United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-82-051 August 1982
Project Summary
Chlorine Dioxide
Disinfection and Granular
Activated Carbon Adsorption
Ben W. Lykins, Jr., and Mark Griese
A pilot plant study was designed to
evaluate chlorine dioxide disinfection
as an alternative to chlorine for drasti-
cally reducing or preventing the pro-
duction of trihalomethanes. Also in-
vestigated was the effectiveness of
post-treatment adsorption by virgin
and reactivated granulated activated
carbon (GAC) for removal of organic
compounds that were either present in
the source water or formed after dis-
infection.
The performance of a 0.38 m3/ min
(100 gpm) pilot plant using chlorine
dioxide disinfection for both raw and
treated water was compared with that
of a full-scale plant applying chlorine
disinfection at similar locations. Two
parallel post-filtration adsorber con-
tactors consisting of virgin in one
contactor and reactivated GAC in the
other were continuously operated and
evaluated until exhaustion. After
about 90 days of operation, the GAC
was educted and reactivated offsite by
an infrared furnace.
Composite samples of spent and
reactivated GAC were collected during
each run, and a representative virgin
GAC sample was taken directly from
the shipping bags for characterization.
Various properties were determined
on these samples to ensure that the
spent GAC had been reactivated to
essentially virgin conditions and to
provide an opportunity to compare the
performances of virgin and reactivated
GAC.
Analyses performed during the pro-
ject consisted of the quantification of
14 volatile and 8 extractable organic
compounds along with qualitative
(detected/nondetected) determina-
tions by mass spectrometer scans of
32 additional volatiles and 54 extrac-
table organics. Additional data were
collected for 9 inorganic metals, total
organic carbon (TOC), and other pa-
rameters such as turbidity, standard
plate count (S PC), and total colif orms.
This Project Summary was develop-
ed by EPA's Municipal Environmental
Research Laboratory, Cincinnati. OH,
to announce key findings of the re-
search project that is fully documented
in a separate report of the same title
(see Project Report ordering informa-
tion at back).
Introduction
An experimental study was initiated
at the waterworks facility in Evansville,
Indiana, to help satisfy a basic need
within the water treatment and supply
industry for producing drinking water
relatively free of organics. This location
provided insight into the feasibility of
implementing an alternative method of
disinfection in a typical surface water
plant. The specific objectives estab-
lished for this project included:
— to develop a water treatment
process using chlorine dioxide as
a disinfectant and to compare its
-------
Table 1. Experimental Modes of Operation for Pilot Plant
Run
Chlorine
Chlorine Dioxide Alum Polymer Mix
Settle
Lime
Chlorine
Filter Carbon Chlorine Dioxide
1
2
3
4
X XX
X X
X
X X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X X
X
X
X
X
X
production of trihalomethanes
with those produced during chlo-
rine disinfection.
— to determine whether any organic
byproducts are formed when using
chlorine dioxide as opposed to
chlorine as a disinfectant.
— to determine the disinfection ef-
ficiency of chlorine dioxide and its
ability to provide residual neces-
sary for full-scale distribution use.
— to determine the effectiveness of
virgin and reactivated GAC for
removing organic compounds that
were either present in the filtered
water or formed after chlorine
dioxide disinfection.
Treatment Method Selection
The initial phase of the project con-
sisted of a control study and three
experimental modes of operation (Table
1). Forthe control study (Run 1), the pilot
plant was operated for 2 weeks using
pre- and post-treatment chlorine disin-
fection and bypassing the GAC
contactors to allow a performance com-
parison with the full-scale plant. This
study revealed that the average concen-
trations of the two plant effluents were
comparable. In addition to average
values, the data for total instantaneous
trihalomethanes (THM's) and TOC were
subjected to the t-test at a 95 percent
confidence level. This test showed no
significant difference for either param-
eter when data from the pilot plant were
compared with those for the full-scale
plant; the full-scale plant could thus be
used as a control for subsequent experi-
mentation.
