United States
Environmental Protection
' Agency
Office of Research and
Development
Washington, DC 20460
EPA/600/R-01/033
June 2001
http://www.epa.gov
Treatment of Arsenic
Residuals from Drinking
Water Removal Processes
<1
s>
\' " f
>~ <>" "«}
-v^M
X » „ •"
As (III)
As(V)
-------
-------
EPA/600/R-01/033
June 2001
Treatment of Arsenic Residuals from
Drinking Water Removal Processes
by
Michael J. MacPhee
Gail E. Charles
David A. Cornwell
Environmental Engineering & Technology, Inc.
Newport News, VA 23606
Contract No. 8C-R613-NTSA
Project Officer
Thomas J. Sorg
Water Supply and Water Resources Division
National Risk Management Research Laboratory
Cincinnati, OH 45268
National Risk Management Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
Printed on Recycled Paper
-------
Disclaimer
The information in this document has been funded by the United States Environmental
Protection Agency (EPA) under Contract No. 8C-R613-NTSAto Environmental Engineering
& Technology, Inc. It has been subjected to the Agency's peer and administrative reviews
and has been approved for publication as an EPA document. Mention of trade names or
commercial products does not constitute an endorsement or recommendation for use.
-------
Foreword
The U.S. Environmental Protection Agency is charged by Congress with protecting
the Nation's land, air, and water resources. Under a mandate of national environmental
laws, the Agency strives to formulate and implement actions leading to a compatible balance
between human activities and the ability of natural systems to support and nurture life. To
meet this mandate, EPA's research program is providing data and technical support for
solving environmental problems today and building a science knowledge base necessary to
manage our ecological resources wisely, understand how pollutants affect our health, and
prevent or reduce environmental risks in the future.
The NatiohalRisk Management Research .Laboratory is the Agency's center for
investigation of technological and management approaches for preventing and reducing risks
from pollution that threatens human health and the environment. The focus of the
Laboratory's research program is on methods and their cost-effectiveness for prevention and
control of pollution to air, land, water, and subsurface resources; protection of water quality
in public water systems; remediation of contaminated sites, sediments and ground water;
prevention and control of indoor air pollution; and restoration of ecosystems. NRMRL
collaborates with both public and private sector partners to foster technologies that reduce
the cost of compliance and to anticipate emerging problems. NRMRL's research provides
solutions to environmental problems by: developing and promoting technologies that protect
and improve the environment; advancing scientific and engineering information to support
regulatory and policy decisions; and providing the technical support and information transfer
to ensure implementation of environmental regulations and strategies at the national, state,
and community levels. The goal of this research effort is to evaluate the effectiveness of
various treatment processes for removing arsenic from residuals produced by arsenic
removal drinking water treatment technologies. '
This publication has been produced as part of the Laboratory's strategic long-term
research plan, ft is published and made available by EPA's Office of Research and
Development to assist the user community and to link researchers with their clients.
E. Timothy Oppelt, Director
National Risk Management Research Laboratory
in
-------
Abstract
The drinking water MCL was recently lowered from 0.05 mg/L to 0.01 mg/L. One concern
was that a reduction in the TCLP arsenic limit in response to the drinking water MCL could
be problematic with regard to disposal of solid residuals generated at arsenic removal
facilities. This project focused on developing a short-.^ of arsenic removal options, for
residuals produced by ion exchange (Ion Ex), reverse osmosis (RO), nanofiltration (NF),
activated alumina (AA), and iron removal processes. Both precipitation and adsorption
processes were evaluated to assess their arsenic removal effectiveness.
In precipitation tests, ferric chloride outperformed alum for removal of arsenic from residuals
by sedimentation, generally resulting in arsenic removals of 88 to 99 percent. Arsenic
removal from the high alkalinity ion exchange samples was poorer. The required iron-to-
arsenic molar ratio for best removal of arsenic in these screening tests varied widely from
4:1 to 191:1, depending on residuals type, and best arsenic removal using ferric chloride
typically occurred between pH 5.0 and 6.2. Polymer addition typically did not significantly
improve arsenic removal using either coagulant. Supernatant total arsenic levels of 0.08
mg/L or lower were attained with ferric chloride precipitation for membrane concentrates and
residuals from iron removal facilities compared to an in-stream arsenic limit of 0.05 mg/L in
place in some states. Settling alone with no coagulant also effectively removed arsenic from
iron removal facility residuals. Even with ferric chloride dosages of 50 to 200 rhg/L applied
to ion exchange regenerants, supernatant arsenic levels after treatment were 1 to 18 mg/L.
Required iron-to-arsenic molar ratios developed in precipitation work could be used by
utilities as guidelines for establishing coagulant dose needs to meet in-stream standards, and
to develop preliminary treatment costs.
Adsorption tests demonstrated the potential for different types of media and resins to remove
arsenic from liquid residuals, but did not assess ultimate capacity. Overall, the iron-based
granularferric hydroxide media evaluated in testing outperformed the aluminum-based media
and ion exchange resin for removal of arsenic. However, activated alumina and the iron-
based media provided comparable arsenic removals of close to 100 percent with an empty
bed contact time (EBCT) of 3-min for most of the membrane concentrates and the settled
iron removal facility residuals. Removal of suspended solids was key to the success of
adsorption for spent filter backwash water and clarifier flush residuals. Arsenic breakthrough
occurred very rapidly forthe ion exchange samples and for one RO concentrate, all of which
had an alkalinity of more than 1,000 mg/L (as CaCO3). This again suggests that alkalinity
significantly interferes with adsorption of arsenic. Based on this work, use of adsorption
media for treatment of arsenic-laden water plant residuals merits further exploration.
Of all of the residuals streams tested, Ion Ex regenerants were the most difficult to treat using
precipitation or adsorption. Disposal of supernatant streams resulting from treatment of
arsenic-laden residuals from ion exchange plants could pose a major challenge. TCLP
arsenic levels in all residuals generated in this work and in full-scale solid media samples
were far below the regulatory limit of 5 mg/L, and in fact were below 0.5 mg/L.
IV
-------
Contents
Foreword jjj
Abstract ... iv
Figures Vjj
Tables .. jx
Acronyms, Abbreviations, and Symbols xi
1. Introduction 1
1.1 Background 1
1.2 Literature Review f
1.3 Project Objectives 4
2. Experimental Procedures 5
2.1 Introduction 5
2.2 Treatment Plant Residuals ... 5
2.2.1 Ion Exchange "...-• 5
2.2.2 Activated Alumina 5
2.2.3 Membrane Filtration 7
2.2.4 Iron-Manganese Removal System 8
2.3 Experimental Design 8
2.4 Test Methods and Materials 11
2.4.1 Precipitation Tests ,... 11
2.4.2 Adsorption Tests 12
2.4.3 Analytical Tests 12
2.4.4 Quality Assurance/Quality Control 13
3. Test Results . 15
3.1 Introduction -....• 15
3.2 Residuals Characterization 15
3.2.1 Arsenic Concentrations 15
3.2.2 Alkalinity, pH, and Total Hardness 15
3.2.3 TDS 18
3.2.4 Total FeandTotal Mn 18
3.2.5 Sulfate 13
3.3 Precipitation Test Results 20
3.3.1 Overview 20
3.3.2 Activated Alumina 20
3.3.3 Ion Exchange Regenerants 20
-------
3.3.4 Reverse Osmosis Concentrates 22
3.3.5 Nanofiltration Concentrates 22
3.3.6 Iron Removal Plant Residuals : 22
3.3.7 Summary of Precipitation Testing 26
3.3.8 Residual Iron and Aluminum Concentrations 31
3.3.9 TCLP Test Results 32
3.4 Adsorption Test Results • 32
3.4.1 Ion Exchange Regenerants 32
3.4.2 RO Concentrates • 33
3.4.3 Nanofiltration Concentrates • • • • 33
3.4.4 Iron Removal Plant Residuals 36
3.4.5. 'Adsorption Test Summary 36
3.5 Comparison of Treatment Processes 38
3.5.1 SFBW (A) and SFBW/ACF (B) , 38
3.5.2 RO (A) and (B) Concentrates 39
3.5.3 Nanofiltration (A) and (B) Concentrates 39
3.5.4 Ion Exchange Regenerant (A) and (B) Composite Streams 39
3.5.5 Activated Alumina Regenerant 41
3.5.6 Summary 41
3.6 Solid Fraction Residuals • .41
4. Sludge Disposal Options 43
4.1 Sludge Production 43
4.1.1 Normalizing Sludge Quantities According to Treatment Process
Type 45
4.2 Federal Disposal Regulations 46
4.2.1 40 CFR 257: Criteria for Classification of Solid Waste Disposal
Facilities and Practices 46
4.2.2 40 CFR 258: Criteria for Municipal Solid Waste Landfills (MSWLF) .. 47
4.2.3 40 CFR 261: Identification and Listing of Hazardous Wastes 47
4.2.4 40 CFR 403: General Pretreatment Regulations for Existing and
New Sources of Pollution 47
4.2.5 40 CFR 503: Standards for the Use or Disposal of Sewage Sludge . 48
4.2.6 Comprehensive Environmental Response Compensation Liability
Act (CERCLA) 48
4.2.7 Hazardous Materials Transportation Act (HMTA) 48
4.3 Residuals Disposal Options 49
4.3.1 Liquid or Semi-Liquid Waste Disposal 49
4.3.2 Solid Media Disposal 50
5. Summary and Conclusions 51
5.1 Summary 51
5.1.1 Project Description 51
5.1.2 Untreated Residuals Sample Characterization 51
5.1.3 Precipitation and Adsorption Test Results 51
5.2 Conclusions 54
5.2.1 Precipitation • 54
5.2.2 Adsorption 54
5.2.3 Solids 57
5.3 Recommendations for Future Work 57
6. References 59
vi
-------
Figures
1-1
2-1
2-2
2-3
2-4
2-5
3-1
3-2
3-3
3-4
3-5
3-6
3-7
3-8
3-9
3-10
3-11
Page
Natural occurrence factors for arsenic in groundwater systems 2
Location of full-scale treatment facilities providing residuals samples 6
Schematic of ion exchange and activated alumina adsorption processes
with regeneration 7
Schematic of membrane and iron-manganese removal filtration process 9
Summary of arsenic residuals treatment plan 9
Coagulant dosage ranges used in precipitation tests 11
Total arsenic concentrations in the untreated liquid residuals 16
Alkalinity, total hardness, and pH of the untreated liquid residuals 17
Total dissolved solids concentrations of untreated liquid residuals ....' 18
Iron and manganese concentration of untreated liquid residuals ..., 19
Total arsenic concentration in the untreated residuals and in the
supernatant after ferric chloride precipitation 26
Total arsenic concentration in the untreated residuals and in the
supernatant after alum precipitation 27
Comparison of percent total arsenic reduction after alum and ferric chloride
precipitation 27
Total arsenic removal achieved per milligram of iron in solution using ferric
chloride precipitation 30
Comparison of iron concentrations in untreated residuals versus
supernatant iron concentrations after precipitation using ferric
chloride 31
Treatment of ion exchange (A) regenerant with iron-based media and
activated alumina 34
Treatment of ion exchange (B) regenerant with iron-based media and
activated alumina 34
VII
-------
3-12 Treatment of reverse osmosis (A) concentrate with iron-based media and
activated alumina 35
3-13 Treatment of reverse osmosis (B) concentrate with iron-based media and
activated alumina 35
3-14 Treatment of nanofiltration (A) concentrate with iron-based media, an ion
exchange resin, and activated alumina 36
3-15 Treatment of nanofiltration (B) concentrate with iron-based media, an ion
exchange resin, and activated alumina 37
3-16 Treatment of iron removal plant spent filter backwash water A (unsettled)
with iron-based media and activated alumina 37
3-17 Comparison of treatment processes for removing arsenic from iron removal
plant residuals—filter backwash and spent filter
backwash/adsorption clarifier flush blend 39
3-18 Comparison of treatment processes for removing arsenic from reverse
osmosis concentrate 40
3-19 Comparison of treatment processes for removing arsenic from nanofiltration
A and B concentrate 40
3-20 Comparison of treatment processes for removing arsenic from ion
exchange A and B regenerant 41
4-1 Residuals production estimates from alum precipitation of wastewaters
containing arsenic 44
4-2 Residuals production estimates from ferric chloride precipitation of
wastewater containing arsenic 44
5-1 Total arsenic concentrations remaining in the supernatant and percent
reduction after ferric chloride precipitation 53
5-2 Total arsenic concentrations remaining in the supernatant and percent
reduction after alum precipitation 53
5-3 Total arsenic concentrations in the column effluent and percent reduction
after iron-based media adsorption using a 3 min EBCT 55
5-4 Total arsenic concentration in the column effluent and percent reduction
after activated alumina adsorption using a 3 min EBCT 55
5-5 Total arsenic concentration in the column effluent and percent reduction
after ion exchange using a 3 min EBCT 56
5-6 Total arsenic concentration in the column effluent and percent reduction
after modified alumina media adsorption using a 3 min EBCT 56
VIII
-------
Tables
Page
1-1 Results of TCLP tests from six utilities 3
1-2 Summary of example residuals characteristics 4
2-1 Liquid-residuals sample description 6
2-2 Concentration factors for different membrane system recoveries 8
2-3 Liquid and semi-liquid residuals stream test matrix 10
2-4 Arsenic removal media tested 12
2-5 Data quality objectives for key measurements 13
2-6 Analysis methods summary for arsenic-containing residuals 14
3-1 Residuals sample characterization 16
3-2 Concentration of arsenic in residuals 17
3-3 Ion exchange run length as a function of influent sulfate concentration 19
3-4 Activated alumina regenerant precipitation results 20
3-5 Ion exchange regenerant precipitation results 21
3-6 RO concentrate precipitation results 23
3-7 NF concentrate precipitation results 24
3-8 Iron removal plant precipitation results 25
3-9 Summary of precipitation testing 28
3-10 Parameters used for calculating arsenic removal versus iron applied (best
ferric chloride precipitation test data) 29
3-11 Alternative evaluation of arsenic removal by precipitation (best ferric
chloride precipitation test data) 30
IX
-------
3-12 Aluminum concentrations in the supernatant following alum precipitation .... 32
3-13 TCLP results from precipitation and settling tests 32
3-14 Summary of adsorption test results 38
3-15 Summary of treatment processes for removing arsenic 42
3-16 TCLP arsenic from solid fraction residuals 42
4-1 Parameters used for calculating residuals production estimates 43
4-2 Estimated sludge production per 1,000 gal of residuals treated by
precipitation 45
4-3 Estimated volume of residuals generated per 1 MG treated 45
4-4 Estimated sludge production for a 1-mgd treatment facility 46
4-5 EPA40 CFR Part261 TCLP limits : 47
4-6 Part 503 pollutant limits for sewage sludge land application 48
5-1 Comparison of treatment processes for arsenic removal 52
5-2 TCLP arsenic from solid fraction residuals 54
-------
Acronyms, Abbreviations, and Symbols
AA Activated alumina
As Arsenic
AVWVA American Water Works Association
AWWARF American Water Works Association Research Foundation
AWWSC American Water Works Association Service Company
BV Bed volumes
CA California
CERCLA Comprehensive Environmental Response Compensation Liability Act
CFR Code of Federal Register
CWA Clean Water Act
EBCT Empty bed contact time
EE&T Environmental Engineering & Technology, Inc.
EP Extraction procedure
EPA United States Environmental Protection Agency
Fe Iron
FeCI3 Ferric chloride
GFH Granular ferric hydroxide
HMTA Hazardous Materials Transportation Act
ID Identification
(on Ex Ion exchange
MCL Maximum contaminant level
Mn Manganese
MSWLF Municipal solid waste landfill
NF Nanofiltration
NM New Mexico
NOF Natural occurrence factor
NPDES National Pollutant Discharge Elimination System
NSF National Science Foundation
QA/QC Quality assurance/quality control
QA Quality assurance
QAPP Quality assurance project plan
RCRA Resource Conservation and Recovery Act
RO Reverse osmosis
SDWA Safe Drinking Water Act
SFBW Spent filter backwash water
SFBW/ACF Spent filter backwash water/adsorption clarifier flush
TBLL Technically based local limits
TCLP Toxicity characteristic leaching procedure
TDS Total dissolved solids
USDOT United States Department of Transportation
WTP Water treatment plant
WWTP Wastewater treatment plant
XI
-------
-------
1. Introduction
1.1 Background
On December 24, 1975, EPA issued the National Interim
Primary Drinking Water Regulations. These regulations
established a maximum contaminant level (MCL) for arsenic
at 0.05 mg/L. Arsenic was designated as a priority for
regulation under the Safe Drinking Water Act (SDWA)
Amendments of 1986, and a decade later, under the SDWA
Amendments of 1996, Congress required EPA to develop a
revised arsenic regulation by January 2001. On June 22,
2000, the USEPA published in the Federal Register a notice
of proposed rulemaking to lower the arsenic MCL to 0.005
mg/L, and on January 22, 2001, a final MCL of 0.01 mg/L
was published.
With reduced drinking water limits, the arsenic-laden
residuals may also become a problem. Arsenic
concentrations in residuals will increase as more arsenic is
removed from raw water during treatment. Enhanced
coagulation is one treatment technique for increasing
removal of arsenic from raw water that will increase the
arsenic content and quantity of residuals. . Higher
concentrations of arsenic in residuals will be of particular
concern if regulatory arsenic limits in residuals are lowered
in response to the new drinking water limit. For example, the
toxicity characteristic leaching procedure (TCLP) arsenic limit
is currently set at 5.0 mg/L, or 100 times the drinking water
MCL of 0.05 mg/L. A proportional reduction would mean that
the TCLP limit would drop to 1.0 mg/L.
Arsenic in residuals can come from two major sources, the
raw water and the treatment chemicals. Based on recent
surveys by Frey and Edwards (1997), locations in the U.S.
that are likely to have high raw water arsenic levels have
been identified. Arsenic occurrence in groundwater systems
is presented in Figure 1-1. The natural occurrence factor
(NOF) is a descriptive variable used by the authors to
differentiate arsenic occurrence patterns geographically. A
ranking system was developed to assign qualitative NOF
levels to individual states in that work. The American Water
Works Service Company (AWWSC) conducted a study to
evaluate the potential impact of contaminants including
arsenic in treatment chemicals on sludge characteristics by
analyzing treatment chemicals from several water treatment
facilities (Dixon et al. 1988). Results showed the presence
of 108 to 122 mg As/kg in a ferric chloride solution, and 214
to 270 mg As/kg in liquid alum.
The handling and disposal of arsenic-laden residuals may be
a problem because various handling and disposal methods
may release arsenic back to the environment. Because
arsenic removal is sensitive to both the pH of precipitation
and the oxidation state, any process that changes pH or
results in a reducing environment may release arsenic from
the solid phase. These processes, including chemical
conditioning during dewatering, storage and lagooning, and
ultimate disposal options such as landfilling, land application,
discharge to sewer, and coagulant recycle, may all contribute
arsenic back to the environment.
1.2 Literature Review
A thorough review of the literature and a search of AWWA's
database, including the last ten years of American Water
Works Association (AWWA) journals and conference
proceedings, yielded relatively few published works that
specifically address characteristics of residuals containing
arsenic and removal of arsenic from those residuals. The
search also included numerous AWWARF publications, three
of which deal with residuals, and proceedings from the
Inorganic Contaminants Workshop (February 2000) held in
Albuquerque, New Mexico. Numerous publications dealing
with treatment of drinking water to remove arsenic were
found; however, limited information was available regarding
characteristics of residuals produced by arsenic removal
processes, or treatment of those residuals streams for
removal of arsenic.
In one residuals characterization effort conducted by NSF,
and partially sponsored by EPA, residuals quality data from
an arsenic removal facility were discussed (Bartley et al.
1991). Cornwell et al. (1992) characterized water plant
-------
Ranked Score
Low <40
Medium 40 - 70
>70
Source: Frey and Edwards, 1997
Figure 1-1. Natural occurrence factors for arsenic in groundwater systems
residuals in terms of inorganic constituents such as arsenic
and presented results of TCLP extractions. Those data,
however, were not from plants designed to remove arsenic.
Hathaway and Rubel (1987) and Clifford and Lin (1986) both
reported results of Extraction Procedure (EP) toxicity tests
performed on residuals containing arsenic. Three recent
EPA publications (Wang etal., 2000; Fields etal., 2000; and
Fields ef a/., 2000) present TCLP results for residuals
collected at arsenic removal treatment facilities, and four
additional recent publications—Chen etal. (1999), Clifford ef
al. (1999), Clifford et al. (1998), and Chwirka
(1999)—address levels of arsenic in residuals. ,
Bartley ef al. (1991) characterized residuals produced at
eight water treatment plants, including one arsenic removal
plant, according to inorganic constituents, including arsenic.
The 18-mgd arsenic-removal plant documented in that study
includes an 8-mgd surface water train and a 10-mgd
groundwater train that treats water from several wells, one of
which is known to be contaminated with arsenic. Water from
the contaminated well is treated with ferric sulfafe and
chlorine applied upstream of a contact tank, and water from
the other wells is aerated and chlorinated. The
aerated/chlorinated water is combined with the contact tank
effluent and filtered. Finished water from the groundwater
treatment train is combined with filtered water from the
surface water treatment train.
Supernatant from the contact basin in the arsenic removal
process is recycled to the head of the surface water train,
and contact basin solids, spent filter backwash water, etc.,
are routed to a wastewater holding tank, lagoons, and a
-------
temporary storage area. Arsenic levels in composite contact
basin solids samples collected over a period of six months
averaged 5,880 mg/kg. Arsenic levels in TCLP extracts
averaged just 0.016 mg/L, rendering the sludge non-
hazardous according to toxicity.
Hathaway and Rubel (1987) described a dried sludge
generated through precipitation of aluminum hydroxide from,
a spent activated alumina regeneration stream that easily
passed the Extraction Procedure (EP) toxicity test in a pilot
study on removal of arsenic from drinking water at the
Fallen, NV Naval Air Station using activated alumina and ion
exchange. The EP toxicity test is essentially the precursor
to the TCLP test. Under the EP toxicity test, a solid waste is
adjusted to a pH of 5.0, modified if necessary to conform to
particle size requirements, and placed in an extractor along
with deionized water for a period of 24 hours. The extract
from the waste is analyzed for.a number of parameters,
including arsenic. The toxicity criterion used to define a
waste as hazardous under the Resource Conservation and
Recovery Act (RCRA) was determined by the Extraction'
Procedure (EP) toxicity test prior to 1990, when that test was
replaced by the TCLP test. A sludge in that study containing
1627 mg/kg of As yielded just 0.036 mg/L As in the extract.
In another study, Clifford and Lin (1991) reported 0.6 mg/L
As in a leachate produced by similar treatment of a spent
alumina regenerant.
aluminum salts or lime contained 1.5 mg/L arsenic when
subjected to the EP toxicity test. Reuse of spent regenerant
was explored in the Albuquerque study, in which arsenic
levels in the reused brine rose to-190 mg/L (AWWA 1999).
Wang etal. (2000) reported TCLP arsenic results for spent
alumina ranging from <0.05 mg/L to 0.066 mg/L in a recently
completed EPA research report. In another EPA research
effort completed this year, Fields et al. (2000) reported
TCLP arsenic levels of less than 0.05 mg/L at an iron
removal facility. Fields etal. (2000) reported arsenic TCLP
concentrations of 0.30 mg/L or lower for residuals collected
at two coagulation/filtration plants and one lime softening
plant in a third research effort sponsored by EPA. TCLP
arsenic levels in more than 30 sludge samples collected
from dewatered sludge lagoons at the two coagulation
plants ranged from below the detection limit to 0.3 mg/L. In
dewatered residuals collected from the softening plant,
TCLP arsenic concentrations were all below the detection
limit of 0.05 mg/L.
Chen et al. (1999) reported TCLP results for arsenic
residuals collected at six different utilities. Data are
summarized in Table 1 -1. The authors noted that the As
levels in the TCLP extract of all but one residuals sample
were well below the existing limit of 5.0 mg/L as well as
much lower limits that could result if the TCLP limit is
reduced in proportion to the drinking water MCL. The
Table 1-1.
Results of TCLP tests from six utilities
Sludge source
Treatment method
Total As
(mg/kg dry solid)
TCLP concentration
(mg/L)
Utility F
Utility G
Utility J ,
Utility L
Utility C
Utility O
Lime softening
Coagulation
Lime softening
Lime softening
Alum coagulation
Fe-Mn removal
Iron coagulation
6.9
2.4
14.8
. ... 24.6
NA
369
338
0.0039
0.0009
0.002
0.028
0.0093
0.0444
1.56
Source: Chen et al. 1999.
NA - Not Analyzed
Three major laboratory and field studies addressing key
issues surrounding arsenic removal by ion exchange have
been conducted by Clifford and his colleagues at the
following locations: Hanford, CA (Clifford and Lin 1986);
McFarland, CA (Ghurye, Clifford, et al. 1999); and
Albuquerque, NM (Clifford, Ghurye, et al. 1997). In the
Hanford work, the extract from dried sludges generated by
treating spent ion-exchange regenerant using ferric or
exception was the iron coagulation sludge from Utility O.
Further, a WET extraction performed on that sludge using
citric acid increased the As level in the extract by ten-fold.
Clifford et al. (1998) addressed removal of arsenic from
spent ion exchange brine containing 3,450 ug/L As using
ferric hydroxide coagulation followed by filtration through a
0.22 urn filter. Ferric chloride dosages ranging from 1 to 50
-------
moles Fe/mol As in the brine were evaluated in that work,
and pH was varied from 5.5 to 8.5. At a pH of 5.5, a molar
ratio of 20:1 was required to lower the As concentration by
99.5 percent to 20 ug/L, consistent with the removal goal.
At pH 6.2, a molar ratio of 50:1 was required to attain similar
results. Molar ratios of 20:1 and 50:1 are approximately
equivalent to FeCI3 dosages of 150 to 375 mg/L.
Table 1-2 provides a summary of example arsenic
concentrations in water treatment residuals reported by
Chwirka (1999). The residuals volumes and arsenic
concentrations shown in the table for various types of
residuals were calculated assuming a raw water arsenic
content and arsenic removal for each treatment technology.
Calculated arsenic concentrations in residuals volumes
generated in each process shown in Table 1 -2 ranged from
0.098 mg/L for membrane technologies to approximately 10
mg/L for activated alumina and ion exchange. On a dry
weight basis, theoretical arsenic concentrations ranged from
165 to more than 14,000 mg/kg. Actual arsenic
concentrations would be site-specific. Based on the
calculated arsenic levels, the author explored the feasibility
of various disposal options.
1.3 Project Objectives
The primary objective of this project was to conduct
laboratory evaluations to determine the effectiveness of
various treatment options for removal of arsenic from
residuals produced by arsenic removal treatment
technologies. An assessment of disposal issues (e.g.,
hazardous, non-hazardous) associated with effective
treatments was also a key part of the research effort. The
approach followed to meet that objective included:
1. Collection of residuals streams and/or solid media
samples from nine different water treatment plants
2. Treatment of liquid waste streams using
precipitation and adsorption processes
3. Performance of TCLP arsenic analyses on solid
media samples and semi-liquid residuals fractions
generated in precipitation tests
Treatment performance was evaluated based on arsenic
removal, and residual arsenic levels in precipitation test
supernatant samples and adsorption column effluent
streams.
Table!-2. Summary of example residuals characteristics
Volume of As concentration
Treatment technology
Conventional coagulation
Softening
Ion exchange
Activated alumina
Iron oxide coated sand
Nanofiltration/Reverse osmosis
Coagulation/Microfiltration
residuals
produced
(gal/MG)
4,300
9,600
4,000
4,200
21,000
664,000
52,600
in residuals
volume
(mg/L)
9.25
4.2
10
9.52
1.9
0.098
0.76
Quantity of
solids produced
(Ibs/MG)
180
2,000
23.4
23.4 (calculated)
23.4 (calculated)
NA
112.6
As concentration in
solids
(mg/kg dry weight)
1,850
165
14,250
14,250 (calculated)
14,250 (calculated)
NA
2,957
Source: Chwirka 1 999.
