United States Office of Water &
Environmental Protection Waste Management SW - 773
Agency Washington D.C. 20460 September 1979
Solid Waste
&EPA Dust and Airborne Bacteria
at Solid Waste
Processing Plants
-------
-------
DUST AND AIRBORNE BACTERIA
AT SOLID WASTE PROCESSING PLANTS
This report (SW-773) was prepared
under contract for the Office of Solid Waste
by D. E. Fiscus, P. G. Gorman, M. P. Schrag, and L. J. Shannon.
U.S. Environmental Protection Agency
1979
-------
An environmental protection publication (SW-773) in the solid
waste management series. 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.
Single copies of this publication are available from Solid Waste
Information, U.S. Environmental Protection Agency, Cincinnati, Ohio 45268.
ii
-------
FOREWARD
This report was prepared for the Municipal Environmental Research
Laboratory, U.S. Environmental Protection Agency, Cincinnati, under
EPA Contract No. 68-02-1871 to the Midwest Research Institute. It
is intended to provide the Office of Solid Waste with a document
that may be disseminated to the public so they may be better Informed
about refuse processing facilities and activities.
This report was written by D. E. Fiscus, P. G. Gorman, M. P. Schrag,
and L. J. Shannon.
ill
-------
PREFACE—METRIC UNITS
This report has been written using the International System of Units
(SI) or metric units. English units in parentheses have been used only
occasionally in the report. Most readers are probably familiar with tht
conversion of feet to meters, cubic feet to cubic meters, and kilogram* to
pounds. A somewhat more unfamiliar term for large weights Is the use of •
megagrams instead of tons. A megagram is 1,000 kilograms, and a kilogram
is 1,000 grams. However, a megagram is only slightly larger'than a ton. A
ton is approximately 9/10 of a megagram.
The standard unit of time in the SI system is the second. Therefore,
airflow rates are expressed in cubic meters per second instead of cubic
feet per minute.
In the English system, dust concentrations are usually expressed as
grains of dust per cubic foot of air. "Grains" may not be famllar to
readers except those who work closely with dust control or air pollution.
Grains are used because dust concentrations are usually quite small, pounds
and ounces are too large a unit of weight to be useful. There are 7,000
grains in a pound. The SI unit most commonly used for dust weight is the
milligram, which is 1/1,000 of a gram.
The folloiwng table includes the conversions from SI units to
English units for the measures used in this report.
Measure
Weight
Air flow
Dust
Multiply Metric SI Units
Megagram
(Mg)
Cubic meters per second
(cu m/sec)
Milligrams per cubic meter
Bacteria Counts per cubic meter
concentration (count/cu m)
by_ To Obtain English Units
1.102 Tons
21.10
Cubic feet per minute
0.000437 Grains per cubic foot
(gr/cu ft)
0.0283 Counts per cubic foot
(count/cu ft)
iv
-------
CONTENTS
St. Louis and Ames 1
The Dust Problem 1
St. Louis 2
Fugitive Dust 2
Cleanup 3
Complaint by a Neighbor . ............. 3
Ames 3
Cleanup and Working Conditions ..... 3
Mechanical Problems 3
Dust Control 3
Other Problems 5
Fire Hazard 5
Microorganisms 5
Test Programs 5
Dust Measurements 5
Microorganisms 5
Dust Collection Systems • 6
Dust Concentrations 6
Solid Waste Processing Plants Tested 6
Test Procedures .......... 6
Dust Concentration Test Results 8
Bacteria Concentrations 10
Solid Waste Processing Plants Tested 10
Refuse-Derived Fuel Plant 10
Incinerator 10
Waste Transfer Station 11
Sanitary Landfill 11
Wastewater Treatment Plant 11
Amount of Waste Handled at Each Plant 11
Other Bacteria Measurement Locations . 12
-------
CONTENTS (Concluded)
Microorganisms Measured
Test Procedures . . . .
Mircroorganism Test Results
Range of Bacteria Concentrations
Waste Processing Plant Comparisons
Dust Collection Systems
Collection Equipment .
Collection Efficiency
Dust Collection Efficiency
Bacteria Collection Efficiency
Design Considerations
Dust Management
Dust Collection and Filtration ....
Cost of a Dust Collection and Filtration System
Page
12
13
14
14
15
17
17
18
18
19
19
19
20
21
FIGURES
Number
1
2
Flow diagram of St. Louis RDF plant
Flow diagram of Ames RDF plant . .
4
7
TABLES
Number
1
2
3
4
5
6
7
8
9
Dust Measurement Results .....
Air Flow Measurement Results
Amount of Waste Handled at Test Plants
Range of Bacteria Concentrations
Ranking Based on Average Bacterial Levels in Descending
Order from Highest to Lowest Concentration
Dust Collection Efficiencies . .
Bacteria Collection Efficiencies
Dust Collection Costs .
Dust Control Cost Model
9
9
12
15
16
18
19
22
23
vi
-------
DUST AND AIRBORNE BACTERIA AT SOLID WASTE PROCESSING PLANTS
Municipal solid waste (MSW) has for years been collected by the
familiar "garbage truck" and hauled to sanitary landfills. Occasionally
waste transfer stations are used where MSW from small collection trucks
is unloaded and then reloaded into larger trucks for transport over long
distances to a landfill.
In the early 1970's, the concept emerged that MSW was not altogether
a waste material but contained valuable resources. Resource recovery
plants where MSW is processed into a refuse-derived fuel (RDF) are now
in operation in several cities and more are being planned.
St. Louis and Ames
The first RDF plants were at St. Louis, Missouri, and Ames, Iowa.
The St. Louis facility was a demonstration project supported by three
cooperative parties: the Environmental Protection Agency (EPA), the
City of St. Louis, and Union Electric Company. The project was com-
pleted in 1975, and the plant is no longer operated. The Ames plant
is an ongoing commercial facility owned and operated by the City of
Ames, Iowa.
The above plants are similar, and each is typical of the RDF
process: the MSW is reduced in size by a shredder or hammermill, and
the shredded MSW is "air classified" to separate the paper and plastic
combustibles (RDF) from the nonflammable "heavies." Metals such as
ferrous (i.e., contains iron or steel) scrap are removed magnetically
and sold as by-products. The air-classified combustible material (RDF)
is combine-fired with coal in electric utility boilers. RDF is usually
10 to 25 percent of the total fuel requirement, with coal constituting
the remainder.
