United States
Environmental Protection
Agency
Water Engineering Research
Laboratory
'Cincinnati OH 45268
Research and Development
EPA/600/S2-85/006 Mar. 1985
Project Summary
Barium and Radium in Water
Treatment Plant Wastes
Vernon L. Snoeyink, Candy K. Jongeward, Anthony G. Myers, and Sharon K.
Richter
Water Treatment plants at nine loca-
tions (10 plants) in Illinois and Iowa
were studied to determine the charac-
teristics and disposal practices for the
sludge, brine, and backwash water
containing radium (Ra) and/or barium
(Ba). The treatment processes in these
10 plants include iron (Fe) and manga-
nese (Mn) removal (3 plants), lime
softening (4 plants), and ion exchange
(3 plants). In the 10 plants, eight had
concentrations of radium in their water
treatment plants wastes and three had
barium. The data are needed to deter-
mine whether special procedures are
required to dispose of such wastes.
For the eight plants having radium in
their wastes, the influent Ra226 con-
centrations ranged from 0.3 to 49
pCi/L. Radium removal averaged 8%,
75%, and 85% for Fe and Mn, lime
softening, and ion exchange, respec-
tively. Theoretically, the amount of
radium removed estimates the quantity
of radium in the waste. The data showed
that backwash water from Fe and Mn
removal plants contained average Ra22*
concentrations that ranged from 21.2
to 106 pCi/L, and average Ra228 con-
centrations ranged from 5.7 to 20
pCi/L. Lime softening sludge, on a dry
weight basis, ranged from <1.2 to 21.6
pCiRa««/g and from <2.4 to 11.7
pCiRa228/g. Ion exchange plant brine
contained peak Ra221 concentrations of
217and1,144pCi/L.
Influent barium concentrations for
the three plants studied ranged from
4.0 to 16.1 mg/L. Lime softening and
ion exchange removed barium to con-
centrations that were below the max-
imum contaminant level (MCL) of 1.0
mg/L. Barium removal during lime
softening is pH dependent. Theoret-
ically, the amount of barium removed
estimates the quantity of barium in the
waste. Barium concentrations in the
brine and rinsewater from two ion
exchange plants that were tested aver-
aged 328 and 1,297 mg/L. Peak con-
centrations of barium from the two
brines were 1,197 mg/L and 5,161
mg/L.
Disposal processes used, for the
plants that were studied, were lagoon-
ing, discharge to sanitary sewers, and
discharge to a water course.
This Project Summary was developed
by EPA's Water Engineering Research
Laboratory. Cincinnati, OH. to announce
key findings of the research project that
is fully documented in a separate report
of the sante title (see Project Report
ordering information at back).
Introduction
Naturally occurring barium (Ba) in
drinking water exceeds the MCL of 1
mg/L in some areas of northern Illinois
and northeastern Iowa. In these same
areas and in some parts of Florida, the
concentrations of radium226 (Ra226) and
radium22B (Ra228) exceed the MCL, which
is 5 pico-Curies (pCi)/L Most of the
contaminated supplies are used by small
communities, many of which do not
presently treat their water to reduce the
concentrations of these substances. Both
Ra and Ba are alkaline earth metals, and
both are found in water as divalent
cations. Their chemical behavior is very
similar, and it is much like that of calcium
(Caa+) and magnesium (Mg2*), the princ-
ipal components of hardness in water.
The MCL's for Ba and Ra were developed
to minimize the attack of these elements
on human bones and the replacement of
-------
the Ca in bones. Their similarity to Ca,
however, means that processes used to
soften water are also very useful for
removing these contaminants from drink-
ing water.
Water treatment in small communities
is often accomplished with iron and
manganese (Fe and Mn) removal plants,
lime softening plants, and ion exchange
plants. The objective of this research was
to characterize the backwash samples
from all three types of plants, as well as
the sludges from the lime softening
plants and the brines from ion exchange
plants. The concentration data were
needed to determine whether special
procedures are required to dispose of
such wastes. The Ba and Ra removal
efficiencies of the plants were also de-
termined.
Materials and Methods
Samples were collected before and
after the processes that were likely to
cause a change in Ra or Ba concentration.
Total alkalinity, total hardness, and Ca
concentrations were determined accord-
ing to Standard Methods for the Examina-
tion of Water and Wastewater (15th ed.,
American Public Health Association, Wash-
ington, D.C., 1980). Total dissolved solids
were measured using a conductivity
meter, and Ba was measured with an
atomic absorption unit. Chloride was
measured using an ion analyzer and an
ion chromatograph. The sludge and back-
wash waters were analyzed for Ra226 and
Ra228 at Argonne National Laboratory by
monitoring gamma radiation; the other
samples were analyzed by the University
of Iowa Hygienic Laboratory by a tech-
nique involving co-precipitation of the Ra
with BaS04 and by alpha counting the
precipitate with an internal proportional
counter.