NA- Not Applicable
-------
2. Experimental Procedures
2.1 Introduction
Three different types of water treatment plant residuals were
evaluated during the project: liquid, semi-liquid, and solid
waste. Precipitation and adsorption removal techniques
were evaluated for removal of arsenic from liquid and semi-
liquid residuals collected at full-scale facilities and shipped to
Environmental Engineering & Technology, Inc. (EE&T) in
Newport News, VA for testing. Settled solids (semi-liquid
residuals) generated from the precipitation tests were
thickened to 6 to 8 percent and analyzed to determine the
TCLP arsenic concentration. Only residuals from the iron
removal facilities generated enough settled solids to perform
TCLP analyses.
TCLP tests were conducted on solid media waste from a full-
scale activated alumina plant and filter media collected at an
iron-manganese removal facility. Ion exchange resin
material used in this project was also evaluated with a TCLP
analysis.
2.2 Treatment Plant Residuals
Residuals were collected from nine drinking water treatment
plants. The water treatment plant residuals evaluated were
generated by the following treatment processes:
• Ion exchange - 2
• Activated alumina adsorption -1
Iron-manganese removal - 2
« Nanofiltration - 2
Reverse osmosis - 2
A total of nine (9) samples were evaluated. Eight different
liquid residuals samples were collected at various locations
across the U.S. (see Figure 2-1) and delivered to EE&T for
testing, and one liquid (AA regenerant) residuals stream was
generated at EE&T's process laboratory. A summary
description of each liquid residuals stream is presented in
Table 2-1 and a process schematic for each full-scale water
treatment process used to generate these liquid residuals is
shown in the following sections along with a brief description
of each treatment process.
2.2.1 Ion Exchange
The ion exchange (Ion Ex) water treatment process is shown
in Figure 2-2. Ion exchange resins are designed to
selectively remove impurities from drinking water. A
chloride-form strong-base anion-exchange resin is used to
remove arsenate (As(V)). The resin must be regenerated
periodically using a brine solution to remove impurities that
accumulate on the ion exchange resin. Regeneration steps
include backwashing the resin and brine regeneration
followed by a final rinse to remove the brine water. All three
regeneration waste streams are typically blended together for
final disposal.
In testing conducted for this project, three different
regenerant waste samples—backwash, brine, and
rinse—were delivered to EE&T for testing in separate
containers. For ion exchange (A), each stream was
analyzed individually and the three waste streams were then
blended together in equal portions (1:1:1) to form a
composite ion exchange sample. The blend ratio was
determined based on the sample volume that was supplied
for testing. The composite sample was used for precipitation
and adsorption testing.
Ion exchange (B) regenerant samples were also collected
from a full-scale WTP during a media regeneration cycle.
The regenerant wastes included backwash water, brine, and
rinse water in separate containers. After analysis of each
individual sample, the regenerant streams were blended into
a single composite sample for testing. The blend was a 4:1:1
ratio of brine, backwash water, and rinse water, respectively.
The blend ratio was determined based on the sample volume
that was supplied for testing. The composite blend sample
used for testing was also analyzed to characterize its quality.
2.2.2 Activated Alumina
A process schematic for a full-scale activated alumina water
treatment system is also shown in Figure 2-2. The
-------
Figure 2-1. Location of full-scale treatment facilities providing residuals samples
Table 2-1. Liquid-residuals sample description
Sample ID Process description
Liquid residuals sample description
AA regenerant* Activated alumina adsorption
SFBW (A) Fe-Mn removal system
SFBW/ACF (B) Adsorption clarifier - Fe-Mn removal system
Sample of activated alumina regenerant
Spent filter backwash water
Composite sample of spent filter backwash water
and adsorption clarifier flush
RO(A)
RO(B)
NF(A)
NF(B)
Ion Ex (A)
Ion Ex (B)
Reverse osmosis
Reverse osmosis
Nanofiltration
Nanofiltration
Ion exchange
Ion exchange
Concentrate
Concentrate . . . .
Concentrate
Concentrate
Composite of ion exchange regenerant waste
streams (brine, rinse, backwash)
Composite of ion exchange regenerant waste
streams (brine, rinse, backwash)
•Sample was generated at EE&T, all other samples were generated by full-scale WTPs.
-------
RAW WATER
SOURCE
SULFURICACID
RINSE WATER •
CHLORINE
RAW WATER
SOURCE
OXIDIZING
PRE-FILTER
ACTIVATED
ALUMINA
-> WASTEWATER:
•SPENT BACKWASH
• SPENT REGENERANT SODIUM
HYDROXIDE
PRODUCT WATER
ANION
EXCHANGE
RESIN
SPENT
BACKWASH/RINSE
SPENT REGENERANT
(BRINE)
PRODUCT/TREATED
WATER
BACKWASH -
WATER
BACKWASH/RINSE
HYDROXIDE REGENERANT
SOLUTION
->• WASTEWATER:
•SPENT ACID
• SPENT RINSE
REGENERANT
BRINE SOLUTION
Figure 2-2. Schematic of ion exchange and activated alumina adsorption processes with regeneration
regenerant stream tested in this study was generated using
the same procedure as for full-scale regeneration, but it was
accomplished in the bench-scale contactor column. The
procedure used for regeneration is described in the following
paragraphs. Regeneration of activated alumina includes the
following sequence—backwashing, caustic soda
regeneration, and final rinse for removal of the caustic
regenerant solution. The waste product from each
regeneration step is typically combined into a common waste
product for disposal.
The activated alumina (AA) regenerant used for testing was
generated at EE&T's process laboratory using a spent AA
media from a full-scale water treatment plant that removes
arsenic from drinking water. The procedure used for the
bench-scale regeneration of the AA media included the
following steps:
1. Load AA media into the bench-scale test column
2. Backwash media with 2.5 bed volumes (BV) of tap
water
3. Flush media with 3 BV of 4 percent caustic soda
4. Rinse media with 10 BV of tap water
5. Combine all three regenerant streams into a
composite sample
6. Adjust pH of sample to 7.0 using sulfuric acid
This procedure was used to collect enough AA regenerant to
perform precipitation testing.
2.2.3 Membrane Filtration
Membrane treatment processes generate two streams—a
permeate (product water) and a concentrate (waste stream).
The two membrane treatment system concentrate streams
evaluated in this study were generated by reverse osmosis
(RO) and nanofiltration (NF). Reverse osmosis and
nanofiltration remove contaminants in the ionic and molecular
size ranges from drinking water. Reverse osmosis is mainly
used to remove salts from brackish water or sea water, and
nanofiltration is used for softening fresh waters and for
removal of disinfection byproduct precursors. Both
processes, however, can be used for removal of trace
inorganic contaminants. A process schematic for a typical
membrane water treatment system is shown in .Figure 2-3.
-------
Concentrate streams collected from two full-scale RO plants
and two full-scale nanofiltration drinking water treatment
facilities were used in this study.
Membrane concentrate samples (both RO and NF) were
spiked with arsenic before conducting laboratory removal
tests because they contained such low concentrations of the
metal. Arsenic in the NF concentrates was measured at
0.005 to 0.013 mg/L, while arsenic levels were below the
detection limit in RO samples.
Actual pilot data generated by EE&T during the preliminary
design phase at RO(A) were examined to determine the
concentration factor (from feed water to concentrate stream)
for arsenic and other constituents. That factor was 5. A
more conservative factor of 10 was applied, consistent with
concentration factors for different membrane system
recoveries described by Mickley etal. (1993) and tabulated
in Table 2-2. Based on a brackish RO system recovery of 85
percent (which was documented in EE&T pilot work), a
concentration factor of 5 to 10 would apply. Assuming a
source water arsenic concentration of 0.05 mg/L (the arsenic
MCL established in 1975), and applying a concentration
factor of 10, a spike dose of 0.5 mg/L was selected for both
RO concentrate streams.
Table 2-2. Concentration factors for different
membrane system recoveries
Recovery Concentration
(percent) factor
50
60
70
80
90
2.0
2.5
3.33
5.0
10.0
Source: Mickley et al. 1993.
Typical system recoveries associated with nanofiltration
system range from 75 to 90 percent (Mickley et al. 1993).
The same conservative concentration factor of 10 was
therefore applied. Assuming a source water arsenic level of
0.05 mg/L, a spike dose of 0.5 mg/L was used.
Brandhuber and Amy (2000) reported comparable rejection
of As (V) by RO and NF membranes (>90 percent) in short-
term (~4-hr) experiments, depending on experimental
condition. The authors alsoJound that As (III) was more
difficult to reject than As (V) and that rejection in RO and NF
systems averaged 67 and 32 percent, respectively. The
objective in this work was not to evaluate the effectiveness
of membranes for arsenic removal, however, but rather to
determine a reasonable concentration factor to use in spiking
membrane concentrate samples with arsenic for testing.
Arsenic (V) was used in spiking work for this project.
2.2.4 Iron-Manganes® Removal System
A process schematic for a typical iron-manganese filtration
system is shown in Figure 2-3. Feed water is passed
through a greensand media bed for removal of oxidized iron
and manganese following oxidant addition. Periodic
backwashing of the greensand media is required to remove
excess iron and manganese, as well as other particulate
contaminants removed from the feed water. Backwashing is
accomplished by reversing the flow of water through the filter
bed to flush out particulates. The backwash waste contains
elevated concentrations of Fe and Mn as well as other
contaminants.
The spent filter backwash residuals stream and spent filter
backwash water/adsorption clarifierflush blend (SFBW/ACF)
evaluated in this project were collected at facilities that also
have a clarification step for removal of solids priorto filtration.
SFBW/ACF (B) was shipped from a water treatment plant in
the Midwestern U.S. that removes iron, manganese, and
arsenic from groundwater using aeration, chlorination,
clarification using an adsorption clarifier, and granular media
filtration. Two separate samples were collected at the
plant—spent filter backwash water and clarifier flush water.
A raw characterization was conducted for both residuals
streams (Appendix A), after which the two samples were
blended (1:1) to obtain a composite sample for arsenic
removal testing. The adsorption clarifierflush and spent filter
backwash water are blended similarly for subsequent
treatment and disposal at the full-scale facility.
2.3 Experimental Design
Various precipitation and adsorption arsenic removal
processes were evaluated for each of the following types of
liquid and semi-liquid residuals streams:
Activated alumina (AA) regenerant
Ion exchange (Ion Ex) regenerant
Nanofiltration (NF) concentrate
Reverse osmosis (RO) concentrate
Spent filter backwash from Fe/Mn removal plant and
adsorption clarifier flush from Fe/Mn removal plant
Limited volumes of residuals shipped from remote plant sites
allowed for a rough screening of all of the treatment options
shown in Figure 2-4, but not a determination of optimal
conditions in each case.
Precipitation tests were conducted using two different
coagulants, alum and ferric chloride. Sulfuric acid, lime, and
-------
FEED •
WATER
ACID/ANTISCAIANT
PRE-TREATMENT
FEED
WWER
*3 *
FEEDPUMP
GREENSAND
MEDIA RLTER
» BACKWASWRINSE
(SPENT WftSTEVWl
BACKWASH/RINSE -
WATER
Figure 2-3. Schematic of membrane and iron-manganese removal filtration process
Arsenic Residuals Stream Residuals Treatment Process
Ion Exchange Brine
AA Regenerant
RO Concentrate
NF Concentrate
SFBW from Fe Removal Plant
and Blend of SFBW and
Adsorption Clarifier Flush
Precipitation
Adsorption
Exchange
Figure 2-4. Summary of arsenic residuals treatment plan
pH Adjust
Alum
Ferric
Fe-Based Media
AA Media
Anion Exchange
Resin
Modified Alumina
Analytes
Liquid Fraction
As
Fe
A]
Solid Fraction
TCLP As
-------
Table 2-3.
Liquid and semi-liquid residuals stream test matrix
Residuals origin
Treatment processes tested
Analyses conducted on liquid fraction
following treatment
Activated alumina regenerant
Ion exchange regenerant
Nanofiltration concentrate
Reverse osmosis concentrate
Spent filter backwash water from Fe
removal plant
Blend of spent filter backwash water
and adsorption clarifier flush from Fe
removal plant
FeCI3 precipitation
Alum precipitation
FeCI3 precipitation
pH adjustment
Fe media adsorption
Activated alumina adsorption
Alum precipitation
FeCI3 precipitation
Fe media adsorption
Activated alumina adsorption
Modified alumina adsorption
Ion exchange
Alum precipitation
FeCI3 precipitation
pH adjustment
Fe media adsorption
Activated alumina adsorption
Gravity settling
Alum precipitation
FeCI3 precipitation
pH adjustment
Fe media adsorption
Activated alumina adsorption
Ion exchange
Gravity settling
Alum precipitation
FeCI3 precipitation
pH adjustment
Fe media adsorption
Activated alumina adsorption
Ion exchange
Total As and Fe
Total As and Al
Total As and Fe
Total As
Total As
Total As
Total As and Al
Total As and Fe
Total As
Total As
Total As
Total As
Total As and Al
Total As and Fe
Total As
Total As
Total As
Total As
Total As and Al
Total As and Fe
Total As
Total As
Total As
Total As
Total As
Total As and Al
Total As and Fe
Total As
Total As
Total As
Total As
sodium hydroxide were also used to adjust pH, when
required. Two to four different types of adsorption
media/exchange resins were evaluated for each untreated
residuals stream (see Figure 2-4). The combination of
treatment techniques used for individual waste samples was
selected based on results of characterization tests which
were used to identify potential interferences. For example,
the effectiveness of ion exchange is reduced by common
ions such as sulfate, which the resin sites prefer to arsenic
(Ghurye era/. 1999).
The general testing approach shown in Figure 2-4 was
modified to eliminate some treatments for some waste
streams as follows:
Because sulfate levels were much greater than 250
mg/L and TDS levels were much greater than 500
mg/L, most wastes; were not treated using ion
exchange.
Modified alumina media was provided near the end
of the test program, so it could only be evaluated
using the nanofiltratipn wastes.
Only ferric chloride (FeCI3) precipitation tests were
conducted on the activated alumina regenerant,
because of its very high starting aluminum
concentration.
Gravity settling was added to the test matrix for the
wastes containing relatively high concentrations of
suspended solids.
10
.
-------
2.4 Test Methods and Materials
All laboratory treatment tests were conducted on site at
EE&T's process laboratory. The test matrix presented in
Table 2-3 shows treatments tested for each residuals
sample, along with samples analyzed.
2.4.1 Precipitation Tests
Precipitation tests were evaluated using a standard jar test.
The jar test system consisted of a Phipps and Bird six-paddle
stirrerwith 2-L square Gator jars. Untreated liquid residuals
samples were dosed with treatment chemicals and mixed for
1 min. The mixing intensity or velocity gradient was 300
sec"1. The coagulant chemicals applied during rapid mixing
included alum or ferric chloride, sometimes along with pH
adjustment chemical and/or polymer to aid settling. In some
cases, only a pH adjustment chemical was added. Rapid
mixing was followed by 30 min of flocculation, during which
the mixing intensity was tapered over the 30-min period (40-
30-15 sec'1).
Following flocculation, the mixer was turned off to allow for
settling of particulate matter. After 10 min of settling
(corresponding to an overflow rate of 0.25 gpm/ft2), samples
were collected for analysis.
Precipitation tests were performed using each of the nine
liquid residuals samples collected. The-chemicals used to
precipitate arsenic from the liquid residuals included alum,
ferric chloride, two polymers, and lime. Required pH
adjustments were made with either sulfuric acid or sodium
hydroxide. Lime and sodium hydroxide were applied in a
single test. The coagulant dose range selected was based
on preliminary screening tests for each residuals sample.
Qualitative screening tests were conducted by applying
various coagulant dosages to 200-mL beakers containing
each liquid residuals stream, mixing for about 30 sec, and
observing floe formation and settling. The coagulant dose
ranges used for alum and ferric chloride precipitation tests
are shown in Figure 2-5.
When sufficient quantities (approximately 100-mL) of settled
solids were generated during precipitation testing conducted
in 2-L Gator jars, the solids were separated from the
supernatant and used for TCLP analysis. Supernatant
samples were analyzed for arsenic and either iron or
aluminum depending on the use of ferric chloride or alum.
_
a.
(0
CO
AA Regenerant
Ion Ex (A)
Ion Ex(B)
RO(A)
i
RO (B)
NF (A)
NF(B)
SFBW(A)
SFBW/ACF (B)
Y/////////////////////////////////////777/
//////A
FeCI3
Alum
25 50 75 100 125
Coagulant dose (mg/L)
150
175
200
Figure 2-5. Coagulant dosage ranges used in precipitation tests
11
-------
2.4.2 Adsorption Tests
Four different adsorption/exchange media were used in
testing:
1. Iron-based adsorption media (Vertell 2000)
2. Activated alumina (APYRON)
3. Anion exchange resin (lonac)
4. Modified alumina (Solmetex)
Table 2-4 shows some pertinent characteristics of the media
and resins used in testing.
consistent with work conducted by Ghurye and Clifford
(Ghurye etal. 1999) on removal of arsenic and nitrate using
ion exchange. The EBCT was also consistent with work
conducted by Simms and Azizian (1997) on removal of
arsenic by activated alumina. Those authors found that run
length was linearly proportional to EBCT in the range of 3 to
12 min using a 14 x 28 mesh AA sample, but preferred to
operate in the 3- to 6-min EBCT range to minimize bed size
and media quantities.
In this project, EBCTs up to 6-min were evaluated for some
residuals samples. Samples were collected hourly over the
Table 2-4. Arsenic removal media tested
Media no. Media type
Trade name
Media properties
1 Iron-based granular ferric hydroxide
Activated alumina
Anion exchange resin
Modified alumina
Vertell 2000
Hawleys, UK
APYRON
Aqua-Bind™ Modified AA
lonac ASBI P
Solmetex Corporation
MetalhX
Size = 0.3 to 1.4 mm
UC<1.6
ES >0.6
Strong base anion
Chloride form
Bead size = 0.3 to 1.2 mm
Size = 0.85 to 1.70 mm
The iron-based media, Vertell 2000, was an early variant of
a granular ferric hydroxide media produced by Hawleys of
the UK. Severn Trent Water in the UK evaluated this media
and the granular ferric hydroxide media GEH, produced by
GEH Wasserchemic Gmb/H&Co. of Germany. Treatment at
the first UK arsenic treatment plant, commissioned in 1999,
consists of adsorption onto GEH followed by disinfection.
Prior to design of that facility, exhaustive pilot trials
concentrated on treatment by adsorption, primarily with AA
and granular ferric medias (Simms et al. 2000). The
APYRON AA is an aluminum-based granular adsorption
media designed to selectively remove both arsenic (V) and
arsenic (III). The third material used in testing was a
standard chloride-form anion exchange resin, while the fourth
test media was a modified alumina that is used for removal
of multivalent anionic metal species.
Adsorption/exchange tests were conducted using a single
2.2-cm diameter glass column filled with 90-mL of adsorption
media. Liquid residuals were pumped through the column at
a rate of 30 mL/min using a peristaltic pump for a period of 6
hours. The corresponding empty bed contact time (EBCT)
was 3 min. Experimental set-up and EBCT times were
6-hour test period and analyzed for total arsenic during all
tests. The test set-up was the same for all media/resins.
Adsorption tests in this work were not run to exhaustion
because of the very limited quantities of liquid residuals
provided for testing.
2.4.3 Analytical Tests
TCLP Tests
TCLP extraction tests were conducted on solid-phase
residuals received from operating arsenic removal plants and
on thickened SFBW/ACF residuals generated in precipitation
tests. The latter were actually semi-liquid samples separated
from jar test liquid supernatant by concentrating them in a
separatory funnel to approximately 6 to 8 percent solids.
Extractions were done in accordance with EPA Method 1311,
as outlined in the Federal Register (1990), and analyses
were conducted using EPA Method 601 OB.
For solid residuals samples, the extraction fluid used was
determined based on the pH of each sample by combining 5
g of the sample with 96.5 mL of reagent water. This solution
was vigorously stirred for 5 min using a magnetic stirrer. If
12
-------
the pH was less than 5.0, an extraction fluid (#1) with a pH =
4.93 ± 0.05 was used. If the pH was greater than 5.0, then
3.5 mL of 1 N hydrochloric acid (HCL) was added to the
solution, it was heated to 50°C and then was held for 10 min.
After the solution was cooled, if the resulting pH was less
than 5.0, extraction fluid #1 was used. If the pH was greater
than 5.0, an extraction fluid (#2) with a pH = 2.88 ± 0.05 was
used.
The solid media samples plus a volume of extraction fluid
equal to 20 times the weight of the sample were added to an
extractor vessel, secured in a rotary agitation device and
rotated at 30 ± 2 rpm for 18 ± 2 hrs. The extract was
acidified with nitric acid to pH less than 2. An acid digestion
was performed on the extract in preparation for arsenic
analysis using EPA Method 601 OB.
The residuals samples generated in precipitation tests were
thickened to 6 to 8 percent solids and filtered through a glass
fiber filter in a pressure filter device. An extraction was
performed on the solids (plus filter). The extract and filtrate
were subsequently combined for arsenic analysis.
Chemical Tests
Bench-scale treatment tests conducted. on the liquid
residuals included two different chemical precipitation
treatments and four adsorption/exchange technologies. Prior
to conducting those arsenic removal tests, each liquid
residuals sample was analyzed to determine total and
dissolved arsenic content. Several other water quality
parameters were also determined to characterize the
samples.
• pH
Alkalinity
Hardness
Conductivity
Total dissolved solids (TDS)
Total iron
• Total manganese
Total aluminum (AA regenerant only)
Sulfate .
Three samples—SFBW/ACF (B), Ion Ex (A), and Ion Ex
(B)—included more than one waste stream. SFBW/ACF (B)
was a blend of adsorption clarifier flush water and spent filter
backwash water. Both ion exchange regenerant samples
included water from backwash, brine, and rinse cycles from
regeneration. Individual waste streams were combined into
composites for testing at EE&T. These composite samples
were .also characterized using the same array of laboratory
tests.
2.4.4 Quality Assurance/Quality Control
A Quality Assurance Project Plant (QAPP) was submitted to
and approved by EPA in February 1999. The report
summarized the data quality objectives for the analytical
determinants for this project. The arsenic measurement was
determined to be the most critical parameter because arsenic
removal was used to compare treatment performance. The
other parameters that were considered key measurements
were total iron, total manganese, total aluminum, and sulfate.
The QA objectives set for these parameters are listed in
Table 2-5.
Table 2-5. Data quality objectives for key measurements
Parameter
TCLP As
Total and dissolved As
Total Fe
Total Mn
Total Al
Sulfate
Sample
Semi-liquid
Liquid
Liquid
Liquid
Liquid
Liquid
Method
EPA 601 OB
EPA 200.7
EPA 200.7
EPA 200.7
EPA 200.7
EPA 300.0A
Method
detection limit
0.002 mg/L
0.002 mg/L
0.010 mg/L
0.005 mg/L
0.050 mg/L
0.350 mg/L
Precision
(percent)
±25
±25
±25
±25
±25
±25
Accuracy
(percent)
75-125
75-125
75-125
75-125
75 - 125
75-125
The characterization tests conducted for each liquid residuals
sample included the following laboratory parameters:
Total arsenic >
• Dissolved arsenic
Project-specific quality assurance objectives were not
established for the remaining water quality parameters
evaluated for characterization of the various residuals
streams, however, the test procedures used for analysis
were either EPA or Standard Methods for the Examination of
13
-------
Water and Wastewater approved methods. The specific
methods used for these water quality parameters are listed
in Table 2-6.
Table 2-6. Analysis methods summary for arsenic-
containing residuals
Parameter Method
Alkalinity
PH
Hardness (total)
TDS
Conductivity
SM 2320B
SM 4500HTB
SM 2340C
SM 2540 C
SM 2510 B
14
-------
3. Test Results
3.1 Introduction
Residuals samples were characterized prior to conducting
precipitation and adsorption tests, and before blending or
spiking with arsenic, if required. Blended composite and
spiked samples were also characterized using the same
array of laboratory tests.
3.2 Residuals Characterization
Table 3-1 provides a summary of the key water quality
results for each sample. For samples that were blended,
only results for the composite samples used in testing are
shown. The concentrate samples collected from the reverse
osmosis and nanofiltration plants had either no arsenic or
very low arsenic concentrations, and therefore had to be
spiked with arsenic prior to testing. These samples were re-
analyzed after spiking to determine the arsenic
concentration. Complete results from all characterization
analyses are tabulated in Appendix Tables A-1 and A.2.
3.2.1 Arsenic Concentrations
Total arsenic levels measured in all untreated residuals
samples are plotted in Figure 3-1. For the reverse osmosis
and nanofiltration samples, spiked arsenic concentrations are
shown. Arsenic concentrations ranged from approximately
0.5 mg/L spiked in the membrane concentrate samples to
around 10 to 25 mg/L in the ion exchange regenerant
streams. The spent filter backwash water and spent filter
backwash water/adsorption clarifier flush blend had total
arsenic levels between the two extremes (about 1.5 mg/L),
and arsenic in the AA regenerant stream was 2.6 mg/L.
Both total and dissolved arsenic levels in the untreated
residuals samples are shown in Table 3-1. EPA Method
200.7, which was used in analyzing total and dissolved
arsenic, includes a digestion step to dissolve all particulate
matter. Nanofiltration and reverse osmosis arsenic
concentrations shown in the table are the measured levels
before spiking. Ninety-three to 99 percent of the arsenic in
the nanofiltration and reverse osmosis concentrate streams
and in the composite ion exchange regenerant samples was
in the dissolved form. In contrast, almost none of the arsenic
in the AA regenerant stream and the SFBW samples was
dissolved.
Concentration Factors
Arsenic levels in the residuals streams were compared to
corresponding source water arsenic levels to determine a
"concentration factor," or the degree to which arsenic levels
were concentrated in the residuals by the various treatment
processes. Results of those calculations are summarized
below in Table 3-2. Data are not included for the RO and NF
samples because they were spiked with arsenic, assuming
a concentration factor of 10.
The concentration factors for the SFBW and SFBW/ACF
samples were 12 and 61, respectively. Concentration of
arsenic of the AA regenerant stream was comparable, with
a concentration factor of 44. The highest concentration of
arsenic occurred in the ion exchange waste streams.
Arsenic levels were 270 and 236 times greater than the
corresponding source water arsenic concentrations for the
composite waste streams (brine, backwash, and rinse
waters) tested. Concentration of arsenic was greater for the
brine streams, which contained higher concentrations of
arsenic than the blends. Clifford et a/. (1998) reported that
arsenic was concentrated by a factor of 144 in a brine.
3.2.2 Alkalinity, pH, and Total Hardness
Alkalinity, pH, and total hardness of the nine liquid-fraction
residuals samples varied significantly. Results are shown
graphically in Figure 3-2. The highest alkalinity of 7,000
mg/L as (CaCO3) was measured for Ion Ex (B). Ion Ex (A)
and RO (A) also had high alkalinities of 950 mg/L and 2,800
mg/L, respectively. The alkalinities of the AA regenerant, the
SFBW stream and SFBW/ACF blend, and the nanofiltration
concentrates were comparable, in the 200 to 400 mg/L (as
CaCO3) range. For the two RO concentrates, alkalinities
were very different, at 600 mg/L (as CaCO3) for RO (B) and
2,800 mg/L (as CaCO3) for RO (A).
15
-------
Table 3-1.