The St. Louis plant used a single shredder and was an unenclosed
plant with all the processing equipment located outside. By contrast,
the Ames plant is completely enclosed with all the processing occurring
inside a metal building. Ames also uses double shredding where two
shredders are installed in series to accomplish size reduction. At
St. Louis, the air flow from the air classifier was discharged directly
to the atmosphere, while at Ames the air discharge is recycled into the
plant.
The Dust Problem
When MSW is shredded, a small portion of the MSW becomes very fine
particulate or dust. This dust tends to be fibrous or linty and is
-------
similar in appearance to the dust collected by a household vacuum cleaner*
Because it Is light and fluffy, this dust is easily blown about if not
contained. Therefore, at both St. Louis and Ames, dust has been a problem.
St. Louis. Figure 1 is a flow diagram of the St. Louis plant.
Refuse collection trucks discharged their loads onto a concrete
tipping floor inside a building. The tipping floor is where the
refuse collection trucks discharged their loads. The refuse 'was
pushed by a front-end loader onto a conveyor belt feeding a shredder.
Shredded refuse was carried by a conveyor belt to the air classifier,
and the air-separated RDF was carried by another conveyor belt to a
storage bin. This RDF was removed from the storage bin and carried
by a conveyor belt to a packer station where large trucks were .loaded
for transporting the RDF to the power plant. There the RDF was
combine-fired with coal in an electric utility boiler. Heavier
(reject material) from the air classifier was carried by a conveyor
belt to a magnetic separator where ferrous metal was removed and
sold for scrap.
Fugitive Dust. The cause of the ambient dust problem at the
St. Louis plant can be attributed to: (1) exhaust from the air
classifier cyclone collector; (2) shredder operation; (3) spillage
from conveyors.
In air classifier, shredded MSW is fed into an upward moving
stream of air which lifts and carries the lighter portion of the MSW
with it while the heavier material falls downward. A cyclone separator
is used to separate the "lifted" material (RDF) from the air stream.
At St. Louis, the air exhaust from the cyclone was vented directly
to the atmosphere. Cyclone separators have been used for many years,
and their principles of operation are well known. Their removal
efficiency is quite high on large particles. Therefore, at St.
Louis, the air classifier cyclone exhaust carried with it small
particles of dust. Tests conducted to measure the rate of the air
classifier cyclone dust emissions showed the average emission rate
to be 27 kg/hr (60 Ib/hr). As this discharged dust was already
airborne, it was quickly blown about the plant area.
The shredder at St. Louis was a large hammermill with a horizontal
shaft fitted with large swinging hammers. MSW entered the shredder from
the top and was torn apart by the action of the hammers until it was
small enought to fall through square grates in the bottom of the shredder.
The action of the hammers inside the shredder created a "windage effect"
which tended to blow dust out of the mill. To reduce the dust discharge,
the conveyors to and from the shredder were enclosed, and the shredder
was fitted with a small fan and cyclone collector as a partial dust
collection system.
Spillage resulted from all the various plant operations, especially
when equipment was cleaned for routine maintenance or when breakdowns
-------
occurred. Waste material, whether It was shredded MSW, RDF, heavies, etc.,
contained dust which was easily blown about the area. Also, small pieces
of paper and plastic bags present in the spillage were picked up by air
currents and blown about. The net result was fugitive dust around the
plant area.
Cleanup. At St. Louis, processing was conducted only during the
regular day shift. During the second shift, a two-man crew was used
to clean the area. Part of the work was to open and clean out the
equipment (primarily the shredder). However, some labor was required
to cleanup the fugitive dust at the plant. Because the immediate plant
area was black-topped, a large water hose was used to clean the equipment'
and the surrounding area.
Complaint by a Neighbor. Even though the cleanup work was the best
that could be expected under the circumstances, the plant received com-
plaints from an adjoining neighbor, an industrial storage depot. This
resulted in the air classifier exhaust air being ducted to a large
settling chamber fitted with a plastic mesh screen. This helped, but
did not eliminate, the fugitive dust problem.
Ames. The Ames facility is still an ongoing municipal plant.
Figure 2 is a flow diagram of the Ames plant which is very similar to
the St. Louis plant; however, there are several major differences.
Ames uses two shredders with magnetic separation between the first and
second shredders; a pneumatic conveying system is used to deliever the
RDF to the storage bin; and an aluminum separation system is used.
Ames has the same type of fugitive dust problems as St. Louis had,
except no neighborhood complaints have been received because all the
processing equipment is enclosed within a building. Therefore, little
fugitive dust escapes to the surrounding neighborhood. The exhaust air
from the air classifier is recycled to the air classifier air intake.
While no dust is exhausted out of the building, this recycling increases
the dust level within the plant.
Cleanup and Working Conditions. Like St. Louis, Ames uses a cleanup
crew. Because the dust is confined within a building, the dust is more
concentrated, resulting in undersirable working conditions. Some plant
employees wear dust masks when working for long periods within the
processing area.
Mechanical Problems. The outdoor plant at St. Louis was cleaned by
the rain and by periodic washdown with a water hose. At Ames, the plant
is enclosed and was not designed for washdown. Therefore, the dust within
the Ames plant may have a greater probability of harming motors and
bearings and of clogging equipment air filters.
Dust Control. After two years of operation, the City of Ames decided
that a plant dust collection system was needed. Plant personnel anticipate
-------
z
o
0£
O
a.
oo
Z
<
0£
k.
V
_*
o
Q_
X
l-
o
c
o
o
^•B
A
^\\
c
_o
a.
k.
0)
0
Q_
-o
o
£
O
JJ
c.
as
a
a;
«3
U
CO
re
o
_<
Pu
I
D
00
-------
that this dust collection system will reduce plant costs by reducing
cleanup labor and equipment maintenance*
Other Problems
Other problems that may be traced to fugitive dust in RDF Plants
are fire hazard and airborne microorganisms.
Fire Hazard. The fire hazard due to the accumulation of dust
around RDF plants is a serious problem.
The dust is readily flammable because it is generated mainly from
flammable paper and plastic. Both the St. Louis and Ames plants have
had several fires. One fire at Ames caused a shutdown of 10 working
days. It was believed to have started in an accumulation of dust and
spillage beneath one of the shredders. One of the important justifi-
cations for the Ames dust collection system was reduction of the fire
hazard.
