PB87- IbO^JI
EPA/600/D-87/113
April 1987
REMOVAL OF HUMIC MATERIAL BY CONVENTIONAL TREATMENT AND CARBON
by
Benjamin W. Lyklns, Jr.
Robert M. Clark
Hrinking Water Research Division
Water Engineering Research Laboratory
Cincinnati, Ohio 45268
Water Engineering Research Laboratory
Office cf Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268

-------
TECHNICAL REPORT DATA
{Please mc/ initntctiont ewr the m*rs* be fort compUtfnt)
V REPORT NO.
EPA/60Q/D*87/113
X AlCiri(NT*S ACCESSION NO.
4. TITLE ANO SUBTITLE
Removal of Humic Material by Conventional Treatment
and Carbon
5. RCrOAT OAT6
April 1987
ft. PERFORMING ORGANIZATION COOE
7. AUTHORIS)
Benjamin W. Lykins, Jr. and Robert M. Clark
8. PERFORMING ORGANIZATION REPORT NO.
fl. PERFORMING ORGANIZATION NAME ANO ADDRESS
Water Engineering Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
10.	PROGRAM ELEMENT NO
11.	CONTRACT/GRANT NO
12. SPONSORING AGENCY NAME ANO ADDRESS
Water Engineering Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45263
13. TYPE OP RfPORT
Proceedinqs
ANO PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Disinfection by-products are among those compounds being considered for refla-
tion under the Safe Drinking Water Act Amendments o* 1986. One of the most signi-
ficant disinfection by-products for those utilities that use chlorine are total tri-
halomethanes (TTHMs).- Pressure is growinq to reconsider the existinq TTH'1 Standard
of 100 ug/L and to lower it"to as yet some unspecified level. Trihalomethane levels
a_s low as 10 ug/L to 50 uq/L may be considered.Aities nay be forced to consider
disinfectants other than chlorine and ta consider treatment ratifications t.nat niqlit
include new actions ranging from improved conventional treatment to granular activated
carbon I-SAC) adsorpti on.~
EPA' Drinking Water Research Division has collected extensive treatment data for
removal of organics including TTHH, their precursors, Total Organic Carbon (TOC), and
Total Organic Halide (TOX) at several water utilities under actual ooeratinq condi-
tions. In these studies,6AC was used at some si tes^c4udinq_ Cincinnati, Ohio;
Jefferson Parish, Louisiana; Manchester, New Hamnshire; and Evansville, Indiana to
determine its ability for removfna those orqanic compounds present after conventional
treatment.
17.
KEY WORDS ANO DOCUMENT ANALYSIS
a. DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COS ATI Field/Group



IB. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (TliU Rtporxj
Unclassified
21. NO. OP PAGES
10
20 SECURITY CLASS iTiiapagt}
Unclassified
22 PRICE
EPA Form 2320-1 (R«*. 4-77) p«cviou» coition 14 OBSOLETE

-------
NOTICE
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.
ii

-------
PEKUVHL nF JC HAFEfllftL BY CDKrtJITICI.AL TJOTFENT AM DARB^
by
Benjamin W. Lykins, Jr.(a) and Robert H. Clark (h)
Disinfection by-products are among those compounds being considered
for regulation under the Safe Drinking Water Act Amendments of 1986. 0)
One of the most significant disinfection by-products for those utilities
that use chlorine are total trihalomethanes (TTHMs). Pressure is growing
to reconsider the existing TTHM Standard of 100 ug/L and to lower It to
as yet some unspecified level. Trihalomethane levels as low as 10 ug/L
to 50 ug/l may be considered. Utilities may be forced to consider dis-
infectants other than chlorine and to consider treatment modifications
that might include new options ranging from improved conventional treat-
ment to granular activated carbon (GAC) adsorption.
Some water utilities are able to meet a TTHM level of 0.10 mg/L
(lflO ug/L} by using properly operated conventional treatment. If,
however, the standard is reduced substantially, adding GAC to conventional
treatment may be the only acceptable treatment orption. The length of time
that GAC can remove THMs to meet a 10 ug/L, 25 ucj/L, 50 ug/L, or 100 ug/L
standard will determine its efficacy as a viable treatment option.
