United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-83-098 Jan. 1984
f/EPA Project Summary
Evaluation of Deep Shaft
Biological Wastewater
Treatment Process at Ithaca,
New York
Donald E. Schwinn, Donald F. Storrier, and Robert Butterworth
s'
The major objectives of this study
were to demonstrate the feasibility of
the Deep Shaft biological treatment
process* and to evaluate its application
for the treatment of municipal wastewa-
ter. A 757-mVday (0.2-mgd) pilot
plant facility was constructed at the
existing wastewater treatment plant
site in Ithaca, New York, for this
purpose.
The Deep Shaft process was evaluated
under a variety of operating conditions
including raw wastewater and primary
effluent as influent sources, constant
and diurnal (varying) flow patterns.
with and without polymer as a flotation
aid, and with alum added for phosphorus
removal. Because partially ground
screenings and abnormally strong
anaerobic digester supernatant are
returned to the main plant headworks
at Ithaca, pilot plant influent characteris-
tics were not the typical domestic raw
wastewater or primary effluent that
had been anticipated when the site was
selected. Numerous operational prob-
lems also hindered the experimental
program; however, 5 mo of reliable
operating and performance data were
obtained on which to draw conclusions
about the process.
This Project Summary was developed
by EPA's Municipal Environmental
Research Laboratory, Cincinnati, OH,
to announce key findings of the research
'Mention of trade names or commercial
products does not constitute endorsement or
recommendation for use.
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
The Deep Shaft biological treatment
process is a high-rate activated sludge
process capable of operating at food-to-
mass (F/M) loadings between 0.5 and
2.0 kg BOD5/day/kg MLVSS. High
volumetric loadings can be achieved
because the system is capable of carrying
and maintaining MLVSS concentrations
between 5,000 and 10,000 mg/L As a
result, the bioreactor volume (aeration
period) is much lower than that needed in
conventional systems.
Deep shafts are self-contained vertical
subsurface aeration reactors normally
between 90 and 250 m (300 and 800 ft)
deep with mean hydraulic retention
times (HRT) of approximately 40 to 60
min for municipal-strength wastewater.
The HRT generally increases with increas-
ing wastewater strength.
Basically, the reactor is divided into
downcomer and riser sections. Raw
wastewater (after screening and degrit-
ting) and return sludge continuously enter
the downcomer section and flow down-
ward. From here the aerated liquid
passes into the riser section and flows
upward. A portion of the mixed liquor
overflows the shaft to a solids separation
process, and the remainder of the mixed
liquor is recirculated to the downcomer
section. Mixed liquor is circulated many
-------
times within the shaft during its residence
in the reactor. Compressed air is injected
into the Deep Shaft to provide the oxygen
needed for treatment and to serve as the
driving mechanism that maintains circu-
lation velocities.
After the design of Eco I, improvements
in its operation and geometry led to the
development of Eco II. Rather than the
one downcomer and one riser in Eco I,
Eco II uses a multi-channel concept
featuring one primary and one secondary
downcomer and one primary and three
secondary risers to minimize anoxic
zones in the shaft and to maximize the
driving force for inducing flotation
clarification. The mixed liquor withdrawal
point was also relocated from the head
tank to the bottom of the largest secondary
riser to increase the dissolved gas
content of the mixed liquor being trans-
ferred to the flotation tank and to provide
maximum treatment time per pass
following initial contact of incoming
substrate, biomass, and injected air.
Based on the experience gained with
Eco I and II, a third generation system,
Eco III, was developed. The Eco II Deep
Shaft reactor configuration optimized the
biological profile inside the reactor and
stabilized the hydraulic flow pattern; the
air supply requirements, however, main-
tained liquid circulation at peak flow
conditions. With the Eco III design, the air
flow rate is controlled to match the in-
fluent wastewater flow; the swirl tank
and an extensive amount of process con-
trol instrumentation are eliminated; and
the flotation unit is modified to strip and
release coarse air bubbles from the mixed
liquor feed stream and to promote better
flocculation of mixed liquor solids.
Description of Ithaca Pilot Plant
Facilities
The Deep Shaft pilot plant (Figure 1)
evaluated at Ithaca was of the Eco II
design. Degritted raw wastewater or
primary effluent was pumped from the
main treatment plant to a splitter box
located in the pilot plant building. The
flow rate to the shaft was controlled by a
pneumatically operated valve that could
be adjusted to provide a constant influent
flow rate or be varied automatically in a
diurnal flow pattern.