Microorganisms. A great many microorganisms such as fungi, bacteria,
virus, and protozoa exist in MSW, some of which are beneficial to man and
some of which are harmful, Microorganisms are certain to be associated
with MSW processing plants. Incomplete studies suggests that many are not
free floating in air but are carried along on dust particles generated when
MSW is processed into RDF. Control of this dust should help control
microorganisms that may be present.
Test Programs
At both St. Louis and Ames, test programs were conducted to measure
the amount and type of microorganisms that existed in and around RDF
plants. These tests were made to determine if a health hazard existed
and to provide more information for better plant designs and systems
to control dust emissions.
Dust Measurement. Extensive dust emission measurements were made
at St. Louis to describe the amount and size of the dust at various
locations around the plant. An interesting discovery was that the
fugitive dust has a higher energy content (heating value) than RDF
itself and, therefore, has an economic value which is lost if the
dust is allowed to escape.
Microorganisms. Extensive measurements were conducted for bacteria.
Although standard dust measurement methods are well known and have been
thoroughly published in reference documents, the measurement of airborne
bacteria carried by dust in and around waste processing facicilities was
a pioneering effort by the Environmental Protection Agency. Very little
previous work has been done in this area.
-------
Dust Collection Systems
Others have benefitted from the St. Louis and Ames experiences,
resulting in improved designs of RDF plants and the use of very efficient
fabric filtration to eliminate dust at its source. Plants where dust
collection was incorporated in the original plant design include Milwaukee,
Wisconsin; Chicago, Illinois; and Rochester, New York. A discussion of
dust collection by filtration along with design considerations aad cost
information are presented in a later section.
DUST CONCENTRATIONS
Solid Waste Processing Plants Tested
To date, dust measurements have been made at the following four
plants: (1) St. Louis, Missouri—RDF demonstration plant; (2) Houston,
Texas—Shredder plant using an experimental air classifiers; (3) Appleton,
Wisconsin—Shredder plant; (4) Cockeysville, Maryland—RDF research plant.
The St. Louis plant has been described in the previous section. The
Appleton plant is similar to St. Louis except no air classifier is used.
Only shredding and magnetic separation are employed, "and the shredded MSW,
less the recovered ferrous metal, is landfilled. The Houston plant
originally was identical in design to the Appleton plant (i.e., only
shredder and magnetic separator); however, for test purposed a new style
of air classifier was installed. The Cockeysville plant produces RDF,
utilizing shredding, air classification, screening, and magnetic separation.
This plant is used as a research facility by a private firm.
Unfortunately, dust emissions measurements had not been made at the
Ames plant, at the time these data were prepared.
Test Procedures
The amount of dust present in the air was measured by passing a known
amount of air through a filter which collects the dust. The clean filter
was carefully weighed before the test. After the test, this same filter,
now dirty with dust, was again weighed. The difference between the two
weights is the amount of the dust collected.
Two types of samplers were used. The first is termed a "Hi-Vol"
sampler. Using this device, a fan pulls a high volume of air through
a filter. This sampler is used to make measurements of the ambient air
inside the plant such as at worker locations.
When air is exhausted through a duct (as in a dust control system), a
second type of smapler was used which is called a "Method Five" sampler,
-------
u
J
o
1 Further
ro
V)
V)
0)
O
e
Q-
0)
>
O
V
C£
O
v>
U
•n
2
o_
i
X.
CO
v>
0
^
U
1/1
E
D
'§
3
<
JJ
UJ
S!
o
O
O
If
CQ
-------
named after the EPA reference method. This sampler also pulls air through
a filter, using a vacuum pump, but a probe, which is a steel tube, is used
to reach inside the air duct to draw a sample. The dust collected, or
dust catch, is the dust deposited on the filter and any dust that may
have settled out in the sampling probe. This dust is carefully removed
from the probe and weighed along with the filter deposit.
In both methods, the dust weight is known, and the volume air sampled
is known. This will yield, through calculations, the weight of dust per
'unit volume of air. In the metric system, the units are milligrams per
cubic meter.
Dust Concentration Test Results
Dust measurements were made at the four different MSW processing
plants (Table 1). The values are averages from several measurements
made at each location and are expressed in metric units. The air flow
rate from the St. Louis and Houston air classifiers and the St. Louis
dust collector cyclone on the shredder were also measured (Table 2).
The remaining locations were at various places within the plants; there-
fore, there is no airflow rate associated with these locations.
The dust concentration in the air exhaust from the Houston Air
classifier settling chamber was much higher than the St. Louis air
classifier cyclone exhaust because a settling chamber is much less
efficient than a cyclone. However, the air from this settling chamber
was not exhausted directly to atmosphere.
The air was first routed to a fabric filter which removed virtually
all the remaining dust before the air was vented to the atmosphere. This
fabric filter is discussed in a later section. The exhaust from the St.
Louis air classifier was cleaned by a cyclone collector which has a
higher dust removal efficiency than a settling chamber; however, the air
from this cyclone was exhausted directly to atmosphere without additional
cleaning by a fabric filter.
The dustiest in-plant location at St. Louis was at the top of the
RDF storage bin. This bin was enclsoed, with a walkway at the top where
a belt conveyor discharged the RDF into the bin.
The in-plant locations at Appleton and Cockeysville had dust
concentrations within the same order of magnitude, ranging in round
numbers, from 1 to 6 mg/cu m. It is very interesting to observe that
of these in-plant locations, the tipping floor at Appleton was the
dustiest and the tipping floor at Cockeysville had the lowest dust
concentration. In both plants, the trucks backed up to a holding pit
into which the MSW was discharged from the collection trucks. However,
there were differences in truck traffic patterns in and out of the
tipping floor and differences in the size and layout of the pits. Also,
the tipping floor at Appleton was relatively open to the rest of the
8
-------
Table 1
DUST MEASUREMENT RESULTS
Plant
Location
Average dust
concentration!
mg/cu m
St. Louis
Houston
Apple ton
Cockeysville
Exhaust from air classifier cyclone 571
Exhaust from shredder dust collection 738
system cyclone
Inside walkway at top of RDF storage bin 17.5
Exhaust from air classifier settling 23,000
chamber (prior to a fabric filter)
In-plant sites:
Tipping floor 5.6
Next to the shredders 3.5
Next to the magnetic separator 1.4
Next to shredded MSW belt discharge 3.4
In-plant sites:
Tipping floor 1.1
Next to the shredder 2.9
Next to the magnetic shredder 1.7
Insider process building which houses an 3.1
air classifier and a large screen separator
Table 2
AIR FLOW MEASUREMENT RESULTS
St
Plant
. Louis
Location
Exhaust from air
classifier
cu m/sec
13.3
Airflow
(cu ft/mln)
(28,000)
cyclone
St. Louis Exhaust from shredder dust 0.5
collection system cyclone
Houston Exhaust room air classifier 21.8
settling chamber
(1,100)
(46,000)
-------
plant area, while at Cockeysville the tipping floor was separated by a
wall from the rest of the plant area.
