitrogen
'- As the molar ratio of MAP is 1(P): 1(N): 1 (Mg) from the moleculer formula
of MgNH,PO<-6H20, magnesium was deficient stoichiometerically to form MAP.
Addition of magnesium to the supernatant to a M'g/P ratio of 1.05/1 caused
formation of MAP crystals, increasing phosphorus removal ratio.
Solubility of MAP declines as pH ascent, as shown in Figure 11, and
the system should be kept at PH 8 or higher, to reserve MAP efficiently.
The supernatant alkalinity should be increased by some, means, because its
pH level was typically within a range from 7.3 to 7.5.
5.000
4.000
be
< 3.000
•8
& 2.000
1.000
0
012345678
pH of solution
A • 1)
FIGURE 11. EFFECT OF pH ON
SOLUBILITY OF MAP
232
-------
The following four methods were studied to increase pH level of the
supernatant:
(D addition of sodium hydroxide '
(D addition of lime (calusium hydroxide)
(3) addition of magnesium hydroxide
© aeration to accelerate decarbonization
Result of lime was less effective than that of sodium hydroxide, requiring
3 times more quantity of feed chemical to keep the same pH level. Result
ol magnesium hydroxide failed to increase pH level efficiently On the
other hand, aeration was found to be an effective method to remove carbon
dioxide dissolved in the supernatant at a high concentration, to go up PH
level to 8.4 in about 2 hours. The pH level adjustment by aeration, however,
was accopanied by the odor problems, and so the exhaust gases should be
deodorized because of the presence of high concentrations of the materials
causing offensive odor. The composition of exhaust gases are shown in
laore 9. '
TABLE 9. COMPOSITION OF EXHAUST GASES
Concentration
Ammonia
Metyl mercaptan
Hydrogen sulfide
Methyl sulfide
Di-methyl di-sulfide*
(ppb)
(ppb)
(ppb)
(ppb)
(ppb)
1,900
, 150
3.6
37
2.3
Based on the above results, it has .been found that MAP can be produced
efficiently by adding magnesium to the supernatant, and by increasing the
system PH to, at'least 8 with addition of sodium hydroxide or aeration. The
MAP process to continuously remove phosphorus from the supernatant has been
then attempted with pilot plant experiments.
233
-------
4-8-2. Pilot Plant Experiments for the MAP Process (Part 1)
Figure 12 and Table 10 show the flow diagram of the pilot plant and
'experimental conditions respectively. The supernatant storage.tank
uh 2 n? volume was equipped with an agitator and also an aeration system
capable of controlling air ratio arbitrarily with a Hertz conver er to
increase pfl'level. The feed water pumps was a variable type, with manual
con ro covering a flow ratio from, 4 to 0.5 nf/hour A NaOH solution was
added automatically to the feed water, to keep.a PH level in a range from ,
3 to 9 0 MgCl2 injection ratio was changed' in accordance with the
phosphorus concentration in the supernatant. The supernatant was passed
from the storage tank to the rapid stirring tank, then transferee! to the
slow stirring tank for crystallization. And, finally, the crystalline
precipitate were separated -from the feed water in the settling tank, and
some part of the precipitates were sent back as the seeds to the rapid
stirring tank.
MgC£> tank , NaOH tank
Blower
Storage tank
Slow stirring tank
Settling tank
n
Rapid stirring
tank
Effluent
Precipitates
Seed crystal
recycling pump
FIGURE 12. FLOW DIAGRAM OF MAP PROCESS
PILOT PLANT (No.l)
TABLE 10. EXPERIMENTAL CONDITIONS OF PILOT PLANT (PART 1)
Flow rate
Dosage of MgCla solution
Molar ratio to PO4-P
Retention time in stirring tanks
Retention time in settling tank
(nrVhour)
(m^/min.)
(min.)
(min.)
0.5
6 ' >
1.05
24
144
234
-------
Figure 13 shows the results of pilot plant study continuous treatment -
with capacity of 0.5 rrf/hour of the supernatant fro, the digestion LnTa
.the Tobu sewage treatment plant. Ortho-phosphate concentration of the
influent fluctuated from 20. to 80 mg/fl , but the effluent Ortho-phosphate
GOT™*/1011 na;fdp,fr°\5 t0 15 mg/£ (1° mg/£ °n the average), that is,
60 to 70% as Ortho-Phosphate removal ratio. The 'crystals, thus, formed
were very fine, mostly deposited on SS surfaces, and sometimes leaked into
the etliuent to, cause high phosphorus concentration Th'e leaked
crystalline precipitates may become soluble ortho-phosphate at a stage of
reduced PH,_and increase phosphorus load of the waste water treatment system
Therefore, it is very important to control SS in the settling tank In the '
experiments, the effhient water ratio must have been controlled at 0 5 m3/
hour (40m/day as LV) or less. Figure 13 includes test results with'the
seed crystal recycling Recycling of the seed crystals, though sometimes
•increased SS in the effluent water, was an effective method of increasing
size or magnesium ammonium phosphate crystals.
The problem observed in- this pilot plant experiments were difficulty
in aeration to adjust PH of the water In the storage tank, frequently
causing precipitation of MAP in the storage tank, and/or decrease in the
purity of MAP because the precipited MAP was withdrawn together with the -
sludge out of the settling tank. As a next step the new pilot plant was
designed and constructed to solve the above problems.
80
70
^ 60
J50
PH 40
30
C?
10
0
Aug. to Sep. Oct.
0 Influent
• Efflueut
* pH
Nov.
9.0
8.5
8.0'
FIGURE 13. RESULT OF MAP PROCESS EXPERIMENT
235
-------
MAP
4-8-3. Pilot Experiments for the MAP Process (Part.2)
Figure 14 illustrates the new pilot plant where aeration, stirring,
and crystallization 'are carried out together in the same reactor tank.
precipitates formed in the reactor column,located in the middle ol the
reactor tank, are separated from SS and settled by gravity. The
experiments are still ongoing, but the results obtained so far have
indicated a prospe-st of producing high-purity MAP. However, phosphorus
removal ratio is not high so far, because of low phosphorus concentration
in the supernatant separated from the digestion tank of standard Activated
sludge process. As a following experiment, we will attempt the case that
supernatant of higher phosphorus concentration, is use'd.
Pump
Effluent
) >
Supernatant
Storage tank
Blower MAp precipitates
FIGURE 14. FLOW DIAGRAM OF MAP PROCESS
PILOT PLANT (No.2)
236
-------
PHOTO 2.
PILOT PLANT
(No.2)
PHOTO 3.
MAP CRYSTALS
RECOVERED
IN THE PLANT
237
-------
5. AFTERWORD
The pilot experiments are still under way, and there is room for
further improvement especially in the case of MAP. Fukuoka City is
planning to improve the sludge disposal process well-matched the biological
phosphorus removal process. This sequence consists of mechanical
thickening of excess sludge, anaerobic digestion , MAP phosphorus fixation,
dewatering, and compost production processes.
The MAP process is desirable not only from the viewpoint of
stabilization of the advanced wastewater treatment process, but also from
the viewpoint of the effective utilization of resources. Magnesium
ammonium phosphate, rich in phosphorus and nitrogen, is expected to find
use as a product for slow-effect fejtilizers. As a newer process, MAP
process, therefore, will be developed further also as a phosphorus
supplier to fully utilize the limited quantity of natural resources in,Japan,
because the phosphorus of necessary volume in Japan is covered with 100%
by import from foreign countries.
REFERENCE: 1) J.Borgerding, Phosphate deposits in digestion systems,
Journal WPCF, vol. 44, No.5, 1972
238
-------
TOXICItY MONITORING METHOD BY BIOSENSOR
by
Eiichi Nakamura
Chief, Water Quality Section
Public Works Research Institute
Ministry of Construction
Tsukuba, Ibaraki 305, Japan
and
Hiroaki Tanaka
Senior Research Engineer
Water Quality Section
Public Works .Research Institute
Ministry of Construction
Tsukuba, Ibaraki 305, Japan
The .work described in this paper "was not funded by
the U.S. Environmental Protection Agency. The contents
do not necessarily reflect the. views of the Agency and
no official endorsement should be inferred.
Prepared for Presentation at:
12th United States/Japan Conference
on
Sewage Treatment Technology
October 1989
Cincinnati, Ohio
239
-------
ABSTRACT
Use of the ammonia biosensor as a tool to monitor toxicity of water is
presented. Since Nitrosomonas are known one of the most sentive classes of
microorganisms in the biological wastewater treatment processes, respirometry
measurement (oxygen uptake) using Nitrosomonas gives more information on the
toxic characteristics of water than using activated sludge.
Ammnoia biosensor was initially developed using Nitrosomonas europaea
(ATCC 25978) to monitor the ammnoia concentration of water. The ammonia
biosensor consists primarily of a receptor (immobilized nitrifier) and a
probe of dissolved oxygen meter (DO probe). Depending on the concentration
of ammonia in water, the rate of oxygen uptake in the receptor will vary.
Comparison of the measurement of output current of a DO probe with the
calibration curve which shows the relationship between the ammonia
concentaration and the output current gives an estimation of the
concentration of ammonia of a sample.
However, if a sample contains a toxic compound against Nitrosomonas
'europaea, the output current of a DO probe would be lowered depending on the
level of toxicity of the compound. Therefore, the level of toxicity can be
estimated by comparing the expected output current which is estimated by
measuring the ammonia concentration with the actual output of a sample which
contains a toxic compound.
Time required to measure the toxicity of a sample is found 10 to 15
minutes. According to the experimental results using ortho-chloro"phenol,
trichloroethylene and tetrachloroethylene, the ammonia biosensor detects
their toxicity with the order of 0.01 mg/1. Although it was known that
ammonia concentration would influence the relative magnitude of output, the
level of toxicity could be expressed reasonably by using the kinetic
parameters. Required time for the sensor to recover from the damage caused
by a toxic compound is found to depend on the magnitude of toxicity
previously applied.
240
-------
INTRODUCTION
A chemical specific approach to regulate the discharge of toxic
compounds can only be possible if the levels of toxicity of all chemical
compounds against selected indicator organisms are understood and al
cje™cal compounds can b* meas"red by available detection methods. However,
it is practically impossible to understand the toxicity levels of 1]uwever'
S. /2mp°^dS- i" add1t1on> available detection methods have not yet
developed for thousands of toxic compounds (1). Therefore, use of the
VPeClf^.appr°aCh to contro1 the discharge of toxic compounds is
to specific group of compounds such as heavy metals.
thf cnemical specific approach, a biological approach has
waPr S^° P"?™! the d1scha^e of toxic compounds to the public
tollasurl thp " f6 ^J10?1?91 approach, living organisms are directly used
to measure the level of toxicity of a compound, or of water even when the
amount or name of compounds contained in the water are completely unknown A
variety of biological testing methods is already developed from a simS°e
respirometry test to a complicated bioassay test (2). Although ?ox c?ty of a
compound against, specific living organism does not necessarily mean that ?he
compound is toxic against all living organisms including men, use of a
™SUre
tact ^EPA developed tnree types of toxicity test, namely microtox toxicity
test, respirometry (oxygen uptake) test, and adenosine triphosphate (ATP
test, to measure the toxicity of wastewater as near real-time and Inexpensive
monitoring tools (3). Since the measurement of rate of oxygen uptake is a
.routine job for a plant operation, the respirometry test is the easiest amona
these methods. However, activated sludge is not necessarily the most 9
sensit;>ve 9rouP of microorganisms to toxic compounds. It is known on the
other hand, that nitrification process becomes upset when the influen?
contains excessive heavy metals or hazardous chemicals (4). Therefore use
of Nitrosomonas as a.tool for a respirometry test may give more sensitive
result of toxicity test than using activated sludge? sensitive
com* Ammonia biosensor is found capable to detect the level of toxicity of
some toxic compounds, though the ammonia biosensor was initially developed to
.monitor ammonia. And also the ammonia biosensor is found to be potentially
used as a continuous monitoring tool of toxicity. This paper describes the
principal mechanisms of the ammonia biosensor and introduces some of the
experimental results how the biosensor responds to a toxic compound!
DEVELOPMENT OF'AMMONIA BIOSENSOR
Structure. Fundamental structure of the ammonia biosensor is shown in
t^*'<"^"?«'V** ?f diss°1v*d oxygen meter (DO
aSfrpn T°Unted Khere "^rifyin9 bacteria (nitrifier) are immobi zed by
acetyl cellulose membrane filters. . This sandwich structure of receotor is
found easier to mount than other immobilizing methods such as qel
tn7r^oZat 10n $ P^gc^""He. And also gel immobilizations are reported
wLn In8 ane9at]vf.effect on the activities of some microorganisms (5}.
When a buffer solution which contains no ammonia is fed to the sensor, there
is a little consumption of oxygen by endogenous respiration of nitrifier
-------
resulting in rather high and steady output current of a DO probe. This
steady output is considered as the base output. In recording the output, the
current is converted to voltage. When a sample contains ammonia, oxygen is
consumed in the receptor resulting in rather lower output. Since the
difference of this output from the base output depends on the concentration
of ammonia in the sample, ammonia concentration can be estimated. The
difference between the sample output and the base output is simply expressed
as "the output" unless otherwise specified.
Electrolyte.
DO probe
0 - ring
Teflon®
membrane
Receptor _
(immobilized
nitrifier)
Figure-'l Fundamental Structure of the Ammonia Biosensor
Receptor(mixed culture). In the initial stage of development of the
ammonia biosensor, various efforts were made to obtain nitrifier from
activated sludge. Mixed liquors were sampled from a wastewater treatment
plant where nitrification was well proceeded. Activated sludge was cultured
by feeding substrate containing ammonium chloride and no^organic compounds.
After three month cultivation, nitrifier seemed well grown in the sludge.
The sludge was immobilized by membrane filters and mounted to a DO probe.
The results of this prototype sensor are shown in Figure-2. It seemes that
the sensor responds quite accurately to the change of ammonia concentration.
The relationship between the output and the ammonia concentration is shown in
Fiqure-3. Although linearity of the relationship can not be maintained in
the region of high ammonia concentration because of oxygen limitation, use of
this prototype sensor seemed possible. However, the prototype sensor was
242
-------
found to respond to organic compounds contained in a sample, because mixed
culture of microorganisms were applied as a receptor. Figure-4 shows an
example where the sensor responded to organic compounds. To overcome the
effect of organic compounds, efforts were made using antibiotics and other
chemicals in vain. The initial development became a complete failure.
«s
v»
Numbers in the figure indicate
-N concentration fed in mg/1.
Figure-2 Recording Results of the Prototype Sensor:
200 _
2 3
NH$ - N (mg/l)
Figure-3 Relationship between the Output and Ammonia Concentration
243
-------
140 r
120 -
O Glucose
0 Peptone
0 '20 40 60 80 100.
Organic Concentration as TOC (mg/1)
Figure-4 Output Response to Organics Using the Prototype Sensor
Receptor(pure culture); Use of pure cultured Nitrosomonas was
considered from tne experience of initial developmentNitrosomonas europaea
(ATCC 25978) was cultured by using Pramer's culture medium (6) as shown in -
Table-!. Receptor structure is shown in detail in Figure-5. Although
optimum pH and temperature where the maximum output was gained were found 9.2
and 35*C respectively, the sensor was tested under pH of 8.0 and temperature
of 30"C to save required amount of a buffer solution as well as energy for
heating. No response to organic compounds was identified by using glucose
and peptone as organics. The amount of nitrifier was found to affect the
relationship between the output and the ammonia concentration. Figure-6
shows how the relationship differs by the amount of immobilized nitrifier
expressed in adenosine triphosphate (ATP) density. It .can be said from
Figure-6 that higher concentration of nitrifier is preferable to measure
lower concentration of ammonia, and lower concentration of nitrifier is
preferable to measure wider range of ammonia concentartion. Although
durability test is still going on, no deterioration effect is identified for
50 days of use.
Table-! ; Pramer's Culture Medium for Nitrosomonas
(NH4)2 S04
Na2 H P04
KH2 P04
Na H C03
Mg S04 . 7H20 .
Ca C12 . 2H20
Fe S04 . 7H20
Distilled water
2.5. g
13.5 g
0.7 g
0.5 g
0.1 g
5 mg
30 mg
1 ,000 ml
244
-------
0
membrane
II
'
*
filters Teflon^
/ \ . membrane
• • • •
*\t • 9
•\Dz • •
* • » »
x
/
x
-,
3lectrol>
v^
te
X-"
cathode
.anode
Nitrosomonas
glass
Figure-5 Detailed Structure of the Receptor of Ammonia Biosensor
400 r-
o—
N1trif1er Concentration
O 40 ,.g. ATP/mm2
Q 10 ng.ATP/mm2
A 2 vg.ATP/imi2
6 8 10
NHj- N (mg/1)
Figure-6 Effect of Nitrifier Concentration on the
Output-Ammonia Concentration Relationship
/"lm°nif Biosensor System. To apply the ammonia biosensor as a
n • • ammonia, supporting devices should be combined. Althouqh
sampling is an important part of whole monitoring processes, no effort has
an^fj1"" 3" eX1^t1n§ Sampl1n9 P^edure maybe app"ed During the
cSn ta t IfTsamolf if ?'^frature of^the sample should be mainfL'ned
San5! H *? Pi ack of Oxy9en, oxygen should be supplied
Standard solut^ns of ammonia should be prepafed for automatic calibration
The output should be expressed in concentartion, which requires themount of
a calculating system as well as a recording system. Figu?e-7 showl a
schematic system of the monitor using the ammonia biosensor. A buffer
SnSlS In SUPP ?d t0 3d^St the PH °f a sample automatically. Air is
supplied to a sample regardless to the dissolved oxyqen concentration nf *
sample. Figure-8 shows an operational example of the monitor
245
-------
CPU
Amplifier
DO Probe
Receptor
Air Pum
Air
Sampling Pump
Sampl e
Recorder
Display
er Bath
Drainage
Buffer Pump
1 2
Standard Solutions
Buffer Solution
Figure-7 Schematic View of the Monitor
Using the Ammonia Biosensor
Output Status of the
Sensor
Reco
D
R
I
V
E
rdln
0~l
I
g(analogue)
OmV F. S.
P-l •
3Y-I
• SV-I
SY-3
SY-4
Operation Status
Time
1
S
Stand-by,
Output Stabili-
zation *
_______ — — —
CAL-1
^-~~^~~~-
Call
Zero a
MV^~M^il«
Ci
brati
djust
^
CAL-2
Ci
IM^^MKB*
CAL-3
Ci
m(3 poits)-Washing
2 poits calibration
M-OO. OOl*
•.•——•*•
•«— ™^™^"^^
s
Sample dose,
Calculation
Figure-8 Operational Example of the Ammonia Monitor
246
-------
MONITORING OF TOXICITY
Effects of toxic compounds. Since the ammonia biosensor uses
Nitrosomonas europaea as a receptor of the sensor, it is expected to have
some effects in the output when supplied with a toxic containing sample.
Three chlorinated organic compounds, ortho-chlorophenol, trichloroethylene
and tetrachloroethylene, were tested to measure their toxicity against
Nitrosomonas europaea under the ammonia concentration of 1.0 mg/1 Since
, aeration or a sample is considered to have an effect to change the
concentration of a test compound, a sample solution was previously aerated
before dosing a toxic compound to a sample solution. Necessary time of
recovery for the ammonia biosensor was examined for ortho-chlorophenol as a
toxic compound. K
Since maximum allowable concentaration (MAC) of each compound below
which there was no effect on the rate of respiration was not precisely
detected, the maximum concentration where the output was the same with the
control was assumed as MAC. Figures 9, 10 and 11 show the results of
experiment for ortho-chlorophenol, trichloroethylene and tetrachloroethylene,
respectively. Relative magnitude of output was calculated by comparing
Sam^n^Utp^ Wlth contro1 output. For ortho-chlophenol, MAC was considered
as 0.034 mg/1. It can be recognized that there is a substantial decrease in
the relative magnitude of output as the concentration of ortho-chlorophenol
increases. Similar tendency is also recognized for other two compounds.
When the ammonia biosensor is fed with a sample containing toxic
compounds, time is required for the sensor to'recover from the effects of
toxics. One. of the experimental results using ortho-chlorophenol as a toxic
compound is shown in Figure-!2. Required time of recovery is increased
proportionally to the increase in toxic concentration.
1.0
f
o
0.5
0
-*• MAC =0.034 mg/1
0
0.5
1.0 1.5 2.0
Ortho-chlorophenol (rag/1)
2J5
Figure-9 Effect of Ortho-Chlorophenol on
Relative Magnitude of Output
247
-------
1.0 ft
MAC = 0.015 mg/1
S 0.5
C
O>
m
0
0 0,5 , 1.0
Trichloroethylene (mg/1)
Figure-10 Effect of Trichloroethylene on
Relative Magnitude of Output
MAC = 0.019 mg/1
•s
01
-------
>>30
0)
>
o
o
OJ
fe 20
q-
0)
10
3
CT
O)
o;
0
•i-
Concentration of orthochlorophenol (mg/1 )
Figure-12 Required Time of Recovery for Ortho-Chlorophenol-
v = Vmax S / ( S + Ks (1 + I / Ki)) (1)
where, Vmax is the maximum rate of reaction S is
of
v = Vmax S / (S + Ks (1 + (I - MAC) / Ki))
where, (I - MAC) = 0 if I is lower than MAC.
(2)
(I - MAC) / Ki = ((1 - p)/p) ((KS + S) / Ks) (3)
^ where, p is the relative magnitude of output as shown in Figures 9 to
249
-------
c;n e
Table-2 Toxicity of Ortho-chlorophenol
Concentration
(mg/1)
0.034
0.068
0.136
0.340
0.680
1.360
2.720
6.800
Relative magnitude
of output, p
1.0
0.932
0.941
0.867
O'.SIO
0.694
0.552
0.185
Toxicity
(I-MAO/K1
0
0.104
0.089
0.219
0.334
0.629
1.157
6.280
Ki = 2.184 mg/1, CV = 0.272
Table-3 Toxicity of Trichloroethylene
Concentration
(mg/1)
Relative magnitude
of output, p
Toxicity
(I-MAO/K1
0.015
0.030
0.060
0.120
0.240
0.480
1.20
1.0
0.961
0.741
0.699
0.480
0.305
0.151
0
0.058
0.498
0.614
.544
.248
1
3.
8.015
Ki = 0.217 mg/1, CV.= 0.073
Table-4 Toxicity of 'Tetrachloroethylene
Concentration
(mg/1)
0.019
0.037 .
0/056
0.074
0.111
0.148
0.370
Ki = 0.120
Relative magnitude
of output, p
1.0
0.977
0.760
0.707
0.566
0.457
0.231
mg/1, CV = 0.074
250
Toxicity
(I-MAC)/Ki
0
0.034
0.450
0.591
1.093
1.694
4.746
-------
CONCLUSION
the experimental results indicated high potential of practical USP of *
ammonia biosensor in the field of toxic Lnagement practlcal use of
CREDITS
?*
8-
poilutlnu;'
REFERENCES
B10'°9lCa'
toxic wate,
3
4. Hockenbury M.R. C.P.L. Grady(1977) Inhibition of nitrification effects
of selected organic compounds; JWPCF, 49, 768 '"'ricanon errects
5. Chibata, I. (Ed.) (1981) Immobilized enzymes; Kodansha, Tokyo
Nitrosomonas in pure culture; j.
* ISfel!!lSi;.J76rSr'(1958)
J W.H.Freeman and Company, San
251
-------
-------
RENOVATION OF AN, EXTENDED AERATION TREATMENT PLANT FOR
BIOLOGICAL NUTRIENT REMOVAL PROCESS
by
T. Matsui*, K. Sato*, K. Moriyama*,
M. Imamichi** and Y. Harada**
C°f r01 De*>artment' Public Works Research
1, Asahi, Tsukuba-Shi,
** Sewage Works Division, Hamamatsu City Hall, 104-6, Motoshiro
Hamamatsu-Shi, Shizuoka-Ken, 430- Japan Motosh2.ro,
The work described in this paper was not funded by
the US. Environmental Protection Agency. The contents
do not necessarily reflect the views of the Agency and
no official endorsement should be inferred
Prepared for Presentation at:
12th United States/Japan Conference
on
Sewage Treatment Technology
October 1989
Cincinnati, Ohio
253
-------
ABSTRACT
F
. INTRODUCTION
operators .
Under these circumstances, in December 1,982, the Environment Agency
obliged to remove nitrogen and phosphorus.
anoxic tank is designed with a total detentxon tune of 12-16 hours.
254 .
-------
°*£i??au? of this Process is in the range of 60-70%
All but one of the plants which was newly built, were oricH™mv
of nitrogen and
o Mass balance of alkalinity in this process.
o Re^°n KSreen rem°Val °f nitr°9en ™* that of phosphorus.
P Relation between ORP control and DO concentration
o Response of this process to load variations.
PLANT DESCRIPTION AND EXPERIMENTAL PROCEDURE
A schematic flow of the plant is shown in Fig. 1. The reactor
255
-------
denitrification reaction was assumed as "aerobic denitrification". The
second basin of 6 hours hydraulic retention ^i- was post-denxtrxf.catxon
zone for nitrate nitrogen leaked out from basin A. The last oxic basin or
£
the anoxic-oxic process.
Aeration/ORP
Control System
Flow | — Influent-
Equalization
Tank j
tGrit Chamber
Effluent
<••
i
— r~
— 4-
-— ¥
,gnal- noRp ^^
Basin A |B
Oxic Anox'c3
330m3r 1 250m3
L[MJ [Mi]1[M2]-[M2]J
t air t
1
C
Oxic
80m3
tai
-* Settl
Tank
/^
[ : Keturn aiuoge
discharge
ing
im3)
— ^-*
Wa
to
X
Upf 1 ow
Sand
Filter
tBiological
Effluent
ste Sludge .
other plant
[MxliSubmerged mixer with air piping
[M2]Submerged mixer
[Bi]:Blower with frequency converter
[B2]:Blower
Pig. 1 Flow chart of the Hitomigaoka- plant
The experiment was commenced in Oct. 1987, and is going on as of Aug.
1989 This paper analyzed the data obtained from Jan. to, Aug. 1988. The
TABLE 1 Operation Conditions
Influent flow
Retention time in the biological reactor
(Basin A; 8 hours, Basin B; 6 hours, Basin
Return sludge ratio
MLSS
BOD-SS loading
TN-SS loading
Sludge retention time
Sludge aeration time
ORP value set in Basin A
900-1100 m /day
15-17 hours
C; 2 hours)
60-90%
2800-3200 mg/1
0.05-0.08 kg-BOD/kg-SS/day
0.01-oTo2 kg«TN/kg«SS/day
17-25 days
10-16 days
100-»125—-150--180--115 —•
125 mV (changed stepwise)
256
-------
RESULTS AND DISCUSSION
Nitrogen Removal
Outline of nitrogen removal
s.
kept in the best condition n ' concentration of the effluent was
60
50
^40-
Z 20-
10-
Jar
°Inf.
0
0 o o . •-
D° D D D aQ o
O
100 .1?-5-/" ' V D a. 125 mV
i^^r*^* S^*,** ^
>. Feb. Mar. Apr. May Jun^Jul'] Au^
T-N +Pff T M t,cff n_ >!.•-._
,20
1
0. c*
O ^ir\
200 z
100
Ort
)t. Ja
:::::::::::/:;:^;:i ^
+ / 0 t
^E~~'A* ••• "AA.- • - -d" •-......
-*^^..i*C?...*t.r....
..A...* * ' * + +
e>* ^ % A"A--f — f----
n. Feb. Mar. Apr. May Jun."
...-.
"Tr'&'d
"WSWftTWfl
Jul. Aug.Sep
o.
a:
0 ,
100
0
t.
.3-.NH4-N-ORP
Fig. 2. Variations of Nitrogen
concentrations in the
influent and the effluent
D-T-N
-QRP
Fig. 3. Variations of nitrogen
concentrations in Basin A
2 or Fig. 3, ORP control can be easily achieved? ' "
257
-------
5 gives the typical behavior of nitrogen in the process, when the
value of l^v is adopted. As mentioned »"«•»
shown in Fig. 5.
16-r
14-
12-
1 8-
z 6-
4-
2
n
_ o
° C
° 0 *
* a. % «
-------
c
o
r<
.Li
« o
If
J
•^
Nitrification
CX7.14
"\^:
Apparent
consumption
c
o
o
Fig. 7. Alkalinity balance in the
simultaneous nitrification-
denitrification reactor
250
200
150-
100-
50-
0
CX3.57
0 20 40
Nitrified nitrogen, C mg/s.
60
Fig. 8. Relationship between nitrified
nitrogen and consumed
alkalinity in Basin A
Phosphorus Removal , •
at different Points in this process are
all
On the other hand, during the efficient
of mv
-------
.
to take place in the settling tank.
ss ^^
higher than 4 mg/1 in the settling tank.
Q.
•a-
12-
10-
8-
6-
4-
2-
ocn D
**
n" °
a a
a +5
a a
n D + «. 4
K: o"**
Stf^.^dtD
°100 ' 120' 140 160 180
ORP tnV
PO,-P + NO,-N
in-R.S. in R.S.
NO,-N in
Basin C
Fig. 11.
Relationships between PO4-P and NO3-N concentrations
ORP Value and DO Concentration
process.
260
-------
O
O
-O 1
£
80 100 120 140 160 180
Measured ORP value mV
200
Fig. 12. Relationship between the measured ORP
value and the measured DO in Basin A
30
2 25
-------
bU-
50-
AQ-
30-
zu-
io-
4 A A March 7 to 8, 1989
A A A ORP value set; 120 mV
a,
V*^ 0 + **>^ + 4, ^
O
200 1~M-
Q. i-
i nn -^ t|—
1UU •
a a.
o a:
en o
8 10 12 14 16 18 20 22 24 2 4 6
Time
-Supplied + Inf. T-N » Inf . BOD * INf . -ORP
air flow
D-TOC
Fig. 15. Variations in the influ'ent load and diurnal
variations in the supplied air flow
CONCLUSIONS
The following conclusions were obtained in this study:
1) It has been proved that in this method, the optimum ORP value for
simultaneous removal of nitrogen and phosphorus exists, and. that ,a
completely mixed flow single reactor, when its ORP value is kept at this
proper value, can provide a simultaneous nitrification and
denitrification reactor, where simultaneous removal of nitrogen and
phosphorus can be achieved.
2) Concretely speaking, it was required to put the NO3-N concentration in
Basin C below 0.5 mg/1 by controlling ORP. The necessary anaerobic
condition in the sludge blanket of the settling tank was assured in this
manner, and phosphorus removal was thus achieved.«
3) Examinations of the relationship between measured ORP values and measured
DO concentrations showed that ORP values are more suitable than DO
concentrations, as a control factor in this method.
4) In the conditions of the optimum ORP value and 16 hours of the retention
time in the biological reactor during a high water temperature period, the
nitrogen removal rate of 90-95%, the effluent T-N of less than 2 mg/1,
the effluent T-P of 0.5-1.0 mg/1 and the effluent PO4-P of 0.5 mg/1 were
achieved.
5) When nitrogen removal is aimed at, then a completely mixed flow single
reactor with an approximately 8 hour retention time and an ORP control
system can achieve the effluent T-N of approximately 10 mg/1.
REFERENCES
1. Robertson, 'L.A. and Kuenen, J.G., "Aerobic denitrification - old wine in
new bottles," Antonie van Leeuwenhoek, 50, pp. 525-544, 1984.
262
-------
2. Kuenen, J.G. and Robertson, L.A., "Ecology of nitrification and
denitrification," In: The Nitrogen and Sulphur Cycles. Society for
General Microbiology Symposium 42, J.A. Cole and S. Ferguson (Eds)js
Cambridge University Press, Cambridge, pp. 161-218, 1987.
263
-------
-------
Development of a High Efficiency Nitrogen Removal System
Using Immobilized Nitrifiers in Synthetic Resin
by
Masanobu Aoki, Deputy Director
Minoru Tada, Deputy Director
Takashi Kimata, Senior Researcher
Shouji Harada, Research Chemist
Yuhko Fujii, Research Biologist
Research and Technology Development Division
Japan Sewage Works Agency
The work described in this paper was not funded by
the U.S. Environmental Protection Agency. The contents
do not necessarily reflect the views of the Agency and
no official endorsement should be inferred
Prepared for Presentation at:
12th United States/Japan Conference
on
Sewage Treatment Technology
October 1989
. Cincinnati, Ohio
265
-------
Development of a High-Efficiency Nitrogen Removal System
Using Immobilized Nitrifiers in Synthetic Resin
Masanobu Aoki, Deputy Director
Minoru Tada, Deputy Director
Takashi Kimata, Senior Researcher
Shouji Harada, Research Chemist
Yuhko Fujii, Research Biologist
Research and Technology Development Division
Japan Sewage Works Agency
ABSTRACT
This technology is a biological nitrogen removal process which uses
immobilized nitrifiers (the IN Process) to reduce the reactor volume
required for nitrogen removal. Nitrifiers are immobilized in 2 to 3 mm
diameter polyethylene-glycol resin. The immobilized nitrifiers in the resin
(Biopellets) are mixed with activated sludge in the nitrification tanks to
promote quick nitrification. Since nitrification is completed at shorter
sludge retention times using Biopellets, the reactor volume required for
denitrification is also reduced. As a result, sufficient nitrogen removal
within the conventional reactor-volume (6-8 HRTs) has become feasible with
this technology. The IN Process was developed through a joint research
project between the Hitachi Plant Construction, Co., Ltd. and the Japan
Sewage Works Agency (JSWA) under a research contract between the Ministry of
Construction and JSWA. This paper introduces selection of sythetic polymer,
polymerization methods, nitrifying activity of Biopellets, separation and
fluidization of Biopellets, and applicability of Biopellets to nitrogen
removal processes.
Keywords; biological nitrogen removal, nitrified liquor recycling activated
sludge process, immobilized nitrifiers, synthetic polymer,
polyethylene-glycol
266
-------
Development of a High-Efficiency Nitrogen Removal System
Using Immobilized Nitrifiers in Synthetic Resin
Masanobu Aoki, Deputy Director
Minoru Tada, Deputy Director
Takashi kimata. Senior Researcher
Shouji Harada, Research Chemist
Yuhko Fujii, Research Biologist
Research and Technology Development Division
Japan Sewage Works Agency
Introduction
Nitrogen removal from municipal wastewater is biologically
using nitrification and denitrification mechanisms. Nitrifiers
r;?^,!!i!r^.n_!0"itrite ^™> *** to nitrate nitrogen.
nitrate nitrogen to nitrogen gas. Two biological
scesses have been applied in full-scale plants in Japan
the nitrified liquor recycling activated sludge system
denitrifying activated sludge system. Although these '
i effective nitrogen removal, the applications have been
larger reactor volume required. To maintain a sufficient
:s in an activated sludge reactor* during periods of low
reactor volume has to be *-"-'•— «--••- -^ -
— uses
for mtrogen removal. Nitrifiers are immobilized in 2 to 3mm
.
volume required for denitrif ication is also reduLd wfthout 4ny carSn
sources, excepting that in the influent wastewater. As a result
nitrogen removal within the conventional reactor-volume (6 - 8
become feasible with this technology.
267
-------
IN Process
Nitrified liquor recycling
T
Biopellets
DN
0 ° 0
N
O O
6 to 8 hr
Return sludge
Conventional nitrified liquor recycling system
Nitrified liquor recycling
IDenitrif ication
tank
Nitrification
tank
16 hr
Return sludge
Figure 1.
Difference between the-IN Process and the conventional
nitrified liquor recycling .system.
a. The selection of polymer resin suitable for immobilizing nitrifiers,
b. Methods for the immobilization of nitrifiers,
c. The separation and fluidization of immobilized nitrifiers,
d. Secondary effluent nitrification using immobilized nitrifiers, and
e. Biological nitrogen removal using immobilized nitrifiers in
activated sludge reactors.
In 1988, a pilot plant study was initiated to evaluate the process
performance of the IN Process.
Selection of the polymer resin, and immobilization method
Polymer resin
The immobilization of microorganisms in polymer resin is a widely
applied technique in drug manufacturing and food processing. The resins
used in those fields are natural polymers (such as agar and carrageenan),
and synthetic polymers (such as acryl-amide and epoxy). Generally, the
characteristics required of the resin for immobilizing microorganisms are:
268
-------
a. to be water soluble, and able to gel at ambient temperatures
b. to have a low toxicity to the immobilized microorganisms during and
after gelling,
c. to have a high dispersion coefficient for the substrate to be
treated in the resin, e
d. to have low biodegradability and to be durable,
e. to be physically strong, and
f. to be inexpensive.
Among the resins used for microorganism immobilization, the natural
resins are less toxic and less expensive compared to synthetic ones.
However, they are physically weak and easily biodegraded within short time
periods, in contrast, synthetic resins have the opposite characteristics.
When applying microorganism immobilization to wastewater treatment the
resin should have good durability to reduce 'operation and maintenance '
costs. This must be the most important point when selecting a desirable
resin. Therefore, it was decided that synthetic resins would be used as the
immobilizing material early in that selection process. The physical
characteristics of polyethylene-glycol (PEG), acryl-amide (AAm) epoxy
(Epx), urethane, and melamine were examined. The results showed that
melamine resin did not gel well at ambient temperatures, and .that it was
gelled1" ^ Ur&thane resin to take on a'spherical shape, even though it
The physical,characteristics and nitrifying activity of PEG, AAm, and
Epx pellets as immobilizing nitrifiers (Biopellets) are compared in Table
if R^^yin9-aCtiVl?,WaS measured as the respiration rate per unit volume
of Biopellets in aerated ammonium nitrogen solution. The composition of the
solution is shown in Table 2. The initial and final respiration rates were
measured just after gelling, and after 80 or more days of acclimation with
ammonium nitrogen solution (Table 2), repectively. Those physical
characteristics did not differ .very much. However, the Biological
characteristics were significantly different. Both the intitial and final
am°ng a11 three resins
269
-------
Table 1. Physical characteristics and nitrifying activity
of pellets immobilizing nitrifiers.
Evaluation item Polyacryl-amide Epoxy Polyethylene-glycol
Compressive strength
(kg/cm2)
Deformation ratio (%)
Degree of swelling (%)
Durability
Initial respiration rate (%)
Final respiration rate
(mgO2/2-pellets/hr)
2.1
49-
101
Good
1 or less
500
2.1 - 5.2
30 - 69
81 - 107
Good
1.8 - 3.3
180 - 500
2.8 - 3.6
33-52
93 - 113
Good
5-15
1100
Note- The initial respiration rate is the ratio of respiration rates'
before and after immobilization. The final respiration rate is
measured after 80, or more, days of acclimation with ammonium
nitrogen solution.
Tablet. Composition of ammonium nitrogen solution.
Chemicals
Concent rat ion
NH4C2
76.1 mg/2 (20 mg-N/£)
Na2HP04*12H2O
23.1 mg/2
NaC2
10.1 mg/2
KC2
4.7 mg/2
Cac22'2H20
4.7 mg/2
MgS04-7N2O
16.7 mg/2
NaHCOa
243.3 mg/2
Note: Sodium bicarbonate is added as a pH buffer.
270
-------
The change in respiration rate with time for PEG and AAm pellets are
shown in Figure 2. The respiration rate of the PEG pellets increased
quickly as compared to the AAm pellets. This means that PEG provides a more
suitable environment for nitrifying bacteria than AAm. As a result of these
physical and biological considerations, PEG was chosen as the resin for
immobilizing nitrifiers.
1000 r
PEG
Pellet diameter :
NH4-N Concentration:
Nitrification rate :
Water temperature :
2.5 to 3 mm
20 mg/e
150 to 240 itigN/2 pellets/h
20°C
80
100 120 140
Time (days)
160
180
200
220
Figure 2. Change of respiration rates with time for PEG and AAm.
Immobilization method
Biopelletes must be produced in large amounts for use in full-scale
treatment plants. From this point of view, four immobilization methods were
evaluated. The methods tested were:
a. intra-tube polymerization,
b. sheet polymerization,
c. drop polymerization, and ,
d. suspension polymerization.
271
-------
The applicability of each method is summarized as follows:
a
Intra-tube
polymerization
Sludge
Immobilizing
material/Gelling
catalyst
L-m^^zzErr-^/jX^
Cylindrical shape
b
Sheet
polymerization
Sludge
Immobilizing
material/Gel 1 ing
catalyst +
L Cutter
Upper belt f^
) - p_Qj>
O — OY- •
Lower belt
Cubic shape
c
Drop
polymerization
Sludge
Immobilizing
material/Gelling
' catalyst +
Sodium alginate
(S>
V
CaCl2 ~4--^^(
Spherical shape
d
Suspension
polymerization
Sludge
Immobilizing
material/Gelling
catalyst
r — r\
l.'ob .-T> — oil
\^y
Spherical shape
Figure 3. Immobilizing methods.
a. Intra-tube polymerization (Figure 3 a)
A mixture of activated sludge (nitrifier enriched), PEG prepolymer, and
gelling catalysts (accelerator and starter) is pumped into a tube. In the
tube gelation of the mixture occurs. The gelled mixture is extruded from
the tube, and then cut into small, cylindrically shaped pellets. Several
tubes must be prepared for mass production. However, there are some
mechanical constraints involved in cutting several bars of gelled mixture at
the same time into a uniform shape.
b. Sheet polymerization (Figure 3 b)
The mixture is cast between two sheets which are separated by a fixed
distance. The-mixture gels there, and then the gelled sheet mixture is cut
horizontally and vertically to produce cubic pellets. Since this process
seems to be easily automated, the sheet polymerization method should be
applicable to mass production.
c. Drop polymerization (Figure 3 c) ,
A mixture of activated sludge, PEG prepolymer, sodium alginate, and
gelling starter is dropped into a calcium chloride solution. The alginate
and calcium, ions react and produce a thin membrane. The mixture is covered
with the membrane and gelation proceeds inside the membrane. An appropriate
sized nozzle producing spherical pellets with a set diameter was determined
through experiments.
272
-------
d. Suspension polymerization (Figure 3 d)
^ Amixt"re of activated sludge, PEG prepolymer, and gelling catalysts is
added to oil. The mixture is dispersed and suspended in the oil and with
agitation forms small drops. In such a suspended condition, gelation
proceeds. This process does not appear to be suitable because the drops
stick to each other and become large clusters during suspension.
It was considered from the above evaluations that only the methods of
sheet polymerization and drop polymerization were applicable to
immobilization. The latter method was chosen temporarily for producing
Biopellets to be used in experiments on treatment performance. This was
because of its mechanical simplicity. The procedures for producing
Biopellets by drop polymerization are shown in somewhat more detail in
Figure 4.
C Conditioning of^\
seed sludge J
Gelling catalyst
C
Preparation of
immobilizing
material
Mixing""')
Immobilizing
material
Gelling^)
Pellets
Figure 4. Procedures to produce Biopellets by drop polymerization.
273
-------
Acclimation of Biopellets
The respiration rates of nitrifiers decrease to 5 to 15% of their
initial rates just after immobilization in PEG. Within two months (or less)
acclimation with wastewater, the rates increase, and then plateau. To
determine the period required for acclimation, the effects of temperature on
nitrifying activity, and in which part of a PEG pellet the nitrifiers
reside, a laboratory study was conducted.
The effects of activated sludge types on acclimation are shown in
Figure 5. Two different types of activated sludge were seeded in PEG
pellets. One was sludge from a nitrified liquor recycling activated sludge
plant, the other was the same sludge but pre-acclimated for 80 days with
ammonium nitrogen solution. The two kinds of pellets were applied to
wastewater treatment at 20 and 30°C. As shown in Figure 5, the respiration
rate of pellets seeded with acclimated sludge increased rapidly at a
temperature of 20°C. By comparison, pellets seeded with the original sludge
needed a longer period to reach the plateau of respiration at the same
temperature. However, at 30°C the pellets did not require such a long
period, and were more rapidly acclimated than nitrifier-enriched pellets at
20°C. The respiration rates after 100 days were 600 to 800
mg02/£-pellets/hr, and there were no differences among the three cases.
These results mean:
~ 1000
I -
W
% 800
600
$ .
* 400
8
•rt
g 200
U
___ A
Sludge from a nitrified
liquor recycling
activated sludge A
plant (30°C) ,-'' A . . t ,
* x^re-acclimated
xx sludge (20°
A
Sludge from a nitrified
liquor recycling activated
sludge plant (20°C)
10 20 30 40 . 50 60 70 80 , 90
Time (day)
100
110
Figure 5. Effects of activated sludge types on acclimation.
274
-------
a.
b.
c.
d.
the acclimation period required depends upon the seed sludges and
temperatures,
the final nitrifying activity is not related to the seed sludge or
temperature,
at higher temperatures, the sludge type seeded with pellets does
not have a significant effect on starting up reactors, and
at lower temperatures, the nitrifier-enriched sludge should be used
as seed sludge for immediate starting up.
Microorganism distribution in a pellet was observed using an electron
microscope. The observed results over time are shown in
Figure 6. The growth of microorganisms in the pellets proceeded as follows:
a. Microorganisms exit dispersively in a pellet just after
immobilization.
b. Microorganisms form colonies in a pellet during increasing
nitrifying activity. The growth of colonies is significantly high
near the pellet's surface (until the 60th day).
c. The density of the colonies increases near the pellet's surface,
where the biological environment is better than that inside,
colonies inside the pellets tend to fade out (from the 60th'to the
200th .day).
d. The microbial layer near and/or on the surface grows to a thickness
of 60 urn, and there seem to be no mircoorganisms inside the pellets
(after the 200th day).
Nitrifying
bacteria
Growth of
colonies
Growth of colonies
near the pellet
surface
50/
-------
From the above observations, it was shown that nitrifying bacteria
exist only near and/or on the pellet's surface after sufficient acclimation,
and that the inside of a pellet works merely as a support medium. This
indicates that smaller pellets produce better nitrification performance when
only a small amount of pellet material is available. The relationships
between pellet diameters and final respiration rates are shown in Table 3.
Smaller pellets showed higher respiration rates per unit pellet volume.
However, respiration rates per unit pellet surface area were approximately
400 mgO2/m2/hour, and almost the same among pellets with different
diameters. This result clearly shows that the pellet diameter must be as
small as possible. The separability of pellets from activated sludge will
determine the suitable diameter for Biopellets.
Table 3. Relationships between pellet diameters and final respiration rates.
Diameter
1.00 - 1.41 1.41 - 1.68 1.68 - 2.00 2.00 - 2.38 2.38 - 2.83
Respiration rate
per unit pellet 1,730
volume
(mg02/£/hr)
1,600
1,480
1,260
1,110
Number of
pellets per unit 1,105
pellet volume
(1000/£)
523
307
179
109
Surface area per
unit pellet
volume
5.00
3.90
3.26
2.73
2.31
Respiration rate
per unit pellet
surface area
(mgOz/m2/hr)
346
411
454
462
481
Separation and fluidization of Biopellets
Separation of pellets
Biopellets have to be separated from treated wastewater or activated
sludge to maintain them in a reactor. Two separation methods, wedge-wire
separation (Figure 7) and gravity separation, were tested. It was thought
before the experiments that pellets would be separated from activated sludge
by a slight difference in the setting velocities of the pellets and the
sludge. But the apparent density of the pellets decreased from 1.06 to
1.03, or less, due to the attached sludge so that stable pellet separation
was difficult.' On the other hand, wedge-wire worked well. It was found
that wedge-wire should be installed at an angle of 30° to 60° from the
water's surface. .
-------
Air bubbles
'l O '
Biopellets i'o/ry'
Mixed liquor
I Wedge-wire
^Titling angle
Figure 7. Biopellet separation with wedge-wire.
To determine the conditions for wedge-wire separation of pellets,
several studies were carried out using a 200 Z reactor. The effects of
wedge-wire opening on the separation of pelltes with different diameters are
shown in Table 4. Each separation test was continued for one hour. Both *
pellet loss ratios and wedge-wire clogging ratios were measured after one
hour operation. For the combination of a wedge-wire with a 1mm opening and
pellets with 2.00 to 2.83mm diameters, pellet loss was less than 0.1%, and
the clogging ratio was less than 5%. When a wedge-wire with a 1.5mm
openning and pellets with diameters of more than 2.83mm were used, the
ratios were the same as with the previous combination. These ratios, 0.1%
loss and 5% clogging, did not cause any operational problems. This result
indicates that the pallet diameter should be at least two times as large as
the wedge-wire opening. '
277
-------
Table 4. Effects of wedge-wire opening on separation of
pellets with different diameters.
Wedge-wire
opening (nun)
0.5
1.0
1.5
Pellet diamter 1.68- 2.00- 2.83 1.68- 2.00- 2.83 1.68- 2.00- 2.83
(mm) 2.00 2.83 < 2.00 2.83 < 2.00 2.83 <
Pellet loss
ratio (%)
Wedge-wire
clogging
ratio (%)
000 0.4 0.1 0 9.1 1.4 <0.1
0 0 0 20 5 0 40 20 <5
Note: The pellet loss ratio is the ratio of the pellet loss measured
after one hour operation to the pellet volume in the nitrification
tank. The wedge«wire clogging ratio is the ratio of the clogged
area to the entire wedge wire area. * .
Operating conditons:
Wedge-wire separation rate = 1.25m3/m2/min.
Air flow rate = 0.08m3/m3 - tank volume/min.
The effects of separation rates and air flow rates on pressure loss at
the wedge-wire are shown in Table 5. Approximately 2m3/hour of air flow
to unit volume of the aeration tank with 3-hour HRT is required for
biological treatment with nitrification. This is based on 25 mg/e influent
ammonium nitrogen and 15% oxygen transfer efficiency. The air flow rate is
equivalent to an aeration intensity of 1.28m3/m2-floor area/min at a
water depth of 4m. Unit area of wedge-wire separated 1.28m /min of mixed
liquor from a aeration tank with 10% pellets by volume at aeration
intensities between 0.08 and 0.16m3/m*/min. When the pellet volume was
increased to 20%, the pressure loss increased, and the separating capacity
of unit area wedge-wire decreased considerably. This means that a larger
wedge-wire area is required for effective separation of mixed liquor at
higher pellet volume ratios. The mixed liquor flow to be separated is equal
to the sum of the inflow rate and recycling rate when a nitrified liquor
recycling system is employed. A IN Process treating 10,000m /day of
wastewater at a 10% pellet volume ratio and 300% recycling rate requires a
wedge-wire separating system with a 20m area.
278
-------
Table 5. Effects of separation rates and air flow rates on
pressure loss at wedge-wire
Inflow rate
Air flow rate
0.96
1.28 1.60 1.92
m3/m2/min. m3/m2/min. m3/m2/min.
No aeration
x/x
0.08 m3/ma/min.
O/A
o/x
x/x
x/x
0.16 m3/m2/min.
o/A
o/x
o/x
0.24
o/o
p/A
x/x
0.40 m3/m2/min.
O/A
O/A
x/x
Note: Pellet volume ratio of 10% / Pellet volume ratio of 20%
& '
o: No operation problem
A: Pellets continue attaching and detaching to the wedge-wire
randomly. This causes, a sudden increase/release of pressure
loss. Due to the fluctuation, the separation'is not be
maintained at a constant rate.
x: Great pressure loss. Continuous operation is not possible.
