SEPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Cincinnati OH 45268
EPA-600/7-78-086
May 1978
Research and Development
Engineering and
Economic Analysis
of Waste to Energy
Systems
Interagency
Energy/Environment
R&D Program
Report
-------
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
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-78-086
May 1978
ENGINEERING AND ECONOMIC ANALYSIS OF WASTE TO ENERGY SYSTEMS
fey
E. Milton Wilson
John M. Leavens
Nathan W. Snyder
John J. Brehany
Richard F. Whitman
The Ralph M. Parsons Company
Pasadena, California 91124
Contract No. 68-02-2101
Project Officer
Harry M. Freeman
Energy Systems Environmental Control Division
Industrial Environmental Research Laboratory
Cincinnati, Ohio 45268
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency, nor does mention of trade
names or commercial products constitute endorsement or recommendation for use.
11
-------
FOREWORD
When energy and material resources are extracted, processed, converted,
and used, the related pollutional impacts on our environment and even on our
health often require that new and increasingly more efficient pollution con-
trol methods be used. The Industrial Environmental Research Laboratory -
Cincinnati (lERL-Ci) assists in developing and demonstrating new and improved
methodologies that will meet these needs both efficiently and economically.
The subject of this report is an evaluation of various systems for con-
verting solid wastes to energy. The information contained herein will be of
interest to those involved in waste-to-fuel research and development programs,
and to those involved in the purchase, design, construction, or operation of
such systems. Inquiries and comments regarding the report should be directed
to the Fuels Technology Branch of the Energy Systems Environmental Control
Division.
David G. Stephan
Director
Industrial Environmental Research Laboratory
Cincinnati
111
-------
ABSTRACT
Waste quantities and characteristics in the U. S. are reviewed and
waste-to-energy conversion technology evaluated. All waste materials, exclu-
sive of those from mining operations, are considered. The technology is
reviewed under the categories of mechanical processing, biological conversion
systems, thermal/chemical systems, and combustion. Important features of
many operating facilities are described and detailed engineering and economic
analyses of seven specific systems are presented. An analysis is also made
of the technology and costs for conversion of pyrolytic off-gas to methane,
methanol, and ammonia. Environmental pollution data are presented where
available and the current control technology briefly reviewed. Conclusions
on the conversion technology are made and research needs considered in a
series of recommendations.
IV
-------
CONTENTS
Foreword iii
Abstract iv
Acknowledgement vi
Section 1. Introduction and Summary 1
Section 2. Methodology 5
Section 3. The Mass Burning Combustion Systems of
Resco-Saugus and Hamilton, Ontario .... 10
Refuse Energy System Company 10
Hamilton, Ontario 25
Section 4. Nashville Thermal Transfer Corporation 45
Section 5. City of Chicago - Commonwealth Edison
Supplementary Fuel System 62
Section 6. Georgia Institute of Technology Mobile
Agricultural Pyrolysis System 99
Section 7. Andco-Torrax Pyrolysis System 133
Section 8. Purox Pyrolysis System 167
Section 9. Occidental Research Corporation Flash
Pyrolysis System 227
Section 10. Capacity - Cost Summary 245
Section 11. General Conclusions 253
Glossary and Abbreviations 260
References 263
Form 2220-1 Technical Report Data 265
Appendix A. Wastes in the United States
Appendix B. Catalogue of Waste-to-Energy Processes
Appendix C. Alternative Uses of Pyrolytically-Formed Syngas
Appendix D. Environmental Control Considerations
Appendix E. SI Units of Measurement
-------
ACKNOWLEDGEMENT
The authors gratefully acknowledge the valuable contributions made by hundreds
of governmental and industrial waste and energy specialists. Each gave freely
of his time in the supplying of the information used in this report and in
reviewing text developed from the data. The number of such individuals is
too large to offer separate credits. The contributions of the original Project
Officer, Mr. James D. Kilgroe, were of great value in the early development
of the program and are sincerely appreciated.
VI
-------
SECTION 1
INTRODUCTION AND SUMMARY
The solid waste generated in just the larger metropolitan areas of the
country could, if converted to energy, supply up to one percent of the
nation's total energy needs. If available agricultural wastes could also be
converted into usable energy, the percentage contribution would be closer to
6%. During the past ten years there have been developed several approaches
for converting wastes into fuel or directly into energy. Recognizing that
there existed much relevant information and many competing technologies for
waste conversion, the EPA in 1975 initiated a project to evaluate the use of
solid waste as a means of supplementing conventional energy sources. The
results of that project are the subject of this report. The purposes of the
project were to:
• Determine the best estimates available concerning the quantities and
characteristics of waste materials, exclusive of mining wastes, in the
United States.
• Survey existing and proposed waste-to-energy technologies, and
identify the ones of most current interest to potential system users.
• Carry out in-depth engineering and economic analyses on selected
systems.
• Recommend research and development needs in the field of waste to
energy systems.
WASTES AVAILABLE FOR ENERGY CONVERSION
Although considerable variation exists between various estimates of
available combustible wastes, there undoubtedly exists an impressive amount of
waste. The most reliable estimates, given in terms of available (recoverable)
dry combustibles, are shown on the next page.
-------
Year _
Early 70's 1980 1990
Waste Stream Tg_ 106 Ton Tg_ 106 Ton Tg 10G Ton
Municipal
Municipal Solid Waste 61.1 67.4 78.0 86.0 98.0 108.0
Sewage Sludge 11.5 12.7 14.0 15.4 16.5 18.2
Industrial 42.3 46.6 48.2 53.1 49.9 55.0
Agricultural 201.2 251.7 316.2 548.5 384.6 424.0
TOTAL 525.1 558.4 456.5 505.0 549.0 605.2
On the assumption that the dry and combustible fraction of most wastes has a
higher heating value (HHV) of 18.61 MJ/kg (8,000 Btu/lb), the total energy
value of U.S. wastes will increase from the 6049 PJ (5.734 x 1015 Btu) of the
early 70's to an estimated 10 214 PJ (9.682 x 1015 Btu) in 1990. In terms of
oil equivalence, this amounts to 910,200,000 to 1,537,000,000 barrels.
TECHNICAL PROCESSES
There are many systems for converting wastes into energy either in opera-
tion commercially, or under some stage of development. Combustion processes
are available for converting wastes directly into heat energy. Other systems
involving various thermo-chemical, mechanical, and biological processes are
either available or under development to convert wastes into more usable
liquid, gaseous, or solid forms of fuel. In most cases surveyed it was found
that the technology in question, rather than being new, was usually just new
to the waste conversion field. Of the systems surveyed, the following were
selected for in-depth engineering and economic analysis. These systems are:
RESCO-Saugus Combustion System
Hamilton, Ontario, Waterwall Incinerator
Nashville Thermal Transfer Corporation
City of Chicago-Commonwealth Edison Supplementary Fuel Plant
Georgia Institute of Technology Mobile Agricultural Pyrolysis
System
ANDCO-Torrax Air Blown Pyrolysis System
Union Carbide Corporation Oxygen Blown Pyrolysis System
Occidental Research Corporation Flash Pyrolysis System
2
-------
The results of the in-depth analyses are contained in the body of the
report as Sections 3 through 9. The results of the broader technology and
waste surveys, along with support information, are in the Appendices. Primary
units in this report will be noted to be in the modernized metric system known
as SI. An explanation of these measurement units is also given in the
Appendix.
In that the decision to select one system over another is so extremely
site specific, depending upon such factors as local demand for the energy
product and characteristics of the particular waste being processed, no
attempt was made to rank the systems on a comparative basis. Much of the
information in this report could certainly be useful in developing such a
comparison. However, to make such a comparison meaningful, the information
contained in this report should be modified by local and regional consid-
erations.
CONCLUSIONS
Conclusions relating to the individual systems studied are in that part
of the report dealing with the subject system. More general conclusions are
contained in Section 11. Major conclusions of the projects were:
• To date there has been very little data published on environmental
emissions and effects from conversion plants. Past and on-going test
results should be immediately documented and issued to those con-
cerned. Applicable regulations and a brief review of the best avail-
able control technology should be included.
• Much developmental work is still needed to establish the best types
of equipment for handling and processing solid wastes. Most of the
existing systems have been adapted from non-solid waste usage.
• The physical and chemical characteristics of the various types of
refuse-derived fuel are still largely unknown. There is need for the
characterization of these fuels, and for the characterization of var-
ious blends of these fuels with fossil fuels.
• More R£D work is needed to develop fundamental information on pyrol-
ysis chemistry and applied studies are needed to advance the state of
the art.
• There is little information available on the combustion characteris-
tics of refuse-derived fuel. The body of the report contains several
specific suggestions for increasing the amount of knowledge in this
area.
• A number of facilities in operation or undergoing start-up tests can
convert wastes to energy in a manner economically attractive to both
the waste generator and the energy purchaser.
As the cost increases for producing energy and for disposing of wastes,
waste-to-energy systems will become even more economically desirable. The
-------
results of this project indicate that although there are problems yet to be
solved, they are certainly not insurmountable. The information provided in
this report will add to the needed solutions and advance the day when solid
wastes can contribute their energy content to the nation's fuel needs,
-------
SECTION 2
METHODOLOGY
INTRODUCTION
Economic, technological, environmental, and public acceptability factors
all influence the eventual success or failure of a commercial scale waste-to-
energy processing facility. The actual sensitivity of each factor in influ-
encing the total system, however, cannot be predicted on considerations of
national averages and any listing of systems in some form of ranking by value
is essentially meaningless.
Planning must nevertheless begin with some more or less uniform guide-
lines, basic assumptions, and data base. The availability of this fundamental
information to local government and industrial organizations therefore serves
a useful purpose in the overall process of reaching a decision on incorpora-
tion of a processing plant into waste management plans. While the general
survey results presented here will assist in understanding the overall field
of concern, examples of specific detailed analyses are necessary to indicate
the methodology to be followed in conducting the local studies. In that it
was logical to select the systems to be so analyzed after review of all
potential systems, no definite number or type of facilities were specified
within the formal contract work statement. The selection process that was
used for the actual systems discussed within Sections 3 through 9 is reviewed
here along with the other methodology applied.
STATUS OF TECHNOLOGY
A review of any technology reveals a similar grouping of individual
projects into such categories as:
• Conceptual processes based on theoretical considerations, fundamental
scientific information, or observations of analogous reactions or
techniques within other technologies.
• Full-scale plants under construction, or completed and start-up
testing still in progress.
• Fully "mature" production plants: the mere fact that a facility has
been in operation for a number of years need not imply that it was
properly designed and that many problems are not being experienced.
Processes for converting wastes to energy forms are to be found in each
of the above categories. In terms of absolute number of reports, the great
-------
majority were found to fall towards the top of the category lifting. Such
basic work is of course essential to eventual commercialization of a new sys-
tem some 8 to 10 years after the initial laboratory experiments. Its use for
any analysis other than the most tentative can lead to erroneous conclusions,
practically always on the optimistic side. Homogeneous, finely divided, and
oftentimes synthetic samples of waste are used under energy-input conditions
that bear no relationship to techniques to be used in production. No informa-
tion is derived on the practical—and costly—problems of materials handling
and environmental controls. The process developer for the early stage work,
typically not knowledgeable in modern construction methods and their costs,
can associate with his process highly favorable economics that will not be at
all truly assessed until undue effort has been expended. Much of the current
small scale research is of high caliber and covers the full range of tech-
nology possibilities, but by its very nature such work does not lend itself
to quantitative process analysis.
The extent of information available on systems beyond the bench-scale
level is a function of the complexity of the overall process and the funding
the developer could allocate to or solicit for the project. This support,
prior to the demonstration plant final design, can be as high as $10 million,
and hence few systems progress to this stage. Processes within these larger
scale capacities are in general well researched. The data available are
characterized, however, by the necessarily limited objectives of the devel-
oper, rather wide deviations under supposedly fixed conditions, and the lack
of some essential measurement (oftentimes as fundamental as the input waste
composition). These limitations of course affect the accuracy of any further
engineering analyses in that theory is not sufficiently advanced to estimate
the effect of process variables or the value to be assigned to missing data
points.
CRITERIA FOR CANDIDATE SELECTION
Review of the pertinent information sources indicated the following
criteria should be used for developing a total listing of potential candidate
systems and for further narrowing down this group to a number suitable for a
report meeting the objectives of the contract:
• The final listing should encompass several types of processing techno-
logy, preferably at least one from each of the fields of mechanical
processing, biological conversion, thermal/chemical conversion, and
combustion.
• The systems should be capable of processing a variety of input waste
materials.
• A variety of output energy forms should be shown as examples.
• The stage of development should be such that process information is
sufficiently reliable to y^ield estimated construction costs for full
scale plants accurate to on the order of ±25%.
-------
• A recent detailed engineering analysis is not available.
* While all systems should be of advanced design, a range of level of
development should be included within the listing of candidates.
An original listing of 13 systems was prepared, meeting most of the
above requirements, as follows:
• The Pompano Beach, Florida, anaerobic digestion system
• Landgard, Baltimore, pyrolysis system
• Ames, Iowa, RDF plant
• Milwaukee Americology RDF plant
• Sanitary landfill methane recovery
• Wet processing RDF (Black-Clawson)
• The RESCO and Hamilton Combustion Systems
• Nashville Thermal Transfer Corporation
• City of Chicago Supplementary Fuel Plant
• Georgia Tech Mobile Pyrolysis System
• ANDCO-Torrax Air Blown Pyrolysis System
• Union Carbide Purox Pyrolysis System
• Occidental Research Corporation Flash Pyrolysis System
This quantity was considered to be an excessive number and would have resulted
in a report of such length that reader interest would have been lost. The
initial six were eliminated and comments on two of these are deserved. Bio-
logical conversion systems are considered to offer significant technical
advantages when the raw wastes have a high water content, yet an example is
not present in the final listing. This results from the fact that the Pompano
Beach anaerobic digestion system is not yet operational-and the experimental
results from methane generation at this level of processing are essential to
any engineering/economic analysis beyond the several detailed reviews of
laboratory scale work now available. The Monsanto Landgard pyrolysis system
at Baltimore is the largest pyrolysis system in the U.S. It is not contained
within the candidates because at the time of the preparation of this report
the facility was undergoing modifications to solve mechanical problems and
to reduce atmospheric emissions. The process yields a hot, low heating value,
gas that is immediately combusted to raise steam, and an example of a pyrol-
ytic steam generator is given in the form of the Andco-Torrax system. In
addition, the Landgard plant has been described adequately in a number of
publications.
-------
Contained within the engineering analyses are two rather detailed
discussions of typical equipment and building requirements of processing
facilities. These details are cited for a refuse-derived supplementary fuel
plant (City of Chicago) and for a pyrolytic system yielding a medium-heating
value gas (the Union Carbide PUROX system). In an extension of this latter
section, Appendix C contains a discussion of the conversion of a pyrolytic
syngas to other products. Attention is directed to the fact that the PUROX
equipment as now offered for sale by UCC only is intended to produce syngas
and the additional analysis is wholly that of The Ralph M. Parsons Company.
INFORMATION SOURCES AND BACKGROUND NEEDS
Information was obtained through (1) letter and telephonic requests to
known investigators; (2) computer-generated retrieval listings of the EPA and
the Smithsonian Science Information Exchange; (3) conference proceedings of
the last 8 years; (4) continuing review of trade journals and newsletters; and
(5) field visits to government, university, and industrial authorities. Each
of these primary sources led to a number of other sources, which in turn
generated additional documents or contacts. A logging, abstracting, follow-
up, storage, and retrieval system was organized for administration of the
information system. Reports were reviewed by specialists in process chemistry,
equipment design, materials handling, biology, economics/cost estimating, and
environmental controls.
In preparing the engineering analyses of the candidate systems, close
cooperation with the facility developer or operator was essential. A number
of the systems have not yet reached a stage of commercial-scale processing
capacity and hence it was necessary to establish probable total plant needs
according to current construction practices and applicable codes and regula-
tions. This was typically accomplished in a series of steps, beginning with
a request to the owner/operator to supply experimental process mass and
energy balance data, the concept for a plant lay-out for his recommended pro-
cessing capacity, and an initial cost estimate for equipment. The project
staff at Parsons then reviewed this information, added necessary interface
and accessory equipment design, and prepared an independent cost estimate
based on actual construction experience and factors utilized within the
Architect-Engineering profession.
The cost elements were derived from vendor discussions and quotations;
trade journal indices for current costs of materials, equipment, and labor;
and values obtained from on-going construction projects within the four oper-
ating Divisions of Parsons. While only energy products were of immediate
concern here, certain of the processing steps yield both fuel and isolated
material commodities from the waste. No attempt was made to allocate costs
between these two functions. Estimates presented indicate all sub-systems
involved in a given facility, and these should be carefully reviewed before
conclusions are sought on relative costs. Those plants yielding a material
product can of course gain income from sales of those materials. Such credits
can have important off-setting effects on capital recovery and operating
expense costs. Because the subject of total plant economics is one where
condensation of information can lead to misinterpretation, no attempt has
been made to give a summary of this area.
-------
Process modifications on the preliminary systems were suggested by
Parsons and comments upon these solicited. After agreement had been reached
on a rather detailed design, capital and operating costs for the facility were
developed with greater accuracy and scaling factors established where appli-
cable to ascertain cost-capacity sensitivities. The developer then reviewed
the final version of the analysis as given in this report. The candidate
systems contain either some items of a proprietary nature or equipment not yet
verified as to its effectiveness at commercial processing capacity, and hence
some uncertainty exists within the estimates. It is nevertheless believed
that the costs presented herein are the most accurate yet published.
Organizations considering the possible financing and construction of a
full-scale waste processing facility must have available to them a staff of
specialists knowledgeable in a wide variety of fields. Administrative
officials or executives need not possess the detailed expertise of the plan-
ning and engineering personnel, yet they should appreciate certain of the
principles upon which decisions will ultimately be based. Review of all of
these speciality areas is outside the scope of this report and only one of
them, environmental pollution abatement, is discussed here. This is done
within Appendix D in that the review of this subject matter is not essential
to the understanding of the processes themselves. Within each of the discus-
sions of the individual candidate systems is presented the best available
information on effluents from the processes.
Economic evaluation is the other area that decision makers must under-
stand at least in general terms. The estimation of true long term net costs
(or profits) is a difficult task that must be approached realistically. All
possible expenditures attributable to the waste processing facility must be
established and revenues should be considered only after adequate market sur-
veys have been completed. Numerous texts on capital investment theory are
available and these should be consulted prior to approval of project funding.
-------
SECTION 3
THE MASS BURNING COMBUSTION SYSTEMS OF RESCO-SAUGUS AND HAMILTON, ONTARIO
Both of these facilities are rather large waterwall combustion systems,
each having its own unique design features, and they are considered together
here as a single candidate type.
REFUSE ENERGY SYSTEM COMPANY (RESCO), SAUGUS, MASSACHUSETTS
Introduction and Summary
This private corporation, a joint venture of Wheelabrator-Frye, Inc.,
and M. DeMatteo Construction Company, owns and operates a mass burning, water
walled combustion system normally handling 1090 Mg/d (1,200 TPD) of solid
waste with an output of 3810 Mg/d (8.4 million lb/day) of steam for indus-
trial use. This system was designed by Wheelabrator-Frye, the exclusive
licensee in the United States and Mexico for the Von Roll technology. The
steam produced is sold to the General Electric Company Lynn Works, directly
across the river from the plant.
Proven technology is used throughout the facility and a high degree of
reliability is incorporated into the design. A thorough waste supply and
product sales analysis was made prior to construction of the plant, assuring
that the facility was adequately funded and could profitably serve both the
local area and the owners.
Conclusions
The Saugus plant of RESCO serves as an excellent example of how waste
materials can be technically and economically converted to energy. Because
the RESCO combustion system is privately owned, not all the details of its
operation are available. Nevertheless, the plans and procedures of this
facility typify elements important to the success of such a plant.
• Waterwall combustion systems for refuse are the best established of
all waste-to-energy processes. It is recommended that governmental
and industrial organizations who have narrowed their conversion
options to inclusion of such processes thoroughly review the rel-
ative advantages of the RESCO facility.
• A thorough and realistic analysis must be made of the quantities and
characteristics of wastes available for energy recovery. Committments
must be obtained for sufficient quantity of waste over the useful
plant life, with financial arrangements mutually beneficial to buyer
10
-------
and seller. RESCO accomplished this with the communities within a
reasonable distance of the plant.
• Because energy is a product subject to all normal economic forces,
purchases must be assured prior to expenditure of large amounts of
capital. The arrangement made by RESCO for sales of steam to General
Electric, in addition to contractual agreements on joint utility and
supply needs, assured a sound revenue base that assured raising of
necessary funds.
• Design was accomplished by a highly qualified engineering organiza-
tion with the freedom to specify equipment and subsystems with high-
performance standards. No attempt was made to either "cut corners"
or to incorporate technological advances not yet proven.
• Start up testing was approached realistically anticipating problems
typical to any new facility. Funds were immediately available to
make any necessary modifications and corrections.
Process Description
A cross section of the facility is presented in Figure 1. Processing
begins with the arrival of refuse collection or transfer trucks at the plant.
As in most waste disposal facilities built in the last decade, all loaded
trucks pass over an automatic recording truck scale to determine and record
load weight for billing and plant operational planning. After weighing, the
trucks dump the raw refuse into a large receiving pit and depart. Refuse
remains in the receiving pit until it is moved by the bridge crane and clam-
shell grapple to the steam boiler charging hopper. The bridge crane and
grapple is manually operated from a control cab from which the operator has
a direct view of all refuse receiving, and steam boiler charging operations.
A shredder is used only to reduce unusually large items. Refuse is fed
to it by the crane and grapple when in the judgment of the operator the
refuse pieces are too large for charging to the boiler. This bulky waste
is loaded, on to a shredder feed conveyor, fed to the shredder, and the
shredded material returned via a chute to the receiving pit. Only a small
portion of the total refuse handled passes through the shredder, a practice
quite typical of European operations.
Refuse as received is moved directly to the boiler charging hopper
without any pre-processing, and this hopper then feeds by gravity on to the
drying section of the grate. It is necessary to maintain the hopper well
filled with refuse to act as a plug to keep the flame from the boiler from
blowing back out the hopper opening.
Refuse fed into the boiler is burned on the three-section Von Roll
grate. The first section provides space and time for the refuse to dry out
prior to combustion. Primary combustion occurs on the second section and
final burnout on the 'third section. The hot gases from combustion pass
through a superheating tube bank, an evaporator tube bank, and an economizer
tube bank before exiting from the boiler. The steam, produced at 4.86 MPa
11
-------
CLINKER AND ASH FUTURE
QUENCH CHANNEL
DRAG CONVEYER
Figure 1. Typical cross section of Resco facility.
-------
(690 psig) and 468°C (875°F) is piped to the General Electric Co., Lynn
Works, 914m (3000 ft.) away.
The flue gas passes from the boiler through a Wheelabrator-Frye electro-
static precipitator to the stack. Auxiliary equipment such as forced draft
fans, pumps, etc., are not shown as part of the process flow diagram.
Bottom ash from the end of the burnout grate and siftings that pass
through the grates are collected in a quench tank. This tank is equipped
with a drag chain conveyor to move the cooled wet ash to a metal and aggre-
gate recovery system. Fly ash from the several boiler tube tanks and the
precipitator is collected by an ash conveyor that discharges the fly ash
through a chute to combine it with the bottom ash as it is conveyed to a
material recovery system.
The material recovery system consists of several ash handling conveyors
and a rotating screen. The large metal pieces making up most of the coarse
material from the rotating screen are conveyed to a metal holding bin where
they are sold to metal salvage firms. The fine screenings are magnetically
scalped to reclaim small pieces of ferrous metal, which are sold. The fine
ash is conveyed to trucks for disposal in a landfill.
In addition to the refuse-to-steam process described above, there are
some related activities that are not in the direct process flow, but never-
theless contribute to a successful operation. These are:
• Auxiliary fuel firing - In order to meet the requirement of maximum
reliability in steam supply demanded by the steam customer, pro-
vision is made for two back-up steam systems. First, each refuse
boiler is equipped to burn oil in order to ensure the availability
of steam if refuse is in short supply and to provide a peaking steam
capacity. Second, two standby oil-fired water tube boilers are pro-
vided which have a total steaming capacity approximately equal to
the average demand on the plant. Usage of these systems will be
dependent on the uniformity of supply of refuse and demand for steam.
• Feedwater - The steam supply contract with General Electric Company
includes a requirement for feedwater return as needed, preheated to
121°C (250°F). This does not eliminate a need for water make-up but
it does reduce the amount needed and simplifies water treatment.
• Oil supply - The steam supply contract also requires that General
Electric provide the auxiliary fuel oil via pipeline. This elimi-
nates a requirement for tankage at the steam plant and shifts the
burden of ensuring oil availability to the steam customer.
• Electric power - G.E. has electric power generating capacity for
80,000 kW, including 20,000 kW of gas turbine generator capacity not
dependent on a steam supply. In addition, G.E. has a 20,000 kW tie
with the Massachusetts Electric Company. These units are able to
supply the 6510 kW connected load of the steam plant. Payment to
13
-------
G.E. for both power and the fuel oil used will be made by supplying
an equivalent additional amount of steam.
Equipment Details
The foregoing process description gives a brief discussion of the flow
of materials through the facility. This section contains a more detailed
description of the major equipment items and an explanation of their oper-
ating characteristics as related to the process.
Scales--
The automatic recording truck scales are activated by a coded plastic
card carried on each truck authorized to dump refuse at the facility. This
card identifies the truck and its owner, for billing and input of correct
empty truck weight. The empty weight of each truck is recorded in the scale
system, making it unnecessary to exit over the scales every time a load is
delivered. Periodic empty weight checks are made to keep the record current.
Receiving Pit--
The refuse receiving pit is a 26 m (85 ft) deep by 61 m (200 ft) long by
12 m (39 ft) wide concrete pit that the designers calculate to hold 6078 Mg
(6700 tons) of refuse when filled to maximum capacity, equivalent to 5.6 days
of steam generator operation. The normal capacity is 2540 Mg (2800 tons),
enough for 2.3 days of operation. The high level of maximum capacity serves
two purposes. It provides storage capacity to cover an unusual or unforseen
situation and it permits future expansion of the facility without increasing
the size of the pit, an expensive undertaking.
Travelling Cranes--
There are two overhead bridge cranes on a common runway over the receiv-
ing pit. Each is equipped with a 2.7 Mg (3-ton) refuse grapple and is con-
trolled by an operator riding with the crane in an air conditioned cab. There
is provision for adding one more crane when the plant is expanded. These
cranes feed the refuse to the boiler charging hoppers or to the shredder feed
conveyor as necessary and in addition do some mixing of refuse in the pit.
Both cranes can serve either of the two boilers.
Shredder--
There is a single 22.7 Mg/h (25 TPH) hammermill that is used to reduce
occasional oversized waste to 30 cm (12 in.) or less size. Agreements with
the cities that deliver their waste to this facility prohibit the dumping of
dangerous materials or those that are excessively large or extremely heavy.
Nevertheless, some items received must be reduced in size in order to be
acceptable for burning. The shredded material is returned to the pit.
Boiler Charging Hopper--
The system of charging refuse to the steam generator furnace in the RESCO
design is a simple gravity chute or hopper that depends on the weight of the
refuse in the hopper to move the refuse out the bottom and on to the furnace
grate. The hopper must be long enough to provide a sufficient plug of refuse
to retain hot gases in the furnace. The hoppers in this facility are equipped
with nuclear bin level indicators that will sound an alarm to warn the crane
14
-------
operators when the bin level is low and there is potentially an insufficient
plug to retain the hot furnace gases. A portion of the hopper nearest the
furnace opening is water cooled by some of the boiler make-up water.
Steam Generators--
The two steam generators are each rated to burn a maximum of 680 Mg
(750 tons) of municipal solid waste per day while producing steam at 4.86 MPa
(690 psig) and 486°C (875°F). Normal operation is 544 Mg (600 tons) per day
per unit and the overall plant operation is planned for this average amount.
At this throughput, average steam output will be 79.4 Mg (175,000 Ib) per
hour per unit, assuming an average waste heat content of 10.47 MJ/kg
(4,500 Btu per Ib).
The steam generators are Von Roll design waterwall boilers with three
level inclined reciprocating grates. The reciprocating grate action both
moves the refuse down the incline and turns it to ensure complete combustion.
Final burnout occurs on the last grate section and the ash is then discharged
into a water sealed hopper for quenching and removal.
Primary combustion air is introduced beneath the furnace grates. The
intake for the primary air is in the refuse pit area, which causes this area
to be constantly under a slightly negative pressure with all air flow inward
and toward the fan. By this simple arrangement, any odors are reported to be
retained in the pit area and ultimately burned away in the boiler.
Secondary air is introduced above the grates through nozzles in the re-
fractory walls. The water walls of the boiler start at a level above the
grates. The furnace is refractory lined at the lower level near the grate
where there is the possibility of a reducing atmosphere. The overfire
secondary air is used to help complete combustion and keep the flue gas tem-
perature within working limits. The gas temperature must be maintained
between 871°C (1600°F) and 1093°C (2000°F) to obtain proper boiler perfor-
mance and to inhibit corrosion of the water tubes of the steam superheater,
generator, and economizer.
The superheater, generator, and economizer are constructed in the form
of vertically-hung tube panels. This design allows the use of mechanical
rappers to clean ash and scale off the tubes.
Standby Boilers--
Two oil-fired, package type water tube boilers are provided for standby
steam production when the main refuse burning boilers are out of service for
any reason and to serve also as peaking units, supplying additional steam
during peak demand periods. These boilers are each rated at 54.4 Mg (120,000
pounds) of steam per hour at a. design pressure of 4.86 MPa (690 psig) and a
temperature of 454°C (850°F). They share major auxiliaries such as feed
water treatment and fuel supply with the main boilers.
Air Pollution Control--
Each refuse boiler is equipped with a Wheelabrator-Lurgi electrostatic
precipitator for removal of particulate matter from the flue gases. These
are two-field type units sized for a boiler discharge of 5660 m^/min
15
-------
(200,000 CFM) of flue gas at a temperature of 220°C (428°F) with a particulate
loading of 2.4 to 4.8 g/m3 (1 to 2 grain/ft3) adjusted to 12 percent CO^.
They are designed for a collection efficiency of 97.5 percent to maintain
emissions to the atmosphere within the allowable 1.77 g/m3 (0.05 grain/ft3).
The design specific collection area (SCA) is 19.42 m? (209 ft2) per 28.3 m3
(1000 ft3) of flue gas.
There is no equipment specifically dedicated to control of oxides of
nitrogen. Instead, production of NOX is curtailed by holding boiler tem-
perature below that required for significant formation of NOX from atmo-
spheric nitrogen. Further, the relatively low heat content of refuse and the
handling of the fuel during combustion precludes the generation of localized
high heat zones and high flame temperatures. Refuse is a low nitrogen fuel
and would contribute little to the production of NOX from, that source. It
should be noted that the air quality standards of the Commonwealth of Massa-
chusetts specifically permit the use of process control to meet emission
requirements, as discussed above, in lieu of special emission control equip-
ment, provided that measurement verifies emissions are below the permitted
maximum levels.
Refuse is a low sulfur fuel with a sulfur content of less than 0.3
percent and normally less than 0.1 percent. Such a low sulfur content in the
fuel will maintain sulfur dioxide emissions below the permitted maximum levels
without any added controls or equipment. The auxiliary fuel oil used also
has an equivalent low sulfur content.
Ash Handling--
The bottom ash and siftings from the grates drop through collection
hoppers to a quench tank below each furnace. Ash is removed from the tank
by a drag chain conveyor and delivered to an ash disposal system. This
method of handling ash is quite common where an ash-producing fuel is used.
It is simple, easily operated without sophisticated controls, and is quite
dependable.
Fly ash is collected from the precipitator and boiler hoppers by a long
enclosed conveyor and dropped through a chute to the ash quench tank for
further processing with the bottom ash. It is a simple system that is facil-
itated by the in-line and comparatively level arrangement of the various hop-
pers so that they can be readily served by one long conveyor. There is pro-
vision to by-pass the quench tank if a market for the fly ash is developed.
The wet ash removed from the quench tank is transported by conveyor to
a material recovery building where salvageable metals are separated from the
ash and facilities are provided to load metals and aggregate on to waiting
trucks for sale. The entering ash is screened in a rotating trommel to
separate bulky metals from the fine metal and aggregate materials. The fine
screenings are scalped by a magnetic separator to remove fine ferrous metals.
Separate storage hoppers are provided for the large metal screenings, fine
ferrous metal, and aggregate material.
16
-------
Water Use--
The in-plant water use and distribution system is designed to maintain
usage and waste water disposal quantities to a minimum. The average water
consumption, mostly to produce steam, is 0.027 m3/s (430 gpm). The average
discharge of waste water is only 0.0008 m3/s (12 gpm), of which 8 percent is
estimated to be sanitary waste. In order .to arrive at this low figure of
total water waste, some prudent design features were incorporated to make
multiple use of water where possible. For example, a portion of the boiler
make-up water is used as cooling water on the boiler feed chutes and ash dis-
charge part of the waste stream. Another lesser example is that the pump'
gland seal water and other miscellaneous waste water streams are drained to
the ash quench tank to form part of its make-up water. The waste water is
pumped to the City of Saugus sewer system.
Existing Operation
This facility started up in late 1975 and there has not been enough
operating experience with it to develop reliable data on its operation.
RESCO, the owners and operators, have stated that the plant is meeting the
design objectives and that initial actual operating and maintenance costs
are within the expected range.
Engineering Evaluation
This facility must be considered as one utilizing a very conservative
engineering approach for the conversion of refuse to useful energy. Its
European mass burning technology has been developed and proven in service
in many installations for years and the key items of equipment, the furnace
and the grates, are of European design. With the adoption of a well devel-
oped waste-to-energy system, the inherent problems of the associated equip-
ment are known and can be guarded against by careful design. For example,
the tendency of a mass burning furnace to have a corrosive reducing atmo-
sphere immediately above the fire is well known. The designers of this
facility have guarded against this condition by lining the lower section of
the furnace with refractory material and by careful control of combustion
air. Any fireside tube wastage noted can be controlled through proper
adjustment.
This facility represents the latest and most recent advance in the
existing mass burning technology. It is larger than any previous similar
installation and the operating steam-pressure and temperature, at 4.86 MPa/
468°C (600 psig/875°F), are slightly higher than normally used in most water
walled systems.
Precipitators--
Stack emission tests have been performed by both the Federal EPA and the
Massachusetts Bureau of Air Quality Control. In both cases the particulate
emissions were within the established legal limits, which indicates that the
boiler emissions are in the expected range and that the precipitators are
functioning as designed. Total solid particulates measured have been in the
range of 0.106 to 0.131 g/Nm3 (0.044 to 0.054 grains/SCF) corrected to 12%
17
-------
C02- The average value in the official EPA Method 5 tests was 0.119 g/Nm3
(0.049 grains/SCF).
Bridge Cranes--
The cranes have been variously quoted as having a 10.5 Mg (13 tons) and
a 2.7 Mg (3 ton) capacity. Both figures are correct. The higher rating is
the gross lift rating of the bridge and includes the weight of the grapple.
The lower rating is the actual weight of refuse that could be lifted by this
crane.
There has been some comment in the industry that there was severe wear
of the bucket hoisting cables when this facility went into operation. If
this is true, it is to be expected in a new installation where crane operators
have not yet gained the experience needed to operate cranes rapidly yet
smoothly. Further, cable life on cranes of this type is normally not very
long, particularly on the bucket closing cable. Cable replacement every two
months is not considered unusual.
Water Walled Systems--
The incinerator furnaces at Saugus are the mass-burning type, i.e., the
refuse is burned on a grate in the same condition as received, without pre-
processing. A comparison of the steam output conditions to the refuse input
indicates a design thermal efficiency of 71.4 percent. Shredded refuse would
have a higher volumetric heat release rate and a thermal efficiency somewhat
greater than this. It is apparent that RESCO has chosen to sacrifice a few
percentage points in efficiency in favor of using a well known and developed
technology.
The choice of mass burning boiler normally require the recovery of metals
to take place after combusion. RESCO does sell the ferrous fraction, but
aluminum is not recovered, as the owners have established to their satisfac-
tion that its recovery is not economically feasible at this time.
Reliability--
With the ability to fire oil as well as refuse in the main boilers and
the existence of two oil-fired standby boilers, the reliability of steam
supply to the customer is assured.
The facility has two waste burning boilers that will both be required in
service to meet the normal average waste load. One can act as standby for
the other in the event of an unscheduled outage. The boilers have been de-
signed carefully and conservatively, using the best information available,
so that unscheduled outages should be rare, but they can and probably will
occur. With this facility in service, the existing landfill, which could
be a back-up disposal facility, is closed. Short term outages of a few days
are amply provided for by the huge capacity of the refuse receiving pit,
which is capable of storing 5.6 days of refuse as it is received on the
average.
The cranes serving the boilers gain reliability by being installed so
that each can serve either furnace. If this should prove inadequate, there
is provision for the future installation of a third crane.
18
-------
Project Development
Although this refuse-to-energy project is based on technology previously
used and even though there are as many as 54 similar previous installations,
the Saugus facility is unique, just as every refuse-to-energy facility is
unique. Most times the conditions that create the need for such facilities
are different in every locality.
There are several basic conditions that nevertheless must be met by
every entrepreneur, whether public or private, in order to establish a
financially successful refuse-to-energy operation. Each of these basic
conditions, and how RESCO met them, are discussed below.
Preliminary Financing--
In order to initiate any sizeable refuse-to-energy project, it is nec-
essary to study the requirements, review the project feasibility, prepare
preliminary designs, and spend a great deal of time in discussion and con-
ferences with interested and involved parties. The initial impetus for this
effort were provided by the M. DeMatteo Construction Company, and General
Electric, with a later involvement by Wheelabrator Energy Systems Inc. in
the form of a joint venture of the two firms under the name of Refuse Energy
Systems Company (RESCO). The key element here was the fact that M. DeMatteo
Construction Co. was the owner of a major landfill, which was the primary
means of waste disposal for the majority of communities in the area, and this
landfill was to be closed for environmental reasons. In order to continue
providing the disposal service, DeMatteo sought other means of disposal.
This vital entrepreneurial effort was accomplished by private interests.
Although the real loss would have been suffered by the communities using the
landfill, they were spared the trouble and expense of the preliminary facility
planning. It is understood that RESCO, the joint venture, now has nearly
$10 million of equity invested in this facility.
A Guaranteed Source of Refuse--
The major investment required to construct a refuse-to-energy conversion
plant demands that the source of refuse, the raw material for the plant opera-
tion, be assured as to quantity and availability. This seems like an ele-
mentary precaution, but it is too frequently overlooked or lightly considered
in facility planning. Refuse collection in a city and surrounding areas is
not always under municipal control and frequently the function is performed
by many private firms with diverse interests.
RESCO obtained long term contractual agreements with ten nearby munici-
palities, including Saugus, to deliver their refuse to the new plant. This
furnished a guaranteed minimum amount of refuse for operation of the facility,
sufficient to justify proceeding with development plans. It also was an in-
dication of much more refuse that would be available to the plant on similar
contracts from other nearby communities, including parts of Boston. There
should be no problem in obtaining enough refuse to operate the plant at the
planned level and perhaps for a future expansion.
19
-------
An Assured Source of Revenue--
If a refuse-to-energy project is to be a financial success, i.e., pay
its own operating expenses and provide a return on capital, adequate revenues
must be derived. Prudent planning requires that there be some assurance of
an adequate level of revenue through some contractual agreement or arrangement
with prospective users of the plant's services. If revenue bond financing is
contemplated, reasonably firm agreements would be mandatory.
RESCO met the need for adequate revenue through (1) a drop charge to the
using municipalities for disposing of their waste, and (2) a firm contract to
sell steam to the General Electric Company, Lynn plant.
The refuse supply contracts with the neighboring municipalities contain
a drop charge clause that sets the initial charge at $14.33/Mg ($13 per tonj
and provides for escalation in accordance with an agreed upon formula. RESCO
cannot arbitrarily raise the price.
When the refuse-to-energy concept was first conceived for this area, the
General Electric Company was very much interested and supported the concept
because they had found it would be necessary to replace two of their existing
steam boilers. It was reasoned that it might be desirable to purchase steam
from an outside source such as RESCO rather than invest in two new boilers.
The result is a firm long term contract between G.E. and RESCO whereby G.E.
is committed to purchase a minimum of 907 Gg (two billion pounds) of steam
per year from RESCO. The RESCO plant is designed to produce steam at the
temperature and pressure required by G.E. The selling price is based on a
formula that includes an allowance for the probable plant investment, fuel
cost, and labor cost for the new replacement boilers, which G.E. now does not
have to buy. The price also contains a slight discount to G.E. to make it
attractive and assure continuation of the arrangement beyond the existing
contract.
The combined revenues from steam sales and drop charges, all firmly com-
mitted, are sufficient to operate the plant, make the lease payments to the
town of Saugus, pay taxes, and afford a profit to RESCO. There is also in-
come from the sale of recovered metals, mostly ferrous metal, but the finan-
cial success of the operation does not depend on it.
Technology--
No matter how well plans are made to receive sufficient refuse and pro-
duce an adequate revenue, no project can be successful if the chosen refuse-
to-energy technology fails to perform as planned. At best, such failure could
result in higher operating costs with reduced profit or even a loss resulting
in default on bonds. At worst, it could mean a total failure to function.
RESCO made certain that the processing technology would not cause trouble to
the project by adopting a thoroughly tried and proven method of extracting
energy from waste.
In this regard, it should be understood that Wheelabrator Energy Systems
Inc. holds the American Rights to the Von Roll technology. This automatically
determined the technology to be used for energy recovery early in the project
20
-------
development without expenditure of time and funds studying and comparing the
various technologies available.
If a new project is to be funded by revenue bonds, as this one was, it
is necessary that the energy recovery methods chosen not only appear to be
technically feasible but it must be demonstrably so to the bond-buying public.
Here too, the choice of proven Von Roll technology met the requirement.
Financing--
The financing of major resource recovery projects has been so frequently
a major stumbling block to any further development that one might expect this
aspect of project development to receive primary consideration. If the pre-
viously described development conditions, preliminary financing, source of
refuse, source of revenue, and technology, have been thoroughly considered
prior to any study of financing, the combined results should indicate clearly
whether or not the project is financially feasible. If it is, there are
several choices of financing methods available. Among them are general obli-
gation municipal bonds, revenue type municipal bonds, and, more recently, the
tax exempt revenue bonds that may be issued by an authority specially con-
stituted for the purpose of constructing environmental protection and waste
disposal facilities.
In this case, RESCO found that the project could be economically justi-
fied. The Town of Saugus created an Industrial Development Authority autho-
rized by Chapter 40D of the General Laws of the Commonwealth of Massachusetts,
which was empowered to issue industrial development revenue bonds on a tax
exempt basis for the purpose of constructing solid waste disposal facilities.
This authority initially issued $15,000,000 worth of short term bonds to
secure funds for construction. Later, as the project approached completion,
$30,000,000 worth of 20-year revenue bonds were issued with the proceeds used
to complete construction, retire the original $15 million bond issue and pay
back some loans made during construction.
The plant is owned, through the Authority, by the Town of Saugus, and
is then leased to RESCO. The amount of the annual lease payment is suf-
ficient to cover the interest and principal amount (debt service) on the
bonds. RESCO is required to maintain a bond reserve fund equal to two years
of debt service.
Financial Analysis
The RESCO facility has an installed capacity to process a maximum of
1361 MG (1,500 tons) of refuse per day with an average throughput conserva-
tively planned at 1088 Mg/d (1,200 tons per day). An estimate by the Rust
Engineering Company, the wholly-owned subsidiary of Wheelabrator-Frye who
was the designer, made in May of 1975, places the gross cost at $38,268,000.
This total consists of the following:
21
-------
Land $ 772,000
Site Development 1,302,000
Refuse Receiving and Handling 3,479,000
Refuse Combustion Building, Chutes, Grates, Furnaces, Etc. . 13,018,000
Stand-by Boilers 610,000
Pollution Abatement Equipment 2,119,000
Ash handling and disposal 1,899,000
Boiler Feedwater Supply and Treatment 565,000
Utility Bridge 1,609,000
Utilities and Miscellaneous 917,000
Direct Construction Costs $26,290,000
Indirect Costs 6,684,000
Real Estate Taxes 313,000
Construction Interest 2,099,000
License and Contracting Fee 1,455,000
Financing and Start-up Costs 1,427,000
Gross Requirement $38,268,000
At this gross cost and an average throughput of 1088 Mg (1,200 tons) per
day (397 Gg or 438,000 tons per year), the capital cost per Mg of daily plant
capacity is $35,156 ($31,890 per ton).
The plant was designed for future expansion and RESCO has made studies
that indicate that the increased amount of waste to make expansion possible
is available in the service area and there is also indication that G.E. could
accept the increased output of steam. If the plant is expanded by the addi-
tion of one more 680 Mg (740 ton) per day rated boiler (544 Mg/d or 600 TPD
avg), the gross cost is estimated by Parsons* as follows:
Land $ NC
Site Development 50,000
Refuse Receiving and Handling 260,000
Refuse Combustion Building 7,000,000
22
-------
Stand-by Boiler $ 325,000
Pollution Abatement Equipment 1,100,000
Ash handling and disposal 400,000
Boiler Feedwater Supply and Treatment 100,000
Utility Bridge NC
Utilities and Miscellaneous 100,000
Added Direct Construction Costs $ 9,335,000
Indirect Costs 2,373,000
Real Estate Taxes NC
Construction Interest 745,000
License and Contracting Fee 516,000
Financing and Start-up Costs 475,000
Total Added Costs (1975 Basis) $13,444,000
*This estimate and the one that follows are entirely the responsibility of
Parsons; the owners of the RESCO facility in no way contributed to the
estimate or commented on their accuracy.
If this estimated additional cost for increased capacity is added to the
$38,268,000 base cost, the total cost of a plant with an average daily capac-
ity of 1633 Mg (1,800 tons) is $51,712,000, and the capital cost per Mg of
daily plant capacity is $31,670 ($28,728 per ton). The reduction in capital
cost per ton from $31,890 to $28,728 for an increase of 50% in capacity is
quite modest, being only about 10%. The variation would be even less apparent
if the present facility had been designed without consideration for future
expansion. This suggests that at an average capacity of 1088 Mg (1,200 tons)
per day, this type of refuse-to-energy plant has already realized most of the
benefit of large size and that further increases can only result in nominal
saving in capital cost per unit of capacity.
An estimate of probable project cost for a plant with half the capacity
of the existing plant (544 Mg/d or 600 TPD average) is considerably more dif-
ficult to make because it involves changes in basic developmental costs as
well as the deletion of costs related directly to capacity. Nevertheless,
in order to illustrate the probable sensitivity of capital cost to plant size,
the following Parsons estimate of cost for an average 544 Mg (600 ton) per
day plant is presented. It is based, as are other estimates, on 1975 costs
and on the cost structure developed for the existing plant.
23
-------
Land $ NC
Site Development 300,000
Refuse Receiving and Handling 1,730,000
Refuse Combustion Building 6,400,000
Stand-by Boiler 325,000
Pollution Abatement Equipment 1,055,000
Ash Handling and Disposal 400,000
Boiler Feedwater Supply and Treatment 265,000
Utility Bridge NC
Utilities and Miscellaneous 200,000
Deleted Direct Construction Costs $10,675,000
Indirect Costs 2,714,000
Real Estate Taxes NC
Construction Interest 852,000
License and Contracting Fee $ 590,000
Financing and Start-up Costs 550,000
Total Deleted Costs (1975 Basis) $15/381,000
If this estimated cost deletion for reduced capacity is deducted from
the $38,268,000 base cost, the total cost of a plant with an average daily
capacity of 544 Mg (600 tons) is $22,887,000, and the capital cost per Mg of
daily plant capacity is $42,052 ($38,145 per ton). This is a substantial
cost increase per unit of refuse handled for a reduction to half of the capac-
ity of the existing plant, being about 19.6%.
It can be seen from a comparison of capital cost per ton for the three
plant sizes, i.e., 5hk , 1088, and 1633 Mg/d (600 TPD, 1,200 TPD, and 1,800
TPD), that the smallest plant suffers from the high cost associated with
small size, while the largest one indicates a diminishing benefit from further
size increases. A similar comparison should be expected in the range of
operating and maintenance costs.
24
-------
HAMILTON, ONTARIO
Introduction and Summary
The Hamilton Solid Waste Reduction Unit (SWARU) is owned and operated by
the Regional Municipality of Hamilton-Wentworth, which maintains offices at
Hamilton City Hall, 71 Main Street, Hamilton, Ontario, Canada. This facility
recovers energy from waste by producing steam from shredded refuse in a water-
wall boiler (incinerator), although there is currently no customer for the
steam. The design capacity is 544 Mg/d (600 TPD) of refuse and 95 888 kg/h
(211,400 Ib/hr) of steam. It is located on level land a short distance from
major industrial plants and the buildings are functional and neat in
appearance.
SWARU has experienced several problems that are discussed in detail with-
in this Section and immediately below. The important observation to be made
is that refuse combustion technology is established and can yield reliable
and economical conversion to energy when all conditions of proper design and
equipment selection are met.
Conclusions
• All waste-to-energy processing plants have rather high capital costs
per unit of capacity, and it is therefore essential to maintain the
design throughput. With problems in any one of the several proces-
sing steps able to markedly affect operating capacity, the design and
financing must be realistically accomplished as a total system.
SWARU was basically conceived of as a waste disposal system and dif-
ferent features would now be incorporated if it were to be designed
today as an energy recovery facility. The semi-suspension moving
grate combustion system used is rather unique for MSW and this plant
has verified the approach is a sound one. Location of a customer for
the steam produced would of course greatly improve plant economics.
• Compromises made in the materials handling equipment have been the
cause of part of the reduced capacity at Hamilton. Thorough testing,
preferably at full scale, should always be made when design changes
are added and equipment suppliers should be notified on all pertinent
characteristics of the waste and interface units. The materials
handling equipment at SWARU can be improved, as can the marginal air
pollution control efficiency, but installation of the proper equipment
originally is always far less expensive than retrofit costs.
Process Description
A schematic process flow diagram is presented as Figure 2, where a step-
by-step description of the process as it was originally designed is shown.
The changes and modifications to the process that have been made to bring it
to its present operating condition are discussed separately.
The process begins with the arrival of a refuse collection truck at the
automatic recording scales, Item 1 on the flow diagram. At this point, the
25
-------
AUTOMATIC
RECORDING
REFUSE
COLLECTION
TRUCK
SHREDDER (4 IN PARALLEL)
13.6 Mg/h (15TPH)
BALLISTIC
REJECT CHUTE
REJECT BIN
MAGNETIC SEPARATOR
BRIDGE ACROSS PIT
(PICKING STATION) / /
Y //
CONVEYOR BOTTOM
REFUSE RECEIVING PIT
FERROUS METAL
'X
SHREDDED REFUSE
SHREDDED REFUSE
544 Mg
(600 ton)
SHREDDED
REFUSE
STORAGE BIN
FERROUS METAL
ELECTROSTATIC
PRECIPITATORS
STEAM
GENERATOR
(2 PER BOILER)
u
BOTTOM
ASH SITTINGS
ASH SCREW CONVEYOR
AIR
ADJACENT ASH
LANDFILL
Figure 2. Schematic process flow diagram of the Hamilton Solid Waste Reduction Unit.
-------
load is automatically weighed and remotely recorded for plant operating
records and billing purposes. The operation is under surveillance from the
central control room by closed circuit TV, and further truck movement is
directed by a system of traffic signals, also controllable from the central
control room.
From the scales, the truck proceeds to the receiving building, entering
by either of two inclined ramps, and discharges its load into the conveyor
bottom receiving pit, Item 2. When unloaded, the truck leaves the building
via one of two exit ramps. Truck movement, except for backing and dumping,
is a straight-through "one-way" operation. In the event of an emergency,
a small amount of space exists at the truck level where loads of refuse can
be dumped directly onto the floor and later moved into the pit by a dozer
or front-loader. Such use of the floor is reserved for emergency situations
only, in order to hold down maintenance costs resulting from blades damaging
the special corrosion-resistant asphalt layer on the concrete slab.
The refuse in the pit is moved toward one end and up a steep incline by
four parallel metal pan conveyors, Item 3, upon which all the refuse rests.
The steep incline serves to limit the amount of waste on the conveyors as
they leave the pit, and, of course, the change in elevation is necessary in
order for the refuse to be transferred from the pit to the shredder.
As the refuse moves slowly up the incline, it passes under a bridge over
the pit, Item 4, from which one or two men using rakes and pitch forks smooth
out the flow and remove all excessively bulky wastes. The items removed are
deposited in two nearby roll-off bins and subsequently sorted for direct
salvage or landfill.
Each of the four pit conveyors delivers refuse to a 13.6 Mg/h (15 TPH)
vertical shaft shredder, Item 5. These shredders reduce the refuse to ap-
proximately 5 cm (2 in.) size or less. With four 13.6 Mg/h (15 TPH) shredders
in parallel, the total rated plant capacity is 54.4 Mg/h (60 TPH). It is
characteristic of vertical shaft shredders that some difficult-to-shred
materials such as iron parts and some rock will be struck by the first whirl-
ing breaker bar and rejected up the ballistic chute. This rejected material
is collected from each shredder in a small bin and taken to the main receiving
floor, where it is hand-sorted for salvageable metals, inert materials for
landfill, and combustibles to be returned to processing.
The shredded refuse from all of the shredders is collected on a single-
troughed belt conveyor, Item 6, and conveyed to a magnetic separator, Item 7.
The separate mechanism requires a new belt about every 3 months, necessitated
by occasional jams in the feed chute. After magnetic separation, the refuse
stream splits, with the separated ferrous metal (containing up to 5% organic
material) conveyed via a troughed belt conveyor, Item 8, to a roll-off bin
outside the building. The remaining shredded material, which can now be
considered as fuel, is conveyed on a troughed belt conveyor, Item 9, outside
the processing building to the storage bin, Item 10-
The shredded fuel is accumulated in the beehive-shaped storage bin, re-
claimed automatically by built-in equipment, and conveyed via trough belt
27
-------
conveyor, Item 11, to the two Babcock and Wilcox "Stirling" waterwall steam
generators, Item 12. Not shown is a flow divider, provided as part of the
fuel conveying system inside the boiler house to proportion the fuel flow to
the steam generators. The present flow divider is being rebuilt to allow
manual control for the fuel flow. This will improve control over the volume
of refuse to each generator.
Flue gas leaves each steam generator at a nominal 286°C (547°F) and
passes through a two-field Lurgi-design electrostatic precipitator, Item 13,
before exiting to atmosphere through a 50 m (165 ft) high stack, Item 14.
Incomplete combustion of flue gases during start-up or when burning exces-
sively wet refuse tends to increase visible stack emissions and the increased
emissions persist until proper operating temperatures are reached in the
boilers. Steam is generated at 1.82 MPa (250 psig), saturated. Some of the
steam is used to drive plant auxiliaries such as conveyor drives, but most of
it it condensed in air cooled condensers on the roof and the condensate is
returned to the boiler. SWARU reports that the air cooled condenser tubes
have recently been leaking profusely. The problem is under investigation,
and a solution is being sought. Continuing leakage results in increased costs
for water conditioning chemicals, and can also lead to serious structural
problems on the roof when the flow of water causes ice masses to form.
Bottom ash and siftings from the boiler grate are designed to be removed
to a silo, Item 16, by a steam-driven pneumatic transport system, Item 15.
The large volume of the silo is intended to slow the moving air and thereby
allow the ash to drop out before the air exits to the atmosphere. The fly
ash silo is currently not in use because (1) the fly ash and bottom ash differ
so much in their characteristics that common removal becomes very difficult;
and (2) the fly ash is so fine that it does not settle in the silo, but es-
capes and disperses in the open air. A report is being prepared for modifica-
tions to the in-plant ash collection system in order to make it operative.
The last item in the process chain is a belt conveyor, Item 17, which
removes ash from the silo and conveys it a short distance to an adjacent
landfill. It is estimated that the landfill can be used until late 1977.
After that, the ash residue will have to be disposed of elsewhere. Additional
uses or means of disposal are under investigation.
Processing Equipment
Refuse Receiving Pit--
This concrete pit, built as part of the building structure, is 12 m wide
by 30 m long by 9 m deep (40 x 100 x 30 ft). The top level of the pit from
which the trucks dump their loads is about 4.6 m (15 ft) above the level of
surrounding terrain.
Conveyors (Item 3)--
The bottom and one end of the refuse receiving pit are completely covered
by four parallel metal slat conveyors, each powered by a 11.2 kW (15 HP)
motor. A 18.6 kW .(25 HP) motor was installed to alleviate a problem of stall-
ing of the conveyors under full load. Its performance proved satisfactory
and three 22.4 kW (30 HP) motors have been ordered to replace the remaining
28
-------
15 HP units. The bottom horizontal sections merge into the inclined,
45-degree end section without any break. Therefore, the bottom part, which
may have a 9 m (30 ft) depth of waste on it, moves just as fast as the in-
clined end part, which has only 30 or 60 cm (1 or 2 ft) of waste on it.
Shredders--
The Item 5 shredders are 149 kW (200 HP) vertical shaft machines rated
at 13.6 Mg/h (15 TPH). These are Tollemache machines sold in the United
States and Canada by the Heil Company; they are basically vertical hammer-
mills with a part of the mill set to relatively close clearance for some
grinding effect. All waste is shredded in a single pass to less than 5 cm
(2 in.) size. There is no secondary shredding.
Other Conveyors--
With the exception of the receiving pit conveyors and the pneumatic ash
removal system, all conveyors used in the facility are heavy duty troughed
belt conveyors built on truss-type frames. None are fully enclosed for dust
control, but all have some form of hood or enclosure. Conveyor belt widths
are appropriate to the quantity of material to be handled. Some of the con-
veyors are steam turbine powered, using the available plant steam.
Shredded Refuse Bin--
This patented unit has a storage capacity of 544 Mg (600 tons), equiv-
alent to 24 hours full-rated operation of the steam generators. The beehive
shape and 21.3 m (70 ft) diameter circular base of this steel bin are in-
tended to prevent any bridging of material. Serious bridging and binding
problems have occurred, however, in the Atlas tank. A proposal by the manu-
facturer suggests that a hollow center cone and a swinging feed chute at the
top of the tank would solve this problem. The proposal is under consideration.
The Atlas reclaiming system utilizes a heavy duty chain around the perim-
eter of the bin, held in place on a recessed track, and driven at a con-
trollable variable speed. This chain, in turn, tows three or more chain and
bucket assemblies that drag on the floor of the bin and tear at the toe of
the piled material. The dragging action of the bucket assemblies has produced
more wear and tear on the concrete floor of the bin than anticipated, a 25 cm
(10 in.) deep groove in the floor thus far. The material drawn from the pile
is dragged across the bin bottom to a grating through which the material falls
to a conveyor below. This conveyor discharges to the transport conveyor,
Item 11, which carries the shredded fuel to the boilers. Metering of fuel
output is accomplished through controlling the speed of the chain and bucket
assemblies.
Boilers--
There are two Babcock and Wilcox "Stirling" power boilers, Item 12, each
rated to burn 272 Mg (300 tons) of shredded refuse per day, producing 47,944
kg/h (105,700 Ib/hour) of steam at 1.82 MPa (250 psig), saturated. This
rating and a 71 percent actual design efficiency are based on an assumed
higher heating value of the refuse of 13.96 MJ/kg (6,000 Btu/lb). The use of
shredded refuse in a water walled boiler of this type allows efficient opera-
tion with only 37 percent excess air. The application of an electrostatic
precipitator for particulate removal becomes practical and comparatively
29
-------
economical in conjunction with this reduced quantity of flue gas and the use
of a small economizer to reduce flue gas temperature to 310°C (590°F).
Figure 3 is a schematic section through this type of boiler as applied
to burning shredded refuse. The shredded refuse enters the boiler house from
the storage bin via a conveyor near the top of the boilers. At this point,
a proportioning device distributes the refuse to one or both boilers, as
required, and the refuse thus proportioned is further divided among three
vertical chutes on the face of each boiler by a swinging spout device. The
refuse drops by gravity to a point just above the grate and there is blown
into the fire box by a blast of air. The lighter parts of the shredded
refuse burn in suspension very quickly, while the heavy parts and inert mate-
rials fly to the opposite side of the fire box and land on the grate. These
heavier parts are carried on the slowly-moving grate toward the front of the
furnace, the speed being adjusted to allow time for burnout with a relatively
thin bed of burning material. After final burn-out, the ash drops into the
bottom ash pit. A portion of the burned material from the grate sifts through
it and drops into the combined air plenum and ash hopper directly below, but
this normally contains little remaining combustible matter.
The over-fire feeding of shredded refuse with a high percentage burned
in suspension makes it possible to maintain a relatively constant furnace
temperature and prevent hot spots and slagging. The suspension burning plus
the resulting thin bed on the grate allows efficient operation with minimum
excess air without creating a reducing atmosphere near the tubes. Because of
these advantages, it is not necessary to provide refractory protection of the
lower boiler tubes and a steadier steam rate can be maintained. The design
efficiency of this boiler installation, even at the comparatively low operat-
ing pressure, is 71 percent. Unburned light fractions of refuse, especially
paper, have been found in the precipitated fly ash. A proposal has been sub-
mitted for the installation of auxiliary air jets to provide over-fire air in
a vortex effect, which would serve to ensure complete combustion of materials
in suspension.
The remainder of the steam generating installation is essentially a
standard water tube boiler, except that the use of refuse as a fuel requires
a somewhat larger installation than would be required to achieve the same
output rating with a fossil fuel.
Air Pollution Control--
The electrostatic precipitators are Wheelabrator-Frye "Lurgi" design
units of the two-field type, designed for inlet conditions of 38.2 m3/s
(81,000 ACFM) of flue gas at 307°C (585°F), with a particulate loading of
0.533 wt-% (on a volumetric basis, 3.18 g/m or 1.5 grains/SCF) corrected to
50% excess air. Design outlet loading is 0.008 wt-% (0.05 g/m3 or 0.0225
grains/SCF), for an efficiency of 98.5 percent. Specific collection area is
0.67 m2/m3 (215 ft2/1000 ft3) of gas.
30
-------
BYPRODUCT
STEAM
SHREDDED
REFUSE
Figure 3. Schematic section of refuse-fired boiler at the
Hamilton Solid Waste Reduction Unit.
31
-------
Existing Operation
Actual Capacity Realized--
A matter of some concern initially was the probable performance and
reliability of the boilers, but these units have thoroughly proven themselves.
There has been no evidence of boiler tube corrosion and very little need for
maintenance shutdown. The grates need an occasional cleaning to remove grit
from the joints, but this has not been a serious or frequent problem. The
boilers have demonstrated their capacity to reach design load, or even a
moderate overload, while burning refuse only. At least 50 percent of burning
is in suspension, according to the operators, though no test figures are
available. The total design capacity of the plant has not been realized,
except for short periods, because of various materials handling problems
discussed below.
Receiving Pit--
The conveyor bottom receiving pit is only partially successful. If the
pit is filled or near full, the weight of refuse on the conveyors is too great
, and the driving motors stall. The replacement of these 15 HP motors with
30 HP motors should solve this problem. An additional conveyor problem is that
they must be stopped due to operational problems. When restarted, the con-
veyors may run freely under the waste, and the bridging must be broken with
an electric hoist mounted over the pit.
At present, refuse is dumped only into the one-quarter to one-third of
the pit nearest the incline feeding the shredders. This reduces the load and
the conveyors are able to function properly. The steep incline at the end of
the pit apparently acts as a metering device to limit the depth of refuse on
the conveyors as they feed the shredders. The inability to maintain a pit
full of refuse makes it impossible to extend operation of the shredder much
beyond the normal working day of the refuse collectors without storing more
refuse at the truck level than was intended. This is one limitation on total
plant capacity.
Shredders--
The use of four 13.6 Mg (15 TPH) shredders, which cannot accept over-
sized waste, rather than a single large 54.4 Mg (60 TPH) shredder capable of
accepting all wastes, was based on an economic decision that the bulky mate-
rial could be separated and land-filled for less total cost than the dif-
ference in cost between the four small shredders and one large one. There
was also a desire to use equipment available in Canada, as these Tollemache
machines were. Data that would prove the merit of this decision are not
available, but there are some design and operating problems that apparently
result from this decision. These are:
• The use of multiple shredders demands some form of multiple feed such
as the conveyor bottom pit, which is not functioning properly at
present.
• The smaller shredders require a much closer control of feed to prevent
overloading and to assure that large items do not slip through.
32
-------
• The shredded refuse from these machines and of the bottom ash from
the boilers are observed to contain some refuse items that are hard
to shred, such as tough rubber products and leather-like plastics.
These seem to get through in much larger pieces than the supposed
5 cm (2-in.) maximum. Whether this is a fault of the machine size,
type, or maintenance is not known.
• In a refuse-shredding operation where ferrous metals are to be re-
covered by a magnetic separator, the ballistic reject feature of the
vertical shaft shredder serves no useful purpose. It only causes a
labor problem in moving and sorting the bins of rejects.
Shredded Refuse Storage Bin--
The greatest single cause of reduced capacity, lost manhours, and general
uncertainty of operation has been the shredded refuse storage bin. Although
the Atlas bin reclaim system has reportedly performed satisfactorily in other
installations, it is not functioning properly here. This apparent contradic-
tion prompted an investigation to learn what the faulty operation really was,
and, if possible, why it was faulty. The major problems that have occurred
are outlined below, followed by a brief history of the engineering, fabrica-
tion, and modification of the storage system. A comparison of the two will
clearly show why the faults occur. Some of the problems are:
• The packed refuse at the bottom of the bin is very difficult to scrape
free and the reclaiming apparatus is unable to handle it, except at a
slow rate.
• If the reclaiming apparatus is speeded up to gain more output, a chain
failure is likely.
• If, for any reason, the shredded refuse is allowed to remain in the
bin for more than 24 hours, it begins to decompose. This adds to the
tendency to pack and the consequent difficulty of removal. In addi-
tion, this change in refuse characteristics is reported to increase
the density of the material, rendering it more difficult to handle
and burn. Most importantly, the incineration of such refuse appears
to result in higher stack emissions.
• There has been a tendency for the shredded refuse to bridge in the
bin. This has nothing to do with the shape of the bin; the shredded
refuse tends to pack down and bridge within itself when attempts are
made to extract it. The sweep buckets compress the material in an
inward radial direction, resulting in formation of a hard core in the
center of the tank. In addition, lighter materials, such as paper,
tend to be more readily removed by the sweep buckets while other
materials are not removed. These remaining materials become compacted
and interwoven, effectively forming a single semi-solid mass.
• The operators report excessive floor and bucket wear.
When the reclaim system was under design, the vendor was provided with
samples of shredded refuse upon which to base his design. These samples
33
-------
indicated a maximum particle size of about 5 cm (2 in.), but in practice this
size is more likely an average, with much material such as rags, plastic,
rope, wire, and the like considerably oversize or hardly shredded at all. The
vendor states that had he been aware of this, the design would have been dif-
ferent and would perform much better in extracting waste from the pile. The
larger pieces, particularly fibrous materials such as rags, tend to remain
behind and bind the remaining shredded material into a tightly packed mass.
In an attempt to reduce the capital investment, a steel center cone that
directs waste toward the periphery of the bin and prevents buildup of a hard
central core, was omitted. The hard central core does built up, as predicted,
and must be removed periodically by hand.
The designers of the plant wanted to drive as much of the equipment as
possible with steam and in order to do this with the bin reclaim system, the
plant designers provided a dual steam turbine drive. This replaced the so-
phisticated, dual DC motor, variable speed, load-sharing drive and control sys-
tem normally provided by the vendor. Turbine controls have never functioned
properly to share the load between two turbines, and speed control is possible
only over a limited range, rather than the 20 to 1 range normally provided.
When load sharing between the two drive turbines proved to be impossible,
one turbine was disconnected and the reclaim system is now driven by only one.
This more than doubles the load on the turbine and the pull of the chain.
This is an obvious explanation for the frequent chain breakage.
There is no explanation for the reported excessive wear on the floor and
buckets other than the fact that the waste contains a certain amount of glass,
dirt, and rocks, and is therefore likely to be quite abrasive.
Although the Hamilton area experiences cold winters, freezing of the
material in the bin has not been a problem. This is possibly due to the good
insulating properties of the shredded material and to the internal heat
generated by decomposition.
Ash Removal--
Both bottom ash and fly ash were originally planned for removal by the
same pneumatic system. No part of this system is presently in use. The
problems were:
• The bottom ash frequently contained large pieces of refuse that had
not been shredded to specification and these oversize pieces plugged
the pneumatic transport pipe. At present, a belt conveyor is used
to remove bottom ash from a small quench tank. This belt dumps the
ash on the ground where a bulldozer pushes it onto a landfill area.
• The air velocity in the pneumatic system was too low to entrain and
transport the heavier ash particles.
• There was no cyclone separator or bag house to trap the lighter
particles of ash, so these particles, mostly from the fly ash, did
not settle out in the ash silo as planned, but were exhausted to the
atmosphere. Several methods of fly ash removal have been tried, but
34
-------
without marked success. This remains an annoying problem, but it
does not affect the overall capacity of the plant.
Electrostatic Precipitators—
There is reason to believe that the specification on the precipitators
would have to be more stringent for future installations. There is no data
available on actual stack emissions, but observations by Parsons employees
and others indicate more visible emissions than could be tolerated on new
installations.
Ferrous Metal Recovery--
The ferrous metal recovery equipment is adequate and working well. There
is no available data on the percentage of recovery achieved, but the actual
quantity of metal recovered is greater than had been expected on analysis of
refuse during the planning phase of the project.
Dust Control--
Most of the conveyors and mechanical equipment have some form of hood
or cover, but there is no real dust control system. The result is that the
plant is dusty and requires frequent cleanup. A part of the dust problem can
be attributed to the ineffective fly ash removal system.
Engineering Comments
In the foregoing sections, the process and plant equipment were described
and the operation was discussed. Most of the faults and problems are suscep-
tible to correction and this facility may yet become an efficient refuse-to-
energy operation.
The operating problems have not prevented utilization of the plant, but
have, forced it to operate at reduced output and with the expenditure of more
manpower than would otherwise have been necessary. Each of the problems in
operating this facility is discussed in detail below for the purpose of learn-
ing from this experience and to suggest some possible improvements or
corrections.
Pit Conveyor Failures--
The inability of the pit conveyors to function properly when the pit is
well filled seems to be caused by the fact that the bottom section, which is
in one unit with the inclined section, must operate at the same speed as the
inclined section. This forces it to scrape under the deep pile of refuse
because the pile cannot move as fast and also tends to ram the refuse very
hard against the inclined conveyors, such that the former could be much slower
than the latter. The ratio of speed could range from 1 to 3 to 1 to 30. With
this one change, admittedly not an easy one, it should be unnecessary to re-
place all conveyors with heavier units.
The operators of the plant, however, do not believe that slowing the
bottom conveyors with respect to the inclined conveyors would work properly.
Instead, they feel that the heavier duty 22.4 kW (30 HP) motors, will provide
the power necessary for successful conveyor operation.
35
-------
This concept of a small refuse receiving pit could be a very successful
design if some of the suggested improvements were made in the conveyor system.
These same improvements have been incorporated in conveyor designs and in
published promotional material by the original vendor of the conveyors
(Ref. 1).
Shredders--
The four relatively small shredders have not, in themselves, been a
problem, but their use forces the operators to provide personnel to pick bulky
waste and difficult-to-shred items from the waste. The income from ferrous
metals salvaged directly from the pit is insufficient to support the additional
manpower required, but this activity is required to protect the shredders
from potential damage- A system of selective refuse collection is presently
under consideration, in an attempt to minimize this problem. The ballistic
reject feature of these small shredders seems of doubtful value.
The most disturbing thing observed concerning the performance of the
shredders is the excessive amount of material that passes through the machine
with little or no shredding. The larger unshredded pieces tend to increase
the percentage of unburned combustible in the bottom ash, causing some dif-
ficulty in ash handling, and creating a problem in the operation of the
shredded refuse storage and reclaim system.
This problem possibly can be rectified by improved shredder maintenance
and by a closer setting of the machine for a finer shredded output, but it
seems probable that the only dependable correction would be to replace the
existing shredders. Normally, shredding to a maximum of 5 cm (2 in.) is
accomplished in two stages. In this case, the 544 Mg/d (600 TPD) capacity
at the desired size could probably be reached by a single large shredder
operating for two shifts. In effect it would be 54.4 Mg/h (60 TPH) shredder
de-rated by one-third to accommodate the fine shred. The type of machine is
open to consideration, but the trend seems to be toward the horizontal
hammermill.
It would be possible to modify the existing operation to install such a
machine, but it would be difficult and costly to do so. It would be necessary
to weigh the potential benefits very carefully.
Although setting the shredders to produce a finer output may, to some
extent, remedy some existing problems, this would also significantly increase
the maintenance requirements of the shredders. The use of a single large
shredder would appear attractive until one considers the additional equip-
ment, such as hoists, required to service such a unit. In addition, a break-
down of the only shredder in the plant could cause operations to cease until
the problem was remedied.
Dust Control--
The plant has no effective dust control and the result is a poor working
environment. There does not appear to be any single piece of equipment or
operation that produces large quantities of dust. All of the conveyors are
covered or hooded, as are the shredders, but there is no means of drawing off
the dust that accumulates inside the enclosures.
36
-------
A dust intake hood could be provided over the pit and all conveyor and
shredder housings could be placed under negative pressure to draw off dust,
but the required equipment would cause this to be a relatively expensive plant
improvement. The lack of dust control does not interfere with operation of
the plant, but there is some suspicion that it is a contributing factor to a
fairly high absentee rate.
The Shredded Refuse Storage Bin--
There does not appear to be any easy, inexpensive modification that could
be made to improve the operation of the shredded refuse storage and reclaiming
system. Since the problems with the system seem to stem from the fact that
some elements of the storage and reclaiming system were never installed, the
obvious solution is to complete the installation to the manufacturer's de-
signs and recommendations. This means installing a center cone; installing
the complete two-motor, DC, variable speed drive and control system; and, if
possible, revising the bucket and chain reclaiming design to improve per-
formance with the larger size shredded material now being processed. All
this would be quite expensive, but, if the plant were ever to be able to meet
a steam demand, it would have to be done.
Selling steam from this plant was not considered to be an essential
feature. The SWARU was conceived strictly as a refuse burning facility. At
present, the storage and reclaiming system has been by-passed and the shredded
refuse is conveyed directly to the boilers to be burned.
Boiler Performance--
The boiler performance has been very satisfactory, with a relatively low
demand for maintenance. It is unfortunate that there have been no tests to
verify boiler efficiency and to measure the relative amounts of waste burned
in suspension and on the grate. Even if manpower and funds were available to
make such tests, the program would have to await improvements in the fuel
handling system so that sufficient control of fuel supply over an extended
testing period could be maintained.
The plant has operated since the summer of 1972, and in 1975 the total
throughput was 43 Gg (48,000 tons), only about 22% of design capacity. This
might seem to be very light duty and therefore easy on the boilers, but the
material handling problems have caused many shutdowns and startups, severe
service for any boiler. There is little evidence of any corrosion that might
be expected to result from this.
The boilers use a gas burner to maintain a dependable steam supply for
steam-driven auxiliaries, and the burner continues to operate at a low level
while burning waste. With this type of boiler in continuous operation on
shredded refuse, it should be possible to shut off the gas burner completely
without serious consequences. Much depends upon a dependable continuous
supply of shredded waste fuel.
Ash Disposal--
The disposal of ash from boiler operations remains a problem, though it
does not affect the performance of the plant. The pneumatic system cannot be
used to remove bottom ash unless there is some change in shredding technique
37
-------
to assure size reduction of all waste material to a maximum size equivalent
to the handling capacity of the pneumatic equipment. The existing bottom ash
handling modification involving a small quench tank, belt conveyor, and dis-
posal area seems to be the best system under the circumstances.
The removal of fly ash could be accomplished by the pneumatic system with
some modifications and improvements. These are:
• Air velocity must be increased, to be certain to entrain all particles
of ash and keep them moving in the pneumatic tube.
• Equipment must be provided to separate the ash from the air at the
storage silo. To meet stringent air quality rules, this should con-
sist of a small diameter, high efficiency cyclone and an automatic
shake-down bag filter. The use of the cyclone ahead of the bag filter
will keep the filter size down and none of the equipment need be large
and costly. The required air flow is not very high.
• The fly ash system will probably require some simple form of cycling
control to sequentially discharge each hopper into the system.
• The use of a vacuum type pneumatic system, rather than a pressure
type, will ensure that the ash does not escape, but remains in the
system. This should materially improve the cleanliness of the entire
plant.
None of the above suggestions solve the problem of what to do with the
dry fly ash after it is collected in the silo. The silo is on the edge of
the landfill, and the ash is currently dumped directly on the landfill area.
As the available landfill reaches capacity, consideration will have to be
given to trucking the ash away. There is a belt-type conveyor in place for
transporting the ash to the fill and it is understood that the operators have
tried introducing water at this point to hold down dust, but the resulting
mix is too heavy for the belt conveyor.
Environmental Considerations--
The environment is affected by stack emissions, blowing ash, and truck
traffic into and out of the plant. The ash problem has already been discussed
in detail and the truck traffic is no real problem because the plant is in a
rather remote industrial area. There is no apparent sound or noise from the
plant. On occasion, however, a whistling was heard outside the plant when
boiler pressure dropped. A change in the pressure relief valve solved the
problem. The only external indication of plant operation is a very minor plume
from the stack. During an inspection by Parsons of the facility, all four
shredders were in operation, all conveyors were operating, the storage and
reclaim system was working, and one boiler was in service. The operation of
the shredders was barely detectable in the offices no more than 30 m (100 ft)
away in the same building.
In spite of the existence of a precipitator with a rated efficiency of
98.5 percent, there were some visible emissions from the stack. This would
38
-------
not be acceptable performance in any new installation. There is no apparent
reason for the visible emissions based on design ratings and assumptions, but
there have been no tests to determine whether or not the particulate emission
from the boilers is within the range for which the precipitators were designed.
The operators report numerous problems with the precipitators, including
clinkering, plugging, and plate distortion. The reasons for these problems
are unknown, but certainly the erratic boiler operation must be a contributing
factor.
Steam Turbine Drives--
There are many pros and cons on steam turbine drive for plant equipment.
Their use may seem attractive from the standpoint of reducing reliance on
expensive external fossil fuel-generated power, but it must be realized that
maintenance will be higher and control is not as easily accomplished as with
electric drives. The operators of this plant are continuing to use the steam
turbine conveyor drives provided with the plant design and do not have any
adverse comments. The shredders are not steam-driven, but unless the large
power-consuming devices, such as shredders and boiler draft fans, are steam
driven, there will be little saving in electric power cost. The serious
problem created by steam turbine drives on the storage bin reclaiming system
has already been discussed.
Auxiliary Fuel—
This plant uses gas as an auxiliary fuel for start-up, for keeping
boilers hot during periods of low refuse availability, and to maintain a
reasonably steady boiler operation. The reported fuel costs for 1975 in-
dicate a gas consumption equivalent of approximately 10 percent of the full
boiler rating. This seems quite high, but is undoubtedly due to the erratic
performance of the refuse fuel handling and storage system. When this system
is improved, gas consumption will certainly be reduced.
Steam Sales—The Hamilton Incinerator Project (SWARU) was developed with
the intent that eventually the steam produced might be sold for revenue, but
the project proceeded without any specific steam customers. To date, there
has been no attempt to sell steam, and no distribution system exists. For
this reason, the plant has never been called upon to demonstrate its ability
to match steam output to a demand. This is probably fortunate, because the
poor performance of materials handling and storage systems would have made
good control of steam output exceedingly difficult. Nevertheless, the per-
formance of the boilers with the shredded waste indicates that good demand
matching should be practical without excessive use of auxiliary fuel, pro-
vided that the fuel handling problems are solved.
Financial Analysis
A meaningful financial analysis is difficult to prepare since steam has
not been sold and the materials handling problems have prevented operation at
an efficient level. If these two important factors are entered into the
analysis by means of assumed figures, then some indication of the financial
potentialities of such a project might be obtained.
39
-------
Capital Costs--
The owners report a capital cost of $9 million when the plant was com-
pleted and went into operation in mid-1972. If an identical plant were to be
placed in operation in early 1976, the capital cost would have escalated by a
factor of 1.6 to about $14,400,000.
As previously stated, it is unlikely that another plant would be built
today to this same design; however, it is believed that improvements in
shredding, materials handling, shredded refuse storage, ash handling, and
dust control could all be realized for about one million dollars. These
changes, though quite costly, would so drastically improve the operation as
to pay for themselves in a short time.
A further important consideration in evaluating the financial feasibility
of such a plant is the potential revenue from the steam produced. In order
to make this revenue possible, it would be necessary to construct a steam dis-
tribution system to transport the steam to potential customers. There are no
real data on this possibility, but there is some indication that, when a
reliable steam supply can be assured, customers for the steam output can be
developed within no more than 3.2 km (2 miles) of the plant. Such a steam
distribution and condensate return system is likely to cost three to four
million dollars at 1976 prices. For the purpose of a conservative financial
analysis, a $4,000,000 figure is assumed.
The total 1976 cost for a fully operating 544 Mg/d (600 TPD), water
walled, revenue-producing incinerator in the manner of Hamilton can be derived
as follows:
Original 1972 Cost $ 9,000,000
Escalation to Early 1976 5,400,000
Plant Improvements 1,000,000
Steam Distribution System 4,000,000
Total 1976 Cost $19,400,000
If the owners and operators of Hamilton make the necessary improvements
and install a steam distribution system, their total investment will be as
indicated above, less the escalation figure, or $14,000,000.
Operation and Maintenance Costs--
As might be expected, the erratic operation of the plant has resulted in
an apparently high requirement for manpower to operate it and some unneces-
sarily high expenses. The total operating staff during 1975 numbered 51 and
was organized as follows:
Manager: 1
Chief Engineer 1
Maintenance Clerk: 1
40
-------
Receptionist: 1
Shift Engineers (Class II): 4
Shift Engineers (Class III) : 4
Shift Engineer Helpers: 4
Floormen: 5
Laborers 21
Mechanics 3
Electricians: 2
Maintenance Helpers: 3
Welder: _1
TOTAL 51
The poor operation of the materials handling and ash handling equipment
shows in the large number of laborers, floormen, and maintenance people re-
quired. It is likely that with equipment improvements, the plant could reach
full capacity with no additional help and might perhaps operate efficiently
with fewer people.
The approximate 1975 operating and maintenance costs are as follows:
Labor $605,000
Water 12,000
Power 42,000
Fuel 203,000
Chemicals 6,000
Equipment Maintenance 236,000
Building Maintenance 13,000
New Equipment 55,000
Equipment Rental 28,000
Miscellaneous 13,000
Debt Repayment 904,000
TOTAL $2,117,000
41
-------
This total operating and maintenance cost was generated while disposing
of a total of 43.5 Gg (48,000 tonsj of solid waste. This is a disposal cost
of $48.62/Mg ($44.10/ton), a very high figure. It is offset only by a rela-
tively small revenue from the sale of recovered ferrous metals. This will be
discussed in a following section.
At 85 percent of design capacity, this plant should be able to dispose
of 168.86 Gg (186,150 tons) of solid waste per year. With this rate of
throughput, the operating and maintenance expenses would be increased, but
not in proportion. An estimate of operating cost at the higher throughput is
indicated below.
Labor Unchanged $ 605,000
Water Increased proportionately .... 46,538
Power Increased proportionately .... 162,881
Fuel Doubled .403,000
Chemicals Increased proportionately .... 23,269
Equipment Maintenance . . No Change 236,000
Building Maintenance . . No change 13,000
New Equipment Accounted for in added debt ...
Equipment Rental .... Doubled 56,000
Miscellaneous Doubled 26,000
Debt Repayment Existing debt - no change .... 904,000
Debt Repayment Plant Improvements at
$1,000,000, 11-1/2%, 10 years . . 175,377
TOTAL EST. 0 § M $2,649,065
At the annual potential throughput of 168.86 Gg (186,150 tons) and the
slightly increased 0 § M cost, as delineated above, the disposal cost drops
to $15.69/Mg ($14.23 per ton) of solid waste before taking any credit for
possible revenues. This forcefully demonstrates how seriously the faulty
operation of a few items of equipment has affected the cost of waste disposal.
The largest single item of 0 § M expense is the repayment of debt. At
43.5 Gg (48,000 tons) per year, this accounts for a disposal cost of $20.76/Mg
($18.83 per ton), while at 168.7 Gg (186,000 tons) per year it is only
$5.36/MG ($4.86 per ton), a substantial difference. This alone illustrates
how much benefit could be obtained by an added expenditure of funds to secure
improved plant operation and a greater throughput.
42
-------
Net Operating Cost--
The annual 1975 operating and maintenance cost was $2,117,000, which is
equivalent to $48.62/Mg ($44.10 per ton) of waste handled. This can only be
reduced by the $1.10/Mg ($1.00 per ton) revenue received from ferrous metal,
or a net of $47.51/Mg ($43.10 per ton). This is the present situation.
If the improvements suggested are made to increase plant utilization to
85%, the estimated 0 & M is $2,649,065, and at the higher throughput the
0 § M cost drops to $15.69/Mg ($14.23/ton). In order to gain income from the
sale of steam, the estimated $4,000,000 steam distribution system must be con-
structed. At 11-1/2 percent for 20 years, a normal rate of interest in Canada,
the annual debt repayment on this new increment of construction would be
$518,819, or $3.08/Mg ($2.79 per ton) of waste handled.
Funding of Investment--The investment in the facility of $8,900,000 was
funded by the sale of debentures at varying interest rates and with periods
of 10 and 20 years. Interest rates varied from 7-3/4 to 8-5/8 percent, with
the average about 8-1/2 percent. The annual debt repayment for 1975 was
$904,000. This is about an interest rate of 8-1/2 percent over a 20 year
period.
Revenues--At the present time, the only revenue available is from the
sale of recovered ferrous metal. The rate of recovery is running at about
4 percent and the market price is at $27.56/Mg ($25 per ton). On this basis,
the ferrous metal revenue amounts to $1.10/Mg refuse ($1.00 per ton). This
rate does not change with throughput.
The potential revenue from sale of steam would be tied to the probable
cost of steam generation in customers' boilers, using fossil fuels and a
comparison factor. Fossil fuels in Ontario presently cost about $1.48/GJ
($1.40 per million Btu) when purchased in quantity. Taking into account
small boiler efficiencies, investment and other operating expenses, potential
customers would certainly find their steam costs running close to $3.00 per
454 kg (1000 Ib) (1.055 GJ or 1,000,000 Btu). A $2.75 figure will be used
here for a conservative analysis, although a figure higher than this could be
economically justified.
At the assumed $2.75 rate, this represents a potential gross revenue
from steam sales of $21.79/Mg ($19.77 per ton) of waste.
The total possible revenue is:
Ferrous metal $1-10 $1-00
Steam 21.79 19.77
TOTAL $22.89/Mg $20.77/ton
Waste Waste
Each of the two boilers is rated at 272 Mg (300 tons) per day and 48 045
kg (105,700 Ib) of steam per hour. If an 85 percent use factor is employed
43
-------
here, as before, then 0.907 Mg (one ton) of waste will produce 3267 kg
(7,188 Ib) of steam.
The total 0 $ M cost therefore becomes $18.76/Mg ($17.02 per ton) of
waste. This compares to a gross revenue of $22.89/Mg ($20.77 per ton) or a
profit of $4.13/Mg (3.75 per ton).
If, for any reason, the plant does not reach the full 85 percent utiliza-
tion suggested here, the 0 5 M cost per ton will increase while the revenue
per ton remains the same. The "break even" point is at a utilization of
138.353 Gg (152,522 tons) per year, or a factor of 69.6 percent.
44
-------
SECTION 4
NASHVILLE THERMAL TRANSFER CORPORATION
INTRODUCTION AND SUMMARY
The Nashville Thermal Transfer Corporation, a "Not for Profit" Tennessee
corporation, owns and operates a mass burning, refuse-fired incinerator in the
city of Nashville that produces steam and chilled water for a district heating
and cooling system for 30 buildings. The corporation, commonly referred to as
THERMAL, has its offices and facilities at 110 First Avenue South, Nashville,
Tennessee.
The facilities of THERMAL are located close to the downtown area of
Nashville and for this reason have been designed as attractive, modern indus-
trial buildings with extensive landscaping. There are two major structures,
the refuse receiving--steam generator building and the water chiller building.
Two boilers, each with a capacity of 327 Mg/d (360 TPD) of MSW, are used.
Steam is produced at 323°C (613°F) and 2.86 MPa (400 psig). The chiller plant
contains two 24.6 MW (7,000 ton) centrifugal chillers driven by steam tur-
bines; the supply header operates at 5°C (41°F) and 1.34 MPa (180 psig).
Numerous combustion and air pollution problems have been experienced and
modifications to the plant are now being completed.
CONCLUSIONS
• As with SWARU, essentially proven technology caused problems in the
case of Nashville and redesign could remedy the situation. A refuse-
fueled centralized heating/cooling system, particularly in a downtown
area undergoing redevelopment, offers important costs savings for
both the MSW disposal agency and clients for the air conditioning.
This is accomplished because of minimum haul costs for collection
vehicles to the waterwall incinerator compared to a landfill and the
savings realized with one large steam/chilled water facility rather
than a small one for each building.
• NTT originally recognized the advantages of the concept, and there-
fore it is unfortunate that a few initial problems compounded into a
major financial situation. The critical problem was one not unique
to waste processing facilities, but inherent in the low-bid require-
ments of governmental purchasing. When it became obvious that air
quality standards could not be met with the wet scrubber system, cash
flow projections could not be realized because of reduced capacity,
increased fossil fuel costs, and customer dissatisfaction. Fire-side
45
-------
tubing problems then added to the conclusion that reorganization and
refinancing was essential to the environmentally and economically
sound operation of NTT. This was expeditiously managed and it is
concluded that, with the addition of the two electrostatic precipita-
tors and other improvements, the facility should be technologically
acceptable. Conclusions cannot yet be made with regard to new eco-
nomics. The zero drop charge plan originally employed by NTT leads to
an unfair apportionment in operating costs, and it is recommended that
such facilities accept revenues on the basis of the volume of refuse
received.
• The lesson to be learned from Nashville is that a bargain in indus-
trial equipment is a rarity. Estimated costs and revenues must be
established by reliable and disinterested parties; a marginal design
should never be considered acceptable.
PROCESS DESCRIPTION
The central heating and cooling plant of THERMAL uses a water wall,
refuse-fired, mass burning incinerator similar to the steam-producing incin-
erator at Saugus, Massachusetts. The technology used is totally American
rather than European and the requirements of a district heating and cooling
system make the total installation considerably different from the Saugus
operation. A simplified flow diagram of the THERMAL operation is shown in
Figure 4. This process description and Figure 4 are based on the THERMAL
facility as originally constructed; modifications are discussed in a following
section.
Municipal refuse arrives at the incinerator plant in large trucks after
being consolidated at distant transfer stations. The trucks discharge their
load into a deep receiving pit and depart. There is no weighing of the loads
because the original concept was to permit no-charge dumping at the plant and
close down the receiving operation each day when sufficient refuse was on
hand. Refuse in the pit is mixed and dumped into the charging chutes of two
327 Mg/d (360 TPD) mass burning boilers by two overhead travelling cranes,
one of which is normally a standby.
Refuse in the charging chute drops by gravity to a four-section Detroit
Reciprocating Grate Stoker. The first section is the charging grate section,
the second and third sections accomplish drying and burning, and the fourth
section completes the burnout. Ash from the last section drops into a hopper,
is quenched with water, and finally drops from the hopper into a truck for
disposal. The ash that drops through the grates is removed by a separate
system, with two chutes dumping into the same truck. These siftings are not
quenched.
Hot gases from the burning refuse pass through a superheater tube bank,
a boiler generating tube bank, and an economizer in a single pass, and then
exit to a cyclone-type dust collector. The heavy particles extracted from
the flue gas are returned to the ash hopper. The flue gas from the cyclone
collector originally passed through a low energy wet scrubber before exiting
to the atmosphere through an induced draft fan (not shown) and stack. Primary
46
-------
STEAM 400 psig 580fF.
SOLID
WASTE
CONTAINERS
Residential and Commercial
COLLECTION TRUCKS
DUST
COLLECTOR
WET SCRUBBERS (4)
ELECTRICITY TO LIGHTS, SMALL PUMPS, ETC.
WATER
SOFTENER
O
CITY WATER MAKE UP
Solid Waste
Air
Steam (High Pressure)
Steam (Medium Pressure)
Water
Stand By Fuels
COOLING
TOWER (5)
MAJOR EQUIPMENT MANUFACTURERS:
(1) Detroit Stoker Company
(2) Babcock and Wilcox (2-135,000 Ib/hr)
(3) Combustion Engmeermg(125,000 Ib/hr)
(4) Air Conditioning Corporation
(5) Marley Company
(6) Carrier Corporation (2 7.000 tons)
STEAM 150 psig 405'F
r
CHILLED WATER PUMPS
NON-CONDENSING TURBINE DRIVEN
Figure 4. Flow diagram of the Nashville Thermal Transfer Corporation.
-------
combustion air is drawn by the forced draft fans from the refuse pit area,
thereby keeping the pit under a constant negative pressure to reportedly trap
and eliminate odoriferous gases in the boilers.
In order to maintain dependable steam output, and for peaking, there is
a standby oil/gas-fired package boiler capable of producing slightly more
steam than one of the refuse-fired boilers. In addition, there is provision
to fire each of the refuse boilers with oil or gas to stabilize combustion
and increase dependability of steam service.
Steam produced by all of the boilers enters a common header at 2.86 MPa
(400 psig) and 325°C (620°F). For the district heating system, this steam is
reduced to 1.14 MPa (150 psig) by non-condensing steam turbine drives which
operate between the two pressures. These steam turbines are used for plant
auxiliaries.
The centrifugal compressors are driven by condensing steam turbines using
the 1.14 MPa (150 psig) steam exhausted from the chilled water pump drives
and cooling tower pump drive. There is always some demand for both chilled
water and steam throughout the year. The plant makes extensive use of steam
drives for most auxiliary equipment as well as the chilled water system.
All of the energy derived from refuse in this facility is used for dis-
trict heating and cooling in 30 downtown Nashville buildings. In order to
accomplish this, it was necessary to install steam distribution and conden-
sate return pipelines, and supply and return chilled water pipelines. Steam
is distributed at 1.14 MPa (150 psig) and condensate return is pumped at low
pressure. Steam use is metered at each branch line to customers' premises.
The chilled water supply header operates at 1.34 MPa (180 psig) and 5°C
(41°F),, with the return operating on residual pressure from customer systems
and a design temperature of 14°C (57°F). The chilled water lines are unin-
sulated. Some customer systems use the pressure difference between the
chilled water supply and return lines for their own internal circulation,
while others use more sophisticated pumped and valved systems.
EQUIPMENT DESCRIPTION AND OPERATION
The successful operation of any processing plant, especially of one that
processes refuse, depends heavily on the performance of the equipment, as well
as on the workability of the flow design. Some of the key items of equipment,
and their performance, are described below.
Boilers
The mass-burning boilers are waterwalled units manufactured by Babcock
and Wilcox. There are two boilers, each with a rating of 327 Mg/d (360 TPD)
of municipal refuse while producing a steam output of 49 442 kg/h (109,000
Ib/hr). Design steam conditions are 2.86 MPa (400 psig) at 323°C (613°F) from
116°C (240°F) feedwater. The boiler design pressure rating is 3.55 MPa (500
psig). The outlet gas temperature averages 221°C (430°F).
48
-------
The grate system is a four-section Detroit Reciprocating Grate Stoker.
The second section receives the refuse from the charging chute "first sec-
tion," and begins to dry it by exposure to the hot atmosphere of the furnace.
Actually, much of the refuse ignites and begins to burn on this second grate.
The primary combustion takes place after the refuse has tumbled from the first
to the second grate, and the final burnout occurs on the fourth grate.
The primary function of the underfire air being blown up through the
grate sections is to provide sufficient oxygen for combustion and its second-
ary purpose is to keep the grate temperature within the operating limits of
the metal grate material. Overfire air is blown into the furnace through the
front and rear wall air jets. Three 10 cm (4 in.) side wall air jets were
installed on each side wall of the boilers, with the air coming from the over-
fire fans, in an attempt to control the corrosive reducing atmosphere that
normally occurs near the level of the burning refuse. This attempt to control
the reducing atmosphere worked to some degree, but both furnaces have been
modified by the addition of silicon carbide refractory protection from the
grate level to a height of 6.1 m (20 ft) on the sidewall tubes. This is a
feature that the furnaces at Saugus, Massachusetts, had from the beginning.
At present, the No. 2 furnace side walls are being metallized with a 0.38 mm
(0.015-in.) coat of aluminum. This begins at the top of the silicon carbide
and extends to 1.2 m (4 ft) below the roof tubes.
During the first few months of operations, the plant operators attempted
to use variation of both underfire air and overfire air to control steam out-
put rather than to adjust air flow to the required boiler conditions. This is
considered to be one of the contributing factors to boiler tube corrosion and
possible damage to the superheaters. The problem has been resolved, as part
of a new capital completion program, by replacing the superheaters with new,
larger units, with heavier gauge tubes. It remains difficult, however, to
maintain draft on the boilers with the installed turbine-driven induced draft
fans.
A performance test of one of the boilers by the manufacturer was con-
ducted from February 10, 1975, to February 27, 1975. Some of the important
averaged test results are tabulated below:
Measured Value
Parameter
Load (98% Capacity)
Refuse Rate
Steam Temperature
Feedwater Temperature
Outlet Gas Temperature
Refuse Higher Heating Value
48 535 kg/h
13.7 Mg/h
301°C
112°C
221°C
11.34 MJ/kg
English
107,000 Ib/hr
15.1 TPH
573°F
233°F
430°F
4875 Btu/lb
49
-------
Measured Value
ST English
Parameter iL —*
Boiler Efficiency 71.9% 71.9-6
Refuse Weight Reduction 78.3% 78. *
Refuse Volume Reduction 92.3% 92.3-6
Total Loss of Combustibles 2.3% 2.3-6
Putrescibles in Ash 0.072% 0.072%
Outlet Particulate Loading at 12% C02 3.17 g/Nm3 1.46 grains/SDCF
In addition to the above tabulated test results, the parts of the report deal-
ing with flue gas analysis and the particulate size distribution are of inter-
est. These are tabulated below:
FLUE GAS ANALYSIS
C02 10.4 vol-%
0 9.5 vol-%
Excess Air 84 vol-%
NO 146 ppm
x
SO 38 ppm
CO 153 ppm
Chloride 110 ppm
FLY ASH PARTICLE SIZE DISTRIBUTION
Particle Size, jum
(Percentage less than stated size)
Test Date
2-21-75
2-21-75
2-23-75
2-26-75
2-26-75
10
56%
32%
44%
26%
38%
5
40%
22%
32%
18%
28%
1
32%
19%
24%
14%
18%
0.3
22%
16%
14%
11%
10%
50
-------
These tests were conducted under controlled conditions and with all
equipment pre-tested and properly adjusted. As such, the reported perfor-
mance, especially boiler efficiency, probably represents the maximum to be
expected from this particular installation. Another factor that strongly
affects the results of the boiler tests is that the fossil-fueled "stand-by"
boiler was also operating at the same time to furnish enough steam to meet
the demand and to smooth out steam output. Therefore, the large uncontrol-
lable swings in steam pressure and temperature that are typical of mass-fired
boilers did not occur in the test.
Boiler Controls
There are both pneumatic and electrical components in the boiler control
system. In this respect, the controls are the usual systems associated with
grate-fired boilers, and, if they are used as designed, they will function
well. Matching the boiler output to steam demand, except for very slow and
gross changes, has not proven possible. This is inherent to the type of fuel
used and the grates employed.
The only abnormal features of the control systems are the use of air,
filtered and dried, from the plant compressed air system for the pneumatic
controls and the direct connection of electrical controls to the plant elec-
trical system. Both of these features are attempts to reduce capital invest-
ment by omitting desirable redundant equipment. The results have been poor.
When the plant air system is being used to run tools for maintenance and re-
pair, the system air pressure sometimes drops low enough to affect the pneu-
matic boiler controls, and the result is an unscheduled shutdown. At present
there is only one primary feeder line to the plant, and occasionally the plant
is affected by a power outage. Both of these conditions will be corrected
during the capital completion program.
Air Pollution Control Equipment
The failure of the pollution control equipment at THERMAL to meet local,
state, and federal emission standards has been as serious a problem as the
boiler tube design one. The original pollution control equipment on each
boiler consisted of a multi-cyclone dust collector for the large particles
and a low energy horizontal wet scrubber. Even after attempts to improve and
modify the operation, the lowest emission level that could be attained was
0.365 g/Nm3 (0.168 grains/DSCF), a little more than double the allowable maxi-
mum of 0.173 g/Nm3 (0.08 grains/DSCF). After extensive investigations, THER-
MAL concluded that for its particular operation, only an electrostatic precip-
itator could produce the desired results.
