&EPA
United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
EPA-600/9-79-O23b
August 1979
Research and Development
Municipal Solid
Waste: Resource
Recovery
Proceedings of the
Fifth Annual
Research Symposium
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific\and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
I
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/9-79-023b
August 1979
MUNICIPAL SOLID WASTE: RESOURCE RECOVERY
Proceedings of the Fifth Annual Research Symposium
at Orlando, Florida, March 26, 27, and 28, 1979
Cosponsored by the University of Central Florida
and the Solid and Hazardous Waste Research Division
U.S. Environmental Protection Agency
Edited by:
Martin P. Wanielista
and
James S. Taylor
Department of Civil Engineering and Environmental Sciences
University of Central Florida
Orlando, Florida 32816
Grant No. 806198
Project Officers:
Stephen C. James
Robert E. Landreth
Solid and Hazardous Waste Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
These Proceedings have been reviewed by the
U.S. Environmental Protection Agency and ap-
proved for publication. Approval does not
signify that the contents necessarily reflect
the views and policies of the U.S. Environmen-
tal Protection Agency, nor does mention of
trade names or commercial products constitute
endorsement or recommendation for use.
ii
-------
FOREWORD
The Environmental Protection Agency was created because of increasing public
and governmental concern about the dangers of pollution to the health and welfare
of the American people. Noxious air, foul water, and spoiled land are tragic
testimony to the deterioration of our natural environment. The complexity of the
environment and the interplay between its components require a concentrated and
integrated attack on the problem.
Research and development is the first necessary step in problem solution;
it involves defining the problem, measuring its impact, and searching for solu-
tions. The Municipal Environmental Research Laboratory develops new and improved
technology and systems to prevent, treat, and manage wastewater and the solid
and hazardous waste pollutant discharges from municipal and community sources;
to preserve and treat public drinking water supplies; and to minimize the adverse
economic, social, health and aesthetic effects of pollution. This publication
is one of the products of that research — a vital communications link between
the researcher and the user community.
The Proceedings present research aimed at minimizing the impact of land
disposal of solid and hazardous wastes. Solutions to specific probelms are sug-
gested.
Francis T. Mayo
Di rector
Municipal Environmental Research Laboratory
-------
PREFACE
These Proceedings are intended to disseminate up-to-date information on
extramural research projects dealing with the disposal of solid and hazardous
wastes. These projects are funded by the Solid and Hazardous Waste Research
Division (SHWRD) of the U.S. Environmental Protection Agency, Municipal Environ-
mental Research Laboratory in Cincinnati, Ohio. Selected papers from work of
other organizations were included in the symposium to identify closely-related
work not included in the SHWRD program.
The papers in these Proceedings are arranged as they were presented at
the symposium. There were two concurrent programs, one primarily for land
disposal and the other primarily for resource recovery. Each program had
multiple sessions. There was an introductory session,, Each of the five ses-
sions includes papers dealing with major areas of interest for those involved
in solid and hazardous waste management and research. Each program, land dis-
posal and resource recovery, is printed in a separate volume,,
The papers are printed here basically as received from the symposium au-
thors,, They do not necessarily reflect the policies and opinions of the U.S,,
Environmental Protection Agency or the University of Central Florida. Hope-
fully, these Proceedings will prove useful and beneficial to the scientific
and engineering community as a current reference on land disposal and resource
recovery.
IV
-------
ABSTRACT
The Fifth Solid and Hazardous Waste Research Division (SHWRD) research
symposium on Municipal Solid Waste: Land Disposal and Resource Recovery was
held at the Marriott Inn in Orlando, Florida on March 26, 27, and 28, 19790
As in previous symposiums, the purpose was similar,, State-of-the-art review
and discussion of ongoing and recently completed research projects were pri-
mary objectives. In addition, the symposium attracted individuals with a
variety of educational backgrounds,and jobs who benefited from an exchange
of ideas and information. The symposium had a dual theme -- Gas and Leachate
in Landfills and Resource Recovery„ The Proceedings are published in two
volumes to correspond with the dual theme. The proceedings are essentially
the papers presented by symposium speakers. The papers are arranged in the
order of presentation. For this volume, the sessions presented are related
to research on resource recovery and are: (1) Introduction, (2) Mechanical
Unit Processes, (3) Chemical Processes, (4) Economic Research Related to
Solid Waste Management, (5) System Studies and Special Projects, (6) Uti-
lization of Recovery Materials, (7) Environmental Impacts of Resource Recovery,
(8) Byconversion.
This report was submitted in fulfillment of Research Grant No. 806198 by
the University of Central Florida under the sponsorship of the U.S. Environmental
Protection Agency. This report covers the period from October 1978 to July 1979,
and work was completed as of July 1979.
-------
CONTENTS
Page
Disclaimer ii
Foreword iii
Preface iv
Abstract v
Contents vii
SESSION I: MECHANICAL UNIT PROCESSES
Pilot Scale Processing Equipment Evaluation for Resource Recovery
Harvey Alter, National Center for Resource Recovery, Inc 40
Mid-Shakedown Evaluation of a Danonstration Resource Recovery
Facility
J. F. Berriheisel, National Center for Resource Recovery 60
Research and Evaluation of Solid Waste Processing Equipment
David Bendersky and Bruce Simister, Midwest Research
Institute 77
Evaluation and Performance of Hanamermill Shredders used in Refuse
Processing
G. M. Savage, G.R. Adiflett, L.F. Diaz and G.J. Trezek, Cal
Recovery Systems, Inc 86
SESSION II: CHEMICAL PROCESSES
Development of Continuous Acid Hydrolysis Process for the Utilization
of Waste Cellulose
vii
-------
Walter Brenner, Barry Rugg, Robert Stanton, Peter Armstrong and
Kuan-Ming Ang, New York University ............................................. 99
Production of Methane from Acid Hydrolysates of Cellulose Wastes
V.R. Srinivason, Louisiana State University ....................................
Stabilization and Characterization of Pyrolytic Oils
Malcolm B. Polk and Metha Phingbodhipakkiya, Atlanta
University
Pyrolytic Oils from Agricultural and Forestry Residues and
Municipal Solid Waste
J. A. Knight, Georgia Institute of Technology ....... , ........ , .......... ,••,..• 126
SESSION III: ECONOMIC RESEARCH RELATED TO SOLID WASTE MANAGEMENT
Some Empirical Evidence on the Effects of User Charges for Solid Waste
Collection Services
Robert J. Anderson, MATHTECH, Inc .............................................. 130
The Desirability of Commodity Futures Trading in Scrap Materials
Roger C. Dower and Elizabeth H. Granitz, Environmental Law
Institute ............................................... <. ......................
An Economic Analysis of Paper Recycling
Roger Bolton, Williams College .................................................
Economic and Institutional Impediments to Conservation and Recycling
Robert C. Anderson, Environmental Law Institute
SESSION TV: SYSTEMS STUDIES AND SPECIAL PROJECTS
Small-Scale and Low Technology Resource Recovery
Gary L. Mitchell and Charles W. Peterson, SCS Engineers .................. ,.,... 179
Compatibility of Source Separation and Mixed-Waste Processing
for Resource Recovery
Harry H. Fielder and Belur N. Murthy. Gilbert & Associates, Inc ................ 197
Forecasts of the Quantity and Composition of Solid Waste
Ralph M. Doggett, International Research and Technology
Corporation
Andrea L. Watson, Bechtel Corporation ......................................... 220
SESSION V: RESEARCH ON RESOURCE RECOVERY
The Development of Testing and Analysis Procedures for Refuse-
Derived Fuel
John Love, University of Missouri-Columbia
Carlton C. Wiles, U.S. Environmental Protection Agency ........................ 225
Calorific Value Determination of Refuse-Derived-Fuels by Large-
Bomb Calorimetry
Duane R. Kirklin, Eugene S. Domalski and David J. Mitchell,
National Bureau of Standards .................................................. 233
viii
-------
SESSION VI: UTILIZATION OF RECOVERED MATERIALS
Technical, Economic and Market Evaluation on the use of Waste
Glass in Structural Clay Brick Manufacture
Ashok K. Gupta, Ratheon Service Company
Laura A. Arozarema, U.S. Environmental Protection Agency 248
The Production and Use of Densified Refuse Derived Fuel
Carlton C. Wiles, U.S. Environmental Protection Agency 274
Status, Trends, and Impediments to Discarded Tire Collection
and Resource Recovery
James F. Hudson, Patricia L. Deese and Douglas Funkhouser,
Urban Systems Research & Engineering, Inc. 293
Advanced Thermal and Chemical Concepts for Improved MSW
Derived Products
N. L. Hecht and D. S. Duvall, University of Dayton
Research Institute 303
SESSION VII: ENVIRONMENTAL IMPACTS OF RESOURCE RECOVERY
Environmental Impact of Resource Recovery
Judith G. Gordon, MITRE Corporation 316
Environmental Assessments of Waste to Fuel Processes
H. M. Freeman, U.S. Environmental Protection Agency 329
The Occurrence of Lead in Municipal Solid Waste
David F. Lewis and Ada Salas, York Research Corporation 339
SESSION VIII: BIOCONVERSION
Start-ups and Operation of the Landfill, Gas Treatment Plant at
Mountain View
M. J. Blanchet, Pacific Gas and Electric Company 346
Anaerobic Digestion of Agricultural Residues - A Technology
Assessment
Tom P. Abeles and David Elsworth, i e associates, inc.
John P. Genereaux, Genereaux Social Science Consultants 353
Gas Recovery from MSW - Sewage Sludge Anaerobic Digesters
Joseph T. Swartzbaugh, Catherine E. Jarvis and Ralph B. Smith,
Systems Technology Corporation 373
ix
-------
PROGRAMS IN LAND DISPOSAL AND RESOURCE RECOVERY AND CONSERVATION
John H. Skinner
Director, Land Disposal Division
Office of Solid Waste
U.S. Environmental Protection Agency
Washington, D.C.
ABSTRACT
This paper describes EPA's p'rograms in land disposal and resource
recovery and conservation. The programs are authorized by the Resource
Conservation and Recovery Act (RCRA) of 1976, which also authorizes the
hazardous waste programs carried out by the Agency. RCRA is implemented
by EPA's Office of Solid Waste. The land disposal and resource recovery
and conservation programs include:
1. development of regulations and guidelines,
2. technology demonstr ations and evaluations,
3. studies and information development, and
4. technical and financial assistance.
The basic purpose and objectives of these activities are summarized and
expected accomplishments and milestones for the next year are presented.
INTRODUCTION
Programs of the U.S. Environ-
mental Protection Agency, Office of
Solid Waste (OSW) are carried
out under the authority of the
Resource Conservation and Recovery
Act of 1976 (RCRA). The basic
objectives of this Act are to pro-
mote the application of improved
solid waste management in order to
protect public health and the envi-
ronment and to conserve valuab le
material and energy resources.
Based on the requirements and pri-
orities in the Act, OSW has devel-
oped programs in these areas:
1. hazardous waste management,
2. land disposal, and
3. resource recovery and
conservation.
Since the subjects of this Symposium
are land disposal and resource re-
covery, this paper will focus only
on those activities. First, speci-
fic projects on land disposal and
resource recovery will be discussed,
followed by a discussion of the
technical and financial assistance
available to help build State and
local government programs.
LAND DISPOSAL
Land disposal activities can be
grouped into two categories:
1. development of regulations
and guidelines, and
2. technology demonstrations
and technical information.
Regulations and Guidelines
The keystone to the disposal
provisions of RCRA are the Criteria
for Classification of Solid Waste
Disposal Facilities (the Disposal
Criteria) called for by Section 4004
of the Act. The Disposal Criteria
will establish national standards
-------
for solid waste disposal by dis-
tinguishing facilities that pose no
reasonable probability of adverse
effects on health or the environ-
ment. Facilities not complying with
these criteria are termed "open
dumps" and are prohibited by the
Act. The Disposal Criteria were
published in proposed form in the
Federal Register on February 6,
1978.Final promulgation is
expected in mid-1979,
Section 4005 of RCRA requires
EPA to publish an inventory of all
disposal facilities in the United
States that are open dumps, i.e.
facilities that do not comply with
the Disposal Criteria. Any facil-
ity found to be an open dump must be
closed or upgraded according to a
State-established compliance
schedule. Financial assistance will
be provided to State solid waste
programs to evaluate facilities for
purposes of the open dump inventory.
A site classification manual is
being developed which will provide a
practical interpretation of the
Disposal Criteria and will be useful
to the States in evaluating facili-
ties for purposes of the open dump
inventory. Also another manual
is being developed describing pro-
cedures and techniques for closing
or upgrading open dumps.
RCRA calls for publication of
the open-dump inventory one year
after final promulgation of the
Disposal Criteria. However, due to
the large number of facilities to
be evaluated and the complexity of
such evaluations, the inventory will
be published in annual installments.
An arrangement has been made with
the Bureau of Census to handle the
data processing and actual inventory
publication.
For the most part the Disposal
Criteria do not specify particular
technologies or operational
approaches but instead establish
performance levels as necessary to
protect public health and the
environment. This was done to allow
different technologies and oper-
ations to be used as the local
situation dictated and to provide
opportunity for new and innovative
approaches. However, RCRA does call
upon EPA to provide guidance on
technologies.
Section 1008 directs the Agency
to develop and publish suggested
guidelines that provide a technical
and economic description of various
solid waste management practices
(including operating practices).
One such guideline describing the
practice of landfilling is under
development at this time. This
guideline will discuss design and
operation of a landfill in order to
protect public health and the envi-
ronment and will recommend practices
for leachate control, gas migration
control and groundwater monitoring.
A working draft of the Landfill
Guideline was circulated last uear
and a proposed version should be
published in the Federal Register
for public comment by March of 1979.
Other guidelines to be devel-
oped in the future will cover land-
farming of solid waste and surface
impoundment disposal. In addition,
preparatory work is underway to
develop guidelines on the manage-
ment of two special waste categor-
ies: coal-fired electric utility
wastes and mining wastes. In
cooperation with the Office of'
Research and Development programs
are being initiated to evaluate
facilities that dispose of fly ash,
bottom ash, and scrubber sludge
from coal-fired utilities and
wastes from metals and phosphate
mining and beneficiation processes
and uranium mining. Thirty to
forty facilities will be evaluated
over a period of two years in order
to assess practices for managing
such wastes.
Finally a guideline is being
developed for the disposal and
utilization of wastewater treatment
plant sludge under the combined
authorities of RCRA and Section 405
of the Clean Water Act. This
guideline will discuss practices
for landfilling sewage sludge, use
of sludge as a soil conditioner on
-------
agricultural lands, thermal pro-
cessing of sludge for energy
recovery, and sludge composting
or heat drying to produce a bulk or
bagged product. These guidelines
will be proposed in mid-1979 with
final promulgation expected in 1980.
Section 405 of the Clean Water
Act requires compliance with such
guidelines by operators of pub E.C-
ly owned wastewater treatment works
and it is expected that these
guidelines will be administered
through the National Pollution
Discharge Elimination System permit
process.
Technology Demonstrations and
Technical Information
The Office of Solid Waste
has several projects underway to
demonstrate and evaluate improved
techniques for disposing of solid
waste on land.
In Tullytown, Pennsylvania,
leachate from an asphalt-lined
landfill is collected and treated
by a number of physical and bio-
logical methods. The leachate
treatment plant has been operating
for several years and a number of
EPA reports have described its
operation. The final project
report was received in early 1979
and should be publicly available
shortly. The treated leachate is
discharged to a nearby river in
conformance with standards estab-
lished by the State.
In 1979, full-scale operation
of another leachate treatment
facility in Enfield, Connecticut
is expected to begin. This facil-
ity, based upon research conducted
by the EPA, Office of Research and
Development and the University of
Illinois, uses an anaerobic filter
to treat leachate. One of the
features of this system is the low
volume of sludge produced. The
final report on this promect is
expected in mid-1980.
A full-scale lined landfill
demonstration is underway in
Lycoming County, Pennsylvania.
After several uears in development,
a landfill employing a PVC liner
to collect leachate began to re-
ceive wastes in early 1978. A
report will be available in 1979
covering among other things, pro-
cedures for installing a liner
sheet covering a large area. Two
other lined landfill demonstrations
in Northern Tier, Pennsylvania are
expected to start in 1979.
In Bangor, Maine the Office of
Solid Waste is participating with
the City in demonstrating a pro-
cess for composting sewage sludge.
This project is based upon the
aerated-pile process for composting
sludge developed by the U.S.
De%artment of Agriculture in Belts-
ville, Maryland. This demonstra-
tion represents the first on-line
application of this process in a
small community. It has proven
very successful and composted
sewage sludge has been sold and
used as a soil conditioner for
over a year. An interim report
was published in 1977 and a final
report is due in 1979.
In Mountain View, California a
very innovative project involving
the recovery of methane from a
landfill is underway. In a joint
effort involving the City, the
Pacific Gas and Electric Companu
and EPA, methane gas is pumped from
a landfill, cleaned up, and injec-
ted into a nearby pipeline supply-
ing gas to residential users.
This project not only prevents the
safety problems that could result
from the migration of gas from the
landfill, but also recovers and
utilizes a valuable resource. The
facility has been operational for
nearly a year, interim reports are
available, and a final report is
expected in mid-1979.
The Office of Solid Waste also
prepares a number of technical re-
ports on various subjects pertain-
ing to land disposal. One such
report that was very well received
-------
was a procedures manual for moni-
toring leachate and groundwater
around disposal facilities. We also
have an ongoing effort to monitor
groundwater at a number of disposal
sites both to improve monitoring
techniques and gain a better
understanding of leachate and
groundwater interaction. Other
technical reports under development
include an assessment of methods to
prevent and control migration of
toxic or explosive gases at disposal
sites and an evaluation of pro-
cedures to prevent hazards to
aircraft from birds atrracted to
disposal facilities.
The Office of Solid Waste
also participates in the review
and evaluation of the wide range
of research and development pro-
iects in the land disposal area
conducted by the Office of
Research and Development and
described in other papers at this
Symposium.
RESOURCE RECOVERY AND CONSERVATION
Resource recovery and conser-
vation activities will be grouped
into three categories for
discussion:
1. activities of the Resource
Conservation Committee
2. guidelines for Federal
agencies, and
3. demonstrations, evalua-
tions and studies.
Resource Conservation Committee
The Resource Conservation
Committee was established by
Section 8002(j) of the Act.. It is
a Cabinet-level committee chaired by
the Deputy Administrator of EPA and
supported by OSW staff. The
Committee is required to conduct "a
full and complete investigation and
study of all aspects of the econo-
mic, social, and environmental
consequences of resource conserva-
tion." To obtain public input for
these studies, the Committee held
several hearings on beverage con-
tainer deposits, solid waste dis-
posal charges, the effect of
existing Federal tax and transpor-_
tation policies on the use of virgin
and secondary materials, recycling
and resource recovery sibsidies,
deposits or bounties on durable
goods, local user fees, litter
taxes, severance taxes, and product
regulation.
The first resource conservation
issue taken up bu the Committee was
beverage container deposits. In its
January 1978 report to the President
and Congress, the Committee pre-
sented a recommended design for
beverage container deposit legisla-
tion.. The Committee did not recom-
mend that such legislation be
passed, however; a decision on this
issue was postponed until further
analyses were completed.
In its July 1978 report to the
President and Congress, the Commit-
tee submitted preliminary analyses
of solid waste disposal charges.
Work is continuing on beverage con-
tainer deposits and solid waste
disposal charges and the results of
these and other analyses will be
presented in the Committee's final
report along with recommendations.
This report is scheduled for com-
pletion in March 1979.
Guidelines for Federal Agencies
On April 23, 1976, EPA issued
Guidelines for Source Separation
for Materials Recovery. These
guidelines require the recovery of
highgrade paper, newsprint, and
corrugated boxes from Federal
facilities. As a result, 175,000
Federal employees in 135 Federal
facilities arel participating in the
paper recycling programs. It is
expected that" another 100,000
employees will be in the program
by October 1979. OSW has worked
with the General Services Admini-
stration in establishing contract
procedures for the sale of the
paper and in developing an educa-
tional program for training
Federal employees in procedures for
-------
recycling office paper. To date,
over 2,000 tons of paper have been
recycled under the guidelines with
a return to the Federal Treasury of
over $125,000. In addition, at
least 15 states and hundreds of
local governments have adopted the
guidelines.
On September 21, 1976, EPA
issued Guidelines for Beverage
Containers. These guidelines
require that a refundable 5-cent
deposit be placed on all containers
for beer and soft drinks sold at
Federal facilities. The deposit is
intended to encourage the return of
containers for either refilling or
recycling. The Department of
Defense concluded a test of the
guidelines at 10 military bases in
June 1978 and will submit its plans
based on this test in early 1979.
The General Services Administration
also conducted a test of the guide-
lines in 8 of its 10 regions. To
date, 13 Federal agencies are
implementing the guidelines. The
OSW monitors agency compliance with
the guidelines and provides assist-
ance upon request.
On September 21, 1978, EPA
issued Resource Recovery Facilities
Guidelines. These guidelines
contain requirements and recommended
procedures for Federal agencies re-
garding establishment and use of
resource recovery facilities. To
date, some degree of implementation
has been reported by Federal agen-
cies in nine metropolitan areas.
Procurement Guidelines are
currently being developed under
Section 6002 of RCRA. Al agencies
procuring with Federal funds
(including State and local govern-
ments, grantees, and contractors as
well as Federal agencies) must
"procure items composed of the
highest percentage of recovered
materials practicable." The guide-
lines for recommended procurement
practices will contain information
regarding suppliers, demand, price,
delivery time, performance, and
certification techniques.
OSW is now gathering data on
products purchased by the govern-
ment that may have high potential
in the use of waste materials.
Studies include:
1) The use of fly ash and
blast furnace slag in cement and
concrete. 2) The use of wastepaper
in insulation and board; waste
rubber in asphalt pavements; and
concrete. 3) The use of wastepaper
and other secondary fibrous mate-
rials in paper products and 4) The
use of composted sewage sludge used
as a soil conditioner and low-grade
fertilizer.
The first guideline, scheduled
to be proposed in April 1979, will
be on the use of fly ash and blast
furnace slag in cement and concrete.
Demonstrations, Evaluations and
Studie¥
The OSW is involved in the
demonstration and evaluation of the
technical, economic and environ-
mental performance of a number of
resource recovery systems.
Two of the early demonstration
projects developed technologies
that are being replicated in a
number of communities. The first,
in Franklin, Ohio, determined the
feasibility of using the wet-
pulping method of separating mixed
municipal solid waste into organic
and inorganic fractions. Follow-on
applications of this technology
will use the organic fraction of
the solid waste as a fuel.
The second demonstration, in
St. Louis, Missouri, developed the
technology of recovering a portion
of the organic fraction of the
mixed municipal waste stream for
use as a supplement to coal in
large boilers. This technology is
being replicated in 10 locations.
An extension of this concept that
includes utilization of sewage
sludge is being demonstrated under
a grant to the State of Delaware.
-------
Two systems for recovering
energy from mixed municipal solid
waste through pyrolysis were
demonstrated in Baltimore, Maryland,
and San Diego County, California,
The Baltimore plant, which produces
steam by combusting the pyrolysis
gases, has had numerous mechanical
and air pollution problems and is
presently undergoing an extensive
modification program. After the
modifications are completed, it is
anticipated that the process will
run in an economically and environ-
mentally sound manner. The San
Diego project also experienced
difficulties and is currently not
in operation. Although it is
doubtful that either of these
systems will be replicated in toto,
much was learned concerning the
pyrolysis of solid waste, and this
information should prove useful to
others considering this process.
Six projects to develop source
separation systems are now nearing
completion. Projects in Marblehead
and Somerville, Massachusetts, were
started three years ago to deter-
mine the feasibility of separate
collection techniques to recover
several materials from municipal
waste streams. These two programs
have shown that it is possible to
recover between 25 and 30 percent
of the residential waste stream in
suburban communities. Approximately
15 to 20 other multimaterial collec-
tion programs have begun in the New
England area as a result of the
increased municipal and industru
interest created by the Marblehead/
Somerville projects. A final re-
port presenting the results of this
project will be issued in 1979.
Four projects were started two
uears ago on: multimaterial col-
lection through private contract;
multimaterial collection from
apartment buildings; source sepa-
ration and materials marketing for
low-density rural areas; and the
use of handicapped laborers for
processing materials. Final reports
will be issued in 1979.
In addition to EPA-supported
demonstrations of resource recovery
systems, OSW is also conducting
evaluations of other commercial-
scale resource recovery facilities:
refuse-derived fuel plants in Lane
County, Oregon, and Chicago,
Illinois; small modular incinera-
tors with heat recovery; waterwall
combustion units in Europe; a plant
for codisposal of sewage sludge and
municipal solid waste in Duluth;
and a pyrolysis system in Frankfort,
Germany. Evaluations of system
components such as air classifiers
and shredders are also underway.
In cooperation with the Office
of Research and Development, OSW is
conducting studies on resource
recovery as required by section
8002 of the Act on the subjects of:
1) compatibility of source sepa-
ration and mixed waste recovery,
2) small-scale technology, 3) re-
search priorities for resource
recovery and 4) tire, glass and
plastics recovery.
FINANCIAL AND TECHNICAL ASSISTANCE
TO STATE AND LOCAL GOVERNMENTS
The Resource Conservation and
Recovery Act places the primary
responsibility of improving solid
waste management on State and local
governments. The Act provides for
a number of financial and technical
assistance programs for assisting
in that effort.
Financial assistance is avail-
able to develop and implement
comprehensive State solid waste
management plans under Subtitle D
of the Act. Guidelines to assist
in the development and implemen-
tation of State solid waste manage-
ment plans were published in pro-
posed form in August 1978 and will
be finalized in June 1979. These
plans should provide for (1)
regulatory programs for environ-
mentally sound disposal of all
other solid wastes, (2) State
programs to encourage resource
recovery and conservation, and
-------
(3) planning for necessary disposal
and recovery facilities. In fiscal
year 1979, $11.2 million was avail-
able for these purposes.
A new financial assistance
program resulting from the
President's Urban Policy Message
will help communities to adequately
assess the feasibility of resource
recovery projects and obtain suffi-
cient consultation and staff for
the preparatory steps in implemen-
tation. This program is scheduled
to begin in fiscal uear 1979 and
$15 million has been appropriated
for the first year. Urban areas
will be eligible for assistance for
"front-end" steps preceding actual
design and construction of resource
recovery projects. This includes
not only mixed waste resource
recovery processing plants but also
projects involving source separation
(i.e., separation at the place of
generation) of waste materials for
recycling. Eligibility for funding
is not limited to large cities, but
a major portion of the funds will
probably go to jurisdictions of at
least 50,000 population. The aid
will go primarily to agencies with
clear responsibility for implemen-
tation as designated in the State
planning process udner Subtitle D.
The first set of applications were
received in December 1978 and
awards are expected to be made in
March or April 1979.
The Act also provides for
technical assistance through the
Technical Assistance Panels Program.
Each EPA regional office has avail-
able a panel of expert consultants
capable of assisting in all areas
of solid waste management. Grants
have been provided to a number of
public interest groups to make State
and local government officials
available to provide assistance as
well. These panels are available
to provide technical assistance
free of charge to State and local
governments and Federal agencies
upon request. Twenty percent of the
funds appropriated under the general
authorization of the Act are used
for this purpose.
The OSW has developed a seminar
series directed mainly at municipal
officials. A first set of seminars
presented alternatives and major
issues in the implementation of
resource recovery. In 1978 the
seminar was conducted in six cities;
approximately 1,000 people attended.
The seminar program will continue
into 1979 and will be expanded to
cover other solid waste management
issues as well.
The Resource Conservation and
Recovery Act provides an opportu-
nity for significant advancements
in solid waste management nation-
wide. The regulations and guide-
lines provide momentum and impetus
for improved practices. Technology
demonstrations, evaluations and
studies will help provide necessary
technical, economic and environ-
mental performance information so
that better decisions can be made.
Financial and technical assistance
programs will provide additional
resources to State and local govern-
ments responsible for implementa-
tion. The Office of Solid Waste
programs in these three areas are
an attempt to provide a balanced
approach to this nationwide effort.
-------
USEPA REGION 4 SOLID WASTE MANAGEMENT ACTIVITIES
Elmer G. Cleveland
U.S. Environmental Protection Agency
345 Courtland Street
Atlanta, Georgia 30308
In the beginning — in this instance
the late 1960s — solid waste management
in Region 4 was at best completely inade-
quate even by standards prevailing at that
time. Before getting into an update on
regional solid waste management activities
we must look at our beginning.
Prior to the passage of the Solid
Waste Disposal Act of 1965, there did not
exist an identifiable state solid waste
program in the region. A survey of land
disposal sites in the late 1960s indicated
over 2500 dumps and fewer than 30 sanitary
landfills which could meet the very loose
requirements that prevailed at that time.
By 1971, total state solid waste program
positions were 42.
State solid waste laws were enacted,
viable programs established, regulations
promulgated, and state plans developed.
Inspection and technical assistance
programs were put into action.
By 1977, over 1000 land disposals met
or exceeded the minimum requirements to be
termed a sanitary landfill and permitted
or approved by the state programs. State
program personnel now exceeds 225 with
increases scheduled to take on the addi-
tional duties associated with hazardous
waste and resource recovery.
While maintaining the successful level
of solid waste management attained, a
combined federal, state and local govern-
mental effort is underway to correct and
eliminate existing and potential environ-
mental hazards from old disposal practices.
New and environmentally sound hazardous
waste processing and disposal facilities
are being designed and constructed.
Some are now in operation.
Resource recovery facilities are in
operation, under construction, and being
planned. Under the President's Urban
Policy; which provides financial assis-
tance for Resource Recovery Project
Development, more than thirty applications
requesting over $3.5 million in Federal
assistance were received from this region.
The Department of Energy has solid waste
demonstration projects in Alabama, Florida
and Georgia.
EPA funding of state solid waste
programs in Fiscal Year 1979 will exceed
four million dollars. Technical assis-
tance through Section 2003 of the
Resource Conservation and Recovery Act
has been delivered to more than 30 local
governments at an approximate cost of
$200,000. This was provided at no cost
to the clients, i.e. local governments.
In the 1980s, we expect to achieve a
level of solid waste management, hazardous
waste control and resource recovery that
should meet our immediate needs and set
examples for the rest of the Country.
-------
CURRENT RESEARCH ON RESOURCE RECOVERY
Albert J. Klee, Ph.D.
Steven C. Lees
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
ABSTRACT
The Resource Conservation and Recovery Act of 1976 (RCRA) charges the U.S. Environ-
mental Protection Agency with the responsibility of providing "technical and financial
assistance for the development of management plans and facilities for the recovery of
energy and other resources from discarded materials and for the safe disposal of discarded
materials, and to regulate the management of hazardous waste." The Solid and Hazardous
Waste Research Division (SHWRD), Municipal Environmental Research Laboratory (MERL),
Cincinnati, Ohio, has many technical responsibilities required for fulfillment of this
Act. Research is performed in a variety of technical, economic, and institutional areas,
reflecting the broad opportunities and requirements in the recovery of resources from
wastes.
Although the SHWRD, Processing Branch, has research projects dealing with hazardous
waste management and control technologies, this paper examines, in an overview fashion,
the current research efforts relating to resource recovery and the improved management of
municipal solid waste (MSW). This research ranges in scope and complexity from short-term
institutional systems analyses to full-scale, long-term demonstrations of developed Resource
Recovery Technologies. To address adequately the many facets of resource recovery, re-
search has necessarily been diverse; however, all efforts lead to the goal of full utili-
zation of the resources available in our refuse. Accordingly, this paper is presented to
bring together the many individual projects into a unified, comprehensive strategy avail-
able for management of these wastes.
The format of this presentation follows the general strategy used to conceptualize,
plan, design, construct, and operate a resource recovery system. Specifically, the follow-
ing general areas are discussed: (1) Waste Forecasting, (2) Systems Analyses, (3)
Processing, (4) Materials Recovery and Reuse, (5) Refuse Derived Fuels, (6) Alternative
Resource Recovery Options, and (7) Pollution Control.
INTRODUCTION lution. As of 1977, only 7 percent of the
waste stream was recovered. Of this, 90
In 1976, annual municipal solid waste percent was source-separated paper. In
generation was estimated to be approximately fact, the business of commercial resource
130 million metric tons. Every individual recovery has accounted for less than 1 per-
in the United States contributes roughly cent of the total wastes generated. Many
1,300 pounds to this figure. By 1985, the resource recovery facilities have not
yearly total is projected to increase to achieved success due to economic, technical,
180 million metric tons. What happens to and institutional constraints. Poor plan-
this waste? By and large the majority is ning, retrofitted equipment, over-rated
destined for land disposal, with no recovery waste flows, and institutional and market
of resources or available energy and with problems have all been blamed, with some
its inherent problems of leachate and gas degree pf validity. The long range solu-
control to avoid water, air, and crop pol- tion to this problem is proper planning,
-------
selection of equipment specifically de-
signed for refuse and implemented in the
optimal process scheme, development of mar-
kets for recovered commodities, and over-
coming of minimizing any institutional or
social uncertainties. Simply stated, re-
source recovery is much more complex than
originally anticipated, and more research
and development is required.
The research performedby the SHWRD is
designed to aid in this long range solution.
The format adopted for this paper attempts
to class the research activities into cate-
gorical areas applicable to amny resource
recovery options. Some research activities
are singular and defy classification, and
the same holds true for some resource re-
covery activities. By the identification of
similarities in processes, the problems en-
countered, their causes and fixes, and the
research requirements, a qualified decision
can be made as to the viability of a total
resource recovery system. It is the purpose
of the SHWRD research program to help in
making these decisions.
WASTE FORECASTING
Many resource recovery facilities have
not had the benefit of actual area waste
surveys prior to design and construction of
the plant. Examples include Ames, Baltimore
County, and the El Cahon facility. Many
other facilities simply relied on "National"
figures as gross estimates of quantity and
composition, as was the case at Franklin's
wet processing plant. The lack of reliable
data on quantity and composition has mani-
fested itself in many negative ways, includ-
ing inadequate total waste flow to fully
utilize the equipment installed. Other ex-
amples include low aluminum loading, un-
usually high moisture and ash contents, or
extreme seasonal fluctuations.
The necessity for accurate quantity/
composition data prior to design has gener-
ally been established. However, numerous
methodologies have been employed to gather
this information, with varied degrees of
accuracy. The initial effort in waste fore-
casting by SHWRD (1) is identifying,
developing, and comparatively evaluating
several methodologies for accuracy in esti-
mating total quantity and waste composition.
Concurrently, the project is determining to
what degree of accuracy these variables must
be estimated to insure proper design of the
facility. Many communities, private organi-
zations, and institutions have been survey-
ed to determine the methodologies used,
their success in predicting actual quantity
and composition, and the identified short-
comings. The output of this research will
be a developed protocol for conducting a
composition and quantity study, including
appropriate "cross-checks" to insure accu-
racy. THis developed protocol will then be
implemented in the context of a field test.
Whereas the above study will provide a
protocol for determining a municipality's
"immediate" waste scenario for establishing
a resource recovery system, the second ef-
fort (2) in this area is to establish solid
waste quantity and composition trends on a
national level. Such factors as increased
use of packaging materials, substitution of
plastic for steel in certain applications,
and beverage container deposits, will be
evaluated and their effects projected on a
national level to the year 1990. The in-
formation obtained will be used in deter-
mining national waste trends and the impli-
cations for collection, disposal, and
resource recovery.
To provide users with the data obtained
in these two projects, along with data on
collection and disposal costs, a solid waste
data base is being developed (3). This proj-
ect will incorporate the methodologies devel-
oped above, solicit additional data needs,
and provide for a mechanism of continual
updating. A questionaire format will be
developed to assist in this update as well
as to track the success or failure of the
forecasting methodologies as implemented.
Ultimately, this solid waste data system
will provide information on many munici-
palities' "immediate" waste scenarios as
well as tracking national solid waste quan-
tity, composition, and collection/disposal
costs. A training program for data gather-
ers will ensure accurate responses to the
questionaire. As State Solid Waste manage-
ment programs are implemented, this system
should provide the information to perform
comparative analyses.
SYSTEMS ANALYSES
Social, economic, and istitutional
factors affecting the implementation of re-
source recovery facilities or improved waste
management techniques are addressed under
Systems Analyses. Perhaps the most compre-
hensive effort in this area is the one
initiated in response to RCRA, Section 8002
(A). Designed to identify the impediments
to economical operation of resource recovery
10
-------
facilities, this effort should identify the
inhibiting factors that have plagued most
operating facilities. These factors will
be quantified, related to research needs,
and ranked to signify which, when removed,
will contribute most to the successful
operation of resource recovery facilities.
Based on these results submitted to the EPA
Administrator, a plan for research, develop-
ment, and demonstration will be submitted
to the appropriate committees of Congress.
Several other mechanisms for improved
waste management and waste reduction are
also being explored. Based on the limited
success of the source separation programs
prevalent in the Northeast, research is
underway to identify the socio-economic
classes of society most likely, and least
likely, to participate in such programs (5).
This research involves analyzing results
from several types of media campaigns and
public information systems to determine the
most effective mechanism for stirring in-
creased participation.
User charges for collection and dispos-
al have been adopted by several cities in
the United States as a mechanism for improv-
ed waste management. The first project
dealing with user charges (6) developed a
descriptive list of the communities using
user charges and categorizing them into five
general catagories. Case studies are being
performed in selected communities represent-
ing each general catagory. To the extent
that data are available, the effects of user
charges on per capita generation rate&y com-
position of refuse, household preferences
for optional levels of service, and alter-
native disposal modes (including resource
recovery) will be measured.
Concurrently, the SHWRD is monitoring
and assisting the City of Seattle's variable
user charge experiment (7)- Along with pro-
viding assistance in experimental design,
development, and implementation of evalua-
tive methodologies, EPA is also assisting
with data gathering and techniques for cost
analysis, data gathering and techniques for
demand and benefit analysis, including the
relevant elasticities of demand, and imple-
mentation of the rate structure. Results
will be evaluated for implications to solid
waste collection, disposal, and source
separation.
Institutional systems designed to favor
the use of virgin materials have been cited
as one major reason operating resource re-
covery facilities have suffered economical-
ly and, on a larger scale, why capital in-
vestments in such systems have been avoided.
The problem appears to include three major
areas: Freight rates for secondary materi-
als, the availability of markets for the
recovered commodities, and the establishment
of standard specification and analysis pro-
cedures for MSW and MSW-derived products.
To quantify the effects of the secondary
material freight rate structure, a theoreti-
cal demand model incorporating transporta-
tion costs as a factor in the demand for
investment is being developed (8). This
model will be adapted to investment deci-
sions by the raw steel-making and ferrous
foundry industries, and will be specified
to permit emperical testing at a future
date.
An economic analysis of scrap futures
markets is underway to determine whether
scrap materials are amenable to futures
trading, shether futures markets would en-
courage capital inflow to the scrap indus-
try, and what the benefits and costs would
be to market participation and to society
(9).
In cooperation with the ASTM E-38 Com-
mitee on Resource Recovery, standard pro-
cedures for sampling and analysis of MSW
(especially RDF and other recovered materi-
als) are being developed (10). Several
candidate procedures and methodologies have
been compiled, reviewed, and consolidated
into an Interim Standard Procedure. Round-
Robin testing techniques are being employed
to test the validity of the procedures.
PROCESSING
Operations on refuse prior to the re-
covery of the energy value of the light
fraction or disposal by landfilling has been
termed "processing." These processing
operations usually involve some form of
communition, separation or classification
of the various components, and preparation
of the combustible fraction for ultimate
end-use. The equipment or unit operations
used in these processing activities are
numerous and not fully understood, especi-
ally in relation to one other when designing
a total process scheme.
The SHWRD has ongoing research to de-
fine the design and operational parameters
needed for evaluation and selection of these
11
-------
equipment items.
The first effort in processing research
(11) is being performed at the National
Center for Resource Recovery, Equipment Test
and Evaluation Facility (ETEF). Various
unit processing hardware, including shred-
ders, air classifiers, screens, conveyors,
and gerrous and non-ferrous separators, are
being evaluated and developed. Numerous
performance characteristics, curves of ef-
fectiveness, and comparative advantages and
disadvantages have been established. Eco-
nomic aspects are also being addressed re-
lative to operating and maintenance costs.
Whereas the previous research has been
performed at pilot scale, several projects
are evaluating and optimizing full-scale
operational systems. The Recovery I facili-
ty in New Orleans is undergoing a series of
test evaluations to establish performance,
maintenance requirements, economics, and
optimal sequencing of the various mechanical
processing equipment (12).
Specific equipment items to be evaluat-
ed include the trommel, air classifier,
ferrous recovery system, aluminum recovery
system, glass recovery system (jigging and
crushing), glass cleanup system (flotation),
glass drying system, and the nonferrous re-
covery system.
To supplement the research results
obtained at Recovery I, several other faci-
lities and selected equipment items are
being evaluated (13). This effort deals
primarily with processes and/or equipment
designed and operated for production of a
fuel or a fuel feedstock. First efforts
have identified the state-of -the-art of
processing equipment and the research re-
quirements. Test plans were then formulat-
ed and field tests conducted to meet some
of these research needs.
Several equipment items have been
identified as almost universal components
of resource recovery facilities. Of these,
the shredder and air classifier are of the
most interest, since the shredder consumes
the most energy of any piece of processing
hardware and the air classifier directly
controls the quantity and quality of the
recovered combustible fraction. To evalu-
ate different shredder designs comparative-
ly, research is being performed at several
operational facilities with different shred-
der designs (14). Several horizontal shaft
hammermills are being evaluated, as well as
some newer designs, including a vertical
shaft ring gear grinder, a reversible rota-
tion horizontal shaft hammermill, and a
vertical shaft hammermill. Performance and
operating data will be obtained as well as
analyses of the MSW processed by the shred-
ders.
To more fully understand teh effective-
ness of the operation of various air classi-
fier designs, a study has been initiated to
document several performance characteristics
of all the designs tested (15). This effort
should also produce a standard protocol and
testing procedures for use in other air
classifier evaluations.
MATERIALS RECOVERY AND REUSE
SHWJID is conducting three studies in
this area in response to RCRA, Section 8002.
The first effort deals with identifying and
prioritizing research needs for energy and
materials recovery (16). This prioritiza-
tion will then be evaluated as to the type
and extent of Federal assistance required to
satisfy the identified research needs.
Secondly, a survey of the state-of-the-
art of glass and plastic recovery will iden-
tify and assess the potential problems and
solutions associated with increased recovery
of these commodities (17). The quantity of
waste generated, recovery methods and tech-
nical feasibility, economics, environmental
impacts, markets, and industry-required
specifications are just some of the factors
to be weighed in this study.
The problems and potential of improved
waste automobile tire management will be
addressed in the third RCRA special study
(18). The current state of recovery and
reuse will be identified, as well as the
prevailing trends, directed by technical and
economic factors..
The SHWRD has sponsored research for
several years on the use of waste glass in
structural clay brick manufacture (19) .
The composition of the waste glass recover-
ed, typically containing 10-30 percent
organics, has been evaluated both technical-
ly and economically relative to effects in
the manufacture and furnace curing of the
bricks. The use of these "waste glass
slimes" now appears technically feasible,
12
-------
and the energy requirements in the brick
curing stage have also been reduced.
REFUSE DERIVED FUELS
One of the most promising energy recov-
ery technologies available is the use of
densified refuse derived fuels (d-RDF) for
combustion with coal in existing utility,
institutional, and industrial stoker boil-
ers, extensive research is being performed
on d-RDF production, including identifying
the optimal processing scheme and quantify-
ing the effects of combustion.
The first efforts in d-RDF production
were to investigate the technical and eco-
nomic aspects of preparation and use (20).
Important characteristics required for a
specification d-RDF were identified and
measured. The first 300 tons produced were
co-fired with coal at the Maryland Correc-
tional Institute, Hagerstown, Maryland.
Based on the results of this test burn,
which produced minimal handling and feeding
problems as well as acceptable emissions,
d-RDF production was continued at the Na-
tional Center for Resource Recovery (ETEF)
to allow for a second, more comprehensive
test burn. Along with the 900-1200 tons
being produced by NCRR, an additional 2000
tons of d-RDF have been purchased. This
second test burn is scheduled to be at the
General Electric Plant, Erie, Pennsylvania
(21). The data gathered during this burn
should provide industry and utilities with
sufficient information to decide as to the
viability of this energy recovery option in
their particular situation.
During the initial phases of production
at NCRR, it became evident that certain
fundamental considerations in d-RDF produc-
tion were lacking. A study was initiated
to address these basic theoretical consider-
ations, and to relate these concepts to
large scale production (22).
Besides co-firing d-RDF in stoker boil-
ers, research is being performed to deter-
mine the requirements for a refuse derived
fuel for use in cement kilns (23) . Fuels
used in cement kilns are typically finely
shredded or pulverized coal, so process
modifications in RDF production will be re-
quired. These processing and preparation
requirements will be addressed in anticipa-
tion of actual field testing.
ALTERNATE RESOURCE RECOVERY OPTIONS
Although the use of densified RDF has
received the most widespread attention as
an energy recovery technology, it is but one
of many alternatives being explored by the
SHWRD.
Exploratory research is being performed
to identify and develop chemical and thermal
processes for improving the fuel quality of
the air-classified combustible fraction (24).
This work has centered on cellulose embrit-
tlement techniques, although other processes
have been evaluated, such as carbonization.
Two RCRA special studies are also being
conducted in this category. The first ef-
fort is reviewing and analyzing systems of
small-scale and low technology solid waste
management, including resource recovery and
resource recovery systems which have special
application to multiple dwelling units and
high density housing and office complexes
(25). The degree of contribution bf these
systems to energy recovery and consercation
is also being evaluated.
The second study in response to RCRA is
investigating and defining the compatibility
of a source separation program with a high
technology resource recovery system (26).
Various levels of source separation and the
related effects are being investigated, with
an economic determination of the optimal
combination of the two approaches.
Getting away from what may be consider-
ed "conventions1" resource recovery as dis-
cussed earlier, several research efforts are
exploring the potential for energy recovery
from MSW through other means.
The feasibility of constructing and
operating landfills to optimize the produc-
tion and recovery of methane gas is under
investigation (27). Various factors are
being defined, such as recovered methane
quality, landfill chemistry as related to
increased biodegration, and waste quantity
and composition effects.
Mehthane production from anaerobic di-
gestion of a mixture of sewage sludge and
MSW is also being explored (28). Gas and
mechanical mixing mechanisms are being
evaluated for effects in the pilot-scale
operation. Several factors, such as reten-
tion time, MSW/sewage sludge mixing ratio,
13
-------
and loading rate are being studied.
Chemical treatment of various urban and
agricultural wastes is being performed for
the determination of the efficacy of the
process in producing methane and organic
acids (29). This research effort involves
conversion of the cellulosic materials into
a fermentable substrate by alkali pretreat-
ment, anaerobic fermentation with selected
acid generation bacteria, and carboxlic
acid recovery by dialysis, coupled with
soluable -'.on exchange polymers.
The SHWRD's largest effort in alterna-
tive resource recovery options is the acid
hydrolysis system, now at the one ton per
day pilot plant level (30). After two years
of study with a 1 liter and 5 liter reactor,
it was determined that cellulosics pretreat-
ed with 1% sulfuric acid at temperatures
around 450''F, resulted in up to 50% conver-
sion to glucose in reaction times as short
as 10-20 seconds. In addition, the surgars
produced appear to be convertible to ethyl
alcohol and single cell proteins.
In light of the problems experienced
with the storage of the pyrolytic oils pro-
duced at El Cajon and elsewhere, our re-
search in utilization and stabilization of
pyrolytic oils has gained in importance.
Ongoing for several years, early research
efforts focused on physical processing of
the oil and characterization of the products
and fractions obtained (31). Subsequent
efforts dealt with chemical processing of
these fractions to obtain commercially
useable products. Research efforts in a
concurrent program have attempted to eluci-
date the mechanism involved in the change
in viscosity of the pyrolytic oils (32). A
combined gas chromatographic-mass spectro-
metric (GC-MS) and liquid chromatographic
(LC) analysis acheme is being developed to
identify the many components of the oil,
and the chemical changes occuring that
produce the increased viscosity. At this
time it appears that recovery of phenols,
and hence production of phenol-formaldehyde
resins, is feasible.
POLLUTION CONTROL
To insure that the energy or material
recovery techniques being developed are
environmentally acceptable and produce a
net positive gain to society and the envi-
ronment, research is being performed to
determine the areas of importance in pollu-
tion control. The SHWRD currently is in-
volved in several research projects in this
area.
The most comprehensive effort, perform-
ed in coordination with the Industrial En-
vironmental Research Laboratory (IERL),
Cincinnati, involves development of air pol-
lution control technology for wastes as
fuels processes (33). Primary efforts have
been in testing and development of improved
fabric filters for a variety of systems,
such as RDF and mass burning. Preliminary
investigations are also being conducted on
design of an improved wet scrubber system.
The potential environmental benefits
and effects of increased resource recovery
are being quantified in another research
project (34). Primary emphasis is placed on
the effects on landfilling and incineration,
and the subsequent reductions in leachate
production and air pollution.
Research is also being conducted on the
emissions from processing and bioconversion
systems (35). Preliminary assessment cri-
teria and pollutant measurement techniques
have been defined, and the pollutants have
been characterized from a processing and a
bioconversion system. Based on these re-
sults, an analysis of applicable control
techniques will be perfromed.
During the operation of the St. Louis
RDF project, cursory sampling at selected
locations revealed higher-than-background
levels of bacteria, causing concern over
the potential health implications. A study
was subsequently performed to gather addi-
tional data at and around the processing
facility (36). Results indicated that there
were increased levels of bacteria present.
However, data does not exist to identify or
quantify absolutely the health effects.
Lead and other trace metal emissions
from facilities burning RDF have recently
received attention due to the sampling and
analyses work performed at Ames, along with
the promulgated 1.5 ug/m» ambient air stand-
ard. A research contract, therefore, was
awarded to identify the sources of lead in
RDF and the chemical composition of these
lead contributing products (37). These
sources will be traced through the process-
ing stages, and any phenomenon of trans-
formation in combustion will be identified
14
-------
and related to emissions. The infromation
developed will aid in determining the seri-
ousness of this problem, and the applicable
control techniques.
CONCLUSION
Realizing that the many specific re-
search areas discussed in this overview will
be covered in greater detail later in this
symposium, some discussions were necessari-
ly brief. Interested parties are urged to
attend the individual presentations, and
for more detail or reports, contact the EPA
project officers referenced and listed at
the end of this paper.
REFERENCES
1. Municipal Solid Waste Protocol, SCS
Engineers, Project Officer, Albert J.
Klee
2. Forecasting The Quantity and Composi-
tion Of Solid Waste, International Re-
search and Technology, Project Officer,
Oscar W. Albrecht
3. Development of A Solid Waste Question-
naire And Data Base, Project Officer,
Oscar W. Albrecht
4. Evaluation Of Impediments To Economi-
cal Resource Recovery, Mathtech, Proj-
ect Officer, Oscar W. Albrecht
5. Source Reduction of Product Packaging
from Households: An Evaluation of Pub-
lic Information Systems, Univ- of Ari-
zona, Project Officer, Haynes Goddard
6. User Charges For Managing Household
Wastes, Mathtech, Project Officer,
Oscar W. Albrecht
7. Evaluation Of The Seattle Variable User
Charge Experiment, University of Wash-
ington, Project Officer, Haynes C.
Goddard
8. Research Design For Freight Rate Study,
Dr. William Whitmore, Project Officer,
Oscar W. Albrecht
9. Economic Analysis Of Scrap Futures Mar-
kets for Stimulation OF Resource Re-
covery, Environmental Law Institute,
Project Officer, Oscar W. Albrecht
Of Interim Standard MSW Procedures,
ASTM, Project Officer, Carlton C. Wiles
11. Materials Recovery And Side By Side
Evaluation Of Selected MSW Preprocess-
ing Equipment, National Center for
Resource Recovery, Project Officer,
Donald A. Oberacker
12. Resource Recovery Unit Process Test
And Evaluation At Recovery I, National
Center for Resource Recovery, Project
Officer, Donald A. Oberacker
13. Preprocessing Equipment For Resource
Recovery, Midwest Research Institute,
Project Officer, Donald A. Oberacker
14. Field Test Evaluation Of Selected
Shredder Designs, Midwest Research In-
stitute, Project Officer, Donald A.
Oberacker
15. Comparative Study of Full Scale Air
Classifiers, Midwest Research Insti-
tute, Project Officer, Stephen C. James
16. Priorities Assessment Of Material nad
Energy Recovery Needs, National Academy
of Science, Project Officer, Oscar W.
Albrecht
17. Resource Recovery Of Glass and Plastic
Wastes, Pacific Environmental Services,
Project Officer, Stephen C. James
18. Status, Trends, and Impediments To Dis-
carded Tire Collection And Resource
Recovery, Urban Systems, Project Offi-
cer, Haynes C. Goddard
19. Technical Evaluation And Market Study
On The Use Of Waste Glass Mixtures In
Structural Clay Brick Manufacture,
Occidental Research Corporation, Proj-
ect Officer, Donald A. Oberacker
20. Preparation Of Refuse Derived Fuels,
National Center for Resource Recovery,
Project Officer, Carlton C. Wiles
21. Effects Of Burning Densified RDF In
Stoker Boilers, Systech, Inc., Project
Officer, Carlton C. Wiles
22. Fundamental Considerations In The Pro-
duction Of d-RDF, University of Cali-
fornia, Carlton C. Wiles
10. Compilation, Development, And Testing 23. Processing And Preparation Of RDF For
15
-------
Cement Kiln Firing, Maryland Environ-
mental Services, Project Officer,
Donald A. Oberacker
24. Concepts For Improving The Fuel Quality
Of RDF, University of Dayton Research
Institute, Project Officer, Albert J.
Klee
25. Small-Scale, Low Technology Study, SCS
Engineers, Project Officer, Donald A.
Oberacker
26. Source Separation/Resource Recovery
Compatibility, Gilbert Associates,
Project Officer, Stephen C. James
27. Methane Recovery From Landfills, Lock-
man & Associates, Project Officer,
Stephen C. James
28. Gas And Mechanical Mixing Of Sewage
Sludge/MSW Mixture, Systech Inc.,
Project Officer, Stephen C. James
29. Methane And Volatile Fatty Acids From
Chemical Treatment Of Agricultural
Residues, Louisiana State University,
Project Officer, Charles J. Rogers
30. Acid Hydrolysis Of Cellulose, New York
University, Project Officer, Charles J.
Rogers
31. Upgrading And Mehtods Of Stabilization
Of Pyrolvtic Oils, Georgia Tech Re-
search Institute, Project Officer,
Charles J. Rogers
32. Development Of Methods For Stabiliza-
tion Of Pyrolytic Oils, Atlanta Univer-
sity, Project Officer, Charles J.
Rogers
33. Air Pollution Control Technology For
Wastes As Fuels Processes, Pedco En-
vironmental, Project Officer, Donald
A. Oberacker
34. Assessment Of The Impact Of Resource
Recovery On The Environment, Mitre,
Project Officer, Albert J. Klee
35. Environmental Assessment Of Wastes As
Fuels Processes, Midwest Research In-
stitute, Project Officer, Carlton C.
Wiles
36. St. Louis Bacteria/ Virus Testing, Mid-
west Research Institute, Project Offi-
cer, Carlton C. Wiles
37. Lead And Other Trace Metals In MSW,
York Research Corporation, Project
Officer, Donald A. Oberacker
16
-------
AN OVERVIEW OF THE DEPARTMENT OF ENERGY'S RESEARCH, DEVELOPMENT
AND DEMONSTRATION PROGRAM FOR THE RECOVERY OF
ENERGY AND MATERIALS FROM URBAN WASTE
Alan S. Cohen
Energy and Environmental Systems Division
Argonne National Laboratory
Argonne, Illinois 60439
ABSTRACT
The Urban Waste Technology Branch of the Department of Energy is conducting a com-
prehensive and diversified research, development, and demonstration program to foster
the widespread use of urban waste as a source of energy and materials. This program is
directed at improving and developing technologies that reprocess waste into fuels, metals,
glass, paper, ammonia, glucose, fertilizers and other energy intensive products. Non-
technical issues relating to institutional, socioeconomic, and legal constraints to re-
source recovery are also being investigated.
This paper briefly describes the technological options available in the field of
resource recovery. Some of the approximately 80 projects comprising the Urban Waste
Technology Branch's research, development, and demonstration program are discussed and
related to the overall objectives of the branch. To facilitate this presentation, pro-
jects are grouped as being primarily related to mechanical, thermal or biological pro-
cesses.
To date the majority of research and development funding has been devoted to bio-
logical processes including anaerobic and enzymatic digestion, energy production and
conservation in water and wastewater treatment, and energy generation and recovery in
sanitary landfills. However, demonstration activities, representing slightly less than
half the program effort, are focused on the thermal/mechanical systems.
DOE activities are coordinated with the EPA and other federal, state and local
efforts in resource recovery to ensure the development of an efficient and comprehensive
national resource recovery program. If successful, approximately 3% of the nation's
energy needs could be supplied by reprocessing wastes.
INTRODUCTION
The Urban Waste Technology (UWT)Branch
of the Department of Energy (DOE) has as
its primary objective the widespread use of
urban waste as a source of energy or ma-
terials, thereby conserving limited fossil
energy and virgin materials resources. A
related objective is to achieve energy con-
servation through reduced energy consump-
tion in the processing of urban wastes. In
pursuit of these objectives the projects of
the UWT Branch focus on five specific areas.
Three are technical process options, me-
chanical, thermal, and biological processes;
the fourth is directed at nontechnical (e>§.,
institutional) aspects of implementation;
and the fifth provides general program
support.
The DOE's authorization for this ef-
fort is contained in the Department of
Energy Authorization Act (1). This act
expands DOE's scope of effort beyond the:
research, development, and demonstration
(RD&D) functions authorized in the Federal
Nonnuclear Research and Development Act
(2). The DOE program in the field of re-
source recovery is coordinated with the
U.S. Environmental Protection Agency's
(EPA) and Department of Commerce's pro-
grams authorized under the Resource Con-
17
-------
servation and Recovery Act (3). This com-
bined federal effort is designed to ensure
that wastes are managed in a manner that
will protect public health and the environ-
ment and to conserve valuable material and
fossil energy resources. Following a brief
overview of activities in resource recover};
this paper will focus on the research, de-
velopment, and demonstration activities of
the WT Branch of the DOE.
OVERVIEW
The recovery of energy and materials
from waste in centralized processing plaits
is still in the early stages of commercial
application (4). In 1977 less than 1% of
the nation's total waste stream was being
recovered. An important reason explaining
this low level of recovery is that some of
the necessary technologies are still in
relatively early stages of demonstration
and commercial application. Also, some
technologies require further development
before they will be ready for demonstration.
Furthermore, those technologies being im-
plemented need operating histories to doc-
ument their technical and economic feasi-
bility.
Figure 1 shows the range of technical
RD&D options in resource recovery. Three
broad technologies are available to recover
energy and energy-intensive materials from
urban wastes: mechanical, thermal and bio-
logical.
Mechanical processing separates wastes
into various components including metals,
glass, and a refuse derived fuel (RDF).
Generally, a mechanical process is a pre-
liminary step to the thermal and biological
technologies. The metallic components,
TECHNOLOGY
TECHNIQUE
PRODUCT
END USE
AS_RECEIVED
WASTE
i— DIRECT-
HEAT
REFUSE DERIVED
FUEL -(RDF)
L- CO-FIRE STEAM
FUELS
_SEWAGE
SLUDGE
OTHER
ANAEROBIC
DIGESTION
ENERGY FROM
LANDFILLS
DISTRICT HEAT
PROCESS HEAT
ELECTRICITY
GAS TURBINE
FEEDSTOCKS CHEMICALS
MATERIALS
REUSE
THERMAL SOIL
CONDITIONER
SYNTHETIC
NATURAL
GAS
Fig. 1 Technical RD&D options
18
-------
glass, and paper fibers are recycled to
displace virgin materials..
In general, mechanical processes are
employed for size reduction and separation
by size, weight, shape, density, and other
physical properties. A typical processing
line would utilize shredding for size re-
duction of raw refuse, followed by some
form of air classification to separate the
particles into a light (organics) and a
heavy (inorganics) material stream. The
light fraction, without further processing,
has generally come to be known as fluff KDF.
A demonstration unit sponsored by the
EPA to produce RDF at St. Louis proved the
basic feasibility of the mechanical separa-
tion processes, transport and storage tech-
niques, and the combustion of fluff RDF to
replace 5% to 27% of the pulverized coal
used in suspension-fired utility boilers.
However, the refinement of equipment com-
ponents and the technical and economic op-
timization of the basic technology still re-
quires a great deal of work.
There is one operating facility recov-
ering and using fluff RDF on a daily basis.
This facility has encountered a series of
economic and technical problems. A second
facility is entering its first year of pro-
duction evaluation prior to finalization of
fuel sales contracts. Several facilities
are in a shakedown phase or under construc-
tion.
The preparation of densified RDF by
pelletizing, briquetting, or extruding is
now being explored and evaluated, and is
particularly adapted for stoker and sprea-
der-stoker furnaces. However, it has not
been demonstrated commercially, and the
costs, handling characteristics, and firiig
characteristics must yet be evaluated.
The anticipated advantages of densified
RDF are greatly improved storage and trans-
portation characteristics.
The production of dust RDF (particles
smaller than 0.15 millimeter) is being de-
veloped in a proprietary pilot-plant pro-
cess. After adding an embrittling chemical,
coarsely shredded waste is pulverized to a
dust-like consistency. The resulting dust
RDF has a higher Btu content than fluff
RDF along with greater density, homogeneity
and decreased moisture content. In addi-
tion, dust RDF may be capable of direct
co-firing with fuel oils. However, the
dust-like composition necessitates special
handling to minimize the danger of an ex-
plosion.
A "wet" mechanical separation process
utilizes hydrapulping technology adapted
from the pulp and paper industry to reduce
raw waste to more uniform size and consis-
tency, followed by a centrifugal, liquid
cyclone process for separating the pulped
mass into light and heavy fraction. The
original EPA sponsored pilot plant at
Franklin, Ohio, is still operating although
it no longer produces low grade fiber for
a roofing felt plant as originally intended
After successful test burns of the 50% or-
ganics, 50% moisture pulp, a full scale
facility designed to burn the pulp to re-
cover steam is now under construction.
Ferrous metal recovery systems are the
most advanced material recovery systems;
paper fiber recovery by both wet pulping
and dry processes has been demonstrated
successfully, and aluminum and glass re-
covery has been demonstrated with limited
success. Efforts are planned to improve
system efficiencies in terms of energy use,
quantity, and quality of material recovered
Combustion techniques burn waste for
the recovery of heat energy. Waterwall
combustors are the most technically deve-
loped energy recovery systems and employ
special grates to burn "as received" urban
waste and recover steam either at saturated
or superheated conditions. Over 250 plants
are operating in Europe and Japan. There
are seven plants operating in the United
States, although three of these were orig-
inally constructed as incinerators. Only
one plant has developed an effective steam
marketing plan and six years after start-
up, has begun steam sales. Worldwide,
there have been a number of technical pro-
blems with the control of corrosion and
erosion being the most serious. The most
recent European designs have solved these
problems but at an increased capital cost.
The more popular U.S. development
seems to be the recovery of RDF for sale
to coal-using facilities. With modifica-
tions, existing boilers can use RDF as a
supplemental fuel. Most development has
been aimed at the large suspension fired
utility boiler and, while test burns have
been encouraging, technical problems have
developed. These are related to burning
characteristics, slagging and environmental
19
-------
control equipment performance. All can be
solved. However, upgrading control devices
such as electrostatic precipitators will re-
quire significant increases in the capital
costs of the system.
Another variation being demonstrated
is the combustion of RDF in a dedicated
boiler as a principal fuel. Normally the
boiler is of spreader-stoker design with
some consideration given to the use of
fossil fuels such as high sulfur coal as a
load leveler and steam production stabilizer.
The only available small-scale system
is a packaged two chamber incinerator with
waste heat recovery. This technique is
practical at the 25 to 100 tons per day
(TPD) scale. In these units partial oxi-
dation occurs in the first section of the
unit and causes a portion of the waste
material to degrade and give off combustible
gases. These gases, as well as products of
combustion and particulate from the first
chamber, flow to a second chamber where
they are combusted with excess air and a
natural gas or oil pilot flame. The com-
bustion products then flow through appro-
priate heat transfer equipment to produce
steam, hot water, or hot air. Four small
cities and more than 60 industrial plants
use the technique with heat recovery equip-
ment today.
Thermal gasification and pyrolysis
systems are also under development with
several systems approaching the full scale
demonstration stage. Specific discussion
is not included on individual techniques
since they are still in the developmental
stages. Fuels from these processes in-
clude gases, oils and chars.
Biological techniques use living or-
ganisms to convert organics into useful
energy forms. These processes are in the
developmental stages. DOE is sponsoring
an anaerobic digestion process that converts
the organics in urban waste to methane under
controlled conditions. This process is at
the proof-of-concept stage and is not ex-
pected to be commercialized until the late
1980s. Another bioconversion process to
recover methane gas from existing landfills
is at or near commercialization. Because
of the explosive nature of the gas, in-
creasing attentio'n has been paid to con-
trolling the migration of the gas. Sub-
sequently, the use of the gas for energy
purposes developed. Today there is one
operating site, two sites in development,
one site under construction and 10 sites
in an advanced planning stage to utilize
the recovered gas. It is estimated that
one trillion cubic feet of methane is po-
tentially recoverable from existing land-
fills with 55 billion cubic feet of pipe-
line quality methane available yearly just
from the 100 largest landfills.
Generally, the mechanical, thermal and
biological resource recovery technologies
can be ranked in order of their stages of
development as shown in Table 1. The DOE
RD&D program addresses all of these tech-
nologies as will be discussed in the next
section.
Table 1. Developmental Stage of Resource
Recovery Technologies
Level of
Develop-
ment
High
Low
Technology
Ferrous Metal Recovery
Anaerobic Digestion of
Municipal Liquid Wastes
. Waterwall and Modular Con-
trolled Air Combustors
. Coarse, Fluff, and Wet
Pulped RDF
. Paper Fiber Recovery
. Landfill Gas Recovery
. Glass, Aluminum and Other
Nonferrous Metal Recovery
. Dust and Densified RDF
. Pyrolysis and Gasifiers
. Anaerobic Digestion of
Solid Waste
. Enzymatic and Fungal
Synthesis
20
-------
Figure 2 shows the range of nontechni-
cal KD&D options in resource recovery.
Three broad categories are of concern: in-
stitutional, socioeconomic, and legal. All
of these are interrelated with the solutions
to nontechnical problems often more diffi-
cult to attain than those to technical pro -
blems.
Institutional issues are many and com-
plex. Perhaps the principal issue is that
cities are conservative, risk avoiders and
politically are more concerned that their
urban wastes disappear than they are with
resource recovery. Additional problems in-
volve waste stream control, difficulty of
regionalizing waste processing, and to this
point, lack of control of land disposal of
wastes although regulations are now man-
dated and in development. The difficulty
of obtaining new landfill sites, however,
is a positive influence, particularly in
more populated areas.
To be economically viable, waste to
energy plants depend on two major sources
of income: the sale of the energy product
and a fee to dispose of waste in the plant
(tipping fee). Additional income can be
realized from the sale of recovered mate-
rials. Where landfill (tipping fees) or
energy costs are high, waste based systems
can compete with fossil fuels. Of the
plants in operation only one enjoys eco-
nomic success.
Furthermore, because of the high front-
end costs and the risks associated with
using a new technology, financing these
facilities may be a problem, particularly
for local government units. Marketing the
energy and materials is essential for the
economic viability of these technologies.
However, many local governments do not have
marketing expertise, and often, governments
cannot legally enter into the long-term
contracts that are essential for using
CATEGORY
INSTITUTIONAL
ISSUE
REGULATIONS
GOVERNMENT INTERACTION
L- PUBLIC/PRIVATE INTERFACE
—[ SOCIOECONOMIC
LEGAL
FINANCING
BENEFIT/COSTS
MARKETS
MANPOWER
WASTE CONTROL
OWNERSHIP
CONTRACTS
OUTPUT
ENVIRONMENTAL/PRICE/IRS
REGIONALIZATION/COORDINATION
INCENTIVES/PROCEDURES
EVALUATE ALTERNATIVES
PRICE SUPPORTS
TRAINING/EDUCATION/TECHNOLOGY
TRANSFER
MODEL CODES/GUIDELINES
GUARANTEES
Fig. 2 Nontechnical RD&D options
21
-------
these capital-intensive technologies. In
addition, electric utilities, which poten-
tially represent a large market for refuse
derived fuel, do not have an economic in-
centive to use this fuel because of fuel-
cost pass through provisions and contracted
long-term fuel supplies.
Innovation has also occurred in the
development of institutional arrangements
to accomplish energy and materials recowry.
These innovations range from the creation
of statewide authorities (such as those in
Connecticut and Delaware), which are charged
with developing and implementing high pri-
ority projects, to formation of not-for-
profit public corporations (such as the
Nashville Thermal Transfer Corporation),
which operate much like utilities. These
state-wide and regional bodies have facili-
tated the organization of regional recovery
facilities by reducing some of the complex-
ities normally encountered in arriving at
intermunicipal agreements. In addition,
several different approaches have been
used to bring together municipalities (the
generators of waste), private industry
(the suppliers of recovery technologies),
and the energy and materials market (the
buyers of recovered resources) as partners
in joint efforts to accomplish the com-
bined public purpose of energy conservation
and waste disposal.
Several innovative approaches to pro-
ject financing have been used. In re-
sponse to local hesitancy or inability
to issue general obligation bonds for re-
covery plants, various forms of revenue
bond financing have been developed. These
financing approaches have been extremely
beneficial because they have forced pro-
ject sponsors to thoroughly investigate
the anticipated performance of proposed
recovery systems, and to assure that pro-
ject revenues are both secured and suffi-
cient to cover all operating and debt ser-
vice costs.
DOE activities in the nontechnical
area are to assess economic, social, in-
stitutional and legal resource recovery
barriers, and also to provide implemen-
tation guidance and direct technical and
financial assistance to nonfederal enti-
ties , including state and local govern-
ments, and private enterprises.
The nontechnical RD&D program may
include evaluations of public attitudes
and institutional barriers to resource re-
covery, and performance of benefit/cost
studies to evaluate the total social costs
of resource recovery and disposal options
including the cost of environmental damages
and externalities such as balance of pay-
ments and national defense. In addition,
the program may also include the develop-
ment of the following: new markets for
recovered resources, product specifications
of recovered materials to enhance their
marketability, methods to properly compare
resource recovery with traditional waste
disposal options, and methods to estimate
the life cycle costs of resource recovery
systems.
These RD&D efforts are being augmented
by a diversified program of economic in-
centives, including grants, contracts,
cooperative agreements, loans, loan guaran-
tees, and price support. Currently, loan
guarantee and price regulations are being
developed. An active program of training,
technical assistance and information trans-
fer is also being initiated. This includes
preparation of case studies, conducting
workshops and seminars, and developing
university and apprenticeship programs.
Although a significant effort is being
made by DOE in the nontechnical area, the
focus of this paper is on DOE activities
in technology RD&D. The previous discus-
sion of nontechnical concerns is included
because this type of RD&D is often ignored
or considered inconsequential. However,
the need for innovation and research in
the nontechnical area is more important
than additional development of some tech-
nologies.
DOE TECHNOLOGY RD&D PROGRAM IN
RESOURCE RECOVERY
The Urban Waste Technology Branch of
DOE has, over the past few years, funded
approximately 80 RD&D projects with about
60 still active. Figure 3 shows a rough
distribution of funding, by percentage,
for various types of efforts. Classifi-
cations of some projects are subjective
and many projects have components that
could be classified differently. There-
fore, the distribution presented should
be viewed as a general indication of the
emphasis of DOE's past RD&D efforts in
resource recovery.
In addition, this distribution repre-
22
-------
sents only a narrow view of DOE's program.
The program emphasis has been shifting as
priorities change and advances are made in
the state of the art. DOE has contracted
with the MITRE Corporation to update its
1975 RD&D priority evaluation (5). This
document will be used to develop future
RD&D priorities for funding.
Future funding of research and de-
velopment projects is expected to remain
constant in real terms over the next few
years, while funding for demonstration
projects will increase significantly.
This prediction is based on congressional
support for demonstration projects as
manifested in authorities granted DOE
for loan guarantees for demonstration pro-
jects (6) and direct financial support for
municipal demonstration projects (7).
With these caveates in mind, Figure 3
can be useful as a basis of discussion on
the types of projects being funded by DOE.
Specific examples are discussed below.
Mechanical Processes. Waste quantifi-
cation and characterization studies, for
the purposes of this paper, are classified
as mechanical research and development
projects. This was done, in part, because
the sizing, separation, and benefaction
functions of mechanical systems are often
determined by the type and quantity of
waste available.
The quantity of wastes generated is a
function of, among other factors, producer
packaging practices, consumer buying pat-
terns, demographic and economic character-
istics of a region, season of the year
and day of the week. Proper measurement
and estimation of wastes generated in a
region would improve process selection and
design decisions.
The composition of the waste will de-
termine its energy content, pollution
characteristics, and corrosion and abrasion
potential. The types of wastes generated
in an area affect the suitability of many
processes for converting the wastes to
energy. In addition, proper character-
ization of the waste can improve equipment
design by providing basic data needed to
evaluate equipment performance.
Standard practice has been to use
either national-average per-capita waste
generation rates and composition factors
or costly and time consuming weighing and
sorting procedures to estimate waste quan-
tities and compositions. Improved estimat-
ing or measurement procedures are sought in
a number of DOE R&D projects.
Franklin Associates is attempting to
develop an improved procedure for estimating
solid waste generation rates based on land-
use zoning in a region. If successful, the
land-use based waste generation factors
would be almost as simple to use as national
average factors but would result in esti-
mates that are based on local demographic/
economic development patterns.
Waste characterization is also an im-
portant component of most of the demonstra-
tion projects funded by DOE. Although
innovative methods for characterizing waste
are not being sought, the data produced and
experiences gained will provide information
that could improve these activities in the
future.
The National Bureau of Standards (NBS),
under sponsorship of DOE and EPA, began a
study in 1977 to develop a method for mea-
suring the gross caloric value of raw refuse
and various types of RDF. In addition, the
study is evaluating the moisture, ash, car-
bon, sulfur and chlorine characteristics of
the waste samples. NBS is also evaluating
the homogeneity and preparation required
for 'bomb calorimetric experiments utilizing
two bomb calorimeters with order of magni-
tude sizes such as 2.5 g and 25 g.
SO-
tS 40
3
CM
o 30
<#
§
• 20
C
a)
Ji 10
Feasibility/Design
Biological R^
_ Thermal
Mechanical
ESS
Research and Developtnent-j-Demonstration
Fig. 3 UWT funding distribution
23
-------
Results on a limited number of samples
of Teledyne National Pellet and Combustion
Equipment Associates ECOFUEL-II "demon-
strated that RDF samples can be processed
for bomb calorimetric experiments to pro-
duce results with a precision (standard
deviation of a measurement) approximately
equal to or better than 1%" (8). The heat
value of the RDFs ranged between 25.20 and
21.93 MJ/kg (10,835 and 9427 Btu/lb). The
ash and sulfur contents were about 12% and
1%, respectively. The degree of prepara-
tion required in order to obtain reprodu-
cible calorific results will be determined
when concurrent tests are made with ,two
sized calorimeters.
The National Center for Resource Re-
covery has been contracted to assemble
and test various components of waste pre-
processing equipment. Studies include
ferrous and other material recovery, RDF
preparation, removal of noncombustibles
during pneumatic transport of RDF, energy
recovery from air classified heavies, and
development of a mobile RDF plant. Systems
Technology Corporation (SYSTECH) has been
contracted to test gas, mechanical and
pump mixing, both separately and in com-
bination, of a high solids concentration
in an existing anaerobic digester. SYSTECH
is also studying the performance charac-
teristics of a coarse trommel in another
project.
In addition, an innovative approach
for size reductions and separation of solid
waste, using a steam cannon similar to the
technology used to make "puffed" foods, is
being developed and tested by Burke, Davoud
and Associates. Recently, tests using the
steam connon on paper feedstock were con-
ducted, and additional tests on other con-
stituents .of urban waste and mixed refuse
are scheduled. Hydrolysis tests on sam-
ples of cannon discharges are being con-
ducted by the U.S. Army Laboratory in
Natick, Mass. A major objective of this
effort is to determine the optimum cellu-
lose particle size for cellulosic enzymatic
hydrolyzation.
If this technique is successful, it
could improve the economics, energy effi-
ciency, and digestion efficiency of ana-
erobic digestion processes such as the one
being tested in Pompano Beach, Florida.
A continual effort is expected to
develop, improve and determine performance
characteristics of waste preprocessing
equipment. Efficient and effective waste
preprocessing is required to optimize the
performance of many resource recovery fa-
cilities.
Thermal Processes. The majority of
thermal process projects involve the com-
bustion of municipal wastewater sludges
in combination with solid waste or alone to
produce energy. Since these projects were
recently awarded as part of two program
research and development announcements,
and consequently, are just being initiated,
no results are available.
Two other thermal process projects
being conducted by Battelle Columbus Labo-.
ratory and Stanford University are inves-
tigating the technical and economic via-
bility of fluidized bed combustion of solid
waste. Another study by Monsanto Research
Corporation will be testing the combustion
properties of an oil/refuse fuel prepared
using a wet pulping process similar to the
one used in Franklin, Ohio.
Although approximately 5% of the RD&D
funds are devoted to research and develop-
ment of thermal systems, the majority of
the 28 demonstration projects funded by
DOE (see Figure 4 for the locations of these
demonstration projects) involve combustion
systems such as waterwall incinerators,
multi-chamber starved-air combustors and
multi-hearth furnaces, or the projects will
involve the production of an RDF for use
in a dedicated boiler or cofired with coal.
At least three of the projects are seriously
considering the demonstration of a pyroly-
sis facility. The Dutchess County, New
York, project has advanced to the engi-
neering design stage and a full scale de-
monstration of the Union Carbide PUROX
System may occur in the early 1980s.
Future research will focus on improv-
ing the operating characteristics of laige-
scale facilities, developing viable pyroly-
sis, gasifier and fluidized bed techniques,
improving the handling, storage and com-
bustion properties of RDF, developing and
evaluating environmental controls for re-
source recovery facilities, and developing
small-scale systems.
Biological Processes. The current
DOE R&D program for biological processes
24
-------
consists of three interrelated program areas
anaerobic and enzymatic digestion of solid
wastes and municipal sludges; production
and conservation in wastewater and water
treatment plants; and production and ex-
traction of methane from sanitary landfills.
By far the largest share of DOE's R&D
dollars have been allocated to anaerobic
and enzymatic digestion projects. More
than half of the biological processes R&D
funds and about 20% of DOE's Urban Waste
Technology RD&D funds over the last few
years have gone to two major projects.
The first of these projects involving
enzymatic hyrolysis has been ongoing since
1976 at the U.S. Army Natick Laboratory.
The objective of this effort is to convert
cellulose in urban wastes to glucose with
further fermentation to alcohols and other
energy products. Numerous biological pro-
jects should benefit from the information
generated by this project. However, enzy-
matic hydrolysis is not expected to be an
economically viable option for converting
waste to energy or energy intensive pro-
ducts in the near future (9).
In 1977 the Energy Research and De-
velopment Administration funded Waste Man-
agement Corporation to design and build a
proof-of-concept anaerobic digestion faci-
lity. The concept is based on laboratory
investigation by Dr. John Pfeffer of the
University of Illinois (10). The facility
was built in Pompano Beach, Fla., and is
referred to as the Refuse Conversion to
Methane (RefCOM) project.
This experimental facility is designed
for a capacity of 100 tons per day of raw
refuse and is expected to produce 0.6 mil-
lion cubic feet of gases (CH^ and C02) per
day. The objectives of the experimental
program are to establish information on
quality and quantity of the gas produced,
establish optimum process operating para-
meters in both mesophilic and thermophilic
modes, and evaluate the technical relia-
bility of the technique.
Design and construction of the faci-
lity was completed in January, 1978, at a
final cost of $3.6 million. Waste Manage-
ment Corporation has been contracted by
DOE (the UWT Branch assumed responsibility
DOE DEMONSTRATION PROJECTS
25
-------
for the project when DOE was formed) to con-
duct a two to four year experimental program.
The RefCOM start-up program began in
March 1978. Since that time Waste Manage-
ment Corporation has been correcting a
number of nagging mechanical problems pri-
marily dealing with the front-end and di-
gester feed systems (11). Some redesign
and fabrication of these systems are being
made. The experimental program will begin
after the start-up problems have been
solved.
DOE has a number of programs underway
that support the RefCOM project. These in-
clude the digester mixing and steam cannon
projects discussed previously and a project
recently awarded to SYSTECH to analyze me-
thods for recovering energy from the diges-
ter sludge.
Of the many anaerobic digestion pro-
jects underway, two are: (1) the design
and construction of an anaerobic digester
to convert fecal material and bedding from
various cages, enclosures, runs, etc. in
the Philadelphia Zoo to a medium Btu fuel
gas and fertilizer; and (2) the development
of a small-scale on-site energy system that
incorporates wind energy, solar energy, and
bioconversion of wastes to produce a stor-
able fuel gas. The latter project is being
conducted at California State University,
Northridge, California.
DOE has an active program to promote
energy production and conservation in
wastewater and water treatment plants. The
anaerobic digestion of sludges at municipal
treatment plants is an established techno-
logy. The gases produced, however, are
often flared resulting in lost energy.
Furthermore, the energy consumption in
water and wastewater treatment facilities
is not known. Estimates varied from 0.25
to 1.0 quads in 1978.
Since 1976 Oak Ridge National Labora-
tory has been developing and analyzing a
packed bed anaerobic digester, referred to
as the ANFLOW pilot plant, that might re-
duce the cost of sewage treatment plants
and the waste sludge disposal problem. The
5,000 gallon-per-day pilot plant is cur-
rently producing methane from sewage sludge.
Tests at a commercial scale treatment plant
with a 50,000 gallons per day capacity are
planned for fiscal year 1979.
A laboratory and small pilot scale pro-
gram to enhance anaerobic digestion by the
addition of activated carbon is being con-
ducted by the Battelle Pacific Northwest
Laboratory. It was found that small quan-
tities of activated carbon do not substan-
tially enhance the production of methane in
unstressed digesters. However, it does in-
crease gas production in stressed digesters.
(12)
In the area of energy production and
conservation in water and wastewater treat-
ment, DOE and EPA are funding a number of
joint ventures. For example, they have a
contract with Hittman Associates to pre-
pare an energy assessment procedure man-
ual. The manual will provide guidance
to planners, design engineers, plant oper-
ators, federal agencies and other interested
parties for energy recovery and conservation
in the design, construction and operation
of municipal wastewater collection and
treatment facilities.
Another project, being conducted by
the Reedy Creek Improvement District in
Florida, is evaluating the use of water
hyacinth in an advanced wastewater treat-
ment lagoon to remove nutrients from the
water in an energy efficient manner. Future
efforts may evaluate the feasibility of
harvesting the hyacinth periodically and
anaerobically digesting it to produce a
fuel gas and a fertilizer. The EPA and
DOE are also assessing the energy con-
servation aspects of a solar assisted
wastewater treatment plant in Wilton,
Maine. In the future, DOE efforts in the
area of energy production and conservation
in wastewater and water treatment plants
will probably involve more joint ventures
with the EPA.
The DOE effort in energy generation,
extraction and utilization from sanitary
landfills began in March, 1978, with a
symposium in Laurel, Md., on the utili-
zation of methane generated in landfills.
The symposium provided one of "the first
opportunities for everyone doing work
related to the utilization of landfill
methane to meet and exchange information"
(13). A second symposium on landfills is
scheduled for March, 1979.
In July, 1978, DOE/UWT issued a pro-
gram research and development announcement
for the development and analysis of alter-
26
-------
native techniques to utilize combustible
gas produced in existing sanitary landfills.
Three projects were selected for funding as
a result of this solicitation.
The Los Angeles County Sanitary Dis-
trict was funded to evaluate and implement
a system to provide landfill gas (40-50%
014) to Rio Hondo College located near the
Puente Hills landfill (14). The gas is
expected to replace the interruptible por-
tion of the college's present natural gas
requirements. The project includes a
boiler conversion study to determine the
practicality and cost of adapting the
boilers to use the landfill gas. A gas
delivery criteria to ensure proper perfor-
mance of the modified boiler will be estab-
lished. If either the modification costs
exceed the estimated benefits or the gas
delivery criteria can not be satisfied,
the project will be terminated. Infor-
mation generated from the study concern-
ing the use of landfill gas as a substi-
tute for interrupted natural gas is expec-
ted to have widespread value.
Pacific Gas and Electric Company (PG&E)
was selected to conduct a comprehensive
analysis of the gas stream and all conden-
sates from the City of Mountain View's
landfill (15). The purpose of the inves-
tigation is to obtain information on the
potential sources of corrosion caused
by landfill gas and to foster acceptance
of the use of landfill gas by industry.
Air intrusion will be evaluated for the
major components (i.e., nitrogen and oxy-
gen) and the potential for CC>2 utilization
from landfill gas will be evaluated. The
PG&E project is significant because it
addresses many generic issues and problems
that have been limiting the use of landfill
gas.
The New York State Energy Research
and Development Authority was selected to
design, develop, install and operate gas
engine generation equipment to demonstrate
direct use of the landfill gas for supply-
ing electricity (16). The electricity will
be used by several New York City Department
of Sanitation facilities including the
maintenance building, tractor repair shop,
and roadway lighting. The data collected
will identify and focus on problem areas
in gas recovery and utilization. This
demonstration at the Fresh Hills site on
Staten Island will also provide information
on life cycle operation of the associated
system components, the reliability of gas
supply, the corrosive nature of the gas,
and the feasibility of large-scale landfill-
gas-electricity projects.
In addition to these three projects,
two of DOE's demonstration projects deal
with the extraction and use of landfill gas.
Adams County and Commerce City in Colorado
will be designing a facility to recover
methane from a network of sanitary landfills
as an energy source for several local in-
dustries, and Lochman and Associates will
be designing a system to recover methane
from landfills to generate electricity in
Glendale, Calif.
All of these programs focus on the use
of landfill gas as a boiler fuel without
upgrading. The gas is used in a boiler as
a substitute for natural gas or to produce
electricity. These projects are designed
to foster the use of landfill gas in the
near future.
In the long run, however, improved
methods and procedures are needed to op-
timize the generation of gas, improve
drilling, collection and upgrading prac-
tices and technologies, and devise economic
and safe ways to store and transport the
recovered gas. This must be done in an
environmentally acceptable manner and re-
sult in competitively priced products.
Toward these program goals a second
competitive solicitation in the area of
optimizing gas production will be issued by
Argonne National Laboratory early in 1979.
In addition, Johns Hopkins Applied Physics
Laboratory is preparing a workbook that
will address all the technical, institu-
tional, environmental, and legal issues
associated with methane recovery from
sanitary landfills. The workbook will also
describe the experiences of the approxi-
mately 14 landfill sites actively consider-
ing or implementing landfill gas recovery
programs.
Future methane from landifll RD&D
will be directed at major technical and
nontechnical problems. The level of
funding and its direction will, in part,
depend on the findings of current DOE
and other programs.
27
-------
SUMMARY
Although all of DOE's projects in the
field of resource recovery from urban wastes
could not be discussed in this paper an
attempt to demonstrate the breadth and
depth of the DOE/UWT program through selec-
ted examples of KD&D activities was made.
Both technical and nontechnical RD&D acti-
vities are being carried out to foster the
improvement and implementation of mechanical,
thermal and biological processes for repro-
cessing urban waste into energy and mate-
rials.
The DOE/UWT program is coordinated
with EPA and other federal, state and local
efforts to ensure the development of an
effective and comprehensive national pro-
gram to recover the energy and material
currently being discarded as waste. If
this program is successful, approximately
3% of the nation's energy needs could be
supplied by reprocessed wastes.
REFERENCES
1. Department of Energy Authorization Act
of 1978 — Civilian Application, Public
Law 95-238, February 1978.
2. Federal Nonnuclear Research and
Development Act of 1974, Public Law
93-577, December 1974.
3. Resource Conservation and Recovery Act
of 1976, Public Law 94-580, October
1976.
4. From (often verbatum) unpublished
planning document prepared by Urban
Waste Technology Branch of DOE, Urban
Waste Technology Program at Argonne
National Laboratory, and DSI Resource
Systems Group, Boston, Mass.
5. MITRE Corp., Energy Conservation Waste
Utilization Research and Development
Plan, July 1975.
6. Department of Energy Authorization Act
of 1978 — Civilian Application,
Public Law 95-238, Title II - General
Provisions, Section 19-Loan Guarantees
for Alternative Fuel Demonstration
Facilities, February 1978.
7. Department of Energy Authorization Act-
of 1978 — Civilian Application, Public
Law 95-238, Title IV-Establishment of
Financial Support Program for Munici-
pal Waste Reprocessing Demonstration
Facilities, February 1978.
8. Kirkland, D.R., et al., Summary of 1977
Fiscal Year Efforts to Establish Test
Procedures for the Determination of
the Gross Calorific Value of Refuse
and Refuse-Derived-Fuels by Oxygen
Bomb Calorimetry, National Bureau of
Standards, Draft Report, 1978.
9. Spano, L.A., Enzymatic Hydrolysis of
Cellulose to Glucose, Quarterly
Reports, U.S. Army Natick Research and
Development Command, Natick, Mass.
10. Pfeffer, J.T., and J.C. Liebman,
Energy from Refuse by Bioconversion,
Fermentation, and Residue Disposal
Processes, Resource Recovery and
Conservation, Vol. 1, pp. 295-313,
1976, and Biological Conversion of
Organic Refuse to Methane, Vol. I and
II, Final Report No. ERDA/NSR/RANN/
SE/AER 73-07872/FR/76/4, Department
of Civil Engineering, University of
Illinois, November 1976.
11. Waste Management, Inc., RefCOM -
Status Report Covering the Start-Up
Phase, March 15 - October 1978. To be
published.
12. Ahlstrom, S.B., and R.R. Spencer,
Assessment of Powdered Activated
Carbon Addition to Anaerobic Digestion
at Salt Lake City, Utah, Pacific
Northwest Laboratory, Richland, Wash.,
April 1978, and monthly progress
reports submitted to DOE/UWT.
13. Proceedings^of a Symposium on the
Utilization of Methane Generated in
Landfills,' sponsored by U.S. Depart-
ment of Energy, hosted by Applied
Physics Laboratory, The Johns Hopkins
University, Laurel, Md., March 1978.
14. Proposed Delivery of Puente Hills
Landfill Gas to Rio Hondo College,
proposal submitted to the DOE in
response to PRDA #EM-78-D-01-5153 by
the Los Angeles County Sanitary
District, August 1978.
28
-------
15. Utilization of Landfill Gas from
City of Mountain View, proposal
submitted to the DOE in response to
PRDA #EM-78-D-01-5153 by Pacific Gas
and Electric Company, August 1978.
16. Landfill Gas-To-Electricity Conver-
sion Project, proposal submitted in
response to PRDA #EM-78-D-01-5153 by
New York State Energy Research and
Development Authority, August 1978.
Dr. Cohen manages the Urban Waste
Technology program at Argonne National
Laboratory, which has provided techni-
cal support to DOE's Urban Waste Tech-
nology Branch during the past year.
This paper reflects his understanding
of the DOE/UWT program and does not
necessarily reflect the opinions and
policies of DOE.
29
-------
SOLID WASTE DISPOSAL RESEARCH ACTIVITIES
OF THE FEDERAL GOVERNMENT IN CANADA
H. MDoij, P. Eng.
Waste Management Branch
Environmental Impact Control Directorate
Environmental Protection Service
Environment Canada
Ottawa, Ontario
K1A 1C8
ABSTRACT
In Canada the Federal government's role is principally to develop solid waste management
technology, to generate and evaluate technical solid waste management information, and
to develop new or improved guidelines or codes of good practice relating to various
waste management aspects. These activities are conducted on a national basis for the
benefit of the industry and regulatory agencies.
Solid waste research activities in Canada are being undertaken primarily at the federal
government level. Current research activities, relating to land disposal in particular,
are underway in each of the following study areas: landfill leachate research; municipal
waste disposal site selection; landfill field investigations; co-disposal studies; land-
fill gas research; and the NATO/COS landfill project. This paper describes Environment
Canada's current research proj ects in these areas of interest.
INTRODUCTION come to grips with the problems facing
both the private sector and the regulatory
Solid waste research activities in Canada agencies. Some provincial governments are
are being undertaken mainly at the federal actively contracting out basic research
level of government, but also at the studies or undertaking similar in-house
provincial government level, in order to investigations, although most others are
30
-------
simply studying localized problems of
disposal or local potentials for resource
and energy conservation. The bulk of the
research effort in Canada is undertaken by
the federal Department of the Environment
(Environment Canada).
The Waste Management Branch of Environment
Canada is very active in the resource
conservation and the municipal and
industrial waste management areas of
responsibility. However, the constraints
of this presentation do not permit a
complete disclosure of all our activities.
Therefore, I have chosen the focus of this
presentation to be the activities of our
Branch relating to the disposal of wastes
onto land.
These activities will be briefly described
under the following headings:
1. landfill leachate research;
2. municipal waste disposal site
selection;
3. landfill field
investigations;
4. co-disposal studies;
5. landfill gas research; and,
6. the NATO/CCMS landfill
project.
LANDFILL LEACHATE RESEARCH
Leachate Production
Although a significant amount of research
work has been accomplished to date on the
subject of leachate generation, or
production, at municipal solid waste
landfills, there are still many basic
questions left unanswered about the actual
process of leaching. Leachate production
is generally an estimated variable,
never really measured directly at a
landfill site.
and
The ability to predict leachate quantity
and quality is probably the most necessary
feature required for the design of
leachate control systems, including
natural attenuation and leachate treatment
systems. All landfill leachate control
systems design work relies heavily on an
accurate input function describing the
leaching behavior of the landfill. An
estimate is only as good as the
"rule-of-thumb" used, and therefore, we
have initiated an attempt to develop a
simple solid waste leaching model at the
University of British Columbia in
Vancouver.
A good quantative, descriptive and
predictive model of the leaching of solid
waste landfills and of the mechanisms
associated with the leaching process may
be important to assess the environmental
impact of leachate on receiving waters; to
aid in the design of any required leachate
treatment facilities; to provide estimates
of the material inputs into an underlying
groundwater system as a first step in
determining leachate concentration there;
and, to assess the likely effects of
co-disposal of various liquid and/or semi-
solid waste in landfills. For the initial
stages of this work a model was chosen
which assumes a simple physical process to
describe the leaching. The initial model
assumes flow-through of infiltrating
water and a mass transfer of soluble
materials from the solid phase to the
liquid phase due to a concentration
difference driving force, as well as conn-
ection within the liquid phase. The model
has been tested for the simple case of
plug flow starting at the top surface of
31
-------
landfill, with a constant rate of
infiltration, insignificant moisture in
the refuse, and a univorm distribution of
leachable materials.
This model provided leachate concentrations
as a function of time. The calculations
require values for the refuse depth, the
rate of infiltration, refuse field
capacity, and three empirically determined
parameters. The model has been fitted to
data from solid waste lysimeter
experiments for six contaminant curves and
a range of operating conditions. The
calculated curves and corresponding data
curves generally showed a good correlation
when the empirical parameters were
suitable adjusted.
The model is currently being further
modified to provide both the quantity and
quality of leachate as a function of time,
and for various patterns of infiltration
and fluid flow. Other variables being
ested are various landfill operational
parameters. A final report on the initial
work should be available within the next
few months. Leachate Control and Treatment
Infiltration into disposed refuse results
in the production of leachate which is
recognized to be potentially hazardous to
the environment. There are a significant
number of documented case histories which
show that the uncontrolled migration of
leachate from improperly located waste
disposal sites can result in extensive
groundwater and/or surface water
pollution. These case histories have been
used to demonstrate the need for the
design of leachate control and treatment
systems at waste disposal sites.
There is a growing trend in regulatory
agencies insisting on some degree of
leachate control including collection and
treatment. This trend is disconcerting in
that many leachate control systems have
not been proven technically, while others
have simply been designed to expedite the
approvals process. Furthermore, some
systems may be failing because of
inadequate design or improper
installation, whereas others are being
over-designed without due regard to the
cost-effectiveness of the systems. Rarely
has the environmental benefit and the
complete cost and benefit of such a system
been evaluated.
Landfill designers, researchers, and
regulatory agencies have became concerned
with problems relating to proper
selection, design, and operation
of leachate control and treatment systems'.
There is a recognized need to develop
practical criteria for the design and
operation of such leachate management
systems.
In response to this need, the Waste
Management Branch and its contractors
sponsored two international round-table
discussions on the topics of leachate
control, held in Vancouver, and leachate
treatment, held in Halifax.
The session on leachate control considered
the topics of natural and engineered
controls to minimize leachate production,
collect leachate" on-site, recirculate
leachate, and rectify leachate migration
problems through aquifer manipulation as a
control procedure. A report is being
prepared on the recommended procedures for
leachate control based on a concensus by
technical experts from the U.S. and
Canada.
32
-------
The purpose of the leachate treatment
session was to discuss and document
current treatment concepts, practices,
and methodologies.
Attendees at this session included
recognized treatment experts from Canada,
the U.S., Germany and Norway. A report is
being prepared to document the discussion
on the need for and the factors affecting
leachate treatment; the treatment of
leachate in the natural hydrogeologic
environment; biological treatment systems
physical-chemical treatment systems; and,
the concensus of expert opinion expressed
at the session.
With regards to in-situ attenuation, it
was recognized that there is a difficulty
in designing for the use of the natural
setting as an attenuation and/or treatment
system, and although it was agreed that
the assimilative capacity of the
hydrogeologic environment for leachate
contaminants is quite significant, ground
water contamination will usually result.
The concern, therefore, must be with the
degree of contamination which will be
allowed, and whether such contamination
actually constitutes pollution.
It was recognized that all constructed
treatment systems require, prior to
design, the setting of an effluent quality
standard. Similarly, if the hydrogeologic
environment is to be designed to accept a
leachate loading, water quality standards
at a finite distance downgradient from the
base of the disposal area must be
established. Mast commonly, the
maintenance of drinking water standards
has been used as a criterion for leachate
discharge in the hydrogeologic
environment.
A realistic leachate discharge objective
would be that leachate must not impair a
reasonable, legitimate, present or
potential off-site groundwater use. The
principal problem which makes it difficult
to define an effluent standard for the
discharge of leachate to the hydrogeologic
environment, is that in general, the
hydrogeologic environment cannot be
improved to allow for an upgrading of the
standard if necessary. At a treatment
facility, such an upgrading can readily be
facilitated.
It coule be argued that all landfill
designs, where leachate is discharged to
the hydrogeologic environment, should be
equipped with contingency plans which are
designed to control the discharge of
leachate to this environment should it be
required. If a contingency plan can
successfully be implemented, a setting of
design standards could be readily done.
With respect to engineered treatment
systems, it was agreed that current
processes can produce effluent liquids
virtually free of contaminants,
notwithstanding the high costs involved.
However, at the same time, rather than
being destroyed or converted to a less
objectionable form, the contaminants are
either retained in a highly concentrated
residual stream or transferred to another
medium, both of which require disposal.
Mast treatment processes have as their
objective, the production of a high
quality effluent liquid without regard for
the residual stream. Since landfills are
frequently the "last step in disposal,
simply transferring leachate contaminants
from one form to another for disposal
"elsewhere" was not considered to be a
satisfactory solution to a leachate problem.
Many solid waste disposal systems are not
presently in a position to finance "best
available technology".
33
-------
Consequently, it is hoped that, for cases
where leachate treatment is clearly deemed
necessary, it will be done according to
the "most practicable technologh" after
consideration of all relevant factors has
been completed. This should be based on
realistic assessment of the potential of,
and expectations for, the disposal area,
be it a surface or a groundwater system,
and the resources at hand to solve the
problem.
Reports on both the control and treatment
sessions will be available soon.
Extensive research into leachate treatment
has been underway for the last few years at
the University of British Columbia, and an
in-depth reporting of this program will be
presented later in this symposium by Dr.
Don Mavinic of their Civil Engineering
Department.
Leachate Toxicity
At the third annual research symposium in
St. Louis, I briefly mentioned our work in
developing a rapid leachate toxicity
testing procedure. The procedure, using
rainbow trout f ingerlings in stoppered BOD
bottles, proved satisfactory for many test
situations.
Toxicity has often been suggested as a
valuable indicator parameter in surface
waters monitoring programs. However,
standard bioassay procedures are expensive
and generally require large volumes of
effluent, and may take up to three days
per sample. The rapid toxicity procedure
was developed to overcome these problems,
and it was demonstrated to be a less
expensive, valuable in-situ monitoring
procedure.
A final report on the set-up and use of
the procedure has been published.
Our continuing interests in leachate
toxicity led us to develop an even simpler
test using Daphnia, and an attemp to
attribute leachate toxicity to
concentrations of specific leachate
contaminants. In addition, work was done
to determine p^ effects on toxicity, and
to assess the effects of landfill
operating variables such as rainfall,
sludge additions, and elapsed time since
disposal, on leachate toxicity.
Equations were developed by statistical
regression analyses of the experimental
data to relate toxicity to contaminant
concentrations. It was found that the
relationship between toxicity and
contaminant concentrations can best be
expressed in terms of the un-ionized
ammonia concentrations as well as the
concentration of zinc in the case with the
Daphnia test, or the copper concentrations
in the case with fish tests.
Furthermore, it was reported that the
adjustment of p" in test procedures can
dramatically reduce the toxicity of a
given sample.
A final report is in preparation.
MUNICIPAL WASTE DISPOSAL SITE SELECTION
Accepting that landfilling will be a
requirement for many years to come, it is
necessary that technology required to
locate and manage these sites be developed
to a high degree.
Over the past year we have been continuing
our development of a selection procedure
for locating suitable landfill sites which
could be used to service small
communities. The procedure will be a
simple, yet comprehensive attemp to
provide a small community with a decision
making process which will assis them in
locating the most suitable site from an
environmental, a public protection, and an
34
-------
operation and management point of view,
without having to rely on outside expert-
ise. A "cook book" approach is being used
in preparing a final report on the
procedure.
The selection procedure works as follows
Potential sites are evaluated against
environmental protection criteria,
including distance requirements to
property boundaries based on acceptable
attenuation distances within the local
soil environment, distance requirements to
downgradient water wells, and distance
requirements to ensure protection of
nearby dwellings from landfill gas
migration problems. Other criteria also
apply.
A pass-fail test is used for each
criteria. Failure to pass any of the
criteria would render the site unsuitable
without special site engineering.
From a public protection point of view,
the major criteria of importance in
addition to environmental and health
criteria are aesthetics, traffic, land use
designations and site completion. The
operation and management criteria ensure
the selection of the most operable and
cost-effective location.
The procedure is based partially on a
variety of procedures documented to date,
and should prove to be a simple, yet
effective decision tool for decision
makers in small communities.
The procedure should be finally available
within the next few months.
LANDFILL FTKTD INVESTIGATIONS
Although a great deal of research has
already been devoted to landfill
technology, the long and inconclusive
arguments between experts over the
acceptability of particular sites clearly
points to the need for additional basic
technical information. These arguments
are most often prevalent at public
meetings or regulatory hearings where
frequently neither side of the argument
can be adequately supported by firm
technical information.
In spite of the large amount of research
which has already been conducted, problems
arise as a result of the difficulty in
showing the applicability of laboratory
data to field conditions and in the
difficulty in studying the processes
directly in the heterogeneous hydro-
geologic environments of the real world.
Because a large amount of detailed
information has already been gathered at
the C.F.B. Borden landfill, as reported in
previous years, this site offered us an
excellent opportunity for further inves-
tigating both the applicability of
laboratory data for making field
predictions and for the field
investigation of contaminant migration
and attenuation processes.
Several activities have been undertaken,
and will be continuing, at this landfill
site. One such activity includes the
further definition of the physical
hydrogeology by supplementing information
from our past studies with some continuous
core sample analysis for detailed
stratigraphy, point dilution testing for
direct velocity measurements, single well
response tests for determining in-situ K
values, and field dispersivity
measurements.
Other activities at the site have included
the geochemical surveying of the
contaminant plume through field
measurements of Eh, pH, DO and specific
conductance taken on samples extracted
from multi-level sampling devices.
35
-------
Trace metal-oxide coating studies are also
being undertaken to determine the
contribution to contaminant levels in the
downgradient soils due to the release of
trace metals from soil oxide coatings
being dissolved under reducing conditions.
Studies have shown that the bulk of the
trace element content of contaminated
groundwater may be derived not from the
waste material itself, but from the
natural manganese and ferric oxide
coatings on the surrounding material
through which the leachate moves. Another
part of the field studies consists of the
evaluation of the degree of anaerobic
decomposition of leachate organics
migrating through saturated soils under
anaerobic conditions. Initial results have
shown a COD reduction of 86% after 1.5
meters of travel and a TOG reduction over
the same distance of 90%.
A comprehensive modelling exercise
completes the long list of activities
being undertaken at the Borden landfill.
The development of a calibrated flow model
is being followed by a simulation of the
distribution of a non-reactive solute,
namely chloride, and then a reactive
species. Lastly, the effects of transient
flow conditions will be examined.
One important aspect of this modelling
exercise is to determine a suitable input
function for the contaminant release from
the landfill itself. In the absence of
historical data from the site, a
successful leaching model such as the one
being developed at the university of British
Columbia may provide necessary information.
If modelling is to have a place at all in
our future work, such as has been
suggested by Weston in their report to EPA
on "Pollution Prediction Techniques for
Waste Disposal Siting", then the Borden
work should be the "proof of the pudding".
CO-DISPOSAL STUDIES
Septic Tank Sludge and Municipal Refuse
Our previous research project, on the
effect of adding septic tank pumpings to
refuse has been completed. The general
finding of this research was that the
addition of septic tank pumpings, in
sufficient quantity, significantly reduced
the mass of contaminants discharged, and
particularly heavy metals, when compared
with a control containing no pumpings.
This effect, attributed largely to
biological activity, was reduced when
rainfall rates were increased from 15 in
(381 urn) to 45 in (1143 mn) per year.
Increased gas production and a greater
initial percentage of methane gas combined
with reduced contaminant mass discharges
at the high septic tank pumping loading
rates made this technique look promising
from both an environmental and an energy
recovery viewpoint. Before proceeding
with a large scale operation, however, it
was felt necessary to determine whether or
not the tie up of metals was permanent or
simply transient in nature.
It was decided, therefore, to continue to
apply water to the existing 9 lysimeters
at the rates which had been previously
used, and to analyze gas and leachates on
a continuing monthly basis.
Railfall applications and leachate
collection have been continued since the
termination of the original contract, and
with only three exceptions, all leachate
contaminant concentrations have continued
to decrease. The exceptions are one tank
which showed increase in colour but no
other increases; another tank showed an
anamalous steady increase in iron con-
centration; and yet another showed
an initial increase in iron contrentation
36
-------
but then a drop to the previous low level.
Generally, pH has increased.
Concentrations of metals such as cadmium
and lead decreased to below detection
limits for all tanks, and ammonia nitrogen
also decreased below detection limits for
most tanks.
There is presently no Indication that
desorption of contaminants is occurring.
However, a continued increase in gas
production in some tanks shows that
biological activity is continuing, and if
biological activity is the mechanism which
is causing metal tie-ups, this indicates
that desorption may not occur for some
tiem yet, if at all.
A final report should be available this
simmer, 1979.
Metal Sludges and Municipal Refuse
Having demonstrated the beneficial effects
of adding untreated biological sludge to
landfills in terms of achieving metal
tie-ups, we decided to undertake further
work to determine whether or not these
previously found beneficial effects would
apply when industrial sludges were added
to the refuse as well.
A series of thirteen lysimeters were
constructed using 12 inch diameter PVC
pipe. Flat bottoms were welded to each and
specially designed sealed tops were
constructed. The necessary additions of
septic tank sludge and an arsenic contain-
ing waste and an electro-plating sludge
were each determined based on past co-
disposal experience.
The tanks were filled and put into
operation in August 1978. Results to date
are still being analyzed, and the study
will continue into 1981.
LASSDFILL GAS RESEARCH
Gas Production and Migration
The objective of our gas production and
migration study is to provide information
about gas composition, pressure, and
production in landfills. Such information
is essential for both an analysis of gas
migration into adjacent soils and an
assessment of gas recovery potentials at
landfill sites. This study, to be
discussed in detail later in this program
by Dr. Grahame Farquhar of the University
of Waterloo, will also involve the use of
field data in the calibration of a
previouslu developed gas transport model.
The calibrated model will be used to
display variations in gas migration
patterns and to evaluate various gas
control and recovery schemes.
Gas Recovery and Utilization
Although it is being recognized that a
municipal solid waste landfill may be a
significant energy source, it is clearly
not feasible to tap smaller sites in
remote locations for low BTU content gas
supplies for consumption elsewhere.
Unfortunately, there is usually no nearby
gas distribution system available. Even
if a pipeline could be found next to every
small site, the economics of cleaning the
gas to pipeline quality could be
prohibitive.
Our study addresses the possibility of
extracting landfill gases year round at
any site and using these gases for an
on-site energy supply. One market which
could readily benefit in Canada from such
a readily available supply is the energy
intensive greenhouse industry, which
presently faces fuel costs of up to
37
-------
$30,000 per year per greenhouse acre.
California studies have suggested that the
best economics lie in the direct use of
the gas on-site with minimal gas
processing and refining. Our 10 month
study is intended to demonstrate the use
of the collected gas to supply the energy
requirements of a small greenhouse
throughout the year. It is planned to grow
tomatoes in this greenhouse situated on a
completed site in St. Thomas, Ontario.
The completed St. Thomas Landfill Site is
divided into two sections. The section
under study was in operation from 1967 to
1978. This section is comprised of
approximately 24 acres of refuse with an
average depth of 40 feet. The gas production
wells, the gas monitoring and pressure
probes, and the various greenhouse
facilities.
The gas production well consists of a 36
inch augered bore hole, logged as 1 foot
of fill, 28 feet of refuse and 3 feet of
native soil. A 6 inch diameter FVC pipe
with an attached 8 feet of stainless steel
well screen (50 slot) was placed in the
center of the augered hole. The hole was
then backfilled with pea gravel, with gas
pressure and monitoring probes inserted at
various depths. The top of the gas
production well was closed using 1 %
feet of 36 inch diameter concrete tile.
Bentonite was placed around the tile and a
concrete plug was used ar the surface. A
concrete cap is placed on top of the
concrete tile.
A rubber adapter is being used to connect
the pipe leading from the gas pump to the
concrete cap, so as to allow pumping
directly off the gravel pack to permit
an assessment of the feasibility of pumping
off such an inexpensive gravel pack
without having to install gas extraction
well hardware. After completion of
pumping from the gravel pack, 6 inch FVC
pipe will be connected directly to the gas
line feeding the gas pump.
The gas recovery system extracts the gas
frcm the production well, passes the gas
through a water knock-out drum and either
pumps the gas to a furnace within the
greenhouse or exhausts the gas to the
atmosphere.
The control on the gas system will be a
pressure release valve. The pump will run
continuously at a set flow rate, and if
teh pressure goes up in the line, the
pressure release valve will allow gases to
exit to the air. When the furnace is not
in operation, all the gas will be released
to the air.
A forced air furnaces has been adapted to
utilize methane gas. A back-up electric
furnace willbe installed in case of
system failure, to prevent crop damage due
to frost.
A total of 46 gas monitoring and pressure
probes were installed. In addition, 3 gas
monitoring and pressure probes were
installed within the gas production well,
and 4 gas monitoring probes were installed
below the greenhouse liner to monitor for
gases coming from the fill underneath the
greenhouse.
The complete system is currently in
operation, and methane concentration and
pressure distributions within the
landfill, changing as a result of varying
pumping rates, are being monitored.
The study will be completed in August, 1979
and a final report should be available
shortly thereafter.
38
-------
NATO/CCM3 LANDFILL PROJECT
A major international effort is underway
among1 the member countries of NATO,
including the U.S. and Canada. The
purpose of the pilot project id to study
hazardous wastes management, and one part
of this project is a landfill study being
supported by the U.S., Canada, Norway, thet
U.K., the Netherlands, and teh Federal
Republic of Germany. The purpose of the
landfilling study is to assess the role of
landfilling in hazardous waste disposal.
As Chairman of this sub-project for the
next three years, I will be reporting on
our deliberations and coordinating input
from the various landfill research
programs being undertaken by each
participating country. All submissions
will be used to prepare a final project
report to NATO.
The results of our work should have a
significant impact on policy planning
within the member countries.
One of the topics to be addressed will be the
the co-disposal of industrial and
municipal refuse. Member countries at
present appear to have opposing view
points on the merits of co-disposal, and
teh resolution of this issue, if at all
possible on the basis of research results,
should prove to be a most interesting
challenge.
Another very important aspect of our joint
international undertaking is the expressed
desire by some member countries to pursue
future research efforts on a cooperative
basis and to establish a formal infor-
mation exchange.
In accordance with this spirit of developing
cooperative international efforts we
have already initiated a coordination
with the Solid and Hazardous Waste
Research Division of the EPA of our
respective landfill gas research studies,
in the hope that we can mutually benefit
from each other's work from the outset,
and avoid costly duplication. It may
even be possible to directly tie into
similar German studies now getting
underway. The benefits of such an
international coordination of efforts
will obviously accrue to us all.
The landfill sub-project will be completed
in late 1980.
39
-------
PILOT SCALE PROCESSING EQUIPMENT EVALUATIONS
FOR RESOURCE RECOVERY
Dr. Harvey Alter
National Center for Resource Recovery, Inc.
Washington, D.C. 20036
ABSTRACT
A pilot scale (ca. 20 tph) pilot plant for investigating the mechanical processing
of municipal solid waste had been established by the National Center for Resource Re-
covery- Its scope was expanded, and its use developed, under a grant from SHWRL-MERL,
U.S. Environmental Protection Agency. The purpose was to investigate particular aspects
of the processing for the recovery of products to specification for reuse. The intent
was not to exhaustively prove or demonstrate operating parameters, but rather to obtain
data useful to designers and implementors, so as to reduce perceived risks of this form
of processing and technology transfer. The pilot plant is described; experimental re-
sults pertaining to investigations of upgrading refuse-derived fuel by screening prior to
air classification, aluminum recovery and jigging preparatory to glass recovery, are pre-
sented.
SCOPE AND OBJECTIVES
The processing of mixed, municipal
solid wastes for the recovery of useable
and saleable materials, and for the prep-
aration of a refuse-derived fuel, usually
employs equipment and unit operations com-
mon in other industries. In spite of this
"technology transfer," there is little
documentation of the performance or utili-
ty of the borrowed devices to assist de-
signers in adopting them for processing
wastes. One reason is that the output pro-
duct from one type of unit operation is of-
ten needed as the input feed to another.
It is difficult to investigate unit opera-
tions singly; a continuous plant, or near-
ly so, is required.
Starting early 1974, the National Cen-
ter for Resource Recovery began to assem-
ble equipment for a continuous pilot plant
so as to investigate the performance of
some particular unit operations. Once
assembled, the effort was supported by
several grants from the Environmental Pro-
tection Agency (1-4).
The overall objective of the experi-
mental work was to record aspects of the
performance of materials recovery unit
operations to assist designers and imple-
mentors of resource recovery. It was not
the objective to rigorously prove or estab-
lish operating parameters for the unit op-
erations but to obtain sufficient data to
guide future systems' design and to reduce
the perceived risks of the technology. This
paper summarizes some of the findings to
give a broad overview of the type and na-
ture of the experimental work. It does not
include all findings and detail nor reports
at all on: the performance of prototype air
classifiers, devices for pneumatic removal
of ash components in RDF, cleaning the re-
covered aluminum product, nor the perfor-
mance of most of the equipment in the glass
recovery circuit'. The interested reader is
referred to other publications and the pro-
ject final report cited in the text.
THE FACILITY AND PROCESS FLOW
The work was conducted at the National
Center's Equipment Test and Evaluation Fa-
cility (ETEF) in the District of Columbia,
located in what was the oversize bulky
waste processing area of the District of
Columbia's operating incinerator. In 1974,
an agreement was signed which permitted use
of part of the building and some of the
equipment for experimental purposes.
40
-------
Packer trucks were diverted from the in-
cinerator tipping floor, as needed, to pro-
vide feedstock for investigations. Unre-
covered materials, or samples after use,
were disposed of in the incinerator pit.
The oversize bulky waste processing
area originally consisted of a tipping
area and steel pan conveyors feeding a
horizontal hammermill. The District per-
mitted temporary modifications of the fa-
cility so that it could evolve into a pi-
lot plant for resource recovery investiga-
tions. The capacity of the shredder, by
itself, is probably of the order of 100
tph. However, the infeed and outfeed con-
veyors originally installed were intended
for bulky waste, of higher bulk density
than packer truck refuse. Consequently,
the capacity of the system with packer
truck refuse was no more than about 20
tph.
The modifications permitted allowed
installation of many different unit opera-
tions. In most instances, there was a
severe space restriction and equipment had
to be installed and located in areas and
configurations that were far from optimum.
The specific arrangements were dictated by
the available space and did not resemble
any attempt at an engineered plant. Only
packer truck refuse collected in residen-
tial neighborhoods was used for the work
described in this report. The description
below omits that portion of the facility
which was used for processing the light
fraction into densified refuse-derived
fuel, described elsewhere (4).
