resource recovery plant implementation
guides for
municipal officials
• planning and overview
•technologies • risks
and contracts • markets
• accounting format •
financing • procurement
• further assistance!
-------
This publication is part of a special series of reports prepared
by the U.S. Environmental Protection Agency's Office of Solid Waste
Management Programs. These reports are designed to assist municipal
officials in the planning and implementation of processing plants to
recover resources from mixed municipal solid waste.
The title of this series is Resource Recovery Plant Implementation:
Guides for Municipal Officials. The parts of the series are as follows:
1. Planning and Overview (SW-157.1)
2. Technologies (SW-157.2)
3. Markets (SW-157.3)
4. Financing (SW-157.4)
5. Procurement (SW-157.5)
6. Accounting Format (SW-157.6)
7. Risks and Contracts (SW-157.7)
8. Further Assistance (SW-157.8)
Mention of commercial products does not constitute endorsement by
the U.S. Government. Editing and technical content of this report were
the responsibilities of the Resource Recovery Division of the Office of
Solid Waste Management Programs.
Single copies of this publication are available from Solid Waste
Information, U.S. Environmental Protection Agency, Cincinnati, Ohio
45268.
-------
Resource Recovery Plant Implementation:
Guides for Municipal Officials
TECHNOLOGIES
This guide (SW-157.2) was compiled
by Steven J. Levy and H. Gregor Rigo
Acknowledgements are made to Robert Lowe for chapter II
of this report, and to Robert Holloway, David Sussman, and
Ivonne Garbe for contributions to other chapters.
U.S. ENVIRONMENTAL PROTECTION AGENCY
1976
-------
CONTENTS
SECTION PAGE
I. INTRODUCTION AND OVERVIEW 1
Overview 1
Energy Recovery Systems 3
Material Recovery Systems 9
Conclusions 11
II. GENERAL CONSIDERATIONS FOR RESOURCE
RECOVERY SYSTEM DESIGN 13
Markets for Recovered Products 13
Waste Generation (Quantity) 14
Waste Composition i 15
System Reliability 16
Plant Location 17
Land Required for the Plant Site 17
Community Acceptance 17
Plant Costs and Revenues 18
III. ENERGY RECOVERY SYSTEMS 21
Energy Balances 21
Waterwall Combustion Systems—Unprocessed Waste 22
Waterwall Combustion Systems—Processed Waste 28
Solid Refuse Derived Fuel Systems 29
Pyrolysis Systems 41
Biological Gasification Systems 55
Waste-fired Gas Turbine Systems 58
IV. MATERIALS RECOVERY SYSTEMS 62
Paper Fiber Recovery 62
Composting 67
Ferrous Metals Recovery 68
Glass and Aluminum Recovery Systems 70
V. READING LIST 79
-------
RESOURCE RECOVERY PLANT IMPLEMENTATION:
GUIDE FOR MUNICIPAL OFFICIALS
Technologies
by Steven J. Levy* and H. Gregor Rigo+
SECTION I
INTRODUCTION AND OVERVIEW
The recent emergence of techniques for converting mixed municipal
waste into marketable products has given municipal and regional
officials a variety of new options for solving their solid waste
management problems. Although these resource recovery systems cannot
be expected to operate at a profit, they are becoming increasingly
competitive with the cost of sanitary landfill ing in many areas of
the country. In addition, although they will not allow a community
to close down its landfill, the life of the landfill can be extended
tremendously by the weight and volume reductions achieved.
The purpose of this technology review is to aquaint the reader
with the available and emerging technology options for processing
of mixed municipal waste for resource recovery.
Although this report focuses only on mixed waste processing
systems to recover materials' and energy, it is important to remember
that other strategies should also be considered and integrated with
such plants for a complete resource recovery and conservation
strategy. This includes particularly examination of waste reduction
and source separation strategies. Fortunately such strategies will
usually be found to be compatible, which allows cities to maximize
recovery and minimize both waste and cost.
For each technology presented in this report the following
information is presented:
*Mr. Levy is Manager of the Demonstration and Evaluation Program
of the Resource Recovery Division, Office of Solid Waste Management
Programs, U.S. Environmental Protection Agency.
+Dr. Rigo is the Principal Engineer for Systems Technology
Corporation, Dayton, Ohio.
-------
Process description
. Product characteristics
Status of development
Energy balance
This information should help officials to determine if systems may be
available to meet their needs; and will help them understand the
technical capabilities and risks associated with various technologies.
This guide will not tell the reader which system, if any, to
select. There is no universally "best" or most economical recovery
technology. Every community facing a resource recovery decision must
consider its own unique set of factors when selecting a course of
action. Factors include available markets and local prices, capital
and operating cost projections, level of risk which they are willing
to assume*, and financing and management alternatives available
for different systems or considerations.
Unfortunately, goals often conflict, making a choice more
difficult. For example, the recovery system with the lowest projected
net cost may involve the highest degree of technological uncertainty.
Or, the system producing products which can be most readily marketed
through firm contracts may have the highest projected costs. The
final decision is subject to specific value judgements which each
community must make on its own. This should normally be done with
the assistance of knowledgeable consultants who can examine in
detail the feasibility of alternative options, including factors
such as marketing, management, and financing.
The most important factor to remember when assessing a technology
is that the system must be able to produce marketable products.
Technology selections should not be made until potential markets
have been identified and the market requirements specified. Some
communities will find that there is only one technical approach that
can simultaneously meet their needs and the requirements of their
markets. Most cities, however, will, have the flexibility to choose
from two or more technologies that meet their market requirements.
the Markets guide of this series (SW-157.3) discusses markets in
detail.
OVERVIEW
To give communities a better idea of the developmental status of various
technologies, the systems described in this report have been classified
into general categories which are defined below. Because resource
recovery systems are rapidly evolving, the categorization serves
only as a general guide rather than providing a precise definition
*Risk is a function of the total dollar investment and the degree
of uncertainty that exists. Factors subject to uncertainty include
availability of waste, reliability and performance of equipment,
product quality, and market demand.
2
-------
of status of development. Primarily, it indicates the degree and scale
of operating experience. It is important that the decision maker
realize that while the degree of operating experience is one detriment
of the technological risk inherent in recovery systems, another factor,
which may be more important, is the performance guarantee that a system
vender may supply. Thus, in a particular situation, a system with an
operating history, which is improperly designed or is secured by
no performance guarantees, may entail more risk for a city than a
newer technology designed and properly warranteed by a qualified vender.
. Commercially Operational Technology
Full scale commercial plants exist which operate continuously.
Thus, there are some operating data available from communities
and engineers already involved in the use of the process. Though
such systems are being commercially utilized, they may be
technologically complex. To operate properly, they will require
maximum use of available information leading to careful design
and operation by knowledgeable professionals. There may be only
limited operating experience with some parts of these plants.
Thus, technological uncertainties may still exist.
. Developmental Technology
Developmental technologies have been proven in pilot operations
or in related but different applications (for example, using
raw materials other than mixed municipal solid waste). There
is sufficient experience to predict full-scale system per-
formance, but such performance has not been confirmed.
System design requires considerable engineering judgment
concerning scale-up parameters and performance projections;
consequently, the level of technical and economic uncertainty
is generally greater than commercially operational technologies.
• Experimental Technology
This category includes new technologies that are still being
tested at the laboratory and pilot plant level. Insufficient
information exists to predict technical or economic viability.
Therefore, such technologies should not be considered by cities
contemplating immediate construction.
The systems described in this guide are further categorized into
energy recovery and material recovery. The two are not mutually
exclusive; most proposed resource recovery systems have aspects of both.
Energy Recovery Systems
Energy recovery technologies are classified in this report as
follows:
. Waterwall Combustion System
Commonly called "waterwall incinerators," this system involves
burning of solid waste in a specially designed furnace jacketed
-------
with water-filled tubes, and incorporating other boiler tubes to
recover heat. In most systems built to date solid waste is
burned without prior processing, on mechanical grates which
move it through the furnace. A relatively recent innovation
to this process involves the burning of shredded solid waste
using semi-suspension type spreader stokers. In both types
of systems little or no supplemental fossil fuel is used
and heat is recovered as steam which can be used directly or
can be converted to electricity.
Solid Refuse Derived Fuel (RDF) System
This designates a processing system employing size reduction and
classification of waste to produce both a combustible fraction
and a "heavies" fraction which may be processed for materials
recovery. This may be either a "wet" or a "dry" process.
These systems are also called "supplemental fuel" systems,
since the combustible fraction would typically be marketed
as a fuel to outside users e.g. utilities and industries, for
use as a supplement to coal (or possibly oil) in their existing
boilers. Some waterwall combustion systems (as mentioned
above) would also involve such a processing system, though
the waste might be shredded more coarsely, and may or may
not be classified. Similarly, some of the pyrolysis systems
may employ elements of this "front end" processing to
prepare waste for the pyrolysis reaction. In this report
the terminology "refused derived fuel (RDF) system" is used
to represent the preparation of a solid fuel to be marketed
to a utility or industry for use as a supplement to a fossil
fuel.
Pyrolysis Systems
Pyrolysis is a broad term given to a variety of processes
where either processed or unprocessed waste is decomposed
by the action of heat in an oxygen deficient atmosphere.
This results in production of combustible gases or liquids
depending on operating conditions. These products may be
either burned immediately to produce steam or, thos.e whose
quality is high enough, may be transported or stored for
use elsewhere.
Biological Conversion Systems
Biological conversion involves the decomposition of solid
waste by bacterial action to produce combustible gases. These
gases could be burned immediately to produce steam, or trans-
ported for use elsewhere if their quality is high enough.
Biological conversion occurs in landfills, and gas wells may
be used to collect the gas if conditions are correct. Alter-
natively, digestion can take place in controlled vessels.
-------
. Waste-Fired Gas Turbine
This technology involves the burning of solid waste in a special
incinerator and the use of the resulting hot gases to drive
a gas turbine for energy production (Brayton Cycle).
Typically, people tend to classify recovery systems by processing
technique, as we have done here. However, since market availability
is the key pre-requisite for selecting a system, it is valuable to
look at technologies in terms of products produced. Viewed in this
way the following array of technologies results:
Product
Steam or Electricity
generated "on-site"
for sale
Technology (Process)
Waterwall combustion system (bulk
burning)
Waterwall combustion system
(processed waste)
Pyrolysis to low Btu gas, which is
burned in afterburner
Bio-conversion to a gas, which is
burned in an afterburner
Production of a solid fuel
for use "off-site" as a
supplement to coal or oil
Production of a gas or liquid
fuel for use "off-site" as a
supplement to oil or gas
Solid Refuse Derived Fuel
system
(RDF)
Pyrolysis to a
Pyrolysis to a
Bio-conversion
gas
medium Btu
liquid
to medium Btu
gas
Table 1 combines technologies and products in a matrix and includes major
locations where implementations have occurred. The table also shows the
status of development of the technologies. Clearly, most of the systems
are still in relatively early stages of development, indicating the
presence of technological risk, and therefore, the need for cities to
proceed cautiously. Status is further discussed below.
Commercially Operational Technology. Combustion of solid waste on
mechanical grates in waterwall furnaces to recover steam is the most
thoroughly proven resource recovery technology. However, most of
the experience has been in Europe rather than the U.S. Approximately
200 of these systems have been installed in Europe, and another 50
exist in Japan, Brazil and elsewhere. Six waterwall combustion
systems are now operating in the United States, where market and'
institutional arrangements have been less attractive than other
countries.
Although this is a proven technology, some technical uncertainties
still exist. Boiler corrosion and air pollution control problems,
-------
TABLE 1
ENERGY RECOVERY TECHNOLOGY AND PRODUCTS
^^\ PRODUCT
^~\^
^\^^
TECHNOLOGY \^
WATERWALL
COMBUSTION
(MASS BURNING)
(SEMI-SUSPENSION)
SOLID FUEL (RDF)
—
PYROLYSIS
GASIFICATION
LOW BTU
MED. BTU
_
LIQUIFICATION
... - . .-_ _ -
CONVERSION
LANIDFILL
—
REACTOR
WASTE FIRED
GAS TURBINE
r—
U
CC
y
111
Ui
Used extensively in
Europe and Japan
Hempstead, N.Y. (C)
Dade Co.,Fla. (D)
St. Louis, Mo. (P.O.)
St. Louis, Mo (D)
Chicago, III (C)
Ames, Iowa (S)
Milwaukee, Wis (C)
Monroe County, N.Y
(D)
Luxembourg (C)
Possible
Possible
Los Angeles, Cal
(O.P)
Possible
Menlo Park, Cal (P)
o> «;
+3 2 —
o a> £•
< 3 a i-
UJ i. C tj
h- o S »
CO ^- +5 at
Braintree, Mass (O)
Harrisburg, Pa (O)
Norfolk, Va (0)
Chicago, III (0)
Nashville, Tenn (O)
Portsmouth, Va (C)
Saugus, Mass (S)
Montreal, Can. (O)
Quebec, Can (O)
Hamilton, Ont. (O)
Tokyo, Japan (O)
Akron, Ohio (D)
Columbus, Oh (D)
Akron, Oh (D)
Baltimore, Md (S)
Grasse:, France (C)
Possible
Possible
Possible
Possible
By-Product
ui | 1
^ | a. ,. £
— S ™ c "5
-" - = n o
8 £! fil
N/At
N/A
Los Gatos, Cal (P.O.
u_
co
O
HI
co
0
N/A
'
N/A
N/A
Bridgeport, Conn. (D)
E. Bridgewater, Mass. (S)
Palmer Twp., Pa (D)
By-Product
N/A
By-Product
N/A
By-Product
N/A
Possible
S. CharlestonlW Va
(P.O.)
N/A
Los
Angeles, Cal (O)
Phoenix, Ariz(S)
Franklin, Oh (P)
Pompano
Beach, Fla (P-D)
N/A
H.
O
5
o
Zj
N/A
N/A
,'N/A
N/A
N/A
San Diego
County,
Cal (P,C)
N/A
N/A
o
or
LLJ
0.
o
o
rr
LU
O
O
LU
5
a.
O
LU
LU
O
<
•z.
LU
cc
LU
0_
X
LU
*Operating status is designated as:
P—Pilot or Demonstration
D—Design
C—Construction
O—Operational
S-Start-up
tN/A-Not Applicable
-------
for instance, have resulted from improper design and/or operation of
some systems. The overall operating experience of waterwall incinerators
varies. Examples can be found of both very successful operations and
those that have experienced problems. Proper design and operation,
utilizing information from the most successful operations is the key
to success.
A recent variation of the standard waterwall combustion system
involves course shredding of waste followed by burning in a special
waterwall boiler. Usually, this would be a "stoker boiler" where the
course shredded waste (which may have had some metals and glass removed)
is mechanically or pneumaticly thrown into a furnace and burned on a
moving grate. There are several of these systems operating now or
scheduled for construction in the near future, but the basic boiler
technology is similar to bulk burning waterwall boilers. Also, such
boilers have been used to burn bark and other waste and are standard
coal burning technology.
Another system in a similar stage of development is the prepara-
tion and use of Refuse-Derived Fuel (RDF). This concept involves
size reduction of solid waste using either hamrnermills (dry) or
hydropulpers (wet) and removal of non-combustibles to produce a
supplementary fuel for use in coal-fired steam generators (boilers).
Although the operating experience of this technology is represented
by only one full-scale demonstration plant and one commercial facility
operational since early 1976, sufficient data have been collected
and observations made to indicate that the concept can be feasible.
RDF can be produced and, according to some electric utility officials,
experience indicates that it can be fired at rates of 5 to 20 percent
of a steam generator's heat output without measurably affecting
boiler operation and short-term maintenance requirements.
However, it should be noted that the design and operating
parameters have not yet been well defined for the most cost-effective
approach to RDF production, storage, transportation, or firing.
The technology will be optimized through the experiences of the second
and third generation plants in operation and under construction.
In addition, it should be noted that many steam generator operators
are wary of the potential adverse effects of RDF on boiler operation and
maintenance. This is understandable given the limited RDF firing
experience to date. Most potential users are concerned enough that
they require RDF firing test periods of from several months to several
years during which they will evaluate the effects of RDF firing. The
RDF purchase agreements usually allow the user to terminate the
contract if sufficient problems are found to make the RDF economically
unattractive. It is expected that as RDF firing experience is gained,
these test period requirements will be dropped from purchase contracts.
-------
Although this discussion so far has centered on fluff RDF
(shredded air classified waste), a number of alternative RDF products
and production systems are discussed in this report. There has been
relatively little experience either with the production or the firing
of these other RDF variations. Because they entail greater uncertainty,
they should be classified as "developmental." The variations include
densification of the fuel into pellets, briquettes, or cubettes for
co-firing with coal in a stoker equipped boiler; use of RDF in oil-
fired boilers, or in conventional stoker or grate equipped boilers;
.use of RDF in cement kilns; and the production of a very fine
powdered fuel (dust RDF) for use alone or slurried in oil.