There is closer agreement between plants for dust concentrations
around the shredder and the magnetic separator. The dust concentrations
shown in Table 1 are averages of from three to eight measurements per
location. Individual measurements vary from the averages, showing that
dust concentrations are not consistent at any given location. It is
expected that dust emissions vary with such factors as the moisture
content of the MSW, the composition of the MSW, and the plant processing
rate. The most important conclusion is that significant dust concentrations
exist at the MSW processing plants tested.
BACTERIA CONCENTRATIONS
Measurements were made to determine the concentration of bacteria
per unit volume of air. For dust alone, various standards have been set
so that judgements can be made concerning the potential harmfulness of
various dust concentrations. However, actual numbers of bacteria are
relatively meaningless in determining whether or not a potential health
hazard exists because no standards have been established for airborne
bacteria. Since bacteria will most likely be present whenever solid
waste is handled, other waste handling plants were tested. This at
least allows comparisons to be made between the bacteria concentrations
at an RDF plant and the concentrations at other similar plants.
Solid Waste Processing Plants Tested
The facilities tested in this research program were: (1) RDF plant—
St. Louis; (2) municipal incinerator; (3) solid wste transfer station; (4)
sanitary landfill; (5) muniicpal wastewater (sewage) treatment plant.
Following is a brief description of each of the five plants tested.
Refuse-Derived-Fuel. The St. Louis RDF plant has been described in
the previous section. Airborne bacteria measurements were made at the
following locations: (1) upwind and downwind from the plant along the
property boundaries; (2) tipping floor where the refuse collection trucks
unload; (3) operator control room; (4) packer station where large trucks
are loaded with RDF; (5) air classifier exhaust.
Incinerator. At the incinerator, incoming refuse trucks are weighed
on a platform scale adjacent to a dump pit. After weighing, the trucks
discharge MSW into the dump pit at one side of a tipping floor. The MSW
is picked up from this pit by an overhead crane and deposited in charging
hoppers, which in turn feed several combustion chambers where the MSW is
burned. Measurements were made at the following locations: (1) upwind
10
-------
and downwind from the plant along the property boundaries; (2) scale
room; (3) tipping floor; (4) overhead crane.
Waste Transfer Station. The waste transfer station has a large
tipping floor very similar to the RDF plant where the refuse collection
trucks discharge their loads. The MSW is then pushed hy a front-end
loader into a hopper feeding a packer. At the packer, a large
hydraulically operated ram pushes the MSW into a large truck. Three
packer stations are located at the end of a truck ramp. These particular
packer stations are almost identical in operation to the one used at the
RDF plant. The large truckloads of MSW are then hauled several miles to
a sanitary landfill. Measurements were made at the following locations:
(1) upwind and downwind from the plant along the property boundaries;
(2) tipping floor, north side; (3) tipping floor, east side; (4) truck
ramp, packer station location.
Sanitary Landfill. The tested landfill is a traditional operation
where trucks enter the area, are weighed on a platform scale, and then
proceed to a designated area (working face) where they discharge their
loads of MSW. Measurements were made at the following locations: (1)
upwind and downwind from the' landfill working face along the property
boundaries; (2) scale; (3) working face, west side; (4) working face,
east side.
Wastewater Treatment Plant. At the wastewater treatment plant sewage
is first received in three primary settling basins. The seage next goes
to an aeration basin where air is injected into the sewage. From there
the sewage flows in sequence to two secondary settling basins, a sludge
thickener, holding tanks, and lastly to a large press where remaining
liquid is squeezed from the solids. The solids removed by the press are
burned in an incinerator. Measurements were made at the following
locations: (1) upwind and downwind from the plant along the property
boundaries; (2) primary settling basin; (3) aeration basin; (4) press
room, operators area; (5) press room basement.
Amount of Waste Handled at Each Point. Waste handling facilities
vary greatly in size. The five plants were selected to be as close as
possible in size to provide better comparisons. The total amount of
waste material received at each plant was recorded on each day of testing.
The average amounts are shown in Table 3.
For the plants handling MSW, the weights shown are the total received
for the day since the plants receive refuse during only one shift per day
which corresponds to the sampling period. However, the sewage treatment
plant received sewage 24 hr/day. The average liquid amount of 2,067,000
liters is received during the 6 to 7 hr period when samples were taken.
This liquid volume converts to approximately 2,000 megagrams (2,200 tons)
on a weight basis.
11
-------
Table 3
AMOUNT OF WASTE HANDLED AT TEST PLANTS
Plant
Averaged daily amount of waste material
received during the test period
Megagrams (Tons) Litters (Gallons)
RDF plant
Incinerator
Waste transfer system
Sanitary landfill
Wastewater treatment plant
163
340
311
788
(180)
(375)
(343)
(869)
2,067,000 (546,000)
Other Bacteria Measurement Locations. For the purpose of making
comparisons, measurements were made at two other special locations. The
first special location was the back of a garbage truck. Two samplers
were attached to the back of the truck, one on each side, and were powered
by a generator attached beneath the truck. The purpose of this sampling
location was to determine the bacteria concentrations in the area where
workers are collecting MSW from city residences and placing it in the
refuse collection truck.
The second location was in the city of St. Louis where a sampler
was set up on the corner of a major downtown intersection. The purpose
of this location was to determine background concentrations of bacteria,
where heavy pedestrian and vehicle traffic was present, in the same city
as the RDF plant. This was an important sampling point for comparative
purposes. Airborne bacteria concentrations are not normally zero in
nature; however, they would be expected to be relatively low in rural
and suburban areas. Therefore, a realistic background location for
comparative purposes is a busy street corner in a metropolitan city like
the one in downtown St. Louis.