EPA's Drinking Water Research Division has collected extensive
treatment data for removal of organics Including TTHM, their precursors,
Total Organic Carbon (TOC), and Total Organic Halide (TOX) at several
water utilities under actual operating conditions. In these studies GAC
was used at some sites including Cincinnati, Ohio; Jefferson Parish,
Louisiana; Manchester, New Hampshire; and Evansvllle, Indiana to determine
its ability for removlnq those organic ccw?
-------
Literature
Conventional Treatment
TTHM precursors can be reduced by proper conventional treatment (coagula-
tion, flocculatlon, sedimentation, and filtration). The extent of reduction
can depend on several factors. The effects of pretreatment processes for
removal of humlc substances are site specific because of raw water quality
variables, treatment plant operating conditions, and treatment plant
dev\gT\.	The "\Vterature indicates that some diversity of findings
exist that make understanding the THM precursor removal process during
coagulation difficult.
Reckhow and Singer (4) reported that alum coagulation of aquatic fulvic
acid removed TOC and THM formation potential proportionately. Jodellah
and Weberobserved that high "levels of TOC removal may yield no selective
removal of THM precursors. Just as there were differences In the findings
of investigators during bench studies, water treatment plants also showed
varying removals for TOC and THM precursors. (®) Under slightly acidic
pH conditions, Edzwald and co-workers (2) reported that despite differences
in raw water quality similar TOC and THM precursor removals were achieved.
GAC Treatment
The specific coagulation process Influences the amount as well as
the THM reactivity of the residual organic inatter remaining after treat-
ment prior to chlorlriatlon.(7) Higher molecular weight organlcs were
rnost effecti^Elv rewcmeti during pretr-eatnEnt and I o^er Ba^ecular
wigit ^rg-anics "'ere effectively rEduced by GAC.f?*-] Jodalleh anc
MEber '.PJ Irdicatec that increased TOC reno^al t-y activated carbon trea--
frent -asLlt-ed dEtrecsed THH tarnation 3n trEated -water.
Proper pre-trsat^ert appears ta benefit activated carbon afi£or?t-!nr»
Randtke and Jepsen reported significant Increases In the adsorption
capacity of organlcs after alum coagulation, Lee and co-workers 0°)
showed that alum coagulation enhanced both carbon adsorption capacity and
the rate of uptake. Semmens and co-workers (UJ observed improved GAC
performance with greater levels of pre-treatment. Weber and Jodellah (3)
noted that alum coagulation Improved overall adsorbablHty of TOC,
Treatment at Research Locations
Various conventional treatment methods were used at the research sites
mentioned previously to remove or reduce the mix of compounds present in the
source water. The type of treatment (.conveatiaaaA and.	the^e
utilities is presented below.
Cincinnati, Ohio
Primary source water for the Cincinnati Water Works 1s the Ohio
River. To aid settling» 17 rog/L of aWox uas, added to tfce ran water.
2
-------
Prior to flocculatlon and clarification, 17 mg/L lime and ferric sulfate
(8,6 mg/L for high turbidity and 3.4 mg/L fop low turbidity) and chlorine
(plant effluent concentration I.ft mg/L free chlorine) were added. Post-
filtration adsorption was evaluated by deep-bed GAC contactors with an
ultimate EBCT of 15.2 minutes.
Jefferson Parish, Louisiana
The Mississippi River provides source water to the Jefferson Parish
treatm3nt plant. Potassium permanganate (0.5 - 1.0 mg/L) was added for
taste and odor control. A catlonlc polyelectrolyte (diallyldimethyl
diammonium chloride; 0.5 - 8.0 mg/L) was added as the primary coagulant
with lime (7 - 10 mg/L) fed for pH adjustment to 8.0 - 8.3. Chlorine
and ammonia (3:1 ratio) were added for chloramlne disinfection {1.4 -
1.7 mg/L residual after filtration). A sand filter was converted to a
post-filter GAC adsorber with about 20 minutes EBCT. In addition, four
G4C pilot columns were operated in series providing 11.6, 23.2, 34.7,
and 46.3 minutes EBCT.