The process flow is shown in Figure
1. The municipal solid waste (MSW) was
first size-reduced in shredder (1); the
particle size distribution of this mate-
rial has been reported (5). In a late
configuration, the shredded MSW was
screened at (2) to remove particulate in-
organic material as a means of investi-
gating the upgrading of the fuel product.
Screen (2) also acted as a vibratory feed-
er to the air classifier (3). Several
different air classifiers were investi-
gated and are described in Reference (6).
The light fraction from the air clas-
sifier was de-entrained in cyclone (4).
In early configurations, when screen (2)
was not used, screen (5) was employed as a
means of investigating the removal of in-
organic fines from the light fraction.
The dust and air effluent from the cyclone
(4) were discharged into a large settling
chamber when it was discovered early that
the effluent contained many small pieces
of paper, approximately 1 cm2 in size.
The settling chamber was a blocked-off, un-
used small room, fitted with fabric fil-
ters over the door. Periodically, the
room was cleaned out. In the course of
some investigations of air classification,
the air discharge from cyclone (4) was re-
cycled back into the air classifier.
Cyclone (4) discharged through a ro-
tary air lock into a pneumatic conveyor so
that the light fraction could be delivered
to feeder-screen (5) and onto further pro-
cessing. A small cyclone was placed over
screen (5).
The heavy fraction from the air clas-
sifier passed under the magnetic belt sep-
arator (6), then past a magnetic head pul-
ley (7) to make certain all (or nearly
all) of the magnetic metals were removed
and recovered. The remainder of the ma-
terial was screened through a rotary screen
(8) and separated into three fractions.
The fraction smaller than 5 cm from
(8) was screened again at (10), a 1.2 cm
vibrating flat deck screen. The oversize
was rejected and the undersize fed to the
Bendelari jig (11). The jig washed out
the organic contaminants and the inorganic
materials were sent to the crusher (12),
operated in conjunction with 20 mesh
screen (13) to reduce all of the material
smaller than 20 mesh. Not shown is the
6.4 mm screen operated in series with
screen (13) so as to remove non-crushable
materials, principally small pieces of
metal.
The underflow from screen (13) dis-
charged into a sump and was pumped, as a
slurry, into hydrocyclone (14) which dis-
charged into froth flotation cells (15).
The float product from (15) overflowed in-
to spiral classifier (16) and was then
screw-conveyed into the oil-fired dryer
(17). The glass recovery circuit is de-
scribed in greater detail later.
The oversize fraction from rotary
screen (8), a material larger than 10.2 cm,
was discharged as waste. The middling
from (8) dropped on a vibratory pan feeder •
and was transported on a belt passing
through the eddy current separator {9) for
aluminum recovery.
41
-------
After (9) , the material was discharged as
a waste.
Equipment Description
Much of the equipment used is de-
scribed in Table 1. The glass recovery
equipment is described in a later section.
SCREENING OF SHREDDED MSW
AS A MEANS OF UPGRADING RDF
Shredded MSW contains crushed rock,
glass, sand and other similar small par-
ticulate inorganic materials. When air
classified, the particulate materials are
likely to be included with the light frac-
tion. The light fraction is a common form
of refuse-derived fuel (RDF) and the inor-
ganic particulates add to the ash (non-
combustible) portion. If the small par-
ticulates could be removed before air
classification, the ash content of the
fuel would be decreased, the heating value
(as-received) increased, and the value of
the fuel enhanced. The particulate ma-
terial contains crushed glass. If the re-
source recovery plant includes glass re-
covery, the process flow should not permit
the glass to be wasted in the air classi-
fier light fraction.
In 1977, when ah experimental verti-
cal zig-zag air classifier was installed,
it was decided to use a vibratory feeder
as the means of providing an even feed of
shredded MSW to the classifier. The feed-
er had a perforated bottom plate so that it
acted as a vibratory screen, thus permit-
ting investigation of the screening of
shredded MSW as a means of reducing the
content of removable inorganic fines (RIF)
in RDF.
Background
The air classifiers used at ETEF were
operated between about 8.6 and 25 m/s
(1700 and 4900 fpm) air velocity. This may
be compared to published values for the
settling velocity of quartz powder (7);
with a specific gravity of 2.65, quartz is
similar to sand and glass. Based on this
comparison, particles as large as 10 mm
may be carried in the air stream at opera-
ting air velocities, consistent with the
computations of Fan (8).
For the work described below, the ex-
perimental zig-zag air classifier was oper-
ated between 9 and 16 m/s to achieve be-
tween a 50/50 and 70/30 split, light/heavy
fraction. The feed for these trials was
the oversize fraction from the screen de-
scribed below. If the material were not
screened, Fan's work predicts that at 15
m/s (=3000 fpm) inorganic particles 3 mm
and smaller will be included in the light
fraction. Also, the light fraction will
contain 10 wt.% of RIF, close to operating
experience. For example, previously re-
ported particle size distributions of
shredded MSW and accompanying air classi-
fier light and heavy fractions (3), show a
larger fraction of material passing 6.4 mm
for the light fraction than for the heavy.
Also, light fraction prepared in an air
classifier operating at the upper part of
the velocity range was found to contain 14
wt.% RIF. Most of this was in the particle
size range less than 10 mesh (2 mm).
The performances of a screen is depen-
dent on many design and operating factors,
some controllable, others not. Further,
methods of determining and expressing
screen efficiency are not uniformly estab-
lished (9). The method chosen here is to
express performances as the efficiency of
recovery, R, of the <6.44 mm fraction from
the shredded MSW feed: R = 100 (f - t)/
[f(l - t)], where f is the wt. fraction of
<6.4 mm material in the feed and t corre-
spondingly in the oversize (10).
The feeder-screen was a Texas Shaker
provided by Triple/S Dynamics Co. It had
a total operating area 2.1 x 4.9 m (7 x 16
ft). However, not all of the area could be
used as a screen in as much as there had to
be an air seal between the air classifier
feed point and the screen to prevent back
leakage of air through the screening medi-
um. The area actually used was measured
as 2.1 x 2.4 m (7 x 8 ft).
The screening surface was 10 gauge
steel punch-plate with 6.4 mm diameter cir-
cular holes on 1.1 cm staggered centers.
The actual open area of this type screen is
48% (11). The screen surface was declined
at 8° so the unit would fit the available
space. There was no adjustment. The vi-
bratory action was provided by a 5.6 kW
(7-1/2 hp) motor with a V-belt drive to
produce 20 c/s at an amplitude of 19 mm.
The screen was fitted with rubber balls and
a tray.
42
-------
Results
can be calculated not to exceed 4.7% (6).
Particle Size Distribution of the
Feed. The particle size distribution of
the shredded MSW feed at ETEF had been mea-
sured and fitted to a Rosin-Rammler func-
tion (5), predicting that 30.4 wt.% of the
oversize fraction to the screen is <6.4
mm, verified by measurement. The average
of the 10 determinations was 30.4% with a
coefficient of variation of 28.4%.
Undersize Fraction in the Oversize.
Samples of feeder-screen oversize product
were collected and the weight fraction of
<6.4 mm material remaining in each was de-
termined (t). The samples were taken at
different feedrates. The results are
listed in Table 2.
Recovery Efficiency. The recovery ef-
ficiency, R, was calculated as above to
over the range of feedrates investigated.
For these calculations, f = 0.304 and t is
determined. The computed values of R are
also listed in Table 2.
Composition of the Screen Undersize.
The feeder-screen undersize fraction con-
tains inorganic particulate material and
some organic materials, which appear to be
residues of food and yard wastes. In or-
der to determine the amount of inorganic
material, the loss on ignition of the
screen undersize was measured. The average
of 14 samples, collected on eight separate
days, was 82.6 wt.% inorganic material with
a coefficient of variation of 8.92%. This
confirms that the screen is removing mostly
inorganic or ash-forming material before it
can enter the air classifier.
The small amount of organic material
included in the screen undersize fraction
represents a potential loss of fuel, which
Scaling and Sizing of Screens
The results reported here can be used
to estimate the size of a screen for a
full-scale resource recovery plant. There
are several ways of scaling screens. How-
ever, with the limited data available, a
simple approach is taken.
Gaudin (12) states the capacity of
vibrating screens to be from 5 to 20
t/ft2-mm aperture - 24 h for ores. Assum-
ing the ratio of densities of shredded
MSW to ore to be 1/10, and screen capacity
is proportional to this ratio, the "rated"
capacity of the 5.1 m2 (56 ft2) screen
used in this work is from 0.18 to 0.73 tph
of shredded MSW (short tons).
The relation between recovery, R%,
and the feedrate, F tph, is represented by
(6), R = 63.6/F0-*2, or that R = 72.6% at
F = 0.73 tph. A properly operated vibrat-
ing screen should have R = 90% or greater
at R = 90%, F = 0.44 tph. Using the Gaudin
scale law, F = 0.44 is the capacity of the
screen. Following Gaudin, the scaling
factor can be calculated (for R = 90%) as
1.2t(MSW)/ft2-mm aperture -24h. Thus, a
similar screen for 50 tph resource re-
covery plant would have an area of 6350
ft2. Vibrating screens are usually de-
signed to have a length twice the width
(10) so this screen would be sized 17 x 34
m (56 x 112 ft), obviously too large for
common sense.
Several design parameters could be
changed to improve the screening efficien-
cy, thus changing the size of the screen.
For example, the 6.4 mm punch-plate used
has an open area of only 47.8%. Screen
cloth with a square hole of the same size
Table 2. Recovery of <6.4 mm Fraction by the Feeder-
Screen for f=0.304, F= Short Tons/hr.,
t=Short Tons, R=% Recovery
Feedrate*, F t R
1.8 tph
2.4
3.7
4.6
6.1
11.6
0.180
.192
.214
.230
.237
.250
49.7%
45.6
37.7
31.6
28.9
23.7
*F in U.S. short tons
43
-------
has an open area of from 60 to 69%, de-
pending on the wire size. If the screen-
ing surface were so changed, the size of
the screen for 50 tph would be reduced to
14.3 x 28.6 m (47 x 94 ft), still too
large for common sense. Other changes in
screen design, such as angle of decline
and vibration intensity, are not likely to
change the scaled dimensions significantly.
Other design features were considered
in order to improve the efficiency of such
a screen. The conclusion was that a flat
screen is not suitable for the purpose,
probably because the paper-like or flat
nature of shredded MSW, causing it to lie
flat on the screen and block the apertures
(6). A rotary screen may be more suited,
as proposed (13) .
ALUMINUM RECOVERY USING AN EDDY CURRENT
SEPARATOR
MSW may contain from 0.2 to 1 wt.%
aluminum depending on whether or not alu-
minum beverage containers are marketed in
the area. When the percentage approaches
1%, about 1/2 of this typically would be
cans. The remainder would be foil, pie
plates, castings, extrusions, and from
miscellaneous sources. The alloy (chemi-
cal) composition of these different items
may vary widely and the value of recovered
aluminum depends on its alloy composition
resembling or matching a standard alloy
composition. Thus, there is advantage re-
covering the aluminum canstock in MSW in a
pure form, to be processed back into alu-
minum suitable for reuse in wrought pro-
ducts such as cans. Recovered Al has high
value.
Test runs were conducted using the
Combustion Power Co. Al Mag 20 Eddy Cur-
rent Separator. As will be shown, the re-
sults indicate that the Al Mag 20 is capa-
ble of handling feedrates and making rea-
sonable to good separations well in excess
of its design capacity of 1 tph. The Al
Mag feed is the 10.2 x 5.1 cm fraction of
the nonmagnetic portion of the air classi-
fier heavy fraction. The product is not
clean aluminum, but a mixture of alloys,
some other nonferrous metals, and miscel-
laneous organic materials. This section
deals only with the recovery of a mixed Al
product using this eddy current device.
Cleanup of the product to a market specifi-
cation is described elsewhere (2,6). The
eddy current separator has been described
(14) .
The procedure was to feed the Al Mag
with heavy fraction that had previously
been magnetically scalped and sized. For
the test trials, whole aluminum cans were
introduced into the heavy fraction along
with the shredded aluminum canstock that
was "native" to the feed. After magnetic
separation, the feed was screened into a
small trommel to produce a 10.2 x 5.1 cm
fraction which concentrated the aluminum.
The mass balance has been reported (2).
Experimental Results
Two series of experimental trials were
conducted. The first was according to the
process flow of Figure 1. The results led
to the conclusion that the 10.2 x 5.1 cm
trommel was undersized, hence inefficient.
In the second series, the middling
product from the trommel was collected and
screened a second time. The second series
is referred to as "double-screen" feedstock.
Some of the experimental results are
plotted in Figure 2 to 5. Figure 2 is a
plot of the amount of organic contamination
in the Al Mag product as a function of the
feedrate for the single-screen feed. These
results should be compared to Figure 3, a
similar plot for the double-screened feed,
illustrating the effect of the undersized
trommel. (A trend line is shown to guide
the readers.)
Figure 4 is a plot of aluminum can re-
covery efficiency as a function of feedrate
for whole cans, combining results from
using both single-screened and double-
screened feeds. Figure 5 is a similar plot
for shredded cans, again showing a trend
line.
Discussion. The particular eddy cur-
rent separator used was designed for a
throughput of abput 1 tph. The results in-
dicate this device is capable of recovering
90% or so of whole cans when operated at
feedrates in excess of four times this de-
sign figure, and of recovering about 65% of
shredded cans when operated at design ca-
pacity. The recovery efficiency falls off
greatly for shredded cans as design capacity
is exceeded.
The performance is summarized in Table
3 reporting average values of some of the
44
-------
Table 3
Summary of Al Mag Performance
at Feedrate up to 2 tph
Single
Screen
Double
Screen
Metals in
Feed, %
12.7
16.0
Metals in
Product, %
78.9
92.7
Metals in
Residue, %
4.1
7.9
Recovery
Whole
cans
92.7
97.0
Shredded
cans
72.6
52.8
Residue
as %
of Feed
88.2
83.2
parameters for feedrates between 0 and 2
tph. This range is chosen to illustrate
some points.
The 10 x 5 cm feedstock to the Al
Mag contained 13 to 16% metal. The Al
Mag concentrated this metal about five
times at a recovery efficiency in excess
of 90% for whole cans and 53 and 73% for
shredded cans. The lower recovery rate
for the double-screened feed is not under-
stood. Note that more than 70% of the Al
Mag feed is residue. The residue contains
on the order of 4 to 8% metal, which is
lost to disposal.
The tentative specification for re-
covered aluminum permits 2 wt.% organic
contaminants. The average contents of or-
ganic contaminants at feedrates up to 2
tph was 8.5% using double-screen feed.
Even the product from using double-screen
feed at throughputs greater than about 1
tph contained more organic material than
permitted by the specification. These re-
sults indicate a clean-up step is required
to recover aluminum to specification.
Some experiments with a double-pass of ma-
terial through the eddy current separator,
and with prototype clean-up devices (air
knife and air table) have been described
(6).
THE RECOVER? OF GLASS
Description of the Glass Recovery Facility
A froth flotation process to recover
container quality glass from municipal
solid waste was investigated; the process
flow is shown in Figure 6. The feedstock
was produced by screening a mixture of the
air classifier heavy fraction (without
magnetic metals), and the underflow of the
feeder-screen to the air classifier.
There are several possible froth flo-
tation processes depending on the particu-
lar choice of equipment and its arrangement.
In each, the basic objectives are the same-
to clean and size the feed to froth flota-
tion cells, float the glass from the waste
and contaminants, and dry the glass. There
have to be intermediate steps of screening,
pumping, and adding and subtracting water.
One particular equipment arrangement was
chosen to do this work described here,
based on some "best judgement" estimates,
the few available literature reports, lab-
oratory work done as part of the National
Center's internal research program, and con-
versations with people in the field. The
process flow chosen was not believed to be
optimum, but rather a starting point for
further investigation. Indeed, several fea-
tures of the process were incorporated so
as to be able to investigate a particular
feature, or option, compared to what may
have been described by others.
The Process Flow and Associated Equip-
ment. The process flow is shown in Figure
6. Each Glass Recovery Unit Operation
(GRUO) is numbered. The numbers refer to
the equipment list and description in Table
4 which summarizes each function.
The equipment was installed for what
was believed to be continuous operation.
Experience indicated this expectation was
optimistic. For example, a large barrier
to continuous operation was the absence of
large surge capacities between the process-
ing steps as the means of evening the flow
of materials. This is particularly impor-
tant in resource recovery processing where
45
-------
the composition of the input waste is tem-
poral, hence the input feedrate to glass
recovery uncontrollable, even if the re-
mainder of the plant operates at all times.
The function of each piece of equip-
ment, and the mass balance for the assem-
bly, have been described (6). In the in-
terest of brevity, only one portion of the
plant is described here - the jig.
The purpose of the jig is to wash and
separate a concentrate of stones and glass
from the feed as one product, a mixture of
nonferrous metals as another, and a waste
fraction of the organic materials (e.g.,
food and yard waste) as another. It was
anticipated that nonferrous metals would
be recovered as a by-product of preparing
the feedstock for glass recovery and that
small pieces of such metal would accumu-
late as oversize material on the sizing
circuit screen, GRUO 1.5. The amount of
such metal separated this way was minor,
probably because little is smaller than
1.2 cm.
Jigging is an essential, if not criti-
cal step in glass recovery, compared to
other pieces of equipment which might be
expected to operate more closely to exper-
ience in the minerals beneficiation indus-
try. Once the jig has performed its func-
tion, the feedstock is similar to the in-
organic mixtures common to other applica-
tions of froth flotation.
THE USE OF A JIG TO PREPARE GLASS
FEEDSTOCK FOR RECOVERY
The feed to the jig is a mixture of
glass, stones, metal, wood, food waste,
and other organic materials. A mineral
jig is not necessarily the only type of
equipment which could be used for this
purpose, but is well suited to the pur-
pose.
"Jigging" action applies a principle
called hindered settling (15,16), the re-
action of particles to movement in rela-
tion to the medium surrounding them. The
medium may be air, water, or some other
fluid. All of the descriptions in this
report refer to water as the jigging me-
dium. Figure 7 shows a Bendelari jig, the
type used for the work reported here.
The length of jig stroke is selected
to cause the jigging medium to exceed the
minimum velocity necessary to liberate the
largest and heaviest particles, Vm. This
velocity can be calculated for spherical
(or near-spherical) particles (17) accord-
ing to: V = 26.32
Vm = velocity (mm/s) , D
... where:
maximum particle
diameter (mm) , and S = specific gravity of
the material to be separated.
The largest and heaviest particles
likely to be separated are nonferrous met-
als, approximately 12 mm in size, ranging
down to glass and stones, <12 mm. The
specific gravity of the metals might be as
high as 9 (e.g., Cu, ignoring an occa-
sional piece of Pb) and certainly not low-
er than 2.
The velocity of the water in the jig
used ranged from 23 to 240 mm/s depending
on operating factors, or that the jig used
was capable of achieving the minimum ve-
locity to separate the largest and heavi-
est particles likely to be found in the
feed. Other operating factors have been
discussed (6) .
Description of the Jig Feed
The feed to the jig is the underflow
from the 13 mm screen, GRUO 1.0, Figure 6.
Measurement of the particle size distribu-
tion showed that >95 wt.% was <10 mm.
Table 5 gives the materials balance.
From this, and the particle size distribu-
tions, it is possible to compute that
about 94 wt.% of the total glass, sand,
rock ceramic fraction (and all of the <6.4
mm underflow from the feeder-screen) are
included in the <13 mm fraction.
Performance Requirements for the Jig
The suitability and function of the
jig must be assessed in terms of its pro-
duct yield and quality. The specification
for the final recovered glass product dic-
tates, in part, a performance specification
for the jig. The recovered glass must be
essentially free of organic materials (0.2
wt.% max.) and metals (0.05 wt.% max. Fe
and 0.1 wt.% max. other), judging from the
proposed specification (18) .
Broken glass, obtained from a nearby
recycling center, was used as ragging.
Ragging for this purpose is essential and
the pieces should not be flat, like broken
glass, but spherical and/or small.
46
-------
Experimental Results
During operation, samples of feed,
hutch and cup product were taken and
tested for particle size, composition and
content of organic materials. Additional-
ly, the jig table overflow was sampled and
examined for glass and nonferrous metals
carryover. Mass balances were determined.
Table 6 tabulates some physical character-
istics of the jig products as well as
giving the mass balance (drained weights).
The content of organic material was deter-
mined by measuring the loss on ignition
(LOI). The results in Table 6 show the
content of organic material of the cup pro-
duct is 5.6 wt.% and the hutch product
0.17 wt.%. These results indicate the jig
is capable of reducing the content of or-
ganic material down to the specification
level required for glass or that elaborate
separation steps beyond the jig are not
needed. Further, the particle size distri-
bution of the hutch product is such that
approximately 60 wt.% is <1 mm in size, or
approximately the required 20 mesh for
froth flotation, which reduces the load on
the impact (or other type) crushing mill.
The mass balance indicates that if the
cup product was discarded, i.e., not used
as part of the feed to froth flotation,
there would be a loss of approximately 17%
of feed, not necessarily of glass. If the
cup and hutch products are combined, the
loading of organic materials to the float
cells is increased from 0.17 (for just the
hutch product) to about 1.1% (for the com-
bined product), assuming no subsequent
losses of organic materials in the sizing
circuit. Combining the two products,
therefore, requires a dispersion float (or
other step) to reduce the 1.1 wt.% to <0.2
wt.% organic content.
FUNDING ACKNOWLEDGEMENT
The work reported here was conducted
under Grant R803901 from the Solid and
Hazardous Waste Research Laboratory, Muni-
cipal Environmental Research Laboratory,
U.S. Environmental Protection Agency,
Donald Oberacker, Project Officer
REFERENCES
1. Alter, H., S. L. Natof, K. L. Woodruff,
W. L. Freyberger and E. L. Michaels,
1974. Classification and concentration
of municipal solid waste. Proceedings,
Fourth Mineral Waste Utilization Sym-
9.
10.
11.
12.
13.
posium. E. Aleshin, ed. IIT Res. Inst.
& U.S. Bu. Mines, Chicago, pp. 70-6.
Alter, H., S. L. Natof and L. C.
Blayden, 1976. pilot studies process-
ing MSW and recovery of aluminum using
an eddy current separator. Proceedings,
Fifth Mineral Waste Utilization Sym-
posium. E. Aleshin, ed. IIT Res. Inst.
S U.S. Bu. Mines, Chicago, pp. 161-8.
Alter, H. and B. Crawford, 1976.
Materials recovery processing research.
A summary of investigations. Report on
Contract 67-01-2944. U.S. Environmental
Protection Agency, Washington, D.C.,
pp.
99
Alter, H. and J. Arnold, 1978. Prepa-
ration of densified refuse-derived fuel
on a pilot scale. Proceedings, Sixth
Mineral Waste Utilization Symposium.
IIT Res. Inst. S U.S. Bu. Mines,
Chicago, in press.
Stratton, F. E. and H. Alter, 1978.
Application of bond theory to solid
waste shredding. J. Environ. Engr.
Div., ASCE. Vol. 104: 93-107.
Alter, H., ed., 1978. Materials re-
covery from municipal solid waste:
Investigation of air classification,
upgrading RDF, aluminum and glass re-
covery. Grant R803901. U.S. Environ-
mental Protection Agency, Cincinnati.
Final Report in press.
Taggart, A. F., 1927. Handbook of
mineral dressing. J. Wiley & Sons, New
York, p. 9-02.
Fan, D. N., 1975. On the air classified
light fraction of shredded municipal
solid waste. Composition and physical
characteristics. Resource Recovery s
Conservation. Vol. 1: 141-50.
Taggart, A. F. pp. cit., p. 7-04.
ibid., pp. 19-191.
ibid., pp. 7-14.
Gaudin, A. M., 1939. Principles of
mineral dressing. McGraw Hill, New York,
p. 162.
Savage, G. and G. J. Trezek, 1976.
Screening shredded municipal solid waste.
Compost Science, pp. 7-11-
47
-------
14. Abert, J. G., 1977- Aluminum re-
covery. ?, status report. NCRR Bul-
letin, Vol. VII, Nos. 2 & 3.
15. Richards, R. H., 1903. Ore dressing.
Engineering & Mining J. Vol. 1^ p.
636.
16. Gaudin, A. M. op. cit., p. 250.
17- Richards, . H. and C. E. Locke,
1940. Textbook of ore dressing.
McGraw Hill, New York, pp. 197-8.
18. Alter, H., 1979. Development of
specifications for recycled materials.
Conservation & Recycling. Vol. 2.
in press.
ACKNOWLEDGEMENTS
Acknowledgement with thanks is made to the District of Columbia, Department of
Environmental Services, for the use of equipment and facilities. Thanks are given to the
various manufacturers who loaned additional equipment. The work described here is the
result of the contributions and hard work of many staff members of the National Center
and engineering co-op students who didn't mind an unusual assignment early in their
careers.
48
-------
Figure 1. Schematic of ETEF Process Flow
MSW
FEED
1
SHREDDER
-4 IN.
2- SCREEN
1/4 IN.
3.
AIR
CLASSIFIER
6.
MAGNET
MAGNET
8- SCREEN
4X2
9 EDDY
SEPARATOR
WASTE
GLASS 4
PRODUCT
17.
->• -1/4 IN.
WASTE
Fe PRODUCT
-> +4 IN.
WASTE
->• A1 PRODUCT
DRYER
16.
SPIRAL
CLASSIFIER
4.
CYCLONE
5.
SCREEN
3/16 IN.
RDF
I0- SCREEN
1/2 IN.
JIG
12.
CRUSHER
l3- SCREEN
20M
14.
HYDROCLONE
15.
FLOAT CELLS
•>DUST
->-3/l6 IN.
WASTE
1/2 X2 IN.
WASTE
•>WASTE
+ 20 M
WASTE
->• WATER 8
SLIME
WASTE
49
-------
Ul
O
CD
C£
O
55
30
25
20
15
10
5
01
i
~&~ O
-O-
-O-
' 4
-cj>- o
o
•*• *»
1
0
O
O
- 4
v
o
o
XX
V
o
\
4UI
0
^ *i
DLL.! orc.ut» vuni /
FPM (INCHES)
o 170 Cft)
> 170 (3^)
a 350 (%)
O 300 (4^)
FIGURE 2
"ALMAG" ORGANIC
CONTAMINATION
OF THE PRODUCT,
FIRST SERIES,
•*
r
1 2-3 4 5
FEED RATE, TPH
-------
25
20
5 15
S
CJ
CO
on.
O
10
UA-
oo
O-
2 3
FEED RATE, TPH
BELT SPEED(GAP)
FPM (INCHES)
O 185
A 185
D 250
O 300
FIGURE 3
"ALMAG" ORGANIC
CONTAMINATION OF
THE PRODUCT, SECOND
SERIES,
-------
FIGURE
100 r-
c_>
UJ
£
80
cn
60
•— «7 \*f* - —•'
•V "
<**" ""-
>^
o
•"" ***
€> i/u o^; 1
+ 170 (^) I FIRST
0 300 (^) J SERIES
-- 185 (3fc) ^
0 185 (^)
n 250 (%)
O 300 (%)
s
+~
SECOND
SERIES
"ALMAG" ALUMINUM CAN
RECOVERY EFFICIENCY
2 3 ^
FEED RATE, TPH
WHOLE CANS,
-------
FIGURE 5
BELT SPEED (GAP)
FPM (IN,)
Ul
6-9
3 60
40
20
FIRST
SERIES
. SECOND
> SERIES
"ALMAG" ALUMINUM CAN RECOVERY
EFFICIENCY SHREDDED CANS,
2 3
FEED RATE, TPH
-------
FIGURE 6. GLASS RECOVERY PROCESS FLOW DIAGRAM
SINKS
54
-------
Figure 7. Cross-Section of Bendelari Jig
JIG BED OR "TOP" WATER ADDITION
a
JIG HUTCH OR
"BACK" WATER
ADDITION
CLEAN-OUT
HUTCH
DISCHARGE
UPPER FRAME
SIEVE
LOWER FRAME
DIAPHRAGM
ECCENTRIC
55
-------
Table 1
Description of ETEF Equipment
ITEM
1. Primary
Shredder
2. Screen Feeder
3. Air
Classifier
4. Cyclone
5. Screen
6. Magnet No. 1
7. Magnet No. 2
8. Trommel
Screens
9. Eddy Current
Separator
10-17 Glass
Recovery Module
MANUFACTURER - MODEL
Williams Patent Crusher Co.
Model 780
Triple/S Dynamics Texas
Shaker (with ball tray)
- several, see ref. (6)
Fisher-Klosterman Model
XQ52 plus Carter-Day rotary
valve Model AN-24 #1803
Kennedy-Van Saun Model A
Eriez SP 630-SEL belt
magnet
Permanent magnet
Combustion Power Co.,
part of Alraag
Combustion Power Co.
Almag Model 20
- see Table 4 -
HP - DESCRIPTION
1000 hp horizontal hammer-
mill, 24-230 Ib. hammers,
grates 9 x 12 in. 720 rpm
(no load)
7 x 20 ft, 10 hp, h in.
punched plate on 7/16 in.
staggered centers
27' h, 8V Si x 6' d barrel.
De-entrainment box 5' x 4'
x 8". Valve 2' d, 2 hp.
3/16 in. sq. openings
4' x 8', 1745 rpm, 31°
6'8" H x 18" w belt
250 fpm, 1 hp, permanent
magnet
head pulley on 16" wide slider
belt conveyor, 250 fpm
3' d x 10' £, 2" section and
4" section 1.5 hp, 27 rpm
25 kw @ 480 V. 2 gpm of
cooling water
FUNCTION
primary shredder
screen shredded MSW to
remove fines
entrainment of light
fraction
screen and feed light
fraction
recovery of magnetic metals
recovery of magnetic metals
screen feed to eddy current
separator
separation of aluminum
-------
Table 4
Description of Glass Recovery Equipment
Ul
GLASS RECOVERY
UNIT OPERATION
(GRUO) NUMBER
1.0
1.1
1.2
1.3/1.5
1.4
1.6
1.7/1.8
1.9/1.10/
1.11/1.12
1.13
1.14
1.15
DESCRIPTION OF
UNIT OPERATION(S)
PERFORMED
Screening at 1/2"
Jigging; sink-float separation;
screening at 1/8"
Screening at 60 M and dewatering
Screening at 1/4" (1.5)
Screening at 20 M (1.3)
Size reduction
Size separation and dewatering
Prefloat separation and con-
ditioning of pulp for flotation
Flotation
Classification & dewatering
Drying
Settling and classification
DESCRIPTION OF
PIECE OF
EQUIPMENT USED
AND MOTOR SIZE
Link belt vibrating
screen (2 hp) 2' x 8'
42" x 42" duplex
Bendelari jig (3 hp)
2' x 4' single deck
vibrating screen (3 hp)
2' x 4' double deck
vibrating screen (3 hp)
1/4" a 20 M screens
Impact hammermill (15 hp)
Cyclone 2-1/2" inlet, 3"
overflow, 7/8" adjustable
apex
Flotation cells size.
No. 12 (2)
Flotation cells size
No. 12 (4)
24" x 13' spiral classi-
fier (3 hp)
30" dia. x 20' horizontal
direct fire dryer (3/4 hp)
Settling tank s rake classi-
fier (3/4 hp)
-------
Table 5
Feed Characteristics - Composition
of Streams, wt.% (Wet)
Material
Stream
<51 mm
Glass,
Rock,
etc.
24.3
Wood
4.1
Metal
3.2
Plastic
4.1
Paper
16.9
<6.4 mm
Fines*
33.7
Other**
13.7
51 x 13 mm
18.9
5.5
9.8
8.4
37.1
Trace
20.3
Ol
CO
<13 mm
26.6
3.5
0.4
2.3
8.2
48.1
10.9
* Composition of <6.4 mm is predominately inorganic (glass, sand, ceramic)
** Includes food waste, leather, yard waste, textiles and unidentifiable
-------
Table 6
ETEF Bendelari Jig Product Characteristics
CUP PRODUCT
(avg. 4.1 kg)
% Organic*
HUTCH PRODUCT
(avg. I = 17.1 kg, II = 6.4 kg)
H20 % Organic*
ORGANIC RESIDUE
(avg. 4.4 kg)
THROUGHPUT, Mg/h
Cup Hutch I Hutch II
% Organic* Product Product Product Residue
28.7
6.0
24.7
0.21
53.3
69.1
29.0
8.2
25.0
.30
59.2
80.3
\o
20.4
3.7
23.5
.11
47.8
72.4
* (1 - LOI) x 100
.15
.42
.18
.15
22.3
4.6
19.8
.07
52.5
83.6
.10
.58
.19
.16
-------
MID-SHAKEDOWN EVALUATION OF A DEMONSTRATION RESOURCE RECOVERY FACILITY
J. F. Bernheisel
National Center for Resource Recovery
1211 Connecticut Avenue, N. W.
Washington, D. C. 20036
ABSTRACT
The National Center for Resource Recovery developed a resource recovery demonstration
facility in New Orleans, Louisiana. This facility processes 750 tons of refuse per day,
and. recovers metals and glass, and produces an unmarketed refuse-derived fuel (RDF). The.
test and evaluation program for this facility began in the latter part of 1977, prior to
the plant's completion in March of 1978, with the development of test plans and procedures.
Testing began during the shakedown period, and will continue for three years. To date,
tests on the trommel show that screening prior to shredding produces a glass- and metals-
rich fraction, and an organics-rich, low-ash fraction which can be further processed into
a fuel. The tests conducted to date with the ferrous recovery system indicate that
entrapped contaminants degrade the product quality. To satisfy the high quality require-
ments for this recovered product, modifications of the facility are being made.
INTRODUCTION
The resource recovery demonstration
facility in New Orleans, Recovery 1, is
currently in shakedown. NCRR participates
in this project to demonstrate technical
feasibility and to obtain performance
information on recovery systems and
equipment. Tests of the equipment
in this demonstration facility are being
conducted under contract with the U. S.
Environmental Protection Agency, Municipal
Environmental Research Laboratory. After
a brief description of the process, the
remainder of the paper describes the
results obtained from the test work con-
ducted to date. This will cover two
efforts. First, the trommel, on which one
test series in considerable depth has been
conducted. Second, the ferrous metals
recovery system, on which the first test
in a comprehensive series and numerous
quick samplings have been made. It is
important to bear in mind that the data
presented results from a testing and
evaluation program in process. More
refined data will be developed as the
program continues.
PROCESS DESCRIPTIONS
The Recovery 1 simplified operational
flow diagram is presented in Figure 1.
Unit operations in the diagram are numbered.
These numbers are.repeated in Table 1,
which gives the equipment description.
Arrows indicate the paths that materials
travel. Where appropriate, light and heavy
components are identified. In Figure 1, it
should be noted that the size of the
individual figures have no relationship to
the physical sizes of the machinery por-
trayed or the complexities of the
equipment.
Reduction Module. The raw, unprocessed
refuse is brought into the Receiving
Building by truck and deposited on the
tipping floor. From this point, the refuse
is pushed into one of the two conveyor pits
by a front-end loader and is moved by the
selected conveyor to the shredding equip-
ment. There, 95 percent by weight of all
large waste objects are reduced to less
than four inches in size, and 99 percent
are reduced to less than five inches in any
direction. There is no absolute top size
limit. The three major functions served by
this processing are: (1) liberation of
60
-------
Figure I
Recovery 1 Simplified Operational
Flow Diagram
o\
-------
TABLE 1. EQUIPMENT ITEM LIST
Equipment Number Equipment Description
1 Truck scale
2 #1 line raw refuse feed conveyor system
3 #2 line raw refuse feed conveyor system
4 Trommel screen and discharge conveyors
5 #1 line shredder and discharge conveyors
6 #2 line shredder and discharge conveyors
7 Magnetic separator (drum type)
8 Magnetic separator (drum type)
9 Ferrous metals concentrator and cleanup
10 Ferrous metals—can compactor and loadout
11 Trifurcated chute
12 Air classifier #1
13 Air classifier #2
14 Landfill loadout system
15 Recovery building feed and discharge conveyors
16 4x2 vibrating screen
17 Heavy nonferrous metals concentrator
18 Impact crusher
19 Vibrating screen
20 Magnetic separator
21 Eddy current separator No. 1
22 Eddy current separator No. 2
23 Air knife
24 Aluminum hammermill
62
-------
TABLE 1. EQUIPMENT TEST LIST (Cont'd)
Equipment Number
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Equipment Description
12-mesh vibrating screen
Aluminum loadout
Rolls crusher
Vibrating screen
Surge bin
Minerals jig
Dewatering screen
Vibrating screen
Rod mill
Hydrocyclone No. 1
Flotation cell unit
Hydrocyclone No. 2
Hydrocyclone No. 3
Hydrocyclone No. 4
Vacuum filter
Glass product dryer
Thermal Ifluid heater
Glass storage and loadout system
composite materials; (2) size reduction to
facilities recovery processing; and (3) a
combination of size reduction and mixing to
meet sanitary requirements of the landfill.
There are two independent shredding
systems. The shredder of each is capable
of processing at the average rate of 62.5
tons per hour (tph). Since there is a
single 'discharge system, both shredders
cannot operate simultaneously at full
capacity.
Shredder Line No. 1 has a trommel
screen (No. 4) preceding the shredder
(No. 5). This is a rotating drum screen,
approximately 45 feet long. Within the
trommel, plastic and paper bags are broken
open by lifters as the containers are
tumbled. The smaller objects fall through
the holes and are transported directly,
without shredding, for recovery processing.
The remaining material is conveyed from the
trommel to the shredder (No. 5) for size
reduction. The shredded material is then
carried on a separate conveyor to magnetic
separator (No. 7). Currently, this shredded
material is sent directly to the landfill
loadout area for transport to the landfill.
This material is the potential refuse-
derived fuel (RDF). The air classifier
(No. 13) has not been installed, pending
evaluation of its cost effectiveness.
Shredder Line No. 2 does not have a
trommel. It is used as backup and for
63
-------
oversized waste. Its flow is shown as a
broken line in Figure 1—the shredded output
is moved to the air classifier area on the
same conveyors as the shredded trommel over-
sized materials. Only a portion of this
flow can be air classified by the air
classifier (No. 12). However, the gates in
the trifurcated chute can be adjusted to
permit the second air classifier (No. 13) to
receive the remainder of the flow when it is
installed.
Ferrous Recovery System. Each
discharge conveyor carries the material
deposited on it, either as trommel underflow
or shredded material, past a drum magnet.
There are two discharge conveyors and, hence,
two magnets (Nos. 7 and 8). The magnetic
materials are deposited on a conveyor where
they fall through a chute into a small air-
classification device, the ferrous metals
concentrator (No. 9). It has two functions:
first, to blow off loose light materials
which have been carried along with the
ferrous materials at the magnetic separation
step and, second, to separate the light
ferrous metals, primarily can scrap, from
heavier ferrous metals. This latter frac-
tion is composed of shredded pieces of
discarded bicycles, auto parts, trash cans,
and other large items. The heavy ferrous
fraction is discharged into a roll-off con-
tainer for later transport to market. After
densification by the compactor (No. 10), the
light fraction is discharged into a railcar
for later transport to market.
Air Classification. The initial
equipment item in the air-classification
area is a trifurcated chute (No. 11). This
is a large, compartmented, steel-plated chute
fitted with hydraulically actuated "flop
gates." Its purpose is to: (1) distribute
the waste stream flows between the two air
classifiers; or (2) send part of the waste
to the landfill if an air classifier mal-
function occurs; or (3) bypass the air
classifier entirely. As the flow diagram in
Figure 1 indicates, the hopper is positioned
between the magnetic drum separators and the
air classifiers.
The principle of air classification is
that the waste material is introduced into
a rising column of turbulent air. Any
object light enough to be buoyed upward by
the rising air current is blown out the top
and delivered to the landfill. Typically,
this material is composed mostly of shredded
bits of plastic, paper, and textiles. This
would be the principal fuel fraction for
energy recovery. Should it become
economically feasible for the New Orleans
facility to produce an energy product, this
fraction could be used directly or be
further processed. (This processing could
include further size reduction, densifica-
tion, or even conversion to gaseous or
liquid fuel.) The air-classified shredded
trommel overs appear to be an excellent
candidate for low-ash fuel (RDF) because ,f
they have a high paper content and should
be almost devoid of fines. As of now, the -
air-classified light fraction is to be
landfilled, as described above.
Those objects which are heavy and fall
in the air classifier, despite the upward
air flow, are collected and conveyed into
the Recovery Building. This heavy, primarily
inorganic, fraction forms a glass, aluminum,
and other nonferrous metals (ONF) concen-
trate. This material is transported into
the Recovery Building for separation,
cleaning, and preparation for shipment to the
user industries. At this point, the ferrous
metals have already been recovered, and the
bulk of the waste (about 65 percent) has
been sent to the landfill.
Within the Recovery Building, there are
two product-oriented submodules. The first,
the aluminum submodule, extracts the light-
gauge aluminum, a small amount of ferrous
metal, and the ONF. The second, the glass
and ONF module, produces additional non-
ferrous metal, and its principal product,
glass.
Aluminum Recovery System. The waste
stream, which is made up of the heavy
fraction from the air classifier, is
deposited on a two-deck vibrating screen
(No. 16). This screen separates the material
into three sizes: larger than 4 inches,
smaller than 2 inches, and between 4 and
2 inches (4 x 2-inch).
The plus 4-inch stream is routed to an
air concentrator (No. 17) for the extraction
of a small amount of nonferrous metals. The
minus 2-inch, glass-rich stream is conveyed
directly to the Glass Recovery System.
The 4 x 2-inch material stays in the
Aluminum Recovery System. It is an aluminum-
rich fraction of beverage cans, some alumi-
num of other types, and a small amount of
other metals, including ferrous material,
that escaped the large drum magnets. There
64
-------
are also glass, plastics, textiles, rubber,
wood, etc. that escape. The entire 4 x 2-inch
stream is processed by an impact crusher to
break friable materials, such as glass and
ceramics. Because aluminum is malleable, it
is distorted, but not broken.
Particles smaller than 1-1/2 inches are
separated on vibrating screen No. 19, This
material, primarily glass, joins the minus
2-inch fraction separated earlier and is
sent to the Glass Recovery System. Any
waste larger than 1-1/2 inches continues in
the flow, passing over a rotating drum mag-
net (No. 20) to remove residual ferrous
metals. Removal of these metals is impor-
tant to the proper operation of the eddy
current aluminum separation device which
follows.
The eddy current separator (No. 21) is
a series of electromagnets set precisely
above and below a belt conveyor. These
produce a magnetic field through which the
aluminum, other conductors, and nonconduc-
tors on the belt are conveyed. The move-
ment of the conductors through the field
creates an electric flow—an eddy current—
within them. This, in turn, generates an
associated magnetic field around the
conductor. The polarities which occur cause
the conductors to be repelled from the
conveyor into a chute. This material drops
into an air-knife classifier (No. 23) for
further separation into three separate
tractions: aluminum can stock, other
metals, and organic carry-over.
Aluminum can stock is the primary
output of this system. Because it is
relatively low in density, it is passed
through a small hammermill (No. 24) for size
reduction of the individual cans and to
increase the density of the product to 15-
25 pounds per cubic foot. This reduces the
product shipping costs. Fines are removed
by screen No. 25. The shredded aluminum is
moved by pneumatic conveyor No. 26 out of
the building to a trailer truck for shipment
to market. A small amount of mixed non-
ferrous fraction is also produced by the
air knife.
Glass Recovery System. The small
fractions (minus 2 and minus 1-1/2 inch
material) obtained by screening in the
aluminum system are conveyed to the glass
recovery system. The glass is again crushed
in the rolls crusher (No. 27). Screen
No. 28 removes larger than 1-inch non-glass
material. The undersized (smaller than
1 inch) fraction is processed by a minerals
jig (No. 30). This piece of equipment,
employed by the mining industry for many
years, uses a vertically pulsed water flow
to separate light and heavy materials. It
produces a light organic fraction, a heavy
glass-rich fraction, and a small amount of
heavy nonferrous metals. After dewatering
by screen No. 31, it is transported to the
landfill.
The glass fraction is pumped as slurry
to the second (or bottom) deck of a two-
deck vibrating screen (No. 32). This
second deck separates at 20 mesh, with
larger particles going to a rod mill
(No. 33) for crushing. The crushed mate-
rial from the rod mill is pumped back to
the screen. This time it enters on the
first, or top, deck. Any material larger
than 1/4 inch slides off the top deck and
is saved. This material is predominantly
nonferrous metal. The minus 20-mesh
material (mostly glass, but containing some
ceramics, stone, shell, and a very small
amount of metal) is pumped as slurry to the
glass cleanup area.
The first step in the glass cleanup is
to remove the water used in previous steps
and the very fine particles (less than
160 mesh). This process, called de-sliming,
is accomplished by a hydrocyclone (No. 34).
The larger-sized solids are mixed with
clean water in a prefloat tank to remove
any remaining organic particles by floating
them to the top of the tank and skimming
them off.
The remainder of the glass cleanup is
effected by froth flotation, a technique
utilizing differences in the chemical
properties of glass and the contaminants to
achieve material separation. The glass and
contaminants are mixed with a chemical
reagent, which adsorbs preferentially to
the surface of the glass. The coated glass
attaches to bubbles formed by agitating the
mixture with air. This glass-rich froth
rises, is swept off the top, and is washed.
Within the froth flotation cells (No. 35)
downstream of the prefloat tank are: (1) a
conditioner tank in which the chemical
reagent is introduced to adsorb on the
glass; and (2) a series of froth flotation
cells where progressively more and more of
the contaminants in the glass are removed.
65
-------
The residue—the material or "tailings"
not floated off—is pumped to the process
water cleanup system. Here, suspended solids
are settles out and discharged to landfill.
Some of the water is recycled in the
plant, while the remainder goes to a lagoon
for reduction of biochemical oxygen demand
(BOD) and chemical oxygen demand (COD) by
oxidation.
The recovered glass is dewatered by a
vacuum filter (No. 39) to a moisture content
of less than 10 percent. The dewatered
glass is then dried in a spiral dryer
(No. 40) to less than one percent moisture.
This device is an enclosed screw conveyor
which has hot (650°F) heat exchange oil or
thermal fluid circulating through the screw
flites and the jacket surrounding the dryer
trough. The heat conducted through the
spiral and trough walls evaporates the mois-
ture in the glass. After drying, the glass
is stored in the glass storage and loadout
system (No. 42).
TROMMEL TEST AND EVALUATION
The efficiency of a trommel to screen
raw refuse was investigated in Great Britain
in 1966.2 Based on that experimental work,
as well as wide experience in the minerals
industry, a trommel was included in the
process design for the recovery plant
planned for the city of New Orleans in
1974. ' At the same time, an experimental
program was initiated by NCRR at its Equip-
ment Test and Evaluation Facility (ETEF) in
Washington, D. C., to provide supporting
data for the design and specification of the
trommel. The experimental results
indicated that the trommel could achieve
separation of the minus 4-inch material with
enrichment of the glass and metal portions
of the MSW.6 Most of the refuse bags were
broken, thus liberating the contents. No
serious blinding of the screen openings
occurred, although textiles did hang on the
wire cloth, they were constantly falling off
and being replaced by other pieces.
One of the chief objectives of this
pilot-scale experiment was to provide data
and insight into sizing a production-scale
trommel. Factors of interest were through-
put capacity, speed of rotation, mass
throughput per unit area or screening
capacity, and angle of declination.
Description of Trommel. The trommel
consists of four main components: a four-
legged base structure, a dust shroud, the
barrel, and the motors and drive system.
The rotating barrel is 9.75 feet in inside
diameter and 46.5 feet in overall length.
It is constructed of 10 cylindrical sec-
tions, each 4.25 feet in length, with solid
inlet and discharge lips attached to the
first and last of the 10 sections. Each of
the 10 sections contains seven removable
screen plates, 4' x 4", constructed of
0.875" thick steel, curved to form the
barrel. Each screen plate has fifty-six
4.75-inch diameter round holes located on
6-inch staggered centers, which corresponds
to a 43.1 percent open screen plate area.
Each plate has associated with it a lifter
which is a blade-shaped steel projection
intended to aid in the tumbling action and
the opening of bags as well as to act as a
screw to propel material forward.
The barrel is declined at five degrees
and is rotated at nominally 11 rpm (45 per-
cent of critical speed) by two 40-hp elec-
tric motors whose drive shafts may be
interconnected to allow both shafts to be
driven by one motor should the other motor
fail.
Trommel Test Description. The
principal objective of this first struc-
tured test of the trommel was to document
its performance at nominal design conditions
with comparison to experimental/design work
where appropriate. The key measures of
performance to be determined were:
(1) mass split of the infeed between
trommel underflow and overflow, (2) com-
position and size characteristics of the
underflow and overflow, and (3) screening
efficiency of the trommel by component.
Additionally, some testing and analytical
procedures were to be evaluated as to their
practicability and validity.
Nominal design conditions are
equivalent to operating the trommel in its
"as-built" configuration at the rated
capacity of 62.5 tph. The objectives were
met by controlling the infeed mass rate to
the trommel, allowing sufficient run time
for start-up transients to be eliminated,
then measuring the mass rates of the
discharging trommel underflow and overflow
to determine mass split of the infeed. Sub-
sequent to the test run, a determination of
66
-------
composition of representative samples of
the underflow and overflow and the fraction
of each component misreporting (-4 inch) to
the overflow was required to compute
screening efficiency by component. Comple-
mentary analyses such as bulk density and
moisture were also performed to further
characterize the feedstock and output
materials.
For Test Run 1, 25,640 pounds of raw
refuse were used; for Test Run 2, 21,300
pounds were used. The measured throughputs
were 62.4 tph and 58.2 tph, respectively.
The throughput rate for each run was
measured and verified by measuring the time
necessary to process the known quantity of
MSW.
Sampling of the raw refuse was done
from the tipping floor. Sampling the
product streams, shredded trommel overs and
trommel unders, was done by running the
respective conveyor systems for a total of
20 seconds and collecting the full flow of
each. All samples were weighed immediately
after collection.
Summary Results. Data were collected
by sampling the two structured trommel test
runs and measuring a number of parameters
of interest. Table 2 shows the moisture
for the trommel overflow, nominally plus
4-inch fraction, and underflow, nominally
minus 4-inch fraction. The moisture in MSW
is concentrated in the yard waste, paper,
and food waste. The moisture in the
trommel fractions is generally consistent
with that observed for raw MSW in New
Orleans. For example, the moisture of the
raw MSW sample, taken in conjunction with
Run 2, was 33.3 percent; the calculated
trommel infeed for Run 2 was 31.6 percent.
The bulk density of the trommel
fractions is also shown in Table 2. It
will be noted that the bulk density of the
underflow is considerably higher than the
overflow. The factor of 4 to 5 Ibs is to
be expected from the difference in composi-
tion and size. Generally, the underflow
contains small, dense material; the mate-
rial in the overflow is larger and lighter.
Table 3 shows the percentage composi-
tion by component category of the trommel
underflow and overflow for the two runs..
These were determined by hand sorting
samples and weighing each component for
each run. Also shown is a calculated
infeed composition for each run. These
data were developed from the mass split and
the composition of the individual fractions.
The Run 1 overflow fraction for the "other
heavy organics" category is approximately
41 percent of the sample. This is due to a
large amount of textiles which comprised
two-thirds of this component. Since the
main objective of the sampling was to deter-
mine where materials reported and at what
efficiency, no attempt was made to adjust
the results for these statistical outliers.
The mass split between the two fractions
is shown in Table 3. This split is approxi-
mately 55 percent to the underflow and
45 percent to the overflow, by wet weight.
One function of the trommel at Recovery 1
is to concentrate metals and glass in the
underflow for recovery. This has been suc-
cessfully achieved for the glass, as shown
in Table 3; however, it is not as marked in
the aluminum and the ferrous metals, due to
the presence of more large items.
The screening efficiency of the trommel
by components is shown in Table 4. Screen-
ing efficiency was determined by analyzing
the particle size of the overflow category.
The material overflow which was smaller than
four inches should have reported to the
underflow. The amount of this misreporting
material was added to the amount of the
underflow material to obtain an estimate
of the total amount of minus four-inch
material for the category. This total was
divided into the amount of the material
reporting to the underflow in each category
to determine screening efficiency.
In the case of aluminum and ferrous
metals, a significant amount of material
larger than four inches was contained in
the raw MSW. A composite or overall effi-
ciency can be calculated as 85 percent for
Run 1 and 78 percent for Run 2. Generally,
the component efficiencies are quite good,
with the exception of the "other heavy
organics" category in Run 1. This run
included the textiles, which affected the
efficiency.
In addition to the data presented here
for the trommel fractions, one other set of
measurements was made. This concerned the
fuel properties of the trommel overflow
after shredding. Samples 1 and 2 were
taken during test Runs 1 and 2, respectively.
These were separate samples of material from
those analyzed for composition. Sample 3
67
-------
TABLE 2. MOISTURE AND BULK DENSITY MEASUREMENTS
Run 1
Run 2
Trommel Underflow
Moisture
Bulk Density
Minimum
Maximum
Average
Number of measurements
Trommel Overflow
Moisture
Bulk Density
Minimum
Maximum
Average
Number of measurements
24.1%
20.5 lb/ft3
25.7
22.6
6
21.0%
4.40 lb/ft3
5.20
4.72
4
26.5%
16.4 lb/ft3
21.9
19.4
4
36.9%
4.69 lb/ft3
6.60
5.46
7
TABLE 3. TROMMEL FRACTION COMPOSITION
(Wet Weight or "As
Underflow (%)
Component Run 1 Run 2
Mass split 58.2 54.7
Paper products 12.0 28.5
Yard waste 14.9 6.9
Plastics 2.1 2.6
Other heavy organics 10.7 11.3
Ferrous metals 3.7 7.3
Alumunim 0.5 1.1
Other nonferrous 0.1 0.2
Glass 24.2 28.3
Stones & ceramics 5.7 2.3
Minus 1/4" fines 26.2 11.6
Received" Basis)
Overflow (%)
Run 1 Run 2
41.8 45.3
47.6 69.4
2.6 7.1
4.5 5.2
40.8 9.9
3.2 6.6
0.1 1.1
0.0 0.0
0.4 0.3
0.0 0.0
0.9 0.4
Calculated
Infeed (%)
Run 1 Run 2
26.9 47.0
9.8 7.0
3.1 3.7
23.2 10.7
3.5 7.0
0.3 1.1
<0.1 0.1
14.3 15.6
3.3 1.2
15.6 6.5
68
-------
TABLE 4. TROMMEL SCREENING EFFICIENCY BY COMPONENT
(% Wet Weight or "As Received" Basis)
Component
Paper
Yard waste
Plastics
Other heavy organics
(food, rubber, leather,
textiles, and wood)
Ferrous metals
Aluminum
Other nonferrous metals
Glass
Stones and ceramics
Minus 1/4" fines
Run 1
73.6
92.3
78.2
51.9
84.8
92.3
100.0
98.8
100.0
97.6
Run 2
61.8
64.1
70.5
80.8
79.8
89.8
100.0
99.1
100.0
97.5
was taken prior to the other two and was
not part of the structured test, and
Sample 4 was taken after the structured
test. Additional samples are being taken
and analyzed as part of the on-going
program. Table 5 shows the heating value,
moisture, ash, and sulfur results for these
samples.
From the data developed by the testing
at Recovery 1 and presented here, it can be
concluded that the trommel achieves two of
its design goals. It concentrates metals
and glass for materials recovery, and it
produces a quality refuse-derived fuel
fraction. While there is a loss of poten-
tial fuel to the trommel underflow, it
would appear that this would be acceptable
in the many situations where a low ash
content is of primary concern. Processing
of the unders fraction could produce addi-
tional material to be used as fuel.
Additional data from these tests can be
found in Reference 7.
FERROUS METALS RECOVERY
Ferrous metals recovery from MSW has
been practiced for a long time. It has
been reported that as of July 1978
facilities in 36 cities were recovering
ferrous metals.8 A first look would
indicate that ferrous metals recovery was a
simple task, given the magnetic quality of
most ferrous metals. A magnetic separator
suspended over an open stream of MSW should
suffice. However, in practice, ferrous
recovery is more difficult than it first
appears.
The difficulty in ferrous recovery is
the contaminants which are brought along
with the ferrous metal during magnetic
separation. Some contaminants are inherent
in the construction of the ferrous items in
MSW. For example, "tin cans," which con-
stitute 85 percent or so of the ferrous
metal in MSW, are sheet steel, covered by a
tin plating with a soldered seam. Also,
some have aluminum tops, and all have
organic lacquer interior coatings. «In
addition, some form of label, either paper
or painted, is usually used. However, the
contaminants which cause the most problems
come from other sources. These are:
(1) loose nonferrous materials from the MSW
chich are carried along with the ferrous
metals during magnetic separation, and
(2) nonferrous materials entrapped in the
69
-------
TABLE 5. RDF CHARACTERISTICS
(Shredded Trommel Overflow)
Btu/lb*
Moisture $
Ash* %
Sulfur* %
Sample No. 1
7624
19.0
10.2
<0.1
Sample No. 2
5563
32.5
5.5
<0.1
Sample No. 3
8174
—
13.6
<0.1
Sample No. 4
7551
20.6
13.6
0.3
Average
7228
24.0
10.7
*Dry weight basis.
ferrous metal. This latter category is
usually residual material left in containers
after use or materials put in the containers
for disposal. All of these contaminants
are primarily organic.
The presence of these contaminants
changes the function of the magnet from
recovering ferrous metals to producing a
magnetic concentrate. This concentrate,
while predominantly ferrous metals, con-
tains sufficient nonferrous material to
require further processing. The extent of
the processing depends upon the original
contaminant level, the processing prior to
magnetic separation, and the requirements
of the user who is buying the ferrous
product. Many approaches have been taken
to beneficiate the magnetic concentrate.
The processing steps used in these
approaches include shredding, air classifi-
cation, screening, and additional magnetic
separation. The approach taken at Recovery
1 and its effectiveness is discussed below.
An example of a user specification is
shown in Table 6. Notice that the specifi-
cation refers only to steel cans. Can lids,
bottle tops, and other light sheet steel
are generally acceptable in small concen-
trations. However, the presence of larger
iron castings or steel forging may not be.
Specifications like this one will require
a separation of ferrous metal recovered from
MSW into a light ferrous fraction, which is
mainly can stock, and a heavy fraction for
most markets. The heavy fraction, which
can be marketed separately if clean enough,
comprises from 10 to 20 percent of the
recovered ferrous product.9'10
Description of Ferrous System. As
discussed earlier and shown in Figure 1,
there are two independent magnetic sepa-
rators. These operate on the two material
streams from the reduction module:
(1) trommel underflow and (2) shredded
trommel overflow.
The trommel unders magnet (No. 8) is a
42-inch diameter drum, 54 inches wide. The
shredded trommel overs magnet (No. 7) is a
48-inch diameter drum, 72 inches wide. Each
of these units is mounted above the head
pulley of its respective feed conveyor. The
arrangement of the magnets and conveyors is
shown in Figure 2. The trommel unders mag-
net is at 15 inches and 26.3 degrees from
its conveyor; the shredded trommel overs
magnet is at 22.5 inches and 24.2 degrees
from its conveyor. (The distance is
measured from the drum surface to the con-
veyor surface on the line joining the
center of the drum and the conveyor head
pulley- The angle is measured between this
line and the vertical.) The rotation of the
drums carries the attached material up and
over to a single discharge belt. The com-
bined magnetic material stream, ferrous
metals plus contaminants, is fed to a two-
stage air cleanup device called the ferrous
concentrator (No. 9). This unit uses an
upward flow of turbulent air and vibration
to remove the loose organic material. Sub-
sequently.- a high-velocity horizontal air
stream separates lighter sheet steel from
heavier castings, forgings, etc. The heavy
ferrous metal is carried by conveyor to a
tote bin. The light ferrous metal is
delivered to the compactor (No. 10) by a
second conveyor. A quality control observer
70
-------
FIGURE 2. MAGNETIC DRUM PLACEMENT
RECOVERED
METAL
CONVEYOR
71
-------
TABLE 6. ORIGIN SPECIFICATION FOR STEEL-CAN BUNDLES
Tin-coated or uncoated scrap steel cans, compressed before or after shredding
to a density of not less than 75 pounds per cubic foot. Maximum bundle size
is 24" x 24" x 60". Steel beverage cans, with or without aluminum tops, may
be included. Must be free of incinerated cans, all-aluminum cans, copper and
other nonferrous metals (except those used in steel can construction).
Aluminum content of bundles must be less than 4 percent. Dirt, plastics,
and loose organics (food wastes, paper, etc.) must be less than 5 percent.
Source: United States Steel Corporation, July 15, 1976.
removes any remaining large contaminants at
this point. The compactor is a 36-inch
diameter hydraulic ram which compresses the
light ferrous into a 12-inch thick disc
with a 25 pound per cubic foot diameter.
These fall by gravity into a railcar below.
Ferrous Test Description. A number of
samplings of the raw MSW on the tipping
floor have been analyzed for ferrous con-
tent. This was done by randomly selecting
a 500- to 1,000-pound sample with the
front-end loader. This is weighed and
moved to an analysis area for immediate
analysis. Samples have been taken of the
feed material to both magnets by stopping
these conveyors and removing the entire
burden for a specified length. This can be
related to material flow by knowing the
belt speed. Samples of the products of the
two magnetic separators have also been
taken. This is done by turning off both
magnets and their discharge conveyor. The
magnet to be tested is then activated under
controlled conditions and a timed sample is
collected on the discharge conveyor. This
sample is removed, weighed, and bagged for
later analysis. Finally, samples have been
taken of the light ferrous product. These
are taken by locking the door of the com-
pactor (No. 10) open so no compaction takes
place. The uncompacted material is
collected in a container for a predetermined
period. No samplings of the ferrous concen-
trator (No. 9) streams have been formally
conducted. These are planned for 1979.
The samples taken as described above
are analyzed. The ferrous metals with their
entrapped nonferrous materials are manually
removed from any loose contaminants. The
ferrous metals are categorized as light and
heavy and each of the three categories
weighed (light, heavy, and loose nonferrous
material). The light ferrous and the heavy
ferrous fractons are each manually cleaned.
This involves freeing entrapped material.
In the case of cans, the paper labels are
removed and any contents are scraped out.
Samples were initially washed and oven
dried, but this step was eliminated when it
was found that only one percent additional
contaminant was removed in this manner.
The American Society for Testing and
Materials has under consideration a test
procedure using an oven at about 500 F to
remove organic contaminants. A comparison
of the two methods will be made.
Summary Results. The composite of all
three ferrous recovery system samplings is
represented by the flow in Figure 3. This
shows the clean light ferrous metal as
comprising 4.8 percent of the incoming MSW.
An additional 0.6 percent of the waste is
heavy ferrous metal. As can be seen, there
is an entrapped contaminant level of
37 percent in the light ferrous, and
43 percent in the heavy-
As indicated in Table 7, the feedstocks
for the two magnetic separators (Nos. 7 and
8) are very similar in general composition.
However, a review of Tables 2 and 3 will
indicate two major differences. First, the
loose contaminants in the trommel unders
contain 40 to 50 percent glass, stones, and
grit. This contributes to the second, which
is a bulk density difference of a factor of
four to five. The shredded trommel overs
have roughly the same density as the
unshredded. The composition of the product
streams shows that both magnets produce
ferrous metals concentrates. However, the
total concentration of ferrous metals is
lower in the shredded trommel overs magnet
product, and the concentration of light
ferrous is lower still. This indicates that
the shredded trommel overs magnet is not
72
-------
FIGURE 3. ESTIMATED FERROUS FLOW
TIPPING FLOOR
/
137.0 Light Fe
-/ 16.5 Heavy Fe
'
TlOO Clean Fe
1 37 Entrapped
/ 11.5 Clean Fe
\ 5.0 Entrapped
1926.5 Contaminants
53.8 /43'0
YIO.S
6.1 / 4-9
I1'2
882.1
i
LANDFILL
HEAVY Fe
i.o 7°-5>
I0-5
6.0 /5'3>
X°-7[
0.1
)
UVE
—
* jrnt x-i-fc
r
mm
MAGlwr — — —
\
16.1 /14.2
I1'9
2.
1.
^j
8 /2'5
I0'3
5
;
1
1
\
57.8 /50-9 )
I6-9
6.7 / 5-9 L _
I0'8 P)
2.0 /
y *
^
i
- Fe CONCE
QUALITY
Figures in
Pounds per Minute
at 62.5 tph
\
J
—<
NTRAT
<
CONTF
— «
'
'73.9 J65-1
\ 8.8
9.5 I8'4
I1"
3.5
V
OT? !r
72.7 /64'4
I8'3
3.5 / 3"1'
i"4
0.4
V
'72.2 /64-2
\ 8.0
2.0 I1'8
I0"
0.2
\.
COMPACTOR
RAILCAR
f 71.2 /57-°
I14'2
J 8.3 / 6.6
] I1'7
1058.5
w
LANDFILL
LANDFILL
'..2 {»-
. /«,
1 ^
3.0
V
LANDFILL
0.5 /°'2
\0.3
^15 A'3
10.2
0.2
\
73
-------
TABLE 7. MAGNETIC SEPARATOR PERFORMANCE
Trommel Unders
Magnet
Shredded Trommel
Overs Magnet
Component
Light magnetic fraction
Light ferrous
Entrapped nonferrous
Heavy magnetic fraction
Heavy ferrous
Entrapped nonferrous
Loose contaminants
Feed
6.3
5.0
1.3
0.7
0.6
0.1
93.0
Product
86.9
76.5
10.4
10.1
8.9
1.2
3.0
Feed
5.7
4.6
1.1
0.6
0.5
0.1
93.7
Product
78.9
69.6
9.3
18.7
12.3
1.4
7.4
performing as well as the trommel unders
magnet. This conclusion is supported by
the comparison of separation efficiencies
shown in Table 8. Efficiency is defined as
the percentage of the component of interest
in the feed which is delivered to the
preferred product stream.
The efficiency of the trommel unders
magnet for both light and heavy ferrous
metals is 89 percent. The efficiency for
the two magnetic fractions is lower, at
81 percent. This indicates that the magnet
is less efficient on those items which
entrap heavy non-magnetic materials, as
would be expected. Also, this may be a
desired result; it keeps some entrapped
contaminants from overloading the cleanup
system. The comparison between the two mag-
nets reveals a marked difference in effi-
ciency. This is due to non-optimum place-
ment of the shredded trommel overs magnet.
The problem has been physical interference
between the magnet and the low bulk density
shredded material. A stronger magnet may
be required.
The efficiency of the ferrous concen-
trator is shown in Table 9. It is lower
than desirable in two critical areas.
First, the separation of the light ferrous
product from the heavy ferrous product is
not good enough. The problem, as shown in
the table, is that some 36 percent of the
heavy magnetic fraction remains with the
light ferrous product. While this amounts
to only five percent of the light ferrous
product by weight, it is a highly visible
contaminant. This is due to the entrapped
nonferrous material contained in this
material. As can be seen in the table, the
majority of the material removed by the
quality control observer is in this category.
The final light ferrous metal product
composition is shown in Table 10. This
shows the product to be 86 percent light
ferrous metal, with the caveat that the con-
taminants discussed earlier which are used
in can manufacture are not counted. The
other contaminants constitute 11 percent of
the product. The specification given in
Table 6 calls for these contaminants to be
less than five percent. The light ferrous
product from Recovery 1 does not meet this
specification. It is not as clean as we
would like to deliver to our customer,
Proler International. It will not meet the
ASTM specifications under consideration.
Processing of trommel unders and
shredded trommel overs at Recovery 1 leads
to a light ferrous metal product which is
not marketable under the specifications
being adopted by users. Neither component
of the fraction removed by the magnets from
the shredded trommel unders or those removed
from the shredded trommel overs will meet
the specifications. The conclusion NCRR
has reached is that additional processing
is required. Further, additional stages of
magnetic scalping or air classifying will
not suffice. Some working of the metal is
required to open enclosures and free
entrapped material. The use of a shredder
for this purpose is being analyzed and
tested. This shredding operation should
be followed by a cleanup stage, either air
74
-------
TABLE 8. COMPARISON OF MAGNET EFFICIENCY
Shredded
Component Trommel Unders Trommel Overs
Light magnetic fraction 81.2 29.9
Light ferrous 89.3 83.0
Entrapped nonferrous 48.6 17.6
Heavy magnetic fraction 80.7 45.9
Heavy ferrous 89.4 51.0
Entrapped nonferrous 47.1 25.0
Loose contaminants 0.2 0.2
TABLE 9. FERROUS CONCENTRATOR EFFICIENCY
Efficiency
Component Light Product Heavy Product
Light magnetic fraction 98.4 1.4
Light ferrous 98.9 0.8
Entrapped nonferrous 94.3 5.7
Heavy magnetic fraction 36.8 63.2
Heavy ferrous 36.9 63.1
Entrapped nonferrous 36.4 63.6
Loose contaminants 11.4 2.9
TABLE 10. LIGHT FERROUS METAL PRODUCT
Composition
Component in Percent
Light magnetic fraction 97.0
Light ferrous 86.3
Entrapped nonferrous 10.7
Heavy magnetic fraction 2.7
Heavy ferrous 2.4
Entrapped nonferrous 0.3
Loose contaminants 0.3
75
-------
classification or magnetic separation. The
latter is preferred, and both would be
better. Such a system is being designed
for implementation at Recovery 1.
ACKNOWLEDGEMENT
This paper is based upon tests
conducted by Perry Bagalman and Kelly
Runyon of NCRR's New Orleans staff.
This testing project is supported in
part by the U. S. Environmental Protection
Agency (Contract No. 68-01-4423) under the
direction of Mr. Donald Oberacker, Project
Officer for the Municipal Environmental
Research Laboratory in Cincinnati, Ohio.
The cooperation of the City of New Orleans,
the project sponsor, and Waste Management,
the project operator, are gratefully
acknowledged. No endorsement of this
paper by any of the above is implied.
REFERENCES
1. National Center for Resource Recovery,
Inc., New Orleans Resource Recovery
Facility Implementation Study - Equip-
ment, Economics, Environment,
Washington, D. C.: National Center
for Resource Recovery, Inc., 1977,
427 pp.
2. Warren, J. L., "The Use of a Rotating
Screen as a Means of Grading Crude
Refuse for Pulverization and
Compression," Resource Recovery and
Conservation, Vol. 3, No. 1,
March 1978, pp. 97-111.
3. Taggart, A. F., Handbook of Mineral
Dressing, New York: John Wiley &
Sons, Inc., 1945, pp. 7-27 - 7-34.
4. National Center for Resource Recovery,
Inc., Resource Recovery and Disposal
Program - Bid Specifications for
Facilities and Operation, prepared for
the City of New Orleans, LA,
Washington, D. C.: National Center
for Resource Recovery, Inc., 1974,
602 pp.
5. National Center for Resource Recovery,
Inc., Materials Recovery System,
Engineering Feasibility Study,
Washington, D. C.: National Center
for Resource Recovery, Inc., 1972,
365 pp., Supplement, 1974, 108 pp.
6. Woodruff, K. L., "Preprocessing of
Municipal Solid Waste for Resource
Recovery with a Trommel," Transactions
of SME, AIME, Vol. 260, September 1976,
pp. 201-204.
7. Trommel Initial Operating Report,
Recovery 1, TR 78-3; Washington, D. C.:
National Center for Resource Recovery,
Inc. , 1978, 26 pp.
8. Kinsey, R. D., and Hildebrand, B. D.,
"Solid Waste Processing Facilities,"
Washington, D. C.: American Iron and
Steel Institute, 1978, 724 pp.
9. Sullivan, P. M., and Makar, H. V.,
"Quality of Products from Bureau of
Mines Resource Recovery Systems and
Suitability for Recycling," Proceedings
of the Fifth Mineral Waste Utilization
Symposium, Chicago, IL: U. S. Bureau
of Mines and IIT Research Institute,
April 13-14, 1976, pp. 223-233.
10. Systems Technology Corporation, "A
Technical, Environmental and Economic
Evaluation of the Wet-Processing System
for the Recovery and Disposal of
Municipal Solid Waste," report prepared
for the U. S. EPA, Office of Solid
Waste Management Programs, Contract
No. 68-01-2211, Washington, D. C.:
U. S. Environmental Protection Agency,
1975, p. 87.'
76
-------
RESEARCH AND EVALUATION OF SOLID WASTE PROCESSING EQUIPMENT
David Bendersky and Bruce Simlster
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
ABSTRACT
This paper summarizes the results of a multiphase study of processing equipment used
in solid waste resource recovery plants, conducted during the period 1976 to 1978. In the
first phase, the research needs were established. The subsequent phases were devoted to
in-plant tests and evaluations of nine shredders, two magnetic separators, one air classi-
fier, and tests of ambient air emissions at two resource recovery plants.
The initial investigation revealed the need for considerable research on solid waste
processing equipment. Forty specific research needs were identified and categorized into
13 areas: general, shredders, magnetic separators, air classifiers, screens, dryers, den-
sifiers, conveyors, storage and retrieval, receiving facilities, controls, fires and ex-
plosions, and economic research needs. Tests of the magnetic separators show that both
systems recovered about 80 percent of the available ferrous metal, but there were consid-
erable differences in the amount of energy consumed and the purity of the recovered metal.
The air classifier tests indicate that there is a critical air to solids ratio for optimum
performance. The air emissions tests show that the processing of solid wastes does not
produce harmful particulates, heavy metals, or asbestos levels. The tests of shredders
were conducted by and are presented in a separate paper by Cal Recovery Systems.
INTRODUCTION Municipal solid waste (MSW) is a dif-
ficult material to handle and process.
The need for better waste disposal Some of the characteristics of MSW that con-
techniques combined with the need for new tribute to the difficulty are: it is a mix-
energy sources has stimulated considerable ture of many different materials, shapes
interest and activity in the development of and sizes; it does not flow well; it con-
processing plants to convert municipal solid tains abrasives; the moisture content varies
wastes into energy products and to recover considerably; it is putrescible, and it
other resources in these wastes. These tends to compact in storage. Furthermore,
plants process the raw waste through vari- much of the equipment being used to process
ous combinations of equipment which include MSW was not originally designed for this
one or more of the following operations: purpose. Thus far, operating experience,
shredding, magnetic separation, air classi- tests and evaluations of waste processing
fication, screening, drying and densifica- plants have been insufficient to provide a
tion. In addition, there is materials han- firm basis for optimum design, selection
dling and other auxiliary equipment associ- and operation of equipment for these plants.
ated with the operations such as receiving
facilities, conveyors, dust collectors, In.light of this situation, the U.S.
storage and retrieval bins, and electrical Environmental Protection Agency contracted
controls. the Midwest Research Institute to conduct
77
-------
a study of processing equipment for the re-
covery of energy and other resources from
municipal solid wastes. The study, initi-
ated in 1976, was divided into three phases.
The first phase was devoted to a study of
the state of the art of waste processing
systems and equipment to delineate the re-
search needs.(1) The second phase of the
study was devoted to in-plant tests of two
magnetic separators, an air classifier, and
air emissions.(1) The third phase involved
in-plant tests of 8 shredders.(2)
RESEARCH NEEDS
The sources of information used to es-
tablish the research needs for solid waste
processing equipment included a literature
review of over 200 relevant documents, a
survey of 54 existing and planned resource
recovery systems and associated processing
equipment; visits to seven resource re-
covery plants; and personal discussions
with representatives of organizations that
have designed resource recovery systems,
manufacturers of waste processing equip-
ment, and operators of resource recovery
plants.
The investigation revealed the need for
considerable research (and development) on
processing equipment for resource recovery
systems. There is a lack of important in-
formation which is needed by systems design-
ers , equipment manufacturers and processing
plant operators. Systems designers need more
performance data about available processing
equipment so that they make choices based
on firm data. Equipment manufacturers need
to know more about the conditions under
which their equipment must operate. And,
plant operators need more information con-
cerning the optimal conditions for operat-
ing and maintaining individual pieces of
equipment and systems, based on firm cost-
effectiveness data.
Forty specific research needs for re-
source recovery processing equipment were
identified, and are listed below. Some
of these research needs have recently been
at least partially satisfied, which are as-
ter ixed, but many still remain.
General
78
1. Determine the optimal arrangement(s)
of solid waste processing equipment.
*2. Study emissions from solid waste
processing equipment.
3. Determine solid waste character-
istics pertinent to equipment
performance.
4. Evaluate new types of waste pro-
cessing equipment.
Shredders
*5. Compare the performance of vari-
ous types of shredders.
6. Determine optimum maintenance pro-
cedures and schedules.
7. Determine most effective hammer
profiles and arrangements.
*8. Establish shred size requirements
for various fuel applications.
*9. Evaluate the effectiveness of
single vs multiple shredders.
Magnetic Separators
*10. Determine the optimal operating
conditions for magnetic separators.
11. Compare the performances of belt
and drum magnets.
Air Classifiers
*12. Determine the optimal conditions
for air classifiers.
13. Compare the performance of the
various types of air classifiers.
Screens
*14. Determine the effectiveness of
prescreening solid wastes.
15. Compare the performance of vari-
ous types of screens.
16. Determine the effects of drying on
separation.
17. Determine the effects of drying on
the storability of RDF (refuse de-
rived fuel).
18. Evaluate the effect of drying on
the combustion characteristics of
RDF.
19. Evaluate the effects of drying on
bacteria and virus in RDF.
-------
Denslfiers
Economics
20. Establish the specification for
densified RDF.
*21. Determine the optimal operating
conditions for densifiers.
Conveyors
22. Determine the performance of vari-
ous types of conveyors for waste
processing.
23. Determine the optimal operating
conditions for conveyors.
Storage and Retrieval
24. Compare the performance of various
storage and retrieval systems for
RDF.
25. Determine the optimal operating
conditions for storage and re-
trieval of RDF.
Receiving Facilities
26. Evaluate the various types of
solid waste receiving facilities.
27. Evaluate segregation of solid
waste prior to processing.
Controls
28. Characterize present processing
plant electrical control systems.
29. Determine effectiveness of pres-
ent electrical control systems.
30. Evaluate present spillage and
dust control systems.
Fires and Explosion Protection
*31. Study the incidents of fires in
waste processing plants.
32. Determine the effectiveness of
present fire protection systems.
*33. Study the incidences of explo-
sions in waste processing plants.
34. Determine the effectiveness of
present explosion protection sys-
tems.
35. Develop improved fire and explo-
sion protection systems for waste
processing plants.
*36. Develop effective and consistent
accounting methods for waste pro-
cessing plants.
37. Compile data on capital costs of
solid waste processing equipment.
*38. Determine equipment operating and
maintenance costs.
39. Determine economic life of solid
waste processing equipment.
40. Perform cost effectiveness analy-
sis of waste processing equipment.
EQUIPMENT TESTS
As a result of the delineation of the
research needs for waste processing equip-
ment, a series of in-plant tests were con-
ducted on magnetic separators, an air clas-
sifier, air emissions, and shredders.
Magnetic Separator Tests
Two magnetic separator systems were
tested. The Outagamie County, Wisconsin
system (Figure 1) uses an Eriez Model V
separator located over a split hopper with
an adjustable blade for impurities control.
At Baltimore County, Maryland, the system
(Figure 2) is designed around a Dings
"Hockey Stick" model separator with a re-
turn belt to remove material dropped at
the air gap.
The comparison of these two systems,
shown in Table 1, indicates that they are
very similar. The main difference is the
energy consumption; the system at Baltimore
County requires more than twice the energy
required at the Outagamie county System.
Both systems recovered approximately 80
percent of the available ferrous metal out
of the input stream, but the recovered ma-
terial at Outagamie County contained more
impurities (nonferrous) than at Baltimore
County. The cleaner material at Baltimore
County does not appear to be as much a re-
sult of the magnetic separator system as
the condition of the input material. The
particle size after shredding is smaller at
Baltimore County,and the material contains
twice as much ferrous. At both sites
79
-------
Path of Attracted Magnetics
Path of Loose Nonmagnetic;
Magnetic Separator (Eriez, Model V)
Electromagnet
Electromagnet
Pulley
1
I
/\
1
i
onmagnetic Magnetic
a ferial Material
Figure 1 - Magnetic Separator System, Outagamie County, Wisconsin
Path of Attracted Magnetics
Path of Loose Nonmagnetic*
Dings "hockey Stick"
Electromagnets
Air Gap (Adjustable)
Nonmagnetics to BeltConv. 8
Belt Conv. 10
Magnetics
(T
Belt Conv. 9
Figure 2 - Magnetic Separator System, Baltimore County, Maryland
TABLE 1. MAGNETIC SEPARATOR SYSTEMS TESTED
Input rate, tons/hr
Input belt speed, m/min
Input belt width, cm
Input belt angle, degrees
Separator belt speed, m/min
Separator belt width, cm
Ferrous in input, %
Ferrous recovered, %
Impurities, %
Ferrous in rejects, %
Energy consumption, kw
Outagamie
County
Baltimore
County
26
65.5
110
30
95
91
5.16
81
1.8
0.9
8.3
28
79
117
30
117
122
10.8
79
0.2
2.6
18.8
80
-------
observations indicated that the impurities
controls were not effective. At Outagamie
County, light paper and plastic floated
across the splitter blade on air currents,
being completely detached from any ferrous
material. At Baltimore County, very little
material was freed at the air gap; the re-
turn belt ran virtually empty.
In both cases, the height of the sep-
arator above the input belt was varied in a
systematic fashion. At Outagamie County,
the length and angle of the splitter blade
in the output hopper was changed. Also
varied at Baltimore County was the width
of the air gap between the second and third
magnets.
Table 2 illustrates the effects of
changing the variables one at a time. At
both sites, there was a linear correlation
between the height of the separator above
the percent ferrous recovered, the smaller
heights produced greater recovery. The
splitter blade tests at Outagamie County
indicated the closer to the vertical the
long blade was positioned, the higher the
rate of recovery. This is explained by ob-r
servations of the recovered ferrous bounc-
ing off of the hopper back plate when it
was released from the magnetic field. With
the long blade in the vertical position,
the least amount of recovered ferrous could
fly back into the nonmagnetics chute. At
Baltimore County, the test settings of the
air gap on the separator did not affect the
performance of the system. The recovery
rate did not change, and the change in the
impurities was only one quarter of a per-
cent.
Air Classifier Tests
The air classifier system at Baltimore
County (Figure 3) was tested at a fixed ve-
locity of 5.3 m/s (1,040 ft/min) with in-
put feed rates from 0.51 to 64.00 mg/hr
(0.56 to 70.55 tons/hr).
The data from the sample analysis is
plotted in Figure 4. The light fraction
curve on this graph shows the weight per-
cent of the input material that traveled
out of the air classifier with the air. Be-
cause the air velocity was held constant,
the larger air/solids ratios were obtained
by reducing the input feed rates. An op-
erating plant producing RDF will operate
as close to the left on the curve as pos-
sible without leaving the horizontal por-
tion. At this velocity setting, the air/
solids ratio is about 25 and results in a
light fraction with a heating value of 15
x 10° J/kg and an ash content of 20 percent.
The Baltimore County plant includes a trom-
mel after air classification which reduces
the fines. These fines constitute a large
portion of the 20 percent ash figure because
they were not removed from the analyzed sam-
ples.
Emissions Tests
Emissions tests were conducted at the
Outagamie County and Baltimore County plants
to determine the nature of the air emissions
produced by processing and handling munici-
pal solid wastes. The emissions of specific
interest were particle concentration, par-
ticle size, trace metal concentration, and
asbestos concentration. Hi-Vol and Acu-Vol
samplers were placed near specific pieces
of equipment to determine the emissions from
that unit.
The highest particulate concentration
reading was at the tipping floor at Outa-
gamie County with a reading of 6.617 mg/Nm^,
which is approximately 66 percent of the
threshold limit value (TLV) of 10 mg/Nm3
(Ref. 5). Analysis of the trace metal in
the particulates indicates that the amount
of toxic metals was well below their respec-
tive TLV's. The sample closest to TLV con-
tained a lead content of 0.018 mg/Nm^ com-
pared to a TLV of 0.150 mg/Nm3, or approxi-
mately a factor of 10 below the TLV. No
asbestos was found at Outagamie County, and
based on the St. Louis and Houston tests
which also showed no asbestos (Refs. 3 and
4), it was decided not to test for asbestos
at Baltimore County.
Shredder Tests
Nine shredders were tested for shred
81
-------
TABLE 2. EFFECTS OF VARIABLE CHANCES
OUTAGAMIE COUNTY
Recovery
efficiency Impurities Ferrous in rejects
Short blade
Height (cm) angle (degrees) (%) (%) (7.)
33 49.5 81 1.79 0.92
35.6 49.5 76 1.50 1.24
38.3 49.5 39 1.69 4.73
33 24'5 71 ' 1.80 1.50
Long blade
angle (degrees)
33 24'5 80 1.95 0.73
33 37-° 72 2.40 1.70
33 49-5 61 2.41
BALTIMORE COUNTY
Air gap (cm)
46 18 78
51 18 '! °'36 2.15
56 18 73 0.28 4>38
64 18 5p °'22 4.80
* *" 'J « 1 ft W'Ji-J
46 3,
79
46 18 7H )-/l2 2.19
/ £ If) (\ '1 £.
46 6 7q '• t6 2.15
°«1« ? hA
^ * O H-
82
-------
Lights
Heavies
Air
Air Input \ I || / Air Input
-\" yj \V ^>
T tX— — Heavies Output
CL
Conveyor 16
Figure 3 - Air Classifier System, Baltimore County, Maryland
90
80
70
60
50
,.,
4°
30
20
10
0
• Light Fraction
o % Comb
a Ash
A Moisture
• Heat Value
16
15 o
14
13 J
,2 |
10
20
30
40
50
60
a- Air/Solids Ratio
70
80
Figure 4 - Light Fraction Analysis, Baltimore County
90
83
-------
size, energy consumption, and hammer wear.*
The plant locations, manufacturers, numbers
and types of shredders tested are given in
Table 3. The results of these tests are
given in a separate paper at this conference
(Ref. 6).
ACKNOWLEDGEMENT
The support of this study by the EPA
Municipal Environmental Laboratory, Cincin-
nati, Ohio, is gratefully acknowledged. The
EPA Project Officers were John 0. Burkle and
Donald A. Oberacker. The cooperation of the
plant operators wherein the various tests
were conducted made this study possible.
TABLE 3. SHREDDERS TESTED
Location
Ames, Iowa
Baltimore County, Maryland
Great Falls, Montana
Odessa, Texas
Outagamie County, Wisconsin
Tinton Falls, New Jersey
CONCLUSIONS
Manufacturer, Model
Number and Type
American Pulverizer, Model 60-90 2 horizontal hammermills
Tracor Marksmen, Model 860
Heil, Model 42F
Newell, Model 68
Alias Chalmers, Model KH 12-8
Carborundum (Eidal), Mo'del 1000
The following conclusions may be drawn
from this study:
The magnetic separator systems
tested recover approximately the
same percentage of the available
ferrous metal. However, there are
significant differences in the pur-
ity of the recovered metal and en-
ergy consumption.
There is a critical air-to-solids
ratio for air classifiers. Above
the critical ratio, the light frac-
tion separation is maximum and re-
latively constant; below the criti-
cal ratio, the light fraction sep-
aration falls off sharply.
The processing of municipal solid
waste does not produce harmful
levels of particulates, heavy metals
or asbestos.
There is need for considerable ad-
ditional research on solid waste
processing equipment.
The shredder tests were conducted by Cal
Recovery Systems, Inc., under subcontract
with Midwest Research Institute.
2 horizontal hammermills
1 vertical hammermills
1 horizontal hammermill
2 horizontal hammermills
1 vertical grinder
REFERENCES
Midwest Research Institute, "Study of
Processing Equipment for Resource Re-
covery Systems," Volume I - State of the
Art and Research Needs, Volume II - Field
Tests of Magnetic Separators, Air Clas-
sifiers, Air Emission, Final Report,
EPA Contract No. 68-03-2387, MRI Proj-
ect No. 4213-D, December 1978.
Cal Recovery Systems, Inc., "Evaluation
of Shredders Used for Size Reduction of
MSW," Final Report, EPA Contract No.
68-03-2589, MRI Subcontract No. 4424-D,
to be issued in March 1979.
Midwest Research Institute, "St. Louis
Refuse Processing Plant: Equipment,
Facility and Environmental Evaluation,"
Final Report,*MRI Project No. 4033-L,
May 1975.
Midwest Research Institute, "Evaluation
of Fabric Filter Performance at Brown-
ing Ferris Industries Resource Recovery
Plant, Houston, Texas," MRI Project No.
4240-L(13), September 1977.
American Conference of Government and
Industrial Hygienists, "Industrial Ven-
tilation—A Manual of Recommended Prac-
tices," 1973.
84
-------
Cal Recovery Systems, Inc., "Evaluation
of Shredders Used for Size Reduction of
MSW," Symposium on Gas and Leachate in
Landfills and Resource Recovery, Orlando,
Florida, March 26-28, 1979.
85
-------
EVALUATION AND PERFORMANCE OF
HAMMERMILL SHREDDERS USED IN REFUSE PROCESSING
G. M, Savage, G, R. Shiflett, L. F. Diaz, and G. J. Trezek
Cal Recovery Systems, Inc.
160 Broadway Suite 200
Richmond, CA 94704
ABSTRACT
This paper presents the results of a comprehensive program to test and evaluate
large-scale shredders used for the size reduction of solid waste. In all, four hori-
zontal hammermills were tested at three commercial sites (Appleton, Wisconsin; Ames, Iowa;
and Cockeysville, Maryland). Both two-stage size reduction (primary and secondary shred-
ders at Ames) and single-stage size reduction (practiced at Appleton and Cockeysville)
were studied as part of this work. Evaluation and interpretation of the data has result-
ed in the development of analytical relationships among comminution parameters and estab-
lishment of levels of performance with respect to energy consumption and hammer wear
associated with size reduction of solid waste.
INTRODUCTION
OVERVIEW
Over the past fifteen years the use
of shredders in the field of solid waste
management has seen steady growth. From
first attempts at using size reduction of
solid waste as the initial step in the
production of a suitable material for com-
posting, the use of shredders has grown
into the field of full-scale, integrated
resource recovery facilities. In between
these events, shredders have seen service
in the areas of ferrous scrap recovery
from solid waste and treatment of refuse
for landfill disposal without the necessi-
ty of utilization of cover material.
Shredding (or synonymously, grinding,
size reduction, milling, or comminution)
of solid waste has taken on an added de-
gree of significance since the initiation
of large, full-scale resource recovery op-
erations. For such facilities, size re-
duction represents the first step in pro-
cessing the waste stream. Consequently,
the unit process of size reduction affects
all equipment involved in downstream ma-
terial handling and separation. In addi-
tion, the shredding operation normally
accommodates 100 percent of the throughput
of waste, whereas other unit processes in
resource recovery plants generally handle
only particular fractions of the waste
stream. As a result, the importance of the
unit process of size reduction is generally
recognized, albeit poorly understood, by
the solid waste industry.
The proliferation of shredders in the
solid waste industry has stimulated inter-
est in their operation, evaluation, and
performance. For example, criteria for
estimation of shredder operation and per-
formance are needed in the initial stage
of design of resource recovery plants.
Also, the plant manager may wish to know
how certain operational changes involving
a shredder (such as a change in size of
the grate openings or variation in shred-
der throughput) may affect shredder opera-
tion (such as energy requirements and size
of product).
The testing and evaluation program
86
-------
described here was funded By the EPA to
extend the present knowledge of size re-
duction of splid waste. Earlier investi-
gations conducted by two of the authors,
Trezek and Savage, and also funded by the
EPA.established the key parameters which
characterize refuse comminution and the
criteria by which they can be evaluated
(1,2,3,4). The underlying motivation of
the present research is to establish pre-
dictive relationships, design criteria,
evaluation techniques, and levels of per^-
formance for large-scale size reduction
equipment. Large scale means shred-
ders with nominal capacities in the range
of 2Q to 60 tons per hour. The informa-
tion contained herein has been developed
for use and consideration of those associ-
ated with solid waste management including
shredder manufacturers, plant designers,
plant operators, and researchers.
The test program involved detailed
measurements of energy and hammer wear
associated with size reduction of solid
waste, which are the two most important
aspects of size reduction due to their
effects on operational cost. In order to
ascertain relationships involving energy
and hammer wear, a rigid, scientific pro-
tocol was established and implemented.
Cooperation of personnel at each site was
instrumental in maintaining a consistent
test procedure among plants.
Site-specific test plans were devel-
oped and were directed toward assessing
the energy consumption and hammer wear
associated with shredding. Both opera-
tional data and levels of performance were
sought for different governing parameters.
In order to adequately assess wear and
energy consumption, the test program
called for collection of samples for de-
termination of product size and moisture
content of shredded refuse. In particu-
lar, these samples were collected during
the interval of energy measurement. Re-
garding hammer wear, the study concen-
trated on the evaluation of various hard-
facing alloys as well as that alloy typi-
cally used at each site. Rates of hammer
wear were determined and interpreted.
The shredders that were tested are
installed at facilities located in Apple-
ton, Wisconsin; Ames, Iowa; and Cockeys-
ville, Maryland.
SITE DESCRIPTIONS
Appleton: The Outagamie County Solid Waste
Shredding Facility located in Appleton,
Wisconsin, operates two Allis-Chalmers
Model KH 12/18 horizontal hammermills in
parallel to shred the approximately 160
tons of refuse received daily, Ferrous
material is recovered after shredding using
a magnetic belt conveyor, and the remain-
ing shredded waste is compacted into trans-
fer trailers and transported to the land-
fill.
Ames: The Ames Resource Recovery System
in Ames, Iowa shreds approximately 180
tons of refuse per day through two Ameri-
can Pulverizer Model 6090 horizontal ham-
tnermills operating in series. Both mater-
ial and energy recovery are practiced at
the Ames plant. Most ferrous material is
magnetically separated from the refuse
after primary shredding, while the remain-
ing material is subjected to secondary
shredding and air classification so as to
recover a refuse derived fuel. The refuse
derived fuel is pneumatically conveyed to
a nearby city-owned power plant where it
is co-fired with coal in the utility's
boilers. The plant also has the capability
of recovering aluminum from the non-
ferrous heavy fraction.
Cockeysville: The Baltimore County Re-
source Recovery Facility in Cockeysville,
Maryland shreds approximately 320 tons of
refuse per day through two Tracor-Marksman
Model A60 horizontal hammermills operating
in parallel. Although the capability
exists for both material and energy re-
covery, currently only magnetic separation
of ferrous material is carried put, and
the remainder of the shredded refuse is
taken to the landfill.
HAMMER MAINTENANCE PROGRAMS
Two of the sites tested utilize ham-
mer retipping as a means of dealing with
hammer wear; the Ames facility is the ex-
ception. The Ames operating procedure
calls for wearing the hammers in each
shredder until they no longer effectively
shred refuse. When that point is reached,
the hammers are removed, scrapped, and re-
placed by new hammers.
Contrary to the procedure followed at
Ames, the Appleton and Cockeysville
facilities utilize hammer retipping as a
87
-------
regular maintenance program. At Appleton,
welding wire is used instead of a welding
electrode. Stoody 110 is used for build-
ing up the hammer surface prior to applica-
tion of one or two passes of Stoody 134
hardfacing. Stoody 134 is the wire equiv-
alent of a Stoody 2134 electrode, which was
used in the wear experiments at Appleton.
MATERIALS AND METHODS
POWER MONITORING INSTRUMENTATION
The power monitoring equipment in-
cludes a Scientific Columbus Model DL34-
2K5A2-AY-6070 watt/watt-hour transducer, a
Houston Instruments Model 3000 chart re-
corder, and a digital dividing circuit.
The transducer itself is adaptable
to single phase, three phase-three wire,
or three phase-four wire systems. For
multiphase systems, both the voltage and
current in each leg are monitored continu-
ously, thus automatically correcting for
the power factor. Two output signals are
provided by the transducer. The first is
an analog current signal which is directly
proportional to power (kw). The second trans-
ducer output is a digital signal directly
proportional to energy consumption. The
digital signal from the transducer is sent
to the dividing circuit, which counts the
pulses and triggers an event marker on a
chart recorder after a predetermined
number of pulses have accumulated.
The divider may be set manually to count
between 1 and 9999 input pulses before
triggering the event marker.
Since the nominal inputs to the trans-
ducer are 120 volts and 6.5 amperes, the
appropriate current and voltage step-down
transformers that were needed for each
test were determined during a survey of
each facility prior to the actual test
visit.
OTHER EQUIPMENT
A Chronos model 3-ST digital chronometer
was used for making any necessary measure-
ments of time. Determination of conveyor
belt speeds (for flowrate measurements) and
time of sample collection required use of
this instrument.
Conveyor belt speeds were determined,
when possible, with a Power Instruments
Inc.,TAK-ETTE model 1707,digital rpm gauge.
An alternative method for obtaining the
belt speed was to measure the time neces-
sary for the belt to traverse a measured
distance, and divide the measured distance
by the elapsed time.
In order to obtain size distributions
for the shredded refuse samples, the sam-
ples were first screened on a set of manu-
ally held screens having square wire mesh
openings of 20.32, 10,16, 5.08, and 2.56
centimeters. Screening of the undersize
from the 2.56 centimer hand held screen
was done on a SWECO model LSI8533333 Vibro-
Energy Rotary Screen. The SWECO was
equipped with a series of square wire mesh
screens having openings of 2.54, 1.59,
0.95, 0.51, 0.27, and 0.13 centimeters.
SIZE DISTRIBUTION ANALYSIS
Samples of shredded refuse were
weighed and dried to a constant weight in a
drying room maintained at 22° C and 65%
relative humidity. After the samples were
dried they were reweighed to permit deter-
mination of the moisture content. Size
analysis of the samples was performed using
manual and mechanical screening. The dried
refuse was placed on the largest of the
manually held screens and shaken until no
further refuse was observed to pass through
the screen. The oversize from the screen
was collected and weighed, and the under-
size was placed on the next smaller screen.
The process was then repeated until all
four manual screens had been used.
The undersize from the 2.54 centimeter
manually held screen was next processed
through the SWECO screens. Material was
screened on the SWECO screens for a time
duration of fifteen minutes.
EXPERIMENTAL PROCEDURES
POWER MEASUREMENTS
Flow rate, samples and power level
data were collected under a number of
different operating conditions for the
purpose of characterizing the shredders.
To guarantee that the flowrate sample coin-
cided exactly with the interval during
which the power was monitored, it was
necessary to accurately determine (1) the dis-
tance from the center line of the shredder
discharge to the center line of the seg-
ment of the discharge conveyor from which
the flowrate sampled was gathered, and
88
-------
2) the speed of the discharge conveyor It-
self. With the shredder sampling distance
and conveyor speed known, it was possible
to calculate the elapsed time needed for
the sample to travel from the shredder to
the point of sample collection.
The mass flowrate through the shred-
der was calculated from the speed of
the conveyor belt and the weight of the
sample removed from a given distance of
belt.
The actual sample used for the size
distribution and moisture analysis was ob-
tained by mixing and subsampling of the
flowrate sample. In order to ensure that
a representative sample for the size dis-
tribution analysis was obtained, the size
of the subsample was chosen according to
the grate spacing of the shredder under-
going testing. In general, the larger the
grate spacing, the larger the sample that
needed to be gathered. An estimation of
the sample size was made using the follow-
ing table:
TABLE 1. SAMPLE SIZE
Grate Spacing
cm (inches)
20
15
10
5
Sample Size
kg (pounds)
10
7.5
5
2.5
(22.1)
(16.5)
(n.o)
( 5.5)
HAMMER WEAR EXPERIMENTAL PROCEDURE
Hammer wear investigations were con-
ducted on base hammer materials as well as
on a number of hardfacing alloys. The
experimental procedure consisted of clean-
ing and weighing hammers prior and subse-
quent to their shredding a measured
amount of solid waste. Quantity of hammer
material lost and tonnage of refuse shred-
ded were used to ascertain the degree of
wear for each type of hammer base material
or hardfacing alloy tested. As is common
in the industry, the degree of wear is
expressed as the weight of material lost
per unit weight of refuse shredded.
The general test procedure called for
testing hardfacing alloys in each shredder
simultaneously thus avoiding as much as
possible the influence of uncontrolled
variables. Where possible four alloys
were tested at one time, one per each row
of hammers.
Hammer wear tests were only conducted
at the Appleton and Cockeysville facili-
ties. Hammer wear data for the plant at
Ames was made available by the City of
Ames. All tonnages used for the wear ex-
periments were based upon actual weight
measurements.
RESULTS AND DISCUSSION
MOISTURE CONTENT OF SHREDDED REFUSE
SAMPLES
Average moisture contents of the
shredded samples showed Appleton refuse
to be significantly wetter than refuse
shredded at either the Ames or Cockeys-
ville facility, namely about 35 percent
for Appleton versus roughly 16 percent
and 17 percent for Ames and Cockeysville.
All moisture contents
quoted are on an air dry basis). Although
water is sometimes used at Appleton to
control dust dispersion, the water spray
was turned off five minutes before be-
ginning the test runs. Consequently, the
above moisture levels for Appleton are
innate moisture contents. Neither Ames
nor Cockeysville used a water spray
system during the test programs.
The moisture content of Appleton ref-
use confirmed visual observations that
Appleton refuse was typically residential
in nature. On the other hand, the refuse
processed at Ames and Cockeysville was
typical of commercial solid waste.
MEASURED THROUGHPUTS
The primary purpose of this study was
to obtain performance and operating data
on large-scale shredders, that is, shred-
ders capable of handling large throughputs
of solid waste. Shredders evaluated in
this study covered the full spectrum of
throughput capacity ranging from an aver-
age of approximately 20 metric tons per
hour (TPH) at the Ames plant to almost 60
TPH at the Cockeysville facility (wet
weight basis). The minimum throughput
measured during the tests was 3.5 TPH at
the Appleton facility while the maximum
throughput was 95.7 TPH measured at the
Cockeysville plant, again on a wet weight
basis. It must be pointed out that both
of the above extreme values were special
89
-------
test runs and therefore are not indicative
of normal plant operation. These wide
variations are a consequence of the
attempt to obtain performance and opera-
tional data over as wide a throughput Base
as plant operation would permit.
SIZE DISTRIBUTION OF SHREDDED SOLID WASTE
Each shredder that was tested pro-
duced a characteristic size distribution
of particles. An overall view of the size
distributions that were obtained from the
samples that were collected from each
shredder is shown in Figure 1. The curves
represent an average size distribution for
each shredder. The reader is reminded
that the input for the Ames secondary
shredder is the output from the Ames
primary shredder.
ESTIMATION OF SPECIFIC ENERGY REQUIREMENTS
Data collected in these experiments
along with that collected by Trezek and
Savage (1,2) enable the development of re-
lationships for the prediction of specific
energy on a wet weight basis, E0w (kwh/
Ton), as a function of both characteris-
tics (X0) and nominal (Xgg) product sizes.
The characteristic size and nominal size
represent that screen size corresponding
to 63.2 and 90 percent cumulative passing,
respectively. Here specific energy is
defined as the quantity of gross minus
freewheeling energy divided by the
throughput of material. Average values
of EQW, X0, and Xgg calculated from the
experimental data are plotted in Figure 2.
The trend of increasing energy require-
ment as a consequence of producing smaller
and smaller particle sizes through size
reduction is quite apparent.
An attempt to develop a functional
relationship between E0 and the size
parameters using standard curve fitting
techniques yielded the following equa-
tions:
-0.90
= 17.91X,
and
ow=35-55X90
-0.81
(1)
(2)
where E0w is expressed in kwh/Ton and X0
and Xgo are expressed in cm. The corre-
lation coefficients for equations 1 and
2 are 0.87 and 0.89 respectively.
These equations were developed with
the aid of data from facilities handling
anywhere from four to ninety metric tons
of refuse per hour and consequently repre-
sent the full gamut of operating conditions
found in actual practice. The equations
should be valuable in estimating net power
requirements for shredders in general.
The term "in general" should not be over-
looked since the actual power requirement
for a particular shredder is a function
of many variables including internal
machine configuration, hammer tip speed,
throughput, as well as size, composition,
and moisture content of the feed.
MOTOR SIZING
With regard to the estimation of the
proper sfze motor for shredding refuse at
a specified rate, it must be remembered
that gross power requirements are composed
of the net power required for size reduc-
tion plus the freewheeling power (otherwise
known as the idle power or power required
when not shredding refuse). Typically,
the freehweeling power represents about
10 percent of the full load rating of the
motor thus leaving the remaining 90 percent
available for size reduction.
An estimation of the net power re-
quired to produce a specified particle
size can be found by multiplying the spe-
cific energy (EQ ) by the anticipated flow
rate (mw). Divicling the product by 0.9
will then produce a rough estimate of the
gross motor power which would be required.
POWER-FLOW RATE-MOISTURE RELATIONSHIPS
To show how flow rate and moisture
affect power requirements for the systems
tested, a multiple regression analysis
was executed for each set of test data.
The multiple regression analyses were
based on the assumption that the exper-
mental results could be represented by an
equation of the form,
w 0-MC)C
Pn =
(3)
where Pn = net power (kwh);
mw = flowrate of refuse through
the shredder (TPH); and
MC = fractional moisture content
of the refuse.
as previously ascertained by Shiflett (5).
The exponents "b" and "c" in equation
90
-------
100
90
80
70
60
50
40
30
20
10
0
J_
Cockeysville
Raw
Commercial
J.
1
10
Screen Size (cm)
Figure 1. Average size distributions of raw & shredded solid waste.
40
~ 30
o
I 20
10
Un. of Cal.
Un. of Cal.
Un. of Cal.
Appleton East
Ames Primary
Ames (Pri.+Sec.)
Cockeysville (Reverse)
Cockeysville (Forward)
-0.90
-0.81
I
0.1
1.0
Product Size (cm)
10
Figure 2. Specific energy consumption as a function of product size
91
-------
3 determine how power requirements change
with flowrate and moisture content, re-
spectively. Positive values for the ex-
ponents "b" and "c" indicate that power
increases with flowrate but decreases with
moisture content. The coefficient "a"
determines the magnitude of the power re-
quirements and is influenced by machine
parameters, such as size of grate open-
ings, internal geometry, and geometry and
number of hammers.
The equations resulting from the
regression analyses are presented in Table
2. An examination of the correlation co-
efficients shown in the table indicates
that there is good agreement between the
assumed form of the curve and the experi-
mental data.
From Table 2, the calculated values
of the exponent "b" can be seen to be pos-
itive and close to unity indicating that
net power requirements are almost directly
proportional to flowrate. The lone ex-
ception occurs for the power equation for
the Ames secondary shredder in which the
exponent "b" is almost two. This high
value of the exponent "b" for the Ames
secondary shredder may be due to the ex-
tremely worn condition of the hammers
within the shredder during the test runs.
The hammers In all the other shredders
that were tested were in good condition.
The calculated values of the ex-
ponent "c" ranged from -7.04 to 1.54.
Positive values of the exponent "c" ob-
tained for the shredders at Appleton and
Cockeysville indicate that power require-
ments decrease with higher moisture con-
tents. The opposite is true for the pri-
mary and secondary shredders at Ames. The
negative values of exponent "c" calculated
for the Ames shredders indicate that the
power increases for increases in moisture
content. Negative values for the ex-
ponent "c" are contrary to other pub-
lished data (1,2,4) and are unresolved
pending further research.
The large variations in the values of
the coefficient "a", that is values span-
ning the range 0.28 to 47.04, indicate the
significant influence of machine character-
istics upon power requirements. Some work
by Shiflett and Trezek (6) indicates that
the amount of material within the shred-
der at any instant in time, denoted as
shredder noldup, may be a basic parameter
for ascertaining the influences of machine
characteristics upon energy consumption
and throughput capacity.
Due to the fact that collection of
data for holdup determinations is a dis-
ruptive process from the standpoint of
testing at an operating facility, we were
only able to collect holdup data at the
Appleton site. The Appleton data fit the
following curves,
Pn = 1.60 H°'96 (4)
1.04
mw = 0.51 H (5)
with correlation coefficients of 0.82 and
0.90, respectively. These curves are simi-
lar in form to those reported in reference
(6) indicating the importance of shredder
holdup as a key parameter for assessing
the effects of machine characteristics
upon energy consumption and throughput
capacity.
SINGLE VERSUS MULTIPLE STAGE SIZE RE-
DUCTION
Some of the data collected at the
Ames and Cockeysville sites can be used
to compare single and multiple stage
size reduction of solid waste. The com-
parison involves examination of gross and
net power consumption (?Q and PR, re- .
spectively), mass flowrates of refuse (m ),
and characteristic product size (X0).
For the purposes of comparison of process-
ing alternatives, the criterion of equiv-
alent product size from both single shred-
ding at Cockeysville and secondary shred-
ding at Ames was used. The data allow a
comparison for which an average charac-
teristic product size (X0) of 1.6 cm was
calculated.
In order to compare the power re-
quirements of single versus multiple
stage size reduction, the moisture con-
tent CMC) as well as the product size
(already chosen as 1.6 cm) must be speci-
fied. For purposes of comparison, the
average moisture content was taken as
20.5 percent for a hypothetical refuse
processed at Ames and Cockeysville facili-
ties. As previously discussed, net
power requirements for size reduction can
be expressed by a relation in the form of
equation 3.
Since we have specified MC as 20.5
92
-------
percent, equation 3 reduces to
. 1.20
-7.04
PN = 0.62 mw (0.795) (6)
and
1.92
-2.84
PN = 0.28 mw CO.795)
(7)
for the Ames primary and secondary shred-
ders, respectively, where m = 21.6 T PH
since that throughput produces the re-
quired average characteristic product size
of 1.6 cm. For shredder #1 at Cockeys-
vilie, the equation for net power (calcu-
lated by combining data for both forward
and reverse directions of rotation) is
= 21.85m
0.90
w
(.0.795)
2.25
(8)
wherein m = 52.2 T PH since that through-
put produces the desired average charac-
teristic product size of 1.6 cm.
The solution of equations 6,7, and
8 allows calculation of the total net
power required to produce the required
product size. The total net power re-
quired at Ames is 320 kw, and that at
Cockeysville is 458 kw. The specific en-
ergy (on a wet weight basis, P|\|/mw) for
Ames and Cockeysville are 14.8 and 8.8
kwh/Tw, respectively,refer to Table 3.
Consequently, the specific energy required
to produce the required product size at
Ames is 168 percent of the energy require-
ment at Cockeysville.
When the freewheeling power of the
primary and secondary shredders at Ames,
53 and 41 kw respectively, is added to the
net power required for size reduction, the
total, or gross, power required is 414,
Likewise, when the freewheeling power for
the Cockeysville shredder (66 kw) is added
to the net power, the gross power required
is 524 kw. On the basis of unit mass,
the gross energy requirements (gross spe-
cific energy) for Ames two-stage shred-
ding and Cockeysville single-stage shred-
ding are 19.2 and 10.0, respectively,
refer to Table 3. Based upon producing
the same particle size, two-stage size
reduction at Ames uses almost twice as
much gross energy as single-stage reduc-
tion at Cockeysville.
The results presented here must be
viewed with caution for two reasons. First
of all, to date this type of information
on large-scale shredders has only been
developed for one specific case. Secondly,
it is not known whether or not the shred-
tiing lines at Ames and Cockeysville have
been optimized to maintain energy con-
sumption at a minimum.
The possibility of optimizing multi-
ple stages of shredding along with methods
for establishing optimization have been
investigated previously (3). Results re-
ported in reference 3 indicate that multi-
ple stages of size reduction can be less
energy intensive than single-stage size
reduction under certain circumstances of
grate opening size and particle size.
HAMMER WEAR
Results of the hammer wear experi-
ments show a range of 0.013 to 0.106 Kg/T
for hammer wear, Table 4. The wide range
of values is a result of the numerous
hammer materials and hardfacing alloys
that were tested.
The greatest rates of wear measured
during the tests were sustained by non-
hardfaced manganese steel, i.e. bare ham-
mers whose hardness as cast was 14 Rc. For
the hardfacing alloys that were tested we
found that as the hardness of the alloys
increased, the degree of wear correpond-
ingly decreased. The limiting factor at
the high range of alloy hardness appears
to be chipping of the welds under high im-
pact loads. The trend of decreased hammer
wear for harder alloys is apparent if one
examines the data from the Appleton west
mill and the Cockeysville #1 shredder,
Figure 3.
The optimum range of alloy hardness
appears to lie within the range 48
-------
TABLE 2. NET POWER AS A FUNCTION OF FLOWRATE AND MOISTURE CONTENT
Shredder
Appleton East
Ames Primary
Ames Secondary
Cockeysville #1 ,
forward rotation
rr\f \sn\if w-i 1 1 f\ -Ul
Power Equation
Pn = 6.33 m °-94(l-MC)
W
Pn = 0.62 mw1<20(l-MC)
.1.92
Pn = 0.28 mw (1-MC)
0.82
Pn = 47.04 m (1-MC)
w
i -ip
D = c /mm ' • ' ° M . Mr A
1.54
-7.04
-2.84
4.79
2.01
Correlation
Coefficient
0.85
0.74
0.93
0.92
0.96
\S\S\f t\\*Jf *JVtlf\f/l>} ll W • T^l-f III \ t I 1\J J
reverse rotation
Cockeysville #1 Pn = 21.85 m°-90(l-MC) 2'25 0.91
combined rotation w
TABLE 3. COMPARISON OF SINGLE-STAGE AND DOUBLE-STAGE SIZE REDUCTION
Cockeysville Ames
Single-Stage Two-Stage
Size of Grate Openings (cm) 20.3x35.6 22.9x25.4 Primary
8.9x12.7 Secondary
Input Particle Size a(cm) 13 13
Intermediate Particle Size a(cm) - 5.0
Final Product Sizea(cm) 1.6 1.6
Flowrate (Tons/Hour) 52.2 21.6
Moisture Content (% air dry) 20.5 20.5
Net Power (KW) 458 320
Freewheeling Power (KW) 66 94
Gross Power 524 414
Net Specific Energy (KWh/Ton) 8.8 14.8
Gross Specific Energy (KWh/Ton) 10.0 19.2
Characteristic size: size corresponding to 63.2% cumulative passing
94
-------
TABLE 4. RESULTS OF HAMMER WEAR EXPERIMENTS
Alloy
Hardness
(RC)
Avg Wt Loss
(w-j)
(kg/Hammer-Ton)
Wear of Full
Hammer Complement
fl
(kg/Ton)
APPLETON
EAST MILL
Stoody 2134
WEST MILL
McKay 48
McKay TiC
Stoody 2134
Amsco Super 20
AMES
Manganese Steel
COCKEY5VILLE
SHREDDER #1
Mangjet
Abrasoweld
McKay 55
Faceweld 12
SHREDDER #2
Manganese Steel
48
38
45
48
56
14
21
36
49
57
14
9.44 x 10'
,-4
4.84 x 10
3.30 x 10
3.60 x 10
2.73 x 10
-4
-4
-4
-4
5.50 x 10
-4
2.88 x 10
1.87 x 10
0.89 x 10
1.57 x 10
-3
r3
-3
-3
0.023
0.023
0.016
0.017
0.013
0.026
4.40 x 10
-3
0.069
0.045
0.021
0.038
0.106
95
-------
the Appleton wear experiments. The test
results for Stoody 2134 for the east and
west mills show wide disparities for wear,
0.023 and 0.017 kg/Ton, respectively
(Table 4). The differences in the rates of
wear possibly point out the effects of
refuse composition and operational varia-
bles, such as hammer geometry, base mater-
ial of hammers, rpm, etc. Simultaneous
testing eliminates these variables and
allows comparison of different hardfacing
alloys without biasing from uncontrolled
variables.
Comparisons of hammer wear data among
sites must take into account the degree of
size reduction since operating experience
has shown that wear increases as product
size of shredded refuse decreases. In
addition to size of the product, size of
the feed material also may have a bearing
on the degree of hammer wear. The fact
that Appleton refuse tended to be more
residential in nature than that at either
Ames or Cockeysville implies that the
characteristic size of the feed for the
Appleton experiments should be on the order
of 12.7 cm. On the other hand, the com-
mercial character of the waste encountered
at Ames and Cockeysville implies that the
feed size should be larger, on the order of
20.3 cm. These are average values for raw
residential and commercial waste which are
based upon previous experience and several
waste composition and sizing studies con-
ducted by Cal Recovery Systems.
In order to account for variations in
the size of the feed material and the
shredded product size among different
sites, a parameter termed the degree of
size reduction (Z0) is introduced. This
term is defined as
=-(F0 - X0)/F0
(9)
where FQ and X0 are characteristic feed
size and product size, respectively.
Values of the degree of size reduction
range from a value of zero corresponding to
no size reduction to a maximum limit of 1.0
corresponding to a product size of zero, or
in other words an infinite amount of size
reduction. The latter limit is, of course,
unachievable in actual practice.
Normalization of the wear data collec-
ted at Appleton's west mill and Cockeys-
ville's shredder #1 is demonstrated in
Figure 4, using average ZQ values and data
from the two curves drawn in Figure 3.
Although only two points are present for
each curve, the general trend of increased
wear at large values of Z0 can be discerned.
Also, the parametric effect of alloy hard-
ness (Re) is apparent, i.e. for a particu-
lar value of Z0, wear is greatest for the
softest hardfacing alloy.
The general conclusion that can be
drawn from the data in Figure 4 is that
hard alloys yield significant reductions in
hammer wear, for example, on the order of
60 percent if an alloy with a hardness of
56 Rc is used instead of an alloy with a
hardness of 28 Rc. A 60 percent reduction
in wear means that the hammers coated with
the harder alloy could be used about 1.8
times longer than those coated with the
softer alloy.