Possible cause for poor scrubber performance may be found by comparing
the scrubber design specification to the particle size distribution reported
in the Babcock and Wilcox boiler performance test results. The scrubbers
were specified to remove 95% of all particulate matter based upon a 5 M™ mean
particle diameter. The particle size data indicated that up to 40% of the
particles were under 5 /m and that the average distribution of such particles
was 28%. Furthermore, these same tests show an average particulate size dis-
tribution of 21.4% under 1.0 Mm and 14.6% under 0.3 /m. The large percentage
51
-------
of particulates in the sub-micrometre range, which is difficult to scrub out,
may be one explanation for the problems encountered. The flue gas, in com-
bining with the scrubber water, created a very active acidic condition that
rapidly corroded the scrubber's internal parts and casing.
THERMAL purchased one electrostatic precipitator that had to be installed,
checked out, and in operation by October 29, 1976, to meet the requirements of
a Compliance Order of the EPA. The specifications indicate a guaranteed out-
let loading of 0.032 g/Nm3 (0.015 grains/DSCF) (adjusted to 12% C02) with an
inlet loading of 100 times that level. This unit should satisfy the air pol-
lution control requirements on one boiler. Installation of an identical pre-
cipitator on the second boiler began about October 1, 1976.
Water Chillers
The water chiller plant contains two 24.6 MW (7,000 ton) centrifugal
chillers driven by steam turbines. They are equipped with variable inlet
vanes, which, in conjunction with turbine speed control, permit a variation
in chiller output from the maximum down to as low as 2.46 MW (700 tons), a
10-1 output adjustment ratio. The operators of the plant report that this
output adjustment ratio is sufficient for their needs.
Initially, one of the chillers experienced excessive vibration, which
limited its use for some time. The trouble was traced to the governor valve
stem, which was replaced, and the same modification was made to the second
chiller.
Steam Turbine Drives
The entire plant was designed to make maximum use of steam turbine drives
on auxiliary equipment. Specific exceptions of this are the forced draft fans
and two 0.014 m^/s (240 GPM) boiler feed pumps driven by electric motors.
This design approach was taken to reduce the need for expensive electric power.
The operators of the plant have indicated that they would not choose steam
turbine drives for auxiliaries again. The major reason would appear to be a
lack of operating flexibility and dependability. Turbine performance varies
considerably with steam pressure, and this may vary over a wide range in a
mass burning boiler. One of the high pressure chilled water distribution
pump drives has already been replaced with an electric motor, and when the
new precipitators are installed, the installation will include new induced
draft fans with electric motor drive.
Most of the turbines on auxiliaries use steam at 2.86 MPa (400 psig) and
316°C (600°F) and exhaust at 1.13 MPa (150 psig). The exhaust steam is used
for the two centrifugal water chiller drives and for distribution into the
district heating system.
Materials Recovery
At present there is no provision for recovery of material resources. The
plant design is consistent with the fact that mass burning incineration leaves
little of practical value to be recovered from the ash, except for the magnetic
52
-------
metals. With the original intent being to generate steam at minimum costs, no
provisions for recovery of the ferrous fraction was made in the design. This
leaves 6% by volume of the original refuse to be landfilled.
Refuse Supply
The refuse is delivered to the facility by Metropolitan Nashville (METRO)
in transfer trucks. Bulky waste, heavy items, and large incombustibles are
supposed to be separated from the waste before delivery, but there is provi-
sion, via by-pass chute, for the cranes to set aside any such items that
arrive. METRO does not pay any drop charge per ton, but does provide an
annual payment for services. Title to the refuse is retained by METRO through-
out the operation. This would be a complicating factor if materials recovery
were attempted.
Studies have indicated that there is a more than adequate supply of
refuse to keep the plant in operation and provide for future expansion. In
that the plant is sized to accept only a part of Nashville's refuse, it is
possible to schedule receipt of refuse to suit the steam demand. This elimi-
nates the necessity for adjustment of refuse supply to steam demand or vice
versa, a problem faced by most refuse-to-steam energy recovery facilities.
Distribution Piping
The steam and chilled water distribution piping is located in a common
trench with other utilities that were installed at the time. Only the steam
supply line is insulated, with an insulation barrier being placed between the
steam and-chilled water lines when the pipes were installed.
The steam supply line is steel, insulated with calcium silicate, and
installed in a prefabricated steel conduit. The chilled water lines are of
ductile iron with an "Enameline" cement lining, and assembled with mechanical
joints.
At present there is only one distribution network, but there is provision
for addition of a second network or extension of the existing system as new
customer areas are developed. All piping is sized for future expansion.
ENGINEERING EVALUATION
This facility was designed, built, and placed in operation with a minimal
expenditure of funds. Outwardly, the incinerator plant is very similar to the
plant operated by RESCO at Saugus, Massachusetts, and many European installa-
tions, but economic considerations forced the designers to take a much less
conservative approach. The result is that the RESCO plant has performed well
from the beginning, while THERMAL's plant has had a multiple of start-up prob-
lems. Most of the problems and their solutions have been discussed and the
capital completion program is intended to provide those solutions. When these
problem areas are cleared up, there should be a considerable improvement in
overall operational dependability and economics. In fact, the modifications
completed thus far have steadily improved plant performance, so much so, that
53
-------
steam deliveries for 1976 increased 57% over 1975, and chilled water deliv-
eries increased 112%.
Boilers
The boiler tests for February 10 to 27, 1975, developed some useful
information such as flue gas analysis and particle size distribution. The
boiler performance data, however, is suspect, because of the operation of a
fossil fueled boiler in parallel with it during the test.
The THERMAL operating staff reports that these mass burning boilers may
experience a mostly uncontrollable swing in output of 6 804 kg/h (15,000
lb/hr) above and below the working level, and that a 6 804 to 9 072 kg/h
(15,000 to 20,000 Ib/hr) change may occur over a 10 minute time period. With
a relatively constant demand for steam, these violent swings in output can
only be accommodated by exhausting or condensing excess steam when the swing
is high and making more steam with fossil fuel when the swing is on the low
side. Either situation is costly and plant economics suffer. Good perfor-
mance of the grate and charging chute to maintain close control of refuse
fed to the boiler can help to reduce the amount of swing, but cannot eliminate
it. It is reported that the existing grate and chute combination does not
afford the needed close control of refuse feed.
Grates
The THERMAL operating staff reports that the grates jam occasionally. It
is their opinion that the grate support arrangements appear to be inadequately
sized for the amount of weight they have to support when the material is wet.
These grates are essentially identical to those that would be used to fire
coal and it is apparent that some modification is necessary to adapt them to
firing unprocessed refuse.
Load Matching
THERMAL's refuse-to-energy operation is the first attempt in the United
States to convert refuse to steam destined solely for a district heating and
cooling system. Such an operation is an exceedingly difficult one to control
for profitable operation. There are three major variables that are basically
uncontrollable, but must be accommodated in the overall system. These are:
Refuse Supply
Usually, when a waste handling facility is constructed, it is planned to
handle all of the waste from a nearby geographical area. The volume of waste
generated in such a service area will vary daily and by season, according to
the habits of the people in the area, and not be generated in relation to
the needs of the plant. THERMAL has avoided this variable by providing a
disposal service for only a part of the refuse generated in Metropolitan
Nashville. Delivery of refuse to the plant can be scheduled within certain
limits.
54
-------
Boiler Performance
The output of a refuse-fired boiler depends on the heating value of the
refuse and its moisture content, both of which may vary greatly in a very
short time. In mass burning boilers, such as those at Nashville, the swings
in output are very wide. In addition, the response of a mass burning boiler
to changes in firing rate is very slow. For these reasons, a mass burning
refuse-fired boiler cannot be used under conditions where it must vary output
quickly to match the steam system load changes. It is claimed that pre-
preparation of the refuse fuel by shredding and then firing it in a somewhat
differently designed boiler can result in the controlled variable output
needed. Though this seems reasonable, it has not been proven in an actual
installation.
Load Variation
The steam load imposed on a producing plant by a district heating and
cooling system will vary according to time of day and season. Sometimes the
daily load variations occur quite rapidly. THERMAL reports that there is some
demand for both steam and chilled water throughout the year, indicating exten-
sive use of blending type air conditioning systems and/or a need for dehumidi-
fication. This tends to smooth out the load fluctuations, but certainly does
not eliminate them. The periods of lowest demand are spring and fall.
THERMAL has no major industrial customers whose demand might be used to smooth
out the total load fluctuation on the plant.
A mass burning boiler installation such as Nashville's must have the
additional capability of burning gas or oil, either along with the refuse or
in a separate standby boiler. Both options may be required in order to meet
the steam demand when the heat content is low, or moisture level of the refuse
is high, and to adjust system output to demand on rapid changes. This expense
for fossil fuel is not always considered in project planning.
Conversely, when steam output is momentarily high because of high heat
content or low moisture of the refuse fuel, there must be some means of con-
densing excess steam. Without a means of recycling the excess steam into the
treated water system, it must be blown off to the atmosphere. This represents
the dual loss of both energy and of treated water used to make t'he steam.
THERMAL has an excess steam condenser but it does not seem to be large enough.
The low total load condition experienced in any district heating and
cooling system in spring and fall results in inefficient use of the total cap-
ital investment in the plant during that period. This is another reason for
attempting to build up a diversified load structure, to make the average
demand more nearly approximate the maximum demand.
Table 1 shows energy sales and fuel usage from June, 1974, through July,
1976. In calendar year 1975, THERMAL incinerated 73.177 Gg* (80,662 tons),
61% of the total capacity of one boiler system. In the first seven months
* All refuse weights reported here are based on periodic calibration of
quantity held within the grapple bucket.
55
-------
TABLE 1. COMPARISON OF ENERGY SALES VERSUS FUEL CONSUMPTION
Month/Year
June 1974
July
August
September
October
November
December
January 1975
February
March
April
May
June
July
August
September
October
November
December
January 1976
February
March
April
May
June
July
Energy Sales
Steam, Ib
1,596,000
6,582,000
3,383,000
7,234,000
12,432,000
15,198,000
17,132,000
13,452,000
23,276,000
17,523,000
24,783,000
8,981,000
6,224,000
5,689,000
5,261,000
12,223,000
17,543,000
27,518,000
37,670,000
40,552,000
28,378,000
27,334,000
14,963,000
14,060,000
8,147,000
7,723,000
404,830,000
(183.63 Gg)
Cooling, ton-hrs
198,890
298,906
325,066
530,722
557,414
382,213
435,297
407,359
422,336
999,188
550,786
1,561,713
2,806,212
2,314,840
2,427,387
2,240,076
1,775,614
1,193,921
709,797
623,280
965,595
1,396,590
1,335,335
1,695,765
2,385,914
2,498,861
31,039,077
(28,158,650 Mg-h)
Fuel Used
Solid Waste, ton
163
0
595
3,257
7,693
6,813
8,190
7,286
6,554
9,401
6,925
333
1,054
7,962
6,610
7,328
8,721
8,827
9,661
10,557
7,302
7,229
7,200
8,379
6,811
6,500
161,351
(146.38 Gg)
Gas, 102ft3
600,300
690,000
923,600
880,300
0
0
220,000
75,568
207,424
193,100
259,118
686,700
556,900
88,984
34,600
23,300
83,070
96,106
61
90
132,728
143,100
20,087
134,700
243,200
282,600
6,575,636
(18.615 x 106 m3)
Oil, gal
0
0
0
96,300
444,200
353,340
137,800
105,807
91,606
6,150
58,206
0
0
0
0
28,350
27,142
32,573
111,944
115,296
141,596
0
0
0
0
0
1,750,310
(6.624 x 103 m3)
-------
of 1976, THERMAL operated at 71% capacity, and reportedly are near 100% (one
boiler) by April 1977. The total operating experience with regard to fuel
usage is summarized as follows:
Fuel
Oil
Gas
Wastes
Percent
of
Total
8.5
23.5
68.0
Amount Used
SI
Units
6.624 x 103 m3
18.615 x 106 m3
146.38 Gg
English
Units
1,750,310 gal
657.564 x 106 ft3
161,351 tons
Heating Value
TJ
258.520
711.141
2 042.911
109 Btu
245.043
674.003
1,936.212
Total
Oil Equivalence
10V
6.624
18.219
52.340
77.183
Barrels
41,674
114,626
329,288
485,588
Steam Driven Auxiliaries
The extensive use of steam auxiliaries, as at Nashville, may result in
some saving in operating cost, but there is a considerable sacrifice in operat-
ing flexibility. Auxiliaries that require high starting and running torque
or very low speed should not be planned for steam turbine drive at all. When
nearly all auxiliaries are steam driven, it is necessary to have an auxiliary
or standby boiler that is equipped with electric auxiliaries in order to build
up steam to fire up the main boilers. Since a standby is needed anyway, this
is not a serious problem, but considerable time is required to bring a boiler
that is using steam-driven auxiliaries up to pressure. During this start-up
period, there is a much greater probability of excessive particulate emissions,
and a plant that at other times meets all air quality standards may be cited
for a violation.
Steam turbine performance is sensitive to steam inlet conditions. If
steam pressure falls due to variation in the heating value or moisture content
of the refuse, the performance of important turbine drives may be so seriously
affected that a boiler malfunction or shutdown will occur. This has happened
at Nashville several times.
Air Pollution Control
The problems with the wet scrubbers and the subsequent correction by the
addition of electrostatic precipitators have already been discussed. The
original tests in September, 1976, of the first precipitator demonstrated that
solid particulate values below 0.07 g/Nm^ (0.03 grains/SCF) at 12% C02 were
being achieved. Retest in March, 1977, indicate that stack effluent is in the
range of 0.017 to 0.019 g/Nm3 (0.007 to 0.008 grains/SCF), an efficiency
greater than 99.5%. What must be made clear, however, is that no air pollu-
tion control equipment of any type can be expected to perform satisfactorily
unless its design is based on a knowledge of the performance of the furnace.
The output flue gas conditions must be known as they actually occur with the
57
-------
fuel used in the furnace. The flue gas measurements taken by Babcock and
Wilcox as part of their tests are valuable in this regard, but the flue gas
conditions from an actual shredded waste fired furnace, such as at Hamilton,
Ontario, are still unknown because no such tests have ever been made. Also
the flue gas conditions from a Von Roll type mass burning furnace are not
generally known because such installations are proprietary and the data are
normally not made available.
PROJECT DEVELOPMENT
The Nashville Thermal Transfer Corporation was the product of a late
1960's drive for urban renewal in downtown Nashville. In response to the
city's needs, Mayor Beverly Briley commissioned a study to determine the fea-
sibility of a district heating and cooling facility for public and private
buildings. The study indicated that such a facility to serve the central city
was practical, and city officials decided to proceed, based on a time table
that called for construction to start in 1972 and for completion in 1974. In
order to undertake such a sizeable job on a tight schedule, Nashville Thermal
Transfer Corporation, a "Not for Profit" corporation, was established.
THERMAL was initially created as the developer and operator of a district heat-
ing and cooling system.
In 1970, about the time THERMAL was organized, it was suggested that the
required steam be produced by burning the refuse from the city, thereby solv-
ing a part of that problem also. Accordingly, a second study that established
the feasibility of deriving energy from refuse was commissioned. THERMAL then
became a refuse disposal organization as well as the operator of a district
heating and cooling system.
The main objectives of the new facility were:
• Provide low cost district heating and cooling for center city
buildings.
• Recover energy in all combustible solid waste not recycled for other
purposes.
• Partially eliminate the need for sanitary landfill.
• Substantially reduce the cost of solid waste disposal.
• Improve water and air quality in urban Nashville by meeting solid
waste disposal, water pollution, and air emission standards with a
central plant that incorporates effective environmental control
equipment.
• Provide for major ferrous metal recycling from incinerator residue.
• Create and operate a solid waste-fueled central heating and cooling
plant project that has a favorable economic and environmental impact
on the community.
58
-------
The project has thus far met many of the stated objectives, and, with the
improvements that are now in progress, should substantially meet all of them.
A review of each objective and of the plant's history to date indicates:
• Low cost heating and cooling - This objective has not been attained.
Delivered prices for steam and chilled water are presently higher than
the equivalent energy charge from the local power company or for pur-
chase of fossil fuel. However, if each customer were to add the
amortized cost of his investment in refrigeration equipment and boilers
to these supposedly low energy costs, the total comparable energy cost
would more nearly approximate the cost of services from THERMAL. In
addition, the improvement in plant performance and increased output
expected from the ongoing modifications should reduce the cost of
heating and cooling services supplied by THERMAL while prices for
fossil fuel and electric power are steadily rising.
• Recover energy from solid waste - This objective was successfully
reached as soon as the plant operated its refuse boilers and sold the
steam product to customers. As the operation improves, this conver-
sion will be more efficient and reach a greater total.
* Partially eliminate the need for sanitary landfill - Every unit volume
of refuse burned in this plant is at least 0.9 of a unit that does not
have to be landfilled. As the plant operation improves and ultimately
approaches its design capacity, the refuse burned (up to 653 Mg or
720 tons per day) becomes the equivalent of a substantial landfill
operation.
• Substantially reduce the cost of solid waste disposal - The plant
certainly has done this for Metropolitan Nashville. In the beginning,
Nashville guaranteed to deliver waste to the plant "at no cost." This
meant no cost to THERMAL but also resulted in "no cost" to Nashville.
Later, Nashville contributed up to $150,000 per year to the operation
to pay for services rendered. The plant does not weigh refuse as
received, so there is no positive record of the number of tons of
refuse burned; but a review of steam and chilled water sales for fis-
cal 1975, just after start-up, indicates a total for the year of about
39 Gg (43,000 tons) and the same sort of review for fiscal 1976 indi-
cates a total consumption of about 72.6 Gg (80,000 tons). The pay-
ment of $150,000 per year seems to be equivalent to a drop charge of
$3.85/Mg ($3.49 per ton) in fiscal 1975, normal for many cities in the
United States. In fiscal 1976 this figure drops to $2.06/Mg (1.87 per
ton), a very low figure. There is some indication that Nashville
might substantially increase the payment to THERMAL.
* Improve water and air quality in urban Nashville - This objective has
been reached in spite of the poor performance of the wet scrubbers.
If each of THERMAL!s customers were to be individually and privately
heated, the total emissions would be greater than now produced by the
Nashville refuse disposal plant. Here, the efficiency of large size
and added control equipment makes the difference.
59
-------
• Provide for major ferrous metal recycling from residue - This objective
has not been attained or even attempted. There is no reason why equip-
ment for ferrous metal recovery cannot be added to the plant if pric-
ing on the metal salvage market indicates that it would be economically
feasible.
• Create a favorable economic and environmental impact on the
community - This objective has been partially attained in improvement
of the environment and in economic benefit to the City. The develop-
ment of reasonable rates for steam and chilled water while assuring
amortization of investment is an objective within reach but not yet
attained. The organization of THERMAL and the construction of its
existing facilities were financed by the issuance of $16,500,000 worth
of revenue bonds, the last of which will mature in June, 2002. When
all bonded indebtedness has been repaid, the facilities will become
the property of Metropolitan Nashville.
FINANCIAL ANALYSIS
In October, 1975, R. W. Beck and Associates issued a feasibility report
that resulted from their analyses, investigations, and studies of the THERMAL
proposal to obtain additional financing. This was a comprehensive study of
all aspects of continued operation of THERMAL. The study generally agreed
with the findings of the Management and their Consultant as to the modifica-
tions and improvements required to make the plant efficient and dependable.
The cost estimated for needed modifications and improvements, combined with
paying off short term loans and some unpaid construction debts, were found to
require an issue of junior lien bonds to an amount of $9,250,000.
The total projected revenues for 1976, based on existing steam and chilled
water rates with no additional customers, was $3,267,000. Without labor and
fuel costs, this total annual revenue increases to $5,096,000 by 1985.
Because present rates are in excess of the rate formula originally written
into customer contracts, there is a delay period built into the rate increase
mentioned above to allow normal inflation to "catch up" with the established
rate.
On the assumption that the $9,250,000 bond issue would become a reality
and the further assumption that the METRO Council would pass an ordinance
increasing the payment of $1,500,000 per year, a pro-forma income statement
was prepared, again on the basis of existing customers only. The income
statement shows that THERMAL can meet the needs of adequate service without
using all of the $1,500,000 per year guarantee from METRO. The actual pro-
jected need for METRO funds ranges from $986,000 in 1976 to $1,407,000 in
1979, with the ten year average between 1976 and 1985 at $1,258,200. If the
average payment were expressed as a disposal fee, it would amount to a dis-
posal fee of $6.59/Mg ($5.98 per ton) when the plant is operating at 80% of
capacity, a reasonable level. However, if the refuse consumption of the plant
remains relatively low, equivalent to the steam and chilled water demand, this
equivalent disposal fee would more than double.
60
-------
If additional customers are added to the system, which is possible at
nominal increase in operating cost, the projected net increase in revenues
available for debt service will reduce the METRO average ten year payments to
$1,026,800 per year, or an equivalent disposal fee of $5.38/Mg ($4.88 per ton),
a very reasonable figure.
THERMAL management has chosen a different method of financing to obtain
the needed extra funds. Their present plans are to secure a loan of
$5,700,000 for construction of environmental protection facilities, which will
be guaranteed by the State of Tennessee, and bank loans totaling $2,300,000,
for a total of $8,000,000. This total is somewhat less than the need indi-
cated by the R. W. Beck report, but the loss will be partially offset by the
elimination of bond issue costs and the necessity of capitalizing the Debt
Reserve Fund.
Funds for the modifications programs presently underway to comply with
EPA orders have been provided by short-term loans from local banks. Loan
terms and interest rates are under negotiation at present; therefore no spe-
cific statement can be made on expected net revenues. With the reduced loan
amount, however, it should be feasible to stay within the projections made by
R. W. Beck even if interest rates are somewhat higher than those of a bond
issue.
LITERATURE
References 2 through 5 contain much of the published information on the
development and operation of THERMAL.
61
-------
SECTION 5
CITY OF CHICAGO-COMMONWEALTH EDISON SUPPLEMENTARY FUEL SYSTEM
INTRODUCTION AND SUMMARY
The City of Chicago Supplementary Fuel Processing Plant is undergoing
start-up testing in the spring of 1977. It is patterned after the City of
St. Louis/Union Electric Company demonstration system that has been operating
during the past four years. The facility processes MSW to make refuse-derived
fuel (RDF), which is then co-fired with coal in the steam generators of a
large electric power generating station. Depending on maintenance experience,
the plant has a capacity of 2250 to 2866 Mg refuse per day (2,480 to 3,160 TPD)
and can produce 1566 to 1996 Mg/d (1,726 to 2,200 TPD) of RDF.
The City of Chicago does not have adequate landfill sites within its
borders to handle its solid waste disposal requirements. It has traditionally
relied on city owned and operated mass burning incinerators to dispose of city-
collected refuse, which averaged 20.9 Gg (23,000 tons) per week in 1972. Con-
cerns with replacing aging units in its existing incineration facilities and
upgrading other units to meet EPA air emission standards led the city to under-
take an exhaustive study in late 1972 to determine what the next increment of
disposal facility should be. Because construction and operating costs were
escalating to new highs, the study criteria included a survey of disposal
technology to determine which system would best satisfy the city's require-
ments. The study found, based on economics, technological effectiveness, and
sociopolitical considerations, that the St. Louis/Union Electric system best
satisfied these requirements. The city then authorized the design and con-
struction of an RDF facility and negotiated an arrangement to sell RDF to the
Commonwealth Edison Company, which would burn it along with coal in the power
boilers of its Crawford Station.
The information presented here has been entirely obtained from records of
the City of Chicago and The Ralph M. Parsons Company. The only published arti-
cle with any level of detail is that of Zralek and Bailey (Ref. 6).
CONCLUSIONS
• Although the waterwall refuse combustion system is considered the most
technologically established waste-to-energy process, use of a refuse-
derived fuel to partially supplement coal in existing furnaces is
expected to become increasingly popular becaus'e of first-cost consid-
erations. The RDF is sold strictly on the basis of its value as a fuel
to a customer already owning the steam generator-ash handling-emission
62
-------
control system; hence, the high capital cost associated with this por-
tion of the equipment need not be borne by the waste processing organ-
ization. This organization becomes a fuel seller rather than an
energy processor-consumer.
• The mechanical processing required to yield the RDF, and which simul-
taneously results in an inorganic-rich fraction, is expected to pro-
vide the impetus for further R§D on commercial recovery of other
materials. If the selling price of the RDF and ferrous metals will
cover the costs for the initial particle size reduction and air classi-
fication, then sufficiently economical methods for isolation of several
non-ferrous metals and glass might be realized. Purity of the frac-
tions now obtained is rather low and it is recommended that research
be continued to achieve a higher degree of component separation.
• The Chicago plant is just now undergoing startup testing. The first
6 to 12 months of operation should reveal information on the perfor-
mance of individual equipment and subsystems. Hopefully, the owner
will share this experience so that new plants will benefit from pre-
vious observations.
PROCESS DESCRIPTION
Flow Diagram
The primary purpose of a refuse-derived fuel (or supplementary solid fuel)
plant is to process the combustible fraction of raw refuse into a transportable
fuel for use typically in a coal-fired, suspension-burning boiler at a large
power plant. The requirement for the fuel for Chicago is that the particles
be approximately 3.8 cm (1-1/2 in.) in size, and light enough to permit good
burnout in suspension burning.
Figure 5 shows the process flow diagram for each train of processing
equipment. Key pieces of equipment at the fuel processing plant and at the
power generating plant are indicated; also shown are the quantities of mate-
rials handled at various points in the system. The raw refuse is delivered
to the tipping floor at a nominal rate of 72 Mg/h (80 TPH) . It is pushed into
the coarse or primary shredder feed conveyor by means of a front loader and
the coarse shredder then reduces the material to particles that have a maximum
dimension of 20.3 cm (8 in.). These particles travel by conveyor into an air
classifier that separates the material into light and heavy fractions. The
light fraction is composed mainly of combustibles, which are readily airborne.
The heavy fraction contains the non-combustible items and those heavy particles
of combustible materials that do not readily burn in a suspension-fired system.
This heavy fraction is conveyed to storage bins for eventual disposal. On the
way to the bins, the stream is passed under a magnetic separator that diverts
the ferrous material into one storage bin, while the remainder goes to another.
Aluminum is not recovered because waste samples indicated insufficient quanti-
ties for economical recycling.
63
-------
60,000 C.FM EXHUUST
(AIR • i 3 TPH MOI5TURC)
FROM REFUSf-J
STACK OSES
£5 ; MOISTURE
- SUPPLEMENTARY FUEL- PROCESSING
POWE.RPLANT.RECEIVING STORAGE AND FEED
• EXISTING POWER PLANT-
SOLID WASTE TO SUPPLEMENTARY FUEL
Figure 5. Flow diagram of the supplementary fuel processing
plant of The City of Chicago - Commonwealth Edison
-------
The light fraction consists of most of the light combustible particles,
such as paper, plastics, shredded pieces of wood, leaves, etc., and some non-
combustible fines; it contains approximately 85% of the organics in the raw
refuse. This fraction leaves the air classifier in the exhaust air stream and
is passed through a cyclone separator, which drops the solid material directly
into the mouth of the secondary or fine shredder. Here the material is fur-
ther reduced in size to an average dimension of 3.8 cm (1-1/2 in.). The
secondary shredder discharge is conveyed to a hopper over the air lock feeder
to the pneumatic transfer system. This system conveys the prepared fuel to
the storage bins or silos, located adjacent to the boiler house at the power
plant. The storage bins provide extra capacity, thus ensuring a proper supple-
mentary fuel flow to the boilers.
Stored supplementary fuel is retrieved by a mechanical system and blown,
as required, into the boilers by pneumatic conveyors that are similar to those
for the powdered coal feed. The feed rate is controlled by regulating the
speed of the mechanical retrieval system. The supplementary fuel consumption
is in fixed proportion, on a heating value basis, to the quantity of coal
being consumed.
Dust control is an important feature of the Chicago RDF production con-
cept. Control and collection systems are integrated into the process and the
recovered material is added to the prepared fuel at the processing plant just
prior to entering the pneumatic transfer system to the storage bins. At the
power plant transfer system, air is bled from the bin through a filter bag-
house and the collected dust returned to the bin.
The Chicago process was designed over three years ago and though much new
work has been done in the RDF field since then, nothing has yet been demonstrat-
ed on a large scale that would warrant making changes in Chicago's RDF pro-
cessing train. The most promising new RDF concept involves trommelling raw
refuse prior to primary shredding. This arrangement results in a significant
reduction in the amount of material requiring shredding, with attendant reduc-
tion in processing power consumption and reduced shredder maintenance. The
concept will be tested functionally and economically on a large scale by the
Recovery I Project, due to go on stream late in 1976 at New Orleans (Ref. 7).
Material Balance
Table 2 shows a mass balance for the RDF processing plant for an input of
907 Mg/d (1,000 TPD) of raw refuse. The raw refuse composition was determined
by testing 680 to 907 kg (1,500 to 2,000 Ib) random samples from 30 standard
refuse collection vehicles. The samples were hand-sorted and the components
classified, weighed, and subjected to physical and chemical analysis. The
test program was conducted during late January and early February of 1973;
some variation in component percentages during other seasons of the year can
be expected. Design was based on these results rather than national average
values.
65
-------
TABLE 2A. RDF MASS BALANCE FOR 907 Mg/d RAW REFUSE INPUT
HHVa
MJ/kg
13.02
13.91
13.51
12.08
33.10
15.51
16.42
9.37
10.28
19.65
4.84
1.65
0.19
Raw Refuse
Component
Corrugated box board
Newspaper
Magazines/books
Misc. paper
Plastics
Textiles
Wood
Leaves, shrubs, grass
Food waste
Rubber & leather
Fines (-2.54 cm)
Metals
Steel
Aluminum
ONFC
Glass, Ceramics,
Stones
Moisture
As Rec'd
Wt. %
3.67
12.73
3.74
24.77
4.09
3.16
2.23
2.06
10.94
1.22
11.45
9.89
(9.07)
(0.50)
(0.32)
10.05
100.00
(25%)
Wt.
Mg/d
33.3
115.5
33.9
224.7
37.1
28.7
20.2
18.7
99.2
11.1
103.9
89.7
82.3
4.5
2.9
91.2
907
(226.8)
Light Fraction
Mg/d
31.7
113.4
31.7
213.2
34.5
27.2
16.3
15.4
90.7
9.1
45.4
1.5
1.8
0.1
13.6
645. 7b
(204. l)b
%
4.9
17.6
5.0
33.0
5.3
4.4
2.5
2.4
14.1
1.4
7.0
0.3
2.1
100.0
(31.6)b
Heavy Fraction
Mg/d
4.9
2.1
2.2
11.5
2.6
1.4
3.9
3.3
8.5
2.0
58.5
80.7
2.7
2.8
77.6
261.4
(22.7)
Iron Recovery
Mg/d
1.4
76.7
78.1
(0.9)
To Disposal
Mg/d
1.5
2.1
2.2
11.5
2.6
1.4
3.9
3.3
8.5
2.0
57.1
4.0
2.7
2.8
77.6
183.3
(21.8)
a. Raw Refuse HHV = 10.19 MJ/kg.
b. Expected moisture content in the storage silos is 30%. The difference is due to moisture loss in the
pneumatic systems and amounts to 14.5 Mg/d. Quantity arriving in silos is 631.2 Mg with HHV of 13.20
MJ/kg.
c.
Other non-ferrous
-------
TABLE 2B. RDF MASS BALANCE FOR 1000 TONS/DAY RAW REFUSE INPUT
HHVa
Btu/lb
5,600
5,980
5,810
5,193
14,230
6,670
7,060
4,030
4,420
8,450
2,080
709
82
Raw Refuse
Component
Corrugated box board
Newspaper
Magaz ines/books
Misc. paper
Plastics
Textiles
Wood
Leaves, shrubs, grass
Food waste
Rubber § leather
Fines (-1 in.)
Metals
Steel
Aluminum
ONFC
Glass, Ceramics,
Stones
Moisture
As Rec'd
Wt. %
3.67
12.73
3.74
24.77
4.09
3.16
2.23
2.06
10.94
1.22
11.45
9.89
(9.07)
(0.50)
(0.32)
10.05
100.00
(25%)
Wt.
T/D
36.7
127.3
57.4
247.7
40.9
31.6
22.3
20.6
109.4
12.2
114.5
98.9
(90.7)
(5.0)
(3.2)
100.5
1000
(250)
Light Fraction
T/D
35
125
35
235
38
30
18
17
100
10
50
1.7
2.0
0.1
15
711. 8b
(225)b
%
4.9
17.6
5.0
33.0
5.3
4.4
2.5
2.4
14.1
1.4
7.0
0.3
2.1
100.0
(3.16)b
Heavy Fraction
T/D
1.7
2.3
2.4
12.7
2.9
1.6
4.3
3.6
9.4
2.2
64.5
89.0
3.0
3.1
85.5
288.2
(25)
Iron Recovery
T/D
1.5
84.6
86.1
(1)
To Disposal
T/D
1.7
2.3
2.4
12.7
2.9
1.6
4.3
3.6
9.4
2.2
63.0
4.4
3.0
3.1
85.5
202.1
(24)
c.
Raw Refuse HHV = 4380 Btu/lb.
Expected moisture content in the storage silos is 30%. The difference is due to moisture loss in the
pneumatic systems and amounts to 16 TPD. Tonnage arriving in silos is 695.8 with HHV of 5674 Btu/lb.
Other non-ferrous.
-------
The flow diagram (Figure 5) indicates that for a 72 Mg/h (80 TPH) pro-
cessing rate, 50.4 Mg/h (55.6 TPH), or 70% of the input material, reaches the
power plant boiler. The Chicago refuse composition data, which was accumu-
lated during a relatively dry season, had an as-received moisture content of
25%. Table 2 shows the light fraction moisture content as 31.6%. This results
from the fact that the water is concentrated in the organic materials and as
inorganics are removed, the percentage of moisture in the remaining material
is higher. In turn, some moisture will be driven off by energy input during
shredding and carried away by the air of the several dust collection and pneu-
matic handling systems. Based on the St. Louis experience, where data were
collected over extended time periods, it is expected that the material fed to
the boilers will have a moisture content of approximately 30%. This indicates
a loss of nearly 1.2 Mg/h (1.3 TPH) of moisture as the material passes through
the system.
The composition of the raw refuse will vary somewhat from day to day, and
noticeably on a seasonal basis. Moisture content of the as-received refuse
will also vary according to the seasons. The percentage of yard waste will
increase in late spring and summer. These factors will cause some variation
in the heating value of the fuel in the storage bins. However, since the quan-
tity of fuel fired will be on a heating value basis, variations will be com-
pensated by adjustment of the quantity of fuel fired.
Another factor affecting the heating value of the fuel is the effective-
ness of the air classifier in separating the light and heavy fractions. Since
the unit involved is the first of its size in this service, the actual per-
formance will be determined after the unit is placed in operation. If the
unit takes off a larger percentage of light fraction, this fraction will con-
tain more inerts and also some larger particles of combustibles. The main
effect of this will be to increase the amount of bottom ash in the furnaces
and also possibly cause some loss of heat because of incomplete combustion of
the larger organic particles in the suspension-fired furnaces.
Energy Balance
Figure 6 presents an energy diagram of the supplementary fuel processing
plant, based on a raw refuse input of 72 Mg/h (80 TPH). The refuse energy
values used are those anticipated from the material composition shown in
Table 2. The HHV of the fuel leaving the processing plant is calculated to
be 13.20 MJ/kg (5674 Btu/lb). This is based on calorimetric tests of the
refuse mass balance components as found in the test program previously
described. Energy content,of the RDF stored in the power plant silos amounts
to 9.183 GJ/Mg (7.896 x 10 Btu/ton) of raw refuse processed.
Table 3 tabulates the energy balance of the system from raw refuse on the
processing plant tipping floor through fuel into the boiler combustion zone.
This shows a net energy availability of 8.208 GJ/Mg (7.058 x 106 Btu/ton) of
raw refuse processed or 1985 TJ (1,881 x 10^ Btu) for a system processing
235.85 Gg (260,000 tons) per year. Also shown is the amount of energy saved
by recycling ferrous metals as opposed to processing metal from virgin mate-
rials; the saving amounts to 2.904 GJ/Mg (2.497 x 106 Btu/ton) of raw refuse
68
-------
ON
VD
RAW REFUSE FEED
72.6 Mg/h, 10.19 GJ/Mg
FUEL FOR
OPERATION
0.147 GJ/Mg
1.13 GJ/h
1
ELECTRICAL POWER *
0.572 GJ/Mg
41.5 GJ/h
4
SUPPLEMENTARY FUEL
PLANT
i
72.9 GJ/h
SUPPLEMENTARY FUEL
50.4 Mg/h, 13.20 MJ/kg
NEW AVAILABLE ENERGY
RECYCLABLE MATERIAL
^| 21 1 GJ/h
IRON
6.25 Mg/h
ENERGY SAVED
RESIDUE TO LANDFILL
14.7 Mg/h
* ELECTRIC POWER GIVEN AS ENERGY IN
ELECTRICITY DIVIDED BY POWER PLANT
THERMAL EFFICIENCY, 33%
NOTE: ENERGY BALANCE DOES NOT OCCUR BECAUSE
OF RECYCLABLE MATERIAL CREDIT AND
MISCELLANEOUS SMALL LOSSES.
TOTAL ENERGY AVAILABLE
= 877 GJ/h
= 12.1 GJ/Mg
RAW REFUSE PROCESSED
Figure 6A. Energy diagram of Chicago supplementary fuel processing plant
(SI Units).
-------
FUEL FOR
OPERATION
0.127 x 106 Btu/ton
ELECTRICAL POWER
0.492 x 106 Btu/ton
1.07x
106 Btu/hr
39.3 x 106 Btu/hr
RAW REFUSE FEED
80 TPH, 8.76 x 106 Btu/ton
700.8 x 10 Btu/hr
I
SUPPLEMENTARY FUEL
PLANT
69.1 x 10° Btu/hr
RESIDUE TO LANDFILL
16.2 TPH
SUPPLEMENTARY FUEL
55.6 TPH, 5674 Btu/lb
631 x 10 Btu/hr
NEW AVAILABLE ENERGY
RECYCLABLE MATERIAL
200 x 106 Btu/hr
IRON
6.89 TPH
ENERGY SAVED
* ELECTRIC POWER GIVEN AS ENERGY IN
ELECTRICITY DIVIDED BY POWER PLANT
THERMAL EFFICIENCY, 33%
NOTE: ENERGY BALANCE DOES NOT OCCUR BECAUSE
OF RECYCLABLE MATERIAL CREDIT AND
MISCELLANEOUS SMALL LOSSES.
TOTAL ENERGY AVAILABLE
= 831 x 106 Btu/hr
= 10.4 x 106 Btu/ton
Raw Refuse Processed
Figure 6B. Energy diagram of Chicago ^supplementary fuel processing plant
(English Units).
-------
TABLE 3A. ENERGY RECOVERED/SAVED FROM 72.6 Mg/h SUPPLEMENTARY
FUEL SYSTEM (235 872 Mg/y RAW REFUSE PROCESSED)
Energy Used and Saved
Energy in Raw Refuse
Energy in Residue to Landfill
Energy in Supplementary Fuel
Fuel Energy Used:
Fuel required to operate vehicles
and front -end
Power required to operate facili-
ties (including Power Plant)
Building Heating Fuel
Total Energy Required
Net Fuel Energy Produced
Energy Saved:
0.0781 Mg of steel at 33.7 GJ/Mg
Tbtal Net Energy
Net Energy Available as Percent of
tfaw Energy in Refuse
Equivalent Mg of 23.26 MJ/kg coal
MJ/Mg
of
Raw Refuse
10 188
1 005
9 183
17
111
180
974
8 209
2 904
11 112
109%
0.43
TJ/y
2 403
237
2 166
4.2
184
42.2
230
1 936
685
2 621
109%
112 672
71
-------
TABLE 3B. ENERGY RECOVERED/SAVED FROM 80 TPH SUPPLEMENTARY FUEL
SYSTEM (260,000 TPY RAW REFUSE PROCESSED)
Energy Used and Saved
10-3 Btu
per Ton
of Raw Refuse
109 Btu/yr
Energy in Raw Refuse
Energy is Residue to Landfill
Energy in Supplementary Fuel
Fuel Energy Used:
Fuel required to operate vehicles
and front-end
Power required to operate facili-
ties (including Power Plant)
Building Heating Fuel
Total Energy Required
Net fuel Energy Produced
0.0861 tons of steel at
29 x 106 Btu/ton
Total Net Energy
Net Energy Available as Percent of
Raw Energy in Refuse (9.555/8/760)
Equivalent tons of 10,000 Btu/lb coal
8,760
864
7,896
15
668
155
838
7,058
2,497
9,555
109%
48
2,278
225
2,053
4
174
40
218
1,835
649
2,484
109%
• 124,200
72
-------
processed. Thus the total energy saved plus that made available is 11 112
GJ/Mg (9.555 x 106 Btu/ton) of raw refuse, or 2 616 TJ (2.48 x 1012 Btu) per
year. This is equivalent to a saving of 112.66 Gg (124,200 tons) of coal
The two power generating units to which the fuel is supplied have a total'
capacity of approximately 900 MW. At a 0.8 utilization factor these units
would normally consume approximately 3,150,000 tons of coal per year.
FACILITY DESCRIPTION
Figure 7 shows an external view of the supplementary fuel plant Figure
8 shows the site plan of the total facility while Figure 9 indicates the pro-
cessing equipment layout. Elevation sections are shown in Figure 10 and
Figure 11 details of the storage bins and material handling equipment on the
power plant site. The overall facility is composed of (1) the fuel processing
plant and (2) the fuel storage and boiler feed system facility. The project
feasibility study and fuel processing plant design were accomplished by
Parsons-Consoer, a joint venture of The Ralph M. Parsons Company of Pasadena,
California, and Consoer-Townsend and Associates of Chicago, under contract to
the Department of Public Works of the City of Chicago. The power plant on-
site facility was designed by John Dolio and Associates of Chicago, under con-
tract to the Commonwealth Edison Company.
Fuel Processing Plant
Capacity Analysis--
The plant is equipped with two independent lines each with a design capac-
ity of 72 Mg/h (80 TPH) . This provides a great deal of flexibility in plant
operation. The city's initial operating plan is to operate one 8-hour shift
5 days per week and process 907 Mg (1,000 tons) of raw refuse per day. Assum-
ing one process line will produce for 7-1/2 hours out of an 8-hour shift, for
a rate of 544 Mg (600 tons) per shift, the second line will have to operate
only 5 hours per day. The majority of maintenance will be conducted on the
second shift. Table 4 shows related material quantities for this mode of
operation.
TABLE 4. SUMMARY OF PLANT PRODUCTION, 907 Mg (1000 tons) SHIFT
Product Item
Raw refuse processed per day (normal operation]
Raw refuse processed per week
Raw refuse processed per year
Supplementary fuel delivered per day
(normal operationl
Supplementary fuel delivered per week
Supplementary fuel delivered per year
Fuel heating value
Fuel heating value per year
Ferrous product recovered per day
(normal operation)
Ferrous product recovered per year
Residue to landfill per day (normal operation)
Residue to landfill per year
Quantity
SI Units
907 Mg
4555 Mg
255 Gg
651 Mg
5157 tig
164 Gg
1.1.19 GJ/Mg
2167 TJ
78 Mg
20.2 fig
185 Mg
-17.6 Ug
linglish Units
1,000 tons
5,000 tons
260,000 tons
696 tons
5,480 tons
181 ,000 tons
11.55 x 106 Btu/ton
2,054,000 x 10b Btu
86 tons
22,500 tons
202 tons
52,500 tons
73
-------
Figure 7. City of Chicago
supplementary fuel plant
-------
On
Figure 8. Site plan of the City of Chicago -
Commonwealth Edison facility.
-------
fcO'
231'
qf
1 , TRUCK SCALE
Jj TRULK SCALE
(
,/«• 09 06) ^
_ -- -^J
riB
3
)l
PROCESSING EQUIPMENT
I PROCESS FEED CONVEYOR
2 DUST CONTROL HOOD
3 PRIMARY SHREDDER
TRANSPORT CONVEYOR
AIR CLASSIFIER
AIR CLASSIFIER CYCLONE
AIR CLASSIFIER INDUCTION BLOWER
SECONDARY SHREDDER
HEAVY FRACTION CONVEYOR
10 WASTE WATER TREATMENT
PNEUMATIC CONVEYOR
DUST CONTROL SYSTEM
(FROM TIPPING FLOOR )
DUST CONTROL SYSTEM
(HOODS AND ENCLOSURES)
DUST CONTROL SYSTEM
( PROCESS AIR )
CYCLONE SEPARATOR
COMPACTOR
HYDRAULIC POWER SUPPLY
DOCK LEVE.LER
MAGNETIC SEPARATOR
20 FERROUS METALS BIN
RESIDUAL MATERIALS BIN
AIR LOCK FEEDER AND HOPPER
23 FUEL CONVEYOR
24 MATERIAL TRANSPORT TRUCK
Figure 9. Floor plan of City of Chicago
supplementary fuel plant.
-------
SOUTH-NORTH SECTION
THROUGH TIPPING FLOOR
NORTH-SOUTH SECTION
THROUGH PROCESSING AREA
. , ^HEAVYFRACTIC
COARSE SHREDDER FINE SHREDDER /.' LAIR CLASSIFIER
DISCHARGE DISCHARGE /• 1 AIR CLASSI HER
BELT CONVEYOR ~> BELT CONVEYOR-1' FEEt CHUTE
FRACTION CONVEYOR
-FINE SHREDDER
- PRIMARY SHREDDER
FEED CONVEYOR -
HORI30NTAL SECTION
PRIMARY SHREDDER
DISCHARGE CONVEYOR -
PRIMARY SHREDDER
PRIMARY SHREDDER I
FEED CONVEYOR - 1
IMCLINED SECTION !
^VEST-EAST SECTION
THROUGH PRIMARY SHREDDER
0' 5' HP1 2p' 30' 49' 50' 60' ?o'
Figure 10. Elevation sections of The City of Chicago
supplementary fuel plant.
-------
00
-SWEEP DRIVE.
ASSEMBLY
PLATFOflM
WHEEL RAIL AROUND
PERIMETER OF SILO
RING PULL SEGMENTS
\\—— ft-Litf VENTFlLTLR A'
•v - VEMTflfLlEF STATION
- ROTARY AIRLOCK ;E£DER
AND INJECTION HS5EM8LI
SECTION A-A
FLOOR PLAN
ID1 IS' 20' 25'
Figure 11. Storage bins with material handling
equipment on the power plant site.
-------
As loads increase, it would be possible to run both lines 2 shifts per
day and increase production to 15-1/2 hours or 2250 Mg (2,480 tons) of raw
refuse processed and 1566 Mg (1,726 tons) of fuel produced. Presuming, on a
very conservative basis, that maintenance is required 1 line-shift for each
2 line-shifts of operation, it is possible to stagger operations so that only
one line is down for maintenance on any shift. Optimistically, operating
experience may show maintenance averaging out to one line-shift of maintenance
for five line-shifts of operation. In this case, the two lines could process
2866 Mg (3,160 tons) of raw refuse and produce 1996 Mg (2,200 tons) of fuel
in 24 hours of operation. Operating experience will establish the reality of
the situation.
Plant Location and Layout--
Based on the 1972 feasibility study, a 22 250 m (5-1/2 acre) site adja-
cent to Commonwealth Edison's Crawford Station was chosen for the processing
plant. The processing plant building is a reinforced concrete structure. The
floor, totally under roof, has a working area of 3307 m2 (35,600 ft2) and is
able to store as much as 1592 Mg (1,750 tons) of raw refuse v/hen piled 3.05 m
(10 ft) high, or nearly 2 shifts' production. A control room provides super-
vision of tipping floor operations and monitors the processing area by closed
circuit television. The whole processing area is serviced by an overhead
bridge crane. The office area is a two-story office and shop-type building
attached to the front of the main structure. This building contains a visi-
tors' entrance to a viewing gallery.
The outside area contains the storage bins and truck loading facilities
for the recovered metals and residue, a compactor for loading fuel or raw
refuse, the exhaust fans and filter baghouses for plant ventilation and dust
control systems, and the electrical substation yard.
Plant Operation--
Trucks entering the processing facility are controlled by traffic signals,
and the entire plant is remotely controlled by appropriate electronic equip-
ment. The flow of material is continuous from recording of the incoming refuse
to finished product with control of the processing being accomplished through a
combination of the rate at which waste is loaded onto the process feed conveyor
from the tipping floor and variation of conveyor speed.
Ferrous metals are recovered from each line at an approximate rate of 18
Mg (20 tons) every 3 hours. Disposal of this material will be determined at
the time a ferrous material sales contract is negotiated. The residue is dis-
posed of as landfill. An 18 Mg (20 ton) load of residue for landfill comes
from each processing line every 1-1/4 hour.
Process Equipment Trains—
Redundancy is provided in the form of two trains of processing equipment.
Dust control is accomplished by hooding or shrouding equipment or areas where
dust is generated and by maintaining negative air pressure in those areas.
The collected air is passed through baghouse filters to remove the dust before
being exhausted to the atmosphere.
79
-------
The most important pieces of equipment installed are described in the
following paragraphs.
The front-end loaders are heavy duty 5 cubic yard machines utilized for
handling the raw refuse on the tipping floor. They are diesel-powered artic-
ulated vehicles with foam-filled rubber tires. Four of these machines are
required, with three normally handling two operating lines and the fourth is
a spare.
The process feed conveyor is a two-section steel Z-bar conveyor, 2.44 m
(96 in.) in width. The first horizontal section is 9.14 m (30 ft) long, and is
mounted in a pit in the floor with the conveying surface 1.83 m (6 ft) below
tipping floor level. It has a reversible hydraulic drive and runs from 0 to
3.05 m (10 ft) per minute. The second section is also reversible and runs at
3 times the speed of the first. The purpose of having two sections operating
with a speed differential is to help break up the refuse and more evenly dis-
tribute the load over the conveyor surface. This is expected to result in a
more uniform feed rate to the shredder. The Z-bar slats are formed from at
least 9.5 mm (3/8 in.) thick plate and are suitably reinforced structurally.
The pit has steel-plated sides sloping into the conveyor at an angle of 30°
from the vertical. The second section sloped approximately 30° from the hori-
zontal that raises the material to the primary shredder mouth. The drive
mechanisms are remotely controlled from the central control room, and are
electrically interlocked with the drive motor of the primary shredder.
The primary or coarse shredder is a horizontal, reversible hammermill
rated for continuous duty at 72 Mg/h (80 TPD); it is direct driven by a 746
kW (1,000 HP) induction motor running at 900 rpm. The machine has a conveyor
entrance with a heavy steel hood to confine heavy material ballistically
thrown by the rotating hammers. The grate openings are set so that all mate-
rial intended for further processing will have a maximum dimension not exceed-
ing 20.3 cm (8 in.). The machine is designed with easy access to those parts
subject to rapid wear, such as hammers, impact plates, grates, and liners.
The design also incorporates easy-opening, hydraulically actuated access doors
in the grinding chamber. The impact plates are mounted in such a manner that
the hammer clearance may be hydraulically adjusted to compensate for wear.
A suitably matched, heavy duty, steel pan-type conveyor is provided to
catch the discharge material from the shredder. (The coarsely shredded refuse
may at times contain metal pieces ejected at high velocity.) This conveyor
discharges onto the coarse material conveyor. The end emerging from under the
shredder is enclosed with a metal dust control shroud that interfaces with a
dust cover on the coarse material conveyor.
The transport conveyor is a troughed rubber belt, 1.83 m (72 in.) wide,
sloped 20° to horizontal, and is electrically driven at a speed of approxi-
mately 1.27 m/s (250 ft/min). It carries the coarse shredded material to the
material intake of the air classifier and has a sectionalized dust hood that
can be opened by section for operational inspection.
80
-------
The air classifier accomplishes air/density separation of the shredded
material at a continuous rated capacity of 72 Mg/h (80 TPH). The system
includes the air classifier proper, a special light material conveying duct, 186
kW *(250 HP) material handling blower, and a cyclone-type material separator.
The coarsely shredded material falls continuously across a steeply slanted
vibrating tray. Below the mid-point, a strong stream of air enters through
slots and passes upward through the vibrationally energized bed of material.
The light fraction is captured by the air stream and is carried out of the
separator and through a cyclone separator, while the heavy fraction continues
to the end of the tray and falls out onto a conveyor. Induced draft air
enters the separator at both the material inlet and the heavy fraction outlet,
while the air entering the tray slots is blown in by forced draft blowers.
Provision is made to "tune" the system for best separation by adjusting the
various air stream velocities. The light fraction drops from the bottom of the
cyclone separator into the secondary shredder. The heavy density material from
the bottom of the air classifier is conveyed to a magnetic separator and then
dropped into transport trucks.
The secondary shredder completes the grinding of the combustible material
to an average dimension of 3.8 cm (1-1/2 in.). It has a continuously rated
capacity of 54 Mg/h (60 TPD), three quarters that of the primary shredder. It
is a heavy duty vertical ring-type grinder, gear driven by two 560 kW (750 HP)
electric motors.
The fuel conveyor receives the discharge of the secondary shredder. It is
a rubberized, troughed and cleated belt 1.37 m (54 in.) wide, and runs at a
speed of 1 m/s (200 ft/min). This conveyor leads to the air lock feed hopper
and is equipped with catwalks and a sectionalized metal dust cover that can be
easily opened for inspection and maintenance.
The air lock feeder has close-clearance rotary paddles, which introduce
the processed fuel into the pneumatic conveyor system pipeline. An enclosed
hopper receives the material from the secondary discharge conveyor to feed the
air lock. A cyclone separator is mounted on this hopper to separate out solid
material conveyed in the pneumatic dust conveyor transporting the material col-
lected from various dust collection systems.
The pneumatic conveyor system carries the processed fuel through two par-
allel 45.7 cm (18 in.) pipes about 488 m (1,600 ft), from the processing plant
to the two storage silos at the power plant. Each line receives air from 448
kW (600 HP) positive displacement rotary blowers, normally discharging at a
differential pressure of 20.7 to 34.5 kPa (3-5 psig). The lines are cross-
connected at the processing plant end so that either blower can operate either
pipeline and either pipeline can feed either storage bin. A branch line can
conduct material to a cyclone separator that feeds a transfer truck loading
compactor.
The heavy density material conveyor is a flat, rubber-cleated belt, with
stationary metal sides to retain the conveyed material. It is a constant
speed, electric-drive machine, and is sized to handle the material quantities
indicated on the flow diagram (Figure 5), plus 25%. The conveyor discharges
81
-------
onto a vibrating conveyor which evenly spreads the material and discharges it
close to the face of the magnetic separator.
The magnetic separator is a standard drum-type permanent magnet that picks
up the ferrous material in the heavy fraction and deflects it into the ferrous
storage bin. The unit is mounted on top of the ferrous and residual materials
retention bin assembly.
The material retention bins are bottom-dumping bins for holding the fer-
rous and residual materials. They are designed as enlarged chutes that receive
the product material streams leaving the magnetic separator, and have suffi-
cient capacity to retain the material, when the doors are closed, a long enough
period of time (about 1/2 hour) to exchange a full truck for an empty one.
The material transport trucks for carrying residue to landfill are open
top dumpers with an 18 Mg (20 ton) minimum capacity each. They are powered by
heavy-duty, diesel tractors.
The dust control systems include three air moving and dust control units
connected to each processing line. One system consists of a hood suspended
over the coarse shredder feed conveyor pit, necessary ducting, a large filter
baghouse, and an exhaust fan. The approximate air-handling requirement is
37.52 TST/S (79,500 CFM). This system maintains proper air quality for workers
in the tipping floor area and also controls dust there. The largest concen-
trated dust generating area is the feed conveyor pit. By exhausting air at
this point, currents are induced that sweep the tipping floor and move dust
generated from truck unloading toward the pit for ultimate removal. The fil-
ter baghouse collects the dust and prevents atmospheric contamination outside
the building. This system only handles fine airborne dust.
A second major system is for air classifier dust control. This unit
handles approximately 31.86 m3/s (67,500 CFM) of air plus moisture (picked up
from the waste) and processes the dust-laden material passed through the air
classifier. Large particles of paper or sheet plastic are kept out of the
baghouse by using a dropout box and a secondary cyclone ahead of the filter
baghouse.
A third and smaller system has a capacity of approximately 7.08 m /s
(15,000 CFM) and handles dust laden air from the various dust hoods and other
collection points in the process train.
The baghouses used are continuously self-cleaning, with antistatic bags
as filters. The baghouses, dropout boxes, and secondary cyclones are equipped
with rotary air locks that pass the collected material into a pneumatic con-
veyor system for transport to the air lock feed hopper, where it is added to
the processed fuel.
Auxiliary Equipment — In addition to the process equipment, several major
support items are essential to operation of the plant.
82
-------
The waste water treatment system provides the tipping area and process
floors with a drainage system that may pick up substantial amounts of solid
materials during periods of heavy water flow, such as washdown operations or
fire-fighting. The drainage water passes through a wastewater pre-treatment
system consisting of scum and solids collecting equipment. The clear water
flows to the sewer system.
A travelling bridge crane serves the process area. It moves overhead
with a variable speed drive and inching control, and is equipped with an 18
Mg (20 ton) hook.
A house air supply of sufficient capacity to handle such items as filter
baghouse operation, instrumentation, air-operated controls, pneumatic tools,
and other maintenance operations is provided. The supply provides 100% redun-
dance in the form of two standard 93 kW (125 HP), 689 kPa (100 psi), packaged
units.
Two truck scales are used to weigh refuse delivery trucks in the entrance
roadway. These are standard weighing scales of 45 Mg (50 ton) capacity, 21.3
m (70 ft) long and 3.05 m (10 ft) wide. They are automatic, including entrance
and exit gates, and are provided with remote data reading, recording equipment,
and manual controls located in the plant control room.
The heating, ventilating, and air conditioning system for the facility con-
sists of a number of independently controlled areas. The tipping floor area is
unheated because of the large quantity of ventilation air required to remove
equipment exhaust gases. The truck entrance and exit doors remain open during
plant operations (except in severe weather) to allow air to sweep across the
tipping floor to the hoods over the conveyor pits. During periods when the
process equipment is not running and the main exhaust system is off, venti-
lation is provided by roof-mounted fan units. Electric heating elements are
embedded in the floor for a distance of 3.7 m (12 ft) around the conveyor pits.
This prevents the formation of floor ice in cold weather close to the pits,
which might be a hazard to the front-end loader operations.
The processing area has large quantities of air passing through it because
of requirements for the air classifiers. Incoming air is warmed to a minimum
of 13°C (55°F) in accordance with the City building code by large heating units
mounted against the inside of the rear process room wall. Wall-mounted gas-
fired space heaters are provided near the working floor levels during periods
when the process equipment is down for maintenance.
The office and shop areas include offices, lunchroom, toilets, and locker
rooms. These are heated and air-conditioned. A central chilled water system
is provided for refrigeration. Other areas are provided with gas-fired heated
and filtered air from a central air handling system. The control room is
heated, ventilated, and air-conditioned, and the visitors' gallery is air con-
ditioned and, in winter, heated by baseboard type electrical heaters.
A fuel supply is required only for the vehicles, and this is provided by
the existing facility at the street maintenance shop already on the property.
83
-------
A fire protection system offering complete sprinkler coverage is provided
to protect the plant, in accordance with local and national codes. All areas
have a wet system except the tipping floor area, which has a dry type because
the room is subjected to freezing temperatures. The tipping floor is also pro-
vided with fire hoses to combat refuse fires anywhere on the floor. Other
areas are provided with wall-mounted hand extinguishers and hoses in accordance
with established standards.
The electric power distribution system within the facility is supplied by
the Commonwealth Edison Company through a transformer station located on the
property. Two 7,500 kVA transformers receive high voltage power and deliver
4,160 V, 3 phase, 60 Hz power to the facility metering panels. The facility
distribution system consists of two duplicate systems of 7,500 kVA each. Tie-
line switching is provided so that either Edison transformer can feed either
load center substation and distribution system. Each system carries one of
the processing lines plus some auxiliary equipment. Motors of 298 kW (400 HP)
or larger are supplied by 4,160-V distribution sections of each system.
The remote control system is located in the control room. Central panels
are provided for each line for independent operation. The material processing
rate is controlled by the rate at which the front loaders place material on the
primary feed conveyor and also by the central process controller adjusting the
speed of the primary feed conveyor. Each line is set up with interlocked con-
trols for automatic sequential start-up and shutdown of each major component of
the line. The central control system can manually override the automatic pro-
gramming. Local start-stop controls are provided at each piece of equipment to
permit emergency shutdown or to facilitate maintenance operation.
Power Plant Facilities
Capacity Analysis--
The boilers of Generating Units 7 and 8 of the Crawford Station have been
modified to receive supplementary fuel. Both units are similar in configura-
tion, with their operating cycles employing separately-fired reheat sections.
The main furnace and the reheat furnace of both units are fitted to fire sup-
plementary fuel. Table 5 shows pertinent generating unit data.
Unit 7 is rated at 238,360 kW with a steaming rate of 658 000 kg/h
(1,450,000 Ib/hr) at 16.201 MPa (2,350 psia) and 565°C (1,050°?). The reheat
section operates at approximately 3.826 MPa (555 psia) and reheats to the
original temperature. The unit is equipped to feed up to 13.6 Mg/h (15 TPH) of
supplementary fuel into the main furnace for a burning rate of approximately
7.27% of the heat requirement. The maximum feed capability to the reheat fur-
nace is also 13.6 Mg/h (15 TPH), but at 10% of the heat release, the feed rate
would be 54.0 Mg/d (59.5 TPD). The total unit would therefore consume 381 Mg/d
(420 TPD), requiring up to 548 Mg/d (604 TPD) of raw refuse to be processed.
Unit 8 is rated at 358 160 kW with a steaming rate of 998 000 kg/h
(2,200,000 Ib/hr). The steam conditions are the same as for Unit 7 except
that reheat pressure is 4.102 MPa (595 psia). The unit is equipped to feed
up to 21.8 Mg/h (24 TPH) of supplementary fuel into the main furnace or a
84
-------
TABLE 5A. CRAWFORD STATION GENERATING UNIT DATA (in SI units)
Item
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Name Plate Rating - kW
Steaming Rate - kg/h
Steam Pressure - MPa
Steam Temperature - °C
Reheat Pressure - MPa
Reheat Temperature - °C
RDF - Max. Firing Rate - Main Furnace - Mg/h/Mg/d
RDF - Max. Firing Rate - Reheat - Mg/h/Mg/d
RDF - Max. Main Furnace Burning Rate -
% Heating Requirement
RDF - Reheat Furnace Firing Rate @ 10%
Heating Requirement - Mg/h/Mg/d
RDF - Total Required at 10% Max. Burning Rate
or Max. Capability - Mg/h/Mg/d
Raw Refuse - Max. Daily Process Requirement - Mg
Unit No. 7
239 360
658 000
16.201
565
3.826
565
13.6/327
13.6/327
7.27
2.25/54.0
15.9/381
548
Unit No. 8
358 160
998 000
16.201
565
4.102
565
21.8/522
21.8/522
7.66
3.12/74.9
24.9/597
858
-------
TABLE 5B. CRAWFORD STATION GENERATING UNIT DATA (in English units)
oo
0\
Item
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Name Plate Rating - kW
Steaming Rate - Ib/hr
Steam Pressure - psia
Steam Temperature - °F
Reheat Pressure - psia
Reheat Temperature - °F
RDF - Max. Firing Rate - Main Furnace - TPH/TPD
RDF - Max. Firing Rate - Reheat - TPH/TPD
RDF - Max. Main Furnace Burning Rate -
% Heating Requirement
RDF - Reheat Furnace Firing Rate at
10% Heating Requirement - TPH/TPD
RDF - Total Required at 10% Max. Burning Rate
or Max. Capability - TPH/TPD
Raw Refuse - Max. Daily Process Requirement - Tons
Unit No. 7
239,360
1,450,000
2,350
1,050
555
1,050
15/360
15/360
7.27
2.48/59.5
17.5/420
604
Unit No. 8
358,160
2,200,000
2,350
1,050
595
1,050
24/576
24/576
7.66
3.44/82.6
27.4/658
946
-------
burning rate of 7.67% of the heating requirement. The feed capability to the
reheat furnace is also 21.8 Mg/h (24 TPH), which is several times the antici-
pated 10% burning rate, but at 10% the feed rate would be 3.12 Mg/h (3 44 TPH)
The total unit burning rate is 24.8 Mg/h (27.4 TPH) or 597 Mg/d (658 TPD)
requiring up to 858 Mg/d (946 TPD) of raw refuse to be processed.
The total RDF consumption of both units running at name-plate rating
would be 978 Mg/d (1,078 TPD), requiring 1 406 Mg (1,550 tons) of raw refuse
to be processed. The average load factor for both units is such that the pro-
cessing plant load will probably not exceed 907 Mg/d (1,000 TPD). At some
later date it might be advantageous to increase the capacity of the main fur-
nace feed systems so that a higher rate of RDF could be burned and thereby
better advantage be taken of the processing plant capacity.
Facility Layout and Features--
The route of the RDF pneumatic transfer lines is approximately 488 m (1,600
ft) long from the interior of the processing plant to the silos. Two storage
silos ar.e provided, one of 1 700 m3 (60,000 ft3) capacity and the other of
2 550 m-' (90,000 ft3). The smaller one feeds Unit 7 and the larger one Unit 8.
Although the average density of material in the storage silos has yet to be well
established, it is anticipated that they will hold 272 Mg (300 tons) and 408 Mg
tons) respectively of RDF. This will permit the processing plant to operate on
one shift per day and the fuel firing to continue around the clock. Firing on
weekends is not contemplated unless there happens to be left-over fuel in the
silos.
The silos shown in Figure 11 are designed by Atlas Systems Inc., of
Spokane, Washington. Atlas also furnishes the patented material retrieval and
handling equipment. The silo is an inverted cone, in the top center of which
is the end of the pneumatic transfer pipeline from the processing plant. The
material leaves the pipeline and falls in a conical pile on the concrete floor.
The air from the pneumatic transfer line is exhausted to atmosphere by an
exhaust fan pulling through a dust filter baghouse.
The mechanical equipment is housed in a room under the silo floor, the
walls of which provide the outer foundations for the silo. The retrieval mech-
anism consists of a four-unit (located 90° apart) sweep-bucket chain that is
dragged around the floor of the silo by a chain drive mechanism circling the
periphery of the floor. The sweep buckets drag the material from the edge of
the pile and cause it to fall onto belt conveyors installed in trenches in the
silo floor. Four of these are provided which feed radially inward (in this
case, but could work outwardly) to discharge the material into hoppers above
the rotary airlocks feeding the pneumatic pipelines to the furnaces. A frag-
mentizer is provided at the discharge of each conveyor to break up any large
agglomerates of caked material that may come down the conveyor. The hoppers
over the airlocks contain impact-sensitive flowmetering devices that measure
the weight of material entering the airlocks. They are equipped with a divert-
ing mechanism for taking weight samples for calibration purposes. These mea-
surements, plus periodic heating value analyses of samples of the material,
form the basis for Commonwealth Edison's payments to the City for fuel value
87
-------
received. The sweep bucket system and conveyors have variable speed electric
drives that are remotely controlled from the boiler control center to regulate
the RDF feed rate to the furnaces.
The silos are connected to the furnaces by four 20.3 cm (8 in.) pneumatic
material conveyors. The smaller silo is connected to Unit 7 and the larger
silo to Unit 8. Each line is powered by its own 74.6 kW (100 HP) positive dis-
placement rotary blower. In addition, each silo system has a diverter valve
and transfer line to the other silo so that, if necessary, fuel may be trans-
ferred between silos. The length of the transfer lines to the boilers is
approximately 152 m (500 ft), including change in elevation.
Modifications to the boilers, to allow for the introduction of supple-
mentary fuel, are relatively minor. They involve simple penetration of the
windbox and furnace wall by the pneumatic fuel transfer lines and mounting of
elevating fuel injector nozzles at the end of the lines, inside the furnace
wall. Some modification is also required in the boiler control system to pro-
vide for regulation of the RDF feed rate and to handle possible changes in the
boiler operating characteristics when firing RDF.
The silos are provided with a complete interior sprinkler system for fire
protection. An ultrasonic height detector is built into each silo for con-
tinuous monitoring of the fuel level. These instruments are also connected to
visual display panels in the processing plant control room as well as in the
power plant control center.
The additional bottom ash resulting from the supplementary fuel does not
place an excessive burden on the existing ash handling system, and the existing
electrostatic precipitators on the boiler exhaust gas systems are expected to
control the additional fly ash generated.
CONSTRUCTION COSTS
The design and construction costs, as given in Table 6 are considered to
be very accurate, being based on fixed price contracts for construction and
equipment procurement. The mid-point of construction for this project was the
fourth quarter of 1975. The costs cover construction costs only and do not
include construction management, working capital, start-up, or other costs
normally considered in financing arrangements. The City acted as its own con-
struction manager for the processing plant and Commonwealth Edison provided
management for the power plant facility construction.
OPERATING COSTS
Operating costs include direct labor, maintenance services and supplies,
utilities, fuel, and residual material disposal costs. The estimates are
reported here on an annual basis and are summarized separately for the pro-
cessing plant and the power plant facility.
88
-------
TABLE 6. DESIGN AND CONSTRUCTION COST ESTIMATE SUMMARY
Item
1. Processing Plant:
• Site Preparation and Below Grade Foundations
• Above Grade Building with Auxiliaries
• Process Equipment
» Process Equipment Installation
• Electrical Equipment and Installation
1
Cost
($000)
1100
6947
2743
1073
1488
Subtotal
Power Plant Facilities
* Prime Construction Contract
• Windbox Modifications and Miscellaneous
Subtotal
Engineering and Construction Supervision*
Total
4228
272
13,351
4,500
1,000
18,851
* Estimated City costs only.
Processing Plant
Table 7 summarizes the Processing Plant Operating costs.
TABLE 7. SUMMARY OF ESTIMATED PROCESSING PLANT
OPERATING COSTS - 256 Gg/y (260,000 TPY)
Item
$/year
Labor
Electric Power
Maintenance Supplies § Services
Miscellaneous Utilities
Residual Disposal
Total
815,000
354,000
353,900
84,500
151,000
1,638,400
Cost per Mg of Raw Refuse - $6.94
Cost per Ton of Raw Refuse - $6.30
89
-------
Direct Labor is the largest item of operating costs. The summary of costs
presented in this section is based on a detailed review of all probable job
assignments related to the plant size and type and an anticipated operating
schedule based on processing 907 Mg/d (1,000 TPD) of raw refuse in one 8-hour
shift, 5 days per week. This will require that the majority of maintenance be
accomplished on a second shift each of the operating days. Table 8 details the
direct labor requirements and costs for the Processing Plant.
TABLE 8, PROCESSING PLANT DIRECT LABOR REQUIREMENTS AND COSTS
Direct Job Costs:
Job Classification
Plant Superintendent
Plant Engineer
Shift Supervisor
Process Operator
Sick Leave § Vacation Man
Front -Loader Operator
Traffic Director
Equipment Monitor
Laborer
Electrical Mechanic
Maintenance Mechanic
Janitor
Clerk/Stenographer
Subtotals
Base
Rate
$/Year*
22,750
18,000
20,800
18,900
20,300
20,300
14,500
16,600
12,600
21,000
20,500
10,400
8,000
No. Personnel
1st
Shift
1
1
1
2
1
3
1
2
3
1
1
1
18
2nd
Shift
1
2
1
2
1
7
Total
1
1
2
2
1
3
1
2
5
2
3
1
1
25
$/Year
22,750
18,000
41,600
37,800
20,300
60,900
14,500
33,200
63,000
42,000
61,500
10,400
8,000
433,950
Other Labor Costs:
Emergency overtime @ 7-1/2%
Labor Subtotal
Fringe Benefits and City Overhead @ 75% of Labor
TOTAL Labor Costs per Annum
52,550
466,500
550,000
816,500
Average Cost per Mg of Refuse--
Average Cost per Ton of Refuse-
$5.46
$5.14
* Wage rates are based on known City of Chicago rates for comparable job
classifications adjusted to 1975 dollars.
90
-------
Table 9 lists the various electrical loads connected with each processing
line. The anticipated average demand at rated load is 2,525 kW per line. In
addition, it is estimated that lighting and miscellaneous other equipment will
average a demand of 700 kW while the plant is running and on the maintenance
shift. The estimated demand at other times is 300 kW. On this basis, the
electrical consumption per year will be 12,491,000 kWh, and the cost/at an
average rate of $0.0283/kWh will be $354,000. The average electrical con-
sumption per Mg of refuse processed is then 52.9 kWh at a cost of $1.50/Mg
(48 kWh per ton at a cost of $1.36/ton).
In that a large maintenance staff is a part of the base payroll, the
Maintenance Supplies and Services item only covers parts and special services,
and is estimated at 1-1/4% of the purchase price of fixed mechanical equipment
and 1% of building costs. In addition, a special cost of $0.44/Mg ($0.40/ton)
of raw refuse processed is added for the rapid wearing parts of the shredders,
such as the hammers, for an annual cost of $104,000. The percentage rate used
on the mechanical equipment estimate is about half of the normal because of the
low utilization rate of the equipment at 907 Mg/d (1,000 TPD) raw refuse
processed.
The maintenance cost for the front loaders is calculated at 5% of the pur-
chase price, or $9,000. In addition, it is assumed that this equipment will
have to be replaced every 7 years. Therefore, a sinking fund payment of
$21,100 is provided for. This fund is anticipated to earn compound interest at
6-1/2%.
TABLE 9. ELECTRICAL POWER LOADING PER PROCESSING LINE
Primary Feed Conveyor Horizontal
Inc1i ned
Primary' Shredder Main
l.uhnak
Primary Disch. Apron Conveyor
Coarse Pisch. Belt Conveyor
Air Classifier Vibrator Drive
Ai r Bloivers (2}
Ai r P:\haust cr
10.
11.
12.
13.
14.
15.
Fine Shredder Main
l.uhpak
Pine Shredder Discharge Conveyor
Pneumatic Pipeline Feeder
Pncuniatic Pipeline Blower Main
Cool i nil Pan
\ir Classifier Induct ion Mm
11)110
IS
5
20
1500
5
20
15
Average
Demand
kiv
(..9
13.8
14.9
1 I .8
14".(i
I! 1 .9
.18. (i
35.0
25.2
l.i. h
91
-------
Within the Miscellaneous Utilities account is fuel for6the front loaders
estimated at $12,000; gas for heating ($0.71/GJ or $0.75/10 Btu) at $43,200;
and water and sewer charges, plus telephone and miscellaneous supplies and
services, at $30,000.
Residue Disposal cost is calculated on the basis of hauling at $0.17 per
Mg-km ($0.25 per ton-mile) for a one-way trip of 11.3 km (7 miles) to the City-
owned Stearns Quarry disposal site, plus costs at the quarry of $0.66/Mg
($0.60/ton). The total cost of residue disposal is estimated at $131,300.
Power Plant Facility
The complete costs of operating and maintaining the power plant facility
are borne by Commonwealth Edison Company. Table 10 gives costs by in-plant
category and by unit of refuse processed, as well as by heating value unit.
TABLE 10. SUMMARY OF OPERATING COSTS AT THE POWER PLANT FACILITY
Item $/Year
Labor 129,800
Electric Power 106,200
Maintenance Labor § Supplies 118,000
Miscellaneous Utilities 8,000
TOTAL 362,000
Cost per Mg of Raw Refuse $1.53
Cost per Ton of Raw Refuse $1.39
Cost per GJ $0.167
Cost per 106 Btu $0.176
The system is highly mechanized and will be controlled from the boiler
control center; it is anticipated that one man will be required to each shift
continuously monitor the operation of both silos.
An extra shift of labor is provided each week at overtime rates in the
event it is necessary to run into the weekend to consume all the fuel processed.
In addition to this, it is expected that a laborer will be assigned to 1 shift
per day for housekeeping purposes for a total of 4 men. Direct labor costs are
then as follows:
3 Equipment Monitors § $16,600 $ 49,800/yr.
1 Laborer @ $12,600 12,600/yr.
Subtotal $ 62,400/yr.
92
-------
1 shift/week Overtime 5
Fringe Benefits and Overhead @ 100% 62 400
TOTAL $129,800/yr.
Cost per Mg of Raw Refuse $0.55
Cost per Ton of Raw Refuse $0.50
The average demand at the power plant site is 359 kW for the various
blowers, fans, conveyors, drives, and lights. The estimated power consumption
is 19.9 kWh/Mg (18.1 kWh per ton) of raw refuse processed, or 4,480,000 kWh/yr.
The cost for electric power at an average of $0.0237/kWh is $106,200 per annum.
It is anticipated that maintenance will be conducted by the existing power
plant maintenance department with additional staff added. Because the equip-
ment will have a reasonably high usage factor, a maintenance cost of $118,000
has been estimated at 5% of the original cost on mechanical equipment and 1% on
the silo building, including labor and supplies.
The Miscellaneous Utilities allowance of $8,000 is for expenses not covered
elsewhere. There is no heating fuel requirement because the silo equipment room
is heated by electric units. Water consumption is negligible. A typical
expense would be the power plant share of the direct wire telephone connection
to the processing plant control room.
REVENUES
Revenues to the City will be sale of recovered ferrous material and sale
of fuel to the Commonwealth Edison Company. While the ferrous metal market
fluctuates considerably, for estimating purposes the yearly income is considered
to be $200,000. The original agreement with Commonwealth Edison Company, made
before energy cost escalations, called for the utility to pay the city $0.30 per
million Btu for the heating value of the RDF deposited in the silos. The sys-
tem includes provision for weighing the incoming material and Edison plans on
measuring heating value of the fuel periodically. The anticipated yearly
revenue from this source, based on 2 167 TJ (2,053,000 x 10& Btu) delivered
would be $616,000.
EXISTING PLANT COST SUMMARY
Costs of raw refuse disposal to the City of Chicago by the system is sum-
marized in Table 11. Not included are costs borne by Commonwealth Edison Com-
pany, which are, of course, reflected in the fuel price.
EFFECT OF PRODUCTION RATE CHANGE
The preceding sections presented the costs for operating the plant at the
rate of 907 Mg/d (1,000 TPD). The original design concept was predicated on a
plant capable of processing a minimum of 1451 Mg/d (1,600 TPD) with a 100%
processing equipment redundancy. The design described fulfills those criteria
93
-------
TABLE 11. OVERALL COST SUMMARY - PRODUCTION RATE OF 907 Mg/d
(1,000 TPD), 236 Gg/y (260,000 TPY)
Item
Amortization:
$18,851,000; 20 yrs @ 4.95%
Operating and Maintenance
Subtotal
Revenues :
Ferrous - $200,000/yr.
Fuel sales - $616,000/yr.
NET COST
Annual Cost, $
1,506,000
1,638,000
3,144,000
816,000
2,328,000
$/Mg
Raw
Refuse
6.38
6.94
13.32
3.46
9.86
$/Ton
Raw
Refuse
5.79
6.30
12.09
3.14
8.95
by providing two identical processing lines, each capable of processing 1451
Mg/d (1,600 tons) of refuse at utilization rate of 83-1/3%. While one line is
operating, maintenance can be performed on the other line. This would permit
processing 377 Gg (416,000 tons) of raw refuse per year. It essentially means
operating one line 20 hours per day 5 days per week. Table 12 summarizes the
plant operating costs under these conditions.
TABLE 12. SUMMARY OF OPERATING COSTS AT PROCESSING RATE
OF 1451 Mg/d (1,600 TPD), 377 Gg/y (416,000 TPY)
Item
$/Year
Labor
Electric Power
Maintenance Supplies and Services
Miscellaneous Utilities
Residue Disposal
Total
Cost per Mg of Raw Refuse .
Cost per ton of Raw Refuse
$5.69
$5.16
1,124,000
340,000
363,000
109,000
210,000
2,146,000
94
-------
The direct labor requirements and costs are shown in Table 13.
TABLE 13. PROCESSING PLANT DIRECT LABOR
REQUIREMENTS AND COST
Base
Rate
$/Year
Plant Superintendent 22,750
Plant Engineer 18,000
Shift Supervisor 20,800
Process Operator 18,900
Sick Leave & Vacation Man 20,300
Front-Loader Operator 20,300
Traffic Director 14,500
Equipment Monitor 16,600
Laborer 12,600
Electrical Mechanic 21,000
Maintenance Mechanic 20,500
Janitor 10,400
Clerk/Stenographer 8,000
Subtotals
No. Personnel
1st
Shift
1
1
1
1
1
2
1
3
3
1
1
1
15
2nd
Shift
1
1
1
1
1
2
2
1
1
1
12
3rd
Shift
1
1
1
*
1
2
2
1
1
8
Total
1
1
3
3
2
4
1
7
7
3
3
1
34
$/Year
22,750
18,000
62,400
56,700
40,600
81,200
14,500
49,800
88,200
63,000
61,500
10,400
8,000
597,350
Other Labor Costs:
Emergency Overtime @ 7-1/2%
Direct Labor Subtotal
Fringe Benefits and City Overhead @ 75% of Labor
Total Labor Costs per Annum
Average Cost/Mg of Refuse $2.98
Average Cost/Ton of Refuse $2.70
* 2 men float to cover 3 shifts.
44,800
642,150
481,612
1,123,762
95
-------
The electric power requirement is 51.3 kWh/Mg (46.5 kWh/ton) for a total
of 19,335,000 kWh/yr. The average cost is $0.0176/kWh for a total cost of
$339,300 per annum. It is of interest to note that the power cost for this
mode of operation is less than for processing 907 Mg/d (1,000 TPD) on a one
shift per day basis. This is because when running one line vs. two, the
demand charge portion of the power bill is reduced enough to more than offset
the additional cost of the energy charge, which is all billed in a low cost
block of the rate schedule.
Maintenance supplies would increase due to additional wear and tear and
is estimated at $362,500/yr.
Miscellaneous utilities will increase because of increased heating costs
resulting from operating the process line three shifts per day. Additional
fuel will be required for the front-loaders, but other items will have negli-
gible increases. The total is calculated at $109,000/yr.
Residue disposal will increase in direct proportion to the increased pro-
duction to $210,000/yr.
Revenues will also increase in direct proportion to production, assuming
a consumer is available. The overall cost summary is presented in Table 14.
TABLE 14. OVERALL COST-SUMMARY - PRODUCTION RATE OF 1451 Mg/d
OR 377 Gg/y (1,600 TPD, 416, 000 TPY)
Item
Amortization: $18,851,000 for
20 yrs @ 4.95%
Operating
Revenues :
Ferrous
and Maintenance
Subtotal
$320,000
Fuel Sales 986,000
NET COST
Annual
Cost - $
1,506,000
2,146,000
3,652,000
1,306,000
2,346,000
$/Mg Raw
Refuse
3.99
5.69
9.68
3.46
6.22
$/Ton Raw
Refuse
3.62
5.16
8.78
3.14
5.64
96
-------
ENVIRONMENTAL CONSIDERATIONS
Within the supplementary fuel processing facility itself, environmental
effects can result from atmospheric emissions, water effluent, and noise
None of these are anticipated to create any particular problems not solvable
within the existing design or with minor modifications to it.
All refuse placed on the tipping floor is processed continuously and none
remains in excess of 8 hours. In the event of any downstream processing mal-
functions, city trucks will remove the material to alternative disposal sites.
Anaerobic digestion of the wastes to odoriferous compounds is limited because
of the short duration of open exposure and the rather high degree of air con-
tact with the low density material. The temporary storage area is under a
slightly negative pressure, with all air being ultimately exhausted through a
baghouse fabric filter system. Similar filter systems collect all dust from
hoods at critical areas of the processing equipment. All baghouse dust is
periodically introduced into the supplementary fuel transport line for com-
bustion in the Commonwealth Edison furnaces.
Operating experience must be gained as to the extent of water contamina-
tion resulting from plant operations. The design incorporates a package treat-
ment unit for removal of scum and suspended solids prior to introduction of the
clear effluent to the sewer system. The level of soluble BOD contamination of
this water is not known at this time. Other package treatment units could fur-
ther reduce the organic level if this proves necessary.
Noises generated by conveyors, fans, and shredders are contained within
the building structure. The highest noise generators are the shredders (2
primary and 2 secondary), located in a room separated from the main receiving
area by a heavy concrete wall. The room is large so that noise is diffused to
a great extent. When maintenance is performed while a shredder is in operation,
workers wear prescribed noise-reducing ear muffs. The shredders are of espe-
cially heavy construction, both in body and covered hood, thus transmitting
less noise than usually found with size reduction equipment. Otherwise, little
is transmitted to the receiving area and will be less than 66 dB(A) at the pro-
perty boundaries as required by the City of Chicago Heavy Manufacturing District
Zone code.
Noise from air compressors for the pneumatic system is led through a
silencer to prevent noise transmission through the air inlet. The shredded
material in the transport pipe also absorbs the sound and the air finally exits
through cyclones and baghouses that also reduce the sound to acceptable levels.
Truck noise on the receiving floor is contained within_the building. Con-
veyor noises are essentially low and contained in the building.
Environmental considerations at the power plant site are similar to other
facilities designed to store, transport, and burn_RDF. Principal experience
has been gained at St. Louis, where rather extensive testing has demonstrated
no undue changes in effluents occur over combustion of100% coal in modern
furnaces Further tests must be conducted to establish the exact extent of
emSsions and the best control technology to employ and no such tests have yet
97
-------
been made at Chicago. The St. Louis data (Refs. 8 and 9), taken at RDF firing
rates varying up to 27 percent on a power output basis, indicate no statisti-
cally significant changes in NO or SO emissions. Average emission of chlo-
rides, for which there are no local or federal air quality standards, was
approximately 30 percent higher during combined firing operations in the 125
MW furnace. Samples for particulate matter examination were taken both before
and after the 97.5 percent efficient electrostatic precipitator. Upstream of
the precipitator, in terms of mass of particulates per unit boiler heat input,
there was no change in particle loading when RDF was added to the furnace.
Downstream, no change occurred up to the design capacity of the boiler. Above
125 MW, however, particulates increased with combined firing. The data can-
not be fully explained yet, although several theories have been developed and
solutions suggested if a utility might desire to operate above design stream
output levels.
Shredded combustibles will not be stored more than two days at the power
plant. Experience at Madison, Wisconsin, and St. Louis indicate that no odors
should develop during this time.
98
-------
SECTION 6
GEORGIA INSTITUTE OF TECHNOLOGY MOBILE AGRICULTURAL PYROLYSIS SYSTEM
INTRODUCTION AND SUMMARY
The Georgia Tech Engineering Experiment Station (EES) has been conducting
R S, D on an air-blown pyrolysis converter for agricultural wastes. The EES
group has proposed that equipment for processing 182 Mg/d (200 TPD) of wet
wastes into 40.9 Mg (45 ton) of a char-oil fuel mixture could be contained on
two standard highway trailers and thus could offer the potential for moving
from source to source of waste on a periodic basis. Design operating condi-
tions were selected from a limited number of experiments from pilot units at
the EES processing 22.7 and 5.5 dry Mg/d (25 and 6 TPD). The developers pre-
pared equipment listings, suggested a suitable configurational layout, and
derived cost estimates for the proposed mobile system. As the most develop-
mental system in the group of candidates, the mobile pyrolysis plant could
only be examined on a rather conceptual basis by Parsons. The experimental
results appear entirely reasonable when compared to similar pyrolysis studies
and have been accepted as the basis for a system that is reviewed here.
Design comments are made and probable costs for a single system and lots of
100 have been estimated by Parsons specialists.
CONCLUSIONS
• The EES agricultural pyrolysis system is the least developed of the
candidates analyzed here. While only limited experimental tests have
been made on which energy and material balances can be based, cellu-
losic pyrolysis chemistry is sufficiently well understood to conclude
the proposed system is technologically sound. R$D remaining to be
conducted will primarily help establish the economics of the mobile
concept, while also increasing knowledge of properties of the fuel to
be produced. If practical hardware can be assembled economically, the
system should prove of great merit in converting farm-related waste
materials to energy.
• It is recommended that component-level scale-up testing be accomplished
prior to building a two or three trailer prototype unit. The latter
effort would be a rather straight forward construction task once fur-
ther R&D demonstrates necessary design features to be used. The fol-
lowing components require study:
- Waste Dryer - Equipment for evaporating water from natural
products is designed on an empirical basis. The dryer would
99
-------
be the largest item in the system and its performance and
energy needs must be determined with a full size unit.
- Pyrolysis Converter - The pilot plant pyrolysis equipment was
intended to yield information on the quantities of gas, liquid,
and char resulting from waste decomposition under varying con-
ditions. Scale-up and incorporation of realistic engineering
features can reveal problems that must be studie4 with an
essentially full size unit. A waste introduction device,
stirrer/air introduction equipment, insulation, product con-
denser, and bottom char release unit all need to be
investigated.
- Combustor - The efficiency and performance of a commercial or
custom combustor for producing the hot drying gas must be
established.
- Power Generator - Moderately extensive use of the off-gas as a
fuel for an internal combustion engine should be conducted to
establish power output and effects on engine maintenance.
« The Parsons' design review is unable to verify the .suggested perfor-
mance capability of the Georgia Tech dryer. While the art of drying
of natural products is quite empirical and very dependent on particle
size distribution of the product to be dried, conventional commercial
equipment is much larger than the unit recommended. A third trailer
will be required if even a slightly larger dryer than suggested by the
EES staff proves necessary.
« Possible problems with the blending of the pyrolytic oil with the char
have not yet been well established. It would be essential to assure
that no possible fire results upon mixing and thus adequate; cooling
of the low thermal conductivity char would be required. The chars
produced from varying waste sources might each demand different blend-
ing techniques to assure oil incorporation and lack of balling or
smearing of the mixed product. Analysis of this problem leads to the
conclusion that perhaps blending should only be utilized in those
rare cases where no market outlet exists for the liquid fraction.
This latter fraction has approximately the same heating value as the
char and net energy yield varies little with the ratio of liquid to
solid. Economically, however, the oil is valued at several times that
of the solid on a unit heating value basis, and hence its isolation
and separate sale should be considered.
PROCESS DESCRIPTION
Overview
The basic processing scheme utilized in the ESS waste-to-energy system
is similar to that used in the PUROX and TORRAX reactors. Cellulosic waste
continuously passes downward through a vertical shaft converter, where it is
first dried, then thermally degraded in a pyrolysis zone heated by upward
100
-------
moving gases leaving the bottom combustion zone. In this latter zone, air is
used to oxidize a portion of the char remaining after the waste has been
decomposed to gaseous and liquid products. The gas, being diluted with nitro-
gen from the air, has a rather low volumetric heating value, but its energy
is fully utilized by (1) burning it within an internal combustion engine that
in turn powers an electrical generator and (2) within an air-gas combustor
whose hot exhaust gases are used as the heating fluid for partially drying
the raw waste. The solid char and liquid products have very good heating"
values and may be used as fuels in systems normally using coal or fuel oil.
The oil may be incorporated into the char whenever a single fuel product is
desired.
The overall process thus converts a low density, low heating value, waste
into valuable fuel that can be economically transported to energy consumers.
An important concept of the EES system is the ability of the processing plant
to move to the many sources of waste, rather than vice versa, with probable
significant advantages in transportation costs in many farming areas.
Detailed operational characteristics of the processes are best examined
through review of the two principal experimental units used at the EES. This
equipment is described and the research data presented in the final reports
for the EPA-supported investigations under Contract 68-02-1485 and Grant
R803430-01-0 and in the publications cited as References 10, 11, and 12.
R § D Facilities--
A simplified flow diagram for both units is shown in Figure 12. The out-
side dimensions of the 5.2 Mg/d (6 TPD) unit are 3 m (10 ft) tall by 1.2 m
(4 ft) on each side, with the inside dimensions being 1.2 m (4 ft) deep by
0.6 x 0.6 m (2 x 2 ft). The 22.7 Mg/d (25 TPD) pyrolyzer is 5.5 m (18 ft) tall
and 1.8 m (6 ft) on each side, with inside dimensions of 2.4 m (8 ft) in depth
with a 1.2 m (4 ft) diameter cylinder.
Feed (moderately pulverized) enters the top of the converter through a
valve and falls onto the top of the bed. Towards the lower portion of this
bed are water-cooled tubes for introduction of combustion air into the reactor.
Only the amount of oxidation of a portion of the char is permitted to occur
that will supply sufficient hot gases for the decomposition of the feed mate-
rial in the intermediate section of the bed. Operating pressure is maintained
at several inches water column below atmospheric in the large unit and several
inches above in the small one. As the 427 to 704°C (800 to 1300 F) gases pass
over the waste material, decomposition and rearrangement reactions occur,
yielding gaseous, liquid, and solid (char) products. The latter pass out
through a mechanical output system and the char is then further transported by
means of a screw conveyor.
The warm gas and aerosols pass upward through the downward moving bed
into mechanical separators and then a liquid fraction is isolated in an air-
cooled condenser. Temperature is adjusted so that water is not permitted to
condense The off gas is combusted with air and vented during R & D tests,
but would be utilized to dry incoming feed and to generate required process
power in the contemplated large mobile system.
101
-------
o
to
WET ^
WASTE ^
DRYER
HOT WAT
rni n WAT
t
w
rn 4,
AIR
T
r
COOLING FAN
PYROLYSIS
CONVERTER
i
HOT
GAS ^
1
1— ^ CHARCOAL
CONDENSER
OIL
fc
AIR
^
r
BURNER
^
HOT GAS TO
ATMOSPHERE
Figure 12. Process flow diagram of EES research unit.
-------
Various bed stirring and agitation devices have been studied In the
latter phases of testing, an L-shaped water-cooled stirrer that also intro-
duced the combustion air was employed. This device has been named the
"AIRGITATOR" by the EES development group. Air delivery by it is through
1.6 mm (0.062 in.) holes spaced 12.7 mm (0.5 in.) apart.
Research Results
Because of the difficulty of maintaining constant conditions within the
pilot plant units and establishing accurate mass balance information, labora-
tory tests have been made to better determine the general trend of product
yields. This work, described in Reference 10, was conducted with sawdust
samples on the order of 2.5 kg (5.5 Ib) that had been dried to a 6 percent
water level. The HHV was 18.85 MJ/kg (8103 Btu/lb). Samples were contained
within a stainless steel tube leading to collection traps and were heated by
an external furnace maintained at temperatures between 540 and 870°C (1004 and
1598°F). Because the time—temperature history of initial solid particles and
decomposition products is different than occurs in a large vertical shaft con-
verter, and because of the absence of air, results cannot be directly used for
design purposes. Excellent guidelines for predicting probable results in pro-
duction equipment are obtained from this work, however.
In Figure 13 the yields of gas, char, water, and oil are shown as a func-
tion of temperature, with the upper total product line indicating good experi-
mental recovery was achieved. The trends are rather typical of results
obtained by other investigators.
Changes in composition of the gas fraction as a function of the tempera-
ture to which the sample was subjected are shown in Figure 14. Calculations
of heating values from these curves show a linear increase from 14.37 MJ/Nnr
(365 Btu/SCF) at 540°C (1004°F) to 16.54 MJ/Nm3 (420 Btu/SCF) at 760°C (1400°F)
and then a slight decrease at the highest furnace temperature.
Heat of combustion of the oil and char was measured in the laboratory.
For the liquid fraction, results vary (water-free basis) from 27,91 to 32.56
MJ/kg (12,000 to 14,000 Btu/lb). The solid had a quite constant heating value,
averaging 33.18 MJ/kg (14,267 Btu/lb). From these data, the heating values of
the individual fractions of the products as a function of temperature have
been plotted in Figure 15. The important fact to be noted here is that the
sum essentially equals the original energy content of the dry waste, demon-
strating that little loss need occur in such a pyrolysis system. In the
practical case, the heat represented by the electric furnace must be supplied
by a portion of the waste, but with properly designed equipment, overall
energy conversion efficiencies can be in the range of 60 to 70-6.
Test results in the pilot units with wood feed are shown in Table 15 and
with peanut hull feed in Table 16. From the results of the experimental work,
it was concluded that:
* The effects of the air/feed ratio on product energy yields appear to
be dominant; changing size and feed material, and the effects of
103
-------
100
95
30
cc
a
u. 25
o
o
cc
£ 20
o
15
_ A „— TOTAL
.1
I
GASES
CHAR
WATER
OIL
I
540°C 650°C 760°C 870°C
FURNACE TEMPERATURE
Figure 13. Pyrolytic product yields from laboratory tests
104
-------
40
35
30
I 25
-J
o
oa
o 20
15
10
C--C4 HYDROCARBONS
I
540°C 650°C 760°C 870°C
FURNACE TEMPERATURE
Figure 14. Gas fraction composition as a function of
temperature.
105
-------
9,000
8,000
y
a
UJ
UJ
£ 5,000
0.
DC
5 4,000
u
a 3,000
*j
03
2,000
1,000
TOTAL
~~cr ~
CHAR
GAS
I
I
I
540°C 650°C 760°C 870°C
FURNACE TEMPERATURE
Figure 15. Heating values of product fractions as a
function of temperature.
106
-------
TABLE 15. SUMMARY OF TEST RESULTS WITH WOOD FEED
Run
No.
4
5
9
10
17
18
19
6
7
8
12
13
15
16
Air/
Feed
(Wgt. ratio)
0.474
0.531
0.704
0.638
0.382
0.580
0.416
0.272
0.558
0.349
0.453
0.378
0.220
0.492
Feed
Rate
kg/h Ib/hr
132 291
112 246
83 184
$3 184
88 195
86 189
84 186
223 491
92 202
102 226
107 237
53 118
157 346
101 222
Char*
Yield
kg
8.5
9.4
5.3
4.7
12.7
4.6
12.6
11.7
7.3
10.2
11.9
13.1
14.7
7.9
Ib
18.7
20.7
11.6
10.4
28.1
10.2
27.8
25.8
16.1
22.5
26.3
28.9
32.4
17.4
Oil*
Yield
kg
2.9
3.0
6.5
5.1
3.7
5.1
6.5
3.5
4.8
5.4
2.7
6.8
5.5
6.2
Ib
6.3
6.7
14.4
11.3
8.1
11.3
14.3
7.8
10.7
12.0
5.9
15.0
12.1
13.8
Lost*
Carbon
kg
7.7
4.8
3.3
6.2
5.8
5.6
0.4
3.4
0.9
3.8
1.9
0.8
2.7
3.5
Ib
17.0
10.6
7.2
13.6
12.9
12.3
0.8
7.4
2.1
8.4
4.3
1.7
5.9
7.8
Mass
Output/
Input
1.02
1.02
0.94
0.98
0.98
0.99
0.91
1.03
0.95
0.98
1.05
0.96
0.97
0.98
Energy
Output/
Input
0.947
0.918
0.995
0.968
0.936
0.946
0.923
0.966
0.939
0.968
0.905
0.990
1.004
0.977
Available**
Energy
(Percent)
65.1
58.2
58.8
62.1
72.5
58.4
65.6
67.7
50.6
47.7
60.1
74.9
78.3
65.4
Comments
No agitation
Agitation
No agitation
Agitation
No agitation
No agitation
02 balance not
good. No agitation
Significant loss
of oil from the
condenser. No
agitation
Agitation
No agitation
02 balance not
good. No agitation
Oil condensing
caused erratic
flow. No agitation
No agitation
No agitation
* All results are presented on a basis of 100 Ib dry sawdust.
**r.nergy available in char/oil divided by total available from sawdust feed.
General notes: (a) Br-d depth of first two runs was 46 cm (18 in.); next five 69 cm (27 in.)-
(39 in.)
(h) All tests with 2 air tubes cxcejif first two with 4 and next throe with 3.
rema ind'jr at 99 cm
-------
TABLE 16. SUMMARY OF TEST RESULTS, PEANUT HULL FEED
o
00
Test
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Feed
Peanut Hulls
Peanut Hulls
Pin Sawdust
Pine Sawdust
Peanut Hulls
Peanut Hulls
Peanut Hulls
Peanut Hulls
Peanut Hulls
Peanut Hulls
Peanut Hulls
Peanut Hulls
Peanut Hulls
Feed
Rate
kg/h
571
390
676
464
494
481
476
408
501
•570
471
490
324
Ib/hr
1260
859
1490
1022
1090
1060
1050
900
1105
1257
1038
1080
715
Char
Yield
21.7
23.9
26.6
24.9
28.8
32.1
22.9
% Oil §
Aqueous
Yield
3.9
8.5
5.7
7.0
7.9
7.2
4.7
Air/Feed
(Wgt. ratio)
0.364
0.265
0.172
0.251
0.227
0.227
0.270
Off-Gas
Temp
°C
97
93
113
141
87
86
88
CHECK OUT "AIRG1TATOR"
40.0
24.9
27.0
28.4
16.1
4.53
23.4
17.8
0.458
0.464
0.539
0.613
79
88
87
83
CHECK OUT MODIFIED "AIRGITATOR
41.4
28.3
3.5
26.2
0.140
0. 190
174
227
°F
207
200
235
285
188
186
190
174
190
188
182
345
440
Bed
Depth
cm
132
132
132
132
132
132
132
89
89
89
89
127
127
in.
52
52
52
52
52
52
52
35
35
35
35
50
50
Agitation
No
No
No
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
No
No
No
Airgitation
No
No
No
No
No
No
No
Yes
No
No
No
No
Yes
Yes
Yes
-------
mechanical agitation, are of minor importance compared with the air/
feed ratio.
• The available energy in the char-oil mixture appears to be a single
function of the air/feed ratio.