Operating conditions:
Pellet diamter > 1.68 mm.
Wedge-wire opening = 0.5 mm
Fluidization of pelJLets
The fluidization test on the pellets was conducted under whole-floor
aeration in a pilot-scale nitrification tank (effective volume = 4.2m3).
The pellet volume ratio to the nitrification tank was set at 15 and 20%.
Air supplied at 1.25m /hour to unit cubic meter of tank volume fluidized
pellets uniformly under both conditions. The air requirement for fluidizing
pellets uniformly increased to 2.5m3/m3-tank/hour when the pellet volume
ratio was 30%. The boilogical air requirement for both BOD removal from and
nitrification of ordinary municipal wastewater is about 2m3/m3/hour at 3
hour HRT in a whole-floor aeration tank. This means that fluidization will
be achieved with air supply to meet the biological need when the pellet
volume ratio is less than 20%.
279
-------
Application of Biopellets to secondary effluent nitrification
*
Temperature largely influences the nitrification of wastewater. In
ordinary activated sludge plants, nitrification proceeds to some extent in
summer, but not in winter. Since partial nitrification often causes
effluent BOD5 to increase, nitrification must be carefully monitored even
in plants targeting only organic removal. Considering such problems in this
section, the application of Biopellets to secondary effluent nitirification
is discussed.
A pilot study was carried out using a 200 L reactor nitrifying
secondary effluent. The pilot plant was as shown in Figure 8. The treated
secondary effluent was pumped from a secondary settling tank in an activated
sludge plant to the pilot plant. Nitrified wastewater in the reactor
containing Biopellets was separated by wedge-wire.
Pellet separator
Effluent
Submerged Feed
pump tank
Secondary settling tank
Pump
compressor
Storage tank
Figure 8. Schematic flow diagram of the pilot plant nitrifying
secondary effluent.
280
-------
The operating conditions were as shown in Table 6. The pellet volume
ratio was 20%,at 2 hour HRT, and 30% at 1 hour HRT. Water temperatures were
between 10 and 12°C. Influent ammonium nitrogen concentrations were 20 to
30mg/e. The experimental results are shown in Figure 9. Complete
nitrification was achieved when the ammonia volumetric loadings were 0.2 to
0.4kg-N/m3/day at 2 hour HRT. Increasing ammonia loadings to a range
between 0.5 and 0.7kg-N/m3/day caused nitrification to .lightly
deteriorate. However, effluent ammonium nitrogen concentrations were still
less than 2mg/£, and more than 95% of the ammonium nitrogen was removed.
When the dissolved oxygen concentrations in the reactor decreased, the
nitrifying efficiency dropped. This might have been caused by oxygen
deficiency.
• ' a.
Table 6. Operating conditions for secondary effluent nitrification.
Run 1
Water temperature (°C) 10 - 12
Pellet volume ratio(%) 20
Retention time (hour) 2
Run 2
10 -12
30
1
Ammonium nitrogen
volumetric loading
(kg/m3/day)
0.2 - 0.4 0.5 - 0.7
281
-------
Run 1
Run 2
100
8 ^ 9°
Si 8°
2 3 7°
!<*;« 6°
IB o =t_
30
25
— 20
15
10
0000- 00Q 00
00 0 ,
_o_a
Influent
Dissolved oxygen
deficiency
Influent
Dissolved oxygen
deficiency
Jan/20 25 Feb/1
15 20
Figure 9. Results of secondary effluent nitrification.
282
-------
The nitrifying rates of the pellets used in the pilot experiment were
measured in a small batch reactor. The pellet volume ratio wS increased by
steps, from 10%, to 20%, to 40%. The results of the batch stSy are X *
shown xn Figure 10. The nitrifying rates were 147mg-N/L-pellets/hour It 10%
pellet volume ratio, 140mg-N/L/hour at 20%, and 83mg-N/L/hour at 40%
increasing the pellet volume ratio means decreasing the nitrifying rate per'
unit volume of pellets. Oxygen transfer to liquid may have controlled tnT
nitrifying rates of the pellets. In-addition, a larger area of wedgj-wirj
was required for pellet separation at higher pellet volume ratios
Therefore, the feasible pellet volume ratio should be limited to 20%
30
20
c
0)
10
Pellet volume ratio
^40%
Reagent :
Water :
temperature
PH :
Secondary effluent
10 toa2°C .
6.7 to 8
(Jan.- 26, '88, Jan. 27, '88)
Time (hr)
Figure 10. Results of batch'experiment.
fetent±0n *ime ^quired for'complete nitrification at 20% pellet
to ^S ? Y °ne h°Ur' ^^ °n thS *«** Stud^ However,^ is
that a Sman^/h OXy9f Concetration in ^ full-scale reactor as high
batCh react- Moreover, maintaining high oxygen
283
-------
Application of Biopellets to nitrogen removal (the IN Process)
Preliminary study
The -application of Biopellets to biological nitrogen removal processes
enables a Significant reduction in the required reactor volume. To confirm
the treatment performance and-to determine operation conditions a
preliminary study was conducted. Biopellets were applied to the nitrified
liquor recycling activated sludge process.
The reactor used was as shown in Figure 11. The reactor consisted of a
denitrification tank, a nitrification tank, and a final settling tank In
the nitrification tank, wastewater was mixed with return sludge, and the
nitrified liquor recycled. Denitrification occurred in the tank, and
simultaneously, some of the organics were consumed as carbon sources for
S^5Steatt£. The mixed liquor then flowed into the nitrification tank,
containing Biopellets. Organic removal by the activated sludge and
SSif ication S both the activated sludge and Biopellets proceeded at this
?ime The wedge-wire installed near the top of the tank-separated nitrified
Sor from thJ pellets. The separated nitrified liquor was partly recycled
to the denitrification tank, and the rest was introduced into a final
settling tank.
Dentrification Nitrification
tank tank
set
0 0
Nitrified
,.T liquor
I recycling
L___J^I_
Kf f 1 tient:
.Biopellets
(Pellet volume ratio:10%)
sludae
1.6m/day Return sludge
Figure 11. Flow diagram of IN Process for preliminary study.
The operating conditions are shown in Table 7. The treated wastewater
was the primary effluent of the municipal wastewater. The retention times
in the denitrification, nitrification, and final settling tanks were 3
hours, 3 hours, and 1.8 hours, respectively. The reactor volume requirement
was almost the same as that for the conventional activated sludge process.
The Sludge return ratio was 100%, and the nitrified recycling ratio was
300%. Twenty liters of pellets were added to the nitrification tank. The
pellet volume ratio was equivalent to 10%.
284
-------
Table 7. Operating conditions for preliminary study of the IN Process.
Nitrification tank
Denitrification tank
Final settling tank
Pellet volume ratio
Sludge return ratio
MLSS
Nitrified recycling
ratio
BOD volumetric loading
T-N volumetric loading
Effective volume: 2002, Retention time: 3
Effective volume: 2002, Retention time: 3
Effective volume: 1202, Retention time: 1.
hours
hours
8 hours
10% of the effective volume of the nitrification
tank
100%
2,000 to>3,000 mg/je
300% .
1.46 kg/m3/day to the denitrification tank
0.31 kg/m3/day to the nitrification tank
Note: Operating condition during winter: Water temperature is 15 to 16°C.
The experimental results are shown in Figure 12. The influent BOD*
suspended solids, and total nitrogen were 120 to 300mg/£, 120 to SMOg/I'-and
30 to 60mg/£, respectively. The average volumetric loadings during the
experimental period were 1.46kg/m3/day for BOD5/ and 0.31kg/m3/day for
total nitrogen. During the experiment, both the -BOD and suspended solids
were removed to 20mg/£, or less, and .the total nitrogen was reduced to less
than 10mg/2. The ammonium nitrogen concentrations remaining in the effluent
were less than 2mg/£. These results demonstrated that Biopellets work well
even in activated sludge, and that the IN Process is feasible as a
biological nitrogen removal process.
285
-------
I
•P
,s
0)
o
OH
CP s— '
300
200
100
Influent
Effluent
Q o
. o ' o
c
o
•H
U
•P
C
c^
O H
O)
(A E
300
250
200
150
100
50
n
Influent
-
*
A
• •
*
Effluent
O r> . n i O i O i
Dec/1 11 21 Jan/1 11 21
Time (day)
c
0
-rl
-P
to
4J
C
O
c
o —
O H
?!
t" -^
60
50
40
30
20
10
n
-
Influent
- * •
& •
• e
Effluent
o 0 • o oo
1 1 1 1
•
1
o
1
o
•P
to
I
8"
30
20
10
influent
Effluent
o
o
n i n
Dec/1 11 21 Jan/1 11 21
Time (day)
Average BOD load : 1.46 kg/m3/day (denitrification tank)
Average T-N load : 0.31 kg/m3/day (nitrification tank)
MLSS : 2,000 to 3,000 mg/Z
Water temperature: 15 to 16"C
Figure 12. Results of preliminary study on IN Process.
The sharing of ammonium nitrogen between the Biopellets and the
activated sludge in summer and in winter is shown in Figure 13. The
nitrification rates of the Biopellets and activated sludge were nearly equal
in summer. The rate of the activated sludge decreased in winter, while the
rate of the Biopellets increased and supplemented activated sludge. The
nitrification rate of the Biopellets in winter was lOOmg-N/L-pellets/hour.
This rate is not much lower than the figure obtained in secondary effluent
nitrification (140 mg-N/L-pellets/hour).
286
-------
•H
•P
s
§
o
2
Activated sludge (winter)
(MLSS:2000mg/l)
Activated sludge (summer)
(MLSS:2500mg/l)
(summer)
ratio:10%)
(winter)
volume ratio:10%)
0 1 2 34 5
Time (hr)
Figure 13. Sharing of ammonium nitrogen between Biopellets and
activated sludge.
Pilot plant study
. c^A3pilot plant study was initiated in November 1988. The plant capacity
is 50m /day. -The primary objectives of this study were:
a. to confirm the treatment performance throughout a year,
b. to clarify operational characteristics, such as fluidization and
separation of pellets and nitrified liquor recycling, and
c. to esatablish design and operation/maintenance procedures.
The pilot plant experiment will last to the end of this year. This
section introduces interim results obtained last winter.
287
-------
A flow diagram and the specifications of the pilot plant are shown in
Figure 14. Raw wastewater is pumped up from a municipal wastewater
treatment plant to the experimental yard in the same plant. After temporary
storage in a feed tank, the raw wastewater is settled in a primary
sedimentation tank. The primary effluent is treated by a scaled-down pilot
IN Process. The process configuration is alomst the same as that for the
preliminary study, but excludes the nitrified recycling system. The
nitrification tank is aerated, making the contents overflow into the
denitrification tank by air-lift effect, to thus recycle the nitrified
liquid. The volume of Biopellets. added into the nitrification tank is 10%
of the tank's effective volume.
«t
The operating conditions last winter are shown in Table 8. The
influent BOD5 and total nitrogen concentrations were 100 to 200mg/£ and 20
to 30mg/£, respectively. The detention times were 4.8 hours for
denitrification, and 3.2 hours for nitrification at the constant inflow rate
of 31.5m3/day. The air flow rate, sludge return ratio, and nitirified
liquor recycling ratio were controlled at 140L-air/min, 40%, and 300%,
respectively. The MLSS concentration fluctuated in the range between 2,000
and 3,OOOmg/2, and the water temperatures were between 10 and 12°C.
288
-------
3§
(0 -P
CN 10
X X CN
CO°°fE
c in ro «3
^ M CO pr)
>»
S
rH
0,
-H
a
•s
§
8
»w u/
58
0)
Ra
289
-------
Table 8. Operating conditions for pilot plant study of the IN Process.
Primary settling tank
Nitrification tank
Denitrification tank
Final settling tank
Pellet volume ratio
Water surface area: 2.9 m2,
Overflow loading : 10.9m3/m2/day
Effective volume: 4.2 m3,
Retention time : 3.2 hours
Effective volume: 6.3 m3,
Retention time : 4.8 hours
Water surface area: 3.8 m2,
Overflow loading : 8.3 m3/m2/day
7.5% of the effective volume of the
nitrification tank.
Nitrified recycling ratio , 40%
BOD volumetric
/day to the denitrification tank
The treatment performance is shown in Figure 15. The average
-.,
Scarification was almost completed. The transparency of the final
effluent was not as good as that from the conventional system. More
consideration should be given to, this point.
290
-------
200 r
150
100
50
Influent
Effluent
20 Feb/1 10 20 Mar/1 10
Figure 15. Results of pilot study on IN Process in Winter.
291
-------
Initial and 0/M costs for the In Process
The initial cost and the operation/maintenace costs for the IN Process
were estimated by the interim results, and were compared with those for the
conventional biological nitrogen removal process. Conditions for the
estimation are shown in Table 9, and the results are shown in Table 10.
Since the reactor volume requirement was reduced to half that for the
conventional system, the initial cost was reduced greatly. The estimate did
not include land cost. This means that the saving should be greater in
larger cities, where land costs are extremely expensive. Furthermore, one
feature on the IN Process is that it can save land area, which is the
biggest advantage for existing treatment plants. In fact, there is no space
for retrofitting advance treatment in most of the treatment plants in large
cities, nor surround them.
Table 9. Conditions for estimation of initial cost,
and operation/maintenance costs.
Conventional biological
IN Process nitrogen removal
process
Flow rate
10,000 m3/day
Influent BOD
120 mg/£
Effluent BOD
20 mg/£ or less
Influent NH4-N
25 mg/£
Effluent T-N
10 mg/£ or less
Lowest water temperature
13°C
Nitrification rate
'96 mgN/£-pellets/hour 0.67 mgN/gSS/hour
Denitrificatipn rate
1.04 mgN/gSS/hour
0.71 mgN/gSS/hour
MLSS
,3,000 mg/£
3,000 mg/£
292
-------
Table 10.
Estimates of initial and O/M costs for the IN Process.
IN Process Conventional biological
nitrogen removal process
Reactor volume (m3)
Retention time (hour)
Nitrification tank (m3)
. Retention time (hour)
Denitrification tank (m3)
Retention time (hour).
Pellet volume (m3)
Initial cost (yen)
Civil .works (yen)
Installations (yen)
Operation and maintenance
cost (yen/year)
Power requirement (yen/year)
Pellet cost (yen/year)
Repairs (yen/year)
Treatment unit cost (yen/m3)
3,330
8
1,330
3.2
2,000
4.8
. 100 (7.5%)
540,000,000
250,000,000
290,000,000
35,000,000
15,000,000
6,000,000
14,000,000
10
6,670
16.
4,170
10
2,500
6
-
790,000,000
480,000,000
310,000,000
46,000,000
31,000,000
15,000,000
13
Note: It is supposed that the air-lift r^von™ c,™i * _.•,. _.i_. _,
.— •—-~ -v,>^jr»^j.j.iiv.) oyocciu uj. xiitiririec
liquor is employed for the IN Process, while a pump recycling
system for the conventional biological nitrogen removal process.
Owing to the higher aeration strength of the nitrification tank
for the IN process (2.02 mVm3. tank/hour, versus 0.50 m3/m3
Process), a greater gas hold up in
is achieved. This simplifies recycling of
liquor by the air-lift method. As a result, the IN "
process requires no driving power for recycling. The estimation
is made in terms of the reactor only. Therefore, the above costs
do not include those for the primary and final settling tanks
293
-------
The estimate of O/M costs did not include the cost for personnel.
Therefore, the costs for agitating denitrification tanks, aerating
nitrification tanks, and maintaining mechanical and electrical devices are
dominant in the estimate. Additionally in the IN Process, the cost for
pellets had to be included because Biopellets would be damaged after a
long-term operation. The duration of Biopellts was assumed to be 5 years.
The results indicated that the O/M costs for the IN Process is less
expensive, (by 3yen/m3-treated) than that for the conventional process.
This is due mainly to the air lift recycling system of nitrified liquor,
which was applied for the IN Process. If pumps are employed for recycling,
the costs will be increase by 4yen/m3, and will be slightly more expensive
compared to the conventional system. From this viewpoint, the establishment
of the air-lift recycling system is important to promote, the application of
the IN Process as well as prolonging the duration of the Biopellets.
s *
Acknowledgements
The authors wish to express their thanks to the following people: Mr.
Aoyama, Deputy Director of the Kitano Sewage Treatment Plant of Hachiouji
City, for his valuable suggestions on the pilot study ongoing in his plant;
Mr. Mizuguchi, Mr. Mori, Mr. Emori, and Mr. Nakamura, co-researchers from
the Hitachi Plant Construction Co., Ltd., for contributing to the
development of the Biopellets and the IN Process; and Mr. Kyosai, Research
Coordinator, Dr. Sato, Section Chief and Dr. Mbriyama, Senior Researcher of
the Water Quality Control Division, Public Works Research Institute,
Ministry of Construction, for serving as advisors during the entire project.
294
-------
CURRENT STATUS AND FUTURE DIRECTION OF ORGANIC TOXICS CONTROL
IN THE WATER ENVIRONMENT «-
DISCHARGE REGULATIONS AND STREAM STANDARDS
by
William F. Brandes
Catherine Crane
Environmental Protection Agency
Office of Water
Permits Division
Washington, D.C.
This paper has been' reviewed in accordance with
The U.S. Environmental Protection Agency's peer
and administrative review policies and approved
for presentation and publication.
Prepared for presentation at:
Twelfth United States/Japan Conference
on Sewage Treatment Technology
Cincinnati, Ohio
October 12-13, 1989
295
-------
I. INTRODUCTION
ecosystems. problems associated with the discharge of
the difficulty of obtaining positive identification of these
analyses o e«luent an,
biological and chemical assessment methods.
II. IEGISLATIVE FRAMEWORK
?s ssss
water pollution control efforts across the country
296
-------
The emphasis changed from strictly water quality-based
standards (biological endpoint-based) to technology-based
standards (wastewater engineering-based). The introduction
of technology-based effluent guidelines shifted the burden
to the discharger to meet the discharge limits achievable
"through already existing, defined control technologies.
The NPDES permit program was established limiting the levels
of pollutants allowable in effluent discharges. The
program is under EPAs' authority and applies to all direct
dischargers. To date, NPDES permits have been issued to
over 65,000 dischargers. This includes more than 50,000
industrial dischargers and over 15,000 municipal dischargers
which release wastewater directly to surface waters. There
are as many as ah additional 100,000 "indirect" industrial
dischargers releasing wastewater to the sewer systems of
publicly owned treatment works (POTWs). These indirect
dischargers are regulated by the POTW through the national
pretreatment program.
Enforcement authority was succinctly defined by making
illegal all discharges to U.S. waters without a permit, and
allowed for civil and criminal penalties in cases of
noncompliance.
A national policy on toxics was established which prohibited
the discharge of toxic pollutants in toxic amounts.
o In 1977 Congress passed the Clean Water Act (CWA) to refine
some of the requirements originally set out in the 1972 law in order
to emphasize the control of toxic pollutants. Among the most
important modifications made to the 1972 requirements were the
revisions to the Effluent Guidelines and Pretreatment programs:
The National Effluent Limitation Guidelines program sets
forth uniform national standards for direct dischargers,
-developed on an industry-by-industry basis utilizing
existing control technologies. in an effort to stimulate
advances in toxic control mechanisms the CWA established a
list of 126 "priority pollutants", produced by 21 categories
of industries, which must be specifically addressed in the
development of effluent limitations guidelines. This list
includes 111 organic compounds, 13 heavy metals, asbestos
and cyanide.
The National Pretreatment Program regulates the introduction
of pollutants from nondomestic sources into sewer systems of
POTWs. ,
.o In 1987, amendments were made to the CWA, known as the Water
Quality Act (WQA), in order to further encourage control of and
research on toxic pollutants and their effects.
297
-------
- Because Congress recognized that progress on chemical
specific controls was slow, a dual approach was developed to
facilitate water quality improvement. Section 303(c)(2)(A)
of the WQA requires that States develop water quality
criteria based on biological monitoring or assessment
methods.
Under section 304(1) of the WQA"each State must prepare
lists of 'its waterways which do not meet water quality
standards. The States must develop individual control
strategies (NPDES permits), to reduce the amount of toxic
substances being released by point sources.'
Increased Congressional concern for toxics is also reflected
in amendments to section 405 which requires EPA to identify
toxic substances in sewage sludge and develop numerical
limits for maximum allowable concentrations. A total of 16
organic chemicals are to be controlled via numeric
limitations.
III. CURRENT STATUS OF ORGANIC TOXICS CONTROL EFFORTS
Technology-Based Standards
The CWA requires the development of national standards for
treatment technologies for direct dischargers, which are implemented
through the NPDES. permit program. Technology-based effluent limits
define a minimum level of control and are imposed at the point of
discharge or "end-of-pipe". Organic toxics are controlled through
Best Available Technology Economically Achievable (BAT).
The most significant effluent limitation guideline to be
developed to date for the control of organic toxics is the organic
chemicals, plastics, and synthetic fibers (OCPSF) effluent limitations
guideline issued in November of 1988. Typically, organic chemical
manufacturing facilities discharge more than 60 toxic substances in
their wastestreams. With at least 400 indirect dischargers and 300
direct dischargers nationally, organic chemicals facilities affect a
large number of receiving waters.
Water Quality-Based Standards
Existing technology-based permits require limits on a specific
set of known pollutants which are treated to a specific level by BAT.
This approach may not address complex effluents where combinations of
chemicals may have additive effects, or effluents containing toxicants
not found on EPA's priority pollutant list. Water quality-based
permitting is used in cases where limits more stringent than
technology-based are necessary to protect the designated use (such as
a fishable river usage) of the receiving waters.
Water Quality Criteria and Advisories. Using experience gained
by States and EPA Regions in toxic impact analysis, EPA is developing
new water quality criteria for high priority toxicants as well as
advisories for toxicants for which more limited information is
298
-------
available. For example, EPA is issuing a criterion for organotin
compounds because these compounds have been found to cause toxic
impacts in harbors where organotin paints are used to protect ship
^ ;>, ?ef c?lteria are bein9 developed for pesticides found: to
pass through typical treatment systems.
. . Whole Effluent Toxi city Toting is often a preliminary step in
the process of setting water quality-based permit limits for toxic
pollutants. Effluent toxicity is measured by exposing test organisms
to varying concentrations of an effluent/water mix to measure the"
aggregate toxic effect on living organisms.
,_ , Toxicity Reduction Evaluation (TRE) . A TRE is initiated if tho
whole effluent toxicity test indicates an unacceptable" re"ve"l of
toxicity. A TRE is a site-specific study conducted in a step-wise
process to narrow the search for effective control measures for
effluent toxicity. These studies are designed to identify the
causative agents of effluent toxicity, isolate the sources of effluent
toxicity evaluate the effectiveness of toxicity control options and
then confirm the reduction in effluent toxicity options, ana
Whole effluent toxicity tests have been used in detecting
tn t^/?flu;?ts °f POTWs across the ^tion. TREs hive then
identifying the causative agents of this effluent toxicitv
6' ^ nUmbSr °^ °a?eS organophosphate pesticides have been
as the cause of toxicity in POTW effluents. One such
compound, diazinon, has been repeatedly identified as either the
primary toxicant, or a substantial contributor to the effluent
t-oxic 1 t
ad
aided
Due to the, unexpected presence of toxic concentrations of '
diazinon in diverse effluents, the National Effluent Toxicity
Assessment Center (NETAC) has conducted a national survey of 28 POTW
inSS;K%b0? *?X1Clty ?aused by diazinon. The resultsof this study
indicate that diazinon is present in 61 percent of the POTWs tested
though toxic levels of diazinon tended to be restricted to the
southern region of the nation. The relatively widespread occurrence
of this compound in POTW effluents would likely have gone undetected
were it not for the use of toxicity testing and TREs. undetected
Pretreatment
„_„,., pretreatment programs require that industrial dischargers to
POTWs reduce the level of pollutants in their process wastewaters
Approximately 1500 of the 15,000 POTWs in the 5 . S . Sre ^iqSSed to
implement pretreatment programs as part of their NPOE?
-der
Prohibited Discharge Stand^r-ri
trucral daaae
structural damage;
include both general and specific
include prbh!bi?SJs agaSJ? ^
from any nondomestic user which cause pass
^.includes prohibitions agains?
V fir& °r exPlosion hazards; 2). corrosive
interference due to flow obstruction; 4)
299
-------
interference due to flow rate or concentration; and 5) interference
due to heat . .
categorical Pretrea^^-nt standards are incorporated into local
permit requirements to limit potentially hazardous discharges.
Categorical standards are developed from effluent limitations
guidelines and identify what level of a specific Pollutant ^ cause
interference with treatment systems and what amount of the pollutant
is removed by typical treatment processes.
T.ocal limits are utilized by the POTW because categorical
standards are minimum requirements which apply to broad classes of
industries and do not necessarily address all industrial discharge
problems that might occur at a given POTW. To prevent site-specific
problems, each POTW must assess all of its industrial discharges and
employ sound technical procedures to develop defensible local limits
that will adequately protect the POTW, its personnel and the .
local limits a POTW must conduct an industrial
waste survey to locate and identify all industrial users as well as
the character and volume of their pollutant discharges. The POTW
then determines which pollutants have the potential for interference,
pass through or sludge contamination. Then for each pollutant of
concern the POTW must determine the maximum loading which can be
accepted by the treatment facility without problems. The POTW must
then implement a system of local permit limits to ensure that the
maximum loadings will not be exceeded.
t
V. Future Directions '
As environmental monitoring and detection capabilities become
more sophisticated, the pervasiveness and consequences of toxic
pollution become increasingly evident. In response to our expanding
awareness of the impacts associated with toxic contamination, control
efforts have been refined to greater levels of efficacy and
comprehensiveness. In order to continue to progress in this area EPA
is both building on established programs and developing new
methodologies .
Technoloav-based effluent limitations guidelines are continuously
being developed for additional industries. Two effluent guidelines
currently under development will have considerable impact on organic
toxics discharges. These guidelines are for the pharmaceutical and
pesticide industries. Both of these industries discharge wide
varieties of organic toxics and few have met BAT control standards.
Water quality-based limits will continue to be a major focus of
pollution control. Point sources and other sources of pollution such
as storm water discharges will be given limits to protect local water .
quality.
Sewage sludge is produced at the rate of 7.6 million tons per
year in the U.S-. , and this rate is expected to double by the year
2000. Sludge may contain unacceptable levels of toxic substances,
300
-------
particularly heavy metals and organics. When properly treated,
sludge can be beneficially used as a soil conditioner and fertilizer.
EPA encourages beneficial reuse of sludge whenever environmentally
feasible. To this end, EPA is developing regulations and procedures
for identifying, managing and reducing the amount of toxic pollutants
in municipal sludge based on sludge use and disposal practices.
Bioconcentratable Contaminants are compounds that have the
ability, due to their chemical properties, to accumulate in fish-and
shellfish tissues in concentrations hazardous to human consumers.
Water quality criteria and effluent guidelines designed to prevent
acute and short-term toxicity to aquatic life may not be sufficient to
protect human health. EPA is exploring methods that will allow
environmental managers to identify and quantitate bioconcentratable
pollutants in effluents, and develop acceptable ambient criteria,
wasteload allocations and effluent limitations for these pollutants.
Sediment Assessment is utilized to evaluate toxic contaminants in
the bottom sediments of lakes, rivers and coastal waters for potential
adverse impacts to biota and water quality. Historically, water
quality-based standards have not been developed to directly protect
sediment quality or the beneficial uses associated with uncontaminated
sediments. Sediment assessment is being analyzed as a means to
target dischargers and develop more stringent effluent limits.
Pollution Prevention reflects EPA's policy to reduce the amount
and toxicity of pollutant's being discharged, either directly or
indirectly through POTWs, to waters of the U.S. Increasingly
stringent controls and better detection methods encourages dischargers
to begin in-plant recycling programs or to substitute less toxic
compounds in their production processes to reduce the levels or
toxicity of their discharges.
301
-------
-------
PERFORMANCE OF FINE PORE AERATION SYSTEMS IN PROCESS WATER
by
«-• .. WffliamC. Boyle, University of Wisconsin-Madison
Richard C. Brenner, U.S. EPA, Risk Reduction Research Laboratory
James J. Marx, Donohue and Associates
Thomas C. Rooney, Sanitary Engineer
This paper has been reviewed in accordance with the U.S
Environmental Protection Agency's peer and administrative
review policies and approved for presentation and publication
Prepared for Presentation at:
Twelfth United States/Tapan Conference
on Sewage Treatment Technology
Cincinnati, Ohio
October 12-13,1989
303
-------
INTRODUCTION
cement of less efficient action systems with fine pore aeration devices can save up to 50
. at in aeration energy costs and has resulted in typical simple pay back periods of two to six
ware As aresult of these very impressive cost savings, more than 1,300 municipal and industrial
Slwater Sent facilities* the United States and Cf ada now use fine pore aeration. In
™°5 the U S Environmental Protection Agency (EPA) funded a CooperativeResearch
men
oiifine pore diffused aeration systems in both clean and process waters, conduct field studies at a
number& municipal wastewater treatment facilities employing fine pore aeration and prepare a
comprehensive manual addressing the design, operation, maintenance, control, and economics ot
fine pore aeration systems.
A Design Manual on Fine Pore Aeration was produced from this effort by U.S. EPA(2).
This presentation summarizes a portion of this manual which discusses the performance of fine
pore aeration systems under process conditions.
The term "fine bubble" diffused aeration is elusive and difficult to define. ^ term "fine
tjore" was adopted to more nearly reflect the porous characteristics of the high efficiency diffusers
mariceSoda?Ty^Sy, fine pore diffuserswill produce a headloss due to surface tension m
S^wlterof greater thai about 5 cm water gauge (w.g.). For the purposes of this paper, fine
pore diffusers are defined as including the following devices:
• Porous ceramic plates, discs, domes, and tubes
• Rigid porous plastic plates, discs, and tubes
• Nonrigid porous plastic tubes
• Perforated membrane tubes and discs
FACTORS AFFECTING PERFORMANCE
*
The performance of diffused aeration systems under process conditions is affected by a
myriad of factors, some of the more important of which are:
• Changes in performance due to fouling, aging, fatigue, etc.
• Wastewater characteristics
• Loading conditions
Process type and flow regime
Basin geometry and diffuser placement
Diffuser performance and characteristics
Mixed liquor DO control and air supply flexibility
Mechanical integrity of the system
Quality of preventive maintenance
304
-------
Manual of Practice FD-13 (3) is a good general reference on the importance of the above
tactors To minimize life-cycle costs of an aeration system, all of these factors must be considered
during design and some must be controlled during operation.
The areas of greatest concern in process water oxygen transfer performance of fine pore
aeration systems are changes in performance due to fouling, aging, etc.; wastewater characteristics;
process type and flow regime; and loading conditions. In a given case, any combination of these
these factors can have a significant effect on the alpha (a) profile of a system, DO control and
changes in aeration system performance with time due to diffuser fouling. Some of these factors
are discussed in greater detail later in this paper.
CALCULATION OF PROCESS PERFORMANCE
A substantial data base exists on the perforrhance of fine pore diffusers in clean water In
designing aeration systems to operate under process conditions, clean water data are corrected to
account for the influences of process water characteristics,«temperature, and pressure. The
corrections are made using the following equations:
OTFf = KLa20 OF 0CT-20) (T p
. where,
- Q V
(1)
OTRf
KLa
V
C*oo20 -
c
a =
P -
ecr-^) =
T
a
Q approx. =
Pb
gamma =
PS
PVT =
C*ST =
oxygen transfer rate under process conditions, mass/time
apparent volumetric mass transfer coefficient in clean water I/time
KLa at 20°C, I/time
volume of water, length3
steady-state dissolved oxygen (DO) saturation concentration attained at infinite time
for a given diffuser at 20<>C and 1 atm, mass/length3
process water DO concentration, mass/length3
(process water Ki,a of new diff user)/(clean water KLa of new diffuser)
(process water C*W(clean water C*oo)
temperature of process water being treated
(Pb + gamma dg - P\rr)/(Ps + gamma dg - PVT)
Pb/Ps (for basin depths <6.1 m or for elevations <600 m)
atmospheric pressure at aeration system location, mass/Cength x time2)
weight density of water at the test water temperature, force/length
standard atmospheric pressure (1.0 atm of air at 100% relative humidity)
mass/(length x time2)
saturated vapor pressure of water at temperature T, mass/(length x time2)
effective depth at infinite time, length
tabular value of DO surface saturation concentration at temperature T and standard
total pressure (1 atm at 100% relative humidity), mass/length3
(process water KLa of a diffuser after a given time in service)/(KLa of a new
diffuser in the same process water)
Since standard oxygen transfer rate in clean water, SOTR, is:
(2)
305
-------
Equations 1 and 2 may be combined to calculate the process water oxygen transfer rate, OTRf:
OTRf = aF(SOTR)0T-20(tpQC*oo20-C)/C*co20 <3>
This equation can be rearranged as follows:
OF(SOTR) = (OTRf C*o<,20 e^/Cr P Q C* C = 0).
Further, since standard oxygen transfer efficiency, SOTE, may be defined as:
SOTE = SOTR/Wo2 . (5)
where,
Wo2 = mass flow rate of oxygen, mass/time
Equation 3 may also be used to describe process water oxygen transfer efficiency, OTEf, by direct
substitution of oF(SOTE), oxygen transfer efficiency under field conditions corrected to.standard
temperature and pressure, and a driving force of C*«>20» fc>r ctF(SOTR).
Although employing clean water SOTR values to estimate oxygen transfer rates in process
water is conceptually straightforward, the estimate of OTRf is subject to considerable doubt
because of the uncertainties contained in a and R These uncertainties are magnified when toe
process water application is based on a basin geometry and process temperature that differ from
those of the clean water test.
Table 1 is a guide for applying Equation 3 and indicates the source of information for the
parameters needed to estimate OTRf.
DIFFUSER FOULING
All fine pore diffusers are susceptible to waterside buildup of biofilms and/or deposition of
inorganic precipitates that can alter the operating characteristics of the diffusers. Porous diffuser
media may also be susceptible to air-side clogging of pores due to particles in the supply air. Air
diffusion media may also suffer changes in properties, and the integrity of the installation may
change with time. These changes, which include alteration of diffuser media properties over time
and leaks in the air piping or around diffuser gaskets, may be caused by type of materials selected,
poor equipment design, improper installation, or inadequate inspection and maintenance.
The rate of diffuser fouling has historically been gauged by the change in backpressure while
in service. Since significant levels of fouling can take place with little or no increase in
backpressure but with substantial reductions in OTE, this provided a crude and qualitative measure
at best.
Better methods of measuring the degree of diffuser fouling and the effectiveness of diffuser
cleaning became available in the early 1980s, These methods include dynamic wet pressure
(DWP), bubble release vacuum (BRV), the ratio of one to the other, and chemical and •
microbiological analysis (4). The practice of employing pilot diffusers that could be removed from
the basin and individually analyzed also came into use during this period. DWP is defined as the
306
-------
pressure differential (headless) in cm w.g. across the diffusion element when operating in a
submerged condition at some specified airflow rate. BRV is a measure of the negative pressure in
cm w.g. required to form and release bubbles in tap water from a localized point on the surface of a
thoroughly-wetted porous element Normally, BRV denotes the average of the point BRV values
for a specific diffuser.
Concurrently, better methods were being developed to measure the oxygen transfer
performance of operating aeration systems. These methods permitted better appraisal of the effects
of fouling and facilitated improved preventive maintenance scheduling. These methods include
inert gas tracers, off-gas analysis, a nonsteady-state technique that uses hydrogen peroxide, and
DO and respiration rate profiles (5-8). Off-gas analysis equipment has been effectively used to
evaluate the adverse effects of fouling on both full-scale diffused air systems and individual
diffusers (5). For the U.S. EPA/ASCE Cooperative Agreement on fine pore aeration (2), off-gas
analyses were used in all field studies. All investigators conducting field studies on this project
were trained in the use of this equipment, all off-gas analyzers were of the same design^ and
calculation methodology was standardized.
TYPES OF FOULING
Fouling can be classified as one of two general types, Type I or Type II. The two types have
distinct characteristics and can occur alone or in combination with variable dominance from
treatment plant to treatment plant and within the same treatment plant from time to time.
Type I Fouling
Type I fouling is characterized by clogging of the diffuser pores, either on the air side by
airborne particulates or on the liquid side by precipitates such as metal hydroxides and carbonates.
In the process of fouling, the areas of the diffusers with the highest local air flux foul more rapidly.
This serves to reduce the flux in the high flow areas and increase it in low flow areas. The
combined effect is to improve uniformity of air distribution. As fouling progresses, the BRV
^coefficient of variation decreases (9).
Eventually, the accumulation of foulant in the pores reduces the pore size and DWP rises
correspondingly. The increase in DWP can exceed the capabilities of the air supply system, and
process air delivery may fall short of requirements. Also, the reduced effective pore diameters
produce smaller bubbles such that OTE does not decline and can actually increase slightly. A
schematic representation of Type I fouling is illustrated in Figure 1. An idealized plot of how OTE
and DWP change with time is shown in Figure 2.
Tvoe n Fouling
Type n fouling is characterized by the formation and accretion of a biofilm layer on the surface
of the diffuser.
Figure 3 is a model of Type II fouling proposed by Costerton (10) based on scanning electron
microscopy data collected on the biofilms that had thicknesses >1 mm. The microscopic work
showed that the biofilms were composed of bacterial cells enmeshed in a matrix of their own
amorphous exopolysaccharides. Inorganic particles were trapped within the bacterial matrix. This
composition has been observed in other biofilms found in natural and industrial aquatic systems
(11). When a biofilm of this structure develops on a fine pore diffuser, the bubble release surface
changes dramatically. Preliminary studies (10) indicate that bubble sizes increase, which may
significantly reduce OTE.
307
-------
Thin biofilm layers are usually not a problem, but as the biofilm thickness increases, BRV and
its coefficient of variation will increase, DWP will increase slightly or not at all, and OTE will
decrease substantially. The net result of nonuniform bubble release and an increase in bubble size
is a substantial reduction in OTE. Figure 4 is an idealized representation of OTE and DWP
changes with time under Type n fouling conditions.
EFFECTS OF FOULING ON PERFORMANCE
As indicated above, fouling processes may affect BRV, DWP, and OTE. Based on fouling
studies at nine municipal plants (12), the effects of diffuser fouling were divided into three
categories as shown in Table 2. the diffusers from the same plant could fit into more than one
category depending on the time in service and operating conditions of the plant.
Equation (1) introduced a new term, F, that describes the impairment of diffuser performance
caused by diffuser fouling or diffuser media deterioration. It has been introduced to separate the
effects of wastewater characteristics (a effects) and diffuser fouling (F effects) on the estimateof
process water oxygen transfer. F generally decreases from a value 1.0 with time in service. The
characteristics of fouling dynamics are site and diffuser specific as hypothesized in Figure 5. At
this time, a linear model describing fouling dynamics has been introduced in the design manual (2)
and that conceptual model is shown in Figure 6. The slope of the F vs. time curve represents the
fouling rate factor, fp, expressed in terms of a unit decrease per month. Therefore:
fF = (F-1.0)/t = [cxF(SOTE)o-ocF(SOTE)t]/aF(SOTE)ot (6)
This model assumes that there is a minimum value of F, Fmjn, that occurs after some critical
service period, tc. It may be reasonably assumed that Fmin approaches the ratio of the oxygen
transfer performance of a coarse bubble diffuser to that of an unfouled fine pore diffuser in
question, both diffusers being in the same physical basin configuration and operating under the
same process loadings. *
F and its functional relationship with time can be estimated by:
• Conducting full-scale OTE tests in process water over a period of time
• Monitoring aeration system efficiency using operational data
• Conducting OTE tests on fouled and new diffusers, ex situ, in an aeration column (2)
The results of preliminary tests conducted at municipal plants under full-scale monitoring of
OTE appear in Table 3. It should be emphasized that these data are very preliminary and based on
a simplified assumption of linear reduction of F with time in service to some minimum value.
MEDIA EFFECTS ON PERFORMANCE
Fine pore aeration systems have sometimes exhibited reductions in OTE that could be
attributed to mechanical failures or material changes rather than a or fouling effects. A number of
examples have been cited where oxygen transfer performance decreased due to gasket failures,
broken support bolts, and cracked plenums and air piping (2).
Changes in diffuser media properties can also affect DWP and OTE. Plasticized PVC (a
thermoplastic elastomer) and EPDM (a thermoset elastomer), the two principal perforated
membrane materials in use, can experience various physical property changes with time when used
as wastewater aeration devices. These changes, however, may differ significantly in nature and
degree between the two materials. Specific, comprehensive descriptions of the materials employed
in media production are therefore essential for prediction of perforated membrane diffuser
308
-------
performance. Conditions that can substantially affect perforated membrane performance and life
includeaoss of plasticizer, hardening or softening of the material, loss of dimensional stability
through creep, absorptive and/or extractive exchange with wastewater, and chemical changes
resulting from environmental exposure. Plasticizer migration can cause hardening and reduction in
membrane volume, resulting in dimensional changes. Absorption by the membrane of various
constituents, including oils, can result in softening of the membrane with volumetric increases and
consequent dimensional changes. Membrane creep, which may be influenced by the above
factors, will reduce OTE in some cases. It may also be accompanied by a reduction in DWP after
cleaning to lower than the original value. This reduction is not recoverable by known maintenance
procedures.
Reported data on performance and changes in characteristics of perforated membranes under
service conditions, although limited, are becoming increasingly available. Table 4 presents the
results of some of these data (2).
EFFECT OF WASTEWATER ON PERFORMANCE
The literature is replete with studies on the effects of wastewater characteristics on oxygen
transier. Both a and the nature and dynamics of fouling are attributed to wastewater properties.
fcurtactante (surface active agents) are believed to play an instrumental role in the depression (and
occasionally the increase) of a in wastewaters. The rise in a as treatment progresses down the
length of plug flow aeration basins has been attributed to surfactant removal from the wastewater
Other wastewater properties may also have an impact on a, including total dissolved solids and
transition elements such as iron and manganese (2).
It is instructive to note that oF values obtained from a significant number of in-process studies
Ait? are wer l ^^ anticipated for fine pore aeration systems in municipal wastewater
Although oF values presented in this table are dependent on several process and design variables
tor ttie specific treatment plants tested, it is apparent that the average mean weighted oF is usually
^\J* J« -
The variability of oF is site specific. Examples of typical variations over a 24-hr period are
presented in Table 6. A review of oF variations at these sites revealed a maximum oFraverage oF
of about 1.2, with a range of 1.08-1.47. The average value of minimum oFraverage oF was
approximately 0.86, with a range of 0.77-0.96.
PROCESS LOADING EFFECTS
. •»
A review of the dynamics of oF(SOTE) in a variety of activated sludge systems suggests that
several process variables affecting oxygen transfer are not clearly identifiable based on our current
knowledge of the process. For example, oF(SOTE) data collected at Madison, WI (13) over an
«UU-day period in the first pass of a three-pass plug flow system demonstrate a significant
variability in oF(SOTE) with time. Some of this variability may be attributed to the properties of
the wastewater (i.e., composition and strength), but cannot account for all of it Multiple linear
regressions of the data including sludge retention times (SRTs), food-to-microorganism (F/M)
loadings, volumetric organic loadings, mixed liquor volatile solids (MLVSS) concentrations and
air-flow rates could account for up to about 60-70 percent of the variability. Similar findings were
1^™ ^ Whittier Narrows treatment plant (14), where 30-74 percent of the variability in
oF(SOTE) could be accounted for by F/M loadings, airflow rates, and time-in-service.
These studies suggest that elements of process loading may have some impact on diffuser
performance. There are theoretical reasons why oF should be a function of SRT, F/M loading, or
MLVSS concentration. Current biological treatment models developed for activated sludge predict
309
-------
substrate concentration as a function of these parameters. Since the substtates are partially
comprised of surfactants, lower substrate concentrations imply lower surfactant levels and higher a
values. The dynamics of biomass accumulation and depletion on diffuser surfaces will very likely
influence F. The impact of these process loading factors on aeration system performance may be
short or long term depending on their combined effects on both a and F.
Data from Table 5 were analyzed to determine whether a relationship between process loading
and oF(SOTE) may exist. A plot for the ceramic diffuser installations in this table, using SRT as
the loading parameter, is given in Figure 7. Although wide variations in system design and
operation, as well as wastewater characteristics, are evident at these sites, it appears a trend does
exist between process loading and oF(SOTE). Nitrifying treatment plants have been highlighted in
this figure to indicate their relative importance to the relationship.
FLOW REGIME EFFECTS
Aeration basin flow regime affects the mixing pattern of the basin and, therefore, the residence
time distribution of the influent wastewater. Because wastewater components may have an impact
on oF(SOTE), it is reasonable to expect that mixing patterns will also affect. oF(SOTE). A study
conducted on the Madison, Wisconsin, ceramic dome diffuser system (13) illustrates this concept
(Figure 8) Single-day oF(SOTE) off-gas profiles are compared as a function of grid position tor
one aeration basin when it was operated first in a step feed mode vs. two months later when it was
operated in a plug flow pattern of three passes in series. The switch from step feed to plug flow
was made approximately half way between the two off-gas test days. In both cases, SRT was
approximately 2.2 days.
The oF profile along the length of any aeration basin will depend on the degree of mixing that
exists in that basin. Typical results for a variety of basin geometries are presented m Table 7. The
site key given in the first column refers to the treatment plants described in Table 5. In general, it
has been found that treatment plants with high length-to-width ratios, operating as plug flow
basins, generate significant oF gradients. Conversely, treatment plants that are short and wide or
that employ step feed configurations exhibit much less variability in oF along the basin length.
DIFFUSER FLOW RATE EFFECTS
The OTE of a fine pore diffuser in clean water decreases as the airflow rate jncreases. Similar
results are reported for process waters (2). A plot of oF(SOTE) versus airflow rate per unit media
surface area for selected municipal plants listed in Table 5 shows this trend (Figure 9). This curve
should not be used for design purposes, since many other variables, as described above, attect
oxygen transfer performance.
CONCLUSIONS
Field research over the last three years has significantly improved our understanding of fine
pore aeration systems. Based on the information available at this time, the following conclusions
can be formulated.
• Fine pore aeration system performance is significantly influenced by diffuser fouling as well
as by changes in media properties. . ,
• Attempts to quantify the effects of diffuser fouling and media changes have been initiated
using the fouling factor, F, and fouling rate, fp.
310
-------
w vaae omnance rom ste to ste and tim
r of factors effect the performance of a fine pore aeration system under process
)Sudf g Ae wa?tewater, process loading, flow regime, and airflow rate.
(F/M) of an operating plant appears to affect fine pore aeration system
106* An mcrease in SRT (decrease in F/M) generally will result in an increase
1. Wesner, G.M., LJ. Ewing, T.S. Linecfc, Jr., and DJ. Hinrichs. Energy Conservation in
Municipal Wastewater Treatment. EPA-430/9-77-01 1, NTIS No. PB81-165391 U S
Environmental Protection Agency, Washington, D.C., 1977.
2. Design Manual, Fine Pore Aeration Systems. EPA/625/1-89/023, U.S Environmental
Protection Agency, Office of Research and Development, Risk Reduction Engineering
Laboratory, Cincinnati, Ohio, September 1989. 6.
3 ' 1988ti°n' ManUal °f Practice FD~13' Water Pollution Control Federation, Washington, D.C.,
4. Boyle, W.C. and D.T.Redmon. Biological Fouling of Fine Bubble Diffusers: State-of-Art
J. Env. Eng. Div., ASCE 109 (EE5):99 1-1005, October 1983.
5 ' Er^?? • DX>T W-'C £?yle' and L' EwinS- Oxygen Transfer Efficiency Measurements in
Mixed Liquor Using Off-Gas Techniques. JWPCF 55(1 1): 1338- 1347, 1983.
6 . Boyle, W.C. and H. J. Campbell, Jr. Experiences with Oxygen Transfer Testing of Diffused
Air Systems Under Process Conditions. Wat. Sci. and Tech. 16 (10/11):91-106, 1984.
7* S°VJ£' J-S-'JJ- McKeown, D. Krause, Jr., and B.B. Benson. Gas Transfer Rate
Coefficient Measurement of Wastewater Aeration Equipment by a Stable Isotope
Krypton/Lithium Technique. In: Gas Transfer at Water Surfaces, Brutsaert, W. and G H
Jirka, Editors, D. Reidel Publishing Co., Dordrecht, Holland, 1984.
8. Mueller, J.A.,R. Sullivan, and R. Donahue. Comparison of Dome and Static Aerators
Treating Pharmaceutical Wastes. In: Proceedings of the 38th Industrial Waste Conference
Purdue University, West Lafayette, IN, May 1983.
9 . Rietii, M.O, W.C. Boyle, and L. Ewing. Effects of Selected Design Parameters on the
K>ulmg of Ceramic Diffusers. Presented at the 61st Annual Conference of the Water
Pollution Control Federation, Dallas, TX, October 1988.
10. Costerton, J.W. Investigations into Biofouling Phenomena in Fine Pore Aeration Devices
Study conducted under Cooperative Agreement CR812167, Risk Reduction Engineering
Laboratory, U.S. Environmental Protection Agency, Cincinnati, OH (to be published).
lj JAV' et ^ Bacterial Films in Nature and Disease. Ann. Rev. Microbiol. 41:435-
464, 1987.
311
-------
19 Raillod C R and K Hopkins Fouling of Fine Pore Diffused Aerators: An Interplant
1 ' C^ptiS' ^mdy conS under Cooperative Agreement CR812167, Risk Reduction
Engineering Laboratory, U.S. Environmental Protection Agency, Cincinnati, Ohio (to be
published).
13. Boyle, W.C. Oxygen Transfer Studies at the Madison Metropolitan Sewerage District
Facilities. Stady conducted under Cooperative Agreement CR812167, Risk Reduction
Engineering Laboratory, U.S. Environmental Protection Agency, Cincinnati, Ohio (to be
published).
14. Stenstrom,M.K. Fine Pore Diffuser Fouling: The Los Angeles Studies. Study conducted
under Cooperative Agreement CR812167, Risk Reduction Engineering Laboratory, U.S.
Environmental Protection Agency, Cincinnati, Ohio (to be published).
312
-------
Parameter
SOTR
de
C
T
t
n
a
P
e
F
TABIJE 1
Guide to Application of Equation 3
Source of Information
Clean water test results.
Clean water test results.
Clean water test results.
Process water conditions.
Process water conditions.
Calculated based on tabulated DO surface saturation values.
Calculated based on site barometric pressure and effective depth data.
Estimated based on experience or on measured values of KJJI in clean and
process'waters using a clean diffuser.
Calculated based on total dissolved solids measurements.
Taken as 1.024 unless experimentally proven to differ.
Estimated based on field experience, field measurements, or laboratory
analysis of diffusers taken from the field.
TABLE2
Fine Pore Diffuser Fouling Categories (2)
Condition
Severe
Moderate
Light
BRV
(cm, w.g.)
>100
40-100
<40
DWP/BRV
<0.3
0.3-0.6
>0.6
F (After 1 year in service)
<0.7
0.7-0.9
>0.9
313
-------
TABLES
Fouling Rates (fp) and Fouling Factors (F) for Selected Treatment Plants (2)
City • Plan!