Radium Removal in Fe and Mn
Removal Plants
Fe and Mn removal plants often treat
their waters by aeration, detention, and
filtration. Ra removal during aeration and
detention has been observed, perhaps as
a result of Ra adsorption on the Mn oxides
or Fe hydroxides. The hydrous oxides of
Fe (III) and Mn (IV) have high sorption
capacities for bivalent metal ions. Re-
moval of Ra by sand, coal, or greensand
media filters is likely attributable to
removal of oxides containing adsorbed
Ra, or to accumulation of these oxides in
the filter followed by sorption of Ra on
their surfaces.
Our data, combined with similar data of
others, show removal efficiencies that
range from a negative value (obtained
from a sample taken at the end of a filter
run) to more than 50%; but values in the
range of 5% to 15% are most common.
Moore and co-workers reported
Mn02 was removed from
the filter. Sufficient data are not available
from this study to show the expected
relationship, however.
Ra226 in the backwash water was found
to be 21 pCi/L at Adair, Iowa, and 106
pCi/L at Stuart, Iowa. The average Ra
removal indicated by backwash concen-
tration was 11%, which was comparable
with the removal figure obtained by
measurement of influent and effluent
concentrations. Total solids levels in the
backwash waters were approximately
2,000 mg/L. These solids could be settled
in a lagoon, and the supernatant could be
recycled to the plant.
Radium Removal in Lime
Softening Plants
The sludges from the lime softening
plants samples in this study were dis-
charged to a sanitary sewer (Bushnell,
Illinois) or to lagoons (Elgin and Colchest-
er, Illinois, and West Des Moines, Iowa).
Sanitary sewers have been noted as
being somewhat unfavorable, since they
simply transfer ahy sludge contaminants
from the water treatment plant to the
sludge of the wastewater treatment plant.
Radioactive contaminants in lagooned
sludges must be considered when choos-
ing methods for reclaiming the land from
permanent lagoon sites. For instance,
home building on such sites is not
recommended because of the potential
hazards from the radon gas.
The softening process may yield Ba and
Ra removal by direct precipitation, co-
precipitation, or adsorption. Other ionic
compounds present in the water and the
pH may influence the mechanism andthe
extent of removal. Removal efficiencies
(Table 1) ranged from 43% to 96%, with
an average of 75%.
The relationship between hardness
and Ra has also been examined and is
expressed in the following equation:
y = 0.4566-^+0.2275
where y = the total hardness removal
fraction, and x = the Ra226 removal
fraction. This equation describes the data
reasonably well, as shown in Figure 1.
A combination of the data collected in
this study with data from the literature
shows Ra226 concentrations in the sludge
ranging from 1,000 to 11,000 pCi/L of
sludge. The concentration per dry gram of
solids was 10 to 20 pCi. Backwash water
concentrations ranged from 6 to 50 pCi/L.
Although the relative quantities of Ra in
backwash and sludge was not deter-
mined, most of the Ra appears in the
sludge. A good estimate of the Ra in the
sludge can be calculated by multiplying
the difference in concentration between
influent and effluent by the volume of
flow through the plant.
Radium Removal in Ion
Exchange Plants
Two ion exchange columns are used in
the water treatment plant at Eldon, Iowa.
Both columns are usually regenerated
with a 100%-saturated (26-37%) by
weight) solution of NaCI, but a shortage of
available salt resulted in the decision to
regenerate one column with a 40%-satur-
ated (10-15% by weight) NaCI solution.
Eldon's spent brine is discharged into a
storm sewer, which then discharges to a
river.
The Ra226 removal obtained by ion
exchange at Eldon was 60%. However,
the regeneration was performed with a
limited amount of 40%-saturated salt
solution and the regeneration was in-
complete. This removal can be compared
with 65% to 85% removals found for
other plants that were incompletely re-
generating their exchange media. The
literature indicates that 81% to 97% Ra
removal can be achieved in well-operated
plants, and that Ra removal is directly
related to hardness removal (see Table 2).
Laboratory studies cited in the literature
using both strong and weak acid resins
have indicated that good removal oc-
curred long after hardness breakthrough
in both types of resin, although operation
through several service cycles with in-
complete regeneration may lead to earlier
Ra breakthrough.