After screening and flow measurement,
the wastewater flowed into the trough at
the influent end of the flotation tank
where it was mixed with floating solids
skimmed off the top of the flotation tank.
Combined influent flow and return float
solids then entered a holding tank
adjacent to the shaft. Solids that sank to
the bottom of the flotation tank were
returned to the influent flow stream.
From the bottom of the holding tank,
influent wastewater and return solids
were piped into the secondary downcomer
section of the shaft. The secondary
downpomer ends slightly above mid-
depth of the shaft where it forms two U-
shaped sections and turns up into two
short secondary risers. Influent wastewa-
ter and return solids entering the
secondary downcomer are consequently
rapidly aspirated into the primary riser
through these two short secondary
risers.
The steel shaft casing had an inside
diameter of 44 cm (17.25 in.) and a depth
of 136 m (446 ft). The inner concentric
primary downcomer had an outside
diameter of 20 cm (8 in.) for slightly less
than the top half of the shaft, an outside
diameter of 30 cm (11.75 in.) for the
remainder of the shaft, and a depth of
133 m (436 ft). The annulus formed
between the outer casing and the inner
primary downcomer constituted the
primary riser.
At the average design flow of 757
mVday (200,000 gpd), the hydraulic
retention time in the shaft was approxi-
mately 39 min. Air was supplied to the
shaft from a 15-kW (20-hp) air compres-
sor through three lines, one injecting air
into the primary downcomer at a depth of
55 m (180 ft) and two injecting air into the
two short secondary risers at a depth of
60 m (196 ft).
The head tank on top of the shaft was
designed for an operating pressure of
2,800 kgf/m2 (4 psig). Off gases plus
foam generated within the reactor
overflowed the head tank and were piped
into the adjacent oxidation tank. The
connecting pipe was submerged approxi-
mately 3 m (10 ft) below the water
surface in the oxidation tank. This liquid
head provided the back pressure in the
head tank to force the aerated mixed
liquor through the secondary riser
withdrawal pipe and out of the shaft.
Mixed liquor was discharged from the
shaft through the largest secondary riser,
bypassing the head tank, directly into a
swirl tank. Here sufficient detention time
was provided (10 to 15 sec at average
flow) to strip any large air bubbles that
would disrupt flotation. Aerated mixed
liquor was piped from the swirl tank into
the flotation tank. The entrance to the
largest secondary riser was slightly
below the point at which influent
wastewater was aspirated into the
primary riser to avoid short-circuiting
untreated or partially treated wastewater
directly to the flotation tank.
Floating solids (float solids) were
drawn by scrapers towards the front end
of the flotation tank and collected on the
beach and pushed over into the influent
trough. There the solids mixed with the
influent and flowed to the holding tank.
Float solids were wasted through a hole
in the beach. The valve to control wasting
was operated manually or automatically
at preset timed intervals. Solids that had
sunk (sink solids) were collected by bot-
tom scrapers and a screw conveyor and
returned to the influent flow. Sink solids
were wasted by opening a valve on the
discharge pipe of the sink recycle pump.
Clarified liquid overflowed an adjustable
weir at the end of the flotation tank into
an effluent trough. Effluent was piped
from there into the building sump for
return to the main plant.
Operating and Performance
Results
Operational Characteristics
Hydraulic characteristics and opera-
tional data are summarized in Tables 1
and 2, respectively. The average influent
flow rate varied between 636 mVd
(168,000 gpd) and 757 mVd (200,000
gpd). The design flow was 757 mVd
(200,000 gpd). The sustained peaking
factor for the diurnal flow periods ranged
from 1.4 to 1.6. The volumetric organic
loading ranged from 2.2 to 4.9 kg BOD5/
day/m3 (137 to 304 lb/day/1000ft3), the
MLSS level from 5,200 to 9,800 mg/L,
and the F/M loading from 0.51 to 1.0 kg
BODs/day/kg MLVSS. The sludge reten-
tion time (SRT) varied from 1.0 to4.0days
based on shaft and head tank solids only.