Frankenmuth. Ml
Green Bay, WI
(2nd operating
period)
Green Bay. Wt
(1st 6 months
rv^K/l
uiuyj
Milwaukee. WI;
Jones Island East
Milwaukee, WI;
South Shore
Madison. WI;
Nine Springs
Monroe. WI
Los Angeles
Comity. CA;
WhitUer Narrows
RkJgewood, NJ
Tmie HI
Service.*
years
3.3
3.0
3.3
6.0
15.0
3.5
3.0
9.3
6.0
Test
Period.
months
13.0
11.7
6.5
6.5
6.0
6.0
30.0
56.7
26.7,
4.5
24.0
5.0
No.
Tests
17
16
9
9
5
5
20
20
37
5
23
48
IF.
month-1
0.029
0°
0.064
0.067
0.028
0.046
0.002
QC
oc
0.043
0.027
0.074
Correlation,
r
•0.52*>
-0.43
-0.66«>
•0.56
-0.65&
-0.22
-
-
•0.81
-0.82t>
-0.15
F(@1
month)
0.97
1.0
0.94
0.93
0.97
0.95
1.0
1.0
1.0
0.96
0.97
0.93
Comments
Ceramic disc diffusers; Basin 5
Ceramic disc diffusers; Basin 6
Ceramic disc diffusers: contact basin
Ceramic disc diffusers; reaeration basin
Perforated membrane tube diffusers;
contact basin
Perforated membrane tube diffusers;
reaeration basin
Ceramic plate diffusers
Ceramic plate diftusers; various basins
Ceramic disc diffusers
Ceramic disc diffusers during bypass of
aerated flow from equalization basin
Ceramic disc diffusers injected with HCI
gas 5 times about every 3 months
Ceramic dome diffusers cleaned often
by hosing or brushing with HCL solution
•As Of 4/1/89.
*> Significant at 95 percent confidence level.
c Zero fouling rale based on visual inspection.
314
-------
00
.5
^ -p
S I
a 2
M
S
«
i
S
S]
II
.c
po»oj K>PJO q«cxi oeo on ca i '^w^
5
o.
55
Q- Q-
S
£
11 11 || |£ is-
u. u. U. U. CL (L (L (L n n
OJ CO
S S
8 5515.1
I
S S §
CO v
o *^
;-
o n
od o o d
S
6
in
o t <3 S
w
CM
» p> •* co
2^ S£
1 • .' |
III
I I
I =
I J
r* . co
I 1
3
S5' o* o^
m 2- O- S **• 2,
^ fill II II II li
v
•i'
1= H
m o Q uj
1 .8
£ 0
b
M
b
I
" i
^o-S E . 5
. c = »
«llf!S£*
315
-------
2
(9
'I
u
I
If
"1
A
i
-a.;
||
fir-
3
£
| 1
C
t
t
X
I
0 j
:
1 =
u7 -
l-
8
Si
DT
a .
I
a. o> e» « g
S'S S 5 S
d d d d d
Io o o «o co
* w o ^ w
CM CB 0 tO K
in n 10 »• r*
d d d d d
in os o> CM CM
c> «r « to h-
S CM CO CM 10
d *- d d d
co to en co CM
o o o o o
O CO CM to O
oo ri co co CM
d d d o o
GO <* ^ CO
v i?) tZ) * .
dodo
o in to •*
CO CO CO CM .
o o o o
N CO 0) CO
co •* •* co .
o o o o
in v- 01 « to
d d r^ d ^
co r- *r to r-
in in to **• co
o o o o o
CO *- CO * en
r* r» cd to o
d d d d *-
}jjj
§ m to p S
1 i 1 1 1
gees*
u. O O I i
i- CM co w in
•- •- to in in •» « «
^ ^ CM r* co co
d d •- d r-' d
CO CM O CO «-; *l
co CM «r en co r^
CM CO CM in » h-
CD co *- •- CM o
d d d d d d
«- to CM to p co
t-. S r- i- «- 00
»- »- T- *- O
d d d d d d
h« CO C5 O OI t
»- , CM cVj CM CO »-;
d d o d d d
f~ O> »- 00
. in r* in in .
d d d d
, S S 5 8 .
d d d d
5 § ? ? .
d d d d
*r jf to co ** «o
r^ ^ »^ O O "-1
^ ^ N f^- OO CD
•» in oi co co o>
d d d d d o
h« *• •* CM ix in
K i^ *- en co CM
d d •-' d d -
•
•* ? s 1 -
g-l n i 1 2- g
= »
a. -.
d o
10 CO
CM" N
5 «.
o o
§ s
co in
co in
d d
« CM
O 0
in CM
CM CO
,d o
0 01
T «
o o
CM r
d d
S Si
d d
S g
d d
d d
CM in
5 2
u o
if t
S| 8
si s
oZ o
2 si 1 1 1 i il II 1
«o N- co cn o *-
CM CO
r- r-
" -
I-
CMIO<->-. d
CM CM
d d
•
CM *
n o>
rl O
a co
CO CM
d d
S ^
d d
rt
d
o
* CO
d
d
fe o.
d ^
t* CM
CM (O
d d
5 S
d d
» < - t-
III
e o o
.2 .»" .Sf S
1 1 1 \
S S S S
CM CM CM CM
«»
O
S E5
CM 0
CM CM
d d
O) «
d d
d d
. to
o
CM
d
to
d
S R
d d
ft 5
o o
to In
o o
U)
2 f
CC §
i" §
=) O
CM CM
I
O
CD
o
<0
Sm <
316
-------
TABLE6
24-Hr OF and oF(SOTE) Variations at Selected Municipal Treatment Plants
Site
No.
4
g
10
12
12
13
17
17
Cily • Plant
Hartford, CT ~
Madison, Wl;
West
RkJgewood, NJ
Los Angeles'
County. CA;
WhitUer Narrows
Los Angeles'
County, CA;
Whittier Narrows
Los Angeles
County, CA;
Whitlier Narrows
Monroe, Wl
Monroe. Wl
— ' ' __
Average
•• i
0.30
0.24
0.46
0.25
0.26
0.4S
0.23
0.39
aF
Minimum
0.23
0.22
0.44
0.21
0.20
0.41
0.19 '
'0.33
Maximum
0.44
0.29
0.59
0.27
0.30
0.50
0.28
0.45
Average
8.3
8.7
10.7
7.8
8.7
12.2
•
oF(SOTE)
Minimum
6.4
7.7
9.5
6.4
6.6
11.1
.
Maximum
11.2
10.4
13.1
8.7
9.9
13.5
'
Position in Basin
Influent pass
Inlet end
Entire basin weighted
Influent grid
Middle grid
Effluent grid
' Influent pass
Effluent pass
' Date lor 6-hr period.
317
-------
TABLE?
oF Profiles for Various Aeration Systems^-6
Site
No. City - Plant
2 Green Bay. Wl
3, Green Bay. Wl
4 Hartford. CT
5 Milwaukee. Wl;
Jones Island East
0 Milwaukee. Wl;
Jones Island West
7 Madison. Wl;
Easl
8 Madison, Wl;
East
9 Madison. Wl;
Wesl
11 Milwaukee. Wl;
Sooth Shore
12 Los Angeles
County. CA;
WNtUjr Narrows
13 Los Angeles
County, CA;
Whittjer Narrows
18 Phoenix, A2;
23rd Ave.
21 Minneapolis, MN;
Metro
22 Minneapolis. MN;
Metro
23 Minneapolis, MN;
Metro
26 Glastonbury, CT
Mean
0.45
0.49
0.37
0.45
0.34
0.32
0.40
0.33
0.64
0.25
0.16
0.29
0.36
•
0.50
0.59
Mm
0.35
0.41
0.18
0.32
0.18
0.24
0.33
0.26
0.50
0.15
0.09
0.25
0.32
-
-
Max Mean
0.55 0.43
0.68
0.49
0.60
0.46
0.44
0.47
0.40
0.92
0.42
0.27
0.34
0.40
'
•
0.43 0.69
•i • —
0.50
0.37
0.58
0.40
0.44
0.64
0.54
0.62
0.30
0.23
0.27
0.36
• "
*
0.54
Znne 3 Total Basin
"Min Max Mean
0.35 0.59 0.40
0.34 0.67
0.28 0.49
0.44 0.79
0.27 0.49
0.29 0.62
0.54 0.78
0.52 0.56
0.47 0.83
0.15 0.40
0.08 0.40
0.23 0.31
t
0.23 0.42
"
0.56 0.77
0.46
0.35
0.60*
0.43
0.52
0.92
0,55
0.64
0.38
0.31
0.25
0.37
0.56
Mm Max Mean
0.31 0.54 0.43
0.30 0.64
0.24 0.45
0.47 0.77
0.25 0.60
0.36 0.76
0.77 1.00
0.52 0.58
0.51 0.83
0.22 0.51
0.17 0.49
0.21 0.30
0.24 0.45
0.37 0.65
0.49
0.36
0.54
0.39
0.43
0.66
0.48
0.63
0.31
0.24
0.27
0.37
0.40
0.52
0.56
•.i •
Oiffuser
Flow
Min Max Type Hegirne
0.36 0.53 PVCPeri. Plug
Memo. Flow
0.36 0.64
0.24 0.48
0.44 0.68
0.23 0.52
0.31 0.57
0.56 0.79
0.44 0.51
0.51 0.75
0.21 0.40
0.11 0.39
0.24 0.31,
0.29 0.45
0.34 0.46
0.45 0.59
0.42 0.67
Tubes
Ceramic
Discs
Ceramic
Domes
Ceramic
Plates
Ceramic
Plates
Ceramic
Domes
Ceramic '
Domes
Ceramic
Discs •
Ceramic
Plates
Ceramic
Discs
Ceramic
Domes
Ceramic
Domes
Ceramic
Domes
Ceramic
Domes
Ceramic
Domes
Rigid
Porous
Plastic
Tubes
Plug
Flow
Step
Feed
Plug
Flow
Plug
Flow
Plug
Flow
Plug
Flow
Plug
Flow
Step
Feed
Plug
Flow
Plug
Flow
Step
Feed
Step
Feed
Step
Feed
Step
Feed
Plug
Flow
• Each zone represents 1«^ «^J^L*,
*> Reaerntion volume nol mdude-
e SOTE tor plaie ditfusers was i
318
-------
FIGURE 1
Schematic Structure of Type I Fouling
Precipitated Inorganic
Foutanl Layer
Small (10.2 mm x 0.6 mm)
Pores in Rigid Diffuser Surface
FIGURE 2
Idealized Plot Showing Effects of Type I Fouling on DWP and OTE
a.
I
Time
319
-------
FIGURES
Schematic Structure of Type H Fouling
Large (0.5 x 0.9mm) Round Aperture*
* in •Leathery" BtoWm V
Structured
Anastomosing
0.6 mm) Irregular
Pores in
Rigid
Oiffuser
Surface
Tight Adhesion
•A Biotilm
Cavernous Spaces
m 17mm
Between Difluser Surface
and Biofilm
FIGURE 4
Idealized Plot Showing Effects of Type H Fouling on DWP and OTE
a.
1,
Ttme
320
-------
aF(SOTE)
aF0(SOTE)
F - 1.0
QF,(SOTE)
aFmin(SOTE)
FIGURES
Hypothetical Fouling Patterns
Linear Fouling Model
Note: Curve Must be Plotted for the Same
Operating Conditions and a Values;
That Is:
F - aF,(SOTE)/aF0(SOTE)
F - F,/F0 - F/1
Time Since Diffuser was Cleaned, t
321
-------
FIGURE 6
linear Fbuling Factor Model
1.0
I
1.0-F 1 .lc»F,(SOTE)/aFn(SOTE)
Time Since Diffuser was Cleaned, t
FIGURE 7
oF(SOTE) vs. SRT for Ceramic Diffuser Facilities
aF(SOTE)/tt, percant
1.5 r-
Numbers refer to plant
identifications in Table S
o Nitrifying
• Non-nitrifying
SRT, days
322
-------
FIGURES
-------
-------
B^DvETG.RAPABII-ITY AND BIODEGRADATION KINETICS ASSESSMENT FOR
TOXIC ORGANIC COMPOUNDS USING OXYGEN UPTAKE MEASUREMENTS
by
Henry H. Tabak
Biosystems Treatment Section
Treatment Assessment Branch
Water and Hazardous Waste Treatment Research Division
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
and
Sanjay Desai and Rakesh Govind
Department of Chemical & Nuclear Engineering
University of Cincinnati
Cincinnati, OH 45221
This paper has been reviewed in accordance with
the U.S. Environmental Protection Agency's peer
and administrative review policies and approved
for presentation and publication.
Prepared for Presentation at:
Twelfth United States/Japan Conference
on Sewage Treatment Technology
Cincinnati, Ohio
October 12-13, 1989
325
-------
BIQDEGRADABILITY AND BIQDEGRADATION KINETICS ASSESSMENT FOR
TOXIC ORGANIC COMPOUNDS USING OXYGEN UPTAKE MEASUREMENTS
* ABSTRACT
!
Electrolytic respirometry involving natural sewage,-sludge and soil
microbiota is being applied to the fate studies of priority pollutant and
RCRA toxic organics to generate data on their biodegradability and on
biodegradation/inhibition kinetics. This paper discusses the experimental
design and procedural steps for the respirometry biodegradation and toxicity
testing approach for individual organics or specific industrial wastes. The
discussion also includes a review of the electrolysis BOD measuring system
inherent in electrolytic respirometry and the factors effecting
respirometric determination and measurement of respiration rate.
A developed multi-level protocol is presented for determination of the
biodegradability, microbial acclimation to toxic substrates and first order
kinetic parameters of biodegradation (n and n') and for estimation of the
Monod kinetic parameter (JL, Ks and Y ) of toxic organic compounds, in order
to correlate the extent and rate of biodegradation of these organics with a
predictive model based on chemical properties and structure of these
compounds. >
Respirometric biodegradability/inhibition and biodegradation kinetic
data are provided for representative RCRA alkyl benzenes, phenolic
compounds, phthalate esters, and ketqnes. Data on the effects of the source
of sludge biomass, temperature, and concentrations of microbial inoculum and
of toxic substrate on the kinetics of biodegradation, are also included.
.326
-------
.INTRODUCTION
Electrolytic respirometry is attaining prominence in biodegradation
m^c,1,^ an?k1Su-Somin§ ?"?.of the more Su1table experimental methods for
measuring the biodegradability and the kinetics*of biodegradation of toxic
organic compounds by the sewage, sludge and soil microbiota and for
determining substrate inhibitory effects to microorganisms on wastewater
treatment systems,
Biodegradation of toxic and hazardous organic compounds holds a great
promise as an important fate mechanism in wastewater treatment and in soil
detoxification Information about the extent and rate of biodegradation s
a prerequisite for informed decision making on the applicability^ of the
Ji2 ^ra?ahrn/Ppr°^h: Unf°rtunately, relatively little quantitative data
are available from which engineering judgement can be made, because of the
large effort required to assess biodegradation kinetics.
Current research in our laboratories has shown that it is possible to
Jhrnnnh Jhfef fi0? kjne^c parameters from oxygen uptake data, obtained
through the use of electrolytic respirometry. This method greatly reduces
the work and expense involved in evaluation of biodegradation kinetics The
ongoing biodegradation studies are concerned with the generation of
biokinetic database so that it can be ultimately used to establish a
possible correlation between molecular substrate configuration
(chemical/physical characteristics) and biomass activity (kinetic
parameters) as an index of biodegradation; The experimental respirometric
testing is also providina data on the mnrpntratinn ipypic nf tnv-.v nw,r>«^,.
• eveis OT T.OXIC organics
respirometry as an approach to measure the biodegradation of selected
organic compounds and generate biokinetic data so that these can be
ultimately used to establish a possible correlation between molecular
substrate configuration (physical/chemical characteristics) and biomass
activity (kinetic parameters) as an index of biodegradation/ The
experimental respirometry testing is also providing data on the
concentration levels of toxic organics inhibitory to microbial activity.
Initially, the inter-laboratory, ring test, Organization of Economic
Cooperation and Development (OECD) studies at the EPA laboratory,
Cincinnati, Ohio, were undertaken to develop confirmatory respirometric
biodegradability testing procedure. Respirometric biodegradabilitv and
biokinetic data were provided for the selected non-inhibitory and non-
adsorbing compounds, tetrahydrofuran, hexamine, pentaerythritol, 1-napthol,
sodium benzene sulphinate, thioglycolic acid and the biodegradable reference
compound, aniline.
tn Hoto s1"J1l?r electrolytic respirometry studies were initiated
to determine biodegradation kinetic parameters for selected representative
toxic compounds of varied classes of organics included in the Priority
Pollutant, RCRA and superfund CERCLA lists, and to demonstrate presence of
327
-------
any inhibitory effects of these organics of specified concentration levels
on the sludge biomass and on the metabolism of^biogenic compounds.
The objectives of the present study were to utilize the electrolytic
respirometry oxygen uptake data to: (1) determine tne'blo^9™dability of
selected RCRA alkyl, chloro, and nitrobenzenes, phenols, phthalates, and
ketones and representative tERCLA leachate toxic organics; (2) generate
information on their acclimated times (t0) and the initiation and
termination time values for the declining growth phase (t^ and t2); (3)
determine their first order kinetic parameters of biodegradation (specific
growth rate constants for the exponential growth phase (n) and for the
declining growth phase (JB'); (4) estimate the Monod kinetic parameters (/ira,
K and Y) of these compounds without initial growth or growth yield
alsumptions; (5) demonstrate presence of any inhibitory effects of these
compounds on the metabolism of the biodegradable reference compound,
aniline; and (6) to correlate the extent and rate of biodegradation of these
compounds with a predictive model based on chemical properties and structure
of these compounds.
The purpose of this study was to obtain information on biological
treatability of the benzene, phenol, phthalate and ketone organics and of
the Superfund CERCLA organics bearing wastes in wastewater treatment systems
which will support development of an EPA technical guidance document on the
discharge of the above organics to POTWs. The study was to generate basic
information on the fate of CERCLA leachate organics. during on-site treatment
and biodegradation and inhibition data for pollutants found in Superfund
site wastewater that could be discharged to POTWs. Respirometric
biodegradability, biokinetic and inhibition data were generated for the
selected RCRA benzene, phenolic, phthalate and ketone compounds.
BACKGROUND
Measurement of Oxygen Consumption
Measurement of oxygen consumption is one of the oldest means of
assessing biodegradability. Time consuming manual measurement of oxygen
uptake (dilution BOD measurements) was replaced gradually by a more direct
and continuous respirometric method for measurement of oxygen consumption in
biochemical reactions, for use in routine examination of sewage and in
control of sewage treatment process.
A rather comprehensive review of the use of respirometers for the study
of sewage and industrial waste and their application to water pollution
problems was published by Jenkins in 1960 (1). Montgomery's (2) review of
respirometric methods summarized the design and application of respirometers
for determination of BOD.
The application of respirometry was gradually directed to research
studies to assess the toxicity and biodegradation of specific wastes or
compounds, to evaluate factors affecting biological growth and to provide an
insight into nitrification reaction. Of the commercial respirometers which
have been developed for respirometric studies, the electrolytic
respirometers were shown to be most applicable for measurement and
328
-------
quantitation of biodegradation activity because they automatically produce
oxygen as needed, thereby eliminating some of the limitations of other
techniques and allowing output data to be collected automatically for direct
recording and processing (3-9). A recent detailed review of respirometric
techniques and their application to assess biodegradability and toxicity of
organic pollutants was published by King and Dutka (10).
Respirometric Biodegradability Testing
Most uses of electrolytic respirometry in biodegradability testing have
been for screening purposes to measure the extent of biodegradation as a
percentage of the theoretical oxygen demand exerted in some time period
(9, 11-17). A more recent study by Painter and King (18) concluded that-a
procedure based on electrolytic respirometry was reliable for assessing
biodegradability, and could serve as an adequate Level I screening test for
biodegradability (19).
A considerable number of .studies using electrolytic respirometry to
determine the biodegradability of wastes and specific organics is available
in published literature and significant data on biodegradation of pollutants
based on oxygen uptake have been generated (2, 4, 5, 20-34).
There are many techniques that have been used to evaluate
biodegradation kinetics and these were reviewed in detail by Howard et al. ,
(35, 36) and Grady (37). These techniques utilize continuous, fed-batch and
batch type reactors for providing data from which kinetic parameters can be
evaluated. The use of batch systems in biotechnology and biological
wastewater treatment represents a less labor intensive, less expensive and
much faster way to model biokinetics.
Hopefully, the kinetic parameters obtained by the above techniques
would be intrinsic, that is, dependent only on the nature of the compound
and the degrading microbial community and not on reactor system used for
data collection. If this condition is satisfied, then the parameters
obtained can be used for any reactor configuration and can be used in
mathematical models to estimate the fate of toxic organics.
Batch techniques have been shown to be successful in obtaining
intrinsic kinetic parameters by applying non-linear curve fitting techniques
to single batch substrate removal curves, provided initial conditions are
selected with proper care (Simkins and Alexander (38, 39), Robinson and
Tiedje (40), Cech et al. (41), and Braha and Hafner (42). Batch systems can
be used with either acclimated or unacclimated biomass for providing
kinetic data and require that samples be taken at discrete time intervals
during the course of biodegradation [Tabak et al. (43), Larson and Perry
(24) and Paris and Rogers '(44)].
329
-------
Use of Electrolytic Respirometry to Generate Biokinetic Data
Dojlido (11), Larson and Perry (24), Tabak et al. (45), Oshima et al.
(46), Gaudy et al. (47, 48), Grady et alI (49-51) investigated the use of
respirometry to generate biodegradation kinetic data. Larson and Perry (24)
showed that the electrolytic respirometer can be used to measure
biodegradation of complex organics in natural waters when specific
analytical methods or radiolabeled materials are unavailable. However, they
used empirical kinetic expressions which were system specific.
Dojlido (11) divided the oxygen uptake curve into seven different
phases and then proposed an empirical model for each phase and evaluated the
biodegradability and toxicity of a test compound by measuring empirical rate
constants and time interval associated with each phase. Various phases were
distinguished by identifying inflection points in the curve through the
plots of the logarithm of the slope versus time.
Tabak et al. (45) and Oshima et al. (46) sought to capitalize on
Dojlido's method of identifying inflection points in order to quantify more
fundamental rate coefficients. In order to identify more fundamental (and
therefore intrinsic) kinetic coefficients they used the. generalized concept
of oxygen uptake by Gaudy and Gaudy (52). Substrate removal was divided
into exponential and declining phases separated by an inflection point, and
the endogenous phase. They coupled substrate removal and cell growth to
oxygen consumption by imposition of an electron balance and consequently
were able to evaluate p from oxygen consumption data up to the inflection
point. They proposed tne use of the lag time as an indicator of how
difficult it is to achieve acclimation to a test compound and their
respirometric studies were carried out with an unacclimated biomass.
The justification for using respirometry to obtain intrinsic kinetic
lies in the concept of oxygen consumption as an energy balance [Busch et al.
(53); Gaudy and Gaudy (52)]. This concept states that all of the electrons
available in a substrate undergoing biodegradation must either be
transferred to the terminal acceptor or be incorporated into new biomass or
soluble microbial products. If the concentrations of the substrate, the
products and the biomass are all expressed in units of chemical demand
(COD), then the oxygen uptake can be calculated in a batch reactor from an
equation relating oxygen uptake to substrate, biomass and soluble products
[Grady et al. (49)]. Furthermore, since biomass growth and product
formation are proportional to substrate removal, this suggests that an
oxygen uptake curve can provide the same information as either a substrate
removal curve or a biomass growth curve. This latter concept has recently
been used by Gaudy et al. (47, 48) to calculate biodegradation kinetics.
Specific growth rates obtained from growth studies as slopes of plots of Ln
(biomass concentration) or Ln(X) versus time at different substrate
concentrations, compared favorably with those obtained from exponential
phase of respirometric oxygen uptake curves as slopes of plots of Ln (d
oxygen uptake)/d time or Ln (dOu/dt) versus time.
it
Studies of Grady et al. (49, 50) have demonstrated that it is possible
to determine intrinsic kinetics of single organic compounds by using only
measurements of oxygen consumption in respirometric batch reactors. With
the use of computer simulation techniques and non-linear curve fitting
330
-------
methods, intrinsic kinetic parameters were obtained from oxygen consumption
data and were shown to be in agreement with those obtained from traditional
measurement of substrate removal (DOC, SCOD, 14C) or cell*growth.
MATERIALS AND METHODS
Experimental Approach
The electrolytic respirometry approach to determine the biodegrad-
ability of the organic test compounds in this study was chosen because of
the specific advantages of the respirometric methods over that of manometric
procedures in tracking oxygen utilization during the exertion of biochemical
oxygen demand (BOD). These advantages are listed in Table 1. General
classification of respirometers based on principle of operation and on
techniques and applications is presented in Tables 2 and 3.
The electrolytic respirometry studies were conducted using an automated
continuous oxygen uptake and BOD measuring Voith Sapromat B-12 (12 unit
system) electrolytic respirometer-analyzer: The instrument consists of a
temperature controlled waterbath, containing measuring units, a recorder for
digital indication and direct plotting of the decomposition velocity curves
of organic compounds; and a cooling unit for the conditioning and continuous
recirculation of waterbath volume. The recorder shows the digital
indication of oxygen uptake and constructs a graph for these values of each
measuring unit. The cooling unit constantly recirculates water to maintain
constant temperature in the waterbath. Each measuring unit as shown in
Figure 1 is comprised of a reaction vessel with a carbon dioxide absorber
mounted in a glass joint flask stopper, an oxygen generator and a pressure
indicator. This measuring unit is interconnected by hoses, forming an air
sealed system, so that the atmospheric pressure fluctuations do not
adversely affect the results.
The activity of the microorganisms in the sample creates a vacuum which
is recorded by the pressure indicator, which triggers the oxygen generator.
The pressure conditions are balanced by electrolytic oxygen generation. The
quantity of the sample, the amperage for the electrolysis and the speed of
the synchronous motor are so adjusted that, with a sample of 250 ml, the
digital counter indicates the oxygen uptake directly in mg/L. The C02
generated is absorbed by soda lime. The nitrogen/oxygen ratio in the gas
phase above the sample is maintained throughout the experiment and there is,
no depletion of oxygen. The recorder-plotter concomitantly constructs an
oxygen uptake graph for the selected values. The oxygen generators of the
individual measuring units are electrolytic cells which supply the required
amount of oxygen by electrolytic dissociation of a copper sulfate solution
combined with sulfuric acid.
The nutrient solution used in these studies was an OECD synthetic
medium (19, 56) consisting of measured amounts per liter of deionized
distilled water of (1) mineral salts solution; (2) trace salts solution,
(3) a solution (150 mg/L) of yeast extract as a substitute for vitamin
solution.
and
331
-------
The microbial inoculum was an activated sludge from The Little Miami
wastewater treatment plant in Cincinnati, Ohio, receiving municipal
wastewater. The activated sludge sample was aerated for 24 hours before use
to bring it to an endogenous phase. The sludge biomass was added to the
medium at a concentration of 30 mg/L total solids. Total volumes of the
synthetic medium in the 500 ml capacity reactor vessels were brought up to a
final volume of 250 ml. «
The test and control compound concentrations in the media were 100
mg/L. Aniline was used as the biodegradable reference compound, at a
concentration of 100 mg/L.
The typical experimental system consisted of duplicate flasks for the
reference substance, aniline, and the test compounds, a single f>ask for the
physical/chemical test (compound control), a single flask for toxicity
control (test compound plus aniline at 100 mg/L each) and an inoculum
control. The contents of the reaction vessels were preliminarily stirred
for an hour to ensure endogenous respiration state at the initiation of
oxygen uptake measurements. Then the test compounds and aniline were added
to it. The reaction vessels were then incubated at 25°C in the dark
(enclosed in the temperature controlled waterbath) and stirred continuously
throughout the run. The microbiota of the activated sludge used as an
inoculum were not pre-acclimated to the substrates. The incubation period
of the experimental run was between 28 to 50 days. A more comprehensive
description of the procedural steps of the respirometric tests is presented
elsewhere (45, 55 and 56). ,
For fully automatic data acquisition, frequent recording and storage of
large numbers of oxygen uptake data, the Sapromat B-12 recorders are
interfaced to an IBM-AT computer via the Metrabyte interface system. The use
of Laboratory Handbook software package allows the collection of data at 15
minute intervals.
The oxygen utilization by the biomass based on the oxygen uptake
velocity (BOD) curves, consisting of the exponential and declining phases of
microbial growth was the basis for measurement of substrate utilization and
growth rate. Figure 2 provides a generalized plot of the substrate
concentration, biological solids concentration (biomass), and oxygen
utilization during exertion of BOD, versus time in a respirometric
biodegradability treating vessel system. A typical relationship between the
BOD and substrate degradation curves as well as between biomass and
substrate concentrations and growth yield is illustrated in Figure 3.
Possible curves of oxygen uptake attributed to the organic test
substrate that can be generated in a respirometric run and which are
dependent on the acclimation time, the extent and rate of bio-oxidation of
the substrate, the presence of cometabolite(s) and the presence of
biocatalytic additive in the nutrient solution are demonstrated in Figure 4.
Factors affecting respirometric BOD determination have been taken into
consideration in the study, and are listed in Table 4.
332
-------
Chemical Analysis
Indirect analysis of culture samples from respirometric vessels during
the study included determination of chemical oxygen demand as soluble and
total COD and of dissolved oxidizable carbon (DOC) with the use of a Beckman
Model 915 B system.
*
Specific substrate analysis of culture samples for the residual parent
compound and the possible intermediate and final products of metabolism was
performed with the use of a Gas Chromatograph Varian 3700, equipped with FID
and EC detectors and with different detector columns (depending on the
substrate to be analyzed) and the Finnigan automated GC/EI-CI Mass
Spectrophotometer system. Samples were analyzed as dichloromethane extracts
of cultures (continuous liquid-liquid extraction) or by direct aqueous
injection.
Suspended solids and biomass were measured gravimetrically (solids
retained on a 0.45 im filter) and by optical density determined
spectrophotometrically (percent transmission read at 540 nm).
-Determination of Substrate Biodegradability from Oxygen Uptake Data
In this study,, biodegradation was measured by three approaches: the
first, as the ratio of the measured BOD values, in mg/L (oxygen uptake values
of test compound minus inoculum control - endogenous oxygen uptake values)
to the theoretical oxygen demand (ThOD) of substrate as a percent; the
second as a percentage of the test compound as measured by dissolved organic
carbon (DOC) changes [OECD Guidelines for Testing of Chemicals (Method DGXI
283/82, Revision 5) (56)]; and the third, as a percentage of the test
compound as measured by specific substrate,analysis.
Graphical representation of percent biodegradation based on the
BOD/ThOD ratio were developed against time for each test compound. The
experimental DOC data for the initial samples and samples for reaction
flasks collected at the end of experimental run were used to calculate the
percent biodegradation based on the percent of DOC removal in the culture
system.
Determination of Kinetic Parameters of Biodegradation
The Monod equation, relating cell growth to biomass and substrate
concentration and the linear law, relating cell growth to substrate removal
are the most popular kinetic expressions which can provide adequate
description of growth behavior during biodegradation of substrate. The
Monod relation states that cell growth is first order,with respect to
biomass concentration (X) and mixed order with respect to substrate
concentration (S) by tbe equation
dX/dt = (SumX)/(Ks + S)
(1)
333
-------
Cell growth is related to substrate removal by the linear law by the
equation
dX/dt = -Y(dS/dt)
(2)
The kinetics of biodegradation were evaluated by quantifying ft , Ks and
Y kinetic parameter's expressed in the equation for the rate of substrate
removal . r^:
' (3)
where X is the concentration of biomass capable of utilizing the organic
substrate and Y is the biomass yield coefficient for the compound and in the
Honod equation;
(if the compound is not inhibitory to its own biodegradation) or by £he
Haldane equation;
+ Ss + (Ss2/Kr)
(5)
if the compound is inhibitory. In these equations, /^ is the maximum
specific growth rate, Ks is, the half saturation coefficient, KT is the
inhibition coefficient and S is the concentration of substrate.
A graphical presentation of the Monod substrate utilization equation is
illustrated in Figure 5 which provides a relationship between rate of
substrate utilization and substrate concentration.
Determination of Rates of Exponential and Declining Growth
The first order kinetic irate constants (specific growth rate
parameters) were determined by the linearization of the BOD curves or
transforming the typical BOD curve to the linear function of time t, by the
relationship of log dO /dt to t, which gives straight lines expressing the
exponential and declining endogenous phases of the BOD curve as shown in
Figures 6 and 7. The slope of the Ln(d oxygen uptake/dt) versus t give
specific rate constants of the exponential growth phase (n values) and the
dec-lining growth phase (/*' values) of the BOD curve as described by Dojlido
(11), Tabak et al. (45), Oshima et al . (46), and Tabak et al . (54, 55).
Acclimation time values (t0) and the time values for the initiation and
termination of the declining growth phase (t1 and t2) for each test compound
were determined from linearized expressions of BOD curves.
Estimation of Honod Kinetic Parameters
The estimations of the Monod Kinetic parameters, maximum specific
growth rate constant, /L, half saturation constant, K and growth yield
constant, Y were determined directly from experimental oxygen uptake curves
without the consideration of initial growth and growth yield assumption
[Jobbagy, Grady and Tabak (57) and Tabak et al. (54)].
/334
-------
If the concentrations of the substrate, the products and the biomass
are all expressed in BOD units, then the oxygen uptake (0 at any time in
the batch reactor may be calculated from **
- (X-XJ - (S -SDO)
(6)
where Sso, Spo and X0 are the concentrations of substrate, products and
cells, respectively, at time zero, and S and S are the concentrations of
substrate and product, respectively, at time t.
To apply equation (4) for the determination of kinetic coefficients,
equations must be available which express the concentrations of soluble
substrate (Ss), soluble product (S ) and biomass (X) as functions of time.
For batch reactors, those equations are:
dSs/dt = -(MyY)SsX/Ks + Ss) (7)
dSp/dt = (Yp/in/Y)SsX/(Ks + Ss) (8)
dX/dt = 0mSsX/(Ks 4. ss) - KsbX/(Ks + Ss) (9)
where Ss = soluble substrate concentration; S = soluble product
.concentration;-Yp = product yield; and b = decay coefficient.
To calculate oxygen uptake in a batch reactor, equations (7), (8) and
(9) must be solved simultaneously, and the resulting values of Ss, S and X
over time are substituted into equation (6). p
Determination of Y Constant
Y - the true yield parameter or the ratio of growth of biomass to
substrate utilization, can be obtained from the experimental oxygen uptake
curve at the initiation of the plateau of the curve as shown in Figure 8,
with the use of equation:
Y- d-Oupt/S0)-Y
(10)
Where Oypt is the cumulative oxygen update value at the initiation of
plateau, S0 is the concentration of substrate at time zero,
product yield. .
and Yp is the
A vertical line is drawn at the point of intersection of the tangents
of the exponential and plateau phases of the curve. The oxygen uptake value
obtained at the point of intersection of the vertical line (drawn through
intersection of tangents) and oxygen uptake curve is the 0 t value -
[cumulative oxygen uptake value at the initiation of plateau].
The Oqpt value is then substituted into equation (10) , for Y
determination. Y is soluble product concentration formed, divided by
initial substrate concentration. In this study, product yield (Y') was
negligible. p
335
-------
Determination of um Constant
The initial estimate of nm is obtained by the technique of Gaudy et
al. (47, 48). If Y is assumed to be zero, equation (8) is eliminated and
if b (decay constant) is assumed to be zero, equation (9) is simplified to
dX/dt = (/imSs)C)/(Ks
(11)
Combining equation (7) and (11) and integrating from Sso to Ss and from X0 to
X gives: -
X = X
Y(SSO - Ss)
(12)
If the assumption that S » Ks, the term SS/KS + S? in equations (7) and (11)
or (1) approaches one, and these systems can be simplified through the use
of equation (12) to give equation
dX/dt = /imX
: *
Integrating equation (13) and combining with equation (2) and than
substituting in equation (6) with P and P0 both equal to zero .gives
Ln[X0
= Ln(X0) + /im(est)
(13)
(14)
The plot of Ln[(X0 + 0U)/((1/Y) - 1)] versus time will give a straight line
with slope /Jm. The accuracy of /Jm(est) will depend upon the size of Ks, but
this value is good enough as an estimate.
fim - the maximum specific growth rate can be determined from
experimental oxygen uptake curve plot in the following manner:
(1) Values of the change of Ou with time (dOu/dt) or slopes are determined
along the entire experimental oxygen uptake. curve as shown in Figure 9.
(2) These dOu/dt (slope) values are then plotted against the cumulative Ou
values for each time interval, as shown in Figure 10.
(3) The slope of the developed linearized form of oxygen uptake curve is
the estimated nm value.
Determination of K= Constant
K - the half saturation constant or the substrate concentration at
which the specific growth rate is 1/2 the maximum specific growth rate can
be obtained from the experimental oxygen uptake curve in the following
mariner:
(1) Value of 0 t can be calculated from the plot of (dOu/dt) versus Ou
provided the Ks value is 1 or less (insignificant in comparison to S0
value) and the plot contains a linear section with the slope nm, as
shown in Figure 11.
336
-------
(2) Other (dO /dt) versus 0 plot in which the slope deviates from /j •
because of larger Ks values (more significant in comparison^ Sm) is
illustrated in Figure 12.
(3) The value of dO/dt is determined at the intercept of the straight line
developed from the plot of dOu/dt versus Ou (Figure 11) which contains
a linear section with slope Mm.
(4) Beginning with the value of 1/2 the intercept value, another straight
line (b) is constructed with the slope 1/2 that of the slope of-
original line (a) whose slope is /Jm.
(5) At the point where line (b) intercepts the declining experimental curve
of the plot, a vertical line from that point of interception can
provide the value of Out on the x axis.
(6) This Out value is then used in the determination" of 1C with the use of
the equation
St = S0 - Out/(l-Yp-Y) = Ks
where S0 = initial substrate concentration and St =.substrate
concentration at time t.
(7) When the Out, Y, Y , and S0 values are plugged into the equation, the
value of St can be calculated - which is the value of K (in systems
where Ks value is 1 or less).
*
Thus the oxygen uptake value Out associated with 1/2 of the estimate of
um is used in Ks' = S0 - Out/(l-Y -Y) to get the estimate of Ks. The major
impact of Ks is upon the shape of the oxygen uptake curve in the region of
the plateau. Comparison of the experimental curves to a family of
standardized curves as an initial estimate provides an initial estimate of
Ks that is sufficient for non-linear curve fitting techniques for
quantitation of the kinetic parameters.
Quantitation of Monod Kinetic Parameters
The methodology for quantitation of the Monod kinetic parameters
requires the use of the above specific methods for estimating them initially
and subsequently followed by computer simulation methods coupled with non-
linear curve fitting techniques and is based on the use of measured values
of initial growth and growth'yield. The method requires the use of the
kinetic equation relating growth rate of biomass in presence of substrate,
the substrate utilization rate, product formation rate and rate of oxygen
consumption from 02 uptake (BOD) curves to calculate and use the theoretical
oxygen consumption data to quantitate the biokinetic parameters.
The determination of the kinetic parameters associated with
biodegradation requires a series of steps. The initial substrate (S ) and
biomass (X ) concentration must be carefully measured in COD units. sthe
ratio of the twb values must lie in a certain range in order to allow
independent evaluation of /im, K and Y [Simkins and Alexander (.38, 39)].
Grady's studies (49, 50) have shown that a SSO/X0 ratio of around 20 works
337
-------
well. The value of Y may be estimated by determining the residual stable
SCOD concentration after substrate depletion (plateau area). It is
numerically equal to the residual SCOD divided by the initial SCOD. The
value of the decay coefficient, b, may be determined by fitting to the
oxygen consumption curve after the plateau when the only activity
contributing to oxygen consumption is endogenous metabolism and cell decay.
Once X0, Sso, Y and b are known, #m, Ks and Y may be determined by non-
linear curve fitting techniques [Grady (49, 50)]. „
The technique involves the calculation of a theoretical oxygen
consumption using oxygen uptake equation (4) and equations for substrate,
product and biomass concentrations (5) - (7) with assumed Monod parameters. .
The residual sum of squared errors (RSSE) associated with the difference in
calculated and experimental oxygen uptake values is used to obtain new
estimates. The above procedure is repeated until a minimum RSSE is found.
The Grid Search technique was selected as a most suitable non-linear
curve fitting technique for application in the determination of the'kinetic
parameters from oxygen uptake data, because it can allow easy discrimination
between local minima and the global minimum RSSE. This technique enables a
comparison between the calculated and experimental oxygen uptake data. The
value of Y is fixed. For this value of Y, a pair of nm and K which give
RSSE is found on a /tm:K plane. The above-procedure is repeated with other
values of Y. Values of /i, Ks and Y which give minimum RSSE associated with
the difference in calculated and experimental oxygen uptake data constitute
the best values of fhe kinetic parameters.
The values of /im, Ks and Y developed from grid search technique, which4
when substituted into equations 5-7, will provide X and S values, which ,
(when substituted into oxygen uptake equation 4) will in turn provide
calculated oxygen uptake values at the region of the plateau, closest to the
experimental oxygen uptake values, with a minimum RSSE, will constitute the
best quantitative kinetic parameter values.
Development of Multi-Level Respirometric Biodegradation Testing Protocol
The oxygen consumption data generated with the 'use of electrolytic
respirometry have been adequately utilized for assessing the biodegradative
activity of sludge microbiota, the biodegradability/toxicity of toxic
organic compounds, as well as for the determination of the intrinisic
kinetic parameters of biodegradation. Methodologies have been developed for
quantitating biodegradability and biodegradation kinetics of representative
classes of RCRA toxic organics, with the resultant development of a
comprehensive multi-level respirometric biodegradation testing protocol
based on oxygen consumption data.
The methods of each successive testing level of the respirometric
protocol are characterized by increased complexity and a consequent higher
cost for performing the tests pertaining to each testing level.
Accordingly, the testing levels can be selected as appropriate to the
research needs.
338
-------
Studies to assess the biodegradability and biodegradation rates of
toxic organics by sludge microbiota with the use of respirometric oxygen
uptake data can involve the use of one of more levels of the protocol,
depending on the amount of information needed for assessing the
biodegradability of the organic toxic pollutant or the toxicant bearing
waste for determination of its fate and rate of biotreatment in the
municipal or industrial waste treatment systems. A logic flow diagram of
the respirometric biodegradation testing protocol, providing a brief
description of each of the testing levels, is shown in Figure 13.
RESULTS AND DISCUSSION
Respirometric biodegradability, biokinetic and Monod kinetic data for
selected RCRA alkyl benzenes, phenols, phthalates and ketones are reported
in this paper. The electrolytic respirometry oxygen uptake data for the
test compounds, the control reference compound aniline, the inhibition and
endogenous control systems were generated revealing the lag phase
„(acclimation phase), the biodegradation (exponential) phase, the different
bio-reaction rate slopes (characteristic of the test compound) as well as
the plateau region at which the biooxidation rate reaches that of the
endogenous rate of microbial activity. Figure 14 illustrates a
representative oxygen uptake curve for aniline and the endogenous controls.
Figure 15 shows the replicate pentaerythritol oxygen uptake curves and the
toxicity control (pentaerythritol plus aniline) curve, Figure 16 illustrates
a representative graphical treatment of the percent biodegradation of
pentaerythritol with time, which was developed for each test compound (OECD
studies).
Based on the biokinetic equations relating growth rate of microbiota in
presence of above compounds, the substrate utilization rate, and rate of
oxygen uptake (BOD) curves, specific growth rate kinetic parameters
(biodegradation rate constants) were derived as slope values of the
linearized plots (plots of the log of DO /dt) of exponential and declining
growth phases of the BOD curve. The acclimation time values (t0), and time
values for the initiation and the termination of the declining growth phases
(tj and t2) for the test compounds and aniline were also generated.
The estimations of the Monod kinetic parameters for benzene, phenol,
phthalate, and ketone compounds reported here, were determined directly from
experimental oxygen uptake curves without the consideration of initial
growth and growth yield assumption.
Respirometric Studies with Selected RCRA Alkyl Benzene Compounds
The biodegradation of benzene, toluene, ethyl benzene, m- and
p-xylenes, tert-butyl benzene, sec-butyl benzene, butyl benzene, cumene, 1-
phenyl benzene and the reference compound, aniline at 100 mg/L concentration
by 30 mg/L sludge biomass (as measured by oxygen consumption by sludge
microbiota in mg 02/L) was followed over a period of 20 days. The
electrolytic respirometry oxygen uptake and BOD curves were generated and
graphical treatment of the percent biodegradation was established for each
339
-------
compound. Figure 17 demonstrates typical oxygen uptake and BOD curves for
p-xylene and p-xylene + aniline and Figure 18 illustrates graphically the %
biodegradation of p-xylene with time.
The percent biodegradation data based on the BOD/ThOD ratios for
benzene, toluene, ethyl benzene, m- and p-xylene and the reference compound,
aniline, are summarized in Table 5. All of the above alkyl benzene
compounds were shown to be biodegradable substrates at concentration levels
of 100 mg/L when exposed to 30 mg/L of activated sludge biomass under the
environmental conditions of the respirometric testing procedure, and within
the period of 20 days of incubation.
The toxicity test control flask respirometric data revealed no
inhibitory effects by these test compounds at the 100 mg/L concentration
levels on the bio-oxidation of aniline by sludge microbiota.
Table 6 summarizes thefsbio-kinetic data for the benzenes studied,
showing the specific growth rate constants for the exponential growth phase
(p values) and for the declining growth phase (/*' values) of the linearized
form of the BOD curves of these compounds, as well as the tft, t, and to
kinetic parameters. Figure 19 shows a typical plot of Ln(dOu/dt) vs. time
for toluene, from which the kinetic parameters were determined.
Table 7 summarizes the Monod kinetic parameter (/im, Ks, Y ) data for
these benzene compounds.
Respirometric Studies with Selected RCRA Phenolic Compounds
The biodegradation of phenol, resorcinol, o-, m- and p-cresols,
catechol, 2,4-dimethyl phenol and the reference compound aniline at 100 mg/L
concentration levels and exposed to 30 mg/L biomass was followed over a
period of 20 days.
All of the phenols were shown to be biodegradable substrates under the
conditions of the respirometric testing procedure. The toxicity test
control flask respirometric data revealed no inhibitory effects by these
compounds at the 100 mg/L levels on the biodegradation of aniline by the
sludge biomass.
Table 8 summarizes the bio-kinetic data for the phenols studied,
showing the specific growth rate constants as well as the t0, t1} and t2
kinetic parameters. Table 9 provides the Monod kinetic parameter data for
these phenolic compounds.
Respirometric Studies with Selected RCRA Phthalate Ester Compounds
Evaluation of the biodegradability and determining of bio-kinetics of
degradation of phthalate compounds, dimethyl phthalate, diethyl phthalate,
dipropyl phthalate and butyl benzyl phthalate was achieved with use of
respirometric oxygen uptake data.
340
-------
rnnrf-?1-1 °f *h?u above Phthalates were shown to biodegradable under the
conditions of the respirometric tests and were shown not to exhibit any
sludge mTcrobTota? at the 10° mg/L levels °" ™"™* biodegradation bjrthe
ann MIabieE-10*and H summarize respectively the biokinetic (first order)
and Monod kinetic parameter data for the selected phthalate esters under
S L
Respirometric Studies with Selected RCRA Ketone Compounds
Respirometric oxygen uptake data from the studies with the selected
ketone compounds, acetone, 2-butanone, 4-methyl-3-pentanone and a cyclic
Ketone, isophorpne were utilized to determine their biodegradability and
biodegradation kinetic parameters. •
All the ketones were shown to be biodegradable at 100 mg/L
anvC?n^vnJeVe-?-in fdia C0"tainin9 30 mg/L biomass and did not exhibit
any toxicity to aniline biodegradation at these concentrations.
Tables 12 and 13 summarize respectively the first order and Monod
kinetic parameter data for these ketones.
CONCLUSIONS
The experimental data of respirometric studies with several classes of
organic compounds definitely demonstrate that it is possible to measure the
SiSS^th11^ (?-rCen£ $1ode9radation - as a ratib of BOD to ThOD) and to
determine the kinetics of degradation of single organic compounds by using
only measurements of oxygen consumption in respirometric batch reactors.
ISre Jl«nS ? + S kinetic parameters determined from oxygen consumption data
were demonstrated to be similar to those based on the measurements of
substrate removal and those made with cell growth data.
The generated data on biodegradation, biodegradation rates and
Siri Ilihii"!; i°n H"et!^s th^ough the use of electrolytic respirometry,
will enable the classification of biodegradability of toxic priority
pollutant and RCRA toxic organic compounds and ultimate projection of the
fate of organic compounds of similar molecular structure to those
experimentally studied by way of the established predictive treatability
models based on structure-activity relationships. «""niy
nf thW1tl! the 6lectro1yt1c resplrometry approach, a data base on the removal
of the above compounds by biodegradation fate mechanism can be adequately
generated to support the development of predictive models on fate and
removal of toxics in industrial and municipal waste treatment systems. A
possible relationship between the kinetic parameters and the effect of
different factors on these parameters, as determined through electrolytic
respirometry and the structural properties of the organic pollutant? cai
eventually facilitate prediction of the extent and the rate of
,^dSrafa*10?u°f,,organ!c chlml«ls In the field of wastewater treatment
systems from the knowledge of the structural properties of the pollutant
orgamcs.
341
-------
A preliminary predictive biodegradation - structure/activity model
based on the group contribution approach was developed from the generated
biodegradation kinetic data (first order kinetic parameters) with the use of
electrolytic respirometry. It is expected that the model will closely_
predict the results found experimentally. In this way, the fate of other
organic compounds may be anticipated without the time and expense of
experimental work.
The electrolytic respirometry biodegradation studies will provide basic
pilot scale treatability information and data which will be used to confirm
methods to predict treatability and the need for pretreatment of
structurally related pollutants (e.g., by structure, anticipated
treatability properties, etc.). This study will thus provide a more
extensive list of pollutants than was covered by experimental data, for
consideration in guiding the Agency to predict the fate of such compounds
without costly experimental testing.
Studies are currently in progress to determine the effect of
temperature, and different sources of sludge biomass (domestic and
industrial wastewater treatment) on the biodegradation kinetics derived from
electrolytic respirometry.
ACKNOWLEDGEMENTS
The authors wish to thank Mrs. Rena M. Howard and Mrs. Diana L.
Redmond, secretaries in the U.S. Environmental Protection Agency's Risk
Reduction Engineering Laboratory, Cincinnati, Ohio, for their excellent and
timely wordprocessing skills in preparing this manuscript for presentation
at the 12th U.S./Japan Conference on Sewage Treatment Technology, October
12-13, 1989 in Cincinnati, Ohio.
REFERENCES
1. Jenkins, D. The use of manometric methods in the study of sewage and
trade wastes, in: P.C.G. Isaac (ed.), Waste Treatment. Pergamon
Press, Oxford, 1960. p. 99-121
2. Montgomery, H.A.C. The determination of biochemical oxygen demand by
respirometric methods. Water Res. 1: 631-662, 1967.