SUMMARY
This study of large-scale shredders has
served to establish levels of performance
that can be expected from horizontal ham-
mermills with respect to size reduction of
solid waste. Due to the intrinsic differ-
ence among shredders and the diverse con-
ditions under which they operate, predic-
tive relationships were developed to allow
interpretation of the experimental results
such that all shredders could be compared
on the basis of producing an equivalent
particle size. Earlier studies (1,2,3,4)
had established: 1) the important govern-
ing parameters that describe the process of
size reduction and 2) the methods of eval-
uating shredder performance. Results of
the present study serve to substantiate
the utility of the evaluative criteria and
the methods of analysis that were estab-
lished during earlier investigations.
The information developed in this
study provides a data base that up to this
point in time had been lacking for shred-
ders operating in the field. Special
emphasis was placed on investigating
methods of minimizing energy consumption
and hammer wear as well as determining the
magnitude of any possible improvements.
Concurrently, governing relationships among
the key parameters of size reduction were
developed to describe the shredders that
were tested.
The need for further research has also
been established as a consequence of this
96
-------
7 -
CM ,
'o 4
s-
IC
Ol
1C
3C
10
Wear Due
To Abrasion
Wear Due
To Chipping
bnder Impact
\
Optimum
Hardness
Range
_ • Appleton West
A Cockeysville #1
03
i.
O)
to
a:
20
Figure 3.
0.06 r-
0.04
0.02
30 40 50
Alloy Hardness (Re)
60
70
Hammer Wear as a Function of Alloy Hardness
0.2
0.4
0.6
0.8
1.0
Degree of Size Reduction (ZQ)
Figure 4. Normalized Hammer Wear
97
-------
program of shredder testing and evaluation.
There are two areas in particular that need
further attention. The first area involves
optimization of shredding lines, which
ultimately would serve to minimize energy
consumption. A test conducted at a site
employing two-stage size reduction would
serve to widen the data base for establish-
ing criteria for optimizing shredder lines.
If such a test could include several grate
bar changes in both the primary and secon-
dary shredders, the optimum grate bar con-
figuration for each shredder could be
established and the appropriate methods of
optimization could be identified.
The second area requiring attention
deals with the effect of machine charac-
teristics, such as size of grate openings,
internal geometry, and geometry and number
of hammers upon specific energy require-
ment, throughput, and particle size. This
second area is directly concerned with
shredder design and manipulation of parti-
cle size. Although some predictive rela-
tionships have been developed as a conse-
quence of this work, the mere fact that
some of the coefficients and exponents in
the equations show a wide range of values
portends that the universal governing rela-
tionship for size reduction has not been
established. Up to this point in time, the
effects of machine characteristics (grates,
internal geometry, number of hammers) have
not been studied in the field. Such an
investigation would serve to coalesce our
present knowledge of size reduction and
subsequently lead to the possible develop-
ment of a universal governing relationship.
ACKNOWLEDGEMENT
This research was supported by the
U.S. Environmental Protection Agency, Muni-
cipal Environmental Research Laboratory,
Contract Number 68-03-2589, Mr. Donald
Oberacker, project officer.
The test program was conducted under
subcontract to Midwest Research Institute,
Mr. David Bendersky, project manager.
REFERENCES
1. Trezek, G. J. and Savage, G. M., Significance of Size Reduction in Solid Waste
Management, EPA-600/2-77-131, July 1977.
2. Trezek, G. J. and Savage, G. M., Size Reduction in Solid Waste Processing, Progress
Report 1973-1976, Report to the Environmental Protection Agency prepared under
Grant No. EPA R 801218.
3. Savage, G. M., Trezek, G J., and Shiflett, G. R., Size Reduction in Solid Waste
Processing - Fine Grinding, Progress Report 1976-1978, Report to the Environmental
Protection Agency prepared under Grant No. EPA R 803034.
4. Trezek, G. J. and Savage, G. M., Results of a Comprehensive Refuse Comminution Study,
Waste Age. July, 1975.
5. Shiflett, G. R., A Model for the Swing Hammermill Size Reduction of Residential
Refuse, D. Eng. dissertation, University of California at Berkeley, 1978.
6. Shiflett, G. R. and Trezek, G. J., Parameters Governing Refuse Comminution,
Resource Recovery and Conservation, In press, 1979.
98
-------
DEVELOPMENT OF CONTINUOUS ACID HYDROLYSIS PROCESS FOR
THE UTILIZATION OF WASTE CELLULOSE
Walter Brenner, Barry Rugg, Robert Stanton
Peter Armstrong and Kuan-Ming Ang
Department of Applied Science
New York University
New York, New York 10003
Charles Rogers
Solids arid Hazardous Waste Research Division
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
ABSTRACT
Cellulosic wastes are being re-evaluated in terms of an underutilized renewable re-
source for both energy production and materials recovery, rather than a solid waste dis-
posal problem. Acid hydrolysis is a potentially attractive route for upgrading the value
of cellulosic wastes by converting them to glucose. The glucose can then serve as an alter-
nate feedstock to petrochemicals for fuels, intermediates and the synthesis of single cell
protein. New York University has carried out considerable work on this developing techno-
logy with emphasis on two crucial steps, namely, a cost effective cellulose waste pretreat-
ment and a high sugar yield acid hydrolysis. Extensive experimental investigations for
achieving these objectives are described. These studies led to the development of a con-
tinuous acid hydrolysis process in order to optimize the acid hydrolysis of was.te cellulose
to glucose. The development of a continuous acid hydrolysis process based on the applica-
tion of a twin screw reactor is discussed and preliminary information is presented on the
establishment and operation of a continuous acid hydrolysis process facility.
INTRODUCTION
Interest in the possible industrial use
of waste biomass for both materials and en-
ergy recovery has been greatly stimulated by
the convergence of two apparently unrelated
problems, i.e., the steadily rising price
of fossil fuels and the even greater magni-
tude of the solid waste disposal problems.
Currently employed methods for dealing with
these solid wastes are admittedly increas-
ingly inadequate for cost effective utiliza-
tion of their latent energy values, material
and especially for cellulose which comprises
a major portion of solid wastes. Haste cellu-
lose conversion to glucose via acid
hydrolysis offers attractive possibili-
ties for.more effective usage of a larae and
continuously-growing yet grossly
underutilized national resource. As is
shown in Figures 1 and 2 the glucose from
cellulosic wastes could be employed for the
manufacture of many volume chemicals which
are presently obtained from petrochemical
feedstocks.
The potential quantities of chemical
feedstocks, etc., which could be obtained
from waste cellulose conversions are quite
significant. Assuming a realistic 32%
available cellulose in the wastes and only
16% conversion to ethylene, the estimated
1550 million tons of farm waste alone would
yield 28 million tons ethylene per annum1'.
This represents twice the U.S. consumption
of this basic building block in 19753).
The economic feasibility of waste cellulose
utilization is of course, greatly dependent
on both waste collection and processing con-
version costs as well as the quantities and
values of the end products obtained.
99
-------
HISTORICAL REVIEW
Acid hydrolysis of cellulose has been
extensively studied for the better part of
a century, particularly in connection with
manufacturing ethanol from wood wastes.
The discovery that cellulose can be hydro-
lyzed in acid solutions and converted to
its monomer, glucose, was first reported
by Bracconnot 160 years ago, in 1819. The
reaction has been experimentally investi-
gated ever since then, mostly on a strictly
empirical basis, in order to develop a cost
effective process for producing sugar from
wood wastes and other sources of waste
cellulose.
Research and development on the acid
hydrolysis of wood wastes was particularly
active during the first .part'of this
century both in the U.S. and Western Europe.
These acid hydrolysis studies went far be-
yond the laboratory scale up to the con-
struction and operation of limited pro-
duction facilities for glucose ancLethanol
in the U.S. and in Western Europe**). The
U.S.S.R. and Japan have also been very
active in the field of wood saccharification.
The growing interest in the utiliza-
tion of waste cellulose for energy produc-
tion and materials recovery has led to the
recognition that the present state of know-
ledge in this area is largely empirical
and that more basic information could be
helpful in the development of a more cost
effective hydrolysis process. Both econo-
mic and technical reasons favor a dilute
acid hydrolysis of waste cellulose to be
conducted at high temperatures and for short
time periods so as to maximize the glucose
yield. Determination of the reaction
kinetics of the waste cellulose conversion
to glucose is necessary for establishing
the optimal reaction conditions and has
therefore been studied in some detai14)5)6)
While the acid hydrolysis of cellu-
lose is heterogeneous, it can be regarded
as a homogeneous reaction provided that the
cellulose reactant is dispersed in the form
of fine particles, e.g. 200 mesh or less.
A kinetic scheme for the acid hydrolysis
of crystalline wood was proposed by Seaman
as long as 30 years ago. This scheme was
used to effectively model the hydrolysis
of cellulose in wood chips and likewise
cotton 1 inters.
Porteous and later Fagen et al.
applied Seaman's model to describe the
formation of glucose from commercially
available waste paper. '5/°) Porteous pre-
dicted a maximum sugar yield of 55% with
o.4% acid at 230°C. The Fagan experi-
ments were carried out with rather small
samples (0.5 gms) of ball milled Kraft
paper. Such kinetic studies are now
deemed of considerable value for the de-
velopment of improved process designs and
economic data of waste cellulose to glu-
cose and/or ethyl alcohol production
facilities.4'5'
The kinetically predicted maximum
sugar yields assume that the cellulose
reactant has appropriate chemical reacti-
vity for the acid hydrolysis. The tech-
nical problems of cellulose hydrolysis
are to a great extent due to the fact that
this is not the case. The lack of an ade-
quate alnount of chemical reactivity in
cellulose is called lack of accessibility.
This is related to the highly inert char-
acter and crystalline organization on a
molecular level of the highly molecular
weight cellulose and also the presence of
lignin. Hydrogen bonding almost certainly
plays a very important role in the struc-
ture of cellulose and may be a key factor
in explaining its chemical inertness.
Over the years, a very extensive
amount of R & D has been concerned with
the development of the both technically
effective and economically viable cell-
ulose pretreatments. The basic approach
has been to reduce the crystal!inity and
disrupt the hydrogen bonding thus rendering
it more accessible to hydrolytic depoly-
merization reactions.7) This should make
it possible to approach the predicted
glucose yields more closely. Many differ-
ent pretreatments have been proposed and
a considerable number of them have been
experimentally investigated.
In general,"mechanical treatments
such as, intens-ive ball-mi 11 ing to
sizes below 60 mesh, have been found to
be technically possible, but at an esti-
mated cost of 6£/lb which is economically
prohibitive. Employment of high energy
ionizing radiation has been shown to be
at least equally effective when the
cellulose is exposed to dosages in the
order of 100 megarads.18' Sugar yields
as high as 70%
100
-------
based on the available cellulose have been
reported for thusly irradiated wood pulp
after dilute acid hydrolysis. The cost of
such large dosages of ionizing radiation is
however, too high for industrial usage.
Other less effective pretreatments which
have been experimentally studied, include
expending of the cellulose to various chem-
ical reagents, heat, etc.
Development of lower cost pretreatment
technology such as would produce mechanical
and/or chemical size and/or crystallinity
reductions, for maximizing cellulose acces-
sibility, is recognized as an essential re-
qureiment for an industrially acceptable
waste cellulose-glucose conversion process.
The pretreatment must furthermore be com-
bined with a high productivity acid hydro-
lysis process so as to optimize the conver-
sion of the pretreated waste cellulose to
glucose.
EXPERIMENTAL BATCH SCALE ACID HYDROLYSIS
STUDIES
Experimental investigations on the
dilute acid hydrolysis of waste cellulose
to glucose have been carried out at the
Department of Applied Science of New York
University over the past four years. The
waste cellulose feedstock employed in these
studies was primarily used newspapers. A
limited amount of work was also conducted
with wood waste. The experimental work was
an evaluation of the cost effectiveness of
various pretreatments for enhancing the
accessibility of the cellulose; determina-
tion of the optimum reaction conditions for
maximizing the sugar yields; techniques for
glucose recovery; also fermentation of glu-
cose to ethanol.
The hydrolysis experiments were ini-
tially carried out batchwise in a 1 liter
stirred autoclave equipped with appropriate
accessories such as electrical heating units,
a quick discharge ball valve for removal of
the reaction mixture after acid hydrolysis
from the autoclave, etc. The data obtained
with the batch 1 liter stirred autoclave
reactor experiments were then analyzed with
respect to the glucose yield at various re-
action conditions. This was followed by
scaling up to a 5 liter stirred autoclave
reactor again with suitable accessory equip-
ment. Quite extensive experimental work
was carried out with this larger reactor in
order to obtain a better understanding of
scale up and other problems for the dilute
acid hydrolysis reaction of the waste cel-
lulose. Stainless steel equipment was em-
ployed for all of these experiments.
Waste newspapers were first subjec-
ted to various preselected pretreatments
and then charged in the form of water slur-
ries with known cellulose concentrations
to a stirred autoclave reactor. These
cellulose slurries were then heated with
continuous stirring to predetermined reac-
tion temperatures and pressures. The re-
quired amount of acid was added only when
the desired reactive conditions had been
attained. Considerable emphasis was
placed on the rapid quenching of the reac-
tion products following completion of a
carefully timed hydrolysis in order to
optimize glucose recovery by minimizing the
formation of acid decomposition products of
glucose. Cellulose concentrations in the
slurries ranged from 5 to more than 20 wt%.
The batch scale hydrolysis experiments
with the two differently sized stirred
stainless steel autoclave reactors were
encouraging in that they showed that glucose
yields up to 50% or more of the available
cellulose values charged could be obtained.
It was furthermore found that economically
acceptable yet technically effective pre-
treatments are possible for obtaining such
glucose yields. The optimum reaction condi-
tions were found to be temperatures of
around 220°C-230°C and reaction times of
less than 30 seconds with about 1 wt% of
sulfuric acid. They agree rather well with
the results of the kinetic rate studies
which were previously reported by Porteous
and Pagan.°)9/
Various pretreatments for the waste
newspaper feedstock were experimentally in-
vestigated in order to enhance the cellu-
lose accessibility to the dilute acid.
These pretreatments consisted of Wiley mill
grinding, industrial dry grinding, hydro-
pulping, hydropulping with ferric ion/hy-
drogen peroxide presoak and hydropulping
followed by exposure to various dosages of
ionizing radiation. Figure 3 compares the
effectiveness of these pretreatments on
glucose yield for 450°F acid hydrolysis
reactions.
The technically most effective pre-
treatment is seen to be the irradiation of
the hydropulped used newspaper feedstock.
The irradiations were carried out at
101
-------
ambient temperatures and in the presence
of air with a 3 MeV Dynamitron electron
beam accelerator at Radiation Dynamics,
Inc., Plainview, L. I., N.Y. service
center. While irradiation dosages rang-
ing from 5 to 50 megarads were investigated,
the 10 megarad dosage was selected as most
cost effective. The combined costs of this
hydropulping/irradiation pretreatment have
been conservatively estimated at 0.3 -
1.1 c/lb waste cellulose feedstock. Figure
4 shows the significant improvement in the
glucose yield obtained by irradiating the
hydropulped waste cellulose.
The irradiation treatment itself was
accomplished rather simply. The hydro-
pulped waste newspapers were placed in the
form of slurries in polyethylene bags and
the bags were heat sealed. Each bag con-
tained about 20 Ibsof a hydropulped waste
newspaper slurry of known cellulose concen-
tration. The bags were then placed on a
conveyor which moved past the beam of the
Dynamitron electron beam accelerator. The
total dosage per pass was 5 megarads.
The effect of changing the hydrolysis
reaction temperatures and times on glucose
yield were experimentally investigated-in
considerable detail for used newspapers
which had been subjected to this combination
hydropulping/irradiation pretreatment. Typ-
ical results of a series of scale experi-
ments which were carried out in the larger
5 liter stirred autoclave reactor, are de-
picted in Fugure 5. The data show that
450°F temperatures and quite short reaction
times in the 20 second range were optimal
for this dilute acid hydrolysis.
Preliminary exploratory studies have
been carried out on recycling the unreacted
cellulose for obtaining additional glucose
product. The unreacted cellulose was re-
covered from the reaction slurry by filtra-
tion and then rehydrolyzed under analogous
reaction conditions. Table 1 shows typical
initial results. The additional amount of
glucose obtained decreases sharply in terms
of total glucose yield with the second
recycle. An alternative procedure would be
mixing of the unreacted cellulose with fresh
used newspaper feedstock in various propor-
tions. The cost effectiveness of such re-
cycling techniques is under study.
More recently the use of various or-
ganic acids in place of the sulfuric acids has
been investigated for this acid hydrolysis.
Preliminary data for the glucose yields ob-
tained with different organic acids are
shown in Figure 6. Maleic acid was found
to be surprisingly effective in enhancing
the glucose yield compared to the other
acids studied to date. Table 2 shows
that maleic acid catalyzed hydrolysis
works best with hydropulped and irradiated
newspapers. Use of this acid which is of
course more expensive than sulfuric acid,
can be expected to ameliorate the equipment
corrosion problems and possibly also the
rate of high temperature degradation of
the glucose product.
The conversion of the glucose product
to ethyl alcohol has also been studied on
a limited scale. Conventional fermentation
techniques were utilized for these experi-
ments. No particular difficulties were
encountered as might be expected in view
of the broad knowledge of glucose to etha-
nol fermentations. Ethanol product char-
acteristics have been determined in accord-
ance with established procedures.
CONTINUOUS ACID HYDROLYSIS STUDIES
Considerable effort is being devoted
to an investigation of possible continuous
processing technology being clearly
preferable to batch reactions for indus-
trial scale waste cellulose to glucose
conversionsl2-16)_ jn-js wor(c nas resulted
in the design and costing of a continuous
waste cellulose to glucose pilot plant
with a 1 ton/day capacity. This pilot
plant is based on the concept of employing
hydropulping and possibly, irradiation for
pretreating a waste cellulose feedstock; and
a screw conveyor and reactor device for con-
tinuously reacting pretreated cellulose in
an aqueous slurry at suitably elevated
temperatures.
The key to successful operation of a
continuous acid hydrolysis process is the
design of the hydrolysis reactor. This re-
actor has to be capable of feeding, convey-
ing and discharging hydrolyzable cellulosic
materials continuously while maintaining
appropriate temperatures in a reaction zone.
Because this hydrolysis requires exposure
of the reactor components to dilute acids
at high temperatures and pressures, all
materials of construction have to be resis-
tant to corrosion,especially in the reaction
zone.
102
-------
A Werner & Pfleiderer ZDSK53 (53mm)
twin screw extruder (Werner & Pfleiderer
Corp., Waldwick, N. J.) was selected
because of its capacity for conveying, mix-
ing and extruding the required amounts of
cellulosic feedstock. This machine allows
accurate control of temperature, pressure,
residence time, etc. within the previously
established acid hydrolysis operating condi-
tions. The key working elements, intermesh-
ing twin screws, eliminate materials build-
up in the processing section and make feasi-
ble close control of residence time, etc.,
with intensive mixing.
For continuous processing this reactor
must be coupled with an appropriate feeding
mechanism for cellulose slurries and a dis-
charge system for reacted material while
maintaining pressure and temperature in the
reaction zone. A steam jacketed crammer
feeder made by Werner & Pfleiderer Corp.,
was interpreted with the twin screw extruder
so as to maximize throughput with preheat-
ing as required. A Kamyr intensive service
2" ball valve (Kamyr Valve Co., Glens Falls,
N.Y.) was selected as the major component
for the design of the discharge system.
Additional ancillary equipment includes a
high pressure steam generator for supplying
energy to the reactor, an acid pump capable
of high pressure injection of acid and a
slurry pump for introducing feedstock into
the crammer feeder.
This equipment was obtained and in-
stalled at the Antonio Ferri Laboratories
of New York University (Westbury, L.I.N.Y.).
Figure 7 shows a floor plan of the facility
which includes space for pulp storage, rou-
tine analysis and an office. A photograph
of the twin screw reactor together with the
crammer feeder and the discharge ball valve
(without actuator) is shown in Figure 8.
Hydropulped recycled newspaper feed-
stock is obtained in slurry form (10$ solids
content) from the Garden State Paper Company
(Garfield, N.J.) Optionally it is irradiated
at the Radiation Dynamics, Inc., Plainview,
L.I., N.Y. service center with a dosage of
10 Megarads. This pulp feedstock is intro-
duced into the reactor by means of the above
mentioned slurry pump and crammer feeder.
The waste cellulose is then conveyed with
heating by the twin screws into the reaction
zone where the required amount of acid is
introduced. Hydrolysis then takes place at
a predetermined temperature and pressure and
the product is promptly discharged.
Two fundamental problems in this reac-
tion scheme have been identified, both re-
lating to maintenance of pressure in a flow
reactor. Firstly, pressure has to be main-
tained at the inlet to prevent egress of
material through the cramner feeder. The
solution has been found to be development of
a dynamic seal in the form of a densified
plug of material within the inlet zone of
the reactor. Secondly, continuous discharge
of the hydrolyzed material has to be accom-
plished while maintaining pressure. This
has been accomplished by the design of a
discharge system which includes a hydrau-
lically powered activator for the Kamyr
val ve.
Preliminary shakedown of this contin-
uous acid hydrolysis system has been com-
pleted. Experiments are underway to charac-
terize the system and to optimize reaction
conditions for maximum glucose yield. Work
has commenced on a study of the material
and energy balances of the system.
SUMMARY AND DISCUSSION
Continuous acid hydrolysis is a poten-
tially attractive technology for energy
production and materials recovery from cel-
lulosiic wastes via glucose. Considerable
progress has been made in the development of
such a process. A 1 ton/day pilot plant has
been constructed based on a twin screw hydro-
lysis reactor. Technical feasibility has
been established. Characterization and opti-
mization studies are in progress. Consider-
ation is being given to utilizing the tech-
nology for methane production as the final
end product.
ACKNOWLEDGEMENT
This study was supported by the U.S.
Environmental Protection Agency under Grant
No. R805239-010 under the direction of
Charles Rogers. The authors would like to
thank Mr. Barry Perlmutter and Mr. Dennis
Nadis for their contributions in the design
and assembly of the continuous reactor lab-
oratory.
REFERENCES
1. Baker, P. and Ruggeberg, W.H.C.,
"Glucose Isomerization - Possible Impact On
Chemical Utilization of Carbohydrates", pre-
sented at the Spring National American
Chemical Society .Meeting, New York, N.Y
April 1976.
103
-------
2. Bracconot, H., Am. Chen. Phys., 12 (2)
(1819) (Fr.).
3. Chemical Engineering and News, January
10, 1977 and October 12, 1977.
4. Converse, A.I., et a., "A Laboratory
Study and Economic analyses for the Acid
Hydrolysis of Cellulose in Refuse to Sugar
and Its Fermentation to Alcohol", Final
Report under PHS Grant No. UL-00597-02.
Thayer School of Engineering, Dartmouth Col-
lege, Hanover, N.H. (1971).
5. Fagan, R.D., Converse, 0. and Grethleih,
H.E., "The Economic Analysis of the Acid
Hydrolysis of Refuse", Thayer School of
Engineering, Dartmouth College, Hanover,
N.H. (1970).
6. Fagan, R.D., Grethelein, H.E., Converse,
A.O. and Porteous, A., Environmental Science
and Technology, 5, 545 (1971).
7 Millet, M.A., et al., "Physical and
Chemical Pretreatments for Enhancing Cellu-
lose Saccharifications", presented at the
Symposium on Enzymatic Conversions of Cellu-
losic Materials: Technology and Applica-
tions", sponsored by U.S. Army Natick Labor-
atories, September^, 1975, Newton, Mass.
8. Lloyd, R.A. and Harris, J.F., "Wood
Hydrolysis for Sugar Production", FPL Rpt.
No. 2029, March 1955, Forest Products Labor-
atory, U.S. Department of Agriculture, Madi-
son, Wisconsin
9. Porteous, A., "Towards a Profitable
Means of Municipal Refuse Disposal", Amer-
ican Society for Mechanical Engineering
Pub!., 67-WAPID-2, (1967).
10. Saeman, J., Industrial Engineering
Chemistry, 37, 43 (1945).
11. Ware, S.A., "Fuel and Energy Production
by Bioconversion of Waste Materials", EPA
Report (600/2-76-148, August 1976, Municipal
Environmental Research Laboratory, U.S.
Environmental Protection Agency, Cincinnati,
Ohio 45268.
12. Personal discussions, N. Kronlund,
Black Clawson Company, Fall 1976.
13. Personal discussions, S. Huss, Garden
State Paper Co., Saddle Brook, New Jersey,
Fall 1976.
14. Personal discussions, M. Cleland,
Radiation Dynamics, Inc., Melville,
Long Island, New York, November 1976.
15. Rogers, C., Brenner, W., and Rugg,
"Radiation Treatment of Solid Wastes", pre-
sented at the First Radiation Processing
Conference, June 1976 (Cerromar Beach,
Puerto Rico; to be published).
16. Rogers, C., Unpublished Report, Nov-
ember 1976, EPA Solids and Hazardous Wastes
Laboratory, Cincinnati, Ohio.
17. Winters, 0., and Eng., M.T., Hydrocar-
bon Processing, p. 125, et al., November
1976.
18. Saeman, J. and Millett, M., "Effect of
High-Energy Cathode Rays on Cellulose",
Industrial Engineering Chemistry, 44, 2848
(1952).
19 Cleland, M.R., Thompson, C.C., and
Malone, H.F., "The Prospects for Very High-
Power Electron Accelerators for Processing
Bulk Materials", Transactions of the First
International Meeting of Radiation Proces-
sing, Radiation Physics and Chemistry,
Vol. 9, Nos. 4-6 pages 547-566.
20. Grethlein, H.E., "Comparison of the
Economics of Acid and Enzymatic Hydrolysis
of Newsprint", Biotechnology and Bioengi-
neering, Vol. XX, pp. 503-525 (1978).
104
-------
TABLE 1
EFFECT OF RECYCLING ON GLUCOSE YIELD FOR DILUTE
SULFURIC ACID HYDROLYSIS AT 450°F
RUN PERCENT GLUCOSE YIELD
450°F Acid Hyrdolysis 51
With one recycle 59
With two recycles 63
TABLE 2
MALEIC ACID CATALYZED HYDROLYSIS OF WASTE CELLULOSE
(FEEDSTOCK: GARDEN STATE PAPER CO. USED NEWSPAPERS)
PPL-TBPATMFNT PERCENT SOLIDS VOLUME ACID* TIME TEMPER- PERCENT GLU-
FEEDSTOC.K PRL-TREATMENT IN DRAG CC SEC ATURE°F COSE YIELD
Hydropulping 5.1 80 60 450 32.5
Hydropulping + irradiation 5.2 80 60 450 60.9
Hydropulping 5.0 160 40 450 30
Hydropulping •+ irradiation 4.8 160 40 450 58.6
Hydropulping 11.0 150 40 420 16.5
Hydropuloing + irradiation 10.1 150 40 420 24.5
Hydropulping 5.0 80 60 450 32.5
Hydropulping + irradiation 4.7 80 60 450 58.7
*
40% concentration
105
-------
ANIMAL FOOD =
CELLULOSE
ACID OR
HYDRO
V
ENZYME
LYSIS
/
GLUCOSE
= HUMAN FOOD
MICROBIAL
CONVERSION
N/
SINGLE CELL PROTEIN
CHEMICAL
RAW MATERIALS
FUEL-ETHANOL
SOLVENTS-ACETONE
CHEMICALS
ANTIBIOTICS
ENZYMES
FIG.1: WASTE CELLULOSE UTILIZATION ROUTES
-------
•* ethylene
butadiene
'ethanol L>acetaldehyde
CELLULOSE |
WASTE+glucose
n-butanol
butyraldehyd<
stic acid —
ace
I
fructose —» mannitol
—»sorbitol —i
acetic anhydride
vinyl acetate
acrytonitrile
glycerol
o
—4
pentoses^* hydroxymethylfurfural
lignins —
*diols
->syn. rubbers
2-ethylhexanol
acetates
acetates
p.v. acetate
p.acrylonitrile
p.urethanes
surfactants
drugs
p.ester,p.u.
p.ester, p. a.
^terephthalic acid-> p.ester
FIG.2: POSSIBLE ROUTES TO PETROCHEMICALS FROM CELLULOSIC
WASTES
-------
o
00
Q hydropulped/ir radiated (10 mr)
A hYdropuljbed/Fe++-H2O2 treated acid;1.3%
o wiley mill ground,acid: 2.25%
x hydropulped,acids 2.25% rt0,
a industrially ground acid=1.3/o
4O
SO
6O
70
80
9O
FIG.3: %GLUCOSE YIELD VS REACTION TIME FOR ACID HYDROLYSIS
OF GARDENSTATE PAPER AT 450° FOR VARIOUS PRETREATMENTS
-------
o
10
o -•
o
VO
III
8
§0
-i n
O
Q
_l
III
O
N
O -•
-x *^HYDROPULPED/ IRRADIATED (10mr)
ACID =0.87%
HYDROPULPED
AGIO = 2.25%
1O 2O 3O 4O SO 6O
FIG.4;%GLUCOSE YIELD VS REATION TIME FOR ACID HYDROLYSIS
^^ « ^^_ ^^^ ^^_ ^|^ ^^_ ^ ^^_ ^^_ IPHk A gp& gp^, ^_^ JK ^^H ^ ^^ ^^Vir p^g
OF GARDENSTATE PAPER AT 450 F
-------
a 425 °F
X475 °F
O 45O °F
1O 2O 3d
REACTION TIME-. SECONDS
SO
FK3.5: ACID HYDROLYSIS OF HYDROPULPED AND IRRADIATED (10 MR)
USED NEWSPAPERS EFFECT OF REACTION TEMPERATURE ON
GLUCOSE YIELD
-------
50.
4O.
30.
0
mm
U
>
U
20.
0
o
o
#10.
Q
•
i
»
>
••••Ml
CO
1
Q
8
u
UJ
1
••••
^.o
c^^
CM
CM
1
Q
0
U
UJ
_j
5
•••••
C5^^
6
1
o
8
o
UJ
5
N.
T
Q
0
Z
UJ
^
-------
-«—i
f
2
Vcrammer
feeder
hydraulically-actuated
ball valve
psi ,
'boiler i
safety '
shower
storage for
paper slurry
laly
control unit 1 J
drying oven
1
«
rtical lab g
T
^
co
•a
oaiances I .h
workbench 1
L
drafting]
board |
1
shelves
file
office
j i
desk
H
5 FEET
FIG.7: NYU/EPA ACID HYDROLYSIS WASTE CELLULOSE TO GLUCOSE
CONVERSION CONTINUOUS PILOT PLANT
-------
6.
7.
8.
9.
.0.
.1.
Steam Jacketed
Crarrmer Feeder
Open Steam Injection
Acid Injection
Barrel Sections
with Heaters
Karnyr Discharge valve
(without activator)
Vent Hood
Melt Thermocouples
Melt Pressure Transducer
Control Panel
Temperature and Pressure
Monitoring
Direction of Flow
Fig. 8. Werner & PfTenderer Twin Screw Extruder Reactor
-------
PRODUCTION OF METHANE FROM ACID HYDROLYSATES OF CELLULOSE WASTES
V. R. Srinivasan
Department of Microbiology, Louisiana State University, Baton Rouge, LA 70803
ABSTRACT
Anaerobic digestion of acid hydrolysates of cellulose presents certain advantages over
direct- conversion of cellulose to methane. Lack of variation of the composition of the
feedstock and the absence of suspended particles make it easier to optimize the environ-
mental conditions for continuous production of methane at shorter residence times. Short
residence times may lead to design of digesters of small size and increased productivity.
Experiments are in progress for continuous generation of methane from glucose with mixed
population of micro-organisms enriched from rumen.
INTRODUCTION
Solid wastes are generated in increasing
amounts in todays affluent and consumer-
oriented society. Most estimates place
the rate of increase in municipal solid
waste generation at about 4% per year
while the rate of population growth per
year is only about 1% (1). The need for
conservation of our natural resources and
the problems of abatement of environmental
pollution have been primarily responsible
for the increased number of investigations
on the recovery of resources and conver-
sion of solid wastes into usable energy or
other products. The composition of
municipal solid wastes varies to a
limited extent, however, nearly 50% of the
waste is usually cellulosic paper products.
The heating value of the waste is generally
between 3000-6000 BTU per pound. Some of
the new technological developments in solid
waste management include: (a) the use as a
supplemental fuel in the coal-fired boilers
(b) pyrolysis-destructive distillation at
high temperatures (1000-2000 C) in an
oxygen-deficient atmosphere to generate
valuable gaseous mixtures and liquids con-
taining organic chemicals, (c) process for
conversion of garbage and combustible re-
fuse into synthetic petroleum yielding
more than one barrel of crude oil per,ton
of refuse, (d) microbial degradation of
cellulosic portion of the refuse for ob-
taining useful products.
MICROBIAL DEGRADATION OF CELLULOSE
Biodegradation of cellulose may be car-
ried out by several methods using different
genuses or species of microorganisms.
Fungi, yeasts as well as bacteria can be
used either singly or as a stable-mixed
culture of organisms. Cellulose may be
fermented aerobically or anaerobically
dependent upon the type of desired products
to be generated. The aerobic process is more
efficient for the growth of organisms while
anaerobic fermentation produces valuable
organic chemicals. Approximately 1.2 moles
of ATP are required to fix 1 mole of carbon
in the cell aerobically while 7-10 moles of
ATP are necessary for fixing 1 mole of car-
bon in the cell by anaerobic processes.
Looking at it from the point of view of
carbohydrate utilization, less than 20%
of the carbohydrate present in the nutri-
ents, is utilized for the generation of the
114
-------
cell mass during anaerobic fermentation (2)_.
Some of the different products which may be
obtained by microbial, fermentation of
cellulose may be summarized schematically
as in Fig. 1.
will mainly discuss the background on
methanogenesis, advantages of our
strategies in the production of methane
and finally our accomplishments during
this period.
Aerobic
Cellulose
t
Hydrolysis
CAcidic or Enzymatic)
f
Glucose
Fermentation
— Anaerobic
Fermentation
Single Cell Protein
Microbial Enzymes
Food
Our laboratory has been interested in
the microbial fermentation of cellulose for
the past several years. We have.success-
fully completed laboratory investigations
on the optimization of conditions for con-
tinuous production of single cell protein
from cellulose. Quite recently under a
contract from USEPA R805239-02 we have
started our investigations on the produc-
tion of methane from acid hydrolysates of
cellulose. The objectives of our studies
may be outlined as follows:
1. To obtain a stable mixed population
of a methanogenic organism with
other compatible anaerobic organisms.
2. Studies on methanogenesis by the
mixed culture from glucose under
different environmental conditions
such as pH and temperature.
3. Experiments on the optimization of
the nutrients in the medium (such as
N, P, etc.) and dilution rate for
continuous production of methane and
maximizing productivity based on
glucose.
4. Studies on the production of methane
with different concentrations of re-
duced sugars obtained by acid
hydrolysis of paper wastes, by the
method developed by Drs. Rugg and
Brenner at New York University, N.Y.
Since these studies have been started
only a couple of months ago, this report
Organic acids, Acetone
Butanol, Ethanol,
Methane
Fuel, Chemicals
BACKGROUND ON METHANOGENESIS
Production of methane by anaerobiosis is
essentially a two step process. First,
biodegradable organic materials such as
carbohydrates and proteins are broken down
to organic acids, CO- and H». Then in the
second step methanogenic organisms convert
the end-products of the first step into CH.
utilizing the energy produced during con-
version for their growth. Weimar and
Ziekus have reported that Methanosarcina
barkerri grown in the presence of H2/C02
alone or mixotrophically with methanol con-
verted 81% and 79% respectively of sub-
strate carbon into methane (3)- Methane-
producing bacteria are not capable of
generating methane directly from complex
carbohydrates or lipids. This in any type
of Biogas digester for the production of
methane, one has to deal with diverse
groups of organisms interacting with one
another. The terminal reaction in the
synthesis of methane during anaerogiosis
may be expressed as
4H2A + C02 = 4A + CH4
where H,A may represent the range of
methanogenic substrates such as formic
acid, acetic acid or methanol or even
hydrogen,(4). The response of the
methanogenic organisms to
115
-------
any change in the environment in the
reactor is more sluggish than other inter-
acting organisms. Hence, it becomes
necessary to maintain the optimum environ-
ments in the reactor for the growth of
methane bacteria. Considerable amount of
information has been published recently
about the growth and physiology of differ-
ent species of methane bacteria. Extensive
investigations are being carried out on
methanogenic bacteria isolated from diverse
environments, such as coastal sediments,
lake muds, decayed tree barks, anaerobic
digestors and cattle rumen. Some general-
izations as to the characteristics of
methanogenic organisms can be made from
these studies. Methane bacteria are obligate
anaerobes. They are the most sensitive
to oxygen of all the organisms known at
present. The optimum temperature of growth
for these organisms depends upon habitat of
the organism from which it was originally
isolated. There are thermophilic and
mesophilic methanogens. The organisms in
general are able to grow and produce
methane only in the range of "p***s 6 to "7.0.
Thus it is imperative that either a good
buffering system has to be present in
the reactor or p must be control led
automatically for the reactor to be oper-
ative. It is also advantageous to adjust
the carbohydrate feed in such a manner
that acids are not formed at a rate
greater than the rate of which they are
utilized by the methanogens. This is one
of the many factors' involved in sub-optimal
operation of a biogas generator. Relative-
ly slow rates of conversion of organic
substances to methane, necessitates the
constuction of large production of
methane. Morris, et al. have summarized
some of the pertinent factors which in-
fluence the anaeoribic fermentation in
biogas generators (Table 1).
USE OF ACID HYDROLYSATES OF CELLUSOSE AS
SUBSTRATES FOR METHANE PRODUCTION
It is common observation in the anaero-
bic digesters in biological waste treat-
ment the normal operating process at times,
suddenly goes "sour" as a response to
environmental change. This is generally
due to increased content of carbohydrates
in the feed. Once such phenomenon occurs,
it is several weeks before the start-up
and normal functioning of the digester
is related. The foremost advantage of
using reducing sugars obtained by acid
hydrolysis of cellulose as substrates for
the biogas generator is maintaining a
reasonable constant feed composition. This
is an important factor if the digester
is operated for extended periods of time
without frequent shut-downs. If cellulose
to glucose by cellulose is used directly
in the digester, then the solubilazation
of cellulose to glucose by celluloytic
organisms becomes the limiting step
which requires long residence times for
digestion. Moreover the particulate
materials settle down to the bottom of
the reactor, and the microorganisms have
atendency to adhere to the suspended
particles. This results in inefficient
transfer of nutrients to the organisms and
which can be avoided by soluable car-
bohydrates. Such difficulties are not
insurmountable as shown by the design
and operation of anaerobic digesters
using agricultural farm residues or
animal feed lot waste. However, from the
author's experience with cellulose fer-
mentation, it is fair to state that
optimazation of a fermentation process
is much less complicated with soluble
substrates than with solids. Optimi-
zation of kinetics of methane produc-
tion is a key step in deciding
Table 1 (5)
Environmental and Operational Factors Effecting Anaerobic Fermentation
Environmental Factors
PH
Alkalinity
Volatile acid concentration
Temperature
Nutrient Availability
Toxic Materials
Operational Factors
Composition of Organic Substrates
Retention Time
Concentration of substrate
Organic loading rate
Degree of mixing
Heating and heat balance
116
-------
whether the process will be economically
viable.
EXPERIMENTAL
Several pure strains of methanogenic
organisms were obtained from Dr. Wolln at
New York Public Health Laboratories,
Albany, NY. The organisms are maintained
and routinely grown in the medium of the
following composition: 100 ml of the
medium contain 0.5 g glucose:0.5 g trypti-
case:0.2 g yeast extract:4 ml of 0.6%
K2HPOt:4 ml of a solution containing
KH2P04 (0.6%) (NH,) SO (0.6%) NaCl (1.2%)
NgS04«7H20 (0.245%) ana CaCl2'2H20 CO.159%).
The medium also contained 5 ml of 8% Na2CO
and 0.05 g Na2S-9H20. 3
•In order to obtain stable population of
mixed organism with methanogenetic ability,
enrichment of organisms from the rumen of
a f istulated cow was executed-.
Rumen fluid was filtered through cheese
cloth and initial enrichment was accomplished
using Difco "Thioglycollate medium".
Organisms were transferred serially after
3 days of growth under oxygen-free nitrogen
containing 5% COj. Methane was determined
by gas chromatography. After nearly 20
transfers, the mixed population still
retained the ability to produce methane.
With this inoculum, methanogenesis was
studied in different media. The organisms
are also grown in media without glucose.
In the absence of glucose methane pro-
duction was higher. This was found to be
due to drastic changes in the pH. In 24
hours pH changed from 7.2 to 4.5. Hence,
the organisms are maintained at present
in a medium containing proteose peptone
as the carbon source and yeast extract
for the growth factors. A continu-
ous culture of the mixed population was
set up in -order to circumvent the inhibi-
tion of methane production by sudden
changes in the pH. Also, the technique
will be used to select a population
which will generate methane from glucose
alone. The same apparatus will be used
to optimize the environmental factors for
methanogenesis from glucose. The apparatus
consists of a 400 ml wide-mouthed bottle
as the culture vessel placed in a water-
bath whose temperature was kept at 38°C.
The inflow of the nutrient medium was
controlled by a peristaltic pump
(Harvard Apparatus Co.) The gas outlet
also served as the exit tube for the excess
liquid in the reactor, thus keeping the
volume constant. The outlet was connected
to a siphon with a head space which split
the gas and liquid streams. The whole
apparatus inclusive of the nutrient
vessel was kept under oxygen-free nitrogen.
The different connections were made with
heavy-walled butyl rubber tubings in order
to ensure no diffusion of atmospheric
oxygen. The nutrient medium per liter
consists of the following composition:
Glucose 10 g., yeast extract 0.2 g:
(NH,) SO, 0.3 g:K2HP04 0.07g; NaH,P04-H70
0.02 g, MgCl2-6H20 0.06 g and trace
elements (Ca+f, Fe4+, Mn++, and Zn++). The
medium also contained 0.5 g of Na2S'9H20
and resazurin 0.002 g.
At present we are still in the process
of enriching and stabilizing the popula-
tion of organisms in the above medium and
we propose to collect quantitative infor-
mation in the next several weeks.
REFERENCES
1. Rimberg, D. (1975) Municipal Solid
Waste Management. Publ. Noyes Data
Corporation, pp. 3-18.
2. Hobson, P. N. (1971) Rumen Micro-
organisms. Progress in Industrial
Microbiology. 9_:41-77.
3. Weimar, P. J. and J. G. Zeikus. (1978)
One carbon metabolism in methanogenic
bacteria. Arch. Microbiol. 119:49-57.
4. Kirsch, E. J. and R. M. Sykes. (1971)
Anaerobic Digestion in biological
waste treatment. Progress in Indust.
Microbiology. 9^:155-237.
5. Morris, G. R., W. J. Jewell and R. C.
Loehr. (1977) Anaerobic fermentation
of animal wastes: Design and opera-
tional criteria in Food, Fertilizer
and Agricultural Residues. Ed. R. C.
Loehy. Publ. Ann Arbor Sci. Michi.
pp. 395-413.'
117
-------
STABILIZATION AND CHARACTERIZATION OF PYROLYTIC OILS
Malcolm B. Polk and Metha Phingbodhipakkiya
Department of Chemistry
Atlanta University
Atlanta, Georgia 30314
ABSTRACT
Capillary gas chromatographic, liquid chromatographic and gas chromatographic - mass
spectrometric analysis procedures have been developed for the pyrolytic oils. The major
components identified were: ethanol; 1-butanol; isobutanol; acetaldehyde; furfural;
furfuryl alcohol; 5-methylfurfural; 2-octanol; phenol; Guaiacol; m- and p-methoxyphenol;
o-, m-, and p-cresol§; Veratrole; isobutyric acid; butyric acid; naphthalene; Estragole;
Anethole;"Eugenol; 2-methoxy-4-ethylphenol; 2-methoxy-4-propylphenol; Isoeugenol;
dimethylphenols; hydroquinone; 2-hydroxy-5-methoxybenzaldehyde; and 2-hydroxyaceto-
phenone. Included among the chemical processes occurring over a.period of time in the
pyrolytic oils are the formation of acetals through the reaction of aldehydes and
alcohols. Under the low pH conditions of the pyrolytic oils, Isoeugenol polymerizes
readily. Other components which might polymerize under the conditions include Eugenol,
Anethole, and Estragole. A fresh sample of pyrolytic oil was hydrogenated in an attempt
to stabilize the oil against polymerization. Hydrogenation of the oil resulted
principally in the conversion of Isoeugenol and Eugenol to 2-methoxy-4-propylphenol.
A phenol fraction was extracted from the pyrolytic oil samples and reacted with
formaldehyde to obtain a phenol-formaldehyde resin.
INTRODUCTION
Every year large amounts of agricul-
tural waste are produced in the United
States. The disposal of this waste mate-
rial is becoming an increasingly more
difficult and expensive matter. Also
there is a critical need for new energy
sources in the United States. Agricul-
tural wastes are largely lignocellulosic
in chemical nature. Pyrolysis is an
excellent approach for the conversion of
agricultural and wood manufacturing wastes
to useful energy or chemical forms.
Pyrolysis of solid agricultural
waste yields a noncondensable gas, a
liquid, and a solid carbonaceous char.
Analysis of the evolved pyrolytic gas
stream indicates the presence of hydrogen,
carbon dioxide, carbon monoxide, methane,
ethane, and ethylene. The individual gas
compositions vary with pyrolysis condi-
tions. The solid material which remains
after pyrolysis is an impure carbon and
ash. The liquid fraction or pyrolytic oil
contains organics and water.
Workers at the Engineering Experiment
Station at the Georgia Institute of
Technology have shown that pyrolysis is
readily utilized for the conversion of
cellulosic and lignocellulosic wastes into
useful fuels and other products. A large
prototype pilot pyrolysis system con-
structed in 1972^ was built to operate
continuously at an input feed rate of 300
to 500 Ibs. of waste per hour. The system
includes a waste receiving bin, a belt
conveyer to the converter, the converter
and char handling system, an off-gas
cyclone, a condenser and a gas burner.
Peanut shells, wood chips, pine bark and
sawdust, municipal wastes, macadamia nut
hulls and cotton gin wastes have been
pyrolyzed in the pilot plant. A large 50
ton/day demonstration plant which is owned
and operated by the Tech-Air Corporation
is located in a wood yard in Cordele,
Georgia and is operated on wastes from the
sawmill.
The pyrolytic oils are mixtures of
neutral compounds and strong and weak acids
with pH's of 3-4. The oils are viscous,
118
-------
sticky liquids at room temperature. The
pyrolytic oils appear to oxidize and/or
polymerize on standing and in the process
undergo an increase in viscosity.
Many investigators have determined
the products of the pyrolysis and distil-
lation of wood. Some of the products
identified are: aphthalene; dimethyl-
naphthalene; 2,3-dimethylanthracene;
2-methoxy-4-methylphenol; Guaiacol;
2-methoxy-4-ethylphenol; 2-methoxy-4-
propylphenol; phenol; cresols; xylenols;
pyrocatechol; 4-methylpyrocatechol;
dimethylpyrogallol; 4-hydroxy-3-methoxy-
toluenef 4-hydroxy-3-methoxystyrene
Eugenol; cis-Isoeugenol; trans-Isoeugenol;
Vanillin; acetovanillone; pyrogallol 1,3-
dimethylether; 4-hydroxy-3,5-dimethoxy-
toluene, 4-hydroxystyrene; 4-hydroxy-3-
methoxystyrene; 4-hydroxy-3-methoxy-l-
vinylbenzene; 4-hydroxy-3,5-dime thoxy-1-
ethylbenzene; benzene; sinasaldehyde;
coniferaldehyde; syringaldehyde; allyl-
syringol; methylnaphthalene; p-ethylphenol;
2,4,6-trimethyIphenol; <*-naphthol;
p-hydroxybenzaldehyde; p-hydroxyace tophe-
none; acetosyringone; butyrolactone;
crotonic acid; maltol; 2-hydroxy-3-methyl-
A2-cyclopentenone; 2,6-dimethoxyphenol;
tiglaldehyde; methyl isopropyl ketone;
methyl ethyl ketone; methyl furyl ketone;
acetaldehyde; furan; 2,3-butandione;
methylfuran; l-hydroxy-2-propanone;
glyoxal; acetic acid; furfural; acetone;
propionaldehyde; methanol; 3-hydroxy-2-
butancine; ethanol; diacetyl; furfuryl
alcohol; formaldehyde; 1-butanol; isobutyl
alcohol; anthracene; and phenanthrene.-*-14
Our objectives were to develop
methods for determining the chemical
composition of the pyrolytic oils imme-
diately after collection and then to
observe any changes in the chemical
composition of the oils as a function of
increasing viscosity. Also, after the
mechanism of the increase in viscosity
of the pyrolytic oils had been determined,
attempts were to be made to stabilize the
pyrolytic oils against whatever changes
they undergo. We also proposed to
evaluate pyrolytic oils as a feedstock for
the recovery and production of chemicals
and to routinely analyze pyrolytic oil
samples submitted by the pyrolytic oil
research group at the Engineering
Experiment Station at the Georgia Insti-
tute of Technology.
ANALYTICAL STUDIES
GC/MS of Condenser Oil Distillate
Qualitative analysis of the condenser
oil distillate by combined gas chromato-
graphy-mass spectrometry was performed on
a DuPont Model 21-490 GC-Mass Spectrometer.
The major components identified were:
ethanol; 1-butanol; isobutanol;
acetaldehyde; furfural; furfuryl alcohol,
5-methylfurfural; 2-octanol; phenol;
Guaiacol; m- and p-methoxyphenol; o-, m-
and p-cresols; Veratrole; isobutyric acid;
butyric acid; naphthalene; Estragole;
Anethole; Eugenol; 2-methoxy-4- propyIphe-
nol; Isoeugenol; dimethylphenols;
hydroquinone; 2-hydroxy-5-methoxybenzalde-
hyde; and 2-hydroxyacetophenone.
Composition of Pistillate as a Function
of Time
Over a period of six months, the al-
cohol, aldehyde, and acid peaks decreased
substantially in area. The decreases in
area are due to the formation of esters
and acetals under acidic conditions over
a period of time. These conclusions are
in agreement with the observation of Knight
that there was an increase in water con-
tent for a pyrolytic oil sample stored
over a period of eight months.-'--'
The peak which contained Estragole
(4-allylanisole) also decreased substan-
tially in area over that same period. That
bit of experimental data caused us to
investigate the facility of cationic
chain-reaction polymerization of other
olefins in the oils which might potentially
undergo polymerization. Anethole, Estra-
gole, Eugenol, and Isoeugenol were
separately added to 3MHC1 solutions.
Isoeugenol formed a solid precipitate
overnight. Eugenol formed a brown,
viscous gel. Anethole and Estragole
formed brown oils.
Capillary Gas Chromatography
The condenser oil filtrate (acetone)
and condenser oil distillate were
analyzed on a Varian 3740 capillary gas
chromatograph with a flame ionization
detector. The chromatograms are shown
in Figures LA and IB. The conditions
were: a 30 meter Supelco 2-3700 glass
capillary grade AA SP 2100 column (internal
diameter ca. 0.25 mm), and an oven temperature
119
-------
program of 40°C-22CrC at 5"C per tain. ThS
He flow rate was 1.0 ml/min. The chromato-
grams were obtained in the direct injection
mode. Peak identifications were made by
noting the enhancement of peak size when
known substances were added to the samples.
Peak assignments and quantitative data are
presented in Tables 1 and 2.
EXTRACTION OF PHENOLS FROM
PYROLYTIC OILS
Phenols may be removed from the pyro-
lytic oil by (1) extracting with water and
discarding the aqueous layer, (2) extract-
ing with 1.5N NaHC03 to remove the
carboxylic acids and (3) extracting with
2.5N NaOH followed by neutralization with
5% HC1 solution. The capillary gas
chromatogram of the phenol extract is shown
in Figure 2. The peaks at retention times
of 16.4 (7.0gc %), 17.9 (5.7gc %), 18.0
(3.3gc %), 19.0 (lO.Ogc %), 21.0
(14.0gc 7,), 24.7 (4.0gc %), and 28.4
(3.9gc %) minutes are phenol, Guaiacol,
o-cresol, m-and p-cresol, 2-methoxy-4-
methylphenol, Eugenol and Isoeugenol,
respectively.
PREPARATION OF PHENOL-FORMALDEHYDE
RESINS
Other investigators have used the
tar!6 and creosote!? fractions from wood
pyrolysis for the preparation of phenol-
formaldehyde resins. The phenol extracts
(35g) were condensed with 42g of formalin
in aqueous barium hydroxide solution. A
low molecular weight reosole (7.9g) was
formed which after additional heating formed
a crosslinked re.sin.
EVALUATION OF PYROLYTIC OILS AS A
FEEDSTOCK FOR THE RECOVERY OF
CHEMICALS
The potential of the pyrolytic oils
as a source of chemicals is excellent. The
valuable chemicals present in the pyroly-
tic oils and their prices are shown as
listed in the August 21, 1978 issue of
CHEMICAL MARKETING REPORTER.18
ACKNOWLEDGEMENTS
The authors gratefully acknowledge
the U.S. Environmental Protection Agency,
Municipal Environmental Research Labora-
tory, Cincinnati, Ohio for support of this
work under Grant No. R-804440020.
10.
11.
12.
13.
REFERENCES
"Pyrolysis of Solid Municipal Wastes,"
E.P.A. Report No. 670/273-039.
Knight, J.A., Tatom, J.W., Bower,
M.D., Colcord, A.R., and Elston, L.W.
"Pyrolytic Conversion of Agricultural
Wastes to Fuels," 1974 Annual Meeting
of the American Society of Agricul-
tural Engineers.
Shafizadeh, F-, Sarkanen, K.V-, and
Tillman, D.A., "Thermal Uses and
Properties of Carbohydrates and
Lignins," Academic Press, New York
(1976).
Reznikow, V.M. and Novskii, Khim.
Prir. Soedin, 11 (1), 77-83 (1975).
CA. 83:44948h.
Vodqinskaya, A.R. and Tischenko, D.V.,
Gidroliz i Lesokhim Prom, 10, No. 5,
9-11 (1957). CA. 51:18604.
X
Shaposhnikov, Y.Z., Kosyukova, L.V.,
and Volkova, E.P., Gidroliz Lesokhim.
Prom. J20 (1), 16-17 (1967).
CA. 66:77124.
Kitoa, K. and Watanabe, Y., Zairyo,
16 (169) 844, (1967). CA. 68:88238.
Yasumity, W. and Koichiro, K.,
Mokyzai Kenkyu 38 (7), 40 (1958).
Gavars, M., Domburgs, G., Liepins, M.,
Tikhomirov, M., Khim. Ispol'z
Lignina, 368-75 (1974).
Chuprova, N.A. and Levin, E.D., Khim.
Tekhr-ol. Polim., No. 3, 78-83 (1972).
Levin, B.C., Tikhomirov, G.V., and
Popva, N.A., Materialy Konf. po.
Itogam. Nauchn. Issled. Rabotza 1964
god, Sibirsk Tekhnol. Inst.
Krasnovarsk, USSR, 54-7 (1965).
CA. 65:5659.
Hartley, R.D., J. Chromatogr. _54_ (3),
335-44 (1971). CA. 74:83835.
Goos, A.W. and Reites, A.A., Ind.
Eng. Chem. 38, 132-35 (1946).
120
-------
14. Hawley, L.F. and Wise, L.E., "The
Chemistry of Wood," The Chemical
Catalog Co., Inc., New York (1926).
15. Knight, J.A., Personal Communication.
16. Smolanka, I.V-, Dokl. i Soobshch.
Uzhgorodsk. Cos. Univ. Ser. Khim.,
No. 4, 3508 (1961).
17. Slavyanskii, A.K. and Piyalkin, V.N.,
Izv. Vysschkh Uchebn. Zabedenii Lesn.
Zh., _5, No. 2, 124-6 (1962).
18. Chemical Marketing Reporter, p. 49-59,
August 21, 1978, Schnell Publishing
Company, Inc., New York.
121
-------
TABLE 1. Condenser Oil Filtrate Capillary Gas Chromatogram Results
Peak
No.
3
4
13
17
21
30
34
37
38
42
48
50
51
58
64
66
69
70
79
Retention
Time (Min)
5.6
5.8
9.2
11.5
13.0
15.3
16.4
17.7
18.0
18.9
20.5
20.9
21.4
23.4
24.7
25.1
25.7
25.9
28.4
GC
%
0.76
1.95
1.81
1.45
0.62
2.18
4.63
5.15
1.02
7.90
1.24
12.98
1.29
2.19
2.82
2.08
1.69
1.17
8.91
Assignment
Isobutyl Alcohol
1-Butanol
Furfural
Furfuryl Alcohol
5-Methylfurfural
2-Octanol
Phenol
Guaiacol
o-Cresol
m-and p-Cresol
Naphthalene
2-Methoxy-4-methylphenol
Dimethylphenol
2-Methoxy-4-ethylphenol
2-Hydroxy-4-methoxyben2aldehyde
Unknown
Eugenol
2-Methoxy-4-propylphenol
Isoeugenol
TABLE 2. Condenser Oil Distillate Peak Enhancement Results from
Capillary Column
Peak
No.
13
17
21
30
34
37
38
42
48
50
51
58
69
70
79
Retention
Time (Min)
9.2
11.5
12.8
15.3
16.4
17.7
17.9
18.8
20.5
20.9
21.4
23.4
25.6
25.9
28.3
GC
%
0.27
0.71
0.36
0.76
5.67
4.71
2.50
5.97
2.72
10.40
3.00
4.83
2.20
1.69
3.08
Assignment
Furfural
Furfuryl Alcohol
5-Methylfurfural
2-Octanol
Phenol
Guaiacol
o-Cresol
m-and p-Cresol
Naphthalene
2-Methoxy-4-methyphenol
Dimethylphenol
2-Methoxy-4-ethylphenol
Eugenol
2-Methoxy-4-propylphenol
Isoeugenol
122
-------
TABLE 3. Current Prices of Chemicals Present in the Pyrolytic Oils
Chemical
Price
$ per Ib.
Comments
Acetaldehyde
Acetic Acid
Acetone
n-Butyl Alcohol
sec-Butyl Alcohol
Butyric Acid
m-Cresol
m,p-Cresol
o-Cresol
Ethyl Alcohol
Eugenol
Furfural
Furfuryl Alcohol
Guaiacol
Hexyl Alcohol
Hydroquinone
Isobutyl Alcohol
Isoeugenol
Methanol
0.20
0.18
0.17
0.21
0.23
0.325
0.98
0.44
0.50
.12
.25
0.515
0.54
2.60
0.24
1.54
0.19
6.85
0.44
1.
4.
technical grade
fermentation grade
synthetic
95-98%
90%
30.5°C m.p.
190 proof
USP
technical grade
mixed isomers
technical grade
synthetic
123
-------
IS 20
RETENTION TIME (Min.l
Figure 1A. Capillary Gas Chromatogram of Pine Condenser Oil
After Filtration in Acetone.
\JI
10
16 20
RETENTION TIME (Mm.)
Figure IB. Capillary Gas Chromatogram of Pine Condenser Oil Distillate.
124
-------
15
20 25
RETENTION TIME (Min.)
30
Figure 2 . Capillary Gas Chromatogram of Phenolic Extract.
125
-------
PYROLYTIC OILS FROM AGRICULTURAL AND FORESTRY RESIDUES AND
MUNICIPAL SOLID WASTE
J. A. Knight
Engineering Experiment Station
Georgia Institute of Technology
Atlanta, Georgia 30332
ABSTRACT
Pyrolytic oils can be produced in varying yields (15 to 25%) by the pyrolysis of agri-
cultural and forestry residues and municipal solid waste. Charcoal and combustible gases
are produced at the same time. Georgia Tech has developed a pyrolysis process, which has
been successfully demonstrated in a field demonstration facility at better than 90% on-time
of rated capacity. The quantities of biomass residues and cull and salvable dead trees in
the United States provide a tremendous resource for the potential production of large
quantities of pyrolytic oil along with the charcoal and combustible gases. The oils contain
a wide spectrum of organic compounds from low-boiling to high boiling, and many of the
compounds are oxygenated. The two major potential applications of the oils are as a fuel
and as a source of chemical materials for industrial applications. The utility of the oils
as a fuel has been demonstrated. In order for the oil to be used as a source of chemicals,
processing techniques must be developed for separating the oil into fractions of the major
chemical species present in the oils.
INTRODUCTION
Large quantities of agricultural,
forestry and municipal wastes are produced
each year in the United States. The proper
utilization of these materials so that they
can be considered a resource rather than
wastes is of extreme importance to the
country. At the same time, the disposal
and environmental problems these wastes
create would be solved. One approach for
the utilization of these materials that has
received a great deal of attention in the
past few years is pyrolysis. Pyrolysis of
lignocellulosic and/or cellulosic material
produces char, pyrolytic oil, water contain-
ing water-soluble organic substances, and
non-condensible gases. The char is primarily
carbon and can be used as a fuel or converted
to activated carbon, to produce gas for use
as a clean burning gaseous fuel or to
synthesize gas for organic synthesis. The
major components of the non-condensible
gases are hydrogen, carbon monoxide, carbon
dioxide and methane along with minor amounts
of the other hydro-carbon gases. The gas
can be utilized on-site as a clean burning
low BTU gaseous fuel. The pyrolytic oils
are clean burning with heating values
approximately sixty percent the heating
values of fuel oils. There is, however,
a great potential for utilizing pyrolytic
oils as a source of chemical materials for
industrial applications and/or as a chemical
feedstock. By upgrading the oils for uses
of greater value than as a fuel the total
economic benefit from waste materials would
be of greater significance to the country.
Also, the utilization of oils produced from
current waste materials as a source of
chemical materials would reduce the demand
on petroleum materials for chemical feed-
stock.
Pyrolytic oils are complex mixtures of
organic compounds ranging from very volatile
to high boiling materials. Many of the
compounds are oxygenated, and the oils
therefore are quite different in their
chemical and physical properties from
126
-------
petroleum. The characterization of these
oils and fractions obtained from them to
determine physical and chemical properties
provides the necessary data for the develop-
ment of the technology to process the oils
into more useful chemical materials for
industrial chemical applications.
GEORGIA TECH VERTICAL-BED PYROLYSIS SYSTEM
Over the past ten years, a vertical-
bed pyrolysis system has been developed at
the Georgia Institute of Technology. Bowen
et. al., have recently described in detail
this development and process [1]. The Tech-
Air Corporation licensed the process for
commercialization in 1970, and in 1975, the
American Can Company acquired Tech-Air as a
wholly-owned subsidiary.
The Georgia Tech pyrolysis process has
been demonstrated very successfully by the
Tech-Air facility at Cordele , Georgia. For
several months in the latter part of 1977
and into mid-1978, Tech-Air operated their
field demonstration facility (40 dry tons/
day) at better than 90 percent on-time of
rated capacity. The Georgia Tech pyrolysis
system can be operated to provide a variable
yield of a given product (i.e., char) with
a corresponding variation in the yields of
the other products (i.e., gas and oil).
The char yields can be varied from about 10
percent to 35 percent. The pyrolytic oil
yield can be varied from about 15 percent
to 25 percent. The gross energy recovery
in the form of products — char, oil and
gas — is approximately 95 percent of the
energy in the dry feed.
POTENTIAL QUANTITIES OF PYROLYTIC OILS
The quantities of biomass residues in
the U.S. are a potential source of signifi-
cant amounts of chemical materials. The
yield of pyrolytic oil from lignocellulosic
materials will vary from 15 to 25 percent
depending upon the feed material and the
conditions of pyrolysis.
Accurate estimates of waste materials
generated in the U.S. are difficult to
obtain. Tillman [2] has recently reported
that 80 million oven dry tons of wood resi-
dues from processing and manufacturing
operations are available annually in the
United States. This quantity of material
has the potential for providing 12 to 20
million tons of pyrolytic oil which is
about 24 to 40 percent of the tonnage of
petroleum used by the chemical industry as
chemical feedstock. In addition to the
available wood residues, there is a poten-
tial source of approximately one billion dry
tons of cull or rough trees and salvable
dead trees in the United States [2]. The
importance and significance of these large
quantities are that utilization of waste
materials and noncommercial timber, which
are unused resources, could have a tremen-
dous impact as a source of chemical mate-
rials as well as providing clean burning
fuels. In addition the utilization of
these unused resources would benefit the
environment and reduce dependence on
imports of foreign oil.
ANALYSIS AND TESTING OF PYROLYTIC OILS
The analysis of pyrolytic oils for
physical and chemical properties and
characteristics can be done by many of the
analytical procedures and techniques that
are utilized for petroleum, vegetable and
other oils. Knight et. al., have reported
on the physical and chemical characteriza-
tion of pyrolytic oils produced by the
Georgia Tech pyrolysis process from a pine
bark-sawdust mixture [3]. Typical proper-
ties that are determined for pyrolytic oils
are density, water content, heating value,
acidity, flash point, pour point, filterable
solids, ash, corrosiveness, viscosity, and
elemental content. The water content of
oils produced by the pyrolysis of agricul-
tural and forestry wastes is particularly
significant. The amount of water in pyroly-
tic oils is determined by the conditions of
the recovery of the condensible organic
phase from the off-gas stream from the
pyrolysis reactor. The water contained in
pyrolytic oils is well-emulsified and does
not separate on standing. In general, the
operating conditions for the recovery of oil
from the off-gas stream of the Georgia Tech
system produce oils with a water content in
127
-------
the 10 to 15 percent range.
The chemical characteristics of the
oil are of significance in the evaluation
of the oils for chemical applications.
Analytical methods and techniques that are
useful for determining the chemical nature
of the oils are gas and liquid chromoto-
graphy, gas chromotography/mass spectros-
copy, and infrared and ultraviolet spectros-
copy.
PROPERTIES AND CHARACTERISTICS OF PYROLYTIC
OILS
Oils produced by the pyrolysis of
lignocellulosic materials are complex
chemical materials with a wide spectrum of
organic compounds. The oils are reddish
brown to black; have a burnt, pungent odor;
are acidic and slightly corrosive; have
heating values about sixty percent the
heating values of fuel oils; have a density
greater than water; and are heat sensitive.
Some properties of pyrolytic oils produced
from a pine bark-sawdust mixture by the
Georgia Tech process are given in Table 1.
TABLE 1. PROPERTIES OF PYROLYTIC OILS
Property
Value
Density
Heating Value
PH
Acid Number
Flash Point
Pour Point
Filterable Solids
Ash
Water Content
9.2 - 9.5 Ibs/gal
9,000 - 11,500 Btu/lb
^ 3.0
30 - 75 mg KOH/g
111° - 121°C
2.7 °C
0.3 - 0.4%
< 0.1%
10 - 15%
The viscosity of pyrolytic oils depends
upon the operating conditions for pyrolysis
and the amount of water condensed with the
oil. The water is well emulsified in the
oil. The Georgia Tech process is generally
operated so that the water content of the
oils is in the 10-15 percent range. These
oils are relatively free flowing, but oils
which are essentially free of water are
viscous, and some have a grease-like con-
sistency at ambient temperature. The oils
are heat sensitive and prolonged heating
will result in increasing viscosity with
eventual formation of solids. A number of
viscosity-temperature curves have been
reported [3].
The oils have been distilled at both
atmospheric and reduced pressure [3]. Due
to the heat sensitivity of the oils, only
55 to 65 percent of the oils can be dis-
tilled. As the distillation proceeds, the
oil in the distilling flask becomes more
viscous, and in the range of 180° to 200°C,
the material in the flask will begin to
smoke with continued heating. Prolonged
heating results in decomposition of the
material in the flask. Fractional distilla-
tion of vacuum distilled pyrolytic oil did
not yield any fractions of close boiling
range. The liquid chromatograms of the
fractions indicated, however, that the more
polar and water soluble components were
concentrated in the low boiling fractions
whereas the less polar components were
concentrated in the high boiling fractions.
Hydrogenation of the oil at 4 and 20
atmospheres with a Pd catalyst showed that
about 2 mg of hydrogen per gram of oil was
absorbed. If one assumes an average
molecular weight of 150 for the oil, then
approximately 0.15 mole of hydrogen is
absorbed per mole of oil.
The pyrolytic oils contain a wide
spectrum of organic compounds from low
boiling to high boiling. Many of the
compounds are oxygenated, and consequently,
a variety of functional groups are present in
the oils. These functional groups contrib-
ute to the chemical properties and reac-
tivity of the oils. The major classes of
organic chemical species found in the pyroly-
tic oils produced from forestry materials by
the Georgia Tech process are: phenolics,
approximately 17-22%; aromatic neutral
compounds, approximately, 34-42%; acids,
approximately 6-7%, and "sugar-type"
substances, approximately 20-24%.
128
-------
POTENTIAL APPLICATIONS OF PYROLYTIC OILS
The pyrolytic oils produced from
agricultural and forestry residues and
municipal refuse have potential applica-
tions as a fuel and as a source of chemical
materials for industrial applications.
The heating values of pyrolytic oils
with 10 to 15 percent water are 60 to 70
percent of the heating values of fuel oils
on a volume basis. The water in pyrolytic
oils reduces the heating value on a mass
basis, but it also reduces the viscosity
of the oils, which aids in the material
handling of the oils and probably aids in
the atomizing of the oil. The pyrolytic
oil produced by Tech-Air at its field
demonstration facility in Cordele, Georgia,
was sold as a. fuel [1]. The oil was sold
for use as a fuel in a cement kiln, a power
boiler and a lime kiln. Pyrolytic oils
have potential use as a fuel for a hot gas
turbine. The utilization of the pyrolytic
oils as a fuel has been demonstrated.
With satisfactory combustion conditions
the oil can be burned cleanly. Since it
has essentially zero sulfur, SOx emissions
are not a problem. It can be admixed with
sulfur-containing fuel oil to reduce the
SOx emissions to acceptable levels. In the
handling and storage of the oil, it is nec-
essary to take into account the slight
corrosive nature of the oils and the
effect of prolonged, relatively high
temperatures on the oils.
The Pittsburgh Energy Research Center
test fired a slurry of number six fuel
oil and a 60-40 mixture of low volatile
wood charcoal and pyrolytic oil. The
slurry was prepared to contain 30 percent
wood charcoal [4]. The slurry was fired
in a 100 HP oil fired, packaged firetube
boiler. The flame stability in all tests
was reported as excellent. The NOX
emissions were lower than those obtained
from firing a coal-oil slurry, and the SO
emissions were proportionate to the con-
centration of sulfur in the slurry. The
results of these combustion tests demon-
strate that the charcoal and pyrolytic oil
can be fired successfully and are ideal
fuels to mix with coal and petroleum fuel
oils to reduce SOx emissions and to conserve
diminishing petroleum reserves.
The potential utilization of pyrolytic
oils as a source of chemical materials for
industrial applications depends upon the
development of processing techniques which
will separate the oils into fractions of
the major chemical species found in the
oils. Research and development work is
currently in progress at Georgia Tech to
develop separation processes for pyrolytic
oils. The phenolic fractions should find
immediate application in the production of
phenolic-formaldehyde resins for the ply-
wood and particle-board industries. The
aromatic neutral fractions should have
potential applications in many diverse
fields such as the rubber, plastics and
asphalt industries. Well characterized
fractions from crude pyrolytic oils will,
no doubt, suggest many industrial applica-
tions for these chemical materials.
REFERENCES
Bowen, M. D., Smyly, E. D., Knight, J.
A., and Purdy, K. R. A Vertical-Bed
Pyrolysis System, pp
Symposium Series, No
94-125, ACS
76, Solid Wastes
and Residues, Jones, J. L. and Radding,
S. B., editors (1978).
2. Tillman, D. A. Wood as an Energy Source.
Academic Press, Inc., New York (1978).
3. Knight, J. A., Hurst, D. R., and Elston,
L. W. Wood Oil from Pyrolysis of Pine
Bark-Sawdust Mixtur e, pp. 169-195.
Fuels and Energy From Renewable
Resources, Tillman, D. A., Sarkanen,
K. V. and Anderson, L. L., editors,
Academic Press, Inc., New York (1977).
4. Demeter, J. J., McCann, C. R., Ekmann,
J.. M. and Bierstock, D. Combustion of
Char From Pyrolyzed Wood Waste,
Pittsburgh Energy Research Center,
PERC/RI-77/9 (1977).
This work was supported in part by grant R804410 from the Municipal
Environmental Research Laboratory, U. S. EPA, Cincinnati, Ohio.
129
-------
SOME EMPIRICAL EVIDENCE ON THE EFFECTS OF USER CHARGES FOR
SOLID WASTE COLLECTION SERVICES
Robert J. Anderson, Jr.(l)
Vice President, MATHTECH, Inc.
P. 0. Box 2392
Princeton, New Jersey 08540
ABSTRACT
A basic proposition of economics is that when the prices of goods and services
increase, other things being equal, the quantity demanded of them will decrease.
Economists have suggested that this relationship between price and demand could be used by
solid waste managers as a tool for encouraging household waste reduction. Waste reduction
would be accomplished by levying charges for solid waste collection and disposal.
This paper presents empirical evidence from Tacoma, Washington, on the sensitivity
of demand for solid waste collection and disposal services to price. It suggests that (a)
households' choices of a service for waste collection and disposal may be sensitive to
price, and (b) that the total quantity of waste generated may not be very sensitive to
price.
INTRODUCTION
Economists have long argued that
solid waste collection and disposal should
be financed through fees levied on
households, businesses, and governments,
with fee levels set equal to the costs of
the solid waste services provided to them.
The rationale for this position is as
follows: If solid waste services are
provided free of charge, waste generators
are encouraged to use too many solid waste
collection services. If, in contrast,
services are priced at the cost of producing
them, service users will balance their
willingness-to-pay for services with the
costs of production of those services. In
general, the higher the charge for the
service, the less it will be used.
Inspite of the near-universal
acceptance by economists of the proposition
that service demands and service price
levels are inversely related, there is
almost no empirical analysis available to
support it. One study, a cross section
study by Stevens(2), provides some evidence
that service demand declines if service
price is increased, other things being
equal. In particular, Stevens finds that
the percentage of households subscribing to
backyard service in communities offering
this option is about .26 percent lower for
each one percent positive difference in the
charge for backyard service. She also finds
that a one percent positive difference in
the charge for more frequent collection
leads to approximately a 0.90 percent
negative difference in the percentage of
households choosing twice weekly collection,
and approximately a .05 percent negative
difference in tonnage of refuse collected.
This paper examines some time series
evidence from a single municipality —
Tacoma, Washington — employing a user
charge. My analysis will show statistically
significant negative relationships between
the number of households choosing a
particular service level and the level of
the charge for the corresponding service.
Estimated service price elasticities vary
from substantially less than one for low
levels of service usage to substantially in
excess of one for high service levels.
These results also will show relatively
little relationship between service prices
and total quantities of waste generated.
Taken together, these results imply that
pricing policy may have a relatively large
effect on households' choices of services
for disposing of solid waste, but a
130
-------
relatively small effect on how much waste
they generate.
The plan of the paper is as follows .
In the next section, the solid waste
collection system and charge policy employed
in Tacoma are briefly reviewed. Tacoma
offers a large number of service levels,
distinguished by the number of containers
picked up, the distance from curbside at
which the containers are placed for
collection, and the number of flights of
stairs above curb level at which containers
are presented for collection.
The section entitled "Model and
Results" describes the statistical models I
have employed to characterize the demand for
collection services in Tacoma. These models
include a multinomial logit model of
qualitative choice to explain choice among
alternative services, and a regression model
to,explain total quantity generated and its
disposition between collection,
self-hauling, and littering.
As noted above, the results
presented above suggest substantial
sensitivity of service usage to price at
high levels of usage (i.e., more than one
can and/or at some distance from the curb)-
Indeed, for all but the basic level of
service (one can at curb-side), my estimates
imply elasticities considerably in excess of
one. If correct, this means that individual
households' demands for service at high
levels of service usage may be somewhat more
elastic than is reflected in the empirical
results of Stevens.
My results on the effect of price on
total quantity of household waste generated
suggest that the effect on total waste
generation in Tacoma is nearly zero. In
particular, the estimates do show some weak
(statistically significant at the 10 percent
level) evidence that quantities presented
for collection decrease, and quantities
self-hauled increase in an offsetting
amount. These two effects roughly cancel,
with the result that total waste generation
appears to remain roughly constant, other
things being equal.
I conclude with a discussion of some
possible implications of my findings. The
most obvious implication is that cost-based
service pricing may result in a substantial
reduction in the demand for solid waste
collection services. This in turn means
that the economic efficiency of collection
could be improved substantially by a charge
policy.
USER CHARGES AND SOLID WASTE COLLECTION
AND DISPOSAL IN TACOMA, WASHINGTON
Collection and Disposal
Tacoma's solid waste disposal system
serves a total population of about 177,000
— 157,000 city residents and 20,000 outside
the city who either haul their own refuse to
the landfill or receive collection services
from haulers who utilize the Tacoma
landfill. About 3.5 percent of refuse
processed at the city landfill originates
outside Tacoma; figures are unavailable on
how much waste originating in Tacoma is
disposed outside the city, but the amount is
probably negligible. The city collects 56
percent jof Tacoma's solid waste, and another
15 percent is hauled by residents to the
landfill. Since 40 percent of the city's
collection is from residential sources,
household solid waste accounts for 37
percent of the total generated in the city
(i.e., 15 percent + 40 percent x 56 percent
= 37 percent), or about 1.2 tons/household
annually. The user fee charged to Tacoma
residents for household collection includes
a charge for disposal of $6.17 per ton.
Commercial haulers are charged at $6.17 per
ton for refuse originating in Tacoma, with a
$3 minimum charge and a $1.25/hour dumping
fee for each hour per load over one hour.
Commerical haulers are charged $9.92 per ton
for refuse originating outside Tacoma, with
a $5 minimum, and private residents from
outside Tacoma are charged $3.15 for refuse
brought to the landfill in cars and $5 for
refuse brought in pickups. Tacoma residents
are allowed to bring household refuse to the
landfill in cars or pickups free of charge.
Tacoma operates a fleet of 24
collection trucks, 22 of which serve
residential routes. Refuse is collected
once a week from each residential customer,
and crews work a five-day week. Trucks are
operated by two-man crews sharing driver and
collector responsibilities. The driver
carries a route book listing name, address,
and current service level of each customer,
and in which a record of extra bags
collected and changes in services are kept.
Three route supervisors handle most customer
requests such as requests for extra
collections or service changes, and reports
of unsanitary conditions. The route
131
-------
supervisor posts were created between 1973
and 1975 at the same time as the collection -crew
size was reduced from three to two. The
change in staff structure is felt to have
resulted in reductions in paperwork, more
efficient routing, and improvements in
safety, equipment operation, and customer
relations and service. Employees of the
Refuse Utility have been organized in the
Teamsters Union since the 1940's; labor
relations have been good in recent years.
The last major restructuring of
refuse collection took place in 1929 when
the Refuse Utility was inaugurated. Refuse
had previously been collected by private
haulers on a purely voluntary basis. Of
some 24,000 households in Tacoma in 1929,
only 6,000 subscribed to collection services.
The majority disposed of their own refuse in
a satisfactory manner. A serious health
hazard was not created from garbage left in
streets, on vacant lots, and alongroads on
the edge of the city. One of the first
first actions of the new utility was a
massive clean-up campaign. A dump site was
created in the swamp on the eastern edge of
the city which was used for over thirty
years. This land was eventually utilized
for industrial development over former
tidal flats.
Today Tacoma's landfill is still
inside city limits, less than 13 kilometers
from any point in the city. Hauling costs
for both the city and for private residents
taking refuse to the landfill are
considerably less than they might be if the
landfills were outside the city, so a great
deal of effort has been devoted to
prolonging use of the present site. The
original landfill in former swampland on the
eastern edge of the city reached capacity in
1960, when the present landfill, a ravine
southwest of the city center, was acquired.
The 81 hectare site of the ravine was
originally expected to serve Tacoraa's
disposal requirements for 10 years, but cost
estimates in the mid-sixties indicated that
to prolong the life of the landfill would
result in considerable savings since no more
suitable sites were available in or near the
city. Consequently, several projects have
been undertaken with the aim of resource
recovery and reduction of the volume of
waste disposed.
User Charges
Tacoma has some 50 years' experience
in financing solid waste collection and
disposal through user fees. The present fee
structure is a combination of a
capacity-based system and a location- or
service-based system.
Residential households living in
single unit or duplex structures receive
mandatory weekly collection at a rate
determined by the number of containers
presented for collection and the level of
carryout service provided. Households
presenting refuse within 7.6 meters of a
legal collection point (usually curbside,
but alleys in some districts) are charged
$2.45 per month for the first can and $1.15
per month for each additional can. Those
presenting refuse between 7.6 and 22.9
meters from collection points are charged
$3.90 for the first can and $2.65 for each
subsequent can. Between 22.9 and 61.0
meters, and over 61 meters, the base prices
are $5.30 and $6.65 and marginal prices are
$4.05 and $5.45, respectively. A charge of
$1.60 times the number of cans is levied for
each flight of up to six stairs between
points of collection and presentation.
These are monthly fees paid along with
electricity, water, and sewage bills on a
monthly of bi-monthly basis. In addition,
occasional extra bags of refuse left
alongside regular cans are collected for a
charge of $.75 each.
The original fee structure
introduced in 1929 varied only according to
pickup location; this was at a time when
almost all households probably demanded one
can per week or less. Since the
introduction of a variable fee f6r quantity
collected in 1938, the marginal rate for
added quantity has steadily decreased as a
proportion of the basic rate for minimum
service, while the marginal rate for added
location-based service has increased
relative to the basic rate.
The "bag" system, whereby residents
would be required to place rubbish in bags
whose price includes the cost of collection
and disposal, was established in 1976. It
was estimated in 1974 that $50,000 to
$100,000 in revenue were lost annually by
132
-------
not charging for bags left for collection by
households who had extra refuse
occasionally, but not often enough to
warrant purchasing an additional can of
weekly service. In April 1974 the Utilities
Service Director put forward a plan to sell
plastic bags through the city's fire
stations for 50 cents each to be used for
these "occasional extras." Consideration
was also given to making a complete
changeover to the bag system for all
household refuse collection. This latter
plan was rejected for several reasons: it
would require changing the city's ordinances
regarding the storage of refuse in
rodent-proof containers; it was felt that
plastic bags were too expensive (13 to 14
cents each); and paper bags would
deteriorate in the damp climate. Because of
problems with arranging for distribution of
bags, the more limited bag system for extras
only was rejected. Also, shops and
supermarkets were generally unenthusiastic
about serving as retail outlets, and fire
stations were not always sufficiently
staffed.
An alternate to the bag system was
finally adopted in May 1976. Collection
crews are given special cards on which to
note the address and number of bags for any
extra pickups. A clerk records this
information along with the account number of
the address where the extras were left, and
the customer is billed at the end of the
month. The price of this service is set at
a level that makes it cheaper for households
to use an extra can all the time if they
have an extra bag of refuse at least twice a
month on average. During the first year
this system was estimated to have cost about
$17,000 while developing added revenues of
about $69,000.
Administration of the System
Tacoma's solid waste system is
constituted as a classified utility under
the Department of Public Works. This is an
intermediate form between a department of
local government and an independent public
utility. Like most public utilities, the
solid waste system must finance its
operations (including capital expenditures,
debt service, and taxes) out of revenue it
collects rather than from general funds, and
its staff, equipment, and land are
autonomous from local government. But
unlike many utilities, its governing board
is the city council rather than a separate
utilities commission.
Tacoma's experience in finding a way
to collect user fees for extra bags of
refuse illustrates some of the
administrative advantages of a combination
of capacity — and location -- based
systems. Because the fee structure has
several parts (varying according to number
of containers, distance from collection
point, etc.) it has been possible to adjust
the relative prices of these items as costs
have changed. In turn, this has meant that
economic constraints have been available in
place of legal constraints such as limits on
quantities or pickup location. One of the
alleged benefits of the bag system is the
elimination of administrative costs
associated with recording service levels and
billing each household separately. But in
order to realize fully this benefit Tacoma
would have to eliminate entirely
location-based variables from its fee
structure, since these require separate
records and billing in any case. For
Tacoma, the card system involved few
additional administrative costs while still
yielding most of the benefits of the bag
system. In general, once a particular fee
structure is adopted it may be costly to
modify service or price. A combination fee
structure is flexible in allowing piecemeal
changes in one of several areas
independently, provided that all such
changes are compatible with all parts of the
current system.
Judging from press reports, records
of council debate, and numbers of
complaints, the Tacoma Refuse Utility has
caused little public dissatisfaction or
political controversy in recent years. The
price of collection and disposal in Tacoma
has been consistently less than in Seattle
and elsewhere in the region. There have
been press reports of sharp debate over
municipal vs. private collection and of
allegations of inefficient service in some
communities in the region, but none in
Tacoma. Since 1970 the Refuse Utility has
recommended two fee increases and one fee
modification (the addition of the charge for
extra bags). On each of these occasions the
council was satisfied that the fee increases
were justified. Each time, the council
considered one proposal to abolish the
133
-------
Refuse Utility in favor of private
collection on a contract basis and another
proposal to apply reduced rates for carryout
service to elderly and disabled residents;
both proposals were rejected each time by
votes of 8 to 1.
MODEL AND RESULTS
To model the demand for solid waste
collection services in Tacoma, I have
employed both a multinomial logit model and
a regression model. The multinomial logit
model is used to characterize choices among
the alternative levels of service offered in
Tacoma while the regression model Is used to
explain data concerning total quantities of
household waste generated.
Choice Among Alternative Services
As explained in the preceding
section of this paper, Tacoma offers its
residents a choice among a number of
different collection services distinguished
by the number of cans presented for
collection and the location with respect to
curbside at which these cans are presented.
In all, approximately 16 different service
levels are offered, ranging from 1 to 6 cans
per collection at various distances from
curbside.(3)
The basic hypothesis on which the
statistical model of service selection used
here rests is that the probability that any
given subscriber will choose a given service
level is related to certain attributes of
that service level and certain Individual
characteristics of the subscriber. In the
particular case at hand, the service
attributes would be such things as its
monthly cost, number of cans, and distance
from the curb; subscriber characteristics
would be such things as family size, income,
and other household characteristics.
The available data report the
frequency with which subscribers chose the
different available services In each of 59
months spanning the period January, 1973 to
November, 1977. The schedule of fees for
the various available services changed once
during this period (in January of 1975), and
the charge for collection of extra bags was
increased from zero to $0.75 per bag In May
of 1976.
The model used here to characterize
service choice data is the multinomial logit
model of qualitative choice. This model, as
noted above, expresses the probability that
an individual will choose a given
alternative as a function of the
characteristics of the alternative
available, the characteristics of the
chooser, and interaction terms between
alternative and choice-maker attributes.
The basic form of the model is as follows:
(1)
e
M
where P.
M.
Xi
the probability that the
jth individual will be
observed to choose the ith
alternative.
number of alternatives.
vector of
parameters.
K^ x 1 vector of attributes
of the ith attribute.
= K,
'vector of
parameters.
x 1
the
vector of attributes
jth Individual.
Yi
W..
= K,
1
parameters.
= K,
vector of
vector of
interaction variables formed
from products (or other
combinations) of alternative
attributes and individual
attributes.
This is the general form of the model I have
estimated using data on service choices in
Tacoma.
Before discussing the precise form
of the model estimated and the statistical
procedures employed, some explanation of why
the multinomial logit model was chosen to
characterize solid waste collection service
choice may be in order. There are, In fact,
134
-------
two reasons. First, it can be shown that
the logit model can be derived from a theory
which explains the distribution of choices
observed in a population of consumers with
differential preferences. In particular, it
can be shown that starting with some
assumptions about the nature of consumers'
preferences and the nature of the random
differences that distinguish individual
preferences, equation (1) can be derived by
assuming that consumers choose the most
preferred alternative available to them.
The probability distribution implied by
equation (1) may thus be interpreted as
predicting the probability distribution of
choices that would be made by a population
of utility maximizing consumers. Second,
among the alternatives available for
modeling choice among distinct alternatives,
tire multinomial logit model is
computationally the most tractable. These
two factors — consistency with economic
theory and computational considerations --
have led to the choice of the multinomial
logit.
Unfortunately, the data available
are not adequate to estimate the logit model
in its most general form (as shown in (1)
above). In particular, the data contain no
information on the characteristics of
individual subscribers in Tacoma that could
be matched with subscriber choices among
services. My working assumption is thus
that the distribution of these
characteristics in the population has
remained roughly constant over time, and
that these characteristics can be treated as
a part of the random variation in individual
preferences. The remaining explanatory
variables in the model are attributes of
alternative services, including price,
number of cans, and distance from the curb
at which cans are placed.
Employing this specification, the
unknown parameters of the model have been
estimated using the method of maximum
likelihood. Estimated parameters and
asymptotic t-ratios are shown below in Table
1. As the results reported in Table 1 show,
the fraction of the total subscriber
population-choosing any particular service
declines with the cost of the service, the
number of cans per collection the service
provides, and the distance from the curb at
which collection is offered. Taken at face
value, these results suggest that households
prefer lower levels of service (e.g., fewer
cans relatively close to curbslde), other
things being equal. While this result may
seem curious at first sight, many
investigators have reported that households
do not like backyard service (due to noise,
spillage, unfamiliar persons in close
proximity to the house) and do not care for
the space and other management problems
posed by additional cans. The negative
signs of the coefficients of number of cans
and distance from the curb could thus
reflect these considerations.
Estimated price elasticities of
demand for each of the services
corresponding to the parameter estimates
shown in Table 1 are reported below in Table
2. These estimates exhibit the pattern that
elasticities increase with service levels.
This is generally what one would expect. At
higher^ levels of service, one has more
alternatives to switch to in the face of a
price increase. For example, households
subscribing to two can service at a distance
of between 25 and 75 feet could consider
switching to one can service at this
distance, to two can service in the curbside
zone, or to one can service in the curbside
zone. Households subscribing to two can
service in the curbside zone can only reduce
their service demanded by switching to one
can service in the curbside zone.
Taken together, the results reported
in Tables 1 and 2 suggest that the demand
for solid waste collection service levels is
sensitive to service price, and that pricing
policy can have a very substantial effect on
the number of households subscribing to
higher levels of service.
Waste Generation
The results reported in Tables 1 and
2 above pertain to choices among the
alternative collection services offered by
the City of Tacoma. The other equations in
my model of the demand for collection
services in Tacoma are equations explaining
the total quantity of waste generated by
households in Tacoma. For these
relationships, I have adopted regression
equations relating estimated waste generated
in Tacoma by households to the price of
collection services, income (retail sales is
used as a proxy for income), number of
135
-------
TABLE 1. ESTIMATED PARAMETERS OF MULTINOMIAL LOGIT MODEL
Service
characteristic
Price
Number of Cans
Distance From Curb
Coefficient
estimate
-0.486187
-1.00216
-0.091494
Standard
error
5.394209E-03
5.685739E-03
2.394175E-04
Asymptotic
t- ratio
-90.1312
-176.259
-382.152
Gradient
-599.292
-227.691
-3209.7
TABLE 2. ESTIMATED PRICE ELASTICITIES USING MULTINOMIAL LOGIT MODEL
Service
(number of
cans, distance)
(1,1)
(2,1)
(3,1)
(4,1)
(5,1)
(6,1)
(1,2)
(2,2)
(3,2)
(4,2)
(1,3)
(2,3)
(3,3)
(1,4)
(2,4)
(3,4)
Price of
service
2.45
3.60
4.85
6.00
7.15
8. 30
3.90
6.55
9.20
11.85
5.30
9.35
13.40
6.65
12.10
17.55
Share of
market
.7809
.1639
.0328
.0069
.0014
.0003
.0125
.0013
.0001
.0000
.0000
.0000
.0000
.0000
.0000
.0000
Estimated
elasticity of
demand
0.2610
1.4634
2.2807
2.8970
3.4714
4.0341
1.8724
3.1804
4.4725
5.7613
2.5768
4.5458
6.5149
3.2331
5.8829
8.5326
Distance key:
1 = 0-25 feet
2 = 25-75 feet
3 = 75-200 feet
4 = > 200 feet
136
-------
subscribers, and season. Three different
equations have been estimated, one for each
of the components of total waste generation.
These three components are total residential
collection, litter (which is assumed to be
generated by households), and the estimated
quantity of waste self-hauled by households
for disposal at the city landfill. The
estimated equations for each of these
components of household solid waste are
summarized below in Table 3.
In broadest terms, the results shown
in Table 3 are consistent with what one
would expect. Increases in the price of
collection services tend to reduce the
quantity of household waste presented for
collection and to increasethe quantity
of self-hauled household waste. Collection
service price seems to have a small negative
effect on litter that is practically insigni-
ficant. Note,however, that only in the
equations fromself-hauling and collection is
the price term significant at the 10% level.
Because of the fact that the
regression equations reported in Table 3
were estimated in the natural logarithms of
the variables shown, each coefficient may be
interpreted as an elasticity. Interpreted
in this fashion, the regression results
imply a relatively low price elasticity of
demand for quantity collected with respect
to average collection charge of
approximately 0.15. If collection charges
were increased across the board by 10
percent, approximately a 1.5 percent
reduction in quantity of waste presented for
collection could be expected, and quantity
self-hauled could be expected to increase by
about 5.4 percent. Since self-hauling
accounts for about 20 percent of household
solid wastes and municipal collection
accounts for about 80 percent (litter is &
negligible fraction), the estimated
percentage change in quantity generated
associated with a 10 percent increase in
price would be approximately zero (i.e., 0.2
x 5.4 - 0.8 x 1.5). This implies that the
TABLE 3. REGRESSION MODEL RESULTS (t-STATISTICS SHOWN IN PARENTHESES)
log
(tons of solid
waste collected)
log (tons)
of solid waste
self-hauled)
log (tons of
litter)
Constant
2.2861
20.4863
-6.4047
log(Deflated Average Service Price) -0.1516
(-1.36Z6)*
log(Deflated Retail Sales)
logfNumber of Subscribers)
Winter Months Dummy Variable
0.0615
(0.3707)
0.5098
(1.7714)*
-0.0745
(-2.9772)*
Extra Bag Service Dummy Variable -0.0319
(-1.4324)*
0.5350
(1.2344)
-1.6283
(-2.5182)*
0.5658
(0.5046)
-0.7935
(-8.1399)*
0.1690
(1.9458)*
-0.7120
(-0.6414)
-2.2052
(-1.3314)*
3.4908
(1.2154)
-0.0114
(-0.0456)
-1.3446
(-6.0437)*
R2
F(5,41)
D.W.
0.1576
2.721
2.06
0.6224
16.167
1.70
0. 5402
11.808
1.87
* Significant at the 0.10 level.
137
-------
main effect of increases in the price of
solid waste collection services is to
increase self-hauling. There is no
reduction in quantity of waste generated
according to these results.
The coefficients of the proxy used
to represent income — retail sales —
indicate that self-hauling and littering are
negatively associated with income levels,
and that quantity presented for collection
is associated positively (but weakly) with
income. The retail sales variable is
statistically significant, however, only for
self-hauling and littering, with
self-hauling and littering decreasing with
retail sales.
Increases in the number of
subscribers (which is equal to the number of
households in Tacoma) result in increases
in all three components of the household
waste stream. Our estimates for collection
and self-hauling (only the coefficient in
the collection equation is significant at
the 10 percent level) imply that waste has
increased less than in proportion to the
number of subscribers. Our estimate for
littering suggests that this component has
increased at over three times the rate of
increase of households, other things being
equal. As has been suggested above,
however, littering is an insignificant
portion of the total, and may not have been
household litter. In addition, the
coefficient of number of households is not
statistically significant.
The coefficient of the dummy
variable for the winter months indicates
that less waste is presented for collection
and less is self-hauled during the winter
months than during other months of the year.
This probably reflects seasonal patterns in
household waste generation, particularly
with respect to yard wastes.
The coefficient of the dummy
variable representing the formal
introduction of extra bag service and
associated increase in the price of this
service in 1976 suggests that the price
increase reduced household collection and
increased self-haul ing, and reduced
littering (perhaps by providing a ready
means to dispose of temporary excess
wastes).
To sum up, the statistical results
presented in this section show that the
choice of service level is sensitive to
price, and that choices are somewhat more
responsive to price changes at relatively
high service levels. This is, I
hypothesize, because there are relatively
more substitution alternatives available to
households at high levels of service (i.e.,
they either can reduce the number of cans
per collection, or the distance from the
curbside at which cans are presented for
collection, or they can self-haul) . My
results also suggest that the quantity of
waste generated seems to be relatively
insensitive to price. However, the choice
between collection and self-hauling does
seem to be sensitive to price.
Let me hasten to caution that the
results presented are far from conclusive.
The data available are relatively rough, and
the statistical procedures and assumptions I
have made in modeling them by no means are
the only ones that could be adopted. If
these assumptions or modeling methods were
changed, it is quite possible that the
results would change also.
CONCLUDING REMARKS
Keeping firmly in mind that all of
the statistical results reported in this
paper are tentative, it is still useful to
discuss the implications for solid waste
management which follow from my results.
Certainly the strongest implication is that
user charges can be used to affect the
demand for levels of service. Increasing
prices on high service levels seem to result
in a reduction in the number of households
choosing high service levels. To the extent
that provision of high service levels is
uneconomic, such a charging policy can
increase the economic efficiency of solid
waste management.
Some idea of the overall difference
in demand for services under a graduated
user charges policy, like that employed in
Tacoma, and under a flat rate policy can be
gotten by comparing predicted service
demands under these two charge systems.
This is done in Figure 1 below, where
predicted percentages of subscribers are
shown on the vertical axis, and the various
138
-------
Percent
80
70
60 —
50 —
40 —
30 —
20
10 —
(1, 1)
Amount by which percentage of subscribers
under present rates exceeds percentage under
flat rate
Amount by which percentage of subscribers
under flat rate exceeds percentage under
current rate
(2, 1) (3, I) (4,l)(5, l)(6,
_
' (3,Z)' (4,2)' (1, 3)'(2, 3) '(3, 3)'(1, 4) ' (Z, 4)' (3, 4) ' (Cans' Distance>
Figure 1. Percentage of households at each service level under present charge rates
and under a flat rate system.
-------
can/distance combinations that constitute
the available services in Tacoma are shown
on the horizontal axis. As can be seen in
Figure 1, the percentage of households
subscribing to basic service is estimated to
be about 17 percentage points higher under
Tacoma's current graduated charge policy
(see the diagonally shaded area in Figure
1). These 17 percentage points reflect a
predicted shift away from higher levels of
service. The predicted net percentage
shifts from each of the higher levels of
service are shown by the stippled areas in
Figure 1.
These calculations suggest that the
aggregate effect of pricing policy on
service level choice may be quite large. If
the costs of providing higher service levels
are large, this suggests that the efficiency
gains from adoption of a cost-based pricing
policy could be substantial.
A second implication of my results
is that service prices have little effect on
the quantity of waste generated by
households. While my results do show some
tendency to reduce tonnage placed for
collection, they also show an approximately
equal increase in tonnage self-hauled to the
landfill.
Overall, these results support the
proposition that households' demands for
solid waste collection and disposal service
are sensitive to price. In addition, these
results suggest that the degree of
sensitivity may be quite high, and therefore
that the solid waste management efficiency
gains typically ascribed to a user charge
policy that sets charges equal to service
costs may be correspondingly great.
NOTES AND CITATIONS
(1) I am indebted to Fritz Efaw of
MATHTECH for his able collection of
the data for this study under EPA
Contract No. 68-03-2634.
(2) Stevens, Barbara. "Pricing Schemes
for Refuse Collection Services: The
Impact on Refuse Generation."
Research Paper No. 154, Graduate
School of Business, Columbia
University, New York, New York,
January 1977.
(3) Virtually no subscribers elected
options to have "flight of stairs"
service. These options are therefore
ignored in my analysis.
140
-------
THE DESIRABILITY OF COMMODITY FUTURES
TRADING IN SCRAP MATERIALS
Roger C. Dower and Elizabeth H. Granitz
Environmental Law Institute
1346 Connecticut Ave., N.W.
Washington, D.C. 20036
ABSTRACT
The highly variable prices for many secondary materials introduce uncertainty
concerning future prices for these commodities. This price uncertainty tends to act
as a barrier to the consumption of scrap materials. This paper proposes commodity
futures trading in scrap materials, especially wastepaper and ferrous scrap, as a
means of lessening the impacts of future scrap price uncertainty. By providing a mar-
ket for managing the risk associated with price uncertainty, and by increasing the
flow of futures market information, futures markets in secondary materials should result
in increased demand for secondary materials, improved credit terms for traders in the
market, and a greater level of investment in scrap processing.
In order for a futures market in scrap to be an effective hedging device and to
provide reliable future market information, it must be shown that scrap commodities
can be successfully traded. Several characteristics of the markets for commodities
in which there is successful futures trading are enumerated and are compared with the
markets for wastepaper and ferrous scrap. In general, the market for ferrous scrap ap-
pears to satisfy the requirements for successful trading if the important grades of
scrap can be more rigorously specified. The case for futures trading in wastepaper
is not as clear.
INTRODUCTION
Secondary material or scrap markets
are characterized by unusual price vola-
tility. By all accounts, this price vari-
ability hinders investment in the second-
ary materials industry and adversely af-
fects recycling levels in the wastepaper
and ferrous scrap industries. This paper,
drawing on research performed for the U.S.
Environmental Protection Agency, presents
the case for a futures market in waste-
paper or ferrous scrap as one alternative
for mitigating the impacts of scrap price
volatility and thereby increasing the
level of recycling.
The format for this paper follows the
report entitled, "An Analysis of Scrap
Futures Markets for Stimulating Resource
Recovery.1^1) First, the underlying causes
of variable scrap prices and the resulting
price uncertainty are discussed and the
groundwork for the evaluation of futures
markets is laid. The next two sections in-
troduce the concept of futures trading in
secondary materials by summarizing the
mechanics of futures trading, presenting
the potential economic impacts of futures
trading, and comparing the characteristics
of several existing futures contracts to
those for scrap materials. The final
section of the paper provides a brief
analysis of the probable effects of
scrap futures trading on the supply and
demand responses of both participants
and non-participants in the market.
UNCERTAINTY AND RESOURCE ALLOCATION
IN SECONDARY MATERIALS MARKETS
Causes of Price Uncertainty for Secondary
Materials
The basic cause of the variation in
141
-------
secondary materials prices, and therefore
the high level of price uncertainty,- can be
found in the inelastic demand and supply
schedules for these materials. Prices
tend to be more variable when demand and
supply functions are unresponsive to price
and when they are subject to large shifts.
This situation is demonstrated in Figure
1. A shift in demand from D to D1 does not
lead to such a large change in price in
Panel A, where the supply function is rel-
atively elastic, as it does in Panel B
where both the demand and supply functions
are inelastic.
Price
TABLE 1. SUPPLY AND DEMAND ELASTICITIES
FOR SECONDARY MATERIALS*
Quantity
Panel A
Price
Quantity
Panel B
Figure 1. Variation in market prices.
Elasticities for secondary materials tend
to be quite low, particularly when com-
pared to demand and supply elasticities
for primary materials. Elasticity esti-
mates for secondary and primary materials
are reproduced in Table 1.
Economic theory demonstrates that the
elasticity of demand for an input to pro-
duction processes depends upon: (1) the
number and availability of substitute ma-
terials; and (2) the cost of the input rel-
ative to the selling price of the final
output. Neither of these reasons is suf-
ficient to explain the low elasticities
for secondary inputs.
Material
Wastepaper
Wood pulp
Scrap steel
Pig iron
Secondary lead
Primary lead
Scrap copper
Primary copper
Supply
Elasticity
0.40
NA
1.12
NA
0.48
1.00
0.32
1.67
Demand
Elasticity"1"
0.08
NA
0.64
NA
0.21
0.87
*References:
' '
Wastepaper
Scrap s
Lead*3)
Scrap copper
(*)
Primary copper *• '
"*"For secondary lead and scrap copper, the
demand elasticity refers to the industry
demand curve for primary and secondary in-
puts combined.
NA: not available.
There are two other, more relevant,
reasons for the low demand elasticities
for secondary inputs. First, many users of
primary and secondary inputs are vertically
integrated into primary material supply
sources as insurance against uncertainty
in input availability and possibly for
favorable tax treatment specific to primary
production. This provides an incentive to
rely upon primary inputs in preference to
secondary materials. Second, demand for
scrap inputs is closely tied to capacity
cycles in the paper and steel industries.
Firms that normally use primary inputs will
increase their demand for secondary inputs
when, because of upward swings in the bus-
iness cycle, their capacity to produce the
primary inputs is strained. A decline in
the demand for paper and- steel products
shifts the demand for scrap inputs back to
more normal levels. The two-level demand
structure and the vertical integration of
user mills implies that the demand for sec-
ondary materials will be less elastic and
subject to larger fluctuations than would
be the case for primary materials.
142
-------
The low supply elasticities are due
in part to (1) rising costs as marginal
supplies of scrap are brought to market,
(2) the fact that most prompt industrial
scrap is sold irrespective of price, and
(3) the low level of inventory holdings
by user mills and scrap dealers/processors.
The Impact of Price Uncertainty on
Resource Allocation
In the face of uncertainty over out-
put price, output and input use levels will
be lower than under certain demand or price
conditions( ' '. Therefore, because of un-
certainty over secondary materials prices,
output from the scrap processing or dealer
sector will be less than if scrap prices
were known with certainty. In terms of
scrap consumers, uncertainty over expected
prices for scrap materials introduces dif-
ficulty in making rational short-run pro-
duction and inventory decisions, and re-
duces the. reliance on secondary
materials as inputs to production proces-
ses. Uncertainty over present and future
prices for scrap has been cited as a major
deterrent to increased recycling of waste-
paper and ferrous scrap.' ' '
The impact of uncertainty is exacer-
bated in the scrap markets by the inability
of producers or consumers to protect them-
selves from the inherent risk of an uncer-
tain market environment. Forward contract-
ing and vertical integration, two common
market mechanisms for mitigating the ef-
fects of demand and supply uncertainty, are
not viable alternatives for the scrap mar-
kets. In addition, price variability and
the inability to cope with price risk, cre-
ates instability in industry profits and
adversely affects the ability of scrap pro-
ducers and consumers to obtain funds from
lending institutions and capital markets.
The stage is now set for the examina-
tion of alternative policy options designed
to rectify the imbalance between secondary
and primary materials use caused by price
uncertainty in scrap markets. The follow-
ing sections of this paper argue that an
organized scrap futures market would pro-
vide a means of forward trading in the
affected commodities, would generate future
prices, and would allow industry members to
hedge some of the. price risk in the markets
for secondary materials.
FUTURES MARKETS AND THEIR ECONOMIC
IMPACT ON THE CASH MARKET
Futures Markets
Futures markets differ from other mar-
kets (for example, cash, contract, or for-
ward delivery markets) on a number of
counts.(10) (1) A futures contract,
which calls for delivery of a commodity at
some future date, is standardized as to
quality and quantity as well as other terms
of delivery. (2) Futures contracts are
traded on organized commodity exchanges
under rules promulgated by the exchange.
For example, the copper contract traded on
the Commodity Exchange, Inc. in New York
calls for delivery of 25,000 pounds of 99.9
percent pure copper cathodes. The delivery
months are January, March, May, June, July,
September and December. Delivery can be
roade from any one of six Comex warehouses
located across the country. (3) Futures
markets are impersonal markets in that
buying and selling is done through a
clearing house, which is generally a part
of the exchange. (4) Futures markets are
extremely competitive markets. All trades
are made by open outcry in the specified
trading area, or "pit," of the exchange.
(5) Although it is a legally binding com-
mitment to deliver or purchase a commodity,
a futures contract does not transfer title
to the commodity. A long position, if a
contract has been sold, can be liquidated
by taking delivery of the commodity or by
taking an offsetting position in the mar-
ket. (6) Although deliveries can be made
on futures markets, generally buyers and
sellers are better off negotiating a cash
market transaction since the terms of
delivery on the futures market do not
usually represent the most profitable terms
for actual sale or purchase. Many con-
tracts traded on futures markets experience
less than 1 percent delivery rates.
Economic Impacts of Futures Markets
The economic impacts of futures trad-
ing can be divided into three general cate-
gories: (1) cash market pricing impacts;
(2) risk management impacts; and (3) mar-
ket information impacts. Each of these
will be discussed below.
Pricing Function of Futures Markets—
By serving as a central market place
for commodities, futures markets lower the
143
-------
transaction costs of trading in an uncer-
tain market environment. Traders buy and
sell futures contracts based on their ex-
pectations concerning current and future
demand and supply conditions. The compet-
itive nature of futures markets assures
that the futures prices will be unbiased
predictions of future cash prices.v ' By
providing a forum for the assimilation and
dissemination of market information and
expectations, futures markets have been
termed price discovery mechanisms.
The futures prices signal decision-
makers when to produce, consume, or hold
stocks and, in a sense, regulate production
and consumption of the commodity.' '
Theoretically, then, futures markets should
stabilize cash market prices as the futures
prices are translated into more orderly
transformations of market positions. At-
tempts to verify this theoretical impact
have reached differing conclusions' '
15,16,1? ) ^ In generai ,.^6 models and
techniques used have not been able to ac-
count for other structural changes that
may have influenced cash price variation.
In an analysis of the impact of futures
trading on the responsiveness of plywood
supply to price signals prepared for this
study, it was shown that supply response
for plywood increased after the initiation
of futures trading. In other words, with
a futures market a given change in price
would result in a larger change in the
quantity of plywood supplied.
On the other hand, it has been argued
that the speculative activity in futures
markets has a destabilizing effect on cash
prices. Theoretical and empirical ^tudies
do not support such an effect.( ' 2 if
speculators base their buying and selling
strategies on the best available market
information, they should actually act to
stabilize cash prices at their true value.
The Risk Management Function of Futures
Markets—
There are two types of direct partici-
pants in futures markets: speculators and
hedgers. Speculators buy and sell futures
contracts in hopes of profiting from favor-
able movements in the future prices; they
have no position in the cash market and
rarely intend to make or take delivery on
the futures market. Speculators add li-
quidity to a futures market by making it
easier for hedgers and other traders to
establish market positions and thus help
to assure a well functioning marketplace.
Hedgers are generally commercial
buyers or sellers of commodities traded
on futures markets, or closely related
commodities, and use the futures market
to manage the risk of an unfavorable
movement in the cash price that would
affect a position they hold or plan to
hold in the cash market. A hedge consists
of taking an opposite position in the
futures market from one held in the cash
market. Since near futures prices and
the cash price of a given commodity tend
to move together, a loss in one market
can be offset by a gain in the other. As
long as prices for a commodity and the
futures prices for another commodity are
highly correlated, the former can be
hedged against the latter. The hedging of
copper scrap against the copper contract
traded in New York or London is an example
of this type of hedging.
Table 2 illustrates a hypothetical
buying hedge, the purpose of which is to
establish a forward price for a planned
or actual commitment in the cash market.
A copper scrap dealer with no scrap on
hand enters into a contract to provide
25,000 pounds of #2 copper scrap to a
user in two months at 48.5
-------
TABLE 2. EXAMPLE OF A BUYING HEDGE
Item Activity
Cash
Futures
April:
Buy
Sell
48.5*
July 64.2*
June:
Buy
Sell*
50. 5*
— —
—
July 66
.2*
*Loss, 2.00*; gain, 2.00*; net loss, 0.
TABLE 3. EXAMPLE OF A SELLING HEDGE
Item Activity
Cash
Futures
April:
Buy
Sell
June:
Buy
Sell*
48.5*
46.5*
July 64.2*
July 62.2*
*Loss, 2.00*; gain, 2.00*; net loss, 0.
Although actual hedging decisions will
depend on a host of factors, these simple
examples serve to demonstrate the risk
management function of futures markets.
Where risk is defined as variance in in-
come and variation in income arises from
price fluctuations, a rational hedging
program will increase income stability for
producers and consumers( 2 °).
Market Information Effects of Futures
Trading—
Traders in futures markets use infor-
mation from numerous sources (exchange and
commission house publications, government
agencies, etc.) to form expectations con-
cerning future market conditions. This
information is reflected in the futures
prices generated by their trading activi-
ties. These prices are reported in the
media and are available to all market par-
ticipants. Widely disseminated price
information should lead to a more re-
sponsive marketplace. The ability of fu-
tures markets to increase the normal flow
of market information results from:
(1) the provision of a central marketing
place; (2) an increase in competition in
the market; and (3) lower transaction
costs of information exchange. As a
result "market prices provide more ac-
curate signals for resource allocation
when there is futures trading in a com-
modity.'^21)
FUTURES TRADING IN SECONDARY MATERIALS
The ability of any futures market to
generate the types of economic impacts
discussed above is highly dependent on
how successfully the market works.
Although the various criteria used
to judge whether a commodity can be suc-
cessfully traded on a futures market tend
to be vague, it is possible to identify
several characteristics that are important
in determining the success of a futures
contract. General characteristics are
presented first, followed by the applica-
tion of more specific criteria for futures
trading in scrap materials.
General Elements Necessary for Successful
Futures Trading
In a broad sense, a successful futures
contract can be defined as one in which
there is sufficient trading volume to en-
sure continued listing of the contract by
the exchange. To ensure hedging demand
there must be a reason for buyers and sel-
lers to want to use a mechanism to manage
price risk. Hedging demand is also a func-
tion of the ability of an exchange to de-
sign a standard contract for trading that
follows accepted industry practices in
terms of quality, quantity, and specifica-
tion of delivery. Furthermore, there must
be a general industry acceptance of the
futures market. Ill-informed or mis-
conceived notions concerning futures trad-
ing may result in a lack of hedging volume.
Once hedging demand is established, specu-
lative trading demand tends to follow.
While it is clear that a need exists
in the wastepaper and ferrous scrap markets
for some type of risk management tool,
general agreement among industry members
that futures markets represent a practical,
145
-------
or even desirable alternative is lacking.
These doubts typically result from the
fears that a futures market would super-
cede the normal marketing procedures for
scrap, and that speculators would enter
the market and drain industry members of
their profits.
Specific Elements for Successful Futures
Trading
Competitive Market Place—
To minimize the possibility of either
buyers or sellers controlling the market
for a commodity, competition should pre-
vail among both buyers and sellers. Lack-
ing complete knowledge of the structure,
conduct and performance of the firms in
the scrap industry, the concentration ratio
is a useful proxy for evaluating the com-
petitiveness of the wastepaper and ferrous
scrap industries. Table 4 lists the four-
firm concentration ratios for different
sectors of the steel, and pulp and paper
industries for 1967 and 1972. Using
Guthrie's criteria, that in industries
where economies of scale are substantial
concentration ratios of less than 30 indi-
cate competitive conditions, it is con-
cluded that all sectors in the paper and
steel industries are relatively competi-
tive, except for pulp mills, sanitary pro-
ducts^ blast furnaces and steel mills.
(2.23 The intense foreign competition for
ferrous scrap undoubtedly increases the
competition in the blast furnace sector
with respect to scrap purchases.
Although no concentration ratios are
available for the seller or scrap dealer
side of the markets in wastepaper and fer-
rous s,crap, it appears that this sector
fits we'll into the competitive mold. Of
the approximately 1800 firms in the U.S.
ferrous scrap industry, over 80% of the
firms have output volumes of less than
30,000 tons.(2 ) In 1963, there were some
2700 wastepaper dealers in the U.S., most
of whom were located near a few major
cities.
Standard Contract Grade—
A futures contract must define a homo-
geneous quality of the commodity to be
delivered. The delivery grade must be spe-
cified within small tolerances, be test-
able, be acceptable to members of the in-
dustry, and be representative of a sizable
portion of the commercial transactions in
the cash market for the particular commod-
ity. The standardization of the commodity
to be delivered assures that each buyer
knows exactly what grade and quality of
the commodity will be delivered.
There is an apparent consensus among
industry members and commodity exchange
officials that the specification guide-
lines which are now used in the wastepaper
and ferrous scrap industries are not suf-
ficiently rigorous to be of use in de-
signing a futures contract. A solution to
this problem is for scrap buyers, sellers,
and exchange officials to meet and outline
formal, concise tolerances in terms of
allowable contaminant levels, and then
formulate descriptions for at least the
major grades of wastepaper and ferrous
scrap. No. 1 Heavy Melting or No. 1
Bundles, appear to be appropriate grades
for ferrous scrap futures contract.
They are relatively homogeneous, have a
history of use in the industry, have a
great deal of buyer confidence, and ac-
count for a large portion of the commercial
transactions in ferrous scrap. For a
wastepaper futures contract, one of the
six or seven grades of corrugated mater-
ials could be chosen for trading. Cor-
rugated medium is an important component
of wastepaper, accounting for approximately
40 percent of all wastepaper consumed in
ttj| manufacture of paper and paperboard.
TABLE 4. CONCENTRATION RATIOS FOR
SELECTED SECTORS OF THE PULP AND PAPER,
AND STEEL INDUSTRIES (4 FIRM RATIOS)*
Item
1967
1972
I. Paper and Pulp:
Pulp Mills 45 59
Paper Mills 26 24
Paperboard Mills 27 29
Sanitary Products 63 63
Folding Boxes 22 28
Set-up Boxes 12 11
Corrugated and
Solid Fiber Boxes 18 18
146
-------
TABLE 4 continued
Item 1967 1972
II. Steel
Blast Furnaces and
Steel Mills 48 45
Steel Pipes and
Tubes 26 23
Grey Iron Foundries 27 24
Steel Foundries NA 24
Steel Wire and
Related Products 24 18
*Reference(2*).