The other three runs were 3 weeks
long and were evaluated to determine
which treatment method would be used
during the long-term operating phase.
The method of disinfection used for Run
2 consisted of chlorine addition to the
raw water and to the GAC effluent just
before it entered the clearwell. For this
and subsequent runs, the GAC con-
tactors were placed in operation.
The instantaneous average data for
the second run showed an increase in
the total trihalomethane(TTHM) concen-
tration from 1.3/ug/Lintherawwaterto
31.3 ;ug/L in the GAC influent. After
passing through the GAC contactors,
the total instantaneousTHM concentra-
tion decreased to 0.1 fjg/L. The average
TOC concentration decreased from 2.6
mg/Lintherawwaterto2.1 mg/Lafter
mixed media filtration, and it was further
reduced to 0.3 mg/L by the GAC con-
tactors.
Total coliform counts were reduced
from an average of 4,800/100 ml in the
second run raw water to <1/100 mL
after pre-treatment disinfection. Very
little regrowth occurred through the
carbon, with the average GAC effluent
containing 1/100 mL. Subsequent disin-
fection reduced the total coliforms to
<1/100mL
Mass spectrometer scans of both the
volatiles and extractables showed only
those instantaneous compounds previ-
ously identified by gas chromatography
(GC) evaluation and verified the removal
of these compounds by GAC treatment.
The third run consisted of a treatment
method using no pre-GAC disinfection.
Chlorine dioxide disinfection was used
only after the GAC contactors. This
method of treatment permitted the study
of organic removal with GAC and with-
out interference from those organic
compounds attributed to disinfection; it
could also be used to determine whether
any byproducts were formed when
chlorine dioxide was used after GAC
treatment.
The TOC concentrations were reduced
by almost 40 percent (4.2 to 2.5 mg/L)
after pilot plant treatment. Complete
treatment, including GAC, reduced TOC
concentrations by 80 percent (4.2 to 0.8
mg/L). Breakthrough occurred in TOC
concentrations in about 1 week at GAC
depths of 6 and 18 in. At 30 and 78 in.
(with 7.7 min of empty bed contact time
at 78 in.), no breakthrough was detected
during the run.
Various samples were stored for 3
days with chlorine and chlorine dioxide.
The average concentrations showed a
large increase in the raw water THM
level when stored with chlorine (1.2 to
247jug/L)and a slight increase in those
samples stored with chlorine dioxide
(1.2 to 2.9 jug/L). The slight increase
with chlorine dioxide is attributed to a
small amount of unreacted chlorine
present after generation. The final THM
concentration after 78 in. of GAC was
38//g/L and 0.7/ug/L, respectively, for
chlorine and chlorine dioxide storage.
Although chlorine dioxide disinfection
after the GAC contactors reduced the
total coliform and SPC to acceptable
values, slime growths were a problem in
the treatment plant because no disinfec-
tant was added to the raw water.
The objectives of the fourth run were
to determine whether any byproducts
were formed when chlorine dioxide was
used as a pre-treatment disinfectant,
and to evaluate GAC removal of any
organic compounds present or formed.
Average concentrations for the instan-
taneous THM's showed the raw water
value of 3.6 fjg/L being reduced to 0.2
//g/L through the pilot plant clearwell.
Very similar values were noted for the
THM's produced when stored with a
chlorine dioxide residual (3.7 //g/L for
the raw water and 1.1 //g/L for the
clearwell effluent).
As in previous runs, the qualitative
mass spectrometer scans did not indi-
cate the formation of any byproducts
relative to this mode of treatment. Also,
no increase in the concentration of
metals occurred after passage through
the activated carbon contactors.