Developmental Technologies. This category includes the types
of special solid refuse derived fuels described above as well as
all types of pyrolysis systems and the recovery of methane from
sanitary landfills.
Numerous pyrolysis systems are being developed. They are
classified as developmental because the processes are still in
the laboratory or pilot plant stages. However, one ,pyrolysis system
(Monsanto's Langard system) is being demonstrated at the 1,000 ton
per day scale in Baltimore, Maryland. This plant is currently
undergoing extensive modifications to correct problems which arose
during its initial start-up period. An assessment of the Systems'
availability cannot be made until the modifications are completed
and their performance is evaluated.
A 200 ton per day demonstration of the Occidental Petroleum
Corporation's pyrolysis process is under construction in San Diego,
California and will begin operations in the fall of 1976. Union
Carbide Corporation has already operated a 200 ton per day oxygen fed
pyrolysis plant in South Charleston, West Virginia. In Europe, one
200 ton per day commercial Torrax pyrolysis plant (Andco Incorporated)
is nearing completion, and two other units are soon to be started.
Several other small-scale pyrolysis systems are currently being
tested.
Communities may wish to consider some of these processes for
implementation, realizing that the technological uncertainties
generally exceed those for commercially operational technologies.
Both Andco and Union Carbide consider development and demonstration
of their respective systems to be far enough along to warrant
marketing of full size plants.
Recovery of methane from sanitary landfills is also considered
developmental because although the technology is not complicated,
it is not yet possible to evaluate the feasibility of such a system.
Long-term monitoring is necessary to determine projected yields
over extended periods of time. Also, more information is needed
to define the parameters (such as depth of fill, soil characteristics,
field moisture, etc.) necessary to predict yields and thus system
economics. Nonetheless, at least one compnay (the NRG Nufuel Company)
8
-------
has sufficient confidence in the process that they are seeking
landfill sites suitable for commercial application.
Experimental Technologies. Anaerobic digestion in controlled
reactors, and direct generation of electricity in a high pressure
gas turbine are presently being pursued at the research level.
Neither of these systems will be ready for commercial consideration
until they are first proven in operating pilot plants.
Materials Recovery Systems
Mechanical processing of mixed or partially concentrated waste
is often combined with energy recovery in comprehensive recovery
plants. However, total materials recovery systems are also possible.
Table 2 is a list of material recovery processes with some brief
notes concerning installations and products.
Commercially Operational Technology
Composting of waste has been practiced commercially in both the
United States and Europe, and thus, can be considered commercially
operational technology. Unfortunately, composting does not have
wide applicability because of the limited market for the humus
product. In the 1960's, many composting systems were built and
operated in the United States. All but the Altoona, Pennsylvania,
plant have been closed because of lack of market for the compost.
Of the unit processes for recovering materials, ferrous metals
recovery is clearly a proven technology. Ferrous metal recovery has
been demonstrated to be commercially available and economically viable.
Systems are in use to recover ferrous metal from incinerator residue,
from coarsely ground solid waste prior to disposal in a shredded waste
landfill, following shredding operations in refuse derived fuel
systems and even from raw unprocessed solid waste in areas where
a high market value and high ferrous metal content make the operation
feasible. The major concern in considering ferrous metal recovery is
to carefully define the market specifications for the project prior
to implementation of the subsystem, so that the equipment can be
designed to extract a marketable product.
Wet processing for fiber recovery is a patented process of the
Black-Clawson Company. This process has been demonstrated in only
on instance to date. A 150 ton per day plant has operated success-
fully on a daily basis in Franklin, Ohio, for approximately five
years. The plant produces a low grade paper fiber which is used
by a roofing felt manufacturer located near the Franklin plant.
However, markets for this low-grade fiber are limited. Despite the
relatively small scale and singularity of the Franklin demonstration,
the technical success of the project suggests that it is near to
-------
TABLE 2.
MATERIALS RECOVERY SYSTEMS LOCATIONS AND PRODUCTS
Product
0
0
\ iy(je
System or Subsystem ^v
Type \.
l~~ Fiber Recovery
J < Wet Separation
*f. "Z-
Q O Dry Separation
rr F
^ o= Composting
^ LU
1°
L Magnetic Separation
!~ Aluminum Recovery
< Wet Processing
[Jj Dry processing
^
CL
o
LU
LU
Q
L_ Glass Recovery
-a *. « £ .2
° § O 1 oj § a
"y -D.E^=5 o is
ir CDP*~OLl U3
5 X^^U.^ := D-
O — — fl5 O CO
tj S < u- C/D a.
Franklin, Oh (O) *
Rome, Italy (0)
Altoona, Pa (O)
others were not
financially viable
St. Louis, Mo (P,O)
Columbus, Oh (O)
Charleston, SC (O)
Atlanta, Ga (O)
San Diego County, Cal (C)
Plus Many Others
rranklin, Oh (P)
Ames, Iowa (S)
Milwaukee, Wis (C)
Monroe Co, NY (D)
New Orleans, La (D)
San Diego County, Cal (C)
New Orleans, La (D)
San Diego County, Cal (C)
Franklin, Oh (P)
""operating status is designated as:
P—Pilot
D—Design
C—Construction
O—Operational
S—Start-up
10
-------
to being commercially operational. A large-scale plant utilizing
the wet pulping technology is scheduled for construction by Black
Clawson in Hempstead, New York. The pulp, however, will be used as
a fuel rather than as a paper fiber.
Developmental Technology. Processes to recover aluminum and
glass must still be considered developmental. A considerable amount
of pilot work is presently underway, and economically viable systems
may soon be available. This will occur through a combination of
optimization of recovery equipment to produce purer materials and
a lessening of industry product specifications as more experience is
obtained in using these materials recovered from municipal solid
waste.
Processes to recover glass and aluminum usually operate on a
pre-concentrated materials stream rich in metals and glass. The
final concentration step in glass recovery technology has focussed
on two basic approaches: froth floatation, which produces a very
pure, small particle, non-color-sorted product; and optical
sorting, which produces a large particle, color-sorted product.
The availability of froth floatation must await its full-scale
demonstration (the first such plant will begin operations in San
Diego County in late 1976) in order to test the economic and technical
viability of the process and in order to produce a sufficient
quantity of glass to test its marketability.
Optical sorting has been demonstrated at Franklin, Ohio by
EPA and the Glass Container Manufacturing Institute. To date,
however, the equipment has not been able to economically produce a
product which can consistently meet the specifications of the glass
industry for ceramic contaminants. Newer, more efficient equipment
will soon be tested. In addition, some manufacturers have begun
to show a willingness to accept glass which does not fully meet the
industry specification.
The most promising final concentration method for recovering
aluminum from solid waste appears to be electromagnetic devices,
referred to as "aluminum magnets." Three companies are currently
developing these devices. The first full-size system to use such
a device is in Ames, Iowa. As of this writing, the equipment was
still going through normal start-up procedures.
Conclusions
Assessing and selecting resource recovery technologies is a
complicated endeavor which the City should undertake with the
assistance of adequate in-house or consulting expertise. A pro-
curement of a recovery system involves many non-technological issues
such as procurement method, management, financing and risk sharing.
11
-------
These issues are discussed in other sections of this series entitled
Resource Recovery Plant Implementation: A Guide for Municipal Officials.
(SW-157.1 to SW-157.8)
However, understanding the basic capabilities and status of
development of technologies is an important link in the implementation
chain. This report presents a background of such information for the
municipal official. Clearly, the local markets and other circumstances
surrounding each situation will influence the attractiveness and
suitability of various technologies. In short, it is difficult to
evaluate technologies in a vacuum. However, this report is intended
to provide factual basic information on various technologies as an
aid to municipal decision-making.
The categorizing of technologies into various stages of development
in this report is an attempt to give cities a rough idea of how much
experience there is with various technologies. However, it is only
a general guide.
As a whole, resource recovery technologies are still in a relatively
early stage of development and entail risks. Such risks include the
possibility that equipment will not perform as designed, that products
will not meet market specifications, and that consequently a city will
suffer an economic penalty. This penalty could range from a major
capital loss for a plant that will not function properly and must
be "written off," to an increase of a few dollars in net costs for
additional operating expenses.
On the other hand, there is sufficient experience with some
technologies and promising early results from some developmental
technologies that the risks involved may not be unreasonable. This
is particularly true of plants designed and backed by knowledgeable,
experienced companies.
Costs of resource recovery are discussed only briefly in this
report (Section II). It is EPA's firm belief that attempts to
predict (and compare) costs of various types of plants in a general
way, apart for local circumstances, is more likely to mislead than
inform. The range of assumptions regarding specific design, reliability,
markets and other factors is too great to make such an analysis
meaningful. We will note only that all but the "experimental"
technologies discussed in this report have been shown or predicted
by reliable engineers to be a roughly "competitive" disposal alternative,
particularly for cities with higher disposal costs, under "reasonable"
assumptions of capital and operating costs and product revenues.
12
-------
SECTION II
GENERAL CONSIDERATIONS FOR RESOURCE RECOVERY SYSTEM DESIGN
To help a community evaluate and select the technology that best
meets its needs, subsequent chapters of this Guide provide descriptions
of available and developing resource recovery technologies. However,
evaluating and selecting a technology are only part of the implementation
process.
There are several additional aspects of a technical nature that
must be considered:
1. Markets for recovered products
2. Waste generation (quantity)
3. Waste composition
4. System reliability
5. Plant location
6. Land required for the plant site
7. Community acceptance
8. Plant costs and revenues
These factors, which apply to all technologies, must be considered
in the design of any system and are discussed briefly below.
Markets for Recovered Products
The successful implementation of a resource recovery system depends
upon the ability to sell the recovered products. Revenues from the
sale of recovered products can help to offset the cost of owning and
operating the plant; without such revenues, the cost of most resource
recovery systems would be prohibitively high.
To be marketable, products reclaimed from energy and materials
recovery systems must have qualities that are acceptable to the user.
Steam and electricity produced from solid waste are similar to those
products from other sources. However, refuse-derived-fuels (solid,
liquid, and gaseous) have characteristics that are different from
conventional fossil fuels. Some of the more important fuel characteris-
tics are: ash content, higher heating value, corrosiveness, viscosity,
and moisture content. Similarly, the quality of recovered materials
must be commensurate with user specifications.
For all products derived from refuse, considerations such as
reliability of supply and quantity are also important. A higher
price can usually be obtained for a certain supply than for an
unreliable source. This is particularly true for fuel processing.
Additional information on markets is available in the Markets guide
of this series (SW-157-3).
13
-------
Waste Generation (Quantity)
The amount of waste that the community generates must be estimated
carefully so that the resource recovery plant (and accompanying elements
in the total solid waste management system, i.e. transfer stations and
sanitary landfills) can be designed at the proper size. An oversized
plant will nave under-utilized equipment and will cost more than necessary.
An undersized plant will not be able to accept the quantity of waste
that must be processed.
There are several ways to estimate waste generation quantities.
Some of these may appear costly; but, considering the potential costs
of over - or under-designing a plant, estimating waste quantifies is
an essential and prudent investment.
Alternative methods for obtaining these data are discussed below.
In many communities (and in particular smaller ones) no weight
records are maintained. A common procedure for determining generation
rates is to count the trucks and then, assuming they are fully loaded,
estimate the tonnage based on the total volume of the trucks entering
the site. Such a procedure can be very misleading and should be avoided.
Communities lacking scales at their disposal sites have several
alternatives that can be utilized instead of volume measurements. The
best approach, short of installing a platform scale, is to reroute
the collection vehicles to an existing scale on a temporary basis.
A highway weigh station or a privately-operated scale, such as at a
grain elevator or trucking firm, may be available. The weighing
schedule should be set up to allow for enough data to span seasonal
and daily variations in generation rates. If only part of a community's
waste will go to the recovery facility, demographic differences should
also be accounted for in the weighing schedule.
In lieu of such a weighing program individual axle weights can
be measured using portable scales at the disposal sites. However,
care must be taken to make an adequate number of weighings, even
though axle weighing is more cumbersome and time consuming than the
use of platform scales.
A final option, one which should be used only as a last resort,
is to utilize national average per capita generation data applied
to the population to be served by the facility. This approach
leaves considerable chance for error because local waste generation
often is significantly different from national averages. In addition,
quoted per capita generation rates may include different waste sources
than those which will go to the recovery facility. (Commercial waste
and construction debris are two examples where confusion could arise.)
Thus, national average data should be avoided as a primary estimate.
14
-------
Waste Composition
Evaluation, selection, and design of any resource recovery system
requires accurate data about waste composition, i.e., what materials
are present in the waste and in what proportions they occur. This is
particularly true where materials recovery subsystems are involved,
as the composition of many valuable components (such as ferrous metal
or aluminum) can vary significantly among different communities.
Waste composition variations such as heat content, moisture content
and ash content can impact on selection and design of energy recovery
components.
Table 3 presents national average'data on waste generation and
composition. The limitations described above on using these data as
a substitute for estimating local waste quantity also apply to their
use for determining composition.
TABLE 3
MATERIAL FLOW ESTIMATES OF RESIDENTIAL
AND COMMERCIAL POST-CONSUMER SOLID WASTE
1973
Material
Paper
Glass
Metals
Plastics
Rubber and leather
Textiles
Wood
Total nonfood product waste
Pounds Per
Capita Per Day
1.36
0.36
0.35
0.14
0.10
0.06
0.13
2.50
Percent
38.9
10.3
9.9
4.1
2.7
1.6
3.6
71.1
Food waste
0.47
13.3
Total product waste
Yard waste
Misc. inorganics
Total
2.97
.50
.05
3.52
84.4
14.1
1.5
100.00
15
-------
However, the solid waste industry, is not in agreement as to how
much, if any, waste sampling for composition should be done. The
major drawback is cost. A waste composition study could cost a
community $5,000 to $20,000. This is a small price compared with the
millions of dollars of capital investment that it affects. However,
some persons argue that the combustible fraction of the waste stream
does not change significantly in percentage from place to place and,
thus, that a facility designed to recover primarily energy (and perhaps
ferrous metals) can be designed without such a composition analysis.
(This would be particularly true where waterwall incineration is involved.)
Although there is merit to this argument, moisture, heating value, and
ash content are important data for design of energy recovery systems and
should be determined.
If recovery of aluminum or glass is being considered, a composition
analysis is far more critical, as these materials vary significantly
in precentage composition from place to place, and quantity in the
waste stream would bear heavily on economic feasibility of recovery.
Clearly, the safest route is to conduct a waste composition analysis,
and EPA believes that the benefits justify the investment. However,
each community will make its own decision based on the cost trade-off
it sees, and the type of recovery technology it expects to employ.
System Reliability
A solid waste management system must accept all the waste that is
generated, and it must accept it when it is generated. Reliability
of the entire solid waste management system is a function of plant
reliability plus the availability of alternative processing or disposal
facilities. Therefore, the system, including the resource recovery plant
and the sanitary landfill, must be designed to operate reliably.
This discussion focuses on the reliability of the resource recovery
plant.
Reliability as defined here is the ability of the plant to accept
and process the community's waste on a regular basis. Reliability is
achieved by a combination of the following:
1. Operational reliability of the equipment;
2. Redundancy of equipment or systems; and
3. Storage capacity, combined with excess processing capacity
to handle backlogs and current demands at the same time.
Excess processing capacity can be achieved by using:
1. Intentionally oversized equipment; and
2. Overtime use of primary processing lines.
The community must specify the degree of reliability it requires
of the resource recovery plant and must communicate this to the system
16
-------
designer. Because reliability is achieved only with an increase in
cost, the degree of reliability desired will have to be evaluated in
terms of capital and operating costs of the system. Differences in
the type or degree of reliability designed into a plant is one reason
for significantly different capital costs of plants which are func-
tionally similar.
There are no simple guidelines on the degree or type of reliability
which is best. However, the decision maker should take care to ensure
that a reasonable degree of reliability is included in the design
even though it may increase initial costs. A "bare bones" facility
could cause operational headaches.
Plant Location
Solid waste processing plants should be located as near as
possible to centers of solid waste generation in order to minimize
haul costs and be readily accessible by major roads where the truck
traffic will present minimal environmental impacts. The location
should also be compatible with market requirements. For example,
waterwall boilers should be located as close as practical to steam
users to avoid large steam transmission losses and costs. When
solid fuels or recovered metals, paper, and glass are being sold
to distant markets, access to rail sidings and major thoroughfares
should be available. The site should be industrially zoned.
Public utilities such as power, gas, water, and sewage should be
available at reasonable installation costs. Truck traffic through
residential areas should be minimized.
Land Required for the Plant Site
The land area required for the plant site will vary with the
type of system, the size of the system, and certain site-specific
constraints such as highway access, height limitations, and typography.
The following data are rough estimates indicating the order of
magnitude of land requirements.
Smaller processing plants (with capacities in the 200 to 500
tpd range) will generally require three to five acres of land.