• While the total energy in the char-oil mixture is a function only of
the air/feed ratio, the relative amounts of char and oil can be
changed significantly by varying the bed depth.
• The integrated mechanical agitation-air supply system, or "AIRGITATOR,"
appears to offer advantages of increased through-put, operating sta-
bility, and off-gas temperature at very low values of the air/feed
ratio.
• The overall mass, energy, and chemical balances appear to be reasonable
and satisfactory.
In that a gaseous fuel has little application as a product in the mobile
concept, and because gas yields increase with temperature, relatively low
pyrolytic temperatures should be selected for maximum energy yields, con-
sistent with suitable rapid reaction rates. Selection of the proper air/feed
ratio, the controlling factor for temperature, therefore becomes the most
critical design factor for practical hardware.
The EES method of selection of the proper air/feed ratio operating point
is a graphical one as represented in Figure 16, the bases of which are the ex-
perimentally observed compositional values from the pilot units. The curve in
the upper right quadrant represents heat consumed in the process, consisting
of 3.49 MJ/kg (1,500 Btu/lb) to evaporate water from the original feed stock
and 0.84 MJ/kg (360 Btu/lb) to process the dry feed. In the example illus-
trated, it is assumed that an agricultural waste containing 50% water (a rather
typical value) is fed to the dryer. To supply the required 4.33 MJ/kg (1,860
Btu/lb) of dry feed, an air/feed ratio of 0.47 would yield an off-gas having
this quantity of heat. The remaining char and oil would have a summed heating
value of 12.79 MJ/kg (5,500 Btu/lb) of dry feed, for a process thermal effi-
ciency of about 65 percent. For wastes having different original moisture
contents, other air/feed ratios would be utilized, it being apparent that there
are significant advantages in starting with a low water content waste.
The 50% water content control point has been used in all design analysis
by Parsons.
MOBILE PYROLYSIS SYSTEM CONCEPT
Sufficient testing was accomplished with the pilot units to ascertain the
nature of the important process variables and the probable net energy yields
when agricultural wastes are used as a feedstock. The EES process developers
then proceeded to develop a preliminary design for a transportable facility
capable of producing pyrolytic char and oil fuel from stockpiles of waste
109
-------
0
UJ
CO
(5
O
CC
u.
UJ
CD
Q -
UJ
CC
O
UJ _
cc
I-
Ul
I
9,000
8,000
7,000
6,000
5,000
4,000
3,000
2,000
1,000
1.0 0.8
0.6
0.4
0.2
AIR/FEED (DRY) Lb/Lb
20 40 60 80 100
FEED PERCENT MOISTURE
1,000
2,000
3,000
4,000
5,000
6,000
ASSUMPTIONS:
1) GROSS HEAT ENERGY
REQUIRED TO PROCESS
ONE POUND DRY FEED =
360 Btu
2) 1500 Btu REQUIRED
IN DRYER TO EVAPORATE
ONE POUND OF WATER
Figure 16. Selection of air/feed ratio.
110
-------
materials located throughout a fairly large region. Design criteria they
selected were:
• No requirements for external utilities.
• Full compliance with all applicable pollution control and OSHA
regulations.
• Processing capacity of 200 tons of wet fuel (50% moisture) per 24-hour
day.
• Capability of being carried on two trailers having maximum dimensions
of 55 feet in length, 8 feet in width, 13-1/2 feet in height, and
weighing no more than 73,000 Ib each.
• Insulation of the converter could be accomplished by means of small
inside shelves to contain char produced within the reactor (and self-
regenerated as required).
DESIGN REVIEW OF CONCEPTUAL MOBILE SYSTEM
Figures 17 and 18 present the basic process flow diagram of the mobile
pyrolysis system suggested by EES, in one possible configuration they have
proposed for the two-trailer system. The equipment list numbering system is
that of EES and has been adopted in this section to avoid any confusion between
the original work and the Parsons design review. That original listing is as
follows:
1. Front end loader
2. Bin conveyor
3. Receiving bin
4. Conveyor mill
5. Hammer mill
6. Drier
7. Feed conveyor
8. Converter
9. Cyclone
10. Condenser
11. Condenser cooling
fan
12. Draft fan
13. Combustion air
fan
14. Off-gas burner
15. Drier fans
16. Burner exhaust
17. Drier exhaust
duct
18. Cyclone
19. Process air
blower
20. Generator
21. Engine
22. Cooling water
radiator
23. .Compressor
24. Conveyor
25. Char oil mixer
26. Char storage
bin
27. Control room
28. Agitator
29. Front end
loader storage
30. Cat walk
31. Engine blower
111
-------
-GRIZZLY BARS
AGRICULTURAL &
FORESTRY WASTES FROM: 200 TPD
COTTON GINS (iNPUT (8.3 TPH| WASTE
PEANUT PROCESSING / * APPROX. 50%
SAWMILLS I MOISTURE CONTENT
SUGAR MILLS
\AAAA
OFF-GAS
FROM TRAILER II
LPG FOR ENGINE STARTUP
(APPROX. 4 MRS I
TO TRAILER II
Figure 17. Process flow diagram trailer No. I.
-------
FLEXIBLE DUCT
FROM TRAILER I
OFF-GAS @ 66°C (150°F|
BURNER
91 Mg/rf (3.8 Mg/h)
100 TPO 14.2 TPH)
WASTE FROM TRAILER I
(2?) AIRGITATOH
FI^(WATER-COOLED)^
©ROTARY TINES
IWATER-COOLEDI
(«B)STftRTUP BURNERS (3)
4B TPD AT
0,64 a
I40LB/FT'!
FINISHED PRODUCT
60% CHAfl/40» OIL
DISCHARGE
TO TRANSPORT
TBUCK/TRAHER
Figure 18. Process flow diagram of trailer No. II.
-------
At the current state of development and from any evidence submitted to
Parsons for review, a high probability exists that the dryer recommended by
EES is insufficient in volume to accomplish the desired drying down to 5%
moisture. If the best commercially available agricultural dryer must be used
in its place, a third processing trailer would have to be used in order to
transport all the equipment to fully process 181 Mg/d (200 TPD).
Each sub-section that follows discusses the functions of the equipment
and presents comments from vendors of such equipment and from specialists
within Parsons.
Item 1 - Front End Loader
In order to have an item of equipment available to assist in the assembly
and disassembly of the process system at each new location, a fork lift truck
is recommended as a substitute for the front end loader. It would be equipped
with a scoop attached to handle the bulk wastes, either manually or hydrauli-
cally controlled. The 5000-lb capacity Hyster Model 50 with 14 x 17.5 tires
would be suitable. Its basic cost is $16,000 ($2,000 extra for hydraulic
scoop control) and the unit would occupy a space of 1.68 x 3.05 m (66 x 120
in.) on the trailer. Manually or hydraulically operated runway rails for on/
off loading would be required.
Item 2 - Bin Conveyor
It is recommended that this item be eliminated and the waste be placed
directly in the bin.
Item 5 - Receiving Bin
The estimated weight of the bin is 2.04 to 2.27 Mg (4500-5000 Ib) and is
one item for which the fork lift would be a necessity (lifting brackets or
lugs would be 2.1 to 2.4 m (7 to 8 ft) off ground level when stowed on the
trailer). A grizzly (coarse bar screen) should be incorporated on the top to
assure that large foreign objects do not enter the hammer mill. Three vendors
submitted prices between $1656 and $2500.
Item 4 - Conveyor to Mill
The conveyor presents no special problems. Its speed should preferably
be automatically controlled to prevent overloading the mill and an overhead
magnet should be used to remove tramp metal that could damage the mill. The
Valley Industrial Supply Co. inclined belt conveyor at $1500 is typical of
those that could be used. It is 30 cm (12 in.) wide and 6.1 m (20 ft) long
and is equipped with a 10 cm (4 in.) diameter tail pulley and a 20 cm (8 in.)
diameter rubber lagged head pulley. The belt is 3-ply Neoprene, reinforced
with 12 gauge galvanized steel bead, and is driven by a 560 W (3/4 H.P.) motor.
Other prices of from $1628 to $4000 have been obtained.
114
-------
Item 5 - Hammer Mill
The Gruendler model 30-2 56 kW (75 HP) mill is recommended over the
Williams C-32 suggested by the developers. It has a more compact design and a
lower weight. Current price is $12,750. Vibration of the mill could create
some problems with items mounted on Trailer No. I; this should be investigated
before a final production design is completed. A rotary feeder should be
incorporated between the mill and the dryer to prevent blow-by.
Item 6 - Dryer
The dryer is one of the major components in the pyrolysis system that
requires further investigation and testing at the full scale component level.
The dryer depicted in the EES concept sketches is a modification of an experi-
mental dryer developed by the Tech-Air Corporation of Atlanta. A unit of this
type was in operation at Cordele, Georgia, drying hogged wood waste at the time
of preparation of this report, but detailed performance data could not be
released to Parsons. Such data should be sought for release prior to any final
design of the mobile system. None of the drying equipment manufacturers con-
tacted are able to supply a unit of the type depicted in the EES concept
sketches or that will meet the necessary space constraints. Based on their
dryer performance experience, several manufacturers have considerable doubt
that a dryer of this design, type, and dimensions could meet the required
performance. Most manufacturers recommended a rotary drum type; these are
well over 2.4 m (8 ft) in diameter to adequately dry typical agricultural
materials. One organization proposed a fluidized bed type that was quite tall
and would have to be folded down for transport. Another company has a com-
bined hammer mill and vertical dryer, but it is over 7.6 m (25 ft) tall.
Generally, it is the belief of commercial dryer manufacturers and of
Parsons that the EES concept dryer would not provide sufficient intimate con-
tact between the moist feed material and the hot drying air. A tumbling action
is needed to break up clumps of material and expose the surfaces of the par-
ticles to the drying air. This is necessary because of the thermal insulating
properties of the feed material and the inherently slow diffusion of the phys-
ically bound water to the surface of the particle. Without sufficient aer-
ation, the drying time is increased, resulting in the larger sized unit.
The EES dryer concept consists of a 91-cm (36 in.) diameter^helical screw
having some paddle arms to produce a mixing and tumbling action in the mate-
rial. The screw moves the moist feed material horizontally along a perforated
trough in the same manner as a screw conveyor functions. Hot drying air under
a slight pressure is forced up through the perforations from the plenum
located below the trough. This action is similar to the action in a fluid bed.
Reportedly, the dryer removes moisture in the feed material from an initial 50
percent to 5 percent total moisture content using hot gas having an inlet tem-
perature of approximately 371°C (700°F), and 430 mVmin. (16 000 CFM) gas flow.
The dryer should be constructed of stainless steel to preclude oxidation of
the metal in the hot moist environment.
The dryer unit as conceived should have counter-flow drying air; however,
the blower as shown in the concept drawings is located at the feed inlet end.
115
-------
Ducting would have to be run to the feed discharge end, consuming more of the
limited space. Further layout and/or rearrangement will be required. The
dryer would require rotary feed valves on both the feed inlet and perhaps the
outlet to prevent escape of the drying gases.
The EES dryer has a scaled width of 1.2 m (4 ft), leaving insufficient
space to park the front end loader, or fork lift truck if used instead, beside
the dryer and still stay within the 2.4m (8 ft) maximum legal width for high-
way vehicles. One or the other of the pieces of equipment will have to be
narrowed. The off-gas burner (Item 14) is considerably larger than that
depicted and will force a rearrangement of equipment on Trailer I. If a spe-
cial permit 3.0 m (10 ft) wide trailer is acceptable, then there will be no
problem provided the dryer is less than 1.5m (5 ft) wide.
If the trailer width is limited to 2.4 m (8 ft), and this is desirable
for use on rural roads, then a third trailer will be required even if testing
verifies the basic concept. The conservative assumption is therefore made
that the three trailer system will be required. The burner and dryer would
be on one trailer and the other trailers would be rearranged to transport the
balance of the process equipment.
Calculations by Parsons indicate a minimum of 15 kW (20 hp) is required
for the screw drive rather than the 746 W (1 hp) shown on the EES concept
drawings. Proposals from dryer manufacturers show total power requirements,
including fans, up to 186 kW (250 hp), which is considerably greater than the
31 kW (41 hp) provided in the EES concept schedule of loads.
The price range for commercial equipment has been established at $27,000
to $180,000. Full scale testing of a final dryer system must be accomplished
to verify performance and feasibility before the trailer conceptual design can
be finalized into a production model. This could be conducted by either manu-
facturers or in a final design program.
Item 7 - Feed Conveyor
This conveyor should be enclosed to prevent wind scattering the dried
waste and should have lifting lugs to permit the fork lift truck to position
it. The belt should have cleats on it to prevent material slippage on the
steep incline.
Item 8 - Converter
The pyrolytic converter is recommended to be cylindrical in cross section
to assure uniform stirring and reaction in both the combustion and decomposi-
tion zones. A sketch of the 2.4m (8 ft) diameter unit is shown as Figure 19.
It should be constructed of Type 304L stainless steel. Until moderate dura-
tion tests of the char-in-shelves concept of insulation proves this system
would indeed work, a design based on the use of a light weight, abrasion
resistant, insulating brick lining must be recommended by Parsons. External
insulation for personnel protection should be contained in an aluminum,
weather-proof jacket.
116
-------
6-IN. FLANGED OPENING
FOR AIRGITATOR
2-6-IN. FLANGED
1-IN. INSULATED AND WATER
TIGHT METAL JACKET
1/i-l N.PLATE
18-IN. MANHOLE
4%- IN. FIREBRICK
1-IN.SUPER-X
INSULATING BOARD
6x6x3/8-IN. (TYPFOR4)
MIXER CONVEYOR
2 FT-6-IN.
Figure 19. Converter section.
117
-------
Stirring and air injection can be accomplished with the "AIRGITATOR" con-
cept; it should be fabricated of stainless steel and water-cooled. Height
restrictions require that the drive motor be demountable or else attached to
the side of the converter and rotation transmitted through a series of geared
shafts.
Char product output is controlled by the speed of rotation of water-
cooled tined drums dropping material onto two screw conveyors. The ducting
about the drums and conveyors might require water cooling to reduce char tem-
perature and should be air-tight to assure ignition of the char does not
occur. An inclined conveyor will be required to move the cooled solid to the
oil blender.
Start-up of the cold converter could perhaps best be accomplished by
bracket-mounted retractable LPG burners such as the Sur-lite Model 2-H118-Ut-
30, priced at $1200. The burners would be of stainless steel construction
with the burner heads housed inside 7.6 cm (3 in.) pipes serving as a protec-
tion against the bed material. Several burners would assure uniform start-up.
If a portable (wand) burner proves adequate and can be operated safely,
the Sur-lite Model PHIT-CH2 is recommended] its current cost is $175.
Item 9 - Cyclone (Off-Gas)
Standard cyclones are typically of too great a height to be used for this
function of removing particulate matter from the converter off-gases. Custom
units of Type 304L stainless steel should be used. A gravity operated tipping
valve and a dust storage bin would be installed under the cyclone(s). Depend-
ing on practical problems that will be studied with the prototype unit, cost
could be in the range of $5000 to $8000.
Item 10 - Condenser
The oil condenser should be fabricated of Type 304L stainless steel.
Tests should be conducted to establish possible fouling and corrosion problems.
Items 11,. 12, 15, 15, 19 and 31 - Fans and Blowers
No particular problems exist with the various fans and blowers. Detailed
sizes and power will have to be specified after final flow and pressure drop
designs can be established. Because of potential corrosion problems, all fans
contacting the off-gas should be constructed of stainless steel. It is recom-
mended that the two dryer fans (item 15) be placed in the exhaust ducts so that
they will handle the cooled air having a lesser volume.
Item 14 - Off-Gas Burner
The Coen Burner Manufacturing Company, an organization having an excellent
reputation for supplying combustion equipment of the type required here, indi-
cates that an envelope size of 4.3 x 1.8 x 2.4 m (14 x 6 x 8 ft) would be
required to supply the 11.6 GJ/h (11 x 106 Btu/hr) needed. The 1371°C (2500°F)
gases would be diluted with air to achieve the desired 371°C (700°F) dryer
118
-------
inlet gas. Their estimated price for the burner, piping train, burner safety
and control system, fans and refractory-lined chamber is $50,000.
Item 16 - Burner Exhaust
This would consist of a vertical steel stack to carry off any surplus hot
gases and to also act as a flare stack. Its top would be 4.1 m (13 ft, 6 in.)
above ground level. Drying air controls would regulate a by-pass damper to
divert excess gases to the stack.
Item 17 - Dryer Exhaust Duct
Ducting must be incorporated between the burner (14) and the dryer (6),
the dryer and fans (15), and between the fans and the cyclone (18). All duct-
ing would have insulation with a weather-proof metal jacket to retain heat and
provide personnel protection.
Item 18 - Cyclone (For Dryer)
The standard cyclone will have to be modified to have an eccentric hopper
with its discharge located to one side of the trailer. The hopper discharge
will have a rotating feeder or gravity dump valve. No other problems have
been encountered or are expected.
Item 20 - Generator
The capacity of the generator will have to be increased from the 150 kW
indicated by EES to 200 kW. This increase is caused by larger motor loads for
process equipment, lighting on trailers, and movable flood lights for night
operations. It also assumes that the Georgia Tech dryer will not be used and
that a commercial rotary one be substituted.
Item 21 - Engine
The specified Waukesh engine is no longer available; the replacement unit
would be Model H2475GU. Some engine manufacturers indicate they might have
difficulty operating on such low heating value off-gas, and that the composi-
tion of the gas is critical to reliable performance. Installation of a LPG
storage tank is recommended which would have a capacity to supply engine and
converter under start-up period and when off-gas is too weak, or not produced
for any reason. Switchover to LPG would be automatic, otherwise the operation
would shut down. The engine and generator should be protected from weather in
a sheet metal hood or enclosure.
The present design will be based on a Kohler generator set, for data
and price. The Kohler price is alsOo lower than Waukesha for one unit and
quantity discount is a minimum of 25%.
It is recommended that the engine chosen be considered one of the compo-
nents for the final development program; it might be operated at ££«*«'-
peratures than normally used to reduce or hopefully, eliminate corrosive
problems from the off-gas. An experimental program for an engine for the EES
119
-------
concept has been conducted by the Georgia Institute of Technology for EPA.
No report has been issued as of May 1977, but it is understood that a 6-cylin-
der spark-ignition truck engine (7.5 compression ratio) performed well with a
dry simulated off-gas. The brake power output was approximately 0.6 of that
when the engine was fueled with gasoline. Tests with actual pyrolysis gas are
essential to developing final operating costs for the system.
Item 22 - Radiator, Converter Cooling Water
The radiator and cooling system would be sized to cool the AIRGITATOR,
converter outlet feed drums, lower hopper bottom, and the water jacketing on
the screw conveyors used for reducing the char temperature to prevent reigni-
tion and possible explosive condition developing when oil is added. The radi-
ator would be a standard commercially available unit with a motor driven fan.
Item 25 - Compressor
The air compressor and air receiver would be a commercially available unit
commonly used in service stations. The compressed air would be used for
instrument control air and powering actuators for controlling the process such
as dampers or valves.
Item 24 - Conveyor, Mixer to Storage Bin
This unit would necessarily be enclosed to prevent wind blowing away the
char. Another conveyor or some means will be required to uniformly distribute
the char-oil into the char-oil storage bin trailer (26).
Item 25 - Mixer
There is still some question as to how best to mix the oil and char prod-
uct. A Littleford Model KM 300D mixer/grinder, 8.5 m3/h (300 ft^/h) capacity,
would be one alternative; this unit costs approximately $20,000. A size 2424
Sprout-Waidron single-hammer attrition mill ($8,000) might also be considered.
With oil currently commanding a much higher sales price per unit heating value
then solid fuels, consideration should be given to eliminating this mixing
operation in most cases.
Item 26 - Product Storage Trailer
In that this item is considered a part of the marketing activity rather
than waste processing, it has not been discussed here, but rather is costed
within the total system operating economics section.
Item 27 - Process Instrumentation and Controls
The EES concept has not progressed to a level where the developers have
had to describe any instrumentation and controls as a part of the transportable
pyrolysis system, but in production units they would be a very necessary and
integral part of it.
120
-------
The plant is intended to pyrolyze agricultural wastes from saw mills
cotton gins, sugar mills, and peanut processing plants, all having varying
degrees of moisture depending on recent weather conditions. Manual (non-
automatic} operation under these conditions, especially with a minimal work
crew, would be unsatisfactory and potentially hazardous. Essential process
variables must be automatically held to practical working tolerances within
their operating ranges in order to assure total system operation with high
energy recovery.
Control and instrumentation are required to provide proper control over
the process, protect the plant from abnormal operating conditions that could
be damaging, and continuously monitor operations against hazardous conditions
developing, such as fire and explosion, for safety protection of personnel.
Parsons has attempted to identify the instruments and controls so that
their cost is reflected in the estimate. Subsequent operations and tests of
the final selected equipment configured for mounting on the trailers will be
needed to verify and check out the instrumentation and controls, and the pos-
sibility exists that the need for some of this equipment could prove to be
unnecessary.
The following control systems are recommended and their cost is included
in the cost estimate:
A dependable burner management system is required to prevent a hazardous
explosive condition. The burner flame must be monitored carefully at the
time of change-over of fuel from LPG to off-gas and switched over when the
off-gas Btu content is high enough to support combustion. The switching action
reverses when off-gas Btu content is too low or in insufficient quantity.
Hot gas at the burner outlet would be monitored and a valve on the bypass
to atmosphere will exhaust excessive gas when the temperature exceeds 370°C
(700°F).
Temperature in and out of the dryer will be monitored and alarmed when
normal operating temperature is not maintained.
Differential pressure across the dryer will control the vent to atmosphere
to maintain a constant dryer pressure.
Current to the hammer mill will be monitored and controls will adjust the
material feed rate so the current draw does not exceed 80% of full load current.
The hammer mill will have a current read out on the control panel.
Differential pressure will be maintained and read out across the converter
and cyclone and will be alarmed on a preset high differential.
Temperature within the converter will be read out at three levels. Tem-
peratures of the hot gas out and the char discharged will also be measured. A
level sensor system will monitor and control feed level.
121
-------
The temperature of radiator water in and out will be measured and will be
alarmed when a preset high temperature is reached. Thermostatic control valves
will regulate cooling water to the Airgitator, water cooled drums in the con-
verter, and the mixer conveyor.
The condenser temperature will be monitored in and out and will be alarmed
when the temperature is not in proper operating range.
The following instrumentation has been identified as the minimum required
for satisfactory operation of the pyrolysis plant:
Control Room
Instrument Read-out and Equipment Control Switch Panel
Control Switches for 24 Equipment Drive Motors
Control Switch for Lighting
Hammer Mill
Current (amperage) flow controls, feeder to hammer mill
Current flow sensor (R)*
Dryer
Temperature - Inlet Hot Air (R)
Temperature - Effluent Air (R)
Air Flow - Drying Air (R)
Pressure Drop of Drying Air (L)*
Converter
Height sensor for material to control flow of dried material into
converter on a go no-go basis (L)
Thermocouples at 3 height levels (R)
Combustion air flow rate to Airgitator (R)
Loss of air flow alarm (R)
Burner
Off-Gas to LPG automatic monitor £ switch-over
Ignition controls
Flame failure safety controls
Excess heat hot gas by-pass to atmosphere
Dryer Exhaust Cyclone
Exhaust gas pressure drop (L)
Converter Off-Gas Cyclone
Off-gas pressure drop (L)
Off-Gas Condenser
Temperature, off-gas from converter (R)
Temperature, leaving condenser (R)
Off-gas pressure drop through condenser (L)
(R) and (L) refer to remote readout and local indicator respectively.
122
-------
Cooling Water System
Temperature, cooling water supply (R)
Temperature, water leaving tines (Drums) (L)
Temperature, water Leaving Airgitator (L)
Temperature, water leaving char cooling conveyor (L)
Engine-Generator
Amperage (R £ L)
Voltage (R $ L)
RPM or Hertz (R § L)
Transfer switch or controls from LPG to Off-Gas (R $ L)
Off-Gas supply pressure (R § L)
Engine oil pressure (L)
Engine oil failure (R)
Detailed estimated costs, presented in summary form later are shown in
Table 17.
TABLE 17. CONTROL £ INSTRUMENTATION COST ESTIMATE
Panel, 2 m (6 ft) (Console) $7,000
Burner Management System *
Off-Gas Analyzer § Switch Over System $10,000
Dryer Air Temperature Loop-Bypass to Atmosphere $7,500
Temperature Loop Across Dryer (3 Points) $3,300
Differential Pressure Across Dryer $1,600
Hammer Mill Current Monitor § Feed Control System $600
Differential Pressure Across Converter $1,600
Temperature Within Converter (3 Points) $3,300
Temperature Within Condenser (3 Points) $3,300
Material Feed Level in Converter (3 Points) $3,800
Cooling Water Radiator Temperature In § Out $2,200
Control Panel for Engine *
Instrumentation $44,200
Installation Labor $30,000
TOTAL COST INSTALLED $74,200
* Cost included with major equipment.
Item 28 - Agitator
A preliminary design of an agitator (AIRGITATOR) has been made to be in
accordance with the Georgia Tech requirements. The unit would be Constructed
of Type 304L stainless steel and be water cooled using water from the cooling
water system and air cooled radiator (22).
123
-------
Item 29 - Front End Loader Storage
See Item 1.
Item 50 - Cat Walks
The extent of cat walks, ladders, and stairs required for access to the
plant during operation have been estimated along with their cost. Complete
peripheral coverage as suggested by the developer does not appear warranted.
The cat walks would be light weight aluminum framing and aluminum grating.
The cat walks can be designed to latch in the raised position and fold down
for the operating position. Removable posts and chains would protect personnel
from falling.
Items 52 and 55 - Trailers I and II
Price and dimensional data have been received from Fruehauf. The 40-ton
Model C40D-J2 trailer would be required in lieu of the suggested Model C25D-J2
because of plant weight. Fruehauf has advised that the trailer could be
extended 6 feet, from 16 feet to 22 feet, in its mid-section, should addi-
tional length be required. During final design of the plant, the loading and
placement of equipment will have to be carefully considered for weight, bal-
ance, and dynamic stability during transit.
Item 54 - Conveyor, Converter to Mixer
The process developers did not identify a conveyor from the converter
discharge to the mixer (25). One is required to horizontally move the char
and then elevate it to the top of the mixer. The exact configuration and
equipment required downstream of the converter is still not defined and the
exact type of char/oil mixer (25) to be used has not been determined.
Item 55 - LPG System
As discussed under items (6) and (21), a LPG system will be required for
start-up and during times when the off-gas is insufficient to support opera-
tional demands. The system would have a capacity of 1100 dm3 (300 gallons),
sufficient for approximately 8 hours of operation. The system would be
refilled after each second start-up. The engine and burner would have to be
equipped with a dual set of burners or an LPG/air dilution-mixer to lower the
heating value of the LPG (88.61 MJ/m3 or 2250 Btu/ft3) to that of the off-gas
(5.91 MJ/m3 or 150 Btu/ft3).
Item 56 - Rotary Feeder
The Dryer (6), the inlet to the Converter (8), and the cyclone discharges
will require air locks to prevent flow of air or off-gas. Rotary feed valves
driven by gear head motors through a roller chain drive would be used.
124
-------
Power Requirements
The transportable pyrolysis system as conceived must be completely inde-
pendent of outside utilities, except for periodic replenishment of LPG fuel
used for start-up. As such, the system must be self-supporting in operation
and supply its own power needs. Parsons has investigated the individual power
requirements of the various pieces of equipment in order to verify the size of
the off-gas engine and the generator that it drives. It is concluded that the
electrical load will be approximately 40 percent greater than that indicated
by EES (120 kW) or 168.5 kW.
Table 18 shows a schedule of process equipment power requirement. The
tabulated data are as close as can be identified at this preliminary stage and
would be subject to modification in the final design phase of development.
TABLE 18. SCHEDULE OF PROCESS EQUIPMENT POWER REQUIREMENTS
Item No.
2
'4
5
6
6A, 6B
7
8
28
8A
11
12
13
15
19
22
22A
23
24A
24
25
25A
31
Equipment Description
Bin Conveyor (Deleted)
Hammer Mill Feeder
Hammer Mill
Dryer Drive
Dryer Rotary Valve Drives, 2 @ O.S hp
Conveyor - Dryer to Converter
Converter Input Rotary Valve Drive
Airgitator Drive
Converter - Tine drive
Condenser Fan
Off -Gas Fan - O.G. to Burner
Combustion Air Fan to Burner
Dry Air from Dryer Fans, 2 @ 30 hp
Process Air Blower - Roots Type
Cooling Water Radiator Fan
Cooling Water Circulating Pump
Instrument Air Compressor
Conveyor - Converter to Grinder-Mixer
Conveyor - Char/Oil from G-M to Storage
Char/Oil Mixer
Oil Spray Feed Pump
Engine Blower - Off Gases to Engine
Process Equipment Power Sub Total
Plant Lighting
Work Area Portable Lighting
Instrumentation and Controls
Total Power
Use Rated Power
Power Required
kW hp
-0- -0-
1.5 2.0
56.0 75.0
7.6 10.0 (1)
0.7 1.0
3.7 5.0
0.4 0.5
2.2 3.0
3.7 5.0
1.1 1.5
4.1 5.5
11.2 15.0 (2)
44.8 60.0
3.7 5.0
3.7 5.0
1.5 2.0
3.7 5.0
2.2 3.0
3.7 5.0
2.2 3.0
0.7 1.0
1.5 2.0
160.0 214.5
2.5
4f\
. 0
2.0
168.5
i nn n
zUU - U
Notes; (1) Not well identified; commercially available dryers range from
(2) NoVdeterminedfdipends on burner combustion chamber pressure.
125
-------
The increased power load results principally from the dryer and its fans
(Items 6, 6A, 6B, 15), combustion air fan (Item 13), plant and work area light-
ing, and instrumentation and controls. No statistical diversity factor can be
established at this time, and to avoid overloading the engine-generator at
plant start-up, the various pieces of equipment should be started sequentially.
The generator capacity should be oversized by approximately 20% to 200kW
to give a reasonable safety margin. Increasing the generator rated capacity
will also increase the horsepower of the off-gas engine, its size, and use of
off-gas. A portion of the final developmental study should be devoted to
optimizing relative energy distribution of the three product streams so that
net energy is maximized. This would be accomplished by variations in the
feed/air ratio.
SYSTEM ECONOMICS
Equipment Cost Estimate
Standard industrial estimating sheets were prepared by engineering con-
struction specialists at Parsons to establish the probable costs of the mobile
pyrolysis system. A summary of the principal elements of cost is presented
below, for both the first prototype unit and for production of a lot of 100.
Several important assumptions were made in deriving the cost estimate.
Most of these assumptions lead towards an estimate that could be at least
$100,000 low if additional developmental work demonstrates them to be false.
Experience demonstrates that estimates prepared prior to completion of final
R S D can be considerably below ultimate costs because of extra features found
necessary and generally increased complexity of the total system. None of the
developmental costs are included within the cost estimate, i.e., a strictly
industrial organization would have to attribute higher costs to each unit to
recover these past expenditures.
The assumptions are as follows:
• A commercial rotary dryer will be required to meet performance require-
ments and a third trailer is required.
• The basic pyrolytic converter as now developed will meet performance
requirements. Stainless steel construction has been priced.
• The internal combustion engine will function reliably with the cor-
rosive low-Btu gas.
• The burner can be packaged to fit the trailer and operate properly
with the off-gas.
• A simple hopper-screw conveyor mixer for the char-oil blending will
suffice.
• An 8-hour LPG storage system is sufficiently large.
126
-------
• The cost of a fuel storage bin (Item 26) and bin conveyor (Item 2) are
not included.
• The controls and instrumentation specified are adequate.
• A 10% procurement cost has been added to all vendor quotations.
• A 15% contingency of total direct costs has been employed to cover
items where design uncertainty exists or where items might have been
inadvertently omitted.
• No escalation has been used over early 1976 prices.
• No sales tax is included.
The cost estimate summary is as follows:
Item
Major Equipment
Loader
Receiving Bin
Conveyors (4)
Hammer Mill
Dryer, with Fans
Converter and Accessories
Cyclones (2)
Condenser
Gas Burner
Process Air Blower
Engine-Generator
Water Radiator, with Accessories
Char-Oil Mixer
Control Room
Engine Blower
Electrical System
Instrumentation and Controls
Trailers (3) Including Catwalks
LPG System and Controls
Painting
Other Equipment
Material Sub-Total
Labor Sub-Total
Sub-Total
Original
(Prototype)
$18,100
3,100
9,100
15,600
115,000
72,500
15,000
8,000
57,400
3,500
41,000
6,
3:
2.
,300
,400
,500
2,000
19,000
74,200
46,000
5,000
3,000
14,200
533,900
51,000
584,900
Production
(100 or more)
453,815 *
38,000
491,815
mix of Quantity discounts, on a w<
iiic «oa.j.uuj miJt U-L 4uo.ii i, -L i./
mately 15% and hence a multiplier of 0.85 was
127
-------
Original Production
Item (Prototype) (100 or more)
Direct Material $30,000 $24,000
Labor 20,000 14,000
Direct Cost Total 634,900 529,815
Engineering 40,000 1,000
Total Direct Cost 674,900 530,815
Contingency § 15% 100,000 80,000
Freight Allowance 4,000 4,000
Grand Total $778,900 $614,815
Operating and System Costs
As with any other waste-to-energy processing system, the analysis of the
total system economics for the mobile agricultural waste pyrolyzer can be
performed with meaningful accuracy only by using actual information obtained
for a specific region. Surveys must be conducted to establish waste quan-
tities, points of origin, and their seasonality. Letters of intent must be
obtained for the supplying of wastes and the purchase of the fuel product,
along with the various local cost factors. Key assumptions can be made in
advance, however, that will permit conclusions to be drawn as to the basic
economic feasibility of the system and that will indicate the sensitivity of
profit to various cost elements. Such an analysis, under several sets of
assumptions, is presented here.
A single value, $800,000, for the investment in the mobile system has
been used. While slightly higher than even the single unit cost derived above,
use of this cost is within the range of accuracy of the estimate and will
allow for a moderate degree of cost escalation. Based on a lO^year useful life
and 8 1/2% interest, the annualized equipment cost is therefore $121,936.
Other costs have been developed under tne specific assumptions listed in
Cases I through IV below. Two fundamental exceptions have been taken with the
preliminary economic analysis made by EES. The nature of the pyrolysis equip-
ment precludes operation on other than a continuous basis and only a 3-shift
day, 7 day per week case has been considered; the unit must be always manned
by at least two personnel while operating. Where EES based their analysis on
waste being delivered at no cost to the processing facility or at a disposal
charge of $3/ton, the Parsons calculations assume that both the no cost and
the drop charge cases might be optimistic assumptions and that a payment to
the waste generator of $5/ton should be considered as the conservative
approach. Such purchase of wastes has become common in the wood products
industry in recent years.
The first two cases analyzed have an operating cycle of 14 days of waste
conversion and then 2 days to move to the next site and set up again; any
128
-------
A stockpile of 2540 Mg (2800 nsof 50 e
for operation of the 200 TPD system. At each site approximate" In
S° "° bS r°
°f »•""
These two case assumptions are as follows:
CASE I CASE II
^actor jOptimistic) (Conservative)
Production days per year 294 238
Product per year, Mg (ton) 12 000 (13,230) 9 716 (10,710)
Sites per year 21 17
Total working days per year 335 272
Cost of waste, $/Mg ($/ton) 0 5.50 (5)
Selling price of product, $/Mg ($/ton) 38.58 (35)* 30.86 (28)*
Annual labor cost, $ 191,280 155,560
Annual maintenance cost, $ 20,000 40,000
Annual transportation cost, $ 33,440 27,680
Annual supply cost, $ 4,000 6,000
* $35/ton corresponds to $1.52/10 Btu and $28/ton to $1.22.
The costs above were derived as follows:
Labor - 2 persons continuously working (24 hrs/day) during the listed
working days per year at a burdened rate of $10/hour. In case I, a
supervisory cost of $30,000 was added and for Case II $25,000.
Maintenance - 4% of capital cost for Case II and one-half of that for the
optimistic case.
Transportation - Tractor with driver rental at $30 per hour for 3 hours
of each working day. To this has been added the annual i zed cost of two
product storage trailers at $8000 each with a 5 year useful life.
Supplies - Estimates for LPG and miscellaneous small parts.
Table 19 shows the summary of annual costs and revenues for the Cases I
and II. It can be seen that for the optimistic case a profit of more than
$90,000 per year could be realized, while a loss of some $290,000 could result
129
-------
TABLE 19. ECONOMIC ANALYSIS SUMMARY
COSTS
Capital Amortization
Waste
Labor
Maintenance
Transportation
Supplies
INCOME
21 x 630 x 35
17 x 630 x 28
PROFIT (LOSS)
PROFIT (LOSS) /Ton raw waste
PROFIT (LOSS)/Mg raw waste
Case I
$121,936
0
191,280
20,000
33,440
4,000
370,656
$463,050
92,394
$1.57
$1.73
Case II
$121,936
238,000
155,560
40,000
27,680
6,000
589,176
$299,880
(289,296)
($6.08)
($6.70)
if all of the several conservative assumptions of Case II indeed were to be
true. The numbers demonstrate the need to operate the equipment for the maxi-
mum possible number of days per year on a continuous basis and the high sen-
sitivity to product sales price and value imputed to the waste.
With Cases I and II establishing the importance of principal key variables
in permitting system profitability, several variations in Case I assumptions
have been made to determine cost sensitivities. In the first (Case III), all
assumptions were held^constant other than the selling price of the fuel. The
upper price of $76.68/ton below was established by assuming that fuel oil
could soon be selling for $0.50/gallon ($3.33/106 Btu) and that the waste-
derived fuel could approach that value as a limit. As yet unreported tests at
the Pittsburgh Energy Research Center demonstrate slurries of the solid fuel in
oil can be successfully fired and hence the assumption would appear to be a
reasonable one. Case III net costs are as follows:
130
-------
Selling Price of Product Unit Net Profit fLossj_
$/Mg $/Ton $/Mg Waste $/Ton Waste
30.86 28.00 (0.004) [0.004)
38.58 35.00 1.73 1.57
60.63 55.00 6.69 6.07
84.52 76.68 12.07 10.95
An annual net profit of $643,820 would be realized at the highest assumed
selling price, a most favorable situation considering the capital investment
is approximately the same if 100 units were to be built. It should also be
noted in the calculations above that a selling price of $28/ton results in
essentially a break-even situation.
In the Case I assumptions, the equipment was producing fuel 80% of the
entire year, and taking into account moves and set-up times had a total utili-
zation factor of 0.92. Agricultural wastes have a high degree of seasonality
and this effect on profitability was examined (Case IV) by calculation of two
reduced utilization factors. During down times it was assumed that no labor
costs would be incurred and that the equipment could be stored for $400 per
month. Case IVA is identical to Case I and the two variations are as follows:
Factor Case IVB Case IVC
Production days per year 266 210
Product per year, Mg (ton) 10 859 (11,970) 8573 (9,450)
Sites per year 19 15
Total working days per year 304 240
Cost of waste, $/Mg ($/ton) 0 0
Selling price of product, $/Mg ($/ton) 38.58 (35) 38.58 (35)
Annual labor cost, $ 172,920 136,200
Annual maintenance cost, $ 18,000 16,000
Annual transportation cost, $ 30,560 24,800
Annual supply cost, $ 4,000 4,000
Storage 800 1,600
Total Costs find. Amort.) $348,216 $304,536
Income 418,950 330,750
Profit 70>734 26'214
Profit/ton raw refuse $!-33 l^'^l
Profit/Mg raw refuse S1-47 $0'69
It can thus be seen that under the basic assumptions of Case I, the
system proves profitable down to a production utilization factor of 0.57 or
a total annual factor of 0.66.
131
-------
ENVIRONMENTAL CONSIDERATIONS
Little specific experimental work has been conducted on effluent composi-
tions or other environmental considerations of the operation of 200 TPD mobile
pyrolysis systems. Many hundreds of hours of pyrolysis and off-gas combustion
have been accumulated on the several units thus far constructed with no visi-
ble emissions being noticed under steady state conditions. In an analysis of
the stack while wood wastes were being used as the feed, Georgia Tech found
the following:
Component Concentration
Oxygen 0.9%
Nitrogen 69%
Carbon Dioxide 7.7%
Carbon Monoxide 30 ppm
Particulates 0.0005 g/Nm^ (0.0002 grains/SCF)
Hydrogen Sulfide 0.009 ppm*
Nitrogen Dioxide 0.04 ppm*
Ammonia 0.09 ppm*
Sulfur Dioxide 0.4 ppm*
Such results are to be expected from the combustion of a clean pyrolysis
gas. NOX could be significantly higher if high temperature combustion occurred.
Gaseous and particulate matter could be emitted from the waste introduc-
tion and char discharge systems of the pyrolysis converter unless proper valv-
ing and pressure differentials are designed into the equipment.
Emissions from the drying system need to be examined for the final dryer-
mechanical separator equipment chosen. Wastes containing large quantities of
fines could require a fabric filter (bag house) emission control unit and
odor levels should be examined in the final configuration. Careful control of
excessive temperatures within the dryer should eliminate this potential prob-
lem other than with unusual wastes that might contain a high degree of
volatile matter.
Through operation of the off-gas condensing system above the dew point
temperature of water, no liquid wastes will be formed at the facility. The
converter should not be permitted to be washed down onto open ground and the
finished product(s) should be protected against leakage by any route into
ground water supplies.
Noise power levels below OSHA regulations can be readily obtained through
proper design.
None detected; value listed as limit of detection.
132
-------
SECTION 7
ANDCO-TORRAX PYROLYSIS SYSTEM
INTRODUCTION AND SUMMARY
The Andco-Torrax system converts municipal refuse into a usable gas by
pyrolysis and then combusts the gas to produce heat for generation of steam.
Noncombustible materials are slagged at temperatures up to 1650°C (3,000°F).
The principal unit is a vertical reactor system that involves an air-blown
partial oxidation pyrolysis process. Refuse from a storage pit, untreated
and unsorted except for large bulky materials, is introduced via a feeder
charged by a crane bucket. The refuse slowly descends through the reactor,
encountering hot gases that dry it, and then thermally convert it to gases,
char, and ash. The char is burned at the bottom of the reactor with air pre-
heated to a sufficient temperature to slag the ash and metals. Figure 20
shows a schematic of the reactor or gasifier as now being built for commercial
application.
A demonstration plant using this principle was designed and constructed
by Andco Incorporated and the Carborundum Company; it was first operated in
the second quarter of 1972. Funding was largely supplied by the Federal EPA,
with additional support from New York State, Erie County, the American Gas
Association, and the developer. Since the summer of 1972 this plant has
operated or been available for use as an engineering development facility to
evaluate various design features and to test the- process on a number of in-
dustrial wastes admixed with municipal solid waste.
Design of the first commercial unit, now built in Luxembourg, utilized
the information obtained from the demonstration plant, but incorporated
several important changes. These include regenerative heat exchange for pre-
heating the gasifier air and an electrostatic precipitator. Three additional
steam generating systems are now being constructed in Europe. The Luxembourg
plant is in start-up phase with a capacity of 200 Mg/d and will produce steam
for use in a turbo-electric generator.
For The Ralph M. Parsons Co. to develop a basis for a range of larger
sizes, Andco Incorporated provided design and cost information utilizing a
300 Mg/d (331 TPD) module. These designs were, in turn, based on the ex-
perience with the Luxembourg plant. Details for a 3 module 900 Mg/d (992 TPD)
plant were developed with an extrapolation to two other sizes, a one module
300 Mg/d (331 TPD) plant and a five module 1500 Mg/d (1,653 TPD) plant.
The base case plant in this report has a capacity of 900 Mg/d (992 TPD)
and it produces 2215 Mg/d (2,442 TPD) of steam at 3.4 MPa (493 psia) and
133
-------
REFUSE
REFUSE
PLUG
DRYING
ZONE
COMBUSTION
AIR
PRIMARY
COMBUSTION
AND
MELTING
ZONE
SLAG
DROPOFF
AND
QUENCH
Figure 20. Schematic of the Andco-Torrax gasifier.
134
-------
385°C (725°F). A conversion efficiency of 76% is estimated with a net thermal
efficiency, taking into account electric power requirements, of 59%.
Residue as slag is produced at the rate of 212 Mg/d (234 TPD). Emis-
sions in the stack gas are 88 ppm HC1, 125 ppm SOX, 115 ppm NOX, and 3 to
4 ppm of hydrocarbons as measured from the demonstration plant. Using an
electrostatic precipitator, particulate matter can be expected to be signif-
icantly below the federal EPA standard of 0.08 gr/SCF at 12% C02.
Construction costs in 1976 dollars is estimated at $30.6 million and
capital requirements at $34.94 million. The annual operating cost is $2.74
million. A net unit cost was determined by using an amortization rate of
8-1/2% over 20 years. Credit was taken for drop charges of 0, $5, and $10
per ton. The net cost of producing steam was then $8.83/Mg ($4.00/1000 Ib)
at zero drop-charge and $4.36/Mg ($1.97/1000 Ib) at $10/ton drop-charge.
Costs as a function of plant size were estimated by extrapolating 900 Mg/d
plant values to a one module (300 Mg/d) and to a five module (1500 Mg/d)
plant. The capital costs are $14.99 million and $51.23 million respectively.
The corresponding unit costs for the 300 Mg/d and 1500 Mg/d plants are, for
a zero drop charge, $12.68/Mg ($5.75/1000 Ib) and $7.58/Mg ($3.46/1000 Ib)
respectively. For a $10/T drop charge, the unit costs are $8.21/Mg ($3.72/
1000 Ib) and $3.11 ($1.42/1000 Ib) respectively. These costs indicate that
the system is competitive with oil-based steam generators and in some areas
with coal based steam generators.
Course shredding of the raw refuse with recovery of materials should be
considered if added equipment costs are more than balanced by revenues from
material sales. Testing the Andco-Torrax reactor with a shredded feed would
be needed.
CONCLUSIONS
• The Andco-Torrax system as a steam generator can now be considered as
an available candidate for installation in a community. With several
commercial-scale facilities now in construction or undergoing start-
up testing, possible clients can review capital and operating cost
data to learn how this approach compares economically with waterwall
combustion systems. Total effluent gases to generate a unit of steam
can be less than in the mass burning incineration case, and, if other
cost factors do not cancel out this advantage, this pyrolysis approach
could become a serious competitor to the older combustion technology.
• Clean up of the rather low heating value off-gas for pipeline trans-
port to a utility cannot be recommended by Parsons based on existing
data. The loss of sensible heat in such an application is another
disadvantage to this approach once suggested by the early developers.
• In communities where no client exists for steam, and electric power
costs are somewhat higher than average, the financial feasibility of
adding a turbine-electric unit should be considered.
135
-------
• The glassy slag residue produced is essentially organic free and can
be used as clean fill or as a base for building materials.
* Pollutant emissions are low enough to meet local, State, and Federal
environmental requirements in most areas of the U.S.
• Cost data presented in this report, although in 1976 dollars, should
permit valid projected comparisons with fossil fuel generated steam,
in that the market value of steam should escalate with the economy.
• High temperature systems can always benefit from improved materials
of construction and R§D should be continued for such improved re-
fractories and metals. Cycling of high temperature processes should
be avoided and the Torrax equipment will enjoy lowest maintenance
costs when operated on a steady basis.
• Capital costs increase almost linearly with plant size because of
modularization of most of the plant components. Unit steam costs
decrease somewhat linearly with increasing plant capacity, with a
flattening of unit costs above about 1200 Mg/d.
LUXEMBOURG PLANT PROCESS DESCRIPTION
In Luxembourg, a 200 Mg/d plant has just been completed and is in the
start-up phases. Figure 21 shows a schematic of the gasifier and secondary
combustion chamber (SCC) while Figure 22 shows a schematic layout of the sys-
tem. This includes the gasifier, SCC, two regenerative heat exchange towers
filled with ceramic/checker-work brick, a waste heat boiler, and an electro-
static precipitator. Air for the reactor is compressed and passed through
one of the regenerative towers while the other is being heated with a portion
of the hot gases from the SCC. Particulate removal is affected by a slagging
in the combustor and by passing all exhaust gas streams through the electro-
static precipitator.
The Torrax installation is part of a complex that includes two conven-
tional grate incinerators to process a total of 600 Mg (661 tons) per day of
municipal waste from the regional area of Luxembourg City. The boiler is a
three-pass vertical combination radiation/convection boiler with the first
of three passes being in the radiation portion. There is a superheater at
the end of the radiation section and an economizer in the third pass. Per-
formance of the boiler with a feed water temperature of 110°C (230°F) is
designed to produce a maximum of 30 Mg/h (33 tons/hr) of steam at 3.4 MPa
(493 psi) and 385°C (725°F). The boiler has been designed for a maximum gas
throughput of 36 800 Nm3/h (22,800 SCFM), a maximum gas temperature of 1370°C
(2500°F), and a nominal gas temperature of 1250°C (2282°F). The steam pro-
duced from the Andco-Torrax unit and the incinerators is led to a 7 MW
turbine-generator producing electricity at 10 kV. The turbine-generator is
designed for the conversion of 600 Mg (661 tons) per day of refuse with a
lower heating value of 8.4 MJ/kg (3,611 Btu/lb). A second turbine-generator
will be installed as refuse production increases. Other details can be found
in Reference 13.
136
-------
REFUSE
REFUSE
PLUG
DRYING
ZONE
COMBUSTION
AIR
PRIMARY
COMBUSTION
AND
MELTING
ZONE
COMBUSTION
AIR
SOLIDS
SLAG
DROPOFF
AND
QUENCH
PYROLYSIS
ZONE
FINAL
COMBUSTION
SECONDARY
COMBUSTION
CHAMBER
Figure 21. Arrangement of Andco-Torrax gasifier-combustion chamber at Luxembourg.
-------
REGENERATIVE
TOWERS
ELECTROSTATIC
PRECIPITATOR
GASIFIER
SECONDARY
COMBUSTION
CHAMBER
WASTE HEAT
BOILER
Figure 22. Layout of Andco-Torrax system at Luxembourg.
138
-------
THREE MODULE PLANT PROCESS DESCRIPTION
Three Andco-Torrax modules are used to process 900 Mg/d (992 TPD) of
municipal refuse. Figure 23 shows a schematic flow diagram of a single module.
The train of equipment shown is duplicated for each module except for the
refuse pit, crane, and the slag collection pit. Refuse is introduced to the
receiving pit and an overhead crane places material into the feed hopper for
each gasifier. The refuse composition is that from the 1975 EPA Report to
Congress (SW-16) and shown in the Purox Section. The analysis can be re-
arranged for use in computations made by Andco as dry combustibles 54.2%, dry
non-combustibles 20.7%, and moisture 25.1% by weight. Slag from each of the
gasifiers and each of the secondary combustion chambers is sent to a single
slag pit from which the residue is conveyed automatically to trucks for
removal. The process requires the following major pieces of equipment:
1. Receiving area and refuse pit
2. Crane - A single crane and clamshell or orange peel bucket can be
used to pick up a load and deliver it to one of the three gasifier
feed hoppers. Two grab buckets may be needed for redundancy where
3 or more gasifier modules are installed.
3. Feeder - Raw refuse from the crane is dropped into a hopper with a
ram feeder mechanism near the top of the gasifier.
4. Gasifier
5. Slag Quench - Slag formed in the hearth area of the reactor runs
into a slag tank filled with water for quenching.
6. Secondary Combustion Chamber (SCC) - To receive and burn the hot
off-gases and entrained char from the gasifier to produce a high
temperature exhaust gas. Additional slagging of fly ash occurs
also, which is quenched.
7. Waste Heat Boiler - To convert the heat in the secondary combustion
chamber exhaust gas to steam.
8. Regenerative Towers - Two checker brick-filled towers operate alter-
nately for heating the combustion air to the gasifier, with heating
of the bricks by a portion of the SCC exhaust gases.
9. Electrostatic Precipitator - For removing particulate matter from
the exhaust gases from the regenerative towers and the waste heat
boiler.
10. Air Moving Equipment - For introducing primary combustion air through
the regenerative towers to the gasifier, secondary air to the SCC,
and exhausting the cleaned exit gases to the flue gas stack.
Off-gases from the gasifier contain hydrogen, carbon monoxide, carbon
dioxide, hydrocarbons, nitrogen, water vapor, and char plus ash particulates.
139
-------
COLD BLAST
FLY ASH T
Figure 23. Schematic flow diagram of Andco-Torrax system.
-------
With the high nitrogen content, the higher heating value of the gas is in the
range of 3.94 to 4.72 MJ/Nm3 (100 to 120 Btu/SCF). For one composition ob-
tained from the demonstration plant test unit, as shown in Table 20, the
higher heating value was 4.25 MJ/Nm3 (108 Btu/SCF). When account is taken
of the sensible heat (427°C or 800°F) and the heating value of the char par-
ticles, the total HHV becomes 6.89 MJ/Nm3 (175 Btu/SCF). This gas is not
useful for chemical synthesis and as a fuel must be burned in a closely
coupled secondary combustion chamber to recover the high sensible heat.
TABLE 20. GAS COMPOSITION
Constituent Vol %
CO
H2
CH4
C2's
CO
2
N2
H-0
2
Total
13.1
12.5
2.0
0.3
10.1
43.5
17.0
98.5
HHV =4.25 MJ/Nm3 (108 Btu/SCF)
Char and sensible heat included in
total heating value to secondary
combustion chamber
HHV (Total) 6.89 MJ/Nm3 (175 Btu/SCF)
HEAT AND MASS BALANCE
Figure 24 shows a mass balance for the various streams in the Andco-Torrax
system, with the values given being based upon one Mg of raw refuse feed.
Figure 25 shows an overall mass flow block diagram for a 900 Mg/d plant. In
Figure 26 is shown a heat balance based on the energy input from one Mg of
refuse, while Figure 27 shows an overall heat balance for a 900 Mg/d plant
with electric power requirements. Table 21 shows the details for electric
power requirements for a 300 Mg/d module. In assessing the thermal per-
formance of the system to produce steam, efficiencies are used with some
variation in definition. The heating value of the fuel is used in the de-
nominator. If there is much moisture, the difference in lower and higher
heating value is quite significant. To make steam, use of the lower heating
value is more realistic and used in Europe, but the higher heating value is
used generally in the U.S. to calculate thermal efficiencies. Values for
141
-------
FUEL
0.004
GASIFIER
AND
SECONDARY
COMBUSTION
CHAMBER
0.512
4.565
0.015
Figure 24.
Generalized mass balance for Andco-Torrax system
Units in Mg of steam per Mg refuse.
142
-------
REFUSE
AIR
900
(9921
4Q16
FEEDWATER
(5086)
2216
(2442)
FUEL
(4.4)
ANDCO-TORRAX
SYSTEM
2216
(2442)
212
STEAM
SLAG
(234)
5295
STACK GASES
(5834)
13
PARTICULATES
(14.4)
Figure 25. Mass balance summary for 900 Mg/d (992 TPD) Andco-Torrax
system. Units in Mg/d (TPD).
143
-------
ELECTRIC POWER
REQUIRED IS 107
kWh/Mg,WITH TH
ENERGY TOPROI
BEING 843MJ/Mg
FUEL
REFUSE
HHV 10335 |_LVH = 8974
1450
FEEDWATER
107.8 ^ — v
1 THE STEAM j
'RODUCETHIS V
V
184
\
>
/
/
""" SLAG 371
f
/
r~
LOSSES 557
GASIFIER
AND
SECONDARY
COMBUSTION
SYSTEM
1 10260
HOT GASES j
>1
f J
HOT BLAST
13
942
669
9318
BOILER
STEAM
7850
1 1
3.4
(49
38
385°C (725°F)
EXIT
GASES
2985
Figure 26. Generalized (1-Mg refuse input) heat balance for
Andco-Torrax System. Units in MJ/Mg refuse.
144
-------
REFUSE
AIR
FEEDWATER
FUEL
9.302
(8.8161
0.012
(0.014)
1.306
(1.238)
0.166
(0.157)
ANDCO-TORRAX
SYSTEM
7.065
(6.696)
0.523
(0.4961 TOWER LOSSES
0.146 _
(0.138)
0.031
(0.029)
2.686
(2.546) *"~
0.335
(0.318)
Figure 27. Heat balance summary for 900 Mg/d (992 TPD)
Andco-Torrax system. Units in TJ/d (109 Btu/day).
145
-------
TABLE 21. MAJOR ELECTRICAL EQUIPMENT USE, 300 Mg/DAY (331 TPD) MODULE
Item
System I.D. Fan
Primary Fan
Secondary Fan
Boiler Feed Pump
Open Water Pump
Closed Water Pump
Electrostatic Precipitator
Peripheral Requirements
Total
Power
kW
247
62
309
185
296
87
26
1 212
HP
331
83
414
248
397
117
35
1,625
1347 x 24 x 3.6
Electrical input = 1212/0.9 = 1347 kW
c • i * u * .c ci * • n
Equivalent Heat for Electric Power = .
=3.6 MJ/kWH
Efficiency of Steam to Electric Power Used = 46%
= 843
MJ
Mg
both cases are given here so that the reader can make whatever comparisons he
desires with other steam producers. Energy values are taken from Figure 26.
The conversion efficiency (i?c) is:
„ rimn _ energy in steam - energy in feed water 7850-1450 _„
TyCIL/rtV) — ; r; — : ;; — .-,_.— . X 1UU = /I. J'o
Lower Heating energy in refuse 8974
r?c(HHV) =
7850 - 1450
10335
x 100 = 61.9%
The net thermal efficiency (rjt) is based upon the amount of energy in to
make the steam less the energy in product steam required to produce the elec-
tric power to drive the various pieces of equipment in the system. This is
divided by the sum of energy in the incoming refuse, energy in supplementary
fuel oil, energy in the water to the waste heat boiler, and energy in the
secondary and primary combustion air. The value is given as
146
-------
net energy out - 7850 - 1450 - 845 _
~ 6U-6°
" sum of energies in 8974 + 184 + 13
flt - 7850 - 1450 - 845 _
T?t(HHV) ' 10555 + 184 + 15 X 10° ~ 52'7*
Shredding and processing the raw refuse may be desirable in order to
recover the revenue-producing materials. Additional equipment will be re-
quired and could be amortized by the material-sale revenues. One method for
simple processing of the entering refuse to recover some of the materials is
to pass the raw refuse through a trommel such has now been installed in
New Orleans (Ref. 7). This serves to break open bags, break glass so that
it is sifted out by the trommel, and separate the smaller steel cans so they
can then be recovered with a magnetic device. Such an arrangement could be
considered for a large plant operation but is not included in the analysis
for this report. A certain amount of glass is desirable in the refuse feed
to assist in the slagging by lowering the melting point.
FACILITIES AND PROCESS EQUIPMENT PERFORMANCE
Equipment for a 900 Mg/d (1,000 TPD) plant has been priced based upon
estimated costs by Andco and Parsons. Figure 28 shows a plan view of the
plant including the various pieces of equipment that were described in the
process diagram of Figure 25. In bringing together three such 500 Mg/d
modules, the building will be extended in length and a raw refuse pit will
be designed in the system to be used for feeding all three gasifiers. Al-
ternative schemes to pit-and- crane raw refuse handling should be evaluated.
A more detailed analysis is required to determine the amount of refuse
to be stored. Andco plans storage for as much as 3 days and this may be
necessary where a reliable flow of steam is required without the use of much
auxiliary oil-firing. Gasifier operation characteristics are such that it
can be turned down by about 15% if the availability of refuse has decreased
and more is not available for a period of time.
Figure 29 shows an elevation of the 300 Mg/d module. An emergency spill
line, not shown, for emergency dumping of gasifier off-gas to an aspiration
burner is to be included. It is estimated that for a 900 Mg/d plant with
sufficient room for movement of trucks, dropping of refuse, and location of
ancillary equipment, 52 376 m2 (8 acres) of land may be needed.
The operation of the plant on an annual basis would require approxi-
mately 2 to 5 weeks shut-down for scheduled maintenance. In addition, there
may be unscheduled down-time, with a total of scheduled and unscheduled
down-time of approximately 4 to 5 weeks. This results in a utilization factor
of approximately 90 percent, although 85 percent would be more in keeping with
this type -of high temperature equipment. Use of three modules adds a degree
of redundancy and therefore more production reliability. Table 22 shows the
summary of plant production flow rates. On a yearly basis, a utilization
factor of 90 percent is used.
147
-------
00
WASTE HEAT BOILER
ISO-
Figure 28. General arrangement plan for 300 Mg/d Andco-Torrax system.
-------
ELECTROSTATIC
PRECIPITATOR
Figure 29. Elevation of the Andco-Torrax system 300 Mg/d module.
-------
TABLE 22. ANDCO-TORRAX 900 Mg/d (992 TPD)
PRODUCTION FLOW RATES*
Item
Refuse
Steam
Residue
Units
Mg/d
Mg/y
Tons/d
Tons/y
Mg/h
Mg/y
Ib/hr
lb/y
Mg/d
Mg/y
Tons/d
Tons/y
Amount
900
295 700
992
325,900
92.32
727 800
203,500
1.605 x 109
212
69,600
234
76,800
*0.9 Utilization Factor
Details of much of the equipment are proprietary to Andco and not com-
pletely available for this report. The following discussion is a general
description of the performance of major equipment items.
Gasifier
In the bottom, or hearth area, preheated air at approximately 1093°C
(2,000°F) comes from the regenerative towers through a hot blast main into a
circular bustle pipe. It then passes through a multiplicity of downcomer-
tuyere arrangements into the hearth area to burn the char descending from the
pyrolysis area. The major section of the gasifier shell is water cooled,
resulting in a protective skull* of solidified slag being formed over the
refractories in the hearth area. This skull protects the refractories from
the high temperatures of approximately 1649°C (3,000°F) and corrosive action
of the molten metals and slag. The molten slag flows into a quench tank that
is periodically dumped, allowing discharge of the shattered glassy aggregate
or frit while the pressure in the reactor remains slightly above atmospheric.
*A hardened layer of slag next to cooled refractory.
150
-------
Quench water is continuously introduced to keep this part of the equipment
cool. Refuse from the pit entering the top is fed into a feed hopper by the
grapple bucket and then automatically fed from the hopper into the gasifier
reactor vessel by a ram mechanism. The gasifier is approximately 2.8m
(9 ft) in diameter and 16.3 m (53 ft) high at the feed mechanism at the top
and 4 m (13 ft) below floor level.
Moist refuse entering at the top of the gasifier moves downward past the
gas exit area with a lantern-type of gas off-take. From this point down, the
refuse dries^and then, in the pyrolysis zone, is heated from 260 to 1093°C
(500 to 2000°F), where the rate of decomposition to gases, oil, char, and ash
increases with temperature. Various oils formed in the low temperature region
of pyrolysis continue to pass down and are cracked into gases and char at the
higher temperatures. Those particles of char and oil that are entrained in
the hot gases are mostly scrubbed out by the descending refuse, thereby re-
cycling down into the higher temperature zone. A portion of the char and ash
particulates pass out with the off-gas.
The heat generated by the burning of char with the preheated air in the
hearth area provides for the drying of the refuse, the heating of the refuse,
heat losses through the walls, and melting of the slag. The heat of chemical
reaction may be slightly endothermic. Most of the heat required is for
raising the temperature and drying the refuse. Gases leave the reactor at
450° to 550°C (800° to 1,000°F).
Fuel Gas System
About 90 percent of the energy content of municipal refuse is contained
in the gas stream that leaves the gasifier. This energy is in the form of
combustible gases, vapors, and entrained particles, and as sensible and latent
heat. The temperature of this gas is approximately 427°C (800°F). Complete
combustion of this gas stream produces about the same volume of products of
combustion per unit of heat released as would be the case with other gaseous
fuels. Because of the high sensible heat content of the gas, it must be
closely coupled with a combustion chamber. Other than a combustion chamber
and steam generator, the products can be utilized if the gasifier is closely
coupled to other apparatus such as cement kilns, lime kilns, or drying
systems.
Composition of the combustible gas stream is dependent on the refuse mix,
but should be approximately that which has been measured at the demonstration
unit (Table 20). The higher heating value of the hot char-laden gas will be
approximately 6.90 MJ/Nm3 (175 Btu/SCF).
Secondary Combustion Chamber (SCC)
The combustible gas-vapor mixture from the gasifier is thoroughly mixed
with a minimum of excess air and burned to completion in the secondary com-
bustion chamber. The fuel gas mixture drawn out of the gasifier lantern
section is typically at a temperature of 315° to 550°C (600° to 1,000°F).
The fuel gas is mixed with ambient combustion air in a high energy mixing
burner at the inlet to the secondary combustion chamber. Gas enters the SCC
151
-------
in a tangential fashion for high turbulence spiral action to further aid com-
plete combustion. The thorough mixing, high temperature, and long residence
time in the refractory-lined SCC insures complete combustion with a minimum
of excess air (approximately 63 percent).
High temperatures maintained in the combustion chamber (1150 to 1260°C
or 2,100 to 2,300°F) causes the fly ash and other inert carryover materials
to melt, fuse, and be slagged out of the stream. Lower particulate and lower
objectionable gaseous emissions are claimed as compared, for instance, with
conventional incinerators. Andco indicates that smaller gas cleaning equip-
ment is required to meet allowable emission codes.
Regenerative Towers
At the demonstration plant, a natural gas direct-fired, silicon carbide
cross-flow shell and tube heat exchanger was used to supply the high tempera-
ture primary air to the gasifier. While this unit has proven reliable, the
cost and the diminishing availability of natural gas make such a unit unat-
tractive for commercial plants. Use of a regenerative type of heat exchanger
results in energy and cost savings.
Two vertical refractory-lined steel shells, filled with checker-work
refractory brick, are used as regenerative heat exchanger towers to recover
a portion of the process heat for pre-heating the primary combustion air as
shown in Figure 30. About 10% by volume of the hot gaseous products of com-
bustion are drawn from the SCC and passed through the top of one of the two
regenerative towers. The induced draft fan downstream from the electrostatic
precipitator causes the flow. Refractories in the tower are heated to a tem-
perature of approximately 1150°C (2,100°F) at the top and 260°C (500°F) at
the base. The waste gas exiting the regenerative tower checker-work is re-
turned to a duct at the inlet of the gas cleaning system (electrostatic pre-
cipitator) . A modulated damper valve in the boiler exit duct controls the
amount of gas used in heating the regenerative tower checkerwork. Combustion
air, on being passed through the hot refractories, is preheated for the
gasifier. The two refractory lined regenerator towers are automatically and
alternately heated and cooled in predetermined controlled cycles. Refractory-
lined and checker brick-filled regenerators of this type have long been used
in iron, steel, and glass making.
During the "blast" cycle, the combustion products from the secondary
combustion chamber are diverted to the second regenerative tower and ambient
process air is introduced at the base of the fully heated regenerative tower,
where it passes up through the checkerwork absorbing the stored heat. The
exit temperature of the air from the checkerwork ranges from 980°C (1,800°F)
to 1150°C (2,100°F). This preheated air then flows to the gasifier hearth.
A constant primary air, or blast, temperature is maintained by blending the
heated air with ambient air before introduction into the gasifier. Automatic
sequencing of the regenerative towers is controlled by temperature sensors.
152
-------
REGENERATIVE
GAS ^Jr
FROM
SECONDARY
COMBUSTION
CHAMBER
HOT BLAST
VALVE
CLOSED
COLD BLAST
VALVE
CLOSED
REGENERATIVE
GAS VALVE
CLOSED
WASTE
GAS TO
CLEANING
SYSTEM
COLD
BLAST
HOT
BLAST
TOGASIFIER
TYPICAL
CHECKER BRICK
WASTE GAS
VALVE
CLOSED
Figure 30. Schematic of regenerative towers,
153
-------
Waste Heat Boiler
The major portion (90 percent) of the volume of combustion products
drawn from the secondary combustion chamber passes through the waste heat
boiler, which is a combination radiation-convection type. As much as 3 kg
of saturated steam is produced for every kg of municipal refuse consumed in
the Andco-Torrax System. Gases leave the waste boiler at approximately 300°C
(592°F) and combine with flow of spent gases from the regenerative towers
before entering the electrostatic precipitator.
It was determined by Andco that the waste heat boiler is physically
smaller but equal in steam generation production to conventional incinerators
of equal refuse capacity for the following reasons:
• Higher inlet gas temperatures.
• Lower gas volume due to lower excess air.
• Slagging of ash in secondary combustion chamber minimizes slag carry-
over to the boiler, permitting use of a large convective tube section.
Additionally, boiler tube cleaning is simpler and is required less frequently
than in the conventional incinerator boilers because of removal of a good
deal of ash and soot in the SCC.
Gas Cleaning System
The waste gases leaving the regenerative towers and the waste heat boiler
are combined before entering the gas cleaning system, which usually consists
of an electrostatic precipitator, sized to handle the maximum gas flow from
the secondary combustion chamber. Because of the lower gas volume resulting
from lower excess air and a lower particulate loading, a smaller precipitator
can be used in the Andco-Torrax system than that for conventional systems to
meet emission requirements.
Depending on the emission requirements in a particular municipality, a
wet scrubber may be preferred. Such equipment can offer installation economy,
as welJ as being able to remove objectionable gaseous constituents in addition
to particulates. Because of the acid gas content of the emissions, gases
throughout the electrostatic precipitator are kept well above the dew point
to prevent corrosion problems both in the precipitator and the stack attached
downstream. Refuse contains a low sulfur content compared to other fuels and
hence the SOX formed is usually sufficiently low enough to meet emission
standards in most areas. Because of the alkaline character of the ash, as
much as 25% by weight being calcium oxide, a considerable amount of SC>2 will
be absorbed in a highly turbulent, refuse fly-ash filled combustion chamber.
Trace emissions measured will be discussed later. The normal dry gaseous
principal components in the flue gas when processing 10.33 MJ/kg (4,445 Btu/lb)
refuse are 80.4 percent nitrogen, 12.4 percent carbon dioxide, and 7.2 percent
oxygen by volume. With moisture, by Parsons calculation, there is 68.2 per-
cent N2, 10.5 percent C02, 6.1 percent 02, and 15.2 percent H20.
154
-------
Process Residue
The slag formed in the hearth of the gasifier and the secondary combus-
tion chamber are quenched, resulting in a black glassy aggregate or frit. A
chemical analysis is described in Table 23. The particle sizes range from
0.4 to 6 nun with most at 2 mm. This material is passed into a slag holding
tank, and several methods are available for transporting it from the tank.
At the demonstration plant, the material flows into a slag pit and a front-
end loader is used to carry it to a pile for later removal by truck. An
automatic method would use drag and belt conveying transfer, with deposition
being made at a continuous rate on horizontal or inclined belt conveyors
feeding a vertical bucket elevator. The bucket elevator directs the material
to external storage bins or stock piles for transport to suitable disposal
areas on an as-needed basis. A third method is the pumping of a suspension
of slag from the main pit to dewatering tanks. The suspended solids quickly
settle out in the dewatering tank for removal by grapple bucket or drag con-
veyor system to a point of disposal. The conveying liquid is pumped back
through the system from the dewatering tanks for reuse in the quench tanks.
TABLE 23. CHEMICAL ANALYSIS OF RESIDUE
Constituent
Si02
A12°3
Ti02
Fe203
FeO
MgO
CaO
MnO
Na20
K20
Cr203
CuO
ZnO
Total
% by Weight
46.0
10.0
0.8
10.0
15.0
2.0
8.0
0.6
6.0
0.7
0.6
0.2
0.1
100-0
bulk density 1.40 g/cm3
true density 2.80 g/cm3
155
-------
Volume reduction to the slag residue from the raw refuse is approximately
95 to 97% and the weight reduction 80 to 85%, depending on the inert fraction
in the refuse.
Industrial Refuse Capability
In addition to normal municipal refuse, the Andco-Torrax System is capable
of handling other wastes., particularly those from industry. These wastes can
be mixed with municipal refuse with minor changes to the equipment and the
operating procedures.
Some of the tests at the demonstration facility include the following:
* Sewage sludge. Undigested sewage sludge with 78 percent water content
was charged with the municipal refuse in quantities averaging 28.5
percent of the total 3.4 Mg/h (3.8 TPH) charge.
• Waste oil. Waste automotive lubricating oil was charged with muni-
cipal refuse in average quantities of 6.1 percent of the 4.0 Mg/h
(4.4 TPH) charge.
• Combined sludge and oil. A test combining sewage sludge with waste
oil and municipal refuse was accomplished. The total feed rate was
3.1 Mg/h (3.4 TPH) of which 30.1 percent was sludge and 3 percent
was oil.
• Tires. Unshredded automotive tires were charged with the normal
refuse. The average addition was 30 tires per hour or about 10 per-
cent of the total 3.0 Mg/h (3.3 TPH) consumption.
• Polyvinylchloride (PVC). Bags filled with PVC plastic waste were
charged with municipal refuse in amounts averaging 7 percent of the
2.9 Mg/h (3.2 TPD) charge.
In all cases no significant changes to the process operation were encountered.
The maximum amount of admixture of special waste has not been determined.
ENVIRONMENTAL FACTORS
Emissions
Waste gas emissions from the Andco-Torrax system will be from a stack
after passing through a gas cleanup system.
Each 300 Mg/d module will require a stack to handle about 57 700 Nm3/h
(35,600 SCFM) of gas at 260°C (500°F). Based on the Luxembourg design con-
straints, a 1.6 m (5 ft) minimum diameter (ID) stack would be required ap-
proximately 50 m (164 ft) in height. Grouping three flues in one stack would
require a minimum 4.6 m (15 ft) diameter single stack.
Height restrictions are based on emission regulations as well as local
requirements. Generally, regulatory agencies provide height restrictions
156
-------
that are a function of concentration per unit area per unit time as a result
of stack emission rates. To appropriately size a stack for use in a specific
site, the necessary local condition requirements should be provided.
Provisions are also made for venting pyrolysis gases from the gasifiers
in the event of emergency shutdowns. For this purpose, a single gasifier that
is being shut down could be vented to a manifold to allow for combustion of
these gases in the adjoining systems. In the event of complete power failure,
an externally ducted spill-stack is included with an aspirator burner.
A self-contained power unit should be provided for a plant this size
to actuate the necessary hardware in an emergency situation. A diesel auxil-
iary power unit for a 300 Mg/d module would produce about 350 kW needed for
cooling water systems, instrumentation, control, and lighting.
Representative sampling of waste gases exiting from the boiler and prior
to gas cleaning has been conducted at the demonstration facility over several
months. The following are typical values obtained from this sampling:
Cl_ (Expressed as HCl)--88 ppm (EPA sampling in a previous period indicated
values as low as 13 ppm; it is expected that these differences are caused
by fluctuations in refuse compositions.]
SOX--125 ppm.
NOX--115 ppm.
Organic acids--0.2-0.6 ppm.
Hydrocarbons--3 ppm.
Aldehydes and Ketones--0.02-0.40 ppm.
Particulates--In that only a spray tower was used, particulate matter was above
emission standards at approximately 1.3 g/Nm3 (0.5 gr/SCF). There were no un-
usual characteristics of the gas stream or particles that would interfere with
the operation of an electrostatic precipitator, baghouse, or venturi scrubber
to significantly reduce particulate matter, certainly below the federal EPA
standard of 0.08 gr/SCF at 12% C02-
Water--The quench water us-ed in the process makes contact with residues that
contain no putrescibles, and hence no organics are added to this water. Sus-
pended solids and pH measured at the demonstration plant were approximately
30-60 mg/dm3 and 7.8, respectively. This waste water is of a quality which,
together with the sanitary wastes from the plant, can be directed to city
sewer systems.
ESTIMATED CONSTRUCTION AND OPERATING COSTS
Construction costs include land, site improvement, the Andco-Torrax
pyrolysis system with its related equipment, contingencies, and engineering
157
-------
and construction fees. Operating costs are those needed for the functioning
of the plant. Financial charges associated with the long term amortization
of plant costs are included in the economic analysis.
Construction and Capital Cost
Construction Cost--
Table 24 presents a summary of the estimated construction cost for a 900
Mg/day plant in 1976 dollars. Equipment cost estimates shown include instal-
lation of a conventional system without site acquisition, preparation, and
development costs. Also, it would not include offsite distribution of steam.
Installed cost was estimated to be $26,160,000 in 1976 dollars.
TABLE 24. ESTIMATED 1976 CONSTRUCTION COST SUMMARY, 900 Mg/d
(992 TPD)
Item
Land
Site Improvement
Andco-Torrax System:
Gasifiers
Regenerative Towers System
Secondary Combustion Chambers
Boilers
Gas Cleanup (ESP's)
Instrumentation/Controls ,
Cooling Tower, Piping,
Pumps, Plant Air
Buildings
Other Items (crane, etc.)
Subtotal
Contingencies (10%)
Engineer £ Construction Management*
Total
Cost ($000)
200
800
10660
incl.
incl .
incl .
incl .
6,500
7,000
1,000
26,160
2,620
1,850
30,630
*Some engineering costs included in equipment costs.
Site acquisition, preparation, and development is based on a land area
of approximately 32 000 m^ (8 acres). The land to be acquired would generally
be in an industrial area with a total cost estimated to be $200,000. Site
preparation includes clearing, utilities, earth work, grading, paving, land-
scaping, and fencing, which would cost approximately $800,000.
Total construction cost was then estimated to be $30,630,000 in 1976
dollars.
158
-------
Capital Costs--
To establish capital requirements, Parsons added interest during con-
struction, start-up costs, and working capital to the construction costs of
$30,630,000. The individual items are given in Table 25, which shows the
total estimated capital cost to be $34,940,000 in 1976 dollars. Interest
during construction is determined by using 8-1/2 percent of the construction
cost for a period of one year. Start-up costs are based upon the fact that
the plant on the average will only be operating 50 percent of the time for
the first year, and the cost during that year in terms of capital will there-
fore be 50 percent of production related costs; this is one-half of $2,161,000
or $1,080,000. The working capital of $660,000 is based upon 25 percent of
the annual operating cost, which is shown later in this section to be
$2,740,000.
TABLE 25. 1976 CAPITAL COSTS, 900 Mg/d (992 TPD)
Item
Construction
Interest during Construction
Startup Costs
Working Capital
Total
Cost ($000)
30,600
2,600
1,080
660
34,940
Operating Costs
Operating costs include labor, electric power, heating fuel, maintenance
supplies and replacement parts, water, residue disposal, insurance, and taxes.
A management fee to an operating organization (or an imputed cost in the case
of government operation to properly allocate expenditures for billing, manage-
ment control, Engineering Department staff, etc.) would oftentimes be included,
but has not been done here because of its highly variable nature. Table 26
shows a summary of the operating costs. Accurate maintenance costs will be
developed as the Luxembourg installation becomes operational and on-stream for
a period of time. Information compiled by Andco and data available from in-
dustry estimates suggest that for the maintenance of a 300 Mg/d plant the
annual cost should be approximately $225,000. For three modules, the cost
would be $675,000, consisting of $387,000 for labor and $288,000 for supplies
and replacement parts. The labor portion is included in the separate labor
estimates.
Fuel oil is used in three places, 2.18 dm^/Mg for the slag trap, 0.44
dm3/Mg for the main burner, and 2.18 dm3/Mg for the combustion chamber slag
trap. Total fuel oil use is 4.80 dm^/Mg (1.15 gal/ton) or 1577 mVy. This
represents 9,700 barrels per year, and with #2 distillate fuel at $15 per
barrel, the cost of fuel oil would be $146,000.
159
-------
TABLE 26. 1976 OPERATING COSTS, 900 Mg/d (992 TPD)
Item
Labor
Electric Power
Heating Fuel
Maintenance Supplies and
Replacements
Water
Residue Disposal
Insurance
Taxes
Total
Annual Cost ($000)
1,188
798
146
288
49
70
201
-
2,740
The water required is approximately 3 percent makeup of 25 m3 of water
required per Mg (6,000 gal/ton). The annual makeup therefore is 0.03 x 25 x
900 x 365 = 246 000 m3/y (65 million gal/yr). At $0.198/m3 ($0.75 per thou-
sand gallons), the annual cost of makeup water is $48,750.
The cost of labor as presented in Table 27 is based on use of three shifts,
with the number of people required being determined from Andco and Parsons
estimates. It was estimated by Andco that for a 900 Mg/d plant, 40 to 45
individuals will be required over three shifts. Table 27 shows an estimate
of 50 people over three shifts as determined by Parsons. Basic pay rates are
typical for a major city in the United States. The total labor direct cost
per hour amounts to $370.50. Using 2,080 hours per year, the total cost is
$771,000 per year, to which overhead and fringe benefits (amounting to an addi-
tional 50 percent, or $385,000) must be added. Overtime is estimated to be
3,000 hours at $7 per hour with a 50 percent premium, or $32,000/year. The
total labor cost is therefore $1,188,000 as noted in Table 26.
Electric power costs result from the use of approximately 1180 kW (see
Table 21) associated with major electric equipment for each module. Total
annual power amounts to 31.9 x 106 kWh/y, for an annual cost, based on 25
mills/kWh, of $799,000. The use of 25 mills/kWh is suggested by Sussman of
the EPA as an average value to be used in the United States. Electric power
costs range from 10 to 40 mills/kWh with some special areas at 60 mills/kWh.
Residue disposal assumes no market for the frit. This is a conservative
estimate since this material is sterile and can be used for a variety of pur-
poses. Possible uses have been studied and evaluated, such as roadway aggre-
gate, black color additive to terrazzo, shot blast media, and decorative panels.
160
-------
TABLE 27. 1976 LABOR COSTS, 900 Mg/d (992 TPD), 3 SHIFTS
Job Description
Superintendent
Chief Operator
Assistant Operator
Crane Operator
Mechanical
Shear Operator
Maintenance Mechanic
Electrician
Scale Clerk
Janitor
Laborers
Secretary /Records
Total
Number of
Positions
1
8
12
7
8
1
3
2
2
1
3
2
50
Basic
Pay
$/hr
11.00
9.00
7.50
7.50
6.50
7.50
7.50
7.50
6.50
6.00
6.00
5.50
Total Pay
$/hr/Position
11.00
72.00
90.00
52.50
52.00
7.50
22.50
15.00
13.00
6.00
18.00
11.00
370.50
Until a definite market is established, it is estimated by Parsons that this
residue will be sent to a closeby landfill at a cost of $1 per Mg. There is
about 212 Mg/d of slag to be disposed, and therefore the total cost per year
is $70,000.
General insurance amounts to 0.6 percent of the construction cost of
$30,600,000 or $183,600. Added to this is personal hazard insurance at 1.5
percent of the direct payroll of $1,188,000, or $17,800, for a total insurance
cost of $201,400.
From Table 26 the total operating cost for the plant is seen to be
$2,740,000 per year.
ECONOMIC ANALYSIS
In this section, an analysis is given for the total unit cost to operate
a 900 Mg/d Andco-Torrax system in terms of dollars per Mg or ton of refuse
(or dollars per Mg or thousand pounds of steam). An amortization cost is
161
-------
included using an 8.5 percent interest rate over a 20 year lifetime after
start-up. With a capital cost of $34,940,000, and an amortization factor of
0.10567, the capital recovery per year is $3,692,000. The annual operating
cost from the previous section is $2,740,000 per year, which yields the total
annual cost of operation of $6,432,000 shown in Table 28.
TABLE 28. NET COST (1976 DOLLARS) TO PRODUCT STEAM(l)
900 Mg/day (1000 TPD)(2) PLANT
Item
Capital Required
Cost of Operation
AmortizationC4)
O&M Cost
Total Cost of Operation
Net Unit CostC5) Based
on Received Refuse
Net Unit CostC5) Based
on Steam Produced
Units
$
$/yr
$/yr
$/yr
$/Ton
refuse
$/Mg
refuse
$/1000 Ib
steam
$/Mg
steam
Drop
Charge (3)
NA
NA
0
5.0
10.0
0
5.5
11.0
0
5.0
10.0
0
5.5
11.0
Value
34,940,000
3,692,000
2,740,000
6,432,000
19.73
14.73
9.73
21.75
16.25
10.75
4.00
2.98
1.97
8.83
6.60
4.36
Notes: (1) Yearly value for steam is 727,800 Mg, or 1.604 x 10y Ib
(utiliz. factor = 0.9)
(2) Nominal value (900 Mg/d = 992 T/day).
(3) Drop charge is treated as a revenue and is in $/Mg or $/ton
raw refuse.
(4) Amortization @ 8-1/2% for 20 yrs.
(5) Net unit cost includes credit for drop charge.
For the present analysis, drop-charges of 0, $5.5 and $11 per Mg (0, $5,
and $10 per ton) are used. The effect of these values on the cost per unit of
product or cost per weight of entering refuse was then determined.
Fron the previously given heat balance, there are 7.85 GJ of energy con-
tent in the steam per Mg of refuse, amounting to 7.065 TJ (6.70 x 10^ Btu) per
day. Also, 2.462 Mg of steam are produced per Mg of refuse, resulting in
727,800 Mg (801,800 tons) of steam per year, using a utilization factor of
0.9.
162
-------
Table 28 shows net costs to produce steam for 900 Mg/d (992 TPD) Andco-
Torrax system. It is interesting to note that with no drop charge, the cost
to produce steam is $4.00/1,000 lb, which is not unreasonable in comparison
with the cost to produce steam by- other methods. With a $10 drop charge, the
cost to produce steam is quite competitive at $1.97/1,000 lb.
COSTS AND ECONOMICS AS A FUNCTION OF PLANT SIZE
Plant sizes used for calculation purposes are based on the number of gasi-
fier modules installed. With each module at 300 Mg/d capacity, cost calcula-
tions presented here include systems for:
One (1) module - 300 Mg/d (331 TPD)
Three (3) modules - 900 Mg/d (992 TPD)
Five (5) modules - 1,500 Mg/d (1,653 TPD)
Construction and operating costs for the 900 Mg/d (992 TPD) case are shown in
the previous section. Because of the modular construction, the cost of the
installed modular parts of the plant will vary linearly with the number of
modules. The cost of buildings, the receiving pit, and the crane will vary
approximately with the 0.6 power of the plant size. Cost of instrumentation/
controls, cooling tower, piping, pumps, plant air and electrical systems will
vary with approximately the 0.8 power of plant size. Land and site improve-
ment costs are given as a 0.8 power variation with plant size.
Using Table 24 as a base, Table 29 shows construction cost estimates
for alternative sizes of 1, 3, and 5 module size plants. Variation of the
total cost is found to be proportional to the 0.83 power of plant size. Capital
requirements for construction and startup are given in Table 30. Although
different exponents are used for several elements, all costs for the 300 Mg/d
and 1500 Mg/d cases were estimated by Parsons using the 900 Mg/d values as
a basis.
Operating costs (O&M) shown in Table 31 vary linearly except for labor,
where the number of personnel for each job description, as shown in Table 27
may not change. For those that change, only a minor increase or decrease
occurs. The variation in labor costs from that calculated for the 900 Mg/d
(992 TPD) was approximately 20% less for the 300 Mg/d (331 TPD) plant and 20%
more for the 1500 Mg/d (1,653 TPD) plant.
It can be seen in Table 31 that the unit costs drop considerably with in-
crease in plant size, graphically shown in Figure 31. With an increase in size
from 300 Mg/d (331 TPD) to 1500 Mg/d (1,653 TPD), unit costs reduce by about
50%, and become almost constant for larger sizes than 1500 Mg/d (1,653 TPD).
163
-------
TABLE 29. ESTIMATED CONSTRUCTION COST SUMMARY WITH PLANT
($000) 1976 DOLLARS
SIZE
Cost Element
Land
Site Improvements
Andco-Torrax System
Gasifiers
Regenerative Tower Sys.
Combustion Chambers
Boilers
Gas Cleanup (ESP's)
Instrumentation/Controls
Cool. Tower, Piping, Pumps
Plant Air
Other Items (crane,
stack, etc. }
Building
Subtotal
Contingencies (10%)
Engineer § Construction
Management^)
Total
Plant Size (Mg/d) ^
300
80
310
3560
incl
incl
incl
incl
incl
2,700
600
3,600
10,850
1,090
820
12,760
900
200
800
10 660
incl
incl
incl
incl
incl
6,500
1,000
7,000
26,160
2,620
1,820
30,600
1500
270
1,130
17 70Q
incl
incl
incl
incl
incl
9,800
1,500
9,500
38,100
3,800
2,900
44,800
(1) One module processes 300 Mg/d.
(2) Some engineering costs included in equipment costs.
TABLE 30. 1976 CAPITAL COSTS ($000)
Plant Size (Mg/d)
Cost Element
Construction
Interest during Construction
Startup Costs
Working Capital
Total
300
12,760
1,080
740
410
14,990
900
30,600
2,600
1,080
660
34,94Q
•-•- -—r— • t-n
1500
44,8,00
3,800
1,730
900
51,330
,' * ' _'— * • • " "
164
-------
TABLE 31. NET UNIT COST (1976 DOLLARS) TO PRODUCE
Item
Capital Required
Cost of Operation
Amortization^)
0 § M Costs
Total Cost of Operation
Steam Production Rate^J
Net
Unit
Cost
Drop Charge
0
$5.5/Mg
$11.0/Mg
Drop Charge
0
$5/T
$10/T
Units
$
$/y
$/y
$/y
Mg/h
(lb/h)
$/Mg
Steam
$/1000 Ib
Steam
Plant Size (Mg/d)
300
14,990,000
1,584,000
1,493,000
3,077,000
30.77
(67,800)
12.68
10.45
8.21
5.75
4.73
3.72
900
34,940,000
3,692,000
2,740,000
6,432,000
92.32
(203,500)
8.83
6.60
4.36
4.00
2.98
1.97
1500
51,230,000
5,413,000
3,778,000
9,191,000
153.9
(339,200)
7.58
5.35
3.11
3.45
2.43
1.42
(1) Net Unit Cost includes credit for Drop Charge.
(2) Amortization is 8-1/2% over 20 years.
(3) On yearly basis, use 0.9 utilization factor.
165
-------
AMORTIZATION 8-1/2%, 20 YEARS, UTILIZATION FACTOR
0.9; CREDITS, ALUMINUM S300/T, STEEL S40/T, DROP
CHARGE AS NOTED, 1976 DOLLARS
STEAM
REFUSE
Mfl
o
t/1
o
o
15 —
14-
13-
12-
11-
10-
9-
8-
7-
6-
5 .
4-
3-
2-
1-
0 —
1000
7-
5-
4-
3-
2-
DROPCHARGE
SO/T
SS/T, SS.SO/Mg
$10/T,$11/Mg
300
I
600
I
900
I
1200
I
1500
I
1800 Mg/d
$
Mg
-30
30-
-20
-10
10-
4-
400
800
1200
1600
2000 T/d
PLANT SIZE
Figure 31. Net unit cost to produce steam with
Andco-Torrax system.
166
-------
SECTION 8
PUROX PYROLYSIS SYSTEM
INTRODUCTION AND SUMMARY
The Purox system, developed by the Union Carbide Corporation (UCC), con-
verts solid wastes to a gas having an HHV of 11.81 to 15.36 MJ/Nm^ (300 to 390
Btu/SCF). The principal unit is a vertical reactor for an oxygen-blown par-
tial-oxidation pyrolysis process. A 181 Mg/d (200 TPD) demonstration plant
has been successfully tested at South Charleston, West Virginia, using the
municipal refuse collected in that area. Although first tested with raw
refuse, operation with shredded refuse was found to be superior.
Refuse is injected near the top of the reactor, slowly descends as a
moist feed, and a counter flow of hot gases then dries out the refuse and con-
verts it to gases, liquids, and char in the pyrolysis zone. The char is
burned with pure oxygen at the bottom (hearth) to produce an ascending mixture
of hot C02 and CO that causes pyrolysis of the cellulose matter. Melting of
inorganics results in a molten slag that runs off into quench water, forming
a black glassy aggregate. The synthetic gas (syngas) leaving, after gas clean-
up and cooling to remove oil, tar and ash, particulate matter, and moisture,
is essentially free of nitrogen and consists mainly of hydrogen, carbon monox-
ide, carbon dioxide, and hydrocarbons. This syngas can be used directly as
a fuel or can be converted to methane, methanol, ammonia, or a light hydro-
carbon fuel. The Purox system is at such a state of development in converting
municipal solid waste to a nitrogen free gas that it has now been offered with
guaranteed operation as a commercial unit.
Construction cost for a 1381 Mg/d (1500 TPD) base plant in 1975 dollars
was estimated at $53.75 million with capital requirements at $62.4 million.
This included use of one spare Purox reactor element. Annual operating costs
were estimated at $7.016 million. Construction costs were determined for two
other sizes, 635 Mg/d (700 TPD) and 1905 Mg/d (2100 TPD) at $26 million and
$74.6 million (using a spare reactor).
A net unit cost was determined for the base plant by using an 8-1/2%
amortization rate and taking credit for steel recovered at $40/ton and alumi-
num recovered at $300/ton and a drop-charge. With no drop charge, the syngas
would cost $3.30/million Btu; for a $5/ton drop charge, $2.53/million Btu; and
for $10/ton drop charge, $1.84/million Btu.
167
-------
Further processing of the syngas is required for subsequent uses. A
compressor and drying system is needed before transporting the gas to the
power plant at some distance, and a compressor is needed also to use the syn-
gas in a gas turbine for producing electric power. For synthesis of methane,
methanol, ammonia, or a light hydrocarbon oil from the syngas, a good deal of
treatment is required to prepare the gas for chemical conversion. Some of
these other gas utilization processes are discussed separately from the Purox
system in Appendix C.
The three sub-systems are shown schematically in Figure 32. Both the
front-end and gasifier are common to all plants, with the syngas being utilized
in one of several processes, with one being shown. The gas is cool, clean, and
dry and can be transported by pipeline to a customer (industrial heating or
utility) several miles away.
Details are presented in the following sections of this report for a
plant receiving 1588 Mg/d (1750 TPD) of municipal refuse over 6 days of each
week. To maintain the necessary continuous operation of the Purox reactors,
the equivalent of 6/7 x 1588 = 1361 Mg/d (1500 TPD) is used for process evalu-
ation. Shredding, air classification, and materials recovery and storage are
performed for 16 hours a day over 6 days, resulting in selection of process
equipment that will handle the entering refuse at a rate of 99.8 to 109 Mg/h
(110 to 120 TPH). Fuel gas is produced at the volumetric rate of 742 000
Nm3/d (27.71 x 106 SCFD) and a heat content rate of 10.8 TJ/d (10.25 x 109
Btu/dayJ. Daily quantities of ferrous metal and aluminum recovered are 111
and 5.4 Mg (122 and 6 tons) respectively.
The front end system is a separate unit and not part of the UCC-furnished
Purox gasifier equipment. This process can be varied in terms of specific
selection and arrangement of equipment. Presented is a Parsons system design
based on experience in commercializing such a unit and where there is avail-
ability of equipment. It is similar to the Chicago supplementary process
plant design reviewed in this report and differs by using one shredder per
line, with addition of an aluminum separator.
In addition to the detailed information on construction, operating, and
maintenance costs presented, an economic analysis by Parsons has been included
to show the net costs of producing a product. This includes amortization at
8-1/2% over 20 years, revenues from recovery of steel and aluminum, and a
range of refuse drop-charges from zero to $10/ton raw refuse. Readers can
develop a set of net costs by substituting their own detailed local values for
expenditures and potential revenues.
CONCLUSIONS
• Sufficient experience has been gained with operation of the South
Charleston facility to conclude that the Purox gasifier is technically
capable of being installed in any community desiring to convert wastes
to a medium heating value gas. Any front end processing equipment
deemed necessary for a given location can be considered as commercially
available.
168
-------
FRONT-END
MATERIALS HANDLING
GASiFIER
GAS UTILIZATION
PRODUCTS
ON
MUNICIPAL
SOLID WASTE •
OTHER WASTES
(SLUDGE)
(OILS)
(FARM&
FOREST)
FUEL
PREPARATION
RECEIVING
SHREDDING
MATERIALS RECOVERY
STORAGE
PUROX
SYSTEM
PARTIAL OXIDATION -PYROLYSIS REACTOR
OXYGEN GENERATOR
GAS CLEANUP
WASTE WATER TREATMENT
GAS
COMPRESSION
AND DRYING
1
f
INDUSTRIAL
OR UTILITY
FUEL GAS
Figure 32. PUROX gas generation and utilization.
-------
• The synthetic gas produced is best used as a fuel for a utility or
other customer having an existing furnace. Electric power can also
be made for sale in a gas turbine combined cycle system. The gas,
having a low nitrogen content because of the nature of the Purox pro-
cess, is technically able to be converted to a variety of compounds.
Of particular importance are methane, methanol, and ammonia. Critical
steps in the synthesis of these materials from the gas must be experi-
mentally investigated before accurate costs can be established.
• Product values should escalate in a similar manner as capital and
operating costs, and it is concluded that current economics presented
here are valid for projected comparison purpose.
• Below about 907 Mg/d (1000 TPD), the unit cost of syngas rises sharply
and there is not much further economics of scale beyond 1361 Mg/d
(1500 TPD).
• A variety of cellulosic or hydrocarbon wastes can be introduced along
with MSW to the gasifier. Sewage sludge dewatered to 25% solids can
be added to at least 10 percent of the weight of MSW.
• Similar emissions, except for 502, Per heating value should be pro-
duced from burning the syngas in utility boilers as compared to natu-
ral gas. The sulfur in the syngas is low, but where necessary can be
removed rather simply.
• With an organic content of approximately 50,000 BOD, condensate from
the Purox gas will typically require treatment before introduction
into sewer systems.
FRONT-END PROCESSING DESCRIPTION
Preparing raw municipal solid waste requires shredding to particle sizes
from 10 to 20 cm (4 to 8 in.). All organic matter is sent to the gasifier.
The process involves recovery of salable materials, with ferrous metals and
aluminum chosen as the most marketable. The capital cost is greater than UCC
has previously shown because of the aluminum recovery subsystem, which requires
an air-classifier. Also, extensive air handling equipment is used to reduce
dust and discharge relatively clean air to the surroundings. Although there
are similarities to the Chicago supplementary fuel process plant, there are
sufficient differences to describe this in relative detail. Aluminum separa-
tion, and small modularized storage bins are used; no secondary shredder is
necessary.
For the purposes of design, the refuse received is assumed to have a
composition as that presented by the EPA in Ref. 15. Table 32 shows the
composition as given by EPA as well as Parsons' estimated values of HHV and
moisture for each component. The average HHV is 10.77 MJ/kg (4,630 Btu/lb)
and the moisture content 25.8%. This composition is a national average and
representative for many major cities. Compositions measured in a specific
community and somewhat different than those used in Table 32 can be substituted
and subsequent dependent process variables and costs easily changed in the
170
-------
TABLE 32. MUNICIPAL SOLID WASTE COMPOSITION, 1975
Component
Paper
Glass
Metals (Total
Ferrous
Aluminum
Other Non-Ferrous (ONF)
Plastic
Rubber $ Leather
Textiles
Wood
Food Waste
Yard Waste
Miscellaneous inorganic
Weighted Average
Wt. %(!)
39-7
9.8
9.6
(8.7)
(0.6)
(0.3)
4.1
2.7
1.6
3.6
13.3
14.1
1.5
100.0
% Moisture (2)
24.3
3.0
6.6
-
-
-
13.8
13.8
23.8
15.4
63.6
37.9
3.0
25.8
Heating Value (2)
MJ/kg
12.84
0.19
1.65
-
-
-
33.10
19.65
15.51
16.51
10.28
9.37
.4.77
10.77
Btu/lb
5,520
82
709
-
-
-
14,230
8,450
6,670
7,100
4,420
4,030
2,050
4,630
(1) From EPA report SW-161, modified by Parsons due to error in report
table.
(2) Value from measurements on like components by Parsons.
several other sections of this report. UCC did not determine the actual com-
position of refuse at South Charleston. The sequence of operations is pre-
sented by a flow diagram in Figure 33, with Table 33 presenting flow rates
for the numbered streams. First, municipal solid waste is delivered to a
tipping floor where a front end loader pushes or places the waste onto one of
two sunken conveyors. Two parallel lines of equipment are used, each with a
capacity of 54 Mg/h (60 TPH), or a total of 109 Mg/h (120 TPH). The Purox
171
-------
OIL
- BUILDING STEAM
- WATER SLOWDOWN
AIR EXHAUSTS
180,000 cfm rri 135,000 cfm
SUBSTATION
BAGHOUSES
;
AIR, DUST
TIPPING FLOOR
CYCLONE
HEAVY ORGANICS,
GLASS, NONFERROUS "
Figure 33. Flow diagram of Purox front end processing system.
172
-------
TABLE 33. COMPOSITION OF MATERIALS IN FRONT-END UNIT STREAMS
OF FIGURE 33. Mg/d (TONS/DAY]
Flowstream
Figure 33
Component
Organics
Ferrous Metals
Aluminum
Glass
Other Metals
Rock, Dirt, etc.