Because a pre-disinfectant is required
to control slime growth in the treatment
plant. Run 3 was eliminated from con-
sideration for further study. The second
and fourth runs were evaluated based
on an estimate of what the consumer
might receive from each of the treatment
modes. In both of these modes, adequate
-------
disinfection was accomplished and
mass spectrometer sea ns of the priority
pollutants did not showthe formation of
any byproducts relative to the disinfec-
tant used other than the THM's. There-
fore, a desire to adhere to the original
study objectives resulted in selecting
chlorine dioxide as a pre- and post-
treatment disinfectant for long-term
evaluation.
Pilot Plant Operation
After the pre- and post-chlorine diox-
ide disinfection operational mode was
selected for long-term studies, four
additional runs were completed. Each
run was approximately 3 months. For all
runs, the performance of the pilot plant
through GAC treatment was compared
with the performance of the full-scale
plant under normal operation (pre- and
post-treatment chlorine disinfection).
Also, two parallel contactors were used
with one always containing virgin GAC,
and the other one containing virgin GAC
during the first run and once, twice, and
thrice reactivated GAC for the remaining
three runs.
The effect of using pre-treatment
chlorine dioxide disinfection for main-
tenance of bacterial water quality and
the reduction of THM's was evaluated
by comparing the sand filter effluent's of
both plants. Terminal THM reduction
through coagulation and settling was
determined by storing all samples for 3
days (estimated residence time in the
Evansville distribution system) at am-
bient temperature with a free chlorine
residual..
The use of pre-treatment chlorine
dioxide disinfection allowed substanti-
ally more of the THM precursors to be
removed by coagulation and settling
than with pre-treatment chlorination.
This was caused by less precursors
being converted to THM's in the pres-
ence of chlorine dioxide thereby allow-
ing more to be available for removal by
coagulation and settling. Therefore, the
THM's formed during post-treatment
chlorination and subsequently in the
distribution system would be reduced. A
comparison of the average terminal
THM concentration for all runs of both
disinfectants showed about 60 percent
less concentration of THM's when chlo-
rine dioxide was used for pre-treatment
disinfection. Though THM precursors
were reduced by using pre-treatment
chlorine dioxide, no difference was
noted in the averageTOC concentration,
regardless of the type of pre-treatment
disinfectant used.
Because of the THM reductions noted
by the use of pre-treatment chlorine
dioxide disinfection, an additional study
was completed to determine the feasi-
bility and benefits of using chlorine
dioxide in Evansville's full-scale plant.
Although chlorine dioxide could be used
as a pre-treatment disinfectant, chlorine
was desired as the post-treatment dis-
infectant because of the size of Evans-
ville's distribution system and the need
to maintain a sufficient disinfectant level
throughout this system. Data generated
from previous runs indicated that main-
tenance of a chlorine dioxide residual
over an extended period of time was
difficult.
The pilot plant GAC contactors were
bypassed to simulate the full-scale plant
as closely as possible. Chlorine dioxide
was introduced into the raw water in
sufficient quantities to maintain a slight
residual through mixed media filtration
(an average addition was 1.1 mg/L, and
the average residual after filtration was
0.1 mg/L). An attempt was made to
keep the total oxidant level below 0.5
mg/L. During this run, the average
concentration of chlorine dioxide plus
chlorite was 0.7 mg/L. The instantan-
eous THM data after filtration again
showed a reduction inTHM's when pre-
treatment chlorine dioxide disinfection
was used. No difference in the bacterio-
logical quality of the two plant effluents
was noticed when chlorinated effluents
were compared.
Each of the parallel contactors con-
tained 545 to 590 kg (1,200 and 1,300
Ib) of a lignite-based GAC. During each
of the four runs, one contactor always
contained virgin GAC and the other
contained reactivated GAC. Analysis of
the GAC showed that the spent carbon
in each reactivation was restored at
least to its virgin state.
Cumulative surface area distribution
curves indicated that reactivated GAC
for all runs had a total surface area
greater than the virgin GAC. This addi-
tional surface area was developed pri-
marily in the small pore region (less
than 30 Angstroms). Also, the iodine
number and cumulative surface area
correlated almost exactly indicating that
the pores were open and maintained.