Sample ID
Residuals sample characterization
Untreated residuals characteristics
Dissolved
Total TDS Total As As Total Fe Total Mn Conductivity Sulfate
pH Alk.* hardness* (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (uS/cm) (mg/L)
AA regenerant
SFBW (A)
SFBW/ACF (B)f
RO (A)*
RO (B)t
NF (A)*
NF (B»
Ion Ex (B)f
Ion Ex (A)f
7.1
7.6
8.1
7.9
7.3
7.1
6.6
9.7
9.0
268
430
197
2,800
600
325
210
7,000
950
13
365
400
460
840
1,560
1,750
86
90
10,240
460
341
14,300
11,750
1,765
1,533
6,240
4,100
2.63
1.41
1.74
<0.002
<0.002
0.013
0.005
24.8
10.5
0.12
<0.002
0.03
<0.002
<0.002
0.007
0.009
24.7
10.3
0.83
78.5
45.9
0.65
0.86
2.16
0.46
<0.01
0.49
0.09
7.52
3.75
0.23
1.11
0.14
0.08
<0.005
__
22,640
900
680
28,500
23,800
3,515
3,080
8,100
12,440
16,338
4.82
97.3
544
—
1,075
1,190
910
—
*mg/L as CaCO3
fAfter blending individual waste streams.
^Before spiking with As
25
20
.1
| 15
I
I
.o 10
CO
15
0
i3
2.5
8 1.5
8
« 0.5
3
ft 0
-------
Table 3-2. Concentration of arsenic in residuals
Arsenic concentration (mg/L)
Sample ID
Ion Ex (A)
Ion Ex (B)
SFBW (A)
SFBW/ACF (B)
AA regenerant
CCQX/U /A\
OrDVV (f\)
,
?999999999999999999999999<
1
•555555655565555555555555?
i
1
.55555555655656656555553 pH
1 "•
i i i i i i i 1 1 i i i i i i i 1 1 i ii
Residuals stream C
10.5
24.8
1.41
1.74
2.63
pH 7.6
oH 8.1
,pH7.1
1
pH 6 6
., _,.., J
p9999993pH79
3! .DH7.3
''VXX/'Sl
XXXJpH 9.0
R9999999999lpH 9 7
7.1
1 1 1 M I 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I
Concentration factor
270
236
61
12
44
&H Alkalinity
E3 Total hardness
10
100 1000
mg/L as CaCO3
10000
100000
Figure 3-2. Alkalinity, total hardness, and pH of the untreated liquid
residuals
The pH of most of the residuals samples was in the 6.5 to 8.0
range. Along with higher alkalinity, ion exchange regenerant
samples exhibited a much higher pH range of 9.0 to 9.7.
The highest levels of total hardness were measured in the
nanofiltration concentrate stream. Those levels,
approximately 1,500 to 1,800 mg/L (as CaCO3) were
comparable to NF concentrate TDS levels. The next highest
total hardness value was associated with RO (B) at 840 mg/L
(as CaCO3). At 840 mg/L (as CaCO3), the total hardness in
that sample was nearly twice the hardness measured in RO
(A). Ion exchange regenerants and the AA regenerant
stream exhibited much lower hardness levels, less than 100
mg/L (as CaCO3), than any of the other residuals streams.
17
-------
3.2.3 TDS
TDS levels of the liquid waste streams before treatment are
plotted in Figure 3-3. TDS ranged from 341 mg/L in
SFBW/ACF (B) to 14,300 mg/L in RO (A). Corresponding
conductivity ranged from 680 uS/cm to 28,500 uS/cm. The
highest levels of total dissolved solids (approximately 10,000
to 15,000 mg/L) were found in the AA regenerant and RO
concentrate streams. SFBWTDS levels were at the low end
of the spectrum at around 300 to 500 mg/L. NF concentrate
TDS levels were also comparatively low (about 1,500 to
1,800 mg/L), and TDS levels in ion exchange regenerant
streams were comparatively high (4,000 to 6,000 mg/L).
3.2.4 Total Fe and Total Mn
Total Fe and Mn concentrations were below detection limits
in the Ion Ex (B) regenerant, and as expected were highest
in the SFBW (A) residuals sample. Iron and manganese
levels in the SFBW and SFBW/ACF blend samples were
78.5 and 45.9 mg/L and 7.5 and 3.8 mg/L, respectively. In all
other samples, except NF (A), total Fe levels were in the 0.5
to 0.9 mg/L range. Similarly, Mn concentrations for the other
samples were in the 0.1 to 0.2 mg/L range, except for RO
(B). Iron and manganese concentrations are shown in Figure
3-4.
3.2.5 Sulfate
The liquid residuals samples had sulfate levels ranging from
less than 100 mg/L in the SFBW and SFBW/ACF blend to
over 16,000 mg/L in the AA regenerant and Ion Ex (A) brine.
Sulfate levels were in between those extremes at around 500
to 2,000 mg/L in the NF and RO concentrates and Ion Ex (B)
brine. If the source water contains <500 mg/L TDS and <150
mg/L sulfate, ion exchange may be a practical treatment
method for arsenic removal (Clifford and Lin 1986; Clifford ef
a/. 1997; Ghurye ef a/. 1999; Clifford et al. 1999). Clifford
(1999) compiled data collected in field studies conducted in
Hanford, CA; McFarland, CA; and Albuquerque, NM that
illustrate the impact of influent sulfate concentration on ion
exchange run length. Those data are tabulated below in
Table 3-3.
As shown in Table 3-3, a run length of 490 BV in Clifford
(1999) was achieved, even with a sulfate concentration of
_
*
CO
Q
16,000
14,000
12,000
10,000
8,000
6,000
4,000
2,000
0
Sample ID
Figure 3-3. Total dissolved solids concentrations of untreated liquid residual
18
-------
100
Iron
Manganese
- 6
- 4 c
- 2
Sample ID
Figure 3-4. Iron and manganese concentration of untreated liquid residuals
Table 3-3. Ion exchange run length as a function of influent sulfate concentration*
Source water concentration
Location
Hanford, CA
McFarland, CA (unspiked)
Albuquerque, NM
McFarland, CA (SO42' spiked)
McFarland, CA (SO42' spiked)
As
50
13
26
13
13
IDS
(mg/L)
213
170
328
259
436
Sulfate
(mg/L)
5
40
82
100
220
Run length
(BV)t
1,500,
1,030$
640
490
250
Source: Clifford (1999).
*Run lengths for ASB-2 type 2 SBA resin regenerated with 20 Ibs NaCI/ft3. When regenerated with 10 Ibs NaCI/ft3, run lengths
decreased by about 25 percent.
fBased on run termination at effluent arsenic concentration of 2 ug/L.
^Extrapolated value based on comparison with IRA 404 performance in McFarland.
19
-------
100 mg/L. The 250-BV run length attained with a sulfate
concentration of 220 mg/L is probably too short for
economical full-scale operation, which is why <150 mg/L
sulfate is suggested as one criterion for selecting ion
exchange for arsenic removal (Clifford, 1999).
3.3 Precipitation Test Results
3.3.1 Overview
The precipitation test results obtained using alum and ferric
chloride are summarized in the following paragraphs and
tables. A comprehensive table of test results is included as
Appendix Table A-3, and Appendix Figures A. 1 through A. 18
3.3.3 Ion Exchange Regenerants
Table 3-5 presents arsenic removal results for precipitation
tests conducted using the ion exchange regenerant streams.
For the Ion Ex (A) composite sample, a ferric chloride dose
of 100 mg/L, equivalent to a molar ratio of Fe:As of 4.4:1,
yielded an arsenic removal of approximately 79 percent at
ambient pH 7.9 (pH resulting from coagulant addition alone)
compared to about 88 percent at reduced pH 6.2. Alkalinity
was also reduced at the reduced pH condition.
Corresponding supernatant arsenic concentrations were 2.36
and 1.28 mg/L. The same ferric chloride dose applied to the
brine component of the composite, however, which contained
about three times as much arsenic, achieved 87 percent
arsenic removal at ambient pH 8.8 compared to 57 percent
Table 3-4. Activated alumina regenerant precipitation results
Untreated regenerant
Total As
(mg/L)
2.6
2.6
2.6
Dissolved
As
(mg/L)
0.12
0.12
0.12
Total Al
(mg/L)
113.0
113.0
113.0
FeCI3
dose
(mg/L)
0
25
50
Fe:As*
4.4
8.8
Coag. pH
(units)
7.1
7.1
7.0
Polymer
No
No
No
Settled
regenerant
(supernatant)
total As
(mg/L)
0.386
0.171
0.154
As
removal
(percent)
85.3
93.5
94.1
*Molar ratio of FeCI3 as Fe applied to untreated regenerant As concentration.
illustrate arsenic removal attained in precipitation work.
While a benchmark of 0.05 mg/L arsenic in the supernatant
was not a treatment goal at the outset of testing, it was used
as a comparison point when treatment results were
evaluated.
3.3.2 Activated Alumina
Table 3-4 presents test conditions including ferric chloride
dosage, molar ratio of ferric chloride as Fe, applied to the
untreated As concentration, and coagulation pH, along with
test results of arsenic concentration remaining in the
supernatant and arsenic removal. Only ferric chloride was
used for precipitation testing conducted with the activated
alumina regenerant, which contained 113 mg/L aluminum.
Appendix Figure A.1 shows those results graphically.
With no chemical addition (gravity settling only), 84.5 percent
of the total arsenic was removed from the activated alumina
(AA) regenerant waste, leaving 0.386 mg/L As in the
supernatant. Arsenic removal increased to about 94 percent
with the addition of 25 to 50 mg/L ferric chloride (Fe:As ratio
of 4.4 to 8.8). Corresponding supernatant arsenic levels in
those tests were approximately 0.15 mg/L.
removal at pH 6.4. Alum tests yielded much poorer arsenic
reductions (11 to 43 percent). Appendix Figures A.2 and A.3
show total arsenic removal and total arsenic remaining in the
supernatant for ferric chloride and alum precipitation tests
conducted on the composite sample.
As shown in Table 3-5, alum and ferric chloride precipitation
tests for the Ion Ex (B) composite sample were conducted at
ambient pH 9.9 and reduced pH 6.2, with alum and ferric
chloride dosages ranging from 50 to 200 mg/L. In ferric
chloride precipitation tests, increasing the ferric chloride
dosage from 50 mg/L to 200 mg/L increased arsenic removal
from 0 to 25 percent without polymer, and to about 30
percent with polymer at reduced pH 6.2. The corresponding
improvement in arsenic reduction was less than 10 percent
for ferric tests at ambient pH 9.9, and about the same for
alum tests conducted at 6.2. Carbonate complexing with the
iron and aluminum in these very high alkalinity samples likely
interfered with arsenic removal. Appendix Figures A.4 and
A.5 present arsenic levels remaining in the supernatant for
ferric chloride and alum tests.
Clifford et a/. (1998) studied removal of arsenic from spent
ion exchange brine containing about 3.45 mg/L As with ferric
20
-------
, c
o
's
Q.
'o
CO
Q.
E
3
C
mO
1
•+2 *Q_
3 'O
0)
2 fe.
§ -g
'S £
Q.
'o
£
Q.
c
S
a)
Q)
?
co t=
<•§<.}
0 ^
•ol| «~
1 1 |sf
co TO 0.2 •=•
CO 3
•£.
co
E
.>»
o
(0 I —
O Q* "
I
<"§.§,
5f
, •*!!
"O CO _f? W *~**
cu Ji\ ro ^t ^^
CO C3) O.S ^-
CO 3
^2-
iu
"o
n
O Q- *—
O 3-
a>
UL
ill
|=P ~o "c
5 ^ 2 < Q-
8 lip
g =» «2
co co
CD a.
2 E
re co
H CO
CO T-
T^ CO
*
T- CO
CO 05
o> in
o o
Z Z
o> in
co in
^ ^
CN <\i
§ 8
T- V™
CO CO
CO N-
t^ co
co co
CO CN
CN T1
O O
z z
05 CM
h~ co
5 5
o o
o o
in
o'
<;.-i
UJ §-
ii
CN O5 CO
O CO CO
T- oo m
°? 05 0
°. ^ CO
O O 0
z z z
•sr oo •fl-
ed CO CD
O T^ T^
in ^ ^
CN
CO
co
a>
^
<
&
c.
o>
CO
O5
CN
O
O
0
V
0
z
in
CO
CM
co
o_
m
CM
CO
o
o'
<"x:
UJ J
<- °
§5
N
O5
oo
o
o
z
in
o
CN
m
h-
§
•c
<
X
UJ
c
' CN OO T— T-
C~^ ... f"> . f~ ^ f~ ^
in <"^ CD co
coincocoincocoin
cocd'stcoiricNco'iri
CNCNCNCNCNCNCNCN
O O O O O O *«j? '«
ZZZZZZc^cf
^" ^"
O)CNO5CNC»CNO>CN
aJcdoJcoCTicdoJcd
ininoopopc)
OOr-T-CNCNCNCM
,-.,-.OOOOOO
SSoooooo
•"•"T-T-CNCNCMCN
co •*
000CN
ojcdoi-cdaicdoicd
SS22SSSS
,-.,-.000000
SSoooooo
IOI"T-T-CNCNCNCN
co
CN
§~
X o
UJ g-
o|
c: •
So
11
§§
8 1
f«
i?
II
§ "c
•• 3
o /-.
2
|S
(0 Q-
"- < cQ
_S? c !~j
0 3 0
0) _ .y
LL CO C
° °l
S-S 8
^ J9 E
21
-------
chloride precipitation. In thatwork, molar ratios of 20:1 and
50:1 (equivalent to ferric chloride dosages of 150 to 350
mg/L) were required to effect 99.5 percent removal of
arsenic. In this project, ferric chloride doses of 460 mg/L to
3,600 mg/L would have been required to achieve equivalent
molar ratios.
3.3.4 Reverse Osmosis Concentrates
Precipitation test conditions and results for RO concentrates
using ferric chloride and alum are summarized in Table 3-6.
In tests conducted using RO (A), increasing the ferric
chloride dose from 25 to 150 mg/L resulted in a dramatic
increase in arsenic removal from less than 10 percent to
greater than 90 percent. Addition of polymer in those tests
had little impact on arsenic removal, while depressing the
coagulation pH from 7.5 to 6.0 yielded a dramatic reduction
in arsenic levels remaining from nearly 0.4 mg/L to less than
0.1 mg/L (equivalent to arsenic removals of 30 and 80
percent). The significant improvement in arsenic removal
may be due to the reduction in alkalinity brought about by the
reduction in pH, and the associated reduction in carbonate
complexing in the highly alkaline concentrate stream. For
the dose and pH conditions evaluated, alum yielded no
arsenic removal. This result is consistent with results of alum
precipitation tests conducted with Ion Ex (B). Appendix
Figures A.6 and A.7 show total arsenic remaining as a
function of coagulant dose achieved using ferric chloride and
alum for RO (A).
For RO (B), ferric chloride and alum dosages of 50 and 100
mg/L, or FerAs molar ratios of 35 and 70 for ferric chloride
and AlrAs molar ratios of 19 and 30 for alum were evaluated.
A ferric chloride dose of 100 mg/L resulted in supernatant
arsenic concentrations of 0.078 at pH 6.2 and 0.132 mg/L at
pH 7.2. For the alum coagulation conditions tested, the best
arsenic reduction attained was about 57 percent. Appendix
Figures A.8 and A.9 show precipitation results achieved
using ferric chloride and alum graphically.
3.3.5 Nanofiltration Concentrates
A summary of precipitation results achieved using ferric
chloride and alum for nanofiltration concentrates NF (A) and
NF (B) is presented in Table 3-7. Figures A. 10 through A. 13
illustrate the impacts of coagulant dose, polymer, and
coagulation pH graphically.
In ferric chloride precipitation tests conducted using NF (A),
lowering the coagulation pH from about 6.5 to 5.0 increased
arsenic reduction by 4 to 12 percent, depending on ferric
chloride dose. Ferric chloride dose had little impact on
arsenic removal at ambient pH 6.5, however, at pH 5.0,
arsenic removal increased from 82 percent with 75 mg/L
ferric chloride to 98 percent with 200 mg/L. Addition of
polymer at pH 5 and 150 mg/L ferric chloride increased
arsenic removal from 76 to 88 percent (corresponding to
supernatant arsenic levels of 0.117 and 0.061 mg/L). The
impact of polymer addition was more significant in alum
tests. With 200 mg/L alum at pH 6.6, arsenic reductions with
and without polymer were 94 and 69 percent, respectively.
Increasing alum dose increased arsenic removal from about
60 percent at 75 mg/L to 94 percent at 200 mg/L. Reducing
the coagulation pH from around 7 to 6, however, had little
effect on removal of arsenic.
While a marginal improvement in arsenic reduction of 5
percent was observed when ferric chloride dose was
increased from 50 mg/L to 150 mg/L, reducing the pH from
6.5 to 5.0 had no impact in tests conducted with NF (B). As
shown in Table 3-7, all ferric chloride precipitation tests
reduced total As to below 0.05 mg/L. The effect of pH was
similar in alum tests, however, the dose effect was much
more significant. Arsenic removals of 40 to 50 percent were
attained with 50 mg/L alum, and a dose of 150 mg/L reduced
arsenic by 93 percent.
As observed for the other types of liquid residuals, on a
weight basis, ferric chloride yielded greater reductions in
arsenic than equivalent dosages of alum. On the basis of
moles of metal applied per mole of arsenic, however,
comparable molar ratios yielded similar results using the two
coagulants. For example, for NF (A), a molar ratio of Fe:As
of 72:1 with polymer reduced arsenic by 86.4 percent to
0.071 mg/L, compared to an arsenic removal of 86.0 percent
at an Al:As molar ratio of 78:1.
3.3.6 Iron Removal Plant Residuals
Results of precipitation tests conducted using spent filter
backwash waters from iron removal plants are summarized
in Table 3-8 and presented graphically in Appendix Figures
A. 14 through A. 18. Arsenic removals of 93 percent or
greater were achieved in precipitation tests conducted with
SFBW (A) using both ferric chloride and alum at dosages of
25 and 50 mg/L. Neither coagulant dose nor coagulation pH
impacted arsenic removal significantly. Polymer also had no
impact on arsenic removal. Supernatant arsenic levels were
reduced to 0.06 mg/L or less in all ferric chloride tests, and
generally below 0.05 mg/L, which is the in-stream domestic
water supply standard in some states including Arizona,
Nebraska, New Mexico, and Nevada (EPA 2000).
Supernatant arsenic levels were 0.1 mg/L or lower in all alum
tests.
In tests conducted with SFBW/ACF (B), increasing the ferric
chloride dose from 25 mg/L to 100 mg/L increased arsenic
removal from 91 to 96 percent, and lowered the supernatant
arsenic concentration from 0.152 mg/L to 0.075 mg/L.
22
-------
c
0
is
IQ.
'o
2
Q.
E
_3
cl"-
o
2
^^
8
c
o
o
O
an
to
0)
2
§?
^ 02
— CD
CO Q.
T-, 2 j5 OT
a) -£ ro < =J
K co E ^ |>
0 3 *" """'
O sOT_
CD
E
f
en 13
o o.'c
O 3-
I
i§i
^ "O C
"ro ?"
**L *—
OT > co
"- &
s^.
-a "S J3 w --••
.CD -g CO < ?J
^ co E Tg p
o §"~ ^
0 .OT,
u.
01
>,
o
a.
*~*
CD ^^ «i—
O f*V ^*
»- —
^
<
CL
LL
-S3 CD —
O OT -3,
CD O H1
IJL "0 .£
0)
Q.
' E
CO o
co 9
^ o
o
CD CO
CM r--
m i^-
0 0
z* *«
-*"
r^- CM
h-^ 00
oo oo
o o
o o
in CD
o> in
•^- In
o o
°|
o o
CO CD
CM CM
CM CM
in m
CM 'CM
o
CD
oo
o
o
CO
5
o
CO
o
z
0
CO
5
o
m
o
o
co
h-
o
COT
CD
>
CM
co
CM
r~-
o
in
CO
CO
co
i
o
1
0
CO
5
Q
in
o
CD
o
"w
0)
>
0
CD
CM
0
in
O CM i-
O CD CM
CO CM CO
oo co 'a-
CD CO O)
CO CO O
o o o
"to mtn *OT
CD Q) CD
in in co
h- N- CD
co co oo
CO CO CO
o o o
o o o
1
co CM -^ m
CO CM T- CM
co oo o> O)
CO CD ^ ^t
o o o o
o o o o
CN (N
O OT O OT
•7 CD -7 CD
^ > *- >
0000
CO CO CO CD
§ § § <2
0000
o o in in
CO
CD
in
co
co
CM
o
o
z
p
1^1
O)
o
in
m
£
CO
oo
T—
O
o
1 —
CO
m
CO
o
m
S
g
p
T—
O
m
0
o
Z
CD
in
CO
o
in
^~
m
v-
CD
in
o
o
co
m
m
CO
S
co O)
CO CO
co in
CM CD
^ 0
•^r co
o o
O 0
Z Z
co co
h^ CO
co oo
co co
o o
o o
T- CM
SCO
co
CM OO
co r-
v- O
0 0
o o
z z
CM CM
h- CD
0 0
I-- h-
0 0
O O
SB-
CD
a: .
•2 2 •§
=d ^ "£
co J? o
CD ° o
co c "
CD c —
c '-s .Q
5"^ "£
c5"§8
£8§
III
E .-^ ^
co S-'o.
CM ' m
• ** Q
II "8 "°
c~3.--=
o a. Q-
'"*** CO ^^
li£ <
8 CO (ft
CO CO
§ Q E
tn Q) ~~
(JJ __ Q
a. co co
W "~ i_
CD — —
H H
r~\ ^ ^~\
C> C\J C3
^" CN ^t
CO l~! CO
o.y 5
O t— O
d CO C
CO O CO
D> O) O>
E E E
23
-------
g
tc
•4—
'5
'o
C
Q.
E
_3
c
££
w .-§
Q) "o.
iZ 'o
S
g) LL
Q.
0)
1
8
LL
•z.
-•
• 1
CO
_Q>
m
P
3 ?
to o ?
*il
.2 ?
TJ 2 IS to —
(D •*-' (0 ^ ^
co c a. § *— •
81
a
E
"o
D.
^-s
rJlE «
o CL c
I
1 1?
"to"
co 1 ~
o o. i—
O .3-
*«
tf§t
LL "o ,fc.
0)
a.
CO ,-N
co 2
CO h-
CM 1^-
co m
CO CM
T- CM
O O
To To
CM T-
N- CO
O5 O)
CO CO
in in
•sr f
CD CM
OO 00
T- m
t^- oo
o o
o o
"to o
^~
oo r--
"fr CO
CM CM
in m
r- i^-
^.
LL
•z.
T- O M-
in co co
h- co m
o CM m
CO CO CM
T- T- CM
o o o
To 'to o
Q) tU *^^
w^ ^_ £*•
T~ T— 00
r^. co CD
CM CM CM
in in in
o o o
o o o
CM M- CO
CM O 00
OO 1^- OO
CO CO T-
co -3- co
O T- O
000
'to o 'to
^" ^"
CO "3- O
'sr co in
CD CD 2
CO CO ^1
O 0 O
o o m
in o -st-
OO CO CO
oo oo co
o co co
Si*^ *J
t~^ ^^
000
"w 'to o
O O CO
|^ CD CO
oo co S
r* ' »^ cj
[^ r^ zi_.
o o o
in in o
in oo T-
o LO oo
oo Is— o>
^" Is1** O5
CO •«- O
O v- O
o o o
o o 'to
^~
CM CO CM
CD m in
CO CO t-
*g** "^ CO
O O O
in in o
O
3
CO
s
o
1
CO
CO
s
fNI
v\
r^ oo
T~" X~
m •*
m co
CO OO
CM CM
o o
CM CM
to to
2 £
O CO
CO CO
CO CO
CM CM
0 0
in in
oo co
CO CM
CO CO
o co
CO CO
o o
0 0
0) 0)
^— ^-
•c- 0
in in
S £
o m
in i^.
Sf
LL
•z.
t- t^~
CD N-
N- CD
CD 1^
T- in
0 0
T- CM
to to
(D 0)
>- >
CO CO
CO CO
8 8
m in
CM 00
CO 00
O) CO
CO CD
O 0
0 O
o o
'to "to
a> a)
^- ^~
in co
co co
T- in
h*- Co
K §
*
m CM
CO CD
h~ OO
CO t^-
CM CD
T- O
0 0
"to "to
CD 0)
>- >-
T- m
CO CO
O) CM
co in
E §
^™"
135 O
in co
G> O)
o in
CM O
o o
o o
CM T-
to to
a) a>
^- ^~
CO CM
"3- CD
in D!
O) ^L
O O
o m
^"" ^~
s
oo
o
o
"1
00
CO
CM
m
o
o
^~
o oo
in CM
co co
co m
r- co
o o
o o
n
O •
-------
c
0
•<=
S
Q.
1
Q.
E
3
•^
o
'S
O_
3 k
Q) •—
*— t
*— Q)
.2 u-
S
'o
'o
CD
Q.
CO
"5.
1
2
1 ^
S
§
CD
CO «
a> >
2
™
CO c^
tj\ f. CD
J2 O (j
^^ CD
CO Q.
T3 "w J2 CO .-^
CD CO CO < =;J
*J g E "co 2>
CD .5 CD *» E
CO CJ Q. O C^
^J."
CD
>»
o
0.
^— -
co' X •—
0 ^ i
1
III
.If
J«* " O
™ fli
QJ Q^
•o^f „
1 Jilt
•° s<2,
1«
1
"o
QL
ro ^C -*^
o o. £•
C_J *^
-
^ j?
O T;
Li,
*co
Hi
^
— S1 CD 2"
O w •?*,
CD o yj
U_ T3 _E_
Ǥi
° •—
Q
CD
Q.
E
co
CO
co CD CM in
•^ CD CO CO
O> O> O3 C3>
•^r oo CD T-
h- -^r o> CM
o o o o
d o cJ ci
• t 1 1
« 0 0 W
CD Z Z 0>
^^ ^™
CD ^" O ^"
Is** Is** CO Is"*
in CD CD CD
in o o o
CM m m in
co •sr o T- oo
N- CO CD O) 1^
O) O> O> O) O>
•^1- CM CO CO T-
co CM m T- co
o o o o o
o o o o o
1 1 1 1 1 ]
CO O O CO CO
a> Z Z .0) «>
co t- o T- in
r»- !>•-• w Is- 10
O O> C35 O5 O>
CO O O O O
OO C3) C3) O3 C35
^ «* ^ T
0-j CD CD CD CD
in o o o o
CN m m in in
5
^
Z
m
u.
co
O) CM CO t-~
od oo in r^
CO OO CO CO
•si- m oo -^r
O) O T T-
v- CN CN CM
0 CD CD CD
O O O O
Z Z Z Z
CO T-; T-; r-
Is- CD I1*- CO
CT> o> in in
C3 ^^ *^* xf
se§§
co co co o h* r**-
T- TT co co co in
O) O> G> O> O> O>
CM o T o o m
in o CD r-- T- N.
TT- T- O O T- O
o o o o o o
o o o o o o
Z Z Z Z Z Z
o CD CM T- r*- T-
CD CD N- CD CD CD
O CD CN CN O> O)
CM co in m T- T-
•^r •sr m m. CD CD
(Q co o> c35 in m
.x co o> o> CD' CD
m o in m 2 R
CM in r- t^ ° °
S
LL.
^
CO
feg
T^ 00 ^]
^j 00 O5 **•!"
m co t>-
co co co T-
O> T- T- T-
o o o o
O 0 0 0
O O O 0
Z Z Z Z
co o> t~~ in
Is- CD CD CD
+— +- -t— -t—
oo [-^ in f2
CD CO O o
CM m oo -
m o in 2
CN in h~ "
CO
S
CD
LL.