Manchester, New Hampshire
The principal water source for the Manchester Water Works is Lake
Massabeslc. Alum and sodlim alumlnate were added for coagulation, pH
adjustment, and alkalinity control at dosage levels averaging about
12 mg/L and 8 mg/L, respectively. Chlorine was added prior to sand
filtration at an average dose of 1 mg/L. At the clearwell, chlorine
was again added in the range of 2 mg/L to 3 mg/L to produce an average
distribution free chlorine residual of 0.5 mg/L. A GAC filter normally
used for taste and odor control was used for post-filtration adsorption
with 23 minutes EBCT.
Evansvllle, Indiana
The Evansvllle Water Works uses Ohio River water as their source.
Chlorine and alum were added before primary settling with average con-
centrations of fi mg/L and 28 mg/L, respectively. A free chlorine residual
of 1.5 mg/L to 2.0 mg/L was maintained after sand filtration. Approximately
12 mg/L of lime was added after primary settling for pH control to 8.0. A
pilot plant operating parallel with the full-scale plant used chlorine
dioxide for disinfection. Average alum and polymer dosages of 12 mg/L
and 0.8 mg/L, respectively were added to the raw water of the pilot plant.
An average lime dose of about 6 mg/L was used for pH control to 8.0. Post
pilot plant GAC contactors had an EBCT of 9.6 minutes.
Results
TOC removal has been suggested as a means of measuring treatment
performance. Although TOC 1s relatively easy to analyze and incorporates
all organics, 1t does not relate to any specific regulatory requirements.
In the following evaluation, however, TOC as a general surrogate parameter
was used to determine the performance of conventional treatment and GAC
adsorption.
3
-------
Removal of instantaneous trlhalomethanes and their precursors to
meet a TTHM standard was also evaluated by using the terminal trlhalomethane
(term THM) parameter. Since the utilities studied used various disinfectants
that effected the trlhalomethane concentrations, term THM (Instantaneous
THM plus THM formation potential) allows a comparison among utilities
by indicating the maximum trlhalomethane 1n the distribution system at
a given time.
TOC and Terminal THM Removal Through Treatment
The TOC raw water concentration at Evansvllle, Indiana varied from
2.8	mg/L to 3.6 mg/L during one operational phase of 85 days duration.
Average raw water TOC concentration was 3.0 mg/L. Average TOC removal
through full-scale conventional treatment was 37 percent and 40 percent
for the pilot plant. Average sand filter TOC concentration was 1.9 mg/L
and 1.8 mg/L for the full-scale and pilot plant, respectively. Evansville's
three-day raw water term THM concentrations ranged from 95 ug/L to 178 ug/L
for an average of 140 ug/L. After conventional treatment the average
concentration was 82 ug/L for the full-scale plant for an average reduction
of 41 percent and 34 ug/L for the pilot plant for a 76 percent reduction.
More efficient THM precusor removal in the pilot plant for this operational
phase was attributed to the addition of a polymer for effective turbidity
removal.
GAC further reduced the TOC concentration until about thirty days
of operation before the GAC effluent tracked just below the filter
effluent (Figure 1). The 3-day term THM concentration of the GAC effluent
was essentially the same as the filter effluent after about 30 days of
operation (Figure 2).
The initia'j raw water TOC concentration at Manchester was 4.6 mg/L
and varied from 3.8 mg/L to 4.8 mg/L with an average concentration of
4.5 mg/L for 130 days of operation. The raw water TOC concentration was
reduced about 47 percent to an average of 2.4 mg/L. Three-day term THM
concentrations for Manchester's raw water at ambient temperature ranged
from 104 ug/L to 191 ug/L (average of 151 ug/L). Precursor removal
through conventional treatment reduced the three-day term THMs to an
an average of 70 ug/L resulting in a 54 percent reduction.