Microorganisms Measured
An objective of the research program was to Look for harmful species
that might exist in MSW. A detailed work plan was prepared and submitted
to a panel of 10 experts in this field. As a result of this work plan and
12
-------
the expert review comments, nine criteria were selected: (1) total
aerobic bacteria; (2) salmonella bacteria; (3) staphylococcus areus
bacteria; (4) total coliform bacteria; (5) fecal coliform; (6) feca;
streptococcus bacteri; (7) klebsiella bacteria and viruses; (8) adeno-
viruses and (9) enteroviruses.
Total aerobic bacteria, which is bacteria that can exist in, air,
was measured to determine the total amount of airborne bacteria that
was present. Likewise, virus was measured to determine the amount of
virus that was present. Salmonella, staphylococcus, streptococcus, and
klebsiella were measured because these are pathogens (disease-causing
bacteria species). Coliforms are bacteria that exist in the intestinal
tract of man and animals, and forms which resemble or are related'to
them. Although not generally pathogens themselves, they indicate that
undesirable fecal material may be present. "Total coliform" refers to a
group or family of bacteria while "fecal coliform" is a species occurring
in several varieties within this groups.
Test Procedures
The field sampling procedures were much the same as for dust. Hi-Vol
samplers, as described in an earlier section, were used to draw air through
a filter which collected the dust. At the end of each test day, the filters
were removed from the samplers, sealed and packed in an ice chest, and
shipped to the laboratory. Analysis of the filters began the next day.
The laboratory procedures used were relatively complex. Briefly, the
dust-laden filters were processed into a slurry (or "paste") and the slurry
was placed onto a nutrient culture media. The samples were then incubated.
The number of bacteria that had grown were counted, differentiating between
different species. A detailed technical discussion of the methods used is
contained in the report "Assessment of Bacteria and Virus Emissions at a
Refuse Derived Fuel Plant and Other Waste Handling Facilities."*
The volume of air sampled through each filter was measured, and the
bacteria concentration per unit volume of air was calculated. In the
metric system, results are expressed in bacteria counts per cubic meter
of air, and in the English system, results are in units of bacteria counts
per cubic foot of air.
* Fiscus, D. E., P. G. Gorman, M. P. Schrag, and L. J. Shannon.
Assessment of Bacteria and Virus Emissions at a Refuse Derived Fuel Plant
and Other Waste Handling Facilities. U.S. Environmental Protection Agency
Report EPA-600/2-78-152. Municipal Environmental Research Laboratory,
Office of Research and Development, U.S. Environmental Protection Agency,
Cincinnati, Ohio, August 1978. 240 pp.
13
-------
Microorganism Test Results
No salmonella, staphylococcus, or virus was found In any of the
samples collected. Klebsiella was found in only four samples and at
very low levels. Therefore, the comparison between waste processing
plants was made for the following four bacteria types: (1) total bacte-
ria count; (2) total coliform; (3) fecal coliform; (4) fecal streptococcus.
Bacteria concentrations varied at each measurement location, which
is not surprising since it is known that dust concentrations also vary
from day to day at MSW processing facilities. Two facts must be kept In
mind concerning the measured bacteria concentrations. Since bacteria
are living organisms, and the laboratory analysis of field samples did
not begin until the day after the samples were collected, bacteria could
have multiplied, or more likely died off, even when every precaution was
taken. Therefore, the bacteria concentrations in counts per cubic meter
of air should not be considered as absolute values. Nonetheless, all
plants were measured in the same way, and the data are useful in making
comparisons between plants.
The following discussion includes the ranges of average bacteria
concentrations found and a comparison or ranking of the various plants*
Because a great number of measurements were made, statements concerning
the comparison of one plant to another will be more useful and more easily
understood than voluminous and complicated tables and graphs of individual
measurement numbers.
Range of Bacteria Concentrations. The range of average concentrations
by category is shown in Table 4.
These values are the ranges of the average results. Of course, some
individual measurements were above and below these average values. Fecal
streptococcus concentrations were higher than were the coliform concen-
trations.
These values are the ranges of the average results. Of course,
some individual measurements were above and below these average values.
Fecal streptococcus concentrations were higher than were the coliform
concentrations.
The downtown bacteria concentrations fell within the range of
concentrations upwind from the plants, and as expected, the downwind
concentrations tended to be higher than the upwind concentrations. An
important result for comparative purposes is that the refuse collection
truck samples had higher average concentrations of total collection truck
samples had higher average concentrations of total coliform and fecal
coliform than any of the waste processing plants.
14
-------
Table 4
RANGE OF BACTERIA CONCENTRATIONS
Average bacteria concentrations (counts/cu m)
Total
bacteria Total Fecal Fecal
Location in category count coliform coliform streptococcus
Upwind at property line
High 8,000 1 0.2 15
Low 500 0.09 0.02 1
Dowtown
High 2,200 0.5 0.15 1.5
Low 950 0.25 0.03 1
Downwind at property line
High 20,000 60 2 55
Low 500 0.12 0.02 1
In-plant
High 400,000 100 11 2,170
Low 220 0.03 0.05 1
Packer truck (refuse
collection truck)
High 90,000 300 170 460
Low 55,000 210 110 410
In summary, while some small values of bacteria counts per cubic
meter found, there were also some large values measured. All
locations had measurable amounts of total bacteria, total coliform, fecal
coliform, and fecal streptococcus. The next step then is to determine
how the individual plants compared to one another.
Waste Processing Plant Comparisons. Table 5 gives a ranking of the
various plants, including downtown and the packer truck, from highest to
lowest average bacteria count.
Some trends become apparent from this ranking. The RDF plant had
higher downwind concentrations than any of the other plants, but it also
had higher upwind concentrations. Therefore, it becomes difficult to
claim that the RDF plant caused a greater increase in downwind bacteria
levels than other types of waste handling facilities. Surprisingly, the
15
-------
downtown location did not have the lowest bacteria concentrations when
compared to plant downwind locations. Depending on the bacteria measured,
either the wastewater treatment plant, the waste transfer station, the
sanitary landfill, or combinations of these three plants had lower
concentrations.