Larger plants (processing over 1,000 tpd) require at least 5 to
10 acres. Trying to squeeze a plant into too small a site can be
very costly, resulting in severe limitations on operating and main-
taining the plant. Therefore, care should be given to providing
adequate space.
Community Acceptance
Resource recovery system planners should be aware of the need
to make resource recovery plants good neighbors. A long history of
poorly operated solid waste disposal facilities has convinced the
17
-------
public that such facilities should be built "somewhere else." Objections
which are most often voiced include increases in truck traffic, spillage
from trucks, noise, harborage of rats and vectors, dust and air
pollution, unsightly plants, etc.
Such factors need not be problems in properly designed recovery
plants. Plant designers should recognize these objections and incorporate
measures to eliminate them. Adequate allowances for attractive
architectural treatment of the buildings and landscaping (both for
decorative purposes and to screen out noise, etc.) are necessary.
Additional acreage, in order, to provide a buffer zone, should also be
considered. Siting should take into account the routes the trucks must
follow. Only commercial thoroughfares should be used and adequate
roadways should be built on the property so that trucks need not
queue up on city streets. Adequate housekeeping of the facility and
grounds must be included in both the design .and the operating procedures.
Sound dampening enclosures should be used to house noisy pieces of
equipment.
The decision maker must anticipate siting problems early and
design a program to deal with them properly. Eliminating the reasons
for objections is not enough, however. The decision maker must also
initiate aggressive communications with the public to prove to them
that their concerns will be met and their interests will be protected.
The caliber, timing, and extent of this effort may be the most critical
task in successfully implementing a new solid waste disposal facility.
The value of professional assistance in conducting this effort should
not be overlooked.
Plant Cost and Revenues
Cost is usually the major factor in decisions about whether to
implement large-scale mixed-waste resource recovery plants. Cost
considerations are also important in formulating State and Federal
policies relating to such implementation. Thus it is important that
sound methods of evaluating and comparing cost figures be used.
Unfortunately, very little useful economic data are available as
no full-scale mixed-waste separation plants have begun regular
operation at this time. In the absence of operating data, cost
projections must be based upon preliminary estimates by consulting
engineers and system development companies; these estimates are derived
from experience with pilot-scale operations and from equipment supplier
quotations.
A major problem in projecting costs has been the general lack
of comparability among cost estimates. There are two apparent causes
for this. First, 'different cost-accounting methods are employed by
various designers, making it difficult to compare cost projections in
proposals from companies bidding on the same contract. Secondly,
18
-------
most estimates have been site-specific and reflect a wide range of
factors which vary from site to site. Capital costs on a 1,000 ton
per day plant may range from $15 to $35 million or more, depending on
the type of system chosen, land and site preparation costs, and
construction costs, including labor, materials, and equipment. Cost
ranges of this magnitude have been experienced even for a given type
of technology.
Annual costs, which include amortized capital cost and operating
and maintenance costs, may vary from $10 per input ton to $25 per
input ton, depending on, among other things, the utilization of capacity,
the interest rate on borrowed funds, wage rates, utility rates, fuel
prices, local taxes, residual waste disposal costs, and assumptions
concerning plant reliability and maintenance costs.
Selling prices for the recovered products are also a great source
of uncertainty. They exhibit large variations among geographic regions
and have been subject to extreme fluctuations over time. Future
negotiable prices for recovered fuels and materials are subject to
additional uncertainties due to technical questions about product
quality.
Considering all of these variables, it is obviously difficult
to provide "typical" costs of various resource recovery plants. We
believe that any such costs would be more likely to mislead than
inform. We will note only that the projected net costs of the 10 or
so plants under design or construction in this country are in the
$5 to $15 per ton range.
Now that several plants are about to begin regular full-size
operation, reliable data will be becoming available. Analysis and
dissemination of these data are high priorities of EPA's Office
of Solid Waste Management Programs.
Until such data is produced, planners and managers should consult,
among others, the persons and literature mentioned in the Resource
Recovery Plant Implementation: Guides for Municipal Officials:
Further Assistance (SW-157.8).One publication, highly recommended
because it illustrates how net plant costs are sensitive to site
specific factors, is entitled Resource Recovery Plant Cost Estimates:
A Comparative Evaluation of Four Recent Dry-Shredding Estimates
(SW-163) by Frank A. Smith, published by EPA's Office of Solid Waste
Management Programs.
One final note of caution about interpreting plant cost informa-
tion. The careful manager will not accept a capital cost figure or
a net cost per ton figure without asking many questions about the
configuration of the system. Typical questions include plant size,
type of technology, plant reliability, redundancy, land cost, method
of transportation of products, etc. (As an aid to comparing cost
19
-------
estimates of different systems, the reader is encouraged to use
the Accounting Format (SW-157.6); which is part of this Implementation
Guide series.)Furthermore, the careful manager will ask whether the
reported costs are based on a preliminary process flow diagram, a final
engineering design, or some other stage in the development of a system.
Obviously, actual costs are the most reliable; estimates based on a
preliminary process flow diagram are far less reliable as predictors
of what actual costs will eventually be; and so on.
20
-------
SECTION III
ENERGY RECOVERY SYSTEMS
Interest in recovering energy from municipal solid waste has
increased sharply in recent years because of the receding avail-
ability and rising cost of conventional fuels, and the continuing
problem of solid waste disposal. Roughly 70 to 80 percent of urban
waste is combustible; reported heating values of raw urban wastes
range from 3500 Btu/lb to 6500 Btu/lb and average about 4600 Btu/lb.
Since the current rate of generation of municipal solid waste is
approximately 3.5 pounds per person per day, each person in the
community discards the energy equivalent to 1.5 pounds of coal each
day. As a result, solid waste is now being regarded as an energy
resource.
The objective of energy recovery systems is to utilize the
heat of combustion (the energy) contained in the waste while
providing a means of reducing the volume of solid waste to be disposed.
This section discusses the following alternative means of
recovering energy from municipal solid waste: the direct generation
of steam in waterwall furnaces; preparation of solid refuse derived
fuels; pyrolysis to produce steam or gaseous, and liquid fuels;
biological gasification systems; and generation of electricity in a
turbine with gases from burning waste. For each of the alternatives
there is a brief process description, a review of the current status of
the process, a calculation of the amount of energy recovered, and
a discussion of any special considerations or product characteristics.
Energy Balances
Energy balances were calculated from data available in the
literature and from vendor contacts. The definition of thermodynamic
efficiency is compatible with the one used in the utility industry:
the ratio of energy produced (exported) to raw energy input. Similar
to utility practice, electrical energy and auxiliary fuel consumption
has been subtracted from the total energy produced to arrive at the
net marketable energy produced. Also, these input energy sources
have been close-looped within the process, that is, it has been assumed
that all external energies were produced within the system. Thus,
for example, for each process the amount of fuel produced which would
be needed to produce the electricity used in running the process is
subtracted. The amount of fuel needed to produce a unit of electrical
energy varies for each system because each of the fuel products
has a different conversion efficiency. By doing so, the reported
thermodynamic efficiency reflects true energy yield from a unit of
solid waste.
21
-------
Three efficiency values are reported. Fuel efficiency ( F)
indicates the percentage of energy in the solid waste contained in the
fuel product after accounting for in-plant energy consumption.
This fuel is then assumed to be used in a boiler to produce steam.
The efficiency of the boiler ( B) is multiplied by fuel efficiency
to yield total system efficiency ( S) for a process converting
solid waste into steam. Thus:
F is the parameter which determines the amount of fuel
produced by various processes.
B specifies the fraction of the fuel which can be converted
into useful work (here assumed to be steam). The fuel pur-
chaser uses this measure to evaluate the relative value of
equivalent amounts of alternative fuels.
S indicates the fraction of the input waste which is converted
into a usable end product (steam). This parameter enables
different types of energy products to be compared on an
equivalent basis.
For the sake of simplicity, the incoming municipal solid waste
has been assumed to have a composition as shown in Table 3 and a
higher heating value of 5000 Btu per pound. Table 4 presents a summary
of the energy efficiencies for the various processes.
Waterwall Combustion Systems - Unprocessed Haste
•Traditionally, the generation of steam from raw refuse has been
accomplished by connecting waste heat boilers to refractory-lined,
stoker-fired incinerators. However, increasingly stringent air
pollution standards created a need for a more effective combustion
unit—the waterwall furnace. This type of furnace has virtually
replaced the use of refractory-lined furnaces because these units
(1) are easier and cheaper to maintain, (2) are smaller and less
costly to build and (3) are more efficient in recovering the energy
available in the solid waste.
Although this technology was developed more than 50 years ago
for use with low grade coal and other types of waste fuels, it
has only been used for municipal solid waste for about 20 years.
However, in Europe and Japan its acceptance has been rapid and
widespread and several hundred units have been built in sizes ranging
from 60 to 2600 tons per day. In the United States and Canada 10
plants have been built,_all since 1967.
Steam is produced at a rate of from one to three pounds per
pound of solid waste,.depending on design and operating conditions,
and the heat value of the solid waste. The steam can be used directly
22
-------
TABLE 4
COMPARISON OF ENERGY RECOVERY EFFICIENCIES
FOR VARIOUS SOLID WASTE ENERGY RECOVERY PROCESSES
Process
Water Wall Combustion
Fluff RDF
Dust RDF
Wet RDF
Purox Gasifier
Monsanto Gasifier
Torrax Gasifier
Oxy Pyrolysis
Biological Gasification*
With use of residue
Without use of residue
Brayton Cycle/combined cycle
Waste Fired Gas Turbine
Net Fuel
Produced
(Expressed as pe
incoming solid wa
70
80
76
64
78
84
26
29
16
Total Amount
Available as Steam
"cent of heat value of
ste)
59
49
63
48
58
42
58
23
42
14
19 plus
12 directly
aselec
* Includes energy recovered from sewage sludge.
tricity
in turbines to drive the major items of equipment in the plant, or
it can be used in a turbo-generator to produce electricity for in-
plant use. There is sufficient steam produced, however, that most
of it is available for off-site use, either as steam or as electricity.
If this excess steam cannot be sold to a nearby industry or utility
or used in other municipal facilities, then it must be condensed and
cooled before it can be recirculated to the furnace.
Process Description. Municipal solid waste is deposited on a
tipping floor or in a large storage pit from which it is transferred
to the furnace feed hopper (Figure 1). From the feed hopper, the
waste is fed onto mechanical grates where it burns as it moves
continuously through the furnace. Noncombustible material falls off
the end of the grate where it is quenched with water and then conveyed
•to trucks or a temporary storage pit. Ferrous metal is routinely
recovered from the residue and in Europe the ash is often used as
a road building material.
23
-------
BOILER
ro
TACK
CONVECTION
SECTION
Figure 1. Typical Waterwall Furnace for Unprocessed Solid Waste
-------
Waterwall furnaces are enclosed by closely spaced water filled
tubes. Water circulating through the tubes recovers heat radiated
from the burning waste. Integrally constructed (attached) heat
recovery boilers generate steam while reducing the temperature (and
the volume) of the exhaust gases. The boilers consist of various
zones or tube packages referred to as heaters, economizers, reheaters,
etc. depending on the function of the particular zone. Marketable
product (steam) is created while permitting the use of smaller gas
cleaning equipment (gas volume is proportional to absolute temperature).
In the combustion process, oxygen (air) is required to burn the
fuel and release heat. Air is introduced into the furnace beneath the
grates (underfire air) to aid in combustion and help keep the grates
cool. Air is also introduced above the fuel bed (overfire air) to
promote mixing of the gases (turbulence) and to complete combustion
in the furnace.
The combustion gases, after being cooled as they pass through
the various boiler sections, are passed through air pollution control
devices (generally electrostatic precipitators) and are then vented
to the atmosphere through a stack.
Status. The use of waterwall furnaces for the recovery
of steam from the combustion of solid waste has been practiced
widely in Europe for over 20 years. Conditions that facilitated
development of steam recovery facilities in Europe include the lack
of availabile land for landfills, the relatively high costs of fossil
fuels, and institutional factors (the responsibility for both refuse
disposal and power generation are often in the hands of one governmental
entity). More than 50 waterwall incinerators are operating in the
Federal Republic of Germany alone.
Application of waterwall incinerator technology in the United
States for the recovery of waste heat has been recently encouraged
by the success of European experience. The first large-scale United
States solid waste burning furnace utilizing waterwalls and
recovering steam is at the U.S. Naval Station, Norfolk, Virginia.
This 360 ton per day plant has operated successfully since 1967. The
steam produced is used to satisfy the station's requirements for
heating and cooling. Other facilities have been successfully operated
for several years, but, for nontechnical reasons, steam has been sold
only intermittently. These facilities are located in Chicago, Illinois;
Braintree, Massachusetts; and Harrisburg, Pennsylvania.
In Nashville, Tennessee, a 720 ton per day facility has been in
operation since 1974. Steam from this plant is distributed through
a utility loop to several dozen large government and private buildings.
During the summer the steam is used to operate a chiller so that cooled
water can be distributed for air conditioning use. This plant has
25
-------
experienced severe design and operating problems. Failure to employ
design features already proven in other plants, largely due to an
attempt to cut costs, is the primary reason for the problems experienced.
Another new steam generating incinerator, which is located in
Saugus, Massachusetts, sells superheated steam to an adjacent industrial
user. The market was obtained before the plant was built. This plant,
which began operating in 1976, was privately constructed as a profit-
making venture. It is owned jointly by a combustion systems manufacturer
and a waste disposal contractor.
The overall operating experience of waterwall combustion systems
in the U.S. and Europe varies. There are examples of both good and
bad operations. That is, some units have performed reliably, been
economically acceptable, and sold steam or electricity to a user on
a regular basis. This is particularly true of units in Europe installed
within the past 5 to 8 years by reliable, experienced companies. Other
facilities which have been either designed or operated poorly or which
have not developed markets for their steam output have exhibited
technical or economic problems.
Waterwall combustion systems or components are available from
a variety of manufacturers. Wheelabrator-Frye (representing the
Von Roll Company of Zurich) and Universal Oil Products (representing
the Josef Martin Company of Munich) are marketing complete systems.
Components (boilers and stokers) are available from Babcox and Wilcox,
Combustion Engineering, Foster-Wheeler, Riley Stoker and Detroit Stoker.
Energy Balance. Figure 2 shows an energy balance for a waterwall
furnace burning mixed municipal solid waste. In a well designed and
operated unit, more than 97 percent of the combustible matter is
consumed to liberate heat for steam generation. European design and
operating practices indicate that approximately 62 percent of the energy
in the refuse can be converted into steam. After accounting for the
energy used to operate the waterwall furnace, 59 percent of the input '
energy is available for sale to a customer. This is among the highest
energy efficiencies of any of the systems discussed in this report.
Recent design changes have been made by Wheel abrator-Frye in the
plant they have installed in Saugus, Massachusetts, which may enable
the waterwall furnace to operate at 70 percent excess air. If these
changes resolve the severe corrosion problems encountered in previous
attempts to operate at low excess air, then up to 67 percent energy
recovery could be realized.
Residues produced from the combustion of refuse in waterwall
incinerators represent approximately 10 percent by volume of the
input waste and 25 to 35 percent of their original weight. Residues
consist of ash, glass, ferrous and nonferrous metals, and unburned
organic materials. Recovery techniques use the unit operations
described in Section IV. Unrecovered residue must be buried in
sanitary landfills to minimize leaching problems.
26
-------
5000 Btu
1 LB MSW
DISSIPATED ENERGY
167 Btu
INTERNAL STEAM USE
167 Btu
t
R/C LOSS
IOO Btu
I
FLUE GAS LOSS
I663 Btu
WATERWALL FURNACE
EXPORT
STEAM
3II2 Btu
2946 BtU.
>7 = 59 %
ASH a UNBURNT CARBON LOSS
125 Btu
.07 LB
Figure 2. Waterwall Furnance Energy Balance
*This balance was based upon data obtained from:
Stabenow, G., Performance of the New Chicago Northwest Incinerator, In Proceedings; 1972 National
Incinerator Conference, New York, June 4-7, 1972. New York, American Society of Mechanical
Engineers, p. 178-194.
Barniske, L., and W. Schenkel, Entwicklungsstand der Muellverbrennungsanlagen mit
Waermeverwertung in der Bundesrepublik Deutschland, In Proceedings; Conversion of Refuse to
Energy, .Montreux, Switzerland, November 3-5, 1975. New York, Institute of Electrical and
Electronics Engineers, p. 91-96.
Stephens, W. C., and R. I. Simon, An Economic and Financing Model for Implementing Solid Waste
Management/Resource Recovery Projects, In Proceedings; Conversion of Refuse to Energy, Montreux,
Switzerland, November 3-5, 1975. New York, Institute of Electrical and Electronics Engineers, p.
422-427.