At Manchester, the TOC concentration of the GAC effluent was about
0.5 mg/L at the start of one evaluation and increased in concentration
until about 35 days of operation before tracking just below the sand
filter effluent (Figure 3). The 3-day term THM concentration appeared to
follow the GAC effluent TOC with an Initial concentration of about 10 ug/L
increasing to about 45 ug/L after 40 days of operation and then tracking
below the sand filter effluent (Figure 4).
In Cincinnati, where ferric sulfate was used as the primary coagulant,
a reduction in the TOC concentration of 41 percent was seen through con-
ventional treatment. Raw water TOC concentrations ranged from 1.9 mg/L to
5.9	mg/L for an average 3.4 mg/L. After conventional treatment the TOC
4
-------
E 2-
u* ^
EC
1-
0-
r *0	fr
*Ek_
¦v
'^«V
/
>~~


J**

RAW WATER
SANO FILTER EFF.
QAC EFF.
0
-i—i—i—i—i—i—i—i—i—i—i—i—f—i—i—j—i—i—r
10 20 30 40 50 BO 70 80 90 rundays
2,894	5,733	3,582	11,576	BED VOLUMES
FIGURE 1. TOC REMOVAL IN PILOT PLANT - EVANSVILLE, IN
-------
180
170
160
150
140
130
120
110
100
90
80
70
60
50
40
30
20
10
0
i—r
3
nr
10
-V
/
.6
I
20
2,894
RAW WATER
SANO FILTER EFF.
CAC EFF.
t—i—i—i—i—i—i—r—1—i—'—i—•—i—i—r
30 40 50 60 70 80 90 ™Joays
5,788	8,682	11,576	BED VOLUMES
FIGURE 2. TERMINAL THM REMOVAL IN PILOT PLANT - EVANSVILLE, IN
-------
V
RAW WATER
SANO FILTER EFF.
GAC EFF.
n—»—!—*—r
10 20 30
1,612
—I	'	1	r
40 50
3,224
¦ i • i • i » i ' i " i 1 r
60 70 80 90 100 110 120 130rummys
4,836	6,448	8,090	9,672 BEO VOLUMES
FIGURE 3. TOC REMOVAL BY CONVENTIONAL TREATMENT
AND GAC ADSORPTION - MANCHESTER, NH
-------
200
190
180
170
160
150
140
130
120
110
100
90
80
70
60
50
40
30
20
10
0
0
V !r^A\
4
i—»—r
fUW WATER
SAHO FILTER EFF.
OAC EFF.

T
t—»—r
i—'—r
T
T
10 20 30 40 50 60 70 BO 90 100 110 120
1»6L2
3,224
4,335
6 ,443
a,090
<>,672
RUNDArS
BED VOLUMES
FIGURE 4. TERMINAL THM AFTER CONVENTIONAL TREATMENT
AND GAC ADSORPTION - MANCHESTER, NH
-------
concentrations ranged from 1.1 mg/L to 3.4 mg/L for an average 2.0
The average percent reduction through conventional treatment was 41. Three-
day term THM concentrations for the raw water ranged from 64 ug/L to 211
ug/L for an average of 146 u^/L. After cOTWfcr.Vlona^ treat-merit the term
THM concentrations ranged from 39 ug/L to 181 ug/L for an average of
89 ug/L, producing an average term THM reduction of 39 percent.
The TOC effluent GAC concentration at Cincinnati was about 0.2 mg/L
at the start of one of the runs and increased to about 1.1 mg/L after
approximately 100 days of operation. As with Manchester and Evansville,
the TOC then tracked just below the sand filter effluent (Figure 5).
The 3-day term THM concentration of the GAC effluent was about 3 ug/L
at the start of an adsorption study and "breakthrough" occurred after
about 50 days of operation. From about day 110, the 3-day term THM
effluent was approximately the same Increment below the sand filter
effluent throughout the 320 day study (Figure 6).
