Energy
Recovery
from
Waste
This report (SW-36d.i)
on solid waste management
demonstration grant #EC-OOZ12
to the City of St. Louis3 Missouri,
was written
by HOENER & SHIFRIN, INC.
U.S. ENVIRONMENTAL PROTECTION AGENCY
1972
-------
2d printing
1973
An environmental protection publication
in the solid waste management series (SW-36d,i)
For sale by the Superintendent of Documents, U.S. Government Printing Office
Washington, D.C., 20402 - Price 40 cents
-------
FOREWORD
THE AMOUNT OF SOLID WASTE collected in the United
States has increased from 2.75 pounds per person per day in
1920 to 5.3 pounds in 1968, and is projected to increase to 8
pounds by 1980. All the while, America's population and the
pressure on natural resources continue to increase. Growth
in population and in the amount of solid waste collected per
person magnify the seriousness of the disposal problem.
Prevailing disposal techniques— burning and dumping—con-
tinue to waste resources and frequently pollute the air and
water. Moreover, even sanitary landfilling, a desirable and
environmentally sound waste disposal technique, does not
contribute to the recovery of resources. With these facts in
mind, the U. S. Environmental Protection Agency has been
studying various solid waste recycling processes for metro-
politan areas-energy recovery, materials separation and reuse,
and chemical conversion processes.
The particular process described in this report is the recovery
of thermal energy by burning shredded residential solid waste
as supplementary fuel in boiler furnaces. This process was
initially studied in 1968 by the City of St. Louis with partial
funding from the Federal solid waste management program.
The results of that study were encouraging; the required
equipment is commercially available, and a power company
serving the metropolitan area, the Union Electric Company,
concurred with the high probability of success of the process
by offering to the City its full cooperation. This included
providing two test boilers and offering to contribute a sub-
------- OCR error (C:\Conversion\JobRoot\000006XG\tiff\2000PW8J.tif): Unspecified error
------- OCR error (C:\Conversion\JobRoot\000006XG\tiff\2000PW8K.tif): Unspecified error
-------
The quantity of ground-up refuse used for fuel is only a
small percentage (nominally 10 percent by heating value) of
the fossil fuel normally fired to the boilers. However, the
two test boilers fitted to burn refuse, although of only
moderate size, can each consume about 300 tons of waste
material per day at the 10 percent burning rate. This ton-
nage is equivalent to the residential solid waste generated by
170,000 people. If it is found feasible to burn larger per-
centages of refuse as fuel, the quantity of refuse disposed of
by this means will increase accordingly.
BACKGROUND
All major metropolitan areas within the United States face
an increasing problem in the disposal of solid wastes. For the
most part, the technology utilized up to this time for solid
waste disposal has not been sufficiently sophisticated to
permit efficient conservation of natural resources. Further,
it is apparent that full advantage has not been taken of even
proven areas of technology in the disposal of solid wastes.
The recovery of waste heat resulting from the combustion of
solid waste materials is not a new concept and has been prac-
ticed in Europe for years. The practice has not been common
in the United States; until recently it was confined to relative-
ly inefficient waste heat boilers installed in conventional refuse
incinerators. A few more sophisticated solid waste incinerators
now are being built, incorporating boilers for the production
of steam. These newer facilities unfortunately are quite
costly and, to be effective as resource recovery units, must
have markets for the steam they are designed to produce.
Such markets are not always readily available.
Consider, however, a large, efficient, existing utility power
plant, already integral with a system for producing, distribu-
ting and marketing electricity. It follows that if a means were
devised to burn refuse as fuel in existing power plant boilers,
-------
without creating significant adverse effects, the matter of
finding a market for the energy available from refuse would
automatically be resolved, since utilities can market essentially
all of the power they have the capability of producing.
The St. Louis project was conceived under the premises that
if residential solid waste were properly prepared, and if it
were to replace only a small percentage of the fuel fired to
coal-fired boilers, the effects would be little, if any, different
than if the fuel were entirely coal. Although coal-fired power
plant boilers are not without operating problems of their own,
a comprehensive study concluded that such problems would
not be significantly increased, if increased at all, by burning
prepared refuse as supplementary fuel. That study was con-
ducted for the City by Horner & Shifrin, Inc. with the close
cooperation of the Union Electric Company. (1)
The concept was found attractive enough for the Union
Electric Company to offer the use of two of its major power-
producing boiler units for full-scale tests, and to contribute a
substantial sum ($550,000) toward the installation of facilities
to implement these tests. Also intrigued by the concept was
the Office of Solid Waste Management Programs of the U. S.