Table 5
RANKING BASED ON AVERAGE BACTERIAL LEVELS IN DESCENDING ORDER
FROM HIGHEST TO LOWEST CONCENTRATION
Total bacteria
count
Total
coliform
Fecal
coliform
Fecal
streptococcus
RDF plant
Incinerator
Downtown
Waste transfer
WWTP*
Landfill
Ambient samples—upwind (and downwind)
RDF plant
Downtown
Incinerator
WWTP
Waste transfer
Landfill
RDF plant
Downtown
Waste transfer
Incinerator
WWTP
Landfill
RDF plant
Incinerator
Waste transfer
Downtown
WWTP
Landfill
Ambient samples—downwind (and downtown)
RDF
Incinerator
Downtown
WWTP
Waste transfer
Landfill
RDF plant
Waste transfer
Incinerator
Landfill
Downtown
WWTP
RDF plant
Waste transfer
Incinerator
WWTP
Downtown
Landfill
RDF plant
Incinerator
Waste transfer
Downtown
WWTP
Landfill
In-plant samples
RDF plant
Packer truck
Incinerator
Waste transfer
WWTP
Landfill
Packer truck
RDF plant
Waste transfer
Incinerator
Landfill
WWTP
Packer truck
RDF plant
Waste transfer
Incinerator
Landfill
WWTP
Waste transfer
Packer truck
RDF plant
Incinerator
Landfill
WWTP
* WWTP = Wastewater treatment plant.
16
-------
For the in-plant locations, the refuse collection packer truck had
higher total coliform and fecal coliform concentrations than the RDF plant*
Comparing the RDF plant to the other waste handling facilities, the RDF
plant had higher in-plant bacteria levels for total bacteria count, total
coliform, and fecal coliform than the other plants. However, the waste
transfer station had higher fecal streptococcus levels than the RDF plant*
The incinerator wastewater treatment plant and the landfill had lower in-
plant concentrations than the RDF plant for all four bacteria types measured*
The conclusions from these test results is that the RDF plant tended to
have higher in-plant bacteria concentrations than many, but not all, of the
other facilities tested. However, the RDF plant tested was a plant with only
a rudimentary dust collection system. If a well-designed dust collection
system were used, then bacteria levels in an RDF plant would be expected to
be comparable to, or less than, levels in other waste handling facilities.
Fortunately, the justification for dust collection is not based on bacteria
control alone. In addition to reduction of bacteria levels, there is also a
need for dust collection systems to reduce the dust levels in RDF plants.
Regardless of any bacteria concentratins, dust removal would be very
beneficial to RDF plants because there are many other problems associated
with dust such as fire hazards, plant maintenance and cleanup, etc., which
would also be ameliorated.
DUST COLLECTION SYSTEMS
Dust collection systems for RDF plants were relatively simple and
were composed of only three major parts as follows: (1) the dust capture
hoods and connecting duct work; (2) the air-moving fan and motor; (3) a
fabric filter, commonly called a baghouse.
Suction is provided by the fan to pull into the capture hoods. The
baghouse then filters virtually all the captured dust from the airstream
which is exhausted to atmosphere. In a very rough sense, a dust collection
system is a large vaccum cleaner.
Baghouses, fans, duct work, and capture hoods come in a variety of
configurations. The most important consideration is correct design and
placement of the capture hoods because a very efficient baghouse will do
an RDF plant no good if the dust is not first captured. For persons
Interested in more technical details, an excellent discussion of dust
collection is presented in the book, Industrial Ventilation.* Competent
engineering services should always be used in the design of dust collection
systes. "Do-it-yourself" dust collection systems have seldom proved adequate.
* Industrial Ventilation, American Conference of Governmental
Industrial Hygienlsts, P.O. Box 453, Lansing, Michgan 48902 (manual
updated periodically; request latest edition if ordering).
17
-------
Collection Efficiency
Baghouses are very efficient for the type of dust found in RDF
plants. EPA has tested two baghouses. One was a small-scale baghouse
at the St. Louis RDF plant where a portion of the air classifier exhaust
air was drawn from the exhaust duct and passed through the baghouse.
The second was a full-scale baghouse, cleaning the total exhaust from
the air classifier settling chamber used at the Houston shredder plant.
At both locations, measurements of dust and bacteria concentrations were
made before and after the baghouses so that their collection (removal)
efficiency could be determined.
Dust Collection Efficiency. Table 6 gives the dust collection
efficiency for the two baghouses.
Table 6
DUST COLLECTION EFFICIENCIES
Average dust Average dust Collection
concentration before concentration after efficiency
Plant the baghouse (mg/cu m) the baghouse (mg/cu m) (%)
St. Louis 300 0.154 99.95
Houston 23,000 2.76 99.99
These two tests are quite interesting to compare because the
concentration of dust in the inlet to the baghouse was quite
different at the two locations. At St. Louis, the air to the baghouse
was the exhaust air from the air classifier cyclone separator. As
discussed earlier, this cyclone separator removed much of the dust from
the air classifier. The Houston facility merely had a settling chamber
installed immediately prior to the baghouse which allowed large pieces
of RDF to drop out of the air -stream but still allowed the fine material
to pass on to the baghouse. Therefore, the Houston baghouse had a much
heavier dust concentration in the incoming air.
The important point is that even though the dust concentrations were
very different, the removal efficiency of the baghouse was almost the same,
being better than 99.9 percent efficient. The high efficiency of baghouses
on dust from RDF plants was expected, based on a study of various published
reports describing baghouses used to collect different types of dusts.
18
-------
Bacteria Collection Efficiency. The baghouses were also very
efficient in collecting bacteria, as would be expected, because it is
believed that bacteria are carried by dust particles. Therefore, the
theory is that if the dust is collected, the bacteria are also collected.
However, even if bacteria are not associated with dust particles, a
baghouse should still removed them very well since baghouses yield high
removal efficiencies on bacteria-size particles, following is the bacteria
collection efficiency efficiency for the St. Louis and Houston baghouses*
Efficiencies are given in Table 7 for total bacteria and for the individual
species.
The bacteria collection efficiency for the Houston baghouse was
slightly less than the dust collection efficiency. However, the bacteria
efficiency is still very high, being better than 98 percent.
Table 7
BACTERIA COLLECTION EFFICIENCIES
Average baghouse collection efficiency
for bacteria (%)
St
Plant
. Louis
Houston
Total
bacteria
count
99.60
98.87
Total
coliform
99.99
98.65
Fecal
coliform
99.95
98.39
Fecal
streptococcus
99.91
98.20
Design Considerations
Dust can be controlled in an RDF plant in a variety of ways. The
traditional dust collection system with capture hoods, suction air, and
fabric filtration is used to collect dust as it is being emitted. How-
ever, it is often preferable to keep the dust from being emitted in the
first place.
