EPA/600/R-07/048
                                                             June 2007
Arsenic Removal from Drinking Water by Iron Removal
              EPA Demonstration Project at
 Big Sank Lake Mobile Home Park in Sank Centre, MN
               Six-Month Evaluation Report
                              by

                          H. Tien Shiao
                       Abraham S.C. Chen
                        Wendy E. Condit
                           Lili Wang

                            Battelle
                     Columbus, OH 43201-2693
                     Contract No. 68-C-00-185
                       Task Order No. 0029
                              for

                         Thomas J. Sorg
                       Task Order Manager

              Water Supply and Water Resources Division
            National Risk Management Research Laboratory
                      Cincinnati, Ohio 45268
            National Risk Management Research Laboratory
                 Office of Research and Development
                U.S. Environmental Protection Agency
                      Cincinnati, Ohio 45268

-------
                                       DISCLAIMER
The work reported in this document was funded by the United States Environmental Protection Agency
(EPA) under Task Order 0029 of Contract 68-C-00-185 to Battelle. It has been subjected to the Agency's
peer and administrative reviews and has been approved for publication as an EPA document. Any
opinions expressed in this paper are those of the author(s) and do not, necessarily, reflect the official
positions and policies of the EPA.  Any mention of products or trade names does not constitute
recommendation for use by the EPA.

-------
                                         FOREWORD

The U.S. Environmental Protection Agency (EPA) 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 National Risk Management Research Laboratory (NRMRL) is the Agency's center for investigation
of technological and management approaches for preventing and reducing risks from pollution that
threaten 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 sub-
surface resources; protection of water quality in public water systems; remediation of contaminated sites,
sediments and groundwater; 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 envi-
ronmental problems by developing and promoting technologies that protect and improve the environment;
advancing scientific and engineering information to support regulatory and policy decisions; and provid-
ing the technical support and information transfer to ensure implementation of environmental regulations
and strategies at the national, state, and community levels.

This publication has been produced as part of the Laboratory's strategic long-term research plan.
It 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.
                                            Sally Gutierrez, Director
                                            National Risk Management Research Laboratory
                                               in

-------
                                          ABSTRACT

This report documents the activities performed and the results obtained from the first six months of the
arsenic removal treatment technology demonstration project at the Big Sauk Lake Mobile Home Park
(BSLMHP) in Sauk Centre, MN. The objectives of the project are to evaluate the effectiveness of
Kinetico's Macrolite® pressure filtration process in removing arsenic to meet the new arsenic maximum
contaminant level (MCL) of 10 |o,g/L, the reliability of the treatment system, the required system
operation and maintenance (O&M) and operator skill levels, and the capital and O&M cost of the
technology. The project also is characterizing water in the distribution system and process residuals
produced by the treatment system.

The Macrolite® CP-213f arsenic removal system at BSLMHP consisted of two 36-in diameter by 57-in
tall contact tanks (205 gal each) and four 13-in diameter by 54-in tall pressure tanks (two for each duplex
unit), all configured in parallel.  Each pressure tank contained 20  in (or 1.5 ft3) of Macrolite® filter media.
The maximum design flowrate was 20 gpm, which yielded at least 20 min of contact time prior to
pressure filtration and at least 5.4 gpm/ft2 of hydraulic loading to  the Macrolite® filters. Because the
system operated in an on-demand configuration, the actual flowrates ranged from  1 to 15 gpm,
corresponding to 27 to 410 min of contact time and 0.3 to 4.1 gpm/ft2 of hydraulic loading.  From July 13,
2005, through January 17, 2006, the system operated for atotal of 617 hr at approximately 3.4 hr/day.
Based on the totalizer readings, the system treated approximately 863,470 gal of water with  an average
daily demand of 4,617 gal during this time period.

Total arsenic concentrations in source water ranged from 20.6 to 36.6 |o,g/L with As(III) being the
predominating species at an average concentration of 23.0 |o,g/L.  Potassium permanganate (KMnO4) was
used to oxidize As(III) and Fe(II) prior to Macrolite® pressure filtration. KMnO4 was selected as the
oxidant because of the presence of elevated total organic carbon (TOC) levels (at 3.2 to 4.8 mg/L) in
source water and high formation potential of disinfection byproducts with the use of chlorine.

After the contact tanks, As(III) concentrations were reduced to an average value of 1.9 |og/L, suggesting
effective oxidation of As(III) to As(V) with KMnO4. Meanwhile, arsenic was present primarily in the
particulate form at an average value of 22.9 |o,g/L, presumably, by being bound to iron particles. During
the first six months from July 13, 2005, through January 17, 2006, total arsenic levels in the  treated water
were reduced to 2.9  to 17.7  |o,g/L (averaged 7.6 |og/L). Out of 24  sampling occasions, arsenic
concentrations exceeded the 10-(ig/L MCL for a total of eight times, with all but one due to  particulate
breakthrough.  Two  samples exceeded total arsenic concentrations of 10 (ig/L due to low KMnO4 dosage,
resulting in incomplete  oxidation of As(III) and Fe(II).  In order to address particulate arsenic
breakthrough, the backwash frequency was increased from every  2,743 gal to every 1,714 gal of
throughput during this time  period.

Total iron concentrations averaged 2,760 |o,g/L in source water, which is above the secondary MCL of 300
Hg/L. With an average soluble iron to soluble arsenic ratio of 100:1, there was sufficient natural iron
present in source water for effective arsenic removal. After the contact tanks, iron was present primarily
in the particulate form,  suggesting effective oxidation even in the presence of elevated TOC levels. Total
iron concentrations in the treated water ranged from <25 to 1,067 (ig/L and averaged 259 (ig/L. An
increase in particulate iron correlated with an increase in particulate arsenic, indicating iron  breakthrough
from the Macrolite® filters.

Total manganese concentrations averaged 144 (ig/L in source water, existing primarily in the soluble form
as Mn(II) at 132 (ig/L.  Afterthe addition of 2.6 to 3.8 mg/L of KMnO4 and after the contact tanks,
manganese was present primarily in the soluble form based on the use of 0.45-(im disc filters, with levels
                                               IV

-------
ranging from 337 to 946 (ig/L before November 15, 2005. The high levels of TOC in source water
appeared to have inhibited the formation of filterable manganese solids. Based on the results of a series
of jar tests, the KMnO4 dosage applied to the treatment system was increased to 5.6 mg/L.  The increased
KMnO4 dosage enabled most manganese to precipitate after the contact tanks,  leaving only 108 to 166
|o,g/L measured as soluble manganese. Further adjustments will be made to the KMnO4 dosing in the next
six-month period to further lower the soluble manganese levels.

During this time period, the backwash water production rates ranged from 2.8 to 7.2%.  The control disc
on top of each  set of duplex units was changed out twice to increase the backwash frequency in order to
address the particulate arsenic, iron, and manganese breakthrough. The backwash frequency was
increased from the initial field setting of every 2,743 gal to every 1,714 gal.  If needed, further
adjustments will be made in the next six-month study period. After November 15, 2006, when the
modified backwash procedure was implemented, total arsenic concentrations in the backwash water
ranged from 114 to 417 (ig/L; total iron concentrations ranged from 14,069 to 77,641 (ig/L; and total
manganese concentrations ranged from 1,595 to 16,178 (ig/L. Using 130 gal of backwash water
produced, this  equates to approximately 0.17 Ib of solids, including 4.4 x 10"4 Ib of arsenic, 0.08 Ib of
iron, and 0.01 Ib manganese, generated per backwash event.

Comparison of the distribution system sampling results before and after system startup showed a
significant decrease in arsenic and iron levels and a significant increase in manganese levels at all three
sampling locations. The distribution water sampling results essentially mirrored the treatment results of
the Macrolite® filters.  Neither lead nor copper concentrations at the sample sites appear to have been
affected by the operation of the system.

The capital investment cost was $63,547, which included $22,422 for equipment, $20,227 for
engineering, and $20,898 for installation. Using the system's rated capacity of 20 gal/min (gpm) (28,800
gal/day [gpd]), the capital cost was $3,177/gpm ($2.21/gpd).

The O&M cost for the Macrolite® CP-213f system included only incremental cost associated with the
chemical supply, electricity consumption, and labor.  The O&M cost was estimated in this report at
$0.43/1,000 gal and will be refined at the end of the one-year evaluation period.

-------
                                        CONTENTS


DISCLAIMER	ii
FOREWORD	iii
ABSTRACT	iv
FIGURES	vii
TABLES	vii
ABBREVIATIONS AND ACRONYMS	ix
ACKNOWLEDGMENTS	xi

Section 1.0: INTRODUCTION	1
       1.1  Treatment Technologies for Arsenic Removal	2
       1.2  Project Objectives	2

Section 2.0: SUMMARY AND CONCLUSIONS	5

Section 3.0: MATERIALS AND METHODS	7
       3.1  General Project Approach	7
       3.2  System O&M and Cost Data Collection	8
       3.3  Sample Collection Procedures and Schedules	8
           3.3.1  Source Water	11
           3.3.2  Treatment Plant Water	11
           3.3.3   Special Study - KMnO4 Jar Tests	11
           3.3.4  Backwash Water	11
           3.3.5  Residual Solids	12
           3.3.6  Distribution System Water	12
       3.4  Sampling Logistics	12
           3.4.1  Preparation of Arsenic Speciation Kits	12
           3.4.2  Preparation of Sampling Coolers	12
           3.4.3  Sample Shipping and Handling	12
       3.5  Analytical Procedures	13

Section 4.0: RESULTS AND DISCUSSION	14
       4.1  Facility Description and Pre-Existing Treatment System Infrastructure	14
           4.1.1  Source Water Quality	14
           4.1.2  Distribution System and Treated Water Quality	16
       4.2  Treatment Process Description	17
       4.3  System Installation	22
           4.3.1  Permitting	22
           4.3.2  Building Construction	22
           4.3.3  System Installation, Shakedown, and Startup	22
       4.4  System Operation	23
           4.4.1  Operational Parameters	23
           4.4.2  Backwash	23
           4.4.3  Residual Management	25
           4.4.4  System/Operation Reliability and Simplicity	25
       4.5  System Performance	26
           4.5.1  Treatment Plant Sampling	26
           4.5.2  Backwash Water Sampling	37
           4.5.3  Distribution System Water Sampling	38
                                            VI

-------
       4.6  System Cost	41
            4.6.1  Capital Cost	41
            4.6.2  Operation and Maintenance Cost	42

Section 5.0:  REFERENCES	44

APPENDIX A:  OPERATIONAL DATA	A-l
APPENDIX B:  ANALYTICAL DATA	B-l


                                      FIGURES

Figure 3-1.  Process Flow Diagram and Sampling Locations	10
Figure 4-1.  Pre-Existing Pump House at BSLMHP, MN Site	14
Figure 4-2.  Existing Well Piping and Pressure Tanks at BSLMHP, MN Site	15
Figure 4-3.  Process Schematic of Macrolite* Pressure Filtration System	18
Figure 4-4.  Photograph of Macrolite® Pressure Filtration System	18
Figure 4-5.  KMnO4 Feed System	20
Figure 4-6.  Kinetico's Mach 1250 Control Valve	20
Figure 4-7.  Backwash Flow Path for One Duplex Unit with Control Disc No. 8 and a
            Throughput of 1,714 gal between Backwash Cycles	21
Figure 4-8.  Concentrations of Arsenic Species at IN, AC, and TT Sampling Locations	30
Figure 4-9.  Total Arsenic Concentrations at TA/TB, TC/TD, and TT Sampling Locations	31
Figure 4-10. Total Iron Concentrations at TA/TB, TC/TD, and TT Sampling Locations	31
Figure 4-11. Total and Soluble Manganese Concentrations at AC Sampling Location	33
Figure 4-12. Total Manganese Concentrations at TA/TB, TC/TD, and TT Sampling Locations	34
Figure 4-13. Jar Test Setup	36
Figure 4-14. Total Phosphorous Concentrations at IN, AC, TA/TB, TC/TD  and TT Sampling
            Locations	38
                                          TABLES

Table 1-1.   Summary of Round 1 and Round 2 Arsenic Removal Demonstration Locations,
            Technologies, and Source Water Quality	3
Table 3-1.   Predemonstration Study Activities and Completion Dates	7
Table 3-2.   Evaluation Objectives and Supporting Data Collection Activities	7
Table 3-3.   Sample Collection Schedule and Analyses	9
Table 3-4.   Summary of Jar Test Parameters	11
Table 4-1.   BSLMHP, MN Water Quality Data	16
Table 4-2.   Physical Properties of M2 Macrolite® Media	17
Table 4-3.   Design Specifications for Macrolite® CP-213f Pressure Filtration System	19
Table 4-4.   System Operation from July 13, 2005 to January 17, 2006	24
Table 4-5.   Control Disc Size and Throughput between Backwash Cycles	25
Table 4-6.   Summary of Arsenic, Iron, and Manganese Analytical Results	27
Table 4-7.   Summary of Other Water Quality Parameter Sampling Results	28
Table 4-8.   Control Disc Sizes and Corresponding Occurrences with High Total Arsenic and
            Iron Concentrations	32
                                             vn

-------
Table 4-9.   Correlations Between Pump Stroke Length, KMnO4 Dosage, and Total and Soluble
            Manganese Concentrations	33
Table 4-10.  Jar Test Results for Macrolite®-Treated Water	35
Table 4-11.  Backwash Water Sampling Results	39
Table 4-12.  Distribution Sampling Results	40
Table 4-13.  Summary of Capital Investment for BSLMHP Treatment System	42
Table 4-14.  O&M Cost for BSLMHP, MN Treatment System	43
                                             Vlll

-------
                              ABBREVIATIONS AND ACRONYMS
AAL          American Analytical Laboratories
Al            aluminum
AM           adsorptive media
As            arsenic

BSLMHP      Big Sauk Lake Mobile Home Park
BTU-hr       British Thermal Units per hour

Ca            calcium
Cu            copper

DO           dissolved oxygen
DOM         dissolved organic matter

EPA          U.S. Environmental Protection Agency

F             fluoride
Fe            iron
FRP          fiberglass reinforced plastic

GFH          granular ferric hydroxide
gpd           gallons per day
gpm          gallons per minute

HOPE        high-density polyethylene
hp            horsepower

ICP-MS       inductively coupled plasma-mass spectrometry
ID            identification
IX            ion exchange

LCR          Lead and Copper Rule

MCL          maximum contaminant level
MDL          method detection limit
MDH         Minnesota Department of Health
Mg           magnesium
Mn           manganese
Mo           molybdenum
mV           millivolts

Na            sodium
NA           not applicable
NaOCl        sodium hypochlorite
NRMRL       National Risk Management Research Laboratory
NTU          nephelometric turbidity units

O&M         operation and maintenance
                                            IX

-------
OIT          Oregon Institute of Technology
ORD          Office of Research and Development
ORP          oxidation-reduction potential
PE            professional engineer
P&ID         piping and instrumentation diagrams
POU          point-of-use
psi            pounds per square inch
PVC          polyvinyl chloride

QA           quality assurance
QAPP         quality assurance project plan
QA/QC       quality assurance/quality control

RO           reverse osmosis
RPD          relative percent difference

Sb            antimony
SDWA        Safe Drinking Water Act
SMCL        Secondary Maximum Contaminant Level

TBD          to be determined
TCLP         Toxicity Characteristic Leaching Procedure
TDS          total dissolved solids
TOC          total organic carbon
TSS          total suspended solids
TTHM        total trihalomethane

V            vanadium

-------
                                  ACKNOWLEDGMENTS

The authors wish to extend their sincere appreciation to the operator, Mr. Bill Tix of the Big Sauk Lake
Mobile Home Park (BSLMHP) in Sauk Centre, MN.  Mr. Tix monitored the treatment system daily and
collected samples from the treatment and distribution system on a regular schedule throughout this
reporting period. This performance evaluation would not have been possible without his efforts.
                                              XI

-------
                                    1.0:  INTRODUCTION

The Safe Drinking Water Act (SOWA) mandates that U.S. Environmental Protection Agency (EPA)
identify and regulate drinking water contaminants that may have adverse human health effects and that
are known or anticipated to occur in public water supply systems. In 1975 under the SDWA, EPA
established a maximum contaminant level (MCL) for arsenic at 0.05 mg/L.  Amended in 1996, the
SDWA required that EPA develop an arsenic research strategy and publish a proposal to revise the
arsenic MCL by January 2000.  On January 18, 2001, EPA finalized the arsenic MCL at 0.01 mg/L (EPA,
2001). In order to clarify the implementation of the original rule, EPA revised the rule on March 25, 2003
to express the MCL as 0.010 mg/L (10 (ig/L) (EPA, 2003). The final rule requires all  community and
non-transient, non-community water systems to comply with the new standard by January 23, 2006.

In October 2001, EPA announced an initiative for additional research and development of cost-effective
technologies to help small community water systems  (<10,000 customers) meet the  new arsenic  standard
and to provide technical assistance to operators of small systems in order to reduce compliance costs.  As
part of this Arsenic Rule Implementation Research Program,  EPA's Office of Research and Development
(ORD) proposed a project to conduct a series of full-scale, on-site demonstrations of arsenic removal
technologies, process modifications, and engineering  approaches applicable to small systems. Shortly
thereafter, an announcement was published in the Federal Register requesting water utilities interested in
participating in Round 1 of this EPA-sponsored demonstration program to provide information on their
water systems. In June 2002, EPA selected 17 out of 115 sites to be the host sites for the demonstration
studies.

In September 2002, EPA solicited proposals from engineering firms and vendors for cost-effective arsenic
removal treatment technologies  for the 17 host sites.  EPA received 70 technical proposals for the 17 host
sites, with each site receiving one to six proposals. In April 2003, an independent technical panel
reviewed the proposals and provided its recommendations to  EPA on the technologies that it determined
were acceptable for the demonstration at each site.  Because of funding limitations and other technical
reasons, only 12 of the 17 sites were selected for the demonstration project.  Using the information
provided by the review panel, EPA, in cooperation with the host sites and the drinking water programs of
the respective states, selected one technical proposal for each site. As of February 2007, 11 of the  12
systems have been operational and the performance evaluation of six systems has been completed.

In 2003, EPA initiated Round 2  arsenic technology demonstration projects that were partially funded with
Congressional add-on funding to the EPA budget. In June 2003, EPA selected 32 potential  demonstration
sites and the Big  Sauk Lake  Mobile Home Park (BSLMHP) in Sauk Centre, MN was one of them.

In September 2003, EPA, again  solicited proposals from engineering  firms and vendors for  arsenic
removal technologies. EPA received 148 technical proposals for the 32 host sites, with each site
receiving from two to eight proposals. In April 2004, another technical panel was convened by EPA to
review the proposals and provide recommendations to EPA with the number of proposals per site ranging
from none (for two sites) to a maximum of four. The final selection of the treatment technology at the
sites that received at least one proposal was made, again through a joint effort by EPA, the state
regulators, and the host site. Since then, four sites have withdrawn from the demonstration  program,
reducing the number of sites to 28.  Kinetico's Macrolite® Arsenic Removal Technology was selected for
demonstration at the BSLMHP site.

-------
1.1        Treatment Technologies for Arsenic Removal

The technologies selected for the Round 1 and Round 2 demonstration host sites include 25 adsorptive
media (AM) systems (the Oregon Institute of Technology [OIT] site has three AM systems), 13 coagula-
tion/filtration (C/F) systems, two ion exchange (IX) systems, and 17 point-of-use (POU) units (including
nine under-the-sink reverse osmosis [RO] units at the Sunset Ranch Development site and eight AM units
at the OIT site), and one system modification. Table 1-1 summarizes the locations, technologies, vendors,
system flowrates, and key source water quality parameters (including As, Fe, and pH) at the 40
demonstration sites. An overview of the technology selection and system design for the 12 Round 1
demonstration sites and the associated capital costs is provided in two EPA reports (Wang et al., 2004 and
Chen et al., 2004), which are posted on the EPA website at
http://www.epa.gov/ORD/NRMRL/wswrd/dw/arsenic/index.html.

1.2        Project Objectives

The objective of the Round 1 and Round 2 arsenic demonstration program is to conduct 40 full-scale
arsenic treatment technology demonstration studies on the removal of arsenic from drinking water
supplies. The specific objectives are to:

           •   Evaluate the performance of the arsenic removal technologies for use on small
               systems.

           •   Determine the required system operation and maintenance (O&M) and operator
               skill levels.

           •   Characterize process residuals produced by the technologies.

           •   Determine the capital and O&M cost of the technologies.