At Jefferson Parish, polymers were used as the primary coagulant. The
raw water (Mississippi River) TOC concentration ranged from 2.9 mg/L to
5.9 mg/L with an average of 4.0 mg/L. After conventional treatment the TOC
concentrations ranged from 2.3 mg/L to 3.8 mg/L with an average of 2.9 mg/L
for a reduction of 27.5 percent. Five-day term THM concentrations for the
raw water ranged from 133 ug/L to 511 ug/L with an average of 281 ug/L.
After conventional treatment the range was 82 ug/L to 364 ug/L for an
average of 175 ug/L. Average 5-day term THM reduction through conventional
treatment was 37.7 percent.
The full-scale GAC adsorber at Jefferson Parish seemed to remove
the TOC concentration steadily for about 160 days. Initial concentration
was 0.2 mg/L Increasing to about 2.0 mg/L (Figure 7). The 5-day terminal
THM GAC effluent concentration for the full-scale system at Jefferson
Parish was about 15 ug/L at the start of one run and, like the TOC*
steadily increased for 140 days (Figure 8).
Effect of EBCT
In the pilot system GAC contactors at Jefferson Parish, 11.6, 23.2,
34.7, and 46.3 minutes EBCT were evaluated using the same source water that
was used for the full-scale system. TOC concentrations "broke-through"
after approximately 30, 50, 80, and 110 days of operation for the 11.6,
23.2, 34.7 and 46.3 minute EBCT contactors, respectively (Figure 9).
The initial term THM GAC effluent concentration at Jefferson Parish
was t5 ug/L and "breakthrough" was noted at about 40, 70, 90, and 110 days
for 11.6, 23.2, 34.7, and 46.3 minutes E8CT, respectively (Figure 10).
GAC for THM Control
Some water utilities are able to maintain their THM concentrations
bel:w the existing promulgated standard of 0.10 mg/L (100 ug/L) by proper
con\entional treatment. If, howevert the standard Is reduced substantially,
9
-------
6-
- 5
° c

z
uj
o
o 2
o
o
o
H-
1-
•&-D—D RAW WATER
SANC FILTER EFF.
OAC EFF.
\
I
I
r
: \
: \
\
/'; g
t! t ftsa •
P8! / • j\f *
u t*. \r
\S
baB0°A
T-T
¦ I 1
100
3,590
t—r
t—r
' I 1
200
17,180
I 1 I I
• I « '
300
25,770
FIGURE 5. TOC REMOVAL BY CONVENTIONAL TREATMENT
AND GAC ADSORPTION - CIMCINHATI, OH
i i i i
RUKDAYS
BED VOLUMES
-------
0	100	200	300	RUNDAYS
3,590	17,180	25,770	BED VOLUMES
FIGURE 6. TERMINAL THM AFTER CONVENTIONAL TREATMENT
AND GAC ADSORPTION - CINCINNATI, OH
-------
6
ro
: n
\
oi
£
J*
E
H
U1
O
z
o
o
o
o
H
/ \ BL^i\
J
O-Q-B RAW WATER
»	SAND FILTER EFF.
GAC EFF.
/I
f
l!