3. Young, J.C., Garner, W., and Clark, J.W. An improved apparatus for
biochemical oxygen demand. Anal. Chem. 37(5): 784; 1965.
4 Liebman, H., and Offhaus, F. Volumetric BOD measurements with the help
of "Sapromat" a new apparatus for determining 5-day BOD and toxicity.
Abwassertechnik. 17: 4-6, 1966
5. Montgomery, H.A.C., Oaten, A.B., and Gardiner, O.K. An automatic
electrolytic respirometer--Its construction and use. Eff1uent Water
Treatment J. 11: 23-31, 1971.
342
-------
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
Young, J.C., and Baumann, E.R. The electrolytic respirometer-I
?I ing oxygen uPtake measurements. Water Res. 10: 1031
Young, J.C., and Baumann, E.R. The electrolytic respirometer-I I Use
AO ?£7£ ollution control plant laboratories,. Water Res. 10:1141-
" 1976.
Fuhs, G.W. Some factors affecting biochemical oxygen demand as
determined in manometric or manostatic devices. Wasser Abwasser-
Forschuna. 5: 161-168, 1968. ^^
Bridie,.A.L.A.M. Determination of Biochemical oxygen demand with
continuous recording of oxygen uptake. Water Res. 3: 157-165, 1969.
j E.F., and Dutka, B.I. Respirometric techniques. In: 6. Britton
and B.J. Dutka (eds.), Toxicity Testing using Microorganisms. Vol. 1.
CRC Press, Inc., Boca Raton, Florida, 1986. p. 76-113.
Dojlido, J.R. Investigation of biodegradability and toxicity of
organic compounds. EPA-600-2-79-163, U.S. Environmental Protection
Agency, Cincinnati, Ohio, 1979.
Howard, P.H., and Banerjee, S. Interpreting results from
biodegradability tests of chemicals in water and soil. Environment
Toxicol. Chem. 3: 551-562, 1984. ~~ - "
Madden, M4, and Tittlebaum, M. Oxygen uptake rates associated with
oLpe^hloropheno1 wastewater. J. Environ. Sci..
321-335, 1984.
Verstraete, W., Voets, J. P., and Vanlocke, R. Three-step measurement
by the Sapromat to evaluate the BOD5, the mineral imbalance and the
toxicity of water samples. Water Res. 8: 1077-1081, 1974.
Yoshimura, K., and Masuda, F. Biodegradation of Sodium Alkyl
Poly (oxyalkylene)-sul fates. Am. Oil Chem. Soc. J. 59: 328-332, 1982.
Young, J.C.; and Affleck, S.B. Long-term biodegradability tests of
organic industrial wastes. Jn: Proceedings of the 29th Industrial
Waste Conference, Purdue University, Extension Series 145, West
Lafayette, Indiana, 1974. r p. 154-164.
Young, J.C., and Baumann, E.R. Demonstration of the electrolysis
method for measuring BOD. Presented at the 45th Annual Conference of
Water Pollution Control Federation, Atlanta, Georgia, 1972.
Painter, H.A., and King, E.F. Environment and quality of life-rinq
test programme 1983-84 - Assessment of biodegradability of chemicals in
v?/,ei^oo/^Sanometr1e respirometry. Final Report, Contract No.
Al/w/83/238; Directorate-General Environment, Consumer Protection and
nHnl!31" Safety> Commission of the European Communities, Report No.
EUR9962EN, 1985.
343
-------
19 OECD, "OECD Guidelines for Testing of Chemicals", Section 3,
Degradation and Accumulation, Method 301C, Ready Biodegradability:
Modified MITI Test (I) adopted May 12, 1981 and Method 302C Inherent
Biodegradability: Modified MITI Test (II), adopted May 12, 1981,
Director of Information, OECD, Paris, France, 1981.
20. Arthur, R.M. Twenty years of respirometry. In: Proceedings of the
Thirty-Ninth Annual Industrial Waste Conference. Purdue University,
West Lafayette, IN, 1984.
21. Halbartschlarger, J., Kohler, H., Szwerinski, H., and Bardtke, D.
Investigations on the biological degradation of chlorinated
hydrocarbons using^dichloromethane (methylene chloride) as an example.
Gwf-Wasser/Abwasser. 125(H.8): 380-385, 1984.
22. Mickey, C.W., and Nagels, J.W. Modifications to electrolytic
respirometer systems for precise determination of BOD exertion kinetics
in receiving waters. Water Res. 19: 463-470, 1985.
23. Klecka, G.M. Fate and effects of methylene chloride in activated
sludge. ADD!. Environ. Microbiol. 44: 701-707, 1982.
24. Larson, R.O., and Perry, R.L. Use of the electrolytic respirometer to
measure biodegradation in natural waters. Water Res. 15: 697-702,
1981.
25. Manios, V., and Balis, C. Respirometry to determine optimum conditions
for the biodegradation of extracted olive press-cake. Soil Biol.
Biochem. 15(1): 75-83, 1983.
26. Nochi, K. Oxygen consumption due to decomposition of chemical
substances. J. Water Waste. 22(11); 1285-1295, 1980.
27. Nochi, K. Oxygen consumption due to decomposition of chemical
substances. J. Water Waste. 26(7); 751-757, 1984.
28. Simpson, J.R., and Nellist, G.R. Development and use of a large-volume
automatic respirometer. Water Pollut. Contr. 69: 596-605, 1970.
29. Therien, N., and Ilhan, F. Relating BOD5 with on-line oxygen uptake
rate measurements using automatic respirometers in view of process
monitoring and control. In: Procedures and Practices in Activated
Sludge Process Control. Vol. 3. 1982. *p. 113-129.
30. Urano, K., and Kato, Z. Evaluation of biodegradation ranks of priority
organic compounds. J. Hazard. Materials. 13: 147, 1986.
31. Wojnowska-Baryla, I., and Young, J.C. Measuring the effect of
biocatalytic additives on treatment process performance. J. WPCF.
55(11): 1373-1380., 1983.
344
-------
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
Pagga, U., and Gunthner, W. Biodegradation and toxicity studies with
microorganisms: A comparison of laboratory tests with a treatment
plant mode system. Jn: Proceedings of the International Symposium
on Principles for Interpretation of the Results of Testing Procedures
in Ecotoxicology, Valbonne, 1980.
Rigin, V.I., Golovin, Y.G., and Tyuneva, G.S. Determination of the
biochemical oxygen demand of natural waters and effluents with the
180-182 198296nerati0n °f °Xygen- Kh1miva i Teknolooiva Vody\ 4(2):
Young, JC. Biochemical oxygen demand: Measurement and application
Iowa State University, Ames, Iowa, Final Report, 1977.
Howard, P. H., Saxena, J., Durkin, P.R., and Ou, L.-T. Review and
evaluation of avai able techniques for determining persistence and
SK? 7? nh?^*1?" ?f chemjcal substances in the environment. EPA
560/5-75-006, U.S. Environmental Protection Agency, Cincinnati, Ohio,
National Technical Information Service No. PB243825, 1975.
n-iM",> r" !!nd Rosenberg> A. A review and evaluation
of available techniques for determining persistence and routes of
?hor?o™T °f*ch®rinal substances I" the environment: An update of
the 1975 Report EPA-560/5-81-011, U.S. Environmental Protection
PB84-168731C119811} National Technical Information Services No.
Grady, C.P.L., Jr. Biodegradation: Its Measurement and
Microbiological Basis. BiotechnoT. Bioenaineer. 27: 660-674, 1985.
th« M" andf A1^xandfr' M- Models'for mineralization kinetics with
the variables of substrate concentration and population density. ADD!
Environ. MicrobioU .47: 1299-1306, 1984. uc"Siuy. HB&I.
Simkins, S and Alexander, M. Non-linear estimation of the parameters
of Monod kinetics that best describe mineralization of several
substrate concentrations by dissimilar bacterial densities. ADD!
Environ. Microbiol. 50: 816-824, 1985. ^^'
Robinson, J.A., and Tiedje, J.M. Non-linear estimation of Monod growth
kinetic parameters from a single substrate depletion curve. ADD!
Environ. Microbiol. 45: 1453-1458, 1983. , ~^'
Cech, J.S., Chudoba, J., and Grau, P. Determination of kinetic
constants^activated sludge microorganisms. Water Sci. lechnol. 1.7:
42. Braha, A., and Hafner, F. Use of lab batch reactors to model
biokinetics. Water Res. 21(1): 73-81, 1987.
43. Tabak, H.H., Quave, S.A., Mashni, C.I., and Barth, E.F.
345
-------
44.
45.
46.
47.
48.
49.
50.
51,
52.
53.
54.
Paris, D.F., and Rogers, J.E. Kinetic concepts for measuring microbial
rate constants: Effects of nutrients on constants. ApjLL. Environ.
Hicrobiol. 51: 221-225, 1986.
Tabak, H.H., Lewis, R.F., and Oshima, A. Electrolytic respirometry
biodegradation studies, CEC/OECD ring test of respiration method of
determination of biodegradability, Ring Test Program 1984. EPA Draft
Final Report, MERL, U.S. Environmental Protection Agency, Cincinnati,
Ohit), August 1984.
Oshima, A., Tabak, H.H., and Lewis, R.F. The evaluation of biological
treatability and removability of toxic organic chemicals by
respirometry. EPA Draft Manuscript, MERL, U.S. Environmental
Protection Agency, Cincinnati, Ohio, 1985.
Gaudy, A.F., Jr., Rozich, A.F.," Garniewski, S., Moran, N.R., and
Ekambaram, A. Methodology for utilizing respirometric data to assess
biodegradation kinetics. Paper presented at the 42nd Annual Industrial
Waste Conference, Purdue University, West Lafayette, Indiana, 1987.
Gaudy, A.F., Jr., Ekambaram, A., and Rozich, A.F. A respirometric
method for biokinetic characterization of toxic wastes. Paper-
presented at the 43rd Annual Industrial Waste Conference, Purdue
University, West Lafayette, Indiana, 1988.
Grady, C.P.L., Jr., Dang, J.S., Harvey, D.M., Jobbagy, A., and Wang,
X.-L. Determination of biodegradation kinetics through use of
electrolytic respirometry. Water Sci. Technol. 21: 957-968, Brighton,
1989.
Grady, C.P.L., Jr., Dang, J.S., Harvey, D.M., and Jobbagy, A.
Evaluation of biodegradation kinetics with respirometric data.
Pollut. Control Fed. 1989: (submitted).
Grady, C.P.L., Jr., Aichinger, G., Cooper, S.F., and Naziruddin, M.
Biodegradation kinetics for selected toxic/hazardous, organic compounds.
Hazardous Waste Treatment: Biosystems for Pollution Control, AWMA,
1989: (in press).
Gaudy, A.F., Jr., and Gaudy, E.T. Biological concepts for design and
operation of the activated sludge process. EPA-17090 FQJ, U.S.
Environmental Protection Agency, Cincinnati, Ohio, 1971.
Busch, A.W., Grady, L., Jr., Rao, T.S., and Swilley, E.L. Short-term
total oxygen demand test. J. Water Pollut. Control Fed. 34: 354-362,
1962.
Tabak, H.H., Desai, S., Govind, R., and Grady, C.P.L. Evaluation of
biodegradability and biodegradation kinetics of organic pollutant
compounds with the use of respirometry. Presented at the 61st Annual
Conference of Water Pollution Control Federation, Dallas, Texas,
October 2-6, 1988.
Water
346
-------
0
"'
57.
HanometHc
347
-------
TABLET
ADVANTAGES OF RESPFOMETWC
1. oxygan uptaka can ba monitorad oontinuou.ly and eon.tantly and .or.
pracisaly.
,. Automation provida. Bu«rical or binary output aat. for air.et
recording er procea.ing (electrolytic).
' ssx&ss,
- jsgss,
5 Th. »anpl.s ar. .ix.fl continuou.ly to provia. uniform contact of
microorganism., substrata and orygan.
«. xo ebamical titrations ara ratjuirad.
7. * continuou. record of O2 uptake i. provided with .om. unit, having
automatic recording device..
a. jermit treatment plant* and in-.tream condition, to be .imulat.d
more clo.ely than in dilution te.t.
-
volume and .ource of .ewage .eed and .eed adaptation.
10. can b. u..d to determine bacterial growth and .ub.trat. removal
coefficient..
-
»•
13. o.afull infor«.tion i. oft.n av.ilabl. in vary .uoh 1... than in 5
day*.
which variabla. can ba eontrollad.
348
-------
TABLE 2 GENERAL CLASSIFICATION OF RESPIROMETERS
Types
1. Manometrk
2. Electrolytic Same.
3. Dissolved 02
depletion
Basic Prinfjply fff O
Determination of O2 weight changes in a dosed system
by measuring or responding to O2 pressure chances at
and ¥0lume " ¥0lume ^ •*
Use of dissolved oxygen probe to make dir«-t
measurements of depletion of dissolved O2 from
TABLE 3 TYPES OF RESPIROMETERS BASED ON TECHNIQUES AND APPLICATIONS
measuring gas
A. Sfnall constant Preyynr* r»SDJromi.t<.r< _
Measurements are made by observing the change in volume of
SSHf " ""^ WUh the feSpi""g llquid' « I«^« 02 is
B. Small constant
°2
C.
s
Lifgt respirometers mi-astiring e*« »»/-h=.^n -
Kecommended especially for studies of treatability.
D. Electrolvti<;
maintained at a constant value
^ 02 ^
349
-------
Diagram of a measuring
unit.
A reaction vessel
B oxygen generator
C pressure indicator
1 magnetic stirrer
2 sample (250 ml)
3 C02 absorber
4 pressure indicator
5 electrolyte
6 electrodes
7 recorder
SCHEMATIC DIAGRAM OF A MEASURING UNIT
FIGURE 1
350
-------
-------
Endogenous Phase
X
>;
(/f
Substrate
^——————
Removal Phase
BOD Curve
Test Compound
Degradation Curve
Time
The schematic diagram for the relationship of S, y and x.
FIGURE 3
352
-------
> «
Time (days)
A - 02 uptake curve of a rapidly oxidizing substrate.
B -
C - .Og uptake curve with an initial lag period
^ miCr°bbta "*
m beffre °^atl'0n othe remaining part of
substrate. Illustration of catabolite repression.
E -
cumulative Or uptake attributed to b
m (A) and (B) curves for a substrate
in presence of only the
control seed organisms.
1?f!I!LE CURVES OF THE OXYGEN UPTAKE
ATTRIBUTED TO THE PROBLEM SUBSTRATE
FIGURE 4
253
-------
z
o
5
Z
1
IS
CO
o
re
&
ra
u
bO
i
e
"re
c
o
n* ^^
is 2-.:s
re
J *
re **
•1
I
I
852
•
O
UJ
(/J
UJ
re
re
— .2
O •-
u.
UJ
m
.B
ra
Ill
- ET "o a
E i
1 §
J I
o 9-
•* *•
0 II
«•- *S
2 -2 S S .2
:s2« 2,
VI
0)
5> To 2
^ ^ —
ts e S
o, « «
0)
bO
.2 -5 5»
S1 S :! e
:§ s
-------
Concentration Range for Respirometer Tests
a
M
0)
4-4
(O
**
W)
J3
a
ra
S,mf/l
Substrate Concentration
Graphical presentation of the "Monod" substrate utilization
equation. The reaction approaches first order at low substrate
concentration and a zero-order at high substrate concentrations
FIGURE 5
355
-------
QC
(D
>s
O
•o
UJ
a
u
c/)
a.
LU
CO
LU
OC
ID
g
LL
C/op/-|/Bui>
356
-------
LU
a
UJ
en
Q_
UJ
LU
02
a. <
•o cc
05 LU
O <
LU
Q.
LU
DC
g
LL
357
-------
gsasaa*-
(I/Bui) uondiunsuoo ue5Axo
o>
UJ ±:
H
UJ CO
X UJ
LL. —
O UJ
§^
g LU
UL OL
O s
UJ
Q P
«
368
O
LL
-------
M
H
(l/Buj) NOIlcMnSNOO N30AXO
HI
S
cc
CL h-
LU CO
O DC
>- LLI
8*
u. O
O
LU
QC
Z)
CD
LL.
359-
-------
€0
V / 3*Vldn N39AXO V
IV /
CO
cc
LU
CO
co
o
CO
S
00
(f) UJ
o
Li.
O
LLI
DC
Z)
O
U-
- 360
-------
id/noa
8
*- *
o
o
7
o
V)
trf
II
o
! 11?
in
CO O O
LU
DC
g
u.
361
-------
ssas
la/rKXi
3
o iq
o o
O
O
7
o
»n
u
o
X
o
u
o
u
o
It
3
D
CO
CO
I
2£
«!g
OB
II
¥•
CO >
fc.%
0
^*
II
CM
LU
DC
O
LL.
362
-------
X O
O_ O •"- VI
o
O
O
O
ex
CO
LU
o
+
o
re
a
at
^£
re
o s-
«-> o.
•?"•
-O.
vt re
a> «o
•!-> E S-
« o cn
i- -^ a>
erg
ce o
u c
O) ••-
0) +j
a c
•o o
4-> n>
c s- v>
I- U
03 a>
o
••-•OS
x o o
O -t— •
co
CO
,
«*-•— o
o a>
> Q)
c a> +j
O r— ITJ
C O (O
H£ g
S- -Q O
OJ -r- C
»•* "^— ^^ -" -^- ^ J^ Q
a 01 *j ^a ja a -^ *j
»™«
O
o
x;
a
a
a>
C31
t
O
S.
to
S-
0»
OJ
•o
o
a.
u-
o
c
o
O)
*->
O)
CO
Q>
OJ OO
i- O>
ai c c
a-^o —
«*-*j tl
o re a>
s-
o
c re
o s- >,
•w at f
re ^ r—
c o —
i- -^ ^
ECO re
i- -o
CD re
••-> s-
aj o>
a a>
re
•o
re
a>
o,
5
C t-
O O
o
••- co
*i »
TO »^—
E -O
u to
u
re re
_«= _E
's ^
s- u
O) (J
•^ «t
0)
a
S-
O
° =
re
re co
o
u
re
•Q "-' a) oj re 3
o _g s _c- <—
i- a> +j -^ Q. re
X- CO
o -a
o
•4->
re
•
a)
a
_. — ' •"* *^» fw
fl) CO 4-a c
_E re •— o> —
X HOC
-""^ O &- +J »r- Q3
cn*->
re 4^ ^3 c & 3
•—••-> Q
363
-------
i- -O
+-> <=
co
s-
to o •
>)
o
'
o
4->
cu
CO -
CU
^s u-g
> T3 C
eu cu co
CO C: CO
E O
«/> a a>
eu -a E
a *->
co J- «
s- o> eu
CU CU S-
-
CU
cu
o
•f—
, +J
CO."
"
o cu o
a LU ,
E •»- •
O CO C
s- «= o
r— Q)
flj S. «J
0-0 ?
+J CU
CO
CD
co
cu
•o
CD
ia
E
O
U
nj
i/)
1/1
o
•r-
.a
o
CO
CO
•o •
CU CO
!,_•*• , r- ••- CO
o eu c
>*- j_ +j r
O C=i- 3 r—
, o
O4->co4->eu *J^Z'*i
•r- s_ E eu *J
i— CU 1C > *
CL+J s- s- o
«X T3 Q. O O1
a c
• +J c co
^- o a c
c 0.0 o
>-. X CO QJ O
cu
c eu**- 3.
o c o
•I- ••- *>
*J I— CO C
nj cu eu
r— 4J CO CO
eu J=
cu
cu
•a S3 ** c
S^1— °
« ^J <0-r-
C CU =>
I *J C r-
C3» co -i- co eu
c eu >
_ o co i- e •>-
i -^ s- eu r— ••- *»
cu to a
en u c-c s-
s- s-
O -U O
•f- cu en
o o
o = eu >
e +j eu a.
m<+- cos.
S- O— »0r— Q>
O « «J *—• C -^
+J O.CO
cu S £ o w i/i s-
•r- O
r- J3
CO X CUr- S-
o.a-a
O»r
O CU I
364
-------
o o
O)
v>
o
•»• ••- 7 «>
«"
ju
•*- u
*> 5
1-
o>
o
Q.
•J; «= °
°.2~
i
en
3J2
•5S
1^'
3°
SS
g 2
*•» $-•--
o 01 2
o «*»
tt)
S|l
««E .
og»
at g
S8
sS
a. a> E
"*S
—.3 3
>• o 2
_» c
a> at
S en
ri « O
01 .t-»T2
>
•o a> <3 o
0«d =
*>r- « vi
— H «A-f~
tt)
a)
>»3 «
JE niO
*3 O *->
o en at
S = =
oi-r ^
J. Ojr
•• —
at
m t
* i
. o
-0
O) O)
re§
o. o
c **
^- a>
o>-o a
o 4-> e re o
Q. C -i- S. U
a. at re 4-> to
3 S **
at o o t_T
>> s. s- o
en o> vi n- S
o re v •>—'
*"• ai *o
o c *» air—
•o re at c re
o E"- >
-c X re re o
•4-> O i. *J E
at .c re JD at
Z VI Q. O S-
•eS
T»t
il
•sZ
VI e
2s*
JQ •>»
01 s. *»
o a>ff
•— l- a u
Oi— *>^ 3 O)
•o .re c^re
o i- a> o
o>
>ir vi
• ^^ AI
"5
5
*
Vt
ai
t. IB
. *••
INI
.a>. x .
re is*^ a» v
?JJi
2 « •*» re
a>
c S vi
«B w —• *^ T3 O
2 vi 4-> at Jj —
re u ••-
t- **-o ~ at
4= ,1 * S
2-5.^15"
5^«8g
53«|«l
** o
»- U
0)
«
—*^ re re -^
•at uj * at
S.X*^ ai h
at s-
C
C
e
n
oo
a.
i- re
re re re
•s1-5:
s. ai re
re e —
re 3 -^ 4-» —
•F- O
e u
re
at
I/I
OJ
*-> Q.
a>
j<: a> ->->
at re
0) >, i-
-C X 4->
*» o c
o>
C - (J
O 01 C
4-> O
TJ re o
ai s-
n a>
re je *»
•St. "Si-
tJ-S e§
re g *J ^.
"•" " a£ re ^
e at JS at
•s a. o
• O *> esit
_•— re o
re.
a»
a1
01
U X O
f o s.
u f-
u 3 S
•"
o ai
o o re
i— -~ ai
a* ** s-
> u +•>
a> •<-
o»
« 3 01 4-> .S
«*- Of— CT> O> c
O C O >, i- S
01 VI >? O 2
O) vt o a) a-
vi o> » .e "^
3 s- a><*- t*J o
Q.-i-' O 3
ai re 01
e re ">
vi ai o i. 03 > u
01 *» -^ ' vi ? a»
"vTSS
. at 3
at re JE
u u u
e re at
O CL4->
U s ^- VI
ai e = J. s.
> f cuOi at
J. vi O Q.-M
3 x at
w •»— at E
i-^5-o2
re JB e e re
a. re gre a.
i a. a> -i-> 4->
e a. a. re at
o re xr- e
e at 3^-
•g
£ o "^ u
C^vjS^
•2-jf.g'
, ^—
S vi
-i*
S>f-o5
reO»^5*'
:»__ *» « n,
•-*! =
«i-i-oS^
3 * at vi .Jj
5s-S*g
J|&W§
o ri at co
.5^ « i. ee -,
•>£ S *
re
at
CO
a> «x»-oi--
.c o e re<4-
1— E « O O
^
re
•ggc.
"•o'i
23-Sajg-
t- <*• W •«-
vt
re
e
VI
VI
5
re
u
re
a>
O)
VI
o>
. CO
ai
o>
o
u
O)
x 3 *»
at at o cr re e
z: J= *j ai-O"-
o Wo» c
r— an*j o s nHJ
o e re — u o 01
•Of i- 4->r-^ C
o *» *j re re •*-
je re vi E u re.^
•U f— .Q I. *J O
o> a> s o o cre —
Z fc. VI I*. 4-> ^ ^
365
-------
0>
E
o>
a.
x
OJ
E
o
in
a)
Di
CA
J= ••-
4J V)
»•=
•a c
a> c/>
o u
tn o
a> a>
o: a.
co
Q)
4->
n>
t-1- •
*> O >»
in o>
-O co o
3 4Jr—
MOO
=1 C
O T3 J=
•r- O O
X S- 0>
O O.-4-"
+•»
•a o
4-> C — I
coo.
01 Z
s--o
a c •«•
a. re c
(O
r— to
ro a> co
= *>Z
w> o to
a> .n
4J O
a> to
owe
•r- IO
«o E o
>, i- J2
r- 0) (O
« +J +J
C C 01
<:— E
a>
*->
to
>,
CO O) T3
jr a)
a> +•> *j
s- a.
3<*- nj
4J O "O •
>— (O C
3 > 3
O £ >> S-
0)
«<- ••- Q) OJ
O M Q. E
M O
ME J-
O) O J= f-
r— S. 4-> O,
O.M- T- w>
E X a>
R) C i- .
c/> a> »—
j*
« to 10 j=
3 4-> «/> 4J
o a>
a> >> > o>
«^- c
ra e— O ••-
cn
n> E E M
o a>
T3 <1J i- O>
••- -a ••- c
3 Q.-I-
cr c M t.
•r- >
_i o s- v>
> a*
ut o> o
>>>=•«->
to cn
•p- (/»
O ••- E
•f- Q. 3
W> >> to i-
>, « OJ 4->
r— ^ I S- W)
«0 C
C >> O) -i-
E Or— >>
O W- 0.1— .
*J S- 3 «
3 O> O C
0--0
O)
0.0
4-> Oi re ••-
o c -o s-
re -r- re *J
o v. o>
••- 3 >> E
O. >>
>» 3 Ol
S- O O
•*-> o r—
§ >> «
0)
a>
s
o o
Q) S_ O)
a) >
o.
Or-
C O
>> re
C Vt O>
f *^ «v* t*i
•r~ M .O.
S- O
O E
01 res
•O O 4J M-
— 00 -r- Q)
o s-
s- a>
•
'3 •— S.-1- to
toeccaio -r-ceu>>a)
o E JQ oi-r- •
re ••- re t. o
« i- i/i >
366
-------
LLI
O/5uo NOIldHnSNOD N30AXO
D
Q
D
LJ
I
D
LLJ
LLJ
X
o
LJ
cr
LL
367
-------
0)
>v
o
o
LJ
O
LJ
C/)
Q.
HI
>
DC
ULJ c
< o
h- T-
O. x
— N O
o
g
CD
LO
LU
DC
O/EuO NOIidWnSNOD N30AXO
O
LL
368
-------
c
«
•
•
.. 1
' C
C
u * g* • * •
. c.
1
1
1
>
1
* I
• I
- • • 1
0 C
in u
cr>
ooa
0
>^
q
13
LU
Q
LU
en
Q.
LU
LU
>
DC
Z5
O
OQ
Qt
o >-
m CE
~ uj
DC
o
LU
Q
g
CD
CD
LU
CE
CD
LL
369
-------
I I
I
_ h.
3
"1
JC
^ *
g
63
\
CL
" UJ
LU
O X
id
LU
X
O
g
CD
UJ
DC
O
LL.
370
-------
T I T - i
\ - 1
I - T
en
LU
I
CL
J ! I ! -t l -I i
= Zf •• |H »
o> CD r* to «n ^
ra -«
LU
2
LU
X
i
Q.
tr
o
LL
g
fee
Q
tt
O
LU
Q
O
00
LU
DC
g
LL.
371
-------
GO
Ul
••»
2«
UJ
M
Ul
CO
S
UJ
O
Ul
UJ 'I'
to 5
^^
UL. ??:
S 8
^^ CO
§ **
a £2-
3 §
>— i
0 P
S §
UJ gj
5^ CD
1 g
UJ CO
r~
it uj
O CJ
^™ II
i °"
GO
in
Ul
fo
^
0)
c
0}
*7T
X
1
a.
O)
c
a>
r—
^
i
E
0)
IM
0)
Ul
CU
c
a>
r—
(2
Q>
C
O)
c
cu
CO
cu
c
£
s=
,
p2.
00 00 00
o in in in
O >— « t— i i— i
in ^j- oo
o co t-n vo
o i— i CM cn
vo
t--. r--.cn
o CM CM cn
O CM CM CM
OO 1— 1 I-H
0 t*. CO r-t
o CM oo in
CO
I-H I-H in
O t— « t— 1 l-H
O CM CM ^*
00 CO «*
o in sf cn
C3 CM CO CO
o o o o
C3 I-H CM CO
•*
cn
CO
vo
cn
vo
CO
t-.
co
o
cn
00
fn^
cn
^.
in
cn
I-H
0
*
00
CO
o
CO
vo
vo
cn
CM
"*
co
cn
in
I-H
r-.
o
CM
O
•-H
O
in
00
CM
00
00
in
CM
CO
in
CM
0
CM
cn
cn
CM
CM
^.
f*K^
O
CM
0
I-H
O
vo
00
in
in
cn
t-.
vo
,3-
00
I-H
cn
CO
i-.
o
0
o
in
00
i-H
CO
o
CM
0
I-H
O
•^
cn
•*
cn
cn
"-
co
o
i«^
CO
CO
CO
f-.
00
r-H
»-H
0
r-~
CO
CO
cn
o
CM
o
i— i
o
00
vo
1^
I-H
o
cn
f-.
f».
00
cn
co
00
00
co
o
cn
CO
CO
cn
o
CM
O
I— 1
o
cn
I--
^f.
0
CM
I-H
o
cn
vo
CO
CO
00
CO
CO
0
in
**
in
cn
o
CM
o
I-H
O
o
I-H
CO
1*^
in
0
in
i-H
o
cn
vo
00
S
00
CO
0
in
**•
in
cn
0
CM
O
I-H
O
I-H
r— 1
r-. vo
vo oo
o o
r-». o
cn co
O t-i
cn cn
in o
vo o
cn I-H
co cn
00 00
co co
o o
co vo
i— i •*
vo r-.
cn cn
o o
CM CM
0 0
I-H i-H
O 0
CM CO
r— 1 i-H
372
-------
o
z
o
a.
o
o
z
NJ
LU
CQ
Q
0
LU
1
—•J
LU
00
o:
£
^•U
£~
*^
u
KH
tL
z
i-t
¥-
o
CQ
u_
o
a.
yz. I
1
to
LU
_J
CQ
JS
1— 1
». 1
(O
-o
_
r—l
1
=1 >»
to
CO
•+•» to
-o
CO
^2
CO
+•»
en o
CM to
CO CO
oo o
l-» OO
CM CO
co cr>
CO I-.
•*^
o in
. 1 — to
-S--3-
0 CD
0 0
"*"*
0 0
r— 1 i— 1
co co
r-H CM
, ^ t %
a> a>
£ £
•r™ «r~
i. i.
O> O> O)
c= o. a.
••-X X
i — LU LU
!~~
° § s s § g K
l^% _A» ^j, . ^^ *
~? 3" * r— 1 O
^ •* — 1 CM
£; CO in *f O I-H OO
m co t^ cn vo CM r>.
co oo oo en to F-( cj
§ S 5! S3 S £ £
m ** CM «a- CM in in
r~« i— i o CM in CM CD
00 CM CM CM CO i-l 0
•«*• *3- CM ^- CM in ^>
0 0 0 0 0 00
m ca ca en o *s- m
t* •* CM CO CM ^»- CO
.
52 *""" *** ^^ 1^" CM CM
S ^ *""' ^H •"" CXJ CM
co co co co co co co
a>
c
M. . [ -°
OJ fl> fl* • * ^^.
CO -Q 0) C C 3 ^
s >, 1 •£ •£ S •?
1 S S i ^ 1 ' g
Tj- CM O
CM tj-
CM CM t-4
co ^ S
CM^ CM t-4
CD to in
o in •— i
co «* in
cr> CM in
f~» en in
CM co t*-
O 0 0
CM CO «*•
O CM to
CM CM CM
CO CO CO
j
Q)
g—
a> to
s= X
O) OJ
rsi .c
c
O) i —
« -° £
S s; 5
S 4J Q.
33 l
<-> CQ -->
*
S_
3
U
a
0
QQ
((••
0
0)
CO
(0
Q.
_^=
^
2
en
5
c
0)
c
o
a.
X
a>
c2
c:
-3
CO
c
o
u
i.
1t
o
i.
en
u
{.
t
o
O)
a.
CO
n
=s.
•
0)
s.
u
o
o
CO
o
e
.>•
c
^.2
CO
o
o
a>
to
-M
0
i.
cn
0
i ,
^ '
u
a>
a.
CO
n
373
-------
LL)
Z
III
o
o
CD
LU
DC
O
374
-------
CO
^^
_
o
o.
o
o
LU
1 1 i
h-4
1 1 1
CO
Q
LU
O
i f i
i
LU
CO
££
0
It
I
C£
Lul
g
2:
o
l~*
LU
JJH
O
o
Lt_
0
oE
2
^5
CO
LU
_J
QQ
J^
J^
£
^•^
»— i
^Ps
T3
*~~*
CO
(O
^'S
o
cE
CD
^2
S1^
CO
o
1
o
t_>
'
0)
(O
S-
CO
o
•3
CO
en
CS
>>.
•o
o
1— 1
IO
in
i— i
10
00
co
o
0
o
*
0)
c
•r™
•r-
c
<:
•— i 00 hsi ' O VO O OlO If) O
CM i— i 10 r--. in CM CM r~- »*• in
CMi-tiniOCOlOCMOOCOCO
o o en co . o o en *a- r~~ in
CO CD O CO to OO CO .^- t— I .1—1
en en en oo >— i i— i CM CM in CM
i— i
*
CMCMCMCMr- 41QVOIOOOVO
OOOOOC3OOCDC3
o<3 o o oca o **• o o
^^•CMCOCM^-COCMCO^l-
.
cu
C O)
NO) 0)
•* C IM C
c o> > r— c . '
CD OJ O) Q) -|J >, (Ui—
J3>>:
£= r— 3 r— i— i CQ C i— O>
N >> 0 >* >>••!-> I ,J=
C J= i— X. X i. 0 E +-> Q-
O) -t-> O I 1 Q) COOCQ^-i
,
• CM
^^
~*
•!->
to •
•o
d) 0)
. •*-> E
fo ^3
S- CO
CO O
J3 0
3
CO O)
•»->
«4- as
O %-
C CO
o .a
•i- 3
-t-> CO
to
S T3
U 0)
-t o fc
Gl U O
+J c>_
*-••-> CO
C CO
26 co O
O C •!-
<-» c i*
i^ .2
-------
CO
a
— >
f-»
y-
f—^
CJ
-»
i— l
O
LU
a.
o
LU
1 —
0
LU
i
1 1 i
CO
CrC
0
U_
^
tw
^<
a
o
LU
£J
i
0
^•"1
CO
u.
0
>-
g
§
CO
*
CO
LU
_J
CD
'(—
1—4
•^ |
5S. >>
T3
T— 1
|
^^ >j
(O
•o
__^
CO
4J°C8
•o
^••^
..-^
CO
-l-> to
"O
*""' '
^_^
CO
t o »J*»
-a
2*
O
Q O
O 1-1
1— S--
O
CO
Q
^•j
O
Qu
f*^
t >
cn
CM
co
oo
r--
•
CM
CO
OO
^
o
^f
o
0
*
CD
i— l
CO
Qi
.^
^_
• r™
C
O
-C CO i I I
a. o; a. o s
in CM
CO O
• •
•— i in
1—4
0 •-«
00 CM
* *
t— i co
i— i
CM •*
—4 CO
r-4 CM
^t" ^"
cn ^
<^
4-^
r— O>
° -^
O Q
4^ «-^
CTJ ^
CJ CM
O '
>
s-
3
O
O
o
CO
C^_
o
a>
CO
r-
a.
.c
4_)
^z
0
s-
o>
_„
Of
•n~
C
. CO
o
a.
X
0)
s-
o
**"
rs
4^
CO
^~
0
u
0)
a)
S-
J=
4->
O
S-
C7>
0
;p*
.Jl
0
CO
a.
CO
u
si'
•
a>
s-
3
' O
a
o
CO
M-
0
CO
CO
of
Q.
.£3
4^
3
0
^
o>
•r-
C
^
u
CO
"O
s-
o
^
c
4-^
CO
^"
o
u
0)
fO
^;
4->
o
^.
m
u
;*~
Ipl
u
0)
ex
CO
II
•h
=1
376
-------
C/)
a
o
o
i— i
o
UJ
a_
a
UJ
h*~
CJ>
UJ
UJ
O
U™
1
a
0£
UJ
UJ
^r
2
•
|i
£
OO
cn
UJ
CO
1 «^
I s
1— 1
•a
oo *&
CO •»->
a>E-0
'_Q )
^ O^
£ S
a>
£^|
O^^^*"O
(TJ
__l
a
o
ex
o
o
f-H
to
10
I-H
10
oo
o
o
o
*
0)
c
•r™
c
co T* oo .-H CM r*. i-»
^r ro r^ ^*- to oo o
o> u> r->. to r-I oo *^
CM •— i f— i »a- .I.:,
-
CM CM I-H O |^« 53 f^j
00 CM i-H I-H CT» 00 tO
cn CM to ^ ivl CM in
1—1 i-H
*
OO 00 CO I-H iO CT> "O\
in *f m •** ** ** Zi
O C3 ^ C3 O C> CD
o o o o »s- m o
O 'O C3 CM ^^ OO O
^
0
£3'
c
oo o
-Q O
V)
C €
" ° c2
*- -M W)
0?°
| § 5
£ § ^
••- £ -o"
1 1 s
i S j=
s *°
*x ' — °
E
H • »
3. *f >-«
377
-------
co
o
^
1
o
0
or
LU
CO
LU
LU
g
3C
CL.
a
o
LU
til
LJ-I
CO
a;
o
LU
-
^
Ss
^s
^5^
CO
*
f—*4
LU
1
CO
J_:
^
•» i
51 <$"
-a
*»^
t— i
=l>>
-o
CO
+?1*
•o
CO
-t-^1 w '
T3
^^
CO
•fJ ro
g>
O
O O
O ft
1— S-
0
C|-
co
•^
^
o
o_
^•^
CO
cn
CM
co
CO
*
CM
CO
00
•*
0
^>
o
0
*
0
I-H
CO
O)
c
•r~
•1-
CU
a} 4->
<0 r"
g- fQ
. \ ^-
.C -(->
a. -c
a.
>» r—
•5 ^
CO -»->
£ CD
0 O
0
o
CM
O
CM
O
CO
00
CM
O
^
»
CM
i— 1
i— 1
CM
0)
to
to
^^2
+J
O.
>>
O.
o
s-
o.
5
CO
oo
CM
CM
1— «
0
00
CM
OO
CM
CM
0
o
CM
CO
CM
CM
CU
<0
"to
-C
4->
x:
CL
,.•
N*
C
CU
_a
r_
^^
•^
CO
^
O)
s-
u
a
o
ca
o
cu
J
CU
J=
+J
i
s-
^ cn
•s
'[ >
c
0
Cu
X
cu
s-
c£
ns
CO
c
0
u
0)
i.
.£.
2
o
s.
cn
o
•r-
•p-
O
cu
Cu
' CO
a.
cu
s-
.0
o
o
CO
o
cu
CO
Cu
J=
o
cn
» F5^
Vrf*
•^
"o
cu
-o
s-
£
c
(O
CO
o
u
OJ
s-
~j
3
0
s-
o
u
•^
/J
*F"
•r"
u
cu
c
VI
II
"a
378
-------
0
1
0
UJ
kw
CO
1 . 1
UJ
<
^J
H-
£
o
UJ
h-
UJ
UJ
oo
C£.
0
u_
- -i— 3
J3 00
__ ^}
S g:
0)
£
giit-g
CO
a
o
a.
z
0
0
i— i
10
LO
*
00
co
0
o
o
•*
(U
c
<:
oo r>» i— i to
IO IO OO CM
•— 1 i— 1 LO IO
* •— • •— • co
£5 o oo ' o
o o r-~ oo
f"^ COa LO t**
CO *O 00 i-™*
*^ ^^ ^i" vo
0 0 O C3
._
*o o o o
** O «<*• O
CO CM CM CM
0)
"«
J»Z CO <~
•M ^= •«-> i—
J= . . 4J J= >,
OL _c a. IM
^^ ^^ rt^
>i i— >j J3
** J= O r—
S- >,
E E
i- 00
-»-> c
00 O
-a u
4->
l«- ro
0 S-
C 00
0 JD
•i- 3
•M 00
&• O)
a» -a
o a>
c: E
0) ° ' *-
to **"
*- +•» oo
J= e "^
•g -3 e
o c -°
™ ° -o
O> ^ "^
9s 3 ••-
VJ • % ^fc.
T" ^^
E i^ «-
1 *.. 1
g 5 0>
- „ II II
II
55. *£ >-CT
379
-------
It— 1
^ «*
•o
CO
a
**•*
o
a.
0
UJ
^^
!••_
r~~
~~
2rf£
O
UJ
/^
t.t_i
UJ
CO
^^
o
L™
^^
L__
s
o
s
£2
CD
»— i
O
1
n
CO
CM
1—1
UJ
CO
>-
r- 1
1
53. >»
t«
-a
^-^
CO
C\J ^^
1 \ J!^
•o
^-l*
CO
T3
CO
-o
t
O
o o
O I-H
1— S-
0
CO
a
^
o
a.
o
0
en
CM
CO
CO
CM
CO
CO
^4"
o
*
•*
o
o
"*
0
t— 1
co
c
c
CO CO
en en
CO * "
in r-t
• *
CM CM
CO m
i— 1 CO
• •
**• CM
cn o
,cn CM
• •
CO CM
0 0
l-~ O
• ' •
CO CM
f— * ^
CM ^-
CM CM
a>
o
0) C
C fO
o +->
a> QQ
U I
«C CM
o
CO
^
1— 1
CO
CM
in
co
•
CM
CM
•
CM
in
CO
1— 1
CM
CM
**
CO
co
o
CO
r-.
o
o
^*
•
in
CM
0
f*.
•
CO
CM
O
CO
CM
CM
CO
CM
a>
o
s-
o
a.
o
CO
*"*
0)
s-
0
a
o
CQ
0
(U
CO
f*
a.
j=.
2
O
i.
O)
^
•r«
0)
c
0
o.
X
(U
s-
Q
c
(O
CO
0
u
O)
1 ^
•T^
s-
.c
1
s-
o>
u
•r-
l*~
u
0)
a.
CO
II
51
3
O
a
o
(U
CO
rtJ
JS
a.
jz
+j
•s.
0
o>
o>
•r—
C
•t—
r— •
U
(U
T3
"~
C
CO
^~
o
u
a>
<«
JC
1
s-
O)
u
•r-
.J^
U
-------
oo
a
o
a.
O
LU
o
LU
a
LU
LMB
o
LU
—1
LU
OO
o
L^
^c
2
LU
H"
LU
g
S
<_>
l~*
LU
1— H
a
0
o
LL.
O
>-
<
g
00
i—
01"^
• E
^™
•"^
i
=if «J
*—
)
(/!
>-°!°
.
o
E
CO
-l->
ft
fj
s
h
*
1*
CO
^_)
ftj
i-
=
C|>
.
ea
en — -o
_j
« 1
l~* 1
LU
03
OO '
Q
O
o
0
o
i-H
to
m
r-H
to
03
CO
CD
0
o
T
-
CO
c
•f—
c
10 en o CM
r-~ r--» t-. <^-
cn cs ^ iv!
•-I CM CM
*O 1-H O • IV. '
CO i—i tf if)
•>*• in ib i— i
10 en m co
co co «ah «»•
o o c> CD
o CD in CD
|v» O CO CO
CO CM i— 1 CM
CM
o
CO
c
o
fO
c
CO
OL
CO CM CO
0 ^ § '
CO C >, s-
O +J +J .C
-M 3 CO Q.
CO CO Z 0
O.i I v>
•
^
•d
CO CO
•4^ E
fB 3
i- OO
+-> c
oo o
-a o
oo CO
o &.
C 00
0 J2
-(J 00
S- 01
co -a
O CO
0 § ^
^ .~ **"
*- -4-> 00
t ._ C 00
o "* °
• fe Is
^ _o
•r~ i_\ -.—
O (a
2- 3 5
4/1 -M >,
§ f> J=
"itf r— O
g .*= cn
I. " •
381
-------
-------
PREDICTION OF BIODEGRADATION KINETICS OF TOXIC ORGANIC COMPOUNDS
Sanjay Desai, Chao Gao and Rakesh Govind
Department of Chemical Engineering
University of Cincinnati
Cincinnati, Ohio 45221
and
Henry H. Tabak
U. S. Environmental Protection Agency
Office of Research and Development
Risk Reduction Engineering Laboratory
Cincinnati, Ohio 45268
This paper has been reviewed in accordance with the
U. S. Environmental Protection Agency's peer and
administrative review policies and approved for
presentation and publication.
Prepared for Presentation at:
Twelfth United States/Japan Conference
on Sewage Treatment Technology
Cincinnati, Ohio
October 12-13,1989
383
-------
PREDICTION OF BIODEGRADATION KINETICS OF TOXIC ORGANIC COMPUNDS
ABSTRACT
Biodegradation is the most important mechanism in controlling the concentration
of chemicals in an aquatic system because it can mineralize toxic pollutants to
harmless forms. So the fate of organic chemicals in an aquatic environment is
dependent on the susceptibility to biodegradation. Because of the large number of
chemicals, it would be expensive and labor intensive to gather this information' in a
reasonable amount of time. Hence, there is need for a prediction method to obtain
these data.
The paper describes a predictive biodegradation—structure/activity model based on
the group contribution approach using biodegradation kinetic data obtained from the
literature and through electrolytic respirometry studies. The model, incorporating
chemical properties and structure of organic compounds, predicts biodegradation
kinetics for related toxic compounds. Biodegradation kinetic data for representative
RCRA toxic organics presented in this paper, were used to establish a pattern of
predictability for biodegradation of related compounds with the group contribution
predictive model.
Results are presented on the development of the structure-activity relationship
for biodegradation using the group contribution approach. With the use of this
approach, reported results of the kinetic rate constants agree within 20% with the
predicted values. Additional compound studies are essential to further extend the
methodology.
384
-------
INTRODUCTION
,h i estimated that 50,000 organic chemicals are commercially produced in
the United States and a large number of new organic chemicals are added to the
JP/°a Sr?nnf fJ^J^X**™?^ **"* of these ^micals in the environment
disposal\iechn\ues Problem- Their presence could be attributed to inadequate
t**» -SlnCe many* °f these hazardous chemicals can be detected in wastewater, their
fate in wastewater treatment system is of great interest. Of the many factors that
fmnnLn,6 Sf6 2J. theus.eucomP°u.nds' mfcrobial degradation is probably the most
important (2). The high diversity of species and the metabolic efficiency of
microorganisms suggest that they play a major role in the ultimate degradation of
these chemicals (3). Biodegradation can eliminate hazardous compounds by
biotransform.ng them into innocuous forms, degrading them by mineralization to
carbon dioxide, water and other simple and harmless molecules. So the information
regarding the extent and the rate of biodegradation of these chemicals is very
important for regulation of their manufacture and use. Due to the large number of
these chemicals, gathering of this information in a reasonable amount of time would
be both expensive and labor intensive. Thus, the only practical way out is to develop
correlations and predictive techniques to assess biodegradability (4). Lack of an
adequate database on biodegradation kinetics has hindered the development of such
techniques.
EXPERIMENTAL TECHNIQUES
There are many techniques for collection of biodegradation data and these are
reviewed in great detail by Howard, et al. (5) and Grady (6). Experimental techniques
for biodegradation data collection fall into three broad categories : continuous fed-
batch and batch reactor systems. '
Continuous culture reactors require an acclimated biomass. They also require long
transitional time intervals for reaching quasi steady state condition (7). So it is time
consuming, tedious and expensive. However, the analysis of data obtained is simpler
because the equations for continuous reactors, operating at steady state, reduce to
algebraic equations that are easily solved. This technique is used more for evaluating
parameters for the design of treatment systems rather than for biodegradation kinetics
.. , Fed-batch reactors have also been used to estimate the kinetic parameters of
biodegradation. In these reactor configurations, a quantity of biomass is added into a
reactor and a substrate stream is continuously added in negligibly small amounts with
respect to the reactor volume. Because of small amounts of both substrate and
microorganisms, a pseudo steady state is achieved. So this technique reduces the time
required and also alleviates the problems associated with changes in the composition
ot the microbial community. This configuration requires an acclimated biomass because
the microbes must be capable of responding instantaneously to the input of substrate
This technique cannot be used to determine the Monod biokinetic parameters, but it is
an excellent procedure to determine the parameters for treatment plant design or
operation. KB
The use of batch cultures in biotechnology and biological wastewater treatment
represents a less expensive and much faster way to model biokinetics in fermenters
385
-------
and in activated sludge tanks (7). The batch method commonly used for a large
number of compounds (10) is one in which the substrate of interest, at different
concentrations, is inoculated with the small amount of biomass. Then the increase in
biomass concentration in each reactor is measured as a function of time. Another
technique focuses on the substrate removal rather than the microbial growth. This is
commonly used in engineering studies. The batch reactors are inoculated with large
quantities of biomass and the substrate removal is measured as a function of time.
Both of these techniques have been widely used to measure parameters such as five
day BOD and COD. Batch reactor can be used with either acclimated or unacclimated
biomass. It requires that samples be taken at discrete time intervals during the course
of biodegradation. If unacclimated biomass is used, the number of samples required
may be large depending on the acclimation time. Tabak, et al. (11) have collected
degradability and acclimation data on 96 compounds by static culture screening
procedure and culture enrichment process. Larson and Perry (12) and Paris, et al. (13)
have done biodegradation studies with unacclimated biomass in the batch reactors and
have evaluated the kinetic parameters.
In the past, the number of data points obtained was small because of manual
sampling. This can be avoided by monitoring oxygen consumption as an indirect
measure of biodegradation using an electrolytic respirometer. The automatic data
collection and recording allows sufficient accumulation of data, that the reliability of
the kinetic parameters evaluated is maximized.
In the respirometric methods of BOD measurement, wastewater samples are kept
in contact with the gas phase source of oxygen. Oxygen uptake by the microogramsms
over a period of time is measured by the changes in volume or pressure of the gas
phase. An alkali is included in the apparatus to absorb carbon dioxide produced during
biodegradation. Samples are usually incubated at constant temperature and are kept
away from light. The latest development in respirometric techniques has been the
advent of an electrolytic respirometer. It supplies oxygen, produced from electrolysis
of water, to the air space above the sample in a completely sealed reaction vessel.
The production of oxygen is triggered due to the pressure changes in the reaction
vessel. Studies by Larson and Perry (12), Young and Baumann (14), Tabak, et al. (15)
and Dosanjh and Wase (16) have shown that the electrolytic respirometer eliminates
most of the technical difficulties associated with other methods for determining BOD.
It is particularly useful for the rate studies because it provides both, a continuous
record of oxygen uptake and it maintains an unchanging atmosphere over the sample
regardless of the length of the test.