This report summarizes the performance of the Kinetico Macrolite® CP-213f arsenic removal system at
BSLMHP in Sauk Centre, MN from July 13, 2005, through January 17, 2006. The types of data collected
included system operation, water quality (both across the treatment train and in the distribution system),
residuals, and capital and preliminary O&M cost.

-------
Table 1-1. Summary of Round 1 and Round 2 Arsenic Removal Demonstration Locations, Technologies, and Source Water Quality
Demonstration
Location
Site Name
Technology (Media)
Vendor
Design
Flowrate
(gpm)
Source Water Quality
As
(HS/L)
Fe
(HS/L)
PH
(S.U.)
Northeast/Ohio
Wales, ME
Bow,NH
Goffstown, NH
Rollinsford, NH
Dummerston, VT
Felton, DE
Stevensville, MD
Houghton, NY(d)
Newark, OH
Springfield, OH
Springbrook Mobile Home Park
White Rock Water Company
Orchard Highlands Subdivision
Rollinsford Water and Sewer District
Charette Mobile Home Park
Town of Felton
Queen Anne's County
Town of Caneadea
Buckeye Lake Head Start Building
Chateau Estates Mobile Home Park
AM (A/I Complex)
AM (G2)
AM(E33)
AM (E33)
AM (A/I Complex)
C/F (Macrolite)
AM (E33)
C/F (Macrolite)
AM (ARM 200)
AM (E33)
ATS
ADI
AdEdge
AdEdge
ATS
Kinetico
STS
Kinetico
Kinetico
AdEdge
14
70W
10
100
22
375
300
550
10
250(e)
38W
39
33
36W
30
30W
19W
27W
15W
25W
<25
<25
<25
46
<25
48
270W
l,806(c)
1,312W
1,615W
8.6
7.7
6.9
8.2
7.9
8.2
7.3
7.6
7.6
7.3
Great Lakes/Interior Plains
Brown City, MI
Pentwater, MI
Sandusky, MI
Delavan, WI
Greenville, WI
Climax, MN
Sabin, MN
Sauk Centre, MN
Stewart, MN
Lidgerwood, ND
City of Brown City
Village of Pentwater
City of Sandusky
Vintage on the Ponds
Town of Greenville
City of Climax
City of Sabin
Big Sauk Lake Mobile Home Park
City of Stewart
City of Lidgerwood
AM (E33)
C/F (Macrolite)
C/F (Aeralater)
C/F (Macrolite)
C/F (Macrolite)
C/F (Macrolite)
C/F (Macrolite)
C/F (Macrolite)
C/F&AM (E33)
Process Modification
STS
Kinetico
Siemens
Kinetico
Kinetico
Kinetico
Kinetico
Kinetico
AdEdge
Kinetico
640
400
340(e)
40
375
140
250
20
250
250
14w
13W
16W
20W
17
39W
34
25W
42W
146W
127(c)
466(c)
1,387W
l,499(c)
7827(c)
546W
l,470(c)
3,078(c)
1,344W
1,325W
7.3
6.9
6.9
7.5
7.3
7.4
7.3
7.1
7.7
7.2
Midwest/Southwest
Arnaudville, LA
Alvin, TX
Bruni, TX
Wellman, TX
Anthony, NM
Nambe Pueblo, NM
Taos, NM
Rimrock, AZ
Tohono O'odham
Nation, AZ
Valley Vista, AZ
United Water Systems
Oak Manor Municipal Utility District
Webb Consolidated Independent School
District
City of Wellman
Desert Sands Mutual Domestic Water
Consumers Association
Nambe Pueblo Reservation
Town of Taos
Arizona Water Company
Tohono O'odham Utility Authority
Arizona Water Company
C/F (Macrolite)
AM (E33)
AM (E33)
AM (E33)
AM (E33)
AM(E33)
AM (E33)
AM (E33)
AM(E33)
AM (AAFS50/ARM 200)
Kinetico
STS
AdEdge
AdEdge
STS
AdEdge
STS
AdEdge
AdEdge
Kinetico
770(e)
150
40
100
320
145
450
90W
50
37
35W
19w
56(a)
45
23(a)
33
14
50
32
41
2,068(c)
95
<25
<25
39
<25
59
170
<25
<25
7.0
7.8
8.0
7.7
7.7
8.5
9.5
7.2
8.2
7.8

-------
                Table 1-1. Summary of Round 1 and Round 2 Arsenic Removal Demonstration Locations, Technologies, and
                                                        Source Water Quality (Continued)
Demonstration
Location
Site Name
Technology (Media)
Vendor
Design
Flowrate
(gpm)
Source Water Quality
As
(MS/L)
Fe
(ug/L)| PH
Far West
Three Forks, MT
Fruitland, ID
Homedale, ID
Okanogan, WA
Klamath Falls, OR
Vale, OR
Reno, NV
Susanville, CA
Lake Isabella, CA
Tehachapi, CA
City of Three Forks
City of Fruitland
Sunset Ranch Development
City of Okanogan
Oregon Institute of Technology
City of Vale
South Truckee Meadows General
Improvement District
Richmond School District
Upper Bodfish Well CH2-A
Golden Hills Community Service District
C/F (Macrolite)
IX (A300E)
POU RO(1)
C/F (Electromedia-I)
POE AM (Adsorbsia/ARM 200/ArsenXnp)
and POU AM (ARM 200)(g)
IX (Arsenex II)
AM (GFH)
AM (A/I Complex)
AM (HIX)
AM (Isolux)
Kinetico
Kinetico
Kinetico
Filtronics
Kinetico
Kinetico
Siemens
ATS
VEETech
MEI
250
250
75gpd
750
60/60/30
525
350
12
50
150
64
44
52
18
33
17
39
37(a)
35
15
<25
<25
134
69w
<25
<25
<25
125
125
<25
7.5
7.4
7.5
8.0
7.9
7.5
7.4
7.5
7.5
6.9
AM = adsorptive media process; C/F = coagulation/filtration; HIX = hybrid ion exchanger; IX = ion exchange process; RO = reverse osmosis
ATS = Aquatic Treatment Systems; MEI = Magnesium Elektron, Inc.; STS = Severn Trent Services
(a)  Arsenic existing mostly as As(III).
(b)  Design flowrate reduced by 50% due to system reconfiguration from parallel to series operation.
(c)  Iron existing mostly as Fe(II).
(d)  Replaced Village of Lyman, NE site which withdrew from program in June 2006.
(e)  Facilities upgraded systems in Springfield, OH from 150 to 250 gpm, Sandusky, MI from 210 to 340 gpm, and Arnaudville, LA from 385 to 770 gpm.
(f)  Including nine residential units.
(g)  Including eight under-the-sink units.

-------
                         Section 2.0:  SUMMARY AND CONCLUSIONS

Based on the information collected during the first six months of system operation, the following
conclusions were made relating to the overall objectives of the treatment technology demonstration study.

Performance of the arsenic removal technology for use  on small systems:

           •   KMnO4, selected over chlorine due to the presence of elevated total organic carbon
               (TOC) levels (i.e., 3.2 to 4.8 mg/L) and high formation potential of disinfection
               byproducts in source water, was effective in oxidizing As (III), reducing its
               concentrations from an average of 23 (ig/L to an average of 1.9 (ig/L after the contact
               tanks.  Arsenic after the contact tanks was converted mostly to particulate arsenic, which
               was ready to be removed by the Macrolite® filters. KMnO4 also was effective in
               oxidizing soluble iron despite the presence of high TOC levels in raw water.

           •   Even at significantly reduced loading rates of 0.3 to 4.1 gpm/ft2, the Macrolite® filters did
               not consistently remove arsenic to the 10-(ig/L level. The removal at the target level was
               achieved during only 16 of the 24 weekly sampling events.  Breakthrough of particulate
               arsenic and iron at concentrations as high as 21.5 and 2,363 (ig/L, respectively, from the
               Macrolite® filters was the main reason for not meeting the target arsenic MCL.

           •   Because of the observed particulate breakthrough, the filter backwash frequency had to
               be repeatedly increased from the initial field  setting of once every 2,743 gal to once every
               1,714 gal during the first six-month study period. As a result, the backwash water
               generation ratios were increased correspondingly from 5.5 to 7.2%.  Further adjustments
               to the backwash frequency will have to be implemented during the next six-month study
               period because of continuing particulate breakthrough.

           •   Oxidation of Mn(II) with KMnO4 was affected by the presence of dissolved organic
               matter (DOM) in raw water, forming fine colloidal particles not retainable by 0.45-(im
               disc filters. A dosage of at least 4.58 mg/L appeared to be needed to form filterable
               manganese solids, thus reducing the "soluble" manganese levels, as measured by using
               0.45-(im disc filters, from as high as 946 (ig/L to below  166 (ig/L after the contact tanks.

           •   Elevated total phosphorous levels ranging from 0.4 to 0.6 mg/L (as P) were detected in
               raw water. Total phosphorous existed primarily as total hydrolysable phosphorous,
               including polyphosphorous and some organic phosphorous  according to EPA Method
               365.4. No organopesticides, however, were present in source water according the EPA
               Method 507. Potential sources of elevated phosphorous included non-point source
               discharge from septic systems, agriculture, and urban runoff.

               The Macrolite® filtration process was able to remove about 80% of total phosphorous.

Simplicity of required system O&Mand operator skill levels:

           •   The daily demand on the operator was about 5 min, which included performing O&M
               activities such as mixing the KMnO4 solution, measuring the KMnO4 consumption,
               installing the hour meter, and working with the vendor to troubleshoot and perform minor
               on-site repairs.

-------
           •   The system experienced some operational issues, which were related primarily to the
               backwash frequency. The throughput between backwash cycles had to be reduced
               repeatedly by using different control discs installed on top of the duplex units.

           •   There was no significant downtime during the first six months of system operation.
               However, the primary well pump developed a leak on January 4, 2006, which was
               repaired within a day. Further, a backwash  malfunction occurred on August 9, 2005,
               which necessitated repairs by the vendor.

Process residuals produced by the technology:

           •   Based on the average of samples taken during several backwash events, the amount of
               solids produced per backwash event was 0.17 Ib, which was composed of approximately
               0.08 Ib of iron and 4.4xlO"4 Ib of arsenic.

Cost-effectiveness of the technology:

           •   Using the system's rated capacity of 20 gal/min  (gpm) (or 28,800 gal/day [gpd]), the
               capital cost was $3,177 per gpm (or $2.21 per gpd). The unit capital cost was
               $0.57/1,000 gal if the system operated at  100% utilization rate. The system's real unit
               cost was 3.47/1,000 gal, based on 3.4 hr/day of system operation and 863,470 of water
               production.

           •   The O&M cost was $0.43/1000 gal, based on labor, chemical usage, and electricity
               consumption.

-------
                                 3.0:  MATERIALS AND METHODS
3.1
General Project Approach
Following the pre-demonstration activities summarized in Table 3-1, the performance evaluation of the
Macrolite® treatment system began on July 13, 2005. Table 3-2 summarizes the types of data collected
and/or considered as part of the technology evaluation process. The overall system performance was
evaluated based on its ability to consistently remove arsenic to below the target MCL of 10 |o,g/L through
the collection of water samples across the treatment train.  The reliability of the system was evaluated by
tracking the unscheduled system downtime and frequency and extent of repair and replacement. The
unscheduled downtime and repair information were recorded by the plant operator on a Repair and
Maintenance Log Sheet.

               Table 3-1. Predemonstration Study Activities and Completion Dates
Activity
Introductory Meeting Held
Request for Quotation Issued to Vendor
Vendor Quotation Received
Purchase Order Established
Letter of Understanding Issued
Letter Report Issued
Engineering Package Submitted to MDH
Permit Granted by MDH
Study Plan Issued
Macrolite® Unit Shipped
Macrolite® Unit Delivered
System Installation Completed
System Shakedown Completed
Performance Evaluation Begun
Date
08/31/04
12/06/04
02/17/05
02/24/05
01/10/05
03/09/05
03/28/05
06/14/05
06/21/05
06/10/05
06/16/05
06/24/05
07/03/05
07/13/05
               MDH = Minnesota Department of Health
           Table 3-2. Evaluation Objectives and Supporting Data Collection Activities
Evaluation Objective
Performance
Reliability
System O&M and Operator
Skill Requirements
Residual Management
Cost-Effectiveness
Data Collection
-Ability to consistently meet 10 |ag/L of arsenic in treated water
-Unscheduled system downtime
-Frequency and extent of repairs including a description of problems,
materials and supplies needed, and associated labor and cost
-Pre- and post-treatment requirements
-Level of automation for system operation and data collection
-Staffing requirements including number of operators and laborers
-Task analysis of preventive maintenance including number, frequency, and
complexity of tasks
-Chemical handling and inventory requirements
-General knowledge needed for relevant chemical processes and health and
safety practices
-Quantity and characteristics of aqueous and solid residuals generated by
system operation
-Capital cost for equipment, engineering, and installation
-O&M cost for chemical usage, electricity consumption, and labor

-------
The O&M and operator skill requirements were evaluated based on a combination of quantitative data
and qualitative considerations, including the need for pre- and/or post-treatment, level of system
automation, extent of preventative maintenance activities, frequency of chemical and/or media handling
and inventory, and general knowledge needed for relevant chemical processes and related health and
safety practices.  The staffing requirements for the system operation were recorded on an Operator Labor
Hour Log Sheet.

The quantity of aqueous and solid residuals generated was estimated by tracking the volume of backwash
water produced during each backwash cycle. Backwash water was sampled and analyzed for chemical
characteristics.

The cost of the system was evaluated based on the capital cost per gal/min (gpm) (or gal/day [gpd]) of
design capacity and the O&M cost per 1,000 gal of water treated.  This task required tracking the capital
cost for equipment, engineering, and installation, as well as the O&M cost for media replacement and
disposal, chemical supply, electricity usage, and labor.

3.2        System O&M and Cost Data Collection

The plant operator performed daily, weekly and monthly  system O&M and data collection according to
instructions provided by the vendor and Battelle.  On a daily basis, with the exception of Saturdays and
Sundays, the plant operator recorded system operational data, such as pressure, flowrate, volume, and
hour meter readings on a Daily System Operation Log Sheet; checked the potassium permanganate
(KMnO4) tank level; and conducted visual inspections to  ensure normal system  operations. If any
problem occurred, the plant operator contacted the Battelle  Study Lead, who determined if the vendor
should be contacted for troubleshooting. The plant operator recorded all  relevant information, including
the problems encountered, corrective actions taken; materials and supplies used, and associated cost and
labor required, on a Repair and Maintenance Log Sheet.  On a weekly basis, the plant operator measured
several water quality parameters on-site, including temperature, pH, dissolved oxygen (DO), and
oxidation-reduction potential (ORP), and recorded the data  on a Weekly Water Quality Parameters Log
Sheet. During the six-month study period, the system was backwashed automatically, except during the
monthly backwash sampling events when the system was backwashed manually to capture the required
backwash samples. Monthly backwash data also were recorded on a Backwash Log Sheet.

The capital cost for the arsenic removal system consisted of the cost for equipment, site engineering, and
system installation. The O&M cost consisted of the cost for chemical usage, electricity consumption, and
labor. Consumption of KMnO4 was tracked on the Daily System Operation Log Sheet.  Electricity usage
was estimated based on the hours of operation and the chemical feed pump motor size. Labor for various
activities, such as routine  system O&M, troubleshooting and repairs, and demonstration-related work,
were tracked using an Operator Labor Hour Log Sheet. The routine system O&M included activities such
as completing field logs, replenishing the KMnO4 solution,  ordering supplies, performing system
inspections, and others as recommended by the vendor. The labor for demonstration-related work,
including activities such as performing field measurements, collecting and shipping samples, and
communicating with the Battelle Study Lead and the vendor, was recorded, but not used for the cost
analysis.

3.3        Sample Collection Procedures and Schedules

To evaluate system performance, samples were collected at the wellhead, across the treatment system,
during Macrolite® filter backwash, and from the distribution system. The sampling schedules and
analytes measured during each sampling event are listed in  Table 3-3. In addition,  Figure 3-1 presents a
flow diagram of the treatment system along with the analytes and schedules at each sampling location.

-------
                    Table 3-3. Sample Collection Schedule and Analyses
Sample
Type
Source Water












Treatment
Plant Water











Distribution
Water







Backwash
Water



Residual
Sludge

Sample Locations'3*
At Wellhead (IN)












At Wellhead (IN),
After Contact Tanks
(AC),
After Tanks A/B
(TA/TB),
After Tank C/D
(TC/TD)


At Wellhead (IN),
After Contact Tanks
(AC),
After Tanks TA/TB
and TC/TD
Combined (TT)


Three LCR
Residences







Backwash
Discharge Line



At Backwash
Discharge Point
No. of
Samples
1












4





3






3








2




2-3


Frequency
Once
(during
initial site
visit)









Weekly





Monthly






Monthly








Monthly




TBD


Analytes
On-site: pH, temperature,
DO, and ORP

Off-site: As(III), As(V),
As (total and soluble),
Fe (total and soluble),
Mn (total and soluble),
U (total and soluble),
V (total and soluble),
Na, Ca, Mg, Cl, F, NO3,
NO2, NH3, SO4, SiO2, P,
turbidity, alkalinity, TDS,
andTOC
On-site: pH, temperature,
DO, and ORP
Off-site: As(total),
Fe(total), Mn(total), SiO2,
P, turbidity, and
alkalinity


Same as weekly analytes
shown above plus the
following:
Off-site: As (soluble),
As(III), As(V), Fe
(soluble), Mn (soluble),
Ca, Mg, F, NO3, SO4, and
TOC
As (total), Fe (total), Mn
(total), Cu, Pb, pH, and
alkalinity






As (total and soluble),
Fe (total and soluble),
Mn (total and soluble),
pH, turbidity, TDS, and
TSS
TCLP Metals

Date(s) Samples
Collected
08/31/04












07/20/05, 07/26/05,
08/02/05, 08/24/05,
08/31/05,09/07/05,
09/14/05, 09/27/05,
10/05/05, 10/12/05,
10/26/05, 11/02/05,
11/09/05, 11/29/05,
12/14/05, 01/10/06,
01/17/06
07/13/05, 08/18/05,
09/20/05, 10/19/05,
11/15/05, 12/08/05,
01/05/06



Baseline
Sampling(b)
02/16/05, 03/23/05
04/19/05, 05/23/05
Monthly Sampling:
07/26/05, 09/07/05,
09/27/05, 11/02/05
11/29/05, 12/15/05,
01/17/06
09/08/05, 09/20/05,
10/12/05, 11/15/05,
12/08/05, 01/10/06


TBD

(a)  Abbreviation corresponding to sample location in Figure 3 -1.
(b)  Sampling events performed before system startup.
TBD = to be determined.

-------
                                                 INFLUENT
        pH, temperature^, DO,
       As speciation, Fe (total and soluble),
                   Mn (total and soluble), -
          Ca, Mg, F, NO3, SO4, SiO2, PO4,
                 TOC, turbidity, alkalinity
 pH, temperature^), DO, ORPW,
As speciation, Fe (total and soluble),
            Mn (total and soluble),-^	,
   Ca, Mg, F, N03, S04, SiO2, PO4,
          TOC, turbidity, alkalinity
                                                                       Sauk Centre, MN
                                                                 Macrolite® Arsenic Removal System
                                                                        Design Flow: 20 gpm
1
^ 	 1 DA:K
r
CONTACT TANKS
                                                                                      Weekly
                                                                      pHM, temperature^), DOM,
                                                                     •As (total), Fe (total), Mn (total),
                                                                      SiO2, PO4, turbidity, alkalinity
         pH, TSS, TDS,
              turbidity,
   As (total and soluble),
   Fe (total and soluble),
   Mn (total and soluble)
Footnote
(a) On-site analyses
                                        DISTRIBUTION
                                           SYSTEM
                                                                             pH
-------
Specific sampling requirements for analytical methods, sample volumes, containers, preservation, and
holding times are presented in Table 4-1 of the EPA- Quality Assurance Project Plan (QAPP) (Battelle,
2004). The procedure for arsenic speciation is described in Appendix A of the QAPP.

3.3.1       Source Water.  During the initial visit to the site, one set of source water samples was
collected and speciated using an arsenic speciation kit (see Section 3.4.1).  The sample tap was flushed for
several minutes before sampling; special care was taken to avoid agitation, which might cause unwanted
oxidation. Analytes for the source water samples are listed in Table 3-3.