Environmental Protection Agency, which partially funded
the original study, and which, together with the Office of Air
Programs, is partially funding the construction and operation
of the facilities required for the full-scale demonstration
program.
PROCESS DETAILS
Residential solid waste, in its raw state, is remarkably hetero-
geneous in appearance. After milling, however, the refuse
(1) Horner & Shifrin, Inc., "Study of Refuse as Supple-
mentary Fuel for Power Plants," St. Louis, Missouri,
1970. 95 p.
-------
TABLE 1
COMPOSITION OF RESIDENTIAL SOLID WASTE
AND COAL SAMPLES
(AS RECEIVED)
Refuse '
Coal +
Proximate Analyses
Moisture
Ash
Volatile
Fixed Carbon
Btu per Pound
Ultimate Analyses
19.69 -31.33
9.43-26.83
36.76 - 56.24
0.61 -14.64
4,171 -5,501
6.20-
9.73-
34.03 -
42.03 -
10.23
10.83
40.03
45.14
11,258 - 11,931
Moisture
Carbon
Hydrogen
Nitrogen
Chlorine
Sulfur
Ash
Oxygen
19.69
23.45
3.38
0.19
0.13
0.19
9.43
15.37
-31.33
- 33.47
- 4.72
- 0.37
- 0.32
- 0.33
- 26.83
- 31 .90
6.20-
61 .29 -
4.49 -
0.83-
0.03-
3.06-
9.73-
9.28-
10.23
66.18
5.58
1.31
0.05
3.93
10.83
16.10
* From three samples of St. Louis residential solid waste, with magnetic
metals removed.
+ From three samples of Union Electric Company coals.
becomes more homogeneous, in both appearance and con-
sistency. Within reasonable limits, the analyses of refuse from
different parts of the country have been found to be surpris-
ingly consistent.
Investigations of the quality of residential solid waste pro-
duced in the St. Louis area, although limited in scope, dis-
cjosed characteristics similar to those found in other parts of
the country. A comparison of the ranges of composition of
refuse and coal for the St. Louis area was made (Table 1). As-
-------
received calorific values of refuse were found to be in the
general range of 4,200 to 5,500 Btu per pound. The ash
content was found to be about twice that of coal. Ash
fusion temperatures were apparently very similar to those of
Illinois bituminous coal. The sulfur content was found to be
only a fraction of one per cent. The chlorine content, how-
ever, was found to be higher than in most coals.
PROCESSING FACILITIES
The solid waste to be processed for use as fuel, at least
initially, will be confined to that produced from households.
No bulky materials, such as appliances, furniture or tires, or
wastes from industrial or commercial establishments will be
processed. However, certain selected industrial and com-
mercial wastes may be processed later.
Refuse processing, to produce 300 tons per day of supple-
mentary fuel, will be accomplished during one 8-hour shift.
Since magnetic metal will be removed, the equivalent quantity
of raw refuse requiring processing will be about 320 to 330
tons per shift. Two-shift operation of the processing facilities
will be required if the supplemental fuel is fired at rates
above the nominal 300-tons-per-day rate. This will still
permit the third shift to be devoted to scheduled maintenance
of refuse grinding (hammermill) and other handling equip-
ment. The hammermill and conveyors have been selected to
provide a nominal production rate of 45 tons of raw refuse
per hour.
Raw refuse is discharged from packer-type trucks to the
floor of the raw-refuse-receiving building. Front-end loaders
are used to push the raw refuse to a receiving conveyor. This
method of handling raw refuse was selected over the pit and
crane method in order to achieve more uniform feeding rates,
to effect economy, and to provide an additional means of
-------
Belt Conveyor
Magnetic Separator
Magnetic Metals
ck
Conveyor
Self-Unloading
Transport Truck
Storage Bin
Belt Conveyor
Belt Conveyor
Belt Scale
Stationary Packer
REFUSE PROCESSING FACILITIES
FIGURE 1
controlling the quality of the material being processed. From
the receiving conveyor, the raw refuse is transferred to an
inclined belt conveyor, which in turn discharges to a vibrating
conveyor, which feeds the hammermill directly (Figure 1).