O
^^
S CD
0 Ll_ fZ?
co co S-
ll
|j 2
§§
§§
O
^r
1 1
§ ll
3
g o
0 ^
fl
CD — t—
LL. < _l
CO CO O
CO CO -ST
CT c CD
Fi t co
0 5 <
u. co o
., 4^ =z
*r ^S C_
O ° o
o .2 c
'•g ~g co
cciS^
•5 o E
CO
CO
c ^
o o
b §i
g §5
CD rn C
.= CO O
CO 3) O
3 is CD
*i£
"D "D ^^-
- CD m
.<
"O —£? ^
S^l
3
c=
_ H- CL
CO
•o
JO
.c
CO
CO
13 «
Q..E
CD CD
Q.L1-
E
-------
Reducing the coagulation pH from about 7 to 6 had little
impact on arsenic removal. For the range of alum conditions
tested (two dosages at two pH levels) arsenic removals of 86
to 89 percent were achieved.
The iron concentration of the untreated SFBW(A) was 78.5
mg/L. Therefore, as indicated in Table 3-8 the molar ratio of
background iron plus iron applied as coagulant to
background arsenic concentration was much higher (5 to 10
times) than the molar ratio of iron applied in the coagulant to
untreated arsenic concentration. Similarly forSFBW/ACF(B)
in which the background iron level was 45.9 mg/L, molar
Fe:As ratios were 2 to 6 times higher when the background
iron was included. Background iron concentrations were
only considered in residuals collected at iron removal
facilities where iron levels in residuals were 45 mg/L or
greater. Iron concentrations were approximately 2 mg/L or
lower in all other samples.
Gravity settling the SFBW and SFBW/ACF blend samples
with no chemical addition reduced arsenic levels by 99.5 and
97.5 percent to well below 0.05 mg/L. Ferric chloride
dosages of 25 to 100 mg/L were added to settled SFBW/ACF
(B) to determine additional achievable arsenic reductions.
Up to 75 percent more arsenic was removed beyond that
achieved through gravity settling alone.
3.3.7 Summary of Precipitation Testing
The effectiveness of alum and ferric chloride precipitation for
arsenic removal was evaluated by conducting laboratory jar
tests using nine different liquid residuals streams. A
summary of untreated and treated total arsenic
concentrations attained for ferric chloride tests for each
residuals stream is presented in Figure 3-5. A similar
presentation of alum precipitation results is shown in Figure
3-6. On a weight basis, ferric chloride outperformed alum for
every residuals stream treated with the exception of NF (A).
Further, ferric chloride precipitation reduced the total arsenic
concentration of six of the nine residuals samples to less
than 0.10 mg/L but to less than 0.05 mg/L for only the SFBW
and NF samples. Exceptions were the AA regenerant and
the two ion exchange regenerants, where supernatant
arsenic levels of 0.15 mg/L (AA), 1.28 mg/L (Ion Ex (A)), and
18.7 mg/L (Ion Ex (B)) were attained. Figure 3-7 shows a
comparison of the arsenic percent removals attained with
25
Untreated residuals
CD
u_
OT
S3.
LL
1
UL
CO
<,
U.
U.
m
i
X
LLJ
X
111
25
20
•
£ 15
10
]§ 5
200 mg/L
pH6.2
Supernatant after ferric
chloride precipitation
X ml oX
E g E §
? i, I i.
IIs
CD
LL.
CO
S.
LL.
a
CO
<
LL.
SB.
LL.
§
o
8-
O
ce
CO D)
Si a)
x o:
c
O
Sample ID Sample ID
Figure 3-5. Total arsenic concentration in the untreated residuals and in the
supernatant after ferric chloride precipitation
26
-------
25
20
1
To 15
t:
8
f
2
to
1
Untreated residuals
KVVN
25
f 20
I
I 15
10
I 5
1X3 53
200 ms'
pH 6.1
Supernatant after
alum precipitation
100 mg/L
pH 5.5
E pi E
a* i£.
/
Sample ID
Figure 3-6. Total arsenic concentration in the untreated residuals and in
the supernatant after alum precipitation
Sample ID
120
100
S:
80
| 60
o
'n
ra 40
I
20
Alum precipitation tests
'/
co
-S
§
o
'•8
120
100
80
60
.9
I
ra 40
i
20
Ferric chloride precipitation tests
CO
SSSSS "2s
u. u. u. O O ?S ,S i?1
f^^^n^n^^llJLU
u^^o:a:> c c
s ^oo
Sample ID Sample ID
Figure 3-7. Comparison of percent total arsenic reduction after alum and ferric chloride precipitation
27
-------
alum and ferric chloride precipitation. The best precipitation
test results achieved for each liquid residuals sample are
presented in Table 3-9. The Table lists the coagulant and
coagulation conditions that yielded the greatest reduction in
arsenic. ;
As shown in Table 3-9, ferric chloride precipitation was
effective for removing 88 to more than 99 percent of arsenic
from all residuals samples except Ion Ex (B). Total arsenic
concentrations remaining in the supernatant ranged from
0.007 to 0.078 mg/L for all samples, except for the activated
alumina regenerant and ion exchange regenerant streams.
The ion exchange and activated alumina regenerants had
much higher initial total arsenic concentrations, so
comparable arsenic reductions of 94 and 88 percent for the
activated alumina and Ion Ex (A) resulted in much higher
supernatant arsenic concentrations (0.154 mg/L and 1.28
mg/L). Table 3-9 shows that for the SFBW, the SFBW/ACF
Table 3-9. Summary of precipitation testing
blend, RO (A), and NF (B), the total arsenic concentration
remaining in the supernatant water was reduced to below
0.05 mg/L, and supernatant arsenic levels less than 0.10
mg/L were attained in precipitation tests for RO (B) and NF
(A).
In general, addition of polymer did not have a significant
impact on arsenic removals achieved using the best ferric
chloride condition alone, butdid result in small improvements
in some cases. The pH that resulted in best arsenic
removals with ferric chloride was in the range of pH 5.0 to
6.7. Greatest benefit in depressing pH for arsenic removal
was achieved with ion exchange regenerants and one
reverse osmosis concentrate stream, which had much higher
alkalinity (1,000 mg/L or greater) than the other residuals
streams. For example, As removal was about three times
higher at pH 6 to 6.3 (82 percent) compared to pH 7.5 (30
percent) for RO (A) with 100 mg/L of ferric chloride.
Residuals stream
Sample ID
AA regenerant
Ion Ex (A)
Composite
Brine
Ion Ex (B)
RO(A)
RO(B)
NF(A)
NF(B)
SFBW (A)
Composite
Settled
comp.
SFBW/ACF (B)
Composite
Settled
comp.
*Based on Fe added as
tCationic LT 22S
tAnionic A3040 LTR
Total
arsenic
cone.
(mg/L)
2.63
10.5
33.2
24.8
0.546
0.663
0.523
0.486
1.41
1.41
1.74
0.043
coagulant
Coagulant
type
FeCI3
FeCI3
FeCI3
FeCI3
FeCI3
FeCI3
FeCI3 or
Alum
FeCI3
FeCI3
None
FeCI3
FeCI3
Best
Dose
(mg/L)
50
100
100
200
150
100
150
150
50
0
75
100
(does not consider Fe
precipitation conditions
Fe:As*
molar ratio
8.8
4.4
1.4
3.7
127
70
133
143
16.4
None
19.9
1,075
Polymer
(mg/L)
0
0
0
0.5t
2t
0
4t
4t
4*
0
0
0
Coagulation
PH
(units)
7.0
6.2
8.8
6.2
6.0
6.2
5.0
7.0 , '
6.2
7.1
7.6
6.2
6.5
Super-
natant
water
arsenic
cone.
(mg/L)
0.154
1.28
4.35
18.7
0.041
0.078
0.060
0.005
0.013
0.007
0.070
0.011
Percent
arsenic
removed
(%)
94.1
87.8
86.9
29.4
92.5
88.2
88.4
98.9
99.1
99.5
96.0
74.4
in the untreated wastewater).
28
-------
The best ferric chloride coagulation conditions for each
residuals sample tested were used to determine the total
arsenic removal achieved as a function of the total amount of
iron that was present in untreated residuals samples, plus the
iron added by ferric chloride addition. Limited volumes of
residuals allowed for a screening of treatment conditions, but
not a determination of optimal conditions in each case. The
parameters used for these calculations are shown in Table
3-10.
level. Observed removals ranged from approximately 0.0005
mol As/mol Fe to 0.05 mol As/mol Fe at treated arsenic
concentrations ranging from 0.0001 mg/L to 0.1 mg/L. By
comparison, removals in this work were similar, ranging from
approximately 0.005 to 0.017 mol As/mol Fe at supernatant
arsenic concentrations of 0.005 mg/L to 0.078 mg/L.
Precipitation results can also be examined using a linear
adsorption isotherm relationship described by Herring et a/.
Table 3-10. Parameters used for calculating the arsenic removal versus iron applied (best ferric chloride precipitation test
data)
Residuals plus Treated settled water
Untreated residuals coagulant (supernatant)
Sample ID
SFBW (A)
SFBW/ACF (B)
Ion Ex (A)
Ion Ex (B)
RO(A)
RO(B)
NF(A)
NF(B)
AA Regenerant
Coag. pH
(units)
7.1
7.2
6.2
6.2
6.0
6.2
5.2
6.2
7.0
FeCI3 dose
(mg/L as Fe)
17
26
34
68
51
34
68
51
17
Fe cone.
(mg/L)
78.50
45.90
0.49
0.01
0.07
0.86
2.16
0.46
0.83
As cone.
(mg/L)
1.41
1.74
10.5
24.8
0.5
0.7
0.5
0.5
2.6
Total Fe* cone.
(mg/L)
95.5
71.9
34.5
61.0
51.0
34.9
70.2
51.4
17.8
Fe cone.
(mg/L)
1.57
2.66
3.51
7.89
0.02
3.22
1.41
0.47
1.15
As cone.
(mg/L)
0.013
0.064
1.28
18.7
0.041
0.078
0.009
0.005
0.154
*Total iron, iron in untreated wastewater plus iron added as FeCI3.
For each residuals sample, Table 3-10 shows the best ferric
chloride dose expressed in mg/L as iron and the iron
concentration in the untreated residuals. The untreated and
treated total arsenic concentrations used for calculating
arsenic removal are also shown in Table 3-10. Figure 3-8
depicts arsenic removal in terms of mg As removed per mg
Fe present (total). The figure shows that the ratio of mg As
removed/mg Fe ranged from 0.007 to 0.267. With the
exception of the AA regenerant and Ion Ex (A and B)
wastewaters, the ratio ranged from 0.007 to 0.023 mg As
removed/mg Fe in solution, or 0.005 to 0.017 mol As/mol Fe.
The amount of iron in solution included the background iron
content of the untreated sample along with the contribution
from ferric chloride added.
Edwards (1994) synthesized all previously published work on
arsenic coagulation in water treatment, calculated moles of
arsenate removed per mole trivalent ion added, and plotted
the calculated results as a function of final treated arsenic
(1996) and McNeill and Edwards (1997). The simplified
isotherm equation described by the authors suggests that the
amount of arsenic adsorbed or removed is primarily a
function of the amount of adsorbent available. The equation
strictly applies for low concentrations of dissolved arsenic
and only as long as surface sites are not saturated by
adsorbed arsenic or by competing species. Table 3-11
shows amount of arsenic removed per amount of iron
removed in ferric chloride precipitation tests, along with the
corresponding adsorption coefficient (K) calculated using the
isotherm relationship. K values in this project ranged from 13
mlvT1 to 105 mM'1, compared to 80 mM'1 to 120 mlvT1
reported by McNeill and Edwards (1997). Thus, the isotherm
relationship may also be useful for evaluating precipitation
experiments conducted on arsenic-containing residuals
samples with higher arsenic concentrations.
29
-------
Q)
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0.03
0.025
0.02
0.015
0.01
0.005
0
100 mg/L
pH6.2
o
a
50 mg/L
pH7.0
•50Hm79/1L ^ 200 mg/L 150 mg/L 150 mg/L 1°0mf
pH7.1 „„„„„„ _u c-, _,,„„ ^u«n P" 6.2
pH 5.2 pH6.2 PH6.0
IXXXXXl KXXXX3
200 mg/L
pH6.2
Sample ID
Figure 3-8. Total arsenic removal achieved per milligram of iron in solution
using ferric chloride precipitation
Table 3-11. Alternative evaluation of arsenic removal by precipitation (best ferric chloride precipitation test data)
mg/L As mg/L As
removed per removed per
As removed As removed mg/L Fe in mg/L Fe K*
(mg/L as Fe) (mg/L) (percent) solution removed (mM'1)
FeCI3 dose
SFBW(A)
SFBW/ACF (B)
Ion Ex (A)
Ion Ex (B)
RO(A)
RO(B)
NF(A)
NF(B)
AA Regenerant
AA Regenerant (accounting for Al)
17
26
34
68
51
34
68
51
17
17
1.397
1.676
6.990
6.100
0.485
0.585
0.514
0.461
2.47
2.47
99.1
96.3
66.6
24.6
92.2
88.2
98.3
99.0
94.1
94.1
0.015
0.023
0.267
0.090
0.009
0.017 i
0.007
0.009
0.139
—
0.015
0.024
0.298
0.102
0.010
0.018
0.008
0.009
0.148
0.011*
63.9
21.1
3.6
0.3
12.9
13.2
46.4
105.4
53.8
3.8*
*Accounts for aluminum.
30
-------
The K values indicate possible interference in precipitating
arsenic from ion exchange residuals. Interference in the high
alkalinity ion exchange regenerant streams is likely due to
carbonate complexing of the iron, and higher iron dosages
would be required to achieve higher arsenic removals. Also,
Clifford et a/. (1999) found that much higher molar ratios of
iron to arsenic were required to successfully remove arsenic
from an ion exchange brine than those applied in this work,
again suggesting that higher iron dosages (more adsorbent)
would be needed. In this work it was not practical to apply
the higher molar ratios, because corresponding coagulant
dosages were approximately 500 mg/L to 3,500 mg/L.
The K value for the AA regenerant sample decreased
substantially when the aluminum removed was considered in
addition to the iron. The K value when aluminum was
accounted for was in line with that for the ion exchange
samples, even though arsenic removal from the AA
regenerant was much better (94 percent compared to 25 to
67 percent). While arsenic in the ion exchange composite
samples was nearly all dissolved, most of the arsenic in the
AA regenerant was incorporated into the solids, suggesting
that precipitation for removal of arsenic from AA is defined by
more than the sorption mechanism, and should focus on
suspended solids removal.
3.3.8 Residual Iron and Aluminum
Concentrations
Analysis for each precipitation test conducted using alum or
ferric chloride included a total metals analysis to determine
the supernatant iron or aluminum concentration remaining.
The iron concentration for each residuals sample was also
determined during the raw characterization testing, while the
aluminum concentration was only measured in the
supernatant from alum precipitation tests. A comparison of
the initial and final iron concentration after precipitation using
the best conditions for arsenic removal is presented in Figure
3-9. The figure demonstrates that the SFBW and
SFBW/ACF blend had very high initial iron concentrations
that were reduced to less than 3 mg/L after ferric chloride
precipitation. Iron concentrations in the other residuals
increased after dosing with ferric chloride for precipitation.
80
I 60
'ro
I
I 40
CD
20
I
0
IQOd I I II
nm
Untreated iron
concentration
Supernatant iron
concentration
Sample ID
Figure 3-9. Comparison of iron concentrations in untreated residuals versus
supernatant iron concentrations after precipitation using ferric chloride
31
-------
For RO (B) and Ion Ex (A) and (B), at the best precipitation
treatment conditions based on arsenic removal, iron levels in
the supernatant were greater than 3 mg/L.
Aluminum concentrations measured in the supernatant
corresponding to the best conditions for arsenic removal
ranged from less than 0.5 mg/L for SFBW (A) to more than
7.0 mg/L for the AA regenerant (see Table 3-12). As would
be expected, the untreated AA regenerant contained a very
high level of aluminum, 113 mg/L. Supernatant aluminum
levels in the ion exchange tests were 4 to 6 mg/L, and were
about 3 mg/L in RO alum precipitation tests. Residual
aluminum concentrations in the supernatant were lowest for
the nanofiltration, SFBW, and SFBW/ACF blend samples,
about 0.4 to 0.8 mg/L, in which alum precipitation yielded
arsenic reductions of 85 percent or higher.
Table 3-12. Aluminum concentrations in the supernatant
following alum precipitation
Aluminum supernatant concentration
Sample ID (mg/L)
AA regenerant*
Ion Ex (A)
Ion Ex (B)
RO(A)
RO(B)
NF(A)
NF(B)
SFBW (A)
SFBW/ACF (B)
7.42
3.73
5.82
2.76
3.09
0.673
0.654
0.429
0.762
*After ferric precipitation
3.3.9 TCLP Test Results
During precipitation testing using alum and ferric chloride,
SFBW/ACF (B) generated a high enough volume of settled
sludge to perform a TCLP analysis. Sludge solids were
separated from supernatant in eight different precipitation
tests conducted on SFBW/ACF (B) to perform TCLP tests.
Also, two sludge samples were collected following gravity
settling of the SFBW wastes without chemical addition. The
test conditions, untreated residuals arsenic levels, and
supernatant arsenic concentrations are shown along with
TCLP results for these tests in Table 3-13. The percent
solids for these thickened residuals samples was in the 6 to
8 percent range. The highest TCLP arsenic concentration
was 0.021 mg/L, which is significantly lower than the existing
EPA TCLP limit of 5 mg/L
3.4 Adsorption Test Results
Because of limited quantities of residuals samples,
adsorption tests could not be run to exhaustion. The purpose
of these tests, therefore, was to assess the potential of
various media to remove arsenic from liquid residuals
streams and not to determine ultimate adsorption capacities
or evaluate media exhaustion.
3.4.1 Ion Exchange Regenerants
Adsorption tests were conducted using both Ion Ex (A) brine
and composite regenerant samples. The adsorption media
used for testing included an iron-based media and an
activated alumina media. Two adsorption tests were
Table 3-13. TCLP results from precipitation and settling tests
Coagulation
Dose pH
Sample ID Coagulant (mg/L) (units)
Precipitation tests
SFBW/ACF (B)
SFBW/ACF (B)
SFBW/ACF (B)
SFBW/ACF (B)
SFBW/ACF (B)
SFBW/ACF (B)
SFBW/ACF (B)
SFBW/ACF (B)
Settling tests
SFBW (A)
SFBW/ACF (B)
FeCI3
FeCI3
FeCI3
FeCI3
Alum
Alum
Alum
Alum
None
None
75
100
75
100
75
100
75
100
—
—
7.2
6.7
6.2
6.1
7.3
7.1
6.1
6.1
7.6
8.1
Untreated residuals
arsenic cone.
(mg/L)
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.41
1.74
Supernatant
arsenic cone.
(mg/L)
0.064
0.110
0.070
0.075
0.194
0.248
0.205
0.214
0.007
0.122
TCLP
arsenic cone.
(mg/L)
<0.002
<0.002
<0.002
<0.002
0.003
<0.002
0.018
0.006
<0.002
0.021
32
-------
conducted using the Ion Ex (A) composite regenerant sample
using iron media EBCTs of 1.5 and 3 min. Ghurye et al.
(1999) used the same EBCTs and found that decreasing the
EBCT from 3.0 to 1.5 min did not greatly alter breakthrough
of As into the product water. Results from adsorption tests
are presented in Figure 3-10. The iron-based media
removed 60 percent of the arsenic from the raw water up to
100 BV for both the 1.5 and 3 min EBCT tests. The
corresponding arsenic concentration after 100 BV was 3.80
mg/L for the 3 min EBCT test. After 100 BV, the arsenic
removal significantly decreased. The effluent arsenic
concentration from the 1.5 min EBCT test was 7.02 mg/L
after 240 BV.
The Ion Ex (A) brine sample was also treated using the iron
media (1.5 min EBCT) and the activated alumina media (1.5
min EBCT). The results from these tests demonstrated that
neither media was effective for removing arsenic from the Ion
Ex (A) brine, perhaps because of the very high alkalinity of
the sample. The total arsenic concentration remaining after
100 BV with AA adsorption was 11.5 mg/L.
Adsorption tests were conducted using only the Ion Ex (B)
regenerant composite sample that had an arsenic content of
24.8 mg/L. Two adsorption tests were conducted using the
iron-based media and activated alumina media at a 3-min
EBCT. The iron media adsorption test was conducted for a
total of six hours (120 BV), and samples were collected after
each hour of operation. The arsenic concentrations in the
effluent are plotted versus the total bed volumes of sample
treated in Figure 3-11. The results indicate that arsenic
removal from the composite sample using the iron adsorption
media was poor. After only 40 BV, the arsenic removal was
less than 35 percent, and 16.7 mg/L of arsenic was
measured in the column effluent. Arsenic reduction declined
to less than 10 percent after 120 BV. The poor arsenic
removal may again be attributable to the very high alkalinity
of the sample.
The activated alumina adsorption test was also conducted for
120 BV, or six hours of operation at the 3-min EBCT. The
activated alumina media removed less than 10 percent of the
arsenic concentration after 40 BV, while only 3 percent of the
arsenic was removed after 120 BV.
3.4.2 RO Concentrates
RO (A) concentrate was treated using both the iron-based
media at 1.5 and 3 min EBCTs and activated alumina media
at a 3 min EBCT. The results of these tests are presented in
Figure 3-12. These data indicate that the iron-based media
with a 3 min EBCT provided the greatest arsenic removal
from the RO (A) concentrate. In that test, the total arsenic
concentration was reduced by 77 percent to 0.119 mg/L up
to 80 BV. The corresponding arsenic concentration for the
1.5-min EBCT test was 0.211 mg/L after 80 BV. By
comparison, activated alumina adsorption at a 3-min EBCT
lowered the total arsenic concentration by just 26 percent to
0.389 mg/L after 80 BV of water was passed through the
column.
The effectiveness of the iron-based media and activated
alumina were also evaluated for removing arsenic from RO
(B) concentrate. Adsorption tests were conducted using
EBCTs of both 1.5 and 3 min for each of the two adsorption
medias. The results from the four adsorption tests are
presented in Figure 3-13. These data indicate that the iron-
based media outperformed the activated alumina media,
resulting in arsenic reductions of 95 percent or greater at 120
BV. Arsenic reduction for the 1.5-min EBCT test decreased
to 84 percent after 240 BV when the arsenic level in the
column effluent increased to 0.106 mg/L.
The activated alumina was also effective for arsenic removal
from the RO (B) concentrate. AA adsorption at a 3-min
EBCT reduced the effluent total arsenic concentration by 89
percent to 0.071 mg/L after 120 BV, compared to 56 percent
at a 1.5-min EBCT. For both test media, increasing the
EBCT increased arsenic removal from the RO (B)
concentrate.
The greatest difference in the quality of the two untreated RO
concentrate samples was alkalinity. While the alkalinity of
RO (B) was 600 mg/L (as CaCO3), the alkalinity of RO (A)
was more than four times as high at 2,800 mg/L. The much
poorer arsenic removal attained in adsorption tests with RO
(A) may be due to interference from the alkalinity.
3.4.3 Nanofiltration Concentrates
Nanofiltration (A) concentrate was treated using an iron-
based media, activated alumina media, ion exchange resin,
and modified alumina. A total of six adsorption tests were
performed; iron-based media (3 and 6 min EBCT), ion
exchange (3 min EBCT), activated alumina (3 and 6 min
EBCT), and modified alumina media (3 min EBCT). The
spiked total arsenic concentration of the nanofiltration
concentrate was 0.486 mg/L prior to treatment. The results
for all six tests are presented in Figure 3-14. Both the iron
media and the activated alumina media tests with 3-min
EBCT provided greater than 90 percent removal of arsenic
up to 120 BV treated, with corresponding effluent arsenic
levels of 0.021 mg/L and 0.034 mg/L, respectively. The
arsenic removal provided by the activated alumina and iron
media were also very similar using a 6 min EBCT; after 60
BV the effluent total arsenic concentration was less than
0.007 mg/L for both.
33
-------
30
Untreated regenerant As cone
Composite - 10.5 mg/L
Brine - 33.2 mg/L
These tests were conducted
using the brine stream
+These tests were conducted
using a composite of the
backwash, brine, and rinse
regeneration streams
100 150 200
No. of bed volumes treated
2-50
Iron-based media*
EBCT = 1.5 min
Iron-based media +
EBCT = 1.5 min
Iron-based media*
EBCT = 3.0 min
Activated alumina
EBCT = 1 .5 m in
300
Figure 3-10. Treatment of ion exchange (A) regenerant with iron-based media
and activated alumina
8
25
20
15
o
| 10
-------
Untreated concentrate As cone. - 0.526 mg/L
Iron-based media
EBCT = 1.5 min
100 150 200
No. of bed volumes treated
250
Iron-based media
EBCT = 3.0 min
Activated alumina
EBCT = 3.0 min
300
Figure 3-12. Treatment of reverse osmosis (A) concentrate with iron-based
media and activated alumina
1" 0.8
8
o
'c
£2 0.4
—
1
i
0.2
Untreated concentrate As cone. - 0.663 mg/L
100 150 200
No. of bed volumes treated
250
Iron-based media
EBCT = 1.5 min
Iron-based media
EBCT = 3.0 min
Activated alumina
EBCT = 1 .5 min
Activated alumina
EBCT = 3.0 min
300
Figure 3-13. Treatment of reverse osmosis (B) concentrate with iron-based
media and activated alumina
35
-------
t 0.8
0.6
g
0.2
Untreated concentrate As cone. - 0.523 mg/L
Iron-based media
EBCT = 3.0 min
Iron-based media
EBCT = 6.0 min
Ion exchange resin
EBCT = 3.0 min
Activated alumina
EBCT = 3.0 min
Activated alumina
EBCT = 6.0 min
40 60 80 100
No. of bed volumes treated
120
140
Modified alumina
EBCT = 3.0 min
—H— •
Figure 3-14. Treatment of nanofiltration (A) concentrate with iron-based
media, an ion exchange resin, and activated alumina
After 40 BV passed through the column, the ion exchange
resin was exhausted. The modified alumina media was also
ineffective for removing arsenic from the concentrate. This
media only achieved 28 percent removal after 120 BV of
sample were treated.
Nanofiitration (B) concentrate was also treated using the
same four test adsorption medias and EBCTs as used for the
NF (A) concentrate. NF (B) concentrate had a spiked total
arsenic concentration of 0.486 mg/L. The test results
showed that both the iron media and activated alumina
media were able to remove greater than 99 percent of the
arsenic, achieving arsenic levels below the detection limit of
0.002 mg/L, using either a 3- or 6-min EBCT (see Figure 3-
15). The ion exchange resin and modified alumina media
removed less than 10 percent of the arsenic up to 120 BV of
sample treated.
3.4.4 Iron Removal Plant Residuals
SFBW (A) (mixed/unsettled sample) was treated using both
the iron-based media (1.5 and 4.5 min EBCT) and activated
alumina media (1.5 min EBCT). The results from these
adsorption tests are presented in Figure 3-16. These data
indicate that neither media was effective for removing arsenic
from SFBW (A). No removal was achieved using the
activated alumina media, and only a limited amount of
removal (24 percent after 80 BV) was achieved using the iron
media (4.5 min EBGT). The very poor arsenic removal for
these tests was attributed to the high solids loading to the
adsorption column; the SFBW (A) was a mixed, non-settled
sample.
Following these tests, the test procedure was modified to
include settling prior to adsorption tests for high solids waste
streams. The settled SFBW/ACF (B) water arsenic
concentration applied to the adsorption column was less than
0.15 mg/L. Ion exchange, iron media, and activated alumina
were used to treat the settled SFBW at an EBCT of 3-min.
The test results show that close to 100 percent of the arsenic
remaining was removed by each media tested. All measured
arsenic concentrations were less than the detection limit of
0.002 mg/L, which is well below an in-stream arsenic limit of
0.05 mg/L that is in place in some states.
