xvEPA
United States
Environmental Protection
Agency
Municipal Environmental Research'
Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-82-027 Sept. 1982
Project Summary
I»
Removal of Organic
Contaminants from Drinking
Water Supply at Glen Cove,
N.Y.: Phase II
Walter A. Feige and Dominick Ruggiero
Phase II continues pilot plant research
to examine treatment alternatives for
the removal of organic contaminants
from ground water. The specific
processes of resin adsorption and
aeration were evaluated for their
ability to remove trichloroethylene,
tetrachloroethylene, cis-1, 2-dichloro-
ethylene, and 1,1, 1-trichloroethane
from the City of Glen Cove's drinking
water supply.
The resin used was Rohm and Haas'
Ambersorb XE-340®*. Major additions
and modifications were made to the
resin testing system investigated
during Phase I to facilitate the study of
cocurrent and countercurrent flow
applications as well as countercurrent
steam regeneration techniques and
also to ensure the presence of pure
steam throughout the regeneration
process. In addition. Phase II included
an evaluation of the feasibility of
regenerating granular activated carbon
(GAC) with steam.
Aeration studies consisted of the
determination of organic contaminant
removal capabilities by different
aeration designs. The relative efficien-
cies of three aeration methods were
compared: diffused aeration vs. packed
aeration vs. induced draft cooling. Air-
'Mention of trade names or commercial products
does not constitute endorsement or recommenda-
tion for use.
to-water ratios (A/W) were varied
with each pilot system.
Flows of 1, 2, and 5 mgd (3785,
7575, and 18,925 m3/d) were chosen
to obtain a range of costs for a full-
scale installation that wishes to
consider diffused aeration, packed
column aeration, or a resin system for
treatment of its ground water.
Project results are being used to
conduct a third phase in which the
technical and economic feasibility of
an aeration-adsorption approach will
be assessed.
This Project Summary was developed
by EPA's Municipal Environmental
Research Laboratory, Cincinnati, OH,
to announce key findings of the
research project that is fully docu-
mented in a separate report of the
same title (see Project Report ordering
information at back).
Introduction
Background
In 1977, the City of Glen Cove, NY,
experienced a water shortage resulting
from the closing of several wells
polluted by the presence of organic
contamination. Realizing that the
contamination could migrate throughout
the aquifer, the city was interested in
finding methods for treating its contam-
inated water. A research project was
funded with the EPA in 1978 to study
-------
two treatment techniques; adsorption of
the contaminants by synthetic resin and
stripping of the contaminants by diffused
aeration. Pilot plant operation took place
for 9 months. The results indicated that
(1) the resi n was capable of lowering the
contaminants to below 1 microgram per
liter (Aig/L), producing an acceptable
water quality for several months and (2)
the aeration system, with the proper
A/W, could reduce contaminant levels
below present New York State guide-
lines. These promising results warranted
the continuation of this project with the
overall objectives for obtaining relevant
design parameters and capital and
operating costs for both removal tech-
niques.
COL
I
§!!
Granular A ctivated
Carbon Columns
Rohm & Haas XE-340
Resin Columns
Adsorption Testing—Flow and Steam Application Modes
Pilot Plant Facilities
The pilot plant consisted of two
separate operations (Figure 1) corre-
sponding to the two methods being
evaluated, adsorption and aeration.
Several changes were made to the pilot
plant equipment used during Phase I
before Phase II testing began. These
modifications and additions included a
change in the size of columns used for
the resin testing, the installation of
activated carbon columns, a larger
steam supply, and the installation of a
packed aeration column and a cooling
tower for air stripping. The same
influent well water flowed to the resin,
carbon, and aeration systems.
Changes to the resin system were
made to facilitate the testing of upflow
and downflow applications, countercur-
rent steam regeneration, and to ensure
that pure steam was present for the
regeneration of the resin. Three new
columns (III, IV, and V), shown in Figure
1, were installed, each 4 in. (0.1 m) in
diameter and 2.2 ft (0.67 m) high. This
height allowed for 2 ft (0.61 m) of resin
depth and 10% freeboard. Water flow
through the resin was in a downward
mode for Column III and upward mode
for Columns IV and V. An electric steam
generator, capable of producing 100
Ib/hr (45 kg/hr), was usedto regenerate
the resin.