Thus it was unlikely that any gradual
degradation of adsorptive capacity would
occur after subsequent reactivations.
Ratios for Brunauer-Emmett-Teller (BET)
surface area/iodine number for an
average of two virgin samples and each
of the three reactivation cycles were
1.08,1.06,1.08, and 1.03, respectively.
Organic loading on the GAC was
evaluated for TOC by comparing the
sum of TOC removed to the sum of TOC
applied per unit weight of carbon. This
loading indicated that the virgin and
subsequently reactivated GAC per-
formed essentially the same (see Figure
1). The average TOC carbon use rate for
both the virgin and the reactivated GAC
was about 0.05 kg/m3 at the steady-
state conditions that occurred after
about 60 days of operation.
Conclusions
Pilot-plant evaluation of chlorine di-
oxide disinfection and GAC treatment
have provided some insight into this
treatment scheme. The use of either
chlorine dioxide or chlorine in raw water
provides similar disinfection qualities
through conventional treatment. Also,
because chlorine dioxide has less de-
mand and possibly strong oxidative
properties, smaller concentrations were
needed to produce effective disinfection.
By using less applied chlorine dioxide,
some of the difference between the cost
of chlorine dioxide and chlorine can
probably be recouped.
Application of chlorine dioxide to the
raw water produced less concentration
of THMS'sinthefinished water than did
chlorine addition (average of 51 //g/Las
opposed to 130 fjg/L for all runs).
Furthermore, chlorine dioxide disinfec-
tion produced no organic byproducts
other than those noted with chlorine
disinfection (i.e., among priority pollu-
tants).
GAC was spent and reactivated three
successive times and essentially re-
stored to virgin state as indicated by
several GAC properties and perform-
ance comparisons with virgin GAC.
Gross estimates of volumetric losses
during GAC offsite reactivation were
about 5 percent, and overall system
losses (eduction, handling, transport,
and reactivation) averaged about 8
percent.
Production of chlorine dioxide did not
proceed stoichiometrically, and thus
some chlorite and chlorine were pro-
duced. Out of the generator, on the
average, the percentage of chlorine
dioxide, free chlorine, and chlorite pro-
duced was 64, 6, and 30, respectively.
Both chlorine and chlorine dioxide were
effectively removed by the GAC and the
chlorite concentration was lowered.
> US GOVERNMENT PRINTING OFFICE 1882-559-017/0751
-------
5
5
5
I
0.05
0.04
0.03
0.02
o.o r
0.0
Virgin GAC
Reactivated GAC
0.0 0.02 0.04 0.06
Sum of TOO applied, mg/kg GAC
Figure 1. Comparison of virgin and once reactivated GAC for TOC removed
vs TOC applied.
0.08
In some situations, the use of chlorine
dioxide as the primary raw water disin-
fectant is a viable alternative to chlorine
disinfection when the THM concentra-
tions must be reduced to meet promul-
gated drinking water standards.
The full report was submitted in ful-
fillment of Cooperative Agreement No.
CR804902 by the City of Evansville,
Indiana, under the sponsorship of the
U.S. Environmental Protection Agency.
The EPA author Ben W. Lykins, Jr. (also the EPA Project Officer, see below) is
with the Municipal Environmental Research Laboratory, Cincinnati, OH
45268; Mark Griese is with the Evansville Waterworks Department, Evans-
ville, IN 47740.
The complete report, entitled "Chlorine Dioxide Disinfection and Granular
Activated Carbon Adsorption," (Order No. PB 82-227 737; Cost: $10.50,
subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
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
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
RETURN POSTAGE GUARANTEED
Third-Class
Bulk Rate
MERL0063240
LOU ft TILLEY
RESIGN V EPA
LIBRARIAN
230 S OEARBQRM ST
CHICAGO IL 60604
------- |