(Moisture) ^ '
TOTALS
HHV, MJ/kg
HHV, Btu/lb
1
CD
Receiving
1060 (1,168)
122 (135)
8 (9)
131 (144)
4.5 (5)
35 (39)
(340) ((375))
1361 (1,500)
10.77
4,630
2
Shredded
Raw Refuse
1060 (1,168)
122 (135)
8 (9)
131 (144)
4.5 (5)
35 (39)
(340) ((375))
1361 (1,500)
10.77
4,630
3*
Ferrous
Recovery
4.5 (5)
111 (122)
-
0.9 (1)
-
-
(4.5) ((5))
116 (128)
0.51
221
4*
Feed to
Surge
1055 (1,163)
12 (13)
2.7 ' (3)
130 (143)
4.5 (5)
35 (39)
(336) ((370))
1239 (1,366)
11.77
5,062
5*
Pur ox
Feed
1055 (1,163)
12 (13)
2.7 (3)
130 (143)
4.5 (5)
35 (39)
(336) ((370))
1239 (1,366)
11.77
5,062
6*
Aluminum
Recovery
-
-
5.4 (6)
-
-
-
-
5.4 (6)
-
-
(1) See Table 32.
(2) Moisture is part of the components shown.
* Items in 3, 4, 5 and 6 are estimations. Aluminum recovery percentage from Combustion Power Co.,
Menlo Park, California.
-------
System is to be operated over seven days with the equivalent of 1361 Mg/d
(1,500 TPD) of delivered raw refuse. The front end plant is designed for
receiving refuse over 6 days per week or 1588 Mg/d (1,750 TPD). The process
equipment depicted in Figure 33 is to be operated over 16 hours or two shifts
in each of 6 days/week, requiring an operational capacity of 100 Mg/h (110 TPH)
or a design capacity of 109 Mg/h (120 TPH), 54 Mg/h (60 TPH) for each process
line.
The raw refuse is conveyed to the shredders, which reduce the feed to
less than 20 cm (less than 8 in.) sized particles. The discharge is then
conveyed to an air classifier that is tuned (air velocity adjustments) to drop
not only the steel cans but as much as 80% of the aluminum cans in the heavy
fraction. The air classifier is used so that aluminum can be recovered.
Otherwise, steel can be recovered directly from the shredder discharge with a
magnetic belt. After recovery of the steel in the heavy fraction with a mag-
netic separator, the remaining material is passed through a trommel where a
fraction in the size range of 10 to 1.6 cm (4 to 5/8 in.) is separated, con-
taining primarily aluminum cans. This fraction is passed through an eddy-
current aluminum can recovery unit. It was estimated that 67% of all the alu-
minum delivered to the plant is recovered. Table 33 shows the amounts of
materials in each of the main streams in Figure 33. Some 5.4 Mg (6 tons) of
aluminum and 111 Mg (122 tons) of steel are recovered daily. Revenues from
aluminum recovered must more than pay for the capital and operating costs of
the air classifier, aluminum separator, and associated equipment. The remain-
ing material from the aluminum separator and trommel consisting of glass,
other non-ferrous metals, and heavy organics are conveyed to surge bins along
with the light fraction from the air-classifier.
The light fraction from the air classifier is separated from air in a
cyclone. The air is discharged to the atmosphere after passing through a
baghouse at the rate of 63.7 m3/s (135,000 CFM). Using a separate fan, air is
collected from various dust hoods in the process building, and passed through
a baghouse at the rate of 84.9 m3/s (180,000 CFM).
Actual feed to the Purox system from the surge bins is 1239 Mg/d (1,366
TPD) with an HHV of 11.77 MJ/kg (5,062 Btu/lb). Experience has demonstrated
that large piles of moist shredded refuse in a bin left for several days may
pack sufficiently to make retrieval with automatic equipment very difficult.
Providing surge capacity for one or two days in relatively small amounts in
multiple bins should present no retrieval problems. Each Purox gasifier module
requires a nominal 317 Mg/d (350 TPD) feed. One surge bin is provided for each
gasifier module and must provide a feed during the one day when no delivery of
refuse is made to the plant, as well as the period in the third shift when the
process equipment is not in operation. This amounts to 32 hours of storage.
The bin has been designed for approximately 36 hours of storage or 476 Mg
(525 tons). Tradeoff analysis for specific cases may lead to increased storage
capacity depending on refuse delivery patterns and gas delivery requirements.
An oil-fired steam generator is used to heat the building and minor pro-
cessing fluids.
174
-------
The major pieces of equipment and their purpose are as follows:
• Weighing scales to measure delivery weights of refuse (not shown)
• Receiving or tipping floor where packer trucks drop their raw-refuse
(municipal solid waste)
• Raw-refuse feed-conveyor to shredder
• Primary shredder that reduces raw waste to particle sizes less than 20
cm (8 in.)
• Air classifier to separate metals from the bulk of paper material as
part of a heavy fraction for ease in recovery of aluminum
• Ferrous metals magnetic separator to recover steel cans and/or iron-
based metals
• Rotary trommel to screen the air-classifier heavy fraction to provide
high concentration of aluminum can material and reject -1.6 cm (-5/8
in.) (mainly glass) and +10 cm (+4 in.) (mainly wood, cardboard, etc.)
• Eddy current aluminum separator unit to recover a relatively pure
aluminum can fraction
• Conveyor of aluminum recovery unit tailings to surge bin
• Main air classifier fan to pass air and light shredded fraction to
cyclone
• Cyclone for de-entraining the light shredded fraction from the air
classifier air stream
• Dust hoods in receiving and process building, and shredder and conveyor
covers maintaining clean air conditions
• Dust hood fan for collection of dust and maintaining negative pressure
in building
• Baghouses to filter dust from air-classifier air and dust hood air
before discharge to atmosphere
• Surge bins for interim storage of all shredded refuse except for re-
covered ferrous metals and aluminum
• Ferrous metal and aluminum storage bins with bottom dump to truck
loading
• Electric power substation for conditioning electric power from utility
lines for use in front-end plant as well as gasifier system and, where
necessary, gas utilization units
175
-------
• Steam generator for heating buildings and process streams where needed
The energy flow in the incoming refuse is 1500 x 2000 x 4630 = 13.89 x
109 Btu/day (14.65 TJ/d) and the energy in the processed refuse feed to the
Purox System is 1366 x 2000 x 5062 = 13.82 x 1(P Btu/day (14.58 TJ/d). Prac-
tically all the organic matter is used, and only that which is trapped with
the steel and aluminum is lost.
PUROX GASIFIER SYSTEM
The Purox gasifier is a vertical shaft furnace or reactor for the purpose
of converting solid waste into a fuel gas by pyrolysis. This synthetic gas
(syngas) consists primarily of hydrogen and carbon oxides with some hydrocar-
bons . There are three general zones of reaction in the vessel, consisting of
drying, pyrolysis, and combustion/slagging, as shown in Figure 34.
The Purox gasifier concept was tested over a period of three years in a
4.5 Mg/d (5 TPD) pilot unit in the UCC laboratories at Tarrytown, New York.
Encouraging test results with the pilot unit prompted UCC to build and operate
a demonstration plant with a nominal capacity of 181 Mg/d (200 TPD) in South
Charleston, West Virginia. That reactor is approximately 3 m (10 ft) ID and
9.1 m (30 ft) high, and has been operating since May, 1974. An extended test
run for 3 months, 7 days per week, showed that the demonstration reactor
operated well with shredded municipal refuse at full and part loading. Gases
were found to have a greater volumetric heating value than was found at Tarry-
town due to more hydrocarbons being generated. Shredding and magnetic separa-
tion of iron are all that is needed in a front end system. It is desirable to
keep at least half the usual amount of glass in the reactor feed because of
the need to reduce the melting point and viscosity of the molten slag formed
in the char combustion zone. In addition, glass encapsulates metals in the
residue.
The moist shredded refuse, with magnetic metals and some aluminum, enters
at the top and slowly descends. A counterflow of hot gases, starting in the
combustion zone at the bottom, dries the refuse, which then decomposes into
synthetic gas, char, and organic liquids in the high temperature pyrolysis
zone. Melting of inorganics and combustion of the char occurs in the hearth
zone of the reactor, producing an ascending gas mixture of CO and C02- Pyrol-
isis-formed oil and char particles carried upward by the hot gases are mostly
scrubbed out by the descending refuse and are thus internally recycled, with
the oil cracking to gases and char. Relatively pure oxygen, from a cryogenic
oxygen producer, is passed into the hearth to burn all the char. Molten slag
is quenched to form a black glassy granular aggregate.
The heat of combustion of the char is sufficient to maintain a temperature
of 1649°C (3,000°F) in the hearth for melting oxides, glass, metals, and other
non-combustibles, and to provide the heat for pyrolysis reactions, heat-of-
vaporization of water in the entering refuse, and heat losses from the vessel.
The syngas produced is cooled to approximately 92-204°C (200 to 400°F) in
drying the refuse.
176
-------
SHREDDED RAW
REFUSE FEED
OXYGEN
HOTCO,C02,|H2,
AND HYDROCARBONS
HOTCO,C02
MOIST SYNTHETIC
FUEL GAS (SYNGAS)
>• DRYING ZONE
^.PYROLYSISZONE
COMBUSTION/SLAGGING ZONE
SLAG
Figure 34. Schematic of Purox reactor.
177
-------
An operational system for a community can use a 181 Mg/d (200 TPD) reac-
tor module or a scaled-up 317 Mg/d (350 TPD) size. Actually, a module can
operate over a wide feed-rate range. Use of the larger-sized module reduces
the number of reactors required in processing refuse at delivery rates in the
907 to 1814 Mg/d (1000 to 2000 TPD) range.
Also associated with each reactor is gas clean-up equipment. Changes
have been made from the original arrangement as a result of operational ex-
perience at South Charleston. A train of equipment associated with each reac-
tor is shown in Figure 35, except for the oxygen plant, water treatment unit,
and cooling tower, which are single units to serve all the modules. The water
treatment unit is discussed separately.
For a 1361 Mg/d (1500 TPD) plant, four plus one spare 317 Mg/d (350 TPD)
modules are used, as shown in the Facilities Section. It is Parsons judgment
that a spare can be eliminated if operational experience can show a 0.9 utili-
zation factor for a set of four modules.
The major pieces of equipment and their purposes are as follows: (See
Figure 35)
• Inclined feed conveyor and leveler.
• Ram-feeder for injecting the mixed refuse material
• Reactor to convert the refuse to syngas and slag
• Water quench tank to solidify and shatter the slag into a frit
• Drag conveyor to transport the slag to a vehicle for hauling to disposal
• Water spray scrubber, with knock-out tank, to cool and remove char,
tar, and ash from reactor off-gas
• Cooler for recycled scrubber water
• Wet electrostatic precipitator to remove fly ash, char, oil, and tar
mist from the cooled and scrubbed reactor off-gas
• Solid-liquid separator system to separate oil and water from the char,
oil, tar, and ash mixture to the reactor; recycle the scrubber water;
and send excess water to waste water treatment
• Condenser to further reduce moisture content
• Combustor flare for emergency disposal of the synthetic gas
• UNOX or equivalent waste water treatment to condition water for accep-
tance in local sewage systems (required only if a Sanitation District
does not allow high BOD discharge into sewage lines)
178
-------
OXYGEN PLANT
COOLING
TOWER
L_
x^n_fc
SLOWDOWN
TO SEWER
24) 200 mg/l BOOg
FROM SEWER
200 mg/l BODg
. HOT WATER
MAKEUP
COOL WATER
TO PROCESS
'18") EQUIPMENT
GAS FROM
PUROX SYSTEM
PILOT FUEL
Figure 35. Train of equipment for Purox gasifier system.
-------
Quantities associated with the numbered streams connecting these pieces
of equipment are given in Table 34.
TABLE 34. PUROX GASIFIER STREAMS*
Stream
No.
5
7
8
9
10
11
12
13
14
15
16
17
Stream
Purox Feed
Purox Reactors
Off-gas
02 to Reactors
Purox Gas from
Condensers
Char, oil, ash
recycle
Slag
Air to 02 Plant
Oxygen from Air
Separation Plant
Nitrogen from
Air Separation
Plant
Electric Power -
Air Compressor
Cooling Water
Tower
Cooling Water
Tower Makeup
Mg/d
1239
862
382
32
27
254
0.75
(28.
32
222
1451
290
1161
Quantity
(TPD)
(1366)
(950) Fuel Gas
(421) Water Vapor
(35) Char, Oil, Ash
Particles
(30) Water Soluble
Organics
(280)
x 106 Nm3/d
03 x 106 scf/d)
(35)
(245)
(1600)
(320)
(1280)
4060 Kw
0.85
0.02
m3/s (13,500 GPM)
m^/s (320 GPM)
Other Information
See Table 33 for
Composition
Total Reactor Off-Gas
1303 Mg/d (1436 TPD)
Includes 11 Mg/d
(12 TPD) N2 + Ar
Saturated with water
vapor at 38°C (100°F)
Mostly char
Includes oxidized
metals and oxides
inherent in paper,
wood, etc.
7/8 for Purox
1/8 for Unox
Gas discharge
For oxygen plant
* Values shown are based on information provided by Union Carbide.
180
continued
-------
TABLE 34 (continued)
Stream
No.
18
19
20
21
22
23
24
25
Stream
Cooling Water
to Process
Cooling Tower
Slowdown
Air Exhaust/
Cooling Tower
Waste Water
Oxygen to Unox
unit
Dilution Water
from Sewer
Discharge to
Sewer
Recycle Spray
Water
Quantity
Mg/d (TPD)
0.85 m3/s (13,500 GPM)
0.003 m3/s (53 GPM)
0.06 x 106 Nm3
(2.2 x 106 CFM)
0.004 m3/s (70 GPM)
27 Mg/d (30 TPD) organics
36 (40)
0.092m3/s (1460 GPM)
0.096 m3/s (1530 GPM)
0.013 m3/s (200 GPM)
Other Information
COD 40,000 to
60,000 mg/dm3
BODs = 200 mg/dm3
BODs = 200 mg/dm3
To off-gas scrubber
Gases leaving the reactor contain approximately 40% moisture by volume
at a temperature of 93 to 204°C (200 to 400°F), with a higher heating value of
the dry gas of approximately 14.57 MJ/Nm3 (370 Btu/scf). The process conver-
sion efficiency is estimated at 70% to 80%, depending on the moisture and
materials content of the refuse. This efficiency does not account for the
energy required for the oxygen plant, pumps, etc. An energy balance and over-
all efficiency showing the fuel energy needed for the electric power is pre-
sented later. Approximately 0.2 ton of C>2 per ton of solid waste is required
for the gas generation process, plus an additional 0.027 ton of oxygen per
ton of refuse for the waste water treatment process.
For 0.90 Mg (1 ton) of feed, 503.1 m3 (18,780 SCF) of 14.57 MJ/Nm3 (370
Btu/SCF) syngas are produced, for a total of 7.33 GJ (6.95 x 106 Btu) of ener-
gy in the syngas. It is of interest to note that over a wide range of refuse
composition variations, the heating value of the gas and the quantity of gas
produced per ton of refuse are inversely proportional. This means that the
total heating value of the gas stream per unit weight of feed material remains
relatively constant. Each 317 Mg/d (350 TPD) reactor and its gas cleanup
equipment will produce 176 x 103 Nm3/d (6.57 million dry SCF/day) of synthetic
gas, with an energy flow of 2.56 TJ/d (2.43 x 109 Btu/day). The composition
of gas leaving the system, shown in Table 35, was determined from tests by
Union Carbide using municipal solid waste, at approximately 25% moisture, from
South Charleston.
181
-------
TABLE 35. TYPICAL DRY PUROX SYNTHETIC GAS COMPOSITION CD
Gas Constituent
H2
CO
C02
CH4
r1 u r 14 *• J
C2H2, C2H4
N_ + Ar
% by Volume
24.0
40.0
24.0
5.6
5,4
1.0
100,0
(1) Gas actually leaves saturated with moisture at 38°C
(100°F).
(2) C2H2 0.7, C2H4 2.1, C2H6 0.3, C3H6 0.2, C4+ 1.6,
oxygenated HC's 0.2.
The feed material can be shredded refuse with only iron removed, or the
shredded combustible fraction following air classification and removal of
other revenue-producing materials. A few percent increase of inorganic feed
changes the off-gas composition insignificantly according to tests run by
Union Carbide at their Tarrytown pilot unit. Parsons estimate that the addi-
tion of biodigested sewage sludge will add a solids content that is approxi-
mately 50% organics and 50% ash with a higher heat content of 10.70 MJ/kg
(4600 Btu/lb). Up to 15% by weight of sludge has been acceptable for design.
Undigested sludge solids contain only 25% to 37% ash and have a higher heat
content of 16.28 MJ/kg (7,000 Btu/lb) as measured by Parsons. Digested sludge
dewatered to 75% moisture will increase the moisture content of the final
sludge-solid waste mix by 6% to 7%. The mixed feed will then contain approxi-
mately 32% to 36% moisture; up to 45% moisture of mixed feed is technically
acceptable. Increased moisture requires an increased amount of oxygen in the
combustion zone of the reactor to provide additional heat for drying and main-
tenance of proper temperatures throughout the reactor. A corresponding in-
crease in C02 occurs. Approximately 0.05 to 0.07 ton of oxygen is required
per ton of refuse to provide the heat necessary to evaporate the moisture in
refuse with a moisture content of 25% to 32%.
The total heat generated in the combustion zone can be accounted for by
the following items requiring heat:
(1) Vaporization of moisture in the waste feed.
(2) Pyrolysis reactions.
(3) Heating and melting of the inorganic residue
182
-------
(4) Heat losses from reactor vessel.
(5) Sensible heat in the off-gas.
In tests at South Charleston, contaminants were measured in the syngas
after passing through the scrubber, electrostatic precipitator, and condenser.
The NOX content was not measurable, being less than 1 ppm. The HC1 was mea-
sured at less than 1 ppm, having been absorbed by the scrubber water. H2S was
measured at levels between 300 and 600 ppm. One ppm or less of carbonyl sulfide
(COS) was detected. A solid particulates level was measured at 0.159 g/Nm3
[0.067 gr/SCF). Contaminants measured are shown in Table 36.
TABLE 36. CONTAMINANTS IN SYNGAS FROM PUROX SYSTEM
AFTER GAS CLEANUpa
Component
Fly Ash
Sulfur as
Waterc
HC1
NO
Amount
125 ppm (0.159 g/Nm^, or 0.067 gr/SCF)
300 to 600 ppm
6% by volume
1 ppm
1 ppm
aAfter the condenser; value will drop after glycol dryer.
Measurement of COS showed 1 ppm or less.
CWater is contaminant if the gas is piped a considerable
distance
This gas requires additional treating before methanation, methanol syn-
thesis, or ammonia synthesis in catalytic converter systems. If the Purox
gas is to be used directly in a power plant, but needs to be transported in a
pipeline, even if only several hundred feet, much of the moisture must be re-
moved, as described later.
Molten slag is collected in a pool at the bottom of the hearth. At the
opening of the hearth to the quench tank, a small weir allows a pool to form
that protects the hearth. After spilling over the weir, the slag falls through
a slag pipe into quench water. A drag conveyor moves the shattered glassy
slag aggregate up an incline to a conveyor and then to a truck for removal to
landfill or storage. The slag has potential for such uses as roadbed fill,
concrete block manufacture, or asphalt aggregate.
The amount of slag will vary depending on the extent of inorganic solids
separation at the raw refuse shredding and separation plant. For the case of
a feed with the ferrous fraction removed, the amount of slag generated is
183
-------
approximately 17% of the feed weight. The volume reduction of raw refuse to
this residue of glassy aggregate is about 98%.
Approximately 0.26 m3/min. [70 gpm) of waste water with a BOD of 40,000
to 60,000 mg/dm3 is purged from the water/solids separator, equaling the amount
of condensate formed from moisture in the Purox reactor offgas. This can be
sent directly to a sewage line if allowable by the particular community sanitary
district. If not, a waste water treatment unit will be needed to reduce the
organic content to approximately 200 mg/dm3. Included in Figure 35 is a Unox
unit to be furnished along with the Purox gasifier system. Oxygen from the
air separation unit is used at the rate of 36 Mg/d (40 TPD) to sustain aerobic
digestion of the organics. For biodigestion to proceed satisfactorily, the
purge stream is diluted 20 to 1 with sewage water from the nearest trunk line.
Discharge from the digester back to the sewage line is at the rate of 5.79
m3/min. (1530 gal/min) with a 200 mg/dm BOD. This water treatment unit is
included in the schedule of equipment. Alternative water treatment methods
are briefly discussed in the section dealing with environmental assessment.
The Unox waste water treatment system consists of covered multi-stage bio-
digestion reactors in which oxygen is fed co-currently with the waste water.
Pumping costs for the sewage and installation of a 7948 nr/d (2.1 MGD) treat-
ment unit adds considerable expense to the plant as well as requiring an addi-
tional 10 110 m2 (2.5 acres) of land. Alternative methods for treating the
relatively small waste water stream have been reviewed by Parsons such as re-
duced pressure stripping, carbon adsorption, or high pressure wet oxidation.
A careful assessment is required of these alternative waste water treatment
systems to ascertain their costs and performance. For this report, the Unox
system is used because of its widespread acceptance and availability.
An energy balance, on a daily basis, for the front end plus the Purox
system shows:
Purox Gas Energy Flow 10.94 TJ (10.37 x 109 Btu)
Raw Refuse Energy Flow 14.65 TJ (13.89 x 109 Btu)
Electric Power* 2.74 TJ (2.60 x 109 Bt-u)
Fuel Oil & Gasoline for 0.26 TJ (0.25 x 109 Btu)
Operation
The net thermal efficiency of the Front End (FE) plus Purox system (Px)
is:
,t (FE . Px) . (10.37 -2.60 -0.25)
* Calculated at 10,000 Btu/kWh.
184
-------
The conversion efficiency of energy in product from energy in raw refuse
is:
TJC (FE + Px) = (y§7|~) 100 = 74.6%
PUROX SYNGAS USE OR CONVERSION
The nitrogen free syngas produced in the Purox process can be converted
to methane, methanol, ammonia, or used as a fuel in the generation of heat or
electric power. Neither of the two other gas-producing partial oxidation
pyrolysis systems that have reached demonstration or semi-commercial status
using municipal solid waste feedstock (Andco-Torrax and Landgard, which are
air blown) produced a gas suitable for conversion to NH3, CH30H, or CH4- There
are some other concepts that have been tested in small or bench scale units to
produce gases, but none of these appear technically or economically feasible at
present. This section describes primarily the preparation of Purox gas for
use as a fuel gas.
The simplest and most cost effective utilization method for the product
is to construct the plant near a large fuel gas user, such as a utility power
plant, and pipe the gas directly to the furnace as a primary or supplementary
fuel. Other uses of the syngas are presented in detail in Appendix C. Five
means of utilization are defined:
(1) Syngas For Delivery to Customers: Setting up a compressor station,
dryer, and pipeline not included in equipment to supply syngas to a
power plant.
(2) Conversion to Methane: Upgrading the syngas catalytically to pipe-
line quality for introduction into an existing gas utility pipeline
distribution system.
(3) Generation, of Electric Power: Using the syngas as a fuel in a con-
ventional gas-turbine combined-cycle system to produce electric
power.
(4) Conversion to Methanol: Converting the syngas catalytically, where-
in a product is derived that has value as a storable fuel for use
in peaking power gas turbines, as an example.
(5) Conversion to Ammonia: Converting the syngas to pure hydrogen, and,
with purified nitrogen from the oxygen plant, the mixture is trans-
formed catalytically to ammonia.
To transport the gas by pipeline to a customer requires dehydration to
prevent ice formation in the winter or liquid condensation during other times
of the year. The most important factor is the proximity of a power plant or
an industrial plant that has a long term demand for the syngas.
A flow diagram of the gas drier unit is shown in Figure 36. The gas is
compressed to 446 kPa (50 psig) and first passed through an air cooler where
185
-------
PURO
PRODUCT
GAS COMPOSITION
MOLE% CONTAMINANTS
H2 24 H2S 400 ppm
CO 40 HCI 1 PPm
C02 24 PARTIC 15 ppm
CH4 5.6 NOX 1 ppm
C2H2 0-7 OIL 100 ppm
C2H4 2.1
c2He o-3
C3H6 0-3
P3H8 0.2
C4 + 1.6
OX HC's 0.2
N? + A 1.0
100.0
AIR COOLER
XGAS @
WATER VAPOR
SO 2
\ i®
\ T
^ DRY SYNGAS L ©
1
INCINERATOR
WATER
I 1 VAPOR
| 4,000 ppmv
H2S
GLYCOL STRIPPER
ABSORBER STACK
T
XXXV f .
- , /- _l '
A ( REBO.LER | ^_
»l 1 1 _L/~ I 1 "-1 1
* rt *
H 1 | XHLJ (VN/N/^*-,
COMPRESSOR | DRYGLYCOL |— 7-^k
^ U c
T®
WASTEWATER
$ ^> H J
^ /^-A
WET GLYCOL
PIPELINE
»• TO UTILITY
BOILER
FUEL FOR REBOILER
Figure 36. Gas compression and drying for piping to utility power plant.
186
-------
approximately two-thirds of the water is condensed, collected, and pumped to
the Unox water treatment unit. The gas is then scrubbed with triethylene
glycol to further dehydrate the Purox gas. In winter, the dew point of the
gas is reduced to -26°C (-15°F) and in the summer, 7°C (45°F). Some of the
Purox gas, 8.8 x 103 Nm3/d (0.33 x 106 SCFD), is used as fuel in the reboiler-
stripper element of the dryer. The flow rate of dry gas is 744 x 103 Nm3/d
(27.71 x 106 SCFD), having an energy content of 10.82 TJ (10.25 x 109 Btu).
In the process of absorbing moisture, some of the H2S will be absorbed
by the glycol and then released in the stripper with the water vapor. A small
burner will be required to convert the H2S to S02. Emissions to the atmosphere
will be 6.8 kg/h (15 Ib/hour) of S02 and 454 kg/h (1000 Ib/hour) of moisture
for the plant. If this emission rate is not acceptable, a small caustic
scrubber can be used to remove the H2S from the stripper emissions.
Power required for the compressor is 3750 kW based on a 446 kPa (50 psig)
discharge pressure. This pressure is minimal for an 8 to 10 km (5 to 6 mile)
pipeline of approximately 20-inch diameter to a utility boiler. A higher
pressure would reduce the pipe size, but increases the power required. A cost
tradeoff analysis is required to determine the optimum pressure and pipe size
for the particular application chosen. A pipeline to the customer is not in-
cluded in costs of construction shown later.
The gas to be delivered to a utility boiler will contain approximately
200 to 400 ppm H2S. A recent study associated with a power utility showed
that this would produce 60 to 75% less S02 emissions than the corresponding
fuel oil being replaced. For certain industrial purposes, where H2S cannot be
tolerated in the fuel gas, a simple Stretford scrubber system would be added
to remove the sulfur and put it in solid form, which can be stored and sold
when a sufficient amount accumulates.
Table 37 shows quantities and characteristics for several of the streams
in Figure 36. Stream 7 is passed into a pipeline for transport to a utility
or industrial furnaces.
FACILITIES
Descriptions are presented here for facilities recommended by Parsons to
convert solid waste to syngas for use in a utility power plant or for chemical
synthesis. The process for recovering resources from solid waste is described
in the flow diagram, Figure 33. Figure 37, Solid Waste to Syngas Conversion
Facility - Perspective View, shows an architectural perspective of a facility.
Whether the gas is piped as a fuel to a utility or to a chemical conversion
plant is not of concern in this section; the user of the gas is presumed to
receive the gas at the plant property line in either case. Included in the
facility are ferrous metal and aluminum recovery systems. Equipment noted is
for purposes of typical design and cost estimation. Similar equipment is
available from several manufacturers and final selection would be based on
competitive bidding to a set of specifications.
187
-------
TABLE 37. PUROX GAS COMPRESSION AND DRYING*
Stream
No.
Stream
Quantity
Other
Information
Gas from Purox System
Compressor Power
Compressor Discharge
Condensate from
Air Cooler
Fuel for Reboiler
Gas to Glycol
Absorber
Gas to Pipeline
Stripper Emissions
753 x 103 Nm3/d
(28.03 x 106 dscf/d)
3400 kW
Same as 1, above, at
446 kPa (50 psig)
19.3 dm3/min.
(5.1 gpm)
8.8 x 103Nm3/d
(0.33 x 106 SCF/d)
757 x 103Nm3/d
(28.2 x 106 SCF/d)
744 x 103Nm3/d
(27.71 x 106 SCF/d)
454 kg/h (1000 Ib/hr)
water vapor and 6.8
kg/h (15 Ib/hr) S02
500 ppm K2S
10.94 TJ/d
(10.37 x 109 Btu/d)
Varies with air
temperature.
400 ppm H2S
10.82 TJ/d
(10.25 x 109 Btu/d)
H2S is burned to
SO 2
* Based on calculations at Parsons
Capacity Analysis
The referenced flow diagram was developed showing material quantities
based on receipt of 1361 Mg/d (1,500 TPD) of solid waste into the system. It
is dictated by several pertinent factors listed below, but not included are
those related to the particular customer and the refuse character and amounts
for a specific community over several seasons.
The syngas generators for the process are constructed in 317 Mg/d (350
TPD) capacity modules. The expected performance of these modules is such that
four of them would be able to convert nominally 1270 Mg/d (1,400 TPD) of pro-
cessed feed material from the raw refuse. The nature of the syngas generators
is such that for best efficiency and least unit operating cost, they should
operate continuously at design capacity. Solid waste collection is conducted
on a 6-day-per week basis for this report and the front-end processing plant
design and costs are predicted on operating two 8-hour shifts per day, 6 days
per week, with the third shift and Sundays being utilized for maintenance
operations. The required 1588 Mg/d (1,750 TPD) of raw refuse processing 6
days per week can be accomplished by two parallel processing lines with a
188
-------
00
Figure 37. Perspective view of solid-waste-to-syngas conversion facility.
-------
nominal capacity of 54.4 Mg/h (60 TPH) each. This will permit the installation
of shredders at about the smallest size that provide good reliability and that
can be coupled with other material handling equipment of a class that will
function with a minimum of problems.
It is assumed that in the course of annual operations of the front-end
processing plant, 3 weeks of full production will be lost because of scheduled
maintenance requiring more than one shift, or to unscheduled downtime for un-
foreseen causes. In addition, because the syngas plant has 5 modules with only
4 operating at a time, it is assumed that only one additional week will be
lost due to unforeseen causes, for a total of 4 per year. This yields a plant
utilization factor of 92%.
Site Characteristics
Figure 38 shows a recommended site plan for the processing plant. The
dimensions shown, with a minimum of 30 m (100 ft) clearance all around, plus
room for expansion, indicate that a plot of 347 by 250 m (1,140 ft by 820 ft),
or approximately 87 000 m^ (21.5 acres), should be procured. The site must
be accessible by a road that is capable of handling heavy truck traffic. The
road should also be on, or shortly connect with, arterial road(s) that pro-
vide access of solid waste collection vehicles from all directions.
The site must be provided with electric power that has capacity to 20,000
kV-A, 60 Hz, 3 Phase, with a minimum of 4160 volts. Water and sewer capabili-
ties are needed for supplying and draining the fire protection system at a
maximum of 3.78 m^/min. (1000 gpm) in emergencies. Normal plant usage would
not exceed 2 271 m3/d (600,000 gpd) for cooling tower makeup, process steam,
housekeeping, sanitation, and some dust control. In addition, there must be
access to a sewage trunk line that carries a minimum of 5.29 m-Vmin. (1400
gpm) during the daily cycle.
Front-End Processing Plant
Front-End Processing Plant Floor Plan and Equipment Layout--
Figure 39 shows the front-end building plan with the process equipment
layout. The plant is composed of three major areas under one roof, and a
fourth outside the building. Under roof are the tipping floor, the processing
area, and the support services area that contains administrative offices,
locker and lunch rooms, wash rooms, and maintenance shop.
The tipping floor receives the solid waste from the collection vehicles
and provides the means for storing it and feeding it to the process equipment
at the desired rate. The minimum clearance height to the roof is 6.1 m (20
ft). The floor shown will be able to store as much as 1361 Mg (1500 tons) of
raw refuse (piled 3 m or 10 ft high), or about one day's production. The
tipping floor is under roof to provide the receiving and handling operations
protection from wet or windy weather, to permit control of litter and dust
during these operations, and to provide a reasonable aesthetic appearance to
enhance the external environmental quality for a refuse handling operation.
The tipping floor area is unheated but has a large ventilating system for dust
control and for removal of air contaminants emitted from the trucks and other
190
-------
t —
SLUDGE COLLECTING PIT -
GAS COLLECTION MAIN
5
PRECIPITATOR —
SLAG CONVEYOR
CONVEYOR
PROCESSED WASTE
STORAGE
CONVEYOR
NoiSNVdxa aaninj
i i
•J — - --TT^^T^ ^
n
ROAD ^
V
SLUDGE
THICKENER -.
/^\ O
MAIN HOLE - -' — 1
^ AND LI FT
1 PUMP STATION
y
f— j — i j— |
5-i O—r
Y T
J 1
UNOX SYSTEM, TWO TRAINS DAF
FOUR STAGES — i
1
61 c^
AIR COMPRESSOR AND
POLYMER ADDITION FACILITY
SURGE TANK,
PUROX CONDENSATE FROM
PUHOX DECANTING UNITS
NEUTRALIZATION AND
o ^c:
1 MIXING T
r- CONTROL!
Q -FILTER
1— j— W ELECTRIC [B j | •* 1 —
o — COLUMN
LrniulpRPSSOR l-tlf PACKAGE
HOIISF 1 1 | -TURBINE OIL CONSO
"OU!>t L-T1IRBINE DUCT
I \ - J / Y REVERSING
\ J { j 1 H.X. PACKAGE
COOLING ' 15 OXYGEN PUMP
TOWER - ( | )
,
>
RE
R
-E
'
'-TL - 35,000 LIQUID
OXYGEN STORAGE TANK
Ti '
FROM SEWAGE
BAH SCREEN GRIT CHAMBER
.„ AND FLOW CONTROL
AIMK
ANEL
CONTROL ROOM, LAB, AND
PIPELINE COMPRESSOR
' ] ] !H UNDERGROUND
! | 1 | OIL STORAGE
u 'J U
N
L EXCHANGER
1 ABSORBER ^ f ° — '
GASOLINE- '" FUTURE PLANT
1=5—] PUMPS EXPANSION
I
TRANSFORMER -
n
EQUIPMENT
PROCESSING
AREA [-CS
""^-^ ^T^
-INSULATION
1
1
JL 1 r-
— ; J
r:
-------
I PROCESSING AREA
FLOOR ELEV. -eO'-O"
[A ' TRUCK ENTRY AND MOBILE
' ' ' EQUIPMENT MAINTENANCE
FLOOR ELEV. O'-Q" i
JElCTTS^T" 'WORKERS'COR^
CONFERENCE!
-------
mobile equipment. An elevated control room is provided that directly oversees
the tipping floor operations and monitors the processing area by closed cir-
cuit television.
The process area contain the waste processing equipment. This is a high
bay area in which the floor level is 6.1 m (20 ft) below that of the tipping
floor. The support services area is located in a one-story office and shop-
type building attached to the main structure. Another area of the plant is
outside adjacent to the processing area wall. This area contains the storage
bins, truck loading facilities for the recovered metals and residue, exhaust
fans, and filter baghouses for plant ventilation and dust control.
Plant Operation--
The control room provided above the tipping floor affords a full view of
truck operations on the floor and the operation of the feed conveyor. The
entire plant operation is monitored from the control room via remote START-
STOP switches with appropriate interlocks for all processing equipment, a
traffic signal system for directing incoming waste trucks, two-way radio to
the tipping floor front-loaders, telephones, and a loud speaker intercom sys-
tem to strategic points in the plant, including the weigh scales.
Organization of each processing line is in the manner of a continuous
step-by-step sequence of operations with each piece of equipment sized to ac-
commodate the volume of material coming to it from the operation immediately
preceding. The flow is continuous from the start of processing to completion.
The rate of processing is controlled by the rate waste is loaded onto the pro-
cess feed conveyor from the tipping floor and by variation of the speed of the
conveyor within the limits of equipment capacity. Flow of a load of solid
waste can be traced in Figure 39, Floor Plan.
Process Equipment Trains--
The initial processing of the raw refuse is accomplished in two identical
and independent trains of processing equipment, other than for a single alumi-
num recovery sub-system. Important processing equipment items, along with
their performance requirements, are described below. The numbers in paren-
theses after the equipment name correspond to the equipment numbers in Figure
39. Sizes of equipment are given and were used in developing cost estimates.
Front-End Loaders are utilized for handling the received raw refuse on
the tipping floor after it is deposited from the collection vehicles. They
stack the refuse in piles and feed it to the shredder feed conveyors as re-
quired.
Each Process Feed Conveyor (1) consists of a two-section, heavy-duty,
steel pan-type conveyor, 2 m (78 in.) in width. The first, or horizontal
section, 24 m (80 ft) long, is mounted in a pit in the floor with the conveying
surface 1.8m (6 ft) below tipping floor level. To prevent damage from mate-
rial dropped from the front-loaders, the conveyor pans must be formed from
steel plate at least 10 mm (3/8 in.) thick and be suitably reinforced struc-
turally. The pit has steel-plated sides sloping toward the conveyor at an
angle of 30° from the vertical. This section discharges onto the second sec-
tion, whose receiving end is below the first section and overlaps it slightly.
193
-------
The Primary or Coarse Shredder (2) is a heavy-duty, horizontal, reversible
hammermill rated for continuous duty at 54 Mg/h (60 TPH). This machine should
be equipped with a heavy metal hood and conveyor entrance configuration to
confine heavy material ballistically thrown by the rotating hammers. The grate
openings are set for producing a shredded material that will have a maximum
dimension of 20 cm (8 in.). The machine must be designed for ease in replace-
ment of those parts subject to rapid wear, such as hammers, impact plates,
grates and liners. Such a design should also incorporate easy-opening, hy-
draulically actuated doors to the grinding chamber.
The Magnetic Separator (9) used employs a suspended rubber belt having
two magnets behind it, one near each end pulley. The material stream being
processed is projected from its conveyor into the air under this magnet/belt
unit. All the material passes through the first magnetic field with the
ferrous material pulled up to the belt while the non-ferrous material falls
by gravity toward a hopper feeding a conveyor. When the ferrous material
passes between the two magnetic fields, it starts to fall from the belt and
air currents separate paper, fabric, or plastics that have been trapped against
the belt. The forward momentum of the ferrous materials carries them into the
second magnetic field which pulls them back to the belt and from there they
are conveyed to a ferrous metal hopper.
Each of two Air Classifiers (4,5,20,21) accomplishes air/density separa-
tion of the shredded material at a continuous rated capacity of 54 Mg/h (60
TPH). Equipment similar to the Chicago Supplementary Fuel Processing plant
was used. Other systems are on the market and, since this is a developing
technology, the final choice should be made based on system evaluation at time
of detailed design. The system includes the air classifier proper, a light
material conveying duct, an air supply blower, a material handling exhaust
blower, and a cyclone type material separator. The light fraction drops from
the bottom of the cyclone separator through an airlock onto the Processed Feed
Material Conveyor (7). The heavy density material from the bottom of the air
classifier is conveyed to the trommel preceding the aluminum separator. In-
stallation of the air classifier is included only if it is economically feasi-
ble to recover aluminum.
The Trommel (14) is a conventional design rotating screen with two stages
of separation, each with different screen openings. The trommel is sized to
process 22.5 Mg/h (25 TPH) of 0.24 to 0.48 g/cm3 (15-30 lbs/ft3) density mate-
rial. It also should have a dust shroud and hoppers to collect the separated
material and feed the respective conveyors. The first stage screen passes
material of less than 1.6 cm (5/8 in.) maximum dimension. The second stage
separation screen passes up to 10.2 cm (4 in.) material, which is the alumi-
num-rich fraction, to the conveyor feeding the aluminum separation unit. The
stream of material not passed through the screens is discharged from the trom-
mel to the Processed Feed Conveyor (7).
The Magnetic Separator (11) is a standard drum-type separator that picks
up the small amount of ferrous in the aluminum-rich stream that was not re-
covered by the first magnetic separator (9) and not lost in the first stage
screen of the trommel (14). Ferrous material is deflected to a conveyor
leading to ferrous bin.
194
-------
The Aluminum Separation Unit (16) is an electromagnetic eddy-current-type
machine that has been under development by several industrial firms. The sys-
tem performance is designed to recover aluminum cans.
Aluminum and Ferrous Material Storage Bins (13 and 18) are elevated,
bottom-dumping, truck-loading bins for holding the recovered metals that are
separated in the plant and must be transported to a customer. There are two
ferrous metal bins (13) associated with one shredder line, and designed for up
to 4 Mg/h (4-1/2 TPH). A truck trailer is filled every 4-1/2 hrs to 5-1/2 hrs.
The aluminum bin should be of 57 m3 (2000 ft3) capacity and will receive an
18 Mg (20 ton) load approximately every 3 working days.
There are three air-moving and Dust-Control Systems (8, 10, and 22) con-
nected to each processing line. System (8) consists of a hood (19) suspended
over the course shredder feed conveyor pit, a large filter baghouse, and an
exhauster fan. The approximate air handling requirements are 35.4 m3/s
(75,000 CFM) for each shredder line. This system produces a sufficient number
of air changes in the tipping floor area to maintain acceptable air quality
and controls dust in the tipping floor area. By air removal at this point, air
currents sweep the tipping floor and move dust generated from truck unloading
toward the pit for ultimate removal. The filter baghouse collects the dust
and prevents atmospheric contamination outside the building. A. second major
system (10) is for air classifier dust control. There are 31.9 m^/s (67,500
CFM) of air flow from each air classifier cyclone (5) leading to a baghouse.
It is desirable to keep occasional large particles out of the baghouse, and a
secondary cyclone plus a dropout box must be provided ahead of the filter bag-
house. A third and smaller system (22), handling approximately 7.1 m3/s
(15,000 CFM}, is in the dust-laden air from the various dust hoods or other
collection points in the process train.
Auxiliary Equipment--
In addition to the process equipment, certain items of support equipment
or systems are important to the facility.
A Waste Water Treatment System (24) is provided for the tipping area and
process area floors drainage system.
A house compressed-air supply of sufficient capacity and pressure is
furnished for filter baghouse operation, instrumentation, air-operated controls,
pneumatic tools, and other maintenance operations.
Truck scales, in the form of dual 50-ton capacity scales in the entrance
roadway, are provided.
Heating, ventilating,and air conditioning is provided in necessary areas
throughout the facility. Unless otherwise noted, all heat is by steam, gener-
ated with an oil-fired boiler located in the mechanical room. The tipping
floor area is unheated because of the large quantity of ventilation air re-
quired to remove mobile equipment exhaust gases. Except in severe weather
the truck entrance and exit doors remain open during plant operations to allow
major air inlet requirements to sweep across the tipping floor from the far-
thest corners to the hoods over the conveyor pits. An additional air inlet
195
-------
is provided by means of wall louvers. Steam pipes are imbedded in the floor
for a distance of 3.6 m (12 ft) around the conveyor pits for use in cities with
cold climates. The processing area has large quantities of air passing through
because of air classifier requirements. The incoming air is heated by units
mounted against the outside process room wall to temperatures at or above 4°C
(40°F), unless the building code of the specific site chosen requires a higher
minimum temperature. Wall-mounted unit space heaters are provided near the
working floor levels for heating during maintenance periods.
Fuel storage is required for the 75 700 dm3 (20,000 gallon) of oil needed
each week for the heating system, during maximum winter load, plus some steam
for the syngas plant wastewater treatment. A 120 000 dm3 (30,000 gallon)
underground storage facility is recommended. The front-loaders and residue
trucks from the syngas plant require on-site diesel fuel storage plus a dis-
pensing pump; a 20 000 dm3 (5000 gallon) storage tank is recommended.
Fire protection must be provided in accordance with national and local
codes. This requires complete sprinkler coverage in all areas. All areas
can have a wet system except the tipping floor area, which requires a dry type
because that room is subjected to freezing temperatures. The tipping floor
must be provided with fire hoses to combat refuse fires anywhere on the floor.
Other areas should be provided with wall-mounted hand extinguishers and hoses
in accordance with established standards. Appropriate outdoor fire hydrants
are also required.
Electrical systems require power to be received at a minimum of 4160
volts. The required capacity can be met by three 7500 kV-A substations for a
total of 22,500 kV-A. Distribution of the 4160-volt circuits is from the out-
door line or substation to indoor 5-kV switchgear and from 5-kV starters to
4160-volt motors. Starters rated at 4 kV are indoor-type and arranged in line
with the 5-kV switchgear. Load center-type unit substations are provided to
transform 4160 V to 480/277 V. The 277/480 V feeders are extended from the
motor control centers to distribution panels. Supplies to 120/208 V power and
lighting panels are provided by dry-type transformers located at each panel
and energized from the 480 V system.
A control system is provided for remote control of the process. Central
panels are provided for each line for independent operation with interlocked
controls for automatic sequential startup and shutdown of each major component
of the line. In the event of shutdown or failure of a component, the rest of
the line is programmed to shut down in proper sequence to avoid material jam-
ming of machinery that cannot be restarted under load. The central control
can also manually override the automatic programming. Local START-STOP con-
trols are provided at each piece of equipment to permit emergency shutdown
or facilitate maintenance operations.
Processed Feed Material Handling
A storage or surge bin is used for each gasifier to provide an even rate
of feed 24 hours per day, 7 days per week from a processing plant that operates
only 16 hours per day, 6 days per week. Storage capacity must equal 36 hours
of consumption, or a minimum of 467 Mg (515 tons) per gasifier. An A-frame
196
-------
shaped bin to prevent bridging of the waste in storage and to permit access to
the underside of the pile for the reclaim equipment permits material removal
on a first-in first-out basis. Each bin is 22.5 m (74 ft) wide at the base
by 30.5 m (100 ft) long. Rotating augers move the material to a belt conveyor.
Purox Converter System
The Purox converter system is that developed as proprietary equipment by
the Union Carbide Corporation. The three subsystems are designed and furnished
by UCC as an integrated system to function in the required manner under war-
ranties and performance guarantees. Operation is normally continuous. Moni-
toring is centralized in a control room located in a building that also houses
a small chemistry laboratory and the product pipeline compressor.
Gasifier--
; Each of these units consists of a fegd material conveyor, a pyrolysis
reactor, and gas cleanup equipment. Five conversion units of 317 Mg/d (350
TPD) capacity are provided. Four units operate continuously with one available
as a standby. The reactors operate at high temperature and should be operated
at design temperature as continuously as possible to minimize expansion/con-
traction cycling damage. In the event that feed supply is interrupted for
a short period, or is furnished at a reduced rate, it is possible to turn
down the conversion rate to as little as 25% of rated capacity and still keep
the reactor at operating temperature and produce a satisfactory gas. However,
the oxygen plant can be turned down to only 60% of design output. Beyond
that requires venting excess oxygen. Use of two oxygen plants, each at half-
size, allows for turn-down to 30% by shutting off one unit.
Oxygen Plant--
The source of oxygen for the Purox system will be an onsite cryogenic
oxygen generating plant. The process involves the liquefaction of air followed
by fractional distillation to separate its components. These plants are well
known so that details of their operating characteristics need not be discussed
here. Liquid oxygen can be transferred to storage facilities that will serve
as a backup oxygen supply in the event of a plant outage. Facilities for
transferring liquid oxygen from a transport vehicle are also included. For
assistance in startup, the design is arranged such that liquid oxygen can be
transferred from the storage tank into the cold box. Location of this plant
should be upwind from any gas producing equipment with possible hydrocarbon or
carbon oxides emissions. The cryogenic plant also supplies oxygen to the
wastewater treatment system. Also, use is made of gaseous nitrogen production
and liquid nitrogen storage for purging reactors and tankage. This nitrogen
supply is critical for a plant safety program.
Wastewater Treatment Plant--
The wastewater treatment plant specified is Union Carbide's UNOX waste-
water treatment system, which uses an oxygen activated sludge process for BOD
reduction. The system consists of covered multistage biodigestion reactors
in which oxygen is fed concurrently with the wastewater. The feed wastewater,
recycled sludge, and oxygen are introduced into the first stage. Mixed liquor
from the last reactor stage flows to a secondary clarifier that allows gravity
separation of the biomass from the treated wastewater. The treated wastewater
197
-------
flows over a weir for discharge to a municipal wastewater treatment plant.
The thickened biomass from the clarifier is recycled to the first stage of the
reactor and excess biomass or sludge is pumped to a thickener prior to trans-
port to a nearby municipal wastewater treatment plant for digestion.
Gas Compression and Drying
Because the syngas leaves the gasifier and cleanup system at approxi-
mately atmospheric pressure, a pumping system must be supplied to deliver the
gas through a pipeline to a user. 446 kPa (50 psig) has been chosen as the
lowest practical pressure that would deliver fuel gas at a high enough pres-
sure for average purposes; if the gas were delivered to an adjacent chemical
conversion or gas turbine-electric plant, a higher pressure would be required.
Three electric-driven centrifugal compressors are used at 1865 kW (2,500 HP)
each, operating in parallel. These machines are rated at 446 kPa (50 psig)
discharge, pumping 635 kg (1,400 Ib) per minute of a gas with a molecular
weight of 25. Any two of these machines will carry the full load, thus pro-
viding sufficient redundancy to ensure full capacity. If the gas is to be
piped any distance farther than to the immediately adjacent property, it must
be dried to prevent condensation problems in the pipeline. A standard glycol
gas drier is recommended. If an t^S level of 400 ppm is not acceptable for
a particular customer, a small sulfur removal unit can be additionally
installed.
ESTIMATED CONSTRUCTION AND OPERATING COSTS
This section presents estimated capital and operation costs for the facil-
ities for converting 1361 Mg/d (1,500 TPD) of solid waste to syngas. While
the syngas produced can be used directly as a fuel in power plant boilers,
it also can be chemically converted to other products as described in Appendix
C. The cost estimates presented in this section are based on mid-1975 prices.
The various cost elements are defined and cost sources are identified.
Capital Costs for Construction
Capital requirements for construction include land acquisition, site
improvements, buildings, equipment, an inventory of working materials and
spares, contingency reserve, engineering and construction management costs,
interest during construction, startup costs, and a working capital fund. A
summary of the estimated capital costs developed for the elements discussed
below appears in Table 38.
Details of the construction costs are described below and are summarized
in Table 39.
Taxes--
While the Contractor may need to pay a State sales tax, usually applica-
tion can be made for a 100% refund. It is assumed that any local taxes or
fees would be absorbed in the continency allowance. If the facility was owned
by a municipality, no property tax liability would accrue. To the extent that
unimproved property is removed from the tax rolls, there would be a minimal
198
-------
TABLE 58. CAPITAL COST REQUIREMENTS FOR PUROX SYSTEM
Item
$ Million
Construction
Interest during construction
Startup costs
Working Capital
Total
53.75
4.30
2.56
1.79
62.40
TABLE 39. ESTIMATED CONSTRUCTION COST SUMMARY (1975 DOLLARS)
FOR PUROX SYSTEM C1)
Item
Costs
($000)
Taxes (None Used)
Land Acquisition
Site improvement
Front-end processing plant
Processed feed material storage and handling facilities
UCC-supplied Purox equipment(2)
Purox equipment installation
Gas pumping and drying station
Subtotal
Contingencies (§10%
Engineering and construction management @ 10% less 6%
engineering on UCC-supplied Purox equipment
Total
468
880
10,066
3,476
17,703
10,675
2,410
45,678
4,567
3,505
53,750
(-1-1 Front end, processed feed handling, and gas compression
unit are Parsons designs.
^ Includes oxygen and waste water treatment units
drying
199
-------
loss in tax revenues to the local government. No taxes were included in con-
struction costs shown but there would be a need to review each specific case.
Land Acquisition--
The land requirement is 87 010 m2 (21.5 acres).
the cost of the site is $468,000.
At a cost of $0.50/ft2,
Site Improvement--
Site improvement costs include such items as clearing and grubbing, rough
grading, and construction of utility mains from the site to the supply mains.
Paving, fencing, and landscaping are also included as site improvements. Site
improvement costs are listed in Table 40. The amounts listed should be consid-
ered as reasonable allowances because the actual costs will be site-specific
and cannot be accurately estimated at this time.
TABLE 40. SITE IMPROVEMENT COSTS FOR PUROX SYSTEM
Item
Clearing and grubbing
Utilities
Earthwork (excavation, backfill, and disposal)
Fine grading
Paving
Landscaping
Fencing
Total
Cost
($000)
40
240
270
40
200
40
50
880
Front-End Processing Plant--
Costs for this portion of the facility are summarized in Table 41. The
cost estimates for the building are based on Parsons estimates for several
similar facilities and are verified by one such facility designed by Parsons
in Chicago and for which actual construction costs are known. Costs for this
building include miscellaneous support equipment not itemized elsewhere, such
as the house air system and fire protection system. The main process equip-
ment and related costs are based on purchase prices of equipment procured on
other projects, modified to reflect inflation escalation or updated quotations
received directly from equipment vendors. Installation costs are also shown.
These include only final electrical connection to the equipment components
from a nearby junction box. The major electrical distribution is included
in the building electrical cost listing. The bridge crane installation in-
cludes only setting the bridge on the rails because the rails and electrical
are included in the building costs.
200
-------
TABLE 41. FRONT-END PROCESSING PLANT FOR USE IN PUROX SYSTEM
Item
Cost
($000)
Building:
Piling allowance
Foundations
Structural
Architectural
Mechanical (except HVAC)
Heating, ventilating, and air conditioning
(not including dust control)
Electrical (including substation)
Subtotal
Process Equipment:
Front-end loaders
Shredders
Conveyors
Air classifier systems
Magnetic separators
Trommel
Aluminum system conveyors
Aluminum separation unit
Material storage and truck loading bins
(installed)
Ferrous and residue trailers and tractors
Pickup truck
Truck weighing scales (installed)
330
861
1,188
461
1,306
440
922
5,508
208
534
398
470
46
85
80
415
116
250
5
85
(continued)
201
-------
TABLE 41 (continued)
Item
Cost
($000)
Process Equipment (continued)
Dust control systems
Bridge crane (installed)
Subtotal
Equipment installation (not included elsewhere)
Spare parts and tools
Total
Total Plant Costs
1,140
120
3,952
448
158
4,558
10,066
Process Feed Material Handling and Storage--
The shredded refuse bins are simple A-frame structures with concrete
foundations. The cost for the five units, including lighting and ventilation,
is $2,125,000. The material receiving and delivering equipment for the storage
bins is listed and costs are shown in Table 42 at a total cost of $1,351,000
installed in addition to the amount above. The total cost for the feed mate-
rial handling and storage equipment is then $3,476,000. This cost is not
listed in Table 41 as part of the Front-End Process Plant but is included in
Table 39 as an item in total construction costs. The storage and material
receiving and reclaim equipment design is unique to the Purox System in that
there is a surge bin for each reactor. If another method of storage is de-
sired, its costs can be substituted for the one presented herein.
UCC-Supplied Purox Equipment--
This item is the mid-1975 quoted price of $17,703,000 for the hardware
supplied by Union Carbide Corporation as the basic Purox system equipment,
listed in Table 39. This price includes the reactors, their feed and residue
conveyors, the gas cleanup train, emergency flare, oxygen plant, and auxiliaries
for waste water treatment. The general scope of supply for the Purox system
(4 units plus one space) is listed below with the number of units in paren-
thesis for a 1361 Mg raw refuse per day (1,500 TPD) plant.
• Purox System 317 Mg/d (350 TPD) reactor (5)
Converter feed conveyor (5)
Feed conveyor refuse leveler (5)
Refuse feeders (5)
202
-------
TABLE 42. FEED MATERIAL HANDLING EQUIPMENT COST
FOR USE BY PUROX SYSTEM
Item
Cost
($000)
Material Receiving Equipment
Receiving conveyor: weather-shielded, -54-in. flat belt
2 required, installed
Distribution conveyor: 54-in. flat belt, 113 ft long
1 per bin installed
Traveling plow on distribution conveyor
1 per bin II $5,200 installed
Inclined conveyor: weather-shielded, 48-in. cleated belt
type with metal sides
2 required, 90 ft long, plut 15% installation
Automatic conveyor sequencing control
Conveyor wiring @ 15% of purchase cost
Total
Material Reclaim Equipment
Bin unloader assemblies: Miller-Hofft
2 required per bin @ $54,000 installed
Bin discharge conveyor: 48-in. wide troughed belt conveyor
110 ft long
2 required per bin installed
Conveyor wiring § 15% of purchase cost
Total Delivery Equipment
Total Feed Material Handling Equipment
192
130
26
100
6
60
514
540
202
95_
837
1,351
Refuse converters (5)
Slag quench tank and conveyors (5)
Electrostatic precipitators (5)
Condensate pumps (10)
Condensers (5)
Solid-Liquid Separation System (5)
Combustors (3)
203
-------
• Wastewater Treatment System (5.78 m3/min. or 2.2 MGD)
Surface aerators, electric motors, gear boxes and skids,
special valves
Purge blower
Instrumentation and controls
• Air Separation Plant (363 Mg/d or 400 TPD]
Interchanger
Liquid oxygen storage
Air compressor and driver
Aftercooler
Expander turbine
Oxygen vaporizer
Air surge tank
Slowdown silencer
Drain vaporizer
Thaw system
Liquid oxygen pumps (4)
Booster compressor
Circuit breaker panel
• Instrumentation
Local and remote control panels
Controllers, control valves, transmitters, and analyzers
associated with supplied equipment
• Engineering Services
Design, specification, procurement, and checkout of supplied
equipment and instrumentation
Installation of Purox Equipment--
The installation of the Purox equipment and related auxiliaries amounts
to $10,675,000 and is shown as a separate item in Table 39. Parsons was
204
-------
unable to estimate this cost in the usual manner due to lack of sufficiently
detailed data. The figure is based on Union Carbide Corporation estimates with
some adjustment by Parsons, based on discussions with UCC personnel Specific
items that this cost must include are:
(1) Utilities, such as steam, cooling water, electric power, and in-
strument air
(2) Electrical substations and switchgear
(3) Buildings, control room, and laboratory
(4) Interconnecting piping, tubing, and wiring
(5) Wastewater treatment auxiliary equipment, such as reactor pumps,
clarifiers, and sludge thickeners
(6) Installation of Union Carbide-supplied equipment
(7) Spare parts and tools
(8) Area lighting
(9) Lift pump and connecting trunks to local sewer system
Contingency--
A reasonable, but conservative, total estimate of construction cost in-
cludes an estimated reserve for contingencies. The higher the contingency
factor, the lower the confidence level in the accuracy of the cost estimate.
Ten percent has been chosen for this facility because the basis for the esti-
mate has been fairly well established, except, perhaps, for the installation
of the Purox equipment.
Engineering and Construction Management--
The costs of preparing construction plans and for management of the proj-
ect during construction are estimated as 10% of the construction costs.
Interest During Construction--
With construction of the facility scheduled over a period of 2 years,
debt service will accrue on funds expended over that period. Previous studies
indicate that interest on total construction cost is generally paid for one-
half of the construction period and is so shown in Table 38.
Startup Costs--
The production startup costs include breaking in a new facility and cor-
recting construction deficiencies. An allowance has been made for the fact
that the facility incorporates new technology that does not have the benefit
of many years of operational experience. It is assumed that operating costs
for the first year will be that for operation under design conditions. How-
ever, the production rate (and hence, income) will be 10% of rated capacity
during the first quarter, increasing to 35% in the second quarter, 65% in the
205
-------
third quarter, and finally, 92% in the fourth quarter, for an overall first-
year rate of 51%. Since the plant does not operate at full capacity, it is
reasonable to take a credit against operating costs for production-rate-related
costs. These include power, mobile equipment fuel, extra raw refuse hauling,
water, and residue disposal, which are described in operating costs below.
Startup costs are estimated at 49% (100-51) less 25% of the production rate-
related costs for a total of $2.56 million, as shown in Table 38.
Working Capital--
Working capital listed in Table 38, like startup cost, is estimated as
a percentage of annual operating costs. For this study, the equivalent of 3
months operating costs, or 25% of annual operating cost, shown in Table 43 is
used.
Operating Costs
Operating costs are those repetitive costs that occur as a result of
operating the facility, including such items as direct labor, maintenance
services and supplies, utilities, fuel, and transport costs. The estimates
are reported on an annual basis and summarized in Table 43.
TABLE 43. SUMMARY OF OPERATING COSTS FOR PUROX SYSTEM
Item
Labor (not including UCC personnel)
Power
Heating fuel and miscellaneous
Process equipment and building maintenance supplies
Mobile equipment, maintenance and replacement
Mobile equipment fuel
Water and sewer
Extra raw refuse hauling
Residue disposal
Insurance
Union Carbide management fee
Total
Annual
Cost
($000)
$2,159
2,307
200
840
92
28
250
56
45
289
750
7,016
206
-------
Direct Labor--
A detailed review of all probable job assignments related to the plant
size, type, location and probable plant operating schedule provides a reason-
able manhour estimate by payroll category for direct labor. The dollar amount
is determined by applying local wage rates for the same or similar tasks,
including adjustments for shift differentials and anticipated overtime, 'ihe
wage rates were determined on the baiss of the U.S. Bureau of Labor Statistics
area wage survey (Bulletin 1850-15). Modifications can be made for specific
localities. Table 44 presents a chart of personnel by category, number re-
quired, hourly rate, and disposition by shift for the front-end processing
plant. Table 45 presents the same information for the syngas plant.
Table 46 summarizes labor costs. Included is a 50% hourly surcharge to
cover fringe benefits and miscellaneous overhead costs such as payroll and
accounting. Administration management is covered in a special cost item
discussed later. The estimated overtime allowance of 3,300 hr/yr will permit
development of a 6-day week operating schedule.
Shown in Table 46 are 6 standby personnel to fill in for vacations, sick
leave, and overtime shifts for front end operators as follows: 1 process
operator, 1 heavy equipment operator, 1 electrician, 1 mechanic, and 2 laborers,
These people are assumed to be in training for a higher labor grade and are
assigned to cover various positions. Also, 5 standby personnel are added for
the Purox plant operation to fill in for vacations, sick leave, and overtime
shifts as follows: 1 reactor monitor, 1 maintenance helper, 1 water treatment
plant monitor, and 2 laborers.
Because the Purox plant operates continuously, it is possible to develop
a rotating shift schedule with an overtime allowance of 2,400 hr/yr. This
is accomplished with 4 men assigned to each shift-day position, as indicated
in Table 45. Annual labor cost for the complete facility is $2,159,000 as
shown in Table 46.
Electric Power--
Electric power costs were calculated on the basis of a 25 mills/kWh rate.
A value for the cost of electric power in specific areas can be substituted
by ratio in the calculations shown below. A typical 12-month operating
schedule was assumed with a connected load of 20,000 hp and an average operat-
ing demand at a rated capacity of 14,800 kW. This is a capacity with both
the front end and Purox system operating. A weighted average was computed
on an annual basis, resulting in the following:
ECeSWy = ACe
Where
E = kWh/ton raw refuse =179
Ce = Average cost per kWh consumed = $0.025
ECe = Average cost per ton raw refuse = $4.47
207
-------
TABLE 44. PERSONNEL REQUIRED FOR FRONT-END PROCESSING PLANT
Total
1
3
4
7
2
1
3
3
1
3
3
4
1
2
38
Position
Superintendent
Shift Supervisor
Process Operator
Heavy Equipment Operator
Equipment Monitor Front-End
and Aluminum
Mobile Equipment Mechanic
Maintenance Mechanic
Maintenance Mechanic
Maintenance Welder
Maintenance Helper
Yardman
Laborer
Janitor
Records Clerk-Steno
Total
Basic Pay
$/hr
10.00
8.90
7.19
7.19
6.35
7.19
7.19
7.19
7.19
6.38
6.38
5.80
5.80
5.00
1st
Shift
1
1
2
4
1
0
1
1
0
1
1
2
0
2
17
2nd
Shifta
0
1
2
3
1
0
1
1
0
1
1
1
1
0
13
3rd
Shift5
0
1
(Mech.)
0
0
0
1
1
1
1
1
1
1
0
0
8
Total
$/Position/hr
10.00
27.10
29.06
50.79
12.91
7.40
21.97
21.97
7.44
19.54
19.54
23.60
5.95
10.00
267.39
O
00
*2nd Shift, add $0.15/hr.
33rd Maintenance Shift, add $0.25/hr.
-------
TABLE 45. PERSONNEL REQUIRED FOR PUROX SYNGAS UNIT
Total
1
1
4
8
4
8
4
4
4
4
4
8
1
4
4
63
Position
Superintendent
Plant Engineer
Shift Supervisor
Process Operator
Heavy Equipment Operator
(Residue)
Reactor Monitor
Feed Material Monitor
Product Compressorman
Maintenance Mechanic
Electrician
Maintenance Helper
Laborer
Instrument Technician
Water Treatment Plant Monitor
Water Treatment Plant Operator
Total
Basic Pay
$/hr
c
c
c
c
7.19
6.55
6.38
7.19
7.19
7.19
6.38
5.80
7.50
6.38
7.19
1st
Shift
1
1
1
2
1
2
1
1
1
1
1
2
1
1
1
18
2nd
Shift3-
0
0
1
2
1
2
1
1
1
1
1
2
0
1
1
15
3rd
Shiftb
0
0
1
2
1
2
1
1
1
1
1
2
0
1
1
15
$/Position/hr
-
-
-
-
29.29
53.46
26.05
29.29
29.29
29.29
26.05
47.46
7.50
26.05
29.29
333.02
Is)
O
2nd Shift, add $0.15/hr.
3rd Shift, add $0.25/hr.
These positions are furnished as part of Union Carbide Corporation's management fee.
-------
TABLE 46. SUMMARY OF LABOR COSTS FOR PUROX SYSTEM
Annual Cost
($000)
Regular Operating Positions:
Front-end processing plant - $267.30 x 2,080 hr/yr
Syngas plant - $333.02 x 2,080 hr/yr
(not including UCC personnel)
Standby personnel - average rate, including
shift differential: front-end plant = $6.86
syngas plant = $6.32
Front-end plant - 6 men x $6.86 x 2,080 hr/yr
Syngas plant - 5 men x $6.32 x 2,080 hr/yr
Overhead and fringe benefits @ 50%
Overtime:
Front-end plant - 3,300 hr x $6.86 x 1.5 (Premium)
Syngas plant - 2,400 hr x $6.32 x 1.5 (Premium)
Total
555
693
86
66
701
34
23
2,159
SWV = Yearly solid waste processed = 515,500 tons
ACe = Total annual power cost = $2,307,000
Maintenance Supplies and Services (not including mobile equipment)--
Because a large maintenance staff is a part of the base payroll, this
item need cover only parts and special services, and is estimated at 2-1/2%
of purchase price of mechanical equipment and 1% of building costs. In addi-
tion, a special cost of $0.14/ton of raw refuse processed is added for the
rapid wearing parts of the shredders, such as the hammers, for an annual cost
of $72,000. The total for maintenance supplies is $840,000.
Mobile Equipment Maintenance--
This item is calculated at 8% per year of the purchase price, or $37,000.
In addition, it is presumed that this equipment will have to be replaced every
7 years resulting in providing a sinking fund payment of $55,000/yr. This
fund is anticipated to earn compound interest at 6-1/2%, which credit was not
used in this report.
210
-------
Process Steam, Heating Fuel, and Miscellaneous--
Steam for heating and other miscellaneous needs requires 13 000 bbl/yr of
fuel oil resulting in a cost of $182,000/yr. Allowing $18,000 for miscella-
neous items, such as telephone and office supplies, the total is $200,000/yr.
Mobile Equipment Fuel—
This requirement is calculated at approximately 60,000 gal, or $25,000 for
the front loaders and 4,000 gal, or $2,100 for the truck fleet, plus $1,000 for
gasoline for the pickup truck.
Water and Sewer--
The estimate of $250,000/yr is based on discussion with water and sanita-
tion district offices. A convervative cost of $0.20/m3 ($0.75/1,000 gal) plus
55% for sewer charge, was established. Water usage in the plant will probably
run in the 2082 to 2271 m3/d (550,000 to 600,000-gpd) range. A special arrange-
ment will probably have to be made considering the nature and needs of the
water treatment plant. The accuracy of this cost is not high because of the
uncertainties in connections, piping, pumps, and charges.
Residue Disposal--
Approximately 81 648 Mg/y (90,000 TPY) of residual material from the pro-
cessing plant will have to be disposed to landfill (or stored and marketed).
This material has a density of approximately 1.28 g/cm^ (80 lb/ft^) for a total
of 64 230 nrVy (84,000 yd-Vyr). Since disposal cost is calculated on a volu-
metric basis, the cost per ton will be less than for raw refuse. An allowance
of $45,000/yr is made for this disposal. Vehicles and personnel for hauling
are provided for elsewhere in the plant costs.
Extra Raw Refuse Hauling--
To keep this plant a full operation, there may be need to bring refuse
from more distant areas, particularly in the winter. Allowance has been made
at the rate of $56,000.
Insurance--
Insurance premiums are normally set by negotiation after inspection of the
specific facility by the underwriters. Representative premiums were established
by discussions with several insurance companies. The rates employed on the
capital items were 0.25% for the front-end processing and material handling
facilities, and 0.6% on the gas producing and handling facilities. General
personnel liability is figured at 1.5% of direct payroll. Insurance premiums
amount to $267,000 and $22,000, respectively, for a total of $289,000 anually.
UCC Management Fee--
The Union Carbide Corporation presently has a policy by which sales con-
tracts on all Purox plants have a reserve clause that provides the right for
UCC to negotiate a contract to provide operating management for the plant. This
fee was determined to be $750,000/yr by UCC for this size plant. Included in
the fees are payroll and overhead costs for the plant superintendent, the plant
engineer, the shift supervisors, and the process operators, all of whom are
provided by UCC.
211
-------
Other Purox Gas Utilization Systems
Construction and operating costs for converting Purox gas to methane,
electric power, methanol, and ammonia are presented in Appendix C. Presented
are costs including values developed above for the front-end and syngas pro-
duction plus the syngas utilization or conversion plant.
ECONOMIC ANALYSIS FOR PRODUCING SYNGAS
For purposes of example and to provide a first order economic comparison
of the alternative systems, the parameter used is a net cost per unit of pro-
duct. This is calculated from unit amortization cost plus unit operating cost
less unit revenues from drop charges, sale of steel, and aluminum. Actual
costs at the time the plant is put in operation can be estimated by applying
escalation factors on the various items shown in the previous section.
An analysis is shown below for the syngas product case. Material can be
found for other product cases in Appendix C.
The basis for obtaining the net costs presented is:
* Municipal solid waste delivery averages 1361 Mg/d (1,500 TPD)
• On a yearly basis, an appropriate utilization factor is used for each
system.
• Drop charges of 0, $5.51, and $11.02/Mg (0, $5, arid $10/T) of raw
refuse.
• Steel revenues of $44/Mg ($40/T) of steel.
• Aluminum revenues of $331/Mg ($300/T) of aluminum.
» Amortization at 8.5% for 20 years.
• All costs are in 1975 dollars.
• Equipment sizes in the front end are based on a process rate of 109
Mg/h (120 TPH) with an equal split between two process lines.
Information is given here to show a possible range of net costs for pro-
ducing a fuel gas using the Purox pyrolysis system. A complete financial anal-
ysis for a given community can be performed taking into account specific costs
and financing methods to be associated with the project. Costs for producing
syngas are more accurate and detailed than for the products described in Appen-
dix C because of design and construction experience on the front end, and cost-
ing experience for the Purox System in recent studies for communities. Calcu-
lation of the unit costs and net cost for syngas production using a $11.02/Mg
($10/T) drop charge case is given in some detail below as an example for one
of the cases evaluated. The plant capital cost is $62.40 million from Table
38, and annual operating cost is $7,016,000 from Table 43. Steel recovery is
111 Mg/d (122 TPD) and aluminum is 5.4 Mg/d (6 TPD). The utilization factor
212
-------
is 0.92 and the product rate is 250 x 106 Nm3 (9.31 x 109 SCF) per year, with
the corresponding yearly delivery rate of raw refuse at 456 957 Mg (503,700
tons). The items of net unit cost* are:
(1) Amort. Cost = C$62.40 x 10^) (Q.1Q567) x 1()6 = ^
9.31 x 109 x 370
(2) Operating Cost = 7>016>000 x 10 = $2.04/106 Btu (1.93)
9.31 x 109 x 370
(3) Steel Rev. = 122 X 365 X °'*2 X 40 X ^ = $0.48/10* Btu (0.45)
9.31 x 109 x 370
,„,,-,. n 6 x 365 x 0.92 x 300 x 106 „,- ,,/m6 D+. fn i^
(4) Aluminum Rev. = - = $0.17/10 Btu (0.16)
9.31 x 109 x 370
(5) Drop Charge ($10/T) = 10 x 505>700 x lo6 = $1.46/106 Btu (1.38)
9.31 x 109 x 370
Therefore, the net cost for the $10/T drop charge case is:
(1) (2) (3) (4) (5)
(1.91 + 2.04 - 0.48 - 0.17 - 1.46)/10° Btu = $1.84/10° Btu ($1.74/GJ)
This corresponds to $12.56/ton ($13.85/Mg) raw refuse.
For a drop charge of $5/ton ($5.51/Mg), the net cost is $2.53/106 Btu
($2.40/GJ) or $17.56/ton ($19.36/Mg) raw refuse. With no drop charge, the net
cost is $3.30/106 Btu ($3.13/GJ) or $22.56/ton ($24.87/Mg) raw refuse. The
net cost can vary widely over the United States or in other areas of the world
due to the different values of off-setting revenues from drop charges, steel,
and aluminum. In some areas, glass may produce sufficient revenues to justify
a recovery system. Drop charges for present methods of disposal have risen to
over $16.53/Mg ($15/ton) in some areas. If a market value for steel, aluminum,
and the final fuel product (in this case, syngas) is known, a drop charge to
break even can be determined and compared to existing drop charges as part ot
the decision-making process a community uses for selection of the best means
of solid waste management.
If the calculation above is extended to a special case where the gas is
sold at the battery limits for $2.50/10* Btu C$2. 37/GJ) (equivalent to about
$15/barrel oil), the drop charge required would be about $5/ ton ($5.51/Mg)
refuse. This is attractive in almost any part of the U.S.
'Calculations are shorn in English units as an aid to un^"J^in^the para
meters involved; the SI units following in parenthesis are in terms of 5/U
213
-------
ENVIRONMENTAL ASSESSMENT
Aqueous Effluents
Cooling tower water requirements are satisfied with city water. The rela-
tively clean blowdown stream from the cooling water system consists of 3.3
dm3/s (53 gpm) enriched in dissolved solids due to the concentration of origi-
nal salts and the additional presence of corrosion inhibitors such as polyphos-
phates or organics. Total dissolved solids are estimated at 1200 rag/dm-3, with
corrosion inhibitors contributing 50 mg/dm3 to this value. The BOD of the
blowdown stream is negligible. This water will be used for ash-quenching make-
up water and waste water dilution. An additional minor blowdown stream, esti-
mated at 0.2 dm3/s (3 gpm), originates from the steam system (oil-fired steam
generator) used for heating purposes; this stream has characteristics similar
to the ones of the cooling water blowdown, and can be similarly disposed of.
A small amount of city water, 0.6 to 1.3 dm3/s, or 10-20 gpm, is used for
cooling, with no contamination occuring during the process.
The heavily polluted aqueous waste stream generated from wet scrubbing of
the pyrolysis gases, consisting of pyrolysis condensates plus water from drying
the refuse feed, has a flow rate of 4.4 dm3/s (70 gpm) and a BOD5 of 50,000 ppm
due to the presence of water-soluble organics such as alcohols, organic acids,
and aldehydes. Table 47 presents measured results of the organic fraction in
the condensate effluent to be treated. The BOD load is too high for a conven-
tional activated sludge system, but could be handled by pretreating using oxy-
gen in place of air, as in the Union Carbide Unox process. Dilution of this
effluent will be required and is achieved by mixing the 4.4 dm3/s (70 gpm)
effluent stream with 92 dm3/s (1,460 gpm) of sewage upstream of the
TABLE 47. CHARACTERISTICS* OF PUROX SYSTEM CONDENSATE EFFLUENT
Organic Compound
Methanol
Ethanol
Acetone
Methyl ethyl ketone
Acetic acid
Propionic acid
Butyric acid
Furfural
Phenol
Benzene
Other
Max. Wt. % in
effluent
1.1
0.5
0.5
0.1
0.8
0.4
0.1
0.5
0.1
0.06
0.38
4.54
* COD max 77,000 mg/dm3; BOD max 52,000 mg/dm3; pH min. 3.7; air
stripping drops COD to 55,000 and BOD to 35,000
214
-------
oxygen-activated sludge treatment. Other alternatives could be considered,
such as the high-pressure oxidation process, using high-pressure air (11 030
kPa or 1,600 psi) at 150 to 315°C (300 to 600°F), or the activated carbon
process. The wet oxidation process could reduce the BOD load to 400 ppm, and
the carbon process to even lower values, but pilot plant tests would be re-
quired before drawing definite conclusions concerning effectiveness and costs.
The concentration of organics is high enough (4.5%) that other chemical or
physical procedures not usually considered in wastewater treatment could be
attempted; examples are reduced pressure distillation, adsorption on nonionic
resins, molecular sieves, or liquid-liquid extraction.
Gaseous Effluents
Table 36 shows contaminants measured by UCC at South Charleston in the
Purox gas. Gaseous effluent streams are also generated in other sections of
the syngas plant. The front end of the plant generates varying amounts of
dust-laden air that is collected from covered conveyor belts and dusthoods and
led to cyclones and baghouses at an estimated flow rate of 80.7 Nm3/s (180,000
SCFM). These remove approximately 4.5 Mg/d (5 TPD) of dust particles from the
air streams and release air to the environment that contains dust amounts well
below applicable particulate standards. The light-fraction material from the
air classifier used in the aluminum recovery process is separated from the air
prior to storage in the surge bin by a de-entraining cyclone. Air from the
cyclones, estimated at 60.3 Nm3/s (135,000 SCFM), is conveyed through a bag-
house prior to venting to the environment. The dust collected at 4.5 Mg/d
(5 TPD) is added to the surge bin as part of the pyrolysis feed.
Air cooling of the water circulating in the cooling tower produces a drift
corresponding to 0.1% of the cooling water loading. The total amount of cooling
water for both the oxygen plant and the Purox system is 852 dm3/s (13,500 gpm),
Therefore, the drift consists of 51 dm3/min. (13.5 gpm) with 61 g/min. (915
grains/min.) of particulates, a negligible amount, being released to the atmos-
phere on evaporation of the water in the droplets.
The oxygen plant separates 1450 Mg/d (1,600 TPD) of air into 290 Mg/d
(320 TPD) of oxygen used in the Purox reactor, and 1160 Mg/d (1,280 TPD) of
nitrogen, which is vented to the atmosphere. A portion of the nitrogen stream
could be utilized in synthetic processes such as the generation of ammonia or
as inert gas.
The oil-fired steam generator used for heating during the cold season and
for supplying steam to the wastewater system consumes 114 to 500 kg/h (250 to
1,100 Ib/hr) of No. 2 fuel oil. According to EPA-supplied emission factors,
the maximum air emissions per hour are 1 kg (2.3 Ib) of particulates 5.7 kg
(12.6 Ib) of sulfur dioxide, 0.3 kg (0.6 Ib) of carbon monoxide, 0.2 kg (0.5 Ib)
of hydrocarbons, and 4 kg (8.8 Ib) of nitrogen oxides.
If the syngas produced is conveyed to a power plant by pipeline, moisture
removal to an acceptable dewpoint is required. This is carried out in an air
cooler and a glycol absorber. Glycol picks up some hydrogen sulfide Capproxi-
mately 100 ppm out of a total of 500 PPm) from the syngas together with the
moisture; on regeneration of the glycol in a reboiler stripper, the hydrogen
215
-------
sulfide would be released to the air with the water, an unacceptable procedure.
The regeneration effluent is therefore passed through a small combuator, which
converts hydrogen sulfide to sulfur dioxide, and released 454 kg/h (1,000 Ib/hr)
of water and 6.4 kg (14 Ib/hr) of sulfur dioxide to the air.
The syngas produced by the facility may have to be vented to the air in
case of emergency. This would occur through a flare combustor that oxidizes
all components to carbon dioxide and water (plus a small amount of sulfur
dioxide) prior to release to the atmosphere. The flare is enclosed and not
visible from the outside.
Solid Wastes
The 227 Mg/d (250 TPD) residue generated by the Rurox process is an inert,
glassy aggregate similar to blast furnace slag. This material is suitable for
many applications such as road building and construction material. It can be
landfilled if suitable markets do not develop.
Noise
Noise control is an integral part of the layout and design of the plant.
Special attention during equipment design and engineering layout will be given
to the air compressor (oxygen plant), gasifier (Purox system), fans, fixed
heaters, and the shredder.
Odor
No noxious odors are expected from a properly run refuse processing plant.
No odors are produced from the pyrolysis reactors or oxygen plant.
Traffic
Traffic generation of a syngas plant is essentially that due to the in-
coming and outgoing wastes. Approximately 280 refuse collection trucks per
day will arrive and depart the plant Monday through Friday, and about 140
trucks will be handled on Saturday. Residue must be hauled from the plant to
the nearby landfill at a rate of one truckload every 2 hours; traffic impact
is expected from this around-the-clock operation if in an industrial or land-
fill area.
Visual
No commercial syngas plant exists on which to base an evaluation of the
visual impact of such a facility. However, a probable syngas plant layout and
artist's perspective are included in this report. The plant can be described
as visually characteristic of industrial type plants. The exposed pyrolysis
reactors and the liquid oxygen plant are to be especially noted in this respect.
While the plant shown is generally devoid of visually attractive architecture
and landscaping, it is doubtful that even careful attention to these disci-
plines would allow such a facility to visually blend with other than industrial
areas. If near a residential area, much of the plant can be enclosed attrac-
tively and safely with, of course, additional expense.
216
-------
Probable Environmental Effects of Proposed Projects
Air Quality--
A syngas plant project will not significantly affect existing air quality
levels. Emissions from the plant consist of vented, dust-free air, stack gas
from an oil-fired steam generator used for heating during the cold'season, and
some sulfur dioxide from the gas dryer, if such dryer is needed. The total
difference in vehicle miles compared to an existing refuse system cannot be
determined until specific plant and disposal sites are selected. Emissions
from the plant have been estimated in Mg/y (tons per year) as 2.7 (3) for par-
culates, 61.7 (68) for sulfur oxides, 8.2 (9) for carbon monoxide, 7.3 (8) for
hydrocarbons and 21.8 (24) for nitrogen oxides. These are not items in the
syngas which is passed through a pipe to a customer.
It is of interest to determine emissions that would occur if the syngas
were burned in a utility type furnace. Based on data furnished by UCC and
generic data for boilers using gaseous fuel, the following Parsons estimates
were computed in Mg/y (tons per year): particulates, 32.6 (36); S02 285 (314);
NOX, 154 (170); hydrocarbons 10.9 (12); and CO nil. Removing H2S from the
syngas by well tested techniques can reduce the S02 to less than 4.5 Mg/y (5
TPY) and with the use of staged combustion the NOX can be reduced considerably.
Water Quality--
Water quality will not be affected by a syngas project. A relatively low
volume of wastewater will be generated and this will be discharged to existing
sewage treatment facilities. Syngas plant effluent will be pretreated before
release to the sewage system to meet criteria established by the sewage plant
operators.
Solid Waste Disposal--
A Purox syngas project will significantly reduce the volume of solid waste
requiring disposal in a landfill. This will increase the lifetime of the exist-
ing landfills. Additionally, the landfill residue will be less likely to leach
into percolating waters and ultimately into the groundwater.
Land Use--
The increased lifetime of existing landfills will reduce the amount of new
lands that must be set aside for landfills. A gas line would have to be con-
structed to link a syngas plant with any steam plant. Construction of this
line could affect numerous land uses along the route. An analysis of the
environmental effects of the pipeline must be addressed in the pipeline route
selection study.
ENERGY BALANCE
Presented in this section are the energies associated with the raw refuse,
fuel and electric power required for operation, and the products. The residue
is inert and does not account for any energy. For the syngas plant, tne energy
balance is given as follows:
217
-------
Fuel
0.26 TJ/d
0.25x109 Btu/day
Raw Refuse
14.65 TJ/d
13.89x109 Btu/day
SYNGAS PLANT
Front End
Gasifier
Compressor/Dryer
0.74x10
, 14.57 MJ/Nm
27.7x10 SCFD, 370 Btu/SCF
10.81 TJ/d
10.25x109 Btu/day
2.97 TJ/d
2.82x109 Btu/day
Heat Required To Produce
11.75MW Elec. Power
The conversion efficiency is
10.81
or
10-25
14.65 w" 13.89
and the net thermal efficiency (English units) is
10.25 - (2.82 + 0.25) _ -^
13.89 "* bl-/1i
DETAILED SUMMARY OF PRODUCTION AND COSTS FOR PUROX SYNGAS
Summary of Product Rates
The rates of MSW delivery and processing are as follows:
Process
Delivery or
Receiving
Front end
Processing
Purox Gasifier
Type of Operation
8 hours, 6 days
16 hours, 6 days
24 hours, 7 days
Daily Weight MSW Processed
1588 Mg (1,750 tons)
1588 Mg (1,750 tons)
1361 Mg (1,500 tons) MSW
equivalent
1239 Mg (1,366 tons)
Shredded Feed
The daily and yearly product rates of various items are shown in Table 48.
Daily rates are for 24 hours/day operation and the yearly rates take into ac-
count a utilization factor of 0.92 based on scheduled and estimated unscheduled
shutdowns.
218
-------
TABLE 48. SUMMARY OF PLANT PRODUCTION FOR SYNGAS AS FUEL
1561 Mg/d (1,500 TPD) PLANT
Product Item
Raw refuse processed per day, Mg (ton)
Raw refuse processed per week, Mg (ton)
Raw refuse processed per year, Mg (ton)
Ferrous product recovered per day, Mg (ton)
Ferrous product recovered per year, Mg (ton)
Aluminum product recovered per day, Mg (ton)
Aluminum product recovered per year, Mg (ton)
Residue to landfill per day, Mg (ton)
Residue to landfill per year, Mg (ton)
Product gas volume per day, 103 Nm3 (106 SCF)
Product gas volume per year, 106 Nm3 (106 SCF)
Gas energy per day, TJ (109 Btu)
Gas energy per year, TJ (1012 Btu)
Quantity
1588 (1,750)
9 526 (10,500)
457 000 (503,700)
111 (122)
37 170 (40,970)
5.4 (6.0)
1 823 (2,010)
222 (245)
74 700 (82,300)
742 (27.71)
249 (9,310)
10.81 (10.25)
3633 (3.444)
Utilization factor of 0.92 for yearly operation values.
Calculation of Net (Unit) Costs
Net costs (or unit costs) have been calculated for syngas based on the
previously described revenue assumptions. A simplified equation characterizing
the calculations is:
Net Cost
Trr
C x F
SWy
OM
n R
- D - Rs -
Where
C =
Capital cost = construction cost +
interest during construction + startup
costs + working capital, $
F = Amortization factor =
i
n
SWy
OM
1 - (1 + i) -n
Interest rate per year
Number of year for amortization
Solid waste delivered, tons/y
Operating and maintenance cost = labor + power
maintenance + production materials + water and
sewage costs.
219
-------
D = Drop charge, $/ton raw refuse
TPD steel
Rs = Steel revenues, $/ton steel x ^-^ refuse
, . , TPD aluminum
Ra = Aluminum revenues, $/ton aluminum x TPD raw refuse
Trr = Tons raw refuse
The net cost in other units can be similarly derived.
A community can make appropriate changes in each item given in (the above
formula to fit other specific cases and determine the corresponding unit costs.
Also, escalation of costs and revenues can be introduced for each item and an
actual net cost determined for design, installation, aid operation on a real
time basis. A period of 4 years is appropriate for accomplishing preliminary
engineering, financing, detailed engineering, construction, and startup.
Because operating costs and revenues will change over the life-time of
the plant, it is recommended that a cash- flow analysis be used to provide a
more accurate picture of eventual net costs (or revenues) . ' The actual market
value for syngas can vary greatly depending on competitive fuel prices, envi-
ronmental requirements, governmental incentives, etc. For these latter reasons,
a cash flow analysis may be difficult to make. The more optimistic view is
that market prices for fuels and materials will rise similarly to construction
costs.
Costs and Economics as a Function of 'Syngas Fuel Plant Size
Costs are estimated for three plant sizes based on multiples of Purox
gasifier modules, each with a 317 Mg/d (350 TPD) capacity. Sizes chosen were
635 Mg/d (700 TPD) (2 modules), 1361 Mg/d (1500 TPD) (4 modules), and 1905
Mg/d (2100 TPD) (6 modules) . The base case previously evaluated was for a
1361 Mg/d (1500 TPD) plant using 4 modules plus one spare. Union Carbide
requested that a spare be available in the two larger size ranges. Costs were
therefore estimated for 4 or 5 modules at the 1500 TPD capacity and 6 or 7
modules at the 2100 TPD capacity.
The construction costs are evaluated for the different capacities based
on the more detailed values determined previously and summarized in Table 49-
For Purox equipment, costs vary somewhat linearly because of modularization.
Other equipment costs vary with the 0.8 power of plant capactiy. For the
operating costs summarized in Table 50, total cost of labor varies only slight-
ly with capacity because of the minimum number of personnel to manage and
operate the plant. Production-related supplies or utilities vary somewhat
linearly with capacity. A Union Carbide management fee is shown as explained
previously.
An estimation was made for either using or not using a spare for the two
larger plant size cases. Because of the increased reliability of having a
spare, a utilization factor of 0.92 (48 out of 52 weeks) was used and 0.85
220
-------
TABLE 49. CONSTRUCTION COSTS TO PRODUCE SYNGAS AS A FUNCTION OF PLANT SIZE
($000) 1975
No. of Modules
Land
Site Improvement
Front End Processing
Plant
Process Feed Storage §
Handling
UCC- Supplied PUROX
Equipment
PUROX Equipment
Installation
Gas Pumping § Drying
Section
SUBTOTAL
Contingency @ 10%
Engineering and
Construction
Management @ 10%,
less 6% UCC-
Supplied Equipment
Plant Size
635 Mg/d
(700 TPD)
2
269
505
5,003
1,390
9,460
5,130
1,384
22,141
2,214
1,646
1361 Mg/d (1,500 TPD)
4
468
880
10,066
2,781
14,160
8,929
2,410
39,694
3,969
3,120
5
one spare
468
880
10,066
3,476
17,703
10,675
2,410
45,678
4,567
3,505
1905 Mg/d (2,100 TPD)
6
612
1,152
13,920
4,171
21,240
12,350
3,333
56,778
5,678
4,403
7
one spare
612
1,152
13,920
4,866
24,780
14,770
3,333
63,433
6,343
4,856
TOTAL
26,001
46,783
53,750
66,859
74,632
-------
TABLE 50. ANNUAL OPERATING COSTS ($000) 1976 FOR SYNGAS PLANT
No. of Modules
Utilization Factor
Labor
Power
Process Equipment and Building
Supplies
Heating Fuel and Miscellaneous
Mobile Equipment and
Replacements
Mobile Equipment Fuel
Water and Sewer
Extra Refuse Hauling
Residue Disposal
Insurance
Union Carbide Management Fee
TOTAL
Plant
635 rag/d (700 TPD)
2
0.80
1,880
1,200
420
150
60
15
125
20
23
121
500
4,514
1360 Mg/d (1
4
Size
,500 TPD)
5
one spare
0.85
2,159
2,307
840
200
92
28
250
56
45
215
750
6,942
0.92
2,159
2,307
840
200
92
28
250
56
45
289
750
7,016
1905 Mg/d
6
0.85
2,440
3,450
1,260
280
130
42
375
75
68
302
850
9,372
(2,100 TPD)
7
one spare
0.92
2,440
3.450
1,260
280
130
42
375
75
68
340
850
9,410
M
K)
-------
for no spare. For the 635 Mg (700 ton) case with two modules only and no spare
the reliability is considered less, at 0.80. '
Figure 40 shows the variations of net unit cost, based on the heating
value of the fuel gas, as a function of plant size with and without a spare.
It should be particularly noted that using a spare costs less in the higher
capacity range because of the increased utilization factor. In other words
the estimated increased reliability of production of fuel gas for sale over
a one year period is sufficient to justify the expense of a spare. Operating
experience will determine whether or not a spare is justified.
Table 51 shows the capital required to build and bring the plant to full
operation.
The economics of plant operation are calculated on the same basis as pre-
viously and results are summarized in Table 52 A§B. Amortization of capital
requirements is based on 8-1/2% interest over 20 years. Credit is taken for
aluminum and steel recovered at $330 and $44 per Mg ($300 and $40 per ton)
respectively. These credits can be easily interchanged with other values and
new net costs calculated in Table 52 A§B.
TABLE 51. CAPITAL COSTS ($000) FOR SYNGAS PLANT
No. of Modules
Construction
Interest During
Construction
Startup Costs
Working Capital
TOTAL
Plant Size
635 Mg/d (700 TPD)
2
26,000
2,080
1,750
1,130
30,960
1360 Mg/d
(1,500 TPD)
4 5
one spare
46,780 53,750
3,740 4,300
2,530 2,560
1,740 1,790
54,790 62,400
1905 Mg/d
(2,100 TPD)
6 7
one spare
66,860 74,630
5,350 5,970
3,280 3,270
2,340 2,350
77,800 86,220
223
-------
O
O
COMPRESSED AND DRIED FOR PIPELINE DELIVERY
CREDITS TAKEN FOR ALUMINUM & STEEL
UTILIZATION FACTOR (UF| WITH SPARE MODULE 0.92,
AND 0.80 AND 0.85 WITH NO SPARE.
SEVEN D/wk OPERATION
NO SPARES
ONE SPARE — ——
DROP CHARGE
— —. __~ $0/Mg(SO/T)
— — — _ _ S5.51/Mg ($5/T)
$11.02/Mg($10/TI
UF 0.92
500
Mg/d
1000
1500
2000
500
1000
1500
2000
TPD
PLANT SIZE
Figure 40. Cost of producing syngas.
224
-------
TABLE 52A. .SUMMARY OF NET UNIT COST TO PRODUCE PUROX SYNGAS AS A FUNCTION OF PLANT
SIZE (SI UNITS)
Item
No. of Modules
Capital ($000)
Amortization ($000/y)
8-1/2%, 20 years
0 § M ($000/y)
Total Cost of operation
($000/y)
Utilization Factor
Refuse Feed (Gg/y)
Product Gas (106 Nm3/y)
Product Gas (GJ/y)
Aluminum § Steel Credits
(106$/y)
Net
Cost,
$/Mg
Net
Cost ,
$/GJ
Drop Charge,
$/Mg
0
5.51
11.02
Drop Charge,
$/Mg
0
5.51
11.02
Plant Size
635 Mg/d
2
30,960
3,272
4,514
7,786
0.80
185.43
101.16
1 474
946
36.88
31.37
25.86
4.64
3.95
3.26
1361 Mg/d
4
54,790
5,790
6,952
12,742
0.85
422.21
230.31
3 356
2,147
25.10
19.60
14.08
3.16
2.46
1.76
5
one spare
62,400
6,594
7,016
13,610
0.92
456.95
249.41
3 633
2,243
24.60
19.09
13.61
3.13
2.39
1.74
1905 Mg/d
6
77,800
8,221
9,372
17,593
0.85
591.04
322.44
4 698
3,003
23.70
19.17
13.65
3. 11
2.42
2.04
7
one spare
86,220
9,111
9,410
18,521
0.92
641.84
349.88
5 102
3,261
23.78
18.27
12. 75
3.00
2.29
1 .60
-------
TABLE 52B. SUMMARY OF NET UNIT COST TO PRODUCE PUROX SYNGAS AS A FUNCTION OF PLANT
(SIZE (ENGLISH UNITS)
Item
No. of Modules
Capital ($000)
Amortization ($000/y)
8-1/2%, 20 years
0 & M ($000/y)
Total Cost of operation
($000/y)
Utilization Factor
Refuse Feed (TPY)
Product Gas (106 SCFY)
Product Gas (106 Btu/y)
Aluminum § Steel Credits
(103$/y)
Net
Cost,
$/T
Net
Cost,
$/106 Btu
Drop Charge,
$/T
0
5
10
Drop Charge,
$/T
0
5
10
Plant Size
700 TPD
2
30,960
3,272
4,514
7,786
0.80
204,400
3,776
1,397,090
946
33.46
28.46
23.46
4.90
4.17
3.44
1,500 TPD
4
54,790
5,790
6,952
12,742
0.85
465,400
8,597
3,181,000
2,147
22.77
17.77
12.77
3.33
2.60
1.86
5
one spare
62,400
6,594
7,016
13,610
0.92
503,700
9,310
3,444,000
2,243
22.56
17.56
12.56
3.30
2.53
1.84
2,100 TPD
6
77,800
8,221
9,372
17,593
0.85
651,500
12,036
4,453,000
3,003
22.39
17.39
12.39
3.28
2.55
1.82
7
one spare
86,220
9,111
9,410
18,521
0.92
707,500
13,060
4,836,000
3.261
21.57
16.57
11.57
3.16
2.42
1.69
-------
SECTION 9
OCCIDENTAL RESEARCH CORPORATION FLASH PYROLYSIS SYSTEM
INTRODUCTION AND SUMMARY
Unique among the fully researched waste-to-energy systems in that a
liquid fuel is directly produced, the flash pyrolysis process of the Occi-
dental Research Corporation (formerly Garrett Research and Development
Company) is now entering the stage of development that will establish the
technological, economic, and environmental feasibility for potential com-
mercialization. The demonstration plant features a high degree of materials
recovery. Its chemical conversion system is characterized by an inert hot
char heat exchange system that prevents any oxidation of the original waste
or the pyrolytic produced fuel.
In 1968, as an outgrowth of research on the conversion of coal to a low-
sulfur fuel oil, Occidental Research Corporation (ORC) began studies on means
to convert the organic portion of municipal refuse to a usable liquid fuel as
well as to recover metals and glass from it. The decision was made in the
early stages of development that the materials should be separated in a rather
high degree of purity so that markets for them could be assured. This ob-
jective has apparently been met through the several R § D programs.
Fundamentals of the conversion process were established .with laboratory
equipment capable of processing 1.4 kg/h (3 Ib/hr). Waste feed, in addition
to municipal refuse, included bark, rice hulls, sewage sludge, animal manure,
and rubber. This work was scaled up to a 3.6 Mg/d (4 TPD) pilot plant where
the critical process variables were investigated, materials handling problems
resolved, and sufficient product produced to establish its properties, in-
cluding those as a fuel in burner test equipment. Information was obtained to
serve as the basis for the design of a 181 Mg/d (200 TPD) plant at El Cajon
(near San Diego), California, now undergoing start-up tests. While the de-
scription presented here is of the new facility, yields and product properties
are necessarily based on work accomplished during the earlier R § D phases.
The most complete published descriptions of the Occidental process are
those cited in References 16 and 17, while References 18 and 19 present other
aspects of the work.
CONCLUSIONS
• Liquid fuels offer significant advantages. Large weights of liquids
can be stored in relatively inexpensive tankage in contrast to gas,
where production (or transport) must be more or less adjusted to
227
-------
usage. Where liquid fuels have low sulfur and ash contents, they can
be utilized in existing gas/oil furnaces having minimum cost pollution
control systems. The Western States in particular have few coal-
fueled boilers and a waste-derived liquid fuel is far more attractive
than a solid RDF. No other waste-to-energy process for yielding a
fuel oil is at all as thoroughly investigated as is the ORC one and
those organizations having narrowed their options to gaseous or liquid
fuels should give detailed consideration to it. Cost analysis should
be made on the basis of locally acceptable fuel oil sales prices
rather than national average energy costs.
• The front end materials processing and recovery systems of the dem-
onstration plant are among the most advanced known. The performance
of such equipment should be reviewed and analyzed to determine its
applicability to other waste-to-energy processes to increase overall
plant revenues.
• Only limited characterization of the pyrolysis oil has been ac-
complished. Statistical variation of input waste composition should
be made and effects on the oil measured. Properties to be examined
should include chemical composition, density, viscosity and pour
points, stability, corrositivity, and the usual combustion tests
applied to new fuels.
• Fuel value remains in the char, but the high ash content renders it
of marginal value at the present time. Uses for this material should
be further explored.
• Waste feedstocks other than MSW should be used at El Cajon once
sufficient information is obtained in the refuse testing program.
• The chemistry of the process leads to a conversion efficiency lower
than most waste-to-energy systems. In no way should this efficiency
be viewed as negating the value of the process. To obtain a barrel
of oil from a ton of waste is of great importance to the nation's
energy program and the scheme should be so judged.
• The demonstration plant can serve as an excellent research tool to
determine product characteristics as functions of waste particle size,
reactor temperature, and residence time. The original small-scale
Occidental research demonstrated that a range of gaseous and liquid
molecular species could result and advantage should now be taken of
the large unit to learn what fuels can be synthesized.
PROCESS DESCRIPTION
The prime concern of this report is waste-to-energy conversion. The
materials recovery portion of the Occidental process, however, cannot be
isolated from the pyrolysis sub-systems of the overall scheme. The equipment
is energy consuming, but the revenues potentially generated from sales of
glass, steel, and aluminum have significant effects on plant economics and
228
-------
thus the required selling price of the pyrolytic fuel oil. To consider only
the energy-producing steps would lead to erroneous conclusions as to the cost
effectiveness of the process.