Phase II included the evaluation of
regenerating GAC with steam. To
facilitate this evaluation, two 6-ft (1.8-
m) high columns, which were used
during the Phase I resin testing period,
were left in place and connected to the
new steam system. These columns
(Nos. land II in Figure 1) were each filled
with a 4-ft depth of activated carbon and
operated in a downward mode. Regen-
Water In
Air In
O
S Air In ^
£ (NoDiffus)
s Air In %
(Diffused)
kkkk
Water
Out
COL I COL II
(Diffused Aeration Column) (Packed Column)
Water Out-** .
Cooling Tower
Air Stripping Tests—Flow Application Modes
Figure 1. Pilot plant testing modes of operation.
eration of these columns was then
attempted in an upward mode.
To continue the evaluation of aeration
designs, a second column identical to
that used in Phase I aeration testing was
installed. Six feet of packing material
purchased from Munters Corporation
was placed in the column. This packing
rested on a ledge 2 ft above the bottom
of the column and air was applied below
the packing. The packing was supplied
in 1 -ft sections and was installed so that
the orientation of the channels in the
packing was altered from layer to layer.
The space between the packing and the
inside of the column.was sealed at each
1-ft interval to prevent air from circum-
venting the packing. Air was supplied to
both the column used in Phase I and the
new column from the same blower
utilized during Phase I.
To evaluate the stripping efficiency of
the cooling tower, a small demonstration
model of an induced draft cooling tower,
provided by Baltimore Air Coil, was
installed. This model is based on a self-
sustaining aerating system requiring no
outside source of energy other than the
water pressure itself.
Relative efficiencies of the three
methods of aeration were determined.
In addition, the packed aeration column
evaluation consisted of determining the
effect of introducing air into the column
below the packing media by two
methods, stone diffusers and a hole in
the air pipe at the center of the media.
The packed column was operated with
the media flooded, partially flooded, and
unflooded. Variables in the testing
program were the air-to-water ratios
(15:1, 10:1, and 5:1), the water depth in
the packed aeration columns, and the
water flow to the induced draft cooling
tower. The water flow to both aeration
columns was maintained at 28 gpm (6.4
mVhr).
Required analyses were performed
onsite with the gas chromatograph
system purchased for Phase I and the
addition of a new recorder purchased'
from Varian Corporation.
-------
Results
Adsorption Studies — XE-
340® Resin
A summary curve of the performance
of the three resin columns appears in
Figure 2 for the major contaminants
present. A water application rate of 4
gpm/ft3 of adsorbent (0.42 Lpm/m3)
was used, which resulted in an empty
bed contact time of approximately 1.9
min. Column III (virgin resin operating in
a downward mode) showed longer
service times than Columns IV and V
(virgin and regenerated resins operating
in the upward mode). In addition,
tetrachloroethylene was more effectively
removed than cis-1, 2-dichloroethylene.
For example, the performance of Column
III observed from Figure 2 shows the
following service times to reach an
effluent level of 5/yg/L: 43 days for cis-
1, 2-dichloroethylene, 110 days for
trichloroethylene, and 142 days for
tetrachloroethylene. The data indicate
the importance of establishing an
effluent target level so that proper
design, operating procedures, and costs
can be determined.
Several steam regenerations took
place and the results showed that, in
most cases, steam stripping removed
the majority of the organics from the
resins. Mass balances indicated about
90% of total organics by weight were
removed. To achieve this stripping
efficiency, approximately 20 to 30 bed
volumes of steam were required.
Methods of condensate disposal were
researched and the results are described
in the complete report. "Superloading,"
a process whereby the aqueous phase
of the condensate collected from one
regeneration cycle was returned to the
resin column next in need of regenera-
tion, appeared successful as observed
by the concentration of organics percent
in the organic layer of the newly
regenerated resin's condensate. A
further positive indication of success
was that the resin's performance
returned to about virgin conditions.
Granular Activated Carbon
Regeneration Studies
The purpose of adsorption testing
using activated carbon was to determine
the feasibility of regenerating spent
activated carbon with steam rather than
the adsorptive ability of the carbon.
Because a hydraulic loading twice the
manufacturer's recommendation was
used to load the carbon in as short a
time as possible, no attempt was made
to evaluate the adsorption results of the
activated carbon or compare them to the
resin.