EXPERIMENTAL SET-UP
The electrolytic respirometer Sapromat B-12, consists of a temperature controlled
waterbath, which contains the measuring units, a recorder for digital indication and
direct plotting of the oxygen uptake curves and a cooling unit. The waterbath has l/
reaction flasks, each connected to the recorder. Each unit as shown in figure, 1
consists of a reaction vessel C, with a carbon diox.de absorber (sodalime) 3 mounted
in a stopper, an oxygen generator B, and a pressure md.cator A. The vessels
interconnected by hoses, form a sealed measuring system which assures that the
barometric pressure fluctuations do not adversely, affect the result The magnetic
stirrer 1 in the sample 2 to be analyzed provides vigorous agitation, thus ensuring an
effective exchange of gases. The activity of the microorganisms in the sample creates
a reduction in the pressure which is recorded by the pressure, indicator It controls
both the electrolytic oxygen generation and the indication and the plotting ot the
measured values.
386
-------
Jjl !• nSUmpIIOnA °f °xyge? by microorganisms creates a reduction in pressure in
°" VeSS?' ^ a -result' the level of °'5% sulphuric acid in the pressu e
drrnt f* *"*' X"168 '" C.°ntac,t wlth " platinum ^trode. This completes the
circuit and I triggers the generation of oxygen by the electrolytic cell. The oxyeen eas
•s provided to the reaction vessel, alleviating the negative pressure ^ The fevel of
ten hC 51"
... - amount of oxygen supplied to the sample is
»n RM ATr6CK-lyi, m """w*™ Per «t« by the recorder. The border is connected to
an IBM AT which records data from the measuring units every 15 minutes onnecrea lo
Figure 1. Diagram of a Measuring Unit
A. Pressure indicator
B. Oxygen generator
C. Reaction vessel
1. Magnetic stirrer
2. Sample (250ml)
3. C(>2 absorber
4. Pressure indicator
5. Electrolyte
6. Electrodes
7. Recorder
BIODEGRADATION KINETICS
A considerable amount of information concerning biodegradation kinetics is
aa?Jc • IStr . P"bl'1!hed literature. Early literature shows widely differing kinetic
rates in different studies. The evaluation and prediction of the extent and rate of
biooxidation is affected by methodological and experimental factors. Regardless of the
intlSiv-^VS"* '™°W '.nLthe measureme"t of biodegradation rates/it is
generally considered associated with microbial cell growth and so most of the models
387
-------
for it are the same as those used to model growth and substrate removal.
Although many models have been proposed for microbial growth, the Monod
relation is the most popular kinetic expression (17). The Monod model, in combination
with the linear law for substrate removal can provide an adequate description ot
microbial growth behavior. It states that the cell growth is first order with respect to
the biomass concentration (X) and mixed order with respect to the substrate
concentration (S)
dX/dt = (S n,m X) / (Ks + S)
Cell growth is related to the substrate removal by the linear law
dX/dt = - Yg (dS/dt)
(1)
(2)
The kinetic parameters of interest are maximum specific growth rate u,m, half
saturation constant Ks, (it is the concentration of substrate when P-=0.5u.m), and the
yield coefficient Ye. The Monod equation has two limiting cases. When the substrate
concentration is much greater than the saturation constant the term (5/Ks+b)
approaches 1.0 and cell growth and substrate removal are zero order with respect to
substrate concentration. When the substrate concentration is much smaller than the
saturation constant, the term (S/Kc+S) approaches (S/KS) and cell growth and
substrate removal are first order with respect to the substrate concentration. Many
researchers have used either of the above two approaches. The characteristics of
these kinetic expressions have been discussed by Simkms and Alexander (18).
The electrolytic respirometer has been mostly used to measure the extent of
biodeeradation as a percentage of the theoretical oxygen demand over a certain period
of time. Several researchers have tried to extract kinetic parameters from oxygen
uptake data. Larson and Perry (12) used empirical kinetic expressions which were
system specific. Dojlido (19) divided the oxygen uptake curve into seven different
phases and then proposed an empirical model for each phase and evaluated the
biodegradability and toxicity of a test compound by measuring.empirical rate constants
and time intervals associated with each phase. Tabak, et al. (15) d.v.ded the ?ubstrate
removal region of oxygen uptake curve into two reg.ons, separated by an inflection
point. The first period was called exponential phase where it was assumed that the
substrate was not limiting and cell growth was occuring, while the second period was
called declining phase and here it was assumed that the substrate was limiting. So
work done to obtain kinetic parameters from oxygen uptake curve has been empirical.
STRUCTURE-ACTIVITY RELATIONSHIP
The structure-activity relationships (SAR) have been widely used in pharmocology
and medicinal chemistry. The different methods and models used are free energy
models, Free-Wilson mathematical model, discriminant analysis cluster analysis, pattern
recognition, topological methods, and quantum mechanical methods.
The free energy model of Hansch, et al. (20) is widely used They incorporated
octanol-water coefficient, log P, in the linear'free energy relationship as a measure of
liooDhilicitv This provided a general SAR model for biological activity. The success or
this model has led many researchers to include additional phys.cochem.cal Parameters
and properties structural, topological and molecular indices. Using similar principles
other researchers have proposed models to include more complex relationships between
the bioactivity and the chemical structure or the properties. Martin (21) has dicussed
388
-------
these models.
subst,tuted ,n various pos.tions, resulting in a series of linear equation^ of the form
to bto to *&£*£«£"? ^ °"ly sr«Hantl^ or qualitative data have
.. the location of the substituent groups, Geating
aLg0"thtm '°. pred!ct "*>*«"»« quantification is required
In the literature, first order rate of biodegradation or five day BOD of chemicals
hlhTH ^"f* wi* Ph»sical o' chemical properties. Paris, etal S,?^)
established a correlate between second order biodegradation rate constant arid the
389
-------
van der Waal's radius of substituent group for substituted anilines and for a series of
para-substituted phenols. Wolfe, et al. (35) correlated second order alkalme hydrolys£
rate constant and biodegradation rate constants for selected pesticides and phthalate
^.fTe^afworkershave observed a correlation between biodegradabihty and
Lilt.:*.. :n^oii» «^anni/watpr nart t on coefficients (log Pi. Pans, et al. Uoj
of esters of 2,4-dichlorophenoxy acetic acid. Banerjee, et a oame a si
relationship for chlorophenols. Vaishnav, et al. (38} correlated ^degradation of 17
alcohols and 11 ketones with octanol-water coefficients using 5-day BOD data
pSer(39) has found a dependence of biodegradation rate on e'ectronicnfaf°"' ™*
Harnmett substituent constant, for a series of anilines and phenols Dearden and
Kfcholso" (40 hav correlated 5-day BOD with modulus difference of atomic charge
acrossTselected bond in a molecule for amines, phenols, aldehydes, car^xyhc acids
haogenated hydrocarbons and amino acids A direct correlat.on be ween the
biodfgradability rate constant and the molecular structure of the chemical has been
used by Govind (41) to relate the first order biodegradation rate constant with the
first oLr molecular connectivity index and by Boethling (42) to "jrfatjthe
biodegradation rate constants with the molecular connectives for dialkyl esters,
carbamates, dialkyl ethers, dialkyl phthalate esters and aliphatic acids.
GROUP CONTRIBUTION APPROACH
Using a group contribution approach, a very large number of chemicals of
interest can bl constituted from perhaps a few hundered functional groups The
Drediction of the property is based on the structure of the compound. According to
tWs method the ^molecules of a compound are structurally decomposed into functional
SouoTor their fragments, each having unique contribution towards the compound
property This ^method is similar to the Free-Wilson method used m medicinal
chemistry.
The biodegradability rate constant k, is expressed as a series f
a)ntribution ai, of each group of the compound. The first order approximation of
TnctionJ representing biodegradation rate constant can be expressed as
Ln (k) =
>
(3)
where Nj is the number of groups of type j in the compound a, is the contribution
of group of type j and L is the total number of groups in the compound.
CALCULATION OF GROUP CONTRIBUTION PARAMETER, o
s
using regression analysis.
If there are n a's and m compounds (n
-------
= 2{Ln(kj)-2Njjaj}2.
1=1 =lj JJ
(4)
The estimates of a's should be such that it produces the least value of S. These
estimates are obtained by differentiating equation 4 with respect to a and setting it
equal to zero
. m n
dS/dak = - 22 {[ Ln(ki) - 2 Njjaj ] Nki> = 0
(5)
This will generate a series of n linear equations which are solved for a's. If N is the
matrix of coefficients of a's and Y is the vector of Ln(k) values then the solution
vector a is given by
a=(N>N)-1(N'Y)
4
where ' denotes the transpose of matrix.
MODEL VERIFICATION
The contribution of several different groups was calculated from the data
obtained from the literature (43,44,45). The data used to calculate these contribution
parameters is given in Table 1. The groups and their contribution parameters are
given in Table 2.
The biodegradation experiments were carried out for cresols, phenol, 2,4—dimethyl
phenol, 2—butanone, acetone, 1—phenyl hexane and butyl benzene using an electrolytic
respirometer Sapromat B-12 (Voith-Morden, Milwaukee, Wl). The chemicals were from
Aidrich chemical company with 99+% purity. These compounds were used to validate
the model, but were not used in the calculation of the group contribution parameters.
The experimental conditions were : temperature 25°C, biomass concentration 30 mg/L
and compound concentration lOOmg^L The biomass was obtained from Little Miami
wastewater treatment plant in Cincinnati, which receives 95% domestic waste. The
biomass was aerated for 24 hours and then used for the experiment. The
biodegradation rate constants for these compounds were determined using the
following simplified Monod form, obtained by assuming that Ks » S and change in
biomass concentration is negligible. -
dS/dt =. - KX0S.
(6)
where K = (p-rn/Yg Ks) and Xo is initial biomass concentration. Integrating equation 6
we get
S = S0exp(-KX0t) (7)
In terms of BOD the above equation transforms to
BOD = ThOD (1 - exp{-KX0t}). .... (8)
The above equation was used for only the rising part of the BOD versus time curve,
and initial data and data for endogenous phase were neglected. Note that the selection
of the rising part of the BOD curve was arbitrary and was based on visual inspection
of the BOD curve. The biodegradation rate constants were also predicted using the
group contribution parameters of Table 1. The comparison of these values is given in
391
-------
Table 3 and Figure 2.
RESULTS AND DISCUSSION
The prediction values of Ln(K) are within 10% of the experimental values. It is
important to emphasize that there are several reasons for the rather large discrepancy
between the predicted and experimental values. These reasons are as follows :
1. The data used for calculating the group contribution parameters were obtained
from the literature (refer Table 1). The experimental BOD curves, obtained from
electrolytic respirometry, were approximated by an exponential relation (refer
equations 6, 7 and 8). The range of curve to which this equation was applied was
selected arbitrarily. This resulted in some error between the experimental data and the
calculated value of the kinetic constant, K.
2. The group contribution method described here is the first order approximation. It
assumes that the contributions of the groups are independent of each other. So the
rates predicted are only expected to be within an order of magnitude.
3. Biomass and experimental methods used in the different studies were different.
The studies, used for determination of group contribution values had acclimated their
biomass to the compound before using it in the experiment, while our biomass was
from domestic wastewater treatment plant and was not acclimated to the compounds
used in the study.
Inspite of these errors, the prediction is within an order of magnitude. With
availability of more data and using a detailed kinetic model rather than a first order
exponential equation, the prediction error may be reduced considerably.
ACKNOWLEDGMENT :This research was supported by co-operative agreement CR
812939-01 from U. S. Environmental Protection Agency.
REFERENCES
1. Blackburn, J.W., Troxler, W.L. Prediction of the fates of organic chemicals in a
biological treatment process — An overview. Environ. Prog. 3: 163, 1984.
2. Howard, P.H., Saxena, J., Sikka, H. Determining the fate of chemicals. Environ.
Sci. Tech. 12: 398,1978.
3. Alexander, M. Biodegradation of chemicals of environmental concern. Science 211:
132,1980.
4. Strier, M.P. Pollutant treatability : A molecular engineering approach. Environ.
Sci. Tech. 14: 28r 1980.
5. Howard, P.H., Banerjee, S., Rosenberg, A. Review and evaluation of available
techniques for determining persistence and routes of degradation of chemical
substances in environment. 560/5-81-011, U.S. Environmental Protection Agency,
1981.
6. Grady, C.P.L., Jr. Biodegradation : Its measurement and microbiological basis.
Biotech. Bioengg. 27: 660, 1985.
7. Braha, A., Hafner, F. Use of lab batch reactors to model kinetics. Wat Res. 21:
73,1987.
8. Chudoba, J., Cech, J.S., Farkac, J., Grau, P. Contro of activated sludge
filamentous bulking ; Experimental verification, of a kinetic selection theory. Wat,
Res. 19:191,1985.
9. Bull, A.T., Slater, J.H. in: AT. Bull, J.H. Slater (eds.), Microbial Interaction and
392
-------
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
n^T"?leIVV°Lil:AuCad!m,lc P:ess«. New York., NY, 1982.
°f
to
• Effects of nutr^VV «n «tK'netl<; concePts,fo/ measuring microbial rate constants
. cnects or nutrients on rate constants. ADpL Env ron Miemhinl «M- wi loa*
Young, J.C., Baumann, E.R. The electrolytic^ respirome er - II -Use ?n water
pollution control plant laboratories. Wat. Res 10-1141 1976
Tabak, H.H., Lewis. R:F.. Oshima A Pi-^olytic' respi'rometry biodegradation
method for J ' ' '
>. Environmental
^^
v
NewYork, 1978". Quan"tatlve
chemistry. 1. Med Ch.m wfl 1976
Desi*n : A Critical IntroducJion? Dekker,
Chemistry. Vo,.
of
.
.....
Dunn, W. J. Ill, Wold, S. Structure-activity' analysed by pattern recognition •
The asymmetric case. J. Med. Ch«m n-vx 1930 p»«ern recognition .
'
Ac?demicBpVessa"NY- ,^6Mole<:ular Connecti.it, in Chemistry and Drug Research,
RMltronships of Drugs' Ed-
. •
n
in water- EPA-
- Realtionships between properties of a series of anilines
^act!ria- ADD|- Environ ^irrfthioi «*".••"»*
LL>' Stecn' W" C' Structure-activity relationships in
H P, ,baCtf ^ ADDl. Environ Mirmhinl 53:971, 1987. ?
D. F., Wolfe, N. L, Steen, W. C., Baugham G L Effect of
molecular structure on bacterial transformation rate constants in p?nd
393
-------
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
<;amnles ADD! Environ. Microbiol. 45:1153, 1983. .
Wolfe NLPr Paris, D.F., Steen, W.C, Baugham GL Correlation of m.crob.a.
degradation rates with chemical structure. Fnv.ron Sc. Tech U.1 431980
KrSriTKrWoife, N. L., Steen, W. C. Microbial transformation of ester of
B^e^ A- J-. S^ «"
Tullis D. L. Development.of a general kinetic model for biodegradation and its
application to chlorophenols and related compounds. Fnviron. Sci. Tech. 18.416,
Vaishnav D. D., Boethling, R. S., Babeu, L. Quantitative structure-
biodegradability relationships for alcohols, ketones and alicychc componds.
Chemosphere. 16:695,1987. romnnnnHs and the rate
Fitter, P. Correlation between the structure of aromatic compounds ana wei rate
of their biological degradation. Collection Czecoslovak Chemical Comm. 49.2891,
Dearien, J.C., Nicholson, R.M. The prediction of biodegradabilities by the use of
quantitative structure-activity relationships :. Correlat.on of biological oxygen
demand with atomic charge difference. Pestici. Sci. 17: 305,1986. ¥9nfM o.
Govind, R. Treatability of toxics in wastewater systems. Hazardous Substances 2.
Boetnnng, R-S. Application of molecular^topology to quantitative structure-
biodegradability relationships. Environ. Tox. Chem. 5: 797,1986. MH^n>tr
Urano, K., Kato, Z. Evaluation of biodegradation ranks of pr.or.ty organic
compounds. J. Hazardous Mtl. 13:147,1986. .
Stover E. L., Kincannon, D.F. Biological treatability of. specific organic
compounds found in chemical industry wastewater. Paper presented at 36th
SiVduV Industrial Waste Conference. Purdue University, West Lafayette, IN, May
12—14 1981
Environ Corporation. Characterization of constituents *"** fdaftaA waste
streams listed in 40 CFR section 261 DRAFT profiles. Report for EPA, 1984.
394
-------
TABLE 1. DATA USED TO CALCULATED GROUP CONTRIBUTION PARAMETERS
No. COMPOUNDS Molecular formula
1 Ethyl alcohol C2H50H
2 Butyl alcohol C4H9OH
3 Ethylene glycol C2H4(OH)2
4 Aceticacid CH3COOH
5 Propionic acid C2H5COOH
6 n-Butyric acid C3H7COOH
7 n- Valeric acid C4H9COOH
8 Adipic acid C4H8(COOH)2
9 Methyl ethyl ketone CH3COC2H5
10 HexamethylenediamineC6H12(l\!H2)OH
11 n-Hexylamine C6H13NH2
12 Mono ethanol amine C2H4(NH2)OH
13 Acetamide CH3CONH2
14 Benzene C6H6
15 Benzyl alcohol C6H5CH2OH
16 Toluene C6H5CH3
17 Acetophenone C6H5COCH3
18 Hydroxybenzoic acid C6H4(OH)COOH
19 Aminobenzoic acid C6H4(NH2)COOH
20 Aminophenol C6H4(NH2)OH
21 2-CI phenol CIC6H4OH
22 1,2-Dichlorobenzene C6H4CI2
23 l,2,3-Tri-CI-benzeneC6H3CI3
K(L/Mg.Hr)*E4
5.749
11.10
40.00
22.22
23.14
10.50
- 6.516
2.781
3.071
2.994
4.425
3.720
4.576
288.3
16.34
8.446
6.408
10.50
6.332
4.179
1000.
678.4
446.7
REF.
43
43
43
43
43
43
43
43
43*
43
43
43
43
44
43
43
43
43
43
43
45
45
45
395
-------
No. COMPOUNDS
Molecular formula K(L/Mg.Hr)*E4REF.
24 l-Glutamic acid C2H4(COOH)CH(NH2)COOH
25 Glycerol
26 iso— Valeric acid
27 Propionaldehyde
28 n-Butylaldehyde
29 Benzaldehyde
30 Acetonitrile
31 Benzonitrile
32 Chloroform
*
33 2,4-Dinitrophenol
34 Nitrobenzene
35 1,2— Dichlorobenzene
36 Tri—Br— methane
37 Br-di-CI-methane
38 Di—Br—CI— methane
CH(CH2OH)2OH
CH(CH3)2COOH
C2H5CHO
C3H7CHO
C6H5CHO
CH3CN
C6H5CN
CHCI3
C6H3OH(N02)2
C6H5N02
C6H4CI2
CHBrS
CHBrCI2
CHBr2CI
24.17
44.78
15.43
4.010
5.106
25.13
1.455
6.326
178.1
93.00
147.3
233.5
1.000
1.000
1.000
43
43
43
43
43
43
43
43
44
44
44
44
45
45
45
396
-------
TABLE 2. GROUP CONTRIBUTION PARAMETERS
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Group
Methyl
Methylene
Ketone
Amine
Acid
Hydroxy
Aromatic CH
Aromatic C
Chlorine
Methylidyne
Nitro
Aldehyde
Nitrile
Bromine
CHS
CH2
CO
NH2
COOH
OH
ACH
AC
Cl
CH
N02
CHO
CN
Br
aJ
-3.460
-0.1354
-0.9650
-4.061
-3.201
-2.983
-0.8340
1.973
-2.352
3.223
-2.448
-3.909
-5.272
-4.657
397
-------
r
TABLE 3. COMPARISON OF ACTUAL AND PREDICTED Ln(K) VALUES
Compound
Actual Predicted
Ln(K) Ln(K)
Error % based
on Ln
o-Cresol -6.0890 -5.8330 4.20
m-Cresol -5.7706 -5.8330 -1.08
p-Cresol -5.8659 -5.8330 0.56
2,4-Dimethylphenol -6.2472 -6.4860 -3.82
Butylbenzene -6.5297 -6.0632 7.14
1-Phenylhexane -6.7983 -6.3340 6.83
398
-------
:LO
: CD
CO
CD
CD
O
.CD
LO
O
CO
LO
O
q
CD
I
o
OJ
CD
CD
I
O
CD
CD
O
00
CD
399
-------
-------
PARTITIONING OF TOXIC ORGANIC COMPOUNDS BETWEEN WASTEWATER
AND WASTEWATER SOLIDS
by
Richard A. Dobbs
Biosystems Treatment Section
Treatment Assessment Branch
Water and Hazardous Waste Treatment Research Division
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
This paper has been reviewed in accordance with
the U.S. Environmental Protection Agency's peer
and administrative review policies and approved
for presentation and publication.
Prepared for Presentat i on at:
Twelfth United States/Japan Conference
on Sewage Treatment Technology
Cincinnati, Ohio
October 12-13, 1989
401
-------
PAPTTTTnMTNfi OF TOXIC ORGANIC COMPOUNDS BETWEEN WASTEWATER
: AND WASTEWATER SOLIDS "
ABSTRACT
An experimental protocol which eliminates the effect of biodegradation
and volatilization losses on the sorption measurement has been developed and
SJplled to uptake of toxic organics by wastewater solids Sorption
capacities obtained for primary, mixed-liquor and digested solids from
municipal wastewater treatment plants have been correlated with
SctaSol/water panition coefficients and with modified Randid indexes. It
has been shown that the correlations are the same for all three types .of
wastewater solids if partition coefficients are calculated on the basis of
organic content of the solids. The correlations developed are useful for
asSing thi role of sorption in the treatment of toxic or hazardous
^^
"™?'of the
impact of toxics on sludge disposal options.
402
-------
INTRODUCTION
Sorption on solids is one of the fundamental processes control linq
removal of toxic organic compounds in biological wastewater treatment
{Jin A J a!S °f hydr°Phobic organic pollutants in wastewater is highly
dependent upon their sorptive behavior. The degree of sorption not only
affects a chemical's mobility, but also is a dominant factor in fate
processes such as volatilization and biodegradation. Organics may
accumulate in sludges at concentrations several orders of magnitude greater
than the raw wastewater concentrations. Concentrations factors of 50-39 000
have been reported (1, 2). ««,vuu
thP nl!!t?l!nnerS0lJ*1°ns*w!!1
-------
(4)
The Freundlich equation is a useful method for treatment of sorption
data. Since X = C0-C , for a given volume of solution, substitution of C0-
Ce for X in Equation (2) yields:
Cr /u vr n fKA
0-Ce/M = KCe \*>)
which gives removal of toxic organic as a function of wastewater solids
concentration. However, to relate the logarithmic form of the Freundlich
equation (Eq. 3) to the equation for the partition coefficient (Eq. 4), one
must assume that the slope (n) is unity. In most cases where the
concentration of toxic organics in the aqueous solution are low, the slopes
are generally close to unity. Deviations can result from difficulties in
measuring the sorption .isotherm caused by biodegradation, volatilization
losses, or analytical problems.
In the present study, partition coefficients were measured for selected
toxic organic compounds on raw, mixed-liquor, and digested wastewater
solids. Sorption data were correlated with fundamental properties of the
compounds studied.
MATERIALS AND METHODS
REAGENTS
Organic compounds were obtained from Eastman Kodak Co., Rochester, NY;
Aldrich Chemical Company, Inc., Milwaukee, VII; and Chem Service Inc., West
Chester, PA.*
ANALYTICAL METHODS
Analytical methods for analysis of the selected organic compounds used
in the validation study are summarized in Table 1.
Volatile chlorinated organics were analyzed using a Tracer LSC-2 sample
concentrator and a Tracer 222 gas chromatograph equipped with a Tracer 700
Hall electrolytic conductivity detector. Pesticides and the cresol were
analyzed using a Varian Model 3700 gas chromatograph equipped with a Van an
CDS-111 chromatographic data system. Dyes were determined with a Beckman
Model 25 spectrophotometer equipped with a clinical sipper system.
COLLECTION OF WASTEWATER SOLIDS
Wastewater solids were collected from municipal wastewater treatment
plants within a fifty mile radius of the Risk Reduction Engineering
Laboratory, Cincinnati, Ohio. Wastewater plants were selected to include a
wide range of domestic/industrial contribution to the total plant flow.
Samples of primary sludge, mixed-liquor solids, and digested sludge were
*Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
404
-------
.
I'eni.rirugation. The collected solid*: (annw\vima+oii/ i/v>/ ~ j . .
wr.^STp£.M;»c:;?J"v^ s.MSrSj'Knn,
Borage perioa required for completion of experimental testing for
USED FOR ANALYSE OF TOX.C
TABLE ,. SUMMARY OF ANA™
Compound
Methyl ene chloride
Chloroform
1 , 1 -di chl oroethyl ene
Carbon tetrachloride
Tetrachloroethylene
Chlorobenzene
Analytical method
.
Purge and trap
Gas chromatography
Conditions
Tenax trap; 3%
Carbowax 1500 on
Chromosorb column
Hall detector
Lindane
Heptachlor epoxide
Dieldrin
Hexane extraction
Gas chromatography
1.5% SP-2250/1.9%
SP-2401 column
Electron capture
detector
2,4-dinitro--cresol
Freon extraction
Gas chromatography
1% SP-1240 DA on
Sepelcoport column
flame ionization
detector
Methylene blue dye
Azo dye 88
Azo dye 151
Visible spectroscopy
665 nm
505 nm
512 nm
405
-------
EXPERIMENTAL PROCEDURE
Experimental Protocol Development
issues to be resolved in the development of a Pro*?
-------
will be found In the wastewater solids as a natural background. If this
occurs, it is easier to correct analytical values of the equilibrium
concentrations in the samples by a constant amount.
' CO
£
co
60
X
1000
100
!
10,
1.0
•>
—AM
ETHOD 1 CONSTANT COMPOUND; VARIED SOLIDS
ETHOD 2 CONSTANT SOLIDS: VARIED COMPOU
o'.oi ' ' ' 4"6'.i ' '
I
/
^
T
'
J. K
' poi
1.0
/I
*
,
y
Kj - 22.2 r - 0.99
K2 - 22.2 r - 0.98
3led - 22.6 r • 0.98
„
10
1 i
ILu
, *
-rr*
1
•JT
hrr
100
RESIDUAL CONC. (C ), mg/1
Figure 1. Comparison of two isotherm methods
for sorption capacity measurements.
A great deal of confusion exists in the literature regarding the rate
of attainment of sorption equilibrium. Sorption kinetics for phenol and
1,4-dichlorobenzene were reported to be exceedingly fast .(equilibrium in < 4
minutes) (7). In the case of polynuclear aromatic hydrocarbons on natural
sediments 35 to 60% of the equilibrium sorption capacity was achieved in
minutes (mixing limited) with apparent equilibrium after a few hours (8).
Based on the conflicting reports in the literature on the rate of
attainment of equilibrium, kinetic measurements on the sorption of methylene
blue dye on mixed-liquor solids were made to determine the proper contact
time for experimental purposes. Mixed-liquor solids from three different
municipal wastewater treatment plants were collected. Rate of sorption of
methylene blue dye from a 10 mg/L solution was measured at 0.1, 0.2, and 0.5
g/L dry weight solids. All three mixed-liquor solids concentrations from
the same wastewater treatment plant used in the isotherm runs are presented
in Figure 2.
407
-------
10
60
B
8
a
M
tn
w
os
1.0
0.10
0.13g/l solids
0.20g/l solids
0.50g/l solids
0
100
200
400
500
600
300
TIME (MIN.)
Figure 2. Kinetics of sorption of methylene
blue dye on mixed-liquor solids.
The kinetic data for the sorption of methylene blue dye are in
aareement with the two box model proposed for hydrophobia pollutants in
water (9). The two box model described sorption dynamics in terms of a
rapid or "labile" exchange and a highly retarded or "non-labi e" sorption
requiring days to weeks to occur. In the case of methylene blue dye and 0.5
g/L mixed-liquor solids dose, 92% of the compound was sorbed in
approximately 30 minutes, while the remainder required seve^l hou^. In
order to minimize biodegradation during the sorption test, an equilibration
period of 6 hours is recommended. The remaining organic compounds were
equilibrated for 6 hours. This avoids overnight contact and permits samples
to be separated from the solids the same day.
The third issue to be resolved in the protocol development studies
involved methods to eliminate biological degradation of test compounds
during the partitioning process. It is obvious that other removal
mechanisms such as biodegradation and volatilization wo"™ haveto be
controlled if accurate sorption data are to be obtained. 9™*™";, inrl,,rfP
measurements of changing aqueous phase compound concentrations would include
the effects of all removal mechanisms resulting in apparent sorption
capacities greater than actually achieved. Volat ization can be eliminated
frbm sorption measurements by using completely filled containers with zero
headspace. The impact of biodegradation on the sorption measurem^!m;"
more difficult to control. Lyophilization followed by dry heat treatment
has been reported to produce biological solids which had flpeculation and
settling properties which were indistinguishable from live biomass (7).
408
-------
These authors claimed the batch experimental technique with nonviable
biomass appeared to give realistic results compared to tests with viable
nJTIXiAniJX ?" *te r?P?rt!d successful use of lyophi 1 ization for control
of biological activity, this technique was evaluated. A batch of mixed-
liquor solids was collected and separated from the supernatant in a
perforate bowl centrifuge. The moist solids were divided into two portions
One portion was freeze-dried and the other was stored at 4-C. Sorption
,„ were measured using 2,4-dinitro-o-cresol. The isotherms were
at pH 3.0 for two reasons. First, the low pH would inhibit or
+h x biological activity in the test with viable biomass. Secondly,
the decrease in pH would convert the 2,4-dinitro-o-cresol to the
undissociated (sprbable) form (pka = 4.35). Results of the isotherm
comparison are shown in Figure 3.
to
to
S
pa
of
o
CO
X
0.01
6.01' ' ' *'6'
RESIDUAL CONC. (C )» Bg/1
Figure 3. Comparison of freeze-dried and viable
biomass for sorption of 2,4-dinitro-o-cresol
at pH 3.0.
409
-------
The use of freeze-dried solids in the isotherm procedure was not
acceptable based on the results shown in Figure 3. The viable biomass
clearly has a significantly greater sorption capacity than the freeze-dned
material. A similar result has been reported for sorption of azo and
triphenylmethane dyes on microbial populations (10).
Alternate'techniques to control biodegradation were evaluated.
Wastewater samples are often preserved prior to analysis by adjusting pH to
3 or less. Preliminary experiments indicated that this approach would
minimize biological activity for primary and mixed-liquor solids but did not
completely inhibit digested sludge. In addition, the technique would be
limited to neutral organics, since pH change would alter charge and
dissociation of acidic and basic organic compounds which,would impact
partitioning.
Two additional methods were evaluated as potential treatments to
eliminate biodegradation during the equilibration period in the isotherm
test Pretreatment of the wastewater solids with cyanide was attempted
based on the reported potent inhibition of methanogensis and carbon
Solism by this aniSn (11, 12, 13). Carbon tj^1;"^"""!6^ 3S
a model compound to evaluate the required contact time to ^activate the
biomass and the ratio of cyanide to dry weight solids required to minimize
biodegradation over the 6-hour period recommended foreg"1]™™!1™- 11C.
Results indicated that treatment of the wastewater solids for 24 hours using
a cyanide to dry weight solids ratio of 0.1-0.2 was optimum to P^yent
biodegradation of the easily degradable carbon tetrachlpride. _In the second
method wastewater solids were pasteurized at 85'C for thirty minutes prior
to being used in the isotherm procedure. The effectiveness of the two
methods for eliminating biodegradation is compared in Table 2 where whole
samples (solid and liquid phases) were analyzed for five chlorinated
solvents initially and after 6 hours of mixing.
TABLE 2.
EFFECT OF CYANIDE ADDITION AND PASTEURIZATION ON BIODEGRADATION
DURING ISOTHERM EQUILIBRATION
Concentration of toxic.
Control
Compound
0 hr
Cyanide addition Pasteurization
6 hr 0 hr 6 hr 0 hr 6 hr
1,1-dichloroethane
Chi orof orm
Carbon Tetrachl or ide
Tetrachl oroethyl ene
Chlorobenzene
106
100
100
105
100
91
103
49
107
109
113
97
100
102
93
100
94
101
100
97
103
101
- 99
100
97
99
100
95
96
106
410
-------
Results in Table 2 demonstrate that both cyanide addition and
mZa^?n T? effe£t1ve f'thpcls to eliminate biodegradation during the
ho v. ? ?"ly «rbon tetrachloride showed biodegradation losses in
the control samples after the 6 hour equilibration period. Pretreatment of
HV1Ud9eS^nd S°11ds by either method will eliminate the effect of
dation on the sorption isotherm. The recommended procedure is to
an5eronin^S ft ^ m*thod °Vhoice» separate fromthe supernatant
^ fh°Uld °0t
of the preliminary experiments a general overall
i- o,,tnnoH <„ c-«; ~;'°n the Petition coefficient (sorption-isotherm)
• is outlined in Mgure 4. .
MEASUREMENT OF SORPTION ISOTHERMS
In the present study, the following general procedure was used. The
moisture content of the solids from the perforate bowl centrifuge was
determined by drying a portion to constant weight at 110'C. A stock slurrv
was prepared to give the desired concentration of solids for the Isothera
test. Equal aliquots of the stock slurry were added to a series of
C^t*;i|uge b?ttlef- A different concentration of the organic compound was
added to each centr fuge bottle from a stock solution of the toxicin
rnethanol. The bottles were completely filled and capped to avoid loss of
voiatiles to the headspace or atmosphere. , Isotherm samples were mixed on a
multiple magnetic stirrer at approximately 100 rpm for six hours. After
equilibration,, the solid and liquid phases were separated by high-speed
centrifugation and analyzed. Blanks, standards, and whole samples (both
an'alwP/fn/nMlvf J T* also Processed through the procedure and
analyzed for quality control purposes.
fnr isotherms were measured using the general procedure described
rhla sele£ted organic compound on the three types of wastewater solids. .
in FigSreS5rU tOXiC Or9an1c compounds investigated are shown
are suSarlze'Sln Table's!^ parameters for sorPt1on °« -Ixed-llquor solids
fnr J?hS?l°n C°?ffi^ents on mixed-liquor solids ranged from 0.056 mg/gm
for methylene chloride to an extrapolated value of 38.9 mg/gm for dieldrin
Mn!^11?^11"1 "ncentrat1o"s of 1.0 mg/L. Slopes (n values) were generally
close to 1.0 and ranged-from 0.57 to 1.28 for sorption on mixed-liqSor
of o ^'to lT?nhf°J f°;pj!?? °" S011f nave been reported to range from a low
of 0.3 to a high of 1.7 (14). Compiled data on 26 chemicals (mostly
Hw0!???* h?J a mean va]ue ?f n of °-87 Wlt" a coefficient of variation of
to HP iiSi tlVnmea?UrediValUe 1s ?ot available, it is frequently assumed
'ohtSnSJ *L !n !; ^x";ient correlation coefficients (r values) were
obtained for all the isotherms measured.
Isotherms for the chlorinated solvents listed in Table 3 were measured
SiyfnS ?ry • -d9? and anaerob1ca1ly digested sludge to complete the
testing of municipal wastewater treatment plant solids. Freundlich
parameters obtained are presented in Table 4.
411
-------
I ISOLATE SOLIDS
FILTRATION
CENTRIFDGATION
[OPTIONAL BUFFER WASH
CYANIDE OR PASTEURIZATION TREATMENT
[EQUILIBRATION |
VARIED AMOUNTS OF SLUDGE
CONSTANT CONCENTRATION OF COMPOUND
mNSTANT AMOUNTS OF SLUDGE
VARIED CONCENTRATION OF COMPOUND
WHOLE SAMPLES
SUPERNATANTS
[ANALYSIS]
^PECTROSCOPY
UV VISIBLE
CHRQMATOGRAPHY_
GAS LIQUID GC-MS
OTHER
TOC TOX TKN etc.
RATE CONSTANTS Ks
FREUNDLICH PARAMETERS X/M, K, n
Figure 4. Protocol for sorptlon of toxics on wastewater solids.
412
-------
cr
Azo Dye 88
Methylene Blue
e o
N«O,S
MO
Heptachlor Epoxide
Azo Dye 151
Lindane
T
CI—C—H
ei
Methylene Chloride
Cl
Dleldrin
Cl
V
2,4-dinltro-o-cresol
I1
Cl— C— H
Cl
Chlorofrom
P
Chlorobenzene
Cl M
Cl— C=C— M
1 » 1 *H4 r« Vil n V«^^^!B«*^ ..__
Cl— C— Cl
Cl
Carbon tetrachloride
Cl Cl
Cl— C=C— Cl
Tetrachloroethyl*
Figure 5. Structures of organic compounds used In sorption
studies.
413
-------
TABLE 3 SUMMARY OF FREUNDLICH PARAMETERS FOR SORPTION OF TOXIC OR6ANICS
. ' ON MIXED-LIQUOR SOLIDS
Compound
Methyl ene chloride
Chloroform
1,1-dichlorethylene
Carbon tetrachloride
Chlorobenzene
2,4-dinitro-o-cresol
Tetrachl oroethyl ene
Lindane**
Heptachlor epoxide**
Dieldrin**
Methyl ene blue dye
Azo dye 151
Azo dye 88
K, (mg/gm)*
0.056
0.094
0.150
0.497
0.285
0.420
0.897
0.762
17.9
38.9
22.6
8.30
1.35
n+
1.28
0.90
0.71
0.57
0.96
0.90
1.12
1.00
1.09
1.11
0.82
1.00
0.65
r#
0.85
0.98
0.96
0.95
0.96
0.96
0.98
0.99
0.98
0.99
0.98
0.99
0.99
*K » sorption coefficient = X/M at Ce = 1.0 mg/L
*n - slope of isotherm plot
*r « correlation coefficient
**K values extrapolated to Ce - 1.0 mg/L for comparison purposes
414
-------
TABLE 4. SUMMARY OF FREUNDLICH PARAMETERS FOR SORPTION OF CHLORINATED
SOLVENTS ON PRIMARY AND ANAEROBICALLY DIGESTED SLUDGES
Compound
Primary sludge
K,(mg/gm) n
Digested sludge
K,(mg/gm) n
Methyl ene chloride
Chloroform
1 , 1 -di chl oroethyl ene
Carbon tetrachloride
Chlorobenzene
Tetrachl oroethyl ene
0.188
0.267
0.142
0.339
0.483
0.596
1.08
0.81
0.54
0.41
1.05
1.31
0.75
0.99
0.93
0.98
1.00
0.98
0.075
0.054
0.130
0.203
0.285
0.698
0.80
0.90.
0.82
1.18
0.76
1.47
0.94
0.61
N. ',
0.96
0.97
0.98
1.00
DISCUSSION OF RESULTS
In soil and sediment sorption, the extent to which an organic compound
partitions itself between the solid and solution phases is expressed in
terms of a parameter K , which is largely independent of the properties of
the soil or sediment fr!6, 17). Once.K has been determined for a soil or
sediment, it is divided by % organic carbon (OC) contained in the soil or
sediment and multiplied by 100 to obtain Koc. In the case of wastewater
solids used in the present study, organic carbon content was based on weight
loss on ignition at 600'C and is defined by the following equation:
K • ifip_
%VSS
(6)
wctre °^ " amount sorbed per unit weight of organic matter in solids and
VSS = volatile suspended solids. Sorption data are usually correlated based
on organic carbon or organic matter. Therefore, the sorption coefficient,
K, must be converted to K for correlation purposes. Thus, we can define a
corrected partition coefficient by the following equation:
1000
(7)
Emphasis on the environmental impact of toxic organic chemicals has
resulted in increased reliance on the physical-chemical properties of these
compounds to assess their environmental behavior. One such parameter, the
octanol/water partition coefficient (Kow), has proved useful as a means for
predicting soil adsorption (18), biological uptake (19), Tlpophilic storage
(20), and biomagnification (21-24). Kow is defined by the following
equation: •
415
-------
(8)
where C - equilibrium concentration of organic compound in octanol layer
and Cw - equilibrium concentration of organic compound in water layer.
Based on the soil-sediment literature, an attempt was made to correlate
sorption on wastewater solids with the octanol/water partition coefficient
for the compounds used in the present study. Sorption coefficients in
Tables 3 and 4, were converted to K values using Equation 6. The Kom
values obtained were used to calculate corrected partition coefficients (K
values) using Equation 7. Log of the average K ' values for sorption of the
toxic organic compounds on the three different types of wastewater solids
are summarized in Table 5 along with the corresponding octanol/water
partition coefficients. Log of the average K ' values obtained for the
three types of wastewater solids were used for each compound for correlation
purposes. Justification for averaging the log Kp' values was based on
replicate isotherms measured for methylene blue dye on the three kinds/of
solids studied. Log K.' values measured for sorption of dye on separately
collected replicate samples of wastewater solids averaged 4.35, 4.57, and
4.41 for primary solids, mixed-liquor solids, and digested solids,
respectively. All individual replicates were within ±0.1 of the average
values. Thus, it was concluded that when sorption coefficients are based on
TABLE 5. AVERAGE LOG Kp' VALUES USED TO CORRELATE SORPTION ON WASTEWATER
SOLIDS WITH OCTANOL/WATER PARTITION COEFFICIENTS
Compound
Log Kp'
Log KQ
Methyl ene chloride
Chl orof orm
1 , 1 -di chl oroethyl ene
Carbon tetrachloride
Chl orobenzene
Tetrachl oroethyl ene
2,4-dinitro-o-cresol *
Lindane
iteptachlor epoxide
Dieldrin
*
2.06
2.15
2.23
2.65
2.64
3.00
2.77
3.41
4.19
4.40
1.26
1.97
2.13
2.64
2.84
2.85
2.88
3.85
5.40
5.48
*K values were taken from reference (25) or were calculated by the
fragment method from reference (26).
416
-------
organic matter in the sorbent, all three types of solids are equivalent.
Methylene blue dye is an ideal model compound for sorption experiments
because it in non-volatile, non-biodegradable and can be determined
accurately in the complex matrices used in the isotherm procedure.
A simple linear regression of log K ' on log Kow yielded a correlation'
coefficient (r) of 0.963. A plot of logHK ' versus log K is shown in
Figure 6. The following equation was obtained for the relationship between
partitioning on wastewater solids and octanol/water partition coefficient:
log Kp' = 1.57 + 0.57 log Kow (9)
If Kow 1s 9iven in numerical form, the equation can be written as follows:
Q)
O
6.0
5.0
4.0
3.0
2.0
1.0
V = 37.2(KOW)
0.57
(10)
lnt.-1.57
Slope*©.57
r-0.96
1.0 2.0
3.0 4.0
5.0
6.0
Log i
Figure 6. Correlation of partition coefficients
with octanol-water partition coefficients.
417
-------
Octane!/water partition coefficients are generally accepted as accurate
estimates of soil sorption coefficients. However, at least two major
difficulties have been described (27). First, the precision of available
K data is low; and secondly, a variety of quantitative linear models
describing the relationship between Kow and soil partition coefficients have
been proposed, hence the choice of the model to be used for a given system
becomes difficult. The same problems are encountered in correlating
sorption on wastewater solids with Kow.
Methods based on the topology of molecules have been found to predict a
broad range of properties and have found applications beyond simple
predictions of chemical behavior (28). The Randid molecular-connectivity
index has a wider range of application than others yet devised. Significant
correlations have been reported between the Randid index and log K (29).
The index was first described by Randid (30) and has since been applied to
prediction of soil sorption coefficients for polycyclic aromatic hydro-
carbons, alkylbenzenes, chlorobenzenes, chlorinated alkanes and alkenes, and
halogenated phenols (27). Several extensive reviews of both the theory and
calculation of molecular and connectivity indexes have been published (29-
33). The conventional first-order molecular connectivity indexes.are
calculated from the non-hydrogen part of the molecule. Each non-hydrogen
atom is described by its atomic 6 value, which is equal to the number of
adjacent non-hydrogen atoms. This index is then calculated from, the atomic
values by the following equation:
2(6,6,)
-0.5
(ID
where 6, and 6- correspond to the pairs of adjacent non-hydrogen atoms and
summation is over all bonds between non-hydrogen atoms. In the present
study heteroatoms such as oxygen and nitrogen are included in the analysis
by using the appropriate valence assigned to the heteroatom. Thus,
6, = Zv-h, . (12)
where Zv is the number of valence electrons in the site and h, is the number
of hydrogen atoms. Therefore, for oxygen Zv - 6 and for nitrogen Z - 5.
In the case of chlorine, a value of 0.690 has been assigned (30). The basic
assumption is that a multiple bond in a molecule should be represented by
the endpoint valencies derived from the formula for 6, as described earlier.
Double bonds are counted as double edges. To illustrate, the modified
Randid index (lxv) for 1,1-dichloroethylene is calculated as follows:
(1) Determine valence weighted graph (omit hydrogen atoms):
Cl -
i;
418
-------
(2) Calculate valence at node or vertex (vertex valence = number of valence
electrons = number of H atoms):
0.69 4 2
Cl - C - C
Cl
° 0,69
(3) Compute product of vertex valencies for each edge:
2.76 8
Cl i C * C
2.76—1
Cl
(4) Compute each edge term as the reciprocal square root product (e.g.,
0.602
0.354/bond
(5) Compute sum of the edge terms. » ••
Therefore, lxv - 2(0.602) + 2(0.354) = 1.912.
Correlation of the modified Randi£ index with the corrected partition
coefficients for sorption on wastewater solids was attempted using the
average Log Kp' calculated from the Freundlich parameters from Table 5 and
the corresponding values for log lxv presented in Table 6.
A simple linear regression of log K ' on lxv yielded a correlation
coefficient (r) of 0.972. A plot of logPK ' versus lxv is shown in Figure
7. The following equation was obtained-for the relationship between
partitioning on wastewater solids and the modified Randic index:
log Kn
1.79 + 0.29 lxv
(13)
Methylene blue dye, and the two azo dyes used for sorption experiments
were not included in the correlation with the modified Randic index because
they were ionic species. Methods exist to treat polar and ionic compounds
and these will be evaluated in future research work.
419
-------
TABLE 6 MODIFIED RANDIC INDEX USED TO CORRELATE SORPTION ON
WASTEWATER SOLIDS
Methylene chloride
Chloroform
1,1-dichloroethylene
Carbon tetrachloride
Chlorobenzene
Tetrachloroethylene
2,4-dinitro-o-cresol
Lindane
Heptachlor epoxide
Dieldrin
1.70
2.08
1.91
2.41
2.51
2.91
3.46
6.17
8.52
8.98
420
-------
. D.
O)
O
7.0
6.0
5.0
4.0
3.0
2.0
1.0
lnt.= 1.79
Slope=0.29
r-0.97
0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0
- V
Figure 7. Correlation of partition coefficients
with modified Randic indexes.
SUMMARY AND CONCLUSIONS-
Sorption of toxic organics on primary sludge, mixed-liquor solids and
digested sludge have been correlated with the octanol water partition
coefficient and with modified Randi£ indexes. The same correlation is valid
for all types of wastewater solids if results are expressed on the basis of
EC* r«S?Ulc content of the solids as measured by weight loss on ignition at
550-600 C. Octanol/water partition coefficients are useful for estimating
uptake of toxic organics on wastewater solids. However, the modified Randid
indexes offer a more fundamental method for assessing sorption of toxic
organic compounds based on structural features of the molecule.
Correlations based on Randic indexes avoid the problems of low precision in
the reported values for Kow and provide at least a theoretical basis for
predicting sorption of complex mixtures of organic compounds.
The relationships developed provides a basis for estimating the removal
of toxic and hazardous organics from wastewater by the sorption process
The equations presented can be used to calculate removal of toxic organic
compounds from wastewater treatment plants if the compounds are non-volatile
and non-biodegradable. In addition, toxic organic compounds sprbed on
421
-------
wastewater solids can impact treatment and disposal of sludges. Toxic and
hazardous compounds can affect anaerobic digestion processes, land
spreading, incineration, and ocean-dumping of sludges. The correlations
developed provide a basis for assessing the impacts of sorbed orgamcs on
the sludge disposal options.
REFERENCES
1.
4.
Dobbs, R.A., Jelus, M., and Cheng, K.Y. Partitioning of toxic organic
compounds on municipal wastewater. treatment plant solids. In:
Proceedings of International Conference on Innovative Biological
Treatment of Toxic Wastewaters, Consortium for Biological Waste
Treatment Research and Technology. Arlington, Virginia, 1986. p. 585.
Dobbs, R.A., Wang, L., and Govind, R. Sorption of organics on
wastewater solids: Correlation with fundamental properties. Accepted
for publication in Environmental Science and Technology, September
1989.
O'Connor, D.J. and Connoly, J.P. The effects of concentration of
adsorbing solids on the partition coefficient. Water Res.. 14: 1517,
1980.
Voice, T.C., Rice, C.P., ,and Weber, W,J. Effects of solids
concentration on the sorptive partitioning of hydrophobic pollutants in
aquatic system. Environ. Sci. Techno!. 17: 513, 1983.
Voice, T.C. and Weber, W.J. Sorbent concentration effects In'
liquid/solid partitioning. Environ. Scj.. Techno!. 19(9): 789, 1985.
Gschwend, P.M. and Wu, S. On the constancy of sediment-water partition
coefficients of hydrophobic organic pollutants. Environ. SCT.. Techno!.
19(1): 90, 1985.
Blackburn, J.W. and Troxler, W.L. Prediction of the fates of organic
chemicals in a biological treatment process. Environ. Progress 3(3):
163, 1984,
Karickhoff, S.W. Sorption kinetics of hydrophobic pollutants in
natural sediments. In: R.A. Baker (ed.), Contaminants and Sediments.
Vol. II. Ann Arbor Science, Ann Arbor, MI, 1980. p. 193.
9. Karickhoff, S.W. and Morris, K.R. Sorption dynamics of hydrophobic
pollutants in sediment suspensions. Environ. Toxicol. Chem. 4: 459,
1985.
10. Michaels, G.B. and Lewis, D.L. Sorption and toxfcity of azo and
triphenylmethane dyes to aquatic microbial populations. Environ.
Toxicol. Chem. 4: 45, 1985.
11. Smith, M.R., Lequerica, J.L., and Hart, M.R. Inhibition of
methogenesis and carbon metabolism in methanosarcins sp. by cyanide.
J. Bact. 162: 67, 1985.
8.
-------
12,
13.
14.
15.
16.
, 17.
18.
19.
20.
21.
22.
23.
24.
Yang, J., Speece, R.E., Parking, G.F., Gossett, J., and Keener, W. The
response of methane fermentation to cyanide and chloroform. Prog
Water Techno! . 12: 977, 1980. ° - a
Fedorak, P.M., Roberts, D.J., and Hrudey, S.W. The effects of cyanide
c iSo- an°gen1c de9radation of phenolic compounds. Water Res. 20:
1986.
Hamaker, J W. and Thompson, J.M. Adsorption. In: Organic Chemicals
in the Soil Environment. Vol.1. Marcel Dekker, Inc. New York, NY,
* y * C. • p • ^ry ,
Rao, P.S.C. and Davidson, J.M. Environmental impact of non-paint
source pollution. Ann Arbor Science Publishers, Inc. Ann Arbor,
Michigan, 1980. p. 23. • .
Lambert, S.M., Porter, P.E., and Schieferstein, H. Weeds 13: 185,
1965.
Karickhoff, S.W., Brown, D.S., and Scott, T.A. Sorption of hydrophobic
pollutants on natural sediments. Water Res. 13: 241, 1979.