Figure 41 shows the process schematic of the Occidental resource recovery
system, basically consisting of a "front end" physical processing and mate-
rials separation section and a pyrolysis/purification section. The functions
and operating characteristics of the equipment within these two sections are
discussed separately below.
Raw refuse composition serving as the basis for design was established
both from experience in pilot plant tests and examination of available survey
information. The values used by ORC in the following table are sufficiently
close to those in the EPA "Third Report to Congress" that no attempt has been
made to adjust them.
Component
Organics (dry)
Magnetic metals
Aluminum
Other metals
Glass
Misc. other solids
Water
Amount, Wt-%
54.4
7.6
0.5
0.3
9.0
3.2
25.0
Daily Input
Total
100.0
Mg
97.7
13.8
0.9
0-5
16.3
5.8
45.4
181.4
Tons
108.8
15.2
1.0
0.6
18.0
6.4
50.0
200.0
The HHV of the as-received municipal waste is 10.70 MJ/kg (4,600 Btu/lb).
Front End System
The first eight steps of processing prepare the raw refuse for the
materials recovery systems. Several fractions are isolated, including one
that is a finely divided organic "fluff" used as the feedstock for the
pyrolysis unit. The elimination from this feedstock of most of the inor-
ganics is an important function of the front end system. The pyrolysis
process itself is not affected by these inerts, but the quality of the
residual char would otherwise be lowered and maintenance costs for the
secondary shredder would be increased.
From storage, unsorted municipal wastes are conveyed to the Primary
Shredder, where size reduction to less than 10 cm (4 in.) is accomplished
in a heavy duty hammermill. A Magnetic Separator then removes 95 percent of
the ferrous metals as the shredded waste is conveyed to the Air Classifier.
229
-------
SHREDDER
DRYER
K>
Cri
o
O
RECYC-AL
EDDY CURRENT
SEPARATOR
"FLASH OIL
PYROLYSIS" COLLECTION
REACTOR SYSTEM
CYCLONE
GAS
WATER
OIL
CHAR
BURNER
ALUMINUM
FERROUS METAL
Figure 41. Schematic of Occidental resource recovery system.
-------
The classifier is of the zig-zag type and was designed by ORC. Organics
entrained with the inorganic fraction from the air classifier are reclaimed
in a later stage of processing. Some 75 percent of the shredded refuse is
taken off in the light (overhead) fraction. Approximately 95 percent of the
original wet organics are recovered in this fraction and 8 percent of the
inerts.
The heavy (underflow) fraction is further treated to recover glass, non-
ferrous metals, and entrained organic material. A Trommel (rotating screen)
is used for the initial separation. The first section, containing 1.2 cm
(0.5 in.) holes, passes the more brittle waste components such as glass,
ceramics, rocks, and bones. Typically the composition of this fraction is
approximately 50 percent glass and it is conveyed to multi-stage Froth
Flotation Tanks after having been ground in a rod mill to a size range of 840
to 44 vim (20 to 325 mesh) . Proprietary chemicals in these tanks cause the
glass particles to have an affinity for air and they rise through the water on
air bubbles while non-glass materials sink. The float material after drying is
99.5 percent £lass and represents about 70 percent of the total glass in the
original refuse.
A second section of the trommel contains holes that are 10.2 cm (4 in.)
in diameter. Material passing through these holes contains 10 percent metal
and is conveyed to the "RECYC-AL" Eddy Current Separator" for recovery of
aluminum. Material greater than the hole size is returned to the primary
shredder feed. A pair of linear induction motors positioned beneath a con-
veyor belt causes non-magnetic electrically conductive materials to be de-
flected into a collection system. A travelling magnetic field is generated
by the motors, inducing eddy currents in metal pieces such as aluminum. A
magnetic field of opposite polarity to that of the motors is produced, re-
sulting in the metal being ejected off the travelling belt. The product
collected consists of about 90 percent aluminum and approximately 60 percent
of the aluminum originally present in the refuse is thus isolated. The 10%
impurities in the aluminum fraction consist of entrapped materials of all
kinds from the grinding operation and objects displaced into the collection
bin by moving aluminum pieces.
The light fraction from the air classifier is conveyed to a Dryer, a
rotary kiln of the type used for removing water from agricultural products,
where the moisture level is reduced to about 3 percent. While not essential
to the pyrolysis conversion step, this drying does help optimize the con-
version and improves separation in the subsequent screening system.
Material not passing through a 1410 /urn (14 mesh) Screen has had its in-
organic content reduced to about 4 percent. The undersize material contains
approximately 65 percent organics and is further purified on an Air Table,
where three fractions are obtained. The light fraction has a high organic
content and is added to the screen oversize material. A heavy glass-rich
fraction is introduced to the glass recovery system. The small intermediate
fraction is landfilled.
High heat transfer rates are important to the rapid pyrolysis process.
Small particles are required and hence the final front end processing step
231
-------
is to pass the organic fraction through a Secondary Shredder, an attrition
mill consisting of counter-rotating disks. The product is quite fine, with
80 percent of it able to pass a 1410 urn (14 mesh) screen. Because of the
potential fire hazard in this operation, a pressurized inert atmosphere is
maintained within the grinder. Power consumption tests demonstrate approxi-
mately equal amounts of power are required in the primary and secondary
stages of grinding. As shown in Figure 42, this amounts to 118 to 148 kJ/kg
(40 to 50 hp-hr/ton) in each stage.
Pyrolysis System
In contrast to the rather high density moving solid bed converters of
the Purox, Torrax, and Georgia Tech pyrolysis systems, the ORC fuel production
process occurs in a rapidly moving gas stream. Carried along by an inert tur-
bulent gas (recycled product gas), the finely divided organics from the
secondary shredder are heated by hot particles also flowing with the gas
stream. These char ash particles are formed and heated in the Char Burner
by combustion of the char that is one of the products of pyrolysis. It is
introduced into the Flash Pyrolysis Reactor at a temperature of approximately
760°C (1,400°F) and at a mass flow rate five times greater than that of the
waste material. Cooling occurs within the reactor so that the actual average
temperature for the conversion process is on the order of 510°C (950°F).
The gas exiting the reactor is passed through a mechanical Cyclone, where
the ash and the newly formed char are separated. As excess ash builds up
during the process, a portion is periodically removed for disposal.
After most particulate matter has been removed from the stream, it is
passed into the Oil Collection System where the temperature is rapidly
quenched to approximately 80°C (175 F). This is accomplished by spraying a
light fuel oil into the gas, effectively stopping any further thermal de-
composition. The liquid fuel then settles to the bottom of a decanter, from
where it can be moved by pipe to storage tanks. A portion of the water formed
in the pyrolysis process is retained with the oil for the purpose of reducing
its viscosity.
After clean-up, the gas is compressed for use as (1) the oxygen-free
transport medium and (2) fuel for preheating the combustion air into the char
heater, the rotary kiln dryer for the coarse-shredded waste, and various
process heat needs. All gas finally exits through an afterburner, heat ex-
changer, and baghouse filter system before it is discharged to the atmosphere.
Typical distribution of the yield of products from the pyrolytic reactor
is shown in Table 53, based on dry material entering the reactor exclusive of
the gas stream and hot ash. Overall product yields for the total system are
discussed below.
232
-------
r~ 10
UJ
N
UJ
o
I-
CE
Q.
UJ
(T
UJ
200
100
50
2
20
10
UJ
z
u
5.01
q
\
o\o
I
PRIMARY
HREDDING*1
\
SECONDARY
"SHREDDING"
HP-HR /TON
30
60
90
120
CD
N
•H
10
o
• H
-P
U
I
CD
o
PH
W)
C
•H
CD
fn
CO
Ofl
•H
kJ/kg
50
100
150
200 250 300 350
TOTAL ENERGY CONSUMED
233
-------
TABLE 53. TYPICAL PRODUCTS OF OCCIDENTAL FLASH PYROLYSIS SYSTEM
YIELDS AT 510°C (950°F), BASED ON DRY WEIGHT OF
FEED TO PYROLYSIS REACTOR
Oil (Dry) - 40% C
H
HHV = 24.66 MJ/kg (10,600 Btu/lb) N
S
Cl
Ash
0
Char - 20% C
H
HHV = 19.0 MJ/kg (8,200 Btu/lb) N
S
Cl
Ash
0
Gas - 30% H2
HHV = 14.96 MJ/Nm3 (380 Btu/SCF) CO
co2
CH4
C2H4
C2H6
C3
V
H2S
HC1
Water - 10%
57.0 wt-%
7.7
1.1
0.2
0.3
0.5
33.2
100.0
48.8 wt-%
3.3
1.1
0.4
0.3
33.0
13.1
100.0
12 vol-%
37
37
6
3
1
1
2
0.8
0.2
100.0
Materials and Energy Balance
The anticipated mass flow output for the process is as follows, based on
unity input of raw refuse in any mass units:
Product
Oil (containing 14% water)
Gas
Quantity
0.256 (0.221 as dry oil)
0.441
234
-------
Product
Char/Ash
Water to Sewer
Residual to Landfill
Ferrous Metal
Glass Cullet
Aluminum
Quantity
0.082
0.019
0.065
0.072
0.061
0.004
Total
1.000
The energy balance of the total system, based on 1 Mg or the parenthetical
values for 1 ton* of input refuse, is as follows:
Refuse
Front End Power
10.70 GJ
(9.2 x 10b Btu)
0.238 GJ
66.1 kWh
Non-Ferrous
Separation Power
Glass
Separation Power
Pyrolysis
(55.0 kWh)
0.028 GJ
7.9 kWh
(7.2 kWh)
0.010 GJ
2.9 kWh
(2.6 kWh)
0.252 GJ
70.1 kWh
System Power (63.6 kWh)
0.024 GJ
6.6 kWh
Afterburner
and Utility Power (5.0 kWh)
OCCIDENTAL
RESOURCE
RECOVERY
SYSTEM
5.43 GJ
(4.67 x 10b Btu}
0.52 GJ
(0.45 x 10b Btu)'
Oil
Char
energy recovery efficiency is based on the assumption that a ^eat equivalent
portion of the product oil would be used to generate electri"7J^ * ^
heat rate of 10,550 kJ or 10,000 Btu/kWh). This energy penalty is assessed
*Note that the English units are not factored conversions from « (which
would imply an input of 1.1023 tons), but for convenience are based on the
convention of 1 (exactly) ton input.
235
-------
against the product and this value then compared to the original energy con-
tent of the refuse. If only the oil is considered as a useful product, this
mathematically (in English units) becomes:
4.67 x 106 - 1.59 x 1Q6
* = 9.2 x 106
If the output energy is considered to include that in the char, a material
that would be difficult to sell as a fuel because of its high ash content,
the energy recovery would increase by 0.52 (0.45 x 106 Btu) and efficiency
would then be 40.5%. Comparison to efficiencies of other processes, even
pyrolysis systems, must be attempted only with a full recognition of the worth
of final products. That some 60% of the originally totally wasted energy is
required to operate a process able to "create" large quantities of a synthetic
fuel oil should not be considered discouraging, but viewed as a factual de-
scription of a given chemical system. Further energy efficiency can be at-
tributed to the savings that result from recovery of glass, ferrous metals,
and aluminum, in that manufacture of new materials in contrast to recycling
old ones is a more energy-intensive process. No consistent set of assump-
tions has been yet developed for quantification of the "inherent" energy in
the recovered materials, but the additional 3.37 GJ/Mg (2.90 x 106 Btu/ton)
sometimes cited by Occidental is an entirely reasonable value and would raise
the efficiency to 67.2% (excluding char).
PRODUCT CHARACTERISTICS
Oil
As with petroleum itself, the oil produced in the ORC process is a com-
plex mixture of molecular weights and structural configurations. While its
chemistry has not been investigated to any great detail, sufficient charac-
terization has been made to establish the probable value of the liquid as a
utility fuel. Key properties of the product are shown on Table 54 along with
those of No.6 fuel oil for comparison.
Important differences between the two oils that can be noted include:
Elemental Analysis--
The high oxygen content of the pyrolytic oil, a result of the largely
cellulosic composition of the original waste, results in a decreased HHV
compared to normal hydrocarbon fuels and causes marked solubility (60%)
capability of the oil. Water is retained to decrease viscosity. The oxygen
content, in addition to the chloride level, results in some acidity of the
product; storage should present no particularly difficult problem and details
of materials to be used will be established during the El Cajon demonstration
plant study. An additional characteristic that thusfar is attributed to the
high oxygen content is that extended high temperature storage causes a further
increase in viscosity and it is recommended the oil be maintained below 71°C
(160°F) until just before atomization. The low sulfur content is a property
of the pyrolytic oil that makes it an attractive RDF. The low ash content,
being markedly less than solid forms of RDF, is another important feature of
the liquid fuel.
236
-------
TABLE 54. TYPICAL PROPERTIES OF NO. 6 FUEL OIL AND OCCIDENTAL'S
PYROLYTIC OIL
Composition, wt-%
C
H
S
Cl
Ash
N
0
Specific Gravity
Heating Value
MJ/kg
MJ/dm3
Btu/lb
Btu/gal
Pour point, °C (°F)
Flash point, °C (°F)
Viscosity
mm2/s at 88°C
SSU at 190°F
Pumping temperature, °C (°F)
Atomization temperature, °C (°F)
No. 6 Oil
87.5
10.5
0.7-3.5
-
0.5
2.0
0.98
42.33
41.47
18,200
148,800
18-29 (65-85)
66 (150)
48
340
46 (115)
104 (220)
Pyrolytic Oil
57.0
7.7
0.2
0.3
0.5
1.1
33.2
1.30
24.66
32.03
10,600
114,900
32* (90*)
56* (133*)
160*
1,150*
71* (160*)
116* (240*)
*Pyrolytic oil containing 14% water, as marketed.
Specific Gravity--
The pyrolytic oil has an unusually high density, some 34% higher than
that of the usual fuel oil. The ORC product has a higher energy content per
volume than any other refuse-derived fuel, a factor that will reduce its
transportation costs relative to other RDF's.
Heating Value--
While even on a volumetric basis the HHV of the pyrolytic oil is 23% less
than that of fuel oil, it is higher than an average coal and if used in con-
junction with a liquid fossil fuel, a substantial portion of the total heat
input to the furnace can be supplied by it without any major modifications
to the system or its steam-generating characteristics.
237
-------
Flow Properties--
The presence of 14 percent water alters flow properties of the pyrolytic
oil sufficiently to permit it being handled with conventional equipment, al-
though the Occidental product remains more viscous than No. 6 oil. The effect
of temperature is greater for the synthetic oil, however, such that the atom-
ization temperatures are only 12°C (20°F) apart.
The combustion properties of oil produced in the pilot plant were briefly
examined in research burners by Combustion Engineering, Inc. Blends of pyro-
lytic oil of 25 and 50 percent by volume with No. 6 oil derived from Alaskan
crude were used. Such blends eventually separate because of the solubility
characteristics of the oxygenated oil, but are stable for several hours. It
was established that ignition stability is equal to the fossil oil alone and
that combustion is successful with properly designed fuel handling equipment.
At air levels over 2 percent excess oxygen, there were negligible quantities
of unburnt carbon in the stack emissions.
Glass
The mixed color cullet recovered from the froth flotation tanks contains
in excess of 99.5 percent glass and represents about 70 percent of the original
glass content of the as-received municipal waste. The material can be drained
dry to 5-10% water content or further dried, depending on relative costs of
heat energy as compared to transportation. A typical particle size distribu-
tion is as follows:
Size
Tyler Mesh Wt-% Cumulative Wt-%
>833 +20 0 0
833 to 495 -20 +32 11 11
495 to 295 -32 +48 41 52
295 to 246 -48 +60 13 65
246 to 147 -60 +100 20 85
147 to 74 -100 +200 12 97
< 74 -200 3 100
This cullet may be directly employed for glass container manufacture, with
some 15 percent less energy being required for melting than with glass raw
materials. Hand blown cruets and molded containers made by a leading glass
manufacturer were free of any "stones" or other defects. The mixed color
composition of this cullet could present some marketing problems and Occi-
dental is continuing to examine glass color sorting techniques. While up
to 20 percent of the mixed cullet may be added to amber color batches and
30 percent to green batches without significantly affecting final product
238
-------
color, approximately 65 percent of container manufacture now calls for flint
(colorless) glass. A large resource recovery facility might produce more mixed
cullet than required by the colored glass industry in the regional area, thus
limiting revenues from this product source.
Metals
The quality of ferrous metals collected by the magnetic separation equip-
ment is typical of the many other facilities using this technique. Residual
food wastes, coatings, and entrained paper products contaminate the metals and
soft metals used for ferrous soldering and corrosion protection also are present.
The ideal market is sales to a chemical de-tinner who in turn sells the steel,
usually as No. 1 Dealer Bundles.
The metal isolated by the Occidental RECYC-AL eddy current device_ contains
about 90 percent aluminum, with the principal impurities being copper, zinc,
iron, and miscellaneous other metels such as tin, lead, nickel, chromium, etc.
Beverage container aluminum must meet stringent specifications in order to be
used with modern production equipment and hence the prime market for this mixed
metal will be the secondary aluminum alloy industry. These manufacturers blend
available salvage stock to produce desired alloys and the Occidental non-ferrous
metal fraction will present no problems for them.
ENVIRONMENTAL CONSIDERATIONS
Recent experience in other facilities has indicated that principal environ-
mental problems occur on start-up of a new full-scale waste processing facility.
This same experience, however, demonstrates that solutions exist within the cur-
rent state-of-the-art and additional expenditure will typically permit opera-
tion of the plant within applicable local and EPA regulations. While develop-
ment tests to date have not shown any effluents or noise are produced for which
design features have not been incorporated into the El Cajon plant, it is
believed that the type of problem that might be encountered should be solvable
with addition of some extra control equipment. A specific EPA experimental
investigation or environmental effects of this facility has been scheduled.
All front end handling and processing steps producing an air stream con-
taining particulate matter are controlled by passing the gas through a baghouse
fabric filter system. Effluents from the char burner and waste drier are passed
through an afterburner, fueled by a portion of the pyrolysis recycled off-gases,
where any combustible matter is exposed to a minimum of 649°C [1,200°F) for
at least 0.5 second under oxidizing conditions. The gas then passes through
another baghouse before being released to the atmosphere. A process heater
within this system supplies heat to the dryer and various process lines.
Estimated emissions from the afterburner baghouse are:
239
-------
Component Concentration
S02 700 ppm (wt)
NO 8 to 1,000 ppm (wt)
HC1 100 ppm (wt)
Particulates 0.12 g/Nm3 (0.05 gr/SCF)
The wide range in the value for nitrogen oxides is a result of the extremes
of assuming (1) only atmospheric nitrogen is fixed according to the known
thermodynamics of this reaction and (2) that in addition all nitrogen entering
the afterburner is involved in the equilibrium.
Two contaminated water streams exist. That from the glass recovery sys-
tem is the larger of the two, but tests have verified that standard floculation
reagent addition, clarification, and filtration brings the water to a quality
level permitting discharge to a sewer system. The second stream, totalling
approximately 3.6 Mg/d (4 TPD) for the demonstration paint, results from the
product quenching and collection system. This effluent can contain up to
100,000 ppm of COD. Limited experiments indicate the organic contaminants are
fully biodegradable, but typical local regulations would forbid discharge of
this liquid directly to a sewer system. Reduction of the COD load would con-
sist of the use of one of several standard biological waste water treatment
systems. Occidental has suggested that in some applications of the recovery
plant sufficient heat might be available for afterburning the entire water
effluent.
Residual solids amount to 13 to 16 percent of the weight of the input
refuse. About half of this is inert ash from the pyrolysis system and the
remainder is rejected material from the air table and glass recovery system.
The inert portion of this latter is approximately 50 percent.
Noise, as with other waste processing systems involving front end treat-
ment, is principally from the size reduction equipment. Sufficient experience
in attenuation of this sound energy has now been obtained that no problems are
anticipated at the demonstration plant.
ECONOMICS
The 181 Mg/d (200 TPD) resource recovery plant at El Cajon is intended to
be a facility for demonstration of the technological and environmental feasi-
bility of the Occidental process. It incorporates a rather high degree of
design versatility and the instrumentation/control system is capable of data
compilation not required for a production plant. The process, largely due to
its advanced materials treatment sub-systems, is acknowledged to have a high
cost sensitivity for input capacity and the demonstration plant's size is
below the level where good returns on investment would normally be obtained.
Accordingly, the discussion of the economics of the system here is limited to
processing capacities of 907 Mg/d (1,000 TPD) and 1814 Mg/d (2,000 TPD). The
composition of the municipal solid waste and the product output ratios are those
previously presented.
240
-------
Operational Characteristics
It is assumed that waste will be received 5-1/2 days per week and proces-
sed 7 days per week. The front end system will be operated 16 hours per day
and the remainder of the plant 24 hours per day. An annual plant utilization
factor of 0.90 has been used. Table 55 shows the annual processing capacities
of the two plants.
TABLE 55. ANNUAL PROCESSING CAPACITIES OF 907 AND 1814 Mg/d
(1,000 AND 2,000 TPD) PLANTS (0.90 UTILIZATION FACTOR)
Component
Organics
Magnetic Metals
Aluminum
Other Metals
Glass
Misc. Solids
Water
Oil (14°s water)
Char/Ash
Solid Residue
Water to Sewer
Flue Gas
Total
Feed Input
Smaller Plant
Gg 103 ton
162.2 178.8
22.7 25. 0
1.4 1.6
0.9 1.0
26.8 29.6
9.5 10.5
74.6 82.2
298.2 328.7
Larger Plant
Gg 103 ton
324.4 357.6
45.4 50.0
2.8 3.2
1.8 2.0
53.6 59.2
19.0 21.0
149.2 164.4
596.4 657.4
Output
Sma 1 1 er
Gg
21.5
1.2
18.2
76.3
24.4
19.4
S.7
131.2
298.2
Plant
103 ton
23.7
1.3
20.0
84.1
26.9
2J .4
6.2
144.6
328.7
Larger
Gg
42.9
2.4
36.4
152.7
48.9
38.8
11.3
262.4
596.4
Plant
103 ton
47.3
2.6
40.1
16S.3
53.9
42.7
12.5
289.3
657.4
Capital Costs
Table 56 lists the various elements of 1976 capital costs for the two
plants. The capital recovery factor of 0.10567 used for the annualized cost is
based on a 20 year useful life of the plant and an 8-1/2% interest rate. Costs
attributable to capital amount to $10.13 and $7.63/Mg of input waste for the
907 and 1814 Mg/d plans respectively ($9.19 and $6.92/ton).
241
-------
TABLE 56. 1976 CAPITAL COSTS FOR 907 AND 1814 Mg/d
(1,000 AND 2,000 TPD) PLANTS
Cost Element
Land
Site Preparation
Design
Construction and Installation
Real Equipment
Other Equipment
Contingencies (@ 10%)
Startup and Working Capital
Financing and Legal
Total Capital Investment
Annual Capital Cost (20 years, 8-1/2%)
Capital Cost, $/Mg
Capital Cost, $/ton
Cost, $ (000)
Smaller Plant
100
35
2,160
12,700
8,100
615
2,371
2,010
514
28,605
3,023
10.13
9.19
Larger Plant
130
46
3,030
19,300
12,400
808
3,571
3,025
775
43,085
4,553
7-63
6.92
Operating Costs
The following factors were used in developing the operating costs:
Labor (incl. benefits)
Fuel
Electricity
Water
Insurance, Fees, and Prof, services
Taxes
Maintenance and repairs (incl. labor)
Parts and supplies
Residue transportation and disposal charge
242
$7.00/h
$0.35/gallon
$0.02/kWh
$0.50/1,000 gallons
$1.00/input ton
0.75% of plant investment
7% of plant investment
0.75% of plant investment
$7.50/ton
-------
Table 57 shows the annual operating costs, indicating per Mg (ton) costs
to be $19.07 ($17.30) and $14.52 ($13.17) for the two capacity plants.
Revenues and Net Operating Costs
Table 58 shows possible revenues that might be realized from the two plants.
Material sales prices are those established by ORC market research and trans-
portation costs based on 25 percent of expected revenues have been allowed. The
pyrolytic oil sales price is estimated to be $1.75/GJ ($1.85/106 Btu) with
transportation costs of $0.008/dm3 ($0.03/gallon).
TABLE 57. 1976 ANNUAL OPERATING COSTS FOR 907 AND 1814 Mg/d
(1,000 AND 2,000 TPD) PLANTS
Cost Element
Labor
Fuel
Electricity
Water
Maintenance and Repairs
Parts and Supplies
Residue Disposal
Overhead and Mobile Equipment Operation
Property Taxes
Insurance, Fees, and Professional Services
Total
Operating Cost, $/Mg
Operating Cost, $/ton
Cost, $ (000)
Smaller Plant
1,604
10
832
56
1,826
195
193
444
195
329
5,684
19.07
17.30
Larger Plant
1,925
20
1,664
112
2,750
295
386
556
295
657
8,660
14.52
13.17
243
-------
TABLE 58. PRODUCT REVENUES
Material
Magnetic Metals
Glass
Aluminum
Oil
Revenue
$/ton
55
21
400
*
Shipping
$/ton
13.75
5.25
100.00
*
Fraction
Recovered
0.072
0.061
0.004
*
Revenue
$/ton input
2.97
0.96
1.20
7.24
Total
$12.37
* Based on $1.85/106 Btu, shipping costs for $0.30/106 Btu, and 4.67 x
106 Btu recovered/ton input.
Net operating costs can be summarized from the above as follows (assuming
np_ revenue from drop charges):
Capital Cost
Operating Cost
Total Cost
Revenue
Net
907 Mg/d (1,000 TPD) Plant 1814 Mg/d (2,000 TPD) Plant
$10.13/Mg
19.07
29.20
13.64
$9.19/ton
17.30
26.49
12.37
$7.63/Mg
14.52
22.15
13.64
$6.92/ton
13.17
20.09
12.37
$15.56/Mg $14.12/ton
$8.51/Mg $7.72/ton
244
-------
SECTION 10
CAPACITY-COST SUMMARY
Ideally, a master figure could be prepared indicating the costs of various
size plants associated with each type of processing technology. Two key fac-
tors detract from developing such presentations. The first concerns the matter
of ranking of systems that has been previously discussed. As an example, an
RDF plant, having no capital costs associated with it for the actual energy
recovery step, would appear to be the most economically attractive process on
such a multi-candidate figure. In reality, selection of the best facility for
a specific location is of course much more complicated than this. The second
difficulty involves the matter of the great variety of credits that can be
claimed for the fuel value, materials, and the charge to be made for refuse
disposal. Proper selection of values can tend to favor particular systems and
presents a bias in the apparent results presented.
Sufficient information exists within this report for construction of such
a figure by interested organizations. Interpretation of inter-system compari-
sons should be approached with caution unless consistent data based on actual
local information is used.
Some of the analyses made for the separate candidates do lend themselves
to a graphical presentation that does aid in a better understanding of waste-
to-energy systems. Such figures are presented within this Section.
Figure 43 shows the range of capital costs as a function of daily proces-
sing capacity for facilities for preparing RDF and using the supplementary fuel
in existing coal-fired boilers. Extrapolation somewhat above 2360 Mg/d (2600
TPD) can be made with rather good accuracy, although practical problems of
collection truck traffic congestion and queuing, waste availability, etc.,
occur near the point and individual studies must be made to establish true
local costs. Economics of scale do not necessarily continue and a 4720 Mg/d
(5200 TPD) plant would approximate the costs of two 2360 Mg/d (2600 TPD) faci-
lities so closely that consideration should be given to siting separate plants
near two centers of waste generation. Extrapolation below about 545 Mg/d (600
TPD) pan lead to high percentage inaccuracies in that the fixed costs of design
and construction, high unit equipment costs, and any degree of processing line
redundancy have great effects on unit costs.
The cost information shown in Figure 43 is based on specific studies con-
ducted by Parsons for 3 Mid-Western clients. Each of the originally-derived
design/cost packages was adjusted to a uniform set of assumptions of pricing
currentness, financial structure and fees, type of equipment, and operating
schedule. One facility, for example, was designed to contain an aluminum
245
-------
NJ
-fa-
28-
26-
24-
22-
B 20~
O 18-
l_- 14-
u 12-
H 10-
o.
5 8-
6-
4-
2-
0
200
400
I
600
800
1000
1200
1400
1600
I
1800
BOO
1000 1500
RAW REFUSE FEED RATE
2000 2200 2400 2600 TPD
2000 2SOO Mg/d
O
>
•a
-18 £
O
-16 8
-15 m
DO
-14 >
-13
-12
O
Hi I
O
-10 w
z
-9 O
-8
Figure 43. Supplementary fuel processing plant (RDF) capital costs.
-------
recovery processing step, and all costs associated with this sub-section were
eliminated so that ferrous metal (and RDF) was the only material resource
isolated at each plant. Another plant was sited at a distance form the power
plant and in this case the costs for trucks and loading facilities were deleted
to make the standard case one of an RDF plant immediately adjacent to the
power client. The assumptions used were as follows:
• Mid-1976 costs
• Daily processing capacity defined to be that possible by operating at
maximum throughput 16 hours per day (maintenance to be performed during
3rd shift and week-ends)
• Processing to consist of receiving, shredding to 100% less than 38 mm
(1.5 in.), removal of magnetic metals, air classification, and trans-
fer to adjacent power plant storage
• Power plant equipment sized to operate boilers continuously and to
consist of receiving piping, beehive-type storage and retrieval system,
and modifications to furnaces to handle supplementary fuels.
• No land costs are included because of the high degree of variability
throughout the nation
• Sum of interest during construction,vorking capital, and start-up costs
is 20% of remaining capital costs
• Design and construction management to be accomplished by a private firm
using ASCE rate structure
The above assumptions resulted in the following capital costs:
Capacity
Mg/d TPD Capital Cost, $
726 800 12 060 000
1742 1920 23 780 000
2322 2560 25 310 000
These values are plotted in terms of both the absolute capital cost and on a
daily unit weight basis. The dashed line added represents the function often-
times found to describe capital costs, where these costs are proportional to
the 0.6 power of the capacity. In the equation Si and 82 are the sizes of
two plants having costs of Cj and C2 respectively. From the limited data used
to develop this figure, the 0.6 power straight line could adequately describe
costs over the range of interest. The curve faired in represents the probable
realistic case where at lower capacity levels, fixed costs of design and high
equipment costs lead to increasing costs per unit and where at some middle
level the greatest change in the economies of scale occur. Eventually at some
247
-------
high capacity, a curve reversal might occur where the size of the operation
leads to cost inefficiencies.
The City of Chicago RDF facility can be used to illustrate the cost sen-
sitivity of varying throughput at a given size plant. The reliability gained
by redundancy and scheduled maintenance down-time is vital to the success of
an operation, but of course the cost of this reliability is reflected in higher
unit costs as processing capacity is reduced. Figure 44 shows this effect over
the range of 725 to 1633 Mg/d (800 to 1800 TPD).
Figure 45 shows probable capital costs for flash pyrolysis facilities for
producing liquid fuel from MSW over an input capacity range of 600 to 2200 TPD.
Until operating experience is gained at the El Cajon plant of Occidental Re-
search Corporation (ORC), the 0.6 exponent capacity-cost curve shown will
serve to estimate costs for such plants.
Within Figure 46 are shown adjusted capital cost estimates for several
waterwall incinerators. The square represents the 600 TPD Hamilton, Ontario,
plant; an additional 3% has been added to the $15.4 million new cost discussed
in this report to adjust that value to mid-1976. The point within the triangle
is based on costs associated with the 720 TPD plant of Nashville Thermal Trans-
fer Corporation. From their original cost of $16.5 million has been deducted
$4 million for the distribution piping network and an inflation factor of 1.24
then applied. The current expenditures of $5.7 million were then added to
yield a probable 1976 cost of $21.2 million. The circles represent 12% in-
flated adjustments made on values within this report for facilities based on
the RESCO, Saugus, design. The 1200 and 1500 TPD values at the same cost re-
sult from the fact that the lower capacity is a nominal one and the possibility
exists that the present plant might be capable of operating at 1500 TPD. The
curve shown for the capital cost per daily ton is based on the smoothed curve
indicating costs are proportional to the capacity to the 0.85 power. Several
other costs available to Parsons fall reasonably well on this same curve.
The capital cost-capacity curves for the two pyrolytic gasifiers dis-
cussed in Sections 7 and 8 are shown in Figure 47. The plotting of Torrax
and Purox costs on the same figure illustrates the care with which such figures
must be read. While it would appear that the six points used in this curve
would fit the same equation, the values used for the Torrax system represent
costs for facilities whose output is steam, while the Purox plants produce a
medium heating value gas requiring additional facilities for energy utilization.
248
-------
o
15-
15'
10-
S 10
u
5-
CAPITAL + O&M
OPERATION AND MAINTENANCE
NET AFTER CREDITS
CAPITAL AMORTIZATION
1
0 500
1000
1500
2000
500
1000 1500
RAW REFUSE FEED RATE
TPD
2000 Mg/d
Figure 44. Costs vs. processing throughput for existing facility
(City of Chicago Supplementary Fuel Plant).
-------
50-4
- 35
45-
40-
(B
O>
*—
g
_j
i
30-
3
20-
h30 |
r-
o
tn
33
1-25 |
-I
O
-20
15-
10-
I
200
400
600 800 1000
500
1000
RAW REFUSE FEED RATE
1200 1400 1600 1800
I
1500
2000
2200
I
2000
TPD
Mg/d
Figure 45. Capital cost vs. capacity of flash pyrolysis facility.
-------
60-1
55-
50-
45-
s
B 40 H
o
li
-I
I
« 35-
fe
8
*30H
g
25-
20-
-35
o
>
TJ
n
O
•D
m
-3,1
O
-1
X
o
I
-25
15-
200
400
600
800
500
I
1000
1200
1400
1600
1800
2000
1000
RAW REFUSE FEED RATE
1500
TPD
Mg/d
Figure 46. Capital cost vs. capacity of waterwall incinerators.
251
-------
to
01
Ki
O
Q.
4
CJ
80 -t
70-
60-
50-
40-
30-
20-
10-
© PUROX, SYNGAS PRODUCT
[3 TORRAX, STEAM PRODUCT
0
200
400
600
800
1000
1200
1400
1600
O
>
O
40 O
c/i
TJ
m
35 3)
§
-30
-25 «>
H
O
I
D
500
1000 1500
RAW REFUSE FEED RATE
1800 2000 2200 2400 TPD
2000 Mg/d
Figure 47. Pyrolytic gasifier capital costs.
-------
SECTION 11
GENERAL CONCLUSIONS
Conclusions and recommendations developed during this project are dis-
cussed in this Section under three categories: Planning Factors, General Sys-
tem Needs, and Waste Material Utilization. The first concerns policy and
planning matters affecting waste-to-energy technology while the second dis-
cusses areas that apply to more than one system. The third lists conclusions
on how the characteristics of U.S. wastes influence their utilization as energy
sources. Conclusions peculiar to a specific facility or type of conversion
system have been presented within Sections 3 through 9,
PLANNING FACTORS
Issuance of a government document presenting probable interactive effects
of economic and trade factors on energy costs is recommended.Private organi-
zations and local governments must prepare an expenditure/revenue analysis
over a 10 or 20-year period to determine the relative advantages of a waste-to-
energy conversion system versus alternative disposal means. Income from the
sales of energy will be an important source of revenue. The local planner now
faces difficulties in locating officially sanctioned and consistent forecasts
of absolute future costs of alternative energy forms and long-term revenue
forecasts are therefore difficult to make. However, sensitivity analysis
would at least reveal the relative effect of a variety of economic factors on
projected cash flow for the several types of waste conversion facilities.
A manual entitled "Standardized Methodology for the Cost Estimation of
Commercial-Scale Waste-to-Energy Qnwj~[£SJ£n±J>^^ great
benefit to a number of organizations. This document must be more than an
accounting system for establishing uniform cost factors. It should enable a
developer utilizing the recommended detailed procedures to approach a client
with sufficient information to permit an unbiased economic comparison of alter-
native conversion processes.
To date there has been very little data published on environmental emis-
sions and effects from conversion plants. Past_and_on-going environmental test_
results should be immediately documented and issued to those concerned, along
with applicable regulations and a brief review p_f_die__bje_sj_j.vailable control
technology. Current projects for compiling information andjmaking new^measure-
ments should be continued and reports issued whenevernew types^ of facilities
or systems become operational. Regulatory hearing boards issuing permits for
new conversion plants often require comparisons of environmental effects of
alternative disposal methods. Such information, including data for sanitary
landfill operations as well as health and safety considerations, should be
made available in a summary document.
253
-------
Nontechnically trained personnel are often required to make key decisions
as to whether a new conversion facility should be financed and constructed.
A document should be prepared to assist decision-makers in understanding the
critical factors to be examined, relative advantages of various alternatives,
and the ramifications of their decisions.Summary reviews of technological,
environmental, and financial considerations should be included.
Even the expert in waste-to-energy conversion technology must presently
review an inordinate quantity of information to maintain his skills. A criti-
cally edited comprehensive annual publication by EPA and ERDA would signifi-
cantly reduce the burden of literature review; the document should not be res-
tricted to existing government-funded projects.
GENERAL SYSTEM NEEDS
Front end processing prior to obtaining energy from waste materials is
largely adapted from other industries. Much remains to be accomplished in
establishing the proper type of equipment to be used and in improving effi-
ciencies . Development activities in the areas of storage and retrieval, size
reduction, drying, transfer, and component separation must be pursued to in-
crease throughput for a given energy expenditure and to recover higher purity
material fractions. In most cases, initial research should be conducted using
100% paper or similar material in order to learn more about the fundamental
controlling parameters of the process. Water content should then be changed
over a range of "0" (dried at 100°C) to 40% to establish the effect of this
common variable. Only after these tests have been completed should greater
heterogeneity be investigated through incorporation of the other components
typical to the particular waste. Investigations should include:
• Concepts for improved receiving and initial transfer of wastes to re-
duce capital, operating, and maintenance costs. Bridging of raw wastes
in storage, and belt transfer at high unit loadings are particular
problems.
• Improvements in equipment for the storage and retrieval of RDF and
means of controlled introduction into furnaces as a function of heating
fraction. A method of continuous flow measurement is also needed.
• Means of reducing energy consumption and high maintenance costs of
size reduction equipment. Current investigations into improving the
technology should be continued and safety (fire and explosion) aspects
further considered.
• Means to avoid empirical adjustment of air classification systems for
isolating the combustible fraction of wastes to give the best possible
yields of RDF. The systems are expensive and need improvements in
both the percentage recovery and purity of the fractions.
• Other processes for separation of mixed wastes into their components,
not necessarity intended for energy recovery, that would influence
overall plant economics and improve efficiency.
254
-------
• Partial removal of water (down to 2 to 10%) from the fuel to Increase
boiler efficiency. Studies need to be conducted to improve the effi-
ciency of dryers for waste materials.
The characteristics of various waste-derived fuels are not yet sufficient-
ly well defined to permit a meaningful comparison of their relative advantages.
An independent laboratory should be engaged to determine the physical, mechani-
cal, chemical, and biological properties of these fuels under conditions of
transport, storage, and usage.The evaluation should include RDF as manufac-
tured; RDF/coal and RDF/oil blends; agricultural waste-derived fuels from
various sources and having varying physical properties; and liquid fractions
from pyrolytic systems. Stability, corrosive properties, health effects, and
applications in energy-releasing equipment must all be studied. It is recom-
mended that research be conducted to modify any adverse properties that may
exist. ,
Preengineered design of transportable sections of a modular processing/
production facility often offers important capital cost savings. Economies
of scale and an experienced work crew at the manufacturing plant reduce costs
and eliminate expensive custom engineering at the client's site. While some
waterwall combustion systems employ this practice to an extent, the concept
has not generally been applied to the waste-to-energy conversion field. It is
Parsons' conclusion that an engineering analysis be conducted to determine
whether significant cost savings from modular design might indeed result.
The design of pyrolytic conversion equipment is hampered by a lack of
relevant data on the complex chemical reactions involved and the effects of
physical variations on this chemistry. This often leads research workers in-
vestigating new variations of a waste pyrolysis process to assume their methods
will offer dramatic breakthroughs in developing new energy sources. Such
claims typically have proven unrealistic when scale-up tests reveal actual
yields and thermal conversion efficiencies. Several EPA-sponsored projects
are now in progress to obtain further information for the generalized pyrol-
ysis process. Other R§D related to specific systems is being conducted at
industrial and university laboratories. Data derived from these studies will
also contribute to further knowledge of the field. On the basis of Parsons'
surveys, it is concluded that much remains to be accomplished in pyrolysis
technology and that government-supported investigations should continue for
a number of years.Periodic publication of up-to-date information on the
nature of the waste-to-energy market and factors leading to commercial profita-
bility will encourage private research.
The RSD recommended above is directed to development of fundamental infor-
mation on pyrolysis chemistry. Applied pyrolysis studies are also needed to
advance the state of the art. It is recommended that the following be pursued:
• The effect of mixed types of wastes on pyrolytic yields and process
economics^should be established.MSW, sewage, commercial, industrial,
and agricultural/forestry wastes should be used as feedstocks in vary-
ing ratios (and water content) to determine operational limits and to
provide direction for future commercial activities.
255
-------
• The effect of wastes as additives to present coal conversion systems
should be determined.
• Superior methods for introduction of wastes into pyrolytic reactors,
particularly those such as fluidized bed systems operating above atmos-
pheric pressure, must be developed. Likewise, removal of char or slag
remains as a problem.
• Use of pyrolysis for processing 10 to 100 Mg/d (11 to 110 TPD) wastes
should be investigated.
• Alternative methods of protection against the temperature extremes and
the corrosive pyrolytic products should be investigated. The thermal
and chemical environment within a pyrolytic reactor has led to selec-
tion of costly construction materials and techniques.
• Combustion devices designed to operate with relatively heterogenous
fuels must be further developed. Those pyrolytic systems yielding
hot off-gases containing suspended solid and liquid combustibles oper-
ate most efficiently when the heterogeneous fuel is immediately com-
busted and the total heat recovered in an exchange system.
Although often associated with the PUROX pyrolysis system, the research
required to establish process economics of converting pyrolysis off-gases to
methane, methanol, or ammonia actually applies to a broader area. The chemis-
try (and economics) involved in converting mixtures of carbon monoxide and
hydrogen to the three named compounds is quite sensitive to the ratio of CO
to \\2 and to the presence of the usually simultaneously present carbon dioxide,
water, hydrocarbons, and sulfur compounds. Much is known of this chemistry,
but pyrolytic off-gases, including those from oxygen-blown reactors, have com-
positions markedly different from the syngas used to date. No accurate pre-
dictions can be made regarding the effect of these differences on catalyst
life, degree of purification required, equipment size, and product yield. To
investigate all controlling parameters for the possible range of pyrolytic off-
gases would be very expensive and therefore is not recommended. What is re-
commended is the statistical design of a test program that would use the PUROX
gas as a base case and permit adjustments around that composition to learn the
principal effects of other off-gases on the economics of conversion to methane,
methanol, and ammonia. Use of an essentially pure mixture of hydrogen and
carbon monoxide, admittedly a far less expensive approach, would result in
meaningless data because of unknown effects of impurities. The study of the
synthesis can be carried out on a benchscale or small pilot plant level since
scaling effects from such research are well established.
Similar to the above synthesis research is the activity devoted to chemi-
cal modification of the waste material and its decomposition products prior
to isolation of a syngas. Many of the reactions taking place are the same as
would occur in the post-treatment mode, but the concept here is to use chemical
reagents to directly yield new gaseous or liquid fuel mixtures from solid
cellulose. Such work is typified by the BuMines research on solid waste con-
version (with CO and/or H2) at moderately high temperatures and pressures.
While low process costs have yet to be demonstrated in the limited laboratory
256
-------
work conducted by several organizations, because of the versatility inherent
in this synthesis concept, particularly the potential for making valuable
liquid products, support of additional work is warranted. This R£D should
consist of further small batch autoclave experiments to learn more about pro-
cess variables, followed by scale-up, perhaps at the ERDA Albany, Oregon faci-
lity, to a continuous reactor capable of processing several hundred pounds an
hour.
It is assumed that EPA will play a continuing role in R£D for overall im-
proved pollution control technology and that research need not be specified
here for assuring that superior equipment and processes become available. The
increasing requirement for the combustion of coal alone will cause R^D to con-
tinue at a significant rate. In very few cases does waste processing create
emissions that might not prove controllable through developments needed by
other industries. In one case, additional testing is recommended to ascertain
if the alkaline ash from wastes tends to chemically absorb sulfur oxides and
if any undesirable properties result from mixed coal and waste ash. In another
area somewhat unique to waste processing, tests should be made to assure that
pathogenic dusts can be adequately removed by the fume hoods and fabric filters
now being specified.
Although the great difficulty in applying generalized data to specific
needs is widely recognized, some decision-making bodies too often ignore this
problem. Additional information on the options available and key operating
characteristics of systems can be found in selected documentation. However,
local/regional studies and analyses must be conducted before construction of
a waste-to-energy processing plant is seriously considered. To rank the can-
didate systems covered in this report on the basis of national average para-
meters is meaningless for local planning. Any group wanting to determine the
feasibility of incorporating an energy/material recovery facility into its
waste management plans must first ascertain whether its own staff is qualified
to perform the necessary analysis, and, if it is not, should retain an expe-
rienced consulting organization.
With biological energy conversion systems, fuels (methane and alcohols
in particular) can be obtained from high water content wastes without the
need for energy-consuming drying processes. Economically successful anaerobic
digestion operations to date have typically employed "free" labor, have used
designs and materials of construction unacceptable to industrialized nations,
and have been used where small quantities of fuel gas are required as an alter-
'native to very high price energy forms. Continuing increases in natural gas
prices and the important advantages of gaseous fuels will result in some lim-
ited methane generation applications on U.S. farms. Support will be needed
for final design and demonstration testing of equipment best suited for this
purpose. The feasibility of large-scale anaerobic digestion of wastes will
be better understood after completion of tests at Pompano Beach (MSW) and at
the Oklahoma plant of Thermonetics, Inc. (animal wastes). Present analysis
indicates that a rather high drop charge is required to offer the methane at
competitive prices and that disposal of the final sludge can present problems.
In mid-1978 a definitive report on the practicality of commercial methane pro-
duction by anaerobic digestion could be prepared. This document should summa-
rize the past and current laboratory work and present engineering conclusions
on production facilities in operation at the time.
257
-------
For many years investigators of hydrolysis/fermentation of cellulosic
wastes have believed that their particular process would result in competi-
tively priced fuel-grade ethanol. Only in rare cases, where waste disposal
costs are extremely high, has production of alcohol proven economically feas-
ible. Because of the large amounts of agricultural wastes in the U.S. and
the ready application of alcohol as a fuel for vehicles and gas turbines,
it is recommended that a low level of fermentation research be maintained
should a significant technological advance occur.
Coal has been used as an industrial fuel for more than one hundred years
and yet research on improved means to efficiently release and use its energy
continues at a high level. Although the high rate of improvement experienced
with coal in the last 50 years cannot be expected with waste-derived fuels,
it appears that much will be learned about superior combustion of such fuels
in the near future. Recommended areas of combustion research for waste-derived
fuels are listed and described below:
• Conduct detailed experiments at new facilities such as those at Ames,
Saugus, and Chicago. Make measurements on gas and ash compositions,
deposition of slag and soot, corrosion and erosion of steam generating
tubes, and observations of flame characteristics.
• Through adaptation of an existing RDF-fired furnace or construction of
special R§D equipment, study effects of injection port geometry, size,
and location; injection air-to-RDF ratioj and varying compositions of
coal and waste.
• Establish the best means to use solid waste-derived fuels as a heat
source for existing oil and gas-fired furnaces. (Note: This need not
imply that the solid fuel must be burned within the present furnace;
external heat recovery systems should be considered.)
• Continue R§D to improve grate-supported (mass burning) combustion
systems. The inherent simplicity of these systems offers a significant
advantage, but they still experience problems of incomplete burn-out,
corrosion, erosion, and nonsteady steam generation.
• Conduct RSD on small energy-recovery incineration to minimize problems
of automatic feed, high maintenance, and excessive emissions. This
would permit profitable use in major commercial and high density housing
projects.
WASTE MATERIAL UTILIZATION
The principal conclusions to be drawn regarding U.S. waste materials are
that (1) the total quantity (and heating value) is impressively large, and
(2) the fraction considered economically available for energy purposes cannot
yet be accurately defined. Present survey information on total national waste
quantities is sufficiently accurate for most purposes. Studies have revealed
that several of the methods popularly employed for estimating availability can
lead to erroneous values. With one method, usually applied by workers still
conducting basic laboratory research, a statement is made such as "the very
258
-------
conservative assumption has been used that only 50 percent of the total agri-
cultural wastes will actually be available." This phrasing has led some liter-
ature reviewers to believe the assumption is truly conservative, when actually
it has no factual basis. Another estimating method assumes all construction
of processing facilities in proportion to the metropolitan population. This
method does not take into account the costs of local alternative disposal, the
need for particular-types of energy forms, financing abilities, or specific
environmental pollution problems. Again, the fraction to be employed is ill-
defined.
While the percentage of total energy needs that could be supplied by U.S.
wastes is highly speculative, even a cursory analysis leads to the conclusion
that futher development of conversion systems should be encouraged. If only
2 percent (a reasonable amount from existing information) of anticipated 1980
energy needs were supplied, this would amount to an annual value of $4.38
billion.
Additional waste quantification at the local and regional level is needed
by governments and organizations considering incorporation of resource recovery
facilities into their waste management systems. Specific local studies must
be conducted on current and projected quantities and compositions of waste
throughout the year, transportation costs, and the nature of the markets for
materials and fuels. Analyses should be continued on the technological and
economic factors affecting the availability of wastes for conversion to energy
forms. Availability of agricultural (farm, logging, and feedlot operations)
wastes are most difficult to determine and prime attention should be directed
to this area. In addition to analytical studies of collection and transporta-
tion costs and the best means of utilizing the energy contained in these wastes,
experiments should be conducted on the efficiency of equipment to gather agri-
cultural wastes and the long-term effect of removing these materials from
their normal place of generation.
Parsons has examined a number of samples of pelletized agricultural
wastes of excellent mechanical integrity, with heating values 50 to 100 percent
higher than the original waste because of water loss and chemical modifications
during the pressing operations. This approach yields a fuel having a higher
boiler efficiency than the original wet waste and which would minimize trans-
portation costs to a user. Further R£D on the densification of agricultural
wastes and characterization of physical and combustion properties is recom-
mended.
Waste materials are primarily cellulosic in nature. Essentially all con-
siderations for obtaining useful energy from these wastes ultimately use total
oxidation of the carbon and hydrogen (whether directly or through burning of
pyrolytic syngas or anaerobically formed methane). Analysis should be direc-
ted toward chemical synthesis from cellulose. Such a process might be more
attractive than combustion from an overall energy standpoint.
Materials recovery is outside the scope of this report. However, it is
impossible to ignore the importance of this area as the required sales price
of energy forms is a function of the total resource recovery plant economics.
R§D of superior materials recovery processes is essential and should be fully
coordinated with waste-to-energy conversion activities.
259
-------
GLOSSARY AND ABBREVIATIONS
ACFM: Actual cubic feet per minute (as opposed to mathematical correction to
a standard temperature and pressure)
Anaerobic digestion: Conversion of organic material through the action of
natural micro-organisms in the absence of air; methane i$ the product of
interest for energy consideration.
BOD: Biochemical Oxygen Demand; a standardized biological method of measuring
the quantity of organic contaminants in water capable of consuming oxygen.
CFM: Cubic feet per minute
COD: Chemical Oxygen Demand; a standardized chemical oxidation method of
measuring the quantity of organic contaminants in water capable of con-
suming oxygen.
DCS: Dry combustible solid
Cubic decimetre, the SI term for the former metric liter; see SI expla-
nation in Appendix E.
EES: Engineering Experiment Station of Georgia Institute of Technology
Fossil fuel: Fuels naturally formed over extensive time periods by the "fos-
silization" of large organic (plant) deposits; includes the various coal
grades, petroleum-derived liquid fuels, and natural gas.
GPD: Gallons per day
GPM: Gallons per minute
HHV: Higher heating value; the heat released upon complete oxidation (combus-
tion) , including the heat of condensation of water vapor.
Hogged fuel: Chopped or shredded wood or other fibrous fuel
IR§T: International Research and Technology Corporation
LHV: Lower heating value; the heat released upon complete oxidation (combus-
tion), not including the heat of condensation of water vapor. Most Euro-
pean work is in terms of LHV.
260
-------
Mass burning: As opposed to "suspension" burning, the fuel combustion occurs
on any of several types of support grates, typically moving.
Mg: Megagram (one million grams or one thousand kilograms), the SI expression
for the former metric tonne (2204.6 Ib); see SI explanation in Appendix E.
MJ: Megajoule, a million joules of energy, equivalent to 947.8 Btu; see SI
explanation in Appendix E.
MRI: Midwest Research Institute
MSW: Municipal solid waste
Mm . Normal cubic meter, the standardized SI gas volume (0°C and 1 atm), equal
to 37.33 standard cubic feet (60°F and 1 atm). See SI explanation in
Appendix E.
ORC: Occidental Research Corporation
Pa: Pascal, the SI unit of pressure; 1 psi equals 6 895 Pa. See SI explana-
tion in Appendix E.
Pyrolysis: Thermal decomposition of organic materials in the absence or near
absence of gaseous oxygen.
RDF: Refuse-derived fuel
RESCO: Refuse Energy System Co., owner-operator of the combustion system at
Saugus, Massachusetts
SCFM: Standard (60°F, 1 atm) cubic feet per minute
SI: le Systeme International d'Unites (modernized metric); see Appendix E
for explanation of specific units and format
SNG: Synthetic natural gas
Supplementary fuel: Fuel, typically derived from wastes, used to supplement
some fraction of fossil fuel in a furnace.
SWARU: Solid Waste Reduction Unit, Hamilton, Ontario, Canada
Syngas: Synthesis gas, a mixture of hydrogen, carbon monoxide, light hydro-
carbons, and carbon dioxide
THERMAL: Nashville Thermal Transfer Corporation
TJ: Terajoule, a trillion (1012) joules of energy, equivalent to 947.8 million
Btu; See SI explanation in Appendix E.
TPD: Tons per day
261
-------
TPH: Tons per hour
TPY: Tons per year
Waterwall boiler: A steam generator in which the combustion zone and hot gas
passes are surrounded by structural walls consisting of a large number of
pipes filled with flowing water absorbing the heat; fossil or waste fuels
may be used.
262
-------
REFERENCES
I. Rexnord, Inc., "The Deep Receiving Pit Concept," Waste Age, March, 1976.
2. Neville, Charles B., and B.A. McDermott, "How District Heating/Cooling
and Solid Waste Disposal Became Part of a Downtown Urban Renewal Project,"
Specifying Engineer, February, 1976.
3. R.W. Beck and Associates, Engineers and Consultants, "Feasibility Report,
Additional Financing for Nashville Thermal Transfer Corporation, Nash-
ville, Tennessee," October, 1975.
4. "Can Nashville Thermal Live Up To Its Original Promise?", Resource Re-
covery and Energy Review, 10-12, Mar/Apr 1976.
5. Engdahl, R.B., "Identification of Technical and Operating Problems of
Nashville Thermal Transfer Corporation Waste-To-Energy Plant," Battelle
Columbus Laboratories Report No. BMI-1947 to U.S. Energy Research and
Development Administration, February 25, 1976.
6. Zralek, R. and E. Bailey, "The City of Chicago and Commonwealth Edison
Company's Watts From Waste Program," Proceedings 1976 National Solid
Waste Processing Conference.
7. "Recovery I ... A Progress Report," National Center for Resource Re-
covery Bulletin, Vol. VI, No. 6, 35-41, Spring 1976.
8. Holloway, J.R., "EPA Resource Recovery Demonstration: Summary of Air
Emissions Analyses," Waste Age, 50-52, August 1976.
9- Shannon, L.J. et al, "St. Louis/Union Electric Refuse Firing Demonstration
Air Pollution Test Report," U.S. EPA Office of Research and Development,
August 1974.
10. Knight, J.A., "Pyrolysis of Pine Sawdust," Presented at the National
Meeting of the American Chemical Society, San Francisco, September 2, 1976.
11. Tatom, J.W., et al, "Clean Fuels from Agricultural and Forestry Wastes -
The Mobile Pyrolysis Concept," Winter Annual ASME Meeting, Houston, Texas,
Nov. 30-Dec. 4, 1975.
263
-------
12. Tatom, J.W-, et al, "Parametric Study for a Pyrolytic System for Produc-
tion of Fuels from Agricultural and Forestry Wastes," 10th Intersociety
Energy Conversion Engineering Conference, Newark, Dela., 1975, Proceedings,
New York, Inst. of Electrical and Electronic Engineers, 1975.
13. Legille, E., F.A. Berczynski, K.G. Heiss, "A Slagging Pyrolysis Solid
Waste Conversion System," Conference Papers, First International Conference
on the Conversion of Refuse to Energy, Montreux, Switzerland, Nov. 1975,
p. 232.
14. Thome-Kozmiensky, Karl J., "Neu Technologien zur Abfallbeseitigung Das
Andco-Torrax-Ver fahren," D Bohn, p. 144, Erick Schmidt Verlag 1977.
15. "Resource Recovery and Waste Reduction," Third Report to Congress, Office
of Solid Waste Management Programs, U.S. EPA Publication SW-161, 1975.
16. Levy, Steven, J., "San Diego County Demonstrates Pyrolysis of Solid Waste,"
U.S. EPA Report SW-80d.2, 1975.
17. Preston, G.T., "Resource Recovery and Flash Pyrolysis of Municipal Re-
fuse," Occidental Research Corporation, Presented at the Institute of
Gas Technology Symposium, Orlando, Fla., Jan. 1976.
18. Morey, B., "Inorganic Resource Recovery and Solid Fuel Preparation from
Municipal Trash," Proceedings Fourth Mineral Waste Utilization Symposium,
84-95, Chicago 1974.
19. MaiIan, G.M. and E.I. Titlow, "Energy and Resource Recovery from Solid
Wastes," Washington Academy of Sciences Symposium, College Park, Maryland,
March 1975.
264
-------
APPENDIX A
WASTES IN THE UNITED STATES
INTRODUCTION AND SUMMARY
An appreciation of what wastes there are, their economic availability,
heat, content, and other characteristics is essential to the planning or eval-
uation of waste-to-energy facilities. To provide an introductory understand-
ing, national data based on a literature review and comments by experts are
summarized and discussed in this section. None of these numbers were origi-
nated by The Ralph M. Parsons Company, although data manipulations were made
to permit comparisons. These figures should not be used as the basis for
designing specific facilities because national averages may bear no relation-
ship to local conditions. Instead, this data should be regarded as a general
introduction to waste quantities and characteristics that will establish a
foundation for the waste-to-energy process evaluations that follow.
Although data on wastes have been collected for many years, the early
definitions of wastes and the measurement techniques were poor, leading to
quantity figures of low accuracy. With the present need to find alternatives
to landfills, and aided by modern data collection techniques, recent investi-
gators have compiled data that are greatly improved, and these newer figures
are used throughout this report.
An attempt has been made to state waste quantities here in terms of dry,
combustible, weights available for potential conversion to energy. Data on
municipal solid waste, however, are historically given as total collected
weight, including moisture and non-combustibles. Where comparisons are being
made with other waste quantities, both methods of specifying the quantity of
municipal solid waste will be used. The subject of detailed technical and
economic availability is one that must be evaluated on a local basis, balanc-
ing the cost of collecting and transporting the waste against the revenue
from its use.
The three types of waste discussed in this chapter are municipal, indus-
trial, and agricultural. A fourth major type of waste, that from mining
operations, is outside the scope of this study. Tabular presentations of the
data, along with the information source, are made in the sections dealing with
the detailed information on each waste category that follow the summary narra-
tive below.
A-l
-------
Municipal Wastes
The two components of municipal waste are municipal solid waste (MSW) and
sewage.
MSW includes household waste, commercial and institutional waste, and
city street sweepings and primings. It has been estimated by Smith (Ref. A-l)
that in 1971, 1.50 kg per capita per day (3.31 Ib/c/d) were collected, total-
ling 113.4 x 106 Mg (125 million tons) per year, including moisture and
incombustibles. About one-fourth of that weight is natural moisture, and
approximately another fourth is glass, metal, and other incombustibles. The
available, dry, combustible weight of municipal solid waste was therefore
61.1 x 106 Mg (67.4 x 106 tons) in 1971. Paper is the most common material in
MSW, with yard and food wastes a distant second and third.
The total weight of MSW, including moisture, metal, and glass, is fore-
cast by Midwest Research Institute (MRI) (Ref. A-2) to grow from 113.4 x
106 Mg/y (125 x 10& TPY) in 1971, to 144.4 x 10$ Mg/y (159.2 x 10$ TPY) in
1980, and 182.2 x 106 Mg/y (200.8 x 106 TPY) in 1990. The corresponding quan-
tities of dry, combustible MSW are 79.9 x 106 Mg/y (88.1 x 106 TPY in 1980
and 105.6 x 106 Mg/y (116.4 x 106 TPY) in 1990.
The average amount of energy contained in raw MSW is estimated to be
10.47 MJ/kg (4500 Btu/lb), sufficiently high to permit unprocessed waste to
be used as a fuel. When a portion of the moisture and incombustibles is
removed in presently available equipment, the heating value of the combustible
fraction isolated is in the range of 13.72 to 14.42 MJ/kg (5900 to
6200 Btu/lb). MRI, in testing 97 samples at the St. Louis-Union Electric
project, found the average as-received heating value to be 10.64 MJ/kg
(4573 Btu/lb), the moisture-free value to be 14.49 MJ/kg (6231 Btu/lb), and
the moisture and ash-free value to be 20.57 MJ/kg (8843 Btu/lb). Depending
on the process used, more energy may be consumed making dry, combustible
solid waste than can be justified on the basis of the more convenient fuel
prepared.
The heating value of municipal solid waste is expected to increase,
primarily due to an increase in the use of plastics.
Sewage sludge, the second component of municipal waste, contains human
and food wastes, residuals from wash water, and in many areas, treated and
untreated industrial wastes. Generation rates of sewage sludge are rising
due to increased disposal of food and other wastes in sewage, and to legisla-
tion requiring more thorough sewage treatment, which adds inorganics to the
waste water. It has been reported by International Research £ Technology
(IR&T) (Ref. A-3) that there were, on a dry basis, 11.5 x 106 Mg/y (12.7 x
106 TPY) of sludge in 1970. They estimate this quantity will increase to
14.0 x 106 Mg/y (15.4 x 106 TPY) in 1980 and 14.0 x 106 Mg/y (18.2 x 106 TPY)
in 1990. Not all of this is combustible, because it contains some dirt and
treatment chemicals. Dry sewage sludge has a high heating value, but in its
A-2
-------
usual state it has such a high water content that it cannot be burned alone.
Other processes for obtaining a fuel gas from sewage sludge are discussed
later in this report.
Industrial Wastes
Industrial wastes, including processing wastes and plant trash, was esti-
mated to total 93.4 x 106 Mg/y (103 x 106 TPY) in 1965. Less than half of
that, about' 41.7 x 106 Mg/y (46 x 106 TPY) was dry, combustible, solid (DCS)
waste. The wood, paper, and allied industries produced the greatest quantity
of process waste, but they are also very active in seeking ways to avoid or
utilize these wastes. The construction and demolition industries are the next
major producers of process wastes. These two industrial groupings were
estimated to produce 88% of the DCS process waste in 1967, or 54% of DCS
industrial waste including plant trash.
Total DCS industrial wastes are forecast to grow slowly, from 42.3 x
106 Mg/y (46.6 x 10& TPY) in 1967, to 48.2 x 10^ Mg/y (53.1 x 106 TPY) in
1980, and 50.0 x 106 Mg/y (55.0 x 106 TPY) in 1990. A decline in wood and
paper industries wastes is expected to be offset by an increase in plant trash,
chiefly wooden shipping cases, cardboard, and paper.
The average heating value of DCS industrial waste has been estimated to
be 20.88 MJ/kg (8976 Btu/lb).
There are also gaseous and liquid industrial wastes. One estimate puts
the total quantity of wet waste liquids and sludges at 15.4 x 10^ Mg
(16.98 x 10^ ton) in 1970, mostly from the chemical and machinery industries.
No reference was found on the total quantity of gaseous industrial wastes.
Agricultural Wastes
Agricultural wastes include crop, livestock, and forestry wastes, all of
which are candidates for some form of conversion to energy. The highest
estimate of the wet-basis quantity of agricultural wastes is a total of
1941 x 106 Mg/y (2140 x 10° TPY) for 1966 (Ref. A-4) . Livestock wastes
accounted for 73 percent of the total in this estimate; crop wastes accounted
for most of the remainder. Many of these wastes are scattered, livestock
manures in particular, so that they are not readily available for energy con-
version. Estimates of the dry, available, wastes in 1970 range from 210.2 x
106 (Ref. A-3) to 585 x 10$ (Ref. A-5) Mg/y (231.7 x 106 to 645 x 10$ TPY).
Based on the lower figure, these are forecast (Ref. A-3) to grow to
316.2 x 106 Mg/y (348.5 x 10& TPY) in 1980 and 385 x 106 Mg/y (424 x 10° TPY)
in 1990 (available DCS wastes).
Heating values of DCS agricultural wastes range from 13.96 to 19.42 MJ/kg
(6000 to 8350 Btu/lb).
Total Wastes
The total dry combustible wastes, summed from the data given in the pre-
ceding paragraphs, is shown in Table A-l. The annual combustible waste
A-3
-------
TABLE A-l. TOTAL WASTE QUANTITIES IN THE UNITED STATES
Year
Waste Stream
Municipal
> Municipal
*• Solid
Waste
Sewage
Sludge
Industrial
Agricultural
Dry Combustible Waste
(From Various Sources Compiled
in this Report)
Early 1970's 1980 1990
SI* Eng.* SI Eng. SI Eng.
61.1 67.4 78.0 86.0 98.0 108.0
11.5 12.7 14.0 15.4 16.5 18.2
42.3 46.6 48.2 53.1 49.9 55.0
210.2 231.7 316.2 348.5 384.6 424.0
Eliassen
(Ref. A-4)
Wet Total
Wastes
1967
SI Eng.
232.2 256
-
99.8 110
1919 2115
EPA First
Report to
Congress
(Ref. A-6) Wet
Total Wastes
1971
SI Eng.
209 230
-
127 140
2159 2380
IR$T (Ref. A- 3)
Dry
Combustible
1967 - 1970
SI Eng.
103.6 114.2
11.5 12.7
41.2 45.4
212.3 234.0
TOTAL
325.1 358.4 456.3 503.0 549.0 605.2 2251 2481 2495
2750 368.6 406.3
*SI units throughout are Tg/y and English are in millions of tons per year.
-------
generation rate is expected to increase by 40% between 1970 and 1980, and by
20% between 1980 and 1990. This declining change in the waste generation rate
is expected because of current efforts to reduce industrial and agricultural
waste generation, and to find uses for those wastes that are generated, thus
removing them from the waste category.
For comparison, the results of some earlier estimates are also shown.
Rolf Eliassen (Ref. A-4), in a report prepared for the Executive Office of the
President, made an estimate of the total quantity of waste based on very
limited data. The estimate overstates the quantities of wastes because some
of the factors used were unmeasured estimates, moisture is included, and the
agricultural quantity is based on the total number of animals, not just those
whose wastes are concentrated. The EPA's First Report to Congress (Ref. A-6)
essentially factored up Eliassen's numbers from 1967 to 1971, with some modi-
fications where newer information had cast doubt on the earlier numbers.
International Research and Technology (Ref. A-3) used these and other earlier
estimates, but cross-checked and refined them by making an analysis covering
each step of the production and consumption process. They were also the first
to give their estimates in terms of totally dry weights. In some respects
this can be misleading, because the weight of wastes that must be handled
include moisture. However, the moisture content varies so much between waste
types that the estimates of the total quantity are more distorted by consid-
ering the wastes with moisture than by considering them dry.
Some of the early estimates, particularly Eliassen's and Anderson's
(Ref. A-5), were widely quoted and used inappropriately. Care should be taken
when considering the quantities of wastes to use the most recent estimates.
The older the estimate, the more likely it is to have a large, inflationary
"guess" factor.
MUNICIPAL WASTES
Classification System
More data have been compiled for municipal wastes than for any other
major type of waste. These data are usually reported in terms of origin or in
terms of composition:
By Origin
Residential - Daily household wastes
Commercial - Waste from stores,
business, offices and
institutions.
Other Municipal - Municipal street
sweepings, tree
trimmings, and catch
basin residue
By Composition
Paper
Glass
Metals
Plastics
Rubber and Leather
Textiles
Wood
Food Wastes (garbage)
Yard Waste
Miscellaneous (dirt, fines)
A-5
-------
These classification systems vary. For example, the "origin" classification
may also include bulky household waste such as appliances and furniture, or
dewatered sewage sludge. The "composition" classification may group glass,
ceramics, and similar inorganics together or may further subdivide the metals.
In either case, there can be problems in comparing data from different sources
because of these variations. As mentioned previously, the weight of these
wastes has historically been given with the moisture content included, and to
avoid incompatibilities with other published data that convention will also
be followed in this section.
Quantities
The earliest studies of the quantity of municipal solid wastes tended to
use the "origin" classification system because it was simple to implement and
provided useful broad planning data. In Table A-2, Combustion Engineering
(Ref. A-7) used data developed by the Refuse Removal Journal to arrive at a
MSW generation rate of 1.68 kg/capita/d (3.7 Ib/capita/d). The National Solid
Waste Survey (Ref. A-8) analyzed data from 6,259 communities in developing a
figure of 2.41 kg/capita/d (5.32 Ib/capita/d). Rolf Eliassen (Ref. A-4)
referred to both of these studies and derived a generation rate of 3.18 kg/
capita/d (7.0 Ib/capita/d). International Research and Technology (Ref. A-3),
and Niessen and Chansky (Ref. A-9) also used the National Solid Wastes Survey
(Ref. A-7) data in preparing their estimate of dry combustible solid waste,
and Roberts et al (Ref. A-10) relied on Niessen and Chansky's work. Thus
there are a number of studies using data from one or two early sources, and,
while they were the best data a-vailable at the time, they were known to have
shortcomings (Ref. A-4, A-7, A-10). Much of the quantity data was estimated
rather than measured, and the definitions of waste types were not uniform
across the country.
Consequently, some revisions were in order. EPA's Frank A. Smith, in an
analysis of a large number of national municipal waste estimates, has
described (Ref. A-l) some of the factors that led to a change from the early
1967 figures used by the Office of Solid Waste Management Programs (OSWMP) of
2.41 kg/capita/day and 172 x 106 Mg/year (5.32 Ib/capita/day and 190 million
TPY) to the newer 1971 figures of 1.50 kg/capita/day and 113.4 x 106 Mg/year
TABLE A-2. MUNICIPAL WASTE GENERATION RATES
kg/capita/day Ib/capita/day
Combustion Engineering
National Solid
Rolf Eliassen
Smith
Wastes Survey
(Ref.
(Ref.
(Ref.
(Ref.
A-7)
A-8)
A-4)
A-l)
1
2
3
1
.68
.41
.18
.50
3.
5.
7.
3.
7
32
0
31
A-6
-------
(3.31 Ib/capita/day and 125 million TPY). He made three major points. First,
the 1968 survey results that provided the basis for the old figures were never
fully analyzed. Second, the figure of 2.41 kg/capita/day includes reported
demolition, construction, and industrial wastes, in addition to residential,
commercial/institutional, and street wastes. The figure for the latter three
items from the 1968 data was about 1.92 kg/capita/day (4.24 Ib/capita/day) .
Third, he located two analyses suggesting that the 1968 survey returns tended
to overestimate the quantities of collected waste, although the extent of the
overestimation is uncertain. One analysis of the results made a comparison
between estimated and measured data, indicating that the measured data was
consistently lower (Ref. A-ll). The second analysis was made by Darnay and
Franklin (Ref. A-12), who compared reported wastes collected to consumption
data. They found that the quantities of various products consumed were not
sufficient to account for the high volume of reported waste collected, but
fit reasonably well if a lower total urban wastes figure was assumed.
The estimates of 1.50 kg/capita/day and 113.4 x 106 Mg/year (3.31 lb/capita/
day and 125 million TPY) are for the year 1971. U.S. EPA's "Third Report to
Congress: Resource Recovery and Waste Reduction" (Ref. A-13) updated these
numbers to 1973, using the same definitions and similar methods of calcula-
tion in order to make the two estimates directly comparable. Total post-
consumer municipal waste increased to 122.5 x 106 Mg (135 million tons), an
8% growth, and per capita generation increased to 1.60 kg/d (3.52 Ib/day), a
6.3% growth. However, there is a warning that these growth rates should not
be used for making forecasts because 1971 was not a very strong year for many
products, whereas 1973 was generally a boom year by comparison.
It is Parsons' conclusion that further refinement of national statistics
of waste quantities will not significantly assist in planning for the disposal
of the wastes or their conversion to energy. There is a large variation in
waste quantities and compositions from city to city, requiring a local
analysis of the potential for energy recovery before any final facility design
can be accomplished.
Composition
Per capita figures are useful for quickly estimating total quantities of
waste generated, but compositional data are much more useful for determining
how best to dispose of or utilize the waste. The results of a number of
waste composition studies are shown in Table A-3. By far the most quoted of
these is "The Nature of Refuse" by W.R. Niessen and S.H. Chansky (Ref. A-9) .
This project collected and summarized the results of 23 sets of apparently
independent sample collection data, and then analyzed them in detail to
develop an estimated annual average national refuse composition, shown in
Table A-3, Column 3. This estimate excluded yard waste because of seasonal
and geographical variation. The Reference A-9 report was updated by Niessen
and A.F. Alsobrook in 1972 (Ref. A-14) . Smith (Ref. A-l) took an analytical,
as opposed to empirical, approach in conducting the materials flow study that
resulted in the 1.50 kg/c/d (3.31 pounds per capita per day) figure mentioned
earlier. The National Center for Resource Recovery (Ref. A-ll) has developed
a municipal waste composition based on a material flow analysis, as has also
A-7
-------
TABLE A-3. COMPOSITION OF MUNICIPAL WASTE*
Percent by Weight
Col limns
Data Source
Reference (A- )
Year
Basis***
Loca 1 i ty
•*" Kinds of Materials
00 Fal'cr
Glass
Me t a 1 s
( Ferrous )
(Aluminum)
(Other \on-Ferrous)
Plastics
Rubber, Leather
Texti les
Wood
Food Waste (Garbage)
Yard Waste
Mi seel 1 aneous
1254
APWA KAISF.R MFSSFN i, MFSSFN f, MFSSFN 5
CIIANSKY CI1ANSKY ALSOBORRK
15 16 9** 9 14
1939 196" 196S 1968 1970
Collection Collection Collection Adjusted to Adjusted to
Data Data Data Material Material
Flows F'lows
New York Long
City Island
21 .9 42.6 50." 55. 1 57 4
5.5 9.6 9.7 8.1 9.0
6.8 S . 5 10.0 8.1 8.4
1 1.6 1.4 1.1 1.1
}:•,.: 1.9 1.4 1.2
) 5.1 2.6 1.9 2.2
2.6 5.2 2.9 2.4 3.1
7.0 10.9 19.1 19.5 20.0
17.6 NA 20.7 15.9
43.0 - 1.7 1.7 5.4
100.0 100.1 11)0.0 100.0 100.0
6 1
NC.RR SMITH
11 1
1971 1971
Materials Materials
Flows P'low
55.2 31.
9 1 9 .
8.6 9 .
7.7 8.
0.35 0.
0 . 39 0 .
1.9 5 .
1.4 2.
2.5 1 .
5 .
) 17.
42.9 19.
) 1.
99. i) 100 ,
s
3
7
5
5
6
5
4
b
4
7
6
3
4
.0
8
SM1T
1
1971
11
Adjusted to
Coll
Data
57 .
10.
10.
5 .
->
1 .
3 .
14.
14.
1.
100.
ection
S
0
1
8
7
6
7
T
6
5
.0
9
FPA
I SMI'
15
1973
,10
Material
F 1 ows
32.
9.
9 .
8.
0.
0.
5 .
2 _
1.
3 .
16.
18.
1 .
100.
8
9
o
2
7
5
7
6
4
6
6
5
4
0
10
1:1' A
(SMITH)
15
1973
Adjusted to
Collection
Data
39.6
10.0
9.5
4.1
2 . 7
1 .6
3.6
13.3
11.1
1.5
100.0
11
NYC
17
1975
Collection
Data
New
City
52.
8.
7
3.
0.
0.
0.
18.
4.
3 .
100
York
5
1
5
2
6
•J
8
1
.9
.6
.0
*** Co 1 Limns g I VCMI on a "ma tor i a 1 flows" o r "adjusted to mate ri a 1 flows" has is i nc 1 ude mo i sture i mmcd i atcly prior to mater! a 1 d i scard ; those gi vcn
on a "collection data" or "adjusted to collection data" basJs assume moisture transfer among iruterhils in collection and storage, but do not
change of moisture for the total .
-------
U.S. EPA (Ref. A-13). These five reports contain the best information
available on national average waste composition and characteristics.
Of the eleven compositions shown in Table A-3, the first five and the
last one were determined from sample collection data, and six through ten
by calculated material flows analysis. Preparing collection data involves
handsorting samples of municipal waste and weighing each component. This
method gives an accurate picture of the waste composition as it is'actually
received, which is important when planning waste disposal or resource recov-
ery systems, but the samples must be carefully chosen to fairly represent the
seasons of the year and the localities from which the collections are made.
Material flows analysis involves calculating from industrial production data
the amount of each material that enters the waste stream, less the amount
recycled, to arrive at the net waste for disposal. This approach is much
more useful for determining national averages than is the method using physi-
cal sampling, which essentially yields a series of spot measurements. However,
it requires estimates of the time interval between production and discarding
for each material, as well as the amount recycled. In addition, there is no
way to estimate food waste, yard waste, or miscellaneous waste by material
flows analysis; these must be determined from sample collection data and
added in.
Moisture Content
There is one other important difference between compositions of waste
calculated on material flows and on a collection data basis, and that is
moisture content. Flows analysis provides an estimate of waste on an "as-
generated" basis, so that the percentage moisture it contains is the percent
that was there at the time immediately prior to disposal. As it is discarded
and mixed with other wastes, a material may gain or lose moisture. Collec-
tion data provide a measurement of waste on an "as-disposed" basis, after
this moisture transfer has taken place. The "as-disposed" figures are to be
used for designing a resource recovery facility, where it is important to
know the characteristics of the materials to be handled. They are not useful
for estimating future quantities of waste, since these projections are based
on paper produced that has a different moisture content. Therefore, the
waste compositions determined from flows analysis or collection data may have
to be adjusted for moisture content, depending on the final use of the data.
Niessen and Chansky (Ref. A-9) give the following information on moisture
content:
Weight Percent Moisture
Materials Flows Basis Collection Data Basis
Kinds of Material "As Generated" "As Disposed"
Paper 8.0 24.3
Glass 2.0 3.0
Metal 2.0 6.6
A-9
-------
Weight Percent Moisture
Materials Flows Basis Collection Data Basis
Kinds of Material "As Generated" "As Disposed"
Plastics 2.0 13.8
Rubber, leather 2.0 13.8
Textiles 10.0 23.8
Wood 15.0 15.4
Food Waste 70.0 63.6
Yard Waste 50-0 37.9
Miscellaneous 2.0 3.0
These factors have been widely used to adjust between flows and collection
data analyses; all of the "adjusted" data in Table A-3 have been developed
on this basis.
The moisture content of mixed municipal solid waste can vary greatly.
In five samplings of refuse collected from Long Island households during a
period when no rain fell, Kaiser (Ref. A-16) found the moisture content to
range between 19% and 42%. The high moisture content refuse was collected
on a Monday morning in June, following a cool, humid period, and contained
grass and leaves from lawn care. A moisture content of 26% was found in
refuse collected on a Friday in February. The residential sources were in
the heating season, and Kaiser comments that this would cause a low moisture
content in the waste paper (Ref. A-16).
The annual average moisture content of refuse was reported by Kaiser
(Ref. A-16) to be 28.3%; by Niessen and Chansky (Ref. A-9) to be 28.3%; and
by Snyder (Ref. A-18) to be 25.1%.
Factors Influencing the Characteristics of Municipal Waste
A review of Table A-3 highlights some interesting statistics. The first
and last columns show a long term trend for New York City, indicating the
decline of ash and the rise of paper as major components of municipal refuse.
New York City is a densely populated area, however, having a higher percentage
of paper waste and a lower percentage of yard waste than the national average.
With the exception of Kaiser's Long Island study, the remaining eight composi-
tional estimates are national averages, arranged in chronological order from
left to right and covering a period from 1968 through 1973. Given the differ-
ent sources and methods of calculation for these estimates, they are surpris-
ingly uniform, with no strongly apparent trends. In fact, removing the ash
content from the 1939 New York City estimate shown in Column 1 of Table A-3
brings the paper content up to 38.4%, the glass up to 9.6%, and the metals up
A-10
-------
to 11.9%, quite in line with current estimates. These figures are percentages,
and give absolutely no indication of the total quantity of waste. Data devel-'
oped by the Public Health Service (reported in Ref. A-15) indicates that total
refuse production in the United States was estimated to be 70 million tons
per year in 1940 and 190 million tons per year in 1970, agreeing closely with
the figures from the National Solid Waste Survey (Ref. A-8), which were devel-
oped under similar assumptions. Therefore, even though the adjusted percent-
age composition has remained relatively stable, the quantity has increased
by more than 170%, or 3.4% per year. Approximately half of this increase can
be attributed to the total population increase and half to the per capita
increase in the amount of refuse generated.
The original sources of the materials found in municipal waste have been
investigated by several researchers (Ref. A-l, A-ll and A-13). Table A-4,
from U.S. EPA's "Third Report to Congress" (Ref. A-13) shows such a material
flow analysis. From this table it can be seen that the paper for containers
and packaging is the largest single contributor to total product waste, with
food waste (garbage) second, and glass containers third. Yard waste, the
largest single contributor to the total municipal waste stream, is outside
the product waste flow.
A number of cyclical variations in waste generation can be identified,
including weekly, yearly, and seasonal. Collections following a weekend may
have above average quantities of yard waste and collections following
Christmas, unusual quantities of packaging materials. Any program for col-
lecting data on waste quantities must take these cyclical patterns into
consideration. This is especially important when discussing yard waste,
which is a large and highly variable component of municipal solid waste.
Niessen and Chansky (Ref. A-9) found a significant annual variation in yard
waste and attributed it to seasonal climatic changes. In regions with little
seasonal change, yard waste is generated all year, and is a larger contributor
to municipal waste than in regions of greater seasonal change where the grow-
ing season is less. Based on limited sample collection data, the seasonal
variations in percent yard waste they found are:
Summer Fall Winter Spring
Northern States 22.9% 6.4% 0.3% 11.1%
Southern States 22.5% 5.2% 6.8% 9.2%
Naturally, in densely populated urban areas or in areas with unusual climato-
logical conditions these averages may not be at all representative; New York
City (Col. II of Table A-4) for example, shows only 4.9% yard waste.
If these figures were stated on a materials flow basis, they would be
30% to 45% higher, because yard waste loses a significant amount of moisture
between the time it is generated and the time it is disposed of. Note that
these figures have been developed from limited data, and that regional and
other differences may also have a large, undetermined effect.
A-ll
-------
fO
TABLE A-4. MATERIAL FLOW ESTIMATES OF RESIDENTIAL AND COMMERCIAL POST-CONSUMER NET SOLID
WASTE DISPOSED OF, BY MATERIAL AND PRODUCT CATEGORIES, 1973*t
Material
Paper
t!l ass
Metals
Fer7-ous
Aluminum
Other nonfcrrous
Plast ics
Rubber and leather
Text iIcs
Wood
Total nonfood
product waste
i;ood waste
Total product
waste
Yard waste
Misc. inorganics
Total
Totals
46.9
As-d i sposed
we i uht §
Mill ion Mi 11 ion
tons Percent tons Percent
52.8
9.9
9. 3
2. 7
1.4
3.6
5.6
3.7
2. 1
4.9
20.5
85.4 63.F
22.4 16.6
107.8
25.0
1.9
80. 1
18.5
1 .4
114.0
19.0
2.0
134.8
"Smith, !•'..A., and I-'.L. Smith, Office of Solid Waste Management I'roy rams , Resource Recovery Division. Hata revised Dec. 1974.
TMct solid waste disposal defined as net residual material after accounting for recycled materials diverted from waste stream.
39.6
10. 3
9.9
4. 1
2.1
1.6
3.6
96.0 71.1
18.0 13.3
84.4
14. 1
1.5
-------
The influence of demographic factors on the generation of residential
solid waste has been investigated by the Los Angeles Bureau of Sanitation
(Ref. A-19), R. G. Davidson (Ref. A-20), and C. R. Rhyner (Ref. A-21). The
Los Angeles study found that the type of dwelling was the most important
factor. The average single family house generates 2.5 to 3 times as much
refuse as the average apartment in a multiple dwelling complex, including
yard waste. Rhyner's study, which covered six southern states plus a portion
of Wisconsin, was unable to form any conclusion on the difference in genera-
tion rates between single family dwellings and apartments because of the
influence of other factors. Rhyner's study excluded yard waste, which was
called the major contributor to seasonal and geographic variations in the
amounts of domestic solid waste, because of its variability.
Davidson and Rhyner both found that the total amount of domestic solid
waste generated per family could be expressed as an amount per family plus
an amount per family member. Paper, organic garbage, and metal increase the
most with increasing household size, while newspapers, plastics, and colored
glass increase the least, according to Rhyner.
Solid waste generation also varies with affluence, as measured by the
size of the dwelling and ownership of air conditioning. Rhyner found that
the more affluent households tended to generate more paper and more total
waste. The Los Angeles study, however, found variations in waste generation
due to economic level, but the impact was small and the pattern insufficiently
clear to permit generalizations.
The amount of solid waste generated by farm households is significantly
less than that generated by urban households, according to Rhyner. In both
the southern states and in Wisconsin he found that farm households generated
significantly less paper and glass than urban households. In Wisconsin, farm
households produced 20% less total waste, and in the South, about 12% less.
In both surveys the number of people in the household account for about a 6%
difference (Ref. A-22).
Ultimate Analysis
An ultimate analysis for typical mixed municipal solid waste given by
Snyder (Re'f. A-18) is shown in Table A-5; Kaiser (Ref. A-16) has made a simi-
lar analysis. Two appealing characteristics encouraging the use of refuse
as a fuel can be seen: the low sulfur and the relatively low chlorine per-
centages. Roberts (Ref. A-10) has pointed out that U.S. municipal solid
waste has a consistent average sulfur content of 0.1% to 0.2%, in contrast
to the 2.5% to 3.5% for typical power plant coals. The chlorine content of
refuse is in the high end of the range for coals now burned (Ref. A-10), but
tests at St. Louis (Ref. A-23) with air classified shredded combustibles
showed that two-thirds of the chlorine is in the form of inorganic chlorides
that normally would not react in a furnace to form HC1. Further experience
is required to establish the quantity of chlorides that will be present in
stack gases. Analysis of refuse in Chicago showed that more chlorine origi-
nates in the rubber (chlorinated elastomers) than in the plastics fraction
(polyvinyl chloride) (Ref. A-18).
A-13
-------
TABLE A-5. ULTIMATE ANALYSIS AND HEATING VALUE FOR
TYPICAL MIXED MUNICIPAL SOLID WASTE
Component
Analysis
(as received)
% by weight
Analysis
(Dry Bases)
% by weight
Moisture
Carbon
Hydrogen
Oxygen
Nitrogen
Chlorine (organic 0.16),
25.1
25.2
3.2
18.8
0.4
0.0
33.5
4.3
25.2
0.5
(inorganic 0. 14)
Sulfur
Metal
Glass, ceramics
Ash
Total
Higher heating value, HHV
Source :
0.3
0.1
8.7
12.2
6.0
100.0
4,400 Btu/lb
Ref. A-18
0.4
0.1
11.6
16.3
8.1
100.0
5,600 Btu/lb
Snyder (Ref. A-18) comments that the carbon content of the combustibles
is low with respect to usual commercial fuels because of the cellulosic
character of the combustibles, where half the weight is in oxygen. Such
items as rubber and plastics contain little or no oxygen and are high in
carbon and hydrogen. Dry paper has an HHV of approximately 18.61 to
20.24 MJ/kg (8,000 to 8,700 Btu/lb), rubber and leather have values in the
17.44 to 29.77 MJ/kg (7,500 to 12,800 Btu/lb) range, and plastics are in
the 37.22 to 41.87 MJ/kg (16,000 to 18,000 Btu/lb) range (Ref. A-18).
The energy content per unit weight depends on the quantity and chemical
characteristics of the individual components, and the amount of moisture
present. Snyder (Ref. A-18) gives a value of 19.31 MJ/kg (8300 Btu/lb) for
the dry, organic portion of municipal solid waste. The addition of inor-
ganics, which have very low heating values, and water, which has no heating
value, reduces the heating value down to the range of 10.12 to 10.70 MJ/kg
(4350 to 4600 Btu/lb) (Refs. A-9, A-18, A-2, A-24).
A-14
-------
Forecasts of Municipal Waste
Four independent forecasts of municipal waste have been identified:
Niessen and Chansky (Ref. A-9); Roberts et al (Ref. A-10); International
Research and Technology (Ref. A-3); and Midwest Research Institute (MRI)
(Ref. A-2). Table A-6 summarizes the results of these projections. The
Niessen and Chansky, the Roberts and the IR§T estimates are now outdated
by the MRI projection, but interesting comparisons can be made.
The Niessen-Chansky projections were made by developing national growth
rate indicators for each of the 10 major categories comprising municipal
refuse, and applying these growth rates to a base pound per capita per day
waste generation figure determined by material analysis. Data were extremely
limited; the base generation figure came from an analysis of one state. In
addition, the optimistic growth projections of the late 1960s gave an average
growth rate, 1970 - 1990, of 1.8% per year. The resulting waste generation
figure for 1990 of 366 x 106 Mg (403 million tons) is more than twice the
current estimate of 182 x io6 Mg (200.8 million tons).
One important factor that Niessen and Chansky tried to take into account
is the difference between generated waste and collected waste. They developed
a 1968 national average figure of 69% collected, and allowed it to increase
in a linear fashion to 95% in the year 2000 (Ref. A-9). The reasonableness
of this assumption is uncertain; it is unknown that a significant percent of
municipal waste disappears as litter or is indiscriminately dumped, but the
effect of this and how it may change in the future is unknown.
The Roberts estimate (Ref. A-10) assumed a growth rate of 1.5% per year
based on the increase in per capita consumption of non-durable goods for the
period 1950 to 1967. Since this period included cycles of both economic
recession and growth, the authors felt it to be reasonably representative.
The base per capita collected rate was determined from the 1968 National Solid
Wastes Survey (Ref. A-8), confirmedj interestingly enough, by Niessen's work
at A.D. Little on which the Niessen and Chansky estimates (Ref. A-9) are
based. The Roberts et al estimate used only the household and commercial
figures reported in the National Solid Wastes Survey, but as has been dis-
cussed earlier, these figures are the result of over-estimates and have been
reduced significantly. Based on the information available at the time,
Roberts projected the 1990 quantity of refuse collected to be 282 x io6 Mg
(311 million tons), 54% more than the current estimate of 132 x io6 Mg
(200.8 million tons) generated.
The IR§T (Ref. A-3) forecasts assumed base levels from the National
Solid Wastes Survey (Ref. A-8) and other sources. Analyses were than made
to determine the percent of total waste that were dry combustibles in 1970,
and how that percentage could be expected to change in 1980 and 1990. The
next task was to determine the future per capita generation rate for each
category of wastes. In the case of household and municipal wastes this was
done in an ad hoc manner since there existed no time series data from which
to construct a trend (Ref. A-3). For commercial/institutional waste cate-
gories, the generation rate was related to factors such as the number of
A-15
-------
Table A-6A. FORECASTS OF MUNICIPAL SOLID WASTES
TOTAL WASTES INCLUDING MOISTURE UNLESS OTHERWISE STATED
(SI Units)
1970 1971 1975 1980 1985 1990
Average Growth Rate
Kg Per Capita Per Day Per Year 1970 - 1990
Niessen § Chansky* (Ref. A-9)
Generated 2.67 3.00 3.35 3.84 1.8%
Collected 1.88 2.24 2.64 3.33
Roberts, et al* (Ref. A-10)
Collected 2.20 2.37 2.55 2.75 2.96 1.5%
IR£T Generated Dry Combustible 1.38 1.64 1.95
MRI - Generated 1.50 1.58 1.71 1.80 1.91 1.3%
^ Population
(Millions) 204.8 207.0 216.5 230.8 246.2 260.7
Tg Per Year
Niessen £ Chansky
Generated 200 237 282 366
Collected 141 177 222 317
Roberts, et al*
Collected 164 187 215 247 282
IR$T* (Ref. A-3) Generated Dry
Combustible 103 138 186
MRI* - (Ref. A-2) Generated 113 125 144 162 182
*Reported Figures. Other values calculated using population figures shown in MRI report.
-------
Table A-GB. FORECASTS OF MUNICIPAL SOLID WASTES
TOTAL WASTES INCLUDING MOISTURE UNLESS OTHERWISE STATED
(English Units)
1970 1971 1975 1980 1985 1990
Average Growth Rate
Pounds Per Capita Per Day Per Year 1970 - 1990
Niessen £ Chansky* (Ref. A-9)
Generated 5.88 6.61 7.39 8.46 1.8%
Collected 4.15 4.94 5.82 7.35
Roberts, et al* (Ref. A-10)
Collected 4.85 5.22 5.63 6.06 6.53 1.5%
Generated Dry Combustible 3.05 3.61 4.30
MRI - Generated 3.31 3.49 3.78 3.98 4.22 1.3%
Population
(Millions) 204.8 207.0 216.5 230.8 246.2 260.7
Millions of Tons Per Year
Niessen § Chansky
Generated 220 261 311 403
Collected 155 ---- 195 245 350
Roberts, et al*
Collected 181 206 237 272 311
IR$T* (Ref. A-3) Generated Dry
Combustible 114 152 205
MRI* - (Ref. A-2) Generated 125 138 159 179 201
*Reported Figures. Other values calculated using population figures shown in MRI report.
-------
white collar workers, restaurant and retail sales, and the number of school
students and hospital patients. Forecasts were made for each factor from
data in the Survey of Current Business and the Statistical Abstract of the
United States. IR£T included household, municipal, sewage, commercial/
institutional, manufacturing plant trash and demolition wastes in their
estimate of MSW. For this report, only the household, municipal and
commercial/institutional wastes are included in MSW, and IR^T's figures
for these wastes are shown in Table A-6. The other wastes are discussed
in other sections of this report.
The IR£T data are on a dry, combustible basis and are therefore not
compatible with the other forecasts. The household wastes component was
assumed to grow at a rate of 2% per year, a very high value, and the com-
bustible portion to increase from 78% in 1970 to 90% in 1990, principally due
to an increase in the use of plastics. The increasingly apparent natural gas
shortage and escalation in petroleum prices since this forecast was made
render these assumptions questionable.
The MRI forecasts (Ref. A-2) have been developed over the past several
years for the Office of Solid Waste Management Programs, EPA. They have
been quoted in the EPA Reports to Congress (Refs. A-25, A-6, A-13) and
elsewhere before being published as Baseline Forecasts of Resource Recovery,
1972 to 1990. The summary table of municipal solid waste generation from
that report is shown here as Table A-7. The basis of these solid waste
generation values is a calculated forecast tonnage based on EPA's estimate of
mixed municipal waste generation and composition for the year 1971, and
updating of this data by independent material-by-material forecasts of waste
generation made by MRI. EPA's municipal waste generation figure for 1971 was
113.4 x 106 Mg (125 million tons) (Ref. A-l); MRI's forecasts produced an
average long term growth rate of 1.3% per year for the per capita generation
figure. High, medium, and low forecasts were made:
Low Estimate Medium Estimate High Estimate
Tg 106 Tons Tg 106 Tons Tg 106 Tons
1975 122 135 127 140 130 143
1980 136 150 145 160 154 170
1985 150 165 163 180 181 200
1990 163 180 181 200 218 240
Adjusted for population growth, the high estimate has a long term per
capita waste generation growth rate of 2.24% per year, the highest of any of
the estimates considered. The low estimate has a long term per capita waste
generation growth of 0.71% per year.
An interesting final point can be observed. Some cities are reporting
a leveling off in per capita waste generation and if this were to achieve a
zero growth rate at, say, the estimated 1975 level of 1.59 kg (3.40 Ibs) per
A-18
-------
TABLE A-7A. MUNICIPAL SOLID WASTE GENERATION BY MATERIAL CATEGORY, 1971 to 1990
(In Megagrams and Percent)
1971
Waste Component
Paper
Glass - containers
other
Total Glass
Ferrous - cans and small items
bulky appliances
other
Total Ferrous
Nonferrous - packaging Al
other Al
Nonferrous - other
V— Total Nonferrous
<£>
Plastics
Rubber/Leather
Textiles
Wood
Sub Total - Manufactured
Products
Food Wastes
Yard Wastes
Misc. Inorganics
Total Solid Waste
Total Organics - as generated
Total Inorganics
Mg
35.
10.
0.
11.
8.
1.
0,
9.
,0.
0.
iL
i.
3 .
5.
1
4
69
19
21
1
113
89
23
.5
1
.9
0
.1
,5
,1
.7
.5
,2
,4
.1
.8
.0
.6
.2
.9
.9
.9
.7
.4
.9
.5
Percent
31,
8.
0,
9.
7.
1.
0
8,
0.
0
1L
1.
3
2,
1
3
61
17
19
1
100
79
20
.3
9
,8
.7
,1
. 3
.1
.5
,5
.2
•J.
.0
.4
.6
.4
.7
.6
.6
. 5
.5
.0
.3
•"
1972
Mg
37
11.
0.
12
8
1
0
10
0
0
0_
1
4
3
1
4
74
20
T 1
1
118
93
25
.6
,1
.9
.0
.4
.5
.1
.0
.6
. 2
A_
-,
. 1
. 1
.7
. 5
.0
.2
. 3
.8
.3
. 3
.0
Percent
31
9
0.
10,
7
1.
0
8.
0.
0,
£.
I ,
5 .
2.
1.
5 .
62.
17.
18.
1
100.
78.
21 .
.8
, 3
.8
,1
, 1
, 3
.1
.5
,5
,2
J5
.0
.5
.6
.5
.6
.6
0
.9
.5
.0
.9
. 1
1975
Mg
38.
12.
1.
13.
9.
1.
0.
10.
0.
0.
(K
1.
5.
3 .
1.
4.
78.
20.
23.
1 .
125.
98.
-
2
i
0
2
0
5
1
6
9
2
j
5
2
3
9
6
5
9
8
9
1
0
1
Percent
30.
9.
0.
10.
7.
1 .
0.
8.
0.
0.
JL
1.
4.
2.
1 .
5 .
62.
16.
19.
1 .
100.
78.
21 .
.6
7
8
.5
•,
2
.1
5
7
1
3
1
\
7
5
7
7
7
1
5
0
3
7
1980
Mg
45
13.
1
14,
10.
1
0
11.
1.
0.
CK
1.
7.
3.
2.
5.
93.
22.
26.
2.
144.
113.
30.
.8
,7
.2
.9
.1
.7
.1
,9
.2
.2
_4_
,8
,6
9
2
2
5
2
6
3
4
6
8
Percent
31
9
0
10
7
1
0
8
0
0
2
1
5
2
i
3
64
15
IS
1
100
78
21
.8
.5
.8
.3
.0
.2
.1
.3
.8
.1
.il
-,
. 3
i
.5
.6
>7
.4
.4
.6
. 1
.6
.5
1985
Mg
52.0
13.7
1.4
15.1
11.3
2.3
0.2
13.8
1.5
0.3
0^5
2.3
10.0
4.5
2.6
5.9
106.2
23.8
29.6
2.6
162.2
128.5
.33.7
Percent
32.0
8.5
0.8
9.3
7.0
1.4
0.1
8.5
0.9
0.2
0,3
1.4
6.2
2.8
1.6
3.6
65.4
14.7
18.3
1.6
100.0
79.2
20.8
1990
Mg
61.0
13.8
1.5
15.3
12.2
2.4
0.2
14.8
1.8
0.3
0^6
2.7
12.0
5.3
3.2
6.7
121.0
25.1
33.0
3.0
182.1
146.2
35.9
Percent
33.5
7.6
0.8
8.4
6.7
1.4
0.1
8.2
1.0
0.2
0.3
1.5
6.6
2.9
1.7
3.7
66.5
13.8
18.1
1.6
100.0
80.3
19.7
Sources: Refs. A-2, A-25
-------
TABLE A-7B. MUNICIPAL SOLID WASTE GENERATION BY MATERIAL CATEGORY, 1971 TO 1990
(In Million Tons and Percent)
I
KJ
1971
Waste Component
Paper
Glass - containers
other
Total Glass
Ferrous - cans and small items
bulky applicances
other
Total Ferrous
Nonferrous - packaging Al
other Al
Nonferrous - other
Total Nonferrous
Plastics
Rubber/ Leather
Textiles
Wood
Sub Total - Manufactured
Products
Food Wastes
Ya I'd Wastes
Misc. Inorganics
Total Sol id Waste
Total Organ ics - as generated
Total Inorganics
Tons
39.