3.3.2       Treatment Plant Water. During the system performance evaluation study, the plant
operator collected samples weekly, on a four-week cycle, for on- and off-site analyses. For the first week
of each four-week cycle, samples taken at the wellhead (IN), afte the contact tanks (AC), and after Tanks
A/B and Tanks C/D  combined (TT), were speciated on-site and analyzed for the analytes listed in Table
3-3 for monthly treatment plant water. For the next three weeks, samples were collected at IN, AC, after
Tanks A/B (TA/TB) and after Tanks C/D (TC/TD) and analyzed for the analytes listed in Table 3-3 for
the weekly treatment plant water.

3.3.3       Special  Study - KMnO4 Jar Tests. Because significantly elevated soluble manganese
concentrations were  measured in the treated water after the Macrolite® filters, a series of jar tests were
conducted at Battelle's laboratories on November 7, 2005, using the treated water taken at the TT location
from the site to determine an appropriate KMnO4 dosage for complete oxidation of Mn(II) and formation
of filterable manganese  solids.

The jar tests consisted of six 1-L jars of the treated water with increasing dosages of KMnO4 ranging from
1.0 to 3.0 mg/L (Table 3-4). One jar was used as a control with no KMnO4 addition. The jars were
placed on a Phipps & Byrd overhead stirrer/jar tester with an illuminated base. The jars were mixed for a
total of 31 min at various mixing speeds: 200 rpm for 1 min immediately after the KMnO4 addition,
followed by 100 rpm for 19 min and 28 rpm for 11 min. After the specified contact time, the supernatant
in each jar was filtered with 0.20-|o,m disc filters and analyzed for arsenic, iron, and manganese using
Inductively Coupled Plasma-Mass Spectrometry (ICP-MS). The pH and ORP values of the content in
each jar also were measured using a VWR Symphony SP90M5 Handheld Multimeter at the beginning
and end of each jar test.  The results of the jar tests are discussed in Section 4.5.1.
                          Table 3-4. Summary of Jar Test Parameters
Parameter
Mix Time (min)(a)
KMnO4 (mg/L)
Jarl
31
0
Jar 2
31
1.0
Jar 3
31
1.5
Jar 4
31
2.0
Jar 5
31
2.5
Jar 6
31
3.0
         (a)  Mixing Speeds: 1 min at 200 rpm, 19 min at 100 rpm, and 11 min at 28 rpm.
3.3.4       Backwash Water. Backwash samples were collected monthly by the plant operator.  One
backwash water sample was collected as one of the tanks in each duplex unit was backwashed during
each of the first six monthly sampling events.  For the first three sampling events, one grab sample was
taken as the bulk of solids/water mixture was being discharged from the sample tap located on the
backwash water discharge line but before the backwash totalizer. Unfiltered samples sent to American
Analytical Laboratories (AAL) for pH, total dissolved solids (TDS), and turbidity measurements.  Filtered
samples using 0.45-(im disc filters were sent to Battelle's ICP-MS laboratory for soluble As, Fe, and Mn
analyses. Starting from November 15, 2005, during the fourth sampling event, the sampling procedure
                                              11

-------
was modified to include the collection of composite samples for total As, Fe, and Mn as well as TSS
analyses. This modified procedure involved diverting a portion of backwash water at approximately 1
gpm into a clean, 32-gal plastic container over the duration of the backwash for each set of duplex tanks.
After the content in the container was thoroughly mixed, composite samples were collected and /or
filtered on-site with a 0.45-(im filters. Analytes for the backwash samples are listed in Table 3-3.

3.3.5      Residual Solids.  Residual solids produced by the treatment process included backwash
solids, which were not collected during the initial six months of this demonstration.

3.3.6      Distribution System Water.  Samples were collected from the distribution system to
determine the impact of the arsenic treatment system on the water chemistry in the distribution system,
specifically, the arsenic, lead,  and copper levels.  Prior to the system startup, from February to May 2005,
four sets of baseline distribution water samples were collected from three residences within the
disribution system. Following the system  startup, distribution system sampling continued on a monthly
basis at the same three locations.

The three homes selected for the sampling had been included in the  Park's Lead and Copper Rule (LCR)
sampling. The homeowners collected samples following an instruction sheet developed according to the
Lead and Copper Monitoring  and Reporting Guidance for Public Water Systems  (EPA, 2002). First draw
samples were collected from a cold-water faucet located upstream of the softener in each home.  (Note
that the samples thus collected were not from a frequently used kitchen or bathroom faucet nor from a
faucet that was commonly used for human consumption.) To ensure collection of stagnant water, the
faucet were not used for at least  6 hr. Dates and times of sample collection and last water usage were
recorded for calculations of the stagnation time. Analytes for the distribution system water are listed in
Table 3-3. Arsenic speciation was not performed for the distribution water samples.

3.4        Sampling Logistics

3.4.1      Preparation of Arsenic Speciation Kits. The arsenic field speciation method uses an anion
exchange resin column to separate the soluble arsenic species, As(V) and As(III)  (Edwards et al., 1998).
Resin columns were prepared  in batches at Battelle laboratories according to the procedures detailed in
Appendix A of the EPA-endorsed QAPP (Battelle, 2004).

3.4.2      Preparation of Sample Coolers. For each sampling event, a sample cooler was prepared
with the appropriate number and type of sample bottles, disc filters, and/or speciation kits. All sample
bottles were new and contained appropriate preservatives. Each sample bottle was affixed with a pre-
printed, colored-coded label consisting of the sample identification (ID), date and time of sample
collection, collector's name, site location,  sample destination, analysis required, and preservative. The
sample ID consisted of a two-letter code for a specific water facility, sampling date, a two-letter code for
a specific sampling location, and a one-letter code designating the arsenic speciation bottle (if necessary).
The sampling locations at the treatment plant were color-coded for easy identification.  The labeled
bottles for each sampling locations were placed in separate Ziplock™ bags and packed in the cooler.

In addition, all sampling-and shipping-related materials, such as disposable gloves, sampling instructions,
chain-of-custody forms, prepaid/addressed FedEx air bills, and bubble wrap, was included. The chain-of-
custody forms and air bills were complete  except for the operator' signature and the sample dates and
times.  After preparation, the sample cooler was sent to the site via FedEx for the following week's
sampling event.

3.4.3      Sample Shipping and Handling. After sample collection, samples for off-site analyses were
packed carefully in the  original coolers with wet ice and shipped to Battelle. Upon receipt, the sample
                                               12

-------
custodian verified that all samples indicated on the chain-of-custody forms were included and intact.
Sample IDs were checked against the chain-of-custody forms and the samples were logged into the
laboratory sample receipt log. Discrepancies noted by the sample custodian were addressed with the plant
operator by the Battelle Study Lead.

Samples for metal analylses were stored at Battelle's ICP-MS Laboratory. Samples for other water
quality analyses were packed in separate coolers and picked up by couriers from American Analytical
Laboratories (AAL) in Columbus, OH and TCCI Laboratories (TCCI) in Lexington, OH, both of which
were under contract with Battelle for this demonstration study. The chain-of-custody forms remained
with the samples from the time of preparation through analysis and final disposition.  All samples were
archived by the appropriate laboratories for the respective duration of the required hold time and disposed
of properly thereafter.

3.5         Analytical Procedures

The analytical procedures described in detail in  Section 4.0 of the EPA-endorsed QAPP (Battelle, 2004)
were followed by Battelle ICP-MS, AAL, and TCCI Laboratories.  Laboratory quality assuarnce/quality
control (QA/QC) of all methods followed the prescribed guidelines. Data quality in terms of precision,
accuracy, method detection limits (MDL), and completeness met the criteria estrablished in the QAPP
(i.e., relative percent difference [RPD] of 20%, percent recovery of 80-120%, and completeness of 80%).
The quality assurance (QA) data associated with each analyte will be presented and evaluated in a QA/QC
Summary Report to be prepared under separate cover upon completion of the  Arsenic Demonstration
Project.

Field measurements of pH, temperature, DO, and ORP were conducted by the plant operator using a
VWR Symphony SP90M5 Handheld Multimeter, which was calibrated for pH and DO prior to use
following the procedures provided in the user's manual. The ORP probe also was checked for accuracy
by measuring the ORP of a standard solution and comparing it to the expected value.  The plant operator
collected a water sample in a clean, plastic beaker and placed the Symphony SP90M5 probe in the beaker
until a stable value was obtained. The plant operator also performed free and total chlorine measurements
using Hach chlorine test kits following the user's manual.
                                              13

-------
4.1
                              4.0:  RESULTS AND DISCUSSION
Facility Description and Pre-Existing Treatment System Infrastructure
Located at 43987 County Road 24 in Sauk Centre, MN, BSLMHP had a water supply system sized for up
to 50 mobile home connections or approximately  100 residents. There were 37 mobile homes in the park
during the study period. Prior to the demonstration study, the facility reported the average daily demand
of 7,500 gpd and the peak daily demand of 16,000 gpd. The system typically operated approximately 6
hr/day. Figure 4-1 shows the pre-existing well house at the facility.

The system was supplied intermittently by groundwater from two wells installed at a depth of
approximately 90 ft below ground surface (bgs). The new well (Well No. 2) was used as the primary well
and the old well (Well No. 1) used as a backup well.  The new well was equipped with a 1 !/2-horsepower
(hp), 4-in submersible pump installed on a 60-ft drop pipe and rated for 25 gpm at 180 ft H2O (or 78 psi).
The pump installed in the backup well reportedly had a similar capacity, but records were no longer
available.  Figure 4-2 shows the existing piping and two 62-gal Champion pressure tanks in the
wellhouse.  There was no disinfection or other treatment at the wellheads, although most residents had
water softeners in their homes.

4.1.1       Source Water Quality. Source water samples were collected on August 31, 2004, and
subsequently analyzed for the analytes shown in Table 3-3. The results of source water analyses, along
with those provided by the facility to EPA for the  demonstration site selection and those independently
collected and analyzed by EPA, MDH, and the vendor are presented in Table 4-1.

As shown in Table 4-1, total arsenic concentrations in source water extracted from both wells ranged
from 17.0 to 32.0 (ig/L.  Based on the August 31,  2004, speciation results, as much as 54% of the total
arsenic, or 13.6 (ig/L, was found to exist as As(III) and 18% existed as particulate arsenic.
                   Figure 4-1. Pre-Existing Pump House at BSLMHP, MN Site
                                              14

-------
            Figure 4-2. Existing Well Piping and Pressure Tanks at BSLMHP, MN Site
Iron concentrations in source water extracted from both wells ranged from 3,000 to 3,400 (ig/L, existing
entirely as soluble iron based on August 31, 2004 results.  A rule of thumb is that the soluble iron
concentration, expressed in mg/L, should be at least 20 times the soluble arsenic concentration, also
expressed in mg/L, for effective removal of arsenic onto iron solids (Sorg, 2002). Based on the best and
worst case scenarios, the results indicate that iron levels are 94 to 200 times higher than total arsenic
levels. As such, there is no need to supplement the natural iron for arsenic removal.  The proposed
treatment process is designed to reduce iron levels in the filtered water to below the secondary MCL of
300
Manganese levels of 130 to 150 (ig/L were elevated above the secondary MCL of 50 (ig/L.  The pH
values ranged from 7.1 to 7.4, which were within the target range of 5.5 to 8.5 for the iron removal
process.  Total organic carbon (TOC) levels at 3.9 to 4.9 mg/L were high and because of the high levels,
KMnO4 was used to oxidize iron and arsenic. The use of KMnO4 should help eliminate the formation
potential of disinfection byproducts, which could occur if pre-chlorination was implemented.  In April
2005, EPA conducted a disinfection byproduct formation test on source water and found that after 96 hr
of maintaining a chlorine residual, the total trihalomethane (TTHM) level was 0.11 mg/L, existing almost
completely as chloroform.  Note that the MCL for TTHM is 0.080 mg/L.  This further confirmed the need
to use an alternate oxidant to chlorine. Ammonia levels at 1.2 mg/L also were elevated and could
significantly increase the chlorine demand should chlorine be used as an oxidant. The turbidity of the
water at 30 NTU was high, presumably caused by iron precipitation during sample collection and transit.
Hardness ranged from 300 to 360 mg/L, silica from 21 to 25 mg/L, and sulfate from <4 to 5.4 mg/L.
Based on the historical data provided by MDH, there was no apparent difference in water quality between
Wells No. 1 and No. 2.  Total arsenic concentrations ranged from 26.0 to 32.0 (ig/L for Well No. 1 and
from 26.0 to 28.0 (ig/L for Well No. 2; total iron concentrations were 3,400 for Well No. 1 and 3,000
(ig/L for Well No. 2; and total manganese concentrations were 140 (ig/L for Well No. 1  and 130 (ig/L for
Well No. 2.
                                               15

-------
                          Table 4-1. BSLMHP, MN Water Quality Data
Parameter
Unit
Date
pH
Temperature
DO
ORP
Total Alkalinity (as CaCO3)
Hardness (as CaCO3)
Turbidity
TDS
TOC
Total N (Nitrite + Nitrate) (as N)
Nitrate (as N)
Nitrite (as N)
Ammonia (as N)
Chloride
Fluoride
Sulfate
Silica (as SiO2)
Orthophosphate (as P)
As (total)
As (soluble)
As (particulate)
As(III)
As(V)
Fe (total)
Fe (soluble)
Mn (total)
Mn (soluble)
U (total)
U (soluble)
V (total)
V (soluble)
Na (total)
Ca (soluble)
Mg (total)
Gross-Alpha
Gross-Beta
Radon
Radium-228
-
°C
mg/L
mV
mg/L
mg/L
NTU
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
Hg/L
"g/L
"g/L
"g/L
Hg/L
"g/L
ug/L
ug/L
ug/L
ug/L
ug/L
"g/L
"g/L
mg/L
mg/L
mg/L
pCi/L
pCi/L
pCi/L
pCi/L
Facility
Well 2
Data
Not
specified
7.4
NS
NS
NS
355
305
NS
NS
4.9
NS
NS
NS
NS
1.5
NS
5.2
24
0.02
26
NS
NS
NS
NS
3,200
NS
140
NS
NS
NS
NS
NS
14
72
30
NS
NS
NS
NS
Kinetico
Well 2
Data
10/14/03
7.3
NS
NS
NS
364
330
NS
NS
NS
NS
NS
NS
NS
<1.0
0.46
<4.0
21.4
0.5
17
NS
NS
NS
NS
3,060
NS
150
NS
NS
NS
NS
NS
13
81
32
NS
NS
NS
NS
Battelle
Well 2
Data
08/31/04
7.1
NS
1.48
-98
363
360
30
338
3.9
NS
0.04
O.01
1.2
<1.0
O.I
<5.0
25
0.1
25.3
20.7
4.6
13.6
7.1
3,078
3,149
150
154
O.I
0.1
0.17
0.1
17
87
35
NS
NS
NS
NS
MDH
Welll
Data
01/25/01 -
08/10/04
7.3
NS
NS
NS
350
310

-------
several different sampling points, including residences and the wellhouse distribution entry piping, ranged
from 18.4 to 28.0 |o,g/L based on MDH treated water sampling data shown in Table 4-1.
4.2
Treatment Process Description
The treatment train for the BSLMHP system includes KMnO4 oxidation, co-precipitation/adsorption, and
Macrolite® pressure filtration.  Macrolite® is an engineered ceramic filtration media manufactured by
Kinetico and approved for use in drinking water applications under NSF International (NSF) Standard 61.
Macrolite® is a low-density, spherical media, designed to allow for filtration rates, as claimed by the
vendor, up to 10 gpm/ft2, a hydraulic loading rate higher than that for most conventional filtration media.
The physical properties of this media are summarized in Table 4-2. The vendor considers Macrolite®
media chemically inert and compatible with chemicals such as oxidants and ferric chloride.
                     Table 4-2.  Physical Properties of M2 Macrolite® Media
Property
Color
Uniformity Coefficient
Sphere Size Range (mm) [mesh]
Nominal Size (mm)
Bulk Density (g/cm3) [Ib/ft3]
Specific Gravity
Value
Variable
1.1
0.21 -0.42 [40
x70]
0.3
0.86 [54]
2.05
Figure 4-3 is a schematic and Figure 4-4 a photograph of the Macrolite® CP-213f arsenic removal system.
The treatment system was operated as an on-demand system and the volume of water treated was based
on water usage.  The well pump turned on when the pressure tank pressure reached 45 psi and shut off at
60 psi. The primary system components consisted of a KMnO4 feed system (with the metering pump
interlocked with a totalizer located after the pressure tanks and prior to the treatment system), two contact
tanks, four pressure filtration tanks (two each within each duplex unit), and associated pressure and flow
instrumentation.  Various instruments were installed to track system performance, including the inlet and
outlet pressure after each filter, flowrate to the distribution system, and backwash flowrate. All plumbing
for the system was Schedule 80 PVC with the necessary valves, sampling ports, and other features. Table
4-3 summarizes the design features of the Macrolite® pressure filtration system.  The major process steps
and system components are presented as follows:

           •  Potassium Permanganate Oxidation - KMnO4 was used to oxidize As(III), Fe(II), and
              Mn(II) in source water. KMnO4 was selected to help reduce the formation potential of
              disinfection by-products due to the presence of high TOC levels in source water.  The
              KMnO4 addition system consisted of a 150-gal day tank,  a Pulsatron metering pump, and
              an overhead mixer (Figure 4-5). The working solution was prepared by adding 0.75  gal
              (or 10 Ib) of KMnO4 crystals with 97% minimum purity into 40 gal of water to form a
              3% KMnO4 solution. During the six-month study period, the 21-in diameter and 31.5-in
              tall KMnO4 tank was re-filled a total of four times when the tank level reached an
              average of 17.4 in. The KMnO4 feed pump was sized with a maximum capacity of 44
              gpd or 6.9 L/hr. However, the pump was flow-paced and  the actual rate of KMnO4
              addition varied based on the influent
                                              17

-------
                                                        Backwash Waste
    Macrolite® CP-213
Arsenic Removal System

Pressure
Tanks



-P
Raw Water from Well


^-





F-
X
'$
^r:




T

F^
ID
x
&
br
                                            s
                                            X
                                            CO
      s
      x
      CO
               to Septic
                                                           Filtered Water
                                                           to Distribution
45-60 psi
          Chemical
           Metering [~
             Pumpr
                           Retention
                             Tanks
S
x
CO
in
x
                                   Two (2) Duplex Macrolite® Filters
                                           Free Standing
Figure 4-3. Process Schematic of Macrolite® Pressure Filtration System
   Figure 4-4. Photograph of Macrolite® Pressure Filtration System
          (1.  Duplex Units, 2. Contact Tanks, 3. Pressure Tanks,
        4. Chemical Day Tank, and 5. Totalizer on Raw Water Line)
                                  18

-------
     Table 4-3.  Design Specifications for Macrolite® CP-213f Pressure Filtration System
Parameter
Value
Remarks
Pretreatment
KMnO4 Dosage (mg/L as [KMnO4])
3.3
Calculated KMnO4 demand based on arsenic, iron, and
manganese in source water; actual demand higher due
to presence of TOC in source water
Contact Tanks
Tank Size (in)
No. of Tanks
Configuration
Contact Time (min/tank)
36 D x 57
H
2
Parallel
20
205 gal each tank
-
-
Based on peak flowrate of 20 gpm; actual contact time
based on on-demand flowrate
Filtration Tanks
Tank Size (in)
Cross-Sectional Area (ft2/tank)
No. of Tanks
Configuration
Media Type
Media Quantity (ftVtank)
Freeboard Measurements (in/tank)
Filtration Rate (gpm/ft2)
Pressure Drop (psi)
Throughput before Backwash (gal)
Backwash Hydraulic Loading Rate
(gpm/ft2)
Backwash Duration (min/tank)
Wastewater Production (gal)
System Design Flowrate (gpm)
Maximum Daily Production (gpd)
Hydraulic Utilization (%)
13 D x54H
0.92
4
Parallel
Macrolite®
1.5
28
5.4
15
2,743
6.5
20
130
20
28,800
56
-
-
-
Between two duplex units and between two filtration
tanks within each duplex unit.
40/60 mesh media
20-in bed depth in each tank
Measurements taken by vendor's contractor on
12/07/05 from top of filtration tank to top of media bed
Based on a 5 gpm system flowrate through each
filtration tank; actual filtration rate based on demand
Across a clean bed
Based on initial field design
Based on a 6-gpm backwash flowrate through each
filtration tank
-
For each tank
Peak flowrate; actual flowrate based on demand
Based on peak flowrate, 24 hr/day
Estimated based on peak daily demand(a)
(a)  Based on historic peak daily demand of 16,000 gpd.

            flowrate to the treatment system. During the first six months of system operation,
            KMnO4 dosages varied from 2.1 to 6.1 mg/L. The operator indicated that the mixer was
            only turned on when the KMnO4 crystals was mixed initially with water in the day tank.