Only a limited amount of cold weather
operational data could be obtained, and
most of that was obtained while treating
primary effluent. Influent wastewater
temperatures ranged from 9° to 25°C.
BOD5, COD, and Suspended
Solids (SS) Removals
Because of the numerous mechanical
breakdowns, excessive foaming, partial
shaft blockages, and operational
problems at the main plant that occurred
throughout much of the study, there
were only a few time periods when a
normal, routine operation was estab-
lished. The following results (Table 3)
obtained during such time frames were
considered to be more representative of
process performance and Deep Shaft ca-
pability and, therefore, were focused on
for evaluation purposes. Percent reduc-
tions represent removals across the pilot
system only and do not account for any.
-------
'oam Transfer Line
Foam
Oxidation
Tank
Secondary
Downcomer
sf~^
Head \~)
Tank
Primary
Downcomer^
. Secondary
Primary Riser
Riser
Intermittent Waste
Bottom Sludge
Figure 1. Schematic of deep shaft system at Ithaca. New York.
Table 1. Pilot Plant Hydraulic Characteristics
Influent Flow Rate
Solids Separation Unitf
Operating Mode*
(Dates)
m3/day (gpd)
Peak/Avg.
Bioreactor\ HRT Overflow Rate
HFtT (min) (hr) m3/day/m2 (gpd/f?)
flaw Wastewater
Constant flow - N.P.
06/10/80-07/10/80
Constant flow W.P.
05/19/80-06/09/80
Diurnal flow - N.P.
07/14/80-07/31/80
08/01 / 80-08/31 /80
09/01/80-09/31/80
10/01/80-10/20/80
02/19/81-02/28/81
03/01/81-03/31/81
04/01/81-04/10/81
05/31/81-06/30/81
07/01/81-07/31/81
08/01 /81 -08/28/81
Primary Effluent
Constant flow - N.P.
It/09/80-11/21/80
Diurnal flow - N.P.
12/07/80-12/12/80
01/24/81 -01/30/81
Diurnal flow - W.P.
12/14/80-01/12/81
Diurnal flow - W.A.
01/13/81-01/23/81
681 (180,000)
703 (186,000)
708 (187,000)
708 (187,000)
711 (188,000)
714 (189,000)
753 (199,000)
726 (192.OOO)
723(191,000)
652(173,000)
738 (195.00O)
636 (168,000)
757 (200.OOO)
703 (186,000)
737 (195,000)
745 (197,000)
745 (197,OOO)
1.0
1.0
1.6
1.6
1.6
1.6
1.4
1.4
1.4
1.4
1.4
1.4
1.0
1.4
1.4
1.4
1.4
50
48
48
48
48
47
45
47
47
52
46
53
45
48
46
45
45
5.2
5.1
5.1
5.1
5.0
5.0
4.7
4.9
4.9
5.5
4.8
5.6
4.7
5.1
4.8
4.8
4.8
18.1 (444)
18.7(459)
18.8 (462)
18.8 (462)
18.9 (464)
19.1 (467)
20.0(491)
19.3 (474)
19.3 (472)
17.4(427)
19.6(481)
16.9(415)
20.1 (494)
18.7(459)
19.6(481)
19.8 (486)
19.8 (486)
*N.P. = no polymer;'W.P = with polymer; W.A. = with alum.
^Based on influent flow.
-------
Table 2. Pilot Plant Operations Data Summary
Operating Mode*
(Dates)
Raw Wastewater
Constant flow - N.P.
06 7 1 0/80-07/10/80
Constant flow-W.P.
05/19/80-06/09/80
Diurnal flow - N.P.
07/1 4/80-07/3 1/80
08/01/80-08/31/80
09/01/80-09/30/80
10/01/80-10/20/80
02/19/81-02/28/81
03/01/81-03/31/81
04/01/81-04/10/81
05/31/81-06/30/81
07/01/81-07/31/81
08/01/81-08/28/81
Inf. Temp. Chem. Dose
(°C) (mg/L)
19-23
16-20 1.5-2.8
24-25
23-25
23-25
27-23
11-14
11-13
11-16
19-22
22-25
24-25
Vol. Org.