11.
1.
12.
8.
1.
0
10
0.
0
0.
1.
4.
3 .
1 .
4.
77.
22.
24.
1,
125.
99.
25.
1
1
.0
.1
9
.7
.1
.7
.6
. 2
.4
. 2
.2
, 3
.8
.6
.0
0
. 1
9
0
1
!1
Percent
31.
8,
0,
9
7.
1,
0
8
0.
0
0.
1
5
->
1 .
5 .
61,
17.
19,
1
100
79.
20.
. 3
.9
,8
.7
.1
. 3
.1
.5
.5
.2
. 3
.0
.4
.6
.4
.7
.6
.6
. 3
.5
.0
. 3
~
1972
Tons
41.
12.
1.
13,
9.
1.
0.
11.
0.
0.
0.
1.
4.
5 .
I .
4.
81 .
22.
24.
2.
1 30 .
102.
27.
5
.?.
0
2
3
7
,1
.1
7
, 2
.4
, 5
r
4
9
.7
6
2
.(,
.0
.4
.8
6
Percent
31.
9.
0.
10.
7.
1.
0.
8,
0.
0.
0.
1.
3 .
2.
1 .
3.
62.
17.
18.
1 ,
100.
~S
21 .
8
3
8
1.
1
3
,1
.5
5
2
3
0
5
6
3
6
,6
0
.'-'
,5
.0
•»
1
1975
Tons
42.
13.
1.
14.
9.
1.
0.
11
1.
0.
0,
1.
5 .
5 .
2.
5
8(1.
23,
2h
2
157
108
29
!
.4
.1
.5
.9
. 7
.1
.7
.0
2
4
.6
,7
.7
•1
. 1
. 5
.0
. 5
. 1
.9
.0
.9
Percent
30.
9.
0.
10.
7 _
1.
0.
8.
0.
0.
0.
1 .
4.
2 .
1.
3 .
62.
16
19
1
100
78
21
6
•7
8
,5
9
2
.1
5
7
1
3
.1
1
7
,5
, 7
.7
,7
. 1
.5
.0
.5
7
1980
Tons
50.
15.
1.
16.
11.
1.
0.
13.
1.
0.
0.
2.
8.
4.
2.
5.
102.
24
29
2
1 59
125
34
6
1
3
4
1
9
1
1
3
2
5
0
4
3
4
7
9
r
. 3
.5
.2
.2
.0
Percent
31.
9.
0.
10.
7.
1.
0.
8.
0.
0.
0.
1.
5.
2
I
3
64
15
18
1
100
78
21
8
5
8
3
0
,2
1
.3
8
I
3
.2
.3
.7
.5
.6
.7
.4
.4
.6
. 1
.6
. 5
1985
Tons
57.3
15.1
1.5
16.6
12.5
2.5
0.2
15.2
1.6
0.3
0.6
2.5
11.0
5.0
2.9
6.5
117.0
26.2
32.7
2.9
178.8
141.6
37.2
Percent
32.0
8.5
0.8
9.3
7.0
1.4
0.1
8.5
0.9
0.2
0.3
1.4
6.2
2.8
1.6
3.6
65.4
14.7
18.3
1 .6
100.0
79.2
20.8
1990
Tons
67.2
15.2
1.7
16.9
13.5
2.7
0.2
16.4
2.0
0.3
0.7
3.0
13.2
5.8
3.5
7.4
133.4
27.7
36.4
3.3
200.8
161.2
39.6
Percent
33.5
7.6
0.8
8.4
6.7
1.4
0.1
8.2
1.0
0.2
0.3
1.5
6.6
2.9
1.7
3.7
66.5
13.8
18.1
1.6
100.0
80.3
19.7
Sources: Refs. A-2, A-25
-------
person per day, the total quantity of municipal solid waste generated would be
influenced by population alone. In that case, the total generated in 1990
would be 151 x 106 Mg (166 million tons), only 83% of the current estimate.
It is not possible at the present time to say whether there is such a trend
independent of economic patterns, but it is interesting to note that, in
chronological order, each of the forecasts examined here used a lower base
and a lower long term growth rate.
Just as the composition of municipal waste is expected to change over
time, the characteristics, specifically the heating value, will also change.
The three groups who made the forecasts discussed in the previous section also
made the following projections of the heating value in MJ/kg (Btu/lb):
1970
1972
1975
1980
1985
1990
Change,
1970-1990
Niessen §
Chansky
(Ref. A-9)
10.58 (4550)
10.79 (4640)
11.00 (4730)
11.53 (4956)
0.94 ( 406)
Roberts §
Wilson
(Ref. A-24)
9.42 (4050)
9.65 (4150)
10.00 (4300)
10.82 (4650)
12.04 (5175)
2.62 (1125)
MR I
(Ref. A-2)
10.47 (4500)
10.82 (4650)
11.16 (4800)
11.63 (5000)
1.16 ( 500)
Many of the municipal refuse components have similar heating values, the
notable exception begin plastic and rubber. A large change in the plastic
fraction would result in a large change in the average heating value of muni-
cipal waste. This is the reason that the Roberts estimate shows such a large
change in heating value, but this estimate was made before the 1974 petroleum
price increases, which changed the outlook for the plastics industry.
Because the amount of heat per unit time that most energy recovery systems
can accept is fixed by their design, the projected increase in heating value
can be expected to yield a corresponding decrease in capacity of any given
system. For example, a unit designed for a peak of 1000 tons per day of
4500 Btu/lb refuse may have a usable capacity of only 900 tons per day if the
heating value of the refuse increased to 5000 Btu/lb.
Municipal sewage contains organic and inorganic materials originating as
human wastes, garbage, and industrial wastes. A common design figure for
domestic (human waste) dry sewage solids is 0.091 kg (0.20 Ib) per person per
A-21
-------
day. Extensive use of garbage grinders in a community will increase this
figure substantially (Ref. A-26). Sewage solids generation rates are highly
variable; one source reports a range of 0.062 to 0.223 kg (0.137 to 0.491 Ib)
per person per day (Ref. A-27). These figures will be further increased if
industrial sewage is handled by the municipal sewage treatment plant. Inter-
national Research and Technology (Ref. A-3) uses an average figure for total
municipal sewage solids of 0.15 kg (0.34 Ib) (dry) per person per day. These
solids are present in the sewage in dilute suspension or solutions. By means
of various concentrating and water reclamation steps, the solids concentration
can be increased to 30% and even higher (Ref. A-2). Until recently, the result-
ing sludge was frequently landfilled or ocean dumped although it also has
some value as a fertilizer (Ref. A-26, A-28). Sludge can also be burned, the
dry solids having a heating value of approximately 23.21 MJ/kg (10,000 Btu/lb).
However, the high moisture content of the raw sludge reduced this heating
value so much that supplementary firing with natural gas or oil is required.
Energy can also be reclaimed from sewage in the form of methane from
anaerobic digestion. The Los Angeles Hyperion Sewage Treatment Plant, for
example, daily produces approximately 127 426 m3 (4.5 x 106 cubic feet) of a
methane-carbon dioxide gas mix having a heating value of 23.63 MJ/m3 (600 Btu
per cubic foot). About two-thirds of this gas is used in the sewage treatment
plant to provide power; the other third is piped to a local electric genera-
ting station, where it supplies approximately 1% of the energy requirement.
International Research and Technology Corporation (Ref. A-3) projected
the generation rate of dry sewage solids to be:
1970 0.15 kg (0.34 lb)/cap/day 11.5 Tg (12.7 x 106 tons) per year
1980 0.17 kg (0.37 lb)/cap/day 14.0 Tg (15.4 x 106 tons) per year
1990 0.18 kg (0.39 Ib)/cap/day 16.5 Tg (18.2 x 106 tons) per year
Some authorities believe these quantities to be too high and a more
recent estimate by Bernard of the quantity of dry municipal sewage sludge gives
much lower annual values (Ref. A-29):
1973 4.3 Tg (4.7 x 106 tons)
1977 4.5 Tg (5.0 x 106 tons)
1985 7.3 Tg (8.0 x 106 tons)
1990 9.1 Tg (10.0 x 106 tons)
The difference between the two estimates can be explained by assuming
that Bernard used 0.091 kg (0.20 Ib) per person per day versus IR§T's 0.15 kg
(0.34 Ib). In addition, Bernard appears to have factored the sludge genera-
tions rate by the recovery rates for primary and secondary treatment, and by
the percentage of the population served by these treatment systems. According
to a 1973 report by the EPA (Ref. A-30), 77.6% of the U.S. population is served
by public sewerage. Bernard indicates that, of the population served, about
25% are served by primary sewage treatment only, and about 72% by a combina-
tion of primary and secondary treatment. Primary treatment can recover 0.054 kg
(0.12 Ib) per person per day; secondary treatment can recover most of the
remaining 0.036 kg (0.08 Ib) per person per day of the total 0.091 kg
(0.20 Ib) per person per day generated (Ref. A-31).
A-22
-------
Bernard anticipates a surge in sludge generation due to legislation
requiring more extensive treatment (Ref. A-27). Another factor in the per
capita increase in sludge generation is an increased standard of living, result-
ing in greater use of garbage grinders and biodegradable single use materials
(Ref. A-32).
INDUSTRIAL WASTES
Types
Industrial wastes can be solid, liquid, gaseous, or sludges. They can be
combustible or incombustible, production waste, or plant trash (office, cafe-
teria, and shipping room waste). However the wastes are defined, they are
usually associated with a specific industry by means of the Standard Industrial
Classification (SIC) Code (Ref. A-33), and are typically categorized in this
manner in compilations and discussions.
The composition of individual industrial process wastes is usually well-
known and homogeneous, and because of this they are frequently recycled. Saw-
mill operations generate large amounts of sawdust, which used to pose a
disposal problem. Now much of the sawdust is recycled into particle board,
paper pulp, and other products. Combustible industrial gases that used to be
vented are now burned to provide process steam. Even natural gas was once a
waste in petroleum drilling and processing. The nature of industrial wastes
is constantly changing as new uses are found for residual materials and
environmental control regulations become more stringent. This means that the
industrial waste stream may be growing smaller even though industrial produc-
tion is expanding, enormously complicating the forecasting process.
Quantities
Data on industrial waste quantities are very limited, particularly on
liquid and gaseous wastes. Only three studies were identified as having made
an independent, comprehensive assessment of industrial waste quantities. The
result of these studies are shown in Table A-8, along with two re-analyses of
data from one of them.
The Combustion Engineering Study, reported in Volume II of "Technical-
Economic Study of Solid Waste Needs and Practices" (Ref. A-7), includes only
solid wastes in 24 SIC Code industry groups, mostly manufacturing. The data
was obtained in some 320 interviews, during which an attempt was made to
identify office waste and general trash, shipping waste, process wastes, and
solid wastes collected by air and liquid cleaning devices. It is not clear,
however, whether the data is for dry solids or for solids with normal moisture.
The interviews also developed the fraction of total waste utilized in any way,
so that the waste quantities reported are those requiring ultimate disposal.
Because industrial production data is often regarded as proprietary and is
therefore unavailable, the waste quantities were related to the number of
employees.
These factors were then multiplied by statistics of industrial employment
to calculate the amount of total waste. In the tabulation shown in Table A-8,
A-23
-------
TABLE A-8. INDUSTRIAL WASTES IN MILLIONS OF TONS PER YEAR
Smith Huffman
(Ref. A-36) (Ref. A-37)
1967 1970
Flry Solid Dry Solid
Combustible Combustible
15-16-17
19
20
21
--
23
24
T r
26
27
28
29
50
51
52
o5
51
55
56
57
58
59
242x
Construction and Deiiiol i t i on
Ordnance
F'ood Products
Tobacco
Text i le Mi 1 1 Products
Appa rel
Iv'ood Products
Furniture and Fixtures
Paper
Printing and Publishing
Chemi ca 1 P roducts
Petroleum and Coal Products
Rubber and Plastic Products
Leather
Stone, Clay and Class
Primary Metal Industries
Fabricated Metal Products
Machinery, except electric
Kicetric and Flectronic Hquipment
Trans port at inn Hqu i pmcnt
Inst ruments
Mi sec 1 1 ancous
Process Wastes
Plant Trash
C IRAN I) TOTAL INDUSTRIAL IVAS'IF.S
Sawmi 1 1 s
19.05
0.56
".15
0.11
1 .08
0 . 56
58.115
1.91
5.09
".61
5. 02
0.3"
2. 16
5. Id
2. 16
1 . ~5
5.85
1.32
1 ."4
0.85
0 . 85
103.21
Incl
105.21
52.80
a - demo
on 1 y
22.151'
0.41
9.57 14.17
0.18
1.55 2.11
0.51 2.99
1~.92 1.86
3.21 1.28
".52 2.8]
10. 55 6. 15
3 .51 9 . 50
()."7 0.50
5.1,5 5.~1
5.(>0 2.88
5.58 1.10
2.22 1.19
3 .51 9.1 5
6. 52
2 . ~ 1 5.61
2.19 1.61
1.5" 0. SI
1.21 0.68
1 Od. 18 88. d2
Incl. hid.
106. IS 88.62
1 1 . 50
tion only
5.15 6. -10- 6.58 7.64- 8.02 1.30
Misc.
0.71 ill. 71 1.10- 1.15 1.21- 1.32 0.71
Misc.
1.58 \ 0.29 0.48- 0.51 0.61- 0.73 io.29
-
I 15. 81 12.76-15.96 8.10- 9.73 15.55
0.46
0.02 ill. 59 9.09- 9.80 6.08- 7.24 10.16
0.01 0.40
5.11 0.07 0.11- 0.16 0.20- 0.25 Misc.
0.01 - - - Misc.
0.09 0.09 0.12- 0.15 0.16- 0.19 0.15
0.06 0.06- 0.09 0.10- 0.16 0.06
0.01 - - - Misc.
2.59 - - - Misc.
0.10 - - - Misc.
5.57 - - - Misc.
0.55 - - - Misc.
0.92 - - - Misc.
0.05 - - - Misc.
1.21 1 .95- 2. 15 2.51-2.99 1.21
16.98 54.78 52.10-54.49 26.61-50.65 50.09
Incl. 11.80 19.85 26.40 N.A.
16.58 51.95-51.5-1 55.01-57.03
15.61
3.85
-
0.75
Misc.
j-0.30
125.65
.
1
0.45
-
Misc.
• Misc.
0.10
31.13
11.75
42.85
1 - Forecast b - construction
on 1 y
-------
minor waste streams from cotton ginning and stockyards were left to be
discussed under agricultural waste, and a major waste stream from supermarkets
was left out because it was part of the municipal waste stream.
Combustion Engineering estimated that 93.66 x 106 Mg (103.24 million tons)
of industrial solid wastes were generated in 1965. More than half that total
was generated in the demolition and wood products industries; saw mills are by
far the largest contributor to industrial solid wastes. As mentioned earlier,
new uses are turning this waste into useful products, and consequently the
amount of saw mill waste is dropping. Saw mills are industrial operations,
but their waste is more akin to agricultural wastes, and is often discussed
under that heading (See Table A-10). Alich (Ref. A-34), for example, reported
that there are 105 x 106 Mg (116 million dry tons per year) of forestry waste,
but goes on to say that two-thirds of that are mill residues. "Of the
76 million dry tons of mill residues, 50 percent is sold for various purposes,
25 percent is used as a fuel without sale, and 25 percent is unused (waste)"
(Ref. A-34). Informal discussion with wood industry officials indicate that
in ten years there will be essentially no wastage, and that the percent sold
will increase.
Niessen and Alsobrook (Ref. A-14) used an approach similar to Combustion
Engineering's, but also attempted to gather data on liquid wastes and on
sludges. The data they reported was in tons per employee per year (TEY).
These have been applied by Parsons to employment statistics developed by the
Bureau of Labor Statistics (Ref. A-35) for 1970 to determine waste quantities.
The calculated waste totals are 80.40 x 106 Mg (88.62 million tons) of solid
wastes and 15.40 x 106 Mg (16.98 million tons) of liquid wastes for 1970, but
these do not include demolition wastes. Taking that into consideration, their
estimate of industrial solid waste is very close to that developed by Combus-
tion Engineering.
The Niessen and Alsobrook TEY factors come from the very narrow base of
northern New Jersey and western New York State. Those authors caution that
the figures may not be reasonably representative of other areas; evidence of
this possibility can be seen in their relatively high figure for Code 20:
Food Products and low figure for Code 24: Wood Products.
IR&T (Ref. A-3) took a narrower view of wastes, more useful for investi-
gating energy recovery possibilities, by studying only dry, combustible solid
waste. IR£T also took a different approach than the other two studies, using
a combination of material flow and input-output analysis to determine the
ultimate disposition of all combustible material, either into products or
into wastes. The waste quantities were aggregated into industrial groups
whose members produce similar waste. Fortunately, these groups tend to be
combinations of SIC Code groups, so the IR§T data retain some comparability
with the other studies.
The IR§T study reported 42.26 x 106 Mg (46.58 million tons) of industrial
combustible solid waste in 1967, less than half that estimated by Combustion
Engineering. In both estimates, wood product waste is the largest single
contributor. Because it is combustible waste in both cases, the estimate
should be similar; there is not sufficient information to account for the
A-25
-------
discrepancy shown. Similar disagreements exist in the food processing and
textile categories, and as a result the close agreement in the paper, printing,
and publishing category is surprising.
Two important re-analyses of the IR£T data have been made. Smith (Ref. A-36)
divided the IR§T waste quantity aggregations to fit the SIC code more closely,
and Huffman (Ref. A-37) updated the quantities to 1970. As a result of Smith's
breakdown, the quantity for sawmill operations can be separated as 12.35 x
106 Mg (13.61 million tons) per year; this is still less than half of the
29.76 x 106 Mg (32.80 million tons) estimated for 1965 by Combustion Engineer-
ing, but may not be unreasonable if the resource recovery of sawdust proceeded
more rapidly than Combustion Engineering estimated in giving the 1975 forecast
at 10.43 x 106 Mg (11.50 million tons).
For liquid and sludge waste, Niessen and Alsobrook developed an estimate
of 15.40 x 106 Mg (16.98 million tons) per year. The major contributing
industries were the chemical products, machinery, and primary metals industries
(Ref. A-14, Table 8). The only other source of data on liquid industrial
wastes that was located was a report put out in 1968 by the U.S. Department
of the Interior, based on 1963 information (Ref. A-38). This report gave a
figure of 8.2 x 106 Mg (9 x 106 tons) of settleable and suspended solids in
industrial waste waters in 1963, primarily from the food products, primary
metals and paper industries.
Characteristics
Smith, in re-analyzing the IR£T waste quantities, also provided a more
useful breakdown of the heat content of these wastes, shown in Table A-9.
Robert G. Schwieger (Ref. A-39) gave some examples of typical industrial
wastes with significant heating value:
Average Heating Value
(As Fired)
Waste MJ/kg Btu/lb
Gases:
Coke-oven ' 45.82 19,700
Blast furnace 2.64 1,139
Carbon monoxide 1.34 575
Refinery 50.71 21,800
Liquids:
Industrial sludge 8.61 - 9.77 3,700 - 4,200
Black liquor 10.23 4,400
A-26
-------
Average Heating Value
(As Fired)
Waste^
Sulfite liquor
Dirty solvents
Spent lubricants
Paints and resins
Oily waste and residue
Solids:
Bagasse
Bark
General wood wastes
Sawdust and shavings
Coffee grounds
Nut hulls
Rice hulls
Corn cobs
MJ/kg
23.26 -
23.26 -
13.96 -
8.37 -
10.47 -
10.47 -
10.47 -
11.40 -
12.09 -
18.61 -
9.77
37.22
32.56
23.26
41.87
15.12
12.09
15.12
17.44
15.12
17.91
15.12
19.31
Btu/lb
10,000 -
10,000 -
6,000 -
3,600 -
4,500 -
4,500 -
4,500 -
4,900 -
5,200 -
8,000 -
4,200
16,000
14,000
10,000
18,000
6,500
5,200
6,500
7,500
6,500
7,700
6,500
8,300
Forecasts
The IR§T study included a forecast of combustible industrial solid waste
categories for the years 1980 and 1990, using an input-output forecasting model
developed by Clopper Almon at the University of Maryland. These forecasts are
shown in Table A-8. There is insufficient comparative data available to per-
mit any comment, other than it is interesting to note that by 1990 plant trash
is projected to be almost half of the total.
AGRICULTURAL WASTES
Types
Agriculture includes both the production of crops and the raising of
livestock; forestry is included, since trees are a form of plant crop. The
principal constituents of agricultural wastes are field crop residues, live-
stock manure, and forest slash, with small quantities of animal carcasses and
A-27
-------
TABLE A-9A. CHARACTERISTICS OF COMBUSTIBLE INDUSTRIAL SOLID WASTES (1967)
SIC Code
15, 16, 17
20
22, 23
24
25
26
27
30
31
19, 21, 28
TABLE A-9B
Construction
Food Products
Textiles and Apparel
Wood
Furniture § Fixtures
Paper
(SI Units)
Gg (dry)/y
1 179
648
264
13 927
413
9 213
Printing & Publishing 366
Rubber § Plastics
Leather
, 29, 32-39 All other
. CHARACTERISTICS OF
138
53
Mfg. 1 098
27 713
TJ/y
27 327
12 556
5 486
324 020
9 601
161 008
6 436
4 115
1 266
26 799
578 617
COMBUSTIBLE INDUSTRIAL SOLID
MJ/kg
23.17
19-38
20-78
23.26
23.26
17.47
17.56
29.84
23.65
24.41
, 20.88
(Average)
WASTES (1967)
(English Units)
SIC CODE 103 Tons (Dry)/year 1012 Btu/year Btu/lb
15,
20
22,
24
25
26
27
30
16, 17 Construction
Food Products
23 Textiles and Apparel
Wood
Furniture £ Fixtures
Paper
Printing § Publishing
Rubber £ Plastics
1,300
714
291
15,352
455
10,156
404
152
25.9
11.9
5.2
307.1
9.1
152.6
6.1
3.9
9,962
8,333
8,935
10,002
10,000
7,513
7,550
12,829
A-28
-------
TABLE A-9B. CHARACTERISTICS OF COMBUSTIBLE INDUSTRIAL SOLID WASTES (1967)
(Cont)
(English Units)
SIC Code 103Tons (Dry)/year 1012 Btu/year Btu/lb
31 Leather 59
19, 21, 28, 29, 32-39 All other Mfg. 1,210
30,548
1.2 10,169
25.4 10,496
548.4 8,976
(Average)
After Smith, Ref. A-36
agricultural chemical wastes. Forest slash includes those wastes left in the
forest during logging operations; logging residues accumulated at saw mills
may be included under agricultural or industrial wastes. Similarly, the
slaughtering and meat packing industry accounts for most of the animal carcass
wastes.
Agricultural wastes are almost entirely organic; exceptions are crop
residues such as rice hulls which contain significant concentrations of silica.
Not all agricultural wastes are directly combustible, however, with fresh
animal manure and wood wastes having very high water content. Because of the
variability of the moisture content of agricultural wastes, data on quantities
and characteristics must be stated on a dry basis for comparability.
Quantities
Estimates of the quantities of agricultural waste are shown in Table A-10.
Crop and animal wastes are generally calculated by applying waste generation
factors to statistics of agricultural production compiled by the U.S. Depart-
ment of Agriculture. For example, Taiganides and Hazen (Ref. A-40) have
reported the following properties of farm animal excreta:
Hens Swine Cattle
Animal weight 1.8 - 2.3 kg 45 kg 450 kg
(4-5 Ib) (100 Ib) (1000 Ib)
Wet manure, weight/day 0.11 kg 3.2 kg 29.0 kg
(0.25 Ib) (7.0 Ib) (64.0 Ib)
Total Solids, % wet basis 29.0 16.0 16.0
Volatile solids, % dry basis 76 85 - 80
A-29
-------
TABLE A-10. SOLID WASTE GENERATION FROM AGRICULTURE AND LOGGING
F.liasscn M»",T
IRef. 1) (Ref. 51
1966 19dd
Wet Wet Dry
Total Total Avail.
Crop Wastes 501 58d 152.1
Livestock Wastes 1418 1054 54 . d
Logging Wastes 25 45 25.3
Total 1942 14d5 210.2
,1 Sawmill Waste - - 12.5
0
Crop Wastes 552 125 Id"."
Livestock Wastes 1565 1110 5S.1
Logging Wastes 25 47 25.9
R.M.P (Irantha)i) Anderson Anderson
Caleu. IRef. 29) IRef. 5) [Ref. 51
1970 1971 1971 1971
Wet Dry Dry Dry
Total Total Total Avail.
TERAGRAMS PIUl YHAR
531 21
125" - 1S1 2-1
29." 50 5
3, S3 50
22.0
MII.I.IUX 01 "IU\S I'l.R U'.AR
590 25
l.iSd - 200 2(i
Poo 1 e
(.Ref. 41)
1973
Dry
Total
509
41
24
5"4
10.2
541
45
26
Kill s Al
(Ret. 42) (Ref.
1975 19
Dry Dry
Total Total
292
55
142 56
361
IS 17
522
56
157 40
ieh Forest Service
34, 43) (Ref. 44)
73 ' 1970 - 1973
Dry
Avail. Total "Available"
252
24
34 1046 41
310
17 17
278
26
38 1153.62 45
2140 Ihi: 251.'
412
342
Sawmill Waste
24 . 5
19
19
19
-------
From this data the following can be calculated:
Cattle
Wet manure, Mg/year/animal 0.0041 1.159 10.596
Volatile solids, Mg/year/animal 0.0009 0.158 1.356
The Statistical Abstract of the United States, (Ref. A-45) gives the following
farm animal population figures for 1970:
Cattle 112 303 000
Swine 56 655 000
Poultry 431 000 000
The resulting total manure and dry organic solids from these animals would be:
Wet Manure Volatile Solids
Cattle 1190 x 106 Mg 152 x 106 Mg
Swine , 65 9
Poultry 2 0.4
Total 1257 x 106 Mg 161.4 x 106 Mg (13%)
This calculation does not include sheep, horses, pets, and possibly some
poultry, so that the total wastes are understated.
The resulting wet basis figure of 1257 x 106 Mg (1385.6 million tons)
agrees reasonably well with the Eliassen (Ref. A-4) estimate of 1563 million
tons and the IR£T (Ref. A-3) estimate of 1140 million tons. The range between
the high and low estimates, 423 million tons per year, is about plus and minus
15% about the average of the three values. For gross calculations, the best
permitted by the available data, this is very close agreement.
The quantities calculated above are wet manure weights for the entire
animal population, not just confined animals. Less than one-fourth of the
total manure generated occurs in sufficient concentrations to make collection
and disposal necessary and economical (Ref. A-3). In addition, manures contain
major amounts of moisture, on the order of 84% moisture for cattle, 75% for
poultry, 90% for sheep and 95% moisture for hogs (Ref. 3). The weight of
dry solids requiring disposal is therefore a small fraction of the total wet
weight generated. This explains why the Alich, IR§T "available" and Poole
estimates are so small. For energy calculation purposes this dry, available
weight should be used, keeping in mind that the wet, available weight must be
used for physical handling calculations.
A-31
-------
Similar calculations can be made for agricultural crop wastes. Table A-ll
gives residue factors for major crops. Crop residues are usually defined as the
above-ground portions of the crop plant that are not harvested. There is
reasonable agreement between most of them, with two notable exceptions. The
maximum ratio of high to low values is 3.2 to 1, except for corn (11 to 1) and
sugar cane (32 to 1). The variation found for corn is quite possibly due to
an error by IR§T in the value assumed for harvest per acre. The IR§T data
imply 0.043 kg/m2 (0.191 tons/acre), while Poole's data imply a value of
0.57 kg/Mg/m2 (2.55 tons/acre), and Knutson's a value of 0.66 kg/m2 (2.96 tons/
acre). The variation found per sugar cane is more likely due to a difference
in the definition of residues. Poole's value specifically excludes bagasse,
considering it a food processing residue. Eliassen's estimate can probably
also be explained this way. Alich and Inman (Ref. A-46) show a range of
aboveground, dry biomass yields of 2.80 to 5.82 kg/m2 (12.5 to 26 tons per
acre) for sugarcane, which indicates that IR^T's residue figure of 6.16 (wet)
and 2.46 (dry) kg/m2 (27.5 tons (wet), 11 tons (dry) per acre) includes bagasse.
The NSF and ERDA sponsored work of Alich (Ref. A-34, A-43) is the most recent
available, and its exacting methodology commends use of the crop residue data
to all interested in this waste source.
The calculation of total residue quantities is made by applying the
Table A-ll factors to statistics of weight or area harvested. Investigators
who have developed these factors have established that three crops - corn,
wheat, and soybeans, in that order - produce two-thirds to three-quarters of
the agricultural crop waste total (Refs. A-4, A-3, A-34, A-41, A-43).
The quantity of forest waste varies according to the logging practices
followed and the definitions used in enumerating residue. When wood prices
are high, tighter control is kept on the generation of residues. The defini-
tion used also have an impact, because the conservative definitions followed
by the U.S.D.A. Forest Survey Teams include only salable wood left in harvested
areas. Wood under 10 cm (4 in.) in diameter and non-salable trees are not
included, nor are residues outside of harvested areas. The estimates by
Eliassen (Ref. A-4), IR£T (Ref. A-3), Poole (Ref. A-41), and Grantham and
Ellis (Ref. A-29) were made on this basis. Ellis (Ref. A-42) included all
residues in harvested areas, recognizing that new wood harvesting technology
such as chipping and hogging, would permit much of this residue to be pro-
fitably used.
An estimate made by the U.S. Forest Service (Ref. A-44) went even further,
estimating that non-commercial timber, forest harvest residues, other "removals"
(land clearing and changes in operational land use), and unused primary manu-
facturing residues total as much as 1046 x 106 Mg (1153.62 million tons) (dry).
Non-commercial timber is a one-time inventory, not an annually recurring
quantity. Unused primary manufacturing residues and other removals due to
changes in land use are outside the logging industry. When the definition of
forest residues is made comparable to the definitions used in other estimates,
the quantities are also comparable. Total forest harvest residues, including
small branches, culls, and bark are estimated to be 113 x 106 Mg (130 million
tons) (dry); residues on the more conservative growing stock basis are 41 x
106 Mg (45 million tons).
A-32
-------
TABLE A-ll. RESIDUE FACTORS FOR MAJOR CROPS
Unit Weight of Dry Residue Per
Unit Weight of Harvest
kg of
(Tons of
Eliassen IR§T Poole Knutson Alich Eliassen
Ref A-4 Ref A-3 Ref A-41 Ref A-47 Ref A-34, Ref A-4
Corn 9.42 5.63° 0.85 1.
Wheat 0.661 0.472 1.50 1.
f\
Oats 0.915 0.929 1.50 1.
Barley 0.733 0.820 1.25 1.
Cotton 3.42 3.20 3.00 1.
Soybeans 1.09 0.554 0.85
Rice 0.531 0.6026 0.85 1.
Sugarcane 0.004 0.129 0.23
Vegetables 0.102 0.114
a. Assuming 60% moisture in residue, except
43
35 1.10
0
0
0
5 2.45
35
for vegetable
b. Calculated by Parsons using Eliassen 's waste rates and
c. Recalculated by Parsons from IR&T data;
d. Recalculated by Parsons from IR§T data;
e. Recalculated by Parsons from IREJT data;
f . Calculated by Parsons from IR$T data
g. Excluded bagasse
original value
original value
original value
Wet
1.0
(4.5)
0.29
(1.3)
0.40
(1.8)
0.40
(1.8)
0.45
(2.0)
0.45
(2.0)
0.67
(3.0)
0.11
(0.5)
0.67
(3.0)
Residue per m2 of Harvest
Residue per Acre of Harvest)
IR§T
Ref A-3
Wet
0.60
(2.69)
0.21
(0.93)
0.41
(1.83)
0.45
(2.0)
0.42
(1.88)
0.22
(1.0)
0.76
(3.4)
6.16
(27.5)
0.75
(3.33)
residues which have
IR&T harvest
was 0.554
was 0.960
was 0.378
rates
Poole
Ref A-41
Dry
0.49
(2.17)
0.32
(1.43)
0.25
(1.13)
0.27
(1.21)
0.48
(2.12)
0.16
(0.71)
0.25g
(1.10)
80%
Knutson
Ref A-47
Air Dry
0.90
(4.0
0.34
(1.5
0.22
(1.0
0.31
(1.4
0.34
(1.5
0.67
(3.0
0.27
(1.2
- 1.0
- 4.5)
- 0.36
- 1.6)
-0.34
- 1.5)
- 0.34
- 1.5)
- 0.45
- 2.0)
-
- 0.83
- 3.7)
-
- 0.49
- 2.2)
-------
The annual quantities of logging wastes reported in Table A-10 are
expected to decline. Twenty years ago, perhaps only 50% of the above-ground
tree reached the consumer as a product; today, an estimated 20% is still
wasted, but in ten years wastage is expected to be near zero.
In developing Table A-10, an attempt has been made to identify a number
of knowledgeable sources rather than quote every possible reference. Given
the varying definitions and bases for calculation, there is reasonable agree-
ment. Once the difference between wet and dry bases is understood, it repre-
sents no problem, but the difference between "total" and "available" is not
always clear. Anderson (Ref. A-5) assumes that only 8% of the total crop wastes
would be available for conversion to energy, citing its wide dispersal. Alich
(Ref. A-34, A-43) assumes that 86% of the total crop waste is available, but
then reports livestock wastes from confined animals only in the total, and
assumes that 76% of that is available. Many authorities reject offhand the
use of agricultural residues as being "too diffuse" for consideration for
conversion to energy, but the quantitative studies to establish the probable
economics are still in progress (Ref. A-43). A simple comparison with urban
waste loadings demonstrates that the situation is one requiring additional
analysis. The highly urbanized County of Los Angeles now disposes of an
average of 0.85 kg/m2/year (3.8 ton/acre/year) while a typical city within it
such as Pasadena currently sends 1.52 kg/m2/year (6.8 ton/acre/year) to land-
fills. These values are "as generated" (water present) and include wastes
from all sources, including demolition rubble. Farm areas generate on a dry
basis (Ref. A-6) 1.01 kg/m2/year (4.5 ton/acre/year) for corn crop wastes,
and 0.67 kg (3.0 tons) each for rice, sorghum, peanuts, potatoes, and sugar
beets. The marginal costs for harvesting equipment to simultaneously cut
(and bale or pelletize) stalks, leaves, and stubble should be low and clearly
transportation costs to central processing plants should be as low as the
comparable urban case.
While this demonstrates that potentially large quantities of agricultural
wastes might be more readily available than previously considered, their
actual usage for conversion to energy (either by combustion or chemical
conversion to fuel) will be affected by factors other than average area
density. Waste generation in farm areas is highly seasonal. Alich (Ref. A-34)
says that 8.4% is generated in the first quarter of the year, 12.0% in the
second quarter, 41.6% in the third, and 38.0% in the fourth. This variation
can place severe constraints on processing plant utilization. The chief
constraint, according to preliminary discussions of field waste conversion
processes with agricultural authorities in Kansas and California, is probably
requirements for the waste in high value applications on the farm itself.
The discing under of crop wastes and manure for benefication of the soil
structure is widely practiced, although no information has yet been found on
minimum quantities truly needed for high crop yields. With increasing feed
costs, farmers are able to sell or use their field wastes (e.g. wheat straw,
corn stalks, and soybean waste) in prepared feeds that incorporate molasses
and urea. Wheat straw is now selling in the $20-$21 per Mg ($22-$23 per ton)
range. Alich (Ref. A-34) gives the following dispositions of crop residues.
A-34
-------
Returned to soil 73.6%
Fed without sale 19.0
Sold 4.0
Fuel 2.8
Wasted 0.6
100%
The quantities given in Table A-10, therefore, do not necessarily represent
true wastes that currently have no economic value, but rather residuals from
primary processes.
Characteristics
Schlesinger, Sanner and Wolfson (Ref. A-48) have calculated the heating
values of the dry volatile solids found in agricultural wastes, and give the
following figures:
Bovine waste: 16.54 MJ/kg (7,110 Btu/lb)
Rice Straw: 14.14 MJ/kg (6,080 Btu/lb)
5 Rice Hulls: 15.37 MJ/kg (6,610 Btu/lb)
Pine Bark: 19.42 MJ/kg (8,350 Btu/lb)
Corder (Ref. A-49) states that non-resinous woods and barks have higher heating
value ranges of 18.61 -19.77 and 17.21 - 22.79 MJ/dry kg (8,000 - 8,500 and
7,400 - 9,800 Btu/dry pound), respectively, while resinous woods and barks fall
in the ranges of 20.00 - 22.56 and 20.47 - 25.12 MJ/dry kg (8,600 - 9,700 and
8,800 - 10,800 Btu/dry pound), respectively. Typical moisture contents for
commercial hogged fuel are cited as 40-55%, while bulk density values fall in
the range of 0.096 to 0.208 g/cm3 (6 to 13 pounds per cubic foot). Pelletized
fuels made from numerous agricultural wastes have been inspected at Parsons,
all of which sank in water, implying that they have a density greater than
lg/cm3 (62.4 lb/ft3).
Distribution
It is of interest to note from the raw SRI (Ref. A-50) data that eight
states contain 67% of the available forest residues, nine states 60% of the
crop residues, and seven states 48% of the manure available. A computer
program developed by SRI is now being applied to the data to ascertain poten-
tial processing sites and the affect of seasonality on such a plant (Ref. A-43).
Eliassen (Ref. A-4) reports that over 70% of the wastes generated are due
to 3 of the 22 crops listed: corn, soybeans, and wheat. Field waste of corn
grown for grain accounts for 46% of the total crop waste. He notes that agri-
cultural experts have estimated "that of the weight of corn crops grown for
canning, about 50% is field waste, about 30% is process waste and less than
20% is actual corn in the can."
A-35
-------
The high degree of regionality associated with agricultural wastes does
not mean that conversion processes cannot be applied to them, but only that
specific analyses should be conducted to establish how best to utilize the
energy inherent in these residues.
Forecasts
IR§T (Ref. A-3) gives a total (dry basis) estimate for 1967 of 210.2 x
106 Mg (231.7 million tons), projected to 309 to 323 x 106 Mg (341 to
356 million tons) in 1980 and 362 to 407 x 106 Mg (399 to 449 million tons)
(dry basis) in 1980. Anderson's base is higher, but he uses less than half
the growth rate used by
A-36
-------
REFERENCES
A-l. Smith, F. A., "Comparative Estimates of Post-Consumer Solid Waste "
U.S. EPA/530/SW-148, May 1975
A-2. "Base Line Forecasts of Resource Recovery, 1972 to 1990," Midwest Research
Institute Report 3736-D, prepared for OSWMP, U.S. EPA, March 1975
A-3. "Problems and Opportunities in Management of Combustible Solid Waste,"
International Research and Technology Corporation (IR§T), prepared for
the National Environmental Research Center, EPA, October, 1972.
A-4. Eliassen, R., "A Comprehensive Assessment of Solid Waste Problems,
Practices and Needs," prepared by Ad Hoc Group for Office of Science
and Technology, Office of the President, May 1969.
A-5. Anderson, L.L., "Energy Potential From Organic Wastes: A Review of the
Quantities and Sources," Bureau of Mines Information Circular 8549, 1972.
A-6. "First Report to Congress, Resource Recovery and Source Reduction,"
(SW-118), 3rd edition, prepared by the Office of Solid Waste Management
Programs, U.S. Environmental Protection Agency, 1974.
A-7. "Technical-Economic Study of Solid Waste Disposal Needs and Practices,"
Vol. 1-1V, Report SW-7c, prepared by Combustion Engineering Inc. for
Bureau of Solid Waste Management, U.S. Department of HEW, 1969.
A-8. Black, R.J., et al., "The National Solid Wastes Survey, An Interim
Report," presented at the 1968 Annual Meeting of the Institute for
Solid Wastes of the American Public Works Association, Miami Beach,
Florida, October 24, 1968.
A-9. Niessen, W.R. and S.H. Chansky, "The Nature of Refuse," Proceedings
of 1970 National Incinerator Conference," pages 1-24, ASME, New York, N.Y.
A-10. Roberts, R.M., et al., "Systems Evaluation of Refuse As a Low Sulfur
Fuel," a report to EPA, Contract CPA 22-69-22, November 1971.
A-ll. "Municipal Solid Waste," National Center for Resource Recovery, Inc.,
Bulletin, Vol. Ill, Number 2, Spring 1973.
A-12. Darnay, A. and W.E. Franklin, "Salvage Markets for Materials in Solid
Wastes," U.S. EPA Report SW-29c, 1972.
A-13. "Third Report to Congress, Resource Recovery and Waste Reduction,"
(SW-161), prepared by the Office of Solid Waste Management Programs,
U.S. Environmental Protection Agency, 1975.
A-14. Niessen, W.R. and A.F. Alsobrook, "Municipal and Industrial Refuse:
Composition and Rates," Proceedings of 1972 National Incinerator
Conference, pages 319-337, ASME, New York, N.Y.
A-37
-------
REFERENCES (CONTINUED)
A-15. Refuse Collection Practice, prepared by the Committee on Solid Wastes,
American Public Works Association, Third Edition, printed by Interstate
Printers and Publishers, Danville, Illinois, 1966.
A-16. Kaiser, E., C.D. Zlit, and J.B. McCoffery, "Municipal Incinerator
Refuse and Residues," Proceedings of 1968 National Incinerator Conference,
ASME, New York, N.Y., 1968.
A-17. "Draft Summary Report, Comprehensive Solid Waste Management Plan for
Refuse Disposal and Recovery of Material and Energy Resources," prepared
by EPA Solid Waste Task Force and Leonard S. Wegman Co., Inc., Consul-
tant, under contract with NYS Department of Environmental Conservation,
October 21, 1975.
A-18. Snyder, N.W., "Energy Recovery and Resource Recycling," Chemical
Engineering, October 21, 1974.
A-19. Hekimian, K.K., R.M. Roberts, E.M. Wilson, "System Engineering Analysis
of Solid Waste Management in The Southern California Association of
Governments Region," Vol. Ill, June 1973.
A-20. Davidson, G.R. Jr., "A Study of Residential Solid Waste Generated in
Low-Income Areas," (SW-83ts), U.S. EPA, 1972.
A-21. Rhyner, C.R., "Domestic Solid Waste and Household Characteristics,"
Waste Age, April 1976.
A-22. Rigo, H.G., "Characteristics of Military Refuse," U.S.A. Construction
Engineering Research Laboratory, Champaign, Illinois (No date).
A-23. Dille, E.K., D.L. Klumb, G.W. Sutterfield, "Recycling Solid Waste for
Utility Fuel and Recovery of Other Resources," Presented at 1973
Frontiers of Power Technology, Oklahoma State University, Stillwater,
Oklahoma.
A-24. Roberts, R.M. and E.M. Wilson, "Systems Evaluation of Refuse as a Low
Sulfur Fuel, ASME Paper No. 71-WA/Inc.-3, presented at the ASME Winter
Annual Meeting, Washington, D.C., November 28 - December 2, 1971.
A-25. "Second Report to Congress, Resource Recovery and Source Reduction,"
(SW-122), prepared by the Office of Solid Waste Management Programs,
U.S. Environmental Protection Agency, 1974.
A-26. "Sewage Treatment Plant Design" American Society of Civil Engineers
Manual of Engineering Practice No. 36, 1959.
A-38
-------
REFERENCES (CONTINUED)
A-27. Bernard, H., "Everything you Wanted to Know About Sludge But Were Afraid
to Ask," Proceeding of the 1975 National Conference on Municipal Sludge,
Management and Disposal, Anaheim, CA, August 18-20, 1975, Information
Transfer Inc., Rockville, MD.
A-28. Vesiline, P.A., Department of Civil Engineering, Duke University,
Treatment and Disposal of Wastewater Sludges, Ann Arbor Science
Publishers Inc., 1974.
A-29. Grantham, J.B. and T.H. Ellis, "Potentials of Wood for Producing
Energy," Journal of Forestry, September 1974.
A-30. "The Economics of Clean Water-1973," U.S. Environmental Protection
Agency, 1973.
A-31. Farrell, J.B., "Overview of Sludge Handling and Disposal," in the
Proceeding of the National Conference on Municipal Sludge Management,
June 1974.
A-32. Wallis, I.G., "The Balance Between Waste Treatment and Waste Discharge
in the U.S., 1957-2000," Journal of the Water Pollution Control
Federation, V46N3, March 1974.
A-33. Standard Industrial Classification Manual, U.S. Government Printing
Office.
A-34. Alich, J.A., and J.G. Witmer, "Agricultural and Forestry Wastes as
an Energy Resource," Presented at the International Solar Energy
Society Annual Meeting, Winnepeg, August 1976.
A-35. "Handbook of Labor Statistics 1973," Bulletin 1790, U.S. Department of
Labor, Bureau of Labor Statistics, 1973.
A-36. Smith, F.A., "Quantities and Energy Content of Combustible Industrial
Solid Wastes," May 1, 1974, U.S. EPA
A-37. Huffman, G.L., "EPA's Program in Environmental Research in Wastes-As-
Fuels," for presentation at the Institute of Gas Technology Symposium
of "Clean Fuels from Biomass, Sewage, Urban Refuse and Agricultural
Wastes," Orlando, Florida, January 29, 1976, U.S. EPA.
A-38. "The Cost of Clear Water" Volume 1, Summary Report, U.S. Department of
the Interior, January 1968.
A-39. Schwieger, R.G., "Power from Waste," Power Magazine, February 1975.
A-40. Taiganides, E.P. and T.E. Hazen, "Properties of Farm Animal Excretia,"
Transactions of the American Society of Agricultural Engineers, Vol. 9
No. 3, 1966, pages 374-376.
A-39
-------
REFERENCES (CONTINUED)
A-41. Poole, A., "The Potential for Energy Recovery from Organic Wastes,"
in The Energy Conservation Papers, R.H. Williams, Ed., Ballinger
Publishing Company, 1975.
A-42. Ellis, T.H., "The Role of Wood Residue in the National Energy Picture,"
published in "Wood Residue as An Energy Source," Forest Products
Research Society, 1975, Proceedings No. P-75-13, pages 17-20.
A-43. Alich, J.A., et al, "An Evaluation of the Use of Agricultural Residues
As An Energy Feedstock: An Extension of Work," Interim Report,
June 1976, ERDA Contract No. E(04-3)-115.
A-44. "The Feasibility of Using Forest Residues for Energy and Chemicals,"
Prepared for the National Science Foundation by USDA Forest Service,
Report NSF-RA-760013, March 1976.
A-45. Statistical Abstract of the United States, Table 1973, U.S. Department
of Commerce, 1974.
A-46. Alich, J.A., and R.E. Inman, "Energy from Agriculture," Reprinted from
the Tenth Intersociety Energy Conversion Engineering Conference (IECEC),
1975.
A-47. Knutson, J., G.E. Miller, and V.P. Osterli, "Crop Residues in California-
Some Factors Affecting Utilization," University of California, Davis,
Division of Argicultural Sciences Leaflet 2872, February 1976.
A-48. Schlesinger, M.D., W.S. Sanner and D.E. Wolfson, "Energy From the
Pyrolysis of Agricultural Wastes," in Symposium: Processing Agricultural
and Municipal Wastes, New York, August 1972, published by the AVI
Publishing Company, Westport, CT.
A-49. Corder, S.E., "Fuel Characteristics of Wood and Bark Affecting Heat
Recovery," in "Wood Residue As An Energy Source," published by the
Forest Products Research Society, 1975, Proceedings No. P-75-13,
pages 30-32.
A-50. Computer output sheets received from Dr. Robert E. Inman, Stanford
Research Institute, Menlo Park, California, November, 1975.
A-40
-------
APPENDIX B
CATALOGUE OF WASTE-TO-ENERGY PROCESSES
INTRODUCTION TO WASTE TO ENERGY PROCESSES
As discussed in Appendix A, many wastes can be converted to energy. This
can be either heat energy for immediate use, or energy for use later, in the
form of fuel. The basic waste-to-energy conversion processes can be con-
veniently categorized into the areas of combustion, thermo-chemical, biologi-
cal, and mechanical systems. While these categories are not mutually
exclusive, they do tend to indicate the primary action taken on the raw waste
and some secondary action may then follow.
Combustion produces heat energy for immediate use. Technically, it is the
only process that releases energy, since all the other processes produce fuel
that must be combusted later when the energy is wanted for use, whether in a
furnace, a car engine, or a stove.
Thermo-chemical processes such as pyrolysis, hydrogasification, and
hydrogenation use heat and/or chemicals to break down complex materials and
produce new solid, liquid, or gaseous fuels.
Biological processes also break down complex materials and produce new
solid, liquid, or gaseous fuels.
Mechanical processes include sorting, size reduction, and drying to
separate the combustible portion of wastes from the remainder. Unlike the
other processes, which make both chemical and physical changes in the wastes
to produce energy or a fuel, mechanical processing makes only a physical
change.
There are some ten to twenty or more variations of each of the four basic
processes given above. None of them are technologically new, although in some
instances their application to the disposal of wastes has only been recently
made. The divisions between these processes are not necessarily clear cut.
Thermo-chemical processes, for example, combust some waste to produce the
heat they need, and mechanical processing is a necessary step in many varia-
tions of the other processes.
A number of reasons exist for the proliferation of waste-to-energy con-
version processes. There are many waste types, many energy/fuel markets, and
many of the processes are technically very promising but not fully developed.
The choice of a process for a given application depends on technical and • •
economic factors, not as they are today, but as they will be in the future
B-l
-------
when the facility is operating. For example, in the days before stringent
air pollution control regulations and when energy costs were low, municipal
incinerators were built for waste disposal without energy recovery. When air
pollution control legislation became effective, the design of incinerators
included steam generators. These systems lowered the exhaust gas temperature
enough to permit electrostatic precipitators to remove the particulate pollu-
tants. These incinerators were still built essentially for waste disposal,
however, and no special effort was made to insure that the quantity, quality,
and location of the steam they produced was suitable for use. Today, as
waste disposal and energy sources both become more expensive, waste disposal
costs can be off-set by the revenue from energy sales. New incinerators are
now being built predominantly with this in mind. In fact, incinerator is now
a misnomer; these new facilities are more correctly called "waste-fired steam
generators."
The costs of waste-to-energy processes vary over a broad range due to
process differences, and because some produce energy while others produce a
fuel. As mentioned at the beginning of this section, combustion is technically
the only process that releases energy, with all the other processes stopping
at producing a fuel. The cost of a combustion facility also typically includes
the energy recovery system, usually a boiler, and may also include an energy
use system, such as electric generators. It is, therefore, generally the most
expensive per ton of waste disposal. However, because a modern combustion
system converts waste to a final energy form in one facility, it may be the
most efficient in terms of energy conversion. In terms of capital costs, the
thermo-chemical processes tend to be the next most expensive, biological
processes next, and mechanical processes least. It is extremely important to
note that these are general tendencies, and that process costs are very
sensitive to the specific situation.
In comparing similar processes, a useful concept is net energy output
efficiency. Basically, efficiency is output divided by input, and expressed
as a percentage. If waste with an energy content of 1000 MJ is put into a
process and 500 MJ comes out, the process is 500/1000 = 50% efficient. Waste
is not the only input to a waste-to-energy process, however; there is also
electrical or mechanical energy and often auxiliary fuel. Simply adding these
energies to the waste input is inaccurate, because, in a closed system, a
portion of the output energy must be diverted to make these inputs. There-
fore these inputs are debits against the gross energy output and the result
is net energy output. Net energy output efficiency is then net energy output
divided by waste energy input, expressed as a percent. This means that effi-
ciency figures will be lower than most developers would like to see, and that
systems that consume more energy than they produce will show negative
efficiencies.
MECHANICAL PROCESSES
The initial steps in waste-to-energy systems are a series of mechanical
processes for controlling and preparing the waste. These processes can con-
stitute an entire system producing a fuel, or can be the first of several
stages that produce a fuel or energy. Mechanical processes can be grouped
under four headings: materials handling, shredding and other size reduction,
sorting and classifying, and drying.
B-2
-------
Treatment of waste makes subsequent steps more efficient and improves
the potential for energy and material recovery. Many wastes are highly
heterogeneous mixtures of materials. The product becomes much more homogeneous
by shredding the waste to achieve a relatively uniform, small particle size;
by sorting and classifying the waste to concentrate the energy fraction; and
by drying the waste to remove water. This improves the ease and efficiency
of energy conversion and increases the quantity of energy that can be
recovered from a given amount of raw waste. In some instances the improvement
in energy recovery is sufficient to justify the extra expense of mechanical
processing. In other cases, processing is justified because it makes a
salable fuel or permits the use of existing boilers and other equipment.
Materials Handling
Materials handling processes include receiving, weighing, moving, and
storing of raw materials, materials in process, process residues, and products
of processing.
Many waste gases contain a significant amount of sensible heat that can
be recovered in a heat exchanger for use. Some waste gases, such as coke oven
gas, can be used as low grade fuels in other processes, while others have
value for the chemical raw materials they contain. In any case, the gas
usually requires removal of contaminants before it can be transported by
pipeline, stored, and eventually used. The cleaning and removal of contami-
nants to make waste gas suitable as fuel, as a chemical raw material, or to
permit disposal to the atmosphere, is very often the most complex and expensive
part of the entire gas handling problem and may govern the ultimate use or
disposal of it.
Liquids can be received at a disposal facility by tank truck or pipeline
and weighed in the truck or measured by a pipeline flow meter. Many waste
liquids are highly corrosive or toxic, requiring care and special materials
and equipment to handle them successfully. Others are very viscous and will
not pour without some application of heat, while still others contain a
heavy load of solids that tend to settle out and plug handling equipment.
There is no single method of handling and disposal of waste liquids that will
satisfy all requirements; the specific waste liquid or class of waste liquids
must be considered on an individual basis.
Solid wastes nearly always arrive at the disposal facility by truck,
which is weighed before and after dumping the waste in order to record quanti-
ties for billing and plant operational records. There are presently two
basic arrangements for receiving the waste. One utilizes a deep concrete pit
into which the waste is dumped from the truck and subsequently handled and
moved into further processing equipment by an overhead travelling crane
equipped with a clamshell or grapple type bucket. The second uses a level
concrete floor upon which the waste is dumped from the trucks and subsequently
piled and moved into further processing by a large rubber-tired front-end
loader. Both arrangements have been used successfully. It is claimed that
the level floor (tipping floor) is more efficient at rates below 36 Mg
(40 tons) per hour, while the pit type operations is more efficient at rates
B-3
-------
greater than this (Refs. B-l, B-2, B-3). There is considerable doubt on this
matter and the designer should carefully consider all economic, operational,
and safety factors before proceeding.
Recovered metals, glass, and residue prior to truck or rail shipment may
be stored temporarily in a metal bin. The real storage problem may be found
with the light fraction of shredded waste. The bulk density is low, approxi-
mately that of the original mixed waste, and hence the required storage volume
is very large. The material has been found to compress in storage and be very
difficult to remove. In straight sided storage structures or silos it has a
strong tendency to bridge, making removal from the bottom difficult and
hazardous. There are several proprietary storage and retrieval systems that
have been or could be used for shredded waste. To date, those that have been
tried have not shown outstanding success or dependability.
The transport methods for solid waste vary considerably, depending on the
method of energy and/or material recovery. The various processing methods
involve one or more types of material handling equipment such as slat or pan
conveyors, belt conveyors, screw conveyors, vibrating feeders, pneumatic
conveyance pipelines, and water slurry pipelines.
Some special handling characteristics of solid waste are:
• High dirt, metal, and glass content causes processed waste to be
abrasive and especially hard on pneumatic conveying equipment. As
processing progresses and more dirt, metal, and glass is removed, this
abrasive characteristic diminishes, but never disappears.
• The tendency to bridge in storage has been discussed above. In
addition, the material tends to migrate into conveyor drive machinery
and clog it unless it is well protected.
• The handling and processing of waste creates dust, which should be
contained and/or removed for health reasons. Dust control equipment
may be a large part of the investment in a processing facility.
• The high moisture content contributes to corrosion of storage and
processing equipment.
• Hazardous chemicals, pressurized containers, and explosives may be
present.
Shredding and Other Size Reduction
Solid wastes have a wide range of particle sizes, making handling and
processing difficult. Primary and secondary size reductions are relatively
expensive, but often necessary. There are eleven basic types of size reduction
equipment commercially available - crushers, cage disintegrators, shears,
shredders, grinders, cutters and chippers, rasp mills, drum pulverizers, disc
mills, wet pulpers, and hammermills (Ref. B-4, B-5, B-6). The general term
"shredder" covers all of these types except for the wet pulpers.
B-4
-------
The most popular type of shredder is the hammermill, which has a series
of hammers, either rigid or swinging, attached to a horizontal or vertical
rotating shaft. The hammers range in weight from 6.8 to 225 kg (15 to 500 Ib)
each and the forces they generate in the hammermill are sufficient to break
almost any waste. The major problem materials are cable, which tends to
become entangled in the hammers; rugs, which absorb the hammer impacts; and
hard steel objects such as automobile crankshafts, which dull the hammers.
Hammermills are used to shred both MSW and forest waste. In the forest
products industry, shredders are called hogs; a hammermill is a no-knife hog.
A knife-type hog, or chipper, is a large, heavy drum with hardened steel
inserts; wood fed into the spinning drum is rapidly chipped. Both the
no-knife and knife-type hogs are used to produce chipped, or hogged, fuel.
The knife-type, being more sensitive to tramp iron and rocks, is usually used
on debarked logs or off-spec lumber, while the no-knife type can handle bark
and logs (Ref. B-7).
MSW can also be shredded by grinders. These are usually vertical shaft
machines with gear-like wheels that grind the refuse against the housing side
walls. The waste flows through the machine assisted by gravity. Vertical
machines, whether grinders or hammermills, generally do not reject hard-to-
shred items, but reduce all objects to a relatively uniform size.
Power requirements for shredding municipal solid waste range from 6.5 to
35 horsepower per ton-hour, with the power requirement increasing as the output
particle size becomes smaller (Ref. B-6). Medium duty mills used for
shredding MSW typically require 20 kW/Mg-h (25 horsepower per ton-hour) of
throughput (Ref. B-6, B-8). A waste shredding facility processing 907 Mg
(1000 tons) per day in one 8-hour shift would require shredders totalling
2.3 MW (3,125 horsepower).
Capital, maintenance, and operating costs for shredders have been
investigated by the Midwest Research Institute for the U.S. EPA (Ref. B-6).
Conclusions are based on very limited data and should not be used for any
detailed design or study projects. As shown in Table B-l, MRI estimated that
the total cost of shredding MSW would vary from $2.28/Mg ($2.07 per ton) at a
capacity of 9 Mg/h (10 tons per hour) to $1.44/Mg ($1.31 per ton) at 45 Mg/h
(50 tons per hour), and rising again to $1.53/Mg ($1.39 per ton) at 91 Mg/h
(100 TPH). These costs are much lower than the costs reported in the EPA
publication "Decision-Makers Guide in Solid Waste Management," which says that
"for projects in which EPA has been involved, costs per ton range from $8.60
up to $10.60 ($9.48 to $11.68 per Mg)" (Ref. B-9). These costs, however, are
total project costs that include hauling and land-filling costs for the
shredded refuse. Other cities have reported lower costs, between $4.52 and
$5.29 per Mg ($4.10 and $4.80 per ton), for similar projects (Ref. B-9).
A number of technical papers have been written on municipal solid waste
shredding (Refs. B-ll, B-12, B-13). "Solid Waste Shredding and Shredder
Selection" (Ref. B-5) has a list of shredder installations as of mid-1974 and
examples of shredder specifications; MRI1s "Fine Shredding Study" (Ref. B-10)
has a 1975 survey of shredders used or to be used for size reduction of MSW
in the United States.
B-5
-------
TABLE B-l. ESTIMATED COSTS OF SHREDDING MUNICIPAL SOLID WASTE,
NOVEMBER 1975
(SI Units)
Capacity
Mg/h
9.07
45.4
90.7
a/
Capital cost—
$/Mg
0.21
0.21
0.21
Maintenance cost—
$/Mg
0.33
0.55
0.77
c/
Operating Cost—
$/Mg
1.74
0.60
0.55
Total cost
$/Mg
2.28
1.44
1.53
a/ Assuming $4409 per Mg-h, zero salvage value, lifetime of 20 years, 6%
interest rate, and 7 hours of running per day;
b/ MRI estimate;
c/ Assuming 21 kWh/Mg, $0.02/KWh, and 1 1/2 men for 8 hours per day at $7 per
hour.
(English Units)
Capacity
tons per hour
10
50
100
a/
Capital cost—
$ per ton
0.19
0.19
0.19
Maintenance cost—
$ per ton
0.30
0.50
0.70
c/
Operating cost—
$ per ton
1.58
0.62
0.50
Total cost
$ per ton
2.07
1.31
1.39
a./ Assuming $4,000 per ton hour, zero salvage value, lifetime of 20 years,
6% interest rate, and 7 hours running per day;
b_/ MRI estimate;
c/ Assuming 25 horsepower per ton/hr, $0.02/kWh, and 1 1/2 men for 8 hours per
day at $7 per hour.
Source: Ref. B-10
The major advantage of dry shredding is the ability of the larger
equipment to handle a wide range of mixed raw waste directly. High operating
and maintenance costs could be considered the penalty for this versatility.
All of the equipment discussed above is for shredding relatively dry
waste. It is also possible to shred waste wet in a machine called a hydra-
pulper, which is essentially an oversize blender. A complete waste disposal
B-6
-------
and resource recovery system using a Hydrapulper and processing wastes in a
slurry has been demonstrated at Franklin, Ohio (Ref. B-14) and is planned for
Hempstead, New York (Ref. B-15).
In operation, MSW is fed into the Hydrapulper, where a high speed
rotating cutter chops the pulpable materials into a 3 to 4% solids-in-water
slurry. This slurry, containing paper, food waste, plastics and most of the
other organics, as well as glass and small pieces of metal, is withdrawn
continuously from the bottom of the pulper. Larger, heavier objects—cans,
stones and large pieces of metal--are ejected through a chute and removed.
Recycled water is introduced into the system through the same chute, so that
the heavy materials must drop down the chute against a counter-current flow
of water, which very effectively carries the lighter materials back into the
Hydrapulper.
A major advantage of the Hydrapulper is that it permits a very high
fraction of the organic, combustible material to be recovered in a homogeneous
form. A major disadvantage is the high volume of water required and the need
for removal of the water before many energy recovery processes could be
efficiently employed.
A semi-wet refuse pulverizing system under development in Japan
performs the classifying function at the same time (Ref. B-16). The system
separates components by utilizing the differences in resistance to destruc-
tion. It consists of a horizontal rotating drum with screens having two kinds
of scrapers rotating at different speeds in both screens. Refuse moving
through the drum is classified by the screens according to the pulverizing
time and screen size, which in turn depend on the material. A limited amount
of water is used to reduce the strength of papers, permitting their extraction
at lower power consumption.
Sorting and Classifying
Sorting and classifying processes include manual sorting, screening,
magnetic metals separation, magnetic eddy current separators (aluminum
magnet), wet and dry density separators, optical sorters, and electrostatic
separators. Many of these systems were originally developed in the mineral,
agricultural, or manufacturing industries. All of them, however, have been
especially adapted for handling wastes.
Manual Sorting--
Hand picking and sorting refuse from conveyors is the traditional means
of separating solid waste. As late as 1968 it was reported to be the most
widely employed separating technique (Ref. B-17). Two prominent manual
sorting facilities were the Lone Star Organics plant of Metropolitan Waste-
Conversion Corporation in Houston, Texas, and Sanitary Refuse Collectors in
Montreal, Quebec (Ref. B-18). Both used human operators to recover salvage-
able materials and bulky wastes; ferrous metals were removed by magnets. Both
plants closed in the early 1970's for economic reasons.
B-7
-------
At the Lone Star plant, it has been reported that approximately 551 to
827 kg (1/2 to 3/4 ton) of newsprint and cardboard could be handpicked from
mixed refuse by one man in an hour, corresponding to 1.4 to 2.2 man-hours
per Mg (1.3 to 2 man-hours per ton) of raw refuse (Ref. B-17).
Clean Steel Company of Carson, California is currently using manual
sorting as part of a car-breaking operation. Stripped, crushed automobiles
were shredded and then passed through a series of screens and other mechanical
separators. Human operators hand picked the product conveyors coming from
each mechanical separator to remove either impurities or the desired salvage
material.
In sorting municipal solid waste there will be few times when manual
sorting can be economically justified.
Screens--
If the materials have definable size differences, a screen may be used
to separate them. The most common form of classifying screen is the trommel,
an inclined, rotating drum with sized holes. It is commonly used to remove
dirt from fuel bark before it is hogged or to sort wood chips (Refs. B-7,
B-19). Similar devices have been proposed for classifying municipal solid
waste into a fraction to be reground and a fraction for further separation
processing (Refs. B-17, B-20). Other dry screen designs include vibrating
tables, rotating discs, and air blown tables (Ref. B-7). Most of these have
been developed in the mineral extraction industries, and design for screening
municipal solid waste is still relatively new and requires experimental
verification (Ref. B-17).
A trommel has a natural drying effect that can be enhanced by adding
a hot air blower, which may be of benefit in reducing clogging of moist
waste (Ref. B-21).
Screens are also used in wet classification processes. The Black
Clawson Company's process being demonstrated at Franklin, Ohio uses a number
of cylindrical screens kept clean by the pulsating action of a rotor turning
inside the screen (Ref. B-14).
Magnetic Metals Separation--
The separation of ferrous (iron) metals from municipal solid waste is
done magnetically. A simple permanent magnet may be sufficient for removing
an occasional nail or other piece of tramp iron from bark or sawdust fuel
passing on a conveyor belt. For removing magnetic metals from municipal
solid waste, however, either a drum type or a belt type system with electro-
magnets is preferred. They are similar in principle with a magnet being
placed inside the drum or behind the belt to hold the ferrous metals while
the remainder of the refuse falls away. Then, as the drum rotates or the belt
moves, the ferrous metals are carried beyond the influence of the magnets and
also fall away. For thorough separation it may be desirable to have a multi-
stage system, because iron and steel tend to carry along pieces of paper,
cardboard and plastic from the refuse (Ref. B-18).
B-8
-------
Eddy Current Separators--
The separation of non-ferrous metals, such as aluminum, copper, and
zinc from municipal solid waste using eddy current effect has been demonstrated
by Occidental Research Corporation (Ref. B-22), Combustion Power Company
(Refs. B-4, B-23), Eriez Manufacturing Company (Ref. B-17), and others. Most
non-ferrous metals and alloys are non-magnetic, so the steady magnetic field
of an ordinary magnet has no effect on them. However, in a moving magnetic
field these metals develop an opposing eddy current field and are repelled.
A linear induction motor placed underneath a conveyor belt carrying the refuse
can provide the necessary field to deflect non-ferrous metals into a collection
bin. In practice, a significant percentage of other materials is also
deflected. Certain metals, notably stainless steel, are not affected. The
repulsive force depends greatly on the size and shape of the metal piece to be
deflected. These factors limit the present usefulness of eddy current
separators.
Density Separators: Dry and Wet--
Density, or gravity, separators include air classifiers, vibrating tables,
dry fluidized beds, light and heavy media flotation, and rising current
separators. They all work by creating a condition in which the relatively low
density materials are carried up by air or a liquid, while the higher density
materials drop out. Because the low density materials tend to be combustibles
and the high density materials tend to be non-combustibles, density classifiers
are a relatively efficient means of separating them.
In an air classifier, refuse is allowed to drop off the end of a conveyor.
As it falls, it is subjected to a flow of air either vertically upward or
horizontally, which blows the lighter materials into a collector while the
heavier materials fall into another collector. Internal arrangements of
baffles, air blowers, collectors, and other equipment can be varied to suit
the job (Refs. B-4, B-18).
Three different refuse-to-energy systems, supported in part by U.S. EPA
grants, that use vertical air classifiers are being developed in St. Louis,
San Diego, and Menlo Park (CA). The results they have achieved using air
classifiers are similar: the light fraction represents 75% to 80% of the
input and is about 90% combustible, with the remaining being light foil and
some fine glass and dirt (Refs. B-22, B-24). The chief drawback of air
classifiers is that they classify materials not only by density but also by
form. A piece of aluminum foil, for example, may rise if it is a sheet but
fall if it is balled up. For best results the input to an air classifier
should be shredded, dried, and screened to a uniform size.
The U.S. Bureau of Mines has done extensive studies with wet and dry
density classifiers, and has developed a prototype materials separation
system (Refs. B-20, B-25, B-26). Stanford Research Institute has developed a
zig-zag air classifier, originally for sorting edible beans, that also shows
promise in classifying municipal refuse. Its chief advantage is that the
refuse is bounced and tumbled at each turn of the separation column, giving
better classification of the lights and heavies (Ref. B-18).
B-9
-------
Rising current separators are very similar in design to air classifiers,
except that water, rather than air, is the separating medium (Refs. B-14,
B-23).
Slightly different techniques are used in froth flotation, heavy media,
and fluidized bed separators, all of which use the buoyancy characteristics
of the materials to be separated. Froth flotation can be used to recover
finely ground glass from a mix of glass chips, rocks, brick, bones, and
heavy plastic. The mix is slurried with appropriate reagents that are
selective for glass; a froth is created, and the glass is floated out of the
mix and washed (Ref. B-22). Heavy media flotation takes advantage of the fact
that heavy solids, such as ferrosilicon, magnetite, or galena, when finely
ground and mixed with water, provide a suspension that closely duplicates
the properties of a true heavy liquid. Specific gravities ranging from
1.24 to 3.4 can be made. Pure aluminum, with a specific gravity of 2.70, can
be made to float or sink, as required. As with many solid waste separation
techniques, practical and economical equipment requires considerable testing
and modification to be truly effective and R§D is continuing (Refs. B-17,
B-23).
Fluidized-bed separators have a bed of fine material, such as sand,
through which air is blown; the bed then behaves very much like a true liquid
(Ref. B-18). Warren Spring Laboratory of the British Ministry of Technology
demonstrated a fluidized-bed separator in 1967, and others have tried the
technique (Ref. B-27).
Density separators used in the agriculture and mineral industries, such
as stoners, Osborne tables, jigs, and Wilfley tables have been tried with
municipal refuse. The usual experience is that municipal refuse is so
heterogeneous that extensive modification to the equipment is necessary for
it to work (Refs. B-4, B-17, B-18).
Other Separation Methods--
The previously mentioned techniques are the principal ones being used
or developed for classifying municipal solid waste. Any physical, chemical
or electrical property that can be measured offers the possibility of develop-
ment of a new sorting system. Two of the more important are optical sorting
and electrostatic separation.
Optical sorting has been developed to separate the three predominant
glass colors: clear (flint), green, and amber. The machine, manufactured
by the Sortex Company of North America, is able to sense and reject material
having light transmission qualities different from a standard glass. Only
one color can be sorted at a time, and the glass must be graded by particle
size. Individual particles drop through a light beam in an optical selection
box, where their color is compared to a filter. When an off-color piece
crosses the beam, electronic circuits cause a pneumatic valve to open and an
air jet blows the piece to one side after it falls clear of the optical
selection box. The classified glass still has a small percentage of particles
that should have been rejected, varying with the mix in the input glass and
the processing rate (Refs. B-4, B-23, B-25).
B-10
-------
Electrostatic separation of conductors and non-conductors has been
demonstrated. Dry waste to be separated is exposed to a high intensity
electrostatic charge and falls on a rotating drum having the opposite charge.
Materials that conduct electricity, such as metals, lose their charge rapidly,
but non-conductors, such as glass and plastics, retain their charge and are
attracted to the drum. As the drum rotates, the conductors fall off into a
bin, but the non-conductors cling to the drum until wiped off into a second
bin (Refs. B-4, B-23, B-28, B-29). Electrostatic separation is size-sensitive,
and would have to be repeated to separate wastes having a range of particle
sizes. Equipment made by at least two manufacturers have been used in industry
successfully.
Drying systems can be classified by the manner in which the material is
heated and moved through the dryer. There are four types of heating systems:
convective, such as a hair dryer; conductive, such as a clothes iron;
radiative, such as an infra-red lamp; and dielectric, such as a microwave
oven. At the present state of the art, only convective systems are of suffi-
ciently low cost to be considered for drying waste.
Among convective dryers, the eight systems for moving the material
through the dryer are individual trays, band or belt conveyors, wiped trays,
vibrating conveyors, rotary drums, fluid beds, airlifts, and sprays. Except
for the first two, all are used for drying waste materials.
Multiple hearth dryers are a stack of wiped trays, each of which has one
or more radial openings. Either the tray or the wiping arm can rotate;
material falling on one tray is pushed around to the next opening where it
falls to the shelf below and is leveled by a raking arm. Hot air is circulated
through the enclosed stack of trays to carry off moisture. This type of dryer
is often used with sewage sludges, which have a high water content; in some
designs the sludge is burned on the lower trays to provide the heat for drying.
Vibrating conveyors, rotary drums, and airlifts are also used for
sorting and can often have a dual function.
Air is used in vibrating conveyors to partially fluidize the waste for
better particle separation. The natural drying effect that this provides can
be further enhanced by using hot, dry air.
A rotary drum trommel screen has a natural drying effect that can be
enhanced by adding a hot air blower (Ref. B-21). Rotary drum driers used to
dry agricultural products and hogged wood fuel are made by a number of
companies. Several of these are proposing systems for preparing dry fuel from
municipal solid waste (Ref. B-30). Rotary drum dryers usually have vanes or
"flights" on the outer shell and down the center of the drum to lift the wet
material and control its fall through the hot gas stream. By controlling the
amount of oxygen present, very high temperatures can be used without combustion
or explosion.
B-ll
-------
An airlift is essentially the same as an air classifier, and if drying
requirements are relatively low, one piece of equipment could perform both
functions. For more drying effect, a recirculating system might be required,
so that only the driest pieces are light enough to be carried out of the
airlift. If this is done, close control on input particle size is necessary.
One company makes a "Hot Hog" that is a hammermill, air classifier and dryer
all in one. Bark or wood is dropped down a stack against a flow of hot air.
At the bottom of the stack is a shredder that chops the feed until the pieces
are sufficiently small and dry to be lifted back up the stack by the hot
gases and out to a separator (Ref. B-7).
A rotary dryer is capable of handling a large throughput of material,
but the evaporative capacity is relatively low. The airlift dryer has a much
higher evaporative capacity, but a small throughput of material. The fluid-
bed dryer has both a high throughput and a high evaporative capacity. Wet
wastes are continuously introduced to the drying chamber and discharged as a
dry product. Hot gas is blown through a distributor plate to partially or
wholly suspend the particles in the hot gas stream. Any particles carried
out by the exhaust gases are caught in a dust collector.
Spray dryers are very much like airlift dryers, and can handle high
volumes of liquids or sludges. The waste liquid is allowed to flow out over
a spinning disc, which flings it off in an umbrella-shaped cloud of droplets.
The hot gas stream is usually introduced in parallel with the droplet cloud,
and drying is accomplished in a fraction of a second. Spray driers work best
when handling large volumes of solutions; for feed rates below 500 kg/h
(1100 Ib/hr), drum dryers become cheaper.
Waste Derived Fuels
Refuse can be burned without auxiliary fuel in its raw, as-received
state, over a rather wide range of compositions. Von Roll of Zurich,
Switzerland, has studied (Ref. B-4) mixtures of MSW, water, and inerts and
established "self burning limits." Selected values from their work, at the
limits of combustion, are as follows:
Water, wt-% Non-combustibles, wt-% Combustibles, wt-%
50 25 25
40 35 25
20 55 25
0 55 45
This information was developed with municipal solid waste in mind, but is
generally applicable to other similar waste derived fuels. Water and non-
combustible materials do not contribute to the heating value of the waste, so
the greater the percentage of combustibles the more efficient the system will
be.
B-12
-------
A number of systems have been developed for processing waste so that it
can be more conveniently handled, stored, and burned. Although the processes
developed by various companies have some proprietary features, they all include
initial shredding, magnetic removal of ferrous metals, and some sort of
classifying system.
One of the simplest fuel-from-waste preparation processes takes bark from
a tree de-barker, passes it over screens to separate the dirt, and then
through a shredder. In order to remove more of the dirt and gravel while
retaining a maximum of the combustible bark, more sophisticated screening and
classifying devices have been developed, including rotary screens and fluid
bed separators (Ref. B-7). At one time, large chunks of bark were hogged
down to 5.0 to 7.5 cm (2 to 3 inch) nominal size for burning on grates, but
today it is becoming more common to hog the bark or wood down to less than
0.5 cm (1/4 inch) nominal size for suspension firing in modern design furnaces.
The heating value of dry, resin-free wood is about 19.31 MJ/kg
(8,300 Btu/lb), with little variation. Resin has a heating value of about
39.32 MJ/kg (16,900 Btu/lb) so resinous woods and bark have higher heating
value than resin-free wood, going as high as 25.03 MJ/kg (10,860 Btu/lb)
(Ref. B-31).
Finely hogged wood has served well as a waste fuel, but it also has close
to 50% moisture, low bulk density, and non-uniform burning characteristics.
In order to remedy these problems, various drying and pelletizing processes
have been developed. One of these utilizes wet grinding to achieve a particle
size of 3 mm (1/8 inch) in diameter and 12 to 18 mm (1/2 to 3/4 inch) long,
and have a heating value of 20.93 MJ/kg (9,000 Btu/lb). In pellet form, the
wastes are hydroscopically stable, ignite at 282°C (540°F), and burn at
2760°C (5000°F) (Ref. B-32).
Municipal solid waste can be similarly processed to produce a fuel, but
additional steps are necessary to remove the ferrous metals and glass. In the
hogged fuel processing, it is usually sufficient to place a permanent magnet
above the flow at a convenient point to catch the occasional piece of tramp
iron, but in a municipal RDF facility a more sophisticated system is required
to capture all the food cans. A classifier and two stages of shredding are
also often necessary. The new RDF facility being built for the City of
Chicago is a good example of this type of system. Details are presented in
Section 5.
Combustion Equipment Associates, Inc. (CEA) has taken this type of
processing one step farther. The first version of their Eco-Fuel was a
confetti-like fluff. To obtain the best suspension burning characteristics,
however, a finer, more uniform product was necessary. This usually means a
very high power requirement for grinding, but by adding a small amount of an
inorganic chemical to what is essentially Eco-Fuel, it can then be ground in
a hot ball mill with a relatively low power requirement. As a result, it is
claimed that Eco-Fuel II can be ground to a size of 0.15 mm (0.006 in.)
economically (Ref. B-33). The combination of chemical agent and hot ball mill,
it is reported, reduces shredding and air classification requirements, uses
less power, promotes recovery of ferrous and non-ferrous metals, and is
B-13
-------
environmentally attractive (Ref. B-34). Eco-Fuel II is a fine, dry,
free-flowing powder with a high bulk density and an improved heating value.
The fineness of grind and the uniformly high heat content facilitate combus-
tion control and result in a complete burning of the organics. The fuel is
flexible; it can be briquetted or slurried in oil (Ref. B-33).
In selecting a form for briquetting Eco-Fuel II, it was found that the
shape of the common home barbecue briquette answered the problems inherent
to coal conveying equipment (Ref. B-35). In creating this briquette, various
binders may be used. Hydraulic pressure is then applied to produce the final
product. The briquettes produced have sufficient anti-roll quality and weight
to hold position on outdoor, utility-type conveyors. This product shape and
stability has been successfully tested for its transport quality in existing
coal conveying systems. In a typical analysis with an initial binding
composition of 10% water, the bulk density was approximately 0.72 g/cm3
(45 lb/ft3), volatile matter around 65%, and total moisture remaining 4%.
CEA is developing a process for incorporating Eco-Fuel II in heavy oil
(Ref. B-33). In Eco-Fuel II and oil mixtures containing up to 40%
Eco-Fuel II, the final fuel product will retain 90% of the original heating
value of the oil in a given volume. For example, No. 6 oil has a density of
0.92 g/cm3 (7.7 Ib/gal) and has an HHV of 39.71 MJ/dm3 (142,450 Btu/gal).
40% Eco-Fuel 11/60% oil mixture has a density of 1.07 g/cm3 (8.9 Ib/gal) and
an HHV of 35.74 MJ/dm3 (128,205 Btu/gal). CEA is developing oil burning
equipment with the capability of combusting the combined fuel product.
Descriptions of Key Projects
On the following 4 pages are presented brief descriptions of typical
projects producing waste-derived fuels.
B-14
-------
NAME - City of Chicago Supplementary Fuel Processing Plant
TYPE - Mechanical MSW processing facility for producing refuse-derived fuel
(RDF), which can be co-fired with coal in utility steam generators.
DEVELOPER - City of Chicago and Commonwealth Edison Company
HISTORY - Lacking suitable landfill sites, the City of Chicago has relied on
incinerators for refuse disposal. Escalating incinerator construction and
operating costs indicated the need for a new solution, and a study undertaken
by the City in 1972 found that the St. Louis/Union Electric RDF system would
best meet their needs. Design of the facility started in late 1973; construc-
tion began in 1974 and testing operations were initiated at the end of 1976.
PROCESS - The process is designed to produce RDF having a maximum 3.8 cm (1-1/2
in.) particle size. The MSW input rate is 72 Mg/h (80 TPH) on each of two
processing lines. Refuse processing capacity is 1152 Mg/d (1280 TPD) in one
8 hour shift or 2304 Mg/d (2560 TPD) in two 8 hour shifts. From each 907 Mg
(1000 tons) of raw refuse processed, 631 Mg (696 tons) of RDF with a 30% mois-
ture content can be delivered, and 78 Mg (86 tons) of ferrous metals recovered.
The higher heating value of the RDF is expected to be 13.20 MJ/kg (5674 Btu/lb).
MSW is weighed and dumped onto a tipping floor and then conveyed to a coarse
shredder. The shredded refuse is divided by an air classifier into a light,
largely combustible, fraction and a heavy fraction containing non-combustibles
and overweight combustibles. The heavy fraction is magnetically separated to
recover ferrous metals and the remainder is landfilled. The light fraction,
containing approximately 85% of the organics in the raw refuse, plus some inor-
ganic fines, is conveyed to a fine shredder whose output has a maximum size of
3.8 cm (1-1/2 in.). The processed RDF is carried by a pneumatic transfer sys-
tem to storage bins adjacent to the power plant boilers. For burning, the
RDF is removed from the storage bins by a mechanical system and blown into two
boilers at Commonwealth Edison's Crawford Station, supplying approximately 7.5%
of the main furnaces' heat requirements. The total RDF consumption of both
units running at nameplate rating would be 978 Mg/d (1078 TPD), requiring 1406
Mg (1550 tons) of raw refuse to be processed. Only minor modifications to the
boilers were required, involving penetration of the windbox and furnace wall by
the pneumatic RDF transfer lines and changes to the boiler control systems. No
changes were required in the ash handling or electrostatic precipitator systems.
ECONOMICS - Design and construction costs, which do not include any amount for
land, construction management, working capital or startup, are estimated to
total $18.8 million. At a production rate of 907 Mg/d (1000 TPD) the annual
costs are expected to be $6,38/Mg ($5.79/ton) for amortization plus $6.94/Mg
($6.30/ton) for operations. These expenses will be offset by revenues from the
sale of RDF and ferrous metals estimated to be $5.37/Mg ($4.96/ton) of refuse
processed, resulting in a net annual cost of $7.85/Mg ($7.13/ton) of refuse
disposed.
B-15
-------
NAME! - Solid Waste Recovery System, Ames, Iowa
TYPE - Refuse-derived fuel (RDF) and materials recovery plant
DEVELOPER - City of Ames, Iowa
HISTORY - As an alternative to continued landfilling, the City of Ames decided
TrTl.972 to develop a resource recovery facility. The source of refuse could be
guaranteed by the city and county, and the city-owned power plant could be the
user of the RDF. Construction began in April, 1973, and the plant was opened
in November 1975; it is the first operational RDF plant and is unique in being
a small facility serving a population of only 65,000, Approximately 136 to 180
Mg/d (150 to 200 TPD) of MSW is available from the city.
PROCESS - Refuse is received on a tipping floor and is conveyed to a coarse
shredder in which 45 Mg/h (50 TPH) of MSW are reduced to an average size of
15 to 20 cm (6 to 8 in.). Following the shredder is a three magnet system
recovering 90 to 95% of the ferrous metals. The recovered metal goes through
a second magnetic system to release trapped non-ferrous material before being
sent to a storage bin. The main flow of refuse from the first magnetic system
continues through a second shredder where it is reduced to a maximum size of
3.8 cm (1.5 in.). The finely shredded material is taken across a vibrating
screen to remove sand and into an air classifier. The light fraction is iso-
lated in a cyclone and passed through an airlock to a storage bin near the
power plant. The heavy fraction contains marketable materials,, primarily alu-
minum, and a separation process designed by Combustion Power Co. is in the
shakedown stage. The remaining material is landfilled.
Ames power plant has three boilers, two of which have spreader-stokers that can
take up to 50% refuse by fuel value with the usual coal; the third is a 33 raw
suspension-fired unit also modified to accept RDF.
ECONOMICS - The capital cost of the facility was $5.6 million. Unit costs of
operation and capital amortization cited by Ames (Waste Age, October 1975) are
$15.71/Mg ($14.25/ton) and credits total $14.39/Mg ($13.05/ton). When operating
at less than rated capacity, the facility would not enjoy such a favorable net
cost of disposal.
B-16
-------
NAME - Americology/Milwaukee RDF and Materials Recovery Plant
TYPE - RDF production and recovery of newspaper, cardboard, iron, aluminum, and
glass from MSW.
DEVELOPER - Araericology Division of American Can Co., Greenwich, Conn.
HISTORY - Milwaukee MSW is to be handled by American Can for a 15-year period.
The facility built by Americology for processing the wastes became officially
operational in May of 1977. The city has an option to buy it after the first
5 years of operation and to share in revenues from the sale of recovered
products.
PROCESS - The facility can process 1089 Mg/d (1200 TPD) of MSW. Two processing
lines shred the refuse to less than 38 mm (1-1/2 in.) and concentrate the organic
fraction through use of air classifiers. About 60% of the raw MSW is expected
to be recovered as RDF, which will be purchased by the Wisconsin Electric Power
Co. for supplemental firing in suspension coal boilers. Aluminum, ferrous
metals, and a glassy aggregate for use by the City's Public Works Department
are also isolated.
ECONOMICS - The plant costs $18 million, including modifications at the power
plant. No operating cost has been disclosed. Americology will be paid $ll/Mg
($10/ton) for MSW disposal, with an escalation clause being in the contract
with the city.
B-17
-------
NAME - Hydrasposal/Fibreclaim Solid Waste Recycling Plant, Franklin, Ohio
TYPE - Wet pulping and separation of MSW
DEVELOPER - Black Clawson Co., New York City and Middletown, Ohio
HISTORY - The city of Franklin, Ohio, desired an improved method of MSW disposal
in 1969 and obtained a two-thirds grant from HEW for construction of a 137 Mg/d
(150 TPD) plant to demonstrate a new process developed by Black Clawson. Con-
struction began in September 1970 and the plant was operational in June 1971.
Black Clawson still operates the plant under contract to the city of Franklin.
PROCESS - The process equipment, normally handling only 36 to 45 Mg (40 to 50
tons") per day, is housed in a 1022 m2 (11,000 ft^) building. Refuse is received
on a tipping floor and then conveyed to a 3.6 m (12 ft) diameter Hydrapulper
powered with a 224 kW (300 HP) motor. Recycled water is mixed in and a 3.5%
slurry is formed. Various washers, perforated plates, liquid cyclones, and
screens then isolate material fractions. Long and short fibers are recovered,
along with ferrous metals and glass. Dewatered non-recoverable organics and
sewage sludge are burned in a fluidized bed Dorr-Oliver FluoSolids reactor. Heat
recovery is not attempted because of the relatively small size of the plant.
ECONOMICS - The EPA 2nd Report to Congress lists a cost of $8.3 million for a
454 Mg/d (500 TPD) plant of this type and calculates an annual operating cost
of $1.5 million. Net unit costs would be $9.75/Mg ($8.83/ton).
B-18
-------
BIOLOGICAL CONVERSION PROCESSES
Biological processes transform organic materials by the action of living
organisms. In the case of interest here, the transformation is one of organic
wastes into a useful form of fuel. The principal processes for doing this
include anaerobic digestion of sewage, municipal solid waste, or manure to
produce methane; and hydrolysis of paper, wood or crop waste to sugar with
subsequent fermentation to ethyl alcohol (ethanol).
Anaerobic Digestion
Anaerobic digestion is a natural decay process that commonly occurs at
the bottom of marshes where there is no oxygen. Marsh gas, containing com-
bustible methane, results from the decomposition of vegetable matter and
bubbles to the surface. This can also occur during sewage treatment, and
tank digesters were early developed to catch the gas for use in street
lighting (Ref. B-36). Because the process produced a fuel gas before natural
gas was available, and because it also stabilized sewage solids and reduced
health hazards, the use of anaerobic digestion grew through the 1920's. Many
sewage treatment plants were self-sufficient in energy and some had an excess
of gas for sale (Refs. B-36, B-37, B-38). However, anaerobic digestion
developed a reputation for unreliability because it can be upset by unskilled
operation, equipment inadequacies, or toxic materials in the waste feed, and
other processes for the disposal of sewage sludge were sought (Refs. B-36,
B-38).
Interest continued in India, where methane production systems were
developed for village homes based on the digestion of cow dung. These systems
made fuel for light and cooking and provided a relatively sterile fertilizer
in the form of digested sludge. Similar digesters are in common use on the
many small pig farms on Taiwan (Ref. B-39).
Generation of methane by anaerobic digestion in the United States is
being given renewed interest. Estimates have been made that 10% to 11% of
the 1970 demand for natural gas could be met by methane derived from sewage
sludge and municipal solid waste; in addition, if all animal and crop wastes
were digested, the above figures would be doubled (Refs. B-40, B-41).
Anaerobic digestion is a wet process performed in stages by two groups
of bacteria. The first group, the acid formers, liquefy the organic wast.e
solids and convert complex organic substances such as carbohydrates and
proteins to simple organic acids. The second group of bacteria, the methane
producers, consume these acids and release carbon dioxide, methane, and
traces of other gases (Refs. B-40, B-41). The chemistry of the process may be
represented by:
Methane
Acid forming producing
C6H10o5 + H20 baCterla > 3C2H,02 JZ££H£i^ 3C02 * 3CH,
Carbohydrates + Water ^Acetic Acid ^Carbon dioxide + Methane
B-19
-------
The mix of gases produced will vary with the type of wastes fed to the
process. The following dry gas compositions have been reported:
Sewage sludge Animal wastes Municipal solid waste
(Ref. B-37) (Ref. B-42) (Ref. B-43)
Methane
Carbon Dioxide (C02)
Nitrogen (N2)
Hydrogen (H2)
Carbon Monoxide (CO)
Oxygen (02)
65 - 70%
30 - 35%
Trace
Trace
Not Mentioned
Not Mentioned
Hydrogen Sulfide (H2S) Trace - 0.2%
54 - 70%
27 - 45%
0.6 - 3%
1 - 10%
0.1%
0.1%
Trace
50%
50%
Not Mentioned
Not Mentioned
Not Mentioned
Not Mentioned
Trace
Some of these bio-gases have significantly more methane than carbon
dioxide, although the chemical reaction predicts that they should be equal.
This occurs because carbon dioxide is much more soluble in water than is
methane, and a portion of it remains in the water required in the process.
All of these gas mixes will burn without purification, and are used
directly by the sewage treatment plants and their industrial customers.
Heating value of these gases is in the range of 20 to 26 MJ/m3 (540 to
700 Btu/ft3), provided principally by the methane which has a heating value
of about 37 MJ/m3 (1000 Btu/ft3). Natural gas is about 98% methane and
methane purified from a biogas mixture can be used directly in natural gas
systems (Ref. B-42, B-44).
Virtually all organic wastes, whether municipal, industrial or
agricultural, can be anaerobically digested. Some materials are easier to
digest than others; gas yields from cow manure digestion, for example, ranged
between 0.09 and 0.20 m3/kg (1.4 to 3.2 ft3/lb) of volatile solids, compared
with 0.38 to 0.57 m3/kg (6.1 to 9.1 ft3/lb) of volatile solids in sewage
sludge digestion. This has been attributed to the difficulty of breaking
down the cellulose in the cow manure (Ref. B-45). The absence of toxic
materials, such as sulfides, heavy metals, and toxic organic compounds is
important, as is the maintenance of an oxygen-free environment.
Anaerobic digestion can take place at two optimum temperature levels,
the mesophilic level, 30°C to 37.5°C (86°F to 99.5°F), and at the thermo-
philic level, 49°C to 51°C (120°F to 124°F). The rates of reaction at the
thermophilic level are faster than those at the mesophilic level, but the
heating necessary to maintain thermophilic temperature make such systems less
economically attractive (Ref. B-41). The correct pH level, optimally between
7.0 and 7.2, is critical, as is the correct carbon: nitrogen: phosphorus
ratio (Refs. B-41, B-44).
B-20
-------
Because anaerobic digestion is a biological process, sufficient time must
be allowed for the bacteria to function. In continuous sewage sludge
digesters, organic solids retention times range between 15 and 20 days, by
which time 90% digestion has been reached. With longer retention times the
digestion rate becomes slower, and over 40 days retention time would be
necessary to attain 95% digestion of the potentially biodegradable material
(Ref. B-46).
Methane-containing biogas is created naturally in many landfills, and the
release of foul smelling gas can be a major source of complaints. Several
organizations have seen this gas evolution as an energy recovery opportunity
and are now extracting and purifying the gas. A description of one of these
methane recovery programs is given at the end of this section.
Anaerobic digestion of sewage sludge is one of the principal means for
sewage treatment. Methane recovery, eclipsed for a while by the availability
and low cost of natural gas, is now receiving new attention. About 0.028m3
(one cubic foot) of biogas is generated every day for each person in the area
served, so it has been recommended that large cities can best practice methane
recovery (Ref. B-47). The Metropolitan Sanitary District of Greater Chicago
has seven treatment plants, five of which has anaerobic digestion systems.
Biogas is used for heating the raw sludge feed, maintaining digester tempera-
ture, and heating the buildings. An excess of 5% to 30% is available for
outside use, and various alternatives are being explored. Energy production,
after satisfying all digester heat requirements and converting excess gas
produced to electrical energy, has been calculated to be twice the electrical
energy required to operate the system (Ref. B-48).
The Santa Clara/San Jose Sewage Plant in California uses biogas to run
large internal combustion engines, and in Orange County, California, gas
turbine generators are powered by biogas (Refs. B-49, B-50). The City of
Los Angeles operates the Hyperion Treatment Plant, a major facility serving
a population of 3 million people. Biogas is used for generating electricity
both in-house and at a neighboring generating plant (Ref. B-37).
Anaerobic digestion has also been proposed for municipal solid waste.
Work done by J. T. Pfeffer for the National Science Foundation (Ref. 43) and
U.S. EPA (Ref. B-41) led NSF to fund the Dynatech Corporation to develop a
preliminary design and economic model for a 1000 TPD plant capable of producing
approximately 104 670 m3/d (317 x 106 ft3/day) (Ref. B-51). NSF also funded
a Mitre Corporation study to provide background information for an "Urban
Trash Methanation Proof-of-Concept Experiment," which reviewed the state of
the art in feed preparation, digestion, sludge dewatering and disposal, and
gas processing (Ref. B-52). As a result of this work, an ERDA-funded facility
is being built at Pompano Beach, Florida, by Waste Management Inc., and
Jacobs Engineering Co. Scheduled to be operational late in 1977, the plant
will have a daily capacity of 91 Mg (100 ton) of refuse (Ref. B-53).
B-21
-------
A similar system has been proposed by D. L. Klass and S. Ghosh of the
Institute of Gas Technology (Ref. B-50, B-54). In both concepts, the organic
fraction, typically about 70% - 75% by wet weight of the total, is isolated
from the refuse. The shredded material, principally paper, is mixed with
sewage sludge and fed into an anaerobic digester. Biogas removed from the
digester is processed to pipeline methane, using conventional gas cleaning
technology, Sludge from the digester is treated, using conventional waste
water techniques (Refs. B-54, B-55). The principal questions to be answered
at the Pompano Beach facility are concerned with the amount of feed processing
necessary and the operating conditions for the digester (Ref. B-52). In all
such plants, an important concern will be the ultimate disposal of the sludge
remaining after the digestion.
According to a 1975 cost analysis made by DynateCh, it should be possible
to produce methane from solid waste and sewage sludge to sell at a price of
$0.074/m3 ($2.09 per thousand cubic feet) (Refs. B-43, B-51). This price
includes penalties for disposal of process wastes and credits for disposal
of wastes input to the process ($11.74/Mg or $10.65/ton). The process has
been criticized as being inefficient because only 35% of the energy content
of the solid waste is recovered as methane and when a penalty is taken for
the energy needed to run the process, an overall energy efficiency of 25%
results (Ref. B-55). The process still produces 2.67 times more energy than
is required to run it, however. In balance, the process is marginally
expensive and requires further development, but could be attractive where waste
disposal charges are high and methane is in short supply. Additional work on
generation of methane from garbage and sewage is being conducted at the
University of Arizona (Ref. B-36).
Digestion of animal wastes to obtain methane is receiving increased
attention in the U.S. B. A. McDonald has designed and operated a prototype
dairy cow manure digestion system for over a year (Ref. B-56) and the Ecotope
Group has prepared a technical and economic feasibility study of anaerobic
digestion for the Washington State Honor Farm (Ref. B-42). Most of these
developers have been concerned with systems for small farms of under 400 cows,
although Fry and Taiganides (Ref. B-44, B-57) have performed work with farms
with several thousand hogs. Increased confinement of cattle in feedlots has
sharply increased the availability of manure in recent years, while at the
same time the use of manure for fertilizer has virtually ceased. New EPA
regulations on pollution from feedlot operations and the shortage of natural
gas have combined to make the production of methane by anaerobic digestion
attractive (Ref. B-37). Calorific Recovery Anaerobic Process Inc. and ERA Inc.
have signed contracts with a pipeline subsidiary of Peoples Gas Co. of
Chicago for 18.4 x 106m3 and 18.1 x 106m3 (650 and 640 million cubic feet) of
methane respectively (Refs. B-38, B-58, B-59). The Federal Power Commision
has approved a price of $0.047/m3 ($1.33 per thousand cubic feet) that the
pipeline company will pay Calorific (Ref. B-60). Bio-Gas of Colorado, has
issued a prospectus for a 5000 cow unit anaerobic fermentation facility
(Ref. B-61), and Jacobs Engineeering Co. states that they are helping a
feedlot operator develop a total manure recycling system that produces methane,
fishmeal, and cattle feed. Research has been started or expanded at a number
of universities, including Cornell (Ref. B-62), Missouri (Ref. B-63), and
Wisconsin (Refs. B-64, B-38).
B-22
-------
Anaerobic digestion of manures is essentially the same as that for
sewage treatment, the most important difference being that cattle manures are
more difficult to digest and produce less gas than sewage. This is because
about 10% of a feedlot steer's diet, or 50% of a dairy cow's diet, is roughage
that is hard-to-digest cellulose and because manure has already essentially
been through an anaerobic process. Yields from MSW are similarly low because
of the cellulose fraction.
The volume of gas produced per pound of volatile solids (VS) added and
the percentage methane in the gas varies according to the material being
digested. Some representative figures from laboratory experiments are given
below (Ref. B-44).
Proportion Gas Produced % Methane
Steer Manure 100% 0.087 m3/kg VS 1.4 ft3/lb VS 65.2
Steer Manure § 50%
Chicken Manure 50% 0.212 3.4 61.9
Chicken Manure 100% 0.312 5.0 59.8
Newspaper $ 10% 0.618 9.9 67.1
Sewage Sludge 90%
While anaerobic digestion is capable of significantly reducing the volume
and biological activity of organic wastes, disposal of large quantities of
supernatant digester liquid and the final sludge remains a major problem. In
sewage treatment, the supernatant is recycled through the waste water treatment,
and the sludge has been either ocean dumped or dewatered and landfilled. More
extensive recycling programs have been proposed for manure digester residuals.
The supernatant is an ideal feed for an algae pond in which small food fish can
be grown for use as cat food or a cattle feed supplement. The digested sludge
is an excellent fertilizer that contains all of the nutrient value of manure
but is odor free and not attractive to flies and other pests; it may also be
possible to make a cattle feed supplement from the dewatered sludge
(Ref. B-61, B-62).
Fermentation
Alcohol (ethanol) produced by fermenting sugars with yeast can be used as
a fuel. Wastes suitable for making ethanol contain cellulose, a principal
material in plants, which must first be converted to sugar by hydrolysis.
Following fermentation of the sugar, the resulting alcohol must be concentrated
by distillation. The process of producing ethanol from waste therefore,
requires three independent maj or steps:
Hydrolyze Ferment Distill
Cellulose or Sugar Dilute Concentrated
Carbohydrates "~ ^Ethanol^Ethanol
B-23
-------
All of these are ancient processes whose most familiar combined use is
the production of beverage liquor. About 1830, the development of continuous
process column still made it possible to produce large quantities of ethanol,
and it began to replace whale oil as a fuel in lamps (Ref. B-65). Slightly
earlier, a major advance in hydrolysis was made when it was found that wood
and other cellulosic materials heated with a strong acid would be converted
to sugar. Traditionally, however, industrial fermentations used molasses
and sugar processing residues which were readily available and did not require
hydrolysis (Ref. B-66).