/s
t i
/Htyvvv] j j
AIhwi ii
r* *\ y	B-a/jMv. r •K
T
20
1,560
T
40
3,120
60
4,630
80
6,240
T
100
7tS00
120 140
0,360 10,120
T
160
12,480
180 RUNDAYS
14,040 BED VOLUMES
FIGURE 7. TOC REMOVAL BY CONVENTIONAL TREATMENT
AND GAC ADSORPTION - JEFFERSON PARISH, LA
-------
o>
a
w
5
DC
Ul
H
5
Q
I
IO
200: ji
:a-o
3,120
4,680
6,240
T	r | i r
140 160 180 runoays
10,920 12,480 14,040 BED VOLUMES
FIGURE 8. FIVE-DAY TERMINAL THM AFTER CONVENTIONAL TREATMENT
AND GAC ADSORPTION - JEFFERSON PARISH, LA
-------
AVG. SANO FILTER * 2.9 irg/L
116 MM. EBCT
O-&-0 n.2 HN. EBCT
~ 34.7 MN. EBCT
W*. EBCT
I
k1

T
80
-i	1	1
100
RUNDAYS
—i—i—|—i—i—i—i—i—r
120 140 160 180 200
T
0
20
T"
40
T
60
FIGURE 9. TOC REMOVAL THROUGH SERIES GAC CONTACTORS -
JEFFERSON PARISH, LA
-------
300
a 200
s
X
¥-
-J
<
z
- i
tn
iu
K
5
Q
I
to
100
AVG. SAND FILTER = 174.8 ug/L
It 6 MIN. EBCT
23 2 MIN EBCT
34.7 MIN. EBCT
46.3 MIN. EBCT
t—i—i—r
100 120
RUNDAYS
1	'	1	«	T
160 180 200
FIGURE 10. FIVE-DAY TERMINAL THM FOR GAC SERIES CONTACTORS -
JEFFERSON PARISH, LA
-------
other treatment alternatives will be required. GAC may be an alternative
worth evaluating. The length of time that GAC can remove tr1halomethanes
to meet a standard of 10 ug/L, 25 ug/L, 50 ug/L, or 100 ug/L will determine
Its efficacy as a viable treatment option.
Since terminal THM values can simulate concentrations 1n the distribution
system, one can estimate the length of GAC operation for meeting THM goals.
Table 1 gives an indication of how long GAC can removal various concentrations
of THMs.
TABLE 1. Length of GAC Operation Before Exceeding Term THM Levels
THM Goals
10 ug/L	25 ug/L	50 ug/L	100 ug/L
Inf,	Inf,	Inf,	Inf,
Location	Day ug/L Day ug/L Day ug/L Day ug/L
Cincinnati, OH	50	75 155 45 208 70 280 150
(3-day term, 15.2
min. EBCT)
Jefferson, Parish, LA	20 80 63 170 103 220
(5-day term, 18.8 min.
EBCT)
Manchester, NH	2	73 16 70 98 65 -
(3-day term, 23 min.
EBCT)
Evansville, IN	-	6 96 56 53
(3-day term, 9.6
min. EBCT)
As can be seen from Table 1, establishing a trihalomethane standard of
10 ug/L will probably negate the use of GAC. Using GAC to meet a ?5 ug/L
standard also may not be feasible. However, at the 50 ug/L trihalomethane
concentration, GAC may be more attractive.
Data Normalization and Predicting THM Concentrations
The above data has shown the performance of GAC over various days of
operation and bed volumes through GAC adsorbers at different locations for
removing term THMs. NormalIzatlon of the data using percent removal shows
that the deep-bed GAC adsorbers used at Cincinnati produced the overall
highest removal rate for terminal THM {Figure 11). Conversely, Evansville
had the lowest percent removal. This may be due, 1n part, because Cincinnati
used a coal-based carbon and Evansville a lignite carbon.
16
-------
CINCINNATI
0-O--0 evansvillE
JEFFERSON PARISH
4 MANCHESTER
400
RUNOAYS
FIGURE 11. TERMINAL THM PERCENT REMOVAL FOR GAC EFFLUENT
-------
TOO has been suggested as a surrogate for predicting THM concentrations.
By removing TOC through GAC adsorption, will THM precursors be selectively
removed? A definite p. tern of TOC and 3-day THM formation potential and
TOC with 7-day THM formation potential was noted 1n Cincinnati for the GAC
effluent, indicating that TOC may be used as a predictive tool at that
location (Figures 12 and 13).
With Jefferson Parish, (another plant using river water as their
source} TOC and 5-day THM formation potential also seemed to follow a
pattern (Figure 14). The correlation for Manchester GAC effluent (lake
water source) with TOC and 3-day term THM Is shown in Figure 15.