        •   Contact - Two 36-in by 57-in fiberglass reinforced plastic (FRP) contact tanks arranged
            in parallel provided at least 20 min of contact time when operating at the design (or peak)
            flowrate of 20 gpm. The longer retention time was designed to aid in the formation of
            manganese particles before Macrolite® filtration.

        •   Pressure Filtration - The filtration system consisted of downflow filtration through two
            sets of dual-pressure filtration tanks arranged in parallel. Each duplex unit was
            comprised of two 13-in by 54-in FRP tanks and a control valve. Each filtration tank was
            filled with approximately 20-in (1.5 ft3) of 40/60 mesh Macrolite® media supported by 3-
            in (0.25 ft3) of garnet underbedding. The standard operation had both tanks of a pair
            online with each pressure tank treating a maximum of 5 gpm for a hydraulic loading rate
                                           19

-------
                 Figure 4-5. KMnO4 Feed System
of 5.4 gpm/ft2. With four tanks online, the maximum system flowrate was 20 gpm.
However, as shown in Figure 4-3, the system had an on-demand configuration with two
pressure tanks located ahead of the treatment system. The actual flowrate through the
system varied based on water demand, but was limited to less than 20 gpm by flow
restrictors located on the duplex units.

The control valve (Kinetico Mach 1250) located on top of each duplex unit (Figure 4-6)
consisted of a gear stack, which determines the throughput between two consecutive
backwash cycles. The control valve consisted of three chambers: inlet, outlet, and
regeneration and only the influent water was measured/recorded by the gear stack.
          Figure 4-6. Kinetico's Mach 1250 Control Valve
                               20

-------
      Backwash Operations - Backwash was a fully automated process triggered by a pre-set
      volume throughput measured by the control valve located on top of each duplex unit.
      The spent filtration tank was backwashed with the treated water from the other tank
      within the duplex unit and the resulting wastewater was discharged to a sanitary sewer.
      The backwash duration for each tank was 20 min from start to finish including 15 min of
      backwash at a flowrate  of 6 gpm and a 5 min filter-to-rinse cycle at 6 gpm. The
      backwash used about 130 gal of the treated water per tank. As discussed in section 4.4.2,
      it was necessary to increase the frequency of backwash from the initial field setting of
      every 2,743 gal to every 1,714 gal over the  six-month study period. Figure 4-7 depicts
      the backwash flow paths for one duplex unit (labeled as Tank A and Tank B), which were
      backwashed on an alternating basis after a pre-set throughput of 1,714 gal. The major
      steps involved in the backwash process are  discussed as follows:
    Tank A
   Throughput
         TankB
        Throughput
System startup using No. 8 control disc
geared to backwash after 1,714 gal of
combined throughput from both Tanks A
andB.
Step 1.  Backwash of Tank A required
after 1,714 gal of combined throughput
from both Tanks A and B.

Step 2.  Tank A backwashed with 130 gal
of treated water from Tank B (which was
not accounted toward the set throughput
of 1,714 gal).
Step 3.  Backwash of Tank B required
after 1,714 gal of combined throughput
from both Tanks A and B.

Step 4.  TankB backwashed with 130 gal
of treated water from Tank A (which was
not accounted toward the set throughput
of 1,714 gal).
Step 5.  Backwash of Tank A required
after 1,714 gal of combined throughput
from both Tanks A and B.

Step 6.  Tank A backwashed with 130 gal
of treated water from Tank B (which was
not accounted toward the set throughput
of 1,714 gal).
Service^ackwash cycles continued as
depicted above.
Key
Throughput through Tanks A and B before Tank A Was Backwashed
Throughput through Tanks A and B before Tank B Was Backwashed
Clean Bed
    Figure 4-7. Backwash Flow Path for One Duplex Unit with Control Disc
         No. 8 and a Throughput of 1,714 gal between Backwash Cycles
                                      21

-------
               Again, both Tank A and Tank B provide the treated water in parallel.  The backwash
               cycles were continuously repeated as shown in Steps 4 through 6 during the treatment
               system operation.  One set of duplex tanks functioned as one unit and always had a
               filtration capacity between 25% (immediately after backwash of one tank at Step 4) and
               75% (right before backwash of the other tank at Step 5).

4.3        System Installation

This section provides a summary of system installation activities including permitting, building
construction, and system shakedown.

4.3.1       Permitting.  Engineering plans for the system permit application were prepared by the
vendor. The plans included diagrams and specifications for the Macrolite® CP-213f arsenic removal
system, as well as drawings detailing the connections of the new unit to the pre-existing facility
infrastructure.  The plans were submitted to MDH on March 28, 2005, and MDH granted its approval of
the application on June 14, 2005.

4.3.2       Building Construction.  The existing well house had an adequate footprint to house the
arsenic treatment system.  The permit approval issued by MDH on June 14, 2005, indicated a need for an
air gap two times the diameter of the  filter-to-waste line and a need for all chemicals to be injected on the
lower half of the influent pipe.  Figure 4-6 shows the chemical injection line located on the top half of the
influent pipe. In addition, MDH required the drain line and sewer connection to have at least a 50-ft
distance from Well No. 1 and Well No. 2 wellheads and at a lower elevation.

4.3.3       System Installation, Shakedown, and Startup. The Macrolite® system was shipped on
June 10, 2005 and delivered to the site on June 16, 2005. A subcontractor to the vendor off-loaded and
installed the system, including piping connections to the existing entry and distribution system. The
system installation was completed by June 24, 2005, and the system shakedown was completed by
JulyS, 2005.

Shakedown activities included disinfection of the contact and filtration tanks and backwash of Macrolite®
filtration media. The bacteriological  test was passed on July 1, 2005. During the startup trip in July, the
vendor conducted operator training for system O&M. Battelle arrived on-site on July 13, 2005, to
perform system inspections and conduct operator training for system sampling and data collection. The
first set of samples for the one year performance evaluation study was taken on July 13, 2005. No major
mechanical or installation issues were noted at system startup; however, several pieces of equipment
shown in the vendor's June  16, 2005  piping and instrumental diagrams (P&ID) were missing and several
installed items did not meet the permit requirements. A list of punch-list items was summarized as
follows:

           •   Install an hour meter.
           •   Install one raw water sample tap.
           •   Install one backwash sample tap.
           •   Install one sample tap after duplex units TA/TB and after duplex units TC/TD.
           •   Install one pressure gauge after duplex units TA/TB and after duplex units TC/TD.
           •   Replace the defective pressure gauge beneath the left most pressure tank.
           •   Install a level sensor on the KMnO4 day tank.
           •   Install a !/2-inch ball valve on the KMnO4 injection tube.
           •   Move the KMnO4 injection port from the top half of the influent pipe to the lower half
               per permit requirements.
                                              22

-------
           •  Verify that the air gap was two times the filter-to-waste pipe between the drain and the
              filter-to-waste pipe.

All punch-list items were resolved by the vendor by September 30, 2005.

4.4        System Operation

4.4.1       Operational Parameters. Table 4-4 summarizes the operational parameters for the first six
months of system operation, including operational time, throughput, flowrate, and pressure.  Detailed
daily operational information also is provided in Appendix A.

Between July 13, 2005, and January 17, 2006, the primary well pump operated for approximately 617 hr,
with an average daily operating time of 3.4 hr/day (compared to 6 hr/day provided by the park owner
prior to the demonstration study) based on the readings of an hour meter installed on the primary well on
September 28, 2005.  Prior to this time period, the operational time was estimated based on the wellhead
totalizer readings and an average well pump flowrate of 25 gpm. The total system throughput during the
first six months was approximately 863,470 gal based on the totalizer before entering the distribution
system.  The average daily demand was 4,617 gal (vs. 7,500 gal provided by the park owner) and the peak
daily demand occurred on July 21, 2005, at 14,300 gal (compared to!6,000 gpd provided by the park
owner).

The flowrates through the CP-213f system varied due to the on-demand system configuration.
Withdrawn from the two pressure tanks located upstream of the system, the on-demand flowrates ranged
from 1 to 15 gpm and averaged 4.4 gpm, corresponding to a contact time of 27 to 410 min compared to a
design value of 20 min. At these flowrates, the hydraulic loading rates to the filter ranged from 0.3 to 4.1
gpm/ft2 compared to the design value  of 5.4 gpm/ft2. Note that Macrolite® filter media is rated for a
maximum hydraulic loading rate of 10 gpm/ft2.

At flowrates of 1 to 15 gpm, the inlet pressure to  the system ranged from 40 to 60 psi (compared to the
pressure tank set points from 45 to 60 psi) and the outlet pressure ranged from 22 to 55 psi. The total
pressure differential (AP) readings across the system ranged from 0 to 25 psi depending on the flowrates.
The AP across Tanks A and B ranged from 0 to 25 psi and across Tanks C and D from 2 to 16  psi based
on inlet and outlet pressure gauge readings.

During this time period, a total number of 431 backwash cycles took place. The throughput values
between two consecutive backwash cycles ranged from 1,714 to  6,857 gal depending on the settings of
the control disc located on top of each set of duplex units. The backwash frequency ranged from 0 to 5
tanks backwashed per day. There was one outlier on August 9, 2005, when over 1,720 gal of backwash
water was produced (equivalent to 13  backwash events in a single day). The vendor's contractor
determined that sediment was lodged in the purge/control valve on one of the duplex units, preventing the
valve from being closed; therefore, the duplex unit was stuck in the backwash mode before the operator
bypassed the system.

4.4.2       Backwash. The backwash was initiated by throughput. The control disc located on top of
each duplex unit determined the throughput before backwash. Table 4-5 summarizes the backwash
frequency based on four control disc sizes installed over the six-month study period. The vendor
switched out the control discs four times (although one was done in error)  due to observations  of
particulate arsenic, iron, and manganese breakthrough through the Macrolite® filters.  Control disc No. 5
geared to backwash after a throughput of 2,743 gal was used from system  startup on July 13, 2005,
through  September 20, 2005. The actual throughput values between two consecutive backwash cycles
averaged 2,449 gal based on the total volume of water treated and the total number of tanks
                                              23

-------
              Table 4-4.  System Operation from July 13, 2005 to January 17, 2006
Parameter
Values
Primary Well Pump (Well No. 2)
Total Operating Time (hr)
Average Daily Operating Time (hr)
Range of Flowrates (gpm)
Average Flowrate (gpm)
617.3(a)
3.4(a)
23-3 l(b)
25(b)
System Throughput/Demand
Throughput to Distribution (gal)
Average Daily Demand (gpd)
Peak Daily Demand (gpd)
863,470
4,617
14,300
CP-213f System - Service Mode
Range of Flowrates (gpm)
Average Flowrate (gpm)
Range of Contact Times (min)
Average Contact Time (min)
Range of Hydraulic Loading Rates to Filters (gpm/ft2)
Average Hydraulic Loading Rate to Filters (gpm/ft2)
Range of System Inlet Pressure (psi)
Range of System Outlet Pressure (psi)
Range of Ap Readings across System (psi)
l-15(c)
4.4(c)
27^10
114
0.3-4.1
2.2
40-60
22-55
0-25
CP-213 System - Backwash Mode
Number of Backwash Cycles
Throughput between Backwash Cycles (gal)
Number of Backwash Cycles Per Day
431(d)
l,714-6,857(e)
0-5
              (a)
    Hour meter installed on September 28, 2005. Run time before this period
    estimated based on wellhead totalizer readings and average well flowrate of
    25 gpm.
    Based on totalizer on raw water line and hour meter readings; excluding data
    from September 29, October 5, and October 6, 2005.
    Based on flow meter readings located on treated water line recorded starting
    September 28, 2005.
    Based on totalizer readings on backwash water discharge line and 130 gal of
    wastewater produced during backwash of each tank.
(e)  Backwash triggered by volume of water treated based on settings of control
    discs located on top of each set of duplex filtration tanks.
              (b)

              (c)
              (d)
backwashed. The number of tanks backwashed per day ranged from 0 to 5 except for the outlier on
August 9, 2005, discussed in Section 4.4.1. Because breakthrough of particulate arsenic, iron, and
manganese was observed, the vendor dispatched its contractor to the site to install a new control disc in an
attempt to curb the particulate breakthrough.  While a higher number disc should have been used, a disc
with a lower number (i.e., No. 2 geared to backwash after a throughput of 6,857 gal) was inadvertently
installed by the contractor between September 21 through 29, 2005.  On September 30, 2005, No. 2 disc
was replaced with a No. 7 disc, which was geared for a throughput of 1,957 gal. The average throughput
for the No. 7 disc was 1,932 gal and the number of tanks backwashed per day ranged from 0 to 5.  For this
reason, control disc No. 8 was subsequently installed on December 7, 2005 to further reduce the
throughput to 1,714 gal.  The actual throughput was 1,684 gal and the number of tanks backwashed per
day ranged from 1 to 5.
                                               24

-------
             Table 4-5.  Control Disc Size and Throughput between Backwash Cycles



Duration
07/13/05-09/20/05
09/21/05-09/29/05
09/30/05-12/06/05
12/07/05-01/17/06



Control
Disc
No. 5
No. 2
No. 7
No. 8
Design
Throughput
between
Consecutive
Backwash
Cycles
(gal)
2,743
6,857
1,957
1,714
Average
Throughput
between
Consecutive
Backwash
Cycles
(gal)
2,449
3,469
1,932
1,684

Number of
Tanks
Backwashed
(No./day)
0-5
0-3
0-5
1-5
Backwash
Water
Generation
Ratio
(%)
5.5
2.8
6.6
7.2
However, after the disc No. 8 changeout, particulate breakthrough continued to be observed. Except for
disc No. 2, the ratios of backwash water generated ranged from 5.5% to 7.2% and averaged 6.4%.

4.4.3       Residual Management. Residuals produced by the operation of the Macrolite® system
included backwash water and associated solids, which were discharged to a nearby septic system for
disposal.

4.4.4       System/Operation Reliability and Simplicity.  During the first six months of system
operation, several instances of total arsenic and iron breakthrough were observed in service mode and the
backwash frequency had to be increased twice by switching out the control valve on top of each set of
duplex units.  The required system O&M and operator skill levels are discussed according to pre- and
post-treatment requirements, levels of system automation, operator skill requirements, preventive
maintenance activities, and frequency of chemical/media handling and inventory requirements.

Pre- and Post-Treatment Requirements. Pre-treatment included KMnO4 addition for the oxidation of
arsenic and iron. Specific chemical handling requirements are further discussed below under chemical
handling and inventory requirements. KMnO4 was selected as an alternative oxidant due to the high TOC
levels in source water and the potential to form disinfection byproducts should chlorine be used as an
oxidant. However, as discussed in Section 4.5.1, it was determined that source water had a relatively
elevated KMnO4 demand, which resulted in some difficulty in controlling the effluent manganese levels
(both particulate and soluble forms) and ensuring that the MnO4" added was completely reduced to form
MnO2 solids.

System Automation. All major functions of the treatment system were automated and would require
only minimal operator oversight and intervention if all functions were operating as intended. Automated
processes included system startup in service mode when the well energized, backwash  cycling based on
throughput, fast rinse cycling, and system shutdown when the well pump shut down. However, as noted
in Section 4.4.1, an operational  issue did arise with the automated system backwash on August 9, 2005.
Due to the small size of the arsenic treatment system, the operational data was collected manually by the
operator mentioned in the next paragraph.

Operator Skill Requirements. Under normal operating conditions, the skill set required to operate the
Macrolite® system was limited to observation of the process equipment integrity and operating parameters
such as pressure and flow. The daily demand on the operator was about 5 min to visually inspect the
system and record operating parameters on the log sheets.  Other skills needed including performing
O&M activities such as replenishing the KMnO4 solution in the chemical drum, monitoring backwash
operational issues, and working with the vendor to troubleshoot and perform minor on-site repairs.
                                              25

-------
For the state of Minnesota, there are five water operator certificate class levels, i.e., A, B, C, D, and E,
with Class A being highest.  The certificate levels are based on education, experience, and system
characteristics, such as water source, treatment processes, water storage volume, number of wells, and
population affected. The operator for the BSLMHP system has a Class D certificate.  Class D requires a
high school diploma or equivalent with at least one year of experience in operating a Class  A, B, C, or D
system or a postsecondary degree from an accredited institution.

Preventive Maintenance Activities. Preventive maintenance tasks recommended by the vendor included
daily to monthly visual inspection of the piping, valves, tanks, flow meters, and other system components.
The pump on the primary well (Well No. 2) developed a leak and had to be shut down temporarily on
January 4, 2006 for repairs.  Meanwhile, Well No. 1 was turned on as the backup well. The leak on the
Well No. 2 pump was repaired the next day and the primary well resumed its normal operation thereafter..

Chemical/Media Handling and Inventory Requirements. KMnO4 addition was implemented since the
system startup on July 13, 2005. The mixing of the KMnO4 solution required only 10 min to complete, as
reported by the operator. The chemical consumption was checked each day as part of the routine
operational data collection.  Several adjustments were made over time to optimize the KMnO4 dosage for
the oxidation of arsenic, iron, and manganese.

4.5        System Performance

The performance of the Macrolite® CP-213f arsenic removal system was evaluated based on analyses of
water samples collected from the treatment plant, backwash lines, and distribution system.

4.5.1       Treatment Plant Sampling. Water samples were collected at five locations across the
treatment train: at the wellhead (IN), after the contact tanks (AC), after the first  set of duplex unit tanks A
and B (TA/TB), after the second set of duplex tanks C and D (TC/TD), and after the two sets of duplex
tanks combined (TT).  Sampling was conducted on 26 occasions (including two duplicate sampling
events) during the first six months of system operation, with field speciation performed on samples
collected from the IN, AC, and TT locations for 7 of the 26 occasions.  Table 4-6 summarizes the arsenic,
iron, and manganese analytical results.  Table 4-7 summarizes the results of the other water quality
parameters.  Appendix B contains a complete set of analytical results through the first six months of
system operation. The results of the water treatment plant samples with the addition of a varying amount
of KMnO4 before and after the November 7, 2005, manganese jar tests are discussed below.

Arsenic and Iron Removal. Total arsenic concentrations in raw water ranged from 20.6 to 36.6 |og/L
and averaged 27.7 |og/L.  As(III) was the predominant species with concentrations ranging from  13.9 to
27.4 |o,g/L and averaging 23.0 |o,g/L (Table 4-6 and Figure 4-8).  Only trace  amounts of particulate As and
As(V) existed in raw water, with concentrations averaging 2.1 and 4.0 |og/L, respectively.  The total
arsenic concentrations measured during this six-month period were consistent with those of the historical
source water sampling (Table 4-1), although the As(III) concentrations were significantly higher,
representing over 83% of the total concentrations in source water (compared to 54% during the August
31, 2004, source water sampling). The existence of As(III) as the predominating arsenic species was
consistent with the low DO concentrations (averaged 1.2 mg/L, Table 4-7) and low ORP values (ranged
from -76 to -23 mV and averaged -46 mV) in source water.

As shown in Table 4-6, total iron concentrations in raw water ranged from 1,069 to 3,758 |o,g/L and
averaged 2,760 |o,g/L. Iron in raw water existed almost entirely in the soluble  form with an  average value
of 2,691 |o,g/L. The presence of predominating soluble iron also was consistent with the presence of
                                              26

-------
             Table 4-6. Summary of Arsenic, Iron, and Manganese Analytical Results
Parameter
As (total)
As (soluble)
As (paniculate)
As(III)
As(V)
Fe (total)
Fe (soluble)
Mn (total)
Mn (soluble)
Sampling
Location
IN
AC
TA/TB
TC/TD
TT
IN
AC
TT
IN
AC
TT
IN
AC
TT
IN
AC
TT
IN
AC
TA/TB
TC/TD
TT
IN
AC
TT
IN
AC
TA/TB
TC/TD
TT
IN
AC
TT
Sample
Count
26
26
17
17
9
7
7
7(3)
7
7
?(a)
7
7
?(a)
7
7
?(a)
26
26
17
17
9
7
7
?(a)
26
26
17
17
9
7
7
7(3)
Concentration (jig/L)
Minimum
20.6
5.6
2.0
2.1
2.9
24.4
2.7
2.0
0.1
18.4
0.1
13.9
0.3
0.4
<0.1
1.7
1.3
1,069
247
12
<25
<25
2,263
<25
<25
110
416
203
187
331
110
108
138
Maximum
36.6
36.1
29.8
17.5
17.7
30.3
8.7
6.2
6.1
26.3
10.9
27.4
5.4
4.4
16.5
o o
J.J
2.4
3,758
3,173
2,363
1,140
1,067
2,954
306
40.7
430
1,506
1,002
971
1,091
145
946
1,062
Average
27.7
26.3
8.8
7.3
7.6
27.0
4.4
3.5
2.1
22.9
2.2
23.0
1.9
1.8
4.0
2.6
1.7
2,760
2,598
394
259
308
2,691
56.3
<25
144
955
648
650
644
132
492
565
Standard
Deviation
4.0
5.5
7.5
4.4
5.7
2.1
2.1
1.5
2.6
2.6
3.9
4.5
1.8
1.5
5.7
0.6
0.3
441
521
635
364
361
255
110
<25
59.7
235
254
257
256
12.2
317
363
         One-half of detection limit used for non-detect samples for calculations.
         Duplicate samples included calculations.
         (a) On December 28, 2005, arsenic speciation results taken at IN, AC, TA/TB, and TC/TD
             locations. Average concentration used for TT location.
predominating As(III) as well as low DO concentrations and low ORP values. Given the average soluble
iron and soluble arsenic levels in source water, this corresponded to an iron:arenic ratio of 100:1, which
was well above the target ratio of 20:1 for effective removal of arsenic (Sorg, 2002).