3.4.5. Adsorption Test Summary
A summary of the best adsorption conditions for each
wastewater sample tested, along with the lowest arsenic
concentration achieved, is presented in Table 3-14, while
Appendix Table A-4 shows all data generated in adsorption
tests. The data indicate that only four of the residuals
samples were successfully treated using the various
adsorption media. These were RO (B) concentrate, NF (A)
and NF (B) concentrate, and settled SFBW/ACF (B).
36
-------
1 0.8
¥
•.c:
03
§ 0.6
8
g
'c
0>
°-4
0.2
20
Untreated concentrate As cone. - 0.486 mg/L
40 60 80 100
No. of bed volumes treated
120
Figure 3-15. Treatment of nanofiltration (B) concentrate with iron-based
media, an ion exchange resin, and activated alumina
Iron-based media
EBCT = 3.0 min
Iron-based media
EBCT = 6.0 min
Ion exchange resin
EBCT = 3.0 min
Activated alumina
EBCT = 3.0 min
Activated alumina
EBCT = 6.0 min
140
Modified alumina
EBCT = 3.0 min
—B—
CD
•^ 1.5
I
I
CO
1
i 0.5
0)
I
50
Untreated SFBW As cone. = 1.41 mg/L
j_
100 150 200
No. of bed volumes treated
250
Iron-based media
EBCT = 1.5 min
Activated alumina
EBCT = 1 .5 min
Iron-based media
EBCT = 4.5 min
300
Figure 3-16. Treatment of iron removal plant spent filter backwash water A
(unsettled) with iron-based media and activated alumina
37
-------
Table 3-14. Summary of adsorption test results
Residuals stream Best adsorption conditions
Sample ID
Ion Ex (A)
Ion Ex (B)
RO(A)
RO(B)
NF(A)
NF(B)
SFBW(A)f
SFBW/ACF (B)
(settled blend)
Total arsenic
concentration
(mg/L)
10.5
24.8
0.546
0.663
0.523
0.486
1.41
0.043
Adsorption media
Iron-based media
Iron-based media
Iron-based media
Iron-based media
Iron-based media or
activated alumina
Iron-based media or
activated alumina
Iron-based media
Iron-based media,
activated alumina, or
ion exchange
EBCT
3.0
3.0
3.0
3.0
3.0
3.0
4.5
3.0
No. of bed
volumes
treated
100
120
80
120
120
120
80
120
PH
9.0
9.9
7.5
7.3
7.1
6.6
7.6
8.1
Arsenic
cone.*
(mg/L)
3.81
22.3
0.119
0.018
0.030
<0.002
1.06
<0.002
Percent
arsenic
reduction
63.7
10.0
77.4
97.3
94.0
99.8
24.8
97.8
*Arsenic concentration remaining in column effluent sample collected after the number of BV listed had passed through the
media.
The table shows that none of the media tested was
successful at removing arsenic from either of the ion
exchange regenerant waste waters. The maximum removal
achieved was 64 percent for the Ion Ex (A) composite,
however, the resulting effluent arsenic concentration was
nearly 4 mg/L.
Adsorption worked best for removing arsenic from the two
nanofiltration concentrates and one of the RO concentrate
samples. Both the iron media and activated alumina were
equally effective for treating the nanofiltration concentrates
(NF (A) and NF (B)), while the iron-based media worked best
for removing arsenic from the RO (B) concentrate. For all
three of these samples, the arsenic concentration was
reduced to less than 0.05 mg/L. Due to the very low arsenic
concentration in the settled SFBW/ACF (B) sample, all three
adsorption/exchange medias tested (iron-based media,
activated alumina, and ion exchange resin) were able to
remove nearly 100 percent of the arsenic. Adsorption
yielded the poorest arsenic removal for the ion exchange
samples and RO (A), which were the three residuals samples
with the highest alkalinity, suggesting that alkalinity was an
interference.
3.5 Comparison of Treatment Processes
The precipitation and adsorption test results were compared
to determine which treatment technique was most effective
for removing total arsenic from each residuals sample.
Treatment comparison was based on the concentration of
total arsenic remaining in the supernatant or column effluent
water after treatment.
3.5.1 SFBW (A) and SFBW/ACF (B)
A total of six treatment processes were used to treat the
SFBW and SFBW/ACF blend (settled and unsettled)
samples. These tests included gravity settling, alum and
ferric chloride precipitation, iron-based media adsorption, AA
adsorption, and anion exchange. The results from these
tests are presented in Figure 3-17. Adsorption was only
effective for SFBW/ACF (B), which was settled prior to
passing it through the adsorption column. Gravity settling
without chemical addition for SFBW/ACF (B) reduced the
total arsenic concentration by 97.5 percent to 0.043 mg/L.
Ferric chloride precipitation was also effective for removing
arsenic from the unsettled SFBW (A) and SFBW/ACF (B)
yielding supernatant concentrations of 0.013 mg/L and 0.064
mg/L, respectively. By comparison, alum precipitation of
SFBW (A) resulted in a supernatant concentration of 0.021
mg/L (98.5 percent reduction). These supernatant arsenic
levels attained through precipitation were near or below the
in-stream standard of 0.05 mg/L that is in effect in some
states.
Overall, the optimal treatment scheme for arsenic removal
from SFBW (A) and SFBW/ACF (B), depending on the
treated total arsenic concentration required, would include
gravity settling to lower the TSS concentration, and possibly
coupling that with either ferric chloride precipitation or an
adsorption process.
38
-------
0.25
o>
.c
'c
'(0
0.2
£ 0.15
.0
IS
I
0.1
.o
'£
* 0.05
I
Untreated SFBW (A) total As concentration: 1.41 mg/L
Untreated SFBW/ACF (B) total As concentration: 1.74 mg/L
<0.002
<0.002
JSJ SFBW (A)
HO SFBW/ACF (B)
Gravity Settling Alum PPT Ferric chloride PPT Fe media AA media
Treatment process
Figure 3-17. Comparison of treatment processes for removing arsenic from iron
removal plant residuals-filter backwash and spent filter backwash/adsorption
clarifier flush blend
3.5.2 RO (A) and (B) Concentrates
RO concentrate samples A and B were each treated using
alum and ferric precipitation and adsorption using an iron-
based media and AA. Treatment results are compared in
Figure 3-18. As shown in the figure, ferric chloride
precipitation was the best treatment for RO (A), yielding a
total arsenic level in the supernatant of 0.015 mg/L, while
adsorption with an iron-based media was best for RO (B).
With the iron-based media, total As in the column effluent
was 0.02 mg/L after 120 BV. It should be noted that while
the iron-based media adsorption treatment provided the best
removal arsenic from the RO (B) concentrate, the final
arsenic concentration was analyzed after only 120 BV. Alum
precipitation and activated alumina adsorption were not
effective for removing arsenic from these two RO concentrate
samples. For both RO concentrate streams, arsenic levels
were reduced below an in-stream level of 0.05 mg/L.
3.5.3 Nanofiltration (A) and (B) Concentrates
Nanofiltration concentrate samples A and B were each
treated using alum and ferric chloride precipitation and
adsorption using all four test medias (iron media, AA, ion
exchange, and modified alumina). The resulting treated
water arsenic concentrations are shown graphically in Figure
3-19. Ferric chloride precipitation lowered the total arsenic
concentration from NF (A) to 0.009 mg/L and from NF (B) to
0.005 mg/L. Alum precipitation was slightly less effective for
arsenic removal from the NF concentrates tested, however,
total arsenic was reduced to below 0.05 mg/L using alum.
Only the iron-based media and activated alumina media were
effective for removing arsenic from the NF concentrates in
adsorption tests. The iron-based media provided the best
total arsenic removal from both NF (A) and NF (B), yielding
effluent As concentrations of 0.021 mg/L and <0.002 mg/L,
respectively. Based on these data, either precipitation or
adsorption would be viable treatment options fortotal arsenic
removal to achieve a total arsenic concentration below 0.05
mg/L.
3.5.4 Ion Exchange Regenerant (A) and (B)
Composite Streams
Due to the very high total arsenic concentrations present in
the Ion Ex (A) and Ion Ex (B) wastewaters (230 to 270 times
the concentrations in the corresponding source waters), the
supernatant and effluent total arsenic concentrations
resulting from precipitation and adsorption treatments were
greater than 1.0 mg/L. A comparison of the total arsenic
concentrations remaining for each treatment option is shown
graphically in Figure 3-20. These data indicate that ferric
chloride precipitation provided the best overall treatment,
however, for the dosages tested, the total arsenic
39
-------
i
£ 0.6
I
e
1 0.4
» 0.2
I
Spiked RO (A) total As concentration: 0.5 mg/L
Spiked RO (B) total As concentration: 0.7 mg/L
Alum PPT Ferric chloride PPT Fe media
Treatment process
AA media
RO (A)
RO (B)
Figure 3-1 8. Comparison of treatment processes for removing arsenic from
reverse osmosis concentrate
I
0.8
0.6
0.4
Sj 0.2
n
I
Spiked NF (A) total As concentration: 0.5 mg/L
Spiked NF (B) total As concentration: 0.5 mg/L
O
B3 NF (A)
CD NF (B)
Treatment process
Figure 3-1 9. Comparison of treatment processes for removing arsenic from
nanofiltration A and B concentrate
40
-------
3U
;entration remaining (mg/
-». NJ NJ
u\ o (n
c
o 10
Total arsen
3 01
Untreated Ion Ex (A) total As concentration: 10. 3 mg/L
_ Untreated Ion Ex (B) total As concentration: 24.7 mg/L
-
u
kXXXX5
m
ND
Ion Ex (A)
Ion Ex (B)
Alum PPT Ferric chloride PPT Fe media
Treatment process
AA media
Figure 3-20. Comparison of treatment processes for removing arsenic from
ion exchange A and B regenerant
concentrations remaining from the Ion Ex (A) and Ion Ex (B)
wastewaters were 1.28 mg/L and 18.7 mg/L, respectively.
Adsorption treatments were ineffective for removing arsenic
from these regenerant streams.
3.5.5 Activated Alumina Regenerant
Only one treatment process, ferric chloride precipitation, was
used to treat the activated alumina regenerant. Total arsenic
in the AA regenerant was lowered by 94 percent from 2.36
mg/L to 0.154 mg/L.
3.5.6 Summary:
Using the test matrix for this work, a summary of the best
treatment technplogy determined for each residuals sample
is presented in Table 3-15. Only three residuals streams (AA
regenerant, Ion Ex (A), and Ion Ex (B)) had treated total
arsenic concentrations that exceeded 0.05 mg/L, which is the
in-stream arsenic standard in some states, in all tests. The
results show that overall, the iron-based coagulants and
adsorption media resulted in greater arsenic reductions than
the aluminum-based coagulant and adsorption media. The
adsorption tests only provided an indication for the potential
of arsenic removal, since exhaustion could not be adequately
assessed using the relatively low number of BVs that could
be treated.
3.6 Solid Fraction Residuals
TCLP tests were conducted on four spent adsorption/filtration
media. The media tested were an activated alumina media
from a full-scale arsenic removal facility, a spent iron-
manganese filter media from a full-scale WTP, and a spent
anion exchange resin from two in-house ion exchange tests.
The anion exchange resins analyzed were collected after ion
exchange tests using SFBW/ACF (B) and NF (A)
concentrate. The results of the TCLP arsenic analyses are
included in Table 3-16.
The maximum TCLP arsenic concentration was 0.203 mg/L,
which is significantly below the current TCLP arsenic limit of
5.0 mg/L. TCLP arsenic levels were in fact below 1.0 mg/L,
which could be the future limit if the TCLP limit is lowered in
proportion to the drinking water MCL. The other solid waste
TCLP arsenic concentrations were at least an order of
magnitude lower. Based on these findings, these media
would not be classified as hazardous wastes.
41
-------
Other researchers have reported similar TCLP results.
Wang etal. (2000) reported TCLP arsenic concentrations of
less than 0.05 mg/L and 0.07 mg/L or less in spent activated
alumina samples collected from roughing filters at two full-
scale activated alumina facilities. Chwirka (1999) reported
no incidences of TCLP failure among eight different
conventional facilities whose residuals were analyzed for
TCLP arsenic. A wide range of arsenic levels in the TCLP
extract was reported (0.0009 mg/L to 1.6 mg/L), however,
and overall arsenic concentrations were higher than those
determined in this work.
Table 3-15. Summary of treatment processes for removing arsenic
Sample ID
Best treatment conditions determined from testing
Total As remaining
(mg/L)
AA regenerant
Ion Ex (A)
Ion Ex (B)
RO (A)
RO(B)
NF(A)
NF(B)
SFBW (A) (settled)
SFBW/ACF (B) (unsettled)
(settled)
None
None
None
Ferric chloride precipitation
Iron media adsorption
Ferric chloride precipitation, iron-based media or AA adsorption
Iron media adsorption, ferric chloride precipitation
Ferric chloride precipitation
Gravity settling (no chemical addition)
Iron media, ion exchange, or AA adsorption
0.154
1.28
18.7
0.041
0.018
0.009, 0.030
O.002, 0.005
0.013
0.043
<0.002
Table 3-16. TCLP arsenic from solid fraction residuals
Solid waste ID
TCLP arsenic
concentration
(mg/L)
Spent activated alumina (full-scale WTP)
Spent Fe-Mn filter media (full-scale WTP)
Spent anion exchange resin (bench-scale SFBW test)
Spent anion exchange resin (bench-scale nanofiltration concentrate test)
0.010
0.004
0.023
0.203
42
-------
4. Sludge Disposal Options
4.1 Sludge Production
In order to quantify the volume of settled solids that could be
expected when treating various types of water treatment
plant residuals streams, using alum or ferric chloride
precipitation techniques, empirical sludge production
equations were utilized (Cornwell 1999). The equations used
were developed for estimating sludge production from the
treatment of raw water for production of drinking water using
chemical coagulants. Equation inputs used for this analysis
include a volume of residuals treated, the total suspended
solids (TSS) concentration in the residuals, and the
coagulant dose used for arsenic removal. The coagulant
dose range used for precipitation testing was between 25
and 200 mg/L, therefore, sludge production estimates for
each coagulant type were calculated using doses of 25, 50,
75, 100, 150, and 200 mg/L. The measured TSS value for
each of the residuals used for estimating sludge production
along with the actual alum and ferric dose range used for
each residuals stream are listed in Table 4-1. SFBW(A) had
the highest TSS of 193 mg/L due to the nature of the
residuals stream, while the NF (A) concentrate and Ion Ex
(B) had TSS concentrations less than 10 mg/L.
The sludge production estimates (dry Ib/MG of residuals
treated) calculated using the empirical equations for alum
and ferric chloride are shown in Figures 4-1 and 4-2,
respectively. Both figures show that the SFBW (A) would
generate the most sludge per volume of residuals treated.
SFBW (A) was generated by backwashing filters that remove
larger suspended particles from drinking water, and therefore
had a higher TSS concentration than the other residuals
analyzed. The RO concentrates, nanofiltration concentrate,
and ion exchange regenerant were all generated by
treatment processes that were designed for removing
dissolved macro molecular or ionic contaminants from
drinking water, meaning the TSS concentration in those
residuals is low compared to the SFBW.
Figures 4-1 and 4-2 illustrate that ferric chloride generates
significantly higher sludge quantities than equivalent doses
of alum (on a weight basis). Results from the empirical
sludge production calculations demonstrate that the amount
of sludge generated using ferric chloride would be 25 to 100
percent higher than the dry weight of the alum sludge
produced using similar applied doses. The minimum and
maximum amounts of dry sludge per volume of residuals
treated for both coagulants are shown in Table 4-2. The
sludge production calculation includes the best coagulant
dose for arsenic removal for both alum and ferric chloride.
The table shows that due to the high doses of ferric chloride
necessary for achieving optimal arsenic removal, the sludge
amounts produced for the different waste steams would
range between 1.0 and 2.0 dry lbs/1,000 gal of residuals
treated.
Table 4-1. Parameters used for calculating residuals production estimates
Measured total suspended solids
concentration Alum dose range tested
Sample ID (mg/L) (mg/L)
Ferric chloride dose range
tested
(mg/L)
SFBW (A)
RO(A)
RO (B)
NF(A)
Ion Ex (B)
193.0
32.5
27.5
1.5
9.0
25 to 50
100 to 150
50 to 100
75 to 200
50 to 200
25 to 50
25 to 100
50 to 100
75 to 200
50 to 200
43
-------
100 150
Alum dose (mg/L)
200
Figure 4-1. Residuals production estimates from alum precipitation of
wastewaters containing arsenic
3,500
3,000
=9 2,500
I
| 2,000
a. 1,500
1 1,000
500
0
50 100 150
Ferric chloride dose (mg/L)
200
SFBW (A)
- 5* -
RO (A)
Concentrate
RO (B)
Concentrate
NF (A)
Concentrate
Ion Ex (B)
Composite regenerant
250
SFBW (A)
— -* -
RO (A)
Concentrate
RO (B)
Concentrate
NF (A)
Concentrate
Ion Ex (B)
Composite regenerant
250
Figure 4-2. Residuals production estimates from ferric chloride precipitation
of wastewater containing arsenic
44
-------
Table 4-2. Estimated sludge production per 1,000 gal of residuals treated by precipitation
Coagulant dose range used for Sludge production estimate
precipitation testing (dry weight)
Sample ID
SFBW (A)
RO(A)
RO(B)
NF(A)
Ion Ex (B)
— No optimal
min.
max.
best dose
min.
max.
best dose
min.
max.
best dose
min.
max.
best dose
min.
max.
best dose
condition was found.
Alum
(mg/L)
25
50
-
100
150
-
50
100
-
75
200
150
50
200
— -
FeCI3
(mg/L)
25
50
50
25
150
150
50
100
100
75
200
150
50
200
—
Alum sludge
(lb/1,000gal)
1.70
1.79
—
0.64
0.82
—
0.41
0.60
—
0.29
0.75
0.56
0.26
0.81
—
FeCI3 sludge
(lb/1,000gal)
1.82
2.03
2.03
0.48
1.53
1.53
0.65
1.07
1.07
0.64
1.69
1.27
0.50
1.75
—
4.1.1 Normalizing Sludge Quantities
According to Treatment Process Type
The calculated sludge production data (Table 4-2) provide
the expected mass of sludge generated per known volume of
residuals treated, however, these data do npt provide a mass
of sludge produced per volume of raw water treated by each
of the different treatment processes. Normalizing these
results provides a better understanding of how much sludge
each treatment process analyzed would be expected to
generate. In order to normalize these data, the following
assumptions were made:
Percentage of residuals generated by each
treatment process (RO, NF, Fe/Mn removal, Ion Ex)
Total treatment plant process (raw water) flow rate
(in this case 1 mgd was used)
Each of these parameters is defined in Table 4-3.
These data show that the membrane treatment processes
would generate a significantly higher volume of residuals
than the Fe/Mn filtration and ion exchange systems. Both
RO and NF would generate approximately 150,000 gpd per
1 mgd treated, compared to 50,000 gpd for Fe/Mn filtration
and 20,000 gpd for ion exchange.
Table 4-3. Estimated volume of residuals generated
per 1 MG treated
Residuals Volume of
Total plant generated residuals
flow rate (percent of generated
(mgd) total flow) (gpd)
Reverse osmosis
Nanofiltration
Fe/Mn filtration
Ion exchange
1
1
1
1
15
15
5
2
150,000
150,000
50,000
20,000
In order to determine the mass of sludge produced per 1 mgd
of raw water treated, the sludge production amounts (dry
lb/1,000 gal) calculated for the best coagulant dose (Table 4-
2) was multiplied by the volume of residuals generated for
each process (Table 4-3). These data are summarized in
Table 4-4.
The table shows that the mass of sludge produced per MG
of raw water treated is highest for the membrane processes
due to the large volume of residuals generated. For
example, the reverse osmosis facility that generated the RO
45
-------
Table 4-4. Estimated sludge production for a 1 -mgd treatment facility
Sample ID
SFBW(A)
RO(A)
RO(B)
NF(A)
Ion Ex (B)
*Best FeCI3 dose found for
Residuals volume
50,000
150,000
150,000
150,000
20,000
removing As from each
Best FeCI3 dose*
(mg/L)
50
150
150
150
200
untreated residuals
Sludge production
using best FeCI3
dose
(drylb/1,000galof
wastewater)
2.03
1.53
1.07
1.27
1,75
sample during precipitation
Total sludge
production
(dry Ib/mil gal raw
water treated)
101.5
229.5
160.5
190.5
35
testing.
(A) residuals would be expected to generate almost 230 dry
Ibs of sludge per MG treated if removal of arsenic from the
concentrate was required. The ion exchange facility (Ion Ex
B) would produce the least amount of sludge at 35 dry Ib/MG
raw water treated.
4.2 Federal Disposal Regulations
There are no existing comprehensive federal regulations that
specifically apply to water treatment plant (WTP) residuals.
There are, however, existing federal regulations that were
developed for biosolids and solid waste disposal. Many
states have adopted all or parts of these federal guidelines
for regulating WTP residuals disposal.
Federal statutory and regulatory requirements for disposal of
liquid and solid WTP residuals were summarized in a recent
publication (Science Applications International 2000). A
summary description of some of the federal regulations that
are currently being adopted by states for applications
involving WTP residuals are as follows:
40 CFR 257: Classification of Solid Waste Disposal
Facilities and Practices
40 CFR 258: Criteria for Municipal Solid Waste
Landfills (MSWLF)
40 CFR 261: Toxicity Characteristic Leaching
Procedure (TCLP) Test
40 CFR 403: General Pretreatment Regulations for
Existing and New Sources of Pollution
40 CFR 503: Standards for the Disposal of Sewage
Sludge
• CERCLA: Comprehensive Environmental Response
Compensation Liability Act
HMTA: Hazardous Materials Transportation Act
The Clean Water Act (CWA), Section 405, established
guidelines for the use and disposal of sewage sludge in order
to protect leaching of contaminants into waterways.
Leaching of metals into groundwater is the primary issue
addressed by CWA Section 405. The framework defined by
CWA Section 405 was also adopted for use in land applied
WTP sludge. The Resource Conservation and Recovery Act
(RCRA) was established primarily to determine toxicity or
hazard potential of a solid waste prior to landfilling in order to
protect land, water, and air from contamination. The RCRA
also provides guidelines concerning the following topics:
Classification of hazardous wastes
Standard for treatment, storage, and final use
Enforcement of standards
Authorization for states to implement regulations
Cradle to grave manifest system
Although developed for biosolids and solid waste, specific
sections of RCRA have been adopted by many states for
regulating WTP residuals end use applications. A summary
of the 40 CFR sections that could apply to WTP residuals are
listed in the following paragraphs.
4.2.1 40 CFR 257: Criteria for Classification of
Solid Waste Disposal Facilities and
Practices
This regulation includes provisions that deal with land
application of a solid waste, including WTP residuals. In
order to comply with Section 405(d) of the Clean Water Act,
the owner or generator of a publicly owned treatment facility
must comply with the guidelines for sludge applications
outlined in 40 CFR 257. The regulation contains specific
criteria governing application of sludge to land for production
of human food-chain crops and limiting annual and
cumulative applications of cadmium and PCBs.
46
-------
4.2.2 40 CFR 258: Criteria for Municipal Solid
Waste Landfills (MSWLF)
The 40 CFR 258 regulation establishes minimum national
criteria for all MSWLF units and for MSWLF that are used to
dispose of biosolids. Biosolids, solid wastes, and WTP
residuals that are placed in a MSWLF must be nonhazardous
as determined by 40 CFR 261, and must not contain free
liquids as determined by the Paint Filter Liquid Tests.
4.2.3 40 CFR 261: Identification and Listing of
Hazardous Wastes
The 40 CFR 261 identifies the solid waste materials which
are subject to regulation as a hazardous waste. A solid is
considered a hazardous waste if it exhibits any of the
characteristics of ignitability, corrosivity, reactivity, ortoxicity
as defined in Subpart C of CFR 261 or if it is listed in Subpart
D of CFR 261. This regulation is pertinent since the final use
options considered for WTP residuals application require a
nonhazardous designation. Since WTP residuals are not
ignitable, corrosive, reactive, or considered hazardous
wastes, the toxicity characteristic leaching procedure (TCLP)
could be used as the primary indicator that a WTP residual
is not a hazardous material. The TCLP regulatory limits
established by 40 CFR 261 are listed in Table 4-5.
Table 4-5. EPA 40 CFR Part 261 TCLP limits
EPA Section 40
Part 261 TCLP limits
Contaminant
EPA Section 40
Part 261 TCLP limits
(mg/L)
Contaminant
Metals
Silver
Barium
Cadmium
Chromium
Lead
HBCiSenMj**"'"'* "'" *~ '' '
Selenium
Mercury
(mg/L)
5
100
1
5
5
*j. -y^^ ^ -,^ K ^Wg -"• •*" -»>y f I
1
0.2
Volatiles
Benzene
Carbon Tetrachloride
Chlorobenzene
Chloroform
1,2-Dichloroethane
1,1-Dichloroethylene
Methyl ethyl ketone
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
1,4-Dichlorobenzene
Semi-Volatiles
o-cresol
0.5
0.5,
100
6
0.5
0.7
200
0.07
0.5
0.2
7.5
200
m-cresol
p-cresol
Cresol (total)
2,4-Dinitrotoluene
Hexachlorobenzene
Hexachlorobutadiene
Hexachloroethane
Nitrobenzene
Pentachlorophenol
Pyridine
2,4,5-Trichlorophenol
2,4,6-Trichlorophenol
1,4-Dichlorobenzene
Herbicides/Pesticides
2,4,-D
2,4,5-TP (Silvex)
Ghlordane
Endrin
Heptachlor
Heptachlor epoxide
Lindane
Methoxychlor
Toxaphene •
200
200
200
0.13
0.13
0.5
3
2
100
5
400
2
7.5
10
1
0.03
0.02
0.008
0.008
0.44
10
0.5
4.2.4 40 CFR 403; General Pretreatment
Regulations for Existing and New
Sources of Pollution
Discharges to the sanitary sewer are subject to EPA's
National Pretreatment Standards and any additional
pretreatment requirements mandated by the state or
wastewater treatment facility. Examples of arsenic limits
from seven states reviewed in a recent USEPA publication
(Science Applications International 2000) range from 0.051
mg/L for Albuquerque, New Mexico to 1.07.mg/L for
Farmington, New Mexico. Residual arsenic levels in this
range were attained through precipitation or adsorption
treatments for all wastewaters examined in this work except
Ion Ex (B). The requirements imposed on a wastewater
treatment facility through a permit and/or local ordinance are
necessary to enable the facility to achieve compliance with
their NPDES permit.
Pretreatment required prior to discharge liquid residuals into
the environment is typically site-specific. Several states
have a surface water quality arsenic standard of 0.05 mg/L
for waters used as public water supplies (Science
Applications International 2000).
47
-------
4.2.5 40 CFR 503: Standards for the Use or
Disposal of Sewage Sludge
This regulation describes comprehensive criteria for the
management of biosolids. Under 40 CFR 503, biosolids are
either land applied in bulk form, sold, or given away.
Application can occur on either agricultural land, forests,
public contact sites, and reclamation sites or on lawns and
home gardens. In order for biosolids to be land applied,
criteria for pollutant limits, pathogens, and vector attraction
reduction must be met. The Part 503 pollutant limits for land
application are given in Table 4-6. All biosolids that are to be
land applied must meet the ceiling concentrations in Table 1
of 503.13. Bulk biosolids that are applied to agricultural land,
forest, public contract sites, or reclamation sites must also
either meet the pollutant limits in Table 3 of 503.13 or be
applied at rates so that the cumulative loading rates in Table
2 of 503.13 are not exceeded. Bulk biosolids that are applied
to lawn or home gardens must meet the pollutant limits in
Table 3 of 503.13. Biosolids that are sold or given away
must either meet the pollutant limits in Table 3 of 503.13, or
be applied so as not to exceed the annual pollutant rates in
Table 4 of 503.13, while still meeting the ceiling
concentrations in Table 1 of 503.13.