Belt conveyors are commonly used in RDF plants. These can be partially
covered to reduce dust emissions. The most important locations to control
are the loading and discharge points of the conveyor, because at these
points the MSW, RDF, etc., are agitated most, creating the dustiest conditions.
However, the plant designer should also consider other types, such as screw
conveyors and drag conveyors which can be totally enclosed. If the covers
and their loading and discharge spouts are tight-fitting and sealed, then
screw or drag conveyors should have very low dust emissions.
19
-------
Another area requiring dust control is storage bins. When material
falls into a storage bin, the material replaces the air in the bin. This
escaping air can carry small dust particles with it. Storage or surge
bins, especially for shredded MSW or RDF, should be covered and the bin
vents should be conveniently located for the dust control duct work.
A technique which has been used effectively in other industrial plants,
and which may have application to RDF plants, is to enclose dusty areas.
This technique helps to keep dust confined to one or a few areas and
prevents contamination of the rest of the plant.
Building ventilation is a very important consideration for the design
engineer. Most RDF plants will have a heating and air conditioning system
providing air for plant offices, operator control rooms, and employee
locker rooms and lunchrooms. Furthermore, in plants where air classifiers
and dust collection systems are used, air will be drawn from inside the
plant and exhausted outside the plant after passing through a baghouse.
This exhausted air must be made up by admitting air into the plant through
fresh air intakes. The fresh air intake for heating and air conditioning,
ventilation, and makeup plant air should be located as far as possible
from any dusty air exhausts. The cleaner the air entering the plant,
the easier will be the task of keeping the plant clean.
Recirculation of dirty air within the plant should also be avoided.
At the Ames facility, the exhaust air from the air classifier was
recirculated into the plant. This was a good idea for eliminating emissions
to the outside atmosphere. However, even though this exhaust is introduced
into the plant near the air classifier air intake, a certain amount of dust
escapes and contaminates the surrounding plant area. At Ames, a baghouse
was installed on the air classifier air exhaust.
Worker protection is another aspect of dust management. Obviously
workers should have clean and well-kept locker room and lunchroom areas.
However, personal protectin devices such as dust masks, etc., should be
available if the plant has especially dusty areas.
For any individual plant, it probably will not be possible to control
all dust emissions without the use of a dust collection system, which will
be discussed next.
Dust Collection and Filtration. Not all plant locations need dust
collection, because dust is not generated at all locations. Use of
enclosed conveyors and enclosed storage bins, for example, can sharply
reduce the need for dust collection. The philosophy of dust collection
is that it be employed wherever dust is being generated. Solid waste lying
at rest does not produce dust. Only when it is moved about is dust
produced. Visual observations in RDF plants indicate that shredded MSW
and RDF are the greatest dust contributors. The various recovered metal
by-products and reject material, such as air classified heavies, are
generally not large contributors to the plant dust burden.
20
-------
Following is a list of plant locations that should be carefully
considered for application of dust collection: (1) shredder inlet and
outlet; (2) any open conveyor transfer point; (3) unenclosed screening
devices used to screen or clean MSW or RDF; (4) storage and surge bin
vents; (5) air exhaust from an air classifier; (6) air exhaust from a
pneumatic conveying system.
Because it has high energy value, the dust collected by a dust
control system should be combined with the RDF.
Also, because dust in RDF plants is flammable, fire detection and
protection systems are important considerations for the designer and
the plant management. Containers of gasoline, cartons of aerosol cans,
etc., may be found in MSW and have, in the past, resulted in shredder
explosions. Therefore, explosion venting and explosion suppression may be
a necessary part of plant design.
Cost of a Dust Collection and Filtration System. Costs vary with
the size of an RDF plant. The 1978 cost a dust collection system at Ames,
Iowa, a nominal 45 megagrams per hour (50 ton/hr) RDF plant, was $166,000.
This is the total installed price including engineering, materials, fab-
rication, and installation.
The dust collection system includes a baghouse on the air exhaust
from the air classifier cyclone and dust collection hoods within the
plant area.
Various engineering estimating methods are available to arrive at
dust collection cost estimates. These methods range from "rule of thumb"
methods to very detailed and specific cost quotations. Of course, the
more information that is available concerning any plant, the more accurate
will be the cost estimate.
Fortunately, dust collection costs are available for two RFD plants:
Ames, Iowa, and Monroe County, New York. Both dust collection systems
were installed in 1978, so cost comparisons may be made without adjusting
for the value of the dollar in different years. Costs for material, labor
and engineering is: (1) Ames—$166,000; (2) Monroe County—$992,000.
These costs do not include the cost of financing and start-up. The
engineering portion of the cost for Monroe County is an estimate.
Since these two plants are very different in size, the question is
how can the dust collection costs for these plants be compared and how can
they be used to predict costs for other plants? Dust emissions probably
depend on the processing rate because it is reasonable to assume that the
more MSW that is processed per hour, the greater is the potential for dust
emissions.
21
-------
The Ames facility ws built and operated for approximately two years
before the decision was made to install a dust collection system. There-
fore, the dust collection system was designed to control the actually
observed emissions resulting from the actual processing rate, not the
original design processing rate. The actual processing rate for the
Ames plant, calculated from plant data, is 23 megagrams per hour (25
tons/hr). This value is calculated from the actual amount of MSW processed
over several months divided by the total actual hours the plant conducted
processing (not including cleanup and maintenance hours after processing
had ceased). The Monroe County plant is not yet in operation, but it is
planned to operate at an actual 63.5 megagrams per hour (70 tons/hr) for
each of two processing lines. Therefore, the total Monroe County plant
processing rate is 127 megagrams per hour (140 tons/hr). These data
allow rule-of-thumb calculations to be made in Table 8.
Table 8
DUST COLLECTION COSTS
Dust
collection
Plant cost
Ames $166,000
Monroe County $992,000
Processing rate
Mg/hr (ton/hr)
23 (25)
127 (140)
Cost per processing
rate
$/Mg/hr
7,217
7,811
($/ton/hr)
(6,640)
(7,085)
These rule of thumb estimates are in relatively close agreement. The
Monroe County cost in terms of dollars per processing rate is approximately
eight percent more than for Ames. This eight percent difference is most
likely due in part to differences in plant design and layout, and labor rate
differences between Iowa and New York.