Initially , trading in a wastepaper or
ferrous scrap futures contract should be
confined to one grade so that buyers of
the contracts will have confidence of re-
ceiving the par grade on futures market
deliveries. A single grade contract would
still be of use to buyers and sellers of
other grades to the extent that the prices
for different grades of wastepaper and fer-
rous scrap move together closely. The con-
tract could still be used for hedging, and
the trader need never consider making or
taking delivery on the market.
Correlation analysis performed on the
price series for individual grades of fer-
rous scrap and wastepaper showed that the
different grades of iron and steel scrap
are highly correlated^ while the individual
grades of wastepaper are not so closely
correlated. A ferrous scrap futures con-
tract in one grade could be used to hedge
against other grades, whereas the case for
wastepaper is not that clear. Trading
separate contracts in each grade of waste-
paper may be feasible, but it would reduce
the number of buyers and sellers in any one
contract and thus impair the liquidity of
the market.
Standard Pricing and Delivery
Arrangements—
For a futures market to be an effect-
ive hedging device, the futures price in
the month of delivery must converge with
the cash price. Although deliveries on
futures markets are relatively uncommon,
it is the threat of delivery that forces
the futures price and cash price to be
equal during the delivery month. If they
are not equal, traders will make or take
delivery to take advantage of the price
differentials until they do become equal.
Therefore, the terms of delivery on a
futures market are important in assuring
the success of a futures contract.
Delivery on a futures market is insti-
tuted at the seller's discretion any time
during the final month of trading in the
contract by presenting to the buyer,
through the clearinghouse, a warehouse
receipt if the commodity is being delivered
out of inventory or storage, or a shipping
certificate if delivery is from current
production. The contract stipulates which
delivery instrument is to be used. The
shipping certificate would seem to be the
appropriate delivery instrument for a
futures contract in wastepaper or ferrous
scrap, since dealer inventories of these
materials are held for only short periods
of time.
Futures markets traditionally use
one of two types of pricing arrangements:
(1) delivered to the market city, if the
seller normally assumes responsibility
for transportation costs; or (2) F.O.B.
a natural shipping point if the buyer
usually pays for transportation. Ferrous
scrap transactions almost always have the
seller paying transportation charges.
For wastepaper there seems to be no stan-
dard industry procedure. Generally,
longer-distance hauls of higher grades of
wastepaper are shipped with the seller
assuming the transportation costs. Pricing
delivered to the market city is recommended
for both ferrous scrap and wastepaper.
There are four main criteria for
selecting a base point from which prices
will be established: (1) there should
only be one base point; (2) the base point
should be a natural merchandizing channel
for the commodity;- (3) it should be a nat-
ural loading point for shippers; and (4)
storage and/or production facilities should
exist so that physical delivery can be
made at a known cost.C26) Although no cen-
tral marketing place exists for wastepaper
and ferrous scrap, the markets for these
commodities are highly concentrated.
Wastepaper buyers and sellers are con-
centrated between Philadelphia and Boston,
with another large market in Chicago. Any
one of the large cities within this region
could be used as a base point. It has been
147
-------
suggested by commodity exchange officials
that New York would be appropriate. Fer-
rous scrap dealers and users are concen-
trated in and around Pittsburgh. Pitts-
burgh would be a likely candidate as a
base point for pricing and delivery of
ferrous scrap.
Inspection procedures for commodities
traded on futures markets vary widely. An
exchange, in designing inspection terms for
a contract, will attempt to follow accepted
industry procedures. In the ferrous scrap
and wastepaper industries, the buyer gen-
erally has the right to inspect and grade
all purchases. The seller has the option
of accepting the decision of the buyer or
taking his materials elsewhere. Ferrous
scrap inspections depend on the grade of
scrap supposedly being delivered; the
higher the grade the less inspection.
Wastepaper deliveries are usually only vis-
ually inspected. For the buyer, the major
criteria in determining how far to carry
the inspection depend on the past per-
formance of the seller.
When it formulates inspection proced-
ures for a scrap futures contract, an ex-
change must make certain that these pro-
cedures are fair and do not favor one side
of the market over the other. Random in-
spections, by a private agency, of those
sellers who are allowed to deliver on the
futures market may be one alternative to
ensure fairness.
IMPACTS OF FUTURES TRADING ON MARKETS
FOR SECONDARY MATERIALS
The class of individuals and organi-
zations whose profit-maximizing behavior
would be affected by futures trading in
scrap can be divided into two groups:
(1) direct participants; and (2) nonparti-
cipants in futures trading. The first
group includes scrap dealers and scrap con-
suming mills who would use the risk manage-
ment function of futures markets. The
second group, or nonparticipants, are those
members of the scrap industry who use the
market information generated by futures
markets to make inventory and production
or consumption decisions.
Participants
Hedging in a futures market is con-
sistent with sound business practices. A
futures market in wastepaper or ferrous
scrap would enable buyers and sellers in
those materials to hedge their present
and future cash positions and to thus
remove that element of business risk as-
sociated with uncertain prices of the
product. The resulting increase in pro-
ducer and consumer income stability would
improve scrap dealers' and consumers'
access to capital since it would lower the
risk premium demanded by lending institu-
tions and financial markets. Hedging may
be of additional use to scrap-consuming
mills in that it would help them to reduce
quantity/availability uncertainties as
well as price risk.
Nonparticipants
One of the principal functions of a
futures market is its forward price discov-
ery role. The forward or futures prices
are published daily and are available to
nonparticipants as well as futures market
traders. The forward prices represent un-
biased estimates of the prices that will
prevail at a future date. To the extent
that these prices are used in forming
dealer and user price expectations, they
will affect inventory decisions, production
plans, and investment programs.
CONCLUSIONS
Our study suggests two crucial, but
not insurmountable barriers to futures
trading in wastepaper and ferrous scrap.
These are as follows. (1) A standard con-
tract grade that is acceptable to industry
members must be defined. Specification
of a tradeable ferrous scrap.grade may be
easier than defining a wastepaper contract
grade; the current grade specifications for
ferrous scrap appear to be more concise
and more widely accepted than those for
wastepaper. (2) A sound futures trading
information base must be established among
industry members so that a hedging demand
for a scrap futures contract is assured. Un-
til scrap consumers and producers fully un-
derstand how futures markets can be used to
their advantage, there will be insufficient
demand for the services of such a market.
In the long run, a successful scrap
futures market should lead to: (a) a
greater reliance on secondary materials, at
least those for which there is futures
trading,, as inputs to production processes;
148
-------
(b) improved credit terms for firms that
use the futures market for hedging pur-
poses; and (c) increased levels of invest-
ment in scrap processing, both by scrap
dealers and user mills. If organized fut-
ures trading in scrap is instituted, and
if the obstacles mentioned above are over-
come, it is likely that the supply of
scrap from dealers and the demand for
scrap by user mills will shift outward and
become more elastic. The net effect on
scrap markets would be greater price sta-
bility and an increase in the quantity of
scrap recovered and reused.
REFERENCES
1. Anderson, R. C. and R. C. Dower. An
Analysis of Scrap Futures Markets for
Stimulating Resource Recovery- EPA-
600/8-78-019.
2. Anderson, R. C. and R. D. Spiegelman.
"Tax Policy and Secondary Material
Use." Journal of Environmental Eco-
nomics and Management 4 (1977): 68-
82.
3. Anderson, R. C. "Economic Incentives
for the Recovery of Secondary Land."
Resource Recovery and Conservation,
2 (1977): 193-209.
4. Fisher, F. M.; P. H. Cootner; and
M. N. Bailey. "An Econometric Model
of the World Copper Industry." Bell
Journal of Economics and Management
Science 3 (1972): 568-609.
5. Pickard, W. C. and R. J. Krumm.
"Joint Aluminum - Copper Forecasting
Model," Proceedings of the Council of
Economics of the American Institute of
Mining, Metallurgical, and Petroleum
Engineers, 1977.
6. Sandmo, A. "On the Theory of the
Competitive Firm Under
Price Uncertainty." American Economic
Review 61 (1971): 65-73.
7. Stevens, G. Y. "On the Impact of Un-
certainty of the Value and Investment
of the Neo-classical Firm." American
Economic Review 64 (1974): 319-36.
8. Albrecht, 0. W. and R. G. McDermott.
Economic and Technological Impediments
to Recycling Obsolete Ferrous Solid
Waste. Springfield, Va. : National
Technical Information Service,
PB 223-034, 1973.
9. Battelle Columbus Laboratories. A
Study to Identify Opportunities for
Increased Solid Waste Utilization.
vol. 8. Prepared for the U.S. En-
vironmental Protection Agency.
Springfield, Va.: National Technical
Information Service, PB 212-731, 1972.
10. Gray, R. W. The Feasibility of
Organized Futures Trading in Resi-
dential Mortgages. Federal Home Loan
Mortgage Corporation Monograph No. 3.
Washington, B.C., 1974.
11. Gray, R. W. and D. J. S. Rutledge.
"The Economics of Commodity Futures
Markets: A Survey." Review of Mar-
keting and Agricultural Economics 39
(1971): 57-108.
12. Peck, A. E. "Futures Markets, Supply
Response, and Price Stability."
Ph.D. diss., Stanford University,
1973. Ann Arbor, Mich.: Xerox
University Microfilms, 1973.
13. Ibid.
14.
15.
16.
17.
18.
19.
Working, H. "Price Effects of Futures
Trading." Food Research Institute
Studies 1 (1960): 3-31.
Gray, R. W. "Onions Revisited."
Journal of Farm Economics 45 (1963):
273-76.
Johnson, A. C. Effects of Futures
Trading on Price Performance in the
Cash Onion Market. U.S. Department
of Agriculture, Economic Research
Service, Technical Bulletin No. 1470,
1968.
Powers, M. J. "Does Futures Trading
Reduce Price Fluctuations in the
Cash Markets?" American Economic Re-
view 60 (1970): 460-64.
Labys, W. C. and C. W. J. Granger.
Speculation, Hedging and Commodity
Price Forecasts^. Lexington, Mass.:
D.C. Heath and Company, 1970.
Brinegar, C.S. "A Statistical Analy-
sis of Speculative Price Behavior."
T?nod Research Institute Studies 9
149
-------
(1970): supplement.
20. McKinnon, R. I. "Futures Markets,
Buffer Stocks, and Income Stability
for Primary Producers." Journal of
Political Economy 75 (1967): 844-
61.
21. Cox, G. C. "Futures Trading and
Market Information." Journal of Pol-
itical Economy 84 (1976): 1215-37.
22. Guthrie, J. A. An Economic Analysis
of the Pulp and Paper Industry.
Pullman, Wash.: Washington State
University Press, 1972.
23. Regan, W. S., R. W. James, and T. J.
McLean. Identification of Opportuni-
ties for Increased Recycling of Fer-
rous Solid Waste. Prepared by the
Institute for Scrap Iron and Steel
for the U.S. Environmental Protection
Agency. Springfield, Va.: National
Technical Information.
24. U.S. Department of Commerce, Bureau
of the Census: Census of Manufactur-
ers, 1976 and 1972.
25. American Paper Institute. 1976-1979
Capacity: Paper, Paperboard, Wood-
pulp Fiber Consumption. New York,
1977.
26. Marcom, Inc. Standardization of
Transportation Costs for Futures Trad-
ing of Coal on a Commodities Exchange.
Prepared for the Federal Energy Of-
fice. Springfield, Va.: National
Technical Information Service, PB
245-398, 1975.
150
-------
AN ECONOMIC ANALYSIS OF PAPER RECYCLING
Roger Bolton
William College
Williamstown, Massachusetts
ABSTRACT
A useful, although partial and incomplete, framework for analyzing paper recycl-
ing is a review of various market imperfections which make the amount of paper recycl-
ing less than socially desirable, or less than "optimal," in terms of economic theory.
The most common imperfections are felt to be: uninternalized pollution which is
greater in production of paper from virgin pulp than from waste paper; discriminating
taxation in favor of virgin sources; uncertainty in .returns from investment in recycl-
ing and use of recycled paper, which inhibits investment; implicit subsidization of
disposal of waste paper after a single use, due to the failure to price accurately
solid waste disposal activities; biases, on the part of paper company managers,
against use of waste paper„ I review briefly these imperfections, the implications for
policy, and some needed directions in new research. Public policy should continue to
move toward elimination of the imperfections, but if that is not possible "second best"
arguments exist for introducing subsidies to recycling to offset the imperfections.
However, these arguments must be used cautiously.
INTRODUCTION
In this paper I plan to review generally
one frame work for an economic analysis of
paper recycling. I organize my discus-
sion around some market "imperfections"
which are widely assumed to make the
level of recycling less than socially
desirable, particularly relative to use
of virgin pulp and relative to disposal
of paper after a single use. In addition
to reviewing the imperfections and the
implications for government policy, I
suggest some directions for new research.
Indeed, the suggestions for research are
the main point of my paper.
This approach is necessarily partial
and imcomplete. The literature on paper
recycling is enormous, and it would do
little good here to review it, or to
introduce one more version of the ubiqui-
tous flow diagram or a precise definition
of "recovery," or to report on the latest
data on prices and quantities. Ample
sources are already available on that.
Particularly good overviews of the eco-
nomic analysis are found in (3,9,25,30,
31,35,A3). My primary goal here is to
focus attention on some needed directions
for further research.
(I intend to limit myself to recycl-
ing of paper as paper, rather than as
energy. This is partly for convenience,
but for other reasons as well. The
imperfections I focus on affectly mostly
the choices between virgin pulp and waste
paper in manufacture of paper products,
although the imperfection in solid waste
disposal pricing also affects energy
recovery. On-the supply side, policies to
affect reuse a.s paper must operate on
paper before it becomes part of "mixed
solid waste;" the technology for recover-
ing paper fiber from MSW produces very
low grade products with limited markets
(45,49). If paper is diverted before it
becomes MSW, reuse as paper is far more
valuable than reuse as energy.
151
-------
(44). This is greatly reinforced by the
fact that reuse as paper is quite consis-
tent with energy recovery anyway; paper
reuse merely postpones, but does not pro-
hibit, eventual recovery as energy (44,
49). And a big increase in recycling
would still not reduce the fuel value of
the remaining MSW much (48). On the other
hand, if paper can be recycled only after
it becomes MSW, energy recovery seems the
only possibility for large scale reuse
and it is what policy should focus on.
Thus, in a sense reuse as paper and energy
recovery are not very inconsistent in the
long run, even when reuse as paper does
not dominate. To the extent recycling.
is economical at all, policy should work
toward reuse as paper and prevent it from
becoming part of MSW. For paper which
does not become MSW, policy should en-
courage energy recovery to the extent it
is economical.
An economist necessarily views re-
cycling in the total context of resource
allocation. He does not see it as solely
a solid waste problem alone, or a resource
problem alone, as some government reports
may do (42)„ He does not see recycling,
or anything else, as an absolute good.
It provides no direct consumer satisfac-
tion, but is a production process—rather,
a sequence of processes—for the produc-
tion of paper products. As such, it has
alternatives, and the economist asks
whether under existing markets and poli-
cies there is an appropriate balance be-
tween recycling and the other processes.
He may also ask whether the total quan-
tity of paper products is appropriate,
and how recycling affects that.
The economist evaluates the relative
quantities of different goods and the
relative reliance on different production
processes. Depending on markets and poli-
cies, there can be too much or too little
recycling. Too much or too little means
relative to what would occur in a "per-
fectly" functioning economy, defined as
one with a competitive price system, no
externalities which are uncorrected by
policy, no uncertainity left uninsured
against, no distortions of prices by
subsidies or taxes, and one which has an
"acceptable" income distribution. Al-
though in principle one could conclude
that imperfections could cause too much
recycling, almost everyone agrees that
the cumulative effect of the existing im-
perfections in this country cause too
little. So from now on I'll always refer
to "too little," or "insufficient," or
"sub-optimal" recycling of paper.
Precisely what is meant by "too
little" is that some other allocation of
resources, with more recycling and less of
some other good(s), would be superior.
Supreior means that the value of "national
income," appropriately defined, would in-
crease and that the new distribution of
income would be no worse than the original
one. Those persons who gain by the change
would gain more than the losers lose, in
the sense that the gainers could compensate
the losers, leaving them unharmed, and
still have gains left over. Also, we
assume that in the new situation, a rever-
sal would not allow gainers to compensate
losers. In technical parlance, such a
change in allocation is said to meet the
Kaldor-Hicks and the Scitovsky criteria
(13,17-19,22,27,38).
In most of the paper, I am going to
assume that the distribution of income
does not change with the reallocation
enough to concern us. In other words, I
assume we can ignore the distributional
changes and make a judgement solely on
the sign—positive or negative—of the
change in national income. I shall make
a few remarks about distribution later.
In this kind of analysis, national income
must be defined broadly to include the
value of all goods and services and activ-
ities (including leisure) which enter
directly into consumers' utility functions
with the values determined by willingness
to pay, regardless of whether there are
any market trades in the goods or activi-
ties—some of the values must be imputed,
in other words. Environmental quality
must definitely be included. Obviously,
this measure of national income is much
broader than "GNP."
152
-------
The value of national income will be
at a maximum if there are no imperfec-
tions in the economy. The proof of this
is a standard result in welfare economics,
and will not be repeated here. In this
case the amount of recycling is just right,
or "optimal." Note that one cannot say
there is too little recycling unless one
can point to a particular imperfection:
one must be able to pair up a misalloca-
tion with an imperfection. Also, one
must always be able to pair it with a
misallocation of the "opposite" kind,that
is, if there is too little recycling
there must be too much of something else0
If one is to argue there is too little
recycling, one must find something else
such that national income would be in-
creased if recycling were increased and
the other thing decreased.
It is generally agreed that several
imperfections cause too little recycling
and too much use of virgin materials; it
is also generally agreed that the excess
production from virgin materials also
causes a sub-optimal amount of environmen-
tal quality and an excess amount of use of
primary energy materials. Thus, the -chief
resource allocation effects are too little
recycling and environmental quality, and
too much use of virgin materials and
energy. Some imperfections also cause
other misallocations, in addition to these.
For simplicity, from now on I shall
refer to paper produced from virgin mater-
ial, and the associated production process,
as V, and the paper produced from recycled
paper as well as the production process
and the required recycling activities,
as R.
SOME PERSPECTIVE
Before reviewing the imperfections,
we must gain some perspective. Increas-
ing the use of recycled paper although
offering some marginal improvements in
overall resource allocation, has an
inherently tough road to hoe. We must
not expect dramatic increases, certain-
ly not quickly, even if we eliminate of
offset all the imperfections, which we
can't do anyway. The research on the
imperfections and their effects has
made that quite clear. The supply and
demand of waste paper have often been
shown to be inelastic, which means that
either an increase in price to suppliers,
or a decrease in price to users, alone,
will not expand the market much (1,2,9,10,
12,34,35). And a shift in either demand
of supply will cause large price changes
which curb the responses from the other
side of the market. What seems to be
required is a simultaneous shift in both
supply and demand functions, so that
quantity can be increased without much
change in price0 This is not out of the
question,, but opportunities, especially
ones soundly based in welfare economics,
for doing this quickly seem very limited.
The longer-run picture is very clouded,
and much of the research done so far has
not dealt with it well,,
It must be remembered that the U.S.,
like some other countries which don't
recycle much (43), has a large and low
cost virgin pulp supply0 Recycling is
much more labor intensive than virgin
production, both in the paper production
process and in the waste paper handling
itself„ The increasing relative price of
labor, common to all advanced economies,
inhibits recycling therefore. Most of
the paper not already recycled is in
the "post consumer waste" category, where
labor costs and household resistance are
high. There are technical limits to the
amount of recycled paper which can be
substituted for virgin pulp in many grades
of paper, and not all of these are trace-
able to consumer ignorance or purchasing
agent biases0 Recycling is traditionally
more important in some grades of paper
than others, but the future growth rates
in those grades are forecast to be lower
than for the other grades (15). The
differential growth rates by grade could
swamp modest increases in the recycling
rates for each grade. (Arithmetically,
of course, one could conceive of the
recycling percentage rising in every
single grade, but the recycling percen-
tage rate for the whole economy falling!)
The bulk of existing paper production
capacity is designed exclusively or
153
-------
predominately for virgin pulp, including
being located near virgin supplies and
far from waste paper supplies; it will
take a long time for investment decisions
at the margin to change the nature of the
capacity and its location to use more
waste paper. Some regional considerations
are also important. The supply of waste
paper is especially abundant in regions
of the country (Northeast, Midwest) which
are now growing more slowly that other
regions (wouth, West), which normally
rely more on virgin pulp and which are
less urbanized, making the supply of
waste paper more costly. And the recent
faster growth in nonmetropolitan, less
urbanized regions of the country (the
"rural renaissance") also seems to work
against increases in the use of waste paper.
MARKET IMPERFECTIONS
Without pretending to draw up an
absolutely exhaustive list, I thing the
main imperfections are as follows:
A. Differential pollution in V and
R (V=virgin, R=recycle). It is apparent
that in most locations, the potentially
damaging "residuals" per unit of output,
from V are potentially more damaging
than those from R. The situation is compli-
cated by the fact that both V and R pro-
duce harmful residuals, in different pro-
portions; R definitely produces more of
some harmful residuals than V (7-9,36,43,
46), and in some locations the potential
damages are probably greater. The liter-
ature on residuals does not disaggregate
by location very much, leaving a great
deal of useful research to be done on
specific locations which might be affected
by broad national policies. The de-inking
necessary to produce higher grades is
especially singled out as generating more
solid wastes and waterborne suspended
solids (43). Bower presented a great deal
of data on the residuals trade-offs in-
volved in choosing a process, specifica-
tion of product, and raw material in the
production of paper (7,8). For example,
in the production of tissue, one can
technically use up to 100% waste paper in
a draft (sulfate) process (brightness
25GEB, waste paper No. 1 mixed). As opposed
to 100% virgin pulp, 100% waste paper would
produce a 50% increase in S(>2 (the SC>2 would
be 0 if sulfur-free fuel oil were the energy
source, but in Bower's model 1% sulfur fuel
oil is assumed, and with that the total
SOj actually rises); a big decrease in
reduced sulfate compounds (hydrogen sul-
fide and organic sulfides) and in parti-
culates and in inorganic solid residuals;
an increase in organic solid residuals;
and a nearly 50-fold increase in suspended
inorganic solids. But the increases and
decreases are quite different if one con-
siders other processes, other brightness
specifications, and other kinds of waste
paper. It is very hard to generalize on
specific residuals (8)„ It would be even
harder to generalize on the damages caused
by discharge into the specific environ-
mental media, with varying assimilative
capacities, in specific locations. And
even Bower's careful work has been criti-
cized for not recognizing the variations
in process specifications from plant to
plant, due to varying engineering and
management practice and skill, and age
of capital equipment (MacAvoy (8).
However, as I have said, there is a
presumption that in the typical location
V produces more harmful pollution (36,
which has an extensive list of references;
46; for a more reserved conclusion, see
31,43). The case is strengthened if one
adds the environmental damage in forestry
(14) to the paper manufacturing effects.
Now, if society does not take steps to
prevent the damages, the damage costs will
be external to producers; the market prices
of V and R will both be below their true
social costs, but the price for V will
diverge more. Thus, paper producers will
be induced to use more V and less R than
is optimal. However, one must recognize
that public policy is moving gradually
toward controlling pollution, which pre-
sumably is reducing the difference in the
price of V relative to R.
B. Discriminating taxation in favor
of V. The chief culprit here is the capi-
tal gains treatment of profit from cutting
of standing
154
-------
timber for paper production, and also in
some cases a lower property tax rate on
timber reserves than on durable equipment,
and some minor advantages of expensing
capital expenditures (1,3,47). The long-
run effect of the tax advantages is to
lower the cost of virgin paper production,
and this seems to be shifted forward on
to buyers, thus expanding the use of V
above what is would be in an optimum
allocation.
However, it appears the effect on
the price of V is rather small„ Coupled
with the usual estimates of low demand
elasticity for V, and of low cross
elasticity between V and R, it appears
the effect on the relative quantities of
V- and R is quite small, probably well
under one per cent. Anderson et a!0
estimated that the effects on the price
of virgin paper was 402%, even under
favorable assumptions, and that the
most likely effect was actually only
about 1% (1). An earlier report for EPA
suggested the effect varied a lot by
type of paper, and could be as high as
2% for linerboard, corrugating medium,
and printing and writing paper; the
report also indicated that the tax effect
on the price could be a large proportion
of the total difference in production
costs of V and R; for example, the tax
effect was estimated as about 70% of
the total differential for linerboard and
corrugating medium, 60% for one kind of
combination boxboard, and only 26% for
printing and writing paper (47). But
even such sizeable numbers may imply
very little effect on relative quantities
of V and R, except possible there could
be sizeable effects in the very long
run. As indicated earlier, we really
can't say what would happen if various
imperfections were removed and we waited
a long time to measure all effects.
C\_ High degree of uncertainity
in returns to capital invested in recycling
operations and production of paper from
recycled material. This is due to the
extreme price fluctuations in the price
of waste paper (2,3,9,10,lOa,35,43-44)„
The fluctuation is one of output prices
for firms, households, and organiza-
tions engaged in collecting and selling
waste paper; the uncertainty in the
rewards to their effort inhibits long-
run capital investment, organizational
effort, development of experience and
routines, and the making of long-term
commitments to supply waste paper. It
should be clear that uncertainty affects
not only business firms, but also house-
holds and non-profit organizations whose
volunteer has usually been a vital input
into waste paper supply. In times of
high prices, organizations have developed
networks of households who are willing
to separate out newspapers and other
forms of waste paper, only to have no
market for the paper a few months when
prices fell so drastically.
This same fluctuation is one of
input prices to paper mills which rely on
waste paper as a raw material. Waste
paper prices fluctuate much more than
paper product prices, so the profit
margins of such mills also fluctuate
a great deal. There are standard
economic arguments on why such uncertainty,
even though it sometimes raises profits
and sometimes lowers them, is less
conducive to long-term investment and
long-term contracts than a stable
profit which is equal (or even lower)
to the average of the fluctuating
profit levels.
Unlike in the effects of pollution
control and tax advantages, we do not
have estimates of just how much this
uncertainty inhibits investment. Our
models for the estimating effects of
uncertainty on profit, and on investment,
are not as well developed as the indus-
trial process and competitive market
models used in the other two areas. Nor
do we have the systematic data on the
attitudes of entrepreneurs and managers
and their opportunities for reducing
uncertainty through diversification,
holding of inventories, etc. A waste
paper dealer might diversify by partici-
pating in more than one secondary material
market; when paper prices are up, perhaps
scrap prices are down, etc. To some
extent households and non-profit
155
-------
organizations might do the same. It
would appear that paper producers, who
buy waste paper as an input, are better
able to hold inventories as buffer stocks,
but it is hard to see them diversifying
into the completely different end-product
industries which use other secondary mater-
ials, e.go, metals„
Fluctuation in waste paper prices
has been attributed to the large shifts
in inelastic demand curves along very
inelastic supply curves» The large shifts
are partly due to the nature of demand
for products using waste paper (e.g.-,
construction paper, insulating materials)
and the oligopolistic nature of the
paper products industry, which means that
fluctuations in demand are coped with
predominately by output fluctuations
rather than (product)price fluctuations„
Waste paper is often the residual or
marginal source of input supply, and
demands for it fluctuate much more than
demands for the virgin pulp input; this
accentuates the instability in paper
output. Paper firms pay large prices
for marginal inputs in times of high
demand for their products, because the
extreme capital intensity and the vertical
integration of most firms puts a premium
on keeping enough inputs flowing to
maintain a mill in operation.
D. Implicit subsidization of dis-
posal of waste paper. In many locations,
household consumers of paper do not pay
the full monetary cost of collecting and
disposing of the paper after use. They
do not pay market prices for municipal
solid waste "management," but rather it
is financed out of general property
or other municipal taxes which do not vary
as the household's demands on the system,
or the costs imposed on it, vary. In
addition, even the businesses, in most
locations, and the households, in some,
who do pay user charges for solid waste
management may not be required to pay
the full social cost of collection and
disposal, because there are some exter-
nalities, chiefly in disposal, not
reflected in charges. For both these
reasons, solid waste disposal is sub-
sidized. As disposal is a complementary
good to V or a substitute for R, the impli-
cit subsidization helps overexpand V and
restrict R,, The subsidization reduces
the incentives for recycling, by prevent-
ing some of the social cost savings from
recycling from flowing to the firms and
individuals who recycle: if nothing need
be paid to dispose of waste, nothing is
saved by recycling instead of disposing.
This is a dominant theme in the lit-
erature on recycling and why it is insuf-
ficient. The argument runs through almost
every bit of general economic analysis
of recycling and is used in almost every
policy proposal for encouraging it. The
precise quantitative significance of
it is less certain. Certainly the quanti-
tative significance must vary greatly
from place to place and from one type
of paper product to another. The geograph-
ical variation is due to the variation in
the correct disposal price which would be
imposed if the imperfection were eliminated,
and also due to the variation in the cost
at which recycling can be substituted for
disposal (this depends on the density of
settlement, the network of supply and
the accessibility to markets, etc.).
The variation from product to product is
due to similar reasons: the disposal
cost and the ease of recycling varies.
The increase in price of disposal
would undoubtedly increase the effective
"price" of R (market price of waste
paper plus the disposal charge saved)
sufficiently, in many locations^ to
make households, firms and organizations
increase the supply of R. This increase
in supply would lower the market price
of R to users, .stimulating some additional
use. In a new equilibrium, the effective
"price" to households, etc. would be higher;
the market price to users would be lower;
the difference is the disposal cost. How-
ever again we have the problem that
inelastic supply and demand would probably
severely limit the increase in quantity
used.
156
-------
As so often in recycling, we find we
should expect considerable spatial varia-
tion in effects. Indeed, a major conclu-
sion from the research in the field,
although one not always made by authors,
is that effective research must move
beyond generalities for all paper, and
for the nation as a whole, and on to a
careful examination of individual regional
and metropolitan markets. In the case of
a disposal fee system, we would expect
the major increases in recycling, if
anywhere, to come in the large, densely
settled urban areas of the Northeast
and Midwest. Why? Because they are the
regions where few cities already charge
user charges, where many cities have a
shortage of land for sanitary landfill,
and where the density of population and the
presence of paper mills relying on waste
paper. Howeverj they are also regions
where a relatively large proportion of
paper is already recycled, which-may
limit the marginal response.
But there should be very great spatial
variation, even within broad regions. It
is imperative, in order to forecast mor-e
accurately the effects on recycling of
eliminating or reducing the imperfection
in disposal pricing, to do more disaggre-
gated—regionally—'analysis.
A rather important factor in the im-
plicit subsidization of disposal is that
many users of paper already pay a user
charge; these are the large business and
commercial and institutional users, who
routinely must pay for collection and
disposal even in cities where house-
holds do not. This point must be
remembered when considering policies to
eliminate or offset the imperfection;
for example, one cannot deduce from the
existence of subsidies for disposal to
households that a general subsidy for
recycling, extended to businesses as
well as households, is justified.
Or, at least, the subsidy should not
be at the same rate for businesses as
for households (3,36).
E^ Biases against use of R in large
vertically integrated paper product
firms. Suspicion of such biases is
based on a variety of hear-say evidence
but also on Cardin's careful study (lOa).
The point here is not simply that the
large paper firms do not use much R in
their operations, nor that their location
and technology makes it difficult to
use much R. That behavior could be
very consistent with sound business
practice and cost minimization, especially
in the short run, when the nature (includ-
ing location) of capacity is fixed.
Rather, what is suspected is that some
managers actually sacrifice profits in
order "discriminate" against R, and in
order to indulge their own prejudices
or to avoid having to make careful
calculations or take moderate risks.
Contributing to such biases are
said to be the lack of experience with R,
the geographical isolation of environ-
mentally damaging production facilities
from the offices of managers, the long-
standing organizational structure of the
firms, and social differences between
the typical manager and the typical
entrepreneur in the recycling business.
There has not been enough research
yet to enable the impartial observer to
reach a sound conclusion. In particular,
we do not know just how strong the biases
are; we don't know how to quantify them,
in the sense of knowing what price
differential between V and R is sufficient
to erode them. It is difficult to separate
biases from normal risk aversion, which
also explains reluctance to use R.
If the biases exist, they should be
most easily detected in the investment
decision by integrated firms. What
technologies, what capital in-
tensities, what locations are chosen when
capacity is expanded? Research on the
investment decisions of the firms has
not been plentiful; it has been slighted
relative to studies of price fluctu-
ations in waste paper markets, to studies
of costs, returns, and pollution in pro
forma, "ideal process" hypothetical case
studies.
157
-------
But it is in the investment decision
that managers of these firms do in fact
have real choices, and it is in data on
those choices that we should find evi-
dence of the trade-offs open to them,
as well as evidence on how they perceive
or mis-perceive the trade-offs.
As a final note on this, we note
that Cardin's work was squarely in a
modern tradition of organizational and
behavioral research on the large frim
(see, as a general example, the work of
Williamson (51)). Considerable empirical
and case-study research has already been
done, for example, on the effects of
the basic organizational structure of
the firm on its production and investment
decisions and on its performance. That
research can be built up for additional
work on the paper industry.
Imperfections Which Go the Other
Way. We would be remiss if we failed to
recognize some imperfections which have
the opposite effect from the ones we have
reviewed so far,, One is the failure of
capital used by municipal governments
to be required to return, in monetary
and/or imputed benefits, its full
social cost. The chief reason is the
Federal income tax exemption of
municipal bond interest; that allows
the budgetary cost of municipal capital
to be below the social opportunity cost.
In addition, we certainly should have
even less faith in municipal managers to
be "efficient" than we have in paper
industry managers. This could bias
society in favor of recycling,
ceteris paribus; while it probably has
limited relevance now, it is something to
worry about if we introduce a wide-
scale program to subsidize municipal
recycling, especially if we wanted to
subsidize municipal capital facilities
in recycling. Indeed, recognition of
this point makes us unenthusiastic about
subsidies of that kind. However, so
far the bias probably operates signifi-
cantly only in municipal "resource
recovery" facilities, in which paper is
generally recovered as energy or very
low grade fiber, rather than in the
forms we are discussing here.
Another possible imperfection is
monopsony power in waste paper markets.
We have come to think of them as very com-
petitive, in sharp distinction to the
oligopolistic markets for paper products.
But it is possible that some paper mills
using R, and some waste paper dealers,
have considerable buying power, especially
in small city markets. Recently a major
recycling paper mill executive commented
that in times of shortages, some dealers
tried hard not to raise the prices they
offered to the public and to maximize the
spread between their buying price (paid
to the public) and their selling price
(received from recycling paper mills).
He pointed out that the dealers' refusal
to raise "door" prices (paid to the public)
denied the supply effect that higher prices
would have had (44). To the trained
economist, this sounds rather like
monopsony. There has been surprisingly
little (none?) research on this possi-
bility, perhaps just because we have
been lulled into seeing a too-sharp
distinction between the structure of
the paper products industry and the
waste paper industry.
A NOTE ON
CONSUMER PREFERENCES
In all the preceding analysis,
there are several important assumptions
about consumer preferences and individual
welfare. First, preferences are given
and unchanging; in particular they are
not affected by any policies. Second,
individual welfare depends only on one's
own consumption, and not on any others.
Third, recycling per se is not valued
directly by individuals. If we abandon
one or more of these assumptions, we add
some other arguments for recycling.
While this is not a step normally taken
by economists, even those most fervently
in favor of recycling, there are some
definite possibilities we should explore.
They call for research on attitudes
and taste formation, rather than the
usual kind of economic research.
One might simply argue that recycling
is a "merit good," and that government
should provide it even
158
-------
though individuals don't want it, or
don't want it now. Maybe their tastes
will change with experience, and they
will value it directly or because it is
a "husbanding" of scarce resources.
Another approach, only a bit less
speculative, is to assume there are
interdependencies in consumption, so
that one person's consumption pattern
directly affects another's welfare. I
have in mind not interdependence due to
physical impacts, as in pollution or
congestion; that is already covered under
the rebric of externalities. Rather, I
have in mind psychological interdepen-
dence—one person simply cares what
another does. This is not usually
allowed for in the usual economic
analysis of recycling, or in anything
else, for that matter—economists
have as yet paid little attention to
such interdependencies. But in principle,
if they exist and private markets cannot
take account of them, we have an additional
imperfection which prevents markets from
achieving results consistent with
consumer sovereignty- The imperfection
would exist, of course, only if high
transactions costs inhibited indivi-
duals from effectively offering financial
inducements to each other to engage in
consumption which was mutually pleasing.
Government could then provide economies
in transactions.
This is still somewhat speculative.
I know of no research on the argument
which is soundly based on theory and
which offers testable hypotheses. How-
ever, there are some recent developments
in economic theory of interest. One
example is Thurow's notion of "individual
societal preferences," according to
which individuals care about others'
consumption of basic goods like education,
medical care, and justice (41). Person
A may feel pleased that Person B
recycles, not because he lives or works
near B and is affected physically
by it, but simply because he is pleased
that B "husbands" scarce resources (this
example is mine, not Thurow's).
Sen (39,40) has also applied the
notion of consumption interdependencies
(or consumption externalities) to
justify collective acction, in this
case national saving. Although the whole
subject is a bit new, there is in
principle nothing objectionable about it.
In a welfare economics fundamentally based
on the notion of individual welfare, we
should accept that individuals' welfare
can depend not only on their consumption
but on others consumption as well. In
practice, of course, it is a social
decision on just what interdepencencies
to recognize in public policy. We might
rule out some as not socially acceptable;
for example, we might rule out mere envy,
or ones which imply racial or sexual
discrimination, or a disutility caused by
a neighbor's painting his house salmon
pink.
In order for this to have any legit-
imacy in recycling, I think we would have
to show the feelings are widespread.
Whether or not they are is highly
debatable, but research is needed. It is
certainly quite clear that a lot of people
do not like to bundle up old newspapers,
and care not a whit whether you or I do
either. They are not bad people; they
may have other attitudes which make them
care deeply whether you or I have good
housing or medical care, or are provided
basic justice, but they simply don't care
about recycling. Some of them may care
about "husbanding" non-renewable resources,
but may not care about paper because it is
a renewable resource; that would be quite
legitimate.
To conclude this review of imperfec-
tions, we note that the approach to the
problem leads to proposed policies and a
particular view of them. There are two
sets of policies: one to reduce or end the
imperfections; the other is to intro-
duce new imperfections, which perhaps
paradoxically., increase the national
income by offsetting the other ones.
It is generally true that the best
approach, which would increase national
income the most, is to eliminate
the original imperfections. Unfortunately,
that is
159
-------
strictly true only if all the imperfec-
tions can be eliminated. If a policy is
constrained—politically or legally—to
tolerate the continued existence of one
or more imperfections, then it may turn
out that adding a new, offsetting one,
is the "second-best" policy—it will
increase national income somewhat, but
not as much as the first-best policy.
Clearly, policymakers certainly cannot
eliminate all imperfections in society
which affect recycling in one way or
another. They must proceed in a more
piecemeal fashon. Therefore, a second-
best analysis is required routinely. It
is useful, therefore, to review secbnd-
best theory and attempt some conclusions
from it.
SECOND-BEST THEORY
A "second-best" problem is one in
which there are imperfections, and a
policy is to eliminate some of them, but
not all. A constraint on the policy is
that some imperfections remain. For
example, we may have pollution exter-
nalities in the production of A, but
we also have a monopoly in A. Our "first-
best" policy is both to correct for exter-
nalities (as with an effluent tax, or
the optimal amount of regualation) and
to eliminate the monopoly. But suppose
the monopoly must remain—it is a politi-
cal fact of life, or a legal requirement,
or whatever. If so, the "second-best"
analysis is required to determine just
what should be done to reduce the exter-
nalities. In general, it is not always
the case that a policy called for by
first-best analysis is called for in
second-best. In our example, a first-
best policy includes effluent taxation,
the rate set at the marginal damages.
In a second-best case, it is easy to see
that it is not necessarily good to tax
the firm's effluent, even at the rate
which would be optimal in the first-best
case. The reason is that the monopoly
restricts the output of A below the
socially optimal level, but the failure
to tax the effluent artifically expands
the output. A tax on effluent will
restrict the output by even more than
the monopoly restricts it, so it will
accentuate the distortion in resource
allocation from the monopoly. We can't
know j3 priori whether that is an improve-
ment unless we know just what increase
in evironmental quality will occur as an
offset to the reduction in output of A.
The balance might go either way: the
reduction in pollution might be greater
than, or less than, enough to offset the
loss of output of A. Put another way,
in the original situation, the overproduc-
tion from the failure to tax effluent
might have just offset the underproduction
from the monopoly, leaving the output of
A about right, (assuming, as always, that
the monopoly had to remain). This case
is discussed in (4). Just how much to
tax the effluent, if at all, cannot be
determined without knowledge of all the
elasticities, the cost functions, and the
marginal damage function, and without
calculations trading off gains versus
losseso The only safe conclusion is that
no simple rule can be laid down (4,16,19,
21).
One is not caught helpless, however.
Mishan (26,28) has suggested some cases
in which certain policies are highly like-
ly to produce increases in national
income, and on his analysis we can build
applications relevant to recycling
policy. Consider the following hypo-
thetical situations, in which we analyze
imperfections and policies affecting
goods A and B. Assume also that the
imperfections and policies affect
only A and B, and no other goods, or
at least predominately so. This may be
the case where A and B are close substi-
tutes for each other, and neither is a
close substitute or complement for
any other good, in either consumption
or production.
Case I. Two imperfections have
roughly offsetting effects on A; for
example, monopoly and uncorrected
pollution in A. This is the example dis-
cussed earlier; the monopoly restricts
A (and expands B), while the uncorrected
pollution expands A (and restricts B),
relative to the optimum. The second-
best policy would include
160
-------
a tax on effluent and a subsidy to firm
A (based on the level of output, and
equal, per unit of output, to the differ-
ence between marginal cost and marginal
revenue at the optimal level of cutout)
(4).
Case II. Imperfections in both A
and B, which tend to offset each other.
For example, monopoly in A, excise tax in
B. The monopoly restricts A, expands B;
the excise tax restricts B and expands A.
The two may offset each other, leaving
the two outputs closer to the optimal
levels than if one imperfection were
eliminated completely or reduced greatly.
This is especially true if A and B are
very close substitutes.
Case III. Imperfections in A, which
accentuate each other. For example, a
subsidy to A and uncorrected pollution in
A. Both these artificially expand A
and restrict B. Now, eliminating only one
of the imperfections would partially
reverse things, and would very likely to
be a'Change for the better. In practice,
we would still have to be careful about
changes resulting in other goods and
services, which we have ruled out in
these examples.
Case IV. Imperfections in A and B,
which accentuate each other. An example
is a monopoly in A and a subsidy in B,
another would be uncorrectd in pollution
in A and an excise tax on B. Here, too,
elimination of just one of the imperfec-
tions in the pair would very likely
improve things.
Now, let us return to Case III,
and assume that we cannot eliminate either
of the distortions. However, it is likely
that we can improve things if we convert
the case into something like Case II,
by adding a new imperfection which off-
sets the combined effect of the two in
Case III. The two imperfections in
A both overexpand A; what is a possible
offset is a subsidy to B, which expands
it. It will be readily seen that this is
analagous to the situation in V and R.
In V, we have several imperfections (tax
advantages, uncorrected pollution,
subsidy through solid waste disposal), all
of which move productions of V in the same
direction—toward above-optimal amounts
and all of which move production of R in
the same direction—toward sub-optimal
amounts. A subsidy to R would help off-
set them.
While the subsidy to R may—probably
will—improve welfare, it can be shown that
the new imperfection is only a second-
best policy, at best. It cannot completely
reverse the losses imposed on society by
the imperfections in R, as would a first-
best policy of eliminating all the imperfec-
tions in V. For one thing, introducing
a subisdy on a product requires, for budge-
tary balance, an increase in government
taxation somewhere in the economy, and
that introduces new imperfections somewhere
else, which may create new losses, and
which dilute the benefit of offsetting
the orignial imperfections in V. The
probability of this can be reduced by
imposing the new taxes in broad-based
way, rather than concentrating them on one
or a few goods.
In addition, the subsidy to R. will
not completely offset the original losses
because it will contribute to an expansion
of the paper products industry beyond the
socially optimal amount. While it corrects
for the overuse of virgin material at the
expense of recycled material, it also low-
ers the price of the final product, and
stimulates an increase in quantity which
has a marginal cost ot the economy greater
than the marginal valuation by consumers.
This basic point is often missed in econom-
ic analyses of recycling, but it is made
by Anderson (3).
Figure 1 illustrates some of the
simple aspects of subsidy to R. The
numbers are hypothetical and are only
examples. The diagram refers to paper
products; D is the demand curve. There
are constant costs; if V is used as
a raw material, the marginal cost and
price, MCV and P , are both $100; if R
is used, MCR and PR are both $90.
161
-------
In the initial situation, there are
no subsidies, so only R is used and P^ =
90; quantity is 100.
Now, assume a subsidy is given to V,
translating to a price advantage of $20
per ton of product; price now falls to
PV , which is $80. Now only V is used,
to the total exclusion of R, and quantity
rises to 110.
Compared to the original situation,
the economy has suffered two kinds of
welfare loss. First, a more expensive
input has replaced a cheaper one on the
original quantity; second, the industry
has overexpanded, with the additional
output having a marginal cost of $100
but a marginal value to consumers of less
than that. The first loss is area abed,
equal to $1000; the second !area is bcfe,
equal to $150 (this is also bghe, total
cost of added output, minus cghf, total
value to consumers of added output).
Thus, total loss is $1150.
Alternatively, we can derive the
$1150 loss as the added cost of producing
the total supply of paper, which is Oaeh
($11,000) minus Odcg ($9,000), less the
FIGURE 1
Cost,
Price
100 a
90 d
80
70
I
i
gl
added value to consumers from increasing
the total supply of paper, which is gcfh
($850). That is $1,150.
Now, assume a subsidy is introduced
on R, translating to a price reduction of
$10; price of paper made with R now falls
to P^ = 80, which is equal to P' Assume
that because the priced are equal, each
gets half the market, which remains at 110.
The average cost of paper products is now
$95 (average of MCy = 100 and MC = 90) ,
so the total costs of producing the paper
which is sold is $10,450, an increase of
$1,450 from the initial position (P=90,
Q=100). The value of the added paper to
consumers is only $850, as in the previous
case, so the loss is now $600.
If at equal prices, only R was used in
the manufacture of paper products, the loss
would be almost eliminated, but not quite.
The cost on the original quantity (100)
would be again 9000 as in the very first
situation, so there are no losses from use
of higher price input instead of lower
price input. However, there is an expan-
sion of the industry of 10 units; they have
a marginal cost of $90, or a total of $900,
while as we have seen the added value to
consumers is only $850. So there is still
a loss of $50. Even though the subsidy on
R is only enough to bring R to par with V,
it does not eliminate the
MC..
MCR = PR
.PV =PR
100 110 120
Quantity
162
-------
welfare loss due to overexpansion which
the original subsidy to V caused.
What would happen if the subsidy to
R were made "too large," that is larger
than the $10 needed to bring R to exact
par with V? Assume the subsidy were
mistakenly made $20, lowering price to R
to P^ =70. The quantity would now
rise to 120, and the loss would increase
(that is, the loss from the very first
position with no subsidies at all). The
added cost to the economy would be 20
units of product times $90 per unit
(remember the MCR is $90), or $1800;
the value to consumers (area under demand
curve) would be only $1600 so the loss
would be $200. .This illustrates the fact
that a subsidy to recycle should be no
larger than the amount needed to off-
set the original imperfection, and even
then there will still be a small welfare
loss compared to a "first-best" position
(the welfare loss in this case will be
larger, of course, the greater is the
price elasticity of demand).
This whole analysis assumes that the
subsidy to R is such that there are no
further distortions in the production of
paper products due to biased input prices.
In particular, the subsidy will have
additional welfare losses if it is based
on capital used in recycling, rather
than on the use of recycled material
per se. This possiblility is mentioned
becaused subsidies to capital are polit-
ically popular, taking the form of
construction grants, low-interest loans,
etc.
INCOME DISTRIBUTION EFFECTS
I promised earlier to deal with
these. I think there are two things to
worry about. One is that subsidies
should be designed to avoid large wind-
fall gaines to recyclers who would
have recycled even without the subsidy.
Improperly designed subsidy schemes
without marginal incentives will have
a rather low increase in recycled paper
per dollar of tax revenue used to pay the
subsidy; that both has an unfortunate
distribution effect and also requires
more distortions elsewhere in the economy
(due to taxes) per marginal ton of recycl-
ing.
The second thing to worry about is
the distributional impacts if we eliminate'
the imperfection due to failure to price
solid waste disposal. Frankly, I don't
think even these impacts are very impor-
tant; they are small amounts of money
and the disadvantage in vertical equity
is partially offset by an increase in
horizontal equity resulting from a fairer,
cost-based price system. But they may be
politically important, and in fact may be
one of the most important reasons why we
move toward "second-best" policy of
subsidizing recycling rather than a first-
best policy including solid waste user
charges. For extended treatment, see (6a).
CONCLUSIONS FROM PREVIOUS
RESEARCH AND RESEARCH AGENDA
The policymaker needs research in
order to guide policymaking. But
financing research—or doing it in-house—
is also a policy. The research on recycl-
ing paper must be guided by the need to
allocate scarce research funds effieiently.
Here I wish to sketch out some suggestions
for further research. Again, I shall
follow my list of major imperfections
affecting recycling paper.
A. Pollution. The relative gap
between V and R is presumably being reduced
grandually. Eventually, it may be elimi-
nated. If so, the pollution question will
become moot, and it is not a wise invest-
ment to do detailed research on relative
pollution loads in V and R. On the other
hand, if the differences are going to
persist, we need much more specific
research than we have had so far—
specific on different locations and the
assimilative capacity of the environ-
mental media in them, specific on the
differences in enforcement, on the lags
in behavioral responses of managers, the
effects of varying ages of plants, etc.
163
-------
B. Taxation. The research we have
so far is very good. It indicates that
the taxation effects are not very big, in
paper. The effects even in the long run,
although uncertain in size, will probably
not have much present value; the effects
will be delayed because the effects on V
will be delayed and because the response
(in the form of investment) of paper firms
to changes in V will be delayed. The pay-
off to eliminating the tax advantages
doesn't seem very big, so the payoff
to further research on the subject doesn't
either. Another factor is that more
research will hardly make a difference
in the political decision; the political
economy is against the change. Removal
of the tax advantages would have a more
concentrated negative effect on landowners
and paper firms but it would have a
diffused positive effect on consumers of
paper products in which recycled material
could be easily substituted. Many consum-
ers of timber products and of paper pro-
ducts in which a lot of V must be used
will of course be hurt.
I may be wrong on the last point.
For many years, we didn't feel the oil
depletion allowances could be eliminated
either, but finally they were. However,
it took a profound energy crisis, which we
are not likely to see in paper.
C. Solid Waste Disposal. Research
on the effects of improper disposal
pricing is one of the highest priorities,
I think. Fortunately, there is a lot
being done, on the effects of a product
charge, of moves toward pricing by
municipalities, etc.
This kind of research is important
for several reasons. First, the effect
on the realtive prices of V and R may
be very high. For some users, the
present system effectively reduces the
price of disposal to zero, which is a
large change. Second, there seems to me
to be a genuine chance that good
research will lead to change. I am not
too hopeful on the product charge; it
seems simply to be too big a jump into
uncharted areas of Federal policy. But
research on municipal pricing would
probably increase the chances of reducing
the present implicit subsidy gradually
in the long run. I believe that if the
effects of municipal pricing are better
understood, it will be adopted more widely.
Cities will see how limited the effects
are on the distribution of income, and
will gain experience on pricing methods.
Many cities are ready for reassurance that
pricing is a legitimate way to resolve
mounting solid waste problems, which on
account of increasing labor and disposal
space costs are becoming more serious.
Cities are also useful groupings of
people in which to capture the consumption
interdependencies I discussed earlier,
which may ease the way to introducing
charges. Furthermore, it is not the case,
as in pollution, that other government
policies are already gradually removing
the imperfection; there is a move to
reduce externalities from open dumps,
but they are only a small part of the
implicit subsidy.
The recent popular movement
("propulism," it has been called) to
reduce property taxes also makes a move
to user charges more feasible, so that
research on the details of execution and
of impact is very valuable.
In addition, there are two gen-
uine contenders for pricing policies:
municipal user charges and the national
product charge. And of course they com-
pete with recycling subsidies as candi-
dates for adoption. There is still
much to be learned about which of
these policies is the best.
I would favor increased municipal
user charges. There are advantages of
efficiency and equity. Charges can be
varied from location to location, which
should contribute to efficiency and to
horizontal equity. They would help
educate people to husband scarce
resources more carefully and thus improve
the internalization of consumption inter-
dependencies. The product charge,
on the other hand, seems to rigid and
too remote from individuals to have the
same advantages. It cannot allow for
164
-------
wide variations in social cost of dispo-
sal, and cannot easily allow for the fact
that many business and commercial users
already pay user charges; levying the
product charge on paper used by them
would indroduce a new imperfection which
would not offset any existing ones.
It would have some of the same adverse
effects on vertical equity (income
distribution) as municipal charges, but
would not have the advantages of increas-
ing horizontal equity—which requires
individuals imposing different costs on
the solid waste managment system to pay
different charges. The product charge
would apparently give a rebate to users
of recycled material, which is a help,
but it would still levy the same charge on
all persons who disposed of waste, no
matter how the social cost of handling
their waste varied.
A recycling subsidy is less desir-
able than municipal charges, because of
the minor distorting effect described
above, because any practical scheme would
probably dispense windfall gains to a few
even as it collected taxes from the gener-
al public, and because in practice it
would be hard to withhold it from users
who already pay user charges for solid
waste disposal. As we saw earlier,
the subsidy properly should not be
available to persons and firms who al-
ready have the appropriate incentives for
recycling.
Many of these things are contro-
versial, and the effects—on distribution
and on degree of recycling and on solid
waste disposal costs—are speculative.
But this is probably the most important
single imperfection in our system, and
the times are ripe for change, so
resolving these questions is the
highest priority in research.
D. Uncertainty in Waste Paper
Markets. The theoretical effect of
uncertainty, caused primarily by the
marked fluctuations (and unpredict-
ability of fluctuations) in the market
price, are well known. One must assume
they might be very significant. But
what is surprising is how little
research has been done on the quanti-
tive significance. Most of the liter-
ature so far merely repeats, with some
embellishment, the standard theory on
uncertainty and investment, and moves
from it to a conclusion that some
public policy is needed to offset the
uncertainty. In some cases, there is
a too-easy jump to the conclusion
that subsidies^ to recycling are needed.
We need to move beyond this stage
and try to quantify the amount of risk
aversion. This may require far more
than routine econometric analysis of
existing data, for we don't have data
for periods when the risk and uncertainty
were lower. We can hardly analyze the
effects of reducing uncertainty by
analyzing data from a period when the
uncertainty has not been reduced.
Perhaps here, as in the case of
research on solid waste user charges as
well, we need some "social experiments,"
similar to the ones on the negative
income tax, housing allowances, etc.
At any rate, we need to probe the
attitudes of participants in recycling
processes, not merely business firms but
also individual households and non-profit
organizations. Particular attention
should be paid to investment decisions
of the large integrated paper products
firms, for research on that also
helps us understand the degree of
"biases" on their part, which is another
important research subject. And, of
course, the attitudes of municipal
officials must be studied. They will be
making some of the most important deci-
sions on recycling, including investment
decisions. Are they unusually averse to
risk, even though they are not employed
by profit-making enterprises? Does it
make a difference how recycling is organ-
ized in municipal government, that is,
does it make a difference whether it is
part of the regular public works func-
tion or a separate, somewhat autonomous
and free-standing public corporation?
How does the risk from recycling corre-
late with other risks in municipal
government—what is the "portfolio
effect" of adding recycling to the
165
-------
municipal managers' package of respon-
sibilities, which has various risks of
financial and political kinds? One can
think of many other questions in this area
which have not been studied much, but which
seem important to answer if we are to
evaluate recycling policies.
Another example is the effects of a
buffer stock stabilization scheme, which
might be operated by a local or higher
level government. Even if municipal
officials don't try to earn profits in such
a scheme, the budget surpluses or deficits
that result will certainly affect their
attitudes and the attitudes of voters.
Yet it is hard to generalize, as Pearce has
pointed out (32,43), on whether a buffer
stock wi 11 make surpluses or deficits,
even if it is clear that it will or will
not produce social benefits.
Even if we learn more about the nature
of uncertainty and risk aversion, the appro-
priate policies are not completely clear.
It is not at all clear, for example, that
an outright subsidy is appropriate. This
point comes out of the theoretical liter-
ature on cost-benefit analysis and public
investment theory (3a). A subsidy is not
appropriate if it does nothing to reduce the
real social cost of risk and uncertainty,
but merely redistributes it. This seems
to be relevant: the risk and uncertainty
of excessive price fluctuations in waste
paper is not only a private risk but is a
social risk. The basic structure of markets
produces uncertainty for the society as a
whole; while a subsidy might .reduce the
part of it which rests on business firms
in the industry, it would not reduce the
total risk. The more appropriate policy
would seem to provide insurance of some
kind, so that spreading the risks over
many firms would reduce the perceived cost
to each one. Even if the fluctuations
are not reduced per se, an insurance
scheme could reduce their real cost to
firms by eliminating the possibility of
concentrated losses to any one firm.
The same sort of thing might be accomplished
by makeing public investments in recycling
facilities, and spreading the costs over
so many taxpayers that there is no real
cost to any of them (4a). Or, a private
insurance scheme in the form of organized
futures markets might be set up (2). All of
these possibilities require further study;
a high research priority is thus how to
translate the theoretical conclusion on the
effects of risk on investment into practi-
cal policies.
E. Biases on Part of Paper Firm
Firm Managers. As said earlier, we
are still unsure whether the biases
reported by Cardin (lOa) and others
are really there, and how important they
are. We don't know how separate they are
from normal risk aversion; that is
why work on investment decisions is
so important, for it will shed light
both on this subject and on the subject
covered just above. We don't know
how long the biases would stand up
before an aggressive policy, to
subsidize recycling or to rationalize
solid waste disposal pricing.
It is important to know more about
these things, not because those who
make policy on recycling can have any
hope of "reforming" the industry,
even assuming for the sake of argument
that it needs reforming, but simply
because the effectiveness of other
policies depends crucially on the response
of the industry. The degree of response,
the speed of response, the spatial
variation in response—all of these are
important. If we were to find that
there are biases in the industry,
recycling policy presumably must accept
their continued existence at least
for awhile. Policy must be shaped
to have an effective impact under the
constraint of the biases; here we have
still another example of second-best
policy and framing it intelligently
requires a lot of second-best analysis.
Let me make a final point. Much
of our previous research, useful as
it has been—has focused on one imperfec-
tion at a time. We need to move toward
research in a wider context, analyzing
the removal of more than one imperfec-
tion or offsetting more than one at the
same time. What, for example, is likely
to be the effect of simultaneously
changing pollution controls in paper
manufacturing and introducing solid
waste disposal charges in municipalities?
166
-------
Of simultaneously introducing charges and
more subsidies for demonstration,
education, and "experiment" projects?
Of simultaneously introducing a futures
market in waste paper and user charges?
Our research has provided very notable
insights into effects of individual
imperfections, and has made fairly clear
the direction—if not necessarily
the magnitude—of effects on recycling
in each case. We still have a way to
go in exploring the effects of imper-
fections, and policies, considered
jointly.
ACKNOWLEDGEMENT S
Some of the research behind this
paper was performed under contract
with the Solid and Hazardous Waste
Research Division, Municipal Environ-
mental Research Laboratory, U.S.
Environmental Protection Agency. I am
especially grateful to Oscal Albrecht,
Haynes Goddard, Fred Smith, Robert C.
Anderson, and Ralph Bradburd for comments
at various stages of the work. I am
solely responsible for any errors of
fact or interpretation.
REFERENCES
1. Anderson, Robert C. and Spiegel-
man, Impact of the Federal Tax
Code on Resource Recovery, U.S.
E.P.A., Washington, 1976 (NTIS)
2. Anderson, Robert C. and Dower,
Roger, An Analysis of Scrap
Futures Markets for Stimulating
Resource Recovery, U.S. E.P.A.,
Cincinnati, 1978 (NTIS)
3. Anderson, Robert C., "Recycling
Policy: Basic Economic Issues,"
in Pearch and Walter (33).
3a. Arrow, Kenneth and Lind, R.,
"Uncertainty and the Evaluation
of Public Investment Decisions,"
American Economic Review, 60:
364-78 (1970)
4. Baumol, William and Gates, Wallace,
The Theory of Environmental Policy.
Prentice-Hall, Englewood Cliffs,
N.J., 1975, pp. 71-3, relying
partly on Buchanan, James,
"External Diseconomies.Corrective
Taxes, and Market Structure,"
American Economic Review, 59:174-7
(1969).~
5. Bingham, Tayler, "Allocative and
Distributive Effects of a Disposal
Charge on Product Packaging,"
in Pearce and Walter (33).
6. Bingham, Tayler, e* al., An
Evaluation of the Effectiveness
and Costs of Regulatory and Fis-
cal Policy Instruments on Product
Packaging, Research Triangle In-
stitute, Research Triangle Park,
N.-C., 1974.
7. Bower, Blair et al., "Residuals
Management in the Pulp and Paper
Indus try,: Natural Resources
Journal, 11:605-23 (1972).
8. Bower, Blair, "Studies of Re-
siduals Management in Industry,"
pp. 275-320, and MacAvory, Paul,
"Comment,"" pp. 321-4, in Mills,
Edwin, ed., Economic Analysis
of Environmental Problems,
Columbia University Press for
National Bureau of Economic Re-
search, New York, 1975.
9. Bower, Blair, "Economic Dimensions
of Waste Recycling and Re-use,"
in Pearce and Walter (33).
10. Butlin, John, "The Prices of
Secondary Materials and Recycling
Effort," in Pearce and Walter (33).
167
-------
lOa. Cardin, Frederick, Secondary
Fiber Recovery Incentive Analy-
sis, U.S. E.P.A., Washington,
1974.
11. Darby, Michael, "Paper Recy-
cling and the Stock of Trees,"
Journal of Political Economy,
81:1253-5 (1973).
12. Deadman, D., "Modelling the U.K.
Waste Paper Market, "Resource
Recovery and Conservation,
3: 237-48 (1978).
13. Dorfman, Robert and Dorfman,
Nancy, "Introduction," in Dorf-
man and Dorfman, eds., Econo-
mics of the Environment, W.W.
Norton, New York, 1972.
14. Environmental Studies Board,
National Academy of Sciences/
National Academy of Engineering,
Man, Materials, and Environment,
MIT Press, Cambridge, 1973
15. Franklin, William £t al., Paper
Recycling: The Art of the Pos-
sible, 1970-1985, Solid Waste
Council of the Paper Industry,
Kansas City, Mo., 1973.
16. Henderson, James and Quandt,
Richard, Microeconomic Theory,
second ed., McGraw-Hill, New
York, 1971.
17. Hicks, James, "The Foundations
of Welfare Economics, Economic
Journal, 1939, p. 696.
18. Kaldor, Nicholas, "Welfare Pro-
positions in Economics," Eco-
nomic Journal. 1939, p. 549.
19. Layard, Richard, and Walter,
Alan, Microeconomic Theory,
McGraw-Hill, New York, 1978.
20. Lingle, Stephen, Recycled
Materials Markets: February
1975, U.S. E.P.A., Cincinnati,
1975.
21. Lipsey, Richard and Lancaster,
Kelvin, "The General Theory of
Second Best," Review of Economic
Studies, 24: 11-32 (1956).
22. Little, Ian, A Critique of Wei-
Economics , second ed., Claren-
don Press, Oxford, 1957
23. Miedema, Allen, Market and Net
Welfare Effects of Secondary
Materials-Biased Input Market
Distortions, Research Triangle
Institute, Research Triangle
Park, N.C., 1975.
24. Miedema, Allen, et aL., The
Case for Virgin Material
Charges; A Theoretical and Em-
pirical Evaluation in the Paper
Industry, Research Triangle In-
stitute, Research Triangle Park,
N.C., 1976
25. Mills, Edwin, The Economics of
Environmental Quality, W.W.
Norton, New York, 1978.
26. Mishan, Ezra, "Second Thoughts
on Second Best," Oxford Economic
Papers, N.S. 14: 205-17 (1962).
27. Mishan, Ezra, "A Survey of Wel-
fare Economics," Economic Jour-
nal, 70:197-256 (1960).
28. Mishan, Ezra, Cost-Benefit
Analysis, second ed., Praeger,
New York, 1976.
29. Nathan, Thomas, "Mining Urban
Ore," EPA Journal, February
1979, 24-6, 40.
30. Page, Talbot, Conservation and
Economic Efficiency, Johns Hop-
kins U.P. for Resources for the
Future, Baltimore, 1977.
31. Pearce, David, Environmental
Economics, Longman, London, 1976.
32. Pearch, David and Grace, R.,
"Stabilising Secondary Mate-
rials Markets: The Case of
Waste Paper," Resources Policy,
June 1976.
33. Pearce, David and Walter, Ingo,
eds, Resource Conservation:
Social and Economic Dimensions
of Recycling. New York U.P.,
New York, 1977.
168
-------
34. Plaut, Thomas, An Econometric
Analysis of Regional Wastepaper
Markets, Regional Science Re-
search Institute, Philadelphia
1978 (Discussion Paper No. 104).
35. Plaut, Thomas and Steiker, Gene,
Characteristics of Wastepaper
Markets ..., Regional Science
Research Institute, Philadel-
phia, 1978 (Discussion Paper
No. 103).
36. Purcell, A., and Smith, F.,
"Energy and Environmental Im-
pacts of Materials Alternatives,"
Resource Recovery and Conserva-
tion, 2: 93-102 (1976).
37. Resource Conservation Committee,
Transcript of Hearing on Solid
Waste Product Charge Issue,
Nov. 17, 1977, U.S.E.P.A.,
Washington.
38. Scitovsky, Tibor, "A Note on
Welfare Propositions in Econo-
mics," Review of Economic
Studies, 9: 77 (1941).
39. Sen, A.K., "On Optimizing the
Rate of Saving," Economic
Journal, 71:479-96 (1961).
40. Sen, A.K., "Isolation, Assur-
ance and the Social Rate of
Discount," Quarterly Journal
of Economics, 81:112-24 (1967).
***
41. Thurow, Lester, "Cash versus
In-Kind Transfers," American
Economic Review, May 1974,
190-95 (vol. 64, No. 2). ***
42. Tillman, D., "Comparison of Two
Government Reports as to their
Approaches to Recycling,"
Resource Recovery and Conser-
vation, 2: 361-4 (1977).
43. Turner, R., Grace, R., and
Pearce, D., "The Economics of
Waste Paper Recycling," in
Pearce and Walter (33).
44. U.S. Citizens Advisory Committee
on Environmental Quality,
A New Look at Recycling, U.S.
G.P.O., 1976.
45. U.S. Council on Environmental
Quality, Resource Recovery;
The State of Technology.
Washington, 1973
46-49. U.S. Environmental Protec-
tion Agency, First, Second,
Third, Fourth (respectively)
Reports to Congress on Resource
Recovery and Source Reduction.
U.S. G.P.O., Washington, 1974,
1975, 1976, and 1977.
50. Whitcomb, David, Externalities
and Welfare, Columbia U. P.,
New York, 1972.
51. Williamson, Oliver, "Managerial
Discretion, Organization Form,
and the Multi-Division Hypo-
thesis," in Robin Marris and
D. Wood, eds., The Corporate
Economy, Harvard U.P., Cam-
bridge, 1971.
52. Wisecarver, Daniel, "The Social
Costs of Input-Market Distor-
tions," American Economic
Review, 64: 359-72 (1974).
***See also McPherson, Michael,
"On the Theory of Endogenous
Preference Change," unpublished
paper, Williams College, 1978.
169
-------
ECONOMIC AND INSTITUTIONAL IMPEDIMENTS
TO CONSERVATION AND RECYCLING
Robert C. Anderson
Environmental Law Institute
1346 Connecticut Ave., N.W.
Washington, B.C. 20036
ABSTRACT
For a variety of reasons it has been suggested that the United States should act
to conserve resources and increase the domestic rate of resource recovery. These
suggestions are buttressed by evidence that in some instances the forces of the free
market do not provide for socially desirable levels of conservation and recycling.
The critical issues for public policy are (1) whether the government should inter-
vene in resource markets and (2) if so, what form the intervention should take.
Examination of these points reveals that existing research and information is not yet
sufficient to provide a conclusive recommendation. One cannot on the basis of existing
information rule out the possibility that the best policy is merely to continue the
status quo.
INTRODUCTION
A widespread perception exists that
this nation is wasteful in its use of
materials and that such wasteful behavior
must be regulated. The perception which
dates at least to the Conservation Move-
ment of the late Nineteenth Century is
underscored by recent calls for the U.S.
to establish a materials policy and to
enact specific programs and policies to
encourage conservation and recycling.'-'-)
With the passage of the Resource Conser-
vation and Recovery Act of 1976 (RCRA),
it became official governmental policy to
"promote the protection of health and the
environment and to conserve valuable ma-
terial and energy resources."(2) Conser-
vation and recycling are to be encouraged
through technical and financial assist-
ance to state and local governments for
new and improved methods of collection,
separation, and recovery of solid waste
as well as through new guidelines for
collection, treatment, separation, recov-
ery and disposal practices, and through
programs of research and development.
The need for governmental inter-
vention to stimulate resource conservation
was argued in years past through reference
to wasteful practices in natural resource
development and material use. As informed
observers are quick to point out, patterns
of resource development and conservation
are governed by market prices and that
what appears to be wasteful behavior
often is not. For example, if natural
gas is flared in an oil field, it's because
it is not profitable to capture and market
the gas. If operators were forced to mar-
ket the gas, society would sacrifice goods
and services having a higher value than the
gas which is captured. Indeed, Hotelling
argued that private market forces general-
ly provide for an optimal amount of re-
source conservation.
Although the same observations con-
cerning alleged wasteful behavior are
still made, a more sophisticated rationale
for actions to promote conservation and
recycling has been advanced in recent
years. This rationale is based upon
failures of the market system to provide
in every instance the correct signals to
those who develop resources and consume
materials. A partial list of the poten-
tially more significant sources of market
failure includes institutional forces such
170
-------
affected materially by subsidies given to
U.S. producers. Previous studies have made
no attempt to estimate the impact of tax
subsidies on the quantity of virgin miner-
als which are produced. The quantity effect
is virtually certain to be larger than the
price effect, again because of the import-
ance of imports in domestic mineral con-
sumption. With only slight changes in the
U.S. producer price of a mineral, large
shifts will occur in the percentage of the
U.S. market which is supplied by domestic
producers.
The impacts on recycling created by
the subsidization of primary mineral pro-
duction were the focus of a major research
project for EPA. (7) in that study we found
the probable quantity impacts on recycling
to be quite small, at most 1 or 2 percent.
Although the approach taken in our econo-
metric modeling of competition between
primary and secondary production tended
to understate substitution possibilities,
the price effects for primary minerals
was probably overstated, implying that
the estimated quanitity impacts may yet
be reasonably accurate.
The tax treatment of standing timber
confers a significant subsidy to the timber
industry. Actually timber producers qual-
ify for two subsidies: immediate expensing
of most costs of growing timber, and cap-
ital gains treatment of the resultant gain
in value upon harvest. Because of the lib-
eral provision for e^ensive growing costs,
most of the value of standing timber
is treated as a capital gain. The maximum
long-run price effect of these subsidies
on the price of standing timber has been
estimated to range from 35 to 45%.W The
impacts are this large principally because
the U.S. is an exporter of timber, and
therefore 'domestic timber prices are de-
termined by the costs of domestic produc-
tion rather than those of world production.
Despite large price effects on tim-
ber, the impacts of the timber industry
subsidies on the recycling of wastepaper
are not likely to be large for a variety
of reasons. First, the price of standing
timber accounts for only about 12 percent
of the cost of woodpulp, and it is wood-
pulp, not standing timber, which substi-
tutes for recycled wastepaper. Second,
the supply of wastepaper does not appear to
be very responsive to changes in price,
having an estimated price-elasticity of
approximately .5.(9) Thlrdj the subst±_
tution between woodpulp and wastepaper is
far from perfect, at least in the short
run as is evidenced historically by vio-
lent swings in the relative price of the
two goods.
The subsidization of primary produc-
tion, when considered in isolation of
other market forces, clearly reduces over-
all economic efficiency. Investments in
virgin material production are more
lightly taxed than are their counterparts
in recycling. As a consequence, the
amount of investment in primary production
is excessive and in recycling it is inad-
equate. As we have seen, the probable
impacts on the quantity of material re-
cycled are relatively minor. The question
is whether the resulting gains in efficien-
cy would repay efforts to eliminate tax
subsidies to primary producers. This ques-
tion has not been investigated in previous
research, and consequently my remarks here
will be brief.
It appears that on the grounds of
administrative costs alone, a case could
be made for removing the special tax sub-
sidies for primary producers. Administer-
ing a complex system of subsidies, such
as is now in place, imposes auditing costs
for compliance and legal costs to adjudi-
cate special situations. These situations
inevitably arise as investors seek to
maximize their tax subsidies.
Removal of the special subsidies for
primary material producers would pose some
difficult problems of equity. Many> if not
most. °f the mining and timber firms have
acquired properties or have made major
investments in company-owned properties
after the tax subsidies were in place. In
making new investments or adding to exist-
ing investments, these companies have
computed an expected rate of return which
includes the subsidy. In fact, the long-
run, after-tax rates of return in mining
and timber operations are much the same
as in general manufacturing or in service
industries. If the special subsidies in
mining and forestry were eliminated, the
existing operations would produce lower
rates of return on existing investments.
In other words, the market value of the
investments would fall until the new post-
tax rates of return were comparable to
those offered by equally risky investments
in other sectors of the economy.
171
-------
as tax subsidies to virgin material pro-
duction, vertical integration in primary
production, labeling requirements for re-
cycled goods, and the regulation of
freight rates at inappropriate levels;
externalities such as the failure to price
correctly the disposal of post-consumer
waste; and informational problems such as
incorrect perceptions as to the quality
of recycled materials.
In their general impact, all such
sources of market failure lead to poor
decisions from the perspective of overall
social well-being. Individual actions
which ordinarily would channel resources
into their most highly-valued uses are
misguided by the various forms of market
failure. It is not sufficient, however,
to argue that because there are market
failures that correction through govern-
mental intervention is desirable. One
must recognize that intervention has its
costs too. Moreover, there often exist
significant legal and political obstacles
which must be overcome before governmental
action can be taken. Finally, it must be
recognized that while there are sources
of market failure which lead to inadequate
conservation of resources and insufficient
resource recovery, there exist counter-
vailing forces such as state severance
taxes, near-monopolistic control of natu-
ral resource development, and subsidiza-
tion of resource recovery, all of which
act to varying degrees to promote con-
servation and recycling. These should
be cons^/tfered in evaluating the necessity
and desirability of intervention to cor-
rect particular forms of market failure.
The plan of this paper is to review
prior research on impediments to conser-
vation and recycling with a broad view as
to the desirability of corrective action.
Because prior research efforts have usual-
ly taken a narrow focus and not examined
critical issues such as administrative and
political obstacles to intervention,
there remain many significant gaps in our
understanding of precisely what actions,
if any, are warranted. Major emphasis
will be given to two of the aforementioned
sources of market failure: the subsidi-
zation of virgin material production
through the federal tax code, and the
failure to price correctly the disposal
of waste materials of both the post-
consumer and hazardous or toxic variety.
TAX SUBSIDIES FOR
VIRGIN MATERIAL PRODUCTION
The federal tax code permits signifi-
cant income tax deductions for the pro-
duction of primary materials, the most
significant deduction historically being
percent depletion.(3) Other tax subsidies
include the expensing of exploration and
development expenditures, the capital
gains treatment of iron and coal royal-
ties, the capital gains treatment of
standing timber and the immediate expens-
ing of many of the costs of planting and
caring for a stand of timber. The total
value of these subsidies exceeds several
billion dollars per year. The major im-
pact of these various tax code provisions
is to reduce the cost of primary materials,
thereby stimulating more rapid development
and greater consumption than would other-
wise occur. Recycled materials are placed
at a competitive disadvantage with respect
to price, serving to inhibit to a certain
degree efforts to recover and reuse ma-
terials, particularly those extracted from
the post-consumer waste stream.
The evolution of percent depletion
as a tax deduction demonstrates that little
thought was given to the probable impacts
of these tax policies.'^' Rather, percent
depletion developed through a continuing
political tug-of-war in the Congress as a
seemingly unending parade of special int-
erest groups in mining sought and finally
gained the prize of percentage depletion
for their industry.
Estimating the quantitative impacts
of tax subsidies such as percent depletion
is difficult for a number of reasons. The
most important of these are the largely
unknown production functions in mining,
the cost conditions for new mineral sup-
plies, the pricing behavior of producers,
and the demand functions for mineral out-
puts. In various studies the maximum po-
tential price impacts of percent depletion
on intermediate mineral outputs which
compete with recycled materials have been
estimated at from 1 or 2 percent up to
about 10 percent, depending upon the
mineral.(5-6) That these price effects
are overstated is almost certain, partic-
ularly in view of the fact that the U.S.
is a net importer of most minerals. Do-
mestic prices are essentially determined
fay conditions on world markets and are not
172
-------
On the grounds of equity towards the
owners of existing investments, it appears
that removal of the subsidies accorded to
those investments would not be justified.
A much preferred approach would be to phase
a new system of taxation in gradually —
making it applicable only to new invest-
ments or new mineral discoveries. Even
so, the equity issue would not disappear
totally. Consider the treatment of an
existing undeveloped mining claim. With
percent depletion> the claim has a far
higher value than if percent depletion
were henceforth disallowed. Eauity for the
owner of the mining claim may dictate that
depletion be allowed on this property. But
then we would continue to encourage the
excessive investment in mining on all ex-
isting mining claims — and there are
several million undeveloped claims on the
public lands in the western states. In
this situation, it appears that the attain-
ment of greater efficiency is in a direct
and irreconcilable conflict with common
perceptions of equity. Furthermore, at-
tempts to attain, even gradually, both ef-
ficiency and equity will almost certainly
require a complicated and administratively
costly system as the subsidies are phased
out.
The political and legal obstacles in
removing the special tax subsidies for tim-
ber and mining would likely be great. No
industry would willingly acquiesce to a
heavier tax burden or see existing invest-
ments marked down sharply in value. Any
attempt to remove the subsidies would al-
most certainly encounter intense lobbying
efforts from industry. Moreover, changing
the tax code would require new tax legis-
lation, and even without industry opposi-
tion this would be a difficult feat to ac-
complish.
Highlighting the analysis thus far,
it appears that the special tax subsidies
of mining and timber operations may have
relatively large impacts on the use of
domestic virgin material supplies relative
to foreign supplies. The impacts on re-
cycling, however, do not appear to be
large. Any attempt to improve efficiency
by removing the special tax subsidies would
produce strong political opposition, par-
ticularly from Congressional representa-
tives from the major mining and timber
states. They and their constituents would
not be pleased with the equity implications
of doing away with the present system of
preferential taxation.
An issue worthy of future research is
whether it is possible to design an admin-
istratively feasible method for eliminating
the special subsidies while simultaneously
preserving intact the value of existing
investments. If such an approach could
be developed, political feasibility would
be enhanced significantly.
PRICING THE COLLECTION AND DISPOSAL
OF WASTE MATERIALS
The collection and disposal of post-
consumer solid waste, commonly known as
garbage or trash, is financed in various
ways throughout the country.'1®) Some
cities and municipalities finance such
activities wholly out of general property
tax receipts. In other areas a flat
monthly fee for collection and disposal is
levied by the city or sometimes by the
system operator if a private contractor
is used. In a small minority of communi-
ties, the fee which is assessed varies with
the quantity of waste generated by a house-
hold. Finally, in some communitiesj
varying levels of service are available
(e.g., curbside pickup or backyard pickup,
and once or twice-weekly pickup), with the
fee varying with the level of service.
From the standpoint of economic ef-
ficiency, these methods of financing and
disposing of post-consumer waste leave
much to be desired. Only rarely does the
household pay a fee which corresponds to
the marginal cost of providing the service.
In many cases, the effective price for
using additional service is zero. This
is undesirable in that the efficient allo-
cation of resources requires that price
must equal marginal cost. The consequence
of such incorrect pricing is that more
waste collection and disposal services
will be demanded than is socially optimal.
Households will have inadequate incentives
to consider alternative options such as
composting of yard wastes (leaves and
grass clippings), separating for recycling
(paper, glass, and metal), or reducing
their consumption of the goods and services
which result in the generation of large
volumes of residual solid wastes..
At least two solutions have been pro-
posed to correct this problem. One is to
institute marginal cost pricing on a local
173
-------
basis. The second is to impose a so-
called "product charge" on the manufacture
of all products destined for consumption.
As will be seen shortly, neither solution
is ideal.
Ideally, the disposal fee should be
set at the marginal cost of collection and
disposal. Therefore, an ideal set of
disposal fees would vary with all of the
factors which affect cost — the quantity
of material disposed of, the amount of
service offered, the local wage rates for
sanitary workers, and the cost of acquir-
ing sites for waste disposal. Needless to
say, the calculation of an ideal price
schedule would be fairly complex.
Establishing the proper fee schedule
is by no means the only difficulty that
will be encountered if one wishes to es-
tablish theoretically efficient prices for
disposal services. A very thorny issue
arises in how the pricing system is to be
administered.(11) Who will keep records
and present bills to households? Several
alternative mechanisms are possible —
records could be kept by sanitation work-
ers; the use of marked plastic bags, sold
only through authorized distributors, could
be required; or a stamp could be required
to be affixed to each container placed at
curbside for pickup. All of these approa-
ches have been tried in local experiments
here and abroad. None of the approaches
has gained widespread use — in fact most
of the experiments ended in apparent fail-
ure. Some systems were administratively
cumbersome and costly while in others
households objected to the inconveniences.
In some cases, fears that illicit disposal
and littering would increase, helped to
curtail the experiments.
In assessing the desirability of in-
stituting marginal cost pricing for solid
waste services, one must have a clear un-
derstanding of the magnitude of the welfare
loss which is created by present pricing
systems. Only if the administrative costs
of a marginal cost pricing system are less
than the inefficiency costs imposed on
society, would corrective action be war-
ranted. Although a vigorous assessment of
the existing welfare losses has not been
made, the available evidence suggests that
only modest losses are being incurred. The
quantity of waste material which is gen-
erated by a household appears to be quite
insensitive to the price of disposal
services. This is to be expected if fac-
tors other than disposal cost such as
tastes and preferences, income, and other
prices, are the principal determinants of
the quantity and composition of those
goods which are purchased and ultimately
diposed.
The product charge represents an
attempt to create, on a national basis, a
charge reflecting average disposal costs
of about one cent per pound for all pro-
ducts which enter the household consump-
tion and disposal stream.(12) xhe charge
would be paid by manufacturers and would
be based upon the virgin material content
of the products. (Alternatively all
materials used would be taxed but a re-
bate would be given for the use of second-
ary materials.) The product charge rep-
resents a sacrifice in efficiency for
savings in administrative costs. Be-
cause the charge would be levied on a
national basis there would be no reflection
of local variation in collection and dis-
posal costs. Furthermore, yard wastes,
which constitute approximately one-half
of all consumer solid waste, would go un-
taxed.
A major problem with the product
charge concept is the fact that it will
require new tax legislation which is no-
toriously difficult to pass in Congress.
The case for a change in the current
system of solid waste disposal pricing
cannot be made until it is shown that the
gains in efficiency in some new system
outweigh the additional administrative
costs which would be incurred. Proponents
of the product charge have not developed
a convincing case that significant effi-
ciency gains would accrue. Likewise, pro-
ponents of marginal cost pricing of solid
waste disposal have not yet designed and
validated an administratively inexpensive
system of charges. Further research on
the former and experimentation on the
latter appear necessary and desirable be-
fore recommendations for change are made.
The disposal of toxic, hazardous, and
polluting substances reflects another type
of pricing failure in solid waste disposal.
Where materials from either the industrial
or post-consumer waste streams have the
capacity to damage the environment or
human health, it may be unwise for dis-
posal to follow the typical pattern of
174
-------
burning or burial in a landfill. Although
solid waste diposal prices could be adjust-
ed to reflect the social costs of including
these materials in the solid waste stream,
a more efficient solution may be to en-
courage separation and different treatment
for the especially toxic, hazardous or
polluting wastes. Just as the Resource
Conservation and Recovery Act of 1976 sets
out special standards for generating,
transporting, storing and disposing hazard-
ous industrial wastes, so also might such
special treatment be accorded these sub-
stances in the post-consumer waste stream.
A good example of this problem is
provided by used lubricating oils, sub-
stances which are particularly likely to
cause unwanted pollution of the water and
air.'(13) Unfortunately, consumers have
long been encouraged by prices and other
market forces to dispose of used oils in
socially undesirable ways. Typically, a
consumer's choices are limited to three
main alternatives: including the oils in
the solid waste stream, burying in one's
backyard, or pouring down a storm sewer.
Occasionally a waste oil recycler will
accept such offerings. Service stations,
which themselves generate and return for
recycling large quantities of used lubri-
cating oils, normally prefer not to accept
the offerings of the "do it yourselfer."
After all, the latter is effectively a com-
petitor to the service station and although
the station is paid for its
used oil, the price may not justify the
inconvenience of accepting a few quarts
of oil from many individuals.
Potential solutions to this external-
ity include (1) the use of a deposit sys-
tem under which used oil must be returned
to a vendor of lubricating oils for a re-
fund of the deposit, (2) the prohibition of
disposal in the municipal waste stream with
the concomitant establishment of local
used oil recycling centers, or requirements
that vendors or service stations accept
used oil, and (3) requiring that service
stations accept oil which is offered by
the do-it-yourselfer.
Because service stations already have
facilities for receiving and storing used
oil, the third alternative would impose
few additional costs on society. More-
over, requiring service stations to accept
used oil produces insignificant admini-
strative burdens upon government agencies.
It may impose slight burdens upon station
operators, but this is balanced to an ex-
tent by the value of the oil which is re-
ceived. The real difficulty with this
alternative is that it requires new legis-
lation. We might note in passing that the
third alternative appears to be gaining
acceptance. A number of states have
passed used oil legislation patterned
after a model statute drafted by staff of
the Environmental Law Institute.(^)
The problems with a deposit system
are that it too would require new legis-
lation and in addition would impose admin-
istrative costs upon those selling lubri-
cating oils. Their opposition to such an
approach has effectively blocked further
consideration of deposit systems.
Prohibitions on disposal in the muni-
cipal waste stream would be very difficult
to enforce, effectively ruling out such
an approach. Furthermore, if enforced}
they could encourage illicit and environ-
mentally more damaging methods of disposal
such as surreptitious dumping in storm
sewers.
OTHER SOURCES OF MARKET FAILURE
In the introduction we noted several
other potential sources of market failure:
vertical integration in primary production,
labeling requirements, railroad freight
rate regulation, and incorrect perceptions
as to the quality of recycled materials.
This section offers a brief review of
these issues, possible solutions, and
recommendations for further research on
these topics.
Vertical integration in primary pro-
duction which can encompass all stages of
activity from the development of sources
of raw material inputs to the manufacture
of goods for final consumption, tends to
inhibit reliance en secondary sources of
supply. Users of raw material inputs pre-
fer to rely upon the certain availability
of firm-owned supplies rather than the
sometimes chaotic markets for recycled
materials. Only when company-owned
supplies are inadequate to meet company
needs is there a great pressure to acquire
raw materials from other sources.
We might ask why vertical integration
has occurred in so many of the industries
which rely upon raw material inputs, and
175
-------
why the vertical integration has favored
primary sources of raw material supply al-
most exclusively- Mancke noted that a
driving force for vertical integration in
the steel industry during the early Twenti-
eth Century was a fear on the part of the
steel firms, that an iron ore cartel could
be established and that such a cartel could
extract the economic rents then accruing
to the steel industry.(15) Aside from the
fact that industrial and post-consumer sup-
plies of scrap constitute a lesser per-
centage of raw material inputs than does
iron ore, the steel industry also must have
realized that capturing control of the
geographically widely dispersed supplies
of scrap steel would be difficult to ac-
complish. Equally difficult would be the
establishment of a scrap iron and steel
cartel. Similar observations probably ex-
plain why paper companies have often inte-
grated backward into the ownership of tim-
ber stands but have not attempted to gain
control of sources of wastepaper supply.
For certain, vertical integration
serves as a deterrent to the recovery of
scrap materials, but it is not clear what,
if any, policy action should be taken. Anti-
trust action could be initiated on the
grounds of restraint of trade. Such an
approach has the advantage that it lies
within existing statutory authorities.
Alternatively, with new legislation, in-
centive systems could be created to encour-
age firms to divest themselves of ownership
of virgin material supplies and to acquire
control of sources of secondary materials.
Labeling requirements for goods which
contain recycled materials can act as a
deterrent to resource recovery, particular-
ly if the labels effectively create mis-
understandings about product quality. Re-
cent legislative action has sought to pro-
hibit labeling requirements which discrimi-
nate against recycled oil. Section 383 of
the recently enacted "Energy Policy and
Conservation Act" reads in part: "No rule
or order of the Federal Trade Commission
may require any container of recycled oil
to also bear a label containing any term,
phrase, or description which connotes less
than substantial equivalency of such re-
cycled oil with new oil."d°) Whether par-
allel action covering other commodities is
necessary or desirable should be studied
further.
Railroad freight rate regulations
allegedly discriminate against recycled
commodities in favor of virgin materials.
A number of studies have been conducted
to probe this issue.(17-19) ^he most con-
vincing of the studies show that on a
weight basis virgin materials do travel
for significantly lower prices than does
the corresponding recycled material. But
to show discrimination requires that one
also assess the costs of service. If
virgin materials happen to cost the rail-
roads correspondingly less per ton to car-
ry (perhaps because they are denser, or
are less damaging to equipment, or are more
easily loaded and unloaded), then there is
no discrimination. Moreover, inasmuch as
railroad freight rates are usually influ-
enced by perceived elasticities of demand
for transportation, commodities for which
demand is thought to be inelastic will
tend to have higher rates than commodities
whose demand is more elastic. Scrap
materials may have relatively inelastic
demands for freight service and as a con-
sequence may experience somewhat higher
railroad transport costs than do virgin
materials.
CONCLUSIONS
In a private free-enterprise economy
decisions as to the amounts of materials
to consume, the amounts to conserve, and
the amounts to recycle are left to deter-
mination by market forces. In our mixed
economy, governmental actions also in-
fluence these decisions. The basic thrust
of the literature on resource use is that
conservation and recycling should be fos-
tered in order to assure a sufficient level
of these worthwhile activities. The need
to foster greater conservation and recy-
cling has historically been based on a
widespread belief that private market
forces would not provide for a sufficient
level of these activities. Economists have
shown that this belief is generally in-
valid; more recently the calls made by
the economics profession for policy action
to encourage resource conservation and
recycling have been based on arguments
that private market decisionmakers receive
incorrect signals and therefore make
socially undesirable decisions.
These incorrect signals may arise from
a variety of institutional and economic
forces — including institutional forces
176
-------
such as tax subsidies to virgin material
production, vertical integration in primary
production, labeling requirements for re-
cycled goods, and the regulation of freight
rates; externalities such as the failure
to price correctly the disposal of post-
consumer waste; and informational problems
such as incorrect perceptions as to the
quality of recycled materials. In general,
all of these sources of market failure lead
to poor decisions concerning resource use.
Although the individual private decision-
maker is maximizing his own welfare, this
does not necessarily lead to socially
optimal decisions when market failures are
present.
From the perspective of public policy,
the critical issue is whether intervention
in resource markets is warranted and if
so, what form the intervention should take.
This paper has attempted to show that this
question is complex and that many subsidi-
ary questions must be resolved in a satis-
factory manner before a particular policy
prescription can be recommended. Some
of these subsidiary issues concern such
dimensions as the administrative costs of
intervention, the equity implications of
intervention, and the political and legal
obstacles to intervention.
This paper reviewed in detail two
sources of market failure — the subsidi-
zation of virgin material production
through the federal tax code and the in-
correct pricing of consumer waste dis-
posal — and examined the desirability of
a variety of alternative policy prescrip-
tions for these problems. Both sources of
market failure tend to inhibit conservation
and recycling, but the analysis here shows
that we are not yet prepared to advocate
a particular solution in the form of fur-
ther public intervention for either of
these problems. The best policy may be
inaction, but that can only be determined
conclusively when more information is in
hand.
REFERENCES
1. Anderson, R. C., "Evaluation of Econ-
omic Benefits of Resource Conserva-
tion," U.S. EPA, 600/5-78-015, 1978.
2. Resource Conservation and Recovery
Act of 1976 (P.L. 94-580) § 1003.
3. Anderson, R. C. and R. D. Spiegelman,
"The Impact of the Federal Tax Code
on Resource Recovery," NTIS, PB 264
886, 1976.
4. Anderson, R. C., A. S. Miller, and
R. D. Spiegelman, "U.S. Federal Tax
Policy: The Evolution of Percentage
Depletion for Minerals," Resources
Policy, V. 3, No. 3, 1977, pp. 165-178.
5. Anderson and Spiegelman, op. cit.
6. Toder, E. J., "Federal Tax Policy and
Recycling of Solid Waste Materials,"
U.S. Treasury Department, Preliminary
Draft Report, Sept. 1978.
7. Anderson and Spiegelman, op. cit.
8. Toder, op. cit.
9. Anderson, R. C. and R. D. Spiegelman,
"Tax Policy and Secondary Material
Use," Journal of Environmental Econ-
omics and Management, V. 4, No. 1,
1977, pp. 68-82.
10. Lanen, W., "Efficiency of User
Charges," presented at Workshop on
Solid Waste Economics, Philadelphia,
Pa., Sept. 19-20, 1978.
11. Stevens, B., "Administrative Costs
of User Fees," presented at Work-
shop on Solid Waste Economics, Phila-
delphia, Pa., Sept. 19-20, 1978.
12. Buchanan, S., "Evaluating the Effi-
ciency of the Solid Waste Charge,"
presented at Workshop on Solid Waste
Economics, Philadelphia, Pa., Sept.
19-20, 1978.
13. Irwin, W. A. and R. A. Liroff, "Used
Oil Law in the United States and
Europe," U.S. EPA 600/5-74-025.
14. Irwin, W. A., "A Model Used Oil Re-
cycling Act," report to Federal
Energy Administration, July 1976.
15. Mancke, R., "The American Iron Ore
and Steel Industries: Two Essays,"
unpublished Ph.D. dissertation, MIT,
1969.
16. 42 USC 6201 (P.L. 94-163).
177
-------
17. Albrecht, 0. W., "Shipping Wastes to
Useful Places," Environmental Science
and Technology, V. 10, No. 5, May
1976, pp. 440-442.
18. Resource Planning Institute, "Raw
Materials Transportation Costs and
Their Influence on the Use of Waste-
paper and Scrap Iron and Steel," NTIS,
PB 223-871, 1974.
19. Moshman Associates, "Transportation
Rates and Costs for Selected Virgin
and Secondary Commodities," NTIS,
PB 233-871, 1974.
178
-------
SMALL-SCALE AND LOW TECHNOLOGY RESOURCE RECOVERY
Gary L. Mitchell, PE
Charles W. Peterson
SCS Engineers
Reston, Virginia
ABSTRACT
A study was conducted to assess the applicability of various approaches to resource
recovery to selected waste generators. The resource recovery systems and technologies
were limited to those operating in the small-scale range, defined as less than 100 tons
per day input, or those approaches considered to be low technology, defined as having
more than 50 percent of operation and maintenance costs associated with labor, i.e.,
labor intensive. The generators included institutions, commercial sources, office
building complexes, multi-unit residences and small cities.
An evaluation of seven potential systems led to the conclusion that two approaches
were technically and economically feasible for application to the waste generators. The
two systems identified were modular incineration with energy recovery and source separa-
tion. A detailed analysis of the application of these two systems to the waste stream
generators led to determination of applicability of either or both approaches to re-
source recovery to each of the generators. It was found that modular incineration is
generally applicable to only the largest examples of the waste generators studied.
Similar conclusions were associated with source separation; however, this approach was
found more applicable to smaller situations than was modular incineration. Recommenda-
tion for future research and development included more thorough waste characterization
of the sources studied, investigation of the effects of building design on resource re-
covery feasibility, and a further study of systems not currently considered as proven
technology.
INTRODUCTION
Approaches to the recovery of mate-
rials or energy from municipal wastes
have been in operation for several years.
This is particularly true of source sepa-
ration and incineration with energy re-
covery. Most examples of these systems
are associated with metropolitan areas
with populations exceeding 100,000.
Little information has, however,has been
compiled on procedures for recovering
resources from specific waste generators
within metropolitan areas or in smaller
cities.
Section 8002(d) of the Resource
Recovery and Conservation Act of 1976
(RCRA) required the U.S. Environmental
Protection Agency (EPA) to conduct a
study of small-scale and low technology
resource recovery. The purposes of the
project were:
• To compile a comprehensive biblio-
graphy on small-scale and low tech-
nology resource recovery systems
t To determine solid waste character-
istics and collection and disposal
practices of selected small waste
generators
• To analyze small-scale and low
technolog'y resource recovery sys-
tems and evaluate applicability of
the most feasible systems to the
various waste generators
t To make recommendations for future
research and development efforts.
DEFINITION AND SCOPE OF PROJECT
Small-scale systems were defined as
technologies which operate at a maximum
capacity of 100 tons per day with less than
179
-------
50 percent of the operating and mainte-
nance costs devoted to labor.* Thus,
these small-scale systems are essential-
ly smaller versions of high technology
approaches to resource recovery. The
small-scale systems' input would be mixed
wastes from the sources identified with
the output including separated materials
and/or recovered energy.
Low technology systems were defined
as having 50 percent or more of the
operating and maintenance costs associat-
ed with labor. No limitations were
placed on the input capacity of low tech-
nology systems. The principal low tech-
nology system in the United States is
source separation; however, many
approaches to material recovery could fit
the definition if they were oriented to
manual labor rather than mechanized
operations.
The project focused on several types
of small volume waste generators includ-
ing the following:
• Institutions
- Hospitals
- Prisons
- Universities
• Office Buildings
• Commercial Sources
- Airports
- Shopping Centers
• Multi-unit Residences
- Garden or Low-rise Apartments
- Mobile Home Parks
• Small Cities
The inclusion of some of these waste
generators was required by the legisla-
tion directing this study. Others were
included because it was felt that they
were potential candidates for resource
recovery operations. Some of the above
sources had easily segregated portions of
the waste streams containing materials
amenable to recovery. Examples include
the recovery of corrugated cardboard from
prisons, universities, and shopping cen-
ters. Some of the waste generators were
considered as likely on-site consumers of
energy recovered from the incineration of
their wastes. These included hospitals,
prisons, universities, airports, shopping
centers and low-rise apartments. Small
cities were included essentially to com-
plete the scope of potential resource
recovery from municipalities.
APPROACH
Waste Characteristics and Generation Rates
In order to assess the applicability
of resource recovery to these waste genera-
tors, information was needed on solid waste
generation rates, waste composition, and
typical collection and disposal practices.
A search of the literature yielded very
little information on the generators with
the exception of small cities, office
buildings, and hospitals. This lack of
published data necessitated limited, infor-
mal, on-site waste characterization studies
supplemented by telephone surveys.
The literature review and data collec-
tion efforts yielded the approximate waste
composition and generation rates shown in
Table 1. It should be noted that some of
the waste generators were divided into sub-
categories, e.g., four different types of
operations at airports and three different
sizes of shopping centers; however, only
one example of each source is shown in
Table 1. Of particular interest were the
units of measure associated with overall
waste generation rates. Most of the liter-
ature reported some factor of waste genera-
tion per day or week, related to a measure
of size or level of activity at the facil-
ity. The size measurement was often based
on the number of persons using the facil-
ity, but also included floor areas (for
shopping centers) and number of paid staff
(hospitals). Composition data was taken
from a limited number of sources in most
cases, as was the overall waste generation
rates. The effort to collect information
on waste stream generators and applicable
resource recovery technologies led to the
development of a topical bibliography.
*Metric units of measure were not used in this project. It was felt that the user com-
munity was more accustomed to English units and that the degree of actual use of this
report would decrease if metric units were reported. The use of English units in lieu of
metric was approved by the Project Officer.
180
-------
TABLE 1. APPROXIMATE WASTE COMPOSITION AND GENERATION RATES
FOR SELECTED WASTE GENERATORS
Percent of Total (Weight Basis)
00
Type of Material
Paper
Corrugated
Glass
Metal
Plastics
Organics
Wood
Miscellaneous
TOTAL
Overall Waste
Generation Rates
Airport
Passenger
Terminal
I71
Regional
Shopping
Center
28
Hospitals
1 40
J 52 J
4
6
5
5
3
6
100
0.5 Ib
per
passenger
per day
1
3
8
2
2
4
100
200
per
sq.
Ib
1000
ft.
of gross
leasable
area
week
per
6
2
15
25
1
12
J
100
2 to 4.5
Ib per
paid
staff
member
per day
Prisons
37
22
1
16
8
10
1
6
J
100
4.5 Ib
per
inmate
per day
Universities
55
10
8
7
3
10
1
7
J
100
1 Ib per
student
per day
Multi-Unit Office Small
Residences Buildings Cities
I35 1
J J
12
10
5
27
]
11
J
>87 J29
J
1 10
7 10
1 3
38
> 4 4
6
100 100 100
2.7 Ib 1.5
per per
Ib 3.5 Ib
per
resident office person
per day worker per day
per
day
-------
The generation rates were found to
be quite variable, depending upon the
type of activity conducted. This was
especially true of hospitals and prisons.
Within hospitals such "heavy care" units
as surgery and maternity had high genera-
tion figures, whereas, "light care" units
such as psychiatric and administrative
units generated much less waste. The
generation rate for prisons is associated
with residential and administrative as-
pects of these institutions and did not
include any wastes generated by indus-
trial or agricultural activities. The
fairly extensive use of disposable items
is reflected in the generation rate and
the quantities of paper and plastics.
The generation rates at universities
depends upon the types of wastes includ-
ed. The generation rate shown does not
include wastes from agricultural or medi-
cal schools or landscaping, demolition,
and construction wastes. Likewise, it
was noted that sources of recyclable
materials at universities, particularly
paper, were easy to identify, with
approximately 80 percent of all paper
coming from office and classroom areas. A
high percentage of this paper is
recyclable.
Office buildings are notable for the
very high percentage of paper. Much of
this is high-grade paper and computer
cards and printout which often command
premium prices. Numerous office build-
ings of governmental agencies and private
firms practice source separation of high-
grade paper. A relatively small amount
of organics is generated by office build-
ings without food service facilities.
The percentage of organics increases
significantly for office buildings with
food service facilities.
Relatively large quantities of
organics are shown coming from multi-unit
residences and small cities. This is due
to the fact that both garbage and land-
scaping wastes are included.
The data in Table 1 are derived from
a relatively small sampling of informa-
tion from the specific sources, with the
exception of hospitals, office buildings,
and small cities. Thus, in assessing the
feasibility of resource recovery in a
specific situation, a waste characteriza-
tion study should be completed to deter-
mine the composition and quantity of wastes
being generated at that location.
Technology System Evaluation
System Criteria—
The next step in this study was to
collect information and evaluate the cur-
rent state-of-the-art approaches to re-
source recovery for the their applicability
to the waste generators selected. Informa-
tion concerning the state-of-the-art,
applicability, and costs associated with
material processing components was obtained
from a variety of sources including pub-
lished material and through direct contact
with equipment users and manufacturers. No
site visits to resource recovery operations
were made during this project. Gathering
of information was not limited to the
United States. As part of the project, the
First idorld Recycling Congress was attend-
ed. The five-day Congress was held in
Basel, Switzerland and included technical
papers and equipment displays from some 20
countries worldwide.
Unit process components considered
technologically proven at the 100 ton per
day level or less were assembled into re-
source recovery systems applicable to the
waste generators. Operational and techni-
cal aspects of each system were analyzed
and a cost analysis developed which was
used to estimate net disposal costs per ton
of input.
For purposes of the study components
of resource recovery systems were consider-
ed technologically proven, if, at 100 tons
per day or less, the component has:
• Operated at full-scale for at least
one year
t Produced the desired product in a
marketable form.
This definition eliminated components
in the pi lot-scale or in shake-down tests.
Likewise, it eliminated those components
generating a product that is not market-
able; i.e., no market exists for the
material or has existed in the very recent
past. The above definition was applied to
various resource recovery system compo-
nents. The components of small-scale sys-
tems initially considered included the
following:
182
-------
acid hydrolysis conversion units
air classifiers
aluminum magnets
composting equipment
froth flotation units
magnetic separators
methane digesters
modular incinerators
pyrolytic units
shredders
trommel screens
Application of the criteria for a
component to be technologically proven
prior to further consideration led to the
elimination of several components. These
are listed below along with the reason
for elimination:
• Acid hydrolysis conversion unit
— considered to be in experimen-
tal or pilot stage.
• Aluminum magnets — unable to
assess effectiveness of the one
operational unit due to small
amounts of aluminum in waste
stream. Other installations con-
sidered to be in shake-down
status.
• Froth flotation -- considered to
be in shake-down status. Al-
though high purity product (99%)
has been achieved, it still does
not meet container industry
specifications.
• Methane digesters — pilot-scale
plant for solid wastes and sewage
sludge in operation. No system
in operation at commercial scale.
• Pyrolysis — two small operating
units considered to be in demon-
stration or shake-down status.
The remaining components were "assembled"
into six, small-scale systems for further
analysis:
• Ferrous Recovery
\
• Compost Preparation
§ Compost Preparation with Ferrous
Recovery
• RDF Preparation with Ferrous
Recovery
• Incineration with Heat Recovery
• Incineration with Heat and Ferrous
Recovery
Application of System and Cost Analyses-
Next, scenarios were developed
describing the features of each of the
above systems, their limitations, and their
applicability. This led to the development
of a cost analysis of each. In order to
evaluate the systems as uniformly as pos-
sible, common assumptions were applied, as
follows:
• All systems to be operated at 100
tons per day input.
• Hauling and disposal costs for non-
recovered waste from the processing
facility were estimated at $7 per
ton.
• Uniform costs
applied.
for labor were
• An average rnid-1977 market value
for recovered material was used.
• The value of energy recovered was
equated to the cost of the least
expensive fossil fuel from which
the same amount of energy could be
recovered.
An example cost analysis is shown in
Table 2. Specific assumptions associated
with modular incineration are shown in the
table.
In the area of low technology systems
only source separation was considered, and
a scenario was developed describing a typi-
cal source separation program applicable to
a small city. The great variability of
source separation, and its applicability to
several of the w.aste generators precluded
the development of scenarios and the re-
sulting cost analyses of more than one
source separation program.
Overall Evaluation—
As a result of an overall evaluation
considering costs and other factors, modu-
lar incineration with heat recovery (with-
out ferrous recovery) and source separation
emerged as the highest rated systems.
Additionally, these were the systems with
the lowest cost per ton. An area of poten-
tial concern for modular incinerators is
air pollution control. While several
183
-------
TABLE 2. EXAMPLE COST ANALYSIS
MODULAR INCINERATOR WITH ENERGY RECOVERY1 (100 TPD)
Amortization
Initial Life Factor Annual
Costs (Years) (8%) Costs
CAPITAL COSTS ($1,000)
Incinerator and boiler, complete
Auxiliary Equipment
Small Front-End Loader
Office Furniture, Refuse Bins
Construction & Land
Building: 9500 ft2 @ $30/ft2
Site Development: 20% of bldg.
Land: 5 acres @ $10,000/acre
in place2 $1800 15
50 5
40
10
396 20
288
58
50
0.117 $210
0.250 13
0.101 40
TOTAL
OPERATING COSTS ($1,000)
Labor:2 4 operators @ $48
1 supervisor @ $16
Supplies: 3% of labor & maint.
Energy:1* Supplemental fuel
Mobile loader
Lighting
Heat building
Maintenance: 3% of total capital costs
Miscellaneous:
$2246
$36.0
3.2
0.7
1.3
(taxes, licenses, insurance, administrative
and management costs) 1% of total initial
capital costs
TOTAL
TOTAL ANNUAL SYSTEM COSTS ($1,000)
COSTS/REVENUES PER TON ($/ton)
System Cost
System Revenue
Net System
Landfi11
Total Net
$263
64
3
41
66
22
$196
$459
$ 17.65
8.08
9.57
2.11
$ 11.68
184
-------
TABLE 2. (Continued)
Footnotes:
1 Data calculated by SCS Enginners from literature and vendor sources.
2 Includes 5 25-TPD units. Incinerators are designed to operate on a 24-hour
basis. The extra unit provides reserve capacity for maintenance.
3 Operators are on duty for 8-hour shifts. The shifts are split to allow for
continuous operation, 5 days per week. Wage rate is $5.80 per hour, which
includes fringe benefits of 15 percent. The supervisor is on duty for one
8-hour shift at $7.80 per hour.
"* Energy: Supplemental fuel is consumed at a rate of 5 percent of the BTU
value of the input refuse. Operating conditions:
• Thermal value of refuse
t Supplemental fuel
• Cost of gas
• Thermal value of therm
5,000 BTU/pound
Natural Gas
$0.2776/therm
100,000 BTU
Mobile Equipment - operation conditions are:
t Gasoline Consumption
- Cost
5 Revenue Factors:
• Percent combustibles in wastestream
t Recovery rate
t Market value: substitute value of
coal
6 Cost Factors:
• Weight Reduction
• Cost to haul to landfill and
disposal
2.5 gallons/hour
$0.60/gallon
90%
$1154/100 tons of combustible
refuse @ 5000 BTU/pound
70%
$7/ton
185
-------
installations have met State standards,
it is possible that some units will re-
quire external air pollution control de-
vices. The approach used in evaluating
the systems is described below.
Those systems with apparent technical and
economic feasibility and desirability
for application to the waste stream gen-
erators were subjectively evaluated. In
order to evaluate the systems, rating
criteria were selected that represent the
characteristics (other than economics) of
greatest concern to smal1 waste genera-
tors. Site-specific factors, such as
public acceptance, were not rated, al-
though they are extremely important. The
criteria selected were essentially quali-
tative in nature and thus were rated
accordingly. Three major categories of
criteria were performance, environmental
acceptability, and marketability of re-
covered product. These were further sub-
divided into more detailed criteria; an
explanation of the approach used to rate
each is shown in Table 3.
Consideration of information avail-
able in the literature and through con-
tacts with system owners and operators
lead to the evaluation of the systems
using the above criteria. The results
are shown in Table 4. A review of the
ratings in Table 4 led to the elimination
of several systems from further evalua-
tion as to their applicability to the
waste stream generators. Reasons for
eliminating these systems are summarized
below:
• RDF — high costs, market uncer-
tainty, and possibile problems
associated with storage and
transport of the material
• Ferrous Recovery ~ high costs
and price fluctuations for re-
covered ferrous
i Compost — high costs, large
capital investment and virtually
no market for the product.
Application of Technologies to Waste
Stream (jenerators"
Summary—
The applicability of modular in-
cineration or source separation or a com-
bination of the two to any of the waste
stream generators was then evaluated on
the basis of the overall system costs. The
inclusion of resource recovery was con-
sidered appropriate and feasible in all
situations where the overall waste manage-
ment costs with the inclusion of resource
recovery were equal to or less than the
waste management costs prior to any change.
Costs for collection and disposal of wastes
from all sources except small cities was
estimated to be $28 per ton, which includes
rental of bulk waste containers. For small
cities the disposal costs of $7 per ton was
used. It was assumed that the city would
have to collect the wastes whether they
used landfill disposal or hauled it to a
resource recovery facility; thus, only the
cost of disposing of unrecovered material
impacted on this analysis. The results of
this analysis indicated situations for
waste stream generators in which one or the
other or both approaches to resource re-
covery was less expensive than current dis-
posal and thus considered feasible. Over-
all feasibility is shown in Table 5.
As can be seen from the table, there
are only a few situations in which either
modular incineration or source separation
or a combination of both were unequivocally
considered feasible for a particular waste
generator. In most instances, resource re-
covery was considered viable only for the
largest examples of each waste generator,
thus exemplifying some economies of scale
associated with these operations. For
those waste generators where the combina-
tion of both approaches to resource re-
covery was indicated as feasible, it should
be understood that the decision was based
upon meeting the more stringent of the re-
quirements for either incineration or
source separation. For example, both modu-
lar incineration and source separation
would be considered feasible in office
buildings if there is at least 4,000
employees.
Only multi-unit residences (low-rise
apartments and mobile home parks) were not
considered amenable to resource recovery.
Neither normally has enough residents to
generate the required quantity of materials
or mixed wastes. Likewise, there are vir-
tually no examples of situations where re-
covered energy could be used in the mobile
home park or in an apartment building, ex-
cept possibly a large apartment complete
with a central boiler/hot water system.
Particular note should be made of the
indicated feasibility of modular
186
-------
TABLE 3. SYSTEM RATING CRITERIA
I. PERFORMANCE
A. Reliability:
High
Medium
Unacceptable
C.
System and components proven to perform dependably and
with minimum down-time.
Description
Proven performance with high reliability
Adequate performance with adequate reliability
Inadequate performance with inconsistent reliability
Degree of Waste Volume Reduction
Rati ng Description
High
Medium
Low
>60%
30-59%
0-29%
Freedom from Maintenance/Simplicity
Rating Description
Simple; minimal skills required for operation; few or no
moving parts
Moderate; intermediate in mechanical complexity; operation
requires some degree of skill and/or training
Complex; involves sophisticated mechanical equipment;
skilled and trained operators required
II. ENVIRONMENTAL ACCEPTABILITY
A. Meets all minimum standards for air, noise, water and land pollution
Rating Description
Acceptable
Unacceptable
B. Maximizes resource recovery within technological limits
Rating Description
High Recovers maximum number of resources; >60% of waste
Complies with minimum standards
Does not meet standards
Medium
Low
Recovers moderate number of resources; 30-59% of waste
Recovers few resources; <29% of was-te stream
III. MARKETABILITY OF RECOVERED PRODUCT(S)
Rating Description
High
Medium
Low
Product(s) have ready markets
Product(s) are somewhat marketable, but prices subject
to cyclical swings
Product(s) difficult to market or have very low value
187
-------
TABLE 4. SYSTEM RATING
00
oo
Waste
Volume
Reliability Reduction J
Ferrous
Recovery Medium Medium
Compost Medium Medium
Compost with
Ferrous
Recovery Medium High
RDF with
Ferrous
Recovery Medium High
Incineration
with Energy
Recovery High High
Incineration
with Ferrous
and Energy
Recovery Medium High
Source
Separation Medium Medium
... - ^\/cf"nm Dn + inn f*v*i + nv*i a f 1 fin TPH c»\/ci'mn^
oybccni ivatiny untcria ^ luu nu oybtuniy
Freedom from
Maintenance; Environmental Resource Marketability Net
i.e. , Simplicity Standards Recovery of Product(s) Cost/Ton*
Low Acceptable Low Low $15.38
Low Acceptable Medium Low 26.70
Low Acceptable High Low 26.05
Low Acceptable High Medium 13.61
Medium Acceptable** High High 11.68
Low Acceptable** High Medium 11.95
High Acceptable Medium Medium 8.16
* Cost of operating system minus revenues plus disposal of non-recovered material.
** May require external air pollution control equipment.
-------
TABLE 5. FEASIBILITY OF RESOURCE RECOVERY SYSTEMS TO WASTE GENERATORS
Waste
Generator
Small Cities
Office Buildings
Ai rports
Low-rise
Apartments
Mobile Home
Parks
Prisons
Hospitals
Universities
Modular
Incineration w/
Energy Recovery
If disposal costs
are more than $12
per ton
If more than 4,000
employees
At major airports
Shopping Centers Yes
No
No
Only the largest;
1,500+ inmates
If more than 500
beds
If more than 5,000
students
Source
Separation
Newspaper—above
13,000 population @
30% participation
Fe, Al, glass and
newspaper—above
22,000 population @
30% participation
High-grade paper
down to 75 employees
Corrugated, some-
times
Corrugated
No
No
Corrugated
No
High-grade paper
Both
No
Yes
Yes
Yes
No
No
Yes
No
Yes
incineration to small cities. The dis-
posal costs of $7 per ton was used for
small cities because it was used earlier
in the cost analyses of the seven systems
originally considered technologically
proven. This cost may be too low, par-
ticularly for small disposal operations;
e.g., 100 TPD, that are truly sanitary
landfills. Likewise, the value of the
energy recovered at any particular loca-
tion may be greater than the $1 per mil-
lion BTU assumed in Table 2. Several
small cities are cost effectively using
this method of resource recovery.
It cannot be overemphasized that the
feasibility of any approach to resource re-
covery is highly site-dependent. The
evaluations made in the report were based
upon assumed costs for sol id waste manage-
ment and values of recovered energy and
materials. There are, and will be, excep-
tions to the findings of this report.
Thus, it is imperative that any waste
generator evaluating resource recovery make
an individual feasibility study conducted
by competent, experienced personnel prior
to making any commitments.
189
-------
Specific Applications--
Situations in which resource re-
covery was considered feasible were pre-
sented as shown in Tables 6 through 13.
These tables were developed to present a
yuide to managers in a preliminary deter-
mination of resource recovery feasibil-
ity. In the example for universities in
Table 6, the application of modular in-
cineration is shown in the upper portion
of the table. A consistent waste compo-
sition was assumed for each of the first
six conditions. This resulted in an
estimated energy content of the waste
measured in BTU per pound. Next, various
waste generation rates were assumed ranging
from 1 to 2 pounds per student per day.