After KMnO4 addition and the contact tanks, As(III) concentrations ranged from 0.3 to 5.4 (ig/L and
averaged 1.9 (ig/L (Table 4-6 and Figure 4-8), suggesting effective oxidation of As(III) to As(V) with
KMnO4.  Particulate arsenic concentrations after the contact tanks ranged from 18.4 to 26.3 (ig/L and
averaged 22.9 (ig/L, representing most of the total arsenic (averaged 26.3 (ig/L) after the contact tanks.
After KMnO4 addition and the contact tanks, total iron concentrations averaged 2,598 (ig/L, existing
almost entirely in particulate form.  This data suggested effective oxidation of arsenic and iron even in the
                                               27

-------
Table 4-7. Summary of Other Water Quality Parameter Sampling Results
Parameter
Alkalinity
(as CaCO3)
Fluoride
Sulfate
Nitrate (as N)
Orthophosphate
(asP)
P (total)
(asP)
Silica
(as SiO2)
Turbidity
TOC
pH
Temperature
Sampling
Location
IN
AC
TA/TB
TC/TD
TT
IN
AC
TT
IN
AC
TT
IN
AC
TT
IN
AC
TA/TB
TC/TD
TT
IN
AC
TA/TB
TC/TD
TT
IN
AC
TA/TB
TC/TD
TT
IN
AC
TA/TB
TC/TD
TT
IN
AC
TT
IN
AC
TA/TB
TC/TD
TT
IN
AC
TA/TB
TC/TD
TT
Unit
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
NTU
NTU
NTU
NTU
NTU
mg/L
mg/L
mg/L
S.U.
S.U.
S.U.
S.U.
S.U.
°c
°c
°c
°c
°c
Sample
Count
26
26
17
17
9
7
7
?(a)
7
7
7(a)
7
7
?(a)
12
12
6
6
6
15
15
12
12
3
26
26
17
17
9
26
26
17
17
9
4
4
4
23
23
14
14
9
23
23
14
14
9
Concentration/Unit
Minimum
352
321
352
352
361
0.2
0.2
0.2
<1
<1
<1
O.05
O.05
O.05
O.05
O.05
O.05
O.05
O.05
0.4
0.1
O.03
O.03
0.1
22.6
22.5
22.5
22.7
21.7
13.0
2.8
<0.1
<0.1
0.1
3.2
3.1
3.0
7.2
7.2
7.2
7.2
7.2
9.3
9.4
9.3
9.2
9.5
Maximum
383
383
378
383
383
0.2
0.2
0.2
<1
<1
<1
0.06
0.06
0.25
O.05
<0.05
O.05
<0.05
O.05
0.6
0.5
0.4
0.2
0.2
29.4
28.6
28.4
28.2
24.7
35.0
6.5
14.0
14.0
11.0
4.8
4.6
4.8
7.5
7.6
7.4
7.5
7.7
14.9
14.1
12.5
12.8
13.8
Average
366
368
369
369
371
0.2
0.2
0.2
<1
<1
<1
0.03
0.03
0.06
O.05
O.05
O.05
O.05
O.05
0.5
0.4
0.1
0.1
0.1
24.3
24.4
24.4
24.6
23.4
30.0
4.0
1.5
2.0
1.9
4.0
3.7
3.9
7.3
7.4
7.3
7.4
7.3
10.7
10.9
10.2
10.3
11.6
Standard
Deviation
11
11
8
9
7
0
0
0
0
0
0
0
0
0.1
0
0
0
0
0
0.04
0.1
0.1
0.1
0.1
1.4
1.3
1.4
1.3
1.0
5.0
1.1
3.4
3.4
3.5
0.7
0.7
0.7
0.1
0.1
0.1
0.1
0.2
1.4
1.5
0.8
0.9
1.5
                                28

-------
      Table 4-7.  Summary of Other Water Quality Parameter Sampling Results (Continued)
Parameter
DO
ORP
Total Hardness
(as CaCO3)
Ca Hardness
(as CaCO3)
Mg Hardness
(as CaCO3)
Sampling
Location
IN
AC
TA/TB
TC/TD
TT
IN
AC
TA/TB
TC/TD
TT
IN
AC
TT
IN
AC
TT
IN
AC
TT
Unit
mg/L
mg/L
mg/L
mg/L
mg/L
mV
mV
mV
mV
mV
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
Sample
Count
23
23
14
14
9
23
23
14
14
9
7
7
?(a)
7
7
?(a)
7
7
7(3)
Concentration/Unit
Minimum
0.7
0.5
0.5
0.5
0.7
-76
-22
-9
-12
5.8
296
295
305
177
178
176
112
114
120
Maximum
3.6
3.1
1.6
1.3
1.2
-23
196
177
183
219
383
330
338
228
197
212
155
133
136
Average
1.2
1.1
0.9
0.9
0.9
-46
62
67
72
106
319
315
319
191
189
191
128
126
128
Standard
Deviation
0.7
0.5
0.3
0.2
0.2
14
56
51
55
78
28.8
10.9
12.0
16.6
6.8
11.5
14.0
6.4
5.3
      One-half of detection limit used for non-detect samples for calculations.
      Duplicate samples included calculations.
      (a) On December 8, 2005, arsenic speciation results taken at IN, AC, TA/TB, and TC/TD locations.
presence of elevated TOC levels at 3.2 to 4.8 mg/L (Table 4-7) in raw water.  Researchers have reported
that Fe(II)-KMnO4 reaction rates are more rapid than KMnO4-DOC interactions (Knocke et al., 1994).
Based on the presence of primarily particulate arsenic and iron after the contact tanks, it appears that the
elevated TOC levels did not have a significant effect on As(III) and Fe(II) oxidation. Note that KMnO4
dosages used during the six-month study period ranged from 2.1 to 6.1 mg/L (as KMnO4) and averaged
4.8 mg/L.  The effects of KMnO4 dosage on Mn(II) oxidation are discussed in the next subsection.

From July 13, 2005, to January 17, 2006, total arsenic concentrations in the treated water ranged from 2.0
to 29.8 ug/L and averaged 7.9 ug/L (Table 4-6). Soluble arsenic concentrations in the treated water
ranged from 2.0 to 6.2 ug/L and averaged 3.5 ug/L. Out of the 26 sampling occasions, total arsenic
concentrations in the treated water exceeded the 10-ug/L MCL for a total of 8 times due to particulate
breakthrough from the Macrolite® filters (Figure 4-9). As shown in Figure 4-10, the elevated total arsenic
concentrations were accompanied by elevated total iron concentrations. The iron concentrations in the
treated water ranged from <25 to 2,363 ug/L and averaged 323 ug/L, with most existing as particulate
iron (Table 4-6). The soluble iron levels were below the method detection limit of <25  ug/L as measured
in water samples filtered with 0.45-um disc filters. On September 7, 2005, the total arsenic concentration
in the treated water exceeded 10 ug/L due to low KMnO4 dosage, which can be seen by the negative ORP
readings across the treatment train, resulting in incomplete oxidation of As(III) and Fe(II).  A study has
shown that Fe(II) complexed with dissolved organic matter (DOM) was very difficult to remove via
oxidation and subsequent precipitation of Fe(OH)3(s). This was due to the formation of colloidal iron that
had a size fraction small enough to pass through 0.2-um disc filters.  However, this phenomenon would
be affected by the concentration and nature of the DOM in water (Knocke et al., 1994).  The formation of
colloidal iron did not appear to be an issue at BSLMHP with primarily particulate iron present after the
                                              29

-------
                                       Arsenic Species at the Inlet (IN)
                                    Arsenic Species After Contact Tanks (AC)
                                    Asenic Speciation After Tanks Combined (TT)
Figure 4-8.  Concentrations of Arsenic Species at IN, AC, and TT Sampling Locations
                                               30

-------
        07/13/05 07/27/05 08/10/05 08/24/05 09/07/05 09/21/05 10/05/05 10/19/05 11/02/05 11/16/05 11/30/05 12/14/05 12/28/05 01/11/06
Figure 4-9.  Total Arsenic Concentrations at TA/TB, TC/TD, and TT Sampling Locations
         07/13/05 07/27/05 08/10/05 08/24/05 09/07/05 09/21/05 10/05/05 10/19/05 11/02/05 11/16/05 11/30/05 12/14/05 12/28/05 01/11/06
                                                   Date


 Figure 4-10. Total Iron Concentrations at TA/TB, TC/TD, and TT Sampling Locations
                                                 31

-------
contact tanks and after the Macrolite® filters (e.g. a size fraction large enough to be retained by a 0.45 jam
disc filter). The increase in particulate iron also corresponded with an increase in particulate arsenic,
indicating iron breakthrough from the Macrolite® filters.

In order to better control particulate breakthrough from the filtration tanks, the control discs located on
top of the two duplex units were replaced twice from Discs No. 5 to No. 7 and, then, from Discs No. 7 to
No. 8 during the six-month duration to allow for more frequent backwash. (Note that Disc No. 2 was
erroneously installed for a short duration before the mistake was caught and corrected). Table 4-8 lists
the disc number, operating duration, total arsenic concentrations exceeding 10 (ig/L, and total iron
concentrations with arsenic exceeding  10 (ig/L. The use of Discs No. 5 and No. 7 resulted in three and
four occurrences, respectively, with arsenic concentrations measured as high as 29.8 (ig/L and iron
concentrations measured as high as 2,363 (ig/L.  Disc No. 8 was installed on December 7, 2005, and the
treated water samples collected during December 7, 2005, through January 17, 2006, contained an
average of 4.0 and 194 (ig/L of total arsenic and iron, respectively, which were the lowest for the six-
month period. However, due to particulate arsenic and iron breakthrough observed on December 14,
2005, control disc No. 8 will be switched out to allow even more frequent backwash during the next six-
month period.
              Table 4-8.  Control Disc Sizes and Corresponding Occurrences with
                         High Total Arsenic and Iron Concentrations
Duration
07/13/05-09/20/05
09/21/05-09/29/05
09/30/05-12/06/05
12/07/05-01/17/06
Control
Disc
No. 5
No. 2(a)
No. 7
No. 8
Occurrence
1
2
3
N/A
4
5
6
7
8
Total Arsenic
Concentration
Exceeding 10 jig/L
TA/TB
N/A
13.8
21.5
TC/TD
N/A
12.3
17.5
TT
17.7
N/A
N/A
N/A
N/A
11.3
N/A
29.8
12.1
11.3(b)
10.1
N/A
N/A
N/A
12.6
12.4(b)
N/A
N/A
17.1
N/A
None
Total Iron Concentration
with Arsenic
Exceeding 10 jig/L
TA/TB
N/A
571
1,052
TC/TD
N/A
465
1,140
TT
482
N/A
N/A
N/A
N/A
336
N/A
2,363
983
978 (b)
547
N/A
N/A
N/A
1,001
l,023(b)
N/A
N/A
1,067
N/A
None
    (a) Incorrect disc inadvertently installed and replaced soon after installation.
    (b) Field duplicate.
    N/A = data not available
Manganese.  As shown in Table 4-6, total manganese concentrations in raw water ranged from 1 10 to
430 |og/L and averaged 144 |o,g/L, which existed primarily in the soluble form at levels ranging from 110
to 145 |og/L and averaging 132 |o,g/L. The manganese levels in raw water exceeded its secondary MCL of
50
Figure 4-11 shows the concentrations of total and soluble manganese after KMnO4 addition and after the
contact tanks over time.  Before November 15, 2005, total manganese levels after the contact tanks
ranged from 416 to 1,126 |o,g/L, 38 to 94% of which was present in the soluble form based on the use of
0.45-(im disc filters.  As noted in the figure and Table 4-9, the KMnO4 dosage was incrementally
decreased from the initial level of 3.8 mg/L to 1.4 mg/L, and then increased to 2.6 mg/L by adjusting the
                                               32

-------
        1,800
        1,600 -<•
                                                                                       0.0
          07/13/05 07/27/05 08/10/05 08/24/05 09/07/05 09/21/05 10/05/05 10/19/05 11/02/05 11/16/05 11/30/05 12/14/05 12/28/05 01/11/06
                                               Date

       Figure 4-11. Total and Soluble Manganese Concentrations at AC Sampling Location
            Table 4-9. Correlations Between Pump Stroke Length, KMnO4 Dosage, and
                          Total and Soluble Manganese Concentrations





Duration
07/13/05 to 08/07/05
08/08/05 to 08/13/05
08/14/05 to 08/30/05
08/3 1/05 to 09/07/05
09/08/05 to 11/15/05
11/16/05 to 11/20/05
11/2 1/05 to 12/04/05
12/05/05 to 01/17/06



Stroke
Length
(%)
33
30
26
15
26
40
38
40


Average
KMnO4
Dosage
(mg/L)
3.8
3.4
3.0
1.4
2.6
5.4
3.1
5.6

Total Mn
at
AC
Location
(mg/L)
634-1,126

Soluble Mn
at
AC
Location
(mg/L)
337
N/A
871-1,097
416-581
676-1,042
850
N/A
468-946
N/A
1,123
1,031-1,506
N/A
108-166
Total Mn
at TA/TB,
TC/TD,
andTT
Location
(mg/L)
428-727
Soluble Mn
at TA/TB,
TC/TD,
andTT
Location
(mg/L)
391
N/A
467-1,010
430-906
548-1,091
1,000
N/A
535-1,062
N/A
432
201-673
N/A
138-202
     N/A = Data not available
stroke length of the paced-pump from 33 to 15%, then to 26%. The KMnO4 dosage was decreased from
the initial level of 3.8 mg/L because elevated total and soluble manganese levels at 996 (average) and
377 (ig/L, respectively, were measured after KMnO4 addition and thought, at the time, to have been
caused by overdosing of KMnO4.  Decreasing the KMnO4 dosage from 3.8 to 3.4 and then to 3.0 mg/L
did not appear to help reduce the manganese concentrations, with total and soluble levels measured, for
example, at 1,097 and 850 (ig/L, respectively, on August 18, 2005.  A further decrease in KMnO4dosage
                                               33

-------
to 1.4 mg/L helped reduce the total manganese levels, which, however, were still higher than those in raw
water at 581 and 416 (ig/L, respectively, on August 31 and September 7, 2005. This low level of KMnO4
addition also caused significantly elevated arsenic and iron concentrations in the treated water due to
incomplete oxidation of As(III) and Fe(II) also discussed in Section 4.5.1. Resuming the KMnO4 dosage
at 2.6 mg/L returned the total manganese concentrations to 676 to 1,042 (ig/L, with most (i.e., 468 to
946 (ig/L) existing in the soluble form, as determined by the use of 0.45-(im disc filters.

The addition of 1.4 mg/L to 3.8 mg/L of KMnO4 during July 13 through November 15, 2005, resulted in
significantly elevated manganese levels not only after the contact tanks, as discussed above, but also after
the Macrolite® filters (ranging from 428 to 1,091 |o,g/L and averaging 722 |o,g/L, Figure 4-12). Further,
manganese in the treated water existed almost entirely (i.e., 535 to 1,062 (ig/L) in the soluble form based
on the use of 0.45-(im filter discs for obtaining the soluble fractions.
            1,800
            1,600
             07/13/05 07/27/05 08/10/05 08/24/05 09/07/05 09/21/05 10/05/05 10/19/05 11/02/05 11/16/05 11/30/05 12/14/05 12/28/05 01/11/06
   Figure 4-12.  Total Manganese Concentrations at TA/TB, TC/TD, and TT Sampling Locations
Mn(II) oxidation by KMnO4 is dependent on the KMnO4 dosage, pH, temperature, and DOM in raw
water. The reaction of KMnO4 with Mn(II) is typically rapid and complete at pH values ranging from 5.5
to 9.0. However, elevated DOM levels can increase the KMnO4 demand due to competition between
these species and resulting kinetic effects (Knocke et al, 1987). Some researchers suggest that DOM can
interfere with the formation of MnO2(s) solids by exerting KMnO4 demand and, possibly, forming
complexes with fractions of Mn(II), thus rendering it less likely to be oxidized (Gregory and Carson,
2003). When modeling the Mn(II) oxidation with KMnO4, Carlson and co-workers (1999) determined
that incorporating a term in the model to account for the DOM demand for MnO4" significantly improved
the prediction of the MnO4" consumption. The incorporation of DOM into the oxidation term to account
for complexation between DOM and Mn(II) also was postulated but no data was collected as part of that
study. Further, high levels of DOM in source  water also can form fine colloidal MnO2 particles, which
may not be filterable by conventional gravity or pressure filters. Knocke et al. (1991) defined colloidal
particles as those passing through 0.20-|om filters and requiring ultrafiltration for removal.
                                                34

-------
The presence of significantly elevated soluble manganese levels after the contact tanks and after the
Macrolite® filters, even with the use of insufficient KMnO4, prompted the speculation that the soluble
manganese measured might, in fact, be colloidal particles that had passed through the 0.45-(im disc filters.
Therefore, jar tests were performed on November 7, 2005, to determine if higher KMnO4 dosages might
help overcome the DOM effect and form larger filterable MnO2 solids in the treated water. Prior to the
start of the jar tests, the additional KMnO4 demand of a Macrolite®-treated water sample (to which 3.0
mg/L of KMnO4 had been added based on the KMnO4 consumption in the chemical day tank during the
week of sampling) was pre-determined by titrating 1 L of the water with a 1-g/L KMnO4titrant. After 2.5
mL of the titrant was added, the water being titrated developed  a dark yellow color, and was filtered, after
about 10 min, with 0.20-|o,m disc filters to remove any suspended solids including MnO2.  The filtrate was
observed to have a pink color, indicating the presence of KMnO4 residual.

