United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
~^
-------
Waddell, and L. Kaplan. 1980. Removing
Potential Organic Carcinogens and
Precursors from Drinking Water. Vol. I.
EPA-600/2-80-130a, U.S. Environ-
mental Protection Agency, Cincinnati,
Ohio). This bed life was based on the
number of days required to reach a
certain breakthrough of trihalomethane
formation potential (THMFP). The
breakthrough was the point at which the
total trihalomethanes (TTHM's) reached
100 yug/L when the effluent was
rechlorinated to a free chlorine residual
of 2.5 mg/L and held for 2 days.
The purpose of the present study was to
extend this bed life by using the European
method of ozonation and carbon
adsorption. This treatment technique
took advantage of aerobic biological
oxidation on granular activated carbon
(GAC). Bed life was said to extend well
beyond the period of pure adsorption, and
the mechanism appeared to be one by
which biodegradable organics were
metabolized by bacteria growing on the
GAC. Further extension of bed life could
reportedly be obtained by ozonation of the
water feeding the GAC adsorbers. This
ozonation was believed to modify the
naturally occurring organics and render
them more readily degradable by the
bacteria.
Consequently, a 22.7-m3/hr (100-
gpm) pilot plant was designed, construc-
ted, and operated to evaluate this process
on groundwater in Miami, Florida. The
pilot plant was located at the Preston
Water Treatment Plant in Hialeah, Florida.
The raw groundwater feed was from
wells 27.4 m (90 ft) deep on the plant site
and contained substantial amounts of
naturally occurring organics--8 to 10
mg/L total organic carbon (TOC) and 600
to 800 /ug/L THMFP.
The pilot lant was designed to study
parallel systems of GAC adsorbers that
were 3.05 m (10 ft) deep, hadanEBCTof
24.8 min, and received ozonated and
nonozonated, lime-softened water that
was recarbonated to pH 8.5 and sand
filtered. The GAC adsorbers were opera-
ted under anaerobic conditions in the
first study and aerobic conditions in the
second study. The absorbed ozone
dosages in these studies were 11.3 and
14.5 mg/L, respectively. The bed life for
each of the two parallel systems was
determined.
In addition to the pilot studies to extend
GAC bed life, bench-scale studies were
conducted to determine:
1) The best of four commercially avail-
able GAC's for removing THM
precursors from the local ground-
water.
2) The bacteria profile of the Preston
Water Treatment Plant, including
the distribution system.
3) The bacteria profile of the bench-
scale GAC work.
4) The ability of GAC to remove agri-
cultural and industrial organic
chemicals from waters high in TOC.
5) The source of chloroethene com-
pounds present in raw groundwater
in the Miami area and in other
locations in Florida.
6) Removal of volatile halogenated
organics from water by aeration
techniques.
These studies are detailed in the
appendices to the main report.
Plant Conditions
Conclusions in this report are based on
the particular groundwater feed and
water treatment practices existing at the
Preston Water Treatment Plant during
the study period. Groundwater
containing high concentrations of TOC
and THM precursors was lime-softened,
breakpoint-chlorinated, and sand-
filtered. Finished water leaving the plant
had a free chlorine residual of 2.5 to 3.0
mg/L for continued disinfection in the
distribution system. The average TTHM
concentration in the finished water was
150 /ug/L. This concentration increased
to 300 /ug/L in the distribution system
because of the continued reaction of free
chlorine residual with precursor
substances remaining in the finished
water.
Both the original research project and
the present study were based on the
possibility of removing enough THMFP at
the water treatment plant to limit further
growth of TTHM in the distribution
system to 100 /ug/L (assuming that the
finished water left the plant with a free
chlorine residual of 2.5 to 3.0 mg/L).
Under the given plant conditions, a
GAC adsorber effluent breakthrough of
approximately 3 mg/L of TOC or 200
/ug/L of THMFP was found to produce the
100 /ug/L TTHM concentration. In this
report, bed life based on these conditions
is referred to as "Criterion 1" bed life.
Procedures
Pilot Studies
The first of the two pilot-plant studies
evaluated the effects of ozonated and
nonozonated influent on the removal of
THMFP with GAC adsorbers operated |
under anaerobic conditions. After the
psuedo-steady-state region of the THMFP
breakthrough curve was reached on the
GAC adsorber effluents, both GAC
adsorbers were oxygenated for aerobic
operation to determine whether the
effective GAC bed life could be extended
by a change to aerobic operation.