Loading^
3.3 (205)
2.5 (154)
2.8(174)
2.8(173)
3.9 (245)
4.4 (275)
3.3 (204)
4.9 (304)
4.4 (270)
2.5(159)
3.7(233)
2.6(164)
(mg/L)
5.700
5,210
7.817
6.408
8.633
6.931
5.700
6.908
7.151
8.261
7.772
9.771
MLSS
(% Vol.)
67
69
64
64
67
70
72
75
75
63
69
59
F/M
Loading^
0.93
0.88
0.91
0.68
0.75
1.00
0.77
0.95
0.79
0.52
0.54
0.51
SRT
(days)
1.9
1.0
1.4
1.9
2.2
1.6
1.9
3.7
3.3
2.5
Primary Effluent
Constant flow-N.P.
11/09/80-11/21/80
15-19
3.4 (214)
6.520
72
0.88
2.6
Diurnal flow - N.P.
12/07/80-12/12/80
01/24/81-01/30/81
Diurnal flow - W.P.
12/14/80-01/12/81
Diurnal flow - W.A.
01/13/81-01/23/81
14-15
14-15
9-17
13-15
0.5-3.0
40.0
3.5(221)
3.4(215)
2.2 (137)
2.2 (137)
7.260
6.757
5.720
6.403
70
68
71
68
0.76
0.5S
0.55
4.0
*N.P. = no polymer; W.P. = with polymer.' W.A. = with alum.
t*ff BODs/day/m3 (Ib BODs/day/1.000 ft3).
\kg BODs/day/kg MLVSS.
removals occurring in the main plant's
degritting unit.
Discussion of Results
Treatment Efficiency
In evaluating the performance of the
Deep Shaft process and its capability to
remove pollutants, as measured by key
parameters such as BODs, COD, and SS,
it should be noted that the influent
wastewater characteristics at the pilot
plant were not the typical domestic raw
wastewater or primary effluent that had
been anticipated when Ithaca was
selected as the site for the demonstration
project. This resulted from such factors
as abundant infiltration/inflow into the
City's collection system, recycle streams
from digester operations at the main
plant, partial grinding of screenings at
the main plant, trickling filter recircu-
lation flows to the primary settling tanks,
and suspected acid waste discharges
from a local industry.
Influent wastewater pollutant concen-
trations were highly variable. The organic
strength of the wastewater was quite low
even with the presence of recycle
streams back to the head end of the plant.
During several periods, the average in-
fluent BOD5 was less than 100 mg/L and
some days it was less than 50 mg/L.
BODs/COD ratios were less than would
be expected for typical domestic waste-
water. The ratios were generally in the
range of 0.3 to 0.4.
The Deep Shaft system effectively
removed BOD5, COD, and SS during
periods when a routine operation was
established and when no major
operational or mechanical difficulties
were encountered. When such difficul-
ties were encountered, process perfor-
mance was adversely affected. Effluent
soluble BODs levels were consistently be-
low 10 mg/L during all periods and condi-
tions of operation.
Primary Effluent Versus Raw
Wastewater Operation
The system appeared to be more
effective in treating raw wastewater than
in treating primary effluent. While
treating primary effluent, the float
blanket was very thin and usually
dispersed with no distinct endpoint; while
treating raw wastewater, it had a much
more healthy appearance, was usually
thick, and often had a distinct endpoint.
This probably resulted from the lower
organic strength associated with primary
effluent and the lack of fibrous material to
assist in forming floe particles that will
more readily float.
Unless required for other reasons, the
use of primary settling tanks is not
necessary or economically advantageous
with the Deep Shaft process. No savings in
Deep Shaft energy requirements will be
realized from the reduction in BOD
concentration achieved with primary
treatment unless the wastewater has a
high organic settleable solids fraction.
Constant Versus Diurnal Flow
Operation
The pilot plant was operated under
constant and diurnal influent flow
patterns on both raw wastewater and
primary effluent feed. The system operated
more efficiently under a constant flow
pattern than under a diurnal flow pattern.
Flow variations appeared at times to
disrupt the float blanket in the flotation
tank. Sometimes, under diurnal flow
conditions, the float blanket would
thicken during the morning, start to
-------
disperse by mid-afternoon, completely
disappear at night, and reappear the
following morning. This pattern may have
resulted from the large variation in BOD5
loading experienced throughout the
course of a typical day. With experience, a
better sludge wasting schedule was
established and the float blanket was
more controllable.