Briggs, G.G. A simple relationship between soil adsorption of organic
chemicals and their octanol/water partition coefficients. In-
Proceedings of the 7th British Insecticide and Fungicide Conference.
city/state, 1973.
Kenga, E.E. Res. Rev. 44: 73, 1972.
Davies, J.E., Barquet, A., Freed, V. H., Hagve, R., Morgade, C.,
Sonneborn, R.E., and Vaclavek, C. Human pesticide poisonings by a fat-
soluble organophosphate insecticide. Arch. Environ. Health 30: 608,
1975.
Lu, P.Y. and Metcalf, R.L. Environmental fate and biodegradation of
benze derivatives as studied in a model aquatic ecosystem. Environ.
Health Persoect. 10: 269, 1975. -
Metcalf, R.L., Kapoor, I. P., Lu, P.Y., Schuth, C.K., and Sherman, P.
Model ecosystem studies of the environmental fate of six organochlorine
pesticides. Environ. Health Persoect. 4: 35, 1973.
Metcalf, R.L., Sanborn, J.R., Lu, P.Y., and Nye, D. Laboratory model
ecosystem studies of the degradation and fate of radiolabeled tri-,
tetra-, and pentachlorobiphenyl compared with DDE. Arch. Environ.
Cont. Toxicol. 3: 151, 1975. - " -
Neely, W.B., Branson, D.R., and Blan, G.E. Partition coefficient to
measure bioconcentration potential of organic chemicals in fish.
Environ. §c_i.. Techno! . 8: 1113, 1974.
423
-------
25. Callahan, M.A., Slimak, M.W., Gabel, N.W., May, I.P., and Fowler, C.F.
Water-related environmental fate of 129 priority pollutants. EPA-
440/4-79-029a and 029b. U.S. Environmental Protection Agency,
Washington, DC, 1979.
26. Hansch, C. and Leo, A. Substituent constants for correlation analysis
in chemistry and biology. John Wiley and Sons, Inc., New York, NY,
1979. p. 18.
27. Sabljic, A. On the prediction of soil sorption coefficients of organic
pollutants from molecular structure: Application of molecular topology
model. Environ. Sci. Techno!. 21: 358, 1987.
28. Rouvray, D.H. Predicting chemistry from topology. Scientific American
225: 40, 1986.
29. Kier, L.B. and Hall, L.H. Molecular connectivity in chemistry and drug
research. Academic, New York, NY, 1976.
30. Randid, M.J. On characterization of molecular branding. J. Am. Chem.
Soc. 97: 6609, 1975.
31. Balaban, A.T., Motoc, I., Bonchev, D., and Mekenyan, 0. Top.. Curr.
Chem. 114: 21, 1983.
32. Sabljic, A. and Trinajstic, N. 'Acta Pharm. Juoosl. 31: 189, 1981.
33. Trinajstic, N. Chemical Graph Theory. CRC Press, Boca Raton, Florida,
1983.
424
-------
MTrnS!nL'S AND REMOVAL OF SELECTED INDICATOR AND PATHOGENIC
MICROORGANISMS DURING CONVENTIONAL ANAEROBIC SLUDGE DIGESTION
by
Vincent P. Olivieri
Lynne Cox and Mohammed Sarai
The Johns Hopkins University
Baltimore, Maryland 21218
and
Jan L. Sykora and Patrick Gavaghan
University of Pittsburgh
Pittsburgh, Pennsylvania
This paper has been reviewed in accordance with
the U.S. Environmental Protection Agency's peer
and administrative review policies and approved
for presentation and publication.
Prepared for Presentation at:
Twelfth United States/Japan Conference
on Sewage Treatment Technology
Cincinnati, Ohio
October 12-13, 1989
425
-------
?1nditcliorsAPP;h°Teraonlyyo 2*10 \ s7e™n wT/efoSn^fov the streptococci
sS"a'were reduced 1.3 to 3.3Orders of magnitude:. The^^ og reduct n ,
fewer significant correlations were found..
426
-------
. . INTRODUCTION
Samples were collected at three wastewater treatment plants in the Baltimore
vicinity to characterize the levels of reduction of selected indicator and
pathogenic microorganisms during anaerobic sludge digestion to provide the
necessary database to prepare and support regulations for the handling of
domestic wastewater sludge including disposal to land. In addition, the study
investigated the relationship between solids, microbial indicators and pathogens
in wastewater and sludge during conventional sludge treatment processes to
evaluate effective monitoring parameters that reflect the pathogen reductions.
METHODS
SAMPLING
Samples were collected after thorough flushing of the available sample
lines to insure a representative sample. Sterile 2 to 5 gal tubs with seal able
tight fitting covers were used for ease of handling and wide mouth to accommodate
the thick samples. After collection, samples were transported immediately to
the Water and Wastewater Microbiology Laboratories at The Johns Hopkins
University in Baltimore and were processed immediately upon arrival. Typical
travel times to the laboratory after sample collection were 20, 40 and 60 minutes
for Back River, Sod Run and Blue Plains, respectively. The routine collection
of samples was begun on August 26, 1987 and continued through September 1988.
In addition, samples for the determination of human parasites were cooled and
shipped by overnight express to Dr. Jan Sykpra at the University of Pittsburgh.
Samples for human enteric viruses were concentrated and the concentrates shipped
frozen to Dr. Robert Safferman at the USEPA laboratory in Cincinnati, Ohio.
Each sample was mixed with a "lightning mixer" for 5 minutes at about 200
revolutions per minute prior to dispensing. Samples were dispensed with
continuous mixing for analysis. Microbiological assays that required subsequent
dilutions were mixed with an equal volume of 1 % peptone and blended at high
speed in a Waring blender for 1.0 minute. Immediately after blending, the
samples were diluted and plated.
CULTURE AND RECOVERY METHODS
A major effort during the preliminary phase was the comparative evaluation
of culture methods for the selected indicator and pathogenic microorganisms.
The intent of this portion of the study was not to develop new methods but to
evaluate existing procedures and techniques to yield the most precise and
accurate determination of the levels of microorganisms. Table 1 shows the
selected microorganisms assayed and the method employed.
CHEMICAL AND PHYSICAL MEASUREMENTS
Total solids (TS) was performed at 103 C according, to Standard Methods
(APHA 1985). Total fixed solids (TFS) were determined at 550 C according to
Standard Methods fAPHA. 1985). Total volatile solids (TVS) were determined by
subtraction (total solids-fixed solids). Suspended solids (SS) was performed
at 103 C with Gooch crucibles containing Whatman 934-AH according to Standard
427
-------
Methods (APHA, 1985). Fixed suspended solids (FSS) were determined at 550 C
according to Standard Methods (APHA, 1985). Volatile suspended solids (VSS) were
determined by subtraction (Suspended solids^fixed suspended solids). The pH
was determined electrometrically according to Standard Methods (APHA, 1985).
RESULTS
LEVELS OF MICROORGANISMS
The logarithmfc mean level per• 100 ml for each of the microorganisms
assayed is shown in Table 2a,b and c. The density of microorganisms for the raw
sewage at the three treatment plants was remarkably similar. Higher logarithmic
mean levels per 100 ml were found for the raw sewage and undigested sludge, while
the plant effluents and digested sludges were lower. Similar trends were
observed for the data presented per unit solids except that the logarithmic mean
levels per unit solids for the plant effluents were not as reduced as that
observed for unit volume. The most numerous indicators were the coliforms with
total coliforms higher than fecal which were higher than E. coli. The fecal
streptococci were about 2 orders of magnitude lower than the total coliform with
fecal streptococci consistently higher than the enterococci. The logarithmic mean
levels of Clostridium and Mvcobacteria were about 3 and 4 orders of magnitude
lower than the coliforms. The slow growing or late Mvcobacteria were
consistently higher than the fast growing Mvcobacteria. The logarithmic mean
levels of coliphage Were generally 2 to 3 orders of magnitude lower than the
levels of coliforms. The logarithmic mean levels observed for phage counted on
£. coli C were similar to the levels observed for male specific phage counted
with Havelaar's strain.
The pathogenic microorganisms were all observed at relatively low levels.
In many of the samples, particularly the digested sludges, and effluents no
pathogens were recovered. The logarithmic mean levels of Salmonella were about
5 orders of magnitude lower than the coliforms but about 2 orders of magnitude
higher than the human enterovirus virus and parasites. Salmonella was
consistently isolated from raw wastewater and sludges and less frequently found
in digested sludges.
The level of human enteric viruses was very low in all samples. The levels
of virus in the raw wastewater was particularly low and suggests that the level
of virus was low in the populations served by the three plants. Only a few
viruses were observed per 100 ml of the" sludge samples.
Parasites were recovered from most of the samples collected from the early
part of the treatment process. The average levels observed in raw sewage were
less than 10/100 ml, but in primary, trickling filter sludge, and activated
sludge, the levels were 29-170,45-150, 14-66 /100 ml, respectively. After
anaerobic digestion, the frequency of the isolation of parasites decreased
slightly. When found, the average level of parasites was 14-60/100 ml. The cyst
concentrations in raw sewage ranged between 0 and 30 cysts per 100 ml, whereas
the concentrations in final effluents ranged from 0 to 1 cyst per 100 ml.
Cysts of Giardia were the most frequently recovered parasite. Helminths
were infrequently found and other human parasites were at best occasionally
observed in the samples of wastewater and sludges.
CORRELATION OF LEVELS OF MICROORGANISMS
The correlation coefficients for the level of selected indicator
428
-------
microorganisms and the level of selected human pathogens for all samples
collected at the three wastewater treatment plants are shown in Table 3. The
levels of microorganisms are presented in 3 different formats; #/100 ml, #/gTS
and #/g VSS. The second number presented below each correlation coefficient was
the number of sample pairs used to calculate the correlation coefficient. The
significance was evaluated by determining the probability that there was no
correlation between the levels of indicators and pathogens, the null hypothesis.
I™ significance was indicated at the 95 % level by an asterisk (*) and at the
99% level by a double asterisk {**).
Significant correlations were observed at the 99% level for the #/ 100 ml
of total coliforms, fecal coliforms, E. coll, fecal streptococci, phage C and
phage H and the #/100 ml Salmonella. Similar significant correlations were found
for human enteric viruses, except that the phage H correlation was not
significant. Only the correlations between the #/100 ml of total coliforms,
fecal streptococci and phage C and the #/100 ml of total parasites were
significant. The correlation coefficients in each case were very low and were
between 0 17 and 0.34 for Salmonella, 0.14 and 0.46 for human enteric virus and
0.18 and 0.19 for total,parasites. No significant correlations were found
between the #/100 ml Of enterococci, Clostridium and Mvcobacteria. It should
be noted that the sample number for these calculations was between 230 and 427
Calculating the correlation coefficient with the microbial densities per gram
of various solids categories yielded poorer correlations". Significant
correlations at the 99% level were observed only for the levels of total
thln#/[So'ml Simnnlif°rmSp £* ^4> fecal streptococci, phage C and phage H and
the #/ioo ml Salmonella. Few significant correlations were found for virus and
E?rSS^eS> • T-r fT1^0311* correlations were found for correlations calculated
with the microbial levels per various solids criteria.
LOG REDUCTION OF MICROORGANISMS THROUGH ANAEROBIC DIGESTION
arP J™n Tna\]£?oldUCtT?nS for'V6 #/1°° ml of the selected microorganisms
are shown ini Table 4. The mean log reduction was calculated from the log
reductions observed for each set of samples collected at each of the three
wastewater treatment plants as shown below. The relative log digester input
MEAN LOG REDUCTION = .S^.^OG_DIGESTER_INPUT:LOG_DIGESTER_OUTPUT)
WHERE. NUMBER OF SAMPLES
LOG DIGESTER INPUT-LOG DIGESTER OUTPUT
FOR PLANT A=
lA3 x °A3>+<#/ioomiA4 x QA4)-i.oG <#/ioo mi^ x OA5>
FOR
FOR
AND'
Q=FLOW IN HGD
Aa,A3,A4,A5,B2fB3,B4,B5,C2,C3, AND C5=SLUDGE SAMPLES AT PLANT A, B. AND C.
th«Mi J«HSttehd bel°-W the Tan log reduct1°n are "the standard deviation for
this value and the maximum and minimum reductions observed during the course of
429
-------
the sampling period. It should be noted that considerable variability was
observed^ fof the mean log reduction values with almost 2 orders of magnitude
range for the mean log reduction of the indicator bacteria, {he ranges for the
log reduction values for the coliform indicators were 1.27 to 1.45, 2.82 to Z.9/
and 1.06 to 1.25 for "plants A, B and C, respectively. Lower log reduction values
were found for the streptococci and phage indicators. Relatively little
reduction of Clostridium and Mvcobacterium were found with no reductions observed
3 P 3Human pathogens were reduced by anaerobic digestion. The mean log
reduction values for Salmonella were 1.92, 3.33 and 1.38 for plants A, B and C.
The mean log reduction for human enteric virus was lower with 1.12, 1.37 and0.4b
for plants A, B and C. Total parasites mean log reduction was similar: 0.90,
1 27 and 0.55 for plants A, B and C. In each case plant B had the highest
reduction in the level of indicator and pathogenic microorganisms.
CORRELATION OF LOG REDUCTION
The correlation coefficients for the linear regression for the log
reductions of selected indicator and pathogenic microorganisms are shown in Table
5 for all samples. Significant correlations at the 99% level were found between
the log reduction of the indicator microorganisms (except for the slow growing
Hvcobacteria which was significant at the 95% level) and Salmonella. The
correlation coefficients ranged from 0.27 for the slow growing MYcobacteria
which was significant at the 95% level to 0.65 for E.colj. which was significant
at the 99% level. The correlation coefficients for log reduction of the
coliforms and streptococci indicators with log reduction of Salmonella were 0.58
to 0.65 and 0.47 to 0.62, respectively. Similar but lower correlations were
observed for all samples with log reductions of human enteric virus. Lesser
correlations were found for total parasites.. However, when the log reductions
for each treatment plant were considered separately, only the log reduction for
E.coli at plant B was significant at the 95% level for the log reduction of
Salmonella.
RELATIVE MICROBIAL BALANCE
The relative microbial mass balance through the three treatment plants
is shown in table 6. The relative mean microbial mass for each selected
indicator and pathogen was calculated from the product of the flow in MGD for
the waste stream and the level of microorganism, #/WQ ml. The data was reported
as the relative microbial mass since no correction was made for the volume term.
The units are # million gallons/100 ml day. The overwhelming majority of the
microbial mass leaves the wastewater treatment plant after conventional secondary
treatment in the liquid effluent. Without terminal disinfection, only 1 to 2
orders of magnitude reduction are accomplished by secondary treatment. The log
reduction in the microbial mass from the raw wastewater after anaerobic sludge
digestion was 2 to 3 orders of magnitude. The microbial mass leaving the plant
by the solids route after anaerobic sludge digestion was about 10 fold lower than
by the liquid route without disinfection.
LOG PROBABILITY . ..
An alternative presentation for the levels of microorganisms in the raw
sewage (open points) and the digested sludge (filled points) for plant A
(circles), B (triangles) and C (squares) is shown in figures 1, 2, 3 and 4 for
fecal coliform, fecal streptococci, coliphage C and Salmonella, respectively .
For each indicator and pathogenic microorganism assayed, the log probability
430
-------
plot approximates a straight line and indicates that the distribution for the
levels of microorgantsms follows a log-normal distribution: As seen in figure
1, 2, 3 and 4, the levels of each of the microorganism in raw wastewater at plant
A, B, and C was similar. The differences found in the raw sewage was small.
However, the levels of microorganisms in the digested sludge for each treatment
plant varied. A visual inspection of each figure shows the approximate log
,#/gTS reduction from raw sewage thru anaerobic sludge digestion (open vs closed
points) for each of the indicator and pathogenic microorganisms assayed. This
visual log reduction differs from the mean log reduction calculation shown in
table 4 and 6. Nevertheless, a 2 to 3 log reduction from raw sewage through
sludge digestion is shownt for the coliform (figure 1) and a 1.5 to 2 log
reduction for fecal streptococci (figure 2). The levels of coliphage (figure
3) also yielded a straight line for the log probability plot and follows a log-
normal distribution in raw sewage and digested sludge. About 2 logs reduction
were observed for phage C (figure 3). Similar log probability plots are shown
for Salmonella in figure 4. Like the indicator bacteria, the log of Salmonella
levels also approxomated a straight line on the log probability plot and appeared
to follow a log normal distribution. The levels of Salmonella for the raw sewage
at each plant were a little more variable than the indicators and reflected the
poorer precision of the multiple tube culture method and the most probable number
calculation used for enumeration. About 2 to 3 logs reduction were observed
from raw sewage through anaerobic digestion.
CONCLUSIONS
Considerable variation was observed for samples collected at the same
sample station at a given treatment plant. The levels of selected microorganisms
and human pathogens in the raw wastewater entering the three treatment plants
were remarkably similar. The variability between treatment plants appeared to
be no more than the variability between type of samples. The levels of the
selected indicators and pathogens appeared to follow the order of densities for
the samples of wastewater and wastewater sludge sampled in this study: total
coliform > fecal coliform > E.coli > fecal streptococci > enterococci > phage
C - phage H > Clostridium > Mvcobacterium > Salmonella > enteric viruses >
parasites.
•The levels of human pathogenic bacteria, viruses and parasites were low
in all the samples compared to the levels of indicators. Salmonella SDP. were
consistently found in raw wastewater, raw sludges, treated sludges and wastewater
effluents. Parasites were found in most of the samples of raw wastewater, raw
sludges, and digested sludges but rarely observed in the wastewater effluents.
Enteric viruses were infrequently recovered from raw wastewater but found in many
of the raw sludge samples and few of the digested sludge samples. Viruses were
not recovered from the wastewater effluent. The low frequency of recovery in
the liquid flows at the treatment plants was probably due to a combination of
low virus densities and low volume (100 ml) assayed.
The anaerobic digestion process yielded reductions in the densities of
selected indicator and pathogenic microorganisms at each of the three treatment
plants. The mean log reduction varied for the three treatment plants for the
same microorganism, but followed the same relative order. Plant B, Sod Run, had
the highest mean log reductions, followed by plant A, Back River, and plant C,
Blue Plains. One to three log reductions were observed for the coliform
indicators thru the anaerobic digestion process, while only 0.2 to 2 logs
431
-------
reduction for the streptococci were found at the three plants. Salmonella were
reduced 1.3 to 3.3 orders of magnitude. While the level's of human virus and
parasites were low and the log reductions difficult to determine, the log
reduction observed were less than those found for the bacteria. Only 0.4 to
1.4 and 0.5 to 1.2 orders of magnitude reduction were observed for virus and
parasites, respectively. Correlations significant at the 95% and 99% level were
found between the log reduction of indicator and pathogenic microorganisms.*
Correlations were generally higher (as high as 0.65) than that observed for the
levels of microorganisms. Again no particular indicator consistently yielded
higher correlations.
ACKNOWLEDGEMENTS
The authors wish to acknowledge the cooperation of Nick francos and staff at the
sludge facility at the Back River Wastewater Treatment Plant in Baltimore,
Maryland, Wayne Ludwig and staff at the Sod Run Wastewater Treatment Plant in
Bel Air, Maryland, and Russel Thomas and Sam Alamine at the Blue Plains
Wastewater Treatment Plant in Washington,DC. This work was supported ,by the
USEPA Project #: CR B135 98.
REFERENCES
Adams, M. 1959. Bacteriophages. Interscience Publishers, New York, New York.
Berg, G. et a]_. 1975. Methods for Recovering Viruses from Sludges (and other
solids). In: U.S. EPA Manual of Methods for Virology, EPA - 600/4-84-013, U.S.
EPA, Cincinnati, Ohio.
Bisson, 0. and V. Cabelli. 1979. Membrane Filter Enumeration Method for
Clostridiura perfrinoens. Appl. Environ. Microbiol.
37:55-66.
Dufour, A.P. 1984. Bacterial Indicators of Recreational Water Quality. Can J
of Public Health 75:49-56.
Engelbrecht, R.N. and C. Haas. 1977. Acid-Fast' Bacteria and Yeast as
Disinfection Indicators: Enumeration Methodology. In: Proc. AWWA Water
Quality Technology Conference, Kansas City, Kansas.
Fox, 0., P.'Fitzgerald, C. Lue-Hing. 1981. Sewage Organisms: A Color Atlas.
The Metropolitan Sanitary District of Greater Chicago, Chicago, Illinois.
Geldreich, E.E. 1981. Membrane Filter Techniques for Total Coliform and Fecal
Coliform Populations in Water. In: Membrane Filtration: Applications,
techniques and problems, B.J. Dutka ed., Marcel Dekker, Inc., New York.
Havelaar, A.H., W.M. Hogenboom and R. Pot. 1984. F. specific RNA bacteriophages
in sewage: methodology and occurrence. Water Sci. Technol. 17:645-655.
Skeat, W.O. (1969) Manual of British Water Engineering. 4th ed. Institution of
Water Engineers, W. Heffer and Sons, Ltd., Cambridge, England.
432
-------
Standard Methods for the Examination of Water and Wastewater. 16th edition.
1985. American Public Health Association, New York, New York.
U.S. EPA. 1978. Microbiological Methods for Monitoring the Environment, Water,
and Waste. Cincinnati, Ohio.
433
-------
TABLE 1. METHODS EMPLOYED FOR THE RECOVERY OF MICROORGANISMS.
Microorganism
Total coliform
Fecal Coliform
E. sail
Fecal Streptococci
Enterococci
Clostridiun
Hvcobacteria
Salmonella
Phoge
Enteric Virus
Protozoan cysts
Helminth Ova
Methods
ra-Endo Plate Count
m-FC Plate Count
m-Tech Plate Count
Plate Count
Membrane Filter
Membrane Filter
Membrane Filter
Multiple Tube Dilution
Agar Overlay, E_. coli c
S.typhimurium UG49 *
Elution and organic
flocculation
Sedimentation with
sucrose floatation*
Sedimentation with
zinc sulfate floatation
References
Standard Methods (APHA 1985}
Geldreich <1981)
Standard Methods (APHA 1985)
Geldreich (1981)
Dufour et al.(1984)
Standard Methods (APHA, 1985)
Dufour et aI (1984)
Bisson et al (1979)
Engelbrecht & Haas (1977)
Standard Methods (APHA, 1985)
Adams (1959)
Havelaar (1984)
USEPA Manual of Methods
for Virology (1984)
Fox et al. (1981)
Fox et al. (1981)
TABLE 2a. LOGARITHMIC MEAN LEVELS PER 100 ML OF BACTERIAL INDICATORS AND
PATHOGENS, HUMAN ENTERIC VIRUS AND HUMAN PARASITES FOR SAMPLES' COLLECTED AT
PLANT A
TC FC EC FS ENT CP MY LTMY SAL PHAGE C HEV PHAGE H TP
RAW
SEWAGE
LM
STD
H
PRIMARY LH
SLUDGE
TFS
AS
OS
HE
STD
H
LM
STD
H
LH
STD
N
LH
STD
H
LH
STD
H
7.99
0.04
24
8.20
0.15
24
8.62
0.15
21
8.59
0.11
23
8.15
0.08
24
7.00
0.15
24
7.34
0.08
24'
7.34
0.15
24
7.61
0.18
21
7.73
0.08
23
7.20
0.11
24
6.38
0.18
24
7.08
0.15
24
7.00
0.18
24
7.34
0.18
21
7.62
0.11
23
6.99
0.11
24
6.15
0.20
24
5.74 5.58 3.79 1.86
0.08 0.08 0.15 0.11
24 24 23 23
4.92 4.52 4.45 3.36
0.15 0.20 0.20 0.11
24 24 23 23
6.18 5.97 4.79 3.18
0.20 0.15 0.28 0.18
21 21 20 20
6.40 6.32 5.04 3.61
0.15 0.08 0.23 0.18
23 23 22 22
5.72 5.64 5.32 4.60
0.11 0.08 0.23 0.15
24 24 23 23
4.82 4.52 3.72 1.57
0.15 0.15 0.11 0.18
24 24 23 23
2.20
0.18
18
3.58
0.20
18
3.75
0.23
16
4.08
0.23
17
4.95
0.20
18
1.96
0.18
17
2.11
0.11
23
1.99
0.18
23
2.66
0.08
21
2.64
0.20
24
1.68
0.15
24
1.26
0.18
24
5.43
0.08
23
4.62
0.15
23
5.11
0.08
20
5.57
0.08
22
5.11
0.15
23
4.64
0.11
23
0.18
0.08
23
0.79
0.15
23
OJ34
0.11
20
1.04
0.15
22
0.40
0.11
23
0.18
0.08
23
5.78
0.23
13
4.54
0.23
13
5.36
0.26
10
5.62
0.23
12
5.26
0.18
13
4.63
0.32
13
0.48
0.08
23
1.32
0.11
22
0.82
0.18
21
0.71
0.18
23
1.34
0.18
24
-0.03
0.04
24
NOTE: LH = LOGARITHMIC MEAN. STD = LOGARITHMIC STANDARD DEVIATION, N
TFS - TRICKLING FILTER SLUDGE, AS - ACTIVATED SLUDGE,
DS - DIGESTED SLUDGE, ME - MIXED EFFLUENT
NUMBER OF OBSERVATIONS.
434
-------
TABLE 2b. LOGARITHMIC MEAN LEVELS PER 100 ML OF BACTERIfll TNnmmws inn
PATHOGENS, HUMAN ENTERIC VIRUS AND HUMAN WWBITB K»^^"cOLLECTH) AT.
TC FC EC FS ENT CP
MY LTMY SAL PHAGE C HEV PHAGE H TP
RAW LM
SEWAGE STD
N
PRIMARY LM
SLUDGE STD
N
TFS
AS
DS
DS
(STORED)
ME
SEPTAGE
'ABLE 2
LM
STD
N
LM
STD
N
LM
STD
N
LM
STD
N
LM
STD
N
LM
STD
N
8.04 7.00 6.93 5.90 5.53 3.83 2.08 2.40 1 60
0.08 0.11 0.11 0.08 0.04 0.15 0.15 0 20
-------
TC
FC
EC
FS
correlation coefficient
. number
EHT CP MY LTHY
PHAGE C PHAGE H
JT/100 HL
*/9TS
*/gVSS
SAL
HEV
TP
SAL
HEV
TP
SAL
HEV
TP
0.32 **
427
0.28 •*
412
0.20 **
423
0.10 *
427
0.01
412
0.04
423
0.09
425
0.03
410
0.04
422
0.28 **
427
0.15 **
412
0.09
423
0.19 **
427
0.06
412
0.06
423
0.27 •*
425
0.07
410
0.05
422
0.17 **
427
0.18 **
412
0.05
423
0.16 **
427
0.03
412
0.05
423
0.06
425
0.53 **
410
0.01
422
0.34
427
0.14
412
0.19
423
0.24
-427
0.08
412
0.21
423
0.27
425
0.20
410
0.24
422
** 0.02
426
** 0.00
411
** 0.00
422
**-0.02
426
-0.01
411
**-0.02
422
** 0.00
424
**-0.01
409
**-0.02
421
-0.03
409
-0.01
412
0.08
405
0.01
409
0.00
412
0.04
405
0.07
407
0.00
410
-0.03
404
-0.06
409
-0.03
412
-0.01
405
-0.05
409
-0.02
412
-0.04
405
-0.02
407
-0.01
410
-0.03
404
0.06
319
-0.03
320
-0.01
313
•0.08
319
-0.03
320
•0.05
313
•0.04
317
•0.03
318
-0.07
312
0.33 **
410
0.46 **
395
0.18 **
406
0.38 **
410
0.15
395
0.04
406
0.36 **
408
0.15 **
393
0.03
405
0.26 **
234 •"
0.07
235
-0.04
230
0.40 **
234
0.02
235
0.08
230
0.29 **
233
0.00
234
0.12
230
TABLE 4 MEAN LOG REDUCTION OF THE 1/100 ML OF SELECTED MICROORGANISMS FROM RAW SEWAGE THRU
CONVENTIONALANAEROBIC DIGESTION AT THREE WASTEWATER TREATMENT PLANTS
EHT
CP
HY
LHY
SAL
PHC
PHH'
HEV
TP
PIAUT A
HEW) LOG
REDUCTION
SID
MX
HIM
M
HAHT_I
XEAH LOG
REDUCTION
STD
MAX
HIM
H
PtAHT C
KEAM LOG
REDUCTION
STD
HAX
HIM
H
1.25
0.41
2.06
•0.13
24.00
2.97
1.12
5.69
1.15
24.00
1.25
0.73
2.85
•0.07
23.00
1.39
0.64
3.08
-0.01
24.00
2.82
1.15
5.34
1.22
24.00
1.06
0.55
2.36
0.04
23.00
1.35
0.57
2.48
0.33
24.00
2.91
1.38
7.19
0.44
• 24.00
1.18
0.79
3.03
-0.60
23.00
1.17
0.60
3.02
0.04
24.00
2.04
0.88
4.60
1.01
24.00
0.43
1.35
6.35
•0.94
23.00
1.10
0.40
1.88
0.43
24.00
2.13
0.88
4.49
0.67
24.00
0.26
0.58
1.53
•1.16
23.00
0.60
1.30
6.48
-0.37
24.00
0.81
0.65
1.95
-0.21
23.00
-0.34
0.74
1.02
-1.31
22.00
0.12
1.01
3.94
-1.46
24.00
0.20
0.46
1.26
-0.48
23.00
•0^54
0.64
0.92
•2.16
22.00
0.93
1.94
5.70
-1.43
23.00
0.65
0.57
1.34
-0.65
18.00
-0.30
0.83
1.57
-1.85
17.00
1.87
0.87
4.04
0.51
24.00
3.33
0.85
5.25
2.10
24.00
1.38
1.04
2.88
-1.32
23.00
1.16
1.14
6.13
0.07
24.00
1.81
0.79
4.42
0.77
23.00
0.40
0.40
1.27
1 -0.70
22.00
1.88
2.08
6.67
0.21
16.00
3.03
0.70
4.18
1.91
13.00
1.34
0.86
2.90
0.42
13.00
1.14
0.32
1.90
0.31
24.00
1.37
0.40
2.19
0.74
23.00
0.45
0.39
1.22
•0.23
22.00
1.01
0.85
2.85
-0.10
24.00
1.22
1.07
3.43
•1.31
24.00
0.55
0.80
2.03
•0.51
22.00
436
-------
7n5Lnr™,^ORRELATION COEFFICIENTS FOR THE LOG REDUCTION OF INDICATOR MICROORGANISMS AND THE
°F HUMAN PATHOGENS FOR ALL SAMPLES COLLECTED AT' THREE WASTEWATER1 TREATMENT
TCR
FCR
ECR
FSR
CORRELATION COEFFICIENT
PROBABILITY THAT H=0
NUMBER
ENTR CPR
HYR
LTHYR
PHAGE CR PHAGE HR
SALR
HEVR
TPR
0.62816
0.0001 **
69
0.46712
0.0001 **
66
0.29146
0.0159 *
68
0.58489
0.0001
69
0.48219
0. 0001
66
0.23363
0.0552
,68
0.64799
** 0.0001 **
69
0.48995
** 0.0001 **
66
0.33649
* 0.005 **
68
0.47003
0.0001 **
69
0.451
0.0001 **
66
0.26595
0.0284 *
68
0.6199
0.0001 **
69
0.64232
0.0001 **
66
0.30236
0.0122 *-
68
0.37327
- 0.002
66
0.49419
0.0001
66
0.02307
0.8553
.65
0.32343
** 0.0081 **
66
0.33542
** 0.0059 **
66
0.1196
0.3426
65
0.27615
0.0498 *
51
0.33927-
0.0149 *
51
0.13059
0.366
50
0.55113
0.0001 **
66
0. 61'254
0.0001 **
63
•0.2197
0.0787
65
0.56131
0.0002 **
38
0.48638
0.002 **
38'
0.10912
0.5143
38
PLANTS6A BRANDTCV* MICR°BIAL BALANCE FROM RAW "ASTEWATER FOR THE SELECTED MICROORGANISMS AT
TC
FC
EC
FS
ENT
CP
MY
LHY
SAL
PHC
PHH
HEV
CALCULATION BASED ON THE HICROBIAL LEVELS AS #/ 100 ML
TP
PLANT A
RAW 10.29
DIGESTER IN 9.15
DIGESTER OUT 7.90
EFFLUENT 9.27
DIGESTED 2.40
EFFLUENT 1.0Z
PLANT B
RELATIVE MEAN LOG HICROBIAL MASS
9.62
8.33
6.95
8.64
2.68
0.98
9.38
8.08
6.73
8.40
2.65
0.98
8.05
6.64
5.47
7.08
2.58
0.98
7.87
6.48
5.38
6.77
MEAN LOG
2.49
1.10
6.15
5.46
5.06
5.97
REDUCTION
1.08
0.18
RELATIVE HEAM LOG
RAW SEWAGE 8.8S
DIGESTER IN 8.46
DIGESTER, OUT 5.49
EFFLUENT 6.97
DIGESTED 3.40
EFFLUENT 1.92
PLANT C
RAW SEWAGE 10.37
DIGESTER IN 9.29
DIGESTER OUT 7.75
EFFLUENT 7.51
DIGESTED 2.62
EFFLUENT 2.87
7.83
7.33
4.51
5.84
3.32
1.99
9t39
8.33
6.98
6.78
2.40
2.60
7.77
7.16
4.25
5.92
3.52
1.84
9.50
8.26
6.83
6.54
2.66
2.95
6.73
6.05
4.01
4.93
2.72
1.81
8.36
6.59
5.89
5.68
2.47
2.68
6.36,
5.74
3.61
4.60
MEAN LOG
2.76
1.77
4.66
4.43
3.61
4.07
REDUCTION
t.05
0.59
RELATIVE MEAN LOG
8.07 6.25
6.48 5.63 ,
5.96 5.94
5.78 5.77
MEAN LOG
2.11
2.29
REDUCTION
0.31
0.48
4.22 4.43
4.27 4.61
4.34 4.46
3.82 4.24
FROM RAW SEWAGE
•0.12 -0.02
0.39 0.20
MICROBIAL M/fsS
H.92 3.26
3.21 3.46
3.01 2.81
2.60 2.61
FROM RAU SEWAGE
-0.09 0.44
0.32 0.65
MICROBIAL MASS
4.31 5.09
4.29 4.90
4.80 5.20
3.97, 4.23
FROM RAW SEWAGE
-0.49 -0.10
0.34 0.86
4.37
3.30
1.42
3.52
2.95
0.85
2.44
2.10
-1.23
1.34
3.67
1.10
4.42
3.52
2.10
2.80
2.32
1.62
7.69
5.82
4.86
6.95
2.83
0.75
5.82
5.09
3.28
5.59
2.55
0.24
7.56
5.69
5.07
5.35
2.49
2.21
8.02
5.98
5.04
6.88
2.98
1.13
6.16
5.31
2.29
5.26
3.87
0.90
8.04
6.03
4.69
5.18
3.35
2.85
3.30
1.85
0.75
3.22
2.55
0.07
1.80
0.95
-0.42
1.81
2.23
-0.00
3.51
2.04
1.59
3.48
1.92
0.03
2.75
2.08
1.08
2.22
1.68
0.53
1.40
1.11
-0.11
0.81
1.51
0.59
3.00
2.10 •
1.51
2.49
1.48
0.50
437
-------
PROBABILITY
Figure 1. Log # fecal coliform/gts vs probability for
samples raw sewage (open points) and digested sludge
(closed points) at plant A (circles, plant B, (triangles)
and C, (squares) from August 1987 thru September 1988.
PROBABILITY
Figure 2. Log # fecal streptococci/gts vs probability
for samples of raw sewage (open points) and digested
sludge (close points) at plant A (circles),,plant B,
(triangles) and C, (squares) from August 1987 thru
September 1988.
438
-------
-2
PROBABILITY
Figure 3. Log # coliphage C/gts vs probability for
samples of .raw sewage (open points) and digested sludge
(closed points) at plant A (circles), plant B, (triangles)
and C, (squares) from August 1987 thru September 1988.
ie
PROBABILITY
Figure 4. Log # Salmonella/gts vs probability for
samples of raw sewage (open points) and digested
sludge (closed, points) at plant A (circles), plant B,
(triangles) and C, (squares) from August 1987 thru
September 1988.
439
-------
-------
IN-VESSEL COMPOSTING OF MUNICIPAL
WASTEWATER SLUDGE
by
Donald Brown, USEPA, Risk Reduction Engineering Laboratory
John Johnston, California State Univerisity-Fresno
, Leslie Beyer, Eastern Research Group
This paper has been reviewed in accordance with
the U.S. Environmental Protection Agency's peer
and administrative review policies and approved
for presentation and publication.
Prepared for Presentation at:
Twelfth United States/Japan Conference
on Sewage Treatment Technology
Cincinnati, Ohio
October 12-13, 1989
441
-------
INTRODUCTION
*
Since the late 1970s, composting has become a more frequently used
method for stabilizing and processing municipal sewage sludge in the United
States The interest in this technology has been extremely rapid, from less
than 10 facilities operating in 1975 to nearly 200 under design or in
operation in 1989. Today much of the compost produced is being used as for
soil improvement in accordance with the U.S. Environmental Protection Agency s
(EPA's) policy that encourages the beneficial use of sewage sludge (49 FR
24358, June 12, 1984).
More recently, because of odor, labor, and materials-handling problems,
a greater number of composting systems are being designed and built which
contain the materials within a vessel. Other reasons for choosing an in-
vesseltf system have been space limitations and adverse climates. In-vessel
systems also differ from other composting systems because they use mechanical
mixers to blend amendments with the sludge prior to composting, and they
extensively use conveyors for materials handling. The evolution of these
in-vessel*systems has been very rapid, and municipalities continue to
encounter problems in dealing with odors, removing moisture, and handling.
materials.
To learn from the experiences-of these mostly first-generation in-vessel
sludge composting systems and their operators, EPA decided in early 1988 to
gather information from 8 of the 13 full-scale facilities operating in the
United States. f '
This paper presents a brief discussion of the findings of this study.
First, a description of the facilities surveyed and the reactor systems used
will be presented. Then the major problem areas discovered by this survey -
materials handling, moisture removal, aeration of curing/storage piles, and
odor control - will be discussed. Finally, capital and operation and
maintenance costs will be presented.
DESCRIPTION OF FACILITIES
The location of the facilities and the type of composting system used at
each facility are shown in Figure 1. The eight facilities studied included
all six types of in-vessel systems operating in early 1988. Akron, Cape May
County, Newberg, Portland, Pittsburgh, and Schenectady were the first
facilities of their kind to be built in this country. Only Cape May County
and Portland were operating prior to 1986.
A survey team conducted a 1.5 to 2-day long site visit at each facility.
The survey team interviewed plant supervisors, operators, and others familiar
with each facility's procurement, construction, and operations. Contract
documents and operating records were also examined when available. The
information collected represents a "snapshot" of each facility taken at the
442
-------
time of the site visit. The data presented are accurate only for the time of
the site visit and may have subsequently changed.
Figure 2 gives a summary of the configuration of each of the facilities.
Composting is often defined as a two stage process, a high-rate stage followed
by a curing stage. The first stage is characterized by high oxygen uptake
rates, high temperatures, rapid degradation of biodegradable volatile solids,
and high potential for odor production. The second stage is characterized by
lower temperatures, reduced oxygen uptake rates, and a lower, but significant,
potential for odor production. No precise definition or distinction exists
between these two stages, however. Moreover, although the first stage of
composting is performed in reactors at all in-vessel facilities, the second
stage can be performed in a reactor, an exterior pile, or both. Because of
the lack of a clear definition, the term "first-step reactor" will be used to
denote the first vessel that the compost enters. "Second-step reactor" will
be used to denote the second vessel the compost enters in series. At several
plants, there is no second reactor. The reactors will be discussed in more
detail in a following section.
All of the facilites haVe exterior piles for curing or storage. Because
no consistent measure of compost stability was used at the facilities, it was "
not possible to define whether material in outside piles was "curing" or just
being "stored." Therefore, the term "exterior curing/storage piles" refers to
any compost piles located outside the reactors, both aerated and unaerated.
AKRON, OHIO
Table 1 summarizes pertinent information for the Akron, Ohio facility,
which uses four Paygro reactors. Each'of the 'reactors is 6 meters wide, 3
meters deep, and 223 meters long. Akron, at 66 dry metric tons per day, had
the largest design, capacity of the facilities surveyed. Akron was operating
at reduced capacity because of complaints about odors by neighbors.
Figure 3 is a schematic of a Paygro reactor. Compost
in the reactors by a series of conveyors. The operation of
similar to an extended aerated static pile facility in that
reactor bin are treated as individual batches. The compost
periodically mixed in place as desired by a machine called
The extractoveyor also removes the compost from the reactor
to the bottom of the reactor through a plenum covered by a
is exhausted from the top of the compost into the building
reactors, and the building air is exhausted by roof fans.
CAPE MAY COUNTY, NEW JERSEY
material is placed
the facility is
sections of each
is then
an "extractoveyor".
Air is supplied
steel grate. Air
which -encloses the
Table 2 summarizes pertinent information for the Cape May County, New
Jersey facility, which uses two Purac reactors. This facility was originally
designed with one first-step, and one second-step reactor. However/to
increase the capacity of the facility, both reactors are now used as the
first-step, and aerated piles located outside the building are used for the
second-step. At the time of the survey, Cape May County was operating at
reduced capacity because of complaints about odors by neighbors. Since the
end of the summer of 1989, the facility has been operating at full capacity.
443
-------
The reactors are rectangular in shape, 8 meters tall, and the walls slope
slightly outward from top to bottom. One reactor has a volume of 850 cubic
meters and the other reactor is 1050 cubic meters.
The reactors at Cape May County have had fires on several occasions,
caused by both spontaneous combustion and welding accidents. Spontaneous
combustion fires occurred when the composting materials were held in the
reactor for too long because some other part of the system slowed or had
topped proc ssing material. Fires can start at surprisingly low temperatures
in dry compost. Cape May found that when the solids content of compost is
greater than 65 percent and tHe temperature exceeds 80'C, it is possible for a
series of exothermic chemical reactions to elevate the temperature to a point
where true ignition and smoldering fires occur.
Figure 4 is a schematic of a Purac reactor. The compost mix is spread
uniformly across the top of the reactor by a series of conveyors. The _
^^
manifold located about 2 meters below the top of- the compost mass. -The
exhaust airflow rate is higher than the supply airflow rate, so some air is
drawn from outside the reactor through the headspace and through the top 2
meters of the compost mass.
CLAYTON COUNTY, GEORGIA
Table 3 summarizes pertinent information for the Clayton County, Georgia
facility which uses two Taulman reactors. The facility has one first-step
reactor!'and one second-step reactor. Clayton County had the smallest design
capacity of the facilities surveyed. Clayton County was processing all of the
sludge being produced by the wastewater treatment plant Both reactors are
f 8 meters tall and 7.6 meters in diameter. Material Discharged from the
reactors is stored in unaerated piles located in a covered shed.
Figure 5 is a schematic view of a Taulman reactor. The compost mix is
uniformly spread across the top of the reactor by a "slinger" which throws the
material ou? from the center. The composting material then moves by gravity
to the bottom of the reactor, where it is removed by a horizontal auger that
rotates around a center pivot. Air is supplied to the bottom of the reactor
through a series of perforated pipes and is exhausted out the top of the
reactor.
NEWBERG, OREGON
Table 4 summarizes pertinent information for the Newberg, Oregon
facility, which uses two Ashbrook-Simon-Hartley (Ashbrook) reactors Each
reactor is 5.5 meters wide, 2.7 meters high, and 19 meters long. This
facility has had major structural failure of the concrete walls of one of the
reactors- the walls buckled due to excessive pressure inside the reactor.
Neither reactor at Newberg is being operating while the failed reactor is
being rebuilt. Operation is scheduled to begin in October 1989.
444
-------
, Figure 6 is a schematic of an Ashbrook reactor. Compost material 'is
loaded into the reactor by a large hydraulically operated door or "ram" The
atmtheSf^ Inl-T^1 thr°Ugh I6ngth °f the reaCt°r' and a bori-ntal auglr
forward ahn,?n°^/eaCt0rvremOVeS th* C°mp°St- The comP°st m^s is moved
forward about 0.25 meters with each push by the ram. Air is supplied to and
exhausted from the reactor through a series of perforated pipes embedded in
the floor of the reactor. Each set of three adjacent pipes can be either air
supply or air exhaust as needed. eitner air
PLATTSBURGH, NEW YORK
* .,.!ableu5 summarizes pertinent information' for the Plattsburgh, New York
facility, which uses two Fairfield reactors. The reactors are 35.4 meters in
diameter and 3 meters deep. Plattsburgh was processing all of thVsludae
being produced by the wastewater treatment plant. Pla'tsburgh has had
numerous odor complaints and is modifying the odor control systems.
Figure 7 is a schematic of a Fairfield reactor. The compost mix is
Placed around the perimeter of the reactor by a traveling bridge The comoost
material is mixed and move\| toward the center of the reactor bTa series of 5$
augers which hang from the traveling bridge. Material is removed from the
center of the reactor. Air is supplied to the bottom of the • reactor by a
ri8Per£"ted Plpes/ Air is e^sted from the top of the compos? into
and ^ ""W air is °
PORTLAND, OREGON
summarizes pertinent information for the Portland, Oregon
r*af,+n~- r- v tW° para*lel Process trains each using three Taulman
reactors Each process train has two first-step reactors, and one larjer
second-step reactor. The first-step reactors are 8.8 meters tall and 13
Tc.lal5 in diame*er- The second-step reactors are 8.8 meters tall and 16
meters in diameter. Compost material is processed in either one of the first-
step reactors, and then conveyed to the second-step reactor -Material
PnrM^r fr?v thVeact?rs " stored in unaerated piles located outside
Portland was^the only facility composting digested sludge. It was also-
processing the largest amount of sludge. Portland is operating at reduced
excels1 ve weear° SSSVl ""^'i --tenance problems, mosUy'rtlated to
accident. p°rtland has also.had reactor fires caused by welding
SARASOTA, FLORIDA
f •TlableJ summarizes pertinent information for the Sarasota, Florida
«?«* £'• uses two Purac reactors. Sarasota was processing all of the
sludge being produced by the wastewater treatment plant. Sarasota has also
tls one'firs? st'en8 Y ^ m°dified the °d°r C°ntro1 s^stems- ^is facility
^*»^^^^^
reactors is stored in unaerated piles located in a. covwed Sad!
445
-------
SCHENECTADY, NEW YORK
Table 8 summarizes pertinent information for the Schenectady, New York
facility, which uses four American Bio Tech (ABT) reactors. There are two
first-step reactors and two second-step reactors contained inside a building.
All of the reactors are 7.5 meters tall, and have a volume of about 500 cubic
meters. This facility has had several problems including compost fires caused
by spontaneous combustion and minor structural failures of the reactor s
exterior steel frame. However, odor complaints are the reason that the
facility is not operating now. It will not be operated until a new odor
treatment system, which is currently being constructed, is operational.
Operation is scheduled to begin in October 1989.
'Figure 8 is a schematic of an ABT reactor. The compost mix is spread
uniformly across the top of the reactor by a series of conveyors. The
composting material then moves by gravity to the bottom of the reactor, where
it is removed by a horizontal auger that spans the reactor width and travels
back-and-forth along the length of the reactor. The reactors are constructed
of corrugated fiberglass walls that hang from an exterior steel frame. Air is
supplied to and exhausted from the reactor through a'series of "air lances
that hang from headers located at the top of the reactor, The operating mode
of the headers alternates between air supply and exhaust, and the operating
mode is reversed every 20 minutes. That is, one set of air lances supplies
air for 20 minutes, and then exhausts air for 20 minutes, while the next set
of air lances exhausts air for 20 minutes, and then supplies air for 20
minutes, and so on.
PROBLEM AREAS
MATERIALS HANDLING
Mixing appropriate proportions of sludge, amendment, and recycle is
essential to create a desirable compost structure with respect to porosity,
moisture content, and energy balance. Proper porosity promotes easier
materials handling and thorough aeration. Proper moisture content and energy
balance support the biological decomposition of the organic material. Table 9
shows the design and actual percent solids in the mix, and the design and
actual volumetric mix ratios. All of the facilities have had to use more
amendment or recycle than expected, and most have had to use a drier mix than
expected, for several reasons. The reasons include: sludge with a lower
percent solids than expected, difficulty conveying the mix at the design mix
ratio, and difficulty with aeration of the mix in the reactor. These
increases resulted in a greater volume of material to be handled than
expected.
Mechanically and operationally, conveyance systems are the dominant
physical features of in-vessel facilities. Conveyors were the second biggest
cause of system problems at the sites (after odor production). Although a .
high degree of automation for conveyor systems is not required, most in-vessel
composting facilities chose automation to minimize labor costs. .
The type of conveyors used at these facilities have been used reliably
in industry for years; none are experimental. Yet due to differences in
446
-------
materials characteristics, conveyors at in-vessel plants have experienced
problems with jamming, breakage, excessive wear, and spillage.
Screw conveyors and belt conveyors have been used with the most success
The few reported problems with scre.w conveyors were related to excessive wear*
and the accumulation of material on flights and housings. Regular cleaning
and lining the housings with ultra-high molecular weight plastic were
effective for preventing problems. Spillage has been the major problem with
belt conveyors. Wet materials adhere to the belts and do not transfer
completely, resulting in spills and labor intensive cleanup problems These
problems were minimized by using belt scrapers or by placing a layer of dry
material, such as sawdust, onto the belt before adding a wet material.
Drag chain conveyors have not performed well, suffering from plugging
and excessive wear. Modifications to these conveyors have .included increasing
the distance between flights, increasing the clearance between the flights and
the conveyor housing, and decreasing the area of the flights themselves
These modifications allow the composting material to push backwards over or
. around the flight, so it won't be compacted. These modifications, however
also reduce the capacity of the conveyor. '
Cleated belts have had a number of problems including excessive spillage
of both wet and dry materials, and do not appear to be suitable for compostina
facilities. These belts, because of their cleated design, cannot be
effectively scraped.
Reactor-discharge devices are key conveyors in the materials-handling
system; if they fail, the reactors cannot -be used. If compost becomes too
dry, it can form into hard chunks; if it becomes too wet, it can form a glue-
like mass. Both conditions make it difficult to move the compost with a
discharge device. „ - '
A few structural failures with discharge devices have been noted but
excessive wear of auger flights has .been the major mechanical problem '
observed. Compost is a surprisingly abrasive material, especially when it is
compressed at the bottoms of vertical reactors. Decreases of as great as 15
percent of flight thickness in 6 months because of wear on the flight face
were reported. Discharge screws originally hard-surfaced with Triten steel on
both the flight edge and face have not suffered excessive wear.
MOISTURE REMOVAL
There are three basic demands on the aeration system: (1) heat removal
(2) moisture removal, and (3) oxygen for biological metabolism. The aeration'
systems at the sites visited appeared to be adequately sized'to meet the
oxygen supply and heat removal requirements; only Schenectady had problems
with elevated temperatures. Table 10 shows the total aeration capacity of the
facilities, the unit capacity per cubic meter of empty reactor volume, and the
length of the path that air must take to pass through the reactor.