At this point, the equipment available
for modular incineration with energy re-
covery was considered. Currently, equip-
ment is available to operate on a contin-
uous basis (24 hours per day) or on a batch
basis (8 hours per day). Both were con-
sidered applicable to universities as it
was assumed that recovered energy would be
in the form of steam connected to an exist-
ing -steam generating plant on the campus.
The different periods of operation are
TABLE b. RESOURCE RECOVERY APPLICABLE TO UNIVERSITIES
MODULAR
INCINERATION
Condition 1
22,000
students
'Condition 2
14.670
students
Condition 3
11,000
students
Condition 4
8,800
students
Condition 5
5,870
students
Condition 6
4,400
students
SOURCE
SEPARATION
Condition 1
9,750
students
Condition 2
6,500
students
Condition 3
4.875
students
MODULAR
INCINERATION
and SOURCE
SEPARATION
Condition 1
23,800
students
Condition 2
15,870
students
Condition 3
11,900
students
Condition 4
9,600
students
Condition 5
6,400
students
Condition 6
4,800
students
On-Slte
Energy
Use
Yes-
$1/10«
BTU
•
•
•
•
N
N/A
•
M
On-Site
Energy
Use
$1/10'
BTU
•
M
•
•
•
Materials
Market
N/A
•
M
m
m
m
High-grade
paper-
$55/ton
M
m
Materials
Market
High-grade
paper-
$55/ton
*
It
•
N
•
BTU/
Ib
Waste
6,000
M
•
•
N
•
BTU/
Ib
Waste
5,370
•
•
N
M
M
TPD
10
M
•
4.4
*
V
4.9
M
•
TPD
11.9
{11 TPD
MOD INC)
m
m
4.8
(4.4 TPD
MOD INC)
M
N
Generation
Rate
1 Ib/
student/
day
1.5 Ib/
student/
day
2 Ib/
student/
day
1 Ib/
student/
day
1.5 Ib/
student/
day
2 Ib/
student/
day
1 Ib/
student/
day
1.5 Ibs/
student/
day
2 Ib/
student/
day
Generation
Rate
1 Ib/
student/
day
1.5 Ib/
student/
day
2 Ib/
student/
day
Waste Composition
Paper 651 Misc 4X
Plastic 31 Other 181
Orqanlcs 101
• •
• N
• •
M •
• m
High-grade paper 15J
Waste Composition
High-grade paper 15X
•
•
•
"
•
Recovery
Rate
24 hr/
day
*
•
8 hr/
day
•
m
50X
Recovery
Rate
24 hr/
day
SOX
•
•
8 hr/
day
SOX
•
•
Existing
System
Cost/Day
$308
•
•
J123
•
•
$137
Existing
System
Cost/ Day
$333
»
N
$134
•
•
New System
Net Cost/
Day !
$275
•
•
$108
-
M
$137
New Systen
Net Cost/
Day
$300
m
•
$127
*
•
190
-------
noted in the column entitled Recovery
Rate. A value for the recovered energy
was then assumed in order to evaluate the
net cost of the solid waste management
system includiny resource recovery.
The following step was to calculate
the minimum number of students that would
generate the required quantity of waste
needed to fuel the incinerator (the six
conditions for modular incineration).
Thus, if an administrator of a university
knew or assumed that the waste composi-
tion was similar to that shown and that
the value of the recovered energy was
also similarly assumed ($1 per million
BTU) and knew what the waste generation
rate was and that it ranged between 1 and
2 pounds per student per day, he could
determine the feasibility of modular in-
cineration with energy recovery.
As an example, if the waste genera-
tion rate at a university was known to be
1.5 pounds per student per day with the
composition and energy costs similar to
that shown on Table 6, it would be pos-
sible to incinerate the wastes and re-
cover energy either on a 24-hour-per- day
basis or 8-hour-per-day basis (conditions
2 and 5 respectively). With approximate-
ly 6,000 students or more, Condition 5 is
met and a batch-type incinerator operat-
ing 8 hours a day appears appropriate and
economically feasible. Note that the
system with energy recovery is estimated
to cost $108 per day as compared to the
existing system cost of $123 per day. If
the university had approximately 15,000
or more students, a 24-hour-per-day
incineration and recovery system appears
to be appropriate and feasible.
Source separation was evaluated in a
similar manner in the center portion of
Table 6. High-grade paper was the only
material that was assumed recoverable
with a price assumed as shown under
Materials Market. Likewise, the concen-
tration of high-grade paper was assumed
as shown. By knowing the overall waste
generation rate (again, between 1 and 2
pounds per student per day) and project-
ing a 50 percent recovery rate, a college
administrator could determine if he
should make a more detailed study of
table. This is essentially a combination of
the two upper evaluations with one major
exception. There was a correction for the
energy content of the waste due to the re-
moval of high-grade paper through source
separation. Note the reduction of energy
content from 6,000 to 5,370 BTU per pound.
The applicability of resource recovery to
the other waste generators is shown in
Tables 7 through 13.
RESEARCH AND DEVELOPMENT NEEDS
The final objective of the project was
to make recommendations for further ex-
penditures of efforts and funds in the area
of research and development associated with
small-scale and low technology resource re-
covery. Three areas were recommended for
further study:
M
• In-depth waste characterization
studies of small waste generators
included in this study, especially
shopping centers, prisons, univer-
sities, and airports.
• Studies to determine how changes in
building design could facilitate
and encourage resource recovery.
Shopping centers appear to be the
most fruitful area for these
efforts.
• Analysis of resource recovery sys-
tems in • the developmental stage
with a likelihood of successful
application to small generators.
The systems recommended for further
analysis included production of RDF
and the application of vermicom-
posting to solid waste management.
When the production of RDF at large
resource recovery facilities in-
creases and the use of this fuel
increases, markets for RDF will
develop and become stable. This
may give- rise to situations where
small generators (probably small
cities) near markets can economi-
cally produce RDF. Vermicomposting
is essentially the only low tech-
nology resource recovery system
that is applicable to such waste
stream generators as prisons, uni-
versities, and small cities.
191
-------
TABLE 7. RESOURCE RECOVERY APPLICABLE TO SHOPPING CENTERS
MODULAR
INCINERATION
Condition 1
BOO .000
SFGLA*
Condition 2
640.000
SFGLA
Condition 3
535,000
SFGLA
Condition 4
400.000
SFGLA
Condition 5
320,000
SFGLA
Condition 6
265,000
SFGLA
SOURCE
SEPARATION
Condition 1
280,000
SFGLA
Condition 2
224,000
SFGLA
Condition 3
190,000
SFGLA
MODULAR
INCINERATION
and SOURCE
SEPARATION
Condition 1
1,560.000
SFGLA
Condition 2
1,250,000
SFGLA
Condition 3
1 ,040.000
SFGLA
Condition 4
630.000
SFGLA
Condition 5
504,000
SFGLA
Condition 6
420,000
SFGLA
On-Slte
Energy
Use
Yes-
$1/10«
BTU
•
•
•
•
N
N/A
On-Site
Energy
Use
$1/10'
BTU
•
•
M
M
•
Materials
Market
N/A
m
m
-
•
•
Corrugated
JlO/ton
N
M
Materials
Market
Corrugated
$10/ton
•
m
m
m
m
BTU/
1b
Waste
7000
M
•
•
II
M
BTU/
Ib
Waste
4116
•
•
«
m
•
TPD
8
»
•
4
•
M
2.8
V
*
TPD
15.6
(10 TPD
MOO INC)
N
•
6.3
(4 TPD
MOD INC)
•
•
Generation
Rate
20 lbs/1000
SFGLA/day
25 lbs/1000
SFGLA/day
30 lbs/1000
SFGLA/day
20 lbs/1000
SFGLA/day
25 lbs/1000
SFGLA/day
30 lbs/1000
.SFGLA/day
20 lbs/1000
SFGLA/day
25 lbs/1000
SFGLA/day
30 lbs/1000
SFGLA/day
Generation
Rate
20 lbs/1000
SFGLA/day
25 lbs/1000
SFGLA/day
30 lbs/1000
SFGLA/day
20 lbs/1000
SFGLA/day
25 lbs/1000
SFGLA/day
30 lbs/1000
SFGLA/day
Waste Composition
Paper 801
Plastic 7. 51
Other 12.51
m
m
m
N
M
Corrugated 52X
•
Waste Cornposltion
Corrugated 52i
-
•
•
m
•
Recovery
Rate
24 hr/
day
•
•
8 hr/
day
•
•
70i
•
M
Recovery
Rate
24 hr/
day
70S
p
•
8 hr/
day
701
*
•
Existing
System
Cost/Day
$224
•
•
$112
m
m
$ 78
•I
II
Existing
System
Cost/ Day
$437
•
•
$176
•
P
New System
Net Cost/
Day
$220
•
m
$104
•
•
$ 78
N
»
New System
Net Cost/
Day
$404
•
N
$163
m
-
SFGLA: Square Feet Gross Leasable Area.
192
-------
TABLE 8. RESOURCE RECOVERY APPLICABLE TO OFFICE BUILDINGS
MODULAR
INCINERATION
Condition
16.000
employees
Condition 2
10,670
employees
Condition 3
8.000
employees
Condition 4
8.000
employees
Condition 5
5.350
employees
Condition 6
4.000
employees
SOURCE
SEPARATION
Condition 1
150
employees
Condition 2
100
employees
Condition 3
75
employees
MODULAR
INCINERATION
and SOURCE
SEPARATION
Condition 1
20,000
employees
Condition 2
13,335
employees
Condition 3
10.000
employees
Condition 4
9,400
employees
Condition 5
6.270
employees
Condition 6
4,700
employees
On-Slte
Energy
Use
Yes-
$1/10'
BTU
•
•
•
M
«
N/A
•
N
One-Site
Energy
Use
Yes-
$1/10'
BTU
•
•
•
•
•
Materials
Market
N/A
m
m
•
•
•
High-grade
paper-
$55/ton
•
•
Materials
Market
HI gh-grade
paper
$55/ton
•
•
*
•
•
BTU/
Ib
Waste
7000
•
•
N
H
•
BTU/
Ib
Haste
4480
•
•
•
m
N
TPD
8
•
M
4
•
«
.075
•
•
TPD
10
(7 TPO
HOD INC)
•
M
4.7
(3.3 TPD
HOD INC)
N
II
Generation
Rate
1 Ib/
employee/
day
1.5 Ib/
employee/
day
2 Ibs/
employee/
day
1 Ib/
employee/
day
1.5 Ib/
employee/
day
2 Ib/
employee/
day
1 Ib/
empl oyee/
day
1.5 Ib/
employee/
day
2 Ibs/
employee/
day
Generation
Rate
1 Ib/
employee/
day
l.S Ib/
employee/
day
2 Ib/
employee/
day
1 Ib/
employee/
day
1.5 Ib/
employee/
day
2 Ib/
employee/
day
Waste Composition
Paper 87 I
Plastic 1.5X
Other 14.51
•
M
•
•
N
High-grade paper 43X
•
«
Waste Composition
High-grade paper 43%
M
•
•
»
N
Recovery
Rate
24 hr/
day
•
8 hr/
day
»
«
70S
•
H
Recovery
Rate
24 hr/
day
701
m
m
8 hr/
day
70*
N
II
Existing
System
Cost/Day
$224
•
•
$112
•
m
$2.10
•
•
Existing
System
Cost/Day
$280
•
•
$132
•
p>
New Systen
Net Cost/
Day
$220
•
•
$104
-
•
$2.10
»
m
New System
Net Cost/
Day
$248
m
m
$118
-
•
193
-------
TABLE 9. RESOURCE RECOVERY APPLICABLE TO AIRPORTS
MODULAR
INCINERATION
Condition 1
80,000
passengers/day
Condition 2
40,000
passengers/day
Condition 3
13,350
passengers/day
Condition 4
32,000
passengers/day
Condition 5
10,000
passengers/day
Condition 6
5,350
passenger/ day
SOURCE
SEPARATION
Condition 1
Airfreight
area, 200
TPD Carqo
Condition 2
Aircraft
Maintenance
Base-1,150
employees
On-S1te
Energy
Use
Yes-
$1/10*
BTU
•
•
-
*
•
Materials
Market
Corrugated
$30/ton
•
BTU/
1b
Waste
esoo
M
•
M
-
M
TPD
10
*
•
4
H
n
.7
1.3
Generation
Rate
.25 lb/
passenger/
day
.5 lb/
passenger/
day
1.5 lb/
passenger/
day
.25 lb/
passenger/
day
.5 lb/
passenger/
day
1.5 lb/
passenger/
day
7 Ibs/ton
cargo/
day
2.2 Ibs/
employee/
day
Waste Composition
Paper 50} Organics 151
Plastic 101 Other 161
Hood 81
B *
H «
• m
m m
m m
Corrugated 321
Corrugated 361
Recovery
Rate
24 hr/
day
•
•
8 hr/
day
M
-
701
701
Existing
System
Cost/Day
$280
•
•
$112
1*
M
$ 9
$ 18
New Systen
Net Cost/
Day
$269
•
•
$107
•
•
$ 9
S 18
TABLE 10. RESOURCE RECOVERY APPLICABLE TO SMALL CITIES
SOURCE
SEPARATION
Condition 1
13,600
people
Condition 2
28,000
people
Condition 1
4,800
people
Condition 2
12,800
people
Condition 1
12.800
peopl e
Condition 2
25.600
people
Condition 1
11,200
people
Condition 2
21 ,600
people
On-S1te
Energy
Use
N/A
M
•
II
•
•
•
•
m
Materials
Market
Newspaper-
J40/ton
baled
M
Newspaper-
$20/ton
loose
m
Fe-$40/ton
AL-J340/ton
Glass-
J30/ton
•
Fe-$40/ton
AL-$340/ton
Glass-
$30/ton
Newspaper-
$40/ton/
baled
•
BTU/
lb
Waste
N/A
•
•
m
m
m
M
•
TPD
14
34
6
16
16
31
14
27
Generation
Rate
2.5 lb/
capital/
day
•
•
•
2.5 Ibs/
capital/
day
H
•
ft
Waste Corrposition
Newspaper 9S
•
•
•
Ferrous 91
AL 10S
Glass 131
m
m
M
Recovery
Rate
so:
301
501
301
501
301
501
301
Existing
System
Cost/Day
$ 29
$ 43
$ 5
$ 11
$102
$115
$116
$136
New Systen
Net Cost/
Day
$ 29
$ 43
$ S
$ 11
$102
$115
$116
$136
194
-------
TABLE 11. RESOURCE RECOVERY APPLICABLE TO PRISONS
MODULAR
INCINERATION
Condition 1
5.500
inmates
Condition 2
4,400
inmates
Condition 3
3,670
inmates
Condition 4
4,400
inmates
Condition 5
1,760
inmates
Condition fi
1,470
inmates
SOURCE
SEPARATION
Condition 1
750
inmates
Condition 2
600
i nmates
Condition 3
500
inmates
MODULAR
INCINERATION
and SOURCE
SEPARATION
Condition 1
2,700
inmates
Condition 2
2,160
inmates
Condition 3
1,800
inmates
Energy
Use
Yes-
$1/10'
BTU
M
N
•
M
M
N/A
•
N
On-Site
Energy
Use
Yes-
Jl/10'
BTU
N
II
Materials
Market
N/A
•
m
p
p
«
Corrugated
$307 ton
•
•
Materials
Market
Corrugated
$30/ton
•
•
BTU/
Ib
Haste
5SOO
II
N
•
•
•
m
p
M
BTU/
1b
Waste
3,960
M
•
TPD
11
M
p
4.4
P
M
1.5
N
•
TPD
5.4
*
m
Generation
Rate
4 Ibs/
inmate/
day
5 Ibs/
inmate/
day
S Ibs/
inmate/
day
4 Ibs/
inmate/
day
5 Ibs/
inmate/
day
6 Ibs/
inmate/
day
4 Ibs/
inmate/
day
5 Ibs/
inmate/
day
6 Ibs/
inmate/
day
Generation
Rate
4 Ibs/
inmate/
day
5 Ibs
inmate/
day
E Ibs/
inmate/
day
Waste Composition
Paper 59X Other 23X
Plastic 81
Orqanics 10J
« •
N •
» •
• H
K •
Corrugated 232
N
H
Waste Composition
Corrugated 23%
•
N
Recovery
Rate
24 hr/
day
•
•
8 hr/
day
M
P
sot
•
•
Recovery
Rate
8 hr/
day
802
p
•
Existing
System
Cost/ Day
$308
P
•
$123
•
•
$ 42
•
"
Existing
System
Cost/ Day
$151
•
•
New System
Net Cost/
Day
$284
•
•
$112
•
*
$ 42
m
M
New System
Net Cost/
Day
$120
m
m
TABLE 12. RESOURCE RECOVERY APPLICABLE TO GARDEN APARTMENTS
MODULAR
INCINERATION
Condition 1
11,000
tenants
'Condition 2
8.800
tenants
Condition 3
7,335
tenants
Condition 4
4,400
tenants
Condition 5
3.S20
tenants
Condition E
2,935 .
tenants
On-Site
Energy
Use
Yes-
$1/10'
BTU
N
•
P
N
•
Materials
Market
N/A
p
p
M
"
m
BTU/
Ib
Waste
5000
•
M
«
M
•
TPD
11
•
m
4.4
p
p
Generation
Rate
2 Ibs/
tenant/
day
2.5 Ibs/
tenant/
day
3 Ibs/
tenant/
day
2 Ibs/
tenant/
day
2.5 Ib/
tenant/
day
3 Ib/
tenant/
day
Waste Composition
Paper 351 Misc lit
Plastic 4.51 Other 22.51
Orqanics 27X
• •
• •
• •
• •
• M
Recovery
Rate
24 hr/
day
•
p
8 hr/
day
P
•
Existing
System
Cost/Day
$308
W
N
$123
p
•
New Systea
Net Cost/
Day
$293
m
m
$115
m
m
195
-------
TABLE 13. RESOURCE RECOVERY APPLICABLE TO HOSPITALS
MODULAR
INCINERATION
Condition 1
1,600
beds
Condition 2
1,067
beds
Condition 3
800
beds
Condition 4
640
beds
Condition 5
800
beds
Condition 6
533
beds
Condition 7
400
beds
Condition B
320
beds
On-Slte
Energy
Use
Yes-
$1/10*
BTU
•
•
•
•
•
•
m
Materials
Market
N/A
m
m
m
m
m
m
»
BTU/
Ib
Waste
7000
•
•
•
M
M
N
•1
TPD
B
•
m
m
4
•>
•
M
Generation
Rate
10 Ibs/
bed/day
15 Ibs/
bed/day
20 Ibs/
bed/day
25 Ibs/
bed/day
10 Ibs/
bed/day
15 Ibs/
bed/day
20 Ibs/
bed/day
25 Ibs/
bed/day
Waste Composition
Paper 401 Hisc 101
Plastic 15X Other 10*
Organic* 25%
• •
• N
• 1*
• •
• m
m •
• •
Recovery
Rate
24 hr/
day
•
•
•
8 hr/
day
M
P
•
Existing
System
Cost/ Day
S224
•
m
m
$112
•
•
•
Net* System
Net Cost/
Day
$220
M
H
N
$104
•
-
-
196
-------
COMPATIBILITY OF SOURCE SEPARATION AND MIXED-WASTE PROCESSING
FOR RESOURCE RECOVERY
HARRY H. FIEDLER
BELUR N. MURTHY
GILBERT ASSOCIATES, INC.
READING, PENNSYLVANIA
ABSTRACT
Source separation is a procedure by which materials may be recovered from a
waste stream and subsequently put to higher use. More specifically, this procedure
is commonly imposed upon municipal solid waste before treatment in a mixed-waste
processing facility.
This study examines the issues surrounding the compatibility of five source
separation scenarios upon four of the most frequently encountered mixed-waste
processes and suggests a methodology by which the characteristics and requirements
unique to a specific community may be used to arrive at the most compatible com-
bination of source separation and mixed-waste processing for that particular
community.
The compatibility evaluation is accomplished by quantifying, to the extent
possible, data pertaining to combinations of source separation and mixed-waste
processing in four major areas of concern - conservation, environmental, economic,
and institutional. The last area is primarily subjective due to its considerable
dependence upon non-quantifiable social and legal considerations.
The major areas of concern are then prioritized depending upon the character-
istics and requirements of the community. Following this, the most compatible
combination from each area of concern for the specific community is identified.
Finally, the prioritization considerations are applied to the identified most
compatible combinations to arrive at the source separation scenario which overall
is most compatible for the community.
INTRODUCTION
Every year, people in the United
'States discard 140 million tons of
residential waste. In so doing, a
number of environmental problems,
ranging from air and water pollution
to the removal of land from productive
use are created. Of the 3.7 pounds of
waste an average American discards
each day, about 85% could be used to
produce energy. Overall, about 25% could
be recovered as material suitable for
reuse or recycling.
Recognizing the energy potential of
this waste stream, the Environmental
Protection Agency (EPA) and other organ-
izations have attempted to initiate actions
- reducing quantity of waste directed to
landfill, recycling usable materials, and
combusting the organic constituents - to
197
-------
recover energy values.
Two principal technologies are
available to achieve this goal, namely,
source separation and mixed-waste pro-
cessing. The EPA has taken the position
that both technologies will be needed
for effective resource recovery, a
position which immediately raises the
question as to whether the two tech-
nologies are compatible. Compounding
this uncertainty, every community
which is interested in ascertaining
the most compatible combination of
technologies, has its own site specific
characteristics and requirements;
accordingly, the technology combination
which is most compatible for one
community is not necessarily the most
compatible for another.
In the following sections, source
separation and mixed-waste processing
technologies will be briefly reviewed.
Following this is a brief discussion
of the most important issues, results
of quantitative analysis, and con-
clusions. Thereafter, a method is
suggested by which communities can
determine the most compatible com-
bination of technologies. "Baselyn",
a hypothetical community, is used as
an example to test the methodology.
Finally, overall conclusions are
presented.
RESOURCE RECOVERY ALTERNATIVES
The alternatives available for
resource recovery can be classified
into source separation and mixed-
waste processing.
Source Separation
In source separation, the primary
responsibility for sorting and re-
covering materials lies with the waste
generator: home owners, businesses,
schools, etc. Currently, the most
appropriate waste materials recov-
erable through source separation are
aluminum, ferrous metals, paper, and
glass.
The common alternatives available for
source separation are:
o recycling centers
o separation of office paper
o separation of corrugated paper
o separate collection of newsprint and
other paper
o separate collection of various
materials
o beverage container deposits
In this study, a typical case of each
of these source separation alternatives,
except the recycling centers, has been
considered for developing and evaluating
the resource recovery compatibility assess-
ment method. The recycling centers were
excluded because the total impact of such
a system on the waste flow of the community
is relatively small, and the cost and
resources invested in such a system are
far greater than for the other methods.
Significant features of the five
source separation scenarios considered are
shown in Exhibit 1. Curb-side collection,
handling and disposal of separated wastes
are considered to be similar to the method
practiced at Somerville and MarbleheadO).
The recovered materials are sold to an
intermediate processor for sorting,
packaging, and resale to manufacturers,
Mixed Haste Processing
Modern mixed-waste processing facil-
ities are complex and capital intensive.
To reduce amortization costs per unit of
municipal solid waste, a long-term commit-
ment is necessary. Sophisticated planning,
management and marketing are required. Most
such facilities recover ferrous metal and
energy in the form of refuse derived fuel
(RDF) or steam. Some facilities also
recover glass, primarily with trommels.
RDF is usually recovered in modern fac-
ilities by air classification.
198
-------
In this study, the following mixed-
waste processes have been selected as
typical commercial alternatives:
o unprocessed waterwall combustion
and ferrous recovery (UWCF)
o processed water-wall combustion
and ferrous recovery (PWCF)
o refuse derived fuel production
and ferrous recovery (RDFF)
o modular incineration (MI)
Significant features of these
alternatives are presented in Exhibit
2. More detailed descriptions of the
mixed-waste processes are available
in the literatureU-4).
"Baselyn". a Hypothetical Community
In order to assess the relative
compatibilities of various source
separation scenarios with commercially
available mixed waste processing
alternatives, a nationally typical
hypothetical community, Baselyn, has
been considered!4).
Baselyn has 108,000 inhabitants,
with an average population density of
1,930 per square mile, and is located
near a major metropolitan area.
There are approximately 24,000
single family homes in Baselyn. Real
cities whose population and density
statistics are similar to those of
Baselyn include Pasadena, California;
Lakewood, Colorado; Waterbury, Conn-
ecticut; Hollywood, Florida; New Bed-
ford, Massachusetts; Ann Arbor, Michigan;
Woodbridge, New Jersey; Albany, New
York; and Canton, Ohio. Approximately
60% of the families own their own home.
The median income is $12,000 per year
and the median education level is 12.4
years.
The city's sanitation department
collects solid waste from all house-
holds in Baselyn. Prior to the source
separation program, the department employed'
60 people and 10 packer trucks (capacity:
20 yd-* each) for waste collection. Each
truck was operated by a crew of three.
In addition, the department employed 60
people for administration, maintenance,
and other duties. Collection and disposal
costs are approximately $5 per household,
per month.
Waste is collected at curbside once
a week. Each resident generates 3.7 Ib
of waste per day (the national average),
so the department collected 200 tons per
day before the source separation program
began. Exhibit 3 presents the composition
of solid waste in Baselyn.
The collection trucks unload at a
transfer-station located within the city
limits. This station, operated by two
people, has a compactor that loads the
waste into 80 yd3 trailers. From the
transfer station, waste is hauled to a
sanitary landfill that the county main-
tains in ;a rural area, 25 miles away.
Baselyn uses approximately 7 acres per
year, and over 150 acres are available
for landfill.
Commercial establishments must
contract with private companies for the
collection of their waste. This waste
is delivered directly to the county
landfill. The county charges the city
and private companies a tipping fee of
$12 per ton.
Ten miles from Baselyn, there is a
materials processor willing to buy paper,
glass, and cans. Current prices are (FOB):
Newspaper $30/ton
Corrugated paper ISO/ton
High grade paplr $60/ton
Mixed glass and" cans $10/ton
The mixed-waste processing facility
is assumed to have a capcity of 1,000 tons/
day, enough to process the combined wastes
from Baselyn and four additional communities
of similar characteristics and size.
199
-------
The effect of the various source
separation scenarios on the quantities
of materials recovered, before the
solid wastes are sent for mixed-waste
processing, is shown in Exhibit 4.
The Baselyn area meets the exist-
ing national ambient air quality stan-
dards.
Fixed and Expanded Service Areas
Since source separation results
in decreased waste quantities entering
the mixed-waste processing facility,
it may be economically desirable to
make up the shortage of waste flow
with mixed solid wastes collected from
an area outside the five communities,
each of which is considered similar to
Baselyn. The effect of including such
an expanded service area on the overall
economics of waste disposal is com-
pared with those of a fixed service
area in this study.
The concepts of fixed and ex-
panded service areas are graphically
represented in Exhibit 5.
COMPATIBILITY: AREAS.OF CONCERN
The entire concept of compat-
ibility hinges upon whose viewpoint
is being examined. As such, the
issues of greatest importance differ
according to the viewpoint of the
observer, be it the plant operator,
municipality, or the nation as a whole.
For example, the plant operator is most
interested in knowing whether source
separation scenarios will decrease his
profitability or increase the amount
of pollutants released to the at-
mosphere. The municipality, on the
other hand, is most concerned about
the cost for disposal of the commun-
ity's solid waste as well as the
future availability of proper land-
fill capacity. Still other issues
are of prime importance to the nation
as a whole, namely, the depletion of
natural resources and net energy
consumption. This list of sample
issues is not complete; however, it
does identify some of the most important
ones. Furthermore, some of the issues
are composites of a number of sub-issues.
For example, plant operation profitability
is dependent upon the institutional issues
of assured supply of solid waste, con-
tractual determination of tipping fee,
and assured revenues from saleable pro-
ducts. The following quantitative
analysis of the combinations of technology
in the conservation, environmental and
economic areas of concern are necessary to
determine the impact of each of the tech-
nology combinations upon an individual
issue. Since the institutional issues
can not be quantified, they will be
treated subjectively.
Conservation
The effect of each source separation
scenario on the net energy recoverable from
the resource recovery system in the fixed
or expanded service area has been cal-
culated from the equation:
°?E = (ES +£EM - EC)/EMSW
where ?E = net energy recovery efficiency
of the resource recovery system.
Es = energy recovered ultimately as
steam from the mixed waste
processing facility.
Em = sum of the net energies con-
served by recycling instead of
manufacturing from original raw
materials - the materials re-
covered from source separation
and mixed waste processing.
EC = energy consumed by the resource
recovery system, i.e., mixed-
waste processing facility
operation, collection and trans-
portation of source separated
and mixed wastes.
= energy content of the municipal
solid wastes being collected in
the service area.
200
-------
The effect of source separation
on the energy recoverable is shown in
Exhibit 6 as steam from the mixed-waste
processing unit.
The overall energy balance is re-
presented by Exhibit 7. In this
exhibit, the quantity of material or
energy in the various exit streams
will vary depending on the particular
combination of source separation and
mixed-waste processing being studied,
and will be zero in some cases. Since
the average composition of the mixed
waste and that of the stream entering
the mixed-waste processing facility are
the same in both fixed and expanded
areas,"7£ for any SS and MWP combina-
tion remains unchanged for the two
areas. The results, shown in Exhibit
8, indicate, in general, that beverage
container recovery and high multi-
material source separation contribute
most to energy conservation.
Environmental Effects
Two factors are considered within
the environmental area of concernr
residual loadings to landfill and air
emissions.
Residual to landfill is a direct
measure of the reduction in tons of
residual directed to final disposal as
a result of each source separation and
mixed-waste processing combination.
Air emissions are determined by
using coefficients for each option
that reflect the quantity of pollutants
emitted to the atmosphere for each ton
of waste processed. Calculations are
made for the following pollutants:
sulfur oxides (SOx), nitrogen oxides
(NOx), particulates, and hydrocarbons.
1) Residual to landfill
The total quantity of waste re-
maining after each source sep-
aration program is the amount
to be directed to landfill or to
the mixed-waste processing
facility. Residual to land-
fill (Exhibit 9) may vary by a
factor of 1.8, from 135 to 239 tons.
Since the residuals are so much
smaller - both in volume and mass -
than the inputs, any of the mixed-
waste processing systems yields a
substantial reduction in landfill
requirement.
2) Air emissions
The following assumptions were used
in analyzing the emissions:
o Particulates produced in pre-
or post-processing of the waste
or residual is insignificant com-
pared to that produced in pro-
cessing and will be captured in
a baghouse or recycled to the
process in the air input steam.
o The excess air assumed in burning
and the standard cubic feet of
flue gas produced per pound of
waste processed were taken to be
as follows:
Technology
Option
Unprocessed
waterwall
Processed
waterwall
RDF
Modular
incinera-
tion
Percent
Excess Air
75
50
50
50
scf/lb
75
65
65
65
The air residuals were ratioed
on the basis of heat inputs, which
were taken as follows:
Source Separation
Multimaterial, high
Multimaterial, low
Newsprint, high
Newsprint, low
Beverage containers
Btu's/Day
7.65 x
8.08 x
8.48 x
8.96 x
9.19 x
109
109
109
201
-------
The results of Exhibit 10 indicate
that source separation scenarios will
result in negligible production of
particulates, S02, NOX and hydrocar-
bons.
For large sources, an applicable
EPA standard for particulates may be
0.1 lb/106 Btu's, which is equivalent
to 0.46 tons/day in the present case.
Hence, all mixed-waste processing
options may require particular emission
controls, with the two incinerator
types requiring the most control.
Emissions of S02 are not gen-
erally a problem, due to the low
sulfur content of the waste. Minimal
cleanup, such as wet particulate re-
moval, should suffice for control.
Alternatively, one may conclude that
the figures are sufficiently close to
the limit so that the errors make the
necessity of control problematical.
Probably, no control will be needed
for S02 to meet legal standards.
The EPA limit on NOX, which is
0.7 lb/!Q6 Btu and which may not apply
to these processes, calculates to 3.22
tons/day.
There does not appear to be any
relevant national hydrocarbon
standard.
In general, only particulate
emissions were high enough to present
potentially significant problems for
all combined alternatives involving
mixed-waste processing facilities.
High newsprint separation at the source
produces a significant reduction in
particulate emissions for all options.
For other pollutants, emissions are
generally low and source separation
produces only a small change in the
ratings.
Economi c
The economic effects of imposing
source separation on a mixed-waste
processing facility were assessed by cal-
culating the total costs of processing
municipal solid waste in fixed and ex-
panded service areas. The total cost was
determined by subtracting revenues from
capital and operating costs. Municipal
ownership of the processing facility via
General Obligation Bonds was assumed.
The following equation was used to compute
the total cost:
CT =((C
ss
(C.
CM H- GU)
) - (Rp +
RF + RG + RA + RS + RR}
where,
C = Collection and transportation
costs of source separated materials
CMWP = Collection and transportation costs
of charge stock to the mixed-waste
processing facility
C. = Labor cost
C«Q = Administrative and overhead costs
C,. = Cost of materials including ex-
pendable supplies and equipment
replacement
C = Utility costs
"TL
'DL
1CT
= Transportation cost of material to
landfill
= Dump cost at the landfill (tipping
fee)
- Interest on capital invested during
construction
= Debt service charges
= Amortized start-up cost
202
-------
R
p = Revenue derived from newsprint
sales
Rp = Reveue derived from ferrous
material sales
RQ = Revenue derived from glass
sales
RA = Revenue derived from alumi-
num sales
RS = Revenue derived from steam
sales
RR = Revenue derived from refuse-
derived fuel (RDF) sales
All costs were converted to mid-
1977 dollars by using escalation or
de-escalation factors and the Chemical
Engineering (CE) plant index. An
annual inflation rate of 6% was
assumed. The annual charge for prin-
cipal and interest is based on a 20
year life of all the facilities with
a 1 year actual construction/start-up
period; therefore, the bond issue is
for a period of 21 years. We assume
a sinking fund is established with
21 equal payments at an interest rate
of 7 percent, and the bond principal
is repaid at the end of the 21 year
period. Costs of $25.85/ton, $5.00/
ton and $12.00/ton were used for
ip> Cj, , and CD-, respectively.
rYous metal sales were set at
$39/ton.
The results of the calculations,
shown in Exhibits 11 and 12, indicate
that high multi-material separation
results in the lowest processing cost.
Institutional
1) Contracts:
Well-written contracts are the
means by which potential adverse
effects of source separation can
be mitigated for a plant operator.
The underlying need for the
contracts is not only maintenance
of profitability for the operator,
but also assurance to the capital
investors that their investment
will be appropriately repaid.
Maintenance of profitability is
predicated upon an assured supply
of solid waste, tipping fee and
acceptance of recycled products at
a specific price. Contractual non-
compliance of any of these elements
can seriously jeopardize, if not
defeat, economic viability of the
operation. Furthermore, the
investors must have a reasonable
assurance that profitability will
be at least adequate to cover prin-
cipal and interest payments over the
life of the project, or else they
will not invest the capital. Since
investment degree of risk is directly
related to interest rate, a proposed
facility without ironclad contractual
safeguards against potential adverse
economic effects resulting from source
separation may still be financed. One
method involves paying a higher in-
terest rate to compensate for the
higher risk, although this method is
not without limitations. Another
method involves issuance of Municipal
Bonds - Revenue or General Obligation.
With the former, the facility revenue
is the source of the principal and
interest payments with the interest
rate dependent upon the degree of
risk. For the latter, the full
faith and taxing power of the mun-
icipality stands behind the principal
and interest payments. As may be
expected, General Obligation Bonds
command a lower interest rate than
Revenue Bonds.
2) Facility Maintenance
Facility maintenance is directly re-
lated to the quantity of glass con-
tained within the solid waste stream
charged to the processing facility!5'
Although this factor is a technical
one, which directly affects operating
costs, it is included within the
institutional area due to the unavail-
ability of reproducible quantitative
data.
203
-------
3) Employment
A low unemployment rate is
desirable for all communities.
Implementation of source sep-
aration will require addition-
al personnel, estimated at 10-
20% depending upon the scenario
elected and degree of adminis-
trative control. In descending
order, source separation per-
sonnel requirements are multi-
material, beverage contain-
ers and newsprint only.
Implementation of source sep-
aration is not expected to
significantly affect a plant
operator's labor force.
4) Community Well-Being
The issue.of whether a combin-
ation of technologies is com-
patible with the concept of
community well-being is somewhat
nebulous. Community well-being,
when expressed as the reduc-
tion in community expense for
solid waste disposal and a
contribution to national re-
cycling efforts, is indeed
enhanced by implementation of
resource recovery methods. The
descending order of compatibili-
ty for the community, and the
plant operator's position are the
same as for the preceding "Employ-
ment" sub-section.
METHODOLOGY FOR ASSESSING COMPATIBILITY
The objective of this section is
to develop a methodology that an in-
terested party can use to determine
which combination of source separation
and mixed-waste processing is most
compatible for their particular needs.
An example is given to illustrate the
use of the methodology.
Methodology
1) Order of Priority
Determine the relative importance
existing between the conservation,
environmental, economic and insti-
tutional areas of concern.
2) Combination Selection
Select from each of the four areas
of concern the most compatible com-
bination of source separation and
mixed-waste processing as measured
by the quantitative evaluator for
the area, viz.,
Conservation - % efficiencies
Environmental - tons/day
Economic - $/ton
Institutional - Subjective
3) Compatibility Determination
Apply the priorities established in
(1) above to the combinations sel-
ected in (2) above. This will result
in weighted combinations from which
the overall most compatible combina-
tion will be selected.
Sample Calculation
"Baselyn", the hypothetical community
previously described, is the basis for the
sample calculation. To simplify the analy-
sis, it is assumed that Baselyn already has
an unprocessed waterwall combustion facility,
complete with ferrous recovery, and that
only a fixed service area is available.
Any other mixed-waste processing facility
could have been selected for this example
because no particular type of facility is
favored in this study.
1) Order of Priority
Determination of the order of import-
204
-------
ance of the four areas of concern
is shown in this section. The
community decision makers nave
decided that conservation is of
little concern to them and should
have little bearing in the final
analysis.
Adequate landfill space - good for
approximately 21 years with the
current population - is available,
national ambient air quality
standards are easily met, and
no prohibitory landfill bills
are anticipated. Taking these
factors into consideration, it
was concluded that environmen-
tal considerations are slightly
greater than the level of im-
portance given conservation.
The community or a private oper-
ator may own and operate the
mixed-waste processing facility.
In either case, firm contractual
commitments are very important.
The people have a sense of com-
munity responsibility and feel
that the project's success is
important. Although the un-
employment rate is close to the
national average, the community
would be pleased if it were less.
Based upon these facts, institu-
tional considerations are felt
to be important and should re-
ceive more recognition in the
final analysis than either con-
servation or environmental con-
siderations.
The national economic pinch is
hurting this middle class com-
munity; accordingly, the
greatest importance has been
assigned to the lowest cost
overall method of solid waste
disposal.
2) Combination Selections
The conservation selection is
accomplished by referring to
Exhibits 6 and 8 and selecting
the best of the source separa-
tion scenarios from the column
representing the unprocessed water-
wall combustion process. Doing this,
the beverage container scenario turns
out to be the best for both the per-
cent Btu recovery as steam and per-
cent net energy efficiency with 65%
and 73%, respectively.
Following the same procedure with
Exhibit 9, the high multi-material
scenario has the best rating with
the lowest residual to landfill figure
of 135 tons/day. From Exhibit 10,
four values - particulates, sulfur
dioxide, oxides of nitrogen and hydro-
carbons - are shown, representing
tons/day of pollutants discharged to
the atmosphere. The data shows that
the high multi-material scenario has
the best rating for all four residuals.
Therefore, for the environmental area
of concern, high multi-material is
the best choice.
The economic choice is obtained by
referring to Exhibit 11 and selecting
the lowest dollars/ton cost. High
multi-material recovery is the choice
at $26.30.
The institutional selection is more
difficult due to its subjective
nature. If the contracts are well-
written and result in as much pro-
tection as possible with appropriate
relief upon the event of non-compli-
ance, there is no real distinction
between the impacts of the source
separation scenarios. However, con-
sidering the factor of degree of
wear in the mixed-waste processing
facility due to glass, it is known
with subjective certainty that the
least quantity is desired.(2) There-
fore, for the contracts portion of
the institutional area of concern,
if ironclad contracts are broken and
commensurate relief is subsequently
granted, the fact that the beverage
container scenario prompted the ab-
rogation would result in the greatest
benefit to the facility operator.
The interests of reduced unemploy-
ment and community well-being are
best served by the high multi-mater-
ials scenario due to the greater
205
-------
number of people employed and in
the quantity of useful material
recovered. Since the employ-
ment and well-being issues are
somewhat overlapping, their
average result is given approxi-
mately the same weight as for the
contractual obligation. The
institutional area best choice is
therefore split evenly between
the beverage and high multi-
material scenarios.
3) Compatibility Determination
This is the point where "it's
all put together". Referring
to the following summary table,
high multi-material recovery is
the most compatible source sep-
aration scenario to combine with
the unprocessed waterwal1 com-
bustion process in Baselyn.
SUMMARY TABLE
Compatibility Determination Matrix
Area
of
Concern
Conser-
vation
Environ-
mental
Economic
Institu-
tional
Priority
Least im-
portant
3rd most
important
Most
important
2nd most
important
Most Compatible
Source Separa-
tion Scenario
Beverage con-
tainer
Multi-material,
high
Multi-material,
high
Equally divided
between high
multi-material
and beverage
container
Although this selection was relatively
clear-cut in Baselyn due to the pre-
ponderance of the high multi-material
scenario in the highest priority rank-
ings, it could be more difficult in
other cases. In such an event, num-
erical ratings applied as priorities
may be useful. Using Baselyn as a
simplified example, if a total impor-
tance of 100% is attributed to the
priority associated with the four
areas of concern, then 40% thereof could
be considered a reasonable quantitative
weight of the "most important" priority
applied to the economic area. Following
this line of reasoning, 30% may be
attributed to the industrial area, 17%
to environmental, and 13% to conservation.
Applying these weights to each of the
individual source separation scenarios in
the third column of the summary table
and adding the results for the like
scenarios, high multi-recovery would be
the most compatible with 72% out of 100%
- the same result obtained subjectively.
CONCLUSIONS
Selection of the optimum resource
recovery system for a community is
currently, at best, a most difficult det-
ermination. The methodology suggested in
this paper provides a way in which one of
the most important components - compati-
bility of source separation and mixed
waste processing - of the optimum sel-
ection process may be accomplished.
Application of this methodology to a
hypothetical community "Baselyn" has
resulted in the conclusion that high
multi-material recovery is the most
compatible source separation scenario to
combine with an unprocessed waterwall
combustion facility with ferrous recovery
in a fixed service area of a community
with the characteristics, requirements
and priorities of Baselyn.
The authors recognize that the current
state of the art of U.S. resource recovery
systems is relatively unsophisticated;
however, they hope that their efforts will
be of assistance to the decision-makers
attempting to optimize the implementation
of resource recovery systems most respon-
sive to the needs of American communities.
206
-------
ACKNOWLEDGEMENT
This study has been sponsored by the U.S. Environmental Protection Agency under
Contract No. 68-03-2645
References
1. U.S. EPA, Office of Solid Wastes, "Fourth Report to Congress: Resource
Recovery and Waste Reduction", EPA Report No. SW-600, August 1, 1977.
2. Wilson, E. M., et al., "Engineering and Economic Analysis of Waste to
Energy Systems", The Ralph M. Parsons Co., Report to U.S. EPA under
Contract No. 68-02-2101, June 1977.
3. Jones, J., "Converting Solid Wastes and Residues to Fuel", Chemical
Engineering, 85(1), p 87-94, 1978
4. Fiedler, H., et al, "Compatibility of Source Separation and Mixed-Waste
Processing for Resource Recovery", Gilbert Associates and Resource
Planning Associates, Report being prepared for EPA under Contract No.
68-03-2645, 1979.
5. "Resource Recovery", Solid Wastes Management, 1979 Sanitation Industry
Yearbook, 16th Edition, p 46, December 30, 1978.
207
-------
EXHIBIT 1
SOURCE SEPARATION SCENARIOS
Case
Features
1. Multi-material
Recovery, High
Multi-material source separation and high-grade office
paper and corrugated paper separation. Waste is
separated into: 1) mixed paper; 2) clear glass and
cans mixed; 3) colored glass and cans mixed; and 4)
remaining wastes. Commercial office and corrugated
papers are separately collected and sold to manu-
facturers. Recovery efficiency ranges from 12 to
60 percent of each type of material in the waste
stream.
2. Multi-material
Recovery, Low
3. Newsprint,
Recovery, High
Multi-material separation of homeowner waste with a
lower recovery of paper, glass and metals. No
recovery of office or corrugated paper in effect.
Waste is separated into: 1) mixed papers; 2) mixed
bottles and cans; and 3) remaining wastes. The
overall recovery efficiency ranges from 12 to 42
percent.
Separate collection of newsprint only. Approximately
60 percent of newsprint in the solid waste stream is
recovered.
4. Newsprint,
Recovery, Low
Separate collection of newsprint only; approximately
20 percent of newsprint is recovered.
5. Beverage Container
Recovery
A beverage container deposit system imposed through
legislation. Residents return containers directly
'to retail stores, or refund facilities. Approximately
90 percent of beverage containers are recovered. The
existing community arrangements for collection and
disposal of the remaining mixed municipal wastes
are not affected.
208
-------
EXHIBIT 2
MIXED-WASTE PROCESSING ALTERNATIVES
Alternative
Features
1.
Unprocessed Water-wall
Combustion and Ferrous
Recovery
2.
Processed Waterwall
Combustion and
Ferrous Recovery
3.
Refuse Derived
Fuel (RDF) Production
and Ferrous Recovery
4. Modular Incineration
Mass burning of collected mixed waste, usually in a
thick bed (up to 34') on a moving grate. Magnetic
separation of ferrous material after incineration.
Metallic tubes in the walls of the incinerator carry
water which is converted to steam. Capacities range
from 50 to 1,200 tons/day of mixed solid waste.
Residue after incineration is usually sent to a land-
fill. Process requires additional pollution control
equipment. Examples: Chicago, Illinois; Harrisburg,
Pennsylvania; Nashville, Tennessee; Saugus, Massachusetts.
Mechanical processing of the mixed waste to concentrate
the combustible fraction and to reduce particle size.
Ferrous material is magnetically separated from the
non-combustible portion of the waste before incineration.
Shredded and classified waste is introduced into the
furnace by mechanical or pneumatic feeder. Burning
waste is semisuspended and falls on a traveling grate
to complete combustion in a thin bed. Steam recovery
is through the water walls. Capacity range: 50 to
1,200 tons/day. Additional pollution control equipment
is required. Examples: Norfolk, Virginia; Akron,
Ohio; Oceanside, New York.
Shredding, tromelling, air classification and magnetic
separation results in separation of a combustible
portion that is either pelletized, briquetted,
extruded, or directly used as fuel in either a dedi-
cated or auxiliary boiler. RDF may be formed wet
or dry. Examples: RDF dedicated - Ames, Iowa;
Akron, Ohio; RDF - auxiliary - St. Louis, Missouri;
Milwaukee, Wisconsin; Bridgeport, Connecticut;
Washington, D. C. (NCRR); Wet RDF (Black-Clawson) -
Franklin, Ohio; and Hempstead, New York. Capacity
range: 100 to 2,000 tons/day. Minimum pollution
control equipment required.
Module capacities: less than 50 tons/day of mixed
waste. Minimum pollution control equipment is
required. Batch and continuous flow units are
available. Preprocessing of mixed waste is not
required. Coils in the incinerator generate steam.
Economically suitable for smaller communities (800
to 200,000 population). Modules provide flexibility
for expansion. Example. Siloam Spring, Arkansas.
209
-------
EXHIBIT 3
COMPOSITION OF SOLID WASTE IN BASELYN
Component
Percentage
Quantity Collected Daily - tons
PAPER
Newsprint
Office
Corrugated
Others
38.9
7.5
3.5
n.o
16.9
77.8
15.0
7.0
22.0
33,8
GLASS
Beer and soft drink
Other
9.8
5.0
4.8
19.6
10.Q
9.6
METAL
Ferrous
Beer & soft drink
Other
Non-ferrous
Beer -4 soft drink
Other
4.9
9.8
4.1
0.8
1.0
3.1
0.5
0.3
8.2
1.6
2.0
6.2
1.0
0.6
REMAINING WASTE*
46.4
100.0
92.8
200.0
* Includes organic materials, wood, plastics, clothing, and other non-durable goods.
210
-------
EXHIBIT 4
SOURCE SEPARATION MATERIALS RECOVERY
Case
1. Multi-material
recovery, high
Source Separated Material* in Tons
per day and Percent of Total Waste Generated
Newsprint
and Other
Paper
Tons %
Corrugated
Tons %
Office
Tons %
Glass and
Metals
Tons %
Total Remaining
Recovered Mixed
Waste Waste
Tons % Tons 3
13.4 6.7 5.5 2.8 1.8 9.0 14.4 7.2 35.1 17.6 164.9 82.4
2. Multi-material
recovery, low
8.3 4.2
5.4 2.7 13.7 6.9 186.3 93.1
3. Newsprint recovery, 9.0 4.5
high
9.0 4.5 191.0 95.5
4. Newsprint recovery, 3.0 1.5
low
3.0 1.5 197.0 98.5
5. Beverage
11.7 9.0 11.7 9.0 188.3 91.0
* Total waste generation in Baselyn is 200 tons per day.
-------
EXHIBIT 5
FIXED AND EXPANDED SERVICE AREAS
5 COMMUNITIES
1,000 TONS/DAY WASTE GENERATED
^
-^
1
mmm^^^m
2
3
4
5
BASELYN _
1,000 1
800 T/D
•ONS/DAY CAP/I
' MIXED
WASTE
PROCESSING
FACILITY
' SOURCE SEPARATED
MATERIALS
200 TONS/DAY
FIXED SERVICE AREA
50 T/D
SOURCE SEPARATED MATERIALS
ADDITIONAL COMMUNITIES
5 COMMUNITIES
1,000 TONS/DAY WASTE GENERATED
BASELYN
1
800
T/D *
'1
200 T/D
MIXED WASTES
MIXED WASTES
MIXED
WASTE
PROCESSING
FACILITY
1,000 TONS/DAY
CAPACITY
250 T/D
EXPANDED SERVICE AREA
212
-------
EXHIBIT 6
CONSERVATION - ENERGY RECOVERY AS STEAM, I
(Fixed or Expanded Service Area)
Mixed Waste Processing Alternative
Source Separation
Scenario: UWCF PWCF RDFF MI_
Multi-Material, High 54 49 49 54
Multi-Material, Low 57 52 52 57
Newsprint, High 60 54 54 60
Newsprint, Low 63 58 58 63
Beverage Containers 65 59 59 65
213
-------
EXHIBIT 7
OVERALL ENERGY BALANCE
TRANSPORTATION
TRANSPORTATION
ho
MUNICIPAL
SOLID >•
WASTE
SOURCE
SEPARATION
UTILITIES
4--1-1-
NEWSPRINT,,
FERROUS
MATERIAL i -
GLASS
ALUMINUM
INTERNAL ENERGY
GENERATION
MIXED
WASTE
PROCESSING
FACILITY
NEWSPRINT , r
FERROUS
MATERIAL < >
GLASS , ,
ALUMINUM "
LANDFILL
|
STEAM
RDF
-------
EXHIBIT 8
ENERGY CONSERVATION - NET ENERGY RECOVERY EFFICIENCY, %
(Fixed or Expanded Service Area)
Source Separation
Scenari q:
Multi-Material, High
Multi-Material, Low
Newsprint, High
Newsprint, Low
Beverage Containers
Mixed Waste Processing Alternative
UWCF PWCF RDFF MI
70
67
68
69
73
65
63
62
64
67
65
63
62
64
67
65
61
61
61
67
215
-------
EXHIBIT 9
ENVIRONMENTAL EFFECTS - RESIDUALS TO LANDFILL,
(tons/day for 1,000 tons of mixed waste generated)
Mixed Haste Processing Alternative
Source Separation
Scenario: UWCF PHCF RDFF MI
Multi-Material, High
Multi-Material, Low
Newsprint, High
Newsprint, Low
Beverage Containers
135
171
196
196
147
135
171
196
196
147
135
171
196
196
147
161
208
238
238
180
216
-------
EXHIBIT 10
ENVIRONMENTAL EFFECTS - EMISSIONS TO AIR,
TONS/DAY FOR 1,000 TONS OF MIXED WASTE GENERATED
Mixed Waste Processing Alternative
Source Separation
Scenario:
Multi-Material, High (a)
(b)
(c)
(d)
Multi-Material, Low
Newsprint, High
Newsprint, Low
Beverage Containers
(a)
(b)
(c)
(d)
(a)
(b)
(c)
(d)
(a)
(b)
(c)
(d)
(a)
(b)
(c)
(d)
UWCF
1.51
0.42
1.16
0.03
1.60
0.45
1.23
0.04
1.68
0.47
1.29
0.04
"1.77
0.50
1.36
0.04
1.82
0.51
1.40
0.04
PWCF
0.50
2.57
1.06
1.12
0.53
2.71
1.23
1.19
0.55
2.69
1.29
1.24
0.58
2.76
1.36
1.31
0.60
2.80
1.40
1.35
RDFF
0.47
2.57
1.00
1.25
0.49
2.71
1.05
1.32
0.52
2.69
i.n
1.38
0.55
2.76
1.17
1.46
0.56
2.80
1.20
1.50
MI
1.25
1.62
1.08
1.12
1.32
1.71
1.14
1.19
1.38
1.80
1.20
1.24
1.46
1.90
1.27
1.31
1.50
1.95
1.30
1.35
(a) Particulates
(b) S02
(c) NOX
(d) Hydrocarbons
217
-------
EXHIBIT 11
ECONOMIC EFFECTS - DISPOSAL COST,
$/TON OF MIXED WASTE GENERATED
(Fixed Service Area)
Source Separation
Scenario:
Mixed Waste Processing Alternative
UWCF
PWCF
RDFF
MI
Multi -Material, High
Multi -Material , Low
Newsprint, High
Newsprint, Low
Beverage Containers
26.3
27.3
28.6
28.8
27.7
39.0
40.1
41.6
42.0
41.5
39.0
40.1
41.6
42.0
41.5
42.3
44.1
45.9
46.4
45.3
218
-------
EXHIBIT 12
ECONOMIC EFFECTS - DISPOSAL COST,
$/TON OF MIXED WASTE GENERATED
(Expanded Service Area)
Mixed Waste Processing Alternative
Source Separation
Scenario: UWCF PWCF RDFF MI
Multi-Material, High
Multi -Material , Low
Newsprint, High
Newsprint, Low
Beverage Containers
26.1
27.2
28.6
28.8
27.8
38.6
39 .-8
41.1
42.0
41.5
38.6
39.8
41.1
42.0
41.5
42.2
44.1
45.9
46.5
45.3
219
-------
FORECASTS OF THE QUANTITY AND COMPOSITION OF SOLID WASTE
Ralph M. Doggett
International Research and Technology Corporation
McLean, Virginia 22102
Andrea L. Watson
Bechtel Corporation
San Francisco, California 94119
ABSTRACT
Five scenarios of future solid waste generation have been developed using a method-
ology linked to an interindustry forecasting model. The estimated quantity and composi-
tion of solid waste in the future is affected by key model parameters. The purpose of
this paper is to indicate the impacts of technology change, materials substitution,
resource recovery initiatives, and average product lifetime on the solid waste load
within the context of the model. The forecasts are made for the year 1990.
INTRODUCTION
Factors affecting the quantity and
composition of future solid waste streams
include technology change, materials subs-
titution, product lifetimes and the extent
of resource recovery. These considera-
tions are the basis of five scenarios of
future solid waste generation that have
been developed using an interindustry
forecasting model of the U.S. economy.
The economic model referred to here is the
Interindustry Forecasting Model of the
University of Maryland (INFORUM).
The starting point for the analysis
is a data base of historical material
flows, for each of 14 materials to 21
product categories. Each product category
has an estimated average lifetime, after
which the product enters the waste stream.
Prompt scrap and other diversions are
accounted for, as is the extent of recyc-
ling. In the forecast mode, the data base
is linked to the economic forecasting model,
which employs an input-output table to
estimate the future demands for each
product, and the flows of materials to
these products, out to the year 1990.
In this paper, the assumptions and
results of each of five scenarios are
presented, beginning with a reference
case, or business-as-usual scenario, and
followed by discussions of scenarios in
which key parameters affecting technology
change, materials substitution, product
lifetime and resource recovery, have
been changed.
THE REFERENCE SCENARIO
An economic growth scenario that
includes a continuation of historical and
current trends in technology change and
materials substitution, serves as a
reference case for our analysis. The
economy, represented by real Gross
National Product, grows at a long-term
average rate of 3.4 percent per year,
while population growth slows to 0.9 per-
cent per year over the forecast period.
The scenario incorporates a significant
amount of materials substitution by
industry. Most notably, transportation
equipment manufacturers are projected to
use significant amounts of aluminum and
plastics in place of steel. Glass and
steel are largely replaced by plastics,
paper and aluminum in the beverage con-
tainers industry, and more and more plas-
tics replace paper in packaging.
Recycling rates for 1990, as shown
in Table 1, are fairly optimistic; 29 per-
cent of all paper entering the waste
stream is recycled, 58 percent of the
aluminum, 62 percent of the steel, and 6
220
-------
TABLE 1. REFERENCE SCENARIO, POST-CONSUMER RECYCLING ASSUMPTIONS0 } 1971 and 1990
Material
Paper and Paperboard
Glass
Metal
Ferrous
Aluminum
Other
Plastics
Rubber and Leather
Texti 1 es
Wood
Concrete
Total Non-Food
Food Waste
Total Product Waste
Yard Waste
TOTAL
Percent
1971
23.2
9.9
52.5
52.8
44.8
51.1
6.7
5.0
56.7
18.8
0.0
27.7
0.0
25.6
0.0
23.7
Recovered
1990
29.0
12.3
61.1
61.8
57.6
61.3
5.9
9.5
54.3
21.2
0.0
28.3
0.0
26.7
0.0
24.9
(1) The figures presented here are composites for materials
recovered from 21 different product categories.
Source: Compiled by International Research and Technology Corporation.
percent of plastics. The high amounts of
aluminum and steel are recovered primarily
from automobiles.
Note that the figures presented in
Table 1 are composites for materials re-
covered from 21 different product cate-
gories. For example, it is assumed that
95% of the steel in automobiles scrapped
in 1990 will be recovered, while only 25%
of the steel in household durables will
be recovered.
Average product lifetimes assumed for
each of the 21 product categories are
presented in Table 2. Six of the product
categories have assumed average lifetimes
of less than a year. These include news-
papers, disposables, and beverage contain-
ers. Automobiles, furniture and household
durables are assumed to last an average of
ten years, while structures last an aver-
age of fifty years before demolition.
The historical materials flow data
base, and the prompt scrap, diversions,
recycling and average product lifetime
assumptions, are all combined with pro-
jections by the economic forecasting
model to estimate the future quantity
and composition of solid waste.
Results for the reference scenario
are presented in Table 3. As can be seen,
high growth rates are projected for net
waste generation of aluminum and plas-
tics, while the amount of steel in the
waste stream increases very slowly. This
is largely the result of substitutions in
the automobile industry, suggested earlier.
Similarly, wood wastes are projected to
grow slowly, as a result of high substi-
tutions of concrete for wood in construc-
tion over the past thirty years. The
fairly rapid growth in the amounts of
rubber, leather and textiles in the
waste stream reflects production trends of
221
-------
TABLE 2. REFERENCE SCENARIO. AVERAGE PRODUCT LIFETIME ASSUMPTIONS (YEARS
Product Category Average Lifetime Product Category Average Lifetime
Newspapers
Books and Periodicals
Writing Paper
Disposables
Beverage Containers
Other Packaging
Consumer and Inst. Products
Furniture
Apparel
Footwear
Household Durables
< 1
< 1
< 1
< 1
< 1
< 1
7
10
4
2
10
El ectri cal /El ectroni cs
Machinery
Automobiles
Other Transp. Equipment
Tires
Batteries
Construction
Ordnance
Other Industrial Applications
Miscellaneous
15
20
10
20
5
3
50
5
2
5
Source: Compiled by International Research and Technology Corporation.
TABLE 3. REFERENCE SCENARIO. PROJECTIONS OF NET SOLID WASTE FROM ALL SECTORS,
1971 AND 1990V') (Million Metric Tons Except as Noted)
Material
Paper and Paperboard
Glass
Metal
Ferrous
Aluminum
Other
Plastics
Rubber and Leather
Texti 1 es
Wood
Concrete
Total Non-Food Products
Food Waste
Total Product Waste
Yard Waste
TOTAL
Tons per $1,000 GNP
Lbs. per capita/day
1971
35.10
10.98
38.26
(35.02)
( 1.17)
( 2.07)
3.22
2.28
1.38
31.69
43.02
165.93
18.68
184.61
20.38
204.99
.19
5.97
1990
53.09
18.54
41.84
(36.87)
( 2.39)
( 2.58)
10.24
5.13
3.48
37.32
101.15
270.83
22.10
292.93
29.84
322.78
.16
7.91
Avge. Annual
Percentage
Growth
2.20
2.80
0.47
.27
3.83
1.17
6.28
4.36
4.99
.86
4.60
2.61
.89
2.46
2.03
2.42
-0.90
1.49
(1) Includes Residential, Commercial, Industrial, Automotive and Other Transportation,
Construction and Demolition Wastes. Excludes Agricultural and Mining Wastes.
Source: International Research and Technology Corporation.
222
-------
the past decade, duHng which the output
of these materials increased almost 70
percent.
The total net waste load grows at a
rate that is significantly lower than that
of real Gross National Product. Conse-
quently, the quantity of waste per dollar
of 6NP is projected to decline at just
under a percentage point per year through
1990. However, the net solid waste load
per capita is estimated to increase
slowly, from just under six pounds per
person per day, to almost eight pounds
per person per day, while per capita 6NP
grows at 2.5 percent per year.
TECHNOLOGY FREEZE
Projections of total net solid waste
in 1990 are dramatically different given
the assumptions of four alternative scen-
arios. These projections are compared in
Table 4. The first alternative scenario,
which we have called Technology Freeze,
uses the economic projection of the
Reference Scenario, except that there is
no substitution of materials in the pro-
duction of goods, and there is no tech-
nology change. However, the prompt scrap,
recycling and product lifetime assumptions
used in the reference scenario were main-
tained. As would be expected, the amounts
of paper, glass, steel, and wood are
greater in the Technology Freeze scenario,
while the amounts of aluminum and plastic
are less. Food and yard waste projections
were held constant between scenarios (the
same population projections are employed),
and the concrete estimates are the same
since this waste comes from structures
erected prior to 1971. The effect of the
TABLE 4. COMPARISON OF NET SOLID WASTE PROJECTIONS FOR 1990
(Million Metric Tons Except as Noted)
Material
Paper and Paperboard
Glass
Metal
Ferrous
Aluminum
Other
Plastics
Rubber and Leather
Textiles
Mood
Concrete
Total Non-Food Products
Food Waste
Total Product Waste
Yard Waste
TOTAL
Tons per $1,000 GNP
Lbs. per capita per day
Reference Technology
Scenario Freeze
53.10
18.54
41.84
(36.87)
( 2.39)
( 2.58)
10.27
5.13
3.48
37.32
60.47
23.53
46.46
41.74)
2.03)
2.70)
8.01
4.81
3.68
40.53
101.15 101.15
270.83 288.64
22.10
22.10
292.93 310.74
29.84
29.84
322.77 340.57
.19
5.97
.20
6.30
Constant
Recycling
56.49
18.97
53.51
(47.64)
( 2.96)
( 2.91)
10.27
5.28
3.48
37.32
101.15
286.48
22.10
308. 58*.
29.84
338.41
.20
6.26
Increased
Recycling
51.66
18.46
39.11
(34.61)
( 2.24)
( 2.26)
10.27
5.08
3.11
37.32
101.15
266.16
22.10
288.26
29.84
318.10
.19
5.88
Increased
Product
Life
52.81
18.18
36.28
(31.50)
( 2.31)
( 2.47)
9.80
5.00
3.39
36.35
80.77
242.60
22.10
264.70
29.84
294.53
.17
5.45
Source: International Research and Technology Corporation.
223
-------
Technology Freeze scenario assumptions on
the total net solid waste projection is
an additional 18 million tons by 1990,
an increase of 5.5% over the Reference
Scenario projection. The materials subs-
titution assumptions of the Reference
Scenario have heavier materials being re-
placed with lighter materials, therefore
resulting in the smaller waste load.
IMPACTS OF RECYCLING
For two of the alternate scenarios,
variations in the recycling assumptions
have been incorporated. Both alternates
employ the technology change, materials
substitution, prompt scrap, and product
life assumptions of the Reference Scen-
ario. However, one scenario, called
Constant Recycling, has no change from the
recycling rates assumed for 1977, and the
other, called Increased Recycling, reflects
an assumed 10% increase in the recycling
rates for all materials from all products
and for all forecast years, over the rates
assumed in the Reference Scenario. These
recycling assumptions are applied at the
point of disposal; the prompt scrap and
other diversion rates were not changed.
The results suggest that the greatest
post-consumer resource recovery potential
lies with metals. The amount of unrecycled
metal in the 1990 waste load is over 11
million metric tons higher in the Constant
Recycling scenario than in the Reference
Scenario, a difference of 28 percent.
Again, automobiles account for the bulk
of this difference: the fraction of alumi-
num recovered from junked automobiles is
assumed to double in the Reference Scen-
ario, from 30 percent in 1971 to 60 per-
cent in 1990. In the Constant Recycling
Scenario, only 39% of the aluminum from
automobiles scrapped in 1990 is recovered.
Similarly, 85 percent of the steel con-
tained in automobiles scrapped in 1971
was recovered. This fraction increases
to 95 percent by 1990 in the Reference
Scenario. However, in the Constant Re-
cycling Scenario, the fraction remains at
88% from 1977 on. The impact on the total
net waste load is a projection that is
15.6 million tons greater than that of
the Reference Scenario, a difference of
4.8 percent.
In the Increased Recycling Scenario,
the change is less dramatic, but still
significant. The difference in the 1990
net waste load projection is only 1.4 per-
cent, but the1amounts of steel, aluminum,
other metals and textiles are 6 to 12 per-
cent lower than the amounts in the Refer-
ence Scenario projection.
INCREASED PRODUCT LIFE
The lowest net solid waste projection
is found in the Increased Product Life
Scenario. For this projection, the
assumed average product lifetime was in-
creased 10% for all products listed in
Table 2 that have average lifetimes of
over one year. All other assumptions of
the Reference Scenario were maintained.
The impact is greatest on the projections
of steel and concrete, materials that are
typically used in products with especially
long average lifetimes. The differences
in the projections for these two materials
alone account for over 90 percent of the
total difference between scenarios. For
all other materials, the differences are
less than 5 percent, since they are typi-
cally incorporated in products with
comparably short average lifetimes.
SUMMARY AND CONCLUSIONS
The five scenarios presented above
depict a range of plausible future net
solid waste loads. The estimates for 1990
vary from a total of 295 million metric
tons assuming increased product lifetimes,
to 341 million tons assuming an absence
of technology change and materials substi-
tution. The projected trends in post-
consumer recycling initiatives can reduce
the 1990 net waste load by 5 percent, and
a modest increase in these initiatives
can reduce the load an additional 1 percent,
for a total reduction of 20 million metric
tons by 1990. It is clear that the extent
of technology change and materials substi-
tution, the effectiveness of recycling
initiatives, and the average lifetimes of
future products, can substantially affect
the quantity and composition of the waste
generated by our society.
224
-------
THE DEVELOPMENT OF TESTING AND ANALYSIS PROCEDURES FOR REFUSE-DERIVED FUEL
Dr. John Love, Jr.
Professor of Mechanical and Aerospace Engineering
University of Missouri-Columbia
Columbia, Missouri 65211
Mr. Carl ton C. Wiles
U.S. Environmental Protection Agency
Municipal Environmental Research Laboratory
26 West St. Clair St.
Cincinnati, Ohio 54268
ABSTRACT
When procedures have not been standardized or are not readily available, establish-
ment of meaningful purchase-sales agreements often become a barrier between user and
producer. The success of RDF resource recovery systems will depend upon the ability to
market the RDF at a price that reflects its true value. Current reluctances to accept
RDF still relate to uncertainties concerning its ultimate effects on boiler surfaces and
associated equipment such as materials handling systems and a general suspicion, perhaps,
that it is an inferior fuel. While there are some bases for caution, some of the problem
relates to the inability of a user and producer to agree to procedures that will establish
the key characteristics of the fuel. If these procedures can be developed and standard-
ized, it not only will provide the basis for sound purchase-sales contracts, but will also
help to better define what RDF is and what it has to offer compared to other fuels. The
following paper describes the efforts of ASTM Subcommittee E38.01 Energy in developing
consensus standards for characterization of a specially processed refuse-derived fuel,
RDF-3.
INTRODUCTION
Various industries have practiced the
reuse and recycling of their residual mate-
rials for a long time and these wastes of
production are becoming a more important
source of raw material for their products,
by-products, or as supplemental fuel as
energy costs continue to rise. The materi-
als reused by manufacturers do not get into
a general waste stream, and thus the char-
acteristics of the scrap are generally well
known.
In the early sixties when formal col-
lection of mixed solid waste from house-
holds was begun by municipalities, because
it was no longer feasible for the indivi-
dual to dispose of his own waste as he saw
fit, it was used as land fill. It was
obvious that several components of the
waste had some commercial value if separ-
ated and precisely identified. Various
non-profit groups found it advantageous to
recover the cans, bottles, and newspapers
for recycling. In general this was accom-
plished at the source, by cajoling those
generating the waste to separate their
waste and deliver it to collection centers,
as a civic duty. However, it is unlikely
that separation at the source, though
highly desirable and economical, will pre-
dominate in the recovery of all the com-
ponents in MSW which have a commercial
value. The major portion of municipal
solid waste collected is a complicated
225
-------
mixture requiring considerable processing
to recover the commercially valuable com-
ponents that must be precisely identified.
At some point the cost of separation
exceeds the value of the recovered frac-
tions; the remainder, if it can be properly
characterized, is suitable for use as a
fuel.
The obvious need to study the problem
of deposition of Municipal Solid Waste led
to the formation, through the efforts of
Dr. Harvey Alter, of the National Center
for Resource Recovery, of ASTM Committee
E38.00 on Resource Recovery. The scope of
this committee is as follows: "The devel-
opment of methods of test, specifications,
recommended practices, and nomenclature;
the promotion of knowledge, and stimulation
of research relating to material and energy
resources, recoverable, or potentially
recoverable, from waste. The waste for
resource recovery is here defined as that
portion of waste which is collected from
industrial, commercial,or households des-
tined for disposal facilities."
The American Society for Testing and
Materials, ASTM, is a non-profit corpora-
tion "formed for the development of stan-
dards on characteristics and performance
of materials,products, systems, and ser-
vices; and the promotion of related know-
ledge." It is essentially a management
system for development of voluntary consen-
sus standards through its standards commit-
tees. The membership of each fcommittee
must be a designated balance between users,
producers, and general interest groups
concerned with the proposed stai.dard. ASTM
procedures insure that its standards are
the result of a full consensus of its mem-
bers by submitting them to a society letter
ballot before publication. ASTM standards
are nonmandatory and only become true stan-
dards when accepted by the interested
parties, or are adopted by an industry or
governmental body by law. ASTM provides
no funds for the operation of its commit-
tees; all the work is done voluntarily by
the members.
ASTM COMMITTEE E38 ON RESOURCE RECOVERY
The committee became fully operational
in April, 1974, when twelve subcommittees
were given the responsibility for develop-
ment of standards relating to a specific
use or fraction of mixed municipal solid
waste. In addition to the executive sub-
committee, E38.90, the following subcommit-
tees are promulgating standards:
E38.01 Energy
.02 Ferrous Metals
.03 Non-Ferrous Metals
.04 Paper and Paperboard
.05 Glass
.06 Construction Materials
.07 Health and Safety
.08 Unit Processes
.09 Terminology
In addition there are subcommittees for
Research and Long Range Planning. The sub-
ject of this paper is confined to the
activities of Subcommittee E38.01 Energy.
ASTM SUBCOMMITTEE E38.01 ENERGY
Dr. John Love, Jr., Professor of
Mechanical and Aerospace Engineering,
University of Missouri-Columbia, was
appointed chairman of the subcommittee in
the spring of 1973. The first year was
spent in trying to determine the scope of
the subcommittee and finding members with
experience and/or expertise in using Muni-
cipal Solid Waste as a commercial product.
Fortunately about forty interested members
attended the first productive meeting in
April, 1974. The scope approved by the
main committee is as follows: "The identi-
fication of and the development of nomen-
clature, standards, specifications, methods
of test and recommended practices relating
to the recovery of energy or fuels and/or
organic feed stocks from waste; the promo-
tion of knowledge, and stimulation of
research relating to direct or indirect
conversion into chemical or biological
energy forms." Waste was defined as above.
Two full days of discussion uncovered what
seemed to be an infinite number of problems
and suggestions of what to do with a packer
truck of MSW of unknown composition. How-
ever, the subcommittee soon divided itself
into two major groups: those in favor of
biological and/or pyrolitic conversion into
chemical feed stocks and those in favor of
direct conversion by using the waste as a
fuel. Regardless of the ultimate use of
MSW it must be completely characterized by
its physical and thermo-chemical properties,
226
-------
if it is to become a commercial product,
and a reliable sampling technique must be
developed to insure the laboratory sample
is representative of a given lot of the
material. The subcommittee began work on
writing initial specifications for the
product, sampling, and analysis of refuse-
derived fuels and/or chemical feed stocks.
DEFINING THE PROBLEM
The useful products likely to be re-
covered from mixed municipal waste are
paper, steel, aluminum, glass and a-resi-
due of mixed organic materials that may be
used as a fuel or chemical feed stock. A
refuse-derived fuel or chemical feed stock
in the broadest sense may be the raw waste
as^collected; the as-collected waste pro-
cessed to a coarse particle size with or
without the removal of glass and ferrous
metals;. or the combustible waste fraction
from which all or most of the inorganic
material has been removed, the organic res-
idue. The combustible waste fraction may
be further processed into powdered, pelle-
tized, liquid, or gaseous fuels.
From the possible forms of refuse-
derived fuels the subcommittee chose to
characterize the combustible waste frac-
tion as this would meet the requirements
of those interested in both indirect and
direct conversion. Impetus was given to
this decision by the fact that Milwaukee
and Chicago were building processing plants
to produce this, type of fuel for consump-
tion in utility boilers, and contractual
agreements required quality and quantita-
tive specifications. This form of refuse-
derived fuel has been tentatively designa-
ted by E38.01 as RDF-3, defined as a
light-weight, shredded material passing
through a two-inch screen which has been
density-classified with most of the inor-
ganic material removed.
The task groups found some thirty-
seven characteristics relating to RDF-3 as
a product for direct conversion fuel.
Eventually these were listed in order of
priority; of course, sampling was number
one on the list, with physical and thermo-
chemical properties following. The task
groups were assigned to write draft stan-
dards for those parameters with a high
priority. Although mo§t of the subcommit-
tee members had little or no experience in
using RDF-3 as a fuel, nearly all of them
were potential users or producers associated
with fuel laboratories and were familiar
with the ASTM standards for coal and coke.
The first draft standards followed the esta-
blished ASTM standards for coal and coke
closely with minor changes based on the
thoughts of the authors, who at this time
were managers of utility laboratories.
Needless to say, these drafts were tho-
roughly challenged as no experimental evi-
dence could be cited to support them, and
the subcommittee became completely bogged
down in endless debate over the details of
the draft standards. Obviously a closely
controlled interlaboratory study, known as
a Round Robin, was the most effective means
of determining the validity of the details
of the proposed draft standards. However,
the committee had no funds for such a
program.
The. lack of acceptable standards for
determining the value of RDF-3, between the
user and producer, is a formidable barrier
to the general use of it as fuel. EPA
recognized that the most expeditious and
effective way to remove this barrier was to
assist subcommittee E38.01 in the develop-
ment and testing of its proposed standards
arrived at through the consensus approach
of ASTM. EPA-Solid and Hazardous Waste
Research in Resource Recovery Division is
therefore cooperating with ASTM by providing
support funding for the subcommittee's
effort. Mr. Carl ton Wiles, second author,
is Project Manager for EPA, with the Chair-
man of E38.01 as Principal Investigator.
An advisory group composed of one po-
tential user, two producers, and two gen-
eral interest members was formed by the
Principal Investigator to assist in deter-
mining the priorities for the interlabor-
atory study and the general area the inves-
tigation should include. It was agreed
that sampling procedures were undoubtedly
the most important, but it was soon dis-
covered that withaut acceptable standards
for determining the physical and thermo-
chemical parameters of the material there
was no way to differentiate between
sampling procedures. Thus the sampling
study will follow the present study.
The problem finally defined was as
follows: develop and test, by Round Robin
procedures, draft standards for the para-
meters included in (a) size distribution
227
-------
determination; (b) modified proximate ana-
lysis; (c) modified ultimate analysis; and
(d) ash fusion analysis.
The parameters included in the various
analyses by definition are:
(a) Size distribution: Results of a
sieve analysis of the air-dried
sample.
(b) Modified proximate analysis:
1. Total moisture
2. Residual moisture
3. Higher heating value on an as
received, moisture free, and
moisture and ash free basis
4. Sulfur
5. Total chlorine
6. Water soluable chloride
7. Ash
8. Volatile matter
(c) Modified ultimate analysis:
1. Total moisture
2. Residual moisture
3. Carbon
4. Hydrogen
5. Nitrogen
6. Oxygen
(d) Ash fusion analysis:
1. Initial deformation
2. Softening
3. Hemispherical
4. Fluid temperatures in both an
oxidizing and reducing atmo-
sphere
THE EXPERIMENTAL PROGRAM
The objective of the experimental pro-
gram is to perfect the protocols used in
the draft standards and to determine for
each a precision statement for the results
by stating the limits of Repeatability
(duplicate results by the same laboratory
on different days) and Reproducibility
(results submitted by different laborato-
ries) .
The initial steps of setting up an
inter-laboratory Round Robin program for a
"new" product such as RDF-3 are formidable.
The precise method of testing must be deter-
mined, a source of uniform representative
samples found, and a list of willing and
able laboratories must be developed. It
was decided that the method of testing
would follow ASTM E-180, Standard Recom-
mended Practice for "Developing Precision
Data on ASTM Methods for Analysis and
Testing of Industrial Chemicals." The
procedures described in this document are
very precise and it states that "Each
analyst is required to perform duplicate
determinations on each sample on each of
two days—." Thus four separate analyses
of a given parameter constitute a "Round".
Some eighteen laboratories were selected
from the utility, governmental, and com-
mercial sectors in keeping with the spirit
of the consensus approach. These labora-
tories were widely separated geographically.
Since RDF-3 is made up of the organic
fraction of MSW including kitchen waste,
the question was raised as to amount of
decomposition, off-gassing, and loss of
moisture that could occur during shipping
that could affect the analysis. To answer
this question, National Center for Resource
Recovery prepared seven two kilogram sam-
ples of RDF-3 for shipment to the most
remote (shipping time) laboratory. All of
the samples were double bagged in three
mil plastic bags securely sealed. Two of
the samples were packed with dry ice in
insulated cartons; thermocouple leads were
brought outside the cartons so that the
temperature could be monitored without
opening. One of these two samples (the
control) was shipped by air freight, the
other by UPS. The control sample was
analyzed when it reached the freezing tem-
perature. The other five samples were
shipped "hot" in uninsulated cartons by UPS.
One of these five samples, containing a
maximum-minimum thermometer, was opened and
analyzed immediately, the others were
stored at 100°F and analyzed at approxi-
mately 14 day intervals. The Proximate and
Ultimate analyses were used to determine if
the time and/or temperatures normally en-
countered during shipping had an effect on
the stability of the samples. No signifi-
cant change was found in the parameters for
the control sample and the "hot" samples.
RDF-3 samples packed as above can be con-
sidered stable for at least eight weeks.
The procedures of E-180 require that
all laboratories use the same precise pro-
tocol; at this time only four standards had
been drafted, it was incorrectly assumed
that ASTM Standards for Coal and Coke would
be suitable for the remaining parameters.
No formal interlaboratory testing had been
done to confirm the existing drafts or the
228
-------
ASTM standards. One draft standard,
"Standard Method of Preparing RDF-3 Samples
for Analysis", had been used by several
utility laboratories to prepare an analysis
sample from a gross sample. The procedure
described in this draft was widely accepted
by the selected laboratories.
Six laboratories (two utility, two
governmental, two commercial) were selected
to make the pilot studies, or Zeroth Round.
No protocols were designated for these
initial studies except those given by ASTM
standards allowing variations from the
standards as the laboratories saw fit, and
using available equipment. The labora-
tories were required to give complete
detailed protocols used; suggest "better"
and less costly methods, and evaluate their
results. The results of the Zeroth Round
were extremely disappointing; there was
little or no consistency in the results or
protocols used, due in part to the poor
sample received, but also due to inexperi-
ence with this material, lack of-proper
equipment, and improvising when equipment
was unavailable. A multitude of unforseen
problems surfaced, but at least this Round
served as a guide to the modifications
needed in the procedures. It was decided
to run a Second Zeroth Round.
At this juncture, a laboratory advisory
group was formed consisting of the repre-
sentatives from each of the Zeroth Round
participants to assist the Principal Inves-
tigatorin planning the Second Zeroth Round
and administration of the larger problem
to come when the subsequent Round Robins
were run. This group met in December, 1977.
The meeting was very productive. The pro-
blems encountered in the Zeroth Round were
thoroughly examined; it was agreed that the
standard methods and apparatus normally
used in coal analysis were not suitable for
the characterization of RDF-3. Several of
these standards need major modifications.
The group agreed to develop these standards
for the subcommittee's perusal. However,
after considerable discussion it was
decided the Second Zeroth Round would be
run without specific protocols except that
the draft standards for the preparation of
the laboratory sample, sieve analysis, and
moisture determination would follow the
draft standards in process of perfection.
The results of the Second Zeroth Round
were promising. The statistical analysis
of the results seemed to indicate we were
on the right track. The protocol for pre-
paring and shipping the samples was being
followed, and several of the laboratories
followed the new draft standards developed
by the advisory group. However, several
problems still remained, but in particular
the normal size of the laboratory analysis
sample as well as its homogeneity was ques-
tioned. Several laboratories were assigned
to do individual experiments to answer
these questions and incorporate their find-
ings into the draft standards. Neverthe-
less the advisory group felt that we had
sufficient information to proceed with the
first full Round Robin with required proto-
cols. The protocols to be used were from
our draft standards, as modified, except
for six parameters for which the ASTM coal
standards seemed to be appropriate.
The First Round samples were prepared
and shipped in June, 1978, to eleven labor-
atories of which eight responded with
results for all the parameters. The statis-
tical analysis of the results indicated that
the laboratories had obtained better repeat-
ability in their analyses, but the reprodu-
cibility, that is, between laboratories,
had not been achieved. These results were
not in an acceptable range that could be
used as a basis, for precision statements.
The advisory group reviewed the precise
protocols used and the new problems unco-
vered, they produced new draft standards
for the problem areas as well as modified
those already in use where difficulties
were encountered. However, we were well
enough along to run the Second Round Robin.
The Second Round samples were prepared
and shipped by NCRR early in December, 1978.
A second representative laboratory sample,
dry passing through a % mm screen, was pre-
pared by one of the laboratories for ship-
ment to all participants for cross-checking
the reproducibility between laboratories,
and to eliminate as far as possible the non-
uniformity of samples shipped by NCRR. The
protocols to be used by all thirteen labor-
atories participating in the Second Round
are from our latest draft standards except
those for volatile matter, carbon, hydrogen,
and nitrogen. The ASTM coal standard for
these parameters are thought to be suitable.
The results of this Round are due to be
returned February 1, 1979, some four weeks
after this date. It is hoped that the
results will be adequate to determine the
229
-------
precision statements, if not the draft
standards will be further modified for the
Third Round.
The details of the laboratory proto-
cols are described in the most recent
Draft Standards (see Appendix A). As
these drafts are subject to reivision at
any time when new problems surface, they
are not available for general use.
ORGANIZATION OF THE SUBCOMMITTEE
The E38.01 Energy subcommittee has
about sixty-five members. It is divided
into nine Task Groups, each with a new
leader who is responsible for the Draft
Standards relating to his general scope.
The members of these groups are engineers,
scientists, educators, and managers, who
are extremely interested and dedicated to
using MSW as a source of energy; their
work is voluntary and contributed to the
subcommittee on their own time and at
their own expense. The major effort, at
present, is to establish RDF-3 as a com-
mercially competative fuel, by providing
a set of standards that will characterize
it by its physical and thermo-chemical
properties.
The broad interest of the subcommittee
is indicated by the scopes of its Task
Groups (see Appendix B).
The subcommittee is also supporting
some of the work of E38.06 Construction
Materials through a subcontract for the
"Development of Test Methods and Specifi-
cations for Resource-Recovered Glass in
Structural Clay Products."
SUMMARY OF ACTIVITIES
At present the major activity and
effort of the subcommittee are devoted to
the development of precision statements
for the Draft Standards for the parameters
listed in the four analyses subject to the
Round Robin study. It is interesting to
note that this does not depend upon the
accuracy of the absolute value found, or
a gross sampling technique. Thus as long
as the sample prepared by NCRR is repre-
sentative of the particular lot of RDF-3
from which it was taken, the Round Robin
results are valid. The uniformity of the
samples is being checked, as stated above,
in the Second Round Robin. This is not to
say the handling, subdivision, and homo-
geneity of the laboratory sample is not
important, in fact, the Round Robin proce-
dure will determine the protocol to be used.
Note, the accuracy of the analytical deter-
mination can only be determined for a
material of known composition.
The subcommittee has developed a Draft
Standard for Recommended Practice for
Describing the Properties of Pyrolitic Fuels.
It is expected to be resubmitted to the
main committee for a letter ballot in the
near future. This standard has not been
the subject of a Round Robin.
The collection and division of a gross
sample is also being studied by one of the
Task Groups.
The protocols for a detailed analysis
of the desirable parameters are time con-
suming and expensive, and although valuable
and necessary for determination of the true
commercial value of RDF-3 delivered over a
long period, usually a thirty-day average
for fossil fuels, they are not useful for an
ongoing day-to-day quality control program.
Through the study of hundreds of analyses
of MSW in the literature, it has been found,
by one of the Task Groups (Macro-sampling),
that if the moisture and ash are known for
a lot of material the higher heating value
can be estimated to perhaps a commercially
acceptable range. Tests are underway to
determine the proper, if any, correlations.
The proposed procedure is simple; the sam-
ple is first dried to a constant weight,
then it is reduced to ash. The only equip-
ment required is an oven with the proper
temperature control. The subcommittee has
tentatively labeled this process Macro-
sampling Technique.
The Macro-sampling Technique is not
intended to replace the detailed analysis,
in fact the correlation depends upon it,
but only to give a rapid indication of the
quality of the RDF-3 delivered on a short
time basis. It could also serve as a con-
tractural parameter for those locations not
equipped to perform the detailed analysis.
FUTURE ACTIVITIES
As mentioned above, the development of
a procedure for obtaining a statistically
sound laboratory sample, say thirty pounds,
230
-------
from a large lot of MSW is indeed a formi-
dable problem. The present sampling
methods for coal and coke are not suitable
because of the heterogeneous nature of MSW.
There is practically no agreement among the
subcommittee members how this can be accom-
plished, or indeed just where in the stream
the gross sample should be collected. The
various particle sizes of the raw material
and the large quantities that must be han-
dled in a commercial operation make quar-
tering, bailing and coring at least unreli-
able. In short, the procedure of obtaining
a statistically sound sample from commercial
quantities of MSW must be reliable and
inexpensive as the sampling cost could
easily exceed the value of the material.
Probably automatic collection at some point
in a free falling stream is the answer.
The collection and division of the gross
sample is our next major project.
Other future studies will include the
determination of volatile toxicants, metals
and trace toxicants in the ash, densifica-
tion, flowability, storabitity, retrieva-
bility, grindability/friability, abrasive-
ness, free swelling index, and combustion
characteristics such as spontaneous combus-
tion point, ignition point and residence
time required for ignition, flash point,
and explosiveness.
One other future project is large
volume calorimetry in which the subcommittee
is cooperating, however, the development
work is being done by the National Bureau
of Standards.
CONCLUSIONS AND RECOMMENDATIONS
These conclusions and recommendations
are those of the Principal Investigator and
not necessarily those of Mr. Wiles.
The RDF-3 supplied by NCRR processed
from MSW collected in Washington, D.C. can
be characterized, on an as-received basis,
as a material passing a one-inch screen
with a higher heating value of 5000 to 6000
BTU/pound mass, with other parameters as
follows: Moisture 17%, Volatile Matter 30%,
Ash 16%, Carbon 20%, Hydrogen 2%, Nitrogen
1%, Oxygen 12%, Sulphur .1%, and Chlorine
•4%. These values are only approximate and
have little value except to show that RDF-3
is certainly suitable as a high quality
fuel with a modest heating value. Certain-
ly it is valuable as a commercial fuel.
I believe the subcommittee, even with
the associated frustrations, will develoo
the required Draft Standards, and the ASTM'
consensus approach is the proper way to
develop commercial standards acceptable to
both the producers and users of a "new"
product, RDF-3. Also although very time
consuming, the use of dedicated volunteers
in developing the Draft Standards will
assure their acceptance by ASTM for publi-
cation.
At this time, it appears that the usual
small, 1 g, analysis sample is suitable for
the thermo-chemical analysis if it is truly
representative of the lot.
I recommend that government agencies
requiring acceptable standards and proce-
dures help develop them through the ASTM
methods by support funding of its committees,
as it has done with E38 Resource Recovery,
and that EPA extend its commitment to cover
the future work of E38.01 Energy.
ACKNOWLEDGEMENT
It is impossible to name all those who
are contributing to the work of the sub-
committee, however, I am especially indebted
to the following: ASTM headquarters and Mr.
Sam Bowman, Staff Representative; Dr. Harvey
Alter, NCRR, for appointing me as Principal
Investigator; the Advisory Group, Dr. Mark
T. Atwood, Tosco Corp;; Dr. Jerome F.
Collins, DOE; Mr. Herbert I. Hollander,
Gilbert Assoc.; Mr. David L. Klumb, Union
Electric Co.; Dr. William L. Young,
Americology; and the Laboratory Advisory
Group, Mr. Richard S. Alberg, Wise. Elec.
Power Co.; Mr. J.E. Attrill, Oak Ridge
Nat. Lab.; Mr. Robert Berkemeyer, Williams
Bros. Proc. Serv; Mr. Richard R. Dlesk,
Commonwealth Edison Co.; Mr. John K.
Kieffer, GiIbert Assoc.; and Mr. David J.
Mitchell, NBS. "'Since March, 1977, the
work upon which this publication is based
was performed persuant to Contract No.
68-03-2528 with the Environmental Protection
Agency. Mr. Carl ton Wiles is the Project
Officer.
231
-------
10.
11.
12.
APPENDIX
THE DRAFT STANDARDS PROMULGATED BY
SUBCOMMITTEE E38.01 ENEKGY TO DATE
Standard Recommended Practice for
Describing the Properties of Pyrolitic
Fuels;
Standard Method of Designating RDF-3
from its Sieve Analysis;
Standard Method of Preparing RDF-3
Samples for Analysis;
Standard Sampling Procedure for RDF-3
Collection and Division of a Gross
Sample;
Standard Method of Test for Residua]
Moisture in RDF Analysis Sample;
Standard Method of Determining Total
Moisture (single stage) in RDF-3
Samples;
Test Method for Chlorine in Refuse-
Derived Fuel (RDF-3);
Standard Test Method for Water-Soluble
Chlorides in Refuse-Derived Fuel;
Standard Method for Calculating Refuse-
Derived Fuel Analysis Data from As-
Determined to Different Bases;
Test Method for Sulfur in Refuse-Derived
Fuel (RDF-3);
Standard Method of Test for Ash in the
Analysis Sample of Refuse-Derived Fuel
(RDF-3);
Standard for Determination of RDF-3
Calorific Value by Bomb Calorimeter.
In addition ASTM standards D3175, D3178-
73, D3179-73, and D1857 are adopted without
change.
11B"
THE TITLES OF THE NINE ACTIVE TASK GRUUPS:
1. Macro-Sampling;
2. Gross Sampling;
3. Laboratory Sample Preparation and
Proximate Analysis;
4. Thermo-chemical Analysis;
5. Physical Characteristics;
6. Metals and Trace Toxicants in Ash;
7. Pyrolytic Conversion;
8. Large Volume Calorimeter;
9. Evaluation of Energy Recovery Systems.
THE LABORATORIES PARTICIPATING IN THE
SECOND ROUND
1. Commonwealth Edison Co.
2. Rochester Gas and Electric Co.
3. Wisconsin Electric Power Co.
4. National Bureau of Standards
5. U.S. Army Const. Eng. Research Lab.
6. Oak Ridge National Laboratory
7. Foster-Wheeler Development Corp.
8. Combustion-Engineering Power Systems
Dept.
9. Gilbert Associates, Inc.
10. Fuel Engineering
11. Williams Bros. Proc. Serv., Inc.
12. Roltech Scientific Services
13. Commercial Testing and Engr. Co.
The first eight laboratories are cooper-
ating in the subcommittee's work gratis. The
remainder, commercial laboratories, are
billing the subcommittee for reimbursement
of only direct cost.
232
-------
Calorific Value Determination of Refuse-Derived-Fuels by
Large-Bomb Calorimetry
Duane R. Kirklin, Eugene S. Domalski and David J. Mitchell
National Bureau of Standards
Washington, DC 20234
ABSTRACT
An oxygen bomb calorimeter has been
designed and constructed at the National
Bureau of Standards which can accommodate a
25 gram sample of refuse or a refuse-
derived-fuel (RDF) for the purpose of
studying the effects of sample processing
on the measured calorific value. The large
calorimeter is an enlarged and modified
version of a conventional-sized calorimeter
also in use at NBS. The large calorimeter
can handle samples ten times larger than
the conventional-sized calorimeter and
therefore can investigate RDF samples with
either minimal or no processing. Calori-
metric re|ults are presented on d(densi-
fled)-RDF carried out in both calorimetric
systems. Moisture- and ash-free (MAF)
calorific values were obtained from six
randomly chosen unprocessed RDF samples and
had a mean value of 10 742 Btu(lb)~J- in
the large calorimeter. Another randomly
chosen analysis sample of unprocessed RDF
was subjected to extensive sample processing
to obtain a "homogeneous" analysis sample
for use in the small conventional-sized
calorimeter. Measurement on a set of ten
"homogeneous" sub-samples gave a mean
calorific value of 10,743 Btu(lb)"1 in the
small calorimeter. The results of this
investigation indicate that the calorific
value of d-RDF is unaffected by the sample
processing technique used at NBS.
The d-RDF utilized in this study was
produced by Teledyne National in Cockeys-
ville, MD but in no way does this imply an
endorsement of this product by the National
Bureau of Standards.
INTRODUCTION
The Resource Conservation and Recovery
Act of 1976, PL 94-580, identifies that
solid waste is a potential source of oil,
gas, or solid fuel which can be converted
into energy and mandates that the Depart-
ment of Commerce provide accurate specifi-
cations for recovered materials. The
National Bureau of Standards will furnish
guidelines, as indicated in this Act, for
the development of specifications which
pertain to the physical and chemical
properties, and characteristics of re-
covered materials with a view toward
replacing virgin materials for various
applications.
The heating value is an important
characteristic and perhaps the most signi-
ficant property of a fuel. Heating values
are used extensively to evaluate the
commercial potential of the fuel and also
to evaluate the performance of incinera-
tors and refuse-fired boilers. The E-38
Committee on Resource Recovery of the
American Society of Testing Materials
(ASTM) is interested in the development of
standard methods of test required for the
establishment of refuse-derived-fuels
(RDF) as an article of commerce. The
American Society of Mechanical Engineers
(ASME) Performance Test Code 33 Committee
is interested in an evaluative method to
determine the performance of large inciner-
ators and boilers. A laboratory procedure
giving representative and reproducible
heating values is necessary to better
equip commercial laboratories to certify
accurately the energy content of RDF and
also to evaluate the compliance of large
incinerators and refuse-fired boiler with
their contract performance specifications.
233
-------
A method must be established for
precise and accurate determination of the
heating value of refuse and RDF to evaluate
adequately its fuel value. Therefore, the
National Bureau of Standards has entered
into a collaborative research agreement
with the U.S. Environmental Protection
Agency and the U.S. Department of Energy to
establish the procedures necessary to
determine the calorific value of refuse and
RDF by bomb calorimetry. The objectives of
this project are to determine the optimum
particle size of samples for combustion
measurements and to establish whether a
conventional (2.5 gram capacity) calori-
meter or a larger (25 gram capacity)
calorimeter will provide more representa-
tive calorific values of RDF. This report
provides a response to both objectives from
the results obtained for a Teledyne National
RDF sample using a conventional (2.5 gram
capacity) bomb calorimeter and a recently
constructed large (25 gram capacity) bomb
calorimeter.
EXPERIMENTAL
Materials
1. Benzoic Acid. Standard Reference
Material 39i was obtained from
the NBS standard samples store-
room. This sample has a certi-
fied energy of combustion of 26
434+3 J-g"1 at standard bomb
conditions and was used to cali-
brate the conventional-sized
calorimeter. All samples were
drawn from the same bottle.
Fisher Scientific Company's
certified benzoic acid with an
energy of combustion of 26 437 +
3 J*g was utilized for the
large calorimeter.
2. Oxygen. Ultra High Purity (UHP)
grade of oxygen was supplied by
Matheson Gas Products. This
oxygen is certified to contain
combustible impurities not ex-
ceeding 0.002 percent and total
impurities of less than 0.05
percent.
3. d-RDF. In February 1977, a 20
kg (44 pound) sample of extruded
RDF pellets was received from
Teledyne National. This labora-
tory sample was from the Baltimore
County Resource Recovery Plant
located in Cockeysville, Maryland.
The collection of extruded pellets
was contained in a plastic bag
enclosed in a cardboard box. The
Teledyne National pellets are
cylindrical in shape having a
diameter of 2.5 cm (1 inch) and
are broken-off lengths of about
2.5 to 7.5 cm (1 to 3 inches).
Sample Preparation
1. Sample Requirements. It is
necessary that a reproducible family of
analysis samples be prepared from a gross
field sample of refuse or RDF. The analy-
sis sample must be representative of the
field sample and unaffected by the labora-
tory techniques utilized to produce these
analysis samples. If it is assumed that
the milling and blending of the gross field
sample produces a homogeneous product, then
the only variational parameter which must
be measured and maintained is the moisture
content. Therefore, the analysis samples
were equilibrated in a constant humidity
container at the average relative humidity
of the bomb calorimetric laboratory. The
temperature and relative humidity of the
laboratory are maintained at 295 K and 45
percent, respectively.
2. Small Calorimeter Samples. A
random sample of extruded Teledyne National
pellets was removed from the field sample
of RDF. The preparation of RDF samples
for the small calorimeter has already been
described [l] and is also provided below.
The pellets were ground in a Quaker City
Mill (Model 4) to a particle size of less
than 1.3 cm (0.5 inch). The particles were
then milled to pass a 2 mm (10 mesh) screen
in a Wylie Micro-Mill. The ground sample
was riffled and then thoroughly blended in
a vee blender.
The ground material was then pressed
into pellets under a force of approximately
10 000 Ibs. Pellets weighed approximately
2.2 grams. These pellets were dried to
constant weight at 105 °C in a drying oven
and weighed again. The samples were then
placed in the constant humidity atmosphere
for approximately 48 hours and weighed for
a third time.
3. Large Calorimeter Samples. Ex-
truded RDF pellets are utilized in the
234
-------
large calorimeter with no further process-
ing by NBS laboratories. The long extruded
pellets were broken into pieces weighing
between 20 and 25 grams and placed in a
constant humidity atmosphere.
Figure 1 shows the size of a benzoic
acid pellets for the large calorimeter,
pellets for both the small and large calori-
meters, and the unprocessed RDF sample as-
received from Teledyne National. The as-
received extruded RDF pellets appear
heterogeneous while the milled and blended
RDF pellets have a more homogeneous appear-
ance.
Calorimetric Apparatus
1. Large Isoperibol Bomb Calorimeter.
The design of the large isoperibol (iso-
thermal-jacket) calorimeter which will
accommodate the large combustion bomb,
described in the following section, is
similar to that of Coops et al. [2,3] and
Gundry ct al.'[4] and is shown in Figure
2. The calorimeter is constructed entirely
on stainless steel. It consists of a
cylindrical calorimeter vessel (25.4 cm
diam, 45.7 cm height (10 inch diam, 18
inch height)) in which three rods support
a concentric cylindrical shield (20.3 cm
' • * > i » a, ,, ., „ :, a
7vr &«n{ "Wraivfr <™ *. National Bureau of Standards
'' t
Pellets of benzoic acid, processed RDF and
unprocessed RDF
Figure 1
235
-------
25 gram capacity Bomb Calorimeter
A pulley to stirrer motor, B one of four stacks attached to the submarine lid to
accommodate (1) fuse circuit, (2) heater, (3) temperature sensor, and (4) pipette,
C submarine lid, D submarine flange with "0" ring, E calorimeter vessel lid, F stirrer
shaft, G shield cover, H stirrer, I calorimeter vessel supports, J shield, K combustion
bomb, L submarine vessel, M calorimeter vessel, N bomb foot.
Figure 2
236
-------
diam, 30.5 cm height (8 inch diam, 12 inch
height)). The calorimeter lid supports a
stirrer assembly and a shield cover. The
shield, shield cover, and stirrer assembly
facilitate the flow of water in the calori-
meter so-that water is moving (downward)
in the space between the bomb and shield,
and (upward) between the shield and wall
of the calorimeter vessel. The combustion
bomb is supported by a foot which has been
welded to the bottom of the vessel to
insure that the bomb will be positioned in
the same manner in the calorimeter for
each experiment. The calorimeter vessel
contains 19 liters of water for each
calorimetric measurement and is housed in
a submarine vessel which has a cover. The
submarine cover is fastened to the vessel
with six bolts and has four vertical ports
for: (1) fuse circuit leads, (2) a heater,
(3) a central stirrer assembly and, (4) a
quartz thermometer. The frequency of
oscillation of a temperature sensitive
quartz crystal is used to determine the
calorimeter temperature. The NBS standard
frequencies of 10 and 100 kHz are utilized.
The entire calorimeter system (bomb,
calorimeter vessel, and submarine compart-
ment) is immersed in a constant tempera-
ture water bath maintained at 30 °C to
within +0.03 °C. The overall volume of
the bath is 280 liters and holds 235
liters of water with the calorimeter
system immersed in it.
2. Large Combustion Bomb. The combus-
tion bomb which accommodates a 25 gram
sample of RDF was purchased from the Parr
Instrument Company and is shown schemati-
cally in Figure 3. The bomb has an overall
height of 35 cm (13.75 in) and has a mass
of 13.8 kg (30.5 Ib) when assembled; the
internal and external volumes of the bomb
are 1.85 and 3.62 liters, respectively. A
platinum crucible (^50 cm-*) was used in
the calibration experiments with benzoic
acid and a stainless steel crucible ('x-SO
cnH) was used in the combustion experiments
with RDF. The bomb body has a wall thick-
ness of 0.953 cm (0.375 in) and an outer
diameter of 11.4 cm (4.5 in).
A photograph of the small and the
large combustion bombs is shown in Figure
4 and allows one to examine the relative
difference in their sizes. Figure 5 is a
photograph of the large calorimeter viewed
from the top. It shows the large bomb
immersed in the calorimeter vessel with
the submarine lid and calorimeter cover
(inverted to display the stirrer and
shield cover).
Calorimetric Procedure
Bomb calorimetric techniques are
well-established and are described in
detail elsewhere [5,6], but a brief
description of the method used in this
work follows. In the thermochemical
investigations, the heat evolved by a
measured amount of RDF burned is compared
with the heat evolved by a measured amount
of a selected standard reaction, using a
fixed calorimeter system with a specific
temperature rise. The standard reaction
is the combustion of benzoic acid under
standard bomb conditions producing a
temperature rise of 3 and 5 degrees in the
small and large calorimeters, respectively.
She energy equivalent of the calorimeter
is the amount of energy produced by the
standard reaction and its accompanying
side reactions divided by the corrected
temperature rise. The observed temperature
rise must be corrected for stirring energy
and thermal leakage from the surroundings.
Multiplication of the energy equiva-
lent (obtained from the calibration experi-
ments) of the calorimeter by the corrected
temperature rise (measured in the RDF
combustion experiment) gives the total
energy produced in the RDF combustion
experiment. This total energy is then
corrected for the various side reactions
and divided by the mass of the RDF sample
to produce the gross calorific value. In
a typical RDF experiment, an equilibrated
RDF pellet is weighed in a tared stainless
steel crucible. The crucible and sample
are supported inside an oxygen bomb. The
sample is in contact with a 10 cm length
of 0.127 mm (.005 in) diameter iron fuse
wire. The bomb also contains 10 cm of
H20 to dissolve the gaseous products of
combustion and maintain an atmosphere that
is saturated with water. The sealed bomb
is then charged with 4.1 MPa (40.8 atm) of
high-purity oxygen. The bomb is lowered
into the calorimeter and the covered
calorimeter is submerged in the constant
temperature water bath. The calorimeter
system is then heated to slightly below
25 °C. Temperature versus time data are
collected to measure the temperature rise
of the calorimeter as a result of the com-
bustion of the RDF sample. Figure 6 is an
237
-------
Inner Arrangement of Large Combustion Bomb
A Grounded Electrode
B Valve
C Split Ring
D Compression Ring
E Drop Band
F Buna-N 0-Ring
G Bomb Head
H Bomb Body
I Handle
J Gas Inlet Tube
K Ungrounded Electrode
L Fuse
Figure 3
M Fuse Attachment Hook
N Ring Holder
0 Ring Support
P Sample Pellet
Q Crucible
238
-------
*S.'f £*'<*/ IttlMJMI,
National Bureau ol Standards
Small and large combustion bombs
Figure 4
239
-------
e Calorimeter System (A view from the top)
Figure 5
240
-------
a
UJ
cr
cr
UJ
a.
2
UJ
i—
e.
Time-temperature curve for a bomb calorimeter experiment.
initial time of an experiment, tj, time at which main reaction
period begins, commenced by ignition of sample, tx, mid-time
determined graphically to calculate the cooling correction,
te time at which main reaction period ends, tf final
temperature of an experiment, 6^ initial temperature of an
experiment, 6f final temperature of an experiment 9j jacket
temperature (i.e.) temperature of water bath), Boo convergence
temperature (i.e., temperature which the calorimeter would
attain in an infinite time if 6. and the rate of stirring remain
constant.
Figure 6
TIME
t.
-------
example of a time-temperature curve for a
typical bomb calorimetric experiment. The
temperature is measured during the period
before the sample is ignited (line ab in
Figure 6), during the reaction period
immediately after the sample is ignited
(line be in Figure 6), and during the
period after the reaction is complete
(line eh in Figure 6). The difference
between the first point in the after-
period and the last point in the fore-
period gives the observed temperature
rise. The slope of the fore- and after-
periods allow one to calculate the portion
of the temperature rise due to stirring
energy and thermal leakage. The submerged
calorimeter must be stirred at a constant
rate to obtain a uniform and meaningful
temperature vs. time curve. A preliminary
experiment was performed to determine the
amount of RDF sample necessary to produce
about the same temperature rise in the
calorimeters as that produced in our
comparison reaction with benzoic acid, (a
three and five degree (K) temperature rise
in the small and large calorimeter, re-
spectively) .
DISCUSSION AND RESULTS
Calibration Experiments
It is necessary to determine the
energy equivalent for both the large and
conventional-sized calorimeter systems.
The energy equivalent of a calorimeter is
the total amount of energy that must be
supplied to the calorimeter to produce a
specified temperature rise in that calori-
meter. The results of these calibration
Expt. No.
Table I. Calibration Experiments of the Bomb Calorimeters
Small Calorimeter Large Calorimeter
1046 1047 1048 1049 1050 2006 2010 2011
AUc°'Th(J'8"1)
"BA(8)
q-BA(J)
q-tgn(J)
q-Fe(J)
q-HN03(J)
q-wc(J)
q-Tfcorr(J)
q-C (J)
corr
Or Total (J)
AT-corr(deg)
E-cal (J-deg"1)
C -cont. (J-deg)
E-si (J-deg"1)
E-si mean (J-deg ]
Std. Dev. (J-deg"1;
Std. Dev. 1 Mean
(J-deg"1)
26410.68
1.635791
43202.35
1.08
6.80
37.13
.02
43247.38
2.969681
14562.97
6.87
14556.10
>
)
26410.68
1.632017
43102.68
1.72
5.90
34.59
.04
43144.93
2.963055
14560.96
6.87
14554.09
26410.
68
1.650145
43581.
0.94
5.77
34.96
-.02
43623.
45
10
2.995277
14563.
96
6.89
14557.
14554.
1.72(0.
0.78(0.
07
75
26410.68
1.639178
43291.81
1.19
5.20
34.85
.03
43333.08
2.976097
14560.37
6.88
14553.49
26410.68
1.640040
43314.57
1.01
5.30
34.80
.02
43355.70
2.977749
14559.89
6.88
14553.01
26411-.3&
14.998983
396143.48
2.19
56.05
183.43
383.16
5.84
-101.60
396672.54
4.601897
86197.61
31.78
86165.83
012%)
0053%)
26411.
36
16.272774
429785.
1.59
38.80
57.87
510.15
-.98
-29.50
430363.
98
91
4.992241
86206.
56
49.88
86156.
86162.
5.00(0.
2.89(0.
68
31
26411
.36
16.289839
430236
1.92
38.82
57.87
510.78
- 1.25
-19.66
430825
.69
.16
4.997143
86214
.30
49.89
86164
.41
0058%)
0034%)
242
-------
experiments are presented in Table I.
Experiment numbers 1046 through 1050 were
performed in the small conventional calori-
meter and 2006, 2010 and 2011 were per-
formed in the large calorimeter. In a
calibration experiment, one accounts for
all energy supplied to the calorimeter
system and measures the resulting tempera-
ture rise. The line identifications in
column 1 of Table I contain the different
sources of energy in a typical calorimetric
experiment. A comparison of the small and
large calorimeter entries will point out
some of the differences between the two
calorimeters. AU °,T, is the energy of
combustion of benioic acid at the specified
reference temperature, T . The small
calorimeter is referenced to 28 °C and the
large calorimeter is referenced to 30 °C.
The heat of combustion of the calorimetric
standard, benzoic acid, is known to one
part in ten thousand. From the mass of
benzoic acid, shown in line 3, the energy
supplied by combustion of benzoic acid can
be calculated and is presented in-line 4,
q-BA. Line 5 shows the electrical energy
that was supplied to ignite the sample.
The platinum fuse wire of the small calori-
meter is heated electrically to its melting
point to ignite the samples. The molterr
platinum fuse solidifies and therefore
supplies no energy to the small calorimeter.
In the large calorimeter a iron fuse wire
was used. The electrical energy necessary
to melt the iron fuse wire must be accounted
for, but in addition, the molten iron burns
in oxygen producing iron oxides. The
quantity of iron fuse wire burned and the
energy of combustion of the iron fuse wire
are used to calculate the energy supplied
by the combustion of iron, q-Fe. Some of
the nitrogen impurities in the oxygen are
converted to nitrogen oxides which react
with water to form nitric acid. The quan-
tity of nitric acid produced is measured
and from its energy of formation, q-HN03
is calculated. In high-precision combustion
calorimetry, it has become standard practice
to apply corrections proposed by Washburn
[7] that produce calorimetric data referenced
to a standard set of conditions. These
corrections to standard states for the
isothermal bomb process at 25 °C and unit
fugacity are known as Washburn Corrections
and are presented as q-wc. The small
calorimeter is designed to operate between
25 °C and 28 °C while the large calorimeter
operates between 25 °C and 30 °C. In
actual practice, the final temperature of
our experimental reaction is not exactly
25 °C or 30 °C and therefore, a small
correction must be applied to the calori-
metric data for high precision work and is
shown as q-Tf corr. in Table I. Due to
the limited quantity of oxygen in the
large bomb, a few tenths of a milligram of
carbon residue were found in the combustion
crucible. The stated energy of combustion
of benzoic acid assumes complete combustion
and therefore a small correction for
unburned carbon was applied to the results.
The algebraic sum of energy from various
sources yields, the total energy supplied
to the calorimeter, Q-Total.
Some additional discussion is warran-
ted in regards to the unburned carbon
residue. Stoichiometrically, 2 grams of
oxygen are required for the combustion of
1 gram of benzoic acid. Normally, 10
grams of oxygen are used for each gram of
benzoic acid to insure complete combustion
in the conventional-sized bomb. Calori-
metrically, it has been found in practice
that 6.5 g of 02 per gram of benzoic acid
is the minimum requirement for complete
combustion in a bomb calorimeter. The
large bomb utilizing a 16.5 gram benzoic
acid sample and 4.1 MPa (40 atm) of
oxygen contains about 100 grams of 02-
This is on the borderline of the minimum
amount of oxygen necessary for complete
combustion in bomb calorimetry. However,
in our experiments the total correction
for unburned carbon is usually less than
40 joules for a 400 000 joule experiment
(1 part/10 000) and if completely neglected
would be acceptable for precision calori-
metry of benzoic acid. The correction for
unburned carbon is given in Table 1 for
the large calorimeter data and presented
as q-C
^ corr
The corrected temperature rise of the
calorimeter, AT-corr, is measured and with
the total energy supplied to the calori-
meter, the energy equivalent, Ecal, of the
calorimeter is determined. The mass of
water introduced into the bomb, the cruci-
ble and the sample may change from experi-
ment to experiment, and when accounted for
in the energy equivalent of the calori-
meter, the energy equivalent of the "empty"
calorimeter, Esi, is obtained.
The important parameters to compare
between the two calorimeters are the energy
equivalents and the total energy capacities
243
-------
of the two calorimeters. The energy
equivalent of the small calorimeter is 14
555 joules per degree while that of the
large calorimeter is 86 162 joules per
degree. This is about six times that of
the small calorimeter. The large calori-
meter has a total energy capacity of
430 810 joules compared to 43 665 joules
of the small calorimeter. Therefore, a
sample ten times larger and thus requiring
less particle size reduction can be used
in the large calorimeter. This is ex-
tremely important since the calorific
value of a solid fuel can now be deter-
mined with only minimal processing and the
suspected chemical changes which are
imposed upon the refuse or RDF analysis
sample as a result of size reduction, can
now be determined quantitatively.
RDF Combustion Experiment
In a calorimetric experiment, the
calculations are done in reverse order in
comparison to a calibration experiment.
The temperature rise of the calorimeter is
measured and with the mean energy equi-
valent of the empty calorimeter plus its
contents, the total energy supplied to the
calorimeter is calculated. In addition to
the corrections applied to the benzoic
acid experiments, a correction is applied
for the formation of sulfuric acid because
of a small amount of sulfur Cv-0.1 percent)
contained in the d-RDF sample. Since the
actual reactants and products in the RDF
experiments are unknown, the Washburn
corrections (used in high-precision
calorimetry), are not applied. In addition,
Table II. MAT Calorific Values in MJ-kg"1 (Btu-lb"1) of
Teledyne National RDF
Small Calorimeter
Expt. No.
1052
1053
1054
1056
1057
1058
1059
1060
1061
1062
V
24.95
(10726)
24.51
(10539)
25.01
(10752)
24.78
(10656)
25.06
(10776)
25.06
(10776)
25.20
(10835)
25.12
(10799)
25.09
(10788)
25.08
(10781)
Range
Mean
Std. Dev.
%Std. Dev.
(10539-10835)
(10743)
(86)
0.80%
Large Calorimeter
Expt. No. Q
2030
2032
2033
2034
2038
2039
25.27
(10866)
26.09
(11216)
24.96
(10732)
24.42
(10501)
24.77
(10650)
24.39
(10485)
(10485-11216)
(10742)
(273)
2.54%
244
-------
complete combustion is assumed in the RDF
experiments. Subtraction of the energy
from all sources except the sample burned
yields the energy supplied by combustion
of the sample alone. Using the mass of
sample burned, the calorific value of the
solid fuel is calculated.
Table II presents the calorific
values obtained on an RDF sample from
Teledyne National. The values presented
from the small calorimeter were obtained
on small samples which required consider-
able processing as outlined in the experi-
mental section earlier. The values
determined on d-RDF samples in the large
calorimeter were obtained on extrusions
with no further processing.
Table II presents moisture- and ash-
free calorific values obtained from these
two types of calorimetric samples. Typical
RDF samples have a moisture- and ash-free
calorific value of between 9 000 and
10 000 Btu(lb)-1. The Teledyne National
samples studied are obviously atypical and
exhibit an unusually high heating value.
Ten experiments in the small calori-
meter on "homogeneous" RDF yielded a mean
calorific value of 10 743 Btu(lb)~l and
a percent standard deviation of 0.80 per-
cent while six experiments on as received
RDF samples with no further processing
yielded a mean value of 10 742 Btu(lb)"1
and a percent standard deviation of 2.54
percent in the large calorimeter. This is
a significant finding when one considers
the large difference in physical appearance
of the two types of samples.