27
-------
Product Characteristics. Steam produced in incinerator facilities
may be used in-house or for district heating and cooling, electricity
generation, or to drive machinery in industrial processes. Steam
customers are usually utilities or large industrial complexes. Steam
temperatures range from 250 F to 1050 F, and pressures range from 150
pounds per square inch guage (psig) to 950 psig. As a rule, higher
temperatures result in a more marketable steam product, but they also
result in larger maintenance expenses. In steam distribution systems,
steam temperatures are kept low to minimize heat losses. In electric
power plants, however, high temperatures and pressures are desirable
because they increase generating efficiency. ,
While steam is an almost universally usable source of energy it
would not be economic to transport it more than 1 or 2 miles. In
addition, the marketing of steam requires a suitable distribution or
delivery system.
If the steam customer has no alternative or standby source of
steam than he must be assured a reliable supply. Providing this
reliability will probably require the installation of a standby
fossil fuel fired boiler for use during emergencies. Of course,
the value of the steam is enhanced by the increase in its reliability.
Waterwall Combustion Systems - Processed Waste
Waterwall furnaces are also being designed to burn coarsely
shredded solid waste. The concept is that by first shredding the
solid waste (and possibly removing the ferrous metal and other
noncombustibles) a more homogeneous and thus more controllable fuel
can be produced. The shredded waste is fed into the furnace by
spreader stokers which propel the waste across the combustion chamber,
where it then lands on a traveling grate. This, type of firing is often
referred to as semi-suspension firing because the waste is ignited
while it is falling through the chamber but combustion is completed
while it rests on the grate.
One such plant is currently in use in Hamilton, Ontario on
municipal solid waste, and several others are in use for burning
industrial wastes. Similar plants have been announced for Akron,
Ohio (construction is expected to begin in 1976) and proposed
for Niagra Falls, New York. Also, the Black Clawson Company has
been contracted by the Town of Hempstead, New York to build, own
and operate a semi-suspension waterwall incinerator that will
burn the wet-pulped fuel produced by their patented wet pulping
process. A similar plant is expected to be built in Dade County,
Florida and a 150 ton per day demonstration plant is in operation
in Tokyo.
In addition to producing a more attractive fuel for the
incinerator, shredding enables recoverable materials which would
28
-------
otherwise be consumed or degraded in the burning process to be extracted
prior to combustion.
The cost of shredding the waste must be balanced against the
benefits of a more uniform fuel and "front-end" materials separation.
Also, there is less experience with these semi-suspension systems
than with burning unprocessed waste on a grate.
Solid Refuse Derived Fuel Systems
Mixed municipal solid waste can be processed to produce a supple-
mentary fuel for fossil-fuel fired steam generators. The wa-ste fuel
is commonly called refuse-derived fuel (RDF). The fuel can/supplement
coal (and possibly oil), and has a heating value of about one half
that of coal.
In this approach a system consists of a processing plant where
the solid waste is shredded and classified, and a separate facility
(market) where the RDF is burned. Utility or industrial steam
generators provide a potential market for the RDF. These steam
generators must be proximate to the source of solid waste and have
adequate capacity, load factor, and ash handling systems. An
attractive feature of this system is that the capital, operating and
maintenance costs of the steam generation and auxiliary equipment are
already being borne by the user of the RDF. This can be particularly
significant in terms of capital costs. However,-the cost of modifying
an existing generator or designing a new generator to fire RDF, as
well as the incremental costs of operation and maintenance attributable
to firing the RDF will be passed back to the waste processing system,
usually by reducing the price paid for the RDF.
Users could be expected to pay the same for RDF as they pay for
the primary fossil fuel, on a fuel value basis. Of course all costs
and savings associated with handling the RDF would be deducted to
determine the net value.
Because the steam generator is designed to fire fossil fuel
primarily, the RDF must have the physical and combustion properties
necessary to make it compatible with the specific boiler-furnace
firing and ash handling system being considered.
Consequently, several types of RDF are being offered to potential
users: fluff RDF, densified RDF, and dust RDF.
Fluff RDF - Dry Processing Systems
Fluff RDF is waste that has been processed so that it will burn
efficiently in suspension as it falls down through the fire-ball
(center of turbulent flame patterns) of a boiler-furnace. It can
be fired into the large utility-class boilers (greater than 500
29
-------
million Btu per hour input or 50 megawatts of power output) including
both suspension fired and cyclone fired boilers, and in certain stoker
and spreader-stoker fired boilers.
Fluff RDF can be defined for purposes of this discussion as RDF
with a particle size of from 1/4 inch to 2 inches (but generally about
1 inch) and with most of the heavy, dense materials, both organic and
inorganic, removed.
The particle size and the degree to which the heavy, dense materials
are removed depends upon the specific characteristics of the boiler-
furnace that will fire the RDF. For example, very large utility
boilers may befable to efficiently burn RDF with a relatively large
particle size and containing some of the wood, rubber, and plastics.
This is because larger furnaces have more turbulent fire-balls and
the RDF particles are suspended in the furnaces and subjected to
the 2200 to 2600 degree temperatures longer.
Because stoker fired boilers have relatively small furnaces,
the RDF has very little time to burn as it falls through the
furnace. Instead, grates at the bottom of the furnace hold the
burning particles until they are completely combusted.
Process Description. The RDF dry processing concept was originally
demonstrated at St. Louis, Missouri where solid waste was passed through
a single shredder to reduce the particle size to 1-1/2 inches. The
shredded material was then injected into an air classifier where a
vertical column of turbulent air separated the light RDF from the
inorganics and the heavy, dense organics that will not completely
burn in suspension. About 80 to 85 percent of the shredded waste was
separated as RDF.
The RDF was fired in two Union Electric Company 125 megawatt
pulverized coal-fired boilers at rates ranging from 5 to 27 percent
on a power output basis.
A typical RDF system may use primary shredding to reduce the
particle size to four to eight inches, followed by an air classifier
that separates from 50 to 85 percent of the shredded material as RDF
(Figure 3). To increase its heat value and reduce its abrasiveness
during pneumatic firing, the shredded material is then passed over a
screen or trommel to remove glass fines. -After screening, the RDF
is passed through a secondary shredder to reduce the particle size
to a range of from 1/4 inch to 2 inches, depending on the requirements
of the specific boiler being considered. The RDF is then stored,
transported, and fired, as required.
Status. Based on the general success of the St. Louis demonstration
project, the fluff RDF approach is being implemented in several cities.
30
-------
WASTE
PRIMARY
SHREDDER
SECONDARY
SHREDDER
HEAVIES TO MATERIALS
RECOVERY AND LANDFILLS
EMBRITTLING
AGENT
AIR
CLASSIFIER
FLUFF
RDF
DENSIFIED
RDF
DUST
RDF
1' Figure 3. This simplified flow diagram shows how the dry processing
approach (no water slurry) can be used to produce fluff, densified, or dust RDF.
-------
Consulting engineers are recommending implementation of the concept
where feasible. In addition, about ten companies are marketing systems
that they will design and construct for a fixed price.
One additional plant is already in operation*;in Ames, Iowa, and
Plants in Milwaukee and Chicago, are scheduled for completion in 1976.
However, the experience with RDF technology is limited. Though a
process of shredding and air classification can produce a sized, mostly
organic fuel, the process is far from optimized. Furthermore, burning
of this fuel as a supplement to coal is still in its infancy. Thus,
willing buyers are not readily available.
Design questions that remain to be answered and some comments
on RDF plant considerations follow:
- Shredding. Is two stage shredding (primary followed by
secondary) actually more cost effective than a single stage?
What is the most cost effective hammer configuration and hammer
retipping material? What is the optimum final RDF particle
size for each boiler firing system?
- Air Classification. Which configuration (design) is most
efficient? What is the optimum degree of removal of non-
burnables for each boiler firing system? Where should
the air classifier be located in the waste processing system—
between the shredders, after secondary shredding, before
or after screening?
- Screening. Is screening cost effective? What kind of screen
is most cost effective? Where should the screen be located--
before primary shredding, between shredders, etc.?
- Storage. Fluff RDF is a very difficult material to handle—
it bridges easily (hangs up in hoppers), has a negative
angle of repose (in a bin, the top of a pile of RDF will not
fall when the RDF directly underneath it has been removed),
it does not flow easily, and it binds like paper-mache1 after
several days of storage and when wet. The best way to handle
RDF would be to keep it in motion at all times. However, this
is not practical, so when storage is required, it should
be stored in a bin in which the unloading device is able to
retrieve the RDF from every point on the bottom on the bin.
In addition, to avoid bridging, the sides of the bin must
flare outward toward the bottom.
- Firing. Erosion of the elbows (curves)'of the pneumatic
firing system can become a major maintenance problem unless
replaceable elbows and an abrasion resistant material are
used at these high wear points. Which abrasion resistant
material is most cost effective? Another consideration
is the optimum elevation of the RDF firing nozzles to enhance
RDF combustion but to minimize ash carryover in the form of
particulate stack emissions.
32
-------
- Stack Emissions. Based on evaluations at St. Louis, sulphur
oxides, nitrogen oxides, mercury vapor, and chloride emissions
are not significantly changed when RDF is fired with coal at
rates of from 5 to 27 percent on a power output basis. In
addition, particulate emissions did not significantly increase
at all RDF firing rates as long as the boiler loading was at
or below its design capacity of 125 mw. However there was an
increase in emissions when the boiler was run at 140 mw using
RDF and coal, as opposed to coal only. Unfortunately, this
boiler is routinely operated at this higher load.
These higher emissions must have resulted from a decrease
in electrostatic precipitator collection efficiency because
the uncontrolled emissions (inlet to the precipitator) did
not increase significantly, even at the higher boiler loading.
It is felt that precipitator efficiency decreased because the
volumetric gas flow rate increases when RDF is used to supplement
the coal. This increase in gas flow rate is due, in part,
to the fact that the moisture content of the RDF (on a heat
value basis) is six times that of coal.
If this decrease in precipitator efficiency results in
emissions that exceed standards, a number of steps could
be taken to correct the problem: (1) RDF could be fired
only at lower boiler loads so that the combined firing
exhaust gases do not exceed the precipitator's design flow
rate; (2) modifications could be made to the precipitator
to increase its collection efficiency; or (3) the RDF could
be dried to reduce the quantity of exhaust gases. Each
situation must be investigated individually to determine
the most cost effective approach.
- Bottom Ash. If an accumulation of unburned RDF organics and
ash would overload the bottom ash handling facilities,
the accumulation could be logically reduced by taking one
of several steps to improve the combustion of the RDF:
(1) reduce the quantity of heavy dense materials recovered
with the RDF by adjusting the air flow rate of the air
classifier during waste processing and thus recovering a
lower percentage of waste as RDF; (2) increase the surface
area of the RDF particles by reducing the RDF particle size;
or (3) increase the retention time of the RDF in the boiler
by raising the elevation of the RDF firing nozzels.
- Boiler Operation. At St. Louis the Union Electric Company
has indicated that they have not experienced any problems
with maintaining power levels, slagging, erosion within the
boiler, fouling, or corrosion.
33
-------
Energy Balance. An energy balance has been developed for a
typical fluff RDF system (See Figure 4). It is based on a system
having two stage shredding; a trommel screen; air classification;
and truck transport to a user 15 miles away. Sixty-two percent of
the raw waste is assumed recovered as RDF.
DISSIPATED ENERGY
44Btu
CONVERSION LOSSES
152 Btu
392 Btj
FUEL
3897 Btu
BOILER LOSSES
1147 Btu
3505 Btu
0.62 LB
74%
BOILER
=69%
HEAVIES AND SCREENINGS
1144 Btu
0.38 LB
ASH
10 Btu
0.04 LB
Figure 4. Fluff RDF Energy Balance
*This balance was based upon data obtained from:
Shannon, L. J., M. P. Schrag, F. I. Honea, and D. Bendersky, St. Louis/Union Electric Refuse Firinq_
Demonstration Air Pollution Test Report, August 1974. Washington, Office of Research and
Development, U. S. Environmental Protection Agency. 108 p.
Proceedings; National Center for Resource Recovery, Inc., Seminar, U. S. Environmental Protection
Agency, Municipal Environmental Research Laboratory, Cincinnati, Ohio, December 3-4, 1975.
Session V, "Unit Processes for Materials Recovery."
Rigo, H. G., Technical Evaluation of the Feasibility of Burning Eco-Fuel at Philadelphia Naval
Shipyard, January 1974. Construction Engineering Research Laboratory, Letter Report E-25. 54 p.
Product Characteristics. RDF is clearly an inferior fuel to
coal in practically every parameter except sulfur content (Table 5).
However, when fired at low rates--!0 to 20 percent of power output-
boiler operation and maintenance problems are not expected to increase
measurably. However, there is only very limited experience to verify
this.
34
-------
TABLE 5.
COMPARISON OF FLUFF RDF AND COAL
Property
Heating Value (Btu/lb)
Bulk Density (Lb/Ft3)
Moisture
Average Size (In.)
Ash
Carbon
Hydrogen
Nitrogen
Oxygen
Sulfur
Per Pound
Fluff RDF
5,000 - 6,500
5-9
20 - 30%
1/4-2
%
19.
28.
6.9
0.6
45.
0.2
Coal
11,500- 14,300
42
3 - 1 2%
%
3-11
6.2 - 81
4.3 - 6.0
1.0-1.7
4.8- 17.4
0.6-4.3
Per Million Btu
Fluff RDF
106
31 - 60 Lb
Lb
29-38
43-56
11 -14
.9-1.2
64-90
2.5 - .35
Coal
106
2- 10 Lb
Lb
2-10
43-70
3-5
.7-1.5
15
.4 - 3.7
Fluff RDF - Wet Processing Systems
Fluff RDF can also be produced by a wet processing system where
the waste is converted into a slurry.
Process Description. Wet processing to produce RDF involves
size reduction, removal of non-combustibles, and dewatering steps.
Solid waste is conveyed to a wet pulping machine (hydropulper) where it
is mixed with water (see Figure 5). The waste material is reduced to an
aqueous slurry by the action of high speed cutting blades located at the
bottom of the pulper. Large items, such as tin cans, rocks and other
inorganics, are ejected through an opening in the side of the hydropulper.
Ferrous metals can be recovered from this material.
The remaining slurry is pumped to a liquid cyclone where heavier
materials such as glass, nonferrous metals, thick plastics, wood, etc.,
are removed by centrifugal action. The heavy fraction can be processed
to recover aluminum and glass. The lighter, mostly paper fraction is
then dewatered to the desired moisture content and used as RDF.
35
-------
CO
TIPPING FLOOR
BARREL PRESS
(THICKENER)
GENERATOR
TO RECOVERY OF
- ALUMINUM
SMALL FERROUS METALS
GLASSIBYCOLOR)
NONFERROUS MATERIALS
RETURN TO PROCESS
RECOVERED
FERROUS METAL
LOW PRESSURE
STEAM
TO PROCESS
Figure 5. Wet Process Energy Recovery System
-------
Although dewatering the s-lurry is expensive, the wet pulping approach
has several advantages over the dry processing approach. Perhaps the
most important is the ability to easily handle sewage sludges. By
blending sewage sludge with the pulped slurry, the resulting mixture
can be simultaneously dewatered for use as fuel. Of course, the
particular sludge being considered must be investigated for materials
(like heavy metals) that may cause air pollution problems during
combustion. However, dewatering the slurry is expensive, especially
if a 20 or 30 percent moisture content is required as may be the case
for use as supplementary fuel in some steam generators. Consequently,
the wet processing approach to producing a supplementary fuel will
probably not be as popular as dry processing. Rather, the RDF produced
by most wet processing systems will probably have a 50 percent moisture
content and be used as the primary fuel in steam generators designed
specifically to handle this material (see discussion on Waterwall
Combustion Systems - Processed Waste).
Nevertheless, at least two communities Memphis, Tenessee, and
Norwalk, Connecticut have seriously studied the feasibility of wet
processing to produce RDF for use as supplementary fuel in existing
coal-fired steam generators.
Other advantages include reduced likelihood of explosions and fires
during size reduction, fewer dust control problems, and greater flexibility
of handling and shipping.
Status. The basic RDF wet processing steps have been demonstrated
in a 150 ton per day EPA demonstration plant located in Franklin, Ohio.
The plant has been consistently processing about 35 tons per day of
waste since 1971. The Franklin plant dewaters a pulped slurry of
nonrecoverable fiber to a moisture content of about 50 percent before
burning it in a fluidized bed incinerator without heat recovery.
Nonetheless, the RDF preparation components have been successfully
demonstrated.
Energy Balance. An energy balance has been developed for a
typical wet processing system (Figure 6). Comparing this to the
energy balance for fluff RDF it can be seen that although a greater
percent of the combustible material is recovered as fuel, its lower
combustion efficiency results in a lower net yield of steam.