The hammermill is a conventional mill with a horizontal
shaft. It has a hammer circle of about 60 inches, and an
interior rotor length of about 80 inches. At the time the
design of the mill was agreed upon, there was no evidence
that any other type of mill could provide either the produc-
tion rate required or the control over particle sizes desired
for purposes of this process. The mill is powered by a direct-
connected 1,250-horsepower, 900-rpm motor. Single-stage
milling was deemed appropriate for purposes of the prototype
installation, although two-stage milling normally is advocated
for this type of operation. The mill has a grate cage which
provides openings of about 2 inches by 3 inches. Tests run
on a similar but smaller mill indicated that essentially all
milled particles would be less than 1-1/2 inches in size, that
96 to 98 percent by weight of the particles would be less than
-------
1 inch in size, and that about 50 percent of the particles
would be less than 3/8 inch in size. Uncompacted bulk
density of the milled material, depending upon moisture
content and composition, was found to be variable, as low as
4 pounds per cubic foot in some cases, and as high as 12
pounds per cubic foot in others. This variation in density
poses problems in equipment design and selection, since some
equipment is designed on a gravimetric basis and others on a
volumetric basis.
From the hammermill, the milled material is discharged to
another vibrating conveyor, feeding an inclined belt conveyor
leading to a storage bin. Magnetic separation is effected at
the head pulley of this belt conveyor, with the magnetic
metals discharged to trucks for sale as scrap.
The storage of milled refuse poses problems requiring
special attention. For the prototype installation, since a
convenient alternate means of refuse disposal was available
(the processing plant was constructed adjacent to an existing
city solid waste incinerator) and since the interruption of
refuse-firing to the boilers would not cause significant opera-
tion problems, it was concluded that only minimum storage
volume for the milled material was necessary. The storage
volume provided, therefore, is only sufficient to permit a
relatively even flow of supplementary fuel to the boilers.
Milled refuse, having laminar characteristics, has a bridging
tendency, and storage bin design must be such that bridging
will be minimized. The most effective means of preventing
bridging appears to be the construction of bins with a greater
cross section at the bottom than at the top. It also is necessary
to provide a bin-unloading device which will remove material
from all parts of the bin bottom without resorting to the use
of hoppers, in which the material almost certainly would
bridge. For this reason, traversing augers are being used to
unload the storage bin.
-------
From the storage bin, the supplementary fuel is conveyed to
a stationary packer for loading into self-unloading trucks for
transport to the power plant. In the prototype project, the
power plant is located about 18 miles from the processing
plant. At the nominal 300-ton-per-day firing rate, only one
25-ton load of supplementary fuel will be delivered to the
power plant every two hours. If it were feasible to locate the
processing facilities near the power plant, it would be possible
to eliminate truck transport by pneumatically conveying the
supplementary fuel directly to the boilers from the storage
bin.
RECEIVING AND FIRING FACILITIES
The self-unloading mechanisms of the transport trailers dis-
charge the supplementary fuel to a receiving bin, from which
the material is conveyed to a pneumatic feeder for transfer to
a surge bin (Figure 2). The surge bin is equipped with four
drag chain unloading conveyors, each of which feeds a
pneumatic feeder. Each of these four pneumatic feeders
conveys the supplementary fuel through a separate pipeline
- Self-unloading Transport Truck
- Receiving Bin
-Boiler Furnace
Surge Bin
n c j Drag Con
Pneumatic Feeder^ 3
Blower
Pneumatic Feeder
/
V
To Precipitator
Bottom Ash
SUPPLEMENTARY FUEL RECEIVING AND FIRING FACILITIES
FIGURE 2
-------
directly to a firing port in each corner of the boiler furnace.
The pipelines are about 700 feet long.
The pneumatic systems are of the high-pressure type, in
which the material to be conveyed is introduced by a rotary
air lock feeder into a pressurized pipeline. The air velocities
of the conveyed particles, depending upon their mass, are
expected to be about 50 to 70 feet per second. The initial
pressures in the pipelines depend upon their length as well as
upon the quantity of material to be conveyed, and will
normally be several pounds per square inch. The boilers are
operated with balanced draft, with a slightly negative pressure
in the furnaces.
The division of responsibility between the City of St. Louis
and the Union Electric Company for this project was deter-
mined to be between the receiving facility and the surge bin.
The operation of the surge bin and the pneumatic boiler
firing systems have been established to be the responsibility
of the utility. The delivery of the supplementary fuel and
its transfer to the surge bin is the City's responsibility.