In the early days of the automobile, alcohols were given serious con-
sideration as motor fuels before being displaced by gasoline. In some
European countries, a shortage of gasoline was relieved by blending it with
as much as 30% alcohol (Ref. B-65). During World War II, acid hydrolysis,
fermentation, and distillation to produce fuel were investigated in Germany
and by the U.S. Forest Products Laboratory (Ref. B-67). About this time an
alcohol plant capable of producing 53 m3/d (14,000 gallons per day) of 95%
ethanol from lignosulfonate chemicals left over from paper making was built
at Georgia-Pacific's Bellingham Division Plant (Ref. B-68). Following the war,
the production of ethanol from ethylene in petrochemical plants became more
economical than the hydrolysis/fermentation process. The Georgia-Pacific
Plant is believed to be the only one of its type remaining in the United States,
excluding alcoholic beverage producers, who cannot by law use synthetic
alcohol in their products.
Wood is composed of about 25% lignin and 75% cellulose chains. Hydrolysis,
either with acid or enzymes, breaks apart the lignin and cellulose chains, and
then further reduces the cellulose to various sugars. Chemically, the reaction
can be generally represented as:
(C6H1005)n + nH20 — ^ nC6H1206
Cellulose + Water = Glucose
Only about 70% of these sugars are readily fermentable, however, so that only
about 52% of the wood can be converted to alcohol - a major limitation on the
process (Ref. B-69). The sulfite pulping process used by Georgia-Pacific's
Bellingham Division is similar to sulfuric acid hydrolysis, except that it is
limited to separating the cellulose chains and sugar production is minimized.
The cellulose becomes paper and the fermentable sugars, amounting to a small
fraction of those potentially available from the cellulose, are made into
alcohol. Yield is about 0.027 m3 of ethanol per Mg of wood (6.5 gallons per
ton) (Ref. B-68).
Since the mid-19601s, a new hydrolysis process based on enzymatic action
has attracted attention. The principal developers have been the U.S. Army
Natick Laboratories, whose original goal was to prevent biological decay of
textiles. They found (Refs. B-70, B-71) that the fungus Trichoderma viride
was especially capable of producing enzymes for breaking cellulose down into
glucose sugar, and have developed pre-pilot plant facilities handling 454 kg
(1,000 pounds) of cellulose per month to investigate the operating parameters.
B-24
-------
Acid hydrolysis is a well-developed technology while enzyme hydrolysis
is still in the research stage. Contact time for the enzyme hydrolysis is
50 hours at 50°C (122°F) compared to 5 hours at 150°C (302°F) for acid
hydrolysis, and the feed material for enzyme hydrolysis must be finely ground
to achieve this rate (Ref. B-66). However, there exists a potential for
higher yields of fermentable sugars from enzyme hydrolysis than from the acid
process, and further development work is continuing (Ref. B-46).
Conversion of the fermentable sugars to ethanol follows the classic
process using Saccharomyces cerevisiae yeast. The chemical reaction is:
C6H1206 Yeast • 2C2H5OH + 2C02
Glucose = Ethanol + Carbon Dioxide.
At the Georgia Pacific Plant, seven interconnected 379 m3 (100,000 gallon)
fermentors are used, with a residence time of 15 to 20 hours (Ref. B-73).
About 95% of the available glucose can be converted. While the yeast can
tolerate up to 15-16% alcohol by volume before being destroyed, the Georgia
Pacific process consumes dilute sugar solutions and does not reach high alcohol
concentrations; the yeast is recovered by centrifuging.
Fermentation can be applied to other wastes that contain or can be con-
verted to sugar. Black Clawson Fiberclaim Inc. is investigating the use of
its Hydrapulper to prepare the cellulose fraction of municipal solid waste for
conversion to ethanol (Ref. B-73). Kraftco produces alcohol and vinegar by
fermenting lactose from cheese whey. Fermentation has not become more popular
because (1) there are other means for recycling these wastes at least as
economically attractive, and (2) high degree of government regulation of
alcohol production facilities (Ref. B-65).
No industrial hydrolysis/fermentation alcohol plants have been built
since the second war, so available cost data are old. A number of re-estimates
have been made, producing total cost figures of $0.08 to $0.25/dm3 (32
-------
NAME - Sanitary Landfill Methane Recovery Program, Los Angeles
TYPE - Methane recovery from landfill. At the Palos Verdes Landfill approxi-
mately 28 300 m3 (1 million cf) of purified methane are recovered daily.
DEVELOPER - NRG NuFuel Company, a joint venture of NRG Incorporated, Phoenix,
and Reserve Oil and Gas Co., Los Angeles; now called Reserve Synthetic Fuels.
HISTORY - The main 696 084 m2 (172 acre) Palos Verdes Sanitary Landfill, owned
and operated by the Los Angeles County Sanitation Districts, was started in
1963. Decomposition of refuse in the landfill created biogas that migrated
to the surface, causing an unpleasant odor. Sanitation Districts' engineers
installed wells and burners to collect and flare the gas. In October 1973,
the Sanitation Districts signed an agreement with NRG for the recovery and
sale of methane from the biogas. The present processing facility was started
in February 1975 and became operational with five wells in June 1975.
PROCESS - The decomposition of refuse in landfills produces biogas, a mixture
of roughly equal parts of methane and carbon dioxide. The biogas production
potential of a landfill depends on many factors. To qualify for the Reserve
program, the landfill must contain at least 1.8 Tg (2 million tons) of refuse
in place, a minimum depth of 12.2 m (40 ft), a high percentage of decompo-
sable organic material with adequate moisture content, a suitable type of
cover, and close proximity to energy markets.
To tap the biogas, wells are drilled into the landfill. Biogas is withdrawn
from the landfill. Biogas is withdrawn from the landfill under vacuum through
an underground collection system and pretreated to remove moisture, hydrogen
sulfide, and other trace contaminants. The gas is then passed through mole-
cular sieves, which selectively remove the carbon dioxide, leaving clean, dry
pipeline quality methane (natural gas). The methane is compressed to specific
pipeline requirements for delivery to the user. In the case of the Palos
Verdes Landfill, enough recovered methane to meet the daily energy needs of
3500 homes goes directly into the distribution system of the Southern Cali-
fornia Gas Company.
ECONOMICS - Cost and revenue data are the proprietary information of the com-
pany. The company's descriptive brocure states that royalty revenues of
12-1/2% of gross sales could be paid to the landfill owner.
(METHANE AND CAfll
DIOXIDE MIX)
B-26
-------
NAME - RefCOM (Refuse Conversion to Methane), Pompano Beach, Florida
TYPE. - Experimental anaerobic digestion facility for studying controlling
parameters for converting MSW to methane.
DEVELOPER - Dr. John T. Pfeffer, University of Illinois, Consultant; Jacobs
Engineering, engineering/construction; Waste Management, Inc., prime contractor.
HISTORY - As discussed within Appendix B, anaerobic digestion of sewage to
methane has been practiced for many years, but conversion of refuse in a
similar manner has only been studied on a laboratory scale. To establish the
factors affecting the ultimate commercial feasibility of such a process, ERDA
has contracted to Waste Management, Inc., for the design and operation of an
experimental facility. Construction began in February 1977 and is scheduled to
be completed by the end of 1977.
PROCESS - Approximately 85m3 (3000 ft3) of methane is expected to be produced
from each 0.9 Mg (1 ton) of processed MSW; the facility is designed for a
capacity of 90 Mg/d (100 TPD). The organic fraction of MSW will be enriched
and then mixed with primary sewage sludge before anaerobic digestion is initiated
in large tanks. Design features for variable adjustments have been incorporated
into the plans.
ECONOMICS - The facility has a cost of $2.8 million, but the plant being of an
R§D nature, this value cannot be used to establish eventual commercial costs.
B-27
-------
THERMAL/CHEMICAL CONVERSION PROCESSES
Overview
The energy content of wastes is most typically utilized by the exothermic
direct and complete oxidation to the elements and oxides (combustion, as
described in the following Section). An alternative method involves the prep-
aration of new fuels through conversion of the rather large number of high
molecular species in the waste to a relatively few low molecular weight com-
pounds plus oftentimes a solid residue of high carbon content. The new
materials require energy for their formation and a kg of total products will
possess a heating value less than the same mass of starting materials. While
isolatable fractions can be obtained having a higher heats of combustion per
unit weight than the original average reactant mixture, this is achieved at
the expense of other fractions having low, or zero, heating values. The loss
in energy is tolerated because of (1) the convenience of the physical or
chemical form of the new fuel or (2) the reduced volume of total gases over
that formed in a typical incineration process, and hence lower capital invest-
ment and air pollution control costs.
The conversions can be accomplished through chemical or thermal processes.
The methods can yield essentially any fuel form desired, or can be used to
synthesize compounds such as ammonia, which otherwise would have consumed fossil
fuels in their normal commercial production. This project is limited to the
following:
• The gaseous, liquid, or solid mixtures resulting from thermal decom-
position schemes with no oxygen present or only sufficient to create
heat to drive the reactions (pyrolysis).
• Reactions designed to give high yields of synthesis gas or "syngas"
(hydrogen-carbon monoxide mixtures).
• Use of carbon monoxide, hydrogen, or water as direct reactants with
waste materials.
• Chemical conversion of thermal decomposition products to methanol,
methane, or ammonia.
The publications in the technology area range from those dealing with
theoretical considerations such as the mechanisms and kinetics of decomposi-
tion of cellulose, through detailed compositional analysis of the products
from laboratory scale reactors, to descriptions of start-up experiences of
plants processing as much as 900 Mg (1000 tons) of waste per day. Investi-
gators include scientists and engineers from government, university, and
industrial organizations. Advances in the field are undoubtedly occurring
more rapidly than in any other waste-to-energy R£D technology. With the
potentiality for processing all forms of organic wastes (municipal, industrial,
sewage, animal, and agricultural) to numerous fuel types over a broad range
of input capacities, several of these systems will undoubtedly enjoy a
moderate degree of commercialization in the next five years.
B-28
-------
The reaction conditions of temperature, pressure, heating time, and
relative reactant quantities can be varied to optimize the yield of a particu-
lar type of fuel or synthesis gas. The actual chemistry involved is compli-
cated because of the large number of simultaneous reactions that may occur,
with each being influenced by the extent of equilibrium of the other.
Measurement of the composition and properties of the final mixture is simple,
however, and research tends to be highly empirical. The principal reactions
involved include the following:
Drying Zone
Solid Waste + Heat -» Dry Waste + H20 (1)
Pyro lysis Zone
Dry Waste + Heat — C + C02 + H20 + Hydrocarbons + H2 + CO (2)
Oxidation and Reduction Zones
C + 02 $ C02 + Heat (3)
C + C02 + Heat J 2CO (4)
C + H20 + Heat £ CO + H2 (5)
C + 2H2 £ CH4 + Heat (6)
2C + 02 £ 2CO + Heat (7)
CO + 3H2 £ CH4 + H20 + Heat (8)
C02 + H20 £ C02 + H2 + Heal; (9)
H2 + 1/202 t H20 + Heat (10)
C02 + 4H2 J CHi, + 2H20 + Heat (11)
2CO + 2H2 J Ofy + C02 + Heat (12)
CO + H2 + Organics ^ Misc. Liquid, Solid and
Gaseous Organics
Synthesis Routes from Syngas^
CO + 3H2 •* CHij + H20 + Heat
CO + 2H2 ->• CH3OH + Heat
N2 + 3H2 •* 2NH3 + Heat
B-29
-------
Pyrolytic Gasifiers
With Sections 6, 7, and 8 are presented details of three pyrolytic con-
version systems (Georgia Tech, Torrax, and Purox) where the temperature and
duration of exposure of the organic waste material have been adjusted to pro-
duce significant yields of gases. That the end energy forms from each is
quite different in spite of the basic same shaft reactor being used in all
cases is indicative of the versatility of this type of processing.
The mobile agricultural pyrolysis system being developed by the Engi-
neering Experiment Station (EES) of the Georgia Institute of Technology
(Ref. B-80) must be independent of outside power. The low heating value gas
produced is therefore totally consumed as a fuel for an internal combustion
engine driving an electrical generator and for a combustor whose hot oxidized
gases are used for drying incoming wastes. The liquid and solid products of
pyrolysis are mixed for sale as a consumer fuel.
The Torrax gasifier (Ref. B-81J produces its necessary heat by air
oxidation of char as does the EES system. The combustion air in this case
has been preheated in an exchange system and temperatures in the oxidation
zone are sufficiently high to melt the inorganic waste fraction to a slag.
The hot combustion gases then pass over essentially dry wastes and decompose
them to a gas mixture that leaves the reactor at approximately 427°C (800°F)
and still contains organic particulate matter. Rather than lose the sensible
heat and expend energy in cleaning up the stream, the reactor off-gas is
immediately combusted and the heat used to generate steam.
Because there are broad applications for a fuel and synthesis gas, the
Union Carbide Corporation has developed its Purox pyrolysis system to yield
a medium-level heating value gas capable of being pipeline transported
(Ref. B-82). This is accomplished by using gaseous oxygen in the char com-
bustion zone, and thus the pyrolytic gases are not diluted with nitrogen.
The gas, having a heating value of approximately 14.57 MJ/Nm3 (370 Btu/SCF)
can serve directly as a utility fuel. It is also capable of serving as the
feedstock for synthesis of a wide range of compounds, including methane,
methanol, and ammonia. Such conversions are discussed in Appendix C.
While the three candidate systems selected for engineering analysis are
among the most advanced pyrolytic gasifiers now available, a number of other
investigators have been active in this field. The largest waste gasifier yet
constructed is the Monsanto Landgard system in Baltimore (Ref. B-83). It was
designed to process 907 Mg/d (1,000 TPD) of MSW by partial air oxidation in a
rotary kiln to yield a gas having an HHV of 3.88 MJ/Nm3 (98.5 Btu/SCF). The
gas was then to be immediately combusted to produce steam in a waste heat
boiler, which is then used for cooling and heating purposes. The plant has
experienced particulate emissions in excess of state regulations and mechani-
cal problems in the kiln; it is presently undergoing modification.
Battelle Northwest (Ref. B-84) has developed a gasifier patterned after
fixed bed coal systems that use a vertical shaft with air combusting the char
in the presence of steam. Feed of approximately 1 Mg/d was introduced through
a rotary air lock. Temperatures at the grate were 810 to 1,080°C (1,490
to 1,976°F), and the HHV of the off-gas was about 6.06 MJ/Nm3 (154 Btu/SCF).
B-30
-------
Battelle has also investigated (Ref. B-85) the use of molten sodium carbonate
as the heating media for the gasification of MSW. At 900°C (1,652°F) a gas
was obtained having an HHV of 19.69 MJ/Nm3 (500 Btu/SCF).
Barber-Colman (Ref. B-86, B-87) developed, in a pilot unit, a unique
pyrolysis system that involved radiant heating with hot overhead tubes in a
horizontal retort. No commercial equipment was ever built. In the concept
the refuse is fed by screw-conveyor into the sealed unit onto a lead bath
kept molten by the heat from the radiating tubes. The liquid lead, flowing
in a channel along the length of the retort, carries the pyrolyzing organics.
At the exit end, char and ash are scraped off the lead, which in turn is
pumped back to the pyrolysis retort. The rapid heat transfer to the refuse
from the lead results in a range of hydrocarbons. The composition of the
product gases were reported (by volume) as 16.4% H2, 29.4% CO, 18.4% C02, 23.33
CHit, 3.4% C2tt^, 0.5% C2H6, 7.9% C$H&, 0.6% C7H8; the HHV was 30.32 MJ/Nm3
(770 Btu/SCF). The cooled gas, with the C6 + condensed out, has a composition
of 18% H2, 32% CO, 20% C02 , 25.5% CH4, 0.5% C2H6, with an HHV of 19.30 MJ/Nm^
(490 Btu/SCF).
For the past several years, Pyrotek (Ref. B-88) has been developing a
fuel gas producer using a pyrolysis unit fed by shredded refuse. The unit
under test is at the bench level and essentially hand fed. The community of
Riverside, California, is purportedly allowing Pyrotek to build a small
demonstration unit in their city (with Pyrotek funds) to test its feasibility.
The laboratory system is a horizontal reactor fed by hand with the shredded
refuse conveyed internally by a vibrating conveyor and heated by hot radiant
tubes above the moving bed. These tubes have been heated electrically, but
they expect that approximately one-third of the pyrolysis gas will be used
in a commercial unit for heating. The gases are scrubbed to remove acid
gases, oil, and solid particulate matter. The volumetric composition of the
gas is 19% H2, 18% N2, 26% CO, 13% CH^, 18% C02, and 4% C^, with an HHV of
13.78 MJ/Nm3 (350 Btu/SCF).
In other reported work, the Urban Research and Development Corporation
(Refs. B-87, B-89) investigated a vertical shaft slagging furnace, using pre-
heated combustion air and yielding a gas having an HHV of 6.26 MJ/Nm3
(159 Btu/SCF); West Virginia University (Ref. B-90, B-91, B-92) has studied
the concept of a fluidized bed gasifier operating at 810°C (1,490°F) and
producing a gas having an HHV of 16.42 MJ/Nm3 (417 Btu/SCF); Texas Tech
University, in a series of studies on energy forms, including ammonia, from
cattle manure (Ref. B-93, B-94, B-95, B-96), has examined experimental condi-
tions best suited for energy recovery from this abundant waste; and the
U. S. Bureau of Mines (Refs. B-97, B-98, B-99), in the gasification phase of
their wide-ranging studies on waste conversion processes, has obtained gases
having an HHV as high as 21.46 MJ/Nm3 (545 Btu/SCF).
Pyrolytic Liquefaction
In the pyrolysis process, thermal energy is used to break the chemical
bonds of large molecular species, yielding fragments and additional products
of reaction. As the temperature to which the original (and typically solid)
material is exposed increases, the rate of decomposition increases, as does
the portion of the total products that are gases. At lower temperatures,
B-31
-------
fragmentation occurs to a molecular size such that, after cooling, liquid
products predominate. Rapid quenching from the pyrolytic temperature to a
temperature where the liquid product is thermally stable is essential to
recovery of significant yields of potentially useful liquid fuels. Because of
the more exacting nature of process development for liquids, little totally
pyrolytic R§D towards such products has been reported. The only process to
yet reach demonstration scale is the Occidental Flash Pyrolysis system, now
undergoing start-up tests in a 181 Mg/d (200 TPD) plant; it is described in
detail in Section 9 and has been optimized to yield an oil. The Georgia Tech
system happens to operate under temperature-time conditions to produce a
significant fraction of a liquid product; it is described in Section 6. Liquid
fuels have significant advantages and further research should be conducted for
developing economical new pyrolytic liquefaction systems.
Work now in progress (Ref. B-100) at the Naval Weapons Center, China Lake,
California, indirectly produces a liquid fuel after an initial pyrolysis step.
Finely divided cellulosic waste is carried by a gas through a hot zone at
760°C (1400°F) and the products rapidly quenched. Approximately 35 to 40 per-
cent of the solid waste energy is converted to "gasoline precursors," various
olefins that could be used as a feedstock for making so-called polymer gaso-
line. Tests have been made by the Navy on straight ethylene as a feed to a
thermal polymerization reactor and 60 percent conversion to liquids have
been obtained per pass. Unleaded samples of the product has an octane rating
of 90. Purification of the crude pyrolysis off-gas mixture and conversion of
the lower olefins to liquids in the boiling point range of gasoline has not
yet been attempted.
Chemical Reagent Systems
The pyrolytic process involves a complex decomposition mechanism and
interaction of various compounds to form new products; it could be generally
characterized as a thermal process where no external reagents are deliberately
added. Such additions can permit an even wider range of products to be formed
and these can be "tailored" to particular forms by the greater degree of con-
trol offered by the techniques of modern chemical synthesis.
The case where a primary product is first obtained by pyrolysis and then
chemically modified is discussed in Appendix C, where syngas from the Purox
system is used as an example of a starting material. Carbon monoxide and
hydrogen are the fundamental reagents for industrial synthesis and gaseous,
liquid, and even solid fuels could be prepared if so desired. The gas mixtures
from pyrolysis are not ideally suited for this chemistry and experiments must
be conducted before accurate economics for the cost of waste-derived fuels can
be established.
Appell et al (Refs. B-101, B-102) have reported on portions of the R§D on
chemical conversion of different cellulosic wastes undertaken by the Bureau
of Mines of the U.S. Department of the Interior. Sawdust, bark, corncobs,
bovine manure, sewage sludge, and MSW have been made into heavy oils through
high pressure and temperature reaction of syngas or carbon monoxide and water.
Conversion of more than 90 percent of the inorganic-free wastes were obtained,
with 40 to 60 percent of the products being oil. The temperature range
B-32
-------
examined was 250° to 425°C (482° to 707°F) and pressure were between 10.34
and 20.68 MPa (1500 to 3000 psi) . The effects of various catalysts and opera-
tion in the presence of high boiling oils on yields and product characteristics
have been investigated. The boiling point range of the products obtained have
been on the order of 175° to 500°C (347° to 932°F) and the viscosities vary
from readily pourable liquids to solids. Typical analyses and heating values
are as follows:
Waste Feed
MSW
Hardwood
Manure
Reaction Temp.
°C °F
Analysis, Wt-%
HHV
360
300
425
680
572
797
H
9.9
7.0
10.2
75.1
72.5
83.4
N
1.4
0.1
4.6
0.1
0.1
0.1
0 MJ/kg Btu/lb
13.5 36.19 15,560
20.2 33.45 14,380
1.7 39.98 17,190
Friedman et al (Ref B-103), in a very preliminary economic analysis of
the BuMines work, estimates that a city having a population of 300,000 or a
cattle feedlot having 200,000 cattle could just break even on the operation of
a conversion plant. This is based on an oil yield of 2 barrels/ton of dry
organic wastes and a 20 year amortization of the investment.
Feldman, also an investigator for the Bureau of Mines, has patented
(Ref. B-104) and reported on (Ref. B-105) the conversion of wastes to pipeline
quality gas. The work is based on analogous studies on the hydrogasification
of coal and lignite, and involves the reaction of cellulosic materials with
hydrogen at temperatures of 500° to 670°C (932° to 1238°F) and pressures of
6.89 to 20.68 MPa (1000 to 3000 psi). Only limited feasibility test have been
conducted, of which these at 650°C and 1 hour are typical:
Synthetic MSW, g
H2 charge, moles
Conversion of carbon to gas, %
Conversion of carbon to hydrocarbons,
Conversion of carbon to CO, %
Conversion of carbon to C02, %
Gas analysis, vol-%, H20 and C02-free
H2
N2
CO
CH4
C2H6
Test 1
10
0.87
60.0
43.4
8.0
8.8
80.9
2.0
3.0
12.1
2.0
Test 2_
20
0.87
66.4
49.7
5.0
11.5
68.8
1.5
3.1
23.0
3.6
Test 5
80
0.46
45.8
23.3
8.0
14.6
25.3
0.8
1.0
71.1
1.0
Descriptions of Key Projects
Five thermal/chemical projects are briefly described on the following
pages.
B-33
-------
NAME - "Landgard" Pyrolysis System
TYPE - Medium temperature, non-slagging, rotary kiln pyrolysis system using
air and supplemental fuel, and recovering steam, glass, ferrous metals, and
ash.
DEVELOPER - Monsanto Enviro-Chem Systems, Inc., St. Louis, MO
HISTORY - Monsanto began investigating waste resource recovery in 1967, and
selecting the rotary kiln pyrolysis system. A 31.7 Mg/d (35 TPD) pilot
plant was operated from June, 1969 to late 1971. In September, 1972, the
City of Baltimore was given an EPA demonstration grant to help fund a 907
Mg/d (1000 TPD) "Landgard" system. Construction was from January, 1973, to
February, 1975. While the pilot plant met the Maryland air pollution limits
of 0.07 g/Nm3 (0.03 gr/SCF), the full-size plant so far has not. Mechanical
problems (primarily with the refractory insulation) have also been experienced.
Monsanto has withdrawn from the project and the city is making efforts to
correct the problems.
PROCESS - Refuse is dumped into a pit, conveyed to two shredders, and then
stored in a 1814 Mg (2,000 ton) capacity bin. Shredded refuse is fed into
the reactor, which is 5.8 m (19 ft) in diameter, 30.5 m (100 ft) long, and
rotates at 2 rpm. During its journey through the kiln the refuse is pyrol-
yzed to a gas and char. The heat is provided by partial burning of the gas
and char and by supplemental oil. Residue temperature is kept below 1093°C
(2000°F) to avoid slagging; the off-gas is controlled to 6.49°C (1200°F) The
off-gases move counter-current to the refuse, leaving the kiln at the feed
end and going into an afterburner where they are burned with air. Two
parallel boilers generate 90 720 kg (200,000 Ib) of steam per hour at 2.38
MPa (330 psig) and saturated conditions, and gases are cleaned in a wet
scrubber. Residual char is landfilled and ferrous metals are recovered. An
overall net conversion efficiency (steam energy/refuse energy) of 42% is
obtained.
ECONOMICS - The capital cost of the plant is now somewhat in excess of $20
million, which leads to a capital cost per Mg of $6.12 ($5.55/ton). Esti-
mated operating costs if the plant were operating normally are $8.38/Mg
($7.60/ton). In 1975 it was estimated the plant could break even without a
drop charge^ but this is considered to be quite optimistic and only success-
ful operation will develop the required revenue and cost information.
CLEAN AIR TO ATMOSPHERE
t,
STACK
RESIDUE
B-34
-------
NAME - Georgia Tech Mobile Pyrolysis System
TYPE - Low temperature air blown pyrolysis system producing char and oil.
DEVELOPER - Georgia Tech Engineering Experiment Station, Atlanta, Georgia
30332.
HISTORY - Studies began in 1967 to develop an alternative to incineration
for disposing of agricultural wastes such as peanut hulls and sawdust. Four
experimental systems ranging in size from 5.4 to 22.7 Mg/d (6 to 25 TPD) dry
input weight have been run. In addition, a 45.4 dry Mg/d (50 TPD) demon-
stration plant built by the Tech-Air subsidiary of American Can Co., a
licensee, has been operated at Cordele, Georgia, for three years on sawdust,
A preliminary design for a 91 Mg/d (100 TPD) dry feed weight mobile system
to fit on two trailer units has been developed under an EPA grant.
PROCESS - Waste is fragmented to 2.5 cm (1 in.) and then dried to 10% mois-
ture. The dried waste enters the converter through an airlock and is pyro-
lyzed as it descends, with partial combustion air provided through a rotating
mixing arm. The desired end product is char, so pyrolysis temperatures are
kept low, between 426-760°C (800-1400°F). Gases and oil vapors are taken
out through a condenser to remove the oil, which can be mixed with the char
in the ratio of 60% char and 40% oil to produce a dry flowing fuel with a
heating value of 27.9-32.6 MJ/kg (12,000-14,000 Btu/lb). The pyrolysis gases
are used to run an engine-generator to operate the system, and to fuel a
burner to provide hot gas for drying the feed waste.
This system will produce approximately 225 kg/Mg (450 Ibs/ton) of oil and
char from 50% moisture waste. The system is designed for use with cellulosic
agricultural wastes: pollution can be adequately controlled by burning the
gases in the burner and using cyclones to collect dust.
ECONOMICS - The developers made a cost analysis for a 91 dry Mg/d (100 TPD)
mobile system operating 250 days per year. A disposal charge of $3.30/Mg
(3.00/ton) of wet waste was assumed along with a product sales price of
$38.60/Mg ($35.00/ton). With their $405,000 assumed capital costs, an annual
net profit of some $300,000 was calculated. Even under more conservative
assumptions given in Section 6 (including $800,000 capital costs), Parsons
concludes the system can be economically attractive.
COARSE
WET
SAWDUST
HOT BURNT GAS
CLEAN EXHAUST
OIL AND CHAR PRODUCT
ELECTRIC POWER TO
OPERATE SYSTEM
B-35
-------
NAME - Andco-TORRAX Solid Waste Energy Conversion System
TYPE - High temperature slagging pyrolysis using pre-heated air and producing
steam and slag.
DEVELOPER - Andco, Incorporated, Buffalo, New York
HISTORY - Torrax Systems Inc. was created in 1969 by Andco Incorporated and the
Carborundum Company. Under EPA sponsorship, a 68 Mg/d (75 TPD) demonstration
plant was put into operation in 1971. Since the summer of 1972, this plant has
operated as a development facility. Various waste types, including sewage, oil
PVC plastic, and tires were tried in combination with MSW. In early 1976, Andco
acquired all of the licensing rights to the process. A 181 Mg/d (200 TPD) plant
in Luxembourg reached the start-up phase in late 1976. A 113 Mg/d (125 'I'FlJj
plant in Grasse, France, and a 181 Mg/d (200 TPD) plant in Frankfurt, Germany,
are under construction.
PROCESS - Andco-TORRAX is a high temperature pyrolysis system, schematically
shown in Figure 23, that reduces wastes to slag and hot gases. The gases are
immediately burned and passed through a boiler to generate steam.
MSW is dumped into a storage pit and transferred by an overhead crane to the top
of a cylindrical gasifier. The refuse descends through the drying, pyrolysis,
and primary combustion zones. The remaining char is burned with pre-heated air
and forms an inert slag with the inorganic materials. The gases are taken from
the gasifier to the secondary combustion chamber at 427-538°C (800-1000°F), are
mixed with the minimum air necessary for complete combustion, and burned to
produce temperatures of 1204-1260°C (2200-2300°F).
Approximately 15% of the combustion gases are used in two regenerative towers to
pre-heat the primary combustion air to 1093°C (2000°F). The remaining gases
are passed through a boiler to produce steam. The Luxembourg facility produces
approximately 2.46 units of steam per unit of refuse input at 3.45 MPa (500 psi)
and 385°C (725°F) to be used for generating electricity. The cooled combus-
tion gases are cleaned by an electrostatic precipitator and exhausted through a
short stack.
The conversion efficiency of the process (energy in the output steam divided by
the amount of energy in the input refuse) is 71.3%. The net thermal efficiency
(energy in the steam less the steam energy required to produce the required
electric power, divided by the sum of all the input energies) is 52.7%.
ECONOMICS - For a 900 Mg/d (1000 TPD nominal) plant, capital costs in 1976 dol-
lars are estimated to be $34,600,000. Annual operating costs are $2,740,000.
Unit costs would be $21.75/Mg ($19.73/ton) of refuse input and at a charge of
$4.36/Mg ($1.97/1000 Ib) for steam and refuse drop charge of $11.00/Mg ($10.OO/
ton), the plant can break even.
B-36
-------
NAME - Union Carbide PUROX Pyrolysis System
TYPE - High temperature slagging pyrolysis system using oxygen to produce a
combustible gas Magnetic metals recovery is recommended/with other mate-
rials recovery being optional.
DEVELOPER - Union Carbide Corporation, New York, N.Y.
HISTORY - Development began in 1967. A 4.5 Mg/d (5 TPD) pilot plant was
™1 ™TfrytT' N-Y" in 197° and tes^d for three years. A 180 Sg/d
(200 TPD) demonstration plant was built in South Charleston, W. Virginia in
1974 and is still operating. It is considered to be a full-size coLerc al
plant, although units up to 318 Mg/d (350 TPD) may be designed.
PROCESS - Solid wastes, following rough shredding and ferrous metals removal,
are loaded into the top of a vertical shaft converter. As the wastes descend
through the converter, they are first dried by the rising hot gases and then
heated to 315-982°C (600-1800°F). Undergoing pyrolysis, the wastes are ther-
mally decomposed in an oxygen-deficient atmosphere, generating fuel gas and
char. The fuel gas rises, and the char drops into the combustion zone, where
it is burned with oxygen. The heat of combustion is sufficient to melt all
inorganic materials such as glass and metal into a slag, which is tapped from
the converter and quenched. The gas is cleaned in a water spray column and
an electrostatic precipitator. Excess moisture is removed in a condenser.
The spent liquor removed from the gas in the spray column is processed in a
decanting system where the oil and solid particulates are removed and the
water recycled. Excess water is fed to a wastewater system for treatment
and disposal to a municipal sewer.
The fuel gas has an average heating value of 14.2 MJ/m3 (370 Btu/scf).
Approximately 70% of the original energy content of the solid waste is re-
covered in the fuel gas, but roughly 1/3 of this energy, in the form of elec-
tric power, is required to operate the oxygen plant. The fuel gas can be
burned in a boiler or chemically converted to methane, methanol, or ammonia.
ECONOMICS - For a proposed 5-converter, 1361 Mg/d (1500 TPD) system, 1976
capital costs are estimated at $62,400,000 including the front-end shred-
ding and magnetic metals recovery systems. Annual operating and amortiza-
tion costs are estimated at $13,620,000. In one of several cases presented
in Section 8, product gas costs are indicated to be $1.841/million Btu with
a $10/ton drop charge.
FUEL GAS PRODUCT
TO COMPRESSION
B-37
-------
NAME - Occidental Flash Pyrolysis Process
TYPE - Conversion of the finely ground organic fraction of MSW to a liquid
fuel by a rapid pyrolytic reaction.
DEVELOPER - Occidental Research Corporation, La Verne, CA.
HISTORY - Occidental (then Garrett Research) began work in 1968 on improved
methods for isolating various fractions from waste materials and converting
the organics to fuels by a proprietary pyrolysis system, A 6 Mg/d (7 TPD)
pilot plant was used to investigate process variables, which after initial
study of a range of conditions, were then optimized to yield an oil for direct
utilization in utility boilers. Construction of a demonstration plant at
El Cajon, California, was begun in August, 1975, and first testing began early
in 1977. The plant handles 181 Mg (200 tons) of raw refuse per day and the
liquid fuel product amounts to 31.8 m^ (200 barrels) daily.
PROCESS - MSW is shredded to less than 10 cm (4 in.) size and magnetic metals are
removed. An air classifier then separates most of the organic fraction from
the inorganic. The classifier underflow stream is used as a feed to a recovery
system. A trommel screen isolates essentially brittle (glass) materials and
non-ferrous metals. Oversize from the trommel is returned to the grinders.
The brittle fraction, ground to 44 to 840 jum (325 to 20 mesh), is pulped in
water in the presence of proprietary chemicals and from a series of froth flota-
tion tanks is produced a material that is 99.5% glass and containing approxi-
mately 70% of the glass in the original waste. The aluminum-rich fraction is
conveyed to the "RECYC-AL" unit, where linear induction motors separate away
about 60% of the original non-ferrous metals by eddy current effects. The
material contains 90 to 95% aluminum.
The light fraction is dried in a rotary kiln and amounts to 55 to 60% of the
weight of the raw MSW. It is further ground so that 80% is less than 1200 jum
(14 mesh) and then pyrolyzed at 500°C (950°F) to convert the organics to gaseous,
liquid, and solid (char) products. Air is excluded from the reactor by using
recycled gas for the carrier gas and heat exchange is accomplished by use of
heated char being carried co-currently in the stream. Quench oil is used to
terminate further decomposition of the product mix. A portion of the gas is
used for fuel in the rotary dryer.
Some 20 percent of the dry feed becomes char having an HHV of 19.10 MJ/kg (8,200
Btu/lb), while 40 percent is converted to oil with an HHV of 24.6 MJ/kg (10,600
Btu/lb). The gas yield is 30 wt-% of the dry feed: it has an HHV of 15.0 MJ/Nm3
(380 Btu/scf). Oh the basis of raw refuse, 38% is recovered as products, 44%
is consumed by the process, and 18% is residue. Assuming the electrical power
required for the plant to be generated by use of an equivalent amount of pyro-
lytic oil, an overall energy recovery of 33% results from this process.
ECONOMICS - Total capital cost for a 907 Mg/d (1000 TPD) plant is estimated
to be $28.6 million or $10.13/Mg ($9.19/ton). Operating costs for this plant
are estimated to be $19.07/Mg ($17.30/ton). Revenues would be $13.64/Mg
($12.37/ton) based on EPA standardized accounting format for a net cost of
$15.56/Mg ($14.12/ton). At the 1814 Mg/d (2000 TPD) plant capacity level,
this net is estimated to drop to $8.51/Mg ($7.72/ton).
B-38
-------
COMBUSTION PROCESSES
Introduction
Incineration of combustible wastes reduces their volume and converts them
chemically into products that usually can be safely discharged into the
environment after processing through applicable pollution control equipment.
Most combustible wastes, whether solid, liquid, or gas can be incinerated,
although handling care must be taken with plastic, toxic, or hazardous waste.
For ideal incineration, combustible waste must be well mixed with air in
the proper proportions, ignited, and retained in a hot area - the combustion
chamber - until it is completely burned. When the waste materials are com-
bustible gases, liquids, or pulverized solids, this is relatively easy to do.
Some wastes, however, have significant percentages of moisture, noncombustibles,
or are not pulverized, and these require special incinerator designs.
The principal products of any combustion are water vapor, oxides of car-
bon and sulfur, and nitrogen. Some of these oxides, for example sulfur
dioxide (S02) are air pollutants. Municipal solid waste, fortunately, has a
low sulfur content. Plastic wastes often contain chlorine, which can result
in the formation of hydrochloric acid (HC1) during combustion. Toxic and
other hazardous wastes usually can be safely incinerated by maintaining a high
temperature and long residence time to insure that they break down completely
into less harmful products. In several cases, however, the oxidized species
are as toxic, or even more so, than the original compounds. Fluorinated hydro-
carbons, for example, whether in the form of aerosol propellant or the non-
stick polymer coating on kitchen utensils, can thermally decompose into com-
pounds having a high order of inhalation toxicity and the fully oxidized form,
hydrogen fluoride, is a gas with a Threshold Limit Value of 8 ppm for a
1-hour exposure.
Combustion of solid and liquid wastes, and occasionally also gaseous
wastes, produces fly ash, which is an air pollutant if allowed to escape. The
amount of fly ash depends on combustion efficiency and the percentage of non-
combustible material in the incoming waste, including coatings on paper stock,
minerals in wood and agricultural crop waste, and dust in blast furnace gas.
Combustion of waste materials releases significant quantities of heat,
which is taken out of the combustion chamber in the combustion products,
through the furnace walls, or transferred back into the incoming waste.
Originally, when the intent of incineration was simply to consume a waste
material, the heat that is produced became a waste. More recently, the need
to have greater control over the combustion process, to reduce the stack gas
temperature to permit the removal of any fly ash, and to avoid the use of
more expensive conventional fuels, has brought about a reawakening of interest
in usefully recovering heat from the combustion of municipal, industrial, and
agricultural wastes.
B-39
-------
Combustion of Municipal Waste
By 1920, more than 200 municipal incinerators were operating in the
United States (B-106), and many of these were equipped with boilers for the
production of steam. A large number failed to operate satisfactorily because
of poor design, unskilled operation, and use of too little auxiliary fuel.
Throughout this period, incinerators were built of refractory brick, with a
waste heat boiler added when heat recovery was practiced. Fly ash control
was always a problem, and when air pollution laws began to tighten, most
incinerators were faced with either shutting down or adding expensive air
pollution equipment.
Meanwhile, incineration with heat recovery was widely practiced in
Europe, with the steam being used for district heating or electric generation.
By removing the heat to make steam, the incinerator exhaust stack gases were
cooled sufficiently to permit the use of high efficiency electrostatic precipi-
tators to remove the fly ash. The walls of these furnaces were made of closely
packed water tubes which formed the steam boiler, resulting in the term
"waterwall" construction. From being an added-on part of the incinerator,
the boiler had become an integral part of the furnace combustion chamber, just
as in any modern fossil fuel steam generating plant.
Waterwall Furnaces
Refractory walled incinerators require a very large air supply, between
100% and 200% in excess of the air theoretically required for combustion
(Ref B-107). Waterwall incinerators require only 25% to 30% excess air
(Ref B-107), an important advantage, because this means that, for a given
capacity, the incinerator and pollution control equipment will be smaller for
waterwall construction.
Currently, there are three basic designs of waterwall incinerators: mass
burning in a thick bed on a moving grate, semi-suspension firing with burnout
on a traveling grate, and supplementary firing in an existing boiler.
The mass burning design follow configurations proven in European opera-
tion. A cross section of a typical facility is shown in Section 3 under the
description of the RESCO, Saugus, plant. The boilers are especially adapted
for. burning low grade fuel with high ash content. Refuse placed in a hopper
falls down a chute and onto the moving grate. The refuse in the chute forms
a plug against flashback from the boiler, with the rate of feed being governed
by the movement of the grate.
There are several types of moving grates, but they all work to agitate
and tumble the wastes to promote complete and uniform combustion. The
reciprocating grate slopes downward from the feed end. Every other grate bar
moves, pushing the burning refuse down the grate and agitating it. As with
most types of grates, there may be as many as three major sections with a step
in between, over which the waste falls to improve mixing.
The reverse reciprocating grate has sliding grate bars that push upwards
and backwards instead of along the path of refuse travel. This design pro-
vides deep beds with more tumbling motion, and is common in large plants.
B-40
-------
The roller grate consists of a number of turning cylinders arranged in
steps, so that the waste is repeatedly tumbled as it is burned.
The traveling grate consists of an endless conveyor that carries the
burning waste through the combustion chamber. Although there may be several
stages, the grate itself does not agitate the waste, so it is not as commonly
used in mass burning furnaces as the other types.
In all mass burning designs, the combustion bed is very thick, as much
as 1 m (3 ft) deep. The grate is adjusted to achieve complete burnout of the
refuse before it drops into the ash discharger, which often includes a
quenching system.
Another type of firing system involves fuel suspension or semi-suspension.
Both require mechanical processing of the waste to concentrate the combustible
fraction and to reduce the particle size of the waste. The shredded and
classified waste is then introduced into the furnace by a mechanical or
pneumatic feeder. The refuse ignites in suspension and because it is
surrounded by air undergoes rapid combustion. In semi-suspension systems, the
burning refuse falls on a traveling grate where it completes burning in a
relatively thin bed. Supplementary firing of wastes in utility boilers can
be accomplished in furnaces designed for burning pulverized coal, since they
are equipped with ash grates. The pulverized, classified refuse is blown
tangentially into the four corners of the furnace, creating a swirl pattern
that promotes complete combustion in suspension. Because of its relatively
heavy ash and slow burning characteristics, the refuse in most cases to date
has amounted to only 10%-15% of the heating value of the total fuel.
The air for supporting combustion is typically preheated and blown into
the furnace. In designs using moving grates with thick combustion beds, most
of the air is underfire air blown up through the grates. Overfire air, blown
in over the burning refuse, is carefully aimed and controlled to provide the
desired flame pattern and residence time. In semi-suspension furnaces, most
of the air is overfire air, but underfire air still is necessary to promote
complete burnout on the grate. In supplementary firing systems, the airflow
direction and quantity is calculated to keep the burning material in turbulent
suspension, with none of the air being supplied through the grate.
The walls of these furnaces are made of panels consisting of vertical
tubes, typically 6.3 cm (2.5 in.) in diameter on 7.6 cm (3 in.) centers with
a bar welded between the tubes. These waterwalls usually extend down to the
grate level, and have a covering of refractory for a short distance above the
grate to protect the tubes and reflect heat back onto the grate. Most of the
tube area is uncovered, because the refractory acts as an insulation, but areas
subjected to wear may also have protective coatings.
The waterwall section is the boiler or steam generator that heats liquid
water into steam. It is followed by a section of tubes hung in the hot gas
flow called the superheat section. The steam coming from the boiler contains
no liquid water droplets, but it would if its temperature dropped only a few
degrees. In addition, there is still significant heat remaining in the flue
gases that could be recovered in the steam. Therefore, in the superheat
B-41
-------
section the steam temperature is raised well above the saturation (vapor)
point, enhancing its ability to do useful work.
Following the superheater is the economizer section, another section of
tubes hanging in the still hot flue gas stream, in which the boiler feed water
is heated to a point just below boiling.
Because of the high fly ash and soot content of incinerator flue gases,
the passages between the tubes of the pendant superheater and economizer can
easily become blocked. Some form of soot blowers or mechanical tube rappers
must be provided to knock the accumulation off the tubes. In addition, the
tubes are more widely spaced than is usual on oil or natural gas fired boilers
in order to reduce the chances of bridging between them.
From the incinerator, the cooled combustion gases go through pollution
control devices to remove any remaining fly ash, through an induced draft fan,
and out the stack. Wet scrubbers and filters were originally used to collect
the fly ash, but a need for greater efficiency in removing particulates
caused electrostatic precipitators to be used almost exclusively for a number
of years. These devices, in which static electricity captures the particles
with an efficiency of over 99%, are still the principal means of pollution
control used, but new filter systems developed are equally effective and are
beginning to be utilized.
In addition to erosion and fouling by fly ash, chemical corrosion can be
a problem when certain types of wastes are used as fuels. Corrosion comes
from at least three sources: oxygen-deficient burning resulting in a reducing
atmosphere; presence of chlorine, certain metals, and other chemicals; and
presence of moisture (Ref. B-107, B-108). Oxygen deficient burning can occur,
despite the provision of excess under-and over-fire air, when uneven distri-
bution of fuel or air results in partly oxidized (burned) gaseous products
such as carbon monoxide. In seeking more oxygen to complete combustion, these
gases reduce the oxygen in the protective layer of metallic oxides on the
boiler tubes, paving the way for other corrosive activity. The cyclic
reduction-oxidation reactions can result in a high degree of tube wastage.
Chlorine from plastics and rubber, and tin and zinc from food containers,
react directly with the iron in the boiler pipes. All forms of corrosion are
greatly enhanced by the presence of moisture, which can occur if the boiler
is allowed to cool, for example, by shutting down on weekends and holidays.
The widely varying results reported on corrosion in refuse fired systems is
undoubtedly a result of the heterogeneity of the waste and widely varying
operating practices of different systems.
Other Municipal Waste Combustion-Energy Recovery Systems
There are other municipal waste combustion systems with energy recovery,
including starved-air incinerators, vortex furnaces, and fluid bed combusters.
They are aimed at overcoming the major disadvantage of a waterwall furnace:
its great size, and consequent high cost.
Starved-air incinerators are made by a number of firms in varying
capacities up to about 13 Mg (14 tons) per day (Refs. B-109, B-110, B-lll,
B-42
-------
B-112), but can be installed in multiple units to achieve the desired plant
capacity. Municipal waste is burned as received in a primary combustion
chamber with insufficient oxygen for complete combustion. The combustible
materials that are not burned are gasified. This gas is then burned in a
secondary combustion chamber with excess air, achieving temperatures suffi-
ciently high to burn out the carbonaceous fly ash. For a unit with a capacity
of 1.3 Mg (2,800 Ib) per hour, the primary combustion chamber measures approxi-
mately 4 m (13 ft) long by 3 m (10 ft) in diameter; the secondary chamber is
much smaller. Overall dimensions are about 9.4 m (31 ft) long, 4 m (13 ft)
wide, and 8.5 m (28 ft) high, including the stack.
These incinerators are designed with a minimum of ancillary equipment,
and most do not have automatic ash handling equipment. They therefore operate
on a 24 hour cycle. In the morning, after being cleaned out from the previous
day, they are preheated with auxiliary fuel, either oil or natural gas. Waste
is fed in by an automatic loading system, consisting of a hopper, a hydraulic
ram, and fire door. Charging normally continues for seven to eight hours and
burndown for another three hours, with auxiliary fuel being used throughout
this period. The unit is then allowed to cool overnight, and is cleaned of
ash in the morning before a new cycle begins. In contrast to large water wall
furnaces, there are few increased maintenance problems due to this cool down,
due to the refractory lining of the small unit and the absence of bare metal
air pollution control equipment.
Because of the two stage burning with limited air in the primary stage,
gas velocities are kept low, avoiding fly ash pickup. The combustion in the
second stage is in the gas phase, permitting a hotter and more complete oxida-
tion. The result is very low stack emissions, averaging 0.07 to 0.19 g/Nm3
(0.03 to 0.08 grains/SCF) (Ref. B-lll). This is below the U.S. Environmental
Protection Agency's limit of 0.19 g/Nm3 (0.08 grains/SCF) for large
incinerators.
Where energy recovery is practiced, a waste heat boiler of either the
water tube or the fire tube type is installed in the flue, although an air
or water heater could also be used (Ref. B-109). Energy recovery can be
regulated by controlling the amount of hot flue gas that is passed through the
boiler, either with mechanical flap valves or aerodynamic valves. The steam
produced is not sufficient for efficient electrical generation, but can pro-
vide process steam for many industrial applications. In a recent test, one
unit produced 21.5 Mg (47,425 Ib) of steam at 791 kPa (100 psig) from 7.5 Mg
(8.3 tons) of refuse and 102.2 m3 (3610 ft3) of natural gas auxiliary fuel.
Weight reduction of raw waste in these incinerators averages 68%, and volume
reduction 93%, comparing very favorably with large waterwall incinerators.
A vortex furnace is a circular horizontal or vertical furnace in which
air is blown in tangentially, creating a swirling vortex. Wastes can be
either finely ground and blown in to burn in suspension, or can be mass
burned, with the hot gases scrubbing over the waste. In either case the object
of the turbulent vortex is to provide good mixing between combustibles and
the air to promote complete and rapid burnout. In order to achieve maximum
residence time, the aerodynamics of the vortex is often arranged so that the
air must follow a spiral along the walls of the furnace and mix with rising
B-43
-------
volatile matter. Only after the air and volatiles are burned to low density
combustion products can they escape through the center of the vortex.
In a vertical cylinder, refuse is fed from the bottom and air is intro-
duced high up through side ports. The air moves in a downward spiral along
the walls to scrub the fuel bed and mix with rising volatile matter.
In the horizontal cylinder, refuse is fed by a hydraulic ram. The com-
bustion air is delivered to the combustion chamber via a manifold and distri-
buted in such a manner as to set up a circular turbulent action. In both
designs, the walls of the furnace are cooled by the incoming air.
During the period of 1967-70, a pilot vortex incinerator plant was built
and tested at the EPA's Solid and Hazardous Waste Research Laboratory,
Cincinnati, Ohio (Ref. B-113). The purpose of the project was to advance the
state-of-the-art by improving combustion efficiency, volume reduction, and
mechanical reliability and to reduce air pollution to an acceptable level.
The completed incinerator was a horizontal cylinder 3.66 m (12 ft) long
and 1.83 m (6 ft) in diameter. Refuse was charged to the incinerator by a
hydraulic ram. The stoking action was accomplished by the incoming refuse
pushing the burning refuse across the floor of the incinerator and finally out
the other side into the residue pit. This kind of stoking eliminates the need
for grates. Any slag that was formed flowed or was pushed into the residue
pit. Heated combustion air was injected so as to set up a cyclone or vortex
action aiding the stoking action by exposing more burning surface to combus-
tion. The desired burning rate was 454 kg/h (1000 Ib/hr), with 90% volume
reduction and 80% weight reduction. The cyclonic movement of the gases down
the combustion chamber was to provide at least 0.5 second residence time at
temperatures in excess of 1315°C (2400°F).
The data collected on the incinerator's combustion chamber operations
indicated it did a very effective job in burning untreated municipal refuse.
Low solid particulate concentration, small particle size, high heat release
rates, and high carbon dioxide concentrations indicated the combustion chamber
performed well. The burning rate was higher than the design specifications,
but was necessary to sustain high combustion temperatures. The result of the
elevated burning rate was a poorer quality residue, even though the volume
and weight reductions were acceptable. Adverse conditions in the stack led to
termination of the project. Stack gas tests indicated that the concentration
of total particulate matter was in excess of established regulations. This
was attributed to a high concentration of water soluble matter in the effluent
that counted as particulates by the test method and the use of an inefficient
cyclone as a control device.
Another vortex incinerator development project was conducted at
Shelbyville, Indiana, by General Electric and the City of Shelbyville with
EPA support (Ref. B-114). Municipal refuse was used as fuel. Finely
shredded combustibles were blown tangentially and at high velocity with
primary air into a horizontal cylindrical furnace to form a vortex. The
facility was built with a refuse receiving apron, primary shredder, rotary
air classifier, secondary shredder, and cylindrical furnace. Pollution
control depended only on a single cyclone for large particulate removal.
B-44
-------
Some 109 Mg (120 tons) per day of raw refuse was accepted, with 23 Mg
(25 tons) per day of heavy materials, such as metals, glass, plastics, rock,
etc. separated. Thus, 86 Mg (95 tons) per day of combustible refuse was
incinerated in the 2.59 m (8-1/2 ft) diameter by 3.96 m (13 ft) long vortex
furnace.
The secondary shredder reduced combustibles to less than 12 mm (1/2 inch)
in size and a pneumatic feeder blew the material into the vortex incinerator.
Secondary air was blown in tangentially along the length of the incinerator
cylinder through inlet ducts. The temperatures reached during combustion were
982 to 1093°C (1800 to 2000°F) . All ash was carried out with the exhaust
gases at high velocity. However, certain heavy items and non-combustibles
that passed through the front-end separator systems were caught in the
incinerator and it was necessary to open the furnace at least once each day
to remove such material.
A private venture in designing and operating a vortex furnace is being
made by ACES in Red Lion, Pennsylvania (Ref. B-115). The system has a front
end consisting of shredding, air classification, and magnetic separation.
The prepared refuse is fed into the top of a vertical, refractory line circular
furnace along with the primary air. As it falls through the furnace, it is
swirled into a vortex by the tangentially blown secondary air. Combustion of
the waste, which averages 34% moisture, takes place in less than two seconds,
at temperatures as high as 1343°C (2450°F). The combination of a turbulent
vortex and a low 20% to 25% excess air results in efficient heat exchange.
Approximately 11% of the input fuel is inerts, largely glass. Ninety four
percent of that comes out the bottom of the furnace as slag; the remaining
6% (less than 1% of the original input fuel) is carried over into the boiler.
For the original tests at Red Lion, the hot gases from the vortex furnace
were taken through a refractory lined tunnel to a standard package firetube
boiler rated at 11 340 kg (25,000 Ib) of steam per hour at saturated condi-
tions and 929 kPa (120 psig). During most of the tests the boiler was
operated in the 8164 to 9072 kg (18,000 to 20,000 Ib) per hour range, exhaust-
ing the gases to the stack at 177°C (350°F). Particulates downstream of a wet
scrubber measured 0.60 g/Nm3 (0.25 grains/SCF) with 6 ppm of hydrocarbons and
no CO.
On a gross heat input basis, the developer claims an efficiency of 64%.
The moisture in the refuse accounted for 7.8% of the lost 36%, 12% was in the
exhaust, and 8% was contained in the hot slag.
Fluid bed furnaces take a different approach to holding wastes in suspen-
sion while they are burned. An inert material, such as sand, is "fluidized"
by blowing air up through it. Through use of the right amount and velocity of
air, the bed can be made to behave very much like a liquid, and refuse fed
into the fluid bed will be brought into close contact with the hot sand.
Very high rates of heat transfer can be obtained because of the contact with
the sand, and the fluidizing air provides all of the oxygen required for
combustion.
B-45
-------
The CPU-400 pilot plant, built and operated by Combustion Power Co. Inc.
in Menlo Park, California, employs the fluidized-bed c'ombustion process.
Shredded and air classified combustibles are burned at high pressure in a com-
bustor, with the combustion gases then used to drive a turbo-electric
generator (Ref. B-24). The combustor is a 6.7 m (22 ft) high by 2.9 m
(9-1/2 ft) diameter cylindrical carbon-steel pressure vessel with dished heads
and lined with fire brick. The cylinder is penetrated by 161 air diffusers.
An oil burner used for preheating the sand bed during cold system startup is
mounted in the top dome of the combustor. Six oil guns penetrate the combustor
immediately above the distributor plate for use during start up.
The fluid bed combustor provides excellent combustion at temperatures
of 816 to 982°C (1500 to 1800°F), resulting in low emission of contaminants.
A train of three separators removes sand and ash particles from the exhaust
gas prior to its entering the turbine. A serious difficulty, not yet elimi-
nated, is carryover of various fine particles that erode the turbine blades.
A moving bed granular filter has recently been studied in this EPA-supported
work, but the equipment experienced a structural failure early in the test
program. ERDA has now accepted the responsibility for further development
of this device.
Both the vortex and fluidized bed furnaces work best when mechanical pre-
treatment of refuse is used to remove heavy materials and non-combustibles that
have bad effects on their operation. Raw refuse before being fed to the
furnace is usually shredded, the ferrous metals removed, and the heavy fraction
isolated by means of an air classifier.
Both systems have also been used with add-on boilers, called waste heat
boilers because their original use was to capture heat from gases in incinera-
tors and other heat generating processes that would otherwise be wasted.
These boilers have upper and lower drums connected by water tubes, and are
made as package units by a number of manufacturers.
Combustion of Industrial and Agricultural Waste
In the constant effort to reduce costs in industrial production, there
is naturally a continuing effort to find ways to recycle or use the process
wastes. Saw mills may have been among the first to use their wastes to
produce energy for running the process, but today many industries practice
energy conversion and recovery. In recent years, interest in waste-to-energy
conversion has increased. This is because of cost increases for competing
fuels, restrictions on more conventional disposal methods such as landfills,
the need for heat removal to make clean up of incinerator flue gas easier,
and the low sulfur content of most process wastes.
In general, solid wastes can be burned on a fuel bed or in suspension.
Liquids are usually introduced into combustion zones as a fine spray. Gases
can be burned as they are produced, although some, like blast furnace gas,
require cleaning to remove particulates. A variety of furnace types can be
used to consume these wastes, including waterwall furnaces, refractory fur-
naces with package boilers, vortex and cyclonic furnaces, fluid bed combustors,
and molten salt baths.
B-46
-------
Waterwall furnaces for industrial wastes are very much like those for
coal or municipal waste; firing can be either on a grate or in suspension,
depending on the waste characteristics. Feed mechanisms for relatively wet
materials such as sugar cane bagasse act like a rotating series of paddles
that toss the chopped bagasse onto a traveling grate. At the other end of
the spectrum, a large carbon monoxide boiler used with an oil refinery needs
only huge ducts and gas ports to carry the hot gas from the catalytic cracker
into the furnace. Wood waste boilers are very common at saw mills; a listing
by just one stoker manufacturer indicates seventy-six installations burning
bark and other wood wastes with their equipment; it is estimated that there
are in excess of 500 such systems in the U.S. A large unit installed by the
Weyerhaeuser Company at Longview, WA, produces 249 480 kg (550,000 IbJ of
steam per hour on hogged fuel using suspension firing with a burnout grate
fed by a vibrating flow splitter. The steam is run through a turbogenerator
to produce electric power and then goes to operate plant processes such as
mill dryers and digesters.
Refractory furnaces were in use before the waterwall type, and are still
used in some applications. Coffee grounds, which have a high moisture content,
are sometimes burned in a cell-type furnace, either on flat grates or on the
furnace floor. Liquids can be sprayed into a refractory furnace by an
atomizing burner, and small carbon monoxide boilers use refractory furnaces
in conjunction with a package boiler. These systems are usually low waste
volume installations that cannot justify the design of a special furnace
and boiler; it is much quicker and easier to use shop-built package units.
Vortex and cyclonic burners hold fuel particles in the burner by centrif-
ugal force until they are completely burned up. Air is blown tangentially
into the furnace; fuel is either fed in with the air (vortex type) or fed in
onto a rotating hearth (cyclonic type). Because these furnaces operate with
a low excess-air ratio with essentially all of the heat release from the fuel
taking place inside the relatively small combustion chamber, combustion gases
can be ducted to a package waste heat boiler or to some other use of hot dry
gases, such as a lumber kiln or veneer dryer. The Goodyear Tire and Rubber
Company has reported success in dealing with a hard-to-dispose-of waste, used
tires. A rotating hearth furnace is used to destruct the tires; the hot
combustion gases are taken through a conventional two-drum design waste heat
boiler with bare tubes in the first section and finned tubes in the second
section. A scrubber is used to clean up the cooled flue gas before it is
discharged (Ref. B-116).
Fluid bed combustors, similar to those used for municipal waste and
sewage sludge, are widely used in industry for drying, converting, and
burning a variety of materials. They are widely used in the pulp and paper
industry to convert wastes into energy, and their use is increasing as
supplies of natural gas decline. Fluidized beds can be made with in-bed heat
exchangers, waterwalls, or package boilers. They have the advantage for many
installations that combustion can be obtained at temperatures that can be
controlled to within 5.5°C (10°F). Carryover of particulates is minimized,
although a scrubber may be required depending on the characteristics of the
waste.
B-47
-------
Another heat recovery combustion system frequently used in the pulp and
paper industry is wet air oxidation. There are numerous examples of wet wastes
that have potential for heat recovery, except that all of the recovered heat
would be required to dry the waste to the point where it could be burned. To
avoid this problem, these wastes can be combined with slightly excess air in
a reactor under elevated temperatures and pressures. Wet air oxidation is
very much a temperature-dependent reaction in that different organic compounds
react at different temperatures. Virtually complete oxidation will take place
at temperatures above 316°C (600°F). At these temperatures, pressures of
approximately 20.7 MPa (3000 psi) are required to keep the necessary amount of
water in the waste liquid (Ref. B-117). Wet air oxidation temperatures cannot
exceed the critical point of water, 374.1°C (705.4°F). Production of steam
at temperatures of more than 287.8°C (550°F) is therefore not practical, as
there must be sufficient temperature differential to give good heat transfer.
However, abundant energy in the form of hot water or low pressure steam can be
made. It can be used in areas such as pre-heating boiler feed water, area
heating, process heat, and drying.
A number of other incineration processes have the potential for energy
recovery, although application is not necessarily yet being made. Molten
salt baths, for example, can be used to destroy many toxic wastes because
the salt bath can chemically alter the wastes besides oxidizing them. Pesti-
cides, low-level radioactive wastes, and explosives are among the other wastes
that can safely be destroyed by this system (Ref. B-118). Maximum temperature
for this process is limited to 980°C (1800°F) to avoid overactivation of the
salt. There is some salt carryover from the molten bath, but the process
developers felt that the problems of designing a boiler to operate under
these conditions were well known and would present no difficulty to heat
recovery. No commercial energy recovery system has yet been built.
Descriptions of Key Projects
Brief descriptions of typical combustion projects are presented on the
following pages.
B-48
-------
NAMJB - RESCO Refuse Burning Steam Generator, Saugus, Massachusetts
TYPE - Mass burning waterwalled combustion system
DEVELOPER - Refuse Energy Systems Company
HISTORY - Feasibility studies were made in 1969 and construction began in
June 1973; initial testing began in late 1975 and the facility is now fully
operational.
PROCESS - Designed by Wheelabrator-Frye, Inc. and built by M. DeMatteo Con-
struction Co., this 1090 Mg/d (1200 TPD) facility is based on proven techno-
logy developed by the Swiss firm of von Roll. MSW from surrounding communi-
ties is received and stored in a pit capable of holding refuse for more than
5 days operation; only bulky wastes are fragmented. Furnace feed hoppers
are supplied by overhead cranes. Two waterwall furnaces (max. capacity =
680 Mg/d each) with 3-section reciprocating grates operate at 900°C (1650°F)
to produce 3810 Mg/d (8.4 million Ib/day) of steam at 4.86 MPa (690 psig)
and 468°C (875°F). It is piped to the nearby G.E. Lynn Works. Two Lurgi
electrostatic precipitators reduce particulate emissions to below 0.05
grains/SCF and the clean flue gas is discharged through a stack 54 m (178 ft)
tall. Standby oil-fired boilers are present to assure a product steam supply.
ECONOMICS - Capital costs were $38,268,000 for the total system. Original
drop charges to the participating communities were $14.33/Mg ($13.00/ton),
with RESCO being permitted to charge more for later cities; a cost escalation
formula is incorporated into contracts. Being a private venture operation,
details of revenues and operating costs are not available.
STEAM
WATER
ELECTROSTATIC
PRECIPITATOR
TO STACK
QUENCH TANK
FLY ASH
TO FERROUS METALS RECLAMATION
B-49
-------
NAME - Hamilton Solid Waste Reduction Unit (SWARU)
TYPE - Semi-suspension fired steam generator burning processed waste
DEVELOPER - Regional Municipality of Hamilton Wentworth, Ontario, Canada
HISTORY - Design was initiated in 1968, with initial start-up in February, 1972,
and commercial operation in June, 1972.
PROCESS - Hamilton is unique in burning prepared refuse partly in suspension,
with burnout on a traveling grate. This technology was taken from the wood
products industries. Incoming refuse is dumped into a pit with a conveyor
bottom, picked over manually to remove large materials, and carried to four
vertical-shaft pulverizers. These are relatively light duty, 150 kW (200 HP]
units, selected on the basis of a study that indicated it would be more economi-
cal to landfill large objects, or those difficult to shred, than to provide
larger shredders. Ferrous metals are recovered magnetically and the remaining
refuse is either sent directly to the furnaces or stored for later use in a
21.3 m (70-ft) diameter Atlas bin. Figure 2 shows a schematic of the facility.
The plant has two Babcock and Wilcox waterwalled furnaces, each capable of burn-
ing 272 Mg (300 tons) per day. The prepared refuse is introduced by a swinging
distribution spout and three parallel pneumatic injection chutes. The light
combustibles burn in suspension, with the non-combustibles and heavy burning
objects burning out in a thin bed on the traveling grate. The ash is landfilled.
SWARU is capable of generating 47 944 kg (105,700 Ib) of steam per hour in each
boiler, at a pressure of 1.82 MPa (250 psig) and saturated conditions. Some
50% to 60% of the steam is used in turbines for running plant equipment. The
remainder is condensed in roof-top air-cooled condensers and recirculated. The
gases leave the boiler at 310°C (590°F) and cleanup is performed by two Lurgi
design electrostatic precipitators in series per boiler.
The major problems in the plant have been in the materials handling system,
where bridging, plugging and spilling of the refuse required design changes.
ECONOMICS - SWARU cost a total of $9 million when it went into operation. It has
been estimated that minimum plant improvements to correct operating problems
would cost an additional $1 million. The 1975 operating and maintenance costs
totalled $2,117,000 for disposing of 43.5 Gg (48,000 tons) of solid waste. This
is a disposal cost of $48.62/Mg ($44.10/ton), a very high figure. If the plant
were to be operated at 85% of capacity, it would be capable of disposing of
168.9 Gg (186,150 tons) of solid waste at an annual cost of $2,649,065 (1975).
This would significantly lower the disposal cost to $15.69/Mg ($14.23/ton)
before taking any credit for possible revenues.
The plant recovers ferrous metals, which are sold, and steam for which there is
no local market at present. If the steam were to be sold, a distribution sys-
tem would have to be built, estimated to cost $4 million in 1976 dollars.
B-50
-------
NAME - Nashville Thermal Transfer Corporation, Nashville, Tennessee
TYPE - Mass burning, waterwalled boilers with reciprocating grates, producing
steam for district heating and cooling.
DEVELOPER - Nashville Thermal Transfer Corporation, 110 First Avenue South,
Nashville, Tennessee, a non-profit Tennessee corporation.
HISTORY - As part of an urban renewal program for downtown Nashville, THERMAL
was created to be the developer and operator of a district heating and cooling
system. In 1970, a study was made that established the feasibility of generat-
ing the required steam by burning refuse. Design was started in early 1971 and
construction in June, 1972. Steam service, using an oil fired boiler, was
initiated in February, 1974, and incineration of MSW began in May, 1974. A
number of operating problems, including water tube and control system failures,
had to be overcome. The most widely publicized problem has been THERMAL' s non-
compliance with air quality standards due to reliance on a wet scrubbing system
for flue gas clean-up. One electrostatic precipitator is successfully operating
and a second is now being installed.
From the refuse pit (see Figure 4), unprocessed solid waste is charged by an
overhead crane to the feed chute of the two furnaces and onto a four section
Detroit Reciprocating Grate Stoker; 653 Mg/d (720 TPD) can be burned. The hot
combustion gases generate steam in a pair of Babcock and Wilcox boilers. Many
of the auxiliaries have steam turbine drives, which do not have the high torque
and rapid response characteristics that would be desirable. Similarly,, the
pneumatic and electric control systems were not originally designed with suf-
ficient independency and redundancy, and some problems have been experienced
with them. Steam is produced at 2.86 MPa (400 psig) and 325°C (620°F) . For
the district heating system this is reduced to 1.14 MPa (150 psig) in the
auxiliary drive turbines. Chilled water is supplied at 1.34 MPa (180 psig) and
5°C
Originally, the flue gases were passed through multi-cyclone dust collectors
and then through wet scrubbers. The scrubbers were specified to remove 95% of
all particulate matter based upon a 5 /im mean particle diameter. However, an
average of 28% of the particulates in the flue gas are under 5 urn, and 21.4%
are under 1 /urn. These small particles are difficult to remove with scrubbers
and the new electrostatic precipitators had to be purchased.
ECONOMICS - The initial cost of the facility was $16,500,000, but the start-up
problems have necessitated a capital completion program estimated to cost an
additional $8,000,000. The start-up problems caused shortfalls in revenues
and overruns in expenses. Upon fulfillment of the capital completion program
and receiving an increase in the annual service payment from Metropolitan
Nashville, it is anticipated that THERMAL1 s operating revenues will be adequate
to meet operating expenses.
B-51
-------
NAME - CPU-400
TYPE - Refuse-fired high pressure fluid bed incinerator coupled to an open
cycle turbo-electric generator system.
DEVELOPER - Combustion Power Company, Menlo Park, California, under contract
to the EPA's Office of Research and Development.
HISTORY - The initial proposal for the CPU-400 concept was submitted to the
U.S. Department of HEW in September, 1966. A preliminary design study was made
between June, 1967; and June, 1968. From mid-1968 to late 1970, subscale
experiments to evaluate basic equipment items were conducted under contracts
with HEW, and later, EPA. Pilot plant design and construction were performed
between mid-1970 and April, 1973. Further contracts have been made between
Combustion Power and EPA for test and development since 1973. In addition,
tests have been made using high sulfur coal for the Office of Coal Research and
ERDA and using wood waste for a major forest products company.
PROCESS - The CPU-400 system has four major sections: solid waste processing,
combustion, gas cleanup, and power generation. Pilot plant testing has been
at the 90 Mg/d (100 TPD) level. The solid waste processing facility includes
a tipping/storage area, two shredders, an air classifier, and materials
recovery systems. The light fraction from the air classifier is taken to a
refractory-lined fluid bed combustor 2.2 m (7.1 ft) in diameter and 4.3 m (14 ft)
high. Two rotary airlock feeders are used to introduce the fuel to the bottom
of the combustor's 61 cm (2 ft) deep bed of sand at a rate of 45.4 kg/min. (100
lb/min.). Air from the turbine compressor at about 400 kPa (58 psia) is used to
fluidize the sand and provide combustion air. The combustion gases are taken
through a gas cleanup system to remove particulates before being passed through
the turbine. A 1000 kW axial flow gas turbine is used to extract energy from
the hot gas stream to drive the air compressor and the electric generator.
The most severe problem in the system has been the failure to achieve adequate
gas cleanup. Carryover of aluminum oxide particles from solid waste formed
deposits on'the first stage turbine stator blades. Subsequent sloughing of the
aluminum oxide deposits resulted in severe downstream turbine blade erosion.
The initial gas cleanup system used three stages of cyclones, but the second
and third stages became plugged and allowed uncleaned gas to pass into the tur-
bine. The second stage has been enlarged and a granular filter developed as
the third stage. Tests of this system indicate that more development work is
necessary.
ECONOMICS - An estimate of costs and revenues for a 12 MW, 544 Mg (600 tons) per
day facility published in September, 1974, set the installed capital cost at
$10.8 million. Total annual cost was estimated to be $2,500,000 and revenues
from the sale of electricity only to be $1,140,000, leaving a net annual cost
of $1,360,000 or $7.61/Mg ($6.90/ton) of refuse disposed. Revised economics
should be made when the design is finalized and verified.
B-52
-------
NAME - Longview, Washington, Plant Power Boiler
TYPE - Wood waste serai-suspension fired steam generator
DEVELOPER - The Weyerhaeuser Company, Longview, Washington
HISTORY - In operation since January, 1976.
PROCESS - Hogged fuel, bark, and other manufacturing wood waste averaging
0.64 cm (1/4 in.) in size, but with pieces up to 10 cm (4 in.), is pre-dried
in a rotary dryer, using wood fines as a fuel. The moisture content of the
wood waste fuel is reduced to approximately 30% to 35%. The drying has been
found to be necessary because the fuel is kept in an outside stockpile, an
enclosed storage system not being considered justified. From the dryer, the
waste wood fuel is fed to the furnace by a vibrating flow splitter. A large
part of the fuel is burned in suspension, with final burnout taking place on
a grate in a manner similar to the MSW waterwall incinerator at Hamilton,
Ontario. The boiler produces 250 000 kg/h (550,000 Ib/hr) of steam at 8.6
MPa (1250 psi) and 510°C (950°F). The steam is expanded through turboelec-
tric generators before being used to heat mill dryers, digesters, and other
plant processes.
According to Weyerhaeuser engineers, electrostatic precipitators cannot be
used on wood-fired furnaces because of resistivity characteristics of the
fly ash, so two-stage mechanical collectors are used for flue gas cleanup.
ECONOMICS - No data have been released on the capital or operating costs, but
they could be considered to be equivalent to other large waste-fueled water-
wall steam generators.
B-53
-------
REFERENCES
B-l. Scott, P. J. and J. R. Homes, "The Operational Characteristics of
Refuse Handling Grates," 1972 National Incinerator Conference.
B-2. Scott, P.J. and J.R. Homes, "The Capacity and Principal Dimensions of
Refuse Storage Bunkers in Modern Incinerator Plants," 1974 National
Incinerator Conference.
B-3. "The Tracer Marksman Solid Waste Processing and Resource Recovery
System," Sales brochure by Tracor Marksman, 1975.
B-4. Van Poolen, L., "Energy Recovery From Solid Waste," Vol. 2, Technical
Report, NASA-ASEE 1974 Systems Design Institute. Final report under
NASA Grant NGT 44-005-114.
B-5. Rogers, H.W. and S.J. Hitte, "Solid Waste Shredding and Shredder
Selection," U.S. EPA Report SW-140, November, 1974.
B-6. Ananth, K.P. and J. Shum, "Fine Shredding Study," Final Report,
Midwest Research Institute, July 1976. EPA-600/2-76-208.
B-7. Dilly, C., "Fuel Cleaning and Grinding," Presented at Wood Residue As
An Energy Source, Denver, Colorado, September 1975, Forest Products
Research Society Proceedings No. P-75-13.
B-8. Schwieger, R.G., "Power From Waste," Power, February, 1975.
B-9. "Decision-Makers' Guide in Solid Waste Management," U.S. EPA
Report SW-500, 2nd Edition, 1976.
B-10. Midwest Research Institute, "Fine Shredding Study," Report No. 3 of
EPA Contract No. 68-02-1324, Task, No. 39, 1975.
B-ll. Gawalpanchi, R.R., P.M. Berthousea, and R.K. Ham, "Particle Size
Distribution of Milled Refuse," Waste Age, September/October, 1973.
B-12. Trezek, G.J. and G. Savage, "Results of a Comprehensive Refuse
Comminution Study," Waste Age, July, 1975.
B-13. Trezek, G.J. and G. Savage, "MSW Component Size Distributions Obtained
from the Cal Resource Recovery System," Resource Recovery and Conserva-
tion, Vol. 2, No. 1, August 1976.
B-14. Herbert, W., "Solid Waste Recycling at Franklin, Ohio," In: Proceedings
of the Third Mineral Waste Utilization Symposium, Chicago, Illinois,
March 14-16, 1972.
B-54
-------
B-15. Landman, W.J. and U.J. Darmstadt, "Energy Recovery From Hydrapulping
at Hempstead," In: Conversions of Refuse to Energy (CRE), Conference
paper for First International Conference and Technical Exhibition,
Montreux, Switzerland, November 3-5, 1975, IEEE Catalog No 75CH0008-2
CRE, p. 589.
B-16. Ito, K. and Y. Hirayama, "Resource Recovery From Municipal Refuse by
Semi-wet Selective Pulverizing System," In: Conversions of Refuse to
Energy (CRE), Conference paper for First International Conference and
Technical Exhibition, Montreux, Switzerland, November 3-5, 1975, IEEE
Catalog No. 75CH0008-2 CRE, p. 354.
B-17. Drobny, N.L., H.E. Hull, and R.F. Testin, "Recovery and Utilization of
Municipal Waste," (SW-10C) for U.S. EPA by Battelle Memorial Institute,
Columbus Laboratory, Contract PH-86-67-265, 1971
B-18. Wilson, G.E., ed., The Treatment and Management of Urban Solid Waste,
Technomic Publishing Co. Inc., Westport, Connecticut, 1972.
B-19. "Morbark Industries, Inc. Facts," Company brochure of Morbark Industries
Inc., Winn, Michigan, 1975.
B-20. Sullivan, P.M., M.H. Stanczyk, and M.J. Spendlove, "Resource Recovery
From Raw Urban Refuse," Bureau of Mines Report of Investigations 7760,
U.S. Department of the Interior.
B-21. "The WCS Process," sales brochure of Waste Control Science-International,
306 Mendocino Ave., Santa Rosa, CA, 1974.
B-22. Preston, G.T., "Resource Recovery and Flash Pyrolysis of Municipal
Refuse," Occidental Research Corporation, Presented at the Institute of
Gas Technology Symposium, Orlando, Florida, January 1976.
B-23. Snyder, N.W., "Progress in Municipal Solid Waste Processing,"
Presented at the American Institute of Chemical Engineers 78th
National Meeting, Salt Lake City, Utah, August 18-21, 1974.
B-24. Combustion Power Company, Inc., "Summary of the CPU-400 Development,"
U.S. EPA Report 68-03-0054, 1974.
B-25. Dean, K.C., C.J. Chindgren, and L. Peterson, "Recovery of the Non-
ferrous Metals From Auto Shredder Rejects by Air Classification,"
Bureau of Mines Solid Waste Research Program, Technical Progress
Report 31, April 1971.
B-26. Dean, K.C., C.J. Chindgren, and L. Peterson, "Preliminary Separation
of Metals and Nonmetals From Urban Refuse," Bureau of Mines Solid
Waste Research Program, Technical Progress Report 34, June 1971.
B-27. Hill, R.M., "Effective Separation of Shredded Municipal Solid Waste,"
Waste Age, October 1974, p. 34.
B-55
-------
B-28. Holman, J.C., J.B. Stephenson and M.J. Adma, "Recycling of Plastic
from Urban and Industrial Refuse/' U.S. Bureau of Mines Report of
Investigation 7955, 1974.
B-29. Dale, J.C., "Recovery of Aluminum From Solid Waste," Resource Recovery,
Jan/Feb/Mar 1974.
B-30. "M-E-C, A World Wide Company,..." Sales brochure for the MEC Company,
Neodesha, Kansas; Bulletin 1075 DE, Guaranty Performance Co., Inc.,
Independence, Kansas.
B-31. Corder, S.E., "Wood and Bark as Fuel," Research Bulletin 14, Forest
Research Laboratory, Oregon State University, August 14, 1973.
B-32. "Woodex," sales brochure of Woodex, Inc., 9720 Wilshire Blvd.,
Beverly Hills, California, 1975.
B-33. Arthur D. Little, Inc., "Eco-Fuel II: The Technology and Economics,"
A report to Combustion Equipment Associates, Revision I, January 1976.
B-34. "Now It's Eco-Fuel II," Resource Recovery, Apr/May/June 1974.
B-35. Haynes, J.H., "Garbage Power, Refuse to Energy Systems Have Great
Potential," Barrens National Business and Financial Weekly,
May 20, 1974.
B-36. Hoenig, S.A., C.F. Russ, "Generation of Natural Gas (Methane) From
Garbage and Sewage," Presented at the National Symposium of Energy and
Materials, Washington, D.C., June 9-11, 1975.
B-37. "Clean Fuels from Biomass, Sewage, Urban Refuse, Agricultural Wastes,"
Symposium Papers, January 27-30, 1976, Orlando, Florida, Institute of
Gas Technology, March 1976.
B-38. Ware, S.A., "Bioconversion State of the Art Report," Prepared by Ebon
Research Systems for Solid and Hazardous Waste Research Laboratory,
U.S. EPA, July 1975.
B-39. Kang, E.K., "The Development of Marsh Gas Production From Hog Waste in
Taiwan, Republic of China," IECEC, 1975 Record, p. 828.
B-40. Proceedings of the Bioconversion Energy Research Conference, University
of Massachusetts, Amherst, June 25-26, 1973 (Report No. NSF-RA-N-73-007.)
B-41. Pfeffer, J.T., "Reclamation of Energy From Organic Waste," University
of Illinois, U.S. EPA Report 670/2-74-016, March 1974.
B-42. Ecotope Group, "Process Feasibility Study: The Anaerobic Digestion of
Dairy Cow Manure at the State Reformatory Honor Farm, Monroe, Washington"
for the State of Washington Department of Econology, January, 1975.
B-56
-------
B-43. Kispert, R.G., S.E. Sadek, L.C. Anderson and D.L. Wise, "Fuel Gas
Production From Solid Waste/' Dynatech R/D Company, NSF Contract C-827
1975.
B-44. Fry, L.J. and R. Merrill, "Methane Digesters For Fuel Gas and
Fertilizer," New Alchemy Institute, Pasadena, California, Newsletter
No. 3, Spring, 1973.
B-45. Golueke, C.G., "Bioconversion Energy Studies at the University of
California Berkeley," Proceedings of Bioconversion Energy Research
Conference, University of Massachusetts, Amherst, June 26-28, 1973.
B-46. Skinner, K.J., "Enzymes Technology," Chemical and Engineering News,
August 18, 1975.
B-47. The American City and County, Vol. 91, No. 5, May 1976, p. 46.
B-48. Mignone, N.A., "Anaerobic Digester Design for Energy Generation,"
Public Works, October 1974.
B-49. Lindsley, E.F., "Methane from waste...How much power can it supply?,"
Popular Science, December 1974, p. 58.
B-50. Klass, D.L. and S. Ghosh, "SNG From Biogasification of Waste Materials,"
Presented at the SNG Symposium I, March 12-16, 1973.
B-51. Kispert, R.G., S.E. Sadek, and D.L. Wise, "An Economic Analysis of
Fuel Gas Production From Solid Waste," Resource Recovery and Conserva-
tion, 1, 1975, Elsevier Scientific Company, Amsterdam, Netherlands,
p. 95-109.
B-52. Bisselle, C., M. Kornreich, M. Scholl and P. Speivak, "Urban Trash
Methanation—Background For a Proof-of-Concept Experiment," MITRE
Technical Report, NSF C-928, 1975.
B-53. Resource Recovery, Jan/Feb 1977, p. 26.
B-54. Ghosh, S., J.R. Conrad, and D.L. Klass, "Materials and Energy
Reclamation From Municipal Wastes," Institute of Gas Technology,
Chicago, Illinois, October, 1974.
B-55. Benzieger, J.B. et al, "Resource Recovery Technology For Urban Decision-
Makers," Prepared for the National Science Foundation, Urban Technology
Center, Columbia University, January 1976.
B-56. Lindsley, E.P., "Bryon McDonald: Cowpower helps run his farm,"
Popular Science, V. 206, May 1975, p. 106-107.
B-57. Taiganides, E.P., "Manure Gas Plants," Bulletin No. F13, National Hog
Farmer, Swine Information Service, May 1963.
B-57
-------
B-58. Chemical and Engineering News, October 21, 1974, p. 24.
B-59. Chemical and Engineering News, June 23, 1975, p. 15.
B-60. Wall Street Journal, October 21, 1975, p. 6.
B-61. "Bio-Gas of Colorado," Company brochure of Bio-Gas, Loveland,
Colorado, 1975.
B-62. Catania, P.J., ed., "Food Fuel Fertilizer," Proceedings of Symposium:
Uses of Agricultural Wastes, University of Regina, Regina Inn, Regina,
Saskatchewan, November 4-5, 1975.
B-63. Gaddy, J.L., E.L. Park, and E.B. Rapp, "Kinetics and Economics of
Anaerobic Digestion of Animal Waste," Water, Air and Soil Pollution,
3(2), D. Reidel Publishing Company, Dordrecht-Holland, 1974, p. 161-169.
B-64. Abeles, T.P., "Energy and Economical Analysis of Anaerobic Digestors,"
in Energy, Agriculture and Waste Management by W.J. Jewell, An Arbor
Science Publishers, Inc., Ann Arbor, Michigan, 1975.
B-65. Energy Primer: Solar, Water, Wind and Biofuels, Portola Institute
Menlo Park, California, 1974.
B-66. Schulz, H.W., M. Neamatalla, G. long and M. Young, "A Pollution Free
System for the Economic Utilization of Municipal Solid Waste For the
City of New York—Phase I," Columbia University, New York, June 15, 1973.
B-67. Encyclopedia of Chemical Technology, 2nd Ed., Interscience Publishers,
John Wiley & Sons, New York, Vol 8, p. 438; Vol. 1, p. 501-531.
B-68. Process information brochure of Georgia-Pacific, Bellingham Division,
Bellingham, Washington, May 1975.
B-69. Meller, F.H., "Conversion of Organic Solid Waste Into Yeast--An
Economic Evaluation," Public Health Service Publication No. 1909,
U.S. HEW, Bureau of Solid Wastes Management, 1969.
B-70. Brandt, D., L. Hontz, and M. Mandels, "Engineering Aspects of the
Enzymatic Conversion of Waste Cellulose to Glucose," Forest Products .
and the Environment, AIChE Symposium Series, Vol. 69, No. 133,
p. 127-133.
B-71. Allen, A.L., "Economic Evaluation of the Enzymatic Conversion of Waste
Cellulose to Glucose," Presented at the AIChE Symposium "Cellulose--
A Valuable Renewal Resource," Los Angeles, California, November 16-
20, 1975.
B-72. Pearl, I.A., Environmental Science Technology, Vol. 2, 1968, p. 676-681.
B-73. NCRR Bulletin, Vol. V, No. 1, National Center for Resource Recovery Inc.,
Winger 1975.
B-58
-------
B-74. Porteous, A., "Bulk Reduction by Incineration, Hydrolysis and Pyrolysis,"
The Recycling and Disposal of Solid Waste, Proceedings of a Course,
University of Nottingham, April 1-5, 1974, Edited by M.E. Henstock,
Pergamon Press, 1975.
B-75. Duggs, D.J., "Fuel From Organic Matter," The Rand Paper Series P-1500,
Rand Corporation, October 1973.
B-76. Scheller, W.A. and B.J. Mohr, "Nebraska 2 Million Mile Gasohol Road Test
Program," First Progress Report, December 23, 1974 to March 31, 1975,
Department of Chemical Engineering, University of Nebraska,
April 2, 1975.
B-77. Chemical and Engineering News, January 1973.
B-78. "Use of 90% Gasoline/10% Ethanol," Sunday Journal § Star, Lincoln,
Nebraska, June 15, 1975.
B-79. Globus, A.R. , "Utilization of Hydrocarbon—Complex Formation in the
Production of Hydrofuel," United International Research, Inc.,
Hauppauge, New York, Presented at the 1975 National Fuels and Lubricants
Meeting, Houston, Texas, September 11-12, 1975.
B-80 Tatom, J.W., et al, "Clean Fuels from Agricultural and Forestry Wastes —
The Mobile Pyrolysis Concept," Winter Annual ASME Meeting, Houston,
Texas, November 30 - December 4, 1975.
B-81. See Ref. 13 in main body.
B-82 Anderson, J.E., "The Oxygen Refuse Converter-A System for Producing
Fuel Gas, Oil, Molten Metal, and Slag from Refuse," 1975 National
Incinerator Conference, ASME Incinerator Division, Miami, Florida,
May 1974.
B-83. Sussman, D.A., "Baltimore Demonstrates Gas Pyrolysis," U.S. EPA
Report SW-75 d.l, Washington, D.C. 1975.
B-84. Hammond, V.L., "Pyrolysis- Incineration Process for Solid Waste
Disposal," Final Report, Battelle Pacific Northwest Laboratories,
Richland, Washington, December 1972.
B-85. Hammond, V.L. and L.K. Mudge, "Feasibility Study of
Salt Technology for Pyrolysis of Solid Waste " EPA
Battelle Pacific Northwest Laboratories, Richland, Washington,
April 1974.
B-86. "Pyrolysis Study of Fuel from Solid Waste," Preliminary Report,
Barber-Colman, NASA Contract NAS 914350, August 1, 1974.
B-87. "Pyrolysis, Pyrolysis System Evaluation Study," Hamilton Standard
Final Report N75- 18722, December 1974.
B-59
-------
B-88. Davis, John D., "Pyrotek Solid Waste Management and Gasification Systems,"
Sixth Annual Western Regional Solid Waste Symposium, San Jose, California,
March 8, 1974.
B-89. Massey, D., et al, "Urban Refuse Incinerator Design and Operation: State
of the Art," Bureau of Engineering Report BER 133-119, University of
Alabama June 1971.
B-90. Alpert, S.B., et al, "Pyrolysis of Solid Waste: A Technical and
Economic Assessment," Stanford Research Institute, West Virginia
University Report WVU-ENG-CHE-73-01, September 1972.
B-91. "Solid Waste: A New Natural Resource," Department of Chemical
Engineering, West Virginia University, May 1971.
B-92. Bailie, Richard C. and Masaru Ishida, "Gasification of Solid Waste
Materials in Fluidized Beds," AIChE Symposium Series, No. 122, Vol. 68,
1972.
B-93. Halligan, J.E., et al, Conversion of Cattle Feedlot Wastes to Ammonia
Synthesis Gas," Texas Tech Univesity, Report EPA-660/2-74-090,
December 1974.
B-94. Halligan, J.E., K.L. Herzog, and H.W. Parker, "Synthesis Gas from
Bovine Wastes," I§EC Process Design and Development, Vol. 14,
January 1975, 64.
B-95. Halligan, J.E. and W.J. Hoffman, "Potential for Solid Waste Use as
an Energy Source in Texas," Texas Tech University, Final Report,
November 1974.
B-96. Smith. G.L., C.J. Albus, and H. W. Parker, "Products and Operating
Characteristics of the TTU Retort," Texas Tech University, 76th
National AIChE Meeting, Tulsa, Oklahoma, March 10-13, 1974.
B-97. Schlesinger, M. D., W. S. Sanner, and D.E. Wolfson, "Pyrolysis of
Waste Materials from Urban and Rural Sources," Proceedings Third
Mineral Waste Utilization Symposium, Chicago, March 14-16, 1972,
423-28.
B-98. Sanner, W.S., et al, "Conversion of Municipal and Industrial Refuse
into Useful Materials by Pyrolysis," Bureau of Mines Report of Investi-
gations 7428, August 1970.
B-99. Corey, Richard C., "Pyrolysis, Hydrogenation and Incineration of
Municipal Refuse—A Progress Report," Proceedings of the Second
Mineral Waste Utilization Symposium, Chicago, Illinois,
March 18-19, 1970, 299-312.
B-60
-------
B-100. Benham, C.B. and J. Diebold, "Conversion of Solid Waste to Fuels "
Naval Weapons Center Report TP 5797, China Lake, California, January
1976.
B-101. Appell, Herbert R. and R.D. Miller, "Fuel from Agricultural Wastes,"
Chapter 8 of "Symposium: Processing Agriculture and Municipal Wastes,"
The Avi Publishing Co., Westport, Connecticut, 1973.
B-102. Appell, H.R., et al, "Conversion of Cellulosic Wastes to Oil," Bureau
of Mines Report of Investigations 8013, 1975.
B-103. Friedman, Sam, et al, "Continuous Processing of Urban Refuse to Oil
Using Carbon Monoxide," Bureau of Mines, 3rd Mineral Waste Utilization
Symposium, Chicago, Illinois, March 14-16, 1972.
B-104. U.S. Patent 3,733,187, "Process for Converting Solid Wastes to Pipeline
Gas," H.F. Feldmann, Assignor to United States of America, May 15, 1973.
B-105. Feldmann, Herman F., "Pipeline Gas From Solid Wastes," AIChE Series,
Vol. 68, No. 122, 1972, 125-31.
B-106. Ilering, R., and S.A. Greeley, Collection and Disposal of Municipal
Refuse, McGraw-Hill, New York, 1921.
B-107. Corey, R.C. (ed.), Principles and Practices of Incineration, Wiley-
Interscience, New York, 1969-
B-108. Roberts, R.M. et al, "Systems Evaluation of Refuse as a Low Sulfur
Fuel," The Envirogenics Co., Contract CPA-22-69-22, November 1971.
B-109. Pyrolytic Incineration Systems, Brochure of the Kelley Company, Inc.,
Milwaukee, Wisconsin.
B-110. "Pyrolytic Heat Recovery," Design News, November 8, 1976, 46-52.
B-lll. "Evaluation of Small Modular Incinerators in Municipal Plants" Ross
Hofmann, Associates, EPA Contract 36-01-3171, 1976.
B-112. "Consumat, An Alternate Energy Source," Sales brochure No. 10-874,
Consumat Systems, Inc., Richmond, Virginia; brief description of incin-
erator given in "A Total Package Concept for Solid Waste Management,"
Public Works, April 1975.
B-113. Thurnau, R.C. and D.A. Oberacker, "High-Temperature Vortex Incinerator,"
U.S. EPA, Report No. EPA-670/2-75-025, February 1975.
B-114. "The Vorcinerator Waste Disposal System," General Electric Co. Brochure,
Shelbyville, Indiana, no date.
B-115. Sales brochure, Adams Creative Environmental Systems, Red Lion,
Pennsylvania, 1975.
B-61
-------
B-116. Moats, E.R., "Operating Experience with a Tire-Fired Boiler," American
Society of Mechanical Engineers Industrial Power Conference, Pittsburgh,
Pennsylvania, May 19-20, 1975, paper no. 75-IPWR-10.
B-117. Flynn, B.L., "Increasing the Thermodynamic Efficiency of Black Liquor
Recovery Operations by the Use of Wet Air Oxidation," Zimpro, Inc.,
Rothschild, Wisconsin, no date.
B-118. "Destruction of Hazardous Wastes by the Molten Salt Process," Brochure
of the Atomics International Division of Rockwell International, Canoga
Park, California.
B-62
-------
APPENDIX C
ALTERNATIVE USES OF PYROLYTICALLY-FORMED SYNGAS
INTRODUCTION
Conversion of solid waste to a syngas has been treated in Section 8.
The base case analyzed was a front end and gasifier to process nominally
1361 Mg/d (1500 TPD) of raw refuse in the Union Carbide Corporation (UCC)
Purox system. Steel and aluminum were recovered, the syngas dried and com-
pressed, and then delivered to a power utility or industrial customer by
short pipeline. Construction and operating costs were developed along with
an economic analysis to show the unit cost of the syngas as a fuel.
In that the discussion within the main body of the text was to be
limited to a specific conversion system, the analysis was taken only through
the step of conditioning the Purox gas. Because this gas has a substantial
component of carbon monoxide and hydrogen with a negligible quantity of
nitrogen, it possesses value as a feedstock for further conversion to other
fuels or chemicals, in addition to being able to be used directly to produce
industrial heat or electric power. It serves, therefore, as an excellent
model for the analysis of possible costs for these alternative uses. Rather
than developing an entirely separate case with new assumptions on composition
and initial costs, it is convenient to directly utilize the information
presented in Section 8 as a starting point. The discussion presented within
this Appendix is entirely the responsibility of Parsons and in no way is it
implied that the Union Carbide Corporation necessarily concurs with the
analysis or that UCC proposes to supply total systems described here.
The several alternative processes are shown schematically as follows:
Pyrolytic
Syngas .
Dry Syngas
•Synthetic Natural Gas (SNG)
Electric Power
•Fuel Grade Methanol
Anhydrous Ammonia
C-l
-------
Ammonia is not produced as a fuel but almost all of it is made from fossil
fuel, primarily natural gas. Its character as an energy form is by direct
relation.
The Purox gas contains about 10% by volume of hydrocarbons. If the gas
is to be converted to methane, methanol, or ammonia, provisions must be made
to process the hydrocarbons by converting them to an appropriate mixture of CO
and H2 (pure H2 for ammonia). Furthermore, the gas has trace contaminants
that can affect catalysts and therefore needs careful experimental evaluation,
which should include a pilot plant test of the gas treating and catalytic
conversion units.
Practically all methanol or ammonia plants that have been constructed use
natural gas and to a smaller extent naphtha as a feedstock. Industrial
partial-oxidation units are now used to convert oil or coal to a syngas
(Koppers-Totsek, Texaco, Shell). The high temperatures of 1539 to 1649°C
(2800 to 3000°F) result in a gas with little or no hydrocarbon content. In
the Purox reactor, the solid waste reduction to gas and char by pyrolysis
takes place at temperatures of 538 to 760°C (1000 to 1400°F) and upwards of
10% to 11% by volume of hydrocarbons are formed. These can be reformed,
otherwise they are used as fuel and their value as a source of H2 and CO is
lost. The Lurgi process for coal produces a similar gas, but to this date
no ammonia synthesis plant has been constructed with the Lurgi gas because
of problems in gas conditioning. Producing methane or methanol will be less
difficult than producing ammonia.
Gas treating expertise and experience will be required in using pyroly-
tic gas for chemical synthesis processes. In particular a pilot gas treatment
and conditioning unit with representative catalysts must be tested at the
Purox demonstration site, otherwise the risks are high in building an expen-
sive commercial chemical synthesis unit.
The information here should only be used as a guide for determining
whether alternative end products are of greater value than the syngas or even
in comparison with different solid waste disposal and processing systems.
Such work has been accomplished specifically in two engineering and economic
feasibility studies in the San Francisco and Denver areas by Parsons
(Ref. C-l, C-2).
METHANE CONVERSION PROCESS DESCRIPTION
The syngas produced by the Purox process must be upgraded to meet
current pipeline gas specifications. Eighty-eight percent by volume of the
syngas is composed of a mixture of CO, C02, and H2, plus 11% hydrocarbons,
with an HHV of 14.57 MJ/Nm3 (370 Btu/SCF). Pipeline gas, which is almost all
methane, has an HHV 2.7 times higher. Methanation has been in existence for
some time to reduce small amounts of CO in a hydrogen rich gas. Conversion
of a syngas to pipeline quality gas by bulk methanation has not been applied
commercially. However, presently occurring changes in the economics of the
natural gas industry has generated great interest in syngas methanation. An
applicable new process, known as the RMProcess, is discussed below.
C-2
-------
A pilot methanation plant has been tested by The Ralph M. Parsons
Company .using gas from a Texaco partial oxidation unit in Montebello
California; oil resids or coal are used as feedstock. This process is a new
approach to catalytic conversion of synthesis gas to methane. Appreciably
reduced capital and operating costs for large-scale operations are projected
as compared to existing schemes because a large recycle gas stream for temper-
ature control purposes is not used and water-gas shift occurs in the same
catalytic bed.
Flow diagram of the process is presented in Figure C-l. Table C-l shows
information regarding some of the several streams in the methanator system.
The feed gas is stream 9 from Figure 35. The syngas is received at approxi-
mately 310 kPa (45 psia) and is compressed to 931 kPa (135 psi).
Condensed water and hydrocarbons are separated from the cooled gas,
which is then passed to an absorption unit to remove acid gases. Hydrocarbons
collected are used as fuel. The concentrate from the absorber unit is sent
to a Stretford process unit where free sulfur is recovered, while releasing
the C02 to the atmosphere. Sulfur is kept to less than 5 ppm in the feed gas
to the methanator reactors.
The cleaned syngas, together with steam, then passes through the bulk
methanator (Ref. C-3), which consists of a series of fixed-bed, adiabatic,
catalytic reactors, as shown schematically in Figure C-2. Both the water-gas
shift and methanation reactions take place simultaneously in the catalyst
beds. Also, remaining light hydrocarbons with molecular weights greater than
CHtt are steam reformed and converted to methane in the same beds. Some
reactions that take place are:
CO + H20 -> H2 + C02 (shift reaction)
C2H5 + 2H20 -> 2CO + 5H2 (reforming)
CO + 3H2 -> CH^ + H20 (methanation)
Because the methanating reactions are exothermic, the gas temperatures
reach 538 to 649°C (1000 to 1200°F). The presence of excess steam in the
catalyst bed prevents carbon deposition on the catalyst at these high tem-
peratures. Heat is removed by generation of high pressure steam in conven-
tional heat exchange equipment. This stream is used in turbines to drive
compressors. The residual low-pressure exhaust steam at both 827 and
414 kPa (120 and 60 psia) is returned to the process to provide water-vapor
for the water gas shift conversion phase of the reaction and for reboiler use
in absorption liquid regeneration. After having passed through the bulk
methanators, a second absorber unit removes C02 formed in the methanator
train, and the gas stream passes through a final stage of dry methanation.
After dehydration, it emerges as pipeline-quality gas ready for injection into
a utility pipeline. Pressures do not exceed 1207 kPa (175 psia) in the
process.
The process is capable of performing satisfactorily with a wide range of
syngas composition. For methane production, the feedstock ideally should
C-3
-------
n
©
SCRUBBER
(COJ
150 psia
55D°F"
2nd
ABSORBER
t®
REBOILER
A
~^V-
\
/
V.
/
\
S
^
U
,
FINAL METHANATOR
I DEHYDRATION UNIT |
Figure C-l. Methanation .process for syngas.