Removal of TOC by GAC may give an Indication of trihalomethane forma-
tion potential (THMFP) removal. By comparing percent THMFP removal to
TOC percent removal for the four utilities evaluated, Figure 16 shows a
45 degree line of equal percent removal. Regression of this data, however,
indicates that removing TOC does not necessarily mean removing an equal
percentage of THMFP. The following equation describes the data from the
four uti1ities:
THMFP = 7.83 + 0.87 TOC	(R2 » .73)
The instantaneous organic halide did not seem to be as good	a predictor
of THM formation potential as TOC (Figures 17 and 18). However,	one
might be able to use the Instantaneous organic halide to predict	GAC
instantaneous THM breakthrough (Figure 19).
Summa ry
Proper conventional treatment can reduce TOC and THM precursors
substantially. For the full-scale systems, average percent TOC removal
was variable, possibly because of the type of coagulant used. Average
percent term THM removal however, seemed to follow a pattern. The river
water sources were about the same with lower average term THM percent
removal through conventional treatment than the lake water at Manchester.
This may be attributed to better TOC removal during conventional treatment
and different TW precursors In the lake water than 1n the river source
water.
Although THM precursors are reduced during conventional treatment,
this reduction will probably not be enough to meet THM concentrations
much below 100 ug/L where chlorine 1s used as the primary disinfectant.
With GAC adsorption, however, additional precursors are removed. For
the utilities evaluated, meeting a SO ug/L THM standard appears to be
possible after GAC treatment.
This paper has been reviewed in accordance with the U.S. Environmental
Protection Agency's peer and administrative review policies and approved
for presentation and publication. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use by
the U.S. EPA.
18
-------
t-2
3-DAY THMFP
TOC
'| | | | | l | | I | | l | i I I I I I I | I » 1 I I I 1 I I ) I » I I I I 1 I I |
o
o
o
o
z
o
m
31
>
bO
100
fl,590
200
17,180
300
25,770
400 rundays
BED VOLUMES
FIGURE 12. COMPARISON OF GAC EFFLUENT 3-DAY THM FORMATION
POTENTIAL AND TOC - CINCINNATI, OH
-------
'l I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 1 I I I (
0	tOO	200	300	runoays
8,590	17,180	25,770	BED VOLUMES
FIGURE 13. COMPARISON OF GAC EFFLUENT 7-DAY THM FORMATION
POTENTIAL ANO TOG - CINCINNATI. OH
-------
Lit
H
O
CL
Z
o
N «-
M 0c
o
u.
5
0
1
10
1,560 3,120 4,680 6,240
7,800
5,260 10,920
T	r
1B0 Rundays
14,040 BEO volumes
FIGURE 14. COMPARISON OF GAC EFFLUENT 5-DAY TERMINAL THM FORMATION
POTENTIAL AND TOC - JEFFERSON PARISH, LA
-------
N
ro
UJ
H
o
a.
5
a
4
n
80
50-
40^
30
$
I ,0-
10
1-3
J—DAY 7HMFP
TOC
0 10 20 30 40 50 60 70
o
o
o
o
z
o
m
z
H
3)
5
o
z
1 3
1.612
3.224
4,836
I • I » I » 1 1 !
BO 90 100 110 120 130 rummys
6,448	8,090	9,672	BED VOLUMES
FIGURE 15. COMPARISON OF GAC EFFLUENT 3-DAY THM FORMATION
POTENTIAL ANO TOC - MANCHESTER, NH
-------
100
90
80
70
60
50
40
30
20
10
0
TOC PERCENT REMOVAL
FIGURE 16. THMFP VERSUS TOC PERCENT REMOVAL FOR GAC EFFLUENT
-------
50 -4
4 40-
ui
i-
o
Q.