In the second pilot-plant study, both
ozonated and nonozonated GAC systems
were oxygenated on initial startup to
compare aerobic operation with the
anaerobic operation of the first study.
Other studies were conducted in the pilot
plant to determine !->ow TOC and THMFP
were affected by absorbed ozone
dosages at various pH ranges.
Throughout all ozone studies, a
complete bacterial profile was made to be
compared with that of conventional
breakpoint chlorination.
Bench-Scale Studies
Both bench- and pilot-scale studies
were conducted at the same time. The
first bench-scale project was designed to
select the best of four commercially
available GAC's for use in the pilot plant.
Selection was based on the greatest I
adsorptive capacity (longest bed life) for
removing THM precursors from the
Preston plant finished water. The best
GAC selected was then evaluated for
bed-life performance on lime-softened
Preston plant water that was first
adjusted to pH 8.5 by recarbonation and
then sand-filtered. This type of water feed
was used to supply both the ozonated
systems and the nonozonated controls.
During the bench-scale studies, a
complete bacteria profile was made for
the Preston plant. This profile included
the distribution system and all bench-
scale systems.
Methods were developed during the
bench-scale studies for continuous
spiking of plant effluents with
agricultural and industrial pollutants.
Studies were then made on the removal
of these compounds by the GAC adsorbers.
Such information will be useful if the
local groundwater supply ever becomes
contaminated, since little is known about
the ability of GAC to remove such
pollutants from high-TOC waters.
Bench-scale studies were also
conducted to determine the source of
volatile, chlorinated ethene compounds
in raw groundwater in the Miami area
and in other locations in Florida.
-------
Results
Chemical Parameters
1. Of the four commercially available
GAC's tested for THMFP removal on
finished water, Calgon Filtrasorb
400* gave the longest bed life and
was selected for the pilot plant study.
2. When a 2.13-m (7-ft) deep bed of
Filtrasorb 400 was tested for
THMFP removal on lime-softened,
sand-filtered water adjusted to a pH
of 8.5, the Criterion 1 bed life was 21
days.
3. In the first pilot plant study, the
ozone- and nonozone-GAC systems
were operated anaerobically for the
first 98 days of the 160-day test.
Based on Criterion 1, the bed life
was 44 days for the ozone-GAC
system and 19 days for the nono-
zone control system.
4. In the first pilot plant study,
changing from anaerobic to aerobic
operation did not appear to alter the
shape of the psuedo-steady-state
portion of the TOC or THMFP
breakthrough curves.
5. In the second pilot plant study, the
ozone- and nonozone-GAC systems
were operated aerobically
throughout the full 172-day test
period. Based on Criterion 1, the bed
life was 25 days for the ozone-GAC
system and 9 days for the nonozone
control system.
6. Because the TOC and THMFP
loadings were different for the two
pilot studies (anaerobic and aero-
bic), direct comparisons were
impractical.
7. Ozone in the ozone contractors
greatly reduces the concentration of
double bond chlorinated compounds.
At an absorbed ozone dose of 14,5
mg/L, the initial concentration of cis
1,2-dichloroethene, vinyl chloride
and chlorobenzene was reduced 90,
91 and 94 percent, respectively. At
an absorbed dose of 4.25 mg/L,
these compounds were reduced 73,
88, and 78 percent, respectively.
8. In a 335-day test of the removal of
spiked organics from finished water,
'Mention of trade names or commercial products
does not constitute endorsement or recommenda-
tion for use
several polyaromatic hydrocarbons
(PAH's), several halogenated
pesticides, and one polychlorinated
biphenyl (PCB) were fed to a 0.76-m
(2.5-ft) deep bed of Filtrasorb 400.
During the test period, this GAC
removed essentially 100 percent of
all the PAH's, 98.6 percent of all the
halogenated pesticides, and 97.5
percent of the PCB.
9. In an aeration system (i.e.,stirred
beaker, aeration tower, spray head),
the decrease in contaminant levels
in water with time is linear if the log
of the effluent concentration is
plotted versus time and if all
compounds present are in the
soluble range.
10. When aeration decreases contami-
nant levels in water with time, the
decrease is independent of other
organic compounds when all
compounds are present in the
soluble state. The decreased con-
centration with time is also inde-
pendent of initial concentration in
the range of solubility.