4.2.6 Comprehensive Environmental
Response Compensation Liability Act
(CERCLA)
The CERCLA, also known as the Superfund Act, was
established to deal with the numerous existing abandoned or
uncontrolled hazardous waste disposal sites that pose a real
threat to public health and safety as well as to the
environment. Prior to the act's passage, USEPA was only
authorized to regulate hazardous waste management at
active and properly closed sites. The Superfund, which is
essentially a pool of money derived from special taxes, forms
the core of CERCLA. Establishment of this fund fulfilled the
primary focus of CERCLA. An expansion of the Superfund
pool that serves to continue cleanup efforts begun under
CERCLA is provided by the Superfund Amendments and
Reauthorization Act (SARA) of 1986. The funds thereof are
used to remediate contaminated sites in accord with RCRA
requirements.
The USEPA is authorized under CERCLA to take necessary
short-term actions to deal with sites posing some immediate
threat to human health or the environment as well as to
implement long-term plans to clean up complex sites, which
are selected on the basis of risk assessments. The
identification of responsible parties is an important part of the
remediation process. Possibly the most noteworthy aspect
of these regulations, however, is that they employ a volume
use basis in assessing cleanup costs, which could potentially
place the liability with a utility whose sludge did not cause the
problem.
4.2.7 Hazardous Materials Transportation Act
(HMTA)
The Hazardous Materials Transportation Act (HMTA) applies
to all beneficial uses requiring transportation of sludge. The
WTP sludge must be determined to be non-hazardous by
RCRA and HMTA in order to transport the material. The
Table 4-6. Part 503 pollutant limits for sewage sludge land application
Table 2 of 503.13 Table 3 of 503.13
Table 1 of 503.13 Cumulative pollutant Pollutant
Ceiling concentrations loading rates concentrations
(mg/kg) (kg/ha) (mg/kg)
Table 4 of 503.13
Annual pollutant
loading rates
(kg/ha/yr)
Arsenic
Cadmium
Copper
Lead
Mercury
Molybdenum
Nickel
Selenium
Zinc
75
85
4,300
840
57
75
420
100
7,500
41
39
1,500
300
17
420
100
2,800
41
39
1,500
300
17
420
100
2,800
2.0
1.9
75
15
0.85
21
5.0
140
48
-------
HMTA also outlines U.S. Department of Transportation
(USDOT) packaging requirements.
4.3 Residuals Disposal Options
The effective removal of arsenic from WTP liquid residuals
streams results in a supernatant or effluent streams that may
meet regulatory criteria for reuse, stream discharge, orsewer
disposal and a sludge or media waste that contains a
concentrated amount of total arsenic. As discussed in the
Federal regulatory review, final land disposal of solid
residuals is dependent on the TCLP arsenic leaching (mg/L)
and total arsenic concentration (mg/kg), as well as other
TCLP or non-metal contaminants regulated by EPA.
Although only a limited amount of sludge solids from
precipitation tests were TCLP tested to determine arsenic
leaching, all samples tested had TCLP arsenic
concentrations well below the 5 mg/L limit. The TCLP
arsenic concentrations of the adsorption media tested were
also significantly lower than the 5 mg/L maximum limit for
arsenic. Based on TCLP arsenic results, these waste
samples would be considered nonhazardous (unless other
contaminants exist that would fail the TCLP analysis).
If a waste material is found to exceed the TCLP arsenic
concentration of 5 mg/L, the liquid or solid material would be
considered hazardous and would require disposal in
hazardous waste handling facilities. If the material is
determined to be nonhazardous, the following disposal
options may apply for liquid or solid media wastes:
Liquid/Semi-Liquid Wastes
> Stream discharge (NPDES permit probably
requires solids removal)
*• Sewer disposal to WWTP
>• Land application
«• MSWLF landfilling (requires dewatering)
Solid Media
*• Land application
»• Landfilling
>• Regeneration/Reuse
Each of these disposal options are summarized in the
following sections. It should be noted that landfill disposal,
sewer disposal, land application, and stream discharge
regulations vary from state to state. Some states have
adopted the Federal regulations for these disposal
applications, while others have developed their own specific
guidelines for disposal.
4.3.1 Liquid or Semi-Liquid Waste Disposal
Stream Discharge
Discharge of WTP residuals to surface water requires a
National Pollutant Discharge Elimination System (NPDES)
permit. NPDES permit requirements are based on stream
flow conditions and provide maximum limits for solids
discharge and'contaminant loadings. The limits established
in the NPDES for specific contaminants are determined by
the water quality criteria established for the receiving water,
ambient levels of the specific contaminants, the established
low flow condition of the receiving water, and the design flow
of the proposed discharge from the arsenic treatment
process (Chwirka 1999). Table 3-15 shows treatments
successful in reducing arsenic levels to 0.05 mg/L or lower,
which is the existing in-stream standard in some states. As
shown, one or more treatment techniques were able to attain
arsenic concentrations of 0.05 mg/L or lower in all residuals
except the ion exchange and activated alumina regenerant
streams.
Sewer Disposal
The quality of WTP residuals allowable for discharge to the
sanitary sewer is dependent on limits imposed by the
wastewater treatment plant receiving the liquid waste. Each
WWTP has an Industrial Pretreatment Program to prevent
unacceptable concentrations of contaminants from entering
the WWTP treatment process. Those guidelines protect the
operation of the WWTP from inhibition of the biological
processes used to treat municipal wastewater, prevent
violations of the WWTP NPDES permit, and prevent
unacceptable accumulation of contaminants in the WWTP
biosolids. The Industrial Pretreatment Program establishes
Technically Based Local Limits (TBLL). The TBLL for
arsenic will typically be limited by contamination of the
wastewater treatment plant biosolids rather than discharge
limitations or process inhibitions (Chwirka 1999).
Land Application
Land application of WTP residuals is dependent on the state
regulatory guidelines. Some states do not allow land
application of WTP residuals. The general criteria for
allowing WTP residuals to be land applied are based on the
following Federal regulations:
EPA CFR 40 261 - TCLP Hazardous Determination
EPA CFR 40 503 - Biosolids Metals Concentrations
EPA CFR 40 257 - Solid Waste Disposal
49
-------
If WTP residuals meet the criteria established by these
Federal regulations, as well as any state or local regulations,
then the material would be allowed for land application. EPA
503 established maximum loading limits for heavy metals
including arsenic. A "clean sludge" limit of 41 mg/kg was
established by EPA 503 for biosolids disposal. Clean sludge
can be land applied with no limitations (Chwirka 1999). A
cumulative arsenic loading limit to soils was set by EPA in
the Part 503 regulations at 36.6 Ibs/acre (41 kg/ha).
Landfill Disposal (MSWLF)
Municipal solid waste landfills have established a set of
disposal guidelines that are similar for most landfill agencies.
The basic guidelines for disposal include the following:
* No free liquids (pass paint filter test)
TCLP nonhazardous (EPA CFR 40 Part 261)
• Non-corrosive, non-reactive, non-ignitable (EPA
261)
Liquid or semi-liquid WTP residuals would require
mechanical or nonmechanical dewatering prior to
acceptance. If the WTP residuals exceeds the TCLP limits
established by EPA 40 CFR 261, then the material would
have to be disposed of in a hazardous waste landfill.
4.3.2 Solid Media Disposal
Land Application
The same regulatory requirements used for sludge disposal
would apply to disposal of adsorption medias. If the material
is determined to be nonhazardous (TCLP limits from EPA 40
CFR 261) and meets the EPA 503 metals limits, then land
application is an option. The ability of the solid media to
blend into the natural soil environment must also be
considered prior to land disposal. Iron-based media may
provide an iron amendment to soils, however, aluminum-
based media and ion exchange resins would most likely not
provide a benefit to soils. Also, under reduced pH
conditions, Fe(lll) could be reduced to Fe(ll), and arsenic
bound to iron complexes could be released to surrounding
soils.
Landfill Disposal
The same criteria discussed for landfilling WTP sludge would
apply to disposal of solid adsorption media. TCLP hazard
evaluation, no free liquids, and determination of corrosivity,
ignitabiiity, and reactivity are each required prior to
acceptance. All solid media samples in this work met the
current TCLP arsenic limit of 5.0 mg/L.
Recycling/Reuse
It is possible that adsorption media may be regenerated by
the manufacturer and reused for similar or different
applications. To determine reuse potential for a specific solid
adsorption media, the manufacturer of the media should be
contacted.
50
-------
5. Summary and Conclusions
5.1 Summary
5.1.1 Project Description
Liquid and semi-liquid residuals streams were collected from
eight operating full-scale treatment plants around the U.S. for
evaluation of several treatment approaches to remove
arsenic.. Spent media samples collected from a ninth plant
were used to generate another liquid stream for testing.
Precipitation processes and adsorption/exchange
technologies that have been demonstrated to be successful
in removing arsenic from potable water treatment plants were
evaluated for their effectiveness in removing arsenic from the
following types of liquid and semi-liquid residuals streams.
Activated alumina regenerant
Ion exchange regenerant
Nanofiltration concentrate
Reverse osmosis concentrate
Spent filter backwash water and spent filter
backwash water/adsorption clarifier flush blend from
Fe/Mn removal plants
Treatment effectiveness was compared based on reductions
in arsenic and residual concentrations of arsenic following
treatment. Residual iron and aluminum levels were also
considered. When sufficient quantities of sludge were
generated in precipitation tests, TCLP analyses were
conducted on the sludge fraction of the samples. Also, TCLP
analyses were conducted on three types of solid media
samples: (1) filter media from an Fe/Mn removal plant, (2)
spent activated alumina, and (3) an ion exchange resin.
5.1.2 Untreated Residuals Sample
Characterization
Untreated liquid residuals streams were characterized
according to the following parameters: total and dissolved
arsenic, total iron and manganese, pH and alkalinity, total
dissolved solids and conductivity, total hardness, and sulfate.
Untreated residuals arsenic concentrations were determined
to assess arsenic removal, and they varied widely from about
0.5 mg/L (spiked) to 1.7 mg/L for all samples except ion
exchange and activated alumina regenerants (Figure 3-1).
As levels in composite ion exchange regenerant samples
were approximately 11 mg/L and 25 mg/L and the activated
alumina regenerant sample contained 2.6 mg/L arsenic.
Other characteristics including pH, sulfate, and TDS were
used to select appropriate treatment options, and were also
important in interpretation of treatment results.
Alkalinity and pH ranged from 197 mg/L to 7,000 mg/L as
CaCO3 and from 6.6 to 9.7, respectively. Both parameters
were highest in ion exchange regenerant wastes. Total
hardness, on the other hand, was comparatively low (less
than 100 mg/L) in ion exchange regenerants and was highest
at around 1,600 mg/L (as CaCO3) in the nanofiltration
concentrates. TDS and conductivity exceeded 10,000 mg/L
and 20,000 uS/cm in the AA regenerant and RO
concentrates. As expected, total iron and manganese levels
were highest in spent filter backwash water and spent filter
backwash water/adsorption clarifier flush blend samples (up
to 78.5 mg/L and 7.5 mg/L), while sulfate levels were lowest
(less than 100 mg/L for SFBW and SFBW/ACF blend
compared to greater than 500 mg/L for all other residuals).
5.1.3 Precipitation and Adsorption Test
Results
Table 5-1 summarizes arsenic results from treatment of all
nine residuals samples. The table shows the minimum total
arsenic concentration remaining in the supernatant or effluent
following treatment. Unless noted otherwise, adsorption test
results are for samples collected at 120 bed volumes during
tests where the empty bed contact time (EBCT) was 3 min.
Precipitation results show the total arsenic concentration
remaining in the supernatant following precipitation using the
best coagulant dose and pH combinations for each residuals
sample.
As shown in Table 5-1, for all residuals samples, precipitation
using ferric chloride was more effective for removing arsenic
than precipitation using alum for the range of test conditions
evaluated. Similarly, the iron-based media produced the
51
-------
Table 5-1. Comparison of treatment processes for arsenic removal
Treated water arsenic remaining (mg/L)
Sample ID
AA regenerant*
Ion Ex (A)
Ion Ex (B)
RO(A)
RO(B)
NF(A)
NF(B)
SFBW (A) (unsettled)
SFBW/ACF (B) (unsettled)
(settled)
Residuals
arsenic
cone.
(mg/L)
2.63
10.5
24.8
0.526
0.663
0.523
0.486
1.41
1.74
0.043
Precipitation
Alum
—
5.98
22.8
0.526
0.286
0.029
0.035
0.021
0.194
—
FeCI3
0.154
1.28
18.7
0.041
0.078
0.009
0.005
0.013
0.064
0.011
Fe-based
media
—
3.60*
22.3
0.252*
0.018
0.021
<0.002
1.18*
—
<0.002
Adsorption
(3 min EBCT, 120 BV)
Activated
alumina
media
—
—
24.0
0.526
0.071
0.034
0.004
1.41* '
—
<0.002
Ion
exchange
resin
—
—
—
—
—
0.535
0.438
—
—
<0.002
Modified
alumina
media
—
—
—
—
— -
0.376
0.452
'
—
—
"Arsenic concentration measured after 120 BV using an EBCT of 1.5 mjn.
lowest effluent arsenic concentrations of the
adsorption/exchange media evaluated for all of the various
liquid residuals streams tested. Further, FeCI3 precipitation
reduced arsenic levels to 0.05 mg/L (in-stream standard for
arsenic for a number of states) or lower in five of the nine
samples tested, and below 0.1 mg/L in six of the nine
residuals samples.
Figures 5-1 and 5-2 show total arsenic concentrations
remaining along with corresponding percent removal for the
best ferric chloride and alum precipitation conditions tested
for each residuals sample. While percent removals indicate
the potential of precipitation to remove arsenic from each
specific residuals stream tested, total arsenic concentrations
remaining in the supernatant (or liquid-fraction) following
treatment are also very important with regard to disposal
options available. As Chwirka (1999) described, the disposal
of liquid residuals containing arsenic to receiving waters will
be subject to compliance with National Pollution Discharge
Elimination System (NPDES) limits, which are determined by
water quality criteria established for the receiving water,
ambient levels of the specific contaminants, the established
low flow condition of the receiving water, and the design flow
of the proposed discharge. Chwirka (1999) also notes that
discharge of arsenic-containing residuals to a sanitary sewer
(the other option for discharge of liquid residuals) is subject
to the established Technically Based Local Limits (TBLL) of
the current Industrial Pretreatment Program, and that the
TBLL for arsenic will typically be limited by the contamination
of the wastewater treatment plant biosolids as opposed to
discharge limitations or process inhibition.
Arsenic removals attained in this work demonstrated that
treatments shown to be effective at removing arsenic from
source waters with relatively low arsenic concentrations were
also successful in removing arsenic from residuals streams
generated from arsenic removal processes. These residuals
streams, of course, contained much higher levels of arsenic
than the corresponding source water (from 12 to 270 times
more arsenic in this study). Similar data covering such a
broad range of liquid residuals streams have not been
previously reported in the literature. Also, these data can be
compared to achievable removal levels reported in the
literature such as 95 percent for coagulation/filtration
(USEPA1999b).
Results of the TCLP analysis are key in dictating disposal
options for solid wastes. TCLP arsenic levels determined for
semi-liquid residuals generated in precipitation tests were all
below the current threshold limit of 5.0 mg/L. TCLP arsenic
levels in media samples from arsenic removal plants were
also well below that limit (Table 5-2).
52
-------
1
3
f 0.8
O)
c
]c
'(D
S> 0.6
c
o
s
£
g
g 0.4
u
_o
ra
75 0.2
i°
200 mg/L
pH 6.2
(As conc./20)
-
_
-
B>o =i 5
6 l~ C « 0
— ^f 0 X Q (O O
-I i CICM10CLOX5
% = oj ^755" a
-'g,^ «I "„ ^^ |§ S ^ KX< ^^
^ ^ p.yv/1 o "^ (D !? ~Q_ OO1 sxO^ rOv
g jL^ (vQJ § -j. § % pvsd yv xx^ oX
rxx
rS88
>^\^
5o<
>vC
>oo
KX>
J^^Cy
>QO
1
Xx
^
Vx
ggggggrogg
lfcz£§§f2,Ji
fe | * I I
m
OT
2
CO
1
100
80
60
40
20
m
LU
en
O>
£
ffi
Sample ID Sample ID
Figure 5-1. Total arsenic concentrations remaining in the supernatant and percent reduction
after ferric chloride precipitation
0.8
£ 0.6
5
§
0.4
s
I
0.2
75 mg/L
pH 7.3
50 mg/L
pH 7.4
200 mg/L
pH 6.1
(As cone.725)
100 mg/L
pH 5.5
(As cono./10)
100 mg/L
pH 7.7
200 mg/L
pH 6.6
200 mg/L
pH 6.6
IWSI
50 mg/L
pH 7.0
100
80
i 60
73
40
20
/
Sample ID Sample ID
Figure 5-2. Total arsenic concentrations remaining in the supernatant and percent
reduction after alum precipitation
53
-------
Summaries of results from adsorption and ion exchange tests
are presented for the iron-based media, activated alumina,
ion exchange resin, and modified alumina in Figures 5-3
through 5-6. Results are shown both in terms of total arsenic
levels detected in adsorption or ion-exchange column
effluents and arsenic percent reduction. Figure 5-3 shows
that the iron-based media was very effective at removing
arsenic for the settled SFBW/ACF blend sample (settled
first), the NF concentrate samples, and one of the RO
concentrate samples overthe total testduration during which
120 bed volumes of water were passed through the column.
Arsenic breakthrough occurred very rapidly (at or before 60
bed volumes) for the ion exchange samples, and for one of
the RO concentrates. Overall, the trends were similar for the
activated alumina tests, although the iron-based media was
more effective than the activated alumina (Figure 5-4).
Figure 5-5 shows that ion exchange was effective only for the
settled SFBW/ACF blend that had a very low arsenic
concentration (0.043 mg/L). IDS and sulfate levels were too
high (>500 mg/L and >250 mg/L) in the NF concentrate for
ion exchange to be effective. Breakthrough also occurred
very rapidly for the modified alumina tests conducted on the
NF concentrate samples (see Figure 5-6).
Table 5-2.
Solid waste ID
TCLP arsenic from solid fraction residuals
TCLP arsenic
concentration
(mg/L)
Spent activated alumina
(full-scale WTP)
Spent Fe-Mn filter media
(full-scale WTP)
Spent anion exchange resin
(bench-scale SFBWtest)
Spent anion exchange resin (bench-
scale nanofiltration concentrate test)
0.010
0.004
0.023
0.203
5.2 Conclusions
This work focused on evaluation of liquid, semi-liquid, and
solid waste streams from five arsenic removal plants and four
membrane plants across the U.S. for removal of arsenic by
precipitation and adsorption. TCLP tests were conducted on
the solid wastes and semi-liquid residuals generated in
precipitation tests. Precipitation and adsorption
investigations were not intended to identify optimal treatment
conditions due to the small quantities of residuals shipped for
testing, but rather to screen treatment options for arsenic
removal capability. Based on the findings, the major
conclusions from this work follow.
5.2.1 Precipitation
For the array of types of residuals samples tested,
precipitation using ferric chloride yielded greater reductions
in arsenic than precipitation using alum. Required dosages,
iron-to-arsenic molar ratios, and pH to achieve the best
arsenic removal varied depending on the residuals stream.
Pertinent findings that stemmed from precipitation work are
as follows:
Ferric chloride outperformed alum for removal of
arsenic from residuals by sedimentation.
The required molar iron-to-arsenic ratio for best
removal of arsenic in these screening tests varied
widely from 4:1 to 191:1, depending on residual type.
Arsenic removals achieved were greater than 88
percent for all but one of the waste streams (Ion Ex
B).
• Generally, polymer addition did not significantly
improve arsenic removal.
Best performance with ferric chloride precipitation
typically occurred between pH 5.0 and 6.2.
Supernatant residual total arsenic levels after ferric
precipitation were between 0.005 mg/L and 0.078
mg/L for all waste streams, except ion exchange and
activated alumina, compared to an in-stream arsenic
limit of 0.05 mg/L in some states. For Ion Ex (A), Ion
Ex (B), and the AA regenerant, those levels were
1.28 mg/L, 18.7 mg/L, and 0.154 mg/L, respectively.
TCLP arsenic concentrations in semi-liquid residuals
generated in ferric precipitation tests were between
<0.002 mg/L and 0.018 mg/L. These values are well
below the current TCLP arsenic limit of 5 mg/L.
• Alkalinity likely inhibited arsenic removal.
Based on these findings, ion exchange regenerant
wastes might be the most difficult to treat for meeting
in-stream standards.
5.2.2 Adsorption
The iron-based adsorption media was the most effective of
the media and resins tested for removing arsenic from the
liquid residuals evaluated in this work. Adsorption tests
demonstrated the potential for different types of media and
resins to remove arsenic fronrvarious residuals streams, but
did not assess media capacity for arsenic adsorption
because tests were not run to exhaustion. Specific findings
from adsorption tests are as follows:
The iron-based media evaluated in adsorption
testing typically outperformed the aluminum-based
media and ion exchange media for removal of
arsenic.
54
-------
SFBW/ACF (B)
RO (A)
-+-
RO (B)
NF (A)
-5*8-
NF(B)
-^-
Ion Ex (A)
-H-
lon Ex (B)
20 40 60 80 100 120 140
No. of bed volumes treated
120
100
a so
1
o 60
'g
S2
S 40
20
SFBW/ACF (B)
H*-
RO (A)
RO (B)
NF (A)
-96-
NF (B)
-e-
lon Ex (A)
•B-
Ion Ex (B)
20 40 60 80 100 1"20 140
No. of bed volumes treated
Figure 5-3. Total arsenic concentrations in the column effluent and percent reduction after
iron-based media adsorption using a 3 min EBCT
SFBW/ACF (B)
-**-
RO (A)
-*-
RO (B)
HB-
NF (A)
-^K-
NF (B)
-e-
lon Ex (B)
-B-
120
20 40 60 80 100 120 140 ° 20 40 60 80 10° 12° 14°
No. of bed volumes treated No. of bed volumes treated
Figure 5-4. Total arsenic concentration in the column effluent and percent reduction after
activated alumina adsorption using a 3 min EBCT
55
-------
0.8
0.6
0.4
0.2
SFBW/ACF (B) 12°
20 40 60 80 100 120 140
No. of bed volumes treated
0 20 40 60 80 100 120 140
No. of bed volumes treated
Figure 5-5. Total arsenic concentration in the column effluent and percent reduction after ion
exchange using a 3 min EBCT
0.8
0.6
ra 0.4
0.2
NF (A)
-**-
NF (B)
120
100
& 80
60
ra
~ 40
s
20
NF (A)
-*-
NF (B)
20
140
20
40 60 80 100 120
No. of bed volumes treated
140
40 60 80 100 120
No. of bed volumes treated
Figure 5-6. Total arsenic concentration in the column effluent and percent reduction after modified
alumina media adsorption using a 3 min EBCT
56
-------
Activated alumina and the iron-based media
provided comparable arsenic removal for the NF
concentrates and the settled SFBW/ACF.
Arsenic removals attained by adsorption using the
iron-based media were 77 percent for RO (A) and
close to 100 percent for RO (B), NF (A), NF (B), and
SFBW/ACF (B) up to 120 bed volumes using an
empty bed contact time of 3 min. Poorer
performance resulted with the ion exchange
regenerant streams, where corresponding arsenic
reductions were 10 and 63.7 percent.
Column effluent total arsenic concentrations below
0.030 mg/L were attained in adsorption tests
conducted for the two NF concentrates, RO (B), and
SFBW/ACF (B).
In order to assess ultimate capacity of adsorption
medias/exchange resins for removal of arsenic, tests
should be run to exhaustion in future work where
possible. Isotherm tests would also be instructive.
Alkalinity may have inhibited arsenic removal.
As for precipitation, ion exchange regenerant may
be the most difficult waste to treat for meeting an in-
stream arsenic standard.
5.2.3 So//cfe
The recent reduction in the drinking water arsenic MCL from
0.05 mg/L to 0.01 mg/L could be followed by a comparable
reduction in the arsenic TCLP limit from 5.0 mg/L to 1.0
mg/L. TCLP arsenic concentrations reported in the literature
indicate that solids from existing arsenic removal facilities
can meet the current limits of 5.0 mg/L. While TCLP arsenic
levels for solid media samples and thickened residuals
samples in this work were all well below 5.0 mg/L, some
facilities could have difficulty in meeting either the current or
some reduced limit upon making treatment process
modifications to remove more arsenic. The following findings
regarding solids stemmed from this work:
All TCLP As concentrations for solid media samples
were well below the TCLP threshold limit of 5.0 mg/L
(0.004 mg/L to 0.203 mg/L). Therefore, disposal of
the solid medias would not be TCLP-limited based
on arsenic.
• Total production of sludge after coagulant addition to
treat residuals for arsenic removal was in the range
of 35 to 230 dry Ib sludge/MG raw water treated, and
the relative order from least to greatest is ion
exchange, SFBW, nanofiltration, reverse osmosis.
5.3 Recommendations for Future Work
Additional work could serve to build on the findings from this
research. Some recommended areas of focus for future
work are listed below:
Determining optimal treatment conditions (chemical
type, dosage, and coagulation pH) in precipitation
tests for activated alumina and for ion exchange
regenerants and other residuals streams with high
alkalinity.
• Defining the role of alkalinity as a possible
interference in arsenic removal (in precipitation and
absorption tests).
Assessing ultimate capacity for arsenic removal in
adsorption tests run to exhaustion.
• Preparing isotherms to define arsenic removal.
• Determining arsenic speciation in residuals samples
and the impact of speciation on removal of arsenic
from residuals.
Investigating the relationship between the TCLP and
California WET test and assessing disposal
implications for arsenic-laden residuals.
57
-------
-------
6. References
Bartley, C.B., P.M. Colucci, T. Stevens. 1991. The Inorganic
Chemical Characteristics of Water Treatment Plant
Residuals. Cooperative Agreement CR-814538-01-
0. Cincinnati, Ohio:USEPA.
Brandhuber, Philip and C. Amy. 2000. Identification of Key
Engineering Parameters Influencing the Treatment
of Arsenic in Drinking Water Via Membrane
Technology. In Proc. of 2000 Inorganic
Contaminants Workshop. Denver, Colo.:AWWA.
Cheng, Robert C., et al. 1994. Enhanced Coagulation for
Arsenic Removal. JAWWA, 86:9:79.
Chwirka, J. 1999. Residuals Generation, Handling and
Disposal. In Arsenic Treatment Options and
Residuals Handling Issues. Draft Final Report.
AWWA, Denver.
Clifford, Dennis. 1999. Ion Exchange and Inorganic
Adsorption. In Water Quality and Treatment. Edited
by Raymond Letterman. New York: McGraw-Hill, Inc.
Clifford, Dennis and C.C. Lin. 1986. Arsenic Removal From
Groundwater in Hanford, California - A Summary
Report. Houston, Texas:University of Houston.
Clifford, Dennis, et al. 1997. Final Report: Phases 1 and 2.
City of Albuquerque, New Mexico using the
University of Houston/USEPA mobile drinking water
treatment research facility. Houston,
Texas:University of Houston.
Clifford, Dennis, G. Ghurye, and A. Tripp. 1998. Arsenic Ion
Exchange Process with Reuse of Spent Brine. In
Proc. of 1998 Annual AWWA Conference. Denver,
Colo.: AWWA.
Clifford, Dennis, G. Ghurye. 1999. Development of an
Arsenic Ion Exchange Process with Direct Reuse of
Spent Brine. JAWWA. Forthcoming.