30-

£C
O
20:
10-
<
Q
I
CO
0-
3—CAYTHMFP
0-£>-£} INST TOX
T—I—i—i i i "I i
—|—i i I l I I I i I | I l i i i
100	200
-80
»
-70
-60
-50
»
•40
»
-30
»
-20
I
-10
- 0
' ' lr
BED VOLUMES
o
X
o
o
z
o
m
z
H
31
a
o
z
FIGURE 17. COMPARISON OF GAC EFFLUENT 3-DAY THM FORMATION
POTENTIAL AND TOX - CINCINNATI, OH
-------
<
p
z
III
5
OL
r\> Z
« o
s
s
oc
o
it
X
H
O
I
100
90-
80-
70j
60-
«
50-
40-
30
20
10H
0
7-OAYTHHFP
INST T09C
*80
-70
60
T
0
i i i—i—i—t—i—i—i—|—i—i—i—i—i—i—i—i—i—|—i—i—i—*—i—i—i—i—i—f
b50 o
m
40
50
H
g
O
o
-t
a
>
to
20 r=
to
1- 0
100
8,590
200
17,180
300 Runoays
BED VOLUMES
Figure is. comparison of gac effluent 7-day thm formation
POTENTIAL AND TOX - CINCINNATI, OH
-------
s
E
nCO
O
III
z
<
0>
80-
«
70-
60-
«
50-
4
40-
30-
5 20-
10-
INST THM
G-Q-0 IHST TOJC
(-80
-70
60
i
I-50 o
o
0" ~ ~ ~ ~
J—I—I—I—I—I—I—I—I—I—I—
0	100
8.590
40
30
•20
»
-10
»
- 0
o
m
5
5
t—i—i—i—i—i—i—i—|—i—i—i—i—i—i—i—i—r—f
200	300 RUNDAYS
17.1 bo	BEO VOLUMES
FIGURE 19. COMPARISON OF GAC EFFLUENT INSTANTANEOUS THM AND TOX -
CINCINNATI, OH
-------
REFERENCES
1.	The Safe Drinking Water Act as Amended by the Safe Drinking Water
Act Amendments of 1986, Public Law 99-339, June 19, 1986.
2.	Edzwald, J. K., Becker, W. C., and Wattler, K. L., "Surrogate Para-
meters for Monitoring Organic Matter and THM Precursors," JAWWA,
Vol. 77, No. 4, 122-132, April 1985.
3.	Weber, W. J. and Jodellah, A. H., "Removing Humlc Substances by
Chemical Treatment and Adsorption," JAWWA, Vol. 77, No. 4, 132-
137, April 1985.
4.	Reckhow, D. A. and Singer, P. C«, "Removal of Organic Hallde Pre-
cursors by Pre-Ozonatlon and Alum Coagulation," Proceedings of AWWA
1983 Annual Conference, Las Vegas, Nevada, June 1983.
5.	Jodellah, A. M. and Weber, W. J., "Controlling Trlhalomethane Forma-
tion Potential by Chemical Treatment and Adsorption," JAWWA, Vol. 77,
No. 10, 95-100, October 1985.
6.	Ohio River Valley Water Sanitation Commission," Water Treatment Process
Modification for Trlhalomethane Control and Organic Substances In the
Ohio River," EPA-600/2-80-028, USEPA, March 1980.
7.	Collins, M. R.t Amy, G. L., and King, P. H., "Removal of Organic Hatter
1n Water Treatment," Journal of Envlronment&l Engineering, ASCE, Vol.
Ill, No. 6, December 1985.
8.	Semmens, M. J. and Staples, A. 8., "The Nature of Organlcs Removed
During Treatment of Mississippi River Water," JAUWA, Vol. 78, No. 2,
76-81, February 1986.
9.	Randtke, S. J. and Jepsen, C. P., "Chemical Pretreatment for Activated
Carbon Adsorption" JAWWA, Vol. 73, No. 8, 411-419, August 1981.
10.	Lee, M.C., Snoeylnk, V. L.» and Crittenden, J. C., "Activated Carbon
Adsorption of Himlc Substances, JAWWA, Vol. 73, No. 8, 440-446, August
1981.
11.	Semmens, M. J., Staples, A. B., Kohensteln, 6., and Norgaard, G. E.,
"Influence of Coagulation on Removal of Organlcs by Granular Activated
Carbon," JAUWA, Vol. 78, No. 8, 80-84, August 1986.
27
-------