Five KMnO4 dosages ranging from 1.0 to 3.0 mg/L were then selected for the jar tests using the same
Macrolite®-treated water sample mentioned above. (These dosages would be in addition to the KMnO4
already added to the water to be treated). After 31 min of mixing time (including 1 min at 200 rpm, 19 at
100 rpm, and 11 min at 28 rpm), the water in the jars was filtered separately with 0.20-(im disc filters and
analyzed for soluble arsenic, iron, and manganese. Table 4-10 summarizes  the results of the jar tests.
                    Table 4-10. Jar Test Results for Macrolite®-Treated Water
Parameter
Potassium Permanganate Added (mg/L)(a)
Mixing Time (min)
Initial pH(b)
Final pH(c) @16.8°C
Initial ORP(b)@16.8°C
Final ORP(c)
Residual KMnO4 (mg/L)(d)
As(soluble)(e)(ng/L)
Fe(soluble)(e)(ng/L)
Mn(soluble)(e)(ng/L)
1
0
31
7.70
7.68
283
353
0.04
5.5
<25
1,090
2
1.0
31
7.80
7.67
292
360
0.01
4.5
<25
102
o
5
1.5
31
7.81
7.70
400
363
0.05
3.3
<25
0.8
4
2.0
31
7.71
7.62
440
369
0.07
3.3
<25
11.0
5
2.5
31
7.74
7.60
509
493
0.35
3.2
<25
399
6
3.0
31
7.76
7.61
521
515
0.63
3.1
<25
469
        (a) Dosage on top of 3.0 mg/L already added to the water prior to jar tests.
        (b) Taken approximately 15 min into jar test.
        (c) Taken at end of 31 min jar test.
        (d) CAIROX® Method 103 (DPD Spectrophotometry) for determination of KMnO4 residual.
        (e) Filtered with 0.20-nm filters.
During mixing, jars No. 2 to 4 formed large brown floes in a pale to dark yellow solution (Figure 4-13).
Jars No. 5 to 6 had smaller brown floes in a dark copper solution. As shown in Table 4-10, the soluble
iron levels in all jars were below the method detection limit of 25 |og/L, suggesting that effective
oxidation and removal of iron had already been achieved prior to the jar tests. Soluble arsenic levels
decreased slightly from 5.5  |o,g/L to 3. 1 |o,g/L in jar No. 6 (the one with the highest KMnO4 dosage 3.0
mg/L). Only soluble manganese concentrations varied significantly, decreasing from 1,090 (ig/L in jar
No . 1 to < 1 (ig/L in j ar No . 3 and then increasing to 469 (ig/L in j ar No . 6 . Knocke et al . (1990) reported
that the kinetics for Fe(II) oxidation are faster than for Mn(II) oxidation when KMnO4 is used as the
oxidant. The relevant stoichiometric equations are shown as follows:

                   3Fe2+ + KMnO4 + 7H2O -> 3Fe(OH)3(s) + MnO2(s) + K+
3Mn2+ + 2KMnO4 + 2H2O
                                                              2K+
                                               35

-------
In the control sample, the soluble manganese level was high due to the slower Mn(II) oxidation kinetics
and the presence of DOM as discussed above. The 1,090 (ig/L of "soluble" manganese in the control
sample confirmed that the manganese most likely was present as colloidal particles since the sample
analyzed had already been filtered with 0.2 (im disc filters.  Increasing the KMnO4 dosage to 1.5 mg/L (on
top of the 3.0 mg/L already added to the water prior to the jar tests) appeared to be sufficient to overcome
the effects of DOM, allowing filterable manganese particles to form.  As a result, only 0.8 (ig/L of
manganese that passed through the 0.2-(im filters was reported as "soluble" manganese.  Further,
increasing the KMnO4 dosage up to 3  mg/L increased the soluble manganese level up to 469 (ig/L,
suggesting that excess KMnO4 was present in the treated water. The presence of KMnO4 was supported
by the elevated residual KMnO4 levels and the elevated ORP readings (see results of jars No. 4 and 5).
                                  Figure 4-13. Jar Test Setup
Based on the jar tests results, it was determined that an additional 1.5 mg/L of KMnO4 was needed to
attain filterable manganese solids.  Therefore, the KMnO4 dosage to the treatment system was increased
on November 15, 2005 for a target dosage of 4.5 mg/L.  The actual dosage after adjusting the stroke
length from 26 to 40% was 5.4 mg/L (Table 4-9). After the increase in dosage, manganese was present
primarily in the particulate form, with concentrations ranging from 1,031 to 1,506 (ig/L. The soluble
manganese was decreased significantly to 108 to 166 (ig/L (Figure 4-11).  (Note that as before, 0.45-(im
filters were used to obtain these treatment results). After November 15, 2005, the speciation results
indicated that approximately 7 to 16% was present as soluble manganese in the Macrolite® treated water:
the total manganese concentrations ranged from 201 to 673 (ig/L and the soluble manganese
concentrations ranged from 138 to 202  (ig/L. Based on an average soluble manganese concentration of
177 (ig/L and total manganese concentration of 673 (ig/L, particulate manganese breakthrough of up to
496 (ig/L was experienced from the Macrolite® filters.  In the next six-month period, further fine-tuning
will be made to the KMnO4 dosing to determine if soluble manganese may be further reduced to less than
the Secondary Maximum Contaminant Level (SMCL) of 50 (ig/L.
                                               36

-------
TOC. TOC levels in raw water were elevated, ranging from 3.2 to 4.8 mg/L. KMnO4 was used as the
oxidant to prevent the formation of disinfection byproducts.  Before November 15, 2005, the effluent
TOC levels ranged from 3.8 to 4.8 mg/L and there was little or no TOC removal across the treatment
train.  After November 15, 2005, the influent TOC level was 3.2 mg/L (average) and the effluent TOC
level was 3.0 mg/L (average) with approximately 6% removal.  Research has shown that only minimal
organic carbon removal occurs (at less than 10%) via KMnO4 oxidation in source water containing Mn(II)
and DOC (Salbu and Steinnes, 1995; Knocke et al., 1990). However, significant DOC removal with
colloidal iron particles produced by Fe(II) oxidation was observed (Knocke et al.,  1994). The
complexation of iron with organic carbon does not appear to be a significant factor at the BSLMHP site as
discussed previously.

Other Water Quality Parameters.  DO levels remained low across the treatment train (with average
values ranging from 1.2 to 0.9 mg/L), but ORP values increased across the treatment train (ranging from -
76 to -23 mV before versus 1 to 196 mV after KMnO4 addition). There were two  outliers on September 7
and October 26, 2005, where the ORP values after the contact tanks were negative. The ORP on
September 7, 2005, was negative because the stroke on the KMnO4 pump was turned down to 15% on
August 30,  2005. The pH in raw water had an average value of 7.3 and the pH in the treated water had an
average value of 7.3. Average alkalinity results ranged from 366 to 369 mg/L (as  CaCO3) across the
treatment train.  Average total hardness  results ranged from 315 to 319 mg/L (as CaCO3) across the
treatment train (the total hardness is the  sum of calcium hardness and magnesium hardness). The water
had an almost even split of calcium and  magnesium hardness.  Fluoride concentrations were 0.2 mg/L in
raw water and after contact tanks and were not affected by the Macrolite® filtration. The average nitrate
concentration was <0.05 mg/L (as N) across the treatment train. There was no detection of sulfate and the
silica concentrations remained at approximately 24 mg/L (as SiO2) across the treatment train.

Orthophosphate was analyzed between July 13, 2005, and October 5, 2005, and there was no detection.
However, total phosphorous analyzed between October 12, 2005 and January 17, 2006, showed an
elevated average of 0.5 mg/L (as P) in raw water and 0.1 mg/L (as P) in the treated water (Figure 4-14).
This indicates a removal rate of approximately 80% most likely through adsorption onto iron solids. The
elevated total phosphorous levels were further confirmed by analyzing a raw water sample taken on
December 14, 2005, for the various phosphorous species according to EPA Method 365.3 by Sierra
Environmental Monitoring, Inc. It was determined that the total phosphorous level in raw water was at
0.58 mg/L (as P), which was present primarily as total hydrolyzable phosphorous at 0.51 mg/L (as P).
According to the EPA Method 365.3, total hydrolyzable phosphorous includes both polyphosphorous and
some organic phosphorous. It also was later confirmed that no organopesticides were present in source
water by EPA Method 507. There were other potential sources for elevated phosphorous in groundwater.
Based on research conducted by the Sauk River Watershed District, the Sauk River and Big Sauk Lake
have sediment, phosphorous, and nitrates caused by non-point source discharges from septic systems,
agriculture, and urban runoff (Post, 2005). The historical monitoring data for the surface water of Big
Sauk Lake shows a maximum total phosphorous level of 0.4 mg/L (as P) (Big Sauk Lake River
Watershed District, 2006) and the Big Sauk Lake is located approximately 1000 ft from the BSLMHP
wellhouse.

4.5.2       Backwash Water Sampling.  Table 4-11 summarizes the analytical results from the six
backwash water sampling events.  For the first three sampling events, only pH, turbidity, TDS, and
soluble As, Fe, and Mn were analyzed for the grab samples collected at the backwash water discharge
line. Soluble arsenic concentrations in the backwash water ranged from 3.5 to 8.5 (ig/L; soluble iron
concentrations ranged from <25 to 63 (ig/L; and soluble manganese concentrations ranged from 560 to
736 (ig/L based on the use of 0.45-(im filters. Starting from November 15, 2005, TSS and total As, Fe,
and Mn also were analyzed for the composite sample  collected using the modified backwash procedure
discussed in Section 3.3.4.  After the modified backwash procedure was implemented, total arsenic
                                              37

-------
             10/05/05 10/12/05 10/19/05 10/26/05 11/02/05 11/09/05 11/16/05 11/23/05 11/30/05 12/07/05 12/14/05 12/21/05 12/28/05 01/04/06 01/11/06
                Figure 4-14. Total Phosphorous Concentrations at IN, AC, TA/TB,
                               TC/TD and TT Sampling Locations
concentrations in the backwash water ranged from 114 to 417 (ig/L; total iron concentrations ranged from
14,069 to 77,641 (ig/L; and total manganese concentrations ranged from 1,595 to 16,178 (ig/L. Note that
November 15, 2005, BW2 data had uncharacteristically high total and soluble As and Fe, and, therefore,
were excluded from all calculations. TSS levels in the backwash water ranged from 102 to 210 mg/L and
averaging 154 mg/L (excluding November 15, 2005 BW2 data that had uncharacteristically high As and
Fe and the January 10, 2006, BW2 data that had uncharacteristically low TSS).  Using 130 gal of
backwash water produced, this equates to approximately 0.17 Ib of solids generated per backwash event
including 4.4 x 10"4 Ib of arsenic, 0.08 Ib of iron, and 0.01 Ib manganese.

4.5.3       Distribution System Water Sampling. The results of the distribution system sampling are
summarized in Table 4-12.  The main differences observed before and after the operation of the system
were decreases in arsenic, iron, and manganese concentrations at each of the three sampling locations.
Arsenic concentrations in the baseline samples ranged from  15.3 to 26.3 |o,g/L. Since the treatment
system started operation, arsenic levels in the distribution system samples ranged from 3.6 to 14.2 |o,g/L
with an average of 6.6 |og/L. Arsenic concentrations mirrored the treatment results after the Macrolite®
filters, except for an outlier at 24.1 |o,g/L on January 17, 2006, when the homeowner did not sufficiently
flush the tap the night before sampling. Total arsenic concentrations exceeded 10 (ig/L at all three
sampling locations on September 7, 2005, due to particulate  arsenic and iron breakthrough from the
Macrolite® filters described in Section 4.4.2.  Iron concentrations in the baseline samples were high,
ranging from 2.1 to 5.0 mg/L.  Since system startup, iron levels in the distributed water decreased
significantly to an average value of 128 |o,g/L (not including  the outlier at DS1 on January 17, 2006).
Particulate breakthrough was observed on September 7, November 29, and December 15, 2005 with
elevated iron concentrations ranging from 532 to 2,363 (ig/L after the Macrolite® filters.  Iron
concentrations in the distribution system during those days ranged from <25 to 279 (ig/L, indicating
                                                38

-------
                                                Table 4-11. Backwash Water Sampling Results
Sampling Event
No.
1
2
3
4
5
6
Date
09/08/05
09/20/05
10/12/05(a)
11/15/05*'0
12/08/05
01/10/06
BW1
TankA/B
KMnO4 Dosage
mg/L
2.6
2.6
2.6
2.6
5.6
5.6
Control Disc
No.
5
5
7
7
8
8
B.
S.U.
7.2
7.3
7.3
7.5
7.4
7.4
Turbidity
NTU
170
160
120
NS
NS
NS
CO
O
X
VI
H
mg/L
576
550
356
54
224
360
NS
NS
NS
102
210
130
Total As
Soluble As
Particulate As
Total Fe
Soluble Fe
Total Mn
Soluble Mn
Hg/L
NS
NS
NS
329
417
363
3.9
3.6
4.4
6.9
0.5
3.3
NS
NS
NS
322
416
360
NS
NS
NS
63,108
77,641
43,384
<25
<25
<25
163
201
128
NS
NS
NS
1,595
16,178
12,265
624
624
685
836
350
341
BW2
Tank C/D
X
e.
S.U.
7.3
7.3
7.3
7.5
7.6
7.6
Turbidity
NTU
120
17
410
NS
NS
NS
If!
O
CO
VI
H
mg/L
544
368
350
346
334
326
NS
NS
NS
348
175
16
Total As
Soluble As
Particulate As
Total Fe
Soluble Fe
Total Mn
1
.—
j=
9
1
Hg/L
NS
NS
NS
1,325
397
114
3.5
8.5
4.3
206
2.9
5.3
NS
NS
NS
1,119
394
109
NS
NS
NS
214,211
75,485
14,069
<25
<25
63
29,992
39
304
NS
NS
NS
3,835
14,159
4,016
560
736
656
1,175
348
376
TDS = total dissolved solids; TSS = total suspended solids; NS = not sampled.
(a)  Manual backwash performed after Tank A/B had just been backwashed; less particles visually observed
(b)  Samples taken on November 15, 2005 re-analyzed with similar results for both samples on this date.
(c)  Modified backwash procedures implemented starting November 15, 2005.

-------
                                                      Table 4-12. Distribution Sampling Results
Sampling Events
No.
BL1
BL2
BL3
BL4
1
2
3
4
5
6
7
Date
02/16/05
03/23/05
04/19/05
05/23/05
07/26/05
09/07/05
09/27/05
1 1/02/05
11/29/05
12/15/05
01/17/06
As at Entry Point
Hg/L
NA
NA
NA
NA
5.5
21.51
17.5
8.4/
7.6
5.61
4.2
9.2
11.7/
12.5
2.8/
2.5
DS1
Stagnation Time
hr
7.0
6.0
6.2
5.8
7.3
8.5
8.3
12.5
8.0
11.3
9.0
S3
S.U.
7.2
7.3
7.0
7.3
7.2
7.4
7.3
7.6
7.4
7.5
7.5
Alkalinity
mg/L
382
362
377
384
365
356
370
361
365
374
383
1/3
£
'
—
U
Hg/L
24.3
21.9
25.3
25.7
5.1
14.2
4.3
6.8
4.1
4.1
24.1
2,649
2,175
2,878
2,578
73
52
72
<25
266
57
1,999
128
130
141
124
722
438
687
976
367
400
923
0.6
0.4
2.4
0.5
0.5
0.3
2.1
0.2
0.9
1.2
1.0
4.1
2.2
3.9
0.7
0.4
0.2
11.0
8.8
6.2
3.9
21.8
DS2
Stagnation Time
hr
8.3
8.3
10.0
7.3
9.3
9.0
7.3
7.0
6.0
8.0
8.5
S3
s.u.
7.4
7.4
7.2
7.3
7.3
7.5
7.4
7.6
7.5
7.6
7.5
Alkalinity
mg/L
374
367
395
370
374
352
361
352
365
374
383
*
QJ
'
—
5
Hg/L
19.8
26.2
15.3
24.2
5.4
12.7
5.1
7.9
3.6
5.7
4.9
2,792
4,986
2,137
2,639
84
<25
127
142
57
184
187
129
147
127
123
617
516
717
950
369
443
267
0.6
0.3
1.6
<0.1
0.4
0.1
0.2
0.1
0.1
0.8
0.2
0.2
2.5
3.4
0.4
0.2
1.7
0.1
0.2
0.2
0.2
0.7
DS3
Stagnation Time
hr
NS
7.3
8.4
8.8
9.3
8.0
9.5
9.3
9.3
9.0
7.5
S3
S.U.
NS
7.5
7.4
7.3
7.3
7.6
7.4
7.6
7.5
7.5
7.6
Alkalinity
mg/L
NS
376
386
379
370
365
374
365
361
374
383
5
£
1
—
5
Hg/L
NS
26.3
24.6
22.6
6.3
13.9
4.2
8.5
3.7
6.3
4.9
NS
2,590
2,751
2,649
162
84
98
37
222
279
342
NS
128
133
119
612
525
659
935
478
468
226
NS
<0.1
0.2
0.1
0.4
0.1
1.1
0.2
1.1
1.0
4.7
NS
1.9
0.4
0.9
0.6
1.4
1.0
0.3
2.4
0.7
3.2
NS = not sampled; NA = not analyzed/applicable.
                                                                         40

-------
settling of iron solids within the distribution system piping.  On January 17, 2006, due to insufficient
flushing of the sampling tap, the iron concentration at DS1 was 1,999 (ig/L while iron concentrations after
the Macrolite® filters were very close to the detection limit of 25 (ig/L. Manganese levels in the
distribution system baseline samples averaged 130 |o,g/L and increased to an average of 569 |o,g/L after the
treatment system became operational. The manganese concentrations in the distribution system mirrored
the results after the Macrolite® filters.

There was no major change in pH values in the distribution  system, which ranged from 7.0 to 7.5 before
system startup and 7.2 to 7.6 after startup. Alkalinity levels in the distribution system ranged from 362 to
395 mg/L (as CaCO3) before and 352 to 383 (as CaCO3) after.

Lead and copper levels in the distribution system did not appear to have been affected by the operation of
the arsenic treatment system. Lead levels in the distribution system ranged from <0.1 to 4.7 |o,g/L with no
samples exceeding the action level of 15 |o,g/L. Copper concentrations ranged from <0.1 to 21.8 |o,g/L
with no samples exceeding the 1,300 |o,g/L action level.

4.6        System Cost

The cost of the system was evaluated based on the capital cost per gpm (or gpd) of design capacity and
the O&M cost per 1,000 gal of water treated. This required the tracking of the capital cost for equipment,
engineering, and installation cost and the O&M cost for chemical supply, electrical power use, and labor.
However, the cost associated with improvements to the building and any other discharge-related
infrastructure were not included in the treatment system cost. While not included in the scope of the
demonstration project, these activities were funded by the demonstration host site.

4.6.1      Capital Cost. The capital investment was $63,547, which included $22,422 for equipment,
$20,227 for site engineering, and $20,898 for installation. Table 4-13 presents the breakdown of the
capital cost as provided by the vendor in its proposal to Battelle dated February  17, 2005.  The equipment
cost was about 35% of the total capital investment, which included the CP-213f filtration tanks,
Macrolite® media, contact tanks, process valves and piping, instrumentation and controls, a chemical feed
system (including a storage tank with a secondary containment), additional sample taps and
totalizer/meters, shipping, and equipment assembly labor.

The engineering cost included the cost for preparing a process design report and required engineering
plans, including a general arrangement drawing, piping and  instrumentation diagrams (P&IDs),
interconnecting piping layouts, tank fill details, an electrical on-line diagram, and other associated
drawings. After certification by a Minnesota-registered professional engineer (PE), the plans were
submitted to the MDH for permit review and approval (Section 4.3.1).  The engineering cost was
$20,227, which was 32% of the total capital investment.

The installation cost included the cost for labor and materials for system unloading and anchoring,
plumbing, and mechanical and electrical connections (Section 4.3.3). The installation cost was $20,898
or 33% of the total capital investment.

Using the system's rated capacity of 20 gpm (or 28,800 gpd), the capital cost was normalized to be
$3,177/gpm (or $2.21/gpd). The  capital cost of $63,547 was converted to an annualized cost of
$5,998/year using a capital recovery factor of 0.09439 based on a 7% interest rate and a 20-year return.
Assuming that the system was operated 24 hours a day, 7 days a week at the design flow rate of 20 gpm
to produce 10,500,000 gal of water per year, the unit capital cost would be $0.57/1,000 gal. However,
since the system operated an average of 3.4 hr/day at just under 4.4 gpm (see Table 4-4), producing
863,470 gal of water during the six-month period, the unit capital cost was increased to $3.47/1000 gal at
this reduced rate of production.