TEST BOILER
The two twin boilers to be used for testing in the St. Louis
project are small when compared to the newer units in the
Union Electric Company system, but they are of modern,
reheat design, and the test results from these units should be
applicable to numerous other similar existing units in service
in many parts of the country. Built by Combustion Engineer-
ing, Inc., whose personnel cooperated fully in the original
study, the units each have a nominal rating of 125 megawatts,
and each will burn about 56.5 tons of Illinois bituminous
coal per hour at rated load. The units are tangentially fired,
with four pulverized coal burners in each corner of the fur-
naces. Since they also are fitted to burn natural gas, they
will be capable of burning refuse with gas as well as with coal
-------
during the test program. The furnaces are about 28 feet by
38 feet in cross section, with a total inside height of about
100 feet (Figure 3).
At full load, the quantity of refuse for each boiler equivalent
in heating value to 10 percent of the coal is about 12.5 tons
SUPERHEATER
ECONOMIZER
FEEDERS
MERAMEC UNIT NO. 1
UNION ELECTRIC COMPANY
COMBUSTION ENGINEERING, INC.
FIGURE 3
10
-------
per hour, or 300 tons per 24-hour day. Refuse will be fired
24 hours per day, but only 5 days per week, since City
residential solid waste collections are scheduled on a 5-day-
per-week basis. No difficulty in boiler operation accruing
from this interrupted refuse-firing schedule is anticipated.
Other than installing a refuse-burning port in each corner of
the furnaces, no modifications to the test boilers were made.
The refuse-burning ports were installed between the two
middle coal burners. No alterations to the pressure parts of
the boilers were necessary. The milled refuse is burned in
suspension, in the same flame pattern as the pulverized coal
or gas. As is typical of utility grade boilers, the furnaces
have no grates. Nonburnable particles, or burnable particles
with sufficient mass to prevent them from being consumed in
suspension, therefore are likely to fall to the bottom ash
hopper.
The prepared refuse is fired at a constant rate. The existing
boiler combustion controls vary the rate of firing of the
pulverized coal or gas to accommodate the heat requirements
of the boilers. Should the boiler go out of service suddenly,
for any reason, an electrical interlock will immediately stop
the feeding of refuse, although the pneumatic blowers will
continue to function in order to clear the pipelines of any
refuse remaining in them.
No modifications were made to either the electrostatic
precipitators or the bottom-ash-handling systems. Although
a significant increase in the quantities of ash was anticipated,
particularly in bottom ash, the existing ash handling facilities
were regarded as adequate, at least for purposes of the test.
Since both fly ash and bottom ash are presently marketed,
tests will be conducted to ascertain changes in ash quality as
a result of burning refuse as supplementary fuel. However,
the low refuse-burning rate is not anticipated to substantially
change the ash characteristics.
11
-------
Combustion Engineering, Inc., in their initial appraisal of the
applicability of the process to the test boilers, concluded that
the air and gas weights would be very close to the original
design conditions, and that the forced draft and induced
draft fans would be adequate.
POTENTIAL BOILER OPERATING PROBLEMS
Although the limitation of burning residential solid waste
as only a small percentage of the total fuel should tend to
minimize any potential additional boiler operating problems,
there are several potential effects which are being given special
attention in the tests. Slagging, corrosion, completeness of
burn-out, and precipitator performance are, therefore, being
thoroughly evaluated.
Some slagging can be expected in most coal-burning boilers,
with the degree of slagging related to coal quality and to
operating procedures. Preliminary investigations indicated
that the ash fusion temperatures of refuse were in the same
range as those of Illinois bituminous coal. Other coals,
particularly those with low sulfur contents, may have higher
ash fusion temperatures. The tests will include observations
of potential slagging tendencies of mixtures of refuse ash and
coal ash.
An evaluation made by Combustion Engineering, Inc., during
the preliminary studies indicated the possibility of a slightly
increased corrosion potential. Probes will be installed at
strategic points in the boiler to assess the effects, if any, of
corrosion. The removal of magnetic metals, along with the
zinc, lead, and tin that often will be bonded to them, is ex-
pected to reduce the potential adverse effect of corrosion and
plating caused by metal oxides.
Since utility boilers are not normally equipped with grates,
there was some concern that larger burnable particles would
12
-------
not be completely consumed in suspension. The small particles
achieved in the type of milling equipment provided is ex-
pected to decrease this possibility, particularly in view of the
large percentage of paper in refuse.