Polymer Addition
Process performance with polymer (a
floatation aid) added under different
operating conditions was compared with
performance without adding polymer.
The goal was to operate the system
without polymer because it represents a
potential significant operating cost.
Adding polymer did not improve effluent
quality. The concentration of solids in the
float blanket did not increase although
the quantity of float solids relative to the
quantity of sink solids increased substan-
tially.
In general, polymer was not needed to
develop a healthy float blanket or a good
quality effluent although, on occasion,
the blanket did become quite thin and
dispersed without it. Contributing factors
probably included one or more of the
following: colder wastewater temperatures,
weak influent strength, lack of fibrous
material, digested or partially digested
solids, and process upsets. At a full-scale
facility, adding polymer would be beneficial
on an intermittent or seasonal basis to
ensure a good, thick, healthy float blanket
at all times.
Alum Addition
While the pilot plant was treating
primary effluent, alum (for phosphorus
removal) was added with and without
polymer. The float blanket became quite
thin and dispersed as most of the solids
sank to the bottom of the flotation tank. A
thin, milky scum appeared on the surface
of the flotation tank. When polymer was
added, it did not significantly improve the
situation.
Alum was not added to the pilot system
while raw wastewater was being treated.
In full-scale applications, the need for
tertiary treatment of Deep Shaft effluent
to obtain phosphorus removal will
depend on the yet-to-be determined
compatibility of in-process alum or iron
addition with Deep Shaft operations
when the system feed is degritted raw
wastewater rather than primary effluent.
Organic and Hydraulic Shock
Loadings
The pilot plant was subjected to a
number of severe organic shock loadings,
sometimes several a week, primarily as
a result of digester operations at the main
plant. Shock organic loadings consisted
of digester supernatant, filtrate from
sludge dewatering, and digested and
partially digested solids from the sludge
holding tank. During startup, before the
primary effluent piping was installed, the
pilot plant received the full impact of
these recycle streams. After the piping
was installed, the influent to the pilot
plant was generally switched to primary
effluent during the "dump" periods.
During a typical digester "dump"
period (lasting approximately 5 hr), the
influent SS load increased to 3,800 mg/L
and the COD concentration increased to
5,000 mg/L. Effluent SS rose to mg/L
from a value below 30 mg/L, and effluent
soluble COD reached a peak of 140 mg/L
from a value of 100 mg/L. The process
was adversely affected only on a short-
term basis and rapidly recovered within
1 2 to 24 hr following cessation of the
"dump" period.
The pilot plant was operated under a
diurnal flow pattern with sustained peak
flows to simulate dry-weather conditions
at a typical domestic wastewater treat-
ment facility. It was not possible, however,
to evaluate the capability of the process to
handle the large instantaneous or short-
term peak flows typically encountered
during wet-weather periods at most
treatment plants. The design of the Ithaca
shaft was such that the total hydraulic
capacity including recycle flows was
1,666 mVday (440,000 gpd). The maxi-
7able 3. Representative Deep Shaft Performance at Ithaca
BODs (mg/L) COD (mg/L)
SS (mg/L)
Period
5/19 - 7/10/80
4/01 - 4/10/81
5/31 -8/28/81*
Weighted avg.
Inf.
91
137
105
102
Eff.
17
17
22
20
%Red.
80
88
79
80
Inf.
261
331
250
263
Eff.
63
64
52
59
%fted.
76
81
79
78
Inf.
172
207
267
231
Eff.
21
28
31
27
%Red.
88
86
88
88
"Effluent BODs results may relect some nitrogenous oxygen demand; nitrification inhibitors were
not used in the BODs analysis although partial nitrification did occur in the system during a por-
tion of this time period.
mum influent flow limit was established
at approximately 1,060 mVday (280,000
gpd).
Sludge Production and
Thickening
Significant amounts of solids had to be
wasted from the sink recycle line as well
as from the float return to properly control
the process. It had been anticipated that
most solids would float, forming a very
thick blanket that would be the primary
source of sludge wasting, and that float
recycle would be the primary mechanism
for controlling the shaft MLSS level. The
relative rates of float and sink solids
wasting varied considerably. At times,
the quantity of sink solids wasted
exceeded that of float solids wasted; at
other times, the opposite was true. This
variation was a function of the quantity
and condition of digester solids received
from sludge processing operations at the
main plant as well as such factors as
polymer addition.