The major problems have been related to moisture removal: condensation
and fog formation above the reactors and formation of a "hardpan" layer
Inadequate aeration for curing/storage piles was a third major problem."
447
-------
Condensation and Fog Formation
In-vessel composting systems must provide for adequate collection of air
after it has passed through the compost. Air can hold four times as much
water at 60°C than at 40°C. Therefore, if the temperature drops before the air
leaves the reactor, it does not remove as much water, and condensation occurs
JiSil the reactor and composting mass. If the air collection piping extends
into the composting mass, exhaust holes in the piping can become exposed due
to natural settling and uneven subsidence. Such exposure can cause odors to
escape saturated air to cool, and condensate to form within the reactor.
Systems with open reactors discharge warm, moist air from the compost
bed into a building or work space above the reactor. In these cases, the
introduction of too much cold outside air into these same areas results in the
condensation of water vapor and the formation of fog The worst case was
found at Pittsburgh, where the fog hampers the stabilization of the product,
promotes corrosion of equipment, and is so thick that operators cannot see to
work.
Hardpan Formation
At least three of the facilities have noted the formation of a "hardpan"
layer at the bottom of the reactor that is nearly impermeable to air.
Formation of a hardpan layer occurs when leachate forms because of "adequate
moisture removal from the reactor. The leachate extracts solubles as it moves
down through the compost and drains to the low points in the reactor. If
these low areas are low in humidity, the leachate dries, cementing material in
the boUom of the reactor into an impermeable layer. At Clayton" County
leachate seeped into the one-centimeter unwashed river stone layer on the
bSttoi of the reactor, cemented the particles together and plugged the holes
in the aeration piping. The stone was replaced "t^^T"?*^" ;a!j!jer
round river rock, and the aeration piping was remounted in the middle rather
than at the bottom of the gravel layer, separating it from the pooling
leachate. At Cape May County a hardpan layer formed which was attributed to
excessive leachate generation. The mix ratio and amendment specifications
were Ranged to obtain a more porous initial compost mix, which significantly
reduced leachate generation. Also, the aeration system was »j£«"* "> fve
the operators better control of the airflow. The hardpan problem has not
recurred since these changes were made.
CURING/STORAGE AERATION
The curing/storage process is an important step in the overall sludge
composting process. The composting process does not stop when material js
removed from the reactor(s); the compost continues to have an oxygen demand.
Ken where compost is processed in two reactors (at Clayton. County, Portland,
slrasota! and Lhenectady), the unaerated curing/storage piles have reheated,
indicating the continuation of biological activity. Depending on the
lability of the compost, oxygen demand may be elevated to the point where the
available oxygen is consumed and anaerobic conditions develop. Movingthe
piles then causes odors to be emitted. At Akron and Cape May County the
exteriofcuring/storage piles are aerated or will be aerated in the future
because of odor problems. At Plattsburgh, the curing/storage piles are
448
-------
odorous, and an evaluation is underway to determine whether they should be
aerated.
ODOR CONTROL
In the U.S. effective odor control is a critical requirement for the -
success of a composting facility. Fear of potential odors can cause public
opposition to proposed plant sites. Qdor complaints can cause a regulatory
agency to shut down or curtail the operations of an existing facility. Plant
odors can also hurt product sales.
Odor control was the greatest problem found at the sites surveyed. Of
the eight composting facilities surveyed in this project, six had received
odor complaints from the local community. At five of these sites, the com-
plaints were serious enough to cause the plant owners to retrofit the facili-
ties with additional odor treatment equipment. Two of the five plants ceased
operations while the new equipment was being installed. Two plants were
operated at reduced capacity to limit odor production while the new treatment
equipment was piloted, designed, and installed. The last plant continued to
operate but at the same time was proceeding with the odor treatment system
retrofits as quickly as possible.
Odors found at compost facilities are caused by a large number of
compounds that originate in both the sludge and the amendment. In-vessel
systems may produce more amendment-related odors than static pile or windrow
systems because the sawdust used at most in-vessel facilities has a greater
surface area than the wood chips usually used in the other systems.
Odor-causing compounds fall into a number of classes that differ in-
their physical and chemical properties. Some are acidic; some are basic.
Some can be destroyed by oxidation; others cannot. These compounds are not
equally soluble in water, nor equally amenable to adsorption. Moreover, the
mix of compounds in the odorous gas stream can change over time, depending on
the composting materials and the reactor operating conditions. For this
reason, treatment systems must employ a broad spectrum of removal mechanisms
to be effective. Treatment systems that rely on only one or two removal
mechanisms such as single-stage wet scrubbers and ozone oxidation chambers
have not been successful.
Table 11 summarizes the odor treatment methods that were found at the
eight plants surveyed. Of all the methods shown in Table 11, bubbling air
through wastewater appeared to be the most successful. Dilution alone was
completely ineffective as a treatment method. Biofilters appeared to remove a
large fraction of odors based on measurements taken at the site. However, the
facility using the biofilters "has still had numerous odor complaints and is
planning to install wet scrubbers to replace the biofilters. Ozonation was
ineffective and replaced by wet scrubbers. Activated carbon was tested at
pilot scale at one location but also appeared to be ineffective.
At Portland, Newberg, and Clayton County, odorous air is bubbled through
wastewater for treatment. Operators at Clayton County have received some odor
complaints, but the odors are, generally not a problem, partly because of the
rural setting of the plant. The Portland and Newberg facilities have not
449
-------
received any odor complaints, but both plants are located in outlying in-
dustrial areas and both compost relatively stable sludges. These mitigating
factors make it difficult to determine the effectiveness of this technique.
The design parameters associated with this technique are not well established,
and a ground-level discharge is created.
At the time of the.site visits, packed tower wet scrubbers were in use
at three of the eight facilities visited. Table 12 lists the odor sources
treated and the chemicals used in these scrubbers. Sarasota installed the
system shown in Table 12 after the site visit. "DeAmine" is a proprietary
chemical.
Single-stage wet scrubbers were ineffective for several reasons.. In
addition to not being able to treat a broad spectrum of compounds, reasons
include overloading the scrubbers and using inappropriate scrubbing chemicals.
Two stages appear to be the minimum number needed for effective odor
treatment. Table 13 lists the planned odor treatment facilities at several of
the plants. Typically, one stage uses an acid, and the other stage uses an
oxidizing agent. Cape May County is planning to try corn syrup as a
surfactant for aldehyde compounds. Packed,tower wet scrubbers were the only
scrubbers observed at the eight plants studied. Mist-type scrubbers have been
used successfully at the Montgomery County Composting Facility, a large
enclosed static pile operation in suburban Washington, D.C.
Although this discussion of odors has concentrated on treatment systems,
an effective odor control system must include more than just a treatment
system. A complete odor control plan, including quantification and
qualification of all odor sources, and proper control of the composting
process, must be'implemented. In addition, no treatment system is 100 percent
effective; some concentration of odorous compounds will be discharged after
treatment. To avoid undesirable effects on local residents, the odorous
emissions must be diluted and/or dispersed so that ambient air concentrations
of the odorous gases fall below detection thresholds.
COSTS
Capital costs for the eight facilities are shown in Table 14. The
facilities are listed by design capacity, from smallest to largest. Costs
shown are the original construction, including change orders, in millions of
dollars; additional expenditures made, since start-up, in thousands of dollars;
and expeditures that the facilities anticipate in the near future, in
thousands of dollars. Note that the majority of additional and future capital
costs are for odor control equipment, and that these costs range up to 40% of
the original cost.
Figure 9 shows annual operation and maintenance costs, expressed as
dollars per dry metric ton, as a function of current operating capacity. This
figure shows an economy of scale for these facilities. Clayton County, the
smallest^facility, appears to have low costs, but this facility obtains its
amendment at no cost. Akron, the largest facility, appears to have high
450
-------
costs, but the cost shown includes sludge dewatering costs which are not
included for the other facilities.
CONCLUSION
In-vessel composting is a total system that comprises a number of
integrally related components, including materials handling, reactors, an
aeration system, exterior curing/storage facilities, an odor control system,
and marketing of the compost product. Inadequate or improper design,
management, operation, or maintenance of any of these components will
seriously impact the total system, and most likely adversely affect the
compost product.
Most of these facilites failed to realize that JL sludge in-vessel
composting system includes much more than just the reactor(s). The major
problems encountered at these facilities were not related to the reactor
itself, but to the other components, including materials handling and odor
control. Indeed, many of the difficulties encountered resulted from a failure
to recognize the totality of the system that must be managed.
Finally, it should be noted that because this study focused on problem
areas, in-vessel composting appears to be plagued with problems. However, all
but two of the facilites discussed were producing and marketing a compost
product. All of, the facilities were meeting EPA requirements for pathogen
reduction, and all were meeting State limits on heavy metals and organics.
ACKNOWLEDGEMENT
The study described in this report has been funded wholly or in part by
the United States Environmental Protection Agency under contract 68-03-334S to
Camp, Dresser, and McKee, Inc. and contract 68-C8-0014 to Eastern Research
Group. This study is fully documented in a report entitled "Technology
Transfer Summary Report: In-Vessel Composting of Municipal Wastewater Sludge",
EPA 625/8-89/016," available from the Center for Environmental Research
Information, U.S. Environmental Protection Agency, Cincinnati, Ohio, 45268.
451
-------
1988 In-Vessel Study
PlattsburgjTjjfai^field)
Tdy (ABT)
May County
Purac)
Sarasota^jfPurac)
FigDre 1. In-Vessel Locations Studied
452
-------
Plant Configurations
Site
AKR
"CAP
CLA
NEW
PLA
POR
SAR
SCH
Horz, Reactor
System
Paygro
Purac
Taulman
-------
o
0)
O
CO
CO
J_
454
-------
HEADSPACE
VENTILATION
PIPING
(to odor control
system) \
FEED CONVEYOR
EXAUSTx
AIR
HEADER
TO ODOR CONTROL SYSTEM
AIRFLOW DIRECTION
7/ ff / /' // J/ / / / / ft /}?/?/ /i
mm
DISHARGE
/SCREW
DISHARGE
CONVEYOR
Figure 4. Purac Reactor
455
-------
TO ODOR
CONTROL
DISCHARGE
SCREWS
INFEED
CONVEYOR
DISTRIBUTOR
AIRFLOW
DIRECTION
COMPOST TO
DISCHARGE CONVEYOR
AERATION PIPING
Figure 5. Taulman Reactor
456!
-------
o
+J
o
to
o>
<0
I
O
O
•O
s.
-Q
x:
W)
«=C
VO
-------
si
o
458
-------
AIR
SUPPLY
BLOWERS
DISCHARGE
SCREW
AIR-SUPPLY
MANIFOLD
\\1A\\\\\\\\\\\\\\
FEED
CONVEYOR
EVACUATION
BLOWERS
TO ODOR
CONTROL
AIRFLOW
DIRECTION
AIR-LANCE
-DISCHARGE
CONVEYOR
Figure 8. American Bib Tech Reactor
459
-------
ANNUAL COSTS
Annual O&M Costs ($$/dmt)
$500
$400
$300
$200
$100
$0
CAP
SAR
CLA
PLA
POR
10 15 20 25 30
Current Capacity (dmt/d)
Figure 9. Annual Operation and Maintenance Costs
460
-------
Akron. Ohio (AKR)
System: Paygro
Design Capacity: 66 dmt/d*
Start-Up Date: 12/86
Type of Sludge: Primary + Waste Activated
20 - 26% Total Solids
> 70% Volatile Solids
Current Capacity:
Current Status:
* dry metric tons per day
19 dmt/d*
Operating at reduced
capacity due to odors
Table 1. Summary of Akron, Ohio Facility
461
-------
Cape Mav County. New Jersey (CAP)
System:
Design Capacity:
Start-Up Date:
Type of Sludge:
Current Capacity
Current Status:
Purac
18 dmt/d*
5/85
Primary + Waste Activated
20 - 30% Total Solids
70 - 80% Volatile Solids
FeCI3 added to primary
6.6 dmt/d*
Operating at reduced
capacity due to odors
* dry metric tons per day
Table 2. Cape May County, New Jersey Facility
462
-------
Clayton County. Georgia (CLA)
System:
Design Capacity:
Start-Up Date:
Type of Sludge:
Current Capacity:
Current Status:
* dry metric tons per day
Taulman
2.6 dmt/d*
8/8.6
Waste Activated
15 - 17% Total Solids
68 - 79% Volatile Solids
AI2(SO4)3 added
1.2 dmt/d*
Processing all sludge
produced by WWTP
Table 3. Summary of Clayton County, Georgia Facility
463
-------
Newbera. Oregon (NEW)
System: Ashbrook-Simon-Hartley
(Ashbrook)
Design Capacity: 3.2 dmt/d*
Start-Up Date: 8/87
Type of Sludge: Oxidation Ditch.
12 - 20% Total Solids
- - - Volatile Solids
Current Capacity: — dmt/d*
Current Status: Not operating due to
structural failure
* dry metric tons per day
Table 4. Summary of Newburg, Oregon Facility
464
-------
Plattsburah. New York (PLA)
System:
Design Capacity:
Start-Up Date:
Type of Sludge:
Current Capacity:
Current Status:
* dry metric tons per day
Fairfield
31 dmt/d*
3/86
Primary + Waste Activated
20 - 25% Total Solids
- - - Volatile Solids
KMnO4 added to sludge
18 dmt/d*
Processing all sludge
produced by WWTP
Table 5. Summary of Pittsburgh, New York Facility
465
-------
Portland. Oregon (FOR)
System:
Design Capacity:
Start-Up Date:
Type of Sludge:
Current Capacity
Current Status:
Taulman
54 dmt/d*
3/85
Digested Primary +
Digested Waste Activated
23 - 28% Total Solids
35 - 50% Volatile Solids
27 dmt/d*
Reduced capacity due
to mechanical problems
* dry metric tons per day
i
Table 6. SUtomary of Portland, Oregon Facility
466
-------
Sarasota. Florida (SA-R)
System: Purac
Design Capacity: 5.7 dmt/d*
Start-Up Date: 8/87
Type of Sludge: t Waste Activated
14% Total Solids
76% Volatile Solids
Current Capacity:
Current Status:
* dry metric tons per day
3.3 dmt/d*
Processing all sludge
produced by WWTP
Table 7. Summary of Sarasota, Florida Facility
467
-------
Schenectadv. New York (SCH)
System: American Bio Tech
(ABT)
Design Capacity: 14 dmt/d*
Start-Up Date: 7/87 .
Type of Sludge: Primary + Waste Activated
20 - 28% Total Solids
73% Volatile Solids
Current Capacity: — dmt/d*
Current Status: Operating at reduced
capacity due to odors
* dry metric tons per day
Table 8. Summary of Schenectady, New York Facility
468
-------
Design vs Actual Mix
AKR
CAP
CLA
NEW
PLA
POR
SAR
SCH
Vol. Mix Ratio*
% Solids
Design
1/0.66/1.33
1/1/1
1/0.35/5
1/0.67/1.67
1/0.4/0.6
1/1/2
Actual
1/1/2
1/0.6/1
1/1.4/1.4
1/0.5/10
1/1.4/3.3
1/1.6/0.6
1/0.56/2.44
Design
40
40
38
40
32
35
35
Sludge/Amendment/Recycle
Table 9. Design and Actual Mix Ratios
Actual
44
37
34
38
4.0
37
40
37
469
-------
Aeration Capacity
AKR
CAP
CLA
NEW
PLA
POR
SAR
SCH
Total
Capacity
(L/s)
110,000
7800
710
2400
15,000
3700
2500
4500
Unit ,
Airflow
(L/m3-s)
6.8
4.1
.0.88
3.0
3.5
0.43
1.2
2.3
Path
*
Length
(ml
2.4-3.0
6.4
7.9
0.6-9.1
2.4-3.0
7.9
7.0
1.8-2.1
Table 10. Aeration Capacity of Reactors
470
-------
Initial Odor Treatment Systems
AKR Dilution (Process Air)
Compost Biofilter (Non-process Air)
CAP 2-Stage Scrubber (Process Air)
1-Stage Scrubber (Non-process Air)
CLA Bubbling into Wastewater Pond
NEW Bubbling into Oxidation Ditch
PLA Single-Stage Wet Scrubber
POR Bubbling into Channel at Headworks
SAR Ozone Contact Chamber
SCH Single-Stage Wet Scrubber
*
/""
Table 11. Initial Odor Treatment Systems
471
-------
Wet Scrubber Installations
Stage Source
CAP
PLA
SAR
SCH
1
1
1
2
1
1
2
3
1
Chemicals
Sludge storage NaOCI+NaOH
Mixing
Reactors
All
All
All
NaOCI+NaOH
NaOCI+NaOH
H2S04
H2SO4+DeAmine
Cooling
H2S04
NaOCI+NaOH
H2SO4+DeAmine
Table 12. Current Wet Scrubber Installations
472
-------
Planned Wet Scrubbers
Stage
Chemicals
AKR
CAP
PLA
SCH
1
2
1
2
3
1
2
1
2
3
H2S04
NaOCI+NaOH
Cooling
H2SO4+Corn Syrup
H9(V
H2SO4+DeAmine
NaOCI+NaOH
Cooling
H2SO4+DeAmine
NaOCL+NaOH
Table 13. Planned Wet Scrubber Installations
473
-------
Capital Costs
Original Additional Future
($$ Millions) ($$ Thous) ($$ Thous)
^ *
CLA
NEW
SAR
SCH
CAP
PLA
POR
AKR
i*r ^r1 •*••••• ^^ • B^
2.0
2.8
4.2
6.7
8.4
13
12
22
80
— -
* 3000*
—
205
•
—
??
200*
470*
—
5300*
??
7500*
Odor Control
Table 14. Capital Costs
474
-------
THE DEVELOPMENT OF RISK ASSESSMENT METHODOLOGIES
FOR USE IN REGULATION OF SEWAGE SLUDGE DISPOSAL
by
Randall J.F, Bruins
Norman E. Kowal
Cynthia SpnVch-MullIn
Steven D. Lutkenhoff
Environmental Criteria and Assessment Office
U.S. Environmental Protection Agency
Cincinnati, Ohio
This paper has been reviewed In accordance with
the U.S. Environmental Protection Agency's peer
and administrative review policies and approved
for presentation and publication.
Prepared for Presentation at:
Twelfth United States/Japan Conference
on Sewage Treatment Technology
Cincinnati, Ohio
October 12-13, 1989
475
-------
INTRODUCTION
Risk assessment 1s a structured procedure for making quantitative
estimates of the potential for harmful agents to cause adverse effects on
human health or the environment. The use of risk assessment methods to
determine appropriate limits for chemicals In environmental media such as
air and water 1s well established. This paper will describe their use for
estimating risk from chemicals 1n sewage sludge, and their potential use for
estimating pathogenic risks as well.
The Environmental Criteria and Assessment Office of the U.S. EPA Office
of Research and Development In Cincinnati, Ohio, has recently developed risk
assessment methodologies which can be used to calculate criteria for
protecting human health and the environment from toxic pollutants In
municipal wastewater sludges (1-4). Since municipal sludges are managed
utilizing a number of different reuse or disposal practices, separate
assessment methods were developed to address land application (Including
distribution and marketing [D&M]), Iandf1ll1ng, land-based Incineration, and
ocean disposal of sludges. Assessment methods to address surface disposal
(I.e., surface Impoundment) of sludges are currently under development.
Each methodology provides procedures for estimating the fate and transport
of sludge-borne pollutants and for evaluating the potential for effects on
humans or other organisms. The methods also describe data requirements for
using these procedures to calculate criteria. Th^ criteria are not
necessarily numerical 1n all cases. In some Instances, the risk assessment
calculations can be used to determine appropriate controls on the design or
operation of the management practice which achieve the same level of
environmental protection. Such management controls are often preferred for
their practicality. In many cases, however, the methodologies suggest that
numerical limits on pollutant levels 1n the sludge should be calculated.
The risk assessment methodologies for each practice describe algorithms
that can be used to calculate whether or not adverse effects would be
expected from a given sludge constituent under a given set of conditions.
Separate algorithms are presented for different pathways of potential
exposure. For example, the methodology for sludge landfUUng develops
separate algorithms for contaminant movement to groundwater and for
volatilization through the soil cap. The algorithms range from simple to
complex. For example, 1n order to prevent phytotoxldty, the land
application methodology states that soil levels toxic to plants should be
determined from the literature for each chemical,, and that sludge
applications should be limited to prevent the occurrence of such levels.
Minimal calculation Is required. On the other hand, the algorithm for the
476
-------
groundwater pathway for sludge landflVMng requires the use of three
computer models In succession to determine the effects of unsaturated zone
e°ChWlJa\equl11br1a- and saturated zone transput
, on potential for human exposure.
Using these procedures, conditions can be established which would
f»fSe effecls* As ^ntloned previously, these cond tlons
of h n fr .^ °?nSlSt of,a S6t Of ^commendations for design or operation
of the Practice For example, a maximum ground slope or minimum depth to
groundwater could be recommended for land application practices In Srder to
prevent surface or groundwater pollution, respectively Often however
numerical criteria which limit the chemicals themselve are derlJed '
Depending on the algorithm, these numerical criteria may take the form of
nf d SI The C!i ter1on 1s the amount Wh1ch co"1d be reapplled annually
without ever exceeding a maximum cumulative amount. dnnuany
huma!?S °r 3n1ma1s may consume the Slud9e
477
-------
son and could be exposed to pure sludge. For these .types of exposures, the
dry-weight pollutant concentration 1n the sludge Is more pertinent than the
pollutant application rate.
Offslte movement of sludge-borne chemicals can also occur through
transport 1n surface runoff or by percolation to groundwater. These
movements are treated as a function of annual areal pollutant application
rate, rather than sludge concentration. A1r concentrations resulting from
volatilization may be a function of sludge concentration (wet weight)
Immediately upon application, but long-term average concentrations would
again be a function of the annual areal application rate.
LANDFILLING
LandfUUng was estimated to account for 15 percent of municipal sludge
disposal 1n 1980 (5) and was estimated at 42 percent 1n 1986 (6). Host
landfUled sludge 1s codlsposed with municipal solid waste, while a lesser
amount (1.3 percent of total sludge disposal In 1986) Is disposed In
sludge-only landfills, or monofUls. The risk assessment methodology for
landfllUng (2) considers only the monofHUng practice; codlsposal will be
dealt with 1n a separate assessment effort. It Is assumed that site design
features such as runon/runoff controls are adequate to protect local surface
waters and that the dally application of cover material prevents any
problems of windblown dust. Numerical criteria are suggested to prevent
hazard from groundwater contamination: design criteria are not recommended
since very few sludge landfills have liners and these liners may rupture.
Numerical criteria are also suggested In order to limit the vapor
concentrations downwind from the site.
Contaminant movement to groundwater and subsequent movement to a point
of human use are predicted using a succession of transport and geochemlcal-
equlllbrlum models. The required Input to the first model (unsaturated zone
transport) Is a leachate concentration. In addition to leachate
concentration, the total mass 1n the landfill of each pollutant must be
estimated In order to determine the maximum length of time over which a
leachate of a given concentration could be produced, since this duration can
affect the predicted maximum concentration at the point of use.
Downwind movement of vaporized contaminant 1s estimated using a
Gaussian air dispersion model. The vapor concentration Immediately above
the source during dally (uncovered) operation, or post-closure, Is estimated
from the leachate concentration using Henry's Law. Thus the leachate
concentration Is required for both of these transport pathways.
For organic compounds, the leachate concentration can be estimated from
the whole-sludge concentration and the partition coefficient of the
contaminant. Inorganic compounds are assumed to be present at the limits; or
their solubility, based on the maximum effluent or leachate levels observed
1n wastewater treatment plants.
478
-------
hu « ?f °a ° be ana1yzed 1n the leachate or whole sludge, as determined
by preliminary screening procedures, could Include various Setals. cyln"de
e te?s (7 55°Sam1neS* chlorinated hydrocarbon pesticides, PCB and phthallte
SURFACE DISPOSAL
Placement of sludge 1n surface Impoundments of some type 1s common-
/J6 'f
-------
OCEAN DISPOSAL
Ocean disposal of municipal sludge by barging 1s currently permitted In
only one location, the newly designated 106-Mile Ocean Waste Dump Site off
the New York/New Jersey coast (9). Ocean disposal (at the previously
designated Hl2-H1le Site") accounted for about 4 percent of U.S. sludge
disposal 1n 1980 (5) and about 5.5 percent In 1986 (6).
The risk assessment methodology for ocean disposal of sludge Is
concerned with both protection of marine life near the site and protection
of humans consuming seafood from that area. Water-column concentrations to
which marine organisms are exposed are estimated from dispersion models. To
calculate short-term concentrations within the dump site Itself, where toxic
effects on marine organisms are most likely, the sludge concentration (wet
weight), as well as speed, dumping rate and positioning of barges, Is the
required Input Information. To estimate long-term, down-range concen-
trations, which are of more Importance for assessing human exposure, only
the total contaminant Input to the dump site on a dally basis Is required.
Methods are not yet available to estimate sedimentation rates for sludge
particles and potential for seafloor accumulation at the 106-Mile site.
Chemicals potentially requiring analysis Include metals, chlorinated
hydrocarbon pesticides, PCB, PAH, phthalate esters and aromatic amines
(7 8). In addition to analysis of Individual constituents, whole-sludge
bloassay methods may be used to determine the degree of dilution required to
prevent aquatic toxldty. While not dealt with 1n these risk assessment
methodologies, which focus on evaluating hazards of Individual chemicals,
these bloassay methods have been required for ocean-disposed wastes under
other authorities (10), and are also useful for sewage sludge.
PATHOGENS
The U S EPA Is currently Devaluating the existing, technology-based
pathogen regulations, which define Processes to Significantly Reduce
Pathogens (PSRP) and Processes to Further Reduce Pathogens (PFRP) (11). A
feasibility study for performing risk assessments for Pathogens was
completed 1n 1985 (12, 13). Based on the feasibility study, the U.S. EPA Is
developing a quantitative risk assessment methodology for pathogens (I.e.,
bacteria, viruses, protozoa and helminths) 1n land-applied sludge. The
methodology examines human exposures to pathogens that could result from the
following pathways:
aerosols generated by sludge application or Incorporation
wind-generated partlculates from soil surface
runoff of precipitation or Irrigation
direct contact with soil or crop
480
-------
movement through ground water to well
use of contaminated ground water for Irrigation
• consumption of crops or animal products
i
CONCLUSION
aswsa t: SWOT,, „
481
-------
REFERENCES
(1)
Office of Water Regulations and Standards,
1986.
, O.C., April,
m 11 S EPA Development of Risk Assessment Methodology for Municipal
{2) SluVe Undflllln? Prepared by the Office of Hea 1th and. nv ro™ental
Assessment Environmental Criteria and Assessment Office, Cincinnati,
Shlo! Tor the Of.f?ce of Water Regulations and Standards, Washington,
D.C.', April, 1986.
m US EPA Development of Risk Assessment Methodology for Municipal
(3) U.S. EPA. ^ve.opm Qfnce Qf Health and
,taT^s«sment? Environmental Criteria and Assessment Office,
_!, Ohio! forthe Office of Water Regulations and Standards,
Washington', D.C., April, 1986.
(4) U S EPA. Development of Risk Assessment Methodology for Ocean
{*) u.a. crn. •«•» _P , .,...„„ DronareH hw the Office of Health
Uiliwiiviiv*1*-* _.---, - f\nf
Washington, D.C., April, 1986.
151 S^ EPJ3SM J£K.^l£l ^rSJtaVSI^arcH Laboratory.
Cincinnati, Ohio, Oct., 1983. p. 1-2.
(6) U S EPA 1989. Standards for the Disposal of Sewage Sludge; Proposed
Rule. Federal Register, 54 (23): 5754.
m U S EPA. Summary of Environmental Profiles and Hazard Indlcles for
{?) Constituents of Municipal Sludge: Methods and Results, Office of
Water Regulations and Standards, Washington, D.C., July, 1*03.
(8) LomnUz, E.. Bruins. R. and Fradkln, L. Blocycle. Vol. 26, No. 7,
Oct., 1985, pp. 52-54.
(9) U.S. Epa. Ocean Dumping: Final Designation of Site. Federal
Register. 49 (88): 19005-19012, 1984.
MO) rnrio nf Federal Regulations Title 40, Part 227, July 1, 1985,
pp. 155-168.
m) nnri* Of Federal Regulations. Title 40, Part 257, 3uly 1, 1985,
pp. 341-342.
482
-------
(12) Fradkln, L. Lutkenhoff,. S., Stara, J.,. Lomnltz, E., and Cornaby B
J. Water Poll. Contr. Fed. 55 (2): 1036-1039, 1985. ornd°y. »>•
(13) U.S.
,; Pathogen Risk Assessment Feasibility Study.
?3^0ft1Ce Jf»ealth and Environmental Assessment.
ff r nCrlter a4!nd Assessment Office, Cincinnati, Ohio and
Office of Water Regulations.and Standards, Washington, D.C., Nov
483
-------
-------
UV DISINFECTION
by
0. Karl Scheible
HydroQual, Inc.
1 Lethbridge Plaza
Mahwah, New Jersey 07430
This paper h.as been reviewed in accordance with
the U.S. Environmental Protection Agency's peer
and administrative review policies and approved
for presentation and publication.
, Prepared for Presentation at:
Twelfth United States/Japan Conference
on Sewage Treatment Technology
Cincinnati, Ohio
October 12-13, 1989
. 485
-------
INTRODUCTION
The use of ultraviolet (UV) radiation for the disinfection of treated
wastewaters is an accepted practice in the United States. There is an
increasing pace of installation at new plants and chlorine contact chambers at
a number of existing plants are being retrofitted with UV equipment. This has
been spurred by the acceptance of the technology as a viable and established
process, and by the negative aspects of chlorination: very often
dechlorination is required to meet increasingly stringent limits on residual
chlorine levels; chlorination can result in the formation of toxic chlorinated
organics; and there are significant concerns over the safety of transporting
and storing chlorine and sulfur dioxide. Conversely, UV has no residual, does
not affect chemical reactivity, and does not have safety concerns that would
affect the surrounding community.
The USEPA disinfection program initiated research and demonstration efforts
in the mid-seventies to develop the UV process as an effective alternative to
chlorination. This support has continued, particularly in the construction
at
grants funding under the EPA Innovative and Alternative Technology program. By
1987 forty-two plants had received I/A funding for UV facilities. Research
efforts were directed to assessing the efficiency of the process, and to
establishing and demonstrating protocols by which the process can be designed
and evaluated. This design procedure will comprise the major portion of this
presentation. Before this, 'however, .some discussion will be given to how UV
works and the status of UV installations in the United States today.
UV DISINFECTION
Ultraviolet radiation disinfection is a physical process, relying on
photochemical changes brought about when far-UV radiation emitted from a source
is absorbed by the genetic material of a cell (deoxyribonucleic acid, or DNA).
The damage to the DNA results in, the inability of the cell to replicate, a
lethal effect. The most effective spectral region for germicidal activity
'corresponds to the maximum absorption spectrum of nucleic acids (Figure 1) .
This is between 250 and 265 nm.C1)
486
-------
Very conveniently, low pressure mercury arc lamps are very efficient in
generating UV light within this range: approximately 85 percent of their light
output is monochromatic at 253.7 hm. They are typically made of Vycor glass,
are 0.75 m or 1.5 m long, and 0.75 cm in diameter. Some disinfection
applications exist that use alternative lamps. The Lewisburg plant in Ohio
uses medium pressure mercury lamps, which have higher intensities and a broader
spectrum of output. A modified low pressure lamp that is capable of higher
intensities than the conventional low pressure lamp is being demonstrated at
the Baldwin, Florida plant. Overall, however, the conventional low pressure
lamps comprise the source of UV energy in effectively all disinfection systems
today.
PHOTOREACTIVATION
Most organisms have an ability to repair and reverse the lethal effects of
UV, a phenomenon broadly termed photoreactivation. It is typically a
photoenzymatic mechanism, and requires concurrent or post-treatment with non-
ionizing radiation in the near UV or visible range. The repair is of a
constant fraction, independent of dose. This is clearly exhibited by data
shown on Figure 2, developed from an EPA study conducted at a New York City
treatment plant.(2) The figure^ shows the residual fecal coliform density as a
f
function of UV loading (Ipm/UV Watt). The data represent analyses of samples
held,in the dark after UV exposure, and the same sample exposed to sunlight for
at least one hour. Photorepair accounted for approximately a 1 log increase in
fecal coliform density. .
Most organisms have the ability to repair, although there are some
pathogens that will not, including the viruses. The total and fecal coliform
groups will exhibit photorepair, while the enterococcus group does not/ In the
design of UV systems, it is important to understand that environmental
conditions at. a treatment plant will be conducive to photoreactivation. As
such, the phenomenon must be considered in sizing the system; generally this
suggests that the system must be sized to accomplish an additional log
reduction beyond the design effluent conditions.
487
-------
STATUS OF UV SYSTEM APPLICATIONS IN THE UNITED STATES
There are a number of ways in which the UV reactors have been designed to
disinfect treated wastewaters. Basically the intent must be to minimize the
loss of UV energy and to maintain a minimum exposure time for the wastewater.
This requires close contact of the wastewater with the UV source, and plug flow
conditions within the reactor itself.
Closed shell quartz reactors have the lamps (with quartz sleeves) fixed
within stainless steel cylinders in full contact with the wastewater. The lamp
spacing is typically 8 to 12 cm (centerline), with the wastewater flow directed
parallel to the quartz. The systems are typically gravity flow, with piped
inlet and outlet. These systems, because of the higher velocities at the
entrance and exit points, tend to exhibit higher advective dispersion,
affecting disinfection performance. The design also provides poor access to
the lamps and quartz for maintenance and repair. These types of units tended
to dominate the market through the mid-eighties, but are currently being used
less frequently.
A "non-contact" configuration uses thin-walled Teflon pipes to carry the
liquid. These are surrounded by unsheathed lamps, placed coaxially with the
Teflon tube. The hydraulics of this arrangement are very good, with low
dispersion plug flow through the long (typically 1.5 to 3 meters), small
diameter (9 cm) pipes. Energy efficiency is generally lower, however, than the
quartz systems, and the systems have been difficult to maintain. These are
generally not considered for new applications.
Open-channel systems rely on the submergence of the quartz and lamp bundle
in an open-channel. The lamps are either fixed in place, with flow directed
perpendicular to the. lamp, or inserted as modules, with the flow parallel to
the lamp. The fixed open-channel units reflect the earlier .designs and
suffered problems with accessibility and maintenance; these are no longer
installed.
488
-------
The newer open-channel designs use a modular arrangement of quartz-sheathed
lamps that can be easily inserted into and removed from the channel. The lamps
are placed vertically into the channel, or horizontally, with the flow parallel
to the lamps. Figure 3 shows an open-channel, horizontally placed, modular UV
design. These systems represent the state-of-the-art in UV equipment,
accounting for the major fraction of new installations and retrofitting of
chlorine contact basins.
The open-channel modular design is best suited to UV process design. The
lamps are placed in the liquid at relatively close spacing, affecting efficient
use of the UV energy. Additionally, the open-channel design itself can be
exploited to favor lower dispersion, plug flow conditions, by forcing a long
liquid pathlength relative to its cross-sectional dimension- The modular
design also allows easy removal of the units for cleaning and repair, the lack
of which hampered the efficient operation of the earlier designs.
Figure 4 gives a perspective on how UV has progressed, particularly with
the types of systems that are being installed. A surveyO) in 1984 identified
53 operating plants. Most were small; 80 perdent had design flows less than 1
mgd (3,900 m3/d). Nearly a third were the non-contact Teflon units, and half
were closed shell reactors. The remainder were open-channel designs, but only
one of these used the modular approach.
In 1988 there were close to 300 plants in operation. Of 177 for which
information was available, 59 percent had design flows less than 1 mgd (3,900
m3/d). Larger systems are being put on-line; however, 38 percent have design
flows between 1 and 10 mgd (3,900 and 39,000 m3/d) and several plants have
capacities greater than 10 mgd (39,000 m3/d). The largest operating plant is
Madison, Wisconsin, with a design peak of approximately 100 mgd (390,000 m3/d).
Quebec City, Canada, has commissioned two plants with a combined design
capacity of 230 mgd (900,000 m3/d).
r
The distribution of the systems reflects the direction of the UV equipment
field. Seventeen percent are Teflon,,with only a few new installations since
489
-------
1984. Only a quarter '(26 percent) of the plants use the closed shell design;
they are still being installed, but the trend is away from them for wastewater
applications. There are no new fixed in-channel units. The trend is obviously
to the open-channel modular systems, which after four-years now comprise'
greater than half the operating plants. Of these, 85 percent utilize the
horizontal configuration and 15 percent the vertical placement.
DESIGN OF UV PROCESS
Through the mid-eighties, reflecting experience with "first-generation"
systems, problems with UV installations related to severe component,
fabrication and installation deficiencies (ballast failures, panel heating,
electrical failures, and poor reliability). This -has changed as the field is
dominated more and more by "second generation" systems. There is a far greater
degree of standardization, and the modular concept is allowing for a greater
degree of flexibility and ability to affect quick repair and maintenance. More
attention is being directed to design protocols that will assure proper sizing
and can assist in the evaluation of existing systems.
A significant effort for the EPA Disinfection Program was the development
and demonstration of a design protocol for the UV process. This was first
described from a large scale pilot study conducted at the New York City Port
Richmond plant(2), and subsequently detailed in the EPA Design Manual^) and
the Journal. WPCF.(4> Several in-field,studies were also conducted to support
specific aspects of the design protocol, and are described in the Manual.
Since then, the design protocol has been used to develop the process sizing and
layout of several new plants. Most recently, the USEPA has funded a pilot
Study to further evaluate UV process design. The field effort, conducted in
Rehoboth Beach, Delaware, has been completed, and the project is now in the
analysis phase.
k-
DESIGN EXPRESSION^
The inactivation of bacteria by UV radiation ,can be approximated by the
first order expression (Figure 5):
N - N0e -
490
-------
where:
N = density after exposure (organisms/100 ml)
N0 = initial density (organisms/100 ml)
k - inactivation rate (cm2//iWatts-sec)
I = intensity at. 253.7nm (/iWatts/cm2)
t = exposure time (second) . »
This is ideal. However, under actual conditions there will be a
distribution of 'exposure times about the mean, the intensity cannot be defined
by direct measure, and particulates will set a base^ level for the effluent
density. The modified model incorporates these design elements (Figure 6):
N =
i -
u
N
where:
x = the characteristic length of the reactor, defined as the average
distance traveled under direct UV exposure (cm)
u = the liquid velocity (cm/sec). This is calculated as
x/(Vv/Q)
where Vv is the void, or liquid volume in the reactor (L) and Q is the
flow rate (Lps)
E = the dispersion coefficient (cm^/sec), which quantifies the breadth of
the residence time distribution about the mean time
K = the inactivation rate (sec'l)
Np = the bacterial density associated with the particulates and unaffected
by exposure to UV (org/100 mL
491
-------
RESIDENCE TIME DISTRIBUTION
The dispersion coefficient is a key parameter defining the hydraulic
behavior of the reactor, providing a measure of the spread of time about the
mean residence time. It is estimated from a residence time distribution (RTD)
analysis of a reactor, such as shown on part (a) of Figure 6. E is estimated
by the expression:
uxcr
E -
26'
where:
a2 - the variance of the RTD (sec2)
u — reactor velocity (cm/sec)
x — characteristic length (cm)
'».
S — mean residence time (sec)
This assumes that the RTD determined for a reactor approximates a normal
distribution. Under ideal plug flow conditions, E approaches zero; with
increased mixing, E will approach infinity, which defines ideal complete mix.
Generally, an E less than 100 cm2/sec would be considered acceptable for
effective design.
Figure 7 gives an example of a hydraulic analysis of a-UV system in West
Virginia. It is an open-channel, horizontal lamp unit, which originally had
over and under baffles to break the velocity of the piped inlet to the channel.
An RTD analysis of this channel showed very high dispersion, with E estimated
at greater than 2,000 cm2/sec. The ratio of tgg to t]_Q, which is the time for
90 and 10 percent of the tracer to pass, respectively, ranged between 2.2 and
6.4. Known as the Merrill Dispersion Index, this will also indicate
significant dispersion at values greater than 2.0. Mixing was occurring in the
reactor, preventing the unit from achieving expected disinfection efficiencies.
492
-------
Removing the baffles provided some improvement, but not sufficient. The
higher velocities from the inlet pipe caused a significant degree of dispersion
to occur in the reactor. The E was estimated to be between 800 and 3,500
cm2/sec, with a Merrill Index of 1.7 to 2.7.
A stilling plate was installed in place of the baffles. This is a
uniformly perforated plate that is able to equalize the velocities entering the
reactor lamp battery, encouraging plug flow. The E was reduced to less than
100 cm2/sec in this case, indicating good plug flow conditions.
Thus, it is important to design a UV disinfection system for plug flow.
.This can be quantified by the dispersion coefficient, which has also been
incorporated into the design expression. RTD analysis should be used to
determine E (and other hydraulic indices), both from the design standpoint and
to troubleshoot performance problems. . . •
*^,
INACTIVATION RATE AND REACTOR INTENSITY
Returning to the design .expression (Figure 6), the inactivation rate
coefficient, K, is described as a function of the average intensity in the UV
reactor:
K
where Iavg is
average reactor intensity (/iWatts/cm2) , and a and b are
coefficients developed from a Log-Log regression of K and Iavg. These are
specific to wastewater conditions, reflecting the sensitivity of the organisms
to UV.
There is no method to directly measure the actual intensity within a
complex, multi-lamp reactor. Rather, a computational scheme (point source
summation method) is employed, based on the number of lamps in a reactor (UV
density) and the absorptive properties of the liquid (UV absorbance
coefficient) . Solutions have been developed and are presented in the EPA
Design Manual for a number of system configurations.
493
-------
Figure 8 shows such an analysis for the symmetrical array of lamps employed
by the modular, open-channel, horizontal lamp systems. Thus for units having a
centerline spacing of 6.35 cm (typical), the UV density, D, is approximately
5,0 UV Watts/Liter; at an absorbance coefficient of 0.4 cm"! (representative of
a "secondary" effluent), the nominal average intensity is approximately 15,000
For design purposes this nominal intensity should be adjusted to reflect
deterioration of lamp output with time and fouling of the quartz surfaces.
Both conditions will obviously attenuate the UV intensity. Typically, the lamp
UV output will be at 70 percent of the nominal rating during the latter half of
its operating life. The quartz surfaces will get dirty, generally from
inorganic deposits such as iron, and calcium and magnesium carbonates. Routine
maintenance of the quartz surfaces (which is a critical O&M task) can typically
keep the transmissibility of the quartz at- an average 8,0 percent of nominal.
Thus, the design intensity can be 50 to 60 percent (70 percent x 80 percent) of
the nominal reactor intensity.
Developing a rate coefficient for a specific application requires
influent/effluent bacterial sampling over a range of intensity levels. This is
typically done by operating two UV pilot units, each with differing average
intensities. Figure 9 shows an analysis conducted at the Port Richmond WPCP in
New York City. (2) In this case,
K
1.45 x 10-5 (Iavg)1.3
This procedure was demonstrated at four additional plants as a special EPA
sponsored project in 1985. The results are summarized in reduced form on
Figure 10. There was sufficient variability to suggest that this type of
information may best be developed for a specific site (particularly larger
plants), although using the combined expression may be adequate for estimating
purposes. The combined data, shown by the heavy dark line, yielded an
expression
K - 1.38 x 10-5 (Iayg)l-28
494
-------
EFFECT OF WASTEWATER PARTIGULATES
Bacteria that are occluded in particulates will not be effected by UV
radiation, and, as such, will comprise the base level of effluent density that
can be achieved for a specific application. The term Np in the design
expression quantifies this effect (Figure 6) by relating it to the effluent
suspended solids:
Np = c SSm
where SS is the suspended solids (mg/L) and c and m are1 coefficients derived
from a Log-Ldg regression of effluent bacterial density and suspended solids.
3 • '
These data must be generated under very high dose levels, such that the
residual density can be attributed to those bacteria which were occluded from
the radiation. An example is shown on Figure 11 for data developed during the
Port Richmond project.(2) The log effluent fecal coliform density is plotted
against the log effluent suspended solids. A linear regression analysis yields
the expression (when transformed):
Np = 0.26 SSl-96
Studies conducted at several plants show considerable variability, although it
appears that these coefficients provide a reasonable estimate of Np.
Calibrated Disinfection Model
The design expression takes on the following format, incorporating the
coefficients required for model calibration (Figure 12):
N = N exp
ux
2E
4E(aI
I ) -,
avg ,1
^ - > ' }\ +
c SS
m
The coefficients a, b, c and m are specific to a given wastewater
application. These can be determined by pilot testing; alternatively, one may
495
-------
use approximations determined from e'arlier studies, recognizing that they may
not be fully representative of actual conditions. In this case, the suggested
coefficients are:
a - 1.38 x ID'5
b - 1.28 ,
c - 0.26 (fecal coliform); 0.9 (total coliform)
m - 1.96
To confirm the model calibration, measured effluent densities can be
compared to the residual predicted by the model. An example is given on Figure
13, developed from an evaluation of the Northfield, Minnesota plant(5), showing
the observed versus predicted fecal coliform densities,. The calibrated model
was subsequently used to determine equipment requirements to upgrade the
system.
SYSTEM SIZING USING THE DESIGN MODEL
The design model is a very effective tool in the sizing of a full-scale
facility and assessing alternative configurations. This is especially the case
when the model has been calibrated by direct testing.
Figure 14 is an example of design curves developed from the model for a
wastewater treatment plant. The figure gives the log N'/No as a function of
the reactor loading, which is defined as the flow (Lpm) per unit UV watt (W).
The sizing assumes using a horizontal lamp, open-channel system, with a 6.35 cm
centerline spacing, and 1.5 m lamps. The solutions are for two banks of lamps
in series and a dispersion coefficient (E) of 100 cm2/sec. The curves
represent different inactivation rates that are defined by varying levels of
the UV absorbance coefficient in the wastewater (recall that these affect the
UV intensity in the reactor).
Consider sizing for the maximum six-hour design condition:
496
-------
Q
TSS
NO
UV Absorbance Coefficient
Design Effluent N
= 42 mgd (160,000 m3/d)
= 30 mg/L
=^600,000 FC/100 mL.'
= 0.55 cm-1
= 800 FC/100 mL
The inactivation rate, K, at this condition was estimated to be 1.0 sec'l.
The .performance requirement (Log N'/N0) is determined by first accounting for
the particulate coliform density,- Np:
Np = 0.26-(30 mg/L)L96 = 2Q4 FC/100 mL
N' = N - Np = 800 - 204 - 596 FC/100BmL
N ' .
Log — = .3.03
o
From Figure 14, the loading is estimated to be 2.0 Lpm/UV Watt. The number
of lamps is computed on the basis of the design flow and a lamp specification
of 26.7 UV Watts/Lamp. This yields 2,070 lamps for the maximum six-hour design
condition. ~ .
With two units in series,'the lamp requirement is 1,035/unit. At 8 lamps
per module, approximately 140 modules are required per unit. This would be
divided to parallel channels to•provide flexibility and control of operations
under varying conditions. A simplified layout is given on Figure 15. This
shows 5 channels, each with 2 banks in series and 26 eight-lamp modules per,
bank. A sixth channel is provided as standby. Each channel is hydraulically
independent, with liquid level control gates at the effluent end of the
channel. The control panels and wireways are generally located near each bank.
SUMMARY
In summary, UV disinfection is an- established technology that is rapidly
gaining acceptance -and is being put into operation at a growing number of
497
-------
plants. The technology configuration that represents current state-of-the-art
is the open-channel system with modular, horizontally% placed lamp banks.
Design protocols have been developed that address key process elements,.
including the dimensional configuration of .the unit, UV intensity, reactor
dispersion and RTD characteristics, and key water quality parameters. The
modeling approach is generic and can be used to characterize most reactor
configurations and wastewater conditions.
REFERENCES
1. Oda, A. Ultraviolet Disinfection of Potable Water Supplies. Ontario Water
Resources Commission, Division of Research, Paper 2012, 1969.
2. Scheible, O.K., M.C. Casey and A. Forndran. Ultraviolet Disinfection of
Wastewaters from Secondary Effluent and Combined Sewer Overflows. EPA-
600/2-86/005, NTIS No. PB86-145182, USEPA, Cincinnati, Ohio, 1986.
3. naatpn Manual: Municipal Wast--ewater Disinfection. USEPA, Office of
Research and Development, Water Engineering Research Laboratory,
EPA/625/1-86/021, Cincinnati, Ohio, October ,1986.
4. Scheible, O.K. "Development of a Rationally Based Design Protocol for the
Ultraviolet Light Disinfection Process," Journal WPCF, Volume 59, No. 1,
j>p 25-31, January 1987.
5. HydroQual, Inc. Report to the City of Northfield, Northfield, Minnesota,
"Evaluation of the UV Disinfection Process at the Northfield Water
Pollution Control Plant," Mahwah, New Jersey, December 1984.
498
-------
2
-J
*,
X
2
Z500
WAVELENGTH (ANGSTROMS - A )
FIGURE 1. RELATIVE GERMICIDAL EFFECTIVENESS AS A FUNCTION
OF WAVELENGTH, (1) .
499
-------
0
-1
-2
-3
-4
-5
-6
-7
1
Uo (Light)
to
Total Coliform , -
Unitl
o to
•t.o
_L
_L
_L
JL
Unit 2
_L
,1 2 3 4 5
Flow/UV Output (LPM/W«ts)
12345
Flow/UV Output (LPM/Watti)
FIGURE 2. PHOTOREACTIVATION EFFECTS FOR TOTAL AND FECAL
COLIFORM AT PORT RICHMOND (2).
500
-------
System UV 2000
UV Modules in
effluent channel.
FIGURE 3. SCHEMATIC OF OPEN-CHANNEL, MODULAR UV SYSTEM
(COURTESY OF TROJAN ENVIRONMENT, INC.,
LONDON, ONTARIO, CANADA),.'
.501
-------
1984
OPERATING PLANTS
• 80% WITH Q < 1.0 MGD
• 35% TEFLON UNITS
• 49% CLOSED SHELL REACTORS
1988
ESTIMATE OF 250 TO 300 OPERATING PLANTS
OF 177 PLANTS SURVEYED
• 59% < 1 MGD " •
• 38% 1 TO 10 MGD
. 3% > 10 MGD
TYPE OF SYSTEMS
17% TEFLON
26% CLOSED VESSEL
• 4.5% OPEN CHANNEL (FIXED)
• 52.5% OPEN CHANNEL (MODULAR)
FIGURE 4. STATUS OF UV INSTALLATIONS
502
-------
N
N =
NO =
k =
T ss:
t =
It =
effluent density (organisms/100 mL)
initial density (organisms/100 mL)
inactivation rate (cm2//iWatts-sec)
intensity at 253.7 nm (/iWatts/cm2)
exposure time (sec)
Dose = /tWatt-sec/cm2
FIGURE 5. UV DISINFECTION EQUATION
503
-------
N= NQe
dispersion
K
K= a I
avg
•avg
Eff.