An earlier study of seven water treat-
ment plants reported that maximum Ra226
concentrations in softener brine and
rinse effluent ranged from 320 to 500
pCi/L. Iowa ion exchange wastes were
found to contain 7.8 to 98 pCi/L of Ra226
in the backwash water, and rinse waste-
waters ranged from 114to1,960pCi/Lof
Ra226. Less than 5% of the Ra was
-------
Table 1. Ra and Hardness Removal Efficiencies in Lime Softening Plants
Location and Date
Ra2" in
Raw Water
(pd/LI
Ra221
Removed (%)
Total
Hardness
Raw Water
fmg/L as CaCOil
Total
Hardness
Removed {%)
IV. Des Moines, Iowa:
8/1/74"
6/5/78t
8/2/83\
Webster City, Iowa:
8/13/74"
2/20/75*
7/26/78t
Peru, Illinois:
2/20/75*
2/25/75*
3/4/75*
Elgin, Illinois:
3/7/75*
3/74/75*
3/27/75*
6/27/83t
Colchester, Illinois:
Bushnell, Illinois:
8/83t
9.3
1.9
6.9
6.1
7.8
1-4
6.5
5.5
5.5
7.5
5.7
3.5
0.3
12.1
12.6
75
43
78
85
96
60
92
70
76
90
86
80
74
45
376
NA
389
507
482
NA
329
278
286
246
243
242
253
698
354
49
NA
61
48
78
NA
47
35
57
60
54
61
67
57
42
•Data from EPA-600/2-77-073.
tOafa fromJAWWA, 61:541, 619. and681.
tData from this study.
removed during backwash. Wastewater
volumes generated were 2% to 10% of
the product water. Wastewater charac-
teristics varied greatly from one plant to
another.
Barium Removal in Lime
Softening Plants
Barium removal is pH dependent, with
an optimum pH occurring in the 9 to 11
range. At Elgin, 90% of the influent
barium (4 mg/L) was removed at pH 9.3.
Barium Removal in Ion
Exchange Treatment Plants
Crystal Lake, Illinois, has an ion ex-
change system composed of three sepa-
rate wells with two strong acid ion
exchange columns for each. The spent
brine is discharged to a sanitary sewer
system, which eventually discharges to
the wastewater treatment plant. Data
indicate that ion exchange plants typically
remove 92% to 99% of the incoming
barium. Laboratory studies have shown
that hardness and Ba will break through
at the same time for a strong acid resin
after operation through several exhaus-
tion-regeneration cycles.
The regeneration of Wells 6 and 8
removed 84% and 153% of the exchanged
Ba, respectively. At Well 6, 6.5 kg of Ba
was contained in the 9,500 gal of waste-
water generated from one regeneration
cycle. Likewise, Well 8 produced 30 kg Ba
in its 10,000 gal of wastewater. Less than
2% of the barium was found in the
backwash water, and the remainder was
in the spent brine and rinse water. Other
researchers report 85% removal or 18 kg
Ba in 9,250 gal of wastewater.
Peak hardness (33,000 mg/L) and Ba
(1,200 mg/L) concentrations in the spent
regenerant coincide at Well 6 (Figure 2),
but the peak hardness (44,000 mg/L)
occurs before the peak Ba (5,000 mg/L)
concentration at Well 8. This phenome-
non may be caused by the greater amount
of Ba on the column before generation at
Well 8 (78.4 g Ba/ft3) compared with that
at Well 6 (40.4 g Ba/ft3). More Ba
accumulates on the resin during the
service cycle at Well 8 because the raw
water Ba concentration is higher and
because less regenerant is applied to the
column per unit volume of water produced
during the service cycle.
Conclusions
Ion exchange plants, lime softening
plants, and Fe and Mn removal plants can
remove Ra from water, with varying
degrees of success. The Fe and Mn
treatment plants removed 0% to 54% of
the influent Ra, which ranged in con-
centration from 6.0 to 49 pCi/L. The
lower removals were more common. The
average removal, including values taken
from the literature, was 23%, but the
mean for this study was only 8%. The total
Ra content of the waters from a single
backwash was 0.02 to 7.6 //Ci, and
concentrations were typically less than
100 pCi/L. The average Ra removal by
the plant was 11 %. Though a relationship
between MnOa accumulation on the
filters and Ra removal was expected,
sufficient data were not available to show
this relationship. Treatment methods
such as lime softening or ion exchange
seem better suited than Fe and Mn
removal plants for removing Ra from
water supplies.