Clifford, Dennis, and C.C. Lin. 1991. Arsenic (ill) and
Arsenic (V) Removal from Drinking Water in San
Ysidro, New Mexico. Cincinnati OH, USEPA.
Cornwell, David. 1999. Water Treatment Plant Residuals
Management. In Water Quality and Treatment.
Edited by Raymond Letterman. New York:McGraw-
Hill, Inc.
Cornwell, David A., et al. 1992. Landfilling of Water
Treatment Plant Coagulant Sludges. Denver,
CO:AWWARF.
Dixon, K.L., R.G. Lee, and R.H. Moser. 1988. Water
Treatment Plant Residuals: A Management Strategy
for the Pennsylvania Region. Vorhees,
NJ:AWWSCo.
Edwards, Marc. 1994. Chemistry of Arsenic Removal
During Coagulation and Fe-Mn Oxidation. JAWWA,
86:9:64.
Fields, Keith, A. Chen, and L. Wang. 2000. Arsenic
Removal from Drinking Water by Iron Removal
Plants. EPA/600/R-00-086. Cincinnati, OH.
Fields, Keith, T. Sorg, A. Chen, and L. Wang. 2000. Long-
Term Evaluation of Arsenic Removal in Conventional
Water Treatment Systems. In Proc. of 2000
Inorganics Contaminants Workshop. Denver,
Colo.: AWWA.
Frey, Michelle and M. Edwards. 1997. Surveying Arsenic
Occurrence. JAWWA, 89:2:107.
Ghurye, Ganesh L., D. Clifford, et al. 1999. Combined
Arsenic and Nitrate Removal by Ion Exchange.
JAWWA, 91:10:85.
59
-------
Ghurye, Ganesh L., D. Clifford, and A. Tripp. 1999.
Combined Arsenic and Nitrate Removal by Ion
Exchange. JAWWA, 91(10):85-96.
Hathaway, Steven W. (deceased) and Frederick Rubel, Jr.
1987. Removing Arsenic from Drinking Water.
JAWWA, 78:8:61.
Hering, Janet G., et a/. 1996. Arsenic Removal by Ferric
Chloride. JAWWA, 88:4:155.
McNeill, Laurie S., Marc Edwards. 1995. Soluble Arsenic
Removal at Water Treatment Plants. JAWWA,
87:4:105.
Mickley, M., R. Hamilton, L. Gallegos, and J. Truesdall.
1993. Membrane Concentrate Disposal. Denver,
Colo.:AWWA.
Science Applications International Corporation. 2000.
Regulations on the Disposal of Arsenic Residuals
from Drinking Water Treatment Plants. EPA/600/R-
00-025. Cincinnati, OH.
Simms, John and F. Azizian. 1997. Pilot Plant Trials on
Removal of Arsenic from Potable Water Using
Activated Alumina. In Proc. of Annual AWWA Water
Quality Technology Conference.
Simms, John, J. Upton, and J. Barnes. 2000. Arsenic
Removal Studies and the Design of a 20,000 m3/d
Plant in the UK. In Proc. of 2000 Inorganic
Contaminants Workshop. Denver, Colo.:AWWA.
USEPA. 1996. Removal of Arsenic from Drinking Water
Treatment Technology. Tom Sorg presented at the
Stakeholders Meeting on Arsenic in Drinking Water,
June 2, 1999. Wynne Miller. 1999. Development of
the Practical Quantitation Limit (PQL) for Arsenic.
Prepared for the June 2-3, 1999 Stakeholders
Meeting on Arsenic in Drinking Water. Washington,
D.C.
Wang, Lili, A. Chen, and K. Fields. 2000. Arsenic Removal
from Drinking Water by Ion Exchange and Activated
Alumina Plants. EPA/600/R-00-088. Cincinnati,
OH.
Wang, Lili, T. Sorg, and A. Chen. 2000. Arsenic Removal by
Full Scale Ion Exchange and Activated Alumina
Treatment Systems. In Proc. of 2000 Inorganic
Contaminants Workshop. Denver, Colo.:AWWA.
60
-------
Appendix A. Raw Characterization, Precipitation, and
Adsorption Data and Precipitation Figures
61
-------
Table A-1. Untreated residuals characterization data: General water quality parameters
Analysis
date
5/16/99
5/16/99
5/16/99
7/19/99
7/19/99
7/26/99
7/26/99
6/29/99
6/29/99
7/16/99
7/16/99
01/28/00
7/12/99
8/4/99
9/10/99
10/01/99
9/10/99
10/01/99
9/10/99
9/13/99
9/27/99
10/19/9
10/12/9
10/20/9
02/10/00
10/14/9
11/11/9
11/15/9
02/09/00
02/09/00
02/09/OC
02/09/OC
02/09/0
02/10/OC
EE&T Sample
ID No.
45AS-FJS-RG1
45AS-FJS-RG2
45AS-FJS-RG3
200AS-FJS
200AS-FJS
200AS-FJS
200AS-FJS
169AS-VOM
169AS-VOM
187AS-NN
87AS-NN (sp III)
020As-NN(spV)
190AS-CRO
90AS-CRO (spk)
243AS-IA
243AS-IA
243AS-IA
243AS-IA
243AS-IA
243AS-IA
243AS-IA
243AS-IA
281AS-FM
281AS-FM (spk)
041AS-FM (spk
285AS-AA1
313AS-PC
313AS-PC (spk
039AS-JL
039AS-JL
039AS-JL
039AS-JL
039AS-JL
039AS-JL
Source
name
;rank Jewitt School
rrank Jewitt School
:rank Jewitt School
:rank Jewitt School
:rank Jewitt School
:rank Jewitt School
:rank Jewitt School
Village of Morton
Village of Morton
Newport News, Va
Newport News, Va
viewport News, Va
Chesapeake, Va
Chesapeake, Va
Indiana American
Indiana American
Indiana American
Indiana American
Indiana American
Indiana American
Indiana American
Indiana American
Fort Myers, FL
Fort Myers, FL
Fort Myers, FL
Act. Alum. - Battel
Palm Coast, FL
Palm Coast, FL
Jerry Lowry
Jerry Lowry
Jerry Lowry
Jerry Lowry
Jerry Lowry
Jerry Lowry
Sample
description
Backwash water
Brine
Rinse
Backwash water no. 2
Brine no. 2
Rinse no. 2
Composite
Backwash water
Supernatant
RO Concentrate
*O Cone spiked (As III)
RO Cone spiked (As V)
RO Concentrate
RO Cone spiked w/ As\
Backwash water
Settled BW
Clarifier flush
Settled CF
BW/ Clarifier (50:50)
Settled 1 (50:50 blend)
Settled 2 (50:50 blend)
Settled 3 (50:50 blend]
Nanofiltration
Nanofilt. spiked w/ As >
Nanofilt. spiked w/ As1
AA Regenerant
Nanofiltration
Nanofilt. spiked w/ As'
Backwash
Brine
Brine (Jug 8 - not mixe
Brine (Jug 9 - not mixe
Rinse
Composite
Sample quality characterization
PH
7.21
8.88 ,
8.48
7.34
8.97
8.43
9.00
7.60 .
-
7.90
7.90 .
8.03
7.30
7.30
8.12
8.12
8.12
8.12
8.12
8.12
8.12
8.12
6.91
7.06
6.90
7.13
6.57
6.59
7.90
9.80
9.80
9.80
9.80
9.70
Alkalinity
(mg/L)
71.5
>5000
300
97
2900
400
950
430
•-
2800
2800
-
600
600
186
-
202
-
197
-
-
.-.
360
325
-
268
240
210
95
9,800
-
-
4,000
7,000
Hardness
(mg/L)
68
260
70
78 .
50
80
90
365
460
460
-
840
840
500
- '
510
- ,
400
-
1700
1560
-
13
1 ,550
1,750
108
80
-
84
86
Conduct
(uS/cm)
250
42,570
15,450
330
26,900
2,500
8,100
900
28,500
28,500
23,800
23,800
670
710
-
680
• -
2,830
3,515
22,640
3,050
3,080
237
15,100
11,460
12,440
.IDS
,(mg/L)
130
18,660
8,250
170
11,600
1,250
4,100
460
14;300
14,300
11,750
11,750
323
358
. -
341
.-••;•
,-
1,418
1 ,765
10,240
1,523
1,533
118
7,550
5,740
6,240
Sulfate
(mg/L)
• • •-„ ,
,, . .'.-.. .. J
•|'?N( «t
•*
-
1075
16338
1220
_
yiu
62
-------
Table A-2. Untreated residuals characterization data: Metals
Analysi
date
5/16/99
5/16/99
5/16/99
7/19/99
7/19/99
7/26/99
7/26/99
6/29/99
6/29/99
7/16/99
7/16/99
01/28/00
7/12/99
8/4/99
9/10/99
10/01/99
9/10/99
10/01/99
9/10/99
9/13/99
9/27/99
10/19/99
10/12/99
10/20/99
02/10/00
10/14/99
11/11/99
11/15/99
02/09/00
02/09/00
02/09/00
02/09/00
02/09/00
02/10/00
EE&T Sample
ID No.
145AS-FJS-RG1
145AS-FJS-RG2
145AS-FJS-RG3
200AS-FJS
-200AS-FJS
200AS-FJS
200AS-FJS
169AS-VOM
169AS-VOM
187AS-NN
187AS-NN (sp 111
020AS-NN (sp V)
190AS-CRO
190As-CRO(spk
243AS-IA
243AS-IA
243AS-IA
243AS-IA
243AS-IA
243AS-IA
243AS-IA
243AS-IA
281AS-FM
281AS-FM (spk)
041AS-FM (spk)
285AS-AA1
313AS-PC
313AS-PC (spk)
039AS-JL
039AS-JL
039AS-JL
039AS-JL
039AS-JL
039AS-JL
Source
name
Frank Jewitt Schoo
Frank Jewitt Schoo
Frank Jewitt Schoo
Frank Jewitt Schoo
Frank Jewitt Schoo
Frank Jewitt Schoo
Frank Jewitt Schoo
Village of Morton
Village of Morton
Newport News, Va
Newport News, Va
Newport News, Va
Chesapeake, Va
Chesapeake, Va
Indiana American
Indiana American
Indiana American
Indiana American
Indiana American
Indiana American
Indiana American
Indiana American
Fort Myers, FL
Fort Myers, FL
Fort Myers, FL .
Act. Alum. - Battelle
Palm Coast, FL
Palm Coast, FL
Jerry Lowry
Jerry Lowry
Jerry Lowry
Jerry Lowry
Jerry Lowry
Jerry Lowry
Sample
description
Backwash water
Brine
Rinse
Backwash water no. 2
Brine no. 2
Rinse no. 2
Composite
Backwash water
Supernatant
RO Concentrate
RO Cone spiked (As III
RO Cone spiked (As V
RO Concentrate
RO Cone spiked w/ AsV
Backwash water
Settled BW
.Clarifier flush
Settled CF
BW/ Clarifier (50:50)
Settled 1 (50:50 blend)
Settled 2 (50:50 blend)
Settled 3 (50:50 blend)
Nanofiltration
Nanofilt. spiked w/ As V
Nanofilt. spiked w/ As V
AA Regenerant
Nanofiltration
Nanofilt. spiked w/ As V
Backwash
Brine
rine (Jug 8 - not mixed)
rine (Jug 9 - not mixed)
Rinse
Composite
Sample quality characterization
Total As
(mg/L)
0.032
37.00
1.700
0.069
33.20
1.240
10.50
1.410
0.007
< 0.002
0.526
0.546
< 0.002
0.663
1.160
0.038
2.450
0.046
1.740
0.122
0.024
0.043
0.013
0.523
0.483
2.630
Q.005
0.486
0.089
34.3
32.4
21.3
12.4
24.8
Total Fe
(mg/L)
0.088
0.894
0.282
1.780
<0.01
0.106
0.490
78.50
0.381
0.067
0.067
-
0.858
0.858
31.00
-
64.60
-
45.90
- -
0.054
.
2.620
2.160
. -
0.831
0.450
0.458
0.084
<0.010
-
-
O.010
<0.010
Total Mn
(mg/L)
< 0.005
< 0.005
0.007
0.060
< 0.005
o.oio
'
' 7.52
'..
0.232
0.232 .
- •
1.110
1.110
2.50
-
5.24
,- •
3.75
-
-,
' -
0.12
0.140
-
0.085
0.084 •
0.081
<0.005
0.006
. ,
-
<0.005
<0.005
Total Al
(mg/L)
-
.
* .
'_
.
. ' '
•
.
- .
.
••_
-
.
,
_
.
.
_
-
_
.
.
_
.
.
113.0
.
_
.-.
_
_
_
_
Disolv As
(mg/L)
0.031
17.70
1.670
0.037
31.80
1.270
10.30
< 0.002
< 0.002
< 0.002
0.501
< 0.002
0.031
0.030
0.029
0.007
0.487
0.117
0.009
0.515
0.094
25.3
14.0
24.7
63
-------
Table A-3. Precipitation test data
Wastewater
ID
SFBW (A)
SFBW (A)
SFBW (A)
SFBW (A)
SFBW (A)
SFBW (A)
SFBW (A)
SFBW (A)
SFBW (A)
SFBW (A)
RO(A)
RO(A)
RO(A)
RO(A)
RO(A)
RO(A)
RO(A)
RO(A)
RO(A)
RO(A)
RO(A)
RO(A)
RO(A)
RO(A)
Sample
description
Backwash water
Backwash water
Backwash water
Backwash water
Backwash water
Backwash v ':er
Backwash water
Backwash water
Backwash water
Backwash water
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO (A) RO concentrate
RO(A)
RO(A)
RO(A)
RO(A)
Ion Ex (A)
RO concentrate
RO concentrate
RO concentrate
RO concentrate
Backwash
Ion Ex (A) ! Rinse
Ion Ex (A)
Ion Ex (A)
Ion Ex (A)
Ion Ex (A)
Ion Ex (A)
Rinse
Composite
Composite
Composite
Composite
Treatment conditions
Alum
(mg/L)
25
50
50
50
-
-
-
-
-
-
-
. -
-
-
-
• -
-
-
-
-
-
100
100
100
150
150
-
-
-
-
-
-
-
-
100
100
Ferric
(mg/L)
-
-
-
-
25
50
50
50
50
-
25
25
50
50
100
100
100
100
100
150
150
-
•
-
-
-
• -
-
-
75
75
100
100
100
-
-
Polymer
(mg/L)
4
-
-
4
4
-
-
4
4
-
-
2
-
2
2
5
2
-
2
-
2
•
2
2
2 ,
2
-
I^aOH
Lime
'(PH)
-
-
-
-
-
- - -
-
Coag.
PH
7.60
7.40
6.00
7.40
7.30
7.07
5.00
7.10
5.C7
5.92
6.00
6.00'
6.00
6.00
7.45
7.50
6.29
6.00
6.00
6.00
6.00
7.70
8.20
6.00
8.20
6.00
6.14
10.75
10.70
6.50
. -
-
7.86
6.19
, 8.87
5.48
Treated characteristics
Total As
(mg/L)
0.074
6.048
0.096
0.021
6.034
0.022
0.056
0.013
0.631
6.231
0.494
0.519
0.304
0.364
0.368
0.388
0.094
0.091
0.097
0.047
6.041
0.526
0.773
0.698
0.730
0.644
0.575
0.483
0.570
< 0.002
0.176
0.387
2.360
1.280
9.310
5.980
Total Fe
(mg/L)
-
-
-
-
3.380
2.880
6.460
1.570
3.900
-
7.100
6.986
8.460
0.739
1.750
1.930
5.480
4.910
0.877
3.520
0.015
-
-
-
-
-
-
""
•"••
0.497
1.670
2.510
5.430
3.510
-
-
Total Al
(mg/L)
0.467
0.531
0.616
0.429
-
-
- .
-
-
-
4::'., •', ,,
. -
-
-
-
-
-
-
. -
-
-
2.760
5.270
6.860
7.490
9.200
-
..,„-,- ,
-
-
s
-
-
2.990
3.730
Arsenic
removal
percent)
94.75
96.60
93.19
98.51
97.59
98.44
96.03
99.08
97.80
83.62
9.52
4.95
44.32
33.33
30.04
26.24
82.13
83.33
82.23
91.39
92.49
6.66
. ji, i " . •'
0.00
0.00
6.60
0.00
6.00
11.54
0.00
" ••.". ;
96.88
89.65
77.24
78.60
87.81
11.33
43.05
j« pJ'UI »
i-"«•"«'I
!l ;- iijifj I
64
-------
Table A-3.
Continued
Wastewater
ID
Ion Ex (A)
Ion Ex (A)
Ion Ex (A)
Ion Ex (A)
Ion Ex (A)
Ion Ex (A)
RO (B) '
RO(B)
RO (B)
RO(B)
RO(B)
RO(B)
RO(B)
RO(B)
RO (B)
SFBW/ACF (B)
SFBW/ACF (B)
SFBW/ACF (B)
SFBW/ACF (B)
SFBW/ACF (B)
SFBW/ACF (B)
SFBW/ACF (B)
SFBW/ACF (B)
SFBW/ACF (B)
SFBW/ACF (B)
SFBW/ACF (B) '
SFBW/ACF (B)
SFBW/ACF (B) 1
SFBW/ACF (B)
AA Regen
AA Regen
AA Regen
NF(A)
NF(A)
NF(A)
NF(A)
NF(A)
Sample
description
Composite
Brine
Brine
Brine
Brine
Brine
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO concentrate
RO concentrate
Composite
Composite
Composite
Composite
Composite
Composite
Composite
Composite
Composite
Composite
Settled Comp.
Settled Comp.
Settled Comp.
Settled Comp.
AA Regenerant '
AA Regenerant
AA Regenerant '
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
Alum
(mg/L)
-
-
-
-
100
100
-
-
•
-
50
50
100
100
-
-
-
-
-
-
-
75
75
100
100
-
-
-
-
-
-
-
-
-
-
-
-
Treatment conditions
Ferric
(mg/L)
-
50
100
100
-
-
50
50
100
100
-
-
-
-
,
25
50 „
75
75
100
100
-
_
-
- .
25
50
75
100
-
25
50
75
75
100
100
150
Polyme
(mg/L)
-
-
-
-
-
-,
-
-
-
-
-
- -
.
-
-
-
-
-
-
-
•
. -
-
•
-
-
-
-
..- ,
-
4
- .
4
-
4
Coag.
PH
6.07
6.35
8.81
6.38
8.19
6.42
6.70
5.78
7.18
6.16
7.01
5.88
7.55
6.26
5.99
. 6.00
.5.97
7.19
6.18
6.65
6.12
7.28
6.07
7.13
6.12
7.32
6.94
6.68
6.54
7.13
7.10
6.95
4.79
6.74
4.90
6.35
4.98
Treated characteristics
Total As
(mg/L)
9.060
29.800
4.350
14.400
32.600
28.700
0.189
0.561
0.132
0.078
0.286
0.570
0.442
0.306
0.719
0.152
0.100
0.064
0.070
0.110
0.075
0.194
0.205
0.248
0.214
0.093
0.018
0.013
0.011
0.386
0.171
0.154
0.071
0.085
0.093
0.143
0.061
Total Fe
(mg/L)
_
10.900
4.900
6.030
.
.
4.420
_
0.087
3.220
7.800
_
4.700
3.930
2.660
2.800
4.890
3.440
_
_•
3.650
2.570
1.910
2.620
0.677
1.150
5.24
0.154
8.68
0.152
8.64
Total Al
(mg/L)
7.320
5.800
4.410
4.880
2.500
3.090
_
_
0.762
0.720
1.150
0.974
11.800 I
7.420
7.620
_
Arsenic
removal
(percent)
13.71
10.24
86.89
56.63
1.81
13.55
71.49
15.38
80.09
88.24
56.86
14.03
33 33
53.85
0.00
91.26
94.25
9632
9598
93.68
95.69
88.85
88.22
85 75
87.70
0.00
58.14
6977
74.42
85.32
93.50
94.14
86.42
82.40
82 22
7039
, 88.34
65
-------
Table A-3.
Continued
Wastewater
ID
NF(A)
NF(A)
NF(A)
NF(A)
NF(A)
NF(A)
NF(A)
NF(A)
NF(A)
NF(A)
NF(A)
NF(A)
NF(B)
NF(B)
NF(B)
NF(B)
NF(B)
NF(B)
NF(B)
NF(B)
NF(B)
NF(B)
NF(B)
NF(B)
NF(B)
NF(B)
NF(B)
Ion Ex (B)
Ion Ex (B)
Ion Ex (B)
Ion Ex (B)
Ion Ex (B)
Ion Ex (B)
Ion Ex (B)
Ion Ex (B)
Ion Ex (B)
Ion Ex (B)
Ion Ex (B)
Sample
description
SIF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
NF concentrate
Composite
Composite
Composite
Composite
Composite
Composite
Composite
Composite
Composite
Composite
Composite
Treatment conditions
Alum
(mg/L)
-
-
-
75
75
100
100
100
150
150
200
200
-
-
-
-
-
-
50
50
75
75
75
100
100
100
150
-
-
-
•
-
-
-
-
50
50
100
Ferric
(mg/L)
150
150
200
-
-
-
-
- .
-
-
- -
50
75
75
100
100
150
-
-
-
. -
-
-
-
-
-
50
50
100
100
200
200
200
200
-
-
-
Polymer
(mg/L)
-
- -
4
4
4
4
4
-
4
4
-
4
0.5
0.5
4
4
0.5
4
0.5
0.5
4
0.5
0.5
4
0.5
0.5
4
-
' -
-
-
-
- .
0.5
0.5
-
-
• -
Coag.
pH
6.20
5.30
5.20
7.20
6.08
7.11
6.11
6.83
6.97
6.04
6.60
6.59
5.06
5.04
6.52
6.31
4.94
6.24
6.04
6.90
6.55
6.79
6.05
6.51
6.75
6.02
6.44
9.90
6.18
9.90
6.15
9.90
6.15
9.90
6.20
9.90
6.15
9.90
Treated characteristics
Total As
(mg/L)
0.094
0.117
0.009
0.197
0.221
0.130
0.162
0.225
0.060
0.073
0.148
0.029
0.030
0.036
0.009
0.006
0.020
0.005
0.235
0.283
0.116
0.157
0.129
0.067
0.087
0.073
0.035
25.8
26.0
25.3
23.3
22.7
18.7
23.2
17.5
26.6
23.5
24.6
Total Fe
(mg/L)
10.0
11.9
1.41
-
-
-
-
-
-
-
-
-
1.22
2.23
4.69
0.699
159 1
0.967
-
-
-
-
-
-
-
-
-
6.14
9.55
5.95
8.81
5.44
7.89
8.17
2.96
-
. -
-
Total Al
(mg/L)
-
-
-
1.220
1.640
1.230
1480
3.810
0.821
1.200
4.970
0.673
-
-
-
-
•-
-
0.649
0.566
0.491
0.796
0.697
0.526
0.525
0.626
0.654
:;
-
-
-
-
-
-
-
-
4.54
4.05
8.59
Arsenic
removal
percent)
80.54
75.78
98.14
62.33
57.74
75.14
69.02
53.42
88.53
86.04
69.36
94.00
93.83
92.59
98.15
98.77
95.88
98.97
!
51.65
41.77
76.13
67.70
73.46
86.21
82.10
84.98
92.80
•
0.00
0.00
0.00
6.05
8.47
24.60
6.45
29.44
0.00
5.24
6.81
66
-------
Table A-3.
Continued
Wastewater
ID
Ion Ex (B)
Ion Ex (B)
ion Ex (B)
Ion Ex (B)
Ion Ex (B)
Sample
description
Composite
Composite
Composite
Composite
Composite
Alum
(mg/L)
100
200
200
200
200
Treatment conditions
Ferric
(mg/L)
- •
-
-
-
Polymer
(mg/L)
-
- '•
-
0.5
0.5
Coag.
PH
6.14
9.90
6.12
9.90
6.12
Treated characteristics
Total As
(mg/L)
23.3
25.5
22.8
26.3
25.5
Total Fe
(mg/L)
_
.
.
_
-
Total Al
(mg/L)
5.59
16.2
5.82
16.7
1.28
Arsenic
removal
(percent)
6.05
0.00
8.06
0.00
0.00
67
-------
Table A-4. Adsorption test data
Wastewater
ID
SFBW(A)
SFBW(A)
SFBW (A)
RO(A)
RO(A)
RO(A)
ion Ex (A)
Ion Ex (A)
Sample
description
Backwash
Backwash
Backwash
Concentrate
Concentrate
Concentrate
Brine
Brine
Media
type
Iron media
Iron media
Activated
alumina
Iron media
Iron media
Activated
alumina
Iron media
Activated
alumina
EBCT
(min)
1.5
_
..
_
-
4.5
_
_
_
-
1.5
_
_
_
-
1.5
-_
_
_
_
-
3
_
_
_
-
3
„
_ ,
_
-
1.5
_
_
-
1.5
.
_
„
-
Sample
time
(hrs)
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
1
2
3
4
5
6
1
2
3
4
1
2
3
4
5
Sample
bed-
volumes
40
80
120
160
200
240
12.2
26.6
40
53.2
66.6
80
40
80
120
160
200
240
40
80
120
160
200
240
20
40
60
80
100
20
40
60
80
100
120
40
80
120
160
40
80
120
160
200
Treated characteristics
Test
PH
7.6
.
-
.
-
7.6
•
-
.
-
7.6
-
-
_
-
7.5
-
_
.
_
-
7.5
-
_
-
-
7.5
-
_
_
_
-
8.9
-
-
-
6.0
-
.
-
-
Total
As
(mg/L)
0.289
0.636
1.180
1.580
1.560
0.286
1.390
1.390
1.410
1.170
1.030
1.060
1.500
1.550
1.540
1.650
1.370
1.420
0.095
0.211
0.252
0.320
0.366
0.398
< 0.002
0.423
0.068
0.119
0.640
0.062
0.116
0.296
0.389
0.473
0.527
11.20
18.40
22.30
23.20
2.410
9.410
11.600
16.400
21.800
Total Fe
(mg/L)
27.8
59.5
111.0
149.0
146.0
28.0
136.0
135.0
141.0
118.0
102.0
105.0
141.0
146.0
145.0
154.0
129.0
136.0
0.209
<0.01
<0.01
<0.01
<0.01
<0.01
0.407
0.078
0.102
<0.01
<0.01
-
-
-
„
-
- •
0.813
0.783
0.753
0.755
- •
-
-
-
-
Total Al
(mg/L)
~
-
-
-
-
_
-
-
-
-
..
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
2.2
2.0
1.0
0.7
0.5
0.3
-
-
-
-
-
-
-
-
-
As
removal
(percent)
79.50
54.89
16^31
0.00
0.00
79.70
1.42
1.42
0.00
17.02
26.95
24.82
0.00
0.00
0.00
0.00
0.00
0.00
81.94
59.89
52.09
39.16
30.42
24.33
100.00
19.58
87.07
77.37
0.00
88.21
77.95
43.72
26.04
10.07
0.00
69.73
50.27
39.73
37.30
92J4
71.65
65.06
50.60
34.33
68
-------
Table A.4.