In summary, a rough order of magnitude (ROM) estimate in round figures
is $8,000 per megagram/hour, yielding $360,000 for a 45 megagram per hour
(50 tons/hr) plant. This estimated cost is only for materials, labor, and
engineering.
In addition to labor, materials, and engineering, there are many other
items in the overall cost: financing, maintenance equipment and spare parts
and supplies, contingencies, and start-up expense. Most of these can be
amortized over the life of the equipment. The yearly operating cost would
then include the amortized cost, utilities, insurance, management, and
administrative labor, maintenance labor, and maintenance supplies. Table 9
illustrates a simplified cost model used to develop yearly costs of a dust
collection system.
22
-------
The financial model of any individual plant will, of course, vary from
the simple model shown here. In summary, cost of a dust collectin system
for an RDF plant would range from $0.50 to $0.75 per megagram ($0.45 to
$0.67/ton), based on 1958 data. No attempt has been made here to estimate
the cost savings resulting from: (1) decreased plant cleanup labor; (2)
decreased plant maintenance; (3) cleaner environment inside and outside the
plant; (4) reduced plant insurance costs due to reduced fire hazard; or (5)
reduced plant downtime due to fires.
Table 9
DUST CONTROL COST MODEL
1978 ($)
Capital investment
Equipment, installation, and construction management
Design engineering about 10%
Subtotal
330,000
30,000
360,000
Financial charges at 25% (interest during construction, 90,000
bond underwriting, legal costs, etc.)
Spare parts and maintenance equipment about 5% 18,000
Start-up expense about 10% (includes acceptance testing) 36,000
Contingencies about 10% 36,000
Total capital investment 540,000
Yearly cost
Depreciation assuming 20-year straight line write-off
Operation and maintenance
Direct labor about 1/3 man-year/year*
Supervision about 1/10 man-year/yeart
Utilities about 1.3% of capital investmant each year
Maintenance supplies about 0.7% of original captial
investment each year
Overhead absorption about 0.35% of original captial
investment each year
Total
27,000
5,000
2,000
7,000
4,000
2,000
47,000
23
-------
Yearly cost per tnegagram of MSW processed
Assume 240 days/year (260 weekdays less 20 days for About 0.6/Mg
holidays and plant downtime) at 7 hr/day; 45 mg/hr
yields 75,600 megagrams (83,400 tons)/year (About 0.55/ton)
* A man-year of direct labor is assumed to cost $15,000.
* A man-year of supervision is assumed to cost $20,000.
The financial model of any individual plant will, of course, vary
from the simple model shown here. In summary, cost of a dust collection
systems for an RDF plant would range from $0.50 to $0.75 per megagram
($0.45 to $0.67/ton), based on 1978 data. No attempt has been made
here to estimate the cost savings resulting from: (1) decreasd
plant cleanup labor; (2) decreased plant maintenance: (3) cleaner
environment inside and outside the plant; (4) reduced plant insurance
costs due to reduced fire hazard; or (5) reduced plant downtime due
to fires.
24
-------
SUMMARY
Processing plants that produce refuse derived fuel (RDF) by
shredding the combustible materials in municipal solid wastes are
in operation in several locations. Dust is a problem whenever the
wastes are agitated, including (a) the tipping floor where trucks are
unloaded, (b) open areas of conveyors, (c) shredders, and (d) cyclone
separators.
Main corrective measures are: (1) containing the dust, by using
covers on conveyors, and by using walls to separate the dustiest areas
from cleaner areas of the plant; (2) providing dust masks where workmen
must enter any dusty area; (3) preventing exhaust of dirty air into
outdoor atmosphere; (4) designing and installing bag filters to capture
the dust before it escapes from the processing equipment.
Efficiency of the bag filter systems was over 99.5 percent in
capturing dust, and over 98 percent in capturing bacteria.
Bacteria count in processing plants for refuse-derived fuel ranged
from 220 to 400,000 counts per cubic meter (cu m), compared to outdoor
atmosphere range of 950 to 2,200 counts per cubic meter at a downtown
location.
Total coliform bacteria airborne inside the plant ranged from 0.03
to 100 counts/cu m, compared to outdoor downtown range of 0.25 to 0.5
counts/cu m. Fecal coliform bacteria ranged from 0.05 to 11 counts/cu m
inside the plant, and 0.03 to 0.15 outdoors downtown. Fecal streptococcus
ranged from 1 to 2,170 counts/cu m inside the plant, and j. to 1.5
counts/cu m outdoors downtown.
The dust measurements were made at refuse-derived-fuel plants at
St. Louis, Missouri; Houston, Texas; Appleton, Wisconsin; and Cockeysville,
Maryland. Bacteria concentrations were measured at the St. Louis plant and
for comparison, at an incinerator site, solid waste transfer station,
sanitary landfill, and municipal sewage treatment plant.
Typical costs for designing and installing dust control systems in
large and small plants producing refuse derived fuel from muncipal solid
wastes were estimated, based on 1978 data.
pal 838
SW-773
* M. WWMMNMMOfftt
25
-------
EPA REGIONS
U.S. EPA, Region 1
Solid Waste Program
John F. Kennedy Bldg.
Boston, MA 02203
617-223-5775
U.S. EPA, Region 2
Solid Waste Section
26 Federal Plaza
New York, NY 10007
212-264-0503
U.S. EPA, Region 3
Solid Waste Program
6th and Walnut Sts.
Philadelphia, PA 19106
215-597-9377
U.S. EPA, Region 4
Solid Waste Program
345 Courtland St., N.E.
Altanta, GA 30308
404-881-3016
U.S. EPA, Region 5
Solid Waste Program
230 South Dearborn St.
Chicago, IL 60604
312-353-2197
U.S. EPA, Region 6
Solid Waste Section
1201 Elm St.
Dallas, TX 75270
214-767-2734
U.S. EPA, Region 7
Solid Waste Section
1735 Baltimore Ave.
Kansas City, MO 64108
816-374-3307
U.S. EPA, Region 8
Solid Waste Section
1860 Lincoln St.
Denver, CO 80295
303-837-2221
U.S. EPA, Region 9
Solid Waste Program
215 Fremont St.
San Francisco, CA 94105
415-556-4606
U.S. EPA, Region 10
Solid Waste Program
1200 6th Ave.
Seattle, WA 98101
206-442-1260
-------
------- |