The shaft MLSS level was controlled
primarily by controlling the sink solids
inventory. The concentration of float
solids was not as high as anticipated, and
the float return rate was much less than
the sink recycle rate. Float solids concen-
trations were about 4 percent, and sink
solids concentrations were around 2
percent. The float return rate was about
76 mVday (20,000 gpd); the sink return
rate was always at least 303 mVday
(80,000 gpd) and often over 379 mVday
(100,000 gpd). Consequently, the process
was primarily controlled by the wasting
and recycle rates for the sink solids.
Process Control
The system was more complicated to
operate effectively than had been original-
ly anticipatedpartly because of the
abnormal number of mechanical break-
downs and instrumentation malfunctions
and also because of the operator under-
standing and judgement needed to make
effective decisions. Daily control variables
included float solids wasting, sink solids
wasting, sink recycle rate, float skimmer
speed, addition of flotation aids, and sink
recycle by pump or gravity return.
Conclusions
Numerous problems including operator
control, mechanical breakdowns, partial
shaft blockages caused by partial grinding
and subsequent return of screenings to
the influent flow at the main plant
headworks, foaming conditions, and
abnormally strong digester return adverse-
ly affected process performance and
-------
often resulted in inconclusive and
inconsistent data. However, sufficient
reliable data were obtained to reach the
following conclusions:
The Deep Shaft process is a high-
rate, high-intensity activated sludge
process utilizing flotation for solids
separation. At the Ithaca pilot plant
site, F/M loadings ranged from 0.5
to 1.0 kg BOD5/day/kg/MLVSS,
SRT's were 1 to 4 days, and
bioreactor HRT's were 45 to 53 min,
exclusive of recycle flows.
During those periods when routine
operation on raw wastewater feed
was established, effluent concentra-
tions of BOD5 and suspended solids
averaged 20 and 27 mg/L, respec-
tively, and percent removals were
80 and 88 percent, respectively.
These percent removals are for the
Deep Shaft process only and do not
account for any reductions that may
have occurred in the main plant
degritting unit. Nitrification inhibi-
tors were not used in the BOD5
analysis, although partial nitrification
occurred near the end of the study.
The process was more effective in
treating raw wastewater than in
treating primary effluent, perhaps
because of the latter's more dilute
nature (lower organic loading) and
lack of fibrous material.
The system operated better under
constant flow conditions than
under diurnal flow variations. The
peak hydraulic capacity of the
system was such that wet-weather
peak flow conditions typical of
Ithaca could not be simulated.
Adding polymer increased the
quantity of the float blanket but did
not significantly improve effluent
quality. Low polymer dosages
caused some sink solids to remain
suspended and, thus, adversely
affected solids separation.
While treating primary effluent,
alum was added for a short time to
remove phosphorus; this caused
solids to sink and apparently adverse-
ly affected the float blanket.
The Deep Shaft process handled
shock organic loadings from solids
processing returns very well and
recovered within 12 to 24 hr after
the shock loads ended.
Ithaca's Deep Shaft system required
significant operator training, atten-
tion, and control. Control of the
solids inventory and maintenance
of a good, thick float blanket
required operator judgment in
wasting float solids, wasting sink
solids, and determining the sink
recycle rate on a day-to-day basis.
Problems at the main plant made
operating the pilot plant substantial-
ly more difficult.
The full report was submitted in
fulfillment of Cooperative Agreement No.
CS806081 by the City of Ithaca, NY, under
the partial sponsorship of the U.S.
Environmental Protection Agency.
Donald E. Sch winn, Donald F. Storrier, and Robert Butterworth are with Stearns &
Wheler, Cazenovia. NY 13035.
Richard C. Brenner is the EPA Project Officer (see below).
The complete report, entitled "Evaluation of Deep Shaft Biological Wastewater
Treatment Process at Ithaca, New York," (Order No. PB 84-110 485; Cost:
$16.00, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield. VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Municipal Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
J.S.OFFICIALJ/AIL;
Official Business
Penalty for Private Use $300
PS 0000329
U S ENVIR PROTECTION AGENCY
REGION 5 LIBRARY
230 S DEARBORN STREET
CHICAGO IL 60604
------- |