The range of calorific values is much
larger on the large unprocessed samples
than on the small processed ones. Calori-
fic values were dispersed over the range
from 10 539 to 10 835 Btu(lb)"1 for the
small calorimeter samples while a range of
10 485 to 11 216 Btu(lb)-1 was observed
for the large calorimeter samples. These
results suggest that the unprocessed
samples are more heterogeneous and that
the calorific values would scatter widely.
However, if a sufficient number of ex-
periments were performed, the mean values
would still be the same (or very nearly
so) for the two types of sample populations.
A closer look at the heating values obtained
in the large calorimeter reveals that five
of the six values are close together and
that one value, 11 216 Btu(lb)"1, lies
further from the mean. However, the
11 216 Btu(lb)-1 value lies only 1.7
standard deviations from the mean, and
thus should be included as part of the
same sample population.
If the high calorific value were to
be omitted, the mean calorific value would
be 10 647 Btu(lb)-1 which is less than one
percent from the mean obtained when 11 216
Btu(lb)"1 is included. This omission
would change the percent standard devia-
tion from 2.54 percent to 1.51 percent.
Several things could account for the
spread of calorific values from the large
calorimeter. In addition to the sampling
problems associated with RDF, the moisture
and ash contents are very important para-
meters-. Table III contains the moisture
and ash contents used to calculate the
moisture- and ash-free results from the
as-determined calorimetric data which are
presented in Table II. From the percent
moisture of samples 1052, 1053 and 1054,
we attribute the fluctuations in the
moisture content to the relative humidity
changes of the room. Therefore, sample
1056 through 1062 were equilibrated in a
constant humidity atmosphere and all had
virtually the same moisture content.
Water was absorbed or evaporated from the
samples during the weighing process but
always at a rate which was not readily
reflected in the corresponding calorific
value. This was accomplished by main-
taining the constant humidity container at
the average relative humidity of the room
(^45 percent). One reason for the good
precision of the small calorimeter results
is that the moisture was virtually con-
stant for all samples and was determined
on the calorimetric sample directly. This
was not true for the large calorimeter
samples. With the small samples, 48 hours
were sufficient for equilibration, but the
as-received extruded pellets for the large
calorimeter had approximately 13 percent
moisture and equilibrated very slowly in
our constant humidity atmosphere. Since
equilibrated samples have the same moisture
content, we feel that separate moisture
experiments would give an average moisture
value indicative of all the equilibrated
RDF samples. Two experiments were carried
out on several extruded pellet fragments
245
-------
each weighed about 5 grams, and yielded
moisture values of 5.71 and 5.92 percent.
Two experiments were performed on extruded
pellets of approximately the same size (^25
grams) as the combustion samples and yielded
moisture contents of 6.82 and 6.30 percent.
We feel that none of the extruded pellets
had equilibrated completely and that the
equilibration was very slow and largely
dependent upon the size of extruded pellets
chosen for equilibration. The moisture
determinations were done on different days
during the period that the calorimetric
experiments were being performed. We feel
that the moisture determinations of the
extruded pellets of approximately the same
size (^25 grams) were more representative
of our calorimetric samples. We, therefore,
used an average moisture of 6.56 percent to
calculate calorific values on a moisture-
free basis. The large uncertainty of the
actual moisture of each calorimetric sample
may be the cause for the large range of
heating values calculated for measurements
carried out in the large calorimeter.
Our earlier experiences in determining
MAF heating values of RDF indicated that,
even with the highly processed and blended
RDF, the ash content varied from sample to
sample. We therefore adopted a procedure
in our laboratories of determining the ash
from the amount of residue remaining in the
bomb after a combustion experiment. This
has allowed us to calculate the actual
"ash" for each calorimetric samples. For
the small calorimeter samples, the actual
moisture and ash contained in each sample
is known and allows us to calculate the
MAF calorific value with a percent stan-
dard deviation of less than one percent.
The ash measurements were determined on
each of the large calorimeter samples but
an average of moisture content values was
used to calculate the MAF calorific values
with a percent standard deviation of
approximately 2.5 percent.
CONCLUSION
We have designed and constructed a
bomb calorimeter which is capable of
handling a sample ten times heavier than
our conventional-sized bomb calorimeter.
Suspected differences in the character of
refuse or RDF samples because of inappro-
priate sampling or intensive size reduc-
tion can be determined from calorimetric
measurements in this large calorimeter.
This isoperibol (isothermal-jacket) calori-
meter is an enlarged and modified version
of our conventional-sized calorimeter with
the same basic principles of operation.
The large calorimeter has been calibrated
to better than one part in ten thousand
and can be considered a precision calori-
meter like our conventional-sized one.
However, the internal volume (1.85 liters)
of the bomb can contain only the minimum
amount of oxygen usually required for
complete combustion of solid samples (4.1
MPa of oxygen for a 25 gram RDF sample).
This large bomb calorimeter has been
calibrated to better than 1 part/10 000
Table III. Residual Moisture and Ash Data
Sample Calorimeter Samples
Expt. No.
1052
1053
1054
1056
1057
1058
1059
1060
1061
1062
% Moisture
3.74
4.15
3.75
3.89
3.82
3.85
3.88
3.96
3.93
3.88
% Ash
11.57
11.50
13.94
11.94
11.87
12.25
11.84
12.08
11.76
12.04
Large Calorimeter Samples
Expt. No. % Moisture % Ash
1
2
3
4
2030
2032
2033
2034
2038
2039
5.71
5.92
6.82
6.30
15.01
13.37
14.90
14.46
13.32
14.18
246
-------
and perhaps is the most precise large bomb
calorimeter being used in this country
today.
One objective of this research pro-
ject is to decide whether or not the
calorific value of refuse is altered by
the amount of particle size reduction.
For a limited number of experiments in
both the large and small calorimeters, we
found that the calorific value of a given
laboratory sample of Teledyne National RDF
was unaffected by particle size reduction.
As-received extruded RDF pellets were
found to have the same mean calorific
value as pellets prepared from 2 nnn (10
mesh) particles which were obtained by the
milling of as-received extruded RDF pellets
followed by several riffling and blending
operations.
The precision (standard deviation)
obtained from pellets prepared from 2 mm
particles was three times better than that
obtained with pellets where no processing
was done in our laboratory. The reason
for this lower precision is not known but
we suspect that it is the result of a
greater degree of heterogeneity in the
unprocessed sample and the imprecision"in
our moisture determinations. We believe
that more precise moisture data can be
obtained if as-received extruded RDF
pellets are equilibrated at a constant
humidity atmosphere and for each calori-
metric sample, a moisture sample also is
selected. If the calorimetric and moisture
sample have the same history and are
weighed concurrently, then the moisture
contents should be comparable. In addi-
tion, it is worth noting that moisture
determinations on pellets prepared from 2
mm (10 mesh) particle were found to yield
constant values after they were air-dried
and equilibrated at constant humidity.
However, particle size reduction does
alter the moisture content and careful
accounting of the moisture is necessary.
We found that the determination of
the ash content by conventional dry ashing
techniques yields a range of values for
"homogeneous," milled and blended 2 mm (10
mesh) particles of RDF. We suggest that
the non-combustible fraction of RDF clings
to the fibrous RDF particles and is not
evenly distributed by the blending process.
Therefore, we adopted a procedure of
determining the ash from the combustion
residue remaining in the bomb.
In summary, it appears the calorific
value of the Teledyne National RDF is
unaffected by the particle size reduction
procedures carried out in our laboratory.
Also, to obtain good precision, one must
apply the ash residue remaining in the
combustion after an experiment to the
calculation of a given calorific value.
To avoid any sample processing for the
large calorimeter experiments, a "twin"
moisture determination pellet must be
selected and run concurrently with the
calorimetric sample. One can draw the
general inference from the calorimetric
results that processing of a sample,
which has been extruded from minus 1.9
cm (0.75 inch) pieces of RDF, down to a
particle size below 0.2 cm (0.08 inch)
can be attained without changes in
chemical composition or representative-
ness.
REFERENCES
[1] Kirklin, D. R., Mitchell, D. J.,
Cohen, J., Domalski, E. S., and
Abramowitz, S., NBSIR 78-1494,
December 1978.
[2] Coops, J., van Nese, K., Kentee, A.,
and Dienske, J. W., Rec. Trav. Chim.
66, 113-130 (1947).
[3] Coops, J., and van Nes, K., Rec.
Trav. Chim. 66, 131-141 (1947).
[4] Gundry, H. A., Harrop, D., Head, A.
J., and Lewis, G. B., J. Chem.
Thermodynam. 1., 321-332 (1969).
[5] Prosen, E. J., Experimental Thermo-
chemistry , Chapter 6, F. D. Rossini,
Editor, Interscience Publishers, New
York, 1956.
[6] Jessup, R. S., NBS Monograph 7,
1960, "Precise Measurement of Heat
of Combustion with a Bomb Calori-
metery," (U.S. Government Printing
Office, Washington, DC 20420).
[7] Washburn, E. W., J. Research NBS 10,
525 (1933), RP546.
247
-------
TECHNICAL, ECONOMIC AND MARKET EVALUATIONS ON THE USE
OF WASTE GLASS IN STRUCTURAL CLAY BRICK MANUFACTURE
Ashok K. Gupta*
Resource Recovery Systems
Raytheon Service Company
Burlington, Massachusetts
Laura A. Arozarena
Solid & Hazardous Waste Research Division
U.S. Environmental Protection Agency
Cincinnati, Ohio
ABSTRACT
This paper gives the results of an investigation conducted
on the use of refuse-derived waste glass in structural clay brick
manufacture. Waste glass was produced at Occidental Research
Corporation's La Verne, California prototype resource recovery
plant. Laboratory testing was performed at the U.S. Bureau of
Mines, Tuscaloosa, Alabama; and commercial testing was performed
at Glen-Gery Corporation's 120,000 bricks per day plant in
Reading, Pennsylvania. For commercial testing, standard size
(2 1/4 inch x 3 5/8 inch x 8 inch) solid bricks, with and without
glass, were formed by stiff-mud extrusion. These were fired in
a commercial tunnel kiln at a 100 F lower temperature than
normally required for firing regular bricks. Bricks containing
glass were of comparable quality to regular bricks fired at
normal temperature. However, these glass-containing bricks were
of a much higher quality (absorption 35 percent lower and
compressive strength 50 percent higher) than the regular bricks
fired at the same temperature. The optimum waste glass-to-clay
ratio was determined to be 15:85. At this ratio, the following
specifications for waste glass were established: particle size
<100 mesh; organic content <8 weight percent; glass content >60
weight percent; water soluble salt content <0.46 weight percent;
and water soluble sulfates <0.035 weight percent.
-.o
The reduction of firing temperature of the tunnel kiln by
100 F can possibly increase brick production by 22 percent.
Assuming a 15 percent increase in production and energy savings,
*Formerly employed by Occidental Research Corporation, Irvine,
California
248
-------
and considering the cost of handling glass at resource recovery
tof ?« Plants, the value of waste glass was calculated to be
?28.10 per ton. It is our opinion that waste glass slimes can
be used in brick manufacture with little or no further processing.
A conceptual process flowscheme was developed to process the
entire waste glass stream for use in brick manufacture. The net
cost to process this stream from 200 TPD of glass-rich residue
was calculated to be $13.06 per ton of glass product.
Marketing survey results show a concentration of brick
plants generally in the areas of existing or potential locations
of resource recovery plants and a good potential for the consump-
tion of approximately half of the waste glass by the brick
industry.
INTRODUCTION
The U.S. Bureau of Mines
(USBM) has established, by
laboratory tests, that the
finely-ground glass mixture from
incinerator residue is both
suitable and beneficial in manu-
facturing structural clay bricks
(1). This laboratory work has
been confirmed by others (2,3),
and the potential value of such
a product to brick manufacturers
seems widely accepted. The
laboratory tests established
that the use of waste-glass
mixtures in structural clay
bricks improves such important
properties as moisture absorp-
tion and compressive strength.
The reported laboratory tests
established that test bars can
be fired at lower firing
temperatures without sacrificing
quality. This results in an
increased brick production rate.
There have been no
commercial-scale tests on glass-
rich mixtures derived from
incinerator residue and, more
importantly, no major tests
prior to this work on glass-rich
mixtures derived from rat refuse
processing. Of the 15 million
tons of glass discarded in
municipal waste annually, only
2 million tons pass through
incinerators (1), and
this tonnage is declining.
The main thrust of this
program was to demonstrate, by
commercial-scale tunnel kiln
testing, that waste glass
slimes derived from raw refuse
processing act as flux in making
bricks and that the glass
reduces the final firing
temperature required in the
manufacture of bricks. Reduc-
tion of firing temperature
signifies increased production
from kilns with subsequent
reductions in production and
energy costs.
In examining the feasibility
of using raw refuse-derived
glass-rich mixtures in structural
clay brick manufacture, both
laboratory- andv commercial-sdale
testing were employed. The
effect of waste glass addition
'to clay bricks was evaluated
using three select clays.
This investigation included two
types of waste glass mixtures
249
-------
which are current or potential
waste products of most resource
recovery plants: glass cullet
feedstock and glass slimes.
The primary mixture
studied, known as "slimes", is
a waste product of resource
recovery plants using the froth
flotation process for cullet
quality glass. Currently, of
six commercial resource recovery
plants committed for cullet
quality glass recovery, four
plants will be using the froth
flotation process (4). About
(10-15%) of the entire waste
glass stream of a resource
recovery plant reports as
slimes.
The second type of waste
glass mixture is the glass-rich
residue produced by most
resource recovery plants which
separate combustibles,
inorganics and metals thorugh
shredding, magnetic separation,
classifying and screening opera-
tions. This glass-rich mixture
is the feedstock for the cullet
quality glass recovery process
and contains most of the glass
found in raw refuse. Because
of economical and/or marketing
limitations, cullet quality
glass recovery from this mixture
is not feasible at some resource
recovery plants.
One of the major findings
of this study was that waste
glass is valuable to the brick
producer. Due to the lack of
any long-term commercial
testing, one brick producer
designated a value of $4-6 per
ton FOB brick plant for waste
glass, which approximates the
replacement cost of their raw
materials (clays). The cost of
raw materials varies from $2
to $5 per ton in the brick
industry. During this investi-
gation, it was concluded that
a greater than $4-6 per ton
value could be assigned to
waste glass due to decreased
brick production cost and
energy savings by addition of
waste glass to the brick mix.
A conceptual process flow-
sheet, mass balance and
economics for dry beneficiation
of 200 tons per day of waste-
glass residue were developed
for brick use and are summarized
below.
Results of the marketing
survey are presented. Lists
of existing brick plants and
existing or potential resource
recovery plants were prepared.
Locations of brick plants
within a 50-100 mile radius
of resource recovery plants
were identified. More than
25% of the brick companies were
contacted by the Occidental
Research Corporation for the
market survey.
TECHNICAL EVALUATION
The technical evaluation
consisted of exploratory,
laboratory, and commercial
testing at: U.S. Bureau of
Mines, Tuscal'oosa, Alabama;
Glen-Gery Corporation,
Shoemakersville, Pennsylvania;
and Glen-Grey Corporation,
Reading, Pennsylvania.
Materials and Methods
Three types of clays were
included in this evaluation,
ranging in composition from 100%
shale body to 100% clay body.
250
-------
Glen-Gery Brick Mix —
The Glen-Gery brick-mix is
a red burning shale generally
known as Red Shale, which
consists of 100% shale body. It
is mined from the Reading,
Pennsylvania area. This shale
clay is fired to 7-9% absorption
at 1920 -1950 F, and was
included in the study because it
is representative of the brick
mix used to produce approximately
half of the structural clay
bricks produced in the United
States.
Michael-Kane Brick Mix —
This brick mix is 100% clay,
and is generally called
Connecticut Red Clay. It is
mined from the Middletown,
Connecticut area from a rather
recent clay deposit. It is fired
fired to 12% absorption at
1800°F with 72-hour soaking at
1800°F in a scove kiln. This
mix was included in the study
because it is 100% clay and the
brick plant is located within
43 miles of the CEA/ORC resource
recovery plant under construc-
tion at Bridgeport.
San Valle Tile Mix —
This mixture of red firing
clays (60%) and shale (40%),
referred to as San Valle Tile
mix, consists of red and grey
clays and brown shale. The
clays and shale are mined sepa-
rately from the neighboring
areas of Corona, California
(from 20 to 90 miles) and mixed
at the plant site in proportions
of 40% red clay, 40% brown shale
and 20% grey clay. The tile mix
is fired to 9-10% absorption at
1860°F with a 10-hour soaking
at 1860 F in a tunnel kiln.
This mix was included for the
potential application of
research results to the sewer
pipe and roof-tile industry,
whose products command higher
unit prices than bricks.
Exploratory Testing
These tests were conducted
to examine the feasibility of
utilizing waste glass in
making structural clay bricks,
to characterize the raw
materials, and to establish
specifications for waste glass
that can be effectively used
in brick manufacture with
minimal waste glass processing.
The two types of waste
glass used for tests are
described in the following para-
graphs .
Waste Glass Mixture I (WGM-I)—
WGM-I is a fine-sized
glass-rich mixture, generally
known as glass slimes, which is
a waste product of resource
recovery plants using the froth
flotation process for recover-
ing cullet quality glass. This
mixture was reclaimed at
Occidental Research Corporation's
(ORC) glass prototype plant at
La Verne, California. A
schematic flowsheet of this
plant is shown in Figure 1.
The processing steps include
trommeling the glass-rich
material (derived by shredding
and air classifying raw muni-
cipal solid waste at 1/2 inch
screen, and processing the
-1/2 inch fraction through an
organic removal circuit (jig
or screw classifier), grinding
circuit and classification
circuit. In the classification
circuit, finer than 150 mesh
251
-------
FEED TO
RECOVERY
PLANT
ALTERNATOR
"RECYC-AL"'
FERROUS SEPARATOR
METAL
^y LINEAR
INDUCTION
MOTOR
REJECT
NON-FERROUS
METAL
ROD MILL
SCREEN CLASSIFIER "B"
ORGANICS
1
t
METALS AND
ORGANICS
THICKENER TANK
VACUUM FILTER
X^
"SLIMES"
SCAVENGERS"
n n n n
ADDITION OF
•SILECT" REAGENT
J) CONDITIONING
TANK
)"RECLEANERS",
•CLEANERS"
TAILINGS
"ROUGHERS"
GLASS
CULLE
CLASSIFIER "C"
DEWATERING/
Figure 1. Flowsheet of ORC's La Verne Prototype Plant
particles are removed from the
feed to flotation circuit by the
combination of a cyclone and
screw classifier. Coarse but
lighter particles, such as
organics which were not removed
in the organic removal circuit,
are also removed with finer
particles. This stream is
processed through a thickener
and a rotary vacuum drum filter,
producing a cake containing 25%
moisture. This material is then
air dried. The balance of the
material in the plant is pro-
cessed through the flotation
circuit which produces high
purity glass cullet used in
manufacturing glass containers.
WGM-I was produced from
glass-rich material obtained
from the Los Gatos, California
and Ames, Iowa resource recovery
plants. In the Los Gatos,
resource recovery plant, raw
municipal solid waste (MSW) is
processed through a hammermill
and an air classifier. The air
classifier heavies were
purchased from the Los Gatos
plant and were trucked to the
La Verne plant site where they
were stored in roll-off bins,
reclaimed and fed manually to
the trommel infeed conveyor.
In the Ames resource recovery
plant, MSW is processed through
two stages of shredding and air
252
-------
classification. The Ames plant
air classifier heavies contained
a large amount of organics and
were processed through additional
screening and air classification
before processing through the
La Verne glass plant.
Table 1 shows the particle
size distribution and organic
content (as determined by the
loss on ignition at 500°C) of
WGM-I produced from the glass
plant feed from Los Gatos,
California and Ames, Iowa. A
screw classifier was used in the
organic removal circuit to
produce this mixture. The higher
organic content of WGM-I pro-
duced from Ames, Iowa feed is
probably due to two-stage
shredding prior to air classifi-
cation at the Ames plant.
Waste Glass Mixture II (WGM) —
WGM-II is a glass-rich
residue derived at most resource
recovery plants by shredding,
air classifying and screening
the heavy fraction from the air
classifier. The "through"
fraction of the screen is
ground to 80% passing -200 mesh,
a particle size considered most
effective in making structural
clay bricks (1). In areas
where resource recovery plants
are not large enough to support
cullet quality glass recovery
or cullet market outlets are
too distant or limited, pre-
sently this glass-rich material
is lahdfilled, unless an
alternative use is found. The
recovery of this material was
selected as an alternative to
landfilling.
TABLE 1. PARTICLE SIZE DISTRIBUTION AND
ORGANIC CONTENT OF WGM-I
Ames, Iowa
Los Gatos, California
Particle
Size
+100
-100+200
-200+325
-325
%
Retained
23.1
26.0
17.9
33.0
Cumulative
% Retained
23.1
49.1
67.0
JOO.O
% LOI*
31.2
5.9
5.5
7.2
%
Retained
21.2
51.5
68.2
100.0
Cumulative
% Retained
21.2
30.3
16.7
31.8
% LOI*
25.0
4.5
5.5
6.5
Total LOI
*LOI = Loss on ignition
12.2
9.7
253
-------
The sample was prepared
from glass-rich material obtained
from the Ames, Iowa resource
recovery plant. Approximately
600 Ib. of this material was
grounded by SELREX Corporation,
Chatsworth, California, in a
5 ft diameter by 6 ft long batch
ball mill for about ]6 hours.
The ground material was screened
at 100 mesh. The +100 mesh
material was discarded (9% of
total) and 200 Ib. of -100 mesh
material was sent to the USBM
for exploratory testing.
Particle size distribution and
organic content (as determined
by loss of ignition at 700°C)
are given in Table 2. Chemical
and physical properties of WGM-I.
and WGM-II are listed in
Table 3.
Clays —
To keep the number of tests
within manageable limits, only
the Glen-Gery brick mix was used
for exploratory testing. This
shale is a mixture of quartz,
illite, chlorite and a small
amount of geothite.
Chemical analyses and pyro-
metric cone equivalent (PCE) of
Glen-Gery brick mix are listed
in Table 4.
Fluxing Tests —
Preliminary fluxing tests
were made to examine the feasi-
bility of utilizing these waste-
glass mixtures in brick manu-
facture. A mixture of ]0%
WGM-I and II and 90% Glen-Gery
brick mix was used. This ratio
was based on a USBM investi-
gation (1). Results plotted
in Figure 2 show that for a
given finite quantity of avail-
able waste glass, the largest
energy savings to a brick manu-
facturer occurs when bricks are
made with 10% waste glass
rather than with higher glass
compositions. Fluxing test
results determined that waste
glass can be used as flux and
established -100 mesh as opti-
mum particle size and <_8% as
maximum tolerable organic
content.
Twenty 1 inch x 1 inch x 7 inch
bars were formed at the USBM
for each composition by using
a laboratory de-airing extrusion
machine. The bars were dried
for 24 hours in air followed
by 24 hours at 212°F. They
were then fired to 1800° and
1850°F at a rise of 100°F per
hour in an electrically-heated
kiln.
TABLE 2. PARTICLE SIZE DISTRIBUTION AND
ORGANIC CONTENT OF WGM-II
Particle Size
+200 mesh
-200+325 mesh
-325 mesh
Loss on ignition -
% Retained
8.8
19.6
71.6
11.1%
Cumulative
% Retained
8.8
28.4
100.0
254
-------
TABLE 3. CHEMICAL AND PHYSICAL PROPERTIES
OF WGM-I AND WGM-II
Chemical/Physical
Properties
Si°2
A1203
Fe2°3
K20
Na20
MgO
CaO
PH
*PCE (Cone)
(temp.,°F)
*PCE = Pyrometric cone
TABLE 4. CHEMICAL
OF GLEN-GERY
Oxide
Si02
A1203
Fe2°3
K2°
Na20
MgO
CaO
Ti02
Weight
WGM-I
54.6
7.1
2.3
1.6
6.8
1.5
8.0
8.6
03-01
2068-2152
equivalent
WGM-II
54.3
3.8
2.0
0.7
7.9
3.4
12.7
10.5
01
2152
ANALYSES AND PCS
BRICK MIX
Weight %
55.8
22.7
8.4
3.6
0.7
1.1
0.2
1.0
Loss on Ignition at 1000"C - 6.0%
PCE - 13
255
-------
|
0.40
10.30
go.20
tu
ui
0.10
-------
TABLE 5. FLUXING FIRING RESULTS ON 10% ADDITION OF WASTE
GLASS MIXTURES I AND II WITH GLEN-GERY BRICK MIX
Waste Glass
Additions
No addition
WGM-I
WGM-II
% Shrinkage
1800° 1850°
2.8
2.5
3.0
4.8
4.5
4.7
% 5-Hr Absorption Compressive
1800° 1850° 1800°
16.6
14.7
13.8
11.7
10.7
10.6
3450
4050
4150
— i - — ^^^•^^—
Strength, psi
1850°
6500
7150
5950
TABLE 6. FLUXING RESULTS ON 10:90 GLASS-CLAY MIXTURE
WITH VARIOUS PARTICLE SIZE GLASS
Waste Glass % Shrinkage % 5-Hr Absorption Compressive Strength,psi
Additions 1800° 1850° 1800° 1850° 1800° 1850°
-100 mesh
-150 mesh
-200 mesh
2.8
3.8
2.8
4.5
4.8
4.7
14.5
13.8
14.0
10.1
9.5
9.5
3300
4150
4150
5750
6700
5800
bricks containing waste glass
with more than 8% organics.
Waste glass samples con-
taining 0% to 12% organics were
made by burning the waste glass
at 700°C and then blending the
burnt waste glass with unburnt
waste glass to desired propor-
tions. Fluxing tests were made
on 1 inch x 1 inch x 7 inch
bars of 90:10 clay mixtures.
Fluxing results are given in
Table 7.
These test bars did not
show any black coring. It was
felt that this effect may be
due to the smaller cross
sectional area of the test bars
permitting complete carbon burn-
out. Therefore, black coring
tests were conducted on standard-
size, hand-molded bricks with
90:10 clay-glass mixture. Waste
glass mixtures used in this test
contained 0%, 5%, 8%, 10%, and
15% organics by weight as de-
termined by loss on ignition
at 700°C. These hand-molded
bricks were fired in the tunnel
kiln during regular production
at Glen-Gery Corporation's
Reading, Pennsylvania plant.
These fired bricks were cut in
half by a diamond saw to exam-
ine the inside of the brick for
black coring. Bricks contain-
ing waste glass with 0%, 5%,
and 8% organics did not show
any black coring, whereas the
bricks containing waste glass
with 10% and 15% organics did
show black coring.
Based on the above results,
the following specifications
were establis-hed for waste '
glass to be used for laboratory
and semi-commercial testing:
1. Particle size - minus
100 mesh
257
-------
TABLE 7. FLUXING TESTS OF 90:10 CLAY-GLASS MIXTURE,
WITH VARIOUS GLASS ORGANIC CONTENT
Water
% Organic Absorption
Content of 24 Hr Cold 5-Hr Boil Sat. Coef.
Waste Glass 1850°F 1950°F 1850°F 1950°F 1850°F 1950°F
0 3.1
2 3.5
5 4.0
7 4.8
8 4.2
12* 3.9
1.
1.
1.
1.
1.
1.
2
5
4
8
7
4
6.3
6.8
7.1
8.6
7.6
6.7
3.
3.
3.
4.
4.
3.
0
8
6
6
0
4
0.
0.
0.
0.
0.
0.
49
51
56
56
55
57
0
0
0
0
0
0
.39
.41
.39
.39
.42
.40
Modulus of
Rupture
psi
1850° 1950°
3798
3357
3347
3476
3604
3136
3933
4347
4275
4502
4374
3818
*Fired separately
2. Organic content <_8% by
weight
3. Glass distribution >_5Q%
(qualitative determi-
nation of the samples
by microscopic exami-
nation)
Specifications for water
soluble salts could not be es-
tablished since test bars did
not effloresce.
Laboratory-Scale Tests
Based on the above specifi-
cations, about 3 tons of waste
glass slimes were prepared for
laboratory and semi-commercial
scale testing. This 3-ton sample
was shipped to Glen-Gery Corpo-
ration, Reading, Pennsylvania.
Raw Materials Procedure —
The waste glass slimes were
produced by the WGM-I processing
method described earlier except
that the organic removal circuit
of the glass plant used a Wemco
reamer jig instead of a screw
classifier. Removal of organics
by the jig was more efficient
than by the screw classifier
and waste-glass slimes contained
less organics (6-8%) than slimes
from the screw classifier
(10-15%). The particle size of
slime was finer in the jig
(1-2%; +100 mesh) than the screw
classifier (16-18% +100 mesh).
The size distribution of
the sample of waste glass
slimes is given in Table 8.
Glen-Gery Corporation
sampled 400 Ibs from the ship-
ment and sent a 300-lb sample
from this to the USBM in
Tuscaloosa for laboratory
evaluation. The balance was
used at Glen-Gery for water
soluble determination and
fluxing tests.
Clays from the three
locations described earlier
were used for this test.
Chemical analyses and pyro-
metric cone equivalents of
258
-------
TABLE 8. SIZE DISTRIBUTION OF
WASTE SLIMES USED FOR LABORA-
TORY AND SEMI-COMMERCIAL
SCALE TESTS
Particle
Size
Mesh
-100+200
-200+325
-325
Wt %
31.0
11.4
57.6
Cumulative
Wt %
31.0
42.4
100.0
LOI at 800°C - 7.9% by wt
Glass Distribution >_50%
Microscopic Determination
these clays and waste glass
slimes are given in Table 9.
Laboratory Procedure —
Mixtures of Glen-Gery,
Michael-Kane and San Valle clays
and 10-, 15-, and 20- percent
waste glass slimes were tempered
in water in a dough mixer and
then formed in 1 inch x 1 inch
x 7 inch bars by stiff-mud
extrusion. After drying at
212°F, 20 bars of each compo-
sition were fired to 1850°F,
1950°F, and 2050°F at a rise of
100°F per hour. Five bars of
each composition were also fired
in the production kilns of the
respective brick plants.
Results of Laboratory-Scale
Tests —
Table 10 shows the results
of Glen-Gery clay with waste-
glass slimes. These results
show that the 15% addition of
waste glass slimes decreased
the boil absorption by 34% and
increased the compressive
strength by 55.7% at 2050°F.
Results of test bars fired in
the production kilns simulated
the firing condition of the
laboratory at 2050°F. The
effect of lowering the firing
temperature by waste glass
addition is shown in Figure 3,
where the addition of 15% water
glass slimes caused a decrease
of 45°F at the 8% absorption
level.
The firing results on San
Valle and Michael-Kane brick
mixes did not show any signifi-
cant effects by the addition
of 10-, 15-, and 20-percent
waste glass slimes. Our scope
of work did not permit us to
investigate other additions of
waste-glass slimes, additional
firing temperatures and the
effect of "soaking" on the
fired product containing waste
glass. Both San Valle and
Michael-Kane mixes are "soaked"
for long periods, 10 hours and
TABLE 9. CHEMICAL ANALYSES AND PYROMETRIC CONE
EQUIVALENT OF CLAYS AND WASTE GLASS SLIMES
Clay
Si0
K20 Na20 MgO CaO Ti02 LOI PCI
Glen-Gery
Michael-Kane
San Valle
Waste Glass
Slimes
55.8
54.1
64.3
59.4
22.
17.
15.
7.
7
1
7
1
8.4
5.7
3.6
3.1
3.6
3.4
2.4
1.2
0.7
2.4
1.8
7.9
1.1
3.0
1.1
1.8
0.2
2.2
0.9
8.6
1.0
0-9
0.8
0.3
36.0
a6.9
a5.6
b7.9
13
2
3-4
04-08
1000°C
3 800°C
259
-------
TABLE 10. RESULTS OF FIRING TESTS ON GLEN-GERY CLAY
WITH VARYING ADDITIONS OF WASTE GLASS SLIMES
Shrinkage
%
No slimes -
1,850°F
1,950°
2,050°
plant run
10 percent
slimes -
1,850°F
1,950°
2,050°
plant run
15 percent
slimes -
1,850°F
1,950°
2,950°
plant run
20 percent
slimes -
1,850°F
1,950°
2,050°
plant run
0.5
2.3
4.8
5.7
2.0
3.7
6.5
6.7
2.5
4.0
6.3
6.5
2.8
3.5
6.2
6.0
Boil
Absorption
%
18.0
13.8
8.2 •
8.0
15.8
11.5
6-6
6.9
14.9
11.6
5.4
7.1
16.4
13.2
7.3
8.1
Soak
Absorption
%
16.8
13.1
6.6
6.5
14.7
10.5
4.8
4.6
13.7
10.4
3.8
3.9
14.9
11.7
5.6
6.1
Compressive
Saturation Strength,
Coefficient psi
0.93
0.95
0.80
0.81
0-93
0.91
0.73
0.67
0.93
0.90
0.70
0.55
0.91
0.89
0-77
0.75
2,575
4,150
7,275
6,750
3,475
5,350
9,450
6,900
4,525
6,225
11,325
10,100
3,150
5,150
10,250
9,000
72 hours, respectively. (The
maintaining of the fired product
at the final firing temperature is
known as "soaking" in the brick
industry.) We believe that the
lack of data precludes any conclu-
sions in regards to the fluxing
ability of waste glass with these
brick-mixes containing predomin-
antly clay. We recommend addi-
tional research in the area so
that the fluxing ability of waste
glass with predominantly clay-
containing brick mixes can be
examined.
Firing temperature *?
Figure 3. Effect of Slimes
Addition on Firing Temperature
with Glen-Gery Clay Mix
260
-------
Laboratory fluxing tests
with 10-, 15-, and 20-percent
waste glass additions in the
Glen-Gery brick mix were also
made at the Glen-Gory Corpora-
tion. Results indicated that a
15% addition of waste glass
slimes improved absorption by
35% and modulus of rupture by
59%.
Based on the results from
the U.S. Bureau of Mines and
Glen-Gery Corporation, a 15%
addition of waste glass slimes
to Glen-Gery brick mix was
determined to be most effective.
Semi-Commercial Scale Tests
Shuttle Kiln Procedures —
Standard size solid bricks
(2 1/4 inch x 3 5/8 inch x 8
inch) with and without glass were
extruded for the tests. The
waste glass slimes addition was
maintained at about 15% by
weight, using premeasured
buckets for supplying the glass
fraction into the brick mix.
Green bricks were made using
commercial-scale equipment. The
brick mix and waste glass slimes
were mixed in a pug mill and
formed into standard bricks by
extruding the mixture through a
die (stiff-mud extrusion) as a
continuous extruding column and
cut into proper size by a reel-
type wire cutter. Bricks with
glass contained 15% moisture
compared to 16% for bricks
without glass.
The extruded bricks were
set on wooden skids and placed
among the tunnel kiln for pre-
liminary drying. Final drying
was done in the shuttle kiln.
The semi-dry bricks were set on
shuttle kiln cars in a ratio of
2/3 regular bricks and 1/3
bricks with a glass addition.
The bricks were dried in
the shuttle kiln for about 36
hours on a small fire (approxi-
mately 600°F) and then all
burners were lit and the auto-
matic firing control was turned
on. The temperature was con-
trolled by four thermocouples;
two positioned at the bottom of
the kiln and the other two
positioned at the top of the
kiln. In all, five sets of
bricks were fired in the shuttle
kiln. Due to thermocouple cal-
ibration problems in the shuttle
kiln, two sets were also fired
in the tunnel kiln with the
regular production. Due to
inaccurate calibration of ther-
mocouples , an additional two
sets, in addition to the origi-
nal three sets planned, were
fired. Tests were planned so
that the first set was fired to
maturing temperature of the
regular bricks; the second set
at somewhat lower maturing temp-
erature; and a confirming third
set. Again, due to calibration
problems, the first three tests
at 1900°F, 1800°F and 1930°F
did not produce matured regular
bricks (8% absorption). After
each firing, representative
samples of bricks, with and
without glass, were selected
and sent to McCreath Laboratories
(an independent ASTM certified
laboratory) in Harrisburg,
Pennsylvania for testing. The
testing at McC-reath Laboratories
included cold, and boiled water
absorption, initial rate of
absorption (suction), compres-
sive strength and efflorescence
tendency.
261
-------
Results of Semi-Commercial Tests —
In the shuttle kiln, the 15%
addition of waste glass slimes
did not significantly affect the
strength and water absorption of
the bricks except at temperatures
of 1950°F and 1970°F, where waste
glass addition increased the
compressive strength of the bricks
by approximately 20%. Interest-
ingly, the results on the two
sets fired in the tunnel kiln
approached a 33% decrease in boil
absorption and up to 76% increase
in compressive strength by the
addition of 15% waste glass
slimes. These bricks, however,
showed high degrees of scumming
and efflorescence which was not
expected and which is not desir-
able mainly because of the
appearance of the bricks.
Scumming and efflorescence
are caused by the presence of
water soluble salts in the brick
mixture. The contents of water
soluble salts in the fired pro-
duct (with glass) and waste glass
slimes were 0.35% and 1.14%,
respectively. Waste glass slimes
contributed approximately half of
the water soluble salts contained
in the fired brick. The content
of water soluble sulfates in the
waste glass slimes was 0.18%.
Commercial Scale Tests
The objective of the tunnel
kiln test was to demonstrate, in
a real plant environment, the
feasibility of lowering the
firing temperature of bricks by
100°F without negatively affect-
ing the quality of the bricks.
Lowering of the firing tempera-
ture implies increased kiln
throughout and reduced energy
cost to the structural clay
brick industry.
Approximately 2.5 tons of
waste glass slimes containing
a low soluble salt content were
sent to the Glen-Gery Corpora-
tion for testing in the 120,000
bricks per day capacity tunnel
kiln. The waste glass slimes
were produced in ORC's La Verne
pilot plant in a similar way as
described earlier except a screw
classifier instead of a jig was
used for the removal of organics
from the glass plant feed. The
jig was employed for slimes used
in laboratory and semi-commer-
cial tests. These waste glass
slimes reported 1.53 percent of
water soluble salts (compared to
1.14 percent for jig slimes), 12
percent organics (compared to 8
percent for jig slimes) and 30
percent glass (compared to 50%
glass in jig slimes). Glass
content of the screw classifier
slimes was determiend by sizing
it into three fractions and then
petrographically analyzing each
size fraction.
This 2.5 tons of waste
glass slimes were prepared by
removing water soluble salts by
additional washing on the rotary
vacuum drum filter, air drying,
screening at 100 mesh and by
blending the -100 mesh fraction
with -100 mesh glass cullet to
bring the glass content to 60
percent by weight of the sample.
Currently, all glass flotation
plants are using a jig for
organics removal and it is
expected that the glass content
of the jig glass product will
contain much higher than 60
percent glass content. The
glass content of the USBM,
Edmonston, Md. pilot plant jig
glass product has been in the
neighborhood of 85-90 percent
glass. Composition of this
262
-------
sample is given in Table 11 and In excess of 4000 extruded
partical size distribution and bricks containing waste glass
chemical analysis is given in slimes were distributed on seven
Tables 12 and 13 respectively. kiln cars. The balance of the
car capacity was filled with
Procedure — standard size solid bricks with-
out waste glass slimes. A fully
The standard size bricks loaded car contains 3600 bricks.
containing 15 percent waste glass
slimes were prepared by following The bricks were dried in a
the same procedure as used for tunnel dryer and fired in the
shuttle kiln tests. BaCC>3 was tunnel kiln (Figure 4) . Cars
added (0.5 Ibs per ton) to the were moved at the rate of 60
brick mixture to reduce the minutes a car length. As the
efflorescence tendency. first car approached the firing
zones, the burners were turned
down to lower the temperature in
the firing zones. Heat treatment
of the bricks is shown in
Table 14.
TABLE 11. COMPOSITION OF WASTE GLASS SLIMES USED FOR
TUNNEL KILN TESTS
Composition Weight Percent
water soluble salts 0.46
water soluble (calcinated @ 1200°F) 0.30
water soluble sulphates 0.035
glass content 60.0
organic (LOI @ 700°C) 6
TABLE 12. PARTICLE SIZE DISTRIBUTION OF WASTE GLASS SLIMES
USED IN TUNNEL KILN TEST
Particle Size Weight Percent Cumulative Weight Percent
+60 mesh 0.6 0.6
-60+120 3.8 4.4
-120+165 7.2 11.6
-165+200 14.6 26.2
-200+325 23.8 50.0
-325+400 5.6 55.6
-400 44.4 100.0
263
-------
TABLE 13. CHEMICAL ANALYSIS OF WASTE GLASS SLIMES
USED FOR TUNNEL KILN TESTS
Oxides
SiO0
2
AL2°3
Fe203
K00
2
Na-0
2
MgO
CaO
TiO0
2
Weight Percent
63.8
4.0
1.7
1.07
10.8
1.44
9.16
0.26
?f
ZONE
27I26E5
24l23l22
ZONE
2
21|20|1S
ZONE
1
!8|l7fT6
5
01
01
a.
a.
I5rt4ii3naii
3^
« P
§5
c
HDI9I8I7I6
5
<
25
SI41312
Ol
fc
Ol
>
CAR POSITION
$ THERMOCOUPLES
FIGURE 4. TUNNEL KILN
TABLE 14. HEAT TREATMENT OF EXPERIMENTAL BRICKS
Car
No.*
59
126
29
100
44
11
29
Zone
1680
1660
1665
1650
1660
1680
1695
2
1660
1665
1650
1660
1680
1695
1700
1665
1650
1660
1680
1695
1700
1705
Zone
1720
1720
1720
1715
1735
1740
1760
3
1720
1720
1715
1535
1740
1760
1760
1720
1715
1735
1740
1760
1760
1775
Zone
1835
1820
1810
1810
1805
1800
1820
H-WMVi^^H^^MBVB
4**
1820
1810
1810
1095
1800
1820
1825
^— MH^HOaVMB^^M
1810
1810
1805
1800
1820
1825
1835
^^^M^M^MPVIVMM
28
1805
1820
1818
1810
1820
1820
1830
fc— ^^^*-"*^""-
29
1820
1805
1800
1805
1800
1810
1820
*Each car contained regular bricks and a few hundred bricks with
15% glass addition.
**During regular production, the temperatures in the fourth zone are
maintained between 1900°F - 1930°F.
264
-------
After the cars came out from
the kiln, the bricks were sampled
for testing at the McCreath Lab-
oratories in Harrisburg, Penn-
sylvania. The brick containing
waste glass slimes were tested
for each car separately, while
regular bricks were tested as a
composite sample.
The pyrometric cone equi-
valent (PCE) tests on shale-slime
mixtures by the U.S. Bureau of
Mines showed a good uniformity in
distribution of waste glass
slimes in the extruded bricks.
compressive strength 9000 -
11000 psi, boiling water absorp-
tion 7% to 10%, IRA 25-35 and
with no efflorescence are con-
sidered good quality bricks.
All the experimental bricks
containing waste glass slimes
which were fired at 100°F lower
temperature were comparable to
good quality bricks. Regular
bricks which were fired along
with glass-containing bricks
showed 30% higher absorption
and 50% lower compressive
strength. Only suction rate
and the efflorescence were not
satisfactory although the
Results of the Commercial Test — values were of no serious
Results from testing the
fired brick are compiled in
Table 15. Regular bricks with
consequence.
TABLE 15. TUNNEL KILN TEST RESULTS
Car 5 Hr.
No. %
59
126
19
100
44
11
29
Composite
Sample of
Regular
Bricks
10
10
10
10
9
9
9
13
WATER
Boil
.0
.5
.0
.4
.8
.4
.1
.1
ABSORPTION
24 Hr.
7.
8.
7.
8.
7-
7.
6.
11.
Cold
7
0
6
1
4
0
5
3
Satur.
Coeff .
0
0
0
0
0
0
0
0
.75
.76
.75
.77
.76
.74
.7
.86
I.R.A.
(Suction
Rate)
45
47
40
48
39
44
35
47
Compressive
Strength
psi
9
9
8
9
10
9
10
6
i
t
i
r
i
r
r
r
360
950
560
090
190
630
920
670
Ef f lorscence
Effloresced
Effloresced
Effloresced
Effloresced
Effloresced
Effloresced
Slightly
Effloresced
No Efflor-
escence
265
-------
Summary of Commercial Tests —
Firing Time—Because of the lower
maturing temperature of bricks
containing glass, these bricks
can be fired in less time than
bricks without glass (1,4,).
This means reduced production
cost due to increased production
rate without any significant
additional cost. A typical
firing curve for the Reading
plant tunnel kiln (used for
testing) is shown in Figure 5.
The preheat and firing zone tem-
peratures are plotted propor-
tional to time or kiln length,
since the charging rate of a
tunnel kiln determines the firing
time. Although the actual tem-
peratures with respect to time
vary from one clay to another,
depending upon the impurities
and composition of the clay; the
curve shown in Figure 5 is also
typical for high temperature,
red-firing clays.
Figure 5 shows that during
firing of regular bricks, the
1800°F temperature is reached
within 18 hours contrasted to
the 1900°F reached in 24 hours.
Allowing for the normal 3-hour
hold period, glass bricks will
mature in 21 hours, 6 hours less
than regular bricks (normal
maturing time, 27 hrs.). This
is based on the normal charging
schedule of one car per hour
(24 cars per day) practiced in
firing solid bricks at the
Reading plant. This would mean
that glass bricks can be charged
at a rate of 1 car every 46.7
((21x60)/27) minutes or 30.8
((24x60)/46.7) cars per day
instead of 24 cars per day of
regular bricks.
Therefore, potential
increase in production using
15% waste glass addition amounts
to 22%. This increase would
amount to 8.8 million bricks per
1940
1850
1800
u. 1760
UJ
oc
1670
ui
0.
1580
1490
1400,
Figure 5.
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
TIME IN HOURS OR KILN LENGTH
FROM PREHEAT TO FIRING ZONE
Typical Firing Schedule of Tunnel Kiln
Used For Waste Glass Test
266
-------
year for the Reading plant which
normally produces 40 million
bricks per year. At an FOB
price of $85.00 per thousand, the
added production output would
increase plant sales by $748,000
per year.
Heat Requirements—The heat
requirements of a tunnel kiln
depend on many parameters, such
as burner type, kiln insulation,
percent excess air, stack temp-
eratures, and efficiency of
recuperation.
The Reading brick plant
consumes 1400 BTU's per Ib of
solid bricks or 2.8 million BTU's
per ton of bricks. The normal
daily energy requirement of the
Reading plant at 300 tons per day
of brick production is 840
million BTU.
Savings in firing time, as
calculated earlier by 15% addi-
tion of waste glass to clay,
amounted to 22%. Therefore, by
the addition of glass waste, the
heat requirement over the total
production will at least be
reduced by 22% of by 184.8
million BTU per day, on the basis
that incremental bricks did not
require any additional work.
ECONOMIC EVALUATION
During this study, a real-
istic value for waste glass
needed to be established by tak-
ing into consideration costs of
handling waste glass and benefits
realized by the utilization of
waste glass in brick manufactur-
ing. The basis of the economic
analysis was as follows:
o Conventional Brick 300 TPD
Plant Capacity (120,000
Bricks Day)
Modified Brick
Plant Capacity
Glass Product
in Brick Mix
Quality of Glass
Product Used
Reduction in
Maturing Temp.
of Bricks
345 TPD
(138,000
Bricks Day)
15%
Air Dry
100°F
This cost/benefit analysis showed
that net value of waste glass to
a brick manufacturer is $28.10
per ton of waste glass (Table 16).
To establish this value, the cost
consideration included, capital
investment (silos and material
handling equipment) and incre-
mental operating costs to the
brick manufacturer. The benefit
consideration allowed for only a
15 percent increase in production
(instead of the possible 22 per-
cent) ; a brick selling price at
$85/1000 (instead of an average
price of $112/1000 bricks as
determined in the market survey),
and an energy savings due to the
15 percent increased production.
In the writer's opinion, the
slimes produced at a commercial
glass recovery plant can be used
in bricks with minimal processing,
if any. It is believed that
drying may not be necessary for
handling purposes and that air
dried material can be handled
satisfactorily.
A conceptual process flow
scheme was developed for dry
processing an entire waste glass
stream for brick use. Based on
the flow scheme, the net cost of
producing a glass product, as
given in Table 17, was deter-
mined to be $13.06 per ton of
waste glass (includes a $4 per
ton for transporation to brick
plant).
267
-------
TABLE 16. VALUE OF GLASS TO BRICK MANUFACTURERS
Incremental Investment by Brick Manufacturer $94,000
Energy Savings $/Ton Glass Product(a' $ 4.85
Credit for Incremental Production (15% More)
$/Ton Glass Product $29.42
Credit for Displaced Clay(d) $ 0.67
Capitalized Charges @ 26.2% Capital Investment $(1.37)
Incremental Operating Costs $(5.47)
Net Value of Glass Product $28.10
a) Based on 15% Increase in production
b) Bricks ass'umed to sell at the price of $85/1000 bricks.
c) Incremental production assumed to be 15%. Possible increase
in production is estimated to be 22%.
d) Clay assumed to cost the brick manufacturers $5/ton.
TABLE 17. DRY PROCESSING OF GLASS-RICH
RESIDUE FOR BRICK USE
o Cost Summary
Capitalized Charges
Annual Operating Cost
(After By-Product Credits)
Freight (@ 45 miles)
Net Cost
$ 4.0I/Ton Product
5.05/Ton Product
4.00
$13.06/Ton Product
268
-------
MARKETING
The market potential for
waste glass utilization in
structural clay brick manufacture
was calculated, taking into con-
sideration the estimated quanti-
ties of waste glass available
and its location relative to the
location of brick manufacturers.
The location of brick plants in
the U.S. is given in Figure 6.
The study results are as follows:
o 45% of resource recovery
plants are or will be within
50 miles radius to brick
producers (Figure 7).
o Over 90% of resource recovery
plants are or will be within
100 miles radius to the brick
producers (Figure 8).
o 103 brick customers were
contacted; 53 responded. Out
of 53 responses:
- 50 companies made face
bricks
40 companies used a tunnel
kiln
23 companies used shale or
shale-clay mixture
average production was
120,000 bricks per day
- average price was $111.93
per thousand bricks.
CONCLUSIONS
Results of the investigation
indicate that waste glass derived
from raw refuse is an effective
flux for shale body clays.
Results were inconclusive on clay
body clays. Although bricks
showed efflorescence tendency, it
was not considered of serious
consequence.
Substitution of 15% waste
glass for clay reduced the
firing temperature by 100°F and
produced good quality bricks.
More than 4000 glass containing
bricks were made in the tunnel
kiln. This could mean possibly
22% increase in production of
the existing tunnel kiln and a
22% energy savings.
Based on 15 percent increase
in production, the value of
waste glass slimes to the brick
producer was calculated to be
$28.10 per ton of slimes. The
waste glass slimes can be used
as flux in bricks without any
additional processing.
The net cost of producing
a glass product was calculated
to be $13 per ton of waste
glass.
The market study indicated
that approximately 45% of the
waste glass generated by resource
recovery plants can be poten-
tailly used by the brick
industry.
ACKNOWLEDGEMENT
This research was supported
by the U.S. Environmental Pro-
tection Agency, Municipal Envi-
ronmental Research Laboratory
under Contract No. 68-03-2469,
Mr. Donald A. Oberacker, Project
Officer.
269
-------
ho
»J
o
Figure 6. Location of Brick Plants in the U.S.
-------
Figure 7. Locations of Existing Brick Plants and Existing/Potential
Resource Recovery Plants Within 50 Mile Radius
-------
•-J
NJ
Figure 8. Locations of Existing Brick Plants and Existing/Potential
Resource Recovery Plants Within 100 Mile Radius
-------
REFERENCES
1. Tyrrell, M.E. and A.G. Goods.
"Waste Glass as a Flux for
Brick Clays." BuMines Report
of Investigation, 7701, 1972.
2. Dr. Gil Robinson's work and
comments. Reported by Harvey
Gershman in Enclosure I of
ASTM E-38.06. Minutes of
Philadelphia Meeting, Nov-
ember 11, 1975.
3. Tyrrell, M.E., I.L. Field
and J.A. Barclay. "Fabrica-
tion and Cost Evaluation of
Experimental Building Brick
from Waste Glass." BuMines
Report of Investigation,
7605, 1972.
4. NCRR Bulletin, Voluem VII,
No. 3, Resource Recovery
Brief, March 1978.
273
-------
THE PRODUCTION AND USE OF DENSIFIED REFUSE DERIVED FUEL
Carlton C. Wiles
U.S. Environmental Protection Agency
26 West St. Clair Street
Cincinnati, Ohio 45268
ABSTRACT
As a resource recovery alternative, the use of a densified form of refuse derived
fuel (d-RDF) is being investigated as a substitute for coal in industrial spreader stoker
boilers. Experiences are summarized from the production of approximately 1700 tons of
1/2 inch diameter pellets using a modified animal feed pellet mill. Storage and other
handling experiences are also discussed.
Approximately 400 tons of the d-RDF have been burned in a spreader stoker equipped
boiler in a heating plant. A full battery of emissions tests were conducted and ob-
servations made to determine the effects on boiler operations. Results of these tests are
presented.
INTRODUCTION
The use of refuse as an energy source
is a potentially attractive resource recov-
ery alternative. While conversion of the
organic fraction of solid waste to liquid
or gaseous fuels (i.e., pyrolytic oils,
alcohols, methane, etc.) has not been com-
mercially implemented, there are facilities
operating today which combust the refuse
either directly for steam recovery or in
combination with fossil fuels for power
generation. The latter involves the proc-
essing of the refuse to remove the combusti-
bles for use in a modified power generating
boiler, usually in combination with coal.
The processed refuse is usually referred to
as refuse derived fuel (RDF). Examples of
RDF plants in operation are Ames, Iowa and
Milwaukee, Wisconsin. These systems, es-
pecially Ames, used the experience gained
from the St. Louis-Union Electric-EPA ref-
use derived fuel demonstration project in
St. Louis, Missouri. There are other RDF
plants in various phases of startup, con-
struction, or design.
The RDF concept in the United States
was originally considered primarily for ma-
jor utilities generating power typically
designed to burn pulverized coal. These
utility boilers often produce more than
400,000 pounds of steam per hour. The use
of RDF, however, does not have to be limit-
ed to this larger user and, in fact, may
offer more value as a fuel for other small-
er power generating facilities, i.e., the
small industrial and institutional boiler
owner. Many of these boilers are stoker-
fed and usually designed to produce 75,000
to 200,000 pounds or more of steam per hour.
This is an attractive alternative. One rea-
son for this involves the inability of the
smaller fossil fuel user to receive quantity
discounts for purchase of primary fuel. He
therefore pays higher prices for his coal
as compared to the larger utility. In this
instance, the RDF may have a higher dollar
value to him, which in turn results in a
higher return to the resource recovery fa-
cility producing the RDF. Another reason
involves the expansion of the potential mar-
ket area for the RDF to smaller communities
that today cannot afford recovery of refuse
as RDF for a larger utility user. Other
marketing factors such as increased flexi-
bility in contract negotiations, especially
in length of commitment, may also increase
the attractiveness of the smaller user as a
market for RDF. Additionally, there are a
greater number of these small power generat-
ing boiler facilities (as compared to
274
-------
the larger utilities) located in both large
and small communities. Many of these coal
burning units are older, prior to 1940, be-
cause newer ones were built or converted to
use oil or gas. The use of RDF in these
facilities, many of which are economically
marginal, may help to reduce the impact of
costly air pollution control equipment.
Refuse derived fuel prepared for the
larger utility boilers is typically the
light fraction of shredded refuse which has
been air classified, screened, or otherwise
processed to remove the noncombustibles.
In this fluffy form it can be pneumatically
fed into the suspension utility boiler. In
contrast, a densified form of RDF is probab-
ly more desirable for use in the stoker fed
boilers. This densified refuse-derived
fuel (d-RDF) may approximate the physical
characteristics of the stoker coal fed to
the boiler. This offers the possibility of
increased flexibility in transporting, stor-
ing, and otherwise handling the RDF. Of
major importance in this respect is the po-
tential capability of mixing the d-RDF di-
rectly with the coal for feeding to the
boiler with only minor or no modifications
required.
The Municipal Environmental Research
Laboratory (MERL) in Cincinnati has major
EPA responsibility for R&D in the recovery
and utilization of municipal solid waste
materials. Naturally, programs have includ-
ed the recovery of energy from solid waste.
This would appear to be the most attractive
resource recovery alternative available to-
day. It is an alternative that could pro-
vide a return sufficient to yield an eco-
nomically successful plant, while helping
to solve the real problem of solid waste
disposal.
While experiences with RDF use were
available and being added to, little infor-
mation was available on d-RDF, especially
concerning its production. Therefore, EPA
implemented a program to investigate the
technical and environmental aspects of pro-
ducing and using d-RDF. If proven to be
technically and economically feasible and
environmentally acceptable, the use of d-
RDF may be an attractive resource recovery
option for many communities. This paper
summarizes the EPA d-RDF program and pre-
sents results of studies conducted thus
far.
EPA PROGRAM IN d-RDF
Although this paper will discuss EPA's
d-RDF program, it is not intended to imply
that the concept originated with EPA. There
have been a number of trials of burning
densified forms of waste materials. How-
ever, despite these successful trials,
little testing results and operational ex-
periences are available. EPA considered
that for the use of d-RDF to become widely
implemented a credible experimental program
was needed to establish the environmental
acceptability and show that the concept is
economically and technically sound. Of ma-
jor importance is the effects on boiler
physical facilities, boiler operations, and
the environment from burning the d-RDF.
The EPA program objective is to provide the
necessary experimental data and operational
experiences to interest potential users to
implement the concept. More specific ob-
jectives can be summarized as follows:
o To establish the equipment modi-
fications necessary to utilize d-RDF in ex-
isting stoker-fired boilers. Modifications
might well include changes in fuel feed de-
vices, grate speed controls, overfire and
underfire air, and dust collectors. The
necessary modifications should be specified
as a function of the boiler type. Hopeful-
ly, experiences will show that only minimal
modifications will be required.
o To establish the characteristics
of the d-RDF required for combustion in
stoker-fired boilers. These must also be
established from a materials handling view-
point. Properties of interest are the size
and shape of the densified fuel, its den-
sity, its moisture content, its mechanical
strength, and its constituents. Refuse
processing required to produce a specifica-
tion d-RDF will require study.
o To establish the operating con-
ditions necessary for a stoker-fired boiler
burning coal and d-RDF. These conditions
will include coal-to-refuse ratio, air
distribution (primary to secondary), air
temperature, excess air, and other char-
acteristics.
o To establish the influence of d-
RDF on the boiler performance. Information
is needed to determine effects on boiler
efficiency, efficiency of dust collection
and other emission control equipment, and
associated factors such as fouling and
corrosion.
275
-------
o To establish the environmental
impact from the use of d-RDF. Influences
on air emissions, ash quantities and quali-
ty, water emissions where applicable, and
similar factors must be determined. The
fate of trace compounds is also important.
o To establish the economic impact
of burning d-RDF. This involves not only
the impact on the user facility, but an es-
timation of the potential economic impact of
d-RDF as a solid waste resource recovery al-
ternative for a community. Cost of produc-
ing a specification d-RDF will have a major
impact upon the economics of a d-RDF sys-
tem.
Briefly, the program involves four
projects. Initially, two concurrent projects
were implemented.
o A research grant was awarded to
the National Center for Resource Recovery
(NCRR) to study production requirements for
d-RDF. The establishment of production
costs and characteristics of a specifica-
tion d-RDF were considered important. Ma-
terial handling during production, trans-
port, storage, and use were also import-
ant. Additionally, NCRR was to produce
d-RDF for combustion tests.
o A competitive contract was award-
ed to Systems Technology Corporation to
establish the combustion characteristics of
the d-RDF and its effects on boiler perform-
ance, environmental impact, and other fac-
tors associated with the use of d-RDF.
o Later in the program, a second
research grant was awarded to the University
of California. This project is supplement-
ing the NCRR grant in looking at the basic
theoretical aspects of producing and using
d-RDF.
o More recently, in an effort to
make more meaningful test burns, a contract
was awarded to Teledyne National to produce
2000 tons of d-RDF. The production is being
done at the Cockeysville, Maryland resource
recovery plant. This project will also pro-
vide additional information on production
processes.
Information presented in this paper is
based upon experiences gained thus far from
conduct of these projects. Approximately
1700 tons of d-RDF have been produced. Of
this approximately 400 tons have been com-
busted with coal in stoker boilers. Addi-
tional combustion tests are being implement-
ed.
PRODUCTION, HANDLING AND STORAGE
CONSIDERATIONS
PRODUCTION
Considerable experience has been gain-
ed in the production of d-RDF. Since 1976,
the National Center for Resource Recovery,
Inc. (NCRR) has pelletized approximately
1400 tons, including approximately 100 tons
of 1-inch diameter pellets. Three hundred
tons of pellets were produced in late 1976
through the Spring of 1977 for combustion
tests at Hagerstown, Maryland (Figure 1).
This process was modified by installing
a zig zag air classifier (in place of the
"Vibrolutriator") which was fed with a large
vibratory feeder. This was done in an at-
tempt to provide a cleaner feed to the
pellet mill and thus reduce die and roller
wear and reduce ash content of the d-RDF.
After these modifications, an additional
1100 tons of d-RDF were produced. Proper-
ties of the d-RDF are summarized in Table 1.
The NCRR processes were limited because of
space and equipment considerations but can
be used to illustrate a process flow. The
production run currently being made by
Teledyne will provide information on a modi-
fied process using a trommel! to screen the
air classified light fraction prior to
secondary shredding. This process is ex-
pected to provide a cleaner feed to the
pellet mill and thus result in a lower ash
d-RDF. Appropriate production and use ex-
perience will determine if this happens.
Most data and observations presented
are based upon the NCRR production.
Feedstock Preparation
The preparation of the feedstock to
the pellet mill is important, as its condi-
tion may have the most significant impact
upon pellet mill performance and product
quality. The process used at NCRR re-
sulted in problems caused by textiles. Al-
though 95 percent of the feed may be minus
1-1/4 inch, textiles predominate in larger
sizes, generally 2 to 4 inches. The tex-
tiles are difficult to shear through the
pellet mill die and tend to bunch up behind
the rolls and slip until the pad breaks
loose and causes the machine to jam.
276
-------
AIR CLASSIFIED
LIGHT FRACTION
PNEUMATIC
CONVEYOR
3/16" SCREEN
Figure 1: NCRR PROCESS FLOW
USED TO PRODUCE INITIAL
d-RDF SUPPLY
CALIFORNIA
PELLET
MILL
TRIPLE "S" VIBROLUTRIATOR
DISCHARGE TO
DUST ROOM
CYCLONE
MOSTLY
INORGANIC
FINES
LIVE-
BOTTOM
BIN
FEEDER
HEIL-TOLLEMACHE
SHREDDER
d-RDF PRODUCT
277
-------
TABLE 1
d-RDF PROPERTIES AS PRODUCED AT NCRR
1/2"
1"
Moisture
(X, D.W.)
Ash
(X, D.W.)
-3/8" Fines
(X. A.R.)
Pellet Density
(g/cc)
Bulk Density
(LB/CF, A.R.)
Mean Length
(mm)
Fall, Spriru)1977 Fall 1977-Fall 1978
19.1 22.9
June 1978
26.6
N/A
1.17
38.6
17.9
23.4
14.9
1.01
35.9
15.9
22.8
22.8
9.6
0.76
27.0
N/A
The particle size distribution (PSD)
of the nontextile feed is not as signifi-
cant a factor as the textiles. Other types
of size reduction equipment such as knife
shredders will provide better control of
textiles. These, however, may have prob-
lems from relatively narrow tolerances, sen-
sitivity to tramp metals and inerts, and
higher maintenance costs.
Schemes to screen the light fraction
to remove inerts, bypass undersize material
around the secondary shredder and concen-
trate the textiles in smaller fraction
would seem to be an easy way to alleviate
some of the operational and maintenance
problems of the NCRR approach. The produc-
tion run at Teledyne may yield some of this
information.
Densification
Actual densification of the refuse was
done in an animal feed mill built by the
California Pellet Mill Company (CPM). The
mill channels input materials between a
rotating 33 inch, i.d., 49 inch o.d., die
and two stationary rollers which force the
feed through the die holes. The die rotates
at 170 rpm and is driven by a 150 horsepower
motor. Both 10 and 13 inch diameter rolls
were utilized, with 85 percent of the 1976-
77 production on 10-inch rolls and virtual-
ly 100 percent of the 1977-78 production on
13-inch rolls. While the larger rolls im-
proved reliability, no impact on pellet pro-
perties was apparent. Dies with 1/2-inch
hole diameters were used except during June
1978, when an alternate die with 1-inch dia-
meter holes was tested. Die selection was
dictated by the objective of making pellets
replicating stoker coal properties.
Application of the pellet mill to MSW
uncovered problems with capacity, operation-
al reliability, and maintenance that were
unexpected or inexperienced in the wide-
spread applications to agriculture, animal
and wood product wastes. One reason for
this relates directly to the differences in
feedstock and machine operations with the
solid waste as compared to the machine's
normal use.
Following are important examples of
these differences:
o The particle size (PSD) of the
feedstock is smaller than the die opening
on all applications except for MSW.
o The MSW feedstock density at 2-4
pounds per cubic foot (pcf) is only half
that in most other application.
278
-------
o The feedstock moisture in other
applications is maintained at lower or more
uniform moistures of 12 to 15 percent by
drying or steam addition.
o The inert and tramp metal levels
in other feedstocks do not approach that of
MSW.
o The pellet product is cooled and
screened to remove the fines for recycling
in most applications. Cooling apparently
aids the integrity of the pellets.
Regardless of these differences, pel-
lets can still be produced from sized MSW
light fraction over a range of throughputs,
moisture, particle size (excepting tex-
tiles) and inert levels.
Production expectations of 10 tons per
hour of 1/2-inch diameter d-RDF pellets
were not realized with the pelletizer. Ex-
periences with the 1/2-inch die indicate
that 5 tons per hour is the maximum through-
put achievable with this mill, while most
production was at lesser rates. One reason
for this would be the difference in the
actual MSW feedstock density (2-4 pcf) com-
pared to 7 to 9 pounds per cubic foot pre-
sumed by the manufacturers. Higher capa-
cities may be achievable with larger dia-
meter pellets and cleaner feedstocks (i.e.,
less of textiles and inerts) with smaller,
more consistent particle size distributions.
Cost of producing the d-RDF is expect-
ed to have a significant impact upon the
economics of the d-RDF concept. Die wear of
the pelletizer was expected to be important
because of costs associated with repair or
replacement.
For the material and inert levels at
NCRR, the die life was on the order of 1000
tons and the roller life 500 tons. However,
since the inert concentration was much high-
er than necessary, it is difficult to extra-
polate these findings. By controlling the
inerts to 8 to 12 percent, the investiga-
tors estimate a production life of 3000 tons
on the die and 1500 tons on the rollers. A
further complication in projecting costs
for die and roller replacement is that the
die may be re-tuned and the roller shells
alone replaced, thus postponing full re-
placement costs.
Although in operation only briefly, to
produce approximately 100 tons, the 1-inch
die seemed more tolerant of oversize tex-
tiles. The 1-inch pellets produced had
lower densities than the 1/2-inch pellets.
This is probably attributable to the taper
of the die used and thus could be improved
by using a die with more taper.
Because the densification, and es-
pecially the throughput of the densifier,
will have a significant impact upon econom-
ics of d-RDF, additional work with various
densifier designs is needed. There is now,
however, considerable experience available
to aid this work. Additionally, better in-
formation is required to determine exactly
what the d-RDF has to be for acceptable use
in the boiler, since this will help deter-
mine the amount of production processing
required.
HANDLING AND TRANSPORTING
Pellets produced at NCRR were trucked
to Haqerstown, Maryland (approximately 75
miles) in 20 cu. yd. drop boxes. Boxes
were covered. Shipments from D.C. to Erie,
Pennsylvania where Phase 2 testing will
take place were made in dump trucks. No
apparent problems have been experienced.
Care must be exercised, however, to ensure
that container beds be clean of trash, or
other materials which might contaminate the
d-RDF. Containers should also be properly
covered to prevent blowing of the material
and/or wetting of the d-RDF.
Movement of the d-RDF from storage for
loading or other reasons has normally been
by front-end loader or similar equipment.
Blending with the coal at Hagerstown and
movement to the existing coal weigh lorry
was done with a system consisting of 2
hoppers, 2 belt conveyors, and a bucket
elevator which discharged into a chute
feeding the lorry. As a precaution, this
was used instead of the plant's only exist-
ing coal silo system. After proper adjust-
ments, the system operated well to blend
the coal:d-RDF and to feed the lorry. This
indication that the coal:d-RDF blends can
be handled by existing plant equipment has
been verified in preliminary trials at the
Phase 2 test site and will be further con-
firmed during detailed testing.
The major problem experienced with the
d-RDF at the boiler plant was dusting. Ap-
propriate precaution to control dusting will
279
-------
be necessary in normal operations.
STORAGE OBSERVATIONS
Storage of d-RDF will be an important
consideration in production and use. Pro-
duction schedules have not corresponded to
combustion tests schedules for various rea-
sons; one of which involved delays in ob-
taining a test boiler. This circumstance
provided good opportunities to observe the
reactions of the d-RDF to various storage
regimes. Storage has occurred at four dif-
ferent locations; Washington, D.C., Hagers-
town, Maryland (2), and Erie, Pennsylvania.
Hagerstown Storage
d-RDF storage for the Hagerstown test
involved both indoor and outdoor storage
over approximately 2 months. Storage pro-
cedures utilized were based upon those
recommended for lignite coal, i.e., keep-
ing storage periods as short as possible
and pile depths to no greater than 6-8
feet. The d-RDF storage in the warehouse
resulted in no apparent problems, although
what appeared to be fungal growth did take
place on the surface of the 6-foot piles.
Some mildly offensive odors were also
noted, but no rodent damage was observed.
Additional storage at Hagerstown was
done by piling the d-RDF pellets 6-7 feet
high onto a concrete pad. The pellets were
covered with a plastic tarpaulin. During
storage, moisture released from the pellets
accumulated on the undersurface of the
plastic and wetted the surface of the d-RDF
pile. This moisture caused pellet deterio-
ration and caking, but only at the surface
of the pile. Pellets exposed to the excess
moisture swelled to cause some minor mate-
rial handling problems.
Although limited storage of pellets in
20 cu. yd. drop boxes caused pellets to
freeze near the edges, they were easily
broken and caused little difficulty.
Washington. D.C. Storage
After the Hagerstown burn tests, the
program was to proceed to a second phase
test burn in a larger more typical boiler.
Because of problems encountered in locating
and arranging an acceptable site, pellets
produced by NCRR had to be stored near the
production site. Precaution dictated that
the storage be outside. Storage was ex-
pected to be short term, but as the diffi-
culties in locating a second phase test site
mounted, delays pushed the storage into
months. Observations made during this stor-
age are interesting.
The original storage plan, dictated by
the expected short term and non-permanent
nature of the storage requirement, involved
the following:
o 1/4 acre sloped asphalt pad;
o 6 foot high contained pile;
o plastic tarp cover anchored by tires
to keep water out and prevent blow-
ing material;
o placement in the pile of net bags
of pellets of known properties;
and
o measurement of temperatures to as-
sure there was no tendency toward
spontaneous combustion.
As observed at Hagerstown, the covers
provided a surface on which moisture gene-
rated from inside the pile would condense
and rain back down on the surface. Mois-
ture levels within the wet cap that formed
over the entire pile reached 130 percent
dry weight or saturation. Although it was
felt leaving the pile uncovered may have
been preferable, concern over blowing paper
was overriding and the plastic tarps were
continued.
Temperature monitoring showed a ran-
dom pattern by depth and location with a
range of 40 to 70°C, nothing above that ex-
pected from biological degradation (Figure
2).
Seven months into storage and with no
problems apparent, the height of the pile
was increased in some areas to 10 to 15
feet. This was dictated by the need for
more storage capacity. In the process, a
portion of the wet cap was covered and the
material thoroughly aerated. As to whether
this event triggered later more active bio-
logical and/or chemical activity is not
known, but suspected.
After this event and during the sum-
mer months, temperature profiles near the
280
-------
Sampled
09 to 12/
Samp!ed
01 to
Sampled
04 to
04/78
TOP VIEW
Mean °C at Top Probe Location
Mean °C at Middle Probe Location
Mean °C at Bottom Probe Location
Highest Probe Temperature- 90°C(surface)
Lowest Probe Temperature- 38°C(base)
"T" indicates probe location
SIDE VIEW
Figure 2: LOCATION OF TEMPERATURE PROBES AND TEMPERATURES
RECORDED IN d-RDF STORAGE PILE
281
-------
top increased with maximums of 85 to 90 C.
In August, 10 months into storage, a series
of smoldering seams of pellets with tempera-
tures above 150°C and hot enough to ignite
plastic or tires surfaced in an area where
the pellets had been repiled. Excavations
and measurements provided a great deal of
data but yielded no conclusive explanation:
o Whereas the as-produced moisture
varied from 15 to 35 percent, seams of pel-
lets with from 4 percent to 131 percent
moisture were uncovered. They ran general-
ly horizontally through the pile at varying
depths and thicknesses (Table 2).
o The hot smoldering seams were
randomly located, often not connected and
adjacent to both very wet and very dry
materials.
o The mulch smell, darkened color
and increased levels of fines in many parts
of the pile indicated that it had compost-
ed.
o The smoldering, charred seams
extended back into portions of the pile
that were still at the original 6-foot
depth, but were more extensive in the re-
piled portions.
o While the angle of repose of the
as-produced pellets was 35 to 70°, when
the stored pellets were excavated, they
often formed stable walls at 90° or more.
These episodes naturally raise ques-
tions as to the cause and to what extent
the particular production, handling, and
storage techniques may have contributed to
the problem. Predictive schemes for the
reactive mechanism(s) involved are impos-
sible to develop for such a heterogenous
material. Possibly brought on by the
aeration and increased moisture of repil-
ing, increased biological decomposition
provided sufficient activation energy to
trigger a chemical reaction and subsequent
higher enthalpies from chemical decomposi-
tion. These hot seams would dry adjacent
pellets and then ignite and begin to oxi-
dize them.
The extended storage and handling
necessitated by the smoldering naturally
caused pellet deterioration. Fines were
significantly increased. By this time a
second phase burn test site had been iden-
tified in Erie, Pennsylania. In order to
verify that the d-RDF would be useable for
the tests, approximately 60 tons (4 truck
loads) were shipped to Erie and burned.
Despite the deteriorated conditions and
relatively large fines content, the test
was extremely successful. The plant offici-
als witnessing the day's operation indicat-
ed minor problems of some odor and dusting.
The remaining NCRR pellets were ship-
ped to Erie, Pennsylvania for test burning
in January-February 1979. Thus, some of
the pellets will have been in storage for
over 12 months at the time of use. Obser-
vations will continue. Additionally, the
Teledyne pellets will be stored in Erie,
Pennsylvania, prior to a separate test burn.
The storage will be for much shorter peri-
ods. Comparison of these pellets and their
condition with the NCRR pellets during com-
bustion will be important. The information
will help better define what "d-RDF" must
be for acceptable use.
Although there is little doubt stor-
age, and particularly the handling in and
out of storage, will increase fines and re-
duce pellet length, density and bulk den-
sity, the result should never approach what
was observed in this program. First, stor-
age beyond four weeks in a commercial op-
eration will probably not be necessary and
the effect of less than a month's storage
will be minimal. Second, proper storage
conditions including shallow piles and a
ventilated cover will minimize the chance
of degradation. Third, screening and cool-
ing the pellets as they are produced will
probably make them more resistent to de-
terioration.
COST CONSIDERATIONS
Costs associated with producing d-RDF
are uncertain and are still being developed.
Factors to be considered include densifier
die and roller wear, equipment maintenance
requirements, the amount of refuse preproc-
essing required prior to introduction into
the densifier, and the final characteristics
required for an acceptable d-RDF.
Cost will, of course, be greatly de-
pendent upon throughout capacities of the
densification equipment. It is evident
that the pellet mill used at NCRR could not
meet expected capacities. However, much
operating experience was gained which can
be used to improve future production faci-
lities. Additionally, alternative densi-
282
-------
TABLE 2
SUMMARY OF d-RDF STORAGE PILE DATA
August 25-September 1, 1978
00
10
Sample
1
2
3
4
5
6
7
8
9
10
Descri pt1 on/Preval ence
1 ft. from base-bagged sample -
21 wk. in pile
4 ft. from base-bagged sample -
21 wk. in pile
6 ft. from base-bagged sample -
21 wk. in pile
Oxidized, friable mat'l - reddish
brown - very localized in pockets
Light brown - typical of 70-80%
of central pile
Prevalent light brown material
along side slope
6-inch thick white strata over-
laying #6
6-inch thick dark strata over-
laying #7
Localized wet dark brown
strata
Localized wet black strata -
Temp.
65°C
N/A
N/A
75°C
68°C
40°C
44°C
58°C
75-85°
Moisture
57%
32%
4%
5%
18%
6%
17%
131%
47%
44%
-3/8" Fines
27%
7%
8%
31%
23%
15%
12%
36%
35%
45%
Mean Length
8.8 mm
11.3 mm
10.8 mm
9.4 mm
9.4 mm
10.9 mm
9.5 mm
7.1 mm
7.5 mm
6.3 mm
crumbly
-------
fication equipment has not been thoroughly
tested with MSW.
Thus far, work has shown that the pel-
lets can be satisfactorily handled in exist-
ing boiler equipment. Some modifications
to stoker feeders and other equipment may
be required but are expected to be minor.
The costs associated with use of the d-RDF
in power generating facilities are also
being developed.
USE OF d-RDF IN COMBUSTION TESTS
The objectives of the combustion tests
involved determining the effects of burn-
ing d-RDF (with or without coal) on the
boiler, its operation, associated facilities
and equipment, and the environment. Infor-
mation, data, and observations discussed
are based primarily on the experimental com-
bustion tests conducted at the State of
Maryland Correctional Institute for Men,
located near Hagerstown, Maryland. The
boiler plant consists of a battery of three
150-psig Erie City boilers. Their design
steam ratings are 78,500, 60,000 and 25,000
Ibs/hr. Each unit is equipped with Hoffman
Combustion Engineering "Firerite" spreader
stokers to distribute the lump fuel in the
furnace. The large coal pieces that do not
burn in suspensions, are combusted on the
surface of Hoffman vibrating grates, and
ash is discharged to the front.
The Erie City boilers have the tube-
and-tile furnaces. The water-walls are com-
posed of wide-spaced nominal 3 1/4-inch
diameter tubes that were later partially
embedded in refractory to approximately 8
feet above the grate. The gases are ex-
hausted from the furnace through a two-
drum boiler bank, consisting of rows of
2 1/4-inch diameter tubes, with two gas
passes. The flue gases are cleaned in a
two-stage multiclone collector. The fly
ash gathered in the first-stage collector
is reinjected into the boiler to complete
combustion of the fly char, and the fly ash
in the second-stage collector is pneumati-
cally transported to disposal. The cleaned
gases are induced through a centrifugal fan
and exhausted to a common breeching and
stack.
Approximately 280 tons of d-RDF were
burned at the Hagerstown test site. Table
3 presents properties of the d-RDF as used
at the site. The following summarizes the
length of combustion testing for the vari-
ous blends:
Blend Ratio
(coal:d-RDF)
1:1
1:2
0:1
Properties of the blends used in the
May tests are shown in Table 4. During the
May tests, one boiler was continuously fir-
ed with coal:d-RDF blends (141 tons of d-
RDF) for a period of 132 hours. Boiler
loads were varied between approximately 17
and 53 percent of boiler load rating. The
test program was designed to acquire three
replications of emissions data for each
blend. Periods for boiler stabilizations
on each blend were reduced as were other
observations because of the limited supply
of d-RDF available for testing.
BOILER PERFORMANCE
Tests were conducted and observations
made to measure the effects of d-RDF:coal
blends on boiler operations. These includ-
ed d-RDF cold flow and hot flow tests; ob-
servations of effects on boiler operations;
effects on ash handling system; air and gas
handling; fouling, slagging, and wastage;
boiler control requirements; mass and ener-
gy balances; and others. Observations are
summarized.
Cold Flow
Tests showed that the d-RDF could be
satisfactorily fed to the boilers through
the existing feeders at settings normally
used for coal firing. Fuel distribution
patterns were acceptable with the d-RDF
being fed to the desired furnace locations.
The spreader performed as designed in that
the largest pellets with the greatest mass
travelled to the rear of the boiler with the
fines falling closest to the spreader.
Hot Flow and Blend Testing
Prior to blend testing, approximately
1 ton of 1/2-inch diameter pellets were
fired without coal to help determine limita-
tions on boiler performance. It was neces-
sary to adjust the fuel trajectory to pre-
vent fuel and flame impingement on the rear
wall. Shortly after starting, the steam
pressure decreased from approximately 153
284
-------
TABLE 3
AVERAGE d-RDF PROPERTIES ON AN AS-RECEIVED AND A MOISTURE AND ASH FREE
BASIS AS USED IN THE HA6ERSTOWN TEST BURNS
to
00
Ul
As Received
% Moisture
% Ash
% Volatile
% Fixed C.
Btu/Lb.
Dry Basis
% C
% H
% N2
% Cl
% S
% Ash
% 02
Btu/Lb.
Fusion
Initial
1st Softening
2nd Softening
Fluid
Mineral Analysis
Phos. Pent, Ox
Silica
Ferric Ox.
Al umi na
Titania
Sodium Ox.
Potassium Ox.
Lime
Magnesium
Sulfur Tri Ox.
Undetermined
December
Average
13.4
19.97
56.54
10.1
6488
43.98
5.29
.35
.40
.40
23.19
30.80
201 8°F
2088°F
2175°F
2275°F
.87
55.52
2.27
13.45
.66
6.82
1.30
10.75
1.14
6.03
1.19
March
Average
12.62
24.41
54.08
8.89
5534
39.17
4.47
.39
.45
.26
27.97
27.30
2040°F
2103°F
2155°F
221 5°F
.73
71.58
2.89
4.43
.99
5.66
.53
7.50
1.12
1.22
1.87
May
Average
12.22
28.75
49.27
9.76
5266
35.63
4.54
.85
.36
.28
33.02
25.33
2005°F
2105°F
2125°F
2225°F
.65
63.65
2.64
8.39
.69
7.53
.91
9.74
1.59
3.20
l.QO
December March May
Average Average Average
MOISTURE & ASH FREE
85.04 85.80 83.38
14.97 14.21 16.62
54.29 54.42 53.36
6.53 6.20 6.75
.43 .54 1.25
.49 .62 .54
.53 .36 .42
38.02 37.87 37.68
9785 8772 8956
-------
TABLE 4
PROPERTIES OF THE COAL:d-RDF BLENDS USED IN MAY*
AS FIRED
VOLUMETRIC BLEND
PARAMETER
Moisture*
Volatiles
Fixed Carbon
Ash
Btu/lb
MJ/kg
Carbon
Hydrogen
Oxygen
Nitrogen
Sulfur
Chlorine
FUSION TEMP. °C
1:0
1.3
22.6
54.2
22.0
11706
1:1
6.6
31.7
38.4
23.3
8988
.2
.5
.3
27.
66.
4.
3.4
1.3
1.2
.05
.9
.1
20
54,
4.1
9.8
1.1
.86
.15
1:2
7.9
38.3
29.1
24.7
8382
19,
47.
4.1
14.2
.9
.66
.21
0:1
16.6
48.6
9.0
25.9
5130
11.9
30.9
3.8
21.8
.6
.23
.33
Init. Def.
1st Soft.
2nd Soft.
Fluid
% Weight Rate d-RDF
% Heat Rate d-RDF
1274
1308
1335
1371
0 35 52
0 20 37
1116
1151
1179
1213
100
100
*Unless otherwise noted, all values are weight percent on a wet basis.
psi to about 149 psi and generally stayed
at this level. The drop occurred because
the feeders were not capable of feeding a
sufficient amount of d-RDF to the boiler to
maintain steam pressure. The steam flow
remained approximately the same at about
30,000 pounds per hour.
Following the successful 0:1 test, a
1:1 blend was fired. During this test
there was frequent clinkering on the grate.
One cause of this clinkering is that when a
fuel with too low an ash fusion temperature
is placed on the grate and agitated, hot
particles with molten edges roll against
each other as the grate is vibrated. As
the particles are quenched, they stick to-
gether and begin to form a slag layer across
the entire surface of the grate. This slag
layer becomes progressively less permeable
than the open portions of the fuel and ash
bed, consequently less air is supplied to
the fuel/ash bed experiencing clinkering.
As fuel is fed into the furnace and falls
on top of the pancake, the airflow is furth-
er retarded and the clinkering condition is
further aggravated. In a short time, smok-
ing becomes severe and the fuel depth over
the clinkering area on the grate increases
appreciably.
The ash fusion temperature analysis of
the coal revealed that the hemispheri^re-
ducing ash fusion temperature was 1477 K
(2170 F). [Detroit Stoker Manufacturing
Company recommends that Eastern bituminous
coals should have a minimum ash fusion tem-
perature of 2300°F (1260°C) for vibrating
grate applications.] Consequently, the
clinkering experienced during the coal only
burns was not surprising. Since continua-
tion of this clinkering during blend firing
would not permit evaluation of the effect
of pellets, procuring a different coal for
286
-------
subsequent tests was Indicated.
Even though the pellets have an ash
fusion temperature less than that of coal,
the furnace operated satisfactorily when
the 100 percent pellets were burned. Ap-
plying the ASME definition [Winegartner,
E.G., Coal Fouling and Slagging Parameters,
American Society of Mechanical Engineers,
1974.] for lignite type ash as a calcium
plus magnesium oxide to iron ratio being
greater than one, indicates that d-RDF will
behave similar to a lignite coal. There-
fore, boiler design rules for lignite
rather than those for bituminous would be
expected to apply when burning d-RDF.
During January, testing switched from
boiler No. 2, a 60,000 Ib/hr boiler with
three spreaders, to boiler No. 1, a 78,500
Ib/hr boiler equipped with four spreaders.
During the January tests, the boiler was
fired with a 1:1 blend for 4 hours and with
100 percent pellets for 2 hours. During the
1:1 firing, the overfire air and fly ash re-
injection jets were adjusted to achieve the
desired flame mixing and low smoke opacity.
The plume was relatively clear during both
the 1:1 and 0:1 blend firings. While firing
with 1:1 blend, the clinkering experienced
during the December tests did not occur.
This improved performance occurred as a re-
sult of the coal having a hemispheric fu-
sion temperature of about 1646°K (2471°F).
The December coal fusion temperature was
1477°K (2170°F).
When the blend ratio was changed to
100 percent d-RDF, the fuel bed remained
clear for the first 20 to 30 minutes of
firing and there was no significant smoking.
Subsequently, windrows were observed on
the fuel bed in line with fly ash reinjec-
tion ports. Each of the windrows was in
line with the double fly ash reinjection
ports (one reinjector for the ash caught in
the tube passes, the other for the first
stage multiclone). The windrowing phenom-
enon could be corrected by spacing the fly
ash reinjection ports uniformly across the
back wall of the furnace instead of placing
them side-by-side. The windrows moved well
through the furnace as the grate vibrated
and remained sufficiently porous for air-
flow. It was further observed that a base
of burned out ash existed under the wind-
rows. This observation is further support
for the conclusion that the fly ash rein-
jection ports had not properly distributed
the ash over the grate.
The furnace volume appeared more than
adequate for combustion. The boiler, how-
ever, could only generate a maximum of
24,500 kg/hr (54,000 Ibs/hr) of steam when
firing 100 percent pellets. This unmodifi-
ed stoker-fed, vibrating-grate boiler
supported 70 percent of the nominal rating
for 2 hours without changing the feeder
when firing 100 percent pellets. The mag-
nitude of this derating is the amount pre-
dicted by the volumetric limitations for
the feeder. Also, the blower capacity was
sufficient to meet the underfire air re-
quirements. There was no apparent reason
why the unit could not support full boiler
capacity if the feeders were sized to pass
the required amount of fuel on a Btu basis.
When a 1:1 blend was fired, there was
no difficulty in maintaining the boiler
steam pressure. However, when 100 percent
pellets were introduced, the steam pressure
fell to about 145 psig while carrying a
54,000 Ib/hr steam demand. This drop in
pressure occurred due to volumetric feeding
limitations of the spreader stokers. After
the initial loss in pressure, the boiler
operated in a stable condition for the rest
of the test.
The principal change required during
the 1:1 and 0:1 blend tests was to reduce
spreader rotor speed such that the throw of
the pellets was approximately 6 inches less
than that for 100 percent coal. This ad-
justment was necessary to prevent rear wall
fuel impingement.
During the combustion tests with 1:1
blend and 100 percent d-RDF, the fireball
was kept well away from the walls of the
furnace by adjustment of the overfire air.
Once these jets were adjusted for minimum
smoke and maximum efficiency for coal only
burning, they continued to meet the mixing
and wall protection requirements when burn-
ing blends and 100 percent pellets. As
viewed from the side of the furnace, firing
both pellets and blends, the bed was well
burned out by the time it approached the
front ash pit. The flame pattern above the
grate indicated that the fuel bed was main-
taining proper porosity and that the com-
bustion was good. With little attempt to
optimize the system, a 10 to 12 percent
carbon dioxide content in the flue gas at
the boiler outlet was readily obtained.
287
-------
Normal Boiler Operation
The cold and hot flow tests showed
that d-RDF could be successfully introduced
into the boiler. The blend also had a dis-
tribution pattern on the furnace floor
which was similar to that of coal. This
finding is not particularly surprising
since the size distributions and material
densities of the coal and pellets were
similar. Therefore, with the same velo-
city and angle of injection into the fur-
nace, pellets and lumps of coal with equal
weight would be expected to carry approxi-
mately the same distance.
There was a severe ash accumulation
on the back furnace wall in line with one
spreader during the May testing. To stop
the slagging, the throw of all pellets was
reduced approximately 6 inches, and the
circular tray in the spreader was pushed in
slightly to reduce the arch of the pellet
trajectories. After this adjustment, the
pellets still carried the same distance
from the feeder and landed at the same point
on the grate. However, the 6-inch throw
reduction eliminated the ash accumulation
on the furnace wall, and minimized smoking.
In December, January, and March, there were
no problems of material impinging on the
side walls. In May, however, the same
spreader had to be adjusted because of its
throwing too far and spraying the side wall
of the furnace. Careful measurement of the
spreader showed that the circular tray had
worked left. Once the tray was properly
aligned, the fuel impingement on the left
side wall was eliminated.
During January and March, boiler No. 2
had a recurring problem of clinkering on
the left side when burning coal. Reports
received after the field testing stated
that this clinkering had been eliminated by
readjusting the spreader circular tray.
Proper adjustment of the spreader
stokers is critical to the successful com-
bustion of coal:d-RDF blends. During part
of the March tests, there was a reoccurring
clinkering in front of one spreader, but
the rest of the fuel bed remained free-burn-
ing. Clinkers formed on top of burnt-out
ash and moved out of the furnace with dif-
ficulty. The back wall of the furnace re-
mained clear throughout the clinkering.
During an unexpected furnace outage caused
by a control loop failure (the d-RDF in the
furnace was not related to this failure),
the furnace spreaders were inspected. While
the other spreaders were clear, the speci-
fic spreader had a heavy accumulation of
partially pyrolized pellets in the feed
throat. Careful measurement of tray posi-
tion from the inside of the boiler indicat-
ed that the pellets were being thrown at
too high an angle out of this spreader,
ricocheting off the refractory feed throat,
and accumulating at a point in the furnace
approximately two-thirds of the way back.
This misadjustment was solved by adjusting
the circular tray. The furnace was then
brought back on line from a cold start with
a 1:2 blend and run continuously for 48
hours with no further clinkering.
Ash Handling
The boilers tested are equipped with
dry pneumatic ash handling systems includ-
ing an ash pit, cyclone separator, and stor-
age silo. Observations regarding ash were
made as to grate performance, bottom ash,
ash storage silo, and reinjection ash flows.
Some relatively minor adjustments were
necessary to the grate operation when fir-
ing blends in order to maintain a proper
bed. As the d-RDF substitution ratio in-
creased, the frequency and/or duration of
the grate pulses were increased to maintain
uniform bed depth. Pulse frequency was the
main control.
As expected, the quantity of bottom
ash as measured in the pit increased with
d-RDF. However, the conventional pneumatic
ash handling system adequately processed
the coal:d-RDF ashes.
In the storage silo, some difficulty
was noted with 100 percent d-RDF firing.
The bottom ash was so fine that it would
not de-intrain properly in the cyclone caus-
ing evential plugging of the ejector.
Observations made of the reinjection
ash flow indicated that the fraction of ash
falling to the ash pit was roughly independ-
ent of the type fuel being used (i.e., coal
only or d-RDF). It was, however, dependent
on the ash content of the blend. Further
observations indicated that the combustion
in these boilers was improved as d-RDF was
substituted for coal. Because of this, the
relative ash flow measurements for various
coal:d-RDF blends are perhaps deceptive.
288
-------
There was a trend toward finer ash with
increasing d-RDF content.
Air and Gas Handling
The boilers tested had three air sys-
tems: underfire air, overfire air, and an
induced draft fan. Observations of operat-
ing requirements for these indicated that
coal:d-RDF blends can be fired in the same
settings as coal only if adequate air is
available to prevent clinkering during load
sheds. The overfire air flow needs to be
increased approximately 100 percent to com-
pensate for the higher volatile content of
the d-RDF as compared to the coal.
Fouling, Slagging, and Wastage
The coal;d-RDF blend firing resulted in
occasional slagging, slight fouling, and
perhaps wastage. However, the boilers were
operated at low loads and high furnace tem-
peratures were not encountered. Even under
these conditions, the slagging and fouling
problems were corrected with boiler adjust-
ments. Metal wastage observations were
complicated because of inadequate exposure
times and methodologies available for the
tests. Wastage rates observed, however,
were low. This observation is encouraging
especially when one considers that experi-
ences at Ames, Iowa, have shown that after
1000 hours of in situ boiler tube exposures
to RDF/coal burning, virtually no corrosion
occurred. Year exposure results are appar-
ently similar.
Energy and Mass Balances
Energy and mass balances for the tests
are summarized in Tables 5 and 6. The high
carbon losses in the refuse are unusual and
account for the low efficiences. In review-
ing the mass balance, an abnormally high
amount of ash reported to the collector.
Such an ash weight distribution is not nor-
mal for expected boiler performance.
Because a significant drop in carbon
content of the collector ash occurred with
increasing d-RDF (for 1:1 to 1:2 blends)
effectively offsetting wet flue gas losses,
the boiler efficiency did not appreciably
change.
A positive result of the tests was the
substantially improved low-load performance
and decreased plume opacity when d-RDF was
substituted for coal.
SUMMARY OF BOILER TESTS
The Hagerstown experience has contri-
buted to available knowledge of blend be-
havior in a spreader stoker. The fuel en-
tered the furnace satisfactorily, burned
well, and met plant energy requirements.
The operative and control differences en-
countered were all manageable, and simple
adjustments usually resolved them. Some
biasing of the air controls was required to
prevent slagging on the fuel bed during
load sheds. The only other limitations on
the boiler operation occurred when the
boiler was operated on 100 percent d-RDF.
During this time, both the spreader and ash
handling systems became capacity limited.
Proper adjustment of the spreader-
stoker is critical to prevent slagging and
fouling. Some slagging and fouling occurred
on the walls slightly above the grates but
was readily removed. The corrosion experi-
ment produced material wastage results com-
parable to what one might expect for coal-
only firing. This test was too short in
duration, however, to permit any definite
conclusions with regard to material wastage.
ENVIRONMENTAL TESTING
The emissions testing performed are
indicated in Table 7. Although the quanti-
tative results for the coal:d-RDF emissions
are significant, principal conclusions are
based on comparisons of coal:d-RDF blends
with coal-only results. Baselines were es-
tablished before and after each blend run
by replicating all tests for coal-only fir-
ing conditions.
Particulate Emissions
Opacity—
Opacity decreased with increased d-RDF
in the blends. Because the boiler plant
tested has a significant amount of carbon
carryover when coal is fired, most of the
plume opacity reduction measured can be
attributed to improved combustion conditions
when the blends were fired.
A Battelle Tenex Sampler for trace
organics was only used three times, i.e.,
at 1:0, 1:1, and 1:2 test blends. The
Tedlar bag provided a record sample of the
flue gas. When the measured emissions pro-
289
-------
TABLE 5
HEAT BALANCE SUMMARY
AS RECEIVED
BLEND
PARAMETER
Fraction of Rating
Excess Air (%)
LOSSES
Dry Gas
Fuel Moisture
H20 for H2 Combustion
Combustibles in Refuse
Radiation
Unmeasured
TOTAL
EFFICIENCY
Fuel %
Blend Flow Ash in
Coal:d-RDF Kg/hr. Fuel
1:0 872 21.9
1:1 1489 23.3
1:2 2035 23.4
1:0 1:1
.17 .33
104 82
17.9 13.7
.1 .9
4.0 5.1
18.3 25.3
3.7 1.8
1.5 1.5
45.5 48.3
54.5 51.7
TABLE 6
ASH MASS BALANCE
Bottom Ash
Ash in Kg/hr.
Fuel Carbon With
Kg/hr. Free Carbon
191 82 89
347 232 238
476 324 341
1:2
.30
99
17.8
1.2
5.4
16.6
1.8
1.5
44.1
55.9
Fly Ash
Kg/hr.
Carbon With
Free Carbon
5 7.7
5 6.8
7 10.2
0:1
.19
113
19.4
4.0
8.1
3.0
3.7
1.5
39.7
60.3
Collector*'
Kg/hr.
Carbon Wi th
Free Carbon
104 219
110 369
145 300
*Note: 1) The collector weight was determined by difference.
290
-------
TABLE 7
EMISSIONS TESTS PERFORMED ON d-RDF/COAL
BLENDS COMBUSTED IN HAGERSTOWN TESTS
Test Method
EPA Method 5
Cascade Impactors
EPA Method 7
Orsat
AID Gas Chromato-
graph
Wahlco Probe
Tedlar Bag
Theta Senors, Inc.
Tri-gas Meter
Leads Northrup
Transmissometer
Draeger Tubes
Measurement
Participate mass flux,
C1, F, S02, S03, and
trace organic and in-
organics.
Particulate size dis-
tribution
Nitrous oxides
C02, 02, CO
Total hydrocarbons
Resistivity
Record sample
C02, S02, N02 (Continu-
ous)
Opacity (Continuous)
CO, C02, SOX
duced unusual results, the Tedlar bag sample
was used to further clarify the data through
gas chromatograph analysis.
Particle Mass Flux—
The particulate mass flux in the flue
gas was reduced with increasing blend ra-
tios, i.e., 1:1 and 1:2 blend firing until
the 0:1 or 100 percent d-RDF firing was
reached. The reductions, however, were not
significant at the 90 percent confidence
level. The data show that when 100 percent
d-RDF is fired, the fuel ash increases more
than offsets the fly ash carbon reductions.
Size Distribution—
The size distribution was shown to
shift toward fines as d-RDF was substituted
for coal. Two reasons for this are the im-
proved burnout of the aerosol (producing
smaller particulates) and the large number
of fine paper platelets in the d-RDF caus-
ing the number of particles formed to in-
crease.
Fly Ash Resistivity-
Table 8 summarizes the effects of sub-
stituting d-RDF for coal on aerosol resis-
tivity. The 1:1 blend fly ash has resis-
tivities around 1010 ohm-cm, within the
range of 108 to 1010 ohm-cm required for
efficient precipitator performance.
TABLE 8
EFFECT OF SUBSTITUTING d-RDF
FOR COAL ON AEROSOL RESISTIVITY
Blend
1:0 <107
1:1 8xl09
1:2
0:1
Gaseous Emissions
Resistivity (n-cm)
March Tests May Tests
6xl07
2xl010
IxlO12
IxlO12
During the Hagerstown tests, data were
collected on emissions of S02/S03, total
oxides of nitrogen, halogens, hydrocarbons,
trace organic, and heavy metals. Results
are summarized below:
SOX-
For 1:2 and 0:1 blend firing, there was
a small, but significant reduction in over-
all S02 emissions with the replacement of
higher sulfur coal by 0.6 percent sulfur d-
RDF. For 1:1 blends, the sulfur reduction
was too little to be distinguished, but as
one would expect, reductions in sulfur e-
missions appear to be porportional to the
reduction of sulfur in the fuel burned.
NOX~
An evaluation of NOX emissions data
revealed that NOX emissions did not change
significantly when d-RDF was substituted
for coal.
Halogens--
Analyses of data on chlorine and fluo-
rine emissions indicate that a significant
291
-------
increase (factor of 16) occurred when d-RDF
is substituted for coal.
Although the chlorine levels in d-RDF
are about the same as some coals burned in
the United States and Europe, the coal:d-
RDF blends emit more chlorine per unit of
energy than most coals because of their
lower heat content.
Hydrocarbons--
The hydrocarbon emissions from the d-
RDF-coal blends did not differ significant-
ly from those for coal only.
Trace Compound Emissions
The flue gases were evaluated for
polycyclic organic compounds and heavy met-
al emissions. Based upon the test results,
the overall emissions of polycyclic organic
compounds for blend firing and coal only
were below the threshold limits being pro-
posed by the National Academy of Science.
The actual amounts of various trace
inorganic elements found for all blends
were generally considered not abnormal.
Some trends were observed. There as a tend-
ency for Br, Mn, Pb, and Sb to increase in
concentration with increasing coal:d-RDF
ratios indicating that the refuse is a con-
tributor of these elements. As, Ni, and Vn
decreased with increasing RDF.
SUMMARY
Tests conducted thus far have been en-
couraging, especially from the standpoint
of operations in the boiler facilities.
Boiler performance has been generally ac-
ceptable while burning blends of d-RDF and
coal; environmental effects appear accept-
able, and handling of d-RDF can be accom-
plished satisfactorily.
Perhaps most discouraging at this time
has been the problem of achieving accept-
able production rates for d-RDF using the
CPM. Since throughput capacities of the
densifier will greatly effect economics,
this is an important consideration. We are
hopeful that the production experience gain-
ed at NCRR and those to be gained at Tele-
dyne will help to ease this situation. Work
using different densification equipment is
also required.
Also important to the economics will
be the final characteristics required for
an acceptable d-RDF. As more processing is
required, costs will be increased. How-
ever, perhaps the d-RDF does not have to be
a well formed pellet (or other shape) and
can contain relatively large amounts of
fines. This would help to increase den-
sifier production. The additional test-
ing planned will help to determine if this
is the case.
Storage of the d-RDF during this proj-
ect, although unusual, has indicated the d-
RDF can be stored for at least a one month
period. Storage recommendations were pro-
vided. Handling was also routinely done.
Additional work would be useful, however,
to demonstrate that blends of d-RDF and
coal can be routinely handled over extended
periods in existing boiler equipment.
Currently the d-RDF program is imple-
menting a second phase of combustion tests
at the General Electric Plant in Erie,
Pennsylvania. We plan to burn approximately
2,800 tons of d-RDF produced by NCRR and
Teledyne. A full battery of tests will be
done to further characterize the effects of
using d-RDF in existing stoker boilers.
Hopefully, results of these tests will de-
monstrate that d-RDF can be successfully used
in existing smaller institutional, indus-
trial, and utility power generating facili-
ties.
ACKNOWLEDGEMENT
Work reported here is primarily a sum-
mary of activities conducted by NCRR under
Research Grant R804150 "Preparation, Use
and Cost of d-RDF as a Supplementary Fuel
in Stoker Boilers," and by Systems Technol-
ogy Corporation under Contract 68-03-2426
"Effects of Burning Densified Forms of Mu-
nicipal Solid Waste Derived Fuels in Indus-
trial, Utility, and Institutional Stoker
Boilers." Dr. Harvey Alter and Jay Campbell
of NCRR and Jerry Degler of Systems Technol-
ogy provided major input to this paper.
Work has also been done in cooperation with
the Industrial Environmental Research Labo-
ratory, Cincinnati. All wish to thank the
State of Maryland personnel for their in-
terest and cooperation.
292
-------
STATUS, TRENDS, AND IMPEDIMENTS TO DISCARDED TIRE COLLECTION AND RESOURCE RECOVERY
James F. Hudson
Patricia L. Deese
Douglas Funkhouser
Urban Systems Research & Engineering, Inc.
36 Boylston Street
Cambridge, Massachusetts 02138
ABSTRACT
As part of its legislative mandate under the Resource Conservation and Recovery Act,
EPA is required to study the problems associated with scrap tire reuse and disposal. USRfcE
is under contract to perform a technical, environmental, and economic assessment of the
various reuse and disposal options, and the use of economic and regulatory approaches to
assure the most efficient use of limited resources.
While the project is in its early stages, it is possible in this paper to describe
most of the available options, and the market structure factors which make them more or
less attractive. The technical options include retreading, increases in tire lifetime,
baling or shredding for improved landfill disposal, material for artificial reefs, use of
ground tires as an additive to coal for fossil fuel power plants, rubber reclaiming, use
of ground tires in pavement and road construction, and a host of minor uses, plus processes
in the development stages. Key impact points for regulations or cricinp mechanisms are
also Identified.
INTRODUCTION
Section 8002h of the Resource Conser-
vation and Recovery Act calls for EPA to
prepare a report to Congress on "discarded
motor vehicle tires which shall include an
analysis of the problems involved in the
collection, recovery of resources including
energy, and use of such tires.'1! As the
basis of that report, the Office of Re-
search and Development has contracted for a
study of the waste tire problem, and this
paper presents some preliminary information
from that study which began in September
1978 and is scheduled to be completed in
September 1979. The paper is divided into
two major sections: the first describes
the flow of passenger car tires, based on
1976 data, and identifies the potential
impact points for economic and regulatory
actions; the second describes the technol-
ogies for reuse and disposal, based on data
from the literature. As the project pro-
gresses, the information contained through-
out will be updated, and it will become
possible to make recommendations for EPA
and Congressional action.
ENVIRONMENTAL IMPACTS, AND MATERIALS FLOWS
AND POINTS FOR POLICY IMPACT
Each year over 200 million tires are
produced to be used as original equipment
on new vehicles, or to replace worn out
tires. Eventually, the tires can no longer
be used for transportation purposes and
must be disposed of. The most common
method of disposal provides landfilling
without size reduction. As much as 90% of
all tires are currently landfilled, once
their useful life is over. This creates an
environmental problem, and represents a
fairly large loss of resources. Although
scrap tires account for no more than 1.5%
of the total solid waste stream, they can
cause significant environmental damage.
Tires landfilled in large numbers tend
to rise to the surface, creating aesthetic
insult and reducing land reuse; they also
provide a breeding space for rats and dis-
ease vectors, causing a public health
293
-------
FIGURE 1
APPROXIMATE 1976 PASSENGER TIRE BALANCE
New Sales
191 million
TIRE
STOCK
net increase in
stock—26 millior
Discards
200 million
Yes, 70%
Retreads
35 million
140 million
Yes, 25%
35 million
Disoosal
No, 30%
60 million
145 million
@
No, 75%
105 million
Resource Recovery
20 million
¥ V T f t
Tire Energy
Landfill Reclaiming Splitting Reefs Recovery
145 million 10 million 3 million 1 million 5 million
T
Other
Productive
Disposal
1 million
o
Impact points for policy analysis
294
-------
hazard. On the other hand, if they are
shredded or sliced, they compact better and
prevent aesthetic and public health damage.
The reason tires are not shredded prior to
landfill may be explained in terms of ex-
ternal costs, since the environmental con-
sequences of dumping are borne by the gen-
eral public, and the individuals who dispose
tires have no incentive to pay the addition-
al costs of shredding and slicing. These
costs have been estimated at $.50 to $2.00
per tire.2
The energy requirements for the average
passenger car tire have been estimated at
seven gallons of oil or its equivalent'—in-
cluding both material and process energy.3
While not all of this can be recovered, the
use of waste tires for energy production is
becoming increasingly common.
Because of the market's failure to
realize the socially optimal tire management
and disposal practices, government has a le-
gitimate role in attempting to reduce the
number of tires discarded, and to promote
environmentally^sound disposal techniques.
This paper is concerned with the role of the
government in this area, in particular with
potential government action. The existine
tire management and disposal systems in the
private sector and the federal government
must be analyzed to accomplish this.
Figure 1 shows the system components,
along with USRSE's estimates of passenger
tire flows for 1976, The impact points
where Federal policies could affect the
system are shown in the figure. Point 1 is
the tire stock and total sales, where in-
creases in the lifetimes of either new tires
or retreads could, potentially, reduce the
resources used in tire production. This
could be accomplished by improved mainten-
ance practices, or improved tire design.
Impact point 2 is inspection for re-
treading after discard. Since retreading
results in the most benefit of any approach
available at present, increased inspection
would probably lead to net benefits which
would be desirable. Tire deposits refunded
to those who turned in old tires would be
one mechanism that might be useful.
Point 3 is related to the results of
the inspection. Standardization of tires,
or improvements in quality control
potentially lead to an advanced retread
technology and market, which could be
induced by federal incentives. Increased
carcass quality or economic benefits for
retreading are examples of federal
incentives.
At Point 4, the disposal decision,
the costs and availability of various dis-
posal options will be crucial. Increased
landfilling costs, and increased availabil-
ity of other disposal methods, will result
in increased resource recovery- Federal
actions are possible on all sides of this
system, ranging from economic approaches
such as product charges
of technologies or enforcement of landfill
standards.
Before recommending specific options
for each of these impact points, it will be
necessary to improve the data on current
materials flows, and analyze the costs,
technical feasibility, and environmental
impacts of the various disposal options.
This work is currently underway.
METHODS FOR RESOURCE RECOVERY, AND FOR RE-
DUCING ENVIRONMENTAL IMPACTS
Extending Tire Lives
Shorter tire lives increase the vol-
ume of tire production, and result in
increasing tire disposal volume.
Therefore, extending tire life may be viewed
as the first step in tire management. As-
suming that roughly 10% of the scrap tires
find "productive" (non-landfill) final uses,
approximately 2.5 million tons of scrap
tires would be landfilled each year. This
corresponds to roughly 1.5% of the total
solid waste stream. However, as discussed
above, scrap tires present a particularly
difficult disposal problem, and also repre-
sent a loss of resources.
Tire consumption—or rather tire mile
consumption—may remain constant, or even
increase with no corresponding increase in
tire production, if the useful life of
tires is extended. Reducing the need for
tire production also reduces resource con-
sumption, as well as the volume of tires -_
entering the waste stream.
Procurement of longer life tires, prop-
er tire maintenance and retreading are the
three major means for extending tire lives.
Each of these will be discussed in turn.
295
-------
Procurement
The purchase of longer life tires re-
duces the number of tires consumed, produced
and eventually disposed of. Currently radi-
al tires achieve the highest mileages, and
have the longest useful lives. Most sources
estimate 40,000 miles for radials, and 20-
25,000 miles for bias and bias belted tires.
However, radials in GSA motor pools in 1975
obtained 24-26,000 miles, while bias and
bias belted tires averaged 15-25,000 miles.
Actual mileages obtained depend on a large
variety of factors—including driving habits,
road composition, and tire maintenance.
According to some sources, the manu-
facture of radials has higher resource and
energy requirements than bias and bias
belted tires, but no actual figures have
been obtained for comparative purposes. If
the mileage gain outstrips the increases in
resource requirements, radials may be expec-
ted to reduce the size of the scrap tire
problem. In addition, radials improve gas
mileage by 5-10%, further reducing energy
requirements and air pollution from mobile
sources.
Radials also represent some marginal
economic benefits for the consumer--slightly
lower tire costs per mile travelled, im-
proved gasoline mileage, elimination of the
time costs of purchasing replacement tires,
as well as the out-of-pocket-costs of re-
placing the tires. The market share of
radial tires has been increasing in replace-
ment use from 23% in 1974 to 36% in 1976.
According to another estimate, radials are
now 67% of original equipment tires and 35%
of replacement tires—accounting for an over-
all market share of 44%. They are particu-
larly important for passenger vehicles,
while trucks and farm equipment still util-
ize bias and bias belted tires for the most
part. The soft sidewalls and radial belts
of radial tires are not suitable for farm
requirements, but are desirable for highway
trucking, largely because of lower heat
gains in travel.
The manufacture and procurement of the
100,000-mile tire would extend tire lives
beyond those of the radial tires. Techno-
logically it is now feasible to manufacture
100,000-mile tires for passenger vehicles,
but product quality and economics make its
use doubtful, despite favorable studies of
the tire by some analysts.
Tire Maintenance
The second major method of extending
tire life involves proper tire care and
maintenance. Correct inflation, wheel
alignment, prompt repair to small damages,
and frequent inspections are some of the
components of tire care. Tire care and
maintenance extends the original life of the
tire, and enhances its retreadability by
maintaining the casing in good condition.
Properly maintained retreads also obtain
higher mileage than those not cared for.
Retreading
Retreading is the third major method
for extending the useful lives of tires.
According to an article in the Retreader's
Journal,^ the number of passenger tires re-
treaded was 30.5 million in 1972, 35.3
million in 1975, and was expected to be
35.5 million in 1976.
Passenger tires are normally retreaded
only once, while truck tires average 3 to 5
retreads. Airplane tires, on the other
hand, may sometimes be retreaded up to 22
to 23 times,6 and routinely average 7 or 8.
The reason for the variation is the relative
loads and levels of care in the three tire
types.
The market share of the retreads has
been decreasing, primarily because of supply
constraints. Retreaders could sell all the
retreads they wanted, but do not apparently
have a sufficient supply of suitable cas-
ings. The proportion of inspected tires
actually accepted for retreading varies a
great deal depending on the quality of the
carcasses inspected, as well as on the
standards of the retreading operation.8 In-
creasing tire collection and inspection
rate would augment the supply of casings,
but improved maintenance is also required
before significant gains can be made. Ra-
dials have been prone to belt-edge separa-
tion, reducing their suitability for re-
treading. The supply constraint in this
area is particularly significant, but im-
proved radial designs and retread tech-
niques should reduce the number of discards
Retreading decreases tire consumption
without affecting mobility. It represents
296
-------
environmental benefits through reducing the
volume of the solid waste stream, resource
savings through lowering the energy and ma-
terials requirements per tire mile, and
substantial savings to the consumer in terms
of expenditures for new and replacement
tires. The average passenger tire retread
requires 8.7 pounds of (retread) rubber,
while a new tire requires about 22 pounds of
rubber (not including process waste).^ If
mileage is doubled through retreading, solid
waste generation is reduced by 13 pounds.
Retread quality control may be a key issue.
Even with the procurement of longer
life tires, proper tire maintenance and
retreading, tires eventually wear out, and
can no longer be used for transportation.
Proper resource recovery and disposal tech-
niques must then be selected to reduce
resource loss and environmental damage.
Resource Recovery - Whole Tire Methods
Breakwaters: Scrap tire breakwaters
represent an increasingly attractive use of
.waste tires." Mats of tires, one-half
wavelength wide, can reduce wave heights
from 4' to 1', and are useful in waves
up to 5'. In one case, a Rhode Island
yacht club built a breakwater 500' x
21' for $2,500 plus donated labor. These
costs appear typical. Because of the energy-
absorbing ability of tires, the breakwaters
require little maintenance and will continue
to be effective for long periods. The po-
tential for this, use is unknown; however, it
should probably remain in the private sector
with the users paying the costs.
Artificial Reefs; Certain species of
fish prefer rocky, partially enclosed hab<-
itats for protection and shelter. Since
many of these fish are sport or commercial
species, the development of artificial reefs
can be of benefit to anglers and the public.
Tires have been used for artificial reefs
because they are inexpensive, inert, and
seem to offer desirable habitat.
Several configurations of scrap tires
have been used for artificial reefs, with
various costs depending on the labor and
materials required to construct and move the
modules.11 Installation costs 'for the
simplest methods have been estimated at ..
30c-40c/tire, plus tire acquisition costs.
Goodyear estimates present use at 500,000
tire per year, but the potential for use of
up to one billion tires has been claimed.
Since one reef in Florida is expecting to
use 500,000 per year, much higher levels can
be expected.13 The cost can often be re-
covered from private angling groups or sport
fishing associations, as contributions of
money or labor.
Artificial reefs are less prevalent on
the West Coast, partially because of more
difficult conditions; fresh water uses
(particularly in artificial reservoirs) are
just beginning.
Crash Barriers and Abutments: Rubber
tires in carefully designed configurations
can be used to considerably reduce crash
impact damage. At present, Goodyear estim-
ates that about 200 million tires (or one'
year's disposal) could be used for such
abutments in well^identified points. This
is an area which EPA could coordinate with
NHTSA and FHWA, since it would involve
highway applications for safety improve-
ments. Cost estimates are necessary.
Heat Recovery; One energy recovery
experiment incinerates tires
heat recovery. This system uses a cyclone
furnace and accepts whole tires with a
capacity of about 36 tons per day. This
would lead to an estimate of about 3 mil-
lion tires/year (.another estimate in a
Goodyear press release is 1 million tires/
year) • This would generate 25,000
pounds of steam per hour at 250 psi.1^ The
economics analysis of this plant is still
underway, but appears unfavorable at pre-
sent.
Resource Recovery - Shredded or Split
Tire Methods
Use of Tires as Road Materials:
Tires appear to provide an attractive
additive for road-building, when properly
prepared. Several options have been
suggested, all using tires which have been
shredded or ground to the correct size,
after metals and textiles have been re-
moved. Devulcanization may also be an aid
in some instances, indicating a potential
market for the reclaimation industry. At
present, reclaim and retread buffings are
products which have consistent particle
sizes and purity.