                                               41

-------
          Table 4-13.  Summary of Capital Investment for BSLMHP Treatment System
Description
Quantity
Cost
% of Capital
Investment Cost
Equipment Cost
Media and Tanks
Process Valves and Piping
Chemical Feed
Chemical Storage and Secondary
Containment
Instrumentation and Controls
Additional Flowmeter/Totalizers
Shipping
Labor
Equipment Total
1
1
1
1
1
1
—
—
—
$8,549
$1,935
$1,150
$680
$1,079
$359
$750
$7,920
$22,422
-
-
-
-
—
—
—
—
35%
Engineering Cost
Labor
Travel
Subcontractor
Engineering Total
—
—
—
—
$15,620
$1,750
$2,857
$20,227
—

—
32%
Installation Cost
Labor
Travel
Subcontractor
Installation Total
Total Capital Investment
—
—
—
—
-
$5,000
$2,913
$12,985
$20,898
$63,547
—
—
—
33%
100%
4.6.2       Operation and Maintenance Cost. The O&M cost primarily included cost associated with
chemical supply, electricity consumption, and labor (Table 4-14). The usage rate for the KMnO4 stock
solution was approximately 7.5 gal or 100 Ib/yr. Incremental electrical power consumption was
calculated for the chemical feed pump. The power demand was calculated based on the total operational
hours throughout the duration of the six-month  study, the chemical feed pump horsepower, and the unit
cost from the utility bills. The routine, non-demonstration related labor activities consumed about 5 min
per day, 5 days a week, as noted in Section 4.4.4. Based on this time commitment and a labor rate of
$21/hr, the labor cost was $0.27/1,000 gal of water treated. In sum, the total O&M cost was
approximately $0.43/1,000 gal. The O&M cost will be verified during the next reporting period.
                                             42

-------
Table 4-14. O&M Cost for BSLMHP, MN Treatment System
Cost Category
Projected Volume Processed (gal)
Value
863,470
Assumption
From 07/13/05 through 01/17/06 (see Table 4-4)
Chemical Usage
Chemical Unit Price ($/lb)
Total Chemical Consumption (Ib)
Chemical Usage (lb/1,000 gal)
Total Chemical Cost ($)
Unit Chemical Cost ($71,000 gal)
$2.07
50
0.058
$103.5
$0.12
97% KMnO4 in a 55-lb pail (approximately 4
gal)
7.5 gal or 100 Ib of KMnO4 per year



Electricity
Electricity Unit Cost ($/kwh)
Estimated Electricity Usage (kwh)
Estimated Electricity Cost ($)
Estimated Power Use ($71,000 gal)
0.067
515
$34.54
$0.04

Calculated based on 617 hr of operation of a
0.17-hp chemical feed pump


Labor
Average Weekly Labor (hr)
Total Labor Hours (hr)
Total Labor Cost ($)
Labor Cost ($71,000 gal)
Total O&M Cost/1,000 gal
0.42
11
$231
$0.27
$0.43
5 min/day; 5 days a week
26 weeks
Labor rate = $2 1/hr

-
                         43

-------
                                 Section 5.0:  REFERENCES

Battelle. 2004. Quality Assurance Project Plan for Evaluation of Arsenic Removal Technology.
       Prepared under Contract No. 68-C-00-185, Task Order No. 0029, for U.S. Environmental
       Protection Agency, National Risk Management Research Laboratory, Cincinnati, OH.

Carlson, Kenneth H., and William R. Knocke. 1999. "Modeling Manganese Oxidation with KMnO4 for
       Drinking Water Treatment." JAWWA 125(10): 892-896.

Chen, A.S.C., L. Wang, J. Oxenham, and W. Condit. 2004. Capital Costs of Arsenic Removal
       Technologies: U.S. EPA Arsenic Removal Technology Demonstration Program Round 1.
       EPA/600/R-04/20 1 . U.S. Environmental Protection Agency, National Risk Management
       Research Laboratory, Cincinnati, OH.

Edwards, M., S. Patel, L.  McNeill, H. Chen, M. Frey, A.D. Eaton, R.C. Antweiler, and H.E. Taylor.
       1998. "Considerations in As Analysis and Speciation." JAWWA 90(3): 103-1 13.

EPA.  2001.  National Primary Drinking Water Regulations: Arsenic and Clarifications to Compliance
       and New Source Contaminants Monitoring. Federal Register, 40 CFR Part 9, 141, and 142.

EPA.  2002.  Lead and Copper Monitoring and Reporting Guidance for Public Water Systems.
       EPA/816/R-02/009. U.S. Environmental Protection Agency, Office of Water, Washington, D.C.

EPA.  2003. Minor Clarification of the National Primary Drinking Water Regulation for Arsenic. Federal
       Register, 40 CFR Part 141.

Gregory, D.,  and K. Carlson. 2003. "Effect of Soluble Mn Concentration on Oxidation Kinetics ."
       JAWWA 95(1) :9
Knocke, William R., Hoehn, Robert C.; Sinsabaugh, Robert L. 1987. "Using Alternative Oxidants to
       ;           ;       ;         ;         O "                    O
       Remove Dissolved Manganese from Waters Laden with Organics." JAWWA, 79(3): 75-79.

Knocke, William R., John E. Van Benschoten, Maureen J. Kearney, Andrew W. Soborski, and David A.
       Reckhow.  1990. Alternative Oxidants for the Remove of Soluble Iron and Manganese. Final
       report prepared for the AWWA Research Foundation, Denver, CO.

Knocke, William R., John E. Van Venschoten, Maureen J. Kearney, Andrew W, Soborski, and David A.
       Reckhow.  1991. "Kinetics of Manganese and Iron Oxidation by Potassium Permanganate and
       Chlorine Dioxide."  JAWWA 83(6): 80-87.

Knocke, William R., Holly L.  Shorney, and Julia D. Bellamy.  1994. "Examining the Reactions Between
       Soluble Iron, DOC, and Alternative Oxidants During Conventional Treatment." JAWWA 86(1):
       117-127.

Post, Tim. 2005. "Pollution Cleanup Cost is Hard to Comprehend." Minnesota Public Radio. Available
       at: http://news.minnesota.publicradio.org/features/2005/10/10jostt impairedcleanup/.

Salbu, B. and E. Steinnes. 1995.  Trace Elements in Natural Waters. CRC Press, Boca Raton, Florida.
                                             44

-------
Sauk River Watershed District. 2006. "Monitoring Our Resources."  Available at:
       http: //www. srwdmn.org/monitoring/html.

Sorg, T.J.  2002. "Iron Treatment for Arsenic Removal Neglected." Opflow, AWWA, 28(11): 15.

Wang, L., W. Condit, and A.S.C. Chen.  2004. Technology Selection and System Design:  U.S. EPA
       Arsenic Removal Technology Demonstration Program Round 1.  EPA/600/R-05/001.  U.S.
       Environmental Protection Agency, National Risk Management Research Laboratory, Cincinnati,
       OH.
                                             45

-------
   APPENDIX A




OPERATIONAL DATA

-------
US EPA Arsenic Demonstration Project at BSLMHP, MN - Daily System Operation Log Sheet
Week
No.
1
2
3
4
5
6
7
Date
07/1 3/05
07/14/05
07/15/05
07/16/05
07/17/05
07/18/05
07/19/05
07/20/05
07/21/05
07/22/05
07/23/05
07/24/05
07/25/05
07/26/05
07/27/05
07/28/05
07/29/05
07/30/05
07/31/05
08/01/05
08/02/05
08/03/05
08/04/05
08/05/05
08/06/05
08/07/05
08/08/05
08/09/05(a' b
08/10/05
08/1 1/05"
08/12/05
08/13/05
08/14/05
08/15/05
08/16/05
08/17/05
08/18/05
08/19/05
08/20/05
08/21/05
08/22/05
08/23/05
08/24/05
08/25/05
08/26/05
08/27/05
08/28/05
Time
21:13
20:10
20:00
NM
NM
18:45
19:10
19:00
18:30
20:00
NM
NM
19:30
20:10
23:15
20:15
18:15
NM
NM
19:05
20:30
23:55
23:55
22:00
NM
NM
21:30
21:30
NM
18:00
20:30
NM
NM
21:00
21:30
20:00
19:15
20:30
NM
NM
20:20
22:10
21:00
21:15
NM
NM
NM
New Well
Hour
Meter
(hr)
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
Dally
Operation
(hr)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Volume to
Treatment
Dally
Volume
(gal)
NA
6,980
6,880
NA
NA
NA
9,125
11,075
14,470
7,680
NA
NA
14,250
4,020
4,030
4,180
3,340
NA
NA
18,670
6,057
4,733
3,635
3,985
NA
NA
12,020
7,195
NA
4,885
5,200
NA
NA
15,090
5,410
3,360
5,860
4,620
NA
NA
11,460
5,140
4,400
3,680
2,970
NA
NA
Average
Flowrate
(gpm)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Pressure Tanks
Pressure
Tankl
(psig)
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
52
46
45
53
45
NM
NM
60
55
NM
50
56
NM
NM
54
45
55
49
54
NM
NM
54
60
46
48
53
NM
NM
Pressure
Tank 2
(psig)
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
72
72
72
72
72
NM
NM
72
72
NM
46
50
NM
NM
49
40
54
45
50
NM
NM
51
59
44
44
50
NM
NM
Pressure Filtration
IN
(psig)
42
54
40
NM
NM
58
40
45
45
48
NM
NM
47
41
58
41
56
NM
NM
58
42
41
49
42
NM
NM
58
55
NM
42
48
NM
NM
46
43
55
42
47
NM
NM
48
50
42
40
48
NM
NM
TA/TB
(psig)
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
44
NM
40
38
NM
NM
42
38
30
36
42
NM
NM
45
46
40
34
43
NM
NM
TC/TD
(psig)
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
38
NM
34
30
NM
NM
34
30
22
30
38
NM
NM
40
40
34
30
40
NM
NM
OUT
(psig)
30
37
30
NM
NM
45
36
22
35
42
NM
NM
45
40
55
40
53
NM
NM
48
40
38
46
40
NM
NM
50
42
NM
40
37
NM
NM
40
39
30
36
43
NM
NM
42
47
40
33
46
NM
NM
AP
Across
System
(psig)
12
17
10
NA
NA
13
4
23
10
6
NA
NA
2
1
3
1
3
NM
NM
10
2
3
3
2
NM
NM
8
13
NM
2
11
NA
NA
6
4
25
6
4
NA
NA
6
3
2
7
2
NA
NA
Volume to
Distribution
Flowrate
(gpm)
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
Daily
Volume
(gal)
NA
7,000
6,700
NA
NA
NA
8,900
10,800
14,300
7,400
NA
NA
13,900
3,900
3,900
3,900
3,200
NM
NM
18,100
5,900
4,500
3,600
3,700
NM
NM
11,600
5,100
NM
4,200
5,000
NA
NA
14,600
5,100
3,400
5,530
4,205
NA
NA
10,315
5,110
3,930
3,640
2,535
NA
NA
Backwash
No. of Tanks
Backwashed
NM
1
2
NM
NM
NM
3
3
4
3
NM
NM
4
1
1
2
1
NM
NM
5
2
1
1
2
NM
NM
4
13
NM
6
1
NM
NM
4
1
2
1
3
NM
NM
6
1
3
0
3
NM
NM
Wastewater
Produced
(gal)
NA
110
240
NA
NA
NA
350
440
490
360
NA
NA
470
130
110
240
120
NA
NA
710
240
120
120
240
NA
NA
490
1,720
NA
740
150
NA
NA
490
120
320
130
380
NA
NA
790
120
370
0
380
NA
NA
KMnO4 Application
KMn04
Tank
Level
(in)
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
30.0
29.6
29.3
28.9
28.7
NM
NM
27.8
27.4
NM
27.1
26.8
NM
NM
25.8
25.4
25.3
24.9
24.6
NM
NM
23.9
23.8
23.5
23.4
23.1
NM
NM
Average
KMn04
Dose
(mg/L)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
3.8
NA
NA
3.4
NA
NA
3.2
NA
NA
2.5
NA
NA

-------
US EPA Arsenic Demonstration Project at BSLMHP, MN - Daily System Operation Log Sheet (continued)
Week
No.
8
9
10
11
12
13
14
Date
08/29/05
08/30/05
08/31/05
09/01/05
09/02/05
09/03/05
09/04/05
09/05/05
09/06/05
09/07/05
09/08/05
09/09/05
09/10/05
09/11/05
09/12/05
09/13/05
09/14/05
09/15/05
09/16/05
09/17/05
09/18/05
09/19/05
09/20/05
09/21 /051"'
09/22/05
09/23/05
09/24/05
09/25/05
09/26/05
09/27/05
9/28/2005(e)
09/29/05
09/30/05'"
10/01/05
1 0/02/05
1 0/03/05
1 0/04/05
1 0/05/05
1 0/06/05
1 0/07/05
1 0/08/05
10/09/05
10/10/05
10/11/05
10/12/05
10/13/05
10/14/05
10/15/05
1 0/1 6/05
Time
21:00
21:00
22:30
21:30
21:15
NM
NM
20:00
21:30
20:15
21:15
20:30
NM
NM
21:00
22:15
23:50
22:00
21:00
NM
NM
20:00
17:30
20:00
20:15
20:00
NM
NM
21:15
20:30
19:15
19:30
21:30
NM
NM
21:30
21:30
23:30
18:30
18:30
NM
NM
NM
18:45
17:15
20:00
20:00
NM
NM
New Well
Hour
Meter
(hr)
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
0.3
3.2
6.1
NM
NM
15.8
18.9
21.0
23.7
26.2
NM
NM
NM
38.1
40.9
43.9
47.2
NM
NM
Dally
Operation
(hr)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
2.9
2.9
NA
NA
9.7
3.1
2.1
2.7
2.5
NA
NA
NA
11.9
2.8
3.0
3.3
NA
NA
Volume to
Treatment
Dally
Volume
(qal)
13,980
7,290
4,530
3,240
3,190
NA
NA
15,955
5,155
4,320
4,750
5,010
NA
NA
10,840
4,675
4,990
3,020
2,715
NA
NA
13,405
4,860
4,940
4,665
2,890
NA
NA
12,350
4,190
3,105
3,255
5,345
NA
NA
14,010
4,493
4,377
2,313
3,617
NA
NA
NA
17,050
3,900
4,280
4,870
NA
NA
Average
Flowrate
(qpm)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
19
31
NA
NA
24
24
35
14
24
NA
NA
NA
24
23
24
25
NA
NA
Pressure Tanks
Pressure
Tankl
(pslq)
50
48
55
48
45
NM
NM
51
50
50
60
54
NM
NM
46
45
48
60
55
NM
NM
51
45
62
60
45
NM
NM
54
50
60
52
50
NM
NM
60
55
65
60
45
NM
NM
NM
50
64
50
50
NM
NM
Pressure
Tank 2
(pslq)
48
43
50
46
42
NM
NM
50
49
46
55
50
NM
NM
44
43
48
60
53
NM
NM
50
48
60
58
43
NM
NM
51
47
60
50
50
NM
NM
57
52
60
56
43
NM
NM
NM
48
60
47
49
NM
NM
Pressure Filtration
IN
(pslq)
44
42
49
43
40
NM
NM
46
45
40
59
55
NM
NM
41
45
43
55
52
NM
NM
46
45
55
53
43
NM
NM
47
45
55
43
45
NM
NM
54
50
57
55
43
NM
NM
NM
42
57
44
45
NM
NM
TA/TB
(pslq)
40
40
48
40
38
NM
NM
42
43
40
55
52
NM
NM
36
40
41
52
50
NM
NM
42
52
46
55
40
NM
NM
42
43
50
40
42
NM
NM
50
46
56
40
38
NM
NM
NM
38
54
38
40
NM
NM
TC/TD
(pslq)
32
30
40
32
36
NM
NM
35
34
32
48
46
NM
NM
30
35
34
46
45
NM
NM
34
46
43
50
32
NM
NM
34
40
44
32
40
NM
NM
50
44
52
42
38
NM
NM
NM
36
53
36
40
NM
NM
OUT
(pslq)
40
39
46
40
38
NM
NM
44
42
40
54
50
NM
NM
38
44
41
52
50
NM
NM
42
44
52
52
38
NM
NM
43
43
51
40
43
NM
NM
52
49
55
50
37
NM
NM
NM
38
54
38
40
NM
NM
AP
Across
System
(pslq)
4
3
3
3
2
NA
NA
2
3
0
5
5
NA
NA
3
1
2
3
2
NA
NA
4
1
3
1
5
NA
NA
4
2
4
3
2
NA
NA
2
1
2
5
6
NA
NA
NA
4
3
6
5
NA
NA
Volume to
Distribution
Flowrate
(qpm)
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
6.0
5.5
1.0
NM
NM
0.0
3.0
1.0
3.0
10.0
NM
NM
NM
9.0
2.5
9.0
2.5
NM
NM
Dally
Volume
(qal)
13,775
5,900
4,095
3,125
3,120
NA
NA
14,625
4,725
4,205
4,275
5,040
NA
NA
9,645
4,255
4,435
2,925
2,700
NA
NA
12,265
4,095
4,495
4,445
3,000
NA
NA
11,670
4,185
2,755
3,840
3,850
NA
NA
12,880
4,060
4,175
2,295
3,350
NA
NA
NA
15,975
3,503
3,862
4,500
NA
NA
Backwash
No. of Tanks
Backwashed
5
4
2
1
2
NM
NM
7
3
1
3
1
NM
NM
5
2
3
0
1
NM
NM
6
5
2
0
0
NM
NM
4
0
0
3
2
NM
NM
7
3
1
0
1
NM
NM
NM
7
3
3
2
NM
NM
Wastewater
Produced
(qal)
660
520
250
140
200
NA
NA
850
390
140
390
120
NA
NA
660
270
390
0
130
NA
NA
770
660
280
0
0
NA
NA
480
0
0
360
250
NA
NA
890
390
170
0
80
NA
NA
NA
870
380
390
250
NA
NA
KMnO4 Application
KMn04
Tank
Level
(In)
22.4
22.0
21.9
21.8
21.6
NM
NM
21.4
21.2
21.1
20.9
20.6
NM
NM
20.1
19.9
19.6
19.5
19.4
NM
NM
18.7
18.5
18.2
18.0
17.9
NM
NM
31.8
31.6
31.5
31.3
31.0
NM
NM
30.3
30.1
29.9
29.8
29.6
NM
NM
NM
28.8
28.6
28.4
28.2
NM
NM
Average
KMnO4
Dose
(mq/L)
2.5
NA
NA
2.1
NA
NA

2.7
NA
NA
2.6
NA
NA

2.7
NA
NA
2.6
NA
NA
NA
2.6
NA
NA

-------
US EPA Arsenic Demonstration Project at BSLMHP, MN - Daily System Operation Log Sheet (continued)
Week
No.
15
16
17
18
19
20
Date
10/17/05
10/18/05
10/19/05
10/20/05
10/21/05
10/22/05
10/23/05
10/24/05
10/25/05
10/26/05
10/27/05
10/28/05
10/29/05
10/30/05
10/31/05
11/01/05
11/02/05
11/03/05
11/04/05
11/05/05
11/06/05
11/07/05
11/08/05
11/09/05
11/10/05
11/11/05
11/12/05
11/13/05
11/14/05
11/15/05
11/16/05
11/17/05
11/18/05
11/19/05
11/20/05
11/21/05
11/22/05
11/23/05
11/24/05
11/25/05
11/26/05
11/27/05
Time
20:30
20:15
20:15
21:30
20:30
NM
NM
20:00
22:15
18:30
18:30
21:30
NM
NM
18:30
18:30
18:00
16:30
19:00
NM
NM
18:30
17:00
19:30
21:15
23:15
NM
NM
17:00
18:00
18:30
18:00
17:30
NM
NM
10:15
18:00
21:00
17:30
18:30
NM
NM
New Wei I
Hour
Meter
(hr)
59.3
62.5
66.1
70.5
74.0
NM
NM
82.4
85.1
87.4
89.9
92.6
NM
NM
100.5
103.1
105.9
108.0
111.0
NM
NM
117.5
120.8
124.4
127.8
130.4
NM
NM
138.0
141.0
144.3
146.8
149.5
NM
NM
159.5
162.9
167.3
169.8
173.4
NM
NM
Daily
Operation
(hr)
12.1
3.2
3.6
4.4
3.5
NA
NA
8.4
2.7
2.3
2.5
2.7
NA
NA
7.9
2.6
2.8
2.1
3.0
NA
NA
6.5
3.3
3.6
3.4
2.6
NA
NA
7.6
3.0
3.3
2.5
2.7
NA
NA
10.0
3.4
4.4
2.5
3.6
NA
NA
Volume to
Treatment
Daily
Volume
(qal)
NA
NA
5,625
6,975
5,530
NA
NA
13,260
4,100
3,570
3,430
3,890
NA
NA
1 1 ,440
4,793
3,357
3,080
4,760
NA
NA
10,057
4,836
5,449
5,168
4,081
NA
NA
10,986
4,743
5,648
3,782
4,226
NA
NA
14,983
5,391
7,070
3,665
5,928
NA
NA
Average
Flowrate
(qpm)
NA
NA
26
26
26
NA
NA
26
25
26
23
24
NA
NA
24
31
20
24
26
NA
NA
26
24
25
25
26
NA
NA
24
26
29
25
26
NA
NA
25
26
27
24
27
NA
NA
Pressure Tanks
Pressure
Tankl
(psiq)
55
58
65
50
54
NM
NM
65
55
52
60
60
NM
NM
55
64
55
65
56
NM
NM
60
54
65
55
56
NM
NM
55
55
55
63
63
NM
NM
62
65
65
58
55
NM
NM
Pressure
Tank 2
(psiq)
52
54
60
48
50
NM
NM
60
52
49
55
55
NM
NM
50
60
50
60
54
NM
NM
55
50
60
45
54
NM
NM
53
50
49
60
60
NM
NM
60
62
60
60
50
NM
NM
Pressure Filtration
IN
(psiq)
48
51
56
52
55
NM
NM
56
50
44
53
54
NM
NM
48
58
46
57
51
NM
NM
48
46
57
41
51
NM
NM
48
47
53
58
56
NM
NM
60
58
58
52
42
NM
NM
TA/TB
(psiq)
45
43
52
48
40
NM
NM
52
46
30
45
50
NM
NM
45
54
42
52
44
NM
NM
44
38
48
40
45
NM
NM
33
42
50
54
54
NM
NM
55
54
52
50
40
NM
NM
TC/TD
(psiq)
44
42
50
45
40
NM
NM
50
44
30
44
50
NM
NM
44
52
40
52
44
NM
NM
44
38
48
39
44
NM
NM
32
42
50
54
52
NM
NM
54
53
52
48
40
NM
NM
OUT
(psiq)
45
45
52
46
40
NM
NM
50
47
31
48
50
NM
NM
46
53
41
53
45
NM
NM
45
40
50
40
46
NM
NM
35
40
49
55
55
NM
NM
55
55
55
50
38
NM
NM
AP
Across
System
(psiq)
3
6
4
6
15
NA
NA
6
3
13
5
4
NA
NA
2
5
5
4
6
NA
NA
3
6
7
1
5
NA
NA
13
7
4
3
1
NA
NA
5
3
3
2
4
NA
NA
Volume to
Distribution
Flowrate
(qpm)
3.0
7.5
2.5
NM
NM
NM
NM
5.0
1.0
15.0
0.0
0.0
NM
NM
0.0
1.0
2.5
2.5
2.5
NM
NM
2.5
6.0
8.0
2.5
1.0
NM
NM
2.5
6.0
5.0
1.0
0.0
NM
NM
7.5
5.0
7.5
1.0
1.5
NM
NM
Daily
Volume
(qal)
16,780
4,390
5,195
6,335
5,055
NA
NA
12,265
3,750
3,360
3,105
3,635
NA
NA
10,480
3,643
3,927
2,930
4,315
NA
NA
8,826
4,749
5,050
4,640
3,705
NA
NA
10,081
4,209
5,095
3,470
3,860
NA
NA
13,405
4,805
6,240
3,250
5,330
NA
NA
Backwash
No. of Tanks
Backwashed
10
2
2
3
2
NM
NM
4
3
1
2
1
NM
NM
5
1
3
1
2
NM
NM
5
1
3
2
2
NM
NM
6
2
3
1
2
NM
NM
9
3
5
2
2
NM
NM
Wastewater
Produced
(qal)
1,260
260
250
360
250
NA
NA
520
350
120
250
120
NA
NA
610
120
360
120
240
NA
NA
690
140
360
240
240
NA
NA
820
240
380
130
250
NA
NA
1,160
390
640
260
250
NA
NA
KMnO4 Application
KMnO4
Tank
Level
(in)
27.3
27.0
26.8
26.4
26.2
NM
NM
25.5
25.3
25.1
24.9
24.8
NM
NM
24.3
24.0
23.8
23.7
23.4
NM
NM
NM
22.6
22.3
22.1
21.9
NM
NM
21.3
20.9
20.4
20.0
19.6
NM
NM
18.3
18.0
17.5
17.3
NM
NM
NM
Average
KMnO4
Dose
(mq/L)
2.6
NA
NA
2.5
NA
NA
3.0
NA
NA
2.8
NA
NA
5.0
NA
NA
3.5
NA
NA