Product Characteristics. Wet process RDF is homogeneous (uniform
composition and particle size) and has an as-received heating value
of approximately 3500 Btu/lb, at a moisture content of 50 percent,
and an ash content of approximately 20 percent. The high moisture
content of the fuel product precludes the use of this product
in many existing boilers (however, it can be used as a primary fuel
in specially designed furnaces). Many industrial and utility boiler
operators are requesting that the material be between 15 and 20 percent
moisture. At this moisture, the RDF can be suspension or spreader stoker
fired or it can be densified (briquetting or pelletizing) for use in
37
-------
MIXING LOSSES
84Btu
CONVERSION LOSSES
222 Btu
BOILER LOSSES
1722 Btu
Figure 6. Wet Process RDF Energy Balance
"This balance was based upon data obtained from:
Wittmann, T. J., et al, A Technical, Environmental and Economic Evaluation of the "Wet Processing
System for the Recovery and Disposal of Municipal Solid Waste," Final Report SW-109c, U. S.
Environmental Protection Agency, 1975. 217 p.
chain grate, underfed, or mechanical or pneumatic spreader stoker
equipped industrial and utility boilers. Drying wet process RDF
requires significant amounts of energy. A 10 percent reduction in
moisture content consumes between 4 and 8 percent of the fuel, depending
on the specific design of the dryers.
Dust RDF
A dust-like RDF has been developed by Combustion Equipment
Associates as a high quality all-purpose fuel. According to the
developer, it can be fired in suspension with coal and oil in most
steam generators with adequate ash handling facilities. In addition,
it may prove feasible to slurry the fuel with oil for firing in
the oil-fired boilers, thus eliminating the need for separate firing
1i nes.
Process Description. Solid waste is first shredded, air
classified, and screened, the same initial steps used to produce
fluff RDF (Figure 3). However, instead of using a secondary shredder
to reduce the particle size further as in fluff RDF production systems,
38
-------
an embrittling agent is added to the coarse shredded material.
This chemical hardens the cellulose fibers so that the paper and cardboard
becomes friable and will shatter upon impact. The treated material
is then run through a ball mill similar to those used by electric
utilities to pulverize coal. The material is pulverized in the ball
mill until it will pass a 100 mesh screen. The dust RDF product
has a particle size of less than 0.15 millimmeter.
Status. According to company officials, a four ton per day pilot
plant operated continuously from Spring 1974 to October 1975.
A 20 ton per hour facility is being constructed at East
Bridgewater, Massachusetts. The fuel from this facility will be
trucked 75 miles to be used as an auxiliary fuel in several oil-fired
steam generators producing industrial process steam.
The dust RDF is to provide 60 percent of the heat input to the
generators. Although the company is investigating the feasibility
of slurrying the dust RDF with oil, the RDF and oil will not be
premixed for this project.
Energy Balance. An energy balance has been developed for a
typical dust RDF system (Figure 7). Eighty percent of the energy
is recovered in the fuel fraction. Its excellent combustion charac-
teristics make it a very efficient fuel so that when fired in a boiler
the steam yield would be 63 percent of the energy value of the incoming
solid waste. This would be the highest yield of any of the systems
examined in this report.
Product Characterists. According to company officials, the dust
RDF has a heating value of 6,900 Btu's per pound, and contains 10
percent ash and two percent moisture. Its bulk density is approximately
25 to 32 pounds per cubic foot. There are no known limits on shelf
life. It is expected that the product can be handled and stored like
a powder using conventional pulverized coal handling equipment.
Although dust RDF has superior combustion properties to fluff
RDF, the production costs are likely to be greater than for fluff
RDF. Also, special care in handling and storage is necessary to
minimize the danger of explosions. The cost trade-off can be
determined only through operating experience with these systems.
Densified RDF
Processes for densifying RDF are being investigated by a
number of organizations. By densifying RDF, it is anticipated
that some of the handling and storage problems of fluff RDF can
be avoided, that stoker and spreader-stokers can more easily fire
processed waste, and that it can be fired at higher rates than
fluff RDF.
39
-------
DISSIPATED ENERGY
884 Btu
CONVERSION LOSSES
136 Btu
BOILER LOSSES
642 Btu
220 E5tu_
FUEL ~!
0.52 LB or 4200 Btu
HEAVIES
800 Btu
3980 Btu ^
80%
BOILER
- 797o
3144 Btu
T]s = 63%
ASH
194 Btu
Figure 7. Dust RDF Balance
*This balance was based upon data obtained from:
p' .
Beningson, R. M., K. J. Rogers, T. J. Lamb, and R. M. Nadkarni, Production of Eco-Fuel -II from
Municipal Solid Waste CEA/ADL Process, in Proceedings; Conversion of Refuse to Energy, Montreux,
Switzerland, November 3-4, 1975. New York, Institute of Electrical and Electronics Engineers, p.
14-21.
Process Description. Densified RDF is produced by pelletizing,
briquetting, or extruding fluff RDF. It is also anticipated that dust
RDF can be densified by adding a chemical binder and processing in a
briquetter.
Status. Densified RDF has been produced in small quantities
at several pilot plants around the country. Several trial burns in
stoker-fired steam generators have been encouraging. However, this
concept has not yet been demonstrated on a commercial scale.
Several areas need to be investigated further:
1. Densifying Costs. What is the cost of densifying RDF? How
rapidly do the dies used in the process need replacement?
How much energy is required? What kinds of production rates
can be achieved with conventional equipment?
40
-------
2, Handling and Storage. Does the densified fuel hold together,
or tend to break up with time or when handled? Are the
materials handling properties improved relative to fluff
RDF so that lower cost storage and handling facilities can
be used? What is the optimum moisture content?
3. Firing. Can densified RDF be premixed with crushed coal for
firing in stoker or spreader-stoker steam generators, thus
avoiding the cost of a separate RDF firing system?
These questions are being addressed by a project recently initiated
by the EPA.*
Product Characteristics. The densified RDF will have basically
the same chemical properties of the fluff or dust RDF feedstock. But,
the bulk density will be increased to about 35 to 42 pounds per cubic
foot, which is similar to that of coal.
Pyrolysis Systems
Pyrolysis is the destructive distillation of the organic fraction
of solid waste. It occurs when organic material is exposed to heat
in the absence or near absence of oxygen. Pyrolysis differs from
incineration in that it is endothermic (heat absorbing) rather than
exothermic. Processes under development use heat from part of the
waste to provide the heat absorbed during pyrolysis and recover the
remaining heat in the form of steam or a gaseous or liquid fuel.
All processes reduce the solid waste to three forms: gases (primarily
hydrogen, methane carbon monoxide and carbon dioxide), liquids (water,
and organic chemicals such as acetic acid and methanol), and solids
(a carbonaceous char). The form and characteristics of the fuel fraction
varies for each of the different processes under development and is a
function of the reaction time, temperature and pressure of the
pyrolysis reactor, the particle size of the feed, and the presence
of catalysts, and auxiliary fuels.
To maximize gas production, reactor temperatures are held in
the range of 1400 F to 3000 F; for oil, temperature is on the order
of 900 F. Pressures range from 1 to 70 atmospheres. Ideally, the
reaction is allowed to take place in the absence of diluting gases
so that the product is the volatile matter of the solid waste. If
air is used in the reactor, the gases produced will be diluted by the
nitrogen in the air (air is approximately 79 percent nitrogen
and 21 percent oxygen). As a result, some processes have been developed
*"Preparation, Use and Cost of d-RDF as a Supplementary Fuel in
Stoker Fired Boilers," Office of Research and Development, U.S. EPA.
41
-------
which use oxygen, thus resulting in a higher heat content fuel gas.
Other systems indirectly transfer the heat to the gasifier to minimize
dilution of the product gas.
Heating solid waste releases gases and leaves a carbon residue
called char. In some reactors, the residue reaches such high temperatures
that the ash and other noncombustibles, such as cans and glass, melt
to form' a slag which can be removed from the reactor in a molten state
and quenched to form a glassy aggregate.
Residues produced from pyrolysis are biologically inactive and
may be safely disposed in sanitary landfills. Solid residues from the
noncombustible portions of the refuse, such as glassy aggregate, may
be used for construction and paving. If the char is not consumed in the
process, it has a higher heating value of approximately 9000 BTU/lb.
Its high ash content (50 percent), however, severely limits its
usefulness. Clearly, failure to consume all the char in the process
represents a loss in energy recovery.
This report describes the four pyrolysis systems which can be
classified as "developmental." There are presently no commercially
operational pyrolysis systems, and there are numerous other systems
which can be considered "experimental." All four of the systems
described have been previously operated on a small pilot scale and
full size plants Of 200 tons per day or larger have been or are being
built. Two of the systems produce low BTU gas which is used "on-site"
to produce steam. The third system produces a medium BTU gas which can
be sold to a nearby industrial user or may be suitable as a chemical
feedstock. The fourth system produces an oil-like liquid fuel which
can be stored and transported for use "off-site" in large industrial
or utility boilers.
Low BTU Gas - Monsanto Langard System
Process Description. The Monsanto Langard system employs a
controlled air primary furnace chamber (pyrolysis) and immediate
combustion of low heat value gases in an afterburner for recovery of
heat (Figure 8). Waste is shredded, conveyed to a storage silo,
and subsequently fed to a rotary kiln where it is pyrolyzed. Fuel
oil is also burned in the kiln to provide some of the heat for the
pyrolysis reaction. The burner is arranged to provide a counter-
current flow of gases and solids, thus exposing the waste to progressively
higher temperatures as it passes through the kiln. The finished residue
is exposed to the highest temperature 1000 C (1800 F) just before it
is discharged from the kiln and quenched in a water-filled tank.
The residuals are split into three fractions, glassy aggregate, ferrous
and char. The glassy aggregate and ferrous materials are recovered
for sale and the char is dewatered and landfilled.
42
-------
AIR POLLUTION CONTROL
WASTE HEAT BOILER
WATER CLARIFIER
STACK
GLASSY AGGREGATE
AND CHAR
MAGNETICS
FOR
SALE
SHREDDING
WATER QUENCH
Figure 8. The Monsanto Landgard System produces a low Btu gas which is immediately
burned on-site for the production of steam.
-------
Gases resulting from the pyrolysis reaction have a high temperature
(1200 F) and low heating value (120 BTU/cubic foot) making off-site
transportation uneconomical; therefore* they are immediately mixed
with air and burned in an afterburner to liberate the heat of combustion.
The combusted gases then pass through waste heat boilers where steam is
generated for distribution.
Product Characteristics. Steam from the Baltimore plant is produced
at up to 200,000 Ibs. per hour. Saturated steam at a temperature of
about 400 F is delivered to the Baltimore Gas and Electric Company's
existing downtown steam loop via a new mile long pipeline.
While this is the most economic arrangement for the Baltimore
facility, other end uses for the gas produced in the kiln might also
be possible. For instance, if a Landgard plant were built immediately
adjacent to a large utility boiler, it. might be feasible to direct
the hot, combustible kiln off-gas directly into the utility's boiler,
thus eliminating the need for a separate afterburner, waste heat
boilers, and air pollution control equipment.
Status. A 1000 ton per day prototype plant has been built in
Baltimore, Maryland. Construction of the plant was completed in
February, 1975 under a turnkey contract with Monsanto. However,
normal operation of the plant has not been possible because a number
of process changes are needed in order to insure proper operation.
Engineers from Monsanto and the City of Baltimore are now working on
a series of medications in an effort to correct the problems
plaquing the plant.
As originally built, exhaust gases are cleaned by means of a
large spray tower. Initial tests of the spray tower showed that it
could not clean the gases sufficiently to meet the required local
and Federal ordinances. All efforts to modify plant operations to meet
the standard have failed, and as a result it has been decided that
additional air pollution control equipment must be added.
Energy Balance. The energy balance for the Languard system is
shown in Figure 9. Here again it is assumed that the energy to produce
the purchased electricity and contained in the purchased quench oil
(7.2 gallons of No. 2 fuel oil per ton of solid waste) was provided
by the system's energy product. Losses in the process include the energy
remaining in the carbonaceous char and conversion losses experienced
in the waste heat boiler. As a result of these losses and provisions
for the system's input energy needs, 78 percent of the energy in the
incoming waste is available in the combustible gas. This gas is
then burned in an afterburner and the heat is recovered as steam in
a "waste-heat" boiler. The burner-boiler combination has a heat
recovery efficiency of 54 percent so the net recovery of energy in
the form of steam is 42 percent of the energy available in the "as
received" solid waste.
44
-------
DISSIPATED ENERGY CONVERSION LOSSES
130 Btu 352 Btu
FLUE GAS 8 R/C LOSSES
I ISC B»u
1 LB MSW
5000 Btu
3
m
to
'
|
PURCHASED
POWER
FRONT
END
SYSTEM
554 Btu
"" ~l ["OIL PURCHASED
I I
Z ' '
° if '
9
4620 Btu
PYROLYZER
I
I
I
I
I ^
3905 Btu
T]f 78 7o
WASTE HEAT
BOILER
2095 Btu
\ 1 17,, =42 %
rjB -- 54%
CHAR
7I5 Btu
Figure 9. Monsanto Landgard Energy Balance *
*This balance was based upon data obtained from:
Sussman, D. B., Baltimore Demonstrates Gas Pyrolysis, Resource Recovery from Solid Waste, First
Interim Report SW-75d.i, U. S. Environmental Protection Agency, 1975. p. 12-13.
.Levy, S. J., San Diego County Demonstrates Pyrolysis of Solid Wastes to Recover Liquid Fuel, Metals,
and Glass. Environmental Protection Publication SW-80d.2. Washington, U. S. Government Printing
Office, 1975. 7 p.
Low BTU Gas - Andco Torrax System
Process Description. The principal components of the Torrax
System are the gasifier, secondary combustion chamber, primary pre-
heating regenerative towers, energy recovery/conversion system, and
the gas cleaning system (Figure 10). The solid waste is charged as
received from the solid waste pit, without prior preparation, into
the gasifier. The gasifier is a vertical shaft furnace designed
so that the descending refuse burden and the ascending high temperature
gases become a counter-current heat exchanger. The uppermost portion
of the descending solid waste serves as a plug to minimize the infil-
tration of ambient air. As the solid waste descends, three distinct
process changes occur. The first is the drying where the moisture is
driven off; the second is the pyrolyzing due to the heat transfer from
the ascending, hot gases to the solid waste; and the third is combustion
in the hearth where the carbonaceous char is oxidized to carbon dioxide,
and melting of the inert fraction of the solid waste.
45
-------
ELECTROSTATIC
PRECIPITATOR
SOLID
WASTE
CTi
STEAM TO
INDUSTRIAL PROCESSES
AIR
COMPRESSOR
REGENERATIVE SLAG TAP
AIR HEATER INERT RESIDUE
SECONDARY COMBUSTION CHAMBER
Figure 10. Jorrax Slagging Pyrolysis System
-------
The heat for pyrolyzing and drying the solid waste and for melting
the inert fraction is produced by the combustion of the carbon char with
2000 F preheated air supplied to the hearth zone of the gasifier. The
heat thus generated melts the inerts to form a molten slag, which
is drained continuously through a sealed tap into a water quench tank to
produce a black, sterile, granulated residue.
As with the Monsanto system, the BTU value of the gas is too low
to make off-site transportation of the gas economic-. Instead the gases
are injected into an after-burner or secondary combustion chamber where
they are burned to completion. The heat which is thus released is then
directed to a waste heat boiler where it is recovered as steam.
A portion of the hot waste gas from the secondary combustion chamber
(about 15 percent) is directed through regenerative towers where its
sensible heat is recovered and used for preheating the process air
supplied to the gasifier hearth. These regenerative towers, successfully
used for many years in the steel industry, but as yet untested for this
system, are two refractory lined vessels containing a high heat capacity
refractory checkerwork material ,x Hot products of combustion from the
secondary combustion chamber and ambient process air are passed through
the towers on a cyclical basis for preheating the 1000 C combustion air.
The remainder of the secondary combustion chamber existing flow is
supplied to a waste heat boiler designed for inlet gas temperatures
of 2100 F to 3000 F.
The cooled waste gases from the regenerative towers are combined
with the exiting flow from the waste heat boiler and are ducted to
a hot gas electrostatic precipitator of conventional design.
Status. The principles of the Torrax process were originally
proven on a 75 ton per day pilot plant operated intermittently since
1971. This plant, located in Erie County, New York has been used
to process municipal solid waste and solid waste/sewage sludge. Test
runs with controlled percentages of waste oil, tires, and PVC plastics
were also run. The pilot plant differs significantly from the above
described system in that the hot blast combustion air is heated using
natural-gas-fired air-to-air heat exchanger instead of the regenerative
towers.
The Carborundum Corporation, which was involved in the original
development of the Torrax process, has recently turned its marketing
rights in the U.S. over to Andco.
A 200 ton per day prototype plant is undergoing startup in Luxemburg
and at least two other plants are also scheduled to be built in Europe
in the near future.
47
-------
Energy Balance. An energy balance for the system is shown in
Figure 11. About 15 percent of the energy value in the solid waste
is utilized to preheat the combustion air or replace the energy needed
to supply the purchased electricity used in the plant. The heat
ultimately delivered to the waste heat boiler is converted to steam
at an efficiency of 69 percent leaving a net system output (as steam)
of 58 percent of the original energy in the solid waste.