297
-------
Joint sealing, another use for waste
rubber, is performed on the New York Thru-
way.15 The flexibility of the rubber adds
life to the joints, reducing necessary
maintenance and ultimately saving on mater-
ials. Thus, the process has net benefits
for New York even though initial materials
costs rise. Goddard estimates that the use
of scrap rubber would be advantageous at
any rubber price up to 29C/pound, lf> which
is at current prices.
The use of rubber in seal coats is
another option. These are placed over
existing streets to cover cracking. The
rubber (mixed in at up to 35%) extends the
life of the seal coat considerably by pro-
viding elasticity to reduce cracking from
temperature and load changes, and also
reduces the speed with which cracks in the
underlayers propogate to the surface. This
has been demonstrated in Phoenix and other
areas by McDonald. Goddard estimates
its value at 60-70<:/pound of rubber used,
though this may be an overestimate.
Rubber has also been suggested as a
constituent in bituminous concrete 6%
bituminous concrete could potentially
contain 1% to 2% waste rubber.18 If this
proved advantageous, most of the presently
discarded scrap tires could be used in this
manner, and the straight asphalt mix could
use the remainder. However, few engineer-
ing benefits have been shown for this
product.
Other options including using rubber
and asphalt or coal tar in an emulsion as a
road dressing for driveways, parking lots,
or streets. Skid resistance is improved,
weathering may also be aided, but no cost
data are available. Goodyear claims to
have
used ground-up tire mixed with an
adhesive to produce a resilient
artificial turf. This springy
surfacing material may be tinted
a variety of colors for use on
playgrounds,..the economics of
production are not favorable, but
the process is under further
consideration.19
One of the most attractive uses is for a
"strain relieving interlayer" between a-new
road surface and the underlying layers.
The flexibility and elasticity of the rubber
reduces cracking and stresses on the upper
layer. Experiments are underway nationally
which employ this use.
Two independent federal studies have
suggested these various asphalt uses as
suitable methods for federal tire disposal,
and deserving of federal promotion. Of the
alternatives studied, Goddard found the
asphalt uses the mosteconomically attrac-
tive, even at present prices of 10$-12<:/
pound. Hedley, et al suggested cryogenic
grinding (see below) which shows recovery as
an asphalt additive as one of the most at-
tractive uses for military waste tires
Ort
which can no longer be retreaded. One
estimate suggests material cost increases
for rubber additives as
• 4.5% for using powdered rubber
additive in resurfacing;
• 5.5% for a rubberized binder in
resurfacing;
• 28.5% for using a rubberized
binder in chip sealing; and
• 14.8% for adding a strain-relieving
interlayer.21
The asphalt industry produces about
5.4 million tons yearly, which has a poten-
tial recovery of over 100 million waste
tires at 25% use with the asphalt.
Embankments: Tire uses in embankments
and retaining walls were mentioned bv one
source, and appear attractive. At this
point, we have only limited informa-
tion on them, and have no data on their
potential use of tires or economic
analysis. It appears that halved tires
can provide sufficient structural support
for embankments which are almost vertical,
when bolted together correctly. Embankments
are constructed of alternating layers of
tires and earth.
Oil Spills; The use of shredded waste
tires, along with polystyrene scrap, for
absorbing spilled oil was described in one
source. The materials have good capacity
for absorbing the oil, can be collected
easily, and can supposedly be converted
through heating into a form similar to
asphalt for reuse.22
298
-------
Tire Splitting: This industry is not
expected to be a major growth sector. The
plants are relatively low-polluting, selling
approximately 35% of their input as new
products, and selling pieces to reclaimers
(as of 1968) . Their input in 1968 consisted
of about 60 million pounds, or about 2.4
million passenger tire equivalents. Thus,
they are likely to continue to be tire
users, but their limited market will keep
them fairly small, accounting for under 2%
of the waste tires for the near future.
Note that these data are several years old,
and deserve updating if EPA plans any major
effort in the area.
Tire Derived Fuel: General Motors has
experimented with the use of shredded tires
as supplementary fuel in a power plant burn-
er. 23 The tire chips were purchased from a
tire outlet company, stored in an old farm
silo, and added to the coal on a conveyor
belt in the ratio of 1 to 9. During the
experiment, shredded tires were obtained at
a cost of $24/ton, which may be compared
with the price of coal — at that time— of
$40 per ton. For a continuing supply of
shredded tires, however, the quotes averaged
$40/ton, while coal prices sagged to around
$30/ton. The use of tire chips required
additional handling and environmental
controls, both increasing the costs of tire
chips relative to coal. One of the diffi-
culties in the GM experiment was that the
furnace was not designed for handling large
chunks of materials — the requirements for
fine particle size implied high shredding
costs. Furnaces designed to handle larger
pieces may be able to obtain shredded tires
at substantially lower costs—at approxi-
mately $20 per ton, and with high BTU value.
Reclaimed Rubber: Probably the best
description of the reclaimed rubber indus-
try's present performance is that in the
U.S. Industrial Outlook24 which states:
An unfavorable cost/price struc-
ture, some old inefficient plants,
and significant environmental prob-
lems have continued to plague the
reclaimed rubber industry ... industry
shipments have declined 50% since
the late 1960 's. The use of re-
claim in radial tires has been
less than in bias-ply tires.
The reclamation industry received about 576
million pounds of rubber in 1968, of which
something like 85% was post-consumer waste,
largely from tires.25 The Commerce estimate
was extrapolated to a present use of
about 300 million pounds, which would
mean disposal of about 10 million
passenger-tire equivalents. Of this, about
2 million came in as waste from the tire-
splitting industry. In other words, recla-
mation is important, but its impact is not
particularly large at present.
One of the major potential uses for
reclamation products is in asphalt additive.
The cleaned, ground rubber is the major
source for small high-quality additive, with
tire buffings from retreaders representing
another major source.26 However, cryogenic
grinding may tend to undercut this market so
that the reclamation industry will continue
to have difficulties selling its product
profitably.
Advanced Chemical Methods; There are
a number of advanced systems for tire recov-
ery which generally involve breaking down
the tire into its components. These, in
turn, can be used as fuels and process
feedstocks. The most advanced of these is
probably the Goodyear/TOSCO pilot plant in
Rocky Flats, Colorado, which uses ''oil
shale recovery technology" to convert the
tires into oil, carbon black, and steel.
They claim that a full-scale plant would
process 8 million tires/year, providing 15
million gallons of oil, 73 million pounds
of carbon black, and 2 million pounds of
scrap steel (for steel-belted tires). This
would be sufficient for the production of
2 million new tires, if all worked well.
Other processes under a number of names
have been developed, each with unique char-
acteristics.28 For example, the destructive
distillation process developed by Firestone
and the Bureau of Mines would convert tires
into char, oil, and gas for use as fuels
or feedstock. Hydrogenation has also been
suggested, as have depolymerization, pyroly-
sis in salts, etc. The similarities and
differences of these processes are not im-
portant to EPA at this time, unless demon-
stration grants are contemplated. They are,
at beet, several years away from full-scale
implementation, and few cost estimates of
any kind are available. The technologies
are available in R & D, but there is no
strong evidence concerning their economics
or practical implementation.
299
-------
Sanitary Landfill Disposal
The sanitary landfill is the last re-
sort for waste tires, as for most other
solid wastes. In most cases, these will be
commercial landfills receiving wastes from
industries—particularly tire dealers, re-
treaders, and auto scrappers—although some
tires do enter the municipal waste stream.
Landfilling tires is a difficult
process, since they do not compact well,
tend to migrate to the surface, and form an
attractive breeding ground for disease
vectors. Thus, size reduction or baling
of tires would be desirable as landfill
practice. Landfill receives the bulk of
waste tires at present.
Size Reduction
Size reduction is required for most
high-value applications of the waste rubber,
and is the first step in the rubber reclaim-
ing industry. Some size reduction is re-
quired for landfill, to allow easy disposal
and reduce the potential for environmental
damage. Recent experiments have suggested
cryogenic grinding for both inplant rubber
waste and for scrap tires, with evidence
that the sizes can be reduced to useful
levels at low costs.^° Though some conven-
tional shredders also appear to work well,
tire shredders and shears range in cost
from a few thousand dollars for some port-
able units up to about $100,000.
300
-------
FOOTNOTES
Resource Conservation and Recovery Act, P.L. 94-580.
2
J.F. Hudson and E.E. Lake. A Decision Paper of EPA Guidelines for Tire Reuse and
Disposal. Cambridge: Urban Systems Research & Engineering, for the EPA's Office of Solid
Waste Management Programs, December 1976.
J.A. Beckman, et al, "Scrap Tire Disposal Procedures". Rubber Chemistry and Tech-
nology. 47 (3);597^624; and G. Crane et al. "Scrap Tire Disposal Procedures". Rubber
Chemistry and Technology. 51:577-599, 1978. ~"
4
R.R. Westerman. Tires: Decreasing Solid Wastes and Manufacturing Throughput.
EPA-600/5-78-009, July 1978.
E.J. Wagner. "Retreadonomics'1. Retreader's Journal, 20(5): 3-9, May 1976.
Smithers Scientific Services, Inc. A Study of the Feasibility of Requiring the
Federal Government to Use Retread Tires. Prepared for EPA-OSWMP, 1974.
Wagner, op.cit.
g
Tire Retreading Institute. Retreading Radial Passenger Tires. Washington, D.C.:TRI
April 1976. (1343 L Street, Washington, D.C. 20005—$2.50); C.C. Humpstone, et al. Tire
Recycling and Reuse Alternatives. IRT for EPA, 1974
9
Wagner, op.cit.
N.W. Ross. Floating Breakwaters. University of Rhode Island, Narragansett, Rhode
Island, 1977.
11R.B. Stone, et al. Scrap Tires as Artificial Reefs. Report SW-119, USEPA, 1974.
L. Colugna and R. Stone (.eds.). Proceedings of an International Conference on
Artificial Reefs. NOAA-7502009. March 1974; and Stone et al, op^cit.
Colugna and Stone, op.cit,
14E.R. Moats. "Goodyear Tire-Fired Boiler", Resource Recovery and Conservation
1(1976) 315; and J.B. Stirbling. "Cyclone Furnaces", Engineering. October 1972.
15T.M. Cleary and W.H. Clark, III. "Old Tires Contribute to Thruway Maintenance",
gublic Works Magazine. July 1973.
301
-------
H.C. Goddard. An Economic Evaluation of Technical Systems for Scrap Tire Recycling.
EPA 600/5-75-019, December 1975.
C.H. McDonald. "An Elastomer Solution for 'Alligator' Pattern, or Fatigue,
Cracking in Asphalt Pavements" presented to the International Symposium on the Use of
Rubber in Asphalt Pavements. May 1971.
18
J.E. Stephens and S.A. Mokrzewski. "The Effect of Reclaimed Rubber on Bituminous
Paving Mixture." Civil Engineering Department, University of Connecticut. Report CE74-75.
March 1974.
19
Rubber Manufacturers Association CRMA). Packet of press releases and articles,
December 3, 1976.
W.H. Hedley, et al. Disposal Study: Tires and Other Polymeric Materials. Monsanto
Research Corporation for Army Natick Development Center, April, 1975. Contract DAAK03-74-
C-0136.
B.D. LaGrone and B.J. Huff. "Use of Waste Rubber in Asphalt Paving." Presented at
the Colorado State University Asphalt Paving Seminar, December 1973. (authors from U.S.
Rubber Reclaiming Company).
22
Beckman, et al. op,cit.
23
R.H. Taggart, Jr. "Shredded Tires as an Auxiliary Fuel." Environmental Activities
Staff, General Motors Technical Center, 1975.
24
U.S. Department of Commerce. 1976 U.S. Industrial Outlook. U.S. GPO, Washington,
D.C., 1976.
R.J. Pettigrew, et al. Rubber Reuse and Solid Waste Management, EPA, 1971 (SW-22c).
Lagrone and Huff, op. cit,
RMA, op.cit.; "GoodyearT-TOSCO Rubber Recycling Plant." Journal of Commerce,
p, 26, 12/24/75, '
28
Beckman, op.cit.
29
Office of Solid Waste Management Programs. Decision-Makers Guide in Solid Waste
Management. EPA, 1976, SW<-500,
R.R. Broton. "Cryogenic Recycling of Solid Waste." Public Works 107(9): 84-85.
September,- 1976; N.B. Frable. "Cryogenic Grinding Saves $$, Recycled Scrap Becomes
Valuable Resource," Rubber World. October, 1976. pp. 66, 68, and 87; "Cryogenic Process
for Recycled Tires." Journal of Commerce, p. 5, 9/7/77; and M.D. Kazarnowicz, et al.
Cryogenic Scrap Tire Processing. Allentown, Pennsylvania: Air Products and Chemicals,
Inc. (paper presented at 112th meeting of the American Chemical Society, October 4-7,1977).
302
-------
ADVANCED THERMAL & CHEMICAL CONCEPTS FOR
IMPROVED MSW DERIVED PRODUCTS
N. L. Hecht & D. S. Duvall
University of Dayton Research Institute
300 College Park
Dayton, Ohio 45469
ABSTRACT
A number of resource recovery projects have been instituted to recover fuel, energy,
and mineral components from refuse. Although a number of these programs have been effec-
tive, the quality of the products recovered could be enhanced to improve their market-
ability. The purpose of this study was to investigate the potential of known processes
that could improve the quality of the fuels or other products derived from the organic
fraction in refuse.
This study concentrated on those processes designed to produce a carbon char and a
powdered fuel from the municipal solid waste. Of particular interest in the production of
carbon char, were those chemical treatments that promote char formation at lower pyrolysis
temperatures. For the production of powdered fuels, the chemical and thermal treatments
which cause cellulose embrittlement were of most interest.
A major accomplishment of this project was the identification and laboratory verifi-
cation of chemical treatments for cellulose embrittlement. As a result of these studies,
the basic requirements were defined for producing a fine powdered fuel from the organic
fraction of MSW.
Information obtained from these studies provided the basis for a preliminary engineer-
ing and economic analysis of a full scale facility to produce a fine powdered fuel.
INTRODUCTION
Due to the many problems associated
with the conventional practices of land-
filling and incineration, a number of new
techniques are being developed to improve
the management of municipal refuse. The
majority of these new techniques are con-
cerned with recovering and utilizing the
valuable materials in solid waste. Resource
recovery is the term applied to the numer-
ous processes and systems being designed to
recover components of waste and convert
them to useful products. Systems have been
developed for energy recovery; compost pro-
duction; fiber, glass and metal recovery;
and alcohol and protein production. The
Major emphasis has been on those processes
designed to recover the thermal energy in
waste. Although many of the systems devel-
oped have been relatively successful, the
quality and consistency of the fuel produced
or the energy generated has not been com-
pletely satisfactory. Higher quality and
greater consistency in the waste-derived
fuel or energy is necessary if it is to be
marketable for large-scale usage.
The purpose of the studies conducted at
the University of Dayton Research Institute
(UDRI) was to investigate the potential of
known processes that could improve the
quality of the fuels and other products de-
rived from the organic fraction in refuse.
Since cellulose products are the major con-
stituents of the organic fraction in refuse,
it seemed likely that a number of the pro-
cesses employed in the pulp, paper, and
textile industries would have considerable
potential for refuse processing.
This study was initiated with a compre-
hensive review of processes for making ref-
use a better fuel. Information was obtained
303
-------
from the open literature, and through per-
sonal contacts. Possible processes for im-
proving the quality of products from the
organic fraction derived from municipal
solid waste (MSW) were determined, and des-
criptions were developed for each process.
To better evaluate the different processes,
an analytical framework for technical and
economic assessment was developed to serve
as a guide for analysis, as well as for the
information acquisition phase of the study.
The literature search identified major ther-
mal and chemical processes used in the pulp
and paper, wood, textile, and resource
recovery industries.
The study, which followed,concentrated
on those processes designed to produce a
carbon char, and a powdered fuel from the
municipal solid waste. Of particular inter-
est for the production of carbon char were
those chemical treatments that promote char
formation at lower pyrolysis temperatures.
For the production of powdered fuels, the
chemical and thermal treatments which cause
cellulose embrittlement were of most
interest.
CARBONIZATION OF CELLULOSE
Thermal decomposition of refuse in the
absence or partial absence of oxygen (pyrol-
ysis) has been used for the production of
useful fuels. Solid, liquid, or gaseous
fuels can be obtained by a variety of pyrol-
ysis processes. The quantity of char,
bitumen-like liquid, and gas produced varies
and is a function of the time-temperature
sequence for each particular process. Com-
mercial pilot and laboratory processes that
utilize a variety of furnace designs have
been developed to produce a wide range of
fuel products. Rotary kilns, vertical and
horizontal shaft furnaces, fluidized bed
furnaces, and a variety of batch type reac-
tors have been employed for fuel production.
A number of these systems may reach commer-
cial status within the next several years.
As indicated, organic liquids, char,
water, and a gas are produced from the
pyrolysis of cellulose. The quantities gen-
erated are controlled by the heating rate,
final temperature and length of exposure to
final temperature. In general, the char
will constitute between 20-40% of the final
product mix, the organic liquids and gas
phase can vary between 10-40%, and water
constitutes the remaining fraction. The
thermal process employed can be designed to
maximize the end products desired from the
cellulose wastes. Higher heating rates and
higher temperature produce larger quanti-
ties of gas and less char. Conversely,
lower heating rates and lower temperature
processes result in increased char
production.
The quantity of char, the composition
of the gases and liquids evolved, and the
necessary reaction temperatures can be sig-
nificantly affected by the presence of
chemical agents in the cellulose materials.
A number of chemical compositions have been
identified which increase the quantity of
, char produced and decrease the amount of
combustible gases and tars formed. Many of
these compounds were developed for flame-
proofing cellulosic materials. The flame-
proofing agents direct the pyrolysis process
toward maximum production of char, minimum
production of tars, and the highest possi-
ble proportions of water and carbon dioxide.
The highly effective compositions are all
soluble in water, and most are salts of
either strong acids or bases. In addition,
many oxidizing agents attack cellulose and
have been found to be effective fire retar-
dants. Oxidized cellulose pyrolyzes
rapidly at lower temperature giving high
charcoal residues.
Although a number of pyrolysis pro-
cesses have been developed to produce
carbon char, the literature search found
only one process using chemical treatment
to promote char formation. In the patented
process developed by Dr. A. Harendza-
Harinxma, sodium aluminate (Na2Al20^) is
used to promote carbon formation at tem-
peratures below 660°F, whereas in most of
the conventional pyrolysis processes
reported in the literature, pyrolysis is
achieved with thermal treatments from 930°F
to 1650°F. In addition, the char yield is
reportedly increased for the sodium
aluminate process
(1)
In Dr. Harendza-Harinxma's process,
municipal refuse and sewage sludge are py-
rolyzed to a char and fuel gas. In this
process, sewage sludge is used as the sol-
vent for a 2-4% solution of sodium alumin-
ate. This solution is mixed with municipal
refuse (two parts sludge to one part refuse).
The slurry mixture is then mechanically
dewatered, resulting in a thickened sludge
which is thermally dried to about 15% mois-
ture content. The dried refuse-sludge
mixture, impregnated with sodium aluminate,
304
-------
is then carbonized for about one hour at
480-570°F.
Engineering Design
A preliminary engineering study of a
plant to produce carbon char from MSW and
sewage sludge was conducted. Based on the
information accumulated, a design for a low
temperature pyrolysis facility was devel-
oped. The proposed facility would process
1000 tons per 20 hour day (50 TPH) for six
days per week. The basic design for this
plant is based on the Black-Clawson wet pro-
cessing system at Franklin, Ohio'2'. The
process flow essentially follows the Black-
Clawson process to dewatering at the hydro-
denser. After partial dewatering at the
hydrodenser, the slurried refuse would be
mixed with sodium aluminate and sewage
sludge in a flash mixer. This slurry mix-
ture is then mechanically dewatered in a
cone press and dried in a rotary carbon-
ization kiln. A flow plan and material
balance for the proposed plant is presented
in Figure 1, and an energy balance is
presented in Table 1.
A plant processing 1000 tons of refuse
per day could generate up to 300 tons of
char per day or about 79,500 tons per year.
Economic Analysis
A preliminary economic analysis was
prepared for the proposed pyrolysis plant.
This analysis is based on the process flow
plan shown in Figure 1. The proposed plant
would have two processing lines each with a
capacity of 25 tons per hour. It is assumed
that the facility would be processing
265,200 tons of refuse annually [1000 x 6 x
52 x 0.85 (on-line availability)]. It is
further assumed that an adjacent landfill
would serve as a backup to the pyrolysis
facility. Revenues for the facility would
be provided from several sources: a) a
tipping fee, b) sludge disposal, c) sale of
aluminum and iron, and d) sale of carbon
char. A summary of the calculations devel-
oped for the capital and operating costs
are presented in Tables 2 and 3 respec-
tively. Consideration of possible revenues
are presented in Table 4. The data pre-
sented in these tables show an estimated
total operating cost of $26.68/throughput
ton of MSW and a potential revenue of
$17.15/throughput ton of MSW plus revenue
from sale of carbon char. For the pyrolysis
facility to break even, the carbon char
would have to sell for $0.017 per pound
($34/ton) in order to cover the $9.53 defi-
cit calculated between total operating cost
and identified potential revenues.
Market Assessment
A major concern is the market potential
for the char residue. If large quantities
are to be produced, then long term valid
markets must be available.
A number of potential markets have been
identified for the char resulting from the
pyrolysis process: a) solid fuel, b) feed
stock for preparing gaseous or liquid fuels
and c) substitute for the carbon now being
used in carbon and graphite products
(activated carbon, charcoal, carbon fillers,
carbon risers, etc.).
It appears that the char can be used as
a solid fuel and as feed stock for synthetic
liquid and gaseous fuels at a market value
of about $0.01/lb. With some additional
processing the char could also be utilized
as an activated carbon, a raw material for
charcoal briquettes, a carbon riser in iron
and steel production and as fill in non-
structural rubber products at a market value
of $0.02-$0.04/lb. The major questions
regarding these latter applications concern
the availability of sufficient market
demand. A plant processing 1000 tons of
refuse per day could more than saturate many
of the potential, market opportunities iden-
tified (79,500 tons/year). It is, therefore,
very apparent that each specific char-
producing plant built must secure a certi-
fied market for its product from one of the
applications identified.
Conclusions
The use of chemical treatments to pro-
mote carbon char formation at lower pyroly-
sis temperatures has potential if a reliable
market for the char is secured. Laboratory
results clearly demonstrated that several
chemical treatments can be employed to gen-
erate char by low temperature pyrolysis.
Both sodium aluminate and ammonium chloride
showed promise as char promoters^). An
analysis of the projected economics for a
low temperature pyrolysis process reveals
that a market providing about $0.017 per
pound for the char is necessary to make the
process economically viable.
305
-------
TABLE 1. MATERIAL AND ENERGY BALANCE CALCULATION FOR
ONE TON OF MUNICIPAL SOLID WASTE
MSU f
10.5
Sludge
M Btu
Pyrolysis
Process
i
Ene
8.73 M Btu Net Energy
Process Output
rgy for Process
4.61 M Btu
^.
4.12 M Btu
Estimated Energy Efficiency
(8.73 - 4.12) M Btu
10.4 M Btu
+ 100 = 44%
tuts.
WHIIE GOODS.
mi STUMPS.
IOCS
Figure 1. Flow Plan for a Plant to Convert MSW to a Carbon Char.
306
-------
TABLE 2. ESTIMATED CAPITAL COSTS (1977 Dollars)
Land (30 Acres at $10,000/acre) 5 300,000
Site preparation (35% of land cost) lOs'oOO
Equipment, installation and facility construction 18,56o'ooO
Engineering and Design (6% of $18,965,000) 1,138,000
Contingencies (10% of $20,103,000) 2,010,000
Start up and work capital 500,000
SUBTOTAL 22,613,000
Legal and financial costs (2%) 452,000
TOTAL PLANT COST 23,065,'000
Municipal revenue bonds for project
(financing costs raise bond requirements
12% abova capital cost of facility) 25,833,000
Estimated life of facility - 20 years
TOTAL INTEREST TO BE PAID (8%) 26,397,655
TOTAL CAPITAL COST 52,230,655
ANNUAL CAPITAL COST 2,611,533
ANNUAL THROUGHPUT (tons) 265,200
Capital Cost per ton 9.85
TABLE 3. ESTIMATED ANNUAL OPERATING &
MAINTENANCE COSTS (1977 Dollars)
Salaries & Benefits $ 923,000
Fuel 75,000
Electricity 963,000
Water s Sewer 111,000
Maintenance 696,000
Residue Removal 190,000
Materials S Supplies 1,070,000
Taxes (.75%) 173,000
Insurance S Management Costs ($1.00/ton) 265,000
Total Annual Operating & Maintenance Costs 4,463,000
Operating & Maintenance Cost per ton 16.83
TABLE 4. POTENTIAL REVENUE SOURCES
Revenue Source Dollars/throughput ton
Tipping fee $ 8.50/ton
Sludge disposal (200 Ibs @ $50/ton) 5.00/ton
Ferrous metal (58 TPD at a net cost of 1.31/ton
$22.55/ton)
Stock for aluminum recovery 18 TPD 2.34
@ net cost of $130/ton
Subtotal 17.15
Carbon Char (295 TPD market values not ?
established)
307
-------
Another major reservation concerning
the use of the carbon char is the lack of
familiarity with it as a product. Potential
consumers are hesitant to commit themselves
to an untried material. To date, the com-
mercialization of char generated from con-
ventional pyrolysis processes has been
disappointing,and this serves to moderate
enthusiasm for a low temperature pyrolysis
process designed to produce a carbon char
product.
POWDERED FUEL
Converting the organic fraction of
refuse into a finely powdered material of-
fers a number of advantages. In powdered
form the refuse is more compatible with the
fuel used in suspension fired boilers; it is
more easily slurried with oil for firing in
liquid fuel units; it is more easily pellet-
ized, and it can be more readily used as a
filler.
A number of thermal and chemical treat-
ments have been described in the literature
for promoting the conversion of the organic
fraction in refuse to a fine powder. Since
the major constituent of the organic frac-
tion is cellulose (75%), the treatments are
based primarily on the technology of cellu-
lose processing. Most of these processes
are chosen to embrittle or degrade the
cellulose by reducing the degree of poly-
merization. In addition to degradation,
certain acid treatments appear to promote
crosslinking of adjacent molecules, which
makes the cellulose rigid and brittle.
Heating cellulose in air up to temperatures
of 400°F will result in embrittlement.
Cellulose treated with formaldehyde also
undergoes embrittlement.
Laboratory Studies at UDRI
During this project a number of chemi-
cals were screened for their ability to em-
brittle paper and other cellulose wastes. A
list of the chemicals studied is compiled in
Table 5. The results of these initial stud-
ies showed that HC1 is an extremely strong
embrittlement agent. Chlorine gas and
thionyl chloride (SOC12) also were found to
be effective embrittlement agents. SO2
proved to be a moderate embrittlement agent.
The effectiveness of SC>2 was greatly en-
hanced by the inclusion of formaldehyde in
the reaction mixture. The experimental
arrangement used in these studies is shown
in Figure 2.
TABLE 5. EMBRITTLEMENT CHEMICALS
EVALUATED
Chemical
Formalin
Formalin + C02
Formalin + NC>2
Formalin
Formalin
Formalin
Formalin
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
+ Formic Acid
+ Acetic Acid
+ SO2
+ HC1
Paraformaldehyde
Methylal + SO2
Acetal + SO2
HC1
S02
SOC12
To better define the quantitative re-
quirements for cellulose embrittlement, a
second series of experiments was conducted
for evaluating the reaction of shredded
newsprint with hydrochloric acid, sulfur
dioxide-formaldehyde and chlorine gas. The
experimental arrangement shown in Figure 2
was used for these experiments.
The results of these experiments showed
that a greater quantity of C12 was required
for a similar degree of embrittlement as
achieved with HC1. In addition, the degree
of embrittlement with HC1 and C12 is almost
unaffected if the quantity of the reactant
gases is in excess or equal to the required
quantity. If the quantity of reactant gas
supplied is less than the required value,
poor embrittlement occurs. In a large scale
process this would mean additional energy
would be required to power the refuse if an
insufficient quantity of reactant was used.
The S02 requirement for sulfur dioxide-
formaldehyde embrittlement was also almost
twice as much as HC1. The degree of embrit-
tlement by this process was relatively poor
and the results showed that the time
required to powder these samples was almost
5 times as much when compared with the pow-
dering time for similar HC1 or C12 treated
samples. The results from these laboratory
experiments showed HC1 to be the most suit-
able reagent as compared to C12 and S02-
A further series of experiments study-
ing the embrittlement of shredded newsprint
with hydrogen chloride was also conducted.
308
-------
250W
Heat
Lamp
• Gas Inlet
Vacuum Connection
Reactor
Flask
/ 0-Ring Seal
^•^. \xGlass Plate
Figure 2. Experimental Arrangement Used for Embrittlement Studies.
using a specially designed pilot size reac-
tor system. The purpose of these experi-
ments was to evaluate process variables with
a pilot reactor. Operational problems with
the system were analyzed throughout the
study and reactor performance was monitored.
The reactor unit used was built by the
University of Dayton Research Institute.
The reactor system (shown in Figure 3) was
designed to process 2 1/2 to 5 Ibs. of
shredded waste samples with a continuous
flow of reactant gases. The reactor was
made from a 30 gallon carbon steel drum.
The interior of the drum was coated with a
20 mil thickness of Emralon 314 (50%
Teflon - 50% Epoxy). The reactor was sup-
ported in a Unistrut frame such that it
could be loaded and unloaded by tilting it
about a horizontal axis between two vertical
Unistrut channels. Provisions were made to
feed the shredded waste from the top and
the lid was sealed with a specially designed
rubber gasket and locking ring to prevent
leakage. The reactor contents were heated
with three 1800 watt drum heaters which were
installed on the outside of the reactor.
Hydrogen chloride was supplied from a 60 Ib.
HC1 cylinder and N2 was drawn from an
existing N2 bank. .A 5 KW air heater was
used to heat the N-
Two glass rotameters
(0. - 3.5 cfm) were used to measure the rate
of the reactant gases flowing through the
reactor. A perforated stainless steel plate
(.33" diameter holes) was installed at the
bottom of the reactor to ensure even gas
flow. A safety relief valve was installed
on top of the reactor. All temperature
measurements were made with a digital
thermometer.
The scrubber was made from a polyvinyl
chloride lined 55 gal. drum in which unre-
acted HC1 in N2 was neutralized in 84 liters
of .15N sodium hydroxide solution. The con-
tacting pattern was improved by dispersing
the unreacted gases through a 6" diameter
fritted glass funnel into the NaOH solution.
Efficiency of the scrubber was tested by
neutralizing a measured flow of HC1 into the
scrubber. The scrubber solution was then
titrated with .IN HC1 and the results were
compared with the flow measurement. A
series of tests showed that the scrubber
309
-------
SOLENOID VALVE
N2 FLOW METER
HCI FLOW METER
lo
I-1
O
SOLENOID
VALVE
3/8"ID POLY-
ETHYLENE
TUBING
—- HCI CYLINDER
AIR
3/4" RUBBER
HOSE
PRESSURE REGULATOR
\3/4" BRAIDED WIRE, TEFLON
X I
LINED TUBING
30 GAL DRUM
(TEFLON
COATED
DRUM HEATER
(1800 Watts)
NaOH
SOLUTION
THERMOMETER
3/8 I.D TEFLON TUBING
55 GAL DRUM
(SCRUBBER)
(FRITTED GLASS
FUNNEL)
SAMPLE
PORT
Figure 3. Process Flow Diagram
-------
accounted for an average of 95 percent of
the HC1.
A ball mill unit was used to powder the
treated shredded paper. The mill consists
of two horizontal hard rubber rollers that
are 2" in diameter. One of the rollers is
chain driven by a 1/4 H/P motor. A 13 qt.
ceramic jar is filled with 30 Ibs. of 1/2" x
1/2" cylindrical grinding pellets and the
treated paper. The charge in the mill is
powdered by rotating the jar between the
two rollers.
From the data acquired for this series
of tests, it is observed that the degree of
embrittlement is maximized at:
•the higher treatment temperatures
•the longer reaction times
•minimum moisture contents
•moderate levels of HC1 concentration
With the reactor unit used for these tests,
the best operating results were obtained at
300°F, six minutes reaction time, and 50
percent (by volume) concentration of HC1
with N2- From some preliminary tests, it is
anticipated that the degree of embrittlement
could be increased by extending the treat-
ment temperature to between 350°F and 400°F
and the reaction time to 8-10 minutes. At
temperatures above these values, spontaneous
combustion and other difficulties may be
encountered. It is also desirable to mini-
mize the quantity of HC1 required. It
should be noted that a 50 percent reduction
in HC1 consumption was achieved for the
tests conducted in the pilot sized reactor
compared to the previous tests conducted in
the laboratory sized reactor.
From these embrittlement studies it
appears that the HC1 serves as a catalyst
for a complex embrittlement reaction in the
cellulose. The presence of heat enhances
the reaction. However, the specific mecha-
nisms are still not completely understood.
Engineering Design
A limited engineering analysis was con-
ducted for a plant to produce a powdered
fuel from refuse. The first phase of this
study was the development of a design for a
plant to process 1,000 tons of refuse per
day. The proposed facility would contain
two processing lines (each 30 ton/hr capa-
city) and would operate 20 hours per day,
six days per week. In the flow plan
developed, the incoming refuse would be
screened in a trommel to separate the minus
4" fraction and open the trash bags. The
oversized (-1-4") would be shredded and both
fractions passed under magnetic separators
to recover the ferrous metal. The non-
ferrous fraction would be screened in a
trommel to separate out the fine fraction
(minus 1/2") prior to air classification.
The heavy fraction from the air classifier
and the fines from the 1/2" trommel are
burned in an incinerator and provide the
thermal energy required for embrittlement.
If aluminum and glass recovery are desired,
this would be accomplished by further pro-
cessing the heavy fraction collected from
the air classifier. Aluminum can also be
recovered from the incinerated residue, but
it would not be of as high a quality. The
light fraction obtained from air classifi-
cation would be embrittled and powdered in
an air-swept ball mill (Hardinge Mill). The
embrittlement is accomplished by treating
the light fraction with heated HC1, in the
Hardinge Mill. The powdered material would
be swept from the mill entrained in the
reactant gas when it is the appropriate size.
The data used for developing the flow plan
for the proposed facility were obtained from
University of Dayton studies^, Combustion
Equipment Associated data reported in U. S.
Patent 3,961,913(5), and NCRR studies for
New Orleans(°). The flow plan established
for a facility to produce fine powdered fuel
from refuse is presented in Figure 4. The
calculations developed for the material bal-
ance used in the flow plan are presented in
Table 6. The flow plan and material balance
developed in this phase of the program
served as the basis for determining equip-
ment specifications and operating require-
ments. This information in turn was used to
develop the economic analysis for the
proposed plant.
Economic Analysis
A preliminary economic analysis was
prepared for the powdered fuel process. The
analysis is based on the process flow plan
developed (Figure 4). Revenues are antici-
pated from: a) tipping fee, b) sale of
recovered metals, c) sale of powdered fuel.
A summary of the calculations developed
for the capital and operating costs are
presented in Tables 7 and 8 respectively.
A summary of potential revenues are
presented in Table 9.
311
-------
RAW REFUSE
BULKY WASTES
WHITE GOODS,
TIRES, IOCS,
FURNITURE, ETC.
REE STUMPS
Figure 4. Resource Recovery Plant to Produce Powdered Fuel.
TABLE 6. MATERIAL BALANCE FOR A POWDERED FUEL PROCESS
1
;
1
IfeftjSC
Onjanic
Fiatfion 750
MftUl 100
Petrous 05
Nnn ferrous 15
Glass and
ollcar inorts ISO
i
• r-
H Ml
1
40
10
9
1
7.5
o
1 r)
7 -
225
67
57
10
138
U
75
485
2J
4
4.5
\\
22
76.5
75
l.S
1.5
!-
i ^
688
13.5
1
12.5
141
Ss
45
0.5
U.5
10.5
Sf
+ ^
64)
U
1
12
130.5
Tnnnal Me
1-/2" classification
3-
,
t.
m u
•jj Jj
C r
M *•
135
Hi!
1
10. S
141
r* f
•*4 U
It
15
,1^-
10
JJLi
i
10.5
141
I.
If
i ^
& i/
55J
1
^
2
-
312
-------
TABLE 7. ESTIMATED CAPITAL COSTS (1977 Dollars)
Land (30 Acres @ $10,000/acre) $ 300 000
Site Preparation (35% of land cost) los'ooo
Equipment, Installation a Facility Construction 15,456^000
Engineering & Design (6% of $15,681,000) ' 952,000
Contingencies (10% of $16,813,000) 1,681,000
SUB TOTAL 18,494,000
Legal fi Financial Costs (2%) 370,000
TOTAL PLANT COSTS $18,864,000
Municipal Revenue Bonds for the Project $21,128.000
(1.12 plant cost)
Annual Interest Rate for 20 years @ 8% 1,095,534
Total Interest to be paid 21,910,669
Total Capital Cost 43,038,669
Annual Capital Cost 2,151,934
Capital Cost Per Ton 8.12
TABLE 8. ESTIMATED ANNUAL OPERATING AND
MAINTENANCE COSTS (1977 Dollars)
Salaries and Benefits $ 872,000
Fuel 100,000
Electricity 437,000
Water and Sewer 3,000
Maintenance 666,000
Residue Removal 250,000
Materials and Supplies 711,000
Taxes (.75%) 142,000
Insurance and Management Costs ($l/ton) 265,000
Total Annual Operating & Maintenance Cost $3,446,000
Operating and Maintenance Cost Per Ton 12.99
TABLE 9. POTENTIAL REVENUE SOURCES
Revenue Source Dollars/Throughput Ton
Tipping Fee 8.50
Ferrous Metal (86 TPD @ a net cost of $22.55/ton) 1.94
Powdered Fuel (550 TPD) ??
Cost/ton 21.11
Cost needed for sale of fuel to break even 10.67
$.01/lb of powdered fuel § 7,500 Btu/lb - 1.33/M Btu
313
-------
From the analysis developed, a total
operating cost per throughput ton of MSW
was calculated to be $21.11. A potential
revenue of $10.44/throughput ton was calcu-
lated for the sale of ferrous and the tip-
ping fee. To break even, the powdered fuel
would have to sell for about $0.01/lb or
about $1.33/million Btu.
Conclusions
The use of powdered fuel prepared from
the embrittlenient of cellulose waste offers
a number of advantages:
a) It is more compatible with powdered coal-
fired boilers and will have better com-
bustion characteristics than
conventional RDF.
b) It is more readily densified into pel-
lets and will produce less wear on the
extrusion dies.
c) It can be slurried with oil for firing
in oil fired units.
d) It can be used as the feed stock for
further chemical or thermal treatments
like hydrolysis, hydrogenation,
gasification, etc.
In the work to date we have identified
a number of chemical treatments for cellu-
lose embrittlement and evaluated their
effectiveness in laboratory studies. In
addition^ •& preliminary design for a resource
recovery plan to produce powdered fuel was
developed. This design plan was the basis
for a preliminary economic analysis. From
the data developed, it was calculated that
the powdered fuel would have to sell for
about $0.01/lb for a facility having a
tipping fee of $8.50.
FUTURE WORK
Laboratory studies at UDRI have demon-
strated that a number of treatments can be
used for the embrittlement of cellulose
wastes to produce a powdered fuel. Although
the process variables have been determined
and a preliminary engineering design and
economic analysis have been established for
the production of a P-RDF, there is need for
further research. A number of studies to
better characterize the P-RDF as a fuel or
fuel component are planned. In addition,
studies to extend the utilization of
embrittlement in resource recovery processes
are planned.
A compilation of the tasks to be ini-
tiated in the next phase of the project are
presented below:
(a) Combustion analysis of P-RDF fuel.
(b) Oil/P-RDF slurry development.
(c) Embrittlement studies on
unprocessed raw refuse.
(d) Embrittlement studies on wood
product and crop wastes.
(e) Investigation of partial embrittle-
ment techniques for preparing RDF
flakes.
(f) Economic reevaluation of
embrittlement processes.
Although P-RDF has been characterized,
performance in burn tests has not been es-
tablished. Therefore, combustion studies
of the powdered fuel prepared from the
embrittlement of RDF will be conducted in a
test unit with a burner for pulverized coal.
Another area of major interest is the
suspension of the P-RDF in an oil. The
handling and shipment of a powdered fuel to
its point of use presents some problems - a
major one being safety hazards arising from
the potential explosion characteristics of
a finely powdered material. A possible
solution to this problem is storing, trans-
porting, and burning the powdered fuel in an
oil slurry. The techniques used for powder
suspension will initially be based on the
work done in developing mixtures of pow-
dered coal in oil. Once an effective P-RD^oil
slurry is developed, mixtures of 10, 25, and
50 percent powder/oil slurries will be pre-
pared. These slurries will be subjected to
combustion tests. It is expected that an
air atomizing burner unit would be used for
the proposed combustion tests.
Another important task will be the
evaluation of the economics for the embrit-
tlement process. The results of the combus-
tion tests and additional research tasks
will be utilized to reevaluate the economics
for powder preparation and its utilization
as a fuel.
314
-------
REFERENCES
1. Harendza-Harinxma, A. Method of Carbon-
izing a Substance Comprising Cellulose.
U. S. Patent 3,961,025, 1976.
2. Wittman, T. J. et al. A Technical,
Environmental and Economic Evaluation
of the Wet Processing System for the
Recovery and Disposal of Municipal Solid
Waste. U. S. Environmental Protection
Agency, Washington, D. C., 1975. 217 pp.
3. Hecht, N. L., Duvall, D. S., Fox, B. L.,
"Investigation of Advanced Thermal-
Chemical Concepts for Obtaining Im-
proved MSW-Derived Products," EPA Report
600/7-78-143, Environmental Protection
Agency, Cincinnati, OH, August 1978.
4. Hecht, N. L. et al. Design for a
Resource Recovery Plant for AMAX Inc.
University of Dayton, Dayton, OH, 1974.
5. Brennemon, R. S. and Clancy, J. J.
Process for Treating Organic Wastes and
Products Thereof. U. S. Patent
3,961,913, June 8, 1976.
6. National Center for Resource Recovery,
Inc., New Orleans Resource Recovery
Facility. Implementation Study: Equip-
ment, Economics and Environment.
National Center for Resource Recovery,
Inc., Washington, D. C., 1977.
315
-------
ENVIRONMENTAL IMPACT OF RESOURCE RECOVERY
Judith G. Gordon
MITRE Corporation
Metrek Division
McLean, Virginia
ABSTRACT
This report summarizes an assessment of the impact of resource recovery on the en-
vironment. The study examined the enviornmental effects that will derive from municipal
solid waste disposal in 1990 and the changes in these effects that will result from im-
plementation of resource recovery from municipal solid waste. The direct effects of
municipal solid waste disposal as well as the secondary effects of substituting recovered
materials for raw materials in the production of steel, aluminum, glass, and energy were
considered. The methodology used in the analysis and the findings are discussed in this
paper.
INTRODUCTION
Interest in resource recovery from mu-
nicipal solid waste has increased in recent
years, and resource recovery is generally con-
sidered a worthwhile goal. Proponents of re-
source recovery maintain that establishment
of resource recovery facilities should be en-
couraged by the Federal government. However,
in the absence of an assessment of the effects
on the environment of resource recovery, it
was not known what these effects would be nor
the net impact of resource recovery on the
environmental effects of municipal solid waste
disposal. This study was initiated with just
such an objective — to determine quantita-
tively the environmental effects of resource
recovery from the perspective of the overall
nationwide impacts.
The objectives of this study were, there-
fore, to determine:
(a) The direct effects on the environment of
municipal solid waste disposal and of
resource recovery from municipal solid
waste,
(b) The secondary effects of resource re-
covery resulting from substitution of
recovered materials for virgin ma-
terials, i.e., environmental effects
and impact on energy, and
(c) The potential benefits of resource
recovery in mitigating the adverse
environmental effects of municipal
solid waste disposal.
The scope of the study was determined
to some extent by the selected definitions
of terms. Municipal solid waste was de-
fined as being comprised of residential and
commercial solid waste, i.e., the urban
refuse that is normally handled by munici-
palities. In resource recovery, recovery
will be achieved by front-end processing
and by mass burning followed by materials
recovery from the residue; source separa-
tion was not considered. Scrap ferrous
metals, aluminum, and glass will be re-
covered from municipal solid waste and will
be substituted for virgin materials in the
production of steel, aluminum products,
and glass containers. The energy content of
municipal solid waste will also be recovered,
replacing some coal combustion in energy
production.
316
-------
TABLE 1
/
1990 SCENARIOS FOR HANDLING OF MUNICIPAL SOLID WASTE
DISPOSAL METHOD/
PROCESS
Mass Burning
Landfill
Resource Recovery
Front-end processing with
materials separation and
combustion of RDF
Mass burning of raw ref-
use with energy recovery
(waterwall incineration)
and materials recovery
from residue
TOTAL
WITHOUT RESOURCE RECOVERY
%
5
95
tons
10,000,000
187,000,000
197,000,000
WITH RESOURCE RECOVERY
%
5
80
7.5
7.5
tons
10,000,000
157,000,000
15,000,000
15,000,000
197,000,000
The analysis was based on the year 1990
which was selected because it is the earli-
est date by which the number of fully opera-
tional resource recovery facilities will be
sufficient for resource recovery to have a
significant impact on the environment. Two
scenarios were developed for municipal solid
waste disposal in 1990 — one without and
the other with resource recovery (Table 1).
METHODOLOGY
Environmental Effects
The factors considered in this study
were air pollutants, water pollutants, land-
fill capacity, and energy conservation.
Analysis of air pollutants was the most
comprehensive because of the availability of
data on pollutant emissions from municipal
solid waste disposal and resource recovery -
related activities! Data sources were pri-
marily the United States Environmental Pro-
tection Agency (1) and studies of the Ames,
Iowa, Solid Waste Recovery System (2,3).
Data were presented for emissions of par-
ticulates, oxides.of sulfur and nitrogen,
carbon monoxide, hydrocarbons, aldehydes,
chlorides, and ammonia as well as for land-
fill generation of carbon dioxide and
methane. Pollutant emissions from munici-
pal solid waste disposal, from materials
production, and from energy production
were quantified.
Analysis of water pollutants per-
tained only to the direct effects of mu-
nicipal solid waste disposal without and
with resource recovery; the secondary ef-
fects of resource"recovery, i.e.. pollu-
tants from materials production and energy
production, were not included in this study.
Data sources were incinerator studies (4,
5,6) and leachate studies (7,8,9); equi-
librium modeling of leachate was used to
project pollutant concentrations in 1990
(10,11,12). Data were presented for vari-
ous metals and other elements as well as
several anions and compounds. Quantities
of pollutants in wastewater discharges to
317
-------
surface waters and in leachate* from land-
fills were estimated.
The requirements for landfill capacity
for disposal of municipal solid waste were
estimated. Estimates were made for the two
scenarios -- without and with resource re-
covery.
The effects of resource recovery on
energy conservation were quantified. Energy
conservation in materials production from
recovered materials rather than from virgin
materials and energy recovery from municipal
solid waste were the aspects analyzed.
Analysis
The methodology developed for quanti-
fying the environmental effects of municipal
solid waste disposal and resource recovery
is based on two components: "effect factors"
and "quantifiers." The effect factor per-
tains to the particular environmental effect,
or component thereof, that is being consid-
ered and provides data on that effect in
terms of unit measure. Examples of effect
factors are the emission factor for an air
pollutant (per ton of municipal solid waste
incinerated or ton of steel produced), the
concentration of a pollutant in landfill
leachate, the density of landfilled raw ref-
use, and the energy requirement for produc-
ing a ton of steel from scrap ferrous metal.
Quantifiers are estimated from available
data and are based on specific assumptions.
Examples of quantifiers are the quantities
of municipal solid waste, leachate, and re-
covered glass, and the quantity of energy
that is recoverable from municipal solid
waste.
An illustration of the methodology is
quantification of the direct effects of
municipal solid waste disposal. The effect
factors required for the determination are
presented in Table 2 (12). The quantifiers
depend on the scenario, i.e.. without or
with resource recovery. The quantities of
pollutants that will derive from disposal
of municipal solid waste (MSW) in 1990 are
calculated from the following equations.
Calculation of Quantities of Air
Pollutants That Will Be Emitted
during Mass Buring of MSW
Qp = Efp x Mj x f
(1)
where Qp = quantity of a pollutant that
will be emitted (tons)
Efp = emission factor for that pol-
lutant (pounds/ton MSW incin-
erated)
Mj = mass of MSW that will be
disposed of by incineration
(tons)
f = conversion factor (from
pounds to tons)
= 1/2000
Calculation of Quantities of &ases
That Will Be Generated in Landfills
in 1991 by MSW Landfilled in 1990
Qp = Gp x f1 x ML x Dp x f2 (2)
where Qp = quantity of the gas that will
be generated in the one year
from the landfilled MSW (tons)
Gp = maximum quantity of gas gen-
eration over 25 years (cubic
feet/ton MSW)
f, = time factor (one year out of
25 years during which gas is
generated)
= 1/25
M. = mass of MSW that will be
L landfilled in 1990 (tons)
Dp = density of the gas (pounds/
standard cubic foot)
fo = conversion factor (from
pounds to tons)
= 1/2000
The total pollutant loading in leachate does
not represent the quantities of pollutants
that eventually enter the ground and/or
surface waters.
318
-------
TABLE 2
PROJECTED CONCENTRATIONS OF POLLUTANTS FROM MUNICIPAL SOLID WASTE (MSW) DISPOSAL IN 1990
• . — — ,
RECEIVING
MEDIUM/POLLUTANT
AIR
Particulates
S0x (as S02)
NO
X
CO
Hydrocarbons
(as CH4)
Aldehydes (as
Formaldehyde)
Chlorides
(as HC1)
C00
2
CH.
4
WATER
Aluminum
Barium
Cadmium
Calcium
Chloride
Chromium
Copper
SOURCE
MASS BURNING
STACK
EMISSIONS
(Ib/ton)
1.02
2.02
1.14 - 3.43
0.69 - 2.07
0.2
0.0004 - 0.01
3.71 - 7.41
NA8
WASTE-
WATER
(ppm)
20.4
5.0
NA
42
30 - 193
0.13
0.02
LANDFILL OF MSW JENERGY RECOVERY FROM RDF
LEACHATEC
(ppm)
NA
NA
0.03
60 - 7200
5 - 2467
1.05
1.30
GASES
(cu ft/ton
in 25 yrs)
4560
4560
STACK
EMISSIONS
(Ib/ton RDF)
1.28
4.43
1.54
NA
0.0036
-0.36
7.81
NA
WASTE -
WATER6
(ppm)
~o
5.0
NA
42
30 - 193
0.13
~0
LANDFILL OR RESIDUE
LEACHATEf
(ppm)
NA
NA
0.00578
26.8
NA
NA
1.26
GASES
(cu ft/ton)
0
0
-------
TABLE 2 (Concluded)
RECEIVING
MEDIUM/POLLUTANT
WATER (cont)
Iron
Lead
Magnesium
Manganese
Nickel
Phenols
Phosphate
Potassium
Sodium
Sulfate
Zinc
r r ' '" ™"™" " - - -.
SOURCE
MASS BURNING
STACK
EMISSIONS
(Ib/ton)
WASTE-
WATER15
(ppm)
0.3
0.5
0
0.30
NA
0.005
2-9
NA,.
NA
32 - 50
1.0
LANDFILL OF MSW
LEACHATEC
(ppm)
810
1.4
17 - 15,600
0.06 - 125
1.20
NA
NA
29 - 3770
34 - 7700
1 - 3770
155
GASES
(cu ft/ton
in 25 yrs)
ENERGY RECOVERY FROM RDF
STACK
EMISSIONS
(Ib/ton RDF)
WASTE -
WATER6
(ppm)
~o
~o
0
~o
NA
0.005
2-9
NA
NA
32 - 50
~o
LANDFILL OF RESIDUE
LEACHATEf
(ppm)
0.0138
0.00425
13.2
2.08
0.0220
NA
NA
NA
NA
11.9
0.138
GASES
(cu ft/ton)
10
N>
O
Source: Gordon, 1978 (Reference 12).
Incinerator wastewater discharged to surface waters.
Concentrations in leachate, not in groundwater.
Portion of cofiring emissions that is attributable to the RDF fraction.
Based on pollutants in incinerator wastewater.
Based on equilibrium composition of a hypothetical liquid-phase/solid-phase system,
%A: Data not available.
-------
Calculation of Quantities of Pollutants
That Will Be Discharged to Surface
Waters in Incinerator Wastewater
Qp = Cp x Qw x
(3)
where Qp = quantity of a pollutant that will
be discharged in incinerator waste-
water (tons)
Cp = concentrations of that pollutant in
Y the effluent
Qw = water requirement (tons/ton MSW in-
cinerated)
M, = quantity of MSW that will be incin-
erated (tons)
Calculation of Quantities of Pollutants
That Will Be Present in Leachate from
Landfilled MSW (based on reported ranges
of concentrations)
QD = CD x Q. x f, x f,
(4)
where Qp = quantity of a pollutant that will
be present in landfill leachate
(tons)
Cp = range of concentration of that pol-
lutant in the leachate (ppm)
Q, = quantity of landfill leachate (109
gallons)
f, = conversion factor (from ppm to
pounds/billion gallons)
= 8345
fy = conversion factor (from pounds to
' tons)
= 1/2000
Calculation of Quantities of Pollutants
That Will Be Present in Leachate from
Landfilled MSW (based on data for a
single landfill)
QD = CD x Q. x f, x f- x f.
(5)
where Qp = quantity of a pollutant that will be
present in landfill leachate (tons)
Cp = concentration of that pollutant in
the leachate
Q, = quantity of leachate from the
one landfill
= 20,000 cubic meters/month
f, = conversion factor (from
month to year)
= 12
f2 = conversion factor (from grams
to tons)
= 1.102 x 10"6
f, = factor for extrapolating from
one landfill to the national
landfill area
= 600,000/300
For calculations of pollutant quan-
tities for the scenario with resource re-
covery, some of the equations are used
more than once ~ with appropriate values
assigned to the effect factors and quan-
tifiers — in order to account for all
sources of a given pollutant (e.g..
leachate will be derived from landfilled
raw refuse as well as from landfilled re-
source recovery residue).
Similar methodology is used to cal-
culate all the other effects including
pollutants emitted during materials pro-
duction and during energy production,
energy conservation in materials produc-
tion from recovered scrap materials,
energy recovery from municipal solid waste,
and landfill capacity requirements. Total
effects of each type (e.g., direct effects
of municipal solid waste disposal and sec-
ondary effects of resource recovery) are
calculated by summing the individual ef-
fects. The net changes that will result
from implementation of resource recovery
are calculated by determining the differ-
ences in effects "that would derive from
the two scenarios.
The equations used in the calculations
are listed and explained in Appendix A of
Reference 12. Appendix A also includes
sample calculations that are based on (a)
the effect factors given in the tables, and
(b) the quantifiers derived from data and
assumptions that are presented in the tables
and accompanying discussion.
321
-------
RESULTS AND DISCUSSION
The effects on the environment of re-
source recovery from municipal solid waste
were determined for a specific scenario for
resource recovery implementation (Table 1),
In addition, numerous other assumptions were
made which directly affected the results of
this study. Consequently, the findings are
valid only for the particular conditions
specified in the assumptions. Any change in
the data and/or the assumptions will alter
the results. The methodology developed in
this study is readily applicable for recal-
culation of the environmental effects of
resource recovery that would result from any
changes in data and assumptions,
Assumptions
It is impossible to list all the assump-
tions in this summary of the study, and the
reader is referred to the original report
(12). Some of the most important assumptions
that were made for the year 1990 were:
• About 197,000,000 tons of municipal
solid waste will be generated in the
United States.
• The scenarios for municipal solid waste
disposal without and with resource re-
covery (Table 1) are valid.
t The effect factors are representative
of actual operating conditions.
• The 1975-to-1990 increases in landfill
area and in leachate from municipal
landfills will be proportional to the
increases in required landfill capacity
under each scenario.
• Recovery efficiencies for ferrous metals,
aluminum, and glass will be the same for
front-end separation and recovery after
mass burning.
• All the materials recovered from municipal
solid waste will be recycled and will re-
place virgin materials in materials pro-
duction.
• Steel, aluminum, and glass will be pro-
duced by the same processes in the same
relative quantities and at the same pro-
duction efficiencies as in 1975.
• Energy requirements for materials
production from virgin and from
recovered materials will be the
same as in 1975.
• Energy recovery from municipal solid
waste will amount to 161.4 x 1012
Btu and will replace 0.8% of the
power that otherwise would be gen-
erated by coal combustion.
Findings
The various effects of implementation
of resource recovery in accordance with
the scenario for 1990 are summarized in
Tables 3, 4, and 5, Table 3 summarizes
the effects on air emissions, Table 4,
the effects on water resources, and Table
5 the miscellaneous effects. The environ-
mental effects of resource recovery from
municipal solid waste are primarily favor-
able.
Table 3 presents the direct effects
and the secondary effects of resource
recovery on air emissions as well as the
summed -- i.e.. net -- effects. Net
emissions of all but three of the air
pollutants considered will be reduced.
Emissions of carbon monoxide will be re-
duced in 1990 by about 2,300,000 tons
and emissions of carbon dioxide, NOX,
and methane by about 310,000 tons, 150,000
tons, and 114,000 tons, respectively.
Smaller reductions will occur in emissions
of SOX ("^23,000 tons) and hydrocarbons
( ^7200 tons). There will be slight in-
creases in the quantities of total par-
ticulates (^1000 tons) and
aldehydes (^2500 tons) emitted and a
larger increase '(••v 140,000 tons) in
chloride emissions. Emissions of par-
ti cul ate and gaseous fluorides will not
be affected by resource recovery.
In Table 4 are listed the direct
effects of municipal solid waste disposal
on the water resources. The quantities
of pollutants that will be discharged to
surface waters in incinerator wastewater
effluent will increase because more mu-
nicipal solid waste will be incinerated
with implementation of the scenario for
resource recovery. With resource re-
covery from municipal solid waste, smaller
322
-------
TABLE 3
SUMMARY OF THE EFFECTS ON AIR EMISSIONS OF IMPLEMENTATION OF RESOURCE RECOVERY
IS ACCORDANCE WITH THE SCENARIO FOR 1990a
POLLUTANT
Particulates , total
SO (as SO )
NO (as NO )
CO
Hydrocarbons
(as CH.)
Aldehydes
(as formaldehyde)
Chlorides (as HC1)
Particulate Fluorides
Gaseous Fluorides
NH3
co2
CH,
DIRECT EFFECTS b
DISPOSAL
OF MSW
+14,850
+40,050
+(17,200 to 34,400)
+(5175 to 15,550)
+1500
-(1900 to 1970)
+(71,800 to 99,450)
0
0
NA
-310,000
-114,000
SECONDARY EFFECTSC
MATERIALS
PRODUCTION
-10,654
-5155
-51
-2,305,000
-5385
f
NA
NA
0
0
-231
NA
ENERGY
PRODUCTION d
+4050
-32,800
-(156,980 to 174,800)
-(210 to 11,140)
-3340
+ 2470
+(81,780 to 110,940)
NA
NA
NA
NA
NET EFFECTSe
+1046
-22,805
-(131,296 to 166,316)
-(2,289,660 to 2,310,965)
-7245
+(2525 to 2595)
+(109,630 to 166,440)
0
0
-231
-310,000
-114,000
10
LO
aTons that will be emitted in 1990.
Direct effects of resource recovery on environmental impact of municipal solid waste disposal. Includes emissions
attributable to refuse-derived fuel (RDF).
Indirect environmental effects of using recovered materials to replace virgin materials in materials and energy produc-
tion. Does not include environmental effects of mining virgin materials or transporting virgin and recovered materials.
Includes pollutants from cofiring of coal and RDF. Credit is given for emissions from coal not burned.
£
Sum of the direct and indirect effects of resource recovery corrected to eliminate double counting of the emissions
attributable to RDF (which are included in both direct and secondary effects). "Plus" indicates an increase in
emissions, "minus" a decrease.
NA: Data not available.
-------
TABLE 4
SUMMARY OF THE EFFECTS ON WATER RESOURCES OF IMPLEMENTATION OF RESOURCE RECOVERY
IN ACCORDANCE WITH THE SCENARIO FOR 1990a
POLLUTANT
Aluminum
Barium
Cadmium
Calcium
Chloride
Chromium
Copper
Cyanide
Iron
Lead
Magnesium
Manganese
Nickel
Phenols
Phosphate
Potassium
Sodium
Sulfate
Zinc
DIRECT EFFECTS OF MSW DISPOSAL
DISCHARGES TO,
SURFACE WATER
+918
+394
NA
+3310
+(2365 to 15,210)
+10
+1
0
+14
+23
0
+14
NA
0
+(158 to 709)
NA
NA
+(2520 to 3940)
+45
POLLUTANTS IN
LANDFILL LEACHATE°
NAd
NA
-3
-(4400 to 540,000)
-(375 to 183,500)
-89
-112
NA
-69,500
-120
-(1210 to 1,170,000)
+5 to -9400
-103
NA
NA
-(2150 to 285,000)
-(2550 to 580,000)
-(25 to 285,000)
-13,250
"Plus" indicates an increase in pollutant quantities,
"minus" a decrease.
Net difference in tons of pollutants that will be
discharged in 1990. Assumes water recirculation
and use of 3 tons of water per ton of incinerated
municipal solid waste or ton of refuse-derived fuel.
Net difference in tons of pollutants that will be 9
present in landfill leachate in 1990. Assumes 90 x 10
gallons of leachate from landfilled raw municipal
solid waste and 1 x 10 gallons of leachate from
landfilled resource recovery residue.
d
NA: Data not available
324
-------
quantities of pollutants will be present
in landfill leachate. The decreases will
range from 3 tons of cadmium, a minimum of
25 tons of sulfate and about 100 tons of
chromium, copper, lead, and nickel to more
than 1,000,000 tons of magnesium. Because
of various factors, (e.g., soil interac-
tions), the total pollutant loading in
landfill leachate does not enter the ground
and/or surface waters. In the absence of
data on leachate contributions to the pol-
lutant loadings of ground and surface
waters, the net effects of resource recov-
ery on the water resources cannot be de-
termined.
The miscellaneous effects of resource
recovery -- specifically the requirements
for landfill capacity and energy considera-
tions — are summarized in Table 5. The
requirement for landfill capacity to dispose
of municipal solid waste and the residues
from mass burning and resource recovery will
be decreased by about 44,260 acre-feet as a
result of the direct .effects of resource re-
covery. The landfill capacity needed with
resource recovery will be about 85 percent
of the capacity that would be required if
there were no resource recovery from muni-
cipal solid waste. The secondary effects
of resource recovery will reduce the quan-
tities of other solid wastes requiring dis-
posal; however, the reductions in quantities
of solid wastes from mining of virgin ma-
terials, from materials production, and from
coal combustion were not quantified in this
study.
Data on energy recovery from municipal
solid waste and energy conservation in ma-
terials production from recovered materials
are summarized in Table 5. Energy derived
from municipal solid waste in 1990 will a-
mount to 161.4 x 1012 Btu (28.8 x 106 bar-
rels of crude oil equivalent). Energy con-
servation that will accrue from substitution
of recovered for virgin materials in the
production of steel, aluminum, and glass
will be about 75.2 x 1012 Btu (13,4 x 106
BCOE). The total energy savings that could
be realized from resource recovery from mu-
nicipal solid waste will therefore be about
236,6 x 1012 Btu or 42,3 x 106 barrels of
crude oil equivalent.
Discussion
The findings of this study confirm
the expectations that the impacts of re-
source recovery on the environment will
be beneficial. Nevertheless, an equally
important benefit derived from this study
is probably the development of a method-
ology for assessing these impacts.
The importance of the methodology
lies in its versatility and its applica-
bility to the study of the environmental
effects of resource recovery. With the
equations developed for this study, it
is possible to calculate environmental
effects easily and rapidly. Therefore,
although the particular findings pre-
sented here may become obsolete as new
data become available, this methodology
provides the means for recalculation of
the environmental effects.
Suggested uses of this methodology
include:
• Filling in gaps in this study that
resulted from unavailable data,
• Extending the analysis to environ-
mental effects not considered in
this study (e.g., the secondary
effects of resource recovery on
the water resources),
• Recalculating and updating the
environmental effects of resource
recovery as additional and better
data become available, and
• Performing sensitivity analyses to
determine the differences that
would result from changes in the
assumptions and other variables.
325
-------
TABLE 5
SUMMARY OF THE MISCELLANEOUS EFFECTS OF IMPLEMENTATION OF RESOURCE RECOVERY
IN ACCORDANCE WITH THE SCENARIO FOR 1990
EFFECT
Requirement for Landfill
Capacity (acre-feet)
Energy Recovery
(Btu)
(BCOE)b
Energy Conservation
(Btu)
(BCOE)b
DIRECT EFFECTS
MSW
DISPOSAL
-44.2603
SECONDARY EFFECTS
MATERIALS
PRODUCTION
b
-75,243,000 x 106
13,436,000
ENERGY
PRODUCTION
b
161,400,000 x 106
28,821,000
TOTAL
-44,260a
•
f.
236,643,000 x 10 Btu
42,258,000 BCOEC
"Minus" indicates a decrease in requirement for landfill capacity.
b,
Solid wastes from mining of virgin materials, from materials production, and from coal combustion were not included
in this study.
r^
"BCOE: Barrels of crude oil equivalent.
-------
ACKNOWLEDGMENTS
This study was prepared for the Process-
ing Branch in the Solid and Hazardous Waste
Research Division of the Municipal Environ-
mental Research Laboratory of the United
States Environmental Protection Agency
in Cincinnati, Ohio under Contract
68-03-2596. Dr. Albert J. Klee was Project
Office. Mr. William Lowenbach performed
the analysis of leachate data and the
equilibrium modeling estimation of leachate
quality.
REFERENCES
1. United States Environmental Protection
Agency. Compilation of Air Pollutant
Emission Factors, 2nd edition. #AP-
42, with Supplements. Office of Air
and Waste Management, Office of
Air Quality Planning and Standards,
Research Triangle Park, North- Carolina,
February, 1972.
2. Hall, J.L., A.W. Joensen, G.A. Severns,
D.B. Van Meter, and H. Shanks.
Emissions from Stoker Fired Boilers-
Using Coal-RDF Mixtures. Paper pre-
sented at the 8th Biennial Waste
Processing Conference, Chicago,
Illinois, May 7-10, 1978.
3. White, Robert. Midwest Research
Institute, Kansas City, Missouri,
Communication of unpublished data,
June 8, 1978.
4. Achinger, W.C., and L.E. Daniels.
Seven Incinerators. #SW-51ts.lj.
Office of Solid Waste, U.S. Environ-
mental Protection Agency, Washington,
D.C., 1970.
5. Brinkerhoff, R.J., and W.C. Achinger.
The Braintree, Massachusetts,
Municipal Incinerator. #SW-108,
pp. 37-39. Office of Solid Waste,
U.S. Environmental Protection Agency,
Washington, D.C., 1973.
6. Weinstein, N.J. Municipal-Scale Thermal
Processing of Solid Wastes. #EPA/530/
SW-133C, pp. 203-212. Office of Solid
Waste, U.S. Environmental Protection
Agency, Washington, D.C., 1977.
10.
Chiffn, E.S.K., and F.B. DeWalle.
Sanitary Landfill Leachates and
Their Treatment. Journal of the
Environmental Engineering Division.
Proceedings of the American Soci"etv
of Civil Engineers. 102(EI2TT -
411-431, April 1976. -
Johansen, O.J., and D.A. Carlson.
Characterization of Sanitary Landfill
Leachates. Water Research (London),
10_: 1129-1134 ("1976), -
McElroy, A.D., S.Y. Chiu, J.W. Nebgen,
A. Aleti, and F.W. Bennett. Loading
Functions for Assessment of Water
Pollution from Nonpoint Sources.
#EPA-600/2-76-151, p. 239. Office
of Air, Land and Water Use, Office
of Research and Development, U.S.
Environmental Protection Agency,
Washington, D.C., May 1976.
Morel, F., and J. Morgan. A Numerical
Method for Computing Equilibria in
Aqueous Chemical Systems. Environ-
mental Science and Technology,
ss-e? (1972).
11.
12.
McDuff, R.E., and F.M. Morel.
Description and Use of the Chemical
Equilibrium Program REDEQL2.
Technical Report EQ-73-02. Keck
Laboratory of Environmental Engineering
Science, California Institute of
Technology, Pasadena, California,
December 1973 (updated July 1975
by J.J. Morgan).
Gordon, J.G. Assessment of the Impact
of Resource Recovery on the Environ-
ment. #MTR-8033. The MITRE Corpora-
tion, Metrek Division, McLean, Virginia,
December, 1978.
327
-------
ENVIRONMENTAL ASSESSMENTS OF WASTE TO FUEL PROCESSES
H. M. Freeman
Fuels Technology Branch
Energy Systems Environmental Control Division
Industrial Environmental Research Laboratory—Cincinnati
Office of Research and Development
U. S. Environmental Protection Agency
ABSTRACT
This paper discusses expected emissions from waste to energy conversion systems.
It provides an overview of waste to energy conversion technology, a discussion of
relevant Federal environmental standards, and a compilation of the results of en-
vironmental assessments of two systems; (1) an RDF combustion system at Ames, IA,
and (2) a partial oxidation pyrolysis unit at South Charleston, WV. Data is provided
for Criteria Pollutants and for various trace elements.
INTRODUCTION
For communities faced with costly mun-
icipal solid waste disposal problems and for
individuals within those communities faced
with constantly increasing costs for heat
and power, the idea of utilizing the waste
stream to supplement conventional energy
supplies is becoming increasingly attrac-
tive. The possibility of converting solid
waste, a potential environmental pollutant,
into a useful and valuable energy source is
appealing. There are on the scene today
several rather well developed systems for
converting wastes to energy, and within the
next five years it is expected that many
other presently undemonstrated technical
options will become available. In the next
decade, the conversion of waste to energy
could become accepted as a waste disposal
option in many of the nation's cities and
could represent a small but very real supp-
lement to conventional energy production
sources.
Conversion of solid waste to energy
offers a means of lessening one very real
environmental problem, solid waste disposal.
However, since conversion units involve fuel
combustion, they are also sources of air
pollution. These air pollution problems must
be recognized and addressed by anyone con-
sidering adopting one of these systems.
This paper discusses what types of air
pollution one might expect from these sys-
tems, the results of several recent en-
vironmental assessments, and what environ-
mental standards one must meet in operating
the plants.
OVERVIEW OF WASTE TO ENERGY CONVERSION
TECHNOLOGY
The various thermal processes for re-
covering energy from solid wastes can be
categorized into three major areas.
Mass Burning Incinerators with Heat
Recovery
The waste is fired in a processed or
unprocessed form in a boiler lined with
water filled tubes that transform the re-
leased heat into steam. Such plants are
presently in operation in. Nashville,
Saugus, MA., Chicago, and Braintree, MA.
Combined Firing Systems
The waste is first processed to re-
move non-combustibles and to reduce the
particles to common sizes, and then is
fired as a supplemental fuel with coal,
oil, or natural gas in a modified indus-
trial or utility boiler. The processed
refuse is sometimes referred to as a refuse
328
-------
derived fuel (RDF). Combined firing plants
are operating in Ames, IA, Chicago and Mil-
waukee. An earlier demonstration operation
was conducted by the Environmental Protec-
tion Agency (EPA) at a Union Electric plant
in St. Louis.
Pyrolysis Processes
The solid waste is converted thermo-
chemically in an oxygen starved environ-
ment into a liquid, gaseous, or solid fuel
product. These fuels are storable and trans-
portable products which can be utilized as
either primary or supplemental fuels. There
are presently no commercial scale pyrolysis
plants in the country. A demonstration
scale plant is being operated by the Union
Carbide Company in South Charleston, WV.
AIR POLLUTION
Waste to energy conversion systems pro-
duce as emissions particulate matter, sul-
fur oxides, nitrogen oxides, hydrogen chlo-
ride, hydrocarbons, carbon dioxide, carbon
monoxides and trace elements. Table 1 lists
some early data on emission factors for poll-
utants from controlled waste combustion
operations. Emission factors are used to
calculate expected emissions from controlled
facilities.
Of these pollutants, particulate matter
is the most significant in terms of present
environmental regulations. Particulate matter
is any material except uncombined water in
the form of solid or liquid in the gas stream.
It consists of.fly ash, dust, aerosols (mi-
croscopic sized particles) and mists. Emiss-
ion rates for particulate matter vary widely
depending on moisture and ash content of the
fuel, unit design, and combustion parameters.
However, uncontrolled rates of from 15 to 24
pounds of particulate per ton of refuse fired
are generally produced by waterwall inciner-
ators.
Gaseous emissions from waste-to-energy
conversion are not presently viewed as a sig-
nificant environmental problem. Refuse is
a low sulfur fuel, averaging only about 0.3
gercent sulfur (as compared to 1 to 3 percent
Eor coal). Consequently, S02 production is
minimal. Significant nitrogen oxides for-
mation usually occurs at temperatures of
above 2000°F. Most waterwall incinerators
operate at much lower temperatures. Although
combustion temperatures are higher in RDF co-
firing situations, the marginal contribution
of the RDF to NOX production appears to be
minimal. Unburnea hydrocarbons and carbon
monoxide are usually present in significant
amounts only if proper combustion is not
taking place. Proper flame turbulance,
ample combustion times and sufficient tern-'
peratures will reduce the quantities of CO
and hydrocarbons to negligible levels.
Chlorine is usually present in solid
wastes as inorganic chlorides in polymers.
It is estimated that chlorine is present in
municipal wastes in concentrations of up to
.5 percent by weight. The combustion of
this chlorine, expecially of the chlorine
contained in polymers, causes HC1 in the
stacks. One of the common findings in
assessments run at solid waste combustion
facilities has been the presence of HC1
in greater quantities than in the stacks
of coal-fired units.
Although data are extremely limited
concerning trace element emissions from
waste-to-energy conversion systems, sever-
al studies have indicated that refuse com-
bustion can produce airborne pollutants not
produced by conventional fuel combustion.
The EPA evaluation of the St. Louis RDF
plant indicated that such potentially haz-
ardous substances as beryllium, cadmium,
mercury, copper, and lead were present in
higher concentrations' in the emission
streams from coal plus RDF combustion than
they were in emissions from coal only com-
bustion. On the other hand, data from an-
other RDF plant at Ames, IA (discussed la-
ter) showed that trace element emissions
from coal + RDF combustion were generally
less than trace element emissions from coal
only combustion. A National Science Foun-
dation study found that in two metropolitan
areas, refuse incineration could account
for major portions of zinc, cadmium, and
antimony observed on airborne particles.
The study also suggested that refuse in-
cineration was a large source of another
toxic element, vapor phase mercury.
WATER POLLUTION? SOLID RESIDUALS
Since waste-to-energy systems usually
use water only for ash quenching and fa-
cility clean-up, they are not typically
seen as a major source of water pollution.
Two exceptions to this general statement
are RDF/coal cpfiring installations in
which water is used to sluice the ash, and
pyrolysis systems in which water is used as
a scrubbing medium for the product gas.
In the cofiring case, sluice water
from the St. Louis demonstration unit was
329
-------
found to exceed state standards for BOD,
dissolved oxygen and total dissolved solids
and to contain higher concentrations of
twelve other pollution parameters than a
sluice stream from a coal-only boiler.
In the pyrolysis case, the untreated
wastewater stream from these units is ex-
pected to have very high concentrations of
BOD, COD, phenols, and other organics and
will certainly require treatment. Such
treatment is included as part of the comm-
ercial-scale units presently being devel-
oped.
If a waste-to-energy system is dis-
charging into a municipal sewer system,
which is the case for most units in oper-
ation today, usually only minimal treatment
such as setting and p** adjustment is re-
quired. If the effluents were discharged in-
to public waterways, more extensive treat-
ment would be necessary.
Characterization of the solid residuals
from waste-to-energy conversion systems has
been minimal to date. These residuals are
produced as bottom ash and processing re-
jects, recovered fly ash and, for high temp-
erature slagging incineration and pyrolysis
units, an inert slag.
Trace elements such as beryllium, mer-
cury, cadmium, and lead have been reported
to be present in the fly ash from coal plus
refuse systems. The extent to which trace
elements are enriched in the fly ash is not
fully understood.
ENVIRONMENTAL STANDARDS
There are in existence Federal environ-
mental standards for major sources of air
pollution built or significantly modified
since December 1971. Sources built before
that time are covered by the environmental
regulations of the state in which they are
located. The Federal standards are titled
the New Stationary Source Performance Stan-
dards (NSSPS). The standards require that
a facility emit not more than a given amount
of certain pollutants based upon existing
state-of-the art for relevant control tech-
nology.
The pollutants covered by the NSSPS are
particulate, NO SO hydrocarbons, and
carbon monoxide.' TheSe substances are comm-
only referred to as the Criteria Pollutants.
The Criteria Pollutants of most interest to
designers and operators of waste-to-energy
conversion systems are particulate matter,
SO , and NO . The principle sources of hy-
drocarbons and carbon monoxide emissions are
automobiles and petroleum processing oper-
ations.
The NSSPS for mass burners addresses
only particulates. The NSSPS for fossil
fired steam generators, the combustion units
for RDF systems include particulate, SO.
and NO . It should be noted here that since
no newxboilers designed to burn RDF have
been put into operation since 1971, the EPA
has not yet said that the standards for
fossil fuel fired steam generators will
apply. However, the Clean Air Act requires
that the standards be revised periodically,
and the stated standards will be revised in
some way to incorporate RDF as a fuel. There
are no NSSPS for pyrolysis systems since
these systems do not have atmospheric emiss-
ions. However, the combustion of the re-
sulting gas could be treated under the "gas"
subcategory for the fossil fuel fired steam
generator. The existing standards for the
various systems are shown in Table 2.
To date no emission standards for trace
element emissions from waste to energy sys-
tems have been developed. None of the EPA
Standards for Hazardous Pollutants* apply
to either incinerators, pyrolysis units, or
fossil fuel fired boilers. With increasing
attention being given by the EPA to toxic
substances, more consideration may be given
in the future to trace element regulation.
ENVIRONMENTAL ASSESSMENTS
The Industrial Environmental Research
Laboratory (IERL), Cincinnati, a part of
EPA's R&D program has the responsibility
for determining emissions from waste-to-
energy conversion systems to insure that
control technology development accompanies
advances in conversion technology. Listed
below are brief discussions of the findings
of assessments at two facilities: a slag-
ging pyrolysis plant operated by the Union
Carbide Company, at South Charleston, WV;
and a utility boiler in Ames, IA in which
RDF is used as a fuel. The latter project
was conducted in cooperation with the U. S.
Department of Energy.
THE UNION CARBIDE PUROX PROCESS
Process Description
Since 1975 the Union Carbide Company
has operated a demonstration scale pyrolysis
330
-------
plant (200 tons of solid waste per day) at
South Charleston, West Virginia. The plant
converts the combustible fraction of solid
waste into a gas having a heat value of app-
roximately 1/3 that of natural gas, and con-
verts the non-combustible part of the solid
waste into a very dense inert slag.
After particle size reduction and ferr-
ous metal removal, the solid waste is in-
jected into the top of a vertical pyrolysis
reactor. Oxygen is injected into the bottom
of the reactor at ratio of about 20% by weight
of the incoming refuse. The oxygen reacts
with char formed from_the refuse to generate
temperatures of 2500° to 3QOO°F in the low-
er zone which converts the non-combustibles
into molten residue, and pyrolyzes the re-
fuse in the upper part of the reactor into
a gas. The gases are passed through a
scrubber and a precipitator prior to being
burned as a fuel gas. A diagram of the Purox
process is shown in Figure 1.
Environmental Assessment
In November of 1977 the EPA carried out
an environmental assessment of the Purox fa-
cility and of a package boiler brought to the
site to combust the product gas. The boiler,
equipped with special burners to burn the
Purox gas, had a rating of 30 x 10 Btu/hr.
Emissions from the combustion of Purox gas
were compared to emissions from burning
natural gas. The results are shown in Tables
3 and 4.
The conclusions of the air emissions
assessment were:
1. The Purox system successfully dem-
onstrates the production of a com-
bustible fuel gas from solid waste.
2. Of the Criteria Pollutants that re-
sult from combustion of the Purox
gas, only NO and particulate show
a significant increase at the out-
let of the boiler.
3. The NO emissions from the Purox
gas if they were present in the
larger size boiler covered by the
NSSPS would exceed the NO standard.
x
4. Particulate emissions are well be-
low the Federal standard.
5. SO. levels do not appear to be a
problem.
Samples on input river water and efflu-
ent scrubber water were taken each test day.
Additionally, during one day grab samples
were taken before and after the Unox water
treatment system which is part of the Purox
process. Results of the analyses of the
samples for general water quality parameters
were much higher in the scrubber effluent
than in the inlet water. The Unox system
did improve most of the general water quality
parameters with the exception of Total Sus-
pended Sediment and Dissolved Oxygen. How-
ever, neither TSS nor BOD would meet second-
ary treatment criteria of 30 mg/liter. Also,
even though the Unox system did decrease the
phenol level from about 90 mg/liter to .7
mg/liter this may not be sufficient to meet
stringent water quality criteria for phe-
nols that may be as low as .001 mg/liter.
At the findings and conclusions of the
assessment are included in a paper "Air
Emissions Assessment - Union Carbide Process,"
presented at the 71st Annual Meeting of the
Air Pollution Control Association, Houston,
Texas, June 26, 1978, and will be included
in an EPA report, "Environmental Assessment
of the Union Carbide Purox Process," in
process for publication by lERL-Ci.
AMES SOLID WASTE RECOVERY SYSTEM
Since 1975 the City of Ames, Iowa, has
operated a system for recovering materials
and energy from solid waste. The system
consists of a processing plant capable of
processing 150 tons/day of solid waste, a
storage facility, and an existing municipal
power plant. Solid waste is shredded and
processed to remove non—combustibles. The
remaining combustible fraction (RDF) is
fired in a stoker fired boiler at the aver-
age rate of 4-5 tons per hour (50% input
on a heat energy input basis). The two
boilers used to burn RDF are rated at 95,000
and 125,000 pounds of steam per hour. Mech-
anical dust collectors are the only type of
air pollution control equipment installed
on the stoker fired units.
During the summer of 1976 an environ-
mental assessment of the Ames plant was
carried out by investigators under an EPA
grant. During the assessment the boiler
load was set at 60, 80, or 100% and the
percent RDF varied from 0 to 50% of the
feed on a heat input basis. Sampling, phy-
sical characterization, and chemical element
analysis were performed on input fuels, ash,
and a compilation of the findings of the
331
-------
Raw
Refuse
Storage
Scale
Shredder
Magnetic
Separator
i
lagnetics)
CO
Off Gas
Fuel Gas
Shredded
Refuse
River Water
Pyrolysis
Reactor
Molten
Material
Water
Quench
Tank
\Recycle
Char/Oil \ Water
Oil
Scrubber
I Solid/
(Liquid
Separation
(System
/aste>
.Water>
ESP
Oil
\B
Fuel
Gas
Cooling Water
II
To Atmosphere
Heat
[Gas
ger
Booster
Blower
Flare |
Combusto
i
To
To Atmosphere
Test
Boiler
Note: "A" denotes flow during "normal" operation,
or as plant was intended to operate.
"B" denotes flow during testing, without
recycle.
Figure 1. Flow diagram for Purox^-vprocess.
-------
effect on stack pollutants of using 20% and
50% RDF is shown in Table 6.
The effects on selected Criterial Poll-
utants of firing RDF as a fuel are shown in
Table 7.
The environmental assessment at Ames
found that the particulate emissions decreased
when RDF was added to the feed. This is con-
trary to earlier findings. It should be
noted, though, that the particular coal used
for these tests had an ash content of 20%
versus an ash content of 17% for the RDF.
NO and SO emissions both decreased as the
percent RD? was increased. Interestingly,
with the exception of lead and copper, all
of the trace metals analyzed appeared in
lower concentrations in the fly ash result-
ing from the coal plus RDF than in the coal
only fly ash.
The complete results of the environmen-
tal assessment at Ames is contained in the
report "Evaluation of the Ames Solid Waste
Recovery System" - EPA Grant #80393, which
is now under review for publication by EPA's
Industrial Environmental Research Laboratory,
Cincinnati.
SUMMARY
While waste to energy conversion offers
an attractive alternative to traditional
solid waste disposal as a means of managing
wastes, there are still environmental ques-
tions which must be addressed. This paper
has discussed some of these questions. The
EPA's assessment programs such as the ones
discussed herein will provide answers to
these questions and insure that waste to
energy conversion will indeed fulfill its
promise as a positive approach to environ-
mental improvement.
*EPA's Standards for Hazardous Pollutants
apply to specific pollutants which have been
found to present clear threats to the public
health. The present list includes asbestos,
benzene, beryllium, mercury, and vinyl chlor-
ide.
333
-------
TABLE 1: POUNDS OF POLLUTANTS DISCHARGED PER TON OF WASTE COMBUSTED
Particulates
Sulfur oxides
Carbon monoxide
Hydrocarbons
Nitrogen oxides
Hydrogen chloride (est.)
Pounds
1.5
1.5
35.0
1.5
2.0
0.6
Source: Compilation of Air Pollutant Emission Factors
U. S. EPA, Office of Air Programs, AP-42, 1971
TABLE 2: STANDARDS OF PERFORMANCE FOR NEW STATIONARY SOURCES
Source Category
Subpart D - Fossil-Fuel
Fired Steam Generators
Subpart E - Incinerators
Affected
Facility
Coal, coal/wood
residue fired boi-
lers >250 million
Btu/hr
Gas, gas /wood
residue fired boi-
lers >250 million
Btu/hr
Incinerators
>50 tons/day
Pollutant
Particulate
Opacity
SO,
NO
X
Particulate
Opacity
SO,
NO
X
Particulate
Emission Level
0.10 lb/106 Btu
20%: 40% 2 min/hr
1.2 lb/10? Btu
0.7 lb/10 Btu
0.10 lb/106 Btu
20%: 40% 2 min/hr
0.20 lb/106 Btu
0.08 gr/dscf cor-
rected to 12% C02
334
-------
TABLE 3: UNION CARBIDE PUROX ENVIRONMENTAL ASSESSMENT
Boiler Flue Gas Particulate and Opacity Data
Fuel Type
Natural Gas
Flue Gas Flow Rate (d Nm /min) 93-108
3
Particulate Concentration (mg/d Nm ) a,/ 0.94-2.40
Opacity (%)b/ 5
Purox Gas
76-80
4.07-14.28
5
aj Includes only "front half" participate weights filter
catch plus acetone rinse of probe) from HVSS (high
volume sampling system) sampling train consistent
with EPA Method 5.
b_/ Average of two, one-hour observations.
TABLE 4: UNION CARBIDE PUROX ASSESSMENT
Boiler Flue Gas Constituents
Fuel Type
Flue Gas Constituents Natural Gas
HO (%) 17.9-19.7
C02 (%) 10.3-11.0
00 (%) 2.5-4.4
z
CO (ppm) 0-15
HC (ppm) 0-2.4
NO (ppm) 70-110
S0_(ppm) 1-1.4
Pur ox
Gas
13-14.3
20.5-20.7
1.1- 5.6
0-289
0
203-504
68-113
335
-------
TABLE 5. WATER QUALITY ANALYSES
Test pH
No. RW SW
A 7.7 5.2
5 7.6 5.0
6 7.2 4.9
7 7.5 4.8
Oil & Grease
(rag/I)
RW SW
<1 50
2.0 52
<1 77
<1 79
TSS^7
(mg/1)
RW SW
11.0 31.5
13.0 112
16.0 88.0
17.0 92.5
Purox E
Influent
COD
(mg/1)
RW SW
10 >4,080
2 4,200
1 4.200
3 4,950
crubber
& Effluent
COD
(mg/1)
RW SW
35 10,900
15 10,850
29 11,480
33 13,150
DO
(ppm)
RW SW
6.5 2.4
3.7 2.8
6.8 3.4
7.1 3.7
Phenols
(mg/1)
RW SW
0.3 118
0.4 118
0.1 128
0.4 131
Turbidity
(JTUs)-7
RW SW
35 280
35 200
<25 130
<25 110
River Water
Flowrate to Scrubber
(1/mln) (CPM)
Not read
337 89
329 87
333 88
W
OJ
Unox Grab Samples Flowrate through
Unox System
RW 7.5
Unox .
in ' 4.7
Unox . ,
5.4
29.8
2.6
12.5
63.0
60.0
-------
TABLE 6. UNCONTROLLED EMISSIONS OF SELECTED TRACE ELEMENTS IN FLY ASH
AT AMES, IOWA, RDF PLANT (AVERAGE VALUES, jig/g)
•••••^•^•••••••••••••••MMMM
Element
Aluminum
Chromium
Copper
Iron
Lead
Selenium
••••••••••••••••^•^MMi
Coal Only
86,149
183
153
86,914
6,733
46
^•^^•MMMMMBH^^BMMH^^MMMMMMMa
Coal Plus
20% RDF
65,137
179
472
22,133
31,684
87
*«»"^^W«MMMMMMMMNW«MMMNW«MMMMMMNMMMMMMMI1MMMM
Coal Plus
50% RDF
55,228
143
379
22,133
22,815
TABLE 7: UNCONTROLLED EMISSIONS OF CRITERIA
POLLUTANTS FROM THE AMES, IOWA, RDF PLANT
Particulate
mg/NnH
gr/dScf
Gases (ppm)
No
so2
Coal Only
9679
4.002
94
2080
Coal Plus
20% RDF
4851
2.006
83
1539
Coal Plus
50% RDF
7597
3.141
60
1494
337
-------
THE OCCURRENCE OF LEAD IN MUNICIPAL SOLID WASTE
David F. Lewis and Ada Salas
York Research Corporation
One Research Drive
Stamford, Connecticut 06906
ABSTRACT
The recycling of the waste products produced by our society has provided gome
exciting solutions to problems associated with solid waste disposal and energy production.
In this regard, a number of municipal solid waste (MSW) treatment systems have been
commercialized. The end product of these processes is a cellulose based refuse derived
fuel (RDF). Combustion of RDF in coal fired utilities can provide useful energy and it
is a particularly attractive fuel because of its low sulfur content. However, some
components of MSW may contain lead and other toxic metals and these may enter the RDF
process stream and become part of the fuel. Lead containing materials are present in
both the combustible and non-combustible fractions of MSW. Efficient methods are
available for the separation of these combustible and non-combustible fractions. However,
the use of lead based pigments in colored papers and printing inks and lead based
stabilizers in plastics can result in significant quantities of lead being intimately
bound to various combustible materials from which RDF is derived.
INTRODUCTION
Municipal solid waste (MWS) is gen-
erated in the United States at a rate of
approximately 4 Ib per person per day and
is growing at a rate of about 1.5% per
year. The management of this volume of
"trash" has become a major problem for
communities, particularly from the point
of view of disposal. Solutions to the
problem have to satisfy both economic and
environmental requirements. Thus we have
seen well worn alternatives such as
incineration and landfill come under close
scrutiny. The results have been reveal-
ing and damaging.
At the same time that these problems
have been widely appreciated, we have also
been brought face to face with a growing
energy problem. Not that this nation's
fuel resources are low or in danger of
being exhausted - they certainly aren't.
The problem is more one of economics
related to the rapidly increasing cost of
using environmentally acceptable fuels
(e.g. oil, gas) or using fuels in an
environmentally acceptable manner
(e.g. coal).
However, there have been active minds
among us who have seen that the problems
of "trash" and energy may have a common
solution, if not in total at least solu-
tion in part. This solution is the recov-
ery of material resources and energy from
municipal solid waste. Not that the ideas
or processes involved are novel, many of
them are not. Rather it is the sum of the
parts or the scale of operations that is
new. Thus municipal solid waste is
divided into combustible and non-combust-
ible fractions. From the non-combustible
fraction, metals and glass are recovered,
and from the combustible fraction is made
refuse derived fuel (RDF). This RDF is a
solid, cellulose based fuel that can be
conveniently used as a supplemental fuel
for coal fired utility boilers. Moreover
338
-------
it is attractive because it has a sulfur
content of less than 0.5% and in the
appropriate proportions could be co-fired
with high sulfur eastern coals without
violation of sulfur dioxide emission codes.
However, we must now look a little in-
to the future where we may face regulations
governing the emission of species such as
lead, mercury, cadmium, zinc, arsenic,
beryllium and other volatile toxic metals.
Does refuse derived fuel contain signifi-
cant amounts of any of these materials?
How much is there, how does it get there,
can it be removed, will the combustion of
that RDF violate emission requirements?
These are the questions, now is the time
to begin generating the answers.
The metal in which we are principally
interested is lead, but during the course
of the investigation on which we have
recently embarked we are also maintaining
a concern about nickel, arsenic and beryll-
ium. Our study is supported by the EPA and
is designed to determine how lead is accum-
ulated in RDF, how much and in what com-
pound forms is the lead present, what
happens to the lead when the RDF is com-
busted and how can the lead be removed from
the RDF and be prevented from reaching the
environment. For the purposes of the
following discussion we shall be primarily
concerned with the problems of accumula-
tion of lead in RDF. However, it is
pertinent to begin by considering the
effect of burning a fuel containing lead.
EMISSION OF LEAD
There is little doubt that the com-
bustion of materials containing volatile
metals, and lead in particular, does lead
to metal emission in the effluent gas
stream. Several studies have shown that
approximately 2/3 of the lead present in
coals can be expected to be released with
the fly ash and 1/3 remains in the bottom
ash. Of course, methods for the removal
of particulates from flue gases have been
well developed, but the volatile metals
present a particular problem. Various
workers have shown that lead emitted in
flue gas is not equally distributed among
the various particles in that gas stream.
Rather it is associated with the finest
particles. This phenomenum arises
because of the volatility of lead and lead
compounds. They are thus present in the
vapor phase in portions of the combustion
zone where the fly ash is formed. Only
when the flue gas cools down is the lead
desposited on the solid phase. Moreover;
because this deposition is a surface phen-
omenuin, the lead becomes concentrated on
those particles having the highest ratio
of surface area to mass, i.e. small parti-
cles. Data has shown that more than 75%
of lead emitted from a certain combustion
process was associated with particles less
than 2 micron in size despite the fact
that these particles accounted for less
than 10% of the total particles emitted.
There is a low efficiency of removal
of small particles, i.e. <1 micron, from
a flue gas stream. In addition, these
small particles are the very ones which
are inhaled and retained in the lungs.
Therefore the presence of lead in fuel may
present.a particular hazard to our health
and is a problem necessitating urgent
investigation.
Let us now turn our attention to RDF
and consider the pathways by which lead
arrives in the municipal solid waste from
which these refuse derived fuels are
derived.
COMPOSITION OF MUNICIPAL SOLID WASTE
The first prerequisite for this
investigation is to aquire an understand-
ing of the composition of MSW. One
quickly realizes that MSW is a heterogeneous
assortment of materials, and its composi-
tion will vary on a seasonal basis and
according to location in both the narrow
and wide sense. The values that are
indicated in Table 1 are derived from a
number of sources, but give a realistic
indication of MSW make-up on a dry weight
basis. It should be appreciated that MSW
is never received or handled dry and that
its moisture content could be in the range
5-30% depending on composition and climatic
conditions. Furthermore, the greater part
of the moisture will be associated with the
cellulose, the putrescible food waste and
yard waste portion of the combustible
fraction.
The combustible fraction of MSW is
composed of paper, plastics, wood, textiles,
putrescible food and yard waste, leather
and rubber. Although this accounts for
55-70% of MSW, a considerable portion does
not enter the fuel process stream of most
RDF manufacturing operations. This is a
339
-------
Table 1
COMPOSITION OF MUNICIPAL SOLID WASTE ON DRY WEIGHT BASIS
Product
Typical Content*, percent
Range of Content, percent
Paper
Glass
Fine glass, dirt, ceramics
Ferrous metal
Plastics
Putrescibles
Corrugated board
Wood
Fabrics
Aluminum
Leather and Rubber
Heavy non-ferrous metals
Miscellaneous
51.7
10.5
10.0
7.6
5.0
4.4
3.5
2.6
1.8
0.1
0.7
0.2
0.9
35
4
9
5
2
2
3
0.4
1
0.6
0.4
0.2
0.5
- 60
- 23
- 20
- 13
- 8
- 25
- 25
- 8
- 9
- 3
- 5
- 0.6
- 15
*Source: Waste Age, September 1978, p. 54. / Max. and min. values from literature data.
result of the physical phenomena that the
various separation processes employ, as well
as the physical characteristics of some of
the components of MSW. Thus rubber tires
are extremely difficult to handle and are
usually manually removed from the process
stream at an early stage. One of the
common separation techniques used in RDF
processing is air classification. This
procedure separates materials according to
their mass per unit surface area and there-
fore in most, but not all, of the MSW
resource recovery processes in operation,
the RDF is derived primarily from the
light combustible fraction, i.e. paper,
plastic and textiles. In these instances
it is a good rule of thumb to say that 2
tons of MSW will produce 1 ton of RDF.
LEAD CONTAINING MATERIALS
IN MUNICIPAL SOLID WASTE
Now that we have an appreciation for
the composition of MSW let us consider the
uses of lead and find out in what compon-
ents of MSW can lead be found.
Lead is used in the United States at
a rate of about 1.5 million tons per year.
Over the last 10 years the consumption of
lead has risen only about 20%. This growth
may be largely offset by the phase out
of lead containing gasoline additives as
they account for about 18% of lead used
Another 72% of lead is used in metal prod-
ucts such as ammunition, alloys, cable
sheathing, tubes and pipes, lead sheet and
foil, type metal, storage batteries, lead
plating, solder, and ballast. Lead is used
in the making of both white and colored
pigments, and this accounts for about 8%
of lead consumption. The remaining 2% is
spread over numerous products.
We now see from Table 2 how lead may
occur in each of the components of MSW that
we have identified earlier. For ease of
use some of these components of MSW have
beep sub-classified. There is also includ-
ed in Table 2 an indication of whether the
occurrence of lead in these items is common
or rare. We cannot be totally objective
about this designation, but when taken in
conjunction with the proportion of that
item in MSW it does allow one to spotlight
areas of concern.
The data contained in Table 2 reveals
some disturbing information - considerable
portions of the combustible fraction of
MSW contain lead. There is a lack of data
giving information as to the lead content
of all the various combustible and non-
combustible fractions of MSW and work in
this area is urgently required. However,
some investigators have determined that the
overall lead content of MSW may be in the
range 1000-5000 ppm and that lead may be
present at up to 350 ppm in the light
combustible fraction. These figures would
indicate that the non-combustible fraction
of MSW contains appreciable quantities of
lead. Nevertheless, because the derivation
of RDF from MSW involves separation of the
combustible and non-combustible fractions,
it is worthwhile to consider these
separation procedures.
340
-------
Table 2
OCCURRENCE OF LEAD IN MUNICIPAL SOLID WASTE
Fraction
Component/
Subcomponent
Form of Lead
Occurrence
Non-Combustible:
Metal:
Heavy Combustible;
Ferrous
Aluminum
Non-Ferrous
Glass:
Ceramics:
Putrescibles:
Food Waste
Yard Waste
Wood:
Leather:
Rubber:
Alloys
Solder
Galvanizing
Paints/Primers
Alloys
Cable Sheathing
Collapsible Tube
Foil and Sheet
Pipes
Solder
Batteries
Ballast
Crystal Glass
Glazes
Enamels
Hopefully none
Pesticides
Fungicides
Horticultural caulk
Paints/Primers
Stabilizers
Rare
Common,High
Rare
Common
Rare
Common,High
Common,High
Common,High
Common,High
Common,High
Common,High
Common,High
Common,High
Common,High
Common,High
Common,High
Hopefully
rare
Rare
Rare
Common
Common
Rare
Rare
Light Combustibles:
Paper;
Newsprint
Kraft Paper
Corrugated board
Paperboard Pigments
Magazines Pigments/Glazes
Colored wrapping paper Pigments
Plastics:
Fabrics:
Stabilizers/Pigments
Common,Low
Common,Low
Common,Low
Common,High
Common,High
Common, High
Common,High
Rare
341
-------
PROCESSING OF MUNICIPAL SOLID WASTE
Many schemes for the processing of
MSW have been devised and commercialized.
It is beyond the scope of this paper to
consider each of these processes in detail.
However,- since many of the processes in-
volve similar steps, the generalized scheme
depicted in Figure 1 will serve to illum-
inate several points.
The basic steps of the process are as
follows. MSW is received on the tipping
floor and oversize bulky waste, e.g.
refrigerators, mattresses, etc. are removed
by hand. The material is then sent to a
shredder where it is reduced to a 4"-6"
size and from there it goes to a screen
where undersize material is removed. The
undersize material is likely to be primar-
ily composed on non-combustibles such as
glass and metal although a certain amount
of food and yard wastes are also removed.
Oversized material is then fed to an air
classifier which is designed to pull off
the light components such as paper, card-
board, rags, plastic, etc. Air classifica-
tion may be performed in more than one
step, but the result should be a process
stream essentially free from metal, glass
and ceramics. A further screening of the
light process stream may then be performed
in order to remove dirt and powdered glass
that remains adhered to the light frac-
tion. The efficiency of this final dirt
removal process depends to a large extent
upon the moisture content of the process
stream, but if the material is dry, an
excellent separation may be accomplished.
The remaining process stream contains
the light combustible fraction of MSW and
is composed primarily of paper and plast-
ics with lesser amounts of rags, wood,
leather and rubber. This then is the
process stream that comprises the RDF or
the RDF precursor. According to the
particular RDF involved, there may be
further size reduction or chemical treat-
ment or compaction. With good separation
FIGURE 1
PROCESS STEPS FOR MSW HANDLING
342
-------
procedures, the amount of non-combustible
materials remaining in the light combust-
ible fraction should be considerably less
than 5% by weight and will be primarily
composed of metals in the form of paper
clips, staples and aluminum foil together
with ground glass, ceramics and dirt.
Since most of the lead in the non-combust-
ible fraction of MSW is likely to be in the
form of bulky items such as pipe, cable
sheathing and storage batteries, it is felt
that adherence to good separation procedr
ures will result in the removal of more than
95% of the lead contained in the non-com-
bustible fraction. However, these figures
should be subjected to experimental
verification.
LEAD IN THE COMBUSTIBLE FRACTION
OF MUNICIPAL SOLID WASTE
The literature contains data indicat-
ing that the light combustible fraction of
MSW contains up to 350 ppm of lead. This
is lead that is intimately associated with
paper and plastics and cannot be removed by
procedures required for the separation of
non-combustible components of MSW. Since
it is suspected that the removal of the
lead in the non-combustible materials is
highly efficient, we anticipate that the
lead used in the paper, printing and
plastic fabrication industries will be of
prime importance with respect to lead
content of RDF. The figure of 350 ppm lead
in RDF represents a value at least one
order of magnitude higher than that for
coal and oil. This gap would about double
when the lead content of the various fuels
is calculated on a Btu basis. Once again
it should be stated that the 350 ppm
figure must be subjected to detailed
investigation and verification.
The data presented in Table 3 has been
gathered to provide an appreciation of the
lead content of various components of the
combustible fraction of MSW.
It must be restated that the values for
the composition of the combustible fraction
of MSW represent those for a single sample
and the composition of a given sample may
vary considerably from those in the table.
However, it can be clearly seen that the
principal contributors to lead in RDF are
the pigments used in the printing of
colored papers and the stabilizers used in
plastics, and of these it it is the former
that is the major problem.
Table 3
COMPOSITION AND LEAD CONTENT
OF COMBUSTIBLE FRACTION OF MSW IN THE U.S.
Component Composition, percent
Corrugated board
Newspapers
Paperboard
Paper packaging
Office paper
Magaz ine s , book s
Tissue paper, towels
Paper plates, cups
Other
13.4
9.0
6.8
6.3
6.1
3.8
2.6
0.7
1.5
Lead component Lead Content, ppm
Pigments
Pigments
Pigments, glazes
6
9
40
5
10
- 50
- 45
- 500
- 10,000
10
- 3500
Yard Waste
Food Waste
Plastic
r~
Wood
Leather, rubber
Textiles
17.8
16.8
5.2
4.6
3.4
2.0
Pesticides,
fungicides
Stabilizers,
pigments
Paints, primers
Stabilizers
10 - 1000
343
-------
Lead chromates and oxides are widely
used in red, orange and yellow pigments
used for the printing of colored papers,
while metallic soaps such as lead naphthen-
ate are incorporated to accelerate drying
formulations applied as paper finishes
such as glazes and sizings. The plastics
industry uses lead sulfate and a variety of
organo lead compounds as additives to
polyethylene, polystyrene and polyvinyl-
chloride in order to promote stability.
It is these plastics that comprise the
bulk of the plastic fraction of MSW.
One further point worth noting con-
cerns the amount of lead in cardboard and
newsprint. Both of these materials might
commonly be fabricated by using recycled
paper. However, as the recycled paper
will contain some color printed material,
it follows that cardboard and newsprint
derived from that recycled paper will be
contaminated with lead.
REMOVAL OF LEAD FROM REFUSE DERIVED FUEL
The combustible and non-combustible
fractions of MSW may be efficiently separ-
ated, and from the data that we have
gathered it seems likely that the major
contributors to lead in RDF are those
components of the combustible fraction that
have lead compounds intimately bound to
them.
It is certain that much more data is
needed concerning the lead content of MSW
and RDF. In particular we need to deter-
mine precisely how much of the lead in the
combustible fraction of MSW comes from the
residual non-combustible fraction and how
much is present as a result of lead inti-
mately bound to the combustible materials.
We also must determine whether the
combustion of RDF containing, let us say,
350 ppm lead is likely to lead to a
violation of future emission statutes.
Although we should not be.presumptuous,
let us consider the removal of lead from
the conbustible fraction of MSW, i.e. the
fraction from which RDF is derived.
Separation techniques might involve a
chemical and/or a physical process.
Chemical separation is not east to envision
without wetting of the RDF and the use of a
subsequent drying step. If one were to use
an aqueous based system the cost of drying
alone can be estimated as being considerably
in excess of $6 per ton of RDF. For a fuel
whose worth may be about $20 per ton, this
clearly represents a serious economic
drawback.
Physical separation would appear more
attractive. Indeed there are several
appropriate methods for separation of the
plastics from paper and these should be
given consideration. However, because of
the use of lead based pigments in colored
papers we may be faced with a problem that
requires separation of various types of
paper. Processes that could accomplish
this on a commercial scale are not visible
on the horizon.
The use of additives to convert the
lead in RDF to non-volatile form during
the combustion process is an interesting
concept, but again a viable system is not
presently known.
Perhaps the best way to remove lead
from the combustible fraction of MSW would
be to ban its use in pigments used for
color printing. This is unlikely to be
popular with those sections of industry
procuding lead based pigment, nor perhaps
with those people using them. However, a
precedent has been set by the voluntary
restriction of the use of lead based
pigments in childrens magazines and comics.
Perfectly good and usable lead free pig-
ments are available to the printing
industry. Manufacturers are unlikely to
assume responsibility for removing lead
from their products and they have a strong
case. However, in this respect, one can
see how astute planning for a national
energy policy can be beneficial to society
as a whole. Restricting the use of lead in
combustible materials likely to end up in
municipal solid waste may be a small seg-
ment of an overall energy plan, but if we
are to decide that our best interests are
to be served by recycling and the efficient
use of energy then let us resolve to remove
trouble at the root and not wait to be
confronted with a problem whose solution is
technically difficult and economically
unattractive.
Funding for this work was provided by U.S.
Environmental Protection Agency, Cincinnati,
Ohio 45268 under Contract No. 69-03-2742.
344
-------
START-UP AND OPERATION OF
THE LANDFILL GAS TREATMENT PLANT AT MOUNTAIN VIEW
M. J. BLANCHET
PACIFIC GAS AND ELECTRIC COMPANY
SAN FRANCISCO, CA 94106
Abstract
The Mountain View landfill gas treatment plant is a cooperative
project involving the City of Mountain View, the Environmental Protection
Agency, and the Pacific Gas and Electric Company. Designed to process
landfill gas at the rate of 1 MMSCFD,* the plant consists of a compression
step followed by adsorption of water, carbon dioxide, and other impurities
on a desiccant, activated carbon, and molecular sieves.
Since its completion in May, 1978, the plant has been in sporadic
operation, accumulating approximately 230 hours of run time. Following
the resolution of mechanical problems in the vacuum cylinder of the com-
pressor and a capacity shortfall due to moisture adsorption by the molec-
ular sieves, a 30-hour test run at design capacity of 1 MMSCFD was made.
During this run, treated gas composition averaged 98% methane.
Introduction
Generation of methane gas, a
by-product of municipal waste de-
composition in landfills, has been
viewed historically as a problem.
In some of the larger landfills in
the U. S., it has been necessary to
collect and flare the gas to pre-
vent its migration into adjacent
properties where it could form
explosive mixtures.
In California, the recovery of
landfill gas was initiated by the
Sanitation District of Los Angeles
County at its Parlos Verdes land-
fill in 1971. Eighteen wells were
constructed to prevent gas migration
into adjacent residential proper-
ties. The gas thus collected was
flared.
The rapid escalation of energy
prices in 1973 caused the potential
of landfill gas as a source of en-
ergy to be recognized, and a suc-
cessful test program to establish
steady-state gas composition and
withdrawal rate was developed for
the Palos Verdes site.
In October of 1973, NRG Nufuel,
now Reserve Synthetic Fuels, Inc.,
and the Sanitation District entered
into a contract to produce and
purify landfill gas at the Palos
Verdes site for a minimum of five
years. A facility to process about
2 MMSCFD of landfill gas was built
in the summer of 1975. (1) Since
January of 1977, this facility has
operated steadily. The pipeline
*1 MMSCFD = one million standard
cubic feet per day.
345
-------
quality gas thus produced is in-
jected into Southern California Gas
Company's distribution grid. Cur-
rently, Reserve has under construct-
ion, at the Operating Industries
landfill in Monterey Park, a plant
to process 8 MMSCFD of raw gas.
Also in the Los Angeles area,
the Department of Water and Power
of the City of Los Angeles, as an
outgrowth of its gas-migration
control program at the Sheldon-
Arietta Landfill, successfully
demonstrated the feasibility of
generating electricity from landfill
gas. An internal combustion engine-
generator set was started in April
1974 and ran until February 1975.(2)
Since then, the department has de-
signed and is currently building a
system to compress, dehydrate and
pipe the full production of the
Sheldon-Arietta site to its Valley
Steam power plant.
Other projects of interest in
Southern California are Southwestern
Portland Cement's low-btu facility
in Azusa, the Wilmington low-btu
facility currently undergoing shake-
down, and the Industry Hills methane
recovery facility which is presently
being started up.
PGandE's Activities
Encouraged by the success of
the programs at Palos Verdes and
Sheldon-Arietta, PGandE surveyed
32 landfills in the San Francisco
Bay Area as potential producers of
gas. Fourteen of these sites were
found to have the potential to pro-
duce a combined total of 21 MMSCFD
of 450 Btu/SCF* gas, equivalent to
1600 barrels of oil per day, or the
average gas requirements of about
41,000 homes. (3) The Mountain View
landfill, operated for the City of
Mountain View by Easley and Brassy
Corporation, showed the greatest
promise for an initial demonstration
project.
In the summer of 1974, PGandE,
the City of Mountain View, and the
Environmental Protection Agency (EPA-)
embarked on a program to investigate
the feasibility of recovery of meth-
ane from existing landfills using
the Mountain View site. The Mount-
ain View site was of interest to
EPA because, unlike the Southern
California projects, it is a shallow
landfill with an average depth of
40 feet as compared to about 150
feet for those in Southern California.
Thus, it is more representative of
fills throughout the U.S.
This program was completed late
in 1976 and reports detailing the
findings of this effort were pre-
pared by PGandE and the City of
Mountain View. (4,5) Based on the
evaluation contained in the PGandE
report, it was decided to proceed
with a demonstration project at
Mountain View to be jointly funded
by PGandE and EPA.
Process Description
A process flow diagram is shown
in Figure 1. Raw landfill gas at the
rate of 1 MMSCFD is extracted from
30 acres with eighteen wells and
compressed to 150 psig in two stages.
Inter-stage cooling with an air-
cooled condenser produces water
which is returned to the landfill at
the rate of 0.3 GPM. From the knock-
out pot where water/gas separation
takes place, saturated gas flows
upward in a cylindrical adsorber
where water vapor, carbon dioxide
and other impurities are adsorbed
on alumina gel, activated carbon,
and molecular sieves. Treatment is
done sequentially in three towers,
with one tower on stream while the
other two are-being regenerated.
Regeneration,involves blow-down to
atmospheric pressure followed by the
application of a vacuum. During a
portion of the blow-down cycle to
atmospheric pressure, gas flows to
the surge drum from which it is
*Btu/SCF = British Thermal Unit per
Standard Cubic Foot
346
-------
RAW IX.
GAS \s*~
1 MMSCFD
—5
«j
\<=.
L
A
f" " '
A | [ A | | A
ODD
S S S
i«m ocir « 000
150 PSIG ft ROD
r1— lYl It n n
1 ^ B B B
M FILTER VRy vRy VR>
| T* 7- KNOCKOUT POT 4> T^^
tJ < Jx T
1 "X. 1 TRFJVTED GAS l^ \
iii ^
H-
AIR COOLED CONDENSER
400 PSIG
r- r- 1
/
1 >VENT
I
1
U
R
G
E
D
R
U
M
v;y
I
COMPRESSOR
FIGURE I! SIMPLIFIED DIAGRAM OF MOUNTAIN VIEW PLANT
SLOWDOWN
-- VACUUM REGENERATION
-------
recycled to the suction of the com-
pressor. From the adsorber, treated
gas is compressed to 400 psig for
injection into PGandE's line 101
which serves the San Francisco
Peninsula. Control is accomplished
with a densitometer which monitors
treated gas specific gravity and ad-
justs cycle time accordingly.
Construction of the treatment
plant was completed in May, 1978.
Table I shows a breakdown of the
plant's costs. EPA's contribution
amounts to $270,000. Corresponding
to yearly revenue requirements of
$288,250 as is usually done in the
utility business, the cost of the
gas is $2.95 per MMBtu. Thus,
treated landfill gas is competitive
with imported Liquefied Natural
Gas and Synthetic Natural Gas from
coal.
Start-Up and Operation
Following completion of the
facilities in May, 1978, the plant
was started-up in early June.
Since then, it has operated sporad-
ically for about 230 hours. This
spotty performance can be attributed
to the following:
. Failure of a desiccant support
screen in one of the adsorbers
resulted in the introduction of
alumina powder in the vacuum
cylinder. The cylinder wall and
rod were scored. New rings,
packing and valves were installed
on the vacuum stage. The cylin-
der was honed. Also, a filter
was installed upstream from the
vacuum stage.
. Moisture adsorption by the mole-
cular sieves caused the plant to
fail to operate at design cap-
acity of 1 MMSCFD. The plant
design requires the adsorber
packing to separate carbon diox-
ide from the raw gas during a
design cycle of twenty minutes.
Carbon dioxide breakthrough was
registered after six to eight
minutes. It is not known when
and how moisture adsorption took
place. It might have happened
while the sieves were in storage
for about one year at PGandE's
Decoto yard or during the early
phases of start-up. In any event,
the presence of moisture and
other volatiles was revealed by
a drying test of sieve samples
from all three adsorbers which
averaged 11.7% volatiles by
weight. Late in November, the
adsorbers were regenerated with
heat generated in a temporary
gas-fired heater rigged up for
this purpose. On the average, each
tower was heated with hot methane
to about 600°F for about ten
hours. Approximately 55 gallons
of condensate per adsorber were
produced during the regeneration
cycle. Samples of the condensate
taken at various stages of the
cycle are being analyzed by West
Coast Technical Service, Inc.
using combined gas chromatography-
mass spectroscopy.
. Inadequacy of the vacuum system
to pull the design vacuum of 26"
Hg. The best that can be achieved
with the system at present is
about 20" Hg.
Four corrosometers installed
on the knockout pots and the air-
cooled condenser have registered
corrosion rates ranging from 15 to
30 mills per year. PGandE's engi-
neers fully expect corrosion to be
eliminated as a practical problem
with the injection of an inhibitor
in the raw gas.
Following heat regeneration of
the adsorbers, the plant was oper-
ated at design capacity of 1 MMSCFD
of raw gas for 30 hours. Treated
gas composition averaged 98% methane.
348
-------
TABLE I
LANDFILL GAS TREATMENT COSTS1}
Equipment Costs, $ Installed Cost, $
Adsorbers 245,000 410,000
Compression2^ 200,000 390,000
Analytical Equipment 10,000
Wells and Collection System 30,000
TOTAL INSTALLED COST 840,OOO6)
YEARLY COSTS $/YR
Maintenance 27,700
Manpower 30,000
Fixed Charges4) 208,300
Feedstock Costs,5) @ $0.16/MMBtu
and 139,000 MMBtu/yr 22,250
TOTAL REVENUE REQUIREMENTS 288,250
Energy Output, MMBtu/yr7) 97,650
Energy Cost, $/MMBtu 2-95
1) Basis: 1 MMSCFD of raw landfill gas with a heating value of 450 Btu/scf.
2) Based on injection pressure of 400 psig.
3) Maintenance = 2% of adsorbers' installed cost plus 5% of compressor's
installed cost.