Regeneration of activated carbon
with steam took place twice for each of
two GAC's using the equipment sized
for the regeneration of resin in Phase II
and no special considerations were
given to the carbon. The carbons were
regenerated using the maximum steam
flow available over a 24-hr period. No
attempt was made, however, to achieve
zero contaminants in the steam con-
densate.
i
? Colum
3V/Day - 77
-.BCT- 1.9m
/
6
V
I
/
t /
r
i
;
/
A
Y
1 1ll (Virgin Downflow
n IV (Virgin Upflow) .
n V (Regen Upflow)
.1 •
in
f
/
^
/
f
/
^
/
/
/
/
/
/
^
.J
*^
Cis- 1,2-Dichloroet'hylene
I I
I
I
3 5 tO 20 30 40
L
/
J
/
t
1
/
/
/
/
/
I
/
/
1
/
/
/
'
1
/
.
Tetrachloroethylene
1 1
0 5 10 20 30 40
Concentration fug/L)
1
I
I
}
//
J
*
r
A
-------
Several steam regenerations were
conducted, and the results showed low
regeneration efficiency when compared
to the resin. Mass balances indicated
that only 25% to 35% of the organics
were stripped from the carbons. In
addition, the effluent from the regener-
ated carbon columns exhibited break-
through contaminant concentrations in
approximately half the service time of
virgin resin or the previous carbon run.
Aeration Studies
Results from the diffused aeration
and packed column studies represent
percent removals for each organic
contaminant at the three A/Wevaluated
(Figure 3). In general, increased A/W
resulted in increased contaminant
removals for the designs tested — the
lowest average removals occurred at an
A/W = 5:1 and the highest occurred at
A/W = 15:1. The packed column,
operated in the unflooded mode with air
introduced through spargers, removed a
greater percentage of each contaminant
at a given A/W than the other systems
evaluated. In addition, removal efficien-
cies for both systems tested were
greatest for tetrachloroethylene (overall
average 84%) and least for cis-1, 2-
dichloroethylene (overall average 67%).
Both designs, however, resulted in
effluent contaminant levels below the
present New York State guideline level
(50 /ug/L per contaminant) 98% of the
time for all contaminants. The induced
draft cooling tower was the least
effective removal method evaluated
and, in fact, was never able to reduce
trichloroethylene levels below the guide-
line level.
Estimated Costs for Full-
Scale Installations
Figures 4 and 5 summarize the
estimated capital and operating costs
for full-scale design for the major
systems evaluated during Phase II. The
estimates indicate that the resin system
has the greatest capital cost and the
diffused aeration system the least.
Approximately 40% to 65% of the capital
cost for the resin system is for the
required piping and valves, particularly
those associated with steam regenera-
tion. Operating costs, however are
greatest for the diffused aeration
system and least for the resin system.
The greater costs for the aeration
systems are a result of the electrical
power required to provide the air. These
costs could be reduced if environmen-
tally acceptable alternative methods of
2
o
I
IUU
90
80
70
60
50
40
30
?n
1
'
t ^
^^,
^
1 ^^"""""^ j
*^~
• Diffused Aeration Column
A Packed Column - S (flooded}
* Packe
D Packe
•" Packet.
d Column - ,5
d Column - r
i Column H (
> (unflooded)
/ (flooded)
unflooded)
t
1
n>
a!
Figure 3.
5:1 10:1 15:1
Air:Water Ratio
Cis-1,2-Dichloroethylene
(Influent Range 28 to 79 ug/L)
IUU
on
Of*
fin
40
in
i
i
*
5:
Lx^'
1 10:
^
^^
1 15
)
:1
/L/C/
90
•35 80
>
o
5 70
*
r-^
j
r^^
•K--'
NKRE
Air:Water Ratio
Trichloroethylene
(Influent Range 117 to
Average percent removals at selected air: water ratios.
4
5:1 10:1 15:1
Air:Water Ratio
1,1,1 -Trichloroethane
(Influent Range 3 to
IUU
90
"(B
I 80
S
3 70
w
c
a> 60
U
a. 50
40
30
i
i
,
^^^—
L^—
|
5:1 10:1 15:1
Air-.Water Ratio
Tetrachloroethylene
(Influent Range 33 to 297 ug/L)
-------
7000
ENR 3383
Cost March 1981
100
3.0 4.0
Flow—M.G.D.