-------
US EPA Arsenic Demonstration Project at BSLMHP, MN - Daily System Operation Log Sheet (continued)
Week
No.
21
22
23
24
25
26
Date
11/28/05
11/29/05
11/30/05
12/01/05
12/02/05
12/03/05
12/04/05
12/05/05
12/06/05
12/07/05
12/08/05
12/09/05
12/10/05
12/11/05
12/12/05
12/13/05
12/14/05
12/15/05
12/16/05
12/17/05
12/18/05
12/19/05
12/20/05
12/21/05
12/22/05
12/23/05
12/24/05
12/25/05
12/26/05
12/27/05
12/28/05
12/29/05
12/30/05
12/31/05
01/01/06
01/02/06
01/03/0619'
01/04/06
01/05/06
01/06/06
01/07/06
01/08/06
Time
17:00
18:30
14:30
18:00
21:00
NM
NM
21:30
20:00
19:00
13:00
18:00
NM
NM
18:00
19:30
19:30
18:00
07:12
NM
NM
21:00
18:00
19:00
19:30
NM
NM
NM
18:30
18:00
19:15
19:30
19:00
NM
NM
10:30
12:30
17:00
10:00
19:30
NM
NM
New Wei I
Hour
Meter
(hr)
183.2
187.2
189.9
193.8
197.9
NM
NM
210.0
213.1
216.7
219.0
224.1
NM
NM
235.4
239.4
242.8
246.3
249.4
NM
NM
266.2
270.9
277.3
281.7
NM
NM
NM
300.8
304.9
309.7
313.5
316.9
NM
NM
331.5
336.2
336.4
341.2
345.8
NM
NM
Daily
Operation
(hr)
9.8
4.0
2.7
3.9
4.1
NA
NA
12.1
3.1
3.6
2.3
5.1
NA
NA
11.3
4.0
3.4
3.5
3.1
NA
NA
16.8
4.7
6.4
4.4
NA
NA
NA
19.1
4.1
4.8
3.8
3.4
NA
NA
14.6
4.7
0.2
4.8
4.6
NA
NA
Volume to
Treatment
Daily
Volume
(gal)
15,655
5,872
4,010
6,390
6,161
NA
NA
18,880
4,700
5,966
4,068
8,590
NA
NA
16,425
5,975
4,790
4,624
4,651
NA
NA
26,100
7,760
9,970
6,250
NA
NA
NA
29,172
6,770
5,908
5,570
4,650
NM
NM
23,050
NA
4,550
7,365
7,290
NM
NM
Average
Flowrate
(gpm)
27
24
25
27
25
NA
NA
26
25
28
29
28
NA
NA
24
25
23
22
25
NA
NA
26
28
26
24
NA
NA
NA
25
28
21
24
23
NM
NM
26
NA
NA
26
26
NM
NM
Pressure Tanks
Pressure
Tankl
(psig)
55
55
54
55
54
NM
NM
55
54
54
65
64
NM
NM
64
55
65
65
55
NM
NM
55
59
62
60
NM
NM
NM
54
65
54
54
55
NM
NM
65
60
65
56
55
NM
NM
Pressure
Tank 2
(psig)
52
50
50
50
50
NM
NM
50
50
50
60
60
NM
NM
60
50
60
60
50
NM
NM
48
55
56
60
NM
NM
NM
50
60
50
47
45
NM
NM
60
55
59
53
52
NM
NM
Pressure Filtration
IN
(psig)
46
46
44
41
44
NM
NM
41
49
44
57
56
NM
NM
59
48
56
55
40
NM
NM
45
53
53
59
NM
NM
NM
42
55
45
44
42
NM
NM
57
58
55
50
48
NM
NM
TA/TB
(psig)
41
42
40
39
40
NM
NM
38
42
39
52
51
NM
NM
56
46
52
49
30
NM
NM
39
48
48
52
NM
NM
NM
37
43
40
40
40
NM
NM
56
53
52
44
40
NM
NM
TC/TD
(psig)
40
40
39
38
40
NM
NM
36
42
36
50
50
NM
NM
54
44
50
47
30
NM
NM
38
47
47
52
NM
NM
NM
36
42
38
39
40
NM
NM
55
52
50
44
40
NM
NM
OUT
(psig)
40
42
40
39
41
NM
NM
37
45
37
52
52
NM
NM
55
45
52
49
30
NM
NM
40
48
48
54
NM
NM
NM
37
45
40
40
40
NM
NM
55
52
50
45
40
NM
NM
AP
Across
System
(psig)
6
4
4
2
3
NA
NA
4
4
7
5
4
NA
NA
4
3
4
6
10
NA
NA
5
5
5
5
NA
NA
NA
5
10
5
4
2
NA
NA
2
6
5
5
8
NA
NA
Volume to
Distribution
Flowrate
(gpm)
1.0
3.0
1.0
7.5
1.5
NM
NM
3.0
5.0
3.0
3.0
6.0
NM
NM
3.0
5.0
2.5
4.0
10.0
NM
NM
6.0
7.5
4.0
4.0
NM
NM
NM
5.0
5.0
3.0
2.0
3.0
NM
NM
2.0
2.0
3.0
2.5
7.5
NM
NM
Daily
Volume
(gal)
14,115
5,230
3,350
5,855
5,250
NA
NA
16,870
4,345
5,445
3,220
7,310
NA
NA
15,230
5,500
4,360
4,170
4,200
NA
NA
23,690
6,970
8,900
5,690
NA
NA
NA
25,740
4,900
5,925
4,845
2,320
NA
NA
21,300
6,080
3,940
6,130
6,000
NA
NA
Backwash
No. of Tanks
Backwashed
8
3
4
2
4
NM
NM
11
1
2
4
5
NM
NM
7
2
1
3
2
NM
NM
10
4
5
3
NM
NM
NM
14
4
3
3
1
NM
NM
12
4
3
4
4
NM
NM
Wastewater
Produced
(gal)
1,020
370
490
250
510
NA
NA
1,390
130
250
480
700
NA
NA
970
250
120
370
210
NA
NA
1,320
470
590
350
NA
NA
NA
1,770
470
350
350
140
NA
NA
1,620
470
350
460
470
NA
NA
KMnO4 Application
KMnO4
Tank
Level
(in)
29.6
29.1
28.8
28.3
27.8
NM
NM
26.0
25.6
25.1
24.7
23.9
NM
NM
22.3
21.7
21.3
20.6
20.4
NM
NM
17.8
17.0
30.6
30.0
NM
NM
NM
27.4
26.8
26.1
25.6
25.2
NM
NM
23.0
22.4
21.9
21.1
20.5
NM
NM
Average
KMnO4
Dose
(mg/L)
4.8
NA
NA
5.1
NA
NA
5.4
NA
NA
5.4
NA
NA
NA
6.1
NA
NA
5.6
NA
NA

-------
                US EPA Arsenic Demonstration Project at BSLMHP, MN - Daily System Operation Log Sheet (continued)
Week
No.
27
28
Date
01/09/06
01/10/06
01/11/06
01/12/06
01/13/06
01/14/06
01/15/06
1/16/2006(n)
01/17/06
Time
18:00
17:00
17:00
19:15
19:00
NM
NM
17:00
21:00
New Wei I
Hour
Meter
(hr)
362.3
366.4
370.5
376.1
380.5
NM
NM
394.7
399.8
Daily
Operation
(hr)
16.5
4.1
4.1
5.6
4.4
NA
NA
14.2
5.1
Volume to
Treatment
Daily
Volume
(qal)
25,483
6,060
6,400
8,635
6,385
NM
NM
21,965
7,757
Average
Flowrate
(qpm)
26
25
26
26
24
NM
NM
26
25
Pressure Tanks
Pressure
Tankl
(psiq)
55
65
65
54
55
NM
NM
65
65
Pressure
Tank 2
(psiq)
50
60
60
50
50
NM
NM
60
60
Pressure Filtration
IN
(psiq)
45
55
59
43
45
NM
NM
59
57
TA/TB
(psiq)
40
51
52
39
38
NM
NM
52
50
TC/TD
(psiq)
38
52
50
38
36
NM
NM
50
50
OUT
(psiq)
40
53
52
38
38
NM
NM
52
52
AP
Across
System
(psiq)
5
2
7
5
7
NA
NA
7
5
Volume to
Distribution
Flowrate
(qpm)
4.0
1.0
12.0
5.0
3.0
NM
NM
8.0
12.5
Daily
Volume
(qal)
21,500
4,835
5,495
7,125
5,465
NA
NA
NA
6,280
Backwash
No. of Tanks
Backwashed
12
5
2
5
3
NM
NM
13
5
Wastewater
Produced
(qal)
1,520
690
250
650
380
NA
NA
1,640
620
KMnO4 Application
KMnO4
Tank
Level
(in)
17.9
17.3
31.0
30.2
29.6
NM
NM
27.6
26.9
Average
KMnO4
Dose
(mq/L)
5.5
NA
NA
5.9
Note:
(a)  On 08/09/05, both sets of duplex filters stuck in backwash mode due to sediment dislodged in purge/control valve, preventing it from closing. System
    bypassed.
(b)  On 08/09/06, a pressure gauge after each set of duplex filters installed.
(c)  On 08/11/05, pressure gauge on pressure tank 2 replaced.
(d)  On 09/21/05, two flow meters, one on  treated water line and one on backwash discharge line, installed although readings not recorded until 09/28/05.
(e)  On 09/28/05 hour meter installed.
(f)  On 09/30/06, pressure gauge changed out for duplex units TC/TD.
(g)  On 01/03/06, totalizer to treatment re-set.
(h)  Totalizer to distribution re-set.
NM = not measured
NA = not available

-------
   APPENDIX B




ANALYTICAL DATA

-------
                                       Analytical Results from Long Term Sampling at BSLMHP, MN
Sampling Date
Sampling Location
Parameter Unit
Stroke Length
Alkalinity
(as CaCO3)
Fluoride
Sulfate
Nitrate (as N)
Orthophosphate (as
P04)
P (total) (as P)
Silica (as SiO2)
Turbidity
TOG
PH
Temperature
DO
ORP
Total Hardness
(as CaCO3)
Ca Hardness
(as CaCO3)
Mg Hardness
(as CaCO3)
As (total)
As (soluble)
As (particulate)
As (III)
As(V)
Fe (total)
Fe (soluble)
Mn (total)
Mn (soluble)
%
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
NTU
mg/L
S.U.
°C
mg/L
mV
mg/L
mg/L
mg/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
07/13/05
IN
AC
TT
33
352
0.2
<1
0.1
<0.05

23.3
25.0


14.9
2.0
-23
383
228
155
36.4
30.3
6.1
13.9
16.5
3,315
2,792
154
133
374
0.2
<1
0.1
<0.05

23.3
3.1

7.5
12.7
0.7
196
330
197
133
29.6
3.3
26.3
1.6
1.7
3,173
<25
996
377
374
0.2
<1
0.3
0.05

22.7
0.6

7.7
12.3
1.1
219
329
197
132
4.3
3.0
1.3
1.7
1.3
157
<25
428
391
07/20/05
IN
AC
TT
33
365


-
<0.05

24.7
23.0

7.3
11.4
2.5
-29
-
-

34.7

-
-
-
2,786
-
139
-
365


-
<0.05

24.4
2.8

7.2
12.3
0.5
85
-
-

26.7

-
-
-
2,766
-
634
-
361


-
<0.05

24.2
0.5

7.2
11.9
0.7
144
-
-

17.7

-
-
-
482
-
561
-
07/26/05
IN
AC
TT
33
365


-
<0.05

23.5
25.0

7.4
10.4
3.6
-40
-
-

26.6

-
-
-
2,864
-
137
-
370


-
<0.05

23.6
2.9

7.3
11.0
1.7
144
-
-

24.8

-
-
-
2,704
-
844
-
365


-
<0.05

23.9
0.1

7.2
11.0
0.9
173
-
-

5.5

-
-
-
45
-
727
-
08/02/05
IN
AC
TT
33
352


-
<0.05

23.8
26.0

7.4
11.2
3.5
-35
-
-

25.7

-
-
-
2,964
-
135
-
365


-
<0.05

24.0
4.7

7.3
12.1
1.0
154
-
-

23.0

-
-
-
2,578
-
1,126
-
374


-
<0.05

23.6
11.0

7.3
12.1
1.2
196
-
-

8.0

-
-
-
666
-
487
-
08/18/05(a'b)
IN
AC
TT
26
352
0.2
<1
<0.05
<0.05

24.1
33.0
4.1
7.2
1.0
0.9
-76
320
188
131
26.4
26.2
0.2
24.1
2.1
2,895
2,954
139
142
365
0.2
<1
<0.05
<0.05

24.2
3.7
3.9
7.3
14.1
0.9
2
317
190
128
23.2
4.8
18.4
2.6
2.2
2,773
<25
1,097
850
361
0.2
<1
<0.05
<0.05

23.9
0.4
4.0
7.3
13.8
0.7
43
323
187
137
5.1
4.8
0.3
3.4
1.4
<25
<25
1,010
1,000
08/24/05
IN
AC
TA/TB
TC/TD
26
352


-
<0.05

29.4
24.0

7.3
12.3
10.4
-48
-
-

30.4

-
-
-
2,764
-
130
-
365


-
<0.05

28.6
2.9

7.4
12.8
0.8
138
-
-

31.5

-
-
-
2,706
-
871
-
361


-
<0.05

28.4
0.7

7.3
12.5
0.7
159
-
-

3.5

-
-
-
<25
-
475
-
374


-
<0.05

28.2
0.2

7.4
12.8
1.1
181
-
-

3.3

-
-
-
<25
-
467
-
(a) Onsite water quality parameters taken on 08/17/05. (b) System bypassed on 08/09/05 and samples not collected that week.

-------
                                           Analytical Results from Long Term Sampling at BSLMHP, MN (continued)
Sampling Date
Sampling Location
Parameter Unit
Stroke Length
Alkalinity
(as CaCO3)
Fluoride
Sulfate
Nitrate (as N)
Orthophosphate (as
P04)
P (total) (as P)
Silica (as SiO2)
Turbidity
TOG
PH
Temperature
DO
ORP
Total Hardness
(as CaCO3)
Ca Hardness
(as CaCO3)
Mg Hardness
(as CaCO3)
As (total)
As (soluble)
As (particulate)
As (III)
As(V)
re (total)
-e (soluble)
Mn (total)
Mn (soluble)
%
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
NTU
mg/L
S.U.
°C
mg/L
mV
mg/L
mg/L
mg/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
08/31/05
-------
                                            Analytical Results from Long Term Sampling at BSLMHP, MN (continued)
Sampling Date
Sampling Location
Parameter Unit
Stroke Length
Alkalinity
(as CaCO3)
Fluoride
Sulfate
Nitrate (as N)
Orthophosphate (as
P04)
P (total) (as P)
Silica (as SiO2)
Turbidity
TOG
PH
Temperature
DO
ORP
Total Hardness
(as CaCO3)
Ca Hardness
(as CaCO3)
Mg Hardness
(as CaCO3)
As (total)
As (soluble)
As (particulate)
As (III)
As(V)
Fe (total)
Fe (soluble)
Mn (total)
Mn (soluble)
%
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
NTU
mg/L
S.U.
°C
mg/L
mV
mg/L
mg/L
mg/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
10/05/05
-------
                                            Analytical Results from Long Term Sampling at BSLMHP, MN (continued)
Sampling Date
Sampling Location
Parameter Unit
Stroke Length
Alkalinity
(as CaCO3)
Fluoride
Sulfate
Nitrate (as N)
Orthophosphate (as
P04)
P (total) (as P)
Silica (as SiO2)
Turbidity
roc
pH
Temperature
DO
ORP
Total Hardness
(as CaCO3)
Ca Hardness
(as CaCO3)
Mg Hardness
(as CaCO3)
As (total)
As (soluble)
As (particulate)
As (III)
As(V)
Fe (total)
Fe (soluble)
Mn (total)
Mn (soluble)
%
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
NTU
mg/L
S.U.
°C
mg/L
mV
mg/L
mg/L
mg/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
11/09/05
IN
AC
TA/TB
TC/TD
26
370
-



0.5
23.9
33.0
-
7.4
10.0
0.9
-38


-
36.6



-
2,549
-
117
-
370
-



0.5
23.6
3.2
-
7.4
10.2
0.8
39


-
36.1



-
2,425
-
1,031
-
365
-



0.1
24.0
0.1
-
7.4
10.2
0.9
65


-
11.3



-
336
-
951
-
370
-



0.1
24.0
0.7
-
7.4
10.4
0.8
68


-
5.7



-
68
-
971
-
11/15/05
-------
                                          Analytical Results from Long Term Sampling at BSLMHP, MN (continued)
Cd
Sampling Date
Sampling Location
Parameter Unit
Stroke Length
Alkalinity
(as CaCO3)
Fluoride
Sulfate
Nitrate (as N)
Orthophosphate (as
P04)
P (total) (as P)
Silica (as SiO2)
Turbidity
roc
pH
Temperature
DO
ORP
Total Hardness
(as CaCO3)
Ca Hardness
(as CaCO3)
Mg Hardness
(as CaCO3)
As (total)
As (soluble)
As (particulate)
As (III)
As(V)
Fe (total)
Fe (soluble)
Mn (total)
Mn (soluble)
%
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
NTU
mg/L
S.U.
°C
mg/L
mV
mg/L
mg/L
mg/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
M9/L
01/05/06
-------