DISSIPATE
361
D ENERGY CONVERSION
Btu LOSSES
i 435 Btu
I
R/C 8 STACK GAS LOSSES
1302 Btu
1
124 Btu PURCHASED 413 Btu
5000 Btu
1LB MSW
S
1 70
0.2!
I
POWER
FRONT
END
SYSTEM
LA
B
3 L
i
G
u
.B
5030 Btu
DISSIPATED
ENERGY '
53 Btu
I37 300
Bfu Btu |
28!
AIR HEATER Btu
COMBUSTOR
1
T
AIR a WATER
71 Btu
n
1
1
1
1
! BOILER
j 4199 Btu ^ 2897 Btu ^
•tJF= 84% ~~ \ | ^s 58%
418 Btu
-rjB =69%
Figure 1 1. Energy Balance For The Torrax System
*This balance was based upon data obtained from:
Stoia, J. Z., Torrax — A Slagging Pyrolysis System for Converting Solid Waste to Fuel Gus,
Carborundum Environmental Systems, Inc., Solid Waste Conversion Division. Niagara Falls, New
York, p. 11-22.
Eerie County — Torrax Solid Waste Demonstration Project, Final Report, May 1974. U. S.
Environmental Protection Agency, Office of Solid Waste Management. 46 p.
Legille, E. etal, A Slagging Pyrolysis Solid Waste Conversion System, On Proceedings: Conversion of Refuse
to Energy, Montreux, Switzerland, November 3-5, 1975, New York, Institute of Electrical and Electronics
Engineers, p. 232-237.
48
-------
Medium BTU Gas - Union Carbide Purox System
Process Description. The key element of the Purox System is a
vertical shaft furnace (Figure 12), wherein shredded solid waste is
fed into the top of the reactor through a piston air lock system while
oxygen is injected into the bottom of the furnace. The solid waste
descends by gravity through the varying temperature zones on its down-
ward passage through the vertical reactor. The oxygen reacting with
char material previously formed from refuse in an upper zone of the
reactor creates a temperature zone in the range of 3000 F in the lower
portion of the reactor. Rising gases cool to approximately 200 F as they
move upward thereby providing the energy for pyrolyzing the incoming
waste in the upper portion of the reactor. Metals, glass and other
materials are transformed into a molten slag by the high temperatures
generated in the lower protion of the reactor. The molten slag mixture
continuously drains into a water quench tank where a hard granular
aggregate material referred to as "frit" is formed.
Product Characteristics. Gases leaving the reactor contain 30 to
40 percent moisture. This is removed in a gas clean up step, along with
ash, tars, and other condensable liquids. The remaining gas contains
approximately 75 percent CO and \\^ in approximately a two to one ratio;
the other 25 percent being comprised of CO?, CH^, No, and organic
compounds. Its heating value is approximately 300 BTU/cu ft.
Status. In 1970, the basic system was assembled in a 5 ton per
day pilot plant at Union Carbide's Technology Development Center in
Tarrytown, New York. Following evaluation of the pilot plant facility, a
200 ton per day Purox System was completed during 1974 in South Charleston,
West Virginia. The West Virginia facility was designed to prove out the
corporation's full-scale modular unit, and it was intended that larger
plants would obtain greater throughput capacity by incorporating modular
additions. Most recently, however, the Union Carbide Corporation has
decided to market a 350 ton per day module so that the unit being tested
is not the one that will be marketed.
Union Carbide is currently concluding the second portion of a three
phase testing and performance evaluation program. The first phase concerned
the receiving, feeding, and pyrolytic conversion of mixed municipal solid
waste without size reduction or sorting. The second phase involved
minimal pre-processing of incoming solid waste, consisting of coarse
shredding and magnetic removal of the ferrous fraction prior to intro-
duction into the pyrolysis reactor. The third phase anticipated to
commence in 1976, constitutes a co-disposal investigation wherein
sewage treatment plant sludges, containing varying moisture contents,
will be mixed in varying proportions with shredded solid waste.
Energy Balance. Energy is consumed in the Purox process primarily
in the shredding of the waste and in producing the 0.2 Ibs. of oxygen
that is required for each pound of solid waste burned (Figure 13). Each
pound of solid waste processed yields about 11.4 cubic feet of gas having
a heating value of about 300 Btu/cu.ft. Because this fuel burns so well,
if used directly in a boiler the combustion efficiency would be on the
order of 90 percent, with a net system efficiency of about 58 percent.
49
-------
ELECTROSTATIC
PRECIPITATOR
GAS
COOLER
tn
o
PRODUCT GAS
TO USER
MAGNETIC
SEPARATOR
PYRO LYSIS
FURNACE
OXYGEN
PRODUCTION
*• WASTEWATER
Figure 12. Union Carbide Purox System produces a medium Btu gas for sale to off site users.
-------
DISSIPATED ENERGY
578 Btu FLUE GAS LOSSES
5000 Btu
1 LB MSW
OXYGEN
0.2 LB
OXYGEN
PRODUCTION
Sii BTu
i
f
PURCHASED
POWER
499 Btu
* n
FRONT
C"M p\
tlNJU
SYSTEM
I i
3722 Btu 3223 Btu
1 1. 4 CUBIC FEET Tj 64% """
OTp AM
o i E.MIVI
BOILER 290' Bf" _
1 — V ^s 58 %
^-J
1?B = 90%
WATER FRIT
0.30 LB 0.20 LB
~50Btu I5I Btu
Figure 13. Energy Balance for the Purox Gasifier
*This balance was based upon data obtained from:
Snyder, N. W., J. J. Brehany, and R. E. Mitchell, East Bay Solid Waste Energy Conversion System, In
Proceedings; Conversion of Refuse to Energy, Montreux, Switzerland, November 3-5, 1975. New
York, Institute of Electrical and Electronics Engineers, p. 428-433.
Bonnet, F. W., Partial Oxvdation of Refuse Using the Purox System, given at Conversion of Refuse to
Energy Conference,Montreux, Switzerland, November 3-5, 1975, but not in Proceedings.
51
U.S. £PA
-------
Despite the excellent quality of the Purox fuel, some communities
in the U.S. that have been considering this system have added at the
back end conventional process technology to produce either ammonia (NH^)
or methanol (CH3OH). Unfortunately, the technical and economic viability
of running such a process on this gas stream remains uncertain.
Liquid Fuel - Occidental Flash Pyrolysis System
Process Description. The Occidental Process (Figure 14) utilizes
two stages of shredding, air classification, magnetic separation, drying,
and screening to produce fluff RDF for the pyrolyse-r feedstock. Representing
about 60 percent of the input solid waste, the fluff RDF is fed along with
hot char into a vertical, stainless steel reactor. The hot char, which is
actually the solid residue remaining after the pyrolysis reaction, provides
the energy needed to pyrolyze the organic material. The material exiting
the reactor consists of a mixture of char and ash and the pyrolysis gases.
By rapidly cooling the gases before they can completely react, a portion of
the gas is condensed into an oil-like liquid fuel. Both the remaining gas
and the char are reused within the system.
In going through the elaborate feedstock preparation steps, a by-product
residual is left which is high in glass and aluminum. This material is
processed by froth flotation to recover a non-color sorted glass cullet
and by linear motor-eddy current cement separators to recover the aluminum.
Product Characteristics. The fuel product will be an oil-like,
chemically complex, organic fluid. The sulfur content will be a good
deal lower than that of even the best residual oils.
The average heating value of the pyrolytic "oil" will be about
(10,500 Btu/lb), compared with 18,000 Btu/lb for typical No. 6 fuel oil.
The lower heating value is due to the fact that pyrolytic oil is lower
in both carbon and hydrogen and containers mcuh more oxygen. A barrel of
oil derived from the pyrolysis of municipal waste contains about 76
percent of the heat energy available from No. 6 oil.
Pyrolytic oil will be more viscous than a typical residual oil.
However, its fluidity increases more rapidly with temperature than does
that of No 6 fuel oil. Hence, although it must be pumped at higher
temperatures than are needed to handle heavy fuel oil, it can be atomized
and burned quite well at 240 F. This is only about 20 F higher than the
atomization temperature for electric utility fuel oils.
The San Diego Gas and Electric Company has agreed to purchase the
fuel for use in one of its existing oil-fired steam-electric power
plants. However, the fuel will first be put through an extensive testing
program to determine its suitability and to determine a price for it.
Status. The first prototype plant is currently under construction
in El Cajon, California. It is being built by San Diego County with
the financial assistance of a demonstration grant from the U.S. Environmental.
Protection Agency and a subsidy from Occidental Petroleum. This 200 ton
52
-------
co
PRIMARY
SHREDDER
AIR
CLASSIFIER
SECONDARY
SHREDDER
TO ALUMINUM
RECOVERY
FROTH
FLOTATION
RESIDUALS
TO LANDFILL TO GLASS COMPANY
FINE SHRED
MAGNETIC
RECOVERY
MIXED COLOR GLASS
PYROLYSIS
REACTOR
PRODUCT
RECOVERY
FUEL
"OIL"
Figure 14. Production of "Oil" from Solid Waste Using the Occidental Process
-------
per day plant was begun in August, 1975 and it is expected that plant
start-up will begin in September, 1976. Several months of start-up
operations will proceed a one year testing and evaluation program.
Energy Balance. An energy balance for the system is shown in
Figure 15. Although electricity and some quench oil is purchased for
the facility, it is assumed in this analysis that these energy inputs
were produced within the system using normal conversion efficiencies
that would result if the pyrolytic fuel product was the prime energy
source. From the figure it can be seen that one pound of solid waste
having a heat value of 5,000 Btu's yield 2050 Btu's of liquid fuel, with
the rest being lost in the residue and char. However, when the 741
Btu's of energy needed to produce the equivalent amount of purchased
energy put into the system is subtracted, only 1309 Btu's (or 26 percent
of the energy in the original pound of solid waste) of fuel remains
available.
DISSIPATED ENERGY
256 Btu
CONVERSION LOSSES
436 Btu
RESIDUE 8 CHAR
3998 Bfu
Figure 15. Occidental Petroleum System Energy Balance *
*This balance was based upon data obtained from:
Flanagan, B. J., Pyrolysis of Domestic Refuse with Mineral Recovery, in Proceedings; Conversion of
Refuse to Energy, Montreux, Switzerland, November 3-5, 1975. New York, Institute of Electrical and
Electronics Engineers, p. 220-225.
Levy, S., The Conversion of Municipal Solid Waste to a Liquid Fuel by Pyrolysis, in Proceedings;
Conversion of Refuse to Energy, Montreux, Switzerland, November 3-5, 1975. New York, Institute of
Electrical and Electronics Engineers, p. 226-231.
54
-------
A conventional boiler using this type of fuel will operate at an
efficiency of 87 percent, so the net amount of energy available from the
original pound of soTid waste, once converted to steam is 1139 Btu's or
23 percent.
Biological Gasification Systems
Anaerobic biological digestion of organic materials is a familiar
and widely used process. Landfill stabilization, domestic sewage stabi-
lization by septic tanks, or municipal sewage sludge digesters all utilize
the same basic process. However, it cannot be said that the process is
well understood. Decomposition of organic materials into methane, water,
and carbon dioxide is the result of the life process of some bacteria which
reside in an obviously complex environment. Some of the effects of varia-
tions in that environment on the health and productivity of the bacteria
are well known while many others are not. Even so, the level of knowledge
and the engineering state of the art are such that the anaerobic digestion
of solid waste could be attractive as an energy recovery and waste disposal
method for the relatively near future. Recent research has greatly expanded
the knowledge about using anaerobic digestion to produce methane from solid
waste.
Anaerobic digestion of wastes requires the action of two types of
bacteria: the acid formers, which are hardy and very resistant to changes
in their environment, and the methane formers, which are strictly anaerobic,
slow growing, and susceptible to upset. There are two steps in the digestion
process: First, the acid formers break down the complex organic materials
into organic acids. Second, the methane formers feed on these organic
acids to produce methane, carbon dioxide, and water.
These biological reations are used to convert refuse into methane
in two different types of systems: landfill gasifiers and biological
reactors. Ther former is considered "developmental" as it is being pursued
in several prototype operations, but the latter is still "experimental."
Landfill Gasifier
Project Description. In California, several landfill operators
are taking advantage of natural phenomena to recover usable methane which
is produced naturally by the decomposing solid waste. The system employs
a deep (over 200 ft. depth) sanitary landfill with impermeable bottom and
gas permeable (porous) daily cover (Figure 16). Cells have been allowed
to attain field capacity (saturated with water) and are then capped. Once
wet, the micro-organisms begin reducing the cellulose in mixed municipal
waste to methane and carbon dioxide. The landfill is equipped with per-
forated well casings which direct the gas into a gas collection system.
Project Characteristics. Gas is primarily methane and carbon
dioxide, but it does contain small amounts of hydrogen sulfide and organic
acids, and the gas stream is saturated with moisture. As a result, unless
the gas is consumed in equipment specifically designed to utilize digester
55
-------
ELECTRICAL RECIPROCATING
GENERATOR ENGINE
ALTERNATE ROUTE-
METHANE AND CARBON DIOXIDE
TO CLEAN UP AND USE
V
en
n
JET
PERFORATED
GAS COLLECTION
TUBE —J
GAS WELL
-IMPERMEABLE
LAYER
Figure 16. Production of Electricity from Landfill Gas
-------
off-gas (as in sewage treatment plants), it must be dehydrated and "sweetened"
(the carbon dioxide is stripped and the acids removed). This method of
energy recovery results in a fuel product which can be obtained from an
existing landfill and may be compatible with utility boilers and residential
fuel requirements.
Status. Pilot or prototype installations are currently recovering
methane from two landfills in Los Angeles, California where both direct
use in a reciprocating engine-generator set and-sweetening to pipeline
quality for residential consumption are being practiced. Also, a research
study, underway at the landfill in Mountain View, California, is attempting
to quantify the various gas recovery parameters so that relaiable technical
and economic surveys can be conducted at other potential sites.
Reactor Gasifiers
Process Description. Reactor-based gasification involves the controlled
introduction of fluff or wet process RDF and sewage sludge into a heated,
well mixed, anaerobic digester where the micro-organisms reduce the
cellulose in the solid waste to methane and carbon dioxide (Figure 17).
The retention time within the reactor will be 5 to 10 days during which the
fastest rate of. decomposition takes place. It would not be economical to
build reactors large enough to hold the material until it is fully
LIGHT FRACTION OUTPUT
DELIVERY TO
FERROUS SCRAP
MARKET
UNRECOVERABLE
RESIDUE TO
SANITARY LANDFILL
GAS OUTPUT*
Figure 17. Biological Gasification of Solid Waste in Reactors
57
-------
digested. Thus, the residue that is removed is not fully digested and
represents about 50 percent of the orginal input weight.- Results from
a pilot plant operated in Franklin, Ohio, indicate that the residue dewaters
easily, and that it contains approximately half of the energy potential
of the input waste. Using vacuum filters and mecahnical presses, it
can probably be dewatered to 55 percent moisture, the same as wet process
fuel. The residue could then possible by used as a boiler fuel in
specially designed boilers. This use would have to be carefully designed
because the residuel is odorous.
Status. The result of laboratory and systems studies indicate
that the technology is promising and would be economic when the price
of gas rises above $2.00 per million cubic feet. Since the intrastate
price of natural gas is above that level in many states today, the process
may prove to be eonomical in the near future. However, the technical
feasibility has yet to be proven at anything above pilot scale experiments.
In addition to the one ton per day gasifier in Franklin, Ohio, Waste
Management, Inc. is designing a 50 tpd prototype unit for construe tion in
Pompano Beach, Florida, under a contract from the Energy Research and
Development Administration.
Among the questions remaining to be answered are: (1.) Is the entire
process economical? (2) Can the reactors operate on mixed municipal solid
waste (people throw out materials such as pesticides which have the capability
of upsetting a digester and preventing it from producing gas)? (3) Is
the existing equipment for mixing the refuse-sludge slurry in the reactors
adequate or is a significant hardware development effort required?
Energy Balance. Figure 18 shows the energy balance for biological
gasification of solid waste mixed with sewage sludge in a reactor. It can
be seen that only one-third of the energy in the "as received" solid
waste is recovered as methane gas. When burned in a boiler having an
efficiency of 85 percent, the net yield is 25 percent. In this analysis, it
has been assumed that energy required to operate the equipment and heat the
digester was obtained by burning the solid residuals recovered from the
digester and the front end system. In addition, recovery of these
residuals adds an additional 633 BTU's of steam per pound of solid waste,
increasing overall system energy yields to 42 -percent. If the residuals
cannot be used as fuel, and system energy requirements are subtracted
from the methane yield, the net energy yield of the system would be
reduced to 14 percent.