When burning pulverized coal and milled refuse in combina-
tion, some increased particulate loading may occur in the
boiler exhaust gases which are cleaned by an electrostatic
precipitator. Therefore, tests will be conducted to determine
the relative precipitator performance. Air pollution tests will
also be conducted on the stack gases to evaluate the effect of
refuse combustion upon gaseous pollutant emissions. It is
expected that the firing of refuse and coal will not produce
discernible increases in either gaseous or particulate emissions
when compared to burning coal alone.
ECONOMICS
Of particular importance are the relative economics of the
application. The basic economic elements are: (1) the total
unit cost of processing, transporting, and firing the refuse as
supplementary fuel, including those costs accruing to the
utility; (2) the costs of alternate means of residential solid
waste disposal; (3) the value of the magnetic materials re-
covered; (4) the value of the milled residential solid waste as
fuel when compared to the value of fossil fuel. Under almost
any situation, the capital cost of the facilities required for
application of the process would be substantially less than
that of conventional incineration facilities. Total unit costs
per ton, including operation, maintenance, and amortization,
could be comparable to those of sanitary landfill when the
landfill is in a location remote from the point of refuse
generation.
Utility fuel costs are rising rapidly, and even under present
conditions the relative value of refuse as fuel could be signifi-
13
-------
cant. The net cost of residential solid waste disposal by this
means, therefore, could be highly attractive in many of the
metropolitan areas of the United States.
APPLICABILITY OF PROCESS
The prototype installation for burning residential solid
waste as supplementary fuel includes the adaptation of
tangentially-fired boilers. Indications are that such units are
the most easily adaptable of the various types in common
use. However, there appears to be no reason why cyclone-
fired boilers could not be adapted, or why front-fired and
opposed-fired units could not be fitted with refuse-burning
ports, even if some modification of pressure parts were to be
required.
Although the original investigations were confined to pul-
verized coal-burning boilers, the test boilers are fitted to burn
natural gas as well, and therefore are capable of burning solid
waste with gas as well as with coal during the testing program.
There is no reason why boilers burning oil also could not be
adapted to burn solid waste, provided they have the capability
of handling both bottom ash and fly ash.
The capability of existing suspension-fired boilers to consume
municipal solid waste as supplementary fuel is great enough
to permit the process to serve as a principal means of solid
waste disposal for many metropolitan areas, even when the
supplementary fuel is fired as only 10 percent of the total
fuel requirement. A 600-megawatt unit, for example, could
at full load consume on the order of 1,200 tons of solid waste
per day at the 10 percent rate. By 1973, the suspension-fired
boilers in the St. Louis area will have the potential capability
of consuming over twice the amount of municipal solid waste
produced in the entire metropolitan area with a population
of over 2,500,000.
14
-------
The process cannot be expected to be applicable in all
metropolitan areas. Some types of fossil-fuel-fired boilers are
not as adaptable as others. Boilers operating essentially as
base load units may be more desirable for burningsolid waste
than those operated at partial loadings. Cooperation between
utilities and municipal corporations sometimes may be
difficult to achieve. The quantities of supplementary fuel
available in some areas may not be sufficient to be of interest
to the utility. The location of the power plant may not
always permit economical application of the process.
In the case of the prototype project, the City of St. Louis is
delivering the prepared residential solid waste to the utility
at no cost for the duration of the tests. Depending upon the
circumstances of a given case, this same arrangement could be
mutually attractive to another combination of a city and a
utility, or a city might be willing to pay the utility a nominal
fee for using the solid waste, or the utility might be willing to
pay the city a nominal fee for the supplementary fuel. Each
situation requires an objective appraisal to determine the
appropriate basis for negotiation.
The process shows promise of having fairly extensive ap-
plication once a satisfactory financial arrangement can be
worked out between the utility and the solid waste supplier,
whether it be a governmental unit or a private organization.
It could provide an economical means of disposing of large
quantities of solid waste, effect recycling by the direct
recovery of electrical energy, conserve natural resources of
fossil fuel and magnetic metal, and, to some degree, assist in
the control of air pollution. The preliminary appraisals
indicate that these potential benefits may be achieved by
effecting sufficient cooperation between municipalities and
utilities to permit use of existing large-scale power plant
facilities, by using commercially available equipment and by
applying already tried and proven basic technology.
ya72342R
*• U. S GOVERNMENT PRINTING OFFICE 1973 — 514-153/204 "| 5
-------
-------
-------
''''"'''".,.«11' "'"'"'..I ']" '.., I "' frrrp" I ' i'1 'ii 7I
------- |