Figure 4. Capital cost comparison.
air introduction can be demonstrated
and used.
Conclusions
Resin Adsorption Studies
1. The test results using Rohm and
Haas' Ambersorb XE-340® indi-
cated that this resin was capable
of reducing the levels of contami-
nants to very low concentrations,
and, in most cases, virgin or
freshly regenerated resins achieved
reductions of each contaminant to
less than 1 /ug/L These results
were achieved for the following
influent ranges: cis-1, 2-dichloro-
ethylene, 38 to 79 pg/L; 1, 1, 1-
trichloroethane, 3 to 7 /ug/L;
trichloroethylene, 1 1 7 to 277
//g/L; and tetrachloroethylene, 33
to 207 /jg/L.
2. The best mode of operation was
downward flow application and
upward steam regeneration.
3. The ability of the resin to be
regenerated with steam, achieving
better than 90% of virgin adsorbent
capacity, was conclusively de-
monstrated. Steam regeneration
techniques, however, require
further development. Approxi-
mately 20 bed volumes of steam at
an application rate of 1 bed volume
/hr were required to regenerate
the resin. The collecting of thecon-
densate, separation of the organic
phase for disposal, and recycling
the condensate to the resin column
next to be regenerated for treat-
ment was determined feasible
during this phase of testing.
Granular Activated Carbon
Regeneration Studies
1. The results of the three 24-hr
regeneration attempts (two on the
ICI carbon and one of the Calgon
carbon) showedthatthe regenera-
tion of activated carbon with
steam, utilizing the steam equip-
ment and conditions designed to
regenerate resin, was not suc-
cessful.
Diffused Aeration Studies
1. Diffused aeration appears to be a
viable treatment option for the
removal of volatile organic con-
taminants from groundwater, and
this approach could be used as a
pretreatment to additional treat-
ment if more stringent water
quality is required than that
achievable by diffused aeration.
2. Aeration can reduce the contami-
nants present in the well water at
Glen Cove to levels below present
New York State guidelines, and to
levels that might be acceptable
throughout the nation. This was
true greater than 98% of the time
for all A/W tested (5:1, 10:1, and
15:1) and for all contaminants
involved.
Packed Column Aeration
Studies
1. Both conclusions made above for
the diffused aeration studies also
apply to the packed column aera-
tion studies.
2. The use of packing material in the
unf looded mode produced the best
results with a slight improvement
over those for diffused aeration at
A/W of 5:1 and 10:1; however, at
increased estimated capital and
operating costs.
Recommendations
1. Evaluate the combination of aera-
tion and adsorption as a possible
method of organic contaminant
removal from groundwater. This
-------
approach would reduce the major
portion of the contamination by
aeration and polish the water by
adsorption.
2. Determine the maximum hydraulic
loading for the resin system. To
date, only one application rate (4
gpm/ft3) has been used. Several
rates should be tested using both
raw and aerated well waters.
3. Continue the development of a
practical and economically feasible
stead regeneration technique. The
proper handling and disposal of
the steam condensate need to be
optimized and the economics
established.
4. Evaluate the process of multi-
stage aeration as an approach to
achieving organic contaminant
removal.
The full report was submitted in
fulfillment of Cooperative Agreement
No. 806355-01 by NebolsineKohlmann
Ruggiero Engineers, P.C., for the City of
Glen Cove, NY, under the partial
sponsorship of the U.S. Environmental
Protection Agency.
40
Diffused Aeration System
Packed Column
Aeration System
1.0
2.0
3.0 4.0
Flow - M.G.D.
Figure 5. Comparison of annual operating cost per thousand gallons treated
(energy cost at 9C/kw-hr).
TheEPA author Walter A. Feige (also the EPA Project Officer, seebelowjis with
the Municipal Environmental Research Laboratory, Cincinnati, OH 45268;
Dominick Ruggiero is with Nebolsine Kohlmann Ruggiero Engineers, P.O.,
New York, NY 10001.
The complete report, entitled "Removal of Organic Contaminants from Drinking
Water Supply at Glen Cove, N.Y.—Phase II," (Order No. PB 82-258 963;
Cost: $12.00, 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
•fr U. S. GOVERNMENT PRINTING OFFICE: 1982/559-092/0525
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
RETURN POSTAGE GUARANTEED
Third-Class
Bulk Rate
------- |