Product Characterisitics. Reactor gas is predominantly carbon dioxide
and methane, in almost equal quantities. The gas has a heat value of about
600 BTU/cu. ft., about 60 percent the value of natural gas.
Waste-Fired Gas Turbine Systems
In gas turbine systems (Figure 19) high pressure gases resulting from
the combustion of solid waste with compressed air are used to drive a gas
*
58
-------
CONVERSION LOSSES
74Btu
239 Btu (STEAM)
R/C 8
, STACK
LOSSES
!725Btu
17 F = 33%
BOILER
Tig- 85%
>
^ '
1 ! -
71 — 2
STEAM
R/C 8 STACK
LOSSES
2103 Btu
INCINERATOR
1402 Btu
STEAM
878 Bfu
Tj - 17%
3 RESIDUE
ASH
Figure 18. Energy Balance for Biological Gasification
"This balance was based upon data obtained from:
Pfeffer, J. T., and J. C. Liebman, Biological Conversion of Organic Refuse to Methane, Semi-Annual
Progress Report covering period 7/1/74 to 12/31/74, Department of Civil Engineering, University of
Illinois at Urbana-Champaign. January 1975. p. 64.
Kispert, R. G., L. C. Anderson, D. H. Walker, S. E. Sadek, and D. L. Wise, Fuel Gas Production from
Solid Waste, Semi-Annual Progress Report, Dynatech R/D Company, July 1974. p. 52-58.
turbine. The only example of a waste-fired gas turbine system is the
CPU-400 under development by the Combustion Power Company. In this system
fluff RDF is burned in a fluidized bed furnace (fuel is burned in an
expanded bed of stones) which keeps temperature and excess air low. The
resulting gases are cleaned of fly ash using inertia! separators and
gravel bed filters. The clean gases, at temperatures around 1450 F,
are introduced into a gas turbine-generator set to produce electricity.
Status. Severe difficulties have been encountered in high temperature
particulate (dust) removal. Additional problems due to condensation of
vapor phase "aerosols" in the gases may prove to be inherent. Extensive
R&D programs are now ongoing and, until they are successfully completed, the
status of the system must be considered as experimental.
59
-------
SHREDDER
01
o
SHREDDED WASTE
STORAGE
BAG HOUSE
GRAVEL BED FILTER
FLUID PRESSURIZED
BED COMBUSTOR
EXHAUST DUCTS
TURBINE EXPANDER
TURBINE COMPRESSOR
Figure ] 9. Gas Turbine Generating System Using Refuse as a Fuel (CPU-400)
-------
Energy Balance. The primary energy product in the gas turbine system
is electricity (Figure 20). About 12 percent of the original energy
is recovered as electricity. This number should not be confused with the
yeilds of the other energy recovery systems, which were calculated on the
basis of steam recovery, as there is a substantial energy loss in converting
steam to electicity. Thus if the steam yield from the Monsanto pyrolysis
system, for instance, were converted to electricity, the efficiency would be
reduced to about 16 percent, not much better than the yield for this system.
In addition to the electrical energy recovered from the Brayton cycle,
there is potentially 19.4 percent more energy which can be recovered as steam.
This would require the use of a waste heat boiler to recover energy from the
gases after they pass through the turbine.
BOILER, COMBUSTOR
a TURBINE
R/C 8 STACK
LOSSES
2707 8TU
DISSIPATED ENERGY
T 44 BTU
OVERALL 31.7% OF THE
INPUT ENERGY IS CONVERTED
INTO USEFUL FORMS
Figure 20. Gas Turbine Energy Balance *
"This balance was based upon data obtained from:
CPU-400 Systems Studies and Preliminary Design, Combustion Power Company, Inc., U. S.
Environmental Protection Agency, Office of Research and Development, Cincinnati, Ohio. 72 p.
61
-------
SECTION IV
MATERIALS RECOVERY SYSTEMS
Materials recovery encompasses methods and procedures for extracting
useful materials from solid waste for return to the economy. The
prime objectives in the development of materials recovery systems are:
(1) to conserve natural resources and energy; ('2) to reduce land
requirements for disposal; (3) to facilitate the preparation of refuse
derived fuels for energy recovery systems.
Materials can be recovered through source separation, hand sorting,
or mechanical separation. This report addresses only mechanical
separation. Mechanical separation methods capable of segregating
solid waste into valuable components have developed, based on techniques
use in the mining and paper industries. These methods are aimed at
minimizing the level of impurities in recovered products so that
maximum dollar value can be obtained for the recovered material.
Material recovery systems have concentrated on the reclamation of
fiber or paper (the most abundant component in solid waste); magnetic
metals (the most easily extractable); aluminum (the most highly
valued); and glass (the most difficult to extract).
Material recovery components are often combined with energy
recovery systems. These systems are designed to serve as total
recovery plants. Section III of this report describes systems which
convert part of the solid waste into a fuel product. The following
subsections describe some of the subsystems used in these solid waste
disposal/energy recovery systems to recover valuable resources.
Paper Fiber Recovery
Paper fiber recovery processes use either wet or dry primary
separation of fibers from mixed municipal waste. The initial separation
steps are similar to those employed in dry and wet RDF production
facilities. In fact, it may be practical in some situations to
establish both a fuel and a fiber market for the paper so that the
actual end use of the product can change in response to changes in
market demand (value).
Wet Processing Concept - Fiber Recovery
The major components of a wet processing system are described in
Section III. The fiber recovery portion of the facility is described
here.
Process Description. Figure 21 is a flow chart of the components
of the fiber recovery subsystem. The feedstock to the fiber recovery
process is the same material as taken to the dewatering presses
in a wet fuel processing system. The hydrapulped solid waste is
62
-------
I I RECYCLE WATER TANK
—•"< '
TIPPINfl FLOOR
TO RECOVERY OF.
ALUMINUM
SMALL FERROUS METALS
GLASS (BY COLOR)
(SEE FIGURE 26)
MAGNETICS
Figure 21. Wet Process Fiber Recovery System
-------
centrifugally separated and light fraction taken to fiber recovery.
The first step in the beneficiation process is to remove all large
particles from the slurry. This is accomplished by screening. All
particles greater than 1/16 inch diameter, including the plastic films,
are removed. The fibrous slurry is then passed through a series of high
efficiency centrifugal cleaners and screens which remove grit. The
material exiting the cleaners and screens is recovered fiber. This
material can be washed and dewatered prior to shipment for use in a
manufacturing process or for further upgrading (cleaning and removal
of shorter fibers).
The economic viability of fiber upgrading processes is determined
by the market for the recovered products. Generally, long-term
contracts must be secured and the price of the upgraded fiber must
be sufficiently higher than unbeneficiated fiber to warrant the extra
expense.
Figure 22 presents a mass balance for the recovery of paper fiber
using wet processing. About 20 percent by weight (dry) of the incoming
municipal solid waste is recovered as marketable fiber. This represents
approximately 50 percent of the paper fiber content of the solid waste
on a dry weight basis.
UNPROCESSABLES
6LB
JUNK i I3LB
IREMOVERJ———>TO MAGNETICS
RECOVERY
SOLID WASTE
MAKE-UP
WATER
3.2 GAL.
10LB RECOVERED FIBER
(•*• 10 LB WATER )
TO GLASS RECOVERY
REJECTS- 12 LB
(GLASS, STONES,
ALUMINUM, GRIT, ETC.)
REJECTS FOR
DISPOSAL 38 LB
X (•» 38 LB WATER)
Figure 22. Mass Balance for Wet Process Fiber Recovery
"This balance was based upon data obtained from: v
Wittmann, T. J., et. al., A Technical, Environmental and Economic Evaluation of the "Wet Processing
System for the Recovery and Disposal of Municipal Solid Waste", Final Report SW-109c, U. S.
Environmental Protection Agency, 1975. 217 p.
64
-------
Status. The fiber recovery process was developed by the Black-
Clawson Company and has been demonstrated with a 150 ton per day
plant built in Franklin, Ohio. This plant has been in continuous
operation since 1971. Fiber recovered at the plant is currently
being sold to a manufacturer of asphalt impregnated roofing shingles.
However, recent combustion tests have established a market for the
fiber as a fuel and henceforth the fiber will be sold to the most
lucrative of the two markets.
Although the wet fiber recovery process has been successfully
demonstrated, the large fluctuation which occurs in the paper fiber
market, and the fact that the fiber is suitable for only low grade
uses has limited the systems economic viability. In fact, as discussed
in Section III Black-Clawson is currently promoting the use of its
wet pulped fuel recovery system instead of its fiber recovery system.
Product Characteristics. The paper fiber recovery at Franklin is
of fairly low quality. It is shipped to its market via a short pipeline
as a slurry containing 4 percent solids. The major contaminent impacting
on its quality is oil and grease. Microbial organisms which survive the
pulping and recovery processes are also a problem in that they severely
restrict the "shelf-life" of the fibre and require that its end use
include a heat treating step where the organisms are killed.
Dry Processing Concept - Paper Recovery
Process Description. Another technique for paper recovery is
displayed in Figure 23. The concept involves the recovery of paper
using a series of air classifiers and rotary screens to remove and
upgrade the paper fraction from shredded solid waste. The paper
fraction removed by air classification is baled and either marketed
in this form or further processed using a wet processing system similar
to the wet process fiber recovery system.
As illustrated in Figure 23, the solid waste is first shredded
followed by the removal of magnetics. An air classifier separates
the paper and plastic from the remaining stream. Further air
classifying removes the plastic fraction from the paper/plastic
stream.
When wet processing is used as the final clean up step, the paper/
plastic fraction is charged into hydrapulpers and converted into a
paper-water slurry. Plastics float on the surface of the hydrapulper
and are removed at regular intervals. Heavy foreign matter and more
plastics are removed by screening.
The fiber-rich pulp is discharged from the pulper to a hydracyclone
to remove small particulate matter (grit). The cleaned slurry is pumped
to a prethickener where a large portion of the water is removed for
recycling to the pulper. This thickened material is conveyed to a
65
-------
SECONDARY
DRUM SEPARATOR
CT>
cn
TO FERROUS
RECOVERY
ELECTRO
MAGNET
I
FLAIL
MILL
TO RESIDUE «-
DISPOSAL
TO ALUMINUM*-
RECOVERY
PRIMARY
DRUM
SEPARATOR
AIR
CLASSIFIER
TO PLASTICS RECOVERY
FLAIL MILL
TO FERROUS RECOVERY
AIR
CLASSIFIER
ROTATING
DRUM SEPARATOR
WASTE
CONTAINER
II ^1
u,
TO FIBER USE
Figure 23. Dry Process Paper and Materials Recovery
-------
dewatering press where the material is concentrated to approximately
38 percent solids and additional water is recycled to the pulper.
The fibrous material is then processed in a refiner (mixing, grinding,
and steaming steps) to remove unwanted paraffins and tar residues from
the fiber product.
Status. This process has been developed by the Cecchini Company
of Rome, Italy. Cecchini presently operates three plants in Italy.
Paper from these plants is used, along with straw, to make a low
grade paperboard. No tests have been conducted to determine if such
a product would be marketable in this country.
Product Characteristics. Like the wet recovery system fiber, the
paper recovered in the dry separation system is of low quality, and as
a result, has limited marketability. It's major contaminant is plastics.
Product yields are lower in this process as substantial amounts of paper
are lost in the air classifying and screening operations. It is estimated
that approximately 23 percent of the input paper is recovered as
marketable fiber.
Composting
Composting of municipal refuse is a method of converting the organic
portion of mixed solid waste into a soil conditioner. This conversion
is accomplished by a well known biological process called aerobic
digestion, the decomposition of organic materials by microorganisms
which require air to live. The humus which results from composted
refuse can improve the tilth and moisture retention characteristics
of poor soils. Clays are only temporarily improved by the addition of
humus, but sandy soils can benefit substantially, especially in dry
climates. Composting of municipal refuse has been practiced in Europe
where intensive agriculture by speciality farmers and other small
landholders is carried out close to large towns and cities.
Three basic methods of composting are distinguishable: windrowing -
digesting of the material in open stacks laid on the ground; tilling
the undigested organics into soil containing mature compost; and
completely mechanized industrial composting plants.
The first two processes require large amounts of land, a condition
which rarely exists near today's American cities. The third requires
mechanical equipment. The windrowing process can require as much as
30 days to achieve a mature compost, while the mechanical process
can go to completion in two to ten days. (In most of the United
States, 10 days are required because of the high paper fraction in mixed
municipal waste.)
Composting processes require moisture addition and mixing to
provide adequate aeration of the material. In addition, efficient
composting requires that the organic components in the solid waste
67
-------
be reduced to small particle sizes and that as much as possible of the
inert materials be removed from the waste stream prior to processing.
The size reduction and inerts cleanup requirements for composting are
almost identical to the processing requirements for production of fluff
RDF. The same equipment can be used for both.
Since similar processing is required for the preparation of refuse
for composting and RDF, and since RDF is expected to be more readily
marketable than compost in an urban economy (near the waste generation
centers), it is unlikely that composting will be able to compete with
energy recovery as a solid waste management tool. Furthermore, composted
refuse is a very low grade fertilizer which cannot compete with available
chemical fertilizers on American farms. Finally, the soil in very
few areas of the United States is in need of the type of soil conditioning
offered by humus. The high processing costs (whether in terms of land
or equipment) and the lack of a suitable market indicates that the
composting of municipal solid waste is not a promising method of urban
solid waste management. The possible exception is its use in sections
of the country where sandy soils exist, solid fuel combustion is
economically prohibitive, and a strong, long-term market for humus
exists.
Ferrous Metals Recovery
Process Description. Ferrous metal reclamation is a subsystem
which can be incorporated into almost all, energy and materials recovery
systems. The technology for extracting ferrous metals is based on
magnetic attraction of ferrous materials and is readily available.
Magnetic separation of ferrous metals from municipal solid waste
generally follows the first stage of shredding. In many sophisticated
resource recovery systems, magnetic separators are also employed
later in the system to recover any ferrous metal that was initially
missed. Particle size does not appear to be critical since existing
equipment can easily remove most ferrous objects which appear in
urban solid waste. Bulky items such as appliances can be either
manually sorted or shredded prior to magnetic separation. Heavy
ferrous objects, such as motor casings, are generally manually separated
in order to protect the size reduction equipment.
Two broad classes of magnetic separators are used in solid waste
processing (Figure 24): suspended types and head pully types. Suspended
type separators, positioned over solid waste feed conveyors, are
used to remove ferrous metals from solid waste which may or may not
have been shredded. The recovered ferrous metal is contaminated with
paper so that air scalping or secondary magnetic separation is needed
to produce a marketable fraction. Head pully type separators are generally
employed as a means of secondary ferrous separation.
68
-------
SUSPENDED
BELT MAGNET
SHREDDED
WASTE
AIR KNIFE-
PAPER BACK TO
PROCESS
HEAD PULLEY
MAGNET
DEMAGNETIZED
WASTE
PAPER a
PLASTIC FILMS
MAGNETICS
Figure 24. Magnetic Separator Configuration
-------
The suspended magnetic separator lifts ferrous metals from the
waste and deposits them on a separate belt. The head pully causes
ferrous metals to follow the conveyor around the head and drop behind
the solid waste stream.
Product Considerations. There are three principal uses of ferrous
scrap in the United States today: detinning, steel production, copper
precipitation. Each of these industries have different physical requirements
and contaminant restrictions for ferrous scrap. These markets are
discussed in the Markets section of this guide (SW-157.3). The actual,
or most likely market for ferrous metal recovered from a proposed plant
should be determined before the plant is designed so that the plant can
be designed to produce a ferrous product that will meet the specifications
of the market.
Recovery rates of 90 to 97 percent of the ferrous material in the
waste stream are possible.
Status. The technology of ferrous metals separation and reclamation
is proven and has been demonstrated in numerous areas. Magnetic separation
is being used in almost all operating and proposed energy and material
recovery systems.
Glass and Aluminum Recovery Systems
Recovery of glass and aluminum from mixed municipal solid waste
would occur after the waste has been processed to remove the bulk of
the organic or combustible waste and ferrous metals. Thus, equipment to
recover glass or aluminum would normally be preceded by one or more
stages of shredding, air classification, magnetic separation and screening.
Thus, glass and aluminum recovery can be viewed as a supplement to other
processing and recovery systems. The separation equipment receives a
mostly nonorganic concentrate containing primarily glass, aluminum, and
nonferrous metals, as well as stones and some leftover ferrous metals.
This stream is often referred to as "heavies". Some residual organics
including food, paper, rubber, plastic, and leather are still in the
feed.
Since separation of one of the desired components (e.g. aluminum)
leaves a component with a heavy concentration of the other (e.g. glass),
glass and aluminum recovery are often viewed as joint recovery operations.
However, recovery of only one or the other of the components is clearly
possible.
Aluminum is difficult to extract because it has no unique physical
characteristic which can be used to easily isolate it from the waste
stream, and because it is a minor constituent (generally less than one
percent of municipal solid waste). It's high value as scrap (approximately
300 per ton) however, makes it a potential recovery target.
70
-------
Glass, on the other hand, has a relatively low scrap value but
because it represents a much larger percentage of the waste stream
(about 9 percent) the value of the glass and the aluminum in a ton of
solid waste are nearly equal. The major problem in recovering glass
is that stones and ceramics are not readily separable from glass and
these materials are a major contaminant in the manufacture of glass.