11. The source of volatile chloroethene
compounds (vinyl chloride, vinyli-
dene chloride, cis and trans 1,2-
dichloroethylene, etc.) in the local
groundwater is the result of the bio-
degradation of parent, volatile
chlorinated degreasing solvents
accidentally spilled on the ground
above the aquifer. The solvents
were trichloroethylene and/or
tetrachloroethylene. Chlorometh-
ane and ethane compounds are the
result of the biodegradation of
1,1,1 -trichloroethene.
12. Field data from an actual contami-
nant spill in the environment
confirmed the results of the labora-
tory biodegradation studies.
Biological Parameters
1. In the GAC beds used to filter raw,
lime-softened, or chlorinated
finished drinking water, large
numbers of bacteria grew and
sloughed off into the water flow.
2. Bacteria isolated from the GAC
adsorbers were common to the raw
water supply and were normal
inhabitants of soil and fresh-water
sediments.
3. Often, many of the bacteria that
made up the total plate counts (and
those most frequently encountered)
were unidentifiable by taxonomic
keys, did not survive subculturing,
or were inert in standard
biochemical tests.
4. Common heterotrophic organisms
recovered from raw and treated
water from the bench-scale tests,
the pilot plant, and the full-scale
treatment plant were species of
Alcaligenes. Psuedomonas, Morax-
el la. Acinetobacter, Micrococcus,
Flavobacterium, and Gram-positive
rods.
5. The composition of the GAC effluent
flora was generally the same as that
of the flora on the GAC surface, but
the species existed in different
proportions. Perhaps these results
reflect different resistances of the
species to dislodgement by the
water flow.
6. No real difference existed in the
numbers or kinds of bacteria
recovered from the GAC adsorber
effluents that received aerated,
oxygenated , ozonated, or untreated
influents. Many of the heterotrophic
bacteria cultured from samples of
the GAC or their effluents were able
to grow well without atmospheric
oxygen.
7. Bacterial populations in the GAC
adsorbers were dynamic. Periodic
sampling and analyses indicated
that the sizes and compositions of
these populations changed
constantly. Probable causes were
succession of species within the
adsorber and introduction of new
species from the raw groundwater
supply.
8. Gram-negative rods, which had no
especially resistant forms, survived
in GAC effluents that were
chlorinated to breakpoint and had
up to 10 mg/L free residual chlorine
at the time of plating.
Conclusions
Ozone applied in the 12-mg/L-range
may be a desirable in-plant primary disin-
fectant for future use in the local south
Florida area. At this concentration ozone
appears to be as effective as current
breakpoint chlorination at 18 mg/L for
pathogen and microorganism control.
-------
and it may provide even better color
reduction. At this dosage, ozone also
effectively oxidizes double-bond
chlorinated compounds that are present
in the raw groundwater. For residual
control in the distribution system, ozone
could be followed by a secondary disin-
fectant that would not form additional
THM's.
If the local groundwater source ever
becomes contaminated with organic
compounds that are not readily removed
by other means, the use of GAC
adsorption in this high-TOC water would
be very effective.
The full report was submitted in
fulfillment of Cooperative Agreement
CR806890 by Florida International Uni-
versity, Miami-Dade Water and Sewer
Authority, and Dade County Department
of Public Health under the sponsorship of
the U.S. Environmental Protection
Agency.
Paul R. Wood, Frances Z. Parsons. Russell F. Lang, Iris L. Payan, Susan S.
Espinet-Tracey are with Florida International University. Miami, FL 33199;
Doris H. Waddell, Richard A. Coates, Raymond D. Diaz. Martha S. Goldberg are
with Miami-Dade Water and Sewer Authority, Hialeah. FL 33010; and Fred W.
Curtis, Jr.. is with Dade County Department of Public Health. Miami. FL33125.
Jack DeMarco is the EPA Project Officer (see below).
The complete report, entitled "Pilot Plant Project for Removing Organic
Substances from Drinking Water." (Order No. PB 84-128 685; Cost: $29.50,
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 Protect/on Agency
Cincinnati, OH 45268
ftUS GOVERNMENT PRINTING OFFICE' 1984-759-015/7294
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
PS 0000329
U S ENVIR PROTECTION AGENCY
REGION 5 LIBRARY
230 S DEARBORN STREET
CHICAGO IL 60604
------- |