-------
TABLE C-1A. METHANATOR FLOW STREAMS*
(SI UNITS)
Stream
number
1
Stream
Purox Gas
Quantity
753 x 103 Nm3/d
Other information
Saturated with water
10
Purox Gas to
Steam Generator
Purox Gas to be
Processed
Packaged Boiler
Steam
Reboiler Steam,
1st Absorber
C02 § H2S to
Stretford Unit
Steam Generation/
methanator
train
Reboiler Steam,
2nd Absorber
C02 from
2nd Absorber
Product Gas, SNG
14.57 MJ/Nm3
10.94 TJ/d
133 x 103 Nm3/d
620 x 103 Nm3/d
21 764 Mg/h
10 873 kg/h
C02 - 11 698 kg/h
H2S - 14 kg/h
21 247 kg/h
19 632 kg/h
21 092 kg/h
220 x 103 Nm3/d
36.51 MJ/Nm3
8.02 TJ/d
vapor at 38°C
Saturated with water
vapor at 38°C
6 894 kPa, 482°C
Slowdown
2 268 kg/h
Exhaust steam
44 kPa
Stretford process
produces pure
sulfur
6 894 kPa, 482°C
448 kPa
- 91.2% by vol.
H2 - 1.2%
CO - 400 ppm
C02 - 4.0%
N2 + Ar - 3.6%
*Based on calculations at Parsons for nominal raw refuse capacity of
1360 Mg/d.
C-5
-------
TABLE C-1B. METHANATOR FLOW STREAMS*
(ENGLISH UNITS)
Stream
number
1
Stream
Purox Gas
Quantity
28.03 x 105 dry scf/d
Other information
Saturated with water
Purox Gas to
Steam Generator
Purox Gas to be
processed
Packaged Boiler
Steam
Reboiler Steam,
1st Absorber
C02 § H2S to
Stretford Unit
Steam Generation/
methanator
train
370 Btu/scf
10.37 x 109 Btu/d
4.95 x 106 dry scf/d
23.08 x 106 dry scf/d
47,980 Ib/hr
23,970 Ib/hr
C02 - 25,770 Ib/hr
H2S - 30 Ib/hr
46,840 Ib/hr
vapor at 100°F
Saturated with water
vapor at 100°F
1000 psia, 900°F
Blowdown
5000 Ib/hr
Exhaust steam
65 psia
Stretford process
produces pure
sulfur
1000 psia, 900°F
Reboiler Steam,
2nd Absorber
43,280 Ib/hr
65 psia
10
C02 from
2nd Absorber
Product Gas, SNG
46,500 Ib/hr
8.2 x 106 scf/d
927 Btu/scf
7.60 x 109 Btu/day
___T - 91.2% by vol.
H2 - 1.2%
CO - 400 ppm
C02 - 4.0%
N2 + A - 3.6%
*Based on calculations at Parsons for nominal raw refuse capacity of
1500 TPD.
C-6
-------
DESULFURIZED
SYNGAS
O T
STEAM IT
o-y
i FE.ED '
n
n
Tt
T.
n
TO
C02REMOVAL
Figure C-2. Process schematic for bulk methanation.
-------
contain six times as many hydrogen atoms in the form of hydrogen gas or water
as there are carbon atoms in the form of carbon monoxide. The model syngas is
relatively low in hydrogen and high in carbon monoxide, which means that a
large amount of process steam must be used. Further, the relatively large
carbon dioxide component in the feed gas requires proportionate amounts of
steam for absorption-liquid regeneration. Certain other petroleum and coal
partial-oxidation processes producing hot gases, in addition to the methana-
tion process, normally yield enough steam for both process requirements and
for driving the compression equipment with steam turbines. With cold gases
from a pyrolysis system, however, a portion of the syngas is required to
generate steam. The amount of process and stripping steam required dictates
the rate of steam production. High pressure steam is produced at 6 894 kPa
(1000 psia) and 482°C (900°F), then expanded to process pressure levels
through turbines to provide the power requirements of the system. The
quantity of syngas diverted to steam generation is 18% of the Purox system
production. More high pressure steam is produced than is usable for driving
compressors. Therefore, a turbo-electric generator is used to generate
electric power and backpressure steam is used for the reboilers.
The SNG produced has the following composition by volume: CHi+, 91.2%;
H2, 1.2%; CO, 400 ppm; C02, 4.0%; N2 + Ar, 3.6%. Moisture is removed so that
the gas has a dew point of -40°C (-40°F). The HHV of the SNG is 36.51 MJ/Nm3
(927 Btu/SCF) and is produced at the rate of 220 x 103 Nm3/d (8.2 x 106 SCF
day) or 8.02 TJ/d (7.60 x 109 Btu/day).
Because a commercial-scale plant using this process has not been built,
accurate detailed costs are not available. Estimates were made based on pre-
liminary design studies. No commercial scale methanator system of any kind
has been built for producing pipeline gas from a synthesis gas.
In generating steam for use in the process, that steam used in a turbo-
electric generator produces 1700 kW. The following is a listing of steam
turbine drives:
Steam Rate
Purox gas blower
Feed Compressor
Product Compressor
Turbo-Electric
Generator
kg/h
7 022
16 774
3 856
Ib/hr
15,480
36,980
8,500
15 359
43 010
- 1022 kW (1370 HP) (Drive)
- 2283 kW (3060 HP) (Drive)
- 574 kW (770 HP) (Drive)
- 2110 kW (produced)
Various electric power drives for pumps and other devices require 330 kW.
Boiler makeup amounts to 2087 kg/h (4,600 Ib/hr). Cooling tower water flow
is 20 m3/min (5300 gpm) and its makeup is 3.8 percent.
C-8
-------
A daily energy balance on the methanator process shows:
Energy in gas feedstock = 10.94 TJ (10.37 x 109 Btu)
Energy in product = 8.02 TJ (7.60 x 109 Btu)
Equivalent heat in electric
power generated = 0.54 TJ (0.51 x 109 Btu))
. , . > net=0.43x!09 Btu
Electric drives = 0.08 TJ (0.08 x 109 Btu))
The conversion efficiency for the methanator is
x 100 = 73.3%
Similarly, for the syngas plant plus methanator it is
l£|g (100) = 54.7%
The overall plant net thermal efficiency, accounting for energy required to
operate the plant, is determined to be
7.6 + 0.51 - 0.08 - 0.25 - 2.82 __ _„
- = 60, I -a
13.89
METHANE CONVERSION CONSTRUCTION AND OPERATING COSTS
Addition of a methanator system to the gasifier plant increases the
complexity. Methanators have not been operated in commercial sizes with
Purox-like gases, which leads to a lower probable utilization factor. Over
92% utilization factor on a yearly basis was used for producing the Purox gas
because of redundancy in equipment. With addition of a methanator system, a
value of 85% is used for the entire plant. This results in 422 Gg/y
(465,000 TPY) of raw refuse that can be processed and 68.0 x 106 Nm3/y
(2.54 x 109 SCFY) of product methane.
Capital Cost
Capital cost for a complete solid waste-to-methane facility would start
with the syngas production facility (front-end plus gasifier) described in
the Purox Section. Added to this is the methanator cost shown in Table C-2.
The methanator should be located adjacent to the Purox plant with a
pipeline required for connection to the nearest natural gas main pipeline.
Land required for the methanator is approximately 20 200 to 24 300 m
(5 to 6 acres). Electric power required is 330 kW, but 1700 kW will be
generated by the excess high quality steam in the plant. Compressors will be
steam driven from plant steam generators. Water supply will be approximately
1500 dm3/min (400 gpm).
C-9
-------
TABLE C-2. ESTIMATED CAPITAL COST REQUIREMENTS FOR METHANE
(1975 DOLLARS)
Cost
Item ($ millions)
Construction
Front end and syngas plant 53.75
Methanation Plant 15.00
Interest During Construction (8-1/2% one year) 5.84
Startup Costs 3.27
Working Capital (25% of annual operating cost) 2.07
Total 79.93
The construction cost is estimated at $15 million for the methanator,
which is considerably less than either the ammonia or methanol plant because
of much lower operating pressures and less gas treatment. The total plant
capital cost given in Table C-2 is $79.93 million with a construction cost
of $68.75 million.
Operating Costs
Operating costs are shown in Table C-3. Labor costs includes operating
labor only, with maintenance including the maintenance labor. A credit is
taken for electric power because of the large amount of steam generated for
reboiler and process use. Not all the steam could be used in drive turbines
for compressors; therefore, a steam-electric generator unit was included to
generate 2110 kW, of which 350 kW was used for various small electric motors,
and the remainder is fed to the Front End/Purox systems. The amount in
excess is 1760 kW and over a year this amounts to 13.1 x 106 kWh, or
$328,000 based on 25 mills/kWh. The same price for electrical power was used
in the syngas plant and therefore the effect is the same as reducing the
amount of power purchased from a utility.
The operating cost per ton of raw refuse, based on a yearly operating
schedule, is $19.55/Mg ($17.74/ton) of raw refuse or $3.30/GJ ($3.49/million
Btu).
Economic Analysis
The effect on net SNG costs of receiving varying drop charges is shown
in the table below. In each case it is assumed that revenues of $331/Mg
($300/ton) of aluminum and $44/Mg ($40/ton) of steel are received. Unit
C-10
-------
TABLE C-3. ANNUAL OPERATING COSTS FOR METHANE
(1975 DOLLARS)
Cost/year
Item ($000)
Methane Unit
Labor 350
Power (328)*
Maintenance 750
Production Materials 320
Water 145
Total (Methane unit only) 1237
Front End § Gasifier (Table 43) 7016
Total Plant 8253
*The methanator has excess power, but the credit taken has been charged to the
Front End and Gasifier and is included in the $7,016,000 shown.
capital costs for the $79.93 million facility are $20.00/Mg ($18.15/ton) of
raw refuse or $3.39/GJ ($3.58/106 Btu) of gas produced. Unit operating
costs are $19.55/Mg ($17.74/ton) of refuse and $3.32/GJ ($3.50/106 Btu).
NET COSTS TO PRODUCE SNG
Raw Refuse Basis Energy Basis
Drop Charge Per Mg Per Ton Per GJ Per 106 Btu
0 $34.66 $31.44 $5.88 $6.20
$ 5.51/Mg ($5.00/ton) 29.14 26.44 4.95 5.22
$11.02/Mg ($10.00/ton) 23.63 21.44 4.01 4.23
It is roughly estimated that SNG produced from coal or LNG transported
from a proposed plant in South Alaska, Indonesia, or North Africa will sell
for $3 to $4 per million Btu or higher in 1975 dollars. This indicates
that SNG manufactured from solid waste could be competitive with other new
sources of SNG or natural gas provided a sufficiently high drop charge is
available.
C-ll
-------
ELECTRIC POWER FROM GAS TURBINE COMBINED CYCLE PROCESS DESCRIPTION
In the combined cycle, use is made of a gas turbine operating in series
with a waste-heat boiler and a steam turbine. Hot exhaust gases from the gas
turbine are a means of producing steam from a waste heat boiler. Previous
studies (Ref. C-l) showed that other cycle schemes with gas turbines would
result in a lower capital cost, but the combined cycle is thermodynamically
the most efficient and results in the least cost per kWh of net electric
power produced.
Data were developed using specifications for two actual gas turbine/
generator assemblies; both units can operate with either gas or liquid fuel.
The gases include syngas or natural gas, while the liquids include petroleum
distillates or methanol. Because of the low heating value per unit volume of
the syngas, the units must be started and brought to operating temperatures
on diesel fuel oil. The unit can be set up to burn gas and oil simultaneously
with the oil automatically increasing to maintain full generator output if the
gas supply varies, is inadequate, or stops, and a power network emergency
arises. In this way, the units can be kept operating at full capacity. Since
startup oil facilities are required in any case, this capability increases
the cost of the installation only by the amount of additional fuel oil
storage required to support continuous oil operation.
A layout of a combined, cycle plant is shown in Figure C-3. Fuel gas must
be compressed to about 1724 kPa (250 psi) to feed the two gas turbines, con-
suming about 5 kW of power when one module is operating at full capacity. For
simplicity, an electric-driven centrifugal compressor is used to pressurize
the feed gas. The heat rate for this combined cycle is approximately 12.55 MJ
(11,900 Btu)/kWh. The net nominal capacity of a module is 29 MW. Net pro-
duction of power to the bus was found to be 36.3 MW for a 1361 Mg/d (1,500 TPD)
plant and subtracting the power to drive the syngas plant (11.2 MW) leaves
25.1 MW to a customer.
ELECTRIC POWER FROM GAS TURBINE COMBINED CYCLE CONSTRUCTION AND OPERATING
COSTS
Other equipment arrangements than given here using a gas turbine have
been studied with respect to capital and operating costs and cost per unit
output. Cases that were studied (Ref. C-l) included combined cycle, gas
turbine only, gas turbine plus steam generation, and steam generation only.
Capital costs for the combined cycle were highest but, because it was most
thermodynamically efficient, it produced electric power at the least cost
per kWh even where other systems had a credit for steam sales. The combined
cycle was chosen for further analysis and the costs are given in the following
text.
This technology is well-known with standard equipment. Costs are based
on passing 751 x 103 m3/d (28.03 x 106 SCFD) of syngas to the combined cycle
plant. It is estimated that there will be two weeks of scheduled downtime
at the same time as that with the syngas system plus one week of unscheduled
shutdown in addition to the two weeks for the syngas system. Thus 47 of the
C-12
-------
CD BOILER TUBE REMOVAL AREA
(2) MOTOR CONTROL CENTER
BOILER AUXILIARIES
LOW-PRESSURE DRUM
© EXHAUST STACK
(G) WASTE HEAT BOILER
HIGH-PRESSURE DRUM
INLET
COMPENSATE STORAGE
WATER TREATMENT
CONTROL ROOM
GAS TURBINE
GENERATOR
EXCITER
SWITCHGEAR
COMPRESSOR HOUSE
COOLING TOWER
CONDENSATE
MAIN STEAM
MAINTENANCE EQUIPMENT
& LAYDOWN AREA
PLAN
Figure C-3. Layout of combined cycle electric power plant.
-------
52 weeks per year are available for operation, resulting in processing
477 703 Mg/y (493,500 TPY) of raw refuse, which corresponds to producing
a net electric power output of 200 x 106 kWh/y. The net electric power
accounts for gas compression and all power needs in the front end and gasifier
units; the latter two units require 88.3 x 106 kWh/y. The electric energy
required for the front end and gasifier units is 197.3 kWh/Mg (179 kWh/ton).
Capital Cost
The essential equipment is listed in Figure C-3. An electric drive is
recommended to compress the pyrolysis gas mainly because of startup require-
ments. With the syngas discharged at atmospheric pressure, it would be best
to locate the compressor at the gas plant and transport the gas by a small
diameter pipe to the turbines. The land required is approximately 12 000 m2
(3 acres) and the plant produces its own electric power for the compressor
and other parts of the plant. There will be need for a utility power line to
the plant for startup purposes and other needs.
Construction cost of the electric power plant is estimated at
$16.21 x 106; adding this to the construction cost of the front end and Purox
systems results in a total of $69.96 x 106. The total capital cost of
$79.96 x 106 shown in Table C-4 includes interest during construction
(equivalent to one year's interest on the construction cost at 8-1/2%),
startup costs, and working capital.
Operating Costs
Operating costs for the electric power unit are shown in Table C-5 as
$1,236,000 per year. No power cost appears as with other plants because
electric motors, lights, instruments and controls, etc., are supplied with
electricity from the generator and accounted for in the net output of
25.1 MW (11.1 MW are used in the front end and gasifier). Operating costs
from the front end and Purox systems of $7,016,000/year are shown in
Section 8. Included in this cost is $2,307,000 for power, which is deleted
as explained. A net operating cost for the front end and gasifier is
$4,709,000 instead of $7,016,000. The total operating cost is then
$5,945,000/yr. Unit operating costs for the entire plant (raw refuse to
electric power) is $13.27/Mg ($12.04/ton) raw refuse or 29.7 mills/kWh.
ECONOMIC ANALYSIS FOR ELECTRIC POWER
With a capital cost of $79.96 million, the unit amortization costs for
the electric power case is $18.87/Mg ($17.12/ton) refuse or 42.2 mills/kWh.
With an operating cost of $13.27/Mg ($12.04/ton) or 29.7 mills/kWh, the total
unit cost is thus $32.14/Mg ($29.16/ton) of raw refuse or 71.9 mills/kWh to
produce a net output of 200 x 106 kWh/y.
Revenues from aluminum and steel recovery from 447 703 Mg/y (493,500 TPY)
raw refuse is $4.90/Mg ($4.45/ton) raw refuse or 11 mills/kWh. Revenues from
$11.02/Mg ($10/ton) or $5.51/Mg ($5/ton) drop charge on the raw refuse results
in an equivalent unit revenue of 24.7 mills/kWh or 12.3 mills/kWh,
respectively.
C-14
-------
TABLE C-4. ESTIMATED CAPITAL COST FOR ELECTRIC POWER
Cost
Item ($ million)
Construction
Front-End Processing and Syngas Plant 53.75
Gas Turbine - Combined Cycle Plant 16.21
Interest During Construction (8-1/2% one year) 5.95
Startup Costs 2.56
Working Capital 1.49
Total Plant 79.96
TABLE C-5. ESTIMATED OPERATING COST FOR ELECTRIC POWER
Cost
Item ($000)
Electric Power:
Labor 499
Power
Maintenance 457
Production Materials 1^8
Water 142
Total Electric Power 1,236
Front-End and Purox System
(Table 43 less electric power) _
Total Plant 5'945
C-15
-------
The net cost is the total cost less drop charges and metal sales given
above. For the highest drop charge case, the net cost is $16.21/Mg
($14.71/ton) raw refuse and for the $5/ton drop charge case, the net cost
is $21.73/Mg ($19.71/ton) raw refuse. For zero drop charge, the net cost
is $27.24/Mg ($24.71/ton). The corresponding costs per unit net electrical
power output is 36.2 mills/kWh, 48.6 mills/kWh, and 60.9 mills/kWh,
respectively. It should be noted that the price of electric power used in
this report is 25 mills/kWh, less than the cost of producing it with the
system evaluated here. In many regions of the U.S. the present cost of
electric power is greater than 30 mills/kWh. With drop charges greater than
$10/ton, this type of plant would be economically acceptable in those regions.
CONVERSION TO FUEL GRADE METHANOL PROCESS DESCRIPTION
The purpose of utilizing pyrolysis gas for synthesis of a fuel grade
methanol is to produce a storable and more transportable fuel. One possible
application is peaking power gas turbine-electric power generators used by
utilities (Ref. C-4). Replacement of petroleum distillate fuel by methanol
requires careful economic analysis. Another application is for internal
combustion engine use. Although isolated tests have shown some potential,
it is not clear that widespread practical application would be possible.
Fuel grade methanol is quite miscible with gasoline because of impurities
present (Ref. C-5).
The CO and H2 in the model syngas fin the ratio of 5:3) can be converted
to methanol by first using a water-gas shift reaction in a catalyst bed so
that the mole ratio of CO to H2 in the gas is slightly less than 1:2, and
then passing this mixture at high pressure through another catalyst bed. The
basic reactions are:
CO + H20 -* H2 + C02
CO + 2H2 -> CH3OH
Also, after scrubbing, C02 is present in a 1:20 ratio to CO and will react
with H2 as follows:
C02 + 3H2 -> CH3OH + H20
Figure C-4 shows a typical process schematic. First, both H2S and C02
are removed from the syngas. After stripping from the absorbent, the H2S is
converted to sulfur by means of the Glaus Process. An alternative method for
sulfur recovery is the Stretford process. The desulfurized gas is then passed
into a shift converter with steam to adjust the CO/H2 mole ratio. The C02
in the gas after shift converter action is removed by absorption and stripping
towers. Water vapor is removed in a condenser before the CO/H2 gas mixture is
passed in the compressor. The pressure is then raised to either the low range
of 5.1 MPa (50 atmospheres] or the high range of 10.2 to 15.3 MPa (100 to
150 atmospheres), depending on the catalyst chosen. The higher pressure
produces a higher yield, but economics will dictate the final choice.
Methanol synthesis takes place on a zinc/chromium oxide catalyst or a
C-16
-------
n
ELECTRIC POWER
SSOR1-
-1
-J
H
r
^
^^-X»^
cc
i cc
>
CONDE
FUEL
GASES
WASTE
WATER
FUEL GASES TO
(STEAM
^GENERATOR
L
METHANOL
PURIFICATION
WATER
AND
HEAVY
ENDS
CRUDE FUEL
COMPRESSORS
PURIFICATION
Figure C-4. Synthesis process for methanol.
-------
copper/zinc/chromium type of catalyst. The zinc/chromium oxide catalyst used
in the high-pressure process is less susceptible to sulfur poisoning and less
costly. A zinc oxide bed can be used to remove traces of sulfur gases to pro-
tect the shift and methanol synthesis catalysts.
The synthesized gas is then passed through a regenerative heat exchanger
to heat the incoming catalytic reactor feed gas and then cooled to condense
the methanol and other hydrocarbon components. Uncondensed gases, particu-
larly N2, Ar, and CH^, with a small amount of CO and C02, are sent to a steam
generator furnace. At this point, the product is crude methanol containing
higher boiling oxygenated hydrocarbons, condensible hydrocarbons, and water.
One licensor of a methanol synthesis system states that sufficient steam is
generated within the process for internal use requirements. However, this is
questionable, because of the amount of steam required to strip the large
quantity of C02 out of the scrubber liquids. More C02 is generated in the
solid waste pyrolysis process than in current commercial petroleum partial-
oxidation processes. Also, the H2/CO mole ratio in the Purox off-gas is
smaller than that from petroleum partial-oxidation conversion systems,
resulting in a larger amount of C02 formed in the water-gas shift reaction.
Upon leaving the condenser after the synthesis reactor, the crude
methanol is approximately 85% methanol. A dewatering column removes water and
heavy ends, producing a fuel-grade methanol with approximately 92% methanol,
5% hydrocarbons, and 3% higher alcohols and esters with a density of
0.80 g/cm3 (6.67 Ib/gal).
Approximately 11 100 kW of electric power are required for the methanol
unit. A detailed design of a unit using Purox-type gas would be needed to
determine whether some of the driving power can be furnished with steam
generated for process use. If steam is to be used for the compressors, then
a steam generator source would be needed for startup, adding an additional
expense. Until this analysis is done, a conservative value for power required
is used.
Production rate for the fuel-grade methanol is 222 Mg/d or 278 m3/d
(245 TPD or 73,500 GPD) for a 1361 Mg/d (1,500 TPD) of refuse plant. The HHV
is 19.93 MJ/dm3 (71,500 Btu/gal).
CONVERSION TO FUEL GRADE METHANOL CONSTRUCTION AND OPERATING COSTS
Technology is available for synthesis of methanol, but because methanol
plants start with natural gas or naphtha as feedstocks, which have been
thoroughly decontaminated, present methanol process plants are not designed
to operate with the gas from Purox-type systems. Certain impurities in the
gas could contaminate catalysts, and the large amount of hydrocarbons may
require consideration of a type of reforming unit. Parsons has determined
on a preliminary basis that the hydrocarbons present will pass through the
system without causing difficulties and appear in solution with the methanol
as a fuel grade liquid (92% methanol). Presence of hydrocarbons is ordinarily
undesirable in the purification process to make industrial grade methanol
(99.85% pure), but this grade is for use in chemical synthesis or as a solvent
and not for fuel, and therefore is of no interest here. An in-depth process
C-18
-------
engineering and facilities study is required to determine more accurately
the details of the plant and associated costs. Presented are costs based on
past experience in methanol manufacture.
The costs are based on production from a 1361 Mg/d (1,500 TPD) plant
With a methanol plant added to the syngas plant, the utilization factor is
assumed to decrease from the 92% to approximately 88%, resulting in processing
482,000 TPY of raw refuse and producing 78,700 TPY (23.83 x 106 GPY) of fuel
grade methanol.
Capital Cost
Capital cost for a complete solid waste to methanol facility would start
with the front end and Purox facility described in Section 8; this is added to
the methanol facility cost as shown in Table C-6.
The methanol plant should be adjacent to the syngas facility. Land
required for the methanol unit is 20 200 to 24 300 m2 (5 to 6 acres), which
will include storage of methanol. This tank could be a small surge type with
a pipeline to a customer not too distant from the plant. As mentioned pre-
viously, a likely customer is a utility with peaking power gas turbines using
distillate fuel. In the TVA power system, as an example, peaking turbines
operate one to three hours per day. Storing methanol at the turbine site and
using a 24-hour production in two hours of combustion, a gas turbine-electric
power generation plant with 220 MW capacity can be operated intermittently
in a peaking power schedule.
The construction cost for the methanol plant in 1975 dollars is approxi-
mately $26.12 million. Total capital costs required are given in Table C-6.
Operating Costs
Operating costs are shown in Table C-7 with details for the methanol unit
and addition of the operating cost for the front-end and Purox units. Labor
includes only operating labor, and maintenance labor is included in the
Maintenance item. There are some disagreement on the amount of outside power
required from different licensors of the process. To be conservative, a power
level of 11 000 kW was used in the methanol unit. Water costs include
effluent discharge fees. The total operating cost for the plant (from raw
refuse to fuel grade methanol) is $11,372,000 per year. This results in a
unit cost of $26.01/Mg ($23.60/ton) raw refuse, or $159.28/Mg ($144.50/ton)
of fuel grade methanol (47.7{/gal, $6.67/106 Btu).
ECONOMIC ANALYSIS FOR METHANOL PLANT
The methodology for arriving at the probable range of net costs for a
syngas-derived fuel has been sufficiently developed previously that details
for the methanol case need not be cited here. The concluding summary
Table C-10 contains the results of the calculations made. For the standa^
case of the $11.02/Mg ($10/ton) drop charge, the net cost would be $0.15/dm
or $7.69/GJ ($0.58/gallon or $8.11/106 Btu). It must be concluded, if
C-19
-------
TABLE C-6. ESTIMATED CAPITAL COST FOR METHANOL
Item
Cost
($ million)
Construction
Front End Processing § Syngas Plant
Methanol Conversion Plant
Interest During Construction
Startup Costs
Working Capital
Total
53.75
26.12
2.22
4.21
2.84
89.14
TABLE C-7. ANNUAL OPERATING COSTS FOR METHANOL
Item
Cost
$(000)
Methanol Unit
Labor
Power
Maintenance
Production Materials
Water
Total, Methanol Unit
Front End £ Purox System
Total Plant
400
2,120
1,130
506
200
4,356
7,016
11,372
C-20
-------
economics is the sole criterion, that fuel grade methanol costs appear too
high for most regions to consider.
If there is a desire to use this grade methanol for automobiles its cost
must be compared with gasoline costs. For a retail price of gasoline at
$0.60/gal, the equivalent price is approximately $5.20/106 Btu as compared to
the net cost of fuel grade methanol cited above of $8.11/106 Btu. If a manu-
facturer's profit and distribution costs are included, the fuel grade meth-
anol 's retail price will be quite high and would not be competitive with
present retail gasoline prices even with a higher drop charge.
Other factors are needed to make a final decision on the use of methanol,
such as cost of changes to make methanol compatible with present automobile
engines, reduction in emission controls, costs needed for separate storage at
service stations, and the cost of preventing injury due to the high toxicity
of methanol vapors.
CONVERSION TO AMMONIA PROCESS DESCRIPTION
Conversion of syngas to ammonia has been seriously considered in several
studies because of the high monetary value of anhydrous ammonia primarily used
as a fertilizer or as a fertilizer base. High prices for ammonia, along with
a high refuse drop charge ($12/ton refuse) has led Seattle to consider the
building of a solid waste-to-ammonia system as possibly economically feasible.
For cities with drop-charges of $2 to $4 per ton, the revenues are insuffi-
cient to allow selling of ammonia at the market price. The market price of
ammonia has recently varied from $100 to $180/ton. This is bulk sale price to
a distributor.
Ammonia synthesis requires quite pure nitrogen and hydrogen which, when
mixed in the ratio of one mole N2 and three moles H2 and passed through a
catalytic bed at approximately 34.5 MPa (5,000 psi) and about 538°C (1000°F),
results in the production of two moles of NH3. Modern synthesis plants can
convert approximately 94% of the nitrogen and hydrogen mixture to ammonia.
The reaction equation is:
N2 + 3H2 = 2NH3
Each plant design may be different in terms of gas compression, cleanup, and
conditioning. In recent years, more attention has been paid to utilizing
equipment requiring less energy with more heat recovery because of the large
amount of power needed to drive the compressors and the large amount of
process steam required for the water-gas shift reactor and reboiler steam for
stripping of acid gases. Attention must be paid to the detailed design to
minimize use of hydrocarbons as compared to that produced from natural gas or
naphtha steam reformers and high temperature partial-oxidation steam systems
that essentially convert all hydrocarbons. An ammonia plant has not yet been
built using gases such as those from Purox or a Lurgi coal gasifier. Further-
more, each new gasifier or feedstock produces trace impurities that may be
deleterious to catalysts. A number of plants have been built for coal gasi-
fication to ammonia, but some of these are no longer in operation (Ref. C-6).
C-21
-------
Examples of partial-oxidation systems are the Koppers-Totzek, Texaco, and
Shell processes (Ref. C-7). If a gas cleanup and treating scheme is designed
for the Purox gas, a pilot unit must be built and tested with the actual gases
from the demonstration plant at South Charleston. The gas conditioning
process has been and will continue to be a subject of research and
development.
In the design of the ammonia plant, a major effort will need to be
directed at converting pyrolysis gas to a pure H2 stream. Several routes are
possible in gas treating. An important item is to use as much of the hydro-
carbons in the gas as a source of hydrogen. Therefore, a steam reforming
unit, not previously used in partial oxidation of oil or coal, is considered
for the system. Use of a reformer to convert the hydrocarbons in the off-gas
from a Lurgi gasifier to CO and H2 has been discussed (Ref. C-8).
There are several choices in treating the syngas. Each requires an acid
gas scrubber initially, and then in the subsequent processing it is necessary
to make choice of one of the following processes:
1. Removal of the hydrocarbons as a separate stream in the purification
step and then steam reform them (heavy hydrocarbons will not reform)
2. Passage of all the gas through a steam reformer that will not convert
the heavier hydrocarbons.
3. Passage of all the gas through a reformer-methanator that produces
which is then steam reformed.
4. Passage of all the gas through a cryogenic fractionator to separate
the components and steam reform the lighter hydrocarbons (German
Linde) .
These schemes are expensive and can increase considerably the cost of
existing types of ammonia processes. Shown in Figure C-5 is the type of
ammonia plant considered here, one that uses a liquid nitrogen wash for final
purification of the gas to be synthesized to NH3.
For present purposes, approximately 10% of the gas input, along with one-
third of the hydrocarbon gas, is considered necessary to provide sufficient
process steam and heat for the ammonia plant. This quantity should be more
accurately determined during a preliminary engineering facility design phase.
High purity nitrogen will be furnished from the oxygen plant, which is modi-
fied from that in which the product is oxygen only. There is more than
sufficient N2 available, but some will be in liquid form for use in the
Cryogenic hydrogen purification section.
A low-pressure compressor is used to pass the gas through an acid gas
absorption system to remove H2S, C02, and traces of HC1. Union Carbide
reports finding 1 ppm or less of HC1 in the Purox gas. The H2S and C02 tail
gas stream is then processed in a Stretford unit to produce sulfur, which is
removed as a solid at the rate of approximately 3.6 Mg/d (4 TPD) . The C02 is
exhausted to the atmosphere at this point. The gas is then heated and passed
C-22
-------
CARBON DIOXIDE
REMOVAL
n
to
SYNGAS PURIFICATION
tRECTISOL PROCESS IS ALTERNATIVE)
H2S.C02
C02
SHIFT CONVERTERS
PUROX SYNGAS
STRETFORD
PROCESS
COMPRESSOR
BENFIELD
PROCESS
H2s,co2
REMOVAL
1 /"*
SULFUR [ |
*•) r— — STEfl
REFOR
IV;
METHANATOR
REFRIGERATOR
I
^
\
^
HYDROCARBON
GAS
FUEL
TO STEAM
BOILER
FROM O,PLAI
,NT—<
\_GNj
RECYCLE SYNGAS
HIGH-
PRESSURE
SEPARATOR
AMMONIA
PRODUCT
Figure C-5. Flow diagram of ammonia plant.
-------
through a water-gas-shift reactor to convert the CO and steam to hydrogen.
Gas from the shift converter consists mostly of H2, C02, hydrocarbons, and
a small amount of CO. Next, the C02 is removed in an amine absorber and the
gas then passed through a methanator where CO is reduced to less than 10 ppm.
From here, the gas is cooled along with a stream from the purge of the final
ammonia recovery tank. A good deal of water is condensed. For more complete
dryness, the gases are passed through a molecular sieve bed. After drying,
the gas is scrubbed with liquid nitrogen, which condenses out the remaining
hydrocarbons leaving a purified stream of H2. This is then mixed with
gaseous N2 from the oxygen plant. In this cryogenic purifier, 99% of the
methane, all other hydrocarbons, 65% of the argon, and about half of the
10 ppm of CO are removed. The hydrocarbon off-gas from the purifier is used
to regenerate the molecular sieve bed and then passed to a steam reformer to
convert the hydrocarbons to a mixture of CO and H2, which is then added to
the main syngas stream going to the shift converters. The heavier hydro-
carbons (GI++) will not reform, however, in presently used steam reformers.
The reason for using a steam reformer is that the Purox gas contains a good
deal of hydrocarbons, and rejecting them all to fuel gas would reduce the
ammonia output considerably. These hydrocarbons are in themselves needed
feedstock for making H2. The mixture of N2 and H2 is then compressed, using
electric motor reciprocating compressors. With the amount of ammonia produced
at less than 363 Mg/d (400 TPD), centrifugal compressors are not considered
economically feasible.
The compressed gases, including recycled gas from the high-pressure
ammonia separator tank, pass through a regenerative heat exchanger and then
through the ammonia synthesis catalytic reactor tubes where N2 is hydrogenated
to produce ammonia. The exothermic heat of the reaction is removed in a steam
generator. This steam is utilized in reboilers and for processing. The mix-
ture of ammonia, hydrogen, and nitrogen from the reactor is cooled in the
regenerative heat exchanger. With further cooling, the ammonia is condensed
and the nitrogen and hydrogen mixture is recycled back through the reactors.
A purge stream is sent back to the purification train to prevent a buildup of
inert gases. The ammonia liquid is dropped in pressure and passed to a
18 144 Mg (20,000 ton) cryogenic storage tank. Such a large tank will allow
storage of approximately 60 days of plant production during the period of low
ammonia usage found in most farming areas during the winter. Choice of
storage type and size depends on use of the ammonia. In some cases where a
surge capability only is desired, storage under pressure in a sphere is
desirable. Such tanks are built up to 454 Mg (500 tons) in capacity. This
occurs where ammonia is transported by ship to areas where farming occurs all
year.
A production rate for ammonia is based on an estimate of the probable
amount of Purox-type syngas that can be converted to hydrogen and then to NH3.
If all the syngas were utilized, there would be 0.85 mole of NH3 per mole of
Purox gas which is equivalent to 0.34 ton NH3 per ton raw refuse or 473 Mg/d
(522 TPD) for a 1361 Mg/d (1,500 TPD) raw refuse feed plant. It is estimated
that 10% of the syngas and one-third of the remaining hydrocarbons and H2, not
converted, are used for generating steam. This results in 0.58 mole NH3 per
mole of syngas, which is equivalent to 0.237 ton NH3 per ton of raw refuse,
or 323 Mg/d (356 TPD) NHs for the plant. A more optimistic estimate of the
C-24
-------
utilization of the hydrocarbons (only 20% for heating purposes) leads to
363 Mg/d (400 TPD) NH3 for the plant. If all of the hydrocarbons are used for
fuel (no reforming), then the production of ammonia would be approximately
227 Mg/d (350 TPD). Until a more detailed preliminary engineering design is
made of the ammonia plant, a value of 323 Mg/d (356 TPD) NH3 is used in deter-
mining the production cost of NH3 for a 1361 Mg/d (1,500 TPD) raw refuse
plant. If the production rate was based on the non-use of hydrocarbons, the
revenues for ammonia would be 30% less than the base used herein.
CONVERSION OF SYNGAS TO AMMONIA CONSTRUCTION AND OPERATING COSTS
The costs presented here are approximations and reflect a great deal more
uncertainty than the presentations for the front-end or the Purox facility.
This is caused by the fact that the syngas of the composition produced has
never been transformed to another product, and must be tested when catalysts
are being used. The technology is available, but the unknown factor is the
complexity, and so is, therefore, the cost of the specific facility required
for the conversion process. This can only be determined by an in-depth
chemical engineering process analysis of the required system.
The costs presented are based on taking syngas from the 1361 Mg/d
(1,500 TPD) unit used as the base case. Because of the lesser system utiliza-
tion factor (0.88) resulting from the addition of the ammonia conversion
plant, only 438 178 Mg/y (483,000 TPY) of raw refuse would be processed and
104 328 Mg/y (115,000 TPY) of ammonia produced.
Capital Cost
Capital cost for a complete solid waste to ammonia facility would start
with the cost of the syngas production facility, to which would have to be
added a complete ammonia conversion unit. The ammonia plant could be adjacent
to the syngas facility or at some distance away. The connection between the
two is a pipeline, whose cost is directly proportional to distance. Land
requirement is 32 400 m2 (8 acres) at a minimum and located such as to
incorporate a rail siding since much of the ammonia distribution from the
plant would be by railway tank car. Electric power up to 20,000 kVA will be
required and a water supply approximately equal to the syngas plant will be
needed. The location should be away from populated areas because of the
possibility of accidental release of ammonia.
The construction cost is estimated at $31.6 million for the ammonia unit
and the total plant construction cost at $85.35 million including the front-
end and syngas units. The capital cost for this total plant system are
presented in Table C-8 and amounts to $100.66 million in 1975 dollars.
Operating Costs
Operating costs are listed in Table C-9 for the ammonia unit and the
syngas plant, including the front-end and gasifier. The same degree of uncer-
tainty applies to these figures as with the capital costs.
C-25
-------
TABLE C-8. ESTIMATED CAPITAL COST REQUIREMENTS FOR AMMONIA
(1975 DOLLARS)
Cost
Item ($ million)
Construction:
Front-end processing and syngas plant 53.75
Ammonia conversion plant 31.60
Interest during construction 7.25
Startup costs 4.84
Working capital 3.22
Total 100.66
TABLE C-9. ANNUAL OPERATING COSTS FOR AMMONIA
(1975 DOLLARS)
Item
Cost
($000)
Ammonia Unit:
Labor
Power
Maintenance
Production Materials
Water
Subtotal
Front-End and Purox System
Plant Total
404
3,374
1,580
345
178
5,881
7,016
12,897
C-26
-------
In the table, labor includes only operating labor; maintenance labor is
on 1Vhe M*intenance item- The operating cost shown is equivalent to
$29.42/Mg ($26.69/ton) o£ raw refuse or $123.62/Mg ($112.15/ton) of ammonia.
ECONOMIC ANALYSIS FOR AMMONIA PLANT
Unit capital amortization cost is $101.95/Mg ($92.49/ton) of ammonia.
With the operating cost of $123.62/Mg ($112.15/ton) ammonia added, the total
unit cost is $225.56/Mg ($204.63/ton) of ammonia produced, or $53.82/Mg
($48.83/ton) raw refuse.
Applying credits due to revenues from drop charges and sale of aluminum
and iron, the net cost can be lowered considerably as shown in summary
Table C-10. As an example, a drop charge of $11.02/Mg ($10/ton) of raw
refuse and sale of recovered aluminum at $331/Mg ($300/ton) and steel at
$44/Mg ($40/ton) yields a net cost for producing ammonia of $158.73/Mg
($144/ton) ammonia. Marketing and wholesale delivery of ammonia can increase
the cost to the manufacturer of ammonia, but was not included; this could be
approximately $28/Mg ($25/ton) of ammonia.
Ammonia manufacture is more acceptable in regions with a high drop-
charge. For this reason, the City of Seattle found it financially feasible
to proceed to a preliminary engineering phase with ammonia manufacture because
the drop-charge used is about $13.23/Mg ($12/ton) raw refuse. Parsons, in a
recent study for the Denver Regional Council of Governments (Ref. C-2), found
that ammonia manufacture was not financially acceptable because of the low
drop-charge available at their landfills. Another method of resource recovery
was recommended.
SUMMARY OF ALTERNATIVE USES OF PUROX
In summary, results of all costs associated with the Purox gasification
system and the addition of conversion to methane, electric power, methanol, and
ammonia are presented in Table C-10. Comparisons can be made, but this may be
meaningless unless the products being compared each have a specific market in
a given area. It is therefore recommended strongly that a thorough engineer-
ing study be made for each community interested in disposing refuse in a
useful manner to determine which of several systems and products would fit
their needs. In particular, net cost of operation or drop-charge to break
even with given markets are items best describing the system rather than
construction or capital costs alone.
C-27
-------
TABLE C-10. NET COSTS(1) TO PRODUCE PRODUCTS USING SYNGAS
1500-TPD - 7 D/Wk, 1750-TPD - 6D/Wk Delivery, 110-TPH - 16 hrs/D Feed Preparation Rate
Revenues: $40/T Steel, $300/T Aluminum, Drop Charge Listed 0,5,10 $/T
Amortization 8.5%, 20 Years
o
i
K)
OO
Product
(A "
Product1 •
Rate
Capital
cost
$ mil lion
Annual
OSM
cost
($OOOJ
$/T Raw refuse ^
Drop
0
charge $/T
5 10
$/Million Btu1-3-1
Drop
0
charge $/T
5 10
$/T Product ^
Drop
0
charge $/T
5 10
mils/kWh(3;i
Drop
0
charge $/T
5 10
Syngas 27.7 x 106
(Fuel Gas) SCFD
Methane 8.2 x 106
(SNG) SCFD
Electric
Power (
Methanol
25.94 MW
245 TPD
62.40 7,016 22.56 17.56 12.56 3.30 2.53 1.84
79.93 8,253 31.44 26.44 21.44 6.20 5.22 4.23
79.96 5,945 24.71 19.71 14.71
89.14 11,372 38.68 53.68 28.68 10.94 9.53 8.11 257 206 176
60.9 48.6 36.2
Ammonia
356 TPD
100.66 12,897 44.38 39.38 34.35
186 165 144
(1) In 1975 Dollars the net cost is defined as .amortization of capital cost plus operating cost less revenues from
steel and aluminum sale, and drop charges.
(2) Net power shown - all power to Front End, Purox System and Gas Compressor furnished by Turbo-Electric Plant.
(3) Alternate unit costs are directly related by ratio of product rate to raw refuse delivery rate of 1,500 TPD.
(4) HHV's of Products: Syngas 370 Btu/SCF, methane (SNG) 927 Btu/SCF, Methanol 21.64 x 106 Btu/Ton. The utilization
factor on a yearly basis is Syngas 0.92, methane 0.85, electric power 0.91, methanol 0.88, ammonia 0.88.
-------
REFERENCES
C-l. "East Bay Energy Resource Recovery System" (EBERRS), The Ralph M.
Parsons Co., Report for Pacific Gas and Electric Co., San Francisco,
1975 in cooperation with the East Bay Municipal Utility District and
Oakland Scavenger Co.
C-2. "Resource Recovery from Solid Waste," Vol. II, The Ralph M. Parsons Co.,
Report for Denver Regional Council of Governments, Denver, 1976.
C-3. White, G. A., T. R. Roszkowski, and D. W. Stanbridge, "The RM Process,"
in Methanation of Synthesis Gas, American Chemical Society, 1975.
C-4. Dory, J. E., "Gasification of Solid Waste: An Alternative to Solid
Fuel," Institute for Energy Analysis, Oak Ridge Associated Universities,
IEA(M)-75-6.
C-5. Reed, T. B., et al., "Improved Performance of Internal Combustion
Engines Using 5 to 30% Methanol in Gasoline," Lincoln Laboratories,
MIT, Cambridge, Massachusetts.
C-6. Slack, A. V. and G. R. James, Ammonia, Part 1, Marcel Dekker, N. Y.,
1975, p 342.
C-7. Ibid, p 293, 334.
C-8. Ibid, p 338.
C-29
-------
APPENDIX D
ENVIRONMENTAL CONTROL CONSIDERATIONS
AIR POLLUTION
Introduction
A brief discussion of air pollution considerations follows that
encompasses:
• The problem of air pollution
• Emission standards
• Air pollution control equipment selection factors
• Air pollution control equipment
• Air pollution control equipment costs
More detailed information can be obtained by examination of the associated
references and by consulting manufacturers of air pollution control equipment.
The Problem of Air Pollution
Air pollution consists of particulate matter or impurities that are sus-
pended in or conveyed by a moving stream of gas or air. An air pollution
problem arises when the concentration of these substances interferes with the
public well-being. The Engineers Joint Council has developed a specific
definition:
"Air pollution means the presence in the outdoor atmosphere of one
or more contaminants, such as dust, fumes, odor, smoke or vapor, in
quantities, or characteristics, and of duration such as to be injur-
ious to human, plant or animal life or to property which unreasonably
interfere with the comfortable enjoyment of life and property."
(Ref. D-l)
The air pollutant problems usually encountered by energy producing
industries include particles, sulfur oxides, nitrogen oxides, and visibility.
Of the mentioned pollutant problems, particles generally cause the most con-
cern. Particles can take many forms and are defined in Table D-l.
D-l
-------
TABLE D-l. DEFINITIONS OF PARTICLE TYPES (Ref. D-2)
Particulate Matter:
Aerosol:
Dust:
Flyash:
Fog:
Fume:
Mist:
Particle:
Smoke:
Soot:
Any material, except uncombined water,
in the form of solid or liquid in the
atmosphere or in a gas stream.
Solid or liquid particles of microscopic
size, such as smoke, fog or mist.
Solid particles larger than colloidal
size and capable of suspension in air
and other gases.
Particles formed as a result of combus-
tion of fuel.
Visible aerosols where liquid is the
dispersed phase.
Particles formed by condensation, sub-
limation or chemical reaction, e.g.,
condensed metal oxides.
Low concentration dispersion of rela-
tively small liquid droplets.
Small, discrete mass of solid or liquid
matter.
Small particles resulting from incom-
plete combustion of combustible
material.
An agglomeration of tarry carbon parti-
cles formed by incomplete combustion
of carbonaceous material.
Sulfur oxides are formed whenever a fuel or fuels containing sulfur or
its compounds are burned. The major portion of these emissions is sulfur
dioxide (S02). Sulfur oxide emission levels vary directly with the amount of
sulfur originally in the fuel utilized. Normally high sulfur fuel will gener-
ate high sulfur emissions and low sulfur fuel will generate low sulfur
emissions, etc.
Nitrogen oxides are formed during usual combustion processes due to
reaction of the nitrogen and oxygen in the combustion air supply. Nitric
oxide (NO) and nitrogen dioxide (N02) formation increases as the temperature
of combustion increases.
D-2
-------
Visibility is not a pollutant in itself, but is a direct result of var-
ious types of particle pollutants. Visibility (or opacity) is categorized as
a pollutant due to the fact there is a general correlation between plume
opacity and the quantity of pollutants released, thereby providing an inexpen-
sive and simple means of enforcement by visual inspection.
Control of the above described air pollutants can take many forms, but
before control can be accomplished a criteria must be established as to what
allowable level of pollutant emissions will be maintained. These criteria are
called emission standards.
Emission Standards for Air Pollution
Emission standards can be categorized into seven general areas. These
seven areas and a brief outline of what they encompass can be found in
Table D-2. Some or all of these emission standards may be applicable to a
particular point source, the choice being dependent on the controlling agency
involved. Consideration of air pollutant emission standards is an essential
step when deliberating over modifications, or new plant design of energy
generating systems.
TABLE D-2. CATEGORIES OF EMISSION STANDARDS
Standards Category
Coverage
1. Visible emissions
Based on emission plume opacity
usually utilizing Ringlemann scale
of obscuring power.
2. Particulate matter concentration Based on either
a) weight concentration of particles
in stack, or
b) relationship between weight of
emitted particles to the total
weight of material processed.
3. Exhaust gas concentration
4. Emission Prohibition
5. Performance standards
6. Regulation of fuel
7. Zoning restrictions
Limits gaseous emissions to a prescribed
level based either on a ratio of the
stack gas on the total weight of
material processed.
Bans certain types of emissions.
Based on either:
a) regulation of process operations
so as to not exceed maximum emis-
sion rate, or
b) equipment design standards.
Limit types of fuels used in burning.
Limits certain of industries from opera-
ting in a particular area.
D-3
-------
Performance standards for new or modified sources have not been pro-
mulgated on the Federal level for many types of waste to energy facilities.
If such is the case for a particular facility, a request for determination
must be made to the regulating agency. After a review of engineering, design,
and equipment data, the source will be categorized under an existing perform-
ance standard, or other provisions made. The new or modified facility will
then have to meet state and local regulations if these are more restrictive
than federal ones. Some selected new source performance standards are shown
in Table D-3 that may be applicable to waste to energy facilities.
When the required emission standards have been discerned for the pro-
posed facility, selection of pollution control equipment that will meet the
emission standards and also be compatible with facility operations can begin.
Air Pollution Control Equipment Selection Factors
The selection of equipment for air pollution control for a particular
application is dependent on many factors, including physical characteristics
of the pollutant, carrier gas characteristics, process factors, operational
factors, emission standards, and costs. They can be described as follows:
Physical characteristics include particle size distribution, weight,
density, shape, hygroscopicity, agglomerating tendency, corrosive-
ness, stickiness, flowability, electrical conductivity, flammability,
and toxicity.
Carrier gas characteristics include temperature, humidity, vis-
cosity, pressure, corrosiveness, toxicity, electrical conductivity,
flammability, and explosiveness.
TABLE D-3. SELECTED NEW SOURCE PERFORMANCE STANDARDS (Ref.D-3)
Fossil Fuel Fired Sewage Plant
Pollutant Steam Generator Incinerators Sludge Incinerators
Grains/Million Cal Heat Input Grains/dscm (Grains/kg dry
sludge input
Particulates 0.18 0.18 0.65 /
N0x 1.4# , 2.2*
S02 0.36# , 0.54* , 0.26*
Opacity 20% - 20%
Notes: # Liquid fuel, * Solid fuel,& gaseous fuel.
D-4
-------
Process factors include gas flow rate, particle density, collector
system efficiency requirements, pressure differential allowed,
product quality requirements, process material rate of flow, and
quality variations in process material.
Operational factors include available floor space, headroom, con-
struction material limitations due to process factors, power
requirements, temperature limitations, pressure drops, maintenance
problems, and waste disposal.
Emission standards are discussed in the subsection entitled
"Emission Standards for Air Pollution."
Costs include purchase, installation, operation, and maintenance
of air pollution control equipment. Costs are further discussed
in the subsection entitled "Air Pollution Control Equipment Costs."
Selection of air pollution control equipment must be made with considera-
tion of changing emission standards and the rapidly changing control tech-
nology. The year data is compiled is very significant since equipment,
material, and labor costs have risen each year.
In selecting air pollution control equipment, collection efficiency is
a major deciding factor. Collection efficiencies for various air pollution
control devices are shown in Table D-4, where it can be seen that there is a
wide range of efficiencies. These efficiencies depend on the selection fac-
tors previously discussed that lead to system compatibility of the control
equipment. An understanding of the different air pollution control devices
and how they operate is needed before deciding on a piece of equipment that
will be compatible with the proposed facility.
Air Pollution Control Equipment
Air pollution control devices are typically designed either to remove
particles or gaseous effluents and are traditionally grouped according to the
means used to remove the undesirable components from an effluent stream.
These devices can be categorized into six distinct classes:
• Mechanical Collectors
• Filters
• Wet Scrubbers
• Electrostatic Precipitators
• Combusters or Afterburners
• Stacks.
D-5
-------
TABLE D-4. AIR POLLUTION CONTROL EQUIPMENT COLLECTION
EFFICIENCIES (Ref.D-4)
Typical efficiency ranges
Equipment Type (on a total weight basis)
ft)
a "*"
Electrostatic precipitator 80 to 99.5
Fabric filtersb 95 to 99.9
Mechanical collector 50 to 95
Wet collector 75 to 99+
Afterburner
Catalytic0 50 to 80
Direct flame 95 to 99
a Most electrostatic precipitators sold today are designed for 98 to 99.5%
collection efficiency.
b Fabric filter collection efficiency is normally greater than 99.5%.
c Not normally applied in particulate control; has limited use because
most particulate matter poisons or desensitizes the catalyst.
Mechanical Collectors--
Mechanical collectors (See Figure D-l) are specifically used for the
removal of particle emissions and are considered the simplest method of parti-
cle control. Removal is effected by either gravity, particle inertia, or
centrifugal force.
Settling chambers slew the effluent gas flow allowing gas particles to
fall out due to gravity.
Inertial separators collect particles on stationary impaction targets and
the most widely used type of pollution control equipment.
Cyclones are a specific type of inertial separator that swirl the efflu-
ent gas stream inside its cylindrical shape, thereby forcing the high density
particles to the sidewalls by centrifugal means and are separated from the
effluent gas stream when they fall to the bottom of the cyclone. Efficiencies
depend mainly upon particle characteristics with removal efficiency decreasing
rapidly with decreasing particle size. These units are most effective in
collecting particles 15 micrometers or larger. Efficiencies vary widely and
range from 20% to 90%. The chief advantages of mechanical collectors are the
D-6
are
-------
o
CLEAN GAS
OUT
Enlarged cutaway shows
Inlet vanes, collecting
cell and discharge tube.
,,;. DIRTY
'"'GAS IN
Figure D-l. Typical cyclonic dust collector.
-------
low purchase and operating costs, with the major deficiencies being low effi-
ciency and large space requirements. In most cases, mechanical collectors are
used as first stages in more complex air pollution control systems. The par-
ticulate matter after collection is usually disposed in landfills.
Filters--
Mechanical screening can be an effective way to trap solid particles. In
this type of unit, the effluent gas is passed through a porous material (a
filter or baghouse installation) that collects particulate matter larger than
the pore (or mesh) size? Fibers used as filtering media include cotton,
Dynel, wool, nylon, Orion, Dacron, and glass, all with variation in weave,
count, finish, etc. Other variations include the size and shape of filters,
arrangement, spacing of bags, and method of cleaning. Systems are generally
efficient for the collection of particles greater than 0.5 micrometer, with
efficiencies over 99% a common occurrence. In operation, the particles are
initially captured and retained on the fibers of the cloth by means of inter-
ception, impingment, diffusion, gravitational settling, and electrostatic
attraction, with the cloth periodically cleaned of entrapped particles. The
chief advantages of a filtering system are high efficiencies with small and
large particles and moderate pressure drop. The disadvantages are 290°C
(550°F) temperature limit, costly bag replacement, and unit size. Figures
D-2 and D-3 show typical systems.
Wet Scrubbers--
Wet scrubbers use the action of a liquid colliding with the pollutant to
remove solids, liquids, and gases from effluent streams (See Figures D-4 and
D-5). Collision causes the pollutant to be absorbed by the wet scrubbing
medium and the resulting sludge or solution is then physically removed from
the system. The units are effective at moderately high temperatures and are
not significantly affected by particle loading. Wet scrubbers are increas-
ingly utilized as a method of air pollution control, with efficiency varying
between 80% and 99.5% for particles ranging from less than lym to 10 urn.
Disposal of the collected material in water without clarification or treatment
may cause water pollution problems. Scrubbers are usually categorized as low
energy and high energy types. Low energy scrubbers have low capital, opera-
tion, and maintenance costs and also lower efficiency, whereas high energy
scrubbers have moderate capital and operation costs and high maintenance
costs, but high efficiency. Other disadvantages of wet scrubbers include cor-
rosion problems, freezing in cold weather, water vapor visibility during cer-
tain weather conditions, and contamination of exhaust gas.
Electrostatic Precipitators--
Electrostatic precipitation is frequently called the Cottrell process
and consists of the use of an electrostatic field for precipitating or remov-
ing solid or liquid particles from a gas in which particles are carried in
suspension (See Figures D-6 and D-7). The basic principles involved are
*0nce a filter cake has built up on the surface, much smaller particles can
be retained.
D-8
-------
DIRTY AIR
Figure D-2. Shaker-type fabric filter.
D-9
-------
Air-shake
cleaning
Filtering
Figure D-3. Flow diagram of a fabric filter.
D-10
-------
Figure D-4. Flooded wall Venturi scrubber.
D-ll
-------
GAS INLET I
VENTURI
THROAT NOZZLES
AGGLOMERATION ZONE
GAS OUTLET
ENTRAINMENT
SEPARATOR
SLURRY
OUTLET
Figure D-5. Flow diagram of a Venturi scrubber.
D-12
-------
High Voltage
Bus Duct
Outlet
Nozzle
"RS"
Discharge
Electrodes
Support
Steel
Transformer
Rectifier
Discharge
Electrode
Rapper
Mechanism
Collecting
Plate
Rapper
Drives
Insulator
Compartment
Collecting
Plate
Rappers
Discharge
Electrode
Weights
Support
Insulator
Collecting
Plates
Inlet
Nozzle
Gas
Distribution
Plates
Hoppers
Figure D-6. Cut-away view of an electrostatic precipitator.
D-13
-------
Insulator
Precipitator
shell
Discharge
electrode
Dust on
precipitator wall
Clean gas
exit
High-voltage-
cable
Rectifier
set
_L
a-c
input
Gas entrance
Collected dust
Figure D-7. Flow diagram showing elements of an electrostatic precipitator.
D-14
-------
particle icnization, normally by a discharge electrode, and then entrapment by
a collecting plate, normally consisting of a grounded electrode. Electro-
static precipitators effectively collect a wide range of particle sizes from
1 to lOOum range. Efficiencies in excess of 99.5% can be achieved in many
cases. A mechanical collector generally precedes an electrostatic precipita-
tor because large particles can cause damage to the discharge electrodes.
Particles collected are usually disposed in landfills. Precipitators are
divided into two general types: 1) single-stage (Cottrell) and 2) two-stage
(Penney), with both operating on essentially the same principles. The differ-
ences in the single-stage and two-stage basically arise in the manner of par-
ticle ionization and the use of high and low voltage, respectively. The sin-
gle-stage is also designed for use in processing large volumes of air. The
advantages of an electrostatic precipitator are its high efficiency and
relatively low operating costs, with the disadvantages being high capital
cost, critical electrode voltage requirements, the constant cleaning neces-
sary, and use of a precleaner, generally a cyclone, for the gas effluent.
Combustors or Afterburners--
Combustors or afterburners are essentially incineration chambers in
which objectional aerosols or gases from some process are oxidized to a less
offensive form, ideally to carbon dioxide (€62) and water. Supplemental heat,
usually required to sustain the oxidation reactions, is supplied by burning
auxiliary fuel in either a preheat chamber or in the reaction chamber itself,
or by passing the incoming pollutant stream through a heat exchanger. Well
designed afterburners, correctly applied and properly maintained, can achieve
over 98% destruction of pollutants. Afterburners can be direct-fired or
catalytic. Direct-fired (or thermal) afterburners are generally refractory-
lined steel enclosures, cylindrical or rectangular in shape, in which the
pollutant stream is mingled with the hot gases of a gas or oil burner elevat-
ing the pollutant stream temperature till oxidation is complete. A catalytic
afterburner operates on the principal fact that many chemical reactions will
proceed to completion at well below normally required temperatures in the
presence of a catalyst.
Stacks and Chimneys--
Historically, a stack's function was to provide natural draft for com-
bustion. Today, stacks are also used as a means of air pollution control by
reducing exhaust gas temperatures and increasing dispersion of contaminants
to achieve lower ground level concentrations. Stacks are not a solution to
the air pollution problem because they do not reduce the amount of pollutant
released to the atmosphere. Limitations on stack emissions to prevent remote
pollution problems and atmospheric and ground-level concentration buildup will
require the use of other methods of air pollution control. Stack operation is
significantly affected by temperature inversions, meteorological conditions,
and adjacent buildings and terrain, which are all difficult to predict.
Stacks today range up to 300m so as to reach above the inversion layer and
provide better dispersion.
D-15
-------
Reinforced concrete chimneys are strong and can withstand higher winds
than other types. The major disadvantage is that the surface may crack due
to temperature and plastic stress conditions. Large chimneys are more eco-
nomical when they are constructed with concrete rather than other building
materials. Masonry chimneys may be constructed from straight or radial
bricks or blocks. They are built with a lining shell of fireclay brick that
will resist temperatures of 500°C (950°F) and an outer shell of ordinary
brick. An annular airspace of 5 to 15 cm (2 to 6 in.) normally separates the
two shells. Linings are also used in concrete chimneys. Vibration damage
near top can be caused by high exit velocities. Steel chimneys have low
initial cost, are easily constructed and are lightweight in comparison to
other designs. Disadvantages are high maintenance costs and unsightly appear-
ance. It is more economical to support steel chimneys by guy-wires than to
make them self-supporting. A thin lining of bonded fused-glass is normally
used on the inside of the steel stack. (Ref.D-2).
Table D-5 shows the various types of collectors mentioned and their
collection range of particles. Table D-6 summarizes the approximate charac-
teristics of air pollution control equipment and Table D-7 summarizes the
advantages and disadvantages of different air pollution control devices.
When an understanding of air pollution control equipment selection oper-
ating factors is achieved, cost comparisons can begin. Air pollution control
equipment costs are discussed in a separate subsection.
Air Pollution Control Equipment Costs
Air pollution control equipment costs can be separated into two main categor-
ies: capital investment or first cost and operating or repeated costs.
Capital cost is the total installed cost of a particular air pollution control
TABLE D-5. TYPES OF COLLECTORS FOR RANGE OF PARTICLE SIZE (Ref.D-5)
Type Useful Size Range, micrometers
Packed Tower 0.001 to 0.5
High Energy Scrubber 0.03 to 5.0
Low Energy Scrubber 0.001 to 0.008 (gas), 1.0 to 15
Dry Cyclone Collector 8 to 1000
Electrostatic Precipitator 0.008 to 30
Cloth Collector 0.008 to 1000
D-16
-------
TABLE D-6. APPROXIMATE CHARACTERISTICS OF DUST AND MIST COLLECTION EQUIPMENT
c- Incl
fabric
tempera
2
O
rt
1 £ Relative
^ " Equipment Type Costa
< »
a- Including necessary aux
udes pressure loss, water p
85°C, synthetic fabrics up
ture gases will be necessar
X rt
0 H-
d H4- C.
HJ fO
Settling Chamber
1. Simple
2. Multiple Tray
Inertial Separators
1. Baffle Chamber
2. Orifice Impaction
3. Louver Type
4. Gas Reversal
5. Rotating Impeller
1
2-6
1
1-3
1-3
1
2-6
Smallest
Particle
Collected
40
10
20
2
10
40
5
Space
Moderate
Moderate
Pressure
Drop
(Inches H20)
0.1-0.5
0.1-0.5
0.5-1.5
1.3
0.3-1
0. 1-0.4
Power Usedc
kW
1000 ftVmin
0.
0.
0.
0.
0.1
0.1
1-0.5
2-0.6
1-0.2
0.1
5-2
Cycl ones
1.
T3 -••
l-i 0 - Q
CO O t/>
< - (D •
S 2
rt 0
"-{ O
CJ X
v~- to
O
CD °
< n
M •
T3
O
fU I
M- -3
r-
0 0
t> o
~. p
CD ,-.
O CT L) •
H' ^-
>—• 3*
CD ^D
O 1
Qa tjn £
O
C_ HI
(-;
o >-•
>— O
•14 F.
o o x
2.
Single
Multiple
1-2
3-6
15
5
Filters
1.
2.
3.
Tubular
Reverse Jet
Envelope
3-20
7-12
3-20
0. 1
0. 1
0.1
Electrical Precipitators
1.
2
One- Stage
Two-Stage
6-30
2-6
0. 1
0. 1
Small
Large
Large
Moderate
Large
0.5-3
2-10
2-6
2-6
2-6
0.1-0.5
0.1-0.3
0.
0.
0.
0.
0.
0.
0.
1-0.6
5-2
5-1.5
7-1.5
5-1.5
2-0.6
2-0.4
Scrubbers
1.
7
Spray Tower
Jet
1-1
4-10
10
2
Moderate
0.1-0.5
0.
-) _
1-0.2
10
S o £
U^ "
0* =r
',/:
(JO
• o y
rt .
0
3.
4.
S.
6.
7.
Venturi
Cyclonic
Inertial
Packed
Rotating Impeller
4-12
3-10
4-10
3-6
4-12
I
5
2
5
7
Mode rat e
Moderate
Sma 1 1
Large
Sma 1 1
10-15
2-8
2-15
0.5-10
->__
0,
0.
0.
->_
10
h-2
8-8
6-2
10
Max. Temp. ,
°C, Standard
Construction Remarks
400°
Large, low pressure drop,
precleaner
Difficult to clean, warpage
problem
400°
Power plants, rotary kilns,
acid mists
Acid mists
Fly ash, abrasion problem
Precleaner
Compact
400°
Simple, inexpensive, most
widely used
Abrasion and plugging problems
400°d
High efficiency, temperature
humidity limits
More compact, constant flow
and
Limited capacity, constant flow
possible
650°
High efficiency, heavy duty,
expensive
Compact, airconditioning
service
Unl imi ted°
Common, low water use
Pressure gain, high velocity
liquid jet
High velocity gas stream
Modified dry collector
Abrasion problem
Channeling problem
Abrasion problem
(Refs. D-6, D-7)
-------
TABLE D-7. ADVANTAGES AND DISADVANTAGES OF COLLECTION DEVICES
Collector
Advantages
Disadvantages
Gravitational
Cyclone
Wet collectors
o
t—
oo
Low pressure loss, simplicity of
design and maintenance
Simplicity of design and maintenance.
Little floor space required.
Dry continuous disposal of collected
dusts.
Low to moderate pressure loss.
Handles large particles.
Handles high dust loadings.
Temperature independent.
Simultaneous gas absorption and
particle removal.
Ability to cool and clean high-
temperature, moisture-laden gases.
Corrosive gases and mists can be
recovered and neutralized.
Reduced dust explosion risk,
Efficiency can be varied.
Electrostatic precipitator 99+ percent efficiency obtainable.
Very small particles can be collected.
Particles may be collected wet or dry.
Pressure drops and power requirements
are small compared to other high-
efficiency collectors.
Maintenance is nominal unless corro-
sive or adhesive materials are
handled.
Much space required. Low collection
efficiency.
Much head room required.
Low collection efficiency of small
particles.
Sensitive to variable dust loadings
and flow rates.
Corrosion, erosion problems.
Added cost of wastewater treatment and
reclamation.
Low efficiency on submicron particles.
Contamination of effluent stream by
liquid entrainment.
Freezing problems in cold weather.
Reduction in buoyancy and plume rise.
Water vapor contributes to visible
plume under some atmospheric
conditions.
Relatively high initial cost.
Precipitators are sensitive to vari-
able dust loadings or flow rates.
Resistivity causes some material to
be economically uncollectable.
Precautions are required to safeguard
personnel from high voltage.
Collection efficiencies can deteri-
orate gradually and imperceptibly.
-------
TABLE D-7. ADVANTAGES AND DISADVANTAGES OF COLLECTION DEVICES (Continued)
Collector
Advantages
Disadvantages
Electrostatic precipitator
(Continued)
Fabric filtration
Afterburner, direct flame.
Afterburner, catalytic.
Few moving parts.
Can be operated at high temperatures
(550? to 850°F.)
Dry collection possible.
Decrease of performance is
noticeable.
Collection of small particles
possible.
High efficiencies possible.
High removal efficiency of submicron
odor-causing particulate matter.
Simultaneous disposal of combustible
gaseous and particulate matter.
Direct disposal of nontoxic gases and
wastes to the atmosphere after
combustion.
Possible heat recovery.
Relatively small space requirement.
Simple construction.
Low maintenance.
Same as direct flame afterburner.
Compared to direct flame: reduced
fuel requirements, reduced
temperature, insulation require-
ments, and fire hazard.
Sensitivity to filtering velocity.
High-temperature gases must be cooled
to 200° to 550°F.
Affected by relative humidity
(condensation).
Susceptibility of fabric to chemical
attack.
High operational cost. Fire hazard.
Removes only combustibles.
High initial cost.
Catalysts subject to poisoning.
Catalysts require reactivation.
(Ref. D-8)
-------
facility and operating costs are those required to run the facility and
replace worn out equipment. These costs are a function of many direct and
indirect variables as discussed in a separate subsection. For a specific
air pollution source, an analysis should include evaluation of all these
variables.
In this report, four types of cost curves were developed as a comparison
of various types of air pollution control equipment:
• Purchase cost
• Installed cost
• Annual operating cost
• Total annualized costs
These cost curves represent the average or typical situation costs that are
generally encountered and should only be utilized as guides for estimating.
Development of the cost curves was accomplished by a survey of existing
literature and data, updating cost figures found to 1976, and cross checking
with known data points. For a particular case, an engineering study of the
actual emission source should be made, thereby generating data by which a
manufacturer of air pollution control equipment can quote purchase and
installed costs and perhaps give a rough estimate of operating costs. It
should be kept in mind that the volume of gas to be cleaned is probably the
single most important factor in determining the cost of an air pollution
control device. System design and process control are other important
factors to be considered. Table D-8 illustrates some conditions which could
affect purchase and installation costs.
Capital costs can usually be split 50/50 between equipment purchase
costs and installation costs. The components of capital cost shown as an
average percent of the capital investment is shown in Table D-9.
An additional 15 - 20% of total capital costs represent the cost of
startup, working capital, capitalized interest, etc.
Purchase cost curves for various type collectors are shown in Fig-
ures D-8 through D-12. Purchase costs are the amounts charged by the manu-
facturer for equipment of standard construction materials. Special fabrica-
tion of equipment to accommodate the particular characteristics of the gas
stream can increase the cost dramatically.
Installed costs (Figures D-13 through D-18) include the purchase
cost, with additional costs being incurred by 1) erection, 2) insulation,
3) freight costs, 4) site preparation, 5) treatment systems (for wet scrub-
bers) and 6) auxiliary equipment (fans, duct work, motors, control instru-
mentation, etc). The installation costs have been assumed to follow those
percentages of purchase costs shown in Table D-10.
D-20
-------
TABLE D-8. CONDITIONS AFFECTING PURCHASE AND INSTALLATION COSTS
Cost Category
Low to Typical Costs
High to Extreme High Costs
Equipment transportation
Plant age
Available space
Instrumentation
Degree of assembly
Degree of engineering
design
Utilities
Collected waste material
handling
Labor
Auxiliary equipment
Corrosiveness
Complexity of start-up
Minimum distance; simple loading and
unloading procedures
Hardware designed into new plant as
an integral part of process
Vacant area for location of control
system
Little required
Guarantee .on performance None required
Control hardware shipped completely
assembled
Standard "package type" control
system
Electricity, water, waste disposal
facilities readily available
No special treatment facilities or
handling required
Low wages in geographical area
Simple draft fan; minimal ductwork
Noncorrosive gas
Simple start-up, no extensive
adjustment required.
Extensive distance; complex procedure
for loading and unloading
Hardware installed into confines of
old plant requiring structural or
process modification or alteration
Little vacant space; extensive steel
support construction and site prep-
aration required
Complex instrumentation required to
assure reliability of control or
constant monitoring of gas stream
Guaranteed high collector efficiency
to meet stringent control
requirements
Control hardware to be assembled and
erected in the field
Control system requiring extensive
integration with process, insulation
to correct temperature and moisture
problem, noise abatement
Electrical and waste treatment facili-
ties must be expanded; water supply
must be developed or expanded
Special treatment facilities and/or
handling required
Overtime and/or high-wage geographical
area
Extensive cooling equipment; ductwork;
large motors
Acidic emissions requiring high alloy
accessory equipment using special
handling and construction techniques
Requires extensive adjustments; test-
ing; considerable downtime.
-------
1,000
100
o
Q
UJ
V)
<
o
cc
Q-
10
I I I I I I I I
I
J I
I I
10 100
GAS VOLUME THROUGH COLLECTOR (m3 /S)
1,000
Figure D-8. Estimated purchase cost of mechanical collectors (1976 dollars).
D-22
-------
10,000
1.000J-
tn
cc
o
Q
P)
O
-------
1.000
o
Q
a)
O
O
cc
0.
10 100
GAS VOLUME THROUGH COLLECTOR (m3 /S)
1,000
Figure D-10. Estimated purchase cost of fabric filters (1976 dollars).
D-24
-------
10,000
1,000
O)
IT
o
Q
V)
O
O
UJ
a>
<
O
cc
a.
10
J I I I I I I I I
1 I I I I
10 100
GAS VOLUME THROUGH COLLECTOR (m3 /S)
1,000
Figure D-ll.
Estimated purchase cost of electrostatic precipitator
(1976 dollars).
D-25
-------
1,000
100
en
£E
o
a
te
o
(J
O
rr
10
I I I I I I I
I
I 111
I I
1 10
GAS VOLUME THROUGH COLLECTOR (m3 /S)
100
Figure D-12. Estimated purchase cost of afterburners (1976 dollars).
D-26
-------
1,000
100 }-
tf>
cc
o
D
8
u
<
10 100
GAS VOLUME THROUGH COLLECTOR (m3 /S)
1,000
Figure D-13. Estimated installed cost of mechanical collectors (1976 dollars)
D-27
-------
10,000
1,000
_
o
D
CO
8
1—I—I I I I I
n—r
_L
J L
J L
10 100
GAS VOLUME THROUGH COLLECTOR (m3 /S)
1,000
Figure D-14. Estimated installed cost of wet scrubbers (1976 dollars].
D-28
-------
10,000
1,000
cc
<
o
Q
u>
8
a
UJ
100
10
i iii
i I i
I 1 I I I I
10 100
GAS VOLUME THROUGH COLLECTOR (m3 /S)
1,000
Figure D-15. Estimated installed cost of fabric filters (1976 dollars).
D-29
-------
10,000
1.000
o
Q
V)
8
a
HI
100
10
J I
100
GAS VOLUME THROUGH COLLECTOR (m3 /S)
I I I I
1,000
Figure D-16. Estimated installed cost of electrostatic precipitator (1976 dollars)
D-30
-------
i.ooo
100
in
oc
o
a
a
LU
10
I i i i i i i i i
_L
i i i i i i i
1 10
GAS VOLUME THROUGH COLLECTOR (m3 /S)
100
Figure D-17. Estimated installed cost of afterburners (1976 dollars).
o-si
-------
1,600
1,400
1,200
1,100
1.000
~ 900
cc
O 800
Q
700
600
400
300
200
100
J . [_
30 60 90 120 150 180 210 240 270 300
STACK HEIGHT (METERS)
Figure D-18. Estimated installed cost of stacks (1976 dollars)
D-32
-------
TABLE D-9. COMPONENTS OF CAPITAL COST (Ref. D-10)
Component
% of Capital Cost
Major control equipment
Auxiliary or accessory equipment
Fiel4 installation equipment
Project management and engineering
Indirect costs (freight, taxes, contractor overhead,
contractor profit)
35%
15%
20%
17%
13%
100%
TABLE D-10. TOTAL INSTALLATION COST FOR VARIOUS TYPES OF CONTROL
DEVICES EXPRESSED AS A PERCENTAGE OF PURCHASE COSTS
(Ref. D-4, D-10)
Equipment Type
Gravitational
Dry centrifugal
Wet collector:
Low, medium energy
High energy*
Electrostatic precipitators
Fabric filters
Afterburners
Low
33
35
50
100
40
50
10
Cost,
Typical
67
50
100
200
70
75
25
percent
High
100
100
200
400
100
100
100
Extreme High
—
400
400
500
400
400
400
High-energy wet collectors usually require more expensive
fans and motors.
Operation and maintenance costs are frequently a substantial percentage
of the total air pollution cost and are usually the most difficult to assess.
The annual operating cost is defined to be the expenditure incurred in
operating a control device at its designed collection efficiency for one
year while maintenance cost is the expenditure required to sustain,the
operation of a control device at its efficiency. Operating and maintenance
costs per year (including maintenance) are shown in Figures D-19 through
D-33
-------
D-23, for various pollution control devices. Annual operating and mainte-
nance costs vary considerably and are dependent on volume of gas cleaned;
pressure drop across the system; operating time; utility costs for electric-
ity, fuel, water, and other raw materials; mechanical efficiencies of
motors, fans, and pumps; and geographical location.
Annualized operating cost is the actual cost per year to operate air
pollution control equipment and includes the depreciation of the capital
investment over the expected life of the equipment, interest rate, and opera-
tion and maintenance costs.
A breakdown of typical annualized operating costs expressed as percent
annual operating costs is shown in Table D-ll. The annualized capital cost
assumptions utilized to develop Figures D-24 through D-28 are:
1) Purchase and installation costs are depreciated over 20 years
2) Straight line method of depreciation (5.0 percent per year)
3) Interest rate of 8.5 percent of the capital investment.
D-3'4
-------
1,000
cc
LU
in
cc
O
Q
fe
8
UJ
u
I-
<
Q
<
-------
1,000
1.000
GAS VOLUME THROUGH COLLECTOR (mj /S)
Figure D-20.
Estimated operating and maintenance cost of wet scrubber
(1976 dollars).
D-36
-------
1.000
cc
ui
o
a
"o
8
10
a.
O
I I T
L I ' I ' I i
10 100
GAS VOLUME THROUGH COLLECTOR (m3 /S)
1,000
Figure D-21.
Estimated operating and maintenance cost of fabric filters
(1976 dollars).
D-37
-------
1.000
cc
UI
cc
cc
4
100
u
z
I-
z
Q
<
a.
o
i i r i i i i
i lit
i i i i i i i
i i i i i i i
1 10
GAS VOLUME THROUGH COLLECTOR (m3 /S)
100
Figure D-22.
Estimated operating and maintenance cost of afterburners
(1976 dollars)
D-38
-------
100
cc
>
cc
IT
<
O
Q
fe
O
u
til
O
O
cc
Ul
1,000
GAS VOLUME THROUGH COLLECTOR
Figure D-23.
Estimated operating and maintenance
precipitator (1976 dollars)
cost of electrostatic
D-39
-------
TABLE D-ll. TYPICAL ANNUALIZED OPERATING COST BREAKDOWN
% of Total Annual
Type Operating Cost
Fixed costs 20%
• Interest, taxes, and insurance (7-12% per
year of capital investment)
» Rent (8-10% per year of total installed
cost, 15-20 years)
• Depreciation (5-10% per year of total
installed cost, 15-20 years)
• Research and development (5% of sales)
Direct Production Costs 60%
• Raw materials
• Operating, supervising and clerical
• Maintenance and repair (M/R) (2-10% of
total capital investment)
• Operating supplies and payroll (10-20%
of M/R)
• Power and utilities (10-20% of produc-
tion costs)
General Plant Overhead 10%
General Administration and office overhead 6%
Contingencies 4%
100%
D-40
-------
1000
v>
CC
o 100
a
8
o
DC
UJ
a.
O
Q
UJ
N
I I
I I I I I I i
I I I I I I
10 100
GAS VOLUME THROUGH COLLECTOR (m3 /S)
1000
Figure D-24.
Estimated annualized operating cost of mechanical collectors
(1976 dollars).
D-41
-------
1,000
10 100
GAS VOLUME THROUGH COLLECTOR (m3 /S)
1,000
Figure D-25.
Estimated annualized operating cost of wet scrubbers
(1976 dollars].
D-42
-------
1.000
Ui
cc
o
D
m
o
8
o
cc
UJ
a.
O
Q
I
100
J L
I I I I I
\ \ \ \ \
J - 1
U..
10 100
GAS VOLUME THROUGH COLLECTOR (m3 /S)
1,000
Figure D-26. Estimated annualized operating cost of fabric filters
(1976 dollars).
D-43
-------
1—I—I I I I 11
1—I—I I I I I I
cc
<
O
Q
en
O
O
(3
CC
UJ
Q.
O
Q
UJ
N
2
J_
I I I I I
10 100
GAS VOLUME THROUGH COLLECTOR (m3 /S)
1,000
Figure D-27. Estimated annualized operating cost of
electrostatic precipitator (1976 dollars).
D-44
-------
1,000
cc
<
o
Q
o
o
o
Q
D
Z
<
-DIRECT FLAME WITH
HEAT EXCHANGER
CATALYTIC WITH
HEAT EXCHANGER
1 10
GAS VOLUME THROUGH COLLECTOR (m3 /S)
100
Figure D-28. Estimated annualized operating cost of afterburners
(1976 dollars).
D-45
-------
WATER POLLUTION
Introduction
A short discussion of the water pollution considerations pertaining to
waste-to-energy conversion systems follows, that encompasses:
• The water pollution problem
• Effluent guidelines
• Typical treatment processes
• Cost considerations
More detailed information can be obtained by examination of the associ-
ated references and by consulting manufacturers of water pollution control
equipment.
Water Pollution Problem
Contaminated waters can occur in waste-to-energy conversion systems
from (1) the spentscrubber liquid of air pollution control equipment, (2)
internal process clean-up systems, or (3) leachate from landfilled residues.
In all three cases control measures must be taken to reduce the level of the
contamination to acceptable levels.
These contaminants can include partially burnt and oxidized material
from the spent scrubber liquid; slag, char, and oils from the various pro-
cesses of waste-to-energy conversion systems; and heat from quench water.
More specifically, the wastewater contamination can consist of suspended and
dissolved solids, chlorides, sulfates, phosphates, hardness, alkalinity, and
pH. In many cases the wastewater discharges will exert an oxygen demand,
both chemical oxygen demand (COD) and biological oxygen demand (BOD) and may
also contain heavy metals.*
Once a conversion system has been analyzed as to what the wastewater
characteristics will probably consist of, the process of choosing control
equipment to meet effluent guidelines and water quality standards can begin.
Effluent Guidelines
The issuance of permits to "point source" dischargers is the basic
regulatory mechanism of water pollution control. These permits are issued,
nationally, by either the Environmental Protection Agency or the state in
which the source is located. Local permits are also a usual requirement for
industrial wastewater dischargers.
For the seven candidate systems chosen for analysis, the wastewater
characteristics are discussed, where known, and the reader is referred
to them for more specific details.
D-46
-------
Effluent guidelines and water quality standards are not available for
all industries, with waste-to-energy systems being included in this category.
When effluent guidelines have not been published, permits for industrial dis-
chargers are based on the best technical judgement of feasible control tech-
nology. Where water quality standards are not available, effluent guidelines
with maximum daily pollutant loads are the basis for the permit. (Ref. D-ll)
Section 301, part (2) (A) of the Federal Water Pollution Control Act states:
"...not later than July 1, 1983, effluent limitations for categories
and classes of point sources, other than publicly owned treatment
works, which (i) shall require application of the best available
technology economically achievable for such category or class, which
will result in reasonable further progress toward the national goal
of eliminating the discharge of all pollutants, as determined in
accordance with regulations issued by the Adminstrator pursuant to
section 304 (b) (2) of this Act, which such effluent limitations
shall require the elimination of discharges of all pollutants if the
Adminstrator finds, on the basis of information available to him
(including information developed pursuant to section 315), that such
elimination is technologically and economically achievable for a
category or class of point sources as determined in accordance with
regulations issued by the Administrator pursuant to section 304 (b)(2)
of this Act, or (ii) in the case of the introduction of pollutant
into a publicly owned treatment works which meets the requirements
of subparagraph (b) of this paragraph, shall require compliance with
any applicable pre-treatment requirements and any other requirement
under section 307 of this Act;..," (Ref. D-12)
Once effluent guidelines and/or water quality standards are known, the
process of choosing a wastewater treatment system to meet them can continue,
with the next step being to identify some possible methods of control.
Typical Treatment Processes
Contaminants in waste water are removed by physical, chemical, and
biological means. Treatment in which the application of physical forces
predominates are called unit operations. Sedimentation is an example of
this type of treatment. Treatment methods in which the removal of contam-
inants is accomplished by the addition of of chemicals or by biological
activity are called unit processes. Precipitation is an example of a chem-
ical unit process and biological oxidation is an example of a biological
unit process.
Activated Sludge--
In the activated sludge process, the waste is stabilized biologically
in some type of holding vessel under aerobic conditions. Diffused or mechan-
ical aeration is used to achieve this aerobic environment. After the waste
is treated in the holding vessel, the resulting biological material is
separated from the liquid in a settling tank. This type of biological
oxidation of wastewater has proven to be one of the most effective methods
for the treatment of both municipal and organic industrial wastes. There are
D-47
-------
various modifications to the activated sludge process, descriptions of which
can be found in the literature.
Wet Oxidation--
Wet oxidation is a physiochemical method for the treatment of waste water
containing oxidizable impurities, consisting of flameless combustion at ele-
vated temperatures of 150 - 350°C (300 - 650°F) and high pressures of 3.1 to
17.2MPa (450 - 2500 psig). Compressed air is fed into the pressure vessel as
the source of oxygen for the reaction process, which consists of hydrolysis
(solubilization of solids and breakdown of long-chain hydrocarbons), mass
transfer (oxygen into solution), and chemical oxidation.
Carbon Absorption--
The absorption process consists of the collection of substances that are
in solution on a suitable interface, typically activated carbon. The effluent
is passed through a bed of activated carbon granules and up to 98% of the
impurities are removed from the water by absorption when sufficient contact
time is provided for this process. The carbon system usually consists of a
number of columns of basins used as contactors and these are connected to a
regeneration system. This process has become more attractive in recent years
due to the development of economical regenerative methods.
Lagoons and Stabilization Ponds--
The two basic types of lagoons are 1) aerated and 2) aerobic-anaerobic.
In an aerated lagoon, the essential function is waste conversion whereby the
BOD content of the effluent is reduced with the contents of the lagoon being
completely mixed and oxygenated by surface aerators or diffusers. Both the
incoming solids and the biological solids do not settle out and therefore
must be removed in a settling basin or tank before discharging the effluent.
In the case of an aerobic-anaerobic lagoon, the contents are not completely
mixed by the aerators and a significant portion of the incoming solids and
waste conversion solids settle to the bottom of the lagoon, eventually pro-
ducing an anaerobic condition at the bottom.
Stabilization ponds are bodies of water, relatively shallow, in which
wastewater is treated. Stabilization ponds (or oxidation ponds) have a
longer detention time than lagoons but are similar in principle. Classifica-
tion is usually by type of biological activity: aerobic, aerobic-anaerobic,
and anaerobic. An aerobic pond is especially suited to strong organic wastes
and brings about rapid stabilization.
Other Methods-- •-.
Since the wastewater effluents from waste-to-energy systems are so
diversified, some having high BOD and COD loadings, other chemical or physi-
cal procedures not usually considered in wastewater treatment could be
attempted; examples are reduced pressure distillation, absorption on non-
ionic resins or molecular sieves, and liquid-liquid extraction, but further
research is needed.
D-48
-------
Leachate Control in Landfilling--
Leaching can be defined as "removal by the action of a percolating
liquid. ' Leachate production in landfills is related to the amount of water
moving through the landfill and is a potential chemical and biological
pollutant of both ground and surface waters. The primary design limitation
for leachate is that the landfill not hinder any current or projected use of
the water resources in the area. Utilization of so-called Class I* disposal
sites would afford excellent protection from leachate emanating from waste-to-
energy conversion systems residue that is landfilled.
More detailed information on the abovementioned treatment processes and
others can be found in the literature.
Cost Considerations
The cost of treating wastewaters from waste-to-energy conversion system
is difficult to assess due to the many available treatment processes and the
widely varied quantities and qualities of the wastewater discharged. In
general, highly mechanized systems tend to have low annual operational costs,
but high initial costs. Conversely, less sophisticated systems might be
built at lower initial costs, but have higher annual operational costs. Once
installed, facilities can incur annual costs over a 20 year period amounting
to five times the initial cost of the facilities.
The use of municipal facilities by industry is subject to the 1972
Amendments to the Federal Water Pollution Control Act which identify such
specific requirements for public treatment of industrial waste as:
• Industrial plants discharging pollutants not susceptible to
treatment by the municipal plants will be required to pretreat
their discharges.
• The costs of providing additional plant capacity for treating
industrial wastes are not eligible for Federal grant funding.
• Industries must pay fairly for treatment services rendered,
including the costs of interceptor and collector services (Ref. D-14).
In general, construction costs for industrial wastewater facilities can
be grossly estimated by assuming approximately $30-35/m3-d treated (Ref. D-15),
ACOUSTIC CONSIDERATIONS
There are two primary considerations in dealing with noise for any
facility. They are employee noise exposure, and the impact that the
* Sites located on nonwater-bearing rocks or underlain by isolated bodies of
unusable ground water, which are protected from surface runoff and where
surface drainage can be restricted to the site or discharged to a suitable
wasteway, and where safe limitations exist with respect to the potential
radius of percolation. (Ref. D-13)
D-49
-------
additional noise caused directly or indirectly by the facility will have upon
the community at large.
Employee noise exposure is regulated by the Department of Labor Occupa-
tional Noise Exposure Standard (OSHA) . The Code of this federal regulation
is: Title 29, Chapter XVII, Part 1910, Subpart G, 36FR 10466, May 29, 1971.
The basic constraints dictated by this standard are as follows:
Permissible Noise Exposure
Duration per Day, Sound Level, dBA*
Hours Slow Response
8 90
6 92
4 95
3 97
2 100
1-1/2 102
1 105
1/2 110
1/4 or less 115
* dBA is the abbreviation for the sound energy level in decibels as measured
by the A- weighted network of a sound meter; this scale approximates the
weighted response of a normal human ear.
When the daily noise exposure is composed of two or more periods of
noise exposure, of different levels, the combined effect should be considered,
rather than the individual effects of each. If the sum of the following
exceeds unity, then, the mixed exposure should be considered to exceed the
permissible limit.
T T T
1 2 n
Where: C = The total time of exposure at a specific noise level.
T = The total time of exposure permitted at that level.
D-50
-------
Exposure to impulse or impact noise should not exceed 140 dBA peak sound
pressure level.
Community noise levels are typically regulated by state and local
agencies, with standards varying from area to area. They also depend on the
location of the proposed facility with regard to zoning restrictions and the
proximity of residential areas.
There are several potential noise problem areas in waste processing
facilities. Some major equipment-related noise sources are: delivery
vehicles, waste moving vehicles, screens, compressors, compactors, shredders,
fans, pumps, furnaces, cooling towers, conveyors, and atmospheric vents.
Other less tangible sources are dumping and receiving bins, sorting bins,
and other areas in which there exist impulse noises from metal to metal,
rock to metal, or solid to metal contact.
Vehicular noise can be reduced by using rubber-tired tractors equipped
with mufflers on their exhaust systems. Noise can be reduced by buying
quieter equipment initially and by acoustically treating existing noisy
equipment by enclosures or lagging. Impulse noise can be reduced by applying
rubber or plastic lining to bins, truck beds, and like surface areas. Com-
bustion roar from furnaces can be reduced through the addition of acoustic
plenums for combustion air intakes. Noise caused by atmospheric vents can be
reduced by fitting the vents with diffusers or silencers.
Finally, noise areas in the facility can be posted for limited access
and thus, by administrative control, the employer can insure that an employee
-is not exposed to excessively loud noises that will present a hearing damage
risk.
If the noise level at the plant boundary still exceeds local ordinances
after initial design and equipment selection efforts, further more refined
scrutiny of noise sources within the facility and further acoustic treatment
would be required. This could be a very costly endeavor and consequently
maximum attention should be employed in site selection and facility design
to insure compliance with all noise regulations prior to construction.
The increase of vehicular traffic on residential surface streets must
also be considered. If there will be an increase of traffic, the annoyance
to the community may be significant. When it is necessary to traverse resi-
dential areas, several approach routes to the facility should be employed,
reducing the impact on any one section of the community. The use of residen-
tial surface streets should be avoided whenever possible because of the
usually very stringent regulations imposed in such areas. These regulations
are stated in terms of L, , the day-night level of the average A-weighted
noise level integrated over a 24-hour period. A 10 dB penalty is assessed
from the hours of 10:00 p.m. to 7:00 a.m. If a local ordinance states that
the maximum permissible L, is 55, a facility may not emit more than an
average of 55 dBA from 7:00 a.m. to 10:00 p.m., and an average of 45 dBA from
10:00 p.m. to" 7:00 a.m.; furthermore, a maximum allowable sound level is
usually imposed.
D-51
-------
REFERENCES
D-l. Bishop, C.A. "EJC Policy Statement on Air Pollution and its Control,"
Chem. Eng. Progr. 53, No. 11, 146-152 (1957).
D-2. Liptak, B.C., Environmental Engineers Handbook - Vol. 2, Air Pollution,
1st. Ed., Chilton Book Co., Radnor, Penna., 1974, p. 210.
D-3. Environmental Reporter Federal Regulators, "EPA Regulations on Standards
of Performance for New Stationary Sources (40 CFR-60)-," 1976.
D-4. Control Techniques for Particulate Air Pollutants, AP-51, U.S. Dept. of
Health, Education, and Welfare, National Air Pollution Control Adminis-
tration, Washington, D.C., January, 1969.
D-5. Lund, H.F., Industrial Pollution Control Handbook, 1st Ed., McGraw-Hill
Book Co., New York, N.Y., 1971, p. 23-1.
D-6. Analysis of Pollution Control Costs, EPA-670/2-74-009, U.S. Environ-
mental Protection Agency, Washington D.C., February, 1974.
D-7. Industrial Ventilation, American Conference of Governmental Industrial
Industrial Hygienists, Edward Bros, Inc., 1974.
D-8 Kerbec, Mathew J., Your Government and the Environment, An Annual
Reference: Vol. I, Arlington, Va., Output Systems Corp., 1971.
D-9. Hesketh, H.E., Understanding and Controlling Air Pollution, Ann Arbor
Science Publications, Ann Arbor, Michigan, 1973.
D-10. Edmisten, Norman G., "A Systematic Procedure for Determining the Cost
of Controlling Particulate Emissions From Industrial Sources," Air
Pollution Control Association Journal, Vol. 2Q, No. 7, July, 1970.
D-ll. EPA, Water Quality Strategy Paper, March 15, 1974, p. 9.
D-12. Environmental Reporter - Federal Laws, "Federal Water Pollution Control
Act and Amendments," 1976.
D-13. Sanitary Landfill, Army Construction Engineering Research Laboratory
National Technical Information Service, AD-773-714, January, 1974
p. 19.
D-14. The Economics of Clean Water - 1973 U.S. Environmental Protection Agency,
Washington, D.C., 1973, p. 49.
D-15. Tihansky, Dennis P., "Historical Development of Water Pollution Control
Cost Functions", Journal of the Water Pollution Control Federation,
Vol. 46, No. 5, May, 1974, pp. 813-833.
D-52
-------
APPENDIX E
SI UNITS OF MEASUREMENT
"SI" is the offical abbreviation, in all languages, for the International
System of Units (le Systeme International d'Unites). It is an improved, co-
herent version of the metric system and has been recognized as the basic system
of measurement by virtually all nations since its adoption in I960- The Inter-
national System has seven basic units from which most of the others can be
derived:
Quantity
Length
Time
Mass
Unit
metre
second
kilogram
Multiply byx
3.048 000*E-01
Amount o f
substance mole
Temperature kelvin
Electric
current
Luminous
intensity
ampere
candela
Symbol To convert from to_
m foot metre
s
kg pound, kilogram 4.535 924 E-01
c
mol
°K
A
cd
Two purely geometrical supplementary units are also used: The radian (rad)
for measuring plane angles, and the steradian (sr) for measuring solid angles.
XThis system of mathematical notation is that basically adopted for computer
operations. An asterisk following the conversion factor indicates an exact
relation, i.e. all numbers to follow are zeros. All factors are written as
a number'greater than one and less than ten, and are followed by the letter
E (for exponent), a plus or minus symbol, and two digits that indicate the
power of ten by which the number must be multiplied to obtain the correct value.
For example 3 523 907 E-02 is 0.035 239 07 while 3.386 389 E+03 is 3 386.389.
E-l
-------
Derived units are expressed as algebraic combinations of the basic and
supplementary units, and may have special names. Some important derived
units are:
Quantity Unit
Symbol
To convert
from
Force
Energy
pound-
newton N = kg-m/s2 force
joule J = N-m
Btu
(International]
Power
Pressure
Heating
value
watt W = J/s
pascal Pa = N/m
joule/
kilogram J/kg
hp
(electric)
psi
Btu/lb
to Multiply by
newton 4.448 222 E+00
joule 1.055 056 E+03
watt 7.460 000*E+02
pascal 6.894 757 E+03
MJ/kg 2.326 000*E-03
There are a number of other units which are not part of the SI but are
nevertheless important and widely used, and therefore considered to be allow-
able units. Examples are the minute (min.), hour (h), day (d), and year (y)
as measures of time; degrees (°) , minutes ('), and seconds ("") as measures of
angles; and degrees Celsius (°C) for temperatures. Trivial names that formed
a part of the old metric system should be avoided. Principal among these are
the litre (which to be consistent with other volumetric units should be called
a cubic decimetre, dm^); the metric tonne, which is properly referred to as
a megagram (Mg) throughout this report; and the micron, correctly now a micro-
metre, /Ltm.
Two important conventions are used in the International System:
• Groups of three digits are separated by a space instead of a comma,
although with four digits the spacing is optional
• Decimal multiples and submultiples of SI units are indicated by appro-
priate letter prefixes, although in certain instances scientific no-
tation may be used:
Scientific
Notation
1015
ID*2
109
106
103
Prefix
peta
tera
giga
mega
kilo
Symbol
P
T
G
M
k
Scientific
Notation
10-
10-
10-
10-
10-
3
6
9
12
15
Prefix
milli
micro
nano
pico
feint o
Symbol
m
M
n
P
f
E-2
-------
Conversions from U.S. customary units to SI must be carried to a suffi-
cient number of digits to keep the accuracy of the original quantity. In
dealing with nominal values, such as "2-inch11 pipe, exact conversions are
used to avoid ambiguities.
The authority followed here for application of SI is American National
Standard Z 210.1, available through the American Society for Testing and
Materials as ASTM E380-76, "Standard for Metric Practice," approved January
19, 1976. An additional guide to the system is "Conversion of Operation and
Process Measurement Units to the Metric (SI) System," American Petroleum In-
stitute Publication 2564, March 1974.
Other conversions used within the report are as follows:
To convert from
acre
atmosphere, standard
barrel, 42-gal
Btu/ft3
Btu/SCF (60°F, 1 atm)
ft
ft3 (standard)
ft3/min
gallon, U.S.
gallon, U.S./min.
grain/SCF
kcal, International
lb/ft3
Ib/gal U.S.
mile
ton, short
to
m2
Pa
m3
MJ/m3
MJ/Nm3
(0°C, 1 atm)
m2
m
3 (normal)
m
m
m3/s
g/Nm3
J
kg/m3
kg/m3
km
Mg
Multiply by
4.046 873 E+03
1.013 250*E+05
1.589 873*E-01
3.725 895 E-02
3.938 080 E-02
9.290 304*E-02
2.831 685 E-02
2.679 102 E-02
4.719 474 E-04
3.785 412 E-03
6.309 020 E-05
2.418 680 E+00
4.186 800*E+03
1.601 846 E+01
1.158 264 E+02
1.609 347 E+00
9.071 847 E-01
E-3
-------
TECHNICAL REPORT DATA .
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-600/7-78-086
3. RECIPIENT'S ACCESSION"NO.
4. TITLE AND SUBTITLE
ENGINEERING AND ECONOMIC ANALYSIS OF
WASTE TO ENERGY SYSTEMS
5. REPORT DATE
May 1978 issuing .date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
E. Milton Wilson, John M. Leavens,
Nathan W. Snyder, John J. Brehany and Richard F. Whitman
8. PERFORMING ORGANIZATION REPORT NO
5495-1
9. PERFORMING ORGANIZATION NAME AND ADDRESS
The Ralph M. Parsons Company
Systems Division
100 W. Walnut Street
Pasadena,. Ca 91124
10. PROGRAM ELEMEN1
EHE 624B
NO.
11. CONTRACT/GRANT NO.
68-02-2101
12. SPONSORING AGENCY NAME AND ADDRESS
Industrial Environmental Research Lab., Cinn, OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final (V75-6/77)
14. SPONSORING AGENCY CODE
EPA/600/12
15. SUPPLEMENTARY NOTES
Project Officer
Harry Freeman (513-684-^363)
16. ABSTRACT
Waste quantities and characteristics in the U.S. are reviewed and waste-to-energy con-
version technology evaluated. All waste materials, exclusive of those from mining
operations, are considered. The technology is reviewed under the categories of
mechanical processing, biological conversion systems, thermal/chemical systems, and
combustion. Important features of many operating facilities are described and
detailed engineering and economic analyses of seven specific systems are presented.
An analysis is also made of the technology and costs for conversion of pyrolytic
off-gas to methane, methanol, and ammonia. Environmental pollution data are presented
where available and the current control technology briefly reviewed. Conclusions
on the conversion technology are made and research needs considered in a series o-f
recommendations.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
COS AT I Field/Group
Wastes*, Heat Recovery*, Energy*, Refuse*,
Pyrolysis, Combustion, Anaerobic Processes,
Pollution, Steam, Agricultural Wastes,
Organic Wastes, Economic Analysis, Fossil
Fuels
Environmental Assessments
Wastes-as-Fuel
Pollution Control
Resource Recovery
Solid Waste Management
13B
10B
12A
3. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
458
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
A U.S. GOVERN KENT PRINTING OFFICE: 1978— 757-140/6853
------- |