Continued
Wastewater
ID
Ion Ex (A)
Ion Ex (A)
RO(B)
RO(B)
RO (B)
RO(B)
SFBW/ACF (B)
Sample
description
Composite
Composite
Concentrate
Concentrate
Concentrate
Concentrate
Settled
composite
Media
type
Iron media
Iron media
Activated
alumina
Activated
alumina
Iron media
Iron media
Ion
exchange
EBCT
(min)
1.5
3
1.5
3
1.5
3
3
Sample
time
(hrs)
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
- 5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
Sample
bed-
volumes
40
80
120
160
200
240
20
40
60
80
100
120
40
80
120
160
200
240
20
40
60
80
100
. 120
40
80
120
160
200
240
20
40
60
80
100
120
20
40
60
80
100
120
Test
PH
9.0
9.0
7.3
7.3
8.0
7.3
8.1
Treated characteristics
Total
As
(mg/L)
0.897
2.280
3.600
4.120
4.620
7.020
0.044
11.600
1.710
2.890
3.810
11.600
0.047
0.095
0.180
0.202
0.263
0.292
0.004
0.012
0.020
0.037
0.051
0.071
< 0.002
0.010
0.036
0.068
0.086
0.106
<0.002
0.003
0.004
0.011
0.013
0.018
< 0.002
< 0.002
< 0.002
< 0.002
< 0.002
< 0.002
Total Fe
(mg/L)
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
0.061
0.097
<0.01
<0.01
-
;
<0.01
0.072
0.189
0.275
0.317
0.366
0.243
0.565
0.047
0.111
0.134
0.218
_
Total Al
(mg/L)
-
-
0.3
0.3
0.4
0.4
0.4
0.3
0.2
0.3
0.4
0.5
0.5
0.6
-
-
-
As
removal
(percent)
91.96
78.29
65.71
60.76
56.00
33.14
99.58
0.00
83.71
72.47
63.71
0.00
92.91
85.67
72.85
69.53
60,33
55.96
99.40
98.19
96.98
94.42
92.31
89.29
100.00
98.49
94.57
89.74
87.03
84.01
100.00
99.55
99.40
98.34
98.04
97.29
100.00
100.00
100.00
100.00
100.00
100.00
69
-------
Table A.4.
Continued
Wastewater
ID
SFBW/ACF (B)
SFBW/ACF (B)
NF(A)
NF(A)
NF(A)
NF(A)
NF(A)
Sample
description
Settled
composite
Settled
50:50 blend
Concentrate
Concentrate
Concentrate
Concentrate
Concentrate
Media
type
Iron media
Activated
alumina
Iron media
Iron media
Ion
exchange
Activated
alumina
Activated
alumina
EBCT
(min)
3
3
3
6
3
3
6
Sample
time
(hrs)
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
Sample
bed-
volumes
20
40
60
80
100
120
20
40
60
80
100
120
20
40
60
80
100
120
10
20
30
40
50
60
20
40
60
80
100
120
20
40
60
80
100
120
10
20
30
40
50
60
Treated characteristics
Test
pH
8.1
8.1
7.1
7.1
7.1
7.1
7.1
Total
As
(mg/L)
< 0.002
< 0.002
< 0.002
< 0.002
< 0.002
< 0.002
0.002
< 0.002
< 0.002
< 0.002
< 0.002
< 0.002
0.003
0.009
0.010
0.013
0.016
0.021
0.003
<0.002
<0.002
<0.002
0.003
0.004
0.246
0.459
0.650
0.690
0.579
0.535
0.007
0.011
0.014
0.021
0.026
0.034
0.004
0.002
0.003
0.002
0.004
0.007
Total Fe
(mg/L)
0.039
0.021
0.023
0.056
0.017
0.030
-
0.273
0.483
0.723
0.959
1.110
1.210
3.550
0.145
0.087
0.138
0.191
0.021
-
-
i
Total Al
(mg/L)
-
0.1
0.1
0.1
0.1
0.1
0.1
-
-
-
0.3
0.3
0.3
0.3
0.3
0.3
0.2
0.2
0.2
0.2
0.1
0.1
As
removal
(percent)
100.00
100.00
100.00
100.00
100.00
100.00
95.35
100.00
100.00
100.00
100.00
100.00
99.43
98.28
98.09
97.51
96.94
95.98
99.43
100.00
100.00
100.00
99.43
99.24
52.96
12.24
0.00
0.00
0.00
0.00
98.66
97.90
97.32
95.98
95.03
93.50
99.24
99.62
99.43
99.62
99.24
98.66
70
-------
Table A.4.
Continued
Wastewater
ID
NF(A)
NF(B)
NF(B)
NF (B)
NF(B)
NF(B)
NF(B)
Sample
description
Concentrate
Concentrate
Concentrate
Concentrate
Concentrate
Concentrate
Concentrate
Media
type
Modified
alumina
Iron media
Iron media
Ion
Exchange
Activated
alumina
Activated
alumina
Modified
alumina
EBCT
(min)
3
3
6
3
3
6
3
Sample
time
(hrs)
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
Sample
bed-
volumes
20
40
60
80
100
120
20
40
60
80
100
120
10
20
30
40
50
60
20
40
60
80
ioo
120
20
40
60
80
100
120
10
20
30
40
50
60
20
40
60
80
100
120
Test
PH
7.1
6.6
6.6
6.6
6.6
6.6
6.6
Treated characteristics
Total
As
(mg/L)
0.128
0.269
0.323
0.355
0.377
0.376
O.002
0.002
<0.002
0.491
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
0.180
0.493
0.587
0.485
0.439
0.438
<0.002
<0.002
<0.002
<0.002
<0.002
0.004
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
0.155
0.358
0.392
0.421
0.439
0.452
Total Fe
(mg/L)
0.961
0.998
0.928
0.998
0.920
0.856
0.152
O.010
0.269
0.359
0.254
0.185
0.118
0.072
0.075
0.097
0.077
0.192
-
_
-
0.167
0.099
0.179
0.462
0.414
0.315
Total Al
(mg/L)
0.3
0.2
0.2
0.1
0.2
0.2
-
-
-
0.1
0.1
0.1
0.2
0.1
0.1
0.3
0.3
0.3
0.3
0.3
0.3
0.2
0.2
0.2
0.2
0.2
0.2
As
removal
(percent)
75.53
48.57
38.24
32.12
27.92
28.11
100.00
99.59
100.00
0.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
62.96
0.00
0.00
0.21
9.67
9.88
100.00
100.00
100.00
100.00
100.00
99.18
100.00
100.00
100.00
100.00
100.00
100.00
68.11
26.34
19.34
- 13.37
9.67
7.00
71
-------
Table A.4.
Continued
\
Wastewater \ Sample
ID t description
Ion Ex (B) i Composite
f
t
i
i
Ion Ex (B) - Composite
i
^
Media
type
Iron media
Activated
alumina
i __
EBCt
(min)
3
3
Sample
time
(hrs)
1
2
3
4
5
6
1
2
3
4
5
6
Sample
bed-
volumes
20
40
60
80
100
120
20
40
60
80
100
120
Test
pH
9.9
9.9
treated characteristics
Total
As
(mg/L)
8.54
16.7
19.3
20.9
21.1
22.3
19.8
22.7
23.6
23.8
23.6
24.0
Total Fe
(mg/L)
12.3
1.50
1.01
0.252
0.726
0.600
'-'
Total Al
(mg/L),
-
9.8
8.2
7.5
6:7
6.2
5.4
As
removaj
(percent)
65.56
32:66
22.18
15.73
14.92
10.08
20.16
8.47
4.84
4.03
4.84
3.23
72
-------
£
'I
0.5
0.4
0.3
°'2
0.1
Untreated As cone. = 2.63 mg/L
pH 7.1
pH 7.1
pH 7.0
1
100
80
60
40
20
0 25 , 50
Ferric chloride dose (mg/L)
0 25 50
Ferric chloride.dose (mg/L)
Figure A-1. Total arsenic rem oval and total arsenic in the supernatant after ferric chloride
precipitation in activated alumina regenerant
10
Untreated As cone. = 10.5 mg/L
0 100
Ferric chloride dose (mg/L)
No polymer
pH 6.1
No polymer
pH 7.9
No polymer
pH 6.2
100
80
- 60
o
"M
I
8 40
20
0 100
Ferric chloride dose.(mg/L)
]No polymer
JpH 6.1
INo polymer
JpH 7.9
No polymer
pH 6.2
Figure A-2. Total arsenic re m oval and total arsenic remaining in the supernatant after ferric chloride
precipitation for ion exchange (A) composite
73
-------
10
Untreated
As cone. -
10.5 mg/L
No po.lymer
pH 8.9
No polymer
pH 5.5
50
40
I.
30
20
10
No polymer
pH 8.9
No polymer
pH 5.5
Alum dose (100 mg/L)
Alum dose (100 mg/L)
Figure A-3. Total arsenic removal and total arsenic remaining in the supernatant after alum
precipitation for ion exchange (A) composite
30
25
20
15
10
!2
TO
]ro 5
"5
Untreated As cone. = 24.8 mg/L
No polymer
pH 9.9
No polymer
pH 6.2
50
100
Ferric chloride dose (mg/L)
200
Figure A-4. Total arsenic remaining in the supernatant after ferric chloride
precipitation for ion exchange (B) composite
74
-------
30
O)
c
'c
'co
£
c
o
I
g
'c
-------
I
0.8
0.6
0.4
0.2
|2
Untreated As cone. = 0.526 mg/L
100
150
Alum dose (mg/L)
Figure A-7. Total arsenic remaining in the supernatant after alum
precipitation for reverse osmosis (A) concentrate
0.6
0.5
0.4
0.3
0.2
0.1
-
-
I
8^
Untreated As
cone. = 0.663 mg/L
I
Y,
1
i_
fvvs NO polymer
DOO DH 6.7
UNO polymer
pH 5.8
'•//JNO polymer
V
-------
=d 0.8
en
E
0.6
0.4
0.2
Untreated As
cone. = 0.663 mg/L
No polymer
pH 7.0
No polymer
pH 5.9
No polymer
pH 7.6
No polymer
pH 6.3
60
50
40
30
=8
o
ra
75 20
10
50 100
Alum dose (mg/L)
50 100
Alum dose (mg/L)
Figure A-9. Total arsenic removal and total arsenic remaining in the supernatant after alum
precipitation for reverse osmosis (B) concentrate
0.3
f 0.25
emaining
P
ro
c
.g
5 0.15
I
I
.o
°c
CD
£2
CO
"co 0.05
P
-»•
Untreated As cone. = 0.523 mg/L
75
100 150
Ferric chloride dose (mg/L)
200
Figure A-1 0. Total arsenic remaining in the supernatant after ferric chloride
precipitation for nanofiltration (A) concentrate
No polymer
pH 7.0
No polymer
pH 5.9
No polymer
pH 7.6
No polymer
pH 6.3
4 mg/L LT 22S
pH 7.0
4 mg/L LT 22S
pH 6.1
4 mg/L LT 22S
pH 6.6
No polymer
pH 6.7
77
-------
0.3
3"
g 0.25
D>
1 0.2
£
2 0.15
(0
S 0.05
o
Untreated As cone. = 0.523 mg/L
75
100 150
Alum dose (mg/L)
200
Figure A-11. Total arsenic remaining in the supernatant after alum
precipitation for nanofiltration (A) concentrate
0.04
0.03
c
'ra
2 0.02
8
| 0.01
"to
Untreated As cone. = 0.486 mg/L
50
75 100
Ferric chloride dose (mg/L)
150
Figure A-12. Total arsenic remaining in the supernatant after ferric chloride
precipitation for nanofiltration (B) concentrate
4 mg/L LT 22S
pH 7.0
4 mg/L LT 22S
pH 6.1
4 mg/L LT 22S
pH 6.6
No polymer
pH 6.7
0.5 mg/L LT 22S
pH 5.0
4 mg/L LT 22S
pH 6.3
78
-------
0.3
f 0.25
O)
c
'g
"ro
o>
c
o
CD
o
c
o
'c
(1)
to
CD
0.2
g 0.15
0.1
3 0.05
Untreated As cone. = 0.486 mg/L
50
75 100
Alum dose (mg/L)
150
0.5 mg/L LT 22S
pH 6.0
0.5 mg/L LT 22S
pH 6.8
4 mg/L LT 22S
pH 6.5
Figure A-13. Total arsenic remaining in the supernatant after alum
precipitation for nanofiltration (B) concentrate
0.25
I"
|
(D
2
ro
0.2
0.15
°-1
0.05
SFBW composite As
cone. = 1.41 mg/L
V
V
0 25 50
Ferric chloride dose (mg/L)
4 mg/L LT 22S
pH 5.6
mg/L LT 22S
pH 7.3
[4
p
pH 7.1
INo polymer
|pH 5.0
INo polymer
]pH 7.1
••B
1
100
80
60 -
40 -
20 -
-
.
<
>
>
•>
>
>
>
>
>
s
>
>
>
>
>
>
>
>
>
>
>
7
/
'
'/
/
',
'
/
/
/
/
/
/
/
£
- -
— .
\ -_
- -
~ ~
$$&
1
W
Ffff
N-M-
-:
4 mg/L LT 22S
pH 5.6
4 mg/L LT 22S
pH 7.3
4 mg/L LT 22S
pH 7.1
No polymer
pH 5.0
No polymer
pH 7.1
0 25 50
Ferric chloride dose (mg/L)
Figure A-14. Total arsenic removal and total arsenic remaining in the supernatant after ferric chloride
precipitation for spent filter backwash (A) composite (unsettled)
79
-------
0.3
0.2
0.1
Untreated As
cone. = 1.41 mg/L
4 mg/L LT 22S
pH 7.5
No polymer
pH 6.0
No polymer
pH 7.4
100
80 -
60 -
40 -
20 -
-
_
-
~
1
1
5<
>c
o*
Qx
Qx
>C
x^
CX
Qxi
O*
O*
x*
X>
X*
X/
X*
x!
X
>c
o*
o*
x/
X*
1
3
|
|
Qj
x5
o*
Qx;
Qx
/c
o*
o*
o*
o*
o*
Sc
x5
o*
x5
Sc
Xx
X
Sc
Sc
Sc
X/
Sc
§
El
^
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
Z,
z
z
z
KXX4 mg/L LT 22S
FVVS oH 7.5
INo polymer
pH 6.0
/// No polymer
VZPH 7.4
0 25 50
Alum dose (mg/L)
0 25 50
Alum dose (mg/L)
Figure A-15. Total arsenic removal and total arsenic-remaining in the supernatant after alum
precipitation for spent filter backwash (A) composite
U. 1 O
0.14
1 0.12
1 0.1
£
a o.os
C
8
e
| 0.06
*12
M
2 0.04
O
H*
0.02
n
_
-
-
-
-
-
_
|
|
5
>r
>/
K
x
>r
>r
8
>r
K
0
>?
v
jX
V
x
y
^
Untreated As
cone = 1 .74 m Q/L
X
X
X
X
x
x
x
x
x
x
x
x
x
x
x
y
x
x
X
x
v
^~i
V
C
V
s<
C
>
V
N^
V
V
s<
V
^
/
X
/
/
/
/
/
/
/
/
n-
s<
s<
£
J>
?>
<
*>
J>
Jx
J>
*>
•>
%
XX> No polymer
XX/" pH 6.0
UT No polymer
| pH 6.7
//AHo polymer
ZZ/lpH 7.2
25
50
75
100
100
80
r 60
s
I to
CO
20
///
'//
No polymer
pH 6.0
No polymer
pH 6.7
No polymer
pH 7.2
Ferric chloride dose (mg/L)
25 50 75 100
Ferric chloride dose (mg/L)
Figure A-16. Total arsenic removal and total arsenic remaining in the supernatant after ferric chloride
precipitation for spent filter backwash (B) composite (unsettled)
80
-------
0.3
Untreated As
cone. = 1.74 mg/L
No polymer
pH 7.2
No polymer
pH 6'.1
100
S2
ro
75 100
Alum dose (mg/L)
75 100
Alum dose (mg/L)
Figure A-17. Total arsenic removal and total arsenic remaining in the supernatant after alum
precipitation for spent filter backwash (B) composite
0.1
0.08 -
s
°
0.06 -
1
0.04 -
0.02 -
"
;
<
/
- ;
p
p
'
)
}
j
s
>
S
5
)
S
)
")
J
X
x Untreated settled As
* cone. - 0.043 mg/L
X
X
X
X
X
X
(
K
><
><
M
^
<
< nn
: P
:
, II
R
d 1
/I rn
VV>jNo polymer
5vQiDH 7.3
UNo polymer
pH 6.9
'//. No polymer
/VVPH 6.7
II 1 1 ||No polymer
II 1 1 llpH 6.5
80
60
I
Q.
C
o
'ts
1
o
40
20
25 50 75 100
Ferric chloride dose (mg/L)
25 50 75 100
Ferric chloride dose (mg/L)
No polymer
pH 7.2
No polymer
pH 6.1
No polymer
pH 7.3
No polymer
pH 6.9
No polymer
pH 6.7
No polymer
pH 6.5
Figure A-18. Total arsenic removal and total arsenic remaining in the supernatant after ferric chloride
precipitation for spent filter backwash (B) composite (settled)
81
-------
-------
Appendix B. QA/QC Results
Results
The QA/QC results contain data from the instrumental
methodologies employed for the analysis of metal ions (As,
Al, Fe, Mn) and sulfate (SCy2'). Valid QA and analytical data
were obtained through the use of duplicate and spiked
samples. The QA/QC results for the key analyses conducted
during the study are provided in Table B-1. This table
reports the relative percent deviation (RPD) of duplicate
analyses, spike recovery percentages, and the continuing
calibration value (CCV). These data quality indicators are
provided for each day that samples were analyzed (for each
measurement) throughout the project. The spike and CCV
values used along with the calibration ranges are listed as
table footnotes.
The QA/QC results presented in Table B-1 indicate that the
objectives outlined by the QAPP were achieved. The RPD
for each QA/QC analysis was less than 25 percent,
demonstrating good analytical precision. The spike
recoveries were within the 75 to 125 percent accuracy range.
The CCV percent recovery data ranged between 95 and 110
percent for each spiked analyte evaluated.
Corrective Actions
During the study there were no deviations from the sampling
procedures outlined by the QAPP. All sampling for analytical
tests performed were conducted at EE&T and either
analyzed by EE&T Laboratory or hand-delivered to James R.
Reed Laboratory for analysis.
There were also no deviations from the analytical procedures
that were outlined in the QAPP. Analytical or calculation
errors, if present, were found and corrected after completion
of each set of analyses. All data were evaluated by the QA
officer to determine if re-analysis was necessary. Overall,
there were no modifications to the original QAPP and any
corrective actions required were taken during the analytical
runs and corrections were made prior to proceeding.
83
-------
Table B-1. QA/QC summary
Analysis date
5/26/99
6/8/99
6/10/99
6/30/99
7/6/99
7/8/99
7/12/99
7/19/99
7/21/99
7/22/99
7/23/99
8/10/99
Analyte
Arsenic
Iron
Manganese
Arsenic
Iron
Manganese
Sulfate
Arsenic
Iron
Arsenic
Iron
Manganese
Arsenic
Arsenic
Iron
Iron
Aluminum
Arsenic
Arsenic
Iron
Manganese
Arsenic
Iron
Manganese
Sulfate
Arsenic
Iron
Arsenic
Arsenic
Iron
Manganese
Arsenic
Arsenic
Arsenic
Iron
Iron
Aluminum
1st analysis
(mg/L)
0.550
0.724
0.552
< 0.002
1.47
0.100
1,808
0.514
0.561
< 0.002
0.774
0.093
< 0.002
1.560
0.277
146
0.258
0.520
1.060
105
9.54
< 0.002
0.654
1.03
19.71
< 0.002
6.34
0.039
33.1
< 0.010
0.010
0.052
0.035
3.60
2.54
3.37
0.217
2nd analysis Duplicate RPD
(mg/L) (percent)
0.545
0.744
0.552
< 0.002
1.45
0.099
1,698
0.518
0.566
< 0:002
0.767
0.092
< 0,002
1.530
0.272
144
0.228
0.517
1.060
105
9.54
< 0.002
0.668
1.05
17.16
< 0.002.
6.40
0.040
33.4
<0.010
0.010
0.051 •
0.033
3.60
2.52
3.38
0.210
0.9
2.7
0.0
N/A
1.4
1.0
6.5
0.8
0.9
N/A
0.9
1.1
N/A
2.0
1.8
1.4
13.2
0.6
0.0
0.0
0.0
N/A
2.1
1.9
14.9
N/A
0.9
2.5
0.9
N/A
0.0
2.0
6.1
0.0
0.8
0.3
3.3
Spike recovery CCV recovery
(percent) (percent)
110
95
98
112
115
100
79.7
103
97
119
99
107
103
121
108
127
100
104
116
101
106
116
96
100
114
101
99
103
101
75
94
75
94
93
101
100
115
95
97
98
104
100
101
87
105
102
103
96
96
108
103
100
102
107
102
103
99
100
102
102
100
100
103
101
105
103
104
105
104
106
105
109
110
100
84
-------
Table B-1. Continued
Analysis date Analyte
8/17/99 Arsenic
Arsenic
Arsenic
Iron
Iron
Iron
Aluminum
8/31/99 Arsenic
Arsenic
Iron
Aluminum
9/1/99 Arsenic
Arsenic
jron
Aluminum
Aluminum
9/17/99 Arsenic
Arsenic
Iron
Manganese
9/29/99 Arsenic
Arsenic
Iron
Iron
10/4/99 Arsenic
Aluminum
10/5/99 Arsenic
Iron
10/12/99 Arsenic
Iron
10/14/99 Arsenic
Iron
Manganese
10/15/99 Arsenic
Iron
Manganese
Aluminum
1st analysis
(mg/L)
0.286
0.720
< 0.002
7.86
0.821
0.107
0.225
0.096
0.044
0.559
0.325
0.471
0.004
< 0.010
0.499
0.170
< 0.002
0.003
0.042
.0.045
0.531
< 0.002
0.600
0.055
0.248
1.14
0.093
3.65
0.147
4.54
< 0.002
1.08
0.864
< 0.002
0.104
0.030
0.219
2nd analysis
(mg/L)
0.286
0.718
< 0.002
7.75
0.827
0.106
0.231
0.095
0.043
0.555
0.344
0.475
0.004
< 0.010
0.509
0.180
< 0.002
0.004
0.040
0.045
0.524
< 0.002
0.594
0.056
0.251
1.16
0.095
3.65
0.158
4.87
0.002
, 1.04
0.862
< 0.002
0.105
0.031
0.215
Duplicate RPD
(percent)
0.0
0.3
N/A
1.4
0.7
0.9
2.6
1.1
2.3
0.7
5.5
0.8
0.0
N/A
2.0
5.6
N/A
25.0
5.0
0.0
1.3
N/A
1.0
1.8
1.2
1.7
2.1
0.0
7.0
6.8
N/A
3.8
0.2
N/A
1.0
3.2
1.9
Spike recovery
(percent)
106
104
103
93
96
95
93
115
102
101
111
120
120
100
122
118
114
110
112
108
106
104
115
93
107
104
105
111
107
106
111
110
116
115
109
114
121
CCV recovery
(percent)
103
105
104
98
98
97
103
103
103
101
102
104
106
103
101
106
101
104
99
100
104
105
102
103
100
92
104
102
100
98
102
101
101
102
101
102
108
85
-------
Table B-1. Continued
Analysis date
10/19/99
10/22/99
10/26/99
10/27/99
11/1/99
11/8/99
11/10/99
11/18/99
11/22/99
11/29/99
12/2/99
12/7/99
12/15/99
12/16/99
1/31/00
Analyte
Sulfate
Arsenic
Arsenic
Iron
Manganese
Aluminum
Arsenic
Arsenic
Aluminum
Arsenic
Arsenic
Iron
Arsenic
Aluminum
Arsenic
Arsenic
Arsenic
Iron
Iron
Manganese
Aluminum
Arsenic
Iron
Aluminum
Arsenic
Aluminum
Arsenic
Aluminum
Sulfate
Arsenic
Iron
Aluminum
Arsenic
Iron
Aluminum
Arsenic
Arsenic
1st analysis
(mg/L)
1,082
0.522
0.020
1.21
0.057
0.206
0.654
0.014
0.282
< 0.002
< 0.002
0.145
0.005
0.181
< 0.002
< 0.002
0.067
0.268
0.184
0.012
0.160
< 0.002
0.131
0.115
< 0.002
0.130
< 0.002
0.288
1,194
0.436
0.446
0.212
0.269
1.00
0.146
0.011
0.494
2nd analysis
(mg/L)
1,091
0.524
0.021
1.21
0.057
0.192
0.647
0.015
0.278
< 0.002
< 0.002
0.145
0.004
0.162
< 0.002
0.002
0.067
0.270
0.186
0.012
0.140
< 0.002
0.127
0.1,28
< 0.002
0.130
< 0.002
0.289
1,221
0.442
0.482
0.249
0.269
0.997
0.163
0.013
0.495
Duplicate RPD
(percent)
0.8
0.4
4.8
0.0
0.0
7.3
1.1
6.7
1.4
N/A
N/A
0.0
25.0
11.7
N/A
N/A
0.0
0.7
1.1
0.0
14.3
N/A
3.1
10.2
N/A
0.0
N/A
0.3
2.2
1.4
7.5
14.9
0.0
0.3
10.4
15.4
0.2
Spike recovery
(percent)
98
110
, 116
111
112
115
112
109
109
110
112
97
109
1.11
101
101
102
111
98
101
1.15
104
93
117
103
115
102
107
98
105
100
105
109
87
117
105
115
CCV recovery
(percent)
101
104
103
100
101
104
103
100
109
101
98
94
102
106
104
100
99
103
99
100
99
101
100
100
101
100
100
100
98
104
106
99
99
100
100
103
101
86
-------
Table B-1. Continued :
Analysis date
2/9/00
2/11/00
2/14/00
2/16/00
2/17/00
2/24/00
3/2/00
3/6/00
3/27/00
Spike value =
CCV value
CCV value =
Analyte
Iron
Arsenic
Iron
Arsenic
Iron
Manganese
Arsenic
Iron
Manganese
Arsenic
Arsenic
Iron
Aluminum
Arsenic
Arsenic
Iron
Aluminum
Sulfate
Arsenic
Aluminum
Arsenic
Iron
Aluminum
Arsenic
Iron
0.5 ppm
1.0 ppm
1.0 ppm
0.3 ppm
1st analysis
(mg/L)
7.03
0.518
0.992
< 0.002
0.203
0.009
34.3
< 0.010
0.006
0.029
0.004
0.299
0.677
19.3
23.9
1.01
0.249
2,118
0.768
5.29
< 0.002
0.037
0.608
32.9
15.5
(As, Fe, Al, Mn)
(S042-)
(As, Fe, Al, Mn)
(S042-)
2nd analysis
(mg/L)
7.07
0.520
0.981
< 0.002
0.202
0.009
34.3
< 0.010
0.006
0.029
0.003
0.297
0.669
19.3
24.0
1.01
0.268
1,864
0.778
5.26
< 0.002
0.035
0.584
33.0
15.5
Calibration
As 0.0
Al 0-
Fe 0-
Mn 0.0
S042' 0 -
Duplicate RPD
(percent)
0.6
0.4
1.1
N/A
0.5
0.0
0.0
N/A
0.0
0.0
33.3
0.7
1.2
0.0
0.4
• 0.0
7.1
13.6
1.3
0.6
N/A
5.7
4.1
0.3
0.0
ranges:
- 2.0 ppm
25 ppm
10 ppm
- 2.0 ppm
10 ppm
Spike recovery
(percent)
100
114
109
106
99
101
100
106
106
105
109
100
121
100
112
111
113
89
104
92
106
109
109
85
99
CCV recovery
(percent)
98
102
100
100
96
98
99
99
101
99
99
99
99
102
101
102
100
92
103
99
101
98
110
102
102
87
*U.S. GOVERNMENT PRINTING OFFICE:2001-650-101/40013
-------
-------
-------
United States
Environmental Protection Agency/ORD
National Risk Management
Research Laboratory
Cincinnati, OH 45268
Please make all necessary changes on the below label,
detach or copy, and return to the address in the upper
left-hand comer.
If you do not wish to receive these reports CHECK HEREd;
detach, or copy this cover, and return to the address in the
upper left-hand corner.
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
Official Business
Penalty for Private Use
$300
EPA/600/R-01/033
------- |