Thus, producing a product which can meet the rigid quality standards
specified by the glass industry is difficult.
Before describing the glass and aluminum recovery systems presently
under development it will be helpful to review some of the unit processes
which are used in these systems.
- Heavy Media Separation. In this process a water suspension
of finely divided particles of heavy minerals (e.g. magnetite
or ferrosilicon) is used to create a fluid having a specific
density which will cause the material being fed to it to
split into "sink" and "float" fractions depending upon the
specific gravities of the particles in the feed. Multiple
separations can be made by using several stages or cells,
each at different specific gravities.
- Eddy Current Separation. This is a dry process for separating
aluminum and other nonferrous metal conductors from non-
conducting materials. In these devices an electrical current
is imposed on a fixed linear motor located beneath a moving
belt. Metal conductors passing through the magnetic field
created by the linear motors are subject to an induced (Eddy)
current which opposes the field created by the linear motor.
The opposing force is strong enough to knock the conductor
off the belt. Non-conductors pass over the linear motors
unaffected.
Combustion Power Company of Menlo Park, California, and
Occidental Research (formally Garrett Research) of La Verne,
California, and the Raytheon Company have developed prototype
aluminum spearation systems using eddy currents. Systems
of this type are reportedly under development that will
include the separation of other nonferrous metals from
aluminum.
- Jigging. This mineral processing technique, is used to
separate materials of different densities. Water is pulsed
through a screen causing material fed onto the screen to
separate. The lighter, material is floated off leaving
the heavier material at the base of the jig. Jigs have been
used in laboratory and pilot scale trials for separation of
aluminum from mixed nonferrous metals.
71
-------
- Electrostatic Separation. This method for dry nonferrous
metal separation is based on differences in the conductivity
of materials. As feed material enter on electrostatic field,
particles become charged and fall on a rotating drum. Con-
ductors immediately lose their charge on the grounded drum and
fall from it while non-conductors retain a surface charge and
adhere to the drum.
- Optical Sorting. Electronic sorting machines are used to
optically separate 1/4 inch to 3/4" inch diameter glass by
color. Glass cullet is fed from a hopper onto a vibrating
feeder (Figure 25). A uniform feed of particles is led to
a grooved belt conveyor which transports pieces in single
file to a separation chamber. Here two photo cells (one
on each side) view the glass. A color plate is situated
opposite the photocell to provide a standard against which
deviations in reflectivity of the glass are measured. Those
particles within a certain range of reflectivity cause a
voltage change in the photocells which in turn triggers a
short blast of compressed air which deflects the particle
from the main stream. This equipment can be set up to
separate transparent particles (glass) from opaque particles
(stones and ceramics), to separate clear glass from colored
glass (amber and green), or possibly, to separate green
glass from amber glass.
- Froth Flotation. This is a standard mineral processing technique
being adapted to glass separation). Froth flotation is
accomplished when an air bubble is attached to a selected
particle having hydrophobic surface characteristics. This
desirable surface property is usually achieved by "conditioning"
the particle using a reagent prior to entering the flotation
circuit.
Following air bubble attachment, the floatable glass particles
are buoyed to the surface to form a froth which can then be
removed by skimmers. Rotors are used to circulate the glass
rich slurry and to provide good air-solids mixing. To achieve
the required residence time, flotation cells are usually
arranged in series with adjacent cells separated by baffles
to reduce "pulp short circuiting."
There are a number of possible configuarations of these unit processes
in combination with grinding and screening to make up a complete recovery
module. Two such systems are described below.
Black Clawson System
Process Description. The first flow scheme is in operation in
Franklin, Ohio, at the Black Clawson fiber recovery plant. The feed
72
-------
FEED BELT
BACKGROUND
SLIDE
LAMP
COMPRESSED
AIR SUPPLY
AIR EJECTOR VALVE
PHOTOCELL
ASSEMBLY
PRODUCT
SPLITTER
SEPARATED PRODUCTS
Figure 25. The Sortex optical sorter is used to color sort glass particles.
to this system is the heavy inorganic fraction (glass, nonferrous metals,
stones and small amounts of ferrous metals and organics) which drops
out of the wet cyclone (see Figure 21). The system as it is currently
laid out is shown in Figure 26. Heavy material from the cyclone is
mechanically dewatered prior to entering the surge storage bin. From
the bin the material is placed on a vibrating screen and the fines and
73
-------
HE/IVY
FRACTION.
HIGH TENSION
ELECTROSTATIC SEPARATOR
OPTICAL SORTERS
(TRANSPARENCY)
(COLOR)
FLINT
GLASS
STONES a CERAMIC
1 NON FERROUS METALS
-FINES
Figure 26. Wet Processs Glass Recovery System
-------
and some dirt with organic residue is washed off; the fines being
arbitrarily defined as anything less than 1/4 inch. This undersize
material will not be color sorted or recovered in any way, and it
is sent to the landfill.
After the screening, the material is magnetically scalped to remove
ferrous metals, and is then conveyed to the heavy media separation unit.
The heavy media separation unit is held at a specific gravity of 2.0
in order to remove any heavy organic materials, specifically plastics,
that have slipped through the liquid cyclone in the main system.
All the floated material is returned to the main plant to be burned
(it would be used as a fuel in a fuel recovery system). The sink
material, that is, the material that has a specific gravity greater
than 2.0, is sent on to the jigging operation for separation of glass
from nonferrous metals - mainly aluminum.
The jigging operation, as set up at the Franklin site, has three
output streams - the lightweight, mostly aluminum can-type stock; the
medium fraction which is mostly glass; and a very hea^y fraction,
composed generally of cast metals, such as brass keys, coins, cast
aluminum, cast zinc, or lead-form material. With the feed material held
for the proper residence time within the jigging operation, good
concentrates of aluminum, glass and heavy metal fractions can be
obtained. The glass fraction is conveyed from the jigging operation
to the rotary kiln dryer to get rid of the excess surface water.
The glass fraction is then carried by a conveyor to the electrostatic
separation unit for removal of any remaining metals.
Material which can be made to carry a charge is pulled out of
the glass rich stream. Some natural stone, residual cast metal materials
and any residual aluminum can stock is thus removed. The use of this
particular device has proved to be very effective for handling materials
ranging in size from 1/4 inch to 1 inch.
The glass fraction coming from the high tension electrostatic
device is then transported by bucket elevators into hoppers which
feed v-shaped belts for the separation of stones and ceramics from
the glass fraction by use of a transparency device. The transparency
device is a relatively new addition to the processing line at Franklin
and is based on the need to remove an extremely high incedence of
ceramic or refractory materials found in the glass fraction. These
refractory materials are unacceptable in the manufacture of glass
containers since their presence causes imperfections in the glass
container which destroy the integrity of the jar or bottle. The
material, once it has been transparency sorted, is then passed
on to a color sorter.
Status and Product Characteristics. In a previous study at
Franklin, the glass composition was segregated into flint, amber
75
-------
and green glass. However, experimentation within the industry determined
that a triple color sorting was not necessary, and that a flint, non-
flint, (amber and green) separation would be sufficient.
While the process appears to satisfactorily sort the glass by
color, achieving high recovery rates and an acceptable product appear to
be contradictory goals. Specifications commonly used by the glass
industry require that the cullet contain a maximum of two stones
per 100 pounds of cullet. As presently operating, the pilot plant is
producing a much lower quality product. Extensive modifications and
tests are being undertaken to remedy the problem. However, at the
present times, color sorting is "developmental" technology.
The Occidental Research Corporation System
Process Description. The Occidental Research Corporation (ORC) has
constructed a pilot glass and aluminum recovery plant which incorporates
froth floatation for glass recovery and eddy current separation of
aluminum (Figure 27).
The material fed to this plant consists of municipal solid waste
that has been shredded to a particle size of 1 inch. The material,
after shredding, has had most of the magnetic metals removed and much of
the organic matter has been removed by air classification. As a result
of this pre-processing, the feed material largely consists of glass,
aluminum, rocks bones, dirt, some magnetic metals, some heavy organics
and other inorganic matter.
The material entering the system flows into a trommel which is a
large rotating cylindrical screen. The large material, which contains
much of the aluminum, passes through the trommel, is conveyed to a
magnetic separate for "tramp" ferrous metal recovery, and then to the
aluminum separator. Here, a linear induction motor, powered by an
alternator generates a force field which acts upon the pieces of aluminum.
The aluminum is very rapidly deflected to the side of the conveyor belt
and is collected.
The small material stream which falls through the trommel screen
openings and is composed of small dense particles, largely glass, is
conveyed to a wet type spiral, classifier. Here the material receives
its first cleaning and the few light organics are removed. The partially
cleansed material then flows by gravity into a rod mill for size reduction.
This sized fraction is pumped through a cyclone and screen where the
large-sized non-glass material (rubber, plastic, etc.) is removed. The
contaminated glass is sized to greater than 200 mesh in a classifier
then flows to a conditioning tank where a proprietary ORC reagent called
"SiLECT" is added. The "conditioned" glass is then sent to a series of
flotation cells called "roughers", "cleaners" and "recleaners". In the
froth flotation cells, the pure glass selectively attaches to bubbles.
and floats to the top of the cells where it is skimmed off, collected,
and sent to a final dewatering classifier. The product glass is then
dried and shipped to market. Rejects from the "roughers" are passed
through "scavengers," dewatered, and then discharged as tailings.
76
-------
MAGNET
TROMMEL
SCREEN
FEED 1
TO
RECOVERY
PLANT
CLASSIFIER "A"
THICKEN-
ING TANK
TO VACUUM
FILTER
JVIAGNETICS
MAGNET v-~~ ~^ REJECTS
(jfrgggScS^
i
NON FERROUS
METALS
SCREEN
- ADDITION OF
"SILECT" REAGENT
CONDITIONING
TANK
METALS AND*
ORGANICS
CLASSIFIER "C"
DEWATERING
Figure 27. The Occidential Research Corporation has set up a glass and aluminum recovery pilot plant
in LaVerne, California utilizing this flow scheme.
-------
Undersize material from the classifier is further processed in a
cyclone and screen * thickener tank and vacuum filter. Water used in the
process is filtered and treated for reuse.
Status. The first full-scale test of this system will be incorporated
in the Occidental Flash Pyrolysis plant now under construction in San
Diego County, California. Until full scale, continuous operational
experience is obtained and market acceptance of the non-color-sorted
cullet has been demonstrated, froth flotation must be classified as
"developmental."
78
-------
SECTION V
READING LIST
Overview
*McEwen, L. B. A nationwide survey of resource recovery activities.
Environmental Protection Publication SW-142.1. Washington, U.S.
Environmental Protection Agency. (In press.)
*U.S. Environmental Protection Agency, Office of Solid Waste Management
Programs. Resource recovery and waste reduction; third report to
Congress. Environmental Protection Publication SW-161. Washington,
U.S. Government Printing Office, 1975. 96 p.
*U.S. Environmental Protection Agency, Office of Solid Waste Management
Programs. Decision-makers guide in solid waste management. Environmental
Protection Publication SW-500. Washington, U.S. Government Printing
Office, 1976. 158 p.
*Smith, F. A. Comparative estimates of post-consumer solid waste..
Environmental Protection Publication SW-148. Washington, U.S.
Environmental Protection Agency, May 1975. 18 p.
Parkhurst, J. D. Report on status of technology in recovery of
resources from solid wastes. [Whittier] , County Sanitation Districts
of Los Angeles, California, January 13, 1976 . 198 p., app.
Energy Recovery
Conference papers; CRE, Conversion of Refuse to Energy; 1st International
Conference and Technical Exhibition, Montreux, Switzerland, Nov. 3-5, 1975.
IEEE catalog no. 75CH1008-2 CRE. DPiscataway, N. JJ , Institute of
Electrical and Electronics Engineers. 615p.
From Waste to Resource Through Processing; Proceedings; 1976 National
Waste Processing Conference, Boston, May 23-26, 1976. New York,
American Society of Mechanical Engineers. 585 p.
*McEwen, L. B., and S. J. Levy. Can Nashville's story be placed in
perspective? Solid Wastes Management/Refuse Removal Journal, 19(8):
24, 28-39, 58, 60, August 1976.
Roberts, R. M., et. al. Envirogenics Company]. Systems evaluation of
refuse as a low sulfur fuel. Washington, U.S. Environmental Protection
Agency, 1971. 2 v. (Distributed by National Technical Information
Service, Springfield, VA, as PB-209 271 - PB-209 272.)
79
-------
+Levy, S. J. A review of the status of pyrolysis as a means of
recovering energy from municipal solid waste. Presented at 3d
U.S. - Japan Conference on Solid Waste Management, Tokyo,
May 12-14, 1976. Washington, U.S Environmental Protection Agency,
Office of Solid Waste Management Programs. 29 p.
*Sussman, D. B. Baltimore demonstrates gas pyrolysis; resource
recovery from solid waste. Environmental Protection Publication
SW-75d.i. Washington, U.S. Government Printing Office, 1975. 24 p.
Davidson, P. E. Slagging pyrolysis solid waste conversion. Engineering
Digest. 21(7):31-34, August 1975.
Anderson, J. E. The oxygen refuse converter - a system for producing
fuel gas, oil, molten metal and slag from refuse. J_n Resource
Recovery Thru Incineration; Proceedings; 1974 National Incinerator
Conference, Miami, Florida, May 12-15, 1974. New York, American
Society of Mechanical Engineers, p. 337-357.
*Levy, S. J. San Diego County demonstrates pyrolysis of solid waste
to recover liquid fuel, metals, and glass. Environmental Protection
Publication SW-80d.2. Washington, U.S. Government Printing Office,
1975, 27 p.
Preston, G. T. Resource recovery and flash pyrolysis of municipal
refuse. J_n_ Clean Fuels from Biomass, Sewage, Urban, Refuse and
Agricultural Wastes Symposium, Orlando, Florida, Jan. 27-30, 1976.
Chicago, Institute of Gas Technology, p. 89-114.
*Hitte, S. J. Anaerobic digestion of solid waste and sewage sludge
to methane. Environmental Protection Publication SW-159. [Washington],
U.S. Environmental Protection Agency, July 1975. 13 p.
Pfeffer, J. T. University of Illino'is, Department of Civil Engineering .
Reclamation of energy from organic waste. Washington, U.S. Environmental
Protection Agency, March 1974. 143 p. (Distributed by National
Technical Information Service, Springfield, VA, as PB-231 176.)
Materials Recovery
*Arella, D. G. Recovering resources from solid waste using wet-processing;
EPA's Franklin, Ohio, demonstration project. Environmental Protection
Publication SW-74d. Washington, U.S. Government Printing Office,
1974. 26 p.
Systems Technology Corporation. A technical, environmental and
economic evaluation of the "wet processing system for the recovery
and disposal of municipal solid waste." Environmental Protection
Publication SW-109c. U.S. Environmental Protection Agency, 1975. 223 p.
(Distributed by National Technical Information Service, Springfield, VA,
as PB-245 674).
80
-------
+Levy, S. J. Materials recovery from post-consumer solid waste.
Presented at 3d U.S.-Japan Conference on Solid Waste Management,
Tokyo, May 12-14, 1976. Washington, U.S. Environmental Protection
Agency. 29 p.
Morey, B., J. P. Cummings, and T. D. Griffin. Recovery of small
metal particles from nonmetals using an eddy current separator -
experience at Franklin, Ohio. Presented at 104th Annual Meeting.
American Institute of Mining, Metallurgical and Petroleum Engineers,
New York City, February 16-20, 1975. 11 p.
Campbell, J. A. Electromagnetic separation of aluminum and nonferrous
metals. Presented at 103d Annual meeting, American Institute of
Mining, Matallurgical and Petroleum Engineers, Dallas,
February 24-28, 1974. 17 p.
Non-ferrous metals recovery...conserving a valuable resource. NCRR
Bulletin. 5(3):67-72, Summer 1975.
McChesney, R., and V. R. Degner. Hydraulic, heavy media, and froth
flotation processes applied to reocvery of metals and glass from
municipal solid waste streams. Presented at 78th National Meeting,
American Institute of Chemical Engineers, Salt Lake City,
August 18-21, 1974. 27 p.
Samtur, H. R. Glass recycling and reuse. IES Report 17. Madison,
University of Wisconsin, Institute for Environmental Studies,
March 1974. 100 p.
Cummings, J. P. Glass and non-ferrous metal recovery subsystem at
Franklin, Ohio - final report. ln_ Proceedings; 5th Mineral Waste
Utilization Symposium, Chicago, April 13-14, 1976. Chicago IIT
Research Institute, p. 175-183.
*Available from: Solid Waste Information Control Section, U.S.
Environmental Protection Agency, Cincinnati, Ohio 45268.
+Available from: Resource Recovery Division, Office of Solid
Waste Management Porgrams, U.S. Environmental Protection Agency,
Washington, D.C. 20460.
81
------- |