4) Fixed charges = 24.8% of total, installed cost on the basis of a 10-year
life. They include: depreciation, interest, profit, federal and state
taxes, insurance, etc.
5) Reflects initial payment of $0.072/MSCF of raw gas to the City of
Mountain View.
6) Of this total, EPA's contribution amounts to $270,000.
7) Based on thermal efficiency of 70% and on-stream factor of 0.85
349
-------
On the basis of data collected
since June, 1978, PGandE is con-
fident that the plant will operate
as designed.
Future Plans
As agreed upon in the contract
signed by PGandE and the City of
Mountain View on November 26, 1975,
PGandE has informed the City of
Mountain View of its plan to in-
itiate the demonstration phase of
the project on January 2, 1979.
The demonstration phase is designed
to establish the technical and
economic viability of the landfill
gas recovery concept. It will last
twelve months, unless both parties
agree to extend it. Following
completion of the demonstration
phase, the plant, if successful, will
be expanded to treat the full pro-
duction of the landfill, or about
5 MMSCFD.
Specifically, the demonstration
phase will involve the following:
. An evaluation of corrosion in
the process equipment and of
ways to minimize it.
. Comprehensive chemical and bio-
chemical analyses of all gas and
condensate streams.
. An analysis of the impact of air
intrusion on the processing and
utilization of gas recovered at
Mountain View.
. Production on a continuous basis
of treated gas that meets design
specifications (less than 3% CC>2,
7 Ibs. of H20 per MMSCF of treat-
ed gas, less than 4 ppm I^S, etc)
. Confirmation of landfill gas re-
covery economics.
PGandE, in collaboration with
Dynatech R&D of Cambridge, Massachu-
setts, has submitted a proposal to
the Department of Energy (DOE) that
encompasses the program for the
demonstration phase. Negotiations
with DOE are expected to take place
soon.
Racognition and Contributions
I am indebted to the following:
1. Messrs. R. Holden and R. Head-
rick of PGandE's Gas System
Design Department.
2. Mr. M. R. Lee of Gas Utilization
Department.
3. Messrs. E. C. Remedies, R. F.
Goldstein, and R. E. Hodgen of
Gas Resources Department.
4. Messrs. A. Garrissere and R.
Savullo of PGandE's San Jose
Division.
5.
6.
7.
Messrs. J. R. Bean and M. D.
Orton of Gas Distribution
Department.
Mr. Ray Boyle of General Con-
struction - Gas.
Mr. W. Culver of the Department
of Engineering Research.
References
1. "Final Environmental Impact Re-
port for NRG Nufuel Go's land-
fill gas processing system,
City of Rolling Hills Estates"
by VTN Consolidated, Inc.,
January 1975.
2. "Methane Recovery Demonstration
Project - Engine Generator Set
Operation Report" by the L.A.
Department of Water and Power,
October 1975.
3. "Methane from Landfills - Survey
of Existing Bay Area Sites" by
J. W. Van Zee for PGandE,
August 1974.
4. "Treatment and Utilization of
Landfill Gas - Mountain View
Feasibility Study"(SW-583)
prepared by PGandE for EPA,
1977.
350
-------
5. "Recovery of Landfill Gas at
Mountain View/Engineering Site"
Study prepared for EPA by the
City of Mountain View, EPA/530/
SW-587d, May 1977.
351
-------
ANAEROBIC DIGESTION OF AGRICULTURAL RESIDUES - A TECHNOLOGY ASSESSMENT*
Tom P. Abeles & David Elsworth
i e associates, inc.
3704 llth Ave. So.
Minneapolis, MN 55407
&
John P. Genereaux
Genereaux Social Science Consultants
370 Summit Ave.
St. Paul, MN 55102
ABSTRACT
Field evaluation of a full scale farm digester was carried out for over a year to
ascertain the engineering design problems needed to bring farm scale systems to commer-
cialization. This work was enhanced by laboratory analysis of a variety of potential
feedstocks including municipal solid waste. On-site inspection of a number of pilot
plants coupled with a literature investigation, economic analysis and farmer survey indi-
cates that these systems are viable on gas-only economics for farm systems as small as
100 dairy cows or the equivalent and perhaps as small as 32 dairy units provided careful
planning is carried out and full use is made of the gas. Financial incentives and
government support would not only enhance the viability but would probably accelerate
adoption of the technology.
Several areas of research and development are in need of attention to assure the
success of these systems and to provide opportunities for better yields and a wider
choice of feedstocks. Simple, low-cost pretreatment processes are not yet available to
allow a variety of organic residues from municiple solid waste to crop residues to be
used effectively and to allow a greater fraction of the organic material to be converted
to energy. More effective tank designs coupled with improved anaerobic populations are
needed. Careful attention must be paid, also, to uses of effluent for refeed, fertilizer
or innovative alternatives such as greenhouses, algal culture and aquaculture.
Very little work has been done on potential social and environmental impacts of
either small scale or large commercial systems for use of organic residues. Potential
changes in farming operations, energy use patterns and animal feed options are in need
of examination. Conversion of methane from organics to methanol may prove viable in the
near term. This may favorably impact the methanol/ethanol/fuels issue now prominent.
INTRODUCTION The process is exothermic as evidenced by
rise in temperature in landfills during
The anaerobic process for conversion methane production; but gas production is
of organic materials to methane and carbon enhanced by the addition of heat which
dioxide is found in nature at ambient reduces the amount of time for conversion
temperatures at the bottom of lakes or at of the material to gas.
37°C in the digestive systems of cattle.
*Work supported in part under USEPA Contract #R-804-457-010 and the Office of
Technology Assessment (U.S. Congress) Contract #OTA-C-78-357.
352
-------
This process has been used for years
in the western hemisphere as a means for
stabilization of sewage prior to disposal.
In the Far East and India, the need for
energy and fertilizer has generated the
development of low cost, simple digesters
for production of gas for use on farms and
in rural homes. With the growing concern
over dwindling energy resources, research,
owner-built digesters and pilot plants
have been established in the U.S. and
abroad to produce energy using farm resi-
dues and organic wastes. The key to the
technology has been in the development of
systems which is amenable to western farm
operations. This means simple systems
without the eastern labor intensiveness
and mechanical systems with the reliability
but not the inherent cost of sewage treat-
ment plants. Economics has to be shown
using the present level of technology since
there is no clear evidence that any re-
search will achieve an order of magnitude
breakthrough in gas production or economics
in the near future.
In the case of cattle and hog wastes,
part of the economics is achieved through
the development of flexible membrane tanks
which use a no-mix plug-flow design and
which can be easily installed by an owner-
operator. This work shows that a key to
economic success is the proper matching
of gas production with load needs or total
use of gas for power and heating where
applicable. Finally, some system cost
analyses are based on use of the effluent in
innovative ways such as algal culture,
aquaculture or animal refeed as well as
fertilizer.
When electricity is produced, signi-
ficant quantities of waste heat are
available. Unless this is used (cogenera-
tion), economics may become negative. The
addition of a greenhouse or other sink
serves as an alternative. The greenhouse
may produce a significant income which
could exceed that of the primary farm pro-
duction. This indicates that careful
attention must be paid to socio/economic
changes as well as bio/physical environ-
mental issues in the introduction of this
technology.
PROCESS DESIGN
Microbiology - In general, one may
consider the biological process as a three
phase operation; pretreatment or
hydrolysis, acidification and methanation.
For farm scale systems the pretreatment
or hydrolysis phase be it acid, basic or
enzymatic hydrolysis, is currently too
costly and premature in its development.
This limits the feedstock to those mater-
ials which are readily usable by the
anaerobic process. In this research the
feedstocks were animal wastes with some
laboratory work done on municiple solid
waste and algae. Other materials which
were considered included waterborn biomass
such as kelp, algae and water hyacinth.
Since the primary area for application was
the Upper Midwest, only algae was given
serious consideration.
The interest in algae was twofold.
First, it has been suggested that one pos-
sible microbiological rate limiting step
was the availability of hydrogen to the
menthanogenic bacteria. Secondly, algae
could serve as a means of concentrating
the effluent in the form of biomass and
could serve as either digester feedstock
or high protein feed. Research focused on
Dcenedesmus and obliques which can evolve
hydrogen during growth. The concept was
to feed the evolved hydrogen to the diges-
ter and enhance the methanogenic process.
The postulated nominal conversion mechan-
isms might be reduced to:
organic matter bacteria CH^ + C02
C0
4H
bacteria
CH
If one looks at the heats of combus-
tion on a molar basis, the hydrogen con-
tains about 17% more potential energy than
the methane which can be produced by the
above conversion pathway. What is unknown
at this point is whether the postulated
conversion is actually on a 4:1 basis or
whether the route is more complex. Time
and funds prevented the development of
this concept beyond the preliminary lab-
oratory phase. It is possible to develop
a solar collector which uses hydrogen pro-
ducing algae as the working fluid. Such
a unit has potentials for enhancing the
fermentation process by biomass, hydrogen
and thermal inputs.
Another area of potential interest is
the possibility that the microbial popu-
lations adapt to substrates. This implies
that not only can the process be carried
out in different biochemical regimes, but
that there is the possibility of obtaining
353
-------
a population which has a higher conversion
yield or efficiency. Work on swine wastes
(2) has shown that good gas yields can be
obtained at alkaline pH which, hithertofore
had been projected as toxic to the micro-
bial process. Recently, work done for the
Gas Research Institute (3) on kelp diges-
tion has shown that selected strains can
be obtained which produce the same volume
of gas per unit biomass and within the
same time at ambient as the optimally
operated mesophillic systems. They also
indicate that currently, kelp does not
yield readily to thermophillic digestion.
Because of the variance in data from
different publications, research was ini-
tiated using thirteen identical digester
units. The quantity was the minimum num-
ber which could have statistical signifi-
cance. Hoi stein dairy waste was used in
a solution of approximately 10% solids for
7 weeks to assure stabilization and then
the mixture was shifted to 63% manure:37%
municipal solid waste for 5 weeks. All
data was then analyzed using a standard
statistical package available through the
University of Wisconsin-Madison's Univac
1110. Tables 1 and 2 indicate the results
for the two feedstocks where ANOVA indi-
cates "analysis of variance". The ANOVA
shows that the effect of time on biogas
production is statistically significant
(p 0.00.), meaning that the digesters be-
have differently from day to day. However,
because the interaction of time x feedstock
-is also statistically significant (p 0.001),
the averages for each 14 day period should
not be combined. In other words, biogas
production in the feedstock 2 phase was
lower than in the comparable phase for
feedstock 1. Table 1 shows the grand mean
for feedstock 1 to be 583 mis/day; for
feedstock w it is 568 mis/day, a difference
of 15 mis/day. While the feedstock , mixed
with the municipal solid waste did oroduce
less gas than the animal wastes alone
are indications that gas production
increased. The experiments
should be repeated with larger concentra-
tions of municipal solid waste and
attempts should be made to select a
favorable microbial population. The
work for the Gas Research Institute focused
on developing specific populations from
decaying kelp on the ocean floor. For
municipal solid waste, one may look toward
species indigenous to landfills. Another
source might be detritous in tropical areas
which may have a predilection toward ligno-
cellulosic materials. Methane production
has been observed in the wood of living
trees which are suffering from decay. This
offers another alternative source of micro-
Dial populations.
Systems Engineering - Based on the
microbiology, a number of systems designs
have been proposed from simple one tank,
unmixed units to multiple tanks which
separate the acid and methanoqenic phases
and are completely mixed. For farm
scale systems, the current state of the
art suggest that a single tank which is
maintained in the mesophillic region (37°C)
is the best design. Work at Cornell (3)
and full scale commercial systems (4)
indicate that simple plug flow units oper-
ating with diluted or undiluted wastes,
maintained at mesophillic temperatures
without mixing,appear to be the most appro-
priate for cattle waste. Smith (5) has
indicated that the energy for mixing does
not seem to warrant the extra cost in terms
of enhanced digestion,either in shortened
retention times or ease of operation. This
is supported by the Cornell studies.
In this study, four systems were in-
vestigated, an intermittent mix pilot plant
Feedstock
1
Table 1: A Comparison of Means, by Feedstock,-Over Time
TIME*
8
10 11 12 13 14
535 643 549 627 641 518 612 507 593 639 524 581 623 569 583
494 572 589 509 582 615 489 533 512 627 534 625 640 638 568
515 608 569 568 612 567 550 520 553633528603632603576"
*Time 1 in this table corresponds to day 34 for feedstock 1, day 69 for feedstock 2;
time 14 corresponds to day 47 and 82, respectively.
354
-------
Table 2: Average Biogas Composition For 13 Digesters
Day No.
CH4/C02 %CH4
%C02
%02
2
9
16
23
30
37
44
51
58
65
72
79
x, 9-79
x, 9-44*
x, 51-79#
2.62
1.45
1.41
1.38
1.21
1.27
1.31
1.13
1.15
1.26
1.34
1.28
1.29
1.34
1.23
39.5
54.9
55.4
54.5
52.4
53.8
53.5
50.5
50.5
52.7
54.3
53.7
53.3
54.1
52.3
15.1
37.8
39.3
39.5
43.4
42.5
40.9
44.5
44.1
41.7
40.4
41.9
41.5
40.6
42.5
6.44
4.32
5.15
3.63
3.07
4.69
4.18
4.56
4.65
4.31
3.46
4.39
4.55
4.23
1.04
1.04
0.94
0.72
0.67
0.92
0.83
0.90
0.95
0.94
0.91
0.90
0.89
0.90
*Represents the average for the period covering steady-state
operation with feedstock 1.
^Represents the average for the period of operation covering
feedstock 2.
Table 3: Comparison of Biogas Production Efficiency
Feedstock 1 Feedstock 2
measured biogas production
mis of biogas/ml influent
%TS in influent
%VS of TS in influent
%VS destruction
ft3 biogas/lb TS«
(m3/Kg TSA)
ft3 biogas/lb VSA
(m3/Kg VSA)
ft3 biogas/lb VSn
(m3/Kg VSn)
11.7*
7.6#
83. 0#
24. 0#
3.79
0.236
4.56
0.285
18.2
1.14
11.4*
4.93
84.7
24.6
3.69
0.229
4.35
0.270
17.7
1.10
*Based on the average biogas production results from the ANOVA.
#No samples of influent or effluent were taken in the feedstock T phase; these
values represent averages from a previous study in which the same type of
influent was used; they also are consistent with results from other, similar
studies.
355
-------
at the University of Wisconsin-Green Bay,
a TOO horse digester constructed from
brewery tanks in Neenah, Wisconsin, a "bag"
digester in Custer, Michigan and a tank
digester in Rice Lake, Wisconsin. The
latter two systems were designed and con-
structed under the supervision of Agri-
cultural Energy Corporation, Luddington,
Michigan. Both are plug flow, mesophillic,
unmixed systems.
During the entire study, all systems
ran under design loading and had no pro-
blems maintaining thermal and biochemical
balance within the digestion unit. Rather,
all suffered severely from mechanical
problems and the lack of adequate engineering
The basic problems came from waste handling
during the feeding cycle of the operation.
The horse digester, which was largely owner
designed, did not take into account the
extensive amount of bedding needed for
show horses. Thus collection and mixing
of manure became such a severe problem
that the unit was abandoned because oper-
ational costs exceeded net benefits from
the gas production.
The pilot plant ran with minimal pro-
blems. The main problem was the reli-
ability of the pump chosen to feed the
tank. This item appeared to be the key
problem with all the systems including
others (outside of this investigation)
with which the authors are familiar. For
example, the Custer digester received manure
from an open feedlot which picked up straw
and animal hairs. Attempts at maceration
led to fouling of the incinerator with anim-
al hairs, while pumping was hindered by long
straw and occasional winter freezeovers
of the feedlot.
Freezing hampered the Rice Lake unit
which had a temporary feed system for
testing turkey wastes. Insufficient
hydraulic pressure to drive the pumps also
became an issue at Rice Lake.
The turkey wastes pointed out handling
problems which will also be prevalent in
broiler operations and feedlots which are
only cleaned periodically. Aged manure
does not mix, sell or pump easily and
materials from unpaved floors tend to
accumulate detritus such as rocks and
other extraneous materials. Additionally,
the aged materials loose volatile nutri-
ents such as nitrogen which lower gas pro-
duction. Also, rice hulls were used as
bedding for the turkeys. This material
does not digest readily as confirmed by
observation and analysis, which showed
over 60% inorganics and fibrous
material.
Figures 1 and 2 are schematics of the
Luddington and Rice Lake systems. The
Luddington unit is a fabric reinforced
"plastic" bag while the Rice Lake unit is
made of galvanized culvert. The bag unit
is low in cost, easy to install and rela-
tively trouble free. It can take cold
climates and is insulated with styrofoam
sheets which float on the surface of the
material. The Rice Lake unit is more
costly to manufacture and has only been in
operation for two years. It is too early
to tell how the metal will behave over
time. Zinc is known to be toxic to anaero-
bic species, but no evidence of inhibition
has been found. Also, corrosion is anti-
cipated to be minimal in the absence of
air as has been experienced in a number of
units with internal metal parts which were
kept oxygen free. It is difficult, though,
to exclude all air when pumping in new
effluent, so problems may arise in the
future.
Both the Custer and Rice Lake units
used the gas produced to drive an engine/
generator set. Though hydrogen sulfide in
the gas and influent did discolor the cop-
per piping in the mechanical sheds, routine
investigation of the internal parts of the
engine after biogas operation indicated
that the scrubbers acted effectively to
protect the units. Waste heat from the
generator units provided the heat for the
digester with excess heat vented. This is
a crucial factor in the economics but not
in systems operation.
SOCIO/ECONOMICS
Social Issues - Outside of technologi-
cal blocks, it became important to identify
those issues which were significant in
encouraging or blocking widespread adoption.
A survey of fanners was conducted in 1974
in Brown County, Wisconsin. A similar
study was conducted in 1978 in Barren
County, Wisconsin, and currently (December
1979), a pretest survey of farm enerqv
needs, uses and attitudes on waste man-
agment was conducted in Ottertail County,
Minnesota. The object of the first two
surveys was to test farmer attitudes to-
ward adoption of digesters, while the final
356
-------
SLURRY PUMPED INTO TANK.
EFFLUENT DISCHARGED INTO LAGOON
BIOCAS SIPHONED Off FOR USE
FLUIP RE-CYCLED TO INTAKE
GARTH COVBBIMO
6ULATIOM
BTAL TANK
• — HEATING PIPE FOR
TEMPERAUJRC CONTROl
BIOGAS RCCVCLEO IM1O
3LURRY PROVinES (HIVING
ACTION (OP-(ION)
DETAILS OF TANK CONSTRUCTION
EACH 6ECTIOM=3PIECE5
AIL IOIMTS CASKETEP^_OF|2'CORKUCATrD METAL,
Figure 1: Rice Lake Fermentation Tank
Figure 2: Earth Supported Flexible Membrane Digester
Operation similar to Figure 1
357
-------
full survey will be carried out in the
spring of 1979 to ascertain potential pro-
blems in carrying out energy conservation
and alternative energy system development
in midwest animal operations. Appendix A
summarizes the factors which appeared
crucial to the adoption of agricultural
innovation.
Several significant factors appeared
outside of the basic attitudinal informa-
tion. First, there is no extant infra-
structure for constructing, servicing and
supporting these systems within the stan-
dard agribusiness context. Next, there
exist few, if any, codes which outline
safety practices and operational procedures
that would allow such parties as banks and
insurance companies to deal with these
systems. Finally, it was uncertain at
the time of the research where government
tal support would come from since few
farms qualified under any of the EPA guide-
lines for pollution abatement. Since then,
DOE and USDA have started exploring funding
mechanisms,and several states have included
provisions within their solar legislative
package.
Economics - Technical feasibility was
never of primary concern; rather, technical
feasibility within the context of economic
viability has been the key issue.
The analysis of methane production
for farms show that it can be a complex
and capital intensive investment if tra-
ditional "turnkey" engineering approaches
are taken. On the other hand, the systems
in the Far East have been proven to be
simple and low cost. Research in this
country by Jewell and others (3) indicates
the possibility of simple, owner-built
units which can significantly change the
capitalization economics. A simple diges-
ter suitable for a 60 cow operation or the
equivalent could cost under $10,000,
including all pumps and controls necessary
for a modern U.S. farm operation.
When examining these non-traditionally
engineered systems, it is found that the
more favorable economics is predicated on
some changes either in farm operation or
life-style. This not only allows for better
energy utilization but may imply
indirect energy and economic benefits that
may outweigh the direct gain from either
the gas value or effluent enhancement. For
example, efficient digester operations are
essential. This implies complete use of
all the outputs such as heat, electricity
and effluent. An engine/generator pro-
duces extensive amounts of waste heat
which can be costly unless a "sink" such
as a greenhouse is found. The ramifica-
tions of a greenhouse for farm income
could be substantial. A quick estimate
shows that the waste heat from a 100 cow
digester-engine/generator can supply suffi-
cient heat for a greenhouse of the size
which could have an income larger than the
dairy producing the waste.
Since economics depends on efficient
gas utilization, farm operations may have
to be changed to accommodate the new
system. For example, few farms are cur-
rently on natural gas in the midwest. Many
of these utilize electricity for such fun'c-
tions as cooking, hot water heating, some
space heating and similar activities. The
cost is likely to affect the small systems
since any changes are reflected in the
system capital cost; yet, it is those
systems which must use the gas most effec-
tively to be economically viable.
The added cost of a generator on a
small system may be even more costly, even
with the ability to credit retail price of
electricity in the economic evaluation.
This may be reflected at several levels.
First, a generator which is sized to burn
all the biogas will probably not have its
output used 100%'of the time. Next, a
generator which operates at reduced loads
(like a base loaded power plant) may place
the farm in the position of being billed
by the utility for peak power utilization.
Finally, if the gas is burned 100% for
power, waste heat will probably be dumped
because it cannot be stored; whereas, a
baseloaded generator with gas left over
will have competition between unburned gas
and waste heat for thermal displacement.
Thus, on small farms, with limited
flexibility, extreme care in design is
necessary. Also, the small farm favors
direct gas use rather than conversion to
power. For larger systems, more flexi-
bility exists for integration and optimum
economics.
Tables 1 and 2 in Appendix B are
summaries of the economic feasibilities of
selected systems across the United States.
Average sized farming operations within
the appropriate regions were chosen using
358
-------
the best available data. For all systems
the following financial assumptions have
been made:
1. An investment tax credit of.7%
applies to the investment in the first
year.
2. An accelerated depreciation of
30% is taken in the first year. Deprecia-
tion of 10% follows for the next seven
years. Depreciation is set against a tax-
able income of $15,000,00, with four
exemptions for the state of Minnesota.
(Tax break of 35%.)
3. All tax credits are amortized over
the twenty-year life period of the equip-
ment.
4. Since the equipment is energy
productive, it is assumed that no property
taxes will be levied against it.
5. Discount rates of 9% over a
twenty-year system life are used.
6. Insurance costs are assumed to be
.5% of investment.
The net tax relief over the twenty-
year life of the equipment is $42.00 per
$1,000.00 of investment per year; insurance
is $5.00 per year.
For large systems (over 200,000 gal.)
which sell refeed, one technician, at
$15,000.00/year, is assured to be required
to operate the plant and accomplish the
more intricate day-to-day management at
this level. Large systems without refeed
are assumed to require one-third man-year
for operation.
It should be emphasized that this is
for mature systems and may have little
application to the prototypes which have
been developed for near future use.
As can be seen from the tables,
individual systems have good feasibility
under high refeed value on fertilizer
enrichment assumptions. Relatively few
are profitable, however, at direct energy
supply levels alone. Exceptions are limited
to large poultry operations, which have
high energy needs and thus provide their
own best market.
Interestingly, when waste heat is
added to direct heat (Table 2, Column 13),
most systems appear feasible.
Climatic factors appear important to
feasibility of direct energy use only in
turkey operations in Minnesota vs. Cali-
fornia and North Carolina. The same was
true of broilers, at least in a comparison
of Arkansas to Minnesota in the smaller
broiler operation shown.
Hog, dairy, and laying hen operations,
do not show significant climatic effects
on overall feasibility.
In fully examining the economic data,
one sees that feasibility is dependent on
utilization of the output from the
system whether it is effluent ot aas and
concomitant by-products. The best return
from the gas would be through cogeneration
provided that the waste heat from the
generator could be recaptured and used
efficiently. A large heat demand is found
in northern areas for such operations as
broilers and other fowl and swine rearing
activities. This can be seen from a com-
parison of the Arkansas and Minnesota data.
The parallels with solar feasibility in
this regard are quite good.
What has not been discussed previously
is the impact that new or other near term
technologies might have on the economic
feasibility. Fuel cells are one such
technology which could enhance overall
efficiency and economics. These units have
a higher power conversion efficiency which
does not require the large recovery and use
of waste heat. Also, unlike generators,
the fuel cells, maintain this conversion
efficiency under partial load,adding
further to economic efficiency.
Another point to note is that many of
the systems show a net return or positive
contribution prior to any tax or deprecia-
tion credits. Many systems in the tables
also show only minor potential profits or
losses. These latter systems could be
enhanced by slight changes in fuel
prices or tax incentives for renewable
energy systems. Thus, for these engineered
(or constructed-for-profit) systems eco-
nomic incentives would make digesters an
economically attrative option. If systems
were owner-built, the economics look even
more favorable.
An important point to re-emphasize,
359
-------
though, is the fact that many systems do
not show profitability on energy alone.
The exeptions are turkey and large boiler
operations which are extremely energy in-
tensive in their operation. Even these
show a potential, energy only, return
without tax credits. The real return
appears to be in the "enhanced"value of the
effluent for use as refeed or fertilizer.
In the U.S., a large portion of the
digester work has been done on dairy and
beef operations, and the data in the
tables shows some interesting facts for
these commercially installed systems. For
systems as small as 32 dairy cows in cold
regions, the loss is only about $300/year
with the current tax structure, while
"energy only economics" show a significantly
larger loss for a California herd size of
over 300 head. First, this implies that a
good end use of the gas is needed. Next,
it indicates that any factors which cut
the net costs to the small dairy farmer
or small farm operator such as owner-built
systems, tax incentives or fuel credits
could make these units a profitable
operation.
CONCLUSIONS
The process of anaerobic digestion is
both technologically and economically
feasible in the U.S. currently. The
basic problems are primarily socio/eco-
nomic in that an extant infrastructure is
not available to design, construct, or
maintain and service these systems. Some
minor engineering problems primarily in
the area of materials handling need to be
solved, but these are laraely a result
of a lack of extensive experience in the
U.S. with intermediate scale systems.
The systems will prove even more via-
ble as energy costs rise and as second
generation systems are developed. Some
of the potential needs for these new sys-
tems are: selected microbial populations
to enhance organic pretreatment and diges-
tion, more efficient conversion systems
for using the biogas, and a supporting infra-
structure including participation by
utilities in optimizing biogas use. Also,
work needs to be done on the use of the
effluent (1) by animals in such areas as
health effects and feed values, (2) for
aquaculture, (3) for algal culture (for use
of both the effluent and waste heat) and
(4) for fertilizer management.
These units could also be very impor-
tant in rural cooperative systems and
integrated farm and community energy
systems which provide energy and waste
management for rural communities or
groups of farms.
REFERENCES
1. Lapp, H. M., et al, Start Up of Pilot Scale Swine Manure Digesters for Methane Pro-
duction, in Managing Livestock Wastes, Proc. 3rd ISLW Conf., Urbana-Champaign,
Illinois, April, 21-24, 1975.
2. Chynoweth, D. P., et al, Biomethanation of Giant Brown Kelp, Macrocystis Pyrifera,
Institute of Gas Technology Symposiu, Clean Fuels from Biomass, Washington,
D.C., 1978.
3. Anaerobic Fermentation of Agricultural Residues, U.S. Department of Energy Rept. #HCP
Jewell, W. J., et al, Byconversion of Agricultural Wastes for Pollution Control
and Energy Conservation, Cornell University, Ithaca, NY, 1976 (TID-27164).
4. Abeles, T. P., et al. Energy and Economic Assessment of Anaerobic Digesters for
Rural Waste Management, OASIS 2000, Rice Lake, Wisconsin, 1978.
5. Smith, K. D., et al, Design and First Year Operation of a 50,000 Gallon Anaerobic
Digester at the State Honor Farm Dairy, Monroe, Washington, Ecotope Group,
Seattle, Washington, 1978.
360
-------
Appendix A:
(Factor)
Factors Underlying the Adoption of
Agricultural Innovations
Cost
1. Initial cost
2. Continuing (operating) costs
Returns
3. Rate of cost recovery
4. Monetary payoff
Profitability
5.
6.
Extent of economic advantage
over alternatives
Replacement status (condition)
of existing equipment
"Efficiency"
7. Saving of time
8. Saving of discomfort
Risk and Uncertainty
9.
10.
Regularity of reward
(Shows results every time)
Divisability for trial
(How easy to try; first,
on a small scale)
Congruence
11
12
13.
Association with main
enterprise (dairying)
Advantage (overall significance
of the practice for the entire
farm program
Pervaiveness (of consequences of
adoption; leads to other changes
or practices)
Communicability
14. Complexity (of understnading
and use)
15. Clarity of results (How clearly
do results show?)
Farm Characteristics
16. Farm size (total acreage,
dairy herd size)
17. Location of farmstead
(proximity to others; coop
unit possibilities)
(Relative Importance/Comment)
Not an important deterrent
Minor significance
Minor significance
Significant
Significant
Significant
Minor significance
Minor significance
Significant
Significant
Significant
Significant
Not an important deterrent
Minor significance
Minor significance
Significant
Significant
361
-------
Farm Financial Position
18. Capital (Curvilinear effect;
Proxies-estimated value of land,
buildings, equipment, and
livestock)
19. Current income (Proxies-gross
farm sales, milk receipts)
Farmer Characteristics
20. Age
21. Education
22. Tenure (owner/renter)
23. Years farmed
24. Attitudes and values
(proxies-affiliation with
farm organizations, activity
with social groups, contact
with "science" via reading
magazines and extension bulletins)
Regulatory and Legal
25. Pollution control (manure
handling rules and policy)
26. Safety requirements
27. Zoning and land use regulations
Institutional and Infrastructure
28. Availability of technical
support (engineering firms,
university extension)
29. Convenience of repair and
maintenance services
30. Supportive (and familiarity with
the innovation) farm finance
and insurance sector
31. Information sources (mass media-
radio, TV, newspapers and magazines;
friends and neighbors - mostly other
farmers; agricultural agencies -
univ. ex., SCS, farm credit agencies,
and the like; vendors and potential
suppliers)
Significant
Significant
Significant
Significant
Significant
Significant predicator of
adopting environmental measures
Significant
Significant
Precondition
Precondition
Condition for widespread
adoption
Condition for widespread
adoption
Condition for widespread
adoption
Significant for diffusion
The factors listed above were adopted, with modifications and expansion from:
Cancian, Frank, "Stratification and Risk-Taking: A theory Tested on Agricultural
Innovation", American SociologicalReview, 1976.
Fliegel, Frederick C. and Joseph Kivlin, "Farm Practice Attributes and Adoption
Rates", Social Forces, 1962.
_, Attributes of Innovations as Factors of Diffusion", American
Journal of Sociology, 1966.
362
-------
Kronus, Carol L. and J. C. van Es., "The Practice of Environmental Quality
Behanior", Journal of Environmental Education, 1976.
Mansfield, Edwin, The Economics of Technological Change. W. W. Norton, 1968.
Pampel, Fred, Jr., and J. C. van Es, "Environmental Quality and Issues of Adoption
Research", Rural Sociology.. 1977.
363
-------
Table 1
System Feasibility of Anaerobic Digestion on Single Livestock
and Poultry Farms in Leading Producer States, 1978.
10
a.
Livestock
State
+J vVi vW
Avg. No.
Turkeys (sold)
Minnesota
124,000
California
98,000
No. Carolina
89,000
Broilers (sold)
Minnesota
52,000m
198,000b
California
56,000m
1,377,000
Arkansas
63,200m
186,000
Swine (sold)
Iowa
1,600
Missouri
1,600
No. Carolina
1,600
Energy Benefitsa/Year
($1,000)
Benefits from By-Products ($l,000/Yr.)
Fertilizer Enrichment
KWh (retail)
G> 40/KWh
9.9
5.9
6.4
0.0
2.8
0.0
9.2
0.0
2.9
0.0
0.0
0.0
BTu
@ $5.4/BTu 10° 2.54/KWh
38.5
20.5
9.0
2.7
8.2
2.6
36.1
1.8
6.2
2.4
2.4
2.4
19.1
13.7
11.7
0.0
1.4
0.0
14.6
0.0
1.6
0.0
0.0
0.0
@ $10/dry ton
12.5
9.8
8.9
.5
2.0
.6
13.7
.6
1.9
1.1
1.1
1.1
Cost of App.
extra K
6.1
4.8
4.4
.2
.8
.2
5.8
.3
.8
.3
.3
.3
Refeed
@ $23/ dry ton
28.2
22.3
22.2
1.2
4.5
1.3
31.0
1.4
4.2
2.4
2.4
2.4
@ $90/dry ton
112.8
89.2
91.0
4.8
18.0
5.2
124.0
5.6
16.7
9.4
9.4
9.4
-------
Table 1 - Continued
Livestock
State
Avg. No.
Energy Benefits3/ Year
($1,000)
Benefits from By-Products ($l,000/Yr.)
Fertilizer Enrichment
KWh (retail)
BTu
@ 4
-------
Table 1 - Continued
Net System Costs &
($l,000/yr.)
Special Costs/Yr.
($1,000)
Sealed
Manure
Pit
9.2C
7.8°
7.4C
2!oc
l.lf
5.7f
1.2C
2.0
Residue
Lagoon
1 1^
.3e
.3e
:U
o.oe
.3e
0.0e
.3
.id
.id
0.0
Use for Waste Btufi
($1,000 @ $4/Btu 10 )
Greenhouse
Heat
9.69
11.29
13.89
0.0
0.0
.7"
9.19
K49
0.0
0.0
.5"
Manure
Drying F<
Refeed
9.69
11.29
13.8g
0.0
0.0
0.0
9.1g
0.0
1.49
0.0
0.0
0.0
Energy re-
placement
& Fertilizer
Enrichment
3r only
(Energy only)
43.9 (30)
18.7 (9)
40.0 (.1)
-1 (-1-5)
-1.4 (-3.4)
-.8 (-1.4)
41.7 (28)
-3.8 (-4.4)
.2 (-1.7)
-2.2 (-2.7)
-2.2 (-2.7)
-2.2 (-2.7)
Energy
Refeed
Value
$Z3
50
23
11
1
-9.4
.3
0
-2.4
-5.5
1.9
1.9
1.9
'&
$90
134
90
80
3.7
4.1
3.9
93
1.8
7.0
8.9
8.9
8.9
Benefits
Energy,
Fertilizer
& Foil Waste
Heat Use 70%
(energy &
heat only)
50.9
26.7
15.4
-
-.3
48.1
-3
1.2
_
-
2
(38)
(17)
(7)
—
(-9)
(35.6)
(-3.6)
(-.7)
__
--
(-1.3)
Energy,
Refeed &
40% Use of.
Waste Heat1
1
$23
51
47
17
—
.6
3.6
-2
-4.9
_-
--
2.7
$90
135
94
86
--
4.2
96.6
2.2
7.6
—
—
9.7
Special
Savings
($l,000/yr)
Manure
Disposal
6. 11
4.81
4.41'
is1
5$
:!'
1.83
1.8J
1.8j
-------
Table 1 - Continued
Special
($1,
Sealed
Manure
Pit
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Costs/Yr.
000)
Residue
Lagoon
0.0
0.0
1.1
0.0
1.2°
0.0
1.2°
0.0
3.2°
Net System Costs & Benefits
($l,000/yr.)
Use for Waste Btu Energy re- Energy,
($1,000 @ $4/Btu 106) placement Fertilizer
fertilizer tnergy & & roil waste
Greenhouse Manure Enrichment Refeed Heat Use 70%
Heat Drying For only Value (energy &
Refeed (Energy only) $23 $90 heat only)
0.0 0.0 -.7 (-1.3) .8 5.1
1-1 0.0 -.1 (-1.3) 2.8 11.3 .6 (-.6)
3.7 3.7 -.6 (-7.3) .4 45.1 2.0 (-4.7)
1.5 1.5 -4.3 (-5.8) -1.7 9.0 -3.3 (-4.8)
4.6 4.6 -.8 (-5.8) -9.5 24.8 2.4 (-2.6)
1.8 0.0 -4.4 (-6.0) -2.1 9.0 -3.1 (-4.7)
4.6 4.6 .2 (-4.8) -6.9 29.9 3.4 (-1.6)
1.8 0.0 -4.4 (-6.1) -1.5 10.4 -3.1 (-4.8)
11.4 11.4 14.3 (-2.5) 17.3 105.0 18.5 (5.5)
i. Manure disposed of as refeed; standard manure spreader used.
j. Limited to systems where manure is disposed of as refeed. Assumes
use of liquid manure spreader at rate of 8.2 hr./mo.
k. Assumes effluent is twice volume of influent.
1. Assumes greenhouse heat only.
m. Assumes refeeding takes place on site and requires no dewatering.
o. Assumes 3-month storage of residue + cost of pump.
p. Assumes 3-month storage of feedstock + cost of pump.
Energy,
Refeed &
40% Use of,
Waste
$23
3.2
1.9
-1.1
-7.6
-1.4
-5.1
-.8
21.9
Heat1
$90
11.7
46.6
9.6
26.7
9.7
31.7
11.1
109.6
Special
Savings
($l,000/yr)
Manure
Disposal
.65
1.3
3.3
.25
.81
.26
.80
.28
2.1
-------
10
o\
00
Table 2
Financial Feasibility of Anaerobic Digestion Systems
on Livestock and Poultry Farms in Leading Producer States
Livestock
State
Avg. No. Sold
per farm
(Inventory)
Turkeys
Minnesota
124,000
California
98,000
No. Carolina
89,000
Broilers
Minnesota
52,000
198,000
California
56,000
1,377,000
Arkansas
63,200
186,000
Swi ne
Iowa
1,600
Missouri
1,600
No. Carolina
1,600
Total
Investment
($1,000)
With
Dewatering
Equipment
345.5
290
286
—
166
—
308
—
136
Without Added
Dewatering Labor Cost
Equipment C$1 >000/Yr,}
227 15.0
191 15,0
187 15.0
31
106
30
190 15.0
45
84
42
42
42
Income Tax
($1,
Witfi
Dewatering
Equipment
14.5
12.1
12,0
7,0
12,9
5,7
Credit/Yr,
000)
Without
Dewatering
Equipment
9.5
8.0
7,9
1.3
4.5
1,3
8,0
1,9
3,5
1.8
1,8
1.8
Insurance
Cost/Yr,
C$1 ,000)
With. Without
Dewatering Dewatering
Equipment Equipment
1,7 1.1
1,5 1.0
1,4 ,9
,2
,8 ,5
.2
1,5 1,0
,2
,7 .4
,2
,2
.2
-------
Table 2 - Continued
CO
vo
Insurance
Livestock
State
Avg. No. Sold
per farm
(Inventory)
Dairy Cows
Wisconsin
(32)
New York
(64)
California
(337)
Laying Hens
Minnesota
(12,922)
(41 ,405)
Georgia
(13,500)
(40,800)
California
(14,300)
(105,200)
Total
Investment
($1 ,000)
With
Dewatering
Equipment
161
-_-
173
173
273
Without Added
Dewatering Labor Costs
Equipment ($l,000/Yr.)
26
26
101
56
90
56
90
56
164 15.0
Income Tax
($1,
With
Dewatering
Equipment
6.8
7,3
7.3
11,5
Credit/ Yr.
000)
Without
Dewatering
Equipment
1.1
1.1
4.2
2.4
3.8
2,4
3,8
2.4
6.9
Cost/Yr,
($1
With
Dewatering
Equipment
,8
.9
,9
1,4
,000)
Without
Dewatering
Equipment
.1
.1
.5
,3
,4
,3
,4
,3
.8
-------
Table 2 - Continued
Overall Costs and Returns ($l,OOQ/Yr,)
Livestock
State
Avg. No. Sold
per farm
(Inventory)
Turkeys
Minnesota
124,000
California
98,000
No. Carolina
89,000
Broilers
Minnesota
52,000
198,000
California
56,000
1,377,000
Arkansas
63,200
186,000
Swi ne
Iowa
1,600
Missouri
1,600
No. Carolina
1,600
Energy
Only
23.4
1.3
4.3
-.4
.6
-.3
30.9
-2.7
1-.6
-1
-1
-.6
Energy
9 $29/Ton
47.8
18.6
6.6
1.2
-3.2
6.1
-4.9
-.7
-.3
3.5
3.5
3.5
& Refeed
9 $90/Ton
131.8
85.3
75.6
4.8
10.3
9.7
8.1
3.5
12.2
10.5
10.5
10.5
Energy and
Fertilizer
Enrichment
@ $10/dry ton
value added
33.9
12.4
2.8
-.3
1.7
.3
32.5
-2.1
3.5
0
0
0
Energy, Fertilizer
Enrichment 70%
Waste Heat Use
All
40.9
16.2
4,9
,- — —
_ — —
.8
38,9
-1.3
4.5
3.8
Energy Only
28
6.5
-3.5
•-»•*•,
__ —
.2
30.1
-1.9
2.6
.3
Energy, Refeed and 40%
Use of Waste Heat
@ $23/dry ton @ $90/dry ton
45,3 129,3
19,8 87
9,8 79
<-*.*.
*-.—
1.7 5.3
1,3 91,7
-.3 2,1
.3 12.8
3.9 10.9
-------
Table 2 - Continued
U)
Overall Costs and Returns ($l,000/Yr.)
Livestock
State
Avg. No. Sold
per farm
(Inventory)
Dairy Cows
Wisconsin
(32)
New York
(64)
California
(337)
Laying Hens
Minnesota
(12,922)
(41,405)
Georgia
(13,500)
(40,800)
California
(14,300)
(105,200)
Energy
Only
-.3
-.3
-3.7
-3.9
-2.5
-4.0
-2.0
-4.1
-11.8
Energy & Refeed
@ $29/Ton
1.8
3.8
6.5
.3
-3.3
-.1
-.7
.5
12.4
@ $90/Ton
6.1
12.3
51.2
11.0
31.0
11.0
38.1
12.4
107.3
Energy and
Fertilizer
Enrichment
@ $10/dry ton
value added
.3
,9
2,9
-2,1
2,6
-2,2
3.8
-2.2
.8
Energy, Fertilizer
Enrichment 70%
Waste Heat Use
All
1.6
5.5
-1,3
5,6
,0
16.0
-1.1
.8
Energy Only
.4
-1.2
-2.8
.8
3.1
1.8
-2.8
-4.0
Energy,
Use
Refeed and 40%
of Waste Heat
@ $23/dry ton @ $90/dry ton
4.2
8,0
.9
-1,4
.6
1.1
1.2
17,0
127
52.7
11,6
32.7
11,7
39,9
13.1
104,1
-------
GAS RECOVERY FROM MSW - SEWAGE SLUDGE ANAEROBIC DIGESTERS
Joseph T. Swartzbaugh, Ph.D.
Catherine E. Jarvis*
Ralph B. Smith
Systems Technology Corporation
245 North Valley Road
Xenia, Ohio 45385
ABSTRACT
The anaerobic digestion of the light organic fraction of municipal solid waste was
studied on a full-scale 100,000 gallon digester. Based on earlier studies which showed
that normal wastewater treatment mixing was inadequate for digestion of MSW, two mixing
modes, a mechanical agitator, and a gas draft tube agitator were alternately employed and
evaluated at various MSW:sewage sludge ratios and loading rates, with total solids from 4
to 10 percent. The results of these studies indicate that materials handling problems
associated with the slurried refuse will restrict the use of this technology for energy
recovery.
INTRODUCTION
Anaerobic digestion has a long history
of use in treatment of sewage sludge. How-
ever, application of this technology to bio-
gasification of municipal solid waste (MSW)
presents several difficulties due to the
nature of the feed. The cleaned organic
fraction of MSW is composed largely of
cellulose. It's susceptibility to microbial
degradation is determined by accessibility
of the cellulose to enzymes, which requires
direct contact between enzyme and substrate.
Cellulose is largely water insoluble and is
composed of long chains of glucose units
linked by 31,4 bonds aggregated into bundles
of microfibrils. The microfibrils contain
amorphous regions which are rapidly hydro-
lyzed during biodegradation, and highly
ordered crystalline regions, which are of
restricted accessibility to enzymes and
limit the rate of degradation. Waste
cellulose materials contain 40 to 60 percent
cellulose plus hemicellulose and lignins.
The degree of lignification also affects the
breakdown of cellulosic materials by shield-
ing the cellulose substrate.
In addition to the chemical properties
of crystallinity and lignification of
cellulose, the physical behavior of the
material in an agitated slurry is also
relevant to anaerobic digestion of MSW.
Cellulose fibers tend to float on the sur-
face of the liquid and adhere to one
another on contact during mixing, forming
increasingly larger mats of fibrous scum.
One other component of municipal solid
waste-derived feeds for anaerobic digestion
which must be dealt with is grit. Even
though the light organic fraction of the
waste has had the majority of glass and
metal removed, a significant amount of
ground glass, sand, and grit remains in the
feed, the abrasive nature of which requires
consideration during design and operation.
The properties of municipal solid waste
are significantly different from sewage
sludge. Thus the direct application of
anaerobic digestion, as practiced in waste-
water treatment, is not feasible. This paper
presents the results of the first full-
scale studies of this technology and
discusses the technical implications for
future applications.
* Now at Pedco Environmental, Inc.;
Cincinnati, Ohio
372
-------
HISTORY OF THE PROJECT
is noted In the results section.
In 1974 through 1975, a laboratory
and full-scale study by EPA was conducted
to evaluate the feasibility of converting
municipal solid waste and sewage sludge
mixtures to methane anerobic
digestion. The light organic fraction of
hydropulped municipal solid waste was
combined with sewage sludge from a municipal
wastewater treatment plant. Two 55 gallon
laboratory digesters were operated at
several MSW to raw sludge ratios and vola-
tile solids loading rates. The results of
these studies indicated that the material
tends to accumulate as a fibrous mat on top
of the digester. Following the laboratory
studies, a pilot study was performed in the
35 foot (10.7 m) diameter, 100,000 gallon
(387,500 £) digester. After 75 days of
operation, the pilot scale study was term-
inated due to the formation of a 2 to 5
foot scum layer at the top of the digester.
Both of these studies indicated that one of
the most important operational parameters
for successful digestion of MSW to sludge
blends is adequate mixing of the contents.
CURRENT STUDY
The current study is being performed in
order to test and compare two methods of
mixing; a mechanical agitator mounted in the
center of the tank, and three gas mixing
units located inside the digester in an
equilateral triangle at approximately half
the radius of the vessel. A series of ex-
periments were conducted using two loading
rates and two blends of organic components
derived from MSW and sewage sludge according
to the outline shown in Table 1.
The digester vessel is equipped with a
Chemineer Model 4HTD10 10 hp mechanical
mixer with 54 inch (1.4 m) agitator blades,
an Aerohydraulics Model 3-12 "expanding
piston" type 5 hp gas mixer with three draft
tubes and a. Vaughan Model 300 "Scum Buster"
40 hp scum breaker pump. A 20 hp 1L12H-CDR
Moyno pump is used for feeding and releasing
effluent from the digester. During each
study, either the mechanical agitator or the
gas mixers were operated at all times except
during sampling. The scum breaker pump was
used during startup of a study to prevent
scum accumulation prior to data collection,
between studies to break up the scum layers,
and during studies if the scum build-up was
excessive. "Scum Buster" use during a study
Figure 1 shows the locations of the
mixing equipment and sampling ports. The
sampling ports are numbered from 1 to 6,
with samples taken from each port at three
levels (top, middle, and near bottom).
Samples are referenced with the port number
followed by the level at which it was taken,
thus solids at 1-3 are from sampling port 1,
level 3 (bottom level).
Monitoring
The objective of this study is to
evaluate mixing methods, therefore, the most
important monitoring parameter is solids
distribution within the digester. There are
6 sampling ports on the digester cover. The
digester contents are sampled bi-weekly at
three depths in each port in order to moni-
tor scum formation and solids distribution
within the vessel. The monitoring schedule
also includes daily gas production, feed
solids, and analyses of the effluent for pH,
alkalinity, volatile acids, total solids,
and volatile solids. Volatile acids and
alkalinities are performed, also bi-weekly,
on 6 of the samples from the ports. These
analyses provide a profile of the conditions
within the digester according to location.
In addition, a weekly gas chromatographic
analysis is performed, and nitrogen and
phosphorous analyses of the feeds are deter-
mined in order to calculate the supplemental
nutrient requirements of the digester.
RESULTS
First Study: Gas Mixing (3:1. 0.08. 4% TS)
The results of the first gas mixing
study using a 3:1 ratio of municipal solid
waste to sewage sludge at a loading rate of
0.08 pounds (1.25 gm/£) of volatile solids
per cubic foot per day are shown in Tables
2 and 3. The numbers shown in these figures
represent a time history of the values
obtained throughout the gas mixing study.
The data reflect conditions during use of
the scum breaker pump. It was intended that
the "Scum Buster" be used during the first
retention time of the study and not after-
ward during data collection. However, the
motor on the gas mixing system compressor
required replacement (due to a faulty motor),
and in the interest of continuing the pro-
ject, the data already collected was used.
373
-------
Tables 2 and 3 show that the solids in
the top layer of the digester are higher
than the solids in the middle or bottom of
the vessel. In addition to the actual
values being slightly higher, the standard
deviation of these values is also greater.
During the course of the study, it was not-
iced that there was a cohesive mass of
floating material moving within the top
layer of the digester. Throughout the
course of the study, this mass of material
was sampled at 1 to 3 ports per sampling,
resulting in higher average values for the
total solids in level 1. This is the reason
for the high mean values as well as the high
standard deviation of these numbers at level
1. Gas production during this study aver-
aged approximately 5,000 cubic feet (141,600
i) per day. The volatile acids to
alkalinity ratio averaged approximately .4
to .5. Gas composition was generally 70
percent methane and 25 percent carbon
dioxide. The feed blend contained 4.1 per-
cent total solids, the effluent average was
3.2 percent total solids. The calculated
volatile solids destruction was 48 percent.
Second Study: Mechanical Mixing
(3:1, 0.08, 4% TS)
Tables 4 and 5 summarize solids distri-
bution results for the first mechanical
mixing study operated at a 3:1 refuse:sludge
ratio of 4 percent total solids and 0.08
pounds VS/ft3/day loading rate. The total
solids in the beginning of the study were
relatively uniform with an average of 4.5
percent TS in the top level. After one week
the samples taken from port 1 showed high
solids of 10 percent. Within two weeks,
samples from two of the ports were high in
total solids, but the remaining ports were
still similar to the samples taken from
levels 2 and 3. Within three weeks, an
extensive 2-foot thick scum layer averaging
25 percent TS existed in the vessel. By
four weeks, the scum layer was 3 to 5 feet
thick at ports 5 and 6, closest to the wall,
and was dry and hard on the top. Scum at
port 4 was 1 foot thick and also drying.
Samples taken from ports 1 and 2 nearest the
mechanical agitator indicated a soft 1 foot
scum layer at those locations. As in the
first study, high total solids correspond to
high volatile solids which are the substrate
for the microorganisms, thus an increasingly
larger amount of food becomes tied up in the
floatable layer.
Visual inspection of the contents
through the sampling ports revealed further
information about the type and extent of
mixing within the digester. The top layer
was moving around the vessel (rotating about
the shaft of the center-mounted agitator).
The only place where the movement was tur-
bulent was near the shaft. The movement
was markedly slower near the walls of the
digester. According to the manufacturer of
the mechanical agitator, the type of mixing
expected for the power of the motor, size of
the blades, and length of the shaft installed
in the digester is generally axial flow with-
in the body of the fluid with a radial
component at the level of the blades, and
slow surface motion without uniform lifting
of solids to the top layer. This is the
type of mixing which the data and visual
observation indicate is the case. In order
to obtain uniform or near-uniform distribu-1
tion of solids, a much larger horsepower
motor would have to be used, 75 hp rather
than 10 hp, according to the manufacturers
recommendations. The smaller power motor
was chosen in order to be comparable to the
power of the gas mixer compressor (5 hp).
Use of the 75 hp motor would also have con-
sumed unduly large amounts of energy which is
not desirable in an energy-recovery process.
The mixing conditions consisted of
continual use of the mechanical agitator and
the recirculating pump with a 4 to 24 hours
per day use of the "Scum Buster" pump.
Initially, the study was run with continuous
use of the scum breaker pump, this was
decreased to 4 hours per day when results
were satisfactory in order to evaluate the
mechanical agitator, however, this was
gradually increased to continuous operation
again as the scum layer accumulated. The
last set of data entered in Figures 4 and 5
reflect conditions after using the "Scum
Buster" 24 hours per day.
Despite the accumulation of floatable
material within the vessel, the digester was
performing satisfactorily. Gas production
during this study averaged approximately
6,660 cubic feet per day. The volatile
acids to alkalinity ratio varied from 0.1
to 0.3. Gas composition was generally 60
percent methane and 35 percent carbon
dioxide. The feed blend averaged 4.62 per-
cent total solids. The calculated volatile
solids destruction was 49 percent. Gas
production averaged 12.9 ft3 per pound of
volatile solids destroyed. These parameters
indicate improved performance compared to
the first study. This is most likely
374
-------
attributable to the fact that the digester
had time to adapt to the feed and become
more stable.
Third Study; Gas Mixing
(3:1, 0.15. 4% TS Feed)
Tables 6 and 7 show solids distribution
during the gas mixing study at the higher
loading rate. A scum layer of 20 percent or
more solids existed in the digester within a
short period of time. The scum layer, con-
sisting mostly of fibrous material, became
1 foot thick after a few weeks, and was
approximately 2 feet thick at the end of two
months. By the end of the study, the top of
the scum was becoming dry, indicating little
if any, turnover of the material. The solids
data show a relatively well-mixed digester
below the scum layer, although the major
portion of the substrate was held in the
scum layer, as shown by the low total and
volatile solids in the rest of the digester.
The scum breaker pump was not used during
this study so that effectiveness of the gas
mixing system alone could be monitored.
The volatile acids to alkalinity ratio
during this study was 0.6 to 0.8, signifying
digester difficulty, probably due to the
high loading rate. However, the ratio
decreased with time indicating some adapta-
tion to the high loading. Gas production in
the beginning of the study increased to
8,000 ft3 per day, however, it was not
measured during the rest of the study due to
meter freezing problems, therefore, the gas
production per pound of volatile solids
destroyed is not available.
Fourth Study; Gas Mixing
(9:1, 0.08. 4%. TS Feed)
The first line of data in Table 8
indicates conditions in the digester before
the study actually began after use of the
scum breaker pump. The total and volatile
solids were uniform throughout the vessel
except at level 1 of port 5, which is
located along the wall. The next few sets
of data show increasingly higher solids at
level 1. The scum layer seemed to spread
from the wall to the center. Within a week,
the scum layer extended over most (if not
all) of the surface, and averaged 18 percent
total solids.
As in the other studies, high total
solids correspond to high volatile solids.
Although the study was not complete, early
data showed a volatile solids reduction of
47 percent. Volatile acid to alkalinity
ratios averaged 0.3 indicating a favorable
balance between the acid and methane form-
ing groups.
Fifth Study: Mechanical Mixing
(9:1. 0.08, 4% TS Feed)
Tables 10 and 11 indicate that an
exceedingly rapid build-up of solids at the
upper level occurs near the walls. A
gradual increase in solids with time does
occur at port 3, but throughout the period
of the study, ports 1 and 2 show no scum
build-up at the upper levels. Again, the
total solids in the well-mixed region show
an exceedingly low total solids concentra-
tion. Also, the volatile solids concentra-
tions appear to indicate that the solids in
the well-mixed region are low in volatiles
as compared with the solids in the entrapped
region. This again corroborates our belief
that the cellulosics are being rapidly
captured and removed from the actively-mixed
region, and thus are not available for
degradation.
Sixth Study; Mechanical Mixing
(9:1, 0.14, 7% TS Feed)
As indicated in Tables 12 and 13, the
total solids distribution shows that ports
1 and 2 remain clear. However, the solids
accumulation is such that ports 3, 4, and 5
each show an exceedingly high solids con-
centration at the upper levels, meaning that
the material in the scum layer is continually
drying. Toward the end of the study, the
material in the scum layer can be seen to be
sufficiently dry to support combustion. In
fact, at some points in the study, it was
impossible to break into the scum layer at
port 4 to withdraw a sample. As before, the
volatile solids concentration appears to be
higher in the unmixed region, while the
total solids concentration is exceedingly
low in the well-mixed portion of the
digester. Even though the feedstock is
entering the digester at 7 percent total
solids, the digester contents in the well-
mixed region appear to be around 1 percent
and rarely higher than 2 percent total
solids. What this means is that the solids
are immediately being removed from the well-
mixed area and immediately adding their bulk
to the scum layer.
375
-------
Seventh Study: Gas Mixing
(9:1. 0.14. 7% TS Feed)
Indications from the data in Tables 14
and 15 are that the gas mixing system does
allow an accumulation of solids throughout
the upper region of the digester. While the
solids in the center portion of the digester
do not achieve the levels seen in the outer
regions of the digester, still the samples
taken from the top level show an unmistak-
able accumulation of solids. Again, the
total solids in the lower regions of the
digester appear to be about 1 percent total
solids. It is interesting to note that the
scum layer had become so dense and so dry
that it was physically impossible to break
through the scum layer at ports 3 and 4 to
obtain samples from the lower regions.
Analysis of the materials drawn from ports
1, 2, and 5, however, indicate that the
digester contents are still well mixed in
those regions while the total solids again
appear to be fairly constant at about 1
percent.
Eighth Study; Gas Mixing
(9:1, 0.20, 10% TS Feed)
Again, the gas mixing system noticeably
allowed an increase in the solids accumula-
tion at the upper levels of the digester.
In spite of the fact that the solids content
in the feed had been increased to an average
of 10 percent, the solids content in the
levels 2 and 3 of the digester indicate the
self-limiting characteristic that we have
noticed earlier so that the total solids in
these regions tend to stay at about 1
percent.
Ninth Study: Mechanical Mixing
(0:1. 0.20, 10% TS Feed)
The mechanical agitating system did
again succeed in keeping the solids reduced
in the upper levels near the center of the
digester. It is noticeable that the total
solids concentration in the well-mixed
region is slightly higher than has been seen
before, undoubtedly due to the fact that the
total solids concentration in the feedstock
is so high. As before, the total solids
throughout the rest of the digester, though
appearing to be well-mixed, indicate that
the digester contents in the active region
are about 1 to 2 percent total solids.
CONCLUSIONS
Based on the results of this study, it
is our contention that it is impossible to
properly agitate a feedstock of higher than
approximately 5 percent total solids util-
izing agitation technology as it has been
commonly applied for wastewater treatment.
The feedstock of gritty, cellulosic material
typical of municipal solid waste is not
amenable to any form of low energy agitation.
The intent of anaerobic digestion of MSW is
energy recovery, and this particular tech-
nology does not result in a high energy
recovery efficiency. The best that can be
expected is 60 to 70 percent of the volatile
solids being converted into a recovered
energy product (CHu). Furthermore, the
process utilizes approximately 50 percent of
the carbon in that converted portion for
producing carbon dioxide, which is of no
value as an energy product. For this reason,
we feel that the process must be restricted
to low energy agitation.
Furthermore, the gritty content of the
material is exceedingly difficult to remove,
and probably can never be fully removed.
This does result in extreme operating pro-
blems for any equipment whose moving parts
are exposed to this material. The Moyno
pump utilized in this study was operated in
several configurations, finally with a
rubber stator and chrome plated rotor. Even
in this configuration, the operating life
time of the Moyno pump is on the order of 30
to 40 days. The "Scum Buster" Pump exper-
ienced severe damage throughout the study.
It was necessary to remove the "Scum Buster"
Pump three times for repair, and at the end
of the study, it was discovered that it had
effectively disintegrated within the
digester. Twice the cutter bar and bottom
plate assembly fell off the "Scum Buster"
and have not been retrieved.
The inspection of the mechanical
agitator system at the completion of the
studies indicated an exceedingly large
build-up of cellulosic material. This
material appears to have adhered to itself,
and formed long rope-like stringers, which
then became wound around the shaft of the
mechanical agitator. It appears that the
action is quite similar to the action of a
spinning wheel in that the fibrous material
tends to adhere to itself and gradually get
wrapped and twisted into threads which
rapidly accumulate. This accumulation is so
great that either regular removal of the
376
-------
material must be effected or the cellulosic
material will accumulate to such an extent
that the hub of the shaft would be as large
in diameter as the agitator blades them-
selves so that, in effect, no mixing at all
would occur. Furthermore, this accumulated
material causes a tremendous strain on the
mechanical agitator drive mechanism (see
Figure 2). Obviously it does not accumulate
in any well-balanced manner and so the
agitator drive mechanism is constantly
trying to turn an ever heavier and ever more
unbalanced shaft. This results in extreme
wear at the gas seal bearing, and makes it
exceedingly difficult to maintain a gas
tight digester. It is our contention, as a
result of this study, that bioconversion of
municipal solid waste utilizing the concept
of a well-mixed anaerobic digester is not a
feasible approach to energy recovery.
ACKNOWLEDGEMENTS
The authors wish to acknowledge the
contributions of Messrs. Robert J. Griffin,
James Thomas, Gerald Ellis, and Jeff McNiece
for the daily operation of the digester
facility.
377
-------
LIST OF TABLES
Table
Number
Description
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Summary of Operating Conditions
Total Solids Distribution for Study 1
Volatile Solids Distribution for Study 1
Total Solids Distribution for Study 2
Volatile Solids Distribution for Study 2
Total Solids Distribution for Study 3
Volatile Solids Distribution for Study 3
Total Solids Distribution for Study 4
Volatile Solids Distribution for Study 4
Total Solids Distribution for Study 5
Volatile Solids Distribution for Study 5
Total Solids Distribution for Study 6
Volatile Solids Distribution for Study 6
Total Solids Distribution for Study 7
Volatile Solids Distribution for Study 7
Total Solids Distribution for Study 8
Volatile Solids Distribution for Study 8
Total Solids Distribution for Study 9
Volatile Solids Distribution for Study 9
LIST OF FIGURES
Figure
Number
Description
Representation of the Digester Showing Location
of Mixing Equipment and Sampling Ports.
View of the Digester Interior After 245 Days of
Operation. (NOTE: The person shown is standing
atop a four-foot thick blanket of accumulated,
undigested feed.)
378
-------
SUMMARY OF OPERATING CONDITIONS
10
»j
vo
Mixing Mode
1.
2.
3.
4.
5.
6.
7.
8.
9.
Gas
Mechanical
Gas
Gas
Mechanical
Mechanical
Gas
Gas
Mechanical
Feed Ratio
(MSW: Sewage Sludge)
3:1
3:1
3:1
9:1
9:1
9:1
9:1
9:1
9:1
Loading Rate
(# Volatile Solids/
ftVday)
0.08
0.08
0.15
0.08
0.08
0.14
0.14
0.20
0.20
Total Solids
In Feed
(%)
4
4
4
4
4
7
7
10
10
-------
TOTAL SOLIDS DISTRIBUTION FOR STUDY 1
(Gas Mixing, 3:1, 0.08, 4% TS Feed)
Level
Top
Middle
Bottom
Date
8/01
8/04
8/08
8/11
8/15
8/18
8/22
8/25
8/01
8/04
8/08
8/11
8/15
8/18
8/22
8/25
8/01
8/04
8/08
8/11
8/15
8/18
8/22
8/25
5
(wall)
6.14
6.01
3.92
2.48
9.52
12.52
9.62
5.55
2.98
3.64
2.81
3.48
3.25
2.41
3.47
2.64
2.40
5.28
3.72
0.93
—
3.60
4.13
3.12
Percent
4
8.68
6.29
3.88
10.29
3.21
3.45
13.21
5.50
2.69
3.84
3.29
3.65
2.76
2.54
5.30
3.36
2.54
4.20
4.84
1.62
5.00
4.69
3.35
2.91
Total Solids by
Port Number
3
3.87
4.38
3.16
5.55
4.21
3.27
10.44
5.56
2.99
6.40
4.83
2.19
2.76
2.47
4.55
2.80
1.82
—
3.94
3.11
3.48
2.50
3.14
3.27
2
7.15
4.35
4.73
2.54
3.25
4.20
6.59
7.94
2.92
3.93
3.23
2.30
2.65
2.30
2.83
3.21
2.77
4.43
3.06
2.89
3.77
3.88
5.42
3.63
1
(center)
1.93
3.23
3.04
2.04
2.83
2.66
5.26
11.41
2.62
2.43
3.00
2.53
2.80
2.58
3.57
3.08
2.92
3.75
3.30
6.02
3.46
2.80
3.37
3.14
380
-------
VOLATILE SOLIDS DISTRIBUTION FOR STUDY 1
(Gas Mixing, 3:1, 0.08, 4% TS Feed)
Percent Volatile Solids by
Port Number
Level
Top
Middle
Bottom
Date
8/01
8/04
8/08
8/11
8/15
8/18
8/22
8/25
8/01
8/04
8/08
8/11
8/15
8/18
8/22
8/25
8/01
8/04
8/08
8/11
8/15
8/18
8/22
8/25
5
(wall)
63.1
57.4
53.3
53.7
61.0
62.1
62.3
54.3
76.2
50.0
67.7
51.9
51.6
51.5
56.2
48.8
75.0
56.2
61.7
66.1
—
50.5
51.0
49.3
4
64.1
55.0
51.5
62.5
52.8
56.7
61.0
56.6
60.4
53.5
53.1
47.6
53.2
52.4
57.3
47.4
58.6
54.6
57.1
60.0
52.2
58.4
54.2
53,4
3
64.0
49.1
52.6
58.4
60.6
56.2
54.9
56.7
59.1
54.5
55.4
44.4
50.9
52.5
52.6
52.2
79.6
—
53.0
48.0
50.0
50.6
58.9
58.5
2
65.7
47.4
59.8
57.9
51.9
55.9
61.7
59.3
61.6
53.2
52.2
48.5
48.0
52.5
48.6
47.8
57.7
48.8
42.4
51.4
52.1
52.0
49.3
51.6
1
(center)
83.7
54.7
48.7
54.0
56.0
52.0
55.6
56.2
56.6
56.5
48.4
51.0
52.1
51.4
54.7
50.0
69.4
47.4
47.5
50.5
51.6
50.0
50.5
52.9
381
-------
TOTAL SOLIDS DISTRIBUTION FOR STUDY 2
(Mechanical Mixing, 3:1, 0.08, 4% TS Feed)
Percent Total Solids by
Port Number
Level
Top
Middle
Bottom
Date
8/29
9/01
9/05
9/08
9/12
9/15
9/19
9/22
9/26
9/29
8/29
9/01
9/05
9/08
9/12
9/15
9/19
9/22
9/26
9/29
8/29
9/01
9/05
9/08
9/12
9/15
9/19
9/22
9/26
9/29
5
(wall)
4.80
8.07
3.79
4.70
4.61
16.02
27.13
26.56
27.31
21.49
3.80
3.25
4.85
3.79
4.52
3.75
3.41
3.55
4.23
3.16
3.46
3.34
4.61
3.83
4.58
3.82
3.12
3.22
3.03
2.96
4
3.94
3.37
4.81
4.14
11.35
13.77
24.62
23.05
18.54
25.03
4.57
4.26
3.45
3.87
3.70
3.34
3.10
3.51
2.86
3,20
3.56
6.27
3.85
3.23
3.29
3.39
3.91
3.26
3.16
3.30
3
5.62
3.98
5.63
4.01
5.03
11.87
24.20
14.26
19.06
31.05
3.44
3.39
4.12
3.82
3.87
3.44
2.36
3.46
3.15
3.73
3.18
— v
3.18
3.82
3.45
3.36
3.40
4. OP
3.00
3.13
2
3.77
6.51
5.00
4.17
3.62
8.74
20.49
23.19
11.36
22.09
3.91
3.32
5.18
3.78
3.33
3.83
2.92
3.97
3.46
3.25
3.62
3.64
3.60
3.62
3.70
5.17
3.91
3.27
2.26
3.38
1
(center)
4.22
3.95
10.11
3.57
3.51
7.99
27.67
22.51
7.77
19.94
2.00
1.95
5.56
3.85
3.67
3.42
3.17
3.74
2.94
3.31
3.61
3.49
3.26
4.72
3.69
3.23
3.40
3.23
3.95
3.26
382
-------
VOLATILE SOLIDS DISTRIBUTION FOR STUDY 2
(Mechanical Mixing, 3:1, 0.08, 4% TS Feed)
Level
Top
Middle
Bottom
Percent Volatile Solids by
Port Number
Date
8/29
9/01
9/05
9/08
9/12
9/15
9/19
9/22
9/26
9/29
8/29
9/01
9/05
9/08
9/12
9/15
9/19
9/22
9/26
9/29
8/29
9/01
9/05
9/08
9/12
9/15
9/19
9/22
9/26
9/29
5
(wall)
55.6
68.5
55.1
60.7
53.7
73.1
67.3
67.5
72.0
71.0
53.9
50.9
53.2
55.5
51.8
50.5
47.7
51.7
53.7
53.9
52.4
54.7
50.9
54.2
52.8
52.3
45.0
51.2
43.0
46.7
4
55.3
50.9
55.5
57.6
63.7
78.3
71.2
75.7
75.3
75.0
57.3
61.3
49.5
54.8
46.7
50.6
42.0
59.1
50.0
47.4
50.0
75.6
47.5
54.2
48.6
52.7
49.0
52.7
51.1
46.0
3
57.4
56.0
56.2
58.3
54.0
71.0
69.7
73.1
78.9
76.3
48.8
51.8
50.9
56.0
51.9
49.4
43.2
55.4
51.4
45.7
49.4
41.1
51.1
55.2
46.8
47.8
41.7
58.4
50.5
45.0
2
54.6
61.0
55.9
55.5
50.5
65.5
70.0
68.3
71.4
74.5
51.0
46.9
53.2
51.5
54.6
50.9
47.0
43.1
50.5
47.1
48.8
51.0
52.4
56.9
52.6
71.7
48.0
41.6
21.1
47.6
1
(center)
55.7
58.0
59.8
52.1
—
69.1
71.1
76.8
68.9
75.4
50.0
46.1
53.9
53.8
46.8
46.0
48.5
42.3
52.6
48.8
50.6
54.2
50.6
53.3
52.5
48.9
50.0
40.9
42.2
45.8
383
-------
TOTAL SOLIDS DISTRIBUTION FOR STUDY 3
(Gas Mixing, 3:1, 0.15, 4% TS Feed)
Percent
Total Solids by
Port Number
Level
Top
Middle
Bottom
Date
11/28
01/16
01/19
01/23
11/28
01/16
01/19
01/23
11/28
01/16
01/19
01/23
5
(wall)
5.4
22.8
22.1
19.7
3.9
4.5
1.6
1.7
3.9
7.3
1'.3
2.9
4
5.0
—
23.8
17.7
3.5
—
1.9
1.8
3.4
—
3.1
1.6
3
4.1
26.7
22.3
21.7
4.7
4.4
5.7
2.0
3.4
4.3
2.0
3.4
2
3.1
22.0
18.3
22.2
2.1
9.3
2.5
1.8
3.8
10.4
4.3
3.0
1
(center)
3.9
21.8
16.5
18.9
3.4
7.3
5.3
2.5
5.0
3.7
1.2
3.4
384
-------
VOLATILE SOLIDS DISTRIBUTION FOR STUDY 3
(Gas Mixing, 3:1, 0.15, 4% TS Feed)
Level
Date
Percent Volatile Solids by
Port Number
5
(wall)
4
3
2
1
(center)
Top
1/16
1/19
1/23
61.6
65.7
83.7
60.8
68.4
65.9
62.6
66.7
56.8
59.8
67.1
56.7
68.9
73.4
Middle
1/16
1/19
1/23
48.1
46.7
48.1
45.8
53.1
52.5
53.8
53.2
49.5
43.1
50.0
44.0
51.8
54.0
Bottom
1/16
1/19
1/23
48.1
44.4
44.2
50.0
51.1
49.6
45.9
43.9
49.5
49.5
48.2
43.3
46.7
385
-------
TOTAL SOLIDS DISTRIBUTION FOR STUDY 4
(Gas Mixing, 9:1, 0.08, 4% TS Feed)
Percent Total Solids by
Port Number
Level
Top
Middle
Bottom
Date
3/23
3/27
3/30
4/03
4/06
4/10
4/13
3/23
3/27
3/30
4/03
4/06
4/10
4/13
3/23
3/27
3/30
4/03
4/06
4/10
4/13
5
(wall)
21.5
19.0
15.2
18.9
18.5
15.1
12.6
2.5
1.7
1.6
0.9
1.3
0.9
0.6
3.2
1.7
1.5
1.1
1.0
0.5
0.8
4
1.3
17.8
22.7
17.7
13.8
16.1
13.4
1.9
1.4
1.6
0.9
1.1
0.9
0.5
1.9
1.6
1.5
0.9
0.9
2.4
0.5
3
2.0
3.6
14.8
20.1
21.2
15.0
14.6
1.9
1.7
1.3
1.1
1.1
0.4
0.7
1.9
1.8
1.3
1.3
1.2
0.9
0.8
2
2.0
2.3
15.6
14.2
17.6
20.0
16.5
1.9
1.5
1.3
1.5
1.0
0.8
0.7
2.4
2.0
1.6
1.4
1.0
1.7
1.9
1
(center)
1.9
18.6
17.1
19.4
18.1
15.2
16.6
1.8
1.8
1.7
1.7
1.1
0.5
0,6
2.2
3.3
1.8
2.3
1.2
0.5
i.i
386
-------
VOLATILE SOLIDS DISTRIBUTION FOR STUDY 4
(Gas Mixing, 9:1, 0.08, 4% TS Feed)
Percent Volatile Solids by
Port Number
Level
Top
Middle
Bottom
Date
3/23
3/27
3/30
4/03
4/06
4/10
4/13
3/23
3/27
3/30
4/03
4/06
4/10
4/13
3/23 .
3/27
3/30
4/03
4/06
4/10
4/13
5
(wall)
71.3
70.6
61.3
59.0
65.5
64.7
68.9
51.1
45.1
48.8
45.0
52.5
62.5
65.0
55.8
51.0
51.1
40.6
43.2
50.0
60.1
4
36.4
61.3
64.3
72.3
68.6
69.5
68.5
53.1
50.0
52.2
36.8
50.0
55.6
— —
53.6
52.8
57.1
42.3
37.0
72.2
64.7
3
53.6
59.7
79.3
71.5
68.3
65.0
68.7
52.9
45.9
57.9
34.5
40.7
70.0
72.7
54.7
41.7
61.8
35.3
46.7
47.1
60.9
2
56.0
58.9
67.8
74.0
70.8
79.0
79.5
51.9
41.9
50.0
42.4
46.2
53.3
63.0
63.8
44.2
50.0
37.5
41.4
52.3
61.3
1
(center)
53.1
71.1
76.2
78.2
—
73.3
78.6
51.2
45.7
48.2
45.7
42.9
53.9
66.7
50.0
47.9
50.9
49.1
43.3
50.0
41.7
387
-------
TOTAL SOLIDS DISTRIBUTION FOR STUDY 5
(Mechanical Mixing, 9:1, 0.08, 4% TS Feed)
Percent Total Solids by
Port Number
Level
Top
Middle
Bottom
Date
4/20
4/24
4/27
5/02
5/04
4/20
4/24
4/27
5/02
5/04
4/20
4/24
4/27
5/02
5/04
5
(wall)
18.8
17.6
14.9
16.9
22.0
1.3
1.3
3.2
0.6
2.2
1.5
3.6
1.2
1.4
1.3
4
25.5
27.4
17.6
21.0
19.7
1.2
1.7
1.3
0.6
0.6
1.3
1.2
2.1
1.8
0.9
3
0.9
0.8
0.7
14.8
13.8
__
0.6
0.9
0.4
1.4
1.3
0.8
0.9
0.9
1.4.
2
0.9
0.7
0.7
0.4
0.7
0.8
0.6
0.6
6.3
0.6
1.3
0.8
0.8
1.0
1.7
1
(center)
0.9
1.0
0.7
0.5
3.7
1.0
0.5
0.7
0.6
0.6
1.0
0.6
0.9
0.7
0.8
388
-------
VOLATILE SOLIDS DISTRIBUTION FOR STUDY 5
(Mechanical Mixing, 9:1, 0.08, 4% TS Feed)
Percent Volatile Solids by
Port Number
Level
Top
Middle
Bottom
Date
4/20
4/24
4/27
5/02
5/04
4/20
4/24
4/27
5/02
5/04
4/20
4/24
4/27
5/02
5/04
5
(wall)
65.1
68.3
70.7
72.0
72.1
40.0
57.1
79.5
52.9
50.9
47.9
50.0
50.0
50.0
43.3
4
69.0
67.8
83.6
76.1
69.0
40.0
63.6
47.6
64.3
60.0
46.7
68.3
61.4
52.3
60.0
3
36.0
42.9
52.0
76.6
62.1
__
66.7
44.8
70.0
57.6
43.8
61.5
42.9
52.6
52.6
2
40.9
47.6
59.1
75.0
43.5
38.1
62.5
42.9
70.0
52.6
36.7
69.6
50.0
68.0
53.0
1
(center)
44.4
48.2
40.0
75.0
69.7
27.8
73.3
39.1
66.7
63.6
47.8
66.7
50.0
61.1
52.0
389
-------
TOTAL SOLIDS DISTRIBUTION FOR STUDY 6
(Mechanical Mixing, 9:1, 0.14, 7% TS Feed)
Percent Total Solids by
Port Number
Level
Top
Middle
Bottom
Date
6/05
6/08
6/12
6/15
6/19
6/22
6/05
6/08
6/12
6/15
6/19
6/22
6/05
6/08
6/12
6/15
6/19
6/22
5
(wall)
2Q.7
34.6
22.1
17.8
31.6
36.0
1.5
1.1
1.8
2.5
1.5
1.7
1.0
3.8
1.9
2.8
1.1
2.2
4
__
34.2
24.6
—
33.9
27.2
__
—
1.7
—
1.5
1.3
—
1.6
—
2.5
1.4
3
30.8
25.6
15.0
23.7
29.3
31.3
2.1
2.0
1.1
1.2
—
—
1.3
1.3
1.3
1.2
—
—
2
1.2
2.6
0.9
15.7
0.7
1.2
0.7
0.9
0.9
3.2
1.0
1.3
1.2
1.2
1.5
0.9
1.2
1.2
1
(center)
0.9
1.7
5.9
0.9
1.3
2.1
0.7
1.1
1.2
0.9
1.0
1.3
1.0
0.9
1.0
0.9
1.7
1.7
390
-------
VOLATILE SOLIDS DISTRIBUTION FOR STUDY 6
(Mechanical Mixing, 9:1, 0.14, 7% TS Feed)
Percent Volatile Solids by
Port Number
Level
Top
Middle
Bottom
Date
6/05
6/08
6/12
6/15
6/19
6/22
6/05
6/08
6/12
6/15
6/19
6/22
6/05
6/08
6/12
6/15
6/19
6/22
5
(wall)
74.5
64.4
60.8
68.2
56.8
64.5
55.8
65.7
64.7
54.8
58.0
61.7
53.3
39.8
61.5
57.5
54.8
62.3
4
__
67.7
66.0
—
66.4
86.9
__
—
58.7
—
40.1
59.0
—
57.7
—
58.5
58.3
3
64.6
61.5
65.5
71.6
59.5
54.0
45.6
58.0
56.7
54.5
—
—
48.5
71.4
61.2
54.8
—
—
2
71.4
73.0
63.3
64.2
60.6
59.5
45.0
67.8
62.9
51.4
56.0
57.7
58.8
68.6
58.9
48.8
57.8
58.8
1
(center)
46.6
69.1
—
54.5
59.6
65.5
57.1
63.1
65.3
49.4
55.4
56.8
55.8
78.9
62.5
52.0
62.5
60.3
391
-------
TOTAL SOLIDS DISTRIBUTION FOR STUDY 7
(Gas Mixing, 9:1, 0.14, 7% TS Feed)
Level
Top
Middle
Bottom
Date
7/03
7/06
7/10
7/13
7/17
7/20
7/03
7/06
7/10
7/13
7/17
7/20
7/03
7/06
7/10
7/13
7/17
7/20
5
(wall)
25.7
31.0
26.6
31.1
26.2
33.3
2.7
1.5
.6
1.9
—
— —
2.9
3.2
.8
1.2
—
—
Percent Total Solids by
Port Number
432
29.5 30.6 13.5
36.5 31.3 24.4
21.5 29.4 17.2
35.8 35.6 28.0
31.3 22.9
37.0 35.3 18.1
1.6
1.2
.8
2.0
1.5
.7
1.0
1.2
.7
1.2
.9
.7
1
(center)
__
19.9
17.3
17.9
15.6
16.6
1.1
1.0
.8
1.6
.8
.8
1.1
1.0
.7
1.1
.7
.7
392
-------
VOLATILE SOLIDS DISTRIBUTION FOR STUDY 7
(Gas Mixing, 9:1, 0.14, 7% TS Feed)
Percent Volatile Solids by
Port Number
Level
Top
Middle
Bottom
Date
7/03
7/06
7/10
7/13
7/17
7/20
7/03
7/06
7/10
7/13
7/17
7/20
7/03
7/06
7/10
7/13
7/17
7/20
5
(wall)
74.2
70.7
58.3
72.1
70.0
69.4
64.0
56.9
52.6
53.2
—
—
63.7
54.6
49.7
55.1
—
—
432
62.3 47.5 76.5
62.4 53.8 74.2
55.6 55.8 73.0
61.8 60.4 72.3
48.8
57.5 57.4 71.3
50.9
54.8
55.1
49.3
52.6
53.0
63.0
53.5
48.9
58.1
59.1
55.9
1
(center)
79.6
74.3
66.5
75.9
74.6
53.6
52.9
50.4
48.6
49.6
54.0
53.7
52.9
52.7
73.5
52.7
66.6
393
-------
TOTAL SOLIDS DISTRIBUTION FOR STUDY 8
(Gas Mixing, 9:1, 0.20, 10% TS Feed)
Percent Total Solids by
Port Number
Level
Top
Middle
Bottom
Date
7/24
7/27
7/31
8/03
8/07
8/10
8/14
7/24
7/27
7/31
8/03
8/07
8/10
8/14
7/24
7/27
7/31
8/03
8/07
8/10
8/14
5
(wall)
35.5
29.8
27.3
32.8
29.1
23.5
15.6
__
—
—
—
—
0.9
1.0
__
—
—
—
—
1.0
1.2
4
__
35.2
34.0
33.7
36.7
14.8
14.8
__
—
—
—
—
1.1
1.0
__
—
—
—
—
1.0
0.9
3
— —
35.6
36.1
40.2
43.3
31.6
16.3
__
—
—
—
—
0.7
1.2
__
—
—
—
—
0.8
1.1
2
1.4
14.8
13.8
22.1
13.4
12.9
23.5
1.7
1.1
—
0.3
1.0
0.9
0.8
1.2
1.5
—
0.4
1.0
1.2
0.7
1
(center)
__ .
—
1.1
18.6
14.0
13.7
22.3
„
1.1
—
0.3
1.1
1.4
0.7
__
1.7
—
0.3
1.1
1.2
0.8
394
-------
VOLATILE SOLIDS DISTRIBUTION FOR STUDY 8
(Gas Mixing, 9:1, 0.20, 10% TS Feed)
Percent Volatile Solids by
Port Number
Level
Top
Middle
Bottom
Date
7/24
7/27
7/31
8/03
8/07
8/10
8/14
7/24
7/27
7/31
8/03
8/07
8/10
8/14
7/24
7/27
7/31
8/03
8/07
8/10
8/14
5
(wall)
61.9
60.4
59.0
58.8
56.0
69.3
78.4
—
—
—
—
48.3
51.2
__
—
—
—
—
53.6
56.3
4
— —
61.6
57.3
63.7
52.5
76.8
80.9
„
—
—
—
—
49.5
50.9
—
—
—
—
47.6
52.6
3
51.5
53.9
58.2
51.7
61.7
78.5
—
—
—
—
46.1
56.0
—
—
—
—
48.2
54.3
2
57.7
74.8
70.0
63.2
78.4
73.5
67.4
56.7
54.7
—
27.9
48.1
49.3
49.0
54.0
54.9
—
38.4
49.1
50.7
50.2
1
(center)
—
57.5
57.4
74.7
81.5
86.4
—
53.9
—
27.3
50.7
67.6
49.0
—
54.7
26.0
49.9
56.8
55.7
395
-------
TOTAL SOLIDS DISTRIBUTION FOR STUDY 9
(Mechanical Mixing, 9:1, 0.20, 10% TS Feed)
Percent Total Solids by
Port Number
Level
Top
Middle
Bottom
Date
8/17
8/21
8/24
8/28
8/31
9/04
9/07
9/11
9/14
8/17
8/21
8/24
8/28
8/31
9/04
9/07
9/11
9/14
8/17
8/21
8/24
8/28
8/31
9/04
9/07
9/11
9/14
5
(wall)
22.6
31.7
26.6
18.3
25.9
21.5
15.6
30.2
19.8
0.7
0.8
0.4
2.6
1,7
2.3
1.9
1.5
1.3
0.7
0.8
1.2
2.2
1.5
1.2
1.5
1.5
1.6
4
14.5
18.4
24.2
31.4
20.1
18.5
17.5
28.8
22.9
1.5
0.8
0.9
1.2
2.3
0.6
1.4
1.5
2.0
0.4
1.0
0.9
1.1
1.5
1.7
1.3
1.7
1.7
3
15.7
12.9
15.5
37.1
24.2
16.9
14.2
37.0
13.8
0.5
1.0
0.7
0.6
1.2
1.1
1.2
1.2
1.2
0.8
1.0
0.8
1.2
1.5
1.3
—
0.9
1.6
2
16.9
16.5
20.0
15.7
0.6
0.9
10.2
37.7
31.3
0.4
0.8
0.7
0.6
0.2
0.7
1.2
2.7
1.2
0.7
1.3
1.0
1.5
1.1
1.1
1.2
1.7
1.9
1
(center)
22.0
15.1
13.5
—
0.9
0.8
1.1
1.5
1.3
0.5
0.8
0.8
0.9
0.9
0.8
1.0
1.2
1.0
0.6
0.9
1.5
2.1
1.2
0.3
—
1.3
1.0
396
-------
VOLATILE SOLIDS DISTRIBUTION FOR STUDY 9
(Mechanical Mixing, 9:1, 0.20, 10% TS Feed)
Level Date
Top 8/17
8/21
8/24
8/28
8/31
9/04
9/07
9/11
9/14
Middle 8/17
8/21
8/24
8/28
8/31
9/04
9/07
9/11
9/14
Bottom 8/17
8/21
8/24
8/28
8/31
9/04
9/07
9/11
9/14
Percent Volatile Solids by
Port Number
5
(wall)
89.8
85.0
88.9
79.5
83.6
83.8
85.1
64.7
77.9
48.0
54.3
48.0
63.4
55.6
63.8
64.7
55.3
55.2
54.3
51.5
69.3
62.3
54.3
44.0
58.6
53.3
52.4
4
82.6
72.1
75.5
86.9
72.8
75.0
78.4
61.2
78.1
56.0
53.6
52.5
51.3
57.8
21.4
53.8
51.4
55.8
43.3
56.7
36.9
53.8
50.9
85.3
69.0
58.5
50.0
3
80.7
73.3
89.2
64.7
61.2
72.9
74.4
63.0
73.7
39.7
61.7
48.6
47.6
83.0
44.8
65.0
55.6
44.4
49.2
54.5
33.4
56.3
53.6
47.1
—
54.2
52.9
2
74.5
62.6
—
73.9
43.6
—
70.9
63.9
67.7
35.0
43.8
50.3
46.5
65.1
37.5
55.6
56.4
50.0
48.2
53.4
50.8
46.1
18.5
57.1
63.9
52.8
1
(center)
80.7
56.5
69.0
—
51.7
57.7
60.0
58.3
55.2
38.2
59.0
51.1
48.0
45.4
45.5
71.4
45.0
52.6
21.2
55.7
56.4
55.3
46.2
41.7
—
43.9
47.6
397
-------
SAMPLING
PORTS
MECHANICAL
AGITATOR
SCUM
BREAKER
PUMP
GAS
GUN
-MECHANICAL
AGITATOR
SCUM
BREAKER
PUMP
•GAS
GUN
398
-------
399
-------
LIST OF ATTENDANTS
Abeles, Tom
Hth Ave. , South
Minneapolis, MN 55407
Arlotta, Salvatore
Wehran Engineering
666 E. Main St.
Middletown, NY 10940
Absher, Susan M.
Environmental Protection Agency
4601 N. Park Ave., #1818
Chevy Chase, MD 20015
Barineau, James M.
Leon Cty- Public Works
Rt. 2, Box 656
Tallahassee, FL 32301
Albrecht, Oscar
SHWRD - MERL
26 W. St. Clair St.
Cincinnati, OH 45268
Basinger, Spencer K.
Oconee Area Plan. & Dev- Comm.
P.O. Box 707
Milledgeville, GA 31061
Alter, Harvey
Natl.Ctr.for Resource Recovery
1211 Connecticut Ave., NW
Washington, DC 20036
Beck, Richard B.
Burns & McDonnell Engr. Co.
7821 E. 80th St.
Kansas City, MO 64106
Ammons, James T.
U.S. Army Corp of Engineers
110 Robin Lane
Huntsville, AL 35802
Beck, William W.
A.W. Martin Assoc.,Inc.
900 W. Valley Forge Rd.
King-Prussia, PA 19406
Anderson, Bernard F.
E.I. DuPont Co.
1610 E. Lafayette Dr.
West Chester, PA 19380
Bernheisel, J. F.
Natl. Ctr. for Resource Recovery
1211 Connecticut Ave., N.W.
Washington, DC 20036
Anderson, Robert C.
Environmental Law Inst.
1346 Connecticut Ave., N.W.
Washington, DC 20036
Blackham, William H.
11738 N.W.. 26th St.
Coral Springs, FL 33065
Anderson, Robert J.
Mathtech
Box 2392
Princeton, NJ 08540
Bnjafield, D. W.
Redland Purle Lmt.
Claydons Lane
Essex, England
Anderson, Sonia
22 Fairway Dr.
Cranbury, NJ 08512
Bolton, Roger E.
Williams College
Dept. of Economics
Williamstown, MA 01267
400
-------
Boretsky, Metodij
U.S. Navy
8302 MacArthur Rd.
Wyndmoor, PA 19118
Chatterjee, Anil K.
SRI International
333 Ravonswood Ave.
Menlo Park, CA 94025
Bromley, John
Waste Research Unit
132 No. Ave.
Abingdon, OXON
Checca, Thomas J.
Post,Buckley,Schuh & Jernigan
13420 SW 77th Ave.
Miami, FL 33156
Brunner, Dirk R.
Environmental Protection Agency
5491 Wasigo Dr.
Cincinnati, OH 45230
Chian, Edward S.K.
Georgia Tech
Daniel Lab.
Atlanta, GA 30332
Brunner, Donald E.
U.S. Navy Civil Engr.
7266 Briarcliff Cir.
Ventura, CA 93003
Lab.
Clark, Dave
New Castle Cty.Public Works
2701 Capital Trail
Newark, DE 19711
Buivid, Michael G.
Dynatech Corp.
34 Bowker St.
Brookline, MA 02146
Clark, Thomas P.
Minn. Pollution Control Agency
3572 Golfview Dr.
St. Paul, MN 55110
Bush, Harvey H.
Greenleaf/Telesca
1451 Brickell Ave.
Miami, FL 33134
Cleveland, Elmer G.
Environmental Protection Agency
345 Courtland St., NE
Atlanta, GA 30308
Carpenter, Charles D.
City of Orlando
1046 W. Gore
Orlando, FL 32805
Cohen, Alan S.
Argonne National Lab.
1543 N. Chickasaw Dr.
Naperville, IL 60540
Carroll, William D.
City of Winnipeg
280 William Ave.
Winnipeg, Manitoba Canada
Cooley, Kevin J.
Post,Buckley,Schuh & Jernigan
3191 Maguire Blvd.
Orlando, FL 32803
Charbonnier, James W.
American Resource Recovery
P.O. Box 698
Jupiter, FL 33458
Cowhey, James J.
Land & Lakes Co.
123 N. Northwest Highway
Park Ridge, IL 60068
401
-------
Cresap, C. C.
Dougharty Cty. Public Works
P.O. Box 1827
Albany, GA 31702
Del Fino, Ronald T.
R.I. Solid Waste Mgmt.
39 Pike St.
Providence, RI 02903
Croke, Kevin
2639 Praire
Evanston, IL 60202
Deluca, Frank
Rt. 3, Box 1954
Odessa, FI 33556
Cunningham, Richard D.
The BF Goodrich Co.
500 So. Main St.
Akron, OH 44318
DeWalle, Foppe
Univ. of Washington
Dept. of Environ. Health
Seattle, Washington 98195
Curtis, Curtis D.
City of Orlando
1046 W. Gore
Orlando, FL 32805
Dietz, Jess C.
Clark,Dietz Engrs.,Inc.
P.O. Drawer 1976
Sanford, FL 32771
Dahl, Thomas E.
Martel Labs., Inc.
851 Snell Isle Blvd., NE
St.Petersburg, FL 33704
Doggett, Ralph
Intl. Research & Tech.
7655 Old Springhouse Rd.
McClean, VA 22101
Daly, Ernest
University of Miami
Dept. Mech. Engr.
Coral Gables, FL 33124
Domalski, Eugene
307 Summit Hall Rd.
Gaithersburg, MD 20760
Davis, Jack E.
Tuscaloosa Testing Lab.Inc.
P.O. Box 1094
Tuscaloosa, AL 35401
Donnelly, Jean A.
Univ. of Cincinnati
2323 Kenlee Dr.
Cincinnati, OH 45230
Davis, Robert
189 Sheridan Ave.
Longwood, FL 32750
Dower, Roger
Environmental Law Inst.
1346 Connecticut Ave.,N.W.
Washington, DC 20036
Deamud, John W.
Reynolds, Smith, Hills
7120 Lake Ellenor Dr.
Orlando, FL 32809
Duggan, James C.
Tennessee Valley Authority
440 Commerce Union Bank Bldg.
Chattanooga, TN 37401
402
-------
Eacott, Robert B.
Govt. of Western Australia
27 Eaton Villa Place
San Carlos, CA 94070
Fitzgerald, Sidney S.
SCA Services
901 Silver Hill Rd.
N.Little Rock, AR 72118
Eden, Burton H.
Combustion Equip. Assoc.,
61 Taylor Reed Place
Stamford, CT 06906
Inc.
Flatt, George
2008 Gibbs Dr.
Tallahassee, FL 32306
Elstrom, Alden
271% Pondella Rd.
N.Ft.Myers, FL 33903
Flower, Franklin B.
Cook College-Rutgers Univ.
P.O. Box 231
New Brunswick, NJ 08903
Emrich, Grover H.
A.W. Martin Assoc.,Inc.
900 W. Valley Forge Rd.
King-Prussia, PA 19406
Flynn, Daniel V.
111. Environ. Protection Agency
603 Eastman Ave.
Springfield, IL 62702
Farrell, Robert S.
ME Dept. of Environ. Protection
Div- of Solid Waste
Augusta, ME 04330
Fowler, Bert
Waste Mgmt. of 111.
P.O. Box 563
Palos Hgts., IL 60463
Fenton, Richard
P.O. Box 2842
St.Petersburg, FL 33731
Francingues, Norman
100 Redbud-Dr.
Vicksburg, MS 39567
Fiedler, Harry H.
Gilbert/Commonwealth
1615 Hampden Blvd.
Reading, PA 19604
Franklin, William E.
Franklin Assoc., Ltd.
8340 Mission Rd., Suite 101
Prairie Vill., KS 66206
Fillip, Alan J.
Vermont Environment Agency
Maplewood Rd.
E. Montpelier, VT 05651
Freeman, Harry
IERL - EPA
Cincinnati, OH 45268
Fisher, Gerald E.
E.I. DuPont
3707 Chevy Chase
Louisville, KY 40218
Friedman, D.
Environmental Protection Agency
401 M Street S.W.
Washington, DC 20460
403
-------
Fuller, Wallace H.
Univ. of Arizona
Soils, Water & Engr.
Tucson, AZ 85721
Gordon, Judith G.
The Mitre Corp.
1448 Aldenham Lane
Reston, VA 22090
Gabbay, S. M.
Occidental Research Corp.
P.O. Box 19601
Irving, CA 92713
Gresh, Gerald
3162 N.E. 7th Dr.
Boca Raton, FL 33431
Gangopadhyay, Tota
Beasy Nicoll Engr.Ltd.
6080 Young St., Suite 512
Halfiax, Nova Scotia
Griffin, Robert A.
111. St. Geological Survey
609 W. Columbia
Champaign, IL 61820
Garmon, Ronald C.
Texaco Inc.
3600 Normandy, Apt. A-9
Port Arthur, TX 77640
Griffith, Lloyd W.
Ross Saarinen Bolton & Wilder
200 E. Robinson St.
Orlando, FL 32801
Garrity, Richard D.
City of Tampa
City Hall Plaza
Tampa, FL 33803
Gupta, Ashok
Raytheon Service Co.
2 Wayside Rd.
Burlington, MA 01730
Geswein, Allen
401 M. Street, SW
Washington, DC 20460
Haigh, C. C.
Beaufort Cty. Public Works
P.O. Box 4279
Beaufort, SC 29902
Giddings, Todd
Todd Giddings & Assoc.
140 W. Fiarmount Ave.
State College, PA 16801
Halvorson,- Thomas G.
Union Carbide Corp.
5535 Mapleton Rd.
Lockport, NY 14094
Goodson, Robert H.
Seaburn & Robertson, Inc.
5510 Gray St., Suite 118
Tampa, FL 33609
Ham, Robert
2130 Chadbourne Ave.
Madison, WI 53705
Gorczynski, Jan
Stevens Elastomeric Prod.
25 Payson Ave.
Easthampton, MA 01027
Hanna, Edward E.
City of Orlando
1046 W. Gore
Orlando, FL 32805
404
-------
Hansen, Warren G.
SCS Engineers
2875 152nd Ave., NE
Redmond, WA 98052
Hentrick, Robert
Systems Tech. Corp.
245 No. Valley Rd.
Xenia, OH 45385
Harp, James D.
Senergy, Inc.
2961 Fitzooth Dr.
Winter Park, FL 32792
Heyden, Peter
Cahn Engineering
Alex Dr.
Wallingford, CT 06904
Hartz, Ken
702 G Eagle Hts.
Madison, WI 53705
Hickey, Peter
New Executive Office Bldg.
Room 3011
Washington, DC 20024
Hasselriis, Floyd
Combustion Equip.
555 Madison Ave.
New York, NY 10022
Hill, David T.
Auburn University
1817 SW 78th Terrace
Gainesville, FL 32601
Haxo, Henry E.
Matrecon Inc.
Box 24075
Oakland, CA 94623
Hoye, Robert
Pedco Environmental, Inc.
11499 Chester Rd.
Cincinnati, OH 45246
Hecht, Norman L.
Univ. of Dayton Research Inst.
300 College Park Dr.
Dayton, OH 45469
Hudson, James F.
Urban Systems Research
36 Boylston St.
Cambridge, MA 02138
Heckler, Dave
Thomas,Dean & Hoskins,Inc.
3808 8th Ave., So.
Great Falls, MT 59405
Hunt, Robert G.
Franklin Assoc., Ltd.
8340 Mission Rd.
Prairie Vill., KS 66206
Helmstetter, Arthur J.
Systems Tech. Corp.
245 North Valley Rd.
Xenia, OH 45385
Hunt, Tim F.
P.B.Co.Solid Waste Authority
120 So. Olive Ave.
West Palm Bch, FL 33401
Hendrix, Sylvan
Koppers-Sprout Waldron Div.
P.O. Box 6221
Lakeland, FL 33803
Husain, Mohammad A.
FL Dept. of Environ. Reg.
1409 Chowkeebin Lane
Tallahassee, FL 32301
405
-------
Iglehart, Cecil
SCA Services, Inc.
8806 Nottingham Pkwy.
Louisville, KY 40222
Landreth, Robert E.
Environmental Protection Agency
26 West St. Clair
Cincinnati, OH 45268
James, Steve
SHWRD - MERL
26 W. St. Clair St.
Cincinnati, OH 45268
Lavigne, Ronald L.
Resource Control Inc.
Quabbin Office Bldg.,Valley Rd.
Barre, MA 01115
Jones, Larry W.
U.S. Army Corp of Engineers
Univ. of Tenn.
Knoxville, TN 37916
Layland, Dwane
Lycoming Co.
48 W. Third St.
Williamsburg, PA 17701
Kinman, Riely N.
Univ. of Cincinnati
415 Stevenson Rd.
Erlanger, KY 42028
Lazarus, Arthur G.
C-E Maguire, Inc.
31 Canal St.
Providence, RI 02903
Kirklin, Duane R.
Natl. Bureau of Standards
Chem/B348
Washington, DC 20234
LaZenby, Mack R.
City of Sanford
P.O. Box 1778
Sanford, FL 32771
Klee, Albert
SHWRD - MERL
26 W. St. Clair St.
Cincinnati, OH 45268
Leamon, Michael R.
Forsyth Cty. Eniron. Dept.
2150 Storm Canyon Rd.
Winston-Salem, NC 27106
Klein, Michael S.
Mgmt. of Res. & Environ.
41C New London Turnpike
Glastonbury, CT 06033
LeCroy, Charles E.
Bay Cty. Solid Waste Dept.
206 South 22 A
Panama City, FL 32401
Knight, James A.
Georgia Inst. of Tech.
2117 Kodiak Dr. NE
Atlanta, GA 30345
Lewis, David
York Research Corp.
1 Research Dr.
Stanford, CT 06906
Laden, Kenneth G.
Dept. of Envir. Services
1321 So.Carolina Ave.,SE #4
Washington, DC 20003
Lincoln, Edward
University of Florida
AG. Engr. Dept.
Gainesville, FL 32611
406
-------
Liskowitz, John W.
NJ Inst. of Tech.
Bunker Dr., HD #2
Belle Mead, NJ 08502
Mavinic, Donald S.
Univ. of British Columbia
1001 Canyon Blvd.
N.Vancouver, B.C.,Canada V7R2K2
Lopez, Ralph A.
Mayes, Sudderth & Etheredge
5750 Major Blvd.
Orlando, FL 32805
Mayo, Franics T.
Environmental Protection Agency
26 West St. Clair St.
Cincinnati, OH 45268
Love, John
Univ. of Mo.-Columbia
Engr. Bldg. 2024
Columbia, MO 65201
Me Dermont, Donald
Post,Buckley,Schuh,& Jernigan
7500 N.W. 52nd St.
Mairni, FL 33166
Loven, Carl
Jamestown Star Rt.
Peakview Dr.
Boulder, CO 80302
Me Guire, Jerry N.
Monsanto Company
800 N. Linberg Blvd.
St. Louis, MO 63166
Lowe, Grady
City of Lake Worth
1301 12th Ave. So.
Lake Worth, FL 33460
McConnell, Wayne
15257 E. Colonial Dr.
Oralndo, FL 32809
Lutton, Richard
U.S. Army Corp of Engineers
P.O. Box 631
Vicksburg, MS 39180
McGregor, K.
Mgmt.Resources & Environ.
41C New London Turnpike
Glastonbury, CT 06033
MacDaniel, Robert D.
California Pellet Mill Co.
1450 Kleppe Ln.
Sparks, NV 89431
McManamy, John
250 Carib Dr.
Merritt Isle., FL 32952
Malone, Philip G.
U.S. Army Corp of Engineers
P.O. Box 631
Vicksburg, MS 39180
Mellott, Philip K.
Nishna Sanitary Serv.,Inc.
P.O. Box 182
Elliott, IA 51532
Mather, Thomas W.
Tim Mather, Inc.
2701 Casey Key Rd.
Nokomis, FL 33555
Meyer, G. L.
USEPA
Ofc. Radiation Progs.
Washington, DC 20467
407
-------
Miller, Carlton K.
Broward County
Room 526, 236 S.E. 1st Ave.
Ft.Lauderdale, FL 33301
Mutch, Robert D.
Wehran Engineering
666 E. Main St.
Middletown, NY 10940
Mitchell, Gary I.
SCS Engineers
11800 Sunrise Valley Dr.
Reston, VA 22091
Myers, Tommy
Waterways Experiment Station
P.O. Box 631
Vicksburg, MS 39180
Mitchell, John W.
Franklin Assoc.,Ltd.
8340 Mission Rd.
Prairie Vill., KS 66206
Neff, Russell R.
Mayes, Sudderth & Ethredge
5750 Mahor Blvd., Suite 200
Orlando, Fl 32805
Montgomery, Dale
Versar, Inc.
3257 Victor Cir.
Annadale, VA 22003
Neisser, Mark
John G. Reutter Associates
9th & Cooper Streets
Camden, NJ 08101
Mooij, Hans
Environ. Impact Control Dir.
Fisheries & Environ. Canada
Ottawa, Ontario, Canada
Nichols, Walter P.
State Of Alabama - Public Health
State Office Building
Montgomery, AL 36130
Moore, Byron G.
City Of Orlando - Public Service
400 W. Livingston St.
Orlando, FL 32802
Noble, David
4827 W. Bradock
Alexandria, VA 22311
Moreau, Raymond L.
FL Res. Recovery Council
2600 Blair Stone Rd.
Tallahassee, FL 32301
Nollet, Anthony R.
AENCO Inc.
Box 387
New Castle, DE 19720
Morgan, William L.
Board of Health
308 N. 16th St.
New Castle, IN 47362
Oberacker, Don
SHWRD - MERL
26 W. St. Clair St.
Cincinnati, OH 45268
Murray, David E.
Reitz & Jens, Inc.
Ill S. Meramec
St. Louis, MO 63105
Oppelt, Edwin T.
Environmental Protection Agency
7939 Shelldale Way
Cincinnati, OH 45242
408
-------
Pace, Robert
1291 Plymouth Place
Jacksonville, FL 32205
Rebholz, Richard P.
City of Fort Walton Beach
Rt.l, Box 91
Mary Esther, FL 32569
Patel, Vijay
Pedco Enviornmental, Inc.
11499 Chester Rd.
Cincinnati, OH 45246
Reese, John A.
FL Dept. of Environ. Reg.
2600 Blairstone Rd.
Tallahassee, FL 32301
Peritz, Leigh A.
Raytheon Service Co.
63 Washington St.
Winchester, MA 01890
Reilly, Bertram
9 Choctaw Trail
Ormond Beach, FL 32074
Peterson, Arnold
Stevens Elastomeric
26 Payson Ave.
Easthampton, MA 01027
Reynolds, Stan L.
Syst.,Science & Software
Box 1620
La Jolla, CA 92038
Pohland, Fred
Georgia Tech - Civil Engineering
4631 North Springs Ct. NE
Atlanta, GA 30338
Rinaldo, Marjory L.
Residuals Mgmt. Tech.
1406 E. Washington Ave.
Madison, WI 53703
Polk, Malcolm
Atlanta University
Dept. of Chemistry
Atlanta, GA 30314
Ritchie, Charles W.
Dept. for Natural Resources
209 Pin Oak Place
Frankfort, KY 40601
Power, Richard M.
SCA Services, Inc.
19 Daley Ave.
Methuen, MA 01844
Robbins, John A.
Sverdrup & Parcel & Assoc.
174 Royal Palm St.
Sebastian, FL 32958
Pullan, Henry
Redland Purle Lmt.
Claydons Lane
Essex, England
Roberson, David L.
State Public Health
State Office Bldg.
Montgomery, AL 36130
Purdy, Kenneth R.
Tech Consultants, Inc.
1485 Leafmore Ridge
Decatur, GA 30033
Rogers, Charles J.
Environmental Protection Agency
461 Merry Maid Lane
Cincinnati, OH 45240
409
-------
Roulier, Mike
SHWRD - MERL
26 W. St. Clair St.
Cincinnati, OH 45268
Searcy, Phillip E.
Post,Buckley,Schuh & Jernigan
3191 Maguire Blvd.
Orlando, FL 32803
Rugg, Barry
New York Univ.
Barney Bldg.
New York, NY 10003
Seelinger, Richard W.
VOP, Inc.
40 VOP Plaza
Des Plaines, IL 60016
Banning, Don
SHWRD - MERL
26 W. St. Clair St.
Cincinnati, OH 45268
Sengupta, Subrata
University of Miami
Dept. Mech. Engr.
Coral Gables, FL 33124
Sarver, Glen E.
B.F. Goodrich Co.
Rt. 3, Box 11
Marietta, OH 45750
Shanks, Howard R.
Ames Laboratory
A207 Phys. Bldg.
Ames, IA 50011
Scaramelli, Alfred B.
Mitre Corp.
4 Carley Rd.
Lexington, MA 02173
Silva, Thomas F.
General Electric
782 Downing St.
Schenectady, NY 12309
Scarpino, P. V-
Univ. of Cincinnati
Dept. of Civil & Env.
Cincinnati, OH 45221
Engr.
Simister, Bruce W.
Midwest Research Inst.
4135 N.E. Davidson Rd.
Kansas City, MO 64116
#374
Schmoeger, Gary A.
Martel Labs.,Inc.
6231 Cedar St.,NE
St.Petersburg, FL 33702
Simpson, William P.
Post,Buckley,Schuh & Jernigan
3191 Maguire Blvd.
Orlando, FL 32803
Schnelle, James F.
C-E Maguire, Inc.
31 Canal St.
Providence, RI 02903
Skinner, John
US EPA (AW 5-64)
401 M St., SW
Washington, DC 20460
Schomaker, Norbert B.
Environmental Protection Agency
65 Bayham Dr.
Cincinnati, OH 45218
Smith, Douglas F.
Dept. Environ. Reg.
825 NW 23rd. Ave., Suite
Gainesville, FL 32601
410
-------
Smith, Frank M.
Leon County
Rt. 2, Box 656
Tallahassee, FL 32301
Surowiec, Joseph T.
State of Georgia
3454 Alison Dr.
Doraville, GA 30340
Smith, Mark E.
Residuals Mgmt. Tech.
4201 Winnequah Rd.
Momma, WI 53716
Sussman, David
907 6th St.,SW,605-C
Washington, DC 20024
Srinivasan, V. R.
Louisiana State University
Baton Rouge, LO 70803
Swartzbaugh, Joseph T.
Systems Tech. Corp.
245 No. Valley Rd.
Xenia, OH 45385
Steeves, W. M.
Reynolds,Smith & Hills
P.O. Box 4850
Jacksonville, FL 32201
Taylor, James
University of Central Fla.
P.O. Box 25000
Orlando, FL 32816
Stenburg, Robert L.
Environmental Protection Agency
26 W. St. Clair St.
Cincinnati, OH 45268
Teer, Ellis H.
Glace & Radcliffe, Inc.
1347 Palmetto Ave.
Winter Park, FL 32789
Stenstrom, Michael K.
Univ. of California, L.A.
7619 Boelter Hall, UCLA
Los Angeles, CA 90024
Terrill, David
Lycoming Co'.
48 W. Third St.
Williamsburg, PA 17701
Steven, Theodore
Alex Dr.
Wallingford, CT 06904
Thomas, Robert A.
Southwest Ga. APDC
P.O. Box 346
Camilla, GA 31730
Stoffel, Clarence M.
Owen Ayres & Assoc.
Rt. 3, Box 16B
Bloomer, WI 54724
Trezek, George
CA Recovery Systems, Inc.
160 Broadway, Suite 200
Richmond, 'CA 94804
Stoll, Bernard J.
USEPA, OSW
401 M St., SW
Washington, DC 20460
Underwood, George T.
Underwood & Assoc.
Ft. Myers, FL 33904
411
-------
Vandervort, John D.
Schlegel Area Sealing Syst.
54 Aldwick Rise
Fairport, NY 14450
Weinberger, Carl S.
Teledyne National
1910 Pot Spring Rd.
Timonium, MD 21093
Vesilind, P. A.
Duke University
Dept. of Civil Engr.
Durham, NC 27706
Weinstein, Jack S.
Mayes.Sudderth & Etheredge
5750 Major Blvd..Suite 200
Orlando, FL 32805
Wagner, Paul L.
Mayes.Sudderth & Etheredge
3370 Vandiver Dr.
Marietta, GA 30066
Welch, Jerome
Kansas Geol. Survey
2606 Bonanza St.
Lawrence, KS 66044
Walsh, James J.
SCS Engineers
5242 Camelot Dr.
Fairfield, OH 45014
Wheeler, William B.
NE Georgia Planning Comm.
115 Lavender Rd.
Athens, GA 30606
Walter, Donald K.
Dept. of Energy
20 MS 2221C
Washington, DC 20545
White, Robert
425 Volker Blvd.
Kansas City, MO 64110
Wanielista, Martin P.
University of Central Fla.
P.O. Box 25000
Orlando, FL 32816
Wigh, Richard J.
Regional Services Corp.
3320 Woodcrest Ct.
Columbus, IN 47201
Ward, George D.
George D. Ward & Assoc.
821 NW Flanders
Portland, OR 97209
Wiles, Carlton C.
Environmental Protection Agency
5853 Brasher Ave.
Cincinnati, OH 45242
Watson, Jack E.
Seaman Corp.
401 Freeman Hill
Tryon, NC 28782
Wilkey, Michael L.
Argonne Natl. Lab.
9700 S. Cass Ave.
Argonne, IL 60439
Weaver, Larry D.
Warner Robins Air Log.Ctr.
113 Glenn Dr.
Warner Robins, GA 31093
Willey, Cliff R.
MD Environ. Service
60 West St.
Annapolis, MD 21403
412
-------
Wilson, Frank L.
Polk Cty. Environ. Services
P.O. Box 39
Bartow, FL 33830
Zimmerman, R. E.
Escor, Inc.
820 Davis St.
Evanston, IL 60201
Wilson, Ronald H.
Stottler Stagg & Assoc.
P.O. Box 272
Winter Haven, Fl 33880
Zumwalt, John B.
Post,Buckley,Schuh & Jernigan
2131 Hollywood Blvd.
Hollywood, FI 33020
Woelfel, Gregory C.
Waste Management, Inc.
900 Jorie Blvd.
Oak Brook, IL 60521
Zwolak, Richard A.
Hillsborough Cty.
700 Twiggs St., Suite 800
Tampa, FL 33612
Wong, Kau-Fui
University of Miami
Dept. Mech. Engr.
Coral Gables, FL 33124
Wood, G. M.
Ontario-Environment
135 St. Clair Ave., W.
Toronto, Ontario Canada
Worrell, William A.
Brown & Caldwell
53 Perimeter Ctr. East
Atlanta, GA 30346
Young, Richard M.
Kansas Geological Survey
2503 Winterbrook Dr.
Lawrence, KS 66044
Yousef, Yousef A.
University of Central Fla.
P.O. Box 25000
Orlando, Fl 32816
Zenzel, Kenneth. M.
Senergy, Inc.
433 Oak Haven Dr.
Altamonte Spr, FL 32701
413
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
i. REPORT NO.
EPA-600/9-79-023b
3. RECIPIENT'S ACCESSIOf*NO.
4. TITLE AND SUBTITLE
MUNICIPAL SOLID WASTE: RESOURCE RECOVERY
Proceedings of the Fifth Annual Research Symposium
5. REPORT DATE
August 1979 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
John Larson, Editor
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
University of Central Florida
Box 25000
Orlando, Florida 32816
10. PROGRAM ELEMENT NO.
1DC818
11. CONTRACT/GRANT NO.
806198
12. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research Laboratory—Gin., OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Symposium March 26-28. 1979
mp
POIS
14. SPONSORING AGENCY CODE
EPA-600/14
15. SUPPLEMENTARY NOTES
See also Municipal Solid Waste: Land Disposal, EPA-600/9-79-023a
Project Officer: Robert E. Landreth (513) 684-7748
16. ABSTRACT
The fifth SHWRD research symposium on land disposal and resource recovery of municipal
solid waste was held at Orlando, Florida on March 26, 27, and 28, 1979. The purposes
of the symposium were (1) to provide a forum for a state-of-the-art review and dis-
cussion of ongoing and recently-completed research projects dealing with the manage-
ment of solid wastes; (2) to bring together people concerned with municipal solid
waste management who can benefit from an exchange of ideas and information; and (3) to
provide an arena for the peer review of SHWRD's overall research approach. These
proceedings are a compilation of the papers presented by the symposium speakers. They
are arranged in order of presentation. Volume I, land disposal, covered four primary
technical areas: Pollutant Identification, Environmental Assessment, Control Tech-
nology, and Pollutant Transport. Volume II, Resource Recovery, covered five primary
technical areas: Evaluation of Equipment, Unit Operations, and Processes; Economics
and Impediments; Systems Analysis and Special Studies; Utilization of Recovered
Materials; and Environmental Impacts of Resource Recovery.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
cos AT I Field/Group
Comminution
Waste Treatment
Size Reduction
Marketing
Prices
Energy
Refuse Maintenance
Shredder Design
Waste-as-fuels
Resource Recovery
Secondary Materials
~5A~
5B
13B
3, DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport!
UNCLASSIFIED
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
21. NO. OF PAGES
424
22. PRICE
EPA Form 2220-1 (9-73)
414
j U.S. GOVERNMENT PRINTING OFFICE, 1979 -657-060/5439
------- |