Lime softening effectively removes Ra
from waters. The influent concentrations
were 0.3 to 24.2 pCi/L. Typical removal
values ranged from 43% to 92%, with an
average of 69% if data from the literature
are included. The Ra removal at the plants
sampled during the summer varied from
45% to 78%, with an average of 75%. Two
correlations presented in the report relate
hardness removal to Ra removal and can
be used to predict Ra removal efficiencies.
The Ra concentrations in the softening
sludges ranged from <1.2 to 21.6 pCi/g
dry solids for Ra226 and from <2.4 to 11.7
pCi/g dry solids for Ra228.
Ion exchange plants also produce an
effluent with low Ra concentrations. The
-------
plant at Eldon removed 60% of the
influent Ra of 47 pCi/L, but sufficient salt
was not applied to completely regenerate
the resin. Common removals reported in
the literature are 65% to 90%, and the
average approaches 85%. Ra concentra-
tions in the brines at Eldon reached 217
and 1,144 pCi/L for NaCI brines at 40%
and 100% saturation with NaCI, respec-
tively.
Ion exchange and lime softening plants
were also analyzed for Ba removal. Ion
exchange at Crystal Lake removed more
than 90% of the influent 9.5 and 16.1
mg/L Ba. Barium concentrations in the
brine at Wells 6 and 8 averaged 328 and
1,300 mg/L, with peak values of 1,200
and 5,200 mg/L, respectively. Lime soft-
ening at Elgin also removed 90% of the 4
mg/L Ba in the influent water.
The full report was submitted in ful-
fillment of Cooperative Agreement No.
CR-808912 by the University of Illinois at
Urbana-Champaign under the sponsor-
ship of the U.S. Environmental Protection
Agency.
o
0.30
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
O West Des Moines. IA (data from this study)
A West Des Moines. IA
D Colchester, IL (data from this study)
V Bushnell, IL (data from this study)
• Webster City. IA
A Peru, IL
Elgin, IL
x Venice, FL
— + Englewood, FL
Curve 1 y = 0.4566 y/x +0.2275
Correlation Coefficient = 0.612
Curve2 y = 0.8977 y/x-0.0376
(Ignore Bushnell and Florida Points)
Correlation Coefficient = 0.875
0.40 0.50 0.60 0.70 0.50 0.90
Total Hardness Removal Fraction y
x
1.00
1.10
Total Radium Removal Fraction
Figure 1. Proposed correlation between fla286 and total hardness removal fractions for lime
softening plants.
-------
Table 2. Hardness and Ra22e Removals in Ion Exchange Plants
Location and Time
of Removal
Ra2" in
Influent
Ipd/L)
Ra226 in
Effluent
(pCi/L)
Ra22S
Removed (%>
Total Hardness
in flaw Water
(mg/L as CaCOd
Total
Hardness in
Product Water
(mg/L as CaCOai
Total
Hardness
Removed
(%)
Hersher, Illinois:*
3/25/75:
Just after regeneration 6.64 1.25
Midpoint 6.94 0.42
Near breakthrough 6.88 2.07
D wight Correctional
Center, Illinois:*
2/13/75-2/14/75:
Just after regeneration 3 0.4
Midpoint 3 0
Near breakthrough 3
Eldon. /ovva/t
8/83:
Just after regeneration} NA 14.2
Midpoint 42.4 NA
Near breakthrough NA 20.1
81.2
93.9
69.9
88.0
92.5
70.7
66.5
NA
52.6
412
427
417
286
284
279
NA
350
NA
18
12
184
16.0
4.1
131.0
175
NA
232
95.6
97.2
55.8
94.3
98.6
53.7
50
NA
33.7
"Data from EPA-600/2-77-073.
]Data from this study.
\4O% saturated brine.
Backwash
Brine
Rinse
10
20
60
70
Figure 2.
30 40 50
Regeneration Time (min)
Barium and hardness concentrations in spent brine at Crystal Lake, IL Well #6.
-------
-------
-------
Vernon L, Snoeyink. Candy K. Jongeward, Anthony G. Myers, and Sharon K.
Richter are with the University of Illinois, Urbana, IL61801.
Richard P. Lauch is the EPA Project Officer (see below).
The complete report, entitled "Barium and Radium in Water Treatment Plant
Wastes,"(Order No. PB 85-165 777'/AS; Cost: $ 10.00, 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:
Water Engineering Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
*• U.S. GOVERNMENT PRINTING OFFICE: 1985-559-016/27052
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
BULK RATE
POSTAGE 8. FEES
EPA
PERMIT No G-3
Official Business
Penalty for Private Use $300
0169064 WERL
US EPA REGION '
LIBRARY
230 S DEARBORN
CHICAGO
ST.
IL
60604
------- |