Coliform
Np= CSS
m
Eff. SS
FIGURE 6. CONCEPTUALIZATION OF UV DESIGN MODEL
504
-------
Baffle Plates
•Stilling Plate
(156- 1" 2000
800-3500 20-90
WlO
2.2-6.4
•1.7-2.7
1.5-2.2
FIGURE 7. WILLIAMSON, WEST VIRGINIA HYDRAULIC ANALYSIS
505
-------
40,000
36.OOO
30,000
23,000
20,000 -
Jg 15,000
™^i
I 10,
000
5OOO
\
ABSORBANCE
COEFFICIENT
UNIFORM LAMP ARRAY
ASSUMES 1OO% LAMP OUTPUT
10O% TRANSMITTANCE
QUARTZ O.D = 2.3CM
_L
O
6 •*-
O
2
O
4 2
CO
Ul
2
3
3 1C
IU
Z
IU
O
8 12 16 20
UV DENSITY (WATTS/LITER)
24
FIGURE 8. NOMINAL INTENSITY SOLUTIONS FOR SYMMETRICAL, UNIFORM LAMP ARRAY.
506
-------
20.0,
10.0
8.0
6.0
5.0
«
3.0
S
S.
i 1.0
tt 0.8
0.6
0.5
0.4
0.3
0.2
° Unit 1
• Unit 2
600 1000 2000 4000 10000 20000 60000
Average Intensity, l.,0 (fj W/cm2)
FIGURE 9. AN EXAMPLE FOR DERIVING AN ESTIMATE OF THE
INACTIVATION RATE FOR FECAL COLIFORMS AS A
FUNCTION OF THE CALCULATED AVERAGE INTENSITY
(2). '•....
507
-------
S
5
CC
6.0
4.0
3.0
2.0
1.5
1.0
0.8
0.6
0.4
0.3
0.2
0.15
1500 2000 3000 4000 6000 10OOO
Average Intensity, !.„, (// W/cmz)
FIGURE 10. COMPARISON OF INACTIVATION RATE ESTIMATES
FROM SEVERAL WASTEWATER TREATMENT PLANTS.
508
-------
I 7
1 •
CO
i 5
e
£ 4
3
o 2
CD
I 0
LLJ
? -1
-2
0.5 1 2 3 4 6 8 10 20304060 100200300
Effluent Suspended Solids (mg/I)
FIGURE 11. EXAMPLE OF DERIVING AN ESTIMATE OF THE RESIDUAL
FECAL COLIFORM DENSITY ASSOCIATED WITH PARTIC-
ULATES AS A FUNCTION OF SUSPENDED SOLIDS (2).
509
-------
N = NQexp|^:
4E(aIaVS } ,1/2\1 . r __m
+ -jj ) |J + c SS
FIRST APPROXIMATION:
a = 1.38 x ID'5
b - 1.28
c = 0.26 (fecal coliform)
=0.9 (total coliform)
m - 1.96
FIGURE 12.' DESIGN EXPRESSION INCORPORATING THE
COEFFICIENTS REQUIRED FOR CALIBRATION
510
-------
4.
Mean Observed = 140
Mean Calculated - 140
0.00
1-20 1.80 3.HJ 3.00 3.60
LOG L CflLC (F. COLI)
FIGURE 13. EXAMPLE OF COMPARISON OF.OBSERVED DENSITIES WITH DENSITIES
PREDICTED BY MODEL (NORTHFIELD, MN).
511
-------
(ON/.N)
512
-------
Overflow weir with
automatic slide gate-
Influent Channel
o
o
/-Stilling
v plate
I I
J I
Lamp
banks
O
4=4
o
Control
panels—?
J I
^-\A/i
Wireways
I I
/—Level control
gates
Effluent Channel
I
I
EIGURE 15. SIMPLIFIED LAYOUT FOR DESIGN EXAMPLE
513
-------
-------
A PRELIMINARY ASSESSMENT OF
HIGH BIOMASS SYSTEMS
by
Arthur J. Condren, Ph.D.
James M. Montgomery, Consulting Engineers, Inc.
Pasadena, California, USA
Bjorn Rusten, Ph.D.
Aquateam Norwegian Water Technology Centre AS
Oslo, Norway
James A. Heidman, Ph.D.
Risk Reduction Engineering.Laboratory
U.S. Environmental Protection Agenc
Cincinnati, Ohio
This paper has been reviewed in accordance with
the U.S. Environmental Protection Agency's peer
and administrative review policies and approved
for presentation and publication.
Prepared for Presentation at:
Twelfth United States/Japan Conference
on Sewage Treatment Technology
Cincinnati, Ohio
October 12-13, 1989
515
-------
INTRODUCTION
The Federal grants program for the construction of municipal
wastewater treatment plants will end on 30 September 1989. With
its termination, municipalities will be seeking plant modifications
which will allow for increased capacity of their existing
facilities with minimal capital outlay. To assist municipalities
in their search for cost-effective solutions to their plant
capacity dilemma, the U.S. Environmental Protection Agency (U.S.
EPA) has undertaken a search for new technologies which may allow
for the realization of this goal. One such technology thus far
identified is what has been termed high biomass systems.
High biomass systems allow for the growth of both fixed film and
suspended biomass in the aeration tanks of activated sludge
systems. The suspended growth concentration is controlled by
adjusting the amount of MLSS recycled in the underflow from the
secondary clarifier. Inert support media is placed in the aeration
tanks and serves as the locus for fixed biomass growth. This_fixed
film growth increases the concentration of biomass in an activated
sludge plant's aeration tanks which, in turn, lowers the food to
mass ratio (F/M) and increases the solids residence time (SRT) of
the system. It also lowers the concentration of biomass being
conveyed to secondary clarifiers since the fixed film biomass is
retained in the system's aeration tanks. Such retention allows for
enhanced levels of solids-liquid separation, especially in
hydraulically overloaded clarifiers. „ In addition to the above, the
presence of the fixed film biomass has been shown to allow for
enhanced wastewater nitrification as well as to yield a better
settling secondary sludge.
CURRENTLY AVAILABLE HIGH BIOMASS SYSTEMS ,
At the present time, there appear to be at least six (6) high
biomass systems commercially available which can be incorporated
into conventional aeration tanks. Linde AG of the Federal Republic
of'Germany (FRG) and Simon-Hartley of Great Britain (Ashbrook-
Simon-Hartley in the USA) both use small, highly reticulated sponge
pads as their inert support media. Bio-2-Sludge (FRG) and Smith
& Loveless, Inc. (USA) use racks of synthetic trickling filter
media to -effect fixed film growth. Ring Lace (Japan) employs a
looped string material as the inert support media and a Chinese
firm uses tassels of a synthetic material attached to a string for
fixed biomass attachment and growth. The latter two inert support
media are tied to racks which are placed in the plant's aeration
tanks.
516
-------
SYSTEM DESCRIPTIONS •
Linde AG/Ashbrook-Simon-Hartley
The basic concept of systems offered by these two manufacturers is
to use highly reticulated sponges as the inert support media for
the fixed film biomass growth. The systems are marketed as Linpor
(Linde AG) and as Captor (Simon-Hartley). Linpor sponges are
approximately cubical with sides of about 10 to 12 mm whereas
Captor sponges are about 12x25x25 mm. The Linpor system includes
solids recycle from the secondary clarifier whereas a Captor system
normally does not include secondary solids return to the aeration
basin. Volume of sponge in Linpor secondary systems approximates
10-30 percent of the aeration tank volume. Biomass accumulation
in the sponges may range from 9-37 g/L (of sponge) , thus accounting
for 25-60 percent of the total biomass in Linpor secondary system
aeration tanks. Measurements of the fixed film biomass in Linpor
systems have indicated equivalent mixed liquor suspended solids
(MLSS-) concentrations of 1,200-3,800 mg/L. Linde AG's cost
estimate for the conversion of an existing conventional activated
sludge plant to a high biomass system is in the range of
$250-350/m of aeration tank volume.
The inert support media in the Linpor system are kept in the
aeration tanks by means of retention screens located approximately
0.3 m upstream of the effluent launder as shown in Figure 1.
Openings in the retention screen are 8-mm holes placed on 12-mm
centers. Forward flow of wastewater through the aeration tank can
cause a maldistribution of sponges in the tank, .with higher
concentrations occurring at the retention screen than at the
influent end of the tank. To normalize the concentration, Linde
AG has installed air lift pumps to convey sponges from the
discharge to the influent end of aeration tanks at installations
where concentration maldistributions have occurred.
Ashbrook-Simon-Hartley has developed a process to squeeze fixed
biomass from sponges in their systems. An air-lift pump lifts the
pads to a conveyor system where they are transported through
pressure rollers. This has been viewed as a means of removing
excess biomass from the pads and producing concentrated waste
secondary sludge solids. The iinde AG system relies on naturally
occurring sloughing of fixed biomass from the sponges and wasting
excess biomass from the return activated sludge (RAS) system.
Ring Lace
The basic concept of a Ring Lace system is the installation of a
poiyvinyl chloridene (PVCE) string on racks in an aeration tank as
shown in Figure 2. Clear spaces above and ' below the racks
approximate 0.5 m each to allow for normal side roll mixing in the
aeration tanks. Each string has numerous attached loops of the
517
-------
Effluent
,,,,."77! v_ '
-------
Effluent *
(
c
c
c
L
L
(
<
(
r
o
c
c
(
<
H
c
-N
Influent
\B) Ring Lace Rack
"Figure 2. Basic Layout of a Ring Lace System.
519
-------
same material, thus greatly increasing the surface area available
for fixed film growth. The European distributor indicated that
typically 25-50 percent of an aeration tank's total volume is
occupied by the Ring Lace racks, and the racks are spaced to yield
a density in this volume of 120-300 m of Ring Lace/m . One
distributor in the US indicated that the racks would typically
occupy from 50 to 80 percent of the basin volume with a density of
about 120 linear meters per mj of the entire basin volume.
Retrofitting a conventional activated sludge plant in Olching
Germany to a Ring Lace system was at a cost of about $2.65/m of
Ring Lace material installed.
Fixed film growth on the inert media has been reported to be 6-7
g/m and thus the equivalent MLSS of fixed biomass in a Ring Lace
system could vary from roughly 200-1,100 mg/1 of total tank volume.
As the fixed film biomass continues to. grow, natural sloughing
occurs. The sloughed material then becomes a part of the MLSS in
the system and is removed by conventional wasting practices. The
level of suspended MLSS is controlled by the amount of solids
recycled from the secondary clarifiers.
Bio-2-Sludge/Smith & Loveless, Inc.
The basic concept of these systems is the inclusion of synthetic
trickling filter media to provide the surface on which fixed film
biomass growth can occur as shown in Figure 3. Volume of media
used in Bio-2-Sludge systems approximates 25 percent of the
aeration tank volume while in Smith & Loveless systems, it is close
to 75 percent., In contrast to the Bio-2-Sludge systems, the Smith
and Loveless system does not recycle settled sludge from the
secondary clarifier back to the aeration tank.
Bio-2-Sludge systems use just about any synthetic media as long as
the openings in the media are at least 2x2 cm. Typical surface
area of the media is in the range of 90-120 itr/m • Racks to hold
the media are constructed to provide approximately 0.5 m of clear
space above the air diffusion system as well as about 0.5 m of
clear space between the top of the racks and the liquid surface in
the aeration tanks. This arrangement allows for normal side roll-
mixing in the tank. If the media begins to plug from excess fixed
film growth, a common practice is to turn on an additional blower
to induce surplus biomass sloughing. Retrofit costs for
Bio-2-Sludge systems have been" reported to be in the vicinity of
$300-350/m3 of media installed.
Chinese system
As mentioned above the Chinese system uses tassels of a synthetic
material which are tied to strings and then mounted in the aeration
tanks. At the present time, there is no additional information
available on this system.
520
-------
Effluent
5
Influent
© TF Media Rack
Figure 3. Basic Layout of a Bio-2-Sludge System.
521
-------
SITE VISITS
During the late spring of 1988, a U.S. EPA evaluation team visited
a number of full scale high biomass systems in the Federal Republic
of Sany. The purpose of the site visits was to view the systems
in opS-ation, collect operational and performance data, learn of
system design details from the system manufacturers and discuss
process operational and maintenance concerns with the treatment
plant staffs.
Facility locations and types visited included, among others, those
presented in Table 1.
Table 1. Site Visit Basic Information.
Type of System
Linde AG
Ring Lace
Bio-2-Sludge
Location .
Freising
Munich
Olching
Schomberg
Calw/Hirsau
Summarized below are data and information on the various types of
high biomass systems from certain of the above facilities.
Freising (Linde AG System)
The Freising activated sludge wastewater treatment plant was
converted to a Linde AG high biomass system in 1984 after a series
of pilot plant studies. It appears that there were three primary.
reasons for the conversion: frequently occurring poor sludge
settleability, limited space at the plant site and higher costs of
alternative technologies. Operational and performance data were
collected for a short period of time before and after conversion
and these are summarized in Table 2.
Before conversion to the high biomass system, which employed a
sponge volume equal to 20 percent of the aeration tank volume, the
plant could only maintain a MLSS concentration of about 2,600 mg/L.
Following conversion, a much higher MLSS concentration could be
maintained which dramatically lowered the F/M of the system and
greatly improved the secondary sludge settleability.
As a follow-up to this historic information, data for calendar year
1987 were collected and analyzed. A summary of this information
is presented in Table 3.
522
-------
Table 2. Historic Freising Data.
Parameter
Solids, g/L
Suspended
Fixed
Total
F/M,
kg/kg-d
SVI ,
ml/g
Effluent BOD5,
mg/L
Conversi
Before
2.6
2.6
0.47
485
48
.on
Afte>r
£±^ l»> v? 4.
,5.1
1.7
6.8
0.17
85
4
Table 3. Freising Data Averages for 1987.
Parameter
.Secondary Influent
Flow, m3/d
.BOD5, mg/L
Secondary Effluent
BOD5, mg/L
Biomass, g/L
Suspended
Fixed
Total
Aeration Tank
Det'n Time, hr
F/M, kg/kg-d
SVI/Aml/g
D.O. , mg/L 1.7
Aeration Rate
m3/kg BODS Removed
Secondary Clarification
SOR , m/hr
R/Q ; %
Value
10,100 •
273
2.59
3.75
6.34
5.5
0.19
114
64
0.34
41
Oxygen
Overflow Rate
Ratio of return/influent flow
thS Fr®isin9 Plant was operated at. 78 percent of its
r capacity and 80 percent of its BOD5 capacity?
Instantaneous influent pH, because of industrial discharges to th4
523
-------
system, ranged from 6.2- 12.0; effluent pH ranged from 6.8-7.5.
An average of 65 percent nitrification was also achieved over a
wastewater temperature range of 10-17 C, even though the D.O.
concentration averaged only 1.7 mg/L.
Munich (Linde AG System)
The Munich Grosslappen plant was retrofitted with the Linde AG high
biomass system to allow for additional treatment whale a parallel
activated sludge plant of equal capacity is being constructed.
Sponge media were also installed to allow for full scale evaluation
of the process for possible inclusion in the new treatment plant.
The existing Munich plant has three banks of aeration tanks and
during the site visit, two of the banks had varying quantities of
sponges in place. Data from the plant are summarized in Table 4.
Table 4. Munich Data for 1987.
Parameter
Sponge Volume in Tanks. %
0 10 .30
Secondary Influent
Flow, mj/d
BODS, mg/L
Secondary Effluent
BOD5, mg/L
TSS, mg/L
Biomass, g/L
Suspended
Fixed
Total
Aeration Tank
Det'n Time, hr
F/M, kg/kg-d
kg BOD5/nr-d
Sludge Age, days
SVI, ml/g
D.O., mg/L
Aeration Rate
m^ Air/itr WW
m /kg BODS Removed
Secondary Clarification
SOR, m/hr
SLR , kg/nr-hr
R/Q, %
RAS, g/L
Waste Activated Sludge
kg/kg BOD5 Removed
165,100
186
21
20
2.81
2.81
1.91
0.84
2.39
1.3
145
2.1
5.4
33
1.08
3.05
76
6.24
1.08
165,060
183
18
16
2.90
•1.20
4.10
1.90
0.57
2.34
2.0
95
2.2
6.1
37
1.08
3.15
73
6.86
1.02
145,800
183
14
13
2.77
1.84
4.60
2.15
0.46
2.11
2.3
90
2.2
8.5
50
0.95
2.82
49
6.94
1.03
'Surface Loading Rate
524
-------
Of interest to note with the Munich data are: 1) the lowering of
the system's F/M by the presence of the fixed film biomass, and 2)
the required increase in aeration rate to address the demand of
additional biomass in the system.
Olching (Ring Lace System)
During the spring of 1988, the Olching plant was operating at 85
percent of its design hydraulic capacity of 35,000 m /day and 83
percent of its design organic loading which was based on a
population equivalent of 240,000. In late 1987, conversion of the
plant to a high biomass system began by adding a new 4,000 rir
denitrification basin and also by adding 252,000 m of Ring Lace
material to each of four 2,070 m3 aeration tanks. The lead
denitrification basin, which contains no Ring Lace and had not yet
been placed in operation at the time of the site visit, is equipped
with paddle mixers and also contains aeration equipment for
nitrification if necessary. A portion of the mixed liquor from the
Ring Lace basins will eventually be recycled to the lead
denitrification basin. At the time of the site visit, influent
flow was sent directly to the aeration basins containing the Ring
Lace. • • . '
The Ring Lace material was strung on racks, with the individual
strands being separated by approximately 50 mm. Spacing between
the racks, which occupied 31 percent of the aeration tank volume,
was about 60 mm. A tensioning system was built into the racks in
case the Ring Lace material elongates with time. In addition to
the above, the equipment supplier had to provide a 10-year process
performance-and equipment guarantee. Required process performance
was based on the plant's discharge requirements of 15 mg/L BOD5,
20 mg/L TSS .and 10 mg/L NH4-N, the latter equating to an
approximate 75 percent level of nitrification.
Because the Olching Ring Lace plant had been in operation for less
.than one year and the required performance test had not been
completed, no operation or performance data were released to the
U.S. EPA . evaluation team. However, the. following general
observations on the Ring Lace system were communicated from the
treatment plant staff.
o Before conversion to the High biomass system, the maximum
operational MLSS concentration that could be achieved was
about 1,500 mg/L, which resulted in the plant operating at
a F/M of 0.6-0.70 kg BOD5/kg MLSS-day. At this loading
rate the MLSS were gray in color and had very poor settling
characteristics. In addition, no substantial nitrification
could be achieved.
o After conversion, the suspended biomass varied from 3,500
to 4,500 mg/1. The fixed biomass on the Ring Lace material
was estimated at 6.5 g/m which is equivalent to an
.additional 790 mg/1 of equivalent basin MLSS. This
525
-------
increase resulted in a MLSS with a rich browi? color
that had greatly improved settling characteristics.
Operating F/M of the system decreased to about-0.2 kg/kg-
day and desire'd levels of nitrification were being
achieved.
o Other high biomass systems were evaluated at the pilot
plant level by the Olching treatment plant staff and they
all yielded effluent qualities equivalent to that realized
by the Ring Lace system. However, the Linde AG system was
not selected because the staff felt that this system might
require additional electrical power for proper operation,
the sponge cubes would be subject to wear by abrasion and,
upon extended levels of mineralization, the sponge cubes
might have a tendency to settle in the aeration tanks. The
Bio-2-Sludge process was not selected because the plant
staff felt that the synthetic trickling .filter media might
be subject to plugging by biomass and/or large
particulates, either of which might lead to anaerobic zones
in the media. Such zones, if they developed, were felt to
potentially limit the nitrification process.
Schomberg (Bio-2-Sludge' System)
The Schomberg wastewater treatment plant was recently expanded in
throughput capacity and simultaneously converted to a Bio-2-Sludge
high biomass system. Both aeration /tank volume and final clarifier
surface area were more than doubled and a denitrification section
was added to the aeration tankage. In the spring of 1988, 'the
plant was operating at 20 percent of its design hydraulic capacity
and 57 percent of its design BOD5 capacity.
Prior to conversion, , the plant was neither hydraulically nor
organically overloaded, but could not meet discharge permit
requirements primarily because of very poor settling MLSS. The
poor settling characteristics would allow the maintenance of only
1,000-1,500 mg/L in the system's aeration tanks in spite of the
very high RAS pumping rate employed. After conversion, overall
plant performance greatly increased as indicated in Table 5.
Installation of the Bio-2-Sludge system at Schomberg along with the
other plant modifications greatly enhanced MLSS settleability as
evidenced by the SVI of 82 ml/g after conversion. Although the RAS
rate at this plant could have been adjusted' to about 50 percent,
a return rate of 100 percent was used. It appears that this is a
common design/operational practice at smaller treatment plants in
certain areas of Germany. Of interest to note is that this plant,
in addition to achieving an effluent containing 5 mg/L BODS,
realized nearly complete nitrification.
. 526
-------
Table 5. Schomberg Data Averages for 1987
Parameter
Secondary Influent
Flow, m /day
BODS, mg/L
NH4-N, mg/L (Approximate)
Secondary Effluent
BOD5, mg/L
NH4-N, mg/L
Biomass, g/L
Suspended
Fixed (Estimated)
Total
Aeration Tank
Detention Time, hr
F/M, kg/kg-d
Kg BOD5/m3-d
SVI, ml/g
D.'O., mg/L
Secondary Clarification
SOR, m/hr
SLR, kg/m -hr
R/Q, %
Value
3,161
120
40
5
0.4
3.96
1.74
5.70
6.7
0.08
0.45
82,-
3.2
0.18
1.42
100
SUMMARY AND CONCLUSIONS
A number of high biomass systems have been installed at wastewater
treatment plants throughout the Federal Republic of. Germany. These
installations were designed to effect improved effluent quality and
it appears that-this goal is being realized.
*
Reasons for selecting high biomass systems over construction of
additional aeration tanks and clarifiers (or other secondary
treatment processes) include reduced space requirements, increased
process stability and capital/operating cost savings.
High biomass systems call for installation of supplemental
equipment over that contained 'in a conventional activated sludge
plant. More .installed equipment generally implies more maintenance
and, to a .certain extent, this occurs with some of the high biomass
systems; In addition, the presence of both suspended and fixed
biomass forms and higher biomass concentrations may require a
certain level of additional operator time to achieve optimum system
performance.
527
-------
The presence of inert support media and higher biomass
concentrations in these systems can increase overall power
consumption. To achieve desired mixing patterns and velocities in
retrofitted aeration tanks, power input may have to be increased
to overcome additional headless brought about by the installation
of the inert support media. Due to endogenous respiration rate
alone the presence of additional biomass will increase system
oxygen requirements which, in turn, will require .additional power
input In addition, high biomass systems generally yield lower
effluent BODS concentrations and/or higher levels of nitrification
which also can effect overall power consumption. Such factors
should be addressed when analyzing high biomass system operating
costs.
Certain system design limitations have been identified in the past,
but many of these have been corrected by the system manufacturers
and/or operators. For example, influent hydraulic surges at Linde
AG plants have caused blinding of the retention screens by the
sponge media. This has been partially corrected by increasing the
pumping rate" of the sponge return system. , Also, abrasion of the
sponge media was stated to result in losses of less than 5 percent
per year, but there does not appear to be a corrective -action for
this system limitation beyond adding new sponges as required. In
Rincr Lace systems, a question on the extent of media stretching
over time persists. Construction of self-tensioning media racks
may address this potential limitation. Plugging of the synthetic
trickling filter media in Bio-2-Sludge systems has been a concern
of certain individuals. It appears that turning on an additional
air blower periodically for a short period of time will induce
additional sloughing of the fixed film biomass, thus negating this
potential problem. There were other potential problems areas
communicated to the U.S. EPA evaluation team, but their relative
importance was deemed minor because solutions had, for the most
part, already been developed and field tested.
Cost of retrofitting a 3,785 m3/day activated sludge plant with
high biomass system equipment appears to be $77,000-260,000,
depending on the process selected. These estimated costs.are_based
on an activated sludge plant with six hours of contact time in the -
system's aeration tanks.
Additional throughput capacity realized by conversion< to a high
biomass system cannot be ascertained from the data and information
thus far amassed. However, it appears conversion can allow for at
least the doubling of the biomass concentration in a system and for
reducing by at least one-third the required HAS pumping rate
compared to a conventional activated sludge plant. These two
factors alone indicate the potential for increased throughput
capacity.
528
-------
Actual capacity increase which can be realized by conversion to a
high biomass system should be established from the conduct of pilot
plant studies. Representatives for both Bio-2-Sludge and Linde
AG stated that to obtain accurate design and operational
information, a pilot plant should have an aeration tank with a
volume of at least 100 m .
.*
ACKNOWLEDGEMENTS
This article was prepared by Arthur J. Condren, PhD of James M
Montgomery, Consulting Engineers, Inc., Pasadena, California USA-
Bjorn Rusten, PhD of Aquateam Norwegian'Water Technology Centre As'
Oslo, Norway; and James A. Heidman, PhD of the U.S. Environmental
Protection Agency, Cincinnati, Ohio, USA. The authors wish to
acknowledge the assistance of the system developers and facility
operators in providing the data and information contained in this
.article. Efforts undertaken on this project were funded wholly or
in part by the U.S. Environmental Protection Agency under Contract
No. 68-03-3429 to James M. Montgomery, Consulting Engineers, Inc
529
-------
-------
EXPERIENCE WITH INTRACHANNEL CLARIFIERS
IN THE U.S.A.
by
James F. Kreissl
Water and Hazardous Waste Treatment Research Division
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency '
Cincinnati, Qhio 45268
Dr. A. T. Wallace
College of Engineering
University of Idaho
•Moscow, Idaho 83843
U.S
Jon H. Bender
Technical Support Division
Office of Drinking Water
Environmental Protection Agency
Cincinnati, Ohio 45268
This -paper has been reviewed in accordance with
the U.S. Environmental Protection Agency's peer
and administrative review policies and approved
for presentation and publication..
Prepared for Presentation at:
Twelfth United States/Japan Conference
on Sewage Treatment Technology
Cincinnati, Ohio
October 12-13, 1989
531
-------
INTRODUCTION
The modern oxidation ditch (OD) which is depicted in
Figure 1, is a high-sludge-age variation of activated sludge or
suspended growth biological treatment system which emanates from
the work of Dr. Pasveer of The Netherlands in 1954. Since the
time of its introduction in 1973 more than 9200 such plants have
been constructed in the U.S., and are widely perceived as the
state-of-the-art treatment in small-to medium- sized communities
(populations/capacities of 5,000 -to 50,000 (0.5 to 5.0 million
gallons per day (MGD)). This status has been attained owing to
the OD's simplicity of operation, reliability of performance and
overall cost-effectiveness.
Intrachannel clarifiers (ICCs) as depicted in Figure 2, were
introduced during the 1980's to replace the conventional external
clarifiers which have been commonly employed for OD facilities in
the U.S. These systems are marketed by a variety of commercial
entities and are of a variety of configurations (1,2), but in
general are physically placed in the ditch and do not employ,
return sludge thickening and pumping. Consequently, T0°
purveyors have claimed:
ICC
1. Lower capital costs owing to elimination of return
activated sludge (RAS) pumping and piping facilities
2. Lower operation and maintenance.(O&M) costs owing to
- the elimination of 'RAS facilities and external
clarifier control problems
' 3. Reduced land area requirements by elimination of
separate aeration and sedimentation structures.
These claimed advantages served the purveyors well during
the early 1980's as an EPA-sponsored study (2) in 1987 revealed
approximately 54 systems in design, construction or operation-, of
which 26 were in the last category. The majority of these
systems were approved under the innovative technology provisions
of the USEPA Construction Grants Program whereby they received an
increased Federal grant with potential guarantees of performance.
In fact, the ICC systems were the second most popular technology
approved under the innovative technology provisions, second only
to ultraviolet light disinfection.
532
-------
The most widely employed version of an ICC is known as a
BMTS'" system. Descriptions of this ICC can be found in
references 1 and 2. For the purpose of this discussion, it is
sufficient to describe the BMTS system as one which has generally
been field constructed integral with the ditch structure, as
shown in Figure 3. The second and almost equally popular .ICC
system is the BOAT™ clarifier. It is described in more detail
in references 1 and 2. The BOAT, as opposed to the BMTS, is
factory fabricated and delivered to the site for installation in
a more conventional OD geometry, and is schematically displayed
in Figure 4. Since these two types of ICC clarifiers constitute
46 of the 54 systems noted in 1987, much of the remainder of this
discussion deals with these two systems as subcategories of ICC
technology.
With any degree of popularity a proportionate degree of
attention is naturally drawn, in this case in the form of
investigative studies and reports. Early literature citations
were generally from the commercial purveyors and already
committed clients and were quite positive in their nature
regarding the costs and perceived benefits of the technology.
Several articles appeared (3,4,5,6) which provided some
performance and comparative cost data, but the circumstances and
assumptions for the data presented were generally undefined. For
example, Parsons (5) described a Florida BOAT system which was
estimated to provide 20 percent capital cost savings over an
external clarifier. Some articles (3,4) published by BMTS
advocates addressed the results of short-term studies which were
subsequently adopted as full-scale design criteria and the basis
for cost savings and other claims which provided the stimulus for
system adoption.
USEPA STUDIES
The USEPA had participated in some of the conceptual and
development studies which led to the widespread adoption of the
ICC concept in its role of assisting new technological ideas.
However, the nature of these developmental.projects merely
identified a few crucial areas of concern, but failed to quantify
the extent of potential problems in full-scale application.
Likewise an early attempt to generically assess the ICC
technology was abandoned owing to difficulties encountered in
providing an unbiased assessment without full-scale information
on systems which approached the ICC concept from significantly
different directions. In a 1984 USEPA Seminar Series an attempt
was^made to assess the status of development of the commercially
available ICC options and to restate generic advantages and
disadvantages of the generic technology (7). The advantages were
cited to be:
533
-------
1. Elimination of separate clarifiers
2. Elimination of sludge return pumps
3. Reduced O&M associated with #1 and #2
4. Elimination of sludge septicity problems
5. Reduced capital and O/M costs.
Disadvantages were listed as:
1. Minimal control over sludge blanket depths and return
sludge flows
2. Less operational flexibility.
3. Difficulty in adding flocculant aids
4. Generally unthickened MLSS wastage
5. Unknown response to high peak flows
6. Unknown control of bulking sludges
7. Unknown scum control effectiveness.
In a 1986 USEPA Seminar Series on ICC technology, an
assessment based 'several short-term site visits was
presented (8). Since most installations had been operational for
only a short period, the findings were presented in a manner
designed to identify potential problems which designers and
operators might minimize through awareness. The most serious
problem identified was the inadequate mixing and aeration
resulting in below critical (1.0 foot/sec.) horizontal velocities
and inadequate dissolved oxygen in the mixed liquor. The latter
problem was correctable, since the -total aeration capacity should
be the same for an ICC systems as it is for a conventional OD.
The other common problem identified was the general undersizing
of sludge handling facilities, e.g., thickeners and drying beds.
Other manufacturer-specific problems noted were a need for better
scum removal for ICCs which completely baffle the ditch water
surface, e.g., BMTS, and the need to improve the design
understanding of propeller mixers which impart horizontal
velocity in systems which separate this function from aeration.
With the advent of significant population of operating ICC
systems a more thorough technology evaluation was initiated in
1987. This evaluation involved the development of a master list
of available information from the operating facilities. From
this list eleven systems were selected for one-day site visits to
collect more extensive information and to assess O&M and
laboratory procedures, competence, etc. The purpose of this
evaluation was to evaluate the true comparative costs of ICC
systems, the design tradeoffs identified by Bender (1), and the
relationships between these factors.
534
-------
RESULTS
Costs
The initial effort on cost comparison dealt with the cost-
effective analyses supplied by the designers/municipalities to
justify their choice of an ICC system over an OD design. In the
Construction Grants regulations a 15 percent cost savings based
on life cycle costs or a 20 percent net energy savings is
generally required to justify designation as a innovative
technology. For a list of eleven projects, comprised of
facilities employing only ICC equipment from the two major ICC
purveyors, capital cost savings of from 4 to 45 percent were
claimed by the engineers. Interpretation of these data was
impossible because of the unequal components. For example, ift
some cases a 10- to 20-hour ICC aeration time was compared to a
24-hour aeration time in the OD, in others no additional cost for
the handling of MLSS vs thickened sludge from a conventional
clarifier was provided, several others used lower aeration/mixing
costs for the ICC than the op despite the fact that the ICC
creates additional headloss in the channel, and one even omitted
preliminary treatment in the ICC.
Even with bid tabulations from 26 projects indexed to a
common date the widespread practice of package bidding and
lumping together of individual contractual subdivisions made
quantitative comparisons impossible. Generally, bid prices
(indexed to the second quarter of 1987) indicated that small
(<0.5 MGD) ICC-based total treatment systems varied between $2.00
and $5.00/gallon of daily capacity, while larger (>2.0 MGD)
systems varied from $1.00 to'$1.SO/gallon. Between 0.5 and 2.0
MGD the scatter reflects the entire spectrum from $1.00 to
$5.00/gallon.
In order to attempt to compare the real capital costs of the
two technologies it was decided to perform quantity takeoffs and
detailed cost estimates from some ICC system designs and replace
those components with conventional OD components using accepted
construction guides supplemented by vendor-installed prices.
This technique was applied to three BMTS plants with capacities
from 0.8 to 1:9 MGD and resulted in the construction costs of
conventional OD being consistently less expensive by 5 to 17
percent. A similar analysis for two BOAT systems (0.25 and
0.88 MGD) reveals an even greater ICC plant cost. However,
comparison of bid tabulations on two recent projects with nominal
plant capacities from 2.0 and 2.4 MGD showed the capital costs of
BOAT systems to be less than a conventional OD by only 3 to 5
percent. Although, there is inconsistency in these data, they do
indicate that the capital cost of properly designed ICC systems
can be,expected to be similar to a conventional OD system.
535
-------
Performance
Performance of operating ICC facilities was compared to
previously analyzed OD facilities to roughly assess whether
significant differences in BOD5 and TSS were apparent. The ICC
data were based on the 17 plants which were apparently operating
at a steady state. The OD data were based on the performance of
29 facilities (8). The'data shown on figures 5 and 6 indicate
that performance is generally comparable, with slightly higher
effluent concentrations likely attributable to the newness of the
ICC technology in terms of developing operational staff
understanding and problems owing to first-generation design
shortcomings in sludge thickening and aeration/mixing systems.
Subsequent analyses of components provide insight into the latter
problems. Although clarifier loadings for both sets of data are
comparable, e.g., actual/design clarifier overflow rates (gal/sq
ft-day) are 357/485 for the OD plants and 357/500 for the ICC
plants, influent wastewater characteristics are not documented.
Therefore, the above comparisons are only qualitative, but lead
one to the conclusion that design overflow rates should be made
at least equal to those of external clarifiers.
Design Tradeoffs
Bender (1) has previously identified the key design
tradeoffs which should be quantified to fully evaluate the ICC
technology against the conventional OD alternative. These key
tradeoffs were identified as:
O Waste sludge concentration - Must the ICC system
provide external thickening to compensate for the
internal thickening 'provided in an external clarifier
to be fully comparable?
O Restriction in aeration channel flow - How much
additional mixing power must bevemployed to compensate
for the increased headless of the ICC placed in the
channel?
O Aeration channel and clarified maintenance - Owing to
the interdependence of the ICC and the aeration
channel, will increased redundancy in design (and cost)
be required to satisfy effluent permit requirements
while performing major O/M tasks?
0 Operational flexibility - Without sludge return
facilities, how.can the O&M staff monitor control and
chemically treat (as necessary) return sludges to
control the activated sludge system?
536
-------
The present study was initiated to provide some field
experience for the purpose of quantifying these issues in
addition to analyzing costs and performance information.
Waste Sludge - - ' "
Early promotional literature (3,4) dismissed this'issue by
asserting that the required surface area of a thickener is
determined only by the mass of solids entering the unit, and was
therefore identical for the MLSS from an ICC system and the
partially thickened (1 to 1.5 percent solids) underflow from an
external clarifier. Although other references (9,10,11)
generally confirm this analysis for determining surface area of
the thickener, they also note limitations on hydraulic loading
and on thickener volume which would indeed equate to additional
construction costs for treatment of significantly more dilute
suspensions. Also, in cases where sludge is removed from the.
plant directly without thickening, additional costs will be
associated with ICC systems owing to increased sludge volume.
This problem, i.e., inadequate sludge thickening capacity,
was noted as a significant performance limitation at 3 of 11
facilities visited and was reported as a significant problem at
several of the other plants contacted by telephone with
subsequent information submission. At one ICC facility visited,
the engineer increased thickener capacity such that with the
added capital cost of thickening the ICC was 67 percent more
costly than for the external clarifier alternative. At the time
of the visit the enlarged thickener was receiving 2,000 to 4,000
mg/L MLSS and yielding an underflow of 2 percent solids.
%
The above information, when combined with the fact that the
average ICC plant-was being loaded at 53 percent of design
organic loading, provides qualitative support for Bender's
concern over the relative thickening requirements for thin MLSS
wasted directly from the ditch or from the ICC with minimal
thickening. Additional analysis of this design issue is
underway.
Channel Headloss -
Of all the issues investigated in the present study, this
one appears to be the most serious due both to the fact that
manufacturers tend to dispute that there is a problem and the
general uncertainties on how to properly address the issue.
Actual channel headlosses have not been adequately characterized.
In systems which employ propeller mixers and diffused air
aeration the wall of rising bubbles also represents an additional
537
-------
headless on which there is little quantitative data. Aeration
energy required for an ICC or conventional OD should be the same
for the same detention time and operating conditions, but the
horizontal-shaft mechanical aeration systems commonly employed,
i.e., surface rotors and disks/ provide both mixing and aeration
in a single mechanical function, making total power requirements
to reach the required operating conditions the only means of
comparison between the two applications. The newer propeller
mixer/diffused aeration designs theoretically permit direct
analysis of each of the two functions.
Six of the ICC plants investigated in this study which
employed horizontal-shaft mechanical aeration had aerator energy
data. The total power required in these six plants varied from
6.2 to 91.6 W/m3 of channel volume. Only one plant was in
steady-state operation with reasonable (75% of design loading)
organic loading, and it required 12.2 W/m3. Ettlich (8) suggests
that 10 to 25 W/m3 is required for ODs to maintain a proper D.O.
(1.0-2.0 mg/L) and V min (1.0 ft/sec), while Hartley (12)
estimates 5 to 10 W/m3 with full baffling. Measurement of total
energy use by 30 O.D. systems in the field yielded a range of 4
to 78 W/m ' which reduced to a range of 6 to 30 W/m3 in the 11
plants which had reasonable organic and hydraulic loading (13).
Koot and Zeper (14) performed a theoretical hydraulic analysis of
OD systems using the Chezy formula and estimated that 5.0 to 7.5
W/m are required to provide 1.0 ft/sec velocity and adequate
aeration. In comparing the above operating ICC facility with
these ranges, no- quantitative determination is possible,
especially since the average channel velocity in this plant has
been measured to be as low as 0.7 to 0.8 ft/sec and the larger
power-use system operated at 82 hours of detention time.
However, the placement of an ICC in the ditch/channel will
undoubtedly increase headless -and, therefore, the power required
to maintain any specified velocity. One promotional report for
the BOAT unit describes studies showing a 15-30% velocity
reduction (15) by their own measurement based on average
velocities in the channel with and without the ICC. Using the
manufacturer's figures and the horizontal-shaft aerator
relationship (12) between power input and average velocity, i.e.,
average velocity varies as the 0.55 power of the power per unit
volume, the increase in power required to overcome a 15 to 30%
average velocity loss is about 25 to 50% of the effective power.
Therefore, to restore average channel velocity to the original
level, approximately 25 to 50% more power must be applied to the
horizontal shaft aeration/mixing system. Since the effects of
these constrictions on oxygen transfer are unknown, this increase
is ascribed only to maintaining required channel velocity..
538
-------
The theoretical analysis of Root and Zeper indicated that
mixing, i.e., maintaining a velocity of 1.0 ft/sec, was
approximately 10 percent (0.3-0.4 W/m ) of the total energy
required for shallow OD systems and about 1 percent (0.1 W/m3)
for deep, short submerged turbine systems, e.g., the Carrousel™
system. Analysis of the operating ICC facilities employing
propeller/mixer these systems yielded only five facilities
operating at reasonable organic loadings (0.3-1.3 of design
loading) with channel detention times between 17 and 29 hours,
i.e, reasonably conventional OD systems. The mixing power demand
for these five facilities varied from 4.6 to 29.6 W/m , or about
10 to almost 100 times the theoretical mixing requirement
computed by Koot and Zeper. The ratio of mixing power to total
power demand varied from 23 to 71 percent, and total power/unit
volume varied from 18 to 54 W/m3.
Even in the early promotional literature-on ICC .systems
inadequate concern over minimum velocities in the presence of
large channel headlosses was apparent. Average channel
velocities were shown to meet the 1.0 ft/sec standard and the
presence of 0.5 ft/sec velocities at a certain point in the
channel was acknowledged and dismissed (4*). Attainment of 1.0
ft/sec velocity in the ditch prior to insertion of an ICC has
also been promoted as a proper test of adequate mixing (6,15).
Field experience at some ICC installations has shown the value of
adjustable mixing systems not only in adapting to different
operating conditions, but also in providing periodic, e.g., once
per week, scouring velocities of at least 1.3 ft/sec to clear
settled s'olids from the bottom of the ditches in ICC systems
which have locations where local velocities are significantly
less than 1.0 ft/sec. Numerous problems from poor distribution
of velocities within the channel have been reported by the
operating ICC facilities.
Clearly, the increased headless in the OD channel due to the
presence of the ICC has substantially"increased the power
requirement for maintaining adequate channel velocities to
prevent biological solids deposition. However, more data will be
needed to fully quantify this effect.' Since the original claim
of energy savings was based on RAS pumping elimination, the above
power increase must be compared against the power saved by
elimination of RAS pumping. Middlebrooks, et al, (16) indicate
that the power required for return and waste activated sludge
pumping and conventional clarifier sludge collection are about 3,
percent of aeration requirements for 1-MGD extended aeration
systems. Also, Wesner, et al (17) indicate that RAS pumping
and extended clarifier mechanisms require approximately 6 percent
of aeration energy requirements. Therefore, energy savings
claimed for ICC systems are apparently non-existent or extremely
minor in nature.
539
-------
Additional O&M Problems
Although several unanticipated O&M problems were discovered
with ICC systems, many were due to design deficiencies. In
addition to those items described above, preliminary treatment
deficiencies were abundantly apparent in most facilities
employing propeller mixers. Early systems employed a »
conventional propeller design which became fouled with rags and
debris and resulted in power demand increases, vibrational
imbalances and frequent removal and cleaning efforts by the O&M
staff. Although replacement with "weedless" blade designs has
reduced this problem, the use of conventional OD preliminary
treatment, e.g., coarse screens and comminutors, has resulted in
scum removal and orifice/baffle opening plugging problems with
certain ICC designs. Improved preliminary treatment with fine
screens with approximately 0.5-inch openings would be beneficial
in reducing high O&M staff requirements for existing systems and
preventing these problems in new facilities. Also, the need to
provide good flow distribution across the channel crossection has
become increasingly apparent. Although problems due to this
inadequacy have been proven soluble, systematic prevention in the
form of design criteria are still forthcoming.
Other design deficiencies in ICC systems which have resulted
in excessive O&M demands include:
1. ICC bottom baffle spacing in BMTS systems - The value
of wider spacing is under investigation at the largest
BMTS system. A frequent problem found at these
facilities has been the buildup of solids on top of the
bottom baffles. This buildup apparently results in an
inordinate percentage of the MLSS being accumulated in
the ICC instead of freely returning to the channel.
These conditions cause an apparent reduction in
effective solids retention time (SRT) in the channel.
and result in deteriorating system performance until
corrected.
2. Effluent orifice pipe design in BMTS - The outlet
orifices of the BMTS system are submerged and can .
result in significant water level changes, e.g., a 9-
inch difference was recorded at one small BMTS plant.
This water level swing can make slotted scum trough
removal systems ineffective and increase O&M labor
requirements. This problem can be solved by designing
auxiliary weirs in the effluent collection box or by
modifying scum collection troughs for a varying water
level.
540
-------
3. Placement and design of propeller mixers and diffusers
- The performance of propeller mixing systems in ICC
systems has been a learning experience for both mixer
suppliers and design engineers. The headlosses
represented by the ICC systems and the banks of
diffusers must be accounted for in determining power
requirements to attain a 1.0 ft/sec minimum ditch
velocity. For systems which vary ditch dimensions such
that velocities vary over a wide range, the deposition
of solids has been controlled through periodic higher
ditch scour velocities.
SUMMARY AND CONCLUSIONS -
ICC^systems appear to be a viable alternative to a
conventional oxidation ditch. Technologically, the ICC concept
can provide secondary treatment if the same overflow rates are
employed as are used in the design of external darifiers. Early
problems with ICC systems appear to be amenable to improved
design procedures which provide better preliminary treatment to
minimize rags and debris, proper mixing to insure a maintenance >
of at least a 1.0 ft/sec minimum velocity, and adequate sludge
handling and thickening capacities. Cost savings are not
inherent with ICC designs, so the relative advantages and
disadvantages of this technology must be thoroughly analyzed for
each potential application.
541
-------
REFERENCES
1.
8,
10,
11.
12,
Bender, J.H. "Assessment of Design Tradeoffs When
Using Intrachannel Clarifiers."
Jour. WPCF, 59, 871-6 (Oct. 1987).
Zirschky, J.H. "Intrachannel Clarification State of the
Art." USEP Technology Transfer Seminar, Field
Evaluation of I/A Technologies. Handout No. CERI-86-22
(Aug. 1986).
Novak, J.T., and Christopher, S.A. "The Use of
Intrachannel Clarification in the Design of Oxidation
Ditches." in Civil Engineering for Practicing and
Design Engineers. 2_, 45-55 (1982) .
Christopher, S.A., and Titus, R.L. "New Design Combines
Concepts to Cut Costs." Pollution Engineering, 16>, 40-
2 (Mar. 1984).
Parsons, W.C. "A New Concept in Clarifiers."
Works Jour., 118. 61-2 (Oct. 1987).
Public
Stensel, H.D. "Oxidation Ditch Modification Shows
Promise." Water/Engineering and Management, 134, 10,
40-2 (May 1987).
Bowker, R.P.G. "In-Channel Clarification." USEPA
Technology Transfer Seminar, New Municipal Wastewater
Treatment Technology, Handout No. CERI-84-16 (April
1984).
Ettlich, W.F. "Performance Capabilities and Design of
Oxidation Ditch Processes" in Proceeding of Operation
and Maintenance of POTW's. USEPA Publication No.
600/9-83-021 (Dec. 1983) .
USEPA. Process Design Manual for Sludge Treatment and
Disposal. USEPA Publication No. 625/1-79-011 (Sept.
1979). ,
Fitch, B. "Batch Tests Predict Thickener Performance."
Chemical Engineering, 78. 83-8 (Aug. 1971).
Scott, K.J. "Continuous Thickening of Flocculated
Suspensions.11 Ind. Eng. Chem. Fundamentals, 9., 422-7
(Mar. 1970). ' • «
Hartley, K.J. "Hydraulics of Horizontal Shaft Oxidation
Ditches.1^ Jour. WPCF, 59, 686-94 (July 1987).
542
-------
13.
14.
15.
16.
17.
Weston, R.F., Inc. "Evaluation of the Energy
Requirements for Oxidation Ditches" USEPA-OWPO Report
No. 18, Contract No. 68-01-6737. •
Koot, A.C.J. and Zeper, J. "Carrousel, A New Type of
Aeration-System with Low Organic Load." Water
Research, • - •
.6, 401-6 (Apr.-May 1972).
Smalley, Wellford and Nalven, Inc. Report to Central
County Utilities. Inc. on Selection of Clarifier System
for CCUI Wastewater Treatment Plant (March 1984).
Middlebrooks, E.J. and Middlebrooks, C.H. Energy
Requirements for Small Flow Wastewater Treatment
Systems. U.S. Army Corps of Engineers Cold Regions
Research and Engineering Laboratory Special Report 79-7
(April 1979).
Wesner, G.M., Gulp, G.L., Lineck, T.S. and Hinrichs,
D.J. Energy Conservation in Municipal Wastewater
Treatment. U.S. EPA Report No. 430/9-77-011 (March
1978) .
543
-------
WAS
SLUDGE RETURN SYSTEM
f
INFLUENT
OXIDATION DITCH
SECONDARY
CLARIFIER
EFFLUENT
Figure 1.
CONVENTIONAL OXIDATION DITCH SYSTEM
INFLUENT
WAS
THICKENER
OXIDATION DITCH
INTRACHANNEL
CLARIFIER
EFFLUENT
OXIDATION DITCH WITH
2. INTRACHANNEL CLARIFIER
544
-------
AERATION ZONE
CLARIFIER
}
' \
t
EFFLUENT
UPSTREAM END WALL
SUBMERGED ORIFICE
DISCHARGE PIPES
CLARIFIER
BOTTOM BAFFLES
DOWNSTREAM
END WALL
AERATION
CHANNEL
Figure 3.
BMTS SYSTEM
(2)
545
-------
AERATION ZONE
A
BOAT CLARIFIER }
^X^,
AERATION CHANNEL
BOAT CLARIFIER
SLUDGE HOPPERS
=¥=
WEIR
INLET
SLUDGE RETURN PORTS
Figure 4.
UNITED INDUSTRIES BOAT CLARIFIER (21
546
-------
EFFLUENT BOD5 PERFORMANCE
5O
40
O
O
CO
l-
-3L
UJ
1U
30
20
10
-i—i—i 1 1
DATA FROM 17
INTRACHANNEL
CLARIFIER PLANTS
(EPA, 1988)
_L
DATA FROM 12
CONVENTIONAL
OXIDATION DITCH
PLANTS
(ETTUCH,1978)
-J—I 1 II
1 5 10 2O 304050607080 90 95 99
Figure 5. PERCENT OF TIME VALUE WAS LESS THAN OR EQUAL TO
547
-------
50
< 40
O
s
co
Q
J
O
> 30
o
Ul
Q
Z
Ul
CL
CO
3 20
Z
Ul
U,
U.
UJ
10
EFFLUENT SUSPENDED
SOLIDS PERFORMANCE
DATA FROM 17
INTRACHANNEL
CLARIFIER PLANTS
(EPA, 1988)
DATA FROM 12
CONVENTIONAL
OXIDATION
DITCH PLANTS
(ETTLICH.1978)
5 1O 2O 3O4O506070 80 90 95
99
Figure 6. PERCENT OF TIME VALUE WAS LESS THAN OR EQUAL TO
4U.S. GOVERNMENT PRINTING OFFICE: 1990-718-159/20162
548
-------
-------
-------
-------
CD
3
O
C
CO
n>
CO
o
o
q
5"
o
'
01
NJ
0)
00
rn
TJ
I
o
o
CD
O
O
CO
o>
•§ m" ~ *"
Isl *
s--3'0- §
??5 2
JS2 §8 g-
= gS 3 3 0)
°- ^ « i? • =
| ^5 3.1
« 8 3 3 S
•"• < o = »
a n
m X
", "1
~